U.S. patent application number 11/694856 was filed with the patent office on 2007-12-13 for method to minimize wet etch undercuts and provide pore sealing of extreme low k (k<2.5) dielectrics.
Invention is credited to Amir Al-Bayati, Hichem M'Saad, Mei-Yee Shek, Derek Witty, Li-Qun Xia, Huiwen Xu.
Application Number | 20070287301 11/694856 |
Document ID | / |
Family ID | 46206138 |
Filed Date | 2007-12-13 |
United States Patent
Application |
20070287301 |
Kind Code |
A1 |
Xu; Huiwen ; et al. |
December 13, 2007 |
METHOD TO MINIMIZE WET ETCH UNDERCUTS AND PROVIDE PORE SEALING OF
EXTREME LOW K (K<2.5) DIELECTRICS
Abstract
Methods of processing films on substrates are provided. In one
aspect, the methods comprise treating a patterned low dielectric
constant film after a photoresist is removed form the film by
depositing a thin layer comprising silicon, carbon, and optionally
oxygen and/or nitrogen on the film. The thin layer provides a
carbon-rich, hydrophobic surface for the patterned low dielectric
constant film. The thin layer also protects the low dielectric
constant film from subsequent wet cleaning processes and
penetration by precursors for layers that are subsequently
deposited on the low dielectric constant film.
Inventors: |
Xu; Huiwen; (Sunnyvale,
CA) ; Shek; Mei-Yee; (Mountain View, CA) ;
Xia; Li-Qun; (Santa Clara, CA) ; Al-Bayati; Amir;
(San Jose, CA) ; Witty; Derek; (Fremont, CA)
; M'Saad; Hichem; (Santa Clara, CA) |
Correspondence
Address: |
PATTERSON & SHERIDAN, LLP
3040 POST OAK BOULEVARD, SUITE 1500
HOUSTON
TX
77056
US
|
Family ID: |
46206138 |
Appl. No.: |
11/694856 |
Filed: |
March 30, 2007 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11668911 |
Jan 30, 2007 |
|
|
|
11694856 |
|
|
|
|
60866770 |
Nov 21, 2006 |
|
|
|
60790254 |
Apr 7, 2006 |
|
|
|
60788279 |
Mar 31, 2006 |
|
|
|
Current U.S.
Class: |
438/780 ;
257/E21.259; 257/E21.261 |
Current CPC
Class: |
C23C 16/401 20130101;
H01L 21/02126 20130101; H01L 21/3122 20130101; H01L 21/02274
20130101; H01L 21/0228 20130101; H01L 21/76831 20130101; H01L
21/02164 20130101; C23C 16/308 20130101; H01L 21/76834 20130101;
H01L 21/31612 20130101; H01L 21/0217 20130101; H01L 21/76829
20130101; H01L 21/31633 20130101 |
Class at
Publication: |
438/780 ;
257/E21.259 |
International
Class: |
H01L 21/312 20060101
H01L021/312 |
Claims
1. A method of processing a film on a substrate in a chamber,
comprising: treating the film by selectively depositing a thin
layer having a thickness of between about 4 .ANG. and about 100
.ANG. and comprising silicon, carbon, and hydrogen on an
oxygen-rich or nitrogen-rich surface of the film, wherein
depositing the layer comprises reacting a precursor comprising Si,
C, and H in the presence of RF power.
2. In the method of claim 1, wherein the precursor is selected from
the group trimethyl silane; tetramethyl silane;
dimethyldimethoxysilane; 1,3-dimethyldisiloxane;
1,1,3,3-tetramethyldisiloxane; hexamethyldisiloxane;
hexamethylcyclotrisiloxane; 1,3,5,7-tetramethylcyclotetrasiloxane
(TMCTS); octamethylcyclotetrasiloxane (OMCTS); and
1,3,5,7,9-pentamethylcyclopentasiloxane.
3. The method of claim 1, wherein the precursor comprises an alkyl
group that is selected to suppress continued growth of the thin
layer.
4. The method of claim 1, wherein the thin layer has a higher
carbon content than the oxygen-rich or nitrogen-rich surface of the
film, and the thin layer provides a carbon-saturated surface layer
on the film.
5. The method of claim 1, further comprising wet cleaning the
substrate after the thin layer is deposited.
6. The method of claim 1, wherein the RF power is applied at a
power level of about 0.109 W/cm.sup.2 or less.
7. The method of claim 1, wherein the pressure in the chamber is
about 1.5 Torr or greater.
8. The method of claim 1, wherein the spacing between a showerhead
in the chamber and a substrate support in the chamber is greater
than about 200 mils.
9. The method of claim 1, further comprising plasma post-treating
the layer, using a gas selected from the group consisting of
O.sub.2, CO.sub.2, N.sub.2O, NH.sub.3, H.sub.2, helium, argon,
nitrogen, and combinations thereof.
10. The method of claim 1, further comprising plasma post-treating
the layer, wherein the plasma post-treating modifies the surface
characteristics of the layer, and wherein the surface
characteristics are selected from the group consisting of surface
tension and surface contact angle.
11. The method of claim 1, further comprising depositing a bottom
anti-reflective coating (BARC) on the thin layer.
12. The method of claim 1, further comprising depositing a barrier
layer by atomic layer deposition or physical vapor deposition on
the thin.
13. (canceled)
14. A method of controlling the thickness of a layer to between
about 4 .ANG. and about 100 .ANG. on a substrate, comprising:
exposing a substrate to a silicon-containing precursor in the
presence of a plasma to deposit a layer on the substrate; treating
the layer after it is deposited with a plasma from NH.sub.3 with or
without H.sub.2, and an oxygen-containing gas selected from the
group consisting of O.sub.2, CO.sub.2, and N.sub.2O; and repeating
the exposing and treating until a desired thickness of the layer is
obtained.
15. The method of claim 14, wherein the layer is deposited on an
oxygen-rich or nitrogen-rich surface of the substrate.
16. The method of claim 14, wherein the silicon-containing
precursor is selected from the group consisting of precursors
containing Si, C, and H; trimethyl silane; tetramethyl silane;
dimethyldimethoxysilane; 1,3-dimethyldisiloxane;
1,1,3,3-tetramethyldisiloxane; hexamethyldisiloxane;
hexamethylcyclotrisiloxane; 1,3,5,7-tetramethylcyclotetrasiloxane
(TMCTS); octamethylcyclotetrasiloxane (OMCTS); and
1,3,5,7,9-pentamethylcyclopentasiloxane.
17. A method of controlling the thickness of a metal line in an
interconnect, comprising: exposing a patterned substrate comprising
an interconnect to a silicon-containing precursor in the presence
of a plasma to deposit a layer in the interconnect; treating the
layer after it is deposited with a plasma from an oxygen-containing
gas selected from the group consisting of O.sub.2, CO.sub.2, and
N.sub.2O, or NH.sub.3 with or without H.sub.2; and repeating the
exposing and treating until a desired thickness of the layer is
obtained to provide a desired thickness of a subsequently deposited
metal line in the interconnect.
18. The method of claim 17, wherein the layer is deposited on an
oxygen-rich or nitrogen-rich surface of the substrate.
19. The method of claim 18, wherein the silicon-containing
precursor is selected from the group consisting of trimethyl
silane; tetramethyl silane; dimethyldimethoxysilane;
1,3-dimethyldisiloxane; 1,1,3,3-tetramethyidisiloxane;
hexamethyldisiloxane; hexamethylcyclotrisiloxane;
1,3,5,7-tetramethylcyclotetrasiloxane (TMCTS);
octamethylcyclotetrasiloxane (OMCTS); and
1,3,5,7,9-pentamethylcyclopentasiloxane.
20. A method of producing a dense dielectric spacer containing
either oxide or nitride, comprising: exposing a patterned substrate
comprising a gate to a silicon-containing precursor in the presence
of a plasma to deposit a layer on the gate; treating the layer
after it is deposited with a plasma from an oxygen or N-containing
gas selected from the group consisting of O.sub.2, CO.sub.2,
N.sub.2O, a nitrogen-containing gas, and NH.sub.3 with or without
H.sub.2; and repeating the exposing and treating until a desired
thickness of the layer is obtained.
21. The method of claim 1, wherein selectively depositing the thin
layer is performed after the oxygen-rich or nitrogen rich surface
of the film is formed during a photoresist removal process.
22. The method of claim 4, wherein the surface of the thin layer is
hydrophobic.
23. The method of claim 1, further comprising wet cleaning the
substrate after depositing the thin layer by exposing the thin
layer to a wet cleaning chemistry comprising hydrofluoric acid
(HF), wherein the thin layer is adapted to substantially prevent
etching of the film during the wet cleaning.
24. The method of claim 23, further comprising depositing a barrier
layer or a BARO layer on the thin layer after wet cleaning the thin
layer.
25. The method of claim 1, wherein the thin layer is sufficiently
dense to prevent the penetration of a material used to form a
barrier layer or a BARC layer into the film.
26. The method of claim 17, wherein the layer is a conformal layer
that is deposited using a precursor containing an alkyl group.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims benefit of U.S. provisional patent
application Ser. No. 60/866,770, filed Nov. 21, 2006, which is
herein incorporated by reference. This application is also a
continuation-in-part of U.S. patent application Ser. No.
11/668,911, filed on Jan. 30, 2007, which claims benefit of U.S.
provisional patent application Ser. No. 60/790,254, filed on Apr.
7, 2006, and of U.S. provisional patent application Ser. No.
60/788,279, filed Mar. 31, 2006, all of which are herein
incorporated by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] Embodiments of the present invention generally relate to the
fabrication of integrated circuits. More particularly, embodiments
of the present invention relate to a process for depositing thin
layers comprising silicon, carbon, and optionally oxygen and/or
nitrogen on low dielectric constant layers.
[0004] 2. Description of the Related Art
[0005] Integrated circuit geometries have dramatically decreased in
size since such devices were first introduced several decades ago.
Since then, integrated circuits have generally followed the two
year/half-size rule (often called Moore's Law), which means that
the number of devices on a chip doubles every two years. Today's
fabrication facilities are routinely producing devices having 0.13
.mu.m and even 0.1 .mu.m feature sizes, and tomorrow's facilities
soon will be producing devices having even smaller feature
sizes.
[0006] The continued reduction in device geometries has generated a
demand for inter layer dielectric films having lower dielectric
constant (k) values because the capacitive coupling between
adjacent metal lines must be reduced to further reduce the size of
devices on integrated circuits. In particular, insulators having
low dielectric constants, less than about 4.0, are desirable.
[0007] More recently, low dielectric constant organosilicon films
having dielectric constants less than about 3.0 have been
developed. Extreme low k (ELK) films having dielectric constants
less than 2.5 have also been developed. One method that has been
used to develop low dielectric and extreme low dielectric constant
organosilicon films has been to deposit the films from a gas
mixture comprising an organosilicon compound and a compound, such
as a hydrocarbon, comprising thermally labile species or volatile
groups and then post-treat the deposited films to remove the
thermally labile species or volatile groups, such as organic
groups, from the deposited films. The removal of the thermally
labile species or volatile groups from the deposited films creates
nanometer-sized voids or pores in the films, which lowers the
dielectric constant of the films, as air has a dielectric constant
of approximately 1.
[0008] Ashing processes to remove photoresists or bottom
anti-reflective coatings (BARC) can deplete carbon from the low k
films and oxidize the surface of the films. The oxidized surface of
the low k films is removed during subsequent wet etch processes and
contributes to undercuts and critical dimension (CD) loss.
[0009] The porosity of the low dielectric constant films can also
result in the penetration of precursors used in the deposition of
subsequent layers on the films, such as BARC layers or
intermetallic barrier layers (TaN, etc.). The diffusion of barrier
layer precursors into the porous low dielectric constant films
results in current leakage in a device.
[0010] Therefore, there remains a need for a method of processing
low dielectric constant films that minimizes damage to the films
from subsequent processing steps, such as wet etch processes and
the deposition of subsequent layers, such as BARC layers and
barrier layers.
SUMMARY OF THE INVENTION
[0011] The present invention generally provides a method of
depositing a thin, conformal pore-sealing surface layer on a low
dielectric constant film on a substrate in a chamber. The method
comprises removing a photoresist from a patterned low dielectric
constant film, and then treating the patterned low dielectric
constant film having any aspect ratio or via dimension by
depositing a thin, conformal layer having a controlled thickness of
between about 4 .ANG. and about 100 .ANG. and comprising silicon
and carbon and optionally oxygen and/or nitrogen on a surface of
the patterned low dielectric constant film. In one embodiment,
depositing the layer comprises reacting
octamethylcyclotetrasiloxane in the presence of a low level of RF
power. Ashing a photoresist depletes carbon from the surface of the
low dielectric constant film, and the surface becomes hydrophilic.
The deposited layer recovers the surface carbon concentration of
the low dielectric constant film after ashing and provides a
hydrophobic surface for the patterned low dielectric constant film.
The wet etch rate of a low dielectric constant film is minimized
when its surface is hydrophobic. The layer protects the low
dielectric constant film from subsequent wet cleaning processes
that may be performed on the substrate and prevents undercuts and
CD loss. The hydrophobic surface provided by the thin layer
prevents moisture adsorption into the low dielectric constant
films.
[0012] A low dielectric constant film surface becomes oxidized and
contains OH groups after photoresist ashing. The surface absorbs
moisture and greatly increases the dielectric constant. Deposition
of the thin layer after photoresist ashing drives out moisture
absorbed in the surface and removes OH groups at the surface of the
low dielectric constant film, and thus recovers the low dielectric
constant. The deposition of the thin layer provides a hydrophobic
sealing layer which prevents further moisture adsorption.
[0013] Octamethylcyclotetrasiloxane (OMCTS) is an example of a
precursor that may be used to deposit the thin layers described
herein. In addition to octamethylcyclotetrasiloxane, precursors
having the general formula R.sub.x--Si--(OR').sub.y, such as
dimethyldimethoxysilane (CH.sub.3).sub.2--Si--(O--CH.sub.3).sub.2,
wherein R.dbd.H, CH.sub.3, CH.sub.2CH.sub.3, or another alkyl
group, R'.dbd.CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group, x
is from 0 to 4, y is 0 to 4, and x+y=4, may also be used to deposit
a thin conformal layer with a suitable process window. Other
precursors that may be used include linear organosiloxanes. The
linear organosiloxanes may include the structure
(R.sub.X--Si--O--Si--R.sub.Y).sub.z, such as 1,3-dimethyldisiloxane
(CH.sub.3--SiH.sub.2--O--SiH.sub.2--CH.sub.3),
1,1,3,3-tetramethyldisiloxane
((CH.sub.3).sub.2--SiH--O--SiH--(CH.sub.3).sub.2),
hexamethyldisiloxane
((CH.sub.3).sub.3--Si--O--Si--(CH.sub.3).sub.3), etc. Other
precursors that may be used include cyclic organosiloxanes
(R.sub.X--Si--O).sub.Y, wherein R.sub.X.dbd.CH.sub.3,
CH.sub.2CH.sub.3, or another alkyl group, and R.sub.Y.dbd.H,
CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group. Cyclic
organosilicon compounds that may be used may include a ring
structure having three or more silicon atoms and the ring structure
may further comprise one or more oxygen atoms. Commercially
available cyclic organosilicon compounds include rings having
alternating silicon and oxygen atoms with one or two alkyl groups
bonded to the silicon atoms. For example, the cyclic organosilicon
compounds may include one or more of the following compounds:
[0014] hexamethylcyclotrisiloxane
(--Si(CH.sub.3).sub.2--O--).sub.3-- cyclic, [0015]
1,3,5,7-tetramethylcyclotetrasiloxane (TMCTS)
(--SiH(CH.sub.3)--O--).sub.4-- cyclic, [0016]
octamethylcyclotetrasiloxane (OMCTS)
(--Si(CH.sub.3).sub.2--O--).sub.4-- cyclic, and [0017]
1,3,5,7,9-pentamethylcyclopentasiloxane
(--SiH(CH.sub.3)--O--).sub.5-- cyclic.
[0018] The thin layer comprises silicon, carbon, and optionally
oxygen. In another embodiment, the precursor may be a silicon and
nitrogen-containing precursor that is used to deposit a thin
conformal layer comprising silicon, nitrogen, and optionally
carbon. The precursor may include linear silazanes and cyclic
silazanes. The linear and cyclic silazanes may include the
structure (R.sub.X--Si--NH--Si--R.sub.Y).sub.z, or the structure
(R.sub.X--Si--NH).sub.Y, wherein R.sub.X.dbd.CH.sub.3,
CH.sub.2CH.sub.3, or another alkyl group, and R.sub.Y.dbd.H,
CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group, x is from 0 to
4, y is 0 to 4, and x+y=4. The cyclic silazanes compounds may
include a ring structure having three or more silicon atoms and the
ring structure may further comprise one or more nitrogen atoms.
Commercially available cyclic silazanes compounds include rings
having alternating silicon and nitrogen atoms with one or two alkyl
groups bonded to the silicon atoms. For example, the cyclic
silazanes compounds may include following: [0019]
1,2,3,4,5,6,7,8-octamethylcyclotetrasilazane, [0020]
1,2,3,4,5,6-hexamethylcyclotrisilazane, [0021]
1,1,3,3,5,5-hexamethylcyclotrisilazane, and [0022]
1,1,3,3,5,5,7,7-octamethylcyclotetrasilazane.
[0023] The thin, conformal layer can be deposited on any blanket or
patterned film containing OH, NH, or NH.sub.2 groups at the
surface, including dielectric films and metallic films with oxide
at the surface (such as Cu/CuO or Al/Al.sub.2O.sub.3), as a
protective layer to prevent moisture adsorption and wet chemistry
etching, or a pore-sealing layer to prevent penetration of
precursors or chemicals. The thin layer can also serve as a
pore-sealing layer for porous dielectric films or metallic films
with OH, NH, or NH.sub.2 groups at the surface.
BRIEF DESCRIPTION OF THE DRAWINGS
[0024] So that the manner in which the above recited features of
the present invention can be understood in detail, a more
particular description of the invention, briefly summarized above,
may be had by reference to embodiments, some of which are
illustrated in the appended drawings. It is to be noted, however,
that the appended drawings illustrate only typical embodiments of
this invention and are therefore not to be considered limiting of
its scope, for the invention may admit to other equally effective
embodiments.
[0025] FIGS. 1A-1E depict schematic cross-sectional views of a
substrate structure at different stages of a process sequence
according to an embodiment of the invention.
[0026] FIG. 2 is a graph showing the dielectric constant (k) of low
dielectric constant films before and after ashing and of low
dielectric constant films having a thin OMCTS layer deposited
thereon after ashing according to an embodiment of the
invention.
[0027] FIG. 3 is a graph showing the wetting angle of low
dielectric constant films before and after ashing and of low
dielectric constant films having a thin OMCTS layer deposited
thereon after ashing according to an embodiment of the
invention.
[0028] FIG. 4A is a sketch of a trench profile (dense array) after
ashing and before wet cleaning according to the prior art. FIG. 4B
is a sketch of a trench profile (dense array) after ashing and wet
cleaning according to the prior art. FIG. 4C is a sketch of a
trench profile (dense array) after ashing and wet cleaning
according to an embodiment of the invention.
[0029] FIG. 5A is a sketch of a trench profile (iso structure/open
area) after ashing and before wet cleaning according to the prior
art. FIG. 5B is a sketch of a trench profile (iso structure/open
area) after ashing and wet cleaning according to the prior art.
FIG. 5C is a sketch of a trench profile (iso structure/open area)
after ashing and wet cleaning according to an embodiment of the
invention.
[0030] FIG. 6 is a graph showing the wetting angle of a thin OMCTS
layer according to an embodiment of the invention versus the length
of time of a helium plasma post-treatment of the layer.
DETAILED DESCRIPTION
[0031] Embodiments of the present invention provide a method of
depositing a thin, conformal layer comprising silicon, carbon, and
optionally oxygen and/or nitrogen on a patterned substrate. In one
aspect, embodiments of the present invention provide a method of
protecting a patterned low dielectric constant film after a
photoresist that has been used to pattern the low dielectric
constant film is removed from the film. In other aspects,
embodiments of the present invention provide a method of
controlling the critical dimension of a metal line in an
interconnect and a method of controlling the thickness of a
deposited layer to between about 4 .ANG. and about 100 .ANG..
[0032] In one embodiment, a low dielectric constant film on a
substrate is patterned using a photoresist and photolithography to
form a vertical interconnect or a horizontal interconnect opening
therein. The low dielectric constant film may be a film comprising
silicon, carbon, and optionally oxygen and/or nitrogen. The low
dielectric constant film may be deposited from a gas mixture
comprising an organosilicon compound, such as an organosilane or
organosiloxane. The gas mixture may also include an oxidizing gas.
In one embodiment, the gas mixture comprises an organosilicon
compound and a porogen, such as a hydrocarbon, that is removed from
the film after the film is deposited to create voids or pores in
the film and lower the dielectric constant of the film. The porogen
may be removed by a UV treatment, electron beam treatment, thermal
treatment, or a combination thereof. Methods of forming porous low
dielectric constant films are further described in commonly
assigned U.S. Pat. No. 6,936,551 and in commonly assigned U.S. Pat.
No. 7,060,330, which are herein incorporated by reference. It is
noted that low dielectric constant films that have other
compositions and/or are deposited from different gas mixtures can
be used in embodiments of the invention.
[0033] It is also noted that films other than low dielectric
constant films can be used in embodiments, such as any films
containing OH, NH, or NH.sub.2 groups at the surface. Generally,
the films that may be used have an oxygen-rich or nitrogen-rich
surface that allows the selective deposition of a thin film
comprising silicon, carbon, and optionally oxygen and/or nitrogen
thereon. As defined herein, an oxygen-rich surface has a Si:O
(silicon:oxygen) ratio of between about 1:1 to about 1:3. As
defined herein, a nitrogen-rich surface has a Si:N
(silicon:nitrogen) ratio of between about 1:1 to about 1:2.
[0034] While the films may be deposited on an oxygen-rich or
nitrogen-rich surface, the films typically do not grow on
carbon-rich surfaces, and thus, the deposition of the films on
oxygen-rich or nitrogen-rich surfaces may be described as selective
deposition processes.
[0035] FIG. 1A shows an example of low dielectric constant film 102
on a substrate 100. FIG. 1B shows a patterned photoresist 104 on
the low dielectric constant film 102.
[0036] The photoresist is then removed from the low dielectric
constant film by stripping or ashing, for example. FIG. 1C shows
the low dielectric constant film 102 after it has been patterned by
the photoresist 104 to form an interconnect 106 and the photoresist
has been removed. A thin, conformal layer 108, i.e., a layer having
a thickness of about 4 .ANG. and about 100 .ANG., comprising
silicon, carbon, and optionally oxygen and/or nitrogen, is then
deposited on a surface of the patterned low dielectric constant
film, as shown in FIG. 1D. The layer may be deposited by reacting a
gas mixture, such as a gas mixture comprising silicon, oxygen, and
carbon, in the presence of RF power. The silicon, oxygen, and
carbon may be provided by an organosilicon compound such as
octamethylcyclotetrasiloxane. The organosilicon compound is
typically introduced into a chamber with a carrier gas. Preferably,
the carrier gas is helium. However, other inert gases, such as
argon or nitrogen, may be used.
[0037] After the layer is deposited, the substrate may be wet
cleaned, such as with a 100:1 HF solution. Then, as shown in FIG.
1E, a layer 110, such as a PVD barrier layer or an ALD barrier
layer, e.g., an ALD tantalum nitride (TaN) layer, may be deposited
on the layer. Alternatively, as shown in FIG. 1F, a layer, such as
a barrier anti-reflective coating (BARC) layer 120, may be
deposited on the layer 108 and fill the interconnect 106.
[0038] The layer comprising silicon, carbon, and optionally oxygen
and/or nitrogen may be deposited in a chemical vapor deposition
chamber or a plasma enhanced chemical vapor deposition chamber by
reacting a gas mixture comprising an organosilicon compound in the
presence of RF power. Examples of chambers that may be used to
deposit the layer include a PRODUCER.RTM. chamber with two isolated
processing regions and a DxZ.RTM. chamber, both of which are
available from Applied Materials, Inc. of Santa Clara, Calif. The
processing conditions provided herein are provided for a 300 mm
PRODUCER.RTM. chamber with two isolated processing regions. Thus,
the flow rates experienced per each substrate processing region and
substrate are half of the flow rates into the chamber.
[0039] During deposition of the layer on a substrate in the
chamber, the substrate is typically maintained at a temperature
between about 150.degree. C. and about 400.degree. C. RF power is
provided at a power level of about 100 W or less, such as between
about 30 W and about 75 W, for a 300 mm substrate. Generally, the
RF power may be provided at about 0.109 W/cm.sup.2 or less, such as
between about 0.033 W/cm.sup.2 and about 0.082 W/cm.sup.2. The RF
power may be provided to a showerhead, i.e., a gas distribution
assembly, and/or a substrate support of the chamber. The RF power
is provided at a high frequency between about 13 MHz and 14 MHz,
preferably about 13.56 MHz. The RF power may be cycled or pulsed.
The RF power may also be continuous or discontinuous. The spacing
between the showerhead and the substrate support is greater than
about 200 mils, such as between about 200 mils and about 1400 mils.
The pressure in the chamber is about 1.5 Torr or greater, such as
between about 1.5 Torr and about 8 Torr.
[0040] The organosilicon compound may be introduced into the
chamber at a flow rate of between about 100 sccm and about 1000
sccm. A carrier gas may be introduced into the chamber at a flow
rate of between about 100 sccm and about 7,000 sccm. The ratio of
the flow rate of the organosilicon compound, e.g.,
octamethylcyclotetrasiloxane (OMCTS, sccm), to the flow rate of the
carrier gas, e.g., helium (sccm), into the chamber is about 0.1 or
greater. The layer may be deposited for a period of time, such as
between about 0.1 seconds and about 600 seconds depending on the
aspect ratio of patterned structure, to deposit the layer to a
thickness between about 4 .ANG. and about 100 .ANG.. Typically, the
layer is deposited for a longer period of time when higher aspect
ratios are used in order to provide a conformal surface.
[0041] It has been found that using the RF power levels, spacing,
pressure, and flow rate ratios described above, a thin, uniform,
conformal layer having a thickness of only between about 4 .ANG.
and about 100 .ANG. can be reliably deposited when a
self-saturating organosilicon compound is used as a precursor to
deposit the layer. A 1 .ANG. thickness range of the layer within a
single 300 mm substrate has been obtained using the conditions
provided herein. As defined herein, a "self-saturating precursor"
is a precursor that deposits one thin layer, e.g., only one
molecular layer of the precursor disregarding the length of
deposition time, on a substrate. The thickness can be controlled by
the choice of the precursor, as different precursors have different
molecular sizes, resulting in different thicknesses for one
molecular layer for different precursors. The presence of the thin
layer hinders the further deposition of additional layers from the
precursor under the processing conditions used to deposit the thin
layer. Generally, the self-saturating precursor may comprise a
methyl group that is selected to suppress continued growth of the
thin layer. OMCTS is a preferred self-saturating precursor as it
contains a large number of methyl groups that result in a
self-saturating deposition of a layer, as the carbon in the methyl
groups provides a carbon-rich film surface that substantially
hinders further deposition thereon. In other words, a conformal
first layer may be deposited from OMCTS because as soon as the
surface of the underlying substrate is covered with OMCTS
molecules, the presence of the Si--CH.sub.3 bonds at the surface of
the deposited layer provides a carbon-rich surface that hinders
further deposition until some of the methyl groups are removed by
some other treatment of the layer. Thus, the deposition of each
molecular layer of OMCTS can be well controlled, which enhances the
step coverage of the final layer.
[0042] Other than octamethylcyclotetrasiloxane, precursors having
the general formula R.sub.x--Si--(OR').sub.y, wherein R.dbd.H,
CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group,
R'.dbd.CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group, x is
from 0 to 4, y is 0 to 4, and x+y=4, may also be used to deposit a
thin conformal layer with a suitable process window. Other
precursors that may be used include linear organosiloxanes and
cyclic organosiloxane. The linear and cyclic organosiloxanes may
include the structure (R.sub.X--Si--O--Si--R.sub.Y).sub.z, or the
structure (R.sub.X--Si--O).sub.Y, wherein R.sub.X.dbd.CH.sub.3,
CH.sub.2CH.sub.3, or another alkyl group, and R.sub.Y.dbd.H,
CH.sub.3, CH.sub.2CH.sub.3, or another alkyl group. Examples of
precursors that may be used include diethoxymethylsilane (DEMS),
hexamethyldisiloxane (HMDOS), and hexamethyldisilane (HMDS). Other
precursors containing Si, C, and H may be used in the process, such
as trimethylsilane, tetramethylsilane, etc.
[0043] X-ray photoelectron spectroscopy (XPS) analysis was
performed on low dielectric constant films that had not been
exposed to an ashing process and on low dielectric constant films
that had been exposed to a photoresist ashing. XPS analysis was
also performed on low dielectric constant films that were exposed
to photoresist ashing and then treated by depositing a thin layer
thereon, with the thin layer being deposited from OMCTS and
comprising silicon, carbon, and oxygen according to embodiments of
the invention. The XPS analysis showed that depositing the thin
layer on the ashed low dielectric constant films provide a higher
carbon content (atomic % carbon) at the surface of those films
compared to the low dielectric constant films that were not treated
by depositing the thin layer thereon. For example, the ashed low
dielectric constant films may have about 3 atomic % carbon, while
the thin layer on the ashed low dielectric constant films provides
about 15 atomic % carbon at the surface. Thus, in one aspect, the
thin layer is a carbon-rich layer. The thin layer may have a carbon
content of between about 5 atomic % and about 30 atomic %. Ashing
depletes the carbon concentration at the surface of low dielectric
constant film, while depositing the thin layer on the ashed low
dielectric constant film recovers the surface carbon
concentration.
[0044] The XPS analysis also showed that the oxygen content at the
surface of the low dielectric constant films treated with the thin
layer was lower than the oxygen content at the surface of the low
dielectric constant films that were not treated with the thin layer
after ashing, as OH groups at the surface of the ashed films was
replaced by the thin layer that comprises carbon. The replacement
of the OH groups at the surface of the ashed films with the thin
layer that comprises carbon also lowers the dielectric constant of
the ashed films. FIG. 2 shows that depositing a thin layer using
OMCTS on the low dielectric constant films lowered the post-ashing
dielectric constant of films subjected to one of three different
ashing processes.
[0045] The wetting angle for low dielectric constant films pre- and
post-ashing (ELK ILD, i.e., extreme low k interlayer dielectric,
and Ashed ELK ILD, respectively, in FIG. 3), and for low dielectric
constant films post-ashing and having a thin OMCTS layer (Ashed ELK
ILD with OMCTS deposition in FIG. 3) thereon was also measured. The
results are shown in FIG. 3. As shown in FIG. 3, depositing the
thin OMCTS layer on the low dielectric constant films post-ashing
increased the wetting angle of the low dielectric constant films.
The increased wetting angle shows that the thin OMCTS layer
increased the hydrophobicity of the low dielectric constant films'
surfaces. Such an increase in hydrophobicity is desirable, as a
hydrophobic surface prevents moisture adsorption into the low
dielectric constant films which can affect film performance or at
least result in a need for time consuming steps to remove the
moisture.
[0046] The effect of the deposition of the thin, conformal OMCTS
layer on the profile of interconnects after a post ash wet clean
was also examined. The trench profiles of regions having high
densities of trenches and low densities of trenches in low
dielectric constant films with and without the thin OMCTS layer
thereon were examined after the films were dipped in a 100:1 HF
solution in a wet clean process.
[0047] FIGS. 4A-4C show the trench profiles of regions having a
high density of trenches. FIG. 4A shows the trench profile after
ashing and before the wet clean. FIGS. 4B and 4C shows the trench
profile after ashing and after the wet clean for low dielectric
constant films without and with the thin OMCTS layer thereon,
respectively. FIG. 4B shows that the wet clean causes a critical
dimension loss of about 30 nm for trenches in a low dielectric
constant film without the thin OMCTS layer thereon. FIG. 4C shows
that such a CD loss was not observed when the low dielectric
constant film had the thin OMCTS layer deposited thereon before the
wet clean.
[0048] FIGS. 5A-5C show the trench profiles of regions having a low
density of trenches. FIG. 5A shows the trench profile after ashing
and before the wet clean. FIGS. 5B and 5C shows the trench profile
after ashing and after the wet clean for low dielectric constant
films without and with the thin OMCTS layer thereon, respectively.
FIG. 5B shows that the wet clean causes an undercut of greater than
about 30 nm for trenches in a low dielectric constant film without
the thin OMCTS layer thereon. FIG. 5C shows that such undercutting
was not observed when the low dielectric constant film had the thin
OMCTS layer deposited thereon before the wet clean.
[0049] Thus, the thin OMCTS layer provides a carbon rich surface,
which in turn provides a hydrophobic surface that prevents the
critical dimension loss and undercutting of the low k films during
wet etch processes.
[0050] It was also found that the thin layers provided according to
embodiments of the invention act as dense, pore-sealing layers that
can prevent the penetration of a material, such as a BARC material
for a subsequently deposited BARC layer, or a PVD barrier precursor
or an ALD barrier precursor, e.g., an ALD TaN precursor, for a
subsequently deposited barrier layer, into porous low k films onto
which the thin layers may be deposited.
[0051] For example, the thin layer may be deposited on a low
dielectric constant film after a via etch and photoresist ashing in
a via first damascene process. Subsequent BARC filling may be
performed on the thin layer. The thin layer provides a pore-sealing
layer that prevents the penetration of the BARC material into the
dielectric film. A dielectric barrier that is between the low
dielectric constant film and an underlying conductive material,
such as copper, may then be etched to expose the underlying
conductive material after a trench etch and photoresist removal.
After the dielectric barrier etching, a reducing chemistry may be
used to clean the conductive surface exposed by the removal of the
dielectric barrier and to remove an oxide from the surface, such as
copper oxide (CuO). The thin layer is then deposited on the
sidewalls of the via and trench. The thin layer provides a
pore-sealing layer that prevents the penetration of subsequent
barrier layer precursors into the low dielectric constant film.
[0052] In embodiments in which a BARC layer is deposited on the
thin layer after wet cleaning the substrate, the thin layer may be
helium (or other inert gas) plasma post-treated to adjust the
carbon concentration at the surface of the thin layer and the
wetting angle of the thin layer. The wetting angle may be decreased
to about 70.degree. C. or less to enhance the wetting and
deposition of the BARC layer. FIG. 6 shows that the wetting angle
decreases with increasing plasma treatment time. Mild processing
conditions, i.e., an RF power of between about 30 W and about 100 W
and a He flow rate of between about 100 sccm and about 10,000 sccm,
are used such that the plasma treatment does not damage the
pore-sealing nature of the thin layer.
[0053] The thin layer may also be helium plasma post-treated before
the deposition of layers other than BARC layers thereon, such as
ALD barrier layers, if the surface wetting or contact angle needs
to be adjusted. The thin layer may be plasma post-treated with
different gases, such as O.sub.2, CO.sub.2, N.sub.2O, NH.sub.3,
H.sub.2, helium, nitrogen, argon, or combinations thereof. The
plasma post-treatment can modify the surface nature and
characteristics of the layer, such as the surface tension and
surface contact angle.
[0054] In another embodiment, a method of controlling the critical
dimension of a metal line in an interconnect is provided. The
method includes depositing a thin layer on a patterned low
dielectric constant film, as described in embodiments above. The
patterned low dielectric constant film may comprise an oxygen-rich
or nitrogen-rich surface before the deposition of the thin layer
thereon. After the layer is deposited, the flow of the precursor
used to deposit the layer, such as OMCTS, is then terminated, and
any remaining precursor is purged from the chamber by introducing
carrier gas only, such as He carrier gas, into the chamber. The
chamber may be purged or pumped or purged and pumped.
[0055] After the chamber is purged and/or pumped, in one
embodiment, an oxygen plasma treatment is performed in the chamber
to treat the layer that is deposited on the substrate from the
precursor and initiate next cycle of deposition, such as an OMCTS
deposition. In another embodiment, an NH.sub.3 plasma treatment
with or without the addition of H.sub.2 can be used if a
nitrogen-doped oxide or SiN layer is desired. The oxygen plasma may
be provided by any oxygen-containing gas capable of generating
oxygen radicals that oxidize the surface of the layer. For example,
the gas may include O.sub.2, CO.sub.2, N.sub.2O, or a combination
thereof. The oxygen-containing gas may be introduced into the
chamber at a flow rate. The oxygen-containing gas may be flowed
into the chamber for a period of time such as between about 0.1
seconds and about 60 seconds depending on the via/trench pattern
profile. The oxygen plasma may be provided by applying a RF power
of between about 50 W and about 1000 W in the chamber at a
frequency of 13.56 MHz. Mixed frequency RF power can be used. To
minimize the impact or damage of plasma treatment on the underlying
layer (such as a low dielectric constant film), a low level of high
frequency RF power is preferred, such as between about 30 W and
about 100 W, which corresponds to between about 0.033 W/cm.sup.2
and about 0.082 W/cm.sup.2.
[0056] The plasma treatment may be terminated by terminating the
flow of the oxygen-containing gas into the chamber. Optionally, the
thickness of the deposited layer is then measured. The flow of the
precursor into the chamber is then resumed to deposit an additional
amount of the thin layer. The chamber is purged and then an oxygen
plasma treatment as described above is also done. Multiple cycles
of deposition, purging, and plasma treatment may be performed until
the desired thickness of layer is obtained. By controlling the
thickness of the layer deposited in the interconnect, the thickness
of a subsequently deposited metal line in the interconnect may be
controlled.
[0057] In another embodiment, a method of controlling the thickness
of a layer to between about 4 .ANG. and about 100 .ANG. on a
substrate is provided. The substrate, which may comprise an
oxygen-rich or nitrogen-rich surface, is exposed to a
silicon-containing precursor in the presence of a plasma to deposit
a layer on the substrate, and then the layer is treated with a
plasma from NH.sub.3 with or without H.sub.2 or with a plasma from
an oxygen-containing gas selected from the group consisting of
O.sub.2, CO.sub.2, and N.sub.2O. The exposure of the substrate to
the silicon-containing precursor to deposit a layer and the
treatment of the layer with a plasma are repeated until a desired
thickness of the layer is obtained.
[0058] In a further embodiment, a method of producing a dense
dielectric spacer comprising either an oxide or a nitride is
provided. The method comprises exposing a patterned substrate
comprising a gate, which may comprise an oxygen-rich or
nitrogen-rich surface, to a silicon-containing precursor in the
presence of a plasma to deposit a layer on the gate and then
treating the layer with a plasma from an oxygen-containing gas or
nitrogen-containing gas selected from the group consisting of
O.sub.2, CO.sub.2, N.sub.2O, a nitrogen-containing gas, and
NH.sub.3 with or without H.sub.2. The silicon-containing precursors
and the plasma treatments provided above with respect to the method
of controlling the critical dimension of a metal line in an
interconnect may also be used for both the method of producing a
dense dielectric spacer and the method of controlling the thickness
of a layer to between about 4 .ANG. and about 100 .ANG..
[0059] While the foregoing is directed to embodiments of the
present invention, other and further embodiments of the invention
may be devised without departing from the basic scope thereof, and
the scope thereof is determined by the claims that follow.
* * * * *