AT&T Archives: Matter Waves, Holden and Germer on Wave Nature and the Da...

transcript



I hope you've got over the shock of

learning that light an electromagnetic

wave behaves in some ways like a stream

of particles the photons because I'm

about to ask you to absorb another punch

I'm going to try to convince you that

particles particles of ordinary matter

behave in some ways like waves I don't

mean by this that the particles bob up

and down they do that sometimes of

course but that's not what I mean I mean

something much more fundamental you'll

pay a visit to this apparatus in this

film and I hope you will be convinced by

what it shows you that a particle can

behave like a wave in a very deep-seated

way maybe after the first punch about

the particle nature of light the second

about the wave nature of particles won't

alarm you too much if it doesn't then

you have something in common with Lois

deploy deploy I suggested back in 1923

that particles might behave like waves

and his suggestion grew out of his

comparison of the way matter behaves

with the way life behaves

he wrote that suggestion into a thesis

he was submitting for a doctor's degree

at the University of Paris but if you

find the waves of matter harder to take

than the particles of light then you're

still in good company dubrow's

examiner's of the university were

reluctant to give him a degree for such

a fool idea he got the degree all right

but only because Albert Einstein

happened to be visiting Paris at the

time and said look this isn't such a bad

idea maybe it's even true and then four

years later Davisson and Germer in New

York City and GP Thompson and Cambridge

England described diffraction

experiments they had done quite

independently of one another showing

that the boys idea was right now

I think you've studied waves enough to

remember that waves can be diffracted

here for example you see waves coming up

to the edge of a barrier and bending

around behind it so that the barrier

doesn't cast a sharp shadow in fact if

you put something in the path of the

waves which is no larger than the

wavelength of the waves the waves Bend

around from both sides and the thing

casts no shadow but if you shoot

particles at something for example the

droplets of paint from a spray gun the

particles either hit the thing or they

don't so that there's always a shadow

this way of casting a shadow by spraying

particles that a thing is used in the

modern electron microscope

here's a small something that you want

to examine in the microscope you spray

it with electrons

you put the beams of electrons through

an electrostatic lands the lands spreads

those beams out

and you let them hit a fluorescent

screen or a photographic plate so that

you get an enlarged shadow of the thing

here is a picture taken in an electron

microscope of some of the tiny bits of

matter that make up smoke they look as

you might expect like the shadows of

little blocks but the edges of the

shadows don't look sharp you see light

and dark bands in them the electrons

making the picture don't seem to behave

quite the way they should if they were

simply particles like those drops of

paint here you see a case of diffraction

diffraction of particles the electrons

in making those diffraction bands the

electrons are behaving like waves you

can get the same kind of shadow when you

place an object a razor blade for

example in a beam of light now I'll talk

in the room for a moment here's a small

source of light and when I line it up on

that razor blade you can see a shadow

through the screen looking closely at

that shadow you see diffraction fringes

these fringes are very much like these

other fringes made by electrons passing

the edges of the smoke bit so the

electrons seem to be behaving like the

light and you might ask the question are

both the electrons and the light

behaving like waves or are they both

behaving like particles but you know the

answer to that one

the band's in the diffracted light come

from the interference of the light waves

with one another looking for

interference phenomena like this is the

best way we know for finding out whether

we are dealing with waves of any sort so

the electrons seem to be behaving like

waves here but it's worthwhile trying to

find some way to check that idea

now let me remind you of the diffraction

of light by a ruled gray here's a thin

sheet of plastic that has very fine

ridges and furrows molded into it

looking at them under a microscope you

see that they are arranged in straight

parallel lines the distance between them

is very short maybe about 20 times the

wavelength of light I have a sheet of

that grating held flat under a piece of

glass when I shoot a beam of light at

braiding the ridges and furrows scatter

the waves of life and the standard waves

add up in just a few directions in most

directions the scattered light cancels

out the ruled face reflects some of the

beam back at an angle which equals the

angle of incidence but it also reflects

some of the beam in other directions and

in those directions the angle depends on

the wavelength the color of the light so

in those directions the light is spread

out into a spectrum with the red light

at the outer end of each spot and the

blue light at the inner end the

wavelength of the light is shortest here

and longest here when I use only the red

light by putting a light filter in you

see what the grating does to light at

the long wavelength end of the visible

spectrum when I shift filters and use

only the blue light you see that the

diffracted spots all shift inward

because the blue light has a shorter

wavelength look at that shift again red

light blue light

now this suggests that if we want to

look for the wave-like properties of

matter we might shoot a beam of matter

at a grating and see how the matter

bounces back if it bounces back every

which way then that's just what we

expect of particles but if it bounces

back in beams then we suspect that the

particles are behaving like waves in

order to get any useful answers we have

to design a grating with lines whose

spacing is comparable with the

wavelength we are looking for

but what wavelength are we looking for

in his original thesis dubrow made a

brilliant suggestion I've already

reminded you that light comes in little

bundles the photons each photon has a

momentum that momentum shows up whenever

the photon hits something for example

and the momentum P of any one of the

photons turns out to be given by H over

lambda where lambda is the wavelength of

the light and H is Planck's constant

dubrow's brilliant suggestion was that

the same relation should hold for the

matter waves in other words solving for

lambda you expect the wavelength to be

given by H over P this is sometimes

called the debris relation now let's get

some numbers into this using the units

with which we are familiar meters

kilograms and seconds in those units

Parkes constant H is about 10 to the

minus 33 you see immediately that we're

going to need a pretty small value of p

to get a wavelength long enough to deal

with to make the momentum MV small we

want a small mass and a small velocity

an electron has a very small mass about

10 to the minus 30 kilogram

if you accelerate the electron through a

potential difference of 100 volts you

will give it a velocity of about 10 to

the 7 meters per second then the

momentum of the electron will be about

10 to the minus 23

then dubrow's relation says the

wavelength of the electron will be about

10 to the minus 10 meter which is about

the size of an atom so you want a

grating with ridges and furrows spaced

by about the diameter of an atom well

that sounds pretty difficult because

after all the grating has to be made of

matter and matter comes and atoms but

the thing can be done here's a model of

what one of the witches on such a

grating might look like the atoms make

little humps along the ridge and when I

put another Ridge next to it spaced by

the diameter of an atom the spacing of

the humps along the ridges is as big as

the spacing between the ridges but when

I've added many lines of atoms you see

that I've just made another sort of

grading little husk regularly arranged

in a plane it's a different sort of

grating from a simple line grating but

it's a grating alright that grating is

the regular arrangement of atoms on the

surface of a crystal and it's already

made for you by the crystal with just

the right spacing to do one experiment

with electrons with a piece of plastic

ruled with lines in two directions like

this graph paper I could do the same

experiment with light if the spacings in

that rating were comparable with the

wavelength of light actually I can get

the same result by taking one of the

light gratings with lines ruled in this

direction and put on top of it a second

with lines ruled in this direction I've

done this here and when I shoot a beam

of light at this plate

I get a symmetrical array of streaks you

would expect that one of the gratings

working separately would give you

streaks in this direction and the other

would give you these streaks but the

cost grading gives you these additional

beams making a square pattern now let me

use those light filters again to show

you what happens to the pattern at

different wavelengths here's what the

pattern looks like when I pass only the

red light when I shift to the filter

that passes only the blue light the

pattern keeps its shape but shrinks

toward the center I'll switch them again

here's the red the longer wavelengths

here's the blue the shorter wavelengths

short form now you know what you might

look for when we accelerate a beam of

electrons by about a hundred volts and

bounce it off the surface of a crystal

if the electrons are behaving like waves

we expect them to bounce off in beams

which make a regular array of spots and

when we increase the accelerating

voltage so that we increase the energy

and the momentum of the electrons and

shorten their wavelength we expect the

pattern to shrink like this

and now you're going to visit dr. Lester

Germer at Bell Telephone laboratories

who has this experiment with electrons

all set up in his research is dr. Gerber

uses an apparatus in which he shoots

electrons out of an electron gun they go

through a tube

bounce off a crystal and come back and

hit a fluorescent screen and make spots

on it dr. Gerber is the man who

collaborated with dr. Davidson in the

earliest experiments that showed the

wave nature of matter 35 years ago this

is the apparatus which just been

described I use it to study the

arrangement of atoms on the surface of a

crystal this is an electron gun

electrons come from this gun and they're

brought together into a narrow beam this

beam strikes the surface of a crystal

some of the electrons of this beam are

bounced off from the surface of the

crystal at various angles and reach a

fluorescent screen they show on the

screen as bright spots from the

positions of these spots we know that

angles through which these electrons

were scattered and from these angles we

can work out the arrangement of the

topmost layer of atoms on this crystal

surface in order to see the diffraction

pattern made up of these electrons we

must turn off the lights these switches

control the voltages in the electron gun

and bring these electrons into a sharp

beam we can now see the diffraction

pattern which these electrons make on

the fluorescent screen this pattern was

made by electrons having a speed

corresponding to 40 volts this is a

symmetrical pattern has the symmetry of

the top layer of atoms of the crystal

surface now I shall increase the voltage

to 47 pattern shrinks I decrease the

voltage the pattern gets bigger again

I'm changing the voltage back and forth

between 4047 the parent expands and

shrinks we are using only this small

range of voltages because if we go to

larger voltages there are some

subsidiary effects which we do not wish

to show the ratio between the sizes of

these two patterns is the ratio of the

square roots of these two numbers 40 and

47 this is because the energy of the

electrons varies directly with the

electron voltage and momentum

varies with the square root of the

voltage correspondingly the wavelength

varies inversely with the square root of

the voltage

I'm changing the pattern back to 40

volts now pattern is bigger again and

now again back to 47 volts to the

smaller pattern I could calculate the

wavelength by finding the angles of the

beams from the police to the spots on

the screen and by knowing the spacings

of the atoms in the crystal and this of

course we do now this calculation of

electron wavelengths is like what dr.

Davidson and I did 35 years ago but our

apparatus was not as convenient as this

one we couldn't see diffraction spots on

a fluorescent screen the old tube which

we had at that time I now have in my

office this is one of our first

experimental tubes we detected reflected

electrons by moving a collector around

inside this can how it works is best

shown in a sketch and I have here a

sketch of the very first tube which was

used this is the sketch this is the

electron gun it's the hot filament the

electron beam is defined by these

pinholes the beam hits the surface of a

crystal this is a box which can be moved

around to different angles and we can

read the current at different angles

just by tipping the whole apparatus this

box moves around under gravity as you

tip the tube here is such a tube the gun

and the crystal and the collecting box

are within this metal chamber and can't

be seen but as you turn the tube this

pointer moves around and it registers

the angle of the collecting box so we

can measure the current to this

collecting box at different angles

in one experiment we got a result like

this this is a plot of the current to

the box as a function of angle this is

the current this is the angle of the box

there is a strong current coming

straight back we don't measure at angles

less than 20 degrees but from 20 degrees

with increasing angle the current is

decreasing then it comes to a new

maximum out here at about 53 degrees and

the current decreases again if we had

had a fluorescent screen this peak at

about 53 degrees would be recognized by

a sharp spot on the fluorescent screen

this is a pretty tedious way of

observing intensity Maxima

it is very much more laborious than our

present method of using a fluorescent

screen originally we were not looking

for confirmation of Blois theory of

matter waves we were doing these

experiments before we had ever heard of

dubois theory but as soon as de Blois

theory came to our knowledge we checked

it against the positions and bulges at

which our diffraction beams came out and

we found that they checked quite well

after finding the first few sharp beams

we made a extensive extensive

exploration and found that there were

many diffraction beams and that they

checked the ROI theory quite thoroughly

and extensively this was the first

experiment in which electron diffraction

was observed the experiment that dr.

Germer has just described to you is

sometimes called the Davidson Germer

experiment now let me describe to you

the experiment that GP Thompson did in

Cambridge England at about the same time

Thompson shot a beam of electrons at

higher voltages right through at them

gold foil and examined the beams coming

out the other side

now Thompson's gold foil was made of

very many tiny crystals not just one if

we represented one of Thompson's

crystals by this square grating we would

have to represent his gold foil by a

great many such gratings turned every

which way

Thompson's beam now although it was was

broad enough to hit lots of them now I

can use my square grating and a beam of

light to prepare you to understand what

he saw one of the little crystals might

give a set of beams like this and

another a set of the same beams but all

turned in a different direction and so

on

well from the whole collection of little

crystals you'd expect the diffracted

beams to make the pattern that I get

when I whirl is grating around

and here's the sort of pattern that

Thompson actually got when a beam of

electrons was deflected by passing

through a thin foil don't be disturbed

by the fact that this pattern is made by

transmitted electrons not by reflected

electrons a transparent grating gives

you just as good a diffraction pattern

for the light that goes through as for

the light that reflects of course in

Thompson's experiment the electrons

going through each crystal are scattered

by layer after layer of atoms not only

the layer at the surface but that just

means that each electron is scattered by

a three-dimensional grating instead of a

two-dimensional Brading a

three-dimensional grating diffracts a

beam of waves in definite directions to

indeed light going through a crystal or

through a foil made of many little

crystals is diffracted that way whenever

the light has a short enough wavelength

a wavelength comparable with the spacing

between the atoms and the crystals

x-rays are light with that kind of

wavelengths here is a photograph taken

by a beam of x-rays shot through a metal

foil the same metal foil that was used

for making the other picture that I

showed you taken with the beam of

electrons comparing those two

photographs you see how alike they are

the electron photograph the x-ray

photograph the spacings of the circles

is the same in both of them I find this

a very convincing proof myself that a

beam of x-rays and the beam of moving

electrons have something in common that

they both have a wave-like property the

photographs aren't exactly alike for one

thing the relative intensities of the

circles aren't the same after all x-rays

and electrons aren't exactly alike

x-rays or x-rays and electrons or

electrons

and the atoms in the foil scatter x-rays

and electrons with different intensities

but both have a wave-like property and

the laws governing the directions in

which they are scattered or the sight

now de Blois wasn't talking especially

about electrons when he made his

suggestion he thought all particles of

matter might behave like waves and soon

after the electron experiments people

devised experiments to show that

dubrow's prediction was correct for

other bits of matter too here is a curve

obtained by Esteban and starin working

in Hamburg in 1929 with beams of helium

atoms they squirted a jet of helium

atoms at the surface of a lithium

fluoride crystal here is the beam of

helium atoms reflected at an angle equal

to the angle of incidence and on either

side of this they found these beams of

diffracted helium atoms the wave-like

properties of neutrons also show up when

they are diffracted from crystals

whenever the momentum of the neutrons

gives them suitable wavelengths here are

pictures made by beams diffracted by a

crystal of sodium chloride a crystal of

common salt the picture on the left was

made by x-rays that on the right was

made by neutrons

today the wave-like behavior of material

particles is well-established

indeed it forms a foundation for our way

of understanding manner the behavior of

individual apps and the ways they join

to form molecules defied explanation

until the wave-like nature of their

electrons and nuclei were discovered

nowadays atoms and molecules are pretty

well understood what an account of that

would need a film all its off

and the fact which Germer helped to

discover that a beam of particles

behaves like a beam of waves with the

wavelengths de Blois predicted is used

in countless researches here Germer

is really turning the Davidson Germer

experiment upside down instead of using

the reflection from crystals to prove

that electrons behave like waves he uses

their reflection today with confidence

that the electrons will behave like

waves since he can trust it he uses it

to find out more about the arrangements

of atoms that are reflecting the

electrons 6 times 10 minus 7 a little

more now coming up slowly

it's enough

they comes to the myosin see a spot

develop

you

you