Geology Reference
In-Depth Information
The affected volume is teardrop shaped. The depth and cross-sectional area of the affected volume varies, depending
on the accelerating voltage of the beam and the density of the sample. The affected volume is usually less than
10 µm
3
, making it one of the smallest analytical volumes that can be quantitatively measured. Generally, the volume
affected by the X-rays increases with increased accelerating voltage and decreases with increasing sample density.
The interaction of the electron beam with the sample
s crystal lattice (internal arrangement of atoms) generates
several phenomena that may be used to characterize the sample. Inelastic scattering occurs when the electron
beam collides with the outer shell electrons in the sample
'
s lattice. Deceleration of the electron beam transfers
energy to the sample lattice and knocks some of the inner shell electrons to higher energy levels within the sample.
When these electrons cascade back down to the inner shells of the atoms, the sample will emit characteristic
radiation (i.e., energy) that is unique to each element in the sample. This is similar to the way characteristic
radiation is generated by X-ray diffraction, except that the mineral is the target instead of the copper anode used in
an X-ray tube. Secondary electrons are also scattered out of the sample and have energies on the order of 50 eV,
much less than characteristic radiation. A magnetic field can be used to attract them to a secondary electron detector
that is usually mounted at the back of the sample chamber. Although secondary electrons are generated throughout
the affected volume, only those generated at or near the surface escape the sample. Thus, secondary electrons are
employed to visualize the details of surface morphology (Figure 10.1.4).
'
Elastic scattering occurs when the electron beam is deflected by positively charged atomic nuclei in the sample lattice,
with some energy loss through heat dissipation and grounding. The deflections cause relatively minor energy losses to
the scattered electrons, but some of the beam can be scattered back out of the sample. These backscattered electrons
have high kinetic energy and are relatively unaffected by magnetic fields in the electron microprobe. They are usually
detected by a solid-state backscattered electron detector mounted coaxially with the electron optical system imme-
diately above the sample. The number of backscattered electrons increases with the increasing average atomic number
(analogous to density) of the sample. Backscattered electrons are employed to visualize compositional variations
within a sample (backscattered electron image) and zonation within individual phases.
As noted above, a fraction of the electron beam collides with the inner shell electrons in the atoms of the sample
lattice with sufficient energy to eject these electrons to higher orbitals and raise the atoms to an excited or ionized
state. Electrons from the outer electron shells cascade down to fill the inner shell vacancies, releasing a discrete
amount of energy, usually in the form of an X-ray photon, i.e., a packet of X-ray energy. Each element has a unique
electron configuration, so the X-rays emitted are characteristic of the atoms in the sample. The amount of energy
released and the ability to detect that energy are related to an element
s atomic mass. Lighter elements such as H, C,
and N are consequently difficult to detect. Most electron microprobes are equipped with two types of X-ray
spectrometers and corresponding detectors: (1) energy dispersive spectrometer (EDS) and (2) wavelength
dispersive spectrometer (WDS).
'
EDS detectors are solid-state detectors that are sensitive to a wide range of X-ray energies. Most microprobes are
equipped with SiLi EDS detectors. These are silicon semiconductors with lithium drifted through them to
Cl
Spot A
Spot B
A
S
B
0
2
4
6
8
10
0
2
4
6
8
10
100
m
μ
Energy (Kev)
Energy (Kev)
Figure 10.1.4. Secondary backscattered electron image (center) of a mineral nucleated at a coal-fire gas vent in
Perry County, eastern Kentucky. The EDS spectra of two different analytical spots reveal a drusy coating of
elemental sulfur (spot A). The spectrum on the left (A) reveals previously formed salammoniac beneath the sulfur
coating. Note that both spectra do not reveal N in the salammoniac because the EDS detector cannot efficiently
gather a signal from this relatively light element. At spot B, only the Cl signal from salammoniac was detectable.
Inconsistencies brought about by irregular surface geometry and penetration of the beam beyond the surface made
calibration of the intensity (vertical height) difficult for accurate elemental quantification. Photo by Paul A.
Schroeder, 2008
.
