U.S. patent application number 11/600062 was filed with the patent office on 2007-03-22 for semiconductor device and method for manufacturing the same.
Invention is credited to Kenji Yoneda.
Application Number | 20070063273 11/600062 |
Document ID | / |
Family ID | 19051861 |
Filed Date | 2007-03-22 |
United States Patent
Application |
20070063273 |
Kind Code |
A1 |
Yoneda; Kenji |
March 22, 2007 |
Semiconductor device and method for manufacturing the same
Abstract
A semiconductor device includes a gate insulating film formed on
a semiconductor substrate, and a gate electrode formed on the gate
insulating film. Nitrogen is introduced into the gate insulating
film, and the nitrogen concentration distribution thereof has a
peak near the surface of the gate insulating film or near the
center of the gate insulating film in the thickness direction. The
peak value of nitrogen concentration in the gate insulating film is
equal to or greater than 10 atm % and less than or equal to 40 atm
%.
Inventors: |
Yoneda; Kenji; (Kyoto,
JP) |
Correspondence
Address: |
NIXON PEABODY, LLP
401 9TH STREET, NW
SUITE 900
WASHINGTON
DC
20004-2128
US
|
Family ID: |
19051861 |
Appl. No.: |
11/600062 |
Filed: |
November 16, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10872403 |
Jun 22, 2004 |
7164178 |
|
|
11600062 |
Nov 16, 2006 |
|
|
|
10195367 |
Jul 16, 2002 |
6773999 |
|
|
10872403 |
Jun 22, 2004 |
|
|
|
Current U.S.
Class: |
257/333 ;
257/395; 257/E21.625; 257/E21.639 |
Current CPC
Class: |
H01L 21/26506 20130101;
H01L 29/518 20130101; Y10S 438/981 20130101; H01L 21/823857
20130101; H01L 21/823462 20130101; H01L 21/28185 20130101; H01L
21/28202 20130101 |
Class at
Publication: |
257/333 ;
257/395 |
International
Class: |
H01L 29/76 20060101
H01L029/76; H01L 29/94 20060101 H01L029/94; H01L 31/00 20060101
H01L031/00 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 18, 2001 |
JP |
2001-217571 |
Claims
1. A semiconductor device comprising: a semiconductor substrate
partitioned into a first region, a second region and a third
region; a first gate insulating film on the first region; a second
gate insulating film on the second region, the second gate
insulating film having a thickness thinner than that of the first
gate insulating film; and a third gate insulating film on the third
region, the third gate insulating film having a thickness thinner
than that of the second gate insulating film, wherein nitrogen is
introduced to the surface portion of the first gate insulating film
such that the nitrogen does not reach the interface between the
first gate insulating film and the semiconductor substrate; and
wherein nitrogen is introduced to the third gate insulating film
such that the nitrogen reaches the interface between the third gate
insulating film and the semiconductor substrate.
2. The semiconductor device of claim 1, wherein the thickness of
the first gate insulating film is equal to or greater than 3.5 nm
and less than or equal to 9 nm.
3. The semiconductor device of claim 1, wherein the thickness of
the second gate insulating film is equal to or greater than 1.0 nm
and less than or equal to 3.0 nm.
4. The semiconductor device of claim 1, wherein a nitrogen
concentration distribution of the first gate insulating film has a
peak near a surface of the first gate insulating film in a
thickness direction, and a nitrogen concentration distribution of
the second gate insulating film has a peak near a center of the
second gate insulating film in a thickness direction.
5. The semiconductor device of claim 1, wherein each peak value of
nitrogen concentrations in the first gate insulating film and the
second gate insulating film is equal to or greater than 10 atm %
and less than or equal to 40 atm %.
6. The semiconductor device of claim 1, wherein the nitrogen
concentration in the second gate insulating film at an interface
between the second gate insulating film and the semiconductor
substrate is equal to or greater than 0.2 atm % and less than or
equal to 3 atm %.
Description
BACKGROUND OF THE INVENTION
[0001] The present invention relates to a semiconductor device and
a method for manufacturing the same, and more particularly to a
method for forming a gate insulating film used in a MIS
transistor.
[0002] A MOS transistor is a typical MOS device. For example, in a
complementary MOS (CMOS) transistor, or the like, a high speed
driving transistor, which is required to have a gate insulating
film of a relatively thin thickness, and a high breakdown-voltage
transistor, which is required to have a gate insulating film of a
relatively thick thickness for handling input/output signals of a
relatively high voltage, are formed on a single semiconductor
substrate.
[0003] The high speed driving transistor is required to have a gate
insulating film having a thickness of about 1 nm to 3 nm, while it
is strongly required to have a high reliability in resisting
against dielectric breakdown and to have a low leakage current.
[0004] A CMOS transistor employs a so-called "dual gate structure",
in which the gate electrode of the P-channel transistor is a P-type
gate electrode obtained by using boron (B) as a dopant, and the
gate electrode of the N-channel transistor is an N-type gate
electrode obtained by using phosphorus (P) as a dopant. Boron,
being a P-type dopant, has a larger diffusion coefficient than that
of phosphorus, being an N-type dopant, whereby during a heat
treatment after the transistor is formed, boron diffuses through
the gate insulating film of the high speed driving transistor to
reach the channel region. The diffusion of boron is called "boron
penetration", and causes various problems in the transistor such as
a substantial variation in the threshold voltage and a
deterioration of the driving ability. The boron penetration is, of
course, more pronounced as the thickness of the gate insulating
film is reduced, and is particularly pronounced when silicon
dioxide (SiO.sub.2) is used for the gate insulating film.
[0005] Furthermore, reducing the thickness of the gate insulating
film also causes an increase in the gate leakage current through
the gate insulating film. Again, where silicon dioxide is used for
the gate insulating film, the conduction mechanism thereof is a
Fowler-Nordheim tunneling current if the thickness is 3.5 nm or
more, and the direct tunneling current becomes dominant if the
thickness is 3.5 nm or less. The gate leakage current increases by
an order of magnitude for every 0.2 nm decrease in the thickness of
the gate insulating film. If the thickness of the gate insulating
film is set to be 2.6 nm or less, the gate leakage current is no
longer negligible.
[0006] As described above, if a thermal oxide film is used for the
gate insulating film, it is no longer possible to suppress the
boron penetration and the gate leakage current. In view of this, an
oxynitride film into which nitrogen is introduced has been used as
a gate insulating film.
[0007] A conventional method for forming a gate insulating film of
a MOS semiconductor device using a silicon oxynitride film will now
be described with reference to the drawings.
[0008] FIG. 12A to FIG. 12C are cross-sectional views sequentially
illustrating the steps of the conventional method for forming a
gate insulating film.
[0009] First, a device isolation region 102 that partitions a
plurality of device forming regions from one another is formed in
an upper portion of a semiconductor substrate 101 made of silicon,
and then a first gate oxide film 103A made of a thermal oxide film
having a thickness of about 7.5 nm is formed entirely across the
upper surface of the semiconductor substrate 101. Then, a resist
pattern 104 having an opening in a second region 202 is formed on
the first gate oxide film 103A, and then a portion of the first
gate oxide film 103A that is included in the second region 202 is
etched away using the resist pattern 104 so that the second region
of the semiconductor substrate 101 is exposed, thereby obtaining a
structure as illustrated in FIG. 12A.
[0010] Then, as illustrated in FIG. 12B, the semiconductor
substrate 101 is subjected to a heat treatment so as to form a
second gate oxide film 105A made of a thermal oxide film having a
thickness of about 2.6 nm in the second region 202. In this
process, the thickness of the first gate oxide film 103A
increases.
[0011] Then, as illustrated in FIG. 12C, the semiconductor
substrate 101 is subjected to a heat treatment in an oxynitriding
atmosphere made of nitrogen monoxide (NO) at a temperature of
900.degree. C. for 30 seconds to several ten minutes so as to
introduce nitrogen into the first gate oxide film 103A and the
second gate oxide film 105A, thereby obtaining a first gate
oxynitride film 103B and a second gate oxynitride film 105B,
respectively. Note that other than nitrogen monoxide (NO),
dinitrogen monoxide (N.sub.2O) or, though rarely, ammonia
(NH.sub.3) may be used in the oxynitriding process using a heat
treatment.
[0012] When nitrogen monoxide (NO) is used, the oxynitriding
process increases the thickness only by 0.3 nm or less. In
contrast, when dinitrogen monoxide is used, it is required to
perform an oxynitriding process under a high temperature of about
1000.degree. C. to 1150.degree. C. for several ten seconds to
several ten minutes, whereby the oxynitriding process with
dinitrogen monoxide (N.sub.2O) increases the thickness by a
substantial amount of up to several nanometers. Therefore, with
dinitrogen monoxide (N.sub.2O), care should be taken for the
process.
[0013] FIG. 13A and FIG. 13B are each a nitrogen concentration
profile in a gate oxynitride film which has been oxynitrided by
using an oxynitriding atmosphere made of nitrogen monoxide (NO),
wherein FIG. 13A is for the first gate oxynitride film 103B and
FIG. 13B is for the second gate oxynitride film 105B. As
illustrated in FIG. 13B, in the second gate oxynitride film 105B
having a thickness of 2.6 nm, the nitrogen atom peak is located
near the interface between the second gate oxynitride film 105B and
the semiconductor substrate 101. The peak concentration is about 4
atm % at maximum, through it varies depending on the oxynitriding
temperature. Note that also when the oxynitriding process is
performed by using dinitrogen monoxide (N.sub.2O), the nitrogen
concentration profile is as that shown in FIG. 13B, and the peak
concentration is, at best, 1 atm %.
[0014] The second gate oxynitride film 105B obtained by the
conventional oxynitriding process has a nitrogen concentration
profile and a nitrogen concentration peak as shown in FIG. 13B,
whereby boron ion implanted into the p-type gate electrode of the
P-channel transistor diffuses through the second gate oxynitride
film 105B relatively easily, though it depends on the heat
treatment temperature, and reaches the channel region in the
semiconductor substrate 101. The diffusion of boron is of course
suppressed as compared with a gate oxide film made only of silicon
dioxide. However, when the thickness is reduced so much as in the
second gate oxynitride film 105B, it is not possible substantially
prevent the diffusion of boron with a nitrogen concentration
profile in which the nitrogen peak concentration is only about 4
atm % and the peak is located near the interface with the
semiconductor substrate 101. This is the first problem in the prior
art.
[0015] Furthermore, with such a silicon oxynitride film, in which
the nitrogen concentration is only about 4 atm % and the nitrogen
atoms are localized near the substrate interface, the nitrogen
content of the film as a whole is not sufficient to change the
dielectric constant and the refractive index of silicon dioxide
(SiO.sub.2), and thus it is certainly not expected to be sufficient
to provide an increase in the electric capacitance or a reduction
in the gate leakage current. This is the second problem in the
prior art.
SUMMARY OF THE INVENTION
[0016] An object of the present invention is to solve these
problems in the prior art and to make it possible to reduce the
gate leakage current while preventing dopant atoms from diffusing
from a gate electrode into a substrate through a gate insulating
film having a thickness that is so reduced that a direct tunneling
current can flow therethrough.
[0017] The present inventor has conducted various researches to
make it possible to suppress the penetration or diffusion of boron
through a thinned gate insulating film while reducing the gate
leakage current. As a result, it has been found that it is
preferred that the nitrogen concentration in the gate insulating
film is sufficiently high with a broad nitrogen distribution in the
insulation film and that a low-temperature process is used for an
improved control of the thickness of the thin film. Moreover, it
has also been found that in order to obtain an increased nitrogen
concentration in an oxynitriding process using nitrogen monoxide or
dinitrogen monoxide, it is necessary to increase the heat treatment
temperature of the oxynitriding process, which is not suitable for
the formation of a very thin gate insulating film.
[0018] Therefore, in order to prevent the penetration of boron
through a gate insulating film and to increase the dielectric
constant and the refractive index by increasing the nitrogen
concentration, thereby reducing the gate leakage current, it is
necessary to form, as a thin gate insulating film, an oxynitride
film that has a broad nitrogen distribution with a nitrogen peak
concentration over 10 atm %.
[0019] Specifically, a semiconductor device of the present
invention includes: a gate insulating film formed on a
semiconductor substrate; and a gate electrode formed on the gate
insulating film, wherein nitrogen is introduced into the gate
insulating film and a nitrogen concentration distribution thereof
has a first peak near a surface of the gate insulating film or near
a center of the gate insulating film in a thickness direction.
[0020] With the semiconductor device of the present invention,
nitrogen is introduced into the gate insulating film, and the
nitrogen concentration distribution thereof has the first peak near
the surface of the gate insulating film or near the center of the
gate insulating film in the thickness direction. Therefore, despite
the reduced thickness of the gate insulating film, it is possible
to prevent the diffusion of dopant atoms from the gate electrode
into the semiconductor substrate while reducing the gate leakage
current. Furthermore, the dielectric constant of the gate
insulating film increases, thereby increasing the resistance to
dielectric breakdown.
[0021] In the semiconductor device of the present invention, it is
preferred that the nitrogen concentration distribution of the gate
insulating film has a second peak near an interface between the
gate insulating film and the semiconductor substrate.
[0022] In this way, at the interface between the gate insulating
film and the semiconductor substrate, the electrical stress
resistance or immunity of the gate insulating film is improved, and
the gate leakage current is reduced.
[0023] In the semiconductor device of the present invention, it is
preferred that a value of the first peak of nitrogen concentration
in the gate insulating film is equal to or greater than 10 atm %
and less than or equal to 40 atm %.
[0024] Moreover, in the semiconductor device of the present
invention, it is preferred that the nitrogen concentration in the
gate insulating film at an interface between the gate insulating
film and the semiconductor substrate is equal to or greater than
0.2 atm % and less than or equal to 3 atm %.
[0025] A first method for manufacturing a semiconductor device of
the present invention includes the steps of: (a) forming a base
gate insulating film on a semiconductor substrate; and (b) exposing
the base gate insulating film to a nitrogen plasma so as to
introduce nitrogen atoms into the base gate insulating film,
thereby forming a gate insulating film from the base gate
insulating film.
[0026] With the first method for manufacturing a semiconductor
device, the gate insulating film is formed from a base gate
insulating film by introducing nitrogen atoms into the base gate
insulating film by exposing the base gate insulating film to a
nitrogen plasma, whereby it is possible to form, as a thinned gate
insulating film, an oxynitride film that has a relatively broad
nitrogen distribution with a nitrogen peak concentration over 10
atm %. As a result, despite the reduced thickness of the gate
insulating film, it is possible to prevent the diffusion of dopant
atoms from the gate electrode into the semiconductor substrate
while reducing the gate leakage current. Furthermore, the
dielectric constant of the gate insulating film increases, thereby
increasing the resistance to dielectric breakdown.
[0027] In the first method for manufacturing a semiconductor
device, it is preferred that in the step (a), the semiconductor
substrate is subjected to a heat treatment in an oxidizing
atmosphere so as to form the base gate insulating film made of an
oxide film on the semiconductor substrate.
[0028] Moreover, in the first method for manufacturing a
semiconductor device, it is preferred that in the step (a), the
semiconductor substrate is subjected to a heat treatment in an
oxynitriding atmosphere without hydrogen so as to form the base
gate insulating film made of an oxynitride film on the
semiconductor substrate.
[0029] As described above, in the step (a), the base gate
insulating film is formed by performing a heat treatment in an
oxynitriding atmosphere without hydrogen, whereby nitrogen atoms
are introduced also into a portion of the base gate insulating film
near the interface between the base gate insulating film and the
semiconductor substrate. As a result, the electrical stress
resistance or immunity of the base gate insulating film is
improved, and the gate leakage current is reduced.
[0030] In such a case, it is preferred that the oxynitriding
atmosphere is an atmosphere containing nitrogen monoxide and oxygen
or an atmosphere containing dinitrogen monoxide.
[0031] In such a case, it is preferred that the method further
includes, before the step (a), a step of implanting impurity ion
that causes an enhanced oxidization effect into the semiconductor
substrate.
[0032] In this way, the thickness of the base gate insulating film
grown on the semiconductor substrate is increased by the
accelerated oxidization effect. Thus, by selectively implanting
impurity ion that causes an accelerated oxidization effect, the
thickness of the base gate insulating film in the region in which
the impurity ion is implanted can be made different from that in
the region in which the impurity ion is not implanted.
[0033] In the first method for manufacturing a semiconductor
device, it is preferred that in the step (a), the semiconductor
substrate is subjected to an oxynitriding atmosphere containing a
nitrogen plasma and an oxygen plasma produced from dinitrogen
monoxide so as to form the base gate insulating film made of an
oxynitride film on the semiconductor substrate.
[0034] In the first method for manufacturing a semiconductor
device, it is preferred that the method further includes, before
the step (a): a first step of partitioning the semiconductor
substrate into a first region and a second region; a second step of
forming a first base gate insulating film made of a thermal oxide
film on the first region and the second region; and a third step of
removing a portion of the first base gate insulating film that is
included in the second region, wherein: in the step (a), a second
base gate insulating film to be the gate insulating film having a
thickness smaller than that of the first base gate insulating film
is formed on the second region of the semiconductor substrate; and
in the step (b), the first base gate insulating film and the second
base gate insulating film are exposed to a nitrogen plasma so as to
introduce nitrogen atoms into the first base gate insulating film
and the second base gate insulating film, thereby forming a first
gate insulating film from the first base gate insulating film and a
second gate insulating film to be the gate insulating film from the
second base gate insulating film.
[0035] In this way, the first gate insulating film and the second
gate insulating film, which have different thicknesses, can be
formed on the first region and the second region, respectively, of
the semiconductor substrate, while nitrogen atoms can be reliably
introduced into each gate insulating film.
[0036] In such a case, it is preferred that: a thickness of the
first gate insulating film is equal to or greater than 3.5 nm and
less than or equal to 9 nm, and a nitrogen concentration
distribution thereof has a peak near a surface of the first gate
insulating film and another peak near an interface between the
first gate insulating film and the semiconductor substrate; and a
thickness of the second gate insulating film is equal to or greater
than 1.0 nm and less than or equal to 3.0 nm, and a nitrogen
concentration distribution thereof has a peak near a center of the
second gate insulating film in a thickness direction.
[0037] In the first method for manufacturing a semiconductor
device, it is preferred that the method further includes, before
the step (a): a first step of partitioning the semiconductor
substrate into a first region, a second region and a third region;
a second step of forming a first base gate insulating film made of
a thermal oxide film on the first region, the second region and the
third region; a third step of implanting impurity ion that causes
an enhanced oxidization effect into the second region of the
semiconductor substrate, after the second step; and a fourth step
of removing a portion of the first base gate insulating film that
is included in the second region and the third region, wherein: in
the step (a), a second base gate insulating film having a thickness
smaller than that of the first base gate insulating film is formed
on the second region of the semiconductor substrate, and a third
base gate insulating film to be the gate insulating film having a
thickness smaller than that of the second base gate insulating film
is formed on the third region; and in the step (b), the first base
gate insulating film, the second base gate insulating film and the
third base gate insulating film are exposed to a nitrogen plasma so
as to introduce nitrogen atoms into the first base gate insulating
film, the second base gate insulating film and the third base gate
insulating film, thereby forming a first gate insulating film from
the first base gate insulating film, a second gate insulating film
from the second base gate insulating film and a third gate
insulating film to be the gate insulating film from the third base
gate insulating film.
[0038] In this way, the first gate insulating film, the second gate
insulating film and the third gate insulating film, which have
different thicknesses, can be formed on the first region, the
second region and the third region, respectively, of the
semiconductor substrate, while nitrogen atoms can be reliably
introduced into each gate insulating film.
[0039] In the first method for manufacturing a semiconductor
device, it is preferred that the impurity ion is fluorine or
silicon, which is implanted into a portion of the semiconductor
substrate near a surface thereof at a dose equal to or greater than
1.times.10.sup.14 cm.sup.-2 and less than or equal to
5.times.10.sup.15 cm.sup.-2.
[0040] In the first method for manufacturing a semiconductor
device, it is preferred that the nitrogen plasma is a high-density
plasma at a temperature in a range from room temperature to
500.degree. C.
[0041] In the first method for manufacturing a semiconductor
device, it is preferred that a peak value of nitrogen concentration
in the gate insulating film is equal to or greater than 10 atm %
and less than or equal to 40 atm %.
[0042] In the first method for manufacturing a semiconductor
device, it is preferred that an oxygen plasma is added to the
nitrogen plasma.
[0043] A second method for manufacturing a semiconductor device of
the present invention includes the steps of: (a) exposing an entire
surface of a semiconductor substrate to a nitrogen plasma and an
oxygen plasma so as to form a gate insulating film made of an
oxynitride film on the semiconductor substrate; and (b) selectively
forming a gate electrode on the gate insulating film.
[0044] With the second method for manufacturing a semiconductor
device, it is possible to form, as a gate insulating film formed on
the semiconductor substrate, an oxynitride film that has a
relatively broad nitrogen distribution with a nitrogen peak
concentration over 10 atm % by exposing the entire surface of the
semiconductor substrate to a nitrogen plasma and an oxygen plasma.
As a result, despite the reduced thickness of the gate insulating
film, it is possible to prevent the diffusion of dopant atoms from
the gate electrode into the semiconductor substrate while reducing
the gate leakage current. Furthermore, the dielectric constant of
the gate insulating film increases, thereby increasing the
resistance to dielectric breakdown.
[0045] In the second method for manufacturing a semiconductor
device, it is preferred that the method further includes, before
the step (a): a first step of partitioning the semiconductor
substrate into a first region and a second region; a second step of
forming a first base gate insulating film made of a thermal oxide
film on the first region and the second region; and a third step of
removing a portion of the first base gate insulating film that is
included in the second region, wherein in the step (a), the entire
surface of the semiconductor substrate including the first base
gate insulating film is exposed to the nitrogen plasma and the
oxygen plasma so as to form a second gate insulating film to be the
gate insulating film having a thickness smaller than that of the
first base gate insulating film on the second region and to
introduce nitrogen atoms into the first base gate insulating film,
thereby forming a first gate insulating film from the first base
gate insulating film.
[0046] In this way, the first gate insulating film and the second
gate insulating film, which have different thicknesses and into
which nitrogen atoms are introduced, can be formed on the first
region and the second region, respectively, of the semiconductor
substrate.
[0047] In such a case, it is preferred that: a thickness of the
first gate insulating film is equal to or greater than 3.5 nm and
less than or equal to 9 nm, and a nitrogen concentration
distribution thereof has a peak near a surface of the first gate
insulating film; and a thickness of the second gate insulating film
is equal to or greater than 1.0 nm and less than or equal to 3.0
nm, and a nitrogen concentration distribution thereof has a peak
near a center of the second gate insulating film in a thickness
direction.
BRIEF DESCRIPTION OF THE DRAWINGS
[0048] FIG. 1A to FIG. 1C are cross-sectional views sequentially
illustrating steps of a method for manufacturing a semiconductor
device according to the first embodiment of the present
invention.
[0049] FIG. 2A to FIG. 2C are cross-sectional views sequentially
illustrating steps of the method for manufacturing a semiconductor
device according to the first embodiment of the present
invention.
[0050] FIG. 3A and FIG. 3B are each a nitrogen concentration
profile in a gate insulating film immediately after the formation
of a second gate insulating film illustrated in FIG. 2A, wherein
FIG. 3A is for a first gate insulating film and FIG. 3B is for the
second gate insulating film.
[0051] FIG. 4A and FIG. 4B are each a nitrogen concentration
profile in a gate insulating film after a plasma nitridation
process illustrated in FIG. 2B, wherein FIG. 4A is for the first
gate insulating film and FIG. 4B is for the second gate insulating
film.
[0052] FIG. 4C, being presented for the purpose of comparison, is a
nitrogen concentration profile that is obtained when the thickness
of the second gate insulating film is increased.
[0053] FIG. 5 is a graph showing the heat treatment time dependence
of the change in threshold voltage due to penetration of boron from
a P-type gate electrode in the method for manufacturing a
semiconductor device according to the first embodiment of the
present invention, in comparison with that of a conventional
example.
[0054] FIG. 6 is a graph showing the relationship between the gate
leakage current and the SiO.sub.2-equivalent thickness value in the
method for manufacturing a semiconductor device according to the
first embodiment of the present invention, in comparison with that
of a conventional example.
[0055] FIG. 7 is a graph showing the relationship between the
nitrogen plasma exposure temperature and the nitrogen concentration
of the insulation film in the method for manufacturing a
semiconductor device according to the first embodiment of the
present invention.
[0056] FIG. 8A to FIG. 8D are cross-sectional views sequentially
illustrating the steps of a method for manufacturing a
semiconductor device according to the second embodiment of the
present invention.
[0057] FIG. 9A and FIG. 9B are each a nitrogen concentration
profile in a gate insulating film after a plasma nitridation
process illustrated in FIG. 8C, wherein FIG. 9A is for the first
gate insulating film and FIG. 9B is for the second gate insulating
film.
[0058] FIG. 10A to FIG. 10C are cross-sectional views sequentially
illustrating steps of a method for manufacturing a semiconductor
device according to the third embodiment of the present
invention.
[0059] FIG. 11A and FIG. 11B are cross-sectional views sequentially
illustrating steps of the method for manufacturing a semiconductor
device according to the third embodiment of the present
invention.
[0060] FIG. 12A to FIG. 12C are cross-sectional views sequentially
illustrating the steps of a conventional method for forming a gate
oxynitride film of a MOS semiconductor device.
[0061] FIG. 13A and FIG. 13B are each a nitrogen concentration
profile in a gate insulating film after the oxynitriding process
illustrated in FIG. 12C, wherein FIG. 13A is for the first gate
insulating film and FIG. 13B is for the second gate insulating
film.
DETAILED DESCRIPTION OF THE INVENTION
First Embodiment
[0062] The first embodiment of the present invention will now be
described with reference to the drawings.
[0063] FIG. 1A to FIG. 1C and FIG. 2A to FIG. 2C are
cross-sectional views sequentially illustrating the steps of a
method for manufacturing a semiconductor device according to the
first embodiment of the present invention.
[0064] First, as illustrated in FIG. 1A, a semiconductor substrate
11 is prepared. For example, the semiconductor substrate 11
includes, in an upper portion thereof, an epitaxial layer (not
shown) having a thickness of about 5 .mu.m and made of P-type
silicon whose specific resistance is about 11 .OMEGA.cm to 14
.OMEGA.cm, and the semiconductor substrate 11 has a specific
resistance of about 0.01 .OMEGA.cm to 0.02 .OMEGA.cm. Then, a
device isolation region 12, which is a shallow trench isolation, is
formed in an upper portion of the semiconductor substrate 11. The
device isolation region 12 partitions the principal surface of the
semiconductor substrate 11 at least into a first device forming
region 51 and a second device forming region 52. Then, although not
shown, a P-type well region and an N-type well region are formed in
the semiconductor substrate 11, and a channel doping process is
performed for adjusting the threshold voltage of the
transistor.
[0065] Then, the surface of the semiconductor substrate 11, in
which the device isolation region 12 has been formed, is subjected
to an SC1 cleaning process with a mixed solution of ammonium
hydroxide (NH.sub.4OH), hydrogen peroxide (H.sub.2O.sub.2) and
water (H.sub.2O) at a temperature of about 50.degree. C., and a
cleaning process with diluted hydrofluoric acid solution (HF), so
as to remove the natural oxide film from the semiconductor
substrate 11. Then, the cleaned semiconductor substrate 11 is
placed into a vertical electric furnace, and the semiconductor
substrate 11 is subjected to pyrogenic oxidization at a temperature
of about 800.degree. C., thereby forming a first gate insulating
film 13A made of silicon dioxide having a thickness of about 5.5 nm
on the principal surface of the semiconductor substrate 11. Then, a
resist pattern 14 having an opening in the second device forming
region 52 is formed on the semiconductor substrate 11 by using a
photolithography method, thereby obtaining a structure as
illustrated in FIG. 1B.
[0066] Then, as illustrated in FIG. 1C, a portion of the first gate
insulating film 13A that is included in the second device forming
region 52 is removed by using the resist pattern 14 as a mask and
using a buffered hydrofluoric acid solution. Then, a piranha
cleaning (=SPM cleaning) process with a mixed solution of sulfuric
acid (H.sub.2SO.sub.4) and hydrogen peroxide is performed to remove
the resist pattern 14. Then, the second device forming region 52 of
the semiconductor substrate 11 is subjected to an SC1 cleaning
process at a temperature of 50.degree. C., as a pre-cleaning
process for the formation of a second gate insulating film. As a
result of the SC1 cleaning process, the thickness of the first gate
insulating film 13A is reduced by about 0.2 nm to be about 5.3
nm.
[0067] Then, as illustrated in FIG. 2A, the semiconductor substrate
11 as illustrated in FIG. 1C is placed into a rapid thermal
processor (RTP), and is subjected to an oxynitriding process in an
oxynitriding atmosphere made of a dinitrogen monoxide (N.sub.2O)
gas at a atmospheric pressure or reduced pressure and at a
temperature of about 900.degree. C., so as to form a second gate
insulating film 15B made of silicon oxynitride having a thickness
of about 1.8 nm in the second device forming region 52 of the
semiconductor substrate 11. In this process, nitrogen atoms are
introduced by the oxynitriding process also into the first gate
insulating film 13A and the thickness thereof increases to be about
5.5 nm. Thus, the first gate insulating film 13A is altered into a
first gate insulating film 13B made of a silicon oxynitride film.
Note that a mixed gas of about 10% by volume of nitrogen monoxide
(NO) and about 90% by volume of oxygen (O.sub.2) may be used,
instead of dinitrogen monoxide, as the oxynitriding atmosphere.
[0068] Then, as illustrated in FIG. 2B, a nitridation process is
performed in which the first gate insulating film 13B and the
second gate insulating film 15B are exposed to a nitrogen (N.sub.2)
plasma. In FIG. 2B, the designation "N*" is used for nitrogen
radicals. A high density plasma of nitrogen plasma is produced by
using, for example, an inductively-coupled plasma system or a
helicon plasma system. In a case where an inductively-coupled
plasma system is used, the substrate is exposed to a nitrogen
plasma for about 90 seconds with a plasma-producing frequency of
about 13.56 MHz, a high-frequency power of about 500 W, a chamber
pressure of about 1.33 Pa and a substrate temperature of about
30.degree. C. Herein, no substrate bias is applied. In a case where
a helicon plasma system is used, the substrate is exposed to a
nitrogen plasma for about 120 seconds with a plasma-producing
frequency of about 13.56 MHz, a high-frequency power of about 500
W, a chamber pressure of about 1.33 Pa and a substrate temperature
of about 30.degree. C. By performing either nitridation process,
the first gate insulating film 13B and the second gate insulating
film 15B are further nitrided and altered into a first gate
insulating film 13C and a second gate insulating film 15C,
respectively.
[0069] Then, a conductive film having a thickness of about 150 nm
and made of polycrystal silicon having a small and randomly
oriented grain structure with a grain diameter of about 20 nm is
deposited by using a CVD method, for example, at a deposition
temperature of 680.degree. C. on the first gate insulating film 13C
and the second gate insulating film 15C, which have been subjected
to the nitridation process with a nitrogen plasma. Then, the
deposited conductive film is subjected to a predetermined
patterning process so as to form gate electrodes 16 from the
conductive film, thereby obtaining a structure as illustrated in
FIG. 2C.
[0070] Then, although not shown, a predetermined ion implantation
process is performed for each of the first device forming region 51
and the second device forming region 52 so as to form an LDD region
and a source/drain region, thereby obtaining a transistor
structure. In this process, the first gate insulating film 13C and
the second gate insulating film 15C are simultaneously formed with
thicknesses of about 5.5 nm and about 1.8 nm, respectively.
[0071] Note that the oxynitriding process in the step of forming
the second gate insulating film 15B illustrated in FIG. 2A may be
performed in an oxynitriding atmosphere containing a nitrogen
plasma and an oxygen plasma produced from dinitrogen monoxide,
instead of using RTP. In this way, the oxynitriding process can be
performed at a low temperature even with dinitrogen monoxide,
whereby it is possible to prevent an unnecessary increase in the
thickness of the first gate insulating film 13B.
[0072] Moreover, before the deposition of the conductive film for
forming gate electrodes illustrated in FIG. 2C, an annealing
process may be performed in a normal-pressure or depressurized
non-reactive atmosphere or small amount of oxygen and non-reactive
atmosphere at a temperature of about 700.degree. C. to about
1000.degree. C. for several ten seconds, in order to stabilize
nitrogen atoms in the first gate insulating film 13C and the second
gate insulating film 15C. Note however that the
nitrogen-stabilizing annealing process is not performed in the
present embodiment.
[0073] The nitrogen concentration profiles of the first gate
insulating films 13B and 13C, and the second gate insulating films
15B and 15C, will now be described.
[0074] FIG. 3A and FIG. 3B are nitrogen concentration profiles in
the first gate insulating film 13B and the second gate insulating
film 15B as illustrated in FIG. 2A, wherein FIG. 3A is for the
first gate insulating film 13B and FIG. 3B is for the second gate
insulating film 15B.
[0075] As illustrated in FIG. 3A, the nitrogen concentration in the
first gate insulating film 13B having a thickness of about 5.5 nm
has a distribution with a peak value of about 1.0 atm % near the
interface with the semiconductor substrate 11. On the other hand,
as illustrated in FIG. 3B, the nitrogen concentration in the second
gate insulating film 15B having a thickness of about 1.8 nm has a
broad distribution in the insulation film with the concentration
peak thereof being located near the center of the insulation film
in the thickness direction (the direction vertical to the substrate
plane). The peak nitrogen concentration is evaluated to be about
1.5 atm % by secondary ion mass spectrometry (SIMS).
[0076] Next, the nitrogen concentration profile after the
nitridation process with an inductively-coupled plasma will be
described.
[0077] FIG. 4A and FIG. 4B are nitrogen concentration profiles in
the first gate insulating film 13C and the second gate insulating
film 15C as illustrated in FIG. 2B, wherein FIG. 4A is for the
first gate insulating film 13C and FIG. 4B is for the second gate
insulating film 15C.
[0078] As illustrated in FIG. 4A, the nitrogen distribution in the
first gate insulating film 13C having a relatively large thickness
includes a first peak near the surface thereof at which the
nitrogen concentration is about 14 atm %, and a second peak near
the interface with the semiconductor substrate 11 at which the
nitrogen concentration is about 1 atm %.
[0079] On the other hand, as illustrated in FIG. 4B, the nitrogen
distribution in the second gate insulating film 15C having a
relatively thin thickness (indicated by a solid line) is the
integrated value of the concentration profile obtained by a plasma
nitridation process (indicated by a broken line) and the
concentration profile obtained by a thermal oxynitriding process
(indicated by a one-dot chain line). Herein, the thickness of the
second gate insulating film 15C is about 1.8 nm, i.e., about 1/3 of
that of the first gate insulating film 13C, whereby the second gate
insulating film 15C has a relatively broad nitrogen distribution in
the film. Moreover, the concentration distribution has a peak
concentration of about 14 atm % near the center of the film in the
thickness direction.
[0080] FIG. 4C, being presented for the purpose of comparison, is a
nitrogen concentration profile that is obtained when the thickness
of the second gate insulating film 15C is set to be about 2.6 nm.
It can be seen that the nitrogen distribution does not
substantially change even if the thickness of the second gate
insulating film 15C is slightly increased to be about 2.6 nm. This
is because the plasma-producing frequency and the high-frequency
power are the factors that influence the nitrogen distribution.
[0081] Herein, as illustrated in FIG. 4A and FIG. 4B, the nitrogen
concentration at the interface between the first gate insulating
film 13C and the semiconductor substrate 11 and that between the
second gate insulating film 15C and the semiconductor substrate 11
are both shown to be about 0.5 atm %. Thus, the nitrogen
concentration at the interface between the gate insulating film 13C
or 15C and the semiconductor substrate 11 is preferably about 0.2
atm % to about 3 atm %, and more preferably about 0.5 atm % to
about 1 atm %. Then, the interface between the gate insulating film
13C and the semiconductor substrate 11 and the interface between
the gate insulating film 15C and the semiconductor substrate 11
have desirable characteristics, whereby the electrical stress
resistance or immunity of the gate insulating films 13C and 15C is
improved, and the gate leakage current through the gate insulating
films 13C and 15C can be reduced. Note that it is undesirable that
the nitrogen concentration at the interface between the gate
insulating film 13C or 15C and the semiconductor substrate 11
exceeds 3 atm %, in which case the threshold voltage of the MOS
transistor may become unstable, and the carrier mobility may
deteriorate.
[0082] The electric characteristics of a P-channel transistor using
the second gate insulating film 15C of the first embodiment will
now be described.
[0083] FIG. 5 shows the heat treatment time dependence of the
change in threshold voltage due to penetration of boron from a
boron-doped P-type gate electrode of a P-channel transistor. As
indicated by a solid line in FIG. 5, with the second gate
insulating film 15C having a thickness of about 1.8 nm, there is no
change in threshold voltage for heat treatment times of up to 60
seconds when the heat treatment temperature is about 1050.degree.
C., and there is no change in threshold voltage even if the heat
treatment time exceeds 75 seconds when the heat treatment
temperature is set to be about 1000.degree. C. For comparison,
broken lines show the heat treatment time dependence of the change
in threshold voltage with a gate insulating film of a conventional
example having a thickness of 2.6 nm, where the threshold voltage
changes by 0.3 V only with a heat treatment for 30 seconds even
when the heat treatment temperature is about 1000.degree. C.
[0084] FIG. 6 shows the relationship between the gate leakage
current and the SiO.sub.2-equivalent thickness value. Herein, the
symbol ".largecircle." is used for a gate insulating film made of
silicon dioxide, the symbol ".DELTA." is used for a gate insulating
film of a conventional example, and the symbol "X" is used for the
second gate insulating film formed by the method of the first
embodiment. For each equivalent thickness value, a thickness value
obtained by ellipsometry is provided in parentheses.
[0085] As shown in FIG. 6, based on the relationship between the
gate leakage current and the equivalent thickness value of a
silicon oxynitride film, which is obtained by changing the
thickness of the second gate insulating film of the first
embodiment, it can be seen that with the gate insulating film of
the present embodiment, the gate leakage current can be reduced by
about an order of magnitude if the equivalent thickness value is
substantially equal to that of the conventional example.
[0086] Note that in the first embodiment, the first gate oxide film
13A is formed as a thermal oxide film, while the second gate
insulating film 15B is formed as a thermal oxynitride film by using
a dinitrogen monoxide atmosphere or a mixed atmosphere of nitrogen
monoxide and oxygen. However, it is of course possible to obtain
similar effects even if the second gate insulating film 15B is also
formed by a thermal oxide film.
[0087] Note however that if the second gate insulating film 15B is
formed by a thermal oxide film, the nitrogen concentration of the
insulation film cannot be increased to a concentration as high as
that of the second gate insulating film 15C of the first
embodiment. Note that "nitrogen concentration" as used herein is
not the peak value, but the integrated value of the nitrogen
concentration in the insulation film. This is because the peak
concentration is determined by the plasma oxynitriding process. The
integrated value of the nitrogen concentration in the insulation
film has a significant influence on the reduction of the
penetration of boron and the reduction of the gate leakage
current.
[0088] Next, the relationship between the nitrogen plasma exposure
temperature and the nitrogen concentration of the insulation film
will be described.
[0089] FIG. 7 shows the nitrogen distribution in the insulation
film made of silicon dioxide and the substrate made of silicon for
different nitrogen plasma exposure temperatures of room
temperature, 300.degree. C., 550.degree. C. and 750.degree. C. Note
that "room temperature" as used herein refers to a temperature of
about 20.degree. C. to about 40.degree. C. As can be seen from FIG.
7, exposure at a temperature around room temperature is preferred
in order to obtain a high nitrogen concentration and a very sharp
concentration distribution in the insulation film. With substrate
temperatures of up to about 500.degree. C., the influence on the
nitrogen concentration and the nitrogen distribution is
substantially negligible. Thus, the process temperature can be set
in the range from room temperature to about 500.degree. C.
[0090] Furthermore, as can be seen from FIG. 7, if the substrate is
exposed to a nitrogen plasma at a temperature of 550.degree. C. or
more, the substrate is significantly nitrided, while the thickness
distribution in the plane of the insulation film and the nitrogen
concentration distribution in the insulation film deteriorate
significantly, although not shown in the figure. Furthermore, when
the temperature is increased to be 600.degree. C. or more, it is
difficult to realize a uniform and shallow nitrogen distribution in
the insulation film due to the decomposition and thermal diffusion
of nitrogen radicals on the insulation film surface. Therefore, the
nitrogen plasma exposure is preferably done at a temperature that
is around room temperature or, at maximum, less than or equal to
about 500.degree. C.
[0091] Note that the exposure may be done at a temperature lower
than room temperature. However, it is recommended that the exposure
is done at a temperature around room temperature because under such
a low temperature, condensation, etc., may occur on the surface of
the substrate. Furthermore, it is appropriate that the plasma
exposure temperature is set to room temperature or a temperature
slightly higher than room temperature in view of the convenience
and stability of the process and the practicability of the
apparatus. As described above, in the first embodiment, the
nitridation process using a plasma performed at a temperature
around room temperature is one important factor.
[0092] In order to introduce a high concentration of nitrogen into
the insulation film (silicon dioxide film) by a process performed
at a temperature in the range from room temperature to about
500.degree. C., which is relatively low as a process temperature,
the nitrogen radical concentration needs to be high. Therefore, in
the present embodiment, a high-density plasma is used as a plasma
source so that a high nitrogen radical concentration can be easily
obtained. Moreover, since hydrogen atoms deteriorate the gate
insulating film characteristics, it is preferred that a nitridation
source gas containing no hydrogen is used for generating nitrogen
radicals.
[0093] Moreover, with a common low-density plasma using microwave,
or the like, a sufficient nitrogen radical supply cannot be
obtained, and it is thus very difficult to introduce nitrogen into
the insulation film at a high concentration.
[0094] Furthermore, the gate insulating film of the first
embodiment requires a plasma that does not cause a charge-up
phenomenon, i.e., a uniform plasma, in order to prevent dielectric
breakdown due to the charge-up phenomenon. According to the present
embodiment, the thickness distribution in the plane of a wafer
having a diameter of about 200 mm is as small as .+-.0.1 nm
(3.sigma., where ".sigma." represents the standard deviation) in a
case where the thickness of the insulation film is 1.8 nm. A
uniform thickness distribution and a nitrogen concentration profile
as described above are necessary in view of the fact that the gate
leakage current changes by as much as an order of magnitude for a
0.2 nm change in the thickness.
[0095] As described above, according to the first embodiment,
despite the reduced thickness of the second gate insulating film
15C, it is possible to prevent the diffusion of dopant atoms from
the gate electrode 16 into the semiconductor substrate 11 while
reducing the gate leakage current. Furthermore, the dielectric
constant of the second gate insulating film 15C increases, thereby
increasing the resistance to dielectric breakdown.
[0096] Note that in the first embodiment, the formation of the
second gate insulating film 15B and the nitridation process by
plasma exposure may be performed by using separate apparatuses, or
may be performed continuously by using a multi-chamber apparatus,
i.e., by forming the second gate insulating film 15B in one chamber
and then performing the nitridation process in another chamber.
Second Embodiment
[0097] The second embodiment of the present invention will now be
described with reference to the drawings.
[0098] FIG. 8A to FIG. 8D are cross-sectional views sequentially
illustrating the steps of a method for manufacturing a
semiconductor device according to the second embodiment of the
present invention.
[0099] First, a device isolation region 22, which is a shallow
trench isolation, is formed in an upper portion of a semiconductor
substrate 21 made of P-type silicon. The device isolation region 22
partitions at least the first device forming region 51 and the
second device forming region 52 from each other. Then, although not
shown, a P-type well region and an N-type well region are formed in
an upper portion of the semiconductor substrate 21, and a channel
doping process is performed for adjusting the threshold voltage of
the transistor. Then, the surface of the semiconductor substrate
21, in which the device isolation region 22 has been formed, is
subjected to an SC1 cleaning process at a temperature of about
50.degree. C. and a cleaning process with diluted hydrofluoric acid
solution (HF), so as to remove the natural oxide film from the
semiconductor substrate 21. Then, the semiconductor substrate 21 is
placed into an RTP, and the semiconductor substrate 21 is subjected
to pyrogenic oxidization in a water vapor atmosphere at a
temperature of about 1000.degree. C., thereby forming a first gate
insulating film 23A made of silicon dioxide having a thickness of
about 7.5 nm on the principal surface of the semiconductor
substrate 21. Then, a resist pattern 24 having an opening in the
second device forming region 52 is formed on the semiconductor
substrate 21 by using a photolithography method, thereby obtaining
a structure as illustrated in FIG. 8A.
[0100] Then, as illustrated in FIG. 8B, a portion of the first gate
insulating film 23A that is included in the second device forming
region 52 is removed by using the resist pattern 24 as a mask and
using a buffered hydrofluoric acid solution, and a piranha cleaning
process is performed to remove the resist pattern 24. Then, the
second device forming region 52 of the semiconductor substrate 21
is subjected to an SC1 cleaning process at a temperature of
50.degree. C., as a pre-cleaning process for the formation of a
second gate insulating film. As a result of the SC1 cleaning
process, the thickness of the first gate insulating film 23A is
reduced by about 0.2 nm to be about 7.3 nm.
[0101] Then, as illustrated in FIG. 8C, the semiconductor substrate
21, in which the second device forming region 52 is exposed, is
exposed to a nitrogen plasma and an oxygen plasma, which are
produced by an inductively-coupled plasma system. Herein, the
plasma gas is a mixed gas of about 95% of nitrogen and about 5% of
oxygen, the frequency is about 13.56 MHz, the high-frequency power
is about 500 W, the chamber pressure is about 1.33 Pa, and the
substrate temperature is about 30.degree. C. With the oxygen plasma
and the nitrogen plasma, the semiconductor substrate 21 is oxidized
and nitrided simultaneously, thereby forming a second gate
insulating film 25C, which is a plasma oxynitride film having a
thickness of about 1.6 nm, in the second device forming region 52.
In this process, the first gate insulating film 23A is altered by
the oxynitriding process with a nitrogen plasma and an oxygen
plasma into a first gate insulating film 23C having a thickness of
7.5 nm.
[0102] Then, a conductive film having a thickness of about 150 nm
and made of polycrystal silicon having a small and randomly
oriented grain structure with a grain diameter of about 20 nm is
deposited by using a CVD method at a deposition temperature of
680.degree. C. on the semiconductor substrate 21, on which the
first gate insulating film 23C and the second gate insulating film
25C have been formed. Then, the deposited conductive film is
subjected to a predetermined patterning process so as to form gate
electrodes 26 from the conductive film, thereby obtaining a
structure as illustrated in FIG. 8D.
[0103] Then, although not shown, a predetermined ion implantation
process is performed for each of the first device forming region 51
and the second device forming region 52 so as to form an LDD region
and a source/drain region, thereby obtaining a transistor
structure.
[0104] The nitrogen concentration profiles of the first gate
insulating film 23C and the second gate insulating film 25C
obtained by an oxynitriding process using an inductively-coupled
plasma according to the second embodiment of the present invention
will now be described.
[0105] FIG. 9A and FIG. 9B are nitrogen concentration profiles in
the first gate insulating film 23C and the second gate insulating
film 25C, wherein FIG. 9A is for the first gate insulating film 23C
and FIG. 9B is for the second gate insulating film 25C.
[0106] As illustrated in FIG. 9A, the nitrogen distribution in the
first gate insulating film 23C has a concentration peak of about 15
atm % near the surface of the film. On the other hand, the nitrogen
distribution of the second gate insulating film 25C illustrated in
FIG. 9B has a broad distribution in the film with a concentration
peak of about 15 atm % near the center of the film in the thickness
direction.
[0107] The nitrogen concentrations in the first and second gate
insulating films 23C and 25C can be controlled by adjusting the
partial pressures of the oxygen gas and the nitrogen gas, which are
plasma sources. Moreover, the nitrogen concentration can be
controlled also by providing a time delay between the introduction
of the oxygen gas and the introduction of the nitrogen gas.
Basically, it is only necessary to simultaneously generate a
nitrogen plasma and an oxygen plasma. Therefore, similar effects
may be obtained using a nitrogen monoxide gas, or the like.
However, since dinitrogen monoxide is strongly oxidizing, it may
not be suitable for the formation of a thin film.
[0108] As described above, according to the second embodiment,
despite the reduced thickness of the second gate insulating film
25C, it is possible to prevent the diffusion of dopant atoms from
the gate electrode 26 into the semiconductor substrate 21 while
reducing the gate leakage current. Furthermore, the dielectric
constant of the second gate insulating film 25C increases, thereby
increasing the resistance to dielectric breakdown.
Third Embodiment
[0109] The third embodiment of the present invention will now be
described with reference to the drawings.
[0110] FIG. 10A to FIG. 10C, FIG. 11A and FIG. 11B are
cross-sectional views sequentially illustrating the steps of a
method for manufacturing a semiconductor device according to the
third embodiment of the present invention.
[0111] First, a device isolation region 32, which is a shallow
trench isolation, is formed in an upper portion of a semiconductor
substrate 31 made of P-type silicon. The device isolation region 32
partitions at least the first device forming region 51, the second
device forming region 52 and a third device forming region 53 from
one another. Then, although not shown, a P-type well region and an
N-type well region are formed in an upper portion of the
semiconductor substrate 31, and a channel doping process is
performed for adjusting the threshold voltage of the transistor.
Then, the surface of the semiconductor substrate 31, in which the
device isolation region 32 has been formed, is subjected to an SC1
cleaning process at a temperature of about 50.degree. C. and a
cleaning process with diluted hydrofluoric acid solution (HF), so
as to remove the natural oxide film from the semiconductor
substrate 31. Then, the semiconductor substrate 31 is placed into
an RTP, and the semiconductor substrate 31 is subjected to
pyrogenic oxidization in a water vapor atmosphere at a temperature
of about 1050.degree. C., thereby forming a first gate insulating
film 33A made of silicon dioxide having a thickness of about 5.5 nm
on the principal surface of the semiconductor substrate 31. Then, a
first resist pattern 34 having an opening in the second device
forming region 52 is formed on the semiconductor substrate 31 by
using a photolithography method. Then, the second device forming
region 52 is implanted with, for example, fluorine ion (F.sup.+) at
a dose of about 5.times.10.sup.14/cm.sup.2 and with an acceleration
energy of about 5 keV while using the first resist pattern 34 as a
mask, thereby obtaining a structure as illustrated in FIG. 10A.
Herein, silicon (Si) ion may be implanted instead of fluorine ion.
When using silicon (Si) ion, it is necessary to optimize the
implantation energy and the dose.
[0112] Then, as illustrated in FIG. 10B, after the first resist
pattern 34 is removed, a second resist pattern 44 covering the
first device forming region 51 is formed by a photolithography
method. Then, a portion of the first gate insulating film 33A that
is included in the second device forming region 52 and the third
device forming region 53 is removed by using a diluted hydrogen
fluoride solution and using the second resist pattern 44 as a
mask.
[0113] Then, as illustrated in FIG. 10C, after the second resist
pattern 44 is removed by a piranha cleaning process, the second
device forming region 52 of the semiconductor substrate 31 is
subjected to an SC1 cleaning process at a temperature of 50.degree.
C., as a pre-cleaning process for the formation of a second gate
insulating film and a third gate insulating film. Herein, the
cleaning process with diluted hydrofluoric acid solution is not
performed because the final process using diluted hydrofluoric acid
solution would selectively etch the first gate insulating film 33A
of the first device forming region 51 to significantly deteriorate
the dielectric breakdown reliability thereof. Then, the
semiconductor substrate 31 is placed into an RTP, and the
semiconductor substrate 31 is subjected to an oxynitriding process
in an oxynitriding atmosphere at a temperature of about 850.degree.
C., which is made of a mixed gas of about 70% by volume of nitrogen
monoxide (NO) and about 30% by volume of oxygen (O.sub.2). As a
result of the oxynitriding process, a second gate insulating film
35B made of silicon oxynitride having a thickness of about 2.2 nm
is formed on the second device forming region 52 of the
semiconductor substrate 31, while a third gate insulating film 36B
made of silicon oxynitride having a thickness of about 1.8 nm is
formed on the third device forming region 53.
[0114] As described above, fluorine ion having an enhanced
oxidization effect is implanted at a high concentration near the
surface of the second device forming region 52 of the semiconductor
substrate 31, whereby the thickness of the second gate insulating
film 35B formed in the second device forming region 52 is larger
than the thickness of the third gate insulating film 36B formed in
the third device forming region 53, into which fluorine ion is not
implanted. Moreover, nitrogen atoms are introduced by this
oxynitriding process also into the first gate insulating film 33A,
which is thus altered into a first gate insulating film 33B made of
a silicon oxynitride film. Note that there is only a slight
increase in the thickness of the first gate insulating film
33B.
[0115] The nitrogen distribution of the first gate insulating film
33B is primarily the pile-up near the interface with the
semiconductor substrate 31, whereas the nitrogen distribution of
each of the second gate insulating film 35B and the third gate
insulating film 36B has a peak near the center of the film in the
thickness direction. Moreover, the peak value of the nitrogen
concentration is about 6 atm % to about 8 atm %.
[0116] Then, as illustrated in FIG. 11A, the semiconductor
substrate 31, on which the first gate insulating film 33B, the
second gate insulating film 35B and the third gate insulating film
36B have been formed, is exposed to a nitrogen plasma, which is
produced by an inductively-coupled plasma system, for about 120
seconds. Herein, the plasma-producing frequency is about 13.56 MHz,
the high-frequency power is about 500 W, the chamber pressure is
about 1.33 Pa and the substrate temperature is about 30.degree. C.
The first gate insulating film 33B is altered by the nitridation
process using a nitrogen plasma into a first gate insulating film
33C made of a silicon oxynitride film having two peaks, i.e., one
peak near the surface and another peak near the interface with the
semiconductor substrate 31. Moreover, the second gate insulating
film 35B and the third gate insulating film 36B are altered into a
second gate insulating film 35C and a third gate insulating film
36C, respectively, each of which is made of a silicon oxynitride
film containing a high concentration of nitrogen and has a broad
concentration distribution in the insulation film with a peak value
of about 15 atm %.
[0117] Then, a conductive film made of polycrystal silicon
germanium having a thickness of about 150 nm and containing about
20 atm % of germanium (Ge) is deposited by using a CVD method at a
deposition temperature of about 550.degree. C. on the first gate
insulating film 33C, the second gate insulating film 35C and the
third gate insulating film 36C, which have been nitrided by a
nitrogen plasma. Then, the deposited conductive film is subjected
to a predetermined patterning process so as to form gate electrodes
37 from the conductive film, thereby obtaining a structure as
illustrated in FIG. 11B.
[0118] Then, although not shown, a predetermined ion implantation
process is performed for each of the first device forming region 51
and the second device forming region 52 so as to form an LDD region
and a source/drain region, thereby obtaining a transistor
structure.
[0119] As described above, according to the third embodiment, it is
possible to simultaneously form, on the semiconductor substrate 31,
the first gate insulating film 33C having a thickness of about 5.5
nm, the second gate insulating film 35C having a thickness of about
2.2 nm and the third gate insulating film 36C having a thickness of
about 1.8 nm. In addition, the second and third gate insulating
films 35C and 36C, which are made of a silicon oxynitride film
having a relatively small thickness, have a broad nitrogen
concentration profile with the nitrogen concentration peak being
located near the center of the film in the thickness direction and
the peak value being as high as about 15 atm %.
[0120] Thus, despite the reduced thicknesses of the second and
third gate insulating films 35C and 36C, it is possible to prevent
the diffusion of dopant atoms from the gate electrode 37 into the
semiconductor substrate 31 while reducing the gate leakage current.
Furthermore, the dielectric constant of each of the gate insulating
films 35C and 36C increases, thereby increasing the resistance to
dielectric breakdown.
[0121] Note that in the third embodiment, a plasma nitridation
process is performed in the step shown in FIG. 11A. Therefore, an
oxidization process may be used instead of an oxynitriding process
for forming the second gate insulating film 35B and the third gate
insulating film 36B, which are formed in the step shown in FIG.
10C. However, when the gate insulating films 35C and 36C are made
of an silicon oxide film obtained by an oxidization process, the
integrated value of the nitrogen concentration of each of the
insulation films 35C and 36C is smaller than that in a case where
it is made of silicon oxynitride. Thus, the nitrogen concentrations
of the second and third gate insulating films 35C and 36C may be
adjusted by selecting a nitridation method in view of the degree of
reduction in the gate leakage current, the degree of suppression of
the boron penetration.
[0122] Moreover, also in a case where silicon oxynitride is
selected as the composition of the second and third gate insulating
films 35B and 36B, the oxynitriding atmosphere is not limited to a
mixed atmosphere of nitrogen monoxide and oxygen, but may
alternatively be a mixed atmosphere of nitrogen monoxide and
dinitrogen monoxide or an atmosphere solely of dinitrogen monoxide.
Any suitable film formation method may be selected according to the
nitrogen concentrations of the second and third gate insulating
films 35C and 36C.
[0123] Furthermore, in the plasma nitridation step shown in FIG.
10C, the plasma source is not limited to a nitrogen gas. For
example, an oxynitriding process with a nitrogen plasma and an
oxygen plasma obtained by using a nitrogen monoxide gas may
alternatively be employed. In other words, the plasma source is not
limited to any particular plasma source as long as a nitrogen
plasma containing no hydrogen can be produced.
[0124] Moreover, while the third embodiment is directed to a method
for forming a gate insulating film having three portions of
different thicknesses in a MOS semiconductor device, a gate
insulating film having four or more portions of different
thicknesses can be formed by forming device forming regions into
which fluorine ion is implanted with different doses, and then
performing the oxynitriding process and the plasma nitridation
process as described above.
* * * * *