Semiconductor Device
The semiconductor device addresses oxidation and low gate threshold issues by using specific barrier layers to prevent electron transport layer oxidation and enhance piezoelectric polarization, achieving stable normally-off operation with increased gate threshold voltage and improved mobility.
Patent Information
- Application Number
- JP2023111629
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2017-12-27
- Filing Date
- 2023-07-06
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2038-11-15
AI Technical Summary
Existing semiconductor devices with HEMT structures face challenges in achieving a normally-off state and have low gate threshold voltage due to issues such as oxidation of the electron transport layer and low carrier mobility from p-type impurities.
The semiconductor device incorporates a back barrier layer containing AlXGa1-X, an electron transit layer with Al a In b Ga (1-a-b), and a top barrier layer with AlYGa(1-Y) to prevent oxidation and enhance piezoelectric polarization, increasing the gate threshold voltage while maintaining low channel resistance.
The solution effectively suppresses oxidation of the electron transport layer, enhances carrier mobility, and increases the gate threshold voltage, enabling stable normally-off operation with improved switching characteristics.
Smart Images

Figure 0007577796000001 
Figure 0007577796000002 
Figure 0007577796000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a semiconductor device including a HEMT (High Electron Mobility Transistor) and a method for manufacturing the same. [Background technology]
[0002] A semiconductor device equipped with a HEMT (High Electron Mobility Transistor) has a problem that it is difficult to make the device normally off due to its structure, and the gate threshold voltage is relatively small. Patent Document 1 discloses an example of a structure that realizes the normally off state, and Patent Document 2 discloses an example of a structure that increases the gate threshold voltage.
[0003] The semiconductor device according to Patent Document 1 includes an i-GaN layer (electron transport layer). An AlN spacer layer (electron supply layer) is formed on the i-GaN layer. An InAlN barrier layer is formed on the AlN spacer layer. A recess exposing the i-GaN layer is formed in the InAlN barrier layer and the AlN spacer layer. A gate electrode is buried in the recess with a gate insulating layer sandwiched therebetween.
[0004] The semiconductor device according to Patent Document 2 includes a p-type back barrier layer formed of a p-type nitride semiconductor. An electron transit layer is formed on the p-type back barrier layer. An electron supply layer is formed on the electron transit layer. A source electrode, a drain electrode, and a gate electrode are formed on the electron supply layer. The gate threshold voltage is increased by the p-type back barrier layer. [Prior art documents] [Patent documents]
[0005] [Patent Document 1] JP 2015-192004 A [Patent Document 2] JP 2016-163017 A
Summary of the Invention
Problems to be Solved by the Invention
[0006] In the semiconductor device according to Patent Document 1, in order to achieve normally-off operation, a recess for exposing the electron traveling layer (i-GaN layer) must be formed. In the process of forming the recess, it is impossible to avoid exposing the electron traveling layer to the outside air. Therefore, there is a risk that the channel resistance increases due to the oxidation of the electron traveling layer. Further, when the gate insulating layer is made of an oxide insulating material, there is also a risk that the electron traveling layer is oxidized due to the material of the gate insulating layer.
[0007] On the other hand, in the semiconductor device according to Patent Document 2, the HEMT must be turned on and off using p-type impurities (holes) introduced into the p-type back barrier layer. The p-type impurities (holes) have the property of relatively low carrier mobility. Therefore, the semiconductor device according to Patent Document 2 has a problem in that the switching characteristics have to be sacrificed.
[0008] One embodiment of the present invention provides a semiconductor device and a method for manufacturing the same that can suppress an increase in channel resistance and increase the gate threshold voltage.
Means for Solving the Problems
[0009] One embodiment of the present invention includes a back barrier layer containing Al X Ga (1-X) N (0 < X ≦ 1), an electron traveling layer containing Al a In b Ga (1-a-b) N (0 ≦ a + b ≦ 1) and formed on the back barrier layer, and a top barrier layer containing Al Y Ga (1-Y) N (0 < Y ≦ 1) and formed on the electron traveling layer, and Al Z Ga (1-Z)An electron supply layer including N(0 < Z ≦ 1), formed on the top barrier layer and having an opening exposing the top barrier layer; a two-dimensional electron gas region formed in a region facing the electron supply layer with the top barrier layer interposed therebetween in a surface layer portion of the electron traveling layer; and a gate electrode layer facing the electron traveling layer with a gate insulating layer formed in the opening of the electron supply layer interposed therebetween. A semiconductor device is provided.
[0010] According to this semiconductor device, a top barrier layer is formed on the electron traveling layer. By this top barrier layer, it is possible to avoid the electron traveling layer from being exposed to the outside air, so that oxidation of the electron traveling layer can be suppressed. As a result, an increase in channel resistance can be suppressed.
[0011] In a structure in which a top barrier layer is formed on the electron traveling layer, an effect in the direction of lowering the gate threshold voltage acts due to piezoelectric polarization generated between the electron traveling layer and the top barrier layer. Therefore, in this semiconductor device, Al X Ga (1-X) The electron traveling layer is formed on a back barrier layer containing N(0 < X ≦ 1).
[0012] According to this back barrier layer, the energy band on the back barrier layer side in the electron traveling layer can be shifted to the high energy side by piezoelectric polarization generated between the back barrier layer and the electron traveling layer. Thereby, the gate threshold voltage can be increased.
[0013] One embodiment of the present invention is a step of forming an electron traveling layer containing Al X Ga (1-X) N(0 < X ≦ 1) on a back barrier layer, and on the electron traveling layer, Al a In b Ga (1-a-b) N(0 ≦ a + b ≦ 1) is formed such that the conduction band energy level at the interface with the electron traveling layer is higher than the Fermi energy level. Al Y Ga (1-Y)A step of forming a top barrier layer including N(0 < Y ≤ 1), and on the top barrier layer, Al is provided such that the conduction band energy level at the interface between the top barrier layer and the electron transport layer is lower than the Fermi energy level. Z Ga (1-Z) A step of forming an electron supply layer including N(0 < Z ≤ 1), oxidizing a part of the electron supply layer to form an oxide, and removing the oxide to form an opening exposing the top barrier layer in the electron supply layer, and increasing the conduction band energy level in a region facing the opening at the interface between the top barrier layer and the electron transport layer to be higher than the Fermi energy level; a step of forming a gate insulating layer on a portion of the top barrier layer exposed from the opening of the electron supply layer; and a step of forming a gate electrode layer on the gate insulating layer, to provide a method for manufacturing a semiconductor device.
[0014] According to this manufacturing method, a top barrier layer is formed on the electron transport layer. This top barrier layer can prevent the electron transport layer from being exposed to the outside air, so that oxidation of the electron transport layer can be suppressed. As a result, an increase in channel resistance can be suppressed.
[0015] In a structure where a top barrier layer is formed on the electron transport layer, an effect in the direction of lowering the gate threshold voltage acts due to piezoelectric polarization generated between the electron transport layer and the top barrier layer. Therefore, in this manufacturing method, an electron transport layer is formed on a back barrier layer including Al X Ga (1-X) N(0 < X ≤ 1).
[0016] According to this back barrier layer, the energy band on the back barrier layer side in the electron transport layer can be shifted to the high energy side due to piezoelectric polarization generated between the back barrier layer and the electron transport layer. Thereby, the gate threshold voltage can be increased.
Brief Description of the Drawings
[0017] [Figure 1]FIG. 1 is a cross-sectional view showing a semiconductor device according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a graph showing the relationship between the thickness and the Al composition ratio of the top barrier layer shown in FIG. [Diagram 3] FIG. 3 is a graph showing the results of a simulation of the effect of introducing a back barrier layer. [Figure 4A] FIG. 4A is a cross-sectional view showing an example of a method for manufacturing the semiconductor device shown in FIG. [Figure 4B] FIG. 4B is a cross-sectional view showing a step subsequent to that of FIG. 4A. [Figure 4C] FIG. 4C is a cross-sectional view showing a step subsequent to FIG. 4B. [Figure 4D] FIG. 4D is a cross-sectional view showing a step subsequent to FIG. 4C. [Figure 4E] FIG. 4E is a cross-sectional view showing a step subsequent to FIG. 4D. [Figure 4F] FIG. 4F is a cross-sectional view showing a step subsequent to FIG. 4E. [Figure 4G] FIG. 4G is a cross-sectional view showing a step subsequent to FIG. 4F. [Figure 4H] FIG. 4H is a cross-sectional view showing a step subsequent to FIG. 4G. [Figure 4I] FIG. 4I is a cross-sectional view showing a step subsequent to FIG. 4H. [Figure 4J] FIG. 4J is a cross-sectional view showing a step subsequent to FIG. 4I. [Diagram 5] FIG. 5 is a cross-sectional view showing a semiconductor device according to a second embodiment of the present invention. [Figure 6] FIG. 6 is a cross-sectional view showing a semiconductor device according to a third embodiment of the present invention. [Figure 7] FIG. 7 is a cross-sectional view showing a semiconductor device according to a fourth embodiment of the present invention. [Figure 8] FIG. 8 is an enlarged view of region VII shown in FIG. [Figure 9] FIG. 9 is a graph showing the time characteristics of the gate threshold voltage of the semiconductor device shown in FIG. [Figure 10A]FIG. 10A is a cross-sectional view showing an example of a method for manufacturing the semiconductor device shown in FIG. [Figure 10B] FIG. 10B is a cross-sectional view showing a step subsequent to that of FIG. 10A. [Figure 10C] FIG. 10C is a cross-sectional view showing a step subsequent to that shown in FIG. 10B. [Figure 10D] FIG. 10D is a cross-sectional view showing a step subsequent to FIG. 10C. [Figure 10E] FIG. 10E is a cross-sectional view showing a step subsequent to FIG. 10D. [Figure 10F] FIG. 10F is a cross-sectional view showing a step subsequent to FIG. 10E. [Figure 10G] FIG. 10G is a cross-sectional view showing a step subsequent to FIG. 10F. [Figure 10H] FIG. 10H is a cross-sectional view showing a step subsequent to that shown in FIG. 10G. [Figure 10I] FIG. 10I is a cross-sectional view showing a step subsequent to FIG. 10H. [Figure 10J] FIG. 10J is a cross-sectional view showing a step subsequent to FIG. 10I. [Figure 10K] FIG. 10K is a cross-sectional view showing a step subsequent to FIG. 10J. [Figure 10L] FIG. 10L is a cross-sectional view showing a step subsequent to FIG. 10K. [Figure 10M] FIG. 10M is a cross-sectional view showing a step subsequent to FIG. 10L. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0018] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
[0019] FIG. 1 is a cross-sectional view showing a semiconductor device 1 according to a first embodiment of the present invention.
[0020] The semiconductor device 1 has a basic configuration including a HEMT (High Electron Mobility Transistor) including a Group III nitride semiconductor.
[0021] 1, the semiconductor device 1 includes a substrate 2. The substrate 2 may be a Si substrate, a SiC substrate, a sapphire substrate, a GaN substrate, or the like. In this embodiment, the substrate 2 is a Si substrate. The substrate 2 includes a first main surface 3 on one side and a second main surface 4 on the other side.
[0022] A nucleation layer 5, a buffer layer 6, a double hetero stack structure 7, and an electron supply layer 8 are formed in this order on the first main surface 3 of the substrate 2. In this embodiment, the nucleation layer 5 contains AlN. The thickness of the nucleation layer 5 may be 100 nm or more and 300 nm or less (for example, about 200 nm).
[0023] The buffer layer 6 may have a single-layer structure consisting of a single buffer layer 6. In this embodiment, the buffer layer 6 has a laminated structure in which a plurality of (two or more) buffer layers are laminated. The plurality of buffer layers may be laminated on the nucleation layer 5 in a sequence in which the Al composition ratio gradually decreases in the lamination direction.
[0024] In this embodiment, the multiple buffer layers include a first buffer layer 9 and a second buffer layer 10 that are stacked in this order on the nucleation layer 5. The first buffer layer 9 is made of Al α Ga (1-α) N (0<α≦1). The second buffer layer 10 contains Al β Ga (1-β) It contains N (0<β<α≦1). The Al composition ratio α may be 0.52, and the Al composition ratio β may be 0.12.
[0025] The first buffer layer 9 may have a thickness of 50 nm or more and 200 nm or less (for example, about 100 nm). The second buffer layer 10 may have a thickness of 50 nm or more and 200 nm or less (for example, about 100 nm).
[0026] The double heterostructure 7 has a laminated structure including a back barrier layer 11, an electron transport layer 12, and a top barrier layer 13, which are laminated in this order from above the buffer layer 6. The back barrier layer 11 is made of Al X Ga (1-X)It includes N (0 < X ≤ 1). The Al composition ratio X may be less than 1 (X < 1). The thickness of the back barrier layer 11 may be 2 nm or more and 2000 nm or less (for example, about 1000 nm).
[0027] The Al composition ratio X of the back barrier layer 11 is a design value that can take various values according to the electrical characteristics (for example, breakdown voltage) of the semiconductor device 1 and is not uniquely determined. The lattice constant of the back barrier layer 11 varies according to the electrical characteristics (for example, breakdown voltage) of the semiconductor device 1 and the Al composition ratio X, so it is not uniquely determined.
[0028] The Al composition ratio X may be 0.01 or more and 0.1 or less. The Al composition ratio X may be 0.1 or more and 0.2 or less. The Al composition ratio X may be 0.2 or more and 0.3 or less. Since the thickness of the back barrier layer 11 varies according to the Al composition ratio X, it is not uniquely determined.
[0029] The electron transport layer 12 contains Al a In b Ga (1-a-b) It contains N (0 ≤ a + b ≤ 1). In this form, the electron transport layer 12 is made of GaN. The a-axis lattice constant (= 3.189 Å) of the electron transport layer 12 is larger than the a-axis lattice constant of the back barrier layer 11 containing Al in addition to GaN in this form. The thickness of the electron transport layer 12 may be 50 nm or more and 300 nm or less.
[0030] The electron transport layer 12 may contain impurity-free Al a In b Ga (1-a-b) It may contain N (GaN in this form). The electron transport layer 12 may contain Al a In b Ga (1-a-b) It may contain N (GaN in this form) added with carbon as an impurity.
[0031] The top barrier layer 13 contains Al Y Ga (1-Y)It includes N (0 < Y ≤ 1). The Al composition ratio Y may be less than 1 (Y < 1). The a-axis lattice constant of the top barrier layer 13 is smaller than the a-axis lattice constant of the electron transport layer 12 (= 3.189 Å). The thickness and the Al composition ratio Y of the top barrier layer 13 may be determined based on the graph in FIG. 2.
[0032] FIG. 2 is a graph showing the relationship between the thickness and the Al composition ratio Y of the top barrier layer 13 shown in FIG. 1. In FIG. 2, the vertical axis represents the thickness, and the horizontal axis represents the Al composition ratio Y.
[0033] In FIG. 2, a curve L connecting a plurality of plotted points is shown. The curve L shows the relationship between the thickness and the Al composition ratio Y of the top barrier layer 13. The coordinate plane formed by the vertical axis and the horizontal axis is divided by the curve L into an upper region RU and a lower region RL.
[0034] The upper region RU is a region where, in a state where the electron supply layer 8 is not formed on the double heterostructure 7, a two-dimensional electron gas region 14 (described later) is formed in the region between the electron transport layer 12 and the top barrier layer 13.
[0035] The lower region RL is a region where, in a state where the electron supply layer 8 is not formed on the double heterostructure 7, a two-dimensional electron gas region 14 (described later) is not formed in the region between the electron transport layer 12 and the top barrier layer 13.
[0036] The thickness and the Al composition ratio Y of the top barrier layer 13 are set to be located in the lower region RL in the coordinate plane of FIG. 2. That is, in a state where the electron supply layer 8 is not formed on the double heterostructure 7, the conduction band energy level EC at the interface between the electron transport layer 12 and the top barrier layer 13 is higher than the Fermi energy level EF (EC > EF).
[0037] Referring to curve L, the thickness of the top barrier layer 13 may be 0.5 nm or more and 50 nm or less. Also, the Al composition ratio Y of the top barrier layer 13 may be 0.05 or more and 0.3 or less. If the thickness of the top barrier layer 13 is 1.0 nm or more, the influence of the leakage of the wave function of the two-dimensional electron gas region 14 (described later) to the top barrier layer 13 side can be kept good.
[0038] Referring to FIG. 1 again, the electron supply layer 8 is formed on the top barrier layer 13. The electron supply layer 8 contains crystalline Al Z Ga (1-Z) N (0 < Z ≦ 1). In this form, the electron supply layer 8 is made of crystalline AlN.
[0039] The a-axis lattice constant of the electron supply layer 8 is smaller than the a-axis lattice constants of the back barrier layer 11, the electron traveling layer 12, and the top barrier layer 13. The thickness of the electron supply layer 8 may be 1 nm or more and 5 nm or less (for example, about 2 nm). An opening 15 for exposing the top barrier layer 13 is formed in the electron supply layer 8.
[0040] The top barrier layer 13 has an exposed portion exposed from the opening 15 of the electron supply layer 8 and a covered portion covered by the electron supply layer 8. The exposed portion and the covered portion of the top barrier layer 13 form an integrally continuous flat surface. The exposed portion of the top barrier layer 13 is not formed so as to be recessed by one step toward the substrate 2 with respect to the covered portion of the top barrier layer 13.
[0041] The conduction band energy level EC at the interface between the electron traveling layer 12 and the top barrier layer 13 is adjusted by the electron supply layer 8. More specifically, the conduction band energy level EC in the region of the electron supply layer 8 facing the interface between the electron traveling layer 12 and the top barrier layer 13 is smaller than the Fermi energy level EF in the steady state (EC < EF). Therefore, in the steady state, a two-dimensional electron gas region 14 is formed in the region of the surface layer portion of the electron traveling layer 12 facing the electron supply layer 8.
[0042] On the other hand, the conduction band energy level EC of the region facing the opening 15 of the electron supply layer 8 at the interface between the electron transit layer 12 and the top barrier layer 13 is higher than the Fermi energy level EF in the steady state (EC>EF). Therefore, in the steady state, the two-dimensional electron gas region 14 is not formed in the region facing the opening 15 of the electron supply layer 8 in the surface layer of the electron transit layer 12.
[0043] A voltage-resistant insulating layer 21 (insulating layer) is formed on the electron supply layer 8. In this embodiment, the voltage-resistant insulating layer 21 has a laminated structure including a planarizing layer 22, a passivation layer 23, and a spacer layer 24, which are laminated in this order from above the electron supply layer 8.
[0044] The planarization layer 22 covers the electron supply layer 8. The planarization layer 22 is formed on the electron supply layer 8 to improve planarity. The planarization layer 22 may contain GaN. The thickness of the planarization layer 22 may be not less than 1 nm and not more than 5 nm (for example, about 2 nm).
[0045] The passivation layer 23 covers the planarization layer 22. The passivation layer 23 may contain SiN. The passivation layer 23 suppresses charge trapping and maintains the insulating properties of the withstand voltage retaining insulating layer 21. The thickness of the passivation layer 23 may be 1 nm or more and 40 nm or less (for example, about 25 nm).
[0046] The spacer layer 24 covers the passivation layer 23. The thickness of the spacer layer 24 is greater than the thickness of the planarization layer 22 and the thickness of the passivation layer 23. The spacer layer 24 is formed to separate the gate electrode layer 32, which will be described later, from the electron supply layer 8. The spacer layer 24 may contain SiO2. The thickness of the spacer layer 24 may be 10 nm or more and 100 nm or less (for example, about 70 nm).
[0047] A through hole 25 communicating with the opening 15 of the electron supply layer 8 is formed in the voltage-resistant insulating layer 21. The through hole 25 of the voltage-resistant insulating layer 21 forms one gate contact hole 26 between itself and the opening 15 of the electron supply layer 8.
[0048] The gate contact hole 26 is formed in a tapered shape with an opening area larger than a bottom area in a cross-sectional view. The opening edge portion of the gate contact hole 26 may have a curved surface facing the inside of the gate contact hole 26.
[0049] A gate insulating layer 31 is formed in the gate contact hole 26. The gate insulating layer 31 includes an oxide insulating material. The gate insulating layer 31 preferably includes an oxide insulating material in an amorphous state.
[0050] The oxide insulating material includes at least one of SiO2, SiON, Al2O3, HfSiO, and HfO2. The gate insulating layer 31 may have a single-layer structure including a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, or a HfO2 layer. The gate insulating layer 31 may have a multilayer structure including at least one of a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, or a HfO2 layer.
[0051] The gate insulating layer 31 may include a SiN layer instead of or in addition to the oxide insulating material. The gate insulating layer 31 may have a laminated structure including at least one oxide insulating material layer selected from the group consisting of a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, and a HfO2 layer, and a SiN layer. In this embodiment, the gate insulating layer 31 is made of amorphous SiO2.
[0052] The gate insulating layer 31 includes a first region and a second region. The first region is formed along the sidewall of the gate contact hole 26. The second region is formed along the surface of the top barrier layer 13. The thickness T1 of the first region of the gate insulating layer 31 may be equal to or greater than the thickness T2 of the second region of the gate insulating layer 31 (T1≧T2). It is preferable that the thickness T1 of the first region of the gate insulating layer 31 is greater than the thickness T2 of the second region of the gate insulating layer 31 (T1>T2).
[0053] The gate insulating layer 31 is extended in the form of a film onto the voltage-holding insulating layer 21 through the inner wall of the gate contact hole 26. As a result, the gate insulating layer 31 covers the surface of the voltage-holding insulating layer 21. In this embodiment, the gate insulating layer 31 covers almost the entire surface of the voltage-holding insulating layer 21.
[0054] A gate electrode layer 32 is formed on the gate insulating layer 31. The gate electrode layer 32 enters the gate contact hole 26 from above the voltage-resistant insulating layer 21. The gate electrode layer 32 covers the voltage-resistant insulating layer 21 and has a covering portion facing the electron supply layer 8 with the voltage-resistant insulating layer 21 in between.
[0055] In the gate contact hole 26, the gate electrode layer 32 faces a sidewall of the gate contact hole 26 with the first region of the gate insulating layer 31 therebetween. In the gate contact hole 26, the gate electrode layer 32 faces the top barrier layer 13 with the second region of the gate insulating layer 31 therebetween.
[0056] The capacitance of a portion of the gate electrode layer 32 facing the sidewall of the gate contact hole 26 across the first region of the gate insulating layer 31 is smaller than the capacitance of a portion of the gate electrode layer 32 facing the top barrier layer 13 across the second region of the gate insulating layer 31. This makes it possible to reduce switching noise.
[0057] The gate insulating layer 31 is formed with a drain contact hole 33 and a source contact hole 34. The drain contact hole 33 is formed in a region of the gate insulating layer 31 that faces the electron supply layer 8. The drain contact hole 33 penetrates the gate insulating layer 31 and the breakdown voltage holding insulating layer 21, exposing the electron supply layer 8.
[0058] The source contact hole 34 is formed in a region different from the drain contact hole 33 in a region of the gate insulating layer 31 that faces the electron supply layer 8. The source contact hole 34 penetrates the gate insulating layer 31 and the breakdown voltage holding insulating layer 21, exposing the electron supply layer 8.
[0059] The drain contact hole 33 is filled with a drain electrode 35. The drain electrode 35 is electrically connected to the electron supply layer 8 within the drain contact hole 33. The source contact hole 34 is filled with a source electrode 36. The source electrode 36 is electrically connected to the electron supply layer 8 within the source contact hole 34.
[0060] When a gate voltage equal to or higher than a predetermined gate threshold voltage Vth is applied to the gate electrode layer 32, in the surface layer portion of the electron traveling layer 12 in the region directly under the gate electrode layer 32, the conduction band energy level EC becomes smaller than the Fermi energy level EF (EC < EF). As a result, a two-dimensional electron gas region 14 is formed in the surface layer portion of the electron traveling layer 12 in the region directly under the gate electrode layer 32, and a current flows between the drain electrode 35 and the source electrode 36.
[0061] On the other hand, when a gate voltage lower than the gate threshold voltage Vth is applied to the gate electrode layer 32, the conduction band energy level EC in the region directly under the gate electrode layer 32 in the surface layer portion of the electron traveling layer 12 becomes larger than the Fermi energy level EF (EC > EF).
[0062] In this case, the two-dimensional electron gas region 14 is not formed in the surface portion of the electron transit layer 12 in the region directly below the gate electrode layer 32. Therefore, no current flows between the drain electrode 35 and the source electrode 36. In this manner, the semiconductor device 1 achieves a normally-off operation.
[0063] Fig. 3 is a graph showing the results of a simulation of the effect of the introduction of the back barrier layer 11. In Fig. 3, the vertical axis indicates the gate threshold voltage Vth, and the horizontal axis indicates the item axis. Here, the gate threshold voltage Vth was examined using a model of normally-on operation. Although the gate threshold voltage Vth has a negative value, the effect of the introduction of the back barrier layer 11 does not substantially change between normally-on operation and normally-off operation.
[0064] 3 shows a first bar graph L1, a second bar graph L2, a third bar graph L3, a fourth bar graph L4, and a fifth bar graph L5. The first bar graph L1 shows a first gate threshold voltage Vth1 of the semiconductor device according to the reference example. The semiconductor device according to the reference example has a structure almost similar to that of the semiconductor device 1, except that the semiconductor device according to the reference example does not include the back barrier layer 11. A detailed description of the semiconductor device according to the reference example will be omitted.
[0065] The second bar graph L2 shows the thickness of the electron transport layer 12 in the semiconductor device 1 set to 100 nm and the Al X Ga (1-X) 4 shows the second gate threshold voltage Vth2 when the Al composition ratio X of N is set to 0.03.
[0066] The third bar graph L3 shows the thickness of the electron transport layer 12 in the semiconductor device 1 set to 50 nm and the Al X Ga (1-X) 3 shows the third gate threshold voltage Vth3 when the Al composition ratio X of N is set to 0.03.
[0067] The fourth bar graph L4 shows the results when the thickness of the electron transit layer 12 in the semiconductor device 1 is set to 100 nm and the Al X Ga (1-X) 3 shows the fourth gate threshold voltage Vth4 when the Al composition ratio X of N is set to 0.06.
[0068] The fifth bar graph L5 shows the results when the thickness of the electron transit layer 12 in the semiconductor device 1 is set to 50 nm and the Al X Ga (1-X) 4 shows the fifth gate threshold voltage Vth5 when the Al composition ratio X of N is set to 0.06.
[0069] With reference to the first bar graph L1, the first gate threshold voltage Vth1 was −4V. With reference to the second bar graph L2, the second gate threshold voltage Vth2 was −3.7V. With reference to the third bar graph L3, the third gate threshold voltage Vth3 was −2.9V. With reference to the fourth bar graph L4, the fourth gate threshold voltage Vth4 was −3.5V. With reference to the fifth bar graph L5, the fifth gate threshold voltage Vth5 was −2.5V.
[0070] From this, it was found that the gate threshold voltage Vth of the semiconductor device 1 increases (shifts in the positive direction) compared to the semiconductor device of the reference example.
[0071] Furthermore, referring to the second bar graph L2, the third bar graph L3, and the fourth bar graph L4 and the fifth bar graph L5, it was found that even when the Al composition ratio X is the same, the gate threshold voltage Vth increases (shifts in the positive direction) as the thickness of the electron transit layer 12 decreases.
[0072] Furthermore, referring to the second bar graph L2, the fourth bar graph L4, and the third bar graph L3 and the fifth bar graph L5, it was found that even when the thickness of the electron transit layer 12 is the same, the gate threshold voltage Vth increases (shifts in the positive direction) as the Al composition ratio X increases.
[0073] As described above, according to the semiconductor device 1, the top barrier layer 13 is formed on the electron transport layer 12. Since the top barrier layer 13 can prevent the electron transport layer 12 from being exposed to the outside air, oxidation of the electron transport layer 12 can be suppressed. As a result, an increase in channel resistance can be suppressed.
[0074] In the structure in which the top barrier layer 13 is formed on the electron transport layer 12, an effect in the direction of decreasing the gate threshold voltage Vth acts due to the piezoelectric polarization generated between the electron transport layer 12 and the top barrier layer 13. Therefore, in this semiconductor device 1, Al X Ga (1-X) The electron transport layer 12 is formed on the back barrier layer 11 containing N (0 < X ≤ 1).
[0075] According to this back barrier layer 11, the energy band on the back barrier layer 11 side in the electron transport layer 12 can be shifted to the high energy side by the piezoelectric polarization generated between the back barrier layer 11 and the electron transport layer 12. Thereby, the carrier mobility can be improved, and at the same time, as shown in FIG. 3, the gate threshold voltage Vth can be increased.
[0076] FIGS. 4A to 4J are cross-sectional views showing an example of a method for manufacturing the semiconductor device 1 shown in FIG. 1.
[0077] Referring to FIG. 4A, first, a substrate 2 having a first main surface 3 and a second main surface 4 is prepared. In this embodiment, the substrate 2 is a Si substrate.
[0078] Next, referring to FIG. 4B, a nucleation layer 5, a buffer layer 6, a double heterostructure layer 7, and an electron supply layer 8 are formed in this order on the first main surface 3 of the substrate 2. The nucleation layer 5, the buffer layer 6, the double heterostructure layer 7, and the electron supply layer 8 are each formed by an epitaxial growth method.
[0079] More specifically, the nucleation layer 5 is formed by epitaxially growing AlN from above the substrate 2. The buffer layer 6 is formed by epitaxially growing AlGaN from above the nucleation layer 5.
[0080] In this embodiment, the step of forming the buffer layer 6 includes a step of forming a first buffer layer 9 and a second buffer layer 10 in this order from above the nucleation layer 5. The first buffer layer 9 is formed from Al α Ga (1-α) The second buffer layer 10 is formed by epitaxially growing Al N on the first buffer layer 9. β Ga (1-β) It is formed by epitaxially growing N.
[0081] The process of forming the double hetero stack structure 7 includes the steps of forming a back barrier layer 11, an electron transit layer 12, and a top barrier layer 13 on the buffer layer 6 in this order.
[0082] The back barrier layer 11 is formed by depositing Al X Ga (1-X) The electron transport layer 12 is formed by epitaxially growing Al N on the back barrier layer 11. a In b Ga (1-a-b) It is formed by epitaxially growing N (in this form, GaN).
[0083] The top barrier layer 13 is formed by depositing Al Y Ga (1-Y) The top barrier layer 13 is formed by epitaxially growing N. The top barrier layer 13 is formed on the electron transit layer 12 so that the conduction band energy level EC at the interface with the electron transit layer 12 is higher than the Fermi energy level EF (EC>EF).
[0084] The electron supply layer 8 is formed by depositing Al Z Ga (1-Z)It is formed by epitaxially growing N (AlN in this form). The electron supply layer 8 is formed on the top barrier layer 13 such that the conduction band energy level EC at the interface of the electron traveling layer 12 and the top barrier layer 13 is smaller than the Fermi energy level EF (EC < EF).
[0085] Next, referring to FIG. 4C, the breakdown voltage holding insulating layer 21 is formed on the electron supply layer 8. The forming process of the breakdown voltage holding insulating layer 21 includes a process of forming a planarization layer 22, a passivation layer 23, and a spacer layer 24 in this order from above the electron supply layer 8.
[0086] The planarization layer 22 is formed by epitaxially growing GaN from above the electron traveling layer 12. The passivation layer 23 may be formed by a CVD method. The passivation layer 23 may contain SiN. The spacer layer 24 may be formed by a CVD method. The spacer layer 24 may contain SiO2.
[0087] Next, referring to FIG. 4D, a mask 41 having a predetermined pattern is formed on the breakdown voltage holding insulating layer 21. The mask 41 may be a resist mask made of a photosensitive resin. The mask 41 has an opening 42 that exposes the region where the through hole 25 is to be formed.
[0088] Next, unnecessary portions of the breakdown voltage holding insulating layer 21 are removed. The unnecessary portions of the breakdown voltage holding insulating layer 21 may be removed by an etching method through the mask 41. Thereby, a through hole 25 that exposes the electron supply layer 8 is formed in the breakdown voltage holding insulating layer 21. After the through hole 25 is formed, the mask 41 is removed.
[0089] Next, referring to FIG. 4E, by an oxidation treatment method, the portion of the electron supply layer 8 that is exposed from the through hole 25 of the breakdown voltage holding insulating layer 21 is oxidized. Thereby, an oxide 43 of the electron supply layer 8 is formed in the through hole 25 of the breakdown voltage holding insulating layer 21. The oxide 43 may contain AlON or Al2O3.
[0090] The oxidation treatment method may be a plasma oxidation treatment method. The plasma oxidation treatment method is performed in an oxygen gas atmosphere until the portions of the electron supply layer 8 exposed from the through holes 25 are all oxidized. The treatment temperature may be 100° C. or higher and 900° C. or lower. The treatment time may be 1 hour or higher and 15 hours or lower. The oxygen concentration in the oxygen gas may be about 30%.
[0091] According to the plasma oxidation method, after the oxide 43 is formed, oxygen in the atmosphere does not or hardly penetrates into the top barrier layer 13. Therefore, the surface of the top barrier layer 13 is not or hardly oxidized.
[0092] 4F, the oxide 43 is removed. As a result, an opening 15 communicating with the through hole 25 of the voltage-resistant insulating layer 21 is formed in the electron supply layer 8, and one gate contact hole 26 including the opening 15 and the through hole 25 is formed.
[0093] Furthermore, with the removal of the oxide 43, the conduction band energy level EC of the region exposed through the gate contact hole 26 at the interface between the top barrier layer 13 and the electron transit layer 12 becomes higher than the Fermi energy level EF (EC>EF).
[0094] The oxide 43 may be removed by an etching method (for example, a wet etching method). The etching solution may be a sulfuric acid hydrogen peroxide mixture (SPM) containing sulfuric acid and hydrogen peroxide. The oxide 43 has an etching selectivity different from that of the electron supply layer 8 and the top barrier layer 13 in that the oxide 43 contains oxygen atoms but does not contain gallium atoms.
[0095] Therefore, almost no part of the electron supply layer 8 and the top barrier layer 13 is removed when the oxide 43 is removed. As a result, the part of the top barrier layer 13 that forms the bottom wall of the opening 15 and the part of the top barrier layer 13 that contacts the electron supply layer 8 form flat surfaces that are flat with each other.
[0096] Furthermore, this step is performed in a state where the electron transit layer 12 is covered with the top barrier layer 13. This prevents the electron transit layer 12 from being exposed to the outside air, thereby preventing oxidation of the electron transit layer 12 and thus preventing an increase in channel resistance.
[0097] Next, referring to FIG. 4G, a gate insulating layer 31 is formed on the top barrier layer 13. In this embodiment, the gate insulating layer 31 is formed so as to be drawn out in a film form onto the voltage-resistant insulating layer 21 via the inner wall of the gate contact hole 26. In this embodiment, the gate insulating layer 31 includes SiO2 in an amorphous state. The gate insulating layer 31 may be formed by a CVD method.
[0098] 4H, a gate electrode layer 32 is formed on the gate insulating layer 31. The gate electrode layer 32 may be formed through a step of forming a conductive layer by a CVD method and a step of patterning the conductive layer by an etching method.
[0099] 4I, a mask 45 having a predetermined pattern is formed on the gate insulating layer 31. The mask 45 may be a resist mask made of a photosensitive resin. The mask 45 has openings 46 that expose regions where the drain contact holes 33 and the source contact holes 34 are to be formed.
[0100] Next, unnecessary portions of the gate insulating layer 31 and the voltage-resistant insulating layer 21 are removed until the electron supply layer 8 is exposed. The unnecessary portions of the gate insulating layer 31 and the voltage-resistant insulating layer 21 may be removed by an etching method using a mask 45. This forms a drain contact hole 33 and a source contact hole 34 that expose the electron supply layer 8. The mask 45 is then removed.
[0101] 4J, a drain electrode 35 is embedded in the drain contact hole 33, and a source electrode 36 is embedded in the source contact hole 34. The drain electrode 35 and the source electrode 36 may be formed through a conductive layer forming step using a CVD method and a conductive layer patterning step using an etching method. Through the steps including those described above, the semiconductor device 1 is manufactured.
[0102] 5 is a cross-sectional view showing a semiconductor device 51 according to a second embodiment of the present invention. In the following, structures corresponding to those described in the semiconductor device 1 are given the same reference numerals and descriptions thereof will be omitted.
[0103] The semiconductor device 51 further includes a cap layer 52 interposed between the top barrier layer 13 and the electron supply layer 8. The cap layer 52 is less susceptible to oxidation than the top barrier layer 13. More specifically, the cap layer 52 is made of a group III nitride semiconductor that does not contain Al. More specifically, the cap layer 52 is made of GaN.
[0104] The cap layer 52 has a thickness that allows a two-dimensional electron gas region 14 to be formed in a region facing the electron supply layer 8 in the surface layer portion of the electron transit layer 12. The cap layer 52 is exposed from the bottom of the gate contact hole 26. The gate insulating layer 31 is in contact with the cap layer 52.
[0105] The cap layer 52 is formed by epitaxially growing GaN on the top barrier layer 13 after the step of forming the top barrier layer 13 and before the step of forming the electron supply layer 8 .
[0106] As described above, according to the semiconductor device 51, the cap layer 52, which is less susceptible to oxidation than the top barrier layer 13, is exposed from the gate contact hole 26 and in contact with the gate insulating layer 31. This makes it possible to suppress direct oxidation of the electron transit layer 12 by the top barrier layer 13 and the cap layer 52, while suppressing oxidation of the region exposed from the gate contact hole 26 and in contact with the gate insulating layer 31. As a result, an increase in channel resistance can be suppressed, thereby improving the stability of the channel resistance.
[0107] 6 is a cross-sectional view showing a semiconductor device 61 according to a third embodiment of the present invention. In the following, structures corresponding to those described in the semiconductor device 1 are given the same reference numerals and descriptions thereof will be omitted.
[0108] The semiconductor device 61 further includes a barrier insulating layer 62 interposed between the top barrier layer 13 and the gate insulating layer 31. The barrier insulating layer 62 is made of an insulating material other than an oxide insulating material and suppresses oxidation of the top barrier layer 13. For example, when the gate insulating layer 31 is made of an oxide insulating material, the barrier insulating layer 62 suppresses oxidation of the top barrier layer 13 by the gate insulating layer 31.
[0109] More specifically, the barrier insulating layer 62 is made of amorphous Al. γ Ga (1-γ) N (0<γ≦1). In this embodiment, the barrier insulating layer 62 is made of AlN in an amorphous state. The barrier insulating layer 62 contacts the top barrier layer 13 at the opening 15 in the electron supply layer 8.
[0110] As a result, the electron supply layer 8 made of crystalline AlN and the barrier insulating layer 62 made of amorphous AlN are located on the same plane (the surface of the top barrier layer 13). The thickness of the barrier insulating layer 62 may be 1 nm or more and 5 nm or less (for example, about 1.5 nm).
[0111] The barrier insulating layer 62 does not have piezoelectricity. In the region of the surface portion of the electron transit layer 12 facing the barrier insulating layer 62, a state in which the conduction band energy level EC is larger than the Fermi energy level EF (EC>EF) is maintained. Therefore, in the steady state, the two-dimensional electron gas region 14 is not formed in the region of the surface portion of the electron transit layer 12 directly below the gate electrode layer 32.
[0112] A mixed crystal layer (not shown) may be formed in the boundary region between the barrier insulating layer 62 and the gate insulating layer 31, in which the insulating material of the barrier insulating layer 62 and the insulating material of the gate insulating layer 31 are mixed. This mixed crystal layer may contain an oxide insulator.
[0113] In this embodiment, the barrier insulating layer 62 is extended in the form of a film onto the voltage-holding insulating layer 21 via the inner wall of the gate contact hole 26 while remaining covered by the gate insulating layer 31. The barrier insulating layer 62 covers the surface of the voltage-holding insulating layer 21. In this embodiment, the barrier insulating layer 62 covers almost the entire surface of the voltage-holding insulating layer 21.
[0114] In this embodiment, the drain contact hole 33 and the source contact hole 34 penetrate the gate insulating layer 31, the barrier insulating layer 62 and the voltage-resistant insulating layer 21 so as to expose the electron supply layer 8.
[0115] The step of forming the barrier insulating layer 62 is performed prior to the step of forming the gate insulating layer 31. In the step of forming the barrier insulating layer 62, the barrier insulating layer 62 is formed on the portion of the top barrier layer 13 exposed from the opening 15 of the electron supply layer 8.
[0116] The barrier insulating layer 62 is formed so as to extend in a film shape onto the voltage-resistant insulating layer 21 through the inner wall of the gate contact hole 26. The barrier insulating layer 62 may be formed by an ALD (Atomic Layer Deposition) method. The processing temperature may be 300° C. or more and 600° C. or less.
[0117] After the step of forming the gate insulating layer 31, an annealing treatment may be performed on the barrier insulating layer 62 and the gate insulating layer 31. The annealing treatment may be performed at a temperature of 500° C. or higher and 900° C. or lower, as long as the barrier insulating layer 62 and the gate insulating layer 31 are not crystallized.
[0118] As described above, according to the semiconductor device 61, amorphous Al is formed between the top barrier layer 13 and the gate insulating layer 31. γ Ga (1-γ) A barrier insulating layer 62 containing N (0<γ≦1) is interposed. This makes it possible to suppress oxidation of the top barrier layer 13.
[0119] Furthermore, the barrier insulating layer 62 can suppress the generation of Ga-O bonds in the region between the top barrier layer 13 and the gate insulating layer 31. Ga-O bonds are known to function as charge traps. By suppressing the generation of Ga-O bonds, the accumulation of charge traps over time can be suppressed. This can suppress undesirable fluctuations (increases) over time of the gate threshold voltage Vth.
[0120] The barrier insulating layer 62 may be applied to the semiconductor device 51. Also, the cap layer 52 may be applied to the semiconductor device 61. That is, the semiconductor device 51 and the semiconductor device 61 may include both the cap layer 52 and the barrier insulating layer 62 in addition to the structure of the semiconductor device 1.
[0121] 7 is a cross-sectional view showing a semiconductor device 101 according to a fourth embodiment of the present invention, and FIG 8 is an enlarged view of a region VII shown in FIG 7.
[0122] The semiconductor device 101 has a basic configuration including a HEMT (High Electron Mobility Transistor) including a group III nitride semiconductor.
[0123] Referring to FIG. 7, the semiconductor device 101 includes a substrate 102. The substrate 102 may be a Si substrate, a SiC substrate, a sapphire substrate, a GaN substrate, or the like. In this form, the substrate is made of a Si substrate. The substrate 102 includes a first main surface 103 on one side and a second main surface 104 on the other side.
[0124] On the first main surface 103 of the substrate 102, a buffer layer 105, an electron traveling layer 106, and an electron supply layer 107 are formed in this order. The buffer layer 105 may contain AlN. The electron traveling layer 106 is made of a group-III nitride semiconductor containing Ga. More specifically, the electron traveling layer 106 contains a In b Ga (1-a-b) N (0 ≦ a + b ≦ 1). In this form, the electron traveling layer 106 is made of GaN. The thickness of the electron traveling layer 106 may be 0.1 μm or more and 3.0 μm or less.
[0125] The electron supply layer 107 contains crystalline Al X Ga (1-X) N (0 < X ≦ 1). The Al composition ratio X is 1 in this form. That is, the electron supply layer 107 is made of crystalline AlN. The thickness of the electron supply layer 107 may be 1 nm or more and 5 nm or less (for example, about 2 nm).
[0126] An opening 108 for exposing the electron traveling layer 106 is formed in the electron supply layer 107. The electron traveling layer 106 has an exposed portion exposed from the opening 108 and a covered portion covered by the electron supply layer 107. The exposed portion and the covered portion of the electron traveling layer 106 form an integrally continuous flat surface. The exposed portion of the electron traveling layer 106 is not formed to be recessed by one step toward the substrate 102 with respect to the covered portion of the electron traveling layer 106.
[0127] A two-dimensional electron gas (2DEG) is formed in the surface layer of the electron transit layer 106 in the boundary region between the electron transit layer 106 and the electron supply layer 107. The 2DEG is formed by piezoelectric polarization and spontaneous polarization of the electron supply layer 107. The piezoelectric polarization is formed due to lattice mismatch between the electron transit layer 106 and the electron supply layer 107.
[0128] On the other hand, in the portion where the opening 108 is formed in the electron supply layer 107, the electron transit layer 106 is not in contact with the electron supply layer 107, and therefore, the 2DEG is not formed. Therefore, the semiconductor device 101 has a normally-off structure.
[0129] 7 and 8, a voltage-resistant insulating layer 110 (insulating layer) is provided on the electron supply layer 107. In this embodiment, the voltage-resistant insulating layer 110 has a laminated structure including a planarizing layer 111, a passivation layer 112, and a spacer layer 113, which are laminated in this order from above the electron supply layer 107.
[0130] The planarization layer 111 covers the electron supply layer 107. The planarization layer 111 is formed on the electron supply layer 107 to improve planarization. The planarization layer 111 may contain GaN. The thickness of the planarization layer 111 may be 1 nm or more and 5 nm or less (for example, about 2 nm).
[0131] The passivation layer 112 covers the planarization layer 111. The passivation layer 112 may contain SiN. The passivation layer 112 prevents charge trapping and maintains the insulating properties of the breakdown voltage retaining insulating layer 110. The thickness of the passivation layer 112 may be 1 nm or more and 40 nm or less (for example, about 25 nm).
[0132] The spacer layer 113 covers the passivation layer 112. The thickness of the spacer layer 113 is greater than the thickness of the planarization layer 111 and the thickness of the passivation layer 112. The spacer layer 113 is formed to separate the gate electrode layer 131, which will be described later, from the electron supply layer 107. The spacer layer 113 may contain SiO2. The thickness of the spacer layer 113 may be 10 nm or more and 100 nm or less (for example, about 30 nm).
[0133] A through hole 114 communicating with the opening 108 of the electron supply layer 107 is formed in the breakdown voltage holding insulating layer 110. The through hole 114 forms one contact hole 115 with the opening 108.
[0134] The contact hole 115 is formed in a tapered shape in which the opening area is larger than the bottom area in a cross-sectional view. The opening edge portion of the contact hole 115 has a curved surface facing into the contact hole 115.
[0135] Referring to FIG. 7, a barrier insulating layer 121 is formed on the electron traveling layer 106 at the opening 108 of the electron supply layer 107. The barrier insulating layer 121 is amorphous Al Y Ga (1-Y) N (0 < Y ≤ 1) is included. The Al composition ratio Y is 1 in this form. That is, the barrier insulating layer 121 is made of amorphous AlN.
[0136] The thickness of the barrier insulating layer 121 may be less than or equal to the thickness of the electron supply layer 107, or may be greater than or equal to the thickness of the electron supply layer 107. The thickness of the barrier insulating layer 121 may be 1 nm or more and 5 nm or less (for example, about 1.5 nm).
[0137] The barrier insulating layer 121 is not piezoelectric. Therefore, at the boundary region between the electron traveling layer 106 and the barrier insulating layer 121, a 2DEG is not formed in the surface layer portion of the electron traveling layer 106. The barrier insulating layer 121 suppresses the bonding of gallium atoms contained in the electron supply layer 107 with oxygen atoms.
[0138] The barrier insulating layer 121 contacts the electron supply layer 107 at the opening 108 of the electron supply layer 107. As a result, the electron supply layer 107 made of crystalline AlN and the barrier insulating layer 121 made of amorphous AlN are located on the same plane.
[0139] The barrier insulating layer 121 is further extended in the form of a film onto the voltage-holding insulating layer 110 via the inner wall of the contact hole 115. This causes the barrier insulating layer 121 to cover the surface of the voltage-holding insulating layer 110. In this embodiment, the barrier insulating layer 121 covers almost the entire surface of the voltage-holding insulating layer 110.
[0140] A gate insulating layer 122 is formed on the barrier insulating layer 121 in the opening 108 of the electron supply layer 107. The gate insulating layer 122 includes an oxide insulating material. The gate insulating layer 122 preferably includes an oxide insulating material in an amorphous state. The thickness of the gate insulating layer 122 may be 5 nm or more and 40 nm or less (for example, about 15 nm).
[0141] The oxide insulating material includes at least one of SiO2, SiON, Al2O3, HfSiO, or HfO2. The gate insulating layer 122 may have a single-layer structure including a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, or a HfO2 layer. The gate insulating layer 122 may have a multilayer structure including at least one of a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, or a HfO2 layer.
[0142] The gate insulating layer 122 may contain SiN in addition to the oxide insulating material. In this case, the gate insulating layer 122 may have a laminated structure including at least one oxide insulating material layer selected from the group consisting of a SiO2 layer, a SiON layer, an Al2O3 layer, a HfSiO layer, and a HfO2 layer, and a SiN layer. In this embodiment, the gate insulating layer 122 is made of an amorphous SiO2 layer.
[0143] The gate insulating layer 122 is extended in the form of a film on the barrier insulating layer 121 and on the voltage-resistant insulating layer 110 via the inner wall of the contact hole 115. As a result, the gate insulating layer 122 faces the electron supply layer 107 with the voltage-resistant insulating layer 110 in between. The gate insulating layer 122 also covers the surface of the voltage-resistant insulating layer 110 via the barrier insulating layer 121.
[0144] In this embodiment, the gate insulating layer 122 covers almost the entire surface of the voltage-resistance insulating layer 110. A mixed crystal layer (not shown) in which the insulating material of the barrier insulating layer 121 and the insulating material of the gate insulating layer 122 are mixed may be formed in the boundary region between the barrier insulating layer 121 and the gate insulating layer 122. This mixed crystal layer may contain an oxide insulator.
[0145] An interlayer insulating layer 123 is formed on the gate insulating layer 122. The interlayer insulating layer 123 may cover almost the entire surface of the gate insulating layer 122. The interlayer insulating layer 123 may have a single-layer structure including a single insulating material layer. The interlayer insulating layer 123 may have a layered structure in which multiple insulating material layers are layered. The interlayer insulating layer 123 may include a SiO2 layer and / or a SiN layer.
[0146] A gate opening 124, a drain opening 125, and a source opening 126 are formed at intervals in the interlayer insulating layer 123. The gate opening 124 is formed in a region of the interlayer insulating layer 123 facing the opening 108 of the electron supply layer 107. The gate opening 124 penetrates the interlayer insulating layer 123 to expose the gate insulating layer 122.
[0147] The drain opening 125 and the source opening 126 are each formed in a region facing the electron supply layer 107 in the interlayer insulating layer 123. The drain opening 125 and the source opening 126 penetrate the interlayer insulating layer 123, the gate insulating layer 122, and the voltage-resistant insulating layer 110, and each expose the electron supply layer 107.
[0148] A gate electrode layer 131 is buried in the gate opening 124. The gate electrode layer 131 has a laminated structure including a gate underlayer 132 and a gate buried layer 133. The gate underlayer 132 may include a TiN layer. The gate buried layer 133 may include a W (tungsten) layer.
[0149] The gate underlayer 132 is formed in a film shape along the inner wall of the gate opening 124 so as to define a concave space within the gate opening 124. The gate buried layer 133 fills the concave space defined by the gate underlayer 132.
[0150] The gate electrode layer 131 has an exposed portion exposed from the gate opening 124. The exposed portion of the gate electrode layer 131 is covered with a barrier electrode layer 134. The barrier electrode layer 134 may have a layered structure including a Ti layer, a TiN layer, and an AlCu layer layered in this order from above the gate electrode layer 131.
[0151] A drain electrode layer 141 is buried in the drain opening 125. The drain electrode layer 141 has a laminated structure including a drain underlayer 142 and a drain buried layer 143. The drain underlayer 142 may include a Ti layer. The drain buried layer 143 may include an Al layer.
[0152] The drain base layer 142 is formed in a film shape along the inner wall of the drain opening 125 so as to define a concave space within the drain opening 125. The drain buried layer 143 fills the concave space defined by the drain base layer 142.
[0153] A source electrode layer 151 is buried in the source opening 126. The source electrode layer 151 has a layered structure including a source underlayer 152 and a source buried layer 153. The source underlayer 152 may include a Ti layer. The source buried layer 153 may include an Al layer.
[0154] The source underlayer 152 is formed in a film shape along the inner wall of the source opening 126 so as to define a concave space within the source opening 126. The source buried layer 153 fills the concave space defined by the source underlayer 152.
[0155] The distance between the gate electrode layer 131 and the drain electrode layer 141 may be greater than the distance between the gate electrode layer 131 and the source electrode layer 151. The distance between the gate electrode layer 131 and the drain electrode layer 141 may be 2 μm or more and 5 μm or less (for example, about 3.5 μm). The distance between the gate electrode layer 131 and the source electrode layer 151 may be 0.5 μm or more and 2.0 μm or less (for example, about 1.0 μm).
[0156] A field plate 160 may be formed in a region around the gate electrode layer 131 on the surface of the gate insulating layer 122. The field plate 160 relieves electric field concentration on the gate electrode layer 131. In this embodiment, the field plate 160 faces the barrier insulating layer 121 with the gate insulating layer 122 interposed therebetween.
[0157] The field plate 160 is covered with the interlayer insulating layer 123. The field plate 160 is formed with a distance of at least 0.1 μm or more from the gate electrode layer 131. The field plate 160 may include a TiN layer.
[0158] In this embodiment, field plate 160 includes a first field plate 161 and a second field plate 162. First field plate 161 is formed in a region between gate electrode layer 131 and drain electrode layer 141 on the surface of gate insulating layer 122. Second field plate 162 is formed in a region between gate electrode layer 131 and source electrode layer 151 on the surface of gate insulating layer 122.
[0159] In this embodiment, a source voltage is applied to the first field plate 161 and the second field plate 162. That is, the first field plate 161 and the second field plate 162 have the same potential as the source electrode layer 151 in this embodiment.
[0160] Either or both of first field plate 161 and second field plate 162 may be electrically released from gate electrode layer 131, drain electrode layer 141, and source electrode layer 151. In other words, either or both of first field plate 161 and second field plate 162 may be in an electrically floating state.
[0161] The width of the first field plate 161 may be greater than the width of the second field plate 162. The width of the first field plate 161 may be 0.5 μm or more. The width of the first field plate 161 may be less than 0.5 μm.
[0162] The distance between field plate 160 and barrier insulating layer 121 may be adjusted by interposing another insulating layer in the region between field plate 160 and gate insulating layer 122. In this case, the distance between field plate 160 and barrier insulating layer 121 may be not less than 30 μm and not more than 100 μm (for example, about 50 μm).
[0163] Fig. 9 is a graph showing the time characteristic of the gate threshold voltage Vth of the semiconductor device 101 shown in Fig. 7. The gate threshold voltage Vth is the gate-source voltage VGS at which the drain current ID starts to flow.
[0164] The graph shown in Figure 9 was obtained by a High Temperature Gate Bias (HTGB) simulation test. In this test, the change in the gate threshold voltage Vth when a given gate-source voltage VGS (for example, about +5V) is applied continuously for 100 hours at a temperature of 150°C is examined.
[0165] 9 shows a first characteristic S1, a second characteristic S2, and a third characteristic S3 (see solid lines). The first characteristic S1, the second characteristic S2, and the third characteristic S3 respectively show the characteristics over time of the gate threshold voltage Vth of the semiconductor device 101. 9 shows a fourth characteristic S4, a fifth characteristic S5, and a sixth characteristic S6 (see dashed lines). The fourth characteristic S4, the fifth characteristic S5, and the sixth characteristic S6 respectively show the characteristics over time of the gate threshold voltage Vth of the semiconductor device according to the reference example.
[0166] The semiconductor device according to the reference example has almost the same structure as the semiconductor device 101, except that it does not have the barrier insulating layer 121. A detailed description of the semiconductor device according to the reference example will be omitted.
[0167] The first characteristic S1, the second characteristic S2, and the third characteristic S3 show the time characteristics of the gate threshold voltage Vth when the barrier insulating layer 121 has a thickness of about 3 nm and the gate insulating layer 122 has a thickness of about 15 nm. The fourth characteristic S4, the fifth characteristic S5, and the sixth characteristic S6 show the time characteristics of the gate threshold voltage Vth when the gate insulating layer 122 has a thickness of about 15 nm.
[0168] 9, the values before 10 hours are not stable, so the following will discuss the range from 10 hours to 100 hours.
[0169] With reference to the first characteristic S1, when the gate-source voltage VGS was applied continuously for 10 hours, the gate threshold voltage Vth was 0.2 V, and when the gate-source voltage VGS was applied continuously for 100 hours, the gate threshold voltage Vth was 0.35 V. The variation rate of the gate threshold voltage Vth was 175%.
[0170] With reference to the second characteristic S2, the gate threshold voltage Vth when the gate-source voltage VGS was applied continuously for 10 hours was 0.3 V, and the gate threshold voltage Vth when the gate-source voltage VGS was applied continuously for 100 hours was 0.45 V. The variation rate of the gate threshold voltage Vth was 150%.
[0171] Referring to the third characteristic S3, when the gate-source voltage VGS was continuously applied for 10 hours, the gate threshold voltage Vth was 0.3V, and when it was continuously applied for 100 hours, the gate threshold voltage Vth was 0.5V. The variation rate of the gate threshold voltage Vth was 167%.
[0172] From the above results, it was found that according to the semiconductor device 101, when a predetermined gate-source voltage VGS was continuously applied for 100 hours, the variation rate of the gate threshold voltage Vth could be suppressed to 180% or less.
[0173] On the other hand, referring to the fourth characteristic S4, the fifth characteristic S5, and the sixth characteristic S6, according to the semiconductor device according to the reference example, when a predetermined gate-source voltage VGS was continuously applied for 100 hours, the variation rate of the gate threshold voltage Vth was 200% or more.
[0174] As described above, according to the semiconductor device 101, an amorphous Al Y Ga (1-Y) N (0 < Y ≤ 1) is formed between the electron transport layer 106 containing Ga and the gate insulating layer 122 containing an oxide insulating material.
[0175] Thereby, it is possible to suppress the electron transport layer 106 and the gate insulating layer 122 from coming into contact with each other. As a result, in the region between the electron transport layer 106 and the gate insulating layer 122, the formation of a Ga-O bond formed by the bonding of gallium atoms and oxygen atoms can be suppressed. Therefore, an increase in the gate threshold voltage Vth over time can be suppressed (see FIG. 9).
[0176] In addition, since the barrier insulating layer 121 is in an amorphous state, the risk of crystal defects can be reduced. Therefore, the gate leakage current caused by the crystal defects of the barrier insulating layer 121 can be suppressed.
[0177] FIGS. 10A to 10M are cross-sectional views showing an example of the manufacturing method of the semiconductor device 101 shown in FIG. 7.
[0178] 10A, first, a substrate 102 is prepared. In this embodiment, the substrate 102 is a Si substrate. Next, a buffer layer 105, an electron transit layer 106, and an electron supply layer 107 are formed in this order on a first main surface 103 of the substrate 102. The buffer layer 105, the electron transit layer 106, and the electron supply layer 107 are each formed by epitaxial growth.
[0179] More specifically, the buffer layer 105 is formed by epitaxially growing AlN on the substrate 102. The electron transit layer 106 is formed by epitaxially growing Al a In b Ga (1-a-b) The electron supply layer 107 is formed by epitaxially growing N (GaN in this embodiment). X Ga (1-X) It is formed by epitaxially growing N (in this form, AlN).
[0180] 10B, a voltage-resistant insulating layer 110 is formed on the electron supply layer 107. The process of forming the voltage-resistant insulating layer 110 includes the process of forming a planarizing layer 111, a passivation layer 112, and a spacer layer 113 in this order on the electron supply layer 107 (also see FIG. 8).
[0181] The planarization layer 111 is formed by epitaxially growing GaN on the electron transport layer 106. The passivation layer 112 may be formed by a CVD method. The passivation layer 112 may contain SiN. The spacer layer 113 may be formed by a CVD method. The spacer layer 113 may contain SiO2.
[0182] 10C, a mask 171 having a predetermined pattern is formed on the voltage-resistant insulating layer 110. The mask 171 may be a resist mask made of a photosensitive resin. The mask 171 has an opening 172 that exposes a region where the through hole 114 is to be formed.
[0183] Next, unnecessary portions of voltage-holding insulating layer 110 are removed. The unnecessary portions of voltage-holding insulating layer 110 may be removed by an etching method using mask 171. As a result, through holes 114 are formed in voltage-holding insulating layer 110. After through holes 114 are formed, mask 171 is removed.
[0184] 10D, a portion of the electron supply layer 107 exposed from the through hole 114 of the voltage-holding insulating layer 110 is oxidized by an oxidation treatment method. As a result, an oxide 173 of the electron supply layer 107 is formed in the through hole 114 of the voltage-holding insulating layer 110. The oxide 173 may contain AlON or Al2O3.
[0185] The oxidation treatment method may be a plasma oxidation treatment method. The plasma oxidation treatment method is performed in an oxygen gas atmosphere until the portions of the electron supply layer 107 exposed from the through holes 114 are all oxidized. The treatment temperature may be 100° C. or higher and 900° C. or lower. The treatment time may be 1 hour or higher and 15 hours or lower. The oxygen concentration in the oxygen gas may be about 30%.
[0186] According to the plasma oxidation method, once the oxide 173 is formed, oxygen in the atmosphere does not or barely penetrates into the electron transit layer 106. Therefore, the surface of the electron transit layer 106 is not or barely oxidized.
[0187] 10E, the oxide 173 is removed. The oxide 173 may be removed by an etching method (for example, a wet etching method). The etching liquid may be a sulfuric acid hydrogen peroxide mixture (SPM) containing sulfuric acid and hydrogen peroxide water.
[0188] The oxide 173 has an etching selectivity different from that of the electron supply layer 107. Therefore, when removing the oxide 173, the electron supply layer 107 is hardly removed. As a result, an opening 108 communicating with the through hole 114 is formed in the electron supply layer 107. The opening 108 forms one contact hole 115 with the through hole 114.
[0189] The oxide 173 does not contain Ga (gallium). The oxide 173 has a different etching selectivity from the electron transit layer 106. Therefore, when the oxide 173 is removed, almost no part of the electron transit layer 106 is removed. As a result, the portion of the electron transit layer 106 that forms the bottom wall of the opening 108 and the portion of the electron transit layer 106 that contacts the electron supply layer 107 form flat surfaces that are flat with each other.
[0190] 10F, a barrier insulating layer 121 is formed on the portion of the electron transit layer 106 exposed from the opening 108. In this embodiment, the barrier insulating layer 121 is formed so as to extend in the form of a film onto the voltage-resistant insulating layer 110 via the inner wall of the contact hole 115. The barrier insulating layer 121 is made of amorphous Al Y Ga (1-Y) N (AlN in this embodiment). The barrier insulating layer 121 may be formed by ALD (Atomic Layer Deposition). The processing temperature may be 300° C. or more and 600° C. or less.
[0191] Next, a gate insulating layer 122 is formed on the barrier insulating layer 121. In this embodiment, the gate insulating layer 122 is formed so as to be drawn out in a film form onto the voltage-retaining insulating layer 110 via the inner wall of the contact hole 115. In this embodiment, the gate insulating layer 122 includes SiO2 in an amorphous state. The gate insulating layer 122 may be formed by an ALD method.
[0192] Next, an annealing process is performed on the barrier insulating layer 121 and the gate insulating layer 122. The annealing process may be performed at a temperature of 500° C. or higher and 900° C. or lower, as long as the barrier insulating layer 121 and the gate insulating layer 122 are not crystallized.
[0193] 10G, a conductive layer 174 is formed on the gate insulating layer 122. In this embodiment, the conductive layer 174 includes TiN. The conductive layer 174 may be formed by a CVD method.
[0194] 10H, a mask 175 having a predetermined pattern is formed on the conductive layer 174. The mask 175 may be a resist mask made of a photosensitive resin. The mask 175 covers the region where the field plate 160 is to be formed.
[0195] Next, unnecessary portions of the conductive layer 174 are removed. The unnecessary portions of the conductive layer 174 may be removed by an etching method through a mask 175. This forms the field plate 160. The field plate 160 may be formed on another insulating layer after another insulating layer is formed on the gate insulating layer 122.
[0196] 10I, an interlayer insulating layer 123 is formed on the gate insulating layer 122. The interlayer insulating layer 123 covers the field plate 160. In this embodiment, the interlayer insulating layer 123 includes SiO2. The interlayer insulating layer 123 may be formed by a CVD method.
[0197] 10J, a mask 176 having a predetermined pattern is formed on the interlayer insulating layer 123. The mask 176 may be a resist mask made of a photosensitive resin. The mask 176 has an opening 177 that exposes a region of the interlayer insulating layer 123 where the gate opening 124 is to be formed.
[0198] Next, the unnecessary portion of the interlayer insulating layer 123 is removed. The unnecessary portion of the interlayer insulating layer 123 may be removed by an etching method through the mask 176. This forms a gate opening 124 in the interlayer insulating layer 123. The mask 176 is then removed.
[0199] 10K, a gate electrode layer 131 and a barrier electrode layer 134 are formed. The step of forming the gate electrode layer 131 includes the step of forming a gate underlayer 132 and a gate buried layer 133.
[0200] The gate underlayer 132 is formed in the form of a film along the inner wall of the gate opening 124 so as to define a recessed space within the gate opening 124. The gate underlayer 132 includes a TiN layer. The gate underlayer 132 may be formed by a sputtering method.
[0201] The gate buried layer 133 is formed so as to fill a concave space defined by the gate underlayer 132 within the gate opening 124. The gate buried layer 133 includes a W (tungsten) layer. The gate buried layer 133 may be formed by a CVD method.
[0202] The barrier electrode layer 134 is formed so as to cover an exposed portion of the gate electrode layer 131 exposed from the gate opening 124. In this embodiment, the step of forming the barrier electrode layer 134 includes the steps of forming a Ti layer, a TiN layer, and an AlCu layer in this order on the gate electrode layer 131. The Ti layer, the TiN layer, and the AlCu layer may each be formed by a sputtering method.
[0203] 10L, a mask 178 having a predetermined pattern is formed on the interlayer insulating layer 123. The mask 178 may be a resist mask made of a photosensitive resin. The mask 178 has an opening 179 that exposes a region in the interlayer insulating layer 123 where the drain opening 125 and the source opening 126 are to be formed.
[0204] Next, the unnecessary portions of the interlayer insulating layer 123 are removed. The unnecessary portions of the interlayer insulating layer 123 may be removed by an etching method through the mask 178. This forms the drain opening 125 and the source opening 126 in the interlayer insulating layer 123. The mask 178 is then removed.
[0205] 10M, the drain electrode layer 141 and the source electrode layer 151 are formed. The steps of forming the drain electrode layer 141 and the source electrode layer 151 include the steps of forming the drain base layer 142 and the source base layer 152, and the drain buried layer 143 and the source buried layer 153, respectively.
[0206] The drain underlayer 142 and the source underlayer 152 are formed in the form of a film along the inner walls of the drain opening 125 and the source opening 126, respectively, so that concave spaces are defined in the drain opening 125 and the source opening 126. The drain underlayer 142 and the source underlayer 152 each include a Ti layer. The drain underlayer 142 and the source underlayer 152 may be formed simultaneously by a sputtering method.
[0207] The drain buried layer 143 and the source buried layer 153 are formed so as to fill the concave spaces defined by the drain underlayer 142 and the source underlayer 152 in the drain opening 125 and the source opening 126, respectively.
[0208] The drain buried layer 143 and the source buried layer 153 each include an Al layer. The drain buried layer 143 and the source buried layer 153 may be formed simultaneously by a CVD method. Through the steps including those described above, the semiconductor device 101 is manufactured.
[0209] Although the embodiment of the present invention has been described above, the present invention can be embodied in other forms.
[0210] For example, in the third embodiment, the barrier insulating layer 62 is made of amorphous Al γ Ga (1-γ)Examples involving N(AlN) have been described.
[0211] However, the barrier insulating layer 62 is made of amorphous Al. γ Ga (1-γ) In place of or in addition to N(AlN), the barrier insulating layer 62 may contain amorphous SiN. γ Ga (1-γ) Instead of N(AlN), a single layer structure made of an amorphous SiN layer may be used.
[0212] The barrier insulating layer 62 is made of an amorphous SiN layer and an amorphous Al layer formed on the amorphous SiN layer. γ Ga (1-γ) It may have a laminated structure including an N(AlN) layer.
[0213] The barrier insulating layer 62 is made of amorphous Al. γ Ga (1-γ) N (AlN) layer and amorphous Al γ Ga (1-γ) It may have a laminated structure including an amorphous SiN layer formed on an N(AlN) layer.
[0214] In addition, various design modifications can be made within the scope of the claims.
[0215] Below, examples of features extracted from this specification and the drawings (FIGS. 7 to 9 and 10A to 10M) are shown.
[0216] Patent document (JP2011-192834) discloses a semiconductor device equipped with a HEMT (High Electron Mobility Transistor). This semiconductor device includes a GaN layer (electron transit layer). An AlGaN layer (electron supply layer) is formed on the GaN layer.
[0217] The AlGaN layer has an opening that exposes the GaN layer. A gate oxide layer containing SiO2 is formed within the opening of the AlGaN layer. A gate electrode layer is formed on the gate oxide layer.
[0218] In a semiconductor device including a HEMT, there is a problem that the gate threshold voltage increases over time. The inventor of the present application has found that a structure in which a gate oxide layer is formed on an electron transport layer contributes to this problem.
[0219] That is, in a structure in which a gate oxide layer is formed on an electron transport layer, a Ga-O bond formed by the bonding of gallium atoms and oxygen atoms is generated in the boundary region between the electron transport layer and the gate oxide layer. This Ga-O bond functions as a charge trap.
[0220] Therefore, the charges flowing through the electron transport layer are trapped and accumulated by the Ga-O bond. As a result, trap levels are formed in the boundary region between the electron transport layer and the gate oxide layer, causing an increase in the gate threshold voltage over time.
[0221] Therefore, hereinafter, a semiconductor device and a manufacturing method thereof that can suppress an increase in the gate threshold voltage over time will be provided.
[0222] [A1] An electron transport layer made of a nitride semiconductor containing Ga, crystalline Al X Ga (1-X) N (0 < X ≤ 1), formed on the electron transport layer and having an opening that exposes the electron transport layer, an electron supply layer, and amorphous Al Y Ga (1-Y) N (0 < Y ≤ 1), a barrier insulating layer formed on the electron transport layer within the opening of the electron supply layer, a gate insulating layer containing an oxide insulating material and formed on the barrier insulating layer, and a gate electrode layer formed on the gate insulating layer.
[0223] According to this semiconductor device, amorphous Al Y Ga(1-Y) A barrier insulating layer containing N(0 < Y ≤ 1) is interposed between the electron transport layer and the gate insulating layer. Thereby, the formation of Ga - O bonds in the region between the electron transport layer and the gate insulating layer can be suppressed by the barrier insulating layer. As a result, the increase in the gate threshold voltage over time can be suppressed.
[0224] [A2] The semiconductor device according to A1, wherein the thickness of the gate insulating layer is larger than the thickness of the barrier insulating layer.
[0225] [A3] The semiconductor device according to A1 or A2, further including a mixed crystal layer formed in a boundary region between the barrier insulating layer and the gate insulating layer, in which the insulating materials of the barrier insulating layer and the gate insulating layer are mixed - crystallized.
[0226] [A4] The semiconductor device according to any one of A1 to A3, further including an insulating layer that covers the electron supply layer, communicates with the opening of the electron supply layer, and has a through - hole formed therein to form one contact hole with the opening of the electron supply layer, and the gate electrode layer enters the contact hole from above the insulating layer.
[0227] [A5] The semiconductor device according to A4, wherein the insulating layer includes a passivation layer formed on the electron supply layer and a spacer layer formed on the passivation layer to separate the gate electrode layer from the electron supply layer.
[0228] [A6] The semiconductor device according to A5, wherein the thickness of the spacer layer is larger than the thickness of the passivation layer.
[0229] [A7] The semiconductor device according to A5 or A6, wherein the passivation layer contains SiN and the spacer layer contains SiO2.
[0230] [A8] A step of preparing an electron transport layer made of a nitride semiconductor containing Ga, and crystalline Al X Ga (1-X)Forming an electron supply layer containing N(0 < X ≤ 1) and having an opening exposing the electron transport layer on the electron transport layer; and in the opening of the electron supply layer, amorphous Al Y Ga (1-Y) Forming a barrier insulating layer containing N(0 < Y ≤ 1) on the electron transport layer; forming a gate insulating layer containing an oxide insulating material on the barrier insulating layer; and forming a gate electrode layer on the gate insulating layer, a method for manufacturing a semiconductor device.
[0231] According to this manufacturing method, a barrier insulating layer containing amorphous Al Y Ga (1-Y) N(0 < Y ≤ 1) is formed between the electron transport layer and the gate insulating layer. Thereby, the formation of Ga - O bonds in the region between the electron transport layer and the gate insulating layer can be suppressed by the barrier insulating layer. As a result, a semiconductor device capable of suppressing the increase in the gate threshold voltage over time can be manufactured.
[0232] [A9] The method for manufacturing a semiconductor device according to A8, wherein the gate insulating layer having a thickness greater than the thickness of the barrier insulating layer is formed.
[0233] [B1] A back barrier layer containing Al X Ga (1-X) N(0 < X ≤ 1); an electron transport layer containing Al a In b Ga (1-a-b) N(0 ≤ a + b ≤ 1) and formed on the back barrier layer; a top barrier layer containing Al Y Ga (1-Y) N(0 < Y ≤ 1) and formed on the electron transport layer; an electron supply layer containing Al Z Ga (1-Z) N(0 < Z ≤ 1), formed on the top barrier layer and having an opening exposing the top barrier layer; a two - dimensional electron gas region formed in a region of the surface layer portion of the electron transport layer facing the electron supply layer with the top barrier layer interposed therebetween; and a gate electrode layer facing the electron transport layer with the gate insulating layer formed in the opening of the electron supply layer interposed therebetween, a semiconductor device.
[0234] [B2] The semiconductor device according to B1, wherein the electron supply layer has a lattice constant such that a conduction band energy level of a region of the surface portion of the electron transit layer that faces the electron supply layer across the top barrier layer is equal to or higher than the Fermi energy level, and a conduction band energy level of a region of the surface portion of the electron transit layer that does not face the electron supply layer across the top barrier layer is lower than the Fermi energy level.
[0235] [B3] The semiconductor device according to B1 or B2, wherein the electron supply layer has a lattice constant smaller than a lattice constant of the top barrier layer.
[0236] [B4] The semiconductor device according to any one of B1 to B3, wherein the electron transport layer has a lattice constant larger than a lattice constant of the back barrier layer.
[0237] [B5] The semiconductor device according to any one of B1 to B4, wherein the top barrier layer has a lattice constant smaller than a lattice constant of the electron transport layer.
[0238] [B6] The semiconductor device according to any one of B1 to B5, wherein the electron transport layer is made of GaN.
[0239] [B7] The semiconductor device according to any one of B1 to B6, wherein the electron supply layer is made of AlN.
[0240] [B8] The semiconductor device according to any one of B1 to B7, further comprising a cap layer made of a Group III nitride semiconductor not containing Al, interposed in a region between the top barrier layer and the electron supply layer.
[0241] [B9] The semiconductor device according to B8, wherein the cap layer is made of GaN.
[0242] The semiconductor device according to any one of B1 to B9, further including a barrier insulating layer made of an insulating material other than an oxide insulating material and intervening in a region between the top barrier layer and the gate insulating layer.
[0243] [B11] The barrier insulating layer is amorphous Al γ Ga (1-γ) N (0 <γ ≦ 1), the semiconductor device according to B10.
[0244] [B12] In the barrier insulating layer, γ is 1, the semiconductor device according to B10.
[0245] [B13] The barrier insulating layer includes amorphous SiN, the semiconductor device according to any one of B10 to B12.
[0246] [B14] A step of forming an electron traveling layer containing Al X Ga (1-X) N (0 <X ≦ 1) on a back barrier layer, and a step of forming an electron traveling layer containing Al a In b Ga (1-a-b) N (0 ≦ a + b ≦ 1), and on the electron traveling layer, a step of forming a top barrier layer containing Al Y Ga (1-Y) N (0 <Y ≦ 1) such that the conduction band energy level at the interface between the top barrier layer and the electron traveling layer is higher than the Fermi energy level, and on the top barrier layer, a step of forming a top barrier layer containing Al Z Ga (1-Z)A step of forming an electron supply layer including N (0 < Z ≤ 1); a step of oxidizing a part of the electron supply layer to form an oxide, and removing the oxide to form an opening for exposing the top barrier layer in the electron supply layer; a step of making the conduction band energy level of a region facing the opening at the interface between the top barrier layer and the electron traveling layer larger than the Fermi energy level; a step of forming a gate insulating layer on a portion of the top barrier layer exposed from the opening of the electron supply layer; and a step of forming a gate electrode layer on the gate insulating layer. A method for manufacturing a semiconductor device.
[0247] [B15] The step of forming the oxide includes a step of oxidizing a part of the electron supply layer by a plasma oxidation method, and the step of removing the oxide includes a step of removing the oxide by a wet etching method. The method for manufacturing a semiconductor device according to B14.
[0248] [B16] The oxide includes AlON or Al2O3. The method for manufacturing a semiconductor device according to B14 or B15.
[0249] [B17] The step of forming the electron traveling layer includes a step of forming the electron traveling layer made of GaN. The method for manufacturing a semiconductor device according to any one of B14 to B16.
[0250] [B18] The step of forming the electron supply layer includes a step of forming the electron supply layer made of AlN. The method for manufacturing a semiconductor device according to any one of B14 to B17.
[0251] [B19] After the step of forming the top barrier layer and prior to the step of forming the electron supply layer, the method further includes a step of forming a cap layer made of a group III nitride semiconductor不含 Al on the top barrier layer. The method for manufacturing a semiconductor device according to any one of B14 to B18.
[0252] [B20] The method for manufacturing a semiconductor device according to any one of B14 to B19, further comprising, after the step of forming the top barrier layer and prior to the step of forming the gate insulating layer, a step of forming a barrier insulating layer made of an insulating material other than an oxide insulating material on a portion of the top barrier layer exposed from the opening of the electron supply layer, and the step of forming the gate insulating layer includes a step of forming the gate insulating layer on the barrier insulating layer. [Explanation of symbols]
[0253] 1 Semiconductor device 8 Electron supply layer 11 Back Barrier Layer 12 Electron transport layer 13 Top Barrier Layer 14 Two-dimensional electron gas region 15 Opening of electron supply layer 31 Gate insulating layer 32 Gate electrode layer 51 Semiconductor Devices 61 Semiconductor Devices EC Conduction band energy level EF Fermi Energy Level
Claims
1. A a In b G (1-a-b) an electron transit layer including N (0≦a+b≦1); A x G (1-x) N (0<X≦1), the electron supply layer being formed on the electron transit layer and having an opening exposing the electron transit layer; a two-dimensional electron gas region formed in a surface portion of the electron transit layer in a region facing the electron supply layer; an insulating layer laminated on the electron supply layer; a tapered through hole defined in the insulating layer so as to communicate with the opening of the electron supply layer and form a contact hole with the opening; a barrier insulating layer that covers a wall surface of the contact hole and the electron transit layer within the contact hole and that covers a surface of the insulating layer outside the contact hole; SiO 2 a gate insulating layer that covers the barrier insulating layer inside the contact hole and faces a surface of the insulating layer outside the contact hole with the barrier insulating layer therebetween; a gate electrode layer disposed in the contact hole and facing the electron transit layer with the gate insulating layer interposed therebetween; the contact hole has, in a cross-sectional view, a lower portion which is defined by the electron supply layer and has a constant inner diameter in a region below a depth position of an intermediate portion of the contact hole, and an upper portion which is defined by the insulating layer and has an inner diameter which gradually increases from the inner diameter of the lower portion in a region above a depth position of the intermediate portion and which is smoothly connected to an upper surface of the insulating layer; an opening edge portion of the contact hole has a curved surface directed toward the inside of the contact hole; the gate insulating layer includes, in the contact hole, a first region extending along a sidewall of the contact hole with the barrier insulating layer therebetween, and a second region extending along the electron transit layer with the barrier insulating layer therebetween; A semiconductor device, wherein the thickness of the first region of the gate insulating layer is equal to or greater than the thickness of the second region of the gate insulating layer.
2. An electron transit layer including Al a In b Ga (1-a-b) N (0≦a+b≦1), an electron supply layer including Al x Ga (1-x) N (0<x≦1), formed on the electron transit layer, and having an opening exposing the electron transit layer; a two-dimensional electron gas region formed in a surface portion of the electron transit layer in a region facing the electron supply layer; an insulating layer laminated on the electron supply layer; a tapered through hole defined in the insulating layer so as to communicate with the opening in the electron supply layer and form a contact hole with the opening; a barrier insulating layer that covers a wall surface of the contact hole and the electron transit layer within the contact hole and that covers a surface of the insulating layer outside the contact hole; a gate insulating layer including SiO 2 , covering the barrier insulating layer within the contact hole and facing a surface of the insulating layer outside the contact hole with the barrier insulating layer therebetween; a gate electrode layer disposed in the contact hole and facing the electron transit layer with the gate insulating layer interposed therebetween; the insulating layer includes a planarization layer, a passivation layer, and a spacer layer stacked in this order on the electron supply layer; the contact hole has, in a cross-sectional view, a lower portion which is defined by the electron supply layer and has a constant inner diameter in a region below a depth position of an intermediate portion of the contact hole, and an upper portion which is defined by the insulating layer and has an inner diameter which gradually increases from the inner diameter of the lower portion in a region above a depth position of the intermediate portion and which is smoothly connected to an upper surface of the insulating layer; the gate insulating layer includes, in the contact hole, a first region extending along a sidewall of the contact hole with the barrier insulating layer therebetween, and a second region extending along the electron transit layer with the barrier insulating layer therebetween; A semiconductor device, wherein the thickness of the first region of the gate insulating layer is equal to or greater than the thickness of the second region of the gate insulating layer.
3. The semiconductor device according to claim 2 , wherein the thickness of said spacer layer is greater than the thickness of said planarization layer and the thickness of said passivation layer.
4. The spacer layer is made of SiO 2 The semiconductor device according to claim 2 or 3, comprising:
5. An electron transit layer including Al a In b Ga (1-a-b) N (0≦a+b≦1), an electron supply layer including Al x Ga (1-x) N (0<x≦1), formed on the electron transit layer, and having an opening exposing the electron transit layer; a two-dimensional electron gas region formed in a surface portion of the electron transit layer in a region facing the electron supply layer; an insulating layer laminated on the electron supply layer; a tapered through hole defined in the insulating layer so as to communicate with the opening in the electron supply layer and form a contact hole with the opening; a barrier insulating layer that covers a wall surface of the contact hole and the electron transit layer within the contact hole and that covers a surface of the insulating layer outside the contact hole; a gate insulating layer including SiO 2 , covering the barrier insulating layer within the contact hole and facing a surface of the insulating layer outside the contact hole with the barrier insulating layer therebetween; a gate electrode layer disposed in the contact hole and facing the electron transit layer with the gate insulating layer interposed therebetween; the contact hole has, in a cross-sectional view, a lower portion which is defined by the electron supply layer and has a constant inner diameter in a region below a depth position of an intermediate portion of the contact hole, and an upper portion which is defined by the insulating layer and has an inner diameter which gradually increases from the inner diameter of the lower portion in a region above a depth position of the intermediate portion and which is smoothly connected to an upper surface of the insulating layer; the gate insulating layer includes, in the contact hole, a first region extending along a sidewall of the contact hole with the barrier insulating layer therebetween, and a second region extending along the electron transit layer with the barrier insulating layer therebetween; the gate electrode layer has a laminated structure including a gate underlayer and a gate buried layer, A semiconductor device, wherein the thickness of the first region of the gate insulating layer is equal to or greater than the thickness of the second region of the gate insulating layer.
6. the gate underlayer includes a TiN layer; 6. The semiconductor device according to claim 5, wherein the gate buried layer includes a W (tungsten) layer.
7. In the surface layer of the electron transport layer, a conduction band energy level of a region facing the electron supply layer is lower than a Fermi energy level; 7. The semiconductor device according to claim 1, wherein a conduction band energy level of said region of said electron supply layer facing said opening is higher than a Fermi energy level.
8. 8. The semiconductor device according to claim 1, wherein the electron transit layer is made of GaN.
9. 9. The semiconductor device according to claim 1, wherein the electron supply layer is made of crystalline AlN.
10. The barrier insulating layer is made of amorphous Al. γ G (1-γ) The semiconductor device according to claim 1 , wherein N (0<γ≦1) is included.
11. The semiconductor device according to claim 10 , wherein the γ is 1.
12. 10. The semiconductor device according to claim 1, wherein the barrier insulating layer contains amorphous SiN.
13. 13. The semiconductor device according to claim 1, further comprising a drain electrode layer and a source electrode layer disposed to sandwich the gate electrode layer in a plan view.
14. The gate insulating layer is made of amorphous SiO 2 The semiconductor device according to any one of claims 1 to 13, comprising:
15. 15. The semiconductor device according to claim 1, further comprising a field plate formed on a surface of said gate insulating layer so as to be positioned around said gate electrode layer.
16. The semiconductor device described in claim 1, wherein the insulating layer includes a planarization layer, a passivation layer, and a spacer layer stacked in this order on the electron supply layer.
17. The semiconductor device of claim 1, wherein the gate electrode layer has a layered structure including a gate underlayer and a gate buried layer.
18. A a In b G (1-a-b) forming an electron transit layer including N (0≦a+b≦1); On the electron transport layer, an Al x G (1-X) forming an electron supply layer including N (0<X≦1); forming a planarization layer, a passivation layer, and a spacer layer in sequence on the electron supply layer, and forming an insulating layer including the planarization layer, the passivation layer, and the spacer layer on the electron supply layer; forming a tapered through hole in the insulating layer to expose a portion of the electron supply layer; a step of oxidizing a portion of the electron supply layer exposed from the through hole of the insulating layer to form an oxide, and removing the oxide to expose the electron transit layer and form an opening defining a contact hole communicating with the through hole, while simultaneously making a conduction band energy level of a region facing the opening on a surface of the electron transit layer higher than a Fermi energy level; forming a barrier insulating layer that covers a wall surface of the contact hole and the electron transit layer within the contact hole and that covers a surface of the insulating layer outside the contact hole; SiO 2 forming a gate insulating layer that covers the barrier insulating layer inside the contact hole and faces a surface of the insulating layer outside the contact hole with the barrier insulating layer therebetween; forming a gate electrode layer on the gate insulating layer in the contact hole; the contact hole is formed so as to include, in a cross-sectional view, a lower portion which is defined by the electron supply layer and has a constant inner diameter in a region below a depth position of an intermediate portion of the contact hole, and an upper portion which is defined by the insulating layer and has an inner diameter which gradually increases from the inner diameter of the lower portion in a region above a depth position of the intermediate portion, and which is smoothly connected to an upper surface of the insulating layer; the gate insulating layer is formed in the contact hole so as to include a first region extending along a sidewall of the contact hole with the barrier insulating layer therebetween, and a second region extending along the electron transit layer with the barrier insulating layer therebetween; A method for manufacturing a semiconductor device, wherein a thickness of the first region of the gate insulating layer is equal to or greater than a thickness of the second region of the gate insulating layer.
19. the step of forming an oxide includes a step of oxidizing a part of the electron supply layer by a plasma oxidation method; 20. The method of claim 18, wherein the step of removing the oxide includes the step of removing the oxide by a wet etching method.
20. The oxide is AlON or Al 2 O 3 The method for manufacturing a semiconductor device according to claim 18 or 19, comprising:
21. 21. The method for manufacturing a semiconductor device according to claim 18, wherein the step of forming the electron transit layer includes the step of forming the electron transit layer made of GaN.
22. 22. The method for manufacturing a semiconductor device according to claim 18, wherein the step of forming the electron supply layer includes the step of forming the electron supply layer made of crystalline AlN.
Citation Information
Patent Citations
Method for fabricating nitride-based transistors with capping layers and buried gates
JP2008507843A
Manufacturing method of compound semiconductor device and compound semiconductor device
JP2011238700A
Always-off semiconductor device and method for fabricating the same
JP2011529639A
MIS type normally-off HEMT element of recess structure having drain current density / transconductance improved greatly
JP2015192004A
Semiconductor device and manufacturing method of the same, power supply device and high-frequency amplifier
JP2016163017A
