• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Dopant impurities are implanted into the active regions 60a / 60b assigned to the field effect transistors, and then titanium silicide layers 87-92 from the titanium layer 86 are deposited on these dope regions. Form. When dopant impurities are implanted into the doped regions, implantation of dopant impurities into the light inactive region 60c which is not assigned to any circuit component by the photoresist implantation mask 76/81 is prevented. In addition, a thick titanium silicide layer 93 is grown on the light inactive region. Even if the titanium silicide layers 87-93 are heat-treated, the thick titanium silicide layer 93 on the light inactive region hardly solidifies, and the interlayer insulating layer 97 hardly peels off from the titanium silicide layer on the light inactive region.
    公开号:KR19980070802A
    申请号:KR1019980002139
    申请日:1998-01-23
    公开日:1998-10-26
    发明作者:후지이구누히로
    申请人:가네꼬히사시；닛뽕덴끼가부시끼가이샤；
    IPC主号:
    专利说明:

    Method for manufacturing semiconductor device without delamination between silicide layer and insulating layer
    BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a manufacturing technology of a semiconductor integrated circuit device, and more particularly, to a semiconductor device without separation between a silicide layer and an insulating layer.
    As the integrated density of semiconductor integrated circuit devices increases, circuit components become smaller. The field effect transistor has a narrow gate electrode and a shallow source / drain region, and the narrow gate electrode and the shallow source / drain region have a problem of high resistance. Since signal transmission is delayed by high resistance, the desired signal processing speed cannot be obtained in an integrated circuit.
    Salicide (self-aligned silicide) structures have been proposed to reduce resistance. Titanium is used for this salicide structure because titanium silicide has the lowest resistance among the current available silicides. 1A-1H illustrate a conventional method of manufacturing a MOS (metal oxide semiconductor) field effect transistor having a salicide structure.
    The conventional method starts by providing a p-type semiconductor substrate 1 and forming an n-type well 2 on the surface of the p-type semiconductor substrate 1. Silicon oxide is selectively grown on the main surface of the p-type silicon substrate 1 to form the field oxide layer 3. The field oxide layer 3 defines an active region 4 assigned to a MOS field effect transistor and an inactive region 5 assigned to a scribe region. The active region 4 and the inactive region 5 are thermally oxidized to grow thin gate oxide layers 6 and 7 and to deposit polysilicon over the entire surface of the final semiconductor structure. Phosphorus is introduced inside the polysilicon layer to reduce resistance.
    A photoresist etch mask (not shown) is patterned onto the polysilicon layer using photolithography techniques, and the polysilicon layer is selectively removed using dry etching techniques. In this way, the polysilicon layer is patterned with gate electrodes 8 and 9 on thin gate oxide layers 6 and 7. Peel off the photoresist etch mask.
    The photoresist implantation mask 10 is patterned onto the final semiconductor structure using photolithography techniques and the n-type well 2 is covered with this photoresist implantation mask 10. Phosphorus is implanted into the p-type silicon substrate 1, and low-doped n-type regions 11, 12, and 13 are formed in the active region 4 and the inactive region 5. The lightly doped n-type regions 11 and 12 are self-aligned with the gate electrode 8. The final semiconductor structure is shown in FIG. 1A. The photoresist ion implantation mask 10 is peeled off.
    The photoresist implantation mask 14 is patterned onto the final semiconductor structure using photolithography techniques, and the n-type well 2 is not covered with this photoresist implantation mask 14. Photoresist implantation mask 14 is an inversion of photoresist implantation mask 10 because photomasks (not shown) are easily designed. For this reason, phosphorus is injected into the inactive region 5. Boron is implanted into the n-type well 2 to form the p-type impurity regions 15 and 16 in the n-type well 2 in self-alignment with the gate electrode 9 as shown in FIG. 1B. . The photoresist ion implantation mask 14 is peeled off.
    Silicon oxide or silicon nitride is deposited over the entire surface of the final semiconductor structure, and the silicon oxide or silicon nitride is anisotropically etched to form sidewall spacers 15 and 16 on both sides of the gate electrodes 8 and 9. do. Silicon oxide is deposited over the entire surface of the final semiconductor structure to form a covering layer 17.
    The photoresist ion implantation mask 18 is patterned on the coating layer 17, and the n-type well 2 is covered with this photoresist ion implantation mask 18. Arsenic ions are implanted into the active region 4 and the inactive region 5 to form high concentration dope n-type impurity regions 19, 20, and 21. The heavily doped n-type impurity regions 19/20 are self-aligned with the sidewall spacers 15 and form source / drain regions 22/23 together with the lightly-doped n-type impurity regions 11 and 12. Source / drain regions 22/23 have an LDD (lightly doped drain) structure. The photoresist ion implantation mask 18 is the inversion of the photoresist ion implantation mask 14 and implants arsenic into the inactive region 5. For this reason, the low concentration dope n-type impurity region 13 is stacked in the high concentration dope n-type impurity region 21. The final semiconductor structure is shown in FIG. 1C. The photoresist ion implantation mask 18 is peeled off.
    The photoresist ion implantation mask 22 is patterned on the coating layer 17, and this photoresist ion implantation mask 22 is the inversion of the photoresist ion implantation mask 18. FIG. Boron is implanted into the n-type well to form highly doped p-type impurity regions 23/24 with the sidewall spacers 16 in a self-aligned manner. The high concentration dope p-type impurity regions 23/24 form the p-type source / drain region 25/26 together with the low concentration dope p-type impurity regions 15/16. The p-type source / drain region 25/26 has an LDD structure. The photoresist ion implantation mask 22 is peeled off.
    An amorphous silicon layer is formed by ion implanting arsenic into the surface of the n-type source / drain region 22/23, the surface of the p-type source / drain region 25/26, and the surface of the polysilicon gate electrode 8/9. 27, 28, 29, 30, and 31). In addition, an amorphous silicon layer 33 is formed in the highly doped n-type impurity region 21 by the implantation of arsenic. The final semiconductor structure is shown in FIG.
    The coating layer 17 is then removed from the final semiconductor structure and sputtering is used to deposit titanium over the entire surface of the final semiconductor structure. This titanium forms a titanium layer 34 and maintains contact with the amorphous silicon layers 27 to 33 as shown in FIG. 1F.
    The final semiconductor structure is placed in a nitrogen atmosphere and heated to below 700 ° C. using rapid thermal annealing technology. Titanium then reacts with amorphous silicon to form titanium silicide layers 35, 36, 37, 38, 39, 40 on silicon / polysilicon layers 22, 8, 23, 25, 9, 26, and 21. , And 41) are formed respectively. The remaining titanium reacts with nitrogen and is converted to titanium nitride layer 42 as shown in FIG. 1G.
    The titanium nitride layer 42 is etched with a wet etching solution containing ammonia water and hydrogen peroxide water. Titanium silicide layers 35 to 41 remain on the silicon / polysilicon layers 22, 8, 23, 25, 9, 26, and 21. These titanium silicide layers 35 to 41 are rapidly heat treated at a higher temperature than the previous rapid heat treatment. Then, a phase change occurs in the titanium silicide layers 35 to 41, and the resistance of the titanium silicide is reduced.
    A silicon oxide layer 43 is formed by depositing undoped silicon oxide over the entire surface of the final semiconductor structure. Borosilicate glass, phosphorosilicate glass, or borophosphosilicate glass is deposited over the undoped silicon oxide layer 43 to form an interlayer insulating film 44. This interlayer insulating film 44 is heated to about 800 占 폚 to increase the density. The undoped silicon oxide layer 43 and the interlayer insulator layer 44 constitute the interlayer insulating structure 45 shown in FIG. 1H as a whole.
    When the titanium layer 34 is converted into the titanium silicide layers 35 to 41 in a self-aligned manner with the amorphous silicon layers 27 to 33, the titanium silicide layers 35 to 41 and the silicon / polysilicon layer become salicide. Form the side structure. This salicide structure reduces resistance and accelerates signal transmission.
    However, the conventional method has a problem that the interlayer insulating structure 45 is likely to peel off from the wide titanium silicide layer 41. This is because the heat treatment for phase change coagulates the titanium silicide layer like an island.
    Therefore, it is an important object of the present invention to provide a method for producing a salicide structure that prevents the interlayer insulating structure from peeling off the silicide layer.
    The present inventor contemplated this problem and studied the coagulation phenomenon. Ion implantation of dopant impurities, such as arsenic, phosphorus, or antimony, into silicon, reports the growth of titanium silicide on ion-implanted silicon, Journal of Applied Physics, 1983, 1860-1864. As it is, dopant impurities interfere with the reaction between titanium and silicon, making the titanium silicide layer thinner. Increasing the ten thousand and one ion implanted dopant impurity, the effect of the ion-implanted dopant for formation of the TiSi 2 (Effects of ion-implantation doping on the formation of TiSi 2) Journal of Vacuum Science and Technologies, reported on page 1984, 264-268 Impedance is important as it is, making the titanium silicide layer thinner. The thin titanium silicide layer is more likely to solidify on the doped silicon than the undoped silicon during the heat treatment. Also, even though the titanium silicide layers are solidified on the active region, these titanium silicide layers hardly peel off because the titanium silicide layers are several hundreds of micrometers or less.
    The present inventors further studied the influence of the area on the thickness of the titanium silicide layer. The titanium silicide layer was thicker on the narrow active region than on the wide inactive region. Amorphous silicon promoted the reaction with titanium, and the thickness of the titanium silicide layer was proportional to the thickness of the amorphous silicon layer. However, if the amorphous silicon was partially etched while etching the coating layer 17 of silicon oxide before deposition of titanium, the amount of etched amorphous silicon was proportional to the area of the amorphous silicon layer. This phenomenon stems from the influence of the field oxide layer. The amorphous silicon layer was narrow in the active region, and the center of the active region was close to the thick field oxide layer. On the other hand, the amorphous silicon layer is wide in the inactive region, and the center of the inactive region is spaced in the thick field oxide layer. When etching the coating layer 17, the etchant was consumed to etch the thick field oxide layer in the active region, and the amorphous silicon layer was less etched. However, the etching solution hardly reached the thick field oxide layer in the inactive region, and was consumed for the amorphous silicon layer. For this reason, the amorphous silicon layer in the active region was thicker than the amorphous silicon layer in the inactive region, and therefore the titanium silicide layer on the active region was thicker than the titanium silicide layer on the inactive region.
    The inventors studied the effect of ion implanted dopant impurities on the adhesion between the interlayer insulating layer and the titanium silicide layer as follows. We grow a thick field oxide layer on the 6 inch silicon wafer 50 (see FIG. 2), which in turn forms an active region and inactive peripheral region 52 represented by the active and hatched areas in the central region 51. Defined. The scribe lines 53 extended like a lattice and did not cover the scribe lines 53 and the inactive peripheral region 52 with a thick field oxide layer. The maximum photoactive area was 50 μm square and the width of the scribe lines 53 was 100 microns. Arsenic was implanted into one of the silicon wafers 50, referred to as a first specimen, with an acceleration energy of 50 KeV and a dose of 3 × 10 15 cm −2 . Boron fluoride was ion implanted into another silicon wafer 50 referred to as a second sample at an acceleration energy of 30 KeV and a dose of 3 × 10 15 cm −2 . Arsenic and boron fluoride were ion implanted inside another silicon wafer, referred to as a third sample, under the conditions described above. In another silicon wafer, referred to as the fourth sample, neither arsenic nor boron fluoride was implanted, but only the center region was doped through ion implantation. Heat treatment of the first to fourth samples to activate dopant impurity implanted with ion, and implanting arsenic into the first to fourth samples with arsenic at an acceleration energy of 30 KeV and a dose of 3 x 10 15 cm -2 . Amorphous silicon layers were formed.
    Silicon oxide was removed from the amorphous silicon layers, and titanium silicide layers were formed similarly to the conventional method. The interlayer insulating layers were deposited over the silicon wafers and heated to 840 ° C. for 10 seconds.
    The present inventors observed the first to fourth samples to determine whether the interlayer insulating layer was peeled off from the titanium silicide layer. The interlayer insulating layer was stripped from the titanium silicide layer on the inactive peripheral region 52 and the scribe line 53 in each of the first sample, the second sample, and the third sample. However, in the active regions of the central zone, the interlayer insulating layer did not peel off from the titanium silicide layers. In particular, in the third sample, the interlayer insulating layer was severely peeled off from the titanium silicide layers. On the other hand, in the fourth sample, the interlayer insulating layer was not peeled off from the titanium silicide layers.
    The inventor measured the resistance of the titanium silicide layer. The 1st sample was 4.6 dl / square, the 2nd sample was 4.3 dl / square, the 3rd sample was 6.1 dl / square, and the 4th sample was 4.0 dl / square. Ion implanted arsenic and ion implanted boron fluoride reduced the thickness of the titanium silicide layers to increase resistance.
    We concluded that ion implanted dopant impurities are undesirable for adhesion and resistance.
    In order to achieve the above object, the present invention proposes to prevent the implantation of ion implanted dopant impurities into the light inactive region.
    According to one aspect of the invention, a method of manufacturing a semiconductor device comprises the steps of: (a) providing a silicon substrate; (b) selectively forming a field insulating layer on the major surface of the silicon substrate to define the narrow active region assigned to the circuit component and the light inactive region not assigned to any circuit component; (c) forming a first ion implantation mask on the major surface to cover the light inactive region but not the narrow active region with the first ion implantation mask; (d) implanting first dopant impurities into the narrow active region to form first doped regions that form part of the circuit components; (e) removing the first implant mask; (f) depositing a metal layer on at least first doped regions; And (g) reacting the metal layer with silicon in the first dope regions while applying heat to form conductive metal silicide layers.
    1A to 1H are cross-sectional views showing a conventional method of manufacturing a semiconductor device having a salicide structure;
    2 is a plan view showing a silicon wafer used for research;
    3A to 3H are cross-sectional views showing a method of manufacturing a semiconductor device having a salicide structure; And
    4A to 4D are cross-sectional views showing another method of manufacturing a semiconductor device having a salicide structure.
    * Description of the symbols for the main parts of the drawings *
    60 p-type single crystal silicon substrate 61 n-type well
    60a / 60b: narrow active area 60c: light inactive area
    62: thick field oxide layer 63/64: gate insulating layer
    65/66: gate electrode 67, 76: photoresist ion implantation mask
    68/69: low concentration dope n-type impurity region
    71/72: p-type impurity region 73/74: sidewall spacer
    75: coating layer 79/80: source / drain region
    77/78: high concentration dope n-type impurity region
    The method of the present invention can be more clearly understood from the following description with reference to the accompanying drawings.
    First embodiment
    3A-3H illustrate a method of manufacturing a semiconductor device embodying the present invention. The method starts with preparing a p-type single crystal silicon substrate 60. N-type dopant impurities are implanted into the surface of the p-type silicon substrate 60 to form the n-type well 61.
    A thick field oxide layer 62 is selectively grown on the main surface of the p-type silicon substrate 60 to a thickness of 300 nanometers. The thick field oxide layer 62 defines the narrow active regions 60a and 60b and the light inactive region 60c in the main surface of the p-type silicon substrate 60. Active regions 60a and 60b are allocated to n-channel type field effect transistors and p-channel type field effect transistors, respectively. However, the inactive area 60c is not assigned to any circuit component. Scribe lines (not shown) are formed in the inactive region 60c.
    Silicon oxide is thermally grown to a thickness of 6 nanometers on the active regions 60a / 60b and the inactive region 60c. These silicon oxide layers in the active regions 60a / 60b function as the gate insulating layer 63/64. Chemical vapor deposition is used to deposit polysilicon to a thickness of 150 nanometers across the entire surface of the final semiconductor structure, covering the silicon oxide layers with a polysilicon layer.
    The photoresist solution is spin coated onto a polysilicon layer and baked to form a photoresist layer. A pattern image for the gate electrodes is transferred from a photomask (not shown) to the photoresist layer to form a latent image in the photoresist layer. This latent image is developed to form a photoresist etching mask (not shown) on the polysilicon layer. In this way, a photoresist etch mask is patterned on the polysilicon layer using photolithography techniques. Using a photoresist etching mask, the polysilicon layer is selectively removed by dry etching, and the silicon oxide layers are also selectively etched. As a result, gate electrodes 65/66 are formed on the gate insulating layers 63/64, respectively.
    Photoresist implantation mask 67 is patterned onto the final semiconductor structure using photolithography techniques. The active region 60a is not covered with the photoresist ion implantation mask 67, but the active region 60b and the inactive region 60c are covered with the photoresist ion implantation mask 67.
    Using the photoresist ion implantation mask 67, phosphorus was implanted into the active region 60a at a dose of 5 x 10 13 cm -2 at an acceleration energy of 30 KeV, thereby lowering the concentration as shown in FIG. 3A. The dope n-type impurity regions 68/69 are formed in self-alignment with the gate electrode 65. After ion implantation, the photoresist ion implantation mask 67 is peeled off.
    Photoresist implantation mask 70 is patterned onto the final semiconductor structure using photolithography techniques. The photoresist ion implantation mask 70 does not cover the active region 60b, that is, the n-type well 61. However, the active region 60a and the inactive region 60c are covered with the photoresist ion implantation mask 70.
    Boron difluoride (BF 2 ) was ion-implanted into the active region 60b at a dose of 5 × 10 13 cm −2 with an acceleration energy of 20 KeV, and the n-type well 61 as shown in FIG. 3B. The p-type impurity regions 71/72 are formed in self-alignment with the gate electrode 66 in the interior. After ion implantation, the photoresist ion implantation mask 70 is peeled off.
    Chemical vapor deposition is used to deposit silicon oxide at a thickness of 70 nanometers across the entire surface of the final semiconductor structure, and the silicon oxide layer is etched again to form sidewall spacers (73/74) on both sides of the gate electrodes 65/66. ). Silicon oxide is deposited to a thickness of 70 nanometers over the entire surface of the final semiconductor structure to form a coating layer 75.
    The photoresist implantation mask 76 is patterned on the coating layer 75 using photolithography techniques, and the active region 60b and the inactive region 60c are covered with the photoresist implantation mask 76. The photoresist ion implantation mask 76 does not cover the active region 60a.
    Arsenic is implanted into the active region 60a at a dose of 5 x 10 13 cm -2 at an acceleration energy of 50 KeV to form high concentration dope n-type impurity regions 77/78. These heavily doped n-type impurity regions 77/78 are self-aligned with sidewall spacers 73, and source / drain regions 79/80 together with low-doped n-type impurity regions 68/69. ). These source / drain regions 79/80 have an LDD structure. Further, arsenic is implanted into the gate electrode 65 to reduce the resistance of the gate electrode 65. However, since arsenic implantation into the inactive region 60c is prevented by the photoresist ion implantation mask 76, no n-type impurity region is formed in the inactive region 60c. The final semiconductor structure is shown in FIG. 3C. After ion implantation into the high concentration dope n-type impurity regions 77/78, the photoresist ion implantation mask 76 is removed.
    A photoresist ion implantation mask 81 is patterned on the coating layer 75 using photolithography techniques. Boron fluoride was ion-implanted into the active region 60b at an acceleration energy of 30 KeV and a dose amount of 3 × 10 13 cm −2 , so as to show high concentration of doped p-type impurity regions 82 / 83 is formed self-aligning with the sidewall spacers 74. The high concentration dope p-type impurity regions 82/83 form the p-type source / drain regions 84/85 together with the low concentration dope p-type impurity regions 71/72. These p-type source / drain regions 84/85 have an LDD structure. In addition, boron fluoride is ion implanted into the gate electrode 66 to reduce the resistance of the polysilicon gate electrode 66. However, the injection of boron difluoride into the inactive region 60c is prevented by the photoresist ion implantation mask 81. After ion implantation, the photoresist ion implantation mask 81 is peeled off.
    The n-type source / drain regions 79/80 and the p-type source / drain regions 84/85 are heat treated at 900 ° C. for 20 minutes in a nitrogen atmosphere. The silicon crystals are then cured to activate the implanted dopant impurities.
    Dry etching is used to remove the cladding layer 75 and sputtering is used to deposit titanium to a thickness of 30 nanometers over the entire surface of the final semiconductor structure. In this way, the titanium layer 86 is laminated on the entire surface as shown in FIG. 3E without ion implantation for amorphous.
    When the titanium layer 86 is heated to 650 ° C. for 30 seconds by rapid heat treatment, titanium reacts with silicon to form titanium silicide layers as shown in FIG. 3F. The remaining titanium reacts with nitrogen to form titanium nitride layer 94.
    The titanium nitride layer 94 is etched using a wet etching solution containing ammonia water and hydrogen peroxide water. Titanium silicide layers 87 to 93 remain on the silicon / polysilicon layers 79, 65, 80, 84, 66, 85, and 60c as shown in FIG. 3G. The titanium silicide layers 87 to 93 are rapidly heat treated at 850 ° C. for 10 seconds to reduce the resistance.
    Silicon oxide layer 95 is formed by depositing undoped silicon oxide over the entire surface of the final semiconductor structure. Borosilicate glass, phosphosilicate glass, or borophosphosilicate glass is deposited over the undoped silicon oxide layer 95 to form an interlayer insulating layer 96. The interlayer insulating layer 96 is heat treated to increase the density. The undoped silicon oxide layer 95 and the interlayer insulating layer 96 constitute the interlayer insulating structure 97 as a whole.
    When the titanium layer 86 is converted into titanium silicide layers 87 to 93 in self-alignment with the silicon / polysilicon layers, the titanium silicide layers 87 to 93 and the silicon / polysilicon layers become salicide structures. To form.
    As can be seen from the above description, the high concentration of the n-type dopant impurity and the highly-concentrated p-type dopant impurity into the inactive region 60c by the photoresist ion implantation masks 76/81 Injection is prevented, and the titanium silicide layer 93 is thickly grown on the inactive region 60c by rapid heat treatment. For this reason, even if the titanium silicide is heated after growth, the titanium silicide does not solidify too much, and the interlayer insulating layer 97 is firmly adhered to the titanium silicide layers 87 to 93.
    Second embodiment
    4A-4D illustrate another method of incorporating the present invention. The method of incorporating the second embodiment is similar to the first embodiment until the semiconductor structure shown in FIG. 3D is completed, and the same representations of the corresponding layers and regions shown in FIG. 3D without detailed description of the layers and regions are shown. Reference signs are attached.
    As soon as the LDD structures for n-channel field effect transistors and p-channel field effect transistors were completed, arsenic was implanted with an acceleration energy of 30 KeV and a dose of 3.0 × 10 14 cm -2 without ion implantation masks. As shown, n-type source / drain region 79, gate electrode 65, n-type source / drain region 80, p-type source / drain region 84, gate electrode 66, p-type source / Amorphous silicon layers 100 to 106 are formed over each of the drain region 85 and the p-type single crystal inactive region 60c.
    Amorphous silicon layers 100-106 are 30 nanometers deep and will facilitate the reaction between silicon and titanium. Although arsenic is implanted into the p-type source / drain region (84/85) and the p-type gate electrode (66), because the p-type dopant concentration is very high, the arsenic is the source / drain region (84/85) and The gate electrode 66 cannot be changed to n-type.
    The coating layer 75 is then removed using dry etching, and titanium is deposited to a thickness of 30 nanometers across the entire surface using sputtering. This titanium forms a titanium layer 107, and the amorphous silicon layers 100-106 remain in contact with the titanium layer 107 as shown in FIG. 4B.
    The final semiconductor structure is placed in a nitrogen atmosphere and heated to 650 ° C. for 30 seconds by rapid heat treatment. Titanium reacts with the amorphous silicon layers 100-106 to form an n-type source / drain region 79, an n-type gate electrode 65, an n-type source / drain region 80, a p-type source / drain region ( 84, the p-type gate electrode 66, the p-type source / drain region 85, and the single crystal inactive region 60c, respectively, form titanium silicide layers 108 to 114 self-aligned. The remaining titanium is converted to titanium nitride layer 115. The final semiconductor structure is shown in FIG. 4C.
    The titanium nitride layer 115 is etched using a wet etching solution containing ammonia water and hydrogen peroxide water. Titanium silicide layers 108-114 remain on the silicon / polysilicon layers 79, 65, 80, 84, 66, 85, and 60c. The titanium silicide layers 108-114 are rapidly heat treated at 850 ° C. for 10 seconds to reduce resistance.
    Undoped silicon oxide is deposited over the entire surface of the final semiconductor structure to form silicon oxide layer 116. Borosilicate glass, phosphorosilicate glass, or borophosphosilicate glass is deposited over the undoped silicon oxide layer 116 to form an interlayer insulating layer 117. The interlayer insulating layer 117 is heat-treated at 840 ° C. for 10 seconds to increase the density. The undoped silicon oxide layer 116 and the interlayer insulating layer 117 as a whole constitute an interlayer insulating structure 118.
    As can be seen from the above description, the amorphous silicon layer 106 promotes the reaction between silicon and titanium, and a thick titanium silicide layer 114 is grown on the inactive region 60c by rapid heat treatment, and the inactive Region 60c is prevented from n-type dopant impurities for n-type impurity regions 77/78 and p-type dopant impurities for p-type impurity regions 82/83. For this reason, even if the titanium silicide is heated after growth, the titanium silicide does not hardly solidify, and the interlayer insulating layer 118 is firmly adhered to the titanium silicide layers 108 to 114.
    Although specific embodiments of the invention have been shown and described, it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the spirit and scope of the invention.
    For example, ion implantation into p-type impurity regions may be performed before ion implantation into n-type impurity regions.
    N-type dopant impurities for the lightly doped n-type regions 68/69 and p-type dopant impurities for the lightly doped p-type regions 71/72 may be ion implanted into the inactive region 60c. This is because the dose is 1/10 or 1/100 of the ion implantation for the high concentration doped regions.
    Monocrystalline silicon and polysilicon may be amorphous by using other impurities, such as elements of group IV such as silicon.
    As can be seen from the above description, according to the present invention, a high concentration of n-type dopant impurities implanted by a photoresist ion implantation mask and a high concentration of p-type dopant impurities are prevented from being injected into the inactive region. Since the titanium silicide layer grows thickly on the inactive region by rapid heat treatment, even if the titanium silicide is heated after the growth, the titanium silicide does not hardly solidify, and the interlayer insulating layer is firmly adhered to the titanium silicide layer. As a result, a salicide structure is provided that prevents the interlayer insulating structure from peeling off from the titanium silicide layer.
    权利要求:
    Claims (10)
    [1" claim-type="Currently amended] (a) preparing a silicon substrate 60;
    (b) selectively forming a field insulating layer 62 on the main surface of the silicon substrate to form a narrow active region 60b assigned to a circuit component and an optical inactive region 60c not assigned to any circuit component. Defining;
    (c) forming the first ion implantation mask (81) on the major surface such that the narrow active region is not covered with the first ion implantation mask (81);
    (d) implanting first dopant impurities into the narrow active region to form first doped regions (82/83) forming part of the circuit component;
    (e) removing the first ion implantation mask (81);
    (f) depositing a metal layer (86; 107) on at least the first dope regions; And
    (g) reacting the metal layer with silicon in the first doped regions while applying heat to form conductive metal silicide layers (90/92; 111/113). In
    And in step (d), the first ion implantation mask (81) covers the optical inactive region (60c) to prevent the injection of the first dopant impurity into the optical inactive region.
    [2" claim-type="Currently amended] 2. The method of claim 1, wherein the first dope regions function as highly doped impurity regions 82/83 of a lightly doped drain (LDD) structure. .
    [3" claim-type="Currently amended] The method of claim 1, further comprising converting the surfaces of the first doped regions into amorphous silicon layers 103/105 between steps (e) and (f). Semiconductor device manufacturing method.
    [4" claim-type="Currently amended] 4. A method according to claim 3, wherein the conversion to the amorphous silicon layers (103/105) is performed by ion implantation.
    [5" claim-type="Currently amended] The method of claim 1,
    (h) forming a gate insulating layer (64) on the narrow active region between steps (b) and (c);
    (i) forming a polysilicon gate electrode (66) on the gate insulating layer between (h) and (c);
    (j) the second ion implantation does not cover the narrow active region 60b with the second ion implantation mask 70 between the step (i) and the step (c) but covers the light inactive region 60c. Forming a mask 70;
    (k) a second dopant impurity is implanted at low concentration into the narrow active region 60b between steps (j) and (c) to self-align the second with the polysilicon gate electrode 66; Forming dope regions 71/72;
    (l) removing the second ion implantation mask (70) between (k) and (c); And
    (m) forming sidewall spacers 74 on both sides of said polysilicon gate electrode between steps (l) and (c),
    Wherein the first doped regions 82/83 are self-aligned with the sidewall spacers 74 to form a lightly doped drain (LDD) structure. Manufacturing method.
    [6" claim-type="Currently amended] 6. The amorphous silicon layers of claim 5, wherein the surface of the first doped regions 82/83 and the surface of the polysilicon gate electrode 66 are formed between (e) and (f). 103/104/105),
    And covering the amorphous silicon layers with the metal layer (107).
    [7" claim-type="Currently amended] 2. The field insulating layer 62 further defines another narrow active region 60a having a first conductive type P opposite to the second conductive type N of the narrow active region. ,
    The other narrow active region and the narrow active region are respectively assigned to a first field effect transistor and a second field effect transistor,
    (h) forming a first gate insulating layer (63) and a second gate insulating layer (64) in said other narrow active region and said narrow active region, respectively, between steps (b) and (c);
    (i) forming a first polysilicon gate electrode 65 and a second polysilicon gate electrode 66 on the first gate insulating layer and the second gate insulating layer between steps (h) and (c). Forming each;
    (j) Between the step (i) and the step (c), the narrow active region 60b and the light inactive region 60c are covered with a second ion implantation mask 67, but the other narrow active region 60a ) Forming the second ion implantation mask 67 so as not to cover;
    (k) a second dopant impurity is implanted at low concentration into the other narrow active region between steps (j) and (c) to self-align with the first polysilicon gate electrode 65 Forming second doped regions 68/69 of a second conductivity type;
    (l) removing the second ion implantation mask (67) between step (k) and step (c);
    (m) The other narrow active region 60a and the light inactive region 60c are covered with a third ion implantation mask 70 between the step (l) and the step (c), but the narrow active region 60b ) Forming the third ion implantation mask 70 so as not to cover;
    (n) a third dopant impurity is implanted at low concentration into the narrow active region between the step (m) and the step (c) to self-align with the second polysilicon gate electrode 66; Forming third doped regions 71/72 of conductivity type;
    (o) removing the third ion implantation mask 70 between steps (n) and (c);
    (p) first sidewall spacers 73 and second sidewalls on both sides of the first polysilicon gate electrode and on both sides of the second polysilicon gate electrode between steps (o) and (c). Forming spacers 74;
    (q) Between the step (p) and the step (c), the narrow active region 60b and the light inactive region 60c are covered with a fourth ion implantation mask 76, but the other narrow active region 60a. ) Forming the fourth ion implantation mask 76 so as not to cover;
    (r) ion implanting a fourth dopant impurity in the other narrow active region at a high concentration between steps (q) and (c) to self-align with the first sidewall spacers Forming fourth dope regions 77/78; And
    (s) further removing the fourth ion implantation mask 76 between step (r) and step (c),
    And the first doped regions of the first conductivity type are self-aligned with the second sidewall spacers (74).
    [8" claim-type="Currently amended] 8. The surface of the first doped regions 82/83, the surface of the first and second polysilicon gate electrodes, and the fourth between the step (e) and the step (f). And converting the surface of the dope regions (77/78) into amorphous silicon layers (100-105).
    [9" claim-type="Currently amended] The method of claim 1, wherein the metal layer is formed of titanium.
    [10" claim-type="Currently amended] 2. The method of claim 1, wherein said first ion implantation mask is formed using photolithography techniques.
    类似技术:
    公开号 | 公开日 | 专利标题
    KR0164072B1|1999-02-01|Method of forming shallow junction in a semiconductor device
    EP0139371B1|1988-07-20|Process for manufacturing a mos integrated circuit employing a method of forming refractory metal silicide areas
    US5341028A|1994-08-23|Semiconductor device and a method of manufacturing thereof
    US5231038A|1993-07-27|Method of producing field effect transistor
    US5234850A|1993-08-10|Method of fabricating a nitride capped MOSFET for integrated circuits
    JP2774952B2|1998-07-09|Method for manufacturing semiconductor device
    US4354309A|1982-10-19|Method of manufacturing a metal-insulator-semiconductor device utilizing a graded deposition of polycrystalline silicon
    US6475888B1|2002-11-05|Method for forming ultra-shallow junctions using laser annealing
    EP0138978B1|1988-03-30|Method of manufacturing a semiconductor device having small dimensions
    KR100326694B1|2002-08-08|Method for manufacturing semiconductor device using lateral gettering
    EP0197531B1|1993-07-28|Thin film transistor formed on insulating substrate
    EP0139467B1|1987-12-16|Method of manufacturing an insulated-gate field-effect transistor
    US6900092B2|2005-05-31|Surface engineering to prevent epi growth on gate poly during selective epi processing
    KR100592401B1|2006-06-22|Self-aligned Power Field Effect Transistors in Silicon Carbide
    US5766991A|1998-06-16|CMOS process utilizing disposable silicon nitride spacers for making lightly doped drain
    US6693025B2|2004-02-17|Local interconnect structures for integrated circuits and methods for making the same
    EP0409561B1|1997-09-17|Methods of manufacturing MIS semiconductor devices
    US4249968A|1981-02-10|Method of manufacturing a metal-insulator-semiconductor utilizing a multiple stage deposition of polycrystalline layers
    JP2978736B2|1999-11-15|Method for manufacturing semiconductor device
    JP3029653B2|2000-04-04|Method for manufacturing semiconductor device
    US5693974A|1997-12-02|Elevated source/drain with solid phase diffused source/drain extension for deep sub-micron MOSFETS
    US5326722A|1994-07-05|Polysilicon contact
    US5858843A|1999-01-12|Low temperature method of forming gate electrode and gate dielectric
    US5350698A|1994-09-27|Multilayer polysilicon gate self-align process for VLSI CMOS device
    US6063704A|2000-05-16|Process for incorporating silicon oxynitride DARC layer into formation of silicide polysilicon contact
    同族专利:
    公开号 | 公开日
    JPH10209291A|1998-08-07|
    JP3003796B2|2000-01-31|
    US6228766B1|2001-05-08|
    CN1107976C|2003-05-07|
    CN1189688A|1998-08-05|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    法律状态:
    1997-01-23|Priority to JP9009933A
    1997-01-23|Priority to JP97-9933
    1998-01-23|Application filed by 가네꼬히사시, 닛뽕덴끼가부시끼가이샤
    1998-10-26|Publication of KR19980070802A
    2001-07-12|Application granted
    2001-07-12|Publication of KR100294131B1
    优先权:
    申请号 | 申请日 | 专利标题
    JP9009933A|JP3003796B2|1997-01-23|1997-01-23|Method of manufacturing MOS type semiconductor device|
    JP97-9933|1997-01-23|
    [返回顶部]