Method for manufacturing a semiconductor device

Abstract

A method for manufacturing a semiconductor device comprises: forming a deposition type interface layer on a substrate; converting the deposition type interface layer into an oxidation type interface layer; forming a high-k layer on the oxidation type interface layer; forming a dipole interface at the interface between the high-k layer and the oxidation type interface layer; forming a conductive layer on the high-k layer; and patterning the conductive layer, the high-k layer, the dipole interface and the oxidation type interface layer to form a gate stack on the substrate.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority from Korean Patent Application No. 10-2019-0138050, filed on October 31, 2019, which is hereby incorporated by reference in its entirety. Technical Field

[0003] Various embodiments of the present invention relate to a method for manufacturing a semiconductor device, and more particularly, to a method for manufacturing a semiconductor device provided with a high-k dielectric material and a metal electrode. Background Art

[0004] As the integration density of semiconductor devices increases, the amount of leakage current through the gate dielectric layer of the transistor increases. To cope with the increase in the amount of leakage current, the gate dielectric layer may be formed of a high-k dielectric material. Summary of the invention

[0005] An embodiment of the present invention relates to a method for manufacturing a semiconductor device capable of improving reliability of a gate dielectric layer.

[0006] According to one embodiment of the present invention, a method for manufacturing a semiconductor device includes: forming a deposition type interface layer on a substrate; converting the deposition type interface layer into an oxidation type interface layer; forming a high-k layer on the oxidation type interface layer; forming a dipole interface at the interface between the high-k layer and the oxidation type interface layer; forming a conductive layer on the high-k layer; and patterning the conductive layer, the high-k layer, the dipole interface and the oxidation type interface layer to form a gate stack on the substrate.

[0007] According to another embodiment of the present invention, a method for manufacturing a semiconductor device includes: preparing a substrate including a first region and a second region; forming a channel layer on the substrate in the second region; forming a deposition type interface layer on the channel layer and the substrate in the first region; converting the deposition type interface layer into an oxidation type interface layer; forming a high-k layer on the oxidation type interface layer; forming a dipole interface at the interface between the high-k layer and the oxidation type interface layer in the first region; forming a conductive layer on the high-k layer; patterning the conductive layer, the high-k layer, the dipole interface and the oxidation type interface layer to form a first gate stack on the substrate in the first region; and patterning the conductive layer, the high-k layer and the oxidation type interface layer to form a second gate stack on the substrate in the second region.

[0008] According to another embodiment of the present invention, a method for manufacturing a semiconductor device includes: preparing a substrate including a first NMOSFET region and a second NMOSFET region; forming a thick deposition type interface layer on the substrate of the first NMOSFET region; converting the thick deposition type interface layer into a thick oxidation type interface layer; forming a thin oxidation type interface layer on the substrate of the second NMOSFET region; forming a high-k layer on the thick oxidation type interface layer and the thin oxidation type interface layer; forming a dipole interface at the interface between the oxidation type interface layer and the high-k layer; forming a gate conductive layer on the high-k layer formed with the dipole interface; and performing gate patterning processing on the substrates of the first NMOSFET region and the second NMOSFET region, respectively, to form a gate stack.

[0009] According to another embodiment of the present invention, a method for manufacturing a semiconductor device includes: forming a silicon germanium channel layer on a substrate; depositing a silicon oxide layer on the silicon germanium channel layer; performing free radical oxidation on the silicon oxide layer to form a high-quality silicon oxide layer; forming a high-k layer on the high-quality silicon oxide layer; forming a conductive layer on the high-k layer; and patterning the conductive layer, the high-k layer, and the high-quality silicon oxide layer to form a gate stack on the substrate.

[0010] According to another embodiment of the present invention, a method for manufacturing a semiconductor device includes: depositing a silicon oxide layer on a substrate; performing free radical oxidation on the silicon oxide layer to form a high-quality silicon oxide layer; forming a high-k layer on the high-quality silicon oxide layer; forming a lanthanum oxide layer on the high-k layer; forming a lanthanum diffused interface between the high-k layer and the high-quality silicon oxide layer. The lanthanum oxide layer is removed from the surface of the high-k layer; a conductive layer is formed on the surface of the high-k layer; and the conductive layer, the high-k layer and the high-quality silicon oxide layer are patterned to form a gate stack on the substrate, wherein the high-k layer is in direct contact with the conductive layer without a lanthanum diffused interface.

[0011] According to another embodiment of the present invention, a semiconductor device includes an N-type transistor and a P-type transistor, wherein the N-type transistor includes: a first high-quality silicon oxide layer on a semiconductor substrate; a first high-k layer on the first high-quality silicon oxide layer; a first gate electrode on the first high-k layer; a lanthanum-diffused dipole interface between the first high-k layer and the first high-quality silicon oxide layer, and the P-type transistor includes: a silicon germanium layer on a semiconductor substrate; a second high-quality silicon oxide layer on the silicon germanium layer; a second high-k layer on the second high-quality silicon oxide layer; and a second gate electrode on the second high-k layer, wherein the second high-k layer is in direct contact with the second gate electrode without a dipole interface.

[0012] The above and other features and advantages of the present invention will become apparent to those skilled in the art from the following specific embodiments in conjunction with the following drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a cross-sectional view showing a semiconductor device according to an embodiment of the present invention.

[0014] FIG. 2A to FIG. 2I is a cross-sectional view illustrating a method of manufacturing a semiconductor device according to an embodiment of the present invention.

[0015] Figure 3 is a cross-sectional view showing a semiconductor device according to another embodiment of the present invention.

[0016] Figures 4A to 4K is a cross-sectional view illustrating a method of manufacturing a semiconductor device according to another embodiment of the present invention. DETAILED DESCRIPTION

[0017] Various embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. However, the present invention can be implemented in different forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to make this disclosure full and complete and to fully inform those skilled in the art of the scope of the present invention. Throughout the disclosure, the same reference numerals refer to the same parts of the various figures and embodiments of the present invention.

[0018] Hereinafter, various embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0019] The drawings are not necessarily drawn to scale, and in some cases, the proportions may have been exaggerated in order to clearly illustrate features of the embodiments. When a first layer is referred to as being "on" a second layer or "on" a substrate, it refers not only to the case where the first layer is directly formed on the second layer or substrate, but also to the case where a third layer exists between the first layer and the second layer or substrate.

[0020] The CMOSFET using the gate-first process may form a gate stack including a high-k layer and a metal electrode. To modulate the threshold voltage of the CMOSFET, the NMOSFET may form a lanthanum oxide capping layer, and the PMOSFET may form a silicon germanium (SiGe) channel layer.

[0021] The germanium (Ge) concentration of the SiGe channel layer may be about 20 at% to 50 at%, and the thickness of the thick gate dielectric layer and the thin gate dielectric layer may be increased due to over-oxidation, or germanium oxide (GeOx) may be formed during the formation of the thick gate dielectric layer and the thin gate dielectric layer. As a result, the reliability of the gate dielectric layer may be deteriorated.

[0022] To improve reliability, a silicon (Si) capping layer can be deposited in situ after depositing the SiGe channel layer, but the deposition of the Si capping layer may sharply reduce the yield. Therefore, in recent years, the thickness increase caused by over-oxidation and the formation of germanium oxide can be suppressed by omitting the Si capping layer and depositing an over-deposited oxide as a thick gate dielectric layer. The deposited oxide may refer to an oxide deposited by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

[0023] However, the reliability of deposited oxide used as a thick gate dielectric layer may be reduced because the amount of impurities and defects in the oxide is too high to cope with the radical oxide used as a gate dielectric layer in a stack of silicon oxynitride (SiON) and polysilicon. In addition, the amount of leakage current caused by direct tunneling of silicon oxynitride (SiON) may increase when a stack of silicon oxynitride (SiON) and polysilicon is applied.

[0024] In addition, the lanthanum oxide capping layer may continue to diffuse into the gate dielectric layer through the post-heat treatment, thereby forming traps, which may deteriorate the reliability of the gate dielectric layer.

[0025] According to the following contents of the present invention, in order to improve the reliability of thick oxide, impurities and defects inside the deposited oxide may be removed by performing a post-treatment (eg, radical oxidation) after the deposition process of the deposited oxide.

[0026] The threshold voltage Vt of the NMOSFET can be modulated by forming a lanthanum oxide capping layer on the high-k layer, then diffusing a dipole-inducing species into the interface between the high-k layer and the interface layer to form a dipole interface, and then removing the lanthanum oxide capping layer. Thus, the formation of traps can be prevented, and the reliability of the gate dielectric layer can be improved.

[0027] Figure 1 is a cross-sectional view showing a semiconductor device 100 according to an embodiment of the present invention.

[0028] refer to Figure 1 , the semiconductor device 100 may include a first transistor T1 and a second transistor T2, which are formed in the first region R1 and the second region R2, respectively. The first and second transistors T1 and T2 may be isolated by an isolation layer 102. In one embodiment, the first transistor T1 may include an n-type transistor, and the second transistor T2 may include a p-type transistor. For example, the first transistor T1 may be an NMOSFET, and the second transistor T2 may be a PMOSFET.

[0029] The first transistor T1 may include a first gate stack 120N formed on a substrate 101, and a first source region 121N and a first drain region 122N formed in the substrate 101 and located on both sides of the first gate stack 120N. The first gate stack 120N may include a first interface layer 104N, a first high-k layer 105N, a first metal electrode 107N, a first low-resistivity electrode 108N, and a first hard mask layer 109N stacked in sequence. The first interface layer 104N may be formed on the highest surface of the substrate 101. The first interface layer 104N may be in direct contact with the highest surface of the substrate 101. The first gate stack 120N may also include a dipole interface 106 between the first interface layer 104N and the first high-k layer 105N. The dipole interface 106 may include an induced dipole substance for modulating the threshold voltage of the first transistor T1. As will be described later, a dipole interface 106 can be formed at the interface between the first interface layer 104N and the first high-k layer 105N by diffusing an induced dipole substance from a dielectric cover layer including the induced dipole substance. The induced dipole substance may include a lanthanide rare earth metal. The threshold voltage of the first transistor T1 may be modulated by the dipole interface 106. The dipole interface 106 may be made of or include a lanthanide rare earth metal. In one embodiment, the dipole interface 106 may be made of or include lanthanum (La). The dipole interface 106 may include a dipole interface diffused by lanthanum.

[0030] The second transistor T2 may include a second gate stack 120P formed on the substrate 101, and a second source region 121P and a second drain region 122P formed in the substrate 101 on both sides of the second gate stack 120P. The second gate stack 120P may include a second interface layer 104P, a second high-k layer 105P, a second metal electrode 107P, a second low-resistivity electrode 108P, and a second hard mask layer 109P stacked in sequence. The second high-k layer 105P may be in direct contact with the second metal electrode 107P without a dipole interface. The second high-k layer 105P may be in direct contact with the second interface layer 104P without a dipole interface. The second transistor T2 may also include a P-channel layer 103, which may be formed on the substrate 101 and below the second gate stack 120P. The P-channel layer 103 may be formed on the highest surface of the substrate 101. The P-channel layer 103 may be in direct contact with the highest surface of the substrate 101.

[0031] The P-channel layer 103 may be crystalline and may contain a large amount of germanium. The P-channel layer 103 may be made of or include silicon germanium. The work function of the second metal electrode 107P may be suitable for the second transistor T2.

[0032] refer to Figure 1The first gate stack 120N may include a first interface layer 104N, and the second gate stack 120P may include a second interface layer 104P.

[0033] The first interface layer 104N and the second interface layer 104P may be made of the same material. For example, the first interface layer 104N and the second interface layer 104P may be made of silicon oxide. The first interface layer 104N and the second interface layer 104P may be made of silicon oxide provided by a combined deposition and post-oxidation process. For example, each of the first interface layer 104N and the second interface layer 104P may be an oxidized silicon oxide. The oxidized silicon oxide may be made of silicon oxide converted from deposited silicon oxide. For example, the oxidized silicon oxide may be a material converted from deposited silicon oxide by a free radical oxidation process. The deposited silicon oxide may be deposited by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

[0034] Since the deposited silicon oxide is converted into oxidized silicon oxide, the first interface layer 104N and the second interface layer 104P can be high-quality silicon oxide (SiO2), which is free of impurities and defects or substantially free of impurities and defects. Therefore, high-quality silicon oxide (HQ SiO2) means that the silicon oxide may not contain impurities and defects. In addition, although the deposited silicon oxide may contain a large amount of impurities and defects in the layer, as the deposited silicon oxide is converted into oxidized silicon oxide through a free radical oxidation process, the impurities and defects can be removed, thereby producing high-quality silicon oxide that is free of impurities and defects or substantially free of impurities and defects.

[0035] The first high-k layer 105N and the second high-k layer 105P may include a high-k material having a high-k constant. The dielectric constant of the material suitable for the first high-k layer 105N and the second high-k layer 105P may be greater than the dielectric constant of silicon oxide (SiO2), which is approximately 3.9. In addition, the physical thickness of the first high-k layer 105N and the second high-k layer 105P may be greater than that of silicon oxide (SiO2) and may have a lower equivalent oxide thickness (EOT) value. The dielectric constant of the first high-k layer 105N and the second high-k layer 105P may be greater than the dielectric constant of the first interface layer 104N and the second interface layer 104P.

[0036] The first high-k layer 105N and the second high-k layer 105P may include metal-containing materials, such as metal oxides, metal silicates, and metal silicate nitrides. The metal oxide may include oxides containing metals such as hafnium (Hf), aluminum (Al), lanthanum (La), zirconium (Zr), etc. The metal oxide may be made of hafnium oxide, aluminum oxide, lanthanum oxide, zirconium oxide, or a combination thereof or include the oxide. For example, the metal oxide may be made of HfO2, Al2O3, La2O3, ZrO2, or a combination thereof or include the oxide or a combination thereof. The metal silicate may include a silicate containing metals such as hafnium (Hf) and zirconium (Zr). For example, the metal silicate may be made of hafnium silicate (HfSiO), zirconium silicate (ZrSiO), or a combination thereof or include the metal silicate or a combination thereof. The metal silicate nitride may be made of hafnium silicate nitride (HfSiON), zirconium silicate nitride (ZrSiON), or a combination thereof or include the metal silicate nitride or a combination thereof.

[0037] The first metal electrode 107N and the second metal electrode 107P may be made of the same material. For example, the first metal electrode 107N and the second metal electrode 107P may be made of or include a metal nitride, such as titanium nitride.

[0038] The first low-resistivity electrode 108N and the second low-resistivity electrode 108P may be made of the same material. For example, the first low-resistivity electrode 108N and the second low-resistivity electrode 108P may be made of or include a tungsten-based material.

[0039] According to another embodiment of the present invention, the first metal electrode 107N may have a low work function, and the second metal electrode 107P may have a high work function. In this article, the low work function may refer to a work function lower than the mid-gap work function of silicon, and the high work function may refer to a work function higher than the mid-gap work function of silicon.

[0040] The first metal electrode 107N may include titanium-rich titanium nitride. The second metal electrode 107P may include nitrogen-rich titanium nitride.

[0041] Nitrogen-rich titanium nitride may refer to titanium nitride containing more nitrogen than the stoichiometric composition of titanium and nitrogen. Nitrogen-rich titanium nitride (N-rich TiN) is titanium nitride containing excess nitrogen. Depending on the composition ratio of titanium and nitrogen, the effective work function of titanium nitride (TiN) may be different. For example, nitrogen-rich titanium nitride (N-rich TiN) may have a P-type effective work function suitable for PMOSFET. Nitrogen-rich titanium nitride (N-rich TiN) can be formed by a physical vapor deposition (PVD) process or an atomic layer deposition (ALD) process.

[0042] Titanium-rich titanium nitride (Ti-rich TiN) may refer to titanium nitride having more titanium than the stoichiometric composition of titanium and nitrogen. Titanium-rich titanium nitride (Ti-rich TiN) may refer to titanium nitride containing excess titanium. Titanium-rich titanium nitride (Ti-rich TiN) may have an N-type effective work function suitable for NMOSFET. Titanium-rich titanium nitride (Ti-rich TiN) may be formed by a physical vapor deposition (PVD) process or an atomic layer deposition (ALD) process.

[0043] The first low-resistivity electrode 108N and the second low-resistivity electrode 108P may be made of or include a low-resistivity metal, such as tungsten. The first low-resistivity electrode 108N and the second low-resistivity electrode 108P may reduce the resistance of the first gate stack 120N and the second gate stack 120P.

[0044] According to the above embodiment of the present invention, the first gate stack 120N may include a dipole interface 106 formed at the interface between the first metal electrode 107N and the first high-k layer 105N. Therefore, the threshold voltage of the first transistor T1 may be modulated. In other words, the threshold voltage of the first transistor T1 may be modulated by the dipole generated in the dipole interface 106.

[0045] Since the P-channel layer 103 is formed below the second gate stack 120P, the threshold voltage of the second transistor T2 can be modulated. In other words, since the P-channel layer 103 includes a large amount of germanium, the energy band gap can be reduced, thereby modulating the threshold voltage to a level suitable for the P-channel transistor.

[0046] Therefore, this embodiment can independently modulate the threshold voltages of NMOSFET and PMOSFET during the integration process of CMOSFET.

[0047] Figure 1 The first gate stack 120N in FIG. 1 is a gate stack of an NMOSFET, and the first gate stack 120N may be a 'La2O3 capping layer-free' gate stack in which the lanthanum oxide capping layer is omitted.

[0048] Figures 2A to 2I is a cross-sectional view illustrating a method of manufacturing a semiconductor device according to an embodiment of the present invention.

[0049] refer to Figure 2A , a substrate 11 is prepared. The substrate 11 may include a plurality of regions in which transistors are formed. These regions may include a first region R1 and a second region R2. The substrate 11 may be made of or include a semiconductor material. The substrate 11 may include a semiconductor substrate. The substrate 11 may include a silicon substrate, a silicon germanium substrate, or an SOI substrate.

[0050] An isolation layer 12 may be formed in the substrate 11. The isolation layer 12 may be formed by a shallow trench isolation (STI) process.

[0051] The first region R1 and the second region R2 may be isolated by an isolation layer 12. The first region R1 may be a region where an NMOSFET is to be formed. The second region R2 may be a region where a PMOSFET is to be formed. Although not shown, a well may be formed in the substrate 11 by a typical well formation process. A P-type well may be formed in the substrate 11 of the first region R1, and an N-type well may be formed in the substrate 11 of the second region R2.

[0052] Subsequently, a P-channel layer 13 may be formed on the substrate 11 of the second region R2. The P-channel layer 13 may be formed on the substrate 11 of the second region R2. The P-channel layer 13 may be in direct contact with the entire top surface of the substrate 11 of the second region R2. The P-channel layer 13 may be made of a crystalline material. The P-channel layer 13 may be made of or include silicon germanium. The germanium concentration in the silicon germanium layer of the P-channel layer 13 may be about 20 at % to about 50 at %. The P-channel layer 13 may be formed only of silicon germanium. The P-channel layer 13 may be formed of crystalline silicon germanium. In one embodiment, the P-channel layer 13 may be formed of a silicon germanium layer with a silicon capping layer on top. The P-channel layer 13 may be formed by a selective epitaxial growth (SEG) process. The thickness of the P-channel layer 13 may be about 1000 Å. or smaller.

[0053] refer to Figure 2B , a deposition type interface layer 14A may be formed over the substrate 11. The deposition type interface layer 14A may be formed on the surface of the substrate 11 in the first region R1 and over the P channel layer 13 in the second region R2. The deposition type interface layer 14A may be made of or include an oxide or an oxide-based material. For example, the deposition type interface layer 14A may be made of or include silicon oxide. The deposition type interface layer 14A may be formed by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process. The deposition type interface layer 14A may be formed to be approximately The deposition type interface layer 14A may be formed to a thickness of about to Thickness.

[0054] As shown above, the deposition type interface layer 14A may be formed by a deposition process rather than a thermal oxidation process. In other words, the deposition type interface layer 14A may include a deposition type oxide. The oxide formed by the thermal oxidation process may be referred to as an oxidized oxide.

[0055] The deposited oxide can be formed to a relatively thicker thickness than the thermal oxide, but a large amount of impurities may exist in the layer. In addition, the deposited oxide can suppress excessive oxidation of the P channel layer 13.

[0056] The deposition type interface layer 14A may be made of or include silicon oxide. The silicon oxide may be a deposition type silicon oxide. For example, the silicon oxide may be formed using an atomic layer deposition process (ALD) or a chemical vapor deposition (CVD) process. In one embodiment, the silicon oxide may be formed using an atomic layer deposition process (ALD). The deposition type silicon oxide may be deposited using a silicon source and an oxidation source. The silicon source may include silane, disilane, dichlorosilane, diisopropylaminosilane (DIPAS), etc., and the oxidation source may include O2, O3, etc. Atomic layer deposition of silicon oxide may be performed several times with silicon source injection, purge, oxidation source injection, and purge as unit cycles.

[0057] refer to Figure 2C , post-processing 15 may be performed. Post-processing 15 may include a free radical oxidation process. The free radical oxidation process may be referred to as post-oxidation. The free radical oxidation process may be performed using oxygen free radicals (O*). The free radical oxidation process may improve the reliability of the deposition type interface layer 14A. For example, impurities may be removed from the deposition type interface layer 14A by a free radical oxidation process. The impurities may be removed by oxygen free radicals. Therefore, the deposition type interface layer 14A may be converted into an oxidized interface layer 14 shown by reference numeral 14.

[0058] Since the oxidized interface layer 14 does not contain impurities, it has excellent film quality with low defects. When the deposited interface layer 14A is silicon oxide, high-quality silicon oxide (HQ SiO2) can be formed by the radical oxidation process 15. The oxidized interface layer 14 can be high-quality silicon oxide.

[0059] According to another embodiment of the present invention, the post-processing 15 for forming the oxidized interface layer 14 may include dry oxidation. In addition, according to another embodiment of the present invention, after forming the oxidized interface layer 14, processes such as plasma nitridation and annealing may be performed.

[0060] The post-treatment 15 can remove impurities and defects in the oxidized interface layer 14, thereby improving reliability.

[0061] refer to Figure 2D, a high-k layer 16 may be formed above the oxidized interface layer 14. The high-k layer 16 may be formed on the oxidized interface layer 14. The high-k layer 16 may be in direct contact with the oxidized interface layer 14. The high-k layer 16 may have a high dielectric constant (high-k). The dielectric constant of the high-k layer 16 may be greater than the dielectric constant of silicon oxide (SiO2), which is approximately 3.9. In addition, the physical thickness of the high-k layer 16 is greater than that of silicon oxide (SiO2) and may have a lower equivalent oxide thickness (EOT) value. The dielectric constant of the high-k layer 16 may be higher than that of the oxidized interface layer 14. The high-k layer 16 may be made of or include metal oxides, metal silicates, metal silicate nitrides, etc. The metal oxide may include oxides containing metals such as hafnium (Hf), aluminum (Al), lanthanum (La), zirconium (Zr), etc. The metal oxide may be made of or include hafnium oxide, aluminum oxide, lanthanum oxide, zirconium oxide, or a combination thereof. For example, the metal oxide may be made of or include HfO2, Al2O3, La2O3, ZrO2, or a combination thereof. The metal silicate may include a silicate containing a metal such as hafnium (Hf) and zirconium (Zr). For example, the metal silicate may be made of or include hafnium silicate (HfSiO), zirconium silicate (ZrSiO), or a combination thereof. The metal silicate nitride may be made of or include hafnium nitride silicate (HfSiON), zirconium nitride silicate (ZrSiON), or a combination thereof. In this embodiment of the present invention, the high-k layer 16 may include HfSiO, which may be formed by an atomic layer deposition (ALD) process.

[0062] A dielectric cap layer 17 may be formed over the high-k layer 16. The dielectric cap layer 17 may be formed on the high-k layer 16. The dielectric cap layer 17 may be in direct contact with the high-k layer 16. The dielectric cap layer 17 may contain an induced dipole substance. The induced dipole substance may diffuse into the space between the high-k layer 16 and the oxidized interface layer 14 through a subsequent process. The dielectric cap layer 17 may include lanthanum. The dielectric cap layer 17 may include lanthanum oxide (La2O3). The dielectric cap layer 17 may be formed by atomic layer deposition (ALD) or physical vapor deposition (PVD). The dielectric cap layer 17 may be a sacrificial layer that is removed after post-heat treatment.

[0063] A mask pattern 18 may be formed over the dielectric capping layer 17 of the first region R1. The mask pattern 18 may be formed in the first region R1, and the second region R2 may be exposed by the mask pattern 18. The mask pattern 18 may be or include a photoresist. A portion of the dielectric capping layer 17 may be exposed by the mask pattern 18, for example, the dielectric capping layer 17 of the second region R2 may be exposed.

[0064] refer to Figure 2E , a portion of the dielectric capping layer 17 may be removed by using the mask pattern 18. For example, HCl may be used to remove a portion of the dielectric capping layer 17. As shown in reference numeral 17B, the dielectric capping layer 17 may remain in the first region R1 and may be completely removed from the second region R2.

[0065] The mask pattern 18 may then be removed. After removing the mask pattern 18, the stack of the oxidized interface layer 14, the high-k layer 16, and the dielectric cap layer 17B may remain in the first region R1. In the second region R2, the stack of the channel layer 13, the oxidized interface layer 14, and the high-k layer 16 may remain.

[0066] refer to Figure 2F , a post-heat treatment 19 may be performed. The post-heat treatment 19 may allow the induced dipole species to diffuse from the dielectric cap layer 17B. The induced dipole species may diffuse in the first region R1 and may be positioned between the oxidized interface layer 14 and the high-k layer 16 of the first region R1.

[0067] The post-heat treatment 19 may form a dipole interface 20 between the high-k layer 16 and the oxidized interface layer 14 .

[0068] The post-heat treatment 19 may include plasma nitridation and annealing. Through the plasma nitridation and annealing, nitrogen may be implanted into the high-k layer 16. For example, HfSiO as the high-k layer 16 may be reformed into HfSiON through the post-heat treatment 19.

[0069] As described above, the high-k layer 16 may be nitrided as long as the dipole interface 20 is formed by the post-heat treatment 19. The dipole interface 20 may include diffused dipole-inducing species. The dipole interface 20 may include lanthanum. The dipole interface 20 may include a lanthanum-diffused dipole interface.

[0070] Furthermore, residual lanthanum of the remaining non-diffused dielectric cap layer 17B may diffuse in a subsequent process, thereby deteriorating the reliability of the transistor.

[0071] Therefore, if Figure 2G As shown, according to an embodiment of the present invention, the dielectric capping layer 17B can be completely removed from the first region R1 by using a mask pattern M1. The mask pattern M1 may include a photoresist pattern, and the mask pattern M1 may cover the first region R1. Therefore, the dielectric capping layer 17B can be completely removed from the first region R1 and the second region R2.

[0072] The stack including the oxidized interface layer 14, the high-k layer 16, and the dipole interface 20 between the oxidized interface layer 14 and the high-k layer 16 may remain on the substrate 11 of the first region R1. The stack including the P-channel layer 13, the oxidized interface layer 14, and the high-k layer 16 may remain on the substrate 11 of the second region R2.

[0073] refer to Figure 2H , after removing the mask pattern M1, a gate conductive layer may be formed on the high-k layer 16. In the gate conductive layer, a metal-containing layer 21 and a low-resistivity layer 22 may be sequentially formed. The metal-containing layer 21 may be made of or include a metal and a metal nitride. In one embodiment, the metal-containing layer 21 may be titanium nitride, and the low-resistivity layer 22 may be or include tungsten.

[0074] A hard mask layer 23 may be formed over the low resistivity layer 22. The hard mask layer 23 may be formed on the low resistivity layer 22. The hard mask layer 23 may be in direct contact with the low resistivity layer 22. The hard mask layer 23 may include silicon nitride.

[0075] refer to Fig.2I , a gate patterning process may be performed. The hard mask layer 23, the low resistivity layer 22, the metal-containing layer 21, the high-k layer 16, the dipole interface 20, and the oxidized interface layer 14 may be sequentially etched by using a gate mask (not shown in the figure). Therefore, a first gate stack NG may be formed on the substrate 11 of the first region R1, and a second gate stack PG may be formed on the substrate 11 of the second region R2.

[0076] The first gate stack NG may include a first oxidized interface layer 14N, a first high-k layer 16N, a first metal electrode 21N, a first low-resistivity electrode 22N, and a first hard mask layer 23N stacked in sequence. The first gate stack NG may also include a dipole interface 20N between the first oxidized interface layer 14N and the high-k layer 16N. The dipole interface 20N may include an induced dipole substance.

[0077] The second gate stack PG may include a second oxidized interface layer 14P, a second high-k layer 16P, a second metal electrode 21P, a second low-resistivity electrode 22P, and a second hard mask layer 23P stacked in sequence. The second gate stack PG may not include a dipole interface 20N.

[0078] The first oxidized interface layer 14N and the second oxidized interface layer 14P may be made of the same material and may have the same thickness. The first high-k layer 16N and the second high-k layer 16P may be made of the same material and may have the same thickness. The first metal electrode 21N and the second metal electrode 21P may be made of the same material and may have the same thickness.

[0079] After the gate patterning process, a process known in the art to which the present invention belongs may be performed. For example, a process of forming a source region and a drain region may be performed. The source region and the drain region may include a first source / drain region 24N / 25N and a second source / drain region 24P / 25P. The first source / drain region 24N / 25N may be formed in the first region R1. The first source / drain region 24N / 25N region may include an N-type dopant. The second source / drain region 24P / 25P may be formed in the second region R2. The second source / drain region 24P / 25P may include a P-type dopant.

[0080] As described above, a first transistor may be formed by forming the first source / drain region 24N / 25N. The first transistor may include a first gate stack NG and the first source / drain region 24N / 25N. The first transistor may include an N-channel transistor, including an NMOSFET.

[0081] The second transistor may be formed by forming the second source / drain region 24P / 25P. The second transistor may include a second gate stack PG and the second source / drain region 24P / 25P. The second transistor may include a P-channel transistor, including a PMOSFET.

[0082] Figure 3 is a cross-sectional view showing a semiconductor device according to another embodiment of the present invention.

[0083] refer to Figure 3 , the semiconductor device 200 may include a first transistor T1 , a second transistor T2 , a third transistor T3 , and a fourth transistor T4 . The first to fourth transistors T1 to T4 may be separated by an isolation layer 32 .

[0084] The semiconductor device 200 may include a first region R1, a second region R2, a third region R3, and a fourth region R4. Each of the first region R1, the second region R2, the third region R3, and the fourth region R4 may include a substrate 31, and the first region R1 and the second region R2, the third region R3, and the fourth region R4 may be separated from each other by an isolation layer 32. The first region R1 may be a region where the first transistor T1 is formed, the second region R2 may be a region where the second transistor T2 is formed. The third region R3 may be a region where the third transistor T3 is formed, and the fourth region R4 may be a region where the fourth transistor T4 is formed.

[0085] The first transistor T1 may include: a first gate stack N1 formed above the substrate 31, and a first source region 44N and a first drain region 45N formed in the substrate 31 and located on both sides of the first gate stack N1. The first gate stack N1 may include a first interface layer 34N, a first high-k layer 37N, a first metal electrode 41N, a first low-resistivity electrode 42N, and a first hard mask layer 43N stacked in sequence. The first gate stack N1 may also include a dipole interface 40N between the first interface layer 34N and the first high-k layer 37N. The dipole interface 40N may include an induced dipole substance for modulating the threshold voltage of the first transistor T1. The threshold voltage of the first transistor T1 may be modulated by the dipole interface 40N. The dipole interface 40N may be made of or include a lanthanide rare earth metal.

[0086] The second transistor T2 may include: a second gate stack P1 formed above the substrate 31, and a second source region 44P and a second drain region 45P formed in the substrate 31 on both sides of the second gate stack P1. The second gate stack P1 may include a second interface layer 34P, a second high-k layer 37P, a second metal electrode 41P, a second low-resistivity electrode 42P, and a second hard mask layer 43P stacked in sequence. The second transistor T2 may also include a P-channel layer 33P, and the P-channel layer 33P may be formed above the substrate 31 below the second gate stack P1. The P-channel layer 33P may be formed on the substrate 31 below the second gate stack P1. The P-channel layer 33P may be in direct contact with the substrate 31 and the second gate stack P1. The P-channel layer 33P may be crystalline and may contain a large amount of germanium. The P-channel layer 33P may be made of or include silicon germanium. The second metal electrode 41P may have a work function suitable for the second transistor T2.

[0087] The third transistor T3 may include: a third gate stack N2 formed above the substrate 31, and a third source region 44N2 and a third drain region 45N2 formed in the substrate 31 and located on both sides of the third gate stack N2. The third gate stack N2 may include a third interface layer 36N2, a third high-k layer 37N2, a third metal electrode 41N2, a third low-resistivity electrode 42N2, and a third hard mask layer 43N2, which are stacked in the order described. The third gate stack N2 may also include a dipole interface 40N2 between the third interface layer 36N2 and the third high-k layer 37N2. The dipole interface 40N2 may include an induced dipole substance for modulating the threshold voltage of the third transistor T3. The threshold voltage of the third transistor T3 may be modulated by the dipole interface 40N2. The dipole interface 40N2 may be made of or include a lanthanide rare earth metal.

[0088] The fourth transistor T4 may include: a fourth gate stack P2 formed above the substrate 31, and a fourth source region 44P2 and a fourth drain region 45P2 formed in the substrate 31 and located on both sides of the fourth gate stack P2. The fourth gate stack P2 may include a fourth interface layer 36P2, a fourth high-k layer 37P2, a fourth metal electrode 41P2, a fourth low-resistivity electrode 42P2, and a fourth hard mask layer 43P2 stacked in sequence. The fourth transistor T4 may also include a P-channel layer 33P2, and the P-channel layer 33P2 may be formed above the substrate 31 and below the fourth gate stack P2. The P-channel layer 33P2 may be formed below the fourth gate stack P2 on the substrate 31. The P-channel layer 33P2 may be in direct contact with the substrate 31 and the fourth gate stack P2. The P-channel layer 33P2 may be crystalline and may contain a large amount of germanium. The P-channel layer 33P2 may be made of or include silicon germanium. The fourth metal electrode 41P2 may have a work function suitable for the fourth transistor T4.

[0089] The P channel layers 33P and 33P2 may be formed by a selective epitaxial growth (SEG) process. The thickness of the P channel layers 33P and 33P2 may be approximately or less. The germanium concentration of the P channel layers 33P and 33P2 may be about 20 at % to about 50 at %. The P channel layers 33P and 33P2 may be formed of only silicon germanium. The silicon capping layer may be omitted.

[0090] refer to Figure 3 , the first gate stack N1 may include a first interface layer 34N, and the second gate stack P1 may include a second interface layer 34P. The third gate stack N2 may include a third interface layer 36N2, and the fourth gate stack P2 may include a fourth interface layer 36P2. The first interface layer 34N and the second interface layer 34P may be thicker than the third interface layer 36N2 and the fourth interface layer 36P2. The first interface layer 34N and the second interface layer 34P may be referred to as thick interface layers, and the third interface layer 36N2 and the fourth interface layer 36P2 may be referred to as thin interface layers.

[0091] The first interface layer 34N and the second interface layer 34P may be made of the same material. The first interface layer 34N and the second interface layer 34P may be made of silicon oxide. The first interface layer 34N and the second interface layer 34P may be made of silicon oxide provided by a combined treatment of a deposition process and a post-oxidation process. For example, the first interface layer 34N and the second interface layer 34P may be oxidized silicon oxide. The oxidized silicon oxide may be a material obtained by converting a deposited silicon oxide by a free radical oxidation process. In other words, the first interface layer 34N and the second interface layer 34P may be a converted silicon oxide provided by a combined treatment of a deposition process and a post-oxidation process. The deposited silicon oxide may be deposited by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

[0092] Since the deposited silicon oxide is converted into oxidized silicon oxide, the first interface layer 34N and the second interface layer 34P may include high-quality SiO2. High-quality silicon oxide (HQSiO2) may not contain impurities. For example, the deposited silicon oxide may contain a large amount of impurities in the layer, but as the deposited silicon oxide is converted by the free radical oxidation process, the impurities are removed.

[0093] The third interface layer 36N2 and the fourth interface layer 36P2 may be made of the same material. The third interface layer 36N2 and the fourth interface layer 36P2 may be made of silicon oxide. The third interface layer 36N2 and the fourth interface layer 36P2 may be made of silicon oxide provided by an oxidation process. For example, each of the third interface layer 36N2 and the fourth interface layer 36P2 may be an oxidized silicon oxide. The third interface layer 36N2 may be formed by directly oxidizing the surface of the substrate 31. The fourth interface layer 36P2 may be formed by directly oxidizing the surface of the P channel layer 33P2.

[0094] As described above, the first interface layer 34N and the second interface layer 34P may be conversion silicon oxide obtained by a combination of a deposition process and a post-oxidation process, and the third interface layer 36N2 and the fourth interface layer 36P2 may be oxidized silicon oxide obtained by a direct oxidation process. The conversion silicon oxide may be thicker than the oxidized silicon oxide.

[0095] The first to fourth high-k layers 37N, 37P, 37N2 and 37P2 may include a high-k material having a high-k constant. The dielectric constant of the first to fourth high-k layers 37N, 37P, 37N2 and 37P2 may be greater than the dielectric constant of SiO2 (which is approximately 3.9). In addition, the physical thickness of the first to fourth high-k layers 37N, 37P, 37N2 and 37P2 is significantly greater than SiO2 and may have a lower equivalent oxide thickness (EOT) value. The dielectric constant of the first to fourth high-k layers 37N, 37P, 37N2 and 37P2 may be greater than the dielectric constant of the first to fourth interface layers 34N, 34P, 36N2 and 36P2.

[0096] The first to fourth high-k layers 37N, 37P, 37N2 and 37P2 may include metal-containing materials, such as metal oxides, metal silicates and metal silicate nitrides. The metal oxide may include oxides containing metals such as hafnium (Hf), aluminum (Al), lanthanum (La), zirconium (Zr), etc. The metal oxide may be made of or include hafnium oxide, aluminum oxide, lanthanum oxide, zirconium oxide or a combination thereof. For example, the metal oxide may be made of or include HfO2, Al2O3, La2O3, ZrO2 or a combination thereof. The metal silicate may include a silicate containing metals such as hafnium (Hf) and zirconium (Zr). For example, the metal silicate may be made of or include hafnium silicate (HfSiO), zirconium silicate (ZrSiO) or a combination thereof. The metal silicate nitride may be made of or include hafnium silicate nitride (HfSiON), zirconium silicate nitride (ZrSiON), or a combination thereof.

[0097] The first to fourth metal electrodes 41N, 41P, 41N2, and 41P2 may be made of the same material. For example, the first to fourth metal electrodes 41N, 41P, 41N2, and 41P2 may include titanium nitride.

[0098] The first to fourth low-resistivity electrodes 42N, 42P, 42N2, and 42P2 may be made of the same material. For example, the first to fourth low-resistivity electrodes 42N, 42P, 42N2, and 42P2 may include a tungsten-based material.

[0099] According to another embodiment of the present invention, the first metal electrode 41N and the third metal electrode 41N2 may have a low work function, while the second and fourth metal electrodes 41P and 41P2 may have a high work function. In this article, the low work function may refer to a work function lower than the mid-gap work function of silicon, and the high work function may refer to a work function higher than the mid-gap work function of silicon. The first metal electrode 41N and the third metal electrode 41N2 may include titanium-rich titanium nitride containing excess titanium. The second and fourth metal electrodes 41P and 41P2 may include nitrogen-rich titanium nitride containing excess nitrogen.

[0100] The first to fourth low-resistivity electrodes 42N, 42P, 42N2, and 42P2 may be made of or include a low-resistivity metal, such as tungsten. The first to fourth low-resistivity electrodes 42N, 42P, 42N2, and 42P2 may respectively reduce the resistance of the first to fourth gate stacks N1, P1, N2, and P2.

[0101] According to the above-described embodiment, a dipole interface 40N may be formed at the interface between the first interface layer 34N and the first high-k layer 37N. Therefore, the threshold voltage of the first transistor T1 may be modulated.

[0102] In the third gate stack N2, a dipole interface 40N2 may be formed at an interface between the third interface layer 36N2 and the third high-k layer 37N2. Therefore, the threshold voltage of the third transistor T3 may be modulated.

[0103] Since the P-channel layer 33P is formed below the second gate stack P1, the threshold voltage of the second transistor T2 can be modulated. That is, since the P-channel layer 33P contains a large amount of germanium, the energy band gap can be reduced, so that the threshold voltage of the second transistor T2 can be modulated to a level suitable for the second transistor T2.

[0104] Since the P-channel layer 33P2 is formed below the fourth gate stack P2, the threshold voltage of the fourth transistor T4 can be modulated. That is, since the P-channel layer 33P2 contains a large amount of germanium, the energy band gap can be reduced, so that the threshold voltage of the fourth transistor T4 can be modulated to a level suitable for the fourth transistor T4.

[0105] The first gate stack N1 and the third gate stack N2 may be gate stacks of NMOSFET, and they may not contain a La2O3 capping layer, ie, the lanthanum oxide capping layer is omitted.

[0106] The first transistor T1 and the third transistor T3 may be N-channel transistors, and the second transistor T2 and the fourth transistor T4 may be P-channel transistors.

[0107] Figures 4A to 4K is a cross-sectional view illustrating a method of manufacturing a semiconductor device according to another embodiment of the present invention.

[0108] refer to Figure 4A , a substrate 31 may be prepared. The substrate 31 may include a plurality of regions in which transistors are formed. These regions may include first to fourth regions R1 to R4. The substrate 31 may be made of or include a semiconductor material. The substrate 31 may include a semiconductor substrate. The substrate 31 may include a silicon substrate, a silicon germanium substrate, or a silicon on insulator (SOI) substrate.

[0109] The isolation layer 32 may be formed in the substrate 31. The isolation layer 32 may be formed by a shallow trench isolation (STI) process.

[0110] The first to fourth regions R1 to R4 may be isolated from each other by an isolation layer 32. The first region R1 and the third region R3 may be regions where an NMOSFET is to be formed. The first region R1 may be a region where an NMOSFET with a thick gate dielectric layer is to be formed, and the third region R3 may be a region where an NMOSFET with a thin gate dielectric layer is to be formed. The second region R2 and the fourth region R4 may be regions where a PMOSFET is to be formed. The second region R2 may be a region where a PMOSFET with a thick gate dielectric layer is to be formed, and the fourth region R4 may be a region where a PMOSFET with a thin gate dielectric layer is to be formed. Although not shown, a well may be formed in the substrate 31 by a typical well formation process. A P-type well may be formed in the substrate 31 of the first region R1 and the third region R3, and an N-type well may be formed in the substrate 31 of the second region R2 and the fourth region R4.

[0111] Subsequently, a P-channel layer 33 may be selectively formed over the substrate 31 of the second region R2 and the fourth region R4. The P-channel layer 33 may be made of or include silicon germanium. The P-channel layer 33 may be crystalline. The P-channel layer 33 may be formed by a selective epitaxial growth (SEG) process. The thickness of the P-channel layer 33 may be 1000 Å. The germanium concentration of the P-channel layer 33 may be about 20 at % to about 50 at %. The P-channel layer 33 may be formed of only silicon germanium. The silicon capping layer may be omitted.

[0112] Subsequently, a deposition type interface layer 34A may be formed. The deposition type interface layer 34A may be formed on the surface of the substrate 31 in the first region R1 and the third region R3. The deposition type interface layer 34A may be formed on the P channel layer 33 in the second region R2 and the fourth region R4. The deposition type interface layer 34A may be made of or include an oxide or an oxide-based material. The deposition type interface layer 34A may be made of or include silicon oxide. The deposition type interface layer 34A may be formed by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process. The deposition type interface layer 34A may be formed to be approximately The deposition type interface layer 34A may be formed to a thickness of approximately To about Thickness.

[0113] As described above, the deposition type interface layer 34A may be formed by a deposition process rather than a thermal oxidation process. That is, the deposition type interface layer 34A may include a deposition type oxide. The oxide formed by the thermal oxidation process may be simply referred to as an oxidized oxide.

[0114] The deposited oxide can be formed to a relatively thicker thickness than the thermal oxide, but may include a large amount of impurities. In addition, the deposited oxide can suppress excessive oxidation of the P channel layer 33.

[0115] The deposition type interface layer 34A may be made of or include silicon oxide. The silicon oxide may be deposition type silicon oxide. For example, the silicon oxide may be formed using chemical vapor deposition (CVD) or atomic layer deposition (ALD).

[0116] refer to Figure 4B , post-processing 35 may be performed. Post-processing 35 may include a free radical oxidation process. The free radical oxidation process may be referred to as a post-oxidation process. The free radical oxidation process may be performed using oxygen free radicals (O*). The free radical oxidation process may improve the reliability of the deposition type interface layer 34A. For example, impurities may be removed from the deposition type interface layer 34A by a free radical oxidation process. Impurities may be removed by oxygen free radicals. Therefore, the deposition type interface layer 34A may be converted into an oxidized interface layer 34 as shown in the figure numeral 34. Hereinafter, the oxidized interface layer 34 is simply referred to as the conversion interface layer 34.

[0117] Since the conversion interface layer 34 does not contain impurities, it can have excellent film quality and low defects. When the deposition interface layer 34A is silicon oxide, high-quality silicon oxide (HQ SiO2) can be formed by the radical oxidation process 35. The conversion interface layer 34 can be high-quality silicon oxide.

[0118] According to another embodiment of the present invention, the post-treatment 35 for forming the conversion interface layer 34 may include a dry oxidation process. According to another embodiment of the present invention, after forming the conversion interface layer 34, a plasma nitridation process and an annealing process may be performed.

[0119] Post-treatment 35 may remove impurities and defects from conversion interface layer 34, thereby improving reliability.

[0120] refer to Figure 4C , the conversion type interface layer 34 may be removed from the third region R3 and the fourth region R4 by using the mask pattern M2. The mask pattern M2 may include a photoresist pattern, and the mask pattern M2 may cover the first region R1 and the second region R2.

[0121] When the conversion interface layer 34 is removed from the third and fourth regions R3 and R4, the surface of the substrate 31 in the third and fourth regions R3 and the surface of the P channel layer 33 in the fourth region R4 may be exposed. The conversion interface layer 34 may remain in the first and second regions R1 and R2.

[0122] refer to Figure 4DAfter removing the mask pattern M2, an oxidized interface layer 36 may be formed. The oxidized interface layer 36 may be formed by directly oxidizing the exposed surface of the substrate 31 in the third region R3 and the exposed surface of the P channel layer 33 in the fourth region R4. The oxidized interface layer 36 may be formed by performing a thermal oxidation process in an oxygen atmosphere. The oxidized interface layer 36 may be thinner than the conversion interface layer 34. The thickness of the oxidized interface layer 36 may be about 1000 Å. or less. The conversion interface layer 34 may be referred to as a 'thick oxidized interface layer', while the oxidized interface layer 36 may be referred to as a 'thin oxidized interface layer'. The terms 'thick oxidized interface layer' and 'thin oxidized interface layer' are used herein as relative terms to compare the thickness of the conversion interface layer 34 and the oxidized interface layer 36. Therefore, relative to the conversion interface layer 34, the oxidized interface layer 36 is a 'thin oxidized interface layer'.

[0123] As described above, the conversion interface layer 34 may be formed in the first and second regions R1 and R2, and the oxidation interface layer 36 may be formed in the third and fourth regions R3 and R4. Due to the materials used to prepare the various layers, the oxidation interface layer 36 is not formed on the conversion interface layer 34 or the isolation layer 32.

[0124] refer to Figure 4E, a high-k layer 37 may be formed on the conversion interface layer 34 and the oxidized interface layer 36. The high-k layer 37 may be formed on the conversion interface layer 34 and the oxidized interface layer 36. The high-k layer 37 may be in direct contact with the conversion interface layer 34 and the oxidized interface layer 36. The high-k layer 37 may have a high dielectric constant (high-k), that is, its dielectric constant is greater than the dielectric constant of silicon oxide (SiO2) (which is approximately 3.9). In addition, the physical thickness of the high-k layer 37 is greater than that of silicon oxide (SiO2) and may have a lower equivalent oxide thickness (EOT) value. The dielectric constant of the high-k layer 37 may be higher than that of the oxidized interface layer 36. The high-k layer 37 may be made of or include metal oxides, metal silicates, metal silicate nitrides, etc. The metal oxide may include oxides containing metals such as hafnium (Hf), aluminum (Al), lanthanum (La), zirconium (Zr), etc. The metal oxide may be made of hafnium oxide, aluminum oxide, lanthanum oxide, zirconium oxide or a combination thereof or include the oxide or a combination thereof. For example, the metal oxide may be made of HfO2, Al2O3, La2O3, ZrO2 or a combination thereof or include the metal oxide or a combination thereof. The metal silicate may include a silicate containing metals such as hafnium (Hf) and zirconium (Zr). For example, the metal silicate may be made of hafnium silicate (HfSiO), zirconium silicate (ZrSiO) or a combination thereof or include the metal silicate or a combination thereof. The metal silicate nitride may be made of hafnium nitride silicate (HfSiON), zirconium nitride silicate (ZrSiON) or a combination thereof or include the metal silicate nitride or a combination thereof. According to this embodiment of the present invention, the high-k layer 37 may include HfSiO, which may be formed by an atomic layer deposition (ALD) process.

[0125] A dielectric cap layer 38 may be formed over the high-k layer 37. The dielectric cap layer 38 may be formed on the high-k layer 37. The dielectric cap layer 38 may be in direct contact with the high-k layer 37. The dielectric cap layer 38 may contain an induced dipole substance. The induced dipole substance may diffuse into the interface between the high-k layer 37 and the conversion interface layer 34 in a subsequent process. The induced dipole substance may diffuse into the interface between the high-k layer 37 and the oxidized interface layer 36 in a subsequent process. The dielectric cap layer 38 may include lanthanum. The dielectric cap layer 38 may include lanthanum oxide (La2O3). The dielectric cap layer 38 may be formed by an atomic layer deposition (ALD) process or a physical vapor deposition (PVD) process.

[0126] Subsequently, a mask pattern 39 may be formed. The mask pattern 39 may include a photoresist. The mask pattern 39 may be formed only in the first region R1 and the third region R3. The second region R2 and the fourth region R4 may be exposed by the mask pattern 39. The mask pattern 39 may expose the dielectric capping layer 38 of the second region R2 and the fourth region R4. The mask pattern 39 may be formed on the dielectric capping layer 38 of the first region R1 and the third region R3. The mask pattern 39 may be in direct contact with the dielectric capping layer 38 of the first region R1 and the third region R3.

[0127] refer to Figure 4F , a portion of the dielectric capping layer 38 may be removed by using the mask pattern 39. The dielectric capping layer 38 in the second region R2 and the fourth region R4 may be removed. For example, the dielectric capping layer 38 may be removed by using HCl. The dielectric capping layer 38 may remain in the first region R1 and the third region R3, while the dielectric capping layer 38 may be completely removed from the second region R2 and the fourth region R4.

[0128] The mask pattern 39 may then be removed. After removing the mask pattern 39, the stack of the conversion interface layer 34, the high-k layer 37, and the dielectric cap layer 38 may remain in the first region R1. After removing the mask pattern 39, the stack of the oxidized interface layer 36, the high-k layer 37, and the dielectric cap layer 38 may remain in the third region R3. In the second region R2, the stack of the P-channel layer 33, the conversion interface layer 34, and the high-k layer 37 may remain. In the fourth region R4, the stack of the P-channel layer 33, the oxidized interface layer 36, and the high-k layer 37 may remain.

[0129] refer to Figure 4G , a post-heat treatment 39 may be performed. The post-heat treatment 39 may cause the induced dipole material to diffuse from the dielectric cover layer 38. The induced dipole material may diffuse in the first region R1 and the third region R3. The induced dipole material may be positioned at the interface between the high-k layer 37 and the conversion interface layer 34 in the first region R1. Therefore, a dipole interface 40 may be formed at the interface between the high-k layer 37 and the conversion interface layer 34 in the first region R1 by the post-heat treatment 39. The induced dipole material may be positioned at the interface between the high-k layer 37 and the oxidized interface layer 36 in the third region R3. Therefore, a dipole interface 40 may be formed at the interface between the high-k layer 37 and the oxidized interface layer 36 in the third region R3 by the post-heat treatment 39. The dipole interface 40 may include diffused induced dipole material. The dipole interface 40 may include lanthanum. The dipole interface 40 may include a dipole interface diffused with lanthanum.

[0130] The post-heat treatment 39 may include a plasma nitridation process and an annealing process. Nitrogen may be implanted into the high-k layer 37 through the plasma nitridation process and the annealing process. For example, HfSiO as the high-k layer 16 may be reformed into HfSiON through the post-heat treatment 19.

[0131] As described above, once the dipole interface 40 is formed by the post-heat treatment 39 , the high-k layer 37 may be nitrided.

[0132] Furthermore, residual lanthanum of the remaining non-diffused dielectric cap layer 38 may diffuse in subsequent processes, thereby impairing the reliability of the transistor.

[0133] According to one embodiment of the present invention, Figure 4H As shown, the dielectric cover layer 38 may be removed from the first region R1 and the third region R3. The dielectric cover layer 38 may be removed from all of the first to fourth regions R1 to R4.

[0134] The stack of the conversion interface layer 34 and the high-k layer 37 may remain on the substrate 31 of the first region R1, and the dipole interface 40 may be formed on the interface between the conversion interface layer 34 and the high-k layer 37. The stack of the P-channel layer 33, the conversion interface layer 34, and the high-k layer 37 may remain on the substrate 31 of the second region R2. The stack of the oxidized interface layer 36 and the high-k layer 37 may remain on the substrate 31 of the third region R3, and the dipole interface 40 may be formed on the interface between the oxidized interface layer 36 and the high-k layer 37. The stack of the P-channel layer 33, the oxidized interface layer 36, and the high-k layer 37 may remain on the substrate 31 of the fourth region R4.

[0135] refer to Fig. 4I , a gate conductive layer may be formed on the high-k layer 37 of each region. The gate conductive layer may be formed on the high-k layer 37 of each region and may directly contact the high-k layer 37 of each region. In the gate conductive layer, a metal-containing layer 41 and a low-resistivity layer 42 may be formed in sequence. The metal-containing layer 41 may be made of or include a metal and a metal nitride. The metal-containing layer 41 may be titanium nitride, and the low-resistivity layer 42 may include tungsten.

[0136] A hard mask layer 43 may be formed on the low resistivity layer 42. The hard mask layer 43 may be formed on the low resistivity layer 42. The hard mask layer 43 may directly contact the low resistivity layer 42. The hard mask layer 43 may be made of or include silicon nitride.

[0137] refer to Figure 4J , a gate patterning process may be performed. By using a gate mask (which is not shown), gate stacks N1, P1, N2, and P2 may be formed in the regions R1 to R4.

[0138] A first gate stack N1 may be formed in the first region R1, and a second gate stack P1 may be formed in the second region R2. A third gate stack N2 may be formed in the third region R3, and a fourth gate stack P2 may be formed in the fourth region R4.

[0139] The first gate stack N1 may be formed by sequentially etching the hard mask layer 43, the low resistivity layer 42, the metal-containing layer 41, the high-k layer 37, the dipole interface 40, and the conversion interface layer 34 in the first region R1. The first gate stack N1 may include a first conversion interface layer 34N, a first high-k layer 37N, a first metal electrode 41N, a first low resistivity electrode 42N, and a first hard mask layer 43N stacked in sequence. The first gate stack N1 may also include a dipole interface 40N between the first conversion interface layer 34N and the first high-k layer 37N. The dipole interface 40N may include an induced dipole substance.

[0140] The second gate stack P1 may be formed by sequentially etching the hard mask layer 43, the low resistivity layer 42, the metal containing layer 41, the high-k layer 37, the conversion interface layer 34, and the P-channel layer 33 in the second region R2. The second gate stack P1 may include a P-channel layer 33P, a second conversion interface layer 34P, a second high-k layer 37P, a second metal electrode 41P and a second low resistivity electrode 42P, and a second hard mask layer 43P.

[0141] The third gate stack N2 may be formed by sequentially etching the hard mask layer 43, the low resistivity layer 42, the metal-containing layer 41, the high-k layer 37, the dipole interface 40, and the oxidized interface layer 36 in the third region R3. The third gate stack N2 may include a first oxidized interface layer 36N2, a third high-k layer 37N2, a third metal electrode 41N2, a third low resistivity electrode 42N2, and a third hard mask layer 43N2 stacked in sequence. The third gate stack N2 may also include a dipole interface 40N2 between the first oxidized interface layer 36N2 and the third high-k layer 37N2. The dipole interface 40N2 may include an induced dipole substance.

[0142] The fourth gate stack P2 may be formed by sequentially etching the hard mask layer 43, the low resistivity layer 42, the metal containing layer 41, the high-k layer 37, the oxidized interface 36, and the P-channel layer 33 in the fourth region R4. The fourth gate stack P2 may include a second oxidized interface layer 36P2, a fourth high-k layer 37P2, a fourth metal electrode 41P2, a fourth low resistivity electrode 42P2, and a fourth hard mask layer 43P2 stacked in sequence.

[0143] As described above, the first gate stack N1 and the third gate stack N2 may include the dipole interfaces 40N and 40N2 , respectively, and the second gate stack P1 and the fourth gate stack P2 may not include the dipole interfaces 40N and 40N2 .

[0144] The first gate stack N1 and the second gate stack P1 may include a first conversion type interface layer 34N and a second conversion type interface layer 34P, respectively. The third gate stack N2 and the fourth gate stack P2 may include a first oxidation type interface layer 36N2 and a second oxidation type interface layer 36P2, respectively.

[0145] The first conversion interface layer 34N and the second conversion interface layer 34P may be made of the same material and may have the same thickness. The first oxidized interface layer 36N2 and the second oxidized interface layer 36P2 may be made of the same material and may have the same thickness. The first conversion interface layer 34N and the second conversion interface layer 34P may be thicker than the first oxidized interface layer 36N2 and the second oxidized interface layer 36P2. The first conversion interface layer 34N and the second conversion interface layer 34P may have a first thickness. The first oxidized interface layer 36N2 and the second oxidized interface layer 36P2 may have a second thickness, and the first thickness may be greater than the second thickness. The first conversion interface layer 34N and the second conversion interface layer 34P may be formed of a material different from the first oxidized interface layer 36N2 and the second oxidized interface layer 36P2. The first conversion interface layer 34N and the second conversion interface layer 34P may have a higher quality than the first oxidized interface layer 36N2 and the second oxidized interface layer 36P2. The so-called higher quality means having fewer impurities or defects.

[0146] The first to fourth high-k layers 37N, 37P, 37N2, and 37P2 may be made of the same material and may have the same thickness. The first to fourth metal electrodes 41N, 41P, 41N2, and 41P2 may be made of the same material and may have the same thickness. The first to fourth low-resistivity electrodes 42N, 42P, 42N2, and 42P2 may be made of the same material and may have the same thickness. The first to fourth hard mask layers 43N, 43P, 43N2, and 43P2 may be made of the same material and may have the same thickness.

[0147] After the gate patterning process, a process known in the art to which the present invention belongs may be performed. For example, a process of forming a source / drain region, etc. may be performed. A first source / drain region 44N / 45N may be formed in the first region R1, and a second source / drain region 44P / 45P may be formed in the second region R2. A third source / drain region 44N2 / 45N2 may be formed in the third region R3, and a fourth source / drain region 44P2 / 45P2 may be formed in the fourth region R4.

[0148] In one embodiment, the first source / drain 44N / 45N and the third source / drain region 44N2 / 45N2 may include N-type dopants, and the second source / drain region 44P / 45P and the fourth source / drain region 44P2 / 45P2 may include P-type dopants.

[0149] According to one embodiment of the present invention, a method for manufacturing a semiconductor device may include preparing a substrate 31 including a first region R1 and a second region R2. The R1 and R2 regions may be defined by an isolation layer 32. Then, a channel layer 33 may be formed on the substrate 31 of the second region R2. The channel layer 33 may be in direct contact with the highest surface of the substrate 31 of the second region R2. The method may also include forming a deposition type interface layer 34A on the channel layer 33 and the substrate 31 of the first region R1. The deposition type interface layer 34A may be formed on the channel layer 33 and the substrate 31 of the first region R1, and may be in direct contact with the channel layer 33 and the substrate 31 of the first region R1. The method may also include converting the deposition type interface layer 34A into an oxidized interface layer 34, and forming a high-k layer 37 on the oxidized interface layer 34. A dipole interface 40 may be formed at the interface between the high-k layer 37 of the first region R1 and the oxidized interface layer 34. The method may also include forming a conductive layer 41 / 42 on the high-k layer 37. The method may also include patterning the conductive layer, the high-k layer, the dipole interface, and the oxidized interface layer to form a first gate stack N1 on the substrate 31 of the first region R1, and patterning the conductive layer, the high-k layer, and the oxidized interface layer to form a second gate stack P1 on the substrate 31 of the second region R2.

[0150] According to one embodiment of the present invention, a method for manufacturing a semiconductor device may include: preparing a substrate 31 including first and second transistor regions (corresponding to R1 and R3), forming a thick deposition type interface layer 34A on the substrate 31 of the first transistor region (R1), and converting the thick deposition type interface layer 34A into a thick oxidation type interface layer 34. The method may also include forming a thin oxidation type interface layer 36 on the substrate 31 of the second transistor region (R3), and forming a high-k layer 37 on the thick oxidation type interface layer 34 and the thin oxidation type interface layer 36. A first dipole interface 40 may be formed at the interface between the high-k layer 37 of the first transistor region R1 and the thick oxidation type interface layer 34. The method may also include forming gate conductive layers 41 and 42 on the high-k layer 37.

[0151] Although the above embodiment of the present invention describes a method for manufacturing a CMOSFET, the concept and spirit of the present invention are not limited to manufacturing only CMOSFET, but can be applied to various semiconductor devices that form N-channel transistors and P-channel transistors. For example, the technology of the present invention can be applied to memory devices such as DRAM, FeRAM, 3D NAND, PCRAM, STT-RAM, etc. The technology of the present invention can also be applied to CMOSFET in peripheral circuits of CMOS image sensors (CIS) that need to operate at high speed.

[0152] According to an embodiment of the present invention, a high-quality interface layer is formed by converting a deposited interface layer into an oxidized interface layer, thereby significantly reducing impurities and defects in the gate dielectric layer, thereby improving the reliability of the gate dielectric layer.

[0153] Additionally, the reliability of the transistor can be improved by removing the dielectric capping layer after forming the dipole interface.

[0154] While the present invention has been described with respect to the specific embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for manufacturing a semiconductor device, comprising: forming a deposition type interface layer on the substrate; removing impurities and defects contained in the deposited interface layer by exposing the deposited interface layer to a free radical oxidation process to convert the deposited interface layer into an oxidized interface layer; forming a high-k layer on the oxidized interface layer; forming a dipole interface at the interface between the high-k layer and the oxidized interface layer; forming a conductive layer on the high-k layer; as well as The conductive layer, the high-k layer, the dipole interface, and the oxidized interface layer are patterned to form a gate stack on the substrate.

2. The method according to claim 1, wherein: Forming the deposition type interface layer on the substrate is performed by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

3. The method according to claim 1, wherein: The deposited interface layer includes deposited silicon oxide.

4. The method according to claim 1, wherein: The oxidized interface layer includes oxidized silicon oxide.

5. The method according to claim 1, wherein: The dipole interface includes a dipole-inducing substance.

6. The method according to claim 5, wherein: The dipole-inducing substance includes lanthanum.

7. The method according to claim 1, wherein: Forming the dipole interface at the interface between the high-k layer and the oxidized interface layer comprises: forming a sacrificial layer containing a dipole-inducing substance on the high-k layer; exposing the sacrificial layer to a post-heat treatment to diffuse the dipole-inducing species into an interface between the oxidized interface layer and the high-k layer; and The sacrificial layer is removed.

8. The method according to claim 7, wherein: The sacrificial layer includes lanthanum oxide.

9. The method according to claim 1, wherein: The gate stack includes a gate stack of an NMOSFET.

10. A method for manufacturing a semiconductor device, comprising: preparing a substrate including a first region and a second region; forming a channel layer on the substrate in the second region; forming a deposition type interface layer on the channel layer and the substrate of the first region; removing impurities and defects contained in the deposited interface layer by exposing the deposited interface layer to a free radical oxidation process to convert the deposited interface layer into an oxidized interface layer; forming a high-k layer on the oxidized interface layer; forming a dipole interface at an interface between the high-k layer and the oxidized interface layer in the first region; forming a conductive layer on the high-k layer; patterning the conductive layer, the high-k layer, the dipole interface, and the oxidized interface layer to form a first gate stack on the substrate in the first region; as well as The conductive layer, the high-k layer and the oxidized interface layer are patterned to form a second gate stack on the substrate in the second region.

11. The method according to claim 10, wherein: The deposited interface layer includes deposited silicon oxide.

12. The method according to claim 10, wherein: Forming the deposition type interface layer on the channel layer and the substrate of the first region is performed by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

13. The method according to claim 10, wherein: Forming the dipole interface at the interface between the high-k layer and the oxidized interface layer in the first region comprises: forming a sacrificial layer containing a dipole-inducing substance on the high-k layer; exposing the sacrificial layer to a post-thermal treatment to diffuse the dipole-inducing species into an interface between the oxidized interface layer and the high-k layer; and The sacrificial layer is removed.

14. The method according to claim 13, wherein: The sacrificial layer includes lanthanum oxide.

15. The method according to claim 10, wherein: The channel layer includes crystalline silicon germanium.

16. The method according to claim 10, wherein: The first region includes an NMOSFET area, and the second region includes a PMOSFET area.

17. A method for manufacturing a semiconductor device, comprising: preparing a substrate including a first NMOSFET region and a second NMOSFET region; forming a thick deposition type interface layer on the substrate of the first NMOSFET region; removing impurities and defects contained in the thick deposited interfacial layer by exposing the thick deposited interfacial layer to a radical oxidation process to convert the thick deposited interfacial layer into a thick oxidized interfacial layer; forming a thin oxide type interface layer on the substrate of the second NMOSFET region; forming a high-k layer on the thick oxide type interface layer and the thin oxide type interface layer; forming a dipole interface at the interface between the oxidized interface layer and the high-k layer; forming a gate conductive layer on the high-k layer having the dipole interface; as well as A gate patterning process is performed to form gate stacks on the substrate of the first NMOSFET region and the second NMOSFET region, respectively.

18. The method according to claim 17, wherein: Forming the dipole interface at the interface between the oxidized interface layer and the high-k layer comprises: forming a sacrificial layer containing a dipole-inducing substance on the high-k layer; exposing the sacrificial layer to a post-heat treatment to diffuse the dipole-inducing species into an interface between the oxidized interface layer and the high-k layer; and The sacrificial layer is removed.

19. The method according to claim 17, wherein: Forming a thick deposition type interface layer on the substrate of the first NMOSFET region is performed by an atomic layer deposition (ALD) process or a chemical vapor deposition (CVD) process.

20. The method according to claim 17, wherein: The thick oxidized interface layer is thicker than the thin oxidized interface layer.

Images