Stuff: the materials the world is made of
by
IVAN AMATO


Basic Books A Division of Harper Collins Publishers

Copyright 1997 by Ivan Amato.

Published by Basic Books A Division of Harper Collins Publishers, Inc.

All rights reserved.  Printed in the United States of America.  No part
of this book may be used in any manner whatsoever without written
permission except in the case of brief quotations embodied in critical
articles and reviews.  For information, address Basic Books 10 East
53rd Street, New York, NY 10022-5299.

FIRST EDITION

Designed by Laura Lindgren

Library of Congress Cataloging-in-Publication Data

Amato, Ivan.

Stuff: the materials the world is made of / by Ivan Amato.  - 1st Ed.

Includes index.  ISBN 0-465-08328-5 1. Materials-Popular works.  2.
High technology-Popular works.

To Mom and Dad Contents

Acknowledgments Introduction: Stuff, Stuff, Everywhere Stuff

PART I

Trial and Error

One: Earth, Air, Fire, and Water Two: New Stuff in the Old World Three:
The Secret Architecture of Stuff

PART II

The Birth of a Superdiscipline

Four: Materials Research Comes of Age Five: The Materials Serengeti
Synthetic Diamond Biomimetic Materials Smart Materials Other
Sightings

PART III

Materials by Design

Six: The Rite of Atomic Masons Spray Painting with Atoms The Road to
Band-Gap Engineering

Seven: Composing Steel

Eight: The Stuff That Dreams Are Made Of

References and Notes Index

Acknowledgements

The path to this book is populated by people to whom I owe an enormous
debt of gratitude.  I thank my teachers throughout life for encouraging
my curiosity and for helping me learn how to learn.  Special thanks to
my fourth-grade teacher, Mr.  Wallace, wherever you and your motorcycle
are.  Special thanks to my ninth-grade biology teacher, Mrs.  Hershey,
whose perpetual enthusiasm probably enriched the lives of more people
than she is aware of.  Thanks to Dave Roberts, a longtime friend whom I
first got to know when he was my college advisor and organic chemistry
teacher at Rutgers College; he convinced me at the impressionable age
of eighteen that I actually might have a working head on my shoulders.
Thanks also to Professor Renee Weber for engraving on my mind that the
stuff of the world is as good an altar as one in a cathedral.  Finally,
in the teacher category I must send heaps of thanks to my one and only
journalism professor, Holly Stocking, who helped me believe that I
could actually become a science writer.

In the writing world I thank Joel Greenberg and Laurie Jackson-Vaughan,
both of them former editors of Science News, for taking me on as an
intern at the much-loved publication in the summer of 1986.  That
experience became the ticket into the science writing world I so badly
wanted to enter.  I thank Curt Suplee of the Washington Post for
actually printing my articles in one of the nation s premier
newspapers, an honor that I at one time thought was never within my
reach.  Thanks also to Nancy Enright of the American Chemical Society
for taking me on as green as I was; to Pat Young, a former editor of
Science News who saw it fit to make materials science a standard topic
of coverage and for making me the reporter to cover it; to Tim
Appenzeller for his demand for clarity and a relentless ability to see
possibilities in stories that many editors would have turned down; and
to Mat Heyman, head of the Business and Public Affairs Division at the
National Institute of Standards and Technology, for his heroic
tolerance of my short attention span.

Since I cannot thank everyone in the science world who I would like to
thank, I will name several people who can serve as representatives:
Stephen Carr of Northwestern University for a fateful conversation
early in my writing career that helped me see how vast and great the
field of materials science and engineering is; the late Cyril Stanley
Smith, whose poetic and philosophic approach to the world of materials
has changed me forever; Federico Capasso, Greg Olson, Mehmet Sarikaya,
Paul Calvert, John Angus, Klaus Zwilsky, and Craig Rogers, who have
helped me in many ways to bring this book from an idea to a thing that
you can throw across the room.  I thank the many hundreds of other
scientists and engineers without whose collective effort I would never
have been able to see the material world in all of its wonder.

I thank Robert Friedel, historian of technology at the University of
Maryland, for the skill and diplomacy with which he read early drafts
of this book, and for the way his mind works.  May my future work
emulate his.

In the book world I thank Ivars Peterson, the tireless reporter at
Science News who proved to me that writing a book is indeed possible. I
thank David Lindley, a book author himself as well as a sometime
colleague at Science magazine, for hooking me up with Susan Rabiner at
Basic Books.  And I thank Susan for ushering this book slowly but
surely from the raw ore that she encountered initially into a handsome
edition that I can rest assured will sit upon my parents bookshelf.

Which brings me to my family, whom I thank most of all.  To my father,
Sol, and my mother, Sylvia, thank you for literally making me and for
helping me to see that the world and our ability to know it at all are
infinitely awesome phenomena.  There is no greater gift that parents
can give to their children.

And, finally, I thank my wife, Mary, and my two young boys, Max and
Simon.  I thank Mary for sharing every single up and down that goes
with writing a book.  I thank her for her infinite supply of
encouragement, goodness, and love.  To Max and his little brother
Simon, I thank you both for coming into this world and for reminding me
every day what really is important.

Introduction: Stuff, Stuff, Everywhere stuff

Stop reading this book.  Instead, hold it in your hand.  Feel its
weight.  Run your finger over a page and feel its smooth texture, its
dryness.  Sense the page s compliance to the pressure of your finger.
See how its whiteness is punctuated with black symbols printed in ink
in a few dozen rows.  If you sniff the page, you even may smell some
aromatic chemicals still fleeing from the ink.  Rub the page between
your thumb and forefinger and hear that familiar papery noise.  Twist a
corner and see how the paper crinkles.  Rip the page and see how the
ragged edges hint of the tiny wood fibers that became entangled at a
mill to become the paper.

Now examine the book s fabric spine to which the pages are attached
with an adhesive.  The stack of pages, in turn, attaches to the back
and front covers of the hardcover edition way of sheets of
heavier-gauge paper that flank the book s 1st and last pages.  At the
bindery a machine pasted these nking sheets to the rectangular pieces
of stiff, heavy paperard that the front and back covers are made of.
These heavcovers, themselves enveloped with a sheet of paper or polymer
impregnated cloth, are attached to each other by a flexible paper or
fabric bridge that ends up directly over the spine so that readers can
open the book.  The soft cover edition is a simpler affair of paper,
ink, and glue.

Soft or hard, this book is made through and through of stuff.

And so is every thing else.  Human beings extract about 15 billion tons
of raw material-that s 30 trillion pounds-from the earth each year, and
from that they make every kind of stuff that you can find in every kind
of thing.  Mined ore becomes metal becomes wire becomes part of a motor
becomes a cooling fan in a computer.  Harvested wood becomes lumber
becomes a home.  Drilled petroleum becomes chemical feedstock becomes
synthetic rubber becomes automobile tires.  Natural gas becomes
polyethylene becomes milk jugs and oversize, multicolored yard toys.
Mined silica sand become silicon crystal becomes the base of
microelectronic chips.  Each kind of stuff is a link to enormous
industrial trains whose workers process the world s raw materials into
usable forms that constitute the items of our constructed landscape.

Each kind of stuff also is a palimpsest of innovations in the use of
materials, some going back to prehistoric times.  The wood-pulp paper
from which books are made today comes from a pedigree of cotton and
linen rags, animal-skin parchment, Nile-reed papyrus, and Sumerian clay
tablets.  The ink, a black pigment made from the ground ash of some
carbon-bearing fuel and then suspended in a rapidly evaporating
solvent, has its roots in crushed ore and charcoal mixed with spit or
animal grease for use on cave walls and faces.  The materials in every
book tell a tale that rivals the one conveyed in its words.

It is the same for every other material thing that you encounter. Train
your attention on the stuff of things rather than on their function. 
What you see is a rich medley of materials: the liquid crystal display
of your laptop computer; the gritty concrete sidewalk on which you are
strolling; the nylon of your raincoat s zipper; the
carbon-fiber-reinforced epoxy polymer of your tennis racket; the Kevlar
polymer in your police force s bullet-proof vests; the oak of your
dresser; the diamond in your engagement ring; the nickel-based
superalloy in the turbine blades in the engine of an airliner you are
flying in; the warm, supple skin of your newborn; the cool, transparent
glass of your office window; the combination of slick, high-density
polyethylene and stainless steel that make up the artificial hip which
a surgeon may have implanted into you; the cotton of the shirt you are
wearing; the aluminum of the can you just drank from.

In a single day the thousands of man-made materials that you encounter,
engage, manipulate, and use display a diversity every bit as wondrous
as that found in living organisms, which are composed of the most
miraculous of all the world s materials-skin, bone, tendon, muscle,
nail, hair, and scads of other biological tissues-all of them honed by
evolutionary engineering into a beautiful marriage of form and
function.

That books and buildings and the things in the world are supposed to be
made of materials suited for their functions is so obvious that it
almost goes without saying.  But things that go without saying long
enough are readily forgotten.  That is why the materials that make up
the world are most often not on people s minds.

When the stuff of the world does comes to the fore, it is often a
consequence of a material failing to serve its intended role.  This
happened in 1988, when the aluminum-alloy skin of an airliner flying
over the Pacific Ocean peeled open and sheared off like the top of a
sardine can, sending a flight attendant to her death.  Articles about
metal fatigue appeared in the New York Times and all over the media.
Aluminum and material failure enjoyed their fifteen minutes of fame. To
the unseen scientists, engineers, and workers who actually transform
the raw stuff of the world into materials like aerospace alloys, the
way materials can go bad is never far from their minds.  Theyand their
bosses, lawyers and insurance companies-know how lives, economies, and
national security depend upon understanding what makes materials the
right stuff or the wrong stuff.

But materials are far too central to our lives and too fascinating to
remain out of mind until they become sensational headlines.  Paying
attention to the materials in your life spawns an endless stream of
fascinating questions (and even more fascinating answers).  Why is a
skyscraper s frame made of steel, not copper or polyethylene?  What is
so special about silicon crystal that it became the foundation of a
society-changing electronics revolution?  What kind of materials does
it take to build a plane that can take off like a jet, accelerate at
20,000 miles per hour, zoom into outer space, and then land two hours
later in an airport on the other side of the planet?

Simply noticing the diversity of stuff in the material world can evoke
an openmouthed sense of wonder akin to visiting a zoo filled with
animals that you had never really seen up close.  That sense of wonder
might even grow into awe when you consider this: the constructed world
brims with metals, ceramics, plastics, fabrics, glass, and thousands of
other specific materials that are found nowhere in the wilderness.

And that sense of awe naturally leads to a question like this one,
which is at the heart of this book: How in the world did humanity learn
to transform the raw stuff of the wilderness into the contemporary zoo
of materials?  Nylon doesn t grow on trees.  It starts out in a barrel
of tarry, smelly coal tar or petroleum.  Someone-in this case a
chronically depressed and brilliant chemist who ultimately committed
suicide-figured out how to pull off the transformation.  Someone had to
figure out how to convert a handful of sand into an optical fiber that
can pipe light around for miles and miles the way a copper wire pipes
electricity.

Aristotle wrote that you cannot know something fully unless you know it
from its beginnings.  This applies to the human use of materials.  So
in this book, I will start from the beginning-about 2.5 million years
ago. That is when our hominid progenitors in the Great Rift Valley of
Africa likely had the original insight that materials as they are found
in the wilderness can be transformed into more useful and empowering
forms.

The original noisy act of materials engineering-striking a stone
against another to produce a sharp, cutting edge where before there had
been a dull surface-is no different in its core from what modern
engineers in Silicon Valley do when they implant boron and phosphorous
atoms (stepping stones for electrical charges) into pristine wafers of
crystalline silicon to create the semiconductor of the microelectronics
revolution.  The goal of both procedures is the same: to alter
materials so that they can perform in new technological ways.

Despite the apparent crudeness of chipping stones to make tools, the
kinship between the original stone flakers and the most modern
materials engineers goes atomically deep.  Silicon atoms, when combined
with oxygen atoms to form specific minerals, is the basis of one of the
most coveted materials of the first stone toolmakers-flint.  Because of
the arrangement of internal mineral grains in flint, the stone can
cleave conveniently in a shell-like pattern that leaves sharp edges.
Moreover, its edges are hard and tough enough to remain sharp even
after cutting and scraping into materials like tendon and bone. Silicon
has been a celebrity element from the beginning.

From the Great Rift Valley, the story of materials progresses through
some 125,000 hominid and human generations (based on a twenty-year
generation) to the present day.  During that vast stretch of time, the
first hominid toolmakers evolved into us, Homo sapiens.  Punctuating
those thousands of millennia were a few major turning points, each
leading to entirely new powers of material transformations that have
changed the course of individual lives and of global history.  (See
Figure 1.)

The use of fire to change the stuff of the world was one of them.  Fire
s heat has the capacity to reach down to the molecular identity of a
material and then to jumble, reorient, and rearrange it to yield almost
magical transformations.  Wood becomes smoke, heat, invisible gases,
and ash.  Dull, crumbly ore becomes lustrous metal.  Wet, malleable
clay becomes hard pottery.  Sand become glass.  It may all have been
accidental chemistry in the hands of the first fire users, but the
consequences were and are world-changing.

Another major turning point was the development of a more deliberate
type of chemistry and its far more extensive ways of reaching into and
transforming the anatomy of stuff.  For centuries alchemists, chemists,
and sundry tinkerers pulverized, mixed, boiled, distilled, roasted,
extracted, electrolyzed, and otherwise fiddled with whatever substance
they could find in the mineral, animal, and vegetable domains.  Out of
this obsessive quest came a myriad of previously unseen
substances-solid, liquid, and gaseous-some of which harbored uses by
themselves or as participants in new chemical reactions.

Along the way, chemists also learned about the limitations of their
business.  Despite the two thousand-year alchemical quest to transform
base materials into gold, no one ever succeeded in transmuting one
element into another.  At least on earth, under normal conditions (that
is, without enormous

FIGURE I

The Stuff of History.  To follow each of the heavy lines across this
diagram Is to trace a word-byte history and projection of four major
classes of materials-metals, polymers-elastomers, composites (whose
properties are derived from two or more materials-such as a polymer
with glass fibers as reinforcement-bound Into a single overall
structure), and ceramics-glasses.  The relative Importance to humanity
of each material class at each historical period is suggested on the
diagram by the proportion of the diagram s vertical axis corresponding
to each material class.  So In 1960, for example, composites were least
important while metals were head and shoulders most Important.  In
2020, however, the four classes of materials are projected to have
roughly equal importance.

MICHAEL F. ASH BY

3:)NVi8odlNI 3AUV138

accelerators), scientists realized they would never become
alchemists-they would never change lead into gold or oxygen into carbon
or any one element into any other element.

This limitation, it turns out, was always more a state of mind than of
reality.  The hard-won insight that the vast menagerie of materials in
the world is a result of a small pantry of elements, that the
personality of each material is the result of an inner hierarchy of
chemical and physical structures, has became the principle of a new
alchemy practiced by people known as materials scientists.  Rather than
gold from lead, their trophies are numerous, ranging from new
lightweight alloys to build a next-generation fleet of super efficient
jetliners to harder than-diamond materials for industrial machine tools
to polymeric drug-soaked medical implants that can release their cargo
into diseased brain tissue.

Yet another major turning point in the human control over the material
world-perhaps the last one possible-is in the offing.  In the past half
century a field known as materials science and engineering has emerged
as a powerful hybrid of many other technical fields.  The practitioners
of this field are coming to a point where they are gaining the ultimate
level of control over the material world.

No longer contented to manipulate stuff, they have tools for seeing,
moving, and understanding individual atoms and the ever larger atomic
collections that become the materials we use.  The emerging ability to
micromanage materials even at the atomic level is giving researchers
unprecedented access to the mostly untapped material wonders of the
periodic table of the chemical elements.  Even compared to today s most
sophisticated materials like the semiconductor in a computer chip or
the gallium arsenide in a CD player s laser, materials scientists say
we still ain t seen nothing compared to the materials that will come.

Materials scientists now recognize that within their new powers lie new
solutions to many of the globe s most vexing problems-among them
pollution, energy supply, housing, transportation, communications, and
poverty: new superconductors that can dramatically increase the
efficiency with which electricity is generated, distributed, and used
could demote petroleum as a factor in geopolitical dynamics; raw
materials for making these superconductors-say, ores containing
unfamiliar elements like yttrium and bismuth-could become strategic
materials; new metallic alloys such as titaniumaluminides and new
composite materials-perhaps ones made of tough ceramic materials laced
with super strong diamond fibers that overcome the ceramic s endemic
vulnerability to catastrophic fracture will be the stuff of new
aircraft that can fly at twenty-five times the speed of sound.  Smart
materials that can respond to external conditions by changing their
color, shape, stiffness, or permeability to air or liquids could become
the stuff of future cities whose buildings are more comfortable and
better able to field sudden violent challenges from earthquakes or
terrorist bombs. Smart materials also could lead to cities whose
infrastructure can sense and even automatically compensate for-the
wounds of corrosion, metal fatigue, age, and the other slings and
arrows of urban decay.

On the down side is that today s wonder materials could become tomorrow
s environmental villains.  It has happened before.  When
chlorofluorocarbons (CFCs) were first invented in the 1930s, their
combination of inertness, non flammability and thermophysical
properties fated them to become materials of choice for technologies
such as refrigeration (which previously relied upon extremely noxious
materials including ammonia and sulfur dioxide), aerosol propellants,
and blowing agents for making lightweight materials such as
styrofoam.

The inventor of CFCs, Thomas Midgely, used to breathe CFCs into his
lungs and then extinguish a candle with a CFCrich exhalation to impress
audiences with the material s benign nature.  Over the course of
decades of ever higher volume use, however, CFC molecules had been
meandering into the stratosphere where, it turns out, they trigger
reactions that destroy ozone molecules.  Ozone molecules absorb much of
the sun s life unfriendly ultraviolet radiation.  So what were once new
wonder materials have become one of the century s most prominent
chemovillains.

Today s materials researchers are far more aware of the Promethean
catch that new materials can harbor, having learned from experience
with CFCs, lead-based paint pigments, asbestos insulation, and many
other materials.  The challenge for today s researchers is to use their
ever more sophisticated knowledge base to more accurately predict the
balance of potential benefits and costs new materials will bring with
them.

In the words of Cyril Stanley Smith, the late materials historian and
Manhattan project metallurgist: Materials themselves have interacted
with mainstream history, for they are the stuff on which virtually all
human activities are based.  As materials researchers discover and
create ever more capable materials, human aspiration and achievement
will soar into new places.

PART I

Trial and Error

Imagine that every material thing in your life suddenly disappears.
Your car.  Your home.  Your clothes.  Your jewelry.  Everything, every
thing, is gone: the steel, concrete, glass, plastic, and thousands of
other materials that made up your modern habitat are nowhere to be
found in ready to-use form.  Naked in the wilderness, you are going to
have to figure out how to transform the raw materials of the world into
forms suitable for making things.  Your face contorts with the anguish
of profound puzzlement.

Your mind focuses on your first goal: covering your nakedness.  The
wilderness becomes a primordial hardware store with resources in the
rough.  The first material you consider for making clothes is some
birch bark, thin and peeling from its tree.  Stitch a few patches of
this stuff together, and just maybe ... But when you get hold of the
bark (whose oil you might later discover to be helpful in starting
fires), you find that the bark is too stiff.  Besides, you have no idea
what you could use to stitch pieces together.  Won t work.  Next, you
catch sight of some large leaves, green and supple.  But you find that
they rip too easily.  Even if you finessed them into a loin covering,
they would rip apart with your first step.  Two trials, two failures.

As you scan the landscape for still another suitable material, a deer
streaks across your path.  Its beautiful hide seems to scream, Wear me.
Notwithstanding the animal rights movement of your former world, you
want nothing more than to hunt down that deer and make clothes out of
it.  You have no gun, no bow, no arrows.  Stones come to mind, and
sharpened sticks.  To find them you scour the forest floor for stones
that have the right feel.  They must be light enough to hold and throw
but heavy enough to stun a deer.  You try hundreds of stones and end up
with a dozen that seem good for hunting.  As for sticks, most are
rotten, and many break too easily when you bend them.  Some are strong
enough, but they are too green to sharpen.

Next to a rotting log, however, lies something far more promising.  It
is an antler shed by a young buck.  Its tips are sharp, hard, and
strong; all at once you intuit the possibilities.  If you could find a
way to haft one of the straighter prongs of the antler onto a wooden
shaft, you would have yourself a nice spear.  The thought pleases you,
but your confidence dips again when you acknowledge that you must still
cut the antler into suitable pieces and also locate a good shaft.  A
sharp-edged stone might work for sawing through the antler... And so it
would go on.  For each task of survival at hand, you would have to find
the kinds of materials, technologies, and techniques suitable for the
job.  It would be a painstaking exercise of trial-and-error, perhaps
punctuated by serendipitous godsends like finding a perfectly straight
antler tip or a nature made stone blade.  By enduring enough trial and
failure, you would learn crucial lessons about the materials in your
environment.  Gradually, necessity transforms you into a craftsman.

Trial and error and accidental discovery: these are the procedures by
which for the last 2.5 million years, humanity and its hominid
progenitors slowly have acquired the skills needed to transform raw
materials into stuff for constructing things-for the first makers of
stone tools; for early metallurgists who learned to combine copper and
tin ores to make more workable and harder bronze alloys; for the
Samurai sword makers, who learned and then ritualized the
transformation of iron into the superior steel edged weapons that made
the Samurai culture the stuff of legends; for the glass makers on the
island of Murano, next to Venice, who happened onto just the right mix
of ingredients and just the right furnace designs to make glassy
materials with an elegance and unprecedented transparency for which the
island become famous five hundred years ago.

Alchemists also practiced trial and error-with perhaps a greater
emphasis on error-in their obsessive search for the philosophers stone,
the esoteric preparation they believed would transform lead into gold.
In one of the most notable instances of trial and error applied to an
impossible goal, they putrefied, boiled, distilled, solidified, and
sublimated all kinds of mineral and organic ingredients in every
possible sequence.  How could these eccentric tinkerers, Isaac Newton
among them, have known that nothing short of nuclear physics in huge
accelerators and reactors could pull off such a fundamental
transformation?

To this day in a scientific age, trial and error remains a cornerstone
in the quest to discover and invent ever more capable materials for
making things faster, stronger, more durable, and otherwise better than
before.

One

EARTH, Air Fire and Water

One day about 2.5 million years ago, a protohuman living in eastern
Africa s Rift Valley made a momentous discovery.  No one knows the
details, but it could have gone something like this:

During a search for rocks suitably sharp-edged for scraping meat from
bones, this hominid-either an Australopithecus or Homo habilus whom we
shall call George-picked up a hunk of dark brown flint by a chalky
cliff face.  Finding it too dull, he threw it down, an exercise he and
his kind had repeated a million times before.  This time, however, it
was different.

The loud report of the stone hitting another stone drew George s
attention.  Right there at the point of impact lay just what he had
been looking for-a sharp-edged flake the size of his palm.  As George
stared at this unbidden gift, his brows furrowed.  A novel shiver of
insight swept over his nearly human brain.  Seizing another piece of
flint, he once again struck rock against rock, and more sharp-edged
flakes laid themselves out in front of his discovery-widened eyes.

In one stroke George unwittingly launched the Paleolithic Age (the Old
Stone Age) and invented materials engineering.  Materials engineers are
the people who figure out how to make the estimated fifty thousand
materials that compose the modern industrial landscape, including the
silicon crystals of computer chips, the ultrapure glass strands of
optical fibers, the nickel based super alloys of jet engines, and the
synthetic rubber of car tires-none of which are found lying around in
the wilderness.  They are made the way George made his first artificial
stone tool: by transforming the raw materials of the world into new and
more useful forms.

Before George s discovery technology had always been the art of using
things pretty much as they were found.  A stick from a flat-topped
acacia tree was good for getting at ants underneath the savanna s hard
earth.  A heavy, egg-shaped lava stone was good for hammering open
wildebeest bones to get at the marrow-rich interiors.  Left intact, the
same bones were good for beating the daylights out of the living flesh
of a rival protohuman.  The Old Stone Age, which lasted until about
fifteen thousand years ago, certainly also was the age of wood, bone,
hide, leaf, shell, hoof, horn, tendon, tooth, tortoise shell, claw,
vine, bark, sap, and of any other easily accessible material that
proved useful in its raw form.

George, our Paleolithic Edison, realized that hominids like him did not
have to be satisfied with the world as they found it.  Bashing one rock
against another can get you a sharp cutting edge where there wasn t one
before.  And with a ready supply of sharp stone tools, George could
scavenge more meat from recently killed prey in the little time he had
before other scavengers like hyenas arrived with their own lust for the
same meat.  Stone flaking meant larger and tougher skinned animals
could be butchered, which made more of the world habitable.  Early
stone technology helped open doors out of Africa to Europe, Asia,
China, and eventually to North and South America.  George s discovery
was worthy of a special Nobel

Prize given every few hundred thousand years: a lifetime of premium
flint to the winner.

The technology of stone flaking, dubbed the first technology by some
paleoanthropologists, probably spread quickly throughout George s own
clan, and then more slowly to other clans.  Toolmakers found that
certain ways of hitting stones together produced more useful shapes and
sharper edges; they found that materials like flint and obsidian
(volcanic glass) were better suited to making tools than crumbly pumice
or overly hard granite.

The clacking of stone against stone was the sound of high technology
for so long that George s descendants had enough time to evolve into
modern humans.  Ice ages came and went; deserts greened into forests
and dried back into deserts; entire animal and plant species emerged
and disappeared; but the stone tools themselves didn t change all that
much.

During all of this time, there were, of course, other Paleolithic
Edisons.  Hominid stone workers at various times and places came up
with variations on the original theme.  But that theme hardly changed
in 2.5 million years: start with a suitable stone blank and remove bits
and pieces until you end up with a scraper, a hammer, an ax, a blade,
or some other tool.  For some 124,000 Paleolithic generations,
innovations in the use of materials were more or less footnotes to
George s happy accident.

During this epoch the effect of materials engineering on the look of
the landscape remained minuscule.  Wherever they went and lived, the
toolmakers littered the landscape with their stone hammers, scrapers,
choppers, hand axes, piles of flint debris, and the bony remains of
their meals.  Like materials engineering, litter is a Paleolithic
invention.  But this litter was small scale.  Even today s best spy
satellites, the ones that can read a license plate from orbit, would
have stood a slim chance of detecting signs of the bipedal creatures
who were slowly expanding their dominion by devising ways of
engineering the stuff of their world.

Human exertion can fracture a stone along weak internal pathways that
developed, like biological traits, during the stone s geological
genesis.  When Paleolithic hominids did this skillfully, they got nice
sharp flakes or beveled edges, all the better than muscle and teeth for
butchering and other daily chores.  But God-given anatomy alone has
never allowed a hominid or human to reach into the microscopic
interiors of raw materials and rearrange their chemical bonds (leaving
aside the miracle by which digestive systems transform food into living
body tissue, biochemical energy, and excrement)-to convert clay into
pottery, ore into metal, or sand into glass.  That requires an entirely
different kind of energy.  It takes fire.

The harnessing of fire s transformative power may have begun in a place
like the limestone Zhoukoudian cave, west of Beijing.  The cave was
first occupied by Homo erectus clans seven hundred thousand years ago
and was still in intermittent use five hundred thousand years later.
There, in a cavern almost two football fields long and fifty yards
wide, hominids made fires.  The charred remains of bones, burned
stones, and wood from about 460,000 years ago-pile twenty feet deep in
some places.  Ironically, using fire for material transformations
apparently did not occur to any of the Homo erectus denizens of the
Zhoukoudian cave, whose use of flame was apparently confined to cooking
and obtaining warmth, light, and protection from other animals.  By the
time the cave was abandoned, however, fire had become part of the
hominids daily experience.

About twenty-five thousand years ago, the modern humans that evolved
from the first fire-using Homo erectus definitely had discovered what
fire can do to the stuff around them.  One of the oldest clay objects
known to have been deliberately hardened by fire was found in what now
is Vestronice, Czechoslovakia.  It is a female figurine-an archaic
Venus.  Formed first out of wet clay, she was endowed by her sculptor
with exaggerated sexual anatomy.  She was a variation on a widespread
figurine theme,

carved in ivory and stone, that had extended back at least a few
thousand years.  But by placing the soft and malleable clay form into
the heat of a fire, the archaic artist became an accidental chemist. He
or she caused the mineral particles of the clay to transform chemically
and physically and bond into a hard and durable material. Clay became
pottery.

Utilitarian need was not a likely impetus for making fired clay
figurines like the Venus of Vestronice.  It seems that the use of fire
to change materials into new forms emerged from a religious or
aesthetic sensibility.  As the late Cyril Smith saw it, necessity has
only recently become the mother of invention.  Ever since the Venus of
Vestronice, he wrote, the first appearance of almost all artificial
materials and of treatments to adjust their properties or shapes to
specific uses appear first in objects of art, or at least in objects of
ceremony in which utility resides in their aesthetic overtones.

For millennia before and after the making of the Venus of Vestronice
and similar prehistoric figurines, the aesthetic sensibility must have
been infecting the entire human population like a benevolent virus.
Across Europe and Russia artists were carving bone and stone into human
and animal forms.  In South Asia they were engraving ostrich shells,
which they may have used as containers, with simple geometric patterns.
In Australia nomadic hunter-gatherers were painting and engraving
rocks.  In caves in Namibia, Spain, France, and elsewhere, they laid
down images of horses, rhinos, ibexes, bisons, and themselves.

In their quest for color, early painters got their black by mixing spit
with charcoal from their fires or with black minerals like manganese
dioxide, which was easily accessible.  White may have come from kaolin,
a chalky clay mineral that much later helped craftsmen in China earn
worldwide envy for their fine pottery and porcelain.  Colorful pigments
came from red ocher (hematite that contains iron) and green, copper
bearing malachite.  The same atoms that gave these minerals their
color-iron and copper, respectively-later became sources for another
kind of fire-born material: metal.

The agricultural revolution that began about ten thousand to fifteen
thousand years ago marked the end of the 2.5-million-year-long Old
Stone Age.  What was once high technology-chipping stones into
tools-was no longer sufficient for new tasks like sowing the land.

There still were no cities or vast agricultural regions forming grids
like huge sheets of graph paper.  There were no major roads, pipelines,
or aqueducts, whose more geometric lines and contours bespeak human
design rather than natural causation.  All such advances were still
thousands of years away.  The human mark on the world remained
intermittent and scant, hardly noticeable amidst vast, untrammeled
expanses.

Agricultural life, however, meant that more people could live in a
smaller region longer than ever before.  In this more settled society
people could assume more specialized roles.  They could observe one
another s skills and innovations far more often.  There probably was
more opportunity than ever before to communicate, commiserate, and
search for solutions to problems.

The variety of materials exploited by the early agriculturists most
likely expanded at a pace unseen since George invented technology in
the first place.  People were beating, scratching, grinding, and
heating materials to reveal previously hidden possibilities of the
things around them.  Flake flint and you get a knife.  Grind up
malachite and you get green pigment.  Bake clay and you get pottery.
What next?

With the advent of agriculture, the technological limitations of stone
must have become irritatingly evident.  Granted, stone was still
suitable for making grindstones, mortars, and pestles the tools for
making real food out of grains.  It also was still the best stuff for
small knives, scrapers, axes, weapons, hammers, and other tools.  But
stone was not a very good mate earth.  air.  fire, and water rial for
the rapidly growing need to store, contain, and protect seeds,
harvests, and other resources.  It was hopeless for transporting
water.

For these tasks early agriculturists did have at least some material
options.  They could shape, carve, sew, or weave hides, gourds, wood,
bark, and reeds into containers and baskets.  But like all materials,
all of these had pros and cons.  Even if the raw materials were
available in sufficient quantities, the resulting implements were
neither waterproof nor insect-resistant.  They could get wet and moldy.
They could crush and break easily.

Perhaps one visionary saw it.  More likely many individuals or groups
in various places at various times must have joined in the proverbial
Aha!  as they made the connection between clay, fire, and the rocklike
pottery utensils and vessels that could emerge from these two
constituents.  Unlike the earlier figurines, the fired clay objects
born of this insight were not merely objects of art and ceremony: they
were useful in everyday life.

Pottery could be watertight.  Insects could not enter pottery vessels
as readily as they could more porous baskets.  The stuff could be made
as long as there was clay, water, and fire.  This was one of those big
ideas that helped initiate a new era of human history: the Neolithic,
or New Stone, Age.

One route of pottery into routine life might be reflected in ancient
pieces of fired clay found in Gambles Cave in Kenya.  These pieces have
wicker like imprints on them.  One explanation for the imprints is that
some clay accidentally got smeared on the outside of an ancient basket
that ended up in a fire.  Or maybe the basket maker used some mud to
strengthen the wicker weave, to waterproof it, or at least to keep
seeds from spilling out of crevices.  Or perhaps someone did in fact
smear the mud on the basket with the intention of making a pottery
vessel.

Whatever the details and wherever the place, the discovery that fire
could transform clay into useful pottery must have been a happy one,
like a huge freshwater lake looming before a desperately thirsty hiker.
Clay, a kind of earth, was readily accessible to early farmers,
particularly along the wet and silty banks of rivers that made the land
of the early agriculturists so fertile.

Suddenly there was a seemingly endless supply of the vessels that were
increasingly vital to the agricultural pioneers.  Unlike any other
material humanity had known, pottery s main ingredient, clay, was
easily molded in all three dimensions using nothing but hands and
water.  Once formed, the soft clay shapes could be heated into hard and
useful pottery implements.  A new material, pottery, was central to the
galloping success of the agricultural societies.

Like a fire in a drought-stricken forest, pottery spread quickly from
its various origins.  Soon fired clay pots, bowls, vessels, pipes, and
other items for cooking, eating, storage, and hauling became so
prevalent that their remnants have become the ancient paper trails by
which historians and anthropologists have re-created who went where and
when.

As ever more people began working with clay and fire, moreover, a form
of positive feedback kicked in to quicken the pace of innovation with
materials.  In different places early potters dug differently shaped
hearths and used different fuels (wood here, dung there) and different
raw materials.  The result was a diversity of pottery types: different
qualities, styles, colors, strengths, and porosities.

Pottery, useful at it was, also had its flaws.  Those same broken
shards of pottery that have attracted armies of scholars and museum
curators are testimony to pottery s great limitation: knock a pottery
bowl and it chips; drop it and it shatters.  The cringe on a
contemporary face witnessing a teetering plate or glass surely mirrors
the pained expression of an early Neolithic potter beholding yet
another shattered masterpiece.

Pottery s fragility arises from the way the countless mineral particles
that make up clay become a single pot by way of fire.  Each clay
particle is a minuscule pebble-a tiny grain that may be a single
crystal or an interlocking conglomeration of many separate mini
crystals  In a hot enough fire, the grains sinter, which means that
they bond together without melting.  The grains do not, therefore, weld
together seamlessly without breaks.  Instead, they end up sporadically
stitched together via chemical bonds between adjacent grains.

The result can be and often is quite strong amidst compressive forces.
Four high-quality Wedgewood teacups can support a tank.  But woe to
those who put a great deal of tension on pottery vessels or teacups:
such tension entails a force that pulls on the chemical stitching
holding the grains together.  That happens when things fall or are hit
or bent.  When a plate falls off of a table and hits the floor, some of
the energy of the impact pulls on the grains of the plate, instantly
opening tiny cracks between grains, which then pry apart more and more
grains at an accelerating rate.  When cracks reach from one end of the
plate to the other, pieces of the plate fly off.  When many cracks
proceed like this simultaneously, the plate shatters.  Some of the
destructive energy ends up as airborne vibrations-the awful sound of
breaking tableware.

Had metal been available, many of the early potters may have become
smiths instead.  Unlike pottery, metal is shiny, reflective, and
smooth.  Often cool to the touch, it doesn t break so easily.  Its
internal anatomy is not quite so rigid as the inside of a ceramic
material.  Most familiar metals dent and bend before they shatter. They
are strong both in compression and in tension, which is why steel
cables and beams can hold up massive structures like the Golden Gate
Bridge and the Empire State Building.  Heat metal and you often can
beat it into sheets, foil, or, in some cases, metallic leaf far finer
than tissue paper.  A cubic inch of heated platinum can be drawn into a
wire so fine that it would wrap around the world twice-more than 52,000
miles.  Heat metals some more and they melt, which means you can pour
metal into molds and cast all kinds of intricate shapes out of them.

The seeds for the Ages of Bronze and Iron, much like those for the New
Stone Age, were always lying around riverbeds, in eaves, and in the
bush.  During their thousands of wandering generations, even the most
ancient of Paleolithic hominids must have happened upon eye catching
nuggets of copper, silver, and gold.  Some may even have found
meteorites rich with a nickel iron alloy, the sort of metallic mix that
later became the material backbone of an industrial revolution, even of
superalloy jet engine parts.

Like the Venus of Vestronice, the first use of metal was likely
confined to embellishment or ceremony.  One of the oldest known metal
objects worked by human hands is a tubular copper bead, probably part
of what was at the time an exotic object-a necklace made out of
metal.

Metal remained a rare sight in Neolithic times because there was not a
lot of it just lying around ready to use and people of that epoch had
no clue that rocks-even rough, dull ones-harbored massive supplies of
shiny, lustrous metals whose properties are as different from stone as
stone s are from wood and flesh.

The discovery of smelting metal from ore most likely was a fateful
accident.  Perhaps it occurred like this: while warming themselves by a
fire on a night with strong, steady winds that could intensify the
flames, some Neolithic herders may have noticed that a searing reddish
liquid was bleeding out from the center of an impromptu cooking furnace
that they built using some greenish malachite, a common copper-bearing
rock.  The combination of fiery heat and the carbon from the burning
fuel was enough to liberate the metal hidden in the green stone.

Unbeknownst to these inadvertent smelters, carbon atoms were
infiltrating microscopic pores and openings in the rock and absconding
with oxygen atoms that used to hold the metal atoms in a vice like
oxide grip.  Copper atoms plus oxygen atoms become copper oxide
molecular structures.  In sufficient quantity such molecular structures
will scale into tiny grains that geologic processes can coalesce into a
rocky hunk of ore.  Oxidebased copper ores minus oxygen equals copper.
Sulfide-based copper ores minus sulfur also can equal metal, but not
without a rotten-egg odor.  Little red rivulets of molten copper might
have hardened into the contours of the ground before the herders eyes.
If the clay-to-pottery transformation was magnificent, the ore-to metal
transformation was divine.  To smelt copper from stone was to make
stone bleed.

So dramatic is the change from ore to metal that people around the
globe attached spiritual, magical, and religious significance to it. If
rock could become metal, then why could not copper or lead become gold?
The alchemical quest was born.  Well into the eighteenth century this
magical component of metallurgy beguiled even Isaac Newton, one of the
greatest scientists and most learned alchemists of all time. Newton
wrote over one million words on the topic of alchemy, whose
practitioners sought, among other things, to transform less magnificent
materials such as lead into what many people still consider

5 to be the most magnificent material of all, gold.

The initial era of metallurgical discovery unfolded between eight
thousand and four thousand years ago.  Some of the earliest known
metallurgists lived in the Zagros Mountains, a desolate and rocky area
stretching from what is now western and central Turkey to the edge of
the central desert of Iran.  The land there was not so suitable for
agriculture, but it was rich with metallic ores.

The use of metallic materials bloomed into diversity in much the same
way that pottery had.  People in various places worked with various
ores and smelting techniques.  The early metallurgists must have been a
doughty but sickly breed.  They had to endure the dark and stale air of
mines; they breathed the poisonous and foul-smelling fumes of roasting
ores; and they withstood searing heats from sun-bright furnaces and the
back-breaking labor of turning raw metal into useful and pretty
implements.  With the advent of smelting, feverish experimentation with
ores, woods, and charcoals, and with furnace designs, blowing devices,
and clays, was carried out, is how one historian imagines metal
technology to have swept the world.  Cyril Smith envisioned this fever
to have played out this way: Practically everything about metals and
alloys that could have been discovered with the use of recognizable
minerals and charcoal fires was discovered and put to use at least a
millennium before the philosophers of classical Greece began to point
the way toward an explanation of them.  In other words, metallurgical
innovation took place well before anyone had ever thought that all
materials, including metals, are made of atoms.

The copper beginnings of metallurgy did not raise the technological
stakes quite enough to inaugurate an historical age the way bronze and
iron metallurgy eventually did.  Copper is soft and malleable, fine for
ornaments and vessels.  It is not so good for weapons or for making
rugged tools.  In a mold of fired clay or stone, molten copper shrinks
as it cools, and it absorbs gas, which is why cast copper objects often
have pits, bubbles, and other imperfections that weaken the product.
Moreover, copper ore was not so easily found in the days before mining
was common.  Its rarity initially made it valuable and expensive, the
stuff of the rich and famous.

Red copper combined with whitish tin, an even softer metal, yields
yellowish bronze.  Bronze is a metallic alloy that is unlike either of
its components: harder than copper or tin, it dLn hold shapes better
than either component, including sharp edges.  Bronze is also easier to
work with than its ingredient metals because its melting temperature is
lower than either of theirs.  Molten bronze flows more readily than
copper alone, so it casts more successfully.  A material that casts
well and can hold an edge is a material able to equip armies with
harder, sharper, and deadlier weaponry.

earth, air, Jire, and water n 27

The original discovery of bronze was probably yet another
well-exploited accident.  Early copper workers could well have hauled
back to their smelting hearths copper ore that included some tin rich
ore, perhaps the brownish-blackish cassiterite.1 Astute observation
eventually could have yielded the following rough rule for smelting
bronze: add one part of black ore to ten or more parts of the green
ore, and then put the mixture into a draft-blown fire.  The result was
a new material that for the first time was strong and tough enough to
replace many uses of stone, which persisted as a staple of the tool
chest.

Because tin-bearing ore was even scarcer than copper ore, bronze
initially was rare and therefore went into making priority implements:
weapons.  It was not a material for a commoner.  For its pivotal role
in making bronze, tin became a militarily and politically strategic
material.  To secure supplies for their military metallurgists, Greek
and Phoenician seamen as long as four thousand years ago undertook long
sea voyages to the southern coast of Britain, where the ground was rich
with tin ore.

Roughly at the same time, Chinese metallurgists were doing the world s
most sophisticated bronze work.  Their bronze often contained 15
percent tin and 85 percent copper, the best proportion for casting
bronze: it produces an alloy three times harder than copper.  Starting
with the ore itself, precision mixing of the components is nearly
impossible since the metal content of even two pieces of the same ore
can vary greatly.  So the Chinese bronze workers must have discovered
their 15 percent solution by smelting copper and tin separately and
then melting the two metals together in many different ratios until
they found a ratio that worked best-trial and error at its best.  With
their optimal bronze alloys they cast beautiful objects like ceremonial
bells with ornate designs and inscriptions first imprinted as negatives
into the fired clay surfaces of the casting molds.  One pyrotechnical
product, pottery, had become an imPortant means of processing
another-metal.

As more sources of metal were found and as mining became more
widespread, bronze eventually became more commonly available, though
rarely in high volumes.  As a class of materials, metal remained
generally exotic.  Most people were surrounded by things that could be
made without using fire.  Wood, stone, mud, thatch, hide, wool: these
were the textures that filled the landscape.

Iron was the first metal for the masses.  As far back as five thousand
years ago, metalworkers must have somehow gotten their furnaces
especially hot on a day when they had charged the furnace with some
iron-rich rock, maybe blood-red hematite.  Perhaps it was a consistent
blast of salty wind sweeping into their furnace from the sea that made
the difference.  From this they obtained a silver-gray metal that they
had never seen before.  It was iron.  One of the oldest known iron
implements is a dagger blade from 1350 B.C. It must have been a highly
prized item because it was found buried in the mummy wrappings of the
entombed Egyptian pharaoh Ritankhamen."

Iron eventually became everyman s metal because its ore is widespread
and plentiful: more soldiers could be equipped with it than with
bronze, and iron tools were more readily available to farmers, potters,
clothes makers, and other workers.  Iron democratized metal.

The down side of iron is that wrenching it from ore requires a
temperature hundreds of degrees higher than that needed to extract
copper or tin.  In fact, the melting temperature was too high for
ancient furnaces, so early iron-workers got their metal by beating it
out of a solid bloom, a conglomeration of metal and rocky slag that
forms when the ore is heated to temperatures high enough to loosen
things up in the ore but not high enough to melt iron.  Only through
sweaty, exhausting ear wracking work with hammers could a smith
consolidate the bits of metal that would form in the bloom like raisins
in a rocky earth, air, f?,re, and water E 29

pudding.  It was then the blacksmith s job to reheat and beat the
consolidated iron into useful things.

The earliest iron turned out no better than bronze.  Often it was
worse, especially for weaponry.  Compared to copper and bronze, it was
a slightly crazy metal, with unpredictably varying qualities, so a
smith never quite knew what to expect from a finished iron implement or
weapon often they came out too soft or too brittle.

The reason for this fickleness is the special role that carbon atoms
play in the properties of iron based metals, including steel.  No one
knew in the early days of iron that carbon atoms from the burning fuel
in their bloomery and smithing furnaces were actually becoming part of
the metal s internal anatomy and changing its properties.  In the
smelting of copper or tin, carbon s role is to remove oxygen without
remaining in the metal.  The iron that smiths wrought from the blooms
and then formed into weapons and implements absorbed varying amounts of
carbon.  Too little carbon resulted in a consistency closer to that of
pure iron, which is soft and malleable.  A man wielding a pure iron
sword would be no match for an opponent swinging a bronze sword.  Too
much carbon in the iron, however, yielded a brittle metal that could
shatter like pottery.

Most often smiths ended up with iron containing somewhere around 1 to 3
percent carbon by weight.  When lucky, they would get somewhere between
.1 and 1 percent carbon.  When that happened, the metal underwent a
dramatic personality change.  It became much harder and tougher, and it
could hold a much sharper edge.  When a smith managed to keep the
carbon content within this range, he made steel.

By around 1500 B.C. the Hittites of Asia Minor and Syria had developed
a sizable iron industry.  They were making enough to equip entire
armies, not just their elite commanders, with iron weaponry of renowned
superiority.  Perhaps their smiths figured out how to convert iron
consistently into steel.

Steel or not, their metallurgical advantage translated into military
success.

Much later, by about the seventh century A.D."  Samurai sword makers in
Japan did master the art of steel making as had their counterparts in
Damascus, Syria.  Their success was passed down from master to
apprentice in precise, secret rituals that were the precursors of the
exact physical and chemical measurements and process control with which
modern steel producers insure uniformity in their products.

These sporadic success stories notwithstanding, steel making remained
an elusive and tricky art.  Steel did not emerge from its status as a
specialty small-volume metal until the middle of the nineteenth
century, when it first became available in tons rather than in
pounds.

Aside from turning clay into pottery and wrenching metal from rock,
fire could also transform mineral sand into glass, the third of the
ancient trinity of materials that were born of flame.

Obsidian, nature s own volcano-made glass, had long been a coveted
material, probably well back into the Paleolithic age.  Natural glass
forms when volcanic heat melts silica-based rocks into liquid that then
cools too rapidly for the atomic constituents, to reorganize into the
tiny crystal grains that comprise a lot of stone and rock.  There are
too many light-absorbing metal atoms and too many light-scattering
defects in a piece of obsidian for it to be transparent.  Instead it is
translucent and usually of a dark brownish or green hue.  No other
material could be made as sharp-edged as obsidian.  But it was
relatively rare, and its edges were easily chipped and broken.

The first people to emulate the volcanic invention of obsidian to make
glassy materials using fire probably lived somewhere in Egypt about six
thousand years ago, in the heyday of copper.  Atop small pottery beads
they applied a coating of colored mineral dust.  The heat of the fire
then melted earth, air, fire, and water w 31

the dust particles enough for them to fuse into an unbroken and
uncommonly smooth coating-a shiny and colorful glaze.

Among the oldest entirely glass objects known are opaque glass beads,
probably 4,500 years old, that were found in Iraq

14 (then part of Mesopotamia) and elsewhere in the Near East.  In the
days of the pharaohs, Egyptians and Mesopotamians had developed a type
of glass craft in which they wound taffy like coils of hot glass around
a clay core.  They then chipped out the core to leave behind hollow and
beautiful glass vessels.

The exact circumstances of the original discovery of glass making
recedes into the same historical mist that shrouds the discovery of
pottery and metal extraction.  There are only stories like this one
from Pliny, a first-century Roman historian.  He credits Phoenician
sailors with the discovery.  As he had heard it, the sailors landed on
a sandy beach, where they built a cooking fire.  To heat a cooking pot,
they propped it up on some blocks of natron, an alkaline material
derived from the ash of plants (and from which the chemical symbol for
sodium is derived).  The sailors found that the combination of heat,
sand, and natron produced a hot liquid stream of an unknown translucent
liquid that cooled and solidified into glass.  As in glass making
today, an ingredient akin to natron serves as a flux that sops up
impurities and lowers the melting temperature of the mix.

Whatever its origin, glass was not a substance for the masses.  Like
gold, the sky-blue lapis lazuli, and translucent white alabaster, glass
was the stuff of Pharaohs, high priests, and nobles.  What helped bring
glass into the realm of common materials was the invention of
glassblowing by craftsman in Syria and Palestine about two thousand
years ago.  By blowing air into a dollop of molten glass gathered at
the end of an iron blow pipe, the workers could make beautiful, hollow
glass vessels.  Within a hundred years Roman glass making was big-time
commerce, even selling glass gladiator-beakers for patrons of the
games.  These were ancient souvenir cups with imprinted scenes of
gladiatorial battles transferred from the ceramic molds into which the
Romans blew the molten glass.

The architectural potential of glass s translucence was not realized
until about the twelfth century with the construction of Gothic
cathedrals.  Intensely colored, backlit portrayals of biblical stories
bathed praying congregations, enhancing the sacral magnificence of the
massive stone cathedrals.  In the following centuries transparent glass
made from purer silica-based ingredients began replacing the dark
wooden shutters and translucent oiled paper or muslin that kept the
climatic elements out of the typical home.  By the mid-fifteenth
century, half of the houses in Venice, Europe s most influential glass
making center, had glass windows.  Two and a half centuries later, the
transition to glass in Europe at last was nearly complete.

Some historians credit the added illumination made possible by glass
with heightening interest in cleanliness and hygiene.  Windows made
dirt more visible.  And thanks to superior mirrors- made with
transparent glass that reflected properly from the thin metal foil on
one side-people could see and understand themselves and their
conditions more accurately than ever before; glass, a miraculous
substance that is at once as solid as a rock and as invisible as air,
shed as much light on people s minds as on their surroundings.
Moreover, the magnifying powers of glass eventually enlightened
scientists as well, enabling them to understand what it is inside of
materials that makes the stuff of the world the way it is.

By sheer volume it was another less lofty set of pyrotechnical
materials-cement, concrete, plaster, brick, and mortar-that changed the
look of the world for ordinary people far more than metal, pottery, and
glass.  Like all materials used by people, the arrival of each of these
had its own discoverers or inventors.

Plaster, for example, goes back at least eight thousand years.  In what
was Anau, Iran, a plaster based on gypsum (a earth, azr, fire, and
water 0 33

crumbly mineral that when pulverized and mixed with water can first be
spread before it hardens) was becoming, as one historian has put it, a
trademark of the Iranian plateau that has lasted right into the
twentieth century.  Likewise lime, or calcium oxide, a. powder made by
burning limestone and that would much later become a primary ingredient
in steel making was formed into an early plaster that ended up on
floors in ancient Turkey.  In the sixth century B.c. bricks, thousands
upon thousands of them, were becoming the architectural trademark of
Babylon, one of the great cities of antiquity.  The builders of
Babylon, and of the rest of ancient Mesopotamia, formed wet clay into
bricks and then fired them to hardness in kilns.  Rome, too, became a
city of brick after much of the city burned to the ground in A.D. 14.
After that, brick became a universal building material for urban
residential structures as well as for imperial buildings.

The Romans pioneered the use of hydraulic, or water

cured, cement.  Its unique chemical and physical properties produced a
material so lasting that it stands today in magnificent structures like
the Pantheon.  Yet the formula was forgotten in the first few centuries
after the fall of the Roman Empire and wasn t rediscovered until 1824
as Portland cement.  One Roman version was based on a burned mixture of
two major components: volcanic ash-called pozzolana-from Mount
Vesuvius, which destroyed Pompeii and nearby towns in A.D. 79; and
calcium carbonate, the stuff of seashells, chalk, and limestone. Adding
water to these sets off a complex set of chemical reactions that
convert the gritty pasty stuff into what is essentially artificial
stone.  The nineteenth-century rediscovery of Roman cement, the
aforementioned Portland cement, is made from a combination of burned
limestone, clay, and water.  It is the single most heavily used
human-made material on earth.  It is visible everywhere as the flesh of
skyscrapers, dams, bridges, roads, patios, and thousands of other
structures.

In addition to the hard materials made from inorganic or mineral
sources-pottery, metal, glass, plaster, bricks, cement new soft
materials were also finding their way into the materials pantry.  This
history of paper started unfolding in about 3000 B.c. Ancient Egyptians
found that they could pound thin bark peeled from river
reeds-papyrus-into sheets.  It was easier to write on the papyrus
sheets than on stone or baked clay tablets, and the result was easier
to carry around.

Besides papyrus, ancient material tinkerers found ways of mashing,
pressing, heating, filtering, and distilling vegetable and animal
tissues to get substances that were tarry, waxy, sticky, and gluey.
There are plenty of biblical references to materials that served as
perfumes and ointments.  Other historical evidence from ancient
shipwrecks and archaeological sites has revealed the remains of salves,
dyes, glues, waterproofing agents, possibly even antiseptic chewing
gum.

By Christ s time the diversity of materials made and used by the world
s modest population would have astounded George, the Paleolithic
inventor of the first material technology.  The transformative power of
fire was raging throughout the globe as people mixed it with earth,
air, and water to create the many substances of their constructed
landscapes.

Two New Stuff in the Old World

In the nineteenth century almost all of the materials people were
making and using had been made and used for several thousand years. But
between ancient times and the 1800s, the scale of production had been
rising.  Animal-, wind-, water-, steam-, and finally electric powered
machinery had been put to ever increasing use to keep up with the
accelerating demand for materials from growing populations. More and
more mines opened.  Smelters, potters, glass makers metallurgists, and
other materials workers were fruitful and multiplied.  Trade in
materials and goods brought the world s People into commercial bonds. 
Production of materials was so intense at times that laws were
intermittently passed to prevent entire forests from disappearing in
furnaces and kilns.

Craftsmanship with materials had long become a matter of national
pride.  So coveted were the secrets of the Venetian glAss houses, which
were sequestered in 1291 on the island of Murano within sight of
Venice, that workers were forbidden to leave the island.  Attempted
escape was punishable by execution.

For some the passion for certain materials was strong enough to inspire
them to try and find out what makes materials tick, to look inside of
inanimate matter in the hope of finding a key to their creation.
Augustus the Strong, King of Poland and monarch of Saxony (located in
what is now southeast Germany) from 1694 to 1735, expressed one of the
more illustrious examples of this curiosity.  And it presaged the kind
of business thinking that would become common in the twentieth century
at places like DuPont and General Electric.  As the materials scientist
and historian David Kingery of the University of Arizona has put it, He
and his court supported one of the first truly modern research efforts:
it was aimed at porcelain manufacture.

The monarch had a passion for the beautiful Chinese and Japanese
porcelains that were flooding European markets in his day.  Ownership
of these ceramic wares was a mark of social stature.  No one in Europe
had a more extensive collection than Augustus, but his interest was not
merely aesthetic: he wanted to create a domestic source of comparable
quality.

To this end he commissioned Count Ehrenfried Walther von Tschirnhaus,
an aristocratic Bohemian trained in mathematics and physics.  He was
known for mineral-melting experiments in which he had used not fuel-fed
furnaces but rather burning mirrors made of high-quality glass lenses
to generate the high temperatures required for the process.

Tschirnhaus first traveled throughout Europe to see what others had
learned.  He then set up a secret laboratory in Meissen, where a hired
staff conducted experiments on many combinations and preparations of
different clays and clay mixtures.  Eventually, Tschirnhaus s chief
researcher, the alchemist Johann Friedrich B6ttger, pursued a promising
lead.  He found that when combined, silica (the stuff of sand, flint,
and certain stones, including quartz) and lime (calcium oxide made by
burning limestone) melted at a lower temperature than either ingredient
alone.  The cooled and dried product appeared whiter and finer than
previous combinations.

Material science has since explained the reason for this result: the
lime serves as a sponge, or a flux, for collecting contaminants in the
silica that otherwise degraded the whiteness, translucency, and texture
of fired wares.  The melting temperature of the mixture was still high
enough (more than 1,350 C) that the Saxon researchers needed hotter
burning furnaces than current furnace designs could muster.  So they
devised them.

The effort paid off.  Tschirnhaus and B6ttger developed a new European
porcelain with the white body, translucency, and sonorous ring of the
Chinese wares, as Kingery, whose students repeated some of the Meissen
experiments, describes them.  The ceramic wares became famous, coveted
and valuable throughout Europe.  The composition and micro structure
are similar to some of the spark plug porcelains developed during the
early years of high-compression automobile and aircraft engines,
Kingery notes.

Another pioneer of the early eighteenth century, Josiah Wedgewood,
whose name is still synonymous with world-class ceramic ware, followed
the Saxony example of systematic process development.  One of his
innovations, developed in the early 1720s, comprised a series of
ceramic cones made of standardized clay mixtures.  By measuring the
shrinkage of these cones as the furnace temperature increased, he had
devised a crudely calibrated but effective pyrometer for measuring the
high temperatures inside a furnace.  The measurements, in turn, helped
him attain an enviable level of quality control.  Before that,
temperature control in a furnace had been entirely a matter of
professional judgment, typically involving the withdrawal and
examination of samples from the furnaces during filings.  It was like
baking a cake in an oven with no temperature dial.

Neither Bbttger nor Wedgewood knew exactly how, but their careful
process control was working because it was coercing the inner anatomy
of their products into favorable material Structures.  Ultimately what
makes steel steel and porcelain porcelain is the way their constituent
atoms and molecules interact with one another and aggregate into ever
larger structures until they form something you can hold in your hand
and use.  It is like the progression from atoms to molecules to
organelles to cells to tissues to organs to entire animals.  This
multilevel, hierarchical anatomy ultimately determines the properties
of every person and every material.  Getting control over this
hierarchical structure always has been the key to making consistently
good materials.

In the 1720s a French contemporary of Wedgewood s, the chemist Ren6
Antoine Ferchault de R6aumur, reasoned his way to perhaps the most
present foreshadowing of this modern picture of material structure.
He conjectured that steel was made of an arrangement of tiny grains
that in turn had a substructure made of parts that he called molecules,
though his concept does not correspond to the modern definition of a
molecule as a specific arrangement of atoms.  His molecules had their
own substructure, some kind of less than-crystalline jumble of spheres.
On the finest scale he imagined a periodic, or crystalline, stacking of
even finer spheres.  When modern materials scientists see the diagram
of nested structures that Reaumur prepared in the 1720s to illustrate
his ideas, they say, Not bad!  (See Figure 2.)

Still, even for later generations of scientists in the nineteenth
century who knew that the properties of materials and the way they
performed depended upon their inner structures, there wasn t much of a
foundation for probing very far into material anatomy.  After all, the
very existence of atoms, much less how they might assemble to become
this or that material,

was a hotly debated topic until the very end of the century.  Moreover,
there were no laboratory instruments for peering deep enough into the
anatomy of materials to know with any precision what the atoms and
molecules actually were doing or how they were arranged.  These
limitations, however, did not hold up the beginnings of a science of
materials.

new stuff the old world

, CT 0

FIGURE 2

Reaumur s Prescient Picture of Stuff.  Rend Antoine Ferchault de
Reaumur, a French chemist of the eighteenth century, envisioned the
internal structure of metals as a series of nested architectural
levels.  This 1722 illustration of Reaumur s view of the internal
structure of steel is remarkably like the generic portrayal of
materials by modern materials scientists.  The dot marked G represents
a grain of steel as seen by the eye.  Below is the grain s structure as
seen under an optical microscope.  The structure includes smaller sub
grains (M) and spaces.  Below this level, Rdaumur surmised that each
subgrain was itself composed of yet smaller sub grains (P) and voids
with the series, perhaps continuing for several additional levels. 
THE

ORIGINAL IMAGE BY RWMUR IS REPRODUCED IN CYRIL SMITH S HISTORY OF

METALLOGRAPHY AND

IS REPRINTED HERE WITH PERMISSION FROM THE MIT PRESS.

The challenge was daunting.  Even knowing about atoms would barely be a
beginning.  Trying to understand materials by way of atoms and
molecules would be like trying to understand how a society works by
looking only at its members individually.  Understanding societies
requires insight into the dynamics of social interaction.  It is the
same for materials.  Atoms and molecules invariably are a material s
fundamental ingredients, but they are part of chemical and physical
features and interactions over a range of architectural levels that
really constitute a material s set of properties.  Any real science of
materials would have to grapple with this complexity on scales that
range from atoms to bridges.

A founding father of modern materials science came not from the ranks
of nineteenth-century chemists with their vague notions of atoms and
molecules; he was an amateur geologist working in Sheffield, England:
Henry Clifton Sorby.1

Sorby came from a family of cutlers in a town with a long tradition in
metallurgy.  Metals were as much a part of his context as wood is to a
carpenter s or words are to a writer s. His early interest in rocks and
minerals led him to discover that thin flat sections, through which he
could shine light, revealed enough gross anatomy of minerals to
distinguish amongst them.  He also knew about the use of acids to
create ornamental patterns on metal items such as razors or to prepare
metal plates used for illustrations in catalogs.  In the heat of the
summer of 1863, Sorby combined these two seemingly disparate lines
slicing stone and etching with acid-into a technique by which he was
able to show the micro structure of iron and steel as it had never been
seen before.

The process became known as metallography because it was like writing
on metal.  It starts with a cut and polished surface of a metal
specimen, followed by a gentle etching step that uses dilute nitric
acid.  Under a microscope Sorby could see the result as a puzzle like
montage with regions of different geometries and appearances that he
was able to relate to the process by which the metal was made, to the
material s starting ingredients, and to the materials strength,
fracture resistance, and other engineering properties.  Though Sorby s
techniques didn t go much beyond his own home at first, by the late
nineteenth century metallography had became a standard method for
studying metals and for controlling their quality.  It still is.

The technique of metallography provided the kind of evidence needed for
understanding how to produce iron or steel of new stuff in the old
world m 41

consistently high quality.  Armed with that kind of knowledge,
materials research stood on the brink of the modern era; rather than
shooting in the dark, it could now turn the light on first.

Sorby s work represented the beginnings of a genuine science of
materials in which a material s composition, internal structure, and
method of preparation are systematically related to the material s
properties and technological performance.  Even today, these
interrelationships comprise the framework of modern materials
researchers.  But Sorby s metallography was only a key to a
well-secured mansion.  To get beyond the foyer would take many more
keys.

The incentives for looking more deeply into the anatomy of materials
were intensifying in Sorby s day.  Iron and steel already had become
materials of strategic military, economic, technological, and social
importance.  Iron and steel were the lifeblood of the Industrial
Revolution, which began in the eighteenth century.  Building more and
better metal was the key to better railroads, stronger armaments, and
more reliable steam engines to power the heavy machinery in factories
and mills.

The production of raw iron (pig iron) in England, the world s biggest
producer throughout much of the eighteenth and nineteenth centuries,
mushroomed between 1740 and 1850 from under 20,000 tons to about 2.5
million tons.  Abraham Darby, the third generation of a famous family
of iron masters in Shropshire, England, gets credit for the first big
iron structure ever-the aptly named Iron Bridge.  It opened in 1779
over the River Severn in Shropshire.  Steel, even though its basic
composition is almost all iron with a small dose of carbon, remained
expensive to make, even after 1850, when English smelters were
producing about 60,000 tons per year in small batches that a single man
could lift.  Even when pooled, that s enough steel for at most a
handful of modern skyscrapers.

In their new technological roles, iron and steel had to perform under
conditions more stressful than metals had ever

42 n S T 0 F F

faced before.  Never did materials have to withstand such high heats
and pressures for so long as when they became the stuff of steam
engines.  Never did materials have to endure the clack eta-clack eta
metal-against-metal stresses that go with trains on rails.  Never
before did materials have to be as predictable, stable, and
standardized as they did in mass manufacturing using interchangeable
parts.  When metal in the form of rails, wheels, pistons, boilers, and
gun parts failed, there were fatal explosions and derailments.  Death
on the rails was far more common in the nineteenth century, even in the
early twentieth century, than death by airplane is today.

The physical limitations of cast-iron cannons in the Crimean War
(1854-1856)-in which England, France, Turkey, and Sardinia defeated
Russia for the domination of southeastern Europe-were a turning point
in the history of steel.  Those limitations caught the attention of
Henry Bessemer, an

Englishman who already had enjoyed considerable commercial success for
inventing the lead of lead pencils.  The lead was, in fact, compacted
graphite, a carbon mineral whose internal structure is marked by
loosely bound molecular sheets that can slide off easily to mark paper.
But high-volume steel production was about to become his more enduring
claim to fame.

Bessemer had developed a new kind of projectile that could travel
farther than cannon balls because it developed a spin as it left the
cannon barrel.  But the new projectile, which he patented in 1854, also
generated more stress on the cannon s iron.  Napoleon 111, the French
Emperor, was interested and bankrolled Bessemer to work on the
projectiles.  In December of 1854, Bessemer successfully demonstrated
that an elongated projectile could be made to spin when shot out of a
smooth-bore gun, but the French commandant who witnessed the trial
questioned whether the cast-iron guns of the day could withstand the
added stress that the new projectiles imposed.  In his autobiography
Bessemer wrote, with a characteristic lack of humility, that this
simple observation was the spark which kindled one of the greatest
industrial revolutions that the present century has to record.

The standard method of steel production at the time was hopelessly
small in scale and variable in result.  A charge of wrought iron-itself
the product of time-consuming labor-was sealed in a clay crucible with
some charcoal and then heated for up to several days.  If everything
turned out right, the amount of carbon needed to make a steel alloy
would wend its way into the metal to produce about 50 pounds of steel
over the course of ten days.  This laborious method precluded
large-scale steel production for war materiel like cannon and ordnance
or for big structures like bridges and buildings.

Bessemer s breakthrough in 1855 was to drop the standard approach,
which involved making steel by heating carbon-free wrought iron with
carbon-bearing fuel.  He tried to turn the process around by starting
with carbon-rich pig iron and using oxygen in an air blast to get the
excess carbon out.  Pig iron, the raw product of iron smelting, got its
name because molten iron often drained from smelting furnaces down
central gullies and then right and left into rectangular molds where
the molten metal cooled and solidified.  The arrangement looked like a
bunch of suckling pigs.  Along with a carbon content of about 4 percent
by weight, pig iron also contained impurities that varied with the type
and quality of ore.  To turn pig iron into steel, about three out of
every four carbon atoms somehow would have to be removed.

That sounds like a hard task for solid pig iron.  In the molten state,
however, oxygen passing through would readily grab onto carbon atoms.
Moreover, the oxygen would bind at least some of the impurities present
in the melt into oxides that would form a rocky slag.  Slag is less
dense than molten metal, SO it floats to the top, where it is readily
skimmed off.  What s left in the melting crucible, called a converter,
is purer metal.

The combination of oxygen and carbon during the blast produced a carbon
monoxide gas that shot up and out of the molten metal, spewing
white-hot flame and sparks like an artificial volcano.  As expected,
the oxygen in the air blast did form oxides with the silicon,
manganese, and some of the other impurities that he wanted to remove.
The process also liberated an unexpected amount of heat that displaced
the need of fuel for keeping the metal in a molten state.

There was far more to the spectacle than sparks and flames.  The
combined advantages of reduced impurities, lower carbon content, and
the heat generation with no-cost air (as opposed to costly fuel) led to
more than just better iron.  As the carbon left the pig iron, the pig
iron became more and more like steel.  If the air blast lasted about
twenty minutes, the amount of carbon left in the iron could produce a
mild steel that would be as workable as a carbon-free wrought iron yet
much stronger and tougher.  A cannon made out of good steel will not
shatter as easily as an iron cannon.

After months developing machinery that would enable workers to operate
the blowing engines that created the air blasts for the converters,
Bessemer publicly introduced his iron-to-steel conversion process on
August 12, 1856.  The paper he read to the British Association for the
Advancement of Science that day was published in its entirety two days
later in the London nmes.

That the process at that point was far from perfect became apparent to
the many iron masters who licensed Bessemer s patent.  The ores of the
Swedish pig iron that Bessemer used for his earlier oxygen-blast
experiments were essentially devoid of phosphorus.  But the domestic
supply of pig iron had plenty of phosphorus, and all it yielded at the
end of a Bessemer conversion was an inferior batch of metal that would
crack during attempts to work it into implements.  The phosphorus would
always end up between the metal s individual grains, weakening the
bonds holding the metal together.  It looked as though Bessemer s
fifteen minutes of fame was up.

Bessemer eventually decided to make his process work by relying on
imported Swedish iron.  It was a pair of cousins, Sydney Gilchrist
Thomas and Percy Carlyle Gilchrist, who in 1875 found the way to disarm
the phosphorus conundrum by adding limestone both to the firebricks of
the converter and to the iron charge inside the converter.  The
phosphorus reacted with the calcium oxide of the burned limestone to
form a slag that could be skimmed from the metal.  That opened the
Bessemer process to much more of the world s iron ore.

Another problem with the original Bessemer process derived from the
oxygen.  There was enough of it in each blow not only to take out the
excess carbon in the pig iron but also to blow holes in the subsequent
steel.  This led to a flawed product known as burnt iron.  In this case
Robert F. Mushet (who later would develop a tungsten-manganese steel
alloy whose especial hardness extended the lives of tools made of it by
five or six times) came to the rescue.  He discovered that adding an
alloy of manganese and iron known as spiegel to the charge would keep
the oxygen under control.

With the subsequent patchwork by Thomas, Gilchrist, Mushet, and others,
the Bessemer process succeeded in turning steel into a high-volume
commodity.  In twenty minutes a large Bessemer converter could
transform 30 tons-60,000 pounds of pig iron into steel.  Global steel
production throttled upward at an astounding rate: from about 500,000
tons in 1870 to nearly 28 million tons by the turn of the century, by
which time other metal entrepreneurs had developed additional methods
for producing large amounts of steel.

The giant leap in iron and steel production delivered huge supplies of
materials with an unprecedented combination of strength, resilience,
hardness, and form ability  There was a proliferation of steel alloys
with various properties that depended on the different proportions of
their constituent elements.  The special toughness of nickel steels
fueled a naval arms race among European powers who were using it as
armor plating for their warships.  More tinkering brought forth
stainless steel,

made with a good portion of chromium, some nickel as well as other
minor constituents.  An alloy like stainless steel, which was resistant
to rusting and corrosion, was dramatic evidence of metallurgy s ability
to bring forth spectacular new properties in basic materials.  No one
understood how the alloying elements went about changing the metal s
properties, but empiricism has always been a fruitful, if not always
efficient, means of discovery.

Iron and steel became the stuff of big and visible things ships
bridges, buildings, even the Eiffel Tower.  These metals, particularly
steel, were the harbingers of skyscrapers and audacious urban skylines.
In 1874 James Eads used steel to build a giant arched bridge across the
Mississippi River at St.  Louis?  Some 50,000 tons of steel went into
the building of a railway bridge over the Firth River in Forth,
Scotland.  For the enormous suspension cables that would hold up his
Brooklyn Bridge, John Roebling in 1883 could choose steel rather than
the iron that Abraham Darby was confined to a century earlier.  In 1890
the second Rand-McNally Building in Chicago was the first building to
be framed entirely in steel.  Hundreds of ever higher buildings and
skyscrapers followed.  Steel also was showing up in hundreds of smaller
everyday items like food cans, home appliances, automobiles, and road
signs.

For all of steel s social impact, the breakthroughs that allowed mass
production of it owed little to academic science.  Sorby s
metallography, for example, did not become part of standard practice
until late in the nineteenth century-well after high-volume steel
production had become commonplace.

Some rough guidance actually had been available much earlier.  Even by
the time Antoine Lavoisier was developing chemistry into a genuine
science, chemists had learned that the difference between iron and
steel pivoted on a small difference in the amount of carbon present in
the metals.  Pig iron contains about 4 percent carbon by weight.
Wrought iron contains almost none.  Steel contains no more than about 2
percent, usu new stuff in the old world H 47

ally less.  Practical steel makers like Bessemer learned to hit the
light compositions and processes by trial and error, not through
chemical analysis.

That began to change in the later nineteenth century.  In 1870, for
example, The Iron and Steel Institute was founded to promote a more
systematic and scientific approach to making and developing steel.  A
decade later Sir William Siemens developed a rival steel making process
that was slower than Bessemer s but afforded more control over the
steel s composition; he remarked that a taste for science had been
awakened among employers, and men who formerly ridiculed the idea of
chemical analysis now speak of fractional percentages of phosphorus
with great respect.

Nineteenth century steel makers were not the only ones discovering the
practical power of chemistry.  When Sorby was inventing metallography
in the 1860s, the chemical industry was in the throes of a massive
transformation from its alchemieal and proto chemical roots.  It was
evolving from a collection of mostly unrelated, small-scale producers
of specific substances for tanners, glass makers dyers, and other
eraftspeople to a nexus of interdependent producers.  One person s
trash became another s feedstock.

As steel production rose, chemistry began to emerge from academic
obscurity to become an industrial and economic force.  New chemical
dyes for the textile industry were the first big products, some of them
quickly overtaking the traditional dye industries such as the indigo
growers of India.  The synthetic dyes were loud signals of chemistry s
miraculous ability to transform stuff.  Many of the new dyes were made
of chemicals like aniline that were obtained by distilling coal tar.
Coal tar was a thick, goopy, unpleasant product of the gas lighting
industry that would have been a total waste product had it not found a
use in preserving millions of wooden ties during the railroad-building
frenzy of the 1830s and 1840s.

Beautiful, colorful, big-business dyes from black, goopy waste were
just one tip of one of chemistry s icebergs.  With increasing frequency
chemists were happening onto new chemicals.  These, in turn, provided
new chemicals for reacting with already known ones.  This gathering
combinatorial.  momentum spurred a dramatic growth in the catalog of
chemicals and materials.  The diversity of transformations emanating
from chemistry labs rivaled the fantasies of any medieval alchemist.
Solids became liquids, which became gases, which became solids again.
Gunk became crystal pure, and crystalline purity became gunk.  Over and
over again in laboratory glassware, colors, textures, and forms changed
with hallucinatory strangeness, degree, and surprise.

As it always had been, the chemists raw materials were of the earth:
ores, minerals, and clays, pulverized into powders; coal and wood,
which, when heated without burning to the point of decomposition,
yielded smoky spirits that would then cool and condense into tars and
liquids such as turpentine; kelp, cotton, and other vegetable matter
burned to yield alkaline ashes, already pivotal for millennia in
metallurgy and glass making and petroleum (known many centuries earlier
by the Islamic alchemists in the Middle East, where it would seep up to
the surface beckoning for attention), which began to serve up a
chemical cornucopia of fuels, sealants, paving and construction
materials, and feed stocks for polymers.

These materials went into furnaces and electrolytic baths, where heat
or electricity would wield their magic or violence.  In kettles and
flasks, hot liquids saturated with pulverized and dissolved minerals or
extracts of the vegetable world would, with sufficient cooling or
evaporation, leave crystalline salts, sugars, fats, oils, and other
purified substances.  Complex liquid brews could be separated into
purified components by distillations in which various liquid components
would boil off at various temperatures and so could be collected apart
from the others as they separately fled through the top of the
container to a collection vessel.  Kerosene, naphtha, benzene,
turpentine, ether, and many other liquids were obtained from coal and
petroleum this way.

From all this tinkering entirely new categories of materials were bound
to emerge, among them a tacky, bouncy, compliant material called
rubber.  For centuries South Americans had air-cured the milky latex
(sap) of a tree known as Hevea brasiliensis into a curiously elastic,
easily shaped, and waterproof material that yielded bowls, shoe soles,
and bouncy balls for games.  Its South American users called it
caoutchouc, which means weeping wood.  In 1770 Joseph Priestly, co
discoverer of oxygen, got a small sample of South American rubber from
a friend.  Priestly dubbed the material rubber when he found that it
could erase pencil marks with a simple rubbing action.  For the next
seventy years people experimented with latex rubber for medical uses
such as making bandages, stomach pumps, and ice bags and for
waterproofing fabrics and leak prone leather water hoses.

Natural rubber had major shortcomings, however.  When cold, it would
become brittle and crack.  When hot, it grew sweaty, flimsy, and gross.
In the mid-1830s rubber got an especially bad reputation in the United
States, where several firms hastily hyped and marketed rubberized
apparel as a new, convenient solution to rainy days-convenient, that
is, in moderate climates; the hot summers and cold winters of the 1830s
revealed the limitations of untreated rubber to dismayed consumers.

Charles Goodyear, then a hardware merchant in Philadelphia wholly
ignorant of the chemical arts, nonetheless was convinced there had to
be some way of overcoming rubber s sensitivity to the weather.  He
spent years mixing rubber with various compounds and solvents while
subjecting the results to various degrees of heat.  Everything was
worth trying-vegetable oils, ink, even chicken fat.

In an apparent breakthrough in 1838, he found that bathing natural
rubber in the vapors from nitric and sulphuric acids seemed to achieve
the desired resistance to disintegration in the heat.  The
cash-strapped Goodyear even convinced the post office to offer a
contract for mail bags made of the acid cured rubber.  But the first
batch of mail bags revealed the Achilles heel of the process: after
spending several days in a hot storage room, the rubber crumbled into
pellets, and the post office withdrew its contract.

No little setback like that was going to thwart the obsessed Goodyear.
And in 1839 he stumbled onto a process at a hot stove in Woburn,
Massachusetts, where a small rubber company had given him space to
tinker.  When he accidentally spilled raw latex rubber mixed with
sulfur onto the stove, it was as though he were a protohuman hominid
discovering the secret of chipping stone.  Rubber mixed with sulfur and
then heated transformed into a stable elastic material unaffected by
the normal range of seasonal temperatures.  He later called the process
vulcanization.  Vulcan, the god of volcanoes, was an appropriate name
for the process, he thought, because its essential ingredients were
sulfur and heat.

Goodyear didn t know it, but the sulfur bonded into chemical bridges
that linked the individual rubber molecules into a huge elastic
network, like a three-dimensional spiderweb, thus preventing the
individual polymer molecules from sliding by one another in hot weather
or from cracking apart in cold weather.  By the end of 1841, Goodyear
had developed a process suitable for industrial-scale production of
vulcanized rubber.

England took the early lead in rubber production, just as it had in
iron and steel; the output of rubber there increased 6,000 percent
between 1840 and the end of the century.  While most of it went into
waterproofing items like clothes, shoes, roofs, and leather fire hoses,
it also found major applications in shock absorption in vehicles and
heavy machinery and electrical insulation in cables.

Eventually, the demand for tires elevated rubber into one of society s
material mainstays.  The first rubber tires were made in the 1840s.
They were solid and made for bicycles.  Toward the end of the century,
men with last names still familiar today-John Boyd Dunlop, Andre and
Deoudard Michelin, for example-developed pneumatic tires that
automotive engineers were introducing roughly at the same time.

As the sight of rubber became more common in the later nineteenth
century, chemists were learning more about its intriguing molecular
makeup.  It was, in fact, a polymer, a material whose constituent
molecules are made of simple chemical building blocks linked into more
or less complicated chains and networks.  Since Paleolithic times,
hominids and their descendants used naturally occurring polymers made
of biochemical building blocks like amino acids and sugar molecules,
which assembled into such useful materials as hair, fur, hide, horn,
and sinew.

Raw rubber is also a natural polymer that comes from trees, not
laboratories.  It is composed of a small hydrocarbon molecule called
isoprene manufactured in large quantities by the enormous collective
industry of latex-producing cells in rubber trees.  And like a piece of
raw flint, natural rubber requires some manipulation-Goodyear s
vulcanization, for example-to make it more useful.

Rubber was just a harbinger of a whole new world of polymers.  A
subsequent polymer, which required more extensive chemical
manipulation, made a quiet and troubled commercial debut in 1869,
aiming not at the needs of a vast military or industrial empire but
rather at a small coterie of sportsmen who were indisposed by an
apparent shortage of ivory for billiard balls.  The incentive
reportedly came from Pheland & Collender, a New York City firm that in
1863 had posted a $10,000 offer for patent rights to anyone who could
come up with a workable substitute for ivory.

John Wesley Hyatt, a printer and mechanic in Albany,

New York, rose to the occasion, taking on the challenge with a quixotic
naivete that clearly predates our current era of hyper specialized team
research.  He began by fiddling with known plastic like materials such
as pressed wood pulp and gum shellac, which were then commonly combined
to make dominoes, checkers, and other items that didn t have to take a
beating-definitely not the right stuff for making billiard balls.  A
journeyman printer, Hyatt then turned to a seat-of-the-pants brand of
chemistry.  He mixed and matched binding substances like glues, resins,
shellacs, and gums with fillers like paper, rags, linen, wood, and
ivory dust.  But none of these led to a new material fit for billiard
balls.

He was set on the right track by (what else?) an accidental finding-a
small, thumbnail-size piece of a sturdy, transparent film he happened
upon in a cupboard of the print shop he was working in.  The film had
formed because of a spill from a bottle of collodion, a viscous, plant
derived substance that was used mostly as an early medium for holding
photographic images and as a surgical dressing.  At the time, printers
often would coat their fingers with the stuff for protection against
the shop s many hot, metallic surfaces.

The collodion that Hyatt and competitors decided to work with came from
cellulose in cotton after it was first treated with strong nitric acid
to produce a nit rated cellulose whose explosive qualities earned it
the name guncotton.  Hyatt found that he could dissolve the nit rated
cellulose in organic solvents like ether (an early and dangerous
general anesthetic) and ethyl alcohol (the kind of alcohol in liquor),
themselves products of the ever more diversified chemical industry. The
initial result was a clear, syrupy liquid that dried as a thin, clear
film.

The material s main use throughout the 1850s and 1860s was as a
photographic emulsion.  When dissolved and mixed with silver chloride
crystals, it could be spread into a thin film onto a glass plate that
then would have the curious ability to capture a photographic image.
Light hitting the crystals would cause the silver atoms to clump into
dark spots-the black of black-and-white photography.  But could
collodion be used for other things?

By trial and error, Hyatt hit upon the formula and process for
celluloid, the first commercially successful plastic.  Isaiah, his
brother and business partner, named it after the cellulose from which
it originally was derived.  The cellulose, first converted into a nit
rated form by soaking it in nitric acid, was ground up and dissolved in
water to form a fine pulp.  At this stage Hyatt found that he could add
pigments and dyes (also products of the infant chemical industry). Next
he added ground gum camphor to the mix in a ratio of one to two, and
then pressed and strained all of the remaining water out.  This
preparation went into a mold and then into an oven heated to between
150-300 F; it was then subjected to high pressure.  In his 1869 patent
Hyatt wrote that the result is a solid material about the consistency
of sole-leather, but which subsequently becomes as hard as horn or bone
by the evaporation of camphor  It had the form ability of wet clay yet
was tougher and more resistant to breakage.

A year after the patent, John and Isaiah started the Albany Dental
Plate Company to manufacture and sell the first celluloid
products-substitutes for the hard rubber dental plates that were most
common at the time.  The celluloid could be color-matched more easily
than the rubber, but it, too, had drawbacks: it warped easily and had a
camphorous flavor, so it never seriously threatened the rubber
standard.

In 1871 the brothers established the Celluloid Manufacturing Company, a
name designed to reflect their ambitions beyond dental plates.  The
company supplied the material in more generic forms-rods, sheets, and
tubes-with the idea that other manufacturers would buy these and
fashion them into a huge variety of finished products.  On short order
the

Hyatt brothers had convinced financial backers that celluloid was a
good risk, and with money from financiers, the company relocated to
downtown Newark, New Jersey, in 1872.

Despite celluloid s tendency to ignite (the Newark factory burned to
the ground on September 8, 1875), celluloid infiltrated many markets
over the ensuing two decades.  With tortoiseshell and mother-of-pearl
coloring, it was a cost-effective substitute for the more expensive
materials in items in the personal hygiene market such as combs,
mirrors, and brushes.  With coloring and texturing it served as a
substitute for ivory and even for fine linen in the form of celluloid
cuffs and collars.

Celluloid even ended up circling back toward its collodion roots when
George Eastman and a chemist colleague, Henry Reichenback, modified
Hyatt s celluloid into a new photographic film.  Unlike collodion,
which had to be painted onto glass plates to work as a photographic
emulsion, this product could be successfully hardened and then rolled
without breaking and flaking.  In the summer of 1889, the first plastic
film went on sale, launching the era of Eastman Kodak s load-and-shoot
camera.

Besides Hyatt, many others jumped into the emerging fray of new plastic
materials.  One of Hyatt s competitors, Alexander Parkes, an English
metallurgist and chemist, had long felt that industrialists could
benefit from a widely available material that, in his words, combined
the properties of ivory, tortoise shell, land] horn but that could be
worked into many forms, as wood and metal could.  Patents were issued
for various plastic materials for uses ranging from waterproofing to
varnishes and lacquers to imitation bearskins to dental plates.

Ironically, celluloid never did serve its original purpose:
substituting for ivory billiard balls.  Its elasticity differed too
drastically from that of ivory.  A game with celluloid balls just felt
too different.  It was a sobering lesson in the intricacies of trying
to substitute new materials for traditional ones.  One plastic did, in
fact, become the first acceptable replacement for new stuff in the old
world N 55

ivory in billiard balls.  Known as Bakelite, it was not so humbly named
after Leo Henrik Baekeland, a flamboyant, Belgiumborn chemist who
invented it by reacting two simple com pounds derived from the chemical
gemische of coal tar near the turn of the century and introduced it
commercially in 1909.

Notwithstanding the success of Bakelite, people came to see celluloid
as a cheap substitute for the real McCoy, according to the historian
Robert Friedel.  Although it did prove itself to be a superior (though
uncomfortably inflammable) material for photography, celluloid s
original stigma as a cheap impostor still clings to the lab-made
polymers that have followed it in the still unfolding Age of Plastics.
Yet, celluloid was the seed of an enormous class of
materials-artificial polymers-that has led the way, for better or for
worse, in altering the texture of the human environment in the
twentieth century.

One of the early uses for polymers was as insulating material for the
nascent electrical industry of the 1870s and 1880s.  Rubber and other
tree-made materials such as gutta-percha were among the first polymers
used in this way.  Bakelite found ready use in that application as
well.  Besides powering the growing population of electrical lights and
machines, the availability of cheap electricity helped to spawn large
quantities of a new metal: aluminum.

Aluminum, first isolated from its ore in tiny amounts in 1825, is
lightweight, akin to silver in color, yet resistant to tarnishing and
corrosion.  Early on the press heralded the new metal as holding the
potential to transform industry and daily life, if only a cheap way to
produce it could be found.

Until the 1880s aluminum was so rare and precious that it was used
chiefly for high-end items obtainable only by the affluent: opera
glasses and jewelry, for example, were often made with the new
material, and Napoleon Ill of France, an early patron of Bessemer, had
a set of tableware made out of it.  While mills were springing up
around the world to produce steel in tonnage quantities, lone
technicians in laboratories were

56 u S T U F F

coaxing aluminum from its ore ounce by precious ounce.  When a

100-ounce apex of aluminum was placed atop the Washington monument with
great ceremony in 1884, it sold for about one dollar per ounce (sixteen
dollars per pound).  That was about the same as silver, which then was
thought of as a precious metal.  Besides having a fitting status for
the national monument, the metal s conductivity made it suitable for
the lightning-rod assembly that it was part of.

Aluminum, the third most common element in the earth s crust, was first
isolated as the great proliferation of machines and railroads was
fueling the ascent of iron production.  In its major ore, a
reddish-brown bauxite, aluminum atoms combine with uncommon tenacity to
oxygen atoms as the mineral alumina (aluminum oxide).  This is the
basic stuff of sapphire crystals.  (Alumina also is better than common
silica-based clay for making tough and refractory ceramic bricks for
industrial kilns.) Without electricity, yanking the metal from the
aluminum ore requires the most reactive substances possible, reactants
so voraciously hungry for oxygen atoms that they would rip them away
from their aluminum partners.  The first reactants found that could do
this were the metals potassium and sodium.  The late chemist and writer
Primo Levi, a Jewish holocaust survivor whom the Nazi s forced to work
in a synthetic rubber plant at Auschwitz, described sodium this way in
his book The Pc?@odic Table:

Sodium is a degenerate metal: it is indeed a metal only in the chemical
significance of the word, certainly not in that of everyday language.
It is neither rigid nor elastic; rather it is soft like wax; it is not
shiny or, better, it is shiny only if preserved with maniacal care,
since otherwise it reacts in a few instants with air, covering itself
with an ugly rough rind: with even greater rapidity it reacts with
water, in which it floats (a metal that floats!), dancing frenetically
and developing hydrogen.

Sodium and potassium are so reactive that in nature they always are
bound with other elements into salts and minerals and never found in
their pure states.  Getting reasonable amounts of sodium and potassium
turned out to be so difficult, expensive, and dangerous that the
aluminum produced with these reactants could not compete in the
marketplace.

Few things fuel innovation more than the rewards of fame and fortune.
In 1886 two young entrepreneurs-Paul L. T. Heroult in France and Martin
Hall in the United States-each working in home laboratories, hit upon
the solution.

The key for both of them was electricity, which Thomas Edison, the
great inventor and master of informed trial-and error was then pushing
forward to help power the growing population of electric lightbulbs.
Electricity, like chemistry and fire, can transform the raw materials
of the world into different and new forms.  Earlier in the nineteenth
century chemists had found that electricity passing between the
electrodes of a battery immersed in baths of chemical solutions often
elicited profound effects.  Beautiful mineral and metallic crystals
often would build up on one electrode.  Gases would often evolve from
another.

Heroult and Hall-who was moved to go after the aluminum trophy by a
quip from his professor about the riches in store for whoever won
aluminum from rock-hit upon the right electrochemical process within
weeks of each other.  First they melted cryolite, a naturally occurring
aluminum salt (a kind of mineral) from Greenland, in a ceramic pot
lined with carbon, which served as one electrode of what amounted to a
giant battery.  The other electrode was in the form of a carbon rod
lowered into the vat.  Heroult made these rods in a high-temperature
firing of powdered coke (roasted coal) bound with tar.

In molten cryolite the alumina ore dissolved like ice in warm water,
yielding an alumina-rich liquid.  The alumina s oxygen and aluminum
atoms could be yanked apart by a strong electrical current passing
between the electrodes.  The oxygen combined with the electrode s
carbon in a combustion reaction.  Meanwhile, the aluminum atoms
coalesced in the cryolite and sank to the bottom, where the metal could
be siphoned off.  A major difference in Hall s technique was that he
used several smaller anodes rather than the one large anode that
Heroult used.  Both filed patents in 1886.

In 1888 Hall established the Pittsburgh Reduction Company, the
progenitor of Alcoa (Aluminum Company of America), in Kensington,
Pennsylvania, and began producing aluminum at a rate of 5 pounds each
day.  And in Europe, the Swiss Metallurgical Company in Neuhausen was
the first firm to use Heroult s electrolytic cells.  Within a few years
a drastic drop in aluminum prices-from twelve to two dollars per
pound-made widespread experimentation with the metal economically
practicable.  Since then, the aluminum production process has changed
largely in detail and scale.

Nevertheless, steel remained the king of metals.  The Journalistic hype
of aluminum as silver from clay didn t quite catch on.  Ironically,
steel makers found that small additions of aluminum reduced the amount
of dissolved air in the molten steel, which in turn reduced flaws in
the solidified metal.  The use of aluminum in steel making rescued
aluminum production from complete un profitability.1

That dependency on steel didn t last long, however.  Armies
experimented with aluminum as a lightweight material for mess kits.
Musical instrument makers dabbled with it to make horns of various
kinds, although the metal s softness doomed it for that purpose.
Aluminum found its way into artificial limbs, bathtubs, jewelry,
tableware, and many other products, none requiring very high volumes of
material.  Pure aluminum was not especially strong, and that relegated
it to uses in which it would not need to withstand much stress or
abuse.

Aluminum alone might not have been the next best thing since Bessemer
steel, but metalworkers knew there was a world of potentially more
useful aluminum alloys just waiting to be discovered.  Throw in, say, a
little copper or nickel, and then cool it slowly or maybe quickly, and
see what you get.

It was Alfred Wilm, then head of the Metallurgical Department at the
erman enter or cienti c esearch near Berlin, who hit paydirt.  In 1902
the German Waffen and Munitionsfabriken, a consortium of German war
munitions companies, asked Wilm to develop a lighter-weight alloy that
might replace the heavier brass alloys so easily drawn into tubes and
therefore so convenient for making bullet and munitions cartridges.
Wilm looked to aluminum as a candidate.  He knew that aluminum itself
was not fit for the job but that an aluminum alloy might work.  It didn
t take long for him to find partial success with an aluminum-copper
alloy.  It had some of the drawability that the Germany munitions
companies wanted, but Wilin didn t think it was strong enough.

In 1906 he embarked on a fateful set of experiments.  To the aluminum
and copper he added some magnesium and manganese which had been
discovered empirically decades earlier to strengthen steel.  When he
tested the resulting alloy for hardness by pressing a steel ball into
it, the resulting impression was deep enough that Wilm assumed his
experimental aluminum alloy was too soft to be useful.  Two days later,
however, he and an assistant checked the result by redoing the hardness
test.  To their amazement, the impression was much shallower.  Somehow
the alloy had gotten harder just by sitting there.  The hardening trend
continued for two more days.  In the end the aluminum alloy was three
times harder (and thereby probably stronger) than after the original
test.

Wilm had discovered age hardening, which metallurgists later attributed
to the development of tiny mineral particles visible in metallographic
preparations-that effectively dammed UP motion amongst the tiny grains
that made up the metal.  The discovery showed that metal is not some
unchanging stuff once it cools into a solid.  Like a person, it can
change with time.  And

60 m S I U F F

that meant that the way you nurture metal can make it better or worse.
(It turns out that the metallurgists who made an airplane engine block
for the Wright brothers, the first men to fly, had unwittingly pulled
off age hardening in an aluminum alloy several years earlier.)

Three years later, Wilm s alloy, called Duralumin, went into commercial
production.  Of the nearly 13 tons produced in 1910, 10 went to the
Vickers Company in England, which used it to build a dirigible called
the Mayfly.  Unfortunately, as workers moved the massive ship from its
construction hangar in preparation for its first flight, it broke in
two, Even though the disaster was the result of mishandling, the
British viewed this German Duralumin as a new material to avoid.

Nonetheless, by 1914 the German Navy had specified Duralumin as the
material for zeppelin airframes.  In flying machines lightweight
materials that are strong enough to serve as structural components are
of paramount importance.  Ninetyseven zeppelins were built, the larger
ones requiring 9 tons of Duralumin.  Aluminum as a material and
aerospace as a technology joined in a common ascent through the ensuing
decades.

The sporadic inventions and discoveries of the nineteenth century-new
polymers like celluloid and Bakelite and new metals like aluminum and
Duraluminwere concrete demonstrations that there were many stones yet
to be unturned--or fortuitously chipped-in the realm of stuff.  In the
twentieth century a new piston-sociological innovation-joined
curiosity, accident, and economic incentive in speeding the engine of
exploration.

The era in which individuals like Goodyear, Hyatt, Wilm, and Baekeland
could spawn industries off of a single material discovery was giving
way to a higher-powered, more systematic approach that relied more on
the collective brain of institutional science.  As Bakelite diffused
into people s homes and aluminum showed its aeronautical mettle
overhead, the race for still better and newer materials was
accelerating in new industrial laboratories that sought to hasten the
transition from theoretical insights to marketable commodities.

Edison set the precedent in 1876 when he opened his laboratory in Memo
Park, New Jersey.  He told the world that its output would be a minor
invention every ten days and a big thing every six months or so.  The
German dye and chemical industry had begun systematic chemical research
in the 1880s, resulting in an economic coup that shifted the center of
chemical innovation from England to Germany.  General Electric set up
its first industrial laboratory in 1900 initially to improve the
company s lightbulb technology.  Two years later E. 1. duPont de
Nemours; and Company followed suit.  Andrew Carnegie, the steel mogul
and railroad baron, did the same and wrote in his autobiography that
nine tenths of all the uncertainties of pig iron making were dispelled
under the burning sun of chemical knowledge.  Eastman Kodak, American
Telephone and Telegraph, Corning Glass Works, and many, many others
soon joined the trend.

This sociological sea change in the mode of materials discovery showed
up most dramatically in the proliferation of new polymers.  In a
laboratory at B. F. Goodrich in 1926, a chemist made polyvinylehloride,
from which he made adhesives and sheets.  Four years later the German
chemical giant 1. G. Farben introduced polystyrene, which could even be
made into extremely lightweight foams.  In 1935 another flrm, Rohm and
Haas, rolled out poly methyl methacrylate (first discovered over sixty
years earlier) as Plexiglas, which was at once transparent, strong, and
tough enough to substitute for glass over the cockpits of military
airplanes.  Two years later DuPont rivaled Plexiglas with its own
Lucite.

Within the next ten years came epoxy resins, Teflon (an accidental
discovery whose nonstick properties are due to the four fluorine atoms
on each of its molecular building blocks), Polyethylenes (which are
ideal for forming containers like milk jugs), polyethylene
tereplithalate (strong and impervious enough to contain pressurized
liquids like soda), and others.  The number and rate of new polymer
discoveries has only increased since then.

Of all of the polymer developments, none was more emblematic than the
invention of nylon at DuPont s Purity Hall in Wilmington, Delaware.10
The process started in 1928, when the company lured a brilliant,
thirty-two-year-old chemist, Wallace Carothers, from his university
bastion at Harvard.  His mission was to unravel the chemical details
behind the formation of polymers, whose general molecular structure at
the time was under intense debate.  Some chemists thought they were
made of small molecules bunched together like minuscule clumps of dust;
others held that polymers were essentially enormous versions of the
small molecules that chemists had gotten used to over the previous
hundred years.  Carothers approached polymers using the most current
techniques and concepts of science, including the new ideas of quantum
mechanics that were beginning to permeate the ranks of chemists. Within
three years of Carothers s arrival, his group scored a hit: they
developed neoprene, one of the first commercial synthetic rubbers.  Far
more resistant to chemicals than vulcanized natural rubber, it was well
suited for use under the hoods of automobiles.  After that, the lab
focused on the quest for artificial textile fibers.  Rayon, pretty much
a fiber version of cellulose from plants, already had reached
commercial production in the 1920s.  Carothers and his colleagues
wanted to un tether themselves from the restrictions of naturally
occurring feed stocks such as cotton-derived cellulose.  They wanted to
start from chemical scratch.

For a model they turned to silk, the strong, resilient, and supple
fiber produced by spiders and insects like the silkworm.  Silk fibers
are made up of protein molecules, which are natural polymers themselves
composed of amino acid molecules linked into chains.  The chemical
linkage between amino acids is known as an amide bond, and that was one
of the chemical domains that Carothers s team decided to reconnoiter.
The proteins that cellular chemistry builds into proteins could also be
called poly amides  So the DuPont team was out to make artificial poly
amides

Each formulation would consist of one or more types of small organic
molecules, somewhat like the amino acids of proteins, that would be
linked into polymers under various reaction conditions.  One day in May
1934, one of Carothers s assistants put a stirring rod into a beaker
containing the syrupy result of a reaction.  When he pulled the rod
out, a long string of the liquid followed up like the tail of a kite
and quickly cured into a silk like fiber.  This became known by polymer
chemists as the rope trick.  That rope trick led to the first nylon
fibers.  By 1937 DuPont researchers spun the first nylon yarn and wove
the first experimental stockings.  That same year Carothers, who had
been suffering from increasingly severe depression, committed suicide
with a dose of cyanide in a Philadelphia hotel room.  It was a mere
three weeks after DuPont had filed for the basic patents on nylon.  In
1938 nylon went into production.

Nylon was instantly popular.  In 1939, when the first nylon stocking
went on sale, DuPont promoted its new silk substitute with fervor,
resorting even to 30-foot-tall mannequin legs touting the new stuff.
World War II deflected this zealous marketing campaign for a while.
When Asian supplies of silk were cut off, the U.S. government turned to
the new synthetic silk nylon-for the tough material it needed for
parachutes, tire cords, flak vests, and rope.  For several years all
nylon that DuPont produced went into the war effort; none was sold in
commercial markets.

After the war the material s combination of silk like toughness and
elasticity as well as its better resistance to chemicals and
electricity made it attractive for all kinds of things.  As a fiber it
made it into everything from shirts to climbing ropes.  But it also
could be molded into hundreds of items, including combs, machine gears,
zippers, and electrical insulators.  Modifications of its chemical
constituents and the way it is processed have continued the
diversification of its roles.  Nylon remains a household name and
continues to be part of DuPont s bread and butter.

The mushrooming of industrial materials laboratories was part of a
larger trend that had begun to accelerate in the first decades of the
twentieth century: the interdisciplinary quest for the key to the
secret inner life of materials.  As physicists dissected the anatomy of
atoms, metallurgists, ceramic experts, chemists, and all kind of
engineers and scientists delved in ever more detail into the
relationship between a material s many leveled structure and its
properties.  Why exactly do ceramic materials break so readily but
metals only dent?  Why does steel resist rusting when you put enough
chromium into it?  What is going on inside of a piece of metal when it
undergoes age hardening?  What makes rubber so elastic?  Can a
carefully constructed crystalline material function like an entire
device, say, like the vacuum tubes of early electronic technology?

Thousands of scientists and engineers in hundreds of different labs
began to look into questions like these.  At best, they were
peripherally aware of one another s pursuits.  Their common thread was
that they were routinely validating the importance of Reaumur s
present hierarchical portrayal of materials in the eighteenth
century, his intuition that a material s essence depends upon how its
ingredients assemble from the atomic level up to the macroscopic level.
They were also validating and extending Sorby s call to observe and
reveal that hierarchy in as much detail as possible.  Like a person s
personality, every material s collection of traits and behaviors comes
from what it is made of and how those ingredients were nurtured into
their present form.  Not only did researchers in the first half of the
twentieth century know this intellectually, but they were also building
the analytical tools and concepts that finally were about to let them
uncover the details in practice.

Three

The Secret Architecture of Stuff

It takes an uncommon technical breadth to traverse matter s vast inner
worlds of atoms and molecules and its equally expansive outer world of
diverse substances and objects.  It requires an intricate blend of
chemistry, physics, mathematics, computer science, engineering, and
whatever other field of knowledge or skill happens to help.  The
ultimate task, after all, is to understand how a common palette of
about ninety types of atoms-the periodic table of the chemical
elements-can harbor earth, air, fire, water, flint, obsidian,
malachite, quartz, copper, bronze, iron, stainless steel, clay,
pottery, porcelain, glass, Portland cement, limestone, rubber,
cellulose, Bakelite, nylon, horn, olive oil, petroleum, spandex,
cotton, silk, electronics-grade silicon, gunpowder, skin, bone, and
countless other materials.

The reductionist approach of trying to understand materials in terms of
atoms and molecules alone is only narrowly instructive.  Atoms and
molecules are indeed the most basic ingredients of materials.  But it s
important not to miss the the secret architecture of stuff forest for
the trees.  Just as a sociologist seeks to understand the dynamics of
human interaction in lesser or greater collectivities under a variety
of conditions, so the chemist tries to grasp the ways in which the
elements of matter interact aggregating segregating, rearranging,
mingling, and repelling to emerge as multi tiered structures, each with
its own set of material traits.

The hierarchical structure of a material is the result of the
interaction of its ingredients.  That is why a little more or less
heat, a different proportion of alloying elements, a finer grade of a
pulverized ingredient, this or that contaminant, and any number of
variations on a standard process can result in what seems like an
entirely different material.

Because these differences in process can produce such wide differences
in materials, achieving the desired result is a lot like cooking.  This
is how a couple of modern aluminum engineers put it:

When you mix pumpkin, spices, sugar, salt, eggs, and milk in the proper
quantities, you can make a pumpkin pie filling.  By adding flour and
adjusting the proportions, you can make pumpkin bread.  Substituting
shortening for the pumpkin and molasses for the milk will yield ginger
cookies.  Each adjustment of the recipe results in a different dessert.
Just as adding enough flour can turn pie filling into bread, adding
enough chromium to steel makes it stainless steel.

The metaphor holds for all of the several hundred steel alloys in use
today.  It holds for the tens of thousands of other materials that go
into the human-made things on earth.

So a process yields a material with a distinctive hierarchical
structure that determines the material s properties and how it will
perform in its various applications.  That is the truism that Henry
Sorby helped raise to a paradigm of metallurgical research and
practice.  In his day a microscope was sufficient for a metallurgist to
discern qualitative differences within a cut, polished, and etched
metal surface.  This was a level of structure above the chemical one. A
chemical analysis could identify the kinds of atoms present in a piece
of metal and the relative proportions of those atoms, but not how those
atoms came together into specific structures.  Metallography helped to
uncover such structures and patterns of material behavior. Over time a
metallographer learns to recognize material meaning in the
mineralogical and metallurgical montages that he sees through his
microscope.  A montage that looks like an aerial view of snow-covered
mountains comes from gray cast iron, the dark outlines of mountain
contours coming from graphite flakes in the metal s micro structure
Another montage has dark, needlelike outlines crisscrossing through a
white background.  That pattern comes from a hard steel whose iron
based crystals have assumed two distinct geometric arrangements known
as austenite and martensite, named after the researchers who first
identified them.  The martensite constitutes the crisscrossing needles
in a background of austenite.

Metallographers eventually learned to correlate these montages with
certain metallurgical formulas, alloying ingredients, and processing
practices: that is, to the various ways you can cook steel.  By the end
of the nineteenth century, metallography, chemical analysis, and other
techniques for looking under a material s skin were becoming standard
tools for guiding industrial metal producers toward consistently good
results.

It was a growing assumption after the turn of the century that advances
in materials would no longer come as easily as they had in the past.
The brazen empiricism that guided the efforts of pioneers like Henry
Bessemer and Charles Hyatt still played a role, but the heyday of the
lone, hit-or-miss tinkerer was over.  Advances grew increasingly out of
precise quantitative observation, measurements, and the kind of
theorizing that physicists and chemists had been honing over the past
century.

Despite the improvements in research methodology, the limitations of
materials technology remained all too apparent.  In the first years of
the twentieth century, for example, the annual average accident rate on
the U.S. railroad system was about four thousand, resulting in
approximately thirteen thousand deaths and injuries per year.  The
blame for most of the carnage went to metal gone bad: broken rails,
wheels, flanges, and axles pushed to failure by a combination of
excessive loads, inadequate maintenance, and inferior iron and steel.

But the toll from inadvertent death paled next to the purposeful
slaughter of World War I, which became a merciless laboratory for
materials research in which the potentially beneficent tools of modern
science-advanced machinery, mass production, precision
manufacturing-were converted into instruments of death.  Even chemistry
diverted some of its attention from the boon of pharmaceuticals toward
fostering a plague of chemical weapons in what some historians have
called the chemists war.  It was falso] an iron and steel war, with
comparative advantage in weapons technology determined by relative
superiority among belligerents in steel-ingot forging, foundry, and
metalworking technologies, notes one metallurgist who later worked on
the unprecedented materials problems related to nuclear energy.
Metallurgists were hidden heroes or perhaps, in the eyes of the
victims, villains-in World War 1, just as they were for the Hittite
armies and Samurai warriors of centuries past.

Tools like metallography-which can help reveal the structural links
between a material s levels, from the invisible atomic base to the
visible superstructure-opened only some of the many windows on material
anatomy.  A metallographer couldn t tell from his microscopic
observations exactly how the atoms in each region or grain of a
metallographic preparation were arranged.  Were the atoms tightly
packed like perfectly stacked oranges, or were they arrayed more
randomly, as though there were lemons and potatoes in the stack?  Or
were crystalline utopias mixed with anarchic dystopias within an
apparently unitary metallographic spot?

Deep internal differences like these, it turns out, have everything to
do with a material s mechanical properties: its strength, its
resistance to fracture, its susceptibility to corrosion or deformation.
At the turn of the century, the key to enhancing understanding of and
control over a metal s internal structure and thereby over its
technological abilities lay in seeing structures finer than those
discernible in Sorby s metallography.

X rays, first observed in 1895, provided one very powerful means.  The
study of crystal structure first developed mathematically in the
eighteenth and nineteenth centuries.  It exploded into experimental
significance in 1912 and 1913, when the father-son team of William
Henry Bragg and William Lawrence Bragg developed a technique called
X-ray crystallography, which uses X rays as a precise tool for
measuring the crystal s structural regularities on atomic scales.  Here
s how it works: as X rays pass into a crystal, they diffract (a kind of
atomic-level reflection) from the crystal s regular arrangement of
atomic columns and planes, which act like mirrors to the incoming X
rays.  When the Braggs let these diffracting X rays expose photographic
film, the result was a diffraction pattern that disclosed the crystal s
regularities or lack thereof.  From the photographic pattern of rings
and spots, arranged more or less concentrically, crystallographers can
calculate things like distances between adjacent atoms (or molecules)
and atomic (and molecular) planes.

In principle the technique provided a zoom lens on the individual
crystal grains visible to metallographers on their optical microscopes.
It enabled the metallurgist to link empirically the properties of a
piece of steel or brass all the way down to its atomic idiosyncrasies.
At the same time, X-ray crystallography the secret architecture of
stuff a 73

underscored the reality that real materials seldom exist in perfectly
crystalline states.  Most often they are conglomerations of many
crystals, just like most stones you might pick up from the ground.  Or,
like glass and polymers and liquids and gases, they have little or no
internal structural order.  Send X rays into a disordered or amorphous
material and they diffract every which way, leaving on the film a
cloudlike image providing a paucity of detailed structural
information.

As the Braggs gave the world a window on the atomic level of material
structure with their X-ray crystallography, Alfred Wilm s discovery of
age hardening was opening another window onto the structure of
materials.  Three researchers at the U.S. National Bureau of Standards
(now called the National Institute of Standards and Technology)
actually teased apart the micro structural details underlying Wilm s
observations.  The NBS researchers discovered that the solubilities of
the alloying elements in Duralumin were much higher in solid hot
aluminum (at the 525 C or so that Wilm had used in his work) than at
room temperature.  Once the metal cooled to room temperature from the
heat treatment, scientists observed some new formations within the sea
of aluminum: tiny crystal islands with chemical compositions such as
CuAl 2 and MgSi; and tiny zinc particles.  The size and distribution of
these islands mattered.  Hard alloys had lots of small particulates.
But a smaller number of larger particulates, or too many of the smaller
particulates, led to weaker metal that fractured more easily.  The NBS
researchers had found a key to what makes aluminum and many other
metals good or bad.

As usual in science, solutions become the problems.  An aluminum alloy
containing a certain distribution of mineral and various metallic
particulates might be harder than one with some other distribution. But
how in fact does structural detail actually harden the metal?  It is
one thing to observe structures that correlate with material properties
like hardness.  It is another to understand the mechanism by which the
structure yields the property in question.  If you understand the
mechanism of, say, age hardening, then you have a better chance of
figuring out how to make harder or stronger metals.

By 1930 metallurgists and physicists had postulated several kinds of
large and small imperfections within and between a metal s grains.
These structural imperfections turned out to be at least as important
to a material s properties as its chemical composition and the
molecular structures within its crystal grains.  As Cyril Smith once
wryly put it, Everything really interesting about materials resides in
the hierarchy of their imperfections.

Like biologists spotting new species in a previously unexplored
ecosystem, an army of scientists, equipped with ever better analytical
instrumentation (including electron microscopes beginning in the 1930s)
were happening onto new kinds of defects and micro structures and
determining the processes by which these formed.  Among the defects is
the absence of an atom from its normal place in a crystal (a vacancy);
another is a single contaminant atom taking the place of an expected
atom in a particular crystal.  It is like finding one tangerine or
grapefruit in a stack of naval oranges.

Contaminants also tend to diffuse more rapidly along larger defects
such as the boundaries between crystal grains and dislocations within
atomic planes of the grains.  You can think of boundaries as defects
since the most perfect piece of metal would be one huge monolithic
crystal with no internal boundaries.  Like dirt between two pieces of
wood glued together, atomic riffraff between metal grains weakens the
intergrain bonds that hold a piece of metal together.  An example of
such dislocation is a line defect, which provides internal slack for
atomic planes to slide by one another.  Line defects help determine a
metal s ductility, form ability and vulnerability to various stresses
and strains.

Another type of defect is known as a screw dislocation,

which results from atoms linking to one another in an upwardly helical
sequence that allows growth to occur more continuously than if one
atomic plane had to be fully completed before the next could start.
Screw dislocations also govern the strength and chemical performance of
materials.  Atoms in the helical structure can be more vulnerable to
the kinds of chemical attacks that lead to corrosion, embrittlement,
and other nasty changes.  Such imperfections also can be good, however,
since it is the screw dislocations, rather than the metal crystals
hosting them, that suffer the violence.  Such balancing acts are
typical of the quest to make good materials.

Such defects and processes have become the meat and potatoes of
research into the mysteries of metallic qualities such as hardness,
softness, ductility, and brittleness.  Growing understanding of them
has opened pathways to making new metallic alloys and ceramic materials
that are harder or more deformable or tougher than previously known
materials.

Although studies and experiments by metallurgists spurred many of the
big advances in materials research, cross-pollination of insights among
chemists, physicists, and other scientists was also generating new sets
of rules for revealing and exploiting the similarities and differences
among materials.  Thanks to this growth of multifaceted knowledge about
how specifle defects, phases (different states of matter, including
different crystallographic forms in a single solid sample), and other
micro structural features affect a material s properties, a more
rational approach to alloy and material development was becoming
possible.

By World War II researchers were routinely taking Henry Sorby s micro
structure-tomacrofunction paradigm to atomic levels of detail.
Metallography, X-ray diffraction, electron microscopy, and other
analytical techniques often provided the data.  There were many tools
for making sense of the data: dislocation theories; thermodynamic
models that described how

76 s S T U F F

different solid phases of different compositions and crystal forms end
up in materials; and ideas about precipitation hardening and related
concepts.  Metallurgy thus became connected to other disciplines
investigating solid materials such as minerals and ceramics, the
mechanical properties of which were affected by the same things.

The convergence of ideas and research techniques gradually brought
members of these related fields together under one umbrella that came
to be known as solid-state physics.  In the 1930s and 1940s this
burgeoning discipline was the intersection where material arenas as
seemingly disparate as polymers, metals, and ceramics converged.

A major by-product of this interdisciplinary ferment was the
proliferation of polymers.  The celluloid plastics were already nearly
sixty years old.  Bakelite had been around for a few decades.  Each
year industry researchers were unveiling new polymers with new
properties and technological possibilities.  The Age of Plastics had
begun.

Many of the polymer players were sophisticated theoretical chemists,
but the development of the materials still advanced largely by dogged
empirical work.  U-y this or that reaction.  Try this or that
temperature or pressure.  Throw in a little metal and see if it
catalyzes different reactions.  Moreover, polymeric micro structure is
rarely crystalline.  It is more like cooked spaghetti.  It is more
complicated and harder to unveil than a Standard metallic or mineral
micro structure  Shine X rays through most synthetic polymers and the
picture you get on the film is a diffuse cloud.  That is just what you
might expect from a jumble of molecules resembling a pasta dish.

Not only were researchers clueless as to how polymer molecules clump
together to form a material, but until the mid-1930s, most chemists
didn t even acknowledge that polymeric materials were made of huge
molecules.  The late Herman Mark-an Austrian refugee who spent his
professional career at the Brooklyn Polytechnic Institute, where he
rose to become one the fathers of polymer science put the situation in
the late 1920s this way: There were fiber chemists; cellulose chemists,
a

Society for Cellulose Research, books on cellulose, journals on
cellulose.  The same thing for silk, the same for wool, for rubber, and
starch.  So there were five different disciplines, strongly represented
by people, by literature, books, societies, and so on.

The modern field of polymers supplied a common thread to this
disjointed quest when the German chemist Herman Staudinger began
floating what was then a laughable idea to most other chemists: he
argued that natural materials such as silk, rubber, and starch, as well
as lab-made materials such as Bakelite rayon and celluloid, were all
composed of huge chain like molecules.  He was proposing the existence
of enormous molecules hundreds or thousands of times larger than the
smaller, more compact molecules with which chemists had become familiar
during the past century.

Herman Mark likened Staudinger to a biologist asking his colleagues to
believe there were elephants 400 feet long and 200 feet high actually
roaming the earth.  A more popular idea for explaining the properties
of polymers (such as their higher viscosity compared to liquids like
water or benzene) was that smaller molecules congregated into larger
structures without actually forming strong chemical bonds.  The
resulting so-called colloidal structures were more like a bunch of
people squeezing together as in a Rugby serum or a New Age hug.

By the mid-1930s, however, most chemists had come around to Staudinger
s way of thinking.  The idea of long molecules made of different
smaller building blocks that link like beads of a necklace was a
powerful unifying principle.  Thanks to Staudinger s daring idea, what
Mark had noticed as five separate disciplines in the 1920s now evolved
into subdivisions of a single, increasingly fertile field.  Wallace
Carothers, an industrial Polymer scientist at DuPont, adopted and
applied Staudinger s theory with momentous practical results: he
invented neoprene (a chemically resistant synthetic rubber) and nylon,
two immensely useful and profitable new materials.

It was just the beginning for polymers, of course.  If you could take
any number of building blocks and then modify and link those in any
number of ways, you had in your hands a fountain of new materials that
could change the industrial and social landscape.  One of the most
famous cinematic sound bites of all time is testimony to this
versatility and impact of polymer science.  In the 1967 film The
Graduate, Dustin Hoffman, playing a recent college graduate who is
alienated and adrift in his parents prosperous, materialistic suburb,
receives one word of career advice from a friend of his father s:
Plastics.  That memorable utterance stood as both a tribute to plastic
s pervasive impact on society and as a sardonic reminder of its dubious
aesthetie and spiritual legacy for the young radicals of the Sixties.

While Staudinger was advancing his path breaking hypothesis in the
mid-1920s, an all encompassing conceptual unification known as quantum
mechanics was under way.  It would join under one theoretical umbrella
metals, ceramics, polymers, and every other material that ever was, is,
or will be.

Quantum mechanics was an irksome answer to the various troubles with
the nineteenth-century belief, stemming from the ancient Greek atomists
Democritus and Leucippus, that atoms were supposed to be hard,
impenetrable spheres-in effect, microscopic billiard balls.  But two
discoveries suggested that atoms did in fact have some kind of
differentiated internal structure: 1) all atoms have electrons (which
are far less massive than the smallest atom), and 2) there is
radioactive emission of particles from atoms.  If atoms had parts, they
could not be teeny spherical monoliths.

There were other reasons to think that atoms had some internal
structure.  For decades, spectroscope-peering scientists were delighted
and mystified by the characteristic set of colors-a spectrum of
light-that different elements emitted and absorbed.  Every time you
light a candle, you witness the emission of colored light from the
excited electrons in the atoms and molecules of the candle s wick and
wax.  The nineteenth-century scientists knew that the colors somehow
could be due to electrical charges in atoms vibrating at different
frequencies.  These oscillations would produce electromagnetic
radiation that could show up as the element s different spectroscopic
colors.

But the idea of atoms with moving parts was hard to reconcile with
other phenomena that made more sense if the atoms were in fact like
little billiard balls.  The century s most brilliant and powerful
physicists were publishing, speaking, and pushing into the scientific
canon the kinetic theory of gases.  Billiardball-like atoms in motion,
they asserted, explained all kinds of things, like the pressure of
gases at different temperatures, the way heat travels through different
materials, and why crystals assume the shapes they do.  More
complicated atoms with moving parts would muddy these apparently
clarified waters.

Was there some picture of an atom that could combine the best of both
worlds, accounting for the moving parts implied by the spectroscopic
observations yet also preserving the more monolithic, billiard-ball
character essential to the prevailing physical theories?

In the fall of 1900 at the University of Berlin, the physicist Max
Planck further confounded orthodoxy with his finding that wavelike,
spread-out forms of energy like light and heat seem to be made of
discrete packages, or quanta, just as matter does.  This insight marked
the birth of quantum mechanics, which has kept theorists ever since
scrunching their faces trying to explain how matter and energy can
behave simultaneously like discrete particles and spread-out waves.

Quantum mechanics pulverized the cornerstones of classical physics,
which had doggedly insisted on a more mechanical model of the universe.
In the traditional model, energy ought to be able to assume any value,
not just discrete ones like specific points on a line.  After all,
doesn t a piece of hot metal smoothly cool so that it takes on every
single temperature between its hottest point and its lowest?  No, said
the quantum mechanicians.  A piece of matter like iron only appears to
cool down continuously without jumps; the discontinuities in the
quantum emissions were indiscernible because a typical piece of visible
matter consists of billions upon trillions upon quadrillions of
individual particles, each of which cools down by its own set of
quantized emissions of energy.  But aggregated in the piece of iron,
these countless particles create a statistical illusion of continuous
cooling.

In 1913 a young Danish physicist named Niels Bohr had an idea of how
these revelations might affect the portrait of atoms.  What if the atom
consisted of a tiny heavy nucleus with featherweight electrons orbiting
around it like planets around a sun?  Bohr s solar system model gave
electrons, which some scientists even thought could be the basic stuff
of all atoms, a definite role in atomic structure.  It also nicely
assimilated the earlier discovery that the vast bulk of atomic mass
resides in a very small positively charged nucleus that spanned only
one one-hundred thousandth or so of the atom s diameter.

The solar system model also suggests that lines and colors seen in
atomic spectra correspond to changes in an electron s orbit around the
atom s nucleus.  When an electron absorbs just so much energy-a
specific quantum of energy-it jumps up to the next orbital.  In so
doing, it prevents the color of light corresponding to that absorbed
energy from reaching a spectroscope, thereby leaving a black line
characteristic of absorption spectra.  An already excited electron
emits just so much light energyagain, a specific quantum-when it
relaxes to a lower orbital.  In so doing, the electron emits light of a
color that shows up in an emission spectrum.  With Bohr s ideas, atomic
structure and behavior seemed to be falling into place.

But pretty as it was, the solar system idea was vague.  it revealed
nothing about why atoms of different elements have such different
chemical behavior.  It failed to explain why electrons orbiting nuclei
don't eventually lose their energy and careen into the nucleus the way
a bird that stops flapping will eventually crash to the earth.  It
couldn t explain why copper was conductive but rubber was not or, for
that matter, why any material had any particular trait.

There were other terribly vexing problems.  Physicists were discovering
that the point like electrons often behaved as though they were
spread-out waves.  Send them through crystals and they would produce
ripply interference patterns much like those of waves passing through a
gallery of poles at a pier.  If electrons were like tiny billiard
balls, they would have produced a much more discrete pattern.  Yet if
electrons were in fact more spread out, like waves, then they didn t
fit neatly into the well-defined orbits that Bohr had in mind.

With light and particles each capable of having characteristics of the
other, the neat and clean Newtonian picture of billiard balls in motion
was banished to fantasyland.  Newtonian physics still worked as a
practical and convenient tool for the large objects (like real billiard
balls) that mechanical engineers and ordinary mortals dealt with.  But
it seemed to have less and less to do with the microscopic domains that
scientists and industrialists were also beginning to understand and
control.

The quantum mechanical portrait of the atom that finally stuck
reflected the maddening dualisms and ambiguities that no one could
dispel from the atomic world.  Add to that Werner Heisenberg s famous
principle of ineluctable uncertainty about matter on these smallest
subatomic scales (you cannot know simultaneously the exact location and
motion of an electron, for example), and all pictures of atoms and
their interactions on these quantum levels were becoming a blur. Light,
long thought to consist of waves of energy, now also had to be thought
of as some kind of particle.  And particles, like electrons and atoms,
had to be thought of also as having wavelike properties.  The location
of electrons around nuclei now could be expressed only as probabilities
of being more likely here than there.  Knowledge of atoms and hence of
matter itself would forever be haunted by uncertainty.

The Austrian physicist Erwin Schrodinger brought some mathematical
exactitude to this portrait with an equation that has made him famous.
His equation sits in the scientific pantheon alongside Einstein s E=MC2
and Newton s F=ma.  Schrodinger s equation has become a mathematical
guide for calculating wave patterns in a space or field infiltrated by
entities like electrons and protons that can influence that field.  The
equation s raw output, which is called a wave function, is a map of the
likelihood that a particle such as the lone electron of a hydrogen atom
will be found in a specific region of space.  The precise size, shape,
and contours of this map depend on the number and locations of other
electrons and atomic nuclei.

With that spin the Schrodinger equation assumes an awesome power, in
principle at least.  As a modern physical chemistry textbook puts it, A
wave function contains ALL THERE IS TO KNOW about the outcome of
experiments that can be done on a system, which includes the making of
all materials.  Within that capacious equation, then, is the chemical
behavior of all atoms and compounds and the answer to whether any
proposed chemical reaction will work or not.  It can demystify all
known chemical behavior and predict chemistry as yet unobserved.  It
can account for how copper was relatively easy to smelt nine thousand
years ago, how six carbon and six hydrogen atoms form into benzene s
planar circular shape, and how a contemporary scientist might be able
to develop a new polymer that can convert red light into blue light for
higher-capacity communications systems.

Schrodinger s equation did not put chemists and materials researchers
out of business because it is, to put it mildly, a mathematical bitch.
It is a computational Mount Everest.  Accurately computing the wave
functions (and thereby the chemical and physical behavior) of even the
simplest atom-the hydrogen atom (one proton and one electron)-was
impossible until the 1960s, when computer power began becoming more
widely available.  Even to solve Schrodinger s equation for a hydrogen
molecule (two protons and two electrons in motion) requires one of
today s most powerful supercomputers.  For molecules with a half-dozen
atoms, each with their fair share of electrons, solving the raw
equation is beyond computational reach.

Yet Schrodinger s equation and its quantum mechanical framework did
probe the insides of materials with an entirely novel kind of lens.
Scientists became adept at developing clever approximations and
shortcuts that cut the equation down to more manageable though less
accurate forms with which they could extract at least qualitative
answers to their questions.  It helped physicists understand how the
electrons in some materials can start to flow in electrical currents
while not in others (see chapter 8).  And the quantum mechanical
framework yielded a flood of insights about how matter and energy in
forms such as light, heat, and electricity can interact in almost
magical ways.

Here, it seemed, was a Rosetta stone for unearthing wondrous new
materials.  Consider what happened just before Christmas in 1947 at
Bell Telephone Laboratory in suburban Murray Hill, New Jersey.  There,
Walter Brattain and John Bardeen watched an oscilloscope trace out the
amount of electricity passing through an inelegant-looking gadget that
Brattain had made.  It consisted of a crystal block of germanium
(peppered with atoms that could accept electrons) on which there were
two tiny gold spots separated by about 4 microns (a hair is about 50
microns in diameter).  (See Figure 9.)

When the physicists applied a tiny electric voltage to one gold
contact, the electronic environment in the crystal changed.

FIGURE 9

The Mother of All Transistors.  In late JL947 researchers at Bell
Laboratories put together inelegant-looking yet revolutionary gadgets
like this germanium based point-contact transistor, which could amplify
electrical signals.  At the bottom of the plastic triangle are two tiny
spots of gold separated by about two thousandths of an Inch.  When the
researchers, who later became

Nobelists for their work, touched these gold contacts to the underlying
slab of germanium and then applied a voltage to one of the contacts,
the amount of current flowing into the germanium crystal through the
other contact

Increased.  This amplification became known as the transistor effect
and became a central pillar of the microelectronics revolution.

BELL LABORATORIES

A crowd of positive charges formed near the crystal s surface.  The
internal result was a new rapid-transit pathway for electrons to flow
through the crystal and out of the other contact.  What the physicists
saw on the oscilloscope when they applied a voltage to the gold contact
was a jump in the amount of current passing through the crystal and out
of the other contact.  They had made a solid-state electronic switch
that exploited the ability of electrons in solids to leave their atomic
anchors and instead become part of a river of electrons with no
allegiance to specific atoms.  A colleague would later suggest that
they call the device a transistor.

The electronic amplification that went on inside a transistor meant
that solid-state devices made out of crystals could serve the same
function as the millions of comparatively large and energy-hungry
vacuum tubes that had been the heart of electronic technology for
decades.  Achieving the amplification effect no longer required the
heating of a metal filament inside of an evacuated glass tube to get
electrons moving: transistors did the job using much smaller currents
and in a far tinier space.

Transistors turned into a godsend for the American Telegraph and
Telephone Company, the parent company of Bell Telephone Laboratories.
AT&T managers had been growing mightily anxious about how they would be
able to meet the growing communications demand of the nation if they
had to rely on vacuum tubes for the millions of switches the growing
network needed.

After rescuing mass telephony, transistors strode forth to lead the
continuing microelectronic revolution, spurred on by the contributions
of a diverse army of researchers.  It may have been physicists who
discerned the quantum mechanical basis for transistor action, but the
microelectronics revolution that began to emerge in the 1950s and 1960s
hinged on the work of metallurgists, chemists, and others who developed
techniques for making crystals of unprecedented purity and perfection
and for miniaturizing solid-state circuitry.

The multi disciplinary effort that developed transistors into a
practical technology provided a preview of the future of technological
innovation.  Peering ever deeper into the substrate of matter, science
discovered that materials as diverse as steel and plastic arise from a
common ground that requires cooperative cultivation by a variety of
scientific researchers.  Their combined efforts continue to yield a
rich harvest of knowledge and new technologies.

PArT II

the Birth of a Superdiscipline

As the twentieth century reached middle age, the era of trial and error
in materials-related research was reaching a point where the law of
diminishing returns became operative.  The days were over when an
individual working solo stood much of a chance of coming up with
dramatically better steel alloys, or magnetic material for motors, or a
semiconductor for switches.  To people running powerful countries, this
was an alarming reality.  One big lesson from World War Il was that
staying on the strategic cutting edge of warfare technology would mean
having a reliable supply of research teams that could create materials
with unprecedented combinations of properties.

The existing pantry of materials had sprung from a multitude of
technical traditions going on in separate but parallel lines, some
extending back thousands of years.  lf so many new materials were
sprouting from such a disjointed approach as they were by the first
half of the twentieth century, what might come out of a concerted
effort to alloy the specific strengths of each materials tradition into
a more systematic super discipline

A yet-to-be-recognized multi disciplinary materials research culture
indeed had been in the midst of forming itself.  The seed crystals of
the culture were set down in places like Los Alamos, New Mexico, during
the Manhattan Project, at Bell Telephone Laboratories in Murray Hill,
New Jersey, at DuPont s polymer research group in Wilmington, Delaware,
at General Electric s research labs in Schenectady, New York, and a
handful of other industrial labs.  There also were a few seeds of the
new culture at academic venues like the Massachusetts Institute of
Technology.

In these places teams of researchers with different training and skills
were churning out new ceramic, polymeric, and metallic materials every
year.  These were places where physicists, chemists, mechanical
engineers, electrical engineers, and others with specialized skills
discovered how badly they needed one another s technical strengths.
However, these pockets of interdisciplinary synergism did not carry
sufficient momentum to trigger a more general trend.

If a new materials research culture was going to thrive, something more
had to happen.  Even as the overall U.S. scientific community was
ballooning in the 1950s, there was still no single field of research
known by something like materials science and engineering.  There were
metallurgists.  There were ceramists and glass specialists.  There were
polymer scientists.  There were solid state physicists adept in
semiconductor science or magnetic materials.  Any sense of unity
between these various categories of researchers was idiosyncratic, not
systematic.

It would have to become systematic, however.  The American future in
military arenas, the space age, jet aircraft (which were only
experimental during World War II), and semiconductor science could not
be assured by masters of any one disci the birth of a super discipline
a 89

pline.  It would take collaborations.  Masters of different disciplines
would have to transcend their traditional differences to forge more
powerful partnerships.  For the sake of national security-always one of
the most powerful drivers of action and mobilizers of resources-a new
field of materials research was more than a luxury: it was a
necessity.

Space exploration called for spacecraft that could reenter the
atmosphere intact and deliver warheads to Cold War adversaries, which
in turn required the development of refractory materials that could
take the punishment of searing rocket thrust and the frictional heat of
atmospheric reentry.  That would take chemists and ceramic engineers
who knew how to make new rugged substances.  It would take physicists
who understood how heat and friction moved through and altered matter.
It would take engineers who could see how the weight, stability,
mechanical, thermal, and other properties of candidate materials might
or might not work into the overall system of, say, an intercontinental
ballistic missile or a manned space vehicle.

Jet engines, which produce far more thrust than propeller planes
powered by internal combustion engines, demand new kinds of metallic
alloys that could take the much higher heat and stresses.  There are no
shoulders in the jet stream.  The added thrust came only when rapidly
spinning turbines could compress air and fuel into the engine s
combustion chamber.  Alloys that could remain strong while spinning
furiously at temPeratures that brought all known metals perilously
close to a taffy like state would be the only way to feel confident
that the turbine blades of a jet engine would not fail
catastrophically.  It would take metallurgists, aeronautical engineers,
physicists, and other specialists to invent the necessary alloys, super
alloys and manufacturing processes.

Solid state semiconductor research and development required the finest
minds in theoretical physics to reveal how electrons move in a crystaPs
regimented labyrinth of atoms, chemists who could develop ways of
producing the world s purest crystals and electrical engineers who
could work these crystalline semiconductor devices into electronic
circuits useful for radios, weapons, telephone switching, and other
technologies.

Postwar optimism about harnessing atomic energy for the peaceful
purpose of generating electricity led to the fanciful notion that
electricity could become too cheap to meter.  But making any
electricity from the energy of nuclear fission would entail managing
the eroding, embrittling, and otherwise adverse consequences that high
amounts and long durations of radiation have even on inanimate
materials.  In 1956 a high-level Atomic Energy Commission advisory
board even made a present recommendation for addressing the many
materials issues that go along with harnessing nuclear energy for power
generation.  The board recommended establishing Materials Research
Institutes in which no single material would be the focus of study, nor
would one single traditional discipline dominate.

Another top-level advisory and planning organ that emerged at this time
was the White House s Coordinating Committee on Materials Research and
Development (CCMRD).  Its very name reflects a conceptual unification
of materials.  Its members were the heads of materials-related research
in each of the federal agencies involved in science and technology. The
charge of the CCMRD was to discuss strategies that would keep a steady
flow of essential innovations in materials available for national
security.  Their recommendation was to create academic environments in
which materials problems could be approached using tools and theories
from whatever scientific or engineering discipline was pertinent.

The message that improving existing materials and creating new ones had
become a skeleton to modern technology development was spreading
throughout the federal science and technology community after World War
11.  The Office of Naval the birth of a super discipline 0 91

Research was hearing from its own Solid State Sciences Advisory Panel
that materials research could address many pressing issues, including
the sometimes catastrophic cracking of the welded-steel hulls of
hundreds of World War II Liberty ships, the corrosion of metals in the
salty ocean environment, and propulsion at sea.  The Air Force
conducted a study in 1957 that recommended the creation of a National
Materials Laboratory much like the Materials Research Institutes that
the Atomic Energy Commission had envisioned.  In such a place, the
development of advanced aerospace materials for jets could be conducted
in a novel interdisciplinary environment better equipped to solve the
new materials challenges.

In 1958 these and many other lines quickened toward an historical
convergence.  On March 18 President Eisenhower s science advisory
committee handed him a short background paper that articulated the
sea-change in the field of materials.  Unique environmental conditions
associated with rocket propulsion, nuclear reactors, space flight, and
vehicle re-entry have established the need for materials which are not
currently available, the paper stated.

The key to the new materials needed was to be found in the kinds of
scientific advances that led to the transistor and to an understanding
of precipitation hardening in metals.

There is a general feeling ... that such advances in sciences as these
can lead to a technology of materials engineering quite different from
the metallurgy, ceramics, and polymer technology of today, the paper
stated.  But to achieve this result it will be necessary to begin to
relate the new fundamental knowledge of matter to the behavior of
highly complex materials.  This in turn will require scientists and
engineers from many different disciplines-organic chemistry, physical
chemistry, metallurgy, and solid-state physics-to associate their
special knowledge and different points of view.  Therein lies a call at
the highest governmental level to form a new field of materials science
and engineering.

The very same call was being heard on the local level as well.  Just
weeks before the president s science advisory committee members sent
their short manifesto to the President, Morris Fine, then a young
professor in the Department of Metallurgy at Northwestern University in
Evanston, Illinois, had written a memo to the school s academic dean.
He and his two departmental colleagues wanted to change the name of
their department to the Department of Materials Science.  There was no
precedent in the academic world for the metallurgy professors
request.

Fine made his case for the departmental name-change in a letter to the
dean.  In many fields the application of scientific discoveries to
technical practice has been limited almost exclusively by the problem
of the materials to be used, Fine wrote.  This is strikingly
demonstrated by the currently important fields of missiles and nuclear
energy, where the scientific principles have been known for a decade or
longer.  In these fields solution of the materials problems is the key
to rapid development.  Solving known problems is only the half of it,
Fine continued.  When materials problems have been solved, new
properties of a particular material discovered, or when new materials
have been found, whole fields of technological importance often are
opened up.  2

As for what the departmental name change would actually mean in
practice, Fine had this to say to the dean.

Traditionally the field of material science has developed along
somewhat separate channels-solid state physics, metallurgy, polymer
chemistry, inorganic chemistry, mineralogy, glass, and ceramics
technology.  Advance in materials science and technology is hampered by
this artificial division of the whole science into separate parts.  The
advantages of bringing together a group of specialists in the various
types of materials and allowing and encouraging their cooperation and
free interchange of their ideas would indeed be very great.

The problem was that, metallurgists physicists, chemists, engineers,
and other materials researchers residing in different disciplines and
departments all had traditionally kept to their own.  Each had
developed different ways of thinking, published their work in different
journals, read different texts, used different jargon, and essentially
lived under different research cultures with different mores and
measures of accomplishment.  At Northwestern, the Materials Science
Department was supposed to be a different animal, a seamless hybrid of
many disciplines with a long history of parallel development and almost
no history of joint development.

Fine, who never left Northwestern after joining in the late 1940s, had
the right kind of experience to back his then audacious plans to open
up the country s first academic materials science department.  He was
one of the few scientists at the time who knew firsthand just how
powerful and fruitful multi disciplinary research can be.  The most
powerful weapon and technology humanity ever had produced came out of
the Manhattan Project at Los Alamos where Fine worked on the plutonium
metallurgy team under Cyril Smith.  After the war, he was recruited
directly from Los Alamos to Bell Telephone Laboratories for the
explicit purpose, he recalls, of bridging the physics and metallurgy
researchers there.  One of the recruiters was William Shockley, who,
with John Bardeen and William Brattain, won the Nobel Prize in 1956 for
inventing the transistor and who later seeded what became known as
Silicon Valley.

Fine s own first efforts at Bell Laboratories focused on the
crystallization of germanium, which was the flesh of the first
transistor, and on silicon, which became the semiconductor material
that changed the world as much as any material in history.

Fine s argument to his dean was persuasive.  On December 29, 1958, the
Department of Metallurgy officially changed its name to the Department
of Materials Science."  A new research tribe was born.  Northwestern
University became one of the first academic departments to deliberately
train students not only as metallurgists or ceramic engineers or
polymer chemists but as materials scientists.

Meanwhile, the six-hundred-word paper by President Eisenhower s science
advisory committee had set a policy machine in motion.  Just as
Northwestern s materials science department began its life in late
1958, all of the federal advisory committees and bureaucratic report
writing began bearing fruit.  At the focus was Herbert York, then the
director of the newly formed Advanced Research Projects Agency (ARPA)
with a Cold War charge to make sure that the country did whatever it
would take to develop intercontinental ballistic missiles.  In a
highpowered meeting between officials from the Atomic Energy Commission
and the Coordinating Committee for Materials Research and Development,
York was convinced that the creation of well-supported
interdisciplinary laboratories (IDLs) for materials research at
universities was a sound course for the government to take and for ARPA
to fund.

For one thing, the creation of such an R&D infrastructure could only
help ARPA!s mission to build ICBMs as well as future systems deemed
important for national security.  After all, planners envisioned ICBMs
as heaving nuclear warheads through the atmosphere into outer space,
shuttling the weapons halfway around the world to some target, and then
plowing again through the atmosphere intact to deliver their
destructive cargo.  If anything, this would be an intensively materials
dependent challenge.

Within a year of taking on the responsibility, the first three IDL
contracts had been drawn up and signed by Northwestern University,
Pennsylvania State University, and Cornell University.  All three
already had adopted the interdisciplinary mind set to some degree, so
the 1DL seeds were sown on fertile soil.

The mission of these labs was about as comprehensive as the birth of a
superaiscipline a 95

lab missions could get.  The original IDL contract stated the objective
as furthering the understanding of the factors which influence the
properties of materials and the relationships which exist between
composition and structure and the behavior of materials.  The
deliberate generality of the contract s language helped bring all
classes of materials-metals, ceramics, polymers, and so onwithin a
single conceptual framework.

In the early 1960s nine more ARPA contracts, three Atomic Energy
Commission (AEC) contracts, and two NASA contracts expanded the nation
s interdisciplinary materials research infrastructure throughout the
country.  By 1972, when the National Science Foundation took over the
IDLs from ARPA and renamed them as Materials Research Laboratories,
materials research had become a genuine field at many major
universities, not just those with special government funding.
Single-discipline departments that for decades had gone by names like
metallurgy or minerals and mining were falling under broader-based
materials departments.

As these laboratories were getting off of the ground, yet more
institutional cornerstones for the new field of materials science and
engineering were poured and cured.  One of these was put into place in
the spring of 1973 when the Materials Research Society (MRS) held its
inaugural meeting at Pennsylvania State University.  Rustum Roy, a
ceramic materials expert at Penn State, founding member of MRS, and
perennially one of the most influential and colorful members of the
materials research community, recalled that just under three hundred
people, including Penn State faculty and students, attended.

A humble beginning, to be sure, but the creation of professional
societies is a hallmark for budding scientific disciplines because it
provides a core from which its members can further develop a community
sensibility.  Since 1960, when those first IDLs were commissioned, the
new breed of interdisciplinary

Pages: 96-99

FIGURE 10

Twentieth Century Stuff.  This fifty-year time line beginning in 1938
shows milestones in the invention, introduction, applications, or
commercialization of various materials and materials making
processes.

> First epoxy resin patent granted.  > Teflon( TM the product of the
polymerization of tetrafluoroethylene, developed.  > High-pressure
polymerization of ethylene introduced.  > W. H. Carothers' invention of
nylon announced.  > Polyvinyl butyral resin sheeting introduced, making
automobile safety glass possible.  > Thermosetting binders
("warm"-pressing ceramics) patented.

> Low-density polyethylene (LDPE) commercially produced.  > Cellulose
acetate molding compound and ethyl cellulose produced.

> Formica( TM and poly vinyl chloride (PVC) produced.  > Bomber noses
of Plexiglas acrylic resin are thermoformed for war planes.  > Film and
sheet vinylidene chloride introduced.  > Jet-molding machines
developed. > Prototype plastic car developed, made of soybeans and wood
fiber processed into plastic, offers improved impact resistance.

> Fiber-forming polyesters developed.  > Polyethylene terephthalate
(PET) invented.  > Superconductivity demonstrated at 15 K. >
Polyacrylonitrile (PAN) fibers (Orlon{TM}) commercially introduced.  >
First polyethylene bottle blown.  > Thermoformed plastic blister
packaging developed.  > First working electromechanical digital
computer developed.

> Saran monofilament extruded.  > Free-blown acrylic aircraft canopy
produced.  > Silicone resins produced.  > E-glass (electrical glass), a
type of fiber glass, patented.  > Injection blow molding process
patented.

> BT-15 airplane with fuselage fabricated in a fibrous-glass-reinforced
plastic sandwich structure fuselage flown.  > First
low-pressure-thermosetting prep regs developed.

> Dielectric preheating introduced.  > Zytel(TM), an extremely tough
nylon molding and extrusion resin, introduced for use in gears and
cams, sports equipment, valves and other products.

> Acrylic introduced in line of dentures and automobile taillight
lenses.  > Polyethylene appears in kitchenware.  > Nylon zipper
introduced.  > Polyester developed.  > ENIAC, the first all-electronic
analog computer, introduced.  Weighing about 30 tons, it required 150
kW to operate its 18,000 vacuum tubes and other devices.

> Vacuum melting technology developed.  > Etched electronic circuit
created.  > Germanium transistor developed, beginning a trend toward
miniaturization.  Its first practical application was in hearing aids.
> Epoxy introduced commercially in the US as an adhesive.

> Acrylonitrile-butadiene-styrene (ABS), used to manufacture luggage,
introduced.

> Screw plastication with automatic transfer molding developed, using a
non reciprocating screw.  > "Doctor-blading" process of forming thin
ceramics patented.

> Commercial production of Teflon( TM begins.  > The International
Standardization Organization establishes Plastics Committee.  > Dacron(
TM polyester fiber for wrinkle-resistant, permanent-press clothing
introduced.  > Commercial production of titanium begins.

> Injection-molded model aircraft hobby kits produced.  > Commercial
production of high-purity, single crystal silicon begins.  > Field ion
microscope developed.

> Catalysts for low-pressure polymerization of ethylene developed.  >
High-impact, rigid poly vinyl chloride developed, making PVC pipe
possible.  > Fiberglass-reinforced polystyrene developed for military
and commercial applications.  > Mylar( TM polyester film introduced,
for recording tape, shrink-wrap packaging, and electrical insulation. 
> Basic oxygen furnace for steel making developed.  > Silicon solar
cell introduced.

> Polycarbonate (PC) invented.  > First Dacron( TM (PET) plant begins
production.  > Chevrolet Corvette uses 41 fiberglass reinforced parts,
including 17 major assembly components.  > Early large-scale
integration circuitry developed by layering wafers of miniaturized
electronic circuits.  > Tungsten-nickel-copper composite first
developed and used for radio-active substance containers.

> Propylene (PP) is polymerized.  > Molybdenum-manganese metal coating
patented for use in vacuum-tight, metal-to-ceramic seals.

> High-density polyethylene (HDPE) developed.  > Hypalon(TM) synthetic
rubber introduced, offering advantages over rubber.  Used in roofing,
auto, home, and industrial applications.  > Cronar(TM) polyester film
base for x-ray and graphic art films developed.  > Hot isostatic
pressing technology developed.

> Polyphenylene oxide (PPO) discovered.  > Lucite( TM acrylic lacquers
introduced.

> Widespread automotive use of urethane foam for seating and safety
dashboards inaugurated.  > Polyacetal resins developed, offering high
strength with toughness of metal for use in hardware, auto parts, cams,
rollers, and zippers.  > Soviets launched Sputnik.  The space race
begins.  > Silicon transistor developed, permitting devices to operate
at higher temperatures and frequencies with greater stability.

> Lycra( TM spandex fiber introduced.  Stretch up to five times its
length and returns to original length.  > Ruby-crystal laser developed.
> First commercial production of micro alloyed columbium steels.

> Commercial production of acetal homo polymer begins.  > Integrated
circuit (IC) developed with multiple silicon transistor circuits on a
single silicon wafer.  > Molybdenum-wire-reinforced-titanium matrix
composite developed.

> Ethylene vinyl acetate copolymers introduced.  > Acetal copolymer
used in automobile brake-cable pulley.  High-impact styrene furniture
legs with through-bolt construction developed, starting trend to
plastic furniture.  > Glass-ceramic patented.  > Amorphous metal alloy
produced. > Synthetic diamond production begins.  > Aircraft with
fiberglass-polyester skin and paper honeycomb core first flown.  >
Advantages of parallel metal grain boundaries for turbine blades
demonstrated.  > Effects of solidification process first reported.

> Tape process of forming thin ceramics patented.

> Polyvinylidene fluoride introduced.  > School furniture introduced,
combining solid and laminated plastic in a tubular steel construction.
> Transparent polucrystalline alumina patented.  > Scanning electron
microscope (SEM) developed.  > Nickel-based super alloys using
oxide-dispersion strengthening announced.  > Gallium arsenide laser
diode developed.

> Polyimides introduced, increasing the thermal endurance of
thermoplastics to 750 degrees F. > Union Carbide process invented for
low-pressure structural-foam processing.  > Nomex( TM flame-resistant
aramid fiber and paper introduced for protective clothing,
highperformance hoses, high-temperature electrical uses.  > Float
process for making glass patented.  > Sintered alumina abrasive grain
patented.

> Polyphenylene oxide (PPO) components for appliances electrical
connectors produced.  > Surlyn( TM ionomer resins introduced, offering
clarity of glass and toughness for use in food packaging, sporting
goods, and auto motives > "Certi-Fired" thick film materials for
electronic circuit miniaturization introduced.

> Polysulfone developed and introduced, finds commercial uses in
electrical components.  > Glass-reinforced styrene-acrylonitrile (SAN)
copolymers appear in automobiles.  > Clysar(TM) shrink film used in
cold packaging applications.  > Process patented for making dense
impregnated, silicon-carbide articles.  > Cobalt-rare earth magnets
developed.

> Noryl(TM), a modified polyphenylene oxide (PPO), introduced.  >
Uniloy machine for high-density polyethylene blow molded milk bottle
production introduced.  > Acrylonitrile-butadiene-styrene (ABS) used on
exterior surfaces of helicopter.  > Kapton(TM) poly imide film,
offering resistance to moisture and extreme temperatures, is developed
and becomes valuable for aerospace use.  > Fiber optics developed.

> Polyarylsulfone introduced as Astrel 360(TM).  > Permasep( TM
permeators introduced as an economical way to desalt water.  > The
first composite aircraft to receive FAA certification, is built.  >
Single crystal turbine blade is demonstrated.

> Polyphenylene sulfide (PPS) introduced as Ryton( TM > Chrome plated
polypropylene used in auto motives > Manned moon landing introduces
plastic components to new applications.  > Riston(TM) photopolymer dry
film resists are introduced as light-sensitive polymer films for
printed circuit board production.  > Very large-scale integration
(VLSI) electronic circuitry developed.

> Corian(TM) introduced as a stain-, scratch-, and burn-resistant
non-porous material for countertops, sinks, etc.

> Boron-epoxy horizontal stabilizer on the F-14 represents the first
advanced composite part produced that was designed as a composite part
and not as a substitute for metal.

> Mechanical alloying developed.  > Metal injection molding developed. 
> Zinc oxide varistor patented.

> Polyethersulfone used on aerospace and automotive applications.  >
Robotics used on the plastics industry as the first high-speed,
machine-mounted, automatic part remover for injection is patented.  >
Imron(TM) polyurethane enamel introduced for automotive applications. 
> Sialons developed.

> Kevlar( TM aramid fiber introduced, which, on a weight basis, is five
times stronger than steel.  Used in sporting goods, industrial
products, bullet-resistant vests, auto motives aerospace, etc.  > NASA
Langley Research Center initiates flight-service program placing carbon
epoxy assemblies into service.

> Carbon-reinforced epoxy upper aft rudders introduced in MacDonnell
Douglas DC-10.  > Acrylic sheet, stiffened with reinforced plastic used
for all exterior body panels in an automobile.

> Reaction-injection molded urethane held on place by a
glass-reinforced polypropylene sheet retainer is used as the front end
of the Chevrolet Monza.

> All-plastic jet designed.  > Plastic microwave cookware becomes
available to the consumer market.  > Amorphous silicon solar cell
introduced.

> Commercial production of linear low-density polyethylene (LLDPE)
begins.  > Polyphenylsulfone (Radel {TM}) introduced.  > First polymer
demonstrating electrical conductivity is synthesized.

> Polyetheretherketone (PEEK), a high-temperature resistant material,
becomes available for aerospace and computer applications.  >
Polyarylates introduced.  > The practical use of optical fibers
begins.

> Volume production of plastics surpasses that of steel.  > The
Gossamer Albatross, made in large part of Mylar( TM polyester film,
becomes the first human-powered aircraft to cross the English Channel. 
> Rynite( TM a heat-resisting, very rigid polyethylene there phthalate
(PET), introduced for use in electronic components, furniture, lighting
fixtures, industrial machines, auto motives > Tungsten-fiber-reinforced
ferro-chromium-aluminum-yttrium used experimentally for jet engine
rotor blades.

> Acid-leaching process introduced for producing 99.6% to 99.9% silica
fibers that resist devitrification up to 1,370 degrees C. Used as
insulation for the Space Shuttle.

> Pyralin(TM) Polyimide coatings used as insulation in semiconductor
chips.  > Superconductivity used in magnetic resonance imaging medical
equipment.

Modified poly imide poly ether-imide) (Ultem{TM}) introduced for use in
high-performance fiber optic components, co extruded food packaging,
and advanced composites.  High-purity polycarbonate resin for
manufacture of compact disks ("CD's") introduced.  >Jarvik-7 artificial
heart designed, made largely of plastics, supports patient for 112
days.  > First space shuttle launched.  > Fusion welding of
silicon-carbide-reinforced aluminum demonstrated.  >
Neodymium-iron-boron magnet developed.

> Conductive composites for computer housings are developed as FCC
regulations require shielding of plastic housed electronics components.
> Alumina-silica-fiber-reinforced aluminum piston for diesel engines
developed.

> Melt-processible liquid crystal polymers (LCP) introduced
(Xydar[TM]).  > Five ship sets of composite horizontal stabilizers
installed on the Boeing 737.  > Plastic fuel tank for U.S. passenger
car is blow molded and treated with sulfonation process to control
hydrocarbon permeability.  > The twin-turboprop Avtek 400{TM] promotes
the art of advanced composites.  > Selar poly amide barrier resins
introduced to blow-molded bottles used for containing chemicals and
hydrocarbons.  > Quasi-periodic crystals discovered.

> Bexloy[TM] engineering resins introduced for low-weight,
high-strength auto body parts to replace metal.  >
Alumina-fiber-reinforced aluminum experimental squeeze casting
introduced.  > Graphite-fiber-reinforced magnesium vac-assisted
investment casting introduced.  > Electronic discharge machining of
metal matrix composites demonstrated.

> The Voyager, a composite aircraft, flies around the world without
refueling.  > Large billets of both silicon-carbide-particle-reinforced
aluminum and silicon-carbon-wire reinforced aluminum produced.  >
Alumina-fiber-preform-reinforced aluminum introduced.  > Molten metal
oxidation method used to produce net shapes.  > Commercial production
begins of alumina-silica-fiber-reinforced aluminum pistons by squeeze
casting.  > Boron-fiber-reinforced aluminum used in sports equipment. 
> Water jet and laser cutting of metal matrix composites
demonstrated.

> Silicon-carbide-particle-reinforced aluminum, optical-grade material
developed.  > Discharge welding of silicon-carbide-fiber and
wire-reinforced aluminum demonstrated.  >
Silicon-carbide-particle-reinforced aluminum formed by standard forming
technique.  > Silicon-carbide-fiber-reinforced titanium and
silicon-carbide-fiber-reinforced titanium aluminum experimental metal
matrix composites developed for temperatures near 2000 degrees

F.

> Polymer semiconductors developed.

> The first all composite business jet, Beech Starship, goes into
commercial production.  > Development of high-purity, low-defect indium
phosphide single crystals, which has high potential for use in
space-based solar cells.

100 0 S T U F r materials researchers have fanned out to nearly every
university, hundreds of government laboratories, as well as thousands
of small and large corporate research laboratories and plants.  The
tribe has spread throughout the land.  This has been true not just in
the United States but throughout the world.

To get an idea of the growth in the field since the MRS was founded,
consider that in 1986, when the Materials Research Society began
publishing the Journal o Materials

Of Research, MRS membership had expanded from its original 1972
membership of 215 to over 4,000.  By 1996 MRS membership was over
12,000.  And MRS is only one of many materials oriented membership
organizations.  A more comprehensive measure of the scope of the field
of materials science and engineering comes from the membership of the
Federation of Materials Societies (FMS), an umbrella organization
representing more than a dozen large and small materials-related
societies and association.  FMS claims about 750,000 members.  The
American Chemical Society (ACS) and its 150,000 members is among the
larger organizations on the roster.  The smaller FMS members, whose
constituencies number in the thousands, are more specialized and go by
names like the American Welding Society and the National Association of
Corrosion Engineers.  There are dozens of other materials related
societies outside of the FMS, so that 750,000 is almost certainly a
conservative estimate.

These materials researchers already have succeeded in changing the way
humanity discovers new materials, incorperating trial and error but
advancing beyond it to a new approach to materials based on the way
their structure and anatomy relate to their properties and
technological performance.  Instead of always limiting designs to the
known performance limits of available materials, engineers have begun
to take the audacious step of specifying the new materials they need
for their ever more ambitious technological designs.  If a the birth of
a super discipline E 101

desired material doesn t exist, then materials researchers ought to be
able to figure out how to make it.

Among the historic precedents for this approach is the Manhattan
Project, whose charge to create nuclear weapons was tantamount to a
challenge to develop new materials like the radioactive metal
plutonium.  Another came a few years earlier in the form of the
government s GR-S Program whose goal was to create almost overnight a
massive industry for making hundreds of thousands of tons of synthetic
rubber to replace embargoed supplies of natural rubber.  President
Kennedy s successful Apollo program confirmed this philosophy again in
the 1960s.  These projects showed how a combination of basic materials
science and engineering research had begun a synergistic spiral toward
ever greater control over the material world.

Despite these gains, materials science and engineering remains an
obscure outpost of the country of science.  People on the street have
heard about biologists, chemists, and physicists-but materials
scientists?  In 1990 this unseen research community made its boldest
bid to earn some name recognition by producing a book-length report,
Materials Science and Enginee7ingfor the 1990s.  Known as the Blue
Book, it makes the case that the U.S. economy is so heavily dependent
on advances in materials that research in the field ought to become a
national priority.

Without new materials and their efficient production, our world of
modern devices, machines, computers, automobiles, aircraft,
communications equipment and structural products could not exist, the
MS&E report declared.

This is really a matter of old news striving for contemporary
attention.  The ancient Phoenicians knew it, traveling to southern
England for the tin they needed to manufacture bronze weapons.  It was
clear to John Roebling in the nineteenth century, who saw in steel the
possibility of suspending a beautiful bridge across the East River from
Manhattan to Brooklyn.  It was clear to Wallace Carothers and his
managers at DuPont

Metallic Forests.  These dendritic tree like micro structures formed
during a welding process Involving a nickel-based alloy, the kind of
mechanically strong alloy used to make the turbine blades In the
engines of airliners.  SYmmetrical structures of this kind are produced
as the already solidified portion of a weld forms side branches while
penetrating into the still molten portions of the weld.  Revealing
structural details of this sort is a critical part of the quest to
improve the performance of materials, Including welded metal parts.  In
this scanning electron microscope photograph, the hot liquid phase has
been expelled to reveal the dendritic structures.  SEM MICROGRAPH

COURTESY OF S. A. DAVID."  L. A. BOATNER, AND R. STEELE OF

OAK RIDGE NATIONAL

in the 1930s, who saw a technological and commercial bonanza in polymer
molecules.  It was a given for aerospace engineers at Northrup and
other aerospace companies in the 1980s and 1990s, who developed more
than nine hundred new materials for the impossibly expensive B-2
Stealth bomber.

But today s avant-garde is often tomorrow s establishment; patiently
pushing the capabilities of technology s flesh and bone, the obscure
visionaries of contemporary materials science might well be the
architects of the future s commonplace miracles.

individual components could sit within a biological cell.  When you
make a graph that shows how the number of transistors on a
fingernail-size piece of silicon crystal has increased over the years,
you get an upward sloping curve that gets steeper and steeper.  If you
were to walk the curve, it would be like walking up a ski ramp.

This feat of microelectronics fabrication would not have happened
without materials scientists.  Crystal growth experts were needed to
make silicon crystals nearly devoid of the tiny defects and
contaminants that had made the first transistors too unreliable for
military planners to rely upon.  It took polymer scientists to develop
the photo lithographic materials required for patterning circuitry onto
the silicon wafers cut from those crystals.  It took the expertise of
ceramic engineers to develop materials to entomb the delicate
microcircuitry so that it can be handled and put into products.

Since the birth of materials science and engineering, materials
researchers have consistently improved materials that have in turn
enhanced the performance of the technologies that use them.  If you
plot the progress of some measure of performance (engineers call such
things figures of merit) over time for each technological function like
electrical conductivity or optical transparency, you get another
ski-jump ramp.  The steepness of the ramp really takes off in each case
when a material receives the sobriquet of advanced material, which
denotes an

FIGURE 12

The Ski Ramps of Progress: The graph shows how the progression from
wood and stone to various metals to the advanced polymers (aramid
fibers) and ceramics (carbon fibers) of today has yielded materials
that grow ever stronger while using ever smaller amounts of mass.
Similarly accelerated rates of improvement in many material properties
in recent years yields simliar ski-ramp curves of progress. 
REPRINTED

WITH PERMISSION FROM MATERIALS SCIENCE AND ENGINEERING

FOR THE 1990S'MAINTAINING COMPETITIVENESS IN THE AGE OF MATERIAL
(WASHINGTON, DC: THE

NATIONAL ACADEMY OF SCIENCES PRESS).

materials research comes of age 0 107

unprecedented leap in superiority over previous products.  (See Figure
12.)

Take the so-called strength-to-density ratio of materials, a measure of
how much of a material it takes to support a given load.  The higher
the ratio, the less material you need.  The strongest thing going in
Antoine Lavoisier s time in the late eighteenth century was good cast
iron.  A cast-iron rod with a square cross section one inch on a side
and weighing 4 pounds per foot could support up to 50,000 pounds (25
tons), or about 330 adults.  Cast iron s strength-to-density ratio
actually is not much better than bronze or, for that matter, wood or
stone.  In the nineteenth century steel started to push the strength-to
density into ski-jump levels.  But not until the 1950s-when scientists
invented polymer-based composites reinforced with

Og8 rA

Z W a

LLI

A, id ibers

Composites

Wood, M. cast onze Iron St.  I

Stone

Aluminum

1800 1900 2000

YEAR

glass and other ceramic fibers and high-strength polymers such as
Kevlar (a poly aramid fiber)-did the ramp get really steep.  Modern
high-strength materials-say, a cable made of the carbon fibers in
military aircraft or high-end golf clubs-could support the same three
hundred people as Antoine Lavoisier s cast-iron rod, but the modern
cable would only have to be one third of an inch on a side and one
sixtieth as heavy.  It also would cost a whole lot more.

The historical trend of another figure of merit-the highest temperature
at which an engine can operate without its metal parts weakening,
melting, or otherwise giving up the ghost also reveals the
technological meaning of materials research.  The higher the operating
temperature, the greater the engine s efficiency.  In 1900 steel steam
engines operated at the boiling point of water, about 212 F. In the
1930s air-cooled airplane engines were running about 380 F. Since then
metallurgical researchers at aerospace engineering companies have
developed super alloys (grown as single monolithic crystals of metal in
the shape of turbine blades) that remain strong enough even above 2,000
F. Putting heat-immune ceramic coatings atop the super alloys could
push that operating temperature up several hundred more degrees.

The litany of ski-jump ramps goes on.  In 1911 when the Dutch physicist
Heike Kamerlingh Onnes chilled mercury to the temperature of liquid
helium (-452 F, or about four degrees above absolute zero or 4 K), he
discovered the phenomenon of superconductivity.  Electricity moves
through a superconductor with no resistance whatsoever.  If you set a
current going in a ring made out of a superconducting material, it will
go around the ring forever.  Unfortunately, if Kammerlingh s mercury
got any warmer than the liquid helium, it stopped being a
superconductor.

For the next seventy-five years researchers searched for materials that
remained superconductive at warmer temperatures, since refrigerating
anything to liquid helium temperatures is costly and difficult.  They
did, but the warmest superconductor they could find-a niobium germanium
intermetallic material discovered in the 1970s-still lost its
superconductivity

2 at417 F (23.3 K).

The shallow upward slant of the previous seventy-five years became a
ski jump in 1986, when two scientists at IBM s Zurich Research
Laboratory started a research wildfire with their discovery of a
ceramic material-a barium lanthanum copper oxide-that remained
superconductive at the unexpectedly warm temperature of about396 F (35
K).  They and others soon learned that the barium lanthanum copper
oxide was just the first of an enormous class of superconductive
ceramic materials.  By 1996 a growing army of materials researchers
discovered ceramic superconductors with transition temperatures as high
as216 F (135 K).  The search for yet warmer superconductors
continues.

There is a catch.  The ceramic superconductors are more like rocks than
like the earlier metallic superconducting materials.  It is hard to
make a rock into a wire, which is the kind of shape you want to work
with for electrical applications.  Nonetheless, researchers have made
experimental cables and electronic circuits made with the high
temperature ceramic superconductors.

Perhaps the most dramatically steep and rapid improvement in a material
is the transparency of glass.  As late as 1965, no light sent into a
thousand-meter length of glass fiber drawn from the best glass
available would make it to the end.  At the time glass fiber was far
more useful for making insulation or as part of a polymer-glass
composite for things like boat hulls than it was for piping light.
Since then, however, materials researchers have developed ways of
manufacturing endless miles of optical fibers so devoid of the material
imperfections that used to darken the light that almost all light put
in one end will shine through the other.

The transparency of optical fibers has increased by one hundred orders
of magnitude-that is, by a factor written with a I followed by one
hundred zeros.  The hair-fine fibers strung across late-model
transoceanic optic fiber cables can carry one hundred and twenty
thousand phone calls at once.  Today s best optical fibers can relay
data at a rate of billions of bits per second, the equivalent of an
entire Encyclopedia Britannica every second.  This is the kind of
ski-jump ramp that leads to talk about information highways.

There are many more ski-jump ramps that show how far materials
researchers have come.  Today, there are neodymium iron-boron-based
magnetic materials a hundred times stronger than the magnetic steel
alloys used at the turn of the century.  Every increase in the magnetic
field leads to smaller and more efficient motors for everything from
heating systems to windshield wiper blades.  There now are tool
materials such as cubic boron nitride and tungsten carbide coated with
synthetic diamond that enable manufacturers to cut, shave, and mill
materials into products a hundred times faster than the tool steels of
the early century.  Tools that work a hundred times faster can produce
a hundred times more product in the same time or the same amount of
product in one one-hundredth the time.

The super discipline of materials science and engineering (MS&E) has
its ski prints all over these ski-jump ramps.  Evidence of the
consequences is manifest throughout today s hightech society.  Whenever
materials researchers have introduced harder, stronger, lighter, or
more electrically conductive materials-or for that matter, any
affordable material that is better in some way than all available
materials-more capable technologies follow.  With better materials come
faster jets, higher capacity communications, more powerful computers,
more energy-efficient cars and appliances, longer-lasting and more
capable medical prostheses, and more woeful and devastating weapons, to
name a few consequences.  (See Figures 13 and 14.)

The rocketlike acceleration in the performance of materials in the past
four decades is due to an unprecedented three-way ma teri ats researck
comes of age 0 111

bootstrapping: the ability t o reveal the structure of materials and
how they form; the ability to form materials into specific micro
architectures and the ability to understand, model, and simulate
material phenomena using math and computers-in short, analysis,
synthesis, and theory, respectively.

Just as diversity exploded in the biological realm during the Cambrian
revolution, so the diversity of analytical tools and methods has
proliferated in materials science in recent decades, giving researchers
a seemingly unobstructed view of the anatomy, formation, and
transformation of materials to atomic levels of detail as they witness
processes at time intervals as short as quadrillionths of a second.

As each member of this triumvirate-analysis, synthesis, and theory-has
diversified and ramified, it has boosted the sophistication of the
other members.  The microelectronics revolution has raised the water
level of all three.  Computers and microelectronics have enabled
researchers to control their instruments with far more finesse, to
gather far more data, and to make sense of data far more quickly; and
they have promoted better control of the reactions and processes by
which ingredients become materials.  Manufacturers constantly test
theorists ideas in an iterative loop that benefits both theory and
practice.  The result: materials science and engineering has achieved
mastery of the material world that was unthinkable in 1960, when the
field was born.

Here is a brief survey of some of the analytical methods, synthetic
methods, and theoretical and computational tools that have fueled this
ascendance.

Analysis.  Henry Sorby could use optical microscopes to see structural
detail down to the point where he could distinguish grains the size of
cells or thereabouts.  By 1912 physicists were using X-ray
crystallography to examine details of the average crystalline structure
of metals and other solids.  In the 1930s the advent of electron
microscopy opened vistas between optical microscopy and X-ray
crystallography.  In 1951 a scientist at the University of Pennsylvania
saw some of the first human glimpses of individual atoms with his
invention-the field ion microscope.  It worked by using intense
electric fields to yank atoms off of the end of a stylus-shaped sample
and then accelerating the atoms into a detector that could reveal the
relative locations of the liberated atoms.

One variation of the field ion microscope essentially shoots movies of
a sample as atomic layers sequentially evaporate off and hit a
detector.  After enough layers have come off, a three dimensional
portrait in atomic detail emerges the way a CAT scanner can construct a
three dimensional image of the inside of a person from many
two-dimensional X-ray slices.

Along the way, too, have come various spectroscopies.  The first of
these techniques goes back to the nineteenth century, when scientists
were first learning to harvest chemical information from the colors of
light emitted from or absorbed by their samples.  Since then the
battery of spectroscopic instruments and methods has grown to
astounding versatility.  All of them

The Sculpture of Ceramic Structure.  Modern synthesis and processing
techniques can produce materials in an endless variety of forms.  Top
right: The scanning electron microscope photograph (SEM) depicts a
particularly sculptural feature from a sample of zinc oxide crystals
produced by a high-temperature growth technique known as chemical vapor
transport.  The process is akin to the formation of Ice crystals when
the moisture In your warm breath blows over a wintry window pane.  Zinc
oxide ceramics are Important electronic materials that are used for
electrical surge protection In devices ranging from large-scale power
distribution systems to consumer electronics such as televisions and
home computers.  Bottom right: This SEM photograph shows a porous micro
structure In potassium tant alate ceramic, whose underlying crystal
structure, known as a perovskite, is a building block for a number of
important compounds, including ferroelectric ceramics that can store
high quality images and high-temperature superconductors.

FIGURE 13: L. A. BOATNER AND R. STEELE, OAK RIDGE NATIONAL
LABORATORY,

OAK RIDGE, TN.

FIGURE 14: L. A. BOATNER, J. 0. RAMEY, AND R. STEELE, OAK RIDGE

NATIONAL LABORATORY, OAK RIDGE, TN.

FIGURE 13

FIGURE 14

measure the wavelengths, intensities, or variations in visible or
invisible light, or in energetic particles such as electrons and
neutrons going in or out of samples.

There is femtoseeond laser spectroscopy, in which laser pulses in the
vicinity of a million billionth of a second can prod and probe the
detailed events that occur as atoms or molecules approach one another,
react, and then diverge.  The result is like a slow-motion picture of a
great play or blunder in a football game, only in this case the detail
helps materials researchers know what structures are worth trying to
emulate or avoid as they try to make new materials or improve old
ones.

There is low-energy electron loss spectroscopy, in which changes in the
speed and direction of slowly moving electrons directed at samples
reveal subtle motions and dynamics occur ring at the surfaces of
materials such as industrial catalysts.

There is laser ablation mass spectrometry, which can take a speck of
sample barely visible to the naked eye, train a laser onto a tiny
subregion of that speck, and then blast part of that region into a
molecular plume.  The mass spectrometer then can weigh and thereby
chemically identify those plume components by monitoring how little or
how much they deflect in a magnetic field.  These kinds of molecular
identifications are part and parcel of developing new materials and the
processes for making them.

Since 1981 a new family of even more powerful microscopes, the
so-called scanning probe microscopes, have been routinely resolving
blurry ghosts of individual atoms so that the tiniest of
imperfections-a single atom out of place in one crystal grain or on a
semiconductor chip-can be discerned.  The scanning tunneling microscope
(STM) was the mother of all scanning probe microscopes.  Invented by
Gerd Binnig and Heinrich Rohrer in Ri1schlikon, Switzerland, at the IBM
Zurich Research Laboratory where the first ceramic high-temperature
superconductor was discovered, their STM, like an old-fashioned record
player writ small, permits imaging of the atomic and molecular
landscapes of surfaces with unprecedented ease and clarity.

The principle behind an STM is to bring an extremely fine stylus close
to an electronically conductive surface so that the electron tunneling
can occur.  Even though there is no direct contact between the stylus
and the surface, electrons tunnel between the two.  The quantum current
dramatically strengthens or weakens upon the slightest difference in
the distance of the stylus s tip from the surface.  As the
computer-controlled stylus effectively sweeps back and forth over the
surface like a blind man s walking stick on a sidewalk, the computer
keeps track of the changing tunneling current.  This electrical record
then can be reconstructed into a three-dimensional image of atomic
resolution complete with operator-chosen false coloration.  The
gorgeous computer-generated pictures from these exercises have changed
the look of scientific journals.

Each of these instruments, as well as a growing roster of variations
for many of them, provides analytical access to almost any structural
level of any material before, during, and after the material s
synthesis, processing, or service life.  The analytical arsenal
available to materials researchers today is nothing short of
miraculous.  Yet the power to see details of materials still
accelerates.  There is a small army of researchers with a special brand
of machismo that expresses itself as a drive to develop instruments and
techniques that push this analytical envelope further and further.  For
them life is like a technical name that tune game, only the boast is
more like I can name that substance with one molecule, or I can
determine the distribution of grain sizes and point defects in that
mite-size sample.

Synthesis.  Even if you know the precise location and activities of all
the atoms in a material, it won t get you a new polymer or a new
metallic alloy.  That takes putting things together, not taking them
apart: it means obtaining raw materials or synthetic ingredients and
then processing them into real materials.  Just as a cadre of
scientists have brought analytical instrumentation and capability to
virtually the ultimate level, so too have makers of chemicals and
materials.

Metallurgists have learned how to use specific heat treatments to
control the way the atomic components of molten metal aggregate,
separate, and fall out into a nexus of crystalline metal grains,
ceramic particles, voids, and other features that constitute a
particular alloy with unique characteristics.

Chemists have gotten to the point where they can probably find a way to
synthesize any molecule that is physically possible.  Some of the most
dramatic displays of molecule building prowess have come from organic
chemists seeking to duplicate in the lab the complex chemical
structures that living cells have brought to an enviably high art.
Materials scientists and engineers need to think beyond that.  Not only
do they need to have suitable atoms and molecules, but they also need
to find ways of assembling these into ever larger structural
arrangements until they get the material with the properties that they
seek.  It is akin to the task of knowing how to start with piles of
bricks, lumber, and shingles and end up with a house.

One level of this pursuit involves synthesizing materials of utmost
purity, an ever more important requirement as engineers demand higher
performance from materials.  Optical fibers were high on the wish list
even in the early 1960s, but every time anyone tried to make them, they
had so many impurities that light would scatter, get absorbed, or
otherwise get quashed into darkness before it could travel a city
block.  In the late 1970s, when fiber makers at Corning Glass Works
nearly eradicated all impurities, they had opened the floodgates of
optical communication.  Purity of ingredients is also of paramount
importance for semiconductor technologies, specialty alloys,
solid-state lasers, and many other materials that are the stuff of
advanced technologies.

Polymer makers have been developing several tactics for controlling how
large molecules--each made of smaller molecular components linked into
chains or networks-assemble into specific arrangements.  Samuel Stupp,
a materials researcher at the University of Illinois, for example,
designs linear molecules so that they automatically fall into an
arrangement like the teeth of a comb.  Then he stitches them together
more securely, using a measured dose of light that activates parts of
each molecule to reach over and bond to its neighbor.  What he gets
from this procedure are so-called photonic materials that can
manipulate light-amplify it or change its wavelength-the way electronic
components manipulate electrons.

Donald Tomalia, an entrepreneurial polymer chemist at the Michigan
Molecular Institute, has been championing a new class of polymers
called dendrimers because of their resemblance to the branching of
trees.  He and a small community of others around the world build one
major kind of dendrimer starting with a small molecular core that they
can link to a specific number of building blocks just like itself to
form a slightly larger structure.  Each of the newly linked components,
in turn, can link to the same number of building blocks.  By repeating
this process a number of times, enormous yet identical polymer
molecules branch out and build up, some up to the size of viruses. They
are so uniform that Tomalia sometimes thinks of dendrimers as a class
of enormous artificial atoms, which he hopes can be combined like real
atoms into the dendrimer versions of molecules and materials.

Until recently, chemistry has been the art of crowd control so that the
average behavior of huge numbers of atoms and molecules would result in
the formation of materials needed to build and run society.  Scientists
are moving beyond that kind of crowd control and getting downright
personal with atoms.  They can now work with this or that atom, Joe or
Jane atom, not merely with enormous numbers of atoms and molecules
whose populations routinely could be denoted by numbers containing
twenty-four zeros (that s a million billion billion).

The scanning tunneling microscope was originally designed for analysis,
but its users will tell you that it doesn t take long to discover that
the stylus can do things, real things, to the surfaces underneath.
Scientists can routinely move individual atoms from place to place,
using them to spell out words, university initials, even creating
simple maps of the world in the ultimate expression of pointillist
writing.  They can pile up a few hundred gold atoms here and a similar
number of copper atoms nearby so that a tiny flow of electrons goes
between the piles.  Beholdthe world s smallest battery!  They can use a
laser beam to lay down fine chromium wires a mere sixty atoms wide to
form diminutive wires for next generation electronic devices.  They can
create even smaller quantum dots whose electronic states can represents
bits of data, dots so small that a stamp size crystal riddled with such
dots could hold the entire Library of Congress.

If thousands of scientists around the world weren t actually making
such things, this all might legitimately be dismissed as the breathless
fantasy of a technophile.  But thousands of scientists around the world
are making these things and showing off their atomic constructions in
enormously magnified pictures projected onto huge screens at meetings
or printing them in spectacular computer-hewn colors in their technical
papers.  Moreover, their skills are only becoming more sophisticated.

Phaedon Avouris, a physicist at an IBM flagship facility in Yorktown
Heights, New York, knows this better than just about anyone.  With an
STM you can image a surface on the nanometer or atomic scale, zoom in
on a particular feature or region of interest, induce modifications on
the atomic scale, and then image what you have done, says the
Greek-born scientist, who grew up near Abdera, where Leucippus and
Democritus first preached their atomism twenty-five hundred years
ago.

Researchers have sliced DNA molecules in half with an STM.  They have
created molecules by lining up atoms in the valleys of crystalline
surface like eggs in an egg box.

Donald Eigler, at yet another IBM facility in San Jose, did this by
lining up seven atoms of xenon, his favorite atom, on a crystal s
surface.  He scored an enormous PR coup for IBM in 1990, when his group
wrote out the company s logo in xenon atoms.  Many, many others also
have used STM tips to draw lines, letters, and other symbols into
crystal surfaces the way children draw in the sand.  With an STM,
though, the lines are drawn with atoms.  Chips of the future may well
be made with machines wielding large numbers of STM tips fashioning
diminutive structures for storing data or computing answers to
problems.  (See Figures 15-20.)

Even before scanning probe microscopes were a gleam in Binnig s and
Rohrer s eyes, Leo Esaki and Raphael Tsu, then also at IBM s Yorktown
Heights facility, proposed another strategy for controlling matter on
nearly atomic scales.  If researchers could come up with a way of
growing crystals that would enable them to mix and match elements known
to form into semiconductor materials while controlling the spacings
between the novel elemental combinations, they would be on their way to
remarkable new materials.  The physics that would go on within the new
material structures, they reasoned, ought to make for faster electronic
components or for devices that

5 might interact with light in novel ways.

One fantastic technique for doing this now, molecular beam epitaxy, is
a kind of atomic spray painting onto surfaces, one atomic layer at a
time.  Every month researchers around the world publish their newest
MBE constructions.  In a stainless steel vacuum chamber of a typical
MBE machine rests a crystal substrate on a temperature-controlled
stage.  Crucibles separated by shutters from the main chamber contain
chemical or mineral precursors.  When heated, these precursors emit
atoms: gallium, arsenic, aluminum, indium, phosphide, or a

$ T Il

FIGURE 15

FIGURE 17

Writing with Atoms.  In one of biggest PR coups of recent years, which
also happened to be a technological tour de force, the physicist Donald
Elgler of IBM s Almaden Research Center in San Jose, along with the
visiting scientist Erhard K. Schweizer from the Fritz-Haber Institute
in Berlin, used a scanning tunneling microscope (STM) to move
individual xenon atoms on top of a surface of crystalline nickel to
create the ultimate pointillist rendition of the

FIGURE JL6

FIGURE JLS

40 W

40 40

4W 40

4W

famous company s logo.  The length of the logo is 660 billionths of an
inch.  This was one of the earliest demonstrations that tools like the
STM held the technical promise of moving individual atoms at will.  The
sequence of four Images shows how the researchers used the STIVI to
transform an Initial scattering of xenon atoms Into a legible acronym,
albeit at cryogenic temperatures and under a strong vacuum.

number of others.  By using a computer to control how long the shutters
are opened or closed, scientists can direct and lay down beams of
different atoms on the substrate surface.  Crystal growers can decide
just how many layers of each element or elemental combination they
want.  The underlying crystal enforces a periodic structure on the
incoming atoms like eggs taking their places on a huge egg carton into
which they are being tossed.  (See Figure 21.)

In the 1980s, as MBE became a more viable and user friendly tool,
researchers really began to see its possibilities: for example, it can
synthesize solids whose electrons can take on energy states otherwise
forbidden by the laws of quantum mechanics, absorb and emit new
wavelengths of light, and interact with each other in different ways.
Like painters getting hold of new colors, materials researchers have
been building thousands of crystal structures with MBE machines,
sometimes just to see what they might get.

More often now they have definite ideas of what they want.  Maybe they
seek new semiconductor materials in which electrons move faster so they
can make faster chips to make faster computers.  Maybe they want new
solid-state laser crystals that emit in colors previously unavailable.
(Red and green

On the other page are two more examples of atomic manipulation.  The
top image depicts the KanJI characters (used In the Japanese and
Chinese languages) for the word atom.  These characters are written not
in ink but rather in the medium of Individual Iron atoms precisely
placed on a copper surface with a scanning tunneling microscope.  Known
as a quantum corral, this circular structure Is made of forty-eight
Iron atoms moved into position on a copper surface by a scanning
tunneling microscope.  Twenty-thousand of these rings could be placed
adjacent to one another across the cut edge of a human hair.  This
circular quantum corral, as well as other corral shapes, can confine
electronic waves that undulate on the copper surface like tiny ocean
waves; electron surface waves are about a trillion times smaller than
ocean waves.  Such atom-built structures could open routes to creating
electron circuits that are smaller and more powerful than today s.

IBM CORPORATION, RESEARCH DIVISION, ALMADEN RESEARCH CENTER

FIGURE 19

have been relatively easy, as is evidenced by the little red and green
indicator lights on consumer electronics.  Blue has been a tougher nut
to crack, though in 1996 blue lasers made of such materials as gallium
nitride had become the trophy for a technological horse race).  In this
way scientists and engineers have been able to create crystalline
structures with specific quantum mechanical features not found
elsewhere; in effect, they have found a way to get atoms and electrons
to do things that were possible in theory but hitherto unrealized in
practice.

One of the most visible and widespread successes of this approach so
far is the lasers that read compact disks.  The lasers are made of
layers of such compounds as aluminum gallium arsenide and gallium
arsenide arranged sandwich style, each layer on the order of tens of
angstroms, a mere handful of atomic layers.  The sandwich structure of
these crystals creates a quantum mechanical context in which an input
of electrons from the wall socket will cause electrons moving within
the crystal to meet with positively charged vacancies in the crystal,
or holes.  When electrons and holes meet, they can make light
together.

Despite these extraordinary feats of atomic and electronic control,
scientists still strive to achieve dominion over the coarser micro
structural levels of materials as well.  They still seek to control the
way a material s ingredients-atoms, mole The MBE Medusa. It takes
machines like this molecular beam epitaxy (MBE) machine to build
materials one atomic layer at a time.  Various ultrapure ingredients
are placed Into the ends of the stainless-steel appendages, where they
are heated so that atoms or molecules bake off and then beam off toward
a starting surface such as a crystal of silicon or gallium arsenide
Inside the MBE s vacuum chamber.  By controlling shutters between the
deposition chamber and the ends of the appendages, researchers can lay
down virtually any sequence of atomic layers.  With such tools
researchers can construct new types of microchips, semiconductor lasers
like the ones In CID players, and other solid-state gadgets.

BELL LABS LUCENT TECHNOLOGIES

materials research comes of age cules, powders, fibers, liquids, or
whatever-aggregate during the heating, melting, solidifying, mixing,
curing, and other processes that go into a material s formation.

Even in the netherworld between atoms and molecules on the one hand and
perceivable pieces of material on the other, scientists are making
headway.  Besides MBE, other deposition techniques with equally
formidable names-metal organic chemical vapor deposition, for example
have opened a vast set of possibilities for this layer-by-layer
approach to synthesizing materials.  Indeed, the catalog of synthetic
techniques is as vast as the analytical catalog.

Chemists like George Whitesides at Harvard University design varieties
of chemical building blocks that self-assemble into preconceived
constructions.  Fraser Stoddart at the University of Birmingham, in
England, has devised systems in which ring-shaped molecules slide over
rodlike molecules, which

7 are then plugged with molecular caps.  The ring can then occupy one
of two sites along the rod, depending on whether it

FIGURE 21

is zapped with the light of one wavelength or the other.  Stoddart
hopes that these systems will provide the makings of molecular machines
or logic components of a molecular computer.  In enormous
self-assembling arrays, these might comprise molecular memory storage
devices so capacious that the entire Library of Congress could be
carried around in one s pocket.  The challenge is to find the right
stuff for conveniently and accessibly stuffing ever larger amounts of
data into ever smaller spaces.

Others are finding ways of controlling the way crystals nucleate and
grow akin to the way living organisms grow and fashion the mineral
hydroxyapetite into structures as intricate as skeletons.  They are
captivated by the idea of emulating nature, growing ceramic automobile
engines in a manner that parallels the way a marine nautilus grows its
beautiful spiral casing or a sea urchin its protective body of
cathedral-like spines.

Among those in this field--called bio mimetic materials research-is
Mehmet Sarikaya of the University of Washington, who may have learned
more from abalone shell-making than from his own professors.  To him,
this pre-Paleolithic materials engineering on the part of the living
kingdom is miraculous.  Sarikaya and his colleagues need look no
further than the nearest forest or for that matter, across the lab
bench or dinner table for proof that atom-by-atom (or molecule
by-molecule) construction of the most sophisticated products known in
the universe can proceed on a vast scale.  Each cell of a tree or a
human being is a tiny factory brimming with engineering know-how. It
uses only readily available and inexpensive raw materials in the form
of oxygen, hydrogen, or carbon dioxide from the air and from water;
from food it extracts carbohydrates, proteins, or ions like calcium. 
Billions of years in the making, the cell s workshop of molecular
machinery acts like a petroleum refinery to process that input into a
huge variety of functional materials: stretchy and enduring vascular
tissue, strong and tough tendons, hard materials research comes of age
N 127

and tough antlers and horns, muscle, bony skeletons, and compliant
skins to support and house the entire operation.

The fantastic productivity of biological materials processing comes
from several levels of multiplications.  In each individual tree.  or
person, trillions of cellular factories work in parallel.  What s more,
these living, breathing factories of biological materials replicate
themselves as long as the raw materials are available and the instinct
or will to do so remains.  That amounts to a planet laden with biology
s bountiful inventory of materials.

For bio mimetic materials research nature serves as a perpetual
in-your-face reminder of just how powerful human materials scientists
could become.  The living kingdom is a mass-production factory for a
variety of astounding materials structural protective, mechanical,
electro chemically driven substances that grow, strengthen, replicate,
heal themselves, and adapt to their conditions, all within a limited,
life-friendly range of physical conditions.  Working in technological
venues free of such natural constraints, materials scientists and
engineers can envision processes of humanly devised materials
production that would outstrip even nature s own genius and
fecundity.9

Theory.  Synthesis is the production of materials.  Analysis breaks
them down to reveal nature s secrets or to see what went right or wrong
in the synthesis.  Theory is both child and parent of synthesis and
analysis; just as it is informed by the antecedent practical results of
synthesis and analysis, so its concepts inform subsequent practical
efforts that can steer researchers away from a lifetime of dead ends or
into a lifelong adventure in fantastic material arenas.

In the first half of the twentieth century, the age-old tria land-error
approach in materials research began to transform with the coming of
ever more sophisticated and detailed theories about all manners of
materials.  Ever vaster amounts of scientific observation were subsumed
under conceptual frameworks that shed light on complex material
phenomena such as the age hardening of aluminum alloys or the formation
of diamonds from carbon atoms.  Rather than relying on the deepest
theories based on quantum mechanics-the early calculations of which
were more often off than on-theories based on material defects,
thermodynamics, and other higher-level concepts proved far more
valuable than quantum approaches for explaining material behavior.  And
they often still do because they are better for depicting more humanly
intuitive levels of material structure.

Quantum mechanical theory, which presumably harbors the deepest
insights into materials, has remained limited as a practical tool by
the computational complexity of its equations.  After all, latent in
Schrodinger s all-encompassing equation is the means of calculating any
atom s or molecule s complete chemical personality: its shape,
stability, and how and when it reacts under different conditions. Since
materials are made of atoms and molecules, the same equation, in
principle, harbors portrayals of material aggregations as big as bricks
and sheets of steel.

But how to domesticate this wild mathematical behemoth?  Paul Dirac, a
contemporary of Schrodinger s, put it this way in 1929: The underlying
physical laws necessary for the mathematical theory of a large part of
physics and the whole of ehernistry are thus completely known ... and
the difficulty is only that the application of these laws leads to
equations much too complicated to be soluble.  Dirac said so much
nearly twenty years before the transistor was invented; though headway
has been made, Schrodinger s equation continues to put up formidable
armor against the computational cannons of the microelectronics
revolution and its most powerful computers.

The reason why is that the exact solution describing the distribution
of electrons around even a simple molecule such as ethanol, a globally
popular nine-atom molecule, takes enough number crunching to give even
today s supercomputers pause.  The simplest mathematical approximation
requires twenty-six functions, one for each electron.  That s just the
beginning, however, because each electron and nucleus affects all of
the others in an exponential web of interactions.

The calculation for ethanol molecules could end up involving more than
five hundred thousand mathematical terms, each one containing a
six-dimensional integral describing the physical features of the
electron or nucleus.  Moreover, solving these huge equations is not a
single gargantuan one pass computation like dividing a pair of
million-digit numbers; rather it is a reiterative task in which the
results of each earlier pass feed into the next one and so on for tens
of thousands or more cycles, or time steps.  Since each cycle
represents a billionth or a millionth of a second of activity in the
life of the molecule, it takes that many before a good idea of the
molecule s actual behavior becomes apparent.

Scientists mostly have surrendered to the complexity of the
calculations by coming up with simplifying approximations.  These have
been more or less artful exercises in self-imposed blinding.  The
strategy is to ignore big hunks of the quantum mechanical equations
that ultimately have little to do with the actual chemical behavior of
atoms and molecules.  Why worry much about the six inner electrons of
an oxygen atom-and the hundreds or thousands of additional mathematical
terms needed to describe them-if it is the atom s outer two electrons
that are responsible for virtually all of its chemical behavior?

That strategy of selective editing was a good idea in principle.  Yet
any approximation that stood a chance of being solved on the earlier
generations of computers was likely to yield calculations and
predictions that were either dead wrong or subject to an unacceptably
wide range of uncertainty or error.

In the mid-1980s there were only a couple of dozen examples of
calculations that had predicted new compounds and reactions.

Most calculations simply made sense of what experimentalists already
had known and observed.  By the mid-1990s the number had grown to
hundreds and appeared on an exponential trajectory reaching into fields
as diverse as rocket fuel development, steel alloy design, drug design,
and semiconductor engineering.  We now can become the world s expert on
a particular molecule or material even though we have never seen it,
remarks Marvin Cohen, one of the most outspoken champions and
practitioners of computational materials research at the University of
California at Berkeley.  As early as 1970, he had predicted
successfully that silicon, a semiconductor at room temperature and
pressure, would behave as a superconductor under certain extreme
conditions.

Cohen and his former graduate student Amy Y. Liu at the University of
Berkeley made headlines in the summer of 1989, when they reported
calculations for a hypothetical material made out of carbon and
nitrogen in a ratio of three to four.  They predicted that the material
might even be harder than diamond, the hardest substance yet known in
the universe.

They calculated many possible crystalline arrangements of the carbon
and nitrogen atoms.  They then used other theories relating atomic
structure to hardness to infer that some of the resulting in-silico
structures ought to represent materials that are at least as hard as
diamond, and maybe harder.

Since this prediction, various researchers have intermittently reported
evidence of making carbon-nitrogen materials nearly as hard as diamond.
Some have gone to extremes: squeezing the ingredients at great pressure
between the jaws of anvils made of diamond or using ion guns to propel
nitrogen atoms into a carbon-rich surface in the hope that the sought
after carbon-nitrogen structures would arise spontaneously. Although
researchers have reported signs of materials that seem as hard as
diamond, their samples have been too tiny to make confident
assessments.  And other theorists have been questioning Cohen s and Liu
s original calculations, coming up instead with alternative crystal
arrangements that might yield the elusive harder-than-diamond stuff.

There are other bridges that link the mind-numbing quantum mechanical
equations with older and more intuitively grasped theories about how
matter and materials behave.  Many of the estimated two thousand
computational chemists and materials scientists spend their days
developing computational strategies and algorithms that serve as those
bridges.  Often they will ease up on the aesthetic appeal of pure
prediction by incorporating experimentally derived data that can help
computers direct their computational efforts more quickly toward
sensible.  With that strategy solid-state scientists have been
extending computational chemistry way beyond small discrete molecules
like ethanol to the architectures of periodic materials such as
crystals with atomic constituencies numbering in the millions.

In addition to using empirical data, computationally inclined
scientists adapt Newtonian ideas to molecular realms, linking classical
physics to thermodynamic accounts of how the subtle redistributions of
energy throughout a system of atoms and molecules will lead to this or
that crystal or less-ordered amorphous phase, grains of this or that
size, and other kinds of micro structural details.  These so-called
molecular mechanics techniques can sidestep computer gagging quantum
mechanical calculations by portraying molecules as, say, sets of balls
linked by springs whose tension and bounce represent the balance of
electron-electron repulsions and electron-nucleus attractions.  Or they
might rely more on matching the energies of particular regions of a mix
with phase diagrams that codify the kind of microanatomy that can form
under those conditions.  These types of computations, of course, often
can first be anchored to quantum mechanical calculations to make sure
they can be grounded in first principles.

Just as the theories of molecular mechanics theories can link to those
of quantum mechanics, so they can link to well developed theories of
thermodynamics and kinetics that account for bulk properties of
materials: when they melt, crystallize, or vaporize.  These
higher-level theories, in turn, link to engineering practice, through
which real materials become useful things.

Computational materials science is getting reliable enough that
researchers have even begun to think of in-comp utero activity as a new
category of scientific activity somewhere between theory and
experiment.  Consider a computational chemistry group at the National
Aeronautics and Space Administration s Ames Research Center in Mountain
View, California, where computational studies on hydrogen atoms
recombining into hydrogen molecules were funneling into engine design
work on the on-again off-again National Aerospace Plane.  The NASP
(which has had several aliases) is envisioned as an aerospace craft
that takes off and lands like a jet but can go into and out of orbit as
well.  Designers have used computational science to help determine
optimum lengths for nozzles and combustion chambers and to establish
more precisely the kinds of hostile conditions that materials of these
engine components will have to endure.  These reactions occur so
quickly in such hostile arenas that only the scantiest of clearly
relevant experimental data from traditional measurements are possible
anyway.  In the last few years, companies specializing in polymers,
ceramics, and materials in general have come to believe that theory can
now provide good enough information that they can do fewer experiments,
reiterates William A. Goddard 111, director of Materials and Molecular
Simulations Center at the Beckman Institute of the California Institute
of Technology.  15

As computers grow more powerful, computationalists grow more adept at
exploiting their scientific potential.  There are visionaries who are
toying with novel approaches that seem to border on the fantastic. Take
the creative computer architecture of MIT researchers Tomasso Toffoli
and Norman Margolus: rather than building computers out of an enormous
number of on off switches-gates that let electrons through or block
their flow-Toffoli and Margolus have been building what they call
94cellular automata machines, or CAMs.  The basic unit of a CAM is a
cell whose own state depends on the collective state of its neighbor
cells.  Programmers stipulate some rule by which the states of neighbor
cells determine the state of each center cell.  The computer then
changes the cells states accordingly, and the procedure is repeated
until a steady state or perpetual static or something in between
emerges from the collective.  This kind of computer architecture-with
discrete cells interacting with neighbors-resembles the way atoms or
molecules behave in collectives.  Tomasso, when he wears a materials
research hat, thinks of his computer as computronium, a universal
chemical element.  In his mind, computronium is a computational
chameleon capable of simulating any material thing or process such as
crystallization or alloy formation with far more physical
verisimilitude than simulations run on more conventional machines.

The triumvirate of analysis, synthesis, and theory has set off an
almost dizzying cycle of positive feedback among materials researchers
over the past thirty or forty years; their interaction has inspired
fresh, diverse efforts ranging from research into what makes the sea
urchin s spine so hard to nothing less than solving major social
problems such as energy-efficient forms of production, pollution, and
transportation.  When viewed collectively, the materials research
community has become something like a vast search team, turning over
every stone and leaf, looking up every tree, and scouring every attic
to see what interesting or useful trinkets of matter can help to change
the world.

FIVE

the MATERIALS sERENGETi Even though materials science and engineering
has been a bona fide field of research for only a few decades, it has
spawned a diversity of products, projects, and practitioners rich
enough to populate its own version of the Serengeti National Park in
Tanzania.

One of the very best times to go on safari is just after Thanksgiving
each year.  That is when ten thousand materials scientists and
fellow-travelers converge on Copley Place in downtown Boston for the
annual winter meeting of the Materials Research Society (MRS).  They
come from every state in the union, Tokyo, Kyoto, Seoul, Moscow,
Berlin, Paris, Manchester, Zurich, Johannesburg, Bombay, Paris, and
hundreds of other places.  They come from universities, government
labs, and industrial research facilities.  High-tech venture capital
firms send representatives.  Hundreds of companies eager to do some
business set up booths and equipment in a massive expo complete with
bowls of chocolate and attractive salespeople to lure potential buyers
of mass spectrometers, nanoscopic ceramic powders, and the hottest new
scanning tunneling microscope.  This town-size conclave fills entire
hotels, overruns restaurants, and injects millions of dollars into
Boston s economy.

Mostly, these masses come for a materials research show and-tell
nonpareil.  They come to talk science, reunite with graduate school
pals, gather intelligence about competitors, and to talk with
long-distance collaborators.  They come to buy and sell instruments,
find jobs, and locate new sources of alumina powders for making
refractory bricks or maybe isotopically purified methane for synthetic
diamonds that are more perfect than any sparkling gem.  Everything you
always wanted to know about materials research and a thousand fold more
can be learned during these weeklong pilgrimages.  An MRS meeting is
car naval for the pocket-protector set.

Because advances in materials have had everything to do with general
scientific progress no matter what the field, this fair is a general
celebration of all that science and engineering have wrought: the
history of microscopy is partly a history of glass making the history
of spectroscopy is partly a history of light-sensitive materials for
the detectors; every computer controlled instrument in every laboratory
owes a debt to semiconductor technology, which is as materials centric
as you can get.

When MRS convention participants pay their registration fee, they get a
telephone book of abstracts describing several thousand talks.  At any
given hour during the meeting, twenty five different researchers will
be talking simultaneously to audiences in twenty-five different meeting
rooms.  The technical torrent rages all week.  Specialized thematic
symposia run through the meeting like subplots of a play.  The symposia
become havens at which each participant can find his or her own
familiar ground.

There are polymer specialists and synthetic diamond aficionados.  There
are users of molecular beam epitaxy machines who create new kinds of
transistors and lasers on a chip by building up special crystals one
atomic layer at a time.  There are solid-state physicists placing bets
that crystals with giant magneto resistance (materials whose electrical
resistance changes sharply in the presence or absence of a magnetic
field) are the key to next generation data storage and handling
technologies.  There is a perennial surfeit of researchers studying a
class of ceramic superconductors that can carry electricity
resistance-free at higher temperatures than all previously discovered
superconductors.

There also are theoreticians sharing ways of coaxing more out of
Schrodinger s equation or of Newtonian descriptions of molecular
activity.  They work in odorless laboratories awash in the glow of
computer screens.  Their equations are the wands with which they
conjure up an endless vista of new materials that will lead to a
technological landscape as far from today s as today s is from that of
the early century.

There are researchers at MRS meetings who rhapsodize in minute
molecular detail about designing new energetic materials, solids, and
liquids that are good as propellants or explosives that pack ever more
bang or thrust into smaller amounts (mind you, all of this has to be
possible with minimal danger of accidentally exploding the stuff during
preparation and handling).  You won t hear the most secret stuff about
energetic materials since the doors of an MRS meeting are open to all
registrants regardless of their security classifications.  But there
are other meetings organized by organizations with names like the
National Defense Preparedness Association where participants need
security clearances and the doors are guarded.

Still other research tribes at an MRS meeting obsess about safe ways of
perpetually entombing radioactive waste into glassy materials that can
take the bombardment of radioactive slings and arrows for millennia.
Another group aims to emulate the arachnid ian silken feats of natural
polymer engineering and use them to make what they call bio mimetic
materials.  For this tribe the exoskeleton of a bess beetle could well
provide the key to a new generation of aircraft: lighter-weight, more
resistant to damage, and harder to detect.

Yet amidst the apparent mishmash of tribal differences lies that unity
of thought and action that Northwestern University s Morris Fine had
seen even in his Bell Lab Days in the 1940s.  It is the same conceptual
unity that provided the very impetus for the birth of the original
Interdisciplinary Labs (IDLs) at Northwestern, Cornell, and the
University of Pennsylvania and for the founding of the Materials
Research Society itself.

As Fine pointed out years ago, material research is not just chemistry
or quantum theory or solid state physics or process engineering or
marketing.  It is all of these as well as any other way of thinking
that helps reveal how the structure, properties, performance, and
processing of materials relate to one another, to technology, and
societal needs.  The founders of MRS knew that the IDLs were merely a
start.  The age-old disciplinary segregations would not buckle easily
even with new academic structures such as the IDLs in place.  The
founders of the MRS wanted to create another kind of forum beyond IDLs
at particular universities; they sought a more general infrastructure
for changing attitudes, a task as daunting as any technical
challenge.

Materials researchers have developed a collective insight into the ways
of materials that often even astounds themselves.  Their scientific
conferences have become show-and-tell marathons in which researcher
after researcher, in darkened hotel meeting halls, flashes annotated
series of slides depicting exquisitely detailed data and images about a
material s structure, functions, and place in the technological world.
Many of these slide shows end up winning the attention of only a
handful of like-minded specialists.  Most quietly recede into oblivion.
Some, however, take root and even grow into great branching
subdisciplines.  And some of these trigger research wildfires that
change the world.

the materials serengeti 0 139

Going beyond this impressionistic helicopter view of the Serengeti of
materials science and engineering means making forays on foot amidst
the different tribes, prides, and packs.  You could probably mingle
with a different group for every day of the year, but a sampling will
suffice to show the diversity on the plain.  Welcome to the tribes of
synthetic diamond, bio mimetic materials, and smart materials and
structures.

SYNTHETIC DIAMOND

The dream to make synthetic diamond spans over a century.  Some of the
nineteenth century s most decorated scientists went to great extremes
to coerce carbon-bearing ingredients into the coveted crystalline
structure of diamond.  In 1880, for example, the Scottish chemist James
Ballantine Hannay filled tubes with various carbon-containing
ingredients such as burned sugar and bone ash, sealed the tubes, and
then heated them until they exploded.  He claimed that tiny hard
particles retrieved from the debris were diamonds.  (He also attributed
a nervous disorder he developed to the frequent fear of injury during
these explosions.) Later analysis of his experimental protocol
virtually precluded the possibility of his actually having produced
diamonds.  Robert Hazen, a geophysicist at the Carnegie Geophysical
Laboratory in Washington, D.C."  speculates that Hannay s assistants
may have spiked his preparations with tiny diamonds with the hope of
getting Hannay to give up his mortally dangerous obsession.

The first documented successes occurred in 1953 and 1954.  The earliest
one was quiet and remained unannounced by cautious researchers at
Sweden s main electrical company.  The other was announced with fanfare
in February of 1955 by the General Electric Company before the
Department of Defense silenced the company for several years in the
interest of national security.  The youngest member of the GE diamond
making teams, the freshly graduated Robert Wentorf, Jr."  even found
that he could convert peanut butter, a rather popular carbon-containing
material in its own right, into diamond grit.  At the time, industrial
diamond was considered a strategically important material since it was
used to make tungsten carbide machine tools. Tungsten carbide machine
tools could cut and mill steel into munitions, tanks, frames, and other
military materiel.  Machinists would have loved to use diamond itself,
which is much harder than the carbide tools.  Unfortunately,
carbon--even in the form of diamond-dissolves in iron and several other
metals at elevated temperatures, thus precluding its use for high-speed
machining of iron-based metallic alloys including steels.  Moreover,
the superlative hardness of diamond means it is very difficult to shape
it into something like a drill bit or milling blade.

The early successes at GE depended on a punishing heat and-press
process designed after nature s own presumed way of making diamonds
deep underground in magmatic cauldrons.  Thousand-ton presses,
themselves wonderful feats of en i

91neering and materials development, and temperatures of 2,500*F (and a
catalyst such as a piece of nickel) hot-squeezed graphite into diamond
grit.  Under these processing conditions, the graphite s carbon atoms
dissolved into the catalyst and then crystallized in the form of tiny
diamonds like salt crystallizing from a saltwater solution.

This transformation on the atomic level is as basic as it gets.  Each
carbon atom in graphite is bonded to three other carbon atoms in a flat
arrangement.  In a piece of graphite, this three-way bonding
arrangement goes on for billions and billions of atoms ahead, behind,
to the right, and to the left.

Carbon Jungle Gyms and Chicken Wire.  The differences between diamond
and graphite goes atomically deep.  Each carbon atom in a diamond Is
strongly bound to four neighbor carbon atoms to create an
everywhere-the same crystal.  This leads to a symmetric crystalline
network of famously superlative strength.  Each carbon atom in a piece
of graphite is bound to only three other carbon atoms, leading to flat,
chicken wire-like sheets of carbon that then loosely stack upon one
another.  The resulting soft, dark material Is the stuff of pencil
leads and lubricants.

the materials serengeti

Each resulting sheet is like an endless expanse of molecular chicken
wire.  Above and below each sheet of molecular chicken wire are other
similar sheets.  The overall structure, therefore, resembles an
enormous stack of sheets of molecular chicken wire.  When you drag a
graphite pencil on a piece of paper, little sheets of molecular chicken
wire get left behind.  (See Figure 22.)

Diamond, on the other hand, is made of an everywhere the-same
structure. Every carbon atom reaches out and bonds tightly and strongly
to four neighbors in a tetrahedral geometry.  Every angle and distance
between carbon atoms is the same.  Making synthetic diamond from
graphite entails the formation of many sheets of molecular chicken wire
into a single infinite molecular jungle gym.  Every carbon must bond to
four others, not only three.  Loosely bound sheets of carbon must
rearrange into a single monolithic structure in which there are no
distinct layers.  Diamond s superlative hardness, compressibility, and
heat conductivity come from a combination of the crystal s supreme
symmetry and from the strength of the carbon-carbon bonds inside.

FIGURE 22

GE filed patents for their heat-and-press process for making synthetic
diamond in 1959, just days before De Beers Consolidated Mines had filed
in South Africa.  De Beers had already been a monopolizer of gem
quality and industrial diamond for decades.  Despite Herculean legal
efforts to prevent it, however, GE broke De Beers s lucrative monopoly
of the industrial diamond market.  The diamond grit industry has since
grown. GE still dominates, but there are other major players, including
De Beers which bit the bullet and licensed GE s technology.  In the
1990s the industrial synthetic diamond market has been approaching the
billion-dollar mark.

Diamond grit has a limited range of applications, almost all of them
based on diamond s hardness and abrasive properties.  If you could make
diamond in larger formats like flat sheets or films, however, the
material s other material talents might find outlets.  The
technological wish list is a long one.  And business analysts predict
that the market for diamond films would make the diamond grit business
seem like peanuts.

Since diamond is the hardest substance known, transparent diamond
coatings would perpetually protect underlying surfaces-finally, a truly
scratch-resistant eyeglass lens, for example.  A diamond film on a
computer s fast-spinning hard disc would protect it from
data-destroying crashes by the read write head hovering a mere
thousandth of an inch above the rapidly spinning disc, a harrowing
situation likened by some to a 747 screaming just feet above the
ground.  Unfortunately, diamonds do not normally come in the form of
films.

Because it absorbs almost no light, diamond offers ideal protection for
optical fibers and space based radiation sensors.  But here, less
perfect diamond like materials might be preferable since they deposit
as a smooth layer rather than as a multifaceted, light-scattering
landscape.  For the same reason engineers expect dianiondlike coats to
serve as lifelong lubrication for such things as ball bearings and gear
shafts.  Diamond could spell the end of a lot of the wear and tear that
costs billions of dollars in automobile and machinery repairs.

Diamond carries heat away at rates second to none, almost never reacts
with anything (and then only at high temperatures) and is an excellent
electrical insulator.  Its chemical inertness and non stickiness make
it a prime candidate for surgical blades and implants such as
artificial hips.  Some scientists even envision semi conductive forms
of diamond that might supersede silicon and gallium arsenide as the
material basis for a yet faster generation of electronic chips. Diamond
would be the best semiconductor if you could make it work, says Rustum
Roy of Pennsylvania State University.  Roy is a founding member of the
MRS and a regular on the blue-ribbon panels that filter through
government agencies.  He was the first to convince government research
managers at the Office of Naval Research to fund work on a way of
making synthetic diamond into the flat and extended forms required for
such applications as diamond electronics.

The first hint of Roy s diamond vision came into view over fifty years
ago, just a year after the first semiconductor transistors were
reported and several years before the General Electric researchers had
developed the high-temperature, high-pressure root.

The demiurge for low-pressure synthetic diamonds was

William G. Eversole of the Union Carbide Corporation (which got its
start in 1898 making calcium carbide for lighting).  He began his
diamond quest in his company lab in 1949.  He scored his first
successes in the period between November 26, 1952, and January
7,1953.1

Eversole s method involved heating small diamond seed crystals
contained at low pressures in a sealed vessel heated to at least 600 C.
Meanwhile, he let methane or carbon monoxide molecules decompose inside
of the heated vessel.  Liberated carbon-containing fragments from the
methane molecules settled onto the seed crystals, slowly and subtly
enlarging them.  The process is now known as chemical vapor deposition
(CVD).  The project-killing rub was that the growth rates were so slow
that it would take many months of continuous deposition before a film
as thick as this page would grow.

Eversole was swimming against the theoretical current of the time,
which held that diamond crystals would fail to start growing under
low-pressure conditions where diamond is the unstable form of carbon
and graphite the stable form.  But Eversole knew that organic chemists
often succeed in making unlikely chemicals.  The trick is to isolate
certain so-called metastable chemical structures before they have a
chance to rearrange into more stable molecular arrangements.  Samurai
sword makers were relying on the same principle when they plunged their
hot blades into their water baths.  That way they locked the mix of
carbon and iron atoms in such a way that the micro structure yielded a
steel edge with famous strength and toughness.  Though Eversole
succeeded in using diamond seeds to engender the growth of new layers
of metastable diamond, the sluggish growth rate spelled doom for the
method s commercial prospects.  The project fizzled out.

For years afterward the party line among diamond-makers was that CVD
wasn t commercially viable, a perception that was strengthened by GE s
1954 patent on the high-temperature, highpressure technique.  It made
more sense.  To make their diamond grains, they subjected graphite to
the enormous pressures and searing temperatures typical of 240 or so
miles below the earth s surface, where natural diamonds form.  Under
such extreme coercion, the carbon atoms in a graphitic arrangement
reshuffle into diamond crystal, the stable form of carbon.  That is how
you get graphite to become diamond.  Right?  Right.

Not until the early 1960s did Eversole s work appear in the patent
literature (it never appeared in a peer-reviewed technical journal).
That was just about when John Angus got started in the field.  Angus,
now a veteran chemical engineer at Case Western Reserve University in
Cleveland, was told by his colleagues that he was crazy to think anyone
could make diamond by letting heated mists of carbon-containing
chemicals deposit onto other surfaces within a vacuum chamber.

But Angus and a handful of others knew too much to dismiss CVD.  If
they could get it to work, CVD diamond-making would be a milder, less
apocalyptic alternative to the GE method.  Still, most would-be CVD
diamond growers abandoned the dream with one look at carbon s phase
diagram.  (A phase diagram is a graph that maps out the gaseous,
liquid, and solid forms for phases] a specific material assumes under
various combinations of pressure and temperature.  Phase diagrams were
born in the nineteenth century, when physicists were first learning the
rules of thermodynamics, which govern the way heat, other forms of
energy, and materials interact.) Carbon s phase diagram suggests that
carbon atoms ought to assume graphitic chicken-wire arrangements under
the relatively low pressures of CVD.

The CVD diamond diehards like Angus reminded themselves, however, that
a phase diagram, like any map, is an abstraction.  Under certain
conditions then yet to be discovered, the unstable form of carbon
diamond-might freeze out before it had time to settle into its
preferred and more stable form graphite  If Eversole had found
conditions suitable for painfully slow growth, perhaps someone else
would find faster-growing conditions.  You really only have to know a
few concepts about how crystals are put together and then make some
intuitive guesses, remarked Robert DeVries, a consultant to companies
that make hard materials and a retired GE diamond research veteran.
Most of these advances are made by intuitive materials scientists
playing around in a sandbox.

Carbon-containing gases like methane-often maligned as

44 swamp gas -and ethylene represent CVD diamond-making sand.  A vacuum
chamber maintained at pressures ranging from one atmosphere to less
than a thousandth as much serves as the box.  Inside this box a source
of molecule-busting energy-such as heat from a red-hot tungsten wire or
from microwaves-decomposes the carbon-bearing molecules carried into
the chamber on a steady flow of hydrogen gas.  Carbon fragments lipped
from these molecules voraciously search for new places to bind.  They
find what they are looking for on target surfaces also in the chamber.
That is where the carbon ends up depositing as a diamond coating.
Carbon atom by carbon atom, diamond films form and slowly thicken on
these surfaces.

Much of the early CVD diamond research proceeded by trial and error:
feed in a little more or less gas, add or subtract hydrogen, raise the
substrate temperature a little.  Playing in sandboxes can be a lot of
work, however.  Not until 1968 did Angus successfully repeat Eversole s
experiments.  In fact, he went further, more than quadrupling the
growth rate.  He found that swamping the carbon-containing gases going
into the CVD chamber with hydrogen gas seemed to hinder or perhaps
reverse graphitic deposits that would otherwise cover the surfaces of
the diamond seeds and prevent further growth.  Nothing much came of
this achievement, at least not immediately.

Three years later, however, Angus described the role of atomic hydrogen
(as opposed to the two atom molecular form of hydrogen) at a meeting in
Kiev.  One of the participants was the Russian scientist Boris
Deryagin, of the Institute of Physical Chemistry in Moscow.  Deryagin
later become a pivotal player in the development of the CVD diamond
game.  Angus s work was the springboard that propelled Deryagin and his
colleagues to the forefront of the low-pressure synthetic diamond
race.

The Moscow researchers actually had begun their own

CVD work by 1956, using molecules such as CBr, and C1" which would shed
their bromine and iodine atoms under conditions of 1000*C and a few
billionths of an atmosphere, a tough pressure to maintain.  They saw
far more promise in Eversole s technically simpler route, which
required low but more reasonable pressures.  By the end of 1969, the
Russians, too, had confirmed Eversole s work.  And they had throttled
up the growth rate more than tenfold, a rate that would have begun
turning industrial heads had convincing word of their success gotten
out.

Both the Angus and Deryagin groups knew that hydrogen seemed to
increase growth rates of the diamond films by frustrating graphite
growth more than diamond growth.  But eventually graphite islands would
still develop on the diamond surfaces, frustrating further growth. 
Only by first cleaning these islands away with reactive hydrogen or
oxygen atoms could the researchers continue diamond growth. Repeating
this sequence yielded more diamond growth.  But this stop-and-go method
was still too slow and awkward to be practical.

Throughout the 1970s Deryagin and his Russian colleagues were learning
in earnest the critical importance of atomic hydrogen.  They found that
the presence of atomic hydrogen during the growth of the diamond
crystals thwarted graphite islands from nucleating on the diamond seed
surfaces by temporarily binding to those surfaces and physically
blocking graphitic carbon carbon bonds from forming.  And if graphite
did form, the hydrogen atoms could break it up by swooping in and
snapping up carbon atoms like hawks grabbing mice.  Deryagin s group
was putting its growing basic understanding to use to find conditions
that would optimize diamond growth.

Most materials scientists were sleepwalking around these developments.
Besides Angus and a handful of others, no one picked up on the
implications of the Soviet work, partly because of what the Soviet
diamond-growers had left out of their papers.  Their technical reports
on the subject were conspicuously devoid of quantitative details about
the pressures and temperatures they used.  They didn t reveal enough
for anybody to reproduce the work.  For other CVD researchers, that
lack of detail initially rendered the Russian papers more skimming
material than study material.

On top of that, Deryagin s group had all but killed its credibility. In
the late sixties and early seventies they became leading players in the
infamous poly water episode, a pseudoscience fiasco now often recalled
as a precedent for the more recent cold fusion saga.  For a time
researchers thought they had discovered polymeric forms of water and
that in principle the entire ocean could become one giant polymer. As a
result of its role in the poly water affair, an aura of charlatanry
surrounded Deryagin s group.

Nevertheless, Japanese researchers, particularly adept at gleaning
nuggets of practical value from the world s scientific archives, took
the Russian work seriously.  A team led by Nobuo Setaka at the National
Institute for Research on Inorganic Materials (NIRIM) made its first
diamond films in 1974 by passing a methane-hydrogen mixture by a hot
metal filament in a chamber kept at low pressures.  Not much new there.
But the Japanese crew was discovering specific conditions that seemed
more promising than those attained in any earlier attempts.

After seven more years of systematic efforts to optimize the rate of
diamond growth, they reported their increasingly successful work to a
gathering of Japan s Carbon Society.  And in a head-turning spate of
developments between 1982 and

Better-than-Nature-Diamond.  Synthetic diamond materials like the one
seen behind the microelectronic circuit above are made by taking
carbon-bearing starting materials such as graphite or methane gas and
rearranging the carbon atoms In these Ingredients into pure diamond.
Nothing conducts heat better than diamond, which makes diamond
particularly good for shunting destructive heat away from hard-to-reach
or particularly critical chips.  The synthetic heat sink above conducts
heat more efficiently than natural diamond because researchers made It
using a carbon source enriched in one of the several forms (isotopes)
of carbon atoms.  In natural sources carbon exists as a mix of these
several atomic forms and natural diamonds reflect that mix.

GENERAL ELECTRIC RESEARCH AND DEVELOPMENT CENTER

the ma teri at s s ereng e ti

1984, they showed the world how to make diamond films with growth rates
of at least several microns per hour.  The magnitude of this rate,
thousands of times faster than Ever soWs and hundreds of times faster
that Angus s or Deryagin s, stunned the research community.  No one had
even come close.  The current worldwide interest in new diamond
technology

FIGURE 23

can be directly traced to the NIRIM effort, says Angus.  As if to
remind the world of this remarkable alchemy, one Japanese researcher
even reportedly made CVD diamond film from sake, the Japanese rice
wine.  The carbon source in this case must have been carbon-bearing
molecules such as ethyl alcohol.  It was CVD diamond s answer to
Wentorf s high-pressure, high temperature peanut-butter stunt.  (See
Figure 23.)

Ever since its quiet beginnings in 1949, CVD research has been taking
the Black Art out of diamond-making by deciphering the detailed
chemical and physical mechanisms underlying metastable diamond growth.
In the late eighties and into the nineties, this search has become an
international technology horse race, with Japan leading in
commercializing the non optimized technology and the United States
having the edge in basic diamond deposition science.  That latter edge
in fundamental science could harbor the most important keys to
unleashing diamond s technological and commercial promise.

Still, many of the physical and chemical details underlying the CVD
process remain mysterious.  Just what types of molecular fragments
break from the carbon-containing gas molecules?  How do they nucleate
into tiny diamond seeds and then aggregate into larger diamonds?  These
unanswered questions have not prevented researchers from discovering
reliable conditions the old-fashioned way-trial and error, lubricated
by an admixture of intuition, experience, and theoretical insight.

By regulating the hydrogen-hydrocarbon ratio, the pressure inside the
chamber, the temperature of the target materials, and the means of
ripping up the carbon-bearing precursor molecules, researchers have
grown diamond films at up to one thousand microns per hour-about a
penny-width-onto relatively large and intricate target areas.  Like
crowded dollops of cookie batter baking on a baking sheet, thousands of
tiny flat diamonds nucleate on the surface and then grow outward into a
continuous poly crystalline film.  One elusive goal, which could
advance diamond beyond silicon and gallium arsenide as the ultimate
semiconducting material, is to grow a film consisting of a single flat
diamond crystal; unfortunately, the irregular and hard-to-control
boundaries between crystallites in a poly crystalline sheet run counter
to the uniformity needed in electronic materials such as silicon.

From their pole position in the practical art of diamond film growing,
the Japanese CVD diamond community has been in a full sprint.  This is
no surprise to Rusturn Roy, a vocal champion of CVD diamond.  He
remembers visiting the lab of one of his former Japanese graduate
students in 1984.  When he showed me the diamond films he was making, I
couldn t believe it, Roy recalls.  Roy convinced the U.S. Office of
Naval Research to begin funding basic CVD diamond research lest the
United States fall hopelessly behind in this critical technology race.
That marked the awakening of the federal government to the promise of
diamond films.

Two years later Roy and a colleague organized the country s first
diamond film technology consortium involving researchers at Penn State
and more than thirty corporations betting that diamond films could
become big business.

In the summer of 1990, the National Research Council, an arm of the
prestigious National Academy of Sciences, which advises the federal
government about scientific and technological issues, publicly placed
bets on CVD diamond in a report, Status and Applications of Diamond and
Diamond-Like Materials: An Emerging Technology.  The ultimate economic
impact of this technology may well outstrip that of high-temperature
superconductors, the NRC panel suggested.  Business analysts have made
predictions that a low-pressure CVD diamond industry could haul in
billions in the coming years.

None of this is a surprise to Angus, who was a pioneer and proponent of
CVD almost twenty years before Soviet and Japanese scientists finally
showed the world the practicability of the method.  Angus is too humble
to say I told you so, but he could say it.  What he has said is that
diamonds are going to

152 a S I U F F

be everywhere.... They ll be in pots and pans, on drill bits, and razor
blades, in copying machines and on hard disks.

Given the diversity of the synthetic diamond research community his
prediction seems a good bet.  Like the potters of an earlier era,
researchers around the world have developed variations on the basic CVD
diamond-growing techniques.  Some are better suited for rapid
deposition.  Other techniques are slower but produce films officer
quality.  Still others don't yield true diamond films but rather
diamond like films.  The smoother texture of these films could make
them attractive for applications in which the tiny sharp facets of
diamond films would cause wear or light scattering: coating ball
bearings or telescope mirrors or radomes of smart missiles.  By scaling
up the experimental CVD apparatus in today s labs, scientists could
usher in an era in which products made with diamond and diamond like
film could become as common as plastic products are today.  (See Figure
24.)

One especially bold group at Lawrence Livermore National Laboratory
once even proposed to make eighteen million carats worth of CVD diamond
as the basis for the world s most sensitive particle detector, which
they claimed would serve well inside of the ill-fated superconducting
supercollider.1They argued that nothing could surpass synthetic diamond
in both its ability to withstand the bombardment of high-energy
particles and the sensitivity with which it detects the impacts of the
particles.  The sensitivity is due to the fact that electrons move far
more quickly in diamond than they do in silicon.  The

Diamond Fields.  Carbon-bearing molecules were broken apart Inside a
chemIcal vapor deposition chamber, and the liberated carbon atoms then
rearranged Into the diamond film revealed here by a scanning electron
microscope.  Diamond films like this can be deposited onto many
different surfaces, even ones with Intricate shapes.  The region framed
off on the left appears at higher magnification on the right, an area
that would just about fit atop the cut edge of a human hair.

THE NORTON COMPANY

the materials serengett a 153

proposal never got the thumbs-up, a rejection that became moot anyway
when Congress killed the super collider project in 1993.  That anyone
could seriously make such a proposal, however, is evidence that the
large-scale production of synthetic CVD diamond is more than just a
pipe dream.

Three years earlier, Science magazine-one of the world s most visible
and prestigious publishers of original scientific research-rated
synthetic diamond as the Molecule of the Year.  Diamonds may soon be
everyone s best friend, wrote Science s editor Daniel E. Koshland.
According to enthusiasts, synthetic diamonds have already or will soon
appear on watch crystals, eyeglasses, optical instruments, audio
speakers, fuel injection nozzles, turbine blades, scalpels, and
semiconductor wafers, to name only a few applications.

Some of the first diamond like carbon products indeed made it to the
market by 1990 when Sony began marketing high-end speakers whose
tweeters contained sound-generating diaphragms coated with diamond like
material.  The diamond presumably

20kv .1.00kx IQ_.@Y 021

enables these components to improve sound quality by displacing audible
distortions to frequencies beyond the range of human hearing.  Seiko
sells watches with scratch-proof diamond coated faces.  Crystallume, in
Memo Park, California, has marketed diamond-coated windows for infrared
scanning systems, which are important in analytical instruments and
missile-guidance systems like those used in the Patriot missiles.  IBM
scientists have looked to a CVD process to fashion light filtering
masks made of patterned diamond films, which they hope to use as a
stencil for making smaller electronic circuitry.  In turn, these
speedier components would enable electronic engineers to build faster
and more powerful computers and communications systems.  The Norton
Company, GE, and others have marketed heat sinks for hard-to-reach or
particularly critical electronic components.

Fired by the promise of CVD diamond films, researchers and inventors
are suffering from a new brand of techno fever  I m addicted to
diamond, admits DeVries, the retired GE diamond research veteran. 
There are a lot of us who have this addiction called diamond fever. 
The relative ease of getting into the fray has something to do with it.
You can almost make this at home with a microwave oven says DeVries."
Several groups in Japan, the United States, and elsewhere have even
used oxygen-acetylene torches, a $250 tool used for welding metal, to
rapidly deposit diamond onto metal surfaces.  Each acetylene molecule
contains a pair of carbon atoms, the raw bits of diamond.

The relatively mild and easy-to-maintain conditions of CVD appear to be
bringing diamond film into the club of readily available materials like
glass, polypropylene, and aluminum.  Yet even as synthetic CVD diamond
makes its own bid to become a twenty-first century wonder material,
many researchers are on the trail of materials that might contest
diamond s preeminence.  Some of these materials may already be the
materzals serengeti 0 155

here, germinating behind a veil of secrecy in government or industrial
laboratories.

Then again, the lack of generally accepted claims of
harder-than-diamond materials might mean that no one has pulled that
off yet.  After all, harder-than-diamond research is an especially
tough business because no one thoroughly understands what makes
materials hard.  Hardness is one of the oldest and yet most poorly
understood of all of the physical properties of solids, according to
the 1990 NRC report on the synthetic diamond technology.

A variety of empirical tests provide simple hardness ranks for
materials: what can scratch what or how much force it takes for a rod
to dent a material.  But these tests say nothing about how the
underlying atomic and molecular structures of those materials relate to
hardness.  Hardness does correlate with such properties as the number
of atoms or chemical bonds within a specified volume of the material:
the more atoms or bonds the harder the material.  But there are so many
exceptions to these rules that they fail to explain hardness in
general.  Besides, correlations often turn out to have no universal
significance.

Nonetheless, some theorists, including Marvin Cohen at the University
of California at Berkeley, have examined hypothetical crystal
structures made with carbon and nitrogen that might be harder than
diamond.  A hold-in-your-hand sample of such a material is eagerly
awaited.

Another contemporary trajectory beyond diamond began in 1990, when GE
rented out New York City s august 21 Club to announce that its
researchers had developed a synthetic diamond that exceeds natural
diamond s already world-record ability to dissipate heat by 50 percent.
Moreover, the synthetic gems, which can take up to one week of press
time to make, could withstand ten times as much laser energy as can
natural gems.

Rather than using normal carbon sources, which contain mostly two kinds
of carbon atoms called isotopes-carbon 12

and carbon 13-in a ratio of roughly one-to-ten, the GE team started
with a carbon source enriched with carbon 12.  The isotopic ratio was
more like a hundred to one.  The carbon isotopes have identical
chemical behavior and bond into graphite or diamond lattices in exactly
the same way.  The only difference between them is that carbon 13
contains one more neutron in its nucleus than carbon 12, so it has a
slightly higher mass.  It turns out that a diamond crystal lattice made
from carbon 12-enriched ingredients can conduct heat faster and
withstand more laser energy than expected.

Decades earlier theorists had hinted at the possible benefits of these
so-called isotopically enriched diamonds.  One of them, Russell Seitz
of Harvard University, claims to be the uncredited inspiration for the
GE super diamond  In 1990 he even bolstered his side of a very public
priority argument with GE by waving Tom Clancy s book, The Cardinal of
the Kremlin, at GE and the press.  A technical cornerstone of the book
depends upon an isotopically enriched diamond window for an X-ray
laser.  In the book Clancy actually acknowledges Russ, a brilliant and
quixotic college dropout in Cambridge, Massachusetts.

Nobody had tried making the isotopically enriched diamond before the GE
team because until recently most researchers expected that any
enhancements in properties would show up only at exotically low
temperatures.  But the GE material struts its super diamond stuff at
everyday temperatures.  These enhanced properties of isotopically
purified diamond could be pivotal for the optical components-mirrors
and beam-steering elements-of especially powerful free-electron
lasers.

The price tag on GE s super diamonds will scare away all but the most
exotic and affluent customers-perhaps only military clients in search
of components for new, more powerful lasers.  But it also could prove a
viable choice as a heat-management safeguard for delicate electronic
devices in nearly impossible-to-reach places such as, outer space or on
the bottom of the ocean.  Keeping orbiting or space-bound electronic
chips cool with super diamonds might eliminate the need for bulky
cooling structures, which sometimes constitute half the weight of
spacecraft.  And every pound not shot into space can translate into
thousands of dollars saved.  For ocean-bottom communications lines,
which often require relays and boosters, using super diamond to rapidly
shunt away device degrading heat would minimize the need for costly and
difficult maintenance expeditions.

The synthetic diamond tribe sees many facets in a gleaming future in
which one of history s most prized and precious materials becomes an
ordinary part of the techno scape

Biomimetic Materials

The superlative properties of diamond were bound to attract the envy of
materials makers who wanted to duplicate or surpass nature s own
ingenuity.  All of that from what might be the simplest of all
materials.  After all, diamond is made of one type of atom, carbon,
each one bonded to four others into a completely uniform crystal. Other
inorganic gems, including garnet, ruby, sapphire, and emerald all of
which involve more than one kind of atom to form more complicated
crystals-have served as benchmarks in the materials engineers unabashed
quest to improve upon nature.

So far nature has a very comfortable lead over humans in engineering
ingenuity.  The earth s natural materials-among them spider silk, skin,
wood, cotton, rubber, cockroach shell, tendon, bone, abalone shell, and
ivory-are as miraculous in their capabilities as life itself They are
materials on which great and small civilizations have risen and fallen
long before anyone had an inkling of how life makes these things.

Until recently, nature was the sole supplier, bestowing precious
materials from its factory, the earth and its organisms.  Acknowledging
nature s genius, a cadre of materials researchers has been forging a
newly recognized field, bio mimetic materials research.  Its
practitioners aim to learn techniques from nature and then beat it at
its own game, the way a novice humbly learns at the master s feet with
an eye to superseding him later on.

The bio mimetic credo is to learn as much as possible about the
material structures within organisms and to exploit that natural
technology into a scientific or industrial context.  You might not want
to make an artificial abalone shell, but you might want to mimic the
abalone s method if making super tough ceramic products for armor or
for lining the pistons in a car s engine.

The name bio mimetic may be new, but the practice is not.  Just a few
years after Leo Baekeland s 1909 introduction of the first fully
synthetic plastic, researchers in German chemical firms began turning
out the first synthetic rubber materials, though falling prices of
natural rubber rendered the man-made material economically inferior.
Until World War II the level of research on synthetic rubber followed
the fluctuating prices of natural rubber.  Natural rubber production
always far outstripped the manufacture of synthetic versions, including
neoprene, the rubber material that Wallace Carothers invented at
DuPont.

That began to change dramatically in October 1942, when the U.S.
government launched a synthetic rubber program (described by some as a
forgotten Manhattan Project), one of this century s most impressive and
unacknowledged successes of chemical engineering.

The program s charge was to secure a domestic source of rubber material
for the war effort in lieu of the natural rubber supplies from
Malaysia, which had been cut off by a strictly enforced Axis embargo.
At the time there simply was no known substitute that could
economically duplicate the combination of toughness, strength,
elasticity, and durability of vulcanized natural rubber.  Without
rubber, however, the war s rolling ma ching the materials serengeti
aery with its belts that conveyed goods or translated power from a
motor to a lathe or saw, and the tires of military vehicles would all
come to an ominous halt.

This was a case where the fate of hundreds of millions of people
depended on the speed with which scientists and engineers could learn
how to mimic nature.  Chemists had long known that the basic unit of
natural rubber was a molecule called isoprene, units of which link into
chain like rubber molecules (polyisoprene), which in turn could
associate with one another to constitute rubber.

In a matter of three years from the start of the synthetic rubber
program, scientists and engineers working on the project had developed
a superior synthetic rubber formulationpolystyrene/polyisobutylene-and
had throttled production up from laboratory scales to pilot scales of
several tons per week to nearly 1 million tons per year by 1945.  This
was a massive collaborative effort involving all the major rubber
companies in the United States and enormous sums of money from the U.S.
Government.  Since then the synthetic rubber industry has grown into a
vast enterprise producing well over 9 million metric tons of synthetic
rubber, more than twice the amount of natural rubber.  Synthetic rubber
might be thought of as the most visible achievement of bio mimetic
materials even though it began well before the field even had a name.

It is easy to see the attraction of the field of biomimetics.  Biology,
after all, has accumulated several billion years worth of awe-inspiring
molecules and materials.  A case in point is human skin, which remains
a perfect fit as a person grows from infant hood to adulthood.  During
hard work, glands squirt sweat onto the skin s surface, where it
evaporates, carrying away excess body heat.  Nerve cells in the skin
detect pressure, temperature, and the texture of objects.  They also
send pain signals to the central nervous system, which then activates
muscles to get the skin away from whatever is causing the pain.  Even
when skin gets damaged despite these safety mechanisms, it repairs
itself.  All this from a thin, pliant, sensory covering that looks to
the eye more or less like a homogeneous sheet.  Bone, tendon, eye lens,
horn, teeth, vascular tissue, beetle cuticle, cocoon silk, and many
other biological materials have similarly awe-inspiring
characteristics.

To Dan Urry of the University of Alabama, any material that can undergo
three billion cycles of expansion and contraction without failing is a
material worth emulating.  That is pree-isely what vascular tissue does
during a lifetime.  Mehmet Sarikaya of the University of Washington is
similarly in awe of the abalone s virtuoso ability to start with the
stuff of chalk and antacid tablets (calcium carbonate), an
intrinsically crumbly, weak ceramic compound, and create a beautiful
iridescent shell tough enough to withstand hammer blows.  Moreover,
since an abalone grows its shell, perhaps it can teach engineers how to
grow tough ceramics into complicated shapes like turbine blades.  (See
Figures 25-28.)

Julian Vincent, a British Zoologist at the University of Reading who
brings a unique clarity and comedic spirit to the field, expresses the
spirit of biomimetics this way: Good scientists borrow ideas, but great
scientists steal them.  A spirited pedagogue, Vincent has been known to
expose his slightly potted belly in airports to show how the sound of
his alternately flexed or relaxed belly supplies rough information
about the micro structure of fat and muscle tissue underneath his
skin.

His playful approach is not unique among his bio mimetic colleagues.
There seems to be a rare research spirit that the field brings out in
its practitioners, a blend of reverence, delight, faith, and
anticipation that their research could well lead to generations of
hightech materials with lifelike sophistication.

But all is not sweetness and light.  In the early 1990s the ubiquitous
Rustum Roy waged a rabid one-man war against the the m a teri a ts s
erenge tin 161

bio mimetic research community, charging it with irresponsible
self-promotion designed to wrest a larger slice of the research dollar
pie than it might deserve.  Roy s aspersions led to published polemics
between him and members of the biomimetics community in such journals
as Nature, Science, and Advanced Materials.

But controversy has not dimmed the ardor of the bio mimetic warriors,
who are still battling toward the holy grail of a new generation of
biologically inspired materials that will rival biology s models in
their versatility and practicability.  Still according to Vincent s
fellow zoologist Steven Vogel of Duke University, they should not
necessarily seek to ape nature but rather to learn its design
principles and apply them in new ways.  Vogel points to the lesson of
Otto Lilienthal, author of a nineteenth-century book Bird Flight as the
Basis of Aviation.  In 1886 be set out to achieve flight by emulating
birds without realizing that they overcome the inherent aeronautical
instability of their bodies with fantastic neural control.  Lilienthal
designed very birdlike hang gliders and was killed in one of his
thoroughly unstable craft, Vogel pointed out in his own book

Life s Devices.

There are other caveats for eager buyers of biology s evolutionary
brand of materials engineering.  For one, biology has done its
engineering within small ranges of physical conditions that can
continually support life.  This is equivalent to transforming a
motorcycle into a car without ever losing the benefit of
transportation, Vogel writes.  Moreover, the very logic of evolution
pushes not toward optimization but toward perpetuating evolutionary
traits that are merely adequate for the survival and reproduction of a
population under prevailing environmental conditions.  So working with
available raw materials, the process has led to hard, soft, stretchy,
sensitive, and other kinds of materials that work just fine.  They are
not necessarily the best materials that could have been made from those
ingredients, but they obviously are sufficient.

FIGURE 25

FIGURE 27

6e

Molluscan Materials Engineering.  In the tough inner layer of Its
shell, an abalone lays down calcium carbonate crystals in a brick like
fashion with an organic mortar composed of carbohydrate polymers and
other bio chemicals (Fig.  25).  The resulting material is both strong
and fracture-resistant, since a crack has to take a torturous course
through the layered structure Instead of speeding through in a straight
line (Fig.  26).  This Is an image of the .  - - - - - -4 - -11 drnwth
nf an abalone (Fig.  27).  The cone like pillars the materials
serengeti v 163

FIGURE 26

FIGURE 28

grow and merge together to yield a full shell.  A synthetic analogue of
the brick-and-mortar shell structure consists of multiple layers made
of a boron carbide-polypropylene mixture alternating with thinner
layers of polypropylene (Fig.  28).  Compared to a monolithic piece of
boron-carbide, the mutillayer architecture yields a material that is
considerably more fracture resistant

MEHMET SARIKAYA

The scientist s freedom from nature s constraints is what inspires bio
mimetic researchers the most.  After all, human engineers can opt to
use cryogenic or kiln temperatures, extremely low or high pressures,
and highly acidic or basic ingredients.  It follows therefore that they
can try to develop, say, faster means of growing biologically inspired
ceramics instead of relying on the slow growth common in life s own
methods for manufacturing bone and shell.

This potential has not been lost on military research managers Nature
has these wonderful solutions and exquisite structures that go far
beyond anything we have now, Michael Marron, then a program manager in
the Office of Naval Research, told Science magazine in 1991. Biological
materials could well serve up the concepts leading to more durable,
survivable materials and structures for naval use- for example, new
underwater adhesives (perhaps based on the byssus that mussels use to
attach themselves to rock), anti corrosion coatings, or lightweight yet
steel-strong composites that make the shells and exoskeletons of
certain beetles so strong.

Nor has the potential been lost on the civilian business sector.
Chemical giants like ICI, Hoechst Celanese, starts-ups like Protein
Polymer Design in San Diego and Adheron are all banking on biology to
point ways to new markets and profits.  Their targets range from bio
ceramic implant materials for grievous bone injuries, composites for
aircraft, and new fibers inspired by spider silk.

The richness of biological materials comes from their hierarchical
design, a point that Eric Baer (a colleague of CVD diamond expert John
Angus at Case Western University) has been making for years.  Living
things exercise exquisite control over chemical composition and
structure at every structural level, from atomic and molecular
components to intermediate structures such as fibers and crystal grains
to visible tissues and components such as tendons, eye lenses,
arteries, skin, and bone.

Baer finds collagen, a protein, to be an instructive example since
biology puts it to multiple uses in connective tissue, tendon, tooth,
and bone.  Collagen is a chain like protein molecule made up of linked
molecular building blocks called amino acids, which in turn are made of
about twenty atoms of hydrogen, carbon, oxygen, nitrogen, and sometimes
sulfur.  Each protein molecule serves as a molecular strand.  These
strands twist into stronger bunches of many strands.  These bunches
twist into even stronger fibers that together twist into strong and
tough collagen cables.  In bunches these cables form anatomical parts
like tendon.  In bone the collagen cables strengthen the hydroxyapatite
mineral that makes up the bone s bulk.  Combined with biological
minerals such as the hydroxyapatite in tooth dentin, they toughen
teeth.  That is why Paul Calvert of the University of Arizona has been
using rat teeth as a model for developing synthetic composites using
synthetic polymers instead of collagen and titania or other tough
ceramics for the bone mineral.  (See Figure 29.)

Calvert admires nuts, too.  This is what a macadamia nut means to
Calvert: The shell of a macadamia nut is an isotropic wood structure,
comprising bundles of cellulose fibers arranged to be locally parallel
but random over longer distances.  Macadamia shells are notoriously
tough, and this can be attributed to the random, fiber-reinforced
structure.  Presumably materials scientists would be able to increase
the toughness of fiber-reinforced polymer composite materials by
duplicating the macadamia nut s various levels of micro structure in
the bulk of the synthetic counterpart.  No one has figured out how to
do that yet, Calvert has noted.

Mehmet Sarikaya found his teachers off the coast of Baja, California:
red abalone shells, which are nearly as fracture resistant as synthetic
ceramics such as zirconia, carbon boride, silicon carbide, and silicon
nitride.  And yet the shell is made almost entirely of calcium
carbonate, the same stuff as chalk and most antacid pills, which
crumble at the flick of a fingernail.

166 a S T II F F

The enhanced properties of the shell derive from the way the mollusk s
cells build calcium carbonate into crystalline micro bricks with
chemical components derived from the seawater around them and cobbled
into a microscopic brick-and mortar structure.  The shell consists of
two main layers in which calcium carbonate is arranged in two different
kinds of crystal structures.  The outer layer consists of calcite; the
inner, tougher, iridescent nacreous layer consists of aragonite.  The
two layers differ only in the relative spacing and angles of the
crystal s calcium and carbonate ions.  In the inner layer the calcium
carbonate forms layers of micro bricks within which each single crystal
is between a quarter and half micron high, or about one one-hundred
thousandth the thickness of a building brick.  Between each of these
layers-and between bricks on the same layer-is a tenfold thinner layer
of organic mortar com posed of carbohydrates and protein molecules
whose precise composition remains unknown.

The structure means that cracks get stopped in their tracks as they
attempt to take cumbersome, tortuous routes around the single crystal
bricks.  Under high magnification, the researchers could see stringy
ligaments, derived from the organic mortar, bridging adjacent bricks
that had separated in cracks.  The ligaments, Sarikaya says, resist and
diffuse crack energy.  In 1994 Greg Olson, a materials researcher at
Northwestern University, adopted these observations as a lead for
making steels that he and his students hope will heal themselves when
too much strain causes their microanatomy to form cracks.

The Hierarchy of Living Stuff.  This classic diagram of the biomimetics
field Illustrates how tough, elastic tendons emerge from a hierarchy of
structural levels beginning with tropo-collagen molecules that twin
into micro fibrils which pack Into subftiblis, which in turn pack into
fibrils, which form into fascicles, which combine into a tendon that
you can see.  Similar atom-or-molecule to-total-thing sequences
underlie nearly all materials.  JOHN KASTELIC AND ERIC BAER

the materials serengeti

In 1989 Ilhan Aksay, Sarikaya s sometime coworker and colleague, had
taken this remarkable structure as a model for making hybrid materials
known as cermets, which are made of one ceramic component and one metal
component.  But instead of mixing the components without much concern
about the micro structure the researchers thought of their boron
carbide as the calcite bricks in the abalone shell and the aluminum as
the organic mortar.  Compared to B 4C/Al cermets, in which there is no
attempt to mimic biological micro structures theirs was 30 percent more
resistant to fracture, and they could see from microscopic examination
of deliberately created cracks that abalone like toughening mechanisms
appeared to be at work.  This enhancement occurred even though the
layers of the cermet material were tens of times thicker than the
calcite layers in the abalone shell.  This told the researchers that
there was plenty more improvement in store for the clever materials
researchers who could more closely

14 replicate the shell s structure.

The lesson is to create laminated structures with highly fibro blasts
Wo r r crimp structure reti plar membrane fascleular membrane

19-A 35A.  100-200 A

500-50004 50-300ju 100- 500)j

SIZE SCALE

ordered micro architectures  The principle is to use one material that
is hard and one that is softer yet capable of forming strong adhesive
interfaces.  Using that molluscan design principle with materials that
are intrinsically much harder than calcite, such as boron nitride or
titanium carbide, remains a possible route to the toughest and
strongest ceramic materials ever made.

Even if researchers flesh out such possibilities in a laboratory, a
small-scale success may not translate to the scaled-up production
required to make a new generation of armor, manufacturing tools, or
engine parts.  This kind of commercial anxiety attends the discovery of
nearly all advanced materials.  They may look good on paper or even in
the lab, but that is not enough for companies to risk the vast sums of
money, credibility, and marketing energy that it takes to convince
potential customers to switch over to a product inspired by a mollusk
shell.  A material can have world-record performance, but if you can t
make it cheaply enough, then it will remain a laboratory curiosity.

Others in the biomimetics tribe have been exploring bio mineralization
Chemist Stephen Mann of the University of Bath is one of them.  He has
focused, for example, on the way the calcite crystals of sea urchin
spines, which look like spires of a Gaudi church, nucleate on cell
membranes.  Cell membranes are made of oily hydrocarbon molecules, not
exactly the chemical venue where you would expect rigid crystals to
grow with any kind of finesse.  What Mann has found is that electrical
charges on the cell membranes serve as nucleation sites so that a bunch
of tiny individual crystals grow perfectly into one another to form a
single complicated crystal.  He now is pushing ahead with his research,
focusing on various systems of synthetic membranes and inorganic
components.

A group of researchers in Israel has been doing the converse-learning
how to use coatings of pre designed hydrocarbon molecules on small
crystals to influence the way in which the crystals grow.  Normally
scientists are unable to control the way in which molecules aggregate
into specific crystal shapes such as flakes or cubes or needles.  Yet
the ability to control relative growth of the different faces of a
crystal could have a lot to do with controlling ice formation on
airplane wings, reducing damage to tissues preserved by freezing them,
or even growing high-tech crystals for detectors or electronics.  What
it takes is knowing the precise physical, chemical, and electronic
contours of the crystal s various faces and the ability to synthesize
chemical compounds that nestle into specific contours.

One of the more daring arenas of the bio mimetic community centers on
the softer world of protein-based polymers, the stuff of blood vessels,
hair, silk, and muscle.  One of the special attractions here is that
the ingredients of these materials are encoded in genes, which
molecular biologists have spent the last forty years learning how to
manipulate.  Not only might biological polymers serve as design models
for new high-tech polymers, but the machinery of life itself might also
be commandeered to carry out the redesigning and manufacturing.

As for models of new polymer materials, Dan Urry of the University of
Alabama has been pushing the envelope as much as anyone.  His successes
were notable enough to earn him R&D magazine s Scientist of the Year
honors in 1988.1 Urry, perhaps echoing some of Roy s caution, thinks
that the living kingdom takes too long to construct the hard parts of
its anatomy.  It takes a sequoia many hundreds of years to create the
several thousand tons of its bulk.  That does not synchronize well with
the needs of modern materials industries, which might have to deal in
tons of materials per day.

He remembers the time in 1978 when he discovered the repetitive
structural motif-a sequence of five amino acids found in elastin, an
elastic protein-based material-that seems to at least partially account
for the wondrous qualities of blood vessels, lungs, and other tissues
that repeatedly stretch and relax.

He has since found ways of chemically modifying these molecular
constructions, which he then can cast into sheets or draw into fibers.
One of his sponsors, the Navy, has been interested in his work since it
could lead to resorb able surgical implants that might prevent the
painful and often life-threatening adhesions that occur on and near
surgical wounds after operations.  In this application it is the
protein polymer s compliance and resilience that are important, and so
it serves its function passively but efficiently.

These materials continue to take Urry s breath away because of how they
act.  He has learned to modify them so that they respond to various
environmental cues such as changes in acidity, temperature, voltage,
salt concentration, and pressure.  When exposed to a flash of light, a
foot-long string of one version of this polymer holding up a weight
will contract and raise that weight by several inches in what appears
to be an almost lifelike response.  Urry can show off similar effects
with different versions of the polymer immersed in fluids or in spaces
in which he can control the temperature: here, potentially, are polymer
materials working as synthetic muscles, either for bionics (human
applications) or robotics.  Here, at the crossroads of the laboratory
and the functioning of the human body, we find contemporary biomimetie
materials research fulfilling the highest hope sand perhaps, some of
the deepest anxieties-of a culture rapidly narrowing the gap between
the organic and the synthetic, between humanity as the product of
natural design and nature as the product of human design.

SMART MATERIALS

Urry s active polymers are a many-talented lot, chafing at any hint of
biornimetic typecasting; these versatile materials are equally adept as
performers in another sexy and contemporary genre known as smart
materials and structures ; metals, ceramics, and other components,
including microprocessors, are also members of this astute and
polymathic league of inanimate structures.

the materiats s erengeti a 171

The field began quietly in 1959, about the same time that Advanced
Research Projects Agency (ARPA) was drawing up legal papers to create
the first three Interdisciplinary Laboratories in an attempt to
institutionalize the modern form of materials science and engineering.
That was when William H. Armistead, an executive at Corning Glass
Works, challenged S. Donald Stookey, the company s most innovative
researcher, to create something truly remarkable.  Armistead, who later
wrote a brief foreword for Stookey s quirky and fascinating
autobiography, wanted a new kind of glass that would automatically
darken in the light and lighten again in the dark.  Somehow light would
have to trigger light-sensitive molecular changes.

Stookey s solution came partly from the light-triggered chemistry of
photography that had made Eastman Kodak so famous.  A decade earlier,
Stookey and some Corning colleagues had experimented with additives
that caused a change in the internal structure of glass in response to
light exposure.  Stookey recalls in his autobiography that the
government delayed the patent for one of these for a few years because
of its potential use in carrying secret messages.  Exposure to sun
light or ultraviolet light through a stencil or a photographic negative
imprints an invisible latent image which can at any later time be
developed to a visible image when the glass is heated to 600 C, Stookey
explains.  Two spooks from the Office of Strategic Services showed up
to collect some samples, though Stookey never learned what they had
done with their booty.

For the photochromic glass that Armistead challenged him to make,
Stookey added silver halide spiked with copper halide to the glass
formula.  These salty inclusions dispersed through the glass like tiny
crystalline islands in a less ordered molecular sea, like grains of
salt in pudding.  The crystallites were so small and transparent that
they hardly absorbed or scattered light.

The material brilliance of this glass formula is a tango of light and
chemistry.  Energy from light knocks an electron from the copper ions
present in the tiny salt islands, and nearby silver ions pick up the
liberated electrons to become neutral silver atoms.  These neutralized
silver atoms then form millions of tiny light-blocking specks
throughout the glass, just as they do in photography to form an image.
When the light dims, the silver atoms in the clumps return their
borrowed electrons to the copper ions, and the clumped silver atoms
disband again, which means the glass clears again.

By the 1970s photochromic glasses had made their way into commercial
items, mostly eyeglasses with a self-adjusting, variable transparency.
According to Donald S. IYotter, a member of the contemporary generation
of Corning s glass and ceramics researchers, if researchers can tame
their temperature sensitivity, which makes these smart materials darken
more intensely on colder days, they could find larger-scale
applications in such places as buildings and automobiles.

At the end of the 1970s, even as photochromic glass was flanking the
noses of thousands of eyeglass wearers, researchers at NASA began
talking up ideas that eventually developed directly into the field of
smart materials.

Richard Claus, a fiber optics expert at the Virginia Polytechnic
Institute and State University in Blacksburg, recalls the time this
way: The basic concept of smart materials evolved back in the late
1970s.  I was working at NASA at the time, and we were trying to
develop nerves and nervous systems for materials.  Spacecraft were
expensive conglomerations of the highest technology, sometimes taking
human beings across the atmospheric threshold into space.  The
researchers wanted better ways of testing materials and components of
spacecraft without always having to take a good portion of the
spacecraft apart, which itself could have ill effects on the materials
involved and add risk to various missions.  The idea we had was to
develop nondestructive sensing applications [essentially, means of
looking inside structures without touching them] to tell when the
structure would not be able to perform the way you would like, Claus
said.

Like the nerves in human skin that send easily interpreted signals of
pain in response to a wound, the sensors that Claus and his fellow NASA
researchers were talking about would detect cracks, de laminations
unusual porosity, excessive strains, and other flaws that could easily
lead to frontpage catastrophe.  One possible way of creating these
artificial nervous systems was to embed optical fibers into the
spacecraft s skin material.  Changes or blockages in the light passing
through the fiber would be the smoking gun indicating that something in
the skin was no longer right.

No one had yet been using terms like smart in reference to spacecraft
rigged with sensors.  Claus dates the beginnings of that kind of talk
to the early 1980s, even tentatively tracing the origin to a
closed-door meeting of Air Force planners known as Forecast 11, the
sequel to Forecast 1, which was convened soon after the emergence of
jet technology during World War II.  World War II had rendered air
power of strategic military importance, and from then on, any
technological edge in aerospace technology could change the balance of
power.  That was the idea behind Forecast I and Forecast II.  The
charge of these meetings was to outline where the Air Force was going
in the next twenty years in terms of materials, structures, and
technologies.  They came up with the top ten technology topics, Claus
said, and one of them was smart materials and structures.

The idea here was to have a plane with a smart skin that could sense
the aeronautical conditions-things like temperatures, pressure, wind
speed-as well as internal strains and maybe even the onset of damage.
Maybe, like living tissue, the smart skin could even automatically
compensate for these changes or even for a direct antiaircraft hit
through a wing.  That could give a pilot that extra second or the
freedom to concentrate on more important things, the sort of advantages
that can mean the difference between life and death for the pilot and
success or failure for a mission.

174 0 S 7 V F F

Initial work focused modestly on smart skins in which optical fibers
would serve as damage sensors.  Gradually people warmed up to the idea,
Claus said.  The main developments revolved around using optical fibers
as sensors embedded in composite materials akin to fiberglass but far
stronger.

As more researchers with different backgrounds began sniffing around in
the field in the 1980s, the concept of smartening up materials and
structures began expanding from its roots in smart skins for aircraft
and spacecraft.  Besides glass fibers as sensors, any material capable
of sensing some thing became part of the mind-set.  The idea was to
develop a multi sensory capability modeled after the human body s
ability to monitor what is happening around it, on it, and within it.
This is how a human body can react to a lion coming at it, dangerously
high carbon dioxide levels in blood, and the scrape wound on a knee.
Any material with some measurable property that changes in response to
stimuli in a mathematically predictable way became part of the
endeavor.  Prototype hardware was validating the concepts while
revealing technical problems, such as the way optical fibers can weaken
composite structures, making them more susceptible to the damage they
are there to detect.

Technophiles have a knack for quickly becoming jaded.  So what if you
can rig a structural beam or aircraft skin with sensors-what else can
you do?  Enter Craig Rogers, who was at VPI, where the country s first
research center devoted to smart materials and structures opened for
business in 1987.  And that s

Smart Stuff.  By building the structural members of vehicles,
airplanes, and bridges with materials that can sense environmental
conditions as well as materials that can respond to those conditions,
engineers have designed materials systems such as concrete and airplane
skins that can behave Intelligently by alerting humans when dangerous
situations arise or by changing their properties In accordance with
changing circumstances.  CRAIG ROGERS

the materials serengeti about when the smart materials and structures
concept began taking off.  (See Figure 30.)

Pivotal to this expansion were piezoelectric ceramic and plastic
materials, which respond to mechanical deformations from pressure and
vibration by creating a change in the voltage across their bulk.
Piezoelectric materials work in reverse, too, a feature that sonar
developers exploited during World War II to develop submarine detection
systems of unprecedented aural sensitivity.  Just as they respond to
mechanical deformation by creating an electrical signal, an electrical
signal fed into them makes them deform or vibrate.  And that means that
they not only can sense things, but they can also do things.  Like
muscles, they can make things happen.

These people stopped looking at walls, girders, beams, airplane skins,
hulls, windows, and other structures as essentially passive things
whose only function was not to fall apart.  Let s put some sensors and
actuators into them.  Let s put a control loop around it, and let s do
things like damp out vibrations in structures, let s eliminate wing
flutter, let s make the material

FIGURE 30

respond to the environment -these became the catch phrases of smart
materials researchers.

By the 1980s the diverse lines of smart-materials research had begun to
converge enough to warrant the honorific discipline.  In 1988 a few
hundred of these scientists and engineers convened the first meeting
devoted to this area.  It kindled a community sensibility that has been
growing ever since.  After that there was a meeting devoted to the
field every half year or so, and other engineering and optical
conferences began to have special sections.  In January 1990 the
premier issue of the field s first journal-Journal of Intelligent
Materials, Systems and Structures-rolled off of the presses.  Craig
Rogers, its editor, became one of the field s most visible champions,
even appearing in a 1991 PBS special called Miracles by Design.  Within
a year Smart Materials, another new journal devoted to the field, this
one edited by Claus, began publishing.

Rogers, who now is dean of the college of engineering at the University
of South Carolina, takes the biological analogy very seriously.  At
more than one scientific conference, he has flashed a slide showing his
young daughter next to a smart cylinder that actively resists buckling:
two examples of intelligent material systems.  Biology is just part of
the formula.  The field s practitioners have developed an eclectic and
audacious way of thinking that draws upon any useful science or
engineering practice.

Like their bio mimetic brethren, smart materials researchers look at
biology for metaphors.  For the prototypical intelligent material
system Rogers envisions structures that have sensory systems that feed
input about the internal state of the system and its external
conditions into some kind of computational component.  This
component-probably a microprocessor or network of microprocessors-would
interpret these signals to plan out and execute a response by
controlling an assortment of muscle like actuators.  Neural network
processors might actually enable the structures to learn proper
responses and even adapt to changing situations.

the materiats serengeti 0 177

With that basic schema smart materials researchers have embarked on a
variety of projects: rope that changes color according to how much
strain its polymeric fibers are enduring; adaptive camouflage materials
that sense the surrounding landscape and then alter the camouflage s
pattern of colors and shades like the skin of a chameleon; and smart
beams, hulls, rivets, and engine mounts, to name only a few.  Smart
beams, as Rogers sees them, are structural members that can change
their stiffness depending on the load or the kinds of vibration they
are subjected to.  And that is a route to making things last longer.
Aeronautical engineers aim to make wings that change their shape, their
aeronautical contours, in accordance with the prevailing conditions;
they thereby hope to steer the craft without the need of flaps.
According to Julian Vincent of the University of Reading, whose work
straddles bio mimetic and smart materials, We are trying to build
intelligence into the constructed world.

When you ask Craig Rogers what good he thinks intelligent material
systems might be, he squints his eyes as he scans over a mental
scenario that he thinks about every day and believes will play out in
the objective world beyond his own hopes.  The concept of instilling
lifelike functions in inanimate objects and artifacts such as advanced
composite materials and home appliances seems to be a vision more akin
to science fiction than present-day realism, he acknowledges.  But he
stresses that an entire community of researchers has collectively come
to this view, that it is not the result of personal vision resulting
from some manifestation of divine intervention.  Yet he still won t shy
away from using prophetic terminology.  Without remorse he has
committed the following to paper: Intelligent material systems will be
manifestations of the next materials and engineering revolution-the
dawn of a new materials age.  31

He expects this revolution to engender major changes in the philosophy
of engineering design.  By replacing material brawn with intelligence,
engineers may no longer have to build structures with tons of extra
concrete, steel, and other kinds of weighty materials to assure
structural integrity.  An intelligent crosswalk or dam or wall could
parry unusually strong strains or vibrations, say during an earthquake,
by sensing and responding to the challenges the way a standing train
com muter bends and twists his knees and ankles to ride out a bumpy
trip.

In time engineers will be able, in effect, to take a structure s
temperature and pulse.  We will soon have the opportunity to ask
structures during their life how they are feeling or where they are
hurt, Rogers says.  If you embedded fibers in materials from the time
they were manufactured until the time they were retired, you would be
able to watch the material be fabricated, you would be able to look at
it during its service lifetime, always monitoring for things like
strain and temperature

1116 and damage.  This is the kind of capability that engineers and
executives of aircraft manufacturers would love to put into their
products.

Though the idea of smart materials and structures may have had its
origins in Air Force discussions, other federal agencies have set their
sights on smartening up their hardware.  The Navy has long imagined
submarines with hulls that could sense the sonar surveillance of enemy
observers and quiver enough to either absorb the sonar energy without
sending back an echo or send back a deceitful signal that would
effectively cloak its whereabouts.  This would be yet one step up from
the advanced silence of Tom Clancy s Red October.

Besides aerospace and submarine settings, the smart materials and
structures idea is infiltrating other more mundane areas.  One company
has made smart rivets whose straight shafts bend back over the part to
be secured by the application of heat rather than force.  Another
markets orthodontic wire made of so-called shape-memory metal that
never has to be adjusted once it is programmed to coerce into proper
alignment the teeth it is attached to.  A high end automotive use for
these alloys is in switches that flip headlight assemblies in and out
of the hood of sports cars.

The magical features of these alloys is that they can be programmed to
assume a particular shape at a particular temperature even after having
been deformed into a very different shape at another temperature.  It
is possible to make a coil of properly programmed metal that would
unfurl to spell out your name when heated with a hair dryer.  The
underlying principle resides in the arrangement of the metal s crystal
grains and the kinds of crystallographic shifts that occur within these
at different temperatures.

OTHER SIGHTINGS

Synthetic diamond, bio mimetic materials, and smart materials are just
a few of the species emerging from the wilds of materials science labs.
The dedicated observer, however, will catch sight of even more recent
and exotic creatures of the scientific imagination.

Fullerenes, for example, are relative newcomers, cousins in carbon of
the more established synthetic diamond.  Their most famous
representative is a ball of sixty carbon atoms that form into a
soccer-ball-shaped molecules known as buckminsterfullerene (C",) or
bucky balls  Its carbon atoms literally bond into the hexagonal and
pentagonal facets of a soccer ball.  As far back as the 1960s,
theorists had postulated their existence, or perhaps wished their
existence because of their extreme display of molecular beauty.  In the
1980s experimentalists with sophisticated analytic instruments began
capturing wispy, blurry snapshots of their actual presence.  Then, in
the fall of 1990, a team of German and American researchers found a way
to make noticeable amounts of them by sending an electric spark between
two graphite electrodes.

Since then fallerenes have been all the rage.  Researchers have made a
wide variety of fullerene balls and cages, including C709 whose
structure resembles a rugby ball, and other much larger ones involving
hundreds, even thousands, of carbon atoms.  There are nested fullerene
structures with carbon balls inside of carbon balls creating structures
resembling onions.  A subpopulation of fullerene researchers focus on
all-carbon bucky tubes which also often show up either in single shell
or nested forms, this time as tubes within tubes within tubes.
Experimentalists have verified that some of these bucky tube structures
can be electrically conductive, which means they could become molecular
wires or even switches for electronic devices even more miniature than
today s.

Richard E. Smalley of Rice University is one of the field s pioneers
who won a Nobel Prize in chemistry in 1996 along with fallere the
pioneers Robert F. Curl, also of Rice, and Sir Harold W. Kroto, of the
University of Sussex in England.  In 1985, he saw some of the first
glimpses of C 60 while vaporizing carbon inside vacuum chambers in the
hope of learning how smaller clusters of atoms-a state of matter
between atoms and big hunks of stuff-behave.  Consider a flat fullerene
sheet, also called a graphene sheet, which is what curves into balls or
rolls up into tubes.  A graphene sheet is really a membrane, a fabric,
one atom thick, made of the strongest materials we expect will ever be
made out of anything, which is also impenetrable, Smalley told a
gathering of chemical engineers in Houston in early 1996, and now we
realize it can be wrapped continuously into nearly any shape we can
imagine in three dimensions.  This has got to be good for something!

Smalley and others envision the bucky balls and bucky tubes and other
shapes hewn from graphene sheets as tinker toys for building nanoscopic
constructions.  If you hang metallic catalysts on these or enzymes, you
might end up with a material for carrying out industrially or medically
important reactions.

This brave new world of curved and tubular carbon could,

like the previously known forms of carbon, spawn a diversity of useful
new materials.  After all, diamond doubles as a mineral symbol of love
(a public relations creation by De Beers that has earned the company
billions of dollars per year) and as a pivotal material in mass
manufacturing.  And graphite is the stuff of pencil leads, dry
lubrication, and reinforcement for composites.

Like biologists discovering a brand-new animal species, fullerene
researchers have been prodding, modifying, and poking fullerenes and
related molecules just to see what the molecules do.  They have
crystallized them along with ions of rubidium to create forms that are
superconductive.  Israeli researchers tried to hook a fluorine atom to
every one of the sixty carbon atoms with the so far unrealized hope of
coming up with a super teflon.  The researchers got close to achieving
the chemical goal, but they didn t end up with anything that seemed
like teflon.  Researchers at Northwestern found a way of using
fullerenes as a kind of carbon adhesive that helps diamond films stick
better to other materials.  Another group of Northwestern researchers
found a way of casting the fullerenes into films that could be quickly
switched into a light absorptive state when hit with a laser beam, just
the sort of thing needed for the nightmarish military scenario of
battlefields thick with potentially blinding laser beams.

Still another tribe of researchers worships at the shrine of newer and
better semiconductor devices; their religion of smaller and faster
will, they hope, usher in an electronic New Jerusalem.  Silicon has
been and remains the stuff of the electronic infrastructure that
innervates modern society so thoroughly now that its sudden
disappearance would be the equivalent of a social nervous breakdown. 
Yet silicon, when properly tainted with charge-friendly impurities such
as boron and phosphorous atoms, is only one member of the class of
materials known as semiconductors.

You also have the three-five and the two-six semiconductors, which are
made by mixing and matching the elements of the periodic table that lie
within the table s corresponding columns.  If you build crystals out of
one part gallium (an element in column three of the periodic table)
with one part arsenic (an element in column five), you get what many
researchers hope will be the next-generation electronic material,
gallium arsenide (GaAs), the preferred material for Cray
supercomputers, since electrons move more swiftly through it than
through silicon.  Moreover, GaAs is host to optical as well electronic
activity.  When electrons and positively charged carriers of current
called holes (because they are really the absence of electrons) meet
inside the material, they can recombine with one another in a
light-releasing embrace.

Each of these three-five or two-six materials-some made of three or
four elements-has its own characteristic band gap, which refers to the
energy difference between their electrons in an excited and mobile
state and those same electrons in the relaxed or valence state, in
which they remain hitched to a particular atom or group of atoms.  What
electrons and photons do inside of these materials depends on the size
of the band gap.  Some of these materials detect radiation of various
wavelengths by converting incoming photons into discernible flows of
electrons.  Others turn electrical energy in the form of moving
electrons into light, which is why these also are known as
optoelectronic materials.  It is this light-emitting property that has
put gallium arsenide, not silicon, at the heart of laser scanners in CD
players that bounce light off of the little hills and troughs on the
plastic compact discs.

Despite that material wizardry, the running joke amongst its developers
is that gallium arsenide is the next-generation material: always has
been, always will be.  The pessimism comes from the difficulty and
expense of producing it with the high quality necessary for making
microelectronic circuitry.  It took years of development and hard work
to develop a reliable method of growing silicon crystal boules from
which technicians could slice wafers that could be diced into millions
of silicon chips and then patterned with microcircuitry to become the
heart and soul of modern technology.  But silicon crystals are made of
one element, not two or more like gallium arsenide, or gallium aluminum
arsenide, or indium phosphide, cadmium mercury telluride, or any other
combination of elements in columns two, three, five, and six that
physicists know will form a new electronic, optoelectronic, or photonic
material.  Still, there are devotees of each of these elemental
combinations.

Other hot materials and trends du jour: liquid crystal polymers augur a
generation of flat computer-screen displays or even television screens
that roll up like posters; like gallium arsenide, electro luminescent
polymers can be made to emit light when stimulated electrically instead
of electrons and holes meeting like lovers, the electricity riles some
of the polymer electrons to excited states that vent by way of light
emission; diagnostic-minded researchers are training the vast and
powerful arsenal of analytical instruments on the material micro
structure of the wood in a Stradivarius violin or in a bronze Chinese
bell to reveal hard-won craft secrets that long predate the scientific
acceptance of the reality of atoms; metallurgists are finding ways of
combining metals in proportions and micro architectures that will yield
the right stuff for technologies like an aerospace plane that flies
twenty times the speed of sound.

The beat goes on: some scientists are beguiled by the prospect of
making ceramic objects like plates, spark plugs, and the tiles lining
the Space Shuttle s exterior by using nanoscale particles of metal or
ceramic particles that are comparable in size to viruses-the tinier the
particle, the more surface it has relative to its interior, and most of
a material s interesting features happen at its surface.  Ceramics made
with these tend to sinter or consolidate at temperatures hundreds of
degrees

FIGURE 31

Very, Very Tiny Grains.  Each of the discernible regions in this
high-resolution Image (from a transmission electron microscope)
corresponds to a tiny grain of palladium metal.  The grains aye on the
order of several nanometers; (billionths of meters) in height, length,
and/or width, which is why materials made with grains this size are
referred to as nanophase materials.  Researchers have shown that the
much smaller grain sizes of these materials (compared to the grain
sizes of powders used to make conventional metal or ceramic materials,
whose grain sizes often are hundreds or more times larger) lead to
important advantages.  One of them is the ability to make ceramic
components In complicated shapes without the difficult and expensive
machining that so fay has kept ceramics out of high-volume applications
where their resistance to heat and wear could make a big difference.
ARGONNE NATiONAL LABORATORY

the m a tert als s ereng e ti 0 185

cooler than with standard powders.  And they are more easily shaped.
These are the kinds of features that can save millions of dollars in
large manufacturing operations.  (See Figure 31.)

There is a computer-wielding, acronym-happy tribe that lives in a world
of theories and equations: quantum mechanical models that describe
where and how electrons move and differential equations that account
for the way a material s atoms and molecules move and jostle at
different temperatures and pressures.  They sometimes taunt their
dirtier-handed brethren by predicting new material structures that will
be in some way superior to any other material on earth: make a certain
arrangement of silicon and nitrogen atoms and you well get stuff that
is harder than diamond; make an odd ring of oxygen atoms and you will
make stuff that can pack more propulsive power than any known fuel.

Some trends solidify; others evaporate as quickly as dew on a summer s
morning.  The field as a whole, however, seems on the verge of a
critical mass of human and technical capability that will result in an
explosion of technological bedazzlement.  Rousing insensate matter into
something like intelligent behavior is not simply science, but an art
as well: consider the rich palette the researcher has in the periodic
table, whose elemental ingredients he or she can combine into an
infinite variety of molecular shapes and sizes and chemical blends. The
resulting material portraits depict our future.  Aping nature or
outsmarting it, man-made matter matters as never before.

PART III

Materials by Design

Materials by trial and error: even an Australopithecine protohuman
living 2.5 million years ago would have been justified in saying, Been
there, done that.  But inventing new materials by doing systematic
scientific research into the chemical and physical bases of the
properties of materials has become the preferred modus operandi of
materials researchers.  Call it materials by design: not been there,
never done that.

To do materials by design is to dream up new materials with a specific
suite of properties, to infer the way you need to put atoms together in
order to make your dream materials, and then to follow up the plan by
actually making the stuff of your dreams.  You might want to make steel
twice as strong as the strongest steel that now exists.  That way you
can use half as much to make each car or skyscraper.  You might want to
make new medical implant materials that corral the body s own repair
mechanisms to regenerate pulverized bone or atrophied muscle.  Maybe
your aim is to create new semiconductor materials that will be to
silicon what silicon became to the vacuum tube.

People have had material dreams like this before, and their dreams have
come true, but only by some combination of hit and-miss experimentation
and sheer luck.  A new and perhaps ultimate level of human control over
the material world is approaching.  In this era a materials researcher
will be able to tap a description of some never-before-seen material
into a computer that has been well stocked with all the relevant corpus
of knowledge about the ways of materials.  And out of the computer will
come the recipe for making the material.  In a fully automated CAD CAM
(Computer Aided Design/ Computer Aided Manufacturing) setting, the
design specification will immediately speed over communications lines
to a factory floor where computer-controlled and roboticized procedures
will transform starting ingredients into the specified material.

This scenario remains utopian, but not by all that much.  There already
are a number of cases in which forward-looking materials researchers
have pushed the envelope of discovery into this materials-by-design
paradigm.  There is no reason to think that they are flukes.  There is
far more reason to believe they are just the first materials designers,
akin to the first stone-flakers 2.5 million years ago or the primitive
metallurgists who first smelted metal from rock ore 10,000 years ago.

The immensity of a shift to a materials-by-design paradigm is hard to
overstate.  To date humanity hardly has tapped the diversity of
materials latent within the periodic table of the chemical elements. If
you begin with sixty or so elements of the table that you might
actually consider for making materials, take every possible combination
of five different elements in this set, and then vary the proportions
of each element in each of these combinations, you already are talking
about tens of billions of materials.  And that is not even considering
the rapid multiplication of this already large number if you consider
six-atom or seven-atom combinations, which is where many materials
scientists think the most important discoveries lie.  Nor does it
include that multiplier that comes with countless different ways you
might process these compositions using different combinations of
temperature, pressure, and other critical manufacturing conditions. The
number of materials humanity has so far probed, much less to put to
use, probably does not exceed a few million!  We have hardly scratched
the surface.

What makes materials by design such a heart-quickening prospect is that
humanity already has stumbled onto big-time, society-changing
materials-stone, bronze, iron, steel, glass, plastic, and
silicon-without the general ability to predict what will happen when
certain combinations of atoms and molecules are put together according
to some particular procedure.  And everybody knows there are mother
lodes in the periodic table just waiting to be discovered.

Francis DiSalvo, a veteran solid-state chemist at Cornell University,
has put it this way: Every material that can possibly come our way is
sitting there in the periodic table.  Every diamond, every
superconductor, every three-five optoelectronic crystal, every speck of
dust, every biological tissue, every material that ever was, is, or
will be is there hidden in the periodic table.

Its a puzzle given to us by God, and there has to be a key; maybe there
are lots of keys, DiSalvo says.  It s one of the most exciting brain
teasers that science will ever have.

Some of those keys, it seems, now are in the hands of a small but
growing cadre of visionary materials researchers who have tasted the
power of materials by design.

SIX

The RITE of ATOMIC MASONS

People love to talk.  That includes Federico Capasso, a leading
materials designer at the part of Bell Laboratories now called Lucent
Technologies, who already has found some of Francis DiSalvo s keys to
materials essential for the high-capacity communication of today and
tomorrow.

Communicating is as basic to human life as breathing and eating.  The
creation of a telephone network at the turn of the century provided
people with a new way to communicate to someone too far away to speak
with in person.  A telephone call is more intimate and immediate than a
letter or even a telegraph message because the two people talking can
hear each other s voices.  You cannot send laughter in a letter or over
a telegraph wire.

By the end of World War II, AT&T and Western Electric (the arm of AT&T
that manufactured telephone equipment) had erected a massive national
mesh of copper wire through which just about anybody anywhere could
reach just about anyone anywhere else.  But the nation s hunger for
telephonic connection was merely whetted.  This was good for the
company.  But Mervin J. Kelly, then executive vice president at Bell
Telephone Laboratories in Murray Hill, New Jersey, where AT&T had
thousands of researchers doing the basic and applied research for
keeping up with future communications needs, looked at the existing
network and got worried.  It was the switches, the millions upon
millions of switches, that unnerved him.

When a person made a phone call in the late 1940s, a series of little
ticking sounds could be heard as a tiny electrical current shot from
the caller s phone and then journeyed through a network of
electromechanical relays that kept handing off the cur rent until it
made it straight into the other person s telephone.  An automatic
switching station sounded like a roomful of people with chattering
teeth.  Before automatic switches, it was a roomful of
people-operators-that switched the right circuits on and off by
plugging and unplugging wires in and out of sockets.  In a mechanical
switch, or relay, a tiny current would loop around a metal coil to
create a magnetic force.  The force would move a small metal arm that
would open or close a circuit by literally completing or interrupting a
conductive pathway for electrons.  It was the movements of those arms
that made the little ticking sound.  Once an unbroken electrical
pathway was established between two specific phones, then voices could
journey over copper wire back and forth in the form of electrical
pulses that could drive a telephone s speaker.  What concerned Mervin
Kelly and AT&T was all of the ticking going on in the millions upon
millions of switches.

They knew that the telephone system would have to expand greatly to
meet the country s growing need to communicate.  The mechanical
switches that were making the miracle of person-to person direct-dial
phone calls possible would not be able to handle the projected traffic.
The switches were too bulky.  Their moving parts wore out.  And they
also needed too much energy to move those little metal arms to make
that ticking sound.

the rite of the atom?,c masons 0 193

Kelly let it be known around Bell Labs that he wanted a replacement for
the pillbox-size mechanical relays.  Make a switch with as few moving
parts as possible-even none, if it were possible.  Reduce the amount of
energy needed to run each relay.  And make them small.  That s what
management wanted.  Electron tubes seemed a ready-made candidate, but
they could only work as a switch after heating a metal electron source
inside the tubes to white hotness.  Moreover, producing them in the
quantities required was impractical.

The conundrum inspired Walter Brattain and John Bardeen-whose
collective talents included theoretical and experimental physics,
chemistry, crystallography, metallurgy, electronics, and precision
mechanics-to find a solution!  By the end of 1947, they found a way to
build all of the functions of a mechanical relay or an electron tube
into the invisible movement of electrons inside a crystal.  In other
words they invented the transistor.

They evaporated a tiny spot of gold onto a crystal of germanium, which
had gained some fame among scientists for its ability to detect faint
radar signals.  When the Bell researchers put a positive charge on the
gold spot atop their germanium crystal, they induced a negative charge
inside the underlying crystal.  That was like clearing a previously
blocked pathway for a stronger flow of electrons through the crystal.
It became known as a point-contact transistor, a solid-state electronic
switch.

The means to meeting Kelly s edict was in hand.  The ticking of the mat
al arm in a mechanical relay could conceivably be replaced by the
silent governance of electric charges.  There were no moving parts in a
transistor, it didn t take much energy to operate, and was potentially
very small.  For their momentous invention, Brattain, Bardeen, and
William Shockley (who invented another type of transistor, the junction
transistor, which displaced the point-contact transistor) won a Nobel
Prize in 1956.

The Nobel Prize committee had no idea how dramatically the new
electronic gadget would change and quicken the world.  It was the
beginning of the end for big, power-gulping vacuum tubes and the
beginning of solid-state electronics.  A transistor, after all, was a
specially made solid object.  It didn t have to be enclosed in glass,
so it didn t contain any air or gas of any kind.  It didn t involve
liquid.  It was solid through and through.  A decade passed before
anyone knew if this seed of solid-state electronics really would
germinate and render vacuum tube electronics a has-been technology.

Ten years after the first transistor, Bell researchers invented the
laser, and by 1976 a large community of researchers was striving to
marry the electronic purity of solid-state devices with the optical
purity of laser light.  When a young Italian physicist named Federico
Capasso arrived at Bell Laboratories in 1976 for what was supposed to
have been a nine month visit, Bell Laboratories had become a Mecca of
physics.  For Capasso, an optical physicist interested in the many ways
that light and solid materials interact, coming to Bell Labs was a
dream.  Written in metal letters on a wall as you enter AT&T s flagship
laboratory in Murray Hill, New Jersey, are these words of Alexander
Graham Bell: Leave the beaten track occasionally and dive into the
woods.  You will be certain to find something that you have never seen
before.  They were words that Capasso took to heart.  He now heads the
Quantum Phenomena and Device Research Department at Lucent
Technologies, a portion of the recently reorganized Bell
Laboratories.

Capasso, a short, ebullient man with unkempt, thinning hair and big
gap-tooth smile, practices on the leading edge of materials science. He
and his colleagues design solid-state materials from atomic scratch,
mostly semiconductor crystals in the same general material family as
germanium and silicon.  These designs rely on the deepest theories of
how atoms and electrons and energy behave-quantum mechanics, that is.

the rite of the atomic masons M 195

Capasso s team uses a miraculous tool called molecular beam epitaxy
(MBE) that can spray-paint a crystal into existence one atomic layer at
a time.  Like the inventors of the transistor, Capasso likes to make
tiny solid pieces of crystals that have no moving parts yet behave like
machines.

Capasso s work at Bell originally concentrated on the detection of
invisible infrared light, which was the region of the spectrum that
traveled well in the optical fibers.  Nine months passed, and Federico
indeed had developed a better feel for the leading edges.  But instead
of returning to Italy, Federico stayed on at Bell Labs.  He never
left.

Capasso soon became a pioneer of band-gap engineering, a term he says
he coined himself in 1983 when he and his research kin had acquired the
kind of control over matter that the term implies.  When Capasso says
band-gap engineering in his strong Italian accent, it sounds like
bang-up engineering.  For now, think of Capasso as an electrician who
fiddles not with wires and wall sockets but with the electrons moving
inside the tiny atomic jungle gyms within artificial crystals of his
own design.  Strange things can happen inside these crystals.  You can
send electrons into them and get pristine beams of laser light to come
out of them.  You can make them into tiny little computers that can
tell you if two numbers are the same or not.

These crystals begin as figments of a prepared imagination.  They lie
latent between the terms in arcane equations of difficult physical
theories.  They come from the researchers intuitions about how
electrons move within solid materials.  They often are born from energy
diagrams, which are rough sketches that show what the territory within
a specific semiconductor crystal might look like to electrons: where
within a crystal an electron encounters barriers, repulsions, or open
doors.  Combining theory, intuition, energy diagrams, and volumes of
data from real materials, Capasso and his researchers test and refine
their ideas.

Band-gap engineers do not do the kind of physics that typically catches
the public s attention, things like creating a new particle of matter
in a gargantuan accelerator or detecting the earliest remnants of the
Big Bang with satellite-born instruments.  Instead they do solid-state
physics.  They study how solid materials are built and how they behave,
a preoccupation that encompasses a good part of our everyday natural
and artificial landscape.

The term band-gap enginee?-ing is a hieroglyph to most people, even to
scientists who have gone through ten years of collegiate training, a
post doc or two, and decades of employment in hightech settings.
Capasso s quick definition doesn t help: We are using the laws of
quantum mechanics to design new materials.  Translation: We are using
theories and calculations about how electrons behave in crystals to
invent new high-tech materials.  What is remarkable about this claim is
that no one has ever actually been able to say with a straight face
that he or she had designed a new material.  People had always stumbled
upon new materials by trial, error, and serendipity, or by varying
details of inherited craft knowledge.

The ability to design and make materials with specific properties is,
for the materials scientist, tantamount to breaking the 4-minute mile
or breaking the sound barrier.  It is a celebration of human
capability.  It is a harbinger of achievements to come.  And since it
is materials that so often limit existing technologies and spawn new
ones, the ability to design materials is a general key to accelerating
technological development.

The general goal of materials designers is to uncover previously
unknown or unrealized properties that matter can have in principle but
has never realized in practice.  This approach to materials marks a
drastic shift from the past.  Since ancient times people have known
that the earth harbors various substances that could be useful in their
raw states or processed to bring out more useful features.  The key to
unlocking the multifarious personalities of matter, however, remained
hidden well into the nineteenth century.  Chemists were learning by
trial and error how to build and break molecules, but they had little
idea why some things worked and others didn t.

In the 1920s, as scientists were still expanding the vast technological
possibilities created by chemically transforming the raw substances of
the earth, physicists began to uncover a deeper layer-the quantum
mechanical-of the mystery of stuff.  Quantum theory began to make
sense, albeit strange sense, of why different atoms and molecules
behaved the way they do.  Quantum theory was a means for the scientific
imagination to get far more intimate with atoms and atoms electrons,
the handmaidens of chemistry, so that they could first explain and then
predict material behavior.  In this way, theorists finally have been
able to make discoveries before experimentalists.

Band-gap engineers like Capasso do in fact mine the labyrinthine
theoretical framework of quantum mechanics for clues to entirely new
materials.  Just how they do it requires a bit of elaboration.  In the
world according to quantum mechanics, the electrons bound to each kind
of atom can take on a set of specific energy levels and no levels in
between.  If an electron in a silicon atom were a car, it could travel
at, say, exactly 10 miles per hour, 17 mph, 28, 43, 58, and 81, but
never at speeds in between.  Give it enough juice at 17 and it goes
instantly to 28 without passing through 18, 19, 20, 21, 22, 23, 24, 25,
26, or 27.  Strange but true.  In a different atom, maybe carbon or
titanium or indium, that same electron would have a different
characteristic set of energies.

Now snorkel a little deeper into the quantum mechanical abyss.  If you
pack zillions of atoms into a solid crystal, the rules of electronic
energy and behavior can change drastically (electrons are like people;
their individual behavior changes when they find themselves in crowds).
For one thing, the number of discrete energy levels allowed for an
electron bound to one of the crystal s atoms increases to the point
where the energy levels smear together into a band of energies-the
valence band.  At the same time, the crystalline venue brings forth a
new, higher band of possible electronic energies-the conduction band-in
which electrons are no longer restricted to stay near this or that
atom.  Instead they can become free agents and flow through the
material like water through a pipe.  Between the valence band and the
conduction band is a no-man s land for electrons, a region of forbidden
energies.  This is the band gap.

In conductive materials like copper and gold, the conduction band is
only half filled, which means that the metal s electrical resistance is
small.  The slightest impetus, such as hooking these materials into a
loop with a tiny battery, will send electrons moving in the conduction
band to form an electrical current.  This little bit of quantum
mechanics happens every time you plug in a lamp or talk on the phone.
In insulators like rubber or mica, on the other hand, the gap is so big
that getting electrons to jump across would require so much input
energy that it would deep-six the material.

Semiconductors, the materials that Capasso spends most of his time
thinking about, are an entirely different kind of electronic creature.
The most famous of these, silicon, is what Bill Brinkman, vice
president of physical science and engineering research at Lucent
Technologies, describes as a material given to you by God.  The band
gap in a semiconductor is too big for electrons to jump without some
external impetus such as a little heat, light, or voltage (which is
like a slope for a river of electrons).  But it is not so big that the
input of energy blasts the material to kingdom come.  What s more, each
semiconductor material like silicon or germanium or gallium arsenide
has a different band gap.  Each has its own personality in the arena of
electronics and related arenas such as optoelectronics (the CCD
detectors in camcorders are optoelectronic devices that turn light
stimuli into electronic signals).

Semiconductors with relatively small band gaps make for supersensitive
heat and radiation sensors.  In the Hubble Space Telescope, it is an
array of such small band-gap crystals that gather the dimmest signs of
light from the big mirrors.

The body heat of a person 50 yards away from a crystal of mercury
cadmium telluride can knock enough electrons into motion within the
crystal to set off alarms.  The National Security Agency undoubtedly is
well versed in the ways of such materials.

Semiconductor materials with somewhat higher band gaps are the stuff of
transistors and computer chips.  In these materials it takes a little
more but not too much more, to get electrons to jump the gap.  Silicon
does this so well at room temperature that it has become the backbone
of a technological revolution.  But gallium arsenide has a way with
light that silicon doesn t. This is especially true when gallium
arsenide is sandwiched between closely related materials with higher
band gaps: for example, aluminum gallium arsenide.

This is one of Capasso s favorite topics.  Once energized into gallium
arsenide s conduction band, electrons can drop back down into the
material s valence band by recombining with a positive charge.  The
flanking materials with high band gaps keep all of the action within
the meat of the sandwich.  This recombination generates light.  Thus do
tiny semiconductor lasers send conversations coded as light pulses
through glass optical fibers.  In every CD player a solid-state laser
that works via this recombination principle sends a beam of light over
the pattern of pits on the disk and then into another solid state
detector that converts the pattern of reflected laser light into the
electrical pattern that drives the sound system s speakers.

Silicon and gallium arsenide are but figureheads of a vast population
of lesser known semiconductors.  By mixing and matching two, three,
four, and even more elements from certain columns of the periodic table
(the third and fifth, and the second and sixth), you get semiconductor
materials with a huge range of band gaps, which means a range of
materials with interesting optical and electronic properties.  If you
can control the way these materials come together in crystals, you open
a practical door to these materials.

To solid-state scientists, this mix-and-match game is a technology
engine that roars with power.  Some three-five materials, as they are
called, are just right for making super fast transistors, which can be
used for speeding up telephone switching systems or computer chips.
Others are good for making the infrared lasers that now send light down
hundreds of thousands of miles of optical fibers.  Others are good for
making the yellow- or green-light-emitting diodes that have changed the
face of consumer electronic displays.  Some are the key to photovoltaic
cells for converting sunlight into electricity.  In this case the light
energy of the sun kicks electrons in the photovoltaic crystal directly
into the conduction band.  Still other semiconductor crystals can
detect radiation.  And those are just the more conventional
applications.  This rain forest of semiconductor diversity has become
the favored locale of thousands of materials scientists, physicists,
chemists, and electrical engineers, all seeking to discover the stuff
that will spawn another Silicon Age.

That opulent diversity notwithstanding, band-gap engineers like Capasso
want to squeeze yet more out of the periodic table than is available
from straightforward combinations of elements.  The specific band gaps
you can get out of these atoms when packed into crystals are merely the
ones that mother nature has seen fit to bestow.  Capasso does not like
to be boxed in like that, especially when his designs for new
transistors, light detectors, or lasers call for a band gap that isn t
in nature s standard catalog.

Soon after Capasso arrived at Bell Labs in 1976, he began to see a way
around these seemingly ineluctable natural restraints on band gaps.  He
sensed that there was an entirely new range of possible band gaps, and
he thought he knew how to find them.  If he was right, the new freedom
to dictate band gaps would lead to a vast new arena of materials, some
of which could be as technologically important as silicon.

One of the biggest goads was the same one that led the managers of
research for the Italian Post Office to send Capasso to Bell Labs-the
gut belief that communication based on electrical pulses traveling down
electrical wires would have to give way to communications systems based
on light pulses traveling down some kind of wires for light.  Major
cities already were running out of underground space for fat copper
cables, yet the communications demands of their citizens were
accelerating relentlessly.  If communicating by light was the wave of
the future, researchers would have to devise components that could
generate the right kind of light (laser light), guide that light
globally, and detect all of the optical commotion.

SPRAY PAINTING WITH ATOMS

In the 1970s the pieces of an optical communications system were
emerging from the primordial soup of materials research.

One of the biggest pieces had come from brilliant work at the country s
premier glass and ceramics research center at the Corning Glass
Company.  In 1970 researchers there demonstrated that they could make
glass fibers pure enough and leakproof enough that light entering one
end would still be detectable many miles down the fiber.  It was an
astounding achievement, akin to making window glass able to transmit
light through miles of thickness.  For years this goal had eluded other
researchers, who couldn t get rid of enough of the physical defects
like tiny bubbles that scatter light in useless directions or chemical
impurities that absorb light.  They always ended up looking through a
glass fiber darkly.

The Corning solution involved evaporating pure silica (the stuff of
sand, which is made of silicon and oxygen atoms) into a rod that they
then could pull like taffy into astonishingly thin fibers.  A crucial
and brilliant detail, however, was to ever so slightly contaminate the
inner layers of the evaporated silica with other atoms.  This resulted
in an inner core of glass surrounded by a cladding made of a purer
glass formulation.  This made the outer cladding a reflective sheath
that would send any light hitting it back into the core of the fiber.
Try as it might, light in the core just could not leak out.

At last optical fibers could serve as long-distance wires for light.
Corning was betting so highly that their fibers would become a mainstay
of future communications systems that they built a multimillion-dollar
factory in 1976 before they even had a single major order.  That put
the pressure on places like Bell Labs and Japan s NTT to devise and
build the lasers and light detectors that would hook onto either end of
these fibers.

Groundwork had been laid already.  The same year the Corning
researchers made their optical fiber breakthrough, Morton Panish and
Izuo Hayashi at Bell Labs had built the kind of solid-state laser that
would inject light into these hair-fine conduits of light.  By
sandwiching a thin layer of gallium arsenide between two layers of
aluminum gallium arsenide, they had made a solid-state semiconductor
laser that worked at room temperature and that could lase continuously
rather than in intermittent spurts.  Previous attempts to make such a
laser had always resulted in tiny imperfections that allowed the energy
in the excited electrons to vent in ways that didn t emit light or that
blasted apart the crystal s internal architecture during continuous
running of the laser.

Bell was the right place for Capasso for many reasons, but the presence
of Alfred Cho was probably the biggest.  By 1980, when the two would
begin the first of many collaborations, Cho had already spent a dozen
years doggedly working out the bugs of an astounding tool called
molecular beam epitaxy (MBE).  Its potential for growing nearly perfect
crystals one atomic layer at a time fueled Capasso s quantum mechanical
imagination like kerosene on hot coals.

At the center of an MBE machine is a stainless-steel chamber that looks
like it could withstand the pressure at the bottom of the ocean. Inside
there are thick glass portholes ringed with heavy duty bolts like
submarine hatches.  Tubes go into the chamber at every angle, and there
are mysterious the rite of the atomic masons m 203

appendages and boxes all over.  Jumbles of wires, which seem to extend
in all directions, terminate in racks of electronic boxes and a control
station.  The machine looks like a metallic Medusa.  Separated by
shutters from the main chamber are small furnaces with crucibles
containing the atomic ingredients needed to make a band-gap engineer s
dreams come true.  When the crucibles are heated, such atoms as
silicon, arsenic, gallium, aluminum, indium, and phosphorus jerk loose
and start to diffuse into the MBE chamber.

By controlling the shutters with a computer, an MBE operator directs
beams of various atoms into the growth chamber and onto the surface of
a crystal template waiting there like an unpainted wall-except that
this wall is a crystal surface resembling an endless atomic-scale egg
carton.  Each barrage of atoms coming from one or more of the crucibles
fills up the corrugations in the crystal surface to produce a brand-new
crystal layer ready to receive the next barrage.  The template crystal
is heated to ensure that the incoming atoms will scramble around until
they find a site at which the chemical attraction is strong enough to
hold it.  Cho himself likens MBE to spray painting with atoms.

The process is inherently slow, though researchers have found ways of
speeding it up.  It typically takes about one hour to grow one
high-quality micron s worth of a crystal.  It would take a few days to
get a crystal as thick as a sheet of paper.  In a year the crystal
still wouldn t span half an inch.  A bonus of this sluggishness,
however, is that MBE operators can decide just how many layers of each
element or elemental combination they want.  Moreover, as they go, they
can leak tiny amounts of atoms known as dopants into the beams.  These
can provide the electronic or optical spice that can make these
constructions scientifically or technologically interesting.

To visualize what happens in an MBE machine, think of selling eighty
thousand general admission tickets to the Super Bowl and then just
opening the gates an hour before game time.

Even if by some miracle the rushing crowd managed to fill all of the
seats in perfect order, that would only represent a tiny fraction of
what needs to happen to grow a crystal.  Imagine now if the stadium had
billions upon billions of seats extending indefinitely in all
horizontal directions and stacked hundreds or thousands of layers high.
For MBE to work well enough for the crystal building purposes
envisioned by Capasso, atoms leaked into the vacuum chamber would have
to completely fill that enormous three-dimensional stadium almost
flawlessly.

Lucky for Cho, atoms are not quite as unruly or as unpredictable as
rowdy football fans.  Over the years he reduced temperature
fluctuations in the instrument, got rid of impurities from the
ingredients and machine, and disarmed other gremlins that would pock
the resulting crystals with too many defects.  The technique was so
good by the mid-1980s that it became the manufacturing tool of choice
for making most of the semiconductor lasers shining invisibly inside
compact disc players.  Over 70 percent of the world supply of compact
disc lasers are made by MBE, Cho likes to point out.  MBE is now so
good that a square centimeter of crystalline real estate-an enormous
expanse of over I quadrillion atoms-can have as few as three atoms out
of place.  That is like searching through a population one hundred
thousand times that of the earth before finding a crazy person.
Solid-state devices of the not-too-distant future may require that kind
of material perfection.

So there was Capasso in the late 1970s within a stone s throw of
perhaps the only machine in the world capable of building a crystal
atomic layer by atomic layer.  It was knowing that MBE was there that
made me really think along the direction of band-gap engineering,
Capasso recalls.  The kind of unprecedented control over matter that
MBE could provide, Capasso intuited, could prove to be the key for
turning a lot of quantum mechanical dreaming into real,
hold-in-your-hand embodiments.  MBE was a tool that could turn
equations, physical intuition, and back-of-the envelope energy diagrams
into the rite of tize atomic masons 0 205

reality.  Moreover, it was just this kind of thought-to-stuff
transformation that eventually helped to take Bell Labs and others to
the threshold of the era of light-based communications.

THE ROAD TO BAND-GAP ENGINEERING

Until Cho and Capasso began collaborating in 1984, no one had ever
deliberately done the kind of fundamental fiddling that band-gap
engineers do.  People had thought about it, done calculations, and made
predictions.  Now MBE was ready to put all of this effort to the test.
The 1980s were for solid-state science a golden era, Capasso says.

To be sure, others had begun paving the way to these good times.  In
the early 1950s William Shockley, one of the three physicists who
invented the transistor, had an inkling of the gold to come.  He
surmised that a semiconductor crystal made out of two different
materials with an atomically abrupt interface-which no one knew how to
make then-would lead to higher performance transistors (electronic
switches) than would be possible with germanium or silicon alone.

In 1957 Harry Kroemer, then at RCA (whose corporate empire was built on
vacuum tubes), took another important conceptual step toward band-gap
engineering.  Kroemer s idea was that by slowly changing the
composition of a semiconductor crystal, you could build what amounts to
an electron propulsion system into the crystal itself.  Electrons
wouldn t just mosey on through-they would rush forward as though they
had entered a river s rapids.  It took thirty years before anyone put
ideas like that to the test.

A few years later, Kroemer, whom Capasso considers an intellectual
father, had another idea.  He suggested that sandwiches of certain
semiconductor materials would actually behave as tiny solid-state
lasers.  At the time, lasers, which were first proposed theoretically
in 1958, were bulky affairs in which the light emerged from excited
electrons in gases.  The first working laser was built two years later.
Laser light is the result of electrons getting pumped up to a higher
energy state and then falling back down to earth with a blaze of
one-color light.  Normal light sources emit many colors or wavelengths,
which is why they appear white.

In 1970 IBM-then one of Bell Lab s few rivals in research and
development-got into the act.  Leo Esaki (another elder of band-gap
engineering who later won a Nobel Prize) and Raphael Tsu, then both at
the IBM Thomas J. Watson Research Center in Yorktown Heights, New York,
published a theoretical paper in IBM s house research journal.  It
became one of the most widely cited sources for a semiconductor
structure called a quantum well, which became for Capasso and other
band-gap engineers what two-by-fours are to carpenters.  (See Figure
32.)

Like water wells or inkwells, quantum wells are good at confining
something.  Unlike wells that people can fall into or that can hold
ink, quantum wells are measured in angstroms (a ten billionth of a
meter).  They are sometimes only a few atomic layers thick.  Under a
souped-up microscope, a quantum well looks like a crystalline sandwich.
The meat is a layer of one material like gallium arsenide, and the
bread is made of another material like gallium aluminum arsenide. These
kinds of wells confine electrons to discrete energy levels much the way
a string fastened between two fixed pegs is constrained to vibrate

Atomic Construction.  Each blob in this electron microscope image
represents an atom of either Indium, gallium, arsenic, or phosphorus.
The more tightly packed group of three layers in the middle is indium
gallium arsenide, flanked on top and bottom by thicker layers of indium
phosphide.  At the boundaries of the different materials there is an
abrupt change in the potential energy.  To an electron passing through
this three-layer molecular sandwich, these potential energy changes, or
steps, amount to an electron launcher.  Such launchers, in turn, can
serve as building blocks for solid-state devices such as radiation
detectors or photomultipliers that depend upon cascades of accelerated
electrons.  This specific structure was grown by Morton Panish at Bell
Laboratories using molecular beam epitaxy.  BELL

LABORATORIES, COURTESY J. M. GIBSON AND S.N.G. CHU

FIGURE 32

only at discrete frequencies.  Some of the electrons in the cur rent
moving though the crystal fall into these quantum wells.  Once in, they
have a hard time climbing out.  That captivity makes them a little
crazy.

A year after Esaki and Tsu floated their theoretical idea, two Soviet
physicists then working in Moscow, Rudy Kazarinov (who ended up working
in Capasso s group) and Robert Suris, had registered another key idea
in an obscure Soviet journal on semiconductors.  They wrote that
quantum wells, if arranged in a series, could be a route to a new type
of solid-state laser.  The wavelength or color emitted by the laser
could actually be chosen dialed in, so to speak.  Electrons would
drop-or tunnel, in the parlance of quantum mechanics from quantum well
to quantum well, losing energy each time by emitting photons of
wavelengths that correspond to the energy jump.  Almost a quarter
century later, this became a central design principle for Capasso s
most prized achievement.  If these Russians were light, what seemed
like an arcane theoretical suggestion by the researchers at IBM could
be the fount of some very practical technologies relevant to optical
communication.

Capasso s first real foray into band-gap engineering began with a
project to make light detectors, a key component for optical
communication.  He was still working at AT&T s facility in Holmdel, New
Jersey, which was fated to become the research and administrative
headquarters for the company s optical communications business,
including its massive business crisscrossing the oceans with undersea
optical cables.

Capasso took on the challenge of trying to develop a new breed of light
detectors that would work well in the specific wavelengths that
traveled best down optical fibers, namely, 1.3 to 1.55 microns, which
are invisible infrared wavelengths.  The most sensitive kind of light
detector available was known as an avalanche photodiode (APD) and was
made out of the old standard-silicon.

In an avalanche photodiode, a process known as impact the r it e
ionization amplifies tiny Ii nals.  The incoming light then gain energy
from the For the bumped electrons, the raging current.  These i
electrons from the silicon band.  The high voltage, ho newly knocked
electrons t( valence band to the condu an avalanche of electrons light
signal going into the g tronic signal.

That is just the sort long-distance optical fiber E

of the glass fiber would be @ wise, the system would have expensive and
complicatec amplify the weakened lig) journey-that prospects s!

would be impracticable.

APDs were clearly n, was problematic.  The wave tronic avalanche within
its that travels best down op with a silicon detector is li a concert.
The materials had t( length of 1.3 microns, Cal that new semiconductor
ma including some with three a

What eventually emerged which are used in long-dis combination of
indium pho But this combination had itis data transmission needed A

major traffic jams: it detects 1.3 micron light but produces a noisy
electronic signal that limits the rate at which it can reliably detect
data.  So for higher-capacity communications, the standard APIs would
work only with stronger light signals that could emerge like a
distinctive voice through the chatter of a cocktail party.  The
requirements were clear: either very strong lasers or an unwieldy
population of repeaters that could doom, for example, a transoceanic
application.  The materials that people were using to make avalanche
photodiodes for light communication at 1.3 to 1.55 microns didn t have
low enough noise, Capasso recalls.  The noise comes from the fact that
electrons are not the only things that get sent into motion when light
hits these detectors.  Positive charges, which are called holes because
they are the same as the absence of an electron, end up streaming in
the other direction.  When they do this, they too knock other electrons
loose, which in turn do the same.  This is counterproductive feedback
and ends up reducing the sensitivity of the device while making it
harder for the real signal to shine through.

It would take some very different thinking to get rid of the noisy
feedback.  As a materials designer, Capasso tended to work backwards,
to imagine the properties he wanted in a material and then to figure
out the atomic arrangement that might actually perform that way.  In
essence, he dreamed up materials and then convinced a crystal grower to
make his dreams come true.  It had always been the other way around:
someone would make a material that happened to have specific
properties, and then people would try to figure out how to put the new
stuff to use.

For the avalanche photodiode challenge, Capasso reasoned that in the
ideal material, electrons would produce an avalanche of other charges
but that the positively charged holes would not.  He knew that such a
material did not exist and that it probably shouldn t even be able to
exist.  The dogma was that if electrons get knocked into motion in a
detector, holes will be sent into motion, too.  Nevertheless, he
persisted.  I asked myself, can I create an artificial material where
only one type of charge carrier creates an avalanche?  Capasso
recalls.

It is amidst challenges of this sort that intuition saves the day.
Capasso says, There was no law that said it could not be done, and
usually in nature where there is something that is not forbidden, you
know that eventually it can happen.  This could be a slogan for
band-gap engineers.  The technological carrot was there, too, according
to Capasso: I knew that if I could come up with such a material, people
would use it in light wave communication.

In the summer of 1978, it all began to coalesce in Capasso s mind.  All
of a sudden one day I got the idea.  We had to make what essentially
would be a launching ramp or a ski ramp for electrons.  In such a
material any electron knocked loose by a photon would take off like a
bat out of hell, kicking up electrons and holes in its wake while the
hole would stir up far less electronic activity.  He even had an idea
of how to realize this situation creating a structure of alternating
layers of aluminum arsenide and aluminum gallium arsenide.  This
material striping was similar to the superlattices that Esaki and Tsu
had suggested years earlier.

Gallium arsenide and aluminum gallium arsenide were natural choices.
For one thing, they were familiar to researchers.  For another, their
quantum mechanical properties (such as band gaps) were quite suitable
for creating the ski ramp that Capasso had in mind.  The abrupt
transitions between the two alternating materials would create energy
steps, which are larger for electrons than for holes.  That means that
the electrons would gain more energy than holes and would be more
effective at creating electron avalanches then hole avalanches.

As soon as he floated the idea, doubters sprang to the fore.  Some said
the electrons wouldn t be able to make it up an energy barrier in front
of the ski ramp that Capasso could see was necessary from the energy
diagrams describing the low noise APD.  And a stranded electron is a
dead end for an avalanche photodiode.  There were a lot of shaking
heads.

All of that nay-saying only spurred Capasso on.  I was so upset that I
said, I am going to solve this, and that is that.  He combined what he
knew and intuited about band gaps with what he could calculate about
how electrons behave in solids.  Then he designed a superlattice-a
stripe of gallium arsenide, a stripe of aluminum gallium arsenide, a
stripe of gallium arsenide, and so on.  Each layer in the blueprint had
a slightly different thickness.  Calculations showed that such a
nuanced architecture would lead to a crystal harboring a series of
regions with different band gaps that collectively would produce ski
ramps for an electron.

At this point in a Capasso project, a technical relay sets into motion.
The blueprint for the crystal goes to the MBE lab, where a wafer of the
designed crystal is grown.  The wafer goes to another colleague, who
dices out tiny pieces of crystalline real estate and attaches metal
leads to the little plots.  The specimens are then ready for physical
testing to see what they can do.

The experiment to see if light triggers an avalanche of electrons-but
not an avalanche of holes-in a piece of crystal is to inject a specific
dose of light on one side and then see if an inordinate amount of
electrons come out the other.  The input output ratio should be akin to
whispering into a mike and then hearing a thunderous sound come out of
the speakers.  In Capasso s design there also should be a lot less
noise from holes.

In the 1980s Capasso and various colleagues indeed came up with a
family of APDs, many of which performed just as predicted.  These
researchers actually had dreamed up materials with capabilities never
before seen and then made them real.  But it was a somewhat
anticlimactic experience.  The problem was that the growth process was
too complicated and inefficient for commerical production.  Although
Capasso rates that experience as one of the greatest satisfactions of
his life, no one else paid much attention.

Starting in the mid-1980s, however, people did begin taking notice.
Eventually some of them, Japanese researchers mostly, jumped all over
it, according to Capasso.  Researchers at three Japanese
companies-Hitachi, NEC, and NTT-began work on multi quantum-well APIs
that were similar to Capasso s or based on the same principle.  A
decade after Capasso s team had first introduced them, the devices
finally were getting close to commercial viability; now AT&T also has
begun developing them.

All of that Sturm und Drang over whether his ski-ramp idea would really
work led Capasso to another insight.  Applying a voltage across a
superlattice changes the energy landscape within the crystal.  He
realized that he could actually transform the energy landscape from one
resembling a series of walls and wells into something that looked like
a staircase.  Says Capasso, This would be an artificial material, which
could be used to make the solid-state equivalent of the photomultiplier
tube.  It is a goal much like Mervin Kelly s call, forty years earlier,
for a solid equivalent of the mechanical relays that were running the
phone network in the 1940s.  A PMT, as a photomultiplier tube is
called, is a tube that multiplies light energy so that you can detect
little bits of it or use less radiation to achieve goals like taking
medical X rays.  It multiplies electrons like an avalanche photodiode,
but there are no holes, and so it is less noisy.  Its drawback is that
it operates under much higher voltages.  (See Figure 33.)

That kind of signal amplification in an MBE-made crystal is just the
kind of material talent that makes for world-class radiation detectors.
By 1995 neither Capasso nor anyone else had actually realized a
solid-state PMT.  But the fact that Capasso could actually came up with
a blueprint for such a thing was a reminder that he had found a
door-band-gap engineering-to a mother lode of scientific and
technological leads.

In the early 1980s, with exquisite control over semiconductor crystal
structure, with quantum mechanics as a fountain of leads, and with the
overall technology goals of Bell Labs as a context, Capasso began to
perform thought experiments, mental simulations of what it is like to
be an electron inside of a crystal: I am an electron sitting in this
quantum well..  .. He was developing an intuition for what would happen
in different superlattices harboring different landscapes of band
gaps.

By the end of the decade, Capasso was giving talks at meetings about
ever more exotic devices grown by MBE, not just detectors and lasers.
One of them was the resonant tunneling transistors in which electrons
tunnei between wells at a handful of specific energy levels.  This
opened a route to a new kind of microelectronic circuitry whose logic
was more powerful than the simple on-off, yes-no, 1-0 binary logic
underlying conventional electronic circuitry based on standard
transistors.

Back-of-the-Envelope Pictures of Energy Inside a Crystal.  Band-gap
engineers like Federico Capasso rely heavily on simplified portrayals
of how the energy landscape inside a complicated crystal changes at
different locations within the crystal.  Top: Each upwardly rising
stage of this energy diagram corresponds to many layers of a crystal
grown so that the amount of energy needed for an electron to jump from
being bound to an atom to being free to conduct continually increases.
Putting several similarly grown stages in sequence leads to an energy
diagram that looks like a sawtooth.  In this case, an electron
continuously gains energy and then falls down the end of a tooth at
which point more energy must be injected to get the electron to travel
up the slope of the next sawtooth.  Bottom: When the same crystal is
Sandwiched between electric contacts and a voltage Is applied, the
electric field tilts the sawtooth landscape downward into a
staircase-like situation.  A free electron created by absorption of
light within the crystal goes down the first step gaining kinetic
energy.  Some of this energy can transfer to some more of the crystal s
electrons, which then begin to move down the next energy step all the
while gaining kinetic energy.  This process Is repeated at each step,
leading to a multiplication of the initial electrons.  The result is a
crystal that behaves like a solid-state photomultiplier: It responds to
incoming photons, or light, with a disproportionately large electric
signal.  BELL LABORATORIES LUCENT TECHNOLOGIES

FIGURE 33

Staircase Solid-State

Photomultiplier

Each of the tunneling energies of resonant tunneling devices could
serve as a definable electronic state, each one a value for a multi
value logic.  So each one of these quantum devices presumably could do
the role of many standard transistors.

It was a nice idea, but using multivalued logic and circuit components
was too radical a departure to find acceptance in any but the most
specialized applications, at least for the moment.  Actually using
these devices entails a new kind of programming, a raft of new
ancillary electronic components that can handle the multi state logic,
and plenty of other unconventional requirements and headaches. Capasso
did prove the principle by using resonant tunneling devices to make a
parity checker, which looks for differences between binary numbers;
this is essentially what a computer does when it checks to see if two
stored documents are the same or if they have some differences. It
takes several dozen conventional electronic components to do the same
job.

These dream-to-hardware feats were just warm-up exercises.  They were
good for training and for demonstrating that the audacious ambition of
designing materials from atomic scratch was not the empty optimism of
deluded visionaries.  In 1986 Capasso began doing some preliminary work
and drawing up energy diagrams for a new kind of solid-state laser
realizable only through band-gap engineering.  In a key experiment he
showed that electrons could cascade from well to well via tunneling if
the energy levels of adjacent wells were properly aligned.  Following
this experiment he aimed to develop a generic strategy for designing
devices that would lase at just about any predetermined wavelength even
if no known semiconductor happened to have the appropriate band gap for
the desired wavelength.

Like the earlier work, this attempt at band-gap engineering started at
the foundation of the contemporary understanding of the material world:
quantum mechanics.  It starts with equations and the little energy
diagrams that serve as the rite of the atomic masons 0 217

architectural floor plans.  These diagrams show rather simply just
where electrons can or cannot go and how hard it is to push them from
one place to another within a complicated semiconductor crystal.  When
he is at work, Capasso probably does not go an hour without scribbling
down an energy diagram into a notebook or blackboard or the back of an
envelope.

For Capasso a picture is worth a thousand equations, and the pictures
seemed to be telling him that a new king of laser, the quantum cascade
laser (QCL), could work.  In conventional semiconductor lasers (like
the ones found in compact disc players) the laser light comes from
excited electrons when they swan dive across the material s band gap
into the valence band, where they recombine with holes.  The energy
difference between the excited and ground state is equivalent to the
energy of the emitted photon, which means that the wavelength of light
is determined by the material s band gap.  Normally, the gap is itself
completely determined by the material s chemical composition.  That is
where the QCL idea differed so radically.  he presumed physics of light
production did not involve holes.  Instead, electrons alone would
produce the light as a by-product of the pathway they would be forced
to take as they traveled through the energy landscape of a precisely
constructed semiconductor crystal.  In the QCL, electrons are
constrained to make transitions between the discrete energy levels of
quantum wells.  So the photons the electrons emit have energies
equivalent to the energy difference between the wells.  The key here is
that the energy difference between wells strongly depends upon how
thick the wells are, a factor that an MBE operator can dial in.

The bottom line of all of this is that the wavelengths emitted by the
electrons therefore can be tuned over a wide range using the same
combinations of materials but with different well thicknesses.  So the
chemical composition of the laser no longer limits the wavelength to a
specific value, as had always been the case.  To build an actual QCL
would amount to an end run around what had been a Berlin Wall in
solid-state laser research: a material s intrinsic band gap no longer
would determine the actual wavelength that would come out of a laser
made of that material.

The blueprint for the quantum cascade laser depicts a
through-the-looking-glass skyscraper-five hundred crystalline stories
tall yet shorter than a mite s eye.  The finished product would be no
more than a few micrometers thick, or about the diameter of a
bacterium.

It was Barbara Sivco who got the task of actually growing the beast.
She had been working for Alfred Cho since 1982.  Before that she had
grown semiconductor crystals, including gallium arsenide, by more
conventional means not so different from growing ice by letting water
freeze.  But in 1994 she got a request that she could hardly believe
from Jerome Faist, a postdoctoral associate from IBM s premier research
facility in Switzerland whom Capasso had recruited.

Like Capasso, Faist was a quantum mechanic who got his Ph.D. in laser
physics and who always wanted to work at the famous Bell Laboratories.
In 1991 Capasso hired him to do the calculations and the hard quantum
mechanical thinking that might lead to the QCL, which at the time was
little more than an energy diagram.  Faist shared an office with Carlo
Sirtori, another quantum mechanics aficionado, and the two worked like
a tag team whose charge was to come up with the detailed blueprints for
the new species of laser.  We started out on the wrong track, Faist
recalls.  Their first idea was to use a standard laser to pump
electrons to the top of the energy landscape within their theoretical
quantum cascade laser; physicists often use one laser to stimulate
another.  But that tack restricted the types of crystal structures that
the researchers could play with.  Even if it worked, it would prove to
be an engineering headache later.  Another laser requires a lot more
space and a rig of carefully focused and aligned mirrors and lenses. 
It would be much more convenient to power a solid-state laser with
electricity than with light from a second laser.  After a year and a
half and little to show for it, not even a paper, Faist was getting a
little nervous.  Like Capasso s, his original contract at Bell was for
a temporary position.

The smell of failure led him to make a fundamental measurement that
would tell him for sure whether the quantum cascade laser was a real
possibility.  He had to measure how long an electron could remain in an
excited state within a pared down predecessor to the QCL.  There was
some danger that the electron lifetime would be too short, Faist
recalls.  In that case the game would be over before it started because
the electron would vent its energy in any number of ways within the
crystal before it spit out any light.

To make the measurement he came up with a relatively simple structure.
It was a series of nine quantum wells that his calculations suggested
should emit dim though measurable light if the lifetimes of the excited
electrons were OK.  At it turns out, his measurement agreed closely
with what earlier theoretical calculations had predicted would be the
lifetime for these excited electrons within a QCL structure.  Faist
recalls, It was an enormous intellectual boost because it gave me
confidence in the predictive power of the theory.

Now he had to grow a structure that actually emitted laser light from
nothing other than electrons clunk-clunk-clunking down an energy
staircase.  As he began to do more detailed micro architectural
designing, Faist quickly realized just how complicated the crystal
skyscraper would have to be.  The first time I showed Federico what I
thought we would have to do, he had a gut feeling that it wouldn t
work, Faist recalls, though Capasso remembers being more upbeat.  Faist
s appointment was scheduled to end soon, and he began to wonder if he,
his wife, and two boys would have to pack up and return to Switzerland
with that disappointing feeling that comes when big investments go
bust.

When he first gave the QCL plans to Sivco in late January 1994, she
almost laughed.  She had grown hundreds and hundreds of crystals using
MBE for a decade, but no one had ever asked her to grow five hundred
layers of anything.  But it took her only an hour to program the
computer and then about three hours to grow the entire microskyseraper.
For a state-of-the-art MBE, five hundred layers is a morning s work.
She scheduled the run for a Friday.

Into the chamber Sivco placed the foundation: a micro slab of indium
phosphide.  Then came the three hours of gentle clicking of
computer-controlled shutters opening and closing.  Hot atoms of
aluminum, indium, arsenic, and gallium streamed from the shutters,
silently building into layer after layer of differently composed
crystal planes.  Each layer had its own specified chemical composition
and width.  Each layer would perform a particular role in a finished
QCL like a gear or a pinion or a gasket in a complicated mechanical
machine.

The atomic masonry of the quantum cascade laser is intricate indeed. In
it is a quantum mechanical staircase of twenty five steps, each one
made of a triplet of quantum wells whose relative proportions of
aluminum, gallium, indium, and arsenic atoms are slightly altered to
make sure that the electrons continue their downward trek.  This is
like steepening the downward slope of an electronic river.  If these
layers actually worked according to the plan, the electrons would emit
light as they tunneled downward from step to step.

In this case the color would be invisible.  The first QCL Capasso s
team designed was for an infrared light source, a fact that would make
the lasers good candidates for technologies that can readily see
through the atmosphere and monitor pollutants or weather conditions.

Even after these twenty-five trios of quantum wells, the blueprint for
the QCL called for another few microns of layer by-layer construction.
Finally, the whole business would get

20-nanometer-thick cap of gallium indium arsenide sprinkled the rite of
the atomic masons N 221

with tin atoms to provide a final conductive pathway for electrons to
reach a metallic overlay.  That overlay is what would allow the
researchers to hook the QCL to the real world.

If all the electrons in a QCL did was hop down an energy stairwell,
few, people would have been so interested.  Like people who hop and rub
their tummies simultaneously, electrons in a QCL are supposed to bound
down stairs while simultaneously emitting photons of specific
wavelengths.  When Faist handed Sivco the plans, no one knew if the new
material would be able to pull off the double stunt.

The fateful Friday arrived.  Sivco prepped the MBE machine and did the
QCL run that had been three years in the making.  What came out was a
brittle little wafer the size of a quarter.  Its shiny, dark,
mirror-smooth surface gave it the look of a single undersized lens for
sunglasses.  (See Figure 34.)

Sivco s MBE performance using Cho s meticulously raised metallic Medusa
is the equivalent of a Yitzhak Perlman rendition of a baroque concerto,
replete with presto bursts packing more perfectly executed notes per
unit time than seems possible.  The slightest irregularities-in the
pumps that maintain the vacuum or in the temperature of the indium
phosphide substrate-the slightest vibrations, or any one of many
unnoticeable disturbances could lead to structural imperfections even
at the three hundred eighty-seventh layer.  The tiniest glitch might
prove fatal.  Not until Faist got his hands on the crystal would anyone
know if the run was a success.

Al Hutchinson, another longtime collaborator of Capasso s who
specializes in prepping samples for experiments, got the wafer from
Sivco and prepared test specimens for Faist.  Executing a number of
jeweler like procedures, he cut tiny stripes from the wafer, etched
away a protective overlay without removing any underlying layers, and
attached metal leads so that experiments could be done.  Faist asked
Hutchinson to prepare only small stripes of material from the wafer
Sivco gave him.  He knew that the larger the stripe, the higher the
likelihood of

222 n S T II F F

FIGURE 34

0015 35KV .  -X33.0 10OPm W037

The Quantum Cascade Layer Gallery.  To the eye the crystal silver
harboring a quantum cascade laser could just as well be a silver of
anything dark and shiny.  Under a scanning electron microscope,
however, the mesa like structure of a QCL overlaid with a gold electric
contact layer can be seen along with a wire leading to a rack of
electronics (above).  The mesa is about 25 microns across, which Is
finer than a human hair.  A much higher-resolution cross-section of a
quantum cascade laser structure (in the mesa) seen in a transmission
electron micrograph reveals a handful of its hundreds of molecular
layers (top right).  Distinct collections of layers in the structure
serve different functions, including accelerating electrons to higher
energies and generating light as electrons race down an energy
staircase, emitting photons all the way.  The energy diagrams depicts
some of the physics going on in these multilayered structures (bottom
right).  In this version of a QCL, which operates at room temperature,
electrons enter a funnel injector that ramps the electron s energy up
so that it enters the top of an energy step.  When the electron then
falls down to the next step, light is emitted.  At the bottom of the
next stair, the electron then enters another injector region that then
places the electron at the top of the next energy step, from which It
falls and emits light-and so on.  By adjusting the widths of the
various layers of the crystal, the researchers can dial in the
wavelength of laser light that comes out as the electron goes down the
energy steps.  BELL LABORATORIES LUCENT TECHNOLOGIES

FIGURE 35

MICROGRAPH OF CROSS-SECTION OF QUANTUM CASCADE LASER

BY TRANSMISSION ELECTRON MICROSCOPY

GaIn As AlInAs

Wells Barriers

(nm) (nm)

0.13 1 ............. ..3.5 3.5 Active 2.8 3-0 Region - 3.0

Electron Injector

C..@y W S.No.  Chu.  AT&T 80 Lab@gorj@

FIGURE 36

"ROOM TEMPERATURE QUANTUM CASCADE LASER

224 a S T V F F

lieht-eatinay defects.  They decided to go for stripes 2 millimeters
long and 10 microns wide, a barely visible splinter.

Faist got the first prepped stripe in the early afternoon.  He put it
into a special chilled chamber that would enable him to cool the
splinter to a few degrees above absolute zero, enabling him to freeze
out atomic vibrations that could divert the electrons from their
intended action.  He connected contacts coming from the frigid crystal
to an electronic box that would allow him to control the voltage and
the current.  A light detector was poised on the other side of the
stripe to detect and measure whatever light might come out.  For a
laser the amount of light coming out should increase as the amount of
electrical current going in increases.  That s just like an ordinary
lamp.  If it is a laser and not just a lamp, however, there will be
some threshold of current at which there is an abrupt increase in the
light output, as though someone suddenly flicked on a floodlight. Faist
had that anxious feeling of being on a starting line waiting for the
sound of the starting gun.

Ready, set, BANG!  Nothing happened.  No matter how far he cranked the
cur rent, no light came out of the stripe.  Faist asked Hutchinson to
get another stripe ready.  It took about an hour.

Ready, set, BANG!  Again, nothing.  In another hour another stripe was
ready for testing.  Again, nothing.  It had been three years since
Faist had begun work on the QCL, and there now was some reason to fear
that he, Capasso, and Sirtori had overlooked some fatal flaw in the
theory. Or maybe something went wrong during the growth of the crystal.
Perhaps they would have to convince Siveo to grow another
five-hundred-layer cake for them.  A hopeful mind blocks such thoughts
and instead conjectures that the problem is simple and fixable.  Faist
decided to try a shorter splinter of the wafer.  That would reduce the
amount of light he could expect, but it also would cut down on the
probability that defects would spoil the day.

Friday had become Friday night.  Faist was thinking about going home to
his wife and two kids.  Then he began cranking up the current on the
small splinter while watching to see if the detector was discerning any
light.  This time he saw a signal!  Light was coming out of that tiny
splinter.  But was it laser light?  At the moment he could only say
that he had made a tiny sophisticated lamp, not a newfangled
solid-state laser.  Then I saw that the light signal went up and up and
up.  I knew then that either the device was burning or lasing, Faist
recounts.  I couldn t believe my eyes.

He was alone in the lab and so couldn t immediately share the elation
of the moment.  So he phoned Capasso with the good news.  Capasso, who
already has enough enthusiasm to perk up a funeral, was gushing.  A
colleague said she imagined Capasso jumping up and down screaming

It s a laser!  like Dr.  Frankenstein.  Fearing that if he blinked the
success would disappear, Faist took data well into the night.

The next day he and Capasso met at the lab and took more data.  There
was no doubt about it, Faist recalls.  The little intuition that became
a detailed set of calculations that became a blueprint for a
five-hundred-layer crystal no thicker than a sperm cell that was made
real by MBE actually confirmed the little intuition-another scientist s
dream come true.  That was a great way to blow the weekend, Faist says.
He, Capasso, and the rest of the team had pulled off an audacious act
of materials design.

A week later Dave Lang, Capasso s immediate supervisor, arranged for
Bill Brinkman and Arno Penzias, the vice president and president of
research at Bell Labs, respectively, to come into the lab and see the
laser in action.  Penzias had won a Nobel Prize for discovering the
everywhere the-same background radiation that cosmologists had
predicted would be present if the universe did in fact begin in a big
bang.  The triumphant band-gap engineers were still shocked enough at
their own success that they considered simply showing data they had
collected earlier rather than flicking the switch again and risking an
embarrassing moment of non lasing  But nothing bad happened.  The
little laser that could lased for the supervisors as it had for Faist
and Capasso.

Brinkman knew that the steep peak on the oscilloscope meant he was
witnessing a laser like no others.  We had talked about making a laser
like this for years, Brinkman notes, recalling earlier discussions with
the researchers.  And there it was.  This was a real tour de force.
This is a step toward the ultimate situation in which you sit at a
computer, design a material, and then let a machine make it for you.

The team wasted little time in shooting a paper about the QCL feat to
Science magazine.  Science, in turn, wasted little time in publishing
it.  The paper arrived on February 14 and appeared in the April 22,
1994, edition.  It is a rare paper that makes it through the gauntlet
of peer review and publication that quickly.

The scope of the list of potential applications increases as

Capasso s team improves the QCL lasers and makes them more convenient
to use.  For one thing, the researchers have gotten these lasers to
work at higher temperatures than in their original experiments.  In
particular, they now can work at temperatures readily maintained with
liquid nitrogen, which anyone can carry around in an insulated
container.  Some designs of the laser even work in sauna temperatures,
though the power output diminishes.  The QCL team also has become adept
at changing the thicknesses of the layers in the active lasing regions
of QCLs and has proved that the resulting wavelengths of light that
come out can be tailored over a wide range, from 3 microns to 13
microns.

This range is suitable for a smorgasbord of potential applications.
High on the list are QCL-based systems for monitoring atmospheric
pollution, since the lasers wavelengths are not absorbed by moisture
and aerosols.  Another potential use is in industrial process control,
since many nasty compounds absorb the rite of the atomic masons v 227

the wavelength of infrared light emitted by the lasers.  There is a
bounty of other QCL applications: central components in systems for
detecting illegal drugs and explosives; tools that diagnose automotive
combustion and emission; collision-avoidance radars in fleets.  of
smart cars ; military countermeasures; medical testing; and basic
research.

The quantum cascade laser is one of the most sophisticated,
future-is-now examples of materials design.  Charles Townes, a Nobel
laureate for the invention of the laser, said of the QCL work that it
opens the door to very important new laser possibilities.  It is an
outgrowth of the postwar ambition to create complicated machines in
single pieces of solid materials by precisely controlling the materials
internal structure.  The ability to design materials in this way is a
harbinger of the new human relationship to matter.

Quantum cascade lasers occupy a leading edge of current technology.
There is an elegance about them, with their pristine crystallinity and
their output of pure laser light.  They are the lacework of the
material world whose mass is measured in tiny fractions of grams.  But
steel, one of the most traditional materials produced globally at the
annual clip of nearly I billion metric tons, also has begun to enter
the growing club of materials that can be designed from atomic scratch.
It seems that steel-the secret of the Samurai, the heart of mythical
Excalibur, the skeleton of urban skyscapes, and the very stuff of the
industrial revolution-has only begun to show its mettle.

SEVEN

COMPOSING Steel

Steel.  Silver-gray, cold, solid, hard, strong.  What else but steel
could have become the metaphor of Superman s superhuman strength? Steel
s strength opened the way for a word like skyscraper.  Without steel
the urban skylines would still look like the pyramid-studded Valley of
Kings in the Nile.  Before steel the only way to build high was to pile
stone upon stone.  For a hundred years architects and engineers have
been relying on skeletons of steel to create the high-rise cityscapes
that are among the most recognized symbols of high-tech modernity. 
Steel s hardness makes it the right stuff for tools that can spin, cut,
and shave at high speeds.  Steel is a vital organ of mass
manufacturing.  It is the stuff of car frames, bulldozer scoops,
appliances, implements of every imaginable kind, and massive caches of
weapons.  Aside from concrete and stone, industry churns out more tons
of steel than any other material.  U.S. steel production in the 1990s
has hovered around 90 million metric tons.

So central has steel become to so many products that uniformity and
reliability have been mantras for those in the steel industry.  Failure
because of flawed materials is simply too expensive.  I-beams for a new
high-rise in San Francisco or

Tokyo must be able to hold the building up even in an earthquake. Steel
used by auto makers had better not cost them customers and embarassing
recall by corroding, cracking, or degrading.  The same goes for
specialty items like the bearing steel used in the fuel turbo pumps of
the space shuttles engines, on which the lives of astronauts and
billion-dollar satellites depend.

The imperative for reliability can make an industry conservative.  Once
a good alloy fits the bill for a specific application, talk of changing
anything becomes heretical.  That conservatism has worked well enough
for erecting cities and furnishing the world with all manner of steel
dependent products.  But in an era like this one, when each new
generation of technologies makes more stringent demands on materials,
the current roster of steel alloys already looks like oldtimers day at
the ballpark.

The wish list for better steel alloys is a long one.  The Navy wants
stronger and tougher steels for the arrestor hooks without which their
jets would be unable to stop as they land on aircraft carriers.  The
automobile industry can always use stronger steels to allow them to
shave fuel-eating pounds from each car.  NASA has called for better
bearing steels for the fuel pumps of its rocket engines.  So vast is
steel s role in modern societies that any significant improvement in
its properties can have revolutionary implications.  If you make an
alloy twice as strong as the best available today, you will need only
half as much metal to build a city (provided the strong alloy is cheap
enough!).  A steel alloy that won t shatter in the chronic presence of
reactive hydrogen atoms will furnish the key to building the future s
fusion-energy reactors.

Ever since steel became a high-volume material for society,
metallurgists have tried to sweat the last drop of technical per co mp
os z ng s the e Im 231

formance out of it.  Mostly, however, new steel alloys have been the
happy and sporadic results of-so what s new?-trial and error. Materials
researchers often refer to this method as the shake and-bake or
heat-and-beat approach to making materials.  Yet there has never been a
science dominated approach to developing steel alloys that has met the
imperious demands of industrial society.

That tried-and-true empiricism, however, has reached a point of
diminishing returns.  The chances of improving steel alloys
significantly by such an approach are growing ever slimmer.  It just
takes too much time, too many trials, too many tests.  Luck is the
limiting factor.  Yet materials researchers know that no one has tapped
steel s ultimate capabilities.

By getting herself born in 1982, Greg Olson s daughter may have helped
to set the history of steel, and of materials development in general,
toward a new destiny.  When she was born, my time line suddenly
expanded much farther into the future, recalls papa Olson, a research
expert in steel metallurgy.  At that time Olson was a materials
scientist in his mid-thirties at the Massachusetts Institute of
Technology, a citadel of materials research.  Before that he had been a
graduate student of Morris Cohen, one of the most revered elders of
materials research.

Olson was trying to establish a track record in the field by uncovering
more details about what makes steel alloys strong and tough.  The
tougher a steel alloy, the more of a beating it can take without
deforming, indenting, or disintegrating.  The stronger a steel alloy,
the more weight it can support.  So if you learn more about what the
iron, carbon, chromium, and other elements in the steel alloy actually
do, you might find keys to making better steels.  Olson was in fact
authoring the kind of papers that would help him move up the academic
ladder from his modest post as a research associate.  Becoming a
father, however, changed his priorities: I suddenly felt that I ought
to be working on things that could make the world better for my
daughter, not just on things that could earn me respect in my academic
world.

What has come of his epiphany is far bigger than what he originally had
in mind.  His long-term goal has become nothing less than to reengineer
the way humanity invents the materials of which society is built-not
just steel, but all materials.  Says Olson, The reality of all of the
materials that we have developed so far is that they have not been
designed.  Trial and error, serendipity, variations of traditional
formulas, with an occasional dash of science-this is how most materials
have been discovered and developed.

The empirical way of making steel is to prepare many alloy compositions
like so many different cake batters.  Each one of these is made by
weighing out stipulated proportions of the alloying elements (80
percent iron, 12 percent chromium to make it stainless, and so on). The
little clinking scraps, spheres, and other shapes with which these
alloying elements start life are placed into a ceramic crucible and
then heated until they melt into the searing molten alloy.  The liquid
metal looks and feels like a miniature sun.  To get near it is to defy
all your survival instincts, which are saying stay away.  The melt, as
metal people refer to the molten brew, is poured into a mold, where it
cools and freezes into the solid alloy.  The cooling may be a simple
affair, or it might involve various tempering or reheating steps.  In a
tempering step, for example, the solid metal might be held for a while
at a temperature of, say, 500 C. That way the metal remains loose
enough inside so that certain internal atomic rearrangements linked to
strength or ductility can occur within the alloy.  A final cooling step
then can lock those new atomic structures in place.

When you set out to make a steel alloy with a particular set of
properties, you don't actually know what you have until you do some
tests.  You need to do some metallography so you can see what internal
structure you actually have.  You need to put a piece of it into a
testing machine that will measure how much tensile force it can handle
before breaking.  You need to put it into another testing machine that
can measure how hard it is.  You might need to cool it down to frigid
temperatures to see how brittle it gets.  If you are systematic enough
about your alloy formulations and property measurements, you may
discern useful correlations between the metal s chemical composition,
the way it is processed, and the physical properties of the resulting
alloys.  If you are really lucky, you will find a winner an affordable,
commercially attractive alloy.

For centuries discovering correlations like those have been the bread
and butter of metallurgy and all materials development.  That is how
early metallurgists discovered bronze five thousand years ago.  That is
how Wallace Carothers discovered nylon in the 1930s.  That is how two
researchers at IBM s research facility in Switzerland discovered
high-temperature superconductors in 1986-all of which proves that
empiricism is not a bad way to go.  Empirically discovered materials
have been plenty good enough for building modern society.

But what was good enough rarely remains so for long.  Steel is a case
in point.  To Olson contemporary steel alloys and most other materials
are highly inefficient.  Their strength, durability, resistance to the
elements, and most other measures of performance fall short of what
steel alloys theoretically are capable of.  To Olson the inefficiency
of materials translates into an across-the-board drag on human
progress.  Bridges collapse, machine bearings wear, tires blow out,
roads crack, metal corrodes.  Engineers have always back-pedaled on
their designs because of the limitations of the available materials.

Olson, perhaps the most avid champion-even philosopher-of materials
design, sees this situation turning completely around.  Engineers no
longer have to be limited by materials that exist, he claims.  They can
design according to the materials they would like to have.  Want to
build a skyscraper 300 stories high?  Get a materials researcher to
invent a steel alloy strong enough to do the job.  Want a semiconductor
laser that emits light at short blue wavelengths so that you can pack
twenty times more information onto a compact disc?  Get a materials
researcher like Federico Capasso to invent a semiconductor structure
that can fill the order.  If we could actually design materials, we
ought to come up with materials that are a whole lot better than any we
had before, Olson says.  If scientists could design and make more
capable materials, the things society makes and needs would go faster,
last longer, pollute less, recycle more completely, store more
information, transmit more data in less time, and generally do more
with fewer shortcomings.

In this framework the feat of materials design that band gap engineers
have been pulling off for years is merely a single embodiment of what
Olson has in mind.  From quantum mechanical theory, band-gap engineers
develop detailed blueprints of new semiconductor crystals.  From the
blueprints crystal growers grow the designed crystals.  And quite often
the crystals work just as predicted.  This is scientific materials
design.

Olson aims to prove this point with steel.  Compared to steel, he says,
No material is in a better position to be designed from scientific
knowledge, and no material is in more need of this approach.  But there
are good reasons why semiconductor researchers have succeeded in
designing crystals well before steel researchers have demonstrated the
same capability with metal.  After all, even the quantum cascade laser
(QCL), in all its five-hundred-layer glory, is a simple material
compared to steel.  The QCL is a member of the most idealized club of
all solid materials: it is a crystal.  Its complexity comes from the
fact that its many layers have different chemical compositions.  Unlike
a diamond crystal, which is everywhere the same, the structure of a QCL
crystal is like a stack of blankets of different thicknesses and
colors.

A steel alloy is another animal.  It is more like a compacted load of
earth with sand particles, pebbles, larger stones, occasional blades of
grass, roots, and God knows what else all jumbled together.  The
performance of each piece of steel depends not only on how its atoms
are arranged into tiny crystal g-rains of different shapes and sizes,
but also on how these accrete, bond, or jam together into ever larger
structures that you could hang a concrete facade onto or make a
high-speed cutting tool out of.  It depends upon different structural
rearrangements and transformations that can occur on the different-size
scales.  It depends upon what little films of minerals and other
chemical structures end up on and between the grains.  You cannot make
steel with a molecular beam epitaxy machine.

Can steel be designed with the same finesse as semiconductor crystals?
Olson has been betting yes for more than a decade.  Steel is probably
unique in the extent to which we know the function of all its
structural levels.  We can have a list of quantitative
structure-property relations [such as the relationship between grain
size and strength], he explains.  This means we have the tools to
design steel alloys.

There are a few hurdles, of course.  There are so many relevant
structure-property relations to consider at one time that the
complexity of the problem has made quitters of previous would-be steel
designers.  Each piece of metal comes from a metallurgical drama that
plays out as an alloy s atomic ingredients are prepared, mixed, made
molten, and then forced to react, coalesce, segregate, and otherwise
interact as they take their final place in the finished alloy.  So much
physics and chemistry go into making steel that no one totally
understands it.  The chemical and physical forces at work invariably
pull the action simultaneously in different directions, forcing the
metallurgist to become a master of trade-offs.  If you add one
ingredient to make the alloy stronger, it becomes more prone to
cracking than before.  That s bad news if you are trying to make a
better metal for the turbine blades inside a jet engine.  Adding a
different ingredient to reduce the alloy s brittleness may result in
greater proneness to corrosion in the rain.  Such an alloy is useless
for anything that needs to serve on, in, or near the ocean.  Finally,
practical metallurgists know that the costs of metal-making and
finished products must hew to commercial limits.  You can make the
world s most wonderful materials, but if no one can afford the stuff,
they are as practical as the Hope Diamond.

Undeterred by these formidable obstacles, Olson was determined to forge
a new approach, a new understanding.  It would take a first-person leap
of imagination: a steel designer would have to become steel the way
Star Trek s Mr.  Spock almost becomes the person with whom he
mind-melds.  To hear Olson speak about steel to his expert colleagues
is to witness a man who has mind-melded with metal.  (See Figure 37.)

Olson s ambition to learn how to design steel is more than personal: he
wants to begin a movement that will change the overall practice of
materials research.  What better way to begin a sustainable movement
than to indoctrinate your students?  That is why Olson s students train
from day one to become materials designers in the only way possible.
They can t learn this approach from books because such books haven t
been written yet.  So Olson s students, along with Olson himself, learn
to become materials designers by being materials designers.

Olson began laying down the foundations for this philosophy of
materials development in 1985 by founding the Steel Research Group
(SRG) at M.I.T. with his mentor, Morris Cohen.  Cohen, along with the
likes of Morris Fine, was among the cadre of researchers who first
defined the field of materials science and

Anatomy of Steel.  Despite Its uniform outward appearance, a piece of
steel has a highly differentiated Internal anatomy.  The micrograph
above is of a metallographic preparation of a piece of martensitic
steel.  Martensite refers to a particularly hard crystalline
arrangement of Iron and carbon atoms that can form when hot steel is
quenched, or rapidly cooled.  The martensite phase shows up here as the
darker regions within a lighter-appearing matrix In which the steel has
not converted Into martensite.

GREG OLSON, ASM INTERNATIONAL

composing steel engineering.  Members of the SRG included researchers
from academic, government, and industry laboratories.  The National
Science Foundation provided the bulk of the initial funding.
Ironically, the steel industry at the time was in such a trough that
some of the SRG s original industry partners would go belly up the very
next year.  Then some of the smaller, more innovative steel companies
such as Carpenter Steel came on board.  Each year overall membership
has shifted.  But the SRG typically has involved several dozen
researchers.  This sounds like a big operation, but like most materials
research it really

FIGURE 37

has been small science.  Since its inception the SRG has spent about
$10 million from government and industry sources.

The SRG seeks to develop a strategy to design steel alloys to order. If
they could succeed with steel, they would try to apply the new design
paradigm to other categories of materials.  In time, Olson hoped,
materials researchers would design all kinds of metals, polymers, and
even concrete (actually one of the more complex materials on earth)
with capabilities surpassing anything resulting from empirical
discovery alone.

First things first.  The initial technical goal of the SRG was to
create steel alloys that are significantly stronger, tougher, and more
corrosion-resistant than any of the empirically discovered alloys
already out there: less steel would be needed in traditional
applications like manufacturing cars; better steels also would increase
the capability, reliability, and safety of spacecraft, aircraft, ocean
craft, and other technologies that use expensive, state-of-the-art
steel.

Consider a single gain that NASA wanted so much that it began pouring
money into the SRG soon after it formed.  This saga began in earnest in
1986, when the price of material failure became graphically ingrained
in contemporary culture.  Over an dover again on TV screens across the
world, the space shuttle Challenger exploded seventy-two seconds into
its ascent.  The culprit was the elastomeric rubber like O-rings
between segments of an external fuel tank.  During the especially cold
launch on that fateful day, the O-ring had gotten too stiff and brittle
to handle the tremendous stresses of takeoff.  Hot combustion gases
blew by the rings and ignited the faelloaded tank.  What many people in
the technical community in and out of NASA knew was that the rubber
O-rings just happened to be one of many potential accidents waiting to
happen.

Before the disaster, just as the SRG was getting off of the ground, a
friend of Olson s at Rockwell International, a high technology firm
with a long history in developing aerospace materials and systems,
suggested a perfect challenge for the would-be steel designer.  There
was another material-based time bomb in the shuttle fleet, a steel
based time bomb. In the heart of the main engines of the spacecraft are
powerful fuel turbo pumps that inject frigid liquid oxygen into the
engines, where the oxygen mixes with the fuel to enable combustion to
occur.  In those turbo pumps are bearings made of a steel alloy known
as 440C.  Engineers chose the alloy for its combination of strength,
toughness, corrosion-resistance, and imperviousness to damage from the
high levels of reactive hydrogen atoms in that environment.

It is a hostile environment at that.  The bearings must support shafts
that spin around thousands of times each minute at cryogenic
temperatures below300 F. All the while the assembly must pump extremely
corrosive liquid oxygen into an adjacent engine that generates 30,000
horsepower in an area that could fit under an automobile s hood.

If these bearings fail, so does the pump.  If the pump fails, the
engines become a bomb.  Even the 440C steel itself would become a fuel.
Everything would ignite.  NASA engineers were concerned about the
potential.  But they apparently had found ways of modifying the alloys
so that the bearing would pass during ground tests of the engines.
Safety entailed frequent inspections and replacements of the parts.

That is why it was not a happy day when NASA engineers discovered that
a shuttle had returned with a cracked steel component in the engine.
Still, with no replacement material available to solve the problem, the
only solution was to inspect and replace the bearings frequently.  NASA
materials engineer Steve Gentz told a Chicago Tribune reporter in 1989
that these inspections had become a significant source of cost as well
as delay between shuttle missions.  A new and more reliable bearing
steel would make a lot of people sleep better.  To Olson it was a
perfect target.  And with new funding from NASA, it became the SRG s
largest project.

Meanwhile, the SRG was running headlong into some daunting obstacles of
the attitudinal, sociological, and cultural sort.  Designing materials
involves such a breadth of techniques, theories, practices, and
mind-sets that it bridges the yawning gulf between science and
engineering.  Olson and his colleagues even have a cute neologism for
the pursuit: scieneering.  To design materials, therefore, meant that a
lot of people had to get together to work as a team.  Moreover,
producing a new breed of scieneers entailed pushing a new kind of
engineering curriculum.

Inspiring teamwork and forging new curricula have never been small
feats at universities.  T)rue, materials research already was born on a
bridge between science and engineering, and M.I.T., Olson s academic
home when the SRG first formed, was in principle receptive to the
linkage.  Still, Olson recalls encountering stiff opposition.  For one
thing, consider the likelihood of convincing a bunch of upwardly mobile
researchers at a high-profile place like M.I.T. to become mere fish in
a larger pond such as the SRG.  Besides, Olson recalls, steel was kind
of iout as a research topic in universities.  Most materials
researchers in the academies had their sights on newer, more exotic
materials such as light-emitting semiconductors and electrically
conductive polymers.

Northwestern University was an exception.  At the time it hosted the
industry-sponsored Steel Resource Center, where steel companies and
researchers could go in search of answers to broad technology and
management problems.  So here was a university with a tradition of
interdisciplinary teamwork in materials research and a strong
commitment to steel.  It was perfect for what Olson had in mind.

Olson, too, seemed perfect for the visions of Stephen Carr, the head of
the university s Materials Research Center (MRC).  Carr already had
been convincing himself for years that materials design would be the
most exciting direction in materials research.  Olson s steel lab was
just the right addition to the MRC to put Northwestern at the
forefront.  Olson and Carr composing steel N 241

talked about their common visions.  In 1988 the flirtation became a
marriage: Olson transplanted himself and about twenty members of his
M.I.T. lab to Northwestern s Materials Research Center.

With backing from his new academic home, Olson now had to confront an
educational divide that transcended locality.  Standard science and
engineering curricula train students to analyze phenomena, to reduce
things to their basic parts: people become organs; organs become cells;
cells become molecules.  Often lost in this kind of reductionism is a
feel for the myriad ways in which the same parts can come together and
produce unexpected results.

For example, how in the world does the property of stainless ness
(resistance to rusting and corrosion) emerge from iron, carbon, and
chromium atoms?  Analytical minds are not fully equipped to excel in
this more synthetic, holistic aspect of materials research.  It takes
more than analytical precision and mathematical acumen to get a feel
for emergent properties that are more than a mere sum of parts.  Worse
than that, Olson believed that the traditional focus on analytical
thinking literally had been disabling generations of students, stunting
the kind of creative thinking and research the field cried out for.

Music composition, it seemed to Olson, was the model that could change
the situation.  When Olson was a graduate student at M.I.T."  he
supplemented his technical program and course work with music
composition.  Composing music is a procedure of synthesis, not
analysis.  To compose music is to put together different instruments,
sounds, tones, silences, and rhythms into a whole that soothes,
agitates, uplifts, saddens, pleases, or otherwise affects the listener.
The composer must orchestrate the composition s progress from the
beginning to the end of the piece.  All the while the composition s
properties-its flow, style, emotive effects-must merge into a singular
performance.

To Olson musical composition holds more than just a metaphoric
resemblance to designing materials.  Musical composition and material
design of the highest creative order partake of similar cognitive
machinery.  Both require a thorough knowledge of the parts involved and
an intuitive grasp of how those materials come together into ever
larger and more complex structures; both demand a sense of how the
parts of a composition can merge into something greater than a sum of
those parts.  Olson s ideal materials designer, then, is an engineer
with the training of a scientist and the soul of a musician.

This musical model of materials design shows up in what Stephen Carr
affectionately refers to as Olson Diagrams (an allusion to the famed
Feynman Diagrams that the late Richard Feynman developed to depict
interactions among elementary particles).  As Olson sees it, his
diagrams are to materials designers what arrangements are to jazz
composers and musicians.  At the top of a typical Olson Diagram for
steel are three headings with a handful of items underneath each: from
left to right they are Processing, Structure, and Properties.  Lines
radiate from the specific items under each heading to specific items
under the adjacent heading.  These lines show how specific processing
steps lead to specific material structures and how specific material
structures relate to the material s physical properties.  (See Figures
38 and 39.)

Take the Olson Diagram for nickel-cobalt steel, the kind of superalloy
used to make the highest strength aircraft landing gear.  Under the
Processing heading is a set of boxes that outline the way molten metal
becomes a solid piece of an alloy: refine the raw molten steel; remove
oxygen from the melt;

Olson Diagrams.  The diagram for an ultrahigh-strength steel alloy
shows how the technologically important properties of materials depend
upon the complex internal structure that emerges as the metal undergoes
processing.  The generality of this conceptual framework for thinking
about materials is made apparent by the second diagram, which depicts
how the important properties of ice cream depend upon specific types of
Internal structures, which again depend upon the way Ingredients are
processed.  GREG OLSON

composing steel

PROCESSM

FIGURE 38

STRUCTURE

PROPERTIES

Lath M its Ni: C;le= Resistance 00: SRO Recovery Resistance b'V)C.  j
Avoid F03C, IV16G, M23CS

GRAIN REFINING DISPERSION

6 1

Microvoid Nucleation Resistance

AUSTE E DISPERSION Stability (Size, Comp.) Amount Dilatation

GRAIN BOUNDARY CHEMISTRY Cohesion Enhancement Impurity Gettering

PROCESSING

STORAGE DISTRIBUTION

-20F

HARDENING

-20F

EXTRUSION & PACKAGING

STIR FREEZING (Rhoocasting)

21F

COOU14G & AGING

40F, >4h

HOMOGENIZATION

1500, 500 psi

PASTEURIZATION

175F, 25s

MIX PREPARATION

FIGURE 39

STRUCTURE

PROPERTIES

solidify; temper the solid metal.  Each process step changes the
internal structure of the metal.  And each structural change increases
or decreases the alloy s strength, toughness, corrosion-resistance, or
form ability  If the molten metal cools at a faster rate, the resulting
grains are smaller, which means the metal is tougher.  Reheating the
cooled metal yields more of a strength-related crystal structure in the
grains.  Some processing details alter the cohesion between grains.
Others affect the types, sizes, and distributions of iron carbides and
other metal carbide particles, which are at the heart of steel s
strength.  Still others determine how well the metal will resist the
embrittling influx of hydrogen atoms into the grains.  This last is
especially important for any steel alloy slated for duty in or near a
rocket engine.

One of the first challenges in designing a new bearing alloy for the
shuttle engine was to develop the relevant Olson Diagram.  The
researchers had a few constraints.  They knew they would have to stick
with a steel alloy akin to the high-chromium stainless-steel alloy
already used for this shuttle application.  The most general
engineering approach would have been to consider all kinds of materials
as candidates: non steel metals, ceramics, polymers, and composites
made by combining different classes of materials.  Considering anything
other than stainless steel for the shuttle engine, however, would have
entailed a host of design novelties and uncertainties.  The novel
approach the team ultimately adopted was the conceptual fusion of a
chromium stainless steel with a high-toughness nickel-cobalt steel for
landing gear.

For the kind of ultrahigh-strength steel needed for a better bearing,
the researchers knew they needed an alloy noticeably tougher than the
440C alloy yet uncompromised in other critical properties such as
resistance to wear and corrosion.

Once the performance of the future alloy was stipulated, the materials
designers began to work backward from the venerable empirical route.
From the performance they could infer which physical properties the
steel must have.  From the properties they could infer the kinds of
internal architecture they had to realize.  And from the architecture
they could infer the details of chemical composition and processing
that might actually create the structure.

Where there is inference, there also is a lot of theorizing and
modeling.  Where there is theorizing and modeling, there is a lot of
calculating, too.  That means computers.  Materials design is in part a
child of microprocessors capable of making quantitative sense of
complicated theories.  One of the most pivotal tools for the SRG goes
by the name ThermoCaICTM.  It is a database and software system for
calculating how an alloy s elements move, align, combine, interact, and
avoid one another as their context changes from liquid metal through
more viscous stages until they lock into place to form a solid piece of
steel.  When integrated with appropriate metallurgical models, it can
calculate the likely size of grains, carbides, and other metallic and
ceramic phases that form in alloys involving up to fourteen different
atomic ingredients.  It can calculate how different cooling rates
affect the outcome.  It was the first time we had a tool like that,
Olson recalls thinking in 1985 when ThermoCalCTM first became available
from the researchers in Stockholm who developed it.  It was one of the
first things that made it possible to think about designing
materials.

ThermoCaICTM itself depended heavily on decades of empirical
observations and thermodynamic data gathered by thousands of
researchers around the world.  From the number crunching with
ThermoCaICTm and a battery of other computational tools emerge
candidate alloy compositions along with a blueprint of the micro
architecture that the alloy ingredients must assume.  The new bearing
alloy candidate that ultimately emerged consisted of seven components:
iron, carbon, nickel,

cobalt, molybdenum, chromium, and vanadium, all interacting to maximize
the alloy s strength, toughness, and hydrogen-resistance.  In shorthand
the alloy is written in weight percentage this way:

55.95Fe-22.6Co-12.OCr-8.6-Ni-0.3Mo-0.25V-0.30C.

Now for a sampling of the details.  As always, iron and carbon form the
basic steel alloy.  The SRG arrived at the specific amount of carbon,
0.3 percent by weight, by carefully examining experimental data that
relates a steel s hardness to the distribution of carbide particles in
the steel.  Any more carbon and the steel would be too brittle.  Any
less would deprive the steel of the hardness necessary to handle the me
talon-metal wear inside the shuttle engine.

The 12-percent chromium component of the alloy makes the steel
stainless by forming an invisible film of chromium oxide on the metal s
surface, preventing oxygen atoms from finding iron atoms at the surface
to form iron oxide, otherwise known as rust.  Even when the surface
oxide scratches away, there is enough chromium for a new oxide surface
to form immediately.  With less chromium, the stainless property would
degrade; with more, strength and toughness would suffer.

Nickel and cobalt-primary ingredients in advanced aircraft landing-gear
steels-are necessary to promote both strength and toughness.  Precisely
how much of these alloying ingredients to include was the question. For
one thing, these elements control the temperature at which an important
structural transformation occurs in the steel during cooling. But the
researchers wanted to keep this temperature high enough to achieve a
specific micro structure related to strength within the alloy s grains.
Calculations with ThermoCaleTM yielded a small range of nickel and
cobalt compositions that would be suitable.  Only one of those
compositions-22.6 percent cobalt and 8.6 percent nickel, it turns
out-would also allow for a desirable toughening mechanism to occur when
the solid metal is tempered at about 500 C. Meanwhile, the new alloy
had to have ultrahigh strength, more than would come from the iron,
cobalt, nickel, and chromium carbides.  From experience and previous
experiments, the SRG knew that molybdenum and vanadium should be able
to do the job by joining with carbon in specific arrangements to form
complex chromium-molybdenum-vanadium car bides.  The trick, however,
was to find the compositions that would lead to the formation of many
more small carbide particles rather than fewer larger ones. The smaller
particles lead to stronger alloys; the larger ones to more brittle
alloys.  The difficulty here is that different amounts of vanadium and
molybdenum have a characteristic driving force for producing the small
carbides at the expense of the larger ones.  Here, the team relied on
both experimental data and thermodynamic calculations to come up with
the specific amounts of the two alloying agents: .3 percent molybdenum
and .25 percent vanadium.

So there it finally was:

55.95Fe-22.6Co-12.OCr-8.6-Ni-0.3Mo4.25V-0.30C.

It was perhaps the first formula of a designed steel alloy.  But was it
in fact any better than the alloy it was designed to beat?  This is
where SRG corporate members like Carpenter Steel come in.  Their
metallurgists can fill the metallic prescriptions that Olson and his
students come up with.  Into a small test crucible went the right
proportions of alloying elements to make the smallest sample they
make-17 pounds.  When the SRG team tested samples from the resulting
17-pound ingot, they were heartened by the results.  The alloy was
indeed twice as tough as the bearing steel NASA had been using.  They
had, in fact, designed a steel alloy with at least some of the right
stuff for the bearing application.

It was a moment of private triumph, but the result had more potential
than immediate value; the prototype material could not be processed in
ways suitable for many applications.  Because of a lower than expected
structural transformation temperature, the alloy required a processing
step at very low temperatures to achieve the desired internal micro
structure  Although bearings still could be fabricated this way, the
processing of alloy compositions would be more difficult and costly
than that of current bearing steels.  That partly explains why the
particular seven-component alloy that Olson, his students, and his SRG
colleagues had designed and that Carpenter Steel had made never made it
into a space shuttle pump.

It didn t make it into the pumps for another reason.  Before the SRG
could try to design a more process able composition, NASA funding for
the project disappeared suddenly as a result of the agency s wrenching
post-Challenger reorganization.  Still, in 1995, NASA did end up
patenting the alloy that the SRG designed.  NASA noted in a technical
brief on the alloy that it represents a revolutionary new breed of
scientifically designed materials.

In the late 1980s Olson began choosing more generic design targets.
Rather than trying to design an alloy whose application would be
limited to specialty applications, like fuel pumps for spacecraft,
Olson decided to aim at alloys that would be more widely used.  It was
getting back to the basics.  The huge technological and economic
potential of such an advance is self-evident in this country, which
produces almost 100 million tons of steel each year.

One project emerged from a continuing collaboration Olson has with the
Newman-Haas Indy car racing team-an outgrowth of his involvement in
amateur sports-car racing.  The chief engineer of the team, Brian
Lyles, wanted to reduce the weight of the steel gears in the cars drive
trains.  Gear steels are harder on the outside so that their teeth won
t wear down and less hard inside so that the gears won t crack from
fatigue.  The gear steels everybody was using had been composing steel
m 249

around since World War II.  Some modest improvement in the performance
of these alloys had been achieved by reducing the amount of
contaminants in the alloys.  But there had been no dramatic
improvements.

In 1991 Olson adopted the goal of improved gear steels as a project for
his materials design class.  Racing cars are not the only place where
improved gears would do some good.  For every pound shaved off of a
gear that goes into a commercial airliner, for example, the economic
payoff would be about eighty dollars.  There are about two tons of
gears on a Boeing 747.  Instead of flying an extra ton of gear steel
all over the place, you could take an extra ten paying passengers.
Similar logic suggested that military aircraft made with the steel
could carry higher payloads or fly farther.

The kernel of the challenge was to make a steel that is harder in the
surface (case-hardened), where the teeth of the gear must withstand a
lot of metal-on-metal contact, and more pliant internally so that the
gear will be able to take the external stresses and internal strains
that come with their use.  By hitting a hardness level known as
Rockwell 70 (named after the test protocol that has become the
standard), the team believed it also would increase the alloy s
resistance to fatigue.  If they could realize these combined
improvements, they stood a good chance of cutting the gear weight in
half.  Creating the hardness gradient entails creating a corresponding
gradient in the amount of carbon in the alloy.  There should be less
carbon inside and more on the outside, by the gear s teeth.

With an Olson Diagram, a century s worth of data, physical modeling and
theorizing, and patient hours in front of computers, the design class
came up with a design.  The strategy was to increase the hardness of
both the case and the core, but without also making the core more
brittle and susceptible to corrosion under stressful conditions.  As in
existing gear alloys, the new alloys would require a carburizing step
to inject extra carbon into the surface of the metal.  Carpenter did a
17-pound melt.  Again, the result was close to the target.  The surface
of the alloy was just shy of the hardness level needed, while the
inside was a bit too hard.

Still, the exercise confirmed Olson s faith that designing steel with
stipulated properties was becoming possible.  In 1996 the SRG sent off
a second prototype with improved properties to the Newman-Haas racing
team for testing.  As often happens in materials development, the work
may prove useful in unintended ways.  After the student project won a
national materials design competition, the properties of the alloy
caught the attention of a Navy research program manager who hopes the
SRG will be able to design stronger and tougher steels for the
components in the landing gear of their Aircraft.

The most futuristic project was inspired by the film Terminator 2:
Judgment Day, in which a future fascistic regime sends a robotic killer
back in time to murder a boy who will grow up to become a military
leader for the opposition.  The robot is made of a bio mimetic
self-healing metallic alloy that can return to its former shape even
after being blown to bits.  After seeing the movie and considering its
potential to inspire students, Olson asked himself a question: Could we
actually make a self healing material?  The challenge went over well in
his design class.

One of the first leads for a design came from the same abalone shells
that had inspired Mehmet Sarikaya of the University of Washington and
others of the bio mimetic persuasion.  The shell s brittle ceramic is
reinforced by a relatively tiny amount of a rubbery protein and
carbohydrate based polymer.  If a crack begins to run through the
ceramic, the rubbery reinforcement bridges the crack and helps resist
further separation.

Olson wondered whether a similar strategy could work for a class of
metals known as high-temperature intermetallic super alloys  Even
though these metals work exceptionally well at the high temperatures
found in places like jet engines, they have a tendency to be brittle at
lower temperatures.  Engineers had told him that they would like to
find a way of reducing that vulnerability.  How could the abalone like
toughening mechanism be realized in a high-temperature intermetallic
alloy?

One possible key resides in a remarkable class of metals called
shape-memory metals-that can be severely bent out of shape but then
returned to their original form by applying heat.  It is a favorite
class of materials of the smart materials tribe.  These metals have
been used to flip up headlights on sports cars and as mechanical
elements in torpedoes.  In a metallic setting a shape memory metal
could behave like a high-strength rubber cement analogous to the
protein-carbohydrate glue of the abalone shell.

Let s say that at low temperatures overloading the metal causes
microscopic cracks to form in an intermetallic alloy containing a
reinforcing distribution of shape-memory metal.  At the low temperature
the shape-memory component can stretch to bridge the cracks just like
rubbery reinforcement in the seashell.  But now, with the alloy heated
up, the shape-memory components would reverse their deformation,
causing the cracks not merely to get no worse but to actually close.
Moreover, by designing the shape-memory alloy so that it actually
creeps less under stress than the surrounding superalloy, the
reinforcement could provide enough clamping to allow rewelding of the
cracks at high temperature, thus completely healing the internal
cracks.

One of the most basic requirements for the design is to identify a
superalloy composition and a shape-memory alloy composition that don't
eat each other up.  The challenge is sort of like putting hot milk and
chocolate syrup together and hoping that the syrupy quality of
chocolate doesn t completely give way to a pure fluid.  The initial
vision is to start with titanium aluminide intermetallic alloys whose
combination of strength and low density (lightness) have made them
important for military aerospace applications.  To achieve the self
healing feature, the SRG proposes to reinforce the matrix metal with a
titanium-nickel shape-memory alloy.  The hard part is finding a way of
getting the right chemical compatibility of the superalloy and the
shape memory phases to avoid combining ingredients completely into
chocolate milk.

The research required in this project runs the gamut.  It includes
fundamental quantum mechanical calculations for determining the factors
favoring stability of the shape-memory alloy.  It includes higher-level
mechanical calculations to reveal the various interactions that will
affect the toughening behavior of the shape-memory alloy when cracks
form in the titanium aluminide matrix.  It also includes some
experimental determinations of how alloying agents such as palladium,
hafnium, and zirconium might yield an alloy in which the self-healing
behavior will be present even at higher temperatures.  That would be
helpful for the hot venues such as jet engines in which these alloys
might be used.

Olson is the first to admit that any aerospace engineer who depended on
an engine part made of an alloy designed in this clean, in-silico
manner without testing the prototypes is the kind of person willing to
bungee-jump without testing the cord.  There will never be substitutes
for making prototypes of the designed alloys, analyzing their
structure, measuring their properties, and monitoring their performance
in their intended applications.  That is why Pratt & Whitney, one of
the world s major jet-engine manufacturers, is part of the project. The
company will make the prototypes.

Since Olson s team is sending designed alloys to Pratt & Whitney, the
expensive and time consuming process of testing prototypes should be
considerably shorter than usual.  Rather than working up the twenty or
so prototypes that come out of a standard empirical approach to
developing a new alloy, the materials design approach ought to conclude
after three prototypes.  This kind of streamlining, Olson hopes, will
reopen imaginations of materials researchers that had long been closed
by fears of development costs.

Metal alloys are not the only kind of material that Olson,

composing steel a 253

Carr, and others have their sights on for this kind of R&D strategy.
Although they see materials design as applicable to nearly all
materials, it might well be best suited to materials such as steel that
already have a long research history and thus an advanced knowledge
base on which to build.  But at Northwestern there are Olson Diagrams
for advanced cements and concretes called polymer-reinforced hydrate
ceramic composites ; case-hard enable rubber-toughened polymers that
could yield sophisticated polymer structures capable of serving as
gears (for such high-volume items as windshield-washer motors and
automatic car-window motors) and other products traditionally
monopolized by metals.

Olson even envisions superalloy ice cream, a material with a potential
annual market of $9 billion.  The Olson Diagram for ice cream looks
like all of the others, except that some of the headings are more
familiar-mix preparation, homogenization, flavor, cooling sensation,
texture, etc.  Just as in any other material, the properties of ice
cream are determined by its internal structure.  For ice cream that
translates into details such as the size and distribution of ice
crystals, fat globules, and air cells (voids) and the way sugars,
salts, fat, and protein have dissolved and distributed in the
milk-cream matrix.  From the perspective of metallurgical and materials
science, ice cream is just like a superalloy.  Because of dietary and
public health trends, food makers have been pressured to remove
traditional food constituents such as fats whose material properties
partly have determined the evolution of processing equipment and
protocols.  Mass producing acceptable fat-free food takes the skills of
materials scientists.  It stands to reason that materials researchers
will probably have an increasingly important role in food industries.
This is a glimpse of the future of materials design, Olson says.  He
says this of the kind of steel alloy design that the SRG group has been
pioneering as well as of Capasso s MBE-grown crystals.  Steel is one of
the most brutish materials in use; band gap engineered crystals are
among the most delicate.  Together, they stand like bookends for the
entire materials spectrum.  So it is reasonable to think that materials
researchers will eventually apply their talents to all the types of
materials in between.

The generic ability to design materials to order would be a
world-historical technological advance.  The ability to design better
materials is the culmination of centuries of empiricism, analysis,
synthesis, and theorizing about materials.  Better materials mean
better technology.

Better materials can also provide the fixes for many of the
technological messes that have emerged from previous, less informed
generations.  But if the current generation of research and development
succumbs to a similar shortsightedness new materials could harbor
dangers more serious than any we face today.  Lead-containing salts
made paint nice and white, but they ultimately became a huge public
health problem.  Chlorofluorearbons (CFC), now pegged as the murderous
chemicals that have been destroying stratospheric ozone molecules, were
deemed benign wonder chemicals when they first became available in the
1930s.  The good news is that materials researchers have learned from
these experiences, and they are far better prepared to avoid repeating
these mistakes.  To be able to design materials from atomic scratch is
to be able to dream up new stuff and to make those dreams come true. It
deserves to stand among the few major turning points in the history of
human control over the material world.  To stone flaking, harnessing
fire, the rise of science, and the emergence of materials science and
engineering, add materials design as the latest and perhaps ultimate
turning point.

Although the ability to design materials is new, Olson knows that it
already has had some effect on his daughter.  Taped on a window in his
office at Northwestern University is one of her first grade quiz
papers. One of the questions is Name two machines.  Her answers: car
and atom probe.  An atom probe is a powerful analytical device that her
father has composing steel 0 256

used to reveal a material s two-dimensional or three-dimensional
anatomy with atomic resolution.  Now the task is to push the materials
design paradigm along so that the periodic table will yield more of its
latent potential for addressing real societal problems and ambitions.
Then a father will have fulfilled his promise to make the world a
better place for his daughter.

EIGHT

The STUFF THAT DREAMS are MADE of

When framed inside a book chapter, Greg Olson and Federico Capasso
might stand out as rare virtuosi among a population of colorless
journeymen.  In fact, they are standouts only for the moment.  They are
like scouts of more dexterous next-link hominids amidst the first
stone-flaking progenitors.  They are representatives of what is to
come.

But those two, because of their respective specialties-solid state
semiconductors and steel-bracket the entire materials spectrum. Capasso
s multilayered crystal structures-grown atomic layer by atomic layer
via molecular beam epitaxy and weighing all but a tiny fraction of a
gram-are among the finest constructions possible in the materials
spectrum.  Olson s steel alloys, though designed in part with atomic
detail, represent the other end of the spectrum, where the
sophisticated, pre designed micro structures of these alloys are
realized by clever control of the micro structures in multi pound or
multi ton batches of alloys.

Moreover, the know-how, methods of synthesis and processing, theories,
and computational and analytical tools on which the successes of these
materials designers depend are improving at an ever accelerating rate.
It is inevitable that the materials design approach that Capasso and
Olson have vindicated in their work will sweep the rest of the
materials spectrum, including polymers, ceramics, composites, and ice
cream.

If you talk to enough materials researchers, read enough of their
papers, hear enough of their presentations, and visit enough of their
laboratories, a surprisingly coherent message emerges from what
otherwise appears to be a community as crowded and cacophonous as a
kitchen s miscellany drawer: the materials research community has
reached a new plateau of unprecedented power in its ability to
understand, control, and manipulate the material world.

Do you need a lightweight alloy that remains strong at the searing
temperatures of a scramjet operating at many times the speed of sound
and able to withstand the frictional offensives of atmospheric
molecules scouring the material like supersonic sandblasting?  Do you
need a polymeric material compatible with brain tissue that in the
presence of specified levels of a specific neurotransmitter will
release pharmaceutical agents previously loaded into the polymer?  How
about concrete ten or twenty times stronger than any available now?
Call in your local materials designer.

Mark Eberhart, a materials researcher at the Colorado School of Mines,
where most work focuses on ways of harvesting the rawest of ingredients
from the world, characterizes the emergence of materials design this
way: I consider what is going on in materials by design as comparable
in scope and importance to the beginnings of the scientific method. Now
we are at a point where we can do computer simulations and use these as
hypotheses in materials development.  This virtual experimentation
saves the time and money once spent on dabbling with physical
prototypes.  This is akin to the evolutionary the stuff that dreams are
made of 0 259

leap in which brains simulated the consequences of potential actions
before hazarding the action itself.  thinking ahead.

The materials design paradigm is permeating the materials science
community, Eberhart observes.  It is simply the way in which we are
training students.  With computers everywhere getting more powerful at
doing the complex calculations required by materials science, the new
ethic seems destined to proliferate.  Materials designers equipped with
these powerful computers are learning to integrate the mountains of
previous theoretical and experimental work with their experience-honed
intuition.

Eberhart got his own start in materials research when he was building
kayaks in the 1970s with DuPont s then new super polymer Kevlar, which
had a tendency to peel away from the kayak s inner foam forms.  When I
called the DuPont guys and asked why the stuff was failing, they didn t
know, Eberhart recalls.  He decided that materials researchers ought to
be able to answer questions like that and that he would become such a
person.

Now, fully smitten by the simulation and materials design bugs,
Eberhart takes on metallurgy problems that only computational
approaches have much hope of tackling.  For him, one particularly
challenging problem is designing titanium aluminide intermetallic
alloys (the same type of metal that the SRG is considering for
self-healing metals) in the on-again, off again (now very off) National
Aerospace Plane (NASP).  Designers of the NASP intended their aerospace
craft to get into and back from orbit as easily and readily as a plane
gets into and out of the jet stream.  That would require materials
capable of withstanding the heat and friction of multiple trans
atmospheric passages.  Yet these same materials also would have to be
as lightweight as possible since every extra pound of the vehicle
represents an enormous cost in terms of money and/or reduced payload.
The problem with titanium aluminides is their tendency to become more
brittle as they absorb hydrogen from their surroundings (this was part
of the problem Olson had taken on with the space shuttle s bearing
steels).  Hydrogen atoms would be abundantly available during each and
every NASP trip.

Even though Congress killed the NASP program in 1994, the very fact
that researchers like Eberhart thought they could build such a vehicle
is testimony to their faith in materials design.  It would fly only if
materials designers systematically invented the right stuff.

The emergence of materials design will be felt not only by the
materials designer and maker, but it will also have social and
geopolitical consequences.  It will become part of the equations
governing the military and economic balance of power, local and global
environmental health, resource management, and the quality of life for
an expanding population.  Those countries with the more talented
materials designers will have the edge.

The drivers of innovation in materials will remain largely the same.
Hypersonic aircraft such as the NASP is only one of countless examples
of how engineers look to new materials to enable new technology.  The
need for new materials runs through anyone s list of high-technology
desiderata.  Faster computation depends in part on new semiconductors
in which electrons move faster.  Faster electronic materials mean
faster computers, which mean more reliable weather forecasting and
climate modeling, quicker design of more fuel-efficient vehicles, and
more fruitful discovery of drugs by molecular modeling.  New kinds of
optical fibers, some of which have rare earth atoms such as erbium that
have the curious ability to amplify light passing through them, mean
more reliable, less complicated long-distance communication systems.
New alloys, composites and ceramic materials can mean a new system of
transportation that moves people faster, more comfortably, with fewer
accidents, with less energy and pollution.  Armor that can the stuff
that dreamsaremadeof a 261

withstand ever more powerful armor-piercing munitions has everything to
do with new materials.  Medical implants, prostheses, and artificial
organs are also on the research agendas of teams of materials
scientists specializing in biology and medicine.  Name the technology
and materials science is there in the forefront.

Pushing old technologies and developing new ones is just one of the
drivers for new materials.  Other factors are the failure of current
materials to live up to expectations; the relentless market pressure to
offer new and improved products; the military need for ever more
capable materials and technologies; the academic honor of creating a
material with superlative properties; and the societal benefit of
having less expensive, more energy-efficient, environmentally safer
materials.

But materials science isn t simply at the mercy of the prevailing
social winds.  The sophistication of researchers like Capasso and Olson
allows them to anticipate and steer some of these driving factors.
Engineers no longer need to design with the limitations of materials in
mind; they can design technologies with materials they would like to
have.

This is a kind of engineering bravado that used to be relegated to the
minds of science fiction writers, who could simply imagine fantastic
materials and technologies.  Of course, the achievements of
contemporary technology often dwarf the wildest imaginings of a century
ago.  Now the distance between thought experiment and real experiment
is often measured in days or weeks rather than in decades.

On the other side of the Janus face of technology is the potential to
design new materials that the world may eventually come to regret.  In
the 1930s Thomas Midgely, a chemist then working for Delco, developed
tetra ethyl lead.  The primary concern was to come up with an antiknock
agent to prevent military planes from stalling out, but it became a
staple for the gasoline industry since it also improved the performance
of automobile engines.  Midgely also invented chlorofluorocarbons

(CFCs) as a refrigerant that would be safer than the ammonia and sulfur
dioxide then in use.  Only decades later, with production and
distribution of these chemicals spanning the globe, did their darker
side become apparent.  Specters of lead poisoning and the destruction
of stratospheric ozone molecules by CFC molecules have been among the
gravest environmental and public health concerns in the late twentieth
century.

There are plenty of others.  The list of materials whose troublesome
properties have come to outweigh their useful ones is quite long.
Traditional lead-tin solders, though convenient to use since their
melting point is low, are on their way to becoming completely phased
out and already are taboo for most plumbing applications, where their
past use has contributed to widespread exposure to toxic levels of
lead.  Asbestos is a fine thermal insulation material but also a
dangerous carcinogen if its fibers are inhaled.  Similar dangers may
lurk in the reinforcing fibers and flakes made of other synthetic
ceramic and mineral materials whose production could increase if
composites become as commonplace as many engineers would like.
Likewise, medical implant materials have improved the quality of life
for millions of people, but some of their negative side effects are
becoming apparent.  In response to an article in Science News about
attempts to make synthetic diamond coatings, one letter writer feared
that an age of synthetic diamond, and the super hard and
degradation-resistant coatings that could come of it, might produce a
persistent waste problem.  After all diamonds are forever.

Nevertheless, there are reasons to be optimistic.  Materials developed
before the environmental movement were driven by convenience, bottom
lines, and ready availability of ingredients.  Environmental issues
were seldom seen on the checklists of materials researchers.  That no
longer is true.  Designing for the environment is becoming integral to
the mind-set of influential members of the materials community, the
regulatory commu the stuff that dreams are made of N 263

nity, and users of the materials.  In Germany, for example, auto makers
are responsible for recycling all of the materials that go into their
cars.

One promoter of this environmental perspective is Robert Laudise,
adjunct chemical director of Lucent Technologies.  In the late 1950s he
helped develop synthetic quartz for radio and electronics technologies,
and he rose through the ranks to become head of Bell Lab s materials
R&D.  One way Laudise has tried to change the materials research
culture is by requiring those under his wing to build environmental
factors into their work at the earliest research stages so that
potential problems will show up sooner rather than later.  And one of
the first things Laudise did in 1993, when he took over the helm of the
Journal of Materials Research (one of the field s most visible
publications), was to organize a special issue devoted to Materials and
the Environment.

Laudise is not alone in his desire to foment a new ethic in the
materials community.  Many companies such as 3M and Alcoa now nip the
development of new materials, products, and technologies in the bud if
there is potential for environmental harm even in the most promising
products or materials.  Companies that are realistic about regulatory
action, public perception, and the growing cost of dirty operation and
waste disposal are actively curtailing their pollution production and
use of hazardous chemicals and materials.  That means they have
researchers, including materials researchers, who are developing a
cradle-to-grave kind of thinking.  This purview extends from raw
materials extracted from the earth to the processing of these materials
into products to the service lifetime of those materials and products
to the end game of recycling and disposal.

Even the enormous military infrastructure, which has often recklessly
ignored the environmental consequences of some of the most hazardous
materials ever created, is taking on a greenish tinge.  The raison d
etre of the Hanford Nuclear Reservation in Washington State has shifted
from making the nation s military plutonium to finding chemical and
materials based solutions to clean up the mess left in the wake of Cold
War weapons production.

This sea change of thought and practice cannot ensure that there are no
potential environmental disasters unwittingly being brewed in
contemporary labs.  But now materials researchers, government planners,
and corporate managers are addressing the environmental and public
health issues they had ignored or punted in the past.

Environmental risks notwithstanding, thinking about the technological
future can still induce the dropped jaw and the starry eyes.  Every
week reports in technical journals like Applied Physics Letters, the
Journal of the American Chemical Society, Luminescent Materials, and
Tissue Engineering nudge and push the materials envelope just a little
more.  The unpublished results from company and government laboratories
could be going even further.

What s more, progress is accelerating in each part of the materials
research triumvirate: synthesis, analysis, and theory.  The result is a
kind of bootstrapping.  Computational tools help steer synthesis and
make sense of analysis.  The ever better ability to synthesize
structures to theoretical order and the data from ever more
sophisticated analyses help refine the theories and simulations running
inside the computers.  Perhaps the most self-reflective example of this
process is the use of silicon-based computer chips to design better and
more powerful silicon chips, which, in turn, will be used to design the
next generation of chips, perhaps ones made out of a material like
gallium arsenide or even synthetic diamond.

It is possible, of course, that all this wondrous wizardry with
materials will merely color the history of technology but not radically
reshape it.  But given the startling technological upheavals of the
past fifty years alone, contemporary materials science is likely to
have as profound an effect on posterity as did that original act of
materials engineering in eastern Ahica s Rift Valley, where the sound
of stone against stone first snapped into the Paleolithic air.

REFERENCES AND NOTES

This book emerged over a period of about five years, during which time
1 covered the field of materials science and engineering as a science
writer.  The book s themes come from a synthesis of thousands of
individual components: visits to laboratories, interviews, technical
papers, oral presentations at meetings, demonstrations, press
conferences, and plenty of bookish research.  To even attempt to list
half of the sources that played some role in this book would, I think,
fatten the spine without adding real value.  In what follows, I offer a
bibliographic essay of sorts that lists some general sources pertinent
to each chapter and/or footnotes referring to the more specialized
sources for particular points or claims.

Introduction: Stuff, Stuff , Everywhere Stuff 1. Cyril Stanley Smith, A
Search for Structure (Cambridge, Mass: MIT Press, 1981).

Chapter One: Earth, Air, Fire, and Water

A good starting point for contemporary thinking about the earliest
hominid relationship to materials is Making Silent Stones Speak: Human
Evolution and the Dawn of Technology (New York: Simon & Schuster, 1993)
by Kathy D. Schick and Nicholas Toth.  Toth also has written a series
of articles in Scientific American on early technology and the use of
materials.  A fascinating and informative source on the use of
materials in earlier times is Smithsonian s nmelines of the Ancient
World: A Visual Chronologyfrom the Origins of Life to 1500 (New York:
Dorling Kindersley, 1993), edited by Chris Scarre.  This oversize book
is exquisitely and expensively brimming with pictures, diagrams, and
graphs prepared by a platoon of authoritative scholars who have made it
usable as a guided tour through the historical use of materials
complete with a pictorial museum s worth of artifacts.

There is a large literature on the early human use of pottery, metal,
and glass-the products of earthy ingredients treated with fire.  The
next stop, after technological encyclopedias and other more general
standards like Jacob Bronowski s The Ascent of Man (Boston: Little,
Brown and Company, 1973), is more and more specialized parts of the
literature.  A general consensus about the outline of this history has
been available for decades, which means that most contributions today
tend to shift or add details.  The writings of Cyril Stanley Smith
provide authoritative details, especially regarding metallurgy, put
into beautiful, sweeping historical and philosophical contexts.  See
his History of Metallography: The Development of Ideas on the Structure
of Metals before 1890 (Cambridge, Mass.: MIT Press, 1988), which was
one of the inspirations for this book.  A Search for Structure
(Cambridge, Mass.: MIT Press, 1981), is a wonderful collection of Smith
s essays on science, art, and history that provides an impressionistic
view of his philosophy of materials.  Also good on metallurgy is
Theodore A. Wertime.  See, for example, his Man s

First Encounters with Metallurgy in Science, 4 December 1964,
1257-1267.  Also valuable to me was an unpublished source entitled
Notes on the History of Metals and Ceramics, which Professor Morris
Fine of Northwestern University compiled for his courses in the early
1970s.

A Short History of Glass (New York: Harry N. Abrams, Inc."  in
association with the Coming Museum of Glass, 1980) is a readable,
authoritative, and informative entry into the history of glass written
by the curator of the Corning Museum of Glass, Chloe Zerwick.  For
early ceramic and fired-clay technology, see anything with the names of
W, David Kingery or Pamela Vandiver, which are usually tough reads but
worth the effort.  An example is Ceramic Masterpieces: Art, Structure,
and Technology (New York: The Free Press, 1986).  Also see collections
of research writeups such as Early Pyrotechnology: The Evolution of the
First Fire-Using Industries (Washington, D.C.: Smithsonian Institution
Press, 1982), edited by Theodore and Steven Wertime, which is a
proceedings of a seminar on early pyrotechnology.

1. The following fictional vignette depicting the birth of Paleolithic
stone tool technology is derived from the modern corpus of
archaeological and paleoanthropological data and theories.
Paleoanthropologists themselves resort to this kind of communication.
See Schick and Toth, Making Silent Stones Speak, 77-78, 147-149,
187-189, and 225-227.  Just which genus of early hominid,
Australopithecus or Homo habilis, was the first to use tools remains a
point of contention.  For a recent articulation of the debate, see
Randall L. Susman, Fossil Evidence for Early Hominid Tool Use, Science,
9 September 1994, 1570-1573.

2. Cyril Stanley Smith.  Metallurgy as a Human Experience, in
Metallurgical Transactions A, April 1975, 606-607.

3. Ibid.  4. Paul Rado, An Introduction to the Technology of

Pottery, 2d ed.  (New York: Pergamon Press, 1988), 2. Some historians
credit hunter-gatherers in Japan about twelve thousand years ago, not
agriculturists in the Fertile Crescent, as the ones to first make
pottery vessels.  See, Scarre, The Smithsonian nmelines, 65.

5. Betty Jo Teeter Dobbs, The Foundation of Newton s Alchemy or The
Hunting of the Greene Lyon (New York: Cambridge University Press,
1975).

6. Theodore A. Wertime, Man s First Encounters with Metallurgy,
Science, 4 December 1964, 1257-1267.

7. See, for example, Cyril Smith s Materials and the Development of
Civilization and Science, Science, 15 May 1965, 911.  The quote comes
from Smith s essay Matter versus Materials: A Historical View, which
was reprinted in A Search for Structure.

S. T. A. Rickard, Man and Metals, vol.  1 (New York: McGraw-Hill,
1932),127.

9. Leslie Aitchison, A History of Metals (New York: Interscience
Publishers, 1960).

10.  See, for example, Bronowski, The Ascent of Man, 128.  11. 
Nicholas Reeves, The Complete Tutankhamun (New York, Thames and Hudson,
1990).

12.  Cyril Stanley Smith, Science, 14 May 1965, 913.  13.  -. A
Searchfor Structure, 1-67.  14.  Zerwick, A Short History of Glass, 8.
15.  Ibid."  12.  See also Pliny the Elder, Natural History: A
Selection (New York- Penguin Books, 1991), 361-363.

16.  Zerwick, A Short History of Glass, p. 25.  17.  Lewis Mumford,
Technics and Civilization (New York: Harcourt, Brace & World,
1963),124.

18.  Ibid.  19.  For history of early plaster, see M. James Blackman,
The Manufacture and Use of Burned Lime Plaster at ProtoElamite Anshan
Uran) in Early Pyrotechnology: The Evolution of the First Fire-Using
Industries, eds.  Theodore and

Steven Wertime (Washington, D.C.: Smithsonian Institution Press 1982),
107-115.

20.  Ibid."  105-135.

Chapter Two: New Stuff in the Old World A good starting point for the
history of nineteenth century technology is the enormous multivolume
classic aptly entitled A History of Technology (Oxford University
Press: New York, 1954-1984), edited by Charles Singer, E. J. Holmyard,
A. R. Hall, and IYevor 1. Williams.  Volumes IV and V focus on the
nineteenth century.  On the early days of polymers, see especially
Robert Friedel s Pioneer Plastic: The Making and Selling of Celluloid
(Madison: University of Wisconsin Press, 1983) and the same author s
excellent booklet-length social interpretation of an exhibit on
material at the Museum of American History: A Material World
(Washington, D.C.: The National Museum of American History, Smithsonian
Institution, 1988).  Stephen Fenichel also has compiled a fascinating
social history of plastics in Plastic: The Making of a Synthetic
Century (New York: Harper Business 1996).

Fred Aftalion s A History of the International Chemical Industry
(Philadelphia: University of Pennsylvania Press, 1991) lays out the
emergence and spread of a chemical industry from its nineteenth-century
roots.  Several chapters on metals in James Edward Gordon s The Science
of Structures and Materials (New York: Scientific American Books, 1988)
have excellent accounts of nineteenth- and early twentieth-century
metallurgy.

For the history of chemistry, see William Brock s Chemistry (New York:
W. W. Norton & Company, 1992).  It is an excellent springboard into the
literature, as is J. R. Partington s 1937 classic A Short History of
Chemistry (New York: Dover Publications, 1989).  Same for Hugh Salzberg
s From Caveman to Chemist (Washington, D.C.: American Chemical Society,
1991) and David Knight s quite readable Atoms and Elements (London:
Hutchinson & Co."  1967).

1. Chloe Zerwick, A Short History of Glass (Harry N. Abrams, Inc."  in
association with the Corning Museum of Glass: New York, 1980),50.

2. David Kingery, Ceramic Materials Science in Society, in Annual
Review of Materials Science (Annual Reviews Inc., 1989),5-8.

3. David Kingery, Looking to the Future, in Ceramic and Civilization:
From Ancient Technology to Modern Science, ed.  David W. Kingery
(Columbus: The American Ceramic Society, 1985), 378.

4. David Kingery and Pamela B. Vandiver, Ceramic Masterpieces (New
York: The Free Press, 1986), 195-206.

5. Cyril Stanley Smith, A History of Metallography: The Development of
Ideas on the Structure of Metals before 1890 (Cambridge, Mass: MIT
Press, 1988).

6. See, for example, Materials Science and Engineering for the 1990s
(Washington D.C.: National Academy Press, 1989).

7. This data come from a graph in HR Shubert s Iron and Steel, in
Singer, History of Technology, vol.  4, 106.

8. These numbers come from H. R. Shubert s The Steel Industry, in
Singer, History of Technology, vol.  5.

9. Robert Friedel, A Material World: An Exhibition at the National
Museum of American History (Washington, D.C.: Smithsonian Institution,
1988), 49.

10.  Gordon, The Science of Structures, 129.  11.  S. B. Hamilton s
Building Materials and Techniques, in Singer, A History of Technology,
vol.  5."  479.  Steel had been used two years earlier in the upper
floors of the first skyscraper ever built, also in Chicago.  See
Singer, History of Technology, vol.  5, 62.

12.  On the history of ideas regarding the role of carbon in metals,
see Smith, History of Metallography.

13.  Siemens is quoted in W. H. G. Armytage, A Social History of
Engineering (Boulder: Westview Press, 1976).

14.  The details about Goodyear s quest are nicely laid out by Stephen
Fenichell in his Plastic: The Makings of a Synthetic Century (New York:
Harper Business 1996).

15.  This account of the earliest synthetic plastics relies heavily on
Robert Friedel s Pioneer Plastic: The Making and Selling of Celluloid
(Madison: The University of Wisconsin Press, 1983).

16.  Friedel points out that the camphor actually remains in the bulk
of the material; see Friedel, Pioneer Plastic, 14-15.

17.  Friedel, A Material World, 49-53.  See also R. Chadwick, New
Extraction Processes for Metals, in Singer, History of Technology, vol.
5, 90-94.

18.  Ibid.  19.  Quoted from Matthew Josephson, Edison, (New York: John
Wiley & Sons, 1959),133-134.

20.  David.  A. Hounshell and John Kenly Smith, Jr."  Science and
Corporate Strategy (New York: Cambridge University Press, 1988),2.

Chapter Three: The Secret Architecture of Stuff The roots of the field
of materials science and engineering are so many and varied that any
account of them will by necessity reflect personal choices.  Much of
this chapter was pieced together from discussions and review articles
appearing originally in venues such as Scientific American as well as
many more technical books and articles.  Still, there are several more
general sources for finding parts of the story.  Cyril Stanley Smith s
A History of Metallography: The Development of Ideas on the Structure
of Metals before 1890 (Cambridge, Mass: MIT Press, 1988) is a
superlative account of the beginnings of materials science before it
even had a name.  Parts of the narrative can be picked out of Smith s
also previously mentioned A Searchfor Structure (Cambridge, 1981).  An
excellent, thorough, but very heavy-going source is entitled Out of the
Crystal Maze: Chapters from the History of Solid-State Physics (New
York: Oxford University Press, 1992), edited by Lillian Hoddeson,

Ernest Braun, Jurgen Teichmann, and Spencer Weart.  A more concise and
handier introduction can be found in Andrew Briggs, ed."  The Science
of New Materials (Cambridge, Mass.: Blackwell, 1992).

1. J. Randolph Kissell and Robert L. Ferry, Aluminum Structures: A
Guide to Their Specifications and Design (New York: John Wiley & Sons,
1995).

2. Cochrane, Rexmond C."  Measures for Progress: A History of the
National Bureau of Standards (U.S. Department of Commerce: 1966).

3. Arden Bement, and appeared in the Metallurgical T@-ansactions A,
March 1987, 363-375

4. Cyril Stanley Smith, personal communication.  5. Herman Mark, oral
history interview archived at the Center for the History of Chemistry
at the University of Pennsylvania.

6. Hermann Staudinger, From Organic Chemistry to Macromolecules: A
Scientific Autobiography Based on My Original Papers, Herman F. Mark
trans.  (New York: WileyInterscience, 1976).

7. Herman Mark, oral history.  8. See, for example, R W. Atkins,
Physical Chemistry, 3d ed.  (New York: 1986),304.

9. This story has been told countless times in many different contexts.
See, for example, S. Millman, ed."  A History of Engineering and
Science in the Bell System: Physical Sciences (19251980) (Short Hills,
NJ: AT&T Bell Laboratories, 1983); and Hans Queisser, The Conquest of
the Microchip (Cambridge, Mass.: Harvard University Press, 1988).
Additional sources include Charles Weiner, How theI)ransistor Emerged,
IEEE Spectrum, January 1973, 24-33; and Mervin J. Kelly, The First Five
Years of the Transistor, Bell Telephone Magazine, Summer 1953, 72-86.

references and notes M 275

Part 11-The Birth of a Superdiscipline

There are a number of first-person written accounts of the events
leading up to and shortly following the birth of an academic discipline
and field of research known as materials science and engineering.  One
of best began as a distinguished lecture given in 1986 by Arden 1.
Bement, Jr."  at a gathering of materials researchers.  Called The
Greening of Materials Science and Engineering and appewing in the
Metallurgical Ransactions A, March 1987, 363375.  Two documents
prepared by Roman J. Wasilewski were complementary to Bement s article.
Development of the Materials Research Laboratory Program in the U.S.
was written shortly before its author passed away and was never
published.  The other was an Outline of the MRL Program that Wasilewski
prepared at the end of 1981 for internal use within the National
Science Foundation.  These two documents were given to me by Lyle
Schwartz, who heads the Materials Science and Engineering Laboratory at
the National Institute of Standards and Technology.  Another good
source for this is Robert Sproull s The Early History of the Materials
Research Laboratories, in Annual Reviews of Materials Science (Palo
Alto, CA: Annual Reviews Inc."  1987).  Advancing Materials Research
(Washington D.C.: National Academy Press, 1987) has several informative
historical chapters by insiders.

1. William 0. Baker, Advances in Materials Research and Development, in
(editors), Advancing Materials Science, eds.  Petre A. Psaras and Dale
Langford (Washington, D.C.: National Academy Press, 1987), 3-24.  The
advisory committee s paper is reprinted in full on page 24.  Quote
appears on page 4.

2. The two-age document that Fine gave the dean was entitled Importance
of Materials Science and Engineering.

3. A memo titled Metallurgy Faculty Meeting records the vote.  Another
memo, dated January 23, 1959, indicates that the relevant dean had
approved the request to change the department s name.

4. Rustum Roy, in the twentieth anniversary coverage in the MRS
Bulletin 18 (September 1993), 73-90.

5. Materials Science and Engineeringfor the 1990s.

Chapter Four: Materials Research Comes of Age The portrait of a field
that emerges in this chapter derives mostly from many personal
interviews and visits with scientists and engineers and from reading
their research and review articles.  One official expression of how
materials science has come of age since its inception is the
book-length report Materials Science and Engineering for the 1990s
(Washington D.C.: National Academy Press, 1989).

1. The examples of ski-ramp curves cited here are derived mostly from
Materials Science and Engineering for the 1990s.

2. See, for example, K. Alex Muller and J. Georg Bednorz, The Discovery
of High-Temperature Superconductors, in Nature, 4 September
1987,"33-1139.

3. Ivan Amato, Turning Polymer Spaghetti into Lasagna, Science, 13
September 1991, 1212, reporting on presentation by Stupp at the Fourth
Chemical Congress of North America in New York City.

4. Phaedron Avouris, personal communication.  5. L. Esaki and R. Tsu,
Superlattice and Negative Differential Conductivity in Semiconductors,
IBM Journal of Research and Development (January 1970), 61-65.

6. See, for example, James L. Wilbur, Amit Kumar, Enoch Kim, and George
Whitesides,

Microfabrication by Microcontact Printing of Self-Assembled Monolayers,
Advanced Materials 6: 600-604.

7. For a nontechnical review of Stoddart and his ilk, see Ivan Amato,
Designer Solids: Haute Couture in Chemistry, Science, 7 May 1993,
593-595.

8. For a review of biornimetic materials, see Ivan Amato,

Heeding the Call of the Wild, in Science, 30 August 1991, 966-968.

9. Julian Vincent, a sociologist and materials scientist at Reading
University in England, personal communication.

10.  Quoted from Impact of Supercomputing Capabilities on

U.S. Materials Science and Technology (National Material Advisory Board
National Research Council, September 1988),38.

11.  Much of the discussion on theory and computational approaches is
derived from the research that I did for the following article: Ivan
Amato, The Ascent of Odorless Chemistry, Science, 17 April 1992,
306-308.  I spoke with a dozen or so practitioners of computational
chemistry and materials science.  Examples of primary sources include
Henry F. Schaefer 111, Methylene: A Paradigm for Computational Quantum
Chemistry, Science, 7 March 1986, 1100-1107; Marvin L. Cohen,
Predicting New Solids and Superconductors, Science, 31 October 1986,
549-553; D.D. Vedensky, S. Crampin, M. E. Eberhart, and J. M.
Maclren,

Quantum Mechanics and Mechanical Properties: Towards Twenty-first
Century Materials, Contemporary Physics 31 (1990): 73-97.

12.  Marvin Cohen, personal communication.  13.  Amy Y. Liu and Marvin
L. Cohen, Prediction of Low Compressibility Solids, in Science, 25
August 1989,841-842.

14.  Charles T. Casale and Bruce Gelin, Growth and Opportunity in
Computational Chemistry (Boston: Aberdeen Group, 1992.)

15.  William Goddard 111, personal communication.  16.  Tommaso Toffoli
and Norman Margolus, Cellular Automata Machines (Cambridge, Mass.:
M.I.T. Press, 1987).

Chapter Five: The Materials Serengeti Twice each year at semiannual
gatherings of its members, the Materials Research Society in Pittsburgh
publishes a phonebook-size tome filled with several thousand abstracts
of the meeting s oral and poster presentations.  These abstract books
and meetings are a particularly relevant example of many similar books
and meetings that reveal the same kind of Serengeti metaphor for the
field.  The three more focused subsections of the chapter are based
almost entirely on personal interviews, visits, and the technical
literature.  What follows are several more general sources for each
section.

Synthetic Diamond.  See Robert Hazen s The New Alchemists: Breaking
Through the Barriers of High Pressure (New York: Times Books, 1993). It
is a nicely written account of the highpressure, high-temperature route
to synthetic diamond and touches lightly on the low-pressure route that
is my focus in this chapter section.  The Materials Research Society
published a special issue of its journal The Journal of Materials
Research 5 (November 1990) that included over three hundred pages of
technical papers that collectively give a flavor of the field. A number
of insiders have written informative review articles.  One is by Robert
C. DeVries: Synthesis of Diamond Under Metastable Conditions, Annual
Review of Materials Science (Palo Alto: Annual Reviews, 1987). John C.
Angus also has presented excellent technical histories of the field. 
See for example, Angus and Cliff C. Hayman s article in Science (August
18, 1988, 913-921) entitled

Low-Pressure, Metastable Growth of Diamond and Diamondlike Phases. Also
see Status and Applications of Diamond and Diamond-Like Materials: An
Emerging Technology (Washington D.C.: National Academy Press, 1990),
another way to get a feel for the synthetic diamond arena.

Biomimetic Materials.  I know of no general texts focusing on either
bio mimetic research or smart materials.  A short review can be found
in Heeding the Call of the Wild by Ivan Amato in Science (August 30,
1991, 966-968).

Smart Materials.  To get your feet wet in smart materials, see Ivan
Amato s chapter Smart Materials in On the Cutting Edge of Technology
(Carmel, Indiana: SAMS, Publishing, 1993) and Animating the Material
World by Ivan Amato in Science, (January 17, 1992, 284-286).

1. For this and plenty of other interesting historical revelations
about the quest for artificial diamonds, see Robert Hazen, The New
Alchemists: Breaking Through the Barriers of High Pressure (New York:
Times Books, 1993).

2. Robert Wentorf, personal communications.  Also, see Ivan Amato,
Diamond Fever, Science News, 4 August 1990, 72-74.

3. Rustum Roy, personal communication.  4. John Angus, History and
Current Status of Diamond Growth at Metastable Conditions, a talk
presented at the 175th meeting of the Electrochemical Society in Los
Angeles, California, on May 8,1989.

5. John Angus, personal communication.  6. DeVries, personal
communication.  7. Rustum Roy, personal communication.  8. John Angus,
personal communication.  9. Rustum Roy, personal communication.  10.
National Research Council, Status and Applications of Diamond and
Diamond-Like Materials: An Emerging Technology (Washington D.C.:
National Academy Press, 1990).

11.  John Angus, personal communication.  12.  H. Kagan, et al."
Diamas: A Compact Diamond-Based Detector for the SSC, unpublished
document.

13.  All quotes in this paragraph from DeVries, personal
communication.

14.  National Research Council, Status and Applications.  15.  That
idea formed the basis of Seitz s 1975 patent claim and showed up in Tom
Clancy s techno-espionage thriller, The Cardinal of the Kremlin, in
which Soviet military scientists,

with the help of brilliant kidnapped U.S. researchers, are on the verge
of developing SDI-type lasers capable of destroying missiles and
satellites from the ground.

16.  Previous lab-made polymers such as celluloid were derived from
existing natural polymers such as the cellulose of plants.

17.  Julian Vincent, personal communication.  Julian Vincent The Design
of Natural Materials and Structures, The Journal of Intelligent
Materials and Structures 1 (January 1990): 141-145; Julian Vincent,
Relationship between Density and Stiffness of Apple Flesh, Journal of
Food and Agriculture 47 (1989):443-462.

18.  See, for example, Advanced Materials 3 (1991): 449-452.  This
includes Rustum Roy s argument and a rebuttal by John Maddox, editor of
Nature.

19.  Steven Vogel, Life s Devices (Princeton: Princeton University
Press, 1988), 315.

20.  Eric Baer, personal communication.  Also see, E. Baer, A. Hiltner,
and H. D. Keith,

Hierarchical Structure in Polymeric Materials, Science, 27 February
1987, 1015-1022; Eric Baer, Advanced Polymers, Scientific American,
October 1986, 179-190.  Eric Baer, James J. Cassidy, and Anne Hiltner,
Hierarchical Structure of Collagen Composite Systems: Lessons from
Biology, Pure & Applied Chemistry 67 (1991): 961-973.

21.  Paul Calvert, personal communication; Paul Calvert, Biomimetic
Materials, unpublished presentation at a conference on Intelligent
Materials in Hawaii, 1990.

22.  Mehmet Sarikaya, personal communication.  23.  Greg Olson,
personal communication.  24.  Mehmet Sarikaya, Katie E. Gunnison, and
Ilhan A. Aksay, pre print of Mechanical PropertyMicrostructural. 
Relationships in Abalone Shell.

25.  James Economy, University of Illinois at Urbana, personal
communication.

26.  Stephen Mann has written many articles on the topic of bio
mineralization  See, for example,

Molecular Recognition in

Biomineralization, Nature" 10 March 1988, 119-124; Molecular Tectonics
in Biomineralization and Biomimetic Materials Chemistry, Nature, 7
October 1993, 499-505;

Crystal Engineering: The Natural Way, New Scientist, 3 October 1990,
42-47.

27.  Lia Addadi and Stephen Weiner, Control and Design Principles in
Biological Mineralization, in Angewandte Chemie, vol.  31 (Weinheim
international English edition, 1992), 153-169.

28.  Robert R. Jones, Dan W. Urry Chosen Scientist of the Year, in
Research and Development Magazine, October 1988, 56-57.

29.  Urry, personal communication.  30.  Donald S. Stookey, Journey to
the Center of the Crystal Ball: An Autobiography (Columbus: The
American Ceramics Society, 1985), 13-15.

31.  Donald S. Trotter, Photochromic and Photosensitive Glass,
Scientific American, April 1991, 124-129.

32.  Richard Claus, personal communication.  33.  See Craig Rogers,
Intelligent Material Systems-the Dawn of a New Materials Age, pamphlet
(Blacksburg: Center for Intelligent Materials Systems and Structures at
Virginia Polytechnic and State University: 1992).  This is a good
source to get a sense of Craig Roger s philosophical views on the field
of smart materials and structures.  34.  One way to a get sense of the
variety of research projects that fall under the rubric of smart
materials is to leaf through the proceedings of one of the field s
meetings.  Some of the examples cited are described in Gareth J.
Knowles, ed."  Active Materials and Adaptive Structures: Proceedings of
the ADPAIAIAAIASMEISPIE Conference on Active Materials and Adaptive
Structures, 4-8 November 1991, Alexandria, VA (Philadelphia: Institute
of Physics Publishing, 1992).

35.  Craig Rogers, personal communication.  36.  Ibid.  37.  Submarine
cloaking was a major topic at the ADPAJ AIAA/ASME/SPIE Conference.

38.  W. Kraschmer, L. D. Lamb, K. Fostiropoulos, and D. R. Huffman,
Solid C60: A New Form of Carbofi, Nature, 27 September 1990, 354-358.

39.  Richard Smalley, Internet posting of a presentation titled From
Balls to Tubes to Ropes: New Materials from Carbon at a meeting of the
South Texas Section of the American Institute of Chemical Engineers in
Houston on January 4, 1996.

Chapter Six: The Rite of Atomic Masons The bulk of this chapter is
based on personal communications and visits with Federico Capasso and
from a large stack of technical papers on the field of band-gap
engineering.  For an introduction to band-gap engineering, see Band-Gap
Engineering: From Physics and Materials to New Semiconductor Devices by
Federico Capasso in Science, 9 January 1987, 172-176; and Federico
Capasso and Alfred Y. Cho, Band-Gap Engineering of Semiconductor
Heterostructures by Molecular Beam Epitaxy: Physics and Applications,
Surface Science 299/300 (1994): 878-891.

1. See, for example, Charles Weiner, How the Transistor Emerged, IEEE
Spectrum, January 1973, 24-33; and Mervin J. Kelly, The First Five
Years of the Transistor, Bell Telephone Magazine, Summer 1953,72-86.

2. This story has been told countless times in many different contexts.
See, for example, S. Millman, ed."  A History of Engineering and
Science in the Bell System: Physical Sciences (19251980) (Short Hills,
NJ: AT&T Bell Laboratories, 1983); and Hans Queisser, The Conquest of
the Microchip (Cambridge, Mass.: Harvard University Press, 1988).

3. In addition to Capasso, I spoke with Jerome Faist, Barbara Sivco,
and Alfred Cho.

4. A fascinating account of the early Corning work on optical fibers is
in Ira Magaziner and Mark Patinkin, The Silent War.  Inside the Global
Business Battles Shaping America s Future (New York: Vintage Books,
1989),264-299.

references and notes R 283

5. L. Esaki, and R. Tsu, Superlattice and Negative Differential
Conductivity in Semiconductors, IBM Journal of Research and
Development, January 1970.

6. R. F. Kazarinov and R. A. Suris, Sov.  Phys.  Semicond.  5
(1971):207.  7. Jerome Faist, Federico Capasso, Deborah L. Sivco, Carlo
Sirtori, Albert L. Hutchinson, and Alfred Cho, Quantum Cascade Laser,
Science, 22 April 1994, 553-556.

Chapter Seven: Composing Steel The bulk of this chapter was based on
personal communications and meetings with Stephen Carr and Greg Olson
and from a large portfolio of Olson s technical papers.  To get a feel
for what materials design means, see

New Steels by Design by Greg B. Olson in Journal of Materials Education
11 (1989): 515-528.  The fullest articulation of the approach appears
in Olson, Carr, Jennings, and Jones, Materials Design Initiative: An
Integrated Research and Teaching Consortium for Systems Design and
Materials, a proposal to the National Science Foundation s Division of
Materials Research, 1994.

1. T. A. Stephenson, C. E. Campbell, and G. B. Olson, Systems Design of
Advanced Bearing Steel, in Advanced Earthto-Orbit Propulsion Technology
1992, eds.  R. J. Richmond and S. T. Wu, NASA Conference Publication
3174, vol.  2 (1992), 299-307.

2. This information comes from Olson s recollection of discussions with
the friend from Rockwell International who told him about the problem
with steel in the pump bearings.

3. Jon Van, New Meaning to Phrase Tough as Steel, Chicago Tribune, 31
December 1989,3.

Chapter Eight: The Stuff That Dreams Are Made Of

1. Mark Eberhart, personal communication.

INDEX

Abalone shells, 126, 160, 162-63, 165-68 Adheron, 164 Advanced Research
Projects Agency (ARPA), 94,95,171 Age hardening, 59-60, 73-74, 128 Age
of Plastics, 55, 76 Agricultural revolution, 20-23 Aksay, Ilhan, 167
Albany Dental Plate Company, 53 Alchemy: gold and, 6, 8, 13, 25; trial
and error in, 12-13 Alcoa (Aluminum Company of

America), 58, 263 Alloys, 9; aluminum, 58-60, 73-74;

nickel-based, 102; steel, 45-46, 110, 230-55;super alloys,108,2-53
Aluminum, 55-59; age hardening of,

59-60, 73-74, 128; alloys of, 58-60, 73-74; cooking metaphor for, 69;
development of, 56-58; in steel production, 58; uses of, 58 Aluminum
gallium arsenide, 124,

211-12 American Chemical Society (ACS), 100 American Telephone and
Telegraph

Company (AT&T), 61,85,191,192, 194,208,213 An-tide bonds, 62-63 Angus,
John, 145-47,149,150,164 Apollo program, 101 Applied Physics Letters,
264 Aristotle, 4 Armistead, William H."  171 Artificial nervous
systems, 172-74 Asbestos, 10, 262 Atomic Energy Commission (AEC), 90,

91,94,95 Atomic hydrogen, 83, 132, 146-47 Atomic structure, 65-83;
biological materials and, 126-27,157-58, 160-63; chemical analysis and,
70, 71; contaminants and, 74-75; dendritic, 102, 117; early theories
of,

38-39; electron microscopy and, 74, 111-13; hierarchy in, 69-70;
imperfections and, 73-74; molecular beam epitaxy and, 119-25, 194-95,
202; periodic table and, 65-69, 188, 189, 199-200; screw dislocation
and, 74-75; synthetic materials and, 117-18,140-41; X-ray
crystallography in, 72-73, 76, 111-12.  See also Metallography;
Molecular structure; Quantum mechanical theory Augustus the Strong, 36
Austenite, 70 Australopithecus, 15 Automobile engines, 248-50,261
Avalanche photodiode (APD), 208-13 Avouris, Phaedon, 118-19

B. F. Goodrich, 61 Babylon, 33 Baekeland, Leo Henrik, 55, 60, 158 Baer,
Eric, 164-65 Bakelite, 54-55, 60, 76, 77 Band-gap engineering, 253-54;
communication networks and, 195-201, 205-27; higher-band gaps and, 199;
quantum cascade lasers and, 217-27, 234; quantum mechanical theory and,
195-201, 216-17; small band gaps and, 198-99 Bardeen'John,83,93,193
Barium lanthanum copper oxide, 109 Bauxite, 56 Beams, smart, 177 Bell,
Alexander Graham, 194 Bell Telephone Laboratories, 83-85,88,

93,138,191-201,202,214,218,263 Bessemer, Henry, 42-45,47,55,70 Binnig,
Gerd, 114,"9 Biontimetic materials, 127,137-38,

157-70; abalone shells and, 126, 160, 162-63, 165-68;
biontineralization and, 168-69; cert nets 167; collagen i n d e x

Biomimetic materials (cont.)

inspired, 165; controversy concerning, 160-61; early development of,
62, 78, 158-59; potential of, 164; protein-based polymers, 169-70;
vascular material, 159-60 Biomineralization, 168-69 Bird Flight as the
Basis of Aviation (Lilienthal), 161 Bismuth, 9 Blue Book (Materials
Science and

Engineeringfor the 1990s), 101 Bohr, Niels, 80-81 Bone, 12,16,19,165
Boron, 5 B6ttger, Johann Friedrich, 36-37 Bragg, William Henry, 72-73
Bragg, William Lawrence, 72-73 Brass, 59 Brattain, Walter, 83, 193
Brattain, William, 93 Bricks: ceramic, 56; limestone, 33 Brinkman,
Bill, 198,225,226 British Association for the Advancement of Science,
44 Bronze: casting, 27; components of, 13,

26-27; smelting, 26-28 Bronze Age, 24, 101 Brooklyn Bridge, 46, 101
Brooklyn Polytechnic Institute, 76-77 B-2 Stealth bombers, 103
Buckyballs, 179-81 Burnt iron, 45

CAD CAM (Computer Aided

Design/ Computer Aided Manufacturing), 188 Calcium carbonate, 33, 160,
162-63,

165-68 Calcium oxide, 33 California Institute of Technology, 132
Calvert, Paul, 165 Cannon balls, 42-44 Capasso, Federico, 191, 194-201,
202,

204-6,208-26,234,253,257-58,261 Carbon: fullerenes, 179-81; in iron
smelting process, 29-30, 42-45, 46-47; in producing synthetic diamonds,
140-41, 144-47, 155-56 Carbon-nitrogen materials, 130 Cardinal of the
Kremlin, The (Clancy),

Carnegie, Andrew, 61 Carnegie Geophysical Laboratory, 139 Carothers,
Wallace, 62-64, 77-78,

101-3,158,233 Carpenter Steel, 237, 247, 248,249-50 Carr, Stephen,
240-42, 252-53 Case-hardened steel, 249 Case Western Reserve
University, 145,

164 Cassiterite, 27 Cast iron, 42-43, 107-8 Celanese,164 Cellular
automata machines (CAMs),

133 Celluloid, 53-55, 76, 77 Celluloid Manufacturing Company,

53-54 Cement, 33, 253 Ceramics-glasses: elements in, 66; of nanoscale
particles, 183-85; piezoelectric, 175-76; properties of, 109-10;
relative importance of, 6-7; superconductivity of, 109.  See also
Glass; Porcelains; Pottery Cermets, 167 Challenger space shuttle, 238,
248 Chemical vapor deposition (CVD),

144-48,151-54 Chemistry: alchemy versus, 6, 8, 12-13,

25; applications of, 47-48, 71; chemical analysis and, 70, 71; new
materials and, 48-49, 116; science of, 46-47 China, bronze in, 27
Chlorofluorocarbons (CFCs): invention of, 9-10, 261-62; ozone and, 10,
254 Cho, Alfred, 202-5, 218, 221 Chromium, 46,246 Clancy, Tom, 156,178
Claus, Richard, 172-74,176 Clay, 18-19, 21-24.  See also Pottery Coal
tar, 47 Cohen, Marvin, 130-31,155 Cohen, Morris, 231,236 Coke,57 Cold.
See Temperature Collagen, 165 Collodion, 52-53,54 Colloidal structures,
77 Colorado School of Mines, 258-59 Communication networks, 191-201;

band-gap engineering and, 195-201,

i n d e x

205-27; lasers in, 194, 199-200, 202, 205-8, 217-27; light detectors
in, 208-13; mechanical relays in, 192, 193; optical fibers in, 200,
201-5, 208; photomultiplier tubes (PMTs) in, 213-16; semiconductors in,
194-95, 198--201, 205; transistors in, 193-94 Composite materials, 9,
253; elements in, 68; environmental concerns for, 262; relative
importance of, 6-7 Computational materials science,

132-33,245-46 Computers: CAD CAM systems, 188;

computational materials science and, 132-33, 245-46; molecular
mechanics techniques and, 131-32 Concrete, 253 Conduction band, 198
Converters, 43, 45 Coordinating Committee on Materials

Research and Development (CCMRD), 90,94 Copper: aluminum--copper
alloys, 59; in bronze, 13, 26-28; first use of, 24; paint and, 19-20;
smelting of, 24-26, 29 Cornell University, 94-95,189 Corning Glass
Works, 61, 116, 171, 172,

201-2 Crimean War, 42 Cryolite, 57-58 Crystallume, 154 Cubic boron
nitride, 110 Curl, Robert F."  180 Cylinders, smart, 176

Darby, Abraham, 41,46 De Beers Consolidated Mines, 142,181 Delco, 261
Democritus, 78 Dendrimers, 117 Dendritic tree like structures, 102,

117 Deryagin, Boris, 146-49 DeVries, Robert, 145,154 Diamonds,
synthetic.  See Synthetic diamonds Dirac, Paul, 128 Dirigibles, 60
DiSalvo, Francis, 189,191 Dopants, 203 Duke University, 161

Dunlop, John Boyd, 51 DuPont, 36,61, 62-64, 77-78,88,101-3,

158,259 Duralumin, 60, 73 Dyes, 4748, 53, 61.  See also Paint

E. I. duPont de Nemours and Company 36, 61, 62-64, 77-78, 88, 101-3,
158,259 Eads, James, 46 Eastman, George, 54 Eastman Kodak, 54, 61, 171
Eberhart, Mark, 258-60 Edison, Thomas, 57,61 Egypt, ancient: glass in,
30-31; metal in, 28; paper in, 34 Eiffel Tower, 46 Eigler, Donald,
119,120 Einstein, Albert, 82 Eisenhower, Dwight D."  91, 94
Electricity: in formation of aluminum

57-58; insulation materials for, 55 Electroluminescent polymers, 183
Electron microscopy, 74, 111-13 Empire State Building, 23 Energy
diagrams, 195 Engines: automobile, 248-50, 261; jet,

89,91 Environmental concerns, 9-10,261-64 Esaki, Leo, 119,206,208,211
Ethanol, 128-29 Ether, 52 Ethyl alcohol, 52 Ethylene, 145-46 Eversole,
William G."  143-47,149

Faist, Jerome, 218--26 Farben (I.G.), 61 Federation of Materials
Societies (FMS), 100 Femtosecond laser spectroscopy, 114 Feynman,
Richard, 242 Fine, Morris, 92-94,138,236 Fire, 5-6; in glass making
30-32; metallurgy and, 23-30, 116; pottery and, 18-19, 21-22; in
vulcanization process, 50, 51.  See also Temperature Flint, 5, 17, 20
Forecast 1, 173 Forecast 11, 173 Friedel, Robert, 55

i n d e x

Fuel pumps, 239, 248 Fullerenes, 179-81

Gallium arsenide, 124, 182-83, 198, 199,

202,211-12,218 Gallium indium arsenide, 220-21 Gambles Cave (Kenya), 21
General Electric Company (GE), 36, 61,

88,139-42,144,145,154,155-57 Gentz, Steve, 239 Germanium:
crystallization of, 93; in niobium-germanium intermetallic material,
109; point-contact transistor based on, 83-85, 193; in semiconductors,
198 Germany: aluminum in, 59,60; dye and chemical industry, 61
Gilchrist, Percy Carlyle, 45 Glass: architectural uses of, 32; early
uses of, 17, 30-32; invention of glassblowing, 31-32; origins of,
30,31; photochromic, 171-72; transparency of, 109-10; Venetian glass,
13, 32,35.  See also Ceramics-glasses Goddard, William A. 111, 132
Gold, alchemy and, 6, 8, 13, 25 Golden Gate Bridge, 23 Goodrich (B.F.),
61 Goodyear, Charles, 49-50,60 Graduate, The (film), 78 Granite, 17
Graphene sheets, 180 Graphite, 42; synthetic diamonds and,

140-42,147,148,150-52 Great Rift Valley (Africa), 5, 15 Greece,
ancient, bronze in, 27 GR-S Program, 101 Guttapercha, 55 Gypsum, in
plaster, 32-33

Hall, Martin, 57-58 Hanford Nuclear Reservation, 264 Hannay, James
Ballantine, 139 Harvard University, 125, 156 Hayashi, Izuo, 202 Hazen,
Robert, 139 Heat.  See also Fire; Temperature Heisenberg, Werner, 81
Hematite, 19,28 Heroult, Paul L. T."  57-58 Hitachi, 213 Hittites,
29-30

Hoechst, 164 Hubble Space Telescope, 198-99 Hughes Research, 205-6
Hutchinson, Al, 221, 224 Hyatt, Charles, 70 Hyatt, Isaiah, 53 Hyatt,
John Wesley, 51-54,60 Hydrogen atoms, 83,132,146-47 Hydroxyapetite,
126

1. G. Farben, 61 IBM, 109,"4,"8-19,154,206,208,

218,233 lee cream, superalloy, 253 ICI, 164 Impact ionization, 208-9
Indium phosphide, 220 Industrial Revolution, 4142 Infrared lasers, 200
Infrared light, 195,200 Institute of Physical Chemistry (Moscow), 146
Insulation: asbestos, 10, 262; polymers as, 55; synthetic diamonds as,
143 Intelligent material systems, 177-78 Intercontinental ballistic
missiles (ICBMs), 94 Interdisciplinary Laboratories (IDLs),

94-95,138,171 Iran, plaster in, 32-33 Iraq,31 Iron, 19; cast iron,
42-43,107-S; disadvantages of, 28-29; Industrial Revolution and, 41-42;
metallography and, 40-41; paint and, 19-20; pig iron, 41,4345,46,61;
smelting, 28-30, 42-45, 46-47; transformation into steel, 13,
29-30,42-45; uses of, 46; wrought iron, 49, 44, 46 Iron Age, 24 Iron
and Steel Institute, 47 Iron Bridge, 41 Isoprene,51,159
Ivory,51-52,54

Japan: Samurai sword makers, 13,30,

144; synthetic diamond research, 148-50,151 Jet engines, 89,91 Journal
of Intelligent Materials, Systems and Structures, 176 Journal of
Materials Research, 100, 263

i n d e x

Journal of the American Chemical

Society, 264 Junction transistors, 193-94

Kamerlingh Onnes, Heike, 108 Kaolin, 19 Kazarinov, Rudy, 208 Kelly,
Mervin, 192-93,213 Kennedy, John F."  101 Kevlar, 108,259 Kinetics, 79,
131-32 Kingery, David, 36,37 Koshland, Daniel E."  153 Kroemer, Harry,
205-6 Kroto, Harold W."  180

Lang, Dave, 225 Laser ablation mass spectrometry, 114 Lasers: in
communication networks,

194,199-200,202,205-8,217-27; infrared, 200; invention of, 194; quantum
cascade (QCL), 217-27, 234; solid-state, 199, 202, 205-6,208,
217-27,234 Latex rubber, 49-50, 55, 158 Laudise, Robert, 263 Lavoisier,
Antoine, 46, 107, 108 Lawrence Livermore National Labora tory,152-53
Lead: paint pigments based on, 10; soldering with, 262; transformation
into gold, 8, 13, 25 Leucippus, 78 Levi, Primo, 56 Life s Devices
(Vogel), 161 Light detectors, 20813 Lilienthal, Otto, 161 Lime, 33,3647
Limestone, 33, 45 Liquid crystal polymers, 183 Liu, Amy Y."  13041
Low-energy electron loss spectroscopy,

114 Lucent Technologies, 191, 194, 198, 263 Lucite, 61 Luminescent
Materials, 264 Lyles, Brian, 248

Malachite, 19-20 Manganese dioxide, 19 Manhattan Project, 10, 88, 93,
101, 158 Mann, Stephen, 168

Margolus, Norman, 132-33 Mark, Herman, 76-77 Marron, Michael, 164
Martensite, 70 Massachusetts Institute of Technology (M.I.T.),
88,132-33,23141 Materials Research Center (MRQ,

24041 Materials Research Institutes, 90, 91 Materials Research
Society

(MRS), 95,

100,135-37 Uate?ials Science and Engineeiingfor the 1990s (Blue Book),
101 Materials science and engineering (MS&E): analytical methods in,

111-15, 127; atomic and molecular structure in, 3741, 65-83; early
process controls in, 36-41; emergence of field, 8-10, 15-16; evolution
of field, 87-103, 106-11; impact of fire on, 18-19, 21-33; limitations
of, 71; materials-by-design paradigm in,
187-89,217-27,234-55,257-65;sociological.  innovation in, 60-64;
synthetic methods in, 115-27; theory in, 127-33; trial-and-error
approach in, 12-13,27 MBE machines, 119,122-25,210,212,

213, 214, 217, 235, 253-54; applications of, 204-5; described, 202-5;
operation of, 203-4; quantum cascade lasers and, 220-25.  See also
Molecular beam epitaxy (MBE) Mechanical relays, 192,193 Mesopotamia,
31, 33 Metallography, 46, 47; improvements in,

70-72; origins of, 40-41 Metallurgy, 13, 23-30; atomic and molecular
structure in, 3741, 65-75; bronze, 26-28; copper, 24-26; early process
controls in, 36-41; heat in, 23-30,"6; iron, 26,28-30,4245, 4647;
origins of, 24-26; steel, 45-46, 110, 230-55.  See also Smelting
Metals, 23-30; characteristics of, 23-24; elements in, 67; first use
of, 24; internal structure of, 38, 39; relative importance of, 6-7;
shape memory 178-79,251-52; smelting, 24-30; strength-to-density ratio
of, 1074.  See also Alloys; names of specific metals i n d e x

Methane, 145-46,148 Michelin, Andre, 51 Michelin, Deoudard, 51 Michigan
Molecular Institute, 117 Microelectronics: emergence of, 83-86,

89-90, 93; impact of, 111.  See also Transistors Midgely, Thomas,
9,261-62 Mirrors, 32 Molecular beam epitaxy

(MBE),

119-25, 194-95, 202; applications of, 122-25; development of, 119-22.
See also MBE machines Molecular mechanics, 131-32 Molecular structure:
biological mate rials and, 126-27, 157-58, 160-63; control of size in
polymers, 117; of ethanol, 128-29; origins of theory of, 38-39; of
polymers, 76-78; ring shaped molecules and, 125; synthetic materials
and, 117-18.  See also Atomic structure; Metallography; Quantum
mechanical theory Molecules, 38, 39 Molybdenum, 247 Montages, 70
Mother-of-pearl, 54 Murano, 13,35 Mushet, Robert F."  45

Nanoscale particles, 183-85 Napoleon 111, 42,55 National Academy of
Sciences, 151 National Aeronautics and Space

Administration (NASA), 95, 132, 230,238,239,247,248 National Aerospace
Plane (NASP), 132,

259-60 National Bureau of Standards, 73 National Defense Preparedness
Association 137 National Institute for Research on

Inorganic Materials (NIRIM), 148-50 National Institute of Standards
and

Technology, 73 National Materials Laboratory, 91 National Research
Council (NRC), 151,

155 National Science Foundation, 95,237 National Security Agency, 199

Natron, 31 NEC, 213 Neodymium-iron-boron-based magnetic materials, 110
Neolithic (New Stone Age), 21-24 Neoprene, 62, 78 Neural network
processors, 176 Newman Haas 248-50 Newton, Isaac, 13,25,82 Nickel, 46,
102 Nickel-cobalt steel, 242-44, 246-47 Nickel steels, 45
Niobiumgermanium intermetallic material, 109 Nitric acid, 53 Northrup,
103 Northwestern University, 92-95,138,

166,181,240-41 Norton Company, 154 NTT, 202,213 Nuts, 165 Nylon, 4,
62-63, 78, 233

Obsidian (volcanic glass), 17,30-32 Ocher, 19 Office of Naval Research,
90-91, 143,

151,164 Olson, Greg, 166,231-55,257-58,260,

261 Olson Diagrams, 242-44, 249, 253 Optical fibers: in artificial
nervous systems 172-74; in communication networks, 200, 201-5, 208;
improvements in, 260; infrared light and, 195, 200; light detectors
and, 208-13; synthesis of, 116,201-2; synthetic diamond protection for,
142; transparency of, 109-10 Optical microscopy, 111-12 O-rings, 238
Oxygen: in aluminum formation, 57-58;

in iron smelting process, 43-44, 45 Ozone, chlorofluorocarbons and, 10,
254

Paint: earliest forms of, 19-20; lead based, 10.  See also Dyes
Paleolithic Age (Old Stone Age), 16-20 Palestine, glassblowing in.  31
Panish, Morton, 202,206 Pantheon,33 Paper, origins of, 34

i n d e x

Papyrus, 34 Parity checkers, 216 Parkes, Alexander, 54 Pennsylvania
State University, 94-95,

143,151 Penzias, Arno, 225 Periodic table, 65-69, 188, 189, 199-200
Periodic Table, The (Levi), 56 Phase diagrams, 145 Pheland & Collander,
51 Philosophers stone, 13 Phoenicia, bronze in, 27, 101 Phosphorus, 5,
44-45, 47 Photochromic glass, 171-72 Photography: celluloid in, 54;
collodion in, 52-53 Photolithographic materials, 106 Photomultiplier
tubes (PMTs), 213-16 Photonic materials, 117 Photovoltaic cells, 200
Piezoelectric materials, 175-76 Pig iron, 41, 43-45, 46, 61 Pittsburgh
Reduction Company, 58 Planck, Max, 79 Plaster, 32-33 Plexiglas, 61
Pliny, 31 Plutonium, 264 PMT (photomultiplier tubes), 213-16
Point-contact transistors, 83-85, 193 Polyamides, 63 Polyethylene, 61
Polyethylene terephthalate, 61-62 Polyisoprene, 159 Polymer-reinforced
hydrate ceramic composites, 253 Polymers-elastomers: development of,

55, 61-64, 76-78; early uses of, 55; elements in, 67; liquid crystal,
183; molecule size in, 117; origins of, 49-55; properties of, 77;
protein based 169-70; relative importance of, 6-7; strength-to-density
ratio of, 107-8.  See also specific types Polymethyl methacrylate, 61
Polystyrene, 61 Polyvinylchloride, 61 Pompeii, 33 Porcelains, 36-38
Portland cement, 33 Potassium, reactivity of, 57

Pottery, 18-23; advantages of, 21-22;

disadvantages of, 22-23; first use of, 18-19,24 Pratt & Whitney, 252
Priestly, Joseph, 49 Protein-based polymers, 169-70 Protein Polymer
Design, 164 Pumice, 17 Pyrometers, 37

Quantum cascade laser (QCL), 217-27,

234; applications of, 226-27; development of, 217-26 Quantum corrals,
122 Quantum mechanical theory, 62, 78--86,

128-30; band-gap engineering and, 195-201, 216-17; birth of, 79; impact
of, 79-80; Newtonian physics versus, 81; solar system model in, 80-81;
spectroscopic observations and, 79-81; wave functions in, 81-83 Quantum
wells, 206-8,220 Quartz, synthetic, 263

Racing cars, 248-50 Rand-McNally building, 46 Rayon, 62, 77 RCA, 205
Reaumur, Reno Antoine Ferchault de,

38-39,64 Red October (Clancy), 178 Refrigeration, 9 Reichenback, Henry,
54 Rice University, 180 Rivets, smart, 178 Rockwell International,
238-39 Rockwell 70,249 Roebling, John, 46, 101 Rogers, Craig, 174-78
Rohm and Haas, 61 Rohrer, Heinrich, 114,"9 Roman Empire, cement and, 33
Roy, Rustum, 95,143,151,160-61,169 Rubber, 49-51; latex,
49-50,55,158;

synthetic, 62, 78, 101, 158-59; uses of, 50-51; vulcanization of, 50,
51

Samurai sword makers, 13, 30, 144 Sarikaya, Mehmet, 126,160,165-67,250
Scanning electron microscopes (SEMs),

102,"2-13

i n d e x

Scanning probe microscopes, 114-15,

119 Scanning tunneling microscopes (STMs), 114-15,"8-19,120-21
Schrodinger, Erwin, 82--83, 128, 137 Schweizer, Erhard K."  120 Science
magazine, 153, 164, 226 Screw dislocation, 74-75 Seiko, 154 Seitz,
Russell, 156 Semiconductors, 5,198-201; changing composition of, 205;
in communication networks, 194-95, 198-201, 205; development of, 88-90;
diamonds as, 143; elements in, 68; three-five, 181-82,199-200; two-six,
181-82, 199-200 Setaka, Nobuo, 148-50 Shape-memory metals, 178-79,
251-52 Shockley, William, 93, 193, 205 Siemens, William, 47 Silica,
36-37,201 Silicon, 5; in semiconductors, 198, 199;

as superconductor, 130-31 Silicon chips, transistors on, 105-6 Silicon
Valley, 5, 93 Silk, 62-63 Silver, 56 Sirtori, Carlo, 218,224 Sivco,
Barbara, 218-21 Skin, 159-60; smart, 172-74 Slag, 43, 45 Smalley,
Richard E."  180 Snmrt Materials, 176 Smart materials and structures,

170-79; biological metaphors in, 172-74, 176; defined, 9; gallium
arsenide (GaAs), 124, 182-83, 198, 199, 202, 211-12, 218; intelligent
material systems, 177-78; liquid crystal polymers, 183; nanoscale
particles, 183-85; photochromic glass, 171-72; piezoelectric ceramic
and plastic materials, 175-76; shape memory metal, 178-79,251--52;
smart beams, 177; smart cylinders, 176; smart rivets, 178; smart skin,
172-74 Smelting: bronze, 26-28; copper, 24-26,

29; iron, 26, 28-30, 42-45, 4 7; steel, 42-46, 110.  See also
Metallurgy

Smith, Cyril Stanley, 10, 19,26,74,93 Society for Cellulose Research,
77 Sodium, 31, 5647 Solder, lead, 262 Solid-state devices: lasers,
199,202,

205-6, 208y 217-27, 234; photomultiplier tubes (PMTs), 213-16; physics
of, 76; transistors, 83-86, 89-90, 105-6,193-94 Solid State Sciences
Advisory Panel, 91 Sonar surveillance, 178 Sony, 153-54 Sorby, Henry
Clifton, 4041,46,47,64.

69,72Y 75, Ill South Carolina University, 176 Soviet Union, synthetic
diamond research,146-49 Space exploration, 89, 101, 132, 230,

23849,244-48,259-60 Spectroscopies, 79-81, 112-14 Sports-car racing,
248-50 Stainless steel, 45-46, 241, 244 State University of Blacksburg,
172-74 Staudinger, Herman, 77-78 Steel, 229-55; alloys of, 45-46,
110,

230-55; band-gap engineering of, 234; Bessemer process, 42-45;
empirical approach to making, 232-33; evolution of chemistry and,
46-47; Industrial Revolution and, 4142; as king of metals, 58; mass
production of, 46; materials-by design paradigm for, 234-52;
metallography and, 4041; self-healing, 250-52; Siemens process, 47;
space exploration and, 230, 238-39, 244-48; sports-car racing and,
248-50; stainless, 4546, 241, 244; structure of, 38, 236-87; super
alloys 242-53; transformation of iron into, 13, 29-30, 42-45; uses of,
46 Steel Research Group (SRG), 236-41,

245-47,250,251,253,259 Steel Resource Center, 240-41 Stoddart, Fraser,
125-26 Stone tools, 5, 12, 13, 15-21 Stookey, Donald, 171
Strength-to-density ratio, 107-8 Stupp, Samuel, 117 Submarines, 178
Sulfur, in vulcanization of rubber, 50

i n d e x

Superalloys, 108,242-53; chromium if ,

246; development of, 245-50; molybdenum and vanadium in, 247; nickel
cobalt steel, 242-44, 24 7; self-healing, 25052 Superconducting
supereollider,.152-53 Superconductors 9 233; ceramic, 109;

discovery of superconductivity and, 108-9; elements in, 68; silicon as,
130-31 Superdiamonds, 154-57 Superiattices, 211, 212, 213 Suris,
Robert, 208 Swiss Metallurgical Company, 58 Synthetic diamonds, 11(),
139-57, 179;

applications of 142-43,152-54; atomic hydrogen and, 146-47; from
chemical vapor deposition (CVD), 144-4S, 151-54; early research on,
139_4@, 148-44; environmental cOncerns for, 262; General Electric
Company research on, 139-42,144, 145, 155-57.  importance of, 151-53;
Japanese research on, 148-50, 151; Soviet Research on, 146-49; super
diamonds 154-57 Synthetic quartz, 263 Syria, glassblowing in, 31

Teflon, 61 Temperature: heat in metallurgy, 23-30,

116; impact on materials, 108; technological disasters and, 238-39;
thermodynamics and, 13142,145, 245-46.  See also Fire Terminator 2:
judgment Day (film), 250 ThermoCalc, 245-46 Thermodynamics, 13132, 145,
245-46 Thomas, Sydney Gilchrist, 45 3M, 263 fin, in bronze, 13, 26-28
nssue Engineering, 264 Titanium-aluminides, 9, 251-52, 259-60 Toffoli,
Tomasso, 132-33 Tomalia, Donald, 117 Tortoise-shell, 54 Townes,
Charles, 227 n net Transistors in communicatio works, 193-94;
development of, 83-85,89-90,93; junction, 193-94; and the miff
oelectronics revolution,

83-86; point-contact, 83--85, 193; on silicon chips, 105-6
Trial-and-error approach, 12-13, 27 Trotter, Donald S."  172 11
schirnhaus, Ehrenfried Walther von,

36-37 Tsu, Raphael, 119, 206,208,211 Tungsten carbide, 110
Ungsten-manganese-steel alloys,

45 lbrkey, 33 Tutankhamen, 28

union Carbide Corporation, 143-44 University of Alabama, 160, 169-70
University of Arizona, 36,165 University of Bath, 168 University of
Berlin, 79 University of California at Berkeley,

130,155 University of Illinois, 117 University of Pennsylvania, 112,
138 University of Reading, 160,177 University of Sussex, 180
126,160,250 University of Washington, Urry, Dan, 160,169-70

Vacuum tubes, 105,194 Valence band, 198 Vanadium, 247 Vascular
material, 15940 Venetian glass, 13,32,35 Venus of Vestronico, 18-19,24
vestronice, Czechoslovakia, 18-19,

24 Vickers Company, 60 Vincent, Julian, 1601161,177 Virginia
Polytechnic Institute, 172-74 Vogel, Steven, 161 Volcanic ash, in
cement, 33 VPI, 174 Vulcan, 50 Vulcanization, 50, 51

Wedgewood, Josiah, 37 Wentorf, Robert, Jr, 139-40, 150 Western
Electric, 191 Whitesides, George, 125 Wilm, Alfred, 5940,73 Wood, 11,
12 World War 1, 71

294 N index

World War II, 63,75,87,88,90-91, York, Herbert, 94

158-59,173,175,191 Yttrium, 9 Wrought iron, 43, 44, 46

Zagros Mountains, 25 Xenon, 119,120-21 Zeppelin,60 X-ray
crystallography, 72-73, 76, Zhoukoudian cave (China), 18

111-12 Zinc oxide ceramics, 112-13