Sputter by S. M.

Rossnagel

deposition for
semiconductor
manufacturing

Sputter deposition, also known as physical deposition of metal thin films used to form vias and lines,
vapor deposition, or PVD, is a widely used as well as the various related thin films which function as
technique for depositing thin metal layers on diffusion barriers, adhesion or orientation layers, or seed
semiconductor wafers. These layers are used layers. Sputter deposition is usually carried out in diode
as diffusion barriers, adhesion or seed layers, plasma systems known as magnetrons, in which the
primary conductors, antireflection coatings, cathode is sputtered by ion bombardment and emits the
and etch stops. With the progression toward atoms, which are then deposited on the wafer in the form
finer topographical dimensions on wafers and of a thin film. Depending on the lithography scheme,
increasing aspect ratios, the broad angular these films are then etched by means of reactive ion
distribution of depositing, sputtered atoms etching (RIE) or polished using chemical-mechanical
leads to poor or nonexistent coverage in deep polishing (CMP) to help delineate circuit features.
features. This has been partially addressed This paper explores the plasma technology relevant to
using directional sputtering techniques such sputter deposition as applied to semiconductor technology,
as collimated sputtering or long-throw the sputtering process itself, and then semiconductor
sputtering. More recently, work originating in applications. In this latter area, recent (less than ten
IBM has moved toward the deposition of films years) developments in sputtering, such as collimated
from metal-rich plasmas fed by sputtering, sputtering, reflow, and ionized deposition, are examined.
a technique known as l-PVD (for ionized Some of these are in wide use in manufacturing; others,
PVD). This technique, based on fairly minor just becoming available, are expected to constitute the
modifications of existing PVD systems, solves sputter-deposition processes of the future.
many of the intrinsic problems of PVD and
appears headed for widespread manufacturing Plasma technology
applications. The most widely used technology for sputter deposition
is based on the magnetron cathode. Originally, physical
sputter deposition utilized dc diodes, which were simply
Introduction parallel plates powered by a power supply of several
Sputter deposition is one of the most widely used kilovolts in a working pressure of several tens to several
techniques for the fabrication of thin-film structures hundreds of mTorr. The negative plate, also known as the
on semiconductor wafers. It is used primarily for the cathode, was bombarded by ions from the plasma set up

^'Copyright 1999 by International Business Maeliines Corporation. Copying in printed form for private use is permitted witlrout payment of royalty provided that (1) eaelr
reproduetion is done williout alteration and (2) {he Journal reference and IBM copyrigiit notice arc inclnded on the first page. The title and abstract, but no other portions,
of this paper may be copied or distributed royalty free without further permission by computer-based and other information-service systeirrs. Permission to republish any other
portion of this paper must be obtained from the Editor. 163
0018-8646/99/$5.00 ® 1999 IBM

IBM J. RES, DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999 S. M. ROSSNAGEL

 capacitor to ground and a series 3-4-turn inductor, are
located along with the series capacitor in the "matchbox,"
which is physically located adjacent to the cathode
position. The inductor is fixed, and both of the capacitors
X X X X X
X*«j,-X (shunt and series) are variable. A control circuit within
,' X \ X X ,' \x the matchbox controller senses the reflected power (from
X ' X
A X .' 'x
the matchbox and plasma back to the power supply) and
X 'V'^
• i X1 /x
adjusts the variable capacitors to minimize the reflected
'-.!<,•' X\ ^ x '"X X '>-' X
X X X X
power. Usually this is done automatically by means of
Electron Electron speed reversible motor drives on the capacitors, but occasionally
speed decreases here
increases laboratory-based systems will have manual controls for the
here tuning network.
In the 1980s, rf diode sputter-deposition systems were
widely used for the deposition of silicon dioxide dielectric
films. By splitting some of the applied rf power between
j .Schematic of motions of electrons in crossed E and B fields. The
the cathode (typically fabricated from silicon dioxide) and
I vertical electric field E is consistent with the presence of a cathode
I located at the bottom of the figure. The magnetic field is oriented the sample pedestal (where the wafer or wafers were
j perpendicularly to the page. mounted), it was possible to deposit films with a moderate
degree of ion bombardment of the growing films. In effect,
the les functioned as partial cathodes in the circuit
and were bombarded by ions from the plasma, albeit at a
lower rate than the primary cathode. This is currently
between these two plates, and cathode atoms were known as bias sputtering and can be useful in helping
dislodged from the metal surface. These atoms could then to densify as well as planarize the depositing film.
deposit on other surfaces inside the vacuum system, Planarization results in a flatter film with smoother
forming films. The dc diodes were characterized by slow- coverage over steps and gaps in the underlying substrate.
deposition rates, high voltages, and low currents, and The principal type of system currently used for high-rate
hence are no longer of interest. They were also inadequate deposition of metals, alloys, and compounds is known as
for the deposition of dielectric films because of charging, the magnetron cathode system. This type of tool uses
arcing, and ver}' low deposition rates. magnetic confinement of electrons in the plasma, which
The second evolution of sputtering technology was to results in a higher plasma density than in either the rf or
replace the dc power supply with an rf supply, generally dc diode systems. The higher plasma density reduces the
operating at a frequency of 13.56 MHz. This change discharge impedance and results in a much higher-current,
eliminated the charging and arcing problems with lower-voltage discharge. As a rough example, an rf diode
dielectrics and also led to slightly higher deposition rates. tool operating at 2 kW might have a peak-to-peak rf
The oscillating rf potential applied to anode and cathode voltage of over 2000 V. A conventional magnetron system
resulted in a modification to the electron motions, which operated at 2 kW might have a dc discharge voltage of
produced better energy coupling to the electrons as well 400 V and an ion current of 5 A to the cathode.
as higher plasma densities [1]. The electron confinement on a magnetron is due to
Because of the high electron mobility in a plasma (rf or the presence of orthogonal E and B fields at the cathode
dc), an rf diode system tends to develop large electron surface. This results in a classic E x B drift for electrons
currents during the positive portion of the applied rf cycle. (the Hall effect), which gives rise to a sequence of
Usually a large capacitor (500-2000 pF) is placed in series cycloidal hopping steps parallel to the cathode face
between the rf power supply and the powered electrode, (Figure 1). As a result, the secondary electrons which are
or electrodes if the anode is not grounded. The large emitted from the cathode because of ion bombardment
series capacitor allows a significant negative bias to are confined to the near vicinity of the cathode. In a
develop on the cathode, typically half of the value of the magnetron, the electric field is always oriented normal to
applied peak-to-peak rf voltage. This bias is then the the surface of the cathode. The transverse magnetic field
acceleration voltage for ions from the plasma, which move is configured so that the £ X B drift paths form closed
much too slowly to respond to the applied rf potentials. loops, in which the trapped, drifting electrons are
In addition to the series capacitor, it is common constrained to circulate many times around the cathode
to use two other tuning components to help match the face.
impedance of the plasma to the output impedance of the The most common design for a magnetron cathode is
164 rf power supply. These components, usually a shunt the circular planar cathode, in which the cathode is simply

S. M. ROSSNAGEL IBM .1. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/M.ARCH 1999
a flat, circular disk, and the E X B drift currents form
circles centered on the disk axis, around the face of the EXB - Magnetic field
drift path -
cathode (Figure 2). The magnitude of the total drift
current can be measured by means of its induced magnetic
field, and results (Figure 3) indicate a ratio of about 3-7 ii
for the circulating current compared to the discharge, or
net current [2J. On average, this ratio is a measure of the
trapping and the number of times a secondary electron Pole-piece assembly
traverses the E x B drift loop prior to leaving the (a)
discharge and arriving at the anode. This ratio, though, is EXB drift path
accurate only for the specific size of cathode used in that
study (150-mm diameter), and more accurately, perhaps,
indicates the length of the drift path for the electrons.
Magnetic
This allows the data to be scaled to larger cathodes or field lines
ones with differing E X B paths.
The magnetron effect is generic: The requirement is
that the E X B drift paths be closed. Several geometric
perturbations have been developed over the years for
specific applications, and examples are shown in Figure 4.
The intrinsic deposition uniformity of a magnetron-type
cathode is not good, though, and this has implications (b)
for semiconductor processing. The conventional, circular
planar cathode is characterized by high levels of erosion
under the £ X B drift path (also known as the etch path),
which is in the form of a ring. At short cathode-to-sample, Schematic of a circular planar magnetron cathode: (a) Side view;
or throw, distances, this ring source shows up as a ring- (b) top Me»
shaped deposition on a planar sample, At moderate throw
distances (of the order of the ring diameter), the profile is
flatter, since the center region fills in somewhat; at large
throw distances, the source functions much as a point
source. Unfortunately, there is no throw distance at which
-
the deposition uniformity approaches that required for
~
>/-•
y*' 0
semiconductor processing, which is of the order of 1-2%
(la). -
In many magnetron deposition configurations, it is . P /
acceptable to physically move the sample during
-
deposition to effectively average out the nonuniform
•

»-- ^'
A' ^-<-'•'
•
deposition flux in order to create a uniform film. In M ay
semiconductor processing, however, this is not an Y^M' - 4 - 10 mTorr
acceptable solution, since it introduces additional motion, j^g^^A
" - O - 20 mTorr
feedthroughs, complexity, and (most significantly) y>%^ - • - 30 mTorr
increased particulate formation to the deposition chamber. ,
An acceptable solution, though, has been to physically 2.4 1 2 3 4 5 6 7 8 10
translate the E X B drift path across the surface of the Magnetron discharge current (A)
cathode to average out the intrinsic nonuniformity. This is
achieved by configuring the magnets behind the cathode
(which form the B field at the cathode surface) to move
Measured EXB total drift currents (the circulating current) as a
by means of a motor drive (Figure 5). function of discharge current The cathode used was Cu, and the
Since the magnets are outside the vacuum system, their working gas, was Ar. Fri»m |21. with pernii.ssion
physical motion does not introduce additional particulate
formation inside the chamber, in addition, the more
uniform erosion of the cathode surface results in better
utilization of the (high-purity) cathode materials, rotating-magnet, heart-shaped E X B drift-path
increasing the time before required maintenance. The magnetron is the primary type of cathode currently used 165

IBM J. RES. DEVELOP. VOL. 4,1 NO. 1/2 JANUARY/MARCH Wm S. M. ROSSNAGEL

 Cathode
Anode

drift path
Magnetic
field

(a)

EX B Anode Magnetic field

drift path lines Cathode

Cathode - Magnetic
field

£ X B drift path
(b) (d)

Various magnetron geometries: (a) Rectangular (top view); (b) S-gun (cross section); (c) cylindrical post; (d) hollow cathode.

in semiconductor processing. Small variations in the reflection of the depositing atoms at other surfaces,
magnet configuration have also been developed to tailor less-than-unity sticking of the depositing atoms, or even
the deposition uniformity under various deposition resputtering from other surfaces in the chamber [4].
conditions (such as changes in throw distance or pressure) Redeposition onto the cathode is generally ignored
or with different cathode materials. As such, a class of because the erosion rate is usually so large. However,
these magnetrons might have 3-10 different magnet in the edge regions (and also at the very center in some
configurations, which are changed depending on the designs), the deposition rate can exceed the erosion rate,
application and tool configuration. resulting in film buildup and eventual flaking or nodule
There are also variations of this general design which formation. Sweeping the £ x 5 drift paths off the edge of
allow the E X B drift path to spill slightly over the edge the cathode results in the elimination of the edge regions
of the cathode. This reduces the intrinsic discharge as a possible contamination source. However, it must be
efficiency and may result in modest increases in the done carefully so as to limit the possibility of sputtering
operating voltage. However, the design reduces the of (and sputter deposition from) the side areas of the
presence of areas near the cathode edge which might not cathode.
be heavily eroded. In a magnetron system, the net erosion Magnetron sputter-deposition systems have evolved
and deposition rates can easily exceed the several-micron- into a near-UHV environment to reduce the influence of
per-minute range. While the deposition is mostly forward, residual gas atoms on the structure of the thin, deposited
there is a small component which can deposit back onto films. While it may seem surprising that there are
166 the cathode surface because of in-flight gas scattering. contamination concerns with films deposited at 1 ^m/min.

S. M. ROSSNAGEL IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH IW)

 the impurity levels even in a moderate vacuum system
can be significant. A deposition rate of 1 /iim/min of Al
corresponds to an atomic deposition rale of about fifty
monolayers of Al per second. In a base pressure of
1 0 " Torr, the arrival rate of background gas atoms is
1 monoiayer/s, and if all of these atoms were incorporated
into the film, it would constitute a 2% impurity. These
numbers scale down linearly with base pressure, so that at
a base pressure of 10 Torr, the impurity flux can still be
two parts in 10 . When this is compared to the purity of Cathode
a typical cathode (99.999%), the effect of the vacuum
system can become very significant.
Current manufacturing-scale magnetron systems are
constructed from stainless steel with a minimum of EXB
etch path
o-rings. They are typically configured with cryopumps
connected directly to their deposition chamber by means
of large-diaraeter valves, and the resultant base pressure is
generally in the low 10^ -Torr range for most cathodes,
and in the 10""-Torr range for Ti. for which the
chemically active nature of the deposited films can
contribute appreciably lo the net pumping speed of the
system.
(b)
The working pressure during sputtering is typically
0.5 to 5 mTorr, which requires a gas flow of many tens of
standard cubic centimeters per minute (seem). Because of
base-pressure considerations, manufacturing-level systems Heart-.shaped magnetron: (a) Cross section .sliowing motor drive and
are not baffled and therefore retain approximately the rotating magnet plate; (h) front view of cathode .surface stiowing Itie
true base pressure of the chamber during deposition. etch path. From [31, with permission.
Their large gas flows, however, can lead to the need for
frequent regeneration of their cryopumps, which is usually
automated and accomplished during routine maintenance.
The throw distance for commercial sputter-deposition
systems varies depending on the material to be deposited, diameter is roughly 50% larger than the wafer diameter,
the application, and the supplier. Since in most tools the with diameters of about 30 cm (12 in.) for 200-mm
substrate position is movable vertically for the purpose of wafers, and 45 cm (18 in.) for 300-mm wafers. There is no
picking up or clamping the wafer, the throw distance is intrinsic limit to the cathode dimensions other than that
usually a controllable variable, and varies from 3 cm at due to structural considerations, and cathodes as long as
a minimum to 1(1 cm. In a variant on sputtering to be several meters In length have been used. However, for
discussed below, known as long-throw deposition, the semiconductor applications, the dimensions approaching
throw distance is increased to 25-30 cm.
the 300-mm generation are somewhat problematic,
The magnetron chambers used for large-scale
because the cathode plate can be somewhat deflected by
semiconductor applications are configured as ports on
the large span and the pressure of the back-side cooling
an integrated-process, load-locked tool, and wafers are
water.
introduced to the deposition chamber via a load lock
The cathodes are water-cooled at a rate of many gallons
(Figure 6). In general, this type of system has its worst
per minute, and arc typically rated at powers of 20-30 kW.
vacuum in the load-lock region, and each successive
There is no fundamental plasma limit to this power
chamber from the load lock has a better (i.e., lower) base
pressure. Such systems are also configured with degassing rating: It is limited primarily by the ability to cool the
stations for the wafers, which are necessary to remove cathode by means of flowing water. The power supplies
water vapor that the wafers absorb from air exposure prior are switching supplies, typically ganged together in 10-kW
to deposition. Generally, the degassing temperature is increments. The supplies have fairly sophisticated arc-
50-100°C higher than the deposition temperature. detection and -suppression circuitry to reduce unipolar
Commercial magnetron cathodes are almost always arc formation, which can result in spitting of microscopic
configured for dc rather than rf power. The cathode droplets onto the wafer substrate. 167

IBM J. RES. DEVELljr. VOL. 43 NO. 1/2 JANUARY/MARCH 1<)'» S. M. ROSSNAGEL

 Sputtering background
Physical sputtering has been known for more than a
Preclean, PVD, etc. High vacuum to UHV
(= 0.5-20 mTorr) lO-'-lQ-' Torr) hundred years and has been in common usage for many
decades. Several good reviews of the topic are available
[5, 6], and it is covered only briefly here. Physical
High vacuum
sputtering is a relatively violent, atomic-scale process in
Transfer module which an energetic particle strikes a solid, resulting in the
(= lO^MO^* Torr)
emission of one or more substrate atoms from the solid.
The dynamics of the collision process depend strongly on
Medium vacuum
the incident energy and mass of the bombarding particle.
Degas and transfer
(= lO-^-lO"' Torr) At relatively low energies, the incident particles do not
have adequate energy to break atomic bonds of the
Cassettes and wafer Low vacuum f
load/unload (= 10^^-10^' Torr) I
surface atoms, and the bombardment process could result
in simply desorbing a few lightly bound gas atoms, perhaps
(a) inducing a chemical reaction at the sample surface, or
nothing at all. At relatively high incident energies, the
bombarding particles travel deeply into the bulk of the
substrate and may cause deep-level disruptions in the
physical structure, but few if any surface atoms are
released. At the moderate energies, typically in the range
from several hundred eV through several keV, the
incident particle can cause substantial numbers of near-
surface broken bonds, atomic dislocations, and ejection
or sputtering of atoms.
The parameter most used to characterize sputtering is
the yield S, which is simply the ratio of the number of
emitted particles to the number of incident ones. The
yield is an average number, including not only the
emission rates from a number of different crystalline
(b) orientations, but, on the atomic scale, a number of
different impact points for the incident particles. In this
moderate-energy range of interest, the yields for Ar^
bombardment of most materials range from 0.1 to about 5,
Schematics ot (a) a manufacturing-scale, integrated-process PVD with the majority of materials in the 0.5-2 range. The
system showing vacuum levels; (b) a widely used PVD cluster
system manufactured by Applied Materials Corporation, Santa yields are energy-dependent in a roughly linear fashion
Clara, CA (with permission) (Figure 7), such that an increase in the incident ion
, T ''ri-Af '
energy of 2X results in an increase in the yield of slightly
less than 2x. It is this near-linear dependence of the yield
that makes sputter deposition roughly power-dependent.
Since the rate of sputtering (and sputter deposition) scales
directly with the bombarding flux (i.e, current), the near-
The cathode materials currently used are generally
linearity of the sputtering yield with voltage leads to the
Al(0.5% Cu), Ti, Ta, and Cu. The purity is generally
essentially linear relation between the deposition rate
99.99% for the Ti and Ta cathodes, and 99.999% or
and the discharge power.
higher for the Al(Cu) and Cu cathodes. The high-purity The emission profile for the sputtered atoms is
disks, typically about 1 cm thick, are diffusion-bonded to characterized by a cosine distribution for most materials.
Al or Cu backing plates which form the vacuum seal as This means that the emission rate at some angle other
well as the cavity for both the magnets and the water than normal (perpendicular) is equal to the normal
cooling. The cathode plate must be insulated from the incidence emission rate times the cosine of the angle from
grounded vacuum system, and this is achieved by means the normal. This is usually drawn as a circle touching the
of a 1-cm-thick ceramic ring and o-rings. These large- impact point, in which the circle is the envelope of the
diameter o-rings (typically 14 in. diameter X 1/8 in. thick) magnitudes of the emission at other angles. Various
limit the base vacuum pressure of the deposition chamber departures from the cosine distribution are seen,
168 to the 10~*-Torr range. depending on the ion energy and the sample structure.

S. M. ROSSNAGEL IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999

As a function of ion energy, the cosine distribution is seen
to be flattened at lower energies ("under-cosine"), in which
there is more emission at lower angles than at the surface
normal, and "over-cosine," or more peaked in the normal
direction at higher energies (Figure 8). There are also
unusual cases of directed emission for specific crystalline
orientations of the target. The angular emission profiles
for a number of orientations are shown in Figure 9.
The cosioelike angular emission profile of the sputtered
atoms, coupled with the extended area of the magnetron
cathode and perhaps some in-ilight gas scattering, implies
that the depositing flux of atoms has a very broad,
almost isotropic nature. This can be extremely useful in
depositing films on unusual, nonplanar surfaces, over
steps or ledges, and on the sides of features. As a result,
sputter-deposited films are smoother than the underlying
surface. However, if the surface is composed of high-
aspect-ratio (aspect ratio = depth/width) holes or
trenches, it is unlikely that the deposition will cover or fill 100 1000 10,000 100,000
the structures entirely. This latter aspect forms the basis Ion energy (eV)
for much of the rest of this paper.
Sputter-deposition rates can, in principle, be obtained
from sputtering yields and some consideration of tool
Phjsical spmtermg yields as a function of Ar ion energy for
geometry and gas scattering. This is almost never done, several materials. From [7]. with permission.
since the details of tool configuration, shielding, shutters,
etc. are very tool-dependent, and the experimental
measurements are easy to perform. Sputtering rates are
often described in terms of the incident power, using units
of A/s/kW. A typical number for Al deposition would be
about 1.0, and for Ti about 0.2. Perturbations of the
system configuration, such as collimation or long throw 3kV
(discussed below), can reduce these rates significantly. To
2kV N
first order, though, the deposition rate can be predicted at
moderate powers from this sort of power-rate indicator, X N

and the rates should also scale roughly with the sputter \ \ •
yields described above. _olkV \ \
Sputter deposition is managed by deposition time, and I / \ i I
there are rarely any in situ diagnostics of deposition rate \i /o .60flV ^ I *
used in manufacturing-scale tools. The rate is calibrated s \
•^-A ^ ri
against time, and then films are deposited for a fixed time N \ \
f 1 • •

period. It should be noted that the linearity of deposition To/

rate with time or power fails at very short deposition times
(very thin films) or very high levels of discharge power.

Patterning techniques for semiconductor

processing
There are two families of circuit-patterning techniques Angular emission distributions for sputtered atoms at several ion
currently used in semiconductor manufacturing. The eneigies. From |8), with permission.
choice of which is used is determined by deposition
capabilities, required feature dimensions and aspect
ratio, and material requirements.
The first of these two classes is based on the deposition etching (RIE) to form circuit elements. These features are
of planar, metallic films, which are then patterned using a then encapsulated with a dielectric (for example, silicon
light-sensitive photoresist and etched using reactive ion dioxide), which is subsequently polished flat. Vias can then 169

IBM L RES. DEVELOP. VOL. 43 NO, 1/2 JANU.'VRY/MARCH IW9 S. .VI. ROSSNAGEL
 -90° •

Polycrystalline Al with Single-crystal (100) Al

strong (100) texture
(b)

0°

. ^: ^ ^ fF\r7->,
• •'

% :

""^••'^i^Si^'-'--""
-90° '90° -90° 0 '90°
Polycrystalline Al with Single-crystal (110) Al
strong (110) texture
(c) (d)

V*

Calculated (solid lines) and measured angular distribution of sputtered Al atoms from similarly oriented polycrystalline and single-crystalline
Al targets. From [9], with permission.

be opened into the dielectric, filled with W by means of deposited. Then, using photolithography and R I E , holes
chemical vapor deposition (CVD), and polished flat prior and trenches are etched into the dielectric. The next step
to the deposition of the next, planar metal film (Figure 10). is to fill the cavities by means of a metal-deposition
This technique is known by several designations, such as process, after which the excess metal is polished off,
R I E - m e t a l patterning, or cloisonne patterning. Because back to the original dielectric surface. This results in an
of the need to reactively etch the deposited metal, this embedded line or via in a planar surface. Another layer
scheme is limited to Al-based conductor systems because of dielectric could then be deposited to form the basis of
of the relatively low vapor pressure of the resultant CuCl the next metallization layer.
product molecules from the R I E process. Metal deposition In many ways, R I E - m e t a l and damascene processes are
for the R I E - m e t a l process is virtually always done with simply the opposite of each other. In the first, the metal is
sputter deposition from magnetron sources. The broad patterned and the dielectric is used to fill in the gaps; in
angular distribution of the sputtered atoms leads to good, the second the dielectric is patterned and the metal fills
continuous coverage over small bumps and ledges on the in the features. However, there are practical issues which
surface, densities of the deposited films are close to bulk come into play in the applications of these techniques,
levels, and with a small degree of annealing, the films such as the previously mentioned difficulty in reactively
display near-bulk electrical conductivity. etching Cu films.
The second class of patterning techniques currently From the point of view of topography, the RIE-metal
used is known as damascene processing, after its similarity processing techniques, which leave behind a freestanding
to ancient jewelry-inlay processes [If]. In damascene metal feature prior to dielectric encapsulation, arc limited
170 processing (Figure 11), a thick, planar dielectric layer is to features with modest aspect ratios, perhaps of the order

S. M. ROSSNAGEl, IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999

 Sill.

1 :i :

^^
1 w •' W
1 S""' i s S- |.|

(b)

Process steps for torming cloisonne wuing stiuctuie (a) Metal Piimary steps in damascene process (a) Oxide deposition, (b)
stack deposition, (b) pliotolithogiaphy, metal RIE. (c) oxide photolithogiaphy and oxide RIE, (c) metal deposition, (d) metal
deposition; (d) oxide CMP. From [10], with permission CMP.

of 1:1, whereas the damascene techniques allow the use of layers to facilitate a desired orientation in a subsequently
much deeper features. However, from the point of view of deposited conducting layer.
metal deposition, there are significant difficulties in the In each case, the general requirement for these layers is
development of viable metal-deposition processes for high- that they be thin and yet conformal within deep features.
aspect-ratio damascene features. Sputtering, which is These requirements tend to preclude conventional sputter
routinely used for RIE-metal depositions, fails at filling deposition as a viable technique because of the wide
features with aspect ratios greater than about 0.5 because divergence of the depositing atoms. As a result, significant
of the wide angular spread of the depositing atoms. The effort has been expended in the exploration of CVD-based
feature that makes PVD attractive for planar films inhibits deposition methods, which tend to be more conformal in
its application to damascene processing. deep features because of the relatively low sticking or
An additional feature for both RIE-metal and incorporation coefficients of the reactants. Sputter
damascene processing is the need for film layers in deposition, however, can still be a viable, less expensive
addition to the principal layers. Examples: Diffusion solution to many of these interconnection applications
barriers are often required on the sidewalls or bottoms of with subtle modifications to the sputtering, transport,
vias or interconnections to prevent the interaction between or surface processes. Since sputtering is a widely used
two materials; a barrier of some kind is needed at the manufacturing-scale technology, it is desirable to
interface between a W stud and an Al metal line to extend it further into future generations.
reduce spiking or void formation at the interface; and a
continuous, hermetic film is needed on the sidewalls of a Directional sputter deposition
silicon dioxide via during CVD deposition of W to prevent Sputter deposition is, on a macroscopic scale, a nearly
the reaction between the oxide and the WF,^ working gas. isotropic deposition process when used at short-throw
Other thin layers may be required to function as adhesion distances with a wide-area cathode. However, on an
layers, seed layers for electroplating, or orientation atomic scale, the sputtered atoms tend to travel in straight 171

IBM J. RRS. DEVRl,OP. VOL. 43 NO. 1/2 JANUARY/MARCH Wn S. M. ROSSNAGKI,

 Long-throw deposition at the edge of a wattr showing (a) good symmetry when oriented parallel to the wafer edge; (b) poor symmetr>' when
oriented radially.

lines without in-flight collisions from the cathode to the 5 cm to 25-30 cm (for 200-mm wafers), which is roughly
sample at the pressures most commonly used. Since the equivalent to the cathode diameter. A requirement of long-
sputtered atoms are virtually all neutrals, it is not possible throw deposition is that the operating gas pressure be low
to redirect their trajectories in flight. However, two enough that in-flight gas scattering is minimal. At 25 cm,
techniques have evolved which can filter the angular this means a working pressure in the low 0.1-mTorr
distribution of the sputtered atoms, resulting in a more range, which is just about the limit for conventional
normal-incidence deposition process. The laterally moving magnetron operation. In the original work, hollow-cathode
sputtered atoms contribute most to the pinching-off electron sources were used to allow even lower-pressure
deposition effect in high-aspect-ratio features. If these operation, but hollow-cathode enhancement is generally
atoms are filtered, the more vertically oriented (i.e., impractical for manufacturing-scale operation using
vertical to the sample plane) sputtered atoms can travel rotating-magnet magnetrons.
more readily down into high-aspect-ratio features. The two With a throw distance of 25 cm, and assuming a purely
filtering techniques are known as long-throw deposition andballistic, unscattered deposition, the angular divergence of
collimated sputter deposition. the arriving atoms is limited to about ±45°, or an effective
aspect ratio of about 1:1. Atoms which travel at higher
• Long-throw deposition angles are unable to reach the sample, and deposit on the
Long-throw deposition is simply an increase in the chamber walls. This reduces the net deposition rate by
cathode-to-sample or throw distance over conventional 7fJ% or so and by implication increases the frequency at
systems. It was first used to mimic evaporation for the which the tooling or shielding inside the chamber must be
deposition into lift-off mask features [12]. In a long-throw cleaned to reduce flaking and particulate formation. Long-
172 system, the throw distance is increased from perhaps throw deposition can be used to fill low-aspect-ratio

S, M. ROSSN.AOEL IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 J.ANUARY/MARCH 1999

features (1:1) or to form reasonably conformal liners at
aspect ratios of about 2:1. There is also the possibility Magnetron cathode
of using a significantly increased sample temperature to
facilitate enough surface mobility to fill higher-aspect-ratio
Sputtered atoms
features. High-temperature processes are discussed below.
Long-throw deposition has an intrinsic directional flaw
as well as a difficulty in scaling to larger wafer sizes Collimator
[13, 14]. The directional flaw is based on the asymmetry of
deposition near the wafer edge. At the centerline of the
system, deposition onto the wafer arrives uniformly from
r" Wafer on chuck
n
the entire cathode in a symmetric manner. However, at
the wafer edge, there is a greater arrival flux from the
center regions of the cathode than from the outer edge.
This results in greater deposition on the inner-facing Oveiall configuiation of collimated sputter deposition.
sidewalls of a trench or via at the wafer edge, compared
to the outer-facing sidewall (Figure 12). The situation can
also be modeled numerically [15]. The result is that at the
wafer edge (i.e., the outermost chip), the asymmetry can packed round holes. Since the collimator is located quite
be as high as 2x at the top of modest-aspect-ratio close to the plasma discharge, it can be subjected to
features, and greater than 5x for the bottom corner. significant heating as a result of bombardment by
The scaling difficulty is based on the inability to operate electrons and sputtered atoms, as well as the heat of
a magnetron discharge at pressures below the O.l-mTorr condensation released by those sputtered atoms. These
range. For the 300-mm wafer generation, the cathode early designs allowed water cooling of the collimator from
diameter is expected to increase to 45 cm, requiring a the outer edge, which limited its edge temperature to
45-cm throw distance to achieve similar directionality. about 100°C and its mid-region temperature to about
While the assumption of no gas scattering was somewhat 150°C. Eventually, use was made of hexagonal holes to
questionable at a 25-cm throw distance, it is clearly very increase its effective transparency.
questionable at 45 cm. This results in a significant level of In the past several years, there has been a trend
in-flight gas scattering and loss of directionality of the toward the use of thin, lightweight, uncooled collimators
depositing atoms. constructed from sheet metal. These collimators are
assembled from either stainless steel or Ti sheet stock and
• Collimated sputter deposition
spot-welded into hexagonal arrays. The collimators drop
Directional filtering similar to long-throw deposition can
into uncooled housings with a minimum of fixtures and
be obtained by interposing an array of collimating tubes
are held in place by gravity. Unfortunately, during high-
between the magnetron cathode and the sample (Figure 13)
power operation, this approach can result in significant
[16]. The collimator functions much like the sidewalls
collimator heating, as high as 500°C at the center, which
of the chamber. It collects atoms which are traveling
in some cases can result in modifications to the films
laterally to the cathode/sample plane. These atoms are
on the substrate. Use is made of Ti collimators for Ti
deposited on the sidewalls of the collimator cells and
deposition to reduce film stress and peeling during
hence not deposited onto the sample. Geometrically,
thermal cycling.
the use of a collimator need not require any significant
increase in throw distance over conventional sputter The filtering of a collimator is directly related to its
deposition, thus removing the very low-pressure constraint aspect ratio; Figure 14 shows schematically the reduction
of long-throw sputtering. in angular divergence of the flux as a function of aspect
The aspect ratio of the collimator can be readily changed ratio. The deposition rate, though, is also significantly
without significantly altering its overall dimensions. reduced, and Figure IS shows the pressure dependence
For practical manufacturing applications, collimator of the deposition rate as a function of collimator aspect
cell diameters are of the order of 1 cm, and therefore ratio. The general result is that the deposition rate is
the height of a 2:1 collimator is only 2 cm. The overall reduced by a factor of about 35% for each unit increase
throw distance increase is then of the order of 3-4 cm; in the collimator aspect ratio.
manufacturing-scale tools outfitted with collimators The converse of this strong reduction in deposition rate
typically use a spacer which is about 4-5 cm high. is a significant deposition rate onto the collimator itself.
Collimators were originally fabricated from solid plates Eventually this leads to either closure of the collimator
of Al or Cu into which were milled arrays of closely holes, or peeling and flaking of thick deposits; neither 173

IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999 S. M. ROSSNAGEL

 Collimated sputtering was originally intended to be used
Surface for filling trenches and vias, but its slow deposition rates
and the increased effective cost due to collimator changes,
cleaning, etc. have limited its use to liner or contact
- Cosindike emission layers. The liner application was first described in 1992,
distribution when thin TiN layers sputtered through a collimator were
found to have adequate conformality to function as
diffusion barrier films [17]. Also widely used is the
Effective deposition contact-layer application, where a Ti layer —200 A thick is
distribution deposited on the bottom of a deep contact hole to make a
less resistive electrical contact to ao underlying layer.
For a 2-cm-high collimator located 2 cm from cathode:
Aspect ratio Emission width (degrees) Heating and pressurization techniques
11 28 (i.e., ±14) Since physical sputtering produces neutral atoms which
21 14
31 11 cannot be easily controlled aside from subtractive filtering,
4 1 7 an alternate approach to using sputter deposition with
high-aspcct-ratio features is to increase the mobility of the
sputtered atoms once they reach the sample, [f the sample
mmmmmmmm temperature is increased to a level of more than half the
melting-point temperature of the depositing film, there is
Schematic of transmitted angular flux through a collimator as a
function of collimator aspect ratio. significant surface diffusion of the depositing atoms as
well as rccrystallization of the deposited film [18]. Since
these are processes limited by an activation energy,
higher temperatures (more than half the melting-point
temperature) result in faster processes. Such use of
1000 higher substrate temperatures is known generically as
"rellow,'' although it is not completely clear whether the
No collimator
500 predominant iriotion on the sample surface is atomic
surface diffusion or the actual microscopic flow of surface
layers.
200
One of the intrinsic advantages of this approach is that
1.1:1 Collimator
small, high-aspect-ratio features can act as sinks for the
mobile material. The bottom of a via or trench has a
concave surface which is ideal as a trap for migrating
50
atoms. Therefore, the smallest high-aspect-ratio features
fill quite rapidly, which is the opposite of what occurs
20 in a directional process such as collimated sputtering.
Reflow techniques are quite strongly dependent on the
4.4:1 Collimator underlying surface structure and species. Simply sputtering
onto a hot surface is more likely to result in the formation
of spherical droplets on the wafer surface than filled vias.
10 1.5 20 25 30 35
Pres.sure (mTorr) Also, since surface diffusion generally has a much lower
activation energy than bulk diffusion, it is critical to keep
the top of a deep via from closing off during sputtering
(due to the wide-angle deposition). If this occurs, any
Deposition i.ile thiimgh a collimator a.s a function of pre-ssure and additional movement of atoms into the via occurs by
I collimatoi aspect latios. From [16], with permission. means of bulk diffusion, which is a much slower process
than surface diffusion.
Reflow deposition has been applied mostly to the
Al(Cu) interconnection system. For moderate levels of
Is desirable from a inanufacturing point of view. At a surface mobility, it is necessary to increase the wafer
practical level, it Is necessary to change the collimator temperature to at least 400°C and as high as 550°C in
about two to three times as often as the cathode, which some cases. Several schemes have been developed for
174 can add significantly to overall expense. reliable filling. One scheme, known generically as a two-

S. M. ROSSN.40HL IBM J. RES. DEVELOP. VOL. 4.^ NO. 1/2 JANUARY;M,4RCH WW

 step process, uses two sequential depositions of Al(Cu) must be well degassed prior to introduction into the
[3]. Tile first is sinort deposition witli a very low wafer chamber. The second concern, primarily for the thermal
temperature, below about 100°C. The function of the film reflow techniques, is the issue of the closing off of the vias
thus deposited is to provide a seed layer which adheres or trenches due to isotropic deposition at a faster rate
well to the substrate, is fine-grained, and is continuous than the reflow can rearrange the surface topography. This
down into deep features. For modest-aspect-ratio samples is primarily a time issue: Sputtering too slowly reduces the
(>2:1), this first step often uses collimated sputter wafer-per-hour throughput.
deposition. The second part of the deposition is carried A third concern is with the differences in physical
out at a wafer temperature of about 500°C, and uses dimensions or the density of features on a surface. A
conventional, noncollimated sputtering at a modest rate. small-diameter, high-aspect-ratio feature will fill much
If the rate is slow enough, the mobility of the surface more rapidly than a very large, low-aspect-ratio feature,
atoms is sufficient to move the atoms down into the deep simply because it takes many more atoms to fill the larger
features before the upper portions of these features can cavity. As a result, features in the middle of a dense field
close off. There is obviously a tradeoff here between of features will fill much more slowly than the ones on the
sample temperature and deposition rate: Higher sample edge of the field, where there is more relative surface
temperatures permit use of a faster deposition rate, but at area to accumulate the depositing and diffusing atoms.
the potential expense of other problems related to the use Finally, there is simply the concern over wafer
of the higher temperatures. temperature. Raising the wafer temperature to 400°C [for
Another process uses the deposition of an Al seed the CVD-secded AI(Cu)] or to 500°C (for the two-step
layer by means of a CVD process [19]. This results in a filling) can lead to significant thermal stress problems in
clean, conformal film which has slightly advantageous the interconnection structure. The thermal expansion
microstructural properties compared to the PVD film. coefficient of silicon dioxide is 20X smaller than that of
The advantage gained by using such a seed layer is seen Al, and this results in the potential for extrusion and other
primarily on the deposits on the sidewalls of trenches stress-relief problems in multiple-layer interconnection
and vias, which, when deposited by sputtering, are often stacks. The high temperatures also preclude the use of
columnar and of sub-bulk density. The CVD Al seed layer polymer-based law-/c dielectrics, which tend to have a
then allows somewhat faster surface diffusion of the much lower temperature limitation, typically 100°C or so;
subsequently deposited PVD AI(Cu), thus permitting a such a limitation would eliminate any possibility of the
reduction in the sample temperature to about 400°C. The use of mobility-enhancement-based filling techniques.
PVD AI(Cu) also contributes small levels of Cu to the
CVD Al film, which can only be deposited in a pure, Ionized sputter deposition of metal and
undoped manner. This incompatibility with doping compound films: l-PVD
precludes the use of CVD Al for primary conductor An alternative to the filtering of sputtered metal atoms to
applications. enhance the net directionality of a metallic deposit is to
Another means for improving the filling of high-aspeet- ionize the majority of the sputtered atoms and form the
ratio features is the use of relatively high pressures after film from metal ions. If the acceleration potential for the
deposition [20]. The technique requires that the sputter- ions is significantly greater than their thermal energy, the
deposited film close off a deep feature, such as a via, ions will arrive at the wafer surface at almost exactly 90°
forming a void. This occurs naturally, and more readily at with a controllable energy. If the percent of the metal
high aspect ratios. Next, the wafer is heated to 400°C and flux that is ionized is made high, the deposition will be
moved into a high-pressure chamber, where Ar gas is primarily directional and the utilization of the sputtered
introduced to a pressure of 600 atm. The force of this atoms from the cathode will be high. This last aspect
pressure on the warm film causes it to move downward relates to the intrinsic inefficiency of collimated or
into the vias, eliminating the voids formed by sputter long-throw sputtering, which results in low deposition
deposition. Once the Al metal reaches the bottom of rates and requires more frequent tool cleanings.
the via, the sample is cooled and the pressure carefully Deposition by means of I-PVD has been practiced in a
decreased. number of ways for many decades. Originally, evaporative
There are some constraints with a mobility-based deposition of metals could be enhanced or the resultant
deposition process. First, the system purity must be quite films modified by partially ionizing the evaporated flux,
high, since the arrival of impurities from the working by passing it through either an electron beam or a weak
gas, the cathode, or the sample will strongly limit the plasma. Other plasma-based deposition processes, such as
magnitude of the surface diffusion, in effect increasing the electron cyclotron resonance (ECR) or even magnetron
activation energy. To minimize this, the base pressure of sputter deposition, can have a small ionization fraction
the system must approach the UHV range, and the wafers for metal atoms sputtered into the discharge. 175

IBM J. RES. DEVELOP. VOL. 4.1 NO. 1/2 JANUARY/MARCH 1999 S. M. ROSSNAGEL
 plasma, this one driven by the rf inductive coil, is
dc magnetron (rotatin »ma met. 29 cm diameter) configured in the region between the cathode and the
p t i v ,• yt^m^ sample. As the sputtered atoms enter this plasma, some
fraction of them are ionized by electron bombardment as
Inductively
they pass through. The ionization fraction can be high
coupled rf
—M; :;©
antenna " ©: .lET because the ionization potential for the metal species is
(1-2 turns, typically 5-7 eV, whereas the ionization potential for the
water-cooled) 1 inert gas species is 15.7 eV (Ar). The relative flux of metal
/ atoms is small compared to the density of the inert gas
atoms (typically 1-5%).
Once the metal ions are formed, they drift within the
/
Wafer chuck
(clamped for dc bias)
n Bias power
plasma, and if they reach the sample sheath, they are
accelerated by the difference between the plasma potential
supply (typically +10 V) and the wafer potential (0 to ^50 V).
Conversely, if the metal ions drift back toward the cathode
sheath, they are accelerated by the magnetron voltage
(typically ^400 V), and are used to sputter more atoms
I-PVD system ba!>ed on inductively coupled rf plasma. From [26], from the cathode. Ideally, the efficiency of utilization
with permission of the sputtered metal atoms can be quite high.
The relative ionization of the sputtered atoms at the
sample location is a function of several parameters. First,
it is sensitive to the density of the rf plasma, which is
In the late 1980s and early 1990s, systems were specifically dependent on the rf power as well as the working pressure
fabricated to optimize the relative ionization of [27], In addition, the metal-ionization level is dependent
metal particles prior to deposition. Early systems used on the amount of time the metal atoms spend in the rf
ECR [21, 22] and were based on either sputtering or plasma. At low pressures, the sputtered atoms pass
evaporation. Another primary direction has been the use through the plasma region rapidly, and the ionization
of dense, inductively coupled rf plasmas in conjunction level is low. As the pressure is increased to several tens
with a metal-sputtering source [23-25], This latter of mTorr, the sputtered atoms can be slowed by gas
technique has proven to be the most robust for collisions, and as a result they spend more time in the
semiconductor manufacturing applications. discharge and are more likely to be ionized. The optimum
pressure depends on system dimensions, but typically is in
An I-PVD system, based on magnetron sputtering and
the 15-30-mTorr range, which is much higher than would
in-flight ionization of the sputtered atoms by means
be used for conventional sputtering.
of a dense, inductively coupled rf plasma, is shown in
The relative ionization at the sample has been measured
Figure 16. Use is made of a conventional 200-mm magnetron
to be as high as 90% at high working pressures, high
source (diameter = 300 mm) having a rotating magnet
levels of rf power to the coil, and relatively low metal-
array, A dense, inert gas plasma is configured in the
sputtering rates. As the metal-sputtering rate is increased,
region between the cathode and the sample by means of a
the relative ionization has been observed to fall. Originally
1-3-turn coil, about 20% larger than the wafer and located
this was thought to be primarily a cooling of the electron
a few centimeters from both the cathode and the sample.
temperature of the rf plasma used for ionization due to
The coil is powered at a frequency of 1,9 or 13.5 MHz and
the presence of large numbers of easily ionized metal
is matched such that each end of the coil is 180° out of
atoms [27], Recent work has also suggested a secondary
phase from the other. The coil may or may not be water- effect caused by the metal flux: the rarefaction of the
cooled; One commercially available tool uses an uncooled working gas within the plasma [28], This was first observed
coil, but most other applications use cooling. The coils are for conventional magnetron sputtering in the late 1980s
powered at 1-3 kW, [29], In I-PVD, the rarefaction results in fewer in-flight
The operating scheme of this type of tool is as follows. collisions for the sputtered atoms, and hence less time in
There are two somewhat separate plasmas set up within the ionization region. It is as though at high magnetron
the chamber, both using the same inert background gas. power the pressure is reduced, and it is this effect which is
The magnetron plasma is a conventional dc plasma, most significant in reducing the relative ionization level.
located close to the magnetron cathode, such that a Deposition by means of I-PVD has been used in
significant flux of ions can strike the cathode, causing semiconductor applications in three primary areas. The
176 sputter emission of the metal cathode atoms, A second most straightforward application has been the deposition

S. M. ROSSNAGEL IBM J. RES. DEVELOP. VOL. 4.1 NO. Ill JANUARY/MARCH 1999
 of Ti at the bottom of contact holes. Originally, this was
done with collimated sputtering, but I-PVD is more 100

efficient at step coverage, even at high aspect ratios

(Figure 17). Commercial suppliers have begun building
tools for this application, and it is rapidly supplanting
80 - •
collimated sputtering. • •
Another semiconductor application of I-PVD has been 60
• • ^ I-PVD Ti
to form diffusion barriers in vias and trenches, again as a ••
• •
replacement for collimated sputtering. At first glance, it
•
would not be expected that a directional process such as 40 - •
I-PVD would be applicable to the conformal films needed •
for liners and seed layers. However, two aspects of the '• •
• • Collimated PVD Ti
deposition make it applicable for modest-aspect-ratio 20 -
- • PVDTi •
features (<5:1 AR). The first is simply that the relative • •
ionization is less than 100%, which means that there is a
small nondirectional component to the deposition. This 4 6 10
component tends to coat the upper sidewalls of the vias. Aspect ratio of via
The second aspect is the ability for the depositing atoms
to resputter the already deposited film. If the sample
potential is made sufficiently negative, the kinetic energy of
the depositing ions will be sufficient to sputter the film Step coverage as a function of aspect ratio for the bottom of a
contact hole, using conventional PVD, collimated PVD, and ionized
and redistribute it within the features. This leads to a PVD. From [3J, with permission.
thickening of the film in the bottom corners of vias, which
is desirable because that location is the most prone to
diffusion barrier failure. A schematic of this is shown in
Figure 18.
A third semiconductor application of I-PVD has been Non-ionized deposit Beveled back
for the filling of features. This application is strongly comer
dependent on the relative ionization of the depositing flux:
Higher levels of ionization correlate with the filling of
higher-aspect-ratio features. I-PVD filling can easily be
applied to either single- or dual-damascene structures
(Figure 19). From a practical point of view, ionization
levels of about 70% can be measured on large tools. This
leads to filling at an aspect ratio of about 2:1 for 0.35-/xm- Resputtered
wide features. Higher aspect ratios or smaller feature sizes onto walls
will require higher levels of ionization. Dielectric
It turns out that at a finer feature size, resputtering of the
deposited film is not desirable. This is seen in Figure 20,
which is a simulation of the effect of resputtering and
redeposition of the sputtered atoms. At a high aspect
ratio, or for features less than about 0.5 /.im wide, cross-
deposition from one side of the trench to the other
results in void formation. The simulation applies at low X:i7:^s^^'': '-' J" ^,

Schematic of a dittusion barrier/liner deposited with I-PVD, show-

temperatures, and may not be applicable if there is
ing the thickening at the bottom comers due to resputtering
adequate sample temperature to allow some degree of
reflow during deposition.

Conclusions
Sputter deposition is likely to continue to be used for sputtering or long-throw deposition, are in widespread
semiconductor applications because of the large, already use, but are likely to be made obsolete in the next
installed base of PVD systems, the simplicity of the several years by the introduction of ionized
process, and the wealth of associated understanding. deposition. It is likely that the near-term applications of
Variations to sputter deposition, such as collimated I-PVD will be primarily for contact layers and diffusion- 177

IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999 S. M. ROSSNAGEL

 Scanning electron microgiaphs illustrating (from left to right) the filling nt dual-damascene features using I-PVD From [30], with permission.

S. M. Rossnagel, "Deposition and Redeposition in

Magnetrons,"/ Vac. Sci. Technol. A 6, 3049-3054 ( 1 '
P. C. Zalm, "Quantitative Sputtering," Surf. & Interface
Anal. 11, 1 (1988).
W. O. Hofer, "Angular, Energy, and Mass Distribution of
Sputtered Particles," Sputtering by Particle Bombardment,
Vol. Ill, R. Behrisch and K. Wittmaack, Eds., Sprlnger-
Vcrlag, Berlin, 1991.
S. M. Rossnagel, "High-Vacuum-Based Processes:
Sputtering," Handbook of Vacuum Science and Technology,
Ch. 5.2, D. Hoffman, B. Singh, and J. Thomas, Eds.,
Academic Press, Inc., Boston, 1997.
W. F. Callaway, C. E. Young, M. Pcllin, and D. M.
(from left to right) on tiench tilling From [31], with permission Gruen, "Laser-Induced Fluorescence as a Tool for the
Study of Ion Beam Sputtering," Handbook of Ion Beam
Sputtering Technology, Ch. 7, J. J. Cuomo, S. M.
Rossnagel, and H. R. Kaufman, Eds., Noyes Publications,
Park Ridge, NJ, 1989.
J. S. Fan, R. S. Bailey, and C. E. Wickersham, Jr., "New
barrier or seed layers. Filing applications are more likely Developments and Applications for Sputtering Targets at
Tosoh SMD," presented at Semicon-China, November
to be addressed by use of elevated-temperature reflow 1997 (published in Reference 3, Ch. 11).
techniques. The use of I-PVD deposition for filling deep lo. T. J. Licata, E. G. Colgan, J. M. E. Harper, and S. E.
features will require some degree of surface mobility to Luce, "Interconnect Fabrication Processes and the
Development of Low-Cost Wiring for CMOS Products,"
help fill in the less dense structures on the sidcwalls, as
IBM J. Res. Develop. 39, 419-435 (1995).
well as reducing the overburden and the required CMP. u. C. W. Kaanta, S. Bombardier, W. Cote, W. Hill, G.
Korszykowski, H. Landis, D. Poindexter, C. Pollard, G.
Ross, J. Ryan, J. Wolff, and J. Gonin, "Dual Damascene:
References
A ULSI Wiring Technology," Proceedings of the 8th
1. B. Chapman, Glow Discharge Processes, John Wiley & International VLSI Multilevel Interconnection Conference,
Sons, Inc., New York, 1980, p. 150. Santa Clara, CA, 1991, pp. 144-152.
2. S. M. Rossnagel and H. R. Kaufman, "Induced Drift 12, J. J. Cuomo and S. M. Rossnagel, "Hollow Cathode
Currents in Circular Planar Magnetrons," /. Vac. Sci. Enhanced Magnetron Sputtering," J. Vac. Sci. Technol A
Technol. A 5, 88^91 (1987). 4, 393-396 (1986).
3. R. A. Powell and S. M. Rossnagel, PVD for 13. I. Wagner, "Sputter Deposition of Ti and TiN Films with
178 Microelectronics, Academic Press, Inc., Boston, 1998. Variable Target-to-Substrate Distance," Proceedings of

S. M. ROSSNAGEL IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH 1999

 the 12th International VLSI Multilevel Interconnection S t e p h e n M. R o s s n a g e l IBM Research Division, Thomas
Conference, Santa Clara, CA, 1995, p. 226. J. Watson Research Center, P.O. Box 218, Yorktown Heights,
14. J. N. Broughton, C. J. Backhouse, M. J. Brett, S. K. Dew, New York 10598 (rossnag@us.ibm.com). Dr. Rossnagel is a
and G. Este, "Long Throw Sputter Deposition of Ti at Research Staff Member at the Thomas J. Watson Research
Low Pressure," Proceedings of the 12th International VLSI Center, working in an interconnection metallurgy and
Multilevel Interconnection Conference, Santa Clara, CA, processing group. He received B.S. and M.S. degrees from
1995, p. 201. Pcnn State University in 1975 and 1977, respectively, and a
15. A. A. Mayo, S. Hamaguchi, J. H. Joo, and S. M. Ph.D. from Colorado State University in 1982 following
Rossnagel, "Across-Wafer Nonuniformity of Long Throw positions on the staff of the Princeton University Plasma
Sputter Deposition,"/ Vac. Sci. Technol. B 15, 1788 (1997). Physics Laboratory and the Max Planck Institute for Plasma
16. S. M. Rossnagel, D. Mikalsen, H. Kinoshita, and J. J. Physics in Munich, Germany. He joined IBM in 1983 at the
Cuomo, "Collimated Magnetron Sputter Deposition," Thomas J. Watson Research Center, where he has worked
/ Vac. Sci. Technol. A 9, 261 (1991). primarily on sputtering and plasma processing. Dr. Rossnagel
17. R. V. Joshi and S. Brodsky, "Collimated Sputtering for has authored more than 100 papers, five books, and nine
Ti/TiN Liners into High Aspect Ratio Contacts/Lines," patents. He received the Peter Mark Award from the
Proceedings of the 9th International VLSI Multilevel American Vacuum Society (AVS) in 1990 and was made a
Interconnection Conference, Santa Clara, CA, 1992, pp. Fellow of the Society in 1994. He chaired the AVS Annual
253-259. Conference in 1996, and was recently elected President-Elect
18. B. Zhao, M. A. Bilberger, V. Hoffman, S.-Q. Wang, P. K. of the AVS for 1998, to serve as President in 1999.
Vasudev, and T. E. Seidel, "A Novel Low Temperature
PVD Planarized AlCu Process for High Aspect Ratio,
Sub-Micron Interconnect," Proceedings of the International
Electron Devices Meeting, IEEE, New York, 1996, pp.
353-356.
19. G. Dixit, W. Y. Hsu, K. H. Hamamoto, M. K. Jain,
L. M. Ting, R. H. Havemann, C. D. Dobson, A. I.
Jeffreys, P. L. Holverson, P. Rich, D. C. Butler, and
J. Hems, "Application of High Pressure Extruded
Aluminum to ULSI Metallization," Semicond. Int.
18, 79-85 (August 1995).
20. P. Kidd, "A Magnetically Confined and Electron
Cyclotron Resonance Heated Plasma for Coating and Ion
Surface Modification Use," J. Vac. Sci. Technol. A 9, 466
(1991).
21. G. Wyborn and T. Guo, "Cool Al, the 0.18 Micron
Technology," Appl. Mater. HP PVD Update 3, 2 (1996).
22. W. M. Holber, J. S. Logan, H. Grabarz, J. T. C. Yeh,
J. B. O. Caughman, A. Sugarman, and F. Turene,
"Copper Deposition by Electron Cyclotron Resonance
Plasma," J. Vac. Sci. Technol. A 11, 1903 (1993).
23. M. Yamashita, "Fundamental Characteristics of a Built-in
High Frequency Coil-Type Sputtering Apparatus," /. Vac.
Sci. Technol. A 7, 151-158 (1989).
24. M. S. Barnes, J. C. Forster, and J. H. Keller, "Apparatus
for Depositing Material into High Aspect Ratio Holes,"
U.S. Patent 5,178,739, January 12, 1993.
25. S. M. Rossnagel and J. Hopwood, "Magnetron Sputter
Deposition with Levels of Metal Ionization," Appl. Phys.
Lett. 63, 3285-3287 (1993).
26. S. M. Rossnagel, "Ionized Magnetron Sputtering for
Lining and Filling Trenches and Vias," Semicond. Int.
19, 99 (February 1996).
27. S. M. Rossnagel and J. Hopwood, "Metal Ion Deposition
from Ionized Magnetron Sputtering Discharge," /. Vac.
Sci. Technol. B 12, 449 (1994).
28. M. Dickson, F. Qian, and J. Hopwood, "Quenching of
Electron Temperature and Electron Density Using
Ionized Physical Vapor Deposition," / Vac. Sci. Technol.
A 15, 340-344 (1997).
29. S. M. Rossnagel, "Gas Density Reduction Effects in
Magnetrons," J. Vac. Sci. Technol. A 6, 19-24 (1988).
30. S. M. Rossnagel, "Filling Dual Damascene Interconnect
Structures with AlCu and Cu Using Ionized Magnetron
Sputter Deposition,"/. Vac. Sci. Technol. B 13, 125
(1995).
31. S. Hamaguchi and S. M. Rossnagel, "Liner Conformality
in Ionized Magnetron Sputter Metal Deposition Process,"
/ Vac. Sci. Techitol. B 14, 2603 (1996).

Received November 13, 1997; accepted for publication

March 20, 1998 179

IBM J. RES. DEVELOP. VOL. 43 NO. 1/2 JANUARY/MARCH l')')9 S. M. ROSSNAGEL