Hafnium oxide film modification method and nonvolatile memory device

Abstract

The present invention provides a hafnium oxide film modification method and the like capable of crystallizing (modifying) the film into an orthorhombic-phase (o-phase)-dominant ferroelectric hafnium oxide (film). The hafnium oxide film modification method comprises: a film formation step (S11) of forming an amorphous hafnium oxide film on a silicon substrate; and a crystallization step (S12) of irradiating the amorphous hafnium oxide film with pulsed laser light having a wavelength in the ultraviolet range so that the amorphous hafnium oxide film is modified into an o-phase-dominant hafnium oxide film and aggregation of the o-phase-dominant hafnium oxide film is prevented.

Description

Hafnium oxide film modification method and nonvolatile memory device

[0001] The present disclosure relates to a method for modifying a hafnium oxide film and a non-volatile memory device.

[0002] A crystallization process is performed by annealing a silicon substrate with a hafnium oxide film. In addition to ordinary resistance heating, rapid thermal annealing (RTA) and the like are also used. An example of the annealing process is described in, for example, Patent Document 1.

[0003] Japanese Patent Application Laid-Open No. 2023-135612

[0004] However, hafnium oxide can exist in multiple crystalline phases, with the paraelectric m-phase (monoclinic phase) being the most stable crystalline phase. Therefore, there is a problem in that it is difficult to suppress the most stable m-phase by the above crystallization treatment and crystallize (modify) hafnium oxide (film) into a hafnium oxide film mainly composed of the ferroelectric o-phase (orthorhombic phase).

[0005] Other objects and novel features will become apparent from the description of this specification and the accompanying drawings.

[0006] A hafnium oxide film modifying method according to one embodiment includes a film forming step of forming an amorphous hafnium oxide film on a silicon substrate, and a crystallization step of irradiating the amorphous hafnium oxide film with pulsed laser light having a wavelength in the ultraviolet range so that the amorphous hafnium oxide film is modified into a hafnium oxide film mainly composed of o-phase and so that aggregation of the hafnium oxide film mainly composed of o-phase does not occur.

[0007] According to the embodiment, it is possible to provide a hafnium oxide film modification method that can crystallize (modify) hafnium oxide (film) mainly composed of ferroelectric o-phase, and a nonvolatile memory device.

[0008] 1 is a flowchart of a hafnium oxide film modifying method according to the present disclosure. FIG. 1 is a schematic diagram of a 1T1C (capacitor type) film. FIG. 2 is a schematic diagram of a 1T (transistor type) film. FIG. 3 is a schematic diagram showing an amorphous hafnium oxide film 20 being formed on a silicon substrate 10. FIG. 4 is a schematic diagram showing a state in which the hafnium oxide film 20 formed on the silicon substrate 10 is crystallized (modified). FIG. 5 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 1). FIG. 6 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 2). FIG. 7 is a schematic diagram showing a state in which the hafnium oxide film 20 formed on the silicon substrate 10 is crystallized (modified). FIG. 8 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 3). FIG. 9 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 4). FIG. 10 is an SEM image of a hafnium oxide film taken by SEM. FIG. 11 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 5). FIG. 12 is a graph showing an X-ray diffraction pattern (experimental results of Experiment 6). 1 is a SEM image of a hafnium oxide film captured by an SEM; 2 is a diagram showing a comparative example; 3 is a diagram showing an embodiment; 4 is a diagram showing a hafnium oxide film 20 modified by the hafnium oxide film modification method of the embodiment; Ｔ＋Ｋ 1 is a schematic configuration diagram of a nonvolatile memory device M to which the above-mentioned method is applied. 2 is a flowchart example of a manufacturing process of the nonvolatile memory device M.

[0009] Specific embodiments will be described in detail below with reference to the drawings. However, the present invention is not limited to the following embodiments. In addition, the following description and drawings have been simplified as appropriate for clarity of explanation.

[0010] <Outline of the Method for Modifying a Hafnium Oxide Film> First, an outline of the method for modifying a hafnium oxide film according to the present disclosure will be described.

[0011] FIG. 1 is a flow chart of the disclosed hafnium oxide film modification method.

[0012] 1 , the hafnium oxide film modifying method of the present disclosure includes a film formation step (step S11) of preparing a silicon substrate 10 (step S10), forming an amorphous hafnium oxide film 20 on the silicon substrate 10, and a crystallization step (step S12) of irradiating the amorphous hafnium oxide film 20 with pulsed laser light having a wavelength in the ultraviolet range (crystallizing the hafnium oxide film). In the crystallization step, the energy density, wavelength, pulse interval, and pulse width of the pulsed laser light are selected so that the amorphous hafnium oxide film 20 is modified into a hafnium oxide film mainly composed of o-phase, and so that aggregation of the hafnium oxide film mainly composed of o-phase (orthorhombic phase) does not occur.

[0013] According to the hafnium oxide film modification method of the present disclosure, the amorphous hafnium oxide film 20 formed on the silicon substrate 10 can be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, i.e., a hafnium oxide film mainly composed of the ferroelectric o-phase.

[0014] The fact that the hafnium oxide film has been crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, i.e., a hafnium oxide film mainly composed of ferroelectric o-phase, can be confirmed by evaluating the hafnium oxide film (crystalline phase) crystallized in step S12 using an X-ray diffraction device (step S13). ２ " represents the amorphous hafnium oxide film 20. In FIG. ２ " represents the crystallized hafnium oxide film 20. In FIG. ２ " represents a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0015] A hafnium oxide film in which the most stable m-phase is suppressed and which has been crystallized (modified) by the hafnium oxide film modification method of the present disclosure, i.e., a hafnium oxide film mainly composed of ferroelectric o-phase, can be applied to nonvolatile memory devices.

[0016] <Background to the Method for Modifying a Hafnium Oxide Film> Next, the background to the method for modifying a hafnium oxide film of the present disclosure will be described.

[0017] In the field of nonvolatile memory devices, PZT (lead zirconate titanate), SBT (strontium tantalate bismuthate), etc. have been used as ferroelectric materials and have been commercialized as 1T1C (capacitor type). Figure 2A is a schematic diagram of the configuration of 1T1C (capacitor type).

[0018] On the other hand, the present inventors have applied the above-mentioned conventional ferroelectric material to 1T (transistor type) for the purpose of increasing density, and have investigated the thinning of the above-mentioned conventional ferroelectric material. Figure 2B is a schematic diagram of the 1T (transistor type).

[0019] However, it is known that when the above-mentioned conventional ferroelectric materials are applied to 1T (transistor type), there is a limit to how thin they can be made, and there is also a problem with their compatibility with silicon substrates.

[0020] Therefore, the present inventors have developed hafnium oxide (HfO), which exhibits ferroelectricity in an ultra-thin film and can be formed directly on a silicon substrate, as a 1T (transistor type) ferroelectric material. ２ ) was considered.

[0021] However, since hafnium oxide can exist in multiple crystalline phases, with the paraelectric m-phase (monoclinic phase) being the most stable crystalline phase, the inventors have investigated a method for crystallizing hafnium oxide that suppresses the most stable m-phase.The inventors have conducted multiple experiments in the course of this investigation.

[0022] As a result, the present inventors discovered a method (the hafnium oxide film modification method of the present disclosure) for crystallizing (modifying) an amorphous hafnium oxide film formed on a silicon substrate into a hafnium oxide film in which the most stable m-phase is suppressed, i.e., into a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0023] Below, experiments 1 to 6 that the present inventors conducted to arrive at the hafnium oxide film modification method of the present disclosure will be described.

[0024] <Experiment 1> Experiment 1 was carried out under the conditions shown in Table 1 below.

[0025] Experiment 1 will be described in detail below.

[0026] <Film Formation in Experiment 1> FIG. 3 is a schematic diagram showing the process of forming an amorphous hafnium oxide film 20 on a silicon substrate 10. As shown in FIG.

[0027] In Experiment 1, an amorphous hafnium oxide film 20 was formed on a silicon substrate 10 by the thermal-ALD method using a film formation device (ALD device (demo machine) manufactured by JSW Afty). ALD is an abbreviation for Atomic Layer Deposition, and is also called atomic layer deposition. Hereinafter, the amorphous hafnium oxide film 20 formed on a silicon substrate 10 by the thermal-ALD method will be referred to as the hafnium oxide film 20. Ｔ It is written as follows.

[0028] Specifically, as shown in FIG. 3, one side of a silicon substrate 10 heated to 120° C. by a heater is brought into contact with flowing raw material gas, oxidizing agent, and inert gas, thereby forming an extremely thin hafnium oxide film 20 having a thickness of 10 nm on the silicon substrate 10. Ｔ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ３ and N as an inert gas. ２ was used.

[0029] <Crystallization (Modification) in Experiment 1> FIG. 4 is a schematic diagram showing the state in which the hafnium oxide film 20 formed on the silicon substrate 10 is crystallized (modified).

[0030] In Experiment 1, a pulsed laser irradiation device (excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) with a XeCl light source (wavelength 308 nm) was used to form a hafnium oxide film 20 on a silicon substrate 10. Ｔ Pulsed laser light with a wavelength in the ultraviolet range 30 Ｘ Pulse laser light 30 Ｘ The wavelength was 308 nm, the pulse interval was 10 ms (100 Hz), the pulse width (FWHM) was several tens of nanoseconds, and the number of shots was 200. Ｘ The energy density is 170mJ / cm 2 , 220 mJ / cm 2, 270 mJ / cm 2 , 400mJ / cm 2 The irradiation was performed 200 times (200 shots) for each.

[0031] Specifically, as shown in FIG. 4, a silicon substrate 10 (hafnium oxide film 20) is transported in the direction of arrow AR1 by a transport mechanism (not shown). Ｔ ) under a nitrogen atmosphere with pulsed laser light for 30 Ｘ Pulse laser light 30 Ｘ The width (width in the transport direction) of the silicon substrate 10 (hafnium oxide film 20 Ｔ ) are conveyed, the adjacent pulsed laser beams 30 Ｘ The silicon substrate 10 (hafnium oxide film 20) is irradiated with the hafnium oxide film 20 so that the irradiated areas overlap. Ｔ The transport speed of the film 10 was set to 0.2 mm / s. Ｘ The hafnium oxide film 20 irradiated with Ｔ The hafnium oxide film 20 Ｔ＋Ｘ In Fig. 4, the symbol A1 indicates a region that has been irradiated with the pulsed laser beam, and the symbol A2 indicates a region that has not been irradiated with the pulsed laser beam.

[0032] <Experimental Results of Experiment 1> Next, the experimental results (evaluation results) of Experiment 1 will be described.

[0033] <Evaluation by X-ray Diffraction Device> Fig. 5A is a graph showing the X-ray diffraction pattern (experimental results of Experiment 1). ＸＴ（１７０） has an energy density of 170 mJ / cm 2 The graph for the case of ＸＴ（２２０） has an energy density of 220 mJ / cm 2 The graph for the case of ＸＴ（２７０） has an energy density of 270 mJ / cm 2 The graph for the case of ＸＴ（４００） has an energy density of 400mJ / cm 2 The graph shows the case.

[0034] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) and was carried out by irradiating the sample with a pulsed laser beam 30 Ｘ Hafnium oxide film 20 irradiated with Ｔ＋Ｘ The crystalline phase was evaluated. X-ray diffractometers are also called XRD (X-ray diffraction). X-ray diffractometers have various measurement methods, and 2θ scan, an asymmetric reflection measurement suitable for evaluating thin films, was performed. Because thin films are evaluated at a small incident angle, it is also called thin film measurement or GI-XRD (Grazing Incidence XRD). The measurement condition for the X-ray diffractometer is an incident angle ω = 0.4 deg.

[0035] As a premise, when crystallization of a hafnium oxide film in which the most stable m-phase is suppressed progresses, that is, when an amorphous hafnium oxide film (exhibiting paraelectricity) formed on the silicon substrate 10 is modified into a hafnium oxide film (exhibiting ferroelectricity) mainly composed of o-phase, the X-ray diffraction pattern of the modified hafnium oxide film obtained by an X-ray diffractometer will have a peak near 2θ=30.35°.

[0036] Referring to FIG. 5A, in Experiment 1, graph G ＸＴ（１７０） , G ＸＴ（２２０） , G ＸＴ（２７０） , G ＸＴ（４００） It can be seen that none of the samples has a peak near 2θ=30.35°.

[0037] That is, in Experiment 1, a hafnium oxide film 20 formed on a silicon substrate 10 Ｔ It can be seen that the hafnium oxide film cannot be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0038] <Experiment 2> Experiment 2 was carried out under the conditions shown in Table 2 below.

[0039] Experiment 2 will be described in detail below.

[0040] <Film formation in Experiment 2> In Experiment 2, an amorphous hafnium oxide film 20 was formed on a silicon substrate 10 by the plasma-ALD method using a film formation device (ALD device (demo machine) manufactured by JSW Afty). Hereinafter, the amorphous hafnium oxide film 20 formed on a silicon substrate 10 by the plasma-ALD method will be referred to as the hafnium oxide film 20. Ｐ It is written as follows.

[0041] Specifically, as in Experiment 1 (see FIG. 3), one side of the silicon substrate 10 heated to 120° C. by a heater was brought into contact with flowing raw material gas, oxidizing agent, and inert gas, and an extremely thin hafnium oxide film 20 having a thickness of 10 nm was formed on the silicon substrate 10 (one side). Ｐ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ２ Plasma is used and N is used as an inert gas. ２ was used.

[0042] <Crystallization (modification) in Experiment 2> In Experiment 2, similarly to Experiment 1 (see FIG. 4), a pulsed laser irradiation device (excimer laser irradiation device (demo model) manufactured by JSW Actina Systems) with a XeCl light source (wavelength 308 nm) was used to crystallize a hafnium oxide film 20 formed on a silicon substrate 10. Ｐ Pulsed laser light with a wavelength in the ultraviolet range 30 Ｘ Pulse laser light 30 Ｘ In Experiment 2, the pulse laser light was 308 nm in wavelength, the pulse interval was 10 ms (100 Hz), the pulse width (FWHM) was several tens of nanoseconds, and the number of irradiations was 200 (200 shots). Ｘ The energy density is 170mJ / cm 2 , 220 mJ / cm 2 , 270 mJ / cm 2 , 400mJ / cm 2 The irradiation was performed 200 times (200 shots) for each.

[0043] Specifically, as in Experiment 1 (see FIG. 4), a silicon substrate 10 (hafnium oxide film 20 Ｐ) under a nitrogen atmosphere with pulsed laser light for 30 Ｘ Pulse laser light 30 Ｘ The width (width in the transport direction) of the silicon substrate 10 (hafnium oxide film 20 Ｐ ) are conveyed, the adjacent pulsed laser beams 30 Ｘ The silicon substrate 10 (hafnium oxide film 20) is irradiated with the hafnium oxide film 20 so that the irradiated areas overlap. Ｐ The transport speed of the film 10 was set to 0.2 mm / s. Ｘ The hafnium oxide film 20 irradiated with Ｐ The hafnium oxide film 20 Ｐ＋Ｘ It is written as follows.

[0044] <Experimental Results of Experiment 2> Next, the experimental results (evaluation results) of Experiment 2 will be described.

[0045] <Evaluation by X-ray Diffraction Device> FIG. 5B is a graph showing the X-ray diffraction pattern (experimental results of Experiment 2). ＸＰ（１７０） has an energy density of 170 mJ / cm 2 The graph for the case of ＸＰ（２２０） has an energy density of 220 mJ / cm 2 The graph for the case of ＸＰ（２７０） has an energy density of 270 mJ / cm 2 The graph for the case of ＸＰ（４００） has an energy density of 400mJ / cm 2 The graph shows the case.

[0046] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) in the same manner as in Experiment 1, and the pulsed laser beam 30 Ｘ Hafnium oxide film 20 irradiated with Ｐ＋Ｘ The measurement conditions for the X-ray diffraction device were an incident angle ω of 0.4 deg.

[0047] Referring to FIG. 5B, in Experiment 2, graph G ＸＰ（１７０） , G ＸＰ（２２０） , G ＸＰ（２７０） , G ＸＰ（４００）It can be seen that none of the samples has a peak near 2θ=30.35°.

[0048] That is, in Experiment 2, the hafnium oxide film 20 formed on the silicon substrate 10 Ｐ It can be seen that the hafnium oxide film cannot be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0049] <Discussion of Experiments 1 and 2> As described above, in neither Experiment 1 nor Experiment 2, a peak appeared near 2θ=30.35° (see FIGS. 5A and 5B). This is thought to be due to a lack of lattice vibrations that would promote crystallization of the amorphous hafnium oxide film. Therefore, in order to compensate for the lack of lattice vibrations, Experiments 3 to 6 were carried out using a pulsed laser irradiation device with a KrF light source, which has greater photon kinetic energy than a XeCl light source.

[0050] <Experiment 3> Experiment 3 was carried out under the conditions shown in Table 3 below.

[0051] Experiment 3 will be described in detail below.

[0052] <Film formation in Experiment 3> In Experiment 3, similarly to Experiment 1, a film formation apparatus (ALD apparatus (demo machine) manufactured by JSW Afty) was used to form an amorphous hafnium oxide film 20 on a silicon substrate 10 by thermal-ALD. Ｔ A film was formed.

[0053] Specifically, as in Experiment 1 (see FIG. 3), one side of a silicon substrate 10 heated to 120° C. by a heater was brought into contact with flowing source gas, oxidizing agent, and inert gas, and an extremely thin hafnium oxide film 20 having a thickness of 10 nm was formed on the silicon substrate 10. Ｔ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ３ and N as an inert gas. ２ was used.

[0054] <Crystallization (modification) in Experiment 3> FIG. 6 shows a hafnium oxide film 20 formed on a silicon substrate 10.Ｔ 1 is a schematic diagram showing the crystallization (modification) of

[0055] In Experiment 3, a pulsed laser irradiation device (excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) with a KrF light source (wavelength 248 nm) was used to form a hafnium oxide film 20 on a silicon substrate 10. Ｔ Pulsed laser light with a wavelength in the ultraviolet range 30 Ｋ Pulse laser light 30 Ｋ The wavelength was 248 nm, the pulse interval was 10 ms (100 Hz), the pulse width (FWHM) was several tens of ns, and the number of irradiations was one (1 shot). Ｋ The energy density is 300mJ / cm 2 , 350 mJ / cm 2 , 400mJ / cm 2 , 450mJ / cm 2 and each was irradiated once (1 shot).

[0056] Specifically, a pulsed laser irradiation device (an excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) was used, and as shown in FIG. 6, a silicon substrate 10 (hafnium oxide film 20) placed in a chamber 41 was moved in the X (+X, −X) direction and the Y (+Y, −Y) direction (directions perpendicular to the paper surface in FIG. 6) by a transport mechanism 40. Ｔ ) is supplied into the chamber 41 through anhydrous quartz 42 under a nitrogen atmosphere. Ｋ As shown in FIG. 6 , the chamber 41 is constructed by combining a plurality of chamber components 41 a, 41 b, 41 c, etc., and is sealed by seals 43 and 44 and a buffer material 45. The buffer material 45 is a Teflon (registered trademark) buffer material provided to prevent direct interference between the anhydrous quartz 42 and the chamber component 41 a (made of stainless steel). In FIG. 5 , reference numeral 47 denotes a valve provided on the nitrogen supply pipe 46, and reference numeral 49 denotes a valve provided on the nitrogen exhaust pipe 48. Nitrogen is supplied into the chamber 41 through the nitrogen supply pipe 46 and exhausted to the outside of the chamber 41 through the nitrogen exhaust pipe 48.

[0057] Pulsed laser light 30 Ｋ The size (cross-sectional size) of the pulsed laser beam 30 is 3 mm x 3 mm. Ｋ After each shot of irradiation, the chamber 41 was moved so that adjacent regions irradiated with the pulsed laser beam 30 would not overlap. Ｋ The hafnium oxide film 20 irradiated with Ｔ The hafnium oxide film 20 Ｔ＋Ｋ It is written as follows.

[0058] <Experimental Results of Experiment 3> Next, the experimental results (evaluation results) of Experiment 3 will be described.

[0059] <Evaluation by X-ray Diffraction Device> Fig. 7A is a graph showing the X-ray diffraction pattern (experimental results of Experiment 3). ＫＴ（３００） has an energy density of 300 mJ / cm 2 The graph for the case of ＫＴ（３５０） has an energy density of 350 mJ / cm 2 The graph for the case of ＫＴ（４００） has an energy density of 400mJ / cm 2 The graph for the case of ＫＴ（４５０） has an energy density of 450 mJ / cm 2 The graph shows the case.

[0060] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) in the same manner as in Experiment 1, and the pulsed laser beam 30 Ｋ Hafnium oxide film 20 irradiated with Ｔ＋Ｋ The measurement conditions for the X-ray diffraction device were an incident angle ω of 0.4 deg.

[0061] Referring to FIG. 7A, in Experiment 3, graph G ＫＴ（４００） , G ＫＴ（４５０） It can be seen that there is a peak near 2θ=30.35°.

[0062] That is, based on the evaluation by the X-ray diffraction device, the amorphous hafnium oxide film 20 formed by the thermal-ALD methodＴ In this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 400 mJ / cm 2 By selecting the above, the hafnium oxide film 20 Ｔ It can be seen that the hafnium oxide film can be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0063] <Evaluation by SEM> The upper part of FIG. 8 shows the hafnium oxide film 20 photographed by SEM. Ｔ＋Ｋ This is an SEM image of the above. SEM is an abbreviation for Scanning Electron Microscope, and is also called a scanning electron microscope.

[0064] Using a SEM (scanning electron microscope manufactured by Hitachi High-Tech), the hafnium oxide film 20 Ｔ＋Ｋ The surface shape was evaluated. Relatively bright areas in each SEM image represent agglomerated areas (agglomerated areas of hafnium oxide), and relatively dark areas represent areas where the hafnium oxide film is thin. Elemental identification of the bright areas revealed that agglomeration had progressed. Note that brightness was not uniform for each image taken, and the evaluation was based on the contrast within the SEM images.

[0065] Therefore, if the SEM image contains relatively bright areas and relatively dark areas, the hafnium oxide film 20 Ｔ＋Ｋ On the other hand, if the SEM image does not include a relatively bright portion and a relatively dark portion (for example, if the image includes only one of the portions), the hafnium oxide film 20 Ｔ＋Ｋ It can be evaluated that there is no aggregation.

[0066] Referring to the upper part of Figure 8, the energy density is 450 mJ / cm 2 In this case, the energy density is 450 mJ / cm 2 The SEM image of contains relatively bright and relatively dark areas (both are mixed). Therefore, the energy density is 450 mJ / cm 2 In this case, the hafnium oxide film 20 Ｔ＋Ｋ It can be evaluated that the hafnium oxide film 20 thus aggregated Ｔ＋ＫThe shape of the hafnium oxide film (mainly, the shape of the hafnium oxide film 20 after the film formation is completed) Ｔ Since the surface shape of the ferroelectric film is not maintained, the ferroelectric film is not suitable as a ferroelectric material for nonvolatile memory devices.

[0067] On the other hand, the energy density is 400mJ / cm 2 In this case, the energy density is 400 mJ / cm 2 The SEM image does not contain relatively bright and relatively dark areas. Therefore, the energy density is 400 mJ / cm 2 In this case, the hafnium oxide film 20 Ｔ＋Ｋ It can be evaluated that the hafnium oxide film 20 does not aggregate. Ｔ＋Ｋ The shape of the hafnium oxide film (mainly, the shape of the hafnium oxide film 20 after the film formation is completed) Ｔ Since the surface shape of the ferroelectric film is maintained, it is suitable as a ferroelectric material for nonvolatile memory devices.

[0068] As described above, according to Experiment 3, the amorphous hafnium oxide film 20 formed by the thermal-ALD method Ｔ In this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 400 mJ / cm 2 By selecting the hafnium oxide film 20 formed on the silicon substrate 10, Ｔ It can be seen that the hafnium oxide film can be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, i.e., a hafnium oxide film mainly composed of the ferroelectric o-phase, and further, aggregation of the hafnium oxide film does not occur.

[0069] <Experiment 4> Experiment 4 was carried out under the conditions shown in Table 4 below.

[0070] Experiment 4 will be described in detail below.

[0071] <Film formation in Experiment 4> In Experiment 4, similarly to Experiment 2, a film formation apparatus (ALD apparatus (demo machine) manufactured by JSW Afty) was used to form an amorphous hafnium oxide film 20 on a silicon substrate 10 by the plasma-ALD method. Ｐ A film was formed.

[0072] Specifically, as in Experiment 1 (see FIG. 3), one side of the silicon substrate 10 heated to 120° C. by a heater was brought into contact with flowing raw material gas, oxidizing agent, and inert gas, and an extremely thin hafnium oxide film 20 having a thickness of 10 nm was formed on the silicon substrate 10 (one side). Ｐ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ２ Plasma is used and N is used as an inert gas. ２ was used.

[0073] <Crystallization (modification) in Experiment 4> In Experiment 4, similarly to Experiment 3, a pulsed laser irradiation device (excimer laser irradiation device (demo model) manufactured by JSW Actina Systems) with a KrF light source (wavelength 248 nm) was used to crystallize the hafnium oxide film 20 formed on the silicon substrate 10. Ｐ Pulsed laser light with a wavelength in the ultraviolet range 30 Ｋ Pulse laser light 30 Ｋ In Experiment 4, the pulse laser light was 300 nm in wavelength, the pulse interval was 10 ms (100 Hz), the pulse width (FWHM) was several tens of nanoseconds, and the number of irradiations was one (1 shot). Ｋ The energy density is 300mJ / cm 2 , 350 mJ / cm 2 , 400mJ / cm 2 , 450mJ / cm 2 and each was irradiated once (1 shot).

[0074] Specifically, a pulsed laser irradiation device (excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) was used, and in the same manner as in Experiment 3 (see FIG. 6), a pulsed laser beam 30 was irradiated onto a silicon substrate 10 (amorphous hafnium oxide film 20) placed in a chamber 41 that was moved in the X (+X, −X) direction and the Y (+Y, −Y) direction (directions perpendicular to the paper surface in FIG. 6) by a transport mechanism 40 via anhydrous quartz 42 under a nitrogen atmosphere supplied into the chamber 41. Ｋ was irradiated.

[0075] Pulsed laser light 30 ＫThe size (cross-sectional size) of the pulsed laser beam 30 is 3 mm x 3 mm. Ｋ After each shot of irradiation, the chamber 41 was moved so that adjacent regions irradiated with the pulsed laser beam 30 would not overlap. Ｋ The hafnium oxide film 20 irradiated with Ｐ The hafnium oxide film 20 Ｐ＋Ｋ It is written as follows.

[0076] <Experimental Results of Experiment 4> Next, the experimental results (evaluation results) of Experiment 4 will be described.

[0077] <Evaluation by X-ray Diffraction Device> FIG. 7B is a graph showing the X-ray diffraction pattern (experimental results of Experiment 4). ＫＰ（３００） has an energy density of 300 mJ / cm 2 The graph for the case of ＫＰ（３５０） has an energy density of 350 mJ / cm 2 The graph for the case of ＫＰ（４００） has an energy density of 400mJ / cm 2 The graph for the case of ＫＰ（４５０） has an energy density of 450 mJ / cm 2 The graph shows the case.

[0078] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) in the same manner as in Experiment 1, and the pulsed laser beam 30 Ｋ Hafnium oxide film 20 irradiated with Ｐ＋Ｋ The measurement conditions for the X-ray diffraction device were an incident angle ω of 0.4 deg.

[0079] Referring to FIG. 7B, in Experiment 4, graph G ＫＰ（３５０） , G ＫＰ（４００） , G ＫＰ（４５０） It can be seen that there is a peak near 2θ=30.35°.

[0080] That is, based on the evaluation by the X-ray diffraction device, the amorphous hafnium oxide film 20 formed by the plasma-ALD method ＰIn this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 350 mJ / cm 2 By selecting the above, the hafnium oxide film 20 Ｐ It can be seen that the hafnium oxide film can be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0081] <Evaluation by SEM> The lower middle part of FIG. 8 shows an image of the hafnium oxide film 20 taken by SEM. Ｐ＋Ｋ 1 is an SEM image of the above.

[0082] As in Experiment 3, a hafnium oxide film 20 was measured using a SEM (scanning electron microscope manufactured by Hitachi High-Tech Corporation). Ｐ＋Ｋ The surface shape was evaluated.

[0083] Referring to the lower middle part of Figure 8, the energy density is 300 mJ / cm 2 , 350 mJ / cm 2 , 400mJ / cm 2 , 450mJ / cm 2 In this case, each SEM image contains relatively bright and relatively dark areas (both areas are mixed). Therefore, at an energy density of 300 mJ / cm 2 , 350 mJ / cm 2 , 400mJ / cm 2 , 450mJ / cm 2 In this case, the hafnium oxide film 20 Ｐ＋Ｋ It can be evaluated that the hafnium oxide film 20 thus aggregated Ｐ＋Ｋ The shape of the hafnium oxide film (mainly, the shape of the hafnium oxide film 20 after the film formation is completed) Ｐ Since the surface shape of the ferroelectric film is not maintained, the ferroelectric film is not suitable as a ferroelectric material for nonvolatile memory devices.

[0084] As described above, according to Experiment 4, the amorphous hafnium oxide film 20 formed by the plasma-ALD method Ｐ In this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 400 mJ / cm 2 By selecting the hafnium oxide film 20 formed on the silicon substrate 10, ＰIt can be seen that although it is possible to crystallize (modify) the hafnium oxide film into a hafnium oxide film in which the most stable m-phase is suppressed, i.e., a hafnium oxide film mainly composed of the ferroelectric o-phase, aggregation of the hafnium oxide film occurs.

[0085] <Considerations on Experiments 3 and 4> In both Experiments 3 and 4, a pulsed laser irradiation device of a KrF light source was used, and the number of irradiations was fixed at one (1 shot), and pulsed laser light 30 with varying energy densities was used. Ｋ This is an example in which irradiation was performed, and the peak is observed around 2θ=30.35° (see FIGS. 7A and 7B).

[0086] In thermal-ALD (Experiment 3), the m-phase was completely suppressed, whereas in plasma-ALD (Experiment 4), the m-phase was slightly crystallized (see Figures 7A and 7B). This difference in film formation technique is expected to be due to the difference in film density. Density tends to be higher in plasma-ALD than in thermal-ALD. Higher density means that lattice vibrations are more likely to propagate.

[0087] In thermal-ALD, the density is relatively low and lattice vibrations are difficult to propagate, so it is presumed that the structure did not shift to the low-symmetry m-phase. Conversely, in plasma-ALD, the density is relatively high and lattice vibrations are easy to propagate, so it is presumed that the structure shifted to the low-symmetry m-phase.

[0088] In addition, in thermal-ALD (Experiment 3), 400 mJ / cm was used to crystallize the material. 2 In contrast, plasma-ALD (Experiment 4) required an energy density of 350 mJ / cm for crystallization. 2 This is also thought to be due to the difference in density.

[0089] In thermal-ALD (Experiment 3), 400 mJ / cm 2 Crystallization proceeded and the m-phase was suppressed. However, as a result of SEM evaluation, 2 On the other hand, in plasma-ALD (Experiment 4), crystallization only progressed at 350 mJ / cm2 above, but 300 mJ / cm 2 Coagulation had already occurred.

[0090] Considering the above, the process that can prevent aggregation, that is, that can suppress the m-phase and crystallize while maintaining the shape of the hafnium oxide film (mainly the surface shape of the hafnium oxide film at the completion of film formation), is the thermal-ALD film formation method, which uses a pulsed laser with a wavelength of 248 nm in the ultraviolet range (KrF light source) at 400 mJ / cm. 2 It is clear that irradiation is required.

[0091] <Experiment 5> Experiment 5 was carried out under the conditions shown in Table 5 below.

[0092] Experiment 5 will be described in detail below.

[0093] <Film formation in Experiment 5> In Experiment 5, similarly to Experiment 1, a film formation apparatus (ALD apparatus (demo machine) manufactured by JSW Afty) was used to form an amorphous hafnium oxide film 20 on a silicon substrate 10 by thermal-ALD. Ｔ A film was formed.

[0094] Specifically, as in Experiment 1 (see FIG. 3), one side of a silicon substrate 10 heated to 120° C. by a heater was brought into contact with flowing source gas, oxidizing agent, and inert gas, and an extremely thin hafnium oxide film 20 having a thickness of 10 nm was formed on the silicon substrate 10. Ｔ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ３ and N as an inert gas. ２ was used.

[0095] <Crystallization (modification) in Experiment 5> In Experiment 5, similarly to Experiment 3, a pulsed laser irradiation device (excimer laser irradiation device (demo model) manufactured by JSW Actina Systems) with a KrF light source (wavelength 248 nm) was used to crystallize the hafnium oxide film 20 formed on the silicon substrate 10. Ｔ was irradiated with pulsed laser light having a wavelength in the ultraviolet region. ＫWavelength: 248 nm, pulse interval: 10 ms (100 Hz), pulse width (FWHM): several tens of ns, energy density: 400 mJ / cm 2 In experiment 5, pulsed laser light 30 Ｋ The number of times of irradiation was changed to 0 times (as-deposition), 1 time (1 shot), 2 times (2 shots), 20 times (20 shots), and 200 times (200 shots), and irradiation was performed for each number of times.

[0096] Specifically, a pulsed laser irradiation device (excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) was used, and similarly to Experiment 3 (see FIG. 6), a silicon substrate 10 (hafnium oxide film 20) placed in a chamber 41 was moved in the X (+X, −X) direction and the Y (+Y, −Y) direction (directions perpendicular to the paper surface in FIG. 6) by a transport mechanism 40. Ｔ ) is supplied into the chamber 41 through anhydrous quartz 42 under a nitrogen atmosphere. Ｋ was irradiated.

[0097] Pulsed laser light 30 Ｋ The size (cross-sectional size) of the pulsed laser beam 30 is 3 mm x 3 mm. Ｋ Each time the pulsed laser beam 30 is irradiated a predetermined number of times, the adjacent pulsed laser beams 30 Ｋ The chamber 41 was moved so that the irradiated areas of the two layers would not overlap. Ｋ The hafnium oxide film 20 irradiated with Ｔ The hafnium oxide film 20 Ｔ＋Ｋ It is written as follows.

[0098] <Experimental Results of Experiment 5> Next, the experimental results (evaluation results) of Experiment 5 will be described.

[0099] <Evaluation by X-ray Diffraction Device> Fig. 9A is a graph showing the X-ray diffraction pattern (experimental results of Experiment 5). ＫＴ（０ｓｈｏｔ） represents the graph when the number of irradiations (number of shots) is 0. ＫＴ（１ｓｈｏｔ） represents a graph when the number of shots is 1. ＫＴ（２ｓｈｏｔ）represents a graph when the number of irradiations (number of shots) is 2. ＫＴ（２０ｓｈｏｔ） represents the graph when the number of irradiations (number of shots) is 20. ＫＴ（２００ｓｈｏｔ） represents a graph in the case where the number of irradiations (number of shots) is 200.

[0100] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) in the same manner as in Experiment 1, and the pulsed laser beam 30 Ｋ Hafnium oxide film 20 irradiated with Ｔ＋Ｋ The measurement conditions for the X-ray diffraction device were an incident angle ω of 0.4 deg.

[0101] Referring to FIG. 9A, in Experiment 5, graph G ＫＴ（１ｓｈｏｔ） , G ＫＴ（２ｓｈｏｔ） , Graph G ＫＴ（２０ｓｈｏｔ） , G ＫＴ（２００ｓｈｏｔ） It can be seen that there is a peak near 2θ=30.35°.

[0102] That is, based on the evaluation by the X-ray diffraction device, the amorphous hafnium oxide film 20 formed by the thermal-ALD method Ｔ In this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 400 mJ / cm 2 By selecting the number of irradiations to be 1 or more, the hafnium oxide film 20 Ｔ It can be seen that the hafnium oxide film can be crystallized (modified) into a hafnium oxide film in which the most stable m-phase is suppressed, that is, a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0103] <Evaluation by SEM> The upper part of FIG. 10 shows the hafnium oxide film 20 photographed by SEM. Ｔ＋Ｋ 1 is an SEM image of the above.

[0104] As in Experiment 3, a hafnium oxide film 20 was measured using a SEM (scanning electron microscope manufactured by Hitachi High-Tech Corporation). Ｔ＋Ｋ The surface shape was evaluated.

[0105] Referring to the upper part of FIG. 10, the energy density is 400 mJ / cm 2In this case, the SEM images of 1 shot, 2 shots, 20 shots, and 200 shots do not contain relatively bright and relatively dark areas. Therefore, the energy density is 400 mJ / cm 2 In this case, even if the number of irradiations is 200, the hafnium oxide film 20 Ｔ＋Ｋ It can be evaluated that the hafnium oxide film 20 does not aggregate. Ｔ＋Ｋ The shape of the hafnium oxide film (mainly, the shape of the hafnium oxide film 20 after the film formation is completed) Ｔ Since the surface shape of the ferroelectric film is maintained, it is suitable as a ferroelectric material for nonvolatile memory devices.

[0106] <Experiment 6> Experiment 6 was carried out under the conditions shown in Table 6 below.

[0107] Experiment 6 will be described in detail below.

[0108] <Film formation in Experiment 6> In Experiment 6, similarly to Experiment 2, a film formation apparatus (ALD apparatus (demo machine) manufactured by JSW Afty) was used to form an amorphous hafnium oxide film 20 on a silicon substrate 10 by the plasma-ALD method. Ｐ A film was formed.

[0109] Specifically, as in Experiment 1 (see FIG. 3), one side of a silicon substrate 10 heated to 120° C. by a heater was brought into contact with flowing source gas, oxidizing agent, and inert gas, and an extremely thin hafnium oxide film 20 having a thickness of 10 nm was formed on the silicon substrate 10. Ｐ The source gas was TEMAH (Tetrakis(EthylMethylAmido)Hafnium), and the oxidizing agent was O ２ Plasma is used and N is used as an inert gas. ２ was used.

[0110] <Crystallization (modification) in Experiment 6> In Experiment 6, similarly to Experiment 5, a pulsed laser irradiation device (excimer laser irradiation device (demo model) manufactured by JSW Actina Systems) with a KrF light source (wavelength 248 nm) was used to crystallize the hafnium oxide film 20 formed on the silicon substrate 10. ＰThe sample was irradiated with a pulsed laser beam having a wavelength in the ultraviolet region. The wavelength of the pulsed laser beam was 248 nm, the pulse interval was 10 ms (100 Hz), the pulse width (FWHM) was several tens of nanoseconds, and the energy density was 400 mJ / cm. 2 In Experiment 6, pulsed laser light 30 Ｋ The number of times of irradiation was changed to 0 times (as-deposition), 1 time (1 shot), 2 times (2 shots), 20 times (20 shots), and 200 times (200 shots), and irradiation was performed for each number of times.

[0111] Specifically, a pulsed laser irradiation device (excimer laser irradiation device (demonstration model) manufactured by JSW Actina Systems) was used, and similarly to Experiment 5 (see FIG. 6), a silicon substrate 10 (hafnium oxide film 20) placed in a chamber 41 was moved in the X (+X, −X) direction and the Y (+Y, −Y) direction (directions perpendicular to the paper surface in FIG. 6) by a transport mechanism 40. Ｐ ) is supplied into the chamber 41 through anhydrous quartz 42 under a nitrogen atmosphere. Ｋ was irradiated.

[0112] Pulsed laser light 30 Ｋ The size (cross-sectional size) of the pulsed laser beam 30 is 3 mm x 3 mm. Ｋ Each time the pulsed laser beam 30 is irradiated a predetermined number of times, the adjacent pulsed laser beams 30 Ｋ The chamber 41 was moved so that the irradiated areas of the two layers would not overlap. Ｋ The hafnium oxide film 20 irradiated with Ｐ The hafnium oxide film 20 Ｐ＋Ｋ It is written as follows.

[0113] <Experimental Results of Experiment 6> Next, the experimental results (evaluation results) of Experiment 6 will be described.

[0114] <Evaluation by X-ray Diffraction Device> FIG. 9B is a graph showing the X-ray diffraction pattern (experimental results of Experiment 6). ＫＰ（０ｓｈｏｔ） represents the graph when the number of irradiations (number of shots) is 0. ＫＰ（１ｓｈｏｔ）represents a graph when the number of shots is 1. ＫＰ（２ｓｈｏｔ） represents a graph when the number of irradiations (number of shots) is 2. ＫＰ（２０ｓｈｏｔ） represents the graph when the number of irradiations (number of shots) is 20. ＫＰ（２００ｓｈｏｔ） represents a graph in the case where the number of irradiations (number of shots) is 200.

[0115] The evaluation by the X-ray diffraction device was carried out using an X-ray diffraction device (X-ray diffraction device manufactured by Rigaku Corporation) in the same manner as in Experiment 3, and the pulsed laser beam 30 Ｋ Hafnium oxide film 20 irradiated with Ｐ＋Ｋ The measurement conditions for the X-ray diffraction device were an incident angle ω of 0.4 deg.

[0116] Referring to FIG. 9B, in Experiment 6, graph G ＫＰ（１ｓｈｏｔ） , G ＫＰ（２ｓｈｏｔ） , Graph G ＫＰ（２０ｓｈｏｔ） It can be seen that there is a peak near 2θ=30.35°.

[0117] That is, based on the evaluation by the X-ray diffraction device, the amorphous hafnium oxide film 20 formed by the plasma-ALD method Ｐ In this case, the pulsed laser light 30 Ｋ Wavelength 248 nm, energy density 400 mJ / cm 2 By selecting the number of irradiations from 1 to 20, the amorphous hafnium oxide film 20 formed on the silicon substrate 10 is converted into a hafnium oxide film 20 in which the most stable m-phase is suppressed. Ｐ That is, it can be seen that the hafnium oxide film can be crystallized (modified) into a hafnium oxide film mainly composed of a ferroelectric o-phase.

[0118] <Evaluation by SEM> The lower middle part of FIG. 10 shows an image of the hafnium oxide film 20 taken by SEM. Ｐ＋Ｋ 1 is an SEM image of the above.

[0119] As in Experiment 3, a scanning electron microscope (SEM) manufactured by Hitachi High-Tech Corporation was used to measure the hafnium oxide film 20 irradiated with the pulsed laser light, which is the experimental result of Experiment 5. Ｐ＋Ｋ The surface shape was evaluated.

[0120] Referring to the lower middle part of Figure 10, the energy density is 400 mJ / cm 2 In this case, the SEM images of 1 shot, 2 shots, 20 shots, and 200 shots each contain relatively bright and relatively dark areas. Therefore, the energy density is 400 mJ / cm 2 In this case, the hafnium oxide film 20 Ｐ＋Ｋ It can be evaluated that the hafnium oxide film 20 thus aggregated Ｐ＋Ｋ The shape of the hafnium oxide film (mainly, the shape of the hafnium oxide film 20 after the film formation is completed) Ｐ Since the surface shape of the ferroelectric film is not maintained, the ferroelectric film is not suitable as a ferroelectric material for nonvolatile memory devices.

[0121] <Considerations on Experiments 5 and 6> In both Experiments 5 and 6, a pulsed laser irradiation device with a KrF light source was used, and the energy density was 400 mJ / cm 2 The pulsed laser light 30 was fixed at 3000 Hz and the number of shots was changed. Ｋ This is an example in which irradiation was performed with 1000 kJ / cm2, and the peak was observed around 2θ=30.35° (see FIGS. 9A and 9B).

[0122] In the case of as-deposition, no crystallization occurred (amorphous) in either thermal-ALD (Experiment 5) or plasma-ALD (Experiment 6).

[0123] In the case of 1-shot, crystallization progressed in both thermal-ALD (Experiment 5) and plasma-ALD (Experiment 6), and the o-phase became the dominant phase (mainly o-phase). In the case of 1-shot, a slight m-phase peak was observed in plasma-ALD (Experiment 6).

[0124] In thermal-ALD (Experiment 5), the o-phase was maintained even when the number of shots was increased to 2 or more.

[0125] On the other hand, in plasma-ALD (Experiment 6), when the number of irradiations was increased to more than two shots, the o-phase peak decreased with increasing number of irradiations, and instead, the m-phase became the main phase (mainly m-phase). This dependence on the number of irradiations in plasma-ALD indicates that the process is different between injecting photons into an amorphous hafnium oxide film and injecting photons into a one-shot hafnium oxide film with a predominantly o-phase structure. Generating lattice vibrations in an amorphous hafnium oxide film did not shift the structure to the m-phase, and the film remained in the predominant o-phase state. However, generating lattice vibrations in an o-phase hafnium oxide film shifts the structure to the m-phase, presumably replacing the o-phase with the m-phase. It is believed that the o-phase does not disappear and the m-phase appears.

[0126] The reason why the crystallization of the m-phase did not progress in thermal-ALD (Experiment 5) is thought to be that the lower density compared to plasma-ALD (Experiment 6) makes it difficult for lattice vibrations to progress.

[0127] Regarding the confirmation of the number of irradiations and surface shape, with plasma-ALD, aggregation (aggregation of hafnium oxide) had already occurred after one shot, and aggregation continued even with an increased number of irradiations. On the other hand, with thermal-ALD, aggregation did not progress even with an increased number of irradiations.

[0128] <Conditions for Modification> Considering the above Experiments 1 to 6, the conditions for modifying an amorphous hafnium oxide film (exhibiting paraelectricity) formed on the silicon substrate 10 into a hafnium oxide film (exhibiting ferroelectricity) mainly composed of o-phase are as follows:

[0129] (Condition 1) An amorphous hafnium oxide film 20 is formed on a silicon substrate by a thermal ALD method. Ｔ The hafnium oxide film 20 Ｔ The thickness of the film may be set appropriately depending on the degree of densification required.

[0130] (Condition 2) Hafnium oxide film 20 formed according to the above condition 1 Ｔand irradiating the amorphous hafnium oxide film 20 with pulsed laser light having a wavelength in the ultraviolet region. Ｔ Hafnium oxide film 20 mainly composed of o-phase Ｔ＋Ｋ and the hafnium oxide film 20 Ｔ＋Ｋ The energy density, wavelength, pulse interval, and pulse width of the pulsed laser beam should be selected so that the aggregation of the particles does not occur. The energy density, wavelength, pulse interval, and pulse width of this pulsed laser beam should be set to 400 mJ / cm. ２ The wavelength is 248 nm (for example, a KrF light source), the pulse interval is 10 ms, and the pulse width is several tens of nanoseconds, but other values ​​are also acceptable. Note that if a CW laser beam is used instead of a pulsed laser beam, it is considered that the o-phase will not be the main component, so it is essential to use a pulsed laser beam.

[0131] The energy density, wavelength, pulse interval, and pulse width of the pulse laser light are selected based on the hafnium oxide film 20 Ｔ＋Ｋ The X-ray diffraction pattern obtained by the X-ray diffractometer has a peak at about 2θ=30.35°, and the hafnium oxide film 20 Ｔ＋Ｋ The SEM image of the hafnium oxide film 20 Ｔ＋Ｋ The pulsed laser beam 30 is focused so that the image does not include relatively bright and relatively dark areas that are evaluated as aggregations of the particles. Ｋ It is desirable to select the energy density, wavelength, pulse interval, and pulse width.

[0132] <Effects Compared to Comparative Examples> Next, the effects of the embodiment will be described in comparison with comparative examples.

[0133] FIG. 11A is a diagram illustrating a comparative example, and FIG. 11B is a diagram illustrating an embodiment.

[0134] In Comparative Example 1, the entire silicon substrate on which the hafnium oxide film is formed is continuously annealed. When the entire silicon substrate on which the hafnium oxide film is formed is continuously annealed, the hafnium oxide film changes into a number of crystalline phases including the m-phase and the o-phase, as shown in FIG. 11A. The "o, m, etc." in FIG. 11A represents this. A hafnium oxide film with such a mixture of a number of crystalline phases is unsuitable as a ferroelectric material for a nonvolatile memory device.

[0135] On the other hand, according to the hafnium oxide film modifying method of the embodiment (see the above conditions 1 and 2, etc.), the amorphous hafnium oxide film 20 formed on the silicon substrate 10 Ｔ The hafnium oxide film 20 in which the most stable m phase is suppressed Ｔ＋Ｋ , that is, a ferroelectric hafnium oxide film 20 mainly composed of o-phase Ｔ＋Ｋ (Modification) to the hafnium oxide film 20 Ｔ＋Ｋ This hafnium oxide film 20 Ｔ＋Ｋ is suitable as a ferroelectric material for nonvolatile memory devices.

[0136] <Nonvolatile Memory Device> Next, the hafnium oxide film 20 modified by the hafnium oxide film modifying method of the above embodiment will be described. Ｔ＋Ｋ An example of a nonvolatile memory device to which the above is applied will be described.

[0137] FIG. 12 shows a hafnium oxide film 20 modified by the hafnium oxide film modifying method of the embodiment. Ｔ＋Ｋ 1 is a schematic diagram of a nonvolatile memory device M to which the present invention is applied.

[0138] As shown in FIG. 12, the nonvolatile memory device M comprises a silicon substrate 10 and at least one ferroelectric gate transistor 50 formed on the silicon substrate.

[0139] The ferroelectric gate transistor 50 includes a source region 51 formed on one surface of the silicon substrate 10 in a state exposed from the surface, a drain region 52 formed on one surface of the silicon substrate 10 at a location away from the source region 51 in a state exposed from the surface, and a hafnium oxide film 53 (hafnium oxide film 20) mainly made of o-phase formed on one surface of the silicon substrate 10 in a state covering the source region 51 and the drain region 52. Ｔ＋Ｋ ), a gate electrode 54 formed on the hafnium oxide film 53, a source electrode 55 electrically connected to the source region 51 via a first contact hole H1 formed in the hafnium oxide film 53 corresponding to the source region 51, and a drain electrode 56 inserted into a second contact hole H2 formed in the hafnium oxide film 53 corresponding to the drain region 52 and electrically connected to the drain region 52.

[0140] Next, an example of a manufacturing process for the nonvolatile memory device M having the above configuration will be described.

[0141] FIG. 13 is a flowchart showing an example of a manufacturing process for the nonvolatile memory device M.

[0142] First, a silicon substrate 10 is prepared (step S10) and cleaned (step S11). The silicon substrate 10 may be, for example, a p-type silicon substrate with a plane index of (100).

[0143] Next, a source region 51 and a drain region 52 are formed in the silicon substrate 10 (step S12). The source region 51 and the drain region 52 are formed on one surface of the silicon substrate 10 by, for example, n-doping, in a state where they are exposed from that surface. Note that an annealing treatment may be performed at this stage, but the annealing treatment may also be omitted.

[0144] Next, an amorphous hafnium oxide film 20 is formed on the silicon substrate 10 by a thermal-ALD method using a film forming apparatus. Ｔ Specifically, an amorphous hafnium oxide film 20 is formed on one surface of the silicon substrate 10 while covering the source region 51 and the drain region 52 (step S13). Ｔ An amorphous hafnium oxide film 20 is formed. ＴThe film thickness is, for example, 10 nm.

[0145] Next, a pulsed laser irradiation device of an ultraviolet light source (for example, a KrF light source) is used to form an amorphous hafnium oxide film 20 Ｔ This crystallizes the amorphous hafnium oxide film 20 (step S14). Ｔ The hafnium oxide film 53 (hafnium oxide film 20) is mainly composed of o-phase. Ｔ＋Ｋ ) is modified to

[0146] Next, contact holes H1 and H2 are formed in the hafnium oxide film 53 mainly made of o-phase (step S15).

[0147] Thereafter, an electrode material is deposited on the hafnium oxide film 53 mainly composed of the o-phase (step S16), and a source electrode, a drain electrode, and a gate electrode are formed (step S17).

[0148] As described above, the hafnium oxide film 20 modified by the hafnium oxide film modifying method of the above embodiment is obtained. Ｔ＋Ｋ It is possible to manufacture a nonvolatile memory device to which the above-described hafnium oxide film modifying method of the above-described embodiment is applied not only to the above-described nonvolatile memory device M, but also to nonvolatile memory devices having other configurations including a hafnium oxide film.

[0149] As described above, according to this embodiment, it is possible to crystallize (modify) hafnium oxide (film) mainly composed of ferroelectric o-phase.

[0150] The invention made by the inventor has been specifically described above based on the embodiments, but it goes without saying that the present invention is not limited to the embodiments already described, and various modifications are possible within the scope of the gist of the invention.

[0151] This application claims priority based on Japanese Patent Application No. 2024-007913, filed January 23, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0152] 10... Silicon substrate 20... Hafnium oxide film 30... Pulsed laser light 40... Transfer mechanism 41... Chamber 42... Anhydrous quartz 50... Ferroelectric gate transistor 51... Source region 52... Drain region 53... Hafnium oxide film 54... Gate electrode 55... Source electrode 56... Drain electrode A1... Pulsed laser light irradiated region A2... Pulsed laser light unirradiated region H1... First contact hole H2... Second contact hole M... Nonvolatile memory device

Claims

1. A method for modifying a hafnium oxide film, comprising: a film forming step of forming an amorphous hafnium oxide film on a silicon substrate; and a crystallization step of irradiating the amorphous hafnium oxide film with pulsed laser light having a wavelength in the ultraviolet region so that the amorphous hafnium oxide film is modified into a hafnium oxide film mainly composed of the o-phase and aggregation of the hafnium oxide film mainly composed of the o-phase does not occur.

2. The amorphous hafnium oxide film is formed by the Thermal ALD method, and the energy density of the pulsed laser light is 400 mJ / cm ２ The method for modifying a hafnium oxide film according to claim 1, wherein the above conditions are satisfied.

3. The method for modifying a hafnium oxide film according to claim 2, wherein the wavelength of the pulsed laser light is 248 nm, the pulse interval is 10 ms, and the pulse width is several tens of ns.

4. The method for modifying a hafnium oxide film according to claim 3, wherein the pulsed laser light is emitted from a pulsed laser irradiation device of a KrF light source.

5. The X-ray diffraction pattern of the hafnium oxide film modified by the crystallization step obtained by an X-ray diffractometer has a peak around 2θ = 30.35°, and the SEM image of the hafnium oxide film modified by the crystallization step obtained by a scanning electron microscope does not include relatively bright portions and relatively dark portions evaluated as aggregation of the hafnium oxide film. The method for modifying a hafnium oxide film according to claim 1.

6. A non-volatile memory device comprising a silicon substrate and a hafnium oxide film formed on the silicon substrate, wherein the hafnium oxide film is formed by: a film forming step of forming an amorphous hafnium oxide film on the silicon substrate; and a crystallization step of irradiating the amorphous hafnium oxide film with pulsed laser light having a wavelength in the ultraviolet region so that the amorphous hafnium oxide film is modified into a hafnium oxide film mainly composed of the o-phase and aggregation of the hafnium oxide film mainly composed of the o-phase does not occur. A non-volatile memory device having a hafnium oxide film mainly composed of the o-phase.

7. A source region formed on one side of the silicon substrate and exposed from that side, a drain region formed on one side of the silicon substrate and at a location away from the source region and exposed from that side, a gate electrode formed on the hafnium oxide film, a source electrode electrically connected to the source region through a first contact hole formed in the hafnium oxide film corresponding to the source region, and a drain electrode electrically connected to the drain region through a second contact hole formed in the hafnium oxide film corresponding to the drain region. The film formation step forms an amorphous hafnium oxide film in a state covering the source region and the drain region on one side of the silicon substrate. The crystallization step irradiates the amorphous hafnium oxide film with pulsed laser light having a wavelength in the ultraviolet region so that the amorphous hafnium oxide film is modified into a hafnium oxide film mainly composed of the o-phase and aggregation of the hafnium oxide film mainly composed of the o-phase does not occur. The non-volatile memory device according to claim 6.

Images