Note: Descriptions are shown in the official language in which they were submitted.
CA 02249157 1998-10-O1
UNIFORM DISTRIBUTION MONOENERGETIC ION IMPLANTATION
FIELD OF THE INVENTION
The present invention relates to ion implantation of an object to create
certain
surface characteristics. More particularly, it relates to a method and
apparatus to implant
ions in an object in a predetermined pattern, said object having three
dimensional (3-D)
topology.
BACKGROUND OF THE INVENTION
At present, conventional ion beam source implantation methods and systems are
used to implant energetic particles into the surface (near surface) of an
object selected for
ion beam implantation. Ion energy levels determine the depth of ion
implantation in the
object. Conventional ion beam source methods and systems generate the ions
from a
plasma source and than use an extraction - acceleration grid assembly to
extract and
accelerate the ions toward the object selected for implantation. However,
these methods
rely on grids which have plane geometry and are located outside the source. As
a result,
all extracted ions travel in the same direction with little divergence.
More recently, plasma source ion implantation (PSII) methods and systems have
also been proposed as a solution for implanting energetic ions into the
surface of objects
with 3D surface topology. In PSII a plasma is produced in a vacuum chamber,
the object
is immersed into the plasma, and then the object is submitted to repetitive
short, negative,
high voltage pulses. Each negative voltage pulse repels the electrons from the
vicinity of
the surface, and accelerate the ions toward the surface of the object
undergoing ion
implantation.
Conventional Ion beam methods and systems are used extensively in the
microelectronic industry for semiconductor doping. Conventional ion beam
methods and
systems, because they are unidirectional have limited industrial applications
outside the
microelectronics industry. The microelectronics industry generally only has to
be
concerned with implanting ions in the flat planar surfaces of chips. Most
other
technologies which can use ion implantation, including the biomedical field,
deal with
-1-
CA 02249157 1998-10-O1
objects that have 3-D topologies such as artificial joints , stems used in
angioplasty etc.
Most of these applications require a precise implantation of ions in uniform
and
predetermined patterns generally at precise concentrations and depths of
implantation.
In order to achieve these requirements of precise concentrations and depths of
ion
S implantation on objects with 3D topology current methods and systems rely on
complicated manipulation of objects in vacuum. Masking of object also becomes
necessary to control the uniformity of implantation, since the sputtering
rates of
materials has strong dependence to the angle of incident ions.
Plasma source ion implantation allows 3D implantation, however, it has three
main disadvantages: (i) an object under implantation is immersed into the
plasma, during
the pause between the implantation periods (between voltage pulses), the
interactions
between the background plasma and the surface implanted can change the surface
properties (e.g. erode the surface); (ii) Since the voltage pulse shape
determines the
energy of the particles implanted and inevitably particles with energies
ranging from few
eV to that determined by maximum applied electric potential are implanted into
the
object; (iii) the extraction surface is determined by the plasma sheath front,
and the
plasma sheath front expands during the voltage pulse, there is no control over
uniformity
(or non uniformity) of the implanted surface. The term sheath front
corresponds to the
boundary between the region where electrons are repelled (also called ion
matrix) and the
plasma.
SUMMARY
The invention uses a 4 ~ steridian concept which allows the implantation of
monoenergetic ions into a three dimensional surface, in the absence of plasma
surface
interaction. In its simplest form, the 4 ~ steridian concept has a single
extraction grid ion
beam source that surrounds the target. The object to be implanted is biased to
a high
negative DC voltage, and is surrounded by a highly transparent control grid
that defines
an equi-potential surface. The ions are supplied from a pulsed or a DC plasma,
localized
between the control grid and the implanter wall. The ions that traverse the
grid are
extracted, accelerated and imparted into the surface of the object to be
implanted with
-2-
CA 02249157 1998-10-O1
energies determined by the biasing voltage and the ion charge state.
Thus, in one aspect, the invention provides an apparatus for implanting ions
in a
predetermined pattern into an object, which object has a 3-D topology. The
invention
provides a vacuum chamber which defines an interior space. The interior space
of the
vacuum chamber includes a first region adjacent to the interior walls of the
vacuum
chamber. The first region is used primarily for plasma production. The
interior space
includes a second region for ion acceleration and the second region is
substantially
surrounded by the first region. A potential distribution control grid defines
the boundary
between the first and second region. The potential distribution control grid
being
pervious to passage of ions such that when an object to be implanted with ions
is placed
at a predetermined position within the second region and is subjected to an
negative
voltage potential, ions from the plasma generated in the first region pass
through the
potential distribution control grid and are uniformly accelerated towards the
object to be
implanted so that all ions drawn through the potential grid are monoenergetic,
having the
same momentum and are directed towards the object to be implanted.
In the preferred embodiment, the potential distribution grid is held at a
constant
potential, i.e. ground and the object to be implanted is subjected to a large
negative DC
voltage to draw ions from the first region into the second region and
accelerate them in
a constant manner towards the obj ect to be implanted.
However, in an alternative aspect, the potential distribution control grid can
have
its voltage varied to either block the passage of ions through it into the
second region or
to conversely facilitate the injection of the ions into the second region and
their
acceleration towards the object to be implanted.
In one aspect of the invention, the potential distribution control grid is a
mesh
with a multitude of openings. The openings being sized to allow the passage of
ions
therethrough, but being of sufficiently small size to prevent the passage of
radiation
through which may be used in the first region to generate a plasma, said
radiation
potentially being microwave radiation.
In another aspect of the invention, the ions are drawn into the second region
in a
predetermined pattern which provides for uniform distribution of the ions over
the
-3-
CA 02249157 1998-10-O1
surface of the object or target being implanted.
In another aspect of the invention, it maintains the second region free of
magnetic
fields.
The invention also provides a method for implanting ions in a predetermined
pattern in an object with three dimensional topology. The method consisting of
the steps
of generating a plasma in a first region, said first region substantially
surrounding a
second region, then injecting the ions from the plasma in the first region
into the second
region and accelerating those ions in a uniform manner towards an object
located at a
predetermined position in the second region. The ions being accelerated in a
uniform
manner such that they all have the same momentum and are thus monoenergetic.
In another aspect of the method of this invention, the step of injecting ions
into
the second region and accelerating them towards the object to be implanted
comprises
subjecting the object to be implanted with ions to a negative DC voltage.
In yet another aspect of the method of this invention, it includes the step
assuring
that the second region remains free of magnetic fields.
In yet another aspect of the method of this invention, the steps of implanting
the
ions in the object in a predetermined pattern involves uniformly distributing
the
implanted ions in the object.
In yet another aspect of the method of this invention, when microwave
radiation
is used to generate a plasma in the first region, steps are taken to prevent
the migration
of the microwave radiation into the second region and thus avoid the creation
of a plasma
in the second region. In the preferred embodiment, the step of preventing the
migration
of microwave radiation used to generate a plasma in the first region into the
second
region consists of providing a mesh barrier with a multitude of openings large
enough to
allow ions to pass therethrough but small enough to prevent the penetration of
the
microwave radiation into the second region.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will be better understood by an examination of the following
description, together with the accompanying drawings, in which:
-4-
CA 02249157 1998-10-O1
Fig. 1 provides a basic schematic view of the major functional components of
the
present invention;
Fig. 2 provides a partial cut away schematic view of a detailed embodiment of
the
present invention; and
Fig. 3 is view of an exterior side of the vacuum chamber of the present
invention
showing the arrangement of a microwave flange bolted to the chamber.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Overview:
The schematic diagram of the 3-D Ion Implanter (3DII) is shown in Fig. 1. 3DII
has a vacuum chamber 1 evacuated to a base pressure bellow 10-7 torr. The
chamber 1
is made of a conducting material. The Implanter has two distinct regions.
Plasma
production region (PPR) 2 and the ion acceleration region (IAR) 3 The boundary
between these two regions is defined by a potential distribution control grid
(PDCG) 4.
The potential distribution control grid has an important influence on the
distribution of
the implanted ion dose. PDCG 4 defines an equipotential surface, and in the
case of
plasma production by microwaves (or rf ) the grid prevents microwave (rf)
field
penetration into IAR 3. The target to be implanted 5 is placed inside the IAR
3, and is
connected to a negative high voltage. On the target, the implanted ion dose
distribution
is affected by the uniformity of ion (plasma) density near PDCG 4 at PPR 2,
and the
potential distribution in IAR 3. Ion production in IAR 3 must be negligible
(to prevent
broadening of the implanted ion energy distribution), and collisions between
ions and
neutrals in IAR 3 must be rare, in other words the ion neutral collision mean
free path in
IAR 3 must be much longer than the distance between the target and the PDCG 4
(to
avoid slowing down of ions, and prevent broadening of the implanted ion energy
distribution). To avoid changing ion trajectories, the magnetic field
intensity in IAR 3
must be negligible. The plasma interaction with the target 5 must be
negligible, in order
to avoid altering the surface properties by low energy ion bombardment.
For a given operating pressure in the chamber, the distance between target 5
and
-5-
CA 02249157 1998-10-O1
the grid 4 must be large enough to avoid arcing between the target S and PDCG
4;
however, the distance should be such that a glow discharge in IAR 3 is not
initiated.
For implantation of ions from a gaseous source, the gas (or a mixture of
gases) is injected
directly into the chamber.
For implantation of ions from solid (or liquid) elements, sputtered or
(evaporated) particles can be injected into the chamber and ionized
subsequently, either
directly or by a buffer plasma. The buffer plasma can be Ar, He or any other
noble gas.
If deposition (and ion mixing) is to be avoided then no direct line of sight
should exist
between the target and the metallic particle source.
For implantation of nonconducting (or conducting) solid elements, one can also
use a crucible to evaporate the element, or one can use magnetron sputtering.
Detailed Description of 3DII:
The schematic of the prototype 3-D Ion Implanter (3DII) is shown in Figs. 2
and
3. The vacuum chamber 1 is evacuated to a base pressure bellow 10-7 torn. The
chamber
1 is made of a conducting, but non magnetic material, and it is held at ground
potential.
In our prototype, we have used a stainless steel cylindrical chamber which has
a10-cm
inner diameter and is 22-cm long, though the chamber can have a different
shape. The
chamber is initially pumped down to belowl0-7 torn, and then it is filled with
an
operating gas (or a mixture depending upon the intended application).
The plasma in PPR 2 can be produced using different methods, however, to
reduce the impurities in the plasma, electrodeless methods have a distinct
advantage. We
use microwave discharge method to produce a substantially uniform plasma in
PPR 2.
The plasma in PPR 2 near PDCG 4 should be substantially uniform to simplify
the
control of the implanted ion dose distribution.
Magnetic fields can substantially improve the particle confinement, and hence
increase the efficiency of the plasma production process. However, the
magnetic field
intensity in IAR 3 must be negligible to avoid influencing ion trajectories
that bombard
the target, and to prevent electron trapping that can result in plasma
production in IAR
3. Permanent magnets have the advantage that they can be used to engineer
complicated
-6-
CA 02249157 1998-10-O1
magnetic field topologies. Sixteen rows of permanent magnets 6 are placed on
the outer
wall of the chamber, there are six permanent magnets in each row, arranged in
alternating
poles. We use a checkerboard configuration, since the magnetic field intensity
drops off
much faster as the order of multipoles increases. This design allows one to
keep the
magnetic field intensity in the IAR 3 to a negligible level. If the permanent
magnets can
produce a magnetic field in the PPR 2 region that is perpendicular to the
microwave
electric field, and has an intensity equal to B = 2Bmf~ W/e (where a and m are
electron
charge and mass, and f~W is the microwave frequency), then ionizing the gas
with
electron cyclotron resonance (ECR) coupling becomes possible. This scheme is
particularly attractive, and is efficient for plasma production at low
pressure ranges (10-5
- 10-4 torr range), where conventional microwave cavity discharges are less
efficient.
The preferred embodiment uses microwaves at 2.45 GHz, and the field intensity
for ECR coupling is 875 G. Such field intensity can readily be produced with
NdFeB
(used in our prototype) and other rare earth permanent magnets. In the
prototype the
1 S resonance zones are about 1.0 cm from the wall. The permanent magnets are
kept near
room temperature to ensure that they retain their magnetization. Each row of
permanent
magnets is sandwiched between two water cooling copper tubes 7. The tubes are
brazed
to the plasma chamber, and the heat transfer between the copper tubes and the
magnets
is enhanced using a thermic paste.
A temperature sensor is used to monitor the temperature of the plasma chamber.
The soft iron layer 8 acts as return winding for the magnetic field. Eight
iron screws 9,
which bolt down the microwave flange 10 to the source chamber, are also used
to transfer
the magnetic flux from the magnets installed on the surface of the flange to
the interior
of the plasma chamber. A quartz window 11-a provides the vacuum seal between
the
plasma chamber and the microwave waveguide. The window is placed in a
different
location l lb if there is a possibility of depositing material on the window.
Instead of
quartz window, any other low loss material, suitable for vacuum sealing, can
also be
used. Microwaves can also be introduced into the chamber 1 by an antenna.
The PDCG 4 is made of a partially transparent conductor (mesh), the mesh size
should be small enough to prevent substantial microwave penetration into IAR
3. In the
_7_
CA 02249157 1998-10-O1
prototype the mesh is made of stainless steel, with a spacing of 0.1 cm and
about 60%
transparency. However, any conducting mesh with somewhat different spacing and
transparency can also be used, provided the spacing remains small compared to
the
microwave wavelength. PDCG 4 is also connected to a H.V. feedthrough 12, so it
can be
biased using an independent power supply 14. Biasing PDCG 4 in turn will
enhance the
extracted ion current from PPR 2; however, biasing PDCG 4 will enhance
sputtering
from its surface. If the presence of these sputtered particles are not
tolerated on the target
5, then the PDCG 4 must be kept at ground potential. The working gas can be
introduced
into the chamber via small port 1 S and a control valve. The same port can be
used to
introduce material by evaporation.
The target to be implanted 5 is placed inside the IAR 3, and is connected to a
negative high voltage 13 via a high voltage feedthrough 12. Ion production in
IAR 3
must be negligible, and the ion neutral collision mean free path must be much
longer than
the distance between the target and the PDCG 4, otherwise the ion neutral
collisions will
slow down the ions, and will broaden the ion energy distribution. There are
three
important factors to be considered in determining the distance between the
target 5 and
PDCG 4. The distance should be large enough to avoid arcing, but kept small
enough to
increase the extracted current (increase the flux), and avoid creating a glow
discharge in
IAR 3. The operating pressure in the prototype implanter has been below 1
mtorr, and the
maximum negative voltage applied to the target has been - 35 kV. A 1.0 cm gap
between
the target S and PDCG 4 has proved to be an appropriate compromise.
While the invention has been particularly shown and described with reference
to
a preferred embodiment thereof, it will be understood by those skilled in the
art that
various changes in form and detail may be made to it without departing from
the spirit
and scope of the invention.
_g_
