Semiconductor device including liner insulating film

The disclosure of Japanese Patent Application No. 2007-013747 filed on Jan. 24, 2007 including specification, drawings and claims is incorporated herein by reference in its entirety.

(1) Field of the Invention

The present invention relates to semiconductor devices and methods for fabricating the same, and more particularly relates to a semiconductor device including MIS transistors of different conductivity types.

(2) Description of Related Art

In recent years, with increases in the degree of integration, functionality and speed of semiconductor integrated circuit devices, miniaturization of transistors has been advanced. A so-called scaling law in which the current driving forces of transistors are improved with advances in miniaturization of transistors has been conventionally effective. Meanwhile, in order to achieve reduction in energy consumption and the long-time use of mobile devices, further miniaturization of transistors and reduction in power consumption have been demanded. In such a field, decreases in the sizes of transistors do not lead to improvement in the driving forces of transistors.

In order to improve the driving force of a transistor, a measure in which a gate insulating film is reduced in thickness is considered. However, a problem that a reduction of the thickness of a gate insulating film causes gate leakage current to increase occurs. To cope with this, a technique has been suggested in which the driving force of a transistor is improved by applying stress to a channel region of the transistor (see, for example, S. Ito, et al., “IEDM 2000”, 2000, p. 247).

FIG. 6 illustrates a cross-sectional structure of a known transistor having a structure in which stress is applied to a channel region of the transistor. As illustrated in FIG. 6 , a gate electrode 204 is formed on a substrate 201 with a gate insulating film 203 interposed therebetween, L-shaped first sidewalls 206 are formed on both lateral sides of the gate electrode 204 , and second sidewalls 207 are formed on both lateral sides of the gate electrode 204 with the first sidewalls 206 interposed therebetween. A liner insulating film 209 is formed to cover the gate electrode 204 and the second sidewalls 207 . The liner insulating film 209 represents a film having tensile stress. With the above-described structure, stress can be applied to the channel region under the gate electrode 204 , resulting in the driving force of the transistor improved.

However, for the known semiconductor device, stress arising from the liner insulating film 209 is not sufficiently transferred to the channel region, and thus the driving force of the transistor cannot be sufficiently improved.

Furthermore, a memory unit or any other unit requires both a P-channel transistor and an N-channel transistor. Meanwhile, the orientation of stress needed to improve the driving force of a P-channel transistor is different from that of stress needed to improve the driving force of an N-channel transistor. In view of the above, if a liner insulating film having tensile stress were formed to improve the driving force of an N-channel transistor, the driving force of a P-channel transistor would be reduced. In order to avoid the reduction in the driving force of the P-channel transistor, the step of removing a part of the liner insulating film located on a region of the substrate on which the P-channel transistor is to be formed (hereinafter, referred to as “P-channel transistor formation region), the step of forming a mask to prevent the formation of a liner insulating film on the P-channel transistor formation region, or any other step is needed.

The present invention is made to solve the above-mentioned known problems, and its object is to allow stress from a liner insulating film to be sufficiently transferred to a channel region and provide a semiconductor device which includes transistors of different conductivity types and in which the need for the step of removing a liner insulating film is eliminated.

In order to achieve the above-described object, a semiconductor device of the present invention is configured such that first MIS transistors each do not have outer sidewalls and second MIS transistors each have outer sidewalls.

More specifically, a semiconductor device of the present invention includes a first MIS transistor and a second MIS transistor both formed on a semiconductor substrate, and a liner insulating film covering the first MIS transistor and the second MIS transistor and applying stress to a channel region of the first MIS transistor along the gate length direction of the first gate electrode. The first MIS transistor includes: a first gate insulating film formed on a first area of the semiconductor substrate; a first gate electrode formed on the first gate insulating film; first L-shaped sidewalls covering both sides of the first gate electrode and part of the semiconductor substrate to each have an L-shaped cross-sectional shape; and first source/drain regions formed on both sides of a part of the first area covered with the first gate electrode and the first L-shaped sidewalls. The second MIS transistor includes: a second gate insulating film formed on a second area of the semiconductor substrate; a second gate electrode formed on the second gate insulating film; second L-shaped sidewalls covering both sides of the first gate electrode and part of the semiconductor substrate to each have an L-shaped cross-sectional shape; outer sidewalls formed on the second L-shaped sidewalls; and second source/drain regions formed on both sides of a part of the second area covered with the second gate electrode, the second L-shaped sidewalls and the outer sidewalls. No outer sidewall is formed on the first L-shaped sidewalls, and the minimum thickness of a part of the liner insulating film located on each of the second source/drain regions is larger than the minimum thickness of a part of the liner insulating film located on each of the first source/drain regions.

According to the semiconductor device of the present invention, no outer sidewall is formed on the first L-shaped sidewalls. This allows a channel region of the first MIS transistor to become closer to the liner insulating film. Therefore, stress can be efficiently applied to the channel region of the first MIS transistor. This allows stress to be more efficiently applied to the first MIS transistor than to the second MIS transistor under conditions that the first and second MIS transistors are covered with the same liner insulating film. As a result, in a semiconductor device in which MIS transistors of different conductivity types coexist, when the driving force of one of the MIS transistors has been improved, the driving force of the other MIS transistor can be prevented from being reduced.

A fabrication method for a semiconductor device of the present invention is directed toward a method for fabricating a semiconductor device comprising a first MIS transistor formed on a first area of a semiconductor substrate and a second MIS transistor formed on a second area of the semiconductor substrate. The method includes the steps of: (a) forming a first gate insulating film and a first gate electrode on the first area and forming a second gate insulating film and a second gate electrode on the second area; (b) forming first sidewalls and second sidewalls in the order away from the first gate electrode to cover both sides of the first gate electrode and forming third sidewalls and fourth sidewalls in the order away from the second gate electrode to cover both sides of the second gate electrode; (c) forming first source/drain regions on both sides of a part of the first area covered with the first gate electrode, the first sidewalls and the second sidewalls; (d) forming second source/drain regions on both sides of a part of the second area covered with the second gate electrode, the third sidewalls and the fourth sidewalls; (e) after the steps (c) and (d), removing the second sidewalls; and (f) after the step (e), forming a liner insulating film to cover the first gate electrode, the first sidewalls, the first source/drain regions, the second gate electrode, the third sidewalls, the fourth sidewalls, and the second source/drain regions, the liner insulating film applying stress to a channel region of the first MIS transistor along the gate length direction of the first gate electrode. In the step (f), the liner insulating film is formed such that the minimum thickness of a part of the liner insulating film located on each of the second source/drain regions becomes larger than the minimum thickness of a part of the liner insulating film located on each of the first source/drain regions.

According to the method of the present invention, stress from the liner insulating film is hardly applied to a channel region of the second MIS transistor. Meanwhile, stress from the liner insulating film is efficiently applied to a channel region of the first MIS transistor. In view of the above, the method of the present invention allows stress from the liner insulating film to be sufficiently transferred to a channel region, and the present invention can provide a semiconductor device which includes transistors of different conductivity types and in which the need for the step of removing a liner insulating film is eliminated.

FIGS. 1A and 1B illustrate a semiconductor device according to a first embodiment of the present invention, in which FIG. 1A is a plan view of the semiconductor device and FIG. 1B is a cross-sectional view taken along the line Ib-Ib in FIG. 1A .

FIGS. 2A through 2C are cross-sectional views illustrating fabrication process steps for the semiconductor device according to the first embodiment of the present invention.

FIGS. 3A through 3C are cross-sectional views illustrating other fabrication process steps for the semiconductor device according to the first embodiment of the present invention.

FIG. 4 is a diagram illustrating the correlation between stress and the effect of improving the driving force of the semiconductor device according to the first embodiment of the present invention.

FIG. 5 is a cross-sectional view illustrating a semiconductor device according to a second embodiment of the present invention.

FIG. 6 is a cross-sectional view illustrating a semiconductor device according to a known example.

A first embodiment of the present invention will be described with reference to the drawings. FIGS. 1A and 1B illustrate a semiconductor device according to the first embodiment, in which FIG. 1A illustrates a plan structure of the semiconductor device and FIG. 1B illustrates a cross-sectional structure thereof taken along the line Ib-Ib in FIG. 1A . As illustrated in FIGS. 1A and 1B , a first MIS (metal-insulator semiconductor) transistor 12 A serving as an N-channel (N-type) transistor is formed on a first area 11 A of a semiconductor substrate 11 made of silicon (Si), and a second MIS transistor 12 B serving as a P-channel (P-type) transistor is formed on a second area 11 B thereof.

The first MIS transistor 12 A is formed on a first active region 14 A of the first area 11 A isolated by a part of an isolation region 13 in the first area 11 A. A first gate electrode 16 A is formed on the first active region 14 A with a first gate insulating film 15 A interposed therebetween. First L-shaped sidewalls 17 A are formed to cover both lateral sides of the first gate electrode 16 A and parts of the top surface of the semiconductor substrate 11 and each have an L-shaped cross-sectional shape.

N-type first source/drain regions 22 A are formed in parts of the first active region 14 A located on both lateral sides of a part thereof covered with the first gate electrode 16 A and the first L-shaped sidewalls 17 A. N-type first extension diffusion layers 23 A are formed in parts of the first active region 14 A located under the first L-shaped sidewalls 17 A. P-type pocket regions (not shown) are formed in parts of the first active region 14 A located under the first extension diffusion layers 23 A.

The second MIS transistor 12 B is formed on a second active region 14 B of the second area 11 B isolated by a part of the isolation region 13 in the second area 11 B. A second gate electrode 16 B is formed on the second active region 14 B with a second gate insulating film 15 B interposed therebetween. Second L-shaped sidewalls 17 B are formed to cover both lateral sides of the second gate electrode 16 B and parts of the top surface of the semiconductor substrate 11 and each have an L-shaped cross-sectional shape. Outer sidewalls 18 are formed on the second L-shaped sidewalls 17 B.

P-type second source/drain regions 22 B are formed in parts of the second active region 14 B located on both lateral sides of a part thereof covered with the second gate electrode 16 B, the second L-shaped sidewalls 17 B and the outer sidewalls 18 . P-type second extension diffusion layers 23 B are formed in parts of the second active region 14 B located under the second L-shaped sidewalls 17 B. N-type pocket regions (not shown) are formed in parts of the second active region 14 B located under the second extension diffusion layers 23 B.

Silicide layers 24 are formed in respective upper parts of the first and second gate electrodes 16 A and 16 B and respective upper parts of the first and second source/drain regions 22 A and 22 B.

A liner insulating film 25 is formed to entirely cover the semiconductor substrate 11 and the first and second MIS transistors 12 A and 12 B. In this embodiment, the liner insulating film 25 represents a film having tensile stress. The tensile stress represents stress pulling, along the gate length direction, a channel region through which electrons travel.

The first and second gate electrodes 16 A and 16 B are formed to cover not only respective parts of the first active regions 14 A and 14 B but also parts of the isolation region 13 . Thus, a plurality of first gate electrodes 16 A are formed on the first area 11 A at fixed intervals w, and a plurality of second gate electrodes 16 B are formed on the second area 11 B at fixed intervals w.

The semiconductor device of this embodiment is configured in the following manner: When the thickness of each of the first and second L-shaped sidewalls 17 A and 17 B is t 1 , the thickness of each of the outer sidewalls 18 is t 2 , and the thickness of the liner insulating film 25 is t 3 , the relationship of 2×(t 1 +t 3 )<w≦2×(t 1 +t 2 +t 3 ) is satisfied. For example, assume that when the interval w between each adjacent pair of electrodes is 150 nm, the thickness t 1 of each of the first and second L-shaped sidewalls 17 A and 17 B is 10 nm, the thickness t 2 of the outer sidewall 18 is 45 mn, and the thickness t 3 of the liner insulating film 25 is 30 nm. In this case, for the first area 11 A, a recess between each adjacent pair of the first gate electrodes 16 A is partially left, and for the second area 11 B, a recess between each adjacent pair of the second gate electrodes 16 B is filled with the liner insulating film 25 . In view of the above, the minimum thickness of a part of the liner insulating film 25 located on each of the second source/drain regions 22 B becomes larger than that of a part thereof located on each of the first source/drain regions 22 A. In other words, the top surface of a part of the liner insulating film 25 located on each of the first source/drain regions 22 A and having the minimum thickness is below the top surface of the first gate electrode 16 A. Meanwhile, the top surface of a part of the liner insulating film 25 located on each of the second source/drain regions 22 B and having the minimum thickness is above the top surface of the second gate electrode 16 B.

With the above-mentioned structure, the semiconductor device of the present invention can provide the following effects. First, since, for the first MIS transistor 12 A, outer sidewalls are not formed on the first L-shaped sidewalls 17 A, the liner insulating film 25 is in contact with the first L-shaped sidewalls 17 A. Since the distance between the channel region under the first gate electrode 16 A and the liner insulating film 25 therefore becomes small, tensile stress can be efficiently applied to the channel region of the first MIS transistor 12 A along the gate length direction.

Meanwhile, for the second MIS transistor 12 B, the outer sidewalls 18 are formed between the liner insulating film 25 and the second L-shaped sidewalls 17 B. Since the distance between the channel region of the second MIS transistor 12 B and the liner insulating film 25 therefore becomes large, tensile stress applied to the channel region of the second MIS transistor 12 B along the gate length direction becomes small.

Furthermore, for the second area 11 B, the recess between each adjacent pair of second gate electrodes 16 B (the second gate electrode 16 B on the second active region 14 B and the adjacent second gate electrode 16 B on the isolation region 13 ) is filled with the liner insulating film 25 . Therefore, the liner insulating film 25 cannot pull the channel region of the second MIS transistor 12 B so that stress is hardly applied to the channel region of the second MIS transistor 12 B along the gate length direction.

As a result, tensile stress is applied to the N-type first MIS transistor 12 A, thereby improving the driving force of the first MIS transistor 12 A. However, tensile stress is not applied to the P-type second MIS transistor 12 B, thereby restraining a reduction in the driving force of the second MIS transistor 12 B.

A fabrication method for a semiconductor device according to the first embodiment will be described hereinafter with reference to the drawings. FIGS. 2 and 3 illustrate fabrication process steps for a semiconductor device according to this embodiment step by step. First, as illustrated in FIG. 2A , an isolation region 13 , such as a shallow trench isolation (STI) structure, is formed in a semiconductor substrate 11 made of Si. Then, a first active region 14 A is formed in a first area 11 A of the semiconductor substrate 11 , and a second active region 14 B is formed in a second area 11 B thereof.

Subsequently, a silicon oxide film that will be gate insulating films and a polysilicon film that will be gate electrodes are deposited on the semiconductor substrate 11 and then selectively etched, thereby forming first gate insulating films 15 A and first gate electrodes 16 A on the first area 11 A and forming second gate insulating films 15 B and second gate electrodes 16 B on the second area 11 B. In this case, recesses between each adjacent pair of the first gate electrodes 16 A and between each adjacent pair of the second gate electrodes 16 B are set to each have a fixed width.

Thereafter, N-type impurity ions are implanted into the first active region 14 A at a dose of 1×10 15 ions/cm 2 through 3×10 15 ions/cm 2 using the first gate electrodes 16 A as a mask. In this manner, first extension diffusion layers 23 A are formed in parts of the first active region 14 A located to both lateral sides of each first gate electrode 16 A to each have a depth of 100 nm or less. Then, pocket regions may be formed by implanting P-type impurity ions into parts of the first active region 14 A located under the first extension diffusion layers 23 A. Subsequently, P-type impurity ions are implanted into the second active region 14 B at a dose of 1×10 15 ions/cm 2 through 3×10 15 ions/cm 2 using the second gate electrodes 16 B as a mask. In this manner, second extension diffusion layers 23 B are formed in parts of the second active region 14 B located to both lateral sides of each second gate electrode 16 B to each have a depth of 100 nm or less. Then, pocket regions may be formed by implanting N-type impurity ions into parts of the second active region 14 B located under the second extension diffusion layers 23 B.

Next, as illustrated in FIG. 2B , a 10-nm-thick first insulating film 17 a made of silicon dioxide (SiO 2 ) is formed to cover the semiconductor substrate 11 and the first and second gate electrodes 16 A and 16 B. Subsequently, a 30- through 50-nm-thick second insulating film 18 a made of silicon nitride (SiN) is formed to cover the first insulating film 17 a.

Next, as illustrated in FIG. 2C , the first and second insulating films 17 a and 18 a are subjected to anisotropic etching. In this manner, first L-shaped sidewalls 17 A are formed to cover both lateral sides of the first gate electrodes 16 A and regions of the semiconductor substrate 11 surrounding the first gate electrodes 16 A, second L-shaped sidewalls 17 B are formed to cover both lateral sides of the second gate electrodes 16 B and regions of the semiconductor substrate 11 surrounding the second gate electrodes 16 B, and outer sidewalls 18 are formed to cover the first and second L-shaped sidewalls 17 A and 17 B.

Subsequently, N-type impurity ions are implanted into the first active region 14 A at a dose of 4×10 15 ions/cm 2 through 5×10 15 ions/cm 2 using the first gate electrodes 16 A, the first L-shaped sidewalls 17 A and the outer sidewalls 18 as masks, thereby forming first source/drain regions 22 A. P-type impurity ions are implanted into the second active region 14 B at a dose of 1×10 15 ions/cm 2 through 3×10 15 ions/cm 2 using the second gate electrodes 16 B, the second L-shaped sidewalls 17 B and the outer sidewalls 18 as masks, thereby forming second source/drain regions 22 B. Thereafter, the semiconductor substrate 11 is subjected to heat treatment at temperatures of 1000 through 1100° C., thereby activating the impurity ions.

Subsequently, respective upper parts of the first and second gate electrodes 16 A and 16 B and respective upper parts of the first and second source/drain regions 22 A and 22 B are allowed to react with a refractory metal so as to be silicided, thereby forming silicide layers 24 .

Next, as illustrated in FIG. 3A , only the first area 11 A is etched using a phosphate solution, thereby selectively removing the outer sidewalls 18 formed on the first L-shaped sidewalls 17 A. Therefore, the width w 1 of the recess between each adjacent pair of the first gate electrodes 16 A is equal to w−2×t 1 , and the width w 2 of the recess between each adjacent pair of the second gate electrodes 16 B is equal to w−2(t 1 +t 2 ). Herein, w represents the interval between each adjacent pair of the first gate electrodes 16 A or between each adjacent pair of the second gate electrodes 16 B, t 1 represents the thickness of each of the first and second L-shaped sidewalls 17 A and 17 B, and t 2 represents the thickness of each outer sidewall 18 .

Next, as illustrated in FIG. 3B , a 30- through 50-nm-thick liner insulating film 25 made of SiN is formed to cover the semiconductor substrate 11 and the first and second gate electrodes 16 A and 16 B. In forming the liner insulating film 25 , the conditions on which the liner insulating film 25 is deposited and other conditions are adjusted such that the liner insulating film 25 has tensile stress. The tensile stress represents stress pulling a channel region formed under the first gate electrode 16 A along the gate length direction.

The thickness of the liner insulating film 25 is set such that the recess between each adjacent pair of the first gate electrodes 16 A is partially left and the recess between each adjacent pair of the second gate electrodes 16 B is filled. In other words, the thickness t 3 of the liner insulating film 25 is set to satisfy w 2 /2≦t 3 ≦w 1 /2.

Next, as illustrated in FIG. 3C , an interlayer dielectric 26 made of SiO 2 is formed to cover the liner insulating film 25 . Subsequently, contact holes are formed to pass through the interlayer dielectric 26 and the liner insulating film 25 and expose the silicide layers 24 associated with the first and second source/drain regions 22 A and 22 B and then filled with a conductive material, thereby forming contact plugs 27 each connected to an associated one of the first and second source/drain regions 22 A and 22 B. Furthermore, if necessary, plugs (not shown) may be formed so as to be electrically connected to the first and second gate electrodes 16 A and 16 B, respectively, and then an interconnect layer (not shown) may be formed.

In the semiconductor device of this embodiment, the first and second gate electrodes 16 A and 16 B are preferably placed such that the gate length direction of each of the first and second gate electrodes 16 A and 16 B (the direction along which current flows from an associated source region to an associated drain region) corresponds to the <100> crystal axis direction of the semiconductor substrate 11 . The reason for this will be described below.

FIG. 4 illustrates the direction of stress applied to a channel region of each of transistors by a liner insulating film and the magnitude of the effect of improving the driving force of the transistor by virtue of the stress. In FIG. 4 , the directions of the arrows correspond to the directions of stresses, and the magnitude of each arrow corresponds to the magnitude of the effect of improving the driving force of the associated transistor.

As illustrated in FIG. 4 , the driving force of an N-type transistor is improved by applying tensile stress to the associated channel region independently of whether the gate length direction thereof corresponds to the <110> crystal axis direction or the <100> crystal axis direction. When the gate length direction corresponds to the <100> crystal axis direction, the effect of improving the driving force is more significant than when the gate length direction corresponds to the <110> crystal axis direction. In view of the above, when the gate length direction of an N-type transistor is allowed to correspond to the <100> crystal axis direction, this can further increase the effect of improving the driving force of the N-type transistor by virtue of a liner insulating film.

On the other hand, the driving force of a P-type transistor is improved by applying compressive stress to the associated channel region independently of whether the gate length direction thereof corresponds to the <110> crystal axis direction or the <100> crystal axis direction. Therefore, when tensile stress is applied to a channel region of a P-type transistor, the driving force thereof is reduced. However, when the gate length direction corresponds to the <100> crystal axis direction, variations in the driving force of a P-type transistor due to stress are hardly caused. In view of the above, when the gate length direction of a P-type transistor is allowed to correspond to the <100> crystal axis direction, this hardly reduces the driving force of the P-type transistor even with a liner insulating film having tensile stress.

The semiconductor device of this embodiment is configured such that the recess between each adjacent pair of the second gate electrodes 16 B is filled with the liner insulating film 25 . Thus, stress from the liner insulating film 25 is hardly applied to the second MIS transistor 12 B. Therefore, the gate length direction of the second gate electrode 16 B does not always need to correspond to the <100> crystal axis direction.

In order to allow the gate length direction to correspond to the <100> crystal axis direction, the first and second gate electrodes 16 A and 16 B need to be patterned with a wafer rotated 45° as compared with the normal case where the gate length direction corresponds to the <110> crystal axis direction.

In this embodiment, a description was given of the case where the conductivity type of a first MIS transistor 12 A is N, the conductivity type of a second MIS transistor 12 B is P, and a liner insulating film 25 has tensile stress. Meanwhile, in a case where the conductivity type of the first MIS transistor 12 A is P, the conductivity type of the second MIS transistor 12 B is N, and the liner insulating film 25 has compressive stress, this can also provide the same effect. In this case, when compressive stress is applied to a channel region of the P-type MIS transistor along the gate length direction thereof, this improves the driving force thereof. Meanwhile, since compressive stress is hardly applied to a channel region of the N-type MIS transistor along the gate length direction thereof, this prevents the driving force thereof from being reduced. The compressive stress in this embodiment means stress compressing a channel region along the gate length direction.

A second embodiment of the present invention will be described hereinafter with reference to the drawings. FIG. 5 illustrates a cross-sectional structure of a semiconductor device according to the second embodiment. In FIG. 5 , the same components as in FIG. 1B are denoted by the same reference numerals as in FIG. 1B , and thus the descriptions thereof are omitted.

As illustrated in FIG. 5 , a first MIS transistor 12 A has I-shaped sidewalls 19 formed between an associated first gate electrode 16 A and associated first L-shaped sidewalls 17 A, and a second MIS transistor 12 B has I-shaped sidewalls 19 formed between an associated second gate electrode 16 B and associated second L-shaped sidewalls 17 B. The I-shaped sidewalls 19 are made of, for example, 5- through 10-nm-thick SiO 2 .

When the I-shaped sidewalls 19 are formed as described above, respective edge parts of the first and second gate electrodes 16 A and 16 B can be protected. This can restrain the leakage current from being generated from the edge parts of the first and second gate electrodes 16 A and 16 B.

In the first and second embodiments, a liner insulating film 25 having tensile stress or a liner insulating film 25 having compressive stress is formed by a known method. A SiN film containing hydrogen is formed, for example, by a plasma chemical vapor deposition (P-CVD) method, and then a SiN film having tensile stress can be obtained by removing hydrogen from the formed SiN film containing hydrogen using ultraviolet irradiation or any other method.

Furthermore, in the first and second embodiments, a description was given of the structure in which a first gate electrode 16 A of another first MIS transistor adjacent to the first MIS transistor 12 A and a second gate electrode 16 B of another second MIS transistor adjacent to the second MIS transistor 12 B cover also parts of the isolation region 13 . However, components covering parts of the isolation region 13 do not always need to be the first and second gate electrodes 16 A and 16 B, and projections need only be formed to project outward. The projections may be formed of, for example, interconnects, dummy interconnects, or dummy electrodes. Alternatively, the projections may be a combination of interconnects, dummy interconnects and dummy electrodes. In this case, the liner insulating film 25 is formed in the following manner: The recess between the first gate electrode 16 A on the first active region 14 A and an adjacent one of the projections on the isolation region 13 is partially left, and the recess between the second gate electrode 16 B on the second active region 14 B and an adjacent one of the projections on the isolation region 13 is filled.

An example in which the interval between each adjacent pair of first gate electrodes is equal to the interval between each adjacent pair of second gate electrodes was described. However, as long as the recess between each adjacent pair of the first gate electrodes can be partially left and the recess between each adjacent pair of the second gate electrodes can be filled, the interval between each adjacent pair of the first gate electrodes may be different from the interval between each adjacent pair of the second gate electrodes.

In the first and second embodiments, silicide layers made of a reaction product of a refractory metal and silicon are formed in upper parts of the gate electrodes made of polysilicon. Alternatively, so-called fully silicided gate electrodes may be formed by siliciding the whole gate electrodes. Alternatively, the gate electrodes may be formed of metal films.

In the first and second embodiments, an example in which the gate insulating film also covers parts of the isolation region 13 was described. However, when a gate insulating film is formed by thermal oxidation or any other method, no gate insulating film covers parts of the isolation region 13 . This structure has no problem. The gate insulating film is not limited to a silicon oxide film, and a high-dielectric-constant insulating film, such as a silicon oxynitride film, HfO 2 , HfSiO 2 , HfSiON, or HfAlO x , may be used as the gate insulating film.

The semiconductor device of each of the first and second embodiments may be any kind of semiconductor device as long as it includes an N-type transistor and a P-type transistor. For example, it may be a static random access memory (SRAM).

As described above, the semiconductor device of the present invention allows stress from a liner insulating film to be sufficiently transferred to a channel region of a transistor, and the present invention can provide a semiconductor device which includes transistors of different conductivity types and in which the need for the step of removing a liner insulating film is eliminated.

The description of the embodiments of the present invention is given above for the understanding of the present invention. It will be understood that the invention is not limited to the particular embodiments described herein, but is capable of various modifications, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, it is intended that the following claims cover all such modifications and changes as fall within the true spirit and scope of the invention.