STACKED STRUCTURE AND METHOD OF MANUFACTURING SAME, AND SEMICONDUCTOR DEVICE

Abstract

[Problem]: The problem of the present invention is to provide a stacked structure excellent in stability of atomic arrangement, a method of manufacturing same, and a semiconductor device using the stacked structure. [Solution]: The stacked structure of the present invention is characterized in that it has an alloy layer A having germanium and tellurium as a main component and an alloy layer B having tellurium and either of antimony or bismuth as a main component, and at least either of the alloy layer A or the alloy layer B contains at least either of sulfur or selenium as a chalcogen atom.

Claims

1. A stacked structure comprising: an alloy layer A having germanium and tellurium as a main component thereof, and an alloy layer B having tellurium and either of antimony or bismuth as a main component thereof, wherein: at least either of the alloy layer A or the alloy layer B contains at least either of sulfur or selenium as a chalcogen atom.

2. The stacked structure according to claim 1, wherein the chalcogen atom is contained in the alloy layer A.

3. The stacked structure according to claim 1, wherein a chalcogen content of the alloy layer A is from 0.05 at % to 10.0 at %.

4. The stacked structure according to claim 1, having the alloy layer A and the alloy layer B stacked alternately in repetition.

5. The stacked structure according to claim 1, wherein the alloy layer A has a cubic crystal structure and the alloy layer B has a hexagonal crystal structure, the alloy layer A is stacked over the alloy layer B, the alloy layer B has a c axis oriented in a stacking direction, and the alloy layer A has a (111) face oriented to a surface adjacent to the alloy layer B.

6. A method of manufacturing the stacked structure as claimed in claim 1, comprising a step of heating each of the alloy layer A and the alloy layer B at a temperature of from 200° C. to 300° C.

7. A semiconductor device, comprising the stacked structure as claimed in claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0064] FIG. 1 is a schematic view (1) showing an example of a low-resistance-state stacked structure.

[0065] FIG. 2 is a schematic view (2) showing an example of another low-resistance-state stacked structure.

[0066] FIG. 3 is a schematic view showing an example of a high-resistance-state stacked structure.

[0067] FIG. 4 is an explanatory view (A) for explaining the mutual diffusion of Ge atoms and Sb atoms.

[0068] FIG. 5 is an explanatory view (B) for explaining the mutual diffusion of Ge atoms and Sb atoms.

[0069] FIG. 6 is an explanatory view (C) for explaining the mutual diffusion of Ge atoms and Sb atoms.

[0070] FIG. 7 shows an X-ray diffusion chart of respective stacked structures of Example 1 and Comparative Example 1.

[0071] FIG. 8 shows an electron microscope image of the stacked structure of Example 1.

[0072] FIG. 9 shows an electron microscope image of the stacked structure of Comparative Example 1.

[0073] FIG. 10 is an explanatory view for explaining the constitution of a semiconductor device of Example 3.

[0074] FIG. 11 shows voltage-electrical resistance characteristics of each of semiconductor devices of Example 3 and Comparative Example 5 from a high-resistance state (RESET phase) to a low-resistance state (SET phase).

[0075] FIG. 12 shows voltage-electrical resistance characteristics of each of semiconductor devices of Example 3 and Comparative Example 5 from a low-resistance state (SET phase) to a high-resistance state (RESET phase).

MODE FOR CARRYING OUT THE INVENTION

(Stacked Structure)

[0076] A stacked structure of the present invention has an alloy layer A and an alloy layer B.

<Alloy Layer A>

[0077] The alloy layer A is formed using germanium (Ge) and tellurium (Te) as a main component.

[0078] In the alloy layer A, atomic arrangement of germanium atoms and tellurium atoms gives the stacked structure two phases called “SET phase” and “RESET phase” different in properties and by applying a voltage to the stacked structure, phase transition occurs between these two phases.

[0079] The term “main component” as used herein means atoms forming an essential unit lattice of a layer. When the layer contains at least either of sulfur or selenium as chalcogen atoms (S atom, Se atom), the term means the chalcogen atoms (S atom, Se atom) and atoms forming the essential unit lattice.

[0080] Although the alloy layer A is not particularly limited, it is preferably a layer having a crystal direction oriented in a certain direction and particularly preferably a layer having a cubic crystal structure and at the same time, having a (111) face located on a surface adjacent to the alloy layer B. More preferably, it is a layer having a face-centered cubic crystal structure and at the same time, having a (111) face oriented to a surface adjacent to the alloy layer B.

[0081] If a layer has such a crystal structure, another layer stacked thereon becomes a template producing orientation with the first layer as an underlayer to facilitate the formation of a superlattice structure of these stacked layers.

[0082] A method of forming the alloy layer A is not particularly limited and it can be selected as needed depending on the purpose. Examples include sputtering, molecular beam epitaxy, ALD (Atomic Layer Deposition), and CVD (Chemical Vapor Deposition).

[0083] The thickness of the alloy layer A is not particularly limited and it is preferably more than 0 nm and not more than 4 nm. The layer having a thickness more than 4 nm sometimes shows its own inherent properties and has an influence on the properties of the stacked structure.

<Alloy Layer B>

[0084] The alloy layer B is formed using either of antimony (Sb) or bismuth (Bi) and tellurium (Te) as a main component.

[0085] The alloy layer B is not particularly limited. It includes a layer formed of SbTe or BiTe having an atomic composition ratio of 1:1 or a layer formed at another atomic composition ratio. Particularly preferably, it is made of either of Sb.sub.2Te.sub.3 or Bi.sub.2Te.sub.3 having an atomic composition ratio of 2:3 from the standpoint of stability of atomic arrangement.

[0086] The alloy layer B is not particularly limited. It is preferably a layer having a crystal direction oriented in a certain direction and more preferably a layer having a hexagonal crystal structure and at the same time, having a c axis oriented in a stacking direction.

[0087] If a layer has such a crystal structure, another layer stacked thereon becomes a template producing orientation with the first layer as an underlayer to facilitate the formation of a superlattice structure of these stacked layers.

[0088] A method of forming the alloy layer B is not particularly limited and it can be selected as needed depending on the purpose. Examples include sputtering, molecular beam epitaxy, ALD, and CVD.

[0089] The thickness of the alloy layer B is not particularly limited and it is preferably from 2 nm to 10 nm because such a thickness facilitates the formation of a c-axis-oriented crystal structure.

[0090] Although the stacked structure is not particularly limited, it has preferably a structure having the alloy layer A and the alloy layer B alternately stacked in repetition from the standpoint of causing phase transition more easily.

[0091] In this case, from the standpoint of forming a stacked structure having orientation, it is preferred to stack the alloy layer A over the alloy layer B with the alloy layer B as an underlayer (undermost layer) and repeat alternate stacking of the alloy layer A and the alloy layer B in this order. When the alloy layer B is located as the uppermost layer of the stacked structure, it functions as a layer preventing oxidation of the stacked structure.

[0092] The stacking number of layers for the stacked structure is not particularly limited and supposing that each of the alloy layer A and the alloy layer B is counted as one layer, the number may be from about 10 to 50.

[0093] From the standpoint of forming a stacked structure having orientation, it is also possible to use a substrate having, as an underlayer of the stacked structure, an orientation control layer formed of any of germanium, silicon, tungsten, germanium-silicon, germanium-tungsten, and silicon-tungsten and form the stacked structure on the orientation control layer.

<Chalcogen Atom>

[0094] In the stacked structure, at least either of the alloy layer A or the alloy layer B contains at least either of sulfur (5) or selenium (Se) as chalcogen atoms.

[0095] The structure containing the chalcogen atoms can not only suppress diffusion of Ge atoms from the alloy layer A to the side of the alloy layer B but also stabilize a phase change of the alloy layer A based on the atomic arrangement of Ge atoms and Te atoms.

[0096] The chalcogen atoms (S atom, Se atom) are substituted by Te atoms in each of the alloy layer A and the alloy layer B. Similar to the chalcogen atoms (S atom, Se atom), Te atoms belong to Group 16.

[0097] An excessively large substitution amount of Te atoms by the chalcogen atoms (S atom, Se atom), however, damages the phase transition-causing properties of the stacked structure, while an excessively small substitution amount makes it difficult to suppress diffusion of Ge atoms in the alloy layer A to the side of the alloy layer B.

[0098] A content of the chalcogen atoms in each of the alloy layer A and the alloy layer B is therefore preferably from 0.05 at % to 10.0 at %.

[0099] Although the chalcogen atoms may be contained in at least either of the alloy layer A or the alloy layer B, they are preferably contained in the alloy layer A from the standpoint of effectively suppressing diffusion of Ge atoms in the alloy layer A to the side of the alloy layer B. It is particularly preferred that their content in the alloy layer A is from 0.05 at % to 10.0 at %.

[0100] A method of forming each of the alloy layer A and the alloy layer B containing the chalcogen atoms is not particularly limited and it is possible to use any method capable of adding the chalcogen atoms to the formation materials of the alloy layer A and the alloy layer B and forming the alloy layer A and the alloy layer B.

(Method of Manufacturing Stacked Structure)

[0101] A method of manufacturing a stacked structure according to the present invention is a method of manufacturing the stacked structure of the present invention and it includes at least a step of heating each of the alloy layer A and the alloy layer B at a temperature of from 200° C. to 300° C.

[0102] To the alloy layer A and the alloy layer B, the descriptions made on the stacked structure including the formation methods of them can be applied, but it is important to heat each of the alloy layer A and the alloy layer B at a temperature of from 200° C. to 300° C.

[0103] This means that the stacked structure having excellent orientation can be obtained by heating at such a temperature.

(Semiconductor Device)

[0104] A semiconductor device according to the present invention is comprised of the stacked structure of the present invention.

[0105] The stacked structure can cause phase transition between two phases having respectively different properties, that is, the SET phase and the RESET phase and this phase transition phenomenon is useful for various devices. In particular, the chalcogen atoms contained in the structure enable it to stably exhibit the device properties which the original atomic arrangement has.

[0106] The semiconductor device is not particularly limited insofar as it has the stacked structure and examples include known phase-change devices and spin electronic devices disclosed in Japanese Patent No. 4599598, Japanese Patent No. 4621897 (Patent Document 1), Japanese Patent No. 5750791, Japanese Patent No. 6124320, Japanese Patent No. 6238495, and International Publication No. 2016/147802.

EXAMPLES

[Stacked Structure]

Example 1

[0107] First, a sapphire substrate (product of SHINKOSHA CO., LTD.) having a thickness of 200 μm was moved to a sputtering apparatus (4EP-LL, product of SHIBAURA MECHATRONICS CORPORATION, having 3 3-inch targets mounted thereon) and sputtering was performed with a silicon material (B-doped Si, product of Mitsubishi Materials Corporation) as a target under the following conditions: vacuum back pressure of 1.0×10.sup.−4 Pa, Ar film-formation gas pressure of 0.5 Pa, temperature of 25° C., and RF power of 100 W to form a 40-nm thick amorphous silicon layer as an underlayer on the sapphire substrate.

[0108] Next, sputtering was performed with an Sb.sub.2Te.sub.3 alloy material (product of Mitsubishi Materials Corporation, purity: 99.9%) as a target under the following conditions: Ar film-formation gas pressure of 0.5 Pa, temperature of 25° C., and RF power of 20 W while maintaining the vacuum back pressure to form a 3.0-nm thick Sb.sub.2Te.sub.3 alloy layer (first layer) on the amorphous silicon layer. After the formation, the Sb.sub.2Te.sub.3 alloy layer was crystallized by heating at 210° C.

[0109] Then, sputtering was performed using an S-atom-added GeTe alloy material (Ge.sub.50Te.sub.47S.sub.3, product of Mitsubishi Materials Corporation, purity: 99.9%) as a target under the conditions of RF power of 20 W while maintaining the vacuum back pressure and the Ar-film formation gas pressure and keeping the temperature at 210° C. to form a 0.8-nm thick GeTe alloy layer (first layer) having an S atom content of 3 at % on the Sb.sub.2Te.sub.3 alloy layer and crystallize it.

[0110] Then, an Sb.sub.2Te.sub.3 alloy layer and a GeTe alloy layer were stacked alternately in repetition under the conditions similar to those for the formation of the first layer, respectively, while maintaining the vacuum back pressure and Ar-film formation gas pressure and keeping the temperature at 210° C. to form a stacked structure having a total of 20 layers stacked one after another, more specifically, having 10 Sb.sub.2Te.sub.3 alloy layers and 10 GeTe alloy layers stacked alternately. The thickness of the second and subsequent Sb.sub.2Te.sub.3 alloy layers, however, was changed from the first layer thickness, 3.0 nm, to 1.0 nm. This means that the first layer of the 10 Sb.sub.2Te.sub.3 alloy layers has a thickness of 3.0 nm and the second to tenth layers each have a thickness of 1.0 nm.

[0111] Finally, under conditions similar to those for the first layer except that the thickness was changed from 3.0 nm to 5.0 nm, an Sb.sub.2Te.sub.3 alloy layer was formed as an anti-oxidant layer on the GeTe alloy layer serving as the uppermost layer of the stacked structure and was crystallized.

[0112] Thus, a stacked structure of Example 1 was manufactured.

Comparative Example 1

[0113] In a manner similar to that of Example 1 except for the use of an 5-atom-free GeTe alloy material (product of Mitsubishi Materials Corporation, purity: 99.9%) as the target material instead of the S-atom-added GeTe alloy material (Ge.sub.50Te.sub.47S.sub.3) to form an S-atom-free GeTe alloy layer, a stacked structure of Comparative Example 1 was manufactured.

(Structural Analysis)

[0114] X-ray analysis of each of the stacked structures of Example 1 and Comparative Example 1 was performed with an X-ray diffraction apparatus (“SmartLab”, product of Rigaku Corporation) by a 2-theta/omega method.

[0115] FIG. 7 shows an X-ray diffraction chart of each of the stacked structures of Example 1 and Comparative Example 1.

[0116] The full width at half maximum (FWHM) at each of the diffraction peaks of (003), (006), (009), (0012), (0015), and (0018) shown in FIG. 7 is shown below in Table 2.

TABLE-US-00002 TABLE 2 Diffraction peak (003) (006) (009) (0012) (0015) (0018) Example 1 2.608886 2.025599 0.462832 0.704485 2.264870 0.845408 Full width at half maximum (FWHM) Comparative 3.954496 3.429469 0.484673 0.748552 2.764585 0.908331 Example 1 Full width at half maximum (FWHM)

[0117] As shown in FIG. 7 and Table 2, the full width at half maximum of the stacked structure of Example 1 at each diffraction peak becomes smaller than that of the stacked structure of Comparative Example 1, revealing that the former one has higher crystallinity.

[0118] The arrow shown in FIG. 7 shows a peak shift of the diffraction peak of the stacked structure of Example 1 viewed from that of the stacked structure of Comparative Example 1.

[0119] The structural analysis of each of the stacked structures of Example 1 and Comparative Example 1 was performed using a scanning transmission electron microscope (“JEM-ARM200F”, product of JEOL Ltd.).

[0120] First, the electron microscope image of the stacked structure of Example 1 is shown in FIG. 8.

[0121] As shown in FIG. 8, it has been confirmed clearly that the stacked structure of Example 1 has a space (refer to the dark-color horizontal line in the drawing) based on Van der Waals bonding between a structural unit of 9 atoms consisting of Ge.sub.2Te.sub.2 and Sb.sub.2Te.sub.3 and another structural unit adjacent thereto.

[0122] Further, it has been confirmed from the results of elemental mapping of a portion of the stacked structure of Example 1 by using an energy dispersible X-ray analyzer that a 5-atom layer consisting of Te—Sb—Te—Sb—Te has thereon a Ge—Te—Ge—Te layer, which coincides well with the atomic arrangement model shown in FIG. 1. This means that even if structural observation is performed by adding external energy, mutual diffusion between Ge atoms and Sb atoms is suppressed by the addition of S atoms and the stacked structure can keep original atomic arrangement.

[0123] Next, the electron microscope image of the stacked structure of Comparative Example 1 is shown in FIG. 9.

[0124] As shown in FIG. 9, it has been confirmed that the stacked structure of Comparative Example 1 becomes a uniform alloy as a result of mutual diffusion between Ge atoms and Sb atoms.

[0125] It has been confirmed from the comparison between the S-atom-added stacked structure of Example 1 and the S-atom-free stacked structure of Comparative Example 1 that a stacked structure having stable atomic arrangement can be obtained and diffusion of Ge atoms can be suppressed by the addition of S atoms.

Example 2

[0126] In a manner similar to that of Example 1 except that the target material was changed from the S-atom-added GeTe alloy material (Ge.sub.50Te.sub.47S.sub.3) to an Se-atom-added GeTe alloy material (Ge.sub.50Te.sub.47Se.sub.3, product of Mitsubishi Materials Corporation, purity: 99.9%) and thereby Se atoms were added to the GeTe alloy layer, a stacked structure of Example 2 was manufactured.

[0127] Structural analysis of the stacked structure of Example 2 in a manner similar to that of Example 1 yielded analysis results similar to those of Example 1. It is thus confirmed that even if Se atoms are added instead of S atoms, a stacked structure having stable atomic arrangement can be obtained and diffusion of Ge atoms can be suppressed.

Comparative Example 2

[0128] In a manner similar to that of Example 1 except that the target material was changed from the S-atom-added GeTe alloy material (Ge.sub.50Te.sub.47S.sub.3) to an Al-atom-added GeTe alloy material (Ge.sub.50Te.sub.47Al.sub.3, product of Mitsubishi Materials Corporation, purity: 99.9%) and thereby Al atoms were added to the GeTe alloy layer, a stacked structure of Comparative Example 2 was manufactured.

[0129] Structural analysis of the stacked structure of Comparative Example 2 in a manner similar to that of Comparative Example 1 yielded analysis results similar to those of S-atom-free Comparative Example 1. It has been confirmed that the Al-atom-added stacked structure of Comparative Example 2 becomes a uniform alloy as a result of mutual diffusion between Ge atoms and Sb atoms.

Comparative Example 3

[0130] In a manner similar to that of Example 1 except that the target material was changed from the S-atom-added GeTe alloy material (Ge.sub.50Te.sub.47S.sub.3) to an S-atom-free GeTe alloy material (product of Mitsubishi Materials Corporation, purity: 99.9%) and at the same time, sputtering was performed by adding an oxygen gas to a sputtering gas (Ar gas) at a gas flow rate ratio of 10:1 (Ar gas:oxygen gas) to add O atoms to the GeTe alloy layer, a stacked structure of Comparative Example 3 was manufactured.

[0131] Structural analysis of the stacked structure of Comparative Example 3 in a manner similar to that of Comparative Example 1 yielded analysis results similar to those of S-atom-free Comparative Example 1. It has been confirmed that the O-atom-added stacked structure of Comparative Example 3 becomes a uniform alloy as a result of mutual diffusion between Ge atoms and Sb atoms.

Comparative Example 4

[0132] In a manner similar to that of Comparative Example 3 except that a nitrogen gas was used instead of the oxygen gas to add N atoms to the GeTe alloy layer, a stacked structure of Comparative Example 4 was manufactured.

[0133] Structural analysis of the stacked structure of Comparative Example 4 in a manner similar to that of Comparative Example 1 yielded analysis results similar to those of the S-atom-free structure of Comparative Example 1. It has been confirmed that the N-atom-added stacked structure of Comparative Example 4 becomes a uniform alloy as a result of mutual diffusion between Ge atoms and Sb atoms.

[Semiconductor Device]

Example 3

[0134] A semiconductor device of Example 3 was manufactured according to the constitution of a semiconductor device 10 shown in FIG. 10. Specific manufacturing conditions will be described below. FIG. 10 is an explanatory view for explaining the constitution of the semiconductor device of Example 3.

[0135] As a structure on the bottom surface side of a stacked structure 18, that having, in a SiO.sub.2 layer 12 on a silicon substrate 11, a lower electrode having a W layer 13 and a TiN layer 14 stacked successively in order of mention is used. The TiN layer 14 has a diameter of 90 nm.

[0136] An underlayer 15 formed of an Sb.sub.2Te.sub.3 alloy was formed, on the surface of the structure on which the TiN layer 14 was formed, in a manner similar to that used for the formation of the first Sb.sub.2Te.sub.3 alloy layer of the stacked structure of Example 1 except that the thickness was changed from 3.0 nm to 5.0 nm.

[0137] Next, a GeTe alloy layer 16 was formed on the underlayer 15 in a manner similar to that used for the formation of the first GeTe alloy layer of the stacked structure of Example 1.

[0138] Next, an Sb.sub.2Te.sub.3 alloy layer 17 was formed on the GeTe alloy layer 16 in a manner similar to that used for the formation of the second Sb.sub.2Te.sub.3 alloy layer of the stacked structure of Example 1 except that the thickness was changed from 1.0 nm to 4.0 nm.

[0139] Further, the GeTe alloy layer 16 and the Sb.sub.2Te.sub.3 alloy layer 17 were stacked alternately in repetition and a stacked structure 18 having eight GeTe alloy layers 16 and eight Sb.sub.2Te.sub.3 alloy layers 17 stacked alternately to have a stacked structure of 17 layers in total including the underlayer 15 was formed.

[0140] Finally, a TiN layer 19 was formed on the Sb.sub.2Te.sub.3 alloy layer 17 serving as the uppermost layer of the stacked structure 18 by sputtering with the sputtering apparatus while using Ti and N as a target (composition ratio: 1:1). The resulting TiN layer 19 serves as an upper electrode.

[0141] Thus, the semiconductor device of Example 3 was manufactured. In the semiconductor device of Example 3, the GeTe alloy layers 16 each contain S atoms at a concentration of 3 at %.

Comparative Example 5

[0142] A semiconductor device of Comparative Example 5 was manufactured in a manner similar to that of Example 3 except that each of the GeTe alloy layers 16 was formed as in the formation of the S-atom-free GeTe layer of the stacked structure of Comparative Example 1.

(Device Properties)

[0143] Device properties of each of the semiconductor devices of Example 3 and Comparative Example 5 were measured by connecting an external power supply thereto and applying a voltage between the upper electrode and the lower electrode.

[0144] FIG. 11 shows a change in the voltage-electrical resistance characteristics of each of the semiconductor devices of Example 3 and Comparative Example 5 from a high-resistance state (RESET phase) to a low-resistance state (SET phase).

[0145] FIG. 12 shows a change in the voltage-electrical resistance characteristics of each of the semiconductor devices of Example 3 and Comparative Example 5 from a low-resistance state (SET phase) to a high-resistance state (RESET phase).

[0146] As shown in FIG. 11, it has been confirmed that there is no large difference between the semiconductor devices of Example 3 and Comparative Example 5 in the voltage-electrical resistance characteristics when a phase change is caused from the high-resistance state (RESET phase) to the low-resistance state (SET phase).

[0147] As shown in FIG. 12, on the other hand, a large difference has been confirmed between the semiconductor devices of Example 3 and Comparative Example 5 in the voltage-electrical resistance characteristics when a phase change is caused from the high-resistance state (RESET phase) to the low-resistance state (SET phase).

[0148] This means that the semiconductor device of Example 3 can cause a phase change at a voltage lower even by 39% than that of the semiconductor device of Comparative Example 5.

[0149] Although not shown in the drawing, the semiconductor device of Example 3 can cause a phase change at a current lower even by 27% than that of the semiconductor device of Comparative Example 5.

[0150] This suggests that since the original atomic arrangement of Sb atoms and Ge atoms can stably be retained in the semiconductor device of Example 3 due to the addition of S atoms even if external energy necessary for memory operation is applied, diffusion of Ge atoms is suppressed and further, the semiconductor device can exhibit and keep its original device properties (phase-change properties).

DESCRIPTION OF REFERENCE NUMERALS

[0151] 10: Semiconductor device [0152] 11: Silicon substrate [0153] 12: SiO.sub.2 layer [0154] 13: W layer [0155] 14, 19: TiN layer [0156] 15: Underlayer [0157] 16: GeTe Alloy layer [0158] 17: Sb.sub.2Te.sub.3 Alloy layer [0159] 18: Stacked structure.