JP5360119B2 - Semiconductor device and manufacturing method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of manufacturing a semiconductor device capable of preventing variations in characteristics of an MOS transistor in which stress is applied on channels. <P>SOLUTION: A manufacturing method of a semiconductor device includes the steps of: forming a gate insulating film on a semiconductor substrate 10, forming a gate electrode 14c on the gate insulating film, forming side walls 15a, 15b on the sides of the gate electrode 14c, forming, after forming the side walls 15a, 15b, holes 10a, 10b in the semiconductor substrate 10 next to the gate electrode 14c using an organic alkali solution or TMAH as an etchant solution and forming source/drain material layers 18a, 18b in the holes 10a, 10b. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a semiconductor device, a manufacturing method thereof, and a semiconductor device evaluation method.

  In recent years, semiconductor devices such as LSIs have been miniaturized, but the performance of MOS transistors due to miniaturization is approaching its limit. Attempts have been made to improve performance. As one of such attempts, there is a method of increasing carrier mobility by applying an appropriate stress to the channel region of the MOS transistor. Although there are various ways of applying stress, in Non-Patent Document 1, a recess is formed in the silicon substrate on both sides of the gate electrode, and a SiGe layer serving as a source / drain is epitaxially grown in the recess, and silicon and SiGe are formed. Distortion is introduced into the channel using the difference in lattice constant. According to Non-Patent Document 1, this structure has a remarkable effect that the drive current of the P-type MOS transistor is improved by 10% or more.

In addition to Non-Patent Document 1, techniques related to the present invention are also disclosed in Patent Documents 1 to 4.
JP 58-35938 A JP 4-180633 JP 7-50293 A International Publication No. 98/40909 Pamphlet T. Ghani et al. "A 90nm High Volume Manufacturing Logic Technology Featuring Novel 45nm Gate Length Strained Silicon CMOS Transistors", IEDM Tech Dig., Pp.978-980, (2003)

  By the way, in the structure disclosed in Non-Patent Document 1, stress is applied from the SiGe layer to the channel as described above, but the magnitude of the stress is not uniform in the gate width direction or varies from transistor to transistor. Therefore, this transistor cannot be mass-produced and widely spread.

  Further, not only the MOS transistor disclosed in Non-Patent Document 1, but also a normal MOS transistor in which a recess for a SiGe layer is not formed on a silicon substrate, a test MOS transistor is manufactured when developing a new device or the like. The characteristics are evaluated. Among the characteristics, the carrier distribution in the channel greatly affects the performance of the transistor. Therefore, it is preferable to directly measure the carrier distribution, but no measurement method has been established so far.

  An object of the present invention is to provide a semiconductor device capable of preventing variation in characteristics of a MOS transistor in which stress is applied to a channel, and a manufacturing method thereof.

  Another object of the present invention is to provide a semiconductor device evaluation method capable of directly measuring a carrier distribution in a channel of a MOS transistor.

According to one aspect of the present invention, a semiconductor substrate, a gate insulating film and a gate electrode formed in this order on the semiconductor substrate, formed in the holes in the side of the semiconductor substrate of the gate electrode, the semiconductor substrate A source / drain material layer having a lattice constant different from the lattice constant of the semiconductor substrate, and the side surface of the hole near the gate electrode is composed of two crystal planes of the semiconductor substrate, and the cross-sectional shape of the side surface is convex A semiconductor device is provided which is characterized by the following.

  According to such a semiconductor device, since the side surface of the hole in which the source / drain material layer is formed is configured by the crystal plane of the semiconductor substrate, Patent Document 1 in which the side surface of the hole is configured by a curved surface that is not the crystal plane. As compared with the above, stress is stably applied to the channel under the gate electrode, and variations in characteristics of the MOS transistor among elements are suppressed.

  In addition, by forming the side surface of the hole with the two crystal planes of the semiconductor substrate and making the cross-sectional shape of the side surface convex, the source / drain material layer formed in such a cross-sectional shape hole On the other hand, a strong stress is generated from the lower surface toward the channel, while the stress becomes weak in the vicinity of the convex apex, and a favorable stress for improving the performance of the MOS transistor can be obtained.

According to another aspect of the present invention, a step of forming a gate insulating film on a semiconductor substrate, a step of forming a gate electrode on the gate insulating film, and a sidewall on the side surface of the gate electrode A step of forming a hole in the semiconductor substrate next to the gate electrode using an organic alkaline solution or TMAH as an etching solution after forming the sidewall, and a lattice constant of the semiconductor substrate in the hole. Forming a source / drain material layer having a lattice constant different from the step of forming a first conductivity type impurity diffusion region in the semiconductor substrate before the step of forming the hole, and the semiconductor Forming a second conductivity type impurity diffusion region deeper than the first conductivity type impurity diffusion region in the substrate, and forming the hole in the step of forming the hole. The method of manufacturing a semiconductor device, characterized by deeper than the object diffusion region is provided.

  According to this method for manufacturing a semiconductor device, since the hole is formed in the semiconductor substrate with the organic alkali solution or the TMAH solution, the crystal plane of the semiconductor substrate appears on the etched surface, and the side surface of the hole is constituted by the crystal plane. The Therefore, compared to the case where the side surface of the hole is configured with a curved surface as in Patent Document 1, the reproducibility of the hole shape is improved, and even when the MOS transistors are integrated on the semiconductor substrate, the hole is formed in the hole. Further, stress is applied from the source / drain material layer toward the channel without variation between elements.

  Further, when forming the hole, the gate electrode may also be etched to reduce its thickness. In that case, a refractory metal layer is formed on the thinned gate electrode, and the refractory metal layer is heated to react with the gate electrode, whereby the entire gate electrode is silicided. Such a gate electrode is called a metal gate. According to the above, the metal gate formation process and the hole formation process are compatible.

  When there is no need to etch the gate electrode in this way, a p-type impurity having an effect of delaying the etching with the TMAH solution or the organic alkaline solution may be introduced into the gate electrode in advance.

  Further, the first conductivity type impurity diffusion region and the second conductivity type impurity diffusion region deeper than the first conductivity type impurity diffusion region may be formed in the silicon substrate, and the hole may be formed deeper than the first conductivity type impurity diffusion region. In this case, the etching rate when forming a hole differs for each impurity diffusion region due to a difference in impurity concentration in the first and second conductivity type impurity diffusion regions and a difference in conductivity type. A plurality of crystal planes appear on the side surface of.

  For example, if the first conductivity type impurity diffusion region is p-type and the second conductivity type impurity diffusion region is n-type, the side surface of the hole is composed of two crystal planes, and the cross-sectional shape of the side surface is It becomes a concave shape bent at the crystal plane.

  On the other hand, when the first conductivity type impurity diffusion region is made p-type and a p-type impurity having a higher concentration is introduced into the second conductivity type impurity diffusion region, the side surface of the hole is constituted by two crystal surfaces, The cross-sectional shape becomes a convex shape bent with these two crystal planes as a boundary.

  Furthermore, an SOI substrate may be used as the semiconductor substrate. When using an SOI substrate, when forming holes by etching with a TMAH solution or an organic alkaline solution, the etching rate decreases near the buried insulating film that constitutes the SOI substrate, and the etching rate varies depending on the depth of the substrate. Become. Thereby, a plurality of crystal planes appear on the side surface of the hole during etching, and the side surface of the hole is constituted by these crystal planes.

  When a silicon substrate is used as the semiconductor substrate, the surface orientation of the surface of the silicon substrate is set to (110), and the gate width direction is set to the [111] direction, so that (111 ) Surface constitutes the side of the hole.

  On the other hand, by setting the plane orientation of the silicon substrate to the same (110) as above and setting the gate width direction to [100], the inclination of the (111) plane viewed from the surface of the silicon substrate becomes gentle, and the gentle (111) ) The side of the hole is configured by the surface.

Further, according to another aspect of the present invention, a step of forming a gate insulating film on a semiconductor substrate, a step of forming a gate electrode on the gate insulating film, and a side wall on the side surface of the gate electrode. Forming a hole in the semiconductor substrate next to the gate electrode using an organic alkaline solution or TMAH as an etchant after forming the sidewall, and forming the hole in the semiconductor substrate. Forming a source / drain material layer having a lattice constant different from the lattice constant, and the side surface of the hole near the gate electrode is composed of two crystal planes of the semiconductor substrate, and a cross section of the side surface A method for manufacturing a semiconductor device, characterized in that the shape is convex .

  According to the semiconductor device of the present invention, since the side surface of the hole in which the source / drain material layer is formed is configured by the crystal plane of the semiconductor substrate, the stress applied from the source / drain material layer toward the channel is reduced. Variations between elements can be prevented, and the reliability of the semiconductor device can be improved.

  Further, according to the method for manufacturing a semiconductor device according to the present invention, a hole is formed in the silicon substrate next to the gate electrode by wet etching using an organic alkaline solution or a TMAH solution as an etching solution. The crystal face appears and the side face of the hole is constituted by the crystal face. Thereby, reproducibility is improved, and even when MOS transistors are integrated and formed on a semiconductor substrate, stress is applied from the source / drain material layer formed in the hole toward the channel without variation between elements.

FIG. 1 is a cross-sectional view of a sample used to investigate the etching selectivity of a TMAH solution for silicon and silicon dioxide in the first embodiment of the present invention. FIG. 2 is a graph obtained by investigating the etching rate of the TMAH solution with respect to silicon and silicon dioxide in the first embodiment of the present invention. FIG. 3 is a graph obtained by investigating the etching rate of the organic alkaline solution with respect to silicon and silicon dioxide in the first embodiment of the present invention. FIG. 4 is a graph obtained by examining the impurity concentration dependency of the etching rate of the TMAH solution in the first embodiment of the present invention. 5A to 5D are cross-sectional views (part 1) of the semiconductor device manufacturing method according to the second embodiment of the present invention in the middle of manufacturing. 6A to 6C are cross-sectional views (part 2) of the semiconductor device manufacturing method according to the second embodiment of the present invention. FIG. 7 is a diagram depicting a recess formed according to the second embodiment of the present invention and drawn based on the SEM image of the recess. FIG. 8 shows an SEM obtained in the second embodiment of the present invention when the depth d of the recess is increased by adjusting the impurity concentration of the source / drain extension and the source / drain region. It is the figure drawn based on the image. FIGS. 9A to 9C are cross-sectional views of the semiconductor device according to the third embodiment of the present invention being manufactured. 10A to 10C are cross-sectional views of the semiconductor device according to the fourth embodiment of the present invention during manufacture. FIG. 11 is a view formed by forming a recess according to the fourth embodiment of the present invention and based on the SEM image of the recess. 12A to 12D are cross-sectional views (part 1) in the middle of manufacturing the semiconductor device according to the fifth embodiment of the present invention. FIG. 13 is a sectional view (No. 2) in the middle of manufacturing the semiconductor device according to the fifth embodiment of the present invention. FIGS. 14A to 14D are cross-sectional views in the middle of manufacturing a semiconductor device according to the sixth embodiment of the present invention. FIGS. 15A to 15D are cross-sectional views (part 1) in the middle of manufacturing the semiconductor device according to the seventh embodiment of the present invention. FIG. 16: is sectional drawing (the 2) in the middle of manufacture of the semiconductor device which concerns on 7th Embodiment of this invention. FIGS. 17A and 17B are cross-sectional views of the semiconductor device according to the eighth embodiment of the present invention in the middle of manufacture. FIG. 18 is a plan view in the middle of manufacturing the semiconductor device according to the eighth embodiment of the present invention. FIG. 19 is a cross-sectional view of a semiconductor device according to a ninth embodiment of the present invention that is being manufactured. FIG. 20 is a plan view in the middle of manufacturing the semiconductor device according to the ninth embodiment of the present invention. 21A to 21D are cross-sectional views (part 1) in the middle of manufacturing a TEG used in the semiconductor device evaluation method according to the tenth embodiment of the present invention. FIG. 22 is a cross-sectional view (No. 2) in the middle of manufacturing the TEG used in the semiconductor device evaluation method according to the tenth embodiment of the present invention. FIG. 23 is a perspective view for explaining a semiconductor device evaluation method according to the tenth embodiment of the present invention. FIGS. 24A and 24B are diagrams in which TEGs used in the tenth embodiment of the present invention are actually measured with a scanning tunneling microscope and drawn based on the concavo-convex images obtained thereby. FIG. 25 is a diagram obtained by actually measuring the carrier distribution of the TEG used in the tenth embodiment of the present invention. FIG. 26 is a cross-sectional view of a TEG used in the semiconductor device evaluation method according to the eleventh embodiment of the present invention. FIG. 27 is a perspective view for explaining a semiconductor device evaluation method according to an eleventh embodiment of the present invention.

  The best mode for carrying out the present invention will be described below in detail with reference to the accompanying drawings.

(1) First embodiment
The recess of the silicon substrate for growing the SiGe layer can also be formed by wet etching using KOH or boiling nitric acid as an etching solution. However, if these etching solutions are used, as shown in FIG. 1 of Patent Document 1, the side surface of the recess becomes a gently curved surface, so that it becomes difficult to control the shape of the recess. Therefore, the shape of the side surface of the recess varies from element to element, and the characteristics of the MOS transistor may vary from element to element.

  If dry etching is used instead of the above wet etching, the surface of the recess is damaged by the plasma, and there is a risk that lattice defects may enter the SiGe layer epitaxially grown on the recess.

  In view of these points, the present inventors have sought an etching solution that can replace KOH and boiling nitric acid, and found that a TMAH (tetramethylammonium hydride) solution can be a suitable etching solution for forming the recess. It was. Furthermore, it has been found that an organic alkali solution obtained by mixing an alkali solution, an alcohol, and water is also suitable as the etching solution.

  Therefore, an experiment conducted by the present inventor in order to investigate the etching characteristics of the TMAH and the organic alkali solution will be described below.

(A) Etch selectivity FIG. 1 is a cross-sectional view of a sample used to investigate the etch selectivity of a TMAH solution to silicon and silicon dioxide. Of these samples, sample S1 was produced as follows.

First, a silicon dioxide layer 2 is formed on a silicon substrate 1 by a plasma CVD (Chemical Vapor Deposition) method using silane (SIH ₄ ), and then LPCVD (Low Pressure CVD) using silane as a reaction gas. A polysilicon layer 3 was formed to a thickness of 100 nm. Thereafter, the surface of the polysilicon layer 3 was exposed to nitric acid and oxidized to form an oxide film 4 having a thickness of about 1.0 nm.

  On the other hand, sample S2 was formed by forming polysilicon layer 3 in the same manner as sample S1, and then exposing the surface to hydrofluoric acid to terminate hydrogen, and oxide film 4 of sample S1 was not formed.

  Thereafter, a TMAH solution having a volume concentration of 5 to 30% obtained by dissolving TMAH in pure water was prepared, and the above samples S1 and S2 were wet-etched with this TMAH solution. And after performing this etching for a predetermined time, each film thickness of the natural oxide film 4 of sample S1 and the polysilicon layer 3 of sample S2 is measured with a film thickness meter, and the measurement result is an initial film of these films. The amount of etching was estimated by comparing with the thickness. The result is shown in FIG.

  The horizontal axis of FIG. 2 shows the etching time with the TMAH solution, and the vertical axis shows the thickness of the polysilicon layer 3 after etching.

  As is apparent from FIG. 2, the etching does not proceed at all in the sample S1 in which the natural oxide film 4 is formed, whereas the etching of the polysilicon layer 3 proceeds with the progress of the etching time in the sample S2 in which the polysilicon layer 3 is exposed. Is done.

  From the above results, it was found that the etching rate of the TMAH solution with respect to silicon dioxide may be assumed to be 0 nm / min, while the etching rate of silicon becomes a finite value. Furthermore, according to another experimental result conducted by the inventor of the present application, it became clear that the etching rate of silicon by the TMAH solution depends on the temperature of the TMAH solution.

  FIG. 3 is a graph obtained by conducting the same experiment as described above using an organic alkali solution instead of the TMAH solution. The organic alkali solution was prepared by adding ammonium hydroxide into pure water to prepare an ammonium hydroxide solution having a concentration of 20 wt% or more, and then adding IPA (isopropyl alcohol) to the ammonium hydroxide solution at a concentration of 2 wt% or more. Made by melting.

  As shown in FIG. 3, it became clear that the organic alkaline solution also selectively etched silicon and not silicon dioxide.

  The organic alkali solution is not limited to the above, and a mixed solution of an alkali solution different from the ammonium hydroxide solution and a heavy alcohol such as IPA may be used as the organic alkali solution.

(B) Impurity concentration dependency of etching rate of TMAH solution In the above-described experiments of FIGS. 2 and 3, no impurity was introduced into the polysilicon layer 3, but the etching rate of the polysilicon layer 3 was changed to the impurity concentration. Presumed to depend. In order to confirm this point, the present inventor conducted the following experiment.

In this experiment, three samples having the same structure as the sample S2 were prepared. Then, arsenic as an n-type impurity and boron as a p-type impurity were ion-implanted into two polysilicon layers 3 of these samples, respectively. The doping amount in the ion implantation was set to 1.0 × 10 ¹⁷ cm ^{−3 to} 2.0 × 10 ²¹ cm ⁻³ . Further, the polysilicon layer 3 in the remaining one sample was not doped without introducing impurities.

  Thereafter, the polysilicon layer 3 of these samples was exposed to the TMAH solution for a predetermined time, and the etching amount of the polysilicon layer 3 was investigated. The result is shown in FIG.

  As shown in FIG. 4, it is clear that the etching rate of the polysilicon layer 3 is faster when an n-type impurity (arsenic) is introduced compared to the case of non-doping. On the other hand, when the p-type impurity (boron) was introduced, it became clear that the etching rate of the polysilicon layer 3 was slower than the non-doped case. Furthermore, according to another experiment conducted by the present inventor, it was also found that when the boron doping amount in FIG. 4 is further increased by 10 times, the etching of the polysilicon layer 3 hardly proceeds.

(2) Second Embodiment Next, a recess is formed in the silicon substrate by utilizing the etching characteristics of the TMAH solution and the organic alkali solution that have been clarified in the first embodiment, and the SiGe layer in the recess is formed as a source / drain. A method for manufacturing a MOS transistor will be described.

  5 to 6 are cross-sectional views of the semiconductor device according to this embodiment in the middle of manufacture.

  First, steps required until a sectional structure shown in FIG.

First, an element isolation trench 10g for STI (Shallow Trench Isolation) is formed on a p-type silicon (semiconductor) substrate 10 whose surface orientation is (001), and then an element isolation insulating film 11 is formed in the element isolation trench 10g. Embed a silicon dioxide layer. Thereafter, when, for example, phosphorus is used as the n-type impurity in the silicon substrate 10, ions are implanted under conditions of an acceleration energy of about 300 KeV or more and a dose amount of 1 × 10 ¹³ cm ⁻² or more, thereby being defined by the element isolation insulating film 11. An n-well 12 is formed in the p-type MOS transistor formation region.

When an n-type MOS transistor is fabricated in addition to the p-type MOS transistor to obtain a CMOS structure, an acceleration energy of 100 KeV or more and a dose of 1 × 10 ¹³ cm are further applied to the silicon substrate 10 in the n-type MOS transistor formation region. For example, boron is ion-implanted as a p-type impurity under the condition of ⁻² or more to form a p-well (not shown). In this case, the p-type and n-type impurities are divided using a resist pattern (not shown) on the silicon substrate 10, and this resist pattern is removed by wet processing after ion implantation.

Next, the surface of the silicon substrate 10 is thermally oxidized to form a gate insulating film 13 made of silicon dioxide having a thickness of about 0.5 to 5.0 nm. Here, as the gate insulating film 13, a gate insulating film obtained by adding a small amount of nitrogen to silicon dioxide may be used. Further, after forming a polysilicon layer 14 on the gate insulating film 13 to a thickness of about 10 to 300 nm by LPCVD using silane, a p-type impurity having a concentration such that the etching of the polysilicon by the TMAH solution does not proceed. Ions are implanted into the polysilicon layer 14. In the present embodiment, boron is adopted as such a p-type impurity, the acceleration energy is about 0.5 to 20 KeV, and the dose amount is about 1 × 10 ¹⁴ to 1 × 10 ¹⁷ cm ⁻ so that the entire gate electrode has a sufficiently high concentration. Ions are implanted into the polysilicon layer 14 under the ^two optimized conditions.

  Next, steps required until a sectional structure shown in FIG.

  First, the polysilicon layer 14 is patterned by photolithography to form a gate electrode 14c. In the embodiment, a process of forming an extension and a pocket without forming a thin spacer after gate processing will be described. However, in order to form an optimum overlap between the extension and the gate, a thin spacer of 5 to 20 nm is formed and then extension and pocket implantation are performed. It is also possible to do this. In addition, a method of forming a spacer only in either nMOS or pMOS is also possible. Any film can be used as long as it has a function as a spacer without being stuck to the film configuration and shape of the spacer.

Subsequently, using the gate electrode 14c as a mask, silicon, for example, boron as a p-type impurity as a p-type impurity, acceleration energy of about 0.2 to 1.0 KeV, dose amount of about 1 × 10 ¹⁴ to 2 × 10 ¹⁵ cm ⁻² , tilt angle 0 to 15 The first and second source / drain extensions 16a and 16b are shallowly formed on the silicon substrate 10 next to the first and second side surfaces 14a and 14b of the gate electrode 14c by ion implantation under the conditions of the above. Pocket injection to suppress the short channel effect at the same location, for example, antimony acceleration energy 30-80 keV, dose amount 1 × 10 ¹³ to 2 × 10 ¹⁴ cm ^-2 , tilt angle 0-35 degrees Do. When BF2 is used for the source / drain extension implantation ion species, the optimum condition is obtained by setting the energy to 1 to 2.5 keV and the dose to double. The optimum conditions described above vary depending on the presence or absence of the spacer and the thickness. When the spacer is present, it is necessary to guide the pocket to deepen energy and guide the extension to a larger amount, and to set the optimum condition. Further, pocket implantation using arsenic, phosphorus, antimony, or the like is possible, and this may be performed before or after extension implantation.

  Thereafter, a silicon dioxide layer is formed as a first sidewall insulating layer 15 on the entire surface by a plasma CVD method using silane under a substrate temperature of about 600 ° C. or less. This first sidewall insulating layer 15 Covers the first and second side surfaces 14a, 14b of the gate electrode 14c. Note that a silicon nitride layer may be formed as the first sidewall insulating layer 15 instead of the silicon dioxide layer.

  Next, steps required until a sectional structure shown in FIG.

  First, the first sidewall insulating layer 15 is etched back by plasma etching, leaving the first sidewalls 15a and 15b on the first and second side surfaces 14a and 14b. Further, in this etching, the portion of the gate insulating film 13 that is not covered with the first sidewalls 15a and 15b is also etched, and the gate insulating film 13 remains only under the gate electrode 14c.

  Further, for example, boron is ion-implanted as a p-type impurity (first conductivity type impurity) into the silicon substrate 10 using the gate electrode 14c and the first sidewalls 15a and 15b as a mask. Thereby, source / drain regions 17a, 17b deeper and darker than the source / drain extensions 16a, 16b are formed in the silicon substrate 10 next to the gate electrode 14c.

  After that, for example, activation annealing is performed under the condition of a substrate temperature of about 950 to 1050 ° C. to activate the impurities in the source / drain regions 17a and 17b. If necessary, this heat treatment may be omitted.

  Next, steps required until a sectional structure shown in FIG.

  First, etching of the silicon substrate 10 is started by immersing the silicon substrate 10 in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C. At this time, as shown in the experimental results shown in FIG. 2, the TMAH solution selectively etches only silicon and does not etch silicon dioxide. Accordingly, in this etching, the first sidewalls 15a and 15b and the element isolation insulating film 11 function as an etching mask, and the silicon substrate 10 in a portion not covered with these is selectively etched.

  Also, as shown in the experimental results shown in FIG. 4, silicon doped with p-type impurities has a slow etching rate with respect to the TMAH solution, so that boron is ionized at a high concentration as p-type impurities in the process of FIG. The implanted gate electrode 14c is hardly etched by this TMAH solution.

  In addition, in the etching with the TMAH solution, the (111) surface of the silicon substrate 10 is clearly displayed instead of the curved surface as in Non-Patent Document 1, so that the (111) surface is represented by the first and second side surfaces 10c, 10d. The first and second recesses (holes) 10a and 10b are formed.

  The depths of the first and second recesses 10a and 10b are controlled by the etching time, and are set to an optimum value of about 20 to 70 nm in this embodiment.

  Further, since the etching proceeds in the lateral direction, the upper end portions 10e and 10f of the recesses 10a and 10b near the gate electrode 14c enter the first sidewalls 15a and 15b by a distance d. As described, the etching rate of the silicon substrate 10 with respect to the TMAH solution depends on the impurity concentration in the silicon. Therefore, the above distance d depends on the impurity concentration of the source / drain extensions 16a and 16b and the source / drain regions 17a and 17b. It becomes controllable by adjusting.

  Further, as shown in the experimental results of FIG. 2, silicon dioxide is hardly etched with respect to the TMAH solution. Therefore, when the recesses 10a and 10b are formed by etching with the TMAH solution, the etching rate of the silicon substrate 10 near the side walls 15a and 15b made of silicon dioxide is lowered, so that the upper end portion 10e is moved to the side wall 15a by the etching. , 15b becomes slower than the other portions, and the above-mentioned penetration amount d can be easily controlled. Furthermore, since the penetration amount d is also determined by the substrate temperature when the first sidewall insulating layer 15 is formed, the penetration amount d can be controlled also by the substrate temperature. The same applies to each embodiment described later.

  Even if an organic alkaline solution is used instead of the TMAH solution, a clean (111) surface can be formed on the first and second side surfaces 10c and 10d, and the amount d of the recesses 10a and 10b can be controlled. .

  Next, steps required until a sectional structure shown in FIG.

  First, the silicon substrate 10 is placed in an epitaxial growth chamber (not shown) to stabilize the substrate temperature. Thereafter, by supplying a silane-based gas or the like into the chamber, a SiGe layer having a Ge concentration of 3 to 30% is selectively epitaxially grown in the recesses 10a and 10b. The SiGe layer grows selectively only on silicon, and does not grow on the element isolation insulating film 11 and the first sidewalls 15a and 15b made of silicon dioxide.

  Thereafter, when the thickness of the SiGe layer measured from the bottom surface of each recess 10a, 10b reaches an optimum value of about 20 to 120 nm, the epitaxial growth is stopped, and the obtained SiGe layer is converted into the first and second source / second / The drain material layers 18a and 18b are used.

  Although the thicknesses of the source / drain material layers 18a and 18b are not limited to the above, the upper surfaces of the source / drain material layers 18a and 18b are formed higher than the surface of the silicon substrate 10 as in the present embodiment. And the bottom of the n-well 12 become longer. This increases the distance between the pn junction on the bottom surface of the n-well 12 and the conductive plugs formed later on the source / drain material layers 18a and 18b, thereby suppressing the junction leak at the pn junction. Thus, the reliability of the transistor can be increased.

  In the above description, the source / drain material layers 18a and 18b are formed after the source / drain regions 17a and 17b are formed. However, the formation order is not particularly limited, and the source / drain material layers 18a and 18b are formed. The source / drain regions 17a and 17b may be formed later.

Next, in the present embodiment, boron as a p-type impurity is optimized in the source / drain material layers 18a and 18b with an acceleration energy of about 0.5-20 KeV and a dose amount of about 1 × 10 ¹⁴ to 1 × 10 ¹⁶ cm ⁻² . Ion implantation is performed under conditions. Thereafter, for example, activation annealing is performed under conditions of a substrate temperature of about 950 to 1050 ° C. to activate the impurities in the source / drain regions 18a and 18b. When in situ doping is performed when forming the source / drain regions 18a and 18b, impurity implantation and heat treatment may be omitted.

  Subsequently, as shown in FIG. 6B, for example, a nickel layer is formed as a refractory metal layer by a sputtering method, and then nickel and silicon are reacted by heat treatment to thereby form the top of the source / drain material layers 18a and 18b. First and second nickel silicide layers 19a and 19b are formed. The nickel silicide layer is also formed on the surface layer of the gate electrode 14c, whereby the gate electrode 14c has a polycide structure. Thereafter, the unreacted nickel layer is removed by wet etching.

  Note that a cobalt layer may be formed as a refractory metal layer instead of the nickel layer.

  Next, steps required until a sectional structure shown in FIG.

  First, a silicon nitride layer is formed as a cover insulating layer 20 on the entire surface by plasma CVD, and then a silicon dioxide layer is formed thick by HDPCVD (High Density Plasma CVD), which has good embeddability. To do. Thereafter, in order to flatten the unevenness formed on the upper surface of the interlayer insulating layer 21 reflecting the unevenness of the gate electrode 10c and the like, the upper surface of the interlayer insulating layer 21 is polished and flattened by a CMP (Chemical Mechanical Polishing) method. Turn into.

  Next, the interlayer insulating layer 21 and the cover insulating layer 20 are patterned by photolithography to form first and second holes 21a and 21b having a depth reaching the nickel silicide layers 19a and 19b. Then, a TiN layer is formed as a glue layer in the first and second holes 21a and 21b and on the upper surface of the interlayer insulating layer 21 by a sputtering method, and a tungsten layer is further formed thereon by a CVD method. Each layer 21a, 21b is completely filled with a layer. Thereafter, excess tungsten layer and glue layer formed on the interlayer insulating layer 21 are removed by polishing by the CMP method, and these films are removed in the first and second conductive plugs 22a, Leave as 22b.

  Thereafter, the process proceeds to a step of forming metal wirings electrically connected to the respective conductive plugs 22a and 22b on the interlayer insulating layer 21, and details thereof will be described.

  As described above, the basic structure of the p-type MOS transistor TR in which the source / drain material layers 18a and 18b are embedded in the recesses 10a and 10b is completed.

  In the MOS transistor TR, the silicon atoms in the silicon substrate 10 are forcibly stretched to match the large SiGe lattice spacing due to the lattice constant mismatch between the source / drain material layers 18a and 18b and the silicon substrate 10. The stress in the direction of the arrow in the figure is applied to the channel below the gate electrode 14c. As a result, the carrier mobility in the channel is improved as compared with the case where no stress is applied, and the driving capability of the MOS transistor can be increased.

  According to the above-described embodiment, since the TMAH solution or the organic alkaline solution is used as the etching solution for forming the first and second recesses 10a and 10b in the step of FIG. One (111) plane automatically appears on the first and second side faces 10c, 10d, and it becomes easy to control the shapes of the recesses 10a, 10b. Therefore, even if the above-described MOS transistors are integrated on the silicon substrate 10, the recesses 10a and 10b are provided for each MOS transistor as compared with Non-Patent Document 1 in which the first and second side surfaces 10c and 10d are curved. Difficult to vary in shape. Thereby, variation in the characteristics of the MOS transistor among elements can be suppressed, and the reliability of a semiconductor device such as an LSI can be improved.

  FIG. 7 is a diagram in which recesses 10a and 10b are formed in accordance with the present embodiment, and drawn based on SEM (Scanning Electron Microscope) images of the recesses 10a and 10b. As shown in this, (111) planes appear on the first and second side surfaces 10c and 10d constituting the side surfaces of the first and second recesses 10a and 10b.

  The distance d that each recess 10a, 10b enters below the first sidewalls 15a, 15b is not particularly limited.

  FIG. 8 is a diagram drawn on the basis of the SEM image when the distance d is increased by adjusting the substrate temperature when forming the first sidewall insulating layer 15.

  When the penetration amount d is increased in this way, the distance between the upper ends 10c and 10d of the recesses 10a and 10b and the channel under the gate electrode 10c is reduced, so that stress is efficiently applied to the channel from the upper ends 10e and 10f. Will be able to. Such advantages can also be obtained in the embodiments described below.

  Note that the present embodiment is not limited to the above. For example, instead of the SiGe layer, the source / drain material layers 18a and 18b may be formed of a metal layer made of a noble metal such as Pt (platinum). In that case, the created transistor TR is a Schottky transistor. The same applies to each embodiment described later.

(3) Third Embodiment Next, a method for manufacturing a semiconductor device according to a third embodiment of the present invention will be described.

  FIG. 9 is a cross-sectional view of the semiconductor device according to the present embodiment during manufacture. In FIG. 9, elements already described in the second embodiment are denoted by the same reference numerals as those in the second embodiment, and description thereof is omitted below.

First, the structure shown in FIG. 5A is completed according to the second embodiment. However, in the second embodiment, a p-type impurity having a sufficiently high concentration is ion-implanted into the polysilicon layer 14 so that the etching of the polysilicon by the TMAH solution does not proceed. However, in this embodiment, the polysilicon layer is formed by the TMAH solution. The polysilicon 14 is ion-implanted with a p-type impurity having such a low concentration that it can be etched partway. In the present embodiment, boron is adopted as such a p-type impurity, and the above polysilicon layer 14 is used under the conditions of an acceleration energy of about 0.5 to 20 KeV and a dose of about 1 × 10 ^{13 to} 5 × 10 ¹⁵ cm ^−3. Ion implantation.

  Thereafter, the structure shown in FIG. 5C is obtained according to the second embodiment described above.

  Next, steps required until a sectional structure shown in FIG.

  First, etching of the silicon substrate 10 is started by immersing the silicon substrate 10 in a TMAH solution having a product concentration of 5 to 30% and a temperature of 0 to 50 ° C. At this time, since the concentration of the p-type impurity introduced into the polysilicon layer 14 constituting the gate electrode 14c was previously reduced, not only the silicon substrate 10 but also the upper surface of the gate electrode 14c is etched in this etching. .

  Then, the etching is stopped when the depth of the first and second recesses 10a and 10b is about 20 to 70 nm and the thickness of the gate electrode 14c is reduced to about 30 to 150 nm. Thereby, as shown in the figure, the first and second recesses 10a and 10b, in which the first and second side surfaces 10c and 10d are (111) planes, are formed, and the height of the gate electrode 14c is the first. A structure lower than the side walls 15a and 15b is obtained.

Subsequently, as shown in FIG. 9B, the first and second SiGe layers to be the first and second source / drain material layers 18a and 18b are formed by performing the process of FIG. 2 recesses 10a, 1
Selectively epitaxial growth in 0b.

Next, in the present embodiment, boron as a p-type impurity is optimized in the source / drain material layers 18a and 18b with an acceleration energy of about 0.5-20 KeV and a dose amount of about 1 × 10 ¹⁴ to 1 × 10 ¹⁶ cm ⁻² . Ion implantation is performed under conditions. Thereafter, for example, activation annealing is performed under conditions of a substrate temperature of about 950 to 1050 ° C. to activate the impurities in the source / drain regions 18a and 18b. When in situ doping is performed when forming the source / drain regions 18a and 18b, impurity implantation and heat treatment may be omitted. Next, as shown in FIG. 9C, a nickel layer is formed as a refractory metal layer on each of the first and second source / drain material layers 18a and 18b and the gate electrode 14c by sputtering. Thereafter, nickel and silicon are reacted by heat treatment to form nickel silicide layers 19a and 19b on the first and second source / drain material layers 18a and 18b made of SiGe layers. This silicidation also occurs in the gate electrode 14c. However, since the thickness of the gate electrode 14c is reduced in advance in the process of FIG. 9A, the silicidation reaches all of the gate electrode 14c, and the gate electrode 14c It becomes a metal gate made of nickel silicide.

  In place of the nickel layer, a cobalt layer, a platinum layer, or a mixture layer thereof may be employed as the refractory metal layer.

  Thereafter, the basic structure of the MOS transistor is completed by performing the process shown in FIG.

  According to the present embodiment described above, the first and second side surfaces 10c, 10d of the recesses 10a, 10b can be configured by a single (111) plane in the same manner as the second embodiment.

  Furthermore, in this embodiment, the concentration of the p-type impurity introduced into the gate electrode 14c is set lower than that in the second embodiment, and the gate electrode 14c is etched at the same time as the recesses 10a and 10d are formed with the TMAH solution. I tried to be low.

  By doing so, the first and second source / drain material layers 18a and 18b are silicided to form the nickel silicide layers 19a and 19b. At the same time, the gate electrode 14c is silicided to become a metal gate. It becomes possible to make the formation processes of the metal gate and the recesses 10a and 10b compatible.

  In the above description, the TMAH solution is used as an etching solution when forming the recesses 10a and 10b. However, the same advantages as described above can be obtained by using an organic alkaline solution instead.

(4) Fourth Embodiment Next, a method for manufacturing a semiconductor device according to a fourth embodiment of the present invention will be described.

  FIG. 10 is a cross-sectional view of the semiconductor device according to the present embodiment during manufacture. In FIG. 10, the elements already described in the second embodiment are denoted by the same reference numerals as those in the second embodiment, and the description thereof is omitted below.

  First, after obtaining the cross-sectional structure shown in FIG. 5C according to the second embodiment described above, as shown in FIG. 10A, the silicon substrate 10, the first sidewalls 15a and 15b, and the gate electrode A silicon dioxide layer is formed to a thickness of about 5 to 100 nm as a second sidewall insulating layer 25 on 14c. The method for forming the silicon dioxide layer is not particularly limited, but in this embodiment, the silicon dioxide layer is formed by a plasma CVD method using silane as a reaction gas. Further, a silicon nitride layer may be formed as the second sidewall insulating layer 25 instead of the silicon dioxide layer.

  Next, steps required until a sectional structure shown in FIG.

  First, the second sidewall insulating layer 25 is etched back by plasma etching to leave the second sidewalls 25a and 25b on the side surfaces of the first sidewalls 15a and 15b. Thus, the side walls 26a and 26b formed of two insulating layers are also called double side walls.

Subsequently, using the second sidewalls 25a and 25b and the gate electrode 14c as a mask, for example, arsenic as an n-type impurity in the silicon substrate 10 is accelerated to about 3 to 20 KeV, and the dose amount is 1 × 10 ^{14 to} 5 × 10 ¹⁵ cm. Ion implantation is performed under the condition of ^-2 . Thereby, in the silicon substrate 10, the first and second impurity diffusion regions 27a of the second conductivity type (n-type) are formed deeper than the source / drain extensions 16a and 16b of the first conductivity type (p-type). 27b is formed.

  Next, steps required until a sectional structure shown in FIG.

  First, etching of the silicon substrate 10 is started by immersing the silicon substrate 10 in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C., and the source of the silicon substrate 10 next to the gate electrode 14c First and second recesses 10a and 10b having a depth of about 20 to 80 nm deeper than the drain extensions 16a and 16b are formed.

  At this time, as is apparent from the experimental results described with reference to FIG. 4, the silicon into which the n-type impurity is introduced has a higher etching rate with respect to the TMAH solution than the silicon into which the p-type impurity is introduced. Therefore, the first and second impurity diffusion regions 27a and 27b into which the n-type impurity (arsenic) is introduced are the first and second source / drain extensions 16a and 16b into which the p-type impurity (boron) is introduced and the source / drain Compared with the drain regions 17a and 17b, the etching with the TMAH solution proceeds faster.

  When the etching rates differ between the two layers as described above, different etching surfaces appear at the boundary between these layers. Accordingly, in the first and second side surfaces 10c and 10d of the recesses 10a and 10b, the (111) plane appears in the portion in contact with the p-type first and second impurity diffusion regions 27a and 27b, whereas the n-type In a portion in contact with the first and second source / drain extensions 16a and 16b and the source / drain regions 17a and 17b, a (111) plane different from the above appears.

  As a result, in the present embodiment, the first and second side surfaces 10c and 10d are each composed of two different (111) planes, and the cross-sectional shapes of the first and second side surfaces 10c and 10d are the lower side of the gate electrode 14c. Thus, the first and second recesses 10a and 10b having a concave shape are obtained.

  Note that the recesses 10a and 10b having such a cross-sectional shape can be formed by using an organic alkaline solution instead of the TMAH solution.

  Thereafter, by performing the process of FIG. 6A described above, as shown in FIG. 10D, a SiGe layer is selectively epitaxially grown in each of the recesses 10a and 10b. The second source / drain material layers 18a and 18b are used.

  Thereafter, the basic structure of the MOS transistor is completed by performing the steps shown in FIGS. 6B and 6C.

  According to the present embodiment described above, as shown in FIG. 10B, deeper than the p-type first and second source / drain extensions 16a and 16b and the source / drain regions 17a and 17b. The n-type first and second impurity diffusion regions 27a and 27b having opposite conductivity types are formed. Due to the difference in conductivity type, when the first and second recesses 10a and 10b are formed with the TMAH solution in the step of FIG. 10C, the first and second side surfaces 10c of the recesses 10a and 10b are formed. Different (111) planes appear in each of 10d, and the first and second recesses 10a and 10b are obtained in which the first and second side surfaces 10c and 10d have a concave shape recessed in the lower side of the gate electrode 14c.

  FIG. 11 shows the recesses 10a and 10b formed according to the present embodiment, and is drawn based on the SEM images of the recesses 10a and 10b. As shown, two different (111) planes appear on the side surfaces of the first and second recesses 10a and 10b.

  According to such a recess shape, as shown by an arrow in FIG. 10D, there is a stress peak slightly deeper than the silicon substrate 10, so that the interface characteristics between the silicon substrate 10 and the gate insulating layer 13 are reduced. It is difficult to deteriorate due to stress, and the reliability of the MOS transistor can be improved while improving the driving characteristics of the MOS transistor.

(5) Fifth Embodiment Next, a method for manufacturing a semiconductor device according to a fifth embodiment of the present invention will be described.

  12 and 13 are cross-sectional views of the semiconductor device according to the present embodiment during manufacture. In these drawings, elements already described in the second to fourth embodiments are denoted by the same reference numerals as those of the embodiments, and the description thereof is omitted below.

  As described below, in this embodiment, an SOI (Silicon on Insulator) substrate is used as the semiconductor substrate.

  First, steps required until a sectional structure shown in FIG.

  First, an SOI substrate 30 in which a buried insulating layer 32 and a silicon layer 33 are formed on a silicon substrate 31 is manufactured by, for example, a bonding method, and an element isolation groove 33g having a depth reaching the buried insulating layer 32 is formed in a silicon layer. 33. Further, a silicon dioxide layer is embedded as the element isolation insulating film 11 in the element isolation groove 33g.

  The plane orientation of the silicon layer 33 is not particularly limited, but in the present embodiment, the silicon layer 33 is formed so that the plane orientation is (001). Further, as the buried insulating layer 32, for example, a silicon dioxide layer having a thickness of 5 to 100 nm is formed.

Next, as shown in FIG. 12B, element isolation is performed by ion-implanting phosphorus as an n-type impurity into the silicon layer 33 under conditions of acceleration energy of about 300 KeV or more and a dose of 1 × 10 ¹³ cm ⁻³ or more. An n-well 34 is formed in a p-type MOS transistor formation region defined by the insulating film 11.

  Next, the surface of the silicon layer 33 is thermally oxidized to form a gate insulating film 13 made of silicon dioxide having a thickness of about 0.5 to 5.0 nm. Here, as the gate insulating film 13, a gate insulating film obtained by adding a small amount of nitrogen to silicon dioxide may be used. Furthermore, after the polysilicon layer 14 is formed on the gate insulating film 13 to a thickness of about 10 to 300 nm by LPCVD using silane, the etching of the polysilicon with the TMAH solution does not proceed. A p-type impurity having a similar concentration is ion-implanted into the polysilicon layer 14.

  Thereafter, the structure shown in FIG. 12C is obtained by performing the steps shown in FIGS. 5B and 5C. In this structure, first and second source / drain extensions 16a and 16b and source / drain regions 17a and 17b are formed in the silicon layer 33 beside the gate electrode 14c.

  Next, as shown in FIG. 12 (d), by etching the silicon substrate 10 in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C., etching of the silicon substrate 10 is started. First and second recesses 33a and 33b having a depth of about 5 to 50 nm are formed in the silicon layer 33 next to the gate electrode 14c.

  From the experimental results shown in FIG. 2, the TMAH solution selectively etches only silicon and does not etch silicon dioxide. Therefore, in this etching, the etching rate of the silicon layer 33 in the vicinity of the gate insulating film 13 made of silicon dioxide and the buried insulating layer 32 is lowered, whereas the etching rate in a portion away from these insulating layers. Speeds up. Due to such a difference in etching rate, the first and second side surfaces 33c and 33d constituting the side surfaces of the recesses 33a and 33b are not formed as a single crystal plane but are constituted by two different (111) planes. The cross-sectional shape is convex.

  The recesses 33a and 33b having such a cross-sectional shape can be formed by using an organic alkaline solution instead of the TMAH solution.

  Thereafter, by performing the process of FIG. 6A described above, as shown in FIG. 13, a SiGe layer is selectively epitaxially grown in each of the recesses 33a and 33b, and the SiGe layer is formed into the first and second sources. / The drain material layers 18a and 18b.

  Thereafter, the basic structure of the MOS transistor is completed by performing the steps shown in FIGS. 6B and 6C.

  According to the present embodiment described above, in the etching with the TMAH solution described with reference to FIG. 12D, the silicon layer 33 is formed in the vicinity of the gate insulating film 13 or in the vicinity of the embedded insulating film 32 constituting the SOI substrate 30. Etching rate decreases. As a result, since the etching rate of the silicon layer 33 varies depending on the depth, the first and second side surfaces 33c and 33d of the first and second recesses 33a and 33b obtained by the etching are different from each other ( 111) plane, and its cross-sectional shape is convex.

  The first and second source / drain material layers 18a and 18b formed in the recesses 33a and 33b having such a cross-sectional shape generate strong stresses on the upper and lower surfaces of the silicon layer 33 as shown by arrows in FIG. In addition, it is possible to obtain a stress distribution in which stress is weakened in the middle of the film where the (111) planes intersect.

(6) Sixth Embodiment Next, a method for manufacturing a semiconductor device according to a sixth embodiment of the present invention will be described.

  FIG. 14 is a cross-sectional view of the semiconductor device according to the present embodiment during manufacture. In these drawings, elements already described in the second to fourth embodiments are denoted by the same reference numerals as those of the embodiments, and the description thereof is omitted below.

  In the fifth embodiment described above, by using the SOI substrate 30, the first and second recesses 30a and 30b having a convex cross-sectional shape are formed. In contrast, in this embodiment, a recess having the same cross-sectional shape as described above is formed using a normal silicon substrate instead of an SOI substrate.

  First, the steps shown in FIGS. 5A to 5C are performed to obtain the cross-sectional structure shown in FIG.

Subsequently, as shown in FIG. 14B, using the gate electrode 14c and the first and second sidewalls 15a and 15b as masks, boron is used as a p-type impurity which has an effect of delaying the etching with the TMAH solution. 10 are ion-implanted to form p-type third and fourth impurity diffusion regions 35a and 35b. As the conditions for this ion implantation, conditions that are higher in concentration and deeper than the p-type source / drain extensions 16a and 16b and the source / drain regions 17a and 17b are employed. In this embodiment, as such conditions, for example, boron implantation under conditions of acceleration energy of about 1 to 20 KeV and a dose amount of about 5 × 10 ¹⁴ to 2 × 10 ¹⁶ cm ⁻² is used. As a result of such ion implantation, p-type third and fourth impurity diffusion regions 35a and 35b are formed which extend deeply into the silicon substrate 10 at a portion deeper than the source / drain regions 17a and 17b.

  Thereafter, activation annealing is performed under the same conditions as in the second embodiment to activate the impurities in the source / drain regions 17a and 17b.

  The order of forming the third and fourth impurity diffusion regions 35a and 35b and the source / drain regions 17a and 17b is not limited to the above. For example, the order may be reversed, and the source / drain regions 17a and 17b may be formed after the third and fourth impurity diffusion regions 35a and 35b are formed.

  Subsequently, as shown in FIG. 14C, etching of the silicon substrate 10 is started by immersing the silicon substrate 10 in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C. First and second recesses 10a and 10b having a depth of about 30 to 120 nm are formed on the silicon substrate 10 next to the gate electrode 14c.

  As described with reference to FIGS. 2 and 3, the etching rate of silicon with a TMAH solution decreases in the etching rate of silicon dioxide or silicon into which high-concentration p-type impurities are introduced. Therefore, in this etching, the etching rate decreases in the vicinity of the gate insulating film 13 made of silicon dioxide and in the vicinity of the third and fourth impurity diffusion regions 35a and 35b into which high-concentration p-type impurities are introduced. On the other hand, the etching rate does not decrease at portions away from these. Due to such a non-uniform etching rate, as in the fifth embodiment, the first and second side surfaces 10c and 10d of the recesses 10a and 10b do not become a single crystal plane, but two different ( 111) plane, and its cross-sectional shape is convex.

  The recesses 10a and 10b having such a cross-sectional shape can also be formed by using an organic alkaline solution instead of the TMAH solution.

  Subsequently, by performing the process of FIG. 6A described above, as shown in FIG. 14D, a SiGe layer is selectively epitaxially grown in each of the recesses 10a and 10b. The second source / drain material layers 18a and 18b are used.

  Thereafter, the basic structure of the MOS transistor is completed by performing the steps shown in FIGS. 6B and 6C.

  According to the present embodiment described above, since the p-type impurity having an effect of delaying the etching with the TMAH solution is introduced into the third and fourth impurity diffusion regions 35a and 35b, in the etching step of FIG. The etching rate of silicon decreases near the third and fourth impurity diffusion regions 35a and 35b and the gate insulating film 14c. As a result, as shown in FIG. 14 (c), the first and second side surfaces 10c and 10d of the first and second recesses 10a and 10b are each constituted by two (111) planes, The first and second recesses 10a and 10b having a convex cross-sectional shape can be formed. This makes it possible to give a favorable stress for improving mobility from the source / drain material layers 18a and 18b formed in the recesses 10a and 10b toward the channel.

(7) Seventh Embodiment Next, a method for manufacturing a semiconductor device according to a seventh embodiment of the present invention will be described.

  15 and 16 are cross-sectional views of the semiconductor device according to this embodiment in the middle of manufacture. In these drawings, the elements already described in the second to sixth embodiments are denoted by the same reference numerals as those of the embodiments, and the description thereof is omitted below.

  In the sixth embodiment described above, as shown in FIG. 14B, the gate electrode 14c and the first and second sidewalls 15a and 15b are used as masks for ion implantation to delay the etching with the TMAH solution. The p-type third and fourth impurity diffusion regions 35a and 35b were formed.

  On the other hand, in the present embodiment, an impurity diffusion region having an effect of increasing the etching rate is formed before the gate electrode 14c is formed.

  First, as shown in FIG. 15A, after forming an element isolation insulating film 11 in the element isolation trench 10g, an n well 12 is formed in a p-type MOS transistor formation region defined by the element isolation insulating film 11. To do.

Next, boron is ion-implanted into the silicon substrate 10 under the conditions of an acceleration energy of about 5 to 30 KeV and a dose of about 1 × 10 ^{13 to} 5 × 10 ¹⁵ cm ⁻³ as an n-type impurity that has the effect of delaying the etching with the TMAH solution. Then, a fifth impurity diffusion region 36 is formed in a portion deeper than a source / drain region to be formed later.

  Next, as shown in FIG. 15B, the surface of the silicon substrate 10 is thermally oxidized to form a gate insulating film 13 made of silicon dioxide, and further, LPCVD using silane is used to form the gate insulating film 13. A polysilicon layer 14 is formed thereon. Thereafter, using the same ion implantation conditions as in the second embodiment, a p-type impurity having a concentration such that the etching of polysilicon by the TMAH solution does not proceed is ion-implanted into the polysilicon layer 14.

  Subsequently, by performing the steps shown in FIGS. 5B and 5C, the source / drain extension 16a is formed on the silicon substrate 10 next to the gate electrode 14c as shown in FIG. 15C. 16b and source / drain regions 17a and 17b.

  Next, as shown in FIG. 15 (d), the silicon substrate 10 is immersed in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50.degree. First and second recesses 10a and 10b having a depth of about 20 to 150 nm are formed on the silicon substrate 10 next to the electrode 14c.

  In this etching, the etching rate of silicon increases in the vicinity of the fifth impurity diffusion region 36 into which n-type impurities having an effect of increasing the etching rate are introduced at a high concentration, whereas p-type impurities are introduced at a high concentration. In the vicinity of the formed source / drain extensions 16a and 16b and the source / drain regions 17a and 17b, the etching rate of silicon decreases. Therefore, for the same reason as in the sixth embodiment, the first and second side surfaces 10c and 10d of the recesses 10a and 10b are not formed as a single crystal plane, but are constituted by two different (111) planes, The cross-sectional shape is convex.

  Note that the first and second recesses 10a and 10b having the same cross-sectional structure as described above can be formed by using an organic alkali solution instead of the TMAH solution.

  Next, by performing the process of FIG. 6A described above, as shown in FIG. 16, a SiGe layer is selectively grown epitaxially in each of the recesses 10a and 10b, and the SiGe layer is formed into first and second sources. / The drain material layers 18a and 18b.

  Thereafter, the basic structure of the MOS transistor is completed by performing the steps shown in FIGS. 6B and 6C.

  According to the present embodiment described above, the fifth impurity diffusion region 36 that has an effect of suppressing the etching of silicon by the TMAH solution or the organic alkaline solution is formed in the silicon substrate 10 deeper than the source / drain regions 17a and 17b. Formed. Therefore, as in the sixth embodiment, when the first and second recesses 10a and 10b are formed by etching, the etching rate varies depending on the distance from the surface of the silicon substrate 10. Accordingly, the first and second recesses 10a and 10b have first and second side surfaces 10c and 10d each composed of two (111) planes, and the first and second recesses 10a and 10b are convex in cross section. Can be obtained. As a result, as in the sixth embodiment, it is possible to give a preferable stress for improving the mobility from the source / drain material layers 18a and 18b formed in the recesses 10a and 10b toward the channel.

(8) Eighth Embodiment FIGS. 17A and 17B are cross-sectional views of a semiconductor device according to an eighth embodiment of the present invention, and FIG. 18 is a plan view thereof. In these drawings, the elements already described in the second to sixth embodiments are denoted by the same reference numerals as those of the embodiments, and the description thereof is omitted below.

  In the above-described second to fourth embodiments, a substrate having a surface orientation (001) is used as a silicon substrate on which a MOS transistor is manufactured, and the gate width direction (extending direction of the gate electrode) is used. Was the [110] direction of the silicon substrate.

  In contrast, in the present embodiment, a silicon substrate having a surface orientation of (110) is employed, and the gate width direction (extending direction of the gate electrode) is set as the [111] direction of the silicon substrate.

  After adopting such a plane orientation and performing the steps of FIGS. 5A to 5C described in the second embodiment, the inside of the TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C. When the first and second recesses 10a and 10b having a depth of about 10 to 100 nm are formed in the silicon substrate 10 next to the gate electrode 14c, the cross-sectional structure as shown in FIG. can get.

  As described above, when the plane orientation of the silicon substrate 10 is (110) and the extending direction of the gate electrode 14c is the [111] direction, the (111) plane of silicon that appears by etching with the TMAH solution is the silicon substrate. It becomes perpendicular to the surface of 10. Therefore, the side surfaces of the first and second recesses 10 a and 10 b constituted by the (111) plane are perpendicular to the surface of the silicon substrate 10.

  FIG. 18 is a plan view after this process is completed, and FIG. 17A corresponds to a cross-sectional view taken along the line II of FIG.

  As shown in FIG. 18, the gate width direction, that is, the extending direction of the gate electrode 14c is the [111] direction, and the plane orientation of the silicon substrate 10 is (110), and such plane orientation is adopted. Thus, the side surfaces of the first and second recesses 10 a and 10 b can be made perpendicular to the surface of the silicon substrate 10.

  Next, by performing the process of FIG. 6A described above, as shown in FIG. 17B, a SiGe layer is selectively grown epitaxially in each of the recesses 10a and 10b. The second source / drain material layers 18a and 18b are used.

  Thereafter, the basic structure of the MOS transistor is completed by performing the steps shown in FIGS. 6C and 6D.

  According to the present embodiment described above, the silicon substrate 10 having the (110) plane orientation is adopted, and the extending direction of the gate electrode 14c is the [111] direction. In this case, the etching surface of the silicon substrate 10 with the TMAH solution or the organic alkali solution becomes the (111) surface in a direction perpendicular to the surface of the silicon substrate 10, and the (111) surface causes the first of the recesses 10a and 10b to be first. The second side surfaces 10c and 10d are configured. Accordingly, as shown by the arrows in FIG. 17B, the first and second source / drain material layers 18a and 18b in the recesses 10a and 10b are uniform from the channel to the channel with little intensity change in the depth direction. Stress can be applied stably.

(9) Ninth Embodiment FIG. 19 is a cross-sectional view of a semiconductor device according to a ninth embodiment of the present invention, and FIG. 20 is a plan view thereof. In these drawings, the elements already described in the eighth embodiment are denoted by the same reference numerals as those of the embodiment, and the description thereof is omitted below.

  In the eighth embodiment, the silicon substrate 10 having the (110) plane orientation is used, and the extending direction of the gate electrode 14c is set to the [111] direction.

  On the other hand, in the present embodiment, the surface orientation of the silicon substrate 10 is (110) as in the eighth embodiment, and the extending direction of the gate electrode 14c is the [100] direction.

  17A and 17B of the eighth embodiment employing such a plane orientation, as shown in FIG. 19, the first and second recesses 10a and 10b have the first and second recesses 10a and 10b. The (111) planes constituting the second side faces 10c and 10d have a gentler inclination angle θ than in the second embodiment described above.

  According to this, since the stress applied to the channel from the first and second source / drain material layers 18a and 18b changes sharply as it approaches the surface layer from a deep position of the silicon substrate 10, A large stress can be applied, and the magnitude of the stress can be easily controlled.

  20 is a plan view of the semiconductor device, and FIG. 19 corresponds to a cross-sectional view taken along the line II-II in FIG.

(10) Tenth Embodiment Next, a semiconductor device evaluation method according to a tenth embodiment of the present invention will be described.

  In order to check whether or not the characteristics of a MOS transistor are as designed, a test transistor called TEG (Test Element Group) was created at the development stage, and the carrier distribution in the channel of the transistor was actually measured. Measure physically. In order to be able to do this, it is necessary to remove the gate electrode and the gate insulating film after the MOS transistor is fabricated to expose the surface of the silicon substrate that will be the channel.

  However, when removing the gate electrode and the gate insulating film, if the silicon substrate is damaged, the carrier distribution in the channel is disturbed, and the obtained measurement value may deviate from the value in the transistor used for actual use. There is.

  Therefore, in order to perform the above evaluation, it is necessary to remove the gate electrode and the gate insulating film while preventing the silicon substrate from being damaged.

  21 and 22 are cross-sectional views showing a method for manufacturing a TEG used in the test method according to this embodiment.

  First, steps required until a sectional structure shown in FIG.

First, an element isolation trench 40g for STI (Shallow Trench Isolation) is formed in a silicon substrate 40 whose surface orientation is (001), and then a silicon dioxide layer is embedded as an element isolation insulating film 41 in the element isolation trench 40g. . Thereafter, phosphorus as an n-type impurity is ion-implanted into the silicon substrate 40 under an acceleration energy of about 300 KeV or more and a dose amount of 1 × 10 ¹³ cm ⁻² or more, whereby a p-type MOS transistor defined by the element isolation insulating film 41 is formed. An n-well 42 is formed in the formation region.

  Next, the surface of the silicon substrate 40 is thermally oxidized to form a gate insulating film 43 made of silicon dioxide to a thickness of about 0.5 to 3.0 nm, and then polysilicon is formed on the gate insulating film 43 by LPCVD using silane. Layer 44 is formed to a thickness of about 20-300 nm. Here, as the gate insulating film 43, a gate insulating film in which a small amount of nitrogen is added to silicon dioxide may be used.

  In the second embodiment already described, in order to prevent the gate electrode from being etched by the TMAH solution, the polysilicon layer 14 (see FIG. 5A) constituting the gate electrode is provided with a p-type impurity that has an effect of delaying the etching rate. Introduced. However, in this embodiment, since the gate electrode is difficult to be removed when the etching rate is reduced in this way, no p-type impurity is introduced into the polysilicon layer 44.

  Next, steps required until a sectional structure shown in FIG.

  First, the polysilicon layer 44 is patterned by photolithography to form a gate electrode 44c.

Subsequently, for example, boron is ion-implanted as a p-type impurity into the silicon substrate 40 using the gate electrode 44c as a mask, and the first and second silicon substrates 40 beside the first and second side surfaces 44a and 44b of the gate electrode 44c. Source / drain extensions 46a and 46b are formed shallowly. The conditions for the ion implantation are the same as those for an actual product MOS transistor. For example, acceleration energy of about 0.2 to 1.0 KeV and a dose of about 1 × 10 ^{14 to} 5 × 10 ¹⁵ cm ⁻² are employed. If necessary, pocket injection of arsenic, phosphorus, antimony, etc. is performed simultaneously. The first and second source / drain extensions 46a and 46b using n-type impurities may be formed shallowly.

  Thereafter, a silicon dioxide layer is formed on the entire surface as a sidewall insulating layer 45 by plasma CVD using silane to a thickness of about 5 to 100 nm. The sidewall insulating layer 45 forms the first and second gate electrodes 44. The side surfaces 44a and 44b are covered. A silicon nitride layer may be formed as the sidewall insulating layer 45 instead of the silicon dioxide layer.

  Next, steps required until a sectional structure shown in FIG.

  First, the sidewall insulating layer 45 is etched back by plasma etching to leave the sidewalls 45a and 45b on the first and second side surfaces 44a and 44b. In this etching, the portion of the gate insulating film 43 not covered with the sidewalls 45a and 45b is also etched, and the gate insulating film 43 remains only under the gate electrode 44c.

Further, in the same manner as an actual MOS transistor for a product, the gate electrode 44c and the sidewalls 45a and 45b are used as a mask, and, for example, boron as a p-type impurity in the silicon substrate 40 is accelerated energy about 1 to 10 KeV, and the dose amount is about Ion implantation is performed under conditions of about 5 × 10 ¹⁴ to 1 × 10 ¹⁶ cm ⁻² . As a result, source / drain regions 47a, 47b deeper and darker than the source / drain extensions 46a, 46b are formed in the silicon substrate 40 next to the gate electrode 44c. If necessary, this impurity implantation may be omitted.

  Thereafter, for example, activation annealing is performed under conditions of a substrate temperature of about 950 to 1050 ° C. and a processing time of 0 to 10 seconds to activate the impurities in the source / drain regions 47a and 47b.

  Next, steps required until a sectional structure shown in FIG.

  First, the silicon substrate 40 is immersed in a TMAH solution having a volume concentration of 5 to 30% and a temperature of 0 to 50 ° C., and the silicon substrate 40 and the gate electrode 44c that are not covered with silicon dioxide are etched. As shown in the experimental results shown in FIG. 2, the TMAH solution has very good etching selectivity between silicon and silicon dioxide. Therefore, in this etching, the reduction of the gate insulating film 43 made of a thin silicon dioxide layer is negligibly small, and the channel under the gate insulating film 43 is not damaged.

  As a result of this etching, first and second recesses 40a and 40b are formed in the silicon substrate 40 in portions not covered with the element isolation insulating film 41 and the sidewalls 45a and 45b.

  Note that this etching may be performed with an organic alkali solution instead of the TMAH solution, and even in that case, the channel is less damaged.

  Thereafter, as shown in FIG. 22, the silicon substrate 40 is immersed in an etching solution obtained by mixing HF (hydrofluoric acid) and HCl at a volume ratio of 1:19, and an element isolation insulating film 41 made of silicon dioxide, By selectively removing the sidewalls 15a and 15b and the gate insulating film 43, the channel 40d whose carrier distribution is to be measured is exposed. The channel 40d is hydrogen-terminated by hydrogen ions contained in the etching solution and becomes chemically active.

  Thus, the basic structure of the TEG in which the channel 40d is exposed is completed.

  Next, a method for evaluating the carrier distribution in the TEG channel 40d will be described with reference to FIG.

  First, the TEG produced as described above is placed in a scanning tunneling microscope (STM), which is a type of probe microscope, and the tip of the probe 50 is in a non-contact state in a plane parallel to the channel 40d. To scan the probe 50. At this time, a predetermined voltage is applied between the probe 50 and the silicon substrate 40, and the magnitude of the tunnel current flowing between the probe and the silicon substrate 40 changes according to the carrier distribution in the channel 40d. . By visualizing the change of the tunnel current, the carrier distribution in the channel 40d can be grasped.

  According to the present embodiment described above, the gate electrode 44c is selectively etched with a TMAH solution or an organic alkali solution as shown in FIG. 21 (d), and then HF and HCl are removed as shown in FIG. The gate insulating film 43 is etched and removed with a mixed etching solution to expose the channel 40d.

  Thus, by using a TMAH solution or an organic alkali solution having high selectivity between silicon and silicon dioxide, only the gate electrode 44c is etched with high etching selectivity without damaging the channel region 44d below the gate electrode 44c. Can be removed. Therefore, in the process of exposing the channel 40d, there is no fear that the carrier distribution in the channel fluctuates, and it is possible to measure the carrier distribution substantially the same as that in the MOS transistor that is actually used, and the performance of the MOS transistor. Can be evaluated with high accuracy.

  Furthermore, according to the present embodiment, the gate insulating film 43 is removed by etching with an etchant containing HF, so that the surface of the channel 40d exposed after the gate insulating film 43 is removed is automatically hydrogen-terminated. . In the carrier distribution measurement by STM, it is preferable that the measurement surface is hydrogen-terminated in order to clearly observe the change in conductivity on the measurement surface. Since termination can be performed, a carrier distribution can be easily measured without a step for hydrogen termination.

  FIGS. 24A and 24B are diagrams in which the surface of the TEG of the present embodiment is actually scanned by STM and drawn based on the concavo-convex image. FIG. 24B is a diagram illustrating FIG. The image brightness is exaggerated.

  FIG. 25 is an image obtained by observing the carrier distribution in the TEG channel 40d shown in FIGS. 24A and 24B by STM.

  As described above, in this embodiment, the channel is hardly damaged when the gate electrode 44c is removed by etching. Therefore, the carrier distribution shown in FIG. 25 is substantially the same as that in the MOS transistor used for actual use. Expected to have the same distribution.

  In contrast to the present embodiment, plasma etching may be employed for removing the gate electrode 44c. However, in the plasma etching, the channel 40d is damaged through the gate insulating film 43 as the etching proceeds due to the kinetic energy of ions in the etching atmosphere, so that the carrier distribution fluctuates, and the MOS transistor as in the present embodiment. It is not possible to accurately evaluate the performance.

  Further, the MOS transistor to be evaluated in the present embodiment is not limited to a type in which a source / drain material layer such as a SiGe layer is formed in the first and second recesses 40a and 40b and stress is applied to the channel. . For example, a normal type MOS transistor that does not have a recess under actual use can also be evaluated. However, even in this type of MOS transistor TEG, when the gate electrode 44c is etched in the step of FIG. 21D, the silicon substrate 40 where the element isolation insulating film 41 is not formed is etched. Recesses 40a and 40b are formed.

(11) Eleventh Embodiment Next, a semiconductor device evaluation method according to an eleventh embodiment of the present invention will be described.

  FIG. 26 is a cross-sectional view of the TEG used in the present embodiment.

  In the tenth embodiment, the channel 40d is observed by STM which is a kind of probe microscope. On the other hand, in this embodiment, the channel 40d is observed using a scanning capacitance microscope.

  In the scanning capacitance microscope, a capacitor is constituted by the probe of the microscope and the channel 40d, and the impurity distribution in the channel 40d is observed by detecting the capacitance value of the capacitor. In this embodiment, in order to form the dielectric layer of this capacitor, the surface of the channel 40d is irradiated with ozone, and a silicon dioxide layer (dielectric layer) 51 as shown in FIG. To a thickness of about 1.0 nm.

  Then, as shown in FIG. 27, the silicon dioxide layer 51 is scanned with the probe 52 while the tip of the probe 52 of the scanning capacitance microscope is in contact with the surface of the silicon dioxide layer 51, and the channel is formed from above the silicon dioxide layer 51. A carrier distribution within 40d is obtained.

  Thus, even when the channel 40d is observed with a scanning capacitance microscope, the silicon substrate 40 is damaged by removing the gate electrode 44c by wet etching using a TMAH solution, as in the tenth embodiment. Therefore, it is possible to measure the same carrier distribution as that in the MOS transistor that is actually used.

  Although the scanning capacitance microscope is used in the above, the same advantages as those of the present embodiment can be obtained by using a scanning spreading resistance microscope instead.

  The features of the present invention are added below.

(Appendix 1) a semiconductor substrate;
A gate insulating film and a gate electrode sequentially formed on the semiconductor substrate;
A source / drain material layer formed in a hole in the semiconductor substrate next to the gate electrode;
Have
The semiconductor device according to claim 1, wherein a side surface of the hole near the gate electrode is constituted by at least one crystal plane of the semiconductor substrate.

  (Additional remark 2) The said side surface of the said hole is comprised by two crystal planes, and the cross-sectional shape of this side surface is concave shape, The semiconductor device of Additional remark 1 characterized by the above-mentioned.

  (Additional remark 3) The said side surface of the said hole is comprised by two crystal planes, and the cross-sectional shape of this side surface is convex shape, The semiconductor device of Additional remark 1 characterized by the above-mentioned.

  (Supplementary note 4) The semiconductor device according to supplementary note 3, wherein a buried insulating layer is formed in the semiconductor substrate under the hole.

  (Additional remark 5) The said side surface of the said hole is comprised with the single crystal plane perpendicular | vertical with respect to the said semiconductor substrate, The semiconductor device of Additional remark 1 characterized by the above-mentioned.

  (Appendix 6) The semiconductor substrate is a silicon substrate, the surface orientation of the surface of the silicon substrate is (110), and the gate width direction of the gate electrode is the [111] direction of the silicon substrate. The semiconductor device according to appendix 5.

  (Supplementary note 7) The semiconductor device according to supplementary note 1, wherein the semiconductor substrate is a silicon substrate, and the side surface of the hole is constituted by a (111) plane of the silicon substrate.

  (Supplementary note 8) The semiconductor device according to supplementary note 7, wherein a surface orientation of the surface of the silicon substrate is (001).

  (Supplementary note 9) The semiconductor device according to supplementary note 7, wherein a surface orientation of the surface of the silicon substrate is (110), and a gate width direction of the gate electrode is a [100] direction of the silicon substrate.

  (Supplementary note 10) The semiconductor device according to supplementary note 1, wherein an upper end portion of the hole enters below the sidewall, and a distance from a channel under the gate electrode is reduced.

  (Additional remark 11) The semiconductor device of Additional remark 1 characterized by the whole said gate electrode being comprised by the refractory metal silicide.

  (Additional remark 12) The said source / drain material layer is a SiGe layer, The semiconductor device of Additional remark 1 characterized by the above-mentioned.

  (Supplementary note 13) The semiconductor device according to supplementary note 1, wherein the source / drain material layer is a metal layer.

(Additional remark 14) The process of forming a gate insulating film on a semiconductor substrate,
Forming a gate electrode on the gate insulating film;
Forming a sidewall on a side surface of the gate electrode;
After forming the sidewall, using an organic alkaline solution or TMAH (tetramethylammonium hydride) solution as an etching solution, forming a hole in the semiconductor substrate next to the gate electrode;
Forming a source / drain material layer in the hole;
A method for manufacturing a semiconductor device, comprising:

(Supplementary Note 15) In the step of forming the hole, the thickness of the gate electrode is reduced by the etchant,
Forming the refractory metal layer on the source / drain material layer and the gate electrode after forming the hole; heating the refractory metal layer to react with the gate electrode; and 15. The method for manufacturing a semiconductor device according to appendix 14, wherein the method includes siliciding the whole.

  (Additional remark 16) The manufacturing method of the semiconductor device of Additional remark 15 characterized by having the process of introduce | transducing p-type impurity into the said gate electrode before the process of forming the said hole.

(Supplementary Note 17) Before the step of forming the hole, a step of forming a first conductivity type impurity diffusion region in the silicon substrate, and a second conductivity type impurity diffusion region in the semiconductor substrate are formed as the first conductivity type impurity. A step of forming deeper than the diffusion region,
15. The method of manufacturing a semiconductor device according to appendix 14, wherein the hole is formed deeper than the first conductivity type impurity diffusion region in the step of forming the hole.

(Supplementary Note 18) In the step of forming the first conductivity type impurity diffusion region, a source / drain extension is formed by introducing a first conductivity type impurity into the silicon substrate while using the gate electrode as a mask. An extension is used as the first impurity diffusion region;
Item 18. The supplementary note 17, wherein the step of forming the second conductivity type impurity diffusion region is performed by introducing the second conductivity type impurity into the silicon substrate using the gate electrode and the sidewall as a mask. Semiconductor device manufacturing method.

(Supplementary Note 19) The step of forming the sidewall includes a step of forming a first sidewall on the side surface of the gate electrode, and a step of forming a second sidewall on the side surface of the first sidewall.
After the step of forming the first sidewall, the method further includes a step of forming a source / drain region by introducing a first conductivity type impurity into the silicon substrate while using the first sidewall as a mask. The method for manufacturing a semiconductor device according to appendix 18.

(Supplementary note 20) The semiconductor according to supplementary note 17, wherein a p-type impurity diffusion region is formed as the first conductivity type impurity diffusion region, and an n-type impurity diffusion region is formed as the second conductivity type impurity diffusion region. Device manufacturing method.

  (Supplementary Note 21) A p-type impurity diffusion region is formed as the first conductivity type impurity diffusion region, and a p-type impurity diffusion having an impurity concentration higher than that of the first conductivity type impurity diffusion region is formed as the second conductivity type impurity diffusion region. The method of manufacturing a semiconductor device according to appendix 17, wherein a region is formed.

  (Supplementary note 22) The semiconductor device manufacturing method according to supplementary note 21, wherein a source / drain extension or a source / drain region is formed as the first conductivity type impurity diffusion region.

  (Additional remark 23) The manufacturing method of the semiconductor device of Additional remark 14 characterized by using a SOI substrate as said semiconductor substrate.

  (Additional remark 24) The manufacturing method of the semiconductor device of Additional remark 14 characterized by using a silicon substrate as said semiconductor substrate.

  (Supplementary note 25) The method for manufacturing a semiconductor device according to supplementary note 24, wherein a substrate having a surface orientation (001) is used as the silicon substrate.

  (Supplementary Note 26) A substrate having a surface orientation of (110) is used as the silicon substrate, and the gate electrode is formed so that the gate width direction is the [111] direction in the step of forming the gate electrode. 25. A method of manufacturing a semiconductor device according to appendix 24, wherein:

  (Supplementary Note 27) Using a substrate having a surface orientation of (110) as the silicon substrate, and forming the gate electrode, forming the gate electrode so that a gate width direction is a [100] direction. 25. A method of manufacturing a semiconductor device according to appendix 24, wherein:

  (Supplementary note 28) The semiconductor device according to supplementary note 14, wherein an amount of an upper end portion of the hole entering below the sidewall is controlled by adjusting a substrate temperature when the sidewall is formed. Production method.

  (Supplementary note 29) The semiconductor device manufacturing method according to supplementary note 14, wherein a mixed solution of an ammonium hydroxide solution and IPA (isopropyl alcohol) is used as the organic alkaline solution.

  (Supplementary Note 30) The method for manufacturing a semiconductor device according to Supplementary Note 14, wherein a SiGe layer is epitaxially grown as the source / drain material layer.

  (Additional remark 31) The metal layer is formed as said source / drain material layer, The manufacturing method of the semiconductor device of Additional remark 14 characterized by the above-mentioned.

(Supplementary Note 32) A step of selectively etching and removing a gate electrode of a MOS transistor formed on a semiconductor substrate by using an organic alkaline solution or a TMAH solution as an etching solution;
Exposing the channel of the MOS transistor by removing the gate insulating film of the MOS transistor by wet etching;
Examining the carrier distribution in the exposed channel with a microscope;
A method for evaluating a semiconductor device, comprising:

(Supplementary Note 33) In the step of exposing the channel, the gate insulating film is removed with an etchant containing hydrofluoric acid,
33. The semiconductor device evaluation method according to appendix 32, wherein a probe microscope is used as the microscope in the step of examining the impurity distribution.

(Supplementary Note 34) The method further includes a step of forming a dielectric layer on the exposed channel.
The semiconductor device evaluation according to appendix 32, wherein, in the step of examining the impurity distribution, the carrier distribution is examined from above the dielectric layer using a scanning capacitance microscope or a scanning spreading resistance microscope as the microscope. Method.

  (Supplementary Note 35) The semiconductor device evaluation according to supplementary note 34, wherein the step of forming the dielectric layer is performed by irradiating the semiconductor in the channel portion with ozone to form an oxide layer. Method.

  (Supplementary note 36) The semiconductor device evaluation method according to supplementary note 32, wherein a silicon substrate is used as the semiconductor substrate.

DESCRIPTION OF SYMBOLS 1, 10, 40 ... Silicon substrate, 2 ... Silicon dioxide layer, 3, 14, 44 ... Polysilicon layer, 4 ... Natural oxide film, 10a, 10b ... 1st, 2nd recess, 10c, 10d ... 1st, 1st 2 side surfaces, 10 g, 33 g, 40 g... Element isolation trench, 10 e, 10 f... Upper end, 11... Element isolation insulating film, 12... N well, 13, 43. , 14c, 44c ... gate electrode, 15 ... first sidewall insulating layer, 15a, 15b ... first sidewall, 16a, 16b ... source / drain extension, 17a, 17b ... source / drain region, 18a, 18b ... first DESCRIPTION OF SYMBOLS 1, 2nd source / drain material layer, 19a, 19b ... 1st, 2nd nickel silicide layer, 20 ... Cover insulating layer, 21 ... Interlayer insulating layer, 21a, 21b ... 1st, 2nd e 22a, 22b ... first and second conductive plugs, 27a, 27b ... first and second impurity diffusion regions, 30 ... SOI substrate, 31 ... silicon substrate, 32 ... buried insulating layer, 33 ... silicon layer, 33a, 33b ... first and second recesses 33c, 33d ... first and second side surfaces 35a, 35b ... third and fourth impurity diffusion regions 36 ... fifth impurity diffusion regions 40a, 40b ... first, 2nd recess, 40d ... channel, 45 ... sidewall insulating layer, 44a, 44b ... 1st, 2nd side, 45a, 45b ... sidewall, 46a, 46b ... 1st, 2nd source / drain extension, 47a, 47b ... Source / drain region, 50 ... STM probe, 51 ... Silicon dioxide layer, 52 ... Probe of scanning capacitance microscope.

Claims (7)

A semiconductor substrate;
A gate insulating film and a gate electrode sequentially formed on the semiconductor substrate;
A source / drain material layer having a lattice constant different from the lattice constant of the semiconductor substrate, formed in a hole of the semiconductor substrate next to the gate electrode;
Have
A semiconductor device, wherein a side surface of the hole near the gate electrode is constituted by two crystal planes of the semiconductor substrate, and a cross-sectional shape of the side surface is convex. Forming a gate insulating film on the semiconductor substrate;
Forming a gate electrode on the gate insulating film;
Forming a sidewall on a side surface of the gate electrode;
After forming the sidewall, using an organic alkaline solution or TMAH as an etchant, forming a hole in the semiconductor substrate next to the gate electrode;
Forming a source / drain material layer having a lattice constant different from the lattice constant of the semiconductor substrate in the hole ,
Before the step of forming the hole, a step of forming a first conductivity type impurity diffusion region in the semiconductor substrate; and a second conductivity type impurity diffusion region in the semiconductor substrate than the first conductivity type impurity diffusion region. A step of forming deeply,
In the step of forming the hole, the hole is formed deeper than the first conductivity type impurity diffusion region . 3. The semiconductor device according to claim 2, wherein a p-type impurity diffusion region is formed as the first conductivity type impurity diffusion region, and an n-type impurity diffusion region is formed as the second conductivity type impurity diffusion region. Method. A p-type impurity diffusion region is formed as the first conductivity type impurity diffusion region, and a p-type impurity diffusion region having an impurity concentration higher than that of the first conductivity type impurity diffusion region is formed as the second conductivity type impurity diffusion region. A method of manufacturing a semiconductor device according to claim 2 . 3. The method of manufacturing a semiconductor device according to claim 2 , wherein an SOI substrate is used as the semiconductor substrate. 3. The method of manufacturing a semiconductor device according to claim 2, wherein a side surface of the hole near the gate electrode is constituted by two crystal planes of the semiconductor substrate, and a cross-sectional shape of the side surface is convex. Forming a gate insulating film on the semiconductor substrate;
Forming a gate electrode on the gate insulating film;
Forming a sidewall on a side surface of the gate electrode;
After forming the sidewall, using an organic alkaline solution or TMAH as an etchant, forming a hole in the semiconductor substrate next to the gate electrode;
Forming a source / drain material layer having a lattice constant different from the lattice constant of the semiconductor substrate in the hole;
Have
A method of manufacturing a semiconductor device, wherein a side surface of the hole near the gate electrode is constituted by two crystal planes of the semiconductor substrate, and a cross-sectional shape of the side surface is convex.

Images