Plasma Scattering of Electromagnetic Radiation: Theory and Measurement Techniques

Plasma Scattering of Electromagnetic Radiation

Theory and Measurement Techniques

Second Edition

Dustin H. Froula

Senior Scientist/Laboratory, for Laser Energetics/University of, Rochester/250 East River Road, Rochester, NY 14623

Siegfried H. Glenzer

Plasma Physics Group Leader, Lawrence Livermore National Laboratory, P.O. Box 808, L-399, Livermore, Ca 94551

Neville C. Luhmann, Jr.

Department of Applied Science, 3001 Engineering III, University of California, One Shields Ave. Davis, Ca 95616

John Sheffield

Institute for Secure and Sustainable Energy, University of Tennessee, 315 Conference Center Building, Henley Street, Knoxville, TN 37996-4138

AMSTERDAM  •  BOSTON  •  HEIDELBERG  •  LONDON

NEW YORK  •  OXFORD  •  PARIS  •  SAN DIEGO

SAN FRANCISCO  •  SINGAPORE  •  SYDNEY  •  TOKYO

Academic Press is an imprint of Elsevier

Table of Contents

Cover image

Title page

Copyright page

Preface

ACKNOWLEDGMENTS

Chapter 1: Introduction

1.1 INTRODUCTION

CHAPTER 1

CHAPTER 2

CHAPTER 3

CHAPTER 4

CHAPTER 5

CHAPTER 6

CHAPTER 7

CHAPTER 8

CHAPTER 9

CHAPTER 10

CHAPTER 11

CHAPTER 12

APPENDIX A

APPENDIX B

APPENDIX C

APPENDIX D

APPENDIX E

APPENDIX F

1.2 PLASMAS

1.3 SYSTEM OF UNITS

1.4 CHARACTERISTIC LENGTHS AND TIMES IN A PLASMA3

1.5 SCATTERING OF ELECTROMAGNETIC RADIATION BY A PLASMA

1.6 RADIATION BY A MOVING CHARGE

1.7 ACCELERATION OF A CHARGE BY AN ELECTROMAGNETIC WAVE

1.8 GENERAL RESTRICTIONS APPLIED TO CALCULATIONS IN THIS BOOK

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 2: Scattered Power Spectrum

2.1 SPECTRAL DENSITY FUNCTION S(k, ω)

2.2 KINETIC EQUATIONS FOR A PLASMA

2.3 POWER SCATTERED [Ps (R, ωs)]

PROBLEMS

SOLUTIONS TO ODD PROBLEM

Chapter 3: Scattering Spectrum from a Plasma Theory

3.1 INTRODUCTION

3.2 DERIVATION OF ne (k, ω) FOR B=0, v=0

3.3 THE FLUCTUATING DENSITY ne (k, ω)

3.4 THE SPECTRAL DENSITY FUNCTION S(k, ω) FOR A COLLISIONLESS PLASMA

3.5 POWER SCATTERED FROM A HOT PLASMA

3.6 COMMENTS ON THE EFFECTS OF VARIOUS INITIAL CONDITIONS

3.7 S(k, ω) FOR A COLLISIONAL PLASMA B=0

3.8 S(k, ω) FROM THE FLUCTUATION-DISSIPATION THEOREM

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 4: Noncollective Scattering

4.1 INTRODUCTION

4.2 THE ORIGIN OF THE DOPPLER SHIFT IN THE SCATTERED FREQUENCY

4.3 COMPARISON OF FIRST AND SECOND ORDER IN β

4.4 THE FINITE TRANSIT TIME EFFECT

4.5 NONCOLLECTIVE SCATTERING FROM A PLASMA IN THERMODYNAMIC EQUILIBRIUM

4.6 NONCOLLECTIVE SCATTERING FROM A MAGNETIZED PLASMA

4.7 COMMENTS ON THE SCATTERED SPECTRUM

4.8 MEASUREMENT OF THE DIRECTION OF THE MAGNETIC FIELD IN A PLASMA

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 5: Collective Scattering from a Plasma

5.1 INTRODUCTION

5.2 S(k, ω), MAXWELLIAN DISTRIBUTION FUNCTIONS

5.3 S(k, ω), Te/Ti ≃ 1, THE SALPETER APPROXIMATION

5.4 EXPERIMENTAL APPLICATIONS

5.5 COLLISIONAL PLASMAS

5.6 TOTAL CROSS SECTION

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 6: Constraints on Scattering Experiments

6.1 INTRODUCTION

6.2 CHOICE OF A SOURCE (λi, Δλi)

6.3 CHOICE OF A SCATTERING ANGLE (θ, Δθ)

6.4 SIGNAL-TO-NOISE RATIO

6.5 RATIO OF SCATTERED POWER TO BREMSSTRAHLUNG RADIATION POWER

6.6 EFFECT OF THE INCIDENT BEAM ON THE PLASMA

6.7 DETECTORS

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 7: Optical Systems

7.1 INTRODUCTION

7.2 GENERAL PROPERTIES OF SPECTROMETERS: INSTRUMENT FUNCTION

7.3 DIFFRACTION GRATING SPECTROMETER: THEORY

7.4 SPECTROMETER: IMAGE DISSECTORS, APPLICATION

7.5 FABRY-PEROT ETALON: THEORY

7.6 FABRY–PEROT ETALON SPECTROMETER: IMAGE DISSECTORS, APPLICATION

7.7 CALIBRATION AND ALIGNMENT

7.8 TYPICAL COLLECTIVE SCATTERING SETUP AND CONSIDERATIONS

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 8: Techniques

8.1 INTRODUCTION

8.2 FAR FORWARD SCATTERING, PHASE SCINTILLATION IMAGING, AND PHASE CONTRAST IMAGING

8.3 THOMSON SCATTERING FROM DRIVEN (NONTHERMAL) WAVES

8.4 DIRECT FORWARD THOMSON SCATTERING

8.5 (ω, k)-RESOLVED THOMSON SCATTERING

8.6 SUBPICOSECOND TIME RESOLVED THOMSON SCATTERING OF STIMULATED RAMAN SCATTERING

8.7 MULTIPLE ION-ACOUSTIC THOMSON SCATTERING DIAGNOSTICS TO MEASURE (Te, ne)

Chapter 9: Industrial Plasmas and Scattering from Energetic Ions

9.1 INDUSTRIAL AND OTHER LOW-TEMPERATURE PLASMAS

9.2 SCATTERING FROM ENERGETIC IONS

9.3 BURNING PLASMAS

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 10: Scattering from a Magnetized Plasma

10.1 INTRODUCTION1

10.2 CALCULATION OF THE SPECTRAL DENSITY FUNCTION S(k, ω)

10.3 S(k, ω), MAXWELLIAN DISTRIBUTION FUNCTIONS

10.4 COLLISIONAL MAGNETIZED PLASMA

10.5 TRANSVERSE MODES

10.6 GENERAL FEATURES OF THE MAGNETIZED SPECTRUM

10.7 TOTAL CROSS SECTION, ST(k)

10.8 HIGH-FREQUENCY SPECTRUM

10.9 LOW-FREQUENCY SPECTRUM

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 11: X-Ray Thomson Scattering

11.1 OVERVIEW

11.2 X-RAY SCATTERING RELATIONS

11.3 X-RAY SCATTERING EXPERIMENTS

11.4 APPLICATIONS

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Chapter 12: Scattering from Unstable Plasmas

12.1 INTRODUCTION

12.2 MICROSCOPIC INSTABILITY THEORY

12.3 SCATTERING FROM A MARGINALLY STABLE PLASMA

12.4 SCATTERING FROM A WEAKLY UNSTABLE PLASMA

12.5 SCATTERING FROM MICROTURBULENCE IN SHOCK FRONTS

12.7 INSTABILITIES IN INERTIAL CONFINEMENT PLASMAS

PROBLEMS

SOLUTIONS TO ODD PROBLEMS

Appendix A: Mathematical Methods

A.1 COMPLEX VARIABLES AND INTEGRALS IN THE COMPLEX PLANE1

A.2 FOURIER TRANSFORMS

A.3 LAPLACE TRANSFORMS

A.4 STABILITY OF LONGITUDINAL PLASMA OSCILLATIONS

A.5 TOTAL CROSS SECTION FOR A STABLE PLASMA

Appendix B: Kinetic Theory of a Plasma

B.1 INTRODUCTION

B.2 CHARACTERISTIC LENGTHS AND TIMES IN A PLASMA

B.3 THE BOLTZMANN EQUATION

B.4 COMMENTS ON THE COLLISION TERM

B.5 KINETIC DESCRIPTION OF SCATTERING FROM A PLASMA

B.6 THE BBGKY HIERARCHY

B.7 THE KLIMONTOVICH HIERARCHY6

B.8 STABLE, HOMOGENEOUS, QUASI-STATIONARY PLASMAS

Appendix C: General Hot Plasma Dispersion Relation

Appendix D: Computation of the Form Factor

PART 1

PART 2

Appendix E: Review of Work on the Scattering of Radiation from Plasmas

E.1 INTRODUCTION

E.2 SCATTERING FROM THE IONOSPHERE

E.3 SCATTERING FROM LABORATORY PLASMAS WITH λi ≅ L AND ωi ≅ ωpe

E.4 SCATTERING FROM A PLASMA CLOSE TO EQUILIBRIUM, B = 0, v = 0, λi ≪ L, ωi ≫ ωpe

E.5 SCATTERING FROM A MAGNETIZED PLASMA CLOSE TO EQUILIBRIUM

E.6 SCATTERING FROM LASER-PRODUCED PLASMAS

E.7 COLLISIONAL EFFECTS

E.8 RELATIVISTIC EFFECTS

E.9 TOTAL SCATTERING CROSS SECTION

E.10 UNSTABLE AND TURBULENT PLASMA

E.11 THOMSON SCATTERING APPLIED TO LASER-DRIVEN PLASMA WAVES

E.12 ABSORPTION OF THE INCIDENT BEAM AND TWO-BEAM SCATTERING

E.13 INDUSTRIAL PLASMAS

Appendix F: Physical Constants and Formulas

PHYSICAL CONSTANTS

CONVERSION FACTORS

FORMULAS

SYMBOLS

SCATTERING FORMULAS

UNITS

Bibliography

Index

Copyright

Academic Press is an imprint of Elsevier

30 Corporate Drive, Suite 400, Burlington, MA 01803, USA

525 B Street, Suite 1900, San Diego, CA 92101-4495, USA

84 Theobald’s Road, London WC1X 8RR, UK

Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands

First edition 2011

Copyright © 2011 Elsevier Inc. All rights reserved

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher.

Permissions may be sought directly from Elseviers Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permissions@elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material.

Notice

No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

Library of Congress Cataloging-in-Publication Data

Plasma scattering of electromagnetic radiation: experiment, theory and computation/John Sheffield … [et al.]. – 1st ed.

p. cm.

Includes bibliographical references and index.

ISBN 978-0-12-374877-5 (alk. paper)

1. Plasma diagnostics. 2. Electromagnetic waves–Scattering. I. Sheffield, John, 1936-

QC718.5.D5P546 2011

530.4’4–dc22

2010030308

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-12-374877-5

For information on all Academic Press publications visit our website at books.elsevier.com

Printed and bound in USA

Transferred to Digital Pringing in 2013

Preface

In this update of the 1975 book, we have added extensive material on industrial and warm dense matter plasmas, energetic particle measurements and instrumentation, collective Thomson scattering, and techniques. We have also expanded the discussion on relativistic theory and added an appendix on computation.

The form of the spectrum of electromagnetic radiation scattered by a charged particle depends on the mass, charge, position, and velocity of the particle. Similarly, the spectrum of radiation scattered by a plasma, an assembly of free charges, depends on the properties of the plasma. By suitable choice of a source of radiation, we can in principle probe any plasmas and measure quantities such as electron and ion densities and temperatures, ionic charge, magnetic fields, wave amplitudes, and plasma instabilities. It is of particular importance that these measurements can usually be made without significant perturbation of the plasma. In order that we may isolate the effect of the scattering, we require an essentially monochromatic source. Because the scattering cross section is small, the source must be powerful. The growing importance, over the years, of scattering as a diagnostic technique has been associated to a great extent with the development of lasers and microwave sources that satisfy these requirements.

In this monograph, we have attempted to cover both theory and experiment in a consistent and detailed fashion. We have made an effort toward providing quantitative results in which units are clearly defined, and there are no 4π’s and c’s lurking in the background. We review the underlying mathematical techniques.

The book is biased on the experimental side in the direction of laboratory high-temperature plasmas because this is our field of research. To illustrate the application of the technique, we describe a number of significant laboratory experiments that have been made over the last few decades. Scattering is a basic tool of the plasma physicist, and to help those who are not spectroscopists, we have included a section on the most commonly used dispersion instrumentation.

This text is written as an introduction text and reference book for graduate students and researchers. We hope that the readers will find answers to their questions at the level to which they are prepared to go.

ACKNOWLEDGMENTS

John Sheffield started writing the original version of this book during a most enjoyable time at the University of Texas in Austin. He completed the book while working at the Culham Laboratory in England. He owes a great debt to the many colleagues from these two institutions who helped him. As always, he remains truly appreciative of the support of his wife Dace.

Dustin Froula completed writing this second edition while on sabbatical at University of California at Los Angeles and would like to thank his colleagues and friends at Lawrence Livermore for the novel ideas and collaborations that have lead to his understanding of Thomson scattering and plasma physics in general. In his life, there have been many influential teachers and mentors but none greater than his mother and father who have truly provided a solid foundation for both life and science. The love and support of his wife, Lynette, and kids, Scott, Adelae, and Victoria, have been invaluable.

Neville Luhmann would like to dedicate this to his parents who instilled in him a love of science and technology.

Siegfried Glenzer would like to thank his family, his wife Anja and his children Helena and Arend, for support during the time when he worked on this book. He is also grateful to his colleagues at the Lawrence Livermore National Laboratory, Los Alamos National Laboratory, Laboratory for Laser Energetics, Rutherford Appleton Laboratory, Ruhr-Universität Bochum and Deutsches Elektronen Synchrotron where he performed most of his Thomson scattering experiments described here. He also likes to thank the Alexander-von-Humboldt foundation and colleagues at the University of Rostock for enabling his engagement with free electron laser experiments.

The authors are very appreciative of the advice and support of Roland Behn, Henrik Bindslev, Alan Costley, Laurent Divol, Tilo Doeppner, Tony Donné, Carsten Fortmann, Alan Garscadden, Gianluca Gregori, Takaki Hatae, Jerry Hughes, Dave Johnson, Chan Joshi, Robert Kaita, Otto Landen, Lynette Lombardo, Atsushi Mase, David Montgomery, Tobin Munsat, Katsunori Muraoka, Allan Offenberger, John Palastro, Hyeon Park, Brad Pollock, Sean Prunty, Ronald Redmer, James Steven Ross, Christophe Rousseaux, Kenji Tanaka, Th. Tschentscher, Kiichiro Uchino, Michael Walsh, Henri Weisen, Paul Woskov, and Chang-Xuan Yu.

We are deeply indebted to our Elsevier colleagues who provided continuing support in the preparation of this book, Gavin Becker, Mohana Natarajan, and Patricia Osborn.

Chapter 1

Introduction

Chapter Outline Head

1.1 Introduction   2

1.2 Plasmas   5

1.3 System of Units   8

1.3.1 The Gaussian System   9

1.3.2 The SI System   9

1.3.3 Maxwell’s Equations and Lorentz Force on a Unit Charge q   9

1.4 Characteristic Lengths and Times in a Plasma   10

1.5 Scattering of Electromagnetic Radiation by a Plasma   11

1.6 Radiation by a Moving Charge   12

1.7 Acceleration of a Charge by an Electromagnetic Wave   14

1.7.1 Low-Velocity Charge (v/c 1) with No Other Forces Acting   15

1.7.2 High-Velocity Charge with No Other Forces Acting   18

1.7.3 Low-Velocity Charge (v/c 1) with Additional Forces Acting   19

1.7.4 High-Velocity Charge with Applied Magnetic Fields   19

1.8 General Restrictions Applied to Calculations in This Book   20

1.8.1 Macroscopic Interaction of Radiation with a Plasma   20

1.8.2 General Restrictions   21

1.8.3 Noncollective and Collective Spectra   23

Problems   24

Solutions to Odd Problems   25

1.1 INTRODUCTION

It is well known that electromagnetic radiation is emitted by an accelerating charge. An important example of this phenomenon occurs when acceleration is induced by an electromagnetic wave. This interaction, when the incident radiation is of sufficiently low frequency ω that ћω is much less than mec², the rest energy of the charge, is generally referred to as Thomson scattering.¹ It is the extension of the theory to include the simultaneous scattering from a large number of free positive and negative charges, that is, a plasma, and the experimental application of scattering that are the topics discussed in this book.

For a single charge, the angular distribution of intensity, the frequency, and the phase of the scattered radiation depend on the orbit of that charge relative to the observer. Equally, for a large group of charges, the scattered spectrum is related to the orbits of all those charges, or rather in practice, to some average taken over the probable behavior of the group. Anticipating the results derived below, we find that from the spectrum of radiation scattered from a plasma, we may in principle determine the electron and ion temperature, the ionization state, densities the ionization state, the direction and magnitude of a magnetic field in the plasma, and in general, information about all the fluctuations (waves, instabilities) within the plasma. In reality, we are of course limited by the radiation sources available to us; the cross section for scattering is so small that measurements on laboratory plasmas were limited until the advent of high-power lasers. The first measurements were by the scattering of radio waves from the ionosphere in the late 1950s. The history of the subject is discussed briefly in Appendix E.

CHAPTER 1

The purpose of this introductory chapter is to remind the reader of some basic properties of a plasma and of the interaction of radiation with a plasma. The conditions under which radiation will penetrate a plasma are established. The discussion is then restricted to situations where the radiation is primarily transmitted. In this situation, we can reasonably deal with the interaction of each charge in the plasma taken independently. The response of a single charge to radiation is then evaluated. The scattered power is found to be inversely proportional to the mass of a charge, and thus, we can see immediately that the scattering is essentially only from the electrons.

Finally, the problem of adding up the scattered waves from the large number of electrons in the scattering volume is discussed in general terms. The scattered spectrum is found to have two parts. The first is the spectrum that would be obtained if there were no charge interactions, the noncollective spectrum. The second is a result of these interactions (collective effects).

CHAPTER 2

In this chapter, a general relationship is derived between the scattered power spectrum and the fluctuations in plasma density. The spectral density function S(k, ω) is introduced.

CHAPTER 3

The general scattered spectrum for an unmagnetized quasi-equilibrium plasma is derived. The effect of collisions on the result is determined.

CHAPTER 4

A derivation is given of the noncollective spectrum for a plasma, including a steady magnetic field. The application of the results is discussed.

CHAPTER 5

The results of the general scattered spectrum (Chapter 3) in the collective regime are analyzed, and their experimental applications are discussed. These are illustrated by reference to some of the significant experimental work in the field.

CHAPTER 6

The constraints and problems that arise in the application of scattering as a diagnostic technique are discussed.

CHAPTER 7

The characteristic performance of various dispersion elements, image dissectors, and detectors is reviewed.

CHAPTER 8

Some interesting applications of scattering are discussed to illustrate the use of Thomson scattering.

CHAPTER 9

Industrial plasmas, scattering from energetic ions, and fusion plasmas are discussed.

CHAPTER 10

A derivation is given of the general scattering spectrum for a magnetized plasma, and the application of the results is discussed.

CHAPTER 11

The use of hard x-rays (high energy photons) is discussed for probing warmdense matter and dense plasmas. Recent applications in the Compton and plasmon scattering regime are presented.

CHAPTER 12

Work on the scattering from unstable plasmas is reviewed, e.g., the scattering from enhanced fluctuations driven by plasma wave turbulence or laser-plasma interaction.

APPENDIX A

A brief review is given of relevant mathematical techniques.

APPENDIX B

The kinetic theory of plasmas is reviewed.

APPENDIX C

A derivation of the general dispersion relation for a hot, magnetized uniform plasma is presented.

APPENDIX D

A brief discussion of computational techniques used to calculate the Thomson scattering spectrum is given. A simple computational method for solving for the scattering spectrum is presented.

APPENDIX E

A historical review of work on the scattering of radiation from plasmas is given.

APPENDIX F

This appendix contains a list of physical constants and significant formulas. The latter includes a list of the various scattered spectra obtained under different approximations.

1.2 PLASMAS

A plasma is an assembly of free electrons and positive ions, which is essentially neutral. Thus, while locally there may be a charge imbalance, in the assembly as a whole, there are to a very good approximation equal number of electrons and positive ions. The term plasma was used by Langmuir (1928) as a description of the ionized state found in an arc discharge. For gases at temperatures² >1 eV (11,600 K), there are many particles in the high-energy tail of the distribution function with sufficient energy to ionize, and consequently large numbers of free charges. The ideal plasma state may be characterized by the following ordering of characteristic scale lengths:

   (1.2.1)

where rc = q²/κT is the distance at which the potential and kinetic energies are equal when two like charges q approach each other; n−1/3 is the average interparticle separation; n for simple 90° Rutherford scattering; and Lp is a representative dimension of the plasma.

.

Within the framework of this definition, there are a wide range of plasma situations apart from that found in the high-temperature gaseous state. The free electrons and holes in a solid and the free ions in a liquid (e.g., salt solution) also constitute a plasma. While it is the application of scattering to gaseous plasmas that is emphasized in this book, most of the results will of course have relevance to work in these other plasmas.

The plasma state has a wide variety of applications; flames and fluorescent lights are plasmas, and the plasma state is involved in gas lasers, magnetic hydrodynamics (MHD) generators, inertial confinement fusion (ICF), and plasma propulsion systems. Further from the earth are the plasmas of the ionosphere, magnetosphere, solar wind, sun, and stars. Plasmas play an important role in the release of fusion energy. The nuclei of the lighter elements (H, ²H, ³H, …) can combine following a collision to produce a heavier element. In this fusion, some of the binding energy is released, as for example in the reactions

   (1.2.2)

Unless the nuclei are moving with sufficient velocity to overcome their Coulomb repulsion, they will simply scatter off each other. (This particle scattering is not to be confused with the scattering of radiation from charges.) For the useful release of fusion power, it is necessary to raise a quantity of some light element to a high enough temperature that fusion can compete with particle scattering. That is, we need a high-temperature plasma of, for example, deuterium or deuterium and tritium. There are currently two main stream approaches to producing fusion plasmas in the laboratory. In magnetic fusion energy (MFE), magnetic fields are used to contain hot plasma from material walls for a long enough time that the nuclei can fuse. Work in this field indicates that a fusion reactor will probably have a plasma of deuterium and tritium of density 10¹⁴–10¹⁵ charges/cm³ at a temperature of 10–20 keV, and that this plasma will be contained by a magnetic field of 2–10 T. Inertial confinement fusion (ICF) is an approach to fusion that relies on the inertia of the fuel mass to provide confinement. Energy from a driver is delivered rapidly to a capsule that heats up and expands. The outward expansion drives an inward force that compresses the fusion material to electron densities above 10²⁵ charges/cm³ and temperatures of 10–20 keV.

At the other end of the density scale, the ionospheric plasma has densities of ~ 10⁵ charges/cm³ and electron temperatures < 1 eV. In between are a wide-range of plasma devices and plasmas that occur in nature. The range of plasmas of interest is indicated in Fig. 1.1.

Figure 1.1 The density and temperature ranges of typical plasmas in nature and in the laboratory.

One problem is common to the investigation of all these plasmas. How do we diagnose the plasma without disturbing it? Conventional diagnostic devices, such as probes for measuring electrostatic and magnetic fields, not only contaminate the plasma but are often too large for the investigation of the microscopic structure of the plasma. In addition, they are not usable in dense plasmas encountered in warm dense matter and ICF research.

It is natural to consider the use of a probing beam of electromagnetic radiation. Ideally, the intensity will be kept sufficiently low so as not to perturb the plasma. Further, in principle, we can choose radiation of a wavelength to fit the characteristic lengths of the plasma under investigation.

The oscillating electric field of the incident beam accelerates each of the charged particles in the plasma, and these charges subsequently reradiate (Fig. 1.2). The interaction is primarily concerned with the electrons because the acceleration of the ions is less by virtue of their greater mass. The spectrum of the scattered radiation depends on the electron density, electron and ion temperatures, the magnetic field, and the plasma oscillations. Before discussing this valuable diagnostic technique, we will discuss the system of units that will be used and then review some basic features of the plasma state and of the interaction of an electromagnetic wave with it.

Figure 1.2 The scattering of radiation by a free charge.

1.3 SYSTEM OF UNITS

The Gaussian system of units will be used for all the theoretical calculations. This choice was made because most of the significant work in this field has been done in this system. However, for the benefit of the reader, all the important results will be given in both Gaussian and rationalized MKS system (see Jackson, 1998, Conversion of Equations and Amounts between SI Units and Gaussian Units. Appendix 4, pp. 783–784). The approach to the definition of unit charge is used to illustrate the difference between the two systems.

1.3.1 The Gaussian System

Here, the unit charge q is defined by Coulomb’s law

   (1.3.1)

is the unit vector in the direction from charge one to charge two.

A unit charge placed a distance r12 = 1 cm from a like charge in vacuum will repel it with a force of 1 dyne. The charge on an electron is 4.8 × 10−10 statcoulomb.

The electric field E of a charge q is defined as the force per unit charge:

   (1.3.2)

This system is unrationalized, i.e., no factor 4π has been introduced into Eqn (1.3.2) to bring out the spherical symmetry of the expression.

1.3.2 The SI System

Here, the unit charge is defined from the unit current, the ampere. The ampere is the current which when flowing in each of two infinitely long parallel wire of negligible cross-sectional area separated by 1 m in vacuum causes a transverse force per unit length of 2 × 10−7 N/m to act between the wires.

With this definition, Coulomb’s law becomes

   (1.3.3)

where the constant is

in rationalized units, F is in Newtons, q1 and q2 is in coulombs, and r12 is in meters. Maxwell’s equations and the Lorentz force are given below for both systems.

1.3.3 Maxwell’s Equations and Lorentz Force on a Unit Charge q

Gaussian

   (1.3.4)

SI System

   (1.3.5)

(For all the discussions in this book, M = 0.)

1.4 CHARACTERISTIC LENGTHS AND TIMES IN A PLASMA

³

The main difference between the plasma state and other states involving macroscopically neutral but microscopically charged systems rests in the freedom of motion of the individual charges. As a consequence of this freedom, charges are able to adjust their positions to screen mutually the electric fields of each other which can result in collective effects that can be probed by techniques described in this book. The potential ϕ around a given charge q in a plasma has the form, for r > q²/κT,

   (1.4.1)

The potential falls away in the characteristic distance, the Debye length λDq,

   (1.4.2)

If a charge in a plasma is displaced, the resultant net electrostatic force is a restoring force. The charge oscillates under the action of this force with the characteristic plasma frequency:

   (1.4.3)

The electron plasma frequency

   (1.4.3)

, the characteristic thermal speed of the particular charge of mass mq.

Finally, we must add the characteristic lengths and times associated with the gyration of the charges when the plasma contains a magnetic field. The cyclotron (gyration) frequency is given by

   (1.4.4)

The electron cyclotron frequency Ωe = 1.76 × 10⁷B(G) rad/s. The cyclotron radius is

   (1.4.5a)

The electron cyclotron radius is

   (1.4.5b)

where v⊥ is the velocity of the charge perpendicular to the magnetic field.

1.5 SCATTERING OF ELECTROMAGNETIC RADIATION BY A PLASMA

The complete calculation of the scattering of radiation by a plasma would be extremely involved, and it is fortunate that for most cases of interest, we are able to decouple the various parts of the problem. This occurs partly because with a wide range of sources available to us, we can avoid the parameter regions where, for example, absorption, reflection, and multiple scattering are important.

Consider a volume V of plasma containing N electrons, charge −e, and N/Z ions of charge Ze. The behavior of this assembly of charges may be described by the Klimontovich equation for each species q (see Section B.7):

   (1.5.1)

where

is the microscopic distribution function, and Fqdrdv represents the number of charges in the range r → r + dr, v → v + dv at the time t.

This is a complicated equation, because the acceleration a involves all the interparticle forces, as well as the effect of external forces. For terrestrial plasmas, we may reasonably restrict the forces to electric and magnetic; therefore, we may use Maxwell’s equations and the Lorentz equation (1.3.4) to determine a. The charges in the volume, external charges, and the electromagnetic wave are all sources of E and B. The following results that are touched on in the next sections help us to simplify the calculation.

1. As a result of the Debye shielding, the influence of each charge is limited so that we may neglect the simultaneous interaction of three or more charges when conditions (1.2.1) are met. This is discussed in more detail in Appendix B.

2. When the incident radiation has a high frequency, ωi ωpe, Ωe, the electromagnetic wave is primarily transmitted, and the attenuation owing to scattering and absorption is small. If the scattering volume is optically thin (no significant losses in crossing it), then we may treat the interaction with each charge in the scattering volume separately. In addition, we limit the incident power so that it will not change the plasma conditions. The total scattered electric field is then obtained as the sum of the individually scattered fields.

1.6 RADIATION BY A MOVING CHARGE

This topic is elegantly discussed in numerous books (Jackson, 1998; Landau & Lifshitz, 1962; Brau, 2004), and the reader is referred to them for a detailed treatment.

We may combine Maxwell’s equations 1.3.5 to obtain

   (1.6.1)

Now, we need to determine the electric field E when J is the current density from a single charge q, i.e.,

   (1.6.2)

It must be remembered that the electric field at a distance R from the charge (see Fig. 1.3) at a time t is related to the behavior of the charge at the previous time t′, the retarded time, where

   (1.6.3)

Figure 1.3 The scattering coordinate system.

The solution of Eqn (1.6.1) (see, e.g., Jackson, 1998, p. 664) is

   (1.6.4)

where β = v/c, ŝ is the unit vector pointing from the charge to the observer, and the term n = ck/ω is the refractive index. Generally, it is assumed that n = 1; however, this is not always the case, for example, for high densities, see Chapter 11. The fields are to be evaluated at the retarded time, i.e.,

In scattering calculations, the condition RL is normally made, and in experiment, the condition is normally met. That is, the observing point P is at a large distance from the charge compared to the characteristic length L over which we observe the motion of the charge.

The first term is therefore dropped in all calculations that follow, and in the denominator of Eqn (1.6.4), we make the approximation RR. It is, however, important to note that this approximation may not be made in Eqn (1.6.3). The retarded time may be approximately written as

   (1.6.5)

where ŝ is now constant in time.

The quantity that is measured is the scattered power, and the scattered power per unit solid angle is given by

   (1.6.6)

where the Poynting vector S is defined as

Substituting from Eqn (1.6.6), we obtain

   (1.6.7)

For low-velocity charges, where |v/c1, the electric field Eqn (1.6.4) is

   (1.6.8)

The scattered power per unit solid angle is

   (1.6.9)

The variation of dPs/dΩ with direction has the characteristic donut shape illustrated in Fig. 1.4a.

Figure 1.4 .

At higher velocities, the full expression for the second term of Eqn (1.6.4) is used, and the most noticeable change is in the increase of power scattered into the direction of v (see Fig. 1.4b). The effect is understood when we realize that the scattered power must still have the symmetrical donut shape in the electron frame of reference. These high-velocity corrections have been observed in scattering from plasmas and are discussed further in Chapters 3, 4, and 5.

1.7 ACCELERATION OF A CHARGE BY AN ELECTROMAGNETIC WAVE

Radiation is scattered by a charge, because the charge is accelerated by the oscillating fields of the radiation and an accelerating charge itself radiates (see Fig. 1.2). Here, we derive the formula for the electric field scattered by a single free electron, but at the same time, we investigate briefly some of the conditions that must be met for the charge to appear to be free.

Consider the case of a plane monochromatic wave incident upon a charge q, whose position and velocity are r(t′) and v(t′). For the sake of consistency, we write these as functions of the retarded time. The incident wave is

   (1.7.1)

ki = 2π/λi, λi, and ωi are the incident wave number, wavelength, and frequency, respectively. Note that Ei0 is not time dependent.

1.7.1 Low-Velocity Charge (v/c 1) with No Other Forces Acting

For a charge with low velocity, we may neglect the effect of the field Bi, and when no other forces act, the equation of motion is

   (1.7.2)

In determining the orbit of the charge, we neglect the influence of the electromagnetic wave, and the unperturbed orbit is simply

   (1.7.3)

We substitute from Eqn (1.7.3) in Eqn (1.6.5) and obtain

   (1.7.4)

With the aid of Eqns (1.7.3) and (1.7.4), we see that

   (1.7.5)

Substitution of Eqns (1.7.5) and (1.7.2) in Eqn (1.6.8) now yields the scattered electric field

   (1.7.6)

that is, the charge radiates the Doppler-shifted electromagnetic wave whose frequency and wave vector are

   (1.7.7)

It is convenient to work in terms of the shift in frequency ω and the shift in wave number k. From Eqn (1.7.7) (see Fig. 1.5),

   (1.7.8)

   (1.7.9)

Figure 1.5 Wave vector diagram for the scattering of radiation.

Equations (1.7.8) and (1.7.9) are statements of the conservation of energy and momentum; they are valid in the long wavelength regions where Compton scattering may be neglected.⁴

The scattering angle is denoted by θ, and it can be seen from Fig. 1.5 that

   (1.7.10)

For a charge for which v/c 1, this reduces to

   (1.7.11)

The scattered power in the unit solid angle dΩ from (1.7.6) and (1.6.7) is

   (1.7.12)

It is clear from the mass dependence that we may neglect the scattering from the positive ions in comparison to that from the electrons.

For an electron, the time-average scattered power in the solid angle dΩ is

   (1.7.13)

The over-bar denotes a time average. The classical electron radius is ro = e²/mec² = 2.82 × 10−13 cm. If the radiation is polarized (see Fig. 1.6),

   (1.7.14)

Figure 1.6 Scattering geometry, showing the relative orientation of the incident polarization ( E vector) and the observer.

If it is not polarized, we may average over ϕo and

   (1.7.15)

The total scattering cross section, the ratio of the total scattered power to the incident power, is called the Thomson cross section and is given by

For high frequencies, the Compton effect is important, and scattering must be treated quantum mechanically. The asymptotic forms of the total cross section are given by Klein–Nishina formulas,

Quantum effects will not be considered in most of this book because hvi mec²and for high-velocity charges, the Doppler shift can be large and v/c → 1. However, the effects can be important in warm-dense matter plasmas, as discussed in Chapter 11.

1.7.2 High-Velocity Charge with No Other Forces Acting

The magnetic field Bi of the incident wave may not be neglected, and the equation of motion has the form

   (1.7.16)

or

where γ = [1 − (v²/c²)]−1/2. The scalar product with v is taken and after suitable rearrangement

   (1.7.17)

Compare this with Eqn (1.7.2). The additional terms contribute scattered radiation polarized in a plane different from the plane of polarization of the incident wave. These high-velocity effects are discussed in more detail in Chapters 3, 4, and 5.

In summary, the electric field of a wave scattered by an electron is given by

   (1.7.18)

which can be expanded as

   (1.7.19)

Note: Normally r(t′) is taken as the solution of Eqn (1.7.16) where the interaction with the incident wave is neglected. Also, the refractive index n is not always unity.

In the case that is normally considered, the incident wave is polarized with its electric field Eio perpendicular to the scattering plane; consequently Eio⋅ŝ . Then, if a polarizer is used to select Es parallel to ê, Eqn 1.7.19 reduces to

   (1.7.20)

.

1.7.3 Low-Velocity Charge (v/c 1) with Additional Forces Acting

It is common practice to neglect all other forces when computing the scattering of radiation from a plasma. The example below is given as a warning that the acceleration is not always given by Eqn (1.7.16) and the effects of all forces must be carefully weighed. In general, the equation of motion of a charge is

where Eo and Bo are additional applied fields (they could be externally applied or result from the interaction of the charges in the plasma) and F represents any other forces acting. Consider the case where Boz is a steady externally applied field, and F is the result of collisions:

where v is the collision frequency for momentum transfer. Let v/c 1. The solution (see Tanenbaum, 1967, p. 70) is

   (1.7.21)

where

   (1.7.22)

It can be seen that the magnetic field Boz may be neglected in the mobility tensor provided Ω = qBoz/mec ωi. Collisions may be neglected if ωi v. However, the effect of the magnetic field and of collisions may not be ignored in the exponent, for they modify the orbit of each charge. We see, for example, that the scattered radiation received by the observer will be coherent only for the time between successive collisions.

In the calculations that follow, these effects will be retained where appropriate in the determination of the orbits of the charges but will not be considered in the calculation of the mobility tensor |W|.

1.7.4 High-Velocity Charge with Applied Magnetic Fields

In the situation in which there is an applied magnetic field Bo,