[MUSIC PLAYING] PRESENTER: Well, in case anyone
here doesn't happen to know, Cliff Shull is the 1994
laureate in physics. [APPLAUSE] Cliff is sharing the honor with
Bert Brockhouse from McMaster. And the Nobel citation,
in fact, quoted Cliff as being the one who told
us where the atoms are, and Brockhouse telling
us what they do. Cliff was born in
Pittsburgh in 1915. There is a-- perhaps
apocryphal-- story that his great skill and
intuition in the laboratory came from his family hardware
and home repair business in which he was quite active,
I understand, as a youth. Got his degree at
Carnegie Tech in 1937, and went off to NYU for
his graduate studies. There, he got into
the beginnings of what you might call
nuclear technology with an MIT connection, as during
his thesis time, he built a Van de
Graaff generator. And of course, Van de Graaff was
a professor here at that time. He did not, however, work
with neutrons at that time. And in fact, during
the war, as well, did not get to work with
developing neutron technology, as he was too valuable
in his work with Texaco-- or what is now Texaco-- working on high-octane
aviation fuel during the war. This time was productive,
however, of course, in a different way. In 1941, he married Martha, with
a very important partnership-- 50-plus years and going on. He did get to work with X-ray
diffraction techniques, which presumably played a
major part in seeding his subsequent
activities with neutrons. After the war, Cliff went
to Oak Ridge, and using the large graphite reactor-- which was developed there-- began his study of
neutron techniques, working on that for many
decades, and of course, resulting in the prize
which he deserves today. I will not get into
the work with neutrons. You'll hear that from Cliff. You'll hear some further
remarks from other people. In fact, I'll just end with some
departmental personal remarks. First of all, I think that
Cliff's colleagues have a deep appreciation to Cliff
for his many contributions to the department,
his pleasant aspect, his dedication to high standards
and to re-doing experiments. And everyone is
extremely delighted with this recognition. Indeed, Cliff won
The Buckley Prize-- a very prestigious prize in
condensed matter physics-- in the mid-50s, and this is
a 0-year overdue recognition by the Nobel Committee. Cliff, with his
characteristic modesty, I think is genuinely
surprised by the news he received a few days ago. And I would just
say that many of us were far less surprised than
he was, since we all feel very, very strongly this
was a very timely and a very, very
well-deserved prize. So we're going to have a few
remarks now before Cliff gets his chance to tell
you about neutrons and how he helped develop
this important field. Chuck Vest, the President,
will say a few things. The Dean, Bob Birgeneau,
who is one of our most active, certainly,
neutron-scatterers today will add something. And then finally, one of
Cliff's students, Dave Moncton, who has flown in from Argonne,
will give a brief perspective from the students' view. So, Chuck. VEST: Thank you
very much, Ernie. I think that moments
like this are extremely important in the life of a
great institution like MIT, and certainly are extremely
important in the life of science period. It's a great honor
to be associated with people who stand
for such excellence in their scientific work and
in their being as humans. There are two things that
particularly struck me as I listened and
talked to people-- I must admit, from a
distance, because I've been out of town the last two days-- and had to sense all this
through the newspapers and the television. And running into a couple
of people, including a former physics student
here from the Department, in the Washington area. The two things that struck
me the most were first-- every single person I have
talked to about this award, within the first one
or two sentences, has used the word "nice." Professor Shull is known
as a nice person, which is another way of saying a warm
human being and somebody who is viewed as being very
supportive of his students and his colleagues, and to
the Institute community. The second thing
that really struck me was a quote in the newspaper. It may have been
from Bob Birgeneau. I'm sorry if I'm attributing
it to the wrong person-- that said that they had
worried that the revolution that Professor Shull
and his colleagues brought about in the use of
neutron scattering to measure properties of
materials and so forth had become so standard,
so well-known, so much the stuff of textbooks
that they were afraid that the person who originated
it might be forgotten. That is a very strong and
very important statement, and I think that one of the
terrific benefits of the Nobel Prize recognition is
that it reminds people-- and particularly young people-- that there are people
behind science, and it is their passion, their
enthusiasm, their commitment to excellence and belief in
the importance of science that really drives it forward. So we thank you very much-- to
use a rather hackneyed phrase, I suppose-- for being such a fantastic
role model for those who follow behind you, Professor Shull. And it's just a very,
very great honor to be a member of
this MIT community and to share this
moment with you. Thank you very much. [APPLAUSE] Bob. BIRGENEAU: As many of my
friends out there are know, this prize is a really
extraordinary pleasure for me, because actually, I began my
research career as a summer student in Bert Brockhouse's
group, Chalk River. And it was because
of Bert Brockhouse-- who also is a nice,
wonderful person-- so this is extraordinary that we had
two such special human beings recognized with one prize. And in fact, my
first publication was with Brockhouse's group,
and embarrassingly, it's still my most referenced publication. [LAUGHTER] After having got
launched by Brockhouse, I then meandered around
the world, actually, and then finally
ended up here at MIT. And while I was learning about
this field from Brockhouse and Brockhouse's
former cohorts, they had told me how there
was this genius who had created the field
in the first place by the name of Cliff Shull, who
was then at MIT, so it was one of my goals to come
to MIT and to meet the great person himself. And I remember first walking
down the hallway the first time I saw Cliff, and Cliff was
so nice and so self-effacing, I couldn't believe-- especially in the
MIT environment-- that this modest person-- [LAUGHTER] --was this great hero
that I had heard about. But indeed, it was. And actually, we had
many years together. We served on a number
of academy committees together, flying back
and forth to Washington. And we shared one extreme
vice, which is both Cliff and I like to smoke cigars. And Cliff had the wisdom to
retire before it was banned. But I presume--
actually, there's still a residue in
his office if you go into his office,
which is now occupied by my and Marc
Kastner's post-docs, who are hoping that some of this
greatness rubs off on them. Besides being a
wonderful colleague and an outstanding
scientist, Cliff also was excellent as a undergraduate
and graduate student teacher, and as a graduate mentor. And he educated many of
the leaders in the field. One of the premier of
those is Dave Moncton, who's Associate Director at
Argonne National Laboratory, and who is Head of the
Advanced Photon Source, which is this wonderful new source
which we and many others hope are going to advance the kind
of science that Cliff has done-- albeit using protons instead
of neutrons, in that case, a level further. So maybe Dave can
make a few comments from the perspective
of someone who used to sit here as
a graduate student. [APPLAUSE] MONCTON: I'm really
overwhelmed by the privilege of being able to stand up
here, have the opportunity to thank Cliff
personally in front of this particular
audience for all that he gave to me as a student. I'd like to speak mostly to
the students in the audience. I guess the faculty members
can listen in if they'd like. The first contact
that I had with Cliff was when I was still
a senior at Cornell. I had applied to only one
school as an undergraduate, and realized that that might
be risky as a graduate student. So I had applied to
a number of places, and I was fortunate to have
gotten accepted at all of them. And I was fortunate, also,
to have gotten scholarships at all of them. So I was having a hard
time making up my mind until a letter
arrived in the mail, and it was a personal
letter from Cliff, who had been busy in the
graduate school office looking over the applications
of graduate students and personally writing to
those people he thought would fit into his laboratory. And there was no
question in my mind, when I read that
letter, that this was where I wanted to work. I don't think I took
my cap and gown off before I got in the car and
drove from Cornell to Cambridge and began working in
the summer of 1970. I think the 1st of
July, essentially. I was so excited about a real
laboratory and real experiments that I couldn't think
of anything better to do that summer. That's an example of Cliff's
fastidiousness, I suppose, that has become legendary. And all of his students
will tell you about that, and it's a gift that I think
he's given all of his students. The other night,
my wife and I were arguing why it was that
when I go to sleep at night, I like to have all the
three-way switches in my house all in the down position. You don't need to have them
all in the down position. As long as they're
up in pairs, then the lights will
all still be off. [LAUGHTER] But it seemed to me that that
was not fastidious enough. I mean, after all,
if someone came in in the middle of the night
and turned on a light and turned it off,
I mean, you'd know if you remembered
where you'd left them. [LAUGHTER] So throughout my career
as an experimentalist, I always tried to emulate. I never successfully emulated
that level of fastidiousness that Cliff brought to bear
on experimental science, but I never regret it. Any time I did something,
I lined a goniometer up to be a 0 of angle when it was
perfectly vertical or perfectly horizontal. It didn't have to be
defined as 0 there, but that just gave you some clue
as to whether it had been moved inadvertently, or the
motors had slipped or what have you, because
you could line it up with various other vertical
clues in the horizon. And those were the
kinds of things that Cliff just did routinely. During my-- I think
early in my second year, Cliff thought it would be a good
exercise for a student to write a Nobel Prize recommendation. So he's told me that he would
like to see Brian Josephson get the Nobel Prize, and would
I go off to the library and take a crack at writing
the case for Brian Josephson? Which I was very enthusiastic
about, and I did that, and I gave it to Cliff. And he puckered his
lips as he usually did and held his pipe
out and said thank you. Of course, that was
the last I saw of it until Brian Josephson,
a few months later, got the Nobel Prize. And I thought, "That was easy!" [LAUGHTER] Little did I know what work goes
on in getting Nobel prizes-- on either side of the equation. The one other gift-- and the most important
gift that I think Cliff gave me as a teacher-- was I guess what I call the
gift of self-confidence. As an experimental
physicist, it's very important-- there
are lots of bleak moments. There are long periods
between good results, and even longer periods
between excellent results, and longer between
publishable results, and longer even between
results that people want to read about
when you publish them. [LAUGHTER] And in those long
waits, you have to be confident that
you have something to bring to bear on the field-- something to offer. And you need to have a lot
of confidence in yourself. And what Cliff did
was, to a large extent, he left me alone, because
he had what I believe-- he can correct me when
he gets his chance-- what I believe is a lot
of confidence in me. I didn't work on a
problem that Cliff gave me as a thesis advisor, and
that was very unusual. And I didn't even
work in his own lab, finally, on my own thesis. I went to Brookhaven and
worked at their reactor. And throughout the entire period
of my graduate school, he was-- especially in the
later years-- almost a secondary figure in
my day-to-day life. But this tremendous confidence
that he had in me translated-- some magical fashion-- into my own self-confidence
as an experimentalist. So it's not the-- I don't know-- the book
learning, in a sense, that I took away from my
graduate education with Cliff. I took it away from MIT,
because goodness knows we had many, many courses. But what Cliff gave me-- and
I think all of his students-- was that self-confidence that
he himself-- and you probably won't get him to admit it, but
he has it in great quantities. Quantities enough that
he is able to share it with so many of us. So thank you, Cliff. [APPLAUSE] PRESENTER: OK,
without further delay, Cliff will tell us
about the development of neutron scattering. SHULL: Thank you, gentlemen. As Ernie has mentioned-- and Bob-- the field
of research that I want to mention this afternoon
goes back a long ways; back to the immediate years
following World War II, back in the late 1940s. And this research deals
with the development of techniques that can utilize
slow neutron radiation. Now, neutrons were
first discovered in 1932 by Chadwick in England. And Chadwick found that if you
take an alpha particle source-- and essentially, the only
alpha particle sources known at that time were from
naturally-occurring radium sources. If you surround a radium
sample with some material-- most usually beryllium-- you find that, emanating
from the beryllium, is an intense radiation alpha. And that was later
investigated by Rutherford and other people-- classic people-- and
labeled alpha particles. Now, it was also found, shortly
after Chadwick discovered-- when Chadwick was arranging
such configurations-- alpha particles sources--
he found that if you further surrounded that source
with other media, that there seemed to be
something coming away from the other media which
was certainly different than the alpha particles
that were originally released from the inner piece of radium. And on investigating
this, he found that it was a very
penetrating radiation, and that was the
surprising thing to him. It's radiation that could
pass through copious thickness of the material, and there was
no known radiation at the time that would do that. So with further studies
of the production of this anomalous radiation
with different materials surrounding the alpha
particle source, he came to think of this as
being some neutral particle type of radiation. The reason being that
a neutral particle would be expected to have
much greater penetration through media than any sort
of charged particle radiation. It was ruled out that it
was an electromagnetic type of radiation from
certain observations. So if it had no charge-- if it were radiation
without charge-- you could think of
it as being somehow neutral in an electrical charge. And he gave it
the name neutron-- the neutron radiation. Well, this was very exciting
to nuclear physics-- what there was of nuclear
physics at the time. Nuclear physics was
very rudimentary then. And further studies--
mostly by Fermi in Italy-- showed that this radiation could
be effectively thermalized by, again, passing it
through and letting it bounce around in
some other medium. And if you observe
what comes out of this third piece of medium,
he obtained what he thought-- what he classified-- as
thermal neutron radiation. How could he
determine that it had this sort of characteristic? By being thermalized, it
implied that the energy was down to thermal energy levels-- thermal energy levels
that characterize atom emission of radiation--
thermal energy radiation. Well, one thing-- it was known
that captured cross-sections-- absorption cross-sections
of materials-- would be expected to follow
a certain energy dependence. The lower the energy, the
greater the absorption. And that, indeed,
was what he found-- that if he put more
and more media there, did more and more acting
upon this neutron radiation, it lowered the energy
of the radiation so that you could observe a
larger absorption cross-section in something else. So this process could
continue until he obtained what he thought was real thermal
energy neutron radiation, where the neutrons had been degraded
in kinetic energy down to a thermal energy level of
some fraction of one electron volt-- some fraction of
one electron volt. So this thermalization concept
was a very important one. It was shortly
recognized that if you had radiation of
such a low energy, that this radiation would also
exhibit a very anomalous wave character. And this goes back to the
heart of quantum mechanics. A particle moving along
is to be considered also as a moving wave. And there's a periodicity
associated with this wave. This first was suggested in
earlier years, most notably by de Broglie-- de Broglie-- who suggested
that there was a wave length associated with moving
electron radiation. Moving electrons should also
exhibit a wave character. And it wasn't long
before people did experiments that
demonstrated sure, indeed, a moving electron does exhibit
a periodicity and a wave length, and that was demonstrated by
diffracting low-energy electron beams from crystals, just
as people with X-rays had demonstrated
many years earlier. So there was this weight
periodicity associated with moving electron particles. Now we have neutron particles
that are moving slowly, and they should have a
characteristic periodicity also. The question comes,
do these neutrons also show diffractive
effects, the same as had been observed with
electrons some years earlier? So there were two experiments--
two experimental groups-- that set out to establish-- to look for this periodicity;
this de Broglie wave character for this neutron radiation. Von Halban and Preiswerk in
France, and Mitchell and Powers at Columbia University
in the United States. Both pursuing
experimentation in 1936. And they both found, indeed,
that this periodicity was there-- that there were
diffractive effects-- excuse me. --from crystalline material
when this neutron radiation-- thermal neutron radiation-- was
applied to these diffracting objects. So now we have a wavelength. We have a kinetic energy. We have an energy-- a kinetic energy--
to be associated with these particles. Well, the experiments
of demonstrating this diffractive effect
were so complicated, and the intensities
that were involved were so small that one could
hardly think of exploiting it as a diffractive tool. X-ray diffraction was in a
well-defined, well-developed state by that time,
and many people were diffracting
X-rays and learning lots of things about crystals
and materials in general. One could think that
maybe neutrons could also serve that purpose, but the
intensities were frighteningly low, and nobody worried about
doing anything seriously in that direction. That all changed a couple
years later in 1939, when nuclear fission was
uncovered by Hahn and Meitner in Germany-- 1939. Now, the exciting thing about
nuclear fission is this. You bring neutrons onto
certain fissile nuclei-- most notably uranium;
normal uranium-- and you find that the
neutron absorbs the neutron and breaks up into fragments,
and more than two neutrons come off from that. One neutron in, more
than two coming off. Now, this conclusion
didn't come automatically. It came with a series of
different types of experiments. But that was very,
very interesting. I put in something, and I get
over twice that same thing out. That immediately suggests
that this process could be used in a cascade manner to
produce more than you feed in, and the process could
skyrocket, cascade, and become cataclysmic-- with more coming out
than you're putting in. Well, it wasn't long after that
that we became engaged in war years, and it was proposed-- thought very
seriously by people-- that maybe this could be
developed into some sort of weapon-- an atomic weapon. Maybe you could have a
configurational system that could be triggered
somehow or other, and it would just diverge. So many people became
involved in that-- Fermi himself, and
many of his colleagues, and very well-known
people became involved in that
through what was called the Manhattan Project,
set up in the early war years. It was finally
demonstrated, in experiments at the University of Chicago-- at the Stagg athletic field-- that, indeed, you could
have a cascading system-- a reproducing system--
that you can think of. The idea was to collect enough
uranium and enough graphite together. The graphite was to serve
as a moderating medium for the extra neutrons that came
from the uranium fissioning, and the graphite would
cascade the energy down until those neutrons would
then cause further fission events in other uranium. Giving, again, two
for every one-- more than two for every one. And that process would
continue and build up. Well, the chance of success of
obtaining such a configuration, we recognize, would increase
as the physical size of this configuration
would get larger, because volume effects go
up as the cube of the size, whereas surface area-- which determine
leakage from a system-- go up only as a
square of the area. So there's an advantage
in increased size. So the test was
going to come when one could get a large enough
configuration of uranium and graphite collected
together into a system. This was obtained,
and it was detected. It was very carefully monitored
with sensitive equipment in this first experiment
at Stagg Field at the University of Chicago. That occurred in December 1942. December 1942. Once it had been demonstrated--
once Fermi had demonstrated that, indeed, you could have a
self-reacting system this way-- the experiment was shut down. They didn't want it
to get out of control. Shut down, and
people immediately started to design
facilities for using that. The thinking at that time was to
construct a reacting system off somewhere else
that would produce some isotopes which were badly
needed during the war time years-- most notably isotopes
of plutonium, which were considered to
be useful in atomic weapons production. So they wanted to build
this test reactor-- which would operate at a
considerable power level now-- as a production test facility
for even larger facilities which were to be
built elsewhere. It was decided to
position this test reactor at Oak Ridge, Tennessee-- when the army had absorbed
a large area there for industrial use. And this reactor
was put on paper. Construction was
carried through, and was completed in an
amazing period of 11 months-- December 1942-- until the
first use of the reactor in November 1943. 11 months in total. And this was all 500
miles away from Chicago, where the first experiments
were carried out. So a whole flock
of physicists were conveyed and sent from Chicago
down to the Oak Ridge area-- and engineers galore,
and construction people. And this amazing development
was pushed through. So this reactor
started to operate at Oak Ridge in November 1943. And I'll show, shortly,
a photograph of that. But let me first project this. This is a chronological
listing of the starting periods of the very early piles-- they were called
piles back "Reactor" as a term came into
vogue later on. Piles. This Oak Ridge graphite pile--
which I was just describing its start-up date was in 1943-- at the same time as the
construction of this reactor was going on-- designing and construction. Another one was under
construction back in Chicago, in what was then called
the Argonne Laboratory. And it differed from the
Oak Ridge graphite pile. This should have actually been
written Clinton graphite pile. Clinton is a town in
Tennessee which is the nearest town to the Oak Ridge facility. So at that time,
this was all Clinton, and Oak Ridge came in later. The Argonne reactor was to
differ from the graphite pile in the Oak
Ridge area by having heavy water instead of graphite
as a moderating medium. The heavy water was to
slow down the neutrons and cause, again,
fissioning in uranium-- the same as in this first
pile, which had graphite as the moderating medium. At later times, in '47,
the Chalk River reactor came into operation. And this is the one
where Dr. Brockhouse did all his research work-- up at this Canadian
government facility. And then later ones in England. This is at England, and
another one at Oak Ridge, and at Brookhaven on Long
Island in this country. My next projection shows a
photograph of this graphite reactor at Oak Ridge. The reactor is a
very large assembly-- a cubic assembly, roughly
20 feet on an edge-- of stacked graphite-- pure
graphite-- components, what were called
graphite stringers. So these graphite stringers were
square, configurational chunks of pure graphite, and
that was a problem to prepare a graphite
of sufficient purity. About 4 or 5 inches on a side,
and 3 or 4 feet in length. Rectangular assemblies,
just stacked together. On the corners of these graphite
sections-- graphite pieces-- there was a diagonal cut. And so when these elements
became stacked together, there would be channels
running down along the length. In these channels, one was to
place the uranium fissioning fuel. The uranium was
pure metal uranium. Many problems associated with
preparing a pure uranium. And the uranium
had to be encased, because uranium will oxidize. And if it gets hot-- which it would in
this operation-- then it would
oxidize very rapidly and become uranium oxide
and become a powder, and essentially collapse. So each of these
uranium pieces, which are roughly an inch in
diameter and 5 inches long-- an inch in diameter,
5 inches long-- had to be encased in
an aluminum shielding can, and sealed up to prevent
air from getting to the uranium and oxidizing the uranium. So we think of this
configuration-- of all this graphite,
tons of it-- stacked up. Not bolted together, or
not clamped together, because the graphite would
become hot and expand, and any banding or
clamping might be broken. And so all of this
graphite with these uranium rods running through
the assembly. Much uranium. I don't have a figure for
the mass of the uranium or the graphite involved,
but you see the scale of it. Now, all of that assembly had
to be surrounded by a shield to prevent radiation
from escaping out to where people might be
operating the assembly or doing something outside. So it had to be
encased in a concrete-- essentially, a concrete
house the whole way around, the concrete being about
5 feet in thickness. 5 feet in thickness. So we see sections
of that around here. We're looking at what's
called the loading face of this assembly. You can perhaps see vestiges of
the channels that are traveling back away from us in the photo. That would represent positions
where the uranium rods were placed, and this loading face
is where those radium rods would have been originally pushed
back into the graphite assembly. So there's a loading
platform here that's on a hydraulic crane assembly. Can go up and down. Here are a couple fellas doing
something at the front face, there. Perhaps pushing some uranium
on back through the assembly. And the control room
is over in here. And for no particular reason,
they can look at the front face here. But there had to be shielding
on this front face as well. Once you had completed
an operation in an area, you had to shield that off. The main shielding
was all around here. At the back side
of the reactor-- the backside-- one
could push these rods-- the uranium rods--
through, and they would fall down into
a plenum chamber and could be collected
together and transported with remote control facilities
over to a chemical processing plant which is next door. The purpose of this
reactor was to produce the first viable
quantities of plutonium. One of the isotopes
of uranium, when it's subjected to this
neutron radiation, transforms itself
into a plutonium-- the next element up in the
periodic table from uranium-- into a plutonium isotope. And from previous measurements
and studies of that isotope, that was potentially a very
good atomic weapon component. Recall we're in the middle-- in the starting period of
the war-- and everybody was desperate for new
developments, here. So the primary
purpose of the reactor was to produce
sizable quantities that the chemists could get at
and purify and study processes, and the physicists could
get out and study processes with the plutonium for
weapons potential use. So there was no experimentation
as such with all this neutron radiation which was bouncing
around inside this facility. There was one experiment that
I know of that was performed, and that dealt with, again,
producing information for further use on the project-- what was called a
pile oscillator. A pile oscillator. Something moves back and forth. The idea is to have a
detector back inside a section of this configuration-- a neutron detector. And you move a sample
that you want to study back and forth in its vicinity. You can determine, with
this device, the capture or absorption cross-section
of this medium for neutron radiation. And that was of great
interest in the technology at that time-- again, for the practical uses to
which the project was devoted. So there was that one
experimental facility, and that's the only use of
the neutrons that I know of, other than to continue
the fissioning process in the uranium
inside the assembly. Well, this was used throughout
the war for producing-- war years, until the war
ended in the summer of 1945-- for producing
quantities of plutonium. As soon as this
reactor was operating, new reactors were under
consideration for construction out in the state of Washington-- at Hanford. Much larger reactors. Much larger piles,
I should say-- which worked to produce
more abundant, more copious quantities of plutonium. So those were under
construction out there when this was operating
for in the first year. But this reactor continued
to operate in that way. The total power
production in this reactor was about 3,500 watts-- 3,500 kilowatts, excuse me. 3,500 kilowatts. 3 or 4 megawatts of
power production. Present day power reactors
operate roughly at 100 megawatts-- 25 or 30 times the operating
power level of this assembly. I didn't mention-- how do you
get this power out of here? Well, the power release
was to be obtained by having air stream
along these channels that surround the uranium
plugs that were permeating the whole assembly-- having airstream along there. So you needed big fans in
the exit side to draw air in. So all through this
building, there was a sucking sound of
air being pulled in here and cooling the reactor. Taking this 3
megawatts of power-- thermal power-- out
and warming up the air, and then the air was
filtered and sent up stacks. Well, at the end
of the war years-- at the end of the war,
in the summer of 1945-- one of the research scientists
at the laboratory here, Ernest Wollan, who had come
down with the original group from Chicago-- Ernest Wollan-- got to
thinking seriously-- as other people did-- how can you use these
facilities as research tools? And Ernie Wollan decided
to see if he couldn't make use of some of these
copious neutrons that were rattling around in here,
trying to bring some out through a collimated
tube, and trying to exploit this diffractive
property of the neutron radiation. So he decided he wanted to
try to set up a diffraction spectrometer in the way that
X-ray people had been doing for 15 or 20 years earlier-- bringing a beam out and
defracting it from samples, seeing what novel things might
be available from such studies. So Ernie arranged
to have sent down from the University of Chicago
an old X-ray spectrometer unit that he had used back up there. He did his thesis work in an
earlier time with Compton, studying X-ray scattering
by gas samples. So in that, he had built
an X-ray spectrometer, and he knew that
was still up there, and he arranged to
have that sent down to Oak Ridge in
the fall of 1945. So he and a colleague
of his, Robert Sawyer-- who was there during
parts of the war years, from University Lehigh-- configured this
spectrometer so that they could try to bring a neutron
beam out from the reactor, now, off to a side, and have
it fall upon this spectrometer. Which, a spectrometer,
in general, is something that you can
control angular positions. Have a detector on it, and you
can move the detector around and determine a
pattern of scattering, or a pattern of diffraction. So that was the idea
in this first effort. And his spectrometer, I show
in my next projection, here. This was the original
spectrometer. The reactor is behind
here, off to the back side. And there has not
been a channel-- an open channel-- through this
graphite assembly that lets neutrons stream out-- those that happen to be coming
in the right direction-- and stream out
through the shield and the face of the reactor. Well, they have to be
contained, so there's another outside shield
built up around here. So the idea was to
place a monochromating crystal in that beam. Neutrons were coming
out and falling on a large, single
crystal of some sort, and being diffracted
from that single crystal. Even if these were different
energy neutrons, as they are-- thermal neutrons
with a thermal-- one wants to select out
a monochromatic section of that spectrum-- a monoenergetic section. And that can be done by
diffraction from a crystal. One wavelength. One energy. So back inside this
secondary shield was this monochromating crystal. Now, about the only
large, sizable crystal that was available to Ernie
and Sawyer at the time was sodium chloride. Sodium chloride
seems everywhere. And people had to
learn the technique of preparing the large
crystals of sodium chloride. The infrared people were
hot on alkali crystals-- very fine, infrared
window material. And so the development people
had developed the technique for preparing large
crystals of alkali halides-- most notably, sodium chloride. So Ernie obtained a
sizable single crystal of sodium chloride,
roughly an inch and a half high and 2 inches long,
and a 1/2 inch thick-- which is an ungodly size for
a single crystal of substance back in those days-- and used that as a
monochromating crystal in here. From that monochromating
crystal, one would have, then, a pure diffractive
beam from it coming out, and that came out through
this secondary shield, crossed over the axis of
this spectrometer, about which I've spoken. Here is this old, original,
University of Chicago X-ray spectrometer bolted down
to the platform on the floor, here. And this neutron beam-- purified neutron beam, now-- coming out and crossing the
axis of this spectrometer. One would place, then, a
further scattering sample or diffracting
sample on this axis so that the beam would come
in and strike the sample, and now be diffracted
around from that sample. So here is a detector. This is the most massive
thing on the assembly. The detector itself
is not large. It's about 2 inches in diameter
and about 15 inches long, and it's a sealed chamber
sealed with an insulated wire along the center, and the
two electrodes, the wire, and the case, and filled
with a special gas-- boron trifluoride. Boron is the important thing. Neutrons are absorbed
copiously in boron, and so the idea was
that a neutron coming in along the axis
of this chamber would be at least
partially absorbed by boron nuclei in
this gas, and that would trigger an electrical
response between the wire and the piece,
which were charged to a high potential
difference between the two. So the chamber is inside
here, and the surrounding here is secondary shielding. The shielding is the massive
part of the assembly, here. All of this shielding to
keep stray radiation away from that detector. But this whole assembly can
be moved around the axis of the spectrometer. Now, this weighs--
this has a mass-- much larger than this
spectrometer shafting can support. So to take off some
of this loading, there were cables running
up to a gimbal joint up on the ceiling, here, so
that this off-axis loading-- mass loading-- was supported
from this universal gimbal joint up on top, here. Well, this had to be
aligned pretty well. And I remember, after I
came into the picture, all the troubles that we had
in keeping it aligned-- making sure that this
operation of angular position here was a smooth operation. But here is shown on here,
now, a typical sample-- a sample in this
circular sample. And this is a powder sample-- random crystallite
direction within the powder. Sealed up between aluminum
windows in a plate. The sample thickness-- you
don't get the impression here-- is perhaps 1/2 inch thick. The neutrons had to go
through perhaps 1/2 inch thickness of the powder sample
in the diffracting process. And the ceiling is just to
keep the powder in there. They're aluminum windows, there. Aluminum is very transparent
to neutron radiation. In operation, here, one
had to turn handles. There are some handles, here,
turning the detector around. Also, separately, controlling
the angular position of the sample. It was all completely
hand-operated. If you wanted to study the
scattering around here, you positioned
this to some point and counted neutrons
that were being detected in this detector
over some period. That period might be
as short as a minute. It might be as long as
10 minutes or 15 minutes. And then after a conclusion
of obtaining that information, go and turn to a new position
and continue the process. Powder diffracting patterns
were very time-consuming in their obtainment. Well, with this
sort of assembly, Ernie and Sawyer
indeed found that they could take powdered
diffractions from specimens. And I show here a powder
pattern obtained with a sample of, again, sodium chloride now. Sodium chloride powder. There's a sodium chloride
monochromator way up in there. That's a single crystal. This is now the powder sample. This is intensity versus
scattering angle position. This was presented in a monthly
project report at the Clinton Laboratory in April 1946. The assembly was
collected together in late 1945 and early
1946 by Wollan and Sawyer. And here are a couple
of diffraction peaks. So these are nice,
well-behaved looking peaks with measurable intensity. They take time, and it was a
chore to obtain such things, but it was available. Notice the background
intensity-- the baseline intensity-- that's
underneath all this structure. There's more of it out here, but
patience ran out at this time. But notice the
level of background versus signal intensity, here. The background is as much
or larger than the signal, and that's a
troublesome feature, and you'd like to do
something about it. Here's another of the
very earliest patterns. Ernie decided to look at water-- H2O. And not only water,
but heavy water-- D2O, and here are patterns
he and Sawyer had obtained in those very early days. Now, the distinction between
deuterium-- heavy water and light, normal water. The interest here was to see
if one could possibly determine steady, hydrogenous materials-- hydrogen-containing
materials-- because-- and this is very important--
such information-- establishing the location
of hydrogen atoms in materials and
material structures-- was completely not
available with X-radiation. Nobody had determined what the
real structure of normal water was, or frozen water or
any other hydrogenous material, because X-rays
were insensitive to hydrogen in hydrogen scattering. That was not
expected, but it was hoped that maybe with
neutron radiation, there would be a difference
enough between deuterium and hydrogen to
show a distinction. Well, these first
patterns-- which, there was no attempt at analyzing these. But they did show differences,
here, of this type that you see. This peak-- you're peaking. And these are liquid
patterns, so you don't expect sharp diffraction lines. But this peak is considerably
more prominent than for deuterium--
the heavy water-- than for the
normal, light water. So this suggests that, again-- that, indeed,
hydrogen and deuterium are scattering
differently, and that one is sensing that in these
diffraction patterns. Well, there was no analysis
of this at the time, but that was the purpose
of the experiments. Again, this was reported in this
same project progress report-- that I had shown earlier-- in April 1946. I came to the laboratory
in June of 1946, a month or two after this
period, joined the group. And Wollan and Sawyer
decided he was going to return to the university. So he came and went
back to Lehigh. And from then on, I
worked very closely with Ernie Wollan in
all of the developments. Well, knowing that
you could successfully take powder
diffraction patterns, then we just decided to
look at many materials, trying to make some
sense out of what was controlling these intensities. What were the
scattering properties of the different atoms for
this neutron radiation? There was something
known about that. There had been a compilation of
what is called total scattering cross-sections for a series
of different atomic nuclear species. But that's not the whole story. That's not the
whole thing that's going to determine these
powder peak intensities. It was recognized that
there are different isotopes for any element-- for an element. Different isotopes
might very well display different
scattering properties. And eventually, one would
have to unravel all of that. Another thing-- many
nuclei have nuclear spin, and nuclear spin-- if you have a nuclear spin, then
you have two nuclear spin state scattering amplitudes-- one where the neutrons
come in this way, and one where the
neutrons come in this way to that nuclear spin. So what that implied-- and in a general
sample, one would have these nuclei arrayed--
and random, of course. There was nothing to say the
spin should be up or down or whatever. Under those conditions,
what that implied-- and the theory had predicted
this from the very earliest-- that this, what we now call
the coherent scattering cross-section, should, indeed-- or may, indeed-- differ from the
total scattering cross-section. We had information
on total scattering, but nothing at all on
coherent scattering. And the coherent scattering
arises and this difference comes in because of
this spin state-- the possible difference
in these two spin state scattering cross-sections,
or scattering amplitudes. So in the long run, all of
that had to be unraveled, and nothing was known about
this distinction between what you were going to see
in a diffraction pattern and what would be predicted
from these total scattering cross-sections. You had to look specifically
at this quantity called the coherent scattering
cross-section, and that might be anything
from a very low value, up to the total
scattering cross-section. So that was the purpose
of looking at many of these early samples-- to see if we could unravel and
understand these intensities and how they were affected
by cross-sections; and, indeed, what the
basic cross-sections were. Well, this was all very
choresome with this hand operation, and we
were all very anxious to try to improve
things so that we didn't have to sit
there through the night and collect the data by
hand and write things down. And we were very fortunate, in
1947, when a couple chaps came down to Oak Ridge as part of
what was called the Oak Ridge Training School. After the war, the
administration management decided to set up a
training school where they would encourage people to
come in there from universities or industrial
laboratories and learn, become familiar with
all the techniques and the new techniques which had
been developed during the war years, and which
were very secret. So these people were to come
in and talk with people, work in the laboratories
in many cases, but mostly to
attend lectures that were to be given by
experts who were brought in for that purpose. So this Oak Ridge Training
School-- which had, perhaps, 150 students at some times-- these were senior
people from universities or from industrial
laboratories that were sent to Oak Ridge for this purpose. And the name Frederick Seitz
may be familiar with you. Seitz, who was a very respected
solid state physicist, was brought there as Director of
this Oak Ridge Training School. Well, two of the
students that came in-- one from the RCA
Laboratories in New Jersey, and the other from Goodrich Tire
Rubber Company over in Akron-- George Morton from RCA and
Bill Davidson from Goodrich were two of the
students, and they became interested in these
neutron diffraction patterns. And in particular, Morton was
a very skilled instrumentation man. He had worked with-- name's Zworykin, who was a
prime developer of electron microscopes in
former years, and RCA was in the business of
selling electron microscopes-- first of the type. So Morton was a very
skilled experimentalist and instrumentation man. And he took on the task-- between the two
of them, they took on the task of automatizing this
slow-operating spectrometer. And they made many improvements. We made many improvements,
there, this first year-- in 1947. This is the same spectrometer. And now, about a year later
than in the earlier photo, here's the basic spectrometer. Nothing changed there. But now here's a nice,
automatic, control, automatic operating
system, here, that drives a worm,
here, with a clock motor and some cam wheels
with some microswitches. So that on demand,
this would now change its angular position. And at the same time, there
was a cadmium shutter assembly that could flop in or
out of the beam that was falling on the sample. And the electronics system
was improved over here on the control cabinet. And the detector-- there was
just general improvement things done to all of the components. Now we had a recorder
to print out data. We didn't have to be there
to see meters print out data recorder. This was what was called
a traffic counter. You've perhaps
seen these gimmicks along the highway, where they
have a tube running across, and gives a signal,
and this thing prints a new number over
here on the side of the road. That was the only
type of recorder-- digital recorder-- that
we could find at the time. All the electronics
in the system-- amplifiers, high-voltage
power supplies-- those all had to be made homemade. There was no commercial
instrumentation of that type available
in this period. So there were improvements
in these amplifiers. Vacuum tubes had gotten better. This was long before the day
of solid state electronics. Tubes had become
better and less noisy. But still, here's
a preamplifier, and you'll notice
it's sitting up here on top of a padded blanket,
because it's microphonic. This preamplifier
is microphonic. You have to separate
it from components, and you don't want to
bump it, or else you will get a false signal
through your circuit. Well, with these
sort of improvements, this changed things in
a very great fashion. One can now put a sample on
here and get things ready and go off and do
something else, and come back the next
morning or something-- some later time--
and look at the data, or change something and
do things automatically. Hey, analysts, spend your time
analyzing things, rather than collecting things. So I can't overestimate
the importance of these
instrumentation features that came in at
that time, and they were due to our having
these two gentlemen as part of the training school-- students as part of the
training school, there-- that affected them. Well, with this,
then, you could think of taking patterns regularly,
and many more over a given time span. We were always interested
in these cross-sections, or amplitudes-- cross-sections, or amplitudes--
defining just what different nuclear species-- what the real scattering
properties were. Because that's data that has
to be obtained empirically, and it can't be predicted
from theory-- or at least, theory was far behind the
stage where it could predict how these things should go. So we were very interested
in studying as many samples and looking for consistency
from one material to another. Does sodium scatter the
same way in sodium chloride as in sodium hydroxide? That wasn't really
an established thing. So we set out to study
these consistencies, looking at chlorine
in different species. And to our dismay, there was
much that was discrepant. And this gave us a
great deal of worry. We were trying, at that time,
to look at the total diffraction pattern-- and you still do that. You look at the diffuse
scattering, as well as the coherence scattering
which is showing up here. There should be some-- if
you combine these two things together to represent the total
scattering information that you're getting-- there should be consistency. We weren't finding that. And to illustrate
that, let me show this. Here are some observations
made on three different forms of pure carbon. Diamond, which is pure carbon. Graphite-- pure carbon. Charcoal-- pure carbon. Now, what you'd first think is
that if I put a diamond sample on here, graphite
and charcoal, I ought to get the
same total scattering if I look at everything that
I can see in the scattering pattern. Turns out that that
isn't the case. It seemed not the case to us. We were not, when we
added everything up in the whole pattern,
including a diffuse scattering and coherent
scattering and peaks. There was not consistency
from one to another. And we looked at other things,
as well as this series. And we were having
extreme trouble in obtaining what we consider
to be consistent information from sample to sample. And I remember a
conversation that we had with Eugene Wigner at the time. Wigner was the Research
Director at the laboratory, and Ernie Wollan
and I bent his ear. And we were telling him
our tale of woe, here, about these inconsistencies that
we were obtaining from sample to sample. And we were very
puzzled about this, and could the great man think
of some help or relief for us? Well, he surprised me
very much by saying, no, he thought that maybe this
is the way it really is. Maybe there's some other
processes, here, going on that we just don't understand. And he gave us great
encouragement to keep on going, and he wasn't taken
aback by the fact that we were getting these
seemingly inconsistent observations. Keep on going-- back
to the laboratory. Well, it wasn't
too long after that that actually, we
found the solution. And the solution
turned out to be this, and we were both
embarrassed by knowing it. Multiple scattering in samples. Multiple scattering in samples. Both Ernie and I had had
previous X-ray pattern experience where you
scatter X-rays from samples and obtain these patterns. And we never encountered
a problem of this type. But here, things
are very different. The neutron penetration
through samples is so much larger
than X-ray penetration through samples there,
because the cross-sections are different. In a neutron case
we're using samples which are orders of
magnitude thicker, larger samples than in the X-ray case. And now you're susceptible
to this problem of multiple scattering
in the sample block. A neutron comes down
to this glob at one position, gets scattered
up through the glob-- partly-- and gets
scattered again. And the result is
that you don't have consistency in the total
scattering in your pattern. So once we recognize that
feature and that fact, and we're kicking
ourselves that it took us so long to recognize it. Then it all became clear to us. We had to place
controls on our sample-- physical, geometrical
conditions-- which we, of course, did. And then things got clarified. Now we were able to get
satisfactory consistency from one sample
to another sample with the recognition of this
troublesome multiple scattering problem. Well, we had this great
interest in knowing whether you could determine
hydrogen atoms in crystals. There was this
earlier observation that I showed on water, but
that was too complicated to interpret. So we searched for crystals
that contained hydrogen, and which might have a nice,
ordered, crystal structure. But it wasn't known whether
the hydrogens were, indeed, ordered in these materials. Well, one of the simplest of the
hydrogen-containing crystals is this-- sodium hydride. Sodium hydride, one-to-one. And if you take an
X-ray pattern of this, you see only the
sodium and you know nothing about where the position
or what the hydrogen atoms are doing. But the X-ray pattern shows
a well-developed pattern characteristic of
a sodium lattice. That's known very nicely
from the X-ray pattern. But the question is, what
are the hydrogen atoms doing? And that's what the
neutrons can hope to show. So here are two patterns
of the same position in the diffraction
pattern from two isotopically different samples-- sodium deuteride,
and sodium hydride. And you see the patterns
are completely different. So this says
immediately that you're seeing the deuterium or the
hydrogen in these patterns. If these were just
looking at sodium atoms, they ought to be the same. Why the difference? Well, the difference
turns out to be that the algebraic sign of
the deuterium scattering is opposite from the
algebraic sign of the hydrogen scattering. And that's why the two patterns
are seemingly reversed, here. Big peak here, little peak. Little peak, big peak. And that persists through more
of the diffraction pattern. So sure enough, here
what we're seeing is the real determination of
the hydrogen atom positions in these two crystals. This was the first and real
indication that one could, and this is a very
important feature. Nowadays, there's a lot of
biological structure work that goes on with neutron diffraction
in neutron diffraction studies, looking at determining where
hydrogen atom positions are in organic polymeric,
biological materials. That's an important
field of application. There are other applications
that I could go on here, but I'm afraid I've
gone too long here. But let me close by showing
one other pattern, here. This is a different
type of neutron pattern. This is what's called
a Laue pattern. In this Laue pattern, one uses
a direct beam of radiation from the reactor-- not being crystal
monochromated-- so that you have all
neutron energies, all neutron wavelengths
falling on a crystal-- a single crystal. Under such conditions,
the different wavelengths and the different atomic
planes in the crystal diffract the radiation
to specific directions, and one obtains
this spot pattern, rather than a peak pattern
as in a powder peak assembly. So we have these
Laue spots here. Well, this was not the first
picture that was obtained. There were quite a
few attempts before getting one that looked
even as good as this. And this has several
interesting features about it. One is that first of
all, it's a photograph. There's photographic film. And now, photographic film
is insensitive to neutron radiation. So you use a trick, here. You put a piece of scintillating
material next to the film. Neutrons come in and are
absorbed in the scintillator. In this case, it was a
sheet of indium metal. The indium then releases--
after capturing neutrons-- releases a charged
particle, and that charged particle reacts nicely
with the photographic emulsion. So the sensitivity comes
from this combination of scintillator and
photographic film. And there was an indium sheet,
here, covering the film. And there was a
section of that cut out in a whole circular region
to let the direct beam weight it so it
wouldn't scintillate and over-cloud the whole film. And we've obviously
gotten it off-center. The whole assembly is
off-center in the exposure. But there are two
other features in here. One-- if you look closely
at these spots, you see-- notice that they all
have a double structure. It's particularly
pronounced here. There's a double. And that puzzles us very much. Why should there be this
doubling of neutron spots? Is that perhaps
connected with the fact that there are two spin states
in the nuclear scattering from this crystal? Well, we were anxious
to explore that and see if we could explain the
origin of this doubling. And as it turned out, after
this exposure was made-- and this exposure took 20
hours, a long-time exposure; 20-hour exposure-- just after this
exposure was completed, the reactor had to shut down for
breakage of one of these fuel element cans. One of the cans
containing the uranium had developed a pinhole,
and the uranium inside it oxidized, and it had essentially
exploded inside the reactor. So there was no way of
operating the reactor under those conditions. They had to get it out of there. And that took some three
weeks for them to cleanse that and get back to operation. So it took us a long time to
ponder over this before we could hope to try to do
something else to try to explain this doubling
of these Laue spots. Well, it turned out, after
we got back in business, that the spot
structure was something associated with a
particular crystal we used. The crystal had a mosaic
layer on its two faces-- entrance and exit. And a mosaic layer
means that the crystal scatters more strongly. So what these spots
represent are, effectively, scattering from the
front side of the crystal and the back side
of the crystal. And that's, somehow or other,
showing up all around here. So that was a
trivial explanation. A good crystal without this
mosaic layer on the surface doesn't show this. But another
interesting feature-- you notice that there's
a band structure, here. Well, there's no way
you would expect that. That turns out to be the world's
first neutron radium graph feature. The band structure
represents the collection between pieces of indium. We didn't have enough indium,
or in large enough area to cover this thing, so we had
strips that we kept together by strips of scotch tape. Scotch tape line. Scotch tape line. Another one here. [LAUGHTER] So this is showing the
radiographing of the Scotch tape that was used to hold
these indium strips together. Well, that set us
off to thinking, what else could we do? If a piece of Scotch tape
shows up here like this, there must be better uses
that one can exploit in this. Indeed, that's the case. People now radiograph large
assemblies, metal castings, or other things with
neutrons, in the way that normal
radiographing is done, and exploiting the particular
penetration properties of the neutron radiation. Well, these were very
exciting times back then, and I often think really
how exciting it was. Almost any corner
you turned, there'd be new vistas showing up. And we explored many,
many corners at the time. But on reflection, I'm sure
that there's no difference today than back then. Almost anything that
one can look at, there is new information
and new things to uncover. So I think the challenges
in present-day physics, present-day science
loom just as important, as prominently as they
did back at this time. So I say that to our younger
friends who are with us here. Thank you very much. PRESENTER: Neutron scattering. Cliff's stories are rich. He actually only
began to scratch the surface of all his
many contributions-- magnetism, polarized neutron
technology, basic neutron properties. I guess remain for the
next installment of that. [LAUGHTER] Cliff has agreed to
answer some questions, so we have time for
perhaps a few, at least. Questions or comments? Bob? BIRGENEAU: Cliff, when and
how did you first get the idea that you could look at
antiparallel magents with this technique? SHULL: That came a year
or two after these things that I've chatted about. And we were set
off on this route by our desire to
look for what was called paramagnetic scattering. Paramagnetic scattering
was something that theorists had predicted
would occur from a material-- a paramagnetic material. And at that time, a paramagnetic
material-- as it is today-- represent a collection of
magnetic centers, where there was randomness in the
direction or the orientation of these magnetic centers-- paramagnetic material. In contrast to a
ferromagnetic material, where they were considered
to be all lined up. Now, in paramagnetics,
with a paramagnetic sample, if one scatters
neutrons from it, you expect to find
a diffuse scattering in the pattern associated
with this random collection of magnetic centers. This was shown in a very
powerful theoretical paper by Halpern and Johnson back in-- years earlier than this
period I was talking about. And so this was a
prediction, and one that was capable of observation
with the technology that was available here. So we were interested
in looking for this paramagnetic scattering
in a paramagnetic media. And the first thing that we
looked at was in a material-- a typical paramagnetic
substance-- hematite. Fe2O3-- hematite. There are strong magnetic ions-- iron ions-- in this substance. And it was known that you
could make or you could measure paramagnetic susceptibility
and obtain certain values of the susceptibility,
and could interpret that-- its temperature dependence-- upon the strength of
these magnetic ions, which were at random. So we looked at that first. And that was present. We could see that. We also looked at another one-- manganous oxide--
where, again, you have a strongly-magnetic
manganese ion present in the material. Manganous oxide-- monovalent,
strong, magnetic ion. And that, again, has a nice
paramagnetic susceptibility. And we looked at that, and
what we found there was there was, indeed, this
diffuse scattering, but it showed some structure
in the diffuse scattering. Instead of showing
a uniformly varying paramagnetic scattering,
it showed some structure. And that suggested-- I had had experience with
short and long-range order in crystals prior to that time. That suggested that if we were
to cool this medium down-- the sample down-- that this
short-range order might develop into coherent long-range order. And sure enough, when we
finally did that experiment, by cooling the manganous oxide
down to a low temperature, the long-range
order had developed. So the origin of the
magnetic scattering all came in the
interest of searching for the paramagnetic
effects, and then the ordered magnetic lattice
scattering followed from that. But that came in
1948, '49, 1950-- a couple years after
these experiments that I've just been describing. PRESENTER: Other questions? Looks like we're coming
to a certain symmetry since you started
with the Clinton pile. And of course, now one hopes
to build the new reactor in Oak Ridge, and
if it succeeds, it will certainly be the Gore-- the Gore reactor. So it will be the Clinton-Gore
reactor in history. [LAUGHTER] The only difference
is the 11 months it took for the Clinton reactor. It'll be 11 years at least
for the Gore reactor. With that, let's thank
Cliff once again. It was wonderful. [APPLAUSE]
