From Fierz-Pauli to Einstein-Hilbert

Σχετικά έγγραφα
6.1. Dirac Equation. Hamiltonian. Dirac Eq.

Higher Derivative Gravity Theories

Symmetric Stress-Energy Tensor

Space-Time Symmetries

Dark matter from Dark Energy-Baryonic Matter Couplings

Torsional Newton-Cartan gravity from a pre-newtonian expansion of GR

Physics 582, Problem Set 2 Solutions

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

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)

A Short Introduction to Tensors

2 Composition. Invertible Mappings

Concrete Mathematics Exercises from 30 September 2016

EE512: Error Control Coding

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

3+1 Splitting of the Generalized Harmonic Equations

Section 8.3 Trigonometric Equations

derivation of the Laplacian from rectangular to spherical coordinates

CHAPTER 25 SOLVING EQUATIONS BY ITERATIVE METHODS

Geometric Lagrangians. for. massive higher-spin fields

Palatiniversus metricformalism inhigher-curvaturegravity

Homework 3 Solutions

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

4.6 Autoregressive Moving Average Model ARMA(1,1)

UV fixed-point structure of the 3d Thirring model

THE ENERGY-MOMENTUM TENSOR IN CLASSICAL FIELD THEORY

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

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

Cosmological Space-Times

Example Sheet 3 Solutions

Homework 3 Solutions

Dr. D. Dinev, Department of Structural Mechanics, UACEG

Inverse trigonometric functions & General Solution of Trigonometric Equations

Tutorial problem set 6,

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

Finite Field Problems: Solutions

상대론적고에너지중이온충돌에서 제트입자와관련된제동복사 박가영 인하대학교 윤진희교수님, 권민정교수님

SCHOOL OF MATHEMATICAL SCIENCES G11LMA Linear Mathematics Examination Solutions

Srednicki Chapter 55

Non-Gaussianity from Lifshitz Scalar

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

Relativistic particle dynamics and deformed symmetry

Section 7.6 Double and Half Angle Formulas

( y) Partial Differential Equations

CHAPTER 101 FOURIER SERIES FOR PERIODIC FUNCTIONS OF PERIOD

THE ENERGY-MOMENTUM TENSOR FOR THE GRAVITATIONAL FIELD.

8.324 Relativistic Quantum Field Theory II

The Simply Typed Lambda Calculus

Toward a non-metric theory of gravity

ST5224: Advanced Statistical Theory II

Matrices and Determinants

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

5. Choice under Uncertainty

Reminders: linear functions

The challenges of non-stable predicates

Elements of Information Theory

Jesse Maassen and Mark Lundstrom Purdue University November 25, 2013

Higher spin gauge theories and their CFT duals

Solutions Ph 236a Week 8

Nonminimal derivative coupling scalar-tensor theories: odd-parity perturbations and black hole stability

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

Forced Pendulum Numerical approach

On the Galois Group of Linear Difference-Differential Equations

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

Fractional Colorings and Zykov Products of graphs

A Short Introduction to Tensor Analysis

Three coupled amplitudes for the πη, K K and πη channels without data

About these lecture notes. Simply Typed λ-calculus. Types

Lecture 2. Soundness and completeness of propositional logic

Hartree-Fock Theory. Solving electronic structure problem on computers

Second Order Partial Differential Equations

Overview. Transition Semantics. Configurations and the transition relation. Executions and computation

Lifting Entry (continued)

1 String with massive end-points

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

Lecture 34 Bootstrap confidence intervals

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.

From the finite to the transfinite: Λµ-terms and streams

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

Strain gauge and rosettes

Riemannian Curvature

1 Lorentz transformation of the Maxwell equations

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

The ε-pseudospectrum of a Matrix

Quadratic Expressions

Eulerian Simulation of Large Deformations

Bounding Nonsplitting Enumeration Degrees

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

The kinetic and potential energies as T = 1 2. (m i η2 i k(η i+1 η i ) 2 ). (3) The Hooke s law F = Y ξ, (6) with a discrete analog

Solutions for String Theory 101

Statistical Inference I Locally most powerful tests

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

6.3 Forecasting ARMA processes

Problem Set 3: Solutions

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

ΠΤΥΧΙΑΚΗ ΕΡΓΑΣΙΑ ΒΑΛΕΝΤΙΝΑ ΠΑΠΑΔΟΠΟΥΛΟΥ Α.Μ.: 09/061. Υπεύθυνος Καθηγητής: Σάββας Μακρίδης

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

Estimation for ARMA Processes with Stable Noise. Matt Calder & Richard A. Davis Colorado State University

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

DESIGN OF MACHINERY SOLUTION MANUAL h in h 4 0.

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

Transcript:

From Fierz-Pauli to Einstein-Hilbert Gravity as a special relativistic field theory Bert Janssen Universidad de Granada & CAFPE A pedagogical review References: R. Feynman et al, The Feynman Lectures on Gravitation, 1995 T. Ortín, Gravity and Strings, 2004 B. Janssen (UGR) Granada, 14 dec 2006 1

Central Idea Orionites with knowlegde of special relativity quantum field theory standard model & QCD NOT general relativity B. Janssen (UGR) Granada, 14 dec 2006 2

Central Idea Orionites with knowlegde of special relativity quantum field theory standard model & QCD NOT general relativity discover gravity F = G N m 1 m 2 r 2 B. Janssen (UGR) Granada, 14 dec 2006 2

Central Idea Orionites with knowlegde of special relativity quantum field theory standard model & QCD NOT general relativity discover gravity F = G N m 1 m 2 r 2 They will try to describe gravity as a special relativistic field theory try to find a bosonic field (graviton) that is responsable for gravity B. Janssen (UGR) Granada, 14 dec 2006 2

Properties of the graviton: Gravity is long range = Massless B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) is universally attractive = even spin (spin 0, 2, 4,...) B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) is universally attractive = even spin (spin 0, 2, 4,...) has deflection of light = not spin 0 spin 0 propagator k 2 only coupling T µ µ k 2 T ν ν electromagnetism (in D = 4): T µ µ = 0 B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) is universally attractive = even spin (spin 0, 2, 4,...) has deflection of light = not spin 0 spin 0 propagator k 2 only coupling T µ µ k 2 T ν ν electromagnetism (in D = 4): T µ µ = 0 theoretical problems for spin 5/2 = spin 2 B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) is universally attractive = even spin (spin 0, 2, 4,...) has deflection of light = not spin 0 spin 0 propagator k 2 only coupling T µ µ k 2 T ν ν electromagnetism (in D = 4): T µ µ = 0 theoretical problems for spin 5/2 = spin 2 = Spin 2 field h µν : Representation theory: 16 = 9 + 6 + 1 with 9 = symmetric & traceless 6 = antisymmetric = spin 1 1 = trace = spin 0 B. Janssen (UGR) Granada, 14 dec 2006 3

Properties of the graviton: Gravity is long range = Massless is static = integer spin (bosonic field) is universally attractive = even spin (spin 0, 2, 4,...) has deflection of light = not spin 0 spin 0 propagator k 2 only coupling T µ µ k 2 T ν ν electromagnetism (in D = 4): T µ µ = 0 theoretical problems for spin 5/2 = spin 2 = Spin 2 field h µν : Representation theory: 16 = 9 + 6 + 1 with 9 = symmetric & traceless 6 = antisymmetric = spin 1 1 = trace = spin 0 = Massless traceless symmetric spin 2 field h µν : B. Janssen (UGR) Granada, 14 dec 2006 3

Outlook 1. Introduction 1: Short review of General Relativity 2. Introduction 2: Electromagnetism as SRFT of spin 1 B. Janssen (UGR) Granada, 14 dec 2006 4

Outlook 1. Introduction 1: Short review of General Relativity 2. Introduction 2: Electromagnetism as SRFT of spin 1 3. Construction of Fierz-Pauli theory 4. Consistency problems with couplings 5. Deser s argument & General Relativity 6. (Philosophical) Conclusions B. Janssen (UGR) Granada, 14 dec 2006 4

1 Short review of General Relativity Einstein & Hilbert (1915): Gravity is curvature of spacetime, caused by matter Interaction between g µν and T µν S = 1 κ d 4 x ] g [R + L matter R µν 1 2 g µνr = κ T µν with R µν = µ Γ λ νλ λ Γ λ µν + Γ λ µσγ σ λν Γ λ µνγ σ λσ ( ) Γ ρ µν = 1 2 gρσ µ g σν + ν g µσ σ g µν T µν = 1 g δl matter δg µν Very geometrical way to understand gravity B. Janssen (UGR) Granada, 14 dec 2006 5

Linear perturbations of Minkowski space Suppose g µν = η µν + εh µν h = h µ µ = η µν h µν B. Janssen (UGR) Granada, 14 dec 2006 6

Linear perturbations of Minkowski space Suppose g µν = η µν + εh µν h = h µ µ = η µν h µν = g µν η µν εh µν R µν 1 2 ε [ 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ ] R µν 1 2 g µνr 1 2 [ ε 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ ( )] η µν 2 h ρ λ h ρλ T µν 0 + ετ µν B. Janssen (UGR) Granada, 14 dec 2006 6

Linear perturbations of Minkowski space Suppose g µν = η µν + εh µν h = h µ µ = η µν h µν = g µν η µν εh µν R µν 1 2 ε [ 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ ] 1 2 R µν 1 2 g µνr [ 1 2 ε 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ ( )] η µν 2 h ρ λ h ρλ T µν 0 + ετ µν Einstein eqn R µν = κ(t µν 1 2 g µνt ) = ] [ 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ κ(τ µν 1 2 η µντ) B. Janssen (UGR) Granada, 14 dec 2006 6

Intermezzo ( Linear perturbations of general solution Suppose g µν = ḡ µν + εh µν h = h µ µ = ḡ µν h µν = R µν R µν + 1 2 ε [ 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ ] T µν T µν + ετ µν Einstein eqn R µν = κ(t µν 1 2 g µνt ) = [ε 0 ] : Rµν = κ( T µν 1 T 2ḡµν ) ] [ε 1 1 ] : 2 [ 2 h µν + µ ν h µ ρ h νρ ν ρ h µρ κ (τ µν 12ḡµντ 12 h T µν + 12ḡµνh ) ρλ Tρλ ) B. Janssen (UGR) Granada, 14 dec 2006 7

Gauge invariance Suppose g µν = η µν + εh µν δx µ = ξ µ, with ξ µ = 0 + εχ µ B. Janssen (UGR) Granada, 14 dec 2006 8

Gauge invariance Suppose g µν = η µν + εh µν δx µ = ξ µ, with ξ µ = 0 + εχ µ = δg µν = ξ ρ ρ g µν + µ ξ ρ g ρν + ν ξ ρ g µρ [ε 0 ] : δη µν = 0 [ε 1 ] : δh µν = µ χ ν + ν χ µ Gauge invariance B. Janssen (UGR) Granada, 14 dec 2006 8

Gauge invariance Suppose g µν = η µν + εh µν δx µ = ξ µ, with ξ µ = 0 + εχ µ = δg µν = ξ ρ ρ g µν + µ ξ ρ g ρν + ν ξ ρ g µρ [ε 0 ] : δη µν = 0 [ε 1 ] : δh µν = µ χ ν + ν χ µ Gauge invariance = δγ ρ µν = ε µ ν χ ρ δr µνρλ = δr µν = 0 B. Janssen (UGR) Granada, 14 dec 2006 8

Gauge invariance Suppose g µν = η µν + εh µν δx µ = ξ µ, with ξ µ = 0 + εχ µ = δg µν = ξ ρ ρ g µν + µ ξ ρ g ρν + ν ξ ρ g µρ [ε 0 ] : δη µν = 0 [ε 1 ] : δh µν = µ χ ν + ν χ µ Gauge invariance = δγ ρ µν = ε µ ν χ ρ δr µνρλ = δr µν = 0 NB: g µν = η µν + εh µν with h µν = µ χ ν + ν χ µ = g µν = η µν since h µν is pure gauge B. Janssen (UGR) Granada, 14 dec 2006 8

Let s forget about all this (for a moment) B. Janssen (UGR) Granada, 14 dec 2006 9

2 Electromagnetism as SRFT of spin 1 Massless spin 1 field A µ : µ µ A ν = j ν B. Janssen (UGR) Granada, 14 dec 2006 10

2 Electromagnetism as SRFT of spin 1 Massless spin 1 field A µ : µ µ A ν = j ν not pos. def. energy density due to non-physical states (A µ : 4 components photon: 2 helicity states) B. Janssen (UGR) Granada, 14 dec 2006 10

2 Electromagnetism as SRFT of spin 1 Massless spin 1 field A µ : µ µ A ν = j ν not pos. def. energy density due to non-physical states (A µ : 4 components photon: 2 helicity states) We can impose Lorenz condition µ A µ = 0 by hand = µ j µ = 0 (charge conservation) B. Janssen (UGR) Granada, 14 dec 2006 10

2 Electromagnetism as SRFT of spin 1 Massless spin 1 field A µ : µ µ A ν = j ν not pos. def. energy density due to non-physical states (A µ : 4 components photon: 2 helicity states) We can impose Lorenz condition µ A µ = 0 by hand = µ j µ = 0 (charge conservation) We would like L such that eqns of motion yield F µ (A) = j µ charge conservation µ F µ (A) = 0 gauge identity implying Lorenz cond. B. Janssen (UGR) Granada, 14 dec 2006 10

Most general Lorentz-invar action, quadratic in A L = α µ A ν µ A ν + β µ A ν ν A µ Eqns of motion: F ν (A) = α 2 A ν + β µ ν A µ = 0 Gauge condition: ν F ν (A) 0 = α = β B. Janssen (UGR) Granada, 14 dec 2006 11

Most general Lorentz-invar action, quadratic in A L = α µ A ν µ A ν + β µ A ν ν A µ Eqns of motion: F ν (A) = α 2 A ν + β µ ν A µ = 0 Gauge condition: ν F ν (A) 0 = α = β The action is [ µ A ν µ A ν µ A ν ν A µ ] L = 1 2 = 1 4 F µνf µν with F µν = µ A ν ν A µ F ν = µ F µν B. Janssen (UGR) Granada, 14 dec 2006 11

Most general Lorentz-invar action, quadratic in A L = α µ A ν µ A ν + β µ A ν ν A µ Eqns of motion: F ν (A) = α 2 A ν + β µ ν A µ = 0 Gauge condition: ν F ν (A) 0 = α = β The action is L = 1 2 [ µ A ν µ A ν µ A ν ν A µ ] = 1 4 F µνf µν with F µν = µ A ν ν A µ F ν = µ F µν Gauge invariance: [ δl δl = µ δ( µ A ν ) δl ] δa ν = µ F µν δa ν µ ν F µν Λ δa ν = δa µ = µ Λ B. Janssen (UGR) Granada, 14 dec 2006 11

Coupling to matter L 0 = 1 2 µφ µ φ 1 4 F µνf µν Eqns of motion: µ µ φ = 0, µ µ φ = 0, µ F µν = 0 Invariant under: δa µ = µ Λ (local U(1)) δφ = ieλφ (global U(1)) B. Janssen (UGR) Granada, 14 dec 2006 12

Coupling to matter L 0 = 1 2 µφ µ φ 1 4 F µνf µν Eqns of motion: µ µ φ = 0, µ µ φ = 0, µ F µν = 0 Invariant under: δa µ = µ Λ (local U(1)) δφ = ieλφ (global U(1)) Noether current for λ: δ λ L 0 ) = ie 2 µλ (φ µ φ φ µ φ = eλ µ j µ (0) j µ (0) preserved on-shell: µj µ (0) ] = [φ i 2 µ µ φ φ µ µ φ B. Janssen (UGR) Granada, 14 dec 2006 12

Coupling to matter L 0 = 1 2 µφ µ φ 1 4 F µνf µν Eqns of motion: µ µ φ = 0, µ µ φ = 0, µ F µν = 0 Invariant under: δa µ = µ Λ (local U(1)) δφ = ieλφ (global U(1)) Noether current for λ: δ λ L 0 ) = ie 2 µλ (φ µ φ φ µ φ = eλ µ j µ (0) j µ (0) preserved on-shell: µj µ (0) ] = [φ i 2 µ µ φ φ µ µ φ j µ (0) is not source term for F µν add coupling term L ea µ j µ (0) B. Janssen (UGR) Granada, 14 dec 2006 12

Coupling term L 1 = 1 2 µφ µ φ 1 4 F µνf µν + ea µ j µ (0) (φ µ φ φ µ φ ) with j µ (0) = ie 2 Eqn of motion: µ F µν = ej ν (0) µ µ φ ie µ A µ φ 2ieA µ µ φ = 0 µ µ φ + ie µ A µ φ + 2ieA µ µ φ = 0 Problem: j µ (0) is not conserved on-shell! 0 = µ ν F µν = µ j µ (0) 0 INCONSISTENCY!!! B. Janssen (UGR) Granada, 14 dec 2006 13

Coupling term L 1 = 1 2 µφ µ φ 1 4 F µνf µν + ea µ j µ (0) (φ µ φ φ µ φ ) with j µ (0) = ie 2 Eqn of motion: µ F µν = ej ν (0) µ µ φ ie µ A µ φ 2ieA µ µ φ = 0 µ µ φ + ie µ A µ φ + 2ieA µ µ φ = 0 Problem: j µ (0) is not conserved on-shell! 0 = µ ν F µν = µ j µ (0) 0 INCONSISTENCY!!! Reason: j µ (0) is Noether current of L 0, not of L 1 Real Noether current: j µ (1) = jµ (0) + eaµ φ φ B. Janssen (UGR) Granada, 14 dec 2006 13

Noether procedure Identify λ and Λ δa µ = µ Λ, δφ = ieλφ B. Janssen (UGR) Granada, 14 dec 2006 14

Noether procedure Identify λ and Λ Then: δl 0 = e Λj µ (0) 0 δa µ = µ Λ, δφ = ieλφ add L ea µ j µ (0) B. Janssen (UGR) Granada, 14 dec 2006 14

Noether procedure Identify λ and Λ Then: δl 0 = e Λj µ (0) 0 δa µ = µ Λ, δφ = ieλφ add L ea µ j µ (0) δl 1 = e Λj µ (0) + e Λjµ (0) e2 µ ΛA µ φ φ 0 add L 1 2 e2 A µ A µ φ φ B. Janssen (UGR) Granada, 14 dec 2006 14

Noether procedure Identify λ and Λ Then: δl 0 = e Λj µ (0) 0 δa µ = µ Λ, δφ = ieλφ add L ea µ j µ (0) δl 1 = e Λj µ (0) + e Λjµ (0) e2 µ ΛA µ φ φ 0 add L 1 2 e2 A µ A µ φ φ L 2 = 1 2 µφ µ φ 1 4 F µνf µν + ea µ j µ (0) + 1 2 e2 A µ A µ φ φ δl 2 = e 2 µ ΛA µ φ φ + e 2 µ ΛA µ φ φ = 0 B. Janssen (UGR) Granada, 14 dec 2006 14

Nota bene: L 2 = 1 2 µφ µ φ 1 4 F µνf µν + ie 2 A µ(φ µ φ φ µ φ ) + 1 2 e2 A µ A µ φ φ = 1 2 ( µφ + iea µ φ )( µ φ iea µ φ) 1 4 F µνf µν = D µ φ D µ φ 1 4 F µνf µν with: D µ φ = µ φ iea µ φ δd µ φ = D µ (δφ) B. Janssen (UGR) Granada, 14 dec 2006 15

Nota bene: L 2 = 1 2 µφ µ φ 1 4 F µνf µν + ie 2 A µ(φ µ φ φ µ φ ) + 1 2 e2 A µ A µ φ φ = 1 2 ( µφ + iea µ φ )( µ φ iea µ φ) 1 4 F µνf µν = D µ φ D µ φ 1 4 F µνf µν with: D µ φ = µ φ iea µ φ δd µ φ = D µ (δφ) Lesson: Lorentz symm & gauge invariance = free theory Gauge invariance & consistent couplings = fully consistent theory coupled to matter B. Janssen (UGR) Granada, 14 dec 2006 15

3 Gravity as a SRFT of spin 2 Massless spin 2 field h µν : ρ ρ h µν = κt µν not pos. def. energy density due to non-physical states (h µν : 4 components graviton: 2 helicity states) B. Janssen (UGR) Granada, 14 dec 2006 16

3 Gravity as a SRFT of spin 2 Massless spin 2 field h µν : ρ ρ h µν = κt µν not pos. def. energy density due to non-physical states (h µν : 4 components graviton: 2 helicity states) Look for action L such that eqns of motion yield H µν (h) = κt µν energy conservation µ T µν = 0 = µ H µν (h) = 0 gauge invariant action B. Janssen (UGR) Granada, 14 dec 2006 16

Most general Lorentz-invar action, quadratic in h L = a µ h νρ µ h νρ + b µ h νρ ν h µρ + c µ h ρ h µρ + d µ h µ h with h = h µ µ = η µν h µν Eqns of motion: H νρ (h) a 2 h νρ + 2b µ (ν h ρ)µ + c (ν ρ) h + c η νρ µ λ h µλ + 2d η νρ 2 h = 0 Gauge condition: ν H νρ (h) 0 = a = 1 2 b = 1 2 c = d B. Janssen (UGR) Granada, 14 dec 2006 17

Most general Lorentz-invar action, quadratic in h L = a µ h νρ µ h νρ + b µ h νρ ν h µρ + c µ h ρ h µρ + d µ h µ h with h = h µ µ = η µν h µν Eqns of motion: H νρ (h) a 2 h νρ + 2b µ (ν h ρ)µ + c (ν ρ) h + c η νρ µ λ h µλ + 2d η νρ 2 h = 0 Gauge condition: ν H νρ (h) 0 = a = 1 2 b = 1 2 c = d The action then is L = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h H µν (h) 2 h µν + µ ν h ρ µ h νρ ρ ν h µρ η µν 2 h + η µν ρ λ h ρλ = 0 (NB: H µν (h) is first order of R µν 1 2 g µνr for g µν = η µν + εh µν ) B. Janssen (UGR) Granada, 14 dec 2006 17

Gauge invariance [ δl δl = µ δ( µ h νρ ) δl ] δh µν = H νρ δh νρ ν H νρ χ δh νρ = δh µν = µ χ ν + ν χ µ (NB: δh µν is first order of δg µν for g µν = η µν + εh µν ) B. Janssen (UGR) Granada, 14 dec 2006 18

Gauge invariance [ δl δl = µ δ( µ h νρ ) δl ] δh µν = H νρ δh νρ ν H νρ χ δh νρ = δh µν = µ χ ν + ν χ µ (NB: δh µν is first order of δg µν for g µν = η µν + εh µν ) Transverse traceless gauge: µ h µν = 0 = h = H µν (h) = ρ ρ h µν = 0 De Donder gauge: µ hµν = 0 with h µν = h µν 1 2 η µνh = H µν (h) = ρ ρ hµν = 0 B. Janssen (UGR) Granada, 14 dec 2006 18

Coupling to matter L = L FP + L φ + κ h µν T µν (φ) Eqns of motion: H µν = κt µν (φ) 2 φ +... = 0 Gauge condition: µ H µν = 0 = µ T µν (φ) = 0 B. Janssen (UGR) Granada, 14 dec 2006 19

Coupling to matter L = L FP + L φ + κ h µν T µν (φ) Eqns of motion: H µν = κt µν (φ) 2 φ +... = 0 Gauge condition: µ H µν = 0 = µ T µν (φ) = 0 Questions: 1. How good is this theory? Experimental verification 2. Is µ T µν (φ) = 0 consistent with 2 φ +... = 0? Consistency problems B. Janssen (UGR) Granada, 14 dec 2006 19

Newtonian limit Point mass: T µν = M δ µ 0 δν 0 δ(x ρ ) = H 00 = κmδ(x ρ ), H ij = 0 = h 00 = Φ, h ii = Φδ ij Φ = M 8π x Action for mass m moving in gravitational potential Φ: 1 } S = m dt{ 1 v2 + [1 + 1 v 2 v2 ]Φ { } 1 dt 2 mv2 mφ 1 4 mv4 + 3 2 mv2 Φ +... (Newtonian potential) = Good agreement for light deflection, time dilatation,... Perihelium Mercury is 0.75 observed value B. Janssen (UGR) Granada, 14 dec 2006 20

4 Consistency problems with couplings L 0 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ Eqns of motion: H µν = 0 µ µ φ = 0 Invariant under: δh µν = µ χ ν + ν χ µ + κ Σ λ λ h µν δφ = κ Σ µ µ φ δx µ = κ Σ µ with Σ µ global symm B. Janssen (UGR) Granada, 14 dec 2006 21

4 Consistency problems with couplings L 0 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ Eqns of motion: H µν = 0 µ µ φ = 0 Invariant under: δh µν = µ χ ν + ν χ µ + κ Σ λ λ h µν δφ = κ Σ µ µ φ δx µ = κ Σ µ Noether current for Σ µ : ] δ Σ L 0 = κ µ Σ [T ν µν (φ) + T µν (h) with Σ µ global symm with T µν (φ) energy-momentum tensor of matter T µν (h) gravitational energy-momentum tensor B. Janssen (UGR) Granada, 14 dec 2006 21

T µν (φ) = µ φ ν φ + 1 2 η µν( φ) 2 T µν (h) = 1 2 µh ρλ ν h ρλ + µ h ρλ ρ h λ ν 1 2 µh ρ h νρ 1 2 ρ h µ h νρ + 1 2 µh µ h + η µν L FP T µν (φ) and T µν (h) preserved on shell: µ T µν (φ) = 2 φ ν φ µ T µν (h) = ν h µρ H µρ B. Janssen (UGR) Granada, 14 dec 2006 22

T µν (φ) = µ φ ν φ + 1 2 η µν( φ) 2 T µν (h) = 1 2 µh ρλ ν h ρλ + µ h ρλ ρ h ν λ 1 2 µh ρ h νρ 1 2 ρ h µ h νρ + 1 2 µh µ h + η µν L FP T µν (φ) and T µν (h) preserved on shell: µ T µν (φ) = 2 φ ν φ µ T µν (h) = ν h µρ H µρ T µν (φ) is not source term for h µν : add coupling term L κh µν T µν (φ) B. Janssen (UGR) Granada, 14 dec 2006 22

Coupling term L 1 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κhµν T µν (φ) Eqns of motion: H µν = κt µν (φ) ] µ µ φ = κ µ [h µν ν φ 1 2 h µφ Problem: T µν (φ) is no longer conserved on shell! 0 = µ H µν = κ µ T µν (φ) = κ 2 φ ν φ 0 (-22) INCONSISTENCY!!! B. Janssen (UGR) Granada, 14 dec 2006 23

Coupling term L 1 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κhµν T µν (φ) Eqns of motion: H µν = κt µν (φ) ] µ µ φ = κ µ [h µν ν φ 1 2 h µφ Problem: T µν (φ) is no longer conserved on shell! Reason: 0 = µ H µν = κ µ T µν (φ) = κ 2 φ ν φ 0 (-22) INCONSISTENCY!!! T µν (φ) is Noether current of L 0, not of L 1 Total energy conserved, not only energy of φ Gravity couples also to own energy non-linear theory B. Janssen (UGR) Granada, 14 dec 2006 23

Gravity self-coupling L 2 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κhµν T µν (φ) + 1 2 κhµν T µν (h) B. Janssen (UGR) Granada, 14 dec 2006 24

Gravity self-coupling L 2 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κhµν T µν (φ) + 1 2 κhµν T µν (h) TOO NAIVE!!! Eqn of motion does not give H µν = κt µν (φ) + κt µν (h) due to contribuition of T µν (h) B. Janssen (UGR) Granada, 14 dec 2006 24

Gravity self-coupling L 2 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κhµν T µν (φ) + 1 2 κhµν T µν (h) TOO NAIVE!!! Eqn of motion does not give H µν = κt µν (φ) + κt µν (h) due to contribuition of T µν (h) T µν (h) too simple for energy-monentum tensor for h µν B. Janssen (UGR) Granada, 14 dec 2006 24

We need L corr = κ h µν L µν κ h( h)( h) such that L 3 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κ hµν T µν (φ) + κ h µν L µν yields B. Janssen (UGR) Granada, 14 dec 2006 25

We need L corr = κ h µν L µν κ h( h)( h) such that L 3 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ yields 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κ hµν T µν (φ) + κ h µν L µν ( ) T µν (h) = L µν σ h ρλ δl ρλ δ σ h µν B. Janssen (UGR) Granada, 14 dec 2006 25

We need L corr = κ h µν L µν κ h( h)( h) such that L 3 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ yields 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κ hµν T µν (φ) + κ h µν L µν ( ) T µν (h) = L µν σ h ρλ δl ρλ δ σ h µν L 3 is invariant upto order κ 2 under δh µν = µ χ ν + ν χ µ + κ χ ρ ρ h µν δφ = κ χ ρ ρ φ B. Janssen (UGR) Granada, 14 dec 2006 25

We need L corr = κ h µν L µν κ h( h)( h) such that L 3 = 1 4 µh νρ µ h νρ 1 2 µh νρ ν h µρ + 1 2 µh ρ h µρ yields 1 4 µh µ h + 1 2 µφ µ φ + 1 2 κ hµν T µν (φ) + κ h µν L µν ( ) T µν (h) = L µν σ h ρλ δl ρλ δ σ h µν L 3 is invariant upto order κ 2 under δh µν = µ χ ν + ν χ µ + κ χ ρ ρ h µν δφ = κ χ ρ ρ φ ) H µν = κt µν (φ) + κt µν (h) (T µ µν (φ) + T µν (h) = 0 B. Janssen (UGR) Granada, 14 dec 2006 25

Most general term for L corr : L corr = 1 2 κ hµν [α µ h ρλ ν h ρλ + β µ h ρλ ρ h λ ν +... ] 20 terms with 16 coefficients determine coefficients by demanding µ T µν (h) = γ ν ρλ Hρλ B. Janssen (UGR) Granada, 14 dec 2006 26

Most general term for L corr : L corr = 1 2 κ hµν [α µ h ρλ ν h ρλ + β µ h ρλ ρ h λ ν +... ] 20 terms with 16 coefficients determine coefficients by demanding µ T µν (h) = γ ν ρλ Hρλ = L corr = 1 2 κ hµν [ 1 2 µh ρλ ν h ρλ ρ h µ λ ρ h λν + λ h ρ (µ ρ h ν)λ +2 λ h ρ (µ ν) h ρλ (µ h ν)λ λ h λ h µν ρ h ρλ λ h λ(µ ν) h + λ h µν λ h + 1 2 µh ν h + η µν L FP ] Eqns of motion: H µν = κt µν (h) + κt µν (φ) ] µ µ φ = κ µ [h µν ν φ 1 2 h m uφ ] Conservation of energy: [T µ µν (h) + T µν (φ) = 0 B. Janssen (UGR) Granada, 14 dec 2006 26

Consistent result upto order κ Right values for perihelium of Mercury previous deficit due to neglecting self-interaction What about order κ 2, κ 3,...? Extremely tedious!! B. Janssen (UGR) Granada, 14 dec 2006 27

Consistent result upto order κ Right values for perihelium of Mercury previous deficit due to neglecting self-interaction What about order κ 2, κ 3,...? Extremely tedious!! WE NEED A MORE GENERAL ARGUMENT! B. Janssen (UGR) Granada, 14 dec 2006 27

5 Deser s argument & General Relativity L 0 = κ h ] ] µν [ µ Γ ρ νρ ρ Γ ρ µν + η [Γ µν λ µργ ρ νλ Γλ µνγ ρ λρ with h µν = h µν 1 2 η µνh Invariant under: δh µν = µ χ ν + ν χ µ δγ ρ µν = µ ν χ ρ Γ ρ µν = Γ ρ νµ Eqns of motion: [ h µν ] : ρ Γ ρ µν 1 2 µγ ρ νρ 1 2 µγ ρ νρ = 0 [Γ ρ µν] : κ ρ hµν + κ λ hλ(µ δρ ν) + 2η λ(µ Γ ν) ρλ η λσ Γ (µ λσ δν) ρ η µν Γ λ ρλ = 0 [Γ ρ µν] is algebraic eqn for Γ ρ µν Constraint on Γ ρ µν B. Janssen (UGR) Granada, 14 dec 2006 28

Constraint on Γ ρ µν: Γ ρ µν = 1 2 κ ηρλ [ µ h λν + ν h µλ λ h µν ] Γ ρ µν is not independent field ( Γ ρ µν is 1st order of Christoffel symbol for g µν = η µν + εh µν ) B. Janssen (UGR) Granada, 14 dec 2006 29

Constraint on Γ ρ µν: Γ ρ µν = 1 2 κ ηρλ [ µ h λν + ν h µλ λ h µν ] Γ ρ µν is not independent field ( Γ ρ µν is 1st order of Christoffel symbol for g µν = η µν + εh µν ) Substitute in [ h µν ]: 2 h µν + µ ν ρ µ h νρ ρ ν h µρ = 0 H µν 1 2 η µνh ρ ρ = 0 ( 1st order of R µν = 0 for g µν = η µν + εh µν ) = L 0 is equivalent to L FP B. Janssen (UGR) Granada, 14 dec 2006 29

We want to find correction to L 0 which takes in account the self-interaction of h µν, such that H µν = κt µν L 1 = κ h ] [ ] µν [ µ Γ ρ νρ ρ Γ ρ µν + (η µν κh µν ) Γ λ µργ ρ νλ Γλ µνγ ρ λρ Eqns of motion: [ h µν ]: R µν (Γ) = 0 [Γ ρ µν]: κ ρ hµν + κ λ hλ(µ δρ ν) + 2(η λ(µ κ h λ(µ )Γ ν) ρλ (η λσ κ h λσ )Γ (µ λσ δν) ρ (η µν κ h µν )Γ λ ρλ = 0 B. Janssen (UGR) Granada, 14 dec 2006 30

We want to find correction to L 0 which takes in account the self-interaction of h µν, such that H µν = κt µν L 1 = κ h ] [ ] µν [ µ Γ ρ νρ ρ Γ ρ µν + (η µν κh µν ) Γ λ µργ ρ νλ Γλ µνγ ρ λρ Eqns of motion: [ h µν ]: R µν (Γ) = 0 [Γ ρ µν]: κ ρ hµν + κ λ hλ(µ δρ ν) + 2(η λ(µ κ h λ(µ )Γ ν) ρλ (η λσ κ h λσ )Γ (µ λσ δν) ρ (η µν κ h µν )Γ λ ρλ = 0 Correction term has no η µν = No contribution to T µν (h) No further correction needed B. Janssen (UGR) Granada, 14 dec 2006 30

We want to find correction to L 0 which takes in account the self-interaction of h µν, such that H µν = κt µν L 1 = κ h ] [ ] µν [ µ Γ ρ νρ ρ Γ ρ µν + (η µν κh µν ) Γ λ µργ ρ νλ Γλ µνγ ρ λρ Eqns of motion: [ h µν ]: R µν (Γ) = 0 [Γ ρ µν]: κ ρ hµν + κ λ hλ(µ δρ ν) + 2(η λ(µ κ h λ(µ )Γ ν) ρλ (η λσ κ h λσ )Γ (µ λσ δν) ρ (η µν κ h µν )Γ λ ρλ = 0 Correction term has no η µν = No contribution to T µν (h) No further correction needed R µν (Γ) = ρ Γ ρ µν µγ ρ νρ + Γλ µν Γρ λρ Γλ µρ Γρ νλ (Surprise!) What is the conecction Γ ρ µν? is Ricci tensor B. Janssen (UGR) Granada, 14 dec 2006 30

Define η µν κ h µν = ĝ µν ĝ µν ĝ νρ = δ ρ µ g µν = ĝ ĝ µν g µν non-linear combination of h µν (infinite series) g µν acts as metric B. Janssen (UGR) Granada, 14 dec 2006 31

Define η µν κ h µν = ĝ µν ĝ µν ĝ νρ = δ ρ µ g µν = ĝ ĝ µν g µν non-linear combination of h µν (infinite series) g µν acts as metric Constraint on Γ ρ µν: Γ ρ µν = 1 2 gρλ [ µ g λν + ν g µλ λ g µν ] Γ ρ µν is Levi-Civita conection R µν (Γ) = R µν (g) [ h µν ] is Einstein eqn R µν = 0 B. Janssen (UGR) Granada, 14 dec 2006 31

Eqn of motion R µν = 0 invariant under general coord transf However L 1 is not! ] add total derivative η µν [ µ Γ ρ νρ ρ Γ ρ µν B. Janssen (UGR) Granada, 14 dec 2006 32

Eqn of motion R µν = 0 invariant under general coord transf However L 1 is not! add total derivative η µν [ µ Γ ρ νρ ρ Γ ρ µν ] = L 2 = (η µν κ h [ ] µν ) µ Γ ρ νρ ρ Γ ρ µν = g µν [ µ Γ ρ νρ ρ Γ ρ µν = g µν R µν (g) [ ] +(η µν κh µν ) Γ λ µργ ρ νλ Γλ µνγ ρ λρ Einstein-Hilbert action! ] [ ] + g µν Γ λ µργ ρ νλ Γλ µνγ ρ λρ B. Janssen (UGR) Granada, 14 dec 2006 32

6 Conclusions Fierz-Pauli L FP is consistent as free theory Couplings to matter = self-coupling of gravity = inconsistent Possible (but difficult) to construct consistent theory upto order κ 2 General Relativity is consistent extension to all orders (unique?) Suprising geometrical interpretation: non-linear function of h µν is metric very elegant & much richer structure B. Janssen (UGR) Granada, 14 dec 2006 33