JP5256618B2 - Semiconductor device and manufacturing method of semiconductor device - Google Patents

Description

  The present invention relates to a semiconductor device and a method for manufacturing the semiconductor device, and more particularly to a semiconductor device using a nitride compound semiconductor and a method for manufacturing the semiconductor device.

  Nitride-based compound semiconductors such as gallium nitride (GaN) are attracting attention as semiconductor materials having good characteristics in terms of high temperature, high output, and high frequency. In addition, since nitride compound semiconductors have a heterostructure of gallium aluminum nitride (AlGaN) and gallium nitride (GaN), the electron mobility can be increased, so that high-speed switching and high current are required. It is suitable for use in a semiconductor device. Such a nitride-based compound semiconductor is used for, for example, a high electron mobility transistor (HEMT).

  By the way, in a nitride-based compound semiconductor, a large amount of deep levels (traps) exist on the bulk crystal or the semiconductor surface. For this reason, during reverse voltage application period or off period to the semiconductor device, for example, carriers are trapped in the trap in the crystal of the semiconductor substrate, and then when the forward voltage is applied to the semiconductor device or turned on, There is a problem that a so-called current collapse phenomenon occurs in which the output current decreases (for example, Non-Patent Document 1).

Until now, as a method for suppressing the current coplus phenomenon, it has been proposed to provide an insulating film made of SiN _{x on} an exposed portion of a semiconductor layer (for example, Patent Document 1).
Binari, SC, Ikossi, K., Roussos, JA, Kruppa, W. Doewon Park Dietrich, HB, Koleske, DD, Wickenden, AE, Henry, RL, "Trapping effects and microwave power performance in AlGaN / GaN HEMTs", IEEE Transactions on Electron Devices, IEEE, Vol. 48, No. 3, p. 465-471, (March 2001) JP-A-2005-286135

However, the process of forming the insulating film composed of SiN _{x in a} manner that effectively suppresses the current collapse phenomenon is complicated and requires various fine adjustments.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a semiconductor device that can easily and satisfactorily suppress a current collapse phenomenon and a method for manufacturing the same.

In order to achieve the above object, a semiconductor device according to the first aspect of the present invention includes:
A substrate,
A compound semiconductor layer formed on one main surface of the substrate;
A drain electrode formed on the compound semiconductor layer;
A source electrode formed on the compound semiconductor layer;
A gate electrode formed on the compound semiconductor layer between the drain electrode and the source electrode;
An insulating film formed so as to cover at least a part of a region between the drain electrode and the gate electrode in a region where the compound semiconductor layer on the compound semiconductor layer is exposed;
A film containing rod-like molecules disposed on the upper surface of the insulating film;
With
A groove extending in a direction connecting the gate electrode and the drain electrode is provided on the upper surface of the insulating film,
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the gate electrode and the drain electrode,
It is characterized by that.

In order to achieve the above object, a semiconductor device according to the second aspect of the present invention provides:
A substrate,
A compound semiconductor layer formed on one main surface of the substrate;
An anode electrode formed on the compound semiconductor layer;
A cathode electrode formed on the compound semiconductor layer;
An insulating film formed to cover at least a part of a region between the anode electrode and the cathode electrode in a region where the compound semiconductor layer on the compound semiconductor layer is exposed;
A film containing rod-like molecules disposed on the upper surface of the insulating film;
With
A groove extending in a direction connecting the anode electrode and the cathode electrode is provided on the upper surface of the insulating film,
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the anode electrode and the cathode electrode,
It is characterized by that.

  The compound semiconductor layer is made of, for example, a nitride compound semiconductor.

  The compound semiconductor layer includes a first compound semiconductor layer and a second compound semiconductor layer, and a two-dimensional electron gas layer is provided in the vicinity of the interface between the first compound semiconductor layer and the second compound semiconductor layer. It is desirable to have.

  The rod-like molecule is formed, for example, as a single layer film or an amorphous film.

  The rod-like molecule has, for example, a double bond or a benzene ring in the main skeleton.

  The rod-like molecule has, for example, an electron donating group or an electron withdrawing group.

  The rod-like molecule is composed of, for example, an organic compound having π electrons.

In order to achieve the above object, a semiconductor device manufacturing method according to a third aspect of the present invention includes:
A compound semiconductor layer forming step of forming a compound semiconductor layer on the substrate;
An electrode forming step of forming a drain electrode and a source electrode on the compound semiconductor layer, and further forming a gate electrode between the drain electrode and the source electrode;
An insulating film forming step of forming an insulating film so as to cover at least a part of a region between the drain electrode and the gate electrode in a region where the compound semiconductor layer on the upper surface of the compound semiconductor layer is exposed;
A groove forming step of forming a groove extending in a direction connecting the said gate electrode and said drain electrode on the upper surface of the insulating film,
A film forming step of forming a film containing rod-like molecules on the insulating film;
With
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the gate electrode and the drain electrode,
It is characterized by that .

In order to achieve the above object, a semiconductor device manufacturing method according to a fourth aspect of the present invention includes:
A compound semiconductor layer forming step of forming a compound semiconductor layer on the substrate;
An electrode forming step of forming an anode electrode and a cathode electrode on the compound semiconductor layer;
An insulating film forming step of forming an insulating film so as to cover at least part of a region between the anode electrode and the cathode electrode in a region where the compound semiconductor layer on the upper surface of the compound semiconductor layer is exposed;
Forming a groove extending in a direction connecting the anode electrode and the cathode electrode on an upper surface of the insulating film;
A film forming step of forming a film containing rod-like molecules on the insulating film;
With
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the anode electrode and the cathode electrode,
It is characterized by that .

  According to the present invention, the current collapse phenomenon can be easily and satisfactorily suppressed.

(First embodiment)
The semiconductor device according to the first embodiment of the present invention will be described below. In this embodiment mode, the present invention is described using a semiconductor device including a high electron mobility transistor (HEMT) as an example.

  FIG. 1 is a diagram schematically showing a state in which a HEMT 111 which is a semiconductor device according to the present embodiment is viewed obliquely from above. In order to clarify the configuration of this apparatus, in the following, after listing examples of rod-like molecules 163 and explaining their behavior, a sectional view of a portion indicated by a dotted line 195, a sectional view of a portion indicated by a dotted line 193, and a dotted line A description will be given with reference to a cross-sectional view of a portion indicated by 191.

  Examples of the rod-like molecule 163 include pyrene, tetracene, phenanthrene, anthracene, chrysene, and pentacene, which are organic compounds having a benzene ring, as shown in FIGS. 2 (a) to 2 (e). Examples of the rod-like molecule 163 include 1,3-butadiene-1-sodium-4-amine which is an organic compound having a double bond in the main skeleton, as shown in FIG. Each of these molecules is a rod-like molecule composed of an organic compound having π electrons, and is polarized in the major axis direction when an electric field is applied from the outside, as indicated by a double arrow in FIG.

The state of such polarization is shown in FIG. For example, in the case of anthracene which is a compound having a benzene ring, when a voltage is applied to the electrodes 1 and 2 arranged at both ends in the long axis direction of the molecule as shown in FIG. All of the 14 π electrons are close to the positive electrode 2 side as indicated by white arrows. As a result, induced dipoles are generated in the molecule as indicated by symbols δ ⁺ and δ ⁻ in the figure.

Alternatively, for example, in the case of 1,3-butadiene-1-sodium-4-amine, which is a compound having a double bond, as shown in FIG. 3 (b), an electron withdrawing group that originally has a tendency to attract electrons and Electron donating groups that originally have a tendency to emit electrons are present at both ends of the long axis of the molecule. Therefore, even when an electric field is not applied from the outside, there is a charge bias in the molecule. On the other hand, as shown in FIG. 3B, when an electric field is generated by the external electrodes 1 and 2, all of the π electrons due to double bonds in the molecule approach the positive electrode 2 side as indicated by the white arrow. . As a result, as shown by the symbols δ ⁺ and δ ⁻ in the figure, a charge bias that is greater than the potential bias in the molecule originally occurs in the molecule.

As the electron-withdrawing group, -Na shown in FIG.
Besides, for example,
It may be replaced by —CH ₃ , —NH ₂ or the like. On the other hand, as the electron-donating group, —NH ₂ shown in FIG.
Besides, for example,
-NO _2, may be replaced by -SO ₃ or the like. Further, the rod-like molecule 163 used in the HEMT 111 does not necessarily need to include both the electron-withdrawing group and the electron-donating group, and may have either one or both. That is, any material having π electrons found in a double bond or a triple bond may be used. Furthermore, the rod-like molecule 163 may be one in which either or both of an electron-withdrawing group and an electron-donating group are bonded to a benzene ring.

  FIG. 4 is a schematic diagram of a cross section corresponding to the dotted line 195 in FIG. 1 of the HEMT 111 according to the present embodiment. As illustrated, the HEMT 111 includes a substrate 141, a buffer layer 143, an electron transit layer 121 composed of a nitride compound semiconductor, an electron supply layer 123 composed of a nitride compound semiconductor, and a source electrode 151. A gate electrode 153, a drain electrode 155, and an insulating film 125.

  The electron transit layer 121 and the electron supply layer 123 form a heterointerface, and a two-dimensional electron gas layer (2DEG layer) 131 is generated in the electron transit layer 121 near the interface. Near the surface of the electron supply layer 123, there exists a surface state 133 that traps electrons due to crystal defects or the like.

  For example, a silicon substrate is used as the substrate 141. In the present embodiment, a substrate formed from a silicon single crystal is used.

The buffer layer 143 is formed on one main surface of the substrate 141, for example, on the substrate 141. The buffer layer 143 is a layer for favorably inheriting the crystal orientation of the substrate 141 between the electron transit layer 121 and the substrate 141. The buffer layer 143 is made of a nitride compound semiconductor. For example, the buffer layer 143 may be formed by alternately laminating Al _K Ga _1-K N (0 <K ≦ 1) and GaN, or from a single layer such as an Al _K Ga _1-K N layer or a GaN layer. You may provide the well-known buffer layer comprised. Note that the buffer layers 143 are preferably stacked alternately rather than being formed as a single layer. This is because by alternately laminating, the electron transit layer 121 can be prevented from warping and cracking to improve crystal quality, and the electron transit layer 121 can be laminated thickly.

  The electron transit layer 121 is formed on the buffer layer 143. The electron transit layer 121 has a function as a so-called current path. The electron transit layer 121 is formed, for example, by stacking GaN on the buffer layer 143 by metal organic chemical vapor deposition (MOCVD).

The electron supply layer 123 forms a heterojunction on the electron transit layer 121. The electron supply layer 123 has a function of supplying electrons. The electron supply layer 123 is made of, for example, a nitride compound semiconductor that has a lattice constant different from that of the nitride compound semiconductor constituting the electron transit layer 121 and has a wider band gap energy. And gallium aluminum nitride (AlGaN). The electron supply layer 123 is formed, for example, by stacking AlGaN on the electron transit layer 121 by metal organic chemical vapor deposition (MOCVD). In addition to the _{AlGaN, Al α M β Ga 1} -α-β N (M is boron (B) or indium (In); 0 ≦ α ≦ 1,0 ≦ β ≦ 1,0 <1-α-β <1 is sufficient.)

  The source electrode 151 is formed on a predetermined region of the electron supply layer 123 (on the main surface of the HEMT 111). The source electrode 151 is formed so as to be in ohmic (low resistance) junction with the electron supply layer 123, for example. In the present embodiment, the source electrode 151 is composed of a titanium (Ti) film and a gold (Au) film formed on the Ti film. The source electrode 151 is formed on the electron supply layer 123 by, for example, forming a Ti film and an Au film by sputtering or the like on the electron supply layer 123, patterning the film into a predetermined shape by dry etching or the like, and performing annealing treatment. The

  The gate electrode 153 is formed on a predetermined region of the electron supply layer 123 (on the main surface of the HEMT 111) with an insulating film 125 interposed therebetween. In the present embodiment, the gate electrode 153 is composed of a nickel (Ni) film or platinum (Pt) film and a gold (Au) film formed on the Ni film or Pt film. The gate electrode 153 is formed on the insulating film 125 by, for example, forming a Ni film (or Pt film) and an Au film by sputtering or the like, patterning the film into a predetermined shape by dry etching or the like, and performing an annealing process. Formed.

  For example, the gate electrode 153 may be formed by forming a film made of a metal such as aluminum or polysilicon on the electron supply layer 123 with an insulating film 125 made of a known material interposed therebetween.

  Further, a material that is in Schottky contact with the electron supply layer 123, for example, a metal such as tungsten (W), palladium (Pd), or molybdenum (Mo), without the insulating film 125 interposed on the electron supply layer 123. May be formed.

  The drain electrode 155 is formed in the same manner as the source electrode 151.

In this embodiment, the HEMT 111 includes an insulating film 125 formed so as to cover the exposed portion of the electron supply layer 123. Insulating film 125 is composed of SiO _2. The insulating film 125 also functions as a protective film that protects the electron supply layer 123.

  FIG. 5 is a schematic diagram of a cross section corresponding to the dotted line 193 in FIG. 1 of the HEMT 111 according to the present embodiment. 4 is substantially the same as FIG. 4, but in this cross-sectional view, a groove 161 provided in the direction connecting the gate electrode 153 and the drain electrode 155 on the upper surface of the insulating film 125, and the major axis direction is the direction. Further shown is a rod-like molecule 163 disposed inside the groove. Note that the rod-shaped molecules 163 do not necessarily have to be entirely accommodated in the grooves 161, and some may be arranged so as to overflow from the grooves 161. The current collapse suppression film is composed of rod-like molecules 163 arranged in this way. The current collapse suppression film is formed, for example, by vacuum-depositing rod-like molecules 163.

2F, when one or both of an electron-withdrawing group and an electron-donating group is provided, a voltage is applied between the gate electrode 153 and the drain electrode 155 so that the gate electrode 153 side is δ ^The rod-shaped molecule 163 is arranged so as to be ⁺ .

  FIG. 6 is a schematic diagram of a cross section corresponding to the dotted line 191 in FIG. 1 of the HEMT 111 according to the present embodiment. As shown in FIGS. 1 and 6, the groove 161 may have a rectangular cross section, and may be a flat surface with a gap between adjacent grooves. Alternatively, the groove 161 may have an inverted triangular cross section and may be provided on the upper surface of the insulating film 125 without a gap, or a gap may be formed between adjacent grooves. The width of the groove 161 is, for example, 1 to 100 μm, preferably 1 to 20 μm. The depth of the groove 161 is less than the thickness of the insulating film 125, and is, for example, 5 nm to 5 μm. The distance between the grooves 161 is 0.1 μm to the distance between the electrodes since at least one groove 161 is required between the electrodes.

  The rod-like molecule 163 has a width of, for example, 2 to 20 angstroms, and is greatly different in structure scale from the above-described groove 161. Nevertheless, by providing the groove 161 as described above, it has been experimentally confirmed that the rod-like molecule 163 adheres to the upper surface of the insulating film 125 with the long axis direction coinciding with the extending direction of the groove. . Although the detailed mechanism is not yet elucidated, it is considered that this is a phenomenon similar to a well-known phenomenon in which liquid crystal molecules adhere to a rubbed substrate in a uniform alignment.

  Next, the operation and effect of the HEMT 111 configured as described above will be described with reference to FIGS. In order to clarify the characteristics of the present invention, the operation and effect of a general conventional HEMT 911 (FIGS. 9 and 10) having no current collapse suppression film will be described in parallel. FIG. 8 is a schematic cross-sectional view of the HEMT 111, which corresponds to the cross section indicated by the dotted line 193 in FIG. FIG. 10 is a schematic cross-sectional view of the HEMT 911 and corresponds to the cross section indicated by the dotted line 995 in FIG.

7 and 9 are diagrams showing the HEMTs 111 and 911 in the off state (the gate electrode 153 has a negative potential and the drain electrode 155 has a higher potential than the source electrode 151). As shown in FIGS. 7 and 9, a switch 185 connected to the source electrode 151 is connected to B (a negative voltage is applied to the gate electrode 153 to turn off the HEMT 111 and the HEMT 911, for example, V _OFF = When 5 V is applied), voltages V _DS and V _DS + V _OFF of several hundred volts are applied to the HEMT 111 and the HEMT 911 (between the drain electrode 155 and the source electrode 151 and between the drain electrode 155 and the gate electrode 153). The If the negative voltage is applied to the gate, otherwise, a current is generated between the drain and the source as is known as a so-called normally-on characteristic, so that the HEMT is turned off. Because it will not be.

  7 and 9, when the switch 185 is switched from B to A, the HEMT 111 and the HEMT 911 are turned on, and an output current flows between the drain electrode 155 and the source electrode 151. At this time, in the conventional HEMT 911 that does not include the current collapse suppression film, the output current decreases due to the current collapse phenomenon.

  Here, the current collapse phenomenon is that when the above-described high voltage is applied to the HEMT in the off state, the electrons 991 are trapped and trapped at the surface level 133 as shown in FIGS. It can be considered that the electric field generated by the electrons decreases the concentration of the electron gas in the 2DEG layer 131.

  Among the surface states 133 widely distributed in the vicinity of the surface of the electron supply layer 123, a place where many electrons are trapped is particularly between the gate electrode 153 and the drain electrode 155 and close to the gate electrode. It is known that the surface level is at a position (a position indicated by a dotted square and an ellipse 835 in FIGS. 9 and 10). Therefore, as shown in FIG. 10, the concentration of the 2DEG layer 131 is decreased in the portion corresponding to the position immediately below the position by the above-described action by the electrons trapped in the surface level at the position. By so reducing the concentration of the 2DEG layer 131 that is part of the electron path, sufficient current does not flow between the drain electrode 155 and the source electrode 151 when switched on. It is considered that the output reduction due to the current collapse phenomenon is such a thing.

On the other hand, in the HEMT 111 according to the present embodiment, the current collapse phenomenon is suppressed as follows. In the HEMT 111, of course, V _DS is much higher than the gate-source voltage V _ON at that time even when the HEMT 111 is in an off state (switch 185 in FIG. 7 is connected to A). Therefore, the potential of the drain electrode 155 is still higher than the potential of the gate electrode 153. Therefore, the rod-like molecule 163 is sandwiched between the drain electrode 155 which is a positive electrode and the gate electrode 153 which is a negative electrode, and is in a state where the induced dipole shown in FIG. 3 is generated. . A state in which an induced dipole is generated in the rod-shaped molecule 163 is schematically shown in FIGS. 7 and 8 by attaching + and − symbols to the rod-shaped molecule 163.

  Using a rod-like molecule capable of generating an induced dipole as a material, and forming a current collapse suppression film while aligning the long axis direction to the direction connecting the gate electrode 153 and the drain electrode 151, it is possible to suppress the current collapse phenomenon well. Confirmed. The assembly of induced dipoles with the same orientation prevents electrons from being trapped at the surface level 133 by some electromagnetic action, pulls out the electrons once trapped at the surface level 133, or Even if the electrons remain trapped, it is considered that the electric field generated by the trapped electrons is relaxed.

  Although elucidation of the details of the above-described electromagnetic action is insufficient, for example, the following action can be considered. That is, of the polarized rod-shaped molecules 163, the portion on the positive electrode side (positively charged portion) of the rod-shaped molecules closest to the gate electrode 153 is within the range indicated by the downward funnel-shaped dotted line 837 in FIG. Influence as a positive charge. The range affected by this includes the position 835 (position indicated by a dotted ellipse in FIG. 8) in which the electrons are most easily trapped in the surface level 133 as described above. Therefore, trapping of electrons by the surface level 133 at the position is suppressed.

  As described above, the HEMT 111 including the current collapse suppression film favorably suppresses the current collapse phenomenon. Moreover, in order to control the direction of the rod-like molecule 163, it is only necessary to dig a groove 161 on the surface of the insulating film 125, and since the size of the groove 161 is at least on the order of μm, special fine processing is required for its formation. In addition, the rod-like molecules 163 are adsorbed on the insulating film 125 by a well-known method as will be described later as an embodiment relating to the manufacturing method. That is, according to the present invention, current collapse can be suppressed very easily.

  In addition, although various studies using an organic semiconductor film as a device have been conducted, in the semiconductor device according to the present invention, even when the current collapse suppression film is composed of organic compound molecules, the film is the semiconductor device itself, That is, it does not function as a part of the current path.

(Second Embodiment)
As a second embodiment, the present invention will be described by taking as an example a semiconductor device including a Schottky barrier diode (SBD). In the present embodiment, the same members as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted. For this reason, in this embodiment, a description will be given centering on differences from the first embodiment.

  FIG. 11 is a diagram schematically showing the SBD 1111 according to the present embodiment as viewed obliquely from above. However, in order to make the drawing easier to see, the insulating film 125 is virtually removed and moved upward. In addition, in order to make it easy to see a later-described groove formed on the insulating film 125, which is a feature of the present invention, a part of the insulating film 125 is hatched only in FIG. Further, in order to make the drawing easier to see, the rod-like molecules 163 that are originally present in large numbers are omitted from the illustration in FIG. 11 is a schematic cross-sectional view at the position of the dotted line 1191 in FIG. 11, and FIG. 13 is a schematic cross-sectional view at the position of the dotted line 1193 in FIG.

  An anode electrode 1151 and a cathode electrode 1155 are formed on a predetermined region (on the main surface of the SBD) of the electron supply layer 123 of the SBD 1111.

  The anode electrode 1151 is formed to have, for example, a Schottky junction with the electron supply layer 123. In this embodiment mode, the anode electrode 1151 is formed on a metal film such as nickel (Ni), platinum (Pt), tungsten (W), palladium (Pd), molybdenum (Mo), and the like. It is composed of a gold (Au) film. The anode electrode 1151 is formed on the electron supply layer 123 by, for example, forming the metal film on the electron supply layer 123 by sputtering or the like, patterning it into a predetermined shape by dry etching or the like, and performing annealing treatment.

  The cathode electrode 1155 is formed to be in ohmic contact (low resistance contact) with the electron supply layer 123, for example. In the present embodiment, the cathode electrode 1155 includes a titanium (Ti) film and a gold (Au) film formed on the Ti film. The cathode electrode 1155 is formed on the electron supply layer 123 by, for example, forming a Ti film and an Au film by sputtering or the like on the electron supply layer 123, patterning them into a predetermined shape by dry etching or the like, and performing annealing treatment. The

  In FIG. 12, which is a cross-sectional view, the anode electrodes 1151 and the cathode electrodes 1155 are alternately arranged. However, as shown in FIG. 11, the anode electrodes 1151 and the cathode electrodes 1155 are alternately formed in the entire apparatus. The anode electrode 1151 and the cathode electrode 1155 have a skewered shape.

  On the upper surface of the insulating film 125, a groove 161 is dug in the direction connecting the anode electrode 1151 and the cathode electrode 1155, as in the previous embodiment. The rod-like molecule 163 is adsorbed on the upper surface of the protective film 125 while making the major axis direction (polarization direction) coincide with the extending direction of the groove to form a current collapse suppressing film. As shown in FIGS. 11 and 13, the rod-like molecules 163 do not necessarily have to be entirely accommodated in the groove 161, and some of them may be adsorbed so as to overflow from the groove 161.

In the case where either or both of an electron withdrawing group and an electron donating group are provided as shown in FIG. 2 (f), a reverse bias voltage is applied to the anode electrode 1151 and the cathode electrode 1155, and the anode electrode 1151 is applied. The rod-like molecule 163 is arranged so that δ ⁺ comes to the side.

  The operation and effect of the SBD 1111 configured as described above will be described from the viewpoint of a problem in a general conventional SBD 1411 (FIGS. 14 and 15) that does not include a current collapse suppression film and its solution.

Although omitted in FIG. 11 for easy understanding of the drawing, a general circuit for operating the SBD is as shown in FIG. 14, and the same applies to the case of the SBD 1111 and the case of the SBD 1411. As shown in FIG. 14, first, the switch 1461 connected to the cathode electrode 1155 is connected to B (V _R : several hundreds V). As a result, a reverse bias (a state where the potential of the anode electrode 1151 is lower than the potential of the cathode electrode 1155) of several hundred volts (a peak value larger than the voltage peak value during forward bias) is applied to the SBD 1111 and the SBD 1411.

Subsequently, when the switch 1461 is switched from B to A, a forward bias voltage (V _F : several V) is applied to the SBD 1111 and SBD 1411, and current flows from the anode electrode 1151 to the cathode electrode 1155 inside the SBD 1111 and SBD 1411. Flowing. At this time, in the conventional SBD 1411, the output current decreases due to the current collapse phenomenon.

  Here, the current collapse phenomenon means that when the above-described high voltage is applied to the SBD in the reverse bias state, the electrons 991 are trapped at the surface level 133 as shown in FIG. It is considered that the electric field generated by the phenomenon reduces the concentration of the electron gas in the 2DEG layer 131.

  Among the surface states 133 widely distributed in the vicinity of the surface of the electron supply layer 123, a place where many electrons are trapped is particularly at a position close to the anode electrode 1151 (position indicated by a dotted ellipse 835 in FIG. 15). It is known to have a certain surface level. Therefore, as shown in FIG. 15, the concentration of the 2DEG layer 131 is reduced in the portion corresponding to the position immediately below the position due to the above-described action by the electrons trapped in the surface level at the position. Since the concentration of the 2DEG layer 131 which is a part of the electron path is reduced as described above, a sufficient current does not flow between the anode electrode 1151 and the cathode electrode 1155 when switching to the forward bias. It is considered that the output reduction due to the current collapse phenomenon is such a thing.

  On the other hand, in the SBD 1111 according to the present embodiment, the current collapse phenomenon is suppressed as follows. When the SBD 1111 is in a reverse bias state, the potential of the cathode electrode 1155 is higher than the potential of the anode electrode 1151. Therefore, the rod-like molecule 163 is sandwiched between the cathode electrode 1155 that is a positive electrode and the anode electrode 1151 that is a negative electrode, and the induced dipole shown in FIG. 3 is generated. . The appearance of the induced dipole is schematically shown in FIG. 12 by attaching + and − symbols to the rod-like molecule 163.

  It has been experimentally confirmed that the current collapse phenomenon can be satisfactorily suppressed by creating a current collapse suppression film using rod-shaped molecules that can generate induced dipoles and aligning the major axis in the direction connecting the anode and cathode. It was. It is considered that the assembly of induced dipoles having the same orientation prevented electrons from being trapped at the surface level 133 by some electromagnetic action.

  Although elucidation of the details of the above-described electromagnetic action is insufficient, the following action can be considered, for example, as in the case of the previous embodiment (in the case of HEMT). That is, of the polarized rod-like molecules 163, the portion on the positive electrode side (positively charged portion) of the rod-like molecule closest to the anode electrode 1151 is within the range indicated by the downward funnel-shaped dotted line 837 in FIG. Influence as a positive charge. The range affected by this includes the position 835 (position indicated by the dotted ellipse in FIG. 12) in which the electrons are most easily trapped in the surface level 133 as described above. Therefore, trapping of electrons by the surface level 133 at the position is suppressed.

  As described above, the SBD 1111 including the current collapse suppression film favorably suppresses the current collapse phenomenon.

  However, unlike the case of the previous embodiment (in the case of HEMT), when the SBD 1111 is in a forward bias state, a forward voltage is applied to the SBD 1111, and thus the anode electrode 1151 and the cathode electrode 1155. The sign of is reversed from that in the reverse bias state. Then, the direction of the induced dipole of the rod-like molecule 163 is also reversed. Therefore, with only the action described above as an example, in the case of SBD, unlike the case of HEMT, it is impossible to extract electrons once trapped in the surface level 133 after being turned on. Conceivable.

  The shape and size of the groove are the same as in the previous embodiment (in the case of HEMT).

(Third embodiment)
As a third embodiment, the present invention will be described by taking a semiconductor device manufacturing method as an example. The procedure is almost the same for both HEMT and SBD, and will be described collectively.

  First, in order to epitaxially grow a semiconductor layer functioning as an apparatus, a base substrate 141 is prepared using silicon, SiC, ceramic, or the like that is a known material.

  Next, a nitride compound semiconductor layer known as the buffer layer 143 is stacked on the substrate 141 by epitaxial growth.

  Further, a semiconductor substrate is obtained by sequentially stacking a first semiconductor layer to be the electron transit layer 121 and a second semiconductor layer to be the electron supply layer 123 by epitaxial growth. The first and second semiconductor layers are both nitride-based compound semiconductors, but have different lattice constants. Therefore, both layers form a heterointerface, and a 2DEG layer 131 is formed in the vicinity of the interface.

  This completes the fabrication of the semiconductor substrate. Thereafter, necessary electrodes are formed on the second semiconductor layer by a known technique such as photolithography in accordance with the specific specifications of the device to be obtained, for example, the specifications of HEMT and SBD. For example, when it is desired to create a HEMT, a source electrode 151, a gate electrode 153, and a drain electrode 155 are formed. When it is desired to create an SBD, an anode electrode 1151 and a cathode electrode 1155 are formed. If necessary, the electrode is Schottky bonded to the semiconductor substrate, ohmic bonded to the semiconductor substrate, or bonded to the semiconductor substrate via the insulating film 125.

  An insulating film 125 is formed on the upper surface so as to cover the surface of the second semiconductor layer. Alternatively, the insulating film 125 may be formed so as to cover the entire surface of the second semiconductor layer including the electrodes.

  A trench 161 having a rectangular or inverted triangular cross-sectional shape is dug on the upper surface of the insulating film 125 by any known method. The depth of the groove 161 is less than the thickness of the insulating film 125, and is, for example, 5 nm to 5 μm. The width of the groove 161 is, for example, 1 to 100 μm, preferably 1 to 20 μm. The distance between the grooves 161 is 0.1 μm to the distance between the electrodes since at least one groove 161 is required between the electrodes. The direction in which the groove 161 extends is the direction connecting the gate electrode 153 and the drain electrode 155 when manufacturing the HEMT, and the direction connecting the anode electrode 1151 and the cathode electrode 1155 when manufacturing the SBD. To.

  A single layer film, a multilayer film, or an amorphous film made of rod-like molecules 163 is formed on the insulating film 125 dug in the groove 161 by a vacuum evaporation method or an evaporation method from a solution.

  When the film is formed in this manner, the orientation of the rod-like molecule 163 is controlled so that the induced dipole that can be generated in the rod-like molecule 163 has an effective orientation for suppressing the current collapse phenomenon. Therefore, the above-described current collapse suppression film on the insulating film 125 has a function of satisfactorily suppressing the current collapse phenomenon.

  When the rod-like molecule 163 is deposited without digging the groove 161 in the insulating film 125, or when the groove 161 is dug in a direction perpendicular to the extending direction of the groove 161, an effect of suppressing the current collapse phenomenon is obtained. It may be small or adverse effects may occur.

  In FIG. 16, in the SBD, when the rod-like molecule 163 is not adsorbed on the insulating film 125, a film composed of the rod-like molecule 163 is provided on the insulating film 125, but the orientation of the rod-like molecule 163 is different, and When the current collapse suppression film formed by controlling the groove 161 so that the rod-like molecules 163 are oriented in the direction connecting the anode and the cathode is provided on the insulating film 125, the on-resistance after applying the reverse bias was measured. Results are shown. It can be seen that the current collapse suppression film suppresses the current collapse phenomenon.

  The drawings are merely schematic diagrams for explanation, and do not reproduce the actual size. For example, the rod-like molecule 163 constituting the current collapse suppression film is exaggerated for easy understanding, and its shape, the number thereof, and the groove 161 dug on the insulating film 125 are illustrated. The ratio of the size and the like are different from the actual device.

  The present invention is not limited to the above embodiment, and various modifications are possible. In constructing the current collapse suppression film, it is important to arrange the rod-like molecules 163 so that the major axis direction thereof is aligned with the direction connecting the electrodes having a large potential difference. Therefore, the constituent elements of the above embodiments are appropriately combined. Also good. For example, in the case of HEMT, the groove 161 may be provided on the entire surface of the insulating film 125 covering the entire electrode, or in the case of SBD, only the exposed portion of the electron supply layer 123 is left leaving the electrode. A groove 161 may be provided over the covered insulating film 125.

  Further, in the case of HEMT, as described above, the problematic electrons tend to be trapped intensively in the vicinity of the gate electrode 153, and therefore, a portion near the gate electrode 153 between the gate electrode 153 and the drain electrode 155. Even if the groove 161 is provided only in this portion and the current collapse suppression film is provided only in that portion, there is a sufficient effect. Similarly, in the case of SBD, a sufficient effect can be obtained by providing the groove 161 only in the portion close to the anode electrode 1151 and providing the current collapse suppression film only in that portion. Further, in the case of HEMT, a larger number of grooves 161 are provided in the portion close to the gate electrode 153 and in the case of SBD near the anode electrode 1151 by narrowing the interval between the grooves 161, and the current collapse is provided inside the groove 161. A suppression film may be provided. Alternatively, a thick current collapse suppression film can be obtained in a portion close to the gate electrode 153 in the case of HEMT and in a portion close to the anode electrode 1151 in the case of SBD. Note that the current collapse suppression film may be in a state where a plurality of layers composed of rod-like molecules 163 are stacked.

  The present invention can be applied to all semiconductor devices in which electrons are trapped in the surface state 133. The present invention can be applied to, for example, a MESFET in addition to the above-described HEMT and SBD. The present invention can also be applied to a semiconductor device having a compound semiconductor layer such as GaAs other than a nitride compound semiconductor.

  Furthermore, if at least the lowermost rod-like molecule is arranged in an appropriate direction, a current collapse suppressing effect can be produced. Therefore, the current collapse suppression film may be in the form of a multilayer film of rod-like molecules 163, polycrystalline, or amorphous.

It is a figure which shows typically a mode that the semiconductor device which functions as HEMT based on the 1st Embodiment of this invention was seen from diagonally upward. It is a figure which shows the specific example of a rod-shaped molecule | numerator. It is a figure which shows a mode that a rod-shaped molecule polarizes with the electric field produced by the positively and negatively charged electrode, and an induced dipole is produced. 1 is a first schematic cross-sectional view of a semiconductor device functioning as a HEMT according to a first embodiment of the present invention. It is a 2nd cross-sectional schematic diagram of the semiconductor device which functions as HEMT based on the 1st Embodiment of this invention. 1 is a third cross-sectional schematic view of a semiconductor device functioning as a HEMT according to a first embodiment of the present invention. It is a figure for demonstrating the OFF state and ON state of the semiconductor device which functions as HEMT based on the 1st Embodiment of this invention. It is a cross-sectional schematic diagram of the device for explaining the operation of the semiconductor device functioning as a HEMT according to the first embodiment of the present invention. It is a figure which shows typically a mode that the conventional HEMT was seen from diagonally upward. It is a cross-sectional schematic diagram of this HEMT for demonstrating operation | movement of the conventional HEMT. It is a figure which shows typically a mode that the semiconductor device which functions as SBD based on the 2nd Embodiment of this invention was seen from diagonally upward. FIG. 6 is a first schematic cross-sectional view of a semiconductor device functioning as an SBD according to a second embodiment of the present invention, and is a view for explaining an operation. It is a 2nd cross-sectional schematic diagram of the semiconductor device which functions as SBD based on the 2nd Embodiment of this invention. It is a figure which shows typically a mode that the conventional SBD was seen from diagonally upward. It is a cross-sectional schematic diagram of the SBD for explaining the operation of the conventional SBD. It is an experimental result which shows the effect which provided the current collapse suppression film | membrane.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Negative electrode 2 Positive electrode 111 HEMT which concerns on 1st Embodiment
121 Electron travel layer 123 Electron supply layer 125 Insulating film 131 Two-dimensional electron gas layer (2DEG layer)
133 Surface State 141 Substrate 143 Buffer Layer 151 Source Electrode 153 Gate Electrode 155 Drain Electrode 161 Groove 163 Rod Molecule 171 Power Supply for Drain-Source Voltage 173 Load Resistance 181 Off Gate Power Supply 183 On Gate Power Supply 185 HEMT Switch 835 A position where electrons are easily trapped 837 A portion where a positive charge effect of a rod-shaped molecule adjacent to the negative electrode reaches 911 Conventional HEMT
991 Electrons trapped at the surface level 1111 SBD according to the second embodiment
1151 Anode electrode 1155 Cathode electrode 1411 Conventional SBD
1461 Switch for SBD

Claims (10)

A substrate,
A compound semiconductor layer formed on one main surface of the substrate;
A drain electrode formed on the compound semiconductor layer;
A source electrode formed on the compound semiconductor layer;
A gate electrode formed on the compound semiconductor layer between the drain electrode and the source electrode;
An insulating film formed so as to cover at least a part of a region between the drain electrode and the gate electrode in a region where the compound semiconductor layer on the compound semiconductor layer is exposed;
A film containing rod-like molecules disposed on the upper surface of the insulating film;
With
A groove extending in a direction connecting the gate electrode and the drain electrode is provided on the upper surface of the insulating film,
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the gate electrode and the drain electrode,
A semiconductor device. A substrate,
A compound semiconductor layer formed on one main surface of the substrate;
An anode electrode formed on the compound semiconductor layer;
A cathode electrode formed on the compound semiconductor layer;
An insulating film formed to cover at least a part of a region between the anode electrode and the cathode electrode in a region where the compound semiconductor layer on the compound semiconductor layer is exposed;
A film containing rod-like molecules disposed on the upper surface of the insulating film;
With
A groove extending in a direction connecting the anode electrode and the cathode electrode is provided on the upper surface of the insulating film,
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the anode electrode and the cathode electrode,
A semiconductor device. The compound semiconductor layer is
Composed of nitride compound semiconductors,
The semiconductor device according to claim 1 or 2, characterized in that. The compound semiconductor layer is
It is composed of a first compound semiconductor layer and a second compound semiconductor layer, and has a two-dimensional electron gas layer in the vicinity of the interface between the first compound semiconductor layer and the second compound semiconductor layer.
The semiconductor device according to any one of claims 1 to 3, characterized in that. The rod-like molecule is
Formed as a single layer film or an amorphous film,
The semiconductor device according to any one of claims 1 to 4, characterized in that. The rod-like molecule is
Has a double bond or benzene ring in the main skeleton,
6. The semiconductor device according to any one of claims 1 to 5 , wherein: The rod-like molecule is
Having an electron donating group or an electron withdrawing group,
The semiconductor device according to any one of claims 1 to 6, characterized in that. The rod-like molecule is
composed of organic compounds having π electrons,
The semiconductor device according to any one of claims 1 to 7, characterized in that. A compound semiconductor layer forming step of forming a compound semiconductor layer on the substrate;
An electrode forming step of forming a drain electrode and a source electrode on the compound semiconductor layer, and further forming a gate electrode between the drain electrode and the source electrode;
An insulating film forming step of forming an insulating film so as to cover at least a part of a region between the drain electrode and the gate electrode in a region where the compound semiconductor layer on the upper surface of the compound semiconductor layer is exposed;
A groove forming step of forming a groove extending in a direction connecting the said gate electrode and said drain electrode on the upper surface of the insulating film,
A film forming step of forming a film containing rod-like molecules on the insulating film;
With
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the gate electrode and the drain electrode,
A method for manufacturing a semiconductor device. A compound semiconductor layer forming step of forming a compound semiconductor layer on the substrate;
An electrode forming step of forming an anode electrode and a cathode electrode on the compound semiconductor layer;
An insulating film forming step of forming an insulating film so as to cover at least part of a region between the anode electrode and the cathode electrode in a region where the compound semiconductor layer on the upper surface of the compound semiconductor layer is exposed;
Forming a groove extending in a direction connecting the anode electrode and the cathode electrode on an upper surface of the insulating film;
A film forming step of forming a film containing rod-like molecules on the insulating film;
With
The rod-like molecule is
At least a portion is inside the groove, and the long axis direction of the rod-like molecule arranged in the groove is arranged to be a direction connecting the anode electrode and the cathode electrode,
A method for manufacturing a semiconductor device.

Images