Ferroelectric device and method for manufacturing same

Abstract

Provided is a ferroelectric device (1) comprising a single-crystal silicon substrate (10), a titanium nitride layer (20), a chromium vanadium layer (30), a platinum electrode layer (40), and a ferroelectric thin film (50), the ferroelectric device (1) being such that the titanium nitride layer (20), the chromium vanadium layer (30), the platinum electrode layer (40), and the ferroelectric thin film (50) are laminated in the stated order on the single-crystal silicon substrate (10).

Description

Ferroelectric device and method of manufacturing same

[0001] This application claims priority to Japanese Patent Application No. 2024-030822, filed on February 29, 2024, the contents of which are incorporated herein by reference.

[0002] Piezoelectric MEMS (Micro Electro Mechanical Systems) elements (hereinafter referred to as "piezoelectric MEMS devices") are characterized by their simple structure and the promise of high energy conversion efficiency. In piezoelectric MEMS devices, it is desirable to provide a single-crystal thin film of oxide ferroelectric material on a single-crystal silicon substrate. This requires a bottom electrode grown as a single crystal.

[0003] Many efforts have been made to form a single-crystal thin film of an oxide ferroelectric on a single-crystal silicon substrate. ３ A method has been developed in which a buffer layer made of zirconium oxide is grown by a technique called MBE (Molecular Beam Epitaxy). A method has also been developed in which zirconium oxide is used as the buffer layer. This method has been realized for mass production (see, for example, Patent Document 1). A method has also been developed in which titanium nitride (TiN) is used as the buffer layer (see, for example, Patent Document 2).

[0004] Patent No. 7011760 Publication JP-A-11-297966

[0005] However, the method of growing a buffer layer by a technique called MBE has not been established as a mass production method. Furthermore, the method of Patent Document 1 requires the combined use of vacuum deposition and sputtering, which poses a problem in terms of cost reduction. Furthermore, the method of Patent Document 2 poses a problem of peeling of the oxide ferroelectric from the single-crystal silicon substrate when applied to the manufacture of piezoelectric MEMS devices with a thickness of several micrometers.

[0006] The present invention has been made in view of the above circumstances, and aims to provide a ferroelectric device that can be efficiently manufactured at reduced cost and that can prevent the oxide ferroelectric from peeling off from the single crystal silicon substrate, and a method for manufacturing the same.

[0007] The present invention has the following aspects. [1] A ferroelectric device comprising a silicon substrate, a titanium nitride layer, a chromium vanadium layer, a platinum electrode layer, and a ferroelectric thin film, wherein the titanium nitride layer, the chromium vanadium layer, the platinum electrode layer, and the ferroelectric thin film are stacked in this order on the silicon substrate. [2] The ferroelectric device according to [1], wherein the titanium nitride layer, the platinum electrode layer, and the ferroelectric thin film have the same crystal orientation as the silicon substrate. [3] The ferroelectric device according to [1] or [2], wherein the chromium to vanadium content ratio (chromium / vanadium) in the chromium vanadium layer is 1 to 3 in molar ratio. [4] The ferroelectric device according to any one of [1] to [3], wherein the ferroelectric thin film has a thickness of 850 nm or more. [5] The ferroelectric device according to any one of [1] to [4], wherein the ferroelectric thin film is a single-crystal ferroelectric thin film or a polycrystalline ferroelectric thin film. [6] The ferroelectric device according to any one of [1] to [4], wherein the ferroelectric thin film is a single-crystal ferroelectric thin film. [7] A method for manufacturing a ferroelectric device, comprising the steps of: forming a titanium nitride layer on a single-crystal silicon substrate by radio-frequency magnetron sputtering; forming a chromium vanadium layer on the titanium nitride layer by radio-frequency magnetron sputtering; forming a platinum electrode layer on the chromium vanadium layer by radio-frequency magnetron sputtering; and forming a ferroelectric thin film on the platinum electrode layer by radio-frequency magnetron sputtering. [8] The method for manufacturing a ferroelectric device according to [7], wherein the ferroelectric device is the ferroelectric device according to any one of [1] to [6].

[0008] According to the present invention, it is possible to provide a ferroelectric device and a method for manufacturing the same that can be efficiently manufactured at low cost and can suppress peeling of an oxide ferroelectric from a single crystal silicon substrate. Note that, in the present invention, even if chromium and vanadium in the chromium-vanadium layer (adhesion layer) diffuse, the effect of suppressing peeling can be obtained.

[0009] 1 is a cross-sectional view showing a ferroelectric device according to one embodiment of the present invention. 2 is a diagram showing the results of analyzing the crystal structure of each layer (film) of the ferroelectric device by X-ray diffraction in Experimental Example 1. 3 is a diagram showing the crystal structure of (100) BiFeO calculated from X-ray 2θ-ω diffraction in Experimental Example 1. ３ 1 is a mapping diagram of the Lotgering factor (LF) at 25 points within the surface of a single crystal silicon substrate at a peak. 2 is a scanning electron microscope image of a cross section in the thickness direction of a ferroelectric device in Experimental Example 1, in which the film formation temperature of the single crystal ferroelectric thin film is 410°C. 3 is an X-ray diffraction pattern of a stacked body consisting of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 4 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 1. 5 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 1. 6 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 7 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 8 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 9 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 10 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 11 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 12 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 13 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 14 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 15 is an X-ray diffraction pattern of a single crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer in Experimental Example 1. 1 ４ N, Mo ２ N, CrN, Cr ２1 is a diagram showing the calculation results of the formation free energy of N, VN, NbN, TaN, YN, AlN, TiN, ZrN, or HfN.

[0033] FIG. 1 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 2.

[0034] FIG. 2 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 2.

[0035] FIG. 3 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 3.

[0036] FIG. 4 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 4.

[0037] FIG. 5 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 5.

[0038] FIG. 6 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 6.

[0039] FIG. 7 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 4.

[0039] FIG. 8 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 5.

[0039] 10 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 9. FIG. 11 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 9. FIG. 12 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 10. FIG. 13 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 10. FIG. 14 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 10. FIG. 15 is a graph showing the results of measuring the frequency dependence of the relative dielectric constant of a ferroelectric device in Experimental Example 11. FIG. 16 is a graph showing the results of measuring the polarization and electric field characteristics of a ferroelectric device in Experimental Example 11. FIG. 17 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 12. FIG. 18 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 12. FIG. 19 is an X-ray diffraction pattern of a ferroelectric device in Experimental Example 13. FIG. 19 is a graph showing the results of measuring the thickness of a single-crystal ferroelectric thin film of a ferroelectric device obtained in Experimental Example 12. FIG. 15 is a diagram showing the results of measuring the degree of orientation of the single-crystal ferroelectric thin film of the ferroelectric device obtained in Experimental Example 12, in Experimental Example 15.16 is a diagram showing the results of observing the surface morphology of the single-crystal ferroelectric thin film of the ferroelectric device obtained in Experimental Example 12 in Experimental Example 16. 17 is a diagram showing the results of measuring the Fe / B ratio of the ferroelectric device obtained in Experimental Example 12 in Experimental Example 16. 18 is an X-ray diffraction pattern of the single-crystal ferroelectric thin film. 19 is a diagram showing the results of observing the surface morphology of the single-crystal ferroelectric thin film of the ferroelectric device obtained in Experimental Example 18 in Experimental Example 19. 19 is an X-ray diffraction pattern of the ferroelectric device obtained in Experimental Example 18 in Experimental Example 19. 19 is a diagram showing the results of measuring the dielectric properties of the ferroelectric device obtained in Experimental Example 18 in Experimental Example 20.

[0010] The following describes embodiments of a ferroelectric device and a method for manufacturing the same according to the present invention. Note that the embodiments are specifically described to provide a better understanding of the gist of the invention, and do not limit the present invention unless otherwise specified.

[0011] [Ferroelectric Device] Fig. 1 is a cross-sectional schematic diagram showing a ferroelectric device according to one embodiment of the present invention. As shown in Fig. 1, the ferroelectric device 1 of this embodiment includes a single-crystal silicon substrate 10, a titanium nitride layer 20, a chromium vanadium layer 30, a platinum electrode layer 40, and a ferroelectric thin film 50. In the ferroelectric device 1, the titanium nitride layer 20, the chromium vanadium layer 30, the platinum electrode layer 40, and the ferroelectric thin film 50 are stacked in this order on the single-crystal silicon substrate 10.

[0012] "Single-crystal silicon substrate" Examples of the single-crystal silicon substrate 10 include substrates for semiconductor devices having a (001) plane and used in electronic devices such as personal computers and mobile terminals, silicon-on-insulator substrates, etc. The thickness of the single-crystal silicon substrate 10 is not particularly limited, but is, for example, 200 μm to 650 μm.

[0013] "Titanium Nitride Layer" The titanium nitride (TiN) layer 20 is a buffer layer formed on one main surface 10a of the single crystal silicon substrate 10. The titanium nitride layer 20 is stable and does not react with the single crystal silicon substrate 10 or metals. The titanium nitride layer 20 also inherits the crystalline information of the single crystal silicon substrate 10 and has the same crystal orientation as the single crystal silicon substrate 10. Furthermore, because silicon and platinum are alloyed at 200°C, providing the titanium nitride layer 20 between the single crystal silicon substrate 10 and the platinum electrode layer 40 can prevent this alloying.

[0014] The thickness of the titanium nitride layer 20 is not particularly limited, but is preferably, for example, 60 nm to 100 nm.

[0015] "Chromium Vanadium Layer" The chromium vanadium (CrV) layer 30 is formed on one main surface 20a of the titanium nitride layer 20, and is an adhesion layer for improving the adhesion between the titanium nitride layer 20 and the platinum electrode layer 40. Because the bonding strength between titanium nitride and platinum is weak, by interposing the chromium vanadium layer 30 between the titanium nitride layer 20 and the platinum electrode layer 40, the adhesion between the titanium nitride layer 20 and the platinum electrode layer 40 is improved. Furthermore, the chromium vanadium layer 30 inherits the crystal information of the single crystal silicon substrate 10 and has the same crystal orientation as the single crystal silicon substrate 10.

[0016] The molar ratio of chromium (Cr) to vanadium (V) (Cr / V) in the chromium vanadium layer 30 is preferably 1 to 3, and most preferably 2. If the molar ratio is less than the lower limit, the chromium vanadium layer 30 is likely to peel off. On the other hand, if the molar ratio exceeds the upper limit, chromium vanadium crystals do not grow.

[0017] The thickness of the chromium vanadium layer 30 is not particularly limited, but is preferably, for example, 6 nm to 10 nm.

[0018] "Platinum Electrode Layer" The platinum (Pt) electrode layer 40 is an electrode layer formed on one main surface 30a of the chromium vanadium layer 30. The platinum electrode layer 40 inherits the crystal information of the single crystal silicon substrate 10 and has the same crystal orientation as the single crystal silicon substrate 10. The thickness of the platinum electrode layer 40 is not particularly limited, but is preferably 20 nm or more, and more preferably 50 nm to 100 nm.

[0019] "Ferroelectric thin film" The ferroelectric thin film 50 may be either a single-crystal ferroelectric thin film or a polycrystalline ferroelectric thin film, but from the viewpoint of the characteristics of the ferroelectric device, a single-crystal ferroelectric thin film is preferable. Hereinafter, a case where the ferroelectric thin film 50 is a single-crystal ferroelectric thin film will be described as an example. The single-crystal ferroelectric thin film 50 is a ferroelectric film formed on one main surface 40a of the platinum electrode layer 40. The single-crystal ferroelectric thin film 50 is made of a compound having a perovskite structure. Examples of compounds having a perovskite structure include BiFeO ３ (hereinafter, also referred to as "BFO"), bismuth iron oxide represented by Pb(Zr,Ti)O ３ Examples include lead zirconate titanate represented by the formula: BiFeO ３ has a perovskite structure and exhibits excellent ferroelectricity, weak ferromagnetism, and ferroelasticity in a thin film state. The single-crystal ferroelectric thin film 50 inherits the crystal information of the single-crystal silicon substrate 10 and has the same crystal orientation as the single-crystal silicon substrate 10. The thickness of the single-crystal ferroelectric thin film 50 is not particularly limited, but is preferably 2000 nm or less, and more preferably 100 nm to 1000 nm.

[0020] According to the ferroelectric device 1 of this embodiment, the chromium vanadium layer 30 is interposed between the titanium nitride layer 20 and the platinum electrode layer 40, resulting in excellent adhesion between the titanium nitride layer 20 and the platinum electrode layer 40. As a result, peeling of the single-crystal ferroelectric thin film 50 from the single-crystal silicon substrate 10 is suppressed, and a ferroelectric device 1 with excellent characteristics is obtained.

[0021] The ferroelectric device 1 of this embodiment can replace all devices equipped with an oxide film having a perovskite structure. In the ferroelectric device 1 of this embodiment, the titanium nitride layer 20, the platinum electrode layer 40, and the single-crystal ferroelectric thin film 50 have the same crystal orientation as the single-crystal silicon substrate 10, so that the dielectric constant is low and the sensitivity is high. Therefore, the ferroelectric device 1 of this embodiment is suitable for use in high-temperature superconductors, magnetic materials, ferroelectric memories, piezoelectric elements, etc.

[0022] [Method for Manufacturing Ferroelectric Device] A method for manufacturing a ferroelectric device according to one embodiment of the present invention includes the steps of forming a titanium nitride layer on a single crystal silicon substrate by radio frequency (RF) magnetron sputtering (hereinafter referred to as the "first step"), forming a chromium vanadium layer on the titanium nitride layer by radio frequency (RF) magnetron sputtering (hereinafter referred to as the "second step"), forming a platinum electrode layer on the chromium vanadium layer by radio frequency magnetron sputtering (hereinafter referred to as the "third step"), and forming a ferroelectric thin film on the platinum electrode layer by radio frequency magnetron sputtering (hereinafter referred to as the "fourth step"). Note that in the method for manufacturing a ferroelectric device according to this embodiment, the first to fourth steps can be performed using a single magnetron sputtering apparatus.

[0023] The method for manufacturing the ferroelectric device of this embodiment will be described below with reference to FIG.

[0024] "First Step" In the first step, a titanium nitride layer 20 is formed on one main surface 10a of a single crystal silicon substrate 10 by radio frequency magnetron sputtering using a magnetron sputtering apparatus. The magnetron sputtering apparatus may, for example, be one equipped with a vacuum chamber, a magnetron sputtering source disposed within the vacuum chamber, and a stage on which a substrate (single crystal silicon substrate) to be deposited can be placed and rotated.

[0025] Here, we explain the radio-frequency magnetron sputtering method. First, plasma is generated in a vacuum. Plasma is an unstable state in which positively charged gas ion atoms (cations) and negatively charged free electrons fly freely. The magnetic force of a magnet placed behind the target traps the cations at high density within a magnetic field. The cations moving around in this magnetic field collide one after another with the target surface, which has a negative potential. The sputtered target particles are then ejected and fly toward the sample. In other words, by utilizing the power of the magnet, plasma can be generated efficiently with little power, and the cations gather in areas with a strong magnetic field and repeatedly collide, resulting in efficient film formation.

[0026] In the first step, titanium nitride (TiN) is used as the target.

[0027] In the first step, the power of the high frequency signal (RF power) is preferably set to 40W to 60W, and more preferably to 50W.

[0028] In the first step, the film formation temperature is preferably set to 300°C to 700°C, and more preferably 400°C to 500°C.

[0029] In the first step, the pressure in the vacuum chamber of the magnetron sputtering device is preferably set to 0.5 Pa to 1 Pa, and more preferably to 0.6 Pa.

[0030] In the first step, the flow rate of the rare gas such as argon flowing into the vacuum chamber of the magnetron sputtering device is preferably 10 sccm to 20 sccm, and more preferably 20 sccm.

[0031] In the first step, the film formation time is preferably 30 to 120 minutes, and more preferably 60 minutes.

[0032] In the first step, it is preferable to remove the native oxide film on one main surface 10a of the single crystal silicon substrate 10 using hydrogen fluoride (HF), so that the titanium nitride layer 20 formed on one main surface 10a of the single crystal silicon substrate 10 can inherit the crystallinity of the single crystal silicon substrate 10.

[0033] "Second Step" In the second step, the chromium vanadium layer 30 is formed on one main surface 20a of the titanium nitride layer 20 by high-frequency magnetron sputtering using a magnetron sputtering device.

[0034] In the second step, a chromium vanadium (CrV) target is used, and the molar ratio of chromium (Cr) to vanadium (V) (Cr / V) in the chromium vanadium target is preferably 1 to 3, and most preferably 2.

[0035] In the second step, the power of the high frequency signal (RF power) is preferably set to 20W to 50W, and more preferably to 30W.

[0036] In the second step, the film formation temperature is preferably set to 300°C to 500°C, and more preferably 400°C.

[0037] In the second step, the pressure in the vacuum chamber of the magnetron sputtering device is preferably set to 0.6 Pa to 2 Pa, and more preferably to 1 Pa.

[0038] In the second step, the flow rate of the rare gas such as argon flowing into the vacuum chamber of the magnetron sputtering device is preferably 10 sccm to 20 sccm, and more preferably 20 sccm.

[0039] In the second step, the film formation time is preferably 1 to 3 minutes, and more preferably 2 minutes.

[0040] "Third Step" In the third step, the platinum electrode layer 40 is formed on one main surface 30a of the chromium vanadium layer 30 by high-frequency magnetron sputtering using a magnetron sputtering device.

[0041] In the third step, platinum (Pt) is used as the target.

[0042] In the third step, the power of the high frequency signal (RF power) is preferably set to 20W to 40W, and more preferably to 30W.

[0043] In the third step, the film formation temperature is preferably set to 300°C to 500°C, and more preferably 400°C.

[0044] In the third step, the pressure in the vacuum chamber of the magnetron sputtering device is preferably set to 0.6 Pa to 2 Pa, and more preferably to 1 Pa.

[0045] In the third step, the flow rate of the rare gas such as argon flowing into the vacuum chamber of the magnetron sputtering device is preferably 10 sccm to 20 sccm, and more preferably 20 sccm.

[0046] In the third step, the film formation time is preferably 20 to 40 minutes, and more preferably 30 minutes.

[0047] "Fourth Step" In the fourth step, the ferroelectric thin film 50 is formed on one main surface 40a of the platinum electrode layer 40 by radio frequency magnetron sputtering using a magnetron sputtering device.

[0048] In the fourth step, a compound having a perovskite structure is used as the target.

[0049] In the fourth step, the power of the high frequency signal (RF power) is preferably set to 20W to 40W, and more preferably to 30W.

[0050] In the fourth step, the film formation temperature is preferably set to 300°C to 500°C, and more preferably 400°C.

[0051] In the fourth step, the pressure in the vacuum chamber of the magnetron sputtering device is preferably set to 0.6 Pa to 2 Pa, and more preferably to 1 Pa.

[0052] In the fourth step, the flow rate of the rare gas such as argon flowing into the vacuum chamber of the magnetron sputtering device is preferably 10 sccm to 20 sccm, and more preferably 20 sccm.

[0053] In the fourth step, the film formation time is preferably 20 to 40 minutes, and more preferably 30 minutes.

[0054] According to the method for manufacturing a ferroelectric device of this embodiment, the first to fourth steps can be performed using a single magnetron sputtering apparatus, allowing each layer (film) to be formed efficiently. Furthermore, since the platinum electrode layer 40 is formed on the titanium nitride layer 20 via the chromium vanadium layer 30, excellent adhesion between the titanium nitride layer 20 and the platinum electrode layer 40 is achieved. As a result, peeling of the ferroelectric thin film 50 from the single-crystal silicon substrate 10 is suppressed, resulting in a ferroelectric device 1 with excellent durability. Furthermore, a thick ferroelectric thin film 50 can be formed on the electrode layer 40.

[0055] Although the embodiments of the present invention have been described in detail above with reference to the drawings, the embodiments are merely illustrative of the present invention. Therefore, the present invention is not limited to the configurations of the embodiments, and design changes within the scope of the present invention are also included within the scope of the present invention. Furthermore, for example, when multiple configurations are included in each embodiment, possible combinations of these configurations are also included, even if not specifically stated. Furthermore, when multiple examples or variations are disclosed in an embodiment as the present invention, possible combinations of configurations across these are also included, even if not specifically stated. Furthermore, configurations depicted in the drawings are also included, even if not specifically stated. Furthermore, when the term "etc." is used, it is used to mean that equivalents are included.

[0056] The present invention will be explained in more detail below by way of experimental examples, but the present invention is not limited to the following experimental examples.

[0057] [Experimental Example 1] A titanium nitride layer, a platinum electrode layer, and a single-crystal ferroelectric thin film composed of BFO were formed in this order on a single-crystal silicon substrate by radio-frequency magnetron sputtering using a magnetron sputtering device, to obtain a ferroelectric device. The native oxide film on one main surface of the single-crystal silicon substrate was removed in advance using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The conditions for forming the platinum electrode layer were: power of the high frequency signal was 30 W, film formation temperature was 500°C, pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, film formation time was 30 minutes, and the thickness of the platinum electrode layer was 100 nm. The conditions for forming the single-crystal ferroelectric thin film were: power of the high frequency signal was 30 W, film formation temperature was 600°C, pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２ The flow rate of the ferroelectric gas was 1 sccm, the film formation time was 420 minutes, and the thickness of the single-crystal ferroelectric thin film was 850 nm. The crystal structure of each layer (film) of the obtained ferroelectric device was analyzed by X-ray diffraction using an X-ray diffraction apparatus (trade name: X'PERT, manufactured by Spectris). The results are shown in Figure 2. From the results shown in Figure 2, it was confirmed that the titanium nitride layer, platinum electrode layer, and single-crystal ferroelectric thin film had the same crystal orientation as the single-crystal silicon substrate. Figure 3 shows the (100) BiFeO calculated from X-ray 2θ-ω diffraction. ３The Lotgering factor (LF) mapping diagram at 25 points on the single-crystal silicon substrate surface is shown. At higher temperatures, the film became slightly cloudy, and a significant decrease in orientation was observed. On the other hand, at lower temperatures, the peak intensity of the (100) orientation increased in the region with a small Bi supply, but the growth of non-oriented grains was small in the region with a slightly large Bi supply around 480°C. (100)BiFeO ３ An epitaxial thin film was obtained. These results indicate that Bi re-evaporation increases rapidly above approximately 460°C, making it difficult to obtain a thin film with a stoichiometric composition, and that the Bi supply rate affects (100)-oriented growth at temperatures below approximately 440°C. However, when a platinum layer was formed directly on the titanium nitride layer, peeling of the ferroelectric thin film was confirmed. Figure 4 shows a scanning electron microscope image of a cross section in the thickness direction of a ferroelectric device formed at a deposition temperature of 460°C. As shown in Figure 4, peeling of the single-crystal ferroelectric thin film from the platinum electrode layer was observed. Furthermore, the crystalline structure of each layer (film) of a laminate consisting of a single-crystal silicon substrate, a titanium nitride layer, and a platinum electrode layer was analyzed by X-ray diffraction using an X'PERT X-ray diffractometer (Spectris Corporation). The results are shown in Figure 5. Crystal peaks of the titanium nitride layer and the platinum electrode layer were confirmed from the results shown in Figure 5. Furthermore, for a ferroelectric device consisting of a single-crystal silicon substrate, a titanium nitride layer, a platinum electrode layer, and a single-crystal ferroelectric thin film, the crystal structure of each layer (film) was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figure 6. From the results shown in Figure 6, it was confirmed that the titanium nitride peak disappeared as the film formation temperature of the single-crystal ferroelectric thin film increased from 370°C to 450°C. This means that the titanium nitride was decomposed, and it was presumed that the nitrogen generated by the decomposition was the cause of the peeling.

[0058] Fe ４ N, Mo ２ N, CrN, Cr ２The calculation results of the free energy of formation of N, VN, NbN, TaN, YN, AlN, TiN, ZrN, or HfN are shown in Figure 7. The results shown in Figure 7 indicate that the decomposition of titanium nitride can be suppressed by inserting Mo, Cr, V, or Ta between platinum and titanium nitride.

[0059] [Experimental Example 2] A titanium nitride layer, an adhesion layer made of tantalum (Ta), and a platinum electrode layer were formed in this order on a single-crystal silicon substrate by radio-frequency magnetron sputtering using a magnetron sputtering device, to obtain a lower electrode. A native oxide film on one main surface of the single-crystal silicon substrate was removed in advance using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The pressure was 1 Pa, the deposition time was 45 minutes, and the thickness of the titanium nitride layer was 60 nm. For the formation of the adhesion layer, the high-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 10 minutes, and the thickness of the tantalum (Ta) adhesion layer was 10 nm. For the formation of the platinum electrode layer, the high-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 30 minutes, and the thickness of the platinum electrode layer was 100 nm. The crystal structure of each layer (film) of the resulting lower electrode was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figures 8 and 9. From the results shown in FIGS. 8 and 9, it was confirmed that the titanium nitride layer, the adhesion layer, and the platinum electrode layer did not have the same crystal orientation as the single crystal silicon substrate.

[0060] [Experimental Example 3] A titanium nitride layer, an adhesion layer made of molybdenum (Mo), and a platinum electrode layer were formed in this order on a single-crystal silicon substrate by radio-frequency magnetron sputtering using a magnetron sputtering device, to obtain a lower electrode. A native oxide film on one main surface of the single-crystal silicon substrate was previously removed using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The pressure was 1 Pa, the deposition time was 45 minutes, and the thickness of the titanium nitride layer was 60 nm. For the formation of the adhesion layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 10 minutes, and the thickness of the adhesion layer made of molybdenum (Mo) was 10 nm. For the formation of the platinum electrode layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 30 minutes, and the thickness of the platinum electrode layer was 100 nm. The crystal structure of each layer (film) of the resulting lower electrode was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figures 10 and 11. From the results shown in FIGS. 10 and 11, it was confirmed that the titanium nitride layer, the adhesion layer, and the platinum electrode layer had the same crystal orientation as the single crystal silicon substrate.

[0061] [Experimental Example 4] A titanium nitride layer, an adhesion layer made of vanadium (V), and a platinum electrode layer were formed in this order on a single-crystal silicon substrate by radio-frequency magnetron sputtering using a magnetron sputtering device, to obtain a lower electrode. A native oxide film on one main surface of the single-crystal silicon substrate was previously removed using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The pressure was 1 Pa, the deposition time was 45 minutes, and the thickness of the titanium nitride layer was 60 nm. For the formation of the adhesion layer, the high-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 10 minutes, and the thickness of the vanadium (V) adhesion layer was 10 nm. For the formation of the platinum electrode layer, the high-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 30 minutes, and the thickness of the platinum electrode layer was 100 nm. The crystal structure of each layer (film) of the resulting lower electrode was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figures 12 and 13. From the results shown in FIGS. 12 and 13, it was confirmed that the titanium nitride layer, the adhesion layer, and the platinum electrode layer had the same crystal orientation as the single crystal silicon substrate.

[0062] [Experimental Example 5] A titanium nitride layer, an adhesion layer made of chromium (Cr), and a platinum electrode layer were formed in this order on a single-crystal silicon substrate by radio-frequency magnetron sputtering using a magnetron sputtering device, to obtain a lower electrode. A native oxide film on one main surface of the single-crystal silicon substrate was previously removed using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The pressure was 1 Pa, the deposition time was 45 minutes, and the thickness of the titanium nitride layer was 60 nm. For the formation of the adhesion layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 10 minutes, and the thickness of the chromium (Cr) adhesion layer was 10 nm. For the formation of the platinum electrode layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time was 30 minutes, and the thickness of the platinum electrode layer was 100 nm. The crystal structure of each layer (film) of the resulting lower electrode was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figures 14 and 15. From the results shown in FIGS. 14 and 15, it was confirmed that the titanium nitride layer, the adhesion layer, and the platinum electrode layer had the same crystal orientation as the single crystal silicon substrate.

[0063] [Experimental Example 6] A single-crystal ferroelectric thin film made of BFO was formed on the lower electrode obtained in Experimental Example 3. In forming the single-crystal ferroelectric thin film, the power of the high-frequency signal was 30 W, the film formation temperature was 420°C to 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２The flow rate of the ferroelectric thin film was 1 sccm, the film formation time was 420 minutes, and the thickness of the single-crystal ferroelectric thin film was 850 nm. The surface of the ferroelectric thin film was photographed with a digital camera and observed. The results are shown in FIG. 16. From the results shown in FIG. 16, it can be seen that the surface of the ferroelectric thin film was rough and had peeled off in some areas, but there was no mirror-finished area.

[0064] [Experimental Example 7] A single-crystal ferroelectric thin film made of BFO was formed on the lower electrode obtained in Experimental Example 4. In forming the single-crystal ferroelectric thin film, the power of the high-frequency signal was 30 W, the film formation temperature was 420°C to 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２ The flow rate of the gas containing ferroelectric gas was 1 sccm, the film formation time was 420 minutes, and the thickness of the single-crystal ferroelectric thin film was 850 nm. The surface of the single-crystal ferroelectric thin film was photographed with a digital camera and observed. The results are shown in FIG. 17. From the results shown in FIG. 17, it was found that only a portion of the surface of the single-crystal ferroelectric thin film formed a mirror finish, and the single-crystal ferroelectric thin film had peeled off in a large area.

[0065] [Experimental Example 8] A single-crystal ferroelectric thin film made of BFO was formed on the lower electrode obtained in Experimental Example 5. In forming the single-crystal ferroelectric thin film, the power of the high-frequency signal was 30 W, the film formation temperature was 420°C to 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２ The flow rate of the gas containing ferroelectric gas was 1 sccm, the film formation time was 420 minutes, and the thickness of the single-crystal ferroelectric thin film was 850 nm. The surface of the single-crystal ferroelectric thin film was photographed with a digital camera and observed. The results are shown in FIG. 18. From the results shown in FIG. 18, it can be seen that the surface of the single-crystal ferroelectric thin film was rough but formed a mirror surface, and there were no areas where the single-crystal ferroelectric thin film had peeled off.

[0066] [Experimental Example 9] For the ferroelectric device using vanadium for the adhesion layer obtained in Experimental Example 7, the crystal structure of each layer (film) was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figures 19 and 20. It was confirmed from the results shown in Figures 19 and 20 that a single-crystal ferroelectric thin film was obtained.

[0067] [Experimental Example 10] For the ferroelectric device using chromium for the adhesion layer obtained in Experimental Example 8, the crystal structure of each layer (film) was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris Corporation). The results are shown in Figures 21 and 22. The results shown in Figures 21 and 22 confirmed that the crystallinity of the single-crystal ferroelectric thin film was low.

[0068] [Experimental Example 11] The frequency dependence of the dielectric constant was measured for the ferroelectric device without an adhesion layer obtained in Experimental Example 1, the ferroelectric device with an adhesion layer made of vanadium (V) obtained in Experimental Example 7, and the ferroelectric device with an adhesion layer made of chromium (Cr) obtained in Experimental Example 8. The frequency dependence of the dielectric constant of the ferroelectric devices was measured using an LCR meter (product name: 4284A, manufactured by Keysight Corporation). The results are shown in Figure 23. Furthermore, the polarization and electric field characteristics of the ferroelectric device without an adhesion layer obtained in Experimental Example 1, the ferroelectric device with an adhesion layer made of vanadium (V) obtained in Experimental Example 7, and the ferroelectric device with an adhesion layer made of chromium (Cr) obtained in Experimental Example 8 were measured. The polarization and electric field characteristics of the ferroelectric devices were measured using a Sawyer-Tower circuit. The results are shown in Figure 24. From the results shown in Figures 23 and 24, it was confirmed that the ferroelectric device obtained in Experimental Example 1 and not provided with an adhesion layer, the ferroelectric device obtained in Experimental Example 7 and provided with an adhesion layer made of vanadium (V), and the ferroelectric device obtained in Experimental Example 8 and provided with an adhesion layer made of chromium (Cr) had ferroelectricity.

[0069] [Experimental Example 12] A ferroelectric device was obtained by forming a titanium nitride layer, an adhesion layer made of chromium (Cr) and vanadium (V), a platinum electrode layer, and a single-crystal ferroelectric thin film made of BFO on a single-crystal silicon substrate in this order by radio-frequency magnetron sputtering using a magnetron sputtering device. The native oxide film on one main surface of the single-crystal silicon substrate was removed in advance using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: radio-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, argon flow rate into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N ２ ) partial pressure to 1 × 10 －４ The conditions were: 1 Pa, deposition time 45 minutes, and the thickness of the titanium nitride layer 60 nm. For the formation of the adhesion layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time 10 minutes, and the thickness of the adhesion layer composed of chromium (Cr) and vanadium (V) was 10 nm. The molar ratio of chromium (Cr) to vanadium (V) was varied between 1:1 and 1:3. For the formation of the platinum electrode layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time 30 minutes, and the thickness of the platinum electrode layer 100 nm. In the formation of the ferroelectric thin film, the power of the high frequency signal was 30 W, the film formation temperature was 420°C to 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２The flow rate of the ferroelectric thin film was 1 sccm, the film formation time was 420 minutes, and the thickness of the single-crystal ferroelectric thin film was 1,100 nm. The surface of the ferroelectric thin film of the obtained ferroelectric device was photographed with a digital camera and observed. The results are shown in FIG. 25. From the results shown in FIG. 25, it was found that although the surface of the ferroelectric thin film had some rough regions, most of it was a mirror-like region. Furthermore, the crystal structure of each layer (film) of the obtained ferroelectric device was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in FIGS. 26 and 27. The results shown in FIGS. 26 and 27 confirmed that the single-crystal ferroelectric thin film had high crystallinity.

[0070] [Experimental Example 13] In the ferroelectric device obtained in Experimental Example 12, the surface of the single-crystal ferroelectric thin film in regions where the molar ratio of chromium (Cr) to vanadium (V) was different between 1:1 and 1:3 was photographed using an atomic force microscope (product name: Nanonavi, manufactured by Seiko Instruments Inc.), and the surface of the single-crystal ferroelectric thin film was observed. The results are shown in Figure 28. From the results shown in Figure 28, it was confirmed that when the molar ratio of chromium (Cr) to vanadium (V) was 2:1, most of the surface of the single-crystal ferroelectric thin film became a mirror finish.

[0071] [Experimental Example 14] The thickness of the single-crystal ferroelectric thin film of the ferroelectric device obtained in Experimental Example 12 was measured using an optical film thickness meter (Filmetrix, F20). The results are shown in Figure 29. From the results shown in Figure 29, it was confirmed that the thickness of the single-crystal ferroelectric thin film was approximately 1100 nm to 1200 nm when the film formation temperature was 420°C to 460°C. Furthermore, for the ferroelectric device obtained in Experimental Example 12, the crystal structure of each layer (film) was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figure 30. From the results shown in Figure 30, it was confirmed that the crystallinity of the single-crystal ferroelectric thin film was high when the film formation temperature was 420°C to 460°C.

[0072] [Experimental Example 15] For the ferroelectric device obtained in Experimental Example 12, the degree of orientation of the single-crystal ferroelectric thin film was measured using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris Corporation). The results are shown in Fig. 31. From the results shown in Fig. 31, it was confirmed that the degree of orientation of the single-crystal ferroelectric thin film increased when the film formation temperature was 420°C to 460°C.

[0073] [Experimental Example 16] The surface of the ferroelectric device obtained in Experimental Example 12 was photographed using an atomic force microscope (product name: Nanonavi, manufactured by Seiko Instruments Inc.) to observe the surface morphology of the single-crystal ferroelectric thin film. The results are shown in FIG. 32. From the results shown in FIG. 32, it was confirmed that the surface morphology of the single-crystal ferroelectric thin film becomes flat when the film formation temperature is 420°C to 460°C. In addition, the Fe / B ratio of the ferroelectric device obtained in Experimental Example 15 was measured using an electron probe microanalyzer (manufactured by JEOL Ltd.). The results are shown in FIG. 33. From the results shown in FIG. 33, it was confirmed that the Fe / B ratio in the single-crystal ferroelectric thin film becomes close to the stoichiometric composition when the film formation temperature is 420°C to 460°C.

[0074] [Experimental Example 17] The ferroelectric characteristics of the ferroelectric device obtained in Experimental Example 12 were measured using a Sawyer-Tower circuit. The results are shown in Fig. 34. From the results shown in Fig. 34, it was confirmed that the ferroelectric characteristics were good when the film formation temperature of the single-crystal ferroelectric thin film was 420°C to 460°C.

[0075] [Experimental Example 18] A titanium nitride layer, an adhesion layer made of chromium (Cr) and vanadium (V), a platinum electrode layer, and a LaNiO ３ A seed layer made of SiO2 and a single-crystal ferroelectric thin film made of BFO were formed in this order to obtain a ferroelectric device. The native oxide film on one main surface of the single-crystal silicon substrate was removed in advance using hydrogen fluoride (HF). The titanium nitride layer was formed under the following conditions: high-frequency signal power of 50 W, film formation temperature of 700°C, pressure in the vacuum chamber of the magnetron sputtering device of 0.6 Pa, flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device of 20 sccm, and nitrogen (N２ ) partial pressure to 1 × 10 －４ The conditions were: 1 Pa, deposition time 45 minutes, and the thickness of the titanium nitride layer 60 nm. For the formation of the adhesion layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 1 sccm, the deposition time 10 minutes, and the thickness of the adhesion layer composed of chromium (Cr) and vanadium (V) was 10 nm. The molar ratio of chromium (Cr) to vanadium (V) was 2:1. For the formation of the platinum electrode layer, the radio-frequency signal power was 30 W, the deposition temperature was 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, the deposition time 30 minutes, and the thickness of the platinum electrode layer 100 nm. In forming the seed layer, the power of the high frequency signal was 30 W, the film formation temperature was 500° C., the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２ ) flow rate was 1 sccm, the film formation time was 30 minutes, and LaNiO ３ The thickness of the adhesive layer consisting of was set to 30 nm. In forming the single crystal ferroelectric thin film, the power of the high frequency signal was 30 W, the film formation temperature was 420°C to 500°C, the pressure in the vacuum chamber of the magnetron sputtering device was 1 Pa, the flow rate of argon flowing into the vacuum chamber of the magnetron sputtering device was 20 sccm, and the flow rate of oxygen (O ２ The flow rate of the ferroelectric thin film was set to 1 sccm, the film formation time was set to 420 minutes, and the thickness of the single-crystal ferroelectric thin film was set to 1,100 nm. For the obtained ferroelectric device, the crystal structure of the single-crystal ferroelectric thin film was analyzed by X-ray diffraction using an X-ray diffraction apparatus (trade name: X'PERT, manufactured by Spectris). The results are shown in FIG. 35. From the results shown in FIG. 35, it was confirmed that the crystallinity of the single-crystal ferroelectric thin film was high when the film formation temperature was 420°C to 500°C.

[0076] [Experimental Example 19] The surface of the ferroelectric device obtained in Experimental Example 18 was photographed with a digital camera to observe the surface morphology of the single-crystal ferroelectric thin film. The results are shown in Figure 36. From the results shown in Figure 36, it was confirmed that the surface of the single-crystal ferroelectric thin film became a mirror finish when the film formation temperature was 420°C to 500°C. Furthermore, compared with Figure 17 and other figures, it is clear that peeling of the single-crystal ferroelectric thin film was suppressed. Furthermore, the crystal structure of each layer (film) of the obtained ferroelectric device was analyzed by X-ray diffraction using an X-ray diffractometer (trade name: X'PERT, manufactured by Spectris). The results are shown in Figure 37. From the results shown in Figure 37, it was confirmed that all layers were epitaxially grown.

[0077] [Experimental Example 20] The dielectric properties of the ferroelectric device obtained in Experimental Example 18 were measured using an impedance analyzer (product name: IM3570, manufactured by Hioki Corporation). The results are shown in Fig. 38. From the results shown in Fig. 38, it was confirmed that the dielectric properties were good when the film formation temperature of the single-crystal ferroelectric thin film was 420°C to 500°C.

[0078] REFERENCE SIGNS LIST 1 Ferroelectric device 10 Single crystal silicon substrate 20 Titanium nitride layer 30 Chromium vanadium layer 40 Platinum electrode layer 50 Ferroelectric thin film

Claims

1. A ferroelectric device comprising a single crystal silicon substrate, a titanium nitride layer, a chromium vanadium layer, a platinum electrode layer, and a ferroelectric thin film, wherein the titanium nitride layer, the chromium vanadium layer, the platinum electrode layer, and the ferroelectric thin film are stacked in this order on the single crystal silicon substrate.

2. A ferroelectric device as recited in claim 1, wherein said titanium nitride layer, said platinum electrode layer, and said ferroelectric thin film have the same crystal orientation as said single crystal silicon substrate.

3. The ferroelectric device according to claim 1, wherein the chromium / vanadium content ratio in said chromium vanadium layer is 1 to 3 in molar ratio.

4. The ferroelectric device according to claim 1, wherein the thickness of said ferroelectric thin film is 850 nm or more.

5. The ferroelectric device of claim 1, wherein the ferroelectric thin film is a single crystal ferroelectric thin film or a polycrystalline ferroelectric thin film.

6. The ferroelectric device of claim 1, wherein the ferroelectric thin film is a single crystal ferroelectric thin film.

7. A method for manufacturing a ferroelectric device, comprising the steps of: forming a titanium nitride layer on a single crystal silicon substrate by radio frequency magnetron sputtering; forming a chromium vanadium layer on the titanium nitride layer by radio frequency magnetron sputtering; forming a platinum electrode layer on the chromium vanadium layer by radio frequency magnetron sputtering; and forming a ferroelectric thin film on the platinum electrode layer by radio frequency magnetron sputtering.

Images