Semiconductor device and semiconductor device manufacturing method

Abstract

In the present invention, there is provided a semiconductor device including: element isolation regions formed in a state of being buried in a semiconductor substrate such that an element formation region of the semiconductor substrate is interposed between the element isolation regions; a gate electrode formed on the element formation region with an gate insulating film interposed between the gate electrode and the element formation region, the gate electrode being formed so as to cross the element formation region; and source-drain regions formed in the element formation region on both sides of the gate electrode, wherein a channel region made of the element formation region under the gate electrode is formed so as to project from the element isolation regions, and the source-drain regions are formed to a position deeper than surfaces of the element isolation regions.

Claims

.[.1. A semiconductor device comprising: semiconductor substrate with an element formation region projecting from a base portion of said substrate and extending in a first direction along said base portion; element isolation regions buried in said semiconductor substrate such that said element formation region of said semiconductor substrate is flanked by said element isolation regions, said element isolation regions having upper surfaces; a gate electrode on said element formation region with a gate insulating film interposed between said gate electrode and said element formation region, said gate electrode crossing said element formation region in a second direction; and source-drain regions formed in said element formation region on both sides of said gate electrode, wherein, said element formation region includes a channel region under said gate electrode, said source-drain regions are positioned within said element formation region so as to extend below said upper surfaces of said element isolation regions, said element isolation regions include depressions in said upper surfaces which extend in said second direction and which flank said channel region, said gate electrode extends into said depressions in said upper surfaces of said element isolation regions and has an upper surface positioned above bottom surfaces of said depressions, and said channel region has an upper surface positioned above said bottom surfaces of said depressions..].

.[.2. The semiconductor device according to claim 1, wherein surfaces of said source-drain regions are at one of a position equal in height to a surface of said semiconductor substrate and a position higher than the surface of said semiconductor substrate..].

.[.3. The semiconductor device according to claim 1, comprising a stress applying film which applies stress to said channel region..].

.[.4. The semiconductor device according to claim 1, comprising a stress applying insulating film covering said gate electrode which applies stress to said channel region..].

.[.5. The semiconductor device of claim 1, further comprising a silicide layer on said element formation region..].

.[.6. The semiconductor device of claim 1, wherein a difference between said channel region and said bottom surfaces of said depressions is from 3 nm to 30 nm, inclusive..].

.[.7. The semiconductor device of claim 1, wherein said element formation region includes SiGe regions in which said source and drain are located..].

.[.8. The semiconductor device of claim 3, wherein said stress applying layer is an SiN film..].

.[.9. The semiconductor device of claim 4, wherein said stress applying layer is an SiN film..].

.Iadd.10. A semiconductor device comprising: a semiconductor substrate having: an element formation region that, in a perspective view of the semiconductor device, extends in a first direction and a second direction, wherein a first extension of the element formation region in the first direction is greater than a second extension of the element formation region in the second direction, a silicon channel region of the element formation region, and silicon germanium source-drain regions of the element formation region; a gate electrode configured to extend, in the perspective view, along the first direction and the second direction, wherein a first extension of the gate electrode in the first direction is less than a second extension of the gate electrode in the second direction, so that: the element formation region crosses the gate electrode, the gate electrode is on the element formation region, and the silicon channel region is between the silicon germanium source-drain regions and under the gate electrode; element isolation regions that, in the perspective view, are: buried in the semiconductor substrate, and between the gate electrode and the semiconductor substrate; and a gate insulating film that, in a vertical sectional view of the semiconductor device, is: between the gate electrode and the silicon channel region, and between the gate electrode and the element isolation regions, wherein: the first direction is orthogonal to the second direction, the element formation region protrudes from a base portion of the semiconductor substrate so that the element formation region, in the vertical sectional view, is between the element isolation regions, the silicon germanium source-drain regions are on both sides of the gate electrode and extend below upper surfaces of the element isolation regions, depressions into the upper surfaces of the element isolation regions flank the silicon channel region, the gate electrode extends into the depressions so that, in the vertical sectional view, an upper surface of the gate electrode is above bottom surfaces of the depressions, and junction positions of the silicon germanium source-drain regions are deeper than surfaces of the element isolation regions..Iaddend.

.Iadd.11. The semiconductor device of claim 10, wherein the element isolation regions are formed of an insulating thin film..Iaddend.

.Iadd.12. The semiconductor device of claim 10, wherein the semiconductor substrate is silicon..Iaddend.

.Iadd.13. The semiconductor device of claim 10, wherein surfaces of the silicon germanium source-drain regions are at one of a position equal in height to a surface of the semiconductor substrate and a position higher than the surface of the semiconductor substrate..Iaddend.

.Iadd.14. The semiconductor device of claim 10, further comprising: a stress applying film that is configured to apply stress to the silicon channel region..Iaddend.

.Iadd.15. The semiconductor device of claim 14, wherein the stress applying film is an SiN film..Iaddend.

.Iadd.16. The semiconductor device of claim 10, further comprising: a stress applying insulating film covering the gate electrode in a manner that applies stress to the silicon channel region..Iaddend.

.Iadd.17. The semiconductor device of claim 16, wherein the stress applying insulating film is an SiN film..Iaddend.

.Iadd.18. The semiconductor device of claim 10, further comprising: a silicide layer on the element formation region..Iaddend.

.Iadd.19. The semiconductor device of claim 10, wherein a difference between the silicon channel region and the bottom surfaces of the depressions is from 3 nm to 30 nm, inclusive..Iaddend.

.Iadd.20. The semiconductor device of claim 10, wherein an upper surface of the silicon channel region, in the vertical sectional view, is above the bottom surfaces of the depressions..Iaddend.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a schematic perspective view schematically showing an embodiment (first embodiment) of a semiconductor device according to the present invention;

(2) FIG. 2 is a vertical sectional view including a line A-A;

(3) FIG. 3 is a vertical sectional view including a line B-B;

(4) FIG. 4 is a vertical sectional view including a line .[.C-C.]. .Iadd.D-D.Iaddend.;

(5) FIG. 5 is a vertical sectional view including a line .[.D-D.]. .Iadd.C-C.Iaddend.;

(6) FIGS. 6A to 6P are manufacturing process sectional views of one embodiment (first embodiment) of a semiconductor device manufacturing method according to the present invention;

(7) FIGS. 7A and 7B are schematic configuration sectional views of an embodiment (second embodiment) of a semiconductor device according to the present invention;

(8) FIGS. 8A and 8B are schematic configuration sectional views of an embodiment (third embodiment) of a semiconductor device according to the present invention;

(9) FIGS. 9A to 9O are manufacturing process sectional views of one embodiment (second embodiment) of a semiconductor device manufacturing method according to the present invention;

(10) FIGS. 10A and 10B are schematic plan layout diagrams of assistance in explaining stresses applied to an active region of a transistor;

(11) FIG. 11 is a schematic configuration sectional view showing a section in a gate width direction of a two-dimensional type (planar) MOS transistor in related art;

(12) FIG. 12 is a schematic perspective view schematically showing a three-dimensional structure of an ordinary MOS transistor; and

(13) FIG. 13 is a vertical sectional view including a line A-A of FIG. 12.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

(14) A preferred embodiment (first embodiment) of a semiconductor device according to the present invention will be described with reference to a schematic perspective view of FIG. 1, a vertical sectional view of FIG. 2 including a line A-A, a vertical sectional view of FIG. 3 including a line B-B, a vertical sectional view of FIG. 4 including a line .[.C-C.]. .Iadd.D-D.Iaddend., and a vertical sectional view of FIG. 5 including a line .[.D-D.]. .Iadd.C-C.Iaddend.. Incidentally, FIG. 1 is a diagram showing a general constitution, in which diagram a part of constituent parts are not shown. In FIGS. 3 and 5, an interlayer insulating film is not shown.

(15) Referring mainly to FIGS. 1 and 2, description will be made in the following of the constitution of a semiconductor device 1.

(16) Element isolation regions 13 for electrically isolating an element formation region 12 in which a transistor is formed are formed in a semiconductor substrate 11. A silicon substrate, for example, is used as the semiconductor substrate 11, and an ordinary STI (Shallow Trench Isolation) structure is used for the element isolation regions 13. Thus, a part of the semiconductor substrate 11 which part is interposed between the element isolation regions 13 is the element formation region 12.

(17) Depressions 15 are formed in upper parts of the element isolation regions 13 formed on both sides of a channel region 14 formed in the element formation region 12 so that the channel region 14 projects from the surfaces of the element isolation regions 13 (see also the vertical sectional view of FIG. 3 including the line B-B and the vertical sectional view of FIG. 4 including the line .[.C-C.]. .Iadd.D-D.Iaddend.). Thus, the depressions 15 are formed in the element isolation regions 13 on both sides of the channel region 14 such that only the channel region 14 projects from the element isolation regions 13. In this case, an amount of projection of the channel region 14 from the surfaces of the element isolation regions 13 at the bottom of the depressions 15 is set at 3 nm to 30 nm inclusive, for example. (And channel width is set at 0.5 m or less). An effective stress is thereby applied in a direction of gate width of the channel region 14 (X-direction). Incidentally, when the amount of projection is less than 3 nm, the channel region 14 is affected by stress in the element isolation regions 13, and thus an effect of forming the channel region 14 in a projecting manner by forming the depressions 15 cannot be obtained. When the channel region 14 is projected higher than 30 nm, the release of the stress in the element isolation regions 13 which stress acts on a lower part of the channel region 14 becomes saturated, and thus further stress does not occur in the channel region 14.

(18) As shown in the vertical sectional view of FIG. 3 including the line B-B, in the channel region 14, effects of direct stress from the element isolation regions 13 are suppressed. That is, in the channel region 14, stress (indicated by arrows B) occurs in a direction of releasing stress (indicated by arrows A) in a part of the semiconductor substrate 11 as the lower part of the channel region 14, which part is directly affected by the stress from the element isolation regions 13. The amount of projection of the channel region 14 can be adjusted by controlling the depth of the depressions 15. Incidentally, the surface of source-drain regions 27 and 28 including the above-described channel region 14 is formed at substantially the same height as the surface of the semiconductor substrate 11 in another region, for example.

(19) It is known that an insulating film (for example a high density plasma (HDP) silicon oxide or the like) buried to form the element isolation regions 13 of the STI structure generally has compressive stress. The stress applied in the direction of gate width of the channel region 14 acts in a direction of degrading mobility. With the constitution of the above-described embodiment of the present invention, effects of the direct compressive stress from the element isolation regions 13 on the channel region 14 are suppressed, and tensile stress occurs in the channel region 14 in a direction of releasing the compressive stress in the part of the semiconductor substrate 11 as the lower part of the channel region 14, which part is directly affected by the compressive stress from the element isolation regions 13. Thus, in the direction of gate width of the channel region 14, the tensile stress acts in a direction of improving the mobility of the transistor.

(20) A gate electrode 22 is formed on the semiconductor substrate 11 with a gate insulating film 21 interposed between the gate electrode 22 and the semiconductor substrate 11. The gate electrode 22 is formed in such a manner as to extend over the channel region 14 and the depressions 15, for example. A high dielectric constant (High-k) film, for example, can be used as the gate insulating film 21. The high dielectric constant film includes for example a nitrided hafnium silicate (HfSiON) film, a hafnium nitride or oxide or oxynitride film, and an aluminum nitride or oxide or oxynitride film. Incidentally, an ordinary silicon oxide film can also be used as the gate insulating film 21. A single-layer structure of a metal or a metallic compound for a metal gate or a laminated structure, for example, can be used for the gate electrode 22. Incidentally, polysilicon can also be used for the gate electrode 22. A silicon nitride film, for example, is used as a hard mask 53.

(21) Offset spacers 23 are formed on the side walls of the gate electrode 22 (including the gate insulating film 21). The offset spacers 23 are formed by an insulating thin film of about 1 nm to 10 nm, for example. An insulating film having etching selectivity with respect to the element isolation regions 13, for example, is used as the insulating thin film, and the insulating thin film is formed of silicon nitride (SiN), for example.

(22) Extension regions 24 and 25 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the offset spacers 23 interposed between the extension regions 24 and 25 and the gate electrode 22. For the extension regions 24 and 25, an n-type impurity such as arsenic (As.sup.+), phosphorus (P.sup.+) or the like is used when an NMOS transistor is formed, for example, and a p-type impurity such as boron (B.sup.+), indium (In.sup.+) or the like is used when a PMOS transistor is formed. The extension regions 24 and 25 are formed with a shallow junction.

(23) Further, side wall spacers 26 are formed on both sides of the gate electrode 22 with the offset spacers 23 interposed between the side wall spacers 26 and the gate electrode 22. The source-drain regions 27 and 28 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the extension regions 24 and 25 interposed between the source-drain regions 27 and 28 and the gate electrode 22. Resistance lowering layers 31 and 32 are formed on the source-drain regions 27 and 28. The resistance lowering layers 31 and 32 are for example formed of cobalt (Co), nickel (Ni), platinum (Pt) or a compound thereof. The compound includes a metal silicide of these metals. As is also shown in the vertical sectional view of FIG. 5 including the line .[.D-D.]. .Iadd.C-C.Iaddend., the source-drain regions 27 and 28 are formed to a position deeper than the surface of the parts of the element isolation regions 13 formed on both sides (direction of gate width) of the source-drain regions 27 and 28. Thus, even when the resistance lowering layers 31 and 32 (the resistance lowering layer 32 is not shown in FIG. 5) are formed by a salicide process, for example, on the surfaces of the source-drain regions 27 and 28 (the source-drain region 28 is not shown in FIG. 5), the resistance lowering layers 31 and 32 do not come close to the semiconductor substrate 11 or are not connected to the semiconductor substrate 11. This prevents a leak of current from the resistance lowering layers 31 and 32 to the semiconductor substrate 11.

(24) Further, an interlayer insulating film 41 is formed over the semiconductor substrate 11 in such a manner as to cover the semiconductor device 1 of the above-described constitution formed on the semiconductor substrate 11. Incidentally, though not shown in the figures, contact parts bonded to the gate electrode 22 and the source-drain regions 27 and 28, wiring connected to each of the contact parts, and the like are formed in the interlayer insulating film 41.

(25) The semiconductor device 1 of the above-described constitution has an advantage in that it is possible to produce stress in the channel region 14 directly under the gate electrode 22 in a gate width direction advantageous to a transistor characteristic (carrier mobility), and thus improve on current Ion of the transistor, so that the performance of the transistor is improved. In addition, because the junction positions of the source-drain regions 27 and 28 are deeper than the surfaces of the element isolation regions, even when the resistance lowering layers 31 and 32 made of a silicide layer for lower resistance are formed on the surfaces of the source-drain regions 27 and 28, a leak of current between the resistance lowering layers 31 and 32 and the semiconductor substrate 11 does not occur. The reliability of the semiconductor device (transistor) 1 is thereby improved. The above-described effects become greater as the semiconductor device (transistor) 1 is reduced in gate width.

(26) Further, as shown in FIGS. 10A and 10B described above, the more the stress in the y-direction originally applied to the channel region, the greater the effect of improvement of the on current Ion. The greater effect can be obtained by applying a stress in the y-direction by forming the source-drain regions of a silicon germanium epitaxial layer or providing a covering of a compressive stress liner for a pMOSFET or by applying a stress in the y-direction by forming the source-drain regions of a silicon carbide epitaxial layer or providing a covering of a tensile stress liner for an nMOSFET. These constitutions will be described later as a second embodiment and a third embodiment of a semiconductor device according to the present invention.

(27) An embodiment (first embodiment) of a semiconductor device manufacturing method according to the present invention will next be described with reference to manufacturing process sectional views of FIGS. 6A to 6P. This manufacturing method will be described as an example of manufacturing the constitution of the semiconductor device 1. Incidentally, FIGS. 6A, 6K, 6L, and 6P are sections in a so-called gate width direction (sections at a position corresponding to the vertical section including the line B-B in FIG. 1 described above), and FIGS. 6B to 6J and 6M to 6O are sections in a so-called gate length direction (sections at a position corresponding to the vertical section including the line A-A in FIG. 1 described above).

(28) As shown in FIG. 6A, element isolation regions 13 for electrically isolating an element formation region 12 in which a transistor is formed are formed in a semiconductor substrate 11. A silicon substrate, for example, is used as the semiconductor substrate 11, and an ordinary STI (Shallow Trench Isolation) structure is used for the element isolation regions 13.

(29) Next, as shown in FIG. 6B, a protective film (not shown) for preventing channeling when performing ion implantation for introducing an impurity into the semiconductor substrate 11 is formed on the semiconductor substrate 11. This protective film is formed by a silicon oxide (SiO.sub.2) film, for example. As an example, a method of forming the protective film is carried out by oxidizing the surface of the semiconductor substrate 11.

(30) Next, impurity injection for performing transistor element isolation and threshold value adjustment is performed by ion implantation. After the ion implantation, the silicon oxide (SiO.sub.2) film previously formed as the ion implantation protective film is removed to expose the surface of the semiconductor substrate 11.

(31) Next, a dummy gate insulating film 51, a dummy gate 52, and a hard mask 53 are formed in order on the semiconductor substrate 11.

(32) The dummy gate insulating film 51, for example an oxide film is first formed with a thickness of about 1 nm to 3 nm on the semiconductor substrate 11. A dummy gate forming film is next formed on the dummy gate insulating film 51. This dummy gate forming film is formed by depositing polycrystalline silicon (polysilicon) to a thickness of about 100 nm to 200 nm, for example. A chemical vapor deposition (CVD) method, for example, is used as a method of forming the dummy gate forming film. Further, a hard mask layer is formed by a silicon nitride film, for example, on the dummy gate forming film. This silicon nitride film is formed to a thickness of 30 nm to 100 nm, for example, by using a chemical vapor deposition (CVD) method, for example.

(33) Next, a resist film (not shown) for lithography is formed on the hard mask layer. A resist suited to an exposure source is used for the resist film. Then, the resist film is exposed to light to form a dummy gate pattern (not shown), and thereafter, with the dummy gate pattern used as an etching mask, the hard mask layer is etched to form a hard mask 53. Optical lithography using for example KrF, ArF, F.sub.2 or the like for a light source or electron beam lithography is used for the light exposure in the above-described lithography. In the etching of the hard mask layer, the hard mask layer can be processed into a line width smaller than that of the pattern of the resist (for example by sliming or trimming) to reduce gate length. Next, the dummy gate pattern formed by the resist film is removed, and with the hard mask 53 formed by the etching process as an etching mask, the dummy gate forming film is processed by dry etching to form a dummy gate 52. The line width of the dummy gate 52 at this time is set at a few nm to a few ten nm. In this etching, the dummy gate insulating film 51 is also etched.

(34) Next, as shown in FIG. 6C, offset spacers 23 are formed on side walls of the dummy gate 52 (a part formed by the dummy gate insulating film 51, the dummy gate 52, and the hard mask 53 will hereinafter be referred to as a dummy gate). The offset spacers 23 are made by forming an insulating thin film of about 1 nm to 10 nm, for example, so as to cover the dummy gate 52, and then etching back the insulating thin film such that the insulating thin film remains only on the side walls of the dummy gate 52. The insulating thin film forming the offset spacers 23 is formed by silicon oxide (SiO.sub.2) or silicon nitride (SiN), for example. The insulating thin film on the semiconductor substrate 11 is removed by the above-described etching back.

(35) Next, extension regions 24 and 25 are formed in the semiconductor substrate 11 on both sides of the dummy gate 52 with the offset spacers 23 interposed between the extension regions 24 and 25 and the dummy gate 52. The extension regions 24 and 25 are formed by ion implantation, for example. An n-type impurity such as arsenic (As.sup.+), phosphorus (P.sup.+) or the like is used when an NMOS transistor is formed, for example, and a p-type impurity such as boron (B.sup.+), indium (In.sup.+) or the like is used when a PMOS transistor is formed. For example, the implantation is performed with a low acceleration energy (100 eV to 300 eV) and a dose of 510.sup.14 (/cm.sup.2) to 210.sup.15 (/cm.sup.2), whereby the extension regions 24 and 25 are formed with a shallow junction.

(36) Next, as shown in FIG. 6D, side wall spacers 26 are formed on both sides of the dummy gate 52 with the offset spacers 23 interposed between the side wall spacers 26 and the dummy gate 52. At the time of etching back for forming the side wall spacers 26, the etching is performed such that the hard mask 53 remains.

(37) Next, as shown in FIG. 6E, source-drain regions 27 and 28 are formed in the semiconductor substrate 11 on both sides of the dummy gate 52 with the extension regions 24 and 25 respectively interposed between the source-drain regions 27 and 28 and the dummy gate 52. Thereafter, an activation process is performed by rapid thermal annealing (RTA) at about 1000 C., for example.

(38) Incidentally, when a PMOS transistor and an NMOS transistor are formed on the semiconductor substrate 11, impurity injection for adjusting the threshold value of the transistors, ion implantation for forming the extension regions 24 and 25, and ion implantation for forming the source-drain regions 27 and 28 are each performed for an NMOS region and a PMOS region separately. For example, a first mask is formed in the NMOS region, ion implantation is performed in the PMOS region, and then the first mask is removed. Next, a second mask is formed in the PMOS region, and ion implantation is performed in the NMOS region. The second mask is thereafter removed.

(39) Next, as shown in FIG. 6F, resistance lowering layers 31 and 32 are formed on the source-drain regions 27 and 28. The resistance lowering layers 31 and 32 are selectively formed on the surfaces of the source-drain regions 27 and 28 by a salicide process. The resistance lowering layers 31 and 32 are for example formed of cobalt (Co), nickel (Ni), platinum (Pt) or a compound thereof. The compound includes a metal silicide of these metals.

(40) Next, as shown in FIG. 6G, a first interlayer insulating film 42 covering the dummy gate 52 and the like is formed.

(41) Next, as shown in FIG. 6H, an upper part of the first interlayer insulating film 42 is removed to expose the surface of the hard mask 53. A chemical mechanical polishing (CMP) method, for example, is used to remove the upper part of the first interlayer insulating film 42. Incidentally, another polishing method may be used, and the upper part of the first interlayer insulating film 42 can be removed by etching back. Incidentally, the chemical mechanical polishing method can planarize the polished surface.

(42) Next, the hard mask 53 and the dummy gate 52 are removed. Dry etching, for example, is used for the removing process. Dry etching damage to the semiconductor substrate 11 is prevented by leaving the dummy gate insulating film 51 in the vapor dry etching. Then the dummy gate insulating film 51 is removed. Wet etching, for example, is used for the removing process. Etching damage to the semiconductor substrate 11 is prevented by performing the removing process by wet etching. As a result, as shown in FIG. 6I, a groove 29 enclosed by the offset spacers 23 is formed. As shown in the sectional view of FIG. 6K in the gate width direction, the channel region 14 formed in the semiconductor substrate 11 directly under the region where the dummy gate 52 is formed is at substantially the same height as the surfaces of the element isolation regions 13. The stress of the element isolation regions 13 is applied to the channel region 14. Thus, the channel region 14 is directly affected by the stress of the element isolation regions 13.

(43) As shown in the sectional view of FIG. 6L in the gate width direction, the above-described wet etching can lower the surfaces of the element isolation regions 13 under the region where the dummy gate 52 was formed by forming depressions 15. Thereby the height of the element isolation regions 13 under the region where the dummy gate 52 was formed can be made lower than that of the surface of the semiconductor substrate 11 (channel region 14). The height of the element isolation regions 13 can be controlled by an amount of wet etching, and an amount of projection of the channel region 14 from the surfaces of the element isolation regions 13 is a height of 3 nm to 30 nm inclusive, for example. The above-described wet etching does not lower the height of the element isolation regions 13 on both sides (in the gate width direction) of the source-drain regions 27 and 28. In addition, the source-drain regions 27 and 28 including the above-described channel region 14 can be maintained at substantially the same height as the semiconductor substrate 11 in another region, for example.

(44) Next, as shown in FIG. 6J, a gate insulating film 21 is formed on the semiconductor substrate 11 within the groove 29. This gate insulating film 21 is actually formed on the inner surfaces of the groove 29 and the surface of the first interlayer insulating film 42. The gate insulating film 21 can be formed by an insulating film such as a high dielectric constant (High-k) film or a silicon oxide film, for example. In this case, the gate insulating film 21 is formed by a high dielectric constant film as an example. In this case, heat treatment (annealing treatment) for modifying the gate insulating film 21 is performed.

(45) Next, as shown in FIG. 6M, a gate electrode forming film 61 is formed on the gate insulating film 21 in such a manner as to fill the inside of the groove 29. This gate electrode forming film 61 is formed by a laminate of metals or metallic compounds for a metal gate or a single-layer structure, for example.

(46) Next, as shown in FIG. 6N, a surplus part of the gate electrode forming film 61 is removed, and a gate electrode 22 is formed by the gate electrode forming film 61 on the semiconductor substrate 11 within the groove 29 with the gate insulating film 21 interposed between the gate electrode 22 and the semiconductor substrate 11. The removing process is performed by a chemical mechanical polishing (CMP) method, for example.

(47) Next, as shown in FIG. 6O, a second interlayer insulating film 43 is formed on the first interlayer insulating film 42 in such a manner as to cover the gate electrode 22. The first interlayer insulating film 42 and the second interlayer insulating film 43 form an interlayer insulating film 41. As shown in the sectional view of FIG. 6P in the gate width direction at this time, stress in the channel region 14 at this time is maintained in a stress state that occurred when removing the top parts of the element isolation regions 13 under the region where the dummy gate 52 had been formed as described above with reference to FIG. 6L.

(48) Though not shown, contact parts electrically connected to the respective source-drain regions 27 and 28, metallic wiring routed to the contact parts, and the like are thereafter formed in the interlayer insulating film 41. The semiconductor device is thereby completed.

(49) The semiconductor device manufacturing method according to the first embodiment has an advantage in that it is possible to produce stress in the channel region 14 directly under the gate electrode 22 in a gate width direction advantageous to a transistor characteristic (carrier mobility), and thus improve on current Ion of the transistor, so that the performance of the transistor is improved. In addition, because the junction positions of the source-drain regions 27 and 28 are deeper than the surfaces of the element isolation regions 13, even when the resistance lowering layers 31 and 32 made of a silicide layer for lower resistance are formed on the surfaces of the source-drain regions 27 and 28, a leak of current between the resistance lowering layers 31 and 32 and the semiconductor substrate 11 does not occur. The reliability of the semiconductor device (transistor) 1 is thereby improved. The above-described effects become greater as the semiconductor device (transistor) 1 is reduced in gate width.

(50) An embodiment (second embodiment) of a semiconductor device according to the present invention will next be described with reference to a schematic configuration sectional view of FIGS. 7A and 7B. FIG. 7A shows a section in a gate length direction, and FIG. 7B shows a section in a gate width direction. The semiconductor device 2 shown in FIGS. 7A and 7B is formed by using a stress applying layer applying stress to the channel region 14 as the source-drain regions 27 and 28 in the semiconductor device 1 according to the first embodiment.

(51) Specifically, as shown in FIGS. 7A and 7B, element isolation regions 13 for electrically isolating an element formation region 12 in which a transistor is formed are formed in a semiconductor substrate 11. A silicon substrate, for example, is used as the semiconductor substrate 11, and an ordinary STI (Shallow Trench Isolation) structure is used for the element isolation regions 13. Thus, a part of the semiconductor substrate 11 which part is interposed between the element isolation regions (not shown) is the element formation region 12.

(52) Depressions 15 are formed in upper parts of the element isolation regions 13 formed on both sides of a channel region 14 formed in the element formation region 12 so that the channel region 14 projects from the surfaces of the element isolation regions 13. Thus, the depressions 15 are formed in the element isolation regions 13 on both sides of the channel region 14 such that only the channel region 14 projects from the element isolation regions 13. In this case, as in the first embodiment, an amount of projection of the channel region 14 from the surfaces of the element isolation regions 13 at the bottom of the depressions 15 is set at 3 nm to 30 nm inclusive, for example. Thereby, effects of direct stress from the element isolation regions 13 are suppressed in the channel region 14. That is, in the channel region 14, stress (indicated by arrows B) occurs in a direction of releasing stress (indicated by arrows A) in a part of the semiconductor substrate 11 as the lower part of the channel region 14, which part is directly affected by the stress from the element isolation regions 13. The amount of projection of the channel region 14 can be adjusted by controlling the depth of the depressions 15.

(53) A gate electrode 22 is formed on the semiconductor substrate 11 with a gate insulating film 21 interposed between the gate electrode 22 and the semiconductor substrate 11. The gate electrode 22 is formed in such a manner as to extend over the channel region 14 and the depressions 15, for example. As the gate insulating film 21, for example, a high dielectric constant (High-k) film can be used, or an ordinary silicon oxide film can also be used. A single-layer structure of a metal or a metallic compound for a metal gate or a laminated structure, for example, can be used for the gate electrode 22. Incidentally, polysilicon can also be used for the gate electrode 22. A silicon nitride film, for example, is used as a hard mask 53.

(54) Offset spacers 23 are formed on the side walls of the gate electrode 22 (including the gate insulating film 21). The offset spacers 23 are formed by an insulating thin film of about 1 nm to 10 nm, for example. An insulating film having etching selectivity with respect to the element isolation regions 13, for example, is used as the insulating thin film, and the insulating thin film is formed of silicon nitride (SiN), for example.

(55) Extension regions 24 and 25 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the offset spacers 23 interposed between the extension regions 24 and 25 and the gate electrode 22. For the extension regions 24 and 25, an n-type impurity such as arsenic (As.sup.+), phosphorus (P.sup.+) or the like is used when an NMOS transistor is formed, for example, and a p-type impurity such as boron (B.sup.+), indium (In.sup.+) or the like is used when a PMOS transistor is formed. The extension regions 24 and 25 are formed with a shallow junction.

(56) Further, side wall spacers 26 are formed on both sides of the gate electrode 22 with the offset spacers 23 interposed between the side wall spacers 26 and the gate electrode 22. Source-drain regions 27 and 28 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the extension regions 24 and 25 interposed between the source-drain regions 27 and 28 and the gate electrode 22. The source-drain regions 27 and 28 are formed by a stress applying layer applying stress to the channel region 14 between the source-drain regions 27 and 28. For example, when the semiconductor device 2 is a p-type FET (field-effect transistor), the source-drain regions 27 and 28 are formed by a silicon germanium layer grown by epitaxial growth, and apply compressive stress to the channel region 14. When the semiconductor device 2 is an n-type FET (field-effect transistor), the source-drain regions 27 and 28 are formed by a silicon carbide layer grown by epitaxial growth, and apply tensile stress to the channel region 14. In either case, it is effective to form the source-drain regions 27 and 28 into an embedded source-drain structure raised from the surface of the semiconductor substrate 11. Resistance lowering layers 31 and 32 are formed on the source-drain regions 27 and 28. The resistance lowering layers 31 and 32 are for example formed of cobalt (Co), nickel (Ni), platinum (Pt) or a compound thereof. The compound includes a metal silicide of these metals.

(57) The source-drain regions 27 and 28 are formed to a position deeper than the surface of the parts of the element isolation regions 13 formed on both sides (direction of gate width) of the source-drain regions 27 and 28. Thus, even when the resistance lowering layers 31 and 32 are formed by a salicide process, for example, on the surfaces of the source-drain regions 27 and 28, the resistance lowering layers 31 and 32 do not come close to the semiconductor substrate 11 or are not connected to the semiconductor substrate 11. This prevents a leak of current from the resistance lowering layers 31 and 32 to the semiconductor substrate 11.

(58) Further, an interlayer insulating film 41 is formed over the semiconductor substrate 11 in such a manner as to cover the semiconductor device 2 of the above-described constitution formed on the semiconductor substrate 11. Incidentally, though not shown in the figures, contact parts bonded to the gate electrode 22 and the source-drain regions 27 and 28, wiring connected to each of the contact parts, and the like are formed in the interlayer insulating film 41.

(59) The semiconductor device 2 provides similar actions and effects to those of the semiconductor device 1, and applies stress effective to improve mobility also from the source-drain regions 27 and 28 to the channel region 14. The semiconductor device 2 therefore improves mobility more than the semiconductor device 1.

(60) An embodiment (third embodiment) of a semiconductor device according to the present invention will next be described with reference to a schematic configuration sectional view of FIGS. 8A and 8B. FIG. 8A shows a section in a gate length direction, and FIG. 8B shows a section in a gate width direction. The semiconductor device 3 shown in FIGS. 8A and 8B is obtained by forming a stress liner film for applying stress to the channel region 14 in the semiconductor device 1 according to the first embodiment.

(61) Specifically, as shown in FIGS. 8A and 8B, element isolation regions 13 for electrically isolating an element formation region 12 in which a transistor is formed are formed in a semiconductor substrate 11. A silicon substrate, for example, is used as the semiconductor substrate 11, and an ordinary STI (Shallow Trench Isolation) structure is used for the element isolation regions 13. Thus, a part of the semiconductor substrate 11 which part is interposed between the element isolation regions (not shown) is the element formation region 12.

(62) Depressions 15 are formed in upper parts of the element isolation regions 13 formed on both sides of a channel region 14 formed in the element formation region 12 so that the channel region 14 projects from the surfaces of the element isolation regions 13. Thus, the depressions 15 are formed in the element isolation regions 13 on both sides of the channel region 14 such that only the channel region 14 projects from the element isolation regions 13. In this case, as in the first embodiment, an amount of projection of the channel region 14 from the surfaces of the element isolation regions 13 at the bottom of the depressions 15 is set at 3 nm to 30 nm inclusive, for example. Thereby, effects of direct stress from the element isolation regions 13 are suppressed in the channel region 14. That is, in the channel region 14, stress (indicated by arrows B) occurs in a direction of releasing stress (indicated by arrows A) in a part of the semiconductor substrate 11 as the lower part of the channel region 14, which part is directly affected by the stress from the element isolation regions 13. The amount of projection of the channel region 14 can be adjusted by controlling the depth of the depressions 15.

(63) A gate electrode 22 is formed on the semiconductor substrate 11 with a gate insulating film 21 interposed between the gate electrode 22 and the semiconductor substrate 11. The gate electrode 22 is formed in such a manner as to extend over the channel region 14 and the depressions 15, for example. As the gate insulating film 21, for example, a high dielectric constant (High-k) film can be used, or an ordinary silicon oxide film can also be used. A single-layer structure of a metal or a metallic compound for a metal gate or a laminated structure, for example, can be used for the gate electrode 22. Incidentally, polysilicon can also be used for the gate electrode 22. A silicon nitride film, for example, is used as a hard mask 53.

(64) Offset spacers 23 are formed on the side walls of the gate electrode 22 (including the gate insulating film 21). The offset spacers 23 are formed by an insulating thin film of about 1 nm to 10 nm, for example. An insulating film having etching selectivity with respect to the element isolation regions 13, for example, is used as the insulating thin film, and the insulating thin film is formed of silicon nitride (SiN), for example.

(65) Extension regions 24 and 25 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the offset spacers 23 interposed between the extension regions 24 and 25 and the gate electrode 22. For the extension regions 24 and 25, an n-type impurity such as arsenic (As.sup.+), phosphorus (P.sup.+) or the like is used when an NMOS transistor is formed, for example, and a p-type impurity such as boron (B.sup.+), indium (In.sup.+) or the like is used when a PMOS transistor is formed. The extension regions 24 and 25 are formed with a shallow junction.

(66) Further, side wall spacers 26 are formed on both sides of the gate electrode 22 with the offset spacers 23 interposed between the side wall spacers 26 and the gate electrode 22. Source-drain regions 27 and 28 are formed in the semiconductor substrate 11 on both sides of the gate electrode 22 with the extension regions 24 and 25 interposed between the source-drain regions 27 and 28 and the gate electrode 22. Resistance lowering layers 31 and 32 are formed on the source-drain regions 27 and 28. The resistance lowering layers 31 and 32 are for example formed of cobalt (Co), nickel (Ni), platinum (Pt) or a compound thereof. The compound includes a metal silicide of these metals.

(67) The source-drain regions 27 and 28 are formed to a position deeper than the surface of the parts of the element isolation regions 13 formed on both sides (direction of gate width) of the source-drain regions 27 and 28. Thus, even when the resistance lowering layers 31 and 32 are formed by a salicide process, for example, on the surfaces of the source-drain regions 27 and 28, the resistance lowering layers 31 and 32 do not come close to the semiconductor substrate 11 or are not connected to the semiconductor substrate 11. This prevents a leak of current from the resistance lowering layers 31 and 32 to the semiconductor substrate 11.

(68) Further, a stress liner film 71 for applying stress to the channel region 14 is formed in such a manner as to cover the semiconductor device 3 of the above-described constitution formed on the semiconductor substrate 11. This stress liner film 71 is formed by a silicon nitride film, for example, and is made by a plasma CVD method, for example. By changing a condition for forming the film, it is possible to form a silicon nitride film having tensile stress or form a silicon nitride film having compressive stress. For example, when the semiconductor device 3 is a p-type FET (field-effect transistor), a compressive stress liner film is used as the stress liner film 71 to apply compressive stress to the channel region 14. When the semiconductor device 3 is an n-type FET (field-effect transistor), a tensile stress liner film is used as the stress liner film 71 to apply tensile stress to the channel region 14.

(69) Further, an interlayer insulating film 41 is formed. Incidentally, though not shown in the figures, contact parts bonded to the gate electrode 22 and the source-drain regions 27 and 28, wiring connected to each of the contact parts, and the like are formed in the interlayer insulating film 41.

(70) In order to form the semiconductor device 3, in the manufacturing method according to the first embodiment, the gate electrode 22 is formed on the gate insulating film 21 within the groove 29, and thereafter the first interlayer insulating film 42 is removed. Next, the stress liner film 71 is formed so as to cover the gate electrode 22 and the side wall spacers 26. Next, the first interlayer insulating film 42 is formed again, and further the second interlayer insulating film 43 is formed. It is desirable to planarize the surface of the re-formed first interlayer insulating film 42.

(71) The semiconductor device 3 provides similar actions and effects to those of the semiconductor device 1, and applies stress effective to improve mobility also from the stress liner film 71 to the channel region 14. The semiconductor device 3 therefore improves mobility more than the semiconductor device 1.

(72) In addition, a stress liner film 71 similar to the stress liner film 71 used in the semiconductor device 3 according to the third embodiment can be formed in the semiconductor device 2 according to the second embodiment. In other words, source-drain regions 27 and 28 made of a stress applying layer similar to the stress applying layer used in the semiconductor device 2 according to the second embodiment can be formed in the semiconductor device 3 according to the third embodiment.

(73) An embodiment (second embodiment) of a semiconductor device manufacturing method according to the present invention will next be described with reference to manufacturing process sectional views of FIGS. 9A to 9O. This manufacturing method will be described as a method of manufacturing the constitution of the semiconductor device 2.

(74) As described above with reference to FIG. 6A, element isolation regions (not shown) for electrically isolating an element formation region 12 in which a transistor is formed are formed in a semiconductor substrate 11. A silicon substrate, for example, is used as the semiconductor substrate 11, and an ordinary STI (Shallow Trench Isolation) structure is used for the element isolation regions 13.

(75) Next, as shown in FIG. 9A, a protective film (not shown) for preventing channeling when performing ion implantation for introducing an impurity into the semiconductor substrate 11 is formed on the semiconductor substrate 11. This protective film is formed by a silicon oxide (SiO.sub.2) film, for example. As an example, a method of forming the protective film is carried out by oxidizing the surface of the semiconductor substrate 11.

(76) Next, impurity injection for performing transistor element isolation and threshold value adjustment is performed by ion implantation. After the ion implantation, the silicon oxide (SiO.sub.2) film previously formed as the ion implantation protective film is removed to expose the surface of the semiconductor substrate 11.

(77) Next, a dummy gate insulating film 51, a dummy gate 52, and a hard mask 53 are formed in order on the semiconductor substrate 11.

(78) The dummy gate insulating film 51, for example an oxide film is first formed with a thickness of about 1 nm to 3 nm on the semiconductor substrate 11. A dummy gate forming film is next formed on the dummy gate insulating film 51. This dummy gate forming film is formed by depositing polycrystalline silicon (polysilicon) to a thickness of about 100 nm to 200 nm, for example. A chemical vapor deposition (CVD) method, for example, is used as a method of forming the dummy gate forming film. Further, a hard mask layer is formed by a silicon nitride film, for example, on the dummy gate forming film. This silicon nitride film is formed to a thickness of 30 nm to 100 nm, for example, by using a chemical vapor deposition (CVD) method, for example.

(79) Next, a resist film (not shown) for lithography is formed on the hard mask layer. A resist suited to an exposure source is used for the resist film. Then, the resist film is exposed to light to form a dummy gate pattern (not shown), and thereafter, with the dummy gate pattern used as an etching mask, the hard mask layer is etched to form a hard mask 53. Optical lithography using for example KrF, ArF, F.sub.2 or the like for a light source or electron beam lithography is used for the light exposure in the above-described lithography. In the etching of the hard mask layer, the hard mask layer can be processed into a line width smaller than that of the pattern of the resist (for example by sliming or trimming) to reduce gate length. Next, the dummy gate pattern formed by the resist film is removed, and with the hard mask 53 formed by the etching process as an etching mask, the dummy gate forming film is processed by dry etching to form a dummy gate 52. The line width of the dummy gate 52 at this time is set at a few nm to a few ten nm. In this etching, the dummy gate insulating film 51 is also etched.

(80) Next, as shown in FIG. 9B, offset spacers 23 are formed on side walls of the dummy gate 52 (a part formed by the dummy gate insulating film 51, the dummy gate 52, and the hard mask 53 will hereinafter be referred to as a dummy gate). The offset spacers 23 are made by forming an insulating thin film of about 1 nm to 10 nm, for example, so as to cover the dummy gate 52, and then etching back the insulating thin film such that the insulating thin film remains only on the side walls of the dummy gate 52. The insulating thin film forming the offset spacers 23 is formed by silicon oxide (SiO.sub.2) or silicon nitride (SiN), for example. The insulating thin film on the semiconductor substrate 11 is removed by the above-described etching back.

(81) Next, as shown in FIG. 9C, dummy side walls 55 are formed on both sides of the dummy gate 52 with the offset spacers 23 interposed between the dummy side walls 55 and the dummy gate 52. At the time of etching back for forming the dummy side walls 55, the etching is performed such that the hard mask 53 remains.

(82) Next, as shown in FIG. 9D, grooves 33 and 34 are formed for source-drain regions 27 and 28 in the semiconductor substrate 11 on both sides of the dummy gate 52. Thereafter, an activation process is performed by rapid thermal annealing (RTA) at about 1000 C., for example.

(83) Next, as shown in FIG. 9E, the source-drain regions 27 and 28 are formed by growing a stress applying layer in the grooves 33 and 34 by an epitaxial growth method. For example, when the semiconductor device 2 is a p-type FET (field-effect transistor), the source-drain regions 27 and 28 are formed by a silicon germanium layer grown by epitaxial growth, and apply compressive stress to the channel region 14. As an example of conditions for forming the film, a processing temperature (substrate temperature) is set at 650 C. to 750 C., and the pressure of a growth atmosphere is set at 6.7 kPa to 13.3 kPa. Dichlorosilane (SiCl.sub.2H.sub.2: DCS), for example, is used as a silicon raw material gas, and germane (GeH.sub.4) is used as a germanium raw material gas. When the semiconductor device 2 is an n-type FET (field-effect transistor), the source-drain regions 27 and 28 are formed by a silicon carbide layer grown by epitaxial growth, and apply tensile stress to the channel region 14. In either case, it is effective to form the source-drain regions 27 and 28 into an embedded source-drain structure raised from the surface of the semiconductor substrate 11.

(84) Next, the dummy side walls 55 are removed to expose the semiconductor substrate 11 between the dummy gate 52 and the source-drain regions 27 and 28, as shown in FIG. 9F.

(85) Next, as shown in FIG. 9G, extension regions 24 and 25 are formed in the semiconductor substrate 11 on both sides of the dummy gate 52 with the offset spacers 23 interposed between the extension regions 24 and 25 and the dummy gate 52. The extension regions 24 and 25 are formed by ion implantation, for example. An n-type impurity such as arsenic (As.sup.+), phosphorus (P.sup.+) or the like is used when an NMOS transistor is formed, for example, and a p-type impurity such as boron (B.sup.+), indium (In.sup.+) or the like is used when a PMOS transistor is formed. For example, the implantation is performed with a low acceleration energy (100 eV to 300 eV) and a dose of 510.sup.14 (/cm.sup.2) to 210.sup.15 (/cm.sup.2) whereby the extension regions 24 and 25 are formed with a shallow junction. Thus, the source-drain regions 27 and 28 are formed in the semiconductor substrate 11 on both sides of the dummy gate 52 with the extension regions 24 and 25 interposed between the dummy gate 52 and the source-drain regions 27 and 28.

(86) Next, as shown in FIG. 9H, side wall spacers 26 are formed on both sides of the dummy gate 52 with the offset spacers 23 interposed between the side wall spacers 26 and the dummy gate 52. At this time, the side wall spacers 26 cover the surfaces of the extension regions 24 and 25. At the time of etching back for forming the side wall spacers 26, the etching is performed such that the hard mask 53 remains.

(87) Next, resistance lowering layers 31 and 32 are formed on the source-drain regions 27 and 28. The resistance lowering layers 31 and 32 are selectively formed on the surfaces of the source-drain regions 27 and 28 by a salicide process. The resistance lowering layers 31 and 32 are for example formed of cobalt (Co), nickel (Ni), platinum (Pt) or a compound thereof. The compound includes a metal silicide of these metals.

(88) Next, as shown in FIG. 9I, a first interlayer insulating film 42 covering the dummy gate 52, the side wall spacers 26, the resistance lowering layers 31 and 32, and the like is formed.

(89) Next, as shown in FIG. 9J, an upper part of the first interlayer insulating film 42 is removed to expose the surface of the hard mask 53. A chemical mechanical polishing (CMP) method, for example, is used to remove the upper part of the first interlayer insulating film 42. Incidentally, another polishing method may be used, and the upper part of the first interlayer insulating film 42 can be removed by etching back. Incidentally, the chemical mechanical polishing method can planarize the polished surface.

(90) Next, the hard mask 53 and the dummy gate 52 are removed. Dry etching, for example, is used for the removing process. Dry etching damage to the semiconductor substrate 11 is prevented by leaving the dummy gate insulating film 51 in the dry etching. Then the dummy gate insulating film 51 is removed. Wet etching, for example, is used for the removing process. Etching damage to the semiconductor substrate 11 is prevented by performing the removing process by wet etching. As a result, as shown in FIG. 9K, a groove 29 enclosed by the offset spacers 23 is formed. As shown in the sectional view of FIG. 6K in the gate width direction, the channel region 14 formed in the semiconductor substrate 11 directly under the region where the dummy gate 52 is formed is at substantially the same height as the surfaces of the element isolation regions 13. The stress of the element isolation regions 13 is applied to the channel region 14. Thus, the channel region 14 is directly affected by the stress of the element isolation regions 13.

(91) As shown in the sectional view of FIG. 6L in the gate width direction, the above-described wet etching can lower the surfaces of the element isolation regions 13 under the region where the dummy gate 52 was formed by forming depressions 15. Thereby the height of the element isolation regions 13 under the region where the dummy gate 52 was formed can be made lower than that of the surface of the semiconductor substrate 11 (channel region 14). The height of the element isolation regions 13 can be controlled by an amount of wet etching, and an amount of projection of the channel region 14 from the surfaces of the element isolation regions 13 is a height of 3 nm to 30 nm inclusive, for example. The above-described wet etching does not lower the height of the element isolation regions 13 on both sides (in the gate width direction) of the source-drain regions 27 and 28. In addition, the source-drain regions 27 and 28 including the above-described channel region 14 can be maintained at substantially the same height as the semiconductor substrate 11 in another region, for example.

(92) Next, as shown in FIG. 9L, a gate insulating film 21 is formed on the semiconductor substrate 11 within the groove 29. This gate insulating film 21 is actually formed on the inner surfaces of the groove 29 and the surface of the first interlayer insulating film 42. The gate insulating film 21 can be formed by an insulating film such as a high dielectric constant (High-k) film or a silicon oxide film, for example. In this case, the gate insulating film 21 is formed by a high dielectric constant film as an example. In this case, heat treatment (annealing treatment) for modifying the gate insulating film 21 is performed.

(93) Next, as shown in FIG. 9M, a gate electrode forming film 61 is formed on the gate insulating film 21 in such a manner as to fill the inside of the groove 29. This gate electrode forming film 61 is formed by a laminate of metals or metallic compounds for a metal gate or a single-layer structure, for example.

(94) Next, as shown in FIG. 9N, a surplus part of the gate electrode forming film 61 is removed, and a gate electrode 22 is formed by the gate electrode forming film 61 on the semiconductor substrate 11 within the groove 29 with the gate insulating film 21 interposed between the gate electrode 22 and the semiconductor substrate 11. The removing process is performed by a chemical mechanical polishing (CMP) method, for example.

(95) Next, as shown in FIG. 9O, a second interlayer insulating film 43 is formed on the first interlayer insulating film 42 in such a manner as to cover the gate electrode 22. The first interlayer insulating film 42 and the second interlayer insulating film 43 form an interlayer insulating film 41. Stress in the channel region 14 at this time is maintained in a stress state that occurred when removing the top parts of the element isolation regions 13 under the region where the dummy gate 52 (see FIG. 9J described above) had been formed.

(96) Though not shown, contact parts electrically connected to the respective source-drain regions 27 and 28, metallic wiring routed to the contact parts, and the like are thereafter formed in the interlayer insulating film 41. The semiconductor device is thereby completed.

(97) The above-described semiconductor device manufacturing method provides similar actions and effects to those of the first embodiment of the semiconductor device manufacturing method, and applies stress effective to improve mobility also from the source-drain regions 27 and 28 to the channel region 14. The semiconductor device 2 therefore improves mobility more than the semiconductor device formed by the semiconductor device manufacturing method according to the first embodiment.

(98) In addition, in each of the foregoing embodiments, a film having stress can be used for the gate electrode 22. For example, when applied to the gate electrode 22, in the case of an n-type MOSFET (field-effect transistor), hafnium, hafnium silicide, tantalum, tantalum silicide or the like may be used to apply tensile stress in the gate length direction of the channel region 14. In the case of a p-type MOSFET (field-effect transistor), titanium, titanium nitride, ruthenium, tungsten or the like may be used to apply compressive stress to the channel region 14. The use of such films can further improve mobility.

(99) It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and alterations may occur depending on design requirements and other factors insofar as they are within the scope of the appended claims or the equivalents thereof.