VIIScanning Electron Microscopy and Associated
Techniques: Overview
Imaging Microscopic Features
The scanning electron microscope (SEM) is an
instrument that creates magnified images which
reveal microscopic-scale information on the size,
shape, composition, crystallography, and other
physical and chemical properties of a specimen.
The principle of the SEM was originally demon-
strated by Knoll (1935; Knoll and Theile 1939)
with the first true SEM being developed by von
Ardenne (1938). The modern commercial SEM
emerged from extensive development in the 1950s
and 1960s by Prof. Sir Charles Oatley and his
many students at the University of Cambridge
(Oatley 1972). The basic operating principle of
the SEM involves the creation of a finely focused
beam of energetic electrons by means of emission
from an electron source. The energy of the elec-
trons in this beam, E 0 , is typically selected in the
range from E 0 = 0.1 to 30 keV). After emission
from the source and acceleration to high energy,
the electron beam is modified by apertures, mag-
netic and/or electrostatic lenses, and electromag-
netic coils which act to successively reduce the
beam diameter and to scan the focused beam in a
raster (x-y) pattern to place it sequentially at a
series of closely spaced but discrete locations on
the specimen. At each one of these discrete loca-
tions in the scan pattern, the interaction of the
electron beam with the specimen produces two
outgoing electron products: (1) backscattered
electrons (BSEs), which are beam electrons that
emerge from the specimen with a large fraction of
their incident energy intact after experiencing
scattering and deflection by the electric fields of
the atoms in the sample; and (2) secondary elec-
trons (SEs), which are electrons that escape the
specimen surface after beam electrons have
ejected them from atoms in the sample. Even
though the beam electrons are typically at high
energy, these secondary electrons experience low
kinetic energy transfer and subsequently escape
the specimen surface with very low kinetic ener-
gies, in the range 0–50 eV, with the majority below
5 eV in energy. At each beam location, these out-
going electron signals are measured using one or
more electron detectors, usually an Everhart–
Thornley “secondary electron” detector (which is
actually sensitive to both SEs and BSEs) and a
“dedicated backscattered electron detector” that is
insensitive to SEs. For each of these detectors, the
signal measured at each individual raster scan
location on the sample is digitized and recorded
into computer memory, and is subsequently used
to determine the gray level at the corresponding
X-Y location of a computer display screen, form-
ing a single picture element (or pixel). In a con-
ventional-vacuum SEM, the electron-optical
column and the specimen chamber must operate
under high vacuum conditions (<10−^4 Pa) to min-
imize the unwanted scattering that beam elec-
trons as well as the BSEs and SEs would suffer by
encountering atoms and molecules of atmo-
spheric gasses. Insulating specimens that would
develop surface electrical charge because of
impact of the beam electrons must be given a con-
ductive coating that is properly grounded to pro-
vide an electrical discharge path. In the variable
pressure SEM (VPSEM), specimen chamber pres-
sures can range from 1 Pa to 2000 Pa (derived
from atmospheric gas or a supplied gas such as
water vapor), which provides automatic discharg-
ing of uncoated insulating specimens through the
ionized gas atoms and free electrons generated by
beam, BSE, and SE interactions. At the high end
of this VPSEM pressure range with modest speci-
men cooling (2–5 °C), water can be maintained in
a gas–liquid equilibrium, enabling direct exami-
nation of wet specimens.SEM electron-optical parameters can be optimized
for different operational modes:- A small beam diameter can be selected for high
spatial resolution imaging, with extremely fine
scale detail revealed by possible imaging strate-
gies employing high beam energy, for example,
. Fig. 1a (E 0 = 15 keV) and low beam energy,
. Fig. 1b (E 0 = 0.8 keV),. Fig. 1c (E 0 = 0.5 keV),
and. Fig. 1d (E 0 = 0.3 keV). However, a nega-
tive consequence of choosing a small beam
size is that the beam current is reduced as the
inverse square of the beam diameter. Low beam
current means that visibility is compromised
for features that produce weak contrast. - A high beam current improves visibility of low
contrast objects (e.g.,. Fig. 2 ). For any combi-
nation of beam current, pixel dwell time, and
detector efficiency there is always a threshold
contrast below which features of the speci-
men will not be visible. This threshold contrast
depends on the relative size and shape of the
feature of interest. The visibility of large objects
and extended linear objects persists when
small objects have dropped below the visibility