About
Community
Bad Ideas
Drugs
Ego
Erotica
Fringe
Abductees / Contactees
Area 51 / Groom Lake / Roswell
Crop Circles and Cattle Mutilations
Cydonia and Moon Mountains
Dreams / Auras / Astral Projection
Flying Saucers from Andromeda
Free Energy
Fringe Science
Government UFO Coverups
Gravity / Anti-gravity
Life Extension
MJ-12 - The Alien-Government Conspiracy
Men In Black
Tesla
Society
Technology
register | bbs | search | rss | faq | about
meet up | add to del.icio.us | digg it

The right beam for the job

----------------------------------------------------------------------------

Title-> Do you have the right beam for the job? High-current-density
electron and ion beams are opening many doors in nanoscale
research as a result of the development of better beam sources.

Authors-> Lindquist, Jay; Rathkey, Doug; Fischer, Phil

----------------------------------------------------------------------------

By Jay Lindquist, Doug Rathkey, and Phil Fischer, FEI Co. Jay Lindquist is
an applications scientist at FEI Co., Beaverton, OR. He received a PhD in
physical chemistry from Univ. of California, Irvine, and previously worked
at Aerojet ElectroSystems, Azusa, CA. Doug Rathkey is marketing manager for
emitter products at FEI. Phil Fischer is responsible for marketing emitters,
columns, and systems at FEI. BEAMS ARE ALL the rage these days, whether
researchers are depositing, etching, analyzing, or machining material
surfaces.

Fundamental to the electron and ion beams used for microanalysis,
micromachining, and microdeposition are the field emission and field
evaporation sources used in high-current-density submicrometer ion and
electron gun designs.

Schottky emission (SE) and cold field emission (CFE) are two of the most
recent electron sources to come into common use. Other electron sources
include LaB, and tungsten filament (see table).

Liquid metal ion (LMI) sources are the choice for producing
high-current-density submicrometer ion beams.

As with most technologies, applications are the real driving force behind
the development of electron and ion beams.

Electron beam applications include: * nanoscale e-beam lithography *
low-voltage scanning electron microscopy * critical dimension measurements *
low-energy electron microscopy * reflection high-energy electron diffraction

(RHEED) * scanning Auger microscopy. Ion

beam applications include: * micromachining and milling * microdeposition of
metals * maskless ion implantation * microstructure failure analysis *
secondary ion mass spectrometry.

SE takes on CFE

SE and CFE are the two high-brightness point cathode technologies of choice
for use in nanometer electron focusing systems. The basic underlying
mechanism for generating electron beams involves the emission of electrons
from a metal surface under the influence of a strong electric field (see
accompanying box). Both SE and CFE sources have similar energy spreads, but
their energy distributions are mirror images (see graph).

In CFE sources, electrons tunnel from various energies below the Fermi
level. The CFE energy distribution terminates on the high-energy side near
the metal's Fermi level.

With SE cathodes, thermally excited electrons escape over a field-lowered
potential energy barrier. The SE energy distribution contains mostly
non-tunneling electrons which are terminated on the low-energy side by the
field-lowered work function barrier.

CFE cathodes are made from (310)-, (111)-, or slightly oxidized (100)-
oriented tungsten with radii of <O.l Km.

SE cathodes usually are ZrO treated, (100)-oriented tungsten emitters that
provide a low work function localized at their emitter apexes. The ZrO and
tungsten layers in a cathode have excellent thermal stability at 1800 K, the
recommended operating temperature.

This high-temperature stability lets you use SE cathodes at pressures in the
high 10 9-torr range, instead of the low 10-10-torr pressures typically
required for proper CFE operation.

The high operating temperature also minimizes the effects of gases that
adsorb on SE cathode surfaces. As a result, residual gas pressures can be
orders of magnitude higher than what is required by CFE sources without
having a significant effect on surface work functions and beam noise.

Carbon does not poison E cathodes. However, carbon contamination
dramatically alters the work function and emission distribution of tungsten
CFE emitters. Only heating in 10-6torr oxygen can remove it.

The relatively large radii emitters used for the SE cathode, compared with
those of the CFE cathode, are quite forgiving with respect to external
voltage breakdowns, which inevitably destroy pointed cathodes. Larger radius
cathodes have a higher tolerance for nanosecond current spike that
resistively heat small ra cathodes to the melting point and trigger a
destructive arc.

A small virtual source size means -100 times less demagnification is
required by the column optics compared with thermionic sources. This reduces
the number of required lenses and, most importantly, allows a larger
effective acceptance angle at the source for a given object aperture
diameter. This results in more current at the target plane for a given
focused beam size.

Applying electron beams

Attaining high-current levels in a submicrometer electron beam at low
voltages (500 eV to 1 keV) is of particular value in secondary electron
microscopy. SEM images must be made at voltages of 1 keV or less when the
sample is electron beam sensitive, as are biological and semiconductor
materials.

At these low beam voltages, relatively high currents are required to bring
out detail and minimize edge effects. Traditional SEM cathodes using
tungsten hairpin filaments and LaB, filaments are severely limited at low
voltages. They simply cannot deliver high enough currents into submicrometer
diameter beams at the low-voltage end.

High brightness cathodes based on field emission technology produce the
current and spot size combinations at the low voltages required for
high-quality and high-magnification SEM analysis of electron beam sensitive
materials.

For example, Schottky field emission electron cathodes, in a compact single
lens column that can be attached to any UHV vacuum system, provide a 25-keV
electron beam with spot diameters as small as 50 nm. SEM instruments with
several magnetic lenses achieve beam diameters as low as 1 nm using Schottky
cathodes.

High-current-density submicrometer electron beams also offer several
advantages in surface analysis techniques, such as scanning Auger
microscopy, for which small focus beam sizes and high-current densifies are
key requirements. Thermionic sources, such as LaB 6 and tungsten filaments,
continue to be used in focusing columns that support less-demanding SEM and
surface analytical applications.

One of the newer applications of electron beams based on SE cathode sources
was reported in the July/ August 1989 issue of Journal of Vacuum Science
Technology. It involves their use in a scanning reflection high-energy
electron diffraction (RHEED) system developed by a team led by T. Isu at
Optoelectronics Technology Research Laboratory in Japan.

The researchers developed their RHEED system using an SE cathode for the
incident electron beam. The system's electron beam can be focused to less
than 100 nm to scan substrate surfaces during growth by molecular beam
epitaxy (MBE). Experiments indicated that the analyzing beam had no
measurable effect on the MBE growth process.

The images obtained were highly sensitive to surface structure and, with
GaAs (001), revealed granular anomalies (from undulations of the surface
with wavelengths of 0.1 to 0.5 um) over the MBE-grown layers. These surface
anomalies developed along the (110) direction during growth at temperatures
of 570 C and lower.

The ability to directly observe island formation and other three-dimensional
structures during MBE growth under practical conditions with RHEED is a
significant advance in molecular beam epitaxial growth-monitoring
technology.

Ions from liquid metals

Producing high-current-density submicrometer ion beams requires a liquid
metal ion source.

With an LMI source, liquid metal (typically gallium) migrates along a needle
substrate. A jet-like protrusion of liquid metal with a cone half angle of
49 deg (Taylor cone) forms at the source tip under the influence of an
electrical field. The source operates in a vacuum of less than 10 torr.
The gallium-to-gallium bonds are broken under this extraction potential and
uniformly ionized without droplet or charge cluster formation.

LMI sources have extremely high brightness levels (10 6 A/cm 2 sr) and a
small energy spread, making them the best available devices for producing
high-current-density submicrometer ion beams.

For example, using a gallium liquid metal ion (LMI) source in a single-lens,
25-keV focused ion column will produce a 100-nm-dia beam with current
densities of up to 0.7 A/CM 2. Typical, commercially produced two-lens
columns create beam diameters smaller than 50 nm and current densities up to
8 A/CM 2.

Although LMI sources typically deliver gallium ion beams other pure
elemental sources also are available, including indium and gold. Several
other beam types can be generated using alloy sources that require mass
separation capabilities in the column. This is particularly important for
semiconductor dopant materials boron, arsenic, phosphorus, silicon, and
beryllium. Finely focused ion beams of these elements can be used to perform
maskless implantation and metal layer patterning with submicrometer
dimensions.

Single-element LMI sources are not practical for certain dopants due to the
dopants' relatively high melting points and low vapor pressures. Liquid
metal alloy sources however, when used with an appropriate carrier metal,
can deliver these dopant ions. In fact, they can produce more than one ion
species, and the ion species of choice can be selected using mass separation
techniques.

Using alloys not only lowers dopant melting points to a practical level for
LMI sources, it also keeps highly volatile elements such as arsenic
chemically bound and usable in a vacuum.

Ion beam power

On the ion column side, two application areas dominate. The first relates to
microfabrication and includes submicrometer focused ion beam (FIB) milling
and micromachining for IC repair, maskless implantation, circuit fault
isolation, and failure analysis of samples. Similar applications involve
modifying optoelectronic devices and fiber optics.

Now that ion beams can be produced with diameters smaller than 50 nm, it
also has become possible to fabricate quantum dot and line structures using
FIB milling techniques, sometimes in the same MBE systems that produced the
original thin films.

Failure analysis applications include:

* making a probe hole and/or a probe pad for electron beam testing of a
metal line beneath several layers

* cross-sectioning a particle or defect by milling, then using SEM
imaging to determine in which layers it is located, thus tracking its
origin in the process (see accompanying photo)

* isolating a structure by cutting lines and traces and measuring device
characteristics on a single transistor

* thin sectioning to assist in sample preparation for x-ray, SEM, and
TEM analyses.

When specific metallization lines are cut to isolate a single transistor
for electrical testing or to test a chip redesign, it is useful to be
able to reconnect the circuitry. FIB milling systems have ion-induced
metal deposition capabilities that make this possible.

Ion-induced metal deposition is performed by using a gas injection
mechanism to deliver metallic vapor directly to the target area. The
focused ion beam decomposes the vapor at the specimen surface and
produces precise submicrometer-diameter tungsten lines and pads.
Ion-induced metal deposition also can be used to deposit tungsten probe
pads for electrical testing.

FIB systems with LMI sources can perform extremely high-performance
imaging and milling. But measuring this performance can be confusing.

There are two kinds of resolution-imaging and milling (minimum feature
size). Unfortunately, researchers tend to use "beam size" and
"resolution" interchangeably (and often incorrectly) without precise
definitions.

Beam size is the full width half maximum (FWHM) of a Gaussian current
distribution, measured by finding the 12% and 88% intensity points as
the beam is swept over a knife edge.

However, the imaging resolutions observed in micrographs and milling
line widths are not necessarily equal.

By manipulating image contrast in the imaging mode, you can resolve
details much smaller than the stated beam size using the FWHM
definition.

The object of a high-resolution FIB system is not only high-resolution
microscopy but also high-resolution milling. With milling, it is the
total current distribution that counts. For example, to capture 90% of
the beam current from a Gaussian beam, the diameter would need to be 1.8
times the measured FWHM.

If it is necessary to mill two fine structures very close together, the
effect of the current distribution's tail will be much more significant
than if only one structure is milled. If you're milling two structures,
it might be necessary to take into account the 90% level of the beam
current for the intensity. If you're milling only one structure, 70%
might suffice.

The point is that no clear, unambiguous relationship exists between
image resolution and the milling resolution of a high-performance FIB
system with an LMI source.

FIBs also can significantly improve surface analysis instrumentation
capabilities, particularly secondary ion mass spectrometry. SIMS can
provide surface composition and depth profiles by sputtering material
from a surface with an ion beam and using a mass spectrommeter to
analyze the ions that are emitted.

Some SIMS applications require small beams and high currents for
high-resolution mapping. A gallium LMI column can provide elemental SIMS
imaging with submicrometer resolution. Small beams are used for
micro-area-selective depth profiling in Auger electron spectroscopy
systems. Only the specific micrometer-square area of interest needs to
be subjected to this destructive technique, leaving the rest of the
critical device undamaged.

FIB micromachining is another application that has much potential,
particularly in the laser and optical fields. Work at Oregon Graduate
Institute, Beaverton, reported by a team of researchers led by R.K.
DeFreez in 1989 (Society of Photo-Optical Instrumentation Engineers, Vol
1043) involves creating optical quality surfaces with complex
topographical features.

As part of this work, the researchers use 250-nm-dia focused-ion beams
to produce sinusoidal gratings, sinesquared gratings, and
two-dimensional gratings (see accompanying photo). The two-dimensional
gratings have a grating period of 1.25 Km.

Research now is aimed at fabricating:

* monolithic dual-micromachined coupled-cavity single-frequency lasers
with wavelength separations continuously variable from 0 to 600 GHz

* linear and parabolic turning mirrors for two-dimensionally coherent
surface emitting arrays of lasers

* total internal reflection mirrors to route light in the plane of the
wafer

* single-wavelength micromachined coupled cavity lasers tunable at high
frequencies

* a methanometer with a continuously tunable micromachined coupled
cavity InGaAs/InP optical source for detecting gas leaks

* curved laser mirrors

* submicrometer arbitrarily profiled diffraction gratings for
distributed feedback and distributed Bragg reflector lasers.

Other applications involve using focused ion beams to produce curved
diode laser mirrors and total-internal-reflection mirrors to direct
light from one optical waveguide to another in the plane of a wafer.
R&D

 
To the best of our knowledge, the text on this page may be freely reproduced and distributed.
If you have any questions about this, please check out our Copyright Policy.

 

totse.com certificate signatures
 
 
About | Advertise | Bad Ideas | Community | Contact Us | Copyright Policy | Drugs | Ego | Erotica
FAQ | Fringe | Link to totse.com | Search | Society | Submissions | Technology
Hot Topics
here is a fun question to think about...
Miscibility
Possible proof that we came from apes.
speed of light problem
Absolute Zero: Why won't it work?
Why did love evolve?
Capacitators
Intersection of two quads
 
Sponsored Links
 
Ads presented by the
AdBrite Ad Network

 

TSHIRT HELL T-SHIRTS

 
www.pigdog.org