U.S. patent application number 12/216224 was filed with the patent office on 2009-01-22 for substrate processing method and substrate processing apparatus.
Invention is credited to Hiroyuki Kanda, Keiichi Kurashina, Tsutomu Nakada, Akira Susaki, Satoru Yamamoto.
Application Number | 20090020434 12/216224 |
Document ID | / |
Family ID | 40263970 |
Filed Date | 2009-01-22 |
United States Patent
Application |
20090020434 |
Kind Code |
A1 |
Susaki; Akira ; et
al. |
January 22, 2009 |
Substrate processing method and substrate processing apparatus
Abstract
A substrate processing method makes it possible to fill
interconnect recesses, such as trenches, with a defect-free
interconnect material by carrying out electroplating directly on a
surface of a ruthenium film as a barrier layer. The substrate
processing method comprises: providing a substrate having
interconnect recesses formed in a substrate surface and having a
ruthenium film formed in the entire substrate surface including
interior surfaces of the interconnect recesses; keeping the
substrate surface in contact with a plating solution for a
predetermined time to adsorb an additive in the plating solution
onto the ruthenium film, and then carrying out electroplating to
form a conductive film on a surface of the ruthenium film.
Inventors: |
Susaki; Akira; (Tokyo,
JP) ; Nakada; Tsutomu; (Tokyo, JP) ; Yamamoto;
Satoru; (Tokyo, JP) ; Kurashina; Keiichi;
(Tokyo, JP) ; Kanda; Hiroyuki; (Tokyo,
JP) |
Correspondence
Address: |
WENDEROTH, LIND & PONACK, L.L.P.
2033 K STREET N. W., SUITE 800
WASHINGTON
DC
20006-1021
US
|
Family ID: |
40263970 |
Appl. No.: |
12/216224 |
Filed: |
July 1, 2008 |
Current U.S.
Class: |
205/239 ;
204/194; 205/291 |
Current CPC
Class: |
H01L 21/67051 20130101;
C25D 17/001 20130101; C25D 21/12 20130101; H01L 21/76877 20130101;
H01L 21/2885 20130101; H01L 21/68728 20130101; C25D 3/38 20130101;
C25D 21/00 20130101; H01L 21/6715 20130101; C25D 5/10 20130101;
H01L 21/67005 20130101; H01L 21/76843 20130101; C25D 5/34 20130101;
H01L 21/76873 20130101 |
Class at
Publication: |
205/239 ;
205/291; 204/194 |
International
Class: |
C25D 3/58 20060101
C25D003/58; C25D 3/38 20060101 C25D003/38; C25D 17/00 20060101
C25D017/00 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 2, 2007 |
JP |
2007-173940 |
Jun 30, 2008 |
JP |
2008-170222 |
Claims
1. A substrate processing method comprising: providing a substrate
having interconnect recesses formed in a substrate surface and
having a ruthenium film formed in the entire substrate surface
including interior surfaces of the interconnect recesses; keeping
the substrate surface in contact with a plating solution for a
predetermined time to adsorb an additive in the plating solution
onto the ruthenium film; and then carrying out electroplating to
form a conductive film on a surface of the ruthenium film.
2. The substrate processing method according to claim 1, wherein
the conductive film is composed of copper or a copper alloy.
3. The substrate processing method according to claim 2, wherein
the plating solution contains a copper ion, a sulfate ion and the
additive.
4. The substrate processing method according to claim 1, wherein
the predetermined time for keeping the substrate surface in contact
with the plating solution prior to the electroplating is not less
than 0.5 second and not more than 60 seconds.
5. The substrate processing method according to claim 1, wherein
the predetermined time for keeping the substrate surface in contact
with the plating solution prior to the electroplating is not less
than 0.1 second and not more than 20 seconds.
6. The substrate processing method according to claim 1, wherein
the predetermined time for keeping the substrate surface in contact
with the plating solution prior to the electroplating is not less
than 0.1 second and not more than 5 seconds.
7. A substrate processing method comprising: providing a substrate
having interconnect recesses formed in a substrate surface and
having a ruthenium film formed in the entire substrate surface
including interior surfaces of the interconnect recesses; keeping
the substrate surface in contact with a plating solution for a
predetermined time to adsorb an additive in the plating solution
onto the ruthenium film, and then carrying out first electroplating
to form an initial conductive film, which covers the entire
interior surfaces of the interconnect recesses, on the surface of
the ruthenium film; cleaning and drying the substrate surface; and
then carrying out second electroplating to allow a conductive film
to further grow on a surface of the initial conductive film.
8. The substrate processing method according to claim 7, wherein
the first electroplating and the second electroplating are carried
out by using the same plating solution.
9. The substrate processing method according to claim 7, wherein
the predetermined time for keeping the substrate surface in contact
with the plating solution prior to the first electroplating is not
less than 5 seconds.
10. A substrate processing apparatus for forming a conductive film
on a surface of a substrate by electroplating, the substrate having
interconnect recesses formed in the substrate surface and having a
ruthenium film formed in the entire substrate surface including
interior surfaces of the interconnect recesses, said apparatus
comprising: a measurement section for measuring time that has
elapsed since the substrate surface has been brought into contact
with a plating solution.
11. The substrate processing apparatus according to claim 10,
wherein the measurement section is comprised of a position detector
for detecting the position of the substrate or a substrate holder,
or a substrate-solution contact detector for detecting contact of
the substrate with the plating solution.
12. The substrate processing apparatus according to claim 11,
wherein the substrate-solution contact detector is comprised of an
optical sensor, a pressure sensor, a conductivity sensor, a
temperature sensor or an ultrasonic sensor, or a combination
thereof.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to a substrate recessing
method and a substrate processing apparatus, and more particularly
to a substrate processing method and a substrate processing
apparatus which are useful for filling a metal (interconnect
material), such as copper or silver, into fine interconnect
recesses provided in a surface (surface to be plated) of a
substrate, such as a semiconductor wafer, to form interconnects in
the substrate.
[0003] 2. Description of the Related Art
[0004] With the progress toward smaller, higher speed and lower
power consuming electronic devices, interconnect patterns in
semiconductor devices are becoming finer, and interconnect
materials are changing from conventional aluminum or its alloys to
copper or copper alloys. The electrical resistivity of copper is
1.67 .mu..OMEGA.cm which is about 37% lower than the electrical
resistivity of aluminum (2.65 .mu..OMEGA.m). Therefore, as compared
to aluminum interconnects, copper interconnects can reduce power
consumption, or can be made finer with the same interconnect
resistance as aluminum interconnects. In addition, the lowering of
interconnect resistance by the use of copper interconnects can
reduce signal delay.
[0005] On the other hand, copper atoms easily move in silicon or in
an insulating film, which can impair the characteristics of a
semiconductor device. Copper interconnects, therefore, need to be
covered with a protective layer called a barrier layer. To produce
a structure of copper interconnects covered with a barrier layer, a
damascene process is widely used which comprises filling copper
(interconnect material) into interconnect recesses, such as
trenches, provided in a surface of a substrate. A barrier layer of
Ti, TiN, Ta, TaN, or the like formed by PVD, CVD or ALD has
heretofore been widely used. The filling of interconnect recesses
with copper is generally carried out by electroplating which is
capable of high-speed film formation.
[0006] FIGS. 1A through 1C illustrate, in a sequence of process
steps, an example of forming a substrate having copper
interconnects by a conventional process. First, as shown in FIG.
1A, an insulating film (interlevel dielectric film) 2, such as an
oxide film of SiO.sub.2 or a film of low-k material, is deposited
on a conductive layer lain which semiconductor devices are formed,
which is formed on a semiconductor base 1. Via holes 3 and trenches
4 as interconnect recesses are formed in the insulating film 2 by
the lithography/etching technique. Thereafter, a barrier layer 5 of
TaN or the like is formed on the surface, and a seed layer 7 as an
electric supply layer for electroplating is formed on the barrier
layer 5.
[0007] Then, as shown in FIG. 1B, copper plating is performed onto
the surface of the substrate W to fill the via holes 3 and the
trenches 4 with copper and, at the same time, deposit a plated
copper film 6 on the insulating film 2. Thereafter, the plated
copper film 6, the seed layer 7 and the barrier layer 5 on the
insulating film 2 are removed by chemical mechanical polishing
(CMP) so as to make the surface of the plated copper film 6 filled
in the via holes 3 and the trenches 4, and the surface of the
insulating film 2 lie substantially on the same plane.
Interconnects composed of the plated copper film 6, as shown in
FIG. 1C, are thus formed in the insulating film 2.
[0008] The copper electroplating is generally carried out by using
an acidic plating solution containing copper sulfate, and supplying
electricity to the seed layer 7 of, e.g., copper from the periphery
of the substrate to allow the plated copper film 6 to grow on the
surface of the seed layer 7. A thickness of the seed layer 7, e.g.,
formed by PVD, is several tens of nm in the substrate surface,
whereas a thickness is not more than several nm in the sidewalls of
the trenches 4. Accordingly, if the substrate is kept in contact
with an acidic copper-plating solution for a long time, the seed
layer 7 can easily dissolve in the plating solution. With the
progress toward finer interconnects of not more than 0.2 .mu.m, for
example, in practical electroplating, a method is generally used
which comprises applying a voltage between the seed layer 7, which
serves as a cathode, and a counter electrode (anode) immediately
before bringing the substrate into contact with a plating solution
to initiate electroplating simultaneously with contact of the
substrate with the plating solution, taking the substrate out of
the plating solution after completion of the plating, and cleaning
and drying the substrate, as shown in FIG. 2.
[0009] As described above, since a seed layer is generally formed
by PVD, a thickness of a seed layer in interconnects of a substrate
is as small as about 10 to 20% of the thickness of the seed layer
in the substrate surface (field area). As interconnects become
finer, a seed layer becomes thinner in a substrate surface in order
to ensure openings in an interconnect area, and the seed layer
becomes further thinner in interconnects. To address the movement
toward thinner seed layer, it has been proposed to devise a voltage
applied between a seed layer and a counter electrode (anode)
(Japanese Patent Laid-Open Publication No. 2003-129297) It has also
been proposed to insert a high-resistance member, having a higher
resistance than a plating solution, between a seed layer and a
counter electrode (anode) (Japanese Patent Laid-Open Publication
No. 2001-323398). With further progress toward finer interconnects,
a thickness of a seed layer in interconnects will become as small
as several nm, and it is considered impossible to form such a thin
seed layer in a continuous form.
[0010] It is known that, when a voltage is applied between the seed
layer 7, which serves as a cathode, and a counter electrode
(cathode) after bringing the substrate into contact with a plating
solution, characteristics in filling of copper into the trenches 4
change. In particular, if the voltage is applied between the seed
layer 7 and the counter electrode (anode) for a long time after the
contact of the substrate with the plating solution, preferential
growth of a plated copper film in the trenches from their bottoms
is suppressed and the plated copper film grows isotropically from
the sidewalls and the bottoms of the trenches. In this case,
growing portions of the plated copper film collide with each other
in the respective trenches, forming voids called "seams" in the
plated copper film embedded in the trenches. The seams may cause
lowering of the reliability of the copper interconnects. The change
in the copper-filling characteristics with the time period from
contact of the substrate with the plating solution to the voltage
application is considered to be related to time taken for an
additive contained in the plating solution to be adsorbed onto the
substrate surface.
[0011] A plating solution for use in plating and filling of
trenches and via holes of a semiconductor device generally
contains, in addition to a metal ion component and a pH adjuster
component, various additives for improving the trench-filling
properties, such as an accelerator, a suppressor and a leveler.
Such additives show their effects when they are adsorbed on a
surface of a substrate. In a submicron trench, an amount of plating
solution per surface area is small as compared to the flat area of
the substrate surface. Accordingly, it takes a longer time for a
certain amount of additive to be adsorbed on the substrate surface
in a trench as compared to the flat area of the substrate
surface.
[0012] A suppressor primarily has the effect of suppressing
deposition of a plated film in a flat area of a surface of a
substrate. When a substrate is left in contact with a plating
solution containing a suppressor, adsorption of the suppressor onto
the interior surfaces of trenches also progresses, thereby
suppressing deposition of a plated film in the trenches.
Accordingly, a plated film grows isotropically also in the
trenches, which can result in incomplete filling of the trenches
with the plated film, with seams or voids being formed in the
plated film.
SUMMARY OF THE INVENTION
[0013] A Ti alloy or a Ta alloy, which is commonly used as a
material for a barrier layer, has a high electrical resistivity,
for example, 12.45 .mu..OMEGA.cm in the case of Ta and 42
.mu..OMEGA.cm in the case of Ti, which are one digit higher than
the electrical resistivity of copper (interconnect material). It is
therefore necessary to form a seed layer as an electric supply
layer on a surface of a barrier layer when filling copper into
trenches by electroplating. A copper seed layer formed by PVD is
generally used as the seed layer. As interconnects become finer, it
is becoming increasingly difficult to form a copper seed layer on
surfaces of trenches. With further progress toward finer
interconnects in the future, the proportion of a thickness of a
seed layer in the width of interconnect becomes larger. It is
therefore feared that a plating solution may hardly enter trenches
during electroplating, resulting in poor filling of the trenches.
Furthermore, the formation of a continuous seed layer will become
more and more difficult.
[0014] A method (direct plating) has therefore begun to be studied
which involves using a barrier layer having a relatively low
electrical resistance and carrying out copper electroplating
directlyon a surface of the barrier layer. Tantalum (Ta), which has
been widely used for a barrier layer, generally has poor adhesion
to an interconnect material, such as copper, for forming
interconnects. This contributes to lowering of the reliability of
interconnects when a Ta barrier layer is used.
[0015] In order to solve these problems, it has been proposed to
use Ru (ruthenium), WNC (tungsten carbonitride) or the like for a
barrier layer. The electrical resistivity of Ru is 7.6
.mu..OMEGA.cm which is considerably lower than the electrical
resistivity of Ta, and therefore the possibility of direct
electroplating on a surface of a ruthenium film is under
consideration. If direct electroplating on a surface of a barrier
layer of ruthenium (ruthenium film) becomes possible, a seed layer
as an electric supply layer becomes unnecessary. This can not only
eliminate a seed layer formation process but can also avoid poor
conduction due to a non-uniform seed layer and can avoid the
formation of voids in an interconnect metal embedded in trenches
due to insufficient intrusion of an electroplating solution into
the trenches.
[0016] In a conventional common electroplating process for filling
copper into trenches using a seed layer, from the viewpoint of
prevention of dissolution of the seed layer and prevention of
defects in copper embedded in the trenches, it is necessary to
apply a voltage between the seed layer, which serves as a cathode,
and a counter electrode (anode) to initiate electroplating almost
simultaneously with bringing the substrate into contact with a
plating solution. It has been found by the present inventors,
however, that when ruthenium is used for a barrier layer and copper
electroplating is carried out in the conventional manner, but
directly on a surface of a ruthenium film as a barrier layer, seams
or voids will be formed in a plated copper film embedded in
trenches.
[0017] The present invention has been made in view of the above
situation. It is therefore an object of the present invention to
provide a substrate processing method and a substrate processing
apparatus which make it possible to fill interconnect recesses,
such as trenches, with a defect-free interconnect material by
carrying out electroplating directly on a surface of a ruthenium
film as a barrier layer.
[0018] In order to achieve the above object, the present invention
provides a substrate processing method comprising: providing a
substrate having interconnect recesses formed in a substrate
surface and having a ruthenium film formed in the entire substrate
surface including interior surfaces of the interconnect recesses;
keeping the substrate surface in contact with a plating solution
for a predetermined time to adsorb an additive in the plating
solution onto the ruthenium film; and then carrying out
electroplating to form a conductive film on a surface of the
ruthenium film.
[0019] The present inventors conducted an experiment in which
copper electroplating was carried out on a ruthenium (Ru) film as a
barrier layer to allow a plated copper film to grow on the
ruthenium film, and the process of the growth of the plated copper
film was analyzed. As a result, it was found that deposition of
particulate copper on the surface of the ruthenium film in the
initial stage of plating causes the formation of voids in the
plated copper film. It was also found that the deposition of
particulate copper depends on the time period from contact of the
substrate with the plating solution to the initiation of
electroplating. Thus, unlike the conventional electroplating for
filling of trenches using a seed layer, the formation of voids in a
plated film can be suppressed and trench-filling characteristics
comparable to the conventional electroplating method using a seed
layer can be ensured by initiating electroplating after bringing a
surface of a substrate into contact with a plating solution and
leaving the substrate in contact with the plating solution for a
certain period of time to adsorb an additive in the plating
solution onto a ruthenium film.
[0020] The conductive film is preferably composed of copper or a
copper alloy.
[0021] This can produce interconnects of copper or a copper alloy
which has been filled into interconnect recesses by
electroplating.
[0022] In a preferred aspect of the present invention, the plating
solution contains a copper ion, a sulfate ion and the additive.
[0023] A ruthenium film as a barrier layer will not dissolve in the
plating solution even when such an acidic plating solution is used
during electroplating.
[0024] The predetermined time for keeping the substrate surface in
contact with the plating solution prior to the electroplating is,
for example, not less than 0.5 second and not more than 60
seconds.
[0025] The predetermined time for keeping the substrate surface in
contact with the plating solution prior to the electroplating is
preferably not less than 0.1 second and not more than 20 seconds,
more preferably not less than 0.1 second and not more than 5
seconds.
[0026] The present invention provides another substrate processing
method comprising: providing a substrate having interconnect
recesses formed in a substrate surface and having a ruthenium film
formed in the entire substrate surface including interior surfaces
of the interconnect recesses; keeping the substrate surface in
contact with a plating solution for a predetermined time to adsorb
an additive in the plating solution onto the ruthenium film, and
then carrying out first electroplating to form an initial
conductive film, which covers the entire interior surfaces of the
interconnect recesses, on the surface of the ruthenium film;
cleaning and drying the substrate surface; and then carrying out
second electroplating to allow a conductive film to further grow on
a surface of the initial conductive film.
[0027] By thus carrying out the first electroplating to conformally
form the initial conductive film, which covers the entire interior
surfaces of interconnect recesses, on the surface of the ruthenium
film, and then carrying out the second electroless plating using
the initial conductive film as a seed layer, it becomes possible to
prevent the formation of voids in a plated film and ensure
trench-filling characteristics comparable to those obtained by the
use of a conventional seed layer. The second electroplating may be
started either after keeping the substrate surface in contact with
a plating solution for a short time to adsorb a small amount of an
additive in the plating solution onto the initial conductive film,
or simultaneously with bringing the substrate surface into contact
with the plating solution.
[0028] Preferably, the first electroplating and the second
electroplating are carried out by using the same plating
solution.
[0029] The predetermined time for keeping the substrate surface in
contact with the plating solution prior to the first electroplating
is preferably not less than 5 seconds.
[0030] The present invention also provides a substrate processing
apparatus for forming a conductive film on a surface of a substrate
by electroplating, the substrate having interconnect recesses
formed in the substrate surface and having a ruthenium film formed
in the entire substrate surface including interior surfaces of the
interconnect recesses, said apparatus comprising a measurement
section for measuring time that has elapsed since the substrate
surface has been brought into contact with a plating solution.
[0031] In a preferred aspect of the present invention, the
measurement section is comprised of a position detector for
detecting the position of the substrate or a substrate holder, or a
substrate-solution contact detector for detecting contact of the
substrate with the plating solution.
[0032] The substrate-solution contact detector is comprised of, for
example, an optical sensor, a pressure sensor, a conductivity
sensor, a temperature sensor or an ultrasonic sensor, or a
combination thereof.
[0033] By carrying out electroplating directly on a surface of a
ruthenium film, having a low electrical resistance, as a barrier
layer according to the method of the present invention, an
interconnect material, e.g., copper, can be securely filled into
interconnect recesses such as trenches without forming defects,
such as voids, in the embedded interconnect metal. Thus, the
present invention makes it possible to enhance the reliability of
interconnects while enjoying the benefit of elimination of a seed
layer formation process.
BRIEF DESCRIPTION OF THE DRAWINGS
[0034] FIGS. 1A through 1C are diagrams illustrating, in a sequence
of process steps, an example of the formation of copper
interconnects by a conventional plating process;
[0035] FIG. 2 is a flow chart showing a conventional plating
process;
[0036] FIG. 3 is an overall plan view of a substrate processing
apparatus according to an embodiment of the present invention;
[0037] FIG. 4 is a plan view of an electroplating apparatus shown
in FIG. 3;
[0038] FIG. 5 is an enlarged sectional view of a substrate holder
and a cathode portion of the electroplating apparatus shown in FIG.
3 (a cross-sectional view taken along line A-A of FIG. 4);
[0039] FIG. 6 is a front view of a plating solution recovering arm
of the electroplating apparatus shown in FIG. 3;
[0040] FIG. 7 is a plan view of the substrate holder of the
electroplating apparatus shown in FIG. 3;
[0041] FIG. 8 is a cross-sectional view taken along line B-B of
FIG. 7;
[0042] FIG. 9 is a cross-sectional view taken along line C-C of
FIG. 7;
[0043] FIG. 10 is a plan view of the cathode portion of the
electroplating apparatus shown in FIG. 3;
[0044] FIG. 11 is a cross-sectional view taken along line D-D of
FIG. 10;
[0045] FIG. 12 is a plan view of an electrode arm portion of the
electroplating apparatus shown in FIG. 3;
[0046] FIG. 13 is a schematic sectional view illustrating an
electrode head and the substrate holder of the electroplating
apparatus shown in FIG. 3 when the electroplating apparatus
performs a plating process;
[0047] FIG. 14A through 14C are diagrams illustrating, in a
sequence of process steps, a process for forming copper
interconnects by a substrate processing method (plating method) of
the present invention;
[0048] FIG. 15 is a flow chart showing the substrate processing
method (plating process) of the present invention;
[0049] FIG. 16 is a control flow chart showing control of the
substrate processing method (plating process) of the present
invention until the initiation of electroplating;
[0050] FIG. 17A through 17D are diagrams illustrating, in a
sequence of process steps, a process for forming copper
interconnects by another substrate processing method (plating
method) of the present invention;
[0051] FIG. 18 is a graph showing the relationship between "bottom
up" and "switch on delay time", as determined in Example 1 with
comparison;
[0052] FIG. 19 is a graph showing the relationship between "bottom
up" and "switch on delay time", as determined in Example 2;
[0053] FIGS. 20A through 20C are SEM photographs of plated copper
films, deposited on various substrates by plating carried out with
a "switch on delay time" of 0 second in Example 2;
[0054] FIGS. 21A through 21C are SEM photographs of plated copper
films, deposited on various substrates by plating carried out with
a "switch on delay time" of one second in Example 2;
[0055] FIGS. 22A through 22C are SEM photographs of plated copper
films, deposited on various substrates by plating carried out with
a "switch on delay time" of two seconds in Example 2;
[0056] FIGS. 23A through 23C are SEM photographs of plated copper
films, deposited on various substrates by plating carried out with
a "switch on delay time" of three seconds in Example 2;
[0057] FIG. 24 is a diagram showing the relationship between
"switch on delay time" and the formation of voids in interconnects
having an interconnect pattern of an interconnect width of 0.09
.mu.m, as determined in Example 3;
[0058] FIG. 25 is a diagram showing the relationship between
"switch on delay time" and the formation of voids in interconnects
having an interconnect pattern of an interconnect width of 0.15
.mu.m, as determined in Example 3; and
[0059] FIG. 26 is a graph showing the relationship between "bottom
up" and "switch on delay time", as determined in Example 4.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0060] Preferred embodiments of the present invention will now be
described in detail with reference to the drawings.
[0061] FIG. 3 is an overall plan view showing a substrate
processing apparatus according to an embodiment of the present
invention. As shown in FIG. 3, this substrate processing apparatus
has a rectangular facility which houses therein two
loading/unloading stations 10 for housing a plurality of substrates
W therein, two electroplating apparatuses 12 for performing
electroplating process and processing incidental thereto, a
transfer robot 14 for transferring substrates W between the
loading/unloading stations 10 and the electroplating apparatuses
12, and plating solution supply equipment 18 having a plating
solution tank 16.
[0062] The electroplating apparatus 12, as shown in FIG. 4, is
provided with a substrate processing section 20 for performing
plating process and processing incidental thereto, and a plating
solution tray 22 for storing a plating solution is disposed
adjacent to the substrate processing section 20. There is also
provided with an electrode arm portion 30 having an electrode head
28 which is held at the front end of a swing arm 26 swingable about
a rotating shaft 24 and which is swung between the substrate
processing section 20 and the plating solution tray 22.
Furthermore, a plating solution recovering arm 32, and fixed
nozzles 34 for ejecting pure water or a chemical liquid such as ion
water, and further a gas or the like toward a substrate are
disposed laterally of the substrate processing section 20. In this
embodiment, three of the fixed nozzles 34 are disposed, and one of
them is used for supplying pure water.
[0063] The substrate processing section 20, as shown in FIG. 5, has
a substrate holder 36 for holding a substrate W with its surface
(surface to be plated) facing upwardly, and a cathode portion 38
located above the substrate holder 36 so as to surround a
peripheral portion of the substrate holder 36. Further, a
substantially cylindrical bottomed splash prevention cup 40
surrounding the periphery of the substrate holder 36 for preventing
scatter of various chemical liquids used during processing is
provided so as to be vertically movable by an air cylinder (not
shown).
[0064] The substrate holder 36 is adapted to be raised and lowered
by the air cylinder 44 between a lower substrate transfer position
A, an upper plating position B, and a pretreatment/cleaning
position C intermediate between these positions. The substrate
holder 36 is also adapted to rotate at an arbitrary acceleration
and an arbitrary velocity integrally with the cathode portion 38 by
a rotating motor and a belt (not shown). A substrate carry-in and
carry-out opening (not shown) is provided in confrontation with the
substrate transfer position A in a side panel of the electroplating
apparatus 12 facing the transfer robot 14. When the substrate
holder 36 is raised to the plating position B, a sealing member 90
and cathode contacts 88 (both to be described below) of the cathode
portion 38 are brought into contact with the peripheral portion of
the substrate W held by the substrate holder 36. The splash
prevention cup 40 has an upper end located below the substrate
carry-in and carry-out opening, and when the splash prevention cup
40 ascends, the upper end of the splash prevention cup 40 reaches a
position above the cathode portion 38 closing the substrate
carry-in and carry-out opening, as shown by imaginary lines in FIG.
5.
[0065] The plating solution tray 22 serves to wet a porous
structure 110 and an anode 98 (both to be described below) of the
electrode arm portion 30 with a plating solution, when plating has
not been performed. The plating solution tray 22 is set at a size
in which the porous structure 110 can be accommodated, and the
plating solution tray 22 has a plating solution supply port and a
plating solution drainage port (not shown). A photo-sensor is
attached to the plating solution tray 22, and can detect brimming
with the plating solution in the plating solution tray 22, i.e.,
overflow, and drainage.
[0066] The electrode arm portion 30 is vertically movable by a
vertical movement motor comprised of a servomotor and a ball screw
(not shown), and swingable (pivotable) such that the electrode head
28 moves between the plating solution tray 22 and the substrate
processing section 20 by a swing motor.
[0067] As shown in FIG. 6, the plating solution recovering arm 32
is coupled to an upper end of a vertical support shaft 58. The
plating solution recovering arm 32 is swingable (pivotable) by a
rotary actuator 60 and is also vertically moveable by an air
cylinder (not shown). The plating solution recovering arm 32
supports a plating solution recovering nozzle 66 that is connected
to a cylinder pump or an aspirator, for example, to draw the
plating solution on the substrate from the plating solution
recovering nozzle 66.
[0068] As shown in FIGS. 7 through 9, the substrate holder 36 has a
disk-shaped substrate stage 68 and six vertical support arms 70
disposed at spaced intervals on the circumferential edge of the
substrate stage 68 for holding a substrate W in a horizontal plane
on respective upper surfaces of the support arms 70. A positioning
plate 72 is mounted on an upper end one of the support arms 70 for
positioning the substrate by contacting the end face of the
substrate. A pressing finger 74 is rotatably mounted on an upper
end of the support arm 70, which is positioned opposite to the
support arm 70 having the positioning plate 72, for abutting
against an end face of the substrate W and pressing the substrate W
to the positioning plate 72 when rotated. Chucking fingers 76 are
rotatably mounted on upper ends of the remaining four support arms
70 for pressing the substrate W downwardly and gripping the
circumferential edge of the substrate W.
[0069] The pressing finger 74 and the chucking fingers 76 have
respective lower ends coupled to upper ends of pressing pins 80
that are normally urged to move downwardly by coil springs 78. When
the pressing pins 80 are moved downwardly, the pressing finger 74
and the chucking fingers 76 are rotated radially inwardly into a
closed position. A support plate 82 is disposed below the substrate
stage 68 for engaging lower ends of the pressing pins 80 and
pushing them upwardly.
[0070] When the substrate holder 36 is located in the substrate
transfer position A shown in FIG. 5, the pressing pins 80 are
engaged and pushed upwardly by the support plate 82, so that the
pressing finger 74 and the chucking fingers 76 rotate outwardly and
open. When the substrate stage 68 is elevated, the pressing pins 80
are lowered under the resiliency of the coil springs 78, so that
the pressing finger 74 and the chucking fingers 76 rotate inwardly
and close.
[0071] As shown in FIGS. 10 and 11, the cathode portion 38
comprises an annular frame 86 fixed to upper ends of vertical
support columns 84 mounted on the peripheral portion of the support
plate 82 (see FIG. 9), a plurality of, six in this embodiment,
cathode contacts 88 attached to a lower surface of the annular
frame 86 and projecting inwardly, and an annular sealing member 90
mounted on an upper surface of the annular frame 86 in covering
relation to upper surfaces of the cathode contacts 88. The sealing
member 90 is adapted to have an inner peripheral portion inclined
inwardly downwardly and progressively thin-walled, and to have an
inner peripheral end suspending downwardly.
[0072] When the substrate holder 36 has ascended to the plating
position B shown in FIG. 5, the cathode contacts 88 are pressed
against the peripheral portion of the substrate W held by the
substrate holder 36 for thereby allowing electric current to pass
through the substrate W. At the same time, an inner peripheral
portion of the sealing member 90 is brought into contact with an
upper surface of the peripheral portion of the substrate W under
pressure to seal its contact portion in a watertight manner. As a
result, the plating solution supplied onto the upper surface
(surface to be plated) of the substrate W is prevented from seeping
from the end portion of the substrate W, and the plating solution
is prevented from contaminating the cathode contacts 88.
[0073] In this embodiment, the cathode portion 38 is vertically
immovable, but rotatable in a body with the substrate holder 36.
However, the cathode portion 38 may be arranged such that it is
vertically movable and the sealing member 90 is pressed against the
surface to be plated of the substrate W when the cathode portion 38
is lowered.
[0074] As shown in FIGS. 12 and 13, the electrode head 28 of the
electrode arm portion 30 includes a housing 94 which is coupled via
a ball bearing 92 to the free end of the swing arm 26, and a porous
structure 110 which is disposed such that it closes the bottom
opening of the housing 94. The housing 94 has a downward-open and
cup-like bottomed configuration having at its lower inside surface
an recess portion 94a, while the porous structure 110 has at its
top a flange portion 110a which can engage with the recess portion
94a. The flange portion 11a is inserted into the recess portion
94a. The porous structure 110 is thus held with the housing 94,
while a hollow plating solution chamber 100 is defined in the
housing 94.
[0075] The porous structure 110 per se is an insulator, but is
constructed so as to have a smaller electrical conductivity than
the plating solution by causing the plating solution to enter its
inner part complicatedly and follow a considerably long path in the
thickness direction.
[0076] In the plating solution chamber 100 and located above the
porous structure 110 is disposed an anode 98 having a large number
of vertically-extending through-holes 98c therein. The housing 94
has a plating solution discharge outlet 103 for discharging by
suction a plating solution in the plating solution chamber 100. The
plating solution discharge outlet 103 is connected to a plating
solution discharge pipe 106 extending from the plating solution
supply equipment 18 (see FIG. 3). Further, a plating solution
supply inlet 104, positioned beside the anode 98 and the porous
structure 110 and vertically penetrating the peripheral wall of the
housing 94, is provided within the peripheral wall of the housing
94. In this embodiment, the plating solution supply inlet 104
comprises a tube having a lower end shaped as a nozzle, and is
connected to a plating solution supply pipe 102 extending from the
plating solution supply equipment 18 (see FIG. 3).
[0077] When the substrate holder 36 is in plating position B (see
FIG. 5), the electrode head 28 is lowered until a gap between the
substrate W held by the substrate holder 36 and the porous
structure 110 becomes about 0.5 to 3 mm, for example, and then the
plating solution supply inlet 104 pours the plating solution into a
region between the substrate W and the porous structure 110 from
laterally of the anode 98 and the porous structure 110. The nozzle
at the lower end of the plating solution supply inlet 104 is open
toward the region between the sealing member 90 and the porous
structure 110. A shield ring 112 of rubber is mounted on the outer
circumferential surface of the porous structure 110 for
electrically shielding the porous structure 110.
[0078] When the plating solution is introduced, the plating
solution introduced from the plating solution supply inlet 104
flows in one direction along the surface of the substrate W. The
flow of the plating solution pushes and discharges the air out of
the region between the substrate W and the porous structure 110,
filling the region with the fresh plating solution whose
composition has been adjusted that is introduced from the plating
solution supply inlet 104. The plating solution is now retained in
the region defined between the substrate W and the sealing member
90.
[0079] Beside the housing 94 is disposed a substrate-solution
contact detector 120 as a measurement section, e.g., comprised of
an optical sensor, for detecting contact of the surface of the
substrate W with the plating solution when the plating solution is
supplied from the plating solution supply inlet 104 to the surface
of the substrate W held by the substrate holder 36. The
substrate-solution contact detector 120 detects contact of the
surface of the substrate W with the plating solution, and inputs
the detection signal into, e.g., a control section, thereby
measuring time that has elapsed since the surface of the substrate
W was brought into contact with the plating solution. For a
dip-type electroplating apparatus, for example, it is possible to
use, instead of the substrate-solution contact detector 120, a
position detector for detecting the position of a substrate or a
substrate holder and to input a detection signal from the position
detector into, e.g., a control section. Instead of an optical
sensor, it is also possible to use apressure sensor, a conductivity
sensor, a temperature sensor or an ultrasonic sensor, or a
combination thereof.
[0080] In order to suppress slime formation, the anode 98 is made
of copper (phosphorus-containing copper) containing 0.03 to 0.05%
of phosphorus. The anode 98 may comprise an insoluble electrode
comprising metal on which iridium oxide is coated. The anode 98 is
electrically connected to an anode of a plating power source 114,
and the cathode contacts 88 are electrically connected to a cathode
of the plating power source 114, respectively.
[0081] When the substrate holder 36 is in plating position B (see
FIG. 5), the electrode head 28 is lowered until the gap between the
substrate W held by the substrate holder 36 and the porous
structure 110 becomes about 0.5 to 3 mm, for example. Thereafter, a
plating solution is supplied from the plating solution supply inlet
104 into the region between the substrate W and the porous
structure 410, so that the plating solution fills the region and is
stored in the region defined by the substrate W and the sealing
member 90 for plating.
[0082] A substrate processing method (plating method) according to
an embodiment of the present invention, carried out by using the
substrate processing apparatus, will now be described with
reference to FIGS. 14A through 16.
[0083] First, a substrate W, as shown in FIG. 14A, is provided
which has been prepared by depositing an insulating film
(interlevel dielectric film) 2 of SiO.sub.2 or a low-k material on
a conductive layer 1a containing semiconductor devices, formed on a
semiconductor base 1, forming via holes 3 and trenches 4 as
interconnect recesses in the insulating film 2 by the
lithography/etching technique, and then forming a ruthenium film 5a
as a barrier layer on an entire substrate surface.
[0084] Then, a substrate W to be plated is taken out from one of
the loading/unloading stations 10 by the transfer robot 14, and
transferred, with a surface to be plated facing upwardly, into one
of the electroplating apparatus 12 through a substrate carry-in and
carry-out opening defined in a side panel. At this time, the
substrate holder 36 is in lower substrate transfer position A.
After the hand of the transfer robot 14 has reached a position
directly above the substrate stage 68, the transfer robot 14 lowers
the hand to place the substrate W on the support arms 70. The hand
of the transfer robot 14 is then retracted through the substrate
carry-in and carry-out opening.
[0085] After the hand of the transfer robot 14 is retracted, the
splash prevention cup 40 is elevated. Then, the substrate holder 36
is lifted from lower substrate transfer position A to
pretreatment/cleaning position C. As the substrate holder 36 is
lifted, the substrate W placed on the support arms 70 is positioned
by the positioning plate 72 and the pressing finger 74, and then
reliably gripped by the chucking fingers 76.
[0086] On the other hand, the electrode head 28 of the electrode
arm portion 30 is in a normal position over the plating solution
tray 22 now, and the porous structure 110 or the anode 98 is
positioned in the plating solution tray 22. At the same time that
the splash prevention cup 40 ascends, the plating solution starts
being supplied to the plating solution tray 22 and the electrode
head 28. Until the step of plating the substrate W is initiated,
the new plating solution is supplied, and the plating solution
discharge pipe 106 is evacuated to replace the plating solution in
the porous structure 110 and remove air bubbles from the plating
solution in the porous structure 110. When the ascending movement
of the splash prevention cup 40 is completed, the substrate
carry-in and carry-out opening in the side panel is closed by the
splash prevention cup 40, thereby isolating the atmosphere in the
side panel and the atmosphere outside of the side panel from each
other.
[0087] Next, the plating step is initiated. First, the substrate
holder 36 is lifted to plating position B while the substrate
holder 36 not being rotated, or being rotated at a preset
rotational speed for plating. Then, the peripheral portion of the
substrate W is brought into contact with the cathode contacts 88,
when it is possible to pass an electric current, and at the same
time, the sealing member 90 is pressed against the upper surface of
the peripheral portion of the substrate W, thus sealing the
peripheral portion of the substrate W in a watertight manner.
[0088] Based on a signal indicating that the substrate W has been
loaded, the electrode arm portion 30 is swung in a horizontal
direction to displace the electrode head 28 from a position over
the plating solution tray 22 to a position over the plating
processing position. After the electrode head 28 reaches this
position, the electrode head 28 is lowered toward the cathode
portion 38 and stopped. At this time, the porous structure 110 does
not contact with the surface of the substrate W, but is held
closely to the surface of the substrate W at a distance ranging
from 0.5 mm to 3 mm. The plating solution is then poured from the
plating solution supply inlet 104 into the region between the
substrate W and the porous structure 110, filling the region with
the plating solution.
[0089] Upon the injection of the plating solution, contact of the
surface of the substrate W with the plating solution is detected
with the substrate-solution contact detector 120. The surface of
the substrate W is kept in contact with the plating solution for a
predetermined time to adsorb an additive(s) in the plating solution
onto the ruthenium film 5a. Thus, as shown in FIG. 16, after
detecting contact of the surface of the substrate W with the
plating solution with the substrate-solution contact detector 120,
the elapsed contact time is measured. The predetermined time for
the substrate surface to be kept in contact with the plating
solution is, for example, not less than 0.5 second and not more
than 60 seconds, preferably not less than 0.1 second and not more
than 20 seconds, and more preferably not less than 0.1 second and
not more than 5 seconds.
[0090] After an elapse of the predetermined contact time, a voltage
is applied from the plating power source 114 to between the cathode
contacts 88 and the anode 98, thereby forming a plated copper film
6 on the surface of the ruthenium film (barrier layer) 5a while
filling copper into the via holes 3 and the trenches 4, as shown in
FIG. 14B. By thus initiating electroplating after bringing the
ruthenium film 5a into contact with the plating solution and
leaving the ruthenium film 5a in contact with the plating solution
for a certain period of time to adsorb an additive(s) in the
plating solution onto the ruthenium film 5a, the plated copper film
6 can be formed on the surface of the ruthenium film 5a while
suppressing the formation of voids in the copper plated film 6 and
ensuring trench-filling characteristics comparable to the
conventional electroplating method using a seed layer.
[0091] By carrying out electroplating directly on the surface of
the ruthenium film 5a, having a low electrical resistance, as a
barrier layer according to the above-described method of the
present invention, the plated copper film 6 can be securely filled
into interconnect recesses, such as trenches, without forming
defects, such as voids, in the embedded copper film 6. Thus, the
present method makes it possible to enhance the reliability of
interconnects while enjoying the benefit of elimination of a seed
layer formation process. The ruthenium film 5a as a barrier layer
will not dissolve even in such an acidic plating solution as
containing a copper ion, a sulfate ion and an additive(s) during
electroplating.
[0092] When the plating process is completed, the electrode arm
portion 30 is raised and then swung to return to the position above
the plating solution tray 22 and to lower to the ordinary position.
Then, the remainder of the plating solution on the substrate W is
recovered by a plating solution recovering nozzle 66. After
recovering of the remainder of the plating solution is completed,
pure water is supplied from the fixed nozzle 34 for supplying pure
water toward the central portion of the substrate W and the
substrate holder 36 is rotated at an increased speed to replace the
plating solution on the surface of the substrate W with pure water.
Rinsing the substrate W in this manner prevents the splashing
plating solution from contaminating the cathode contacts 88 of the
cathode portion 38 during descent of the substrate holder 36 from
plating position B.
[0093] After completion of the rinsing, the washing with water step
is initiated. That is, the substrate holder 36 is lowered from
plating position B to pretreatment/cleaning position C. Then, while
pure water is supplied from the fixed nozzle 34 for supplying pure
water, the substrate holder 36 and the cathode portion 38 are
rotated to perform washing with water. At this time, the sealing
member 90 and the cathode contacts 88 can also be cleaned,
simultaneously with the substrate W, by pure water directly
supplied to the cathode portion 38, or pure water scattered from
the surface of the substrate W.
[0094] After washing with water is completed, the drying step is
initiated. That is, supply of pure water from the fixed nozzle 34
is stopped, and the rotational speed of the substrate holder 36 and
the cathode portion 38 is further increased to remove pure water on
the surface of the substrate W by centrifugal force and to dry the
surface of the substrate W. The sealing member 90 and the cathode
contacts 88 are also dried at the same time. Upon completion of the
drying, the rotation of the substrate holder 36 and the cathode
portion 38 is stopped, and the substrate holder 36 is lowered to
substrate transfer position A. Thus, the gripping of the substrate
W by the chucking fingers 76 is released, and the substrate W is
just placed on the upper surfaces of the support arms 70. At the
same time, the splash prevention cup 40 is also lowered.
[0095] All the steps including the plating step, the pretreatment
step accompanying to the plating step, the cleaning step, and the
drying step are now finished. The transfer robot 14 inserts its
hand through the substrate carry-in and carry-out opening into the
position beneath the substrate W, and raises the hand to receive
the plated substrate W from the substrate holder 36. Then, the
transfer robot 14 returns the plated substrate W received from the
substrate holder 36 to one of the loading/unloading stations 10.
Thereafter, the plated copper film 6 and the ruthenium film
(barrier layer) 5a on the insulating film 2 are removed by chemical
mechanical polishing (CMP) so as to form interconnects composed of
the plated copper film 6 in the insulating film 2, as shown in FIG.
14C.
[0096] A substrate processing method (plating method) according to
another embodiment of the present invention, carried out by using
the above-described substrate processing apparatus, will now be
described with reference to FIGS. 17A through 17D.
[0097] First, a substrate W, as shown in FIG. 17A, is provided
which has been prepared by forming trenches 4a as interconnect
recesses in an insulating film (interlevel dielectric film) 2a of
SiO.sub.2 or a low-k material, and forming a ruthenium film 5b as a
barrier layer on a substrate of the insulating film 2a.
[0098] As in the above-described embodiment, after bringing the
porous structure 110 to a position as close to the surface of the
substrate W as about 0.5 to 3 mm, a plating solution is injected
from the plating solution supply inlet 104 into the region between
the substrate W and the porous structure 110 to fill the region
with the plating solution. Upon the injection of the plating
solution, contact of the surface of the substrate W with the
plating solution is detected with the substrate-solution contact
detector 120. The surface of the substrate W is kept in contact
with the plating solution for a predetermined time to adsorb an
additive(s) in the plating solution onto the ruthenium film 5b. The
predetermined time for keeping the substrate surface in contact
with the plating solution is generally not less than 5 seconds,
preferably not less than 20 seconds, e.g., 30 seconds.
[0099] After an elapse of the predetermined contact time, a voltage
is applied from the plating power source 114 to between the cathode
contacts 88 and the anode 98, thereby conformally forming an
initial plated copper film 6a, which uniformly covers the entire
interior surfaces of the trenches 4a, on the surface of the
ruthenium film 5b, as shown in FIG. 17B.
[0100] Next, as in the above-described embodiment, after rinsing
the plated surface of the substrate W with pure water, the
rotational speed of the substrate holder 36 and the cathode portion
38 is increased to spin-dry the substrate W, e.g., at 1500 rpm for
30 seconds, thereby obtaining a dried substrate W, as shown in FIG.
17C, having the initial plated copper film 6a formed conformally on
the ruthenium film 5b.
[0101] Next, plating is carried out using the initial plated copper
film 6a as a seed layer to allow a plated copper film to grow on a
surface of the initial plated copper film 6a, thereby filling a
plated copper film 6 into the trenches 4a, as shown in FIG. 17D. In
particular, after bringing the porous structure 110 to a position
as close to the surface of the substrate W as about 0.5 to 3 mm, a
plating solution is injected from the plating solution supply inlet
104 into the region between the substrate W and the porous
structure 110 to fill the region with the plating solution while
applying a voltage from the plating power source 114 to between the
cathode contacts 88 and the anode 98, thereby allowing a plated
copper film to grow on the surface of the initial plated copper
film 6a.
[0102] After filling the plating solution into the region between
the substrate W and the porous structure 110 by injecting the
plating solution from the plating solution supply inlet 104, it is
also possible to keep the surface of the substrate W in contact
with the plating solution for a short time to adsorb a small amount
of an additive(s) in the plating solution onto the initial plated
copper film 6a, and then apply a voltage from the plating power
source 114 to between the cathode contacts 88 and the anode 98.
[0103] The subsequent procedures are the same as described above,
and hence a description thereof is omitted.
Example 1
[0104] A substrate sample was prepared by forming an SiO.sub.2 film
over a silicon substrate, forming a groove-shaped interconnect
pattern (interconnect width not less than 0.1 .mu.m) in the
SiO.sub.2 film, and then forming a ruthenium film as a barrier
layer on the entire substrate surface. A copper sulfate plating
solution, which is commonly used for forming a plated copper film
in the process of forming interconnects in a semiconductor device,
and contains an accelerator, a suppressor and a leveler as
additives in appropriate amounts, was used as a copper-plating
solution. An insoluble anode coated with iridium oxide was used. A
series of electroplating tests was carried out on the same
substrate samples under the same plating conditions but varying the
time period from contact of the substrate with the plating solution
to the initiation of electroplating, thereby allowing a plated
copper film to grow on a surface of the ruthenium film, including
the interior surfaces of trenches. A cross section of each sample
after plating was observed under a scanning electron microscope to
measure a thickness "a", shown in FIG. 18, of the plated copper
film in trenches and a thickness "b", shown in FIG. 18, of the
plated copper film in the substrate surface. The trench-filling
selectivity ratio a/b was calculated from the measured thicknesses
"a" and "b" of the plated copper film for all the test samples to
determine the relationship of the selectivity ratio with the time
period from contact of the substrate with the plating solution to
the initiation of electroplating. The graph with the mark "A" in
FIG. 18 shows the relationship between the time period from contact
of the substrate with the plating solution to the initiation of
electroplating (switch on delay time) and the trench-filling
selectivity ratio [bottom up (a/b)].
[0105] For comparison, after depositing a copper seed layer by PVD,
plated copper film was formed on a surface of the copper seed
layer, and the trench-filling selectivity ratio a/b was calculated
from the measured thicknesses "a" and "b" of the plated copper
film. The graph with the mark "O" in FIG. 18 shows the relationship
between the time period from contact of the substrate with the
plating solution to the initiation of electroplating and this
trench-filling selectivity ratio (a/b). The copper seed layer was
formed on a surface of a silicon substrate having a interconnect
pattern of an interconnect width of 0.2 .mu.m.
[0106] The above-described two substrates differ in the
interconnect structure and the aspect ratio. Further, different
types of additives are used and thicknesses of plated films differ
each other. Accordingly, simple comparison of the values cannot be
made.
[0107] As can be seen from FIG. 18, when copper plating is carried
out on the copper seed layer, the trench-filling selectivity ratio
[bottom up (a/b)] generally decreases, i.e., the trench-filling
characteristics become worse, with an increase in the "switch on
delay time". In contrast, when copper plating is carried out
directly on the ruthenium film, the trench-filling selectivity
ratio [bottom up (a/b)] generally increases, i.e., the
trench-filling characteristics become better, with an increase in
the "switch on delay time". Further, the cross-sectional
observation revealed that as the "switch on delay time" increases,
the form of plated film in the initial plating stage changes from a
particulate form to a continuous film. These facts indicate that
when carrying out copper plating directly on a surface of a
ruthenium film, it is advantageous to initiate electroplating after
bringing a surface of a substrate into contact with a plating
solution and leaving the substrate in contact with the plating
solution for a certain period of time in order to suppress the
formation of voids or seams in a plated copper film embedded in
trenches and thereby ensure the reliability of interconnects.
Example 2
[0108] Substrates, each having an interconnect pattern of an
interconnect depth of 0.25 .mu.m and an interconnect width of 0.1
.mu.m, 0.2 .mu.m or 0.25 .mu.m and having a 3 nm-thick surface
ruthenium film formed by CVD, were prepared. The sheet resistance
of the ruthenium film was about 150 .OMEGA./sq. Copper plating of a
surface of each substrate was carried out in an electrolytic amount
corresponding to a plating thickness of 55 nm, using a plating
solution having a copper concentration of 50 g/l, a sulfuric acid
concentration of 80 g/l and a chlorine concentration of 50 ppm, and
containing an accelerator, a suppressor and a leveler as additives.
A series of electroplating tests was carried out on each substrate
at a current density of 5 to 40 mA/cm.sup.2 by varying the time
period from contact of the substrate with the plating solution to
the initiation of electroplating (switch on delay time) as follows:
0 second, 1 second, 2 seconds and 3 seconds. With reference to the
interconnect pattern, the trenches had a curved cross-sectional
contour at the openings (see FIGS. 24 and 25).
[0109] FIG. 19 shows the relationship between the time period from
contact of the substrate with the plating solution to the
initiation of electroplating (switch on delay time) and the
trench-filling selectivity ratio [bottom up (a/b)] shown in FIG.
18. As shown in FIG. 19, the selectivity ratio [bottom up (a/b)]
generally increases with an increase in an immersion time (switch
on delay time), and this is marked in the interconnect pattern of
an interconnect width of 0.1 .mu.m. This is considered to be due to
the fact that an increase in the period of time during which the
substrate is immersed in the plating solution without application
of a voltage, may lead to increased adsorption of the accelerator
(containing sulfur), contained as an additive in the plating
solution, onto the surface of the ruthenium film in the
interconnects.
[0110] FIGS. 20A through 20C are SEM photographs of plated copper
films as deposited on the substrates when the "switch on delay
time" was 0 second; FIGS. 21A through 21C are SEM photographs of
plated copper films as deposited on the substrates when the "switch
on delay time" was 1 second; FIGS. 22A through 22C are SEM
photographs of plated copper films as deposited on the substrates
when the "switch on delay time" was 2 seconds; and FIGS. 23A
through 23C are SEM photographs of plated copper films as deposited
on the substrates when the "switch on delay time" was 3
seconds.
[0111] As can be seen from FIGS. 20A through 23C, a longer
immersion time (switch on delay time) results in a denser crystal
deposition. This also is considered to be due to increased
adsorption of the accelerator onto the ruthenium film. It is
expected that as the crystal deposition becomes denser, the
formation of voids in interconnects will be reduced.
Example 3
[0112] Substrates, each having an interconnect pattern of an
interconnect depth of 0.25 .mu.m and an interconnect width of 0.09
.mu.m or 0.15 .mu.m and having a 3 nm-thick surface ruthenium film
formed by CVD, were prepared. The sheet resistance of the ruthenium
film was about 150 .OMEGA./sq. Copper plating of a surface of each
substrate was carried out in an electrolytic amount corresponding
to a plating thickness of 55 nm, using a plating solution having a
copper concentration of 50 g/l, a sulfuric acid concentration of 80
g/l and a chlorine concentration of 50 ppm, and containing an
accelerator, a suppressor and a leveler as additives. A series of
electroplating tests was carried out on each substrate at a current
density of 5 to 40 mA/cm.sup.2 by varying the time period from
contact of the substrate with the plating solution to the
initiation of electroplating (switch on delay time) as follows: 0
second, 3 seconds, 5 seconds, 7 seconds and 1 minute.
[0113] FIG. 24 shows the relationship between the time period from
contact of the substrate with the plating solution to the
initiation of electroplating (switch on delay time) and the
formation of voids in the interconnects having the interconnect
pattern of an interconnect width of 0.09 .mu.m; and FIG. 25 shows
the relationship between the time period from contact of the
substrate with the plating solution to the initiation of
electroplating (switch on delay time) and the formation of voids in
the interconnects having the interconnect pattern of an
interconnect width of 0.15 .mu.m.
[0114] FIGS. 24 and 25 each illustrate a plated copper film 6b,
which is to become interconnects, embedded in trenches 4b formed in
an insulating film (interlevel dielectric film) 2b which is covered
with a ruthenium film 5c.
[0115] As shown in FIG. 24, in the case of the interconnect pattern
of an interconnect width of 0.09 .mu.m, the formation of voids V in
the interconnects (plated copper film) was observed when an
immersion time (switch on delay time) was 0 second or one minute.
It is considered in this regard that when the immersion time
(switch on delay time) is 0 second, the amount of the accelerator
(containing sulfur), contained as an additive in the plating
solution, adsorbed onto the surface of the ruthenium film in the
interconnects is too small, whereas when the immersion time (switch
on delay time) is one minute, the amount of the suppressor,
contained as an additive in the plating solution, adsorbed onto the
surface of the ruthenium film is so large as to lower the plating
rate, resulting in the formation of voids in the interconnects. In
the case of the interconnect pattern of an interconnect width of
0.15 .mu.m, on the other hand, no formation of voids in the
interconnects was observed for any switch on delay time, as will be
appreciated from FIG. 25.
Example 4
[0116] A substrate having an interconnect pattern of an
interconnect width of 0.08 .mu.m and an interconnect depth of 0.22
.mu.m and having a 2 nm-thick surface ruthenium film formed by CVD
was prepared. The sheet resistance of the ruthenium film was about
250 .OMEGA./sq. Copper plating of a surface of the substrate was
carried out in an electrolytic amount corresponding to a plating
thickness of 40 nm, using a plating solution having a copper
concentration of 50 g/l, a sulfuric acid concentration of 80 g/l
and a chlorine concentration of 50 ppm, and containing an
accelerator, a suppressor and a leveler as additives. A series of
electroplating tests was carried out on the substrate at a current
density of 5 to 40 mA/cm.sup.2 by varying the time period from
contact of the substrate with the plating solution to the
initiation of electroplating (switch on delay time) as follows: 0
second, 2 seconds, 5 seconds, 10 seconds and 20 seconds. With
reference to the interconnect pattern, the trenches had a
relatively sharp cross-sectional contour at the openings.
[0117] FIG. 26 shows the relationship between the time period from
contact of the substrate with the plating solution to the
initiation of electroplating (switch on delay time) and the
trench-filling selectivity ratio [bottom up (a/b)] shown in FIG.
18. Since the trench-filling selectivity ratio [bottom up (a/b)]
varies depending on the interconnect width, the interconnect depth
and the electrolytic amount, proper comparison must be made under
the same experimental conditions.
[0118] As can be seen from FIG. 26, the trench-filling selectivity
ratio [bottom up (a/b)] is large when the "switch on delay time" is
0 to 5 seconds, whereas it is small when the "switch on delay time"
is more than 10 seconds and reaches saturation at the "switch on
delay time" of about 20 seconds. The trench-filling selectivity
ratio [bottom up (a/b)] is an index of filling of trenches. Thus,
in the case of the plating solution used in this Example, the
immersion time (switch on delay time) is preferably not more than 5
seconds, and more preferably not more than 2 seconds for better
filling of trenches. This is because adsorption of the accelerator
(containing sulfur), contained as an additive in the plating
solution, onto the ruthenium film reaches saturation in 2 to 5
seconds, while adsorption of the suppressor, contained as an
additive in the plating solution, onto the ruthenium film reaches
saturation in 20 seconds.
[0119] Thus, when the immersion time (switch on delay time) is 2 to
5 seconds, because of the large adsorption of the accelerator, the
trench-filling selectivity ratio [bottom up (a/b)] is large,
indicating plating of good trench-filling characteristics. When the
immersion time (switch on delay time) is 20 seconds, on the other
hand, conformal plating will be performed in the interconnects due
to the large adsorption of the suppressor.
Example 5
[0120] Based on the results of Example 4, an experiment was
conducted to determine if void-free plating is possible when
conformal plating is carried out in combination with plating of
good trench-filling characteristics. First, a substrate having an
interconnect pattern of an interconnect width of 0.08 .mu.m and an
interconnect depth of 0.22 .mu.m and having a 2 nm-thick surface
ruthenium film formed by CVD was prepared. The sheet resistance of
the ruthenium film was about 250 .OMEGA./sq. After keeping the
substrate in contact with the same plating solution as used in
Example 4 for 20 seconds ("switch on delay time" of 20 seconds),
copper plating of a substrate surface with the plating solution was
carried out in an electrolytic amount corresponding to a plating
thickness of 25 nm (at a current density of 5 to 40 mA/cm.sup.2),
thereby conformally forming a plated copper film in the
interconnect pattern. Thereafter, the substrate was rinsed with
water for 60 seconds and then spin-dried at 1500 rpm for 30
seconds, thereby removing the plating solution from the substrate.
Next, with the immersion time (switch on delay time) of 0 second,
electroplating of the substrate surface was carried out in an
electrolytic amount corresponding to a plating thickness of 500 nm
(at a current density of 5 to 40 mA/cm.sup.2), using the same
plating solution as used in the first electroplating. On
cross-sectional observation of the plated substrate, no void was
found in the interconnects.
[0121] While the present invention has been described with
reference to the embodiments thereof, it will be understood by
those skilled in the art that the present invention is not limited
to the particular embodiments described above, but it is intended
to cover modifications within the inventive concept. For example,
though the use of copper as an interconnect material has been
described, it is also possible to use a copper alloy instead of
copper.
* * * * *