Transcript SEM Microcharacterization

SEM Microcharacterization
• SEM characterization modes:
– Microscopy
– Electron Beam Induced Current
– Cathodoluminescence
– Energy dispersive X-Ray Spectrum Analysis
– Electron beam lithography
Fundamental Physics I
Trivia: SEM working principles were outlined in 1942 by Zworykin, but it was not until 10 years later that a
working machine was assembled in Cambridge University.
h
h
h
1.22
e 



(nm)
mv
2mE
2mqV
V
The SEM operates with electrons having
energy 20 – 30 keV. For 20 KeV, the De
Broglie wavelength e = 0.0087 nm.
The interaction of the electrons with a given material produces secondary electrons, backscattered electrons,
characteristic and continuum X-Rays, Auger electrons, photons, and electron-hole pairs
Fundamental Physics II and Applications
Re can be found out from the empirical expression
Re 
4.28  10 6 E 1.75

(cm)
Where  is the sample density, and E is the energy in keV
The interaction of the electrons with a given material produces secondary electrons, backscattered electrons,
characteristic and continuum X-Rays, Auger electrons, photons, and electron-hole pairs
SEM imaging parameters
• Magnification M = (length of CRT display) / (length of sample area scanned). In
modern machines magnifications up to 200, 000 can be achieved.
• Resolution as low as 1 nm can be achieved, which is usually limited not by the
wavelength of the electrons but by the diameter of the focused electron beam
and electron scattering in the sample from the valence and the core electrons.
Due to electron scattering the original collimated beam gets broadened.
• Contrast of the SEM depends mostly on the sample topography since most of
the secondary electrons are emitted from the top 10 nm of the sample. The
contrast C depends on angle as:
C = tan d, where  is the angle from normal incidence. At 45° angle, d = 1°
causes change in contrast by 1.75%.
The contrast in backscattered electrons can come from the difference in atomic
number Z.
• The SEM operates in a very different manner from optical microscope, in that
electrons even away from the detector are attracted, amplified, and displayed
on the CRT. Thus the image displayed in the CRT is not a true image of the
sample.
SEM working parts I
Trivia: SEM was discovered in 1942 by V. K. Zworykin, but it was not until 10 years later that a fully
functional microscope was developed by researchers at Cambridge University
• The basic SEM consist of an Electron
gun, and a few focusing lenses, and
detector. For EDS an X-Ray detector is
also used
• The pressure inside the chamber is
maintained at ~10-8 Torr vacuum.
• Microscopes are usually operated in the
voltage range of 20 – 30 keV, but for
insulating samples 1 kV or less can be
used. For insulating samples a thin metal
coating can also be used.
• The standard electron detector is an
Everhart-Thornley design (scintillator
followed by a photomultiplier tube) that is
capable of amplifying electron currents by
almost a million times.
SEM working parts II: Electron sources
• There are mainly three types of
electron sources
– Tungsten hairpin filament: This is simply a tip
that is heated to an extremely high temperature
of ~2500 C to make electrons have high
enough energy to overcome the surface work
function of ~4.5 eV
– To get higher electron current stable materials
with lower work function is preferred. LaB6 as
polycrystalline powder is used to reduce the
work function to about half that of the tungsten
metal and significantly increasing the current
– In field-emission guns, an extremely high
electric field is applied to have the electrons
“tunnel” through the barrier into vacuum.
These could be operated as “cold” or they
could be operated at higher temperature, when
they are called Schottky emitters. The later
ones are easier to clean and maintain.
Imaging Results I: Effects of source
Imaging Results II: EDS
Other modes associated with SEM
• EDX: It is part of electron probe microanalysis, and is based on the
detection of the energy of the X-Rays that are generated from the
electron interaction with matter. This is a widely popular technique
for microanalysis since X-rays of all the energy range can be
simultaneously detected. The detector consists an FET that is
capable of resolving the energy of X-ray into pulses of different
peaks based on the EHPs created by the X-Ray.
• EBIC: In this technique, the incident electron beam is used to create
electron-hole pairs that constitute the current. If the EHPs recombine
quickly due to higher density of recombination center (poor material
quality) then the magnitude of current drops
• CL: Here the EHPs generated as a result of the incident electron
beam recombine to give similar information as a PL spectrum.
Advantage here is that materials with different bandgaps can be
probed easily. Also, some depth profiling is possible.
Calculation of EBIC current
The number of EHPs generated is given as:
Here Eehp is the average energy necessary to create one EHP, Ebs is the
mean energy of the backscattered electrons, and α is the backscattering
coefficient.
The generation rate of the EHPs is given as:
Here Vol is the volume in which the EHPs are generated and equal to 4/3r3. r
can be either Re/2 or Ldiff, the diffusion length of the minority carriers, depending
on which one is greater.
The density of excess carriers is given as: n = Gn, where n is the electron
concentration in a p-type material, and n is the minority carrier lifetime.
Problems on SEM microcharacterization
Assume Ebs = 0.9 E, α = 0.1 and Eehp for Si is 3.64 eV. Also assume minority
carrier diffusion length is smaller than the radius of the generation volume, Re.
Ion probe techniques
• Uses a variety of different
materials to produce ions
such as Cs, O2 or Ga as
they are most suitable for
secondary ion yield.
• Used in two common
techniques: SIMS and
RBS
SIMS 1
• Uses Ions to hit the material and produce secondary ions
• The secondary ions are selected by means of a tandem electric and
magnetic filter so that a narrow range of ions with correct charge/mass
(q/m) ratio can emerge out
• The mass resolution m/m can be up to 40,000 so that elements
differing in mass of 0.003% can be distinguished
• This process is destructive, but highly accurate provided a reference
sample for comparison exists
• This is the only method that gives the actual dopant density and not
just carrier concentration in a semiconductor
• All elements can be analyzed in this technique
• Lateral resolution down to micron size is possible. Depth resolution
down to 5 – 10 nm can achieved.
• Most sensitive of all beam techniques with detection limits down to
1014 cm-3.
SIMS 2
• Two major types – Ion Microprobe and Time of flight (TOF)-SIMS. The
first is also called dynamic SIMS where a complete depth profile can
be done, and the later is for static SIMS as only a few monolayers are
removed in pulsed fashion and detected.
• All SIMS modes other than TOF does serial screening of the q/m ratio.
TOF SIMS displays everything together based on the time taken to
reach the detector (a fixed path length)
RBS
• This process uses Light atoms typically He ions with energy 1 – 3 MeV to
bombard the surface of the sample and measure the energy of the
backscattered ions.
• Typically suited for heavy metals on light substrates, i.e. metal contacts
• Gives masses of elements in the sample up to a depth of 10 nm to a few
microns
• This is non-destructive technique which is an advantage over SIMS
• Depth resolution is on the order of 10 nm or so
• The detection limit is in the range of 1017 – 1020 /cm-3 , much less than SIMS
Focused ion beam (FIB) technique
• In this technique, finely focused Ga+ ions are used to etch away selected
regions in a circuit or a micro/nanostructure. The beam energy is 5 – 50 keV,
which is lower than that of electrons in SEM. The process is called ion-milling.
• Common applications are for etching materials so that they are suitable for
imaging in optical microscope or even TEM
• The minimum spot size of ions is ~10 nm, which is much larger than SEM.
• The region affected by the beam can be imaged by the secondary electrons
just like in SEM. Note that the primary beam is no longer made of electrons.
• The technique can also be used to deposit metals like W, Pt, or Au. This is
very important for rectifying small circuit errors or joining nanostructure to
large metal pads for rapid device prototyping.
• The metals are deposited by delivering selected gases very close to the
beam, which then gets adsorbed on the surface, get decomposed by the Ga+
ions and are deposited on the surface.