Professional Documents
Culture Documents
OSA - Century of Optics
OSA - Century of Optics
of Optics
1 2 3
5
4
8
OSA Century of Optics
7
On The Cover 12
Left to right from top left:
17
18. Data encryption iStock.com/
Danil Melekhin
OSA Century of Optics
OSA Century of Optics
OSA History Book Committee
Copyright 2015 by The Optical Society (OSA). All rights reserved. No part of this book may be
reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, recording, or by any information storage and retrieval system without the written
permission of OSA, except where permitted by law.
ISBN: 978-1-943580-04-0
INTRODUCTION
Introduction
Paul Kelley 3
PRE1940
Introduction: Early Technology
Carlos Stroud 9
Optics in the Nineteenth Century
Jeff Hecht 11
Spectroscopy from 1916 to 1940
Patricia Daukantas 17
Government and Industrial Research Laboratories
Carlos Stroud 23
Camera History 1900 to 1940
Todd Gustavson 31
OSA and the Early Days of Vision Research
Patricia Daukantas 38
Evolution of Color Science through the Lens of OSA
Roy S. Berns 43
19411959
Introduction: Advances in Optical Science and Technology
Paul Kelley 49
Inventions and Innovations of Edwin Land
Jeff Hecht 51
Birth of Fiber-Optic Imaging and Endoscopes
Jeff Hecht 53
Xerography: an Invention That Became a Dominant Design
Mark B. Myers 57
U.S. Peacetime Strategic Reconnaissance Cameras, 19541974: Legacy of James
G. Baker and the U-2
Kevin Thompson 64
History of Optical Coatings and OSA before 1960
Angus Macleod 68
19601974
Introduction
Jeff Hecht 79
The Discovery of the Laser
Jeff Hecht 81
Table of Contents v
Postwar Employment Bubble Bursts
Jeff Hecht 85
Gas LasersThe Golden Decades, 19601980
William B. Bridges 88
Discovery of the Tunable Dye Laser
Jeff Hecht 94
Remembrances of Spectra-Physics
David Hardwick 97
The Birth of the Laser Industry: Overview
Jeff Hecht 100
Lasers at American Optical and Laser Incorporated
Bill Shiner 101
Solid-State Lasers
William Krupke and Robert Byer 103
Semiconductor Diode Lasers: Early History
Marshall I. Nathan 107
Lasers and the Growth of Nonlinear Optics
Jeff Hecht 114
Early Years of Holography
Jeff Hecht 119
History of Laser Materials Processing
David A. Belforte 124
Brief History of Barcode Scanning
Jay Eastman 128
Developing the Laser Printer
Gary Starkweather 134
History of the Optical Disc
Paul J. Wehrenberg 138
Interferometric Optical Metrology
James C. Wyant 143
Half a Century of Laser Weapons
Jeff Hecht 149
KH-9 Hexagon Spy in the Sky Reconnaissance Satellite
Phil Pressel 153
CORONA Reconnaissance Satellite
Kevin Thompson 157
Laser Isotope Enrichment
Jeff Hecht 161
Lasers for Fusion Research
John Murray 166
History of Laser Remote Sensing, Laser Radar, and Lidar
Dennis K. Killinger 175
19751990
Introduction
Michael Bass 183
vi Table of Contents
The Shift of Optics R&D Funding and Performers over the Past 100 Years
C. Martin Stickley 185
Through a Glass Brightly: Low-Loss Fibers for Optical Communications
Donald B. Keck 189
Erbium-Doped Fiber Amplier: From Flashlamps and Crystal Fibers to 10-Tb/s Communication
Michel Digonnet 195
Advent of Continuous-Wave Room-Temperature Operation of Diode Lasers
Michael Ettenberg 199
Remembering the Million Hour Laser
Richard W. Dixon 203
Terabit-per-Second Fiber Optical Communication Becomes Practical
Guifang Li 209
Applied Nonlinear Optics
G. H. C. New and J. W. Haus 213
Linear and Nonlinear Laser Spectroscopy
M. Bass and S.C. Rand 218
Optical Trapping and Manipulation of Small Particles by Laser Light Pressure
Arthur Ashkin 223
High-Power, Reliable Diode Lasers and Arrays
Dan Botez 227
Tunable Solid-State Lasers
Peter F. Moulton 232
Ultrashort-Pulse Lasers
Erich P. Ippen 237
Ground-Based Telescopes and Instruments
James Breckinridge 244
Space Telescopes for Astronomy
James Breckinridge 249
Contact Lenses for Vision Correction: A Journey from Rare to Commonplace
Ian Cox 253
Excimer Laser Surgery: Laying the Foundation for Laser Refractive Surgery
James J. Wynne 257
Intraocular Lenses: A More Permanent Alternative
Ian Cox 262
Spectacles: Past, Present, and Future
William Charman 265
Major Milestones in Liquid Crystal Display Development
Shin-Tson Wu 269
1991PRESENT
Introduction
Govind Agrawal 277
Birth and Growth of the Fiber-Optic Communications Industry
Jeff Hecht 278
Telecommunications Bubble Pumps Up the Optical Fiber Communications Conference
Jeff Hecht 282
THE FUTURE
Far Future of Fibers
Philip Russell 327
View of the Future of Light
Steven Chu 329
The 100-Year Future for Optics
Joesph H. Eberly 331
Future of Energy
Eli Yablonovitch 332
Future of Displays
Byoungho Lee 333
Biomedical OpticsThe Next 100 Years
Rox Anderson 334
Lasers and Laser Applications
Robert L. Byer 336
Optical Communications: The Next 100 Years
Alan Willner 338
INDEX 341
Introduction
Paul Kelley
T
his book describes progress in optics during the period from 1916 to 2016, the rst
hundred years of The Optical Society (OSA). Before we begin, let us consider how much
the rate of advancement has increased over this period. A sense of this can be found in the
OSA membership and publication statistics. There were 30 Charter Members in 1916, and in
1917 the membership was 74. The Society grew in the 1920s, but in the depression decade of the
1930s the membership was fairly static at about 650. The membership rose sharply with the
onset of World War II, roughly doubling by the end of the war. Government funding of science
and technology and the increased use of optics in industry stimulated further growth so that by
1960, the year of the laser, the membership stood at 2600. The development of the laser further
enhanced this growth, and by the ftieth anniversary of the Society in 1966 there were 4500
members. In the 1980s, OSA passed the 10,000 member mark, and today the organization has
19,000 members. The Society has endeavored to include all of optics. However, for a number of
reasons, including growth and divergence of interests, several subelds have left the organization.
Because of this it is hard to do justice here to some topics in optics.
While this volume does not intend to discuss progress in optics before 1916 in any depth, it is
useful to consider where the eld stood at the beginning of the period. Optics is the science and
technology of light. As such, it is concerned with the generation, manipulation, and use of light.
Light and the tools of optics are our principal means of directly sensing our world and allow us to
vastly expand our knowledge of the universe and the microscopic world. While optics has a very
long history, its inuence became particularly strong toward the end the nineteenth century. The
invention of the electric light changed the way we lived by extending our nighttime activities of
work, study, and pleasure. Eyeglasses, still and motion picture cameras, and other optical
instruments had widespread impact on our lives. The industries that provided these devices set
the stage for the founding of OSA.
The development of optical spectroscopy led in 1913 to Bohrs quantum theory of the atom.
At about the same time, Einsteins theory of blackbody radiation and the photoelectric effect gave
us an understanding of the quantization of light. The extension of quantum mechanics into
molecular physics and condensed matter physics provided the basis for much of the progress in
twentieth century physical science and technology, including the invention and development of
the laser.
At the start of OSA, principal areas of interest to OSA members included optical instru-
ments, vision, optical materials, lens technology, theoretical optics, and the photographic
process. The practical nature of most of these subjects reected the backgrounds of the founders.
In the 1920s and 1930s spectroscopic instrumentation was under rapid development. The use of
photocells with vacuum tube ampliers overcame many of the limitations of photographic
recording of spectra. New photocathode materials were developed to extend spectra ranges, and
the photomultiplier tube was invented in 1934. Silver-halide-based photographic materials were
developed with improved sensitivity and spectral range, and color photography became practical
and widespread. CCD image sensors replaced lm in the 1990s, bringing further improvement in
sensitivity and dynamic range in photography. World War II saw the development of innovative
camera lens designs for use in reconnaissance and the widespread use of antireection coatings.
During the war, infrared spectroscopy became vital in the production of articial rubber and
custom fuels. Analytical instrumentation using spectroscopy spread rapidly in the chemical
3
industry at the end of the war. This period also saw the introduction of new civilian applications of
optics such as instant photography, the Xerox copier, and the ber endoscope.
Astronomy has seen a number of innovations in the last hundred years. The Schmidt wide-eld-of-
view camera was invented in 1930, and early versions were built at Hamburg Observatory and Palomar
Observatory in the mid-1930s. The Schmidt camera and various variants are widely used in sky
surveys, and a modied version was designed to track earth satellites. As astronomical telescopes
became larger to provide greater light-gathering power and resolution, stability and weight of
monolithic reectors became serious problems. A segmented-mirror telescope design was proposed
in 1977, two versions of which have been operated at Mauna Kea since the early 1990s. Since then,
more segmented telescopes have been deployed by astronomers. Laser guide stars are being used to
correct the optical wavefront for effects of atmospheric turbulence. The Hubble telescope, which uses a
Ritchey-Chretien Cassegrain wide-eld design, has been operating in earth orbit since 1990.
One of the most important uses of light is illumination. While Edisons incandescent lamp was a
welcome replacement for gas and oil lamps, it was inefcient and not very long lasting. The
uorescent lamp was commercialized in the 1930s. The need for 24-hour production in wartime
factories led to the widespread use of uorescent lighting, and by the early 1950s it had surpassed
incandescent lighting in the United States. In order to reduce energy consumption, new uorescent
lamp congurations were designed in the 1990s to mimic the incandescent lamp. Today uorescent
lighting is being replaced by even more efcient LED lighting. First developed as a cousin of the
semiconductor laser in the 1960s, LEDs were not considered useful for illumination because of the
absence of a blue source. This problem was solved in the mid-1990s. When fully deployed, the
worldwide energy savings will be about 5 PWh/yr.
1960 began the age of the laser. The rst laser had ruby as the active medium. Other pulsed solid-
state lasers were developed that year, and in December came the HeNe laser, the rst continuously
operating system. After that, new lasers were invented at a rapid pace, including high-power gas lasers
at wavelengths from the infrared to the ultraviolet as well as continuously operating solid-state lasers.
Most lasers used optical or electrical excitation (pumping) of the active medium. Perhaps the most
signicant early (1962) invention was the semiconductor diode laser, which operated with very high
efciency through electrical excitation. After considerable development, continuous operation was
achieved at room temperature, cementing the great practical value of this system. While individual
semiconductor lasers were not particularly powerful, they were small and could be fabricated in one-
and two-dimensional arrays for use in optical pumping. Broadly tunable lasers were invented; early
ones used dyes but were supplanted by solid-state systems. The tunable laser was valuable for general
spectroscopy and is essential in ultrafast science. Diode-pumped rare-earth ber lasers have successfully
competed with gas lasers for a number of high-power industrial applications.
Because of the availability of lasers as sources of very intense light, it became possible to induce a
nonlinear response of material to radiation. Following the rst report of second harmonic generation in
1961, many nonlinear phenomena were observed, including stimulated inelastic light scattering,
parametric oscillation and amplication, and self-action (four-wave mixing) effects. Parametric
processes have been important in the understanding of entanglement and other quantum optics
phenomena. Octave frequency combs and optical solitons are a consequence of self-action. Nonlinear
frequency conversion is often used to extend the wavelength range of laser radiation.
Over the fty-plus years since 1960, the laser has seen a wide variety of applications. Military uses
include laser targeting and tracking; laser weapons have also been tested. In nuclear energy, lasers have
been built to test concepts in inertial connement fusion and for uranium isotope separation. Industrial
lasers such as CO2, diode-pumped solid-state, and diode-pumped ber lasers are used for welding,
marking, machining, and other industrial processes, representing business of greater than $2 billion
dollars per year. This is about 25% of the laser market. Applications such as ber optical communi-
cation, optical storage, photolithography, and laser printing are on a similar scale. Access to worldwide
information at very high bandwidth has changed the way people work and live in many ways. The
Internet, cable television, video on demand, cell phone networks, and many other information sources
depend on ber optical connectivity. Fabrication of microelectronic devices with feature sizes
approaching 10 nm using excimer laser lithography has led to a mass market for inexpensive, powerful
4 Introduction
computers. Sales of microprocessor-based devices approach a trillion dollars per year. In medical optics,
lasers are used in a variety of diagnostic and therapeutic applications, including refractive surgery of the
eye (LASIK) and optical coherence tomography.
While it is hard to predict the future, it is apparent that rapid progress in optical science and
technology is continuing. New ways of generating and applying ultrashort pulses are being found.
Novel ber structures and plasmonic devices are being actively studied. As nanofabrication techniques
are developed, it seems possible that a variety of sub-wavelength optical devices will be made. Such
devices would function much like electronic devices. Optics should continue to play an important role
in our understanding of the theory of entangled states and the development of quantum computing and
quantum cryptography.
Introduction 5
PRE1940 19411959 19601974 19751990 1991PRESENT
PRE-1940
T
his section of our centennial history of optics addresses two tasks: setting the stage by
describing the situation at the beginning of our highlighted period, and then summarizing
the changes that occurred. The beginning and end of our period are both quite special
years in political and economic history. The United States was just entering the Great War, as
World War I was called in 1916; and in 1940 it was on the inevitable path leading to its entry into
World War II. It is not an exaggeration to say that the course of civilization was dramatically
altered by each of these events, and the course of optical research and technology was no less
altered.
In a very real sense modern instrumental optics began in a series of developments in
Germany led by Carl Zeiss, Ernst Abbe, and Otto Schott. In his essay Jeff Hecht reviews these
and other earlier developments that formed the basis for the rapid developments in our eld in
the rst half of the twentieth century. The dawn of the new century found Germany recently
unied and growing quickly in industrial output, Great Britain at the peak of her imperial era,
and the United States, fresh from its victory in the SpanishAmerican War, rapidly becoming the
worlds leading industrial power. Technical inventions such as a practical light bulb, the
telegraph and telephone, phonograph, motion picture camera, and projector changed the way
people lived. There was a great deal of optimism looking forward to the new century of
continued progress. There were a series of worlds fairs and exhibitions in which the latest
inventions were touted. Perley G. Nutting, the prime mover in the founding of The Optical
Society, apparently constructed the very rst neon sign and exhibited it at the Louisiana Purchase
Exhibition in 1904, proudly proclaiming NEON in glowing light.
It was in this heady environment that optics entered the twentieth century. Optics was
centrally involved in two scientic revolutions that shook condence in the foundations of the old
Newtonian science that had served the science and industry of the nineteenth century so well:
Einsteins relativity and quantum mechanics. Patricia Daukantas reviews the advances in
spectroscopy up to 1940 and their importance to the development of quantum theory and
astronomy. Today it is difcult to imagine carrying out precision spectroscopic measurements
without a laser, a computer, or a photomultiplier or photodiode. Photographic plates had to
sufce, unless you used Albert Michelsons technique of calibrating dark-adapted students. That
proved adequate for him to resolve the 1.7 GHz ground state hyperne splitting of sodium by
measuring the drop-off of the visibility of the fringes in his interferometer illuminated by
uorescence from sodium. By 1940 the new quantum theory was in place, and Paul Dirac and
Erwin Schrdinger had developed a quantum version of electrodynamics. The basic ideas
underlying modern quantum optics were in place awaiting the development of optical technology
that would allow controlled experiments one atom and one photon at a time. As we will see in
later chapters in this volume, these technological developments followed in the second half of the
twentieth century following the development of the laser.
Prior to the twentieth century, science and engineering were carried out mostly by university
professors and amateur scientists working mostly alone with only their own funds or perhaps a
rich patrons municent interest. This changed completely in the new century, rst by the
establishment of a number of industrial and governmental research laboratories, and then by
governmental science and engineering funding agencies following World War II. I review the
founding of these laboratories and their central importance to twentieth century optics.
9
A very important optical industry has a history that almost exactly spans the rst century of the
existence of The Optical Society: lm-based photography. Todd Gustavson recounts the history of
photography, concentrating particularly on the rst 40 years of the twentieth century. A lot of optical
instrumentation is fairly specialized in its application, with but a few thousand to a few tens of
thousands of units sold. With the introduction of George Eastmans Brownie camera in 1900, optics
became mass market with sales of hundreds of thousands to millions. The economics of optics was
completely changed, and with that technology changed equally rapidly.
A second mass-market development in optics was the production of affordable eyeglasses. Bausch
and Lomb sold 20 million in 1903, and American Optical was not far behind. This supported rapid
progress in vision research, which Patricia Daukantas reviews. From the founding of OSA to today this
has remained a central concern of the Society and its members. As the average human lifespan increased
due to improvements in sanitation, nutrition, and medical science, age-related vision problems became
more important, and this eld of optics responded with rapid developments.
The development of color photography and color printing as mass industries required standardi-
zation of color measurements and the development of a better understanding of color vision. Roy Berns
recounts these developments with particular emphasis on the role of OSA and its committees.
This series of essays takes us up to the beginning of World War II, after which the climate for
research and development in optics changed dramatically into something approximating its current
form.
T
he nineteenth century laid the foundation for modern optics and for the establishment of
The Optical Society in 1916. Optical science had come a long way from Newtons
pioneering Optiks, but much remained to be learned. In 1800 Newtons particle theory
of light still held sway, the interference of light had not been recognized, and the rest of the
electromagnetic spectrum was undiscovered. Only the wealthy and elite used spectacles, poor
glass quality limited the use of refractive optics, and the worlds largest telescope was a 1.2-m
reector built by William Herschel in 1789 that required frequent repolishing.
Robert Bunsen matched wavelengths that they measured in the lab with solar lines (Fig. 3). Astronomers
William and Margaret Huggins then showed that stellar spectra included lines found in sunlight, and they
measured the Doppler shift of Sirius, the rst stellar motion detected on Earth.
Spectroscopy also opened a new window on atomic physics. In 1885, Swiss mathematician Johann
Balmer discovered a numerical pattern in a series of visible hydrogen wavelengths measured by ngstrm.
The wavelengths were equal to a constant multiplied by the quantity n2 n2 22 where n was an integer.
Balmer used the formula to predict additional wave-
lengths in the ultraviolet, which William Huggins
and Hermann Wilhelm Vogel conrmed in the
spectra of white stars. Later, Johannes Rydberg
developed a more general formula that explained
other series of lines.
Those patterns remained a mystery until Niels
Bohr recognized them as transitions between a
limited number of electron orbits in the hydrogen
atom and then developed the Bohr model of
hydrogen in 1913, a major step on the road to
quantum theory.
Optical Instruments
The poor quality of optical glass limited optical
instruments at the start of the nineteenth century.
In 1757, John Dollond had combined crown and
int glass to make the rst achromatic lens, but he
lacked high-quality glass and accurate dispersion
measurements. Eighteenth century astronomers
Fig. 3. Astronomical spectroscopy in the nineteenth
had turned to reectors for a better view of the century required attaching a spectrometer to the
sky. The worlds largest telescope in 1800 was a telescope and viewing the dispersed spectrum with
reector with a 1.26-m mirror and 12-m focal the eye [14].
Spread of Spectacles
Although Benjamin Franklin is famed for inventingor at least popularizingbifocals in 1784, few
people of his time wore spectacles. They were expensive, and visual science was not advanced enough to
give a precise correction.
Thomas Young has been called the father of physiological optics based on his 1801 paper On the
mechanism of the eye [8,9]. He developed an optometer to measure visual accommodation, analyzed
References
1. A. Einstein and L. Infeld, The Evolution of Physics (Simon & Schuster, 1961).
2. R. Wimmer, Essays in Astronomy (D. Appleton, Company, 1900). Public domain.
3. A. B. Buckley, Through Magic Glasses, and Other Lectures (Appleton, New York, 1890).
4. http://www.aip.org/history/cosmology/tools/tools-refractors.htm
D
uring the rst quarter century of The Optical Society (OSA), spectroscopy led to major
insights into atomic and molecular physics and paved the way for important practical
applications. Optical spectroscopy existed for decades before the formation of OSA, but
it was empirical and descriptive in its nature. Spectroscopists had carefully measured the
wavelengths of spectral lines associated with various elements, but the subatomic mechanisms
that created these lines were not yet fully understood.
Twenty-four years later, as the world lurched toward the second all-encompassing war of
the twentieth century, the spectroscopic ngerprints of atoms and molecules had provided vital
evidence for the emerging quantum theory. Experimentalists rened their techniques and
discovered previously unknown phenomena.
Wolfgang Paulis rule for equivalent electrons and his exclusion principle;
Friedrich Hunds correlation of spectral terms with electron congurations and his correlation of
multiplet components to series limits; and
the determination by George Uhlenbeck and Samuel Goudsmit of the contribution of electron spin
to the complexity of spectra, and their postulation of the half-integral quantum numbers of
fermions.
Nearly simultaneously in 1925, Werner Heisenberg and Erwin Schrdinger formulated their
matrix and wave mechanics formalisms, and quantum theory blossomed. Two years later, Heisenberg
came up with his uncertainty principle, which partially explains spectral line broadening (but is
certainly not the only cause of it).
Advances in Molecular
Spectroscopy
While some physicists occupied themselves with
subatomic structures, other physicists and chemists
investigated new spectroscopic phenomena in mole-
cules. The nineteenth-century observations of uo-
rescence by G. G. Stokes led to the American R. W.
Woods discovery of resonance radiation of vapors
in 1918.
Wood (see Fig. 2), for whom an OSA award is
named, began his career with detailed investigations
of the spectra of iodine, mercury, and other elements Fig. 2. R. W. Wood. (Courtesy of The Observatories
of the Carnegie Institution for Science Collection at the
in gaseous form. As a biographer wrote, Wood Huntington Library, San Marino, California.)
discovered resonance radiation and studied its
many puzzling features with great thoroughness
and amazing experimental ingenuity [8].
By far the biggest boost to molecular spectros-
copy during this time period was C. V. Ramans
discovery of the inelastic scattering of lightthe
effect that came to bear his name. During his
European trip in 1921, Raman (see Fig. 3), a native
of India, spied the wonderful blue opalescence of
the Mediterranean Sea and, as a result, was inspired
to study the scattering of light through liquids [9]. In
1928, he and a colleague, K. S. Krishnan, discovered
the inelastic scattering of photons now known as the
Raman effect.
Lacking lasers, Raman and Krishnan had to use
sunlight passed through a narrow-band photo-
graphic lter as a monochromatic light source.
Early scientists who studied Raman scattering used
mercury arc lamps or gas-discharge lamps as their
sources. Nevertheless, in the 1930s scientists used
Raman spectroscopy to develop the rst catalog of
molecular vibrational frequencies. The technique,
however, would not reach its full owering until the
development of the laser in the 1960s.
Optical spectroscopy also played an important
role in the understanding of nuclear structure. Al-
though A. A. Michelson had observed hyperne
Fig. 3. Chandrasekhara Venkata Raman.
structure as far back as 1881, it lacked an interpre- (Massachusetts Institute of Technology, courtesy AIP
tation until 1924, when Pauli proposed that it Emilio Segre Visual Archives.)
References
1. W. F. Meggers, Spectroscopy, past, present, and future, J. Opt. Sci. Am. 36, 431443 (1946).
2. A. Sommerfeld, The model of the neutral helium atom, J. Opt. Sci. Am. 7, 509515 (1923).
3. H. N. Russell and F. A. Saunders, New regularities in the spectra of the alkaline earths, Astrophys.
J. 61, 3869 (1925).
4. F. S. Brackett, Visible and infra-red radiation of hydrogen, Astrophys. J. 56, 154161 (1922).
5. L. A. Du Bridge and P. A. Epstein, Robert A. Millikan, in Biographical Memoirs (National Academy
of Sciences, 1959), p. 260.
6. H. W. Babcock, Ira S. Bowen, in Biographical Memoirs (National Academy of Sciences, 1982),
Vol. 53, p. 92.
7. P. Swings, Edlns identication of the coronal lines with forbidden lines of Fe X, XI, XIII, XIV, XV;
Ni XII, XIII, XV, XVI; Ca XII, XIII, XV, A X, XIV, Astrophys. J. 98, 116128 (1943).
8. G. H. Dieke, Robert Williams Wood, in Biographical Memoirs (National Academy of Sciences,
1993), Vol. 62, p. 445.
9. C. V. Raman, The molecular scattering of light, Nobel lecture, 11 December 1930, online at www.
nobelprize.org/nobel_prizes/physics/laureates/1930/raman-lecture.pdf.
10. W. F. Meggers and K. Burns, Hyperne structures of lanthanum lines, J. Opt. Sci. Am. 14, 449454
(1927).
11. W. F. Meggers, C. C. Kiess, and F. J. Stimson, Practical spectrographic analysis, Scientic Paper 444,
Scientic Papers of the Bureau of Standards 18, 235255 (1922).
A
common impression is that each of the many types of lasers was invented in an
industrial research laboratory. While one can dispute the accuracy of that statement in a
few cases, there is no argument that industrial and governmental research laboratories
were the locations of much of the development of optics in the twentieth century.
The concept of an industrial research laboratory emerged just before the beginning of the
twentieth century. The rst industrial optics research laboratory was Carl-Zeiss Stiftung,
founded in 1889, in Jena, Germany, by Ernst Abbe. It grew out of earlier collaboration by
Abbe, Otto Schott, and Carl Zeiss, and quickly became the source of optical glass and precision
optical instruments for most of the world [1]. This German success did not go unnoticed and
helped to stimulate the founding of other laboratories. The contributions of industrial and
governmental laboratories in the twentieth century were truly incredible, and this essay briey
reviews how these various laboratories came to be; but it will leave, for the most part, their
enormous range of inventions and discoveries to be described in the later essays in this volume.
Several factors led to the rise of industrial and government research laboratories at the
beginning of the twentieth century. The harnessing of steam power, and then electricity, led to
mass-consumer-product industries that had sufcient resources to support basic research
laboratories. In 1903 Bausch & Lomb sold 20 million spectacle lenses and 500,000 photographic
lenses per year; Eastman Kodak sold 150,000 Brownie cameras in 1900, the rst year it was sold;
and by 1914 General Electric sold 88.5 million lamps in the United States alone [2]. The general
public saw the night lit up by electric lights; radio, telephone, and motion pictures changed the
way people lived and perceived the future. Thomas Edison, George Westinghouse, and Nikola
Tesla captured the popular imagination as scientic geniuses who would develop new technolo-
gies that would revolutionize industry. Everything was aligned to enable and encourage large
investments in basic research. Small laboratories for quality and process control existed before,
but not industrial and governmental research laboratories whose task was to develop whole new
technologies and products that had never existed.
Following the Civil War, industry grew rapidly in the United States. The new companies
were receptive to change and optimistic about future technologies, so much of the early
development of industrial laboratories occurred in the United States. In 1900 General Electric
(GE) established the rst industrial basic research laboratory in Schenectady, New York, an
outgrowth of Edisons earlier laboratories.
General Electric characterizes the nature of this laboratory:
The lab was the rst industrial research lab of its kind. Prior to the formation of the GE
Research Lab the only industrial research labs were German pharmaceutical labs. In the
German labs like Bayer scientists and researchers worked independently and competed with
one another. At General Electric in Schenectady, New York engineers and scientists were
encouraged to share information and assist with problem solving. They were given great
nancial support to buy materials. The best machinists and craftsmen were employed to
help build prototypes. From the tungsten light bulb to the computerized hybrid car it is no
wonder that the Schenectady lab produced a great proportion of our world's technology [3].
23
While the General Electric laboratory was not focused on conventional optics, it did work on
illumination and the development of x-ray sources. William Coolidges x-ray tube designs were
instrumental in leading to the development of radiology, and his discovery of a method to make
tungsten ductile provided a long-life lament for incandescent light bulbs. Soon GE was selling
them by the millions, and Irving Langmuirs studies of monatomic lms on laments led to GEs rst
Nobel Prize. Most important, the GE Research Lab set the standard that other industrial labs used
as a model.
In 1918 the Westinghouse Research Laboratory was established with goals and organization much
like those of the earlier General Electric laboratory. In particular, this research laboratory was separate
from any manufacturing facility. Again, the early work in this laboratory was not devoted to optics,
although it was soon working in optical spectroscopy, a pursuit that it maintained for most of the
century. One notable contribution to optics from this Pittsburgh laboratory was that it provided the
rst job for Brian OBrien, who was the rst permanent director of the University of Rochesters
Institute of Optics. OBrien, working with Joseph Slepian, developed the rst lightening arrestors,
which are commonly used today [4].
In 1915 the Eastman Kodak Research Laboratory was founded, and before World War I (WWI)
broke out, laboratories were established at Dupont, Standard Oil (Indiana), U.S. Rubber, and Corning
Glass. Bausch & Lomb did not have a formal research laboratory at that time but were soon central to
the United States efforts in optical research and development. After WWI Major Fred E. Wright wrote
the following in a Journal of the Optical Society of America article [5]:
Before this country entered the war, it was realized that the making of optical glass might
prove to be a serious problem. Prior to 1914, practically all of the optical glass used in the
United States had been imported from abroad; manufacturers followed the line of least
resistance and preferred to procure certain commodities, such as optical glass, chemical dyes,
and other materials difcult to produce, direct from Europe, rather than to undertake their
manufacture here. The war stopped this source of supply abruptly, and in 1915 experiments
on the making optical glass were underway at ve different plants: The Bausch & Lomb
Optical Co. at Rochester, N.Y.; the Bureau of Standards at Pittsburgh, Pa.; the Keuffel &
Esser Company at Hoboken, N.J.; the Pittsburgh Plate Glass Company at Charloi, Pa.; the
Spencer Lens Company at Hamburg, Buffalo, N.Y. By April, 1917, the situation had become
acute; some optical glass of fair quality had been produced, but nowhere had its manufacture
in adequate quantities been placed on an assured basis. The glass-making processes were not
adequately known. Without optical glass, re-control instruments could not be produced;
optical glass is a thing of high precision, and its manufacture, accurate control is required
over all the factory processes. In this emergency the Government appealed to the Geophysical
Laboratory of the Carnegie Institution of Washington for assistance. This laboratory had been
engaged for many years in the study of solutions, such as optical glass, at high temperatures, and
had a corps of scientists trained along the lines essential to the successful production of optical
glass; it was the only group in the country with a personnel adequate and competent to
undertake a manufacturing problem of this character and magnitude. A group of their scientists,
with writer [Major Wright] in charge, was accordingly placed in April 1917, at the Bausch &
Lomb Optical Company, and took over virtual direction of the plant.
The effort succeeded, and the United States became a serious player in optics and optical
instrumentation, no longer depending on European supplies and technology.
The military importance of precision optics in WWI was enormously enhanced by two technolog-
ical developments: (1) machining of artillery barrels was much more precise than ever before so that
shells could be directed much more accuratelyif you knew with enough accuracy where your target
was located; and (2) military aircraft, which required bomb sights and aerial cameras for the airplanes
and ground-based binoculars and telescopes for the anti-aircraft artillery. Another development that
one does not usually associate with optics was the invention of camouage to hide ships, airplanes, and
land-based targets from the improved optics. Abstract artists were brought in to design the patterns,
The idea of a national bureau of standards was presented at an auspicious hour. America in the
year 1900 thought well of itself. The hard times of 189395 were all but forgotten in the aura
of prosperity and sense of achievement that energized the Nation. Industry and invention
boomed and business ourished as never before. The prophets at the turn of the century
unanimously agreed on the good years to come.
The major exception to the policy of avoiding the building of government laboratories was in
the development of the atomic bomb. After preliminary studies by NDRC and OSRD, it
became clear that a colossal program would be needed, and Bush recommended to Secretary
Stimson that the Army take over the responsibility. The result was the formation of Manhattan
Engineering District by the Corps of Engineers. Bush with Conant as his deputy, maintained an
active scrutiny of the enterprise.
This was the foundation of science and engineering administration in the U.S. as it exists up until
now. All of the developments in optics in the second half of the century grew up in this environment.
Optics during the war was overseen by Division 16, Optics and Camouage of the NDRC. It was led by
George Harrison. Paul Kelley describes elsewhere in this volume the optical developments during this
period. Well before the war was over, Bush started to plan how the momentum of research could be
sustained with new peacetime goals. President Roosevelt asked him to make recommendations on
government policies for combating disease, supporting research, developing scientic talent, and
diffusing scientic information. Four committees were set up to generate recommendations. On the
basis of these recommendations Bush submitted a report titled ScienceThe Endless Frontier, which
laid out the proposals for organizing post-war science and technology. The argument for the
government to continue supporting research after the war was summed up in the report: To create
more jobs we must make new and better and cheaper products. We want plenty of new, vigorous
enterprises. But new products and processes are not born full-grown. They are founded on new
principles and new conceptions which in turn result from basic scientic research. Basic scientic
research is scientic capital.
References
1. Carl-Zeiss-StiftungCompany prole, information, business description, history, background
information on Carl-Zeiss-Stiftung. http://www.referenceforbusiness.com/history2/79/Carl-Zeiss-
Stiftung.html.
2. R. Kane and H. Sell, Revolution in Lamps: A Chronicle of 50 Years of Progress, 2nd ed. (Fairmont Press,
2001), p. 37, table 21.
3. General Electric Research Lab, http://www.edisontechcenter.org/GEresearchLab.html.
4. C. R. Stroud, Jr., Brian O'Brien, 18981992, A Biographical Memoir (National Academy of Sciences,
2010).
5. F. E. Wright, War-time development of the optical industry, J. Opt. Soc. Am. 2, 1 (1919).
6. C. E. K. Mees, The Kodak Research Laboratories, Proc. R. Soc. Lond. 8 135, 133147 (1948).
7. C. Stroud, Jewel in the Crown (Meliora Press, 2004), p. 18.
8. R. Sobel, RCA (Stein and Day, 1986).
9. http://www.nist.gov/nvl/upload/MP275_06_Chapter_I-__AT_THE_TURN_OF_THE_CENTURY.
pdf.
10. V. Bush, Pieces of the Action (William Morrow, 1970), p. 32.
11. J. B. Wiesner, Vannevar Bush 18901974, Biographical Memoir (National Academy of Sciences, 1979).
Introduction
The photographic process, announced in 1839 by the Frenchman Louis Jacques Mand
Daguerre, captured and xed the images that were viewed through a camera obscura. This
was accomplished through a combination of mechanics (the camera), optics (to improve the
image), and chemistry (to sensitize and process the image). Over the next forty years, improve-
ments made to all aspects of the processcameras, shutters, lenses, and chemistryled to
cheaper and simpler image making, generating a growing interest for the nonprofessional
photographer.
The technicalities of early photography required the photographer to sensitize media shortly
before exposure and then process the image immediately afterward. Although this system was
ne for the professional, it was generally too cumbersome and time-consuming for most
amateurs. On 13 April 1880, George Eastman of Rochester, New York, patented a machine
for coating gelatin dry plates. The following January, with the nancial backing of Rochester
businessman Henry Strong, he formed the Eastman Dry Plate Company, one of the rst
commercial producers of light-sensitive photographic emulsions. With reliable plates now
available, companies worldwide began manufacturing cameras designed specically to use them.
Eastmans business expanded ve years later with the introduction of his American Film, a
paper-supported stripping lm intended for the professional market. It was not well received by
the professionals, who considered it to be rather difcult to process. Undeterred, Eastman instead
used it in a new small box camera he named the Kodak. Introduced in 1888, the Kodak was an
easy-to-use detective camera, a box-style, point-and-shoot camera meant for the novice
photographer. Eastmans camera required no adjustments, which was atypical of the time, but
the real innovation was after exposing the lm: the camera was shipped back to the company for
processing and reloading, marking the beginning of the professional photo-nishing industry.
This novel feature was marketed with the advertising slogan You press the button, we do the
rest, which established the companys business model: the promotion of cameras as the means
to selling highly protable lm and processing. Twelve years later, the Brownie camera was
added to the camera line; its $1.00 selling price made photography available to just about
everyone.
Brownie
Introduced by Eastman Kodak Company in 1900, the Brownie camera was an immediate public
sensation due to its simple-to-use design and inexpensive price. (See Fig. 1.) Now nearly anybody,
regardless of age, gender, or race, could afford to be a photographer without the specialized
knowledge or cost once associated with the capture and processing of images. An important
aspect of the Brownie cameras rapid ascendancy in popular culture as a must-have possession
was Eastman Kodak Companys innovative marketing via print advertising. The company took
the unusual step of advertising the Brownie in popular magazines instead of specialty photogra-
phy or trade magazines with limited readership. George Eastman derived the cameras name
31
from a literary character in popular childrens
books by the Canadian author Palmer Cox.
Eastmans astute union of product naming, with
a built-in youth appeal, and inventive advertising
placement had great consequence for the rise of
modern marketing practices and mass consumerism
in the twentieth century.
The Brownie was designed and manufactured
by Frank A. Brownell, who had produced all of
Eastman Kodaks cameras from the beginning. The
use of inexpensive materials in the cameras con-
struction and George Eastmans insistence that all
distributors sell the camera on consignment enabled
Fig. 1. Brownie camera. Eastman Kodak the company to control the cameras $1 price tag
Company, Rochester, New York, ca. 1901. Gift of Ansel and keep it within easy reach of consumers pocket-
Adams, 1974.0037.1963. books. More than 150,000 Brownies were shipped
in the rst year of production alone, a staggering
success for a company whose largest single-year
production to date had been 55,000 cameras (the
No. 2 Bullet, in 1896). The Brownie launched a
family of nearly 200 camera models and related
accessories, which over the next 60 years helped to
make Kodak a household name.
Institute of Optics
World War I changed the optical landscape in the United States. The industry relied on German
manufacturers for the supply of high quality optical glass, optics, and engineers. A number of steps were
taken to remedy the situation, the rst being establishing The Optical Society (OSA) in 1916. Under the
leadership of Perley G. Nutting, and with the support of optical scientists in Rochester, the optical
center of the United States, the OSAs mission was to promote and disseminate knowledge of optics and
photonics. This was accomplished with published journals and by holding conferences, thus establish-
ing a network of information exchange. The University of Rochester, with nancial support from B&L
and Eastman Kodak Company, established the Institute of Applied Optics (now known as the Institute
of Optics) in 1929. The president of the University, Reverend Benjamin Rush Rhees, hired Rudolf
Kingslake, graduate of the Imperial College of Science and Technology in London, where he studied
under Alexander Eugen Conrady, to teach at the new school. Kingslake became the head of Eastman
Kodak Companys Optical Design Department in 1937, a position he held until retiring in 1968.
Kingslake continued to teach at the University of Rochester during his Kodak years; he continued
teaching at the university into the 1980s.
Leica A
Starting about 1905, when he worked at the rm of Carl Zeiss in Jena, Germany, Oskar Barnack
(18791936), an asthmatic who hiked to improve his health, tried to create a small pocketable camera
to take on his outings. At the time, cameras using the most common format of 13 18 cm (5 7 in.)
were quite large and not well suited for hiking. Around 1913, Barnack, by then an employee in charge
of the experimental department of the microscope maker Ernst Leitz Optical Works in Wetzlar,
designed and hand built several prototypes of a small precision camera that produced 24 36-mm
images on leftover ends of 35-mm motion picture lm. Three of these prototypes survive. The most
Contax I (540/24)
The successful introduction of the Leica camera was not lost on Zeiss Ikon AG of Dresden, Germany.
Formed in 1926 as the merger of Contessa-Nettel, Goerz, Ernemann and Ica, Zeiss Ikon was the largest
camera manufacturer in Europe. Zeiss was one of the leading manufacturers of optical devices, with its
roots dating back to optician Carl Zeiss. Zeiss began as a lens and microscope manufacturer in 1847.
He hired physicist Ernst Abbe in 1866 as research director; Abbe designed the rst refractometer in
1868, a device used to measure the index of refraction of optical glass. Abbe hired Otto Schott in 1883
to develop new types of glass necessary for reducing reection in microscope objectives, then hired Paul
Rudolph to design photographic lenses with glass developed by Schott. After the passing of Carl Zeiss
in 1888, Abbe bought out Zeisss son Roderich and established the Carl Zeiss Foundation. Unusual in
its day, the Zeiss foundation was partially owned by its workers. Many of the classic lenses used in
photography, such as the Anastigmat (1890), Planar (1895), Unar (1899), and Tessar (1902),
originated at Zeiss, under the direction of Paul Rudolph.
The Zeiss Ikon catalog of 1927 listed over 100 camera models from the small pocket-sized
Piccolette roll lm camera to the Universal Jewel professional folding dry plate camera (Ansel Adams
used one). Its camera line included the Deckrullo focal plane shutter models and the Miroex reex.
And like Eastman Kodak Company, along with cameras Zeiss Ikon sold a complete line of
photography equipment for darkroom and motion picture projection. With the introduction and
success of the Leica from one of its smaller competitors, Zeissconsidered to be the gold standard of
camera makersneeded to come up with a better version of the precision 35-mm camera. The answer
was the Contax, introduced in 1932. (See Fig. 4.) On paper it was exactly that, a better Leica. The
Contax used a built-in coupled rangender, with a longer base than the Leicas, for more accurate
focusing, vertical-traveling focal plane shutter, with speeds to 1/1250 s, which was more than twice as
fast as the Leicas 1/500. The Contax had a removable back for easy loading, in contrast to the Leica,
which rather awkwardly loaded through its removable bottom plate. And most important, the
Contax used Zeiss lenses, which were far superior to those used by the Leica. But there was one
problem: the Contax was an unreliable picture taker, with most of the problems relating to its shutter.
Kodak Retina
August Nagel, of Contessa Nettel, dissatised with
his companys merger with Zeiss Ikon, left and
formed a new company, Nagel Werke in 1928.
Eastman Kodak Company purchased Nagel Werke
in 1932, becoming Kodak AG, the companys Ger-
Fig. 4. Contax I (f). Zeiss Ikon AG, Dresden, man manufacturing arm. In 1934, Eastman Kodak
Germany, ca. 1932. Gift of 3M; ex-collection Louis Company introduced the Retina, its rst precision
Walton Sipley. 1977.0415.0004.
35-mm camera, designed to compete with the
Leica. Unlike the Leica and Contax, the Retina
was a folding 35-mm camera with a permanently
mounted lens. Introduced with the Retina was the
Kodak 35-mm daylight loading lm magazine,
which became the standard used on just about every
35-mm camera. The Kodak lm magazine used a
built-in heat-sealed velvet light trap still in use
today. Prior to this, the other 35-mm cameras used
their own unique lm magazines, tted with some
type of light trap mechanism connected in some way
to the bottom of the camera (Leica) or with separate
supply and take-up housing (Contax).
Kodak AG went on to produce some 50 differ-
ent models of the Retina camera through the
mid-1960s.
Conclusion
Camera research and development largely went on hold during World War II. Much of the German
photo manufacturing industry was destroyed by the end of the war. The post-war era also saw the
division of the Zeiss factories, split between East and West Germany. The low cost of post-World-War-
II German labor had a direct impact on American manufacturing, causing most U.S. makers to
concentrate on inexpensive point-and-shoot cameras only. And the U.S., in trying to strengthen Japan,
helped re-establish the edgling camera manufacturing there, laying the seeds for what became the
premier camera manufacturing power for the rest of the century.
B
y the second decade of the twentieth century, scientists studying human vision had come a
long way from the days of the ancient Greeks, who debated whether light rays shot
themselves out of the eyeball or emanated from objects in the visual eld [1]. Neverthe-
less, the whole area of vision, especially the retinas reaction to light, remained an important topic
of research as The Optical Society (OSA) was organizing itself.
In the early days of the OSA, scientists had come to realize that vision sat at the intersection
of three elds: physiology, for the anatomy of the eye; physics, for the action of stimuli on the eye;
and psychology, governing how the conscious brain interprets the eyes sensations [2]. Reecting
the interdisciplinary nature of the subject, vision-related articles published in 1920 were
distributed among 58 different journals from elds ranging from physics and engineering to
zoology and pathology.
Between the two world wars, the scientists studying photochemistryincluding two who
would become OSA Honorary Membersprogressed from the simple eyes of sea creatures to the
complexities of the human visual system. Researchers learned that the retina contains vitamin A,
leading to generations of parents telling their children, Eat your carrotstheyre good for your
eyesight! The new understanding of the eye paved the way for advances in vision correction and
optical instruments.
Photochemistry: Hecht,
Hartline, and Wald
During the 1920s and 1930s, three scientists whose
talents bridged the elds of physics, chemistry, and
biology made invaluable contributions to our un-
derstanding of the molecules that react in the pres-
ence or absence of light.
Born in an Austrian town now part of Poland,
but raised in the United States, Selig Hecht (1892
1947) (Fig. 3) explored the photochemistry of vision
by studying animals whose visual systems are much
simpler than those of humans: the worm Ciona and
the clam Mya. Those organisms reactions to light
were slow enough that they could be measured
without sophisticated apparatus [6].
Hecht began his studies of the photoreceptor
process immediately after receiving his Ph.D., when
he spent a summer at the facility now known as the
Scripps Institution of Oceanography. There he inves-
tigated the sensitivity of Ciona to light. As he moved
among several institutions in the United States and Fig. 2. Leonard Thompson Troland, OSA
England, he studied the rate at which visual purple president from 1922 to 1924. (AIP Emilio Segre Visual
Archives.)
(now known as rhodopsin) decomposes upon expo-
sure to light [7], the bleaching of rhodopsin in solu-
tion [6], and (with Robert E. Williams) the spectral sensitivity of human rod vision [8]. Hecht ended up at
Columbia University, where, with his frequent collaborator Simon Shlaer, he built an instrument for
Lasting Consequences
Many of the discoveries about the eye as a visual system did not bear practical fruit until after the
interwar (19161940) period. The studies of sensitivity performance and contrast thresholds of the
human eye formed the basis of everything from television and computer displays to the design of
highway signs, which must be read in mere milliseconds for safetys sake [16,17]. That early twentieth
century work continues to enhance many aspects of our twenty-rst century life.
References
1. J. P. C. Southall, Early pioneers in physiological optics, J. Opt. Sci. Am. 6, 827842 (1922).
2. L. T. Troland, The Present Status of Visual Science, Bulletin of the National Research Council of the
National Academy of Sciences (U.S.A.) (1922), Vol. 5, No. 27, pp. 12.
T
he Optical Society (OSA) was the dominant professional society in the evolution of color
science, both through its many technical committees and through the Journal. This
chapter highlights some of the many signicant activities and publications that occurred
through the 1950s.
OSA established the Committee on Colorimetry in 1919 chaired by I. G. Priest from the
National Bureau of Standards and during its rst year circulated a preliminary draft [1]. The
committees rst report was published in the Journal in 1922, authored by the current
chairman and president of the Society, L. T. Troland [2]. This remarkable 64-page report
outlined the basis of photometry and colorimetry, including visibility and color-matching
function data (referred to as the OSA excitation curves), terminology for visual description,
chromaticity diagrams, complementary wavelengths, standard illuminants, color temperature,
optimal color lters for trichromatic color reproduction, visual colorimetry, and transforma-
tion of primaries. All of these concepts would be central to establishing the 1924 V visibility
curve and the 1931 CIE colorimetric system, XYZ and xyY. The Colorimetry Committee was
a driving force in the evolution of modern colorimetry, culminating with the book The Science
of Color published in 1953 [3]. The book indicates the breadth of expertise of the committee
and that color science is multi-disciplinary as it includes physics, optics, physiology, psycho-
physics, and history beginning with our rst use of colored materials hundreds of thousands of
years ago.
The rst color order system that was based on extensive psychophysics was the Munsell
system. The Munsell Value scale quantied visual compression by establishing the relation-
ship between incident light and perceived lightness [4]. It has been used to support Stevens
exponential model of visual compression and relate luminance factor to CIE lightness, L*. An
OSA committee performed extensive research leading to the current denition of the Munsell
system [5]. These data were used by Adams to derive the precursor to CIELAB [6]. The
Munsell system is a cylindrical system, and as a consequence, neighboring samples are not
equidistant. In addition, samples of constant hue vary in either lightness or chroma, but not
both simultaneously as occurs in common coloration. In the late 1940s an OSA committee,
chaired by D. B. Judd from the National Bureau of Standards, was established to develop a
new color order system where samples were equidistant in all three dimensions based on a
regular rhombohedral crystal lattice structure to [7]. The OSA Uniform Color Scales were the
result thirty years later. Both systems are still used to develop and evaluate colorimetric-based
color spaces for visual uniformity.
Any quantitative color description of objects depends on measuring the spectral reectance
factor. A breakthrough occurred during the 1930s when A. C. Hardy, a professor at the
Massachusetts Institute of Technology, developed the rst recording spectrophotometer whose
illumination geometry was optimized for measuring materials via an integrating sphere where
the specular component could be included or excluded, the latter correlating with the
appearance of glossy materials [8]. General Electric manufactured the Hardy spectrophotom-
eter. By the late 1940s, it was possible to interface the instrument to an automatic tristimulus
integrator [9], and as a result, color measurements were reported as a spectral graph and CIE
43
tristimulus values. One drawback of this approach was the high cost. Hunter made color measure-
ment much more accessible with the development of a color-difference meter using color lters and
three photodetectors, rst presented at an OSA Annual Meeting in 1948 [10].
When the CIE system was promulgated in 1931, there were three standard sources, A, B, and C,
representing incandescent, sunlight, and daylight, respectively. Source C was produced by ltering
incandescent lighting with bluish liquid lters. Such a light was very decient in UV and short-
wavelength visible radiation compared with natural daylight. Measurements of daylight, principal
component analysis, and a very clever approach to calculate the eigenvector scalars for a specic
correlated color temperature resulted in the CIE D series illuminants [11, 12]. Today, CIE illuminants
D50 and D65 are used extensively in color reproduction and color manufacturing, respectively.
All specications include tolerances, and as early as 1932 [13], the Journal began publishing
research demonstrating the CIE systems lack of uniformity with respect to color discrimination,
research proposing linear and nonlinear transformations that improved correlation, and psychophysi-
cal data from discrimination experiments. At the forefront of this research was D. L. MacAdam, a
student of Hardy at MIT, who went on to have a distinguished career at the Eastman Kodak Research
Laboratories. In the early 1940s, he built an apparatus to measure color-matching variance that
resulted in the MacAdam ellipses, still used as a discrimination dataset [14]. His research and
leadership resulted in the 1960 uv and 1976 uv uniform chromaticity scale diagrams and the 1976
L*a*b* and L*U*V* uniform color spaces.
An interesting research topic was designing color reproduction systems that could be related to
colorimetry by linear transformation. During the late 1930s, Hardy and Wurzburg [15], MacAdam
[16], and Yule [17] laid the groundwork for todays color management for both additive and
subtractive imaging systems.
We all use manufactured products meeting a color specication. Predicting and controlling a recipe
is invaluable for coloration systems where the colorants and media both absorb and scatter light. The
theory proposed in 1931 by P. Kubelka and F. Munk and published in the Journal in 1948 [18]
continues to be used successfully in textiles, plastics, and coatings. In 1942, J. L. Saunderson
demonstrated its effectiveness for the coloring of plastics, particularly by accounting for refractive
index discontinuities at the surface [19].
Today, color science has evolved from tristimulus XYZ, through L*a*b* and L*u*v*, to color-
appearance spaces such as CIECAM97s and CIECAM02. A key requirement of such spaces is
accounting for the effects of chromatic adaptation. Such research began in the 1950s and the seminal
experiments by R. W. Burnham, R. M. Evans, and S. M. Newhall from Eastman Kodak remain reliable
and viable data [20].
I will end my highlight tour with Ref. [21], which describes how MacAdam created separation plates
for printing both the color gamut of a set of offset printing inks and a spectrum. A 19-page article
appeared in the 3 July 1944 issue of Life magazine, titled Color: it is the response of vision to wave
lengths of light [22]. This remarkable article includes colored images of a dispersed spectrum, additive
and subtractive mixing, principles of selective absorption of colored lters, spectral reectance curves of a
lemon and a tomato, the Hardy recording spectrophotometer, the visible spectrum, the Munsell system,
an afterimage demonstration using the American ag, and several other optical illusions. The 1931 CIE
system was used to calibrate the color separations where dominant wavelength represented the spectral
hues and, in turn, mixtures of the printing inks. Incredibly, I have MacAdams copy of the article. The
article summarizes color science and, indirectly, the tremendous impact the OSA has had on its evolution.
References
1. 1919 Report of the Standards Committee on Colorimetry, J. Opt. Soc. Am. 4, 186187 (1920).
2. L. T. Troland, Report of Committee on Colorimetry for 192021, J. Opt. Soc. Am. 6, 527591
(1922).
3. Committee on Colorimetry of The Optical Society of America, The Science of Color (Thomas Y.
Crowell, 1953).
49
the military, and infrared spectroscopy played a vital role in the rubber manufacturing process. The entry
of Perkin-Elmer and Beckman into the spectrometer business was motivated by the use of their equipment
in rubber manufacturing and fuel rening. Chemists, biologists, and other scientists soon came to
embrace the use of physical measurements, most particularly optical spectroscopy in the infrared
region. In 1950, the rst Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy
(Pittcon) was held. Optical techniques continue to play a central role in this enormous conference,
which in 2015 had 16,000 attendees, 925 exhibitors, and more than 2000 sessions.
In 1957 ber endoscopes were used for medical imaging by Hirschowitz employing bundles of clad
bers developed by Peters and Curtiss at Michigan [2,3]. In 1930 Heinrich Lamm demonstrated the
concept of imaging through ber bundles, H. H. Hopkins developed the berscope using coherent ber
bundles in the early 1950s [4], and, also in the early 1950s, A. C. S. van Heel proposed the use of
cladding to avoid crosstalk between bers. Fiber endoscopes are now widely used in clinical medicine,
and ber optical communication relies on the use of clad bers.
In 1961 Xerox announced the rst Xerox copier, which was based on an invention by Chester
Carlson in 1938. The basic idea was to use optical transfer to produce an electrostatic pattern or image
on a drum. This pattern then attracted black material (toner), which could be transferred to paper.
Other printing technology developments in the 1940s and 1950s included phototypesetting, inkjet
printers, and dye sublimation printing. A somewhat related area, photolithography of semiconductor
circuits, was initially developed by Andrus and Bond at Bell Labs [5,6]. This was based on techniques
used to make printed circuits. In one of its rst large-scale applications, the printed circuit had been used
during World War II for proximity fuses. The work of Andrus and Bond was quickly followed by
efforts at Texas Instruments and Fairchild to miniaturize silicon circuits, an effort that would lead to the
microelectronics revolution.
The most revolutionary invention in the century of optics, the laser, was rst realized just after this
period ended. Its precursor, the maser, came in the 1950s. Gordon, Zeiger, and Townes reported [7] the
operation of the ammonia maser in 1954; this was followed by the development of solid state masers
used in radio astronomy [8]. In 1958 Schawlow and Townes published a paper [9] describing the
physics of masers and lasers and a proposed method for making a laser. The next year a conference was
held at Shawanga Lodge in New York State, where further discussions were held concerning the
possible operation of the laser [10]. The race was on.
References
1. From Classical to Modern Chemistry: The Instrumental Revolution, P. J. T. Morris, ed. (Royal Society
of Chemistry, London, 2002).
2. B. I. Hirschowitz, Endoscopic examination of the stomach and duodenal cap with the berscope,
Lancet 1, 10741078 (1961).
3. L. E. Curtiss, B. I. Hirschowitz, and C. W. Peters, J. Opt. Soc. Am. 47, 117 (1957). Paper FC63 at the
OSA Annual Meeting.
4. H. H. Hopkins and N. S. Kapany, A exible berscope using static scanning, Nature 173, 3941
(1954).
5. J. Andrus, Fabrication of semiconductor devices, U.S. patent 3,122,817 (3 March 1964).
6. J. Andrus and W. L. Bond, Photoengraving in transistor fabrication, in F. J. Biondi et al., eds.,
Transistor Technology, Vol. III (D. Van Nostrand, Princeton, 1958), pp. 151162.
7. J. P. Gordon, H. J. Zeiger, and C. H. Townes, Molecular microwave oscillator and new hyperne
structure in the microwave spectrum of NH3, Phys. Rev. 95, 282 (1954).
8. J. A. Giordmaine, Centimeter wavelength radio astronomy including observations using the maser,
Proc. Natl. Acad. Sci. 46, 267276 (1960).
9. A. L. Schawlow and C. H. Townes, Infrared and optical masers, Phys. Rev. 112, 1940 (1958).
10. Quantum Electronics, C. H. Townes, ed. (Columbia University, 1960). Shawanga Lodge Conference
Proceedings.
E
dwin Land was the Thomas Edison of twentieth-century optics, a prolic inventor and
entrepreneur. His milestone introduction of instant photography, at an Optical Society
spring meeting in New York on 21 February 1947, often overshadowed his other
contributions, ranging from 3D movies to surveillance satellites.
Lands rst transformative invention was the plastic sheet polarizer in 1928, when he was
not yet 20. Fascinated by polarization, he tried growing large sheets of iodoquinine sulfate, a
polarizing material invented in the nineteenth century. That did not work, but he found he could
make polarizing sheets by applying an electric or magnetic eld to align tiny crystals of the
material, then embedding them in a celluloid lm. Later he invented a process for making
polarizing sheets by stretching the plastic to align the polarizing crystals. Those plastic sheet
polarizers became the foundation of the Polaroid Corporation.
Land also invented a polarizing lter system that he hoped could solve a major highway
safety problemheadlights blinding other drivers at night. He proposed applying polarizers
aligned one way to headlights and orthogonal polarizers to windshields. Light scattered
from the environment would lose its polarization, so the windshield polarizer would transmit
it. But the polarized windshield would block light directly from the headlights, so only a
few percent would reach the drivers eyes. It sounded great, but the auto industry never
embraced it.
Instead, the polarized lm found other applications. In 1934, Eastman Kodak contracted to
buy it for photographic lters. Kodak was also interested in polarizing sunglasses, but Land got a
better deal from American Optical and in 1935 signed a contract to supply them with polarizing
lm bonded to glass for sunglasses.
Meanwhile, Land invented polarization-based stereoscopy for 3D movies. The rst genera-
tion of 3D movies projected overlapping images in two colors, which viewers watched through
glasses with red and green or red and blue lters. Land realized that glasses with a pair of
polarizers, one horizontal and the other vertical, could give the same effect for overlapping
images projected in horizontal and vertical polarization. A short polarized 3D lm at the
Chrysler Pavilion was a hit at the 1939 New York Worlds Fair. World War II interrupted 3D
movie development but created a need for stereoscopic surveillance imaging that was met by the
vectograph, a transparency-based process invented by Joseph Mahler and Land at Polaroid.
Polarized 3D movies returned after the war to produce a brief boom in the early 1950s, including
the rst color 3D lm, Bwana Devil.
A prescient question asked by Lands young daughter during a 1943 vacation launched his
quest for instant photography. Why couldnt she see the photo he had taken right away? Lands
logical mind realized it was a matter of chemistry, so he invented a self-developing lm that
combined exposure and processing of the negative and transfer to a positive. In early versions,
the photographer pulled a paper tab or leader after exposure, starting a series of events. Inside the
camera, a pair of rollers pressed the positive and negative sheets together and spread a processing
uid between them. This then emerged from the camera and, after a brief specied waiting time,
the photographer pulled the two sheets apart to display the image. Afterward, brushing a nal
coating across the image could preserve it.
51
The rst Polaroid cameras had input rolls of negative and positive monochromatic lm. Color
lm followed in the late 1950s. Polaroid introduced lm packs combining both types in the early
1960s, simplifying handling. Instant photography delighted amateurs, and also found many other
applicationsnotably, recording oscilloscope traces in research labs. Theodore Maimans notebook
recording the rst laser includes Polaroid prints of laser pulse traces.
Lands success lay in hiding the messy chemistry inside the lm package. The most rened version
was the SX-70 color lm introduced in 1972, in which each photo was a separate dry plastic package
ejected by the camera after exposure. The image area was pale green when ejected, then took on its nal
color over several minutes. It marked the pinnacle of Polaroids instant-photography success; a 1977
effort to introduce Polavision instant movies was a commercial failure.
Behind the scenes, Land was a pioneer in optical surveillance from aircraft and satellites. In 1952 he
served on a panel that recommended ying a spy plane at 70,000 feet over the Soviet Union to
photograph military facilities. He drew on that experience in 1954, when he was named to the steering
committee that proposed the U-2 spy plane, which performed exactly that mission, collecting the rst
reliable data on Soviet nuclear and missile activity. Land was among the scientists that President
Eisenhower assembled days after the 1957 Sputnik launch to discuss its implications. That led to Lands
involvement in the Corona series of photographic surveillance satellites, described elsewhere in this
book, which provided hard evidence that debunked the myth of a missile gap, a key step in stabilizing
Cold War tensions.
F
iber-optic imaging had a surprisingly long prehistory before its birth as an important
optical technology in the 1950s. One fundamental building block, the concept of light
guiding by total internal reection, was already well over a century old. A second, the idea
of image transmission through arrays of light guides, went back decades. But it took the
invention of low-index cladding to successfully launch ber-optic imaging and endoscopes.
Swiss physicist and engineer Daniel Colladon was the rst to describe light guiding by total
internal reection in 1842 [1]. He demonstrated the effect by illuminating a water jet, an
experiment later repeated by John Tyndall. French physicist Jacques Babinet noted that light
guiding could also be seen in bent glass rods, but he gave no details. Light guiding in water jets
helped light up the luminous fountains of the great Victorian exhibitions in the late nineteenth
century, and by the early 1900s, glass and quartz light guides were illuminating microscope slides
and the mouths of dental patients [2].
The late nineteenth century also saw the rst interest in remote viewing, or what we now
call television. Henry C. Saint-Ren, who taught physics and chemistry at a small French
agriculture school, realized that one way to transmit an image was to project it onto one end of an
array of thin glass rods so it could be viewed at the other end of the bundle. He recognized that
light would mix within each rod, so the rods had to be tiny to give a good image. In 1895, he
wrote to the French Academy of Sciences: The whole array gives a complete illusion of the
object if the diameter of each point does not exceed 1/3 millimeter when the viewer is at a distance
of one meter from the image [3]. The idea was simple and elegant but probably was impractical
at the time, and no further records of his work have been found.
In 1926, a British pioneer of mechanical television re-invented the concept. John Logie Baird
led a patent on a method to produce an image without the use of a lens by assembling an array
of thin transparent tubes. His patent also covered using thin rods or tubes of glass, quartz, or
other transparent material [which] could be bent or curved, or in the case of very ne quartz bers,
could be exible [4]. He tried to transmit images through an array of 340 metal tubes of 0.1-in.
diameter and 2-in. length but abandoned it in favor of spinning disks for mechanical television.
At almost the same time, a young American radio engineer and inventor named C. W. Hansell
thought of a new way to read instrument dials that were out of sight. In a notebook entry dated 30
December 1926, he outlined his plans for using a exible bundle of glass bers. When his employer,
the Radio Corporation of America, applied for a patent, he expanded on his original idea,
proposing to use ber bundles in periscopes, endoscopes, and facsimile transmission. Crucially, he
realized that the bers on the two ends had to be aligned in the same pattern to transmit the image
properly. The patent issued in 1930 [5], but by then Hansell had moved on to other ideas.
The rst person to make an image-transmitting bundle was a medical student named
Heinrich Lamm at the University of Munich in Germany. Lamm had studied with Rudolf
Schindler, who had developed a semi-rigid gastroscope that could be bent up to 30 deg. Lamm
thought a bundle of glass bers would be much more exible and persuaded Schindler to buy him
some glass bers from the Rodenstock Optical Works in Munich.
Lamm combed the glass bers so they lined up from end to end of the bundle and projected
an image of a lamp lament onto one end. In 1930 he recorded an imperfect but recognizable
53
image on the other end (Fig. 1). It was
enough to prove the principle, although
Lamm conceded that the images were not
bright or sharp enough to be usable. He
tried to apply for a patent, but the German
Patent Ofce told him that a British version
of Hansells patent had just issued.
Lamm described his experiment, but
could go no further [6]. The world was
sinking into the Depression, and soon
Lamm had to ee Nazi Germany. World
War II followed. The concept of ber
image transmission did not reappear until
around 1950when three people devel-
oped it independently, two of them well
connected in optics and the third an inde-
pendent inventor.
The postwar Dutch navy turned
to one of its leading optics specialists,
Abraham C. S. van Heel, to develop a new
type of periscope as it tried to rebuild its
submarine eet. The German optics indus-
try was in ruins, and neither the United
States nor Britain wanted to share their
Fig. 1. Heinrich Lamm, M.D., combed thin glass bers and periscope technology with Holland. A
packaged them in a short bundle (a), then focused the image professor at the Technical University of
of a light bulb lament (b) onto one end. The bers were well Delft, van Heel thought he could solve the
enough aligned to transmit a recognizable image of the lament problem by guiding light through thin rods
(c) to the other end. Both laments are shown in negative images.
(Courtesy of Michael Lamm, M.D.)
of glass or plastic. But his experiments with
bare bers initially got nowhere because of
light leakage and scratching.
In neighboring Denmark, engineer and inventor Holger Mller Hansen, like Hansell, wanted to
peer into inaccessible places. He thought of using a exible ber bundle to transmit images after looking
at insects segmented eyes. An avid experimenter, he rst tried drawing his own bers, then bought
some bers to test. He also discovered that light leaked between bers if they touched but realized that
he could solve that problem if he clad the ber with a material having a lower refractive index.
However, when he sought a material with index close to one, the best candidate he could nd was
margarine, which did not work well.
Meanwhile, in 1951, British optical physicist Harold H. Hopkins found his inspiration at a dinner
party where a physician discussed the horrors of trying to use a rigid endoscope [7]. Hopkins decided
that a bundle of exible glass bers could do a better job and applied for a research grant to support a
research student. When the money came through, he assigned the project to a young student from India,
Narinder Kapany.
Hansells patent had been forgotten and expired in 1947. But the Danish Patent Ofce found it
after Mller Hansen led his own application in 1951, and rejected the ling. With no support and no
luck in nding a good cladding material, he gave up and turned to another invention. With more
support, van Heel and Hopkins persevered.
When van Heel sought help with his ber periscope design, the Dutch government referred him to
Brian OBrien, OSA president in 1951 and director of the University of Rochesters Institute of Optics.
The two knew each other as leaders in the parallel worlds of American and European optics; at the time,
van Heel headed the International Commission on Optics. As it happened, OBrien had already been
experimenting with light guiding, and he recommended cladding the outside of the ber with a lower-
index material, so no dirt or scratches spoiled the total reection, and light could not leak out if bers
Introduction
Xerography, or electrophotography, was one of the great inventions of the twentieth century.
It was invented in 1938, 78 years ago, and remains in wide use today. The copier has become
a common presence in our workplace, and its availability is assumed. Prior to its invention an
ofce worker would type an original with sheets of carbon paper and copy paper sandwiched
behind it in the typewriter carriage. Legibility limited the number of copies that could be
made. If more copies were required, the typing process would be repeated or a master would
be typed and offset printing would be employed. The xerographic copier radically changed
all of that work and created a whole new communication chain between ofce workers
and their organizations with the multiple copies of a copy sharing the remarks of the
respondents.
Xerographys creation and application closely parallels the 100-year history of The Optical
Society. It was invented as a novel imaging system, which had no existing competitors. One of its
rst public demonstrations was at The Optical Societys Annual Meeting held in Detroit,
Michigan, on 22 October 1948 [1,2]. Although seen as highly novel, the observers could not
see the future value of the technology. That was not unusual: the leading industrial laboratories
of the time had previously been offered the opportunity for the development and commerciali-
zation of the technology, but all had declined [3].
It would be 1959, owing to the combined efforts of Battelle Memorial Institute and the small
company Haloid that would become the Xerox Corporation, when the Xerox 914 copier made
its phenomenal market introduction. It would take the efforts of the inventor Chester Carlson,
the Battelle Memorial Institute, and Xerox people over a period of 21 years to reach this 914
successand what a success it was! It is estimated that in 1955 before the introduction of the 914
about 20 million copies per year were made worldwide, largely by typing carbons. In 1964, ve
years after the introduction of the xerographic copier, 9.5 billion copies per year were made, and
in 1985 the number had grown to 550 billion [4]. The revenues of the small Haloid-Xerox
Corporation based on the 914 and the follow on products would grow at a 44% rate
compounded annually for the decade 1960 to 1970 to be greater than $1.5 billion. It was the
fastest sustained corporate growth rate in history up to that time.
When invited to write this brief chapter on the history of the invention of xerography, the
author was confronted with the question of what more can be usefully said that has not been
previously written. Two comprehensive books were published on the subject in 1965 by the key
early participants, namely, Xerography and Related Processes by John Dessauer and Harold
Clark [5] and Electrophotography by Roland Schaffert [6]. There are at least four other texts
[710] written by practitioners over the period 1984 to 1998 as well as numerous scientic
papers and popular press reviews in the same period written by scientists who researched the key
processes during the further development of the technology. The value the author brings is that of
an early participant in the decade following the introduction of the 914. These are the
observations of a young scientist joining Xerox in 1964 to work with the individuals from
Xerox and Battelle who created that rst product success.
57
Fig. 1 Six-step process.
The Invention
Xerography is a photoelectric imaging process that creates high-delity copies. It is distinguished for its
ability to image directly onto plain paper without the use of wet chemical agents, which were common
to silver halide and other sensitized paper photography. How xerography works is demonstrated in the
following six-step process (see Fig. 1.):
1. An insulator photoconductive sheet attached to an electrode substrate is uniformly electrostatically
charged.
2. The photoconductive sheet is imagewise exposed with light. The electrical conductivity of the
photoconductors exposed areas is greatly increased and the surface charges are discharged through
the photoreceptor, leaving a latent electrostatic image on the unexposed areas.
3. Pigmented polymer particles charged to the opposite polarity of the latent image are cascaded
over the surface. The pigmented particles are electrostatically attracted and tacked to the
charged image area, whereas the particles do not stick to the uncharged areas. The latent image
is now visible.
4. Plain paper is placed on top of the powder image, and a charge is applied to its back surface with
sufcient voltage to de-tack and transfer the image to the plain paper.
5. The plain paper is stripped away from the photoreceptor surface with the image.
6. The polymer toner image on the paper is fused by heat. The photoreceptor surface is cleaned and
readied for the next imaging.
This six-step process is the formulation of the basic Chester Carlson 1938 invention as led in
his patent application of 4 April 1939 and which was issued in 1942 [11]. The process has been so
robust over time that it still is the core design of all xerographic copiers and printers produced, 77
years later.
The rst commercial implementation of this process was the Xerox Model A processor introduced
in 1949 (Fig. 2). It was a totally manual operation where the operator carried out each of the above
process steps. As a new Xerox employee, the author was introduced to xerography with this machine by
working through all of the steps described above. The experience was reminiscent of an introductory
physics lab, interesting to the technically trained but bothersome for ofce workers.
The time between these products, 1949 to 1959, required intensive improvements by the Battelle
and Xerox teams in both process physics and materials. The major new challenge to realizing the
potential of this technology was the automation of the process steps requiring their systems integration
The Inventor
Chester Carlson by every measure is the
model for the aspirations of all indepen-
dent inventors: he created a great inven-
tion that had tremendous societal benets
as well as providing him with great per-
sonal wealth. He is the individual inven-
tors dream.
His story is compelling. He grew up
as an only child in a family of very limited
resources. In his early years he became the Fig. 2. Xerox Model A, 1949. (Courtesy of Xerox
sole provider for his parents. Living in a Corporation.)
suburb of Los Angeles, he worked his way
through two years at the Riverside Junior
College, from which he transferred to
California Institute of Technology for his
nal two years and graduated with a
degree in physics. He started his career
in Bell Labs in New York, but was laid off
during the depression. He became a pat-
ent attorney after attending the New York
University law school. It was his work as
an attorney that drove his sense of pur-
pose to nd a solution to the need for
copies.
Chester Carlson rst led a patent
application for his invention in October
1937 and reduced it to practice in October
1938 reproducing the image 102238
ASTORIA. At this time he had funded an
assistant, Otto Kornei, to help with the
Fig. 3. Xerox 914, 1959. (Courtesy of Xerox Corporation.)
laboratory work. This experimental pro-
cess was his basis for working out the basic six-step process that was the core of his invention. The
photoconductor they employed was amorphous sulfur, and they developed the image with dyed
lycopodium powder. Charging was done by rubbing a cloth imparting a triboelectric charge on the
sulfur lm and shaking the powder in a container to impart a triboelectric charge of the opposite sign.
The developed image was fused by heat from a Bunsen burner.
Chester Carlson would contact over 20 companies to try to establish interest in his invention over
the period of 1938 to 1944, with no success. In 1944 he had the opportunity to describe the invention to
Russell Dayton of the Battelle Memorial Institute. Dayton was visiting Carlson seeking counsel on an
unrelated patent matter and became interested in the idea. Battelle and a yet to emerge small company,
Haloid, would transform his ideas into a phenomenal success.
Working with both of these organizations, Carlson would relocate to Rochester, New York, and
make the Haloid (to become Xerox) labs his professional home. He would maintain an ofce there
through the 1960s. His original xerographic patent would expire in 1955 just before the introduction of
the 914. He actively protected his and Xeroxs interests by ling over 20 additional xerographic patents,
with the nal one granted in 1965. The author recalls his presence in the labs. He was a highly honored
gure for the new and growing research staff. He was very shy, so few knew him well personally. When
he was seen walking the hallways, on second glance he would be gone, a ghostlike gure.
Chester Carlsons wealth from his invention would reach $150 million. At the time of his passing in
1968, that amount would be worth over $1 billion in todays money. He spent the nal period of his life
giving away his wealth to causes that supported peace and social justice.
There was a benet to the expansion of xerography by offerings of new competition. Xerox
established through its relationship to Carlson, Battelle, and its own investments a patent position
that limited competitive offerings. This patent exclusion was set aside in 1974 by a consent decree
agreement with the U.S. Federal Trade Commission. It required that Xerox license to all competitors
its xerographic patents for period of ten years and any new patents issued in that interval. This
created an explosion of competitive offerings particularly from Japan.
An important advance in 1969 was the invention of computer-driven laser writing onto a
xerographic photoreceptor [14]. This opened a new market for xerography in electronic imaging
and printing. Xerox introduced the 9700 in 1977, which printed single-sheet, 300-spi (samples
per inch), single-sheet images at 120 pages per minute. Hewlett Packard introduced desktop laser
300-spi printing in 1984 working at eight pages per minute. Both products revolutionized their
respective market places. Most importantly, the application of xerography was transformed from its
analog imaging role to become part of the emerging digital imaging future.
Canon introduced the concept of a low-cost personal copier with a customer-replaceable
consumable cartridge in 1982. They creatively collected all of the high-maintenance elements of
the xerographic processes into a customer-replaceable unit, thereby removing the need for frequent
References
1. OSA Annual Meeting, Detroit, Michigan, 22 October 1948.
2. R. M. Schaffert and C. D. Oughton, Xerography: a new principle of photography and graphic
reproduction, J. Opt. Soc. Am. 38, 991998 (1948).
3. J. H. Dessauer, My Years at Xerox, The Billions Nobody Wanted (Manor, 1975), p 31.
4. D. Owen, Making Copies: at rst, nobody bought Chester Carsons strange idea, but trillions of
documents later, his invention is the biggest thing in printing since Gutenberg, Smithsonian Mag.,
August 2004.
5. J. H. Dessauer and C. H. Clark, Xerography and Related Processes (The Focal Press, 1965).
6. R. Schaffert, Electrophotography (The Focal Press, 1965).
7. M. Scharfe, Electrophotography Principles and Optimization (Research Studies Press, 1984), Vol. 3.
8. J. Mort, The Anatomy of Xerography, Its Invention and Evolution (McFarland, 1989).
9. L. B. Schein, Electrophotography and Development Physics (Springer, 1992).
10. P. Borsenberger, Organic Photoreceptors for Xerography (Marcel Dekker, 1998).
11. C. Carlson, Electrophotography, U.S. patent 2,297,691 (6 October 1942).
12. E. Gareld, Citation index for scientic information, Science Watch 4(2), 8 (1993).
13. J. M. Utterback, Mastering the Dynamics of Innovation (Harvard Business Press, 1996), pp. 2326.
14. J. C. Urbach, T. S. Fisli, and G. K. Starkweather, Laser scanning for electronic printing, Proc. IEEE 70,
597618 (1982).
15. M. Stolka, D. M. Pai, and J. E. Yanus, Imaging system with a diamine charge transport material in a
polycarbonate resin, U.S. patent 4,265,990 (5 May 1981).
16. M. Smith, M., C. F. Hackett, and R. W. Radler, Overcoating the photoconductive layer with a charge
transfer compound of aromatic polynuclear structure; xerograph, U.S. patent 4, 282, 298 (4 August 1981).
J
ames G. Baker contributed to optics, optical design, and, as this chapter describes, was a
pivotal player during the development and deployment of the U-2, and the optics of the U-2.
To briey mention some of his contributions outside of the U-2 is itself a challenge. He
graduated from Harvard in 1942 with a Ph.D. in Astronomy and Astrophysics, advised by
leading astronomer Harold Shapely, and went on to make innovative contributions for nearly 70
years including developing ray tracing and optical design code using the second largest computer
ever built (the rst one was delivered to Richard Feynman for the Manhattan Project). He not
only designed large format cameras for reconnaissance but also fabricated and tested the
large aspheric components personally. He is perhaps best known in the public for his design
of the BakerNunn tracking cameras and for designing and supporting the fabrication of
the rst freeform surface in mass production as part of the Polaroid SX70 camera, to name a
few examples.
This chapter features his work not only as the optical designer for the optics for the U-2, but
also his lesser known contributions as a leading member of the group that convinced then
President Eisenhower to authorize the U-2 program. The sources for this chapter were selected to
be as original as possible, and are dominantly CIA reports that were developed by the CIA
History Staff in the 1980s and released as classied reports within the CIA. These were later
declassied with redactions when the existence of the National Reconnaissance Ofce (NRO)
became known to the public in the late 1990s. All of the material in this chapter comes from
Bakers personal les that were made available to the author by the Baker family.
Bakers involvement in reconnaissance cameras began in 1941, when he was invited by
Major George Goddard to spend two months at the Wright Field in Dayton, Ohio [1]. Perhaps
the most succinct introduction to Bakers role in the U-2 and related programs is from an NRO
press release that announced the rst Pioneers of National Reconnaissance on 18 August
2000. The release states: James G. Baker, Ph.D.A Harvard astronomer, Dr. James Baker
designed most of the lenses and many of the cameras used in aerial over-ights of denied
territory, enabling the success of the U.S. peacetime strategic reconnaissance policy [2].
To write only on his technical accomplishments for reconnaissance cameras would overlook
a key role Baker played in bringing President Eisenhower to authorize the U-2 to carry the
camera. The rst section of the chapter will highlight Bakers roles in that arearoles that often
consisted of leading key technology committees, which led to the authorization of the U-2
program specically as described in [3]. In the context of the U-2 program, these roles began in
1951 with the establishment of what came to be called the BEACON HILL Study Group, named
for the location of the study group headquarters on Beacon Hill, in Boston. The group was made
up of chairman Carl Overage, a physicist at Kodak, Baker, Edward Purcell from Harvard, and a
total of 12 others that included Edwin Land of Polaroid, Richard Perkin of Perkin-Elmer, and
64
signicantly, Lt. Richard Leghorn from the Wright Air Development Command, who later became the
founder of ITEK where the CORONA program was developed in later years. This group
toured airbases, laboratories, and companies every weekend for two months in January and February
of 1952. From there the members invested three months preparing a classied document they presented
on 15 June 1952the BEACON HILL Report. The report, with 14 chapters, discussed various
technologies from radio to photography including infrared and microwave reconnaissance
systems. One of the key recommendations from the report was the need to develop high-altitude
reconnaissance.
Reaction to the BEACON HILL Report came a year later, in the summer of 1953, after Dwight D.
Eisenhower became president. The specic timing of the presidents interest was driven by an early report
of a new Soviet intercontinental bomber, designated Bison by NATO. This was a B-52 class bomber
(the B-52 was just entering production in the U.S.). This report was validated at the Moscow May Day air
show. In July of 1953, the Intelligence Systems Panel (ISP) was established, chaired by Baker, to advise
both the Air Force and the CIA on ways to implement the construction of high-ying aircraft and high-
acuity cameras. In parallel, during World War II (WWII), Baker had established a full-scale optical
laboratory, the Harvard University Optical Research Laboratory. After the war, Harvard asked that the
laboratory end its relationship with the university and it was moved to Boston University to become the
Boston University Optical Research Laboratory (BUORL), with the move funded by the Air Force. Baker,
however, elected to stay at Harvard where he continued to design lenses for use in photoreconnaissance.
BUORL was destined to become ITEK in 1957 under the leadership of Richard Leghorn.
At the rst meeting of the ISP on 3 August 1953 the discussion centered on the fact that the best
intelligence on the interior of the Soviet Union was based on German aerial photos taken near the end
of WWII. Discussions continued to review incremental modications that either were being attempted
or planned to create a high-altitude airframe from existing production aircraft. At the third ISP
meeting on 2425 May 1954, a critical outcome was to establish that to be successful, a high-altitude
aircraft would need to y above 70,000 feet, something that could not be achieved with modications
to existing airframes. The other pivotal event at this meeting was that the panel learned of a
lightweight, high-ying aircraft that was being developed at Lockheed Aircraft Corporation. Baker
dispatched a member of the panel to learn more about the project. The plane was conceived by the
now legendary Kelly Johnson, leader of the Skunk Works, who had designed essentially a single
engine jet powered glider, which was called at the time the Lockheed CL-282. On 24 September 1954
Baker convened the ISP panel to discuss the new airplane. The panel moved to support the CL-282,
but the Air Force, which had been aware of the CL-282, had already made a decision not to fund the
development of the aircraft.
Somewhat independently, on 26 July 1954, President Eisenhower commissioned another panel of
experts, led this time by James Killian, then the president of MIT. This panel had 42 of the nations
leading scientists, including Baker, segmented into three project groups. This group met 307 times over
nine months and included eld trips and conferences. Baker was a member of the Project 3 committee,
which was led by Edwin (Din) Land of Polaroid. Land believed the optimal committee size was one that
could t into a taxi and, as a result, this was a small group consisting of Baker and only a few others,
including notably mathematician John W. Tukey. In mid-August 1954, Land and Baker went to
Washington where Land was shown the details of the CL-282, after which he is quoted as having
phoned Baker to say, Jim, I think we have the plane you are after. Following a somewhat convoluted
path that was dominantly political and too lengthy to describe here, Land and Killian met directly with
President Eisenhower in November 1954 and the president directed that CL-282 be developed by the
CIA. Even with the presidents support, the competitive situation was complicated, but a key deciding
factor in the end was that Kelly Johnson promised to deliver the plane in eight months for $22 million,
which he did, under budget. A nal contract was signed on 2 March 1955 with Lockheed to deliver 20
planes between July 1955 and November 1956. To give some perspective on the priority of the project,
Richard Bissell of the CIA wrote a check to prestart the work and mailed it to Kelly Johnson.
With this background on how the U-2 airframe, a version of which is shown in Fig. 1 [3], came to
be authorized, this section will present Bakers work on some of the lenses that were considered or used
on cameras that ew on the U-2. This material is based on [4] and from the article written by Baker [5].
U.S. peacetime strategic reconnaissance cameras, 19541974: Legacy of James G. Baker and the U-2 65
Fig. 1. U-2R (World Air
Power Journal, Vol. 28, Spring
1997 published by AIRtime
Publishing, 10 Bay Street,
Westport, Conn. 06880).
To frame the challenge, the dominant aerial cameras that were used in WWII were the Fairchild
K-19 and K-21 framing cameras with focal lengths from 24 to 40 inches. In the period that the U-2 was
authorized, a typical ground resolution was 78 meters when ying at 10,000 meters. For the U-2, due
to the new objects of interest, there was a need for 3-meter ground resolution from >20,000 meters, or a
4 improvement. In the mid-1940s, Baker, working with Richard Perkin of Perkin-Elmer, had
developed a 48-inch focal length scanning camera that was installed in a B-36 that resolved two
white softballs on a green from 10,000 meters. However, this camera weighed more than a ton and the
weight budget for the U-2 was near half of this.
Baker began work on a radical new camera in October 1954, but quickly realized that it would
take more than a year to design, even with his computer access, whereas the plane needed a camera well
before this. Consulting with Richard Perkin, the decision was made to base the improved camera on the
Hycon K-38. This camera, with weight reduction implemented by Perkin-Elmer and improved optical
design developed by Baker in a few weeks, became the A-1 camera working at f/8 that was used in the
rst ights in mid-1955. A high-impact innovation at this stage was that instead of ying three cameras,
one down-looking and two oblique, Rod Scott of Perkin-Elmer developed a rocking mount to gather
the oblique and down-looking images with one camera.
As soon as there was a plan set for a camera to support the early U-2s, Baker began work on a
totally new concept, the B-camera. This was a 36-inch focal length f/10 lens with aspheric surfaces,
personally polished and tested by Baker. The use of aspheric lenses was essentially unheard of in this era
and is one of the reasons Bakers lenses set a new standard for high-acuity cameras. Developed in
Fig. 2. (a) Layout of a proposed Camera-C (this version at f/11), (b) an assembled Camera-C, 240 EFL, with a
nal conguration at f/12.
66 U.S. peacetime strategic reconnaissance cameras, 19541974: Legacy of James G. Baker and the U-2
collaboration with Rod Scott of Perkin-Elmer, the B-camera used only one panoramic imaging lens with
18 18 inch format frames. This lens, and variations on it, became a key component of all cameras
throughout the U-2 program.
Independently, Bakers concept for the ultimate U-2 camera, called the C-camera (see Fig. 2 [6]),
was a 240-inch focal length lens to be operated at f/20. However, in conversation with Kelly Johnson he
realized this format would never be small enough or light enough for the U-2. Eventually he developed a
180-inch focal length lens operating at f/13.85. While this design would typically have taken years to
complete in that era, his state-of-the-art computer allowed it to be completed in 16 days. However, in a
test ight of the Hycon manufactured lens, the conclusion was that the 5 longer focal length made the
lenses too sensitive to vibration. Apparently this result was never relayed to Baker, who learned of it
years later. When he learned of the source of the decision to not use the C-camera, he wrote a terse letter
stating he had solved that, should they have bothered to ask.
References
1. The New England Section of OSA, Highlighting past projects and camera systems in aerial
photography, 23 October 1997, from J. G. Baker personal les.
2. Press release from the NRO, 18 August 2000, from J. G. Baker personal les.
3. From U-2 The Second Generation, a reprint from World Air Power Journal (AIRtime Publishing, Spring
1997), Vol. 28, from J. G. Baker personal les.
4. G. W. Pedlow and D. E. Welzenbach, The CIA and the U-2 Program, 19541974 (Center for the Study
of Intelligence, 1998), pp. 1766.
5. J. G. Baker, The U-2 B-camera, its creation and technical capabilities, Proceedings of the U-2
Development Panel, The U-2 History Symposium, held at the National Defense University, 17
September 1998.
6. Photo from J. G. Baker personal les.
7. Personal correspondence between Bill McFadden (formerly of HYCON), and R. Cargill Hall, Air Force
History Support Ofce, on 10 September 1997, from J. G. Baker personal les.
8. R. Cargill Hall, The Eisenhower administration and the cold war: framing American astronautics to
serve national security, draft manuscript, 10 January 1994, from J. G. Baker personal les.
9. Personal correspondence between J. Baker and R. Cargill Hall, comments on the draft manuscript, 13
December 1993, from J. G. Baker personal les.
U.S. peacetime strategic reconnaissance cameras, 19541974: Legacy of James G. Baker and the U-2 67
19411959
Introduction
The full history of any scientic subject is impossibly complex, and any account can only be a
simplied one. Like other technologies, optical coatings developed over a broad front in many
countries with many workers and over a long time. Some discoveries were made and then
forgotten and rediscovered later; others were simultaneous but independent. This account is
intentionally heavily biased toward The Optical Society, and so, although we will try to retain
some breadth in the story, we will concentrate on those workers who were signicant in the
Society. Others, many of whom we will not mention, were also involved in and made signicant
contributions to the eld.
Beginnings
No one knows exactly when the technology of optical coatings started. As far as optical
instruments are concerned, the earliest was probably the simple mirror, and by 2000 B.C.
mirrors were common all over the world. Early mirrors were made from anything that could be
polished, and their reectance was simply that of the particular material. Obsidian, jade, bronze,
silver, or gold, even pots of water, were all used. The idea of using a coating to improve the
reectance was a later development. We know that the Romans employed many different
techniques for mirror manufacture, including some that we can classify as thin lms. Glass was a
common substrate. Mass production of cheap mirrors involved pouring molten lead over glass,
yielding irregular fragments that had somewhat raised reectance from the lead that stuck to the
glass, but quality was generally poor. Better, but more expensive, glass mirrors had lms of
mercury or gold leaf. Metal mirrors often carried layers of polished tin. Outstandingly clear glass
was developed in Murano in the middle of the fteenth century, and the production of what we
would describe as the rst modern mirrors followed soon after. These mirrors carried a coating
that was primarily a mercury amalgam of tin, although small amounts of other metals were also
sometimes added. Thus by the sixteenth century there was a well-established thin lm coating
industry but the coatings were solely of metals.
The development of interference coatings took rather longer. Of course, nature was rst in
the use of thin lm interference. Color in transparent thin lms must have been observed at a
very early stage of human development, but it was Isaac Newton who, in the late seventeenth
and early eighteenth century, painstakingly established the relationship between lm properties
and perceived color [1]. He realized that the same effects he saw in his thin lms were
responsible for many colors in nature and, mistakenly, thought that such effects were
responsible for all colors. Not much happened in thin lm optics from then until the beginning
of the nineteenth century.
Two major events in the early 1800s were the 1802 proposal by Thomas Young that light is
a wave [2,3] and the publication in 1810 of Goethes great book on color [4].
68
Young was not the rst to propose a wave theory for light, and, indeed, for a time the theory was
not generally accepted. It took the 1818 work of Fresnel on diffraction [5] to convince the eld. The
wave theory of light paved the way for the understanding of interference phenomena. Fresnel and
Poisson developed the idea of the absentee half-wave and the quarter-wave perfect anti-reection
coating [6]. By the end of the nineteenth century interference in thin lms was well understood, had
been recognized in nature, and was known to be responsible in the form of tarnish layers for an increase
in the transmittance of high-refractive-index lenses.
Goethes book contained in its rst edition a chapter by Seebeck, missing from subsequent editions,
dealing with experiments on precipitates of silver chloride where illumination was followed by
exhibition of reection of the very colors used for illumination. Wilhelm Zenker [7] realized that
this was an interference phenomenon that could be used in photography and also recognized that half-
wave spacing of repeated features should give high reectance at the corresponding wavelength.
Zenkers work was the precursor to the Lippmann emulsion that won Gabriel Lippmann the 1908
Nobel Prize in Physics.
Metallic reecting coatings had also considerably developed during the nineteenth century. Justus
von Liebigs [8,9] development of a wet chemical deposition process for silver in the middle of the
century had transformed the production of reectors of all kinds. Interferometers required beam-
splitters with semi-transparent reectors. Astronomy adopted mirrors constructed from stable glass
with silver coatings rather than the older, somewhat unstable, speculum metal. Sputtering was
sometimes used, but the general view was that it tended to distort the substrates and so it was not
much in favor. Then an important paper by Pohl and Pringsheim in 1912 [10] suggested a vacuum
process using what was then called distillation, but nowadays thermal evaporation, for mirror coatings.
A great advantage of this method was that with a substrate exhibiting a sufciently high quality of
surface nish, the coatings would immediately form a mirror of equal quality without any further need
of polishing.
At the beginning of the twentieth century, thin lm applications were largely in photographic
emulsions and in metallic reectors. There was as yet no real need for other kinds of optical coatings.
Also, strangely enough, although mirrors were much in demand, there seems to have been no great rush
to adopt Pohl and Pringsheims technique.
Early Efforts
The rst volume of the Journal of The Optical Society of America appeared in 1917. The second issue
(numbered as 2 and 3) contains two papers that would appear of great signicance to us today but,
from their citations, seem to have received little notice at the time. The rst paper, on what we would
now describe as an interference optical coating, was by Herbert Ives [11], who modied the treatment
of a Lippmann emulsion to produce a narrowband reecting lter of high efciency that we would
now call a notch lter. In the same issue Otto Stuhlmann [12] described his technique for depositing
metallic mirrors and beamsplitters by thermal evaporation from wire sources. Later, in volume 2,
Frederick Kollmorgen describes a spinning process for the protection by lacquer of silver lms,
solving the then current problem of applying a thin, uniform lm to protect silver surfaces from
tarnishing [13].
In the late 1920s and early 1930s, John Donovan Strong pioneered work in the deposition of
optical coatings. 1930 he joined the California Institute of Technology and teamed with Charles
Hawley Cartwright to investigate the deposition of an enormous number of metals and dielectrics [14].
By early 1931 Strong had coated, with quartz-protected silver, a 6-in. (15.24-cm) reector. The
following year he replaced the coating with one of aluminum. Aluminum had two great advantages. It
strongly reected the ultraviolet and it had an innate environmental resistance due, Strong was sure, to a
lm of oxide that naturally formed over the surface.
Meanwhile progress was being made in Germany and France. At the Carl Zeiss company in Jena
Alexander (Oleksandr) Smakula [15], of Ukrainian origin, developed an anti-reection coating for
lenses, which for several years remained a close secret and was used primarily for military applications.
War Years
Optical instruments of all kinds including binocular telescopes, submarine periscopes, range nders,
telescopic gun sights, and aircraft bomb sights were required for World War II. The performance of all
of these could be much improved, especially for use at dusk or dawn, by the addition of anti-reection
coatings. All the participants on either side in the war were involved in anti-reection coatings, yet they
were treated everywhere as highly secret.
Richard Denton joined the Frankford Arsenal in early 1942. In 1935 the staff consisted of eleven
people. By the end of 1943 the staff numbered 1100. His account of his experiences at the Arsenal [26]
paints a vivid picture of the rapid problem solving and innovation that was required by the needs of the
conict. Anti-reection coatings represented only a part of his responsibilities. Magnesium uoride had
been found most satisfactory, and soon virtually all optics were being coated to improve their
transmittance.
Postwar Years
Now, after the war, the subject expanded rapidly. Part of the reason was the impetus given by the war
effort to the eld. Many people were involved in optical coating and found it an attractive and
rewarding eld. But also optics was ready for it. Great improvements could be produced by coating
camera lenses. High performance could be obtained from reecting coatings, avoiding the unpleasant-
ness and unpredictability of the wet chemical processes. Interference lters could be made as easily for
one wavelength as another and had enormous energy grasp. Thin lm polarizers showed high efciency
without the need for expensive crystals. There were, of course, many military needs, but all of optics
was expanding. The chemical industry needed infrared instrumentation, and astronomy needed
telescopes and instrumentation and especially narrowband lters for increasing contrast of diffuse
nebulosities. Binoculars, photographic cameras, microscoscopes, surveying equipment, and naviga-
tional equipment all showed vastly improved performance with anti-reection coatings.
Optical coatings had developed in Germany during the war, and now the results were being
brought back to the United States. In 1946 Howard Tanner, who had been with the U.S. Naval
Technical Mission in Europe together with Luther Lockhart, both of the Naval Research Laboratory,
published a paper on some of the German anti-reection coatings. One of the coatings they described in
detail was a three-layer one based on a quarter-wave of intermediate index next to the substrate,
followed by a half-wave of high index and then nally a quarter-wave of low index. This gives high
performance over the visible region. It was further analyzed by Lockhart and Peter King [36], and the
idea of the half-wave layer that broadens the anti-reection performance has since appeared in coating
after coating and in many publications and patents.
Accurate calculation of the properties of coatings was of considerable interest, and a good number
of the contributions to the Journal at this time were theoretical and concerned optical property
calculation. Robert Mooney had two papers in the 19451946 volume of the Journal [38]. Antonin
Conclusion
Optics has long reached the stage where optical systems without coatings are unthinkable. Thin lm
coatings play a variety of roles. In many cases they enable optical components and systems better to
perform their function that may be quite different from that of their optical coatings. The anti-reection
coating improves transmission and reduces glare, but the function of the system might be to magnify
distant objects. Enabling applications were the main driver for optical coatings in the very early days.
Later, with the appearance of the narrowband lter and the thin lm polarizer, we begin to see
components whose critical performance is purely that of the thin lm system, thus extending the role of
coatings well beyond that of a purely enabling technology. By 1960 that extension of the role of optical
coatings was becoming clear.
References
1. I. Newton, Opticks or a treatise of the reections, refractions, inections and colours of light (The Royal
Society, 1704).
2. T. Young, On the theory of light and colours (The 1801 Bakerian Lecture), Philos. Trans. R. Soc.
Lond. 92, 1248 (1802).
3. T. Young, Experiments and calculations relative to physical optics (The 1803 Bakerian Lecture),
Philos. Trans. R. Soc. Lond. 94, 116 (1804).
4. J. W. von Goethe, Zur Farbenlehre (1810).
5. H. d. Senarmont, E. Verdet, and L. Fresnel, eds., Oeuvres completes d'Augustin Fresnel (Imprimerie
Impriale, Paris, 18661870).
6. Z. Knittl, Fresnel historique et actuel, Opt. Acta 25, 167173 (1978).
7. R. Gther, The Berlin scientist and educator Wilhelm Zenker (18291899) and the principle of color
selection, Proc. SPIE 3738, 2029 (1999).
8. J. Liebig, Ueber die Producte der Oxidation des Alcohols, Aldehyd, Ann. Pharmacie 14(2), 134144
(1835).
9. J. von Liebig, Ueber Versilberung und Vergoldung von Glas, Ann. Chemie Pharmacie 98(1), 132139
(1856).
10. R. Pohl and P. Pringsheim, ber der Herstellung von Metallspiegeln durch Destillation im Vakuum,
Verhand. Deutsche Phys. Gesell. 14, 506507 (1912).
11. H. E. Ives, Lippmann color photographs as sources of monochromatic illumination in photometry and
optical pyrometry, J. Opt. Soc. Am. 1(2-3), 4963 (1917).
12. O. Stuhlmann, The preparation of metallic mirrors, semitransparent and transparent metallic lms and
prisms by distillation, J. Opt. Soc. Am. 1(2-3), 78 (1917).
13. F. Kollmorgen, Protection of silvered surfaces, J. Opt. Soc. Am. 2, 1617 (1919).
14. C. H. Cartwright and J. Strong, An apparatus for the evaporation of various materials in high vacua,
Rev. Sci. Instrum. 2, 189193 (1931).
Introduction
Jeff Hecht
P
hysics as a whole boomed in the middle of the twentieth century, but optics remained a
seemingly sleepy backwater compared with hot elds such as nuclear physics, electronics,
and astronautics. Yet the seeds of two technological revolutions were growing quietly,
fertilized by the generous government research funding that had fueled the rapid expansion of
physics. One was the development of space optics for surveillance satellites, which in time would
stabilize the uneasy balance of nuclear power. The other was the birth of the laser, which brought
new excitement and ideas to optics.
The development of spy satellites was among the deepest of military secrets in 1960. The
effort had begun quietly in 1955, as military and intelligence ofcials realized that satellites
might offer a new window on the Soviet Unions nuclear activities. That priority grew more
important with the Soviet Sputnik launch in 1957, which both showed that spaceight was
possible and established the precedent that satellites above the atmosphere could y over
countries without violating their airspace. Advanced optics were as crucial to the effort as
rockets; without good optics, the satellites could not record images of the ground clearly
enough for intelligence analysts to interpret them. Just weeks after Sputnik, the U.S. started a
crash optics program called CORONA, described in this section by Kevin Thompson, which
eventually succeeded in lming Soviet nuclear activity from space, helping to ease nuclear
tensions. The Hexagon program that followed, described by Phil Pressel, built on CORONAs
success.
The laser was an outgrowth of a military program seeking higher-frequency microwave
sources that led Charles Townes to develop the maser, then to think of how to extend the
principle of amplifying stimulated emission to even higher frequencies. Laser light brought
dramatic new possibilities to opticsmonochromatic and coherent light that could be concen-
trated into a beam of energy.
Irnee DHaenens, who assisted Ted Maiman in making the rst laser, may have been the rst
to call the laser a solution looking for a problem, and it was a cute joke in the early 1960s. But
in reality the laser opened the door to solving a host of previously intractable problems. One
series of articles in this section tells of the development of new varieties of lasers, made from
gases, new types of solids, semiconductors, and organic dyes in solution. Another article tells
how companies began manufacturing lasers for others to use.
The laser also opened up whole new elds of endeavor, covered in other articles in this
section. The intensity of laser light revealed nonlinear effects that had previously been
impossible to observe. The coherence of laser light made practical a radically new form of
truly three-dimensional imaging called holography. Lasers offered precise new ways of
measurement, from remote sensing to ultra-precise metrology. Laser beams could cut or drill
materials, print words on paper or record data on optical disks, or read printed patterns to
automate checkout at stores.
Lasers soon launched whole new government programs, described in other articles in this
section. Concern about nuclear attack led to efforts to develop laser weapons that could destroy
targets at the speed of light, a program that would wax and wane with the arms race and progress
(or lack of it) in building high-power lasers until the present day. The lasers ability to focus
intense energy onto pinpoint spots led to research on laser fusion, both as a way to generate
79
energy and to simulate nuclear weapons. The lasers narrow linewidth and tunability led to efforts to
enrich isotopes, both for nuclear reactors and to make bombs.
And the echoes of laser ideas, stimulated in the early years of the laser revolution, also resonate
through the remaining sections of this history.
80 Introduction
19601974
A
lbert Einstein planted the seed that grew into the laser when he realized the possibility of
stimulated emission in 1916, the year The Optical Society (OSA) was founded. Experi-
ments in the 1920s conrmed the existence of stimulated emission, then called negative
absorption, but it seemed only a matter of academic interest. Russian physicist Valentin
Fabrikant in 1939 proposed using stimulated emission to amplify light but did not pursue the
idea at the time.
Charles Townes made the rst major step toward the laser at Columbia University in 1951
when he proposed isolating excited ammonia molecules in a resonant cavity so stimulated
emission could oscillate at microwave frequencies. In 1954, Townes and his student James
Gordon demonstrated the rst maser, shown in Fig. 1, a word he coined from microwave
amplication by the stimulated emission of radiation. Microwave masers soon became
important as high-frequency oscillators and low-noise ampliers.
With millimeter waves and the far infrared then vast terra incognita, the next logical step was to
develop stimulated emission at infrared and optical wavelengths. The key requirements were a
medium with energy levels that could be inverted to produce stimulated emission in the optical
band, a way to produce a population inversion, and a cavity in which the light waves could oscillate.
That took some serious rethinking, and in the summer of 1957 Townes began a systematic
analysis of how to build what he called an optical maser. In essence, he formulated the physics
problem that had to be solved to develop the laser. As part of his investigation, in late October
Townes talked with Gordon Gould, a graduate student under Polykarp Kusch, about optical
pumping, which Gould was using to excite thallium vapor for his dissertation research. Optical
pumping was new, and Townes thought it might produce an optical population inversion. The
two talked twice, then went their separate ways.
Townes enlisted the aid of his brother-in-law, Arthur Schawlow, who worked at Bell Labs
and had experience in optics. Schawlow proposed using a pair of parallel mirrors to form a
FabryPerot resonator for the laser. They initially considered using thallium vapor as the active
medium, but Schawlow decided potassium vapor was more promising, so they focused their
attention on that system, and also noted that solids could be optically pumped. Reviewers at Bell
Labs, where Townes was a consultant, urged them to analyze cavity modes, which they included
in their pioneering paper, Infrared and optical masers, in the 15 December 1958 Physical
Review [1], which laid the groundwork for early laser development.
They did not know that Gould had jumped on the idea earlier. At age 37, he was growing
impatient with his dissertation. Gould had worked with optics before, and within weeks after
talking with Townes he described a FabryPerot laser resonator in a notebook that he had
notarized on 13 November 1957, shown in Fig. 2. Filled with dreams of becoming an inventor,
he left Columbia, talked with a patent lawyer, and holed up in his apartment with a pile of
references to work out his plans for what he called the LASER. Gould had solved the laser
problem on his own, and in time he would develop an extensive catalog of potential laser
transitions. But neither he nor Townes and Schawlow were close to building a working laser.
They had the blueprint, but nding the right material was a serious problem.
Alkali metal vapors were attractive because they are simple systems easy to describe in
theory. They did not offer much gain, but they looked promising for a proof-of-principle physics
experiment. Townes thought it would make a good dissertation project, as the microwave maser
81
had been for Gordon, and put two of his
students, Herman Cummins and Isaac
Abella, to work on it.
Schawlow pursued optical pumping
of solids, a natural because Bell Labs was
deeply involved in solid-state physics.
Schawlow initially focused on synthetic
ruby, which was also being used in
solid-state microwave masers and was
readily available at Bell. However, the
spectroscopy of ruby discouraged him.
The red transitions which had looked at-
tractive turned out to be three-level transi-
tions terminating in the ground state, mak-
ing it hard to invert the population. More-
over, other Bell researchers had found that
the red emission was inefcient, so he
began looking for other candidates.
As word of the laser circulated around
Bell, others developed their own ideas. Ali
Javan proposed a novel scheme for excit-
ing a gas laser with an electric discharge in
a mixture of helium and neon. The helium
would absorb energy from the discharge,
producing an excited state with energy
Fig. 1. Townes and Gordon with ammonia maser. (AIP very close to a neon transition. Collisions
Emilio Segre Visual Archives, Physics Today Collection.) would excite the neon to a metastable
upper laser level, which would then emit
on a transition to a level well above the ground statea four-level system that looked attractive for
continuous laser emission.
Gould, meanwhile, had gone to work at a defense contractor, Technical Research Group Inc., to
support himself while working on his laser ideas. He had hoped to keep his ideas secret, but eventually
worked out a deal to share patent rights with TRG, which helped him develop a patent and write a grant
for research on building a laser. In early 1959, Gould and TRG president Larry Goldmuntz pitched
their proposal to the Advanced Research Projects Agency, then less than a year old and chartered to
explore daring new ideas. ARPA was so impressed that they approved a contract for $999,000more
than triple the $300,000 TRG had requested.
By then, publication of the SchawlowTownes paper had put the laser into public view, interesting
other researchers in trying to make one. The ARPA contract was serious money at the time, intended to
support efforts to demonstrate laser action in a number of media. Laser development was becoming a
race, but it would not be an easy one.
The rst public reports on laser experiments came at a 1518 June 1959 conference on optical
pumping at the University of Michigan. Worried that the Pentagon might classify all laser research, not
just its TRG project, Bell Labs management encouraged Javan to describe his work both at the meeting
and in Physical Review Letters. Javan reported some progress in understanding energy transfer in
helium-neon discharges in experiments he had begun with William Bennett. Gould described his ideas
and hinted at the size of TRGs military program but was vague on details. Meanwhile, Gould was
having trouble getting the security clearance he needed to work on the TRG project because of his past
involvement with communists.
September saw a meeting much better remembered, the rst Quantum Electronics Conference at
Shawanga Lodge in High View, New York. Sponsored by the Ofce of Naval Research, it was the rst in
a series of biennial meetings that became the International Quantum Electronics Conference. Only two
speakers at the 1959 meeting talked about lasers. Javan described the early stages of his helium-neon
References
1. A. L. Schawlow and C. H. Townes, Infrared and optical masers, Phys. Rev. 112, 19401949 (1958).
2. A. Javan, W. R. Bennett, Jr., and D. R. Herriott, Population inversion and continuous optical maser
oscillation in a gas discharge containing a HeNe mixture, Phys. Rev. Lett. 63, 106110 (1961).
3. T. Maiman, Stimulated optical radiation in ruby, Nature 187, 493 (1960).
4. J. Hecht, Beam: The Race to Make the Laser (Oxford, 2005).
O
ptics prospered along with other areas of physics and engineering as American research
universities grew after World War II. Military programs encouraged universities to
expand basic research, both in hope of developing new defense technology and to train
specialists for defense research at government agencies or defense contractors. Over the years
from 1938 to 1953, military support of university physics research soared by a factor of 20 to 25,
after adjusting for ination.
These programs provided both bright new ideas and bright people to help launch the laser
era in optics. The Columbia Radiation Laboratory, founded in 1942 at Columbia University to
develop new microwave tubes for 30-GHz radar, received $250,000 a year after the war from the
Army Signal Corps to continue microwave research in Columbias physics department. At the
time, that was enough to support a staff of 20 and nearly as many graduate students, as well as to
pay several faculty members over the summer. Charles Townes headed the radiation lab from
1950 to 1952, during the time he conceived of the microwave maser.
Military research dollars also produced new physicists. American universities had graduated
about 150 new physics Ph.D.s annually just before the war, and the number dropped steeply
during the conict. But from 1945 to 1951 the number of physics Ph.D. graduates doubled every
1.7 years, reaching about 500 per year, as shown in Fig. 1. Seeing where the jobs were, postwar
students concentrated on experimental physics. Engineering likewise boomed in the postwar
years, with 159,600 bachelors degrees awarded from 1946 to 1950, more than from 1926
through 1940.
Dwight Eisenhower had seen part of that growth as president of Columbia University from
1948 to 1951, but as President of the United States he cut military research spending in 1953, and
the number of physics Ph.D.s remained in the 500600 range through the 1950s. The cuts led
universities to scale down their programs. Boston University went further, shutting the optics lab
it had inherited from Harvard; veterans of that group became the nucleus of the Itek Corpora-
tion, founded in 1957 by Richard Leghorn with funding from the Rockefeller family.
Eisenhower changed course after the Soviet launch of Sputnik 1 on 4 October 1957 stunned
the American physics community, the Pentagon, and politicians. Fearing the U.S. was falling
behind in an arms race in space, his administration boosted funding for physics and engineering
research and education. The money brought quick results. The number of Ph.D.s graduating
from American universities rose exponentially from about 500 in 1960 to some 1600 in 1970,
faster than the growth of Ph.D.s in any other eld. The number of American universities offering
Ph.D.s in physics climbed from 52 in 1950 to 78 in 1960 and reached 148 in 1970. The number
of undergraduate degrees in physics also climbed, from 1000 in 1945 to a peak above 6000 in
1968. Engineering degrees also increased. The numbers reected both growth in overall college
enrollment and an increase in the fraction of students studying physics and engineering. It did not
include the Postwar baby boom, who started to graduate from college in 1968.
The arms race, the space race, fast-growing industrial labs, and a booming technology
industry created unprecedented demand, particularly for physicists. A 1964 report from the
American Institute of Physics found that in 1960 only 17,300 trained physicists were available to
ll some 29,000 physics-related jobs in the U.S. Its not clear how many of the excess jobs went
unlled or were lled by people lacking physics degrees, but the decit seemed formidableand
the gap was projected to reach 20,000 by 1970.
85
Fig. 1. Number of Ph.D.
physicists graduating from
American universities annually,
showing the dramatic postwar
boom and post-1970 decline.
( 2002 by The Regents of the
University of California. All rights
reserved [1].)
A tripling of government research and development funding from 1955 to 1965 helped propel the
boom, with defense and space programs leading the way. The birth of the laser and increasing military
use of electro-optics pumped up spending on optical research and development, and in 1962 OSAs
Needs in Optics Committee concluded that existing training programs could ll only a quarter of the
need for 3500 new optics specialists in the coming ve years [2].
Yet by the mid-1960s, the well-oiled machinery of growth had begun hitting serious bumps in the
road. Doubts were growing about Americas escalating involvement in Vietnam, and opponents were
raising questions about the presence of military research on university campuses. Budget watchers
worried that the country could not afford to continue pumping more money into basic research while
ghting a war. Pentagon auditors found that military spending on basic research yielded a disappoint-
ing return on investment and urged focusing narrowly on mission-oriented research and development.
Congress began pressing to cut military spending on basic research, and spending on new research
buildings was stopped in early 1967, forcing some creative nancing to build the new Optical Sciences
Center at the University of Arizona [2]. Congress complained that too much research money was going
to a few elite universities, and too little to other Congressional districts. Topping off the trend, the
Manseld amendment in 1969 barred Pentagon spending on research lacking direct military applica-
tions, although those restrictions were later eased.
Universities also began re-examining their military research policies, pushed by faculty and student
protests. In 1967 Columbia, an early hotbed of protests, divested its Electronics Research Laboratory,
which became the Riverside Research Institute. More would follow. Stanford in 1970 split off the
Stanford Research Institute, later SRI International, and in 1972 MIT divested its Instrumentation
Laboratory, which became the Charles Stark Draper Laboratory. The most important split for the
optics world probably was the University of Michigans 1972 divestiture of its off-campus Willow Run
Laboratories, the birthplace of laser holography and optical signal processing.
In retrospect, it should have been obvious that the rapid growth powered by the space and arms
races could not continue, but students recruited with promises of well-paying jobs were caught by
surprise. Recruitment advertisements, which had fattened campus newspapers at elite schools like
Caltech, began evaporating after 1967. Job fairs at physics conferences shrank. Only 253 jobs were
advertised at the American Physical Societys 1968 annual meeting, but nearly 1000 applicants showed
up, and over 1500 people received Ph.D.s that year. Two years later, 1010 job-hunters chased 63 jobs at
the APS April meeting. American physics had indeed reached a crisis by 1970, exactly when the 1964
report had predicted, wrote MIT historian David Kaiser [1]. But the crisis was a shortage of jobs rather
than of physics graduates.
Inevitably, graduate enrollment shrank, and the number of new physics Ph.D.s dropped from a
peak of 1600 at the start of the 1970s to about 1000 per year at the end. Physics research continued
growing, but at a much slower pace. One measure of research, the number of abstracts published each
year in Physics Abstracts, increased about 3% a year from 1971 to 1999only a quarter of the 12%
References
1. D. Kaiser, Cold war requisitions, scientic manpower, and the production of American physicists after
World War II, in Historical Studies in the Physical and Biological Sciences, Vol. 33 (University of
California Press, 2002), Part 1, pp. 131159.
2. S. Wilks, from the History of OSA (to be published).
3. M. K. Fiegener, Science and Engineering Degrees: 19662010: detailed statistical tables (National
Science Foundation, June 2013).
B
y all rights, gas lasers should have been discovered long before 1961, likely by accident.
Einsteins 1917 classic paper derived the relationship among spontaneous emission,
stimulated emission, and absorption, but only considered a system in thermodynamic
equilibrium (guaranteed not to oscillate). It remained only to ask: What if the system were not in
thermodynamic equilibrium? Yet despite countless experiments looking at the absorption of
radiation in gas discharge tubes (not in thermodynamic equilibrium), the rst gas laser had to
wait for Ali Javan of Bell Telephone Laboratories.
Fig. 7. An experimental
gas-dynamic CO2 laser
developed by AVCO
Corporation circa 1968. The
output was over 50 kW.
Summary
The two decades ending in the 1980s were the heyday of gas laser development. Today, the world of gas
lasers is much quieter, with only a few types remaining, with mostly carbon dioxide in the factory and
some excimers and argon ion lasers in ophthalmologists ofces.
A list of literature citations for the thousands of gas lasers implied by this chapter would be longer
than the chapter itself. The interested reader is referred to guides to that literature, such as [1].
Reference
1. R. J. Pressley, ed., Handbook of Lasers (CRC Press, 1971 and subsequent editions).
T
he narrow-emission bandwidth of laser light quickly attracted the attention of spectro-
scopists in the early 1960s, but that narrow linewidth came at a costthe wavelength
was xed. Laser researchers found that they could shift the xed wavelength somewhat
by applying magnetic elds to the laser, they developed tunable parametric oscillators, and
eventually they found a few laser lines that were tunable. But those arrangements were
cumbersome and their range limited. As a student in the mid-1960s, spectroscopist Theodor
Hnsch felt a sense of frustration that he had no way to tune lasers to wavelengths that were
interesting.
What spectroscopists really wanted was a laser that could be tuned across a broad range of
interesting wavelengths. The rst such tunable laser, the organic dye laser, was discovered by
accident in research on Q-switching ruby lasers. The rst Q switches were active devices based on
Kerr cells or rotating mirrors, but in early 1964 the rst passive Q switches were developed using
saturable absorbers. Later that year, Peter Sorokin at the IBM Watson Research Center showed
that certain organic dyes dissolved in solvents made simpler and more convenient saturable
absorbers.
After that success, Sorokin found himself with a large collection of dye compounds that had
been prepared for the saturable absorber experiments. The dyes had interesting properties
including strong uorescence, so he decided to try producing stimulated Raman scattering. He
red pulses from a big Korad ruby laser into a dye that had never been tested in Q switching. The
rst experiment produced a black smudge on a photographic plate, but it was late Friday
afternoon and he had to leave. Monday morning, 7 February 1966, he told his assistant Jack
Lankard they should try aligning a pair of mirrors with the dye cell before they red the laser
again. Jack came back from developing the plate with a big grin on his face. There was one place
in the plate that the emulsion was actually burnt, Sorokin later recalled. They knew it was laser
action because the bright line was at the peak of the dye uorescence
Word of their experiments traveled slowly; Sorokin chose to publish his results in the
March 1966 issue of the IBM Journal of Research and Development because he liked the
editor, but it was not widely read. That gave two other groups a chance to independently invent
the dye laser.
The idea of a dye laser came to Mary Spaeth, then at Hughes Aircraft Co., about the same
time Sorokin was working on his experiment. She recalls, I was sitting on my bed with my two
year old daughter on my lap, two months pregnant with my second daughter, and about 20
papers spread out in front of me. I had been studying dyes that had been used for many years
for photographic purposes. In particular, I was studying models for how they are excited and
how they transfer energy from one molecule to another in the photographic process. The
excited states of these dyes have a geometry very similar to their ground states, so they have
very strong absorption spectra. I suddenly realized that if a dye could be put in a suitable
solvent, you could have an enormous population inversion after illumination by a short-pulse
laser. It was just like the light bulb pictures you see in the funny books. Boing! There it was,
clear as day.
She also realized that because dyes have huge numbers of rotational states, they should have
a broad gain bandwidth, so that placing dispersive optics in a laser cavity with the dye solution
should allow wavelength tuning. But rst she wanted to try exciting the dye with pulses from a
94
ruby laser. It was not part of her job, so it
took her months to make arrangements to
pump dyes with a ruby laser in Dave
Bortfelds lab. As she sat epoxying a dye
cell together, Bortfeld entered the room
and threw a paper airplane at her. She
recalls, I looked at him to try to gure
out why he had done that. As I unfolded
the airplane, I found it was a copy of
Sorokins paper, which Bortfeld had
just spotted. She knew the dyes, so she
instantly realized what it was about. We
decided, what the heck, we were working
independently, and we continued on
our way.
Expecting the dye to emit at a wave-
length a little longer than 700 nm, she did Fig. 1. Peter Sorokin with the ashlamp-pumped dye
not set up a detector, guring she would be laser in 1968. (Courtesy of International Business Machines,
able to see the laser spot on a magnesium International Business Machines Corporation.)
oxide block. However she didnt see any-
thing. I was about eight months pregnant, I had trouble reaching the knobs on the oscilloscope, it was
7 in the evening, and I was very tired, she recalls. Bortfeld told her to go home, while he set up a
photodetector and tried again. He called later that evening to tell her it had worked.
In further tests, they changed dye cells and moved their optics and found the oscillation
wavelength of one dye changed from 761 to 789 nm when they tried cells from 8 mm to 10 cm
long, and mirror spacing from 10 to 40 cm. They sent a paper to Applied Physics Letters, which
received it 11 July 1966 and published it in the 1 September issue. It was the rst report to show that
dye laser wavelength could be changed, although it was not yet practical tuning. Spaeth did not get
the chance to explore tuning further. Hughes management had no interest in dye lasers, and she had a
difcult childbirth, so her immediate priorities became recovering and dealing with two small
children.
Fritz Schaefer wrote that his group at the Max Planck Institute in Germany was unaware of either
effort when they stumbled upon the dye laser while studying saturation in a different group of organic
dyes. A student was testing the effects of increasing the dye concentration by ring ruby pulses into
the solution, Schaefer wrote, when he obtained signals about one thousand times stronger than
expected, with instrument-limited risetime[s] that at a rst glance were suggestive of a defective cable.
Very soon, however, it became clear that this was laser action. They may have learned of Sorokins
work after submitting a paper on their results which Applied Physics Letters received on 25 July, two
weeks after Spaeths paper. (After revisions received by APL on 12 September, Schaefers paper was
published in the 15 October 1966 issue, citing Sorokins paper but not Spaeths.) Like Spaeth, they
reported wavelength changes, in their case arising from changes in dye concentration.
Sorokin soon demonstrated ashlamp pumping, shown in Fig. 1, which proved important because
it could pump dyes across a broader range of wavelengths than the ruby laser. In 1967 Bernard Soffer
and Bill McFarland at Korad replaced one cavity mirror with an adjustable diffraction grating to make
the rst continuously tunable dye laser. They tuned across 40 nm and also reduced emission linewidth
by a factor of 100. At last, spectroscopists had a broadly tunable laser, and they soon were busy
exploring the possibilities.
Triplet-state absorption in the dyes limited pulse duration to nanoseconds in those early pulsed
lasers, but in 1969 Ben Snavely from Eastman Kodak and Schaefer found that adding oxygen to the
solvent could quench triplet absorption. Snavely then teamed with Kodak colleagues Otis Peterson and
Sam Tuccio to develop a continuous-wave (CW) dye laser. They rst investigated prospects for
pumping with intense plasma light sources, then tried pumping with an argon-ion laser. That required
longitudinal excitation and liquid ow to keep the dye solution cool, deplete triplet states, and avoid
Remembrances of Spectra-Physics
David Hardwick
I
t was a cold February morning in Minnesotareally cold! The year was 1963 at the
Honeywell Research Center, and the author, only recently graduated from college, helped
some visitors bring in their product to demonstrate. Herb Dwight, one of the ve founders of
Spectra-Physics, and Gene Watson, their star salesman, had stayed overnight in Minneapolis and
left their laser in the back of a station wagon. When their Model 110 HeNe laser was brought
into the lab, steam was pouring off every surface, betting the change from below zero to room
temperature. The unit was turned on and, miracle of miracles, a sharp red 632.8-nm beam
emerged. It does not seem like much now, but the author was blown awayhaving only too
recently tried to build such a laser himself. With the optics of the time and his limited
understanding of the process, achieving the necessary alignment proved difcult indeed. And
here were these guys, tanned by the California sun and braving the frigid temperatures, showing
us pallid northerners in the depth of winter a commercial product that worked.
Some months later, convinced that he wanted to join the world of lasers, the author headed
west to join the company. Just before he set out, a call came in requesting that he stop at the JILA
lab in Boulder, Colorado, to demonstrate a laser to Dr. John Hall, a future Nobelist. That laser,
drop shipped to the author in Denver, did not work. It turned out that the power supply on
switch was not wired in and the author was too clueless to determine the problem. The next day
another laser arrived and was demonstrated to Dr. Hall and his staff, thus completing the
authors rst sales call.
Early Spectra-Physics lasers consisted of a tube lled with a HeNe gas mixture at a pressure
of a few Torr placed in an optical cavity with mirrors at either end and a power source, which
was radio-frequency (RF) coupled into the gas. Radio-frequency coupling avoided the necessity
of placing anodes and cathodes in the tube itself; cathodes available at the time quickly
deteriorated, and the tube would go from a healthy pink glow to a sickly bluedeath by gas
poisoning!
The Model 130 was introduced in 1963, a foot-long ten-pound laser that looked for all the
world like a lunch box complete with leather handle. Cost considerations demanded that DC
power be used instead of RF coupling. The tube was terminated with optical windows set at
Brewsters angle, and the confocal mirror cavity was protected from the outside world with
exible rubber boots. The problem was the cathodes were borrowed from neon sign
technology and were designed for use at pressures 10 that of the laser tube. These little metal
tubes, terminated with a ceramic disc and lled with some rare-earth oxide mixture, simply did
not last very long; the neon was quickly sputtered away, and a few-hundred-hour lifetime was
considered good. What to do?
The authors bosses, Arnold Bloom and Earl Bell, asked him to follow up on a paper by Urs
Hochuli of the University of Maryland in College Park describing aluminum cathodes for use in
HeNe lasers. This assignment led to the authors rst real project at Spectra. A visit to Hochuli
in College Park resulted in Spectras machine shop fabricating a few aluminum cathodes, tubes
a few inches long and an inch in diameter, allowing some HeNe tubes to be made. The results
were very promising. So promising, in fact, that in a few months, the neon sign cathodes were
abandoned and only aluminum cathodes were used. Some 50 years have passed, HeNe lasers
are still being manufactured, and to the authors knowledge aluminum cathodes remain the
standby.
97
That technology became the Model 130, which had quite a long life as a Spectra product. Early
devices delivered about 0.5 mW at 632.8 nm; they cost $1525, a solid value at the time, although today
a laser pointer producing much more power can be purchased for a few dollars. The Model 130 found
many applications, ranging from serving as a pointer in Arthur Schawlows lecture room to guiding a
gigantic borer with a ten-foot-diameter cutting face in a tunnel being drilled through a hillside in
Llanelli, Wales.
Spectra-Physics was a wonderful place to grow up in the laser world. The ve founders provided
leadership, presented real opportunities to those younger and dumber, and created an enjoyable work
environment. As an example, when it came time to crate the hundredth laser for shipment, work was
halted, a keg of beer was produced, signicant others were invited, and the factory oor witnessed a
party celebrating the event. Now, when millions of lasers in thousands of different congurations are
produced worldwide, it is fun to remember when coherent light was rare and customers clamored for
the rst chance to employ it in their experiments.
Spectra-Physics was also a place where the workdays seemed to run on foreverit was the
employees choice to work overtime, not a company demand. The author recalls ddling in his lab late
one night in 1964 when Earl Bell, a company founder, called out and asked him to come next door to
his lab. He had a three-meter-long, large-diameter laser tube attached to a vacuum system and tted
with various gas sources. As usual, he was experimenting with different gases to investigate their laser
potential. There was a very bright beam coming out of the tube and Earl asked what color it was. The
answer was obviousa very intense green! Earl said, I thought so but couldnt really tell as I am quite
color blind! Thus the author was the second person, after Earl, to see an ion lasera mercury-ion
laser. The gain was amazingEarl took a Kennedy half-dollar out of his pocket and held it in the mirror
position at the end of the tube, and the laser ickered on and off as he brought the mirror into
alignment.
After Earls discovery, Bill Bridges at Hughes built a pulsed argon-ion laser. Earl quickly followed,
and soon the continuous wave argon-ion laser, now ubiquitous, came on the scene. Spectra quickly
commercialized it with the refrigerator-sized Model 135 argon-ion laser and power supply. Only a few
dozen were made; they were RF-coupled, temperamental, and short-lived. The author remembers many
miserable days at a Paris university trying to coax usable power out of one of these monsters during the
dog days of August 1968, when all the more intelligent Parisians had left town for the seaside.
Spectra-Physics actively sought to sell their lasers in Europe from very early days. They employed a
salesman stationed in Switzerland who visited universities and company laboratories, selling many
large HeNe lasers at prices favorable to the company. However, there was a problem: European
countries had rm tariff barriers that greatly increased the costs of buying American lasers. The solution
was to set up manufacturing inside the tariff borders. When Herb Dwight asked if anyone was
interested in setting up such an assembly operation, the author quickly volunteered and, in a couple of
months, moved to Scotland with his small family to do so, choosing a site in Glenrothes Fife, just north
of Edinburgh. With the help of the Spectra team, friends of Herb at the local Hewlett-Packard factory,
Scottish government representatives and a host of others, Spectras rst Scottish-built Model 130 was
shipped three months later, in late 1967. During three years based in Scotland, the team demonstrated
and sold Spectra lasers throughout Europe, from nearby England to far-off Athens and north to
Stockholm. It was a great adventure!
Back to Mountain View, California, and the author had a new assignment to be product manager
for the Spectra-Physics Geodolite Laser Distance Rangender, working with Ken Ruddock, one of the
ve company founders. The Geodolite was based on a 25-mW HeNe laser that was amplitude
modulated at ve different frequencies while the return from the target was phase-detected. A one-inch
telescope broadcast the beam, and an eight-inch Cassegrain telescope gathered the return signal.
The team used the Geodolite for several ground-based and aerial applications, including ice
roughness measurement and wave height determination from various air platforms including a
Lockheed TriStar, Convair 990, and Douglas DC-3. For the author, it was the travel gig of a lifetime.
He was was in Barbados with the BOMEX project and a NASA team when Neil Armstrong landed on
the moon. Unfortunately, there was no live television feed to the island, so the team listened on the radio
and celebrated with the local brew! As an aside, the very next day Thor Heyerdahl pulled into
98 Remembrances of Spectra-Physics
Bridgetown Harbor after having been rescued from the failed Ra rafting attempt across the Atlantic,
and the team was there to greet him. Other remote sites visited with the Geodolite included Ireland, the
Shetland Islands, Hawaii, the north slope of Alaska, and Brazil. On the ground, the team used the
Geodolite to survey in the primary markers for the Batavia, Illinois, accelerator.
Ken Ruddock was a great director and a lot of fun to work with. The Spectra team was testing the
Geodolite in airborne applications using the open cargo bay of a rented DC-3 on a hot day ying over
the central California valley. Unfortunately, the plane was owned by a chicken raiser, who used it to
ship many thousands of baby chicks from his farm to customers located all over the western United
States. These chicks leave a powerful odor, which was endured for many ight hours, but there was
compensation: the team was on one of those ights the day Spectra-Physics became a public company.
Ken turned to the author and said, I think I have just become a millionaire!
The author also worked for Bob Rempel, a founder and our rst president. Bob was a Ph.D.
physicist by degree but a tinkerer and mechanical engineer in his heart. He had strong ideas as to how
products should be built and expected all those in his sway to follow his lead. The authors favorite
vignette about Bob was his deep love of the Allen head bolt. Such fasteners were used in every possible
conguration in all Spectra products. Of course, to use such a bolt, one needed to have the correct Allen
head driver on hand. Somehow they were never at hand, and this dearth of drivers drove Bob up the
wall. One day, in a t of pique, he showed up in the lab areas with many boxes of these small drivers
and scattered them loosely over every conceivable work surface. With a satised smile, he took his
leave, saying as he left, there, that should x the problem!
Life at Spectra-Physics was full, challenging, and instructional. The author worked at one time or
another for each of the ve founders. Though young and dumb, he was treated as an equal partner and
was generously given the right to make mistakes and the encouragement to contribute ideas and energy
to build a successful Spectra-Physics. The founders of Spectra-Physics are owed a debt of gratitude that
cannot be fully paid off.
Remembrances of Spectra-Physics 99
19601974
C
ompanies large and small began making lasers after Ted Maiman announced the ruby
laser. The big companies had large industrial research laboratories and the resources
needed to develop a new technology. The little companies, many formed after Maimans
report, had energy, enthusiasm, and exibility. Both would play important roles in the laser
industry.
Money, expertise, and military contracts gave some companies a head start. Hughes Aircraft
started with Maimans design, as well as an Air Force contract to develop laser radars and
rangenders. The much smaller Technical Research Group already had an ARPA contract to
develop lasers based on Gordon Goulds patent applications and were the rst outside group to
replicate Maimans laser. Bell Labs had a formidable laser research group. Other big companies
including American Optical, IBM, General Electric, Raytheon, Varian, and Westinghouse began
investigating lasers, with their own funds or with military contracts.
American Optical, Hughes, and Raytheon became important early laser manufacturers, but
most other big companies never made many lasers. As part of the AT&T regulated phone
monopoly, Bell Labs had to license its patents. GE, IBM, Varian, and Westinghouse focused on
other products.
A wave of small companies also set out to build lasers. Maiman left Hughes to found a laser
group at a short-lived company called Quantatron in Santa Monica. When Quantatrons backers
soured on lasers, Maiman founded Korad Inc. with investment from Union Carbide and key
people from Hughes and Quantatron. Lowell Cross, Lee Cross (no relation), and Doug Linn left
the University of Michigans Willow Run Laboratory in 1961 to establish Trion Instruments Inc.
in Ann Arbor to build ruby lasers they had developed while at Michigan. Narinder Kapany
added lasers to the product line of Optics Technology, which he founded in 1960 to make optical
bers and other optical equipment.
Several books and articles, listed below, tell about the early days of laser development. In the
essays that follow, two industry veterans recount their adventures as young men working in the
very young laser industry in the early 1960s.
Bibliography
1. J. L. Bromberg, The Laser in America 19501970 (MIT Press, 1991).
2. J. Hecht, Lasers and the glory days of industrial research, Opt. Photon. News 31, 2027
(2010).
3. T. Maiman, The Laser Odyssey (Laser Press, 2000).
4. R. Waters, Maimans Invention of the Laser: How Science Fiction Became Reality (CreateSpace
Independent Publishing Platform, 2013).
100
19601974
A
merican Optical (AO) entered the laser business early through its interests in optical
glass and optical bers. Elias Snitzer, whom AO had hired to work on ber optics, made
the rst glass laser in 1961 by doping glass with neodymium, drawing it into a long, thin
rod and cladding the rod with lower-index glass to guide light along the rod by total internal
reection, just as in an optical ber.
The author started at AO in 1962 as a technician working for the companys chief
metallurgist, George Granitsis, who was investigating potential use of lasers for welding. They
were in the same building in Southbridge, Massachusetts, as Eli Snitzer, so the author also was
assigned the task of testing new laser glasses for Eli. Everyone was excited about lasers, and the
author remembers AO putting out a press release touting that the company would become the
IBM of the laser industry.
Those were fun days. Glass was easier to make in large rods than other solid-state lasers, so
larger and larger powered lasers were made, such as the one Eli is working on in Fig. 1. When
Shiner worked in Elis laser lab, they had two big metal wastebaskets. One said Eli and one
said Bill. The ashlamps that pumped the glass lasers sometimes blew up, so when they
charged the power supplies for them, they put the wastebaskets over their heads in case the lamp
failed. When the lamps exploded, the glass would hit the metal wastebasket. These wastebaskets
were also the rst form of laser eye protection.
AO made the rst Sun-powered laser, using a huge mirror to focus sunlight onto a
neodymium-glass rod. AO produced the rst laser capable of ranging off the Moon with a
group from Harvard University, using a glass laser and an amplier. The company also had a lot
of early military contracts and for a time held the worlds record for producing the most energy in
a single laser pulse, 5000 J, which was classied at the time. The authors lab had glass lasers that
put out 1500 to 3000 J per pulse, and they had to pump the rod with many times that energy, as
the efciency was about 2% wall plug. The resulting heat caused thermal expansion that
sometimes blew up the glass rods. They also built the rst large glass oscillator-amplier systems
for KMS Fusion and the Lawrence Livermore National Laboratory to use in the rst laser fusion
experiments back in the late 1960s.
The author also did some early medical laser applications work with Dr. Charles Koester,
some of which in retrospect was rather weird. He worked with a doctor at the Delaware
Veterans Hospital who was working on a new procedure to stop ringing in the ear that was
plaguing Vietnam veterans. The standard procedure was to drill a hole to the brain with the
patient alert and knock out brain audio receivers until the ringing stopped. Many times more
brain tissue was destroyed than required. The laser application was to map the cochlea of the
inner ear with a ber laser to knock out the receptors rather than to knock out the receivers in the
brain. Monkeys were trained to respond to sound by pulling on a lever when they heard a sound
at a certain frequency to avoid receiving a slight shock. This technique thus established a map of
the threshold of sound as a function of frequency for the monkey. The side of the monkeys face
was shaved, the diaphragm was folded back, and the ber laser was inserted in the inner ear of
the monkey. The procedure was to locate the ber laser at a precise location and re it to
eliminate a receptor. In the cochlea the receptors are at a precise location as a function of
101
frequency. After the procedure the monkey
was tested to determine which receptor was
eliminated. Many times as the diaphragm
was removed to reach the inner ear, the
seventh cranial nerve would be damaged,
creating distortion of the monkeys face.
The experiments went very well and the
Veterans hospital called in the press. Photos
were taken of the doctor, the monkey, the
laser, and the author.
The author was very proud of his
contribution to the project; the photos
went out over the Associated Press wire.
When he came back to AO he was called
into the presidents ofce, and the author
thought he was going to be congratulated
for his contribution. Instead, he almost got
Fig. 1. Elias Snitzer with glass laser. (Courtesy of the Snitzer
family.) red. The company made eyeglasses, and
the company slogan was about products to
enhance and protect the physical senses: animal groups from all over the country were calling,
complaining about the photos showing the author with the poor monkey with a shaved head and
distorted face.
AO later bought a small company called Laser Incorporated in Briarcliff Manor, New York,
headed by Tom Polanyi, which had developed an industrial carbon dioxide laser. They moved the
personnel to Framingham, Massachusetts, and consolidated it with AOs laser group. However, like
most other large companies, AO found it hard to make enough money from lasers to generate a prot
and decided to close the laser division. At that time in June of 1973 the author was application manager
and Albert Battista was engineering manager in the AO Laser Division. The two of them teamed up and
purchased the business from AO and renamed it Laser Inc. They did quite well and grew sales to several
million dollars, making the company quite protable. In 1980 they sold Laser Inc. to Coherent, and it
became the most protable division of Coherent for the next three years.
This article was adapted from an interview by Jeff Hecht, 18 May 2012.
Solid-State Lasers
William Krupke and Robert Byer
16 May 1960 marks the beginning of the laser era, in particular the era of the solid-state laser.
On this date Dr. Ted Maiman and his colleagues at the Hughes Research Laboratories in
Malibu, California, demonstrated the rst ever laser, a ruby laser. The work leading up to this
event is described elsewhere in this section, and in more detail in Joan Lisa Brombergs The Laser
in America, 19501970, published in 1991 [1]. Ruby would be the rst in a large family of solid-
state lasers.
George F. Smith [2], a Hughes manager at the time, wrote the following: Maiman felt that
a solid state laser offered some advantages: (1) the relatively simple spectroscopy made the
analysis tractable, and (2) construction of a practical device should be simple. Maiman initially
considered making a gadolinium laser in a gadolinium salt, but soon turned to synthetic ruby, a
form of sapphire (Al2O3) doped with trivalent chromium ions, which he knew from his earlier
work on microwave masers.
Maiman resolved doubts about rubys quantum efciency, but producing a population
inversion was a problem because the laser transition terminated in the ground state. When he
calculated requirements for laser operation based on gain per pass and mirror reectivity, Smith
wrote, He concluded that the brightest continuous lamp readily available, a high pressure
mercury vapor arc lamp, would be marginal. A pulsed xenon ash lamp, on the other hand,
appeared promising.
Crucially, ruby offered a way to demonstrate the laser principle using commercially
available materials, a ruby crystal made for use in precision watches, and a helically coiled
ash lamp made for photography. Maimans success surprised many others working on the
laser. Looking back, Arthur L. Schawlow wrote, I was surprised that lasers were so easy to
make. Since they had never been made, it seemed likely that the conditions needed might prove
to be very special and difcult to attain. It was also surprising that the earliest laser was so
powerful [3]. He told Optics News [4], I thought if you could get it to work at all it might put
out a few microwatts or something like that, and here he was getting kilowatts.
Schawlow and others had realized the attractions of a solid-state laser, but had focused their
attention on continuous-wave (CW) lasers, which consisted of a four-level system, with the
lower laser level above the ground state. Maiman showed that pulsed operation could be easier
and could produce attractively high instantaneous power. His ruby laser was reproduced within
weeks at other labs, and use of his ashlamp-pumping approach quickly led to the demonstra-
tion of other solid-state lasers.
Peter P. Sorokin and Mirek Stevenson at IBM had been working on their own approach to
solid-state lasers at the IBM Watson Research Laboratory. In Sorokins words [5]: The most
valuable and stimulating aspect of the SchawlowTownes article [6] was the derivation of a
simple, explicit formula applicable to a general system, showing the minimum rate at which atoms
must be supplied to an excited state for coherent generation of light to occur. The formula showed
that this rate (actually a measure of the necessary pump power) was inversely proportional to the
longest time that uorescence from the excited state could be contained between the two cavity end
mirrors in the parallel-plate geometry proposed by Schawlow and Townes.
When Sorokin searched for suitable materials, he concentrated on those suitable for four-
level laser action. Fluorite (CaF2) looked attractive as host material because of its optical quality,
so he searched the literature for suitable emission lines from ions doped into CaF2. Looking back,
103
he wrote, It was strongly felt that a
suitable ionic candidate should display
luminescence primarily concentrated in a
transition terminating on a thermally un-
occupied state. It was also felt that there
should be broad, strong absorption bands
that could be utilized to populate the
uorescing state efciently with broad-
band incoherent light. These two require-
ments generally dene a four-level optical
pumping scheme.
His search found spectral data that
Fig. 1. Peter Sorokin and Mirek Stevenson adjust their identied two promising four-level sys-
uranium laser at IBM. (Courtesy of AIP Emilio Segre Visual tems in CaF2: trivalent uranium and diva-
Archives, Hecht Collection.) lent samarium. He and Stevenson ordered
custom-grown crystals of uranium- and
samarium-doped CaF2 grown by outside vendors, and started experimenting with them. Then hearing
Maimans results stimulated a change in course.
Sorokin recalled, We quickly had CaF2:U3+ and CaF2:Sm2+ samples still in hand fabricated
into rods with plane-parallel silvered ends, purchased a xenon ashlamp apparatus, and within a few
months time successfully demonstrated stimulated emission with both materials. The materials CaF2:U3+
and CaF2: Sm2+ thus became the second and third lasers on record. When cooled to cryogenic temperatures,
both systems operated in a striking manner as true four-level lasers. Threshold pumping energies were
reduced from that required for ruby by two or three orders of magnitude. Our demonstration of this
important feature stimulated subsequent intensive research efforts in several laboratories to nd a suitable
rare earth ion for four-level laser operation at room temperature. (See Fig. 1.)
Heavily-doped dark or red ruby (as opposed to the pink ruby used by Maiman) also has four-
level transitions, on satellite lines arising from interactions of chromium atoms. In 1959, Schawlow had
recognized the lower levels could be depopulated at cryogenic temperatures, but did not pursue it for a
laser at the time. He and others returned to the system, and in February 1961, after the four-level
uranium and samarium lasers were reported, Schawlow and G. E. Devlin [7] and, independently, Irwin
Wieder and L. R. Sarles [8] reported achieving four-level laser action in the satellite lines of dark ruby at
cryogenic temperatures.
The trivalent neodymium ion, Nd3+, rst demonstrated in late 1961, proved to be the preferred ion
for constructing a room temperature four-level laser. L. F. Johnson and K. Nassau at Bell Telephone
Laboratories [9] rst demonstrated laser emission on that line in a neodymium-doped calcium
tungstate crystal. In the same year Elias Snitzer at American Optical Company [10] reported achieving
similar room temperature laser action in neodymium-doped glass. Interestingly, Snitzers laser was in a
glass rod clad with a lower-index glassa large-core optical berbut the importance of that
innovation would not be realized for many years. Not until 1964 did J. E. Geusic (Fig. 2) and his
colleagues at Bell Laboratories [11] report robust room temperature laser action in neodymium-doped
yttrium aluminum garnet (YAG), the crystal destined to be the dominant solid-state laser material for
commercial and industrial laser applications to the present time.
Once rare earth ions were identied as a particularly fertile group of materials for near-infrared
and visible lasers because of their characteristically narrow-band uorescence transitions, an explosion
of demonstrations of optically pumped solid-state lasers ensued, beginning in 1963. Rare-earth ions
included the trivalent thulium, holmium, erbium, praseodymium, ytterbium, europium, terbium,
samarium ions, as well as divalent dysprosium and thulium ions; these ions were doped into a variety
of crystalline host materials. Z. J. Kiss and R. J. Pressley [12] give an excellent review of solid state laser
development up to 1966.
All of the early solid-state lasers described so far have relatively narrowband laser transitions
offering very limited spectral tunability. There also was growing interest in developing solid-state
lasers, preferably four-level lasers operating at room temperature, with broadband laser transitions
that would allow wide spectral tunability for scientic and commercial laser applications. The rst such
solid-state lasers were realized in 1963, when. L. F. Johnson, R. E. Dietz, and H. J. Guggenheim [13] of
Bell Telephone Laboratories identied divalent nickel, cobalt, and vanadium in magnesium uoride
crystals as four-level laser gain media for widely tunable lasers in the near-infrared spectral range. Peter
Moulton details the development of these and later tunable solid state lasers elsewhere in this section.
The ve or six years after Maimans successful demonstration were immensely fruitful for solid-
state and other lasers, recalled Anthony Siegman of Stanford University. The eld was just exploding.
And it turns out if you look into it, essentially every major laser that we have today had actually been
demonstrated or invented in at least some kind of primitive form by 1966 (OSA Oral History Project,
May 2008).
The latter part of the 1960s and the 1970s saw the identication of many new crystalline host
materials doped with rare-earth and transition metal ions, described by A. A. Kaminskii [14]. Over the
same periods, the most promising of these solid-state lasers were developed technologically and
industrialized.
The next seminal advance in the history of solid-state lasers was replacing the pulsed or CW
discharge lamps used to pump the rst generation of solid state lasers with emerging semiconductor
light sources, including light-emitting diodes (LEDs) and later semiconductor laser diodes (LDs).
Lamps are inherently broadband pump sources, generally spanning the whole visible spectrum, so they
can pump many different materials, but solid-state laser materials have distinct pump bands, so
inevitably much of the light would not excite the laser transition. In contrast, LEDs have bandwidths of
about 20 nm, and laser diodes of about 2 nm. Adjusting the mixture of elements in a compound
semiconductor can shift the peak emission wavelength to match many absorption lines, such as the
808-nm absorption line of neodymium. As long as a suitable pump band is available, this generally
increases coupling of pump radiation to the laser gain medium and signicantly decreases deposition of
waste heat in the gain medium. Generally, diode lasers are preferred for their higher efciency and
output power.
Diode pumping has a long history. In 1964 R. J. Keyes and T. M. Quist [15] reported transversely
pumping a U3+:CaF2 crystal rod with a pulsed GaAs laser diode, with the entire laser enclosed within a
liquid helium-lled dewar. M. Ross [16] was the rst to report diode pumping of a Nd:YAG laser in
1968, using a single GaAs diode in a transverse geometry. Reinberg and colleagues at Texas
Instruments [17] used a solid-state LED to pump a YAG crystal doped with trivalent ytterbium at
cryogenic temperatures.
Early progress in diode laser-pumped solid-state lasers was limited by the need for cryogenic
cooling and by the low powers of the diode lasers. It was not until 1972, nearly a decade after the
pioneering experiments, that Danielmeyer and Ostermayer [18] demonstrated diode laser pumping of
Nd:YAG at room temperature. Room temperature CW operation was rst demonstrated in 1976.
Powers of diode-pumped solid-state lasers increased with the powers of the pump diodes and with the
development of monolithic arrays of phase-locked diodes in 1978.
References
1. J. L. Bromberg, The Laser in America: 19501970 (MIT, 1991).
2. G. F. Smith, The early laser years at Hughes Aircraft Company, IEEE J. Quantum Electron. QE-20,
577584 (1984).
3. A. L. Schawlow, Lasers in historical perspective, IEEE J. Quantum Electron. QE-20, 558 (1984).
4. A. Schawlow, Bloembergen, Schawlow reminisce on early days of laser development, Optics News,
March/April 1983.
5. P. P. Sorokin, Contributions of IBM toward the development of laser sources1960 to present, IEEE
J. Quantum Electron. QE-20, 585 (1984).
6. A. L. Schawlow and C. H. Townes, Infrared and optical lasers, Phys. Rev. 112, 1940 (1958).
7. A. L. Schawlow and G. E. Devlin, Simultaneous optical maser action in two ruby satellite lines, Phys.
Rev. Lett. 6(3), 96 (1961).
8. I. Wieder and L. R. Sarles, Stimulated optical emission from exchange-coupled ions of Cr+++ in Al2O3,
Phys. Rev. Let. 6, 95 (1961).
9. L. F. Johnson and K. Nassau, Infrared uorescence and stimulated emission of Nd3+ in CaWO4, Proc.
IRE 49, 1704 (1961).
10. E. Snitzer, Optical maser action of Nd3+ in a barium crown glass, Phys. Rev. Lett. 7, 444 (1961).
11. J. E. Geusic, H. M. Marcos, and L. G. Van Uitert, Laser oscillations in Nd-doped yttrium aluminum,
yttrium gallium and gadolinium garnets, Appl. Phys. Lett. 4, 182184 (1964).
12. Z. J. Kiss and R. J. Pressley, Crystalline solid state lasers, Appl. Opt. 5, 14741486 (1966).
13. L. F. Johnson, R. E. Dietz, and H. J. Guggenheim, Optical laser oscillation from Ni2+inMgF2 involving
simultaneous emission of phonons, Phys. Rev. Lett. 11, 318 (1963).
14. A. A. Kaminskii, Laser Crystals, Vol. 14 of Springer Series in Optical Sciences (Springer-Verlag, 1981).
15. R. J. Keyes and T. M. Quist, Injection luminescent pumping of CaF2:U3+ with GaAs diode lasers,
Appl. Phys. Lett. 4, 50 (1964).
16. M. Ross, YAG laser operation by semiconductor laser pumping, Proc. IEEE 56, 19 (1968).
17. A. R. Reinberg, L. A. Riseberg, R. M. Brown, R. W. Wacker, and W. C. Holton, GaAs:Si LED pumped
Yb doped laser, Appl. Phys. Lett. 10, 11 (1971).
18. H. G. Danielmeyer and F. W. Ostermayer, Diode-pump-modulated Nd:YAG laser, J. Appl. Phys. 43,
29112913 (1972).
I
n 1958 Arthur Schawlow and Charles Townes [1] published a seminal paper suggesting how
to extend maser action to the visible spectrum to make a laser. Only two years later in 1960
Ted Maiman [2] made the rst working laser by exciting R-line emission of ruby with a
ashlamp. Shortly thereafter Peter Sorokin and Mirek Stevenson [3] reported a four-level laser in
uranium-doped calcium uoride, which had a much lower excitation threshold, and Ali Javan [4]
reported the helium-neon gas laser, which used radio frequency (RF) excitation.
All these lasers suffered from inherent shortcomings, they were large, bulky, and very
inefcient at transforming excitation energy into coherent light. Overcoming these difculties
would be crucial because most applications of lasers require compact, highly efcient devices.
Semiconductors offered the possibility of high efciency and compactness, but it was by no
means obvious how to make a semiconductor laser. Many people proposed ideas, but there was
no experimental work. John von Neumann was the rst to suggest light amplication by
stimulated emission in a semiconductor in an unpublished paper in 1953 [5], ve years before
Schawlow and Towness groundbreaking paper. Von Neumann suggested using a p-n junction to
inject electrons and holes into the same region to achieve stimulated emission, but the scientic
community was unaware of his idea. In 1958, months before Schawlow and Townes, Pierre
Aigran also proposed stimulated emission from semiconductors in an unpublished talk [6]. At
about the same time N. G. Basov, R. M. Vul, and Yu. M. Popov [7] made a similar suggestion.
None of these ideas led to any experiments, perhaps because they did not specify what
semiconductor or structure or electronic transitions to use.
M. G. Bernard and G. Durafforg [8] then put forth a condition for lasing when electrons
dropped from the conduction band to the valence band: the difference between the quasi-Fermi
level of electrons in the conduction band, EFn, and that of the holes in the valence band, EFp, must
be greater than the photon energy (EFnEFp > h). More to the point, Basov and co-workers [9]
suggested that recombining electrons and holes could produce stimulated emission. However,
their work attracted little attention because they said nothing about the crucial matter of which
semiconductor to use.
W. P. Dumke [10] in early 1962 pointed out that indirect semiconductors such as silicon and
germanium would not work as lasers because the gain from conduction to valance band
transitions is not sufcient to overcome the loss from free carrier absorption, which is intrinsic
to the material. In contrast, the gain for interband transitions in direct materials such as GaAs is
large enough to overcome the loss. That prediction has stood up until the present time,
notwithstanding the work of Kimerling and co-workers [11] who made a laser in Ge, which
was made quasi-direct by stress caused by epitaxial growth on Si.
By far the most inuential work leading to the GaAs injection laser was the observation of
interband emission from forward biased GaAs p-n junctions at 900 nm at room temperature and
at 840 nm at 77 K. This was rst reported at the March 1962 American Physical Society Meeting
by J. I. Pankove and M. J. Massoulie [12]. At the same meeting Sumner Mayburg and co-workers
[13] presented a post-deadline paper claiming 100% emission efciency of 840 nm radiation
from a p-n junction at 77 K. However, their evidence was indirectthat the light at 840 nm was
visible to the eye, indicating that it was very intense, and its intensity was linear with injection
107
Fig. 1. IBM scientists
observe electronic
characteristics of their new
gallium arsenide direct injection
laser. From left to right: Gordon
J. Lasher, William P. Dumke,
Gerald Burns, Marshall I.
Nathan, and Frederick H. Dill, Jr.
The picture was taken on 1
November 1962. (Courtesy of
International Business
Machines Corporation,
International Business
Machines Corporation.)
currentand less than totally convincing. At about the same time D. N. Nasledov and co-workers [14]
in the Soviet Union reported about 20% line narrowing of the radiation from a forward biased GaAs
p-n junction. It was an interesting result but was not stimulated emission.
A few months later in June 1962 R. J. Keyes and T. M. Quist [15] presented direct evidence of the
high efciency of the GaAs p-n junction light at the Durham, New Hampshire, Device Research
Conference. They measured light intensity as a function of current with a calibrated light detector and
found near 100% efciency for the conversion of electrical energy to optical energy. This work got wide
attention, with an acount published the day after the conference presentation in The New York Times.
The management at several industrial research laboratories took notice, and activity in GaAs emission
increased substantially.
It was barely four months later that laser action in GaAs was reported at four separate
laboratories within ve weeks of one another. The rst two reports were published simultaneously
on 1 November 1962. R. N. Hall, G. E. Fenner, J. D. Kingsley, T. J. Soltys, and R. O. Carlson [16]
from General Electric in Schenectady, New York, had a received date 11 days before M. I. Nathan,
W. P. Dumke, G. Burns, F. H. Dill, Jr., and G. J. Lasher [17] from IBM in Yorktown Heights,
New York (see Figs. 1, 2, and 3). The GE paper was more complete in that it demonstrated an
actual laser structure, shown in Fig. 1(a) of that paper (not reproduced here). The laser oscillated in
the plane of the junction and emitted coherent light from the polished end faces. On the other hand
the IBM paper reported line narrowing in an etched diode. One and a half months later two more
papers from different laboratories were published: N. Holonyak, Jr., and S. F. Bevacqua [18]
from General Electric in Syracuse, New York, and T. M. Quist, R. H. Rediker, R. J. Keyes, W. E.
Krag, B. Lax, A. L. McWhorter, and H. J. Zeiger [19] from Lincoln Laboratory in Lexington,
Massachusetts.
All four lasers operated at 77 K in a pulsed mode with a pulse length of about 100 ns and a
repetition rate of about 100 Hz, and the emission of three of them was about 840 nm. The GE Syracuse
work was different from the others in that the laser light was visible, near 660 nm, and the laser material
was a semiconductor alloy, GaPAs. It was remarkable in that the GaPAs material was polycrystalline,
but still recombination radiation was so efcient that it lased. The IBM group achieved full-edged
pulsed laser operation at room temperature and continuous operation at 2 K in short order as reported
in several papers in the January 1963 issue of the IBM Journal of Research and Development [2026].
A key advance of the IBM group was the rst use of cleaved ends of the lasers by R. F. Rutz and F. H.
Dill [27]. This greatly simplied the fabrication process.
The publication of the four papers from GE, IBM, and Lincoln Lab launched a tidal wave of
research activity on semiconductor lasers. Just about every industrial and government research
laboratory and many university laboratories initiated work in the area.
The threshold current density of early semiconductor lasers operating at 77 K was several thousand
A/cm2. The threshold current was so high that the laser could operate only under short (100 ns)
excitation. When the lasers [28] were cooled to 4.2 K, the threshold went down to less than 100 A/cm2
and the laser operated continuous wave (CW). As the temperature was increased, the threshold current
This was left to H. Kressel and H. Nelson [35], who in 1967 reported an AlGaAs/GaAs single-
heterojunction laser (structure shown in Fig. 1(b) from that paper [not reproduced here]) with its active
region in the p-type region of the GaAs. Because of the improved guiding and reduced absorption of the
AlGaAs the lasers threshold current density was 8000 A/cm2, a factor of two to three times lower than
the best homojunction lasers at the time. Shortly thereafter similar work was done by Hayashi, Panish,
Foy, and Sumki [36,37], who obtained a threshold current density as low as 5000 A/cm2. However,
these results were not good enough to obtain CW operation at room temperature.
Room-temperature continuous operation would take a further advance, namely, the double-
heterojunction laser, shown in Fig. 1(c) from that paper (not reproduced here), in which the large-gap
AlGaAs material is on both sides of the junction, providing better mode guiding and reduced loss on
both sides of the junction. The heterojunctions also conne the electrons and holes to a thin region,
yielding higher gain. The rst double-heterojunction lasers were made by Alferov, Andreev, Portnoi,
and Trukan [38] in 1968. These lasers had threshold current density as low as 4300 but were not yet
CW. In 1969 Hayashi, Panish, and Sumski [36] reported the achievement of double-heterostructure
AlGaAs/GaAs lasers with a threshold as low as 2300 A/cm2 [39] By the following year (1970) they had
reduced the threshold down to 1600 A/cm2 and obtained CW operation at room temperature [40].
Alferovs group (see Fig. 4) achieved CW room temperature operation at about the same time in a
stripe-geometry laser [41].
At this point it was clear that the semiconductor laser was a device with many important
applications. Research and development toward this end have continued and expanded since then.
N
onlinear optical effects were seen long before the laser was invented. In 1926, Russians
Sergey Vavilov and Vadim L. Levishin observed optical saturation of absorption when
they focused bright microsecond pulses to power densities of kilowatts per square
centimeter. Vavilov introduced the term nonlinear optics in 1944, and during World War II
Brian OBrien put saturation to practical use in his Icaroscope to spot Japanese bombers
attacking with the sun behind them. The bright coherent light from the laser opened new
possibilities.
Peter Franken (Fig. 1) realized them as he sat in packed sessions on lasers at OSAs spring
meeting in early March of 1961. His mind wandered as speakers droned about applications in
communications and eye surgery. Seeking something really unusual, he calculated the intensity
of a 5-kW laser pulse focused onto a 10-m spot. His answer was megawatts per square
centimeter, with electric elds of 100,000 V/cmonly three or four orders of magnitude below
the electric eld inside an atom.
I realized then that you could do something with it, Franken recalled in a 1985 interview
[1]. Further calculations showed the elds should be able to produce detectable amounts of the
second harmonic. Excited, he left the meeting and hurried back to the University of Michigan,
where he and solid-state physicist Gabriel (Gaby) Weinreich began planning an experiment. He
rented a ruby laser from Trion Instruments, a small Ann Arbor company that was the rst to
manufacture them, and got Wilbur Pete Peters to set up a spectrograph and camera for
measurements. Weinrich told him to re the laser into crystalline quartz, which can produce the
second harmonic because it lacks a center of inversion.
They needed a long time to get usable results. Alignment requirements were demanding, and
harmonic conversion was so inefcient that 3-J, 1-ms pulses containing about 1019 photons
yielded only about 1011 second harmonic photons. Nonetheless, their photographic plate clearly
showed the small second harmonic spot. They submitted their paper in mid-July, a little over
four months after the meeting, and it appeared in the 15 August Physical Review Letters
without the faint second harmonic spot, which an engraver had removed because it looked like a
aw in the photo [2].
Optical harmonic generation experienced a breakthrough in 1961. At that time, we were
all thinking photons, and you cant change the frequency of a photon, recalled Franken. But
working with Willis Lamb at Oxford University in 1959 had taught Franken that classical
electromagnetic wave theory applied to light, so he had realized that nonlinearities might
generate optical harmonics. The faint second harmonic spot that never made it into print
launched modern nonlinear optics.
Frankens results caught the eye of Joe Giordmaine, who just two months earlier had begun
exploring the effects of ruby laser pulses on various materials at Bell Labs. He began testing Bells
large stock of crystals left from World War II research and within a few weeks was seeing more
harmonic power than Franken had. When he tested crystals of potassium dihydrogen phosphate
(KDP) he was surprised to nd that second harmonic emission was not just in the direction
of the ruby beam, but in a ring centered on a different direction, and that the second harmonic
was many times higher at some angles than others. He had discovered the importance of phase
114
matching the fundamental and second harmonic
beams. It did not work in quartz, but it did in
birefringent crystals such as KDP. Bob Terhune
independently discovered phase matching at the
same time at the Ford Motor Co. Research
Laboratory.
At Harvard, Nicolaas Bloembergen (Fig. 2)
gathered John Armstrong, Peter Pershan, and Jac-
ques Ducuing to work on nonlinear optics after he
saw a preprint of Frankens paper. Armstrong and
Ducuing began experiments, and all four worked on
theory. Bloembergen wrote the differential equa-
tions describing harmonic generation, but solving
the nonlinear problems posed a formidable task.
The group spent several intense and exciting months
from July 1961 to early 1962, dividing the task
among themselves and working closely with
Bloembergen.
The result was a 22-page detailed analysis of
light interactions in nonlinear dielectrics, published
in Physical Review in September 1962 [3]. It was
by no means the last word, but it was a very complete Fig. 1. Peter Franken. (OSA Historical Archives.)
rst word, says Armstrong, whose name was rst
in alphabetical order. The codication of nonlinear
interactions including harmonic generation and
parametric conversion had a huge impact in the
young eld.
Meanwhile, experiments with high-power, sin-
gle-pulse Q-switched ruby lasers at Hughes Air-
crafts Aerospace group revealed an unexpected
nonlinear anomaly. In early 1962, Eric Woodbury
and Won Ng measured output power at several
hundred megawatts, far more than expected, when
they used a Kerr-cell Q-switch lled with nitroben-
zene. Puzzled, they did other experiments, but the
light nally dawned when measured power dropped
to the expected level after they inserted narrow-pass
lters centered on the 694.3-m ruby line. Further
measurements revealed unexpected light on three
near-infrared lines, the strongest at 766 nm, a
weaker one at 851.5 nm, and a barely detectable
line at 961 nm. The increments were roughly equal
in frequency units.
They reported what they thought was a new
type of laser action, but it was up to Robert
Hellwarth and Gisela Eckhardt of Hughes Re- Fig. 2. Nicolaas Bloembergen. (Photograph by
search Labs to suggest the infrared lines were Norton Hintz, courtesy AIP Emilio Segre Visual
coming from stimulated Raman scattering by the Archives, Hintz Collection.)
nitrobenzene in the Q-switch. Experiments quickly
conrmed that, and Hellwarth later developed a full theoretical model. It was a landmark discovery in
nonlinear optics, showing that light interacted with molecular vibrations to stimulate scattering at
Stokes-shifted wavelengths. Soon afterward, Terhune and Boris Stoicheff separately observed anti-
Stokes emission.
pulse [4]. In 1970 Bob Alfano and Stan Shapiro at GTE Laboratories in Bayside, New York,
demonstrated more frequency spreading in glass and crystals [5]. The higher the power, the broader
the bandwidth, and over the years the effect spread the spectrum enough to make white-light
supercontinua.
In 1973, Akira Hasegawa and F. Tappert took another important step, extending the concept of
self-trapping to describe optical temporal solitons in optical bers [6]. Nonlinear phase modulation and
dispersion interact such that pulse duration and frequency chirp increase and decrease cyclically along
the length of the ber, periodically reconstructing the original pulse. Hasegawa, Linn Mollenauer, and
others later showed that solitons could transmit signals through optical bers.
Modern nonlinear optics has come a long way from its roots, yet the fundamental groundwork
remains solid. To this day, every time I make a discovery in nonlinear optics, I look at [Bloembergens]
paper and hes done it, says Robert Boyd of Rochester. He put the whole eld together in 18
months. That feat earned Bloembergen the 1981 Nobel Prize in Physics.
Nonlinear optics is used in consumer products. Second harmonic generation turns the invisible
1.06-m line of neodymium into a bright 532-nm green beam. Its hard to believe you can buy these
things. If you think of whats inside, its just amazing, says Garmire. Harmonic generation also nds
cutting-edge laboratory applications, generating pulses of attosecond duration or with wavelengths in
the extreme ultraviolet or x-ray bands. Self-phase modulation together with mode locking produces
femtosecond pulses and frequency combs. The more we try to do with optics, the more we have to think
about nonlinearities. Like the laser that was essential to its birth and its applications, nonlinear optics
seems to be everywhere.
Note: This chapter was adapted from [7].
References
1. Peter Franken oral history interview, http://www.aip.org/history/ohilist/4612.html
2. P. A. Franken, A. E. Hill, C. W. Peters, and G. Weinreich, Generation of optical harmonics, Phys. Rev.
Lett. 7, 118120 (1961).
3. J. A. Armstrong, N. Bloembergen, J. Ducuing, and P. S. Pershan, Interactions between light waves in a
nonlinear dielectric, Phys. Rev. 127, 19181939 (1962).
T
he idea of holography came to Dennis Gabor while he was waiting for a tennis court on
Easter Day in 1947. Born in Hungary in 1900, Gabor had earned a Ph.D. in electrical
engineering from the Technical University of Berlin, then moved to Britain when Hitler
came to power. In 1947, he was working at the British Thoms-Houston Company in Rugby and
wondering how to improve the resolution of electron microscopes.
Waiting for his tennis match, he wondered how to overcome the imperfections in electron
optics that limited resolution. Why not take a bad electron picture, but one that contains the
whole information, and correct it by optical means? he recalled later. He rst thought of
illuminating an object with coherent electrons, so interference between electrons scattered from
the object and those not deected would record the phase and intensity of the wavefront. If he
recorded the interference pattern and illuminated it with coherent light, he thought he could
reconstruct the electron wavefront and generate a high-resolution image.
Lacking a way to record electron interference patterns, Gabor tried using light as a model,
although he had not worked with optics before. The best available coherent source at the time
was a high-pressure mercury lamp, but its coherence length was only 0.1 mm, and ltering it
through a pinhole left only enough light to make 1-cm holograms of 1-mm transparencies.
Nonetheless, he made recognizable holographic images in 1948 (Fig. 1), a dozen years before
Theodore Maiman made the rst laser.
Gabors report in Nature in 1948 [1] raised the possibility of three-dimensional (3D)
imaging, generating considerable attention, and helped him land a professorship at Imperial
College in London; but progress was slow, his design generated twin overlapping images, and the
short coherence lengths of available light sources limited imaging to small transparencies. By the
mid-1950s, Gabor and most others had largely abandoned holography.
The revival of holography grew from a completely independent direction: classied military
research on synthetic aperture radar launched in 1953 at the University of Michigans Willow Run
Laboratory. The following year, a young engineer named Emmett Leith who had studied optics
at Wayne State University began developing an optical system to perform Fourier transforms of
radar data collected by ying over the target terrain. He and Wendell Blikken started with incoherent
optics, but Leith later said many of their problems just melted away when they considered
coherent light in 1955. They did not need much coherence and they eventually found that focusing all
the light from a point source onto another point would sufce for radar processing.
In September 1955, Leith realized that the light waves diffracted from the data record were
replicas of the original radar signals converted to optical wavelengths. That led him to a theory that
mirrored Gabors wavefront-reconstruction holography but shrank the radio waves to optical
wavelengths rather than stretching electron waves to optical lengths. He knew nothing about other
research in holography until a year later, when he discovered a paper by Paul Kirkpatrick and
Hussein M. A. El-Sum in the Journal of The Optical Society of America (JOSA) [2].
Holography intrigued Leith, but the radar project kept him too busy to experiment until
1960, when Willow Run hired Juris Upatnieks as a research assistant in the optics group. Born in
Latvia in 1936, Upatnieks ed with his family when Soviet troops occupied Latvia in 1944. They
spent years as refugees in Germany before moving to the U.S. in 1951. He had a fresh degree in
electrical engineering from the University of Akron (Ohio) but lacked a security clearance, so he
could not work on the radar project.
119
Fig. 1. Dennis Gabors rst hologram. (Reprinted by permission from Macmillan Publishers Ltd.: Nature 1948.)
Another major imaging advance came was the invention of rainbow holograms by Steve Benton
at Polaroid in 1969. Seeking to make brighter images, he produced reection holograms that displayed
depth only in the horizontal plane, the only one in which our eyes see parallax. This allows the
hologram to diffract the whole visible spectrum, spread across a range of angles to produce a rainbow
of colors. Easily visible under normal lighting, such holograms can be embossed onto metal lms and
they have become the most widely used holograms.
In the early 1970s in San Francisco, Lloyd Cross developed a variation on rainbow holography that
offered an illusion of motion. He produced the holograms in a two-stage process. First, he took
conventional photographic transparencies as he moved around a person or object, and then he recorded
rainbow holograms of the series of transparencies as successive narrow stripes on lm. Finally, the lm
was mounted in a 120-deg arc or a 360-deg cylinder.
The viewers eyes saw different frames, giving the parallax that the brain interprets as depth. If the
model moved between frames, a viewer saw the movement while moving around the curved hologram.
Cross formed a company called Multiplex to make the holograms; the best known one shows Pam
Brazier blowing a kiss to the viewer (Fig. 4).
In October 1971, when the holographic imaging boom was in full ower, Dennis Gabor received
the Nobel Prize for his invention and development of the holographic method. Many in the optics
community felt that Leith and Upatnieks should have shared the prize for reviving holography with
lasers and their solution of the twin image problem.
In his book Holographic Visions [8], science historian Sean Johnston blames George W. Stroke,
who in 1963 started a holography program on the Michigan campus that came to compete with Leiths
work at Willow Run. Stroke eventually left Michigan carrying a grudge and claiming that his work was
more important. This was long a common view in the optics community.
However, in her dissertation on the history of holography written at Cambridge University [9],
holographer Susan Gamble argues that the problem was that Leith and Upatnieks worked at a military
lab. Michigan students had protested Willow Runs military projects, and in 1971 opposition to the
Vietnam War was widespread in Europe. The Nobel committee may well have decided that awarding a
Nobel Prize for military work would send the world the wrong message.
If some optical Rip Van Winkle from 1970 woke up today after his long nap, he might ask,
Whatever happened to holography? Holographic imaging never came to movies or television, and
the holographic telepresence of convention speakers is based on the old Peppers Ghost illusion
rather than real holograms. Yet holographic displays have found some specialized niches. Furthermore,
holograms are used in industry in many ways that go unrecognized, such as holographic optics, and
security imprints on packaging and some currencies. We may never watch wide-screen movies in
glorious holovision, but who would have expected us to be carrying holograms in our pockets on credit
cards?
Note: This chapter is adapted from [10].
I
n the 100 years of OSA, laser technology has played a part for more than 50 years and
industrial laser materials processing has played a part for more than 40 years. This capsule
view presents the highlights of these years.
Prior to 1970, a handful of commercial laser suppliers, located mostly in the United
States, attempted to satisfy requests from a number of industrial manufacturers that showed
an interest in the possibility of a laser materials processing solution to a unique production
problem. A 1966 publication stated, This year will mark the beginning of an accelerated
growth for lasers. Many of the early problems involved in their use are nearing solution. In
the commercial markets, the applications will center on welding and other high-power CO2
and neodymium YAG (yttrium aluminum garnet) lasers : : : [1]. Interestingly, this otherwise
optimistic report ended with the statement, The markets for lasers will gradually develop
over the next few years, but they are not nearly as imminent or as large as is frequently
quoted.
One reason behind this disparity may be found in the premise that the laser was born fully
grown, a view held by many who read about the amazing possibilities for this powerful energy
source, as evidenced by the commonly quoted line that lasers are a solution looking for a
problem [2]. Industrial manufacturers that approached these scientic laser companies were
from many different industries: glass, with interest in cutting at plate glass [3]; mining, with
interest in rock drilling [4]; packaging, with interest in cutting steel rule dies [5]; aircraft engines,
with interest in processing turbine engine components [6]; sheet metal cutting [7]; paper, for
cutting and slitting paper [8]; and microelectronics, with accelerating interest in trimming
resistors and printed circuits [9] and cutting/scribing ceramic substrates [10]. Of these, only
the latter two advanced to widespread industrial utilization stages in the late 1960s, pushed by
soaring growth in the microelectronics industry. The others, all technically good applications,
languished for a few years, fullling the prophecy cited above, as the laser suppliers struggled to
develop devices with more power or better beam quality with improved reliability and mainte-
nance procedures.
The most economically successful applications drawing attention from a wide segment of the
worlds media were the use of a CO2 laser beam to cut woven fabric for made-to-order mens
suits [11] and the use of a pulsed ruby laser beam to drill holes in diamond dies used as wire
drawing dies [12]. The latter was the rst industrial laser processing machine to be exhibited in
the Smithsonian Museum in Washington, D.C.
While technical and economic cases can be built to explain the slow commercial success of
the laser as a manufacturing process tool, widespread implementation of laser processes was
inhibited to a degree by published articles. These articles were headlined, for example, Death
rays benet mankind, a phrase that can be attributed to a number of journalists searching for
attention-grabbing headlines in the early 1970s. Implementation was also stalled because of the
unfortunate labeling, by engineering societies and the U.S. government, of laser processing
systems as a nonconventional materials processing technology.
One anecdote that illustrates the former is this authors personal experience. While
negotiating the purchase of a high-power CO2 laser welding machine by a Fortune 500 company,
he was startled to hear a company ofcial sanction the purchase because he was impressed by
successful laser cataract surgery performed on his brother-in-law.
124
Thus, the industrial laser suppliers of the early 1970s were faced with an additional selling
burden, easing the concerns of uninformed, risk-wary buyers, and reassuring potential buyers that
their lasers were reliable and safe. A common selling tactic was to identify a laser champion as the
potential customer and to educate this person to be an inside sales advocate. Many of these
champions became laser industry advocates through their willingness to publish complimentary
articles.
Overcoming the nonconventional tag took many years [13], and it was not until the late 1980s that
this sobriquet was dropped by those charged with producing industry statistics. The 1970s, a period
that saw the blooming of several industrial laser suppliers, is considered by most analysts to be the
beginning of the industrial laser market, with annual revenues for laser sales ramping from $2 million to
$20 million in the rst decade of the market, an almost 26% compound annual growth rate (CAGR).
Several applications drove this growth: thin gauge sheet metal cutting [14], microelectronic package
sealing [15], cooling hole drilling in aircraft turbine engine blades and vanes [16], steel-rule die board
cutting [17], and semiconductor wafer dicing [18]all applications that continue successfully today.
An interesting footnote to the early beginnings of the industrial laser material processing era is that
these applications, and many that rose to prominence later, were accomplished using lasers that can best
be called industrialized scientic lasers, which were controlled by analog programmable controllers
or tape reader numerical control (NC) devices. MIT scientists developed numerical control for
machining in the 1970s, and it became commonly used in the 1980s. This technology was a major
contributor to the growth of lasers for industrial material processing applications. The evolution to
computer numerical control (CNC) [19] and the industrial development of minicomputers in the 1980s
and the microprocessor in the 1990s vaulted the industrial use of lasers to annual growth rates in the
mid-teens.
Through the 1980s and 1990s, solid-state lasers led by Nd:YAG devices and ultra-reliable low-
power, sealed-off CO2 units remained the backbone of the industrial laser materials processing
industry. On a smaller scale, excimer lasers were used mostly in semiconductor processing [20] and
metal [21] and non-metal applications in the manufacture of medical devices. These lasers had evolved
from the scientic designs of the 1970s into ruggedized, reliable, low-maintenance products that were
being integrated by system manufacturers into material processing products acceptable to a broad
range of global consumer product manufacturing companies.
The utilization of industrial lasers, very much advanced in the U.S. in the rst two decades of the
technology, was due in great part to the marketing prowess of domestic equipment suppliers. This is
counter to some international views, mainly in Europe, that the U.S. government, through the
Department of Defense (DOD), funded the development of the laser products that were being used
in commercial industrial applications. In reality, the industrial laser and systems suppliers of the 1970s
and 1980s were essentially a part of a bootstrap industry, self-funded in terms of equipment and
applications development. What little funding owed from the U.S. government through its DOD
Manufacturing Technology programs was focused on laser applications that could improve or repair
defense products. In part, this lack of a national initiative to support progress in manufacturing
stultied the growth of the industrial laser economy.
Stepping into the void left by this modest industrial laser program, the government of Japan in the
1980s and Germany (supported in part by the European Union) undertook university-based efforts to
understand and improve the laser beam/material interaction on a broad range of materials. In Japan,
most of the effort focused on dening and improving the process of laser cutting sheet metals [22],
specically stainless steel, at that time a major industry in that country. As a result, increased output
power from new types of CO2 lasers, improved gas-assist nozzle signs, and purpose-built cutting
systems entered the market from a number of suppliers, rst in Japan to a large number of custom
cutting job shops and then exported to the international markets. In addition to this effort, the
Japanese government funded a major program for exible manufacturing, which had as a part the
development of a very-high-power CO2 laser that vaulted the selected supplier to the top of the CO2
power chain.
In the late 1980s, almost concurrent with the laser cutting development in Japan, European CO2
laser suppliers [23] made efforts to expand their markets by improving their product lines. This
Introduction
It is not an overstatement to say that barcodes are nearly everywhere you lookvirtually every
product you purchase at a supermarket, hardware store, liquor store, book store, or elsewhere
carries a universal product code (UPC) barcode printed on the package or an attached label.
Most package delivery services, including Federal Express, UPS, and the United States Postal
Service, use barcodes on packages for tracking purposes. As a consequence we can track whether
the book we ordered from Amazon has shipped, and at any time we please, know where our
book is on the route from Amazon to our front door.
Barcode scanners are equally ubiquitous. Scanners are at most check-out counters where we
shop. Some of us even carry a barcode scanner with us wherever we goin the form of an app on
our smartphone. One smartphone app can build a grocery shopping list by simply scanning
barcodes on empty packages before they go into the recycling bin.
This article provides an illustrated overview of the history of barcode scanning, beginning
with the development of the various barcode symbologies, and following through the develop-
ment of the scanning devices used to read the barcodes. Since the barcode industry has been very
competitive, little information was published in technical journals. Inventions were either
patented or treated as trade secrets. This article will illustrate the history of barcode scanning
based on key patents issued in the eld. Figure 1 illustrates by year the number of patents issued
that include either of the terms barcode or bar code. Issued barcode patents rose from a
trickle in the early 1980s to a high of 265 patents in 2003.
Barcode Symbologies
The rst mention of encoding information into printed dark bars and white spaces was disclosed
in U.S. patent 1,985,035 submitted by Kermode, Young, and Sparks in 1930. The patent was
ultimately issued on 18 December 1934 and assigned to Westinghouse. The invention described a
card sorting system for organizing electric bill payments by geographic region, thus simplifying
the work of accurately tabulating customer payments.
The rst true barcode was a circular bullseye symbol invented by Silver and Woodland
(see Fig. 2). The two disclosed their invention to the U.S. Patent Ofce in 1949 and their patent,
numbered 2,612,994, was issued on 7 October 1952. The patent contained claims covering a
circular bullseye symbol on an item and an apparatus to read the symbol.
In the late 1960s a group of supermarket chains began to realize efciencies could be gained
with a more automated checkout process. Several checkout methodologies were formulated and
subsequently studied resulting in a recommendation to adopt an 11-digit product identication
code. This effort ultimately resulted in the formation of the UPC Symbology Committee in March
1971. The committee was charged with selecting a symbology concept and providing a detailed
specication for the selected symbology. The Symbology Committee also worked with suppliers
of optical readers for the selected symbology.
The symbol ultimately adopted was the UPC symbol found on most products today, as
shown in Fig. 3. In the U.S. the leading digits of a symbol, which identify a manufacturer, are
128
licensed by GS1 US, a private rm respon-
sible for maintaining the assignment of
manufacturers identication numbers.
The following ve digits are assigned by
a manufacturer for each product it pro-
duces. The nal check sum digit is used to
ensure the data integrity of the scanning
and decoding processes.
Numerous other symbologies have
been developed over the years for other
applications ranging from inventory con- Fig. 1. Number of patents issued (including either of the
trol through military logistics to package terms barcode or bar code).
tracking by delivery companies. Some of
these, such as Code 3 of 9 (aka Code 39)
and Interleaved 2 of 5, are purely numeric
codes. Others, such as Code 93 and Code 128,
are full alphanumeric codes. Examples of these
one-dimensional (1D) symbologies are illustrated
in Fig. 4.
The need for labels containing ever-increasing
amounts of data led to the development of stacked
codes and two-dimensional (2D) codes. A complete
discussion of these higher information density
symbologies is beyond the scope of this article.
Examples of higher information density 2D symbol-
ogies are shown in Fig. 5.
Supermarket Barcode
Scanners Fig. 2. First true barcode using a circular bullseye
symbol.
In 1971, RCA began the rst system test of a
bullseye scanner at a Kroger supermarket in Cin-
cinnati, Ohio. This test and others continued
through early 1974. The rst full-scale implemen-
tation of supermarket checkout scanning began at
Marsh Supermarkets in Troy, Ohio, when a pack
of Wrigleys chewing gum was scanned by a laser
checkout scanner on 26 June 1974. The scanner,
jointly developed by NCR and Spectra Physics,
Inc., is described in U.S. patent 4,064,390 (the
390 patent) issued on 20 December 1977 and
assigned to Spectra Physics. One of the original
scanners, Spectra Physics serial number 006, from 0 12345 67999 5
the rst Marsh Supermarket installation is now on Cnty. Check
Manufacturer # Product #
Code Digit
display at the Smithsonian Institute in Washing-
ton, D.C. Fig. 3. UPC symbol.
These initial supermarket scanners were enor-
mous in comparison to the laser scanners common in todays checkout counters. The scanner was very
large and sat directly on the oor. Its scanning window was at the end of a grocery conveyor that sat on
top of the checkout counter. The scanners dimensions were 30 inches high 12 inches wide 18 inches
deep. The scanner is aptly described as being about equally comprised of optics, mechanics, and
electronics.
Before beginning a discussion of the optical path through this scanner, is it useful to consider
factors involved in scanning a UPC barcode symbol. The UPC symbol was designed so that it could be
scanned by a simple X conguration scanning pattern. As a result, the UPC symbol is split into two
halves that can be scanned in two separate scanning passes. In order to ensure that the two halves are
assembled in the correct order, a check digit and design features such as differing start and stop bar
patterns for the left- and right-hand halves of the symbol are included in the UPC symbology
specication. Figure 6 illustrates that the beam labeled A scans through the entire left half of the
label, while the beam scanning down and to the right (B) scans through the complete right half of the
label. In principle, these two scans produce a scanning signal which allows the entire label to be decoded
by the scanning system.
Figure 7 from the 390 patent illustrates a portion of the optical path in the Spectra-Physics
scanner. A 24-facet optical polygon, denoted by R, provides a mechanism that produces orthogonal
horizontal and vertical scan lines on a product (the cube at the top of the illustration). A laser beam
entering at the bottom right of the gure is directed by mirror 60 through a slot in the polygon mirror
assembly to mirror 82. This mirror subsequently sends the beam to mirror 84, through beamsplitter 86
and lens 88 to mirror 42 and on to lens 90. Lenses 88 and 90 form a relay telescope used in generating
vertical scan lines. After lens 90, the beam is deected by the polygon mirror and reected by fold mirror
94 through the scanner window 34 to impinge on the product. Light scattered from the barcode label on
the product follows a retro-directive path back through the optical system and ultimately impinges on a
photodetector (not shown).
Vertical scan lines are generated in a similar manner and follow a similar beam path as the
horizontal scan lines, however, each beam from beamsplitter assembly 54, 56, 58 makes two reections
from two separate polygon mirrors. An ingenious arrangement of facet tilt angles of sequential polygon
mirrors results in three vertical scan lines for each horizontal scan line. The slots in the face of the
polygon assembly are designed so that only one horizontal or vertical scan line passes through the
scanning window at any given time.
A large fractional horsepower AC motor rotated the 390 scanner polygon at 3400 RPM
producing scanning speeds of 8000 in./min. The retro-directive light collection path utilized aspheric
collection optics to minimize spherical aberration and coma. Narrow-band optical lters rejected
ambient light. These design features resulted in breathtaking, state-of-the-art, scanning performance. It
was possible to literally throw a ve-stick pack of chewing gum spinning across the scanning window
and have its barcode label decode on the rst pass! Now, nearly 40 years later, present day supermarket
A B
Handheld Barcode
Scanners
Scanners used in supermarket applications quickly
moved to laser scanning due to the high scanning
0 12345 67999 5
speed and large depth of focus available from such Fig. 6. Simple X conguration scanning pattern.
devices. Initial industrial applications of barcodes,
such as inventory control and tracking work in
process, had signicantly lower performance
requirements and required lower price points. Ini-
tially simple barcode wands were used for these
purposes. An early barcode wand is described by
Turner and Elia in U.S. patent 3,916,184 assigned
to Welch Allyn, Inc. (the 184 wand). The 184
wand utilized an incandescent bulb or LED and a
ber optic bundle to illuminate the barcode symbol
through an opening in the case. A simple two-lens
system and photocell or photodiode produced an
electrical signal representative of the barcode sym-
bol as the wand was manually scanned across the
label. Apertures in the two-lens system controlled
the depth of eld and eld of view (i.e., resolution of
the barcode label) of the wand.
Since wands were in contact with the label
during scanning, the label became degraded when
scanned multiple times. Another common problem
with wands was that paper lint would accumu-
late in the entrance opening and degrade scanning
performance. To improve on early wands, Bayley of
Hewlett Packard suggested the use of a sapphire ball
lens in the opening of the wand in U.S. patent
4,855,582. Hewlett Packards commercial product Fig. 7. A portion of the optical path in the Spectra-
based on this patent had a compact hermetic elec- Physics scanner.
tronic package that housed the illumination LED
and a photosensor. The highly integrated design was cost effective and very rugged, an important
requirement for any handheld device in an industrial or warehouse environment.
The contact nature of barcode wands was a disadvantage in many industrial environments since
the label was often read several times during a manufacturing or inventory process, or in package
tracking. These applications drove the development of non-contact handheld scanners. An early
example is described in U.S. patent 4,560,862, rst disclosed to the Patent Ofce in 1983. The concept
of this patent is illustrated Fig. 8. A rotating polygon with concave mirrors scans an image of an
incandescent source across a barcode symbol. The illuminated scanning plane is then imaged back
along the optical path to a beamsplitter which directs the returning light through a relay lens, aperture
stop and eld stop to a photodetector. The curved mirrors on the polygon have various radii, thus
producing multiple temporally multiplexed focal planes on the photodetector due to rotation of the
polygon. The commercial device utilized eight spherical mirrors on the polygon and was housed in a
gun shaped housing for convenient handling, and used a trigger for selection of a barcode label to be
read.
Eastman and Boles disclosed the rst laser diode based xed-beam handheld laser scanner to the
patent ofce in 1983, resulting in issuance of U.S. patent 4,603,262 in July 1986. The xed-beam
scanner, similar in size to a childs squirt gun and the rst to use surface mount electronics to reduce size
and weight, was scanned by the users wrist motion. The laser diode operated at 780 nm, so its light was
not readily visible to a user. Consequently a visible marker beam propagated coaxially with the laser
beam to enable the user to point the scanner at a barcode label. The scanner had no moving parts other
than its trigger button, so it was very rugged and capable of operating after a drop from a second-story
window onto a concrete sidewalk with no ill effects.
Both of the above devices were quickly eclipsed by HeNe-based moving beam handheld laser
scanners. U.S. patent 4,409,470 by Shepard, Barkan, and Swartz disclosed a narrow-bodied laser bar
code scanner that became successful in the early to mid-1980s as Symbol Technologies LS-7000. The
advent of low-cost visible laser diodes quickly led to the availability of rugged handheld laser scanners
in the late 1980s and early 1990s, as described in U.S. patents 4,760,248; 4,820,911; and 5,200, 579. In
order to avoid the strong patent position of Symbol Technologies in handheld laser barcode scanners,
Rockstein, Knowles, and their colleagues invented a triggerless handheld barcode scanner as
described in U.S. patent 5,260,553. This device automatically began scanning when a barcode symbol
was in close proximity. Several examples of visible laser diode barcode scanners are shown in Fig. 9, in
approximate chronological order from left to right.
I
nventors usually realize that any good idea owes some debt to earlier technological develop-
ments. The laser printer is no exception. In 1938, Chester Carlson, a struggling patent attorney,
needed a way to copy patents other than by hand. That led him to develop a technology now
known as xerography from which the company Xerox was born. The word xerography comes
from the Greek words xeros and graphein which mean respectively dry and writing. The
laser printer, as we now know it, depends on this wonderful imaging capability.
Xerox introduced the rst real copier in 1959 and called it the 914, with the number
standing for the largest paper the machine could copy. Despite warnings by market experts to
the contrary, the 914 became one of the most protable products ever produced in the Western
world. Xerox started developing many different kinds of imaging machines. One of the most
interesting and advanced for its time was a limited-volume product called LDX for Long Distance
Xerography.
As a young engineer coming to Xerox in 1964, one of the challenges the author was given
was to see if the LDX system could be made faster. The LDX system as built in the middle 1960s
was a design with limited extensibility. A line scan cathode ray tube (CRT) was used with an
imaging lens to scan an original document. The light was picked up by a light sensor and sent
over a 56-kilobaud (kBd) line to a receiver at a location perhaps hundreds of miles away. This
sort of bandwidth was not readily available but could be purchased if needed. The receiving
station also had a line scan CRT whose beam was modulated to generate a variable-intensity
light signal that a lens imaged to expose a xerographic drum similar to that used in a copier. The
problem was that the CRT used for exposure was pushed hard to get enough light output. It took
many seconds to print a document, and there was a real desire to go much faster. The immediate
challenge was to nd a better way.
Being a graduate student at the University of Rochester Institute of Optics, the author was
using a new light source: the heliumneon (HeNe) laser, invented in 1961. Its main advantage
was its brightness or radiance. Because the laser beam was highly conned rather than a
Lambertian radiator, its radiance was thousands of times higher than the CRT. The red beam
was a concern for current photoreceptors in the copiers, but as a bright, deectable light source, it
had no peer. The author set about to see what might be done with the laser as an illuminator for
the print and perhaps even the scan station.
A key advantage of the CRT was the fact that magnetic or electrostatic elds could deect
the electron beam on the screen. Laser beams, as someone has described them, are stiff and so
they need something to deect them. The only practical solution was putting several mirror facets
on a rotating disk. Using 10 to 20 or more facets greatly reduced the required rotational speed.
However, the mirror facets and rotational axis had to be kept within a very few arc seconds of
each other while rotating at several thousand revolutions per minute. This is an exceedingly
difcult requirement for a cost-effective commercial product. The author built a laser facsimile
prototype with a modied 914 (720 series) copier to scan an original and print the results. His
skilled colleague Robert Kowalski built electronics generating about 1000 V to drive a special
Pockels-cell beam modulator. Switching 1000 V in a small fraction of a microsecond even with a
small capacitance was not trivial.
The two researchers clamped, taped, and otherwise assembled a scan and print breadboard
to the 720 copier with a special red-sensitive drum and made some laser fax copies in 19681969.
134
The lack of precision in the scanning mirror left bands in the images, but the demonstration showed
what a laser system could do. However, a way had to be found to make a precise scanner without
spending $20,000 each.
After thinking about the precision requirements for several days, the author came upon an idea
while sketching the problem on a piece of paper. It looked as though a cylinder lens would solve the
problem. If it would, it was puzzling why no one else had discovered it. A 12-in. (30.48-cm)-long
cylinder lens was ordered, which arrived the next day by air. What was the result? Eureka! It solved
the scanner problem. A scanner with perhaps 1 or 2 arc min of error could perform a task that
would have required 1 arc sec precision. The scanner was now going to be very inexpensive.
Today, such a simple six-sided polygon and motor system for a personal laser printer costs less than
$5$10.
About this time, the author began to wonder about an idea after talking with a couple of other
people. Why not forget the input scanner and use a computer to generate the signal patterns for a print
station only?
Up to this time, every part the author and Robert Kowalski had used was already part of their
laboratory equipment since no spending on this effort was permitted. Furthermore, about this time a
more serious, non-technical issue arose.
The authors immediate manager got wind of his idea and stated in no uncertain terms that this was
a bad idea and that he wanted all work on it stopped. This was the beginning of a real challenge: To
continue the project or let it go? The author decided to continue working on it less obviously. The
situation was heading to a real confrontation when, one day in early 1970, the author read in the
company newsletter about a new research center being started in Palo Alto, California. He called one
person he knew in the starting group to ask how to tell them about the project and described what was
being worked on. They decided to y the author out to California to make a case for the new printer
technology.
The trip was a rousing success. A group also becoming part of the new Palo Alto Research Center
(PARC) was working on a personal computer that bit-mapped text and graphics onto a display much
like todays Macs and PCs. They needed a way to render their pixel-oriented screen image to paper. The
new laser printer was a natural t to their needs. They were willing to take the author into their
organization, but there was one problem: management in Rochester would have to approve a
transfer. The author promised to nd a way to get this done.
Upon the authors return to Rochester, his manager refused to permit the transfer to PARC.
Technically, this was a violation of company policy. After some stressful discussions the author took the
issue to a more senior level. Eventually, after some tense but productive discussions, George White, an
energetic and future-oriented Xerox vice president, approved the transfer to PARC, and the author
moved his young family to California in early January of 1971. Thus began work in earnest on the laser
printer.
Spearheaded by the visionary genius of Jack Goldman, PARC was a great place to build this
machine as well as being a font of other great technologies. The invaluable Bob Kowalski from the
Webster, New York, Xerox facilities was hired. John Urbach, now deceased, provided a lot of
encouragement as well as nancial support. He reported to one of the best managers and mentors
anyone could have, Bill Gunning, who helped the author set realistic and important goals for the rst
printer and provided very wise counsel.
The group decided to build a prototype that would print at one page per second and at a spatial
density of 500 laser points per inch in both the fast and slow scan directions. A solution to the poor red
sensitivity of standard Xerox photoreceptors emerged from a major optical system design error in the
Xerox 7000 duplicator that did not show up until early production. The only practical way to remedy
this optical system problem was to replace the usual bluegreen-sensitive photoreceptor with one more
sensitive in the red part of the spectrum on the drum of the 7000. This error was a truly fortuitous event
allowing the laser printer work to proceed. It is unlikely that the printer would have had the necessary
backing if it alone had required a special photoreceptor.
The Xerox 7000 with the red-sensitive drum was going to be used to print one page per second
using a HeNe laser. This meant generating at least 20 million points per second from the scanner. The
A
merican inventors including David Paul Gregg and James Russell originated some key
optical storage concepts in the late 1950s and early 1960s, but initially envisioned
writing with electron beams and reading by directing laser beams through the material
to detectors on the other side. The concepts of a rotating disc and reective media made optical
storage a real possibility [1]. Rotating the disc and moving the optical pick-up (OPU) radially
gave the required two-dimensional access to the data surface. Reective media meant the emitters
and detectors could be on the same side of the disc, greatly easing optical alignment. Burying the
data surface in a transparent disc made the media robust in the hands of the consumer.
By the early 1970s, growing interest in read-only optical discs for Hollywood movie
distribution led to product development. A partnership between MCA and Philips, MCA
DiscoVision, introduced the rst consumer laser video disc (later called Laservision) in the
United States at the end of 1978. It used HeNe gas lasers to read molded or embossed pits on a
30-cm disc, the size of a vinyl LP record. Video information was encoded as a variable distance
between the edges of pits in a spiral track, yielding a frequency-modulated analog signal as the
disc rotated past the laser spot.
The details of the tracking process were quite complex, and it took longer than expected to
develop a reliable and low-cost process to mass produce the discs. Philips made the rst fully
playable disk in 1976, but it took an intense engineering effort to launch the rst qualied mass
production started at a factory in Blackburn, England, in 1981. The discs showed less wear than
VHS tape, and image quality was better, but those advantages were not enough for Laservision to
outcompete tape, which was less expensive and recordable (although most customers did not use
that aspect). In the end, VHS tape thoroughly dominated consumer video distribution until the
arrival of DVD in the mid-1990s.
In 1974 Philips Research Laboratories and the Philips Audio Division began developing an
optical audio-disc system. Their design thinking, further detailed below, is an excellent
example of system integration using the best of current technologies and additionally antici-
pating probable future developments in component technology, specically digital processing
power of consumer integrated circuits and wavelength reduction of solid state lasers. The
project grew internally in Philips, and it was decided that analog signal recording would not
work well enough and that a fully digital technique was a better approach. The magnitude of
the development effort made it attractive to have partners, and after some negotiation, an
agreement was reached with Sony in 1979. In-depth technical discussions were started,
focusing primarily on the error-correction signal processing. The contributions from both
companies resulted in a system standard which forms the physical basis of the compact disc
(CD) as we know it today.
Early in the project the disc size was chosen as 120 mm and called compact disk because it
was smaller than the 300-mm Laservision disc. They knew that the available and affordable solid
state lasers for the playback devices would give them about 1 mW at approximately 800 nm, and
designed the optical system accordingly (see Fig. 1) [2]. The laser beam passed through a 1.2-mm
transparent substrate to read data marks embossed onto the aluminized disk surface. The
embossing makes the data marks reective phase objects.
After dening the CD-A disc standard, Philips and Sony set up a licensing organization
which Philips still administers. Licensees receive a copy of the Red Book which details the
138
standard and optical performance metrics.
The physical standard focuses entirely on
the removable optical disc. The only con-
straint on the disc player is that it must be
able to read and play back standard-
format discs. A great advantage of this
sort of standard is that it allowed open-
ended growth in the capabilities of disc
players. For example, todays inexpensive
players transfer data at 16 times the
1.41 Mb per second of initial players. The
optics, servos, and electronics could han-
dle twice that rate, but that would require
spinning the polycarbonate disk at 6400 to
16,000 revolutions per minute, reaching
speeds where the centrifugal force could
shatter the plastic disc, a very disconcert-
ing experience for the user.
Several aspects of optical disc system
design are brilliant. One example is writing
data tracks as a very long spiral rather
than concentric circles, allowing mass-
produced players to read data by following
the track rather than creating it. Injection
molding can replicate discs accurately and
inexpensively, so this shifts the costs of
achieving the required precision to the
mastering machine, which is amortized
over millions of replicated discs. That also
allowed most players to play discs with
track pitch reduced to squeeze up to 99
minutes of music onto a disc originally
designed for 74 minutes. Inspired choices
of eight-to-fourteen modulation coding
and cross-interleaved ReedSolomon
error-correction code made the system re-
silient to random bit errors that if uncor- Fig. 1. Artists rendering of playback optics in rst Philips CD
rected could blow out speakersvital product, CD100. Size was 12 mm x 45 mm. Philips Technical
because replicated disks had raw byte error Review 40(6), 150 (1982).
4 5
rates of 10 to 10 . Establishing 2352-
byte blocks for CD-audio discs left room
for the error-correction codes needed to meet computer requirements of bit-error rates less than 1012,
allowing development of CD-ROM for computer storage.
Industry Anecdotes
The author was deeply involved in developing those systems, so he saw the dynamics that shaped their
history. As a Senior Researcher in the R&D Division of Ampex Corporation in the 1970s, he was
offered the opportunity to lead technical development of a either re-writable magneto-optic media or
write-once media. He chose the write-once group because it seemed that write-once media were
certainly as useful as ink and paper and that the dye polymer media and drives could be produced at
much lower cost than the M-O media and drives. These guesses turned out to be correct in the long run.
What was not realized at the time was that the changes in computer operating systems required to
manage read-only and write-once media would be very slow in coming. Those le-system enhance-
ments were not standardized and implemented until the late 1980 and 1990s, when software developers
nally understood that the utility and low cost of CD-ROM, and later CD-R, made them necessary
system components.
By the mid-1980s the author was on the other side of the fence, as Manager of Optical Storage
at Apple Computer. His initial goal in joining Apple had been to develop CD-A and CD-ROM for use
with Apples computers. Steve Jobs really liked optical storage and therefore provided good support
to the CD effort. At the time, M-O developers believed the unlimited re-writeability and removability
of M-O media made it more attractive than conventional magnetic hard disk drives for computer use.
After Steve left Apple, rumors spread that his new company called NeXT was going to used M-O
drives instead of magnetic discs in its new computer. That worried Apple management, which had
great respect for Steves product judgment, so the authors group began working with a major
Japanese electronics company on M-O drives for Apple computers. As the possible performance and
costs were learned, analysis showed that computer performance would not be adequate with only a
M-O drive. The slower access time and transfer rates of M-O drives would make the computers too
sluggish for the market. Subsequent developments indicate that dropping the M-O disc was the
correct choice.
cases the lens is actually a dual optic with a high-numerical-aperture (NA) annular zone giving a
small 650-nm spot and a smaller-NA region focusing 780 nm light to a larger spot.
Decades of research and development have dramatically reduced size and increased capabilities.
Figure 2 and Tables 1 and 2 compare size and specications of optical drives from 1988 to 2010. For
demonstration, the top lid of the 2008 drive has been removed, and an unnished 120 mm disc
(metallization layer not yet applied) has been placed on the spindle. The optical pickup is visible
through the still transparent disc.
The bar was raised even further in 2006 with introduction of the Blu-Ray drive product, based on
the development with 405-nm lasers. Evolution in every aspect of the technology, as shown in Table 2,
created a dual-layer 120-mm disc with 50-gigabyte capacitywhich would have been unthinkable four
decades earlier. Blu-Ray can support high denition video with four times as many pixels as NTSC/PAL
video.
The future of optical disc use and development will be strongly affected by other technologies. Will
consumers accept the lower-quality video distributed over the Internet or insist on the quality delivered
by a 120-mm HD Blu-Ray disc? Optical discs with properly made media are as archival as silver halide,
so what role will they play in archiving the data our society continues to generate at an accelerating
rate?
References
1. A. Kees, I. Schouhamer, and I. Immink, The CD story, J. Audio Eng. Soc. 46, 458465 (1998).
2. M. G. Carasso, J. B. H. Peek, and J. P. Sinjou, The Compact Disc Digital Audio system, Philips Tech.
Rev. 40(6), 150156 (1982).
L
asers have made truly revolutionary changes in optical metrology. The lasers small
source size and narrow linewidth made it so much easier to obtain good contrast
interference fringes that applications of interferometric optical metrology have increased
immensely during the 50 years since the laser was rst developed. Single-mode frequency-
stabilized lasers provided a standard for dimensional metrology, while ultra-short pulsed lasers
have enabled high-resolution range nding.
The laser has greatly enhanced the testing of optical components and systems. Before the
laser, the use of interferometry in optical testing was limited because either the interferometer
paths had to be matched or the source size had to be very small to have good spatial coherence,
and the lters needed to reduce spectral width left very little light for measurements. Once the
laser was introduced, Bob Hopkins from the Institute of Optics at the University of Rochester
was quick to realize how much laser light could improve the testing of optical components [1],
and he encouraged other researchers to design laser source optical interferometers [26]. By 1967
lasers had become common in optical testing [7,8]. Figure 1 shows a laser unequal path
interferometer (LUPI) designed by John Buccini and manufactured by Itek in the late 1960s
and early 1970s.
Abe Offner from Perkin Elmer was quick to realize that adding null correctors to laser
interferometers would allow measurements of optical components with aspheric surfaces [9]. Null
correctors are a combination of lenses and mirrors having spherical surfaces, but when used in the
proper way they produce an aspheric wavefront that matches the surface of an aspheric optic,
producing interferograms with straight equally spaced fringes when the tested aspheric surface is
perfect. Unfortunately, that use of null correctors received horrible publicity after initial orbital
tests of the Hubble Space Telescope showed its optics could not be brought to the expected sharp
focus. Analysis of the awed images showed that the primary mirror had an incorrect shape.
A commission headed by Lew Allen, director of the Jet Propulsion Laboratory, determined that the
null corrector used to test the primary mirror had been assembled incorrectlyone lens was
1.3 mm from its proper position [10]. That caused the null corrector to produce an incorrect
aspheric wavefront, so using it to test the primary mirror led to fabricating the mirror with the
wrong shape. In correcting the error, the cost was more than a billion dollars to design and
fabricate additional optics and install them on the Hubble telescope from the space shuttle.
Fig. 4. Three-dimension contour maps showing shape of vibrating surface as a function of time.
Frequency Combs
An important recent development is the use of frequency comb lasers for determining the absolute
distance to an object. In 2005 John Hall and Theodor Hnsch shared half the Nobel Prize in physics
for development of laser-based precision spectroscopy, including the use of frequency comb lasers.
Frequency comb lasers [36] have the potential to revolutionize long-distance absolute measure-
ments by allowing better than sub-micrometer accuracy of distances up to, and possibly beyond,
10,000 km. Comb lasers are pulsed (ultrafast) mode-locked lasers with a precisely controlled repetition
rate and pulse phase. Stabilizing the output of a femtosecond laser provides a spectrum of well-dened
frequencies. The periodic pulse train of a femtosecond laser generates a comb of equally spaced
frequencies for multi-wavelength interferometry. It is possible to link the time-of-ight domain of long-
distance measurement with an interferometric measurement to obtain nanometer accuracy. The basic
concept is to use this incredibly regular pulse structure to measure a distance in units of the pulse
separation length. For accuracies down to the 10-m level, it is sufcient to use Time of Flight
measurement [37,38]. Sub-wavelength accuracy in the nanometer range can be obtained using spectral
interferometry where the distance is obtained by determining the slope of the phase as a function of the
optical frequency [39,40]. It is believed that distances of 500 km can be measured to accuracies better
than 50 nm.
It continues to be a very exciting time for the use of lasers in optical metrology. With the
combination of new lasers, modern detectors, computers, and software, the capabilities and applica-
tions of metrology are astonishing.
References
1. R. E. Hopkins, Re-evaluation of the problem of optical design, J. Opt. Soc. Am. 52, 12181222
(1962).
2. T. Morokuma, K. F. Neen, T. R. Lawrence, and T. M. Ktlicher, Interference fringes with long path
difference using HeNe laser, J. Opt. Soc. Am. 53, 394395 (1963).
3. R. M. Zoot, Laser interferometry of small windows, Appl. Opt. 3, 985986 (1964).
4. R. M. Zoot, Laser interferometry of pentaprisms, Appl. Opt. 3, 11871188 (1964).
5. K. M. Baird, D. S. Smith, G. R. Hanes, and S. Tsunekane, Characteristics of a simple single-mode
He-Ne laser, Appl. Opt. 4, 569571 (1965).
6. D. R. Herriott, Long-path multiple-wavelength multiple-beam interference fringes, J. Opt. Soc. Am.
56, 719721 (1966).
7. U. Grigull and H. Rottenkolber, Two-beam interferometer using a laser, J. Opt. Soc. Am. 57,
149155 (1967).
8. J. B. Houston, Jr., C. J. Buccini, and P. K. ONeill, A laser unequal path interferometer for the optical
shop, Appl. Opt. 6, 12371242, (1967).
9. A. Offner, A null corrector for paraboloidal mirrors, Appl. Opt. 2, 153155 (1963).
10. L. Allen, The Hubble Space Telescope optical systems failure report, NASA Technical Report, NASA-
TM-103443 (1990).
11. J. N. Dukes and G. B. Gordon, A two-hundred-foot yardstick with graduations every microinch,
Hewlett-Packard Journal 21, 28 (August 1970).
12. R. Crane, Interference phase measurement, Appl. Opt. 8, 538542 (1969).
13. J. F. Ebersole and J. C. Wyant, Collimated light acoustooptic lateral shear interferometer, Appl. Opt.
13, 10041005 (1974).
T
he laser concept emerged at an ideal time to stimulate the emission of military research
contracts. In early 1958, President Dwight Eisenhower established the Advanced
Research Projects Agency (ARPA) to handle the high-risk, high-payoff projects that
cautious military bureaucrats had been avoiding. That May, ARPA director Roy Johnson told
Congress that his agencys work might lead to a death ray. That would be the weapon of
tomorrow, a step beyond the hydrogen bomb, able to destroy nuclear-armed ballistic missiles
before they reached their targets.
Thus it was no wonder that ARPA welcomed Gordon Gould and Lawrence Goldmuntz
with open arms when they came bearing a proposal to build a laser in early 1959. As Gould
told the author many years later, Ray guns and so on were part of science ction, but
somebody actually proposing to build this thing? And he has theoretical grounds for believing
its going to work? Wow! That set them off, and, those colonels, they were just too eager to
believe. (See Fig. 1.)
Charles Townes and Arthur Schawlow were the rst to propose the laser publicly, but their
vision was a modest-power oscillator. Gould had realized that the amplication of stimulated
emission in an oscillator might allow a laser to generate high power and concentrate light to a
high intensity. His pitch to ARPA was laden with bold ideas. He said a laser pulse could mark
military targets and measure their ranges for other weapons. He predicted that laser beams could
be focused to be 10,000 times brighter than the Sun, enough to trigger chemical reactions.
Ultimately, he suggested, lasers might be powerful enough to destroy targets or ignite nuclear
fusion.
Paul Adams, who handled ARPAs optics projects, loved the plan, and a review panel
thought prospects for laser communications, target designation, and range nding were good
enough to justify the $300,000 grant requested. Adams was so enthusiastic that he pushed
through a $999,000 contract for a bigger program at TRG Inc., the company Goldmuntz
headed. Then the Pentagon tossed a monkey wrench into the works by classifying the laser
project and denying Gould a security clearance because of his youthful dalliance with commu-
nism. He could not work on the project he had created.
The press also focused on the idea of laser weapons. When Ted Maiman announced he had
made the rst laser, reporters asked if the laser was a death ray. After trying to duck the
question, he nally admitted he could not rule out the possibility. When he returned to
California, he found the Los Angeles Herald carrying a headline in two-inch red type: L.A.
Man Discovers Science Fiction Death Ray.
After Maimans success, ARPA expanded its program to study laser mechanisms, materials,
and beam interactions with targets. The Air Force gave Maiman a contract to develop ruby
lasers, and other military labs started their own laser projects. The armed services focused on
near-term applications in missile guidance and communications; ARPA focused on high-energy
laser weapons.
Although many physicists were skeptical, they also hesitated to oppose Pentagon plans.
After weapon scientists said nuclear re-entry vehicles were so sensitive to thermal shock that laser
heating might shatter them, ARPAs laser-weapon budget was boosted to $5 million. Air Force
Chief of Staff General Curtis LeMay jumped on the laser bandwagon, saying on 28 March 1962
that beam directed energy weapons would be able to transmit energy across space with the
149
speed of light and bring about the technological
disarmament of nuclear weapons. The Air Force
Systems Command budgeted $27 million for a ve-
year Project Blackeye to develop ground-based
anti-satellite lasers and perhaps a space-based laser
weapon.
But early laser technology was not up to the
task. American Optical pushed neodymium-glass
lasers to generate 35-J pulses, but thermal effects
shattered the rods. The same happened to ruby
rods when Westinghouse pushed Q-switched pulse
energy to 60 to 80 J. Discouraged, ARPA scaled
down its solid-state laser weapon program around
1965.
By that time, the carbon-dioxide laser was
showing hints that gas lasers could reach high
Fig. 1. Gordon Gould. Courtesy of Geoffrey Gould, powersand could conduct away troublesome
1940.
heat. C. Kumar N. Patel generated 200 watts
continuous wave from CO2 at 10 m in mid-1965. That was enough to satisfy his research needs,
but it only whet the appetites of military labs, which began scaling CO2 lasers to impractical sizes.
Hughes reached 1.5 kW using a 10-m oscillator followed by a 54-m amplier.
The real breakthrough to high-energy lasers was the gasdynamic laser, developed by Arthur
Kantrowitz and Ed Gerry at the Avco Everett Research Laboratory near Boston. They knew that
sustained laser power would have to reach a megawatt to damage a military targetand gured they
might reach that level by drawing 0.1% of the energy from a rocket engine, which could generate a
gigawatt by burning chemical fuel to generate hot CO2. Expanding the gas through special nozzles at
supersonic speed produced a population inversion. It was a very simple thing, but not a very efcient
laser, recalled Gerry. First demonstrated in 1966, the gasdynamic laser was kept classied until
1970. By then Avco had exceeded 100 kW, although Gerry was only allowed to report 50 kW at
the time.
That power level attracted interest from the armed forces, and Avco built three 150-kW
gasdynamic lasers, one for each of them. Moving targets proved a challenge. When the Air Force
tried to hit a drone ying gure-eight patterns, the beam locked onto a weather tower and melted it. In
1973, the laser nally shot down a weakened drone. The next step was squeezing a 400-kW gasdynamic
laser into a military version of a Boeing 707 to make the Airborne Laser Laboratory. Two years after an
embarrassingly public failure in 1981, it nally shot down an air-to-air missile over the Naval Weapons
Center in China Lake, California. That was the end of the line for the gasdynamic laser, a monster
of such size and complexity that critics called it a
ten-ton watch.
After the Big Demonstration Laser built by
TRW exceeded 100 kW, the Navy focused its
attention on chemical lasers because moist air
transmits better at the 3.6- to 4.0-m band of
deuterium uoride. In 1978, the 400-kW Navy
ARPA Chemical Laser (NACL) became the rst
chemical laser to shoot down a missile in ight.
TRW then built the rst megawatt-class laser, the
Mid-Infrared Advanced Chemical Laser (MIR-
ACL) (Fig. 2). The giant laser, nished in 1980,
could emit 2 MW, but only for seconds at a time.
Focusing that tremendous power through the air to
Fig. 2. MIRACL. Courtesy of U.S. Army Space and a moving target proved an overwhelming chal-
Missile Defense Command. lenge, and by the early 1980s the armed services
Reference
1. J. Hecht, A half century of laser weapons, Opt. Photon. News 20(2), 1421 (2009).
I
n 1965, Central Intelligence Agency Director John McCone laid down a challenge to a
selected few companies with experience in designing cameras for the intelligence community.
He wanted a new generation of surveillance satellites that combined the broad area coverage
of CORONA with the high resolution of the KH-7 GAMBIT.
Thus was born what would eventually become the KH-9 Hexagon spy satellite. It was the
last lm-based orbiting reconnaissance camera for the United States government. It was a
marvel of engineering achievements that resulted in a ne optical instrument that was capable
of taking stereo photographs of the entire earth as well as concentrating on small areas of
interest and able to distinguish objects two to three feet in size from an altitude of 90 miles
above the earth. The system would become an invaluable asset and provided intelligence
information credited with persuading President Nixon to sign the SALT-1 treaty in 1972. It
was also acknowledged at the time to have been the most complicated system ever put into
orbit. The rst launch was on 15 June 1971 and the last of 19 successful missions sadly
exploded 800 feet above the pad on 18 April 1986 just a few months after the tragic Challenger
explosion.
The vehicle weighed 30,000 pounds, was 60 feet long and 10 feet in diameter, and each of
the two cameras carried 30 miles of lm. The lm traveled at speeds up to 204 in./s at the focal
plane and was perfectly synchronized to the optical image captured by a constantly rotating
scanning camera. The exposed lm was periodically returned to Earth in four re-entry vehicles
caught by an Air Force C-130 over the Pacic. A photograph of the entire vehicle and a schematic
diagram of the vehicle are shown in Figs. 1 and 2, respectively.
The story started out as the author was working for the Perkin-Elmer Corporation, and with
a small group who studied the concept for over a year. The results were presented to the CIA at
night in an innocuous-looking safe house in Washington, DC. Albert Bud Wheelon, the rst
CIA Deputy Director of Science and Technology (from 1963 to 1966) said that the agency
thought highly of the groups concept.
The group then spent an extremely intense six weeks writing a proposal. It culminated in
May 1966 when Perkin-Elmer CEO Chester Nimitz, Jr., the son of the famous World War II
admiral, stood up at the end of the nal proposal presentation to the CIA, put his foot up on a
table and said, We want this fg job and were gonna get every fg agency and every
fg engineer from here to Florida. We recognize the importance to national security and
were capable of doing the job. It was a memorable event.
A second memorable event came ve months later on 10 October 1966, when the group was
told to gather at 10 a.m. in the large engineering room, in an isolated and secure area across the
street from one of Perkin-Elmers two main plants. Group vice president Dick Werner, the
groups program manager Mike Maguire, and contract specialist Charley Hall walked in shortly
after 10. They were all dressed in stylish suits. In those days everyone wore ties and jackets,
although the latter were soon discarded as each day progressed. As they reached the front of the
room Dick reached into his right inside jacket pocket and took out one of the longest cigars
imaginable. The rst words out of his mouth were We won. A great cheer went up from the
153
group. Dick and Mike then each spoke a
few words of praise for the great team
effort along with wishes for success in this
new adventure.
As soon as the meeting broke up,
group members immediately made phone
calls. Many called their wives to say that
Fig. 1. Photo of the Hexagon vehicle (minus 2 re-entry lm the group had won a big program that
capsules).
would keep them employed for a long
time.
Some employees called their stock-
brokers to buy as much Perkin-Elmer stock
as they could afford. Of course, this was
illegal as it was trading using insider infor-
mation. The next day a secretary went
around asking everyone if they had pur-
chased shares and if so, how many. This
list was eventually given to the Perkin-
Elmer legal department, and all who had
bought stock expected to be reprimanded
Fig. 2. Schematic of the entire Hexagon vehicle. and possibly made to sell the shares or void
the purchases. But nothing further was
heard, and it turned out to be a lucrative investment, especially for those who had the courage to
invest serious funds. The company stock split seven times in the next dozen years.
Hiring a skilled technical staff was difcult because the program was top secret, so potential
candidates could not be told the nature of the program or the specic tasks to which they would be
assigned. In addition, completing the required background and security checks took from four months
to a year, and permanent employment depended on clearing security screening.
New hires were told not to discuss or even speculate with others what the program was about.
While awaiting their clearances, most of them worked on unclassied projects in a non-secure part of
the building called the tank. It also was called the mushroom patch, because the people working
there were kept in the dark and fed a lot of crap.
Everyone in the tank eagerly awaited their security clearance. Dick Carritol, a systems and
servomechanism engineer, recalls being called to the security ofce. I was given a bunch of documents
to read and sign. I remember being awed by the words I was reading. It seemed like I was being told
more than I needed to know. After 40 years the memory is a little hazy, but I do remember something
like this: : : : a study program leading to the design and development of a photo reconnaissance
satellite, to conduct covert operations for the CIA, under cover as the Discoverer Program. This high
resolution system is to carry out search and surveillance missions over the Sino-Soviet Bloc : : : the
program name is FULCRUM. (It later was changed to Hexagon.)
Carritol continues: The documents droned on about not revealing, acknowledging, or comment-
ing on the existence of the program, the program name, the customers name, or any of the participants
in the program. This ban on discussion included everyone from ones family and friends all the way to
others on the program with the proper security clearance but without an explicit need to know.
When I had nished all the reading and signing, the security ofcer asked if I was surprised. I
didnt have a feeling of surprise. I felt numb. I had just read a lot of words and concepts that I had never
considered before. Covert Operations, Under Cover, Search and Surveillance of the Sino-Soviet Bloc,
and compartmentalized security clearances were all new and quite foreign to me. I had a lot to learn!
No, I didnt feel surprise, I felt like I had just joined the Big Leagues.
The design environment in the late 1960s was very different from that of today. Computers were
large general-purpose mainframes which received input on punched cards and produced output on
magnetic tape or an impact printer. Analysis programs were limited to early versions of NASTRAN (for
mechanical structural analysis) and SINDA (for thermal analysis).
Reference
1. P. Pressel, Spy in the sky: the KH-9 Hexagon, Opt. Photon. News 24(10), 2835 (2013).
T
he CORONA program came at a time when classied optics programs were in their
steepest ascent toward a mission to literally save the world. But very few people realized it
at the time because it was among the most classied of all classied programs. Outside of
a team of fewer than 100 scientists, at one point only six people, including President Eisenhower,
were aware of the work that together with the U2 surveillance plane helped save the world from
nuclear war. Signicantly, a single person was behind the success of both CORONA and the U2
missions: Richard Bissell of the CIA.
Initiated just weeks after the Soviet Sputnik launch, CORONA was at the cutting edge of
technology and a remarkably visionary program. It anticipated that the high-altitude U2 could be
brought down, as it would be in 1960. Its crucial role was to cast the light of knowledge onto the
dangerous shadows of speculation about Soviet capabilities. At one point, advisors told
Eisenhower that the U.S. needed 10,000 nuclear warheads to catch up. The U2 and CORONA
together provided hard evidence that if there was a missile gap, it was the Soviets who were
behind. The rst successful CORONA mission acquired ten times more information than all of
the preceding U2 missions combined. Eisenhowers visionary program was a credit to his
presidency, and kept President Kennedy from overreacting to the Cuban missile crisis in 1962.
The saga of CORONA has been the subject of a number of good books since its
declassication in 2004. A major reference for this article was ITEK and the CIA [1], which
offers a substantial, factual account of the CORONA program. The most readable history of
CORONA, which covers many of the technical and operational issues, is Eye in the Sky: The
Story of the CORONA Spy Satellites edited by Day, Logsdon, and Latell [2], in the Smithsonian
History of Aviation Series. Another important resource for this essay was a plenary talk given at
the 2004 SPIE annual meeting by (the late) Robert S. Hilbert, one of the principal optical
engineers on CORONA for nearly a decade before becoming the leader of Optical Research
Associates. The author worked with him for nearly 20 years.
CORONA, like the U2, proceeded from concept to ight hardware in a matter of months, an
incomprehensible pace today. The multidisciplinary team of engineers and scientists were armed
primarily with slide rules and engineering judgment, and they had only limited computer
simulation capabilities. But they were unencumbered by any signicant management or budget
constraints and were driven by genuine personal urgency to move ahead at a pace that was
perhaps matched only by the earlier U2 program at the Burbank Skunk Works. The engineering
team, fortuitously, had been together for some years. Nearly all had worked at a reconnaissance
research facility at Boston University. The university was in a nancial crisis when Eisenhower
commissioned CORONA and was disbanding the reconnaissance group, which was quickly
bought by the newly formed Itek Corporation, formed with funding from David Rockefeller.
Rockefeller was an outspoken conservative who decided that if he would not implement his
vision of a better world politically, he would create it by backing key technologies that enabled
his goals. He was a visionary who saw that gaining knowledge of the unknown was a key to
ensuring the future. At the time, Eisenhower was crippled by having no information at all about
vast expanses of adversarial countries. This lack of knowledge led to speculation that potential
adversaries had vast arsenals, as well as strong pressure from the military, the press, and the
public to arm the U.S. well beyond its means. Eisenhower made a key decision, that knowledge at
any monetary cost was the best option.
157
Rockefellers role was vital because the president could not directly ensure that Itek had the
nancial resources needed for the program. Because Eisenhowers key military advisors knew nothing
about CORONA, he was continually challenged as being indecisive in ways that were clearly rational in
light of the super-secret project. As one of the six people briefed on the program outside of Itek,
Rockefeller understood this. However, he was the only Rockefeller briefed, and Itek needed so much
nancing that he had to involve his brothers. This led to some suspense in the story of Itek, but in the
end all the Rockefellers investedand reaped the nancial benets by a timely exit from Itek before
Perkin-Elmer won a vital contract for the follow-on Hexagon (Big Bird) program.
Edwin Land, the founder of Polaroid, was a second key technology advisor and an important link
between the optics community and the president. At a time when the Air Force was pushing for a rst-
of-its-kind crash program in electronic imagery from space, it is likely, but unveried, that Land kept
the CORONA mission rmly based in lm (although the lm was to come from Kodak). Although the
program was Eisenhowers highest priority, its classication level made it impossible to get priority
access to new technology, in particular a critical polyester base lm from Kodak. After the project
stalled because it lacked the special lm they needed, Bissell quietly intervened and a large batch
suddenly arrived.
The exposed lm had to be returned to Earth for processing, so it was jettisoned in a capsule that
was supposed to be caught in the air by a C-130 aircraft. To make sure the lm did not fall into the
wrong hands, the capsules had salt plug seals that dissolved in an hour to drop them to the bottom of
the sea. Only the lm returned to earth, so each mission needed a new camera. The logistics of this were
staggering.
The CORONA program became the denition of perseverance, determination, and perhaps
desperation. The crash program went through a long series of failures, often with the rocket simply
blowing up on the launch pad, a problem not related to CORONA. That might be expected at the
beginning of the space age, but for a year it set a grueling pace for the scientists. Bob Hilbert would
typically arrive at the ofce between 10 a.m. and noon for technical meetings and exchanges and then
work through to midnight. At midnight, he would put on his optics engineer hat and work on computer
simulations until 4 a.m. because the computer time was too expensive at other hours. His wife always
had his dinner prepared when he arrived, at 4:15 a.m., seven days a week.
The stakes were raised after the Soviet Union shot down a U-2 over Siberia on 1 May 1960,
stopping ights that had been the best source of surveillance data. On 10 August, the fourteenth
CORONA launch successfully orbited a capsule carrying an American ag, but the recovery aircraft
ew in the wrong direction. Fortunately, a Navy ship was able to retrieve the capsule. The next launch
came on 18 August, carrying a camera that operated successfully and ejected lm that was successfully
recovered.
The composite graphic in Fig. 1 gives a good overview of the CORONA equipment. Instead of
stabilizing the capsule by spinning it in orbit, which would make photography difcult, Itek scientists
stabilized it with small microjets. The camera itself needed to move back and forth in a pendulum-like
motion to image from side to side. These requirements prevented use of the Fairchild camera used for
imaging in the Korean War, so Itek had to design their own based on earlier ideas for a panoramic camera
for imaging large swaths of the ground by sweeping in a cross-track direction as the satellite orbited.
The chosen orbit was a northsouth one synchronous with the sun to provide maximum high-
latitude coverage during daylight. Initial designs used an oscillating lens to focus the image onto a
curved platen carrying the photographic lm. Traditional aerial photography generally used long focal
lengths to produce large-scale images to record sufcient detail with the limited resolution of
photographic lm. However, the size and weight restrictions of early satellite systems limited the
focal length and the amount of lm that could be carried to orbit. CORONA had to achieve very high
resolution in a compact system constrained by lm handling and dynamic limitations.
Robert Hopkins of the Institute of Optics suggested a Petzval-type design to meet the camera
resolution requirements. Itek engineers directed by Walter Levison, Frank Madden, and Dow Smith
generated a novel Petzval design that mounted primary and large-aperture imaging components in a
constantly rotating lens barrel and put the lower-tolerance eld attening components near the focal
surface in a lightweight oscillating arm that dened the image location. These two assemblies operated
synchronously to wipe the image across the photographic lm. The lm was advanced when the lens
was rotating in a non-image collecting part of the cycle and was dynamically located relative to the lens
just at the time of exposure by rollers attached to the oscillating eld attener assembly.
The result was a minimum-weight camera that could t across the width of the spacecraft and
allowed the inclusion of two cameras to provide stereo coverage of the entire imaging swath. The
optical components also needed to exhibit appropriate lateral shifts during the panoramic scan to
provide image motion compensation and reduce along track blur in the recorded image. Additional
optics recorded stellar index images on the lm to aid geo-location of targets. The result was a
remarkable synthesis of optical, mechanical, and electrical systems that were the most complicated, and
eventually reliable, systems of their kind to be incorporated in a spacecraft at the time.
Figure 2 shows a test exposure taken from an aircraft ying over Manhattan, which illustrates the
strong distortion of the wide-panorama photos. One of Bob Hilberts key responsibilities was the
optical design and manufacture of the rectier lens based on a concept credited to Claus
References
1. J. E. Lewis, Spy Capitalism: Itek and the CIA (Yale University Press, 2002).
2. D. A. Day, J. M. Logsdon, and B. Latell, Eye in the Sky: The Story of the Corona Spy Satellites
(Smithsonian Institution Press, 1998).
T
he idea of laser isotope enrichment grew from the lasers ability to concentrate its output
power in a narrow range of wavelengths. Different isotopes of the same element are very
hard or impossible to separate chemically, but the difference in their masses leads to
differences in their spectra, which in principle can be used to selectively excite one isotope and
isolate it by some photo-induced process.
The rst proposal came from the Atomic Energy Commissions (AECs) Mound Laborato-
ries in Miamisburg, Ohio, which in 1961 began a classied investigation of using lasers to enrich
the concentration of ssionable uranium-235. Others independently proposed laser uranium
enrichment. A company called Radioptics proposed it to the AEC in 1963 and later unsuccess-
fully sued the AEC for violating their trade secrets. A French group received a patent in France in
1965, and by the time a U.S. version of the patent issued in 1969 the idea was looking attractive.
The impetus came from the development of the tunable dye laser and the growth of nuclear
power. The U.S. depended on the gaseous diffusion process developed during World War II to
enrich U-235 concentration to the levels needed for atomic bombs. Gaseous diffusion is energy-
intensive, expensive, and raises U-235 concentration only a small amount on each pass. Laser
enrichment offered to reduce cost, improve efciency, and increase recovery of U-235.
At the Avco-Everett Research Laboratory, Richard Levy and G. Sargent Janes developed a
two-step process to enrich U-235. First a dye laser would selectively excite U-235 atoms in
uranium vapor, then an ultraviolet laser would ionize the excited U-235 atoms, so they could be
collected [1]. (Figure 1 shows the process.) Avco lacked money to develop the technology, so they
formed a joint venture with Exxon Nuclear, hoping to build a private uranium enrichment
business.
Avco-Everett founder Arthur Kantrowitz initially worried that laser enrichment might open
the door to nuclear proliferation. At rst glimpse it seems like its a garage operation. A garage
operation for separating uranium isotopes is a frightening thing, he recalled in a 1985 interview.
He imposed special security restrictions but eventually realized this is not an easy way to make a
bomb. It might be an easy way to make 1000 bombs, but it is not a terrorist operation because
of its technical complexity [2].
In 1972 the AEC launched competing laser uranium enrichment projects at its Los Alamos
and Livermore laboratories.
John Emmett, director of Livermores laser program, chose to try selective excitation of
U-235 atoms in uranium vapor with the relatively well-developed tunable dye laser. That
paralleled the Avco approach but was based on earlier work by Ray Kidder of Livermore. They
proposed a two-step process, starting with using visible output of a narrow-band dye laser tuned
to excite U-235, then ionizing the excited uranium atoms. In early 1973 Livermore hired three
developers of the rst continuous-wave dye laser from Eastman Kodak, Ben Snavely, Otis
Peterson, and Sam Tuccio, to start and manage the program. It seemed like an exciting thing to
do at the time, Snavely recalled many years later, an opinion echoed by the other two.
At Los Alamos, Reed Jensen and John Lyman chose to try selective enrichment in UF6, the
compound used in gaseous diffusion, which sublimes at about 55 deg Celsius and is easier to
handle than uranium vapor. They found a large isotope shift in a 16-m absorption band of UF6
and discovered that ultraviolet photons could photodissociatiate excited UF6 molecules, precipi-
tating solid UF5 from the gas phase reaction and releasing free uorine into the gas. Developing
161
the process would require nding a nar-
rowband 16-m laser that could generate
enough power to dissociate 235UF6. Los
Alamos chose C. Paul Robinson to be the
director of the program to solve all those
problems.
At Livermore, Snavely clashed with
Edward Teller and particularly recalled
Tellers disapproval of a metal-vapor pro-
cess that eventually was adopted for the
Atomic-Vapor Laser Isotope Separation
(AVLIS) program (see Fig. 3). When
Snavely told him he expected the process
to succeed by the end of September, Teller
grumbled, You mean by the 31st of Sep-
tember? Snavely ignored him, and Teller
Fig. 1. The Avco-Everett scheme for laser enrichment of pointedly said, You know September has
uranium required the combination of four laser beams to produce only 30 days. Snavely then replied, Yes,
the desired wavelengths to select U-235. (AVCO Research I knew that, but I wasnt sure that every-
Laboratory, courtesy AIP Emilio Segre Visual Archives, Physics body knew it, and Teller threw him out of
Today Collection.) his ofce. Yet Snavely recalled that after he
succeeded, Teller made a point of congrat-
ulating him when they met at a University
of California ceremony.
Livermore was the rst to report ura-
nium enrichment in June 1974 at the In-
ternational Quantum Electronics Confer-
ence in San Francisco. They illuminated a
beam of hot uranium vapor with a dye
laser emitting near 590 nm, selectively
exciting U-235 atoms that then were ion-
ized with ultraviolet light from a mercury
arc lamp [3]. Figure 2 shows the enriched
uranium oxidized to form yellowcake
visible in the bottom of a test tube. That
process would not scale to mass produc-
tion, but Richard W. Solarz and Jeffery A.
Paisner later found a way to coherently
pump the selected isotope all the way from
the ground state to an autoionization state
(Rydberg level), permitting cost-effective
isotope separation.
Meanwhile, Los Alamos developed a
two-step process in which a 16-m source
rst excited vibration of cooled UF6 mole-
cules containing U-235 and then a 308-nm
xenon-chloride laser removed a uorine
atom from the excited UF6. The resulting
Fig. 2. Four milligrams of uranium with its U-235 UF5 precipitated as a solid that could be
concentration enriched to 3% by a dye laser process at Livermore ltered from the gas. Developing the cool-
is visible at the bottom of this test tubethe rst time this much
ing process was a major accomplishment;
uranium was enriched by lasers. 1975 photo from the Lawrence
Livermore National Laboratory. (Lawrence Livermore National it required owing UF6 diluted with a
Laboratory.) noble gas through a supersonic nozzle to
Acknowledgment
Thanks to Otis Peterson for assisting with this essay.
References
1. R. H. Levy and G. S. Janes, Method of and apparatus for the separation of isotopes, U.S. patent
3, 772, 519 (13 November 1973).
2. Arthur Kantrowitz, interview by Robert W. Seidel for the Laser History Project, 25 September 1985.
3. Anonymous, Report from San Francisco, Laser Focus 10(8), 1025 (1974).
4. Anonymous, Molecular process enriched milligrams of uranium two years ago at Los Alamos, Laser
Focus 14(5), 3234 (May 1978).
5. George Palmer and D. I. Bolef, Laser isotope separation: the plutonium connection, Bull. Atom. Sci.
40(3), 2631 (March 1984).
6. A. Heller, Laser technology follows in Lawrences footsteps, Sci. Technol. Rev., 1321 (May 2000),
https://www.llnl.gov/str/Hargrove.html.
L
aser fusion research began [1] at several establishments shortly after the rst laser
operated in 1960. John Nuckolls of the Lawrence Livermore National Laboratoroy and
others around the world quickly recognized that the laser had the potential to concentrate
power to the extreme levels required for small-scale fusion tests. Theoretical analysis showed
[1,2] that achieving fusion and signicant energy yield with the easiest targets to ignite, a mixture
of deuterium and tritium (DT), would require imploding them to extremely high density
perhaps ten thousand times normal liquid densitywith nanosecond-scale pulses in the kilojoule
to megajoule range. Producing the extreme pressure and fuel implosion velocity required to reach
the required density would require irradiance of 1014 W/cm2 with lasers expected to be available
in the near term. The challenge was to achieve signicant energy yield at a size that looked
reasonable for laboratory experiments.
Two basic concepts for laser-driven fusion explosions were quickly developed, as shown in
Fig. 1. The direct-drive implosion uses laser energy that impinges directly on a spherical target
containing DT fuel within an ablator shell that absorbs laser energy and expands, compressing
the remaining ablator and fuel to a small volume in the center of the target and heating it to
initiate DT fusion. The indirect-drive implosion absorbs the laser energy on the inside of a heavy
metal cavity or hohlraum, producing soft x-rays that illuminate the ablator and implode the fuel
capsule as in the direct-drive fusion.
The direct-drive implosion requires extremely uniform irradiance to achieve spherical
symmetry. Indirect-drive fusion eases that requirement by converting the laser light to soft
x-rays that with proper design uniformly irradiate the central capsule. X-ray absorption in the
ablator is also simpler and less subject to nonlinear processes than laser absorption. However,
indirect drive couples only 10%20% of the drive energy to the fuel capsule, so it needs a higher
laser drive energy.
Laser sources for such small targets should store energy from a long pump pulse and deliver
a carefully shaped nanosecond pulse. Development of the Q-switch and the neodymium-glass
laser were important milestones, providing a nanosecond pulse source and an amplier that
could be made in large sizes and had rather low gain so that it did not break into spontaneous
oscillation from stray light before the nanosecond extraction pulse. Those developments
encouraged Ray Kidder of Livermore to estimate that a pulse of at least 100 kJ lasting less
than 10 ns might be able to ignite a small amount of DT fuel [1].
The glass laser is not a perfect solution, however, and in the early years of inertial fusion
many other options were explored. The photolytically pumped iodine laser at 1.3 m was
identied as a promising fusion driver as soon as it was demonstrated in the early 1960s. The
gas medium makes the laser less limited by nonlinear processes and much less expensive than a
solid. The Asterix laser system [3] at the Max Planck Institute for Quantum Optics in Garching,
Germany, and the Iskra laser system [4] at the Research Institute of Experimental Physics in
Sarov, Russia (formerly Arzamas-16), were used in fusion research. Asterix, now operating in
Prague, Czech Republic [5], produces up to 1 kJ in 350 ps, with frequency conversion to 657
and 438 nm. Iskra-5 reached 120 TW in 12 beams in 1991. Pumping a photolytic iodine laser
with explosive-driven light sources, looked very appealing as a low-cost (but single-shot) route
to megajoule energies [6], but precision control proved too difcult for use in fusion
experiments.
166
Fig. 1. In a direct-drive target, laser beams illuminate a fuel capsule uniformly. In an indirect-drive target, they
illuminate the inside of a heavy metal hohlraum surrounding the target and are converted to soft x-rays. The x-rays then
implode the fuel capsule.
The 10.6-m carbon dioxide laser initially seemed an excellent candidate, with high efciency, the
potential for large ampliers in large sizes, and relatively inexpensive construction. The Antares project
(see Fig. 2) [7] at the Los Alamos National Laboratory directed nanosecond CO2 pulses of up 40 kJ on a
fusion target from two nal ampliers, each with 12 roughly square 30-cm subapertures. Unfortu-
nately, the long wavelength of the CO2 laser proved a severe handicap because laser-plasma instabilities
scale with the square of the wavelength, so they are two orders of magnitude larger at 10.6 m than at
1.06 m; therefore CO2 laser fusion was abandoned in 1985.
The 248-nm krypton uoride laser has also been explored as a fusion driver. The short wavelength
is desirable for target interaction, but optics that far in the ultraviolet are difcult to develop. The KrF
laser has broad bandwidth, which is desirable for beam smoothing in direct-drive fusion. At the power
levels needed for fusion, it generates pulses of 100 ns or longer, which must be optically compressed to
the few nanosecond pulses required for fusion. The Nike laser system [8] at the Naval Research
Laboratory has explored KrF technology by stacking 56 pulses through an amplier to give up to 4 kJ
on target in 4 ns, and the Ashura laser system [9] at the Electrotechnical Laboratories, Tsukuba, Japan,
has operated with up to 2.7 kJ in 20-ns target pulses. Figure 3 shows the 6060-cm nal amplier of the
Nike system.
The neodymium glass laser emerged as the most versatile and successful laser system for fusion
research. A major advantage was that its 1.06-m pulses can be converted efciently to the second and
third harmonics at 532 and 355 nm, which proved less vulnerable to laser-plasma instabilities than
longer wavelengths. Xenon ashlamps excite neodymium ions in the glass, which drop to the upper
level of the 1.06-m laser transition. The transition has a lifetime of 300400 ms and a gain cross-
section high enough that energy can be extracted efciently in short pulses with uences tolerable for
laser optics.
Early glass laser systems used cylindrical rods similar in concept to the rst laser, a small
cylindrical rod of ashlamp-pumped synthetic ruby crystal. The Deln laser system [10] at the
Lebedev Institute, Moscow, Russia, used a large array of cylindrical rods serving as subapertures
within a single beamline. Ampliers that used zig-zag laser beam propagation through large laser
glass slabs were also explored [11].
Fusion experiments in the U.S. began in the early 1970s, with three laboratories building a series of
neodymium-glass lasers initially operated at 1.06 m.
Moshe Lubin established the Laboratory for Laser Energetics at the University of Rochester in
1970 and built the four-beam Delta laser in 1972. When the labs new building was completed in 1978,
the six-beam Zeta laser began operation, performing experiments for universities, government agencies,
and industry.
The promise of laser fusion also attracted a private company, KMS Fusion, founded by physicist
and entrepeneur Keeve M. Siegel in Ann Arbor, Michigan. KMS built its own glass laser, and had some
early experimental success, but the company ran short of money. Siegel suffered a fatal stroke while
asking Congress for government support in 1975, and KMS Fusion survived for a time on government
contracts.
John Emmett and Carl Haussmann led development of a series of glass lasers for fusion
experiments at Livermore. The one-beam, 10-J Janus laser conducted the rst fusion shots in 1974.
The one-beam Cyclops laser followed, a prototype of one beam in the 20-beam Shiva laser. The two-
beam Argus laser came on line in 1976, followed in 1977 by Shiva, which reached 10 kJ.
The most popular design for modern neodymium glass lasers with apertures larger than 10-cm is
the Brewsters-angle slab amplier shown in Fig. 4. A laser beam polarized in the plane of the gure
sees no loss when it strikes the slab surfaces at Brewsters angle, and the slab faces are also easily
accessible for ashlamp pumping. Early examples [12] used circular disks of glass, forcing elliptical
beam proles. More modern designs use elliptical or rectangular slabs so that the laser beam can be
circular or square.
Many large glass fusion lasers have been built with those ampliers, such as Gekko [13] at Osaka
University, Japan; Vulcan [14] at the Rutherford-Appleton Laboratory, Didcot, UK; Omega [15] at
the University of Rochester; Phebus at the Commissariat a lEnergie Atomique, Limeil-Valenton,
France; and the sequence of lasers [16] leading to the Nova laser at Livermore completed in 1984. There
Fig. 4. A Brewsters angle slab amplier using neodymium glass. The laser beam sees no loss if it propagates
through this series of slabs with polarization in the plane of the gure. (Courtesy of Lawrence Livermore National
Laboratory.)
have been many others [17]. Nova was the largest of its generation, with ten 46-cm beamlines able
deliver up to 30 kJ at 351 nm in shaped pulses of a few nanoseconds duration for indirect-drive
experiments.
The Omega Upgrade laser at Rochester [15] began experiments in 1995. It delivers 30 kJ in 20-cm
diameter beams at 351 nm in a 64-beam geometry optimized for direct-drive targets. The beams use a
technique [18] called smoothing by spectral dispersion (SSD) to smooth the irradiance to give a very
uniform prole on the target.
The largest fusion laser system now operating [19] is the National Ignition Facility at Lawrence
Livermore National Laboratory (LLNL). Figure 5 is an artists sketch of the facility. It contains 192
laser beamlines of 40-cm square aperture and was designed to irradiate targets with pulses to 1.8 MJ at
the third harmonic (351 nm), and to have very exible output pulses for a wide variety of target
experiments [20].
NIF irradiates indirect-drive targets with conical arrays of beams that illuminate three rings of 64
beam spots each on the inside of a cylindrical hohlraum. This allows experimenters to tune the x-ray
distribution within the hohlraum to optimize target implosions. The NIF beam arrangement can also be
used to drive some direct-drive targets [2123]. SSD smoothing is available if required.
Each beamline includes sixteen slabs, with the beam making four passes through the nal amplier
(see Fig. 6) before exiting and being diretected into the target chamber. Such multipass ampliers reduce
the number of intermediate ampliers and reduce cost of the facility, though they are harder to design
and control than the single-pass amplier chains used for most fusion laser systems in the past. Each
preamplier module in NIF injects about 1 J into each of four adjacent beamlines. The oscillator that
drives the preampliers is a ber laser that uses modulators and other hardware derived from those
developed for ber-optic communications systems.
The Laser Megajoule (LMJ) project under construction [23] by the Commissariat a lEnergie
Atomique at Le Barp near Bordeaux, France, will have amplers similar to NIF, but will have 240
beamlines with 18 slabs each, and somewhat higher energy output capability. An eight-beam prototype
called Ligne dIntgration Laser (LIL) is currently operating.
Omega Upgrade, NIF, and LMJ also will have the capability to deliver kilojoule-class, petawatt-
power picosecond beams to target from beamlines that use grating compression of frequency-chirped
pulses [24]. This capability allows them to explore an advanced target design [25] called the fast
ignition target that uses the main laser output to compress a target, and a separate petawatt
picosecond beam to heat the central spot of the target sufciently for ignition. Target implosion
simulations suggest that such targets will offer higher net gain (fusion energy out divided by laser energy
in) than conventional targets, highly desirable for future applications of laser fusion to energy
production. Other laser facilities also have experimental programs investigating fast ignition. Petawatt
beams are also useful for other experiments such as x-ray backlighting of imploding targets.
The National Ignition Facility succeeded in delivering pulses of more than 1.8 mJ to targets in
2012. However, that design energy proved insufcient to ignite fusion targets. Further experiments
have increased yield, and Livermore researchers are focusing on improving target compression and
reconciling theory with experimental results.
Researchers have long hoped to use laser fusion for electric power generation. The HiPER project
[26] in the European Community, FIREX [27] in Japan, and LIFE [28,29] in the U.S. are all exploring
energy applications of advanced laser fusion concepts. These projects are developing concepts for
high-average-power facilities to follow NIF and LMJ, either with advances from NIF/LMJ-like
technologies or with advanced diode-pumped solid-state lasers that offer higher efciency and better
thermal properties. Large slabs of laser-grade transparent ceramics [30,31], if developed in time, would
be very valuable for advanced laser fusion projects since they offer the laser and thermal properties of
References
1. R. E. Kidder, Laser fusion: the rst ten years, Proc. SPIE 3343, 1034 (1998).
2. J. D. Lindl, Inertial Connement Fusion (Springer, 1998). Most of the technical content can also be
found in the review article by the same author: J. D. Lindl, Development of the indirect-drive approach
to inertial connement fusion and the target physics basis for ignition and gain, Phys. Plasmas 2, 3933
4024 (1995). See also http://www.osti.gov/bridge/servlets/purl/10126383-6NAuBK/native/10126383.
pdf and Lindl and Hammel IAEA NIC plan: http://re.pppl.gov/iaea04_lindl.pdf.
3. H. Baumhacker, G. Brederlow, E. Fill, R. Volk, S. Witkowski, and K. J. Witte, Layout and performance
of the Asterix IV iodine laser at MPQ, Garching, Appl. Phys. B 61, 225232 (1995).
4. V. I. Annenkov, V. A. Bagretsov, V. G. Bezuglov, L. M. Vinogradski, V. A. Gadash, I. V. Galakhov,
A. S. Gasheev, I. P. Guzov, V. I. Zadorozhnyi, V. A. Eroshenko, A. Yu. Ilin, V. A. Kargin, G. A. Kirillov,
G. G. Kochemasov, V. A. Krotov, Yu. P. Kuzmichev, S. G. Lapin, L. V. L'vov, M. R. Mochalov, V. M.
Murugov, V. A. Osin, V. I. Pankratov, I. N. Pegoev, V. T. Punin, A. V. Ryadov, A. V. Senik, S. K. Sobolev,
N. M. Khudikov, V. A. Khrustalev, V. S. Chebotar, N. A. Cherkesov, and V. I. Shemyakin, Iskra-5 pulsed
laser with an output power of 120 TW, Sov. J. Quantum Electron. 21, 487 (1991). See also G. A. Kirillov,
V. M. Murugov, V. T. Punin, and V. I. Shemyakin, High power laser system ISKRA V, Laser Part.
Beams 8, 827831 (1990) and G. A. Kirillov, G. G. Kochemasov, A. V. Bessarab, S. G. Garanin, L. S.
Mkhitarian, V. M. Murugov, S. A. Sukharev, and N. V. Zhidkov, Status of laser fusion research at
VNIIEF (Arzamas-16), Laser Part. Beams 18, 219228 (2000).
5. http://www.pals.cas.cz/laboratory/.
6. V. P. Arzhanov, B. L. Borovich, V. S. Zuev, V. M. Kazanski, V. A. Katulin, G. A. Kirillov, S. B. Kormer,
Yu. V. Kuratov, A. I. Kuryapin, O. Yu. Nosach, M. V. Sinitsyn, and Yu. Yu. Stolov, Iodine laser
pumped by radiation from a shock front created by detonating an explosive, Sov. J. Quantum Electron.
22, 118 (1992).
7. J. Jansen, Review and status of Antares, IEEE Pulsed Power Conference (IEEE, 1979), http://www.
iaea.org/inis/collection/NCLCollectionStore/_Public/16/076/16076157.pdf; Antares main: http://
library.lanl.gov/cgi-bin/getle?00258820.pdf; Antares phase II http://library.lanl.gov/cgi-bin/getle?
00307486.pdf; http://library.lanl.gov/cgi-bin/getle?00258902.pdf. See also H. Jansen, A review of the
Antares laser fusion facility, in Proceedings of 1983 IAEA Technical Committee Meeting on ICF
Research (Osaka University, 1984), pp. 284298 and P. D. Goldstone, G. Allen, H. Jansen, A. Saxman,
S. Singer, and M. Thuot, The Antares facility for inertial fusion experiments-status and plans, in Laser
Interaction and Related Plasma Phenomena, H. Hera, and G. Miley, eds. (Plenum, 1984), Vol. 6,
pp. 2132.
8. http://www.nrl.navy.mil/ppd/nike-facility. R. H. Lehmberg, J. L. Giuliani, and A. J. Schmitt, Pulse
shaping and energy storage capabilities of angularly multiplexed KrF laser fusion drivers, J. Appl. Phys.
106, 023103 (2009) and M. Karasik, J. L. Weaver, Y. Aglitskiy, T. Watari, Y. Arikawa, T. Sakaiya,
J. Oh, A. L. Velikovich, S. T. Zalesak, J. W. Bates, S. P. Obenschain, A. J. Schmitt, M. Murakami, and
H. Azechi, Acceleration to high velocities and heating by impact using Nike KrF laser, Phys. Plasmas
17, 056317 (2010), http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA521400.
9. Y. Owadano, I. Okuda, and Y. Matsumoto, Overview of Super-Ashura KrF laser program, Fusion
Eng. Design 44, 9196 (1999).
10. N. G. Basov, A. P. Allin, N. E. Bykovskii, and B. L. Vasin, Deln-l laser-driven thermonuclear facility:
operating assembly and development trends, Trudy FIAN 178, 388 (1987).
11. M. E. Brodov, V. P. Degtyarova, A. V. Ivanov, P. I. Ivashkin, V. V. Korobkin, P. P. Pashinin, A. M.
Prokhorov, and R. V. Serov, A study into characteristics of a triple-pass amplier using a neodymium
glass slab, Kvant. Elekt. 9, 121125 (1982).
12. S. W. Mead, R. E. Kidder, J. E. Swain, F. Rainer, and J. Petruzzi, Preliminary measurements of x-ray
and neutron emission from laser-produced plasmas, Appl. Opt. 11, 345352 (1972).
13. http://www.ile.osaka-u.ac.jp/research/csp/facilities_e.html.
L
idar and remote sensing grew from developments in optical spectroscopy, optical
instrumentation, and electronics in the 1930s to 1950s. Starting in 1930, searchlights
were directed upward and atmospheric scattering was measured with a separately located
telescope. Starting in 1938, pulsed electric sparks and ashlamps were used in searchlights to
measure cloud base heights. Middleton and Spilhaus introduced the term LIDAR (for Light
Detection and Ranging) in 1953.
The laser revolutionized lidar and launched laser remote sensing. In 1962 Louis Smullen of
MIT and visiting scientist Giorgio Fiocco (who had worked on radar at Marconi) detected
backreection from the Moon using 50-J, 0.5-ms pulses from a Raytheon ruby laser transmitted
through a 12-inch telescope together with a 48-inch receiving telescope and a liquid-nitrogen
cooled photomultiplier at MIT Lincoln Laboratory. (See Fig. 1.) The signal that returned after
2.5 s was very weak, including only about 12 photons, and had to be recorded by photographing
a double-beam oscilloscope trace using vast amounts of Polaroid lm and time. The project
was called Luna-See, probably reecting its difculty. The following year a newly invented
rotating mirror Q-switch shortened a 0.5-J ruby pulse to 50 ns for a series of lidar studies of the
upper atmosphere. The rst use of the term lidar referring to such a laser radar system was used
by Goyer and Watson in 1963 and by Ligda in 1964.
During the next decade advances in laser technology drove improvements in laser remote
sensing. Richard Schotland in 1964 detected the concentration of a gas in the atmosphere for the
rst time by temperature-tuning the wavelength of a ruby laser across a water vapor absorption
line. This was the rst Differential-Absorption Lidar (DIAL) system.
Other groups went on to detect other species. After a detailed theoretical analysis of lidar
techniques by Byer and Kildal in 1971, Hinkley and Kelley showed experimental detection of air
pollutants using tunable diode lasers in 1971, and Byer and Garbuny detailed DIAL requirements
for pollution detection in 1973. Karl Rothe and Herbert Walthers group in Germany used DIAL
with tunable dye lasers to detect NO2 and in 19741976 Ed Murray, Bill Grant, and colleagues at
SRI detected the gas with a tunable CO2 laser. Menzies and Hinkley in 1978 measured
atmospheric gases with a laser absorption spectrometer (LAS), two waveguide CO2 lasers, and
stripchart recorders mounted in a plane (see Fig. 2). In 1979, they measured atmospheric gases
with the balloon-borne Laser Heterodyne Radiometer shown in Fig. 3. Sune Svanbergs group at
the Lund Institute mapped the mercury emission from coal-red power plants in a seminal DIAL
study in the 1980s, Jack Bufton at NASA Goddard measured atmospheric CO2 in 1983, Ed
Browell at NASA Langley measured water vapor and ozone in the atmosphere and the ow of
Sahara Desert dust from Africa to the Southeast United States, and Nobuo Sugimoto and
Kazuhiro Asais group measured similar Asian dust ow.
DIAL also performed landmark environmental observations. In 1993, Bill Heaps group at
NASA Goddard and Stuart McDermids group at JPL tracked variations of stratospheric ozone
levels in time and space for the rst time, validating data suggesting an ozone hole collected
by solar occultation instruments on NASA satellites in the 1980s. The satellite sensors had
detected the hole years earlier but had not transmitted the data to the ground because the
software considered the measured ozone levels too low to be accurate. The problem was
175
Fig. 1. Photo of Luna-See, the rst laser radar measurement of a laser beam backscattered from the Moon
(white speck at the upper left) in May 1962 at Lincoln Laboratory by MIT Prof. Louise Smullen (left), Raytheon
laser scientist Dr. Stanley Kass (middle), and visiting radar scientist Dr. Giorgio Fiocco (right). (Courtesy MIT
Museum.)
Fig. 5. Good lidar friends attending a banquet dinner at the 17th International Laser Radar Conference in
Sendai, Japan in 1994. (Left to right) bottom: Takao Kobayashi, Pat McCormick, Chet Gardner, Dennis Killinger,
Jack Bufton; top: Akio Nomura, Osamu Uchino, Hiromasa Ito, Yasuhiro Sasano, Kazuhiro Asai, Toshikazu Itabe.
detection techniques such as femtosecond absorption spectroscopy. It is hard to predict the future, but it
is certain that major technical improvements will occur : : : they always have. As the technology
continues to improve and laser remote sensing and lidar techniques become more widely accepted,
we will nd uses for lidar in applications not yet imagined.
It is sobering to recall that 40 years ago we thought that the main use of lidar and laser remote
sensing was going to be akin to Star Trek where Spock scans the distant planet surface with a laser
beam and tells the Captain that there are two humanoids on the planets surface and one has a bad
kidney. Who would have guessed back then that one of the huge commercial successes for lidar today
would be mapping of urban buildings and geological features, nding buried Mayan ruins, mapping
wind elds for wind farms, detecting and mapping global climate change gases and pollutants in the
atmosphere, and laser sensing of pharmaceuticals and chemicals at close ranges.
Introduction
Michael Bass
I
n 1980, just 20 years after the rst laser was demonstrated and about 10 years after the way
to make low loss optical bers was discovered, two miracles took place: one that lots of
people noticed and that some recall and another that few noticed and that changed the course
of human history. At the Winter Olympics at Lake Placid, New York, the Miracle on Ice in
which the U.S.A. mens hockey team beat the much vaunted Soviet Union team was seen by tens
of millions on televisionlots of people noticed. However, the television broadcast of the
Olympic Games, including the hockey match, was transmitted over an optical communications
system using diode lasers and ber optics. Virtually no one noticed this miracle at the time, but
many billions would be affected by the technology. Optics changed the world and commu-
nications would never be the same. This section presents the pivotal events and technologies
leading to optical ber communications becoming practical.
Perhaps a few people in the mid-1970s could have foreseen that ultra-low-loss optical bers
and diode lasers would enable optics to take over the world as the dominant means of
communications. Optics did just that. Not only are billions of kilometers of ber optics
communication cables in use with diode lasers as the light sources but progress continues as the
demand for more and more information-carrying capacity continues to grow. New techniques
for multiplexing are still being developed to enable higher throughput.
The invention of the laser and the demonstration of nonlinear optics spurred a greatly
renewed interest in optics. In the period 19751990 that interest blossomed into many major
applications and scientic breakthroughs. Nonlinear optics beneted from demonstration of
excellent new materials for use in both the visible and the infrared. Periodically poled nonlinear
material had been described as early as 1962 but was nally demonstrated in this period. It
turned out that the periodically poled material was often a more efcient harmonic generator
than its single-crystal index-matched version. These materials and greatly improved engineering
made optical parametric oscillators and ampliers available for applications requiring wave-
length tunable sources. Nonlinear optics also made possible achieving ultrashort pulses,
6 picosec in this period (today 67 attosec) and supercontinuum pulses with spectral content
exceeding an octave in frequency.
The list of applications of optics that developed in this period is too long to list in its
entirety here. However, a few are worth mentioning because they are so common that the
outstanding optics and optical design that makes them possible can be easily overlooked. They
are the bar code scanner, the CD/DVD player, the laser printer, the laser pointer, the laser cut,
the drilled or welded part of a nished product, the laser-marked product, the variable-focus
spectacle lens, self-darkening spectacle lenses, soft contact lenses, the optical mouse, and the
remote control for an appliance, as well as the display screens of televisions, computers, and
mobile phones.
Between 1975 and 1990 developments of new lasers and their applications spurred
demonstration of new medical innovations. The LASIK technique for vision correction based
on the use of an excimer laser was developed and has now been used on 30,000,000 patients.
Optics and ber optics have made detecting pathologies in patients more reliable and less
invasive. Laparoscopic surgeries are performed today with minimal cuts because ber optic
endoscopes or miniaturized cameras can be inserted to give the surgeon vision of the problem
that must be dealt with. Photodynamic therapy in which a laser is used to excite a dye that
183
preferentially locates in tumorous tissues is another area in which optics and medical treatment have
come together.
During this period spectacular progress was made in optical astronomy. The Hubble Space
Telescope was launched and, after its optics were repaired, it performed spectacularly. It provided data
on the content of the universe such as the number of galaxies and the presence of dark matter
surrounding galaxies. Ground-based telescopes were designed and built that took advantage of
adaptive optics to build large-aperture, segmented-mirror instruments that can minimize atmospheric
distortions and provide superb images. These telescopes could be much larger than space telescopes
and could gather more light from distant objects. Using image processing techniques and modern
computers, it is now possible to link optical telescopes to greatly enlarge their effective aperture.
Whenever the eld of optics is mentioned to non-optics people in the eld of optics, they
immediately think of their eyeglasses or contact lenses. And why not? Almost everyone will use
spectacles or contacts at some point in his or her life and if they live long enough will have an implanted
lens as part of cataract surgery. Progress in these areas has been remarkable. Contact lenses were
invented that allow air to pass through, enabling long periods of comfortable wearing. In addition,
contact lenses can now provide astigmatic correction. Spectacle lenses with continuously variable
strength eliminated the need for bifocal lenses with a sharp delineation between near and distance
viewing sections. Then photochromic lens materials became available enabling the wearer to no longer
need different spectacles indoors and outdoors; the lenses would lighten and darken according to the
ambient light environment.
By 1990 optics included light sources from continuously operating very stable lasers to lasers
producing pulses as short as a few picoseconds (now a few tens of attoseconds). Optics included
components small enough to be swallowed to 30-meter-diameter segmented telescope mirrors.
Displays were getting so small as to be worn in a head-mounted device or so large as to be seen
by 100,000 people in a stadium. Most interesting and important was that applications of optics beyond
those that aid vision had become part of everyday life and so ubiquitous that most went unnoticed.
184 Introduction
19751990
I
n the earliest days of the past century, advancements in optics were led by newly created
optics companies: Kodak and its research laboratory, Bausch & Lomb, and the American
Optical Company. George Eastman led the effort to found the Kodak Research Laboratory
in 1912 because he saw the connection between optical science and development of new
products. The Institute of Optics at the University of Rochester was not founded until 1929,
after ten years of discussions. As for government, Thomas Edison urged in 1915 that a national
laboratory be formed to attack issues faced by the U.S. Navy. While this resulted in the
establishment of the Naval Research Laboratory in 1923, the (Physical) Optics Division was
not formed until after World War II.
In July 1945 during the closing days of World War II, Vannevar Bush, the Director of the
Ofce of Scientic Research and Development, in response to a request from President Franklin
Roosevelt issued an extensive report entitled Sciencethe Endless Frontier, which urged the
government to establish and fund a broad program in science and applied research to ght
disease, develop national security, and aid the public welfare. It urged that basic science and long-
term applied research be supported in universities, that nearer-term applied research and
development be funded in industry, and that military research be increased and tied to university
and industry R&D programs as appropriate. It estimated the cost of this program to be $10
million at the outset rising to perhaps $50 million within ve years. One of the recommendations
was to create the National Science Foundation
Congress created the Ofce of Naval Research (ONR) in 1946 with the Naval Research
Laboratory being its principal operational arm. In light of the wartime success in developing the
proximity fuse, the Division of Ordnance Research was transferred from the National Bureau of
Standards to create the Armys Diamond Ordnance Fuse Laboratory. The Army also created a
laboratory for electronics research at Ft. Monmouth in New Jersey. The Air Force was spun out
of the U.S. Army in 1947, leading to the creation of the Wright-Patterson Air Force Base
Laboratories in Dayton, Ohio; the Air Force Cambridge Research Laboratory in Cambridge,
Massachusetts, which had Infrared Optics as one of its major divisions; and the Air Force
Weapons Laboratory in Albuquerque, New Mexico. Further, the MIT Radiation Laboratory at
MIT, which was so successful during the war in radar development, was expanded and relocated
near the small town of Lincoln, Massachusetts, and renamed the MIT Lincoln Laboratory. All of
these played a major role in modern optics and laser development.
Corporate labs were established and grew after the war. Some of them were at GE, Bell Labs,
RCA Laboratories, Hughes Research Laboratory, Westinghouse Research Laboratory,
Raytheon, Texas Instruments, Perkin-Elmer, and Boeing. Figure 1 is an aerial photo of the
iconic Bell Holmdel Laboratory. The growth of corporate labs was aided by scal help that
resulted from the Vannevar Bush report and two events that accelerated the science and
technology of and funding for optics dramatically: the launch of the Soviet Sputnik in 1957
and the demonstration of the laser in 1960.
In 1958 in direct response to Sputnik, President Eisenhower created the Advanced Research
Projects Agency (ARPA) within the Defense Department. One of the U.S.s limitations was a lack
of broad and deep materials capability. Thus, ARPA initiated the Interdisciplinary Laboratories
185
(IDL) program in 1960 to ensure that
chemists, physicists, and electrical and me-
chanical engineers work together to solve
the difcult research problems in materials
development. This program led to the cre-
ation of the eld of materials science.
The 12 universities funded in this program
were MIT, Harvard, Cornell, Illinois,
Stanford, University of Pennsylvania,
Maryland, Brown, Chicago, Northwestern,
Purdue, and University of North Carolina.
A major success of the IDL program was
the development of the science and technol-
ogy of electronic materials, especially III-V
Fig. 1. Aerial view of Bell Holmdel Laboratory. (Courtesy of materials such as GaAs and ternary and
AT&T/Bell Labs.)
quaternary mixtures of them. These mate-
rials systems have been the success story of
diode lasers and photonics more generally, and the scientists who went on to industrial laboratories to
develop these materials systems for specic applications in optics were likely trained in one of the IDLs.
With government funding enabling universities to supply highly skilled people to industry who
would lead in the revolution in optics brought on by the laser, we will concentrate on that history
because it is in many ways symbolic of the transitions that took place in basic research in optics.
This is not to say that other subjects such as advances in still and motion picture photography, CCD
cameras, polaroid photography, electrophotographic (xerographic) copiers, laser printers, point-of-
sale scanners, optical storage devices, laser machining, and optical communication systems
could not show the same transitions; it is just that the laser revolution presents the changes most
powerfully.
Simultaneously with the initiation of the IDL program was the demonstration of the rst laser. This
occurred at an industrial research laboratory using internal fundsthe Hughes Research Laboratory
(HRL) in Malibu, California, on 16 May 1960. As soon as other corporate labs heard in July of Ted
Maimans success, their efforts accelerated. TRG, a small company on Long Island, New York, had
been funded by ARPA in 1959 to the tune of $990,000 for laser development and is thought to be the
rst to duplicate Maimans result. A number of military labs including MIT Lincoln Laboratory
immediately initiated laser programs. The author was a 1st lieutenant in the U.S. Air Force at that time
stationed at the Air Force Cambridge Research Laboratory (AFCRL) in Bedford, Massachusetts. He
and Rudolph Bradbury had a ruby laser like Maimans operating by November 1960. A request of
$392 was made for the purchase of capacitors and ashlamps. This request was immediately approved,
as everyone was excited about the prospects of having an operating red laser!
Military labs like AFCRL, Wright-Patterson Air Force Base, and Air Force Weapons Laboratory
(AFWL) typically had sufcient funding not only to fund their own projects but also to fund industrial
and university proposals in areas of laser R&D that they deemed important. So the decade of the 1960s
was one of intense laser activity, especially in the development of laser range nders and target
designators at HRL and other companies, coherence studies of partially coherent lasers at Rochester
and Brandeis and at TRG, the phenomenon of mode locking that was discovered in Nd:glass lasers by
Tony DeMaria of United Technology Research Center in Connecticut, the development of parametric
oscillators using LiNbO3 at Bell Labs by J. Giordmaine and R. C. Miller and at Stanford by Steve
Harris, the study of the dynamics of laser operation at the University of Rochesters Institute of Optics
by Mike Hercher, and laser-induced damage to ruby and glass at HRL by Connie Guiliano and at
American Optical Company by Charles Koester. These damage studies were funded by ARPA, but the
other efforts (with the exception of the research on parametric oscillators at Bell Labs) were funded with
military laboratory and ONR monies.
Meanwhile, with corporate funding at Bell Labs, Kumar Patel developed the CO2 laser in 1964,
and Joe Geusic developed the Nd:YAG laser in the same year; both lasers are still workhorses today.
186 The Shift of Optics R&D Funding and Performers over the Past 100 Years
At American Optical Company, Elias
Snitzer developed the rst Nd:glass rod
laser as well as a Nd:glass ber laser.
About that same time, Bill Bridges of HRL
achieved lasing of argon and krypton.
While these achievements were extremely
noteworthy, looking back at that decade,
the most signicant achievements for
the U.S. telecommunications industry were
the developments of GaAs homojunction
(diode) lasers in 1962 at GE by Robert N.
Hall and N. Holonyak, Jr., and at IBM by
Marshall Nathan using corporate funds,
and by T. M. Quist and R. J. Keyes at MIT
Lincoln Laboratory, which had block Fig. 2. Aerial view of IBM Watson Laboratory. (Courtesy
of IBM ResearchZurich. Unauthorized use not permitted.
funding by the U.S. Air Force. Initially, Copyright owner is IBM Zurich at http://www.zurich.ibm.com/
these lasers had to be cooled to liquid imagegallery/.)
N2 temperatures or below and could oper-
ate only as pulsed devices. It took the insight of Herb Kroemer of Varian Associates in Palo Alto,
California, using corporate funds, to realize that if one formed a heterojunction at both sides of the
homojunction where lasing was occurring, the greater bandgap at the heterojuction would prevent
carrier diffusion away from the homojunction, thus leading to the rst continuous-wave diode laser a
year later. Kroemer received the Nobel Prize in 2000 for this achievement. Figure 2 is an aerial photo of
the IBM Watson Laboratory.
With ARPA funding, Roy Paanenen at Raytheon demonstrated a 100-W argon laser that required
a huge ow of cooling water. Also at Raytheon, Dave Whitehouse was the rst to demonstrate a 1-kW
laser with a longitudinal gas-ow CO2 system that seemed as large as a tennis court. Ed Gerry, with
ARPA funding, at AVCO/Everett Research Laboratory developed a owing gas-dynamic CO2 laser that
had the potential for smaller size and ultra-high power because the waste heat in the gaseous medium
could be removed by owing the gas transversely out of the laser resonator. AVCO/Everett with
continued ARPA funding went on to achieve very-high-power operation of the CO2 laser as well as
high-peak-power pulsed operation of rare gas lasers.
As the powers that were achieved by the CO2 laser were high enough to fracture the transparent
materials that were then available, a new effort had to be made to develop better optics for such lasers.
Consequently, the author departed AFCRL in 1971 for ARPA to lead efforts to develop highly
transparent windows and reecting and anti-reecting coatings. The best of the window materials that
were developed were ZnSe and ZnS, and BaF2 by Raytheon (Jim Pappis). Coating development was led
by Maurice Braunstein at HRL and resulted in thorium-containing coatings with reection coefcients
exceeding 99%. Supporting university and industrial contractors were involved in these programs, with
their roles ranging from modeling of optical distortions in high-power windows to development of
techniques to measure absorption coefcients as low as 0.00001 cm-1.
In the 1970s and 1980s, changes began to occur in the corporate world that led the corporations to
reduce funding of research. First, Wall Street and the stock market expected companies to make their
numbers on a quarterly basis as failure to do so would result in stock prices dropping. This led to
corporations investing their money in the short term to the detriment of funding research that paid off
mostly in the long term. Second, it was becoming apparent to management that these labs were perhaps
more of a drain on prots than the corporation could afford as the research labs did not seem able to
convert research results to products that would boost sales. Third, the rise of globalization meant that
these companies faced competition around the world that had not mattered previously. Fourth, the U.S.
Congress had initiated the Small Business Innovative Research Program to fund product development
at businesses with fewer than 500 employees. Each agency of the federal government that had R&D
funds was (and is) required to set aside 2.5% of these funds for such awards. In 1995 this amounted to
$950 million for product development by small businesses. While this is small compared to what U.S.
The Shift of Optics R&D Funding and Performers over the Past 100 Years 187
corporations spend annually for R&D, the availability of such funding attracted people to leave
corporate research laboratories to develop their new ideas rather than attempt to do so in the corporate
environment.
At this point, it is natural to ask, Why werent the research labs more efcient at developing new
products? It seems that the researchers were just not close enough to the companies customers to
know what was needed or what could be improved upon [1]. So large companies began cutting back
their research laboratories in the 1970s1990s, if not eliminating them altogether, and moving their
best R&D people nearer to the front line. Instead of looking for major breakthroughs such as a laser,
they concentrated instead on, as the Economist writes, tinkering with todays products rather than pay
researchers to think big thoughts. More often than not, rms hungry for innovation look to mergers
and acquisitions with their peers, partnerships with universities, and takeovers of venture-capital-
backed start-ups [1].
The several changes mentioned above led to a shift of basic research and long-term applied research
to universities and, to a smaller extent, government laboratories. While various government agencies
still fund individual investigator proposals in optics, there has been a dramatic growth in Multi-
University Research Initiatives (MURIs)designed to tackle important long-range development
objectives. MURIs involve universities and private companies that would be likely to commercialize
the developments of the research done in the MURI. These MURIs thus take on development efforts
that, 30 years ago, would have been done by a company that had its own research laboratory to
perform the fundamental work necessary to develop the new product.
Reference
1. http://www.economist.com/node/8769863.
188 The Shift of Optics R&D Funding and Performers over the Past 100 Years
19751990
T
echnological breakthroughs develop through years of scientic collaboration and inno-
vation, each discovery built upon the failures and successes of earlier work. Such was the
case with the work on the rst low-loss optical ber. What began with three Corning
scientists searching for a communications solution ultimately created what is now known to be a
key to the Information Age.
In 1948, Claude E. Shannon [1] proved that optical carrier frequencies provided greater
bandwidth than radio or microwave frequencies. But the technology of the day had not yet
caught up with the science. Those looking to apply Shannons work lacked a suitable light
source, modulator, and detector technology as well as any kind of transmission conduit.
Then in 1960, Ted Maiman [2] demonstrated the rst laser. A few laboratories saw it as a
source for optical communications with the bandwidth that Shannon described and began to
research that application. However, it could not be implemented because at that time, a suitable
transmission conduit for light had not yet been invented.
Corning learned of the growing interest in optical communications on 17 June 1966, when
one of its scientists, William Shaver, brought back a request from the British military. They
wanted a single-mode ber (100-m diameter with a 0.75-m core) with a total attenuation of
less than 20 dB/km. This was prior to any publication, such as the Kao and Hockham paper [3],
suggesting that optical bers could be used as a practical communications conduit. The very best
bulk optical glasses of the day had attenuation of around 1000 dB/km. The British request
required an improvement in transparency of 1098 to reach the 20 dB/km goal. Given the science of
the time, it was seemingly impossible. But within Cornings culture of scientic innovation
particularly when it came to discovering new applications for glassan impossible goal was
merely a problem yet to be solved.
This particular problem was handed to Robert Maurer, a physicist known for his work on
light scattering in glasses. Though Bob did not know it at the time, he actually had begun his ber
work a decade earlier. He published two denitive works in 1956 [4] and 1960 [5], indicating
that Cornings ame-hydrolysis fused silica had the lowest Rayleigh scattering of all glasses he
had measured.
These studies were built upon the discoveries of two giants within Cornings history, Frank
Hyde [6] and Martin Nordberg [7]. In 1930, Hyde demonstrated that when vapors of silicon
tetrachloride were passed through a ame in the presence of oxygen, they would hydrolyze to
form a ne powder of very pure silicon dioxide that could be fused into very pure silica glass. He
noted that the normal glass impurities that give rise to absorptive losses in the glass were low.
Nine years later, Nordberg added titanium tetrachloride to Hydes process and formed a very-
low-expansion doped fused silica glass.
While these processes had been used at Corning for years, Bob took them in innovative
directions that, ultimately, laid the foundation for the Corning groups invention of low-loss
optical ber. Always the contrarian, and inuenced by his earlier work on light scattering, Bob
189
and a summer intern made a rod-in-tube
(RIT) ber (Fig. 1)the best known pro-
cessing method at that timeusing Corn-
ings fused silica as the cladding. He
purposely added an impurity to the fused
silica to raise the refractive index of the
core, Nordbergs titanium doped silica, and
obtain light guidance. Losses were still very
high, but Bob was encouraged enough to
request two additional scientists, Peter
Schultz and Donald Keck (the author).
Peter took a fresh look at Hydes
ame hydrolysis process. He built a small
Fig. 1. Illustration of RIT and thin-lm processes for making boule furnace and began making various
an optical ber preform. (Courtesy of Corning Incorporated.)
doped fused silicas and measuring their
properties. Based on Bobs earlier results, the group of three focused their efforts exclusively on fused
silica bers made by ame hydrolysis. They continued the counterintuitive approach, adding an
impurity to the pure fused silica to raise the refractive index and create the ber core.
So began a time of trial and error. No human endeavor progresses more rapidly than can be
measured. The group began to systematically measure and identify the sources of their optical losses.
They knew absorptive losses were one source, and they struggled to examine the impurities introduced
in the ame hydrolysis glasses that could cause absorption. The best analytic equipment of the day
could measure impurity levels only to the parts-per-million level, and parts-per-billion were needed. An
attempt was also made to evaluate losses in a few centimeters of bulk glass, but this still could not
produce the losses in an actual ber that had gone through all the processing steps. Making their own
bers was the only way to get a thorough understanding of optical losses.
Optical absorption from formation of reduced-titanium (Ti3+) color centers during the high-
temperature ber drawing step accounted for about half of the ber loss. At rst the losses were
annealed away by heat-treating the bers at 800C to 1200C. Unfortunately this treatment drastically
weakened the bers as a result of surface crystallization. The other half of the loss originated from light-
scattering defects at the corecladding interface. No publication of the day ever mentioned this most
signicant source of loss. The Corning group believed that this loss originated during the RIT process
from dirt in the lab environment.
With each failure a little more was learned until an idea was hit upon that proved to be the key: the
traditional RIT method was abandoned and a new approach was invented. Rather than inserting a core
rod, the group decided to directly deposit a thin layer of core glass inside a carefully ame-polished
cladding tube (Fig. 1). This produced intimate contact between core and clad materials and, it was
hoped, would get rid of the scattering defects observed in the RIT ber.
For those who believe that excellent work can be done only with the very latest equipment, take
note of the Corning lab pictured in Fig. 2. The equipment was crude but effective. A portable lathe
headstock held the rotating cladding tube in front of the ame hydrolysis burner. The burner produced
a soot stream containing titania-doped silica. Initially the soot would not go into the 56-mm hole in
our cladding tube. One of the group spotted the lab vacuum cleaner. Putting this at the end of the
cladding tube beautifully sucked soot from the ame and deposited a uniformly thin layer onto the
inside tube surface. This coated tube was then placed in the ber draw furnace where the soot sintered
into a clear glass layer, the hole collapsed to form a solid rod containing the doped core, and the entire
structure was drawn down into ber.
Measuring that rst low-loss ber was an unforgettable experience. It was late afternoon, and, after
heat-treating a piece of the groups latest ber, the author positioned it in the attenuation measurement
apparatus. With a viewing telescope he could observe and position the focused He-Ne laser beam on the
ber end. When the laser beam hit the ber core, a blindingly bright returning laser beam was produced.
It took a moment to realize that the laser was being retro-reected off the far end of the ber and coming
back through the optical system.
The brilliant laser beam emanating from the end of the ber was so dramatically different from
anything previously seen that it was apparent something special had occurred. With considerable
anticipation, the author measured the ber loss, and to his delight and surprise it was 17 dB/km. With
little sense of history, Donald Kecks excitement was registered in his now fairly well-known lab-book
entry: Whoopee! (Fig. 3).
In 1970 the result was announced to the world when Bob presented the Corning groups paper
Bending losses in single-mode bers at an Institution of Electrical Engineers Conference in London
on analog microwave technology [8]. In that paper, he mentioned that the ber had a total attenuation
of only 17 dB/km, prompting scientists at the conference to remark that at least their 2-in. helical
microwave guides could be lled with lots of optical bers. We also submitted our paper to Applied
Physics Letters, and it was initially rejected! The reviewer commented, It is rather difcult to visualize
an amorphous solid with scattering losses below 20 decibels per kilometer, much less the total
attenuation. Eventually, however, the paper was published [9]. (See Fig. 4.)
The Corning group had done it, but they were far from done. Though revolutionary, their
breakthrough ber solution was not exactly robust. Only small preforms could be made, and the
heat treatment required to achieve low attenuation made the bers brittle. Also, the preferred ber
design had shifted to multi- rather than single-mode. The larger core diameter was believed necessary
to more easily couple light into the ber from the relatively crude semiconductor lasers of the day.
To make such bers, Peter, our colleague Frank Zimar, and the author invented another ame
hydrolysis approach later dubbed outside vapor deposition. In this method, rst core and then
cladding soot were deposited onto a removable rotating rod to build up a porous soot preform. Because
of the lower temperature in this process, Peter found he could incorporate new dopants that had
vaporized in the higher-temperature boule process. One of these dopants was germania, a glass former
like silica.
In June 1972, the rst ber incorporating germania was drawn in the core. The group was
obviously on the right track, as the bright light of the draw furnace was still visible through the end of a
kilometer of ber on the wind-up drum. The loss measured was only 4 dB/km, no heat treatment was
needed, and ber strength was excellent. This was the rst truly practical low low-loss ber.
This writing marks the 42nd anniversary of the Corning groups invention of low-loss optical ber.
With more than 1.6 billion kilometers of it wrapped around the globe, a world has been created that is
dependent upon reliable, speed-of-light access to people and information anywhere, anytime, through
almost any device of their choosing. The dramatic increase in users has brought with it unprecedented
demand for bandwidth. Several sources, including a University of Minnesota Internet Trafc Study and
Cisco, have estimated that the average Internet trafc today worldwide is 150 Tb/s and growing at
about 50% per year.
This growth rate is not surprising. Collectively we have moved from simple audio to increasing
video content in our communications. Estimates are that two-thirds of the mobile data trafc will be
video by 2015 as social networking continues to explode. People sending data is one thing, but
machines-talking-to-machines (M2M) as is happening increasingly is yet another. The latter will
overtake the former in just two or three yearsall this without even considering potential new data-
generating applications. We are already seeing the deployment of ber-enabled remote sensors to
monitor our environment. Power lines and highway and civil structure monitors provide an optical ber
safety net supporting the infrastructure we rely upon every day. Emerging biomedicine and biotech-
nology applications ranging from transmission of x-ray data to real-time high-denition video for
remote surgeries to the potential petabytes involved in DNA data transmission and analysis are still in
the future. It is now well established that creative people will invent new ways to use the bits if
technology can provide improved cost of transmitting the bit.
The amount of information that can be transmitted over a single ber today is staggering.
Commercial core networks today operate at 50 Tb/s on a single ber, and as reported at OFC
2012, scientists are achieving in their labs record data rates of more than 305 Tb/s.
While this capacity is enormous, ber bandwidth is niteperhaps only 10 times higher
than todays core network trafc level. Our current demand for bandwidth will most likely exceed
our capacity before 2030. This would require a beginning over-build of the core networks even
as we nish the build-out of the local loop! We should not be surprised if the 1.6-billion-
kilometer ber network of today will be but a fraction of that which will exist in just a couple of
decades.
But beyond all the bits and bytes, the most important story of the communications revolution
brought about by optical ber may well be the one about improving human lives. All of us who have
worked and continue to work in optical ber communications technology have truly made the world a
better placeand for that we should be proud.
When asked about glass, most people still picture something breakable that shatters when dropped.
But low-loss optical ber has shown us that hair-thin strands of glass lled with light are strong enough
to help people all over the world shatter long-held assumptions and break down centuries-old political
and cultural walls.
In 2000, the United Nations created the Millennium Project, aimed at lifting millions of people in
the developing world from impoverishment, illness, and death. One of the primary methods for
achieving that objective was to deploy the benets of optical ber technology for their education and
economic betterment.
The International Telecommunications Union continues to track progress toward that end. In 2011
they reported that today, thanks to optical ber, more than two billion people around the world are
instantaneously and simultaneously accessing the Internet, virtually 75% of the worlds rural popula-
tion has cell phone coverage, and more than 60% of the worlds countries have a National Research
and Education network.
We have come a long way since we rst stood on the shoulders of those giants of early optical
communications. Today the optical ber network has become the lifeblood of our society, providing the
medium through which commerce and culture are being simultaneously created and communicated on
a personal and global scale. We can never be sure just what the future of optical communications holds,
but given the remarkable history of low-loss ber, it is fairly certain to be a future full of light.
References
1. C. Shannon and W. Weaver, The Mathematical Theory of Communication (University of Illinois Press,
1949).
2. T. H. Maiman, Stimulated optical radiation in ruby, Nature 187, 493404 (1960).
3. C. Kao and G. Hockham, Dielectric-bre surface waveguides for optical frequencies, Proc. IEE 113,
11511158 (1966).
4. R. D. Maurer, Light scattering by neutron irradiated silica, J. Phys. Chem. Solids 17, 4451 (1960).
5. R. D. Maurer, Light scattering by glasses, J. Chem. Phys. 25, 12061209 (1956).
T
he deployment of the worlds optical telecommunication network starting in the 1980s
was a major change of paradigm in modern society that enabled the Information Age.
From a technical standpoint, of the many technologies without which this colossal
achievement would have never seen the light of dayfrom frequency-stable laser sources to
efcient low-noise detectors, division wavelength multiplexers, optical lters, and low-noise
high-speed electronicsperhaps none was as decisive and challenging as the ber-optic amplier
(FOA) in general, and the erbium-doped ber amplier (EDFA) in particular. Like the optical
ber itself, the EDFA had no good alternative; had it not existed, no other component would
have been available, then or now, to perform its vital function as nearly perfectly as it does.
The basic idea of transmitting data encoded on light carried by optical bers dates back to at
least the 1960s. Early incarnations of optical communication links used electronic repeaters that
periodically detected, amplied, and remodulated the traveling light signals. Such repeaters
worked adequately for high-speed communications over planetary distances, but they required
power and costly high-speed electronics. By then the potential of replacing them with optical
ampliers, devices that would amplify the modulated signals without the need for electronics,
had already been formulated. Optical ampliers already existed, and they offered, at least on
paper, multiple advantages, including an unprecedented bandwidth in the multiterahertz range.
Yet it took nearly three decades of gradually intensifying research in numerous laboratories
around the world to turn this concept into a reality, which involved, among other things,
developing a practical optical amplier utilizing a ber as the gain medium.
From the start, the development of FOAs was riddled with challenges. To be successful in a
communication network, an amplier had to meet tough criteria. It had to provide a high, nearly
wavelength independent gain over a broad spectral range while also incorporating an efcient
means of mixing the excitation source with the incoming signal, being internally energy efcient,
preserving the single-mode character of the trunk ber, and inducing negligible crosstalk between
channels. In later years other requirements were added to this list that further complicated the
task. In retrospect, it is easy to trivialize the now well-known solutions to these problems. But
back in the 1970s and 1980s when these problems were being tackled, there was nothing obvious
about them, and, as in other scientic pursuits, many potential solutions were proposed, tested,
and discarded.
The rst report of amplication in a ber appeared in a famous article published in 1964 by
Charles Koester and Elias Snitzer in The Optical Societys (OSAs) Applied Optics, just four years
after the demonstration of the rst laser [1]. This historic amplier consisted in a 1-m Nd-doped
glass ber coiled around a pulsed ashlamp and end-probed with 1.06-m pulses. This visionary
device already contained several of the key elements of modern FOAs, including a clad glass ber
doped with a trivalent rare earth, an optical pump, and means of reducing reections from the
ber ends to avoid lasing. It provided a small-signal gain as large as 47 dB, which is remarkable
considering that it came out so early in the history of modern photonics. For his many
195
contributions to the elds of FOAs and lasers, Elias Snitzer was awarded the OSAs Charles H. Townes
Award in 1991 and the John Tyndall Award in 1994.
Like almost all the laser devices of the time, this ber amplier was side-pumped: the pump was
incident on the ber transversally. This made the device bulky, inefcient, and ultimately impractical.
The concept of a ber amplier in which the pump is end-coupled into the ber emerged years later as
part of efforts carried out at Stanford University to develop a compact ber amplier. This work
involved end-pumping Nd-doped crystal bers with an argon-ion laser. This work demonstrated that
end-pumping could produce sizeable gain (5 dB) from a very short ber (cm).
The second key improvement was the introduction of the wavelength-division-multiplexing
(WDM) coupler to mix the pump and the signal and end-couple them simultaneously into the gain
ber. The advantages of this technique were overwhelming: it made it possible to efciently inject, with
a compact and mechanically stable device, both the pump and the signal into the gain medium. It took
several years before it was adopted, in part because commercial WDM couplers were almost
nonexistent. It is now the standard technique used in the vast majority of FOAs.
Another concept critical to the performance of FOAs in general, and bench-tested rst with EDFAs,
is that they should use a single-mode ber. Although in recent years new ndings have suggested that
the data transmission capacity could be increased by using multimode bers, in current telecommuni-
cation links a single-mode FOA offers two key advantages, namely, a higher gain per unit pump power
due to the higher pump intensity and the elimination of modal coupling, which would otherwise induce
time-dependent losses at the trunk ber/FOA interfaces.
It was known as early as the 1980s that the third communication window, centered around
1550 nm, was the most promising candidate for long-haul ber links, because in this spectral range both
the loss and dispersion of conventional single-mode silica bers are minimum. Trivalent erbium ions
(Er3+) in a variety of amorphous and crystalline hosts had also long been known to provide gain in this
wavelength range, so this ion was a natural candidate. David Payne, who would receive the OSAs John
Tyndall Award in 1994 for his pioneering work on EDFAs, and his team at the University of
Southampton were rst to demonstrate this potential experimentally with the report of the rst EDFA
in 1987 [2,3]. This was followed later the same year by a similar paper from Bell Laboratories. These
milestone publications provided experimental proof that single-pass gains exceeding 20 dB were readily
attainable in single-mode Er-doped bers (EDFs) end-pumped with the best laser wavelengths available
at the time, namely, 670 nm and 514.5 nm. Another key property that made Er3+ so attractive is that the
lifetime of its 1550-nm transition is unusually long (510 ms); hence the population inversion and the
gain essentially do not respond dynamically at the very high modulation frequencies of the signals
(unlike semiconductor ampliers). The important
consequence is that the crosstalk between signals
being amplied simultaneously in an EDFA can be
exceedingly small (see Fig. 1) [4], a crucial property
for communications.
The EDFA seemed to be a great candidate, but
several issues, some perceived to be critical by the
communication community, made it difcult to be
accepted right away. In fact, it took nearly another
decade of detailed engineering and the development
of several parallel technologies (diode lasers and
fused WDM couplers, in particular) to make this
device a reality.
To be practical, the EDFA had to be pumped
with a semiconductor laser. Over time several pump
Fig. 1. Measured crosstalk between two channels, sources and wavelengths were investigated. This
characterized by the peak-to-peak gain variation battle was one of the most technically challenging
induced in a rst signal (channel A) by a second signal and interesting in the history of the EDFA. The
(channel B) sinusoidally modulated at frequency f.
[C. R. Giles, E. Desurvire, and J. R. Simpson, Opt. Lett. proliferation of inexpensive GaAs diode lasers in
14, 880882 (1989)]. the 800-nm range in the electronic products of the
196 Erbium-Doped Fiber Amplier: From Flashlamps and Crystal Fibers to 10-Tb/s Communication
late 1980s led to substantial research on 800-nm
pumping, in particular at British Telecom Research
Laboratories. However, they found that the gain
efciency was low. The reason was later identied
as the unfortunate presence of excited-state absorp-
tion around 800 nm in Er3+, a limitation that could
not be sufciently reduced by adjusting the pump
wavelength or the glass composition.
Much of this research soon focused on two
pump wavelengths only, namely, 980 nm from the
then-emerging strained GaAlAs laser technology
and around 1480 nm from InGaAsP diode lasers.
The prevalent thinking was initially that since an
EDFA pumped at 1480 nm is nearly a two-level
laser system, it should be difcult to invert and
exhibit a poor noise performance. The demonstra-
tion of the rst EDFA pumped at 1.49 m by Elias
Snitzer in 1988 quickly changed this perception.
The following year saw the rst report of an EDFA Fig. 2. Measured gain and gain coefcient in an
pumped at 1480 nm with an InGaAsP diode laser, at EDFA pumped at 980 nm and 1480 nm. Circles, 980 nm;
NTT Optical Communication Laboratory in Japan. triangles, 1480 nm. (Reproduced with permission of the
Institution of Engineering and Technology.)
This spectacular result (12.5 dB of gain for 16 mW
of absorbed pump power) put the EDFA on a new
track by establishing that a packaged FOA was within reach. For a short while pumping at 980 nm was
the underdog, in part because it had a higher quantum defect than 1480-nm pumping, hence an
expected lower efciency, and in part because of the lower maturity of the strained GaAlAs technology.
But 980-nm pumping nevertheless eventually won. Stimulated emission at 1480 nm turned out to be a
serious penalty, which gave a lower pump efciency and noise performance than with a 980-nm pump.
M. Shimizu and his team at NTT illustrated this compromise clearly in a cornerstone paper [5] that
compared the gain of an EDFA pumped at either wavelength (Fig. 2). The gain and the gain per unit
pump power (11 dB/mW!) were all substantially higher with 980-nm pumping, and the transparency
threshold was lower. This new understanding triggered a substantial R&D effort in the semiconductor
laser community, which ultimately lead to the commercialization of reliable, high-power, long-lifetime
diode lasers at 980 nm.
Many other important engineering issues were addressed through the mid-1990s. Two teams
contributed to this major effort more prominently than any other, namely, David Paynes group at the
University of Southampton [2,3] and Emmanuel Desurvive, rst at the AT&T Bell Laboratories, then at
Columbia University and Alcatel in France [6]. Many other substantial contributions came out of academic
and industrial laboratories around the world, especially in the U.S., UK, Denmark, Japan, and France.
A signicant fraction of the research was consumed by the quest for ever greater gain bandwidth,
lower noise, and a gain that is nearly independent of signal polarization, signal power, and number of
channels. To increase the bandwidth a number of ingenious solutions were implemented, ranging from
hybrid EDFs concatenating bers of different compositions and slightly offset gain spectra to adjusting the
level of inversion to produce preferential gain in the C band (15301565 nm) or L band (15651625 nm)
or designing the EDF so that it does not guide well above 1530 nm to produce efcient gain in the S band
(14601530 nm). This effort was greatly complicated by the parallel need for a uniform gain (or at gain
spectrum) so that all channels have a similar power and signal-to-noise ratio at the receiver. Here too,
clever solutions were conceived, from using passive lters to hybrid EDFAs, gain clamping, and the use of
telluride bers. This last approach produced a gain with a remarkable bandwidth of 80 nm (see Fig. 3) [7].
Later renements produced EDFAs with a gain atness well under 1 nm over wide bandwidths [8].
The EDFA rose from the status of research device to stardom remarkably rapidly, a resounding
manifestation of its practical importance, exceptional performance, and timeliness. The rst commer-
cial EDFA appeared in 1992. By 1998 over 40 companies were selling EDFAs; the count ultimately
Erbium-Doped Fiber Amplier: From Flashlamps and Crystal Fibers to 10-Tb/s Communication 197
peaked above 100. Research on communica-
tion systems followed suit, leading to the demon-
stration of increasingly large and high-performance
experimental and deployed systems. As one of
many examples illustrating the phenomenal per-
formance of communication links utilizing EDFAs,
in a particular experiment a total of 365 signals
were simultaneously recirculated 13 times around a
500-km ber loop containing ten EDFAs (one
every 50 km). At the output the power imbalance
between channels was as low as -7 dB and the bit-
error rate only 10-13. This system accomplished a
remarkable total optical reach of 6850 km and a
Fig. 3. Measured gain spectrum of a 0.9-m-long total capacity as high as 3.65 Tb/s. Deployed links
tellurite EDFA at various input-signal power levels. The now exceed 10 Tb/s over even longer distances.
gain in the 15351570-nm range was compressed by In the early 2000s, following the saturation of
using higher-power input signals. [Y. Ohishi, A. Mori,
M. Yamada, H. Ono, Y. Nishida, and K. Oikawa, the telecommunication industry and the sharp de-
Opt. Lett. 23, 274276 (1998)]. cline in the worlds markets, a signicant percentage
of the optical communication task force redirected
its vast technical expertise to other areas of photonics. This concerted effort gave the EDFA and other
FOAs a second carrier in spectacular new applications, especially ber sensors and high-power ber
lasers. Using an FOA to amplify the output of a ber laser, in a now widely used conguration called the
master-oscillator power amplier (MOPA), turned out to be the most energy-efcient way to produce
extremely clean and spectrally pure laser outputs up to enormous power levels. Today, ber MOPAs
are the worlds brightest light sources, to a large extent thanks to the superb properties of ber
ampliers. Yb3+, in particular, rapidly became the workhorse of high-power ber lasers for its low
quantum defect and high quenching-free concentration. Power scaling posed signicant challenges,
including efcient coupling into the gain ber of the high required pump powers, wavelength
conversion due to stimulated Brillouin and Raman scattering, optical damage, and photodarkening.
These challenges were met with a number of clever engineering solutions, including large-mode-area
bers (in which the signal intensity, and hence nonlinear effects and optical damage, are reduced) and
acoustic anti-guiding bers (in which the spatial overlap between acoustic and optical modes, and hence
the nonlinearity, are reduced). Commercial ber lasers utilizing MOPA congurations now offer
average powers up to the 100-kW range, a feat that would not have been possible without the superb
attributes of FOAs.
References
1. C. J. Koester and E. Snitzer, Amplication in a ber laser, Appl. Opt. 3, 11821186 (1964).
2. R. J. Mears, L. Reekie, I. M. Jauncey, and D. N. Payne, High gain rare-earth doped bre amplier
operating at 1.55 m, Proceedings of OFC, Reno, Nevada, 1987.
3. R. J. Mears, L. Reekie, I. M. Jauncey, and D. N. Payne, Low-noise erbium-doped bre amplier
operating at 1.54 m, Electron. Lett. 23, 10261028 (1987).
4. C. R. Giles, E. Desurvire, and J. R. Simpson, Transient gain and cross talk in erbium-doped ber
ampliers, Opt. Lett. 14, 880882 (1989).
5. M. Shimizu, M. Yamada, M. Horiguchi, T. Takeshita, and M. Okayasu, Erbium-doped bre ampliers
with an extremely high gain coefcient of 11.0 dB/mW, Electron. Lett. 26, 16411643 (1990).
6. M. Desurvire, Erbium-Doped Fiber AmpliersPrinciples and Applications (Wiley, 1994).
7. Y. Ohishi, A. Mori, M. Yamada, H. Ono, Y. Nishida, and K. Oikawa, Gain characteristics of tellurite-
based erbium-doped ber ampliers for 1.5-mm broadband amplication, Opt. Lett. 23, 274276
(1998).
8. M. J. F. Digonnet, Rare-Earth-Doped Fiber Lasers and Ampliers (Marcel Dekker, 2001).
198 Erbium-Doped Fiber Amplier: From Flashlamps and Crystal Fibers to 10-Tb/s Communication
19751990
Advent of Continuous-Wave
Room-Temperature Operation of
Diode Lasers
Michael Ettenberg
T
he story of getting to room temperature continuous-wave (CW) operation of semicon-
ductor diode lasers will start when the author arrived at RCA Labs with a fresh Ph.D. in
June of 1969. RCA had decided that GaAs would be the next important semiconducting
material in the solid-state electronics business after germanium, which at the time was the most
prevalent transistor material. While silicon transistors were already being manufactured, GaAs
transistors would be far superior, and RCA Research Lab researchers would concentrate their
efforts on GaAs and related compounds to leapfrog silicon. The choice had some validity. GaAs
was a direct-band semiconductor and thus had shorter electron hole lifetimes and a larger
bandgap, making possible transistors with higher speed, less temperature dependence, higher
operational temperature ranges, and smaller size. While all this is true, silicon became the pervasive
electronic device material for a variety of good reasons that will not be detailed here. But GaAs and
its related direct bandgap materials could do something that silicon could not do, that is, emit light
efciently. So RCA Labs moved its GaAs efforts to develop LEDs and diode lasers.
The authors rst assignment was to grow AlAs epitaxially on GaAs single-crystal
substrates via vapor-phase epitaxy, where Al is transported by passing HCl gas over Al and
As is supplied by breaking down arsine. After AlAs growth characterization, devices became of
interest. Since it seemed easier to make devices out of new materials than to create the materials
themselves, the author joined a group headed by Henry Kressel working on laser diodes. These
small devices were fascinating, as they were able to put out large amounts of reasonably
directed light, albeit they could be seen only with a night vision scope. There were four
relatively large research efforts at the time: Bell Labs the largest by far, Standard Telecom-
munications Laboratory (STL) in England, RCA Labs, and the Russian effort, about which less
was known, mainly due to the cold war. At IBM, GE, and Lincoln Labs, even though diode
lasers were rst demonstrated there, research efforts were not substantial. The research
projects at Bell Labs and STL were considerable, supported by telephone usage; telephone
companies were utilities at time. The telephone giants rst saw lasers as a potential source for
free-space communications, a secondary effort compared with microwave transmission in air
and pipes until Charles Kao envisioned optical communications in bers [1] and research at
Corning demonstrated low-loss optical bers in 1970 [2]. Then the laser efforts intensied.
RCAs research was driven by other applications such as optical disc recording and playback
and military usage. Since RCA had an Aerospace and Defense division, the diode laser efforts
could be justied, but RCA as a corporation was focused on television, and lasers were not a
mainline effort. The research was about half supported by the corporation and half by
government research contracts. The rst applications of laser diodes were military in nature,
and RCA decided to make the devices commercially so they could supply them to their defense
customers and potentially lower the price by supplying them for other commercial uses. In
1969 RCA became the rst commercial supplier of laser diodes, although it was a miniscule
business, especially for a multi-billion-dollar corporation.
199
The author was introduced to diode
lasers by Herb Nelson, who invented liq-
uid-phase epitaxy (LPE) the process used
to fabricate lasers throughout their initial
development, well past the rst CW
demonstrations and many years beyond,
through the rst several years of CD player
manufacture. Today almost all lasers are
made by metal organic chemical vapor
deposition (MOCVD): a much better con-
Fig. 1. First liquid-phase epitaxy (LPE) growth apparatus for trolled process and one that can be readily
creating laser diode (tipping furnace). (H. Nelson, RCA Rev. scaled up to multiple large wafers. What
24, 603 [1963]. Courtesy of Alexander Magoun.) Herb demonstrated was called a tipping
furnace (as shown in Fig. 1); it was a
tubular furnace about six inches in diameter mounted in a metal cage. The cage was in turn mounted
in a seesaw arrangement at the center of the furnace so the furnace could be rocked back and forth or
tipped. Inside the furnace was a sealed quartz tube with hydrogen owing through it and a carbon boat;
at one end of the boat was a small polished single-crystal GaAs wafer of about a square centimeter, and
at the other end of the boat was a polycrystalline GaAs wafer with a glob of gallium on it. The process
started with the furnace being heated to about 800C with the polycrystalline GaAs and Ga side lower,
and some time was allowed so the GaAs could go into solution in the Ga until saturation; then the
furnace was tipped the other way and the saturated Ga rolled onto the single-crystal wafer. Next
the furnace was cooled and the GaAs in solution precipitated onto wafer to form an epitaxial layer on
the single crystal. This epitaxial layer was much superior in terms of contaminants and defects to the
underlying substrate and was also superior in terms of its luminescent properties and ability to make
lower-threshold, more efcient lasers. Al was added to the Ga glob so that AlGaAs alloys could
be grown. Al and Ga atoms are about the same size; therefore dislocations caused by lattice parameter
mismatch would not be formed when AlGaAs was grown on GaAs. In addition, adding Al to GaAs
raised the bandgap and lowered the index of refraction of the alloy compared to GaAs, which proved to
be crucially important to the creation of low-threshold efcient lasers. The temperature was controlled
by hand, using a variable transformer to control the current to the furnace and a 01000C dial
thermocouple readout. It was amazing how such a crude growth system could produce such
sophisticated devices, but Herb understood the materials and was an artist. Later the process was
brought under better control using carbon boats with a wafer that slid under multiple bins to allow
growth of multiple layers with controlled composition and remarkable submicrometer-thickness
accuracy.
The rst diode lasers were simply millimeter-sized cubes of GaAs containing a diffused pn junction
with polished faces for mirrors that operated at liquid nitrogen temperatures with multi-amp very-low-
duty-cycle short pulses applied. It was remarkable that these devices lased, considering that all prior laser
types required tens of centimeters of cavity length and mirror reectivity greater than 95%. The gain in
GaAs per unit length was exceptional, thus allowing the gain to exceed the loss even though the reectivity
at the natural mirror surface of GaAs is only about 30%. The applied current to these rst devices was
many tens of thousands of amperes per square centimeter at liquid nitrogen temperatures; the threshold
current increased exponentially as the temperature increased, so room temperature CW lasing was a long
way away. It was found early on that GaAs cleaved nicely on the 100-crystal plane, so the lasers were
grown on single-crystal wafers cut on the 100 plane. Then the mirror facets could be easily formed after
the wafers had been thinned, metalized on the n and p sides by cleaving into bars. Next the bars were
sawed into individual dies about 400 m in length between the mirrors and 100 m wide. The sawn
roughened sides prevented lasing from occurring crosswise to the mirrors.
There were three important steps to room-temperature CW lasing: the addition of heterojunctions
to the laser structure, the double heterojunction, and nally, the stripe contact. In 1969 independently
and simultaneously, Kressel and Nelson [3] and Panish et al. [4] published papers demonstrating that
adding heterojunctions of AlGaAs on GaAs and diffusing the pn junction a micrometer or so from the
heterojunction formed a light waveguide. This waveguide conned the light, creating electron/hole
recombination to that waveguide as illustrated in Fig. 2 [5]; consequently, the threshold current could
be reduced to about 10,000 amps/cm2, still a factor of 10 or so away from what would be needed for
CW operation. The reduction in threshold from the simple p-n homojunctions came from the fact that
the light and the recombination of electron and holes was conned to a smaller volume, thus requiring
less current to invert the population to the point of lasing. These single-heterojunction devices were the
rst laser diodes to go into production, becoming optical proximity sensors for the sidewinder missile.
Art DAsaro and colleagues [6, 7] at Bell Labs developed the stripe contact, shown in Fig. 3, which
is a necessary and enduring feature for laser diodes, because it not only stops the cross lasing in a simple
manner but facilitates the heat sinking of the device with unpumped regions all along the laser cavity.
The nal and most important step came from Alferov et al. [8]. Alferov was one of the leaders in the
eld and came to United States to visit RCA and Bell Labs, among others. The visit was memorable,
because it was very strange. We sat in a small ofce and discussed the progress of lasers. Alferov had a
large heavyset man with him who said very little and seemed to know little about lasers; it was surmised
that he was KGB. Alferov did not disclose the double-heterojunction laser structure nor was he shown
much because the work was partially supported by Department of Defense. It was learned later that he
did discuss the double-heterojunction work at Bell Labs, probably because they were more open. There
was a race to achieve CW operation. Bell and RCA Labs were neck and neck, but Bell had the stripe-
contact technology and learned about the double-hetrojunction structure. As a result, Hayashi and
Fig. 3. Schematic of a
typical CW heterojunction laser,
drawn upside down to show the
stripe contact. Diffraction
causes the vertical spreading of
the beam. Reproduced with
permission from Fig. 6 of H.
Kressel, I. Ladany, M.
Ettenberg, and H. Lockwood,
Physics Today 29(5), 38 (1976).
(Copyright 1976, American
Institute of Physics.)
References
1. C. Kao, IEEE Meeting in London, 1966.
2. F. P. Kapron, D. B. Keck, and R. D. Maurer, Radiation losses in glass optical waveguides, Appl. Phys.
Lett. 17, 423425 (1970).
3. H. Kressel and H. Nelson, Close-connement gallium arsenide P-N junction lasers with reduced optical
loss at room temperature, RCA Rev. 30, 106113 (1969).
4. M. B. Panish, I. Hayashi, and S. Sumski, A technique for the preparation of low-threshold room-
temperature GaAs laser diode structures, IEEE J. Quantum Electron. 5, 210 (1969).
5. Figure 5 from H. Kressel, H. F. Lockwood, I. Ladany, and M. Ettenberg, Heterojunction laser diodes
for room temperature operation, Opt. Eng. 13(5), 417422 (1974).
6. Figure 6 from H. Kreseel, I. Ladany, M. Ettenberg, and H. Lockwood, Light sources, Physics Today
29(5), 3842 (1976).
7. J. E. Ripper, J. C. Dyment, L. A. DAsaro, and T. L. Paoli, Stripegeometry double heterostructure
junction lasers: mode structure and cw operation above room temperature, Appl. Phys. Lett. 18, 155
157 (1971).
8. Zh. I. Alferov, V. M. Andreev, E. L. Portnoi, and M. K. Trukan, AlAs-GaAs heterojunction injection
lasers with a low room-temperature threshold, Sov. Phys. Semicond. 3, 11071110 (1970).
9. I. Hayashi, M. B. Panish, P. W. Foy, and S. Sumski, Junction lasers which operate continuously at room
temperature, Appl. Phys. Lett. 17, 109110 (1970).
10. M. Ettenberg, A new dielectric facet reector for semiconductor lasers, Appl. Phys. Lett. 32, 724725
(1978).
11. R. L. Hartman, N. E. Schumaker, and R. W. Dixon, Continuously operated (Al, Ga)As double
heterostructure lasers with 70 C lifetimes as long as two years, J. Appl. Phys. 31, 756 (1977).
12. M. Ettenberg, Electron. Lett. 14, 615 (1978).
13. H. F. Lockwood, H. Kressel, H. S. Sommers, Jr., and F. Z. Hawrylo, An efcient large optical cavity
injection laser, Appl. Phys. Lett. 17, 499501 (1970).
14. P. L. Derry, A. Yariv, K. Y. Lau, N. Bar-Chaim, K. Lee, and J. Rosenberg, Ultralowthreshold graded
index separateconnement single quantum well buried heterostructure (Al, Ga)As lasers with high
reectivity coatings, Appl. Phys. Lett. 50, 17731775 (1987).
I
n the late 1960s, Bell Labs had a problem. The nations demand for long-distance
telecommunications services was steadily increasing, but the technologies then in use
coaxial cable and point-to-point microwave transmission through the aircould not keep
up with the pace. The major reductions in optical ber waveguide losses reported in the early
1970s were therefore of great interest. The lowest-loss regions of these bers were in the 0.8 to
0.9 m range, which could in principle be accessed by devices built using the GaAs-GaAlAs
material system. Thought was given to the possible use of GaAs light-emitting diodes (LEDs), but
it was immediately obvious that semiconductor lasers would be much better sourcesif they
could be developed reliably in commercial quantities. One could easily imagine an efcient GaAs
laser that could couple a milliwatt of optical power into a ber with a core diameter of about
50 m. Thus was dened the rst generation of ber-optic telecommunication systems.
In the late 1960s and early 1970s, the author was a young supervisor working on the
development of LEDs for Bell System applications. In that process, he learned quite a bit about
the physics, technology, and transfer-to-volume manufacture of III-V semiconductors. One result
of this program was the successful implementation of green-emitting GaP LEDs for nighttime dial
illumination in the handset of the Dreyfuss-designed Trimline phone. Something like 100 million
of these sets were subsequently produced.
In 1973, the author transferred to a small exploratory development group working on
semiconductor lasers. The group had beneted from an excellent research effort that happened
just down the hall. Most notable was the demonstration in 1970 of a continuously operating
room-temperature GaAs-AlGaAs heterostructure semiconductor laser [1] (see Fig. 1). However,
these broad-area lasers had high operating currents (around 400 mA) and very short lives (they
were sometimes referred to as ashbulbs), but they showed the way forward!
The groups choice of a laser structure for initial development consisted of four planar
epitaxial layers grown sequentially by liquid-phase epitaxy (LPE) on a GaAs substrate. We
inhibited lateral carrier ow by using proton bombardment to dene a stripe-geometry
wherein only a narrow stripe, 10 250 m, was electrically pumped (see Fig. 2). These
stripe-geometry lasers became the workhorses of the early Bell Labs semiconductor laser
development. They allowed the sorting out of many reliability and device performance issues. In
a typical week, half a dozen or so wafers were processed into some thousands of lasers. Fast
turnaround made it possible to quickly and systematically iterate device, processing, and material
innovations.
Many of the early stripe-geometry lasers had very erratic properties. Some would lase for a
time but would then suddenly become inoperable. Others would die slowly. Still others would
not work from the outset. Typical continuous-wave operating lifetimes at room temperature
were on the order of minutes to days. Many devices also had other undesirable characteristics,
for example, nonlinear light output versus current. It was clear that the group had a very difcult
development project on its hands! Some thoughtful observers, including one key Bell Laborato-
ries vice president, opined that success was unattainble.
Important clues to improvements came in early 1973 from an experiment in which
windows were fabricated on the substrate side of stripe-geometry lasers in such a way that
spontaneous emission (and scattered stimulated emission if present) from the stripe region of the
laser could be observed with an infrared optical microscope. Dark-line defects (DLDs), which
203
grew in a lasers active region during operation,
were observed and were determined to be the prin-
cipal failure mechanism in devices that stopped
working in the rst 100 hours or so [2].
This paper correctly stated that the combina-
tion of low-strain processes and extreme cleanliness
in materials growth should provide a dramatic
increase in laser life. It galvanized a large technical
community such that it seemed that everyone in the
world with an electron microscope then decided to
investigate this area. A picture was, in this case,
worth many thousand words!
The Bell group subsequently worked hard to
understand and eliminate localized modes of degra-
dation, including those associated with DLDs in the
long narrow-lasing region of the laser and those
associated with mirror surfaces. Subsequent experi-
ments showed that DLDs identical to those seen in
lasers could be generated by optical pumping of
undoped and unprocessed laser material, thus con-
rming that DLD initiation and growth could result
from properties of laser material that were not
associated with proton bombardment, p-n junction
dopants, or contact metallization technology.
Many improvements in LPE growth technology
and its automation were also made during this
Fig. 1. Izuo Hayashi, holding a heat absorbing period. Fundamental difculties with this batch
device, points to the location of a broad-area process made it stubbornly difcult to reproducibly
semiconductor laser designed by Bell Laboratories control, but it was greatly improved in the skilled
scientists. (Bell Laboratories/Alcatel-Lucent USA Inc.,
courtesy AIP Emilio Segre Visual Archives, Hecht hands of the Bell groups crystal growers.
Collection.) By late 1974, with continuing work on many
technology fronts, the reliability situation had im-
proved considerably, and selected lasers had been operating continuously for more than a year at room
temperature (typically 30C). On the basis of the data obtained, the group was able to conclude that
continuous room-temperature operation of these devices as lasers with power outputs exceeding
1 mW per laser face for times in excess of 100,000 h is possible. This was an important feasibility
demonstration. However, it served to reinforce the urgency of nding ways to condently accelerate
diode aging so that lasers tested for short periods could be installed in the eld with the expectation that
they would last for decades.
others to insist that those who by much hard work made inventions practical be honored. In
particular, I wanted recognition for the team at Bell Telephone Laboratory. They had increased
the mean time to failure at room temperature of the double-heterostructure GaAs-based laser from
several minutes in 1970 to an extrapolated 8 million hours in 1978. Redikers efforts along with
others led to B. C. DeLoach, R. W. Dixon, and R. L. Hartman receiving the IEEE Gold Medal for
Engineering Excellence in 1993 for this work (see Fig. 5).
The groups efforts in the 1970s, 1980s, and 1990s were aimed at Bell System applications in long-
distance, high-volume voice, data and video transmissionboth on land and undersea. Today,
essentially all terrestrial and undersea telecommunications, data, and television trafc above the local
distribution level is carried in ber using lasers as sources. The Internet would not be possible without
these laser devices. Undersea cables with long repeaterless spans (approaching 10,000 km) now often
have the high-performance lasers that encode digital information only at the land ends. Much simpler
continuously operating lasers, which carry no signal information, are used to pump ber ampliers that
are periodically spaced under the sea. Data rates in a single ber, using very-high-speed modulation and
wavelength division multiplexing, in high-volume applications, can approach 1 Tb/s20,000 times
higher than the groups initial 45-Mb/s rates!
The program also supported what was then called ber-to-the-home, or colloquially the last
mile. This application took longer to become a reality because of the breakup of the Bell System and
the high costs of serving individual customers. It was pleasing, and a little nostalgic, when about ve
years ago Verizon brought their laser-based FiOS product to the authors home. On the consumer
products side, it has been extremely satisfying to witness the unexpectedly fast and widespread
application of lasers in products such as printers and CD/DVD players and/or the dramatic price
reductions made possible by these high-volume applications. Through the efforts of thousands of
scientists and engineers throughout the world, both the programs the author worked on and their
subsequent applications have succeeded beyond his wildest dreams.
The author is grateful to each one of the scores of professional scientists, technologists, and many
others who contributed to the success of the Bell Laboratories semiconductor laser development
program during the last decades of the twentieth century. It was fun being along for the ride.
Terabit-per-Second Fiber
Optical Communication Becomes
Practical
Guifang Li
H
umans used optical signals intuitively for the purpose of communication in ancient
times. Modern day optical communication systems are instead based on the funda-
mental understanding of information theory and technological advances in optical
devices and components. The Optical Society (OSA) played a vital role in making ber-optic
communication practical for the information age.
It is well known that the capacity of a communication channel is constrained by the Shannon
limit, W log2(1 + S/N), where W is the spectral bandwidth and S/N is the signal-to-noise ratio
(SNR). The bandwidth of a communication channel is proportional to the carrier frequency, which
is on the order of 200 THz for visible or near-infrared light. Therefore, a small fractional
bandwidth around the optical carrier can provide a capacity much larger than the limited capacity
supported by the spectrum of radio-frequency (RF) waves or microwaves [1]. The SNR of a
communication channel is proportional to the received power and inversely proportional to the
noise and distortion. The invention of the laser, which can produce high-power coherent optical
radiation at the transmitter, fueled the migration from RF/microwave communication to optical
communication. In fact, the rst patent on lasers (more precisely masers) by Nobel Laureates
Charles Townes and Arthur Schawlow, both OSA Honorary Members, was entitled Maser and
maser communication systems.
To make optical communication practical, however, the received optical power (not only the
transmitted power) must be much stronger than the noise. This requires a low loss optical
transmission channel. The loss in free-space transmission is determined by diffraction, which is
much larger than that of RF/microwave in appropriate cables. Fortunately, light can also be
guided by total internal reection, a phenomenon known since the mid-nineteenth century. An
optical ber with a high-index core surrounded by a lower-index cladding can support guided
modes inside the dielectric cylindrical waveguide that propagate without experiencing radia-
tive loss [2]. As a consequence, the loss of the optical ber is dominated by material loss. Glass
bers were initially deemed impractical for communication systems, as the measured attenuation
was >1000 dB/km.
In 1966, Kao and Hockham showed that the measured losses were due to impurities rather
than fundamental loss mechanisms and, without impurities, glass bers could achieve losses
below 5 dB/km. They also identied that fused silica ber could have the lowest losses. OSA
Fellow Dr. Charles Kao was awarded the 2009 Nobel Prize in Physics for his groundbreaking
achievements concerning the transmission of light in bers for optical communication, which
has fundamentally transformed the way we live our daily lives. It is the invention of the silica
optical ber and the semiconductor laser with signicantly long life that ushered in the era of
modern optical communication. (These inventions are described in separate essays in this section
of this book.)
The rst-generation ber-optic communication system in the 1980s used multimode
bers and 0.8-m multimode FabryPerot semiconductor diode lasers, supporting a data rate
209
of 45 Mbit/s [3], which was orders of
magnitude larger than that of the micro-
wave cable systems then in use. Since then,
the capacity of optical ber communica-
tion systems has grown in leaps and
bounds. Throughout its history, ber-op-
tic communication has invented and rein-
vented itself many times over, as shown in
Fig. 1, making terabits per second (Tb/s)
practical. For example, the second-gener-
ation ber-optic communication system
operated at 1310 nm using single-mode
bers and single-mode semiconductor di-
ode lasers. This brought about two
Fig. 1. History of ber-optic communication systems. improvements over the rst-generation
(Courtesy of Tingye Li, Alan Willner, and Herwing Kogelnik.) systems. First, the 0.3-dB/km loss of opti-
cal ber at 1310 nm is much lower than
3 dB/km at 870 nm, which helped to over-
come noise. Second, 1310 nm is the zero-dispersion wavelength for standard single-mode ber. All of
these different stages of technology development overcame different physical limitations of the optical
communication system, pushing capacity toward the Tb/s fundamental limit. The physical limitations
for ber-optic communication arise from noise and distortion.
First let us focus on the sources of noises, which are closely related to modulation formats. Before
1980, the modulation format for optical communication systems was intensity-modulation direct
detection, which is thermal noise limited with a sensitivity of thousands of photons/bit. In an effort to
overcome thermal noise, the third-generation optical communication systems moved to 1550 nm,
which is the minimum-loss wavelength for single-mode bers, to increase the received optical power. As
the additional power budget allowed gigabits-per-second transmission, distortions due to ber
dispersion could sometimes be the limiting factor. So in some third-generation systems, dispersion-
shifted bers for which the zero-dispersion wavelength was shifted to 1550 nm through proper design
of the ber index prole were used. In such systems, the capacity was still limited by thermal noise.
Thus, starting from the mid-1980s, the optical communications community embarked on the develop-
ment of coherent detection. Phase-shift keying (PSK) using coherent homodyne detection is limited by
the shot noise of the local oscillator, and for binary PSK the sensitivity is 9 photons/bit, two orders of
magnitude better than the thermal noise limit. However, coherent optical communication did not
advance into commercial deployment because (1) phase locking and polarization management of the
local oscillator was too complex and unreliable, and (2) the advent of the erbium-doped ber amplier
(EDFA) made it unnecessary.
As early as 1964, rare-earth metal-doped glass ber was proposed and demonstrated as a gain
medium for optical amplication [4]. However, it was not until the late 1980s when two groups
published work demonstrating high-gain EDFAs for ber-optic communicationrst by the group led
by David Payne [5] and then by Emmanuel Desurvire [6]that EDFA revolutionized the eld of optical
communication. Payne and Desurvire received the John Tyndall Award from OSA in 1991 and 2007,
respectively. In terms of noise performance, optical pre-amplication (using an EDFA in front of the
photodetector) changes the dominant noise source to the amplied spontaneous emission of the EDFA
rather than the thermal noise of the photodetector. The fourth-generation optical communication
system employed pre-amplied direct detection, which has a sensitivity of 39 photons/bit. (An essay on
ber optical ampliers is in this section of this book.)
In fact, the gain bandwidth of an EDFA is 3 THz, much wider than the single-channel
bandwidth, which is limited by the speed of electronics. As a result, EDFAs enabled the fth
generation of wavelength-division-multiplexed (WDM) optical transmission systems. In these sys-
tems independent data streams are simultaneously transmitted on multiple wavelength channels in a
single ber and amplied together in a single EDFA, similar to frequency-division multiplexing in
T
he recent ftieth anniversary celebrations marking the invention of the laser and
the birth of modern nonlinear optics were major historical milestones. Theodore
Maimans observation of laser action in ruby in May 1960 [1] provided the essential
tool that enabled Peter Frankens team at the University of Michigan to perform their
legendary 1961 experiment in which they saw optical second harmonic generation for the
rst time [2]. From this small beginning, nonlinear optics has grown into the vast and vibrant
eld that it is today.
The Optical Society Centennial provides an opportunity to reect on developments in
nonlinear optics in the intervening years and, specically, to focus on some of the highlights in the
development of the eld between 1975 and 1990. The theoretical foundations of optical
frequency mixing were laid by Nicolaas Bloembergens Harvard team in a seminal 1962 paper
[3], which was prescient for introducing innovative ideas that strongly inuenced later devel-
opments in the eld; some specic examples will be mentioned later. In 1979, Nicolaas
Bloembergen (see Fig. 1) was awarded The Optical Societys Ives Medal, the societys highest
award. He won a quarter share of the 1981 Nobel Prize for his contribution to the development
of laser spectroscopy, in addition to his pioneering work on nonlinear optics.
By the early 1970s, many of the conceptual foundations of nonlinear optics had been laid,
and a remarkable number of crude experimental demonstrations of techniques that are now
routine had been performed. Progress over the ensuing decades was often prompted by advances
in laser technology and, crucially, in materials fabrication. Suddenly it would become possible to
implement an experiment so much more effectively than previously that it would soon become an
established laboratory technique, or might even form the basis of a new commercial product.
A major achievement of the period was the fabrication of layered crystalline structures in
which phase-matching is determined by the periodicity of the layers. Remarkably, this quasi-
phase-matching (QPM) technique was originally suggested in the 1962 Harvard paper
mentioned earlier [3], and it is a prime example of a principle that took more than two decades
of gestation between original inspiration and nal fruition.
Quasi-phase-matching materials have periodically reversed domains, each one coherence
length thick. The nished product is like a loaf of sliced bread in which alternate slices (of
anisotropic crystal) are inverted (see Fig. 2). The problem is that each slice has to be only a few
micrometers thick, so it would be a little thin for ones breakfast toast! It took more than two
decades to develop the sophisticated crystal growth techniques needed to fabricate media with
such thin layers. Today, QPM is routine; indeed many researchers have abandoned traditional
birefringent phase-matching altogether. The most well-known QPM medium is perhaps period-
ically poled lithium niobate (abbreviated PPLN and pronounced piplin), and practical devices
of high conversion efciency are commercially available. In 1998, Robert Byer (see Fig. 1) and
Martin Fejer were awarded The Optical Societys R. W. Wood Prize for seminal contributions
to quasi-phase matching and its application to nonlinear optics. More recently, in 2009, Robert
Byer received the Ives Medal, The Optical Societys most prestigious award.
The need for tunable coherent light sources to replace tunable dye lasers drove the
development of solid-state devices; these are not subject to messy chemical spills, and the tuning
ranges achievable in a single medium can extend from the ultraviolet to the mid-wave infrared
(35-m) regimes.
213
Fig. 1. Images of ve scientists who have made major breakthroughs in the development of nonlinear optics.
From top left to bottom right they are: Nicolaas Bloembergen, Robert L. Byer, Chuangtian Chen, Linn F. Mollenauer,
and Stephen E. Harris. (Bloembergen, Mollenauer, and Harris photographs courtesy of AIP Emilio Segre Visual
Archives, Physics Today Collection; Byer photograph courtesy of AIP Emilio Segre Visual Archives, Gallery of
Member Society Presidents; Chen photograph courtesy of Professor Chen Chuangtian.)
An important nonlinear optical process for creating a wideband coherent light source is optical
parametric generation. This is essentially sum-frequency generation (the generalized version of second
harmonic generation) running in reverse. A high-frequency pump wave drives two waves of lower
frequency, known as the signal and the idler; in photon language, the pump photon divides its
energy between the signal and idler photons. Without a seed to dene a particular frequency band, the
signal and idler grow from noise, with frequencies determined by the phase-matching conditions. An
optical parametric amplier is a device of this kind with a signal or idler seed to x the operating
frequency. If the gain is high, the conversion efciency can be quite large, even for a single-pass system.
However, the efciency can be greatly improved by placing the nonlinear medium within a well-
designed cavity, creating an optical parametric oscillator (or OPO).
The rst OPO was demonstrated by Giordmaine and Miller as early as 1965, but subsequent
progress was slow, largely because nonlinear crystals of the necessary high quality were not available.
Indeed, the OPO is another example of a device where technological capability lagged seriously behind
References
1. T. Maiman, Stimulated optical radiation in ruby, Nature 187, 493494 (1960).
2. P. A. Franken, A. E. Hill, C. W. Peters, and G. Weinreich, Generation of optical harmonics, Phys. Rev.
Lett. 7, 118119 (1961).
3. J. A. Armstrong, N. Bloembergen, J. Ducuing, and P. S. Pershan, Interactions between light waves in a
nonlinear dielectric, Phys. Rev. 127, 19181939 (1962).
4. G. A. Magel, M. M. Fejer, and R. L. Byer, Quasiphasematched secondharmonic generation of blue
light in periodically poled LiNbO3, Appl. Phys. Lett. 56, 108110 (1990).
5. C. M. Bowden and J. W. Haus, eds., Nonlinear optical properties of materials, J. Opt. Soc. Am. B 6
(April 1989).
6. A. M. Johnson, R. H. Stolen, and W. M. Simpson, 80 singlestage compression of frequency doubled
Nd:yttrium aluminum garnet laser pulses, Appl. Phys. Lett. 44, 729731 (1984).
7. J. Hewett, Ultrashort pulses create ultrabroad source, historical archive, Optics.org.
8. J. W. G. Tisch, Imperial College Attosecond Laboratory; reproduced from G. H. C. New, Introduction
to nonlinear optics, Cambridge University Press 2011, with permission.
S
pectroscopy has been a fundamental part of optics ever since Newton rst showed that
white light could be dispersed into its constituent colors and later when Young showed that
light was wavelike and provided a grating with which to measure its wavelength. The role
of The Optical Society (OSA) in spectroscopy during the pre-laser era is described in an essay
entitled Spectroscopy from 1916 to 1940 in an earlier part of this book. The rst experimental
demonstration of a laser, a ruby laser, was made by Theodore Maiman in 1960, and soon after,
in 1964, a Nobel Prize was awarded for prior theory on the topic to Charles Townes, Nikolay
Basov, and Alexander Prokhorov. Additionally, parametric nonlinear optics was discovered by
Peter Franken in 1961. The combination of lasers and nonlinear optics made possible incredible
advances in spectroscopy leading to linear and nonlinear laser spectroscopy. Developments in
this eld were so numerous that this short account can only hope to capture the principal events
of an important chapter in optics and OSA history.
Almost immediately upon the invention of the laser, scientists recognized that the two most
obvious features of laser light, its high intensity and its spectral purity, were far beyond anything
that had been available before. In less than a year following Maimans ruby laser, Franken took
advantage of its high intensity to demonstrate optical second harmonic generation and open up
the eld of nonlinear optics. This would lead to numerous nonlinear spectroscopies mentioned
below. Different designs also permitted wide-ranging variations in the type of output obtainable
from lasers. Very pure single-frequency light was created with continuous-wave lasers and very
broad, supercontinuum sources were created with ultrashort pulse lasers. The availability of
lasers with large or small bandwidths and short or long pulse durations enabled the development
of dozens of new and powerful approaches to precision optical measurements.
Laser Spectroscopy: An
Enabling Science
The transition from spectroscopic research in the
period 19602000 to its many applications had a
long gestation period. D. Auston disclosed a method
of generating single cycles of terahertz radiation in
the 1980s. However, applications such as imaging
through plastics and ceramics with terahertz waves
would not become routine until the beginning of the
twenty-rst century. Similarly, as early as 1980,
T. Heinz and Y. R. Shen found that second har-
monic generation was allowed on the surfaces of
centro-symmetric media but forbidden in their inte-
rior. IBM exploited this interaction to inspect silicon
wafers for electronic circuits, but decades passed
before species-specic structural and dynamic stud-
ies became popular with chemists. By the 1990s,
experiments in the research groups of S. Harris and
B. P. Stoicheff had established that opaque materials
could be rendered transparent through quantum
Fig. 4. Claude Cohen-Tannoudji. (Photograph by
interference. This had immediate impact on spec- Studio Claude Despoisse, Paris, courtesy AIP Emilio
troscopy and the generation of short wavelength Segre Visual Archives, Physics Today Collection.)
The Future
Over its 100-year lifespan, The Optical Society has
Fig. 5. John Hall. (Courtesy of AIP Emilio Segre
Visual Archives, Physics Today Collection.) been led by many accomplished scientists, many of
whom were spectroscopists. It is partly for this
reason that the society has been able to maintain a prominent role throughout an explosive period
of scientic history that relied on precise spectral tests of new theories. Spectroscopists contributed to
but also beneted from and were nurtured by the emphasis on fundamental science and the open,
relaxed style of the Society, where many disciplines intersect. The vibrancy of OSA has rested on
personal relationships fostered by the Society across ideological boundaries. OSA has followed a
tradition of internationalization that began long before globalization made it necessary. Past president
Art Schawlow understood how important international connections were for spectroscopy and science
in general. He knew that when it came time for visitors from China, New Zealand, Canada, and Ireland
to return home, they would inevitably take home part of his magic recipe for having fun with great
science. They had learned that You dont need to know everything to do good research. You just have
to know one thing that isnt known, and of course you also had to be a spectroscopist! By sharing this
attitude, Art was a great ambassador for the eld of spectroscopy and for OSA itself. The rich history of
both, and his encouraging message, accumulated in the hearts of his students and visitors. Current and
future OSA members will sustain the unique strengths of the Society that account for its remarkable
spectroscopic legacy and its future contributions.
Acknowledgement
Photos were provided by S. Svanberg, J. Hecht, H. van Driel, and the OSA archives. The authors wish
to thank J. Eberley for a critical review.
T
he invention of the laser has made possible the use of radiation pressure to optically trap
and manipulate small particles. The particles can range in size from tens of micrometers
to individual atoms and molecules. Laser radiation pressure has also been used to cool
atoms to exceptionally low temperatures, enabling a new branch of atomic physics. See [1] for an
extensive summary of the many varieties of work done with laser radiation pressure.
Inspired by a long interest in radiation pressure, in 1969 the author focused a TEM00 mode
laser beam of about 30-m diameter on a 20-m transparent dielectric latex particle suspended in
water. Strong motion in the direction of the incident light was observed. If the particle was off
axis, at the edge of the beam, a strong gradient force component to the light force pulling the
particle into the high-intensity region on the axis was observed. The particle motion was closely
described by these two force components: one called the scattering force in the direction of the
incident light and the other the gradient force in the direction of the intensity gradient. With
these two components, and using two oppositely directed beams of equal intensity, it was
possible to devise a stable three-dimensional all-optical trap for conning small particles.
Particles moving about by Brownian motion that entered the fringes of the beam were drawn
into the beams, moved to the equilibrium point, and were stably trapped. If the axial gradient
force is made to exceed the scattering force, and this can be done, then a single-beam trap is
possible, as shown in Fig. 1.
Because this was the rst example of stable optical trapping, this discovery was submitted to
Physical Review Letters. Since single atoms are just small neutral particles and should behave
much as single dielectric spheres, it was postulated that trapping of single atoms and molecules
should also be possible. At Bell Labs, if one wanted to submit a paper to Physical Review Letters
one had to pass an internal review by the prestigious theoretical physics department to preserve
the Labs good name. So the author submitted a manuscript and it was rejected. Upon the
recommendation of his boss, Rudi Kompfner, the inventor of the traveling-wave tube, the paper
was resubmitted and was accepted with no problem [2]. A second theoretical paper was
submitted to Physical Review Letters in 1970 on acceleration, deceleration, and deection of
atomic beams by resonance radiation pressure [3]. This was followed by a number of experi-
ments on optical traps for micrometer-size solid spheres or liquid drops demonstrating optical
levitation against gravity in air and as a function of pressure down to high vacuum and for
various beam convergence angles. By using optical levitation in conjunction with feedback
stabilization of the levitated particles position, it was possible to study the wavelength depen-
dence of the optical levitation forces with dye lasers. A series of complex size-dependent
resonances were observed that were found to be in close agreement with MieDebye electro-
magnetic theory calculations. These results are probably the most exact conrmation of
Maxwells theory for light scattering by transparent dielectric spheres. The frequencies of these
resonances allow one to determine the particle size and index of refraction to six or seven
signicant gures. Using the position stabilization technique it was possible to perform a modern
223
version of the Millikan oil drop experi-
ment for accurately determining the elec-
tric charge of a single electron.
Optical trapping of atomic vapors in
high vacuum is more difcult than trap-
ping macroscopic particles. One needs
some form of damping for lling and
holding atoms in an optical trap. Work
was started in the early 1970s on acceler-
ating, decelerating, and deecting atoms
Fig. 1. A single-beam optical trap for a high-index, with applications such as velocity sorting
transparent sphere. The laser beam is tightly focused such and isotope separation. T. Hnsch and
that the axial component of the gradient force exceeds the
A. Schawlow wrote an important early
scattering force. E0 is the equilibrium point at which the sphere
is trapped. paper on optical cooling of atoms using
the Doppler shift in a six-beam geometry
for use in precision spectroscopy. They
did not consider the possibility of optical trapping. In Russia, V. S. Letokhov and V. G. Minogen did
experiments trying to stop sodium beams with chirped counterpropagating light beams, but failed.
They were intending to trap atoms in a trap tuned a half-linewidth below resonance where cooling is a
maximum. W. D. Phillips and H. W. Metcalf, inspired by Ashkins rst paper about atoms, also started
work on atom slowing. They soon realized that the slowing difculties experienced by Letokhov and
Minogen were due to optical pumping, and in 1982 they successfully used a beam-slowing method
based on a tapered magnetic eld to completely stop the beam at a nal temperature of about 0.1 K.
In 1978 Bjorkholm, Freeman, and the author carried out an experiment using tuning far from
resonance that demonstrated dramatic focusing and defocusing of an atomic beam caused by the optical
gradient forces [4] (Fig. 2). These striking results suggested that atom trapping would be possible if
proper cooling could be achieved. It was realized that optical heating of atoms was a problem in
achieving stable traps for cold atoms even for optimal tuning at a half-linewidth below resonance,
where the cooling rate is a maximum, due to saturation. However, it was shown that deep trapping
potentials were possible for two-beam traps and one-beam traps by tuning far-off resonance where
saturation is greatly reduced. Two papers by Ashkin and Gordon addressed the details of laser cooling
and heating and showed various ways of achieving adequate Doppler cooling.
In 1983 Steve Chu was transferred to our Holmdel Lab from the Murray Hill Lab. He was an
experienced atomic physicist, but he did not know much about trapping at the time. He became
interested and decided to join John Bjorkholm and the author in an attempt to trap atoms using lasers.
This was at a time when we had some new bosses who decided that atom trapping would not work, and
they tried unsuccessfully to discourage Bjorkholm and Chu from working with the author on this
project. In spite of this pressure, Chu was given a quick lesson in atom trapping, and an effort was made
to demonstrate the rst optical trap for atoms. The rst experiment was aimed at creating a collection of
Fig. 2. Experimental
demonstration of the focusing
and defocusing of an atomic
beam caused by the optical
gradient force. (a) Laser tuned
below resonance; atoms
attracted to high intensity
regions. (b) Laser tuned above
resonance; atoms repelled from
high intensity regions. (Redrawn
from A. Ashkin, [IEEE J. Sel.
Topics Quantum Electron. 6(6),
Nov./Dec., 2000.])
224 Optical Trapping and Manipulation of Small Particles by Laser Light Pressure
very cold atoms capable of being conned in the shallow atom traps. The experiment was based on the
theoretical ideas proposed by Hnsch and Schawlow, mentioned earlier, and it worked beautifully. It
provided a cloud of atoms having a temperature of about 240 K, as expected, which is ideal for
trapping. That cooling technique has become to be known as optical molasses. Now that it was
possible to generate cold atoms, Bjorkholm suggested trying a single-beam gradient trap in spite of its
small size. The trap worked awlessly, and shortly afterward, in December 1986, the work was
featured on the front page of the Sunday New York Times. Surprisingly, a new trapping proposal by
Dave Pritchard of MIT appeared in the same issue of Physical Review Letters as our trapping paper. It
was for a large-volume magneto-optic scattering-force trap rendered stable via a quadrupole Zeeman-
shifting magnetic eld. The magneto-optical trap (MOT) is a relatively deep trap and is easily lled
because of its large size. As later shown, it did not even require any atomic beam slowing.
Shortly after the atom trapping experiment, Chu left Bell Labs for Stanford and continued his atom
trapping work. At Bell Labs, Bjorkholm and Ashkin turned to other work. Use of MOT traps
dominated over dipole traps for atom work for about the next ten years. In 1997 the Nobel Prize
in Physics was awarded to Chu, Phillips, and Cohen-Tannoudji for cooling and trapping of atoms.
In the lab, with the help of Joe Dziedzic, the author started looking at the use of focused laser beams
as tweezers for the trapping and manipulation of Rayleigh particles. They made a surprising discovery
one morning when while examining a sample that had been kept in solution overnight. Wild scattering
was seen emanating from the focus of the trap. A joke was made about having caught some bugs. On
closer examination it turned out that that this had happened. Bacteria had contaminated the sample,
and they had fallen into the trap. The sample was placed under a microscope where the trapping could
be observed in detail. In fact, the trap could be maneuvered to chase, capture, and release fast-
swimming bacteria with green argon-ion laser light. If the laser power was turned up, opticution was
observed; that is, the cell exploded. It was found that infrared YAG laser power was very much less
damaging. Samples of E. coli bacteria obtained from Tets Yamane of Murray Hill were seen to
reproduce right in the trap. Internal-surgery was performed in which the location of organelles was
rearranged and the organelles were attached in new locations. The visco-elasticity of living cells
cytoplasm and the elasticity of internal membranes were also studied. This early work was the start of a
new, unexpected, and very important application of laser trapping. A Nobel Prize winner at Bell Labs
mentioned, amusingly in retrospect, that the author should not exaggerate by predicting that
trapping would someday be important for the biological sciences.
Meanwhile, work to better understand optical molasses cooling of atoms was carried out at NIST
and Stanford. Importantly, at NIST Phillips had made the surprising discovery of cooling to tem-
peratures as low as 40 K in optical molasses. This was of great interest to those racing to achieve
BoseEinstein condensation (BEC) at very low temperatures and high densities. Anderson et al. won
this race in 1995 using evaporative cooling from a magnetic trap reaching a temperature of about
170 nK at a density of 21012 atoms/cm3 with a loss of evaporated atoms by a factor of 500 from an
original 107 atoms. Eric Cornell, Carl Weiman, and Wolfgang Ketterle received the 2001 Nobel Prize in
Physics for the experimental demonstration of BEC.
The Nobel Committee in their 1997 press releases Addendum B on additional material mainly
for physicists says To become really useful one needed a trap deeper than the focused laser beam trap
proposed by Letokhov and Ashkin and realized by Chu and coworkers in optical molasses experi-
ments. On the contrary, far-off-resonance traps built according to Ashkins design are the traps used in
virtually every current BoseEinstein experiment.
The story of the application of tweezer traps to biophysics and the biological sciences is more
straightforward [57]. After the early work of the author on living cells, Ashkin and collaborators and
Steven Block with Howard Berg showed the usefulness of optical tweezers for studying single motor
molecules such as dynein, kinesin, and rotary agella motors. Block and his co-workers continue to
extend tweezer techniques to DNA replication and protein folding at even higher resolution (fractions
of an angstrom) and lower force levels using super-steady optically levitated low-noise traps held in a
helium gas environment.
Light-pressure forces are probably the smallest controllable and measurable forces in nature. Other
low-force techniques such as atomic force microscopy (AFM) have their unique features but cannot
Optical Trapping and Manipulation of Small Particles by Laser Light Pressure 225
function deep inside living cells, for example. Looking to the future, one expects the interesting work on
motors and protein folding to continue. Perhaps we will see optical tweezers serving as gravitational
wave detectors. Large improvements in atomic clocks have been made in the past using atomic fountain
techniques. Recently another breakthrough has been made using ultracold optical lattice clocks
approaching a stability of one part in 1018. This achievement in time keeping by NIST has many
potential applications.
The study of light is fundamental to physics. As such, one expects that applications of optical
trapping and manipulation of particles by laser light pressure will continue well into the future.
The importance of using lasers for the trapping and cooling of atoms has been recognized by a
number of prizes and awards, including the Nobel Prizes mentioned above. In addition, Arthur Ashkin
has been recognized for his work in that eld by The Optical Society (OSA) with the Charles H. Townes
award in 1988, with the Ives Medal/Quinn Award in 1998, and by being elected an Honorary Member
of OSA in 2010.
Many thanks to John Bjorkholm for his help in editing this essay.
References
1. A. Ashkin, Optical Trapping and Manipulation of Neutral Particles Using Lasers: A Reprint Volume
with Commentaries (World Scientic, 2006).
2. A. Ashkin, Acceleration and trapping of particles by radiation pressure, Phys. Rev. Lett. 24, 156159
(1970).
3. A. Ashkin and J. M. Dziedzic, Observation of resonances in the radiation pressure on dielectric
spheres, Phys. Rev. Lett. 38, 13511354 (1977).
4. J. E. Bjorkholm, R. R. Freeman, A. Ashkin, and D. B. Pearson, Observation of focusing of neutral
atoms by the dipole forces of resonance-radiation pressure, Phys. Rev. Lett. 41, 13611364 (1978).
5. K. C. Neuman and S. M. Block, Optical trapping, Rev. Sci. Instrum. 75, 27872809 (2004).
6. K. Dholakia, P. Reece, and M. Gu, Optical micromanipulation, Chem. Soc. Rev. 37, 4255 (2008).
7. D. G. Grier, A revolution in optical micromanipulation, Nature 424, 810816 (2003).
226 Optical Trapping and Manipulation of Small Particles by Laser Light Pressure
19751990
T
he long-lived diode lasers demonstrated at Bell Laboratories in 1977 produced only a
couple of milliwatts (mWs), good enough for ber-optical communications and later for
compact disc reading. Other applications, such as high-speed optical recording, required
quasi-continuous-wave (CW) powers in the 50100-mW range delivered reliably in a single
spatial mode.
Since the reliable power is closely related to the optical power density that can damage the
emitting facet, designs were needed for enlarging the laser spot size both transversely (i.e., in a
direction perpendicular to the plane of the grown layers) and laterally, while maintaining a
single spatial mode. In conventional double-heterojunction devices, for which single transverse
optical-mode operation is ensured, the main challenge was to create single-mode structures of
large lateral spot size. This was realized by introducing mode-dependent radiation losses in
so-called antiguided structures, in either the lateral or the transverse directions, on both sides of
the dened lateral waveguide. Laterally antiguided diode lasers [1] emitting single-mode peak
powers in the 5080-mW range at 20%50% duty cycle enabled RCA Laboratories in 1980 to
realize high-speed optical recording. At about the same time, Hitachi Central Research
Laboratory reported single-mode CW powers as high as 40 mW employing optimized
transversely antiguided double-heterojunction devices [2].
In 1980 a breakthrough occurred in high-power diode-laser design with the implementation
of the large-optical-cavity concept for increased spot size in the transverse direction [1]. These
structures provided transverse spot sizes about 60% larger than in double-heterojunction
devices, enabling record-high reliable powers [1]. As a result, the constricted double-hetero-
junction, large-optical-cavity laser became the most powerful single-mode commercially avail-
able diode laser between 1981 and 1986.
In the 1980s the maximum reliable CW power was only about 25% of the maximum
achievable power set by catastrophic optical-mirror damage. Mirror damage in diode lasers is
caused by thermal runaway at the mirror facets due to increased light absorption and non-
radiative recombination with increased drive current [3]. Solutions to suppressing damage
required nonabsorbing regions at the mirror facets. As early as 1978 researchers from NEC
Laboratories showed that Zn diffusion provides nonabsorbing regions at the mirror facets. This
led to a fourfold increase in the maximum achievable CW output power. Then, in 1984
researchers from RCA Laboratories demonstrated mirror-damage suppression by creating, via a
single etch-and-regrowth cycle, two-dimensional (2D) waveguiding structures at the mirror
facets that were transparent to the laser light. Those devices [3] provided peak output powers
of 1.5 W, a fourfold increase over the highest previously reported. However, the early
nonabsorbing-mirror approaches were impractical to implement. It took over ve years before
practical nonabsorbing-mirror lasers were developed and became commercially available.
Around 1982, interest arose in replacing ashlamps with diode-laser arrays as pumps for
solid-state lasers. This drive picked up steam with the advent of quantum-well diode lasers
since much lower threshold currents could be achieved than in standard double-heterojunction
lasers. In early 1983 researchers from Xerox PARC reported very high (>2.5 W CW) CW
power quantum-well lasers with optimized facet coatings [3]. Thus, they achieved an eightfold
227
increase over the maximum CW power
reported from double-heterojunction
lasers due both to the use of quantum
wells and the use of low-reectivity di-
electric facet coatings. The facet coatings
also prevented attack and erosion of the
cleaved facets in air, enhancing device
reliability [3]. By the mid-1980s large-
aperture, high-power, reliable diodes
lasting at least 10,000 hours became com-
mercially available [3] from Spectra Di-
ode Laboratories Inc., a start-up company
spun off from Xerox PARC. Later that
decade, quantum-well laser optimization
employing a single, thin quantum well in
a large-optical-cavity connement struc-
ture resulted in front-facet, maximum
CW wall-plug efciency as high as 55%.
Quantum-well lasers turned out to offer
a solution for practical nonabsorbing-mirror
lasers. Researchers from the University of
Illlinois at Urbana discovered that impurity
diffusion causes lattice disordering of multi-
quantum-well structures, leading to struc-
tures of higher bandgap energy than the
energy of light generated in undisturbed
multi-quantum-well structures [4]. In 1986,
by using impurity-induced disordering, re-
searchers from Xerox PARC and Spectra
Diode Laboratories achieved nonabsorbing-
Fig. 1. First diode-laser bar to emit 100-W CW power at room mirror structures at the mirror facets [4].
temperature: (a) schematic representation, (b) CW output as a This led to dramatic improvement in
function of drive current. (D. R. Scifres and H. H. Kung, High-
power diode laser arrays and their reliability, Chap. 7 in Diode
maximum CW power from large-aperture
Laser Arrays, D. Botez and D. R. Scifres, eds. [Cambridge devices and was reected in similar im-
University Press, 1994].) provements in the reliable CW power
output from single-mode devices. This
approach led in 1990 to the rst 100-mW CW commercially available single-mode diode laser. An
alternative nonabsorbing-mirror approach was developed at IBM Zurich Laboratories [3,4]. This
approach, called the E2 process, consisting of complete device-facet passivation via in situ bar cleaving
in ultrahigh vacuum and deposition of a proprietary facet-passivation layer, led to reliable operation of
single-mode AlGaAs lasers at 200-mW output power. Today these are the two main nonabsorbing-mirror
approaches for multi-watt, reliable operation of both single-stripe lasers and laser bars.
In the early 1990s single-stripe laser and laser-bar development for pumping solid-state lasers
started in earnest. For single-stripe, facet-passivated devices of ~400-m-wide aperture, Spectra Diode
Laboratories reported maximum CW power of 11.4 W with reliable CW power of ~4 W [3]. Monolithic
laser bars [Fig. 1(a)] composed of an array of 80 separate facet-passivated lasers emitted 100-W CW at
room temperature [Fig. 1(b)] [3]. Laser-bar operation in quasi-CW mode at low duty cycles allowed
effective heat removal; thus, permitting maximization of the energy per pulse and consequently quite
suitable for pumping solid-state lasers. Researchers from Lawrence Livermore National Laboratories
(LLNL) reported highly stable high-peak-power, quasi-CW operation after 1 billion shots from 1-cm-
long bars [4]. Laser bars were further stacked in 2D arrays to deliver the high powers needed for
effective solid-state-laser pumping. Heat removal was a challenging task, and several approaches were
developed [3,4]. At the time, the most efcient way to remove heat from 2D arrays was the silicon-based
References
1. D. Botez, D. J. Channin, and M. Ettenberg, High-power single-mode AlGaAs laser diodes, Opt. Eng.
21(6), 216066 (1982).
2. N. W. Carlson, Monolithic Diode-Laser Arrays (Springer-Verlag, 1994).
3. D. R. Scifres and H. H. Kung, High-power diode laser arrays and their reliability, Chap. 7 in Diode
Laser Arrays, D. Botez and D. R. Scifres, eds. (Cambridge University Press, 1994).
4. R. Solarz, R. Beach, B. Bennett, B. Freitas, M. Emanuel, G. Albrecht, B. Comaskey, S. Sutton, and
W. Krupke, High-average-power semiconductor laser arrays and laser array packaging with an
emphasis on pumping solid state lasers, Chap. 6 in Diode Laser Arrays, D. Botez and D. R. Scifres, eds.
(Cambridge University Press, 1994).
5. D. Botez, Design considerations and analytical approximations for high continuous-wave power,
broad-waveguide diode lasers, Appl. Phys. Lett. 74(21), 31023104 (1999).
6. D. Botez, High-power Al-free coherent and incoherent diode lasers, Proc. SPIE 3628, 210 (1999).
7. M. Kanskar, T. Earles, T. J. Goodnough, E. Stiers, D. Botez, and L. J. Mawst, 73% CW power
conversion efciency at 50 W from 970 nm diode laser bars, Electron. Lett. 41(5), 245247 (2005).
8. G. Yang, G. M. Smith, M. K. Davis, D. A. S. Loeber, M. Hu, Chung-en Zah, and R. Bhat, Highly
reliable high-power 980-nm pump laser, IEEE Photon. Tech. Lett. 16(11), 24032405 (2004).
9. D. Botez, Monolithic phase-locked semiconductor laser arrays, Chap. 1 in Diode Laser Arrays,
D. Botez and D. R. Scifres, eds. (Cambridge University Press, 1994).
10. H. Yang, L. J. Mawst, and D. Botez, 1.6 W continuous-wave coherent power from large-index-step
(n0.1) near-resonant, antiguided diode laser arrays, Appl. Phys. Lett. 76(10), 12191221 (2000).
W
hile the wavelength of any laser can be varied, lasers get classied as tunable when
their tuning range becomes a substantial fraction of their center wavelength.
Despite having lower optical gain than narrow-line rare-earth doped crystal
lasers such as Nd3+-doped YAG, tunable lasers are desirable for a number of reasons. In
laser-based spectroscopy, laser tuning allows one to access spectral features of interest, while
in laser propagation through the atmosphere, tuning can be used to avoid atmospheric
absorption lines. A large tuning range implies the ability to generate and amplify short
pulses of light. The development of practical and efcient tunable solid state lasers has led to
a scientic revolution and an emerging industrial revolution in laser processing of materials,
based on the generation of electromagnetic pulses with femtosecond and recently attosecond
duration.
Most broadly tunable lasers employ ions from the 3d portion of the periodic table.
Figure 1 presents so-called conguration-coordinate diagrams that help explain the broad
tunability of 3d ions. The diagrams are a greatly simplied schematic representation of the
combined energy of the laser-active ion and its environment as a function of the positions of the
atoms surrounding the ion. In equilibrium, the overall energy is minimized, and the system
energy increases as the coordinate deviates from the equilibrium position. Deviation occurs as a
result of the always present vibrations of the atoms, which appear even at the lowest tempera-
tures from the uncertainty principle of quantum mechanics. The left-hand diagram shows the
case where, when the ion energy level changes from a ground state to an excited state, the
equilibrium position for the conguration coordinate is unchanged. The right-hand side shows
the case where the equilibrium position does change.
An important concept regarding the linewidth of the transitions between the ion ground and
excited states is the FranckCondon principle. Stated in classical terms, when an active ion
undergoes a transition, it occurs so quickly that the atomic surroundings do not move, as shown
by the vertical arrows in the diagrams. The left-hand diagram is representative of the type of
narrow-linewidth transitions among levels of the rare-earth ions, since changing the electronic
state of the spatially compact wavefunctions of the rare earths has negligible effect on the
surrounding atoms.
The electronic wavefunctions of 3d ions have a larger spatial extent than those of rare-earths
and have a stronger interaction with their environment. The case illustrated in the right-hand
diagram shows what happens with a strong interaction, exciting the electronic level leads to a
new equilibrium position. As is evident from the arrows, the energy associated with ground-to-
excited-state transitions does vary with the displacement, leading to a large spread in energies
and hence a large linewidth. The energies for the absorption (ground-to-excited) transitions are
generally distinct and higher than those for emission and possible laser operation (excited-to-
ground transitions). As a result, even with only two electronic transitions, one can observe four-
level laser operation (as shown by the numbers in Fig. 1) as the peak absorption and emission
wavelengths do not overlap. These types of transitions are often referred to as vibronic, a
concatenation of vibrational and electronic.
After the demonstration of the ruby laser, and around the same time as the development of
rare-earth-doped lasers, there were demonstrations of the rst broadly tunable solid state lasers,
based on 3d-ion transitions. In particular, in 1963 L. F. Johnson and co-workers at Bell Labs
232
Fig. 1. Conguration-
coordinate diagrams for two
cases of paramagnetic-ion
transitions.
reported optical maser oscillation from Ni2+ in MgF2 involving simultaneous emission of phonons,
which, translated to now-accepted terminology, would be laser operation on vibronic transitions.
Subsequent work by the same group showed operation on vibronic transitions from Co2+ ion in MgF2
and ZnF2 around 17502150 nm, prism-based tuning, albeit in discontinuous segments from Ni:MgF2,
and operation from V2+-doped MgF2 around 1100 nm. The major drawback to these rst vibronic
lasers was that, because of thermally induced non-radiative processes, relatively low-threshold opera-
tion with lamp pumping required cooling of the laser crystals to cryogenic temperatures. The author,
working at MIT Lincoln Laboratory in the 1970s, became aware of the early Bell Labs work and
realized that the use of lasers, rather than lamps, as pump sources could greatly reduce the engineering
complexity of the systems. In particular, Nd-doped solid state lasers operating around 1300 nm proved
effective in pumping both Ni2+ and Co:MgF2 lasers. In the subsequent work, he had some success with
the Co:MgF2 laser, which proved capable of tuning from 16302080 nm at LN2 temperatures and
17502500 nm at room temperature. Other 3d systems he studied showed clear evidence of a problem
that has plagued many tunable solid state lasers: excited-state absorption (ESA). For most ions there are
a number of 3d levels above the rst excited state, i.e., the upper laser level. Depending on the positions
of the levels in the conguration-coordinate diagram, it is possible that, for the desired laser wavelength,
induced transitions to one or several of these levels may be possible. The net cross section that
determines the laser gain is the cross section for transitions to the lower laser level minus the cross
section for transitions to the higher-lying states, and this reduces laser efciency and can even prevent
laser operation.
The announcement of room-temperature, 750-nm-wavelength-region, tunable laser operation
from Cr3+-doped BeAl2O4 (alexandrite) in 1979 re-ignited interest in Cr3+-doped lasers beyond ruby.
At rst, laser operation was thought to be, like ruby, on a narrow-line transition but spectroscopic
investigation showed that it was in fact a vibronic. However, the gain in alexandrite lasers is relatively
low, limiting applications, and today the most widespread use of alexandrite is in lamp-pumped, long-
pulse lasers used for a variety of medical applications. The majority of other Cr3+-doped tunable
materials studied showed low conversion of pump to laser power, generally attributed to ESA. One
class (colquirite structure) of materials, rst developed at Lawrence Livermore National Laboratory,
includes the crystals LiCaAlF6 (LiCAF) and LiSrAlF6 (LiSAF) and was shown to have relatively weak
ESA and thus high efciency. However, the thermo-mechanical properties of the colquirite host crystals
(with thermal conductivities 10%20% of the sapphire and alexandrite host crystals) signicantly limit
their ability to generate high average powers free of signicant thermo-optic distortion of the output
beam and, ultimately, free of fracture to the laser material.
While listening to a presentation on a particular type of color-center laser the author noted the
simplicity of that system: there were no excited states above the upper laser level that could cause
ESA. A subsequent review of the periodic table showed that one 3d ion, Ti3+, has only a single 3d
electron. The ve-fold degenerate free-space state for that electron placed in a typical crystal, to rst
order, splits into a three-fold degenerate ground state, 2T2, and a doubly degenerate upper state, 2E.
Any higher-lying states result from transitions that take the single electron out of the 3d shell and
could be so high in energy as to not create ESA. There were reports on the basic spectroscopy of
Ti3+-doped Al2O3 (Ti:sapphire) with data on absorption and uorescence. Given the superior thermo-
mechanical properties of sapphire, proven with the ruby laser, it looked to be a good choice for a
Ti3+ host.
The author obtained crystal samples from Robert Cobles group at MIT, where they had been
studying the diffusion of oxygen in sapphire by using the oxidation state of Ti as a tracer. (Coble was
the developer of the rst transparent ceramics, paving the way for sodium arc lamps and, later, laser-
quality ceramics.) The authors measurements of the absorption cross section and uorescence spectra,
shown in Fig. 2, showed much broader emission than earlier reports. When one converts the emission
data to gain cross section (also plotted in Fig. 2), multiplying by the necessary (wavelength)5 correction,
the tuning range is unusually broad. The spectral breadth of the emission results, in part, from
JahnTeller splitting of both the ground and upper levels of the ion, leading to a more complicated
conguration coordinate than shown in Fig. 1, where both the ground and excited states have multiple-
energy versus displacement curves. The author also determined the uorescence decay time and found a
room-temperature value of 3.15 s. The short lifetime seemed to indicate low quantum efciency, but if
one estimates the radiative value based on the strength of the measured absorption in the material, as
well as on optical gain measurements, one nds high quantum efciency, on the order of 80% at room
temperature. The short lifetime and associated high gain cross section (in the range 34 10-19 cm2)
result from the trigonal symmetry of the Ti3+ site in sapphire, which acts to strongly activate the dipole-
forbidden 2E 2T2 transitions.
The author rst obtained laser operation from the material in May of 1982 and reported the results
in June at the Twelfth International Quantum Electronics Conference in Munich. There was a delay in
publication in a fully refereed journal until 1985 while the author worked, unsuccessfully, to patent the
system, became engaged in other technical work, and left MIT Lincoln Laboratory to help start a
company. The results published in 1985 included demonstrations of pulsed laser operation with lamp-
pumped, dye-laser pumps, frequency-doubled, Q-switched, Nd:YAG laser pumps, and continuous-
wave (CW) operation with argon-ion-laser pumps, with cryogenic cooling used to obtain true-CW
operation. In laser operation, tuning experiments showed that the observed tuning range and that
predicted by uorescence measurements were in good agreement, conrming that ESA was not a factor
in laser operation.
The rst commercial Ti:sapphire laser product, an argon-ion-laser pumped CW device, was
introduced by Spectra-Physics in 1988 and was followed shortly after by one from the authors
company, Schwartz Electro-Optics, that included an option for a single-frequency, ring-laser congu-
ration. Early applications of the products included use as a diode-laser substitute in the development of
Ultrashort-Pulse Lasers
Erich P. Ippen
Introduction
A particularly remarkable aspect of lasers is their ability to emit shorter ashes (pulses) of light
than achievable with any other means. This ability has, over the years, advanced the
observation and measurement of events from the nanosecond timescale down to the picosec-
ond (1012), femtosecond (1015), and even attosecond (1018) timescales. To use such pulses has
required the development of new methods for measuring and characterizing the pulses
themselves on ultrafast timescales beyond the reach of electronics. These methods have, in
turn, made it possible to study ultrafast phenomena in ways that produced completely new
insights into the evolution of such phenomena in physics, chemistry, and biology [1]. As
ultrashort-pulse laser technology has developed, its other characteristics such as the high peak
power and ultrabroad bandwidth packed into a short pulse have also found important
applications. The compression of even very modest amounts of pulse energy into femtosecond
durations produces sufciently high peak power for precision machining and micro-surgery
without unwanted damage and for nondestructive nonlinear methods of microscopy that
produce three-dimensional (3D) biological imaging with micrometer resolution. The ultra-
broad bandwidths associated with femtosecond pulses have made possible 3D medical imaging
via optical coherence tomography (OCT), simultaneous creation of many wavelength-
multiplexed optical communication channels with only one source, and major advances in
precision spectroscopy and optical clocks [2,3].
The Optical Society (OSA) played a major role in supporting the eld, starting with its
creation of the rst International Conference on Picosecond Phenomena in 1978 (name changed
in 1984 to Ultrafast Phenomena to reect the emergence of femtosecond science and technology).
Held every two years since then (for the 19th time in 2014, the year of this writing) with
continuing OSA support, this successful conference has provided perhaps the greatest testament
to the continuous technological development and widespread impact of the eld with its 19-
volume series of hardcover proceedings [4]. OSA journals became the primary source of
publications on ultrafast optics and photonics. Multiple sessions on ultrafast optics and its
applications every year at conferences like CLEO, QELS, IQEC, and OFC have been essential to
advancing the technology and its applications to science and engineering.
gigawatt range had been demonstrated, to a great extent by the Bradley group at Imperial College.
As the pulse-forming dynamics of dye systems began to be studied in detail, the following question
arose: How were such short pulses generated with saturable absorber dyes having much longer recovery
times? In Nd:glass lasers, pulses were shown in studies to build up from noise, with the saturable
absorber selecting the most intense pulse and determining the nal duration by its recovery time. Dye-
laser pulses were getting much shorter. This could happen, according to the insight of G. H. C. New,
because, although bleaching the saturable absorber could only shape the leading edge of the pulse
shorter than its recovery time, the trailing edge could be shaped by rapid saturation (depletion) of the
dye gain medium. By 1975 all of these analyses were put into the subsequently very inuential steady-
state analytical descriptions, by Haus, of fast and slow saturable-absorber mode-locking that
predicted shapes, durations, and stability [8,9,10].
mode-locked (CPM) geometry in 1981 at Bell Labs reduced pulse durations to the 100-fs level and
further improved stability. The interplay between self-phase modulation and internal dispersion was
analyzed theoretically and optimized experimentally via prism pairs to reduce durations further to
below 30 fs. Rapid progress was made by several groups, and with amplication and external
compression, a record duration of 6 fs, a record that lasted more than a decade, was achieved.
Amplied systems, pumped by either 10-Hz frequency-doubled Nd:YAG lasers (Fig. 4) or by kHz
copper-vapor lasers, further extended the capability of femtosecond technology and its range of
applications. The experiments leading to the 1999 Nobel Prize for chemistry [1] were achieved with
this early femtosecond dye-laser technology.
were achieved by Delfyett and co-workers with pulse compression and semiconductor optical
amplication. Stable, transform-limited pulse generation with semiconductor diodes has, however,
for the most part depended on external-cavity-controlled picosecond sources. Pump-probe investiga-
tions revealed that ultrafast nonequilibrium carrier dynamics in a semiconductor make the generation
of pulses shorter than 1 ps problematic.
Color-Center Lasers
An important capability for early 1.5-m-wavelength ultrafast research was provided by the CW
color-center laser. First mode-locked by synchronous pumping, the KCl color center laser was thrust
into further prominence by Mollenauers demonstration at Bell Labs that it could produce
femtosecond pulses by operating as a soliton laser. This was achieved by coupling the laser
output into an anomalously dispersive, soliton-shaping, optical ber, the output of which was then
coupled back into the laser. It was soon discovered, however, that soliton formation in the ber was
not necessary since this coupled-cavity approach also worked with normal-dispersion ber.
Experiments at MIT further revealed the underlying pulse-shortening mechanism to be the
interference of each pulse with a copy of itself that had been self-phase modulated in the ber.
This method, dubbed additive-pulse mode locking (APM), was shown to be compatible with the
Haus fast-absorber model. Recognized as a means of creating an articial fast absorber out of
reactive nonlinearity in a lossless dielectric, APM then stimulated the application of this technique to
a variety of other lasers [11].
Fiber Lasers
Interest in ber lasers developed rapidly after demonstrations at Southhampton of efcient optical
amplication in low-loss bers doped with rare earths. The key mechanism for ultrashort-pulse
generation in ber lasersnonlinear polarization rotationwas also found to be describable by the
fast-absorber model of Haus developed in the context of APM analysis. Earliest progress was made
using Nd:ber lasers, in both actively mode-locked and passively mode-locked congurations. By 1992
pulse durations as short as 38 fs had been generated at 1.06 m in a Nd:ber laser utilizing nonlinear
polarization rotation and prism pairs for dispersion compensation. By the turn of the century, however,
development of the much more efcient Yb:ber laser led to considerably higher powers at 1 m
wavelengths, with similarly short pulses and more compact geometries. In the late 1980s the attention
of researchers also turned to Er:ber lasers for wavelengths being used for optical ber communications
and where bers were anomalously dispersive, permitting soliton pulse shaping and shortening. Sub-
picosecond pulses were rst achieved, at NRL and at Southhampton in gure-eight geometries that used
a nonlinear loop mirror for intensity modulation and pulse stabilization, and then, at MIT and
Southhampton, in the ring geometry stabilized by nonlinear polarization rotation that achieved
common usage. The MIT stretched-pulse laser achieved shorter pulses and higher pulse energies and
was soon commercialized. Although not geared to the high-power applications of Yb:ber lasers,
Er:ber lasers continue to be pursued for silicon photonics, ber-based communications, and a variety
of eye-safe applications.
References
1. A. H. Zewail, Nobel Prize in Chemistry, 1999.
2. T. W. Hnsch, Nobel Prize in Physics, 2005.
3. J. L. Hall, Nobel Prize in Physics, 2005.
4. Ultrafast Phenomena IXVIII, Springer Series in Chemical Physics (Springer, 19782012).
5. S. L. Shapiro, ed., Ultrashort Light Pulses, 2nd ed., Vol. 18 of Topics in Applied Physics, (Springer-
Verlag, 1984).
6. M. A. Duguay and J. W. Hansen, An ultrafast light gate, Appl. Phys. Lett. 15, 192194 (1969).
Ground-Based Telescopes
and Instruments
James Breckinridge
B
y 1916, the American astronomer George Ellery Hale (see Fig. 1), a founding member of
The Optical Society (OSA), had designed and built an optical solar telescope on
Mt. Wilson and measured the strength of magnetic elds on the Sun using his new
invention: the solar magnetograph. This opened a new era in astronomy and demonstrated to all
the merits of adding optical physics to astronomy. In 1916 the Mt. Wilson Observatory, under
the direction of Hale, had just completed the 60-inch reecting telescope and it was becoming
productive. Hale hired George Ritchey to gure the 60-inch mirror with a hyperbolic primary
and secondary to extend the eld of view (FOV) of the standard Cassegrain telescope. Most
astronomical telescopes today use this optical conguration.
Hales career started out at the University of Chicago, where he met A. A. Michelson (OSA
Honorary Member) in 1889 when he arrived at the University of Chicago. Hale nominated
Michelson for the Nobel Prize in Physics in 1907. In 1916 Hale, director of Mt. Wilson
Observatory, was elected vice-president of OSA. Later (in 1935) he would be awarded the
Frederic Ives Medal. Obsessed with optical astronomy since childhood, Hale graduated from
MIT in physics and studied solar physics at Harvard. Hale recognized the advantages of
reectors and in 1908 used a 60-inch-diameter glass disk given to him by his father to build
the worlds largest telescope on Mt. Wilson in southern California. By 1916 Hale had obtained
funds from John D. Hooker, a Chicago philanthropist, and he was building, once again, the
worlds largest telescope: the 100-inch, dedicated in 1917. The 100-inch Hooker ground-based
telescope is the same size as the Hubble Space Telescope of today. By 1935, Hale had sold the
Rockefeller Foundation on supporting the design and construction of a 200-inch telescope and
set off for a third time to build the worlds largest telescope. George Ellery Hale engaged private
nancial support for optical telescopes from wealthy barons of the industrial revolution: Yerkes,
Carnegie, Hooker, and Rockefeller. Figure 2 shows Hale with Andrew Carnegie in 1910. Hale
established the tradition of private support that continues today with the Keck telescopes, Sloan
Digital Sky Survey, and others.
Using new sensitive photographic emulsions developed by C. E. K. Mees (for whom the OSA
Mees Medal is named), Edwin Hubble (shown in Fig. 3) imaged several Cepheid variables in the
Andromeda Galaxy (M-31). The average luminosity of these variables is constant. Therefore, a
measurement of the brightness of these very faint objects in M31 gives a direct measure of the
distance. The measured distance was well outside our galaxy, demonstrating that spiral nebulae
were outside our galaxy and thus proving that the universe was very large indeed! Hubble went
on to show that the universe was expanding, thus providing fundamental evidence for todays
big bang cosmology.
In 1930 an Estonian optician, Bernard Schmidt, developed his Schmidt camera for the
imaging of large areas of the sky. For the rst time, astronomers could make wide-FOV
surveys needed to study the large-scale structure of our galaxy and to create catalogs of
spectral types and variable stars in an efcient manner. The rst large-aperture Schmidt
cameras were the 40-cm-aperture at Mt. Palomar (1936) and the 60-cm at Case Western
Reserve University (1939).
244
In 1946 Aden Meinel (1982 Ives Medalist,
1952 Lomb Medalist, and OSA President) built
the rst high-speed Schmidt camera and discov-
ered the OH bands in the IR spectrum of the
atmosphere using recently declassied infrared-
sensitive photographic emulsions. James Baker
(OSA Ives Medalist) improved on Schmidts de-
sign to create the BakerNunn camera for wide-
angle observations of articial satellites passing
rapidly overhead.
Hale conceived the 200-inch telescope shortly
after the dedication of the 100-inch telescope in
1917. The task of raising funds, keeping the vision
alive, and preparing conceptual designs occupied
most the 1920s. By 1928 Hale secured a grant of
$6 million from the Rockefeller Foundation to
complete the design and begin construction of the
200-inch telescope on Mt. Palomar. The Corning
Glass Works, an OSA Corporate Member,
working over a ten-year period, developed the
technology and cast the Pyrex primary mirror.
Construction of the observatory facilities began Fig. 1. George Ellery Hale, astronomer and
in 1936 but was interrupted by the onset of World founding member of the OSA. Credit Huntington
War II. The telescope was completed and dedicat- Library, San Marino, California. (The University of
ed in 1948. Ira Bowen (1952 Ives Medalist) Chicago Yerkes Observatory, courtesy AIP Emilio
Segre Visual Archives.)
rened the optical system and the grating spectro-
graphs and rebuilt the mirror support system.
The telescope was not open for scientic use until 1949, and the rst astronomer to use it was
Edwin Hubble.
John Strong (1956 Ives Medalist and OSA President), demonstrated the advantages of using an
evaporative aluminum coating on the 100-inch telescope in 1936. Before this, chemically deposited
silver was used, which degraded rapidly to limit the faintest magnitude that could be recorded. The
reectivity of silver degrades signicantly within a few days. Al coatings on mirrors are robust and with
proper care retain high reectivity for years. This increase in telescope transmittance enabled
astronomers to record stars several magnitudes fainter than before.
During World War II, most optical astronomers were involved in the war effort. Scanners, detectors,
photomultipliers, mirror coatings, manufacturing methods for large glass mirrors, and high-speed
cameras were just a few of the technologies developed by optical astronomers during this period.
At the end of the war optical astronomers returned to civilian jobs. The new infrared-sensitive
photographic lms developed during the conict were now used to extend astronomical discoveries into
the infrared. Photomultipliers were used to make precision measurements of stellar brightness and
color. These data improved our understanding of stellar evolution and reddening (absorption) due to
interstellar matter.
The National Science Foundation was founded in 1950. Its earliest research center was the Kitt
Peak National Observatory founded in 1955 operated under a board of directors from several
university astronomy departments. Aden Meinel, an astronomy professor and optical scientist from
the University of Chicago, was selected to be the founding director. The purpose of the observatory was
to provide astronomical telescope time on a peer-review selection basis to all astronomers in the U.S.
Under Meinels direction the observatory developed the process for the thermal slump of a Pyrex mirror
around a conformal mold (used in the 82-inch telescope), created a rocket program for UV spectros-
copy of stellar objects, developed the worlds largest solar observatory (the 60-inch McMath-Pierce),
developed a 50-inch robot telescope for photoelectric photometry, and laid the groundwork for the rst
program in observational infrared astrophysics.
In 1960 Meinel left the Kitt Peak National Observatory to become the director of Steward
Observatory. There he led the academic program, developed a 92-inch telescope for the University of
Arizona on Kitt Peak Mountain and led an initiative to establish a national center of excellence in
optical sciences and engineering, focused on many issues related to technology for astronomical
telescopes and instruments. In 1964 funding became available, and the University of Arizona
established the Optical Sciences Center under Adens leadership. Aden established a distinguished
faculty composed of A. F. Turner (Ives Medalist), R. R. Shannon (1985 OSA President), R. V. Shack
(David Richardson Medalist), J. C. Wyant (2010 OSA President), and Roger Angel (OSA Fellow).
Figure 4 shows Aden Meinel in 1985 while at NASA/JPL. In 1973 Aden resigned from the directorship
to continue research in solar thermal energy, and Peter Franken (OSA Wood Prize and OSA President)
became director.
In the late 1970s Roger Angel (OSA member) experimented with spin casting Pyrex mirrors for
astronomical telescopes. This development has led to a family of 8-meter ground-based telescopes,
which are revolutionizing our astrophysical understanding of the universe around the world.
In 1920 optical physicist A. A. Michelson (OSA Honorary Member) made the rst measurements
of the diameter of a star using a white-light spatial interferometer mounted to the top of the 100-inch
telescope. Atmospheric seeing and telescope stability prohibited useful data using the photographic
plates of the time, and both he and his colleague F. G. Pease resorted to visual observations of
ickering fringes to measure the diameter of stars. Breckinridge (OSA Fellow) recorded the rst direct
images of the fringes more than 50 years later. C. H. Townes (1996 Ives Medalist and Nobel
I
n 1946, Lyman Spitzer of Princeton University proposed the construction of a space
telescope for astrophysics, and Princeton astronomers launched several balloon-borne
telescopes (Stratoscope project) to operate in the dry excellent seeing provided by the upper
stratosphere to demonstrate the value of space science.
At the very beginning of NASA, Nancy Roman, Lyman Spitzer, and Art Code laid out a
space satellite program that envisioned a series of modest-aperture telescopes for UV and optical
astronomy [the Orbiting Astronomical Observatory (OAO)] and an R&D program leading to a
large space telescope. The seeds of the Hubble Space Telescope were sown 35 years before its
launch.
In 1962 the worlds rst space telescope was launched, and it recorded the UV spectrum of
the Sun. The OAO program became a series of three space telescopes. The rst OAO was to carry
experiments, and observing time was to be shared between the two university groups that
produced the instruments. However, when that satellite was launched, it almost immediately self-
destructed before the scientic instruments could be turned on.
NASA quickly organized an additional launch using ight spares of the satellite and the
scientic instruments. That satellite was successful and is referred to now as OAO-2. It was
launched 7 December 1968, carried 11 UV telescopes, and operated until 1973. OAO-2
discovered that comets are surrounded by enormous halos of hydrogen several hundred
thousand kilometers across and made observations of novae to nd that their UV brightness
often increased during the decline in their optical brightness.
OAO-3 (Copernicus) was orbited in August of 1972 and carried an 80-cm-diameter telescope
for UV astronomy. OAO-3 successfully operated for 14 years and established an excellent
reputation for the highest-quality astronomical data at the time. The Copernicus mission played
a large role in winning the support of the wider astronomical community for space astronomy, not
only because of the very high-quality data it produced, covering the UV to below the Lyman limit,
but also because of the serious commitment Spitzer and his Princeton colleagues showed to making
the data available and easily interpretable. Complete spectra were obtained for only about 500
stars, very modest by todays standards. But the scientic impact of those spectra was huge!
The concept for a series of four large telescopes, called the Great Observatories, evolved at
NASA starting in the 1980s. In order of increasing wavelength they were Compton Gamma Ray
Observatory (CGRO), Advanced X-Ray Astrophysics Facility (AXAF), now called Chandra, the
Hubble Space Telescope (HST), and the Space Infrared Telescope Facility (SIRTF), now called
Spitzer. Optical Society (OSA) members had a major role in the development of AXAF, Chandra,
and Spitzer.
HST started out as the Large Space Telescope (LST) with a 3-meter aperture. Soon the reality
of the launch vehicle capacity set in and NASA issued a request for information to the industry
for a 2.4-meter-diameter telescope. Three optics companies, all corporate members of OSA,
responded with feasibility studies: Eastman Kodak, Itek, and Perkin-Elmer. Perkin-Elmer was
selected as the primary telescope provider. NASA recognized that the longest lead item in the
procurement would be the primary mirror and directed Perkin-Elmer to fund Eastman Kodak to
provide a back-up mirror. This mirror is now at the Smithsonian Air and Space Museum. Corning
manufactured both of the ultra-low expansion (ULE) honeycomb 2.4-meter mirror blanks. PE
was responsible for the telescope, and Lockheed Sunnyvale was the spacecraft system integrator.
249
Large was dropped from the LST name during its development, and later it was renamed after
Edwin Hubble to become the HST. NASA Headquarters issued a competitive-science solicitation for
instruments. These UV/optical/IR science instruments were designed to be replaced on-orbit.
The HST became the worlds rst scientic instrument with the capability to be serviced multiple
times on-orbit. The instruments selected were the Wide-Field Planetary Camera (WF/PC), the Faint
Object Camera (FOC), the Faint Object Spectrograph (FOS), the Goddard High Resolution Spectro-
graph (GHRS), and the High Speed Photometer (HSP). The HST primary mirror was maintained near
room temperature. That combined with the poor IR detectors at the time prohibited an infrared
astronomy instrument.
HST was scheduled for launch in 1986 soon after the Challenger mission that ended in disaster.
The shuttle eet was grounded for 32 months, delaying the HST launch to late April 1990. By the end of
May 1990 it was discovered that the telescope could not be focused, and in June the error was suggested
to be spherical aberration. NASA headquarters formed two teams. One, the ofcial NASA optical
failure review board led by Dr. Lew Allen (a retired four-star general and JPL director) had membership
and support from Optical Society Fellows Roger Angel, Bob Shannon, John Mangus, Jim Breckinridge,
and Bob Parks. This team investigated the root cause of the error. The other board, the Hubble
Independent Optical Review Panel (HIORP) was led by Optical Society Fellow Duncan Moore. Optical
Society Fellows Aden and Marjorie Meinel, Dietrich Korsch, Dan Schulte, Art Vaughan, and George
Lawrence, among others, were members. The HIORP had broad membership from the optics and
astronomy communities and was charged with making recommendations on how to x the error. The
nations optics community came together to establish that the error was on the primary. Nine optics
groups composed of many Optical Society Members and Fellows across the country made independent
measurements on the PE test apparatus hardware and on digital images recorded by the hardware on-
orbit. The recording of star images across the eld of view and at different telescope focus settings
provided a diverse set of image data for the new prescription retrieval algorithms. For the rst time, the
on-orbit optical prescription was determined precisely. The intensity of this work is evidenced by the
fact that it was completed over a ten-week period to meet the instrument rebuild schedule for a repair
mission launch.
An accurate value for the telescope primary-mirror conic constant and the fact that the error was
isolated to the primary enabled corrective optics to be integrated into a newly built WF/PC2 (designed
and built by NASA/JPL), and a new optical system called COSTAR. COSTAR was designed and built
by Ball, an Optical Society Corporate Member. Both instruments were inserted into HST on the rst
repair mission. The COSTAR optical system replaced the HSP instrument. This new optical
system corrected the wavefront for the Faint Object Spectrograph (FOS), the Faint Object Camera
(FOC), and the GHRS. In 1997 the IR
system NICMOS was launched replacing
COSTAR to give the telescope its rst IR
capability to 2-m wavelength.
Today, at the one hundredth anniver-
sary of The Optical Society, the HST has
been successfully operating for 26 years.
By far it is the most productive scientic
UV/optical instrument ever known, a spec-
tacular monument to the space optics com-
munity and the many dedicated Optical
Society members who saved the mission
from disaster. Figure 1 is a photo of the
HST in orbit taken from the space shuttle
after a service mission. One of the most
famous and spectacular photos taken by
HST is shown in Fig. 2. It is the so-called
pillars of creation in the Eagle Nebula
Fig. 1. The HST in orbit. (Image courtesy of NASA.) where stars, and by implication their
A
lthough the rst practical contact lens was described in 1888 [1], glass-blown shells
formed individually to rest on the sclera and vault across the cornea were the norm until
the 1930s. The advent of polymethyl methacrylate (PMMA) made it possible, in a
method pioneered by William Feinbloom [2], to process an all-plastic lens that could be tted by
custom molding or trial tting from a range of premade lenses. This reduced the weight and cost
of lenses while improving comfort and wearing times. It was not until 1948 that Kevin Tuohy, an
optician, made the rst corneal contact lens [3]. Accidentally cutting through a scleral shell at the
edge of the optic zone, Tuohy tried the small-diameter lens that was left on his own eye and
quickly realized that a lens tted within the cornea could be more comfortable and provide longer
wearing times than a scleral shell. The realization by Smelser and Ozanics that oxygen for corneal
metabolism came directly from the atmosphere led to a major shift to corneal contact lenses
because the t could be adjusted to replenish the oxygenated tear lm with every blink, thus
extending comfortable wearing times from just a few hours. The contact lens market expanded
with commercially available corneal contact lens designs enabling the correction of myopia,
hyperopia, astigmatism, and even novel bifocal designs for presbyopia correction.
Otto Wichterle (Fig. 1) was a brilliant Czech polymer chemist who made the worlds rst
soft contact lenses from his newly invented HEMA hydrogel material [4]. This 38% water
content material was highly exible, oxygen permeable, and signicantly more comfortable than
the rigid PMMA corneal contact lenses that were available. Although working behind the Iron
Curtain, an American patent company acquired the intellectual property rights from Wichterle
and licensed them to Bausch & Lomb (B&L). The company licensed both the material and the
novel spincasting manufacturing technique that Wichterle had developed in his own kitchen.
The prototype for this production method was built from an erector set, powered by the electric
motor from his phonograph (Fig. 2). Henry Knoll, a physicist working at B&L and one of a team
assigned to developing the Wichterle prototypes, pointed out the difculty in working with this
hydrogel material. The rst lens we released commercially was called the C series lens, we built
the A series and the B series but neither would stay on the eye after a few blinks. Management
said if the third design didnt work we would give up on the project. The C-series contact lens
design (Fig. 3) tted the eye, and although the optics were compromised by the wildy aspheric
posterior lens surface produced by the spincasting manufacturing process, the lens was a
commercial success when launched in 1971 following FDA approval. The dramatically improved
comfort changed the contact lens industry in the U.S., and ultimately the world, with rigid
corneal contact lenses today accounting for less than 10% of the lenses tted worldwide. Otto
Wichterle was recognized for his great contributions to the world of optics when he was awarded
the R. W. Wood Medal by The Optical Society (OSA) in 1984.
Initially available only in spherical powers to correct myopia and later hyperopia, soft
lenses to correct astigmatism were rst introduced in the U.S. in the early 1980s. Unlike rigid
lenses which mask the astigmatic component of the cornea, soft lenses conform to the
253
Fig. 1. Otto Wichterle, Czech polymer
chemist, inventor of the rst hydrogel
material to be used in making soft contact
lenses. Wichterle was responsible for
making the rst usable soft contact lenses in
his lab behind the Iron Curtain. (AIP Emilio Fig. 2. A model of the rst spin-casting machine
Serge Visual Archives, Physics Today that Otto Wichterle used to make the rst soft contact
Collection.) lenses in his kitchen.
underlying corneal shape, requiring a method of stabilization and orientation to be built into the
physical shape of the lens. The most successful designs used an increasing thickness prole in the
vertical meridian of the lens, allowing the squeeze force of the upper eyelid to stabilize the lens on
the eye between blinks. Multifocal soft lenses designed to correct presbyopia were introduced by
B&L and CIBA VISION in 1982. B&L used its early experience with signicant spherical aberration
in its rst lenses for myopia to help manufacture a lens with sufcient spherical aberration to expand
the depth of eld of the wearer. Ironically, after spending years trying to eliminate spherical
aberration inherent in the spincast lens product, B&L was purposely designing it in the lens with
the PA1 bifocal.
A major issue with soft contact lenses over the 1970s and 1980s was combating adverse ocular
responses related to deposition of protein and lipid on lens surfaces from the tear lm. This required
daily cleaning and disinfection routines and impacted the longevity of the lenses, prescribed as a single
pair to be worn daily for as long as they lasted, typically a year or more. A second issue was transmitting
sufcient oxygen from the atmosphere through lenses to ensure an adequate physiological environment
for the cornea. Many patients had their lens wear curtailed from insufcient oxygen being available to
the eyes during wearing. This was also the time of continuous wear, a modality where patients wore
their contact lenses constantly, with removal as needed for cleaning (typically every 30 days in the early
1980s) [5]. Although convenient, continuous wear only exacerbated the issues of deposition, reduced
lens life, and caused a signicant increase in ocular adverse responses due to reduced oxygen availability
to the cornea. In 1982, a small company in Denmark started cast molding hydrogel contact lenses and
packaging them in small plastic blisters with foil covers. All other companies delivered their lenses
individually, stored in a small glass serum vial, packaging that dated back to the original B&L lens.
Danalens was the rst disposable contact lens and lit the fuse on a major upheaval in the contact lens
industry (Fig. 4). Johnson and Johnson, sensing an opportunity to enter the lucrative contact lens
254 Contact Lenses for Vision Correction: A Journey from Rare to Commonplace
market in the U.S., acquired the Danalens produc-
tion process and a small contact lens company
called Vistakon whose hydrogel lens material was
already approved by the FDA. Within ve years
Vistakon launched the rst disposable lens in the
United States (1987). Launched as a continuous-
wear lens to be replaced weekly, the marketplace
eventually dictated its use as a daily wear only (no
overnight wear) lens with a biweekly replacement
schedule. Although the oxygen permeability of these
new lenses was no better, the fact that patients could
buy them for only a few dollars each (previously
patients would typically pay hundreds of dollars for
a pair of lenses) and replace them frequently made
them a rapid success. Toric and multifocal options
soon followed as companies invested in the
manufacturing capacity necessary to process these
complex designs for a low cost. As manufacturing
technology improved and cost of goods decreased,
the option of a truly disposable lens, one that was
worn once and then discarded, became a reality.
Vistakon again led the industry by launching the
rst daily disposable contact lens in 1994. Although
the cost of each lens was less than one dollar to the
patient, the high annual cost prohibited rapid adop-
tion of daily disposables, and it was another decade
before this modality made any signicant inroads
into the marketplace. Fig. 3. The rst commercially available soft lens in
In the intervening years, others were still the U.S., the Bausch & Lomb C Series. (Courtesy of
chasing the ultimate in convenience, a lens that Andrew Gasson.)
was so physiologically compatible with the eye
that it could be worn continuously for 30 days
without the risk of adverse ocular responses. The massive oxygen permeability of silicone elastomer
led researchers to develop lenses made from this material in the late 1970s, with Dow Corning being
the most well-known manufacturer to try this alternative material. Although physiologically
successful, silicone elastomer lenses had one undesirable and potentially dangerous aw: their
rubber-like nature generated negative pressure under the lens during wear and resulted in the lens
sticking to the eye. The only path forward was a hybrid material, a silicone hydrogel. Although
seemingly simple, material scientists were essentially trying to mix oil and water and maintain a
transparent material. B&L, the rst company to bring soft hydrogel contact lenses to the market in
1972, were also the rst to develop a commercially viable silicone hydrogel lens. This lens provided
four times the oxygen transmission of hydrogel lenses, and it was approved for up to 30 days of
continuous wear in 1999. Clinicians immediately noted that highly oxygen transmissive lenses
eradicated signicant adverse responses related to oxygen deprivation at the cornea, but they were
slow to adopt silicone hydrogel lenses due to the up to 30 days continuous-wear indication awarded
by the FDA. Experience over the years had shown that corneal ulcers, or microbial keratitis, was the
single most signicant adverse response associated with continuous wear, with the FDA limiting
approval of all hydrogel lenses to six nights maximum in 1989 over their concern with incidence
levels. Clinicians and companies now recommend silicone hydrogel lenses for daily wear or extended
wear with monthly or more-frequent replacement, but the largest area of growth within the contact
lens industry is the daily wear modality.
Currently available soft lens materials provide excellent physiological compatibility with the eye,
and the cornea specically, when worn in a daily wear modality, and so the focus of the industry has
Contact Lenses for Vision Correction: A Journey from Rare to Commonplace 255
moved to improving end-of-day comfort through
design and material formulation, as well as im-
proved optical performance. This last development
has been driven by the development of clinically
applicable HartmannSchack wavefront sensors.
Porter et al. [5] measured the wavefront error of
the eye of a large contact-lens-wearing population,
identifying that the Strehl ratio of the eye can be
signicantly improved by correcting at least the
major higher-order wavefront aberrations. This
technique proved to be an ideal method to evaluate
the optical performance of contact lenses on and off
the eye, and OSA members led the development of
standards for reporting the optical aberrations of
eyes. Ideally, individual prescription contact lenses
should be made for each eye based on wavefront
measurements performed in a clinical setting, en-
abling correction of all higher-order aberrations for
improved low-light vision. Although the feasibility
Fig. 4. Examples of the rst disposable soft contact of this concept has been demonstrated by Marsack,
lens. Although the Danalens lens design and material the challenge for industry is to deliver these custom-
were not unique, the packaging and delivery concept
were innovative and ultimately changed the way
optics contact lenses in the same low-cost, dispos-
contact lenses were sold the world over. (Courtesy of able paradigm that patients and clinicians are cur-
Andrew Gasson.) rently using. In the meantime, at least one manufac-
turer (B&L) is altering the inherent spherical aber-
ration of their spherical and toric contact lens
products using aspheric optical surfaces to minimize the spherical aberration magnitude of the eye
with the lens in place and improve the quality of vision under low-illumination conditions.
References
1. R. M. Pearson and N. Efron, Hundredth anniversary of August Mllers inaugural dissertation on
contact lenses, Surv. Ophthalmol. 34, 133141 (1989).
2. R. B. Mandell, Contact Lens Practice, 4th ed. (Charles C. Thomas, 1988).
3. K. M. Tuohy, Contact lens, U.S. patent 2,510,438, led 28 February 1948. Issued 6 June 1950.
4. O. Wichterle and D. Lim, Hydrophilic gels for biological use, Nature 185, 117118 (1960).
5. J. Porter, A. Guirao, I. G. Cox, and D. R. Williams, Monochromatic aberrations of the human eye in a
large population, J. Opt. Soc. Am. A 18, 17931803 (2001).
256 Contact Lenses for Vision Correction: A Journey from Rare to Commonplace
19751990
Physics of Ablation
Ablation occurs when the laser uence is
such that the energy deposited in a volume
of tissue is sufcient to break the chemical
and physical bonds holding the tissue to-
gether producing a gas that is under high
pressure. The gas then expands away from
the irradiated surface, carrying with it
most of the energy that was deposited into
the volume that absorbed the energy. If the
absorption depth is sufciently shallow
and the pulse duration is sufciently short,
the expanding gas can escape from the
surface in a time that is short compared
with thermal diffusion times, leaving a
clean incision with minimal collateral
damage. These conditions are readily sat-
ised by a short pulse of short-wavelength
light having sufcient energy/unit area,
given that protein and lipids are very
Fig. 1. Three scanning electron micrographs of laser- strong absorbers of ultraviolet light.
irradiated turkey cartilage, recorded from different perspectives
and with different magnication. In the bottom micrograph, arrows
indicate the regions irradiated with 193-nm light and 532-nm light.
For each wavelength, the uence/pulse and number of pulses of Next Steps
irradiation are given.
To develop practical innovative applica-
tions, Srinivasan, Blum, and the author
needed to collaborate with medical/surgical
professionals. To interest these profes-
sionals, they etched a single human hair by
a succession of 193-nm ArF excimer laser
pulses, producing an SEM micrograph
(Fig. 2), showing 50-m-wide laser-etched
notches.
While IBM was preparing a patent
application, Srinivasan, Blum, and the
author were constrained from discussing
Fig. 2. Scanning electron micrograph of a human hair etched their discovery with people outside IBM.
by irradiation with an ArF excimer laser; the notches are 50 m wide. But a newly hired IBM colleague, Ralph
Linkser, with an M.D. and a Ph.D. in
physics, obtained fresh arterial tissue from a cadaver, and Linsker, Srinivasan, Blum, and the author
irradiated a segment of aorta with both 193-nm light from the ArF excimer laser and 532-nm light from
the Q-switched, frequency-doubled Nd:YAG laser. Once again the morphology of the tissue adjacent to
the irradiated/incised regions, examined by standard tissue pathology techniques (Fig. 3), was
stunningly different, with irradiation by the 193-nm light showing no evidence of thermal damage
to the underlying and adjacent tissue [3].
258 Excimer Laser Surgery: Laying the Foundation for Laser Refractive Surgery
This experimental study on freshly
excised human tissue conrmed that exci-
mer laser surgery removed tissue by a
fundamentally new process. Srinivasan,
Blum, and the authors visionthat exci-
mer laser surgery would allow tissue to
be incised so cleanly that subsequent
healing would not produce scar tissue
was more than plausible; it was likely,
subject to experimental verication on
live animals.
Excimer Laser Surgery: Laying the Foundation for Laser Refractive Surgery 259
uncorrected visual acuity of 20/20, the patients myopia was denitely reduced. One serious drawback
of RK was that the depth of the radial incisions left the cornea mechanically less robust. The healed eye
was more susceptible to fracture under impact, such as might occur during an automobile collision.
Trokel speculated that the excimer laser might be a better scalpel for creating the RK incisions.
Upon learning of Srinivasan, Blum, and the authors discovery of excimer laser surgery, Trokel,
who was afliated with Columbia Universitys Harkness Eye Center in New York City, contacted
Srinivasan and brought enucleated calf eyes (derived from slaughter) to the Watson Research Center on
20 July 1983. Srinivasans technical assistant, Bodil Braren, participated in an experiment using the ArF
excimer laser to precisely etch the corneal epithelial layer and stroma of these calf eyes. The published
report of this study is routinely referred to by the ophthalmic community as the seminal paper in laser
refractive surgery [5].
To conduct studies on live animals, the experiments were moved to Columbias laboratories. Such
experiments were necessary to convince the medical community that living cornea etched by the ArF
excimer laser does not form scar tissue at the newly created surface and the etched volume is not lled in
by new growth. The rst experiment on a live rabbit in November 1983 showed excellent results in
that, after a week of observation, the cornea was not only free from any scar tissue but the depression
had not lled in. Further histological examination of the etched surface at high magnication showed
an interface free from detectable damage.
LEsperance, also afliated with Columbia, thought beyond RK and led a patent application
describing the use of excimer laser ablation to modify the curvature of the cornea by selectively
removing tissue from the front surface, not the periphery of the cornea. His U.S. patent 4,665,913 [6]
specically describes this process, which was later named photorefractive keratectomy (PRK).
Soon ophthalmologists around the world, who knew of the remarkable healing properties of the
cornea, were at work exploring different ways to use to excimer lasers to reshape the cornea. From live
animal experiments, they moved to enucleated human eyes, then to blind eyes of volunteers, where they
could study the healing. Finally, in 1988, a sighted human was treated with PRK and, after the cornea
had healed by epithelialization, this patients myopia was corrected.
Development of an alternative technique, known as laser in situ keratomileusis (LASIK) com-
menced in 1987. In LASIK, a separate tool is used to create a hinged ap at the front of the cornea,
preserving the epithelial layer and exposing underlying stroma, which is then irradiated and reshaped
by the ArF excimer laser. After such irradiation, the ap is repositioned over the irradiated area, it
adheres rather quickly, and the patient is soon permitted to blink, while the surgeon makes sure that the
ap stays in place. No sutures are required. The ap acts like the corneas own bandaid, minimizing
the discomfort of blinking. LASIK offers the patient much less discomfort than PRK and much more
rapid attainment of ultimate visual acuity following surgery. For these reasons patients prefer LASIK to
PRK, and far more LASIK procedures are performed than PRK procedures.
However, patients whose corneas are much thinner than average are not good candidates for
LASIK, because a post-LASIK cornea is mechanically weaker than a post-PRK cornea, making the
cornea more susceptible to impact or high-acceleration injury. In fact, the U.S. Navy accepts candidates
into training programs for the Naval Air Force who had their visual acuity improved by PRK, but it
does not accept candidates who had LASIK.
260 Excimer Laser Surgery: Laying the Foundation for Laser Refractive Surgery
required for certain aviators. With further renements in so-called custom wavefront-guided laser
refractive surgery, soon there may be a time when patients undergoing laser refractive surgery may
expect to achieve visual acuity of 20/10.
Public awareness and interest in laser eye surgery was intense even before FDA approval. On 30
January 1987, The Wall Street Journal published an article entitled Laser shaping of cornea shows
promise at correcting eyesight, and on 29 September 1988, The New York Times published its rst
article on PRK, entitled Laser may one day avert the need for eyeglasses. Subsequent articles in the
press dealt with the progress in the research on PRK, the formation of three U.S. companies to market
this procedure and approval by the FDA in 1995. At this point, the surgical procedure was discussed at
length in all the popular media, including The Washington Post, The San Francisco Chronicle,
Newsweek, and The New York Magazine. On 11 October 1999, Time magazine published a cover
story entitled The laser x.
In August 1998, The National Academy of Sciences issued a pamphlet entitled Preserving the
Miracle of Sight: Lasers and Eye Surgery, the stated purpose of which was to show The Path from
Research to Human Benet. One section describes the rst experiments that were done at IBM
Research and, subsequently, at Columbia University, leading to the development of PRK [7].
As for the size of the business of laser refractive surgery, at a typical cost of $2000/procedure,
patients have spent more than $90 billion on PRK and LASIK through the end of 2012.
Srinivasan, Blum, and the author opened the door to this revolution in eye care through their
seminal discovery and subsequent transfer of the technology to the medical/surgical profession. The
OSA presented this group with the R. W. Wood Prize in 2004 for the discovery of pulsed ultraviolet
laser surgery, wherein laser light cuts and etches biological tissue by photoablation with minimal
collateral damage, leading to healing without signicant scarring. In 2013, Srinivasan, Blum, and the
author received the National Medal of Technology and Innovation from President Obama and the Fritz
J. and Dolores H. Russ Prize from the National Academy of Engineering.
References
1. R. Srinivasan and V. Mayne-Banton, Self-developing photoetching of poly(ethylene terephthalate)
lms by far-ultraviolet excimer laser radiation, Appl. Phys. Lett. 41, 576578 (1982).
2. R. Srinivasan and W. J. Leigh, Ablative photodecomposition: the action of far-ultraviolet (193 nm)
laser radiation on poly(ethylene terephthalate) lms, J. Am. Chem. Soc. 104, 67846785 (1982).
3. R. Linsker, R. Srinivasan, J. J. Wynne, and D. R. Alonso, Far-ultraviolet laser ablation of
atherosclerotic lesions, Lasers Surg. Med. 4, 201206 (1984).
4. R. Srinivasan, J. J. Wynne, and S. E. Blum, Far-UV photoetching of organic material, Laser Focus 19,
6266 (1983).
5. S. L. Trokel, R. Srinivasan, and B. Braren, Excimer laser surgery of the cornea, Am. J. Ophthalmol.
96, 710715 (1983).
6. F. A. LEsperance, Jr., Method for ophthalmological surgery, U.S. patent 4,665,913 (19 May 1987).
7. R. Conlan, Preserving the miracle of sight: lasers and eye surgery, in Beyond Discovery: The Path from
Research to Human Benet (National Academy of Sciences, 1998). http://www.nasonline.org/
publications/beyond-discovery/miracle-of-sight.pdf.
Excimer Laser Surgery: Laying the Foundation for Laser Refractive Surgery 261
19751990
B
efore the 1950s, cataracts, a loss of transparency of the human lens causing blindness, had
been treated using procedures such as couching and various forms of intra- and
extracapsular lens extraction (ICCE, ECCE). Minimizing surgical complications and
attaining good postoperative vision were the primary goals of the surgery. Correction of post-
operative aphakia with spectacles was less than satisfactory for patients; their quality of vision
was impacted by the magnication, visual aberrations, and eld loss inherent in the high-
powered positive lenses required to correct the post-surgical eye. Contact lenses provided a
superior optical alternative to spectacles, but mobility in the elderly patients typically undergoing
cataract surgery was a real problem, as contact lenses needed to be inserted and removed
every day.
Sir Harold Ridley (Fig. 1) is universally accepted as the father of intraocular lenses (IOL).
He was the rst to conceptualize a lens that could be surgically implanted in the eye to
compensate for the loss of optical power that occurs when the cataractous lens is removed.
Noting that ghter pilots injured during the early years of World War II with Plexiglass splinters
permanently lodged in their eyes showed no adverse responses, he designed a polymethyl
methacrylate (PMMA) optic to replace the cataractous lens in the eye. In 1949 he performed
the rst surgery to implant a plexiglass intraocular lens. Although the prescription was far from
ideal due to errors in the calculation of the refractive index of the natural lens, the surgery was
considered a success [1]. Ridley IOLs were used in hundreds of similar surgeries over the next
decade, with successful outcomes reported in about 70% of cases. Difculties in maintaining the
lens location in the posterior chamber of the eye and centered on the pupil were the main causes
of failure. Amazingly, although a small number of visionary surgeons followed Ridleys lead in
the use of intraocular lenses to correct for cataract extraction, it would not be until the late 1980s
before it became the preferred method of correction.
From the 1950s through the 1980s, the history of IOL development would be a leap-
frogging of technologies in the placement of the IOL in the eye, IOL mechanical design, surgical
technique, and diagnostic equipment for measuring the intraocular length of the eye. During this
period the lens material of choice was PMMA, with rigid metal or PMMA haptics requiring a
large incision size, polypropolene haptics being introduced to help with centering the lens as the
capsular bag collapsed during the healing process [2].
In 1984, the rst silicone IOL lens, designed by Marzocco and introduced by STAAR, was
brought to the marketplace. The huge advantage of this exible lens was that it could be
introduced through the incision into the eye in a folded conguration, allowing a decrease in the
surgical incision size. The incision length is related to the induction of post-surgical corneal
astigmatism [3], so this signaled the beginning of a drive toward smaller incision sizes that
continues to this day. Ridleys original incision was essentially the full diameter of the cornea,
while today incisions can be as small as 2 mm, using a dedicated injector to fold and introduce the
lens through the incision. It was not until the early 2000s that convergence of these technologies
brought a standard of procedure that is the norm in the United States even today [2]. This
involves a cataract extraction in the capsule via phacoemulsication under topical intracameral
anasthesia. The replacement IOL is a exible, one-piece lens with a square posterior edge
262
(to reduce posterior capsule opacication),
introduced through a 3.0-mm or smaller
incision in the cornea and placed fully
within the capsular bag, with a slight vault
against the posterior surface of the capsule.
Having spent 50 years developing this
procedure to be the preferred option for all
cataract surgeries, even in children, the
industry moved its sights to optimizing the
optical performance of IOLs. In 1989
David Atchison identied the considerable
increase in spherical aberration created by
removing the natural lens and recom-
mended spherical surfaced lens forms that
would correct the majority of this aberra-
tion [4]. He followed this with the sugges-
tion that using aspheric surfaces would not
be benecial, due to the aberrations in-
duced by tilt and decentration of the nal
IOL after healing. Not to be deterred,
Antonio Guirao and several colleagues,
including Pablo Artal and Sverker Norrby,
measured the image quality of the normal
population with age and then of the typical
psuedophakic population. Led by Norrby,
an IOL was developed to correct the aver-
age spherical aberration of the post-surgi-
cal IOL implanted eye. The lens, released Fig. 1. Sir Harold Ridley, universally accepted as the father
to the market by Abbott Medical Optics of IOLs, being the rst to devise, produce, and implant the
(AMO) as the TecnisIOL, was designed rst PMMA IOL. ( National Portrait Gallery, London. Sir
with an aspheric anterior lens surface and (Nicholas) Harold Lloyd Ridley by Bassano Ltd., half-plate lm
negative, 19 May 1972, NPG x171529.)
consideration of the typical decentrations
that occur with IOL surgical placement
and postoperative healing. A rapid response from Alcon provided lenses that corrected a portion of
the spherical aberration of the eye and IOL in combination, and Bausch and Lomb provided a spherical
aberration-free IOL design, ignoring the spherical aberration inherent in the aphakic eye. All three
lenses met with successful use by surgeons around the world, the more technology minded exploring the
concept of using all three lenses along with Zernike analysis of corneal topography measurements to
determine which lens would come closest to nullifying the spherical aberration of an individual eye.
The next challenge was correcting near vision in the pseudophakic eye, which of course, has no
accommodation after removal of the natural lens. Early attempts at multizonal IOLs for correcting
presbyopia demonstrated marginal success due to poor image quality and led to withdrawal from the
market by the early 1990s, but in 1997 AMO released a simultaneous refractive multifocal lens
(distance, intermediate, and near zones of the design were within the patients pupil under normal
illumination) that gained traction in the marketplace until the early 2000s, when complaints of reduced
contrast and halos at night led to a reduction in use [2]. About this time Alcon introduced a diffractive
bifocal IOL design, based on patents bought from 3M but updated with a smaller optic zone (only the
central 3.6 mm encapsulated the bifocal diffractive element) and an apodized energy prole. The lens
had greatest near power at the center of the pupil (equal distance and near), and a shift biased toward
distance power moving from the center to the periphery of the optic zone, with all light focused at
distance outside the 3.6-mm central diffractive zone. Under its marketed name of ReSTOR, this product
met with great enthusiasm when presented to clinicians and continues to grow in popularity, especially
in the latest version, which has a lower add power (reduced from +4 D in the original design to +3 D).
References
1. H. Ridley, Intra-ocular acrylic lensespast, present and future, Trans. Ophthalmol. Soc. UK 84(5),
514 (1964).
2. J. A. Davison, G. Kleinmann, and D. J. Apple, Intraocular lenses, Chap. 11 in Duanes
Ophthalmology (CD-ROM), W. Tasman and E. A. Jaeger, eds. (Lippincott Williams & Wilkins, 2006).
3. K. Hayashi, H. Hayashi, F. Nakao, and F. Hayashi, The correlation between incision size and corneal
shape changes in sutureless cataract surgery, Ophthalmology 102, 550556 (1995).
4. D. A. Atchison, Optical design of intraocular lenses. I. On-axis performance, Optom. Vis. Sci. 66,
492506 (1995).
S
pectacles probably have a longer history than any other optical device, apart from
magniers, and their development has continued throughout the era of The Optical
Society (OSA). A fascinating aspect of this history is that spectacle lens design and
technology involve not only optical solutions to the visual needs of the wearer but also
considerations of comfort, fashion, and appearance. In particular, the diameter of lens required
to t any frame may put serious constraints on the optical characteristics of the lens.
The optics of the human eye should form an image of the outside world on the light-sensitive
retina. Since objects of interest may lie anywhere between distant and relatively close distances of
the order of arms length or less, either the depth of focus of the eye must be very large or, more
realistically in view of the eyes relatively large maximal numerical aperture, 0.25, an active
focusing mechanism is required. Focusing is achieved by active changes in the shape of the elastic
crystalline lens, a process known as accommodation. With accommodation relaxed, the eye
ought to be focused for distance, when it is called emmetropic.
Unfortunately, our evolutionary development has left us with two problems. First, the ocular
dioptics may not form a sharply focused image of distant objects, so that the eye suffers from
ametropia. If the optics are too powerful, the image lies in front of the retina, and the eye is
myopic (short-sighted); if too weak, the image lies behind the retina and the eye is hyperopic
(often erroneously called long-sighted). Evidently the myopic eye can focus clearly on near
objects and the hyperopic eye may be able to increase its power by accommodation to focus both
distant and some near objects. The second problem is that while accommodation was adequate
to the needs of our short-lived ancestors, most of us are now living too long for accommodation
to remain effective in the later part of life. The objective amplitude of accommodation (i.e., the
maximum change in ocular power) for each of us declines steadily from the early teenage years to
reach zero at about 50, when the individual becomes fully presbyopic. Thus, older uncorrected
emmetropes and hyperopes inevitably have poor near vision, although myopes have less
difculty. Almost all older individuals need some form of optical assistance if they are to see
both distant and near objects clearly, the only exceptions being a few happy anisometropic
individuals, having one near-emmetropic eye and one mildly myopic eye.
By 1916, at the time when the OSA was founded, basic spectacle lens design was reasonably
well understood. A variety of types of bifocals were available, including the fused form, where the
bifocal near segment was made of int glass and the distance carrier was made of crown so that
the add effect could be obtained with a lens having no surface discontinuities. Prisms had been
introduced by Von Graefe and Donders to help those with convergence problems. Tints of
various colors and transmittances were available (indeed, as early as Christmas Eve 1666, the
great diarist Samuel Pepys was writing I did buy me a pair of green spectacles, to see whether
they will help my eyes or no). After seven centuries of development, could spectacle lenses be
improved further?
Spectacle lens design and the materials used have, in fact, advanced to a surprising degree
during the OSA century. The earliest relevant paper in the OSAs brave new agship
publication, Journal of The Optical Society of America, appeared in the rst volume under
the title The reected images in spectacle lenses [1]. These reections may interfere with the
265
wearers vision but are generally considered to be most important from the cosmetic point of view. Since
for normal incidence the reectance at the surface of a lens of refractive index n is (n1)2/(n+1)2, the
problem increases as the lens index is raised. Single-layer and multi-layer coatings have, in recent
decades, provided a solution, but questions remain on the optimal coating characteristics, since under
conditions of spectacle use ngerprints and other dirt may, on the lens, be more obvious on the coated
lens, and regular cleaning is required. It is, incidentally, of interest that as late as 1938 Tillyer, in a
discussion on optical glasses given at an OSA symposium on optical materials, still thought it worth
commenting more light gets through the lens when it is tarnished slightlyan earlier, less controlled
form of lens coating!
The question of lens index is also, of course, of great importance in relation to lens thickness and the
consequent appearance of the spectacles when worn. Surface power is given by (n1)/r, where r is the
surface radius. Thus, for any required corrective power, the difference between the two surface
curvatures of a meniscus spectacle lens will be reduced if its index is increased. This means that a
positive lens can have smaller central thickness and a negative lens will have reduced edge thickness for
any given lens diameter. This is of particularly cosmetic value for high myopes wanting a frame that
demands a large lens diameter. Depending upon the material density, the weight of the thinner lens may
also be reduced. Thus, over recent decades there have been continuing and successful attempts to
produce materials of higher refractive index, in both glass and plastic. Whereas traditional crown and
int glasses had indices of 1.52 and 1.62, respectively, materials are now available with indices up to 1.9.
Refractive index and density are, however, not the only consideration with lens materials.
Dispersive characteristics are also important, since when directing the visual axis away from the lens
center the wearer is effectively looking through a prism, resulting in transverse chromatic aberration
and color fringing around objects. Thus, as well as having high index and low density, the ideal lens
material should have as high a constringence (Abbe number, V-Value) as possible. Currently glasses of
refractive index 1.8 have a constringence of about 35.
A major advance in materials was the appearance of plastic lenses. Although polyethyl methacry-
late (PMMA, Plexiglass, Perspex) had been introduced before the second world war, it was relatively
soft and easily scratched. The breakthrough came with a wartime development, CR39, a polymerizable,
thermosetting plastic with a refractive index (1.498) similar to that of crown glass and a V-value of 58.
Importantly, it had better scratch resistance than PMMA, a high impact resistance, and half the density
of crown glass. The rst ophthalmic lenses in the material were produced by Armorlite in 1947. Lenses
can be either surfaced or molded. Demands for still higher impact resistance led to the introduction of
polycarbonate lenses in the late 1950s, rst for safety eyeware and later, as optical quality improved, for
all powers of ophthalmic lens. Polycarbonate is a thermoplastic, and lenses can again be made by either
molding or surfacing techniques. Its index (1.586) is a little higher than crown glass but its V-value (30)
is lower: since the scratch resistance is not high, the surface is usually protected by a hard coating, such
as thermally cured polysiloxane. The specic gravity and UV transmittance are low. Other higher index
plastics are now available. Various hard and anti-reection coatings can be applied to all these plastic
lenses, whose many attractive features have given them a dominant position in the spectacle market.
Ultimately gradient-index media may nd a role in spectacle lens design [2].
From the design point of view, the advent of computers has allowed the impact of aspherization
on the performance of single-vision lenses to be explored in considerable detail [3]. Such work has
revealed that aspherization widens the range of lens forms that yield zero oblique astigmatism as
compared to those lying on the Tscherning ellipse. Modern ray-tracing techniques have also greatly
beneted the design of progressive addition (varifocal) lenses. These are lenses for presbyopes in which
the discrete power zones of traditional bifocals and trifocals are replaced by a smooth variation in
power across the lens surface, from that appropriate for distance vision to that for near, with good
vision for intermediate distances between the distance and near zones and an absence of visible dividing
lines on the lens surface. First proposed by Aves in 1907, with his elephants trunk design, the rst
successful lenses of this type were the French Varilux designed by Maitenaz (Essilor) and, in the U.S., the
Omnifocal (Univis). Since then numerous variations have been produced. Optically, the challenge is
that the shorter the progressive corridor between stable distance and near corrections, the narrower the
corridor and the greater the unwanted astigmatism in neighboring lens areas (Fig. 1). Since the visual
References
1. W. B. Rayton, The reected images in spectacle lenses, J. Opt. Soc. Am. 1(56), 137148 (1917).
2. S. P. Wu, E. Nihei, and Y. Koike, Large radial graded-index polymer, Appl. Opt. 35, 2832 (1996).
3. D. A. Atchison, Spectacle lens design: a review, Appl. Opt. 31, 35793585 (1992).
4. C. W. Fowler, Method for the design and simulation of progressive addition spectacle lenses, Appl.
Opt. 32, 41444146 (1993).
5. V. J. Ellerbrock, Report on survey of optical aids for subnormal vision, J. Opt. Soc. Am. 36, 679695
(1946).
6. L. L. Sloan, Optical magnication for subnormal vision: historical survey, J. Opt. Soc. Am. 62,
162168 (1972).
T
he earliest display of moving images was the motion picture projector, in which light
from a bright lamp was passed through an image on a lm that was then imaged onto a
screen. In the 1920s and 1930s the rst black and white television broadcasts were made
and viewed on small black-and-white cathode ray tube displays. Such a display was achieved by
writing a visible image on a phosphor screen with an electron beam. It required a vacuum tube
and high voltage electronics, yet it produced a reasonable image. Over time cathode ray tube
displays became larger and capable of color images. They also became very heavy, bulky, and
power hungry, though they had good color rendition. However, they were all there were, and the
industry developed color CRTs with screen sizes as large as 1 m in diagonal dimension.
Alternative displays were tried such as plasma screens (an array of tiny, energy-hungry plasmas
that excited special phosphors for each color that quickly were bleached by the UV in the plasma)
or micro-mirror scanner displays. However, all of these were supplanted by the advent of the
liquid crystal display, the LCD. Today these displays dominate the display marketplace due to
their ability to be used in all sizes, from as small as a wristwatch to over 2.8-m-diagonal television
screens. LCDs can be reective, requiring just ambient light to be viewed, transmissive, requiring
a backlight to enable viewing, or transreective, in which a pixel is split into reective and trans-
missive subpixels. In either case their advantages of light weight, lower energy demand, and
scalability have won LCDs a dominant place in todays display marketplace. This essay explores
how that happened.
Liquid crystal is a mesogenic phase existing between crystalline solid and isotropic liquid.
In 1888, Austrian botanist Friedrich Reinitzer and German physicist Otto Lehmann discovered
such an anisotropic liquid crystal. However, in the early days only a few compounds with a
liquid crystal phase were available, and their melting points were quite high. Moreover, to
utilize its large optical anisotropy the liquid crystal has to be aligned and an external eld
applied. Before the optically transparent and electrically conductive indium-tin-oxide (ITO)
lm was available, an alternative way to align a liquid crystal was by applying a magnetic eld.
Therefore, in the rst few decades major research focused on magnetic-eld-induced molecular
reorientation effects. But the electromagnet required to align the liquid crystals was too bulky
to be practically useful. Then in the 1930s Russian scientist V. Fredericksz and colleagues
started to investigate the electro-optic effects in nematic liquid crystals. Some basic concepts
were formulated such as the Fredericksz transition threshold and order parameter, which
described the crystalline state of a liquid crystal. In the 1950s and 1960s, the dynamic behavior
of a liquid crystal cell subjected to an external force, such as a magnetic eld or electric eld,
was investigated by C. W. Oseen, F. C. Frank, J. L. Ericksen, and F. M. Leslie. These concepts
and models provided the foundation for the rapid development of the useful electro-optic
devices that followed.
In the 1960s, American scientists George Heilmeier, Richard Williams, and their collea-
gues at RCA (Radio Corporation of America) Labs developed the dynamic scattering
mode and demonstrated the rst LCD panel [1]. This opened a new era for electronic
displays. Heilmeier was credited with the invention of the LCD. In 2006, he received the
OSA Edwin H. Land Medal, and in 2009 he was inducted into the National Inventors Hall of
269
Fame. However, the dynamic scattering LCD, which utilized the electric-current-induced electro-
hydrodynamic effect, was intrinsically unstable. Also, its contrast ratio was poor and power
consumption was high. As a result, it had a short life and was ultimately abandoned as a practical
display technology.
In the 1970s, to overcome the instability, poor contrast ratio, and high operation voltage of the
dynamic scattering mode display, Martin Schadt and Wolfgang Helfrich, and James Fergason
independently, invented the twisted nematic (TN) effect and steered LCD in a new and productive
direction. TN is regarded as a major invention of the twentieth century. In 1998, James Fergason was
inducted into the National Inventors Hall of Fame. In 2008 Schadt, Helfrich, and Fergason received the
IEEE Jun-Ichi Nishizawa medal in recognition of their outstanding contribution.
Also in the 1970s, a landmark equally important to TN was the development of stable liquid
crystals called cyanobiphenyls by George Grays group at Hull University [2]. Amazingly, these positive
dielectric anisotropy ( 15) materials are still being used in some wristwatches and calculators in
2016. Meanwhile, to obtain a uniform domain new liquid crystal alignment techniques were developed.
Among them, buffed polyimide deserves special mention because it enables large panel LCDs to be
fabricated. This technique is still commonly used in modern LCD fabrication lines. Liquid crystals need
a small pre-tilt angle (35) to guide their reorientation direction when activated by an electric eld.
Otherwise, different domains could be formed, which caused spatially inhomogeneous electro-optic
behaviors. In addition to TN, vertical alignment (VA) and in-plane switching (IPS) were invented in the
1970s. In TN and VA cells, the electric eld is in the longitudinal direction, while in an IPS cell the
electric eld is in the lateral direction, also called the fringing eld. These three modes form the bases of
modern LCD technologies. TN is used in notebook computers and personal TVs in some aircraft
because of its low cost and high transmittance; multi-domain VA is widely used in high-denition TVs
because of its unprecedented contrast ratio; and IPS is commonly used in mobile displays, such as
iPhones and iPads, because of its robustness to external mechanical pressure allowing use in touch
screens.
Another crucial development in the 1970s was the thin lm transistor liquid crystal display (TFT
LCD) led by Bernard Lechner at RCA and Peter Brodys group at Westinghouse. In 1972, a group at
Westinghouse led by A. G. Fisher demonstrated that a color TV could be made by integrating red (R),
green (G), and blue (B) spatial color lters with liquid crystal pixels as intensity modulators [3]. Each
color pixel was independently controlled by a TFT. This combination of TFT and LCD enabled high
information content and became the foundation of todays display industry. In 2011, three TFT
pioneersBernard Lechner, Peter Brody, and Fang-Chen Luoreceived the IEEE Jun-Ichi Nishizawa
medal, and in 2012 Heilmeier, Helfrich, Schadt, and (the late) Brody received the prestigious National
Academy of Engineerings Charles Stark Draper Prize to recognize their engineering development of
LCD utilized in billions of consumer and professional devices.
The early TFTs developed by Brody and his colleagues were based on cadmium selenide (CdSe),
which was never commercialized because of high off-current and reliability issues. Today, most LCDs
use silicon TFTs: amorphous silicon for large panels [>10-in. (25 cm) diagonal], poly-silicon for small-
to-medium panels such as iPhones/iPads, and single-crystal silicon for micro-displays. Recently, oxide
semiconductors, e.g., InGaZnO2 with mobility about 20 higher than that of amorphous silicon, have
been attempted in TFT LCDs by major display producers. The high mobility of oxide semiconductors
helps to shrink TFT feature size, which in turn leads to a larger aperture for higher backlight
throughput.
In the 1980s, passive matrix and active matrix addressed LCDs were pursued in parallel. In the
passive matrix camp, a new LC mode called super-twisted nematic (STN; twist angle >90) was
developed to steepen the voltage-dependent transmittance curve to increase information content.
However, the viewing angle, contrast ratio, and response time of STN are far from satisfactory. In
the active matrix camp, Seiko, Epson, and several Japanese display leaders invested heavily in active
matrix TFT-LCD production facilities. In the meantime, new high-resistivity uorinated liquid crystals
were developed; this technology is required for active matrix operation to avoid image ickering. After
nearly a decade of erce competition, active matrix outperformed passive matrix and is commonly used
in display products.
Introduction
Govind Agrawal
T
his section covers the 25-year period extending from 1990 to 2014. This period is often
referred to as the Information Age because of the advent of the Internet during the early
1990s. It is also the period during which computer technology became mature enough
that it became difcult to imagine life without a computer. These developments affected quite
dramatically both the eld of optics and The Optical Society devoted to serving it. The articles in
this section make an attempt to document the advances made during this recent period and how
they impacted the functioning of The Optical Society.
The most dramatic story of the 1990s is the exponential growth in the capacity of optical
communication networks, fueled by the advances such as wavelength division multiplexing
and erbium-doped ber ampliers. A set of three articles provides the sense of history of this
period. In the rst one, Jeff Hecht discusses the birth and growth of ber-optic communication
industry starting in 1970 when Corning rst announced the invention of the low-loss ber. In
the second of his articles, Jeff Hecht describes how the telecommunication industry grew so
rapidly during the 1990s that it led to a telecom bubble in the stock market that burst
eventually in 2001. In the third article, Rod Alferness, who was at the forefront of this
revolution taking place during the 1990s, provides his perspective on the evolution of optical
communication networks since 1990.
A set of six articles provides a avor of how the eld of optics is evolving in the twenty-rst
century. They cover diverse research areas ranging from integrated photonics to biomedical
optics to quantum information. The rst article by Radha Nagarajan focuses on the recent
advances in the area of integrated photonics that are behind the revival of the telecommunication
industry after bursting of the telecom bubble in 2001. It is followed by Phillip Russells article
on the new wave of microstructured optical bers. Russell was the rst one to make bers known
as photonic crystal and photonic bandgap bers. Here is your chance to hear the history from the
inventor himself.
The third article in this section, by Wayne Knox, covers the history of ultrafast laser
technology. Knox has been involved with ultrafast lasers for a long time and knows their history
well. The fourth article is devoted to advances in biomedical optics. Greg Faris describes in this
article both the in vivo and in vitro applications made possible by recent advances in the area of
biomedical optics. In the next article, David Hagan and Steven Moss focus on novel optical
materials that are likely to revolutionize the twenty-rst century. The last article by Carlton
Caves is devoted to the history of the emerging eld of quantum information.
It was difcult to choose among a wide range of topics, and many could not be included
because of space limitations, among other things. It is my hope that the reader will gain an
appreciation of how the eld of optics is evolving during the twenty-rst century.
277
1991-PRESENT
F
iber-optic communications was born at a time when the telecommunications industry had
grown cautious and conservative after making telephone service ubiquitous in the United
States and widely available in other developed countries. The backbones of the long-
distance telephone network were chains of microwave relay towers, which engineers had planned
to replace by buried pipelines carrying millimeter waves in the 60-GHz range, starting in the
1970s. Bell Telephone Laboratories were quick to begin research on optical communications
after the invention of the laser, but they spent the 1960s studying beam transmission through
buried hollow confocal waveguides, expecting laser communications to be the next generation
after the millimeter waveguide, on a technology timetable spanning decades.
Cornings invention of the low-loss ber in 1970 changed all that. Bell abandoned the
hollow optical guide in 1972 and never put any millimeter waveguide into commercial service
after completing a eld test in the mid-1970s. But telephone engineers remained wary of
installing ber without exhaustive tests and eld trials. Bell engineers developed and exhaustively
tested the rst generation of ber-optic systems, based on multimode graded-index bers
transmitting 45 Mb/s at 850 nm over spans of 10 km, connecting local telephone central ofces.
Deployment began slowly in the late 1970s, and soon a second ber window opened at 1300 nm,
allowing a doubling of speed and transmission distance. In 1980, AT&T announced plans to
extend multimode ber into its long-haul network, by laying a 144-ber cable between Boston
and Washington with repeaters spaced every 7 km along an existing right of way.
Yet by then change was accelerating in the no-longer stodgy telecommunications industry.
Two crucial choices in system design and the breakup of AT&T were about to launch the
modern ber-optic communications industry. In 1980, Bell Labs announced that the next
generation of transoceanic telephone cables would use single-mode ber instead of the copper
coaxial cables used since the rst transatlantic phone cable in 1956. In 1982, the upstart MCI
Communications picked single-mode ber as the backbone of its new North American long-
distance phone network, replacing the microwave towers that gave the company its original
name, Microwave Communications Inc. That same year, AT&T agreed to divest its seven
regional telephone companies to focus on long-distance service, computing, and communica-
tions hardware.
The submarine ber decision was a bold bet on a new technology based on desperation.
Regulators had barred AT&T from operating communication satellites since the mid-1960s.
Coax had reached its practical limit for intercontinental cables. Only single-mode ber trans-
mitting at 1310 nm could transmit 280 Mb/s through 50-km spans stretching more than
6000 km across the Atlantic. AT&T and its partners British Telecom and France Telecom set
a target of 1988 for installing TAT-8, the rst transatlantic ber cable. More submarine ber
cables would follow.
In 1982, MCI went looking for new technology to upgrade its long-distance phone network.
Visits to British Telecom Research Labs and Japanese equipment makers convinced them that
single-mode ber transmitting 400 Mb/s at 1310 nm was ready for installation. AT&T and
Sprint soon followed, with Sprint ads promoting the new ber technology by claiming that callers
could hear a pin drop over it.
278
Fueled by the breakup of AT&T and intense competition for long-distance telephone service, ber
sales boomed as new long-haul networks were installed, then slumped briey after their completion.
The switch to single-mode ber opened the room to further system improvements. By 1987,
terrestrial long-distance backbone systems were carrying 800 Mb/s, and systems able to transmit 1.7
Gb/s were in development. Long-distance trafc increased as competition reduced long-distance rates,
and developers pushed for the next transmission milestone of 2.5 Gb/s. Telecommunications was
becoming an important part of the laser and optics market, pushing development of products including
diode lasers, receivers, and optical connectors.
Fiber optics had shifted the telephone industry into overdrive. Two more technological revolutions
in their early stages in the late 1980s would soon shift telecommunications to warp speed. One came
from the optical world, the ber amplier. The other came from telecommunicationsthe Internet.
Even in the late 1980s, the bulk of telecommunications trafc consisted of telephone conversations.
(Cable television networks carried analog signals and were separate from the usual world of
telecommunications.) Telephony was a mature industry, with trafc volume growing about 10% a
year. Fiber trafc was increasing faster than that because ber was displacing older technologies
including microwave relays and geosynchronous communication satellites. Telecommunications net-
works also carried some digital data, but the overall volume was small.
The ideas that laid the groundwork for the Internet date back to the late 1960s. Universities began
installing terminals so students and faculty could access mainframe computers, ARPANET began
operations to connect universities, and telephone companies envisioned linking home users to main-
frames through telephone wiring. Special terminals were hooked to television screens for early home
information services called videotex. But those data services attracted few customers, and data trafc
remained limited until the spread of personal computers in the 1980s.
The rst personal computer modems sent 300 bits/s through phone lines, a number that soon rose
to 1200 bits/s. Initially the Internet was limited to academic and government users, so other PC users
accessed private networks such as CompuServe and America Online, but private Internet accounts
became available by 1990. The World Wide Web was launched in 1991 at the European Center for
Nuclear Research (CERN) and initially grew slowly. But in 1994 the number of servers soared from 500
to 10,000, and the data oodgates were loosed. Digital trafc soared.
By good fortune, the global ber-optic backbone network was already in place as data trafc
started to soar. Construction expenses are a major part of network costs, so multi-ber cables were laid
that in the mid-1980s were thought to be adequate to support many years of normal trafc growth.
That kept the Information Superhighway from becoming a global trafc jam as data trafc took off.
The impact of ber is evident in Fig. 1, a chart presented by Donald Keck during his 2011 CLEO
plenary talk. Diverse new technologies had increased data transmission rates since 1850. Fiber optics
became the dominant technology after 1980 and is responsible for the change in slope of the data-rate
growth.
Acknowledgment
Part of this material is adapted, with permission, from Jeff Hecht, City of Light: The Story of Fiber
Optics (Oxford, 2004).
F
iber-optic ampliers and wavelength-division multiplexing (WDM) developed almost
perfectly in phase with the explosive growth of the Internet in the 1990s. The new optical
technology promised the bandwidth needed to carry fast-growing Internet trafc. Initially
the parallel advances of optical and Internet technology seemed an ideal match. Unfortunately,
that pairing ignited a speculative bubble that went out of control, creating trillions of dollars of
vastly inated stock valuation that vanished when the bubble collapsed.
An earlier chapter describes how ber became the backbone of the global telecommunica-
tions network. The roots of the Internet go back to the late 1960s, when low-loss bers were still
in development. The Defense Advanced Research Projects Agency (DARPA) (then called ARPA)
began funding computer links among university and government laboratories.
Further Readings
1. L. Endlich, Optical Illusions: Lucent and the Crash of Telecom (Simon & Schuster, 2004).
2. J. Hecht, City of Light: The Story of Fiber Optics, Revised and Expanded Edition (Oxford University
Press, 2004).
3. O. Malik, Broadbandits: Inside the $750 Billion Telecom Heist (Wiley, 2003).
4. B. McLean and P. Elkind, The Smartest Guys in The Room: The Amazing Rise and Scandalous Fall of
Enron (Portfolio-Penguin, 2004).
5. A. Odlyzko, Internet growth, myth and reality, use and abuse, http://www.dtc.umn.edu/~odlyzko/
doc/internet.growth.myth.pdf.
Introduction
Optical communication networks have played a critical role in the information/communication
revolution and in turn have fundamentally changed the world and daily life for billions around
the globe. Without cost-effective, high-capacity optical networks that span continents and connect
them via undersea routes, the worldwide Internet would not be possible. Optical access systems,
both ber/cable and ber-to-the home, are also essential to bring broadband access to that global
Internet to homes and businesses. Increasingly important, ubiquitous broadband optical networks
provide the high-bandwidth backhaul essential for wireless access networks that enable
todays smartphone users. These networks also provide the always available broadband access
that will make cost-effective and energy-efcient cloud services available to all in the future.
All this has been made possible because, as capacity demand has grown exponentially
following the advent of the Internet, optical technology has made possible a dramatic reduction
in the cost per bit carried over an optical ber, allowing cost-effective capacity scaling. On
average, transmission capacity over a single ber has increased at a rate of 100-fold every ten
years over the last thirty years. As a result, as trafc has grown and is aggregated at the ingress
and disaggregated at egress nodes, new higher-capacity generations of long-haul and metro
optical systems have been deployed at a total cost that has grown sub-linearly relative to capacity.
Of course, the advantage of the optical frequencies for communication is the inherent ability to
serve as a carrier for very-high-bandwidth information. Fiber provides an extremely attractive
transmission medium that offers both ultra-low loss and low chromatic dispersion. The latter
results in minimal pulse spreading, resulting in low inter-pulse (bit) interference after transmission
over large distances. At its most basic implementation, an optical transmission system requires an
optical source whose generated dc optical signal can be modulated with information at the
information bandwidth of interest, a ber, and an optical detector and supporting electronics.
Figure 1 captures the progress of the hero research transmission experiments [1]. Shown is
the maximum information capacity carried on a single ber versus the year the research results
were achieved. For this review, it is convenient to describe the research progress in ber optic
transmission capacity in three waves or eras. In what follows, we use those generations, each
enabled by a set of critically important optical component technology innovations, to provide an
overview of the advances in optical communications since 1990.
At the start of the 1990s, commercially deployed systems provided per ber capacities of
about 1 Gb/s. They were used primarily in long-haul intercity applications to carry highly
aggregated voice service. At that time, increase in capacity demand was still driven mostly by
population growth as well as some increase in new services such as fax. The wavelength window
utilized was the minimum chromatic dispersion 1.3-m window. To increase time division
multiplexed bit rates (TDM) for xed distance between electrical regenerations, both signal
strength relative to noise and quality of the detected signal with respect to pulse-to-pulse
287
interference are important. To mitigate the reduced
receiver power at higher bit rates, research focused
on moving to the lowest-loss wavelength window
around 1.55 m. Unfortunately, for the standard
single-mode ber then available, chromatic disper-
sion at 1.55 m was signicant. For systems that
employed directly modulated lasers that exhibit
wavelength chirp during the change from the
on to off state, that dispersion caused prob-
lematic pulse spreading interference.
Three technology advances were instrumental
in strongly mitigating these limitations to enable
increased TDM rates. To avoid chromatic disper-
Fig. 1. Reported research transmission systems sion, it was essential that the semiconductor laser
experiments showing maximum transmission capacity operating at 1.5 m be truly single frequency. That
over a single ber vs. the year of the research results. capability was provided by the distributed feedback
(DFB) laser, which could also be directly modulated
to provide information encoding. In addition, as TDM rates increased, external optical waveguide
modulators that provided high-speed optical information encoding without the chirping effects
proved to be essential for data rates above several gigabits per second.
Finally, high-gain, high-bandwidth avalanche photodiodes (APDs) to provide reasonable
optical to electrical conversion efciency were also needed for high-speed TDM systems. The
combination of single-frequency lasers operating at 1.5 m, signal encoding with external interfer-
ometer waveguide modulators, and detection with APDs resulted in record transmission experi-
ments (216 Gb/s over 100-km spans) in the early 1990s that led to commercially deployed 10 Gb/s
systems in the late 1990s.
Wavelength division multiplexed (WDM) transmission systems employ multiple wavelengths, each
separately encoded with information that is passively multiplexed together onto a single single-mode
ber, transmitted over some distance, and then wavelength demultiplexed into separate channels whose
information is detected and received. While such systems had been proposed earlier, they had not
initially gained popularity because of the need for a regenerator for each wavelength at repeater sites.
Compared with increasing capacity via TDM, the approach did not scale capacity as cost effectively
as TDM.
The ber amplier totally changed the value proposition of WDM systems. While not a pulse
regenerator, the optical amplier provides relatively low-noise 2030-dB amplicationsufcient to
compensate for transmission loss over 50100 km of low loss ber. Most importantly, the optical
amplier can simultaneously amplify multiple wavelengths, each carrying a high-capacity TDM signal.
Notably, there is no mixing of signals, and amplication can be achieved for signals with arbitrarily
high information rates. Both erbium-doped and Raman-based ber ampliers have been developed,
with the former being the commercial workhorse. The erbium-doped ber amplier gain peaks at about
1.55 mwell aligned to the ber loss minimum.
Besides the ber amplier, the other key enabling technologies for WDM transmission systems are
the wavelength multiplexing and demultiplexing devices and single-mode lasers whose wavelength can
be precisely matched to the mux/demux wavelength response. For large wavelength counts, waveguide
grating routers based on silica waveguide technology are typically employed. Figure 2 shows the 80-
wavelength output from an early silica-based arrayed waveguide router. High-power (100-mW
output power) semiconductor pump lasers are required. Fiber amplied transmission systems are
essentially analog systems where amplier noise from each repeater accumulates, as does dispersive and
nonlinear pulse spreading. Careful dispersion management is very important. Zero-chirp optical
modulators are especially important for signal encoding to leverage the cost effectiveness of the
amplier over longer distances without electrical regenerators.
The rst WDM commercial systems, deployed in terrestrial long-haul applications in the mid-
1990s, employed eight wavelengths at 2.5 Gb/s, a tenfold improvement over the single-channel systems
previously available. As multiplexing devices and amplier performance was improved, the number of
wavelengths was soon doubled and then quadrupled. In the research lab, work focused on WDM for
higher TDM rates, 10 Gb/s and beyond.
The rst WDM systems were deployed over existing standard single-mode ber. However, to
reduce the phase-matched nonlinear mixing effect of ber at its zero-dispersion wavelength, so called
non-zero-dispersion shifted bers were developed. Such bers could be used as the transmission ber
or in the repeater site as a dispersion compensating ber to undo dispersion accumulation. In this
case the transmission ber has sufcient dispersion over the transmission distance to avoid four-wave
mixing but produces pulse spreading that is undone by the compensating ber.
Undersea lightwave systems were an important driver and early adopter of ber amplied WDM
transmission systems that were especially attractive because they avoid undersea high-speed electronics,
which reduced the lead time for reliability testing. In addition, properly designed WDM transmission
systems offered the potential for future capacity growth by increasing the wavelength bit rate or the
number of wavelengths. The rst such system, a transatlantic system, included 16 wavelengths at
2.5 Gb/s each with repeater spacing of 100 km.
In research labs around the world, as multiplexing devices and amplier performance were
improved and techniques to mitigate dispersive and nonlinear transmission impairments developed,
single-ber transmission capacity results were improved, sometimes quite dramatically, every year.
These extraordinary hero transmission systems experiments became the highlight of the post-
deadline session of The Optical Society (OSA, and IEEE Photonics and Communications) sponsored
Optical Fiber Conference (OFC) each year. Increased capacity in transmission systems experimental
results over the years (Fig. 1) were achieved by increasing the per wavelength bit rate from 2.5 Gb/s to
10 Gb/s to 40 Gb/s to 100 Gb/s. Key issues that needed to be addressed included demonstrating high-
speed electronics, modulators, and receivers at the higher rates; mitigating nonlinear ber; and
managing dispersive effects. Total capacity was also increased by increasing the number of wave-
lengths. This was achieved either by increasing the bandwidth of the amplier or by nding ways to
reduce the wavelength spacing without reducing the information rate/wavelength, resulting in
improved spectral efciency.
The adoption of WDM transmission led to wavelength-based recongurable optical networks that
provide wavelength-level, cost-effective network bandwidth management. That evolution is shown
schematically in Fig. 3. Initially WDM was employed over linear links where all wavelengths were
aggregated onto the ber at one node and carried with periodic amplication to an end node. However,
in real networks, especially as the distance achievable without electronic regeneration has been
increased, the sources and destinations of trafc require off and on ramps for trafc entering between
large metropolitan areas. Optical wavelength add/drop multiplexers provide those high-capacity on/off
ramps with a full wavelength of capacity and allow all other wavelengths to pass through the node
beneting from the amplication. While initially these were xed in number and which wavelengths
were added/dropped, these modules are now fully remotely recongurable with respect to both the
number of channels and which wavelengths are added/dropped.
Networks are not linear but are meshed to enhance resilience to equipment failures and ber cuts.
They require branching points where several ber routes coming into a major metropolitan area
connect to several exiting routes and also drop/add wavelengths at the node. In this case optical switch
modules, referred to as optical cross-connects, which, in a wavelength-selective manner connect
wavelength channels from one input ber route to a particular output route, are employed.
Automated, recongurable optical switch cross-connects have become essential elements in todays
WDM optical networks to effectively manage bandwidth capacity as demands increase and change.
The enabling technologies for recongurable wavelength add/drop multiplexers and cross-connects are
electrically controlled optical switches, either broadband or wavelength selective, together with
components known as wavelength multiplexers/demultiplexers. A variety of technologies have been
used for optical switch fabrics, including micro-mechanical (MEM), liquid crystal, and thermo-optical
waveguide switches. Integrated modules that include wavelength demultiplex/demultiplex (demux/
mux) together with optical space switches are also commercially available. Commercial wavelength
recongurable optical networks have been widely deployed at both national and metropolitan levels.
Integration, both monolithic and hybrid, has been important to cost effectively achieve the functional
complexity required for modern optical networks.
An important advantage of optical networks is the potential to upgrade the bit rate per wavelength
without the need to deploy new optical networking elements. The inherent bit rate independence of
optical ampliers (other than the possible need for higher pump power), optical switch fabrics, and mux/
demux elements has allowed carriers to upgrade properly designed recongurable optical networks,
initially operating at 10 Gb/s, to 40 Gb/s and 100 Gb/s by changing out only the ingress transmitters and
egress receiversa signicant advantage of optical networks. Express wavelengths can now be carried
cross continent without going through costly electronic regenerators, while along the way trafc can be
optically dropped and added to fully utilize the high-bandwidth-ber pipe.
At the time of this writing, commercial recongurable optical networks available and deployed for
national and metro applications have capacities of 10 Tb/s (100 wavelengths at 100 Gb/s) with fully
recongurable wavelength add/drop capability. Transoceanic commercial systems are operating at
capacities of 4 Tb/s.
Acknowledgements
In this short historic overview, scope and space have not allowed proper citations [2]. My thanks to the
large global communitymany of whom are members of OSAwho have contributed to the
extraordinary progress in optical networks described here.
References
1. Adapted from R. W. Tkach, Scaling optical communications for the next decade and beyond, Bell
Labs Tech. J. 14, 39 (2010).
2. Suggested further reading for recent overview and update: Special issue on the Evolution of Optical
Networks, Proc. IEEE 100(5) (2012).
Integrated Photonics
Radhakrishnan Nagarajan
A
n essay on the history of integrated photonics invariably starts with the seminal paper by
Miller [1]. In 1969 the idea was way ahead of its time, and many of the components
needed to make such an integrated circuit a reality had yet to be invented. Hayashi and
Panishs demonstration of the continuous wave (CW) room temperature operation of a
semiconductor laser, a critical device for the photonic integrated circuit (PIC), was still a year
away [2]. Optical transport, where PICs nd their applications, got its somewhat fortuitous start
in 1970 as well with the report of a low-loss optical ber by the group at Corning [3].
There is always some personal bias in presenting the historical evolution of any technology.
Figure 1 graphically shows one such historical progression of PIC complexity, as measured in the
number of integrated components on a single InP substrate, with time. The details of the devices
and references presented in Fig. 1 are in [4]. InP and its alloys are the material of choice in
fabricating light emitters for optical transport applications. This is due to the low-loss window at
1550 nm and the low-dispersion point at 1300 nm in the standard silica optical ber.
For the rst decade or so after the demonstration of the CW laser in the GaAs system, InP
lasers started to mature. In the mid-1980s there was active work in the area of opto-electronic
integrated circuits (OEICs), where the integration of electronic devices such as HBT (hetero-
junction bipolar transistor) and FET (eld effect transistor) with laser diodes and photodetectors
was pursued. In the late 1980s three-section tunable DBR (distributed Bragg reector) lasers were
introduced. This was also when electro-absorption modulators (EAMs) integrated with distrib-
uted feedback (DFB) lasers were demonstrated. The trend continued with more complicated (four
and ve section) tunable laser sources that were also integrated with an EAM or a semiconductor
optical amplier (SOA). The next step was the demonstration of the arrayed waveguide grating
(AWG) or PHASAR (phased array) router integrated with photodetectors for multi-channel
receivers or with gain regions and EAM for multi-frequency lasers and multi-channel modulated
sources. One of the most complex PICs reported in the last century was a four-channel optical
cross-connect integrating 2 AWGs with 16 MZI (MachZehnder interferometer) switches. At
this stage the most sophisticated laboratory devices still had component counts below 20 while
those in the eld had component counts of about 4.
The trend in low-level photonic integration continued into the 2000s with one of the larger
chips reported being a 32-channel WDM channel selector. In 2003, ThreeFive Photonics
reported a 40-channel WDM monitor chip, integrating 9 AWGs with 40 detectors. MetroPho-
tonics reported a 44-channel power monitor based with an echelle grating demultiplexer. The
commercial development of both chips was subsequently discontinued. The rst successful
attempt at a commercial large-scale photonic integrated chip (LS-PIC) was made in 2004 when
Innera introduced a 10-channel transmitter, with each channel operating at 10 Gbit/s. This
device with an integration count in excess of 50 individual components was the rst LS-PIC
device deployed in the eld to carry live network trafc. This was quickly followed in 2006 by a
40-channel monolithic InP transmitter, each channel operating at 40 Gbit/s, with a total
component count larger than 240, and aggregate data rate of 1.6 Tbit/s. The complementary
40-channel receiver PIC also had an integrated, polarization independent, multi-channel SOA at
the input.
2004, the year when the rst commercial large-scale photonic integrated circuit was
deployed, proved to be a watershed year for silicon photonics as well when Intel demonstrated
293
Fig. 1. Historical trend and timeline for monolithic, photonic integration on InP (without including vertical
cavity InP devices). The vertical scale is linear, and the red lled circles start at 1 and go to 240. The
trend shows an exponential growth in PIC complexity in recent years. Unlike silicon ICs where the transistor
count is a universal metric, there is no unique benchmark for complexity in photonic integration. For this
exercise, we have counted a functional unit (which may be a combination of other optical elements) as a
device. For example, an MZI is counted as 1 and not as 3. Likewise an AWG is counted as 1 irrespective of
the fraction of the PIC real estate it occupies.
the rst gigabit per second optical silicon (Si)-on-insulator (SOI) modulator [5]. Si as a platform for
optical integration dates back to the 1980s [6,7]. In [6] can be found an excellent review of the early
years of Si photonics. Unlike InP, Si has a centro-symmetric crystalline structure and does not exhibit
the linear electro-optic effect that is commonly used for modulating light in InP. Most Si modulators
are based on the carrier plasma effect, change of refractive index with carrier accumulation, or
depletion. Although this is a weak material effect, the capacitor structure, which allows for a large
effective charge transfer, improves the efciency considerably [6]. Although there are reports of
integrated Ge lasers on Si substrates [8], for the most part the light sources for Si photonics are made
of InP and are integrated using hybrid techniques [9].
In Fig. 1 we saw the progression of PIC complexity thru the 2005 timeframe. Although some of
the PICs, such as the switches and CW sources, were modulation format agnostic, for the most part
these operated using OOK (onoff keying). Figure 2 shows the progression of PICs used for
advanced modulation formats such as QPSK (quadrature phase shift keying) used in optical
coherent communication. The details of the devices and references presented in Fig. 2 can be
found in [10].
Coherent optical communication development started in the mid-1980s. After a gap of more than
ten years, in the mid-2000s the eld went through a revival with the availability of high-speed Si ASICs
and advanced digital signal processing algorithms that eliminated the need for ultra-stable optical
sources and analog phase/frequency/polarization tracking of the optical carrier at the receiver. Early
coherent receiver PICs were all single channel. They were designed for binary phase shift keying
(BPSK) modulation format. BPSK is similar to QPSK except that there are no data in the quadrature
component of the signal. A simple, single-stage MZ modulator (MZM) may be used to generate a
BPSK signal. BPSK signals have lower spectral efciency but better noise margin for longer
transmission distances. There were early attempts to integrate a LO (local oscillator) on the receiver
PIC as well. A multi-channel PIC with I/Q MZM integrated with an optical source was reported in
2008. There have been a number of variants on the DQPSK and QPSK (with external LO) receiver
PICs reported since then. The DQPSK PICs also have the polarization components integrated onto the
same substrate. The rst multichannel, dual polarization, QPSK receiver PIC with an integrated LO
per wavelength was reported in 2011. Unlike the rst phase of the history of integrated photonics
discussed in Fig. 1, the evolution of coherent PICs shown in Fig. 2 has devices on both the InP and Si
platforms.
References
1. S. Miller, Integrated optics: an introduction, Bell Syst. Tech J. 48, 20592069 (1969).
2. I. Hayashi, M. Panish, P. Foy, and S. Sumski, Junction lasers which operate continuously at room
temperature, Appl. Phys. Lett. 17, 109111 (1970).
3. I. Kapron, D. Keck, and R. Maurer, Radiation losses in glass optical waveguides, Appl. Phys. Lett. 17,
423425 (1970).
4. R. Nagarajan, M. Kato, J. Pleumeekers, P. Evans, S. Corzine, S. Hurtt, A. Dentai, S. Murthy, M. Missey,
R. Muthiah, R. Salvatore, C. Joyner, R. Schneider, M. Ziari, F. Kish, and D. Welch, InP photonic
integrated circuits, J. Sel. Top. Quantum Electron. 16, 11131125 (2010).
5. D. Samara-Rubio, L. Liao, A. Liu, R. Jones, M. Paniccia, O. Cohen, and D. Rubin, A gigahertz silicon-
on-insulator Mach-Zehnder modulator, in Optical Fiber Communication Conference (Optical Society
of America, 2004), post-deadline paper 15.
6. G. Reed, W. Headley, and C. Png, Silicon photonics: the early years, Proc. SPIE 5730, 596921 (2005).
Background
In the early 1990s there was a good deal of excitement about three-dimensional periodic
structures in which light cannot exist at frequencies within a photonic bandgap (PBG) [1]. Henry
van Driel (Optical Society Fellow, University of Toronto) even compared the atmosphere at a
packed-out Quantum Electronics and Laser Science (QELS) session on PBGs (on the afternoon of
the last day of the conference) to 1969 Woodstock! At that time it occurred to the author that, if
one could create a two-dimensional PBG crystal of microscopic hollow channels in the cladding of
an optical ber, low-loss guidance of light in a hollow core might be possible [2,3]. The challenge
would be to design a suitable structure and not least work out a way of making it (in pioneering
work at Bell Laboratories in the early 1970s, primitive structures with a small number of large
hollow channels had been made, the aim being air-clad glass ber cores [4]). (See Fig. 1.)
Actually, the rst hints that total internal reectionthe workhorse of conventional ber
opticsmight not be the only way to guide light had emerged in 1968 with the little known
theoretical work of Melekhin and Manenkov in the Soviet Union [5], followed by a more detailed
studyagain purely theoreticalby Yariv and Yeh at Caltech in 1976 [6]. Their idea was to
create a cylindrical Bragg stack from concentric tubular layers of alternating high and low
refractive index. Rays of light traveling within a certain range of conical angles would be Bragg
reected back into the core for all azimuthal directions. The trick then was to choose a core
diameter that supports a Mie-like resonance at conical angles where the cylindrical Bragg stack
has a radial stop-band, resulting in a low-loss guided mode (note that such Bragg bers do not
possess a PBG since light is free to propagate azimuthally).
The operating principle of both of these proposals is closely linked to anti-resonant reecting
optical waveguiding (ARROW), in which light is partially conned by a structure of one or more
pairs of anti-resonant layers. Originally proposed by Duguay (AT&T Bell Laboratories) in 1986,
these are essentially FabryPerot cavities operating off resonance so that they reect light
strongly back into the core [7,8]. When the number of such layers becomes large the ARROW
structure begins to resemble a Bragg waveguide; i.e., the anti-resonance condition coincides with
the presence of a radial stop-band [9].
Although solid-core versions of Bragg bers have been produced using modied chemical
vapor deposition (MCVD) (at IRCOM in Limoges, France) [10], for guidance in a hollow core
one is up against the need for the radial stop-band to appear at values of axial refractive index
less than 1. This means that individual layers must be very thin (~0.69 , where is the vacuum
wavelength), enhancing the effects of dopant diffusion during ber drawing and further reducing
the already weak index contrast. Small index contrast also has the drawback that, for good
connement, a large number of periods is needed and the structure must be highly perfect to
avoid leakage through defect states in the cladding layers.
The ideal structure would consist of a series of concentric glass layers with air between
them, but of course this would not hold together mechanically. A possible compromise is to
fabricate a structure of rings held together with thin glass membranes, but the losses so far
reported are quite high [11]. One could think of increasing the index contrast using two solid
297
materials, but here the problems are
extreme for another reason. Pairs of
drawable glasses with compatible melt-
ing and mechanical properties, a large
refractive index difference and high op-
tical transparency are hard to found.
More exotic combinations of chalcogen-
ide and polymer overcome the mechani-
cal problems, offering moderately low
losses even though the absorption is ex-
tremely high in the polymer layers. Nev-
ertheless, the company Omniguide has
achieved 1 dB/m at 10 m in such Bragg
bers [12], which are now used in laser
surgery [13].
Breakthroughs and
Applications Fig. 2. Clive Day working with his three-legged drawing tower
at the Post Ofce Research Laboratories in Martlesham (UK) in
This work led to the discovery of endlessly the 1970s. (Courtesy Dr. Clive Day and the Post Ofce Research
single-mode (ESM) PCF, which, if it guides Centre, Martlesham Health, UK.)
at all, supports only the fundamental guided
mode [14]. There is a story behind the
publication of this result. Submitted to
Optics Letters, the manuscript received
lukewarm or negative reviews and was
initially rejected. Feeling that justice was on
their side, the group appealed to the editor,
Anthony Campillo, who took a look at it
and decided to accept it. Currently (October
2015), with more than 1700 citations, it is
one of the most frequently cited in the eld.
ESM behavior is also a feature of ridge
waveguides formed by etching a thin lm
of dielectric material so as to produce a
raised strip, and in fact Kaiser points this
out in his 1974 paper [4]. The reason is
simple: thinner structures support modes
with lower refractive indices, which means
that the fundamental mode of the thicker
ridge will be trapped by the equivalent total Fig. 3. Maryanne Large, Martijn van Eijkelenborg and Alex
internal reection. Compared to planar Argyros drawing polymer PCF at the University of Sydney.
ridge waveguides, however, ESM-PCF is (Photograph by Justin Digweed.)
These techniques have thrown up a large number of useful devices, including long-period gratings,
rocking lters, helical bers, and the remarkable photonic lanterns now used to lter out atmospheric
emission lines in ber-based astronomy [32,33]. Based on all-solid multi-core bers, these devices
perform the astonishing feat of adiabatically channeling each mode of a multi-mode ber into separate
single-mode bers.
Applications of the new ber structures continue to emerge, an obvious highlight being broadband
light sources millions of time brighter than incandescent lamps and extending into the UV, pumped by
Q-switched Nd:YAG microchip lasers or Yb-doped ber lasers at 1-m wavelength. These are now to
be found in many laboratory instruments, including commercial microscopes. New types of sensing,
ber have emerged, some of them reminiscent of the original single-material bers of Kaiser (e.g., the so-
called Mercedes ber [34]). Hollow core PCF has perhaps opened up the greatest number of new
opportunities. For example, it is being employed as a microuidic system for monitoring chemical
reactions, in which guided light is used both to photo-excite and to measure changes in the absorption
spectrum [35]. (See Fig. 6.) Compared to conventional microuidic circuits, the quantity of liquid
required is very small, the long path-length means that very small absorption changes can be detected,
and the high intensity achievable in the narrow core for moderate optical power means that reactions
can be rapidly initiated. PCF is also being used in many other optical sensors, with applications in
environmental detection, biomedical sensing, and structural monitoring.
The unique ability of hollow-core PCF to keep light tightly focused in a single mode in a gas is
creating a revolution in nonlinear optics. For the rst time it is possible to explore ultrafast nonlinear
optics in gases in a system where the dispersion can be tuned by changing the gas pressure and
composition [36]. Raman frequency combs spanning huge ranges of frequency, from the UV to the mid-
IR, can be generated at quite modest power levels [37,38]. Atomic vapors of, e.g., Rb and Cs can be
incorporated into the hollow core, permitting experiments on EIT and few-photon switching [39].
Hollow core also adds a new dimension to the important eld of optical tweezers: the absence of
diffraction means that radiation forces can be employed to transversely trap and continuously propel
dielectric particles over curved paths many meters in length [40].
In Conclusion
The Optical Society, through its conferences and publications (especially Optics Letters and Optics
Express), has played and continues to play a major role in promoting a disruptive technology that,
through delivering orders of magnitude improvement over prior art, seems likely over the next decades
to have an increasing impact in both commercial and scientic research.
References
1. E. Yablonovitch, Photonic band-gap structures, J. Opt. Soc. Am. B 10, 283295 (1993).
2. P. St.J. Russell, Photonic-crystal bers, J. Lightwave Tech. 24, 47294749 (2006).
3. P. St.J. Russell, New age ber crystals, IEEE Lasers Electro-Opt. Soc. Newsletter 21, 11 (2007). http://
2photonicssociety.org/newsletters/oct7/21leos05.pdf.
T
he eld of femtosecond lasers was in a difcult state in January 1984. Lasers that
generated pulses of 100 fs or less in duration were few and far between, but there were a
growing number of research applications they could be applied to. For example, at Bell
Laboratories in Holmdel, New Jersey, David A. B. Miller and Daniel S. Chemla were very
interested in studying the excitonic nonlinear optical response and electro-optic properties of
GaAs-based quantum wells, which were rather new back then. The author was a post-doc with
that group and was able to take advantage of the magnicent femtosecond laser labs that had
been developed by Richard L. Fork and Charles V. Shank to work on the generation of infrared
femtosecond pulses, which were perfect to use to study the dynamics of GaAs-based quantum
wells. A few years before, Chuck Shanks group had developed the rst colliding pulse mode-
locked laser that reliably gave pulses of great stability and always shorter than 100 fs around
625 nm wavelength [1]. They had built a multi-stage dye-cell amplier system pumped by a
frequency-doubled Q-switched Nd:YAG laser at 10 Hz rate, producing millijoule pulse energies
that were more than intense enough to generate a beautiful white-light continuum. Pumped by an
argon laser with a few watts of green light, the dye laser produced average powers of a few tens of
milliwatts in a train of femtosecond pulses as long as the dye jets were behaving well. Bad
behavior included clogging, popping hoses squirting dye all over the lab. And, of course, the dye
would eventually turn bad and have to be changed. So, given that the laser was generating only
one color of light in the visible at low power and was running on 40 kW of electrical line power,
while using ve gallons per minute of chilled water, it was very difcult to imagine how such a
laser technology could be useful in the world someday.
The development of Ti:sapphire lasers by Peter Moulton while at MIT Lincoln Labs and
subsequent demonstration of Kerr-lens mode locking by Wilson Sibbetts group [2, 3] were a
tremendous advance for the eld, offering much higher powers and near-infrared tunability as
well. Chirped-pulse amplication, by Gerard Mourous group at the University of Rochester in
1985 [4], led to widely scalable oscillator-amplier systems of great variety and complexity.
Simultaneously, development of erbium and then later ytterbium ber gain media together
with the development of cheap high-power laser diode pump sources were driven strongly by
demand during the telecommunications bubble that peaked in March 2000 with the NASDAQ
briey hitting 5000. Combining these advances in solid-state as well as ber technologies now
has made possible a new generation of practical ultrafast compact laser sources that are
offered by more than 30 commercial suppliers, many of which are still in search of their
killer application. Figure 1 shows the state of the ultrafast-laser eld in 1995, plotting
shortest pulse width as a function of photon energy. We can see that the attosecond short-
wavelength frontier had been identied, but not explored yet, and note the tremendous
advances in that eld have been driven by science and technology developments in many elds
since then.
The Optical Society (OSA) has been at the forefront in promoting ultrafast laser technolo-
gy through its various journals and conferences. In 1995 a CLEO (Conference on Lasers and
Electro-Optics) tutorial entitled Ultrafast Optical Power Supplies was given by the author
[5], which reviewed the progress of the eld and laid out some of the challenges for laser
304
developers. Figure 2 shows an Ultrafast
Catch-22 that seemed to exist then and
still seems to be true today. With the rapid
developments in source technologies and
materials in the late 1990s, it appeared
that it would be possible to develop com-
pact reliable sources of femtosecond
pulses covering a variety of parameter
ranges; however, few commercial applica-
tions had been developed, and therefore
there were few incentives to invest in those
technologies. Figure 3 shows that a wide
range of applications require a wide range
of versatile sources, and no single laser
can satisfy all of them; therefore, individ-
ual unit volumes remain low. In 1996 a
plenary talk was given by the author at Fig. 1. Survey of the ultrafast laser eld in 1994. The short-
CLEO titled Ultrafast Epiphany: The wavelength attosecond frontier had been identied, but not
Rise of Ultrafast Science and Technology explored.
in the Real World [6]. The Epiphany was
that ultrafast lasers could actually be useful for things beyond the obvious ones in high-speed
measurements. This is indeed the most important consideration about the use of ultrafast laser
technology. In some cases, there may be absolute value in the use of ultrafast laser technology. In such
a case, there is simply no other way to carry out a certain application without the use of femtosecond
lasers. Those cases may not be very numerous. But in most of the other cases, there is competing
technology, and then femtosecond laser technology has to offer enhanced value but at a price that is
commensurate with the increased value that it offers. Most ultrafast laser oscillators still cost
$50$150K today, so they need to add a lot of value to justify that expense.
A number of applications for femtosecond technology were predicted by the author in 1995 and
1996; it might be interesting to see how those predictions have come out. The rst known commercial
application of femtosecond technology was coherent phonon generation and detection for multilayer
thin lm metrology, by Rudolph Instruments in New Jersey. For this, an OEM laser source was
developed by Coherent, Inc. In 1995, the author predicted that a high-power chirp-pulse amplied
femtosecond laser would be mounted on a truck and used by the military forces. Today, indeed such a
truck has been developed and sold by Applied Energetics for detection and detonation of IEDs
(improvised explosive devices). The TeraMobile project has taken atmospheric propagation of
femtosecond pulses truly throughout the globe in search of applications. In 1995, the author
predicted that ultrafast electro-optic sam-
pling systems would be commercially
available, and indeed such systems are
available from Ando and others. In
1996, the author predicted that ultrafast
sources would power new generations of
two-photon microscopes, and several
companies now offer these, including
Zeiss/IMRA and BioRad/Spectra-Physics,
but they are not yet widely used in clinical
practice. In 1995, the author predicted
that someday there would be commercial
terahertz radiation spectrometers. Indeed,
this area has advanced tremendously,
Fig. 2. The incentive to invest in development of practical
with commercial systems available from real-world femtosecond lasers comes from the applications.
seventeen companies [7]. Applications for Lasers and applications must be developed in parallel.
C
all it what you will: biomedical optics, biophotonics, optics in the life sciences, or lasers
in medicine; light, lasers, and optics have played a tremendous role in biology and
medicine over the last few decades, and this role is growing. This chapter covers
activities on biomedical optics for in vivo and in vitro applications. Additional material on
biomedical optics can be found in the chapter by Jim Wynne on LASIK.
Optical methods are used in medicine and biology for both diagnostics and therapeutics.
Important aspects of optical methods for these applications include the ability to use multiple
wavelengths to perform spectroscopy (i.e., detect or stimulate specic transitions to provide
molecular information) or to perform multiplexing with multi-color probes, the ability to
penetrate tissue (particularly in the near infrared), the ability to produce changes in molecules,
and the potential to produce low-cost and portable instrumentation.
Clinical use of optical methods has a long history. Early methods relied on the observers eye
for imaging through human tissue, with reports of detection of hydrocephalus (accumulation of
cerebrospinal uid within the cranium, 1831) [1], hydrocele (accumulation of uid around the
testis, 1843) [2], and breast cancer (1929) [3]. The advent of the laser and microelectronics
enabled applications such as retinal surgery using argon lasers in the 1960s [4] and pulse
oximetry in the 1970s [5]. However, the largest growth in biomedical optics methods began in
the 1990s, where advances in lasers, image sensors, and genetic modication led to the advent of
many new biomedical optics methods, among them optical coherence tomography (OCT) [6], in
vivo diffuse optical imaging, multi-photon microscopy [7], revival of coherent anti-Stokes
Raman spectroscopy (CARS) microscopy [8], photoacoustic imaging, bioluminescence imaging
[9], green uorescence protein as a marker for gene expression [10], and bioimaging using
quantum dots [11,12].
Fig. 6. Label-free coherent Raman scattering microscopy showing (a) myelinated neurons in mouse brain,
(b) sebaceous glands in mouse skin, (c), single frame of coherent anti-Stokes Raman movie acquired at 30 Hz, and
(d) image of penetration trans-retinol in the stratum corneum. All scale bars are 25 m [27].
References
1. R. Bright, Diseases of the brain and nervous system, in Reports of Medical Cases Selected with a View
of Illustrating the Symptoms and Cure of Diseases by a Reference to Morbid Anatomy (Longman, Rees,
Orms, Brown and Green, 1831), Vol. II, Case CCV, p. 431.
2. T. B. Curling, Simple hydrocele of the testis, in A Practical Treatise on the Diseases of the Testis and of
the Spermatic Cord and Scrotum (Samuel Highley, 1843), pp. 125181.
3. M. Cutler, Transillumination as an aid in the diagnosis of breast lesions, Surg. Gynecol. Obstet. 48,
721729 (1929).
4. F. A. LEsperance, Jr., An opthalmic argon laser photocoagulation system: design, construction, and
laboratory investigations, Trans. Am. Ophthalmol. Soc. 66, 827904 (1968).
5. J. W. Severinghaus and Y. Honda, History of blood gas analysis. VII. Pulse oximetry, J. Clin. Monit.
3, 135138 (1987).
6. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte,
K. Gregory, C. A. Puliato, and J. G. Fujimoto, Optical coherence tomography, Science 254, 1178
1181 (1991).
7. W. Denk, J. H. Strickler, and W. W. Webb, Two-photon laser scanning uorescence microscopy,
Science 248, 7376 (1990).
8. A. Zumbusch, G. R. Holtom, and X. S. Xie, Three-dimensional vibrational imaging by coherent anti-
Stokes Raman scattering, Phys. Rev. Lett. 82, 41424145 (1999).
9. C. H. Contag, P. R. Contag, J. I. Mullins, S. D. Spilman, D. K. Stevenson, and D. A. Benaron, Photonic
detection of bacterial pathogens in living hosts, Mol. Microbiol. 18, 593603 (1995).
10. M. Chale, Y. Tu, G. Euskirchen, W. W. Ward, and D. C. Prasher, Green uorescent protein as a
marker for gene expression, Science 263, 802805 (1994).
11. W. C. W. Chan and S. Nie, Quantum dot bioconjugates for ultrasensitive nonisotopic detection,
Science 281, 20162018 (1998).
12. M. Bruches, Jr., M. Moronne, P. Gin, S. Weiss, and A. P. Alivisatos, Semiconductor nanocrystals as
uorescent biological labels, Science 281, 20132016 (1998).
13. A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. Elzaiat, Measurement of intraocular distances by
backscattering spectral interferometry, Opt. Commun. 117, 4348 (1995).
14. K. K. Ghosh, L. D. Burns, E. D. Cocker, A. Nimmerjahn, Y. Ziv, A. E. Gamal, and M. J. Schnitzer,
Miniaturized integration of a uorescence microscope, Nat. Methods 8, 871878 (2011).
15. M. D. Duncan, J. Reintjes, and T. J. Manuccia, Scanning coherent anti-Stokes Raman microscope,
Opt. Lett. 7, 350352 (1982).
16. N. G. Horton, K. Wang, D. Kobat, F. Wise, and C. Xu, In vivo three-photon microscopy of subcortical
structures within an intact mouse brain, in 2012 Conference on Lasers and Electro-Optics (CLEO),
OSA Techincal Digest (Optical Society of America, 2012).
17. E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych, J. S. Bonifacino, M. W. Davidson,
J. Lippincott-Schwartz, and H. F. Hess, Imaging intracellular uorescent proteins at nanometer
resolution, Science 313, 16421645 (2006).
18. M. J. Rust, M. Bates, and X. Zhuang, Sub-diffraction-limit imaging by stochastic optical
reconstruction microscopy (STORM), Nat. Methods 3, 793795 (2006).
I
t is a somewhat daunting task to speculate on optical materials for the next century. Before
proceeding, it is perhaps useful to imagine how someone may have tried to write such an
essay 100 years ago. Looking back at volume 1 of the Journal of The Optical Society of
America, discussion of materials was limited to photographic emulsions, metallic lms, and
color lters. Of course, an optical scientist of that time could have had no inkling of the
revolutions that were to follow (lasers, semiconductor electronics, ber optics, to name but a
few) that would transform our concept of optics, give birth to the eld of photonics, and in
many ways redene what we mean by an optical material. Although it is hard to imagine that
the twenty-rst century could be as revolutionary as the twentieth century was for the eld of
optics and photonics, it certain that things will change in ways that we cannot imagine. With
that in mind, this essay focuses on some recent advances in materials that in our opinion are
promising. Whether they will signicantly impact our eld well into the twenty-rst century,
time will tell.
Even in the last few decades, the face of photonic materials research has changed markedly.
Thirty years ago, the eld was dominated by the development of new bulk materials, such as new
IR glasses, nonlinear crystals, or doped laser crystals, while today research in new photonic
materials has more emphasis on advances at the nano or micro scale that can result in materials
with new or enhanced properties. There is also a great deal of research in integration of different
photonic materials for enhanced functionality, resulting in exible photonic platforms, infrared
photonics devices, semiconductor-core bers and integration of III-Vs, organics, or carbon
electronics into silicon electronic platforms. The tremendous growth in the breadth and depth of
the eld of optical materials resulted in The Optical Societys decision to launch a new journal
devoted to the subject, Optical Materials Express, in 2011.
315
dots, also fall into this category, although
in this case the partial connement
results in relatively small modications
to the electronic properties. Nevertheless,
quantum-well materials have already
become the materials of choice for semi-
conductor lasers and are the basis of
the important quantum-well infrared
photodetector (QWIP) devices. Quantum
wires and quantum dots offer the possi-
bility of improved laser and detector
materials, while quantum dots also offer
signicant efciency improvements for
solar cells and for displays. Improve-
Fig. 1. Oblique-view electron micrograph of a woodpile ments in mid-infrared detector materials
photonic-crystal polymer template (black to dark gray) coated with
based upon advances in strained-layer
Al:ZnO (bright gray.) (Frlich and Wegener, Opt. Mater. Express
1(5), 883889 [2011].) superlattice structures and nBn-type
structures are also likely.
Nanoscopic metal particles exhibit
properties markedly different from those of bulk metals. These nanoplasmonic materials [2] have
gained a great deal of attention since the discovery of surface-enhanced Raman scattering (SERS) in the
1970s. Beneting from recent advances in nanofabrication techniques, research in nanoplasmonics has
recently been very successful in using noble metal (especially silver and gold) nanostructures to control
light elds well beyond the limit of diffraction. Such control has already contributed to enhancing light
interaction with tiny amounts of matter down to the single-molecular level. This enhancement, where
the plasmonic particles effectively act as nanoscopic antennas that collect and redirect electromagnetic
elds may nd applications in diverse elds, including infrared detection, solar cells, and nonlinear
optics. Recent work has focused on materials for plasmonics other than silver and gold, including
oxides and nitrides, particularly TiN. Other compounds, alloys, and nanostructured materials are likely
to prove useful for plasmonic applications.
A second category encompasses cases where micro or nano structure provides enhanced function-
ality of known photonic materials, for example, ceramics and advanced polymer composites. Ceramic
fabrication processes provide the properties of crystals with the functionality of amorphous materials,
enabling large parts to be formed that are relatively strain free and have homogenous doping relative to
single crystals in applications where high thermo-mechanical performance and large apertures are
needed. This is leading to improved laser gain media with superior optical quality, with engineered
index and doping proles that make possible diode-pumped solid-state lasers in the 100-kW range.
Similarly, optical ceramics are now offering advantages in applications such as efcient lighting, solar-
energy harvesting, and radiological and nuclear detection. Optical polymer nanocomposites (OPNs),
composites of nanoscopic inorganic particles in a polymer host, have emerged as a promising eld
thanks to advances in optical polymer materials, nanoparticle synthesis, and nanoparticle functionaliza-
tion and dispersion techniques. OPNs have the potential to fulll a broad range of photonic functions
including highly scattering materials for backlighting of liquid crystal displays, narrowband lters,
integrated magneto-optic and electro-optic devices, and optical amplication and lasing.
Third, metamaterials [2] are periodic composite materials of the type shown in Fig. 1 that may have
bulk properties that are very different from the component materials, for example, negative-index
metamaterials. The origins of this eld can be traced back to research in the 1950s on microwave
engineering for antenna beam shaping; articial materials have recently regained a huge interest
triggered by attractive theoretical concepts such as superlensing and invisibility at optical frequencies.
Metamaterials often employ plasmonic nanostructures, providing a close connection between the two
elds. The strong local elds that occur in these materials can be used to strongly modify the nonlinear
properties of the component materials. For example, second and third harmonic generation (SHG and
THG) may be strongly enhanced and nonlinear optical refraction and absorption may be strongly
Advances in Optical
Materials Integration and
Processing
Just as interesting and groundbreaking as the
advances in new materials is the research in inte-
gration of different photonic materials for enhanced Fig. 2. Crystalline-silicon-core optical ber with
functionality, resulting in exible photonic plat- silica cladding. (Ballato et al., Opt. Express 16(23),
forms; infrared photonics systems; semiconductor- 1867518683 [2008].)
core bers; and integration of SiGe, SiC, SiGeC, and
III-Vs and of organics or nanocarbons into silicon
electronic platforms. Additionally, new processing
methods such as direct laser writing are resulting in
new photonic platforms that were not previously
possible.
Infrared materials are notoriously difcult to
process, causing integrated mid-infrared devices to
be extremely challenging to fabricate. Progress in
development of materials for such applications
has slowly evolved to the point where interesting
integrated devices based on chalcogenides are now
being produced [5]. Chalcogenides, being com-
posed of weakly covalently bonded heavy ele-
ments, have bandgaps that are in the visible or
near-infrared region of the spectrum, and low
vibrational energies make them transparent in the Fig. 3. A exible microdisc resonator on polymer
mid-infrared. They can also act as hosts for rare- substrate. (Copyright 2012, Rights Managed by
Nature Publishing Group.)
earth dopants. Advances in processing using CHF3
gas chemistry etching have now resulted in As2S3
rib waveguides with losses as small as 0.35 dB cm 1. Chalcogenide bers, although studied since the
1980s, still have not shown improvement over heavy-metal oxides for mid-infrared transmission,
but as ber draw capabilities improve, many other materials are becoming possibilities for bers in
this wavelength range, for example, the demonstration of a ber with a crystalline silicon core,
shown in Fig. 2. Additionally, developments in photonic-crystal bers, where in some cases most of
the optical mode does not overlap with the material, provides yet more avenues for optical bers for
new wavelength ranges using materials for which implementation in traditional bers would be
impossible. As photonics becomes more pervasive in practical systems, researchers are nding
materials platforms for devices and interconnects to meet industry needs. For example, patterning of
photonic devices on mechanically exible polymer substrates has produced high-quality exible
photonic structures, an example of which is shown in Fig. 3.
Laser processing of traditional materials provides yet another avenue for new platforms for devices
and interconnects, even though the materials themselves are not new. For example, femtosecond direct
laser writing [7] relies on nonequilibrium synthesis and processing of transparent dielectrics with short-
pulse lasers, which open up new ways to create materials and devices that are not currently possible
with established techniques. The main advantage remains in the potential to realize three-dimensional
(3D) multifunctional photonic devices, fabricated in a wide range of transparent materials. This
Summary
Advances in optical materials over the last thirty years have resulted in both evolutionary and
revolutionary advancements of optics, optoelectronics, and photonics. However, this short article
cannot begin to cover the areas that we expect to be impacted by optical materials. Advances in
optical materials have begun to impact biophotonics and biomedicine with promise for improvements
in human health and the treatment of disease [8]. The impact of advanced optical materials on solar
cells is briey discussed above but is not discussed in detail. Advances in manufacturing for
inexpensive solar cell materials including amorphous silicon, materials containing organic dyes,
and nanopatterning may speed their integration into power infrastructure. Work on developing
quarternary and quinternary materials including dilute nitride materials may enhance efciencies in
high-efciency multi-junction solar cells. Advances in optical materials will have a broader impact on
energy consumption and sustainability through development of new, more efcient devices and
applications such as photochromic and electrochromic materials for climate control in buildings and
vehicles. Optical materials, including LCDs and organic light-emitting diodes (OLEDs) have led to a
revolution in display technology. This will likely continue, resulting in even better displays, monitors,
and TVs with brighter colors, blacker blacks, better contrast, better resolution, and wider eld of
view using new OLEDs or organic/inorganic composite LEDs incorporating rare earth and other
materials. Polymer and organic/inorganic systems that enable wearable electronics and optoelec-
tronics, including materials for neuroprosthetics incuding retinal imaging, are likely to become
important. In short, we expect advances in optical materials to pervade almost every aspect of human
life. The future of optical materials is bright.
References
1. Feature issue on Nanocarbon for Photonics and Optoelectronics, Opt. Mater. Express 2(6) (2012).
2. Focus issue on Nanoplasmonics and Metamaterials, Opt. Mater. Express 1(6) (2011).
3. A. Frlich and M. Wegener, Spectroscopic characterization of highly doped ZnO lms grown by
atomic-layer deposition for three-dimensional infrared metamaterials, Opt. Mater. Express 1(5),
883889 (2011).
4. J. Ballato, T. Hawkins, P. Foy, R. Stolen, B. Kokuoz, M. Ellison, C. McMillen, J. Reppert, A. M. Rao,
M. Daw, S. Sharma, R. Shori, O. Stafsudd, R. R. Rice, and D. R. Powers, Silicon optical bre, Opt.
Express 16(23), 1867518683 (2008).
Q
uantum information science (QIS) is a new eld of inquiry, nascent in the 1980s,
founded rmly in the 1990s, exploding in the 2010s, now established as a discipline for
the twenty-rst century.
Born in obscurity, then known as the foundations of quantum mechanics, the eld
began in the 1960s and 1970s with studies of Bell inequalities. These showed that the predictions
of quantum mechanics cannot be squared with the belief, called local realism, that physical
systems have realistic properties whose pre-existing values are revealed by measurements. The
predictions of quantum mechanics for separate systems, correlated in the quantum way that we
now call entanglement, are at odds with any version of local realism. Experiments in the early
1980s demonstrated convincingly that the world comes down on the side of quantum mechanics.
With local realism tossed out the window, it was natural to dream that quantum correlations
could be used for faster-than-light communication, but this speculation was quickly shot down,
and the shooting established the principle that quantum states cannot be copied.
A group consisting of quantum opticians, electrical engineers, and mathematical physicists
spent the 1960s and 1970s studying quantum measurements, getting serious about what can be
measured and how well, going well beyond the description of observables that was (and often
still is) taught in quantum-mechanics courses. This was not an empty exercise: communications
engineers needed a more general description of quantum measurements to describe commu-
nications channels and to assess their performance. These developments led, by the early 1980s,
to a general formulation of quantum dynamics, capable of describing all the state changes
permitted by quantum mechanics, including the dynamics of open quantum systems and the state
transformations associated with the most general measurements. An important advance was a
quantitative understanding of the inability to determine reliably the quantum state of a single
system from measurements.
The 1980s spawned several key ideas. A major discovery was quantum-key distribution,
the ability to distribute secret keys to distant parties. The keys can be used to encode messages
for secure communication between the parties, conventionally called Alice and Bob, with the
security guaranteed by quantum mechanics. In addition, early in the decade, physicists and
computer scientists began musing that the dynamics of quantum systems might be a form of
information processing. Powerful processing it would be, since quantum dynamics is difcult
to simulate, difcult because when many quantum systems interact, the number of probability
amplitudes grows exponentially with the number of systems. Unlike probabilities, one cannot
simulate the evolution of the amplitudes by tracking underlying local realistic properties that
undergo probabilistic transitions: the interference of probability amplitudes forbids; there are
no underlying properties. If quantum systems are naturally doing information processing that
cannot be easily simulated, then perhaps they can be turned to doing information-processing
jobs for us. So David Deutsch suggested in the mid-1980s, and thus was born the quantum
computer.
As the 1990s dawned, two new capabilities emerged. The rst, entanglement-based
quantum-key distribution, relies for security on the failure of local realism, which says that
there is no shared key until Alice and Bob observe it. This turns quantum entanglement and the
320
Fig. 1. (a) Coding circuit for Shor nine-qubit quantum code. An arbitrary superposition of the 0 and 1 (physical)
states of the top qubit is encoded into an identical superposition of the 0 and 1 (logical) states of nine qubits. (b) Error
(syndrome) detection and error-correction circuit for Shor nine-qubit code. Six ancilla qubits are used to detect a bit ip
(exchange of 0 and 1) in any of the nine encoded qubits, and two ancilla qubits are used to detect a relative sign
change between 0 and 1 in any of the nine encoded qubits. Correction operations repair the errors. The code detects
and corrects all single-qubit errors on the encoded qubits and some multi-qubit errors.
associated failure of local realism from curiosities into a tool. The second capability, teleportation, lets
the ubiquitous Alice and Bob, who share prior entanglement, transfer an arbitrary quantum state of a
system at Alices end to a system at Bobs end, at the cost of Alices communicating a small amount of
classical information to Bob. Surprising this is, because the state must be transferred without identifying
it or copying it, both of which are forbidden. Sure enough, the classical bits that Alice sends to Bob bear
no evidence of the states identity, nor is any remnant of the state left at Alices end. The correlations of
pre-shared entanglement provide the magic that makes teleportation work.
These two protocols fed a growing belief that quantum mechanics is a framework describing
information processing in quantum systems. The basic unit of this quantum information, called a qubit,
is any two-level system. The general formulation of quantum dynamics provides the rules for preparing
quantum systems, controlling and manipulating their evolution to perform information-processing
tasks, and reading out the results as classical information.
The mid-1990s brought a revolution, sparked by discoveries of what can be done in principle,
combining with laboratory advances in atomic physics and quantum optics that expanded what can
be done in practice. The rst discovery, from Peter Shor, was an efcient quantum algorithm for
factoring integers, a task for which there is believed to be no efcient classical algorithm. The second
was a proposal from Ignacio Cirac and Peter Zoller for a realistic quantum computer using trapped
ions. This proposal drew on a steady stream of advances that promised the ability to control and
manipulate individual neutral atoms or ions, all the while maintaining quantum coherence, and
applied these to the design of the one- and two-qubit gates necessary for quantum computation. The
third discovery, quantum error correction, was perhaps the most surprising and important nding
about the nature of quantum mechanics since its formulation in the 1920s. Discovered independently
by Peter Shor and Andrew Steane, quantum error correction (Fig. 1) allows a quantum computer to
compute indenitely without error, provided that the occurrence of errors is reduced below a
threshold rate.
Denitely a eld by 2000, QIS galloped into the new millennium, an amalgam of researchers
investigating the foundations of quantum mechanics, quantum opticians and atomic physicists building
on a legacy of quantum coherence in atomic and optical systems, condensed-matter physicists working
on implementing quantum logic in condensed systems, and a leavening of computer scientists bringing
an information-theoretic perspective to all of quantum physics.
QIS researchers are implementing the fundamental processing elements for constructing a
quantum computer in a variety of systems: ions trapped in electromagnetic elds, controlled by
laser pulses and herded to interaction sites by electric elds (Fig. 2); circuit-QED, in which super-
conducting qubits are controlled by microwaves in cavities and transmission lines; neutral atoms
cooled and trapped, interacting via cold collisions or by excitation to Rydberg levels; impurity atoms,
vacancies, and quantum dots in semiconductor or other substrates, controlled electronically or
photonically; and photonic qubits processed through complicated linear-optical interferometers,
capable of implementing efcient quantum computation provided that they are powered by single-
photon sources and the photons can be counted efciently. As experimenters develop these basic
elements for quantum information processing, theorists integrate them into architectures for full-scale
quantum computers, including quantum error correction to suppress the deleterious effects of noise
and of unwanted couplings to the external world that destroy quantum coherence. An active research
effort explores the space of quantum error-correcting codes to nd optimal codes for fault-tolerant
quantum computation.
Other researchers investigate exotic architectures for quantum computation, such as topological
quantum computation, which encodes quantum information in many-body systems in a way that is
naturally resistant to error, obviating or reducing the need for active quantum error correction. A prime
candidate uses as qubits the quasi-particle excitations known as non-Abelian anyons, neither bosons
nor fermions, but occurring naturally in fractional quantum-Hall states. Braiding of the anyons is used
to realize quantum gates.
Experimenters verify the performance of quantum-information-processing devices using quantum-
state and quantum-process tomography, techniques invented by quantum opticians to identify a
quantum state when one can generate the same state over and over again. The inefciency of these
tomographic techniques drives a search for more efcient ways to benchmark the performance of such
devices.
Computer scientists explore the space of quantum algorithms, searching for algorithms that
perform useful tasks more efciently than can be done on a classical computer and seeking to
understand generally the class of problems for which quantum computers provide an efciency
advantage. One class of problems, present from the beginning of thinking about quantum computers,
O
ver the next century it seems likely that glass optical bers, in many as-yet-uninvented
forms, will continue to penetrate more and more deeply into science, technology,
engineering and their applications.
Ultra-Low-Loss Fiber
Perhaps there will be hollow-core photonic crystal bers, with specially treated ultra-smooth
internal surfaces, that offer transmission losses of 0.001 dB/km in the mid-infrared. Such ultra-
low loss will allow extremely long repeaterless communications spans (perhaps more than
20,000 km) and greatly simplify long-haul communications by rendering the ubiquitous Er-
doped ber amplier, with its thirst for expensive pump lasers, largely redundant. All the world's
oceans may then be spanned by single continuous lengths of such ber: Sydney to Los Angeles,
Auckland to Lima, or Sao Paolo to London. The resulting greatly reduced cost of long-haul
communications will make access to the World Wide Web a realistic and cost-friendly possibility
for all the world's populations. Of course, this may also entail the development of a range of new
sources, modulators, and detectors for the mid-infrared, but semiconductor science and tech-
nology will certainly meet this challenge.
The extremely low loss of these bers and the lack of optical damage in the empty core might
also allow them to be used in power distribution systems. They will thus replace old-fashioned
electrical power lines, which will vanish from the landscape in many countries, replaced by
underground ber optical power cables carrying light generated by the highly efcient laser
power stations of the future. These high-power bers will be so ultra-lightweight (a 100 km
length with the newest high-strength carbon ber coatings will weigh only 10 kg and have a
transmission loss of 0.1 dB, i.e., a loss of 1%) that they could be suspended vertically in the
atmosphere using computer-controlled balloons placed at regular intervals. Spiraling up into the
sky, they will deliver megawatts of optical power to the Earth's surface from Sun- or fusion-
driven lasers in space.
Domestic power outlets of the future may also be based on light, delivered via low-loss
optical bers. Such a power socket might consist of a low-loss optical ber that, when a plug is
inserted, sends a signal to a computer-controlled network specifying the amount of power
required. Fiber power delivery to remote devices, using highly efcient laser diodes, will have
become ubiquitous, providing an elegant and cost-effective replacement for awkward and often-
unreliable electrical supply cables and batteries.
Sensing Systems
In an exotic sensor system of the future, a small sensing particle is picked up using laser
tweezers and propelled into a length (which might be kilometers long) of hollow-core optical
ber. Enclosed and protected by the glass sheath, the particle can be propelled along a exible
path even through harsh environments. It can be held stationary or moved backward and
forward by varying the power ratio between counterpropagating optical modes, and its position
327
monitored using time-domain reectometry or (to interferometric precision) using laser Doppler
velocimetry. It can also be optically addressed in many different ways, permitting sensitive measure-
ments of external parameters with high spatial resolution. A further exotic particle type, made possible
by future advances in semiconductor nanofabrication, is a micrometer-scale optoelectronic microbot
that is powered by the propelling light and capable of sending signals back to the ber input using light
of a different wavelength or perhaps via a radio signal. It will be designed to sense many different
physical quantities, including acting as a small microphone for detecting vibrations in inaccessible or
harsh environments, as a point source for illumination or probing, as a light detector, or as a probe for
local oscillating electric or magnetic elds. Perhaps the microbot could, by varying its orientation (if
non-spherical) or its reection coefcients against the incoming light, swim freely to and fro in the
optical eld upon instructions coded into the counterpropagating laser elds.
In the future it may be essential to monitor radiation levels and other parameters close to the core of
a nuclear fusion reactor. Electronics cannot be used and solid-core bers darken rapidly upon exposure
to high levels of radiation. Flying particle sensors in hollow-core bers will provide a solution: light
generated by a radioluminescent particle is relayed back to the ber input, providing a direct measure of
radiation level, as well as other parameters.
Medicine
Endoscopy systems of the future will be multi-functional, enabling surgeons to carry out keyhole
diagnosis, treatments, and surgery using a thin exible cable containing a multi-core microstructured
optical ber with many advanced functions built into it. Such a ber will be able to deliver drugs
(perhaps photo-activated for treating all kinds of conditions including invasive cancer) in precise
amounts through a hollow channel, transmit many different wavelengths of light appropriate for
diagnosing the health of tissue, deliver selectable wavelengths of high-power laser light for tissue cutting
and blood coagulation, and produce deep-UV light for killing cancerous cells. Each system is likely to
have as standard a multi-mode ber microscope for high-resolution structured light imaging of tissue
at many different wavelengths. It will also have a built-in distributed electrically controllable transducer
system (with feedback provided by optical bend and twist sensors) that will allow the ber to be twisted,
turned, coiled, and bent at the surgeons command.
So there you have ita future where glass bers will play an ever-increasing role in society and
everyday life. Do some of these applications seem outrageous? Just think what has been achieved over
the past half century in optical ber communications. Maybe they are not outrageous enough : : : .
N
iels Bohr, the great Dane wisely noted, Prediction is very difcult, especially about the
future, while the American philosopher of the twentieth century, Yogi Berra quipped,
You can observe a lot by just watching. To be asked to write seriously about what
we can expect from light-based technologies over the next hundred years is serious foolishness.
With this caveat, here are some predictions of what light will allow us to see and do in that future.
The interferometers of Michelson of 100 years ago are superseded by matter interferometers
that use light as beamsplitters and mirrors to measure the interference of atom matter-waves. The
precision of the MichelsonMorley experiment 100 years ago saw no measurable shift of
distances l /l 3 10-9 parallel and perpendicular to the motion of the Earth. With atom
interferometers, the precision improves by 19 orders of magnitudethe equivalent of measuring
a change in the distance to the nearest star 3 light years away to one millionth of the width of a
human hair. Gravity-wave astronomy becomes a reality, and spacetime distortions due to
quantum uctuations of the vacuum enlarged during the epoch of ination are mapped directly.
Photostable, near-infrared optical probes smaller than the average protein are routinely used
to label and observe the molecular interactions of RNA strands and dozens of proteins
simultaneously with sub-millisecond time resolution. While tissue is relatively transparent at
these wavelengths, light is strongly scattered. Adaptive optics using multi-megapixel arrays and
ultrafast correction methods are used to restore full optical resolution, peering centimeters into
tissue. Voltage-sensitive versions of these probes record the real-time individual ring of billions
of synapses in the human brain. Coupled with full knowledge of the Human Connectome, we
now understand, at the circuit wiring level and at the molecular level, human consciousness and
self-awareness. This understanding has allowed us to signicantly slow the progression of
various forms of dementia.
Optical probes allow us to track the expression levels and location of the full suite of RNA
expression in time and space within individual cells in live tissue. DNA sequencing identication
methods based on optics help us identify many diseases and greatly reduce misdiagnoses. Optical
methods of understanding the genetic mutations that cause many cancers are routinely used to
develop targeted drug therapies and in helping recruit the human immune system to cleanse the
body of oncogenes with minimal side effects.
To handle the stupendous computing needs of the achievements listed above, quantum
computers, quantum simulators, and nanoscale memory are widely used. We use them to
simulate complex systems with sufcient detail to discover improved room temperature super-
conductors. We use this computational prowess to understand how our brains perceive and how
we analyze and respond to stimuli, as well as to perform massive simulations that reliably predict
climate change caused by human-generated greenhouse gas emission.
Solar power is the lowest-cost source of energy in many parts of the world. This energy is
beginning to be distributed across oceans via ultra-high DC voltage lines in undersea cables
capable of moving tens of gigawatts of power greater than 4000 km with less than 5% loss.
Regions of the world with poor solar irradiation and reduced winter solar generation are
supplied with clean energy.
Unfortunately, the integrated carbon emission by 2065 was not reduced quickly enough.
With our deeper understanding of climate change, the errors of not heeding early warning
signs are starkly seen. The advanced visible and infrared Earth monitoring sensors and orbiting
329
atom-wave gravity gradiometers allow us to measure with remarkable precision how the climate is
changing. The demonstration of reliable long-term weather predictions allows us to forecast with
condence the climate of 2100 and 2200. Just as exposure to carcinogens such as asbestos or cigarette
smoke can trigger a series of multiple mutations that lead to cancer many decades later, we now realize
that greenhouse gas emissions put our world on an extremely disruptive and destructive course for a
signicant fraction of the population.
Is this last prediction too dire? Possibly, but I also believe there is hope. While science alone will not
change political policy, the massive use of optical technologies will provide compelling evidence (and
compelling predictions) to convince a vast majority of people and governments of the world to make the
necessary investments for future generations. In addition, the near future is ripe with the promise of
understanding the human brain and body at breathtaking new levels, again with optics-enabled
technologies. These advances will not only lead to better health and longer and better life spans. With
our optics-enhanced ability see the future, we will likely observe that global altruism and compassion
will serve our own self-interest exceedingly well.
Of course, what happens beyond 50 years is very difcult to predict. The rst power ight by the
Wright Brothers was in 1903, and we landed men on the moon in 1969. All that can be reliably foreseen
is that there will be many wondrous surprises in optics in the next 100 years.
T
he most interesting part of a 100-year future is the last three-quarters of it, following the
arrival of the predictable stuff. Even obvious insights can quickly look sillythink of the
condent predictions of personal airplanes for commuting to work made in the 1930s
and 1940s, while weve managed only bigger highways and longer-lasting trafc jams since then.
Meanwhile, entire generations of music playing systems arrived unpredicted, became universally
adopted, and are already forgotten. How many futurists imagined xerography, or personal
computers, or intelligent telephones that are also cameras and computers, to say nothing of the
FANG teamFacebook, Amazon, Netix, and Google?
What we need is an unconstrained view of the future of optics, and Quantum Optics is nearly
ideal for this because we think we know what it is, but its still far from fully explored. The
meaning of quantum mechanics itself is steadily debated while more and more optical processes
are being given quantum properties. On the near horizon, and easy to connect to current research
themes, one expects to see and possibly benet from optical control of cars and roadways,
photon counting without photon annihilation, wide uses for optical entanglement both quantum
and classical, quantum optical networks for secure identity hacking, the development of
powerful sources of squeezed light, 4-photon down-conversion crystals and quantum-commu-
nicating telescope arrays, in addition to inexpensive consumer items such as invisibility cloaks
that will t in ladies purses.
Farther out, but inevitable, will be lethal hand-held optical weapons and wide-area satellite
monitoring of their use. Entirely speculative, but more fascinating, will be fundamental
discoveries employing quantum optical sensitivity, including: (i) experimental proof that a
connection between quantum mechanics and gravity cannot exist, (ii) detection of coherent
quantum opto-galactic signals pervading space, (iii) discovery of the origin of quantum
randomness, (iv) prediction of the longest possible electromagnetic wavelength and its detection,
(v) real-time optics for in-vivo whole-body DNA correction, (vi) verication of the macroscopic
limit to quantum superposition, and (vii) reliable quantum-optical disassembly and recovery of
bio-systems, allowing practical teleportation. In the end, all of these projections will turn out to
be too conventional. To reorient a remark attributed to Steve Jobs, and thinking of Marie Curie,
the optical scientist doesnt know what shell be most thrilled to nd until she nds it.
331
THE FUTURE
Future of Energy
Eli Yablonovitch
C
ivilization is presently in the hunter/gatherer mode of energy production. Nonetheless,
the continual drop in cost of solar panels will lead to an agrarian model in which energy
that is harvested from the Sun, optically, will satisfy all of societys needs.
Solar panels are optical. By recognizing the optical physics in solar cells, scientists are, for the
rst time, approaching the theoretical limit of 33.5% efciency from a single bandgap.
At the same time, solar panels have dropped in price by a factor of approximately three times
per decade, for the last four decades, cumulatively a 100-fold reduction in real price. Since solar
panels are manufactured in factories under controlled conditions where continuous improvement
is possible, these panels will continue to drop in price until solar electricity becomes the cheapest
form of primary energy (likely to occur around 2030). At that point, solar electricity will become
cheap enough to be converted into fuels, which can be stored summer to winter. The creation of
fuel requires panels that are three to four times cheaper than todays already depressed solar
panel cost, while maintaining the highest efciency.
The highly successful petroleum industry is over 150 years old. It has taken advantage of
technology, but it appears resistant to disruptive technical changes that could sweep it away, as
so many industries have been irrevocably changed or entirely eliminated by the advance of
technology. Nonetheless, the application of solar electricity to create fuel could sweep away the
petroleum exploration industry, which the author calls the hunter/gatherer mode.
Future solar cells will all have direct bandgaps, allowing them to be very thin. The cost of the
material elements composing the cell will be small, since a lm as thin as 100 nm can fully absorb
sunlight using light trapping. Even if the chemical elements were to be expensive, there would be
so little material used in such thin photovoltaic lms that the cost would be low. Indeed, there are
methods to produce free-standing, highest-quality, single-crystal thin lms economically.
The key to high performance from a solar cell is external luminescence efciency, an insight
which has produced record open-circuit voltage and power efciency. This has everything to do
with light extraction, in agreement with the mantra a great solar cell needs to also be a great
light emitting diodeagain the application of optics.
Solar electricity in the open eld will be brought to nearby locations where it will be used for
the recycling and electrolysis of CO2 solutions. There have been great strides in electrolysis,
which can produce various proportions of H2, CH4, and higher hydrocarbons as products. The
carboncarbon bond is particularly prized, since such compounds can be readily converted into
diesel fuel and jet fuel. The study of such selective electro-catalytic surfaces is still in its infancy.
Even if only H2 were ever to be produced, there are industrial methods of using H2 to reduce CO2,
and make useful liquid fuels, among many other products.
The ability to create fuels would increase the size of the photovoltaic panel industry at least
tenfold, allowing the adoption of new cell technology, which is better than the current outdated
1950s crystalline silicon solar cell technology.
Thus we see that the application of optical science in making solar cells more efcient and
lower in cost will produce a revolution in mankinds energy source, playing a role analogous to
the agricultural revolution of 10,000 years ago.
332
THE FUTURE
Future of Displays
Byoungho Lee
D
isplays have been created as a way to convey information. From 2D to 3D, display
technology has been evolving to cope with the complexity of the information we try to
deliver. But what comes next? Based on current research progress in the eld, it is
possible to predict that in the next decade we will be reading news from newspaper-like exible
displays with real-time videos (instead of still pictures) and live internet feeds. But if we go even
further and predict what displays are going to be like 100 years from now, we can expect that
displays will substantially affect the way we live.
The year is 2116, and as his windows turn from opaque to transparent, Mark wakes up
feeling the sun in his face. Marks house already knows that he is awake and the coffee is already
brewing. As Mark looks out to an awakening New York, he is presented with the weather
forecast as well as a reminder about his dinner with his girlfriend. While taking his shower, Mark
likes to read the morning news in the shower-box glass door. In the kitchen Mark is distracted by
the football highlights being shown on the table-top display when he gets a call from his mother.
It is a hologram call. She is having trouble with the new robot vacuum cleaner she was given for
Christmas. Mark then activates the 3D interactions mode, and his 3D image appears in his
mothers house where he can show her how to x her problem. Marks smartwatch buzzes,
telling him that he should leave home if he wants to catch the subway on time. He then transfers
the call to his watch and continues to see his mother through his contact lenses. As an architect,
Mark has always struggled to visualize and interact with his creations in three dimensions, so he
is excited to work on his new interactive desk with a built-in volumetric 3D exible transparent
display. To get a better perspective of what a clients structure is going to look like, Mark uses the
virtual reality feature on his contact lenses and walks around the structure xing the last details.
He then invites his boss and clients to his virtual model, where they can look at it together and
talk over details using a 3D virtual reality call. The client is happy, and Mark could not be
happier. He copies the design documents to his foldable transparent screen. Before folding it, he
checks the status of his own house with the display. His house seems a little bit dark. He opens
the curtain with the Internet of Things menu of the display and orders his robot cleaner to clean
the living room. In addition, since he wants to invite his girlfriend to his home after dinner, he
adjusts the temperature of a nice bottle of wine. Now everything is perfect!
Technology development goes faster and faster, and predicting 10 years later often looks
meaningless. However, predicting 100 years later might be easier because a century is plenty of
time to pass through the trial and error stage, and we can expect that what we originally
imagined as a technology will have come true in real life. The whole idea of displaying
information that started from peoples imagination will be implemented, and we might hope
that all the bugs will be worked out in 100 years. Think of a seamless display technology like
perfect, anytime, completely life-like augmented reality, where users see appropriate virtual
images overlapped with real scenes at any time and at any place. 100 years is enough to make that
possible. The only limitation would be the lack of our imagination rather than an incomplete
technology.
333
THE FUTURE
T
he previous century of biomedical optics strongly suggests that our technology and
capability will be much improved in the next 100 years. Today we have articial light
sources emitting thousands to billions of watts that are routinely used to treat children;
photodynamic therapy drugs designed to hit specic molecular targets; reading an individual
persons genetic code using molecular-optical probes; changing brain functions by inserting light-
activated genes into mammals; and reading human brain activity with light, to name just a few
current capabilities.
But what comes next, next, and next? Some doctors, including this author, have been
accused of being often wrong, but never in doubt. With that caveat, what follows is certainly
what will happen during the next 100 years.
Optical diagnostics will improve, miniaturize, proliferate, become mainstream, replace
conventional biopsies, guide medical and surgical therapy in real time, and then be fully
integrated via the extension of what we now call robotics. Optical systems already provide an
unprecedented combination of high-speed imaging, resolution, point-of-care molecular assays,
and minimally invasive access deep inside the body. By 2040, optical diagnostics will be
comparably as different as todays smart phones are from the telephones of 1985an equal
time gap. What will drive this? At the least, cancer detection, surgical guidance, instant diagnosis
of infections including their antibiotic sensitivity, and the need for common lab tests done quickly
on a single drop of blood, probably as a smart phone app. By 2050, user-friendly optical
diagnostics will be nearly everywhere in medicine, surgery, school, public, and home. Data and
decision analysis will be rapid, highly automated, almost free, and simultaneously personal and
widely shared.
Most of our optical treatments using lasers and light-activated drugs aim to destroy some
undesirable target. But light also stimulates, modulates, heals, controls, or creates. By 2065,
the tables will have turnedmost of the therapeutic realm of biomedical optics will be non-
destructive. An early example now is optogenetics. Rhodopsin genes linked to specic promoter
sequences are used to express light-activated action potentials in neuronal systems. The technique
started as a way to study brain function. By 2025, it will provide a cure for blindness from the
genetic disease retinitis pigmentosa. This is just the rst example of a designer optical interface
with our central nervous system. Other examples will hail from the natural and somewhat
enigmatic phenomenon of photobiostimulation, in which light activates mitochondria, the
cellular power plant that produces ATP. Apparently every cell in our bodies has at least one
photoreceptor system, and probably several. During this century, light will be used to activate
much more than transfected neurons, mitochondria, or naturally occurring photosystems. There
will be a steady trend to use light for controlling biological systems. Microscale implanted optical
machines will be developed, powered, and controlled by light. Think, designer tattoos.
Optical technology itself will benet directly and greatly from biology! The rst live-cell laser
was demonstrated only a few years ago. Useful optical components occur in natural organisms,
including waveguides, gain media, energy storage and transfer, charge separation, quantum-level
light detectors at body temperature, and narrow-band emitters. We use a lot of optical devices to
study biology, but the ow of capability between optics and biology is ultimately a two-way
334
street. Can you imagine using optical components that respond to their environment, self-align,
replicate, and/or repair themselves (because they are alive)? This revolution has already started, by
making optical components from natural biomaterials. Some useful optical cyborgs will be around well
before 2115.
The past 100 years has seen a steady trend in optics and electronics, toward smaller and smaller
devices. Enzymes, RNA, and other macromolecules are incredibly agile nanomachines that specically
manipulate other molecules. Combining three current trends of (a) ever-smaller-devices, (b) designer
molecular biology, and (c) near-eld optics, one comes up with diagnostic and therapeutic, nanoscale,
inside-you, optical robots that work in concert with our natural nanomachinery. This will lead to the
design of circulating, biocompatible, harmless, controllable, self-reporting, intervention-capable
cyborgic devices that are the size of your cells or smaller. At the end of this century, such things
will be in clinical trials. It will be impossibleand irrelevantto decide if they are devices, drugs, or
diagnostics. Eventually, even the FDA will stop caring about that.
Energy, global warming, and environmental change are all, at heart, biomedical optics problems.
Evolution came up with photosynthetic algae and forests that are barely 1% energy efcient, yet they
are the only power source for life on the planet (except for a few, very weird organisms). Can we do
better than photosynthesis? A delocalized, efcient, solar-driven, self-repairing, replicating, energy-
generating, non-polluting equivalent of photosynthesis is sorely needed. Like it or not, we have become
shepherds of this world. A century ago, Mark Twain famously quoted a friend : : : everybody
complains about the weather, but nobody does anything about it! A century from now, global
warming may be viewed as an uncontrolled but positive feasibility experimentyes, we can change the
weather! Other global challenges will be faced and attacked using biomedical optical technologies. By
2115, people themselves may have the option of being photosynthetic. What if food were plentiful and
free? What if people were healthy for a very long time? Traditionally, species populations are controlled
by disease, famine, and unfortunately for us, war. Population control is probably going to be an even
bigger issue in 2115. Maybe biomedical optics will help that, somehow.
Finally, there is optical exobiology. Bioscience has been fundamentally limited by looking at life,
well, here. Optical telescopes are the tool that recently allowed us to detect many other planets, orbiting
many other stars. Exobiology is likely to be a robust science by 2115, and surely it will depend on much
better optics. Someone or some team will use optical spectroscopy to probe whats on those planets.
Telescopes now look at a small patch of sky for a small patch of time, with limited spatial and spectral
resolution. Why not look at all of it, all the time, with detecting life in mind? If life is found, bioscience
will take a giant leap forward thanks to optics.
T
he year 2015 was declared by the United Nations to be the International Year of Light
and light-based technologies. The opening ceremonies not only celebrated the present but
also acknowledged the past and hinted at what was in store for the future. In the modern
world, 50 years after the demonstration of the laser, light impacts everything we do from
communicating, to manufacturing, to health care. This is not surprising, because 50 to 100 years
is the adoption cycle of a new technology for widespread use by society. Just think for a moment
about railroads, electrication, air transportation, the national highway system, electromagnetic
communication from the radio, television, and the Internet.
So what about the future of lasers and laser technology? We are now six years into the x-ray-
laser age, and x-ray lasers based on linear accelerators are being constructed around the world.
What will the characteristics and applications of the x-ray laser be 50 years from now? We can
expect that, like the radio and the laser, in 50 years the x-ray laser will be integrated into wide use
by society in applications such as precision medical imaging, protein structure determination,
and coherent transmission of information at rates 105 times higher than with visible light. We can
also expect advances in x-ray power that will allow for controlling matter at the high densities
suitable for small-scale inertial fusion power generation. The eld of x-ray nonlinear interactions
will be extended from x-ray to gamma ray frequencies suitable for probing nuclear energy levels
and for pumping gamma ray lasers.
Laser-driven accelerators will open up a host of applications in the future. Going from
Klystrons to laser-driven accelerators reduces physical device scale by 5 orders of magnitude.
Accelerators could even be made as all-solid-state devices on a wafer scale. For example, a few-
centimeter-long accelerator will generate MeV-energy electrons at a mode-locked laser repetition
rate of 100 MHz and would be ideal for treating patients. Such an accelerator, if tted into a
catheter, would revolutionize radiation medicine. This same technology could enable an all-solid-
state scanning electron microscope of centimeter length that is driven by compact ber lasers.
A 1-m laser accelerator with 1 GeV electrons of 10-attosec duration at a 100-MHz repetition
rate is ideal for driving a free-electron laser (FEL) that operates at x-ray frequencies. The 100-
MHz repetition rate allows the consideration of an FEL laser with a resonator to match the 100-
MHz period. Using, for example, diamond mirrors, this sync-pumped FEL opens the door to
upconverting a comb of modes from the visible to x-ray frequencies. This in turn leads to
opportunities for precision clocks, precision spectroscopy, and attosecond-timing resolution
measurements in the hard-x-ray region, as well as eld strengths adequate to ionize the vacuum.
Imagine the vacuum as the ideal nonlinear medium for future experiments.
High-average-power lasers have opened the door to new applications. As the power level
increases in the future to approach and exceed the 1-MW level, new and surprising applications
are enabled. For example, a laser of 15-MW average power operating at 100 pulses per second,
located on the ground, will enable the launching of satellites into low earth orbit, each with a
mass of greater than one ton. A laser of 35-MW average power operating at a 15-Hz repetition
rate is ideal for driving a laser inertial fusion power plant with a 1-GW electrical output. When
that happens, laser energy will become the carbon-free energy of choice: stars burning under
control on the surface of the earth.
In the future, if laser propulsion were used to launch hundreds of 2-m-diameter telescopes
and the telescopes were directed into formation as a constellation of satellites, then optical
336
telescopes of 1000-m diameter and greater would be possible. How would these mirrors be aligned?
Again the laser offers the solution through the use of precision clocks and precision interferometry to
locate each 2-m mirror to better than 1/100 of an optical wave in spacetime. Such a telescope array
would enable detailed studies of exoplanets using precision spectroscopy based on laser frequency
combs.
It seems appropriate that in 2060, 100 years after the demonstration of the laser, the amazing laser
will continue to serve society across multiple dimensions from energy to manufacturing to health and
the environment.
Optical Communications:
The Next 100 Years
Alan E. Willner
O
ver the past few decades, the eld of optical communications has produced astounding
scientic and engineering feats. In addition, it has helped transform the way society
functions since the Internet as we know it could not exist without it. Given the exciting
nature of optical science and the ubiquity of communications in our world, there is much reason
to hope that this rate of technical progress and impactful applications will continue for many
decades to come.
The following predictions might capture the future of our eld, or just tickle the imagination.
We know that technological advances have made the transmission of enormous amounts of
data across the planet commonplace, with the exponential growth in capacity continuing into the
future. Many past advances in transmission capacity have utilized the multiplexing of multiple
data-carrying optical waves with each beam inhabiting a unique optical parameter, such as is
done with different wavelengths. Although recent research experiments have shown signicant
capacity increases due to space-division multiplexing, we are just scratching the surface. Basic
optical science tells us that the spatial domain has an enormous number of orthogonal spatial
states, and we will nd new ways to exploit space to enable many orders of magnitude
improvement. Whatever the technology, we will have an endless cycle of thinking we have
enough capacity followed by the panic of needing more, followed by innovation. We will be
feverishly following a Moores Law-like growth, and always worried that we are coming to
fundamental physical limitsbut not.
We are always intrigued by the single photon itself. Present single-photon systems are fairly
limited in terms of data rate, transmission distance, complexity, and cost. However, utilizing
future advances in quantum repeaters and high-speed single-photon sources and detectors, we
will be able to control and communicate using single photons for many types of low-power, long-
distance, and secure systems.
It is quite likely that advances in the coming decades in the performance and mass
production of photonic integrated circuits will enable optics to be ubiquitously deployed
wherever and whenever it can bring benet to the system, just as we use electronic integrated
circuits today without thought. Furthermore, optics will bring low-loss and high-bandwidth
connections between and within computer chips. Furthermore, with future advances in highly
nonlinear devices, optics will perform specic signal processing operations and logic functions
alongside electronics to enable higher speed and lower power consumption, such that electronics
and optics will be used in a hybridized and harmonized fashion. In some applications, optics will
not even need electronics to process data.
Optical networks have enabled many users to communicate with each other very efciently.
However, these networks are still made up of discrete nodes, such that data is sent away from one
node and independently received by a different node, without different nodes actually interacting as
a single unit. Indeed, think of a computer chip. It is a brain, with many operations occurring in
parallel but all working toward a single end goal. With advances in highly accurate optical clocks,
networks covering large geographic areas will be designed to act like a large computer brain and
synchronously communicate and process data efciently. Distances will truly disappear.
338
For the past 100 years, radio has been king of the free-space communications world, with optics
barely registering an impact. However, with the constant increase in needed capacity, optical links will
become commonplace. Indeed, with the future ubiquity of solid-state lighting, almost any bulb can be
used for communications.
Sir Charles Kao, the Nobel Laureate credited with proposing that low-loss glass can be used for a
communication system, had said that silica might last 1000 years as the medium of choice. So, going out
on a limb, is it possible that silica ber will give way to a new material with lower loss and lower
nonlinearity? Such materials have been envisioned, and the economics may one day demand that a new
type of ber be adopted and laid around the world.
Since there has been exponential growth in the ber transmission capacity and the demand for that
capacity, our eld is now cemented as being essential for economic and societal growth. For the past few
decades, ber transmission capacity has increased 100 every decade. We have seen names y
byMega, Giga, Tera, and now even Petabits/sec on a single ber. Will this continue? In 100 years and
ifa big if!advances continue at the same pace, we will see words like Exa, Zetta, Zotta, and even
Brontobits/sec (1027 bits/sec). It is thrilling to imagine the enabling technologies and potential
applications for such capacity.
If past is prologue, either the above-mentioned or other transforming advances will occur. If this
happens, the exponential growth in the capacity of communication systems will enhance our ability to
interact with each other, our environment, and machines in unforeseen ways.
Index 341
Bass, Michael, 183, 218, 219 Brackett, Frederick Sumner, 18 Universal Jewel professional folding dry
Bates, Frederick J., 27 Bradbury, Rudolph, 186 plate camera, 35
Battelle Memorial Institute, 57, 6061 Bragg bers, 297, 298 Wide-Field Planetary camera (WF/PC),
Battista, Albert, 102 Brandeis University, 186 250
Baumeister, Philip, 73 Braren, Bodil, 260 See also photography; surveillance
Bausch, John Jacob, 15 Braunstein, Maurice, 187 imaging
Bausch & Lomb, 10, 15, 23, 24, 25, 33, Brazier, Pam, 122, 122 Campillo, Anthony, 299
70, 71, 72, 185, 253, 254, 256 Breckinridge, James, 244, 246, 249, 250 Canon, 62, 63
BBO (beta barium borate), 215 Brewster, David, 11 carbon dioxide lasers, 9293, 92, 102,
BEACON HILL Report, 65 Brewsters angle, 91, 97, 169 124, 150, 163, 167, 168, 186, 187
BEACON HILL Study Group, 64 Brewsters angle slab amplier, 168169, carbon monoxide lasers, 92
Beckman, 50 169 carbon nanotubes, 315
Beecher, William, 15 Bridges, William B., 8893, 91, 98, 187 Carl Zeiss Co., 69
Belforte, David A., 124 Brillouin scattering, 116 Carl Zeiss Foundation, 35
Bell, Earl, 89, 90, 97, 98 British Telecom Research Laboratories, Carl-Zeiss Stiftung, 23
Bell and Howell, 15 197, 278, 280 Carlson, Chester, 50, 57, 58, 5960, 61,
Bell Holmdel Laboratory, 185, 186, 224 Brody, Peter, 270 61, 62, 134
Bell inequalities, 320 Bromberg, Joan Lisa, 103 Carlson, R.O., 108
Bell Telephone Laboratories, 25, 50, 81, Browell, Ed, 175 Carnegie, Andrew, 244, 246
82, 84, 88, 89, 91, 92, 96, 100, 104, Brown, Gordon, 145 carotenoids, 41
105, 114, 116, 177, 185, 186, 196, Brown University, 186 carrier frequency sweep, 215
199, 201, 204, 205, 215, 218, 223, Brownell, Frank A., 32 carrier leakage, 229
227, 232233, 239, 240, 278, 284, Brownie camera, 10, 3132, 32 Carritol, Dick, 154
297, 304 Buccini, John, 143 CARS spectroscopy (coherent anti-Stokes
Bennett, William, 82, 84, 88, 89, 89, 91 Bufton, Jack, 175, 179, 180 Raman spectroscopy), 219, 308
Benton, Steve, 122 Bunsen, Robert, 13 Carswell, Alan, 177
Berg, Howard, 225 BUORL (Boston University Optical Cartwright, Charles Hawley, 69, 70, 71
Bernard, M.G., 107 Research Laboratory), 65 Case Western Reserve University, 244
Berns, Roy S., 10, 43 Burbank Skunk Works, 157 Cataln, Miguel A., 17
beta barium borate (BBO), 215 Bureau of Standards, 20, 24, 25, 26, 27, cataract surgery, 124, 184, 262264, 312
Bevacqua, S.F., 109 43, 185 cathode ray tubes (CRTs), 269
Biacore, 312 Burnham, R.W., 44 Cats Eye Nebula, 18
bifocals, 14, 184, 253, 254, 263, 265, 266, Burns, Gerald, 108, 108 Caves, Carlton, 277, 320
268 Burns, Keivan, 20 CDs (compact discs), 138, 140, 141, 141,
Big Bird, 156, 158 Bush, Vannevar, 27, 28, 28, 29, 185 142
Big Demonstration Laser, 150 Byer, Robert L., 103, 213, 214, 336 cellular control, 312
binary phase shift keying (BPSK) cellulose nitrate lm, 15
modulation, 294295 Central Intelligence Agency (CIA), 55, 65,
binoculars, 14, 24, 71 C 153, 157
bioluminescence, 311 C-camera, 66, 67 Central Laser Facility, 235
BIOMED meeting, 313 C-series contact lenses, 253 Centre National de la Recherche
biomedical optics, 277, 308313, Cabellero, Doris, 72 Scientique, 72
309312, 334335 cadmium selenide, 270 ceramic fabrication processes, 316
See also ophthalmic surgery; vision calcium, spectrum, 18 ceramics, 97, 124, 171, 221, 234, 316
correction Caltech, 86, 297 CERN (European Center for Nuclear
Biomedical Optics Express (journal), 313 Cambridge Research Laboratory, 185, 186 Research), 279
BioRad/Spectra-Physics, 305 camera lenses, 3, 33, 35 CGHs (computer generated holograms),
Birks, Tim, 299, 300 cameras, 10, 15, 3336, 36, 37 145
Bison (Soviet bomber), 65 A-1 camera, 66 CGRO (Compton Gamma Ray
Bissell, Richard, 65, 157, 158 aerial cameras, 24, 25, 66 Observatory), 249
Bjorkholm, John, 224, 225, 226 automatic exposure (AE) control, 36 chalcogenide bers, 317
blackbody radiation, 3, 12 BakerNunn camera, 245 chalcogenides, 317
BlazePhotonics, 300 Barnack camera, 35 Chan, Kin Pui, 180
Blikken, Wendell, 119 Brownie camera, 3132, 32 Chance, Britton, 309, 309
Block, Steven, 225 C-camera, 66, 67 Chandra X-ray Observatory, 249, 251,
Blodgett, Katharine, 70 Contax I, 33, 3536, 36 251
Bloembergen, Nicolaas, 115, 115, 213, Deckrullo focal plane shutter cameras, Chanin, Marie, 177178
214, 221 35 Charles Stark Draper Laboratory, 86
Bloom, Arnold, 89, 90, 97 Faint Object Camera (FOC), 250 Charman, William, 265
Blu-Ray, 142, 142 Fairchild K-19, 66 Chebotayev, V., 220
Blum, Samuel E., 257, 258, 259, 260, 261 lm, 10, 15, 34, 39, 51, 52 chemical elements, 19, 21
Boeing, 150, 151, 185 folding Pocket Kodak (FPK), chemical lasers, 93, 150, 151
Bohr, Niels, 3, 13, 17, 329 3233, 32 chemical oxygen-iodine laser (COIL), 151
Bohr atom, 3, 13, 17 Homos stereo camera, 34 Chemla, Daniel S., 304
Boll, Franz Christian, 41 K-19 camera, 66 Chen, Chuangtian, 214, 215
BOMEX project, 98 Kodak Retina camera, 36 Chiao, Ray, 116
Bond, W.L., 6 Leicas, 3335, 35 chirp, 117, 171, 215, 224, 238, 288
Borde, C., 220 lenses, 3, 33, 35 chirp-pulse amplied femtosecond lasers,
Bortfeld, Dave, 95 Miroex reex camera, 35 305
BoseEinstein condensation, 219, 221, Model A, 35 chirped-pulse amplication, 235, 242, 304
225 Pocket Kodak, 32 Chraplyvy, Andrew, 211
Boston University, 85 Polaroid process, 49, 52, 158 chromatic dispersion, 211, 280, 283, 288
Boston University Optical Research Polaroid SX70 camera, 64 chromophores, 309
Laboratory (BUORL), 65 reconnaissance cameras, 6467 Chu, Steven, 220, 221, 221, 222, 224, 225,
Botez, Dan, 227, 231, 231 Schmidt camera, 4, 244, 245 329
Bowen, Ira S., 18, 18, 20, 245 Simplex camera, 34 Churchill, Winston, 205
Boyd, Robert, 117 Super Kodak Six-20, 3637, 36 CIA (Central Intelligence Agency), 55, 65,
BPSK modulation (binary phase shift 35-mm precision cameras, 34 153, 157
keying modulation), 294295 Tourist Multiple camera, 34 CIBA VISION, 254
342 Index
CIE system, 43, 44 ber-optic communications, 4, 50, Cross, Lowell, 100
CIECAM02, 44 209210, 210, 227, 230, 278281, Crosswhite, H., 218
CIECAM97s, 44 279, 281 CRTs (cathode ray tubes), 269
Cirac, Ignacio, 321 future trends in, 338339 CRU International, 280
CL-282 (aircraft), 65 Internet, 4, 63, 133, 142, 191, 193, 207, Crystalens, 264
Clark, Alvan, 14 211, 277, 279, 280, 282, 283, 285, CSF, 90, 91
Clark, Harold, 57 286, 287, 333 Cummings, Stuart, 264
Clark-MXR, 306 low-loss bers for, 189193, 190193, Cummins, Herman, 82
CLEO (Conference on Lasers and Electro- 241, 282 Currie, Mal, 88, 89
Optics), 178, 237, 259, 279, 285, modems, 279, 282 Curtiss, Lawrence E., 50, 55, 56
300, 304, 305, 313 optical communications networks, 183, custom wavefront-guided laser
climate change, 329330 186, 189, 193, 195, 199, 205, 209 refractive surgery, 261
coatings 211, 215, 237, 277, 289292, 289, Cyclops laser, 168
anti-reection coatings, 3, 49, 69, 70, 290, 338
7172, 73, 266
interference coatings, 6870
quantum communications, 323
telecommunications industry, 282286 D
mirrors, 68, 69, 187, 245, 329 telephony, 26, 203207, 204, 206, 207, Dagor lens, 33
optical coatings, 3, 6873, 142 279, 282 Daguerre, Louis Jacques Mand, 31
COBE space telescope, 252 terabit-per-second ber, 209211, 210 Dlibard, J., 221
Coble, Robert, 234 World Wide Web, 279, 282 Danalens, 255, 255
Code, Art, 249 compact discs (CDs), 138, 140, 141, 141, Danielmeyer, H.G., 105
Cohen-Tannoudji, C., 221, 221, 225 142 dark-eld microscope, 312
coherent anti-Stokes Raman (CARS) Compton, Karl, 27 dark-line defects (DLDs), 203
spectroscopy, 219, 308 Compton Gamma Ray Observatory DARPA (Defense Advanced Research
coherent anti-Stokes Raman spectroscopy (CGRO), 249 Projects Agency), 29, 151, 282
(CARS) microscopy, 308 computer generated holograms (CGHs), DAsaro, Art, 201
Coherent, Inc., 102, 305 145 DAST (4-dimethylamino-N-methyl-4-
Coherent Laser Radar Conference, 178 computers stilbazolium), 215
coherent lidar, 178 personal computers, 135, 141, 279, data transmission, 196, 215, 279280,
coherent light, 79, 88, 98, 107, 108, 114, 282, 331 279
119, 213, 214 quantum computers, 320323, 329 Daukantas, Patricia, 9, 10, 17, 38
coherent optical communication, 210, Conant, James B., 27 Davidson, Gil, 179
211, 294295, 295 condensed-matter physics, 3, 206, 323 Davis, Doug, 177
coherent phonons, 305 Conference on Electron Device Research, Day, Clive, 298, 299
coherent population trapping, 217 88 Day, D.A., 157
Coherent Technologies, 178 Conference on Laser Radar Studies of the Dayton, Russell, 60
coherent Raman microscopy, 310, 311 Atmosphere, 178 DCFs (dispersion-compensation bers),
COIL (chemical oxygen-iodine Conference on Lasers and Electro-Optics 211
laser), 151 (CLEO), 178, 237, 259, 279, 285, death ray, 149
Cold War, 4950, 52, 85, 116, 151, 156, 300, 304, 305, 313 Deckrullo focal plane shutter cameras, 35
157, 164, 199 Conrady, Alexander Eugen, 33 Defense Advanced Research Projects
See also spy satellites; surveillance contact lenses, 183, 184, 253256, 254, Agency (DARPA), 29, 151, 282
imaging 255, 260, 262, 333 Dehmelt, H.G., 220
Colladon, Daniel, 53 Contax I camera, 33, 3536, 36 Deln laser system, 167
colliding-pulse mode-locked (CPM) Contessa Nettel, 36 Delfyett, P.J., 241
geometry, 239240 continuous-stream inkjet, 62 DeLoach, B.C., 207
colliding pulse mode-locked lasers, 304 continuous wave argon-ion laser, 98 Delta laser, 168
color-center lasers, 215, 241, 333 continuous-wave (CW) dye lasers, 9596, DeMaria, Tony, 186, 237
color-matching function data, 43 103, 161 dementia, advances in treatment, 329
color measurements, standardization, 10 continuous-wave femtosecond laser Denisyuk, Yuri, 121
color order system, 43 systems, 239242, 241 Denton, Richard, 70
color photography, 3, 10, 33, 34 continuous-wave (CW) room-temperature Department of Energy (DOE), 29, 164
color printing, 10 diode lasers, 199202, 200, 201 Depot of Charts and Instruments, 26
color science, 4344 continuous wear contact lenses, 254255, Derr, Vernon, 178
color television, 270 256 designer optical interfaces, 334
Colorado State University, 178 Convert, G., 90 Dessauer, John, 57, 6162, 61
colorimetry, 43 Coolidge, William, 24 Desurvive, Emmanuel, 197, 210
Columbia Electronics Research Copernicus mission, 249 Detch, J.L., 91
Laboratory, 86 copiers, 57, 6263, 134 Deutsch, David, 320
Columbia Radiation Laboratory, 85 xerography, 5763, 5861, 134137, Devlin, G.E., 104
Columbia University, 30, 40, 81, 84, 85, 136, 137 Dexheimer, John, 285
86, 197, 219, 261 copper-vapor lasers, 96, 163, 240 DFB (distributed feedback) lasers, 288,
Columbia University Harkness Eye Center, copper-vapor pumped dye lasers, 163, 164 293
260 Cornell, Eric, 221, 225 DHaeens, Irnee, 79, 83
Commissariat a lEnergie Atomique, Cornell University, 186 DIAL system (Differential-Absorption
169, 170 Corning, 189, 190, 191, 199, 267, 277, Lidar system), 175
Committee on Medical Research 278, 280, 284, 300 Dieke, G., 218
(CMR), 27 Corning Glass Works, 24, 245, 248 Dietz, R.E., 105
communications, 327 CORONA program, 52, 65, 79, 153, Differential-Absorption Lidar (DIAL)
bandwidth, 4, 191192, 211, 280, 291 157160, 159 system, 175
coherent optical communication, 210, coronium, 19 diffraction, 69
211, 294295, 295 COSTAR optical system, 250 diffraction grating, 12
continuous-wave (CW) room-tempera- couching, 262 diffuse optical imaging in vivo, 309
ture diode lasers, 199202, 200, 201 Cox, Ian, 253 digital holographic microscopy, 311
data transmission, 196, 215, 279280, Cox, Palmer, 32 digital signal processing (DSP), 211
279 CPM (colliding-pulse mode-locked Digonnet, Michel, 195
erbium-doped ber amplier (EDFA), geometry), 239240 Dill, Frederick H., Jr., 108, 108, 109
195198, 196198, 210, 230, 277, Cross, Lee, 100 4-dimethylamino-N-methyl-4-stilbazolium
280, 281, 288 Cross, Lloyd, 122 (DAST), 215
Index 343
diode-laser bars, 229 EDFA (erbium-doped ber amplier), Exxon Nuclear, 161
diode laser-pumped solid-state lasers, 195198, 196198, 210, 230, 277, Eye in the Sky: The Story of the CORONA
105106 280, 281, 288 Spy Satellite (Day, Logsdon, &
diode lasers, 105 Edison, Thomas, 4, 15, 23, 34, 185 Latell), 157
continuous-wave room-temperature Edln, Bengt, 18 eye surgery. See ophthalmic surgery
diode lasers, 199202, 200, 201 Einstein, Albert, 3, 12, 81, 88 eyeglasses, 10, 11, 1415, 265268, 267
high-power diode lasers, 227231, Eisenhower, Dwight D., 29, 49, 52, 64, 65, astigmatism, 15
228230 85, 148, 157, 158, 185 bifocals, 14, 184, 253, 254, 263, 265,
InGaAsP diode lasers, 197 EIT (electromagnetically induced 266, 268
long-lived diode lasers, 206207 transparency), 217 frames for, 15
mirror damage in, 227 Ektaprint 100 copier, 63 lenses, 184, 265, 266
semiconductor diode lasers, 4, El-Sum, Hussein M.A., 119 for low-vision patients, 267
107111, 199, 209, 210, 240241 electric power polarizing sunglasses, 51
diode-pumped neodymium-slab laser, 151 laser fusion for, 171
diodes, LEDs, 4, 26, 105, 133, 178, 199,
203, 271, 272, 318
solar power, 329, 332
electricity, 11 F
Dirac, Paul, 9 electro-absorption modulators (EAMs), Fabrikant, Valentin, 81
direct-detection lidar, 178 293 FabryPerot resonator, 81
DiscoVision, 138 electromagnetic radiation, 11 Faint Object Camera (FOC), 250
dispersion-compensation bers (DCFs), electromagnetically induced transparency Faint Object Spectrograph (FOS), 250
211 (EIT), 217 Fairchild K-19, 66
display technology, future trends in, 333 electromagnetism, 11 Fano interference, 217
disposable contact lenses, 255 electron microscope, 119, 204, 336 Faraday, Michael, 11
distributed Bragg reector lasers, 293 electron spin, 18 Faris, Gregory, 277, 308
distributed feedback (DFB) lasers, 288, Electronic Array Raster Scanner (EARS), fast ignition target, 171
293 136 Faust, W.L., 92
Dixon, Richard W., 203, 205, 207, 207 electrophotography, 57 Feinbloom, William, 253
DLDs (dark-line defects), 203 See also xerography Fejer, Martin, 213
DNA Electrophotography (Schaffert), 57 FELs (free-electron lasers), 151, 336
genetic modication, 311 Electrotechnical Laboratories (Japan), 167 femtosecond absorption spectroscopy, 180
microarrays, 312 Ellerbrock, V.J., 267 femtosecond direct laser writing, 317318
Dobrowolski, George, 72 ELT (Extremely Large Telescope), 248 femtosecond lasers, 147, 238, 239242,
DOE (Department of Energy), 29, 164 emission lines, 12 241, 304, 305, 306
Dollond, John, 13 Emmett, John, 161, 168 Fenner, G.E., 108, 109
dominant designs, 62, 63 end-pumping, 196 Fergason, James, 270
Donders, F.G., 265 endlessly single-mode (ESM) PCF, 299 fermions, 18, 322
Doppler-free laser spectroscopy, 220 endoscopy, 50, 55, 310 FETs (eld effect transistors), 293
Dorpat Observatory, 14 ber-optic endoscope, 50, 55, 56 ber ampliers, 195196, 288
dot-com boom, 283 future trends in, 328 ber attenuation, 280
double heterojunction lasers, 110, 111, energy, future trends in, 332 ber-based astronomy, 301
201, 201, 227, 228 Energy Star 6, 272 ber-grating compressors, 216
Dover printer, 136, 137, 137 engineering, post-World War II statistics, ber lasers, 241242
Dow Corning, 255 85, 87 ber-optic communications, 4, 50,
drop-on-demand inkjet, 62 Enron, 284, 286 209210, 210, 227, 230, 278281,
DSP (digital signal processing), 211 entangle-based quantum-key distribution, 279, 281
duality of light, 12 320321 ber-optic connectivity, 4
Duguay, Michel, 238, 297 Epson, 270 ber-optic endoscope, 50, 55, 56
Dumke, William P., 107, 108, 108 Epstein, Ivan, 72 ber-optic image scramblers, 55
Dupont, 24 erbium, 280, 304 ber-optic imaging, 5356
Durafforg, G., 107 erbium-doped ber amplier (EDFA), ber-to-the-home, 207
DVDs, 141, 141, 142 195198, 196198, 210, 230, 277, ber-optic amplier (FOA), 195198, 282
Dwight, Herb, 97, 98 280, 281, 288 bers
dye lasers, 95, 304 erbium-doped lasers, 106 Bragg bers, 297, 298
dye sublimation printing, 50 erbium ions, 104, 196 chalcogenide bers, 317
dynamic grating spectroscopy, 219 Ericksen, J.L., 269 dispersion-compensation bers (DCFs),
dynamic light scattering, 312 Ernst Leitz Optical Works, 33, 34 211
dysprosium ions, 104 ESM-PCF, 299 ber structure, 301
Dziedzic, Joe, 225 Essilor, 266 future trends in, 327328
ether, 11 glass bers, 5354, 55, 187, 195, 209,
ether wind, 1112 210, 297, 328
E Ettenberg, Michael, 199 high-power bers, 327
E-Tek Dynamics, 284 European Center for Nuclear Research hollow-core photonic crystal bers,
Ealey, Mark, 247 (CERN), 279 297, 299, 300, 301, 327, 328
EAMs (electro-absorption modulators), European Conferences on Biomedical low-loss bers, 189193, 190193,
293 Optics (ECBO), 313 241, 278, 282
EARS (Electronic Array Raster Scanner), europium ions, 104 Mercedes ber, 301
136 Evans, R.M., 44 microstructured optical bers, 277,
Eastman, George, 10, 15, 25, 31, 33, 185 evaporated dielectric coatings, 70 297301, 298301, 328
Eastman, Jay, 128 excimer laser, 183 multi-core bers, 301, 328
Eastman Dry Plate Co., 31 excimer laser ablation, 260, 306 non-zero dispersion-shifted bers, 280,
Eastman Kodak Co., 23, 25, 31, 36, 44, excimer laser lithography, 4 289
51, 95, 161 excimer laser surgery, 257261, 258, 259, photonic bandgap bers, 277
Eastman Kodak Research Laboratory, 24, 306 photonic crystal ber (PCF), 298299,
25, 27, 3334, 185 excitation curves, 43 299, 317, 327
Ebbers, Bernie, 286 exclusion principle, 18 rod-in-tube bers, 55, 190
ECBO (European Conferences on exobiology, 335 single-mode bers, 5556, 191, 206,
Biomedical Optics), 313 exoplanets, 252 210, 279, 301
ECCE (extracapsular lens extraction), 262 extracapsular lens extraction (ECCE), 262 terabit-per-second ber, 209211, 210
Eckhardt, Gisela, 115 Extremely Large Telescope (ELT), 248 ultra-low-loss bers, 327
344 Index
berscope, 50 GaAs injection laser, 107109 Great Britain, 9
eld effect transistors (FETs), 293 Gabel, Conger, 96 Great Observatories, 249, 252
lm-based photography, 10 Gabor, Dennis, 119, 122 green uorescent protein, 311
lm, photographic. See photographic lm GALEX space telescope, 252 green gap, 271
Fiocco, Giorgio, 175, 176 GAMBIT system, 160 Gregg, David P., 138
FIREX project, 171 Gamble, Susan, 122 Grotrian, Walter, 1819
rst-generation lasers, 205 GaPAs, 109 ground-based telescopes, 244248,
Fisher, A.G., 270 Garbuzov, Dmitry Z., 111 245248
Fizeau interferometer, 144, 144 Gardner, Chet, 178, 179 group velocity dispersion (GVD), 215
ame-emission spectroscopy, 20 Garmire, Elsa, 116, 117 Gschwendtner, Al, 177
ame hydrolysis, 190 gas-dynamic lasers, 9293, 92, 150 GTE Laboratories, 117
ashlamp-pumped picosecond systems, gas lasers, 8893 Guggenheim, H.J., 105
237239 ionized gas lasers, 9091, 91 Guiliano, Connie, 186
ashlamp pumping, 84, 95, 95, 103, 169, GE Hitachi Nuclear Energy, 165 Guinand, Pierre Louis, 13
280 Geffcken, Walter, 70 Guirao, Antonio, 263
owing gas-dynamic carbon dioxide Gekko laser, 169 Gustafson, Ken, 116
lasers, 187 gelatin dry plates, 31 Gustavson, Todd, 10, 31
uorescence correlation spectroscopy, 312 Gemini ampliers, 235, 235 GVD (group velocity dispersion), 215
uorescence microscope, 312 gene chips, 311
uorescence recovery after photobleaching
(FRAP), 311
gene expression, 311
General Electric Co. (GE), 2324, 70, 100, H
uorescent lamp, 4, 271 108, 109, 165, 185, 187, 199 Hagan, David J., 277, 315
uorite, 103 genetics Hale, George Ellery, 244, 245, 245, 246
uorophores, 309, 311 genetic modication, 311 half-integral quantum numbers, 18
FOA (ber-optic amplier), 195198, 282 optogenetics, 334 Hall, Charles, 153
FOC (Faint Object Camera), 250 Geodolite Laser Distance Rangender, 98 Hall, Freeman, 178
folding Pocket Kodak (FPK) camera, 99 Hall, Jan, 300
3233, 32 Georgia Tech, 177 Hall, J.L., 219, 222
Ford Motor Co. Research Laboratory, 115 germania, 191 Hall, John, 97, 147
Ford Scientic Research Center, 177 germanium, 107, 110, 199 Hall, Robert N., 108, 109, 187
Fork, Richard L., 216, 304 Gerry, Ed, 150, 187 Haloid Co., 57, 60, 61
Forster, Don, 88 Geusic, J.E., 104, 105, 186 Hamburg Observatory, 4
Frster resonance energy transfer (FRET), GHRS (Goddard High Resolution Handbook of Physiological Optics
311 Spectrograph), 250 (Helmholtz), 15
Fort Belvoir, 72 Giant Magellan Telescope (GMT), 248 handheld barcode scanners, 131132
FOS (Faint Object Spectrograph), 250 Gilder, George, 284, 285 Hnsch, Theodor, 94, 96, 147, 220, 220,
four-level lasers, 83, 104 Giordmaine, Joe, 114, 116, 117, 186, 214 221, 224, 225, 300
four-wave mixing (FWM), 211, 219 glass Hansell, C.W., 53
Fowler, Alfred, 17 anti-reection coatings, 3, 49, 69, 70, Hardesty, Mike, 178, 180
Foy, P.W., 111 7172, 73, 266 Hardwick, David, 97
FranckCondon principle, 232 optical glass, 13, 23, 24, 33, 35, 101, Hardy, A.C., 43, 44
Frank, F.C., 269 189, 266 Hardy, John, 247
Franken, Peter, 114, 115, 213, 218, 219, photo-thermo-refractive (PTR) glass, Hardy spectrophotometer, 43
246 318 Harris, Stephen E., 186, 214, 216, 217,
Frankford Arsenal, 70 quality for lenses, 13, 14 221
FRAP (uorescence recovery after rare-earth metal-doped glass ber, 210 Harrison, George, 20, 21, 28
photobleaching), 311 glass bers, 5354, 55, 187, 195, 209, 210, Hartline, Haldan Keffer, 40, 40, 41
Fraunhofer, Joseph von, 12, 13 297, 328 Hartman, R.L., 207, 207
free-electron lasers (FELs), 151, 336 glass fusion lasers, 169 HartmannSchack wavefront sensors, 256
free-space solid-state lasers, 242 glass lasers, 101, 104, 150, 166, 167, 168, Harvard College Observatory, 14, 14
Fredericksz, V., 269 186, 237, 238, 239 Harvard great refractor, 14, 14
Freeman, R.R, 224 glass mirrors, 68, 245 Harvard University, 101, 177, 186
frequency combs, 4, 117, 147, 221, 300, GMT (Giant Magellan Telescope), 248 Harvard University Optical Research
301, 337 Goddard, George, 64 Laboratory, 65
frequency-resolved optical gating (FROG), Goddard High Resolution Spectrograph Hasegawa, Akira, 117, 215
238 (GHRS), 250 Hass, Georg, 72
Fresnel, Augustin-Jean, 11, 12, 69 Godowsky, Leopold, Jr., 34 Haus, H.A., 239
FRET (Frster resonance energy transfer), Goethe, J.W. von, 68, 69 Haus, J., 213
311 Goetze, Richard, 17 Haussmann, Carl, 168
Freulich, Rod, 180 Goldman, Jack, 62, 135 Hayashi, Izuo, 111, 201202, 204, 293
FROG (frequency-resolved optical gating), Goldmuntz, Lawrence, 82, 149 HBTs (hetero-junction bipolar transistors),
238 Goldsworthy, Michael, 165 293
fuels, 332 Gordon, E.I., 91 Heaps, Bill, 175, 177
Fuji-Xerox, 63 Gordon, James, 7, 81, 82, 82, 215, 224 Heard, H.G., 92
Fujimoto, James, 238, 240, 309, 310 Goudsmit, Samuel, 18, 83 Heavens, Oliver, 72, 84
Fujitsu Laboratories, 281 Gould, Gordon, 81, 82, 83, 100, 149, 150 Hecht, Jeff, 9, 11, 51, 53, 79, 81, 85, 94,
FULCRUM program, 154 governmental and industrial research 100, 102, 114, 119, 149, 161, 277,
fullerenes, 315 laboratories. See industrial and 278, 282
fused ber bundles, 55, 56 governmental research laboratories Hecht, Selig, 3940, 40
fusion research, with lasers, 166172, governmental funding agencies, 9, Heilmeier, George, 269
167171 185188 Heinz, T., 221
FWM (four-wave mixing), 211, 219 Graham, Clarence H., 40 Heisenberg, Werner, 18, 40
Granit, Ragnar, 41 Helfrich, Wolfgang, 270
Granitsis, George, 101 helium, model for neutral atom, 17
G Grant, Bill, 175 helium-mercury ion laser, 9091, 91
GaAlAs lasers, 197, 203, 240 graphene, 315 helium-neon lasers, 4, 84, 8889, 89, 90,
GaAs-GaAlAs heterostructure graphite, 315 97, 98, 107, 120, 134, 135, 136,
semiconductor lasers, 203, 204 gravity-wave astronomy, 329 138, 190
GaAs homojunction (diode) lasers, 187 Gray, George, 270 Hellwarth, Robert, 115
Index 345
Helmholtz, Hermann, 15 hydrogen, Bohr model, 13 International Quantum Electronics
Henderson, Sammy, 178 hydrogen-uoride chemical lasers, 93 Conference, 82, 162, 234
Heraeus Corp., 155 hydrogen-uoride optical parametric International Symposium on Remote
Hercher, Michael, 116, 186 oscillator, 163 Sensing of Environment, 178
Herriott, Donald, 84, 88, 89, 89 hyper-contrast optical systems, 252 International Telecommunications Union,
Herschel, William, 11, 14 hyperne splitting, 9, 20 193
Herschel space telescope, 252 hyperne structure, 19 Internet, 4, 63, 133, 142, 191, 193, 207,
Herscher, Mike, 96 211, 277, 279, 280, 282, 283, 285,
Hertz, Heinrich, 11, 12 286, 287, 333
hetero-junction bipolar transistors (HBTs), I Internet of Things, 333
293 IBM, 100, 108, 108, 110, 187, 199, 221 intracapsular lens extraction (ICCE), 262
heterodyne Doppler lidar, 177 IBM Watson Research Center, 84, 94, 103, intraocular lenses, 262264
heterodyne interferometry, 143, 247 104, 187, 187, 257 iodoquinine sulfate, 51
Hewlett-Packard, 62, 63, 98, 131 IBM Zurich Laboratories, 228 ionized gas lasers, 9091, 91
Hexagon program, 79, 154, 158, 160 Icaroscope, 114 ionography, 62
Hexagon spy satellite, 153156, 154156, ICCE (intracapsular lens extraction), 262 iPhone 6, 272, 272
158 ICG (indocyanine green), 309 Ippen, Erich P., 96, 216, 232, 239, 240
Heyerdahl, Thor, 98 ICLAS (International Coordination IRAS (Infrared Astronomical Satellite),
Hicks, Will, 55, 56 Group on Laser Atmospheric 251252
hierarchical self-assembly, 318 Studies), 178 IRCOM (France), 297
high-average-power lasers, 336 illumination Iskra laser system, 166
high-power diode lasers, 227231, uorescent lamp, 4, 271 isosulfan blue, 309
228230 incandescent light bulbs, 4, 24 isotope enrichment, 161165, 162164
high-power ber lasers, 106, 126, 198 solid-state lighting, 339 ISP (Intelligence Systems Panel), 65
high-power bers, 327 ILRC (International Laser Radar Itabe, Toshikazu, 179
high-power gas lasers, 4 Conference), 178, 179 ITEK and the CIA (Lewis), 157
High Speed Photometer (HSP), 250 image scramblers, 55 ITEK Corp., 65, 143, 157, 158
Hilbert, Robert S., 157, 158, 159 imaging barcode scanners, 133 ITEK Optical Systems, 247
Hillotype, 33 imaging machines, xerography, 5763, Ito, Hiromasa, 179
HIORP (Hubble Independent Optical 5861, 134137, 136, 137 ITO lm (indium-tin-oxide lm), 269
Review Panel), 250 Inaba, Humio, 180 Ives, Herbert, 69
HiPER project, 171 incandescent bulbs, 4, 24 Izatt, Joe, 313
Hirschowitz, Basil I., 50, 55, 56 indium-tin-oxide (ITO) lm, 269
indocyanine green (ICG), 309
Hitachi Central Research Laboratory, 165,
227 industrial and governmental research J
Hochuli, Urs, 97 laboratories, 9, 2330 JahnTeller splitting, 234
Hockham, G., 189, 209 Infrared Astronomical Satellite (IRAS), James Webb Space Telescope (JWST), 252
hohlraum, 166, 170 251252 Janes, G. Sargent, 161
Holland, Leslie, 72 infrared materials, 317 Janus laser, 168
hollow-core photonic crystal bers, 297, infrared optical microscope, 203204 Javan, Ali, 8283, 84, 88, 89, 89, 107
299, 300, 301, 327, 328 infrared spectroscopy, 3 JDS Uniphase, 284
holmium ions, 104 infrared thin lm, 72 JDSU Corp., 230, 285
holographic interferometry, 144145 InGaAsP diode lasers, 197 Jelalian, Al, 177
holography, 79, 119122 inkjet printers, 50, 62, 63 Jensen, Reed, 161
computer generated holograms inner-quantum number, 17 Jersey Nuclear-Avco Isotopes, 163
(CGHs), 145 InP-based lasers, 206, 293, 294 Jet Propulsion Lab (JPL), 175
phase-shifting interferometric, 145 Institute of Optics (University of Jewett, Frank B., 27
reection holography, 121, 122 Rochester), 25, 33, 54, 134, 143, JHPSSL (Joint High Power Solid State
time-averaged holography, 145 158, 168, 169, 170, 185, 186, 304 Laser), 151, 151
two-wavelength holography, 145 instrumental optics, 9 JILA, 97, 221
Holonyak, N., Jr., 109, 187 integrated photonics, 277, 293295, 294, Jobs, Steve, 140
Homos stereo camera, 34 295 Johnson, A.M., 216
Homer, Howard, 71 Intel, 293 Johnson, Kelly, 49, 65
homodyne interferometry, 143144 Intelligence Systems Panel (ISP), 65 Johnson, L.F., 104, 105, 232233
Hooker, John K., 244 intensity-comparison with standards Johnson, Roy, 149
Hopkins, Harold H., 50, 54 method, 20 Johnson and Johnson, 255
Hopkins, Robert, 143, 158 intensity-modulation direct detection, 210 Johnston, Sean, 122
HRL (Hughes Research Laboratory), 88, interference coatings, 6870 Joint High Power Solid State Laser
91, 98, 100, 103, 115, 185, 186, 187 interference phenomena, 69 (JHPSSL), 151, 151
HSP (High Speed Photometer), 250 interferometers, 9, 12, 69, 70, 143146, Jones, Frank, 71
Hubble, Edwin, 244, 245, 247 144, 247, 293, 322, 329 Journal of Applied Physics, 84
Hubble Independent Optical Review Panel interferometric optical metrology, Journal of Display Technology, 271
(HIORP), 250 143147 Journal of Lightwave Technology,
Hubble Space Telescope (HST), 4, 13, 143, interferometry 291292
184, 247, 249250, 250, 251, 252 heterodyne and homodyne interferom- Journal of the Optical Society of America
Huffaker, Milt, 178 etry, 143144, 247 (JOSA), 20, 38, 40, 41, 44, 56, 69,
Huggins, Margaret, 13 metrology and, 143147 119, 121, 221, 265, 312, 315
Huggins, William, 13 phase-shifting interferometry, 143, 144, JPL (Jet Propulsion Lab), 175
Hughes Aircraft Co., 94, 98, 100, 115 146147 Judd, D.B., 43
Hughes Research Laboratories (HRL), 88, stellar interferometers, 247 JWST (James Webb Space Telescope), 252
91, 98, 100, 103, 115, 185, 186, 187 International Conference on Picosecond
Hull University, 270 Phenomena, 237 K
Hund, Friedrich, 18 International Conference on Ultrafast K-19 camera, 66
Hunter, Max, 151 Phenomena, 237 Kaiser, David, 86
Hunter, R.S., 44 International Coordination Group on Kaminskii, A.A., 105
Huygens, Christiaan, 11 Laser Atmospheric Studies (ICLAS), Kantrowitz, Arthur, 150, 161
Hycon, 67 178 Kao, Charles, 189, 199, 209, 339
Hycon K-38, 66 International Laser Radar Conference Kapany, Narinder, 54, 56, 100
Hyde, Frank, 189 (ILRC), 178, 179 Karrer, Paul, 41
346 Index
Kass, Stanley, 176 Large Optics Demonstration Experiment diode lasers. See diode lasers
Kay, Alan, 136 (LODE), 151 diode-pumped neodymium-slab lasers,
Kazarinov, R.F., 110 large-scale photonic integrated chip 151
Keck, Donald B., 189, 190, 191, 279 (LS-PIC), 293 distributed Bragg reector lasers, 293
Keck Ten-Meter-Diameter Telescope Large Space Telescope (LST), 249 distributed feedback (DFB) lasers, 288,
Project, 248 laser ablation, 92, 257, 258, 260, 306 293
Kelley, Paul, 3, 28, 49, 116, 179 laser-bars, 228 double heterojunction lasers, 110, 111,
Kepler mission, 252 laser-based phase-shifting Fizeau 201, 201, 227, 228
Kepler space telescope, 252 interferometer, 144, 144 dye lasers, 95, 304
Kerr-effect lensing, 236 laser-based spectroscopy, 147, 232 erbium-doped lasers, 106
Kerr-lens mode-locked lasers, 242, 304 laser cooling, 221 excimer lasers, 183, 257261, 258, 259
Kerr nonlinearity, 211 laser diode pump, 304 femtosecond direct laser writing, 317
Kessler Marketing Intelligence, 280 laser diodes, 105, 199, 200, 201, 202, 293, 318
Ketterle, Wolfgang, 221, 225 304, 318, 327 femtosecond lasers, 147, 238, 239242,
Keuffel & Esser Co., 24 laser Doppler velocimetry, 328 241, 304, 305, 306
Keyes, R.J., 105, 108, 109, 187 Laser Focus (magazine), 121, 259 ber lasers, 241242
KH-7 GAMBIT, 153 laser fusion experiments, 101 ashlamp-pumped picosecond systems,
KH-9 Hexagon spy satellite, 153156, laser guide star, 4, 29, 178, 247, 248, 248 237239
154156, 158 Laser Heterodyne Radiometer, 175, 177 owing gas-dynamic carbon dioxide
Khokhlov, Rem V., 116 The Laser in America (Bromberg), 103 lasers, 187
Kidder, Ray, 161, 166 laser in situ keratomileusis (LASIK), 5, four-level laser action, 83, 104
Kiess, C.C, 20 183, 260, 261, 306, 308, 312 free-electron lasers (FELs), 151, 336
Killinger, Dennis K., 175, 177, 178, 179, Laser In Space Technology Experiment free-space solid-state lasers, 242
180 (LITE), 177 frequency comb lasers, 147
Kimmerling, L.C., 107, 110 Laser Inc., 102 fusion research with, 166172
Kinetoscope, 34 laser-induced-breakdown spectroscopy future trends in, 336337
King, Peter, 71 (LIBS), 178 GaAlAs lasers, 197, 203, 240
Kingslake, Rudolf, 33 laser-induced continuum structure, 217 GaAs-GaAlAs heterostructure semi-
Kingsley, Jack, 108, 109 laser-induced uorescence (LIF), 177, 178 conductor lasers, 203, 204
Kirchhoff, Gustav Robert, 1213 Laser Interferometer Gravitational-wave GaAs homojunction (diode)
Kirkpatrick, Paul, 119 Observatory (LIGO), 12 lasers, 187
Kirtland Air Force Base, 248 laser isotope enrichment, 161165, 162164 GaAs injection lasers, 107109
Kiss, Z.J., 104 Laser Megajoule (LMJ) project, 170, 171 gas-dynamic lasers, 9293, 92, 150
Kitt Peak National Observatory, laser oscillation, 90, 91 gas lasers, 8893
245, 246 laser printers, 134137, 136, 137, Gekko lasers, 169
Kleinman, David, 117 183186 glass fusion lasers, 169
KMS Fusion, 101, 168 laser printing, 4, 62 glass lasers, 101, 104, 150, 166, 167,
Knight, Jonathan, 299, 300 laser radar, 175178 168, 186, 237, 238, 239
Knoll, Henry, 253 laser radiation pressure, 223 helium-mercury ion lasers, 9091, 91
Knox, Wayne H., 96, 277, 304, 305, 306 laser spectroscopy, 218219, 221222 helium-neon lasers, 4, 84, 8889, 89,
Knutson, J.W., Jr., 91 laser trapping, 225 90, 97, 98, 107, 120, 134, 135, 136,
Kobayashi, Takao, 179 laser unequal path interferometer (LUPI), 138, 190
Kodachrome lm, 34 143, 144 high-average-power lasers, 336
Kodak AG, 36 laser video disc, 138 high-power diode lasers, 227231,
Kodak camera, 31 laser weapons, 149152 228230
Kodak Co., 61, 158, 185 lasers, 3, 4, 9, 50, 52, 79, 163, 209, 218 high-power ber lasers, 106, 126, 198
Kodak Research Laboratories. See Airborne laser, 151 holography and, 120121
Eastman Kodak Research AlGaAs lasers, 202, 228 hydrogen-uoride chemical lasers, 93
Laboratory Alpha laser, 151 industrial growth, 100
Kodak Retina camera, 36 at American Optical Co., 10, 15, 51, 55, industrial lasers, 124126
Koester, Charles, 101, 186, 195 56, 100, 101102 InGaAsP diode lasers, 197
Kollmorgen, Frederick, 69 applications, 4 InP-based lasers, 206, 293, 294
Kompfner, Rudi, 223 argon-ion lasers, 91, 91, 95, 96, 98, interferometric optical metrology,
Korad Inc., 100 196, 225, 234 143147
Kornei, Otto, 59 Ashura laser system, 167 Iskra laser system, 166
Korolkov, Vladimir I., 111 Asterix laser system, 166 for isotope enrichment, 161165,
Korsch, Dietrich, 250 carbon dioxide lasers, 9293, 92, 102, 162164
Kossel, Walther, 17 124, 150, 163, 167, 168, 186, 187 Joint High Power Solid State Laser
Kowalski, Robert, 134, 135 chemical lasers, 93, 150, 151 (JHPSSL), 151, 151
Krag, W.E., 109 chemical oxygen-iodine lasers (COILs), Kerr-lens mode-locked lasers, 242, 304
Kressel, Henry, 111, 199, 200 151 krypton uoride lasers, 167
Krishnan, K.S., 19 chirp-pulse amplied femtosecond krypton-ion lasers, 91
Kroemer, Herb, 110, 187 lasers, 305 laser-based precision spectroscopy, 147
Krupke, William, 103 colliding pulse mode-locked lasers, 304 laser-induced-breakdown spectroscopy
krypton uoride lasers, 167 color-center lasers, 215, 241, 333 (LIBS), 178
krypton-ion lasers, 91 continuous wave argon-ion lasers, 98 laser isotope enrichment, 161165,
Kubelka, P., 44 continuous-wave (CW) dye lasers, 162164
Kusch, Polykarp, 81 9596, 103, 161 Laser Megajoule (LMJ) project, 170, 171
continuous-wave femtosecond systems, laser unequal path interferometer
239242, 241 (LUPI), 143, 144
L copper-vapor lasers, 96, 163, 240 Ligne dIntgration Laser (LIL), 170
Labuda, E.F., 91 copper-vapor pumped dye lasers, 163, liquid-phase epitaxy (LPE), 200, 200,
Lamb, Willis, 96, 114 164 203, 204, 206
Lamb shift, 220 Cyclops laser, 168 live-cell lasers, 334
Lamm, Heinrich, 50, 5354 Deln laser system, 167 as manufacturing process tool, 124
Land, Edwin, 49, 5152, 64, 65, 158 Delta lasers, 168 materials processing with, 124126
Langmuir, Irving, 24 development, 8184, 8893 matrix-assisted laser desorption/ioniza-
Lankard, Jack, 94 diode laser-pumped solid-state lasers, tion (MALDI), 312
Large, Maryanne, 299 105106
Index 347
medical applications. See medical ultrashort-pulse lasers, 96, 237242, inelastic scattering, 19
applications 239242 particle theory, 11
mercury-ion lasers, 91, 98 vibronic lasers, 233 quantization, 3
Mid-Infrared Advanced Chemical vision correction. See vision correction as trigger for changes in cells, 312
Lasers (MIRACLs), 150, 150 Vulcan lasers, 169 wave nature, 1112
million hour paper, 205 weapons, 149152 waveparticle duality, 12
mirrors, 81, 88, 89, 90, 90, 91, 91, 94, Yb:ber lasers, 242 wave theory, 11, 69
96, 97, 103, 130, 131, 132, 132, ytterbium-doped lasers, 106 light-emitting diodes (LEDs), 4, 26, 105,
143, 200, 202, 235, 336337 yttrium aluminum garnet (YAG) lasers, 133, 178, 199, 203, 271,
molecular gas lasers, 9293, 92 104, 105, 124, 125, 186, 225, 240, 272, 318
Navy ARPA Chemical Laser (NACL), 242, 257, 258, 259, 301, 304 light guiding, 53, 54
150 Zeta lasers, 168 light in ight, 238, 239
neodymium:ber lasers, 241 Laservision, 138 light waveguide, 201
neodymium-glass ber lasers, 187 Lasher, Gordon J., 108, 108 lighting. See illumination
neodymium-glass lasers, 104, 150, 166, LASIK (laser in situ keratomileusis), 5, Ligne dIntgration Laser (LIL), 170
167, 186, 237238 183, 260, 261, 306, 308, 312 LIGO (Laser Interferometer Gravitational-
neodymium-glass rod lasers, 187 lasing without inversion (LWI), 217 wave Observatory), 12
neodymium-YAG lasers, 104, 105, 124, Latell, B., 157 Lincoln Laboratory. See MIT Lincoln
125, 186, 240, 242, 257, 258, 259, lateral inhibition, 40 Laboratory
301, 304 Lawrence, George, 250 linear ion trap, 322
NIF lasers, 170, 171 Lawrence Livermore National Laboratory linear spectroscopy, 218219
Nike laser system, 167 (LLNL), 96, 101, 161, 162, 163, Linkser, Ralph, 258
noble-gas ion lasers, 91 164, 165, 166, 169, 170, 228, 233 Linn, Doug, 100
nonlinear optics and, 114117 Lax, B., 109 Lippmann, Gabriel, 69
Nova lasers, 169, 170 LBO (lithium borate), 215 Lippmann emulsion, 69
Omega lasers, 169 LCD TV, 272 liquid crystal display (LCD), 269272,
Omega Upgrade lasers, 170, 171 LCDs (liquid crystal display), 269272, 271, 272, 318
Phebus lasers, 169 271, 272, 318 liquid-phase epitaxy (LPE), 200, 200, 203,
photolytically pumped iodine lasers, 166 LDs (semiconductor laser diodes), 105, 204, 206
picosecond lasers, 237239 132 Lister, Joseph (son; surgeon), 14
printers, 134137, 136, 137 LDX (Long Distance Xerography), 134 Lister, Joseph Jackson (father), 14
for propulsion, 336337 Lebedev Institute (Russia), 167 LITE (Laser In Space Technology
pulsed argon ion lasers, 91, 98 Lechner, Bernard, 270 Experiment), 177
pulsed dye lasers, 96, 238239 LED lighting, 4 lithium borate (LBO), 215
pumped dye lasers, 95, 163, 164, 177, LEDs (light-emitting diodes), 4, 26, 105, lithography, 4, 50, 318
234 133, 178, 199, 203, 271, 272, 318 live-cell lasers, 334
Q-switching ruby lasers, 94, 115, 116, Lee, Byoungho, 333 LLNL (Lawrence Livermore National
116 Leghorn, Richard, 65, 85 Laboratory), 96, 101, 161, 162, 163,
quantum cascade lasers (QCLs), 176, Lehmann, Otto, 269 164, 165, 166, 169, 170,
178, 318 Leica cameras, 3335, 35 228, 233
quantum-well lasers, 202, 227, 228 Leith, Emmett, 119121, 120, 122 LMJ project (Laser Megajoule project),
radio-frequency coupling, 97 Leitz, Ernest, II, 35 170, 171
rare earth ber lasers, 4 LeMay, Curtis, 149 local realism, 320
remote sensing, laser radar, and lidar, length-of-line method, 20 Lockhart, Luther, 71
175178, 176, 177, 179, 180, 180 lens index, 266 Lockheed, 65
room-temperature GaAs-AlGaAs het- lenses, 13 Lockheed CL-282 (aircraft), 65
erostructure semiconductor lasers, achromatic lens, 13 Lockheed Sunnyvale, 249
203 for cameras, 3, 33, 35 Lockwood, H.F., 202
ruby lasers, 83, 84, 88, 94, 95, 100, contact lenses, 183, 184, 253256, 254, LODE (Large Optics Demonstration
103, 114, 115, 116, 116, 121, 124, 255, 260, 262, 333 Experiment), 151
149, 175, 186, 218, 232, 234 eyeglasses, 184, 265, 266 Logsdon, J.M., 157
semiconductor diode lasers, 4, intraocular lenses, 262264 Lohmann, Adolf, 145
107111, 199, 209, 210, 240241 lens index, 266 Lomb, Adolph, 25
separate connement heterojunction photochromic lenses, 267 Lomb, Henry, 15
quantum well lasers, 202 prism lenses, 265 long-distance telephone, 26
Shiva lasers, 168 lensless photography, 121 Long Distance Xerography
single-stripe lasers, 228, 229, 229, 231 Leonberger, Fred, 300 (LDX), 134
solid-state lasers, 4, 84, 101, 103106, Leslie, F.M., 269 Los Alamos Laboratory, 161162, 163
125, 126, 131, 178, 227, 228, 231, LEsperance, Francis L., 259, 260 low-loss bers, 189193, 190193, 241,
242, 316 Letokhov, V., 220, 224, 225 278, 282
soliton lasers, 241 Lett, P., 221 low-vision patients, 267
at Spectra-Physics, 89, 90, 91, 9799 Leviathan mirror, 14 LPE (liquid-phase epitaxy), 200, 200, 203,
spectroscopy with, 96 Levishin, Vadim L., 114 204, 206
stretched-pulse lasers, 242 Levison, Walter, 158 LS coupling, 18
stripe-geometry lasers, 111, 203 Levy, Richard, 161 LS-PIC (large-scale photonic integrated
Sun-powered lasers, 101 Li, Guifang, 209 chip), 293
in telescopes, 184, 245248, 251, 252 Li, Tingye, 211 LST (Large Space Telescope), 249
10-J Janus lasers, 168 LIBS (laser-induced-breakdown Lubin, Moshe, 168
three-section tunable DBR lasers, 293 spectroscopy), 178 Lucent Technologies, 283285
titanium:sapphire lasers, 234, 235, 236, Lick Telescope, 14 Lumire brothers, 34
242, 304 lidar, 175178 Luna-See project, 175, 176
tunable dye lasers, 4, 9496, 95, 161 Lidar Pancake, 177 Lundegrdh, Henrik, 20
tunable quantum cascade lasers, 176 LIF (laser-induced uorescence), 177, 178 Luo, Fang-Chen, 270
tunable solid state lasers, 105, 232236, LIFE project, 171 LUPI (laser unequal path interferometer),
233235 light 143, 144
types, 4 coherent light, 79, 88, 98, 107, 108, LWI (lasing without inversion), 217
ultrafast-laser technology, 304306, 114, 119, 213, 214 Lyman, John, 161
305, 306 as electromagnetic radiation, 11 Lyman, Theodore, 17
ultrashort lasers, 306 illumination, 4, 24, 271, 339 Lyon, Dean, 71
348 Index
M Mehr and Mahler, 14 mirrors, 151, 155
MacAdam, David, 20, 44 Meinel, Aden, 245, 246, 247, 250 astronomy, 69, 245, 247, 251, 252
MacAdam ellipses, 44 Meinel, Marjorie, 250 coatings, 68, 69, 245, 329
Macenka, Steve, 252 Melekhin, V.N., 297 early history, 68
Macleod, Angus, 68 Mellon Institute (University of Pittsburgh), glass mirrors, 68, 245
Madden, Frank, 158 71 lasers, 81, 88, 89, 90, 90, 91, 91, 94, 96,
magnesium uoride, for anti-reective MEMS (micro-electro-mechanical 97, 103, 130, 131, 132, 132, 143,
coatings, 7071 systems), 310 200, 202, 235, 336337
magnetism, 11 Menyuk, Norman, 179 Leviathan mirror, 14
magneto-optic (M-O) recording, 140 Menzies, Bob, 175, 177 metallic mirrors, 68, 6972
magneto-optical trap (MOT), 220221, Mercedes ber, 301 in telescopes, 245
225 Mercer, G.N., 91 MIT, 86, 96, 116, 175, 186, 206, 220,
Magnuson, Warren, 29 mercury-ion laser, 91, 98 240, 241, 242
Maguire, Mike, 153, 154, 156 metal nanoparticles, 311 MIT Lincoln Laboratory, 109, 116, 175,
Mahler, Joseph, 51 metallic mirrors, 68, 6972 177, 185, 186, 187, 199, 206,
Maiman, Theodore, 52, 73, 79, 8384, 84, metamaterials, 316, 316 233, 304
100, 103, 104, 107, 119, 149, 186, Metcalf, H., 220, 224 MIT Ultrafast Optics Lab, 240
189, 213, 215, 218 metrology, interferometric, 143147 MIT Wavelength Tables (Harrison), 21
Maitenaz, 266 Meyerhof, Otto, 41 MLIS program, 164
Maker, P., 219 Michelson, Albert, 9, 12, 19, 144, 244, M-O (magneto-optic) recording, 140
MALDI (matrix-assisted laser desorption/ 246 mobile display, 272
ionization), 312 Michelson interferometer, 12, 247, 329 mode locking, 147, 186, 237, 238, 239
Malus, Etienne-Louis, 11 MichelsonMorley experiment, 12, 12, mode patterns, 55
Manenkov, A.B., 297 329 Model A camera, 35
Mangus, John, 250 micro-electro-mechanical systems modems, 279, 282
Manhattan Project, 29 (MEMS), 310 modied chemical vapor deposition
Mannes, Leopold, 34 microbots, 328 (MCVD), 297
Martinot-Lagarde, P., 90 microuidics, 301, 311, 318 molecular gas lasers, 9293, 92
Marzocco, B., 262 micromachining, 306 molecular imaging, 308
masers, 50, 79, 81, 82, 83, 85, 103, 107, micrometer-scale optoelectronic molecular laser isotope enrichment, 165
209, 233 microbots, 328 molecular physics, 3
Massoulie, M.J., 107 microscopes, 3, 14, 15, 34, 35, 53, 237, molecular ruler, 311
master-oscillator power amplier 301, 309313, 323 molecular spectroscopy, 1920
(MOPA), 163, 198 atomic force microscopy (AFM), Mollenauer, Linn F., 214, 215, 241
Mather, John, 252 225226 Mooney, Robert, 71
Mathias, L.E.S., 92 coherent anti-Stokes Raman spectros- Mooradian, Aram, 178
matrix-assisted laser desorption/ionization copy (CARS) microscopy, 308 Moore, Duncan, 250
(MALDI), 312 dark-eld microscope, 312 MOPA (master-oscillator power
matrix TFT-LCD, 270 digital holographic microscopy, 311 amplier), 163, 198
Mauna Kea telescope, 4 electron microscope, 119, 204, 336 Morley, Edward, 12
Maurer, Robert, 189190, 191 uorescence microscope, 312 Mosaic Fabrications, 56
Max-Planck Institute for Quantum Optics, infrared optical microscope, 203204 Moscow State University, 116
95, 166, 300 multi-mode ber microscope, 328 Moss, Steven C., 277, 315
Maxwell, James Clerk, 11, 33 multi-photon microscope, 305 MOT (magneto-optical trap), 220221,
Mayburg, Sumner, 107 nonlinear microscope, 308 225
Mayer, Herbert, 72 optical microscopes, 14, 257 motion picture lm, 15
Mayne-Banton, Veronica, 257 phase-shifting interference microscope, motion pictures, 34
MCA, 138 144, 145 Moulton, Peter F., 105, 232, 233, 234, 304
McCone, John, 153 photoacoustic microscope, 310 Mourou, Gerard, 235, 242, 304
McCormick, Pat, 176177, 179 photoactivated localization microscopy movies, 51, 52, 72, 138
McDermid, Stuart, 175 (PALM), 311 Mt. Palomar observatory, 4, 18, 244, 245
McFarland, Bill, 95 stochastic optical reconstruction Mt. Wilson observatory, 18, 244, 247
McFarlane, R.A., 92 microscopy (STORM), 311 multi-core bers, 301, 328
MCI Communications, 278 two-photon microscopes, 305 multi-layer dichroic reector, 202
MCI Worldcom, 285, 286 microstructured optical bers, 277, multi-megapixel arrays, 329
MCVD (modied chemical vapor 297301, 298301, 328 multi-mode ber microscope, 328
deposition), 297 microwave masers, 81, 83 multi-photon microscope, 305
McWhorter, A.L., 109 Mid-Infrared Advanced Chemical Laser Multi Speed Shutter Co., 34
medical applications, 306 (MIRACL), 150, 150 Multi-University Research Initiatives
biomedical optics, 277, 308313, military laboratories, 186 (MURIs), 188
309312, 334335 military optics, 49, 55, 64, 79 multifocal contact lenses, 254, 255
excimer laser ablation, 260, 306 anti-reection coatings, 69 multiplets, 17
excimer laser surgery, 257261, 258, ber-optic image scramblers, 55 Multiplex, 122
259, 306 fused ber bundles, 55, 56 Munk, F., 44
future trends in, 328 laser weapons, 149152 Munsell Value scale, 43
imaging, 50, 309, 328 See also spy satellites; surveillance Murray, Ed, 175
intraocular lenses, 262264 imaging Murray, John, 166
LASIK technique, 5, 183, 260, 261, Millennium Project, 193 Myers, Mark B., 57
306, 308, 312 Miller, David A.B., 304 MZ modulator (MZM), 295
medical instruments, 55, 91 Miller, R.C., 91, 116, 186, 214 Mller Hansen, Holger, 54
photodynamic therapy, 183184, 309, Miller, S., 293
312, 334
photorefractive keratectomy (PRK),
Miller, W.C., 70
Millikan, Robert A., 18 N
260, 261 million hour paper, 205 NACL (Navy ARPA Chemical Laser), 150
radial keratotomy (RK), 259260 miniaturization, 310 Nagarajan, Radha, 277, 293
medical imaging, 50, 309, 328 Minogen, V.G., 224 Nagel, August, 36
medical optics, 4 MIRACL (Mid-Infrared Advanced Nagel Werke, 36
Mees, C.E.K., 25, 26, 33, 34, 244 Chemical Laser), 150, 150 nanocarbon, 315
Meggers, William F., 17, 18, 20, 20, 21 Miroex reex camera, 35 nanocones, 315
Index 349
nanodiamond, 315 Nike laser system, 167 ONR (Ofce of Naval Research), 29, 82,
nanofabrication, 5 Nimitz, Chester, Jr., 153 185
nanoparticles, 309, 312, 316 NIST (National Institute of Standards and OOK (onoff keying), 294
metal nanoparticles, 311 Technology), 26, 177, 221, 225, Operation Paperclip, 72
plasmonic nanoparticles, 315 226, 300 ophthalmic surgery, 306
quantum dots, 312, 315316 nitrogen lasers, 92 biomedical optics, 277, 308313,
semiconductor nanoparticles, 312 Nixon, Richard M., 153 309312, 334335
nanoplasmonic materials, 316 nLight Inc., 230 cataract surgery, 124, 184, 262264,
nanoporation, 312 noble-gas ion lasers, 91 312
nanoscale memory, 329 noble metals, 312, 316 excimer laser ablation, 260, 306
nanoscopic metal particles, 316 Nomura, Akio, 179 excimer laser surgery, 257261, 258,
nanostructuring, 315 non-zero dispersion-shifted bers, 280, 259, 306
nanosurgery, 312 289 intraocular lenses, 262264
nanotubes, 315 nondestructive testing, holographic, 45 LASIK technique, 5, 183, 260, 261,
narrowband interference lters, 70 nonlinear frequency conversion, 4 306, 308, 312
NASA (National Aeronautics and Space nonlinear microscope, 308 photorefractive keratectomy (PRK),
Administration), 29, 175, 176, 177, Nonlinear Optical Properties of 260, 261
249, 250, 252 Materials, 215 radial keratotomy (RK), 259260
NASA Goddard, 175, 177 nonlinear optics, 114117, 183, 213217, ophthalmoscope, 15
NASA Langley, 175, 176 219220, 238 OPNs (optical polymer nanocomposites),
Nasledov, D.N., 108 applied, 213217, 214217 316
Nassau, K., 104 lasers and, 114117 OPO (optical parametric oscillator), 214
NASTRAN program, 154 parametric nonlinear optics, 218 Optech Corp., 177
Nathan, Marshall I., 107, 108, 108, 110 Nonlinear Optics (Bloembergen), 116 optical astronomy, 184, 247, 248, 249,
National Academy of Sciences, 261 nonlinear phenomena, 4 252
National Aeronautics and Space nonlinear refraction, 215 optical bistability, 215
Administration (NASA), 29, 175, nonlinear spectroscopy, 215, 219221 optical ceramics, 316
176, 177, 249, 250, 252 Nordberg, Martin, 189, 190 optical clock transitions, 220
National Bureau of Standards (NBS), 20, Norrby, Sverker, 263 Optical Coating Laboratory Inc., 284
24, 25, 26, 27, 43, 185 Nortel, 284, 286 optical coatings, 3, 6873, 142
National Defense Research Committee Northrop Grumman, 151 anti-reection coatings, 3, 69, 70
(NDRC), 27, 28, 49 Northwestern University, 186 Blu-Ray, 142
National Ignition Facility, 170, 170, 171 Nova laser, 169, 170 computer-aided design, 73
National Institute of Standards and NRL (Naval Research Laboratory), 71, early history, 68
Technology (NIST), 26, 177, 221, 167, 185, 298 optical coherence tomography (OCT), 5,
225, 226, 300 NRO (National Reconnaissance Ofce), 309
National Reconnaissance Ofce 64 optical communications, 183, 186, 189,
(NRO), 64 NSF (National Science Foundation), 29, 193, 195, 199, 205, 209211, 215,
National Science Foundation (NSF), 245 237, 277, 289292, 289, 290, 338
29, 245 NTT, 197 future trends in, 338
Naval Research Laboratory (NRL), 71, nuclear structure, optical spectroscopy, 19 terabit-per-second ber, 209211, 210
167, 185, 298 nuclear technology optical diagnostics, 334
Navy ARPA Chemical Laser (NACL), 150 fusion research with lasers, 166172, optical discs
NCR, 129 167171 history, 138142, 141, 142
NDRC (National Defense Research laser isotope enrichment, 161165, writable and re-writable discs, 139140
Committee), 27, 28, 49 162164 optical exobiology, 335
near-infrared optical probes, 329 Three Mile Island nuclear accident, 164 Optical Fiber Communications
nebulium, 18 null correctors, 143 Conference (OFCC), 211, 283284,
negative-index metamaterials, 316 Nutting, Perley G., 9, 25, 25, 27, 33, 283, 284, 286, 289, 291
Nelson, Herb, 111, 200 38, 39 optical glass, 13, 23, 24, 33, 35, 101, 189,
Nelson, Jerry, 248 266
neodymium-doped calcium tungstate, 104 O optical imaging, in vivo, 308310
neodymium-doped glass ber, 195 O-Series Leica camera, 35, 35 optical instruments, 1315, 23
neodymium-doped optical amplier, 280 OAO (Orbiting Astronomical optical interferometers, 143
neodymium-ber lasers, 241 Observatory), 247, 249 optical Kerr effect, 215
neodymium-glass ber lasers, 187 OBrien, Brian, 24, 54, 55, 114 optical levitation, 223
neodymium-glass lasers, 104, 150, 166, OCT (optical coherence tomography), 5, optical masers, 81
167, 186, 237238 309 optical materials, 315318, 316, 317
neodymium-glass rod, 101 octave frequency combs, 4 Optical Materials Express (journal), 315
neodymium-glass rod lasers, 187 Odlyzko, Andrew, 283 optical microscopes, 14, 257
neodymium ion, 104 OEICs (opto-electronic integrated optical modulation spectroscopy, 219
neodymium-YAG lasers, 104, 105, 124, circuits), 293 optical molasses, 220, 225
125, 186, 240, 242, 257, 258, 259, OFCC (Optical Fiber Communications optical networks, 338
301, 304 Conference), 211, 283284, 283, optical parametric generation, 214
neon sign, 9 284, 286, 289, 291 optical parametric oscillator (OPO), 214
Neugebauer, Gerry, 252 Ofce of Naval Research (ONR), 29, 82, optical phase conjunction, 215
New, G.H.C., 213, 239 185 optical pick-up (OPU), 138
New Ideas Manufacturing, 34 Ofce of Scientic Research and optical polymer nanocomposites (OPNs),
Newhall, S.M., 44 Development (OSRD), 27 316
Newton, Isaac, 68 Offner, Abe, 143 optical pumping, 81
NeXT, 140 OLEDs (organic light-emitting diodes), Optical Research Associates, 157
Next Generation Space Telescope (NGST), 318 The Optical Society (OSA), 17, 19, 20, 25,
252 Omega laser, 169 27, 33, 38, 40, 57, 70, 84, 120,
Ng, Won, 115 Omega Upgrade laser, 170, 171 178, 213, 219, 222, 237, 246,
NGC 6543, 18 Omnifocal lenses, 266 291, 304
NGST (Next Generation Space Telescope), Omniguide, 298 areas of interest, 3
252 onoff keying (OOK), 294 biomedical optics and, 312313
NICMOS system, 250 On the mechanism of the eye color science, 4344
NIF laser, 170, 171 (Young), 14 Committee on Colorimetry, 43
350 Index
Committee on Needs in Optics, 86 Panish, M.B., 111, 200, 201202, 293 motion pictures, 15, 34
membership, 3 Pankove, J.I., 107 movies, 51, 52, 72, 138
The Science of Color, 43 Pappis, Jim, 187 Polaroid process, 49, 52, 158, 186
Uniform Color Scales, 43 parametric nonlinear optics, 218 speckle photography, 145
optical solitons, 4, 25 parametric oscillators, 186 three-dimensional movies, 51
optical spectroscopy, 3, 17, 19, 21, 24, 50, parametric processes, 4 See also cameras; spy satellites
175, 218, 220, 335 Parker, J.T., 92 Photography by laser (Scientic
optical surveillance. See spy satellites; Parks, Bob, 250 American), 121
surveillance imaging Parsons, William, 14 photolithography, 4, 50, 312
optical trapping, 223226, 224, 311, 313 particle theory of light, 11 photolytically pumped iodine laser, 166
optical tweezers, 222, 225, 226, 301, 311, particle tracking, 312 photometry, 43
327 Paschen, Friedrich, 17 photomodication of cells, 312
optics, 277, 284 passive optical network (PON), 291 photomultiplier tubes, 3
adaptive optics, 29, 151, 178, 184, 247, Patel, C.K.N., 92, 150, 186 photomultipliers, 26, 245
248, 248, 329 Pauli, Wolfgang, 18, 1920, 206 photonic bandgap (PBG), 297, 298
biomedical optics, 277, 308313, Payne, David, 196, 197, 210, 280 photonic bandgap bers, 277
309312, 334335 PBG (photonic bandgap), 297, 298 photonic crystal ber (PCF), 298299,
future trends in, 331 PCF (photonic crystal ber), 298299, 299, 317, 327
industrial and governmental research 299, 317, 327 photonic integrated circuit (PIC), 293, 338
laboratories, 9, 2330 Pease, F.G., 246 photonic lanterns, 301
microuidics and, 301, 311, 318 Pepys, Samuel, 265 photonic materials, 315
military optics, 49, 55, 56, 64, 69, 79, Perilli, 283 photoreceptors, 40, 63, 134, 135
149152 periodically poled lithium niobate (PPLN), photoreconnaissance. See spy satellites;
nonlinear optics, 114117, 183, 213 surveillance imaging
213217, 219220, 238 periscope, 53 photorefractive keratectomy (PRK), 260,
physiological optics, 14, 15 Perkin, Richard, 64, 66 261
quantum optics, 4, 9, 166, 222, 300, Perkin-Elmer Corp., 50, 66, 90, 143, 153, phototypesetting, 50
321, 331 155, 185, 249 Physical Review Letters (journal), 82, 83,
R&D funding, 9, 185188 Pershan, Peter, 115 114, 115, 223, 225
optics (history), 35 personal computers, 135, 141, 279, 282, physicists, post-World War II statistics, 85,
pre-1800, 11 331 86
pre-1940, 34, 944 Peters, C. Wilbur Pete, 50, 55, 114 physiological optics, 14, 15
19411959, 4973, 8587 Peterson, Otis, 95, 161 PIC (photonic integrated circuit), 293, 338
19601974, 79180 petroleum industry, 332 picosecond lasers, 237239
1970s status, 8587 Pfund, August Hermann, 70 pillars of formation (star formation),
19751990, 183236 PHASAR routers, 293 250251, 251
1991present, 277323 phase change recording, 140 piplin, 213
future trends in, 327339 phase-shift keying (PSK), 210, 291, Pittsburgh Conference on Analytical
Optics Express (journal), 312, 313 294295, 295 Chemistry and Applied
Optics in the Life Sciences (meeting), 313 phase-shifting interference microscope, Spectroscopy, 50
Optics Letters (journal), 299, 312 144, 145 Pittsburgh Plate Glass, 24
Optics Technology, 100 phase-shifting interferometric holography, Planck, Max, 12
opticution, 225 145 Planck space telescope, 252
Optiks (Newton), 11 phase-shifting interferometry, 143, 144, plasmonic nanoparticles, 315
opto-electronic integrated circuits 146147 plastic sheet polarizer, 51
(OEICs), 293 phased array routers, 293 plutonium, laser isotope enrichment,
Optoelectronics Research Center (ORC), Philips, 138, 139, 140 163164
299 Philips Audio Division, 138 PMD (polarization-mode dispersion), 211
optogenetics, 334 Philips Research Laboratories, 138 Pocket Kodak camera, 32
optometer, 14 Phillips, W., 220, 221, 224, 225 Pohl, R., 68
Orange Book (optical discs), 140 Phebus laser, 169 Polacolor, 49
Orbiting Astronomical Observatory photo-nishing industry, 31 Polanyi, Tom, 102
(OAO), 247, 249 photo-thermo-refractive (PTR) glass, 318 polarization, 11, 5152, 143, 146, 169,
ORC (Optoelectronics Research Center), photoablation, 261 197, 210, 211, 241242, 291, 294,
299 photoacoustic imaging, 309 295
organic/inorganic composite LEDs, 318 photoacoustic microscope, 310 polarization-based stereoscopy, 51
organic light-emitting diodes (OLEDs), photoactivated localization microscopy polarization-mode dispersion (PMD), 211
318 (PALM), 311 polarized reection, 11
organic photoreceptors, 63 photoactive pigment electrography, 62 polarized windshields, 51
Osaka University, 169 photobiostimulation, 334 polarizing sheets, 51
Oseen, C.W., 269 photocathode materials, 3 polarizing sunglasses, 51
OSRD (Ofce of Scientic Research and photochromic lenses, 267 Polaroid Corp., 51, 65, 122, 158, 280
Development), 27 photocopiers, 50 Polaroid process, 49, 52, 136, 158
Ostermayer, F.W., 105 photodynamic therapy, 183184, 309, Polaroid SX70 camera, 64
Overage, Carl, 64 312, 334 Polavision instant movies, 52
Oxford University, 114 photoelectric effect, 3, 12 Pollard, Marvin, 55
oxide semiconductors, 270 photographic emulsions, 31 polycarbonate, 266
Ozanics, V., 253 photographic lm, 10, 15, 34, 39, 51, 52 PON (passive optical network), 291
photographic lters, 51 Popov, Yu. M., 107
photography, 3, 10, 15 Porro prisms, 14
P in the 1800s, 31 Porter, J., 256
Paanenen, Roy, 187 cellulose nitrate, 15 Portnoi, E.L., 111
Paisner, Jeffery A., 162 color lm, 34, 52 Porto, S., 218
Pake, George, 62 color photography, 3, 10, 33, 34 Post Ofce Research Laboratories (UK),
PALM (photoactivated localization dry plates, 15 298
microscopy), 311 lm, 10, 15, 34, 39, 51, 52 potassium dihydrogen phosphate, 114,
Palmer, Roger C., 133 instant photography, 51 116
Palo Alto Research Center (PARC), 135, Kinetoscope, 34 PPLN (periodically poled lithium niobate),
136 lensless photography, 121 213
Index 351
praseodymium ions, 104 quantum-well infrared photodetectors Rider, Ron, 136
Preserving the Miracle of Sight: Lasers (QWIPs), 316 Ridley, Sir Harold, 262, 263
and Eye Surgery (National quantum-well lasers, 202, 227, 228 Rigden, J. Dane, 88, 89, 89, 90
Academy of Sciences), 261 quantum-well materials, 316 Rigrod, W.W., 89
Pressel, Phil, 79, 160 quantum wells, 228, 304, 315, 316 RIT method, 190, 190
Pressley, R.J., 104 quantum wires, 316 Ritchey-Chretien Cassegrain wide-eld
Priest, I.G., 43 quasi-phase-matching (QPM) technique, design, 4
Princeton University, 249 213 Ritchey, George, 244
Pringsheim, P., 69 qubit, 321, 321, 322, 322 Riverside Research Institute, 86
printers Quist, T.M., 105, 108, 109, 187 RK (radial keratotomy), 259260
inkjet printers, 50 QWIPs (quantum-well infrared Robinson, C. Paul, 162
laser printers, 134137, 136, 137 photodetectors), 316 Rochester Optical Society, 25
printing technology, 50 Rockefeller, David, 157158
prism lenses, 265
prisms, 12, 14, 21, 89, 120, 216, 233, 240, R Rockefeller family, 85
Rockefeller Foundation, 244, 245
241, 265, 266, 267 radial keratotomy (RK), 259260 rod-in-tube bers, 55, 190
Pritchard, David, 96, 220 radiation pressure, 223 Rohlsberger, R., 222
PRK (photorefractive keratectomy), 260, radio communication, 26 Roman, Nancy, 249
261 Radio Corporation of America (RCA), 26, room-temperature GaAs-AlGaAs
Problems in Nonlinear Optics (Khokhlov 53, 129, 269, 270 heterostructure semiconductor
and Akhmanov), 116 radio technology, World War I, 2526 lasers, 203
Project 3 committee, 65 radioastronomy, 50 Roosevelt, Franklin D., 27, 28, 185
Project Blackeye, 150 Radioptics, 161 Rosenberg, R., 96
Prokhorov, Alexander, 218 Raman, Chandrasekhara Venkata, 19, 19 Ross, M., 105
PSK (phase-shift keying), 210, 291, Raman effect, 19 Rossell, Henry Norris, 18
294295, 295 Raman frequency combs, 301 Rothe, Karl, 175
PTR glass (photo-thermo-refractive glass), Raman spectroscopy, 19, 218, 219, 308, Rouard, Pierre, 70, 72
318 310 Royal Observatory (Greenwich), 26
Pulkovo Observatory, 14 Ramsey, Norman, 220 rubber manufacturing, 50
pulse compression, 216, 216 Rand, S.C., 28 ruby lasers, 83, 84, 88, 94, 95, 100, 103,
pulsed argon ion laser, 91, 98 rare earth ber lasers, 4 114, 115, 116, 116, 121, 124, 149,
pulsed dye lasers, 96, 238239 rare earth ions, 104 175, 186, 218, 232, 234
pumped dye lasers, 95, 163, 164, 177, 234 rare-earth metal-doped glass ber, 210 ruby masers, 83
pumping (lasers), 4 rare gas-halide excimers, 92 Ruddock, Ken, 98, 99
Purcell, Edward, 64 ray guns, 149 Rudolph, Paul, 35
Purdue University, 186 Rayleigh, Lord, 12 Rudolph Instruments, 305
Raytheon, 100, 178, 185, 187 Runge, Peter, 96
RCA (Radio Corporation of America), 26,
Q 53, 129, 269, 270
Rupprecht, Hans, 110
Russell, Henry Norris, 18
Q-switching ruby lasers, 94, 115, 116, 116 RCA Laboratories, 185, 199, 201, 227 Russell, James, 138
QAM (quadrature amplitude modulation), re-writable discs, 139140 Russell, Phillip, 179, 277, 297, 300, 300,
291 Reagan, John, 176 327
QCLs (quantum cascade lasers), 176, 178, reconnaissance cameras, 6467 Rutherford Appleton Laboratory (UK),
318 See also spy satellites; surveillance 169, 235, 235
QD LEDS (quantum-dot LEDs), 272 imaging Rutz, R.F., 109
QIS (quantum information science), reconnaissance satellites Ryan, John, 286
320323, 321, 322 CORONA program, 52, 65, 79, 153, Rydberg, Johannes, 13
QPM (quasi-phase-matching) technique, 157160, 159 Rydberg constant, 220
213 KH-9 Hexagon spy satellite, 153156, Rydberg formula, 17
quadrature amplitude modulation (QAM), 154156, 158
291 Sputnik, 52, 73, 79, 85, 157, 185
quadrature phase-shifted keying (QPSK), recording spectrophotometer, 43 S
291, 294 rectier lens, 159 Saint-Ren, Henry C., 53
quadropole trap, 220 Red Book (optical discs), 138139 samarium-doped calcium uoride, 104
Quantatron, 100 Rediker, R.H., 109, 206 samarium ions, 104
quantization of light, 3 Reeves, Will, 300 Sarles, L.R., 104
quantum algorithms, 321, 322 reection holography, 121, 122 Sasano, Yasuhiro, 179
quantum cascade lasers (QCLs), 176, 178, refractometer, 35 satellites. See spy satellites; surveillance
318 refractors, 14 imaging
quantum communications, 323 Reinberg, A.R., 105 Saunders, Frederick A., 18
quantum computers, 320323, 329 Reinitzer, Friedrich, 269 Saunderson, J.L., 44
quantum-conned semiconductors, 178, Reintjes, J., 216 scanners, for barcodes. See barcode
315316 remote sensing, 175178 scanners
quantum-dot (QD) LEDs, 272 Rempel, Bob, 89, 99 Schadt, Martin, 270
quantum dots, 308, 312, 315316, 322 Renhorn, Ingmar, 178 Schaefer, Fritz, 95
Quantum Electronics Conference (High Research Institute of Experimental Physics Schaffert, Roland, 57
View, NY), 82 (Russia), 166 Schawlow, Arthur, 50, 8183, 92, 96, 98,
quantum error correction, 321, 321 residual spectrum method, 20 103, 104, 107, 149, 209, 220, 221,
quantum information, 222 resonance radiation pressure, 223 222, 224, 225
quantum information science (QIS), resonant Raman spectroscopy, 218 Schindler, Rudolf, 53
320323, 321, 322 ReSTOR lens, 263 Schmidt, Bernard, 244
quantum-key distribution, 320 retina, 40, 41 Schmidt camera, 4, 244, 245
quantum mechanics, 3, 9, 1718, 232, retinal, 41 Schotland, Richard, 175
320323, 331 retinene, 41 Schott, Otto, 9, 14, 23, 35
quantum optical sensitivity, 331 ReZoom lens, 264 Schott and Sons, 14, 15, 70
quantum optics, 4, 9, 166, 222, 300, 321, Rhees, Benjamin Rush, 33 Schott Glass, 248
331 rhodopsin, 39, 41 Schrdinger, Erwin, 9, 18
quantum simulators, 329 Richard, Jules, 34 Schroeder, Harold, 72
quantum theory, 3, 9, 13, 17, 18, 21 Richards, A. Newton, 27 Schuda, Felix, 96
352 Index
Schulte, Dan, 250 smoothing by spectral dispersion (SSD), resonant Raman spectroscopy, 218
Schultz, Peter, 190, 191 170 time-domain laser spectroscopy, 219
Schwartz Electro-Optics, 234 Smullen, Louis, 175 transient grating spectroscopy, 238
The Science of Color (Optical Society of Snavely, Ben, 95, 161, 162, 163 spectrum, 12
America), 43 Snitzer, Elias, 56, 101, 102, 104, 187, Spencer, William, 62
Scifres, Carol, 229 195196, 197, 280 Spencer Lens Co., 24
Scifres, Donald R., 229, 229, 285 Soffer, Bernard, 95 spin-orbit coupling, 18
Scott, Rod, 66 SOFIA telescope, 252 spincasting manufacturing technique,
SDI (Strategic Defense Initiative), 151 soft contact lenses, 253, 256 253, 254
second-generation lasers, 205 SOI modulator (silicon-on-insulator Spitzer, Lyman, 249, 252
second harmonic generation (SHG), 4, 117, modulator), 294 Spitzer telescope system, 249, 251, 252
213, 214, 218, 221, 238, 316, 318 solar cells, 332 SPM (self-phase modulation), 117, 215
second order nonlinear interactions, 215 solar panels, 332 Sprint, 278
secret keys, 323 solar power, 329, 332 Sputnik, 52, 73, 79, 85, 157, 185
segmented telescope, 3 Solarz, Richard W., 162 spy satellites, 79
Seiko, 270 solid-state lasers, 4, 84, 101, 103106, CORONA program, 52, 65, 79, 153,
self-developing lm, 51 125, 126, 131, 178, 227, 228, 231, 157160, 159
self-phase modulation (SPM), 117, 215 242, 316 KH-9 Hexagon spy satellite, 153156,
self-trapping, 117 diode laser-pumped solid-state lasers, 154156, 158
semiconductor circuits, 50 105106 Sputnik, 52, 73, 79, 85, 157, 185
semiconductor diode lasers, 4, 107111, free-space solid-state lasers, 242 See also surveillance imaging
199, 209, 210, 240241 tunable lasers, 105, 232236, 233235 SRI International, 86
semiconductor laser diodes (LDs), 105 solid-state lighting, 339 Srinivasan, R., 257, 258, 259, 260, 261
semiconductor lasers, million hour solid-state masers, 50 SSD (smoothing by spectral dispersion),
paper, 205 soliton laser, 241 170
semiconductor nanoparticles, 312 solitons, 4, 25, 117, 215, 216 STAAR, 262, 264
sensing particles, 327328 Soltys, T.J., 108 Standard Oil (Indiana), 24
sensor systems, 327328 Sommerfeld, Arnold, 17, 40 Standard Telecommunications Laboratory
separate connement heterojunction SommerfeldKossel displacement, 17 (STL), 199
quantum well lasers, 202 Sony, 138, 140, 141 Stanford Research Institute, 86
SERS (surface-enhanced Raman Sorokin, Peter, 94, 103, 104, 104, 107 Stanford University, 96, 105, 186, 196,
scattering), 316 Space Infrared Telescope Facility (SIRTF), 220, 225
Shack, R.V., 246 249 Starre Optical Range, 29
Shank, Charles V., 96, 216, 239, 241, 304 space race, 85 Starkweather, Gary, 134, 135
Shannon, Claude E., 189 Spaeth, Mary, 9495, 96, 96 Steane, Andrew, 321
Shannon, R.R., 246, 250 special relativity, 12 STED (stimulated emission depletion
Shannon limit, 209 speckle photography, 145 microscopy), 311
Shapiro, S.L., 117, 238 spectacles. See eyeglasses Steinvall, Ove, 178
Shaver, William, 189 spectra, 2021 stellar spectra, 13
She, C.Y., 178 chemical elements, 21 stereoscopic surveillance imaging, 51
Shen, Y.R., 221 infrared spectral lines, 18 Stetson, Karl, 145
Shenstone, Allen G., 21 multiplets, 17 Stevenson, Mirek, 84, 103, 104, 104, 107
SHG (second harmonic generation), 4, 117, singlets, doublets, and triplets, 17 Steward Observatory, 246
213, 214, 218, 221, 238, 316, 318 SommerfeldKossel displacement, 17 Stickley, C. Martin, 185, 186
Shimizu, Fujio, 116 stellar spectra, 13 still photography, 34
Shimizu, M., 197 Spectra Diode Laboratories Inc., 228, 229 Stimson, F.J., 20
Shiner, Bill, 101102 Spectra-Physics, 89, 90, 91, 9799, 121, stimulated emission depletion microscopy
Shiva laser, 168 129, 130, 131, 234, 305 (STED), 311
Shlaer, Simon, 39 spectral multiplexing, 308 Stitch, Malcolm, 84
Shor, Peter, 321 spectral reectance factor, 43 STL (Standard Telecommunications
short-wave radio, 26 spectrometers, 50, 305 Laboratory), 199
Sibbett, Wilson, 242, 304 spectrophotometers, 43, 44 STN (super-twisted nematic), 270
Sieder, Irwin, 104 spectroscopic instruments, 20 stochastic optical reconstruction
Siegel, Keeve M., 168 spectroscopy, 9, 218 microscopy (STORM), 311
Siegman, Anthony, 105 applied spectroscopy, 20, 4950 Stoicheff, Boris, 115, 116, 221
SILEX process, 165 astronomy and, 13, 13, 1819 Stokes, G.G., 19
Silex Systems Ltd., 165 atomic physics and, 1213, 13 Stolen, R.H., 215, 216
silicon-on-insulator (SOI) modulator, 294 coherent anti-Stokes Raman (CARS) STORM (stochastic optical reconstruction
silicon photonics, 242, 293, 294 spectroscopy, 219, 308 microscopy), 311
silicon TFTs, 270 with continuous-wave dye lasers, 96 Strategic Defense Initiative (SDI), 151
Simplex camera, 34 Doppler-free laser spectroscopy, 220 Stratoscope project, 249
Simpson, W.M., 217 dynamic grating spectroscopy, 219 Stratton, Samuel W., 27
SINDA program, 154 early history, 1721 Straus, Josef, 285
single-mode bers, 5556, 191, 206, 210, femtosecond absorption spectroscopy, Strehl ratio, 256
279, 301 180 stretched-pulse lasers, 242
single molecular detection, 311 ame-emission spectroscopy, 20 Strickland, D., 235, 242
single-photon systems, 338 uorescence correlation spectroscopy, stripe-contact technology, 201, 203
single-stripe lasers, 228, 229, 229, 231 312 stripe-geometry lasers, 111, 203
SIRTF (Space Infrared Telescope Facility), laser-based spectroscopy, 147, 232 Stroke, George W., 122
249 laser-induced-breakdown spectroscopy Strong, Henry, 31
Skunk Works, 65 (LIBS), 178 Strong, John, 69, 70, 71, 245
Slepian, Joseph, 24 laser spectroscopy, 218219, 221222 Stroud, Carlos, 9, 23, 96
Smakula, Alexander, 69, 70, 72 linear spectroscopy, 218219 structured light imaging, 328
Small Business Innovative Research nonlinear spectroscopy, 215, 219221 Struve, Horst, 165
Program, 187 optical modulation spectroscopy, 219 Struve, Wilhelm, 14
Smelser, G.K., 253 optical spectroscopy, 3, 17, 19, 21, 24, Stuhlmann, Otto, 69
Smith, Dow, 158 50, 175, 218, 220, 335 sub-Doppler laser cooling, 222
Smith, George F., 103 quantum mechanics and, 1718 subshells, 17
Smith, Richard G., 205 Raman spectroscopy, 19, 310 Sugimoto, Nobuo, 175
Index 353
Sullivan, Walter, 84 Kitt Peak National Observatory, 245, TPF (two-photon-induced uorescence),
Sumski, S., 111 246 238
Sun-powered laser, 101 Large Space Telescope (LST), 249 transient grating spectroscopy, 238
Super Kodak Six-20 camera, 3637, 36 laser propulsion, 336337 Tretyakov, Dmitriy N., 111
super-twisted nematic (STN), 270 lasers in, 184, 245248, 251, 252 TRG (Technical Research Group Inc.), 82,
supercontinuum, 216, 216 Mt. Palomar observatory, 4, 18, 244, 84, 100, 149, 186
supermarket barcode scanners, 129131 245 Trion Instruments Inc., 100, 114
superresolution, 311 Mt. Wilson Observatory, 18, 244, 247 triplet-state absorption, of dyes, 95
surface-enhanced Raman scattering Next Generation Space Telescope tristimulus integrator, 43
(SERS), 316 (NGST), 252 Trokel, Stephen, 259, 260
surface plasmon resonance, 312 refractors, 14 troland (unit), 38
surgery, 306 SOFIA telescope, 252 Troland, Leonard Thompson, 38, 39, 43
biomedical optics, 277, 308313, Space Infrared Telescope Facility Trukan, M.K., 111
309312, 334335 (SIRTF), 249 Truman, Harry, 29
excimer laser surgery, 257261, 258, space telescopes, 249252, 250, 251 TRW, 150
259, 306 spectroscopy and, 13, 13 Tuccio, Sam, 95, 161, 163
intraocular lenses, 262264 Spitzer telescope system, 249, 251, 252 Tukey, John W., 65
LASIK technique, 5, 183, 260, 261, Thirty Meter Telescope (TMT), 248 tunable dye lasers, 4, 9496, 95, 161
306, 308, 312 television, 53, 270 tunable optical parametric oscillators, 176
nanosurgery, 312 Teller, Edward, 162 tunable quantum cascade lasers, 176
photorefractive keratectomy (PRK), 10-J Janus laser, 168 tunable solid state lasers, 105, 232236,
260, 261 terabit-per-second ber, 209211, 210 233235
radial keratotomy (RK), 259260 TeraMobile project, 305 Tuohy, Kevin, 253
See also ophthalmic surgery terbium ions, 104 Turner, Arthur Francis, 70, 72, 246
surveillance imaging Terhune, R., 115, 219 Twain, Mark, 335
19541974, 6467 Tesla, Nikola, 23 twisted nematic (TN) effect, 270
CORONA program, 52, 65, 79, 153, Tessar lens, 32, 33, 35 two-photon-induced uorescence (TPF),
157160, 159 tetrahertz radiation spectrometer, 305 238
KH-9 Hexagon spy satellite, 153156, Texas Instruments, 50, 105, 185 two-photon microscopes, 305
154156, 158 TFT LCD (thin lm transistor liquid two-wavelength holography, 145
Sputnik, 52, 73, 79, 85, 157, 185 crystal display), 270272, 271 TwymanGreen interferometer, 144
stereoscopic surveillance imaging, 51 Thack, Robert, 211 Tyndall, John, 53
U-2 spy plane, 49, 52, 6467, 66, 157, Thelen, Alfred, 70
158
See also spy satellites
theory of entanglement, 323
theory of special relativity, 12 U
Svanberg, Sune, 175 thermal evaporation, 69 U-2 spy plane, 49, 52, 6467, 66, 157, 158
Sweden NDRI, 177 thin lm coatings, 73 U-235, laser isotope enrichment, 161
SX-70 color lm, 52 thin lm interference, 68 Uchino, Osamu, 179
Symbol Technology, 132 thin lm polarizers, 71 Uhlenbeck, George, 18
synthetic rubber, 4950 thin lm transistor liquid crystal display ultra-low-loss bers, 327
(TFT LCD), 270272, 271 ultrafast electro-optic sampling systems,
thin lms, 72 305
T third-order nonlinear interactions, 215 Ultrafast Epiphany: The Rise of Ultrafast
Talanov, Vladimir, 116 35-mm precision cameras, 34 Science and Technology in the Real
Talon Gold, 151 Thirty Meter Telescope (TMT), 248 World (CLEO paper), 305
Tanner, Howard, 71 Thomas, L., 178 ultrafast-laser technology, 304306, 305,
Tappert, F., 117, 215 Thompson, Kevin, 64, 79, 157 306
TAT-12, 282283 Thomson, J.J., 12 ultrafast manufacturing systems, 306
TDM PON technology, 291 Thomson-CSF, 139 ultrashort lasers, 306
Teague, Walter Dorwin, 37 three-dimensional movies, 51 ultrashort-pulse lasers, 96, 237242,
Technical Research Group Inc. (TRG), 82, three-level lasers, 83 239242
84, 100, 149, 186 Three Mile Island nuclear accident, 164 Unar lenses, 35
TecnisIOL lens, 263 three-section tunable DBR lasers, 293 uncertainty principle, 18
telecom bubble, 277, 304 ThreeFive Photonics, 293 United States Army Signal Corps, 72
telecommunications industry, 282286 thulium ions, 104 United States Enrichment Corp., 164, 165
telephony, 26, 203207, 204, 206, 207, time-averaged holography, 145 United Technology Research Center, 186
279, 282 time-domain laser spectroscopy, 219 Universal Jewel professional folding dry
teleportation, 321 time-domain reectometry, 328 plate camera, 35
telescopes, 4, 11, 1314, 184, 249252, tipping furnace, 200, 200 University of Arizona, 176, 246, 248
250, 251 titanium:sapphire laser, 234, 235, 236, University of Arizona, Optical Sciences
Advanced X-Ray Astrophysics Facility 242, 304 Center, 86
(AXAF), 249 TMT (Thirty Meter Telescope), 248 University of Chicago, 29, 186
Chandra X-ray Observatory, 249, 251, TN effect (twisted nematic effect), 270 University of Illinois, 62, 178, 186, 228
251 Tolman, Richard C., 27 University of Maryland, 97, 186
Compton Gamma Ray Observatory Tomsk Laser Institute (Russia), 178 University of Michigan, 213, 306
(CGRO), 249 Tonucci, R.J., 298 University of Michigan, Willow Run
Extremely Large Telescope (ELT), 248 Topics in Biomedical Optics (BIOMED Laboratories, 86, 100, 119, 120, 122
Giant Magellan Telescope (GMT), 248 meeting), 313 University of North Carolina, 186
Great Observatories, 249, 252 topological quantum computation, 322 University of Pennsylvania, 186
ground-based telescopes, 244248, toric contact lenses, 255, 256 University of Pittsburgh, Mellon Institute,
245248 toric intraocular lenses, 264 71
Hubble Space Telescope (HST), 4, 13, Toschek, P.E., 220 University of Rochester, Institute of
143, 184, 247, 249250, 250, 251, Total Quality Movement, 63 Optics, 25, 33, 54, 134, 143, 158,
252 touch panels, 271 168, 169, 170, 185, 186, 304
James Webb Space Telescope (JWST), Tourist Multiple camera, 34 University of Southampton, 196, 197, 242
252 Townes, Charles, 50, 79, 81, 82, 82, 85, University of Toronto, 116, 177
Keck Ten-Meter-Diameter Telescope 103, 107, 116, 149, 209, 218, University of WisconsinMadison, 230
Project, 248 246247 Univis, 266
Kepler space telescope, 252 up-conversion gating, 238
354 Index
Upatniek, Juris, 119120, 120 Watson Research Center, 84, 94 Wright-Patterson Air Force Base, 186
UPC symbol, 128, 129 wave nature of light, 1112 writable and re-writable discs,
UPC Symbology Committee, 128 waveparticle duality, 12 139140
Ur-Leica camera, 35 wave theory of light, 11, 69 write-once read-many-times (WORM)
uranium, laser isotope enrichment, wavefront reconstruction, 121 media, 140
161163 wavelength-division multiplexing (WDM), Wu, Shin-Tson, 269, 271
uranium-doped calcium uoride, 104 210211, 280, 282283, 284, Wurzburg, E.L., Jr., 44
Urbach, John, 135 288289, 290, 291, 293 Wyant, James C., 143, 246
U.S. Department of Energy (DOE), 29, 164 wavelength-division-multiplexing (WDM) WYKO Corp., 144
U.S. National Bureau of Standards, 20, 24, coupler, 196, 210, 211 Wynne, James J., 257261, 308
25, 26, 27, 43, 185 Webb, Watt, 310, 311
U.S. Naval Observatory, 26
U.S. Rubber, 24
Wehrenberg, Paul J., 138, 140
Weiman, Carl, 221, 225 X
Weinreich, Gabriel, 114 x-ray tube, 24
xerography, 5763, 5861, 134
V Weisner, J.B., 27
Welch Allyn, Inc., 131 Xerography and Related Processes
vacuum ultraviolet spectroscopy, 20 Welford, Walter, 72 (Dessauer), 57
van Driel, Henry, 297 Wenzel, Robert, 163 Xerox 914, 57, 59
van Eijkelenborg, Martijn, 299 Werner, Christian, 178 Xerox 7000, 135136
van Heel, Abraham C.S., 50, 54 Werner, Dick, 153, 154 Xerox 9700 Electronic Printing System,
VanderLugt, Anthony, 120 Western Electric Research Laboratories, 137
Varian Associates, 100, 187 25 Xerox copiers, 50
Vasicek, Antonin, 7172 Westinghouse, George, 23 Xerox Corp., 50, 57, 63, 134,
Vaughan, Art, 250 Westinghouse Research Laboratory, 24, 135, 137
Vavilov, Sergey, 114 26, 100, 128, 150, 185, 270 Xerox Model A processor, 58, 58
Vavilov State Optical Institute, 121 WF/PC (Wide-Field Planetary Camera), Xerox PARC, 227, 228
vectograph, 51 250
VHS tape, 138
vibronic lasers, 233
Wheelon, Albert Bud, 153 Y
White, Alan, 88, 89, 89 Yablonovitch, Eli, 332
videotex, 279 White, George, 135, 136
Virtual Journal for Biomedical Optics, 313 YAG lasers (yttrium aluminum garnet
white-light continuum, 304 lasers), 104, 105, 124, 125, 186,
visibility, 43 white-light supercontinuum, 300, 300
vision, 3839, 39 225, 240, 242, 257, 258, 259, 301,
Whitehouse, Dave, 187 304
vision correction, 306 Wichterle, Otto, 253, 254
contact lenses, 183, 184, 253256, 254, Yahashi, I., 111
wide-eld-of-view camera, 4 Yale University, 91
255, 260, 262, 333 Wide-Field Planetary Camera (WF/PC),
excimer laser surgery, 257261, 258, Yamane, Tets, 225
250 Yariv, A., 297
259, 306 Williams, Richard, 269
intraocular lenses, 262264 Yeh, P., 297
Williams, Robert E., 39 Yerkes Observatory, 14, 244
LASIK (laser in situ keratomileusis), 5, Willner, Alan E., 338
183, 260, 261, 306, 308, 312 Young, Thomas, 11, 14, 68, 69
Willow Run Laboratories, 86, 100, 119, ytterbium-doped lasers, 106
photorefractive keratectomy (PRK),
260, 261
120, 122 ytterbium ber, 304
Wilson, Joseph C., 61, 61 ytterbium-ber lasers, 242
radial keratotomy (RK), 259260 windshield polarizer, 51
in vitro methods, 310312 ytterbium ions, 104, 105
Winker, David, 176 yttrium aluminum garnet (YAG) lasers,
in vivo imaging, 308310 WIRE space telescope, 252
See also eyeglasses; ophthalmic surgery 104, 105, 124, 125, 186, 225, 240,
WISE space telescope, 252 242, 257, 258, 259, 301, 304
vision research, 10, 3841 WMAP space telescope, 252
Vistakon Co., 255 Yule, J.A.C., 44
Wood, Robert, 34
visual reception, 3839 Wood, R.W., 19, 19
vitamin A, 40, 41 Woodall, Jerry, 110 Z
Vogel, Hermann Wilhelm, 13 Woodbury, Eric, 115 Zeiger, H.J., 50, 109
von Fraunhofer, Joseph, 12, 13 Workshop on Optical and Laser Remote Zeiss, Carl, 9, 14, 23, 35
Von Graefe, A., 265 Sensing, 178 Zeiss, Roderich, 35
von Neumann, John, 107 World War I, 15, 24, 25, 33, 49 Zeiss (company), 15, 33
Vul, R.M., 107 World War II, 3, 26, 41, 4950, 51, 85, Zeiss Foundation, 35
Vulcan laser, 169 185, 245 Zeiss Ikon AG, 35
vulcanite, 15 aerial cameras, 66 Zeiss/IMRA, 305
optical coatings, 7071 Zeldovich, Boris Ya., 116
W World Wide Web, 279, 282 Zenker, Gabriel, 69
Zernike, Frits, 54
Wald, George, 4041, 41 WORM media (write-once
Wallop, Malcolm, 151 read-many-times media), 140 Zeta laser, 168
Walther, Herbert, 175 Worokin, Peter, 84 Zimar, Frank, 191
Wang, Charles C., 177, 179 Wratten & Wainwright, 33 zinc germanium phosphide (ZGP), 215
Warburg, Otto, 41 Wright, Fred E., 24, 25 Zoller, Peter, 321
Watson, Gene, 97 Wright Air Development Command, 65 Zuev, Vladimir, 178
Index 355