Materials Characterization

MT21005

Electron Microscopy
 Topics Covered
 History
 General concepts
 Wavelength-accelerating voltage correlation
 Basic components of electron microscope - Electron Gun
- Electromagnetic lenses
- Scanning Coils
 Aberrations
Max Knoll and Ernst Ruska Manfred von Ardenne develops first SEM
invented the electron microscope (1938)
in 1931

• Electron microscopes
were developed in
the 1930s to enable
us to look more
closely at objects
than is possible with
a light microscope.
 Cambridge Scientific Instruments
releases the first commercial SEM
(1965)

Charles Oatley begins SEM development

at Cambridge University, UK (1948)
 Introduc
•
tion
When thinking about light microscopy we tend
to ignore most of the interactions between the
light and the specimen.
• It is sufficient that enough light is transmitted
through or reflected from the specimen that the
image can easily be seen.
• The assumption is generally made that the
specimen is unchanged and for most specimens
this is a reasonable assumption.
• However, the interaction of electrons with the • It is important (in order to appreciate the way in
material through which they pass may have which an electron microscope works and the
more serious consequences. meaning of the information which it provides)
• Real possibilities are that the specimen will be that we understand nature of the possible
heated by the electron beam and that chemical interactions between
changes may take place. • Electron beam and the other parts of
the microscope (e.g. lenses or camera)
• Electrons and the specimen.
TEM fundamental
design
A simplified ray diagram of a TEM consists of
• An electron source
• Condenser lens with aperture
• Specimen
• Objective lens with aperture
• Projector lens
• Fluorescent screen
Electronic shell structure
• The nucleus carries a positive charge and is
surrounded by a number of negative electrons
which exactly neutralize this charge.
• When atoms are close to one another in a solid
most of their electrons remain ‘localized’ that is
they can be considered to remain associated with
a particular atom.
• Some outer ones will be shared, to an extent
which depends on the type of bonding with
neighboring atoms.
• The innermost (K shell) electrons are the most
tightly bound, and they would need to be given
approximately 20 keV before they could leave the
atom.
• It is common to define the zero of the energy scale as the potential energy of a free electron far from
any atom.
• The energies of localized electrons are then negative.
• Alternatively, spectroscopists refer to a positive ‘binding energy“ or the energy of the atom with the
specified electron missing, which is the negative of the energy.

Two alternative representations of the first three electron shells around a molybdenum atom. The
innermost (K shell) electrons are the most tightly bound, and they would need to be given
approximately 20 keV before they could leave the atom.
Electronic band
structure •For atoms which are bonded to others
the energy level diagram must be
modified because the Pauli exclusion
principle forbids shared electrons to
occupy the same states.
• This leads to the development of energy
bands, the most common of which are
referred to as the valence and
conduction bands.
• In a metal the conduction band holds
the familiar ‘sea of electrons’ which are
responsible for conduction and much of
the bonding.
 Important for EM
• Inner shell electrons: Highest binding energy, nearest to nucleus (usually in K or L shell),
sharply defined energy, localized.
• Outer shell electrons: Lowest binding energy, outermost occupied shell.
• Conduction band: Shared electrons, range of possible energies, delocalized.
• Outer electrons are fairly readily detached from their atoms since only a small amount of
energy to be supplied.
• It is this easy availability and low mass which makes ‘free’ electrons so useful.
 Wavelength-accelerating
• voltage
Wavelength correlation
of the electron depends on the potential difference, or accelerating voltage: At the accelerating
voltages which are most useful for electron microscopy, (2x10 V upwards) the electrons are accelerated
4

to a velocity which is a significant fraction of the velocity of light and relativistic effects are quite
important.

c = 2.998*108 ms-1 e = 1.602*10-19 C h = 6.62*10-39 Js

me = 9.108*10-31 kg

1
1.5
λ =[ ] 2
( 𝑉 +10 −6 𝑉 2 )
Basic components of electron microscope
 Electron
•
source (gun)
Of the many ways of encourgaing electrons to leave a solid so that they may be accelerated towards
the specimen, two have proved particularly useful in the construction of electron guns.
• The most widespread system uses thermionic emission from a heated filament.
• At temperatures in excess of 2700K, a tungsten wire emits an abundance of both light and
electrons.

• In a light bulb only the light is used but in an electron gun the electrons are accelerated across a
potential difference of tens or hundreds of kilovolts to generate a beam of electrons of controlled
energy (and hence of known wavelength).
• A piece of tungsten, usually a wire 0.1 mm
diameter bent into a hairpin, acts as the
cathode.
• This filament is heated by the passage of a
current to about 2800K while being held at
a high negative potential with respect to
the anode (A) and the rest of the
microscope.
• Electrons thermionically emitted from the
filament are accelerated rapidly towards the
anode and a beam of high energy electrons
is emitted through the circular hole at its
centre into the microscope column.
• The addition of a Wehnelt cap (W), which is
held at a voltage slightly more negative than
the filament enables the diameter of the
area at the end of the filament which emits
electrons to be controlled.
• The crossover diameter is effectively the
size of the electron source.
• A small current flows when the potential is applied but
before the filament current is high enough to give rise
to thermionic emission.
• This is known as the dark current since it flows before
the filament is hot enough to emit light and to
overcome its binding energy.
• As the current through the filament is increased there
is an initial rise in the emitted electron beam current.
• This eventually saturates however and there is no
point in passing more than the critical current through
the filament since this merely increases the
temperature of the filament (thus reducing its lifetime)
without giving rise to any additional beam current.

• It is best to use a filament material with as high a melting point and as low a work function as possible.
• Tungsten has a high melting point (3653 K) and a work function which is much the same as most metals (4-5 eV) and
is the most widely used filament material.
• Tungsten filaments give a brightness of about 109 Am-2 sr-1. (Beam current density per unit solid angle).
• Brightness is therefore a measure of how many electrons per second can be directed at a given area of the specimen.
• The thermionic gun is satisfactory for many purposes but is limited in the brightness of the beam it can
 Example of Tungsten
thermionic guns

• Tungsten wire (Tm = 3653K, φ = 4.5 eV, Brightness ~109 Am-2sr-1)

 LaB6
• The brightness can be increased by a factor of 10 or more (from
Tungsten) if LaBthermionic
6 with a small work function of 3.0 eV, is used (𝑩~𝟏𝟎 𝟏𝟎

Am sr ).
-2 -1

• guns
It is lanthanum hexaborate, a nonoxide ceramic. It has high melting point.
It need high level of vacuum to prevent from high temperature oxidation.
• Electron guns which use LaB6 are quite common on microscopes which
are used for analytical or high resolution work since in both of these
fields a high brightness is desirable.

Single crystal of LaB6 (φ = 3 eV,

Brightness ~1010 Am-2 sr-1)
• Synthesis is difficult. Electron
beam is confined to narrow
wavelength range, which
eliminates the spherical
aberration.
• Maintenance is difficult and the
advantage it gain is limited.
Therefore, it is not popular to
use.
 Field emission
• If still higher brightness is required, then the field emission gun is used.
• guns
If a metal surface is subjected (FEG)
to an extremely high electric field (> 10
V/m) there is a high probability
9

that an electron can leave the surface without needing to be given the amount of energy represented by
the work function.
• This is because of the effect predicted by quantum mechanics and known as tunnelling.
• The result is that many more electrons can be drawn from a piece of tungsten than is possible using
thermionic emission and the brightness can be increased in excess of 1013 Am-2 sr-1.
• The field emission current depends very strongly on
the applied field.
• In order to apply a high field, the emitter (usually
tungsten) has to be prepared in the form of a sharp
point.
• The diameter at the point must be about 0.1 μm,
which is orders of magnitude finer than a pin, so the
emitter is a rather delicate structure.
• For this fine point to be preserved in use it must be
operated in an environment with very few ions and
this dictates the use of ultra-high-vacuum (UHV)
techniques.
 Advantage of FEG
• An important feature of field emission sources is that the
emitted electrons have very well defined energies.
• Whereas the electrons from a thermionic source
inevitably have an energy spread of 1-2 eV, the electrons
from a cold FEG have a much smaller energy spread,
usually less than 0.5 eV.
• Thus for both analytical and high resolution electron
microscopy, field emission sources are important not
just because they provide a high brightness beam but
because they provide a ‘clean’ monochromatic supply of
electrons.
• Electron beam energy distribution of cold FEG on HD-
2700 (Hitachi High-Technologies Corp.) measured with
Gatan electron energy-loss spectroscopy detector (Gatan
Inc.; Acceleration voltage: 200 kV, acquisition time: 1 s,
dispersion: 0.05 eV/pixel).
Comparing
electron guns
Electromagnetic lenses
 Electromagnetic lenses
• A beam of electrons could be focused by either an
electrostatic or a magnetic field.
• Both types of field have been used in electron
lenses but the electromagnetic lens is by now
virtually universal in commercial electron
microscopes.
• The key to an understanding of what is essentially a
very simple lens is the direction of the force which
acts on a moving electron in a magnetic field.
• If an electron moving with velocity 𝒗 experiences a
magnetic field of strength 𝑩, then it suffers a
force 𝑭 = 𝒆(𝐁 ∧ 𝒗) in a direction perpendicular
to both the direction of motion and the magnetic
field.
 Forces on electrons in
• A typical EM lenses
electromagnetic lens is designed to provide a magnetic field almost parallel to the
direction of travel of the electrons.
An electron entering the lens experiences a magnetic field 𝑩 which can be resolved into
components 𝑩𝒂𝒙 along the axis of the microscope and 𝑩𝒓𝒂𝒅 in a radial direction.
•

Initially the electron is unaffected by 𝑩𝒂𝒙 , which is parallel to its direction of travel, but
experiences a small force of magnitude 𝑭𝒓𝒂𝒅 = 𝑩𝒓𝒂𝒅ev from the small radial component.
•
 Forces on electrons in
•
EM lenses
This radial force causes the electron to travel in a helix along the lens.
• As soon as it starts to spiral it has a component of velocity 𝒗𝒄𝒊𝒓𝒄𝒖𝒎

force of magnitude 𝑭𝒂𝒙 = 𝑩𝒂𝒙𝒆𝒗𝒄𝒊𝒓𝒄𝒖𝒎 in a axial direction.

perpendicular to the plane of the paper and therefore experiences a

• The helical path follows a tighter and tighter radius.

• The effect is that a parallel beam of electrons entering the lens is
caused to converge to a point exactly as light is focused by a glass lens.
 How the EM lenses works?
• If the magnetic field only extends over a short distance along the axis, then
the lens behaves as a ‘thin lens’.
• A coil consisting of a large number of turns of wire is wound on a soft iron core
(pole piece) which has only a very small accurately machined air gap across which
the field is produced.
• By varying the current passing through the coil (typically in the range 0-1 A) the
magnetic field strength and hence the focal length of the lens can be varied.
How the EM lenses works?
• An important feature for which there is no analogy in the
light microscope is the spiraling of the electrons as they
travel through an electromagnetic lens.
• Since it is very rare for the electron to travel an integral
number of turns of the spiral as it passes through the
lens in general there is a rotation of the image caused by
the lens.
• This is not a distortion since the image is otherwise
unaffected.
• Cunning design of electron-optical systems in modern
microscopes sometimes involves using lenses in
combinations which cancel out the image rotation.
Scanning • Electromagnetic fields are
coils also used to deflect the
entire beam of electrons to
scan the beam back and
forth in an electron
microscope.
• For these applications, the
field needs to be
perpendicular to the electron
beam, but much smaller
fields sufiice for these
smaller deflections so the
coils are quite small.
• Most microscopes will
contain a dozen or more such
coils, designed to enable the
microscopist to optimize the
position of the beam in the
column of the microscope.
Aberrations
in •EM
In the electron
microscope (TEM in
particular), resolution
limited by ONLY
diffraction cannot be
obtained because of the
lens aberrations.
• Whereas in a light
microscope it is possible
to correct both chromatic
and achromatic
aberrations by using
subtle combinations of
lenses, this is very
difficult using electron
lenses and has only been
seriously attempted in the
1990s.
 Corrections of
• Chromatic aberrationsaberrations
can be virtually eliminated by using electrons of a very
small range of wavelengths (by using FEG electron sources).
• It is not possible to eliminate the monochromatic aberrations, principally
spherical aberration.
• The spherical aberration is caused by the lens field acting inhomogeneously
on the off-axis rays.
• The rays which are "parallel" to the optic axis but at different distances from
the optic axis fail to converge at the same point.
• The further off-axis the electron is, the more strongly it is bent back toward
the axis.
• A point object is imaged as a disk of finite size, which limits the ability to
magnify detail, because features are degraded by the imaging process.
• A point 𝑷 is imaged as a disk with a minimum radius in the plane of "least
confusion" and as 𝑷4 with an intense central bright region with a
surrounding halo in the image plane.
Effect of spherical
aberration
• A point source imaged by a system
with negative (top), zero (center),
and positive (bottom) spherical
aberration.
• Images left of the center column are
defocused toward the inside;
images right of the center column
are defocused toward the outside.
• Only the central point is a dot; the
image above and below it appears
as a disc.
• The amount of blur i.e. degradation of resolution
caused by spherical aberration on the object plane,
𝒓𝟐 = 𝑪𝒔𝜷𝟑
• 𝑪𝒔 is spherical aberration coefficient and 𝜷 is the
angle between electron beam and the optical axis
i.e. the maximum semi-angle of collection of the
objective lens aperture.
• 𝑪𝒔 has the dimensions of length.

optical axis, the value of 𝑪𝒔 is always positive.

• For a rotationally symmetric lens with respect to the

• The practical way of minimizing this is to restrict the

electrons to paths very near the optical axis, i.e. near

aperture (by reducing 𝜷).

the centre of the lens, by using a small objective
 • The use of a small aperture reduces spherical aberration but makes the difraction-limited resolution (𝒓𝟏) worse
• There is an optimum size of aperture (i.e. value of 𝜷) for which the net resolution is smallest.
• Net resolution 𝒓 = 𝒓𝟏 + 𝒓𝟐
𝟑
• Minimizing 𝒓 with respect to 𝜷, 𝜷𝒐𝒑𝒕 = 𝟎. 𝟔𝟕𝝀–𝟒 𝑪𝒔 𝟒; 𝒓𝒐𝒑𝒕= 𝟏.
𝟏 𝟏
𝟏

𝟐𝟏𝝀 𝟒 𝑪𝒔𝟒
• The resolution can be improved by reducing the factor 1.21 to as low as
0.7 in favourable circumstances.
• Using the optimum aperture it is now possible with a good TEM to
resolve two points about 0.2 nm apart (approximately the separation of

• Since it is necessary to keep 𝜷 small in order to reduce the effect of

atoms in a solid).

spherical aberrations, electron microscopes always gain the advantage of

a large depth of field.
• For EM, 𝒉 = 𝟎.𝟔𝟏𝝀 , as 𝜷 is reduced the depth of field increases very
𝜷𝟐

• One way to minimize spherical aberration is to use a short focal length lens (i.e. small 𝑪𝒔).
rapidly.

• Spherical aberration coefficient (𝑪𝒔) is decreasing as the relative 𝜽 = 𝜷 𝒙 𝑳 increasing (𝑳 = Effective

length of lens)