JP4865076B2 - Manufacturing method of semiconductor device - Google Patents

Description

This invention relates to a method of manufacturing a semiconductor device, particularly, to a method of manufacturing a semiconductor device capable of improving the characteristics of the device by nitrogen implantation technique.

  Conventionally, in order to suppress the short channel effect of a MOS transistor, it is known to form a shallow source / drain junction surface of a MOS transistor. In order to suppress the short channel effect of a P-channel MOS transistor (hereinafter referred to as a PMOS transistor), it is effective to use a P-type or N-type doped electrode as an electrode material of the PMOS transistor. In order to suppress the short channel effect of the N channel MOS transistor (hereinafter referred to as NMOS transistor), it is effective to use an N-type doped electrode as the electrode material of the NMOS transistor. Further, by utilizing this fact, in a CMOS (Complementary MOS) transistor composed of an NMOS transistor and a PMOS transistor, an N-channel MOS transistor (hereinafter referred to as an NMOS transistor) uses an N-type doped gate electrode, and the PMOS As the transistor, a dual gate CMOS transistor using a P-type doped gate electrode has been proposed.

Next, a method for forming a shallow source / drain junction surface of a PMOS transistor according to the prior art will be described with reference to FIGS. 145 to 147 are diagrams showing conventional PMOS transistors. 145 to 147, 1 is an N-type silicon substrate, 2 is a gate oxide film formed on the N-type silicon substrate 1, 3 is a gate electrode formed on the gate oxide film 2, and 4 is on the gate electrode 3. 5 is a side wall oxide film formed on the side wall of the gate electrode 3, 6 is a P type source / drain region formed on the N type silicon substrate 1 with a channel region 10 interposed therebetween, 7 Is an element isolation oxide film formed on the N-type silicon substrate 1. Conventionally, in order to form a shallow junction surface of the source / drain region 6, boron fluoride ions (BF ₂ +) having a mass number larger than that of the boron ions (B +) as shown in FIG. Had been injected. Alternatively, in order to prevent the channeling phenomenon of boron ions, silicon ions (Si +) or germanium ions (Ge +) are implanted into the source / drain regions 6 as shown in FIG. Then, boron ions (B +) were implanted into the source / drain regions 6 as shown in FIG. 146 (b).

  In addition, when the source / drain junction surface is formed shallowly, there arises a problem that the sheet resistance of the source / drain region increases, so that titanium having high conductivity is formed in the source / drain region as shown in FIG. Providing silicide 8 has been proposed.

  FIG. 148 shows an example of a conventional dual gate CMOS transistor. In FIG. 148, 11 is a P-type silicon substrate, 12 is an element isolation oxide film formed on the P-type silicon substrate 11, 13 is an N-well formed in the P-type silicon substrate 11, and 14 is a P-type silicon substrate 11. A P-well formed therein, 15 is a gate oxide film formed on the P-type silicon substrate 11, and 16 is a polysilicon film formed on the gate oxide film 15, and is doped P-type. Reference numeral 17 denotes a polysilicon film formed on the gate oxide film 15, which is doped N-type. Reference numeral 18 denotes a tungsten silicide film provided on the polysilicon film 16 and the polysilicon film 17. By providing the tungsten silicide film 18 on the polysilicon film 16, the gate electrode of the polycide gate structure of the PMOS transistor is formed. Further, by providing the tungsten silicide film 18 on the polysilicon film 17, the gate electrode of the polycide gate structure of the NMOS transistor is formed. 19 is an oxide film provided on the gate electrodes of the PMOS transistor and NMOS transistor, 20 is a side wall oxide film provided on the side walls of the gate electrodes of the PMOS transistor and NMOS transistor, and 21 is a P-type doped PMOS transistor. The source / drain region is formed in the N well 13 with the gate electrode interposed therebetween. Reference numeral 22 denotes a source / drain region of an N-type doped NMOS transistor, which is formed in the P well 14 with a gate electrode interposed therebetween.

  Next, a method of manufacturing the dual gate CMOS transistor shown in FIG. 148 will be described. First, an element isolation oxide film 12 is formed on the main surface of a P-type silicon substrate 11, and an N well 13 serving as a PMOS transistor forming region and a P well 14 serving as an NMOS transistor forming region are formed (FIG. 149). Next, an oxide film 15a is formed, a polysilicon film 9 is deposited by CVD, and then a tungsten silicide film 18a is deposited by sputtering (FIG. 150). Next, the PMOS transistor formation region is covered with a resist 25, and arsenic ions (As +) are implanted into the polysilicon film 9 in the NMOS transistor formation region (FIG. 151). Next, as shown in FIG. 152, after removing the resist 25, the formation region of the NMOS transistor is covered with a resist 26, and boron fluoride ions are implanted into the polysilicon film 9 in the formation region of the PMOS transistor (FIG. 152). Next, after removing the resist 26, an oxide film is deposited by CVD (not shown), and the oxide film, tungsten silicide film 18a and polysilicon film 9 are formed into the shape of the gate electrode by using photolithography and anisotropic etching. Then, an oxide film 19, a tungsten silicide film 18, and polysilicon films 16a and 17a are formed (FIG. 153). Next, an oxide film is deposited by CVD (not shown), and etched back to form a sidewall oxide film 20 on the side wall of the gate electrode (FIG. 154). Next, the formation region of the PMOS transistor is covered with a resist 27, and arsenic ions are implanted into the formation region of the NMOS transistor (FIG. 155). Next, after removing the resist 27, the formation region of the NMOS transistor is covered with the resist 28, and boron fluoride ions are implanted into the formation region of the PMOS transistor (FIG. 156). Next, after removing the resist 28, heat treatment is performed to activate the implanted ions, and the N-type doped polysilicon film 16, the P-type doped polysilicon film 17, and the N + type A source / drain region 22 and a P + -type source / drain region 21 are formed to complete a dual gate CMOS transistor having a polycide gate structure (FIG. 157).

  FIG. 158 is a diagram showing another example of a conventional dual gate CMOS transistor. In FIG. 158, 11-17 and 20-22 are the same as or correspond to the conventional example shown in FIG. Reference numeral 23 denotes a titanium silicide film formed in a self-aligned manner on the source / drain regions 21 and 22 and the polysilicon films 16 and 17. A structure obtained by siliciding the surface of the polysilicon film and the source / drain region as a gate electrode in a self-aligning manner is called a salicide (self-aligned silicide) structure and is a source / drain junction surface. This is effective in suppressing the sheet resistance of the source / drain region, which becomes a problem when the layer is formed shallow. In particular, titanium silicide is a promising material as a high heat resistance wiring of a semiconductor device because it has a low specific resistance even in metal silicide and has good adhesion.

  Next, a manufacturing method of the dual gate CMOS transistor shown in FIG. 158 will be described. First, an element isolation oxide film 12 is formed on the main surface of a P-type silicon substrate 11, and an N well 13 serving as a PMOS transistor forming region and a P well 14 serving as an NMOS transistor forming region are formed (FIG. 159). Next, after sequentially depositing an oxide film 15a and a polysilicon film (not shown) on the N well 13 and the P well 14, as shown in FIG. 160, a polysilicon film is formed by using photolithography and anisotropic etching. Is patterned into the shape of a gate electrode to form a polysilicon film 8. Next, as shown in FIG. 161, after a sidewall oxide film 20 is formed on the side wall of the polysilicon film 8, the formation region of the PMOS transistor is covered with a resist 25, and the source / drain region of the NMOS transistor and the polysilicon film 8 are covered. Arsenic ions are implanted (FIG. 161). Next, as shown in FIG. 162, after removing the resist 25, the formation region of the NMOS transistor is covered with the resist 26, and boron fluoride ions are implanted into the source / drain region of the PMOS transistor and the polysilicon film 8 (FIG. 162). ). Next, after removing the resist 26, titanium is sputtered on the entire surface and heat treatment is performed to react silicon and titanium, and the titanium silicide film 23 is formed on the source / drain regions 21 and 22 and the polysilicon films 16 and 17. Is formed (FIG. 163).

  As described above, in the dual gate CMOS transistor, in order to connect two gate electrodes of different materials, that is, the P-type doped polysilicon film 16 and the N-type doped polysilicon film 17 are connected. A method is adopted in which the gate electrode has a polycide gate structure of a polysilicon film and a tungsten silicide film, or the gate electrode is silicided. In particular, when a salicide structure is incorporated in a dual gate CMOS transistor, the increase in sheet resistance in the source / drain region, which becomes a problem when the source / drain junction surface is formed shallow in order to suppress the short channel effect, can be mitigated. it can.

  As one of semiconductor devices, there is a thin film transistor (hereinafter referred to as TFT) using polysilicon, which is an important device as a load transistor of a highly integrated SRAM or a driving transistor for a liquid crystal display. However, due to the demand for higher integration and higher performance of TFT application elements, miniaturization of TFT itself, improvement of electrical characteristics, and improvement of reliability are required.

  Important issues for TFT miniaturization are suppression of the short channel effect caused by diffusion of impurity ions forming the source / drain regions into the channel region, and improvement of hot carrier resistance.

  Next, the structure of a conventional TFT will be described. Here, for convenience, a P-channel MOS-TFT (hereinafter referred to as a PMOS-TFT) will be described as an example. FIG. 164 is a sectional structural view showing a conventional PMOS-TFT. In the figure, 101 is a semiconductor substrate, 102 is an insulating film formed on the semiconductor substrate, 103 is formed on the insulating film, is a P-type doped gate electrode, and 104 is formed on the gate electrode. A gate insulating film, 105 is a polysilicon layer formed on the gate insulating film, 105a is a channel region formed in the polysilicon layer 105, 105b is a P-type source region formed in the polysilicon layer 105, 105c is a P-type drain region formed in the polysilicon layer. FIG. 165 is a bird's eye view of the upper part of the TFT shown in FIG.

Next, a manufacturing method of the TFT shown in FIG. 164 will be described. First, a high temperature oxide film is deposited on the semiconductor substrate 101 by a CVD method or the like to form an insulating film 102. Next, a non-doped polysilicon layer 103a is deposited on the insulating film 102 by a CVD method or the like, and a P-type doped polysilicon layer 103a is implanted by implanting ions that form P-type such as boron (B) ions. (FIG. 166). Next, the resist 107 is patterned in the shape of the gate electrode, and the polysilicon layer 103a is anisotropically etched to form the gate electrode 103b (FIG. 167). Next, after removing the resist 107, a gate insulating film 104 is formed by gate oxidation, and a non-doped polysilicon layer is deposited on the gate insulating film 104 by a CVD method or the like. Thereafter, arsenic ( As) is ion-implanted (not shown). Next, a resist is deposited, and patterning is performed using a photoengraving process so as to leave a region to be a channel region, a source region, and a drain region, and then anisotropic etching is performed using the resist as a mask to form a desired shape. A polysilicon layer 105 is formed (FIG. 168). Next, after removing the resist formed in the previous step, a resist 108 is formed again on the channel region, and boron fluoride (BF ₂ ) is ion-implanted using the resist 108 as a mask (FIG. 169). Next, a heat treatment for activating the implanted impurities is performed to form the gate electrode 103, the source region 105b, and the drain region 105c, and the TFT shown in FIG. 164 is completed.

  Further, there is a nonvolatile semiconductor memory device as one of the semiconductor devices. Among them, there is known an EEPROM (Electrically Erasable and Programmable Read Only Memory) capable of freely programming data and electrically writing and erasing data. ing. This EEPROM has the advantage that both writing and erasing can be performed electrically, but it has the disadvantage that it is difficult to achieve high integration because it requires two transistors in the memory cell. In view of this, there has been proposed a flash EEPROM in which a memory cell is composed of a single transistor, and written information charges can be erased collectively. These are disclosed, for example, in US Pat. No. 4,686,619.

  FIG. 170 is a sectional view showing a conventional stacked gate type flash EEPROM. The structure of a conventional flash EEPROM will be described with reference to FIG.

  170, a drain region 208 and a source region 209 are formed on the main surface of a P-type silicon substrate 201 so as to sandwich a channel region 215 with a predetermined interval. A floating gate electrode 203 is formed on the channel region 215 through a thin oxide film 202 having a thickness of about 100 mm. A control gate electrode 205 is formed on the floating gate electrode 203 via an interlayer insulating film 204 so as to be electrically isolated from the floating gate electrode 203. The floating gate electrode 203 and the control gate electrode 205 are formed of a polysilicon layer, and are doped P + type by implantation of boron (B) or the like. The thermal oxide film 216 is formed so as to cover the P-type silicon substrate 201 and the floating gate electrode 203 and control gate electrode 205 made of a polysilicon layer. On the thermal oxide film 216, a smooth coat film 212 made of an oxide film or the like is formed. 214 is a wiring layer made of an aluminum alloy or the like.

Next, the operation of the flash EEPROM will be described with reference to FIG. First, in a write operation of a flash EEPROM using CHE (Channel Hot Electron), a voltage VD1 of 6V to 8V is applied to the drain region 208, and a voltage V _G1 of 10V to 15V is applied to the control gate electrode 205. Due to the application of the voltages V _D1 and V _G1 , some of the high energy electrons generated in the vicinity of the drain region 208 and the oxide film 202 are caused by the electric field due to the voltage V _G1 applied to the control gate electrode 205. It is attracted and injected into the floating gate electrode 203. When electrons are accumulated in the floating gate electrode 203 in this way, the threshold voltage V _TH of the control gate transistor becomes higher than a predetermined value. A state in which the threshold voltage VTH is higher than a predetermined value is a written state, which is called a “0” state.

In addition, with reference to FIG. 172, the writing operation of the flash EEPROM using SHE (Substrate Hot Electron) will be described. For example, in an N channel type flash EEPROM formed in a P well 222 in an N type silicon substrate 221, a drain region 208 and a source region 209 are grounded, and a voltage V _G2 of 10V to 15V is applied to the control gate electrode 205. To do. Further, a voltage VB2 of -5V to -10V is applied to the substrate electrode 223. By applying the voltages V _G2 and VB2, a forward bias is applied to the PN junction formed by the N-type silicon substrate 221 and the P well 222, and an on-current is generated. Some of these electrons are attracted and injected into the floating gate electrode 203 by the electric field generated by the voltage V _G2 applied to the control gate gate electrode 205.

In addition, with reference to FIG. 173, the writing operation of the flash EEPROM using the FN (Fowler-Nordheim) tunnel phenomenon will be described. For example, in FN writing at the drain end, a voltage VD3 of -10 V to -12 V is applied to the drain region 208, the control gate electrode 205 is held at the ground potential, and the source region 209 is held in a floating state. Electrons in the floating gate electrode 203 pass through the thin oxide film 202 by the FN tunnel phenomenon due to an electric field caused by the voltage V _D3 applied to the drain region 208. As the electrons are accumulated in the floating gate electrode 203 in this way, the threshold voltage V _TH of the control gate transistor is increased.

Next, the erase operation will be described. A voltage V _S of 10V to 12V is applied to the source region 209, the control gate electrode 205 is held at the ground potential, and the drain region 208 is held in a floating state. Due to the electric field caused by the voltage V _S applied to the source region 209, electrons in the floating gate electrode 203 pass through the thin oxide film 202 by the FN tunnel phenomenon. In this way, the electrons in the floating gate electrode 203 are extracted, whereby the threshold voltage V _TH of the control gate transistor is lowered. This state in which the threshold voltage V _TH is lower than a predetermined value is an erased state and is referred to as a “1” state.

Further, in the read operation, the voltage of 5V VG4, the drain region 208 voltage V _D4 of 1V~2V is applied to the control gate electrode 205. At this time, the determination of “1” or “0” described above is performed depending on whether a current flows in the channel region of the control gate transistor, that is, whether the control gate transistor is in an on state or an off state. As a result, information is read out.

Further, the coupling ratio in the flash EEPROM will be described with reference to FIG. Since the flash EEPROM has a gate electrode having a two-layer structure, the voltage applied to the control gate electrode 205 is applied to the channel region via the floating gate electrode 203. That is, even if the accumulated charge amount of the floating gate electrode 203 is the same and the same potential is applied to each terminal, the potential of the floating gate electrode 203 varies depending on the structure of the interlayer insulating film 204 and the oxide film 202. The potential V _FG of the floating gate electrode 203 includes the threshold voltage V _TH , the floating gate electrode 203 and the control gate electrode 205 in addition to the potential applied to each terminal such as the control gate voltage V _CG , the source voltage V _S and the drain voltage V _D. capacity (C _FC), the capacity of the floating gate electrode 203 and the substrate 201 (C _FB), the capacitance between the floating gate electrode 203 and the source region 209 (C _FS) and the capacitance between the floating gate electrode 203 and the drain region 208 (C _FD ) and is approximately given by

V _FG = C _FC V _CG / C _TOTAL + C _FD V _D / C _TOTAL + (C _FD + C _FB ) V _S / C _TOTAL + C _FB V TH / C _TOTAL + Q _FG / C _TOTAL (1)
Q _FG = C _FC (V _FG- V _CG ) + C _FD (V _FG- V _D ) + C _FS (V _FG- V _S ) + C _FB (V _FG- V _TH- V _S )
However, C _TOTAL = C _FC + C _FD + C _FS + C _FB
From the expression (1), the potential V _CG of the control gate electrode 203 affects the potential V _FG of the floating gate electrode 203 in a form multiplied by C _FC / C _TOTAL called a coupling ratio. That is, when the coupling ratio is increased, the potential of the floating gate electrode 203 is increased even if the potential applied to the control gate electrode 205 is the same. Therefore, if the coupling ratio is large, even if the same potential is applied to the control gate electrode 205, the potential of the floating gate electrode 203 becomes high. Therefore, the larger the coupling ratio, the easier it is to control the transistor operation by the potential applied to the control gate electrode 205.

In the semiconductor device described above, when data writing and erasing are performed using the FN tunnel phenomenon, the oxide film 202 is destroyed with a certain probability, so that the reliability of the element is lowered.
Further, when electrons tunnel through the oxide film 202, the electrons injected into the oxide film 202 are trapped in the oxide film 202 with a certain probability, and an interface state is generated at the interface between the silicon substrate 201 and the oxide film 202. To do. Due to the generated interface state, the reliability of the oxide film 202 is lowered, causing a problem that the threshold voltage is changed and the current driving capability is lowered. In addition, since a high potential is applied to the floating gate electrode 203, the source region 209, or the drain region 208 at the time of data writing or erasing, the interface between the drain region 208 and the oxide film 202 or the source region 209 and the oxide film 202 is high. An electric field is generated. In particular, since adjacent memory cells share the drain region 208, a potential is also applied to the drain region 208 of the non-selected cell when data is written. Since the control gate electrode 205 in the cell non-selected state is held at the ground potential, a high electric field is generated between the floating gate electrode 203 and the drain region 208 in such a non-selected cell. As shown in FIG. 175, band-to-band tunneling occurs, and electron / hole pairs are generated. The generated holes are injected into the oxide film 202 with a certain probability, creating an interface state at the interface between the silicon substrate 201 and the oxide film 202, and reducing the reliability of the oxide film 202.

In order to prevent such a decrease in the reliability of the oxide film 202, a method for suppressing the generation of interface states at the interface between the silicon substrate 201 and the oxide film 202 has been proposed. When the oxide film 202 is subjected to RTN (Rapid Thermal Nitridation) treatment and nitrogen is contained in the oxide film 202, the nitrogen terminates dangling bonds in the oxide film 202, so that charges are trapped in the oxide film 202. Can be prevented. The RTN process is a process of performing an annealing process for a very short time in a reactive gas atmosphere containing nitrogen such as ammonia (NH ₃ ). As a result, nitrogen is taken into the silicon substrate 201 and the oxide film 202.

  FIG. 176 shows a conventional embedded channel type flash EEPROM. 176, reference numerals 201 to 205, 208, 209, 215, and 216 denote the same or corresponding parts as those in FIG. Reference numeral 217 denotes an N-type impurity layer formed in the channel region 215, and reference numeral 218 denotes a P-type impurity layer formed below the N-type impurity layer 217. The N-type impurity layer 217 and the P-type impurity layer 218 form a buried channel layer. In such a buried channel type flash EEPROM, unlike the surface channel type flash EEPROM, a high electric field is not applied between the source region 209 or the drain region 208 and the oxide film 202. The occurrence of tunneling can be suppressed. Therefore, generation of holes due to band-to-band tunneling can be prevented during data writing or erasing, so that holes can be prevented from being injected into the oxide film 202.

JP-A-3-46238 JP-A-3-44075 JP 62-177919 A

The conventional MOS transistor has the following problems.
Conventionally, a method for forming a source / drain region of a PMOS transistor has been formed by implanting boron fluoride ions in order to form a shallow source / drain junction surface. Therefore, there is a problem that fluorine contained in boron fluoride ions hinders the reaction between titanium and silicon when titanium silicide is formed, and a good titanium silicide film cannot be formed.

  In addition, since the conventional method for forming the source / drain regions of the PMOS transistor is preamorphized by silicon ions or germanium ions, high-temperature heat treatment for crystal recovery is required. Accordingly, there is a problem in that the recovery of the crystal becomes insufficient and the junction leakage current increases due to the reduction of the heat treatment, which is a condition for forming the source / drain junction surface shallow.

  Further, in order to form the source / drain junction surface of the NMOS transistor shallowly, it is necessary to reduce the heat treatment. As a result, there is a problem that the crystal recovery is insufficient and the junction leakage current is increased.

  Further, in the conventional source / drain region formation method, both the PMOS transistor and the NMOS transistor have a problem that impurities implanted by the heat treatment for activation diffuse and it is difficult to form a shallow junction. .

  In addition, in the dual gate CMOS transistor, a boron ion penetrates the gate oxide film from the P-type doped gate electrode of the PMOS transistor and enters the channel region during the heat treatment process, thereby changing the threshold voltage of the transistor. was there.

  In particular, in a dual gate CMOS transistor having a polycide gate structure, arsenic ions from the N-type doped gate electrode and boron ions from the P-type doped gate electrode are diffused in the silicide during the heat treatment process. Thus, there is a problem that the work function of the gate electrode is changed and the threshold voltage of the transistor is changed.

  Furthermore, in the conventional NMOS transistor and PMOS transistor, since the gate electrode is doped with an impurity, the gate oxide film deteriorates due to the diffusion action of the impurity during the heat treatment process. There was a problem that it was not possible to get enough.

Further, the conventional TFT has the following problems due to the miniaturization in recent years.
Due to the heat treatment after the source / drain implantation, the impurities in the source region and the drain region are thermally diffused and diffused to the channel region, so that punch-through occurs and the original transistor operation is not performed.

  When the electric field applied to the drain end is increased in the off state, hot carriers are generated and the reliability of the element is deteriorated.

The conventional flash EEPROM has the following problems.
Conventionally, an RTN process has been used as a method for introducing nitrogen into the oxide film 202. Since the RTN process is often performed in an ammonia atmosphere, as shown in FIG. 177, the oxide film 202 contains not only nitrogen but also hydrogen. This hydrogen doping has a problem that the reliability of the oxide film 202 is lowered. Furthermore, there is a problem that hydrogen and nitrogen are also implanted into the silicon substrate 201 due to the manufacturing method.

  Further, in the RTN process, the silicon substrate 201 is exposed to a high temperature of about 1100 ° C., and since the process is performed in a short time, the temperature around the silicon substrate 201 is rapidly changed. Therefore, there is a problem that a temperature distribution is generated in the surface of the silicon substrate 201, and slit-like defects are generated due to a difference in expansion coefficient.

  Further, in the conventional flash EEPROM, since impurities such as boron are implanted into the control gate electrode or the floating gate electrode, the implanted impurities penetrate into the interlayer insulating film 204 or the oxide film 202 during the heat treatment process, and the interlayer insulating film There is a problem that the film quality of the film 204 or the oxide film 202 deteriorates.

  In particular, when boron penetrates the oxide film 202 and enters the channel region 215, there is a problem that the threshold voltage of the flash EEPROM is changed.

  Further, the potential applied to the control gate electrode 205 is applied to the floating gate electrode 203 in a form multiplied by the coupling ratio. For this reason, it is necessary to apply a potential to the control gate electrode 205 while reducing the potential due to the coupling ratio. For example, when writing is performed in a device having a coupling ratio of 0.5, 5V is applied to the floating gate electrode 203. If applied, a voltage of about 10 V must be applied to the control gate electrode 205. That is, in order to ensure the same operation, the smaller the coupling ratio, the higher the voltage that must be applied to the control gate electrode 205, which makes it difficult to reduce the power supply voltage of the flash EEPROM.

  Conventionally, in order to improve the coupling ratio, a method of using a nitride film having a relative dielectric constant higher than that of the oxide film as the interlayer insulating film 204 has been proposed. However, if the interlayer insulating film 204 is formed only of the nitride film, the leakage current There is a problem that becomes larger. In order to prevent this leakage current problem, a composite film of a nitride film and an oxide film may be used as the interlayer insulating film 204, but eventually the interlayer insulating film 204 becomes thick and the coupling ratio cannot be increased. was there.

  Further, when considering miniaturization of elements, it is difficult to form a shallow source / drain junction depth by diffusion of impurities implanted by source / drain implantation. there were.

  Further, in the conventional buried channel type flash EEPROM, since it is difficult to form a buried channel layer shallow due to diffusion of impurities implanted into the buried channel region, a source / source voltage is applied at the potential applied to the control gate electrode 205. There is a problem in that it becomes impossible to control the drain-to-drain current, and the source / drain breakdown voltage characteristics such as punch-through deteriorate.

  The present invention has been made to solve the above-described problems, and forms a shallow source / drain junction surface, prevents lateral diffusion of impurities implanted in the source / drain region, and is doped in the gate electrode. Semiconductor device capable of suppressing penetration of boron ions, suppressing diffusion of impurities doped in the gate electrode, improving hot carrier resistance, improving reliability of oxide film and interlayer insulating film, and lowering power supply voltage, and manufacturing the same It aims to provide a method.

A method of manufacturing a semiconductor device according to the present invention includes a step of forming an oxide film on a semiconductor substrate, a step of forming a polysilicon film on the oxide film, a step of implanting first nitrogen ions into the polysilicon film, After implanting the first nitrogen ions, patterning the polysilicon film to form a gate electrode, patterning the oxide film to form the gate oxide film, forming the insulating film on the side surface of the gate electrode, After forming the insulating film, implanting second nitrogen ions into the semiconductor substrate with the gate electrode interposed therebetween, forming a nitrogen doping region in the semiconductor substrate, and after forming the insulating film, the projected range of the second nitrogen ions A step of implanting impurity ions into the semiconductor substrate with a larger projection range across the gate electrode to form source / drain regions including a nitrogen-doped region, and a step of implanting impurities into the semiconductor substrate After a heat treatment, and a step of precipitating the nitrogen into the gate oxide film.

  Since the present invention is configured as described above, the following effects can be obtained.

  In the method of manufacturing a semiconductor device according to the present invention, nitrogen ions are implanted, and impurity ions having a range larger than the range of nitrogen ions are implanted to form the source / drain regions. A nitrogen doping region can be formed.

1 is a cross-sectional structure diagram of a PMOS transistor according to a first embodiment of the present invention. It is a figure which shows the profile of the depth direction of a source / drain region. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. FIG. 6 is a manufacturing process diagram of the PMOS transistor according to the first embodiment of the present invention. It is a cross-sectional structure diagram of a PMOS transistor of the present invention. It is a figure which shows the profile of the depth direction of a source / drain region. FIG. 4 is a cross-sectional structure diagram of a PMOS transistor according to a second embodiment of the present invention. It is a figure which shows the profile of the depth direction of a gate electrode and a gate oxide film. It is a manufacturing process figure of the PMOS transistor by the 2nd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 2nd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 2nd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 2nd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 2nd Embodiment of this invention. It is a figure explaining the injection | pouring conditions of nitrogen. It is a figure which shows the result of having evaluated the reliability of an oxide film. It is a figure which shows the nitrogen injection amount dependence of the variation | change_quantity of the threshold voltage by hot carrier injection | pouring of a PMOS transistor. FIG. 6 is a cross-sectional structure diagram of a PMOS transistor according to a third embodiment of the present invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a manufacturing process figure of the PMOS transistor by the 3rd Embodiment of this invention. It is a cross-sectional structure diagram of a PMOS transistor of the present invention. FIG. 6 is a cross-sectional structure diagram of an NMOS transistor according to a fourth embodiment of the present invention. It is a figure which shows the profile of the depth direction of a gate electrode and a gate oxide film. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 4th Embodiment of this invention. It is a figure which shows the nitrogen injection amount dependence of the variation | change_quantity of the threshold voltage by hot carrier injection | pouring of an NMOS transistor. FIG. 9 is a cross-sectional structure diagram of an NMOS transistor according to a fifth embodiment of the present invention. It is a figure which shows the profile of the depth direction of a source / drain region. It is a manufacturing process figure of the NMOS transistor by the 5th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 5th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 5th Embodiment of this invention. It is a manufacturing process figure of the NMOS transistor by the 5th Embodiment of this invention. 1 is a cross-sectional structure diagram of an NMOS transistor of the present invention. It is sectional drawing of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 6th Embodiment of this invention. It is sectional drawing of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a figure which shows the profile of the depth direction of a gate electrode and a gate oxide film. It is a figure which shows the profile of the depth direction of a gate electrode and a gate oxide film. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is a manufacturing process figure of the dual gate CMOS transistor by the 7th Embodiment of this invention. It is sectional structure figure of TFT by the 8th Embodiment of this invention. It is a figure which shows the impurity profile in the a-a 'cross section of PMOS-TFT. It is a figure which shows the impurity profile in the b-b 'cross section of PMOS-TFT. It is a manufacturing-process figure of TFT by the 8th Embodiment of this invention. It is a manufacturing-process figure of TFT by the 8th Embodiment of this invention. It is a manufacturing-process figure of TFT by the 8th Embodiment of this invention. It is a figure which shows the impurity profile in the a-a 'cross section of NMOS-TFT. It is a figure which shows the impurity profile in the b-b 'cross section of NMOS-TFT. FIG. 14 is a cross-sectional structure diagram of a TFT according to a tenth embodiment of the present invention. It is a figure which shows the impurity profile in the a-a 'cross section of PMOS-TFT. FIG. 30 is a manufacturing process diagram of a TFT according to the tenth embodiment of the invention. FIG. 30 is a manufacturing process diagram of a TFT according to the tenth embodiment of the invention. FIG. 30 is a manufacturing process diagram of a TFT according to the tenth embodiment of the invention. It is a figure which shows the impurity profile in the a-a 'cross section of NMOS-TFT. FIG. 16 is a cross-sectional structure diagram of a TFT according to an eleventh embodiment of the present invention. It is a figure which shows the impurity profile in the a-a 'cross section of PMOS-TFT. FIG. 38 is a manufacturing process diagram of a TFT according to the eleventh embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the eleventh embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the eleventh embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the eleventh embodiment of the present invention. It is a figure which shows the impurity profile in the a-a 'cross section of NMOS-TFT. FIG. 34 is a bird's eye view of a TFT according to a twelfth embodiment of the present invention. It is A-A 'sectional drawing of TFT. It is B-B 'sectional drawing of TFT. It is a figure which shows the impurity profile in the a-a 'cross section of TFT. It is a figure which shows the impurity profile in the b-b 'cross section of TFT. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a TFT according to the twelfth embodiment of the present invention. FIG. 34 is a sectional structural view of a flash EEPROM according to a thirteenth embodiment of the invention. It is a figure which shows the profile of the depth direction of a gate electrode. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a thirteenth embodiment of the present invention. FIG. 34 is a cross-sectional structure diagram of a flash EEPROM according to a fourteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a fourteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a fourteenth embodiment of the present invention. It is a manufacturing process diagram of a flash EEPROM according to a fourteenth embodiment of the present invention. FIG. 28 is a sectional structural view of a flash EEPROM according to a fifteenth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a flash EEPROM according to a fifteenth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a flash EEPROM according to a fifteenth embodiment of the present invention. FIG. 34 is a cross-sectional structure diagram of a flash EEPROM according to a sixteenth embodiment of the present invention. FIG. 38 is a manufacturing process diagram for a flash EEPROM according to a sixteenth embodiment of the present invention; FIG. 38 is a manufacturing process diagram for a flash EEPROM according to a sixteenth embodiment of the present invention; FIG. 38 is a manufacturing process diagram for a flash EEPROM according to a sixteenth embodiment of the present invention; FIG. 38 is a manufacturing process diagram for a flash EEPROM according to a sixteenth embodiment of the present invention; FIG. 38 is a cross-sectional structure diagram of a flash EEPROM according to a seventeenth embodiment of the present invention. It is a figure which shows the profile of the depth direction of a drain region. FIG. 38 is a manufacturing process diagram of a flash EEPROM according to a seventeenth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a flash EEPROM according to a seventeenth embodiment of the present invention. FIG. 38 is a manufacturing process diagram of a flash EEPROM according to a seventeenth embodiment of the present invention. FIG. 38 is a cross-sectional structure diagram of a flash EEPROM according to an eighteenth embodiment of the present invention. FIG. 38 is a cross-sectional structure diagram of a flash EEPROM according to a nineteenth embodiment of the present invention. FIG. 38 is a cross-sectional structure diagram of a flash EEPROM according to a nineteenth embodiment of the present invention. It is a sectional view of a conventional PMOS transistor. It is a sectional view of a conventional PMOS transistor. It is a sectional view of a conventional PMOS transistor. It is sectional drawing of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is sectional drawing of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a manufacturing process figure of the conventional dual gate CMOS transistor. It is a cross-sectional structure diagram of a conventional TFT. It is a bird's-eye view of the conventional TFT. It is a manufacturing process figure of the conventional TFT. It is a manufacturing process figure of the conventional TFT. It is a manufacturing process figure of the conventional TFT. It is a manufacturing process figure of the conventional TFT. It is sectional drawing of the conventional flash EEPROM. FIG. 6 is a cross-sectional structure diagram for explaining the operation of the flash EEPROM. FIG. 6 is a cross-sectional structure diagram for explaining the operation of the flash EEPROM. FIG. 6 is a cross-sectional structure diagram for explaining the operation of the flash EEPROM. FIG. 3 is a cross-sectional structure diagram for explaining a coupling ratio of a flash EEPROM. FIG. 4 is a cross-sectional structure diagram for explaining band-to-band tunneling generated in a flash EEPROM. It is a cross-sectional structure diagram of a conventional buried channel type flash EEPROM. It is a figure which shows the manufacturing method of the conventional semiconductor device.

(Embodiment 1)
Next, an embodiment of the present invention will be described. FIG. 1 is a sectional view showing a PMOS transistor according to a first embodiment of the present invention. In FIG. 1, reference numerals 1, 4 to 7 and 10 are the same as or correspond to those in FIG. A hatched portion 30 indicates a nitrogen doping region and exists in the P + -type gate electrode 35 and the gate oxide film 36. FIG. 2 is a diagram showing profiles in the depth direction of the P + -type gate electrode 35 and the gate oxide film 36 of the PMOS transistor shown in FIG. 2 that nitrogen is deposited in the gate oxide film 36. FIG. Here, the precipitation of nitrogen indicates a state in which nitrogen is trapped at a certain position and the concentration becomes high.

Next, a manufacturing process of the PMOS transistor shown in FIG. 1 will be described. First, after an element isolation oxide film 7 is formed on the N-type silicon substrate 1 by a normal element isolation process, an oxide film 36a of about 100 mm is formed by thermal oxidation, and a polysilicon film 35a of about 2000 mm is formed by CVD. (FIG. 3). Then implanting nitrogen ions should come projected range centered on top of the polysilicon film 35a under the condition of ^{20keV, 4 × 10 15 / cm} 2 ( Figure 4). Next, boron ions are implanted into the polysilicon film 35a under conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 5). Next, an oxide film having a thickness of about 2000 mm is deposited by CVD (not shown), and the oxide film and the polysilicon film 35a are patterned into the shape of the gate electrode using photolithography and anisotropic etching. 4 and the gate electrode 35b are formed (FIG. 6). Next, an oxide film of about 800 mm is deposited by CVD (not shown) and etched back to form the sidewall oxide film 5 and the gate oxide film 36b, and then boron fluoride ions are added to the source / drain regions at 20 keV. Injection is performed under the condition of 4 × 10 ¹⁵ / cm ² (FIG. 7). Next, a source / drain region 6 and a gate electrode 35 shown in FIG. 1 are formed by activating the implanted impurities by applying a heat treatment at 850 ° C. for about 20 minutes. During this heat treatment, the nitrogen doped on the upper portion of the gate electrode 35b is thermally diffused, but the nitrogen is segregated in the gate oxide film 36b, and as shown in FIG. 2, the gate oxidation having a nitrogen concentration peak is present. A film 36 is formed.

  Here, the nitrogen injection conditions during the manufacturing process will be described in more detail. If the standard deviation of the projected range RP of nitrogen is ΔRP, the projected range RP of nitrogen will be in the gate electrode 35 at a position above the position of 5 × ΔRP from the interface between the P + -type gate electrode 35 and the gate oxide film 36. (FIG. 8). If the position is lower than this condition, the gate oxide film 36 may be damaged by nitrogen implantation.

  In the above description, the doping to the gate electrode 35 and the doping to the source / drain region 6 are performed in separate steps, but the doping to the gate electrode 35 is also performed in conjunction with the doping to the source / drain region 6. There is no problem. Further, boron fluoride may be ion-implanted for doping the gate electrode. Furthermore, in the above-described embodiment, the case of only the PMOS transistor has been described. However, the PMOS transistor may be added to a part of the CMOS transistor, or the manufacturing process of the PMOS transistor may be added to a part of the manufacturing process of the CMOS transistor. Good.

  Next, the effect of the invention in this embodiment will be described. Since the gate electrode 35 is doped with nitrogen, diffusion of boron is suppressed. In other words, the diffusion mechanism of nitrogen is the same as that of boron, and the diffusion coefficient is larger than that of boron. Therefore, when nitrogen is interdiffused with boron, nitrogen occupies the vacancies that are diffusion paths first. As a result, diffusion of boron can be suppressed, and as a result, penetration of boron into the channel region 10 can be suppressed, and fluctuations in threshold voltage can be effectively suppressed. Further, by doping nitrogen using an ion implantation method, the depth and concentration distribution of nitrogen can be easily controlled.

  Nitrogen is deposited on the gate oxide film 36 by doping nitrogen on the gate electrode 35 and performing heat treatment. As a result, the silicon oxide film / silicon interface state is reduced, the reliability of the gate oxide film 2 is improved, and the hot carrier resistance is effectively improved. Here, FIG. 9 shows the result of evaluating the reliability of the gate oxide film of the conventional MOS transistor and the MOS transistor in which nitrogen is implanted into the gate electrode by the constant current stress method. FIG. 9 is a diagram showing an improvement in the reliability of the oxide film by nitrogen implantation. From FIG. 9, it can be seen that when nitrogen is implanted into the gate electrode 35 and nitrogen is deposited on the gate oxide film 36, the dielectric breakdown resistance is improved and the reliability of the gate oxide film is improved. FIG. 10 shows the dependency of the amount of change in threshold voltage due to hot carrier injection in the PMOS transistor on the nitrogen injection amount. Figure 10 shows the change in threshold voltage after applying a constant stress voltage for 1000 seconds.The increase in the amount of nitrogen injected into the gate electrode 35 decreases the change in threshold voltage. It can be seen that doping the gate oxide film 36 with nitrogen improves the hot carrier resistance of the PMOS transistor.

The nitrogen concentration peak in the nitrogen doping region 30 inside the gate electrode 35 and the gate oxide film 36 is preferably set in a range of ˜10 ¹⁹ / cm ³ to ˜10 ²¹ / cm ³ . Therefore, the amount of nitrogen ions implanted during the manufacturing process may be set in the range of ˜10 ¹⁴ / cm ² to ˜10 ¹⁶ / cm ² . When the nitrogen concentration peak is lower than ˜10 ¹⁹ / cm ³ , the above effect cannot be obtained. When the nitrogen concentration peak in the gate oxide film 36 is higher than ˜10 ²¹ / cm ³ , the mobility of channel electrons is deteriorated. As a result, the MOS transistor characteristics deteriorate.

(Embodiment 2)
Next, another embodiment of the present invention will be described with reference to the drawings. FIG. 11 is a sectional view showing a PMOS transistor according to the second embodiment of the present invention. In FIG. 11, 1 to 7 and 10 are the same as or correspond to those in FIG. Reference numeral 30 denotes a nitrogen doping region formed inside the source / drain region 6. FIG. 12 is a diagram showing a profile in the depth direction of the source / drain region 6 of the PMOS transistor shown in FIG. From FIG. 12, it can be seen that the junction surface of the source / drain region 6 is not doped with nitrogen, and the nitrogen doped region 30 exists inside the source / drain region 6 formed by doping with boron.

Next, a manufacturing process of the PMOS transistor shown in FIG. 11 will be described. First, an element isolation oxide film 7 is formed by a normal element isolation process, and then a 100% oxide film 2a is formed on the N-type silicon substrate 1 by thermal oxidation, followed by 5 × 20 phosphorus / phosphorus by CVD. About 2000 cm of polysilicon film 3a doped with about ³ cm ^{3 is} deposited, and then about 2000 mm of oxide film 4a is deposited by CVD (FIG. 13). Next, the oxide film 4a and the polysilicon film 3a are patterned into the shape of the gate electrode by photolithography and anisotropic etching to form the oxide film 4 and the gate electrode 3 respectively (FIG. 14). Next, an oxide film of about 800 mm is deposited by CVD (not shown) and etched back to form a sidewall oxide film 5 (FIG. 15). Next, nitrogen ions are implanted under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² (FIG. 16). Next, boron ions are implanted under conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 17), and the implanted impurities are activated by applying a heat treatment at 850 ° C. for about 20 minutes. At the same time as forming the P + -type source / drain region 6 shown, a nitrogen doping region 30 is formed.

Next, another manufacturing process of the PMOS transistor shown in FIG. 11 will be described. Since FIG. 13 to FIG. 15 are the same as the manufacturing steps described above, the description thereof is omitted. Next, nitrogen ions are implanted under the conditions of 12 keV and 2.5 × 10 ¹⁵ / cm ² while rotating the N-type silicon substrate 1 at an incident angle of 30 ° (FIG. 18). Next, boron ions are implanted under the conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 19), and the implanted impurities are activated by applying a heat treatment at 850 ° C. for about 20 minutes. At the same time as forming the P + -type source / drain region 6 shown, a nitrogen doping region 30 is formed.

  Here, the nitrogen injection conditions in the manufacturing process will be described. The nitrogen injection condition is that the nitrogen is projected with energy such that the projected range of nitrogen is smaller than the projected range of boron. This is to prevent defects that occur during nitrogen implantation from occurring at the junction surface between the source / drain region 6 and the N-type silicon substrate 1 and the occurrence of junction leakage current during device operation.

  In the above description, an N-type gate electrode is used as the gate electrode. However, a P-type gate electrode or a gate electrode having a laminated structure of metal silicide and polysilicon may be used to lower the sheet resistance of the gate electrode. As shown in FIG. 20, in order to reduce the resistance of the source / drain region 6, a titanium silicide film 8 may be formed on the source / drain region 6 using a titanium salicide step after the step shown in FIG. . Further, in the above embodiment, boron ions are implanted when the source / drain region 6 is formed. However, when the titanium silicide film 8 as shown in FIG. 20 is not formed in the source / drain region 6, the source / drain region 6 is formed. Boron fluoride may be ion-implanted. In the above embodiment, only the PMOS transistor is shown. However, the PMOS transistor may be added to a part of the CMOS transistor, or the manufacturing process of the PMOS transistor may be added to a part of the manufacturing process of the CMOS transistor. Good.

  Next, the effect of the invention in this embodiment will be described. Since the P + -type source / drain region 6 is doped with nitrogen, the diffusion of boron is suppressed. In other words, the diffusion mechanism of nitrogen is the same as that of boron, and the diffusion coefficient is larger than that of boron. Therefore, when nitrogen is interdiffused with boron, nitrogen occupies the vacancies that are diffusion paths first. As a result, the diffusion of boron can be suppressed, and the junction surface of the source / drain region 6 can be formed shallow. Further, in this embodiment, in order to prevent the adverse effects caused by nitrogen implantation, nitrogen is implanted with an energy such that the projected range of nitrogen is smaller than the projected range of boron. However, the diffusion of boron is illustrated in FIG. As shown in FIG. 3, the end of the nitrogen distribution at the time of nitrogen implantation is sufficiently suppressed even if it is not deeper than the end of the boron distribution at the time of boron implantation.

  Further, when nitrogen is doped by an ion implantation method, the N-type silicon substrate 1 is made amorphous, and the channeling phenomenon during the subsequent boron ion implantation can be suppressed, so that the junction surface of the source / drain region 6 is formed shallow. It becomes possible. In addition, since amorphization by nitrogen is not as amorphous as by germanium and silicon ion implantation, high temperature heat treatment for recovering the crystal is not required, and it is effective by forming a shallow junction surface. Further, since the source / drain region 6 can be formed without implanting boron fluoride into the source / drain region 6, when the resistance of the source / drain region is reduced by using a salicide process, fluorine in the boron fluoride can be formed. Therefore, it is possible to eliminate the hindrance to the silicidation reaction and to form a good metal silicide film. Further, by doping nitrogen with an oblique rotation ion implantation method, the lateral diffusion of boron can be further suppressed, and as a result, the effective gate length of the transistor can be increased.

The nitrogen concentration peak in the nitrogen doping region 30 inside the source / drain region 6 is preferably set in a range of ˜10 ¹⁹ / cm ³ to ˜10 ²¹ / cm ³ . Therefore, the amount of nitrogen ions implanted during the manufacturing process may be set in the range of ˜10 ¹⁴ / cm ² to ˜10 ¹⁶ / cm ² . When the nitrogen concentration peak is lower than ˜10 ¹⁹ / cm ³ , the above-mentioned effect cannot be obtained. When the nitrogen concentration peak is higher than ˜10 ²¹ / cm ³ , the activation rate of boron decreases, and the source / drain region 6 Resistance increases.

(Embodiment 3)
Next, another embodiment of the present invention will be described. FIG. 22 is a sectional view showing a PMOS transistor according to the third embodiment of the present invention. In FIG. 22, 1, 2, 5-7, 10, 30, 35, and 36 are the same as or correspond to those shown in FIG. The present embodiment is an invention in which the first embodiment and the second embodiment are combined.

Next, the manufacturing process of the PMOS transistor shown in FIG. 22 will be described. First, after an element isolation oxide film 7 is formed on the N-type silicon substrate 1 by a normal element isolation process, an oxide film 36a of about 100 mm is formed by thermal oxidation, and a polysilicon film 35a of about 2000 mm is formed by CVD. Form (FIG. 23). Next, the polysilicon film 35a and the oxide film 36a are patterned into the shape of the gate electrode by photolithography and anisotropic etching to form the polysilicon film 35b and the gate oxide film 2 (not shown). Next, an oxide film of about 800 mm is deposited by CVD (not shown), and etched back to form sidewall oxide film 5 and gate oxide film 36b (FIG. 24). Next, nitrogen ions are implanted into the polysilicon film 35b and the source / drain regions under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² so that the center of the range comes above the polysilicon film 35b (FIG. 25). Next, boron ions are implanted into the polysilicon film 35b and the source / drain regions under the conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 26). Next, a source / drain region 6, a gate electrode 35 and a nitrogen doping region 30 shown in FIG. 22 are formed by activating the implanted impurities by applying a heat treatment at 850 ° C. for about 20 minutes. During this heat treatment, the nitrogen doped on the upper portion of the gate electrode 35b is thermally diffused, but the nitrogen is segregated in the gate oxide film 36b, and as shown in FIG. 2, the gate oxidation having a nitrogen concentration peak is present. A film 36 is formed. The conditions for nitrogen implantation into the source / drain region 6 and the gate electrode 35 are as described in the first embodiment and the second embodiment, respectively.

Next, another manufacturing process of the PMOS transistor shown in FIG. 22 will be described. First, after an element isolation oxide film 7 is formed on the N-type silicon substrate 1 by a normal element isolation process, an oxide film 36a of about 100 mm is formed by thermal oxidation, and a polysilicon film 35a of about 2000 mm is formed by CVD. Form (FIG. 27). Next, nitrogen ions are implanted under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² so that the center of the range comes above the polysilicon film 35a (FIG. 28). Next, boron ions are implanted into the polysilicon film 35a under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 29). Next, the polysilicon film 35a is patterned into the shape of a gate electrode by photolithography and anisotropic etching to form a gate electrode 35b (not shown). Next, an oxide film of about 800 mm is deposited by CVD (not shown), and etched back to form sidewall oxide film 5 and gate oxide film 36b (FIG. 30). Next, nitrogen is ion-implanted into the source / drain regions under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² (FIG. 31). Next, boron ions are implanted into the source / drain regions under conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 32). Finally, heat treatment is performed at 850 ° C. for about 20 minutes. Since the action in this heat treatment has been described in detail in the first manufacturing process, description thereof is omitted here.

  In this second manufacturing process, nitrogen and boron are implanted twice into the gate electrode 35. An oxide film of about 2000 mm is deposited by CVD before patterning the gate electrode 35, and then patterned. As a result, an oxide film serving as a stopper film for introducing impurities into the gate electrode 35 may be formed on the gate electrode, and the subsequent ion implantation of nitrogen and boron may be performed only in the source / drain region 6. In this manufacturing process, boron ions are implanted before patterning the gate electrode 35. However, the process may be omitted and the gate electrode 35 may be doped by boron ion implantation into the source / drain region 6.

  In the above embodiment, only the PMOS transistor is shown. However, the PMOS transistor may be added to a part of the CMOS transistor, or the manufacturing process of the PMOS transistor may be added to a part of the manufacturing process of the CMOS transistor. Further, as shown in FIG. 33, in order to reduce the resistance of the gate electrode 35 and the source / drain region 6, the titanium salicide process is used after the process shown in FIG. Alternatively, titanium silicide 8 may be formed.

  Next, the effect of the invention in this embodiment will be described. Since the gate electrode 35 is doped with nitrogen, diffusion of boron is suppressed as described in detail in the second embodiment, and boron can be prevented from penetrating the gate oxide film 36 and entering the channel region 10. The fluctuation of the threshold voltage can be effectively suppressed. Nitrogen is deposited on the gate oxide film 36 by doping nitrogen on the gate electrode 35 and performing heat treatment. As a result, the silicon oxide film / silicon interface state is reduced, the reliability of the gate oxide film 36 is improved, and the hot carrier resistance is effectively improved. Further, since the source / drain region 6 is doped with nitrogen, the diffusion of boron is suppressed as described in detail in Embodiment 1, and the junction surface of the source / drain region 6 can be formed shallow. Become. In addition, although the number of processes increases, it is possible to change and optimize each nitrogen profile by separately performing the nitrogen doping process to the gate electrode 35 and the source / drain region 6, so that the gate oxidation of boron can be performed. The penetration of the film 36 and the diffusion in the source / drain region 6 are further effectively suppressed.

(Embodiment 4)
Next, another embodiment of the present invention will be described. FIG. 34 is a sectional structural view showing an NMOS transistor according to a fourth embodiment of the present invention. In the figure, 4, 5, 7, and 10 are the same as or equivalent to those shown in FIG. Reference numeral 40 denotes a P-type silicon substrate, and hatched portions 30 denote nitrogen doping regions, which are present in the N + -type gate electrode 41 and the gate oxide film 42. Reference numeral 43 denotes an N − type source / drain region formed with the channel region 10 interposed therebetween, and reference numeral 44 denotes an N + type source / drain region formed adjacent to the N − type source / drain region. The N− type source / drain region 43 and the N + type source / drain region 44 constitute an NMOS transistor having an LDD (Lightly-Doped Drain) structure. FIG. 35 is a diagram showing profiles in the depth direction of the N + -type gate electrode 41 and the gate oxide film 42 of the NMOS transistor shown in FIG. 35 that nitrogen is deposited in the gate oxide film. Here, the precipitation of nitrogen indicates a state in which nitrogen is trapped at a certain position and the concentration becomes high.

Next, the manufacturing process of the NMOS transistor shown in FIG. 34 will be described. First, after forming the element isolation oxide film 7 on the P-type silicon substrate 40 by a normal element isolation structure, an oxide film 42a of about 100 mm is formed by thermal oxidation, and a polysilicon film 41a of about 2000 mm is formed by CVD. Form (FIG. 36). Then, on top of the polysilicon film 41a of the nitrogen ions should come projected range centered 20keV, 1 × 10 ¹⁶ / cm is implanted at ² conditions (Figure 37). Next, arsenic ions are implanted into the polysilicon film 41a under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 38). Next, the polysilicon film 41a is patterned into the shape of the gate electrode by photolithography and anisotropic etching to form the gate electrode 41b (not shown). Next, arsenic ions are implanted into the N-type source / drain region under the conditions of 50 keV and 4 × 10 ¹³ / cm ² while rotating the P-type silicon substrate 40 at an incident angle of 45 ° (FIG. 39). Next, an oxide film of about 800 mm is deposited by CVD (not shown) and etched back to form sidewall oxide film 5 and gate oxide film 42b (FIG. 40). Next, arsenic ions are implanted into the N + type source / drain regions under the conditions of 50 keV and 4 × 10 ¹⁵ / cm ² (FIG. 41).

  Finally, the implanted impurities are activated by applying a heat treatment at 850 ° C. for about 20 minutes, and the N − type source / drain region 43, N + type source / drain region 44, gate electrode 41 and nitrogen shown in FIG. 34 are activated. A doping region 30 is formed. During this heat treatment, the nitrogen doped on the upper portion of the gate electrode 41b is thermally diffused, but the nitrogen is segregated in the gate oxide film 42b, and as shown in FIG. 35, the gate oxidation having a nitrogen concentration peak is present. A film 36 is formed. Also in this embodiment, the conditions for nitrogen implantation into the gate electrode 41 are as described in the second embodiment. In other words, the projected range RP of nitrogen is in the gate electrode 41 at a position above 5 × ΔRP from the interface between the N + -type gate electrode 44 and the gate oxide film 42 when the standard deviation is ΔRP. Set to come.

In the above description, the gate electrode 41 doped N-type is formed by implanting arsenic into the polysilicon film. However, a doped polysilicon film doped with phosphorus at about 5 × 10 ²⁰ / cm ³ is used. Alternatively, an N-type doped gate electrode 41 may be formed. Further, in the present embodiment, only the NMOS transistor is shown, but the NMOS transistor may be added to a part of the CMOS transistor, or the manufacturing process of the NMOS transistor may be added to a part of the manufacturing process of the CMOS transistor. Good.

  Next, the effect of the invention in this embodiment will be described. Since nitrogen is doped on the upper portion of the gate electrode 41, nitrogen is deposited on the gate oxide film 42 by the subsequent heat treatment. As a result, the silicon oxide film / silicon interface state is reduced, the reliability of the gate oxide film 42 is improved, and the hot carrier resistance is effectively improved. The evaluation of the reliability of the gate oxide film 42 is as described in FIG. 9 in the first embodiment. Further, FIG. 42 shows the dependency of the threshold voltage on the nitrogen injection amount due to hot carrier injection in the NMOS transistor. Fig. 42 shows the change in threshold voltage after applying a constant stress voltage for 1000 seconds.When the amount of nitrogen injected into the gate electrode 41 is increased, the change in threshold voltage decreases. It can be seen that doping nitrogen and depositing nitrogen on the gate oxide film 42 improves the hot carrier resistance of the NMOS transistor.

Further, the nitrogen concentration peak in the nitrogen doping region 30 inside the gate electrode 41 and the gate oxide film 42 is desirably set in a range of ˜10 ¹⁹ / cm ³ to ˜10 ²¹ / cm ³ . Therefore, the amount of nitrogen ions implanted during the manufacturing process may be set in the range of ˜10 ¹⁴ / cm ² to ˜10 ¹⁶ / cm ² . When the nitrogen concentration peak is lower than ˜10 ¹⁹ / cm ³ , the above-described effect cannot be obtained. When the nitrogen concentration peak in the gate oxide film 42 is higher than ˜10 ²¹ / cm ³ , the mobility of channel electrons deteriorates. As a result, the MOS transistor characteristics deteriorate.

(Embodiment 5)
Next, another embodiment of the present invention will be described. FIG. 43 is a sectional view showing an NMOS transistor according to the fifth embodiment of the present invention. In the figure, 2, 3, 4, 5, 7, 10, 30, 40, 43 and 44 are the same as or equivalent to those shown in FIG. That is, in the NMOS transistor according to the present embodiment, the nitrogen doping region 30 is formed inside the N + type source / drain region 43. FIG. 44 shows a profile in the depth direction of the N + -type source / drain region 43 of the NMOS transistor. As shown in FIG. 44, the junction surface of the N + type source / drain region 43 is not doped with nitrogen, and the nitrogen doped region 30 exists inside the N + type source / drain region 43 formed by doping with arsenic. I understand.

Next, a manufacturing process of the NMOS transistor shown in FIG. 43 will be described. First, after forming an element isolation oxide film 7 on a P-type silicon substrate 40 by a normal element isolation process, an oxide film 2a of about 100 mm is formed by thermal oxidation, and phosphorus is deposited on the oxide film by a CVD method at 5 × 10 5. A polysilicon film doped to about ²⁰ / cm ^{3 is} formed to about 2000 mm, and an oxide film of about 2000 mm is formed on the polysilicon film (not shown). Next, the oxide film and the polysilicon film are patterned into the shape of the gate electrode by photolithography and anisotropic etching to form the oxide film 4 and the gate electrode 3 respectively (FIG. 45). Next, arsenic ions are implanted into the N − type source / drain region under the conditions of 50 keV and 4 × 10 ¹³ / cm ² while rotating the P-type silicon substrate 40 at an incident angle of 45 ° (FIG. 46). Next, an oxide film of about 800 mm is deposited by CVD (not shown) and etched back to form the sidewall oxide film 5 and the gate oxide film 2, and further, nitrogen ions are formed in the N + type source / drain regions. Is injected under the conditions of 10 keV and 2 × 10 ¹⁵ / cm ² (FIG. 47). Next, arsenic ions are implanted into the N + type source / drain region under the conditions of 50 keV and 4 × 10 ¹⁵ / cm ² (FIG. 48). Next, the implanted impurities are activated by applying heat treatment at 850 ° C. for about 20 minutes, and the N − type source / drain region 43, the N + type source / drain region 44 and the nitrogen doping region 30 shown in FIG. Form.

  Here, the nitrogen injection conditions in the manufacturing process are as described in the first embodiment. In other words, the nitrogen injection condition is that the nitrogen projection range is implanted with energy that is smaller than the arsenic projection range.

  In the above description, the gate electrode is formed by depositing phosphorus-doped polysilicon. However, the gate electrode may be formed by implanting N-type impurities after depositing non-doped polysilicon. Further, in order to reduce the sheet resistance of the gate electrode, a gate electrode having a laminated structure of metal silicide and polysilicon may be used. As shown in FIG. 49, in order to reduce the resistance of the source / drain region, titanium silicide 8 may be formed on N + type source / drain region 44 using a titanium salicide process after the step shown in FIG. Good. In the above embodiment, only the NMOS transistor is shown. However, the NMOS transistor may be added to a part of the CMOS transistor, or the manufacturing process of the NMOS transistor may be added to a part of the manufacturing process of the CMOS transistor. Good.

  Next, the effect of the invention in this embodiment will be described. Since nitrogen is doped in the N + type source / drain region 44, arsenic diffusion is suppressed. That is, since the relationship between boron and nitrogen in Embodiment 1 has been described in detail with respect to the relationship between arsenic and nitrogen, the diffusion of arsenic can be suppressed by interdiffusing nitrogen and arsenic. It is possible to form a shallow junction surface in the drain region.

Further, the nitrogen concentration peak in the nitrogen doping region 30 of the N + -type source / drain region 44 is preferably set in the range of ˜10 ¹⁹ / cm ³ to ˜10 ²¹ / cm ³ . Therefore, the amount of nitrogen ions implanted during the manufacturing process may be set in the range of ˜10 ¹⁴ / cm ² to ˜10 ¹⁶ / cm ² . When the nitrogen concentration peak is lower than ˜10 ¹⁹ / cm ³ , the above-mentioned effect cannot be obtained. When the nitrogen concentration peak is higher than ˜10 ²¹ / cm ³ , the activation rate of arsenic decreases, and N + type source / The resistance of the drain region 44 increases.

(Embodiment 6)
Next, another embodiment of the present invention will be described. FIG. 50 is a cross-sectional view of a dual gate CMOS transistor showing the sixth embodiment of the present invention. In FIG. 50, 10 to 14, 20, 21, and 23 are the same as or correspond to those in FIG. A hatched portion 30 indicates a nitrogen doping region, which exists in the source / drain region 21 in the PMOS transistor, in the P + type polysilicon film 50 and the gate oxide film 47, and in the NMOS transistor, the N + type polysilicon film 51 and It exists in the gate oxide film 48. 52 is an N − type source / drain region formed in the P well 14 with the channel region 10 interposed therebetween, and 53 is an N + type source / drain region formed adjacent to the N − type source / drain region. A titanium silicide film 23 is formed on the P + -type polysilicon film 50, the N + -type polysilicon film 51, the P + -type source / drain region 21 and the N + -type source / drain region 53 to form a two-layer gate. In addition to constituting the electrodes, the resistance of the source / drain regions is also reduced.

Next, a manufacturing process of the dual gate CMOS transistor shown in FIG. 50 will be described. First, after forming an N well 13 and a P well 14 on a P type silicon substrate 11, an element isolation oxide film 12 is formed on the P type silicon substrate 11 by a normal element isolation process. Next, an oxide film 49 of about 100 mm is formed by thermal oxidation, and a polysilicon film 55 of about 2000 mm is deposited by CVD (FIG. 51). Next, nitrogen ions are implanted under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² so that the center of the range comes above the polysilicon film 55 (FIG. 52). Next, the PMOS transistor formation region is covered with a resist 60, and arsenic ions are implanted into the polysilicon film 55 in the NMOS transistor formation region under conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 53). Next, after removing the resist 60, the NMOS transistor formation region is covered with a resist 61, and boron ions are implanted into the polysilicon film 55 in the PMOS transistor formation region under conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 54). ). Next, after removing the resist 61, the polysilicon film 55 is patterned into the shape of the gate electrode of the PMOS transistor and NMOS transistor by photolithography and anisotropic etching to form a polysilicon film 50a and a polysilicon film 51a (see FIG. Figure 55). Next, the PMOS transistor formation region is covered with a resist 62, and arsenic ions are applied to the N − -type source / drain region on the P-well 14 while rotating the P-type silicon substrate 11 at an incident angle of 45 ° and 50 × 4 × 10 ^13. Inject under the conditions of / cm ² (FIG. 56). Next, after removing the resist 62, an 800-nm oxide film is deposited by CVD (not shown) and etched back to form the sidewall oxide film 20, the gate oxide film 47a, and the gate oxide film 48a (see FIG. Figure 57). Next, the PMOS transistor formation region is covered with a resist 63, and arsenic ions are implanted into the N + type source / drain region on the P well 14 under the conditions of 50 keV and 4 × 10 ¹⁵ / cm ² (FIG. 58). Next, after removing the resist 63, the NMOS transistor formation region is covered with the resist 64, and nitrogen ions are implanted into the source / drain regions on the N well 13 under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² , and then boron ions Is injected under conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 59).

  Next, after removing the resist 64, the implanted impurities are activated by applying a heat treatment at 850 ° C. for about 20 minutes, so that the source / drain regions 21, the P + -type polysilicon film 50, N shown in FIG. A -type source / drain region 52, an N + type source / drain region 53, an N + type gate electrode 51 and a nitrogen doping region 30 are formed. In addition, during this heat treatment, nitrogen doped on the polysilicon film 50a and the polysilicon film 51a is thermally diffused, but in the gate oxide film 47a and the gate oxide film 48a, nitrogen is segregated and the nitrogen concentration peak is reached. The existing gate oxide film 47 and gate oxide film 48 are formed (not shown). Next, titanium of about 500 mm is deposited by sputtering, and heat treatment is performed at 700 ° C. for about 30 seconds, so that P + type polysilicon films 50 and 51, P + type source / drain regions 21 and N + type source are applied. By forming titanium silicide film 23 on / drain region 53 and removing unreacted titanium on the oxide film (not shown), the dual gate CMOS transistor shown in FIG. 50 is formed.

Next, another manufacturing process of the dual gate CMOS transistor shown in FIG. 50 will be described. First, formation is performed up to FIG. 51 by the manufacturing process described above. Next, the PMOS transistor formation region is covered with a resist 60, and nitrogen ions are implanted under the conditions of 25 keV and 1 × 10 ¹⁶ / cm ² so that the range center is located on the polysilicon film 55 (FIG. 60). Next, the resist 60 is left as it is, and arsenic ions are implanted into the polysilicon film 55 under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 61). Next, after removing the resist 60, the NMOS transistor formation region is covered with the resist 61, and nitrogen ions are implanted under the conditions of ¹⁵ keV and 4 × 10 ¹⁵ / cm ² so that the range center is located on the polysilicon film 55. (Figure 62). Next, with the resist 61 as it is, boron ions are implanted into the polysilicon film 55 under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 63). Next, after removing the resist 61, the polysilicon film 55 is patterned into the shape of the gate electrode of the PMOS transistor and the NMOS transistor by using photolithography and anisotropic etching, and the polysilicon film 50a and the polysilicon film 51a are respectively formed. Form (FIG. 64). Since the following steps are the same as the above-described manufacturing steps (FIGS. 56 to 59 correspond in the drawing), the description thereof is omitted here.

  In the above two manufacturing processes, the process of doping the polysilicon film 50a with boron and the process of doping the source / drain region with boron are performed separately. However, the polysilicon film 50a is doped with the source / drain region. This may also be performed in the doping step. Also, the step of doping the polysilicon film with arsenic may be performed as the step of doping arsenic into the N − type source / drain region or the N + type source / drain region.

  Next, the effect of this embodiment will be described. In the PMOS transistor region, since the P + type polysilicon film 50 and the P type source / drain region 21 are doped with nitrogen, the effects described in detail in the first and second embodiments are obtained. Further, since the N + type polysilicon film 51 is doped with nitrogen in the NMOS transistor region, the effect described in detail in the fourth embodiment can be obtained. In the manufacturing method in which the step of injecting nitrogen ions into the polysilicon film 50a and the step of injecting into the N + type polysilicon film 51a are performed in separate steps, the ions are injected into the polysilicon film 50a and the polysilicon film 51a. Each nitrogen profile can be optimized depending on the nature of the ions, boron penetration from the P + type polysilicon film 50 in the PMOS transistor region and the gate oxide / silicon substrate interface state in the NMOS transistor region. The generation is suppressed more effectively.

(Embodiment 7)
Next, another embodiment of the present invention will be described. FIG. 65 is a cross-sectional configuration diagram of a dual gate CMOS transistor showing a seventh embodiment of the present invention. In FIG. 65, 10-14, 20, 21, 50-53 are the same as or correspond to those in FIG. A tungsten silicide film 70 is formed on the P + type polysilicon film 50. The gate electrode of the PMOS transistor is formed by the two-layer structure of the tungsten silicide film 70 and the P + -type polysilicon film 50, and the nitrogen doping region 30 indicated by oblique lines exists in the gate electrode and the gate oxide film 47. An oxide film 19 is formed on the tungsten silicide film 70. Reference numeral 71 denotes a tungsten silicide film, which is formed on the N + type polysilicon film 51. The gate electrode of the NMOS transistor is constituted by the two-layer structure of the tungsten silicide film 71 and the N + -type polysilicon film 51, and the nitrogen doping region 30 indicated by oblique lines exists in the gate electrode and the gate oxide film 48. An oxide film 19 is formed on the tungsten silicide film 71. 66 is a diagram showing a profile in the depth direction of the gate electrode and the oxide film 47 of the PMOS transistor, and FIG. 67 is a diagram showing a profile in the depth direction of the gate electrode and the gate oxide film 48 of the NMOS transistor. . FIG. 66 shows that a nitrogen concentration peak exists at the interface between the P + -type polysilicon film 50 and the tungsten silicide film 70 in the gate electrode of the PMOS transistor, and nitrogen is deposited in the gate oxide film 47. . Further, as shown in FIG. 67, in the gate electrode of the NMOS transistor, there is a nitrogen concentration peak at the interface between the N + type polysilicon film 51 and the tungsten silicide film 71, and nitrogen is precipitated in the gate oxide film 48. I understand.

Next, a manufacturing process of the dual gate CMOS transistor shown in FIG. 65 will be described. First, after forming an N well 13 and a P well 14 on a P type silicon substrate 11, an element isolation oxide film 12 is formed on the P type silicon substrate 11 by a normal element isolation process (not shown). Next, an oxide film 49 of about 100 mm is formed by thermal oxidation, and a polysilicon film 55 of about 2000 mm is deposited by CVD (FIG. 68). Next, a tungsten silicide film 72 of about 1000 mm is deposited by sputtering (FIG. 69). Next, nitrogen ions are implanted under conditions of 40 keV and 1 × 10 ¹⁶ / cm ² so that the center of the range comes near the interface between the polysilicon film 55 and the tungsten silicide film 72 (FIG. 70). Next, the PMOS transistor formation region is covered with a resist 60, and arsenic ions are implanted into the polysilicon film 55 in the NMOS transistor formation region under conditions of 120 keV and 4 × 10 ¹⁵ / cm ² (FIG. 71). Next, after removing the resist 60, the NMOS transistor formation region is covered with a resist 61, and boron ions are implanted into the polysilicon film 55 in the PMOS transistor formation region under conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 72). . Next, after removing the resist 61, an oxide film of about 2000 mm is deposited by CVD (not shown), and the oxide film, tungsten silicide film 72 and polysilicon film 55 are formed on the gate electrode by photolithography and anisotropic etching. Patterning is performed to form an oxide film 19, a tungsten silicide film 70a, a tungsten silicide film 71a, a polysilicon film 50a, and a polysilicon film 51a (FIG. 73). Next, the PMOS transistor formation region is covered with a resist 62, and the conditions of 50 keV and 4 × 10 ¹³ / cm ² are obtained while rotating the P-type silicon substrate 11 at an incident angle of 45 ° to the N − -type source / drain region. (FIG. 74).

Next, after removing the resist 62, an 800-nm oxide film is deposited by a CVD method and etched back to form a sidewall oxide film 20, a gate oxide film 47a, and a gate oxide film 48a (not shown), The PMOS transistor formation region is covered with a resist 63, and arsenic ions are implanted into the N + type source / drain region under the conditions of 50 keV and 4 × 10 ¹³ / cm ² (FIG. 75). Next, after removing the resist 63, the NMOS transistor formation region is covered with the resist 64, nitrogen is implanted into the source / drain region of the PMOS transistor under the conditions of 10 keV and 2 × 10 ¹⁵ / cm ² , and then boron ions are implanted at 10 keV, 4 Injection is performed under the condition of × 10 ¹⁵ / cm ² (FIG. 76). Next, after removing the resist 64, the implanted impurity is activated by applying a heat treatment at 850 ° C. for about 20 minutes, and the tungsten silicide films 70 and 71, the P + type polysilicon film 50, N shown in FIG. A + type polysilicon film 51, a source / drain region 21, an N − type source / drain region 52, an N + type source / drain region 53 and a nitrogen doping region 30 are formed. During this heat treatment, nitrogen doped in the interface between the polysilicon film 50a and the tungsten silicide film 70a and the interface between the polysilicon film 51a and the tungsten silicide film 71a is thermally diffused, but the gate oxide film 47a and the gate oxide film 48a. The nitrogen is segregated therein, and a gate oxide film 47 and a gate oxide film 48 having a nitrogen concentration peak are formed as shown in FIGS. 66 and 67.

Next, the effect of the invention in this embodiment will be described. Since nitrogen is doped in the vicinity of the interface between the P + type polysilicon film 50 and the tungsten silicide film 70 and in the vicinity of the interface between the N + type polysilicon film 51 and the tungsten silicide film 71, And diffusion of arsenic into the tungsten silicide film 71 are suppressed. In other words, since the diffusion coefficient of nitrogen is larger than that of boron or arsenic, nitrogen occupies the diffusion path first, which results in diffusion of boron into the tungsten silicide film 70 and diffusion of arsenic into the tungsten silicide film 71. It suppresses and effectively suppresses the fluctuation of the threshold voltage caused by the work function change due to the mutual diffusion of boron and arsenic. In this embodiment, the nitrogen doping region 30 is formed inside the source / drain region 21, but boron fluoride ions are implanted into the source / drain region 21 under conditions of ²⁰ keV, 4 × 10 ¹⁵ / cm ² , for example. In this case, the nitrogen doping region 30 inside the source / drain region 21 may not be formed.

(Embodiment 8)
FIG. 77 is a sectional view of a PMOS-TFT showing the eighth embodiment of the present invention. In the figure, reference numerals 101 to 105 denote the same or corresponding parts as those of the conventional TFT shown in FIG. A hatched portion 110 indicates a nitrogen doped region which is a region doped with nitrogen. 78 shows the impurity profile in the depth direction in the section aa ′ in FIG. 77, and FIG. 79 shows the impurity profile in the depth direction in the section bb ′ in FIG. 78 and 79 that the nitrogen doping region 110 is formed up to the channel region outside the source end surface and the drain end surface.

Next, a manufacturing process of the TFT shown in FIG. 77 will be described. First, after forming the insulating film 102 on the semiconductor substrate 101, a non-doped polysilicon layer is deposited by about 2000 mm by the CVD method. Next, boron is ion-implanted into the non-doped polysilicon layer to form a P-type doped polysilicon layer, and then the polysilicon layer is patterned into a gate electrode shape by photolithography and anisotropic etching. Then, the gate electrode 103 is formed (not shown). Next, after forming a 100-inch gate insulating film 104 by thermal oxidation, a non-doped polycrystalline silicon layer is deposited by about 2000 mm by CVD. Next, in order to control the threshold voltage, arsenic is ion-implanted into the non-doped polycrystalline silicon layer under the conditions of 50 keV and 1 × 10 ¹² to 1 × 10 ¹³ / cm ² , and an N-type doped polysilicon layer is formed. Form. Next, the polycrystalline silicon layer is patterned using a photoengraving process and anisotropic etching so as to leave a region to be a channel region, a source region and a drain region, thereby forming a polycrystalline polysilicon layer 105 having a desired shape. (Fig. 80). Next, a resist 107 is provided in the channel region using a photoengraving process, and nitrogen is ion-implanted under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² while rotating the semiconductor substrate 101 at an incident angle of 15 ° to 60 °. (Figure 81). Next, boron fluoride is ion-implanted under the conditions of 10 keV and 4 × 10 ¹⁵ / cm ² (FIG. 82), and the implanted impurities are activated by applying heat treatment at 850 ° C. for about 20 minutes. At the same time as forming the P-type source region 105b and drain region 105c as shown, a nitrogen doping region 110 is formed.

  Here, the relationship between the nitrogen implantation conditions and the source / drain implantation conditions will be described. The nitrogen implantation energy is set so that its range RP is smaller than the range RP of boron fluoride. If the nitrogen doping region 110 is formed deeper than the source / drain junction surface, crystal defects formed at the time of nitrogen implantation are included in the depletion layer formed at the source / drain junction surface, and the junction leakage current is reduced. It is a factor to generate.

  In the above description, boron fluoride is ion-implanted into the gate electrode, but there is no problem even if boron is used. There is no problem even if an N-type gate electrode is used instead of a P-type gate electrode. Further, although boron fluoride ions are used for the P-type source / drain regions, boron ions may be used. In the above embodiment, the case of the P-channel MOS thin film transistor has been described. However, there is no problem even if the P-channel MOS thin film transistor is added to a part of the CMOS thin film transistor or the process is added to a part of the CMOS process.

  Next, the effect of the invention in this embodiment will be described. Since the source region 105b and the drain region 105c are doped with nitrogen, diffusion of boron is suppressed. In other words, the diffusion mechanism of nitrogen is the same as that of boron, and the diffusion coefficient is larger than that of boron. Therefore, when nitrogen is interdiffused with boron, nitrogen occupies the vacancies that are the diffusion path first. As a result, boron diffusion can be suppressed. Therefore, the lateral diffusion of boron into the channel region is suppressed by the action of nitrogen, the effective gate length can be increased, and punch-through due to the short channel effect can be prevented from occurring. Moreover, the lateral diffusion of boron is further suppressed by obliquely injecting nitrogen.

(Embodiment 9)
In the eighth embodiment, the case where the present invention is applied to a PMOS-TFT has been described. In the ninth embodiment, the present invention is applied to an N-channel MOS-TFT (hereinafter referred to as NMOS-TFT). Will be described. In the case of configuring the NMOS-TFT, the conductivity type of the implanted impurity in FIG. 77 may be reversed from that in the case of configuring the PMOS-TFT. That is, the gate electrode 103, the source region 105b, and the drain region 105c are doped N-type, and the channel region 105a is doped P-type. FIG. 83 shows an impurity profile in the depth direction in the aa ′ section when the TFT shown in FIG. 77 is formed of an N channel type, and FIG. 84 shows an impurity profile in the depth direction in the bb ′ section. 83 and 84, it can be seen that the nitrogen doping region 110 is formed up to the channel region outside the source end face and the drain end face.

Next, an NMOS-TFT manufacturing process according to the present invention will be described. Since this is basically the same as the manufacturing process of the PMOS-TFT described in detail in Embodiment 8, FIGS. 80 to 82 are used for the drawings. However, since the impurity implantation conditions are different, when the impurities here are different from those in the seventh embodiment, those described in parentheses in the drawing are adopted. First, after forming the insulating film 102 on the semiconductor substrate 101, a non-doped polysilicon layer is deposited by about 2000 mm by the CVD method. Next, arsenic is ion-implanted into the non-doped polysilicon layer to form an N-type doped polysilicon layer, and then the polysilicon layer is patterned into a gate electrode shape by photolithography and anisotropic etching. Then, the gate electrode 103 is formed (not shown). Next, after forming a 100-inch gate insulating film 104 by thermal oxidation, a non-doped polysilicon layer is deposited by about 2000 mm by CVD. Next, in order to control the threshold voltage, boron fluoride is ion-implanted into the non-doped polysilicon layer under the conditions of ²⁰ keV and 1 × 10 ¹² to 1 × 10 ¹³ / cm ² , and P-type doped polysilicon is formed. Form a layer. Next, the polysilicon layer is patterned using a photoengraving process and anisotropic etching so as to leave a region to be a channel region, a source region, and a drain region, thereby forming a polysilicon layer 105 having a desired shape (FIG. 80). Next, a resist 107 is provided in the channel region using a photoengraving process, and nitrogen is ion-implanted under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² while rotating the semiconductor substrate 101 at an incident angle of 15 ° to 60 °. (Figure 81). Next, arsenic is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 82), and the implanted impurities are activated by applying a heat treatment at 850 ° C. for about 20 minutes to form an N-type source region. Simultaneously with the formation of 105b and drain region 105c, a nitrogen doping region 110 is formed (FIG. 82).

  Also in the present embodiment, the relationship between the nitrogen implantation conditions and the source / drain implantation conditions is the same as in the eighth embodiment. That is, the nitrogen implantation energy is set so that its range RP is smaller than the arsenic range RP.

  In the above description, arsenic implantation is used for the gate electrode, but there is no problem even if phosphorus is used. There is no problem even if a P-type gate electrode is used instead of an N-type gate electrode. In addition, although arsenic is used for the N-type source / drain regions, phosphorus ion implantation may be used. In the above embodiment, an N-channel MOS thin film transistor has been described. However, there is no problem if the NMOS thin film transistor or a part of the CMOS process is added to a part of the CMOS thin film transistor.

  Next, the effect of the invention in this embodiment will be described. Also in the present embodiment, as in the eighth embodiment, since the N-type source / drain regions are doped with nitrogen, diffusion of arsenic or phosphorus is suppressed. In other words, the details of the relationship between boron and nitrogen in Embodiment 8 can also be said to the relationship between arsenic and nitrogen, or phosphorus and nitrogen, so that diffusion of arsenic is suppressed by interdiffusion of nitrogen and arsenic. it can. Therefore, the lateral diffusion of arsenic or phosphorus into the channel region is suppressed by the action of nitrogen, the effective gate length can be increased, and punch-through due to the short channel effect can be prevented. Further, the lateral diffusion of phosphorus or arsenic is further suppressed by obliquely injecting nitrogen.

(Embodiment 10)
FIG. 85 is a cross-sectional view of a PMOS-TFT showing the tenth embodiment of the invention. In the figure, reference numerals 101 to 103, 105, and 110 denote the same or corresponding parts as those of the TFT shown in FIG. However, in the present embodiment, the nitrogen doping region 110 indicated by the hatched portion is formed not only in the source region 105b and the drain region 105c but also in the polysilicon layer 105 and the gate insulating film 111. FIG. 86 shows an impurity profile in the depth direction in the section aa ′ of FIG. The impurity profile in the depth direction in the bb ′ cross section of FIG. 85 is the same as FIG. 86 that nitrogen is deposited in the gate insulating film 111. FIG.

Next, the manufacturing process of the PMOS-TFT in this embodiment will be described. Up to the gate electrode 103 is formed by the process described in the eighth embodiment. Next, after forming a 100Å gate insulating film 111a by thermal oxidation, a non-doped polycrystalline silicon layer 106 is deposited by about 2000 法 by CVD (not shown). Next, nitrogen is ion-implanted into the non-doped polycrystalline silicon layer 106 under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² while rotating the semiconductor substrate 101 at an incident angle of 15 ° to 60 ° (FIG. 87). Further, in order to control the threshold voltage, arsenic is ion-implanted into the polysilicon layer 106 under conditions of 50 keV and 1 × 10 ¹² to 1 × 10 ¹³ / cm ² (not shown), and the polysilicon layer 106 is photoengraved. Using the process and anisotropic etching, patterning is performed so as to leave a channel region, a source region, and a drain region, thereby forming a polysilicon layer 105 having a desired shape (FIG. 88). Next, a resist 107 is provided in the channel region using a photoengraving process, and boron fluoride is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 89), and heat treatment is performed at 850 ° C. for about 20 minutes. In addition, by activating the implanted impurities, a P-type source region 105b, drain region 105c and nitrogen doping region 110 shown in FIG. 85 are formed. Further, during this heat treatment, nitrogen implanted into the polysilicon film 105 is thermally diffused, and nitrogen is segregated in the gate insulating film 111a, so that the gate insulating film 111 having the nitrogen doping region 110 is formed.

  Here, the relationship between the nitrogen implantation conditions and the source / drain implantation conditions is the same as in the first embodiment. That is, the nitrogen implantation energy is set so that its range RP is smaller than the range RP of boron fluoride.

  In the above manufacturing process, the rotational oblique implantation method of nitrogen is used. However, vertical implantation may be performed, and nitrogen may be diffused into the channel portion of the side wall portion of the gate electrode 3 by a subsequent heat treatment.

  Next, the effect of the invention according to this embodiment will be described. Since nitrogen is segregated in the gate insulating film 111, the interface state of the polysilicon layer / silicon oxide film (gate insulating film) is reduced, and the reliability of the gate insulating film 111 is improved. That is, hot carriers generated at the drain edge due to the reduction of the interface state can be suppressed from being trapped in the gate insulating film 111, and the hot carrier resistance can be effectively improved. Further, since the source / drain regions are also doped with nitrogen, the effect described in detail in Embodiment 8, that is, the occurrence of punch-through due to diffusion of impurities constituting the source / drain regions can be prevented.

(Embodiment 11)
In the tenth embodiment, the case where the present invention is applied to a PMOS-TFT has been described. In the eleventh embodiment, the case where the present invention is applied to an NMOS-TFT will be described. In the case of configuring the NMOS-TFT, in FIG. 85, the conductivity type of the implanted impurity may be reversed from that in the case of configuring the PMOS-TFT. That is, the gate electrode 103, the source region 105b, and the drain region 105c are doped N-type, and the channel region 105a is doped P-type. FIG. 90 shows an impurity profile in the depth direction in the section aa ′ when the TFT shown in FIG. 85 is formed of an N-channel type. The impurity profile in the depth direction in the bb ′ cross section is the same as FIG. 90 that nitrogen is deposited in the gate insulating film 111. FIG.

Next, the manufacturing process of the NMOS-TFT in this embodiment will be described. Since this is basically the same as the manufacturing process of the PMOS-TFT described in detail in Embodiment 10, FIGS. 87 to 89 are used for the drawings. However, since the impurity implantation conditions are different, if the impurities here are different from those in Embodiment 10, those described in parentheses in the drawing are adopted. Up to the gate electrode 103 is formed by the process described in the eighth embodiment. Next, after forming a 100Å gate insulating film 111a by thermal oxidation, a non-doped polysilicon layer is deposited by about 2000Å by CVD. Next, nitrogen is ion-implanted into the non-doped polysilicon layer under the conditions of 10 keV and 2 × 10 ¹⁵ / cm ² while rotating the semiconductor substrate 101 at an incident angle of 15 ° to 60 ° (FIG. 87). In order to control the voltage, boron fluoride is ion-implanted into the polysilicon layer under the conditions of 30 keV and 1 × 10 ¹² to 1 × 10 ¹³ / cm ² (not shown). Next, the polysilicon layer is patterned using a photoengraving process and anisotropic etching so as to leave a region to be a channel region, a source region, and a drain region, thereby forming a polycrystalline polysilicon layer 105 having a desired shape ( (Figure 88). Next, a resist 107 is provided in the channel region using a photoengraving process, and arsenic is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 89), and heat treatment is performed at 850 ° C. for about 20 minutes. By activating the implanted impurity, an N-type source region 105b, drain region 105c, and nitrogen doping region 110 are formed. Further, the nitrogen implanted into the polysilicon film 105 during this heat treatment is thermally diffused, and the nitrogen is segregated in the gate insulating film 111a to form the gate insulating film 111 having a nitrogen doping region (FIG. 89). Through the above steps, the TFT shown in FIG. 85 is completed.

  Next, the effect of the invention according to this embodiment will be described. Since nitrogen is segregated in the gate insulating film 111a under the channel region, the interface level of the polysilicon layer / silicon oxide film (gate insulating film) is reduced, and the reliability of the gate insulating film 111 is improved. That is, hot carriers generated at the drain edge due to the reduction of the interface state can be suppressed from being trapped in the gate insulating film 111, and the hot carrier resistance can be effectively improved. Further, since the source / drain regions are also doped with nitrogen, the effect described in detail in Embodiment 8, that is, the occurrence of punch-through due to diffusion of impurities constituting the source / drain regions can be prevented.

(Embodiment 12)
FIG. 91 is a sectional view of a PMOS-TFT showing the twelfth embodiment of the present invention. In the figure, reference numerals 101, 102, 105, 110, and 111 denote the same or corresponding parts as those of the TFT shown in FIG. However, in the present embodiment, the nitrogen doping region 110 exists in the gate insulating film 111 below the gate electrode 120 and the channel region 105a. FIG. 92 shows an impurity profile in the section aa ′ in FIG. 92 that nitrogen is deposited in the channel region of the gate insulating film 111. FIG.

Next, a manufacturing method of the TFT shown in FIG. 91 will be described. First, after an insulating film 102 is formed on the semiconductor substrate 101, a polysilicon layer 120a is deposited by about 2000 mm by a CVD method. Next, nitrogen is ion-implanted into the polysilicon layer 120a under conditions of 10 keV and 2 × 10 ¹⁵ / cm ² (FIG. 93). Next, boron fluoride is ion-implanted into the polysilicon layer 120a (FIG. 94). Next, the polysilicon layer is patterned into the shape of a gate electrode by a photolithography process and anisotropic etching to form a gate electrode 120b. Then, after forming a gate insulating film 111a of 100Å by thermal oxidation, a polysilicon layer is 2000Å about by CVD, arsenic into the polysilicon layer in order to control the threshold voltage 30keV, 1 × 10 ¹² ~ Ion implantation is performed under the condition of 1 × 10 ¹³ / cm ² (not shown). Next, the polysilicon layer is patterned using a photoengraving process and anisotropic etching so as to leave a region to be a channel region, a source region, and a drain region, thereby forming a polysilicon layer 105 having a desired shape (FIG. 95). . Next, a resist 107 is provided in the channel region using a photoengraving process, and boron fluoride is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² (FIG. 96), and heat treatment is performed at 850 ° C. for about 20 minutes. In addition, by activating the implanted impurities, a P-type source region 105b and drain region 105c are formed. Further, during this heat treatment, nitrogen implanted into the gate electrode 120 is thermally diffused, and nitrogen is segregated in the gate insulating film 111a, so that the gate insulating film 111 having a nitrogen doping region is formed (FIG. 96).

  Next, the effect of the invention according to this embodiment will be described. Since the gate electrode 120 is doped with nitrogen, boron can be prevented from diffusing during heat treatment for impurity activation and penetrating through the gate insulating film 111 and entering the channel region 105a. That is, as described in detail in Embodiment 7, this is due to the effect of preventing diffusion of nitrogen. Further, nitrogen is deposited in the gate insulating film by doping nitrogen into the gate electrode 120 and then performing heat treatment. As a result, the effect as described in Embodiment Mode 10, that is, generation of interface states in the gate insulating film due to hot carrier injection is suppressed, and the reliability of the gate insulating film 120 is improved.

(Embodiment 13)
In the twelfth embodiment, the case where the present invention is applied to a PMOS-TFT has been described. In the thirteenth embodiment, the case where the present invention is applied to an NMOS-TFT will be described. When configuring an NMOS-TFT, in FIG. 91, the conductivity type of the implanted impurity may be reversed from that when configuring a PMOS-TFT. That is, the gate electrode 120, the source region 105b, and the drain region 105c are doped N-type, and the channel region 105a is doped P-type. FIG. 97 shows an impurity profile in the depth direction in the aa ′ cross section when the TFT shown in FIG. 91 is formed of an N channel type. 91 that nitrogen is deposited in the gate insulating film 111 below the channel region 105a.

  The manufacturing process of the NMOS-TFT in the present embodiment is basically the same as the manufacturing process of the PMOS-TFT described in detail in the twelfth embodiment, and the ionic species and the reverse conductivity type used in the PMOS-TFT. These ion species may be used. The ion implantation conditions for forming the NMOS-TFT are described in the above embodiment, and the description thereof is omitted here.

  Also in this embodiment, as in Embodiment 12, since the gate electrode 120 is doped with nitrogen, the arsenic doped in the gate electrode diffuses during the heat treatment for activating the impurities, and the gate insulation Injection into the film 111 can be prevented. In addition, since nitrogen is precipitated in the gate insulating film 111 during this heat treatment, it is possible to prevent the generation of interface states in the gate insulating film 111 due to hot carrier injection, that is, to improve the reliability of the gate insulating film. Can do.

(Embodiment 14)
FIG. 98 is a bird's-eye view of a dual gate CMOS-TFT showing the fourteenth embodiment of the present invention. 99 shows an AA ′ cross section of FIG. 98, that is, a cross-sectional structure of a PMOS-TFT, and FIG. 100 shows a BB ′ cross section of FIG. 98, that is, a cross-sectional structure of an NMOS-TFT. 98 to 100, 101 denotes a semiconductor substrate, and 102 denotes an insulating film. The hatched portion 110 is a nitrogen-doped region, 125 is a non-doped polysilicon layer, 126 is a tungsten silicide (WSi ₂ ) layer, and 127 is a P-type polysilicon layer. The gate electrode of the PMOS-TFT is formed by the three-layer structure of the layer 127. 128 is a gate insulating film, 129 is a polysilicon layer having a channel region 129a, a P-type source region 129b and a drain region 129c. Reference numeral 130 denotes an N-type polysilicon layer, which forms a gate electrode of an NMOS-TFT with a three-layer structure of a non-doped polysilicon layer 125, a tungsten silicide layer 126, and an N-type polysilicon layer 130. 131 is a gate insulating film, 132 is a polysilicon layer having a channel region 132a, an N-type source region 132b and a drain region 132c. The nitrogen doping region 110 exists in the tungsten silicide layer 126, the P-type polysilicon layer 127, the gate insulating film 128, the N-type polysilicon layer 130, and the gate insulating film 131. 101 shows an impurity profile in the section aa ′ in FIG. 99, and FIG. 102 shows an impurity profile in the section bb ′ in FIG. 101, the peak of the nitrogen concentration distribution is present in the interface between the P-type polysilicon layer 127 and the tungsten silicide layer 126 and in the gate insulating film 128 in the gate electrode of the PMOS-TFT. Further, as shown in FIG. 102, in the gate electrode of the NMOS-TFT, the peak of the nitrogen concentration distribution exists in the interface between the N-type polysilicon layer 130 and the tungsten silicide layer 126 and in the gate insulating film 128.

Next, a manufacturing process of the TFT shown in FIG. 98 will be described. First, after forming the insulating film 102 on the semiconductor substrate 101, a polysilicon layer 125a is deposited by about 500 mm by CVD. Next, a 500-nm tungsten silicide layer 126a is deposited on the polysilicon layer 125a by sputtering, and a polysilicon layer 135 is deposited on the tungsten silicide layer 126a by about 1000-mm (FIG. 103). Next, nitrogen is ion-implanted near the interface between the polysilicon layer 135 and the tungsten silicide layer 126a. In the present embodiment, the nitrogen ion implantation condition may be set to about 40 keV and 2 × 10 ¹⁵ / cm ² (FIG. 104). Next, the region to be the PMOS-TFT is covered with a resist, and arsenic ions are implanted into the region to be the NMOS-TFT. Conversely, the region to be the NMOS-TFT is covered with a resist, and boron fluoride is ion-implanted into the region to be the PMOS-TFT. FIG. 105 is a cross-sectional structure diagram of the TFT after implantation. Next, after patterning the polysilicon film 135, the tungsten silicide layer 126a, and the non-doped polysilicon layer 125a into the shape of the gate electrode, a 100-nm gate oxide film is formed by thermal oxidation, and a polysilicon layer is deposited by about 2000 mm by the CVD method. Next, ion implantation for threshold voltage control is performed for each of the PMOS-TFT and NMOS-TFT regions, and then the polysilicon layer is patterned to form a polysilicon layer 129 and a polysilicon layer 132 (FIG. 106). .

Next, a region resist 140 other than the source region 132b and the drain region 132c of the NMOS-TFT is provided, and arsenic is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² . A top view of the TFT at this stage is shown in FIG. Further, heat treatment is performed at 850 ° C. for about 20 minutes to activate arsenic ions, thereby forming the source region 132b and the drain region 132c of the NMOS-TFT. Next, the resist 140 is removed, a resist 141 is provided in a region other than the source region 129b and the drain region 129c of the PMOS-TFT, and boron fluoride is ion-implanted under the conditions of 30 keV and 4 × 10 ¹⁵ / cm ² . A top view of the TFT at this stage is shown in FIG. Further, heat treatment is performed at 850 ° C. for about 20 minutes to activate boron ions, thereby forming a PMOS-TFT source region 129b and an NMOS-TFT drain region 129c.

  By the way, in the heat treatment step for activating the source / drain regions, impurities contained in the gate electrode are also diffused. However, due to the effect of nitrogen doped in the vicinity of the interface between the tungsten silicide layer 126 and the polysilicon layer 127 and in the vicinity of the interface between the tungsten silicide layer 126 and the polycrystalline silicon layer 130, boron and arsenic into the tungsten silicide layer 126. The diffusion is suppressed, and the variation of the threshold voltage due to the change in the work function of the gate electrode is effectively suppressed.

(Embodiment 15)
FIG. 109 is a sectional structural view showing a stacked gate type flash EEPROM according to a fifteenth embodiment of the present invention. In FIG. 109, 201, 205, 208, 209, 212, and 215 indicate the same or corresponding parts as in FIG. 206 is a side wall oxide film provided on the side walls of the control gate electrode 205 and the floating gate electrode 221, 212a is provided in the smooth coat film 212, a contact hole for connecting to the drain region 208, and 213 is on the smooth coat film 212. And a titanium alloy film such as titanium nitride (TiN) provided on the wall surface of the contact hole 212a, 214 an aluminum alloy wiring layer provided on the titanium alloy film 213, and a hatched portion 219 is a nitrogen doping region. An oxide film 220 has a thickness of about 100 mm. Reference numeral 221 denotes a floating gate electrode made of a polysilicon film, and has a thickness of about 1000 mm. 222 is an interlayer insulating film made of a composite film of a nitride film and an oxide film, and has a thickness of about 200 mm. The nitrogen doping region 219 exists in the oxide film 220, the polysilicon film 221 and the interlayer insulating film 222. FIG. 110 is a view showing a nitrogen profile in the depth direction of the control gate electrode 205, the interlayer insulating film 222, the floating gate electrode 221 and the oxide film 220 of the flash EEPROM shown in FIG.

Next, a manufacturing process of the flash EEPROM shown in FIG. 109 will be described. First, after forming a well region and an element isolation oxide film (not shown) in a predetermined region of a P-type silicon substrate 201, an oxide film 220a of about 100 mm is formed on the entire surface, and about 1000 mm is formed on the oxide film 220a. A polysilicon film 221a is formed (FIG. 111). Next, nitrogen is ion-implanted into the polysilicon film 221a under conditions of 10 keV and ˜4 × 10 ¹⁵ / cm ² (FIG. 112). At this time, the projected range RP of nitrogen is in the polysilicon film 221a at a position above 5 × ΔRP from the interface between the polysilicon film 221a and the oxide film 220a, where the standard deviation is ΔRP. (Fig. 113). If the position is lower than this condition, the oxide film 220a may be damaged by nitrogen implantation.

Next, boron is ion-implanted into the polysilicon film 221a under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 114). Next, an interlayer insulating film 222a made of a composite film of an oxide film and a nitride film is formed on the polysilicon film 221a by about 200 mm. Thereafter, about 2500 mm of polysilicon film 205a is formed on interlayer insulating film 222a (FIG. 115). Next, a resist 225 is formed in a predetermined region on the polysilicon film 205a, and anisotropic etching is performed using the resist 225 as a mask, thereby performing polysilicon film 205a, interlayer insulating film 222a, polysilicon film 221a, and oxide film 220a. Is patterned (FIG. 116). Thereby, the control gate electrode 205, the interlayer insulating film 222b, the floating gate electrode 221b, and the oxide film 220b are formed (FIG. 117). Next, after removing the resist 225, a resist 226 is formed so as to cover a portion to be a source region of the memory cell, and arsenic (As) is applied to the main surface of the silicon substrate 201 using the resist 226 and the control gate electrode 205 as a mask. Ions are implanted under conditions of 35 keV and 5 × 10 ¹⁵ / cm ² (FIG. 118). Thereafter, the resist 226 is removed.

Next, a resist 227 is formed so as to cover a portion to be a drain region of the memory cell, and arsenic is 35 keV and 1 × 10 ¹⁶ / cm ^{2 on} the main surface of the silicon substrate 201 using the resist 227 and the control gate electrode 205 as a mask. Ions are implanted under conditions (FIG. 119). Thereafter, the resist 227 is removed. Next, an oxide film 206a is formed on the entire surface by about 2000 mm (FIG. 120), and anisotropic reactive ion etching is performed to form a sidewall oxide film 206. The width of the sidewall oxide film 206a formed in this way in the channel direction is 2000 mm, which is almost the same as the thickness of the oxide film 206a shown in FIG. Therefore, the width of the sidewall oxide film 206 in the channel direction can be easily controlled by adjusting the thickness of the oxide film 206a. After the sidewall oxide film 206 is formed, the implanted impurity is activated by performing heat treatment at 850 ° C. for about 60 seconds to form the source region 209 and the drain region 208. By this heat treatment, boron and nitrogen implanted into the floating gate electrode 221b are diffused, but nitrogen is diffused faster than boron, and only nitrogen is deposited in the oxide film 220b and the interlayer insulating film 222b, and the oxide film 220, A nitrogen doping region 219 is formed in the floating gate electrode 221 and the interlayer insulating film 222 (FIG. 121).

  Next, after the smooth coat film 212 is formed in a thickness of about 5000 to 15000 by using the CVD method, the surface of the smooth coat film 212 is flattened by performing a heat treatment at a temperature of 800 to 1000 ° C. by the reflow method. The smooth coat film 212 is formed of, for example, a PSG film, a BPSG film, a nitride film, a non-doped oxide film, or a laminated film thereof (FIG. 122). Next, a contact hole 212a having a diameter of about 0.6 μm to 1.5 μm is provided in a portion located in the drain region of the smooth coat film 212 (FIG. 123). Next, a titanium alloy film 213 made of titanium nitride for electrically connecting to the drain region is formed on the side surface of the contact hole 212a and the smooth coat film 212 (FIG. 124). Finally, an aluminum alloy wiring layer 214 having a thickness of about 10,000 mm is formed on the titanium alloy film 213 using a sputtering method, and the titanium alloy film 213 and the aluminum alloy wiring layer 214 are patterned using a photoengraving technique and a dry etching technique. As a result, a bit line composed of the titanium alloy film 213 and the aluminum alloy wiring layer 214 and electrically connected to the drain region 208 is formed. In this way, the flash EEPROM shown in FIG. 109 is completed. Note that source / drain implantation may be performed simultaneously with the resist 225 as a mask in the step shown in FIG.

Next, the effects of the invention of the flash EEPROM according to this embodiment will be described. In this embodiment, nitrogen is ion-implanted into the floating gate electrode 221 and nitrogen is deposited in the oxide film 220 and the interlayer insulating film 222 by subsequent thermal diffusion, so that there is no hydrogen doping as seen in the RTN process. . Therefore, when writing and erasing operations are performed using FN tunneling due to the nitrogen effect in the oxide film 220, traps and interface states due to hot carrier injection and holes generated due to interband tunneling are generated. Occurrence of the traps and interface states caused by this can be suppressed, and further, since the oxide film is not deteriorated due to hydrogen doping, the reliability of the oxide film 220 is improved and the probability of the initial failure of the flash EEPROM is reduced. . At the same time, the reliability of the interlayer insulating film 222 is improved due to the nitrogen effect in the interlayer insulating film 222. Further, when the reliability of the interlayer insulating film 222 is improved, the interlayer insulating film 222 can be made thinner, so that the capacitance (C _FC ) between the control gate electrode 205 and the floating gate electrode 221 can be increased. . In other words, even if the same potential is applied to the control gate electrode 205, a device with a large coupling ratio applies a larger electric field to the channel and improves the current drive capability. The applied potential can be reduced, and the power supply voltage can be reduced.

  Further, since the floating gate electrode 221 is doped with nitrogen, diffusion of boron is suppressed. In other words, the diffusion mechanism of nitrogen is the same as that of boron, and the diffusion coefficient is larger than that of boron. Therefore, when nitrogen and boron are interdiffused, nitrogen occupies the vacancies that are the diffusion path first. As a result, boron diffusion can be suppressed. Therefore, penetration of boron into the channel region 215 and implantation into the oxide film 220 can be suppressed, and fluctuations in threshold voltage can be effectively suppressed.

  Further, in the manufacturing process of the flash EEPROM according to the present embodiment, nitrogen is doped by an ion implantation method, so that unlike the RTN process, the silicon substrate 201 is not exposed to a rapid temperature change. For this reason, generation | occurrence | production of a stripe-form defect can also be suppressed.

  In addition, since it is necessary to apply heat at the time of nitrogen doping in the RTN process, there is a possibility that nitrogen diffuses over a wide range of the silicon substrate 201. However, since nitrogen is doped by an ion implantation method in this embodiment, the nitrogen is diffused. There is no need to perform a heat treatment step during implantation. Therefore, heat treatment can be performed after patterning the gate electrode, which is effective when it is not desired to diffuse nitrogen into the source region 209 and the drain region 208.

(Embodiment 16)
FIG. 125 is a sectional view showing the memory cell portion of the stacked gate type flash EEPROM according to the sixteenth embodiment of the invention. 125, reference numerals 201 to 203, 206, 208, 209, and 215 denote the same or corresponding parts as in FIG. Further, the structure of the part not described other than the memory cell part is the same as that shown in FIG. A hatched portion 219 indicates a nitrogen doping region. 222 is an interlayer insulating film made of a composite film of a nitride film and an oxide film, and has a thickness of about 200 mm. 223 is a control gate electrode made of a polysilicon film, and has a thickness of about 2500 mm. The nitrogen doping region is present in the interlayer insulating film 222 and the control gate electrode 223.

Next, a manufacturing process of the flash EEPROM shown in FIG. 125 will be described. First, after forming a well region and an element isolation oxide film in a predetermined region of a P-type silicon substrate 201 (not shown), an oxide film 202a of about 100 mm, a polysilicon film 203a of about 1000 mm, an oxide film, An interlayer insulating film 222a of about 200 mm and a polysilicon film 223a of about 2500 mm made of a nitride film composite film are formed in this order (FIG. 126). Next, nitrogen is ion-implanted into the polysilicon film 223a under conditions of 10 keV and ˜4 × 10 ¹⁵ / cm ² (FIG. 127). At this time, as described with reference to FIG. 113 of the fifteenth embodiment, the projected range RP of nitrogen is 5 × ΔRP from the interface between the polysilicon film 223a and the interlayer insulating film 222a, where the standard deviation is ΔRP. It is set so as to be in the polysilicon film 223a at a position above that position.

Next, boron is ion-implanted into the polysilicon film 223a under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 128). Thereafter, the flash EEPROM shown in FIG. 125 is completed through the manufacturing steps shown in FIG. 116 of the fifteenth embodiment. However, in the heat treatment process for impurity activation in the present embodiment, nitrogen doped in the control gate electrode 223 is deposited on the interlayer insulating film 222.

  Next, the effects of the invention of the flash EEPROM according to this embodiment will be described. Also in the present embodiment, the effect shown in the fifteenth embodiment, that is, the reliability of the interlayer insulating film 222 can be improved and the power supply voltage of the element associated therewith can be reduced. In addition, by implanting nitrogen into the control gate electrode 223, boron doped in the control gate electrode can be prevented from diffusing during heat treatment, and boron can be prevented from being implanted into the interlayer insulating film 222.

(Embodiment 17)
FIG. 129 is a sectional structural view showing a memory cell portion of the stacked gate type flash EEPROM according to the seventeenth embodiment of the invention. In FIG. 129, reference numerals 201, 206, 208, 209, 215, 219 to 223 denote the same or corresponding parts as those in the embodiment 109 or FIG. Further, the structure of the part not described other than the memory cell part is the same as that shown in FIG. The present embodiment is an invention that combines the fifteenth embodiment and the sixteenth embodiment.

Next, a manufacturing process of the flash EEPROM shown in FIG. 129 will be described. After performing the manufacturing steps shown in FIG. 115 of the fifteenth embodiment, nitrogen is ion-implanted into the polysilicon film 223a under the conditions of 10 keV and ˜4 × 10 ¹⁵ / cm ² (FIG. 130). Next, boron is ion-implanted into the polysilicon film 223a under the conditions of ²⁰ keV and 4 × 10 ¹⁵ / cm ² (FIG. 131). Thereafter, the flash EEPROM shown in FIG. 129 is completed through the manufacturing steps shown in FIG. However, in the heat treatment process for impurity activation in this embodiment, nitrogen doped in the floating gate electrode 221b is deposited in the oxide film 220b and the interlayer insulating film 222b, and at the same time, doped in the control gate electrode 223b. Nitrogen is also deposited on the interlayer insulating film 222b.

  Since the effects of the invention according to the present embodiment are as described in detail in the fifteenth and sixteenth embodiments, the description thereof is omitted here.

(Embodiment 18)
FIG. 132 is a cross-sectional structure diagram showing a memory cell portion of a buried channel type flash EEPROM according to an eighteenth embodiment of the present invention. 132, reference numerals 201 to 205, 208, 209, 215, 217, and 218 denote the same or corresponding parts as in FIG. Reference numeral 206 denotes a sidewall oxide film provided on the side walls of the control gate electrode 205 and the floating gate electrode 203. The hatched portion 219 is a nitrogen doping region and is formed inside the N-type impurity layer 217.

  Next, a manufacturing process of the buried channel type flash EEPROM shown in FIG. 132 will be described. First, a well region and an element isolation oxide film are formed in a predetermined region of a P-type silicon substrate 201 (not shown). Next, nitrogen is ion-implanted into the silicon substrate 201 in such a range that the depth from the main surface of the silicon substrate 201 becomes smaller than 500 mm (FIG. 133). Next, N-type impurities such as arsenic or phosphorus are implanted at a range such that the depth from the main surface is 500 mm or less (FIG. 134), and then P-type impurities such as boron are introduced from the main surface. The injection is carried out at a range such that the depth is 500 mm or more (Fig. 135). In other words, the nitrogen implantation condition is to perform implantation with such energy that the range of nitrogen is smaller than the range of arsenic. Next, an oxide film 202a of about 100 mm, a polysilicon film 203a of about 1000 mm, an interlayer insulating film 204a of about 200 mm made of a composite film of an oxide film and a nitride film, and a polysilicon film 205a of about 2500 mm are formed in this order on the entire surface. (Fig. 136). Since the subsequent manufacturing steps are the same as those shown in FIG. 116 and subsequent drawings of the fifteenth embodiment, description thereof is omitted here. However, in the present embodiment, the impurities implanted into the N-type impurity layer 217 and the P-type impurity layer 218 are activated by the heat treatment process in the fifteenth embodiment, and the nitrogen doping region 219 is also formed at the same time. . Note that since the N-type impurity layer 217 is formed so as to cover the nitrogen doping region 219 under the above-described impurity ion implantation conditions, defects generated during nitrogen ion implantation are caused between the N-type impurity layer 217 and the P-type impurity layer 218. Since it does not occur at the junction surface and the junction leakage current does not increase, there is no need to worry about adverse effects caused by nitrogen implantation.

  Next, the effect of the invention according to this embodiment will be described. Since the region shallower than the N-type impurity layer 217 is doped with nitrogen, diffusion of arsenic is suppressed. That is, nitrogen has a diffusion mechanism that is the same as that of arsenic and has a diffusion coefficient larger than that of arsenic, so that diffusion of arsenic during the heat treatment step can be suppressed. Further, boron diffusion of the P-type impurity layer 218 can be suppressed by the same mechanism. Therefore, since the thin N-type impurity layer 217 can be formed, punch-through can be suppressed in the buried channel type flash EEPROM. Further, the thickness of the N-type impurity layer 217 can be controlled to a desired value depending on the nitrogen implantation conditions.

(Embodiment 19)
FIG. 137 is a sectional view showing a stack gate type flash EEPROM according to the nineteenth embodiment of the present invention. In FIG. 137, reference numerals 201 to 205, 208, 209, and 215 denote the same or corresponding parts as in FIG. Reference numeral 206 denotes a sidewall oxide film provided on the side walls of the control gate electrode 205 and the floating gate electrode 203. The hatched portion 230 is a nitrogen doping region formed inside the drain region 208. FIG. 138 shows a profile in the depth direction of drain region 208 of the flash EEPROM shown in FIG. FIG. 138 shows that the junction surface of the drain region 208 is not doped with nitrogen, and the nitrogen doped region 230 exists inside the drain region 208 formed by doping with arsenic.

Next, a manufacturing process of the stack gate type flash EEPROM shown in FIG. 137 will be described. First, after forming a well region and an element isolation oxide film in a predetermined region of a P-type silicon substrate 201 (not shown), an oxide film 202a of about 100 mm, a polysilicon film 203a of about 1000 mm, an oxide film, An interlayer insulating film 204a of about 200 mm, a polysilicon film 205a of about 2500 mm, and an oxide film 207a of about 1000 mm made of a composite film of nitride films are formed in this order (FIG. 139). Next, the oxide film 202a, the polysilicon film 203a, the interlayer insulating film 204a, the polysilicon film 205a, and the oxide film 207a are patterned into the shape of the gate electrode to form the oxide film 202, the floating gate electrode 203, the interlayer insulating film 204, and the control gate. An electrode 205 and an oxide film 207 are formed. Next, the source formation region is covered with a resist 225, and nitrogen is ion-implanted into the drain formation region under conditions of 10 keV and ˜8 × 10 ¹⁵ / cm ^{2 using} the resist 225 and the oxide film 207 as a mask (FIG. 140). Next, arsenic is ion-implanted under conditions of 35 keV and 5 × 10 ¹⁵ / cm ² (FIG. 141). In other words, the nitrogen implantation condition is to perform implantation with such energy that the range of nitrogen is smaller than the range of arsenic. Thereafter, the resist 225 is removed. Since the subsequent manufacturing steps are the same as those shown in FIG. 119 and thereafter in the fifteenth embodiment, description thereof is omitted here. However, in the present embodiment, the impurity implanted into the source region 209 and the drain region 208 is activated by the heat treatment step in the fifteenth embodiment, and the nitrogen doping region 230 is also formed at the same time. Note that the drain region 208 is formed so as to cover the nitrogen doping region 230 under the impurity ion implantation conditions, so that defects generated during nitrogen ion implantation do not occur at the junction surface between the drain region 208 and the silicon substrate 201. Since the junction leakage current does not increase, there is no need to worry about the harmful effects caused by nitrogen implantation.

  Next, effects of the invention according to the present embodiment will be described. Since the drain region 208 is doped with nitrogen, arsenic implanted into the drain region 208 during the heat treatment process can be prevented from diffusing. Therefore, it is possible to form a shallow PN junction surface with the silicon substrate 201 in the drain region 208, and to suppress short channel effects such as punch-through. Further, since the short channel effect can be suppressed, the element can be miniaturized.

In addition, since the diffusion of arsenic implanted into the drain region 208 is suppressed by nitrogen doped in the drain region, the overlap region between the oxide film 202 and the drain region 208 due to the lateral diffusion of arsenic is reduced, The capacitance (C _FS ) between the control gate electrode 205 and the drain region 208 becomes small. Therefore, the coupling ratio (C _FC / C _TOTAL ) can be increased, and the potential difference between the potential (V _CG ) of the control gate electrode 205 and the potential (V _FG ) of the floating gate electrode 203 is reduced. That is, even when the same potential is applied to the control gate electrode 205, a device having a larger coupling ratio applies a larger electric field to the channel region 215, and current drive capability is improved. Therefore, in order to obtain the same effect, the larger the coupling ratio, the smaller the voltage (V _CG ) applied to the control gate electrode 205, and the power supply voltage can be reduced.

(Embodiment 20)
FIG. 142 is a sectional view showing a stack gate type flash EEPROM according to the twentieth embodiment of the present invention. 142, reference numerals 201 to 205, 208, 209, and 215 denote the same or corresponding parts as in FIG. Reference numeral 206 denotes a sidewall oxide film provided on the side walls of the control gate electrode 205 and the floating gate electrode 203. Reference numeral 231 denotes a nitrogen doping region formed inside the source region 209.

  In the manufacturing method of the stacked gate type flash EEPROM shown in this embodiment mode, the nitrogen doping step shown in Embodiment Mode 19 may be performed before the source implantation step. Also in this embodiment, as in the case of the nineteenth embodiment, the nitrogen injection condition is to implant with energy such that the range of nitrogen is smaller than the range of arsenic.

  Also in this embodiment, the same effect as the effect of the invention shown in the nineteenth embodiment can be obtained.

(Embodiment 21)
FIG. 143 is a sectional view of a stack gate type flash EEPROM shown in the twenty-first embodiment of the invention. 35, reference numerals 201 to 205, 206, 208, 209, 230, and 231 denote the same or corresponding parts as those in the embodiment shown in FIG. 137 or FIG. 142. The present embodiment is an invention that combines the nineteenth embodiment and the twentieth embodiment. That is, a nitrogen doping region 230 is formed inside the drain region 208, and a nitrogen doping region 231 is formed inside the source region 209.

  In the manufacturing process of the stacked gate type flash EEPROM shown in this embodiment mode, the nitrogen doping process shown in Embodiment Mode 19 may be performed after the step of patterning the gate electrode (FIG. 144).

  In the present embodiment, since the nitrogen doping region 230 and the nitrogen doping region 231 are provided inside the drain region 208 and the source region 209, respectively, the effects shown in the nineteenth embodiment or the twentieth embodiment are more remarkable. appear. Further, in this embodiment mode, nitrogen may be implanted without providing the oxide film 207, and the control gate electrode 205 may be doped with nitrogen.

  Note that a region doped with nitrogen may be selected in combination with the above embodiment according to a desired effect.

  6 source / drain region, 8 titanium silicide film, 23 titanium silicide film, 30 nitrogen doping region, 35 gate electrode, 36 gate oxide film, 41 gate electrode, 42 gate oxide film, 44 N + type source / drain region, 43 N -Type source / drain region, 47 gate oxide film, 48 gate oxide film, 50 polysilicon film, 51 polysilicon film, 52 N-type source / drain region, 53 N + type source / drain region, 105 polysilicon layer, 105a channel region, 105b source region, 105c drain region, 110 nitrogen doping region, 111 gate insulating film, 120 gate electrode, 126 tungsten silicide layer, 128 gate insulating film, 131 gate insulating film, 208 drain region, 209 source region, 217 N-type impurity layer, 218 P-type impurity layer, 219 Nitrogen doping region, 220 Oxide film, 222 Interlayer insulation film, 223 Controller Rugate electrode, 230 nitrogen doped region, 231 nitrogen doped region.

Claims (7)

Forming an oxide film on a semiconductor substrate;
Forming a polysilicon film on the oxide film;
Implanting first nitrogen ions into the polysilicon film;
After implanting the first nitrogen ions, patterning the polysilicon film to form a gate electrode, patterning the oxide film to form a gate oxide film;
Forming an insulating film on a side surface of the gate electrode;
After forming the insulating film, implanting second nitrogen ions into the semiconductor substrate across the gate electrode to form a nitrogen doping region in the semiconductor substrate;
After forming the insulating film, impurity ions are implanted into the semiconductor substrate with a projection range larger than the projection range of the second nitrogen ions, and the source / drain regions including the nitrogen doping regions are formed. Forming, and
A method of manufacturing a semiconductor device, comprising: performing a heat treatment after the step of injecting impurities into the semiconductor substrate to deposit nitrogen in the gate oxide film.   The method of manufacturing a semiconductor device according to claim 1, wherein the source / drain regions are P-type.   The method of manufacturing a semiconductor device according to claim 1, wherein the impurity ions are boron ions.   The method of manufacturing a semiconductor device according to claim 1, wherein the impurity ions are boron fluoride ions.   The method of manufacturing a semiconductor device according to claim 1, wherein the source / drain regions are N-type.   The method for manufacturing a semiconductor device according to claim 1, wherein the impurity ions are arsenic ions. The method for manufacturing a semiconductor device according to claim 1, wherein a nitrogen concentration peak in the nitrogen doping region is in a range of 10 ¹⁹ / cm ³ to 10 ²¹ / cm ³ .

Images