U.S. patent application number 12/658670 was filed with the patent office on 2011-04-28 for transition metal alloys for use as a gate electrode and devices incorporating these alloys.
Invention is credited to Nathan Baxter, Robert S. Chau, Mark Doczy, Kari Harkonen, Teemu Lang.
Application Number | 20110097858 12/658670 |
Document ID | / |
Family ID | 34136454 |
Filed Date | 2011-04-28 |
United States Patent
Application |
20110097858 |
Kind Code |
A1 |
Doczy; Mark ; et
al. |
April 28, 2011 |
Transition metal alloys for use as a gate electrode and devices
incorporating these alloys
Abstract
Embodiments of a transition metal alloy having an n-type or
p-type work function that does not significantly shift at elevated
temperature. The disclosed transition metal alloys may be used as,
or form a part of, the gate electrode in a transistor. Methods of
forming a gate electrode using these transition metal alloys are
also disclosed.
Inventors: |
Doczy; Mark; (Beaverton,
OR) ; Baxter; Nathan; (Portland, OR) ; Chau;
Robert S.; (Beaverton, OR) ; Harkonen; Kari;
(Kauniainen, FI) ; Lang; Teemu; (Helsinki,
FI) |
Family ID: |
34136454 |
Appl. No.: |
12/658670 |
Filed: |
February 12, 2010 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11640034 |
Dec 15, 2006 |
|
|
|
12658670 |
|
|
|
|
11202541 |
Aug 11, 2005 |
7193253 |
|
|
11640034 |
|
|
|
|
10641848 |
Aug 15, 2003 |
7030430 |
|
|
11202541 |
|
|
|
|
Current U.S.
Class: |
438/197 ;
257/E21.409 |
Current CPC
Class: |
H01L 29/6659 20130101;
H01L 21/823842 20130101; C22C 32/0084 20130101; H01L 21/82345
20130101; C22C 14/00 20130101; H01L 21/28088 20130101; C04B
2235/723 20130101; C22C 27/00 20130101; C04B 2235/722 20130101;
H01L 29/4966 20130101; C04B 35/5618 20130101; C22C 16/00
20130101 |
Class at
Publication: |
438/197 ;
257/E21.409 |
International
Class: |
H01L 21/336 20060101
H01L021/336 |
Claims
1. A method comprising: depositing a layer of insulating material
on a substrate; and depositing a layer of an alloy over the
insulating layer, the alloy including approximately 20 to 50 atomic
percent of a transition metal, the transition metal selected from a
group consisting of titanium, zirconium, tantalum, and hafnium,
approximately 30 to 60 atomic percent of carbon, and up to
approximately 20 atomic percent of aluminum.
2. The method of claim 1, wherein the alloy has a work function in
a range of approximately 3.8 eV to 4.4 eV.
3. The method of claim 2, further comprising subjecting the alloy
layer to a temperature up to approximately 900.degree. Celsius,
wherein the work function does not significantly shift at the
temperature.
4. The device of claim 1, further comprising subjecting the alloy
layer to a temperature up to approximately 900.degree. Celsius,
wherein the alloy does not react substantially with the insulating
material.
5. The method of claim 1, wherein the alloy layer is deposited by
one of chemical vapor deposition, physical vapor deposition, and
atomic layer deposition.
6. The method of claim 1, further comprising adjusting the
composition of the alloy during deposition of the alloy layer.
7. The method of claim 1, wherein the alloy further includes up to
5 atomic percent of a residual material, the residual material
comprising at least one of oxygen, nitrogen, and chloride.
8. The method of claim 1, further comprising depositing a layer of
a conductive material over the alloy layer.
9. The method of claim 8, wherein the conductive material comprises
one of aluminum and poly-silicon.
10. The method of claim 8, wherein the thickness of the allow layer
is between approximately 50 and 100 Angstroms, and the thickness of
the conductive material layer is between approximately 500 and 2000
Angstroms.
11. The method of claim 8, wherein the alloy layer functions as at
least one of a barrier layer and an etch stop.
12. The method of claim 1, further comprising etching the layer of
alloy and the layer of insulating material to form a gate electrode
stack.
13. The method of claim 12, further comprising forming source and
drain regions in the substrate.
Description
CLAIM OF PRIORITY
[0001] This application is a divisional of U.S. patent application
Ser. No. 11/640,034, filed Dec. 15, 2006, now abandoned, which is a
divisional of U.S. patent application Ser. No. 11/202,541, filed
Aug. 11, 2005, now U.S. Pat. No. 7,193,253, which is a divisional
of U.S. patent application Ser. No. 10/641,848, filed Aug. 15,
2003, now U.S. Pat. No. 7,030,430.
FIELD OF THE INVENTION
[0002] The invention relates generally to integrated circuit
devices and, more particularly, to metal alloys that can be used as
a gate electrode in a transistor.
BACKGROUND OF THE INVENTION
[0003] Illustrated in FIG. 1 is a conventional MOSFET (Metal Oxide
Semiconductor Field Effect Transistor) 100. For an n-type MOSFET,
the transistor 100 includes n-type source and drain regions 120a,
120b, respectively, formed in a p-doped substrate 110. A gate
electrode 130 is disposed between the source and drain regions
120a, 120b, this gate electrode being separated from the substrate
110 and the source and drain regions 120a-b by a gate insulating
layer 140. Insulating layers 150a, 150b further isolate the gate
electrode 130 from the surrounding structures. Conductors 160a,
160b (e.g., conductive traces) may be electrically coupled with the
source and drain regions 120a, 120b, respectively. If a sufficient
voltage is applied to the gate electrode--i.e., the "threshold
voltage"--electrons will flow from the source to the drain, these
mobile electrons concentrated in a thin "inversion layer" 170
extending between the source and drain regions 120a, 120b. Of
course, those of ordinary skill in the art will recognize that the
complementary MOSFET--i.e., the p-type MOSFET--will have a similar
structure (p-type source and drain regions on an n-type substrate),
and that a CMOS (Complementary Metal Oxide Semiconductor)
integrated circuit will utilize both n-type and p-type MOSFETs (or,
more generally, both NMOS and PMOS devices).
[0004] In conventional MOSFET devices, the gate electrode 130
typically comprises a poly-silicon material, whereas the gate
insulating layer comprises Silicon Dioxide (SiO.sub.2). To increase
circuit density and improve device performance, it may be desirable
to scale down the thickness of the gate insulating layer 140 (often
referred to as the "gate oxide"). As the thickness of the gate
oxide is scaled down, it may be necessary to use a material having
a higher dielectric constant--i.e., a "high-k dielectric"--as the
gate oxide in order to maintain sufficient capacitance while also
preventing failure by electron tunneling. Integration of a
poly-silicon gate electrode onto a high-k gate oxide has, however,
proven difficult due to interactions between the poly-silicon gate
material and the high-k insulating material. Furthermore, as the
thickness of the gate insulating layer 140 is further scaled down
(e.g., below about 20 Angstroms), it may be desirable to use an
alternative material to poly-silicon as the gate electrode, in
order to eliminate the thickness contribution of poly depletion to
the gate oxide (i.e., to eliminate that portion of a poly-silicon
gate electrode that becomes depleted of free charges and, hence,
adds to the effective thickness of the gate insulating layer).
[0005] Use of a metal gate electrode can eliminate the
above-described effects of depleted poly-silicon in the gate
electrode, and a metal gate electrode may also enable further
scaling down of the gate oxide thickness. However, use of metal
materials as the gate electrode in NMOS and PMOS devices has also
proven difficult. To optimize the performance of a transistor, the
metal used at the gate electrode should be selected to provide a
work function that will achieve a sufficiently low (but non-zero)
threshold voltage for the transistor (e.g., 0.2 V to 0.3 V). Many
metals have a suitable work function (a value representing an
energy level of the most energetic electrons within the metal), but
are thermally unstable at high temperature. Process flows for
transistors can often reach temperatures of 900.degree. C. and,
therefore, during subsequent processes, the work function of these
metals may shift to unsuitable values. Furthermore, at elevated
temperatures, these metal gate materials may react with the gate
insulating layer, thereby degrading its insulating properties.
Other metals are thermally stable at the temperatures present in
transistor process flows; however, these metals have work functions
that are inadequate for high performance transistors.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] FIG. 1 is a schematic diagram illustrating an embodiment of
a conventional MOSFET transistor.
[0007] FIG. 2 is a schematic diagram illustrating the composition
of one embodiment of a transition metal alloy that can be used as a
gate electrode.
[0008] FIG. 3 is a block diagram illustrating one embodiment of a
method of forming a gate electrode including a transition metal
alloy.
[0009] FIGS. 4A-4E are schematic diagrams illustrating further
embodiments of the method of forming a gate electrode, as shown in
FIG. 3.
[0010] FIG. 5 is a block diagram illustrating another embodiment of
a method of forming a gate electrode including a transition metal
alloy.
[0011] FIGS. 6A-6E are schematic diagrams illustrating further
embodiments of the method of forming a gate electrode, as shown in
FIG. 5.
DETAILED DESCRIPTION OF THE INVENTION
[0012] For high performance transistors, it is desirable to have a
low, but non-zero, threshold voltage--i.e., the gate voltage at
which electrons (or, in the case of p-type devices, holes) begin to
flow between the source and drain regions. Many factors can affect
the threshold voltage of a transistor, and one of these factors is
the work function of the gate electrode material. However, for many
transistor designs, most of the other factors that can affect the
threshold voltage are "locked in" by design constraints, such that
the primary factor determining the transistor's threshold voltage
is the work function of the gate electrode. Thus, selection of the
gate electrode work function plays a significant role in setting
the threshold voltage of an optimized, high performance (e.g., high
switching speed, high drive current, etc.) MOSFET device. As noted
above, the "work function" of a material, such as a metal,
describes the energy level of the most energetic electrons within
the material.
[0013] The performance of a transistor is governed not so much by
the absolute value of the gate electrode material's work function,
but rather by the relationship between the gate electrode's work
function and the work function of the underlying substrate material
(e.g., Silicon, Gallium Arsenide, Silicon on insulator, etc.).
Negative channel semiconductor devices--i.e., those that rely on
the movement of electrons during activation--require an "n-type
work function", whereas positive channel semiconductor
devices--i.e. those that rely on the movement of electron
vacancies, or holes, during activation--require a "p-type work
function." A gate electrode has an n-type work function if the
electrode material's work function is near (e.g., within +/-0.3 eV)
of the energy level of the underlying substrate material's
conduction band. Conversely, a gate electrode has a p-type work
function if the electrode material's work function is near the
energy level of the substrate material's valence band. Generally,
for semiconductor materials, the valence band is the highest range
of electron energies where electrons are normally present, whereas
the conduction band is a range of electron energies above the
valence band where electrons are free to accelerate under
application of an electric field (and, thus, create an electric
current).
[0014] By way of example, Silicon has a conduction band energy
level of approximately 4.1 eV and a valence band energy level of
approximately 5.2 eV. Thus, for a Silicon substrate, the gate
electrode of a negative channel MOSFET (or NMOS) device would have
an n-type work function of approximately 4.1 eV (+/-0.3 eV), and
the gate electrode of a positive channel MOSFET (or PMOS) device
would have a p-type work function of approximately 5.2 eV (+/-0.3
eV). Note that energy levels falling between the valence band and
conduction band energy levels are often referred to as "midgap"
energies (e.g., returning to the above example using Silicon, the
midgap energies are those falling between approximately 4.4 and 4.9
eV). Also, the difference between the work function of the gate
electrode material and the work function of the substrate material
is often referred to as the "flatband" energy. However, because the
work function of the substrate is often fixed--i.e., it is one of
the "locked in" features of a transistor design--the term
"flatband" is sometimes used in a manner that is synonymous with
the term "work function."
[0015] Disclosed herein are embodiments of a transition metal alloy
having either an n-type work function or a p-type work function.
The disclosed transition metal alloys may be used as the gate
electrode in CMOS integrated circuits, and embodiments of methods
of forming a gate electrode using such a transition metal alloy are
also disclosed below. In one embodiment, the transition metal
alloys are thermally stable at elevated temperatures (e.g., greater
than 900.degree. C.)--i.e., their work function does not shift or,
in other words, there is no appreciable flat band shift. In another
embodiment, the transition metal alloys, when used as a gate
electrode, do not react with the underlying gate insulating layer
at elevated temperature. In a further embodiment, a transition
metal alloy comprises a transition metal carbide.
[0016] Turning now to FIG. 2, illustrated is one embodiment of a
transition metal alloy 200 that may have an n-type work
function--or, alternatively, a p-type work function--that is
suitable for use as a gate electrode in MOSFET devices (e.g.,
either n-type or p-type). The transition metal alloy 200 includes a
transition metal 210, Carbon (C) 220 (or other suitable element, as
described below), and a dopant 230.
[0017] Generally, the transition metal 210 is selected to provide
either an n-type work function or a p-type work function. In one
embodiment, the transition metal comprises one of Titanium (Ti),
Tantalum (Ta), Zirconium (Zr), and Hafnium (Hf). In another
embodiment, the transition metal may comprise any one of the
aforementioned elements or one of Chromium (Cr), Molybdenum (Mo),
Tungsten (W), Vanadium (V), and Niobium (Nb). In one embodiment,
the transition metal 210 comprises between 20 and 50 atomic percent
("at %") of the alloy 200. In yet another embodiment, the alloy 200
includes two or more of the above-listed transition metals.
[0018] Carbon 220 may be another component of the transition metal
alloy 200. In one embodiment, carbon comprises between 30 and 60 at
% of the alloy 200. Some transition metals, when alloyed with
Carbon to form a transition metal carbide (or other alloy including
Carbon with no or minimal amounts of carbide), will be
characterized by good thermal stability--i.e., at least some
material properties of these transition metal carbides, such as the
work function, do not degrade at elevated temperature. It should be
understood, however, that Carbon is but one example of an element
that can be alloyed with a transition metal to improve thermal
characteristics. In addition to Carbon, other elements that may be
alloyed with a transition metal to achieve enhanced thermal
properties include, for example, Nitrogen (N), Silicon (Si),
Germanium (Ge), and Boron (B), or various combinations of these
elements.
[0019] The transition metal alloy 200 also includes a dopant 230,
as noted above. The dopant is introduced into the transition metal
alloy to adjust or alter certain characteristics of the alloy. In
one embodiment, the dopant is added to affect the work function of
the alloy (e.g., to enhance either the n-type or p-type work
function characteristics). In another embodiment, the dopant is
added to alter another property of the transition metal alloy, such
as conductivity. The dopant may, in one embodiment, comprise up to
20 at % of the transition metal alloy. In one embodiment, the
dopant comprises Aluminum (Al)--e.g., for an n-type work
function--and in another embodiment, the dopant comprises Platinum
(Pt)--e.g., for a p-type work function. Other possible dopants
include Silicon, Gallium (Ga), and Germanium (Ge), as well as many
of the transition metals.
[0020] As shown in FIG. 2, the transition metal alloy 200 may
further include other, residual substances 240. The residual
elements are typically present in the transition metal alloy in
relatively small amounts as a result of the deposition process or
other fabrication steps. In one embodiment, the residual substances
comprise approximately 5 at % or less of the alloy's composition.
Common residual substances include Nitrogen and Oxygen, as well as
halide impurities such as chloride. It should, however, be
understood, as suggested above, that elements such as Nitrogen may
be a desired component of the alloy 200 in other embodiments.
[0021] In one embodiment, the transition metal alloy 200 is a
transition metal carbide comprising between 20 and 50 at %
Titanium, between 30 and 60 at % Carbon, and up to 20 at %
Aluminum. Such a transition metal carbide may provide an n-type
work function. In a further embodiment, the transition metal
carbide comprises approximately 37 at % Titanium, approximately 55
at % Carbon, and approximately 4 at % Aluminum, as well as
approximately 4 at % Oxygen and 1 at % or less Nitrogen. Of course,
those of ordinary skill in the art will appreciate that these are
but a few examples of transition metal alloys that may be useful as
a gate electrode material.
[0022] In one embodiment, the transition metal alloy 200 is
thermally stable at elevated temperature, and in a further
embodiment, the transition metal alloy 200 has a thermally stable
work function at elevated temperature, as measured on a high-k
dielectric gate material, such as ZrO.sub.2, HfO.sub.2, or
Al.sub.2O.sub.3. In one embodiment, the transition metal alloy 200
exhibits thermal stability--e.g., the alloy's work function does
not shift to a midgap energy level--at a temperature up to
approximately 900.degree. C. When the transition metal alloy 200 is
used as a gate electrode in a MOSFET device, this enhanced thermal
stability enables the gate electrode structure to withstand the
high post-deposition processing temperatures that such device are
routinely subjected to in the manufacturing process. Thus, the work
function--and ultimately the performance--of the transistor will
not degrade during later stages of a CMOS process flow.
[0023] Illustrated in FIG. 3 is an embodiment of a method 300 of
forming a gate electrode using a transition metal alloy. Also, the
method 300 of FIG. 3 is further illustrated by the schematic
diagrams of FIGS. 4A through 4E, and reference should be made to
these figures, as called out in the text.
[0024] Referring to block 310 in FIG. 3, a gate insulating layer is
deposited on a substrate. This is illustrated in FIG. 4A, which
shows a layer of gate insulating material 401 that has been
deposited on a surface of a substrate 410. In one embodiment, the
gate insulating material 401 comprises a high-k dielectric
material, such as ZrO.sub.2, HfO.sub.2, or Al.sub.2O.sub.3. The
layer of gate insulating material 401 may be deposited using any
suitable technique. In one embodiment, the gate insulating material
layer 401 is formed by a blanket layer deposition process, such as
chemical vapor deposition (CVD) or physical vapor deposition
(PVD).
[0025] In one embodiment, where an NMOS device is to be formed, the
substrate 410 comprises a p-doped substrate (and the source and
drain regions that are formed will comprise n-doped regions). In
another embodiment, where a PMOS device is to be formed, the
substrate 410 comprises an n-doped substrate (and the source and
drain regions that are formed will comprise p-doped regions). In
one embodiment, the substrate 410 comprises Silicon (e.g., single
crystal Silicon); however, it should be understood that, in other
embodiments, the substrate 410 may comprise another material (e.g.,
GaAs, Silicon-on-insulator, etc.).
[0026] A layer of a transition metal alloy is then deposited over
the gate insulating material layer, as set forth in block 320. This
is illustrated in FIG. 4B, where a layer of a transition metal
alloy 402 has been formed over the layer of gate insulating
material 401. In one embodiment, the layer of transition metal
alloy 402 comprises the transition metal alloy 200 shown and
described above with respect to FIG. 2. In a further embodiment,
the layer of transition metal alloy 402 comprises a transition
metal carbide and, in yet another embodiment, the transition metal
alloy layer 402 comprises a transition metal carbide including
Titanium, Carbon, and Aluminum. In yet a further embodiment, the
transition metal alloy layer 402 comprises approximately 37 at %
Titanium, 55 at % Carbon, and 3 at % Aluminum, as well as
approximately 4 at % Oxygen and 1 at % or less Nitrogen.
[0027] The transition metal alloy layer may be deposited using any
suitable technique. In one embodiment, the transition metal alloy
layer 402 is deposited to a thickness of between approximately 500
and 2,000 Angstroms. The transition metal alloy layer 402 is, in
one embodiment, deposited by a PVD process using a TiC(Al) target
in a single chamber having an Argon gas environment (introduced at
a flow rate of 10-200 sccm) at a pressure of between 1 and 100
mTorr with the substrate maintained at a temperature between
0.degree. and 450.degree. C. Deposition may take place using a DC
power source in the 1 kW to 40 kW range and, in another embodiment,
pulsed DC power at a frequency of between 1 and 100 kHz is
utilized. In a further embodiment, the gate electrode material is
deposited by a PVD process using a Ti(Al) target in a first chamber
and a Carbon target in a second chamber, wherein the gate electrode
is formed by alternating thin (e.g., 5-10 Angstroms) layers of
Ti(Al) and Carbon (e.g., 5-10 Angstrom layers to a final thickness
of between 500 and 2,000 Angstroms).
[0028] In another embodiment, the transition metal alloy layer 402
is deposited by a CVD process. Deposition is performed in a CVD
chamber at a pressure of 0.25 to 10.0 Torr and at a temperature of
between 150.degree. C. and 600.degree. C. Precursors introduced
into the deposition chamber include TiCl.sub.4 vapor at a flow rate
of 10 to 1000 sccm, TMA (trimethylaluminum, or Al(CH.sub.3).sub.3)
vapor at a flow rate of 10 to 1000 sccm, as well as ammonia
(NH.sub.3) at a flow rate up to 10 sccm. Also, an inert gas, such
as N.sub.2, Argon (Ar), or Helium (He), may be introduced into the
deposition chamber (e.g., as either a transport agent or as a
purging agent) at a flow rate up to 4,000 sccm.
[0029] In yet a further embodiment, the transition metal alloy
layer 402 is deposited using an atomic layer deposition (ALD)
process.
[0030] Returning to FIG. 3, etching is then performed to create the
gate electrode stack, as set forth at block 330. This is
illustrated in FIG. 4C, where both the transition metal alloy layer
402 and gate insulating material layer 401 have been etched to
create a gate electrode stack 405. The gate electrode stack
includes a gate electrode 430 overlying a gate insulating layer
440. Any suitable etching processes may be employed to etch the
transition metal alloy layer 402 and gate insulating material layer
401 in order to form the gate electrode 430 and gate insulating
layer 440. Although not shown in the figures, it will be
appreciated by those of ordinary skill in the art that a patterned
mask layer may be formed prior to etching.
[0031] Referring to block 340, additional features may then be
formed on the substrate to create an NMOS or PMOS device. This is
illustrated in FIGS. 4D and 4E, where additional features have been
created to form a device 400. Referring first to FIG. 4D, source
and drain regions 420a, 420b are created. For an NMOS transistor,
the source and drain regions 420a, 420b will comprise n-type
regions, whereas for a PMOS transistor, the source and drain
regions will comprise p-type regions. An ion implantation process
may be employed to create the source and drain regions 420a, 420b.
Depending upon the composition of the gate electrode 430, it may be
desirable to deposit a hard mask layer 490 (shown in dashed line)
over the gate electrode 430 during ion implantation in order to
inhibit the implantation of ions at the gate electrode.
[0032] Turning to FIG. 4E, insulating layers 450a, 450b have been
formed, and further ion implantation has been performed to create
deeper source and drain regions 420a, 420b. The insulating layers
450a, 450b also function as masks to inhibit the implantation of
ions in the underlying substrate 410, thereby forming source and
drain extensions 422a, 422b, respectively. Ion implantation on the
gate electrode material may be inhibited using a hard mask, as
noted above. Other features that may be formed include conductive
interconnects (e.g., see FIG. 1, items 160a-b), as well as a
passivation layer.
[0033] In an alternative embodiment, which is set forth in block
350 in FIG. 3, the composition of the transition metal alloy layer
402 is adjusted during the deposition process. As noted above, in
one embodiment, the gate electrode may comprise alternating thin
layers of different substances (e.g., Ti(Al) and C). Other
variations in the deposition process (e.g., pressure, temperature,
pulse rate, process sequences, precursors, etc.) may affect the
final composition of the transition metal alloy. In one embodiment
(see block 350), variations in the deposition process are
introduced to alter the composition of the transition metal alloy.
For example, this additional alloying of the gate electrode
material during deposition can be used to convert the work
function--e.g., from an n-type to a p-type--which can provide for a
simplified process flow (i.e., by avoiding deposition of two
separate gate electrode materials for n-channel and p-channel
devices). As will be appreciated by those of ordinary skill in the
art, conversion of the work function is but one example of the gate
electrode's characteristics that can be altered through this
additional alloying and, further, that such variations in the
deposition process can be used to adjust other chemical and
electrical properties of the gate electrode material.
[0034] In the embodiments illustrated with respect to FIGS. 3 and
4A-4E, the transition metal alloy functions as a gate electrode.
However, the disclosed transition metal alloys are not so limited
in function and application. In addition to use as a conductor and
gate electrode, the disclosed transition metal alloys may, in other
embodiments, function as a barrier layer and/or an etch stop.
Embodiments of the use of the disclosed transition metal alloys as
a diffusion barrier and/or etch stop (as well as part of a gate
electrode) are shown and described below with respect to FIGS. 5
and 6A through 6E.
[0035] Thus, turning now to FIG. 5, illustrated is another
embodiment of a method 500 of forming a gate electrode using a
transition metal alloy. The method 500 of FIG. 5 is further
illustrated by the schematic diagrams of FIGS. 6A through 6E, and
reference should be made to these figures, as called out in the
text. Also, many of the drawing elements shown in FIGS. 5 and 6A-6E
are the same or similar to those shown in FIGS. 3 and 4A-4E,
respectively, and like reference numerals are used to refer to the
same items in FIGS. 5 and 6A-6E.
[0036] Referring to block 310, a layer of gate insulating material
is formed on a substrate. This is illustrated in FIG. 6A, which
shows a gate insulating material layer 401 that has been deposited
on a substrate 410, as described above. Again, the gate insulating
material layer 401 may, in one embodiment, comprise a high-k
dielectric material (e.g., ZrO.sub.2, HfO.sub.2, Al.sub.2O.sub.3,
etc.
[0037] As set forth at block 550, a relatively thin layer of a
transition metal alloy is deposited over the gate insulating
material layer to "set" the threshold voltage of the transistor.
This is also illustrated in FIG. 6A, where a thin layer of a
transition metal alloy 602 has been deposited over the gate
insulating material layer 401. In one embodiment, the thickness of
the transition metal alloy layer 602 is between 25 and 100
Angstroms.
[0038] In one embodiment, the layer of transition metal alloy 602
comprises the transition metal alloy 200 shown and described above
with respect to FIG. 2. In a further embodiment, the layer of
transition metal alloy 602 comprises a transition metal carbide
and, in yet another embodiment, the transition metal alloy layer
602 comprises a transition metal carbide including Titanium,
Carbon, and Aluminum. In yet a further embodiment, the transition
metal alloy layer 602 comprises approximately 37 at % Titanium, 55
at % Carbon, and 3 at % Aluminum, as well as approximately 4 at %
Oxygen and 1 at % or less Nitrogen. The layer of transition metal
alloy 602 may be deposited using any suitable blanket deposition
technique, such as PVD, CVD, or ALD, as described above.
[0039] A layer of a conductive material is then deposited over the
layer of transition metal alloy, as set forth in block 560. This is
illustrated in FIG. 6B, where a layer of conductive material 603
has been deposited over the transition metal alloy layer 602 to
form a gate electrode. The conductive material layer 603 may
comprise any suitable conductive material, such as poly-silicon or
Aluminum. In one embodiment, the thickness of the conductive
material layer 603 is between 500 and 2,000 Angstroms.
[0040] The conductive material layer 603 may be deposited using any
suitable deposition technique. In one embodiment, the conductive
material layer 603 is deposited using a blanket deposition process,
such as CVD or PVD. In another embodiment, the conductive material
layer 603 is deposited using a selective deposition technique. In
this embodiment, the transition metal gate electrode and gate
insulating layer are first patterned by etching, and the conductive
material is selectively deposited on the transition metal gate
electrode using a selective deposition technique, such as
electroplating or electroless plating.
[0041] Etching is then performed to create the gate electrode
stack, as set forth at block 330 in FIG. 5. This is illustrated in
FIG. 6C, where both the conductive material layer 603, transition
metal alloy layer 602, and gate insulating material layer 401 have
each been etched to create a gate electrode stack 605. The gate
electrode stack 605 includes a conductive layer 680 overlying a
transition metal alloy layer 635 which, in turn, overlies a gate
insulating layer 440. Any suitable etching processes may be
employed to etch the conductive material, transition metal alloy,
and gate insulating material layers 603, 602, 401, respectively.
Although not shown in the figures, it will be appreciated by those
of ordinary skill in the art that a patterned mask layer may be
formed prior to etching.
[0042] Additional features may then be formed on the substrate to
create an NMOS or PMOS device, as set forth at block 340. This is
illustrated in FIGS. 6D and 6E, where additional features have been
created to form a device 600. With reference to FIG. 6D, source and
drain regions 420a, 420b are created by ion implantation. As
previously noted, a hard mask may not be needed in this embodiment,
where the conductive material layer 680 (e.g., poly-silicon) may
itself function as a hard mask during ion implantation. Referring
now to FIG. 6E, insulating layers 450a, 450b have been formed.
Further ion implantation results in the formation of deeper source
and drain regions 420a, 420b, and the insulating layers 450a, 450b
inhibit the implantation of ions in the underlying substrate 410,
such that source and drain extensions 422a, 422b are created, all
as described above. Other features that may be formed include
conductive interconnects (e.g., see FIG. 1, items 160a-b), as well
as a passivation layer.
[0043] As suggested above, the disclosed transition metal alloys
are not limited in function to their role as a conductor and gate
electrode. In one embodiment, the disclosed transition metal alloys
may further function as a barrier layer material and, in another
embodiment, these transition metal alloys may function as an etch
stop, as will now be described in more detail.
[0044] For some embodiments, the transition metal alloy 200 may not
be a "good" conductor. In this case, it may be desirable to using a
relatively thin layer of the transition metal alloy--as shown in
FIG. 6C--to achieve a desired threshold voltage with the optimal
work function characteristics of the transition metal alloy, and
then form another layer on top of the transition metal alloy of a
relatively good conductor, such as Aluminum. However, many metals
such as Aluminum that are good conductors will react with (e.g.,
lose electrons to) the underlying gate insulator material. Thus, in
this instance, the thin layer of transition metal alloy functions
as a barrier layer.
[0045] In another embodiment, as noted above, a relatively thin
layer of a transition metal alloy can function as an etch stop. As
an etch stop, a thin transition metal alloy layer is again applied
to "set" the threshold voltage of the transistor. A conductive
material that more readily etches--e.g., poly-silicon--is then
deposited over the thin transition metal alloy layer (see FIGS. 6B
and 6C). Etching of the conductive layer (e.g., poly-silicon) stops
at the thin transition metal alloy layer, and another etch process
is performed to remove any remaining transition metal alloy.
Materials such as poly-silicon are very amenable to anisotropic
etching processes and, further, because the transition metal alloy
layer is relatively thin as compared to the overall thickness of
the gate electrode (i.e., the electrode stack including the
transition metal alloy layer and the conductive material layer),
this layer is easy to remove. Note also that poly-silicon can
function as a hard mask material during ion implantation and,
therefore, where the conductive material layer 680 comprises
poly-silicon (or other similar material), a hard mask (see FIG. 4D,
item 490) may not be necessary during formation of the source and
drain regions by ion implantation.
[0046] Embodiments of a transition metal alloy that may be used in
the gate electrode of a transistor--as well as methods of forming a
gate electrode for a transistor--having been herein described,
those of ordinary skill in the art will appreciate the advantages
of the disclosed embodiments. The disclosed transition metal alloys
can function as a gate electrode in NMOS or PMOS devices, and these
transition metal alloys are thermally stable--e.g., the alloy's
work function does not significantly shift nor does the alloy react
with the underlying insulating material--at temperatures up to
900.degree. C. Thus, the disclosed transition metal alloys can be
integrated into existing CMOS process flows, where processing
temperatures can approach 900.degree. C. Also, using one of the
disclosed transition metal alloys as the gate electrode in a
transistor allows for further scaling down of the thickness of the
gate oxide. Further, in addition to functioning as a conductor and
gate electrode, the disclosed transition metal alloys can also
serve as a barrier layer and/or etch stop.
[0047] The foregoing detailed description and accompanying drawings
are only illustrative and not restrictive. They have been provided
primarily for a clear and comprehensive understanding of the
disclosed embodiments and no unnecessary limitations are to be
understood therefrom. Numerous additions, deletions, and
modifications to the embodiments described herein, as well as
alternative arrangements, may be devised by those skilled in the
art without departing from the spirit of the disclosed embodiments
and the scope of the appended claims.
* * * * *