Contents

Preface to the Reissue of the Materials Characterization Series ix

Preface to Series x
Preface to the Reissue of Characterization of Optical Materials xi
Preface xii
Contributors xiv

INTRODUCTION 1

PART I: INFLUENCE OF SURFACE MORPHOLOGY AND

MICROSTRUCTURE ON OPTICAL RESPONSE

CHARACTERIZATION OF SURFACE ROUGHNESS

1.1 Introduction 9
1.2 What Surface Roughness Is 10
1.3 How Surface Roughness Affects Optical Measurements 14
1.4 How Surface Roughness and Scattering Are Measured 14
1.5 Characterization of Selected Surfaces 20
1.6 Future Directions 23

CHARACTERIZATION OF THE NEAR-SURFACE REGION

USING POLARIZATION-SENSITIVE OPTICAL TECHNIQUES
2.1 Introduction 27
2.2 Ellipsometry 29
Experimental Implementations of Ellipsometry 29, Analysis of
Ellipsometry Data 32
2.3 Microstructural Determinations from Ellipsometry Data 34
Temperature Dependence of the Optical Properties of Silicon 34,
Determination of the Optical Functions of Glasses Using SE 35,
Spectroscopic Ellipsometry Studies of SiO2/Si 37, Spectroscopic
Ellipsometry for Complicated Film Structures 38, Time-Resolved
Ellipsometry 40, Single-Wavelength Real-Time Monitoring of Film
Growth 41, Multiple-Wavelength Real-Time Monitoring of Film
Growth 42, Infrared Ellipsometry Studies of Film Growth 44

imo-exarhos-00fm.indd v 12/24/09 11:19 AM

 THE COMPOSITION, STOICHIOMETRY, AND RELATED
MICROSTRUCTURE OF OPTICAL MATERIALS
3.1 Introduction 49
3.2 Aspects of Raman Scattering 50
3.3 III–V Semiconductor Systems 51
3.4 Group IV Materials 56
3.5 Amorphous and Microcrystalline Semiconductors 59
Chalcogenide Glasses 60, Group IV Microcrystalline Semiconductors 63
3.6 Summary 66

DIAMOND AS AN OPTICAL MATERIAL

4.1 Introduction 71
4.2 Deposition Methods 72
4.3 Optical Properties of CVD Diamond 74
4.4 Defects in CVD Diamond 76
4.5 Polishing CVD Diamond 79
4.6 X-ray Window 80
4.7 Summary 81

PART II: STABILITY AND MODIFICATION OF FILM

AND SURFACE OPTICAL PROPERTIES

MULTILAYER OPTICAL COATINGS

5.1 Introduction 87
5.2 Single-Layer Optical Coatings 89
Optical Constants 90, Composition Measurement Techniques 91
5.3 Multilayer Optical Coatings 106
Compositional Analysis 107, Surface Analytical Techniques 108,
Microstructural Analysis of Multilayer Optical Coatings 109
5.4 Stability of Multilayer Optical Coatings 111
5.5 Future Compositional and
Microstructural Analytical Techniques 113

CHARACTERIZATION AND CONTROL OF STRESS

IN OPTICAL FILMS
6.1 Introduction 117
6.2 Origins of Stress 119

vi Contents

imo-exarhos-00fm.indd vi 12/24/09 11:19 AM

 6.3 Techniques for Modifying or Controlling
Film Stress 124
Effect of Deposition Parameters 124, Effect of Ion-Assisted
Deposition 127, Effect of Impurities 127, Effect of Post
Deposition Annealing 128
6.4 Stress Measurement Techniques 130
Substrate Deformation 130, X-Ray Diffraction (XRD) 133,
Raman Spectroscopy 134
6.5 Future Directions 136

SURFACE MODIFICATION OF OPTICAL MATERIALS

7.1 Introduction 141
7.2 Fundamental Processes 142
Ion–Solid Interactions 142, Defect Production, Rearrangement,
and Retention 143
7.3 Ion Implantation of Some Optical Materials 145
Glasses and Amorphous Silica 145, α-Quartz (SiO2 ) 147,
Halides 148, Sapphire (α-Al2O3) 149, LiNbO3 152,
Preparation of Optical Components by Ion Implantation 153

LASER-INDUCED DAMAGE TO OPTICAL MATERIALS

8.1 Introduction 157
8.2 Laser Damage Definition and Statistics 158
Defining Damage 158, Collecting Damage Statistical Data 159,
Types of Damage Probability Distributions 160, Identification of
Pre-Damage Sites 160, Changing the Damage Threshold 161
8.3 In Situ Diagnostics 165
Photothermal Techniques 165, Particle Emission 168
8.4 Postmortem Diagnostics 170
Surface Charge State 170, Surface Phase and Structure Analysis 171
8.5 Future Directions 174

APPENDIX: TECHNIQUE SUMMARIES

1 Auger Electron Spectroscopy (AES) 181
2 Cathodoluminescence (CL) 182
3 Electron Energy-Loss Spectroscopy
in the Transmission Electron Microscope (EELS) 183
4 Energy-Dispersive X-Ray Spectroscopy (EDS) 184
5 Fourier Transform Infrared Spectroscopy (FTIR) 185
6 Light Microscopy 186

Contents vii

imo-exarhos-00fm.indd vii 12/24/09 11:19 AM

 7 Modulation Spectroscopy 187
8 Nuclear Reaction Analysis (NRA) 188
9 Optical Scatterometry 189
10 Photoluminescence (PL) 190
11 Photothermal Displacement Technique 191
12 Raman Spectroscopy 193
13 Rutherford Backscattering Spectrometry (RBS) 194
14 Scanning Electron Microscopy (SEM) 195
15 Scanning Transmission
Electron Microscopy (STEM) 196
16 Scanning Tunneling Microscopy
and Scanning Force Microscopy (STM and SFM) 197
17 Static Secondary Ion
Mass Spectrometry (Static SIMS) 198
18 Surface Roughness: Measurement, Formation by
Sputtering, Impact on Depth Profiling 199
19 Total Internal Reflection Microscopy 200
20 Transmission Electron Microscopy (TEM) 202
21 Variable-Angle Spectroscopic Ellipsometry (VASE) 203
22 X-Ray Diffraction (XRD) 204
23 X-Ray Fluorescence (XRF) 205
24 X-Ray Photoelectron Spectroscopy (XPS) 206
Index 207

viii Contents

imo-exarhos-00fm.indd viii 12/24/09 11:19 AM

 Preface to the Reissue of the Materials
Characterization Series

The 11 volumes in the Materials Characterization Series were originally published

between 1993 and 1996. They were intended to be complemented by the Encyclope-
dia of Materials Characterization, which provided a description of the analytical tech-
niques most widely referred to in the individual volumes of the series. The individual
materials characterization volumes are no longer in print, so we are reissuing them
under this new imprint.
The idea of approaching materials characterization from the material user’s
perspective rather than the analytical expert’s perspective still has great value, and
though there have been advances in the materials discussed in each volume, the basic
issues involved in their characterization have remained largely the same. The intent
with this reissue is, first, to make the original information available once more, and
then to gradually update each volume, releasing the changes as they occur by on-line
subscription.

C. R. Brundle and C. A. Evans, October 2009

ix

imo-exarhos-00fm.indd ix 12/24/09 11:19 AM

 Preface to Series

This Materials Characterization Series attempts to address the needs of the practical
materials user, with an emphasis on the newer areas of surface, interface, and thin
film microcharacterization. The Series is composed of the leading volume, Encyclope-
dia of Materials Characterization, and a set of about 10 subsequent volumes concen-
trating on characterization of individual materials classes.
In the Encyclopedia, 50 brief articles (each 10 to 18 pages in length) are presented
in a standard format designed for ease of reader access, with straightforward tech-
nique descriptions and examples of their practical use. In addition to the articles,
there are one-page summaries for every technique, introductory summaries to group-
ings of related techniques, a complete glossary of acronyms, and a tabular compari-
son of the major features of all 50 techniques.
The 10 volumes in the Series on characterization of particular materials classes
include volumes on silicon processing, metals and alloys, catalytic materials, inte-
grated circuit packaging, etc. Characterization is approached from the materials user’s
point of view. Thus, in general, the format is based on properties, processing steps,
materials classification, etc., rather than on a technique. The emphasis of all volumes
is on surfaces, interfaces, and thin films, but the emphasis varies depending on the
relative importance of these areas for the materials class concerned. Appendixes in
each volume reproduce the relevant one-page summaries from the Encyclopedia and
provide longer summaries for any techniques referred to that are not covered in the
Encyclopedia.
The concept for the Series came from discussion with Marjan Bace of Manning
Publications Company. A gap exists between the way materials characterization is
often presented and the needs of a large segment of the audience—the materials user,
process engineer, manager, or student. In our experience, when, at the end of talks or
courses on analytical techniques, a question is asked on how a particular material (or
processing) characterization problem can be addressed the answer often is that the
speaker is “an expert on the technique, not the materials aspects, and does not have
experience with that particular situation.” This Series is an attempt to bridge this gap
by approaching characterization problems from the side of the materials user rather
than from that of the analytical techniques expert.
We would like to thank Marjan Bace for putting forward the original concept,
Shaun Wilson of Charles Evans and Associates and Yale Strausser of Surface Science
Laboratories for help in further defining the Series, and the Editors of all the indi-
vidual volumes for their efforts to produce practical, materials user based volumes.

C. R. Brundle and C. A. Evans, Jr.

imo-exarhos-00fm.indd x 12/24/09 11:19 AM

 Preface to the Reissue of Characterization
of Optical Materials

Though there have been advances in the materials and in the design and manufactur-
ing of integrated optical devices since this volume was originally published, the basic
principles of the techniques and protocols for analysis and characterization of optical
materials discussed here remain valid. There have, of course, been technical advances
for many of these techniques (ease of use, software, spatial resolution, and extension
of wavelengths into the UV for optical methods). Applications in the new, carbon-
based materials, such as nanotubes and graphene, have developed. After this re-issue
in a form close to the original, it is our intent that updates, covering the advances that
have occurred, will be released as downloads as they become available.

C. R. Brundle and C. A. Evans, December 2009

xi

imo-exarhos-00fm.indd xi 12/24/09 11:19 AM

 Preface

The design and manufacture of advanced optical materials have been driven by a
multidisciplinary approach from which new components and integrated optical
devices have evolved. The diversity of optical materials—including metals, poly-
mers, glasses, ceramics, semiconductors, and composites—provides a challenge to
the analyst charged with characterizing optical surfaces and interfaces and developing
associated structure–property relationships. Analysis of most optical materials must
include not only measurements of optical properties, but a determination of the fun-
damental surface and interfacial material properties as well. For example, correlating
the optical response of a material with its microstructure, residual interfacial stress,
phase purity, and surface-roughness can lead to a refinement of processing methods
in order to secure the optimum material for a particular application.
This volume—one component of the Materials Characterization Series: Surfaces,
Interfaces, Thin Films—focuses on the kind of information derived from the princi-
pal analytical methods currently used to characterize optical materials. This infor-
mation is useful for identifying the key parameters that control the optical response
of a material. The theory and methodology of the analytical methods used for
the surface and interfacial characterization of optical materials are discussed in
the lead volume of the series, Encyclopedia of Materials Characterization, and are
summarized in the appendix of this book. The intent of this volume is not to
discuss instrumental methods in great detail, but to provide the necessary back-
ground information to permit a constructive dialogue to be initiated between the
researcher with a specific problem to address and the technical specialist skilled in
applying a particular method.
This book consists of two major sections, an introductory chapter, and an exten-
sive appendix, which summarizes the analytical methods pertinent to the character-
ization of optical materials. The book covers both crystalline and amorphous mate-
rials—with applications from the far infrared (∼1 mm wavelength) to the vacuum
ultraviolet (∼100 nm wavelength) regions of the spectrum—and presents a succinct
discussion of the type of information obtainable using various key surface character-
ization methods. The extensive list of references for each chapter may be consulted
to gain a more thorough understanding of a particular subject area.
The topics covered and organization of this volume grew from a number of
discussions with Marjan Bace of Manning Publications Company, who persuaded
me to consider editing a text on the surface and interfacial characterization of
optical materials. His direction and advice were greatly appreciated. I wish to
express my gratitude to all contributing authors, who took time from their busy
schedules to participate in writing this volume. I hope the readers will find our

xii

imo-exarhos-00fm.indd xii 12/24/09 11:19 AM

 efforts of practical value. Our Managing Editor, Lee Fitzpatrick, also of Manning
Publications Company, was instrumental in persuading the authors to submit their
chapters in a timely manner so that the production schedule could be met. Her
efforts in dealing with all involved were invaluable. Finally, I wish to acknowledge
the Pacific Northwest Laboratory, operated for the U.S. Department of Energy by
Battelle Memorial Institute under contract DE-AC06-76RLO 1830, for allowing
me the time and resources to complete this project.

Gregory J. Exarhos

Preface xiii

imo-exarhos-00fm.indd xiii 12/24/09 11:19 AM

 Contributors

Michael F. Becker Laser-Induced Damage to Optical

University of Texas-Austin Materials
Austin, TX
Jean M. Bennett Characterization of Surface
Naval Air Warfare Center Roughness
China Lake, CA
Gregory J. Exarhos Introduction
Battelle Pacific Northwest Laboratory
Richland, WA
E. N. Farabaugh Diamond As an Optical Material
NIST
Gaithersburg, MD
Albert Feldman Diamond As an Optical Material
NIST
Gaithersburg, MD
Trevor P. Humphreys The Composition, Stoichiometry,
North Carolina State University and Related Microstructure of
Raleigh, NC Optical Materials
Gerald E. Jellison, Jr. Characterization of the Near-Surface
Oak Ridge National Laboratory Region Using Polarization-Sensitive
Oak Ridge, TN Optical Techniques
Peter M. Martin Multilayer Optical Coatings
Battelle Pacific Northwest Laboratory
Richland, WA
Carl J. McHargue Surface Modification of Optical
The University of Tennessee-Knoxville Materials
Knoxville, TN
Robert J. Nemanich The Composition, Stoichiometry,
North Carolina State University and Related Microstructure of
Raleigh, NC Optical Materials

Bradley J. Pond Characterization and Control of

S. Systems Corporation Stress in Optical Films
Albuquerque, NM
L. H. Robins Diamond As an Optical Material
NIST
Gaithersburg, MD
D. Shechtman Diamond As an Optical Material
Technion
Haifa

xiv

imo-exarhos-00fm.indd xiv 12/24/09 11:19 AM

 0

Introduction
gregory j. exarhos

The development of improved optical materials requires a thorough understand-

ing of their surface and subsurface structure and associated chemistry so that opti-
cal properties can be modified in a controlled manner and the materials durability
enhanced. Advanced surface analytical methods are used to probe these materials to
discern structure–property relationships which are used for designing materials with
a specific response. This volume is intended to provide the researcher an introduction
to the principal methods used for the surface and interfacial analysis of materials used
for optical applications.
Optical materials discussed include bulk solids, thin films, and multilayer dielec-
tric coatings which find use in applications requiring reflection, refraction, absorp-
tion, emission, scattering, or the diffraction of infrared, visible, or ultraviolet light
having wavelengths from about 100 nm to 10 mm. Included in the variety of materi-
als covered by this definition are metals, glasses, polymers, semiconductors, ceramic
oxides, carbides and nitrides, and diamond. A cadre of surface analytical techniques
is required for the characterization of such a diverse collection of materials. Further-
more, methods used to analyze a specific class of optical materials may be entirely
inappropriate for a different materials class. Once an appropriate analysis method or
methods have been identified, the information derived from the measurements can
be used in several ways, such as assuring that the optical material or device meets
desired specifications and controlling interactively processing parameters during
manufacture. In a more fundamental approach, empirical models that relate struc-
tural properties of the material and chemical bonding to the intrinsic optical response
can be developed and refined. Such models provide a basis for optimizing processing
parameters to achieve a targeted optical response. The goal of this volume is to review
the most important methods for characterizing the surface and interfacial properties
of optical materials and to demonstrate for each method the kind of information
obtained and how to interpret it.
The surface and interfacial properties of an optical material can be probed at
length scales ranging from micrometers to tenths of nanometers. Microscopic
defects which control surface smoothness and coating homogeneity contribute to
the scattering of light and associated degradation of the optical response. Likewise,
structural perturbations at the atomic level can alter optical properties through

imo-exarhos-00intro.indd 1 12/23/09 8:36 AM

 modifications to the complex refractive index of the material. To characterize defects
at this level, high-resolution diffraction measurements coupled with the evaluation
of localized chemical bonding using molecular spectroscopic techniques may be
required. Such an approach can provide information regarding the structural phase
of the material, the degree of surface/interfacial stress present, and the nature of
the chemical bonding associated with a particular structure. Therefore, analytical
probes having sensitivity at increasing levels of spatial and depth resolution are
required in order to understand the physical and chemical properties that control
the associated optical response of the material. These issues are addressed in Part I
of this volume where correlations between the optical response of a material, sur-
face morphology, and associated microstructure are described.
Surface roughness is a critical parameter that can degrade the performance of
optical materials through light scattering processes. In addition, modification to the
optical response can result from rough surfaces in which chemical attack from the
ambient environment (water) is accelerated due to the increased surface area. The
assessment of surface roughness is therefore critical to the behavior and long term
stability of optical materials. Surface topology is a recurrent theme throughout this
volume beginning with its introduction and definition in the first chapter and con-
tinuing in later chapters, including those that focus on diamond coatings and laser
damage phenomena.
A powerful technique for characterizing the optical response of materials relies
on measuring the change in the properties of light reflected from a surface or inter-
face. Ellipsometry has evolved into a powerful noninvasive surface characterization
tool, and its application to studies of optical materials is reviewed in this volume.
Instruments are currently available that directly interface to vacuum chambers for
real-time in situ characterization during the deposition of optical coatings or modi-
fication of surfaces through ion irradiation. Data acquired using this technique
can be interpreted on the basis of various empirical models (Effective Medium
Approximation) designed to extract microstructural information intrinsic to the
surface under investigation. Such measurements complement the electron beam
microscopy techniques used to evaluate microstructure but have the advantages of
being nondestructive and amenable to materials analysis in real time. Ellipsometric
measurements are relatively easy to perform, but considerable effort is required to
interpret the measured data. Chapter 2 provides insight into the kinds of informa-
tion that can be extracted from ellipsometric measurements of optical surfaces and
indicates limitations of the technique. The quality of information inferred from
these optical reflection methods is strongly dependent upon the model used to
interpret the data. New algorithms have been developed relating the microstruc-
ture of a material to the optical response. One example, based upon a finite element
model of the microstructure in a thin film, appears in a recent article by Risser and
Ferris.1 The technique of using ellipsometric methods to provide more extensive
microstructural information about surfaces and interfaces will continue to mature
as the models used to interpret the data become more refined.

2 INTRODUCTION

imo-exarhos-00intro.indd 2 12/23/09 8:36 AM

 Surface morphology and microstructure act to perturb the optical response of
materials, but the nature of chemical bonding, the deviations from stoichiometry,
and the presence of impurities can influence the intrinsic optical properties to a
greater extent. Standard surface probing techniques such as Auger spectroscopy
(AES), X-ray photoemission spectroscopy (XPS), secondary ion mass spectroscopy
(SIMS), and Rutherford backscattering spectroscopy (RBS) are very useful for the
elemental identification and concentration depth profiling of thin films. However,
subtle changes in chemical bonding, which can produce large changes in the optical
response, often are difficult to quantify using these techniques, particularly when
the material under investigation is nonconducting or when bound hydrogen is an
integral component of the material.
Significant advances in the application of vibrational spectroscopy to the analysis
of surfaces and interfaces have been reported during the past several years. Infrared
absorption and inelastic light scattering (Raman spectroscopy) are two principal non-
destructive methods that have been used extensively for the characterization of opti-
cal materials. These techniques are used routinely to analyze semiconductor materials
for optoelectronic applications (Chapter 3) and multilayer dielectric films used as
filters or mirrors (Chapter 5). In many cases, the optical response of the material
to be examined can be used to enhance the sensitivity of the photon-based analytic
methods. A paper by Friedrich and Exarhos2 describing several Raman enhancement
methods for characterizing dielectric thin films should be consulted for a more com-
plete discussion of this subject.
The optical response of a material or multilayer coating can be modified either
intentionally during processing as a means to improve performance or unintention-
ally during use as a result of environmental degradation. Changes in the optical
properties can be traced to physical or chemical alteration of the surface in bulk
materials or interfacial layers in multilayer coatings. Variations in refractive index
may be achieved during processing by ion bombardment at selective energies which
serves to implant impurity species at depths proportional to the ion energy, alter
the surface stoichiometry at larger doses, or modify the surface microstructure. Part
II of this volume deals with modifications to optical materials and the correlation
of optical properties to measured perturbations in the surface and interfacial layer
characteristics.
The design and development of specialty optical coatings is a principal thrust in
the optical materials area: Multilayer dielectric stacks are used to regulate the trans-
mission and reflection properties of a surface by means of interference phenomena.
Variations in individual layer composition, which control refractive index and layer
thickness, are used to develop antireflection (AR) and high reflection (HR) mirrors,
which are used in high power laser systems and for controlling light transmission
in optical windows. In many of these applications, irreversible changes in transmit-
tance can be introduced as a result of an applied stress (mechanical, thermal, or
chemical) which alters the chemical and/or physical properties of the surface. Sur-
face analytical techniques are important for characterizing these changes in order

INTRODUCTION 3

imo-exarhos-00intro.indd 3 12/23/09 8:36 AM

 for us to understand how a coating fails. For example, sputter-deposited coatings
often are found to exhibit a columnar grain microstructure. Diffusion of water
through the intergranular channels will lead to either reversible changes in the layer
index, depending on the amount adsorbed, or irreversible changes, if the water
chemically interacts with the dielectric material. The mechanism for this type of
failure has been substantiated using a variety of surface analytical techniques. To
minimize this problem, processing parameters have been modified to produce a
fine-grained microstructure which retards diffusion of water and minimizes chemi-
cally reactive sites. Chapter 5 discusses the key methods used to characterize optical
coatings.
Materials deposited as thin films usually exhibit some degree of interfacial stress
arising in part from the difference in thermal expansion coefficient or lattice mis-
match between the film and substrate material. This interfacial stress is distinguish-
able from the inherent stress in a thin film that varies with crystallite grain size,
phase homogeneity, and the presence of nonstoichiometric phases or impurities.
The magnitude of such stresses can be influenced by changing film deposition
parameters or through physical or chemical interaction of the film with the envi-
ronment. Both the real and imaginary parts of the refractive index change with
stress, and large stresses may cause the performance of a multilayer coating to
deviate from the design parameters. Chapter 6 discusses the causes of interfacial
stress in optical thin films and techniques for characterizing the stress. Knowledge
of the degree of stress and the stress homogeneity in these coatings is necessary for
us to understand their optical response, coating–substrate adherence, and stability
during use, particularly in chemically reactive environments or under high fluence
irradiation.
The final chapter concerns the performance of optical surfaces, thin films, and
multilayer coatings exposed to high-energy pulsed-laser irradiation. The identity
and concentration of chemical impurities, surface morphology, microstructure,
and thickness and stress homogeneity are important parameters which control
laser damage thresholds in these materials. Photothermal methods for identifying
predamage sites are introduced as relatively new surface-sensitive techniques that
have significant advantages over other methods commonly used to identify likely
surface regions for damage.
The appendix summarizes the key surface analytical techniques used for the sur-
face and interfacial characterization of optical materials. Owing to the diversity of
materials which comprise this category, an arsenal of possible methods is required. In
addition to the methods summarized in the lead volume of this series, Encyclopedia of
Materials Characterization, various surface sensitive microscopy techniques designed
specifically for optical materials are discussed. These include total internal reflection
microscopy and photothermal deflection methods.
Since optical materials belong to a relatively large number of different materials
classes, the reader is encouraged to consult other volumes in the Materials Charac-
terization Series to get a different perspective on surface analytical methods applied

4 INTRODUCTION

imo-exarhos-00intro.indd 4 12/23/09 8:36 AM

 to a specific material. Particular volumes most relevant to the subjects presented
herein are Characterization of Metals and Alloys, Characterization of Ceramics, Char-
acterization of Polymers, Characterization in Silicon Processing, and Characterization in
Compound Semiconductor Processing.

References

1 S. M. Risser and K. F. Ferris. Materials Letters. 14, 99–102, 1992.

2 D. M. Friedrich and G. J. Exarhos. Thin Solid Films. 154, 257–270, 1987.

INTRODUCTION 5

imo-exarhos-00intro.indd 5 12/23/09 8:36 AM

imo-exarhos-00intro.indd 6 12/23/09 8:36 AM
 Part I

Influence of Surface Morphology and

Microstructure on Optical Response

imo-exarhos-01.indd 7 12/23/09 8:48 AM

imo-exarhos-01.indd 8 12/23/09 8:48 AM
 1

Characterization of Surface Roughness

jean m. bennett

Contents
1.1 Introduction
1.2 What Surface Roughness Is
1.3 How Surface Roughness Affects Optical Measurements
1.4 How Surface Roughness and Scattering Are Measured
1.5 Characterization of Selected Surfaces
1.6 Future Directions

1.1 Introduction

Surface roughness is becoming increasingly important as optical instruments have

much higher performance specifications, microelectronic circuits become more com-
pressed with more information packed into a smaller area, and optical and magnetic
disks have increased information storage density. Surface characterization techniques
have now advanced to sophisticated levels to meet these needs.
Surface characterization is too large a subject to be covered adequately in a
chapter of this length. Thus, only the basics will be given here, including references
where more information can be obtained. In Section 1.2 we first define what we
mean by surface roughness—what it is and also what it is not. Examples are given
of a metal reflector and a glass surface to show how the processing of a material
can affect its surface roughness. In Section 1.3 we describe how surface rough-
ness affects optical measurements. Techniques for measuring surface roughness and
scattering are given in Section 1.4. Section 1.5 contains examples of different types
of surfaces and lists appropriate techniques for characterizing them. Finally, in Sec-
tion 1.6 we note future directions for surface characterization and suggest places
where more work is needed.

imo-exarhos-01.indd 9 12/23/09 8:48 AM

 Several excellent review articles and books have been written about surface charac-
terization. The 1982 book edited by Thomas1 contains a detailed discussion of stylus
instruments and statistics appropriate for machined surfaces. The tutorial, Introduc-
tion to Surface Roughness and Scattering,2 is a simple yet comprehensive treatment of
characterization of optical surfaces, including measuring instruments, surface statis-
tics, and elementary theory. Reference 3 contains extensive references to the entire
field of surface characterization and over 100 selected, previously published articles
on the subject.
There are many excellent references to theoretical articles relating surface rough-
ness to scattering. The landmark book is The Scattering of Electromagnetic Waves from
Rough Surfaces, by Beckmann and Spizzichino.4 Stover5 has written a more recent
book on optical scattering. Other books and articles on scattering theory and mea-
surements are referenced in the anthology.3

1.2 What Surface Roughness Is

Surface roughness can take many forms. Often it consists of tiny scratches in ran-
dom directions remaining after polishing, but it can also be the grooved structure
produced by diamond-turning a metal mirror, the grain relief on a polished metal
mirror such as molybdenum, the distinct random machining marks left on a finished
metal part to give it a diffuse patina, the tiny parallel grooves on a glass surface that
has been precision-ground, or even a few large scratches or pits (digs) sometimes
caused by improper handling. Some materials such as ceramics or silicon carbide
contain voids, while others—including aluminum or beryllium—have hard inclu-
sions in their bulk; all these things can appear on the surface. Optical or magnetic
films applied to surfaces can add additional roughness. With modern deposition
techniques, optical dielectric films such as silicon dioxide, titanium dioxide, zinc
sulfide, and magnesium fluoride are generally extremely smooth and contour the
surfaces onto which they are deposited. However, silver, gold, and copper films, in
particular, tend to be slightly lumpy, adding a fraction of a nanometer roughness to
the surface. Other metal films such as aluminum, platinum, nickel, and rhodium are
smoother since they are finer-grained. Magnetic films applied to aluminum or plas-
tic surfaces and used to store information in dense arrays are generally quite rough
compared with optical films. Optically black materials that absorb or scatter light are
even rougher. These coatings have steep slopes and contain tiny holes acting as light
traps because of multiple scattering.
The processing of a material can greatly affect its surface roughness. Here we
give two examples, first a metal reflector and then a polished glass surface. For the
metal reflector, consider a piece of polycrystalline copper such as oxygen-free, high-
purity copper. This material can be cut to shape, rough finished, and then given a
matte finish with a fine grade of sandpaper or emery paper. In this case, the surface
looks slightly dull and contains a large number of scratches oriented in all direc-
tions. Alternately, the final finish can be made with a precision grinding wheel that

10 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 10 12/23/09 8:48 AM

 Figure 1.1 Nomarski micrographs of four different types of optical surfaces.

produces tiny parallel grooves. The surface looks burnished, but there are colored
bands produced by diffraction from the grooves when the piece is viewed at certain
angles. If the copper surface is optically polished on a pitch lap with a fine grade of
abrasive in a liquid slurry, a smooth, shiny surface can be produced that contains
tiny randomly oriented scratches and, sometimes, imbedded polishing compound.
After polishing, the surface can be chemically etched to remove the subsurface
damage generated by the polishing operation. Orange peel (waviness) is introduced
by preferential etching of the grains and uneven etching of the entire surface. A
grid of straight lines reflected from the surface appears wavy. Single-point diamond
turning can produce a shiny surface with a minimum of subsurface damage. On the
surface are closely spaced grooves made by the cutting diamond as well as coarser
and deeper grooves caused by vibrations between the machine, tool, and surface;
also, large grains in the material form a mottled surface structure (see Figure 1.1).
If aluminum, stainless steel, or beryllium had been considered, alloying material
in the form of tiny hard particles dispersed throughout the material produce tiny
bumps or pullouts (holes) in a polished surface. These increase the scattering level
and, of course, the measured surface roughness.
In the second example, a piece of optical glass, which could be a window, beam
splitter, filter, or witness sample, is first cut to shape by sawing or grinding. In some
cases, glass can be molded to a size slightly larger than that of the finished shape
and then fine-ground. The ground surface has a matte finish. Progressively finer
grades of abrasive in water are used when the part is ground against an iron lap
to bring the shape closer to the finished shape; the matte finish appears smoother.
After the final grinding operation, polishing begins with a pitch lap and a fine
abrasive mixed with water to make a slurry. Tiny surface asperities are polished flat
but pits remain, giving the surface a gray appearance. As more material is removed,
the pits gradually disappear, the grayness diminishes, and tiny scratches appear, ori-
ented in all directions according to the random motion of the polishing machine.

1.2 WHAT SURFACE ROUGHNESS IS 11

imo-exarhos-01.indd 11 12/23/09 8:48 AM

 To the unaided eye the surface appears shiny; the scratches can only be seen in a
special differential interference contrast or Nomarski microscope (see Figure 1.1
and Section 1.4). To produce a superpolished surface, one can perform an addi-
tional polishing operation using a pitch lap and a slurry of very fine abrasive par-
ticles in water. The final polishing operation is done with pure water, and the only
abrasive is what is imbedded in the lap. Polishing primarily occurs by a combina-
tion of mechanical and chemical actions. In the chemical part, water is the active
reagent. During polishing, the outermost surface layer (perhaps a few atomic lay-
ers) dissolves and a thin film of pure silicon dioxide is redeposited, which covers
up the tiny surface scratches. Thus, the final superpolished surface is scratch-free
and extremely smooth. It appears featureless, even in the best optical microscope.
Figure 1.1 shows micrographs of four different types of optical surface finish,
all taken with a differential interference contrast or Nomarski microscope (see Sec-
tion 1.4). The featureless surface, polished fused quartz, is the smoothest with the
lowest scatter. This type of surface finish is desired for the highest-quality optics.
The three surfaces whose micrographs are shown on the right would all look shiny,
while the scratched surface on the far left would have a grey cast caused by the
scratches and pits.
The foregoing discussion applies to what surface roughness is. What it is not
includes anything unintentionally added to a surface—fingerprints, dust, other
particulates, surface contamination in the form of pollutants in the air, or oil
films. Although each of these, in principle, can be removed from surfaces, it may
be difficult or even impossible to remove them without damaging the underlying
surface. Note that a surface can be intentionally covered with single or multiple

Figure 1.2 Dimensions of various types of surface features. The height scale
is ~1000 times smaller than the lateral scale.

12 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 12 12/23/09 8:48 AM

 dielectric or metal films. These contour the underlying surface structure, adding
tiny bumps in some cases, and may complicate the removal of unwanted surface
contamination.
Surface roughness has two main attributes: roughness heights (or depths) and lat-
eral dimensions. A scratch, for example, has a depth of a few tenths of a micrometer
and a width of a few micrometers. The tiny crystallites that make up optical films
have heights of a few tens of nanometers and lateral dimensions of a fraction of a
micrometer. Some types of film structure look like tiny pancakes on a surface. Figure
1.2 shows three types of surface features: thin films; polishing marks, grain structure,
and scratches; and waviness or orange peel. The scale for the heights is orders of mag-
nitude smaller than for the lateral dimensions. For example, for the middle drawing,
the maximum peak-to-valley amplitude could be 100 nm and the maximum lateral
dimension 1 mm = 1000 µm = 106 nm. The dimensions of surface features can range
from atomic sizes to many micrometers.
Figure 1.3 shows the lateral dimensions of different types of surface features
plotted on a logarithmic scale of surface spatial wavelengths which are separa-
tions of surface features or their lateral dimensions, as measured along the surface.
Surface structure can be divided into three general groups according to the lateral
dimensions. Surface microroughness (often called roughness) has lengths up to
approximately 1 mm and includes thin films, polishing marks, scratches, and grain
structure. Surface waviness, or mid spatial frequency roughness, has lengths from
a few millimeters to perhaps 1 cm. Chemically polished surfaces such as those on
silicon wafers exhibit mid spatial frequency roughness, commonly called orange
peel. The overall surface shape, often called optical figure (departure from a perfect
surface of the desired shape) or form in the machining industry, has lengths from

Figure 1.3 Graphs of the lateral dimensions of different types of surface

roughness.

1.2 WHAT SURFACE ROUGHNESS IS 13

imo-exarhos-01.indd 13 12/23/09 8:48 AM

 the centimeter range to the size of the piece, as indicated in Figure 1.3. Of course,
some objects such as tiny lenses are smaller than 1 cm, so their figure and waviness
will overlap. In this chapter, the emphasis is primarily on roughness rather than
waviness or figure because we are mainly interested in the interaction between a
rough surface and a light beam.

1.3 How Surface Roughness Affects Optical Measurements

All surface roughness scatters light. The character of the roughness—heights and
spatial wavelengths—determines the intensity of the scattering and its angular dis-
tribution. As an example, an aluminum-coated glass mirror normally scatters ∼0.1%
of the light that it reflects. In other words, the ratio between the total scattered light
and the total reflected light (specular reflectance plus scattering) is 0.1%. This surface
would have a roughness of ∼1.6 nm rms.
If we use this roughness as a ballpark figure, the total reflectance at normal
incidence of an opaque mirror is reduced by about 0.1%. If light is reflected at
nonnormal incidence, the specular reflectance is higher and the scattering losses
lower. The transmission of a transparent piece of glass is also reduced by scattering
from both surfaces.
If intensities are being measured, surface roughness scatters light out of the beam.
However, if polarization properties are being measured, the effect of the surface rough-
ness on a beam of plane polarized, circularly polarized, or elliptically polarized light
at nonnormal incidence as in an ellipsometer or polarimeter6 is negligible if the
amount of surface roughness is small (i.e., the surface appears shiny).7, 8 If the sur-
face is so rough that multiple scattering occurs, the ellipsometric and polarimetric
parameters are affected.9
Transmission measurements that depend on measuring an angle of deviation,
as, for example, when determining the refractive index of a glass by measuring the
angle of minimum deviation, are unaffected by surface roughness as long as it is
small enough not to distort the slit image being observed. Striae, bubbles, and other
types of inhomogeneities in the bulk of the glass are more serious problems in this
type of measurement.
In waveguides where light is traveling within an optical fiber, surface roughness
greatly affects the amount of light unintentionally coupled out of the fiber. In fact,
this principle is used in the total internal reflection microscopy (TIRM)10 technique
to observe surface imperfections (see Section 1.5).

1.4 How Surface Roughness and Scattering Are Measured

A wide variety of methods are available for measuring surface roughness and the
light scattering the roughness produces.2 These methods can be grouped into those
that give pictures of surfaces (and sometimes quantitative information) and those
that yield quantitative statistical information about the surfaces. Figure 1.4 shows

14 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 14 12/23/09 8:48 AM

 Figure 1.4 Techniques for measuring surface roughness in various spatial
wavelength regions.

a diagram of measurement techniques that are suitable for various types of rough
surfaces. This figure is similar to Figure 1.3 except that the measurement techniques
have been added to the types of surface roughness. The various techniques are dis-
cussed in this section.
Among the methods that give pictures of surfaces are microscopes ranging from
optical microscopes, scanning electron microscopes, and transmission electron
microscopes to scanning probe microscopes. These latter instruments can be used
to produce topographic maps of surfaces on an atomic scale, both laterally and ver-
tically.11, 12 If the surface is conducting, a scanning tunneling microscope (STM)
can be used to produce a topographic map of the electron density. Most optical
surfaces are nonconducting, and maps and profiles of the surface topography can
be obtained by using an atomic force microscope (AFM). Some of the scanning-
probe microscopes are now commercially available (from Digital Instruments, Inc.,
Park Scientific Instruments, and WYKO Corporation). The transmission elec-
tron microscope (TEM) and a newer scanning transmission electron microscope
(STEM) are excellent for giving pictures of thin film surfaces or their cross sec-
tions. However, a surface replica must be made and shadowed at a suitable oblique
angle in order to show the roughness structure on a film surface. Cross sectioning a
film is even more difficult. The scanning electron microscope (SEM) requires steep
surface slopes to produce an image with good contrast, and much effort and exper-
tise are required to show the shallow but distinct structure on a diamond-turned
metal surface.
A differential interference contrast or Nomarski (light) microscope2, 13–15 is far
superior to an SEM for observing roughness structure on smooth surfaces using

1.4 HOW SURFACE ROUGHNESS AND SCATTERING ARE MEASURED 15

imo-exarhos-01.indd 15 12/23/09 8:48 AM

 magnifications ranging from 100× to 1000×. The lower magnifications are gener-
ally better because surface slopes are larger and the contrast in the image is bet-
ter. Nomarski micrographs of four different types of optical surfaces are shown in
Figure 1.1.
For smooth surfaces, two kinds of surface characterization methods give quan-
titative information about the surface roughness: optical and mechanical profilers
that take profiles along a line or make topographic maps of an area, and light-
scattering methods that can give information about surface statistical properties
such as the rms roughness or power spectrum but no surface topography. Some of
the general types of optical and mechanical profilers that are commercially avail-
able are listed in Tables 1.1; they have been described in detail in Reference 2. Here
we mainly comment on their similarities and differences. The obvious advantage
of any optical profiler is that, since it is noncontact, it cannot damage the surface.
All commercially available optical profilers are user-friendly, measurements can be
taken easily and rapidly, and data reduction is rapid. The height sensitivities of
the interferometer-based instruments are in the subnanometer range. The main
disadvantages of the optical profilers are that (1) their lateral resolutions are limited
by the properties of the optical systems and by the light beams illuminating the

Quantity Reference
Instrument Type Principle Measured Surface
Bauer 100 Optical Reflection of two Curvature No
noncontact light beams

Chapman MP 2000 Optical Differential Slope No

noncontact interference
contrast

WykoTOPO-2D Optical Mirau Phase of inter- Yes

noncontact interferometer ference fringe

Zygo Maxim-3D Optical Fizeau or Mirau Phase of inter- Yes

noncontact interferometer ference fringe

Continental Optical Optical Pencil-beam Slope No

Corp. Long Trace noncontact interferometer
Profiler

UBM UB 16 Optical Autofocus of CD Mechanical No

noncontact optical head displacement

Rank Taylor Hobson Mechanical Diamond stylus Mechanical No

Talystep contact displacement

Table 1.1 Commercially available optical and mechanical profile

16 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 16 12/23/09 8:48 AM

 surfaces and (2) the maximum step height that can be measured is less than half
of the incident wavelength. The 1–2-µm lateral resolution is not sufficient to
resolve fine scratches and other tiny surface details. As a consequence, roughness
values obtained using optical profilers tend to be smaller than those measured with
mechanical probe-type instruments on the same surfaces. Figure 1.5 illustrates this
situation schematically. The wiggly line is a profile taken with a mechanical profiler
using a 1-µm radius stylus, giving it submicrometer lateral resolution.16 The bar
graph and shading represent values that would be measured on the same surface
using an optical noncontact profiler that had 2.5-µm lateral resolution. Note that
the rms roughness value was decreased from 0.47 to 0.32 nm when going to the
optical profiler. The height-measuring range, lateral resolution, and profile length
of several commercially available profilers are shown in Figure 1.6.
Many other types of optical and mechanical profilers are described in the lit-
erature; about 30 different optical profilers are mentioned in References 2 and 3.
Standards for calibrating optical and mechanical profilers are being developed at the
National Institute for Standards and Technology (contact T. V. Vorburger) and at
the National Physical Laboratory (contact A. Franks at the Division of Mechanical
and Optical Metrology). These consist of surfaces having sine wave, square wave, or
random profiles, or some combination of these, varying in amplitude and surface
spatial wavelength.
Total integrated scattering (TIS) and angle-resolved scattering (ARS) can be used
to make maps of scattering as a function of position on a surface and to give the

Figure 1.5 Illustration of the averaging of a surface profile by an optical

profiling instrument. The 0.47-nm rms original roughness was
changed to 0.32-nm rms when the profile was averaged in 2.5-µm
segments. (From Reference 2.)

1.4 HOW SURFACE ROUGHNESS AND SCATTERING ARE MEASURED 17

imo-exarhos-01.indd 17 12/23/09 8:48 AM

 Figure 1.6 Characteristics of surface profilers and light scatter measuring in-
struments. The horizontal bars give surface spatial wavelength
ranges; numbers in parentheses are the range of rms heights that
can be measured with each instrument. For the TIS instruments,
the upper and lower bars are for wavelengths of 0.6328 and 10.6
µm, respectively. (Adapted from Reference 2.)

rms roughness, power spectral density function, and other statistical properties
of surfaces. The relation between scattering and the various statistical quantities
depends on an assumption about the form of the surface roughness. If it consists
of tiny polishing marks that are uniformly distributed over an isotropic surface, the
roughness heights are small compared to the wavelength, and the lateral dimensions
of the surface roughness are much larger than the wavelength, then scalar and vector
scattering theories provide relations between the scattering and surface statistics.
The TIS from an opaque surface can be calculated from a simple relation obtained
from scalar scattering theory. For light incident normally on the surface, the expres-
sion is17, 18
TIS ≈ (4πδ/λ)2 (1.1)
where δ is the rms roughness and λ is the wavelength. This equation has been shown
to predict correctly the scattering from an aluminized polished glass sample having
a roughness of about 3 nm rms in the wavelength range from ∼0.4 to 1.0 µm.17 At
present, there are no commercially available instruments for measuring TIS.
The expression for angle-resolved scattering derived from vector scattering theory
is more complicated19–21 and includes wavelength scaling (proportional to λ–4),
cosine of the angle of incidence, cosine squared of the scattering angle, functions

18 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 18 12/23/09 8:48 AM

 of the optical constants of the surface and polarization of the incident and scattered
light, and the surface power spectral density function. All quantities in the expression
can be calculated from known parameters with the exception of the surface power
spectral density function, which must be measured for the particular surface. Com-
mercially available instruments (such as those from TMA Technologies) can be used
to measure angle-resolved scattering at a variety of laser wavelengths in the viable and
infrared spectral regions.
Many instruments for measuring TIS and ARS have been described,2, 3 and
there are over 100 articles reporting on TIS and ARS measurements in the open
literature.2, 3
Unfortunately, the wavelength scaling of the scattered light for most surfaces does
not obey either Equation 1.1 or the ARS relations at shorter wavelengths in the
ultraviolet 22, 23 and longer wavelengths in the infrared. More theoretical and experi-
mental studies are required in order for us to understand the dominant scattering
mechanisms.
The TIS and ARS scatter measuring instruments are suitable for measuring sur-
face features covering the entire surface such as microirregularities remaining from
the polishing process. Isolated scratches and digs are not identified by these tech-
niques. There is an official U.S. military specification MIL-0-13830A24 for assessing
the sizes of scratches by visually comparing them to “standard scratches” ruled on
glass blanks. It has been shown by Young25 and Johnson26 that near-angle scatter-
ing from scratches of various widths can be duplicated by scattering from a group of
closely spaced parallel grooves. This provides the possibility of having an objective
reference with which to compare scattering from scratches on optical surfaces. There
is also a commercial instrument (the image comparator microscope manufactured by
Sira in England) that can be used to identify scratches by measuring the amount of
light they scatter out of the collecting optics of the instrument.27 Officially, though,
scratches on optics in the U.S. must still be visually inspected according to the mili-
tary specification.
Many machined surfaces have roughnesses whose heights are not small compared
to the wavelength of the illuminating laser beam, generally a He–Ne laser. However,
it is sometimes possible to use near-angle scattering to compare similar surfaces, for
example, for quality control purposes, by calibrating the instrument with a surface
whose roughness has been measured by a different instrument such as a mechanical
profiler. A commercial instrument of this type is available.28
As mentioned previously, the interferometer-based optical profilers are limited
to step heights of half of the incident wavelength. For rougher surfaces, there are
optical stylus profilers (such as the UB 16 Precision Optical Length Measurement
System manufactured by UBM-USA and the RM 600 Laser Stylus manufactured
by Optische Werke G. Rodenstock) based on the auto-focusing system of the com-
pact-disk optical head with a height sensitivity of ∼0.05 µm, a lateral resolution
of ∼1 µm, and a dynamic range of up to 500 µm. These instruments provide a
convenient extension of optical profilers into the rough surface domain of, for

1.4 HOW SURFACE ROUGHNESS AND SCATTERING ARE MEASURED 19

imo-exarhos-01.indd 19 12/23/09 8:48 AM

 example, machined surfaces and paper surfaces, provided that the local slopes are
less than a few degrees. A recent study29 indicates that care must be taken in inter-
preting microroughness values measured on bulk scattering materials using auto-
focus instruments.
Confocal microscopy is another alternative for producing topographic maps of
surfaces whose heights are large compared to the focal range (depth of focus) of
a microscope objective, that is, > ∼1 µm. A tiny area on a surface is imaged on a
detector masked by a pinhole. If the area is in focus, the detector will have maxi-
mum signal output. If the area is out of focus, the detector signal will be lower. Two
options are possible: (1) to move the imaging lens and detector until the signal is
maximized and note the distance the lens was moved or (2) to calibrate the reduced
signal level to correspond to the amount of defocus of the surface. Confocal optical
microscopes have been used by the group at the University of Oxford for measuring
surface topography whose heights are in the micrometer regime.30–33 Other optical
systems that act like pseudo-confocal microscopes and measure focus differences cor-
responding to surface height variations have been described by Sawatari and Zipin34
and by Fainman et al.35
Surfaces that have steep-sloped roughness where multiple scattering can occur pro-
duce humps in the angle-resolved scattering curves at certain angles.36–38 These same
surfaces can enhance light scattered directly back into the incident beam by a factor
of two, giving the so-called “opposition effect.”39 Theory predicts that enhancement
in the retroscattered light can also occur when the surface roughness is much smaller
and there are no multiple reflections.40
Although roughness values can be measured by a variety of techniques, care should
be taken when comparing measurements made with different instruments since the
surface spatial wavelength ranges may be different. Church has done extensive work
in this area, starting in 1975, and has published nearly 50 papers on the subject.
A bibliography of his papers is included in a recently published book of reprints.3
Two of Church’s recent papers with co-workers compare optical and mechanical
measurements of surface finish41 and discuss instrumental effects in surface finish
measurement.42

1.5 Characterization of Selected Surfaces

Tables 1.2 shows examples of six surfaces or groups of surfaces along with suggested
ways of characterizing them. All of the materials in the first group (aluminum-
coated glass mirror, diamond-turned copper mirror, and polished molybdenum
mirror) are opaque and highly reflecting; they can all be characterized by using
the same techniques, which have been discussed above. The surfaces can first be
inspected in a Nomarski microscope to see whether they contain isolated defects
that would affect the measurements or whether they require cleaning. If the dia-
mond-turned copper mirror was turned on-center, it would have a small defect at
that point that would affect profile or scattering measurements. The surfaces can be

20 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 20 12/23/09 8:48 AM

 Material Characterization Technique

Aluminum-coated glass mirror Nomarski microscope for inspection or photograph

Diamond-turned copper mirror Optical or mechanical profiler for surface profile
Polished molybdenum mirror Optical profiler and/or STM for topographic map
TIS and ARS for scattering
Interferometer for optical figure, shape, or form
Polished glass substrate Nomarski microscope for inspection
Antireflection-coated window Optical or mechanical profiler for surface profile
Optical profiler and/or AFM for topographic map
Interferometer for optical figure
Optical disk Nomarski microscope (high magnification) for
inspection or photograph
Optical profiler (high magnification) or mechanical
profiler (special stylus) for surface profile
Scanning electron microscope? for photograph
Optical profiler and/or AFM for topographic map
Magnetic disk Nomarski microscope for inspection (surface
cleanliness) or photograph
Long-scan optical or mechanical profiler
TIS for uniformity check
Optical black baffle Nomarski microscope for inspection (surface
uniformity)
Infrared TIS or ARS for uniformity
Reflectance versus wavelength for efficiency
Machined aluminum surface Nomarski microscope for inspection (surface
uniformity and defects)
Optical or mechanical profiler for surface profile and
surface statistics
Infrared TIS or ARS for uniformity and statistics

Tables 1.2 Suggested methods for characterizing different rough surfaces.

profiled with an optical noncontact profiler or a mechanical contact profiler to

determine their rms roughness. If the aluminized glass mirror contains a network
of tiny scratches remaining from the polishing process, the roughness value mea-
sured by the optical profiler will be smaller than that measured by the mechanical
contact profiler; the latter can resolve tiny scratches with a sharp stylus (∼1 µm
radius and ∼1 mg loading). One must be careful not to damage a soft surface
(aluminum or copper) by using too large a loading. Molybdenum is harder, so
scratching during mechanical profiling is not so much of a problem. Topographic
maps can be made by using some types of optical profilers. If much higher lateral

1.5 CHARACTERIZATION OF SELECTED SURFACES 21

imo-exarhos-01.indd 21 12/23/09 8:48 AM

 resolution is desired, the STM or AFM will give excellent topographic maps of
small areas of the surface from fractions of a micrometer to ∼100 µm on a side. The
STM can only be used on conducting materials, so aluminum with its ∼3 nm thick
native oxide coating or slightly oxidized copper must be profiled with an AFM.
TIS and ARS can give information about surface scattering; maps can be made of
the scattering variations on the sample if it is possible to translate either the sample
or the probe beam. If the surface contains only uniform, tiny scratches over the
entire surface, the scattering measurements can be converted into an effective rms
roughness (TIS, ARS), autocovariance function (ARS), and power spectral density
function (ARS).2 It is important that the theory being used to calculate surface
roughness from scattering be valid for the type of roughness that is measured on
the surface. Optical figure, shape, or form can be measured interferometrically
using well-established techniques.
The second pair of surfaces in Tables 1.2—glass substrates and antireflection
coated windows—are transparent and have low reflectances. Thus, some techniques
are inappropriate. A Nomarski microscope can show dirt, large scratches, and
water marks, but is less sensitive to tiny scratches and surface microstructure since
the reflectance is low. An alternate inspection technique, TIRM,10 is much more
sensitive but requires a laser, polarizer, coupling prism, and microscope. Profiles
of the surfaces can be made with optical or mechanical profilers, and topographic
maps can be made with some optical profilers. The AFM can also be used to map a
surface on a much finer lateral scale. Optical figure can be measured interferometri-
cally. TIS and ARS instruments are not useful since the samples are transparent
and have low reflectance.
It is difficult to characterize glass or plastic lenses, particularly small ones or those
that have small radii of curvature, because of their low reflectance, curved shape, and
focusing properties. The surfaces can be visually inspected for obvious defects such as
scratches and, if the curvature is not too strong, forward scattering can be measured
either as TIS or ARS. However, since the scattering and reflections are coming from
both surfaces and from the bulk of the material, it is not simple to determine an rms
roughness value from the scattering measurement. A mechanical profiler is a good
choice for obtaining surface roughness information because it is insensitive to the low
reflectance of the surface.
Returning to Tables 1.2, an optical disk contains tiny information pits, approxi-
mately ∼1 µm wide and 0.2 µm deep. These can be seen in a Nomarski microscope at
high magnification and can be profiled with an optical profiler at high magnification.
A topographic map can also be made with one type of optical profiler. The pits can
be profiled with a mechanical profiler using a special shovel-shaped stylus whose nar-
row dimension is ∼0.1–0.3 µm. It may be possible to use an SEM to view the pits if
the bare aluminum surface is studied before it is coated with transparent plastic. An
AFM should give a good picture of the tiny pits.
The surface of a magnetic disk is rough by optical standards since the magnetic
coating has steep-sloped grains. This type of surface can be inspected for uniformity

22 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 22 12/23/09 8:48 AM

 in the Nomarski microscope, but neither an optical profiler nor a mechanical profiler
has good enough lateral resolution to show the true shapes of the magnetic grains.
However, because a long-scan optical profiler can be used to show the surface wavi-
ness and also any isolated large grains, it may be useful for quality control. TIS is
good as a uniformity check if the entire surface can be mapped.
An optical black baffle is similar to a magnetic disk in that the black material is
rough with steep-sloped structure. In addition to using a Nomarski microscope and
infrared TIS or ARS to check for uniformity, one can measure the specular reflec-
tance throughout the visible spectral region as well as in the infrared to be sure that
it is low.
A machined aluminum surface is rough, but the surface slopes are much smaller
than those for a magnetic disk or an optical black baffle. For this reason, it is possi-
ble to use an optical or mechanical profiler to get a good surface profile and surface
statistics, and an optical profiler to obtain an area map. The Nomarski microscope
is helpful for inspecting the surface to look for defects and check for cleanliness.
TIS or ARS can be used with an infrared wavelength for obtaining surface statis-
tics and checking for uniformity. The wavelength should be much larger than the
roughness heights.

1.6 Future Directions

In the future, the STM, AFM, and other atomic probes will be used more to enable
us to better understand surface roughness on an atomic scale and relate it to light
scattering. There will be more long-scan profilers that have better lateral resolution
and can measure surface heights in the waviness regime. Measurements made by
these instruments will be used to improve the methods for making optical surfaces.
Subsurface damage and anisotropic grain structure will be studied both from theo-
retical and from experimental points of view in order for us to better understand
their effects on optical scattering and to minimize the scattering as much as possible.
Finally, more theoretical work should be done for us to better understand the lack of
wavelength scaling of the TIS and ARS into the ultraviolet and vacuum ultraviolet
and into the infrared spectral regions.

References

1 Rough Surfaces. (T. R. Thomas, Ed.) Longman, London, 1982.

2 J. M. Bennett and L. Mattsson. Introduction to Surface Roughness and Scattering.
Optical Society of America, Washington, DC, 1989.
3 J. M. Bennett. Surface Finish and Its Measurement. Optical Society of America,
Washington, DC, 1992.
4 P. Beckmann and A. Spizzichino. The Scattering of Electromagnetic Waves from
Rough Surfaces. Pergamon Press, London, 1963.

1.6 FUTURE DIRECTIONS 23

imo-exarhos-01.indd 23 12/23/09 8:48 AM

 5 J. C. Stover. Optical Scattering: Measurement and Analysis. McGraw-Hill, New
York, 1990.
6 R. M. A. Azzam and N. M. Bashara. Ellipsometry and Polarized Light. North-
Holland, New York, 1977.
7 J. R. Blanco, P. J. McMarr, and K. Vedam. “Roughness Measurements by
Spectroscopic Ellipsometry.” Appl. Opt. 24, 3773–3779, 1985.
8 J. R. Blanco and P. J. McMarr. “Roughness Measurements of Si and Al by
Variable Angle Ellipsometry.” Appl. Opt. 30, 3210–3220, 1991.
9 I. Ohlídal and F. Lukes. “Ellipsometric Parameters of Rough Surfaces and of
a System Substrate–Thin Film with Rough Boundaries.” Optica Acta. 19,
817–843, 1972.
10 P. A. Temple. “Total Internal Reflection Microscopy: A Surface Inspection
Technique.” Appl. Opt. 20, 2656–2664, 1981.
11 H. K. Wickramasinghe. “Scanned-Probe Microscopes.” Sci. Am. 261 (4),
98–105, 1989.
12 R. Pool. “The Children of the STM.” Science. 247, 634–636, 1990.
13 G. Nomarski. “Microintexrférometre différentiel à ondes polarisées.” J. Phys.
Rad. 16, 9S–13S, 1955.
14 G. Nomarski and A. R. Weil. “Application à la métallographie des méthodes
interférentielles à deux ondes polarisées.” Rev. Metall. (Paris). 52, 121–134,
1955.
15 D. L. Lessor, J. S. Hartman, and R. L. Gordon. “Quantitative Surface
Topography Determination by Nomarski Reflection Microscopy.” J. Opt.
Soc. Am. 69, 357–366, 1979.
16 J. M. Bennett and J. H. Dancy. “Stylus Profiling Instrument for Measuring
Statistical Properties of Smooth Optical Surfaces.” Appl. Opt. 20, 1785–
1801, 1981.
17 H. E. Bennett. “Scattering Characteristics of Optical Materials.” Opt. Eng.
17, 480–488, 1978.
18 J. M. Bennett and L. Mattsson. Introduction to Surface Roughness and
Scattering. Optical Society of America, Washington, DC, 1989, pp. 50–53.
19 J. M. Elson, H. E. Bennett, and J. M. Bennett. “Scattering from Optical
Surfaces.” In Applied Optics and Optical Engineering, Vol. 7. (R. R. Shannon
and J. C. Wyant, Eds.) Academic Press, New York, 1979, pp. 191–244.
20 J. M. Elson and J. M. Bennett. “Vector Scattering Theory.” Opt. Eng. 18,
116–124, 1979.
21 J. M. Bennett and L. Mattsson. Introduction to Surface Roughness and
Scattering. Optical Society of America, Washington, DC, 1989, pp. 53–55.

24 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 24 12/23/09 8:48 AM

 22 L. Mattsson, J. Ingers, and J. M. Bennett. “Wavelength Dependence of
Angle-Resolved Scattering in the EUV-Visible. I. Experimental Results.”
Appl. Opt. 1992. In press.
23 J. Ingers, L. Mattsson, and J. M. Bennett. “Wavelength Dependence of
Angle-Resolved Scattering in the EUV-Visible. II. Comparison Between
Theory and Experiment.” Appl. Opt. 1992. In press.
24 “Optical Components for Fire Control Instruments; General Specification
Governing the Manufacture, Assembly, and Inspection Of.” Military
Specification MIL-0-13830A. 11 Sept. 1963. See especially the
accompanying drawing No. 7641866, revision L, “Surface Quality Standards
for Optical Elements (Scratch).”
25 M. Young. “Objective Measurement and Characterization of Scratch
Standards.” In Scattering in Optical Materials. (S. Musikant, Ed.) Proc. Soc.
Photo-Opt. Instrum. Eng., Vol. 362, 1982, pp. 86–92.
26 E. G. Johnson, Jr. “Simulating the Scratch Standards for Optical Surfaces:
Theory.” Appl. Opt. 22, 4056–4068, 1983.
27 L. R. Baker. “Inspection of Surface Flaws by Comparator Microscopy.” Appl.
Opt. 27, 4620–4625, 1988.
28 R. Brodmann, O. Gerstorfer, and G. Thurn. “Optical Roughness Measuring
Instrument for Fine-Machined Surfaces.” Opt. Eng. 24, 408–413, 1985.
29 L. Mattsson and P. Wagberg. “Assessment of Surface Finish on Bulk
Scattering Materials. A Comparison Between Optical Laser Stylus and
Mechanical Stylus Profilometers.” Precision Eng. 1992. In press.
30 D. K. Hamilton and T. Wilson. “Surface Profile Measurement Using the
Confocal Microscope.” J. Appl. Phys. 53, 5320–5322, 1982.
31 T. Wilson, A. R. Carlini, and D. K. Hamilton. “Images of Thick Step
Objects in Confocal Scanning Microscopes by Axial Scanning.” Optik. 73,
123–126, 1986.
32 H. J. Matthews, D. K. Hamilton, and C. J. R. Sheppard. “Surface Profiling
by Phase-Locked Interferometry.” Appl. Opt. 25, 2372–2374, 1986.
33 C. J. R. Sheppard and H. J. Matthews. “The Extended-Focus, Auto-Focus
and Surface-Profiling Techniques of Confocal Microscopy.” J. Modern
Optics. 35, 145–154, 1988.
34 T. Sawatari and R. B. Zipin. “Optical Profile Transducer.” Opt. Eng. 18,
222–225, 1979.
35 Y. Fainman, E. Lenz, and J. Shamir. “Optical Profilometer: A New Method for
High Sensitivity and Wide Dynamic Range.” Appl. Opt. 21, 3200–3208, 1982.
36 K. A. O’Donnell and E. R. Mendez. “Experimental Study of Scattering from
Characterized Random Surfaces.” J. Opt. Soc. Am. A. 4, 1194–1205, 1987.

REFERENCES 25

imo-exarhos-01.indd 25 12/23/09 8:48 AM

 37 E. R. Mendez and K. A. O’Donnell. “Scattering Experiments with Smoothly
Varying Random Rough Surfaces and Their Interpretation.” In Scattering in
Volumes and Surfaces. (M. Nieto-Vesperinas and J. C. Dainty, Eds.) Elsevier,
North-Holland, Amsterdam, 1990, pp. 125–141.
38 J. C. Dainty, N. C. Bruce, and A. J. Sant. “Measurements of Light
Scattering by a Characterized Random Rough Surface.” Waves in Random
Media. 1, S29–S39, 1991.
39 Z.-H. Gu, R. S. Dummer, A. A. Maradudin, and A. R. McGurn.
“Experimental Study of the Opposition Effect in the Scattering of Light
from a Randomly Rough Metal Surface.” Appl. Opt. 28, 537–543, 1989.
40 A. R. McGurn. “Enhanced Retroreflectance Effects in the Reflection of
Light from Randomly Rough Surfaces.” Surf. Sci. Reports. 10, 357–410,
1990.
41 E. L. Church, J. C. Dainty, D. M. Gale, and P. Z. Takacs. “Comparison
of Optical and Mechanical Measurements of Surface Finish.” In Advanced
Optical Manufacturing and Testing II. (V. J. Doherty, Ed.) Proc. Soc. Photo-
Opt. Instrum. Eng., Vol. 1531, 1991, pp. 234–250.
42 E. L. Church and P. Z. Takacs. “Instrumental Effects in Surface Finish
Measurement.” In Surface Measurement and Characterization. (J. M. Bennett,
Ed.) Proc. Soc. Photo-Opt. Instrum. Eng., Vol. 1009, 1988, pp. 46–55.

26 CHARACTERIZATION OF SURFACE ROUGHNESS Chapter 1

imo-exarhos-01.indd 26 12/23/09 8:48 AM