JP2011233810A - Photoelectric conversion element and manufacturing method thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a photoelectric conversion element capable of improving the photoelectric conversion efficiency, and provide a manufacturing method of the photoelectric conversion element.SOLUTION: The photoelectric conversion element comprises: a first gallium nitride layer produced by growth at low temperature; and an indium nitride quantum dot formed on a surface of the first gallium nitride layer. The manufacturing method of a photoelectric conversion element comprises: a first process of producing a gallium nitride layer by growth at low temperature; and a second process of forming an indium nitride quantum dot on a surface of the gallium nitride layer formed by the first process.

Description

  The present invention relates to a photoelectric conversion element typified by a solar cell, a light detection element, and the like and a method for manufacturing the photoelectric conversion element, and more particularly to a photoelectric conversion element using quantum dots and a method for manufacturing the photoelectric conversion element.

  Solar cells have the advantage that the amount of carbon dioxide emission per unit of power generation is small and fuel for power generation is unnecessary. Therefore, research on various types of solar cells has been actively promoted. Currently, single-junction solar cells having a pair of pn junctions using single-crystal silicon or polycrystalline silicon are the mainstream among solar cells in practical use. However, since the theoretical limit of photoelectric conversion efficiency of single-junction solar cells (hereinafter referred to as “theoretical limit efficiency”) is only about 30%, a new method for further improving the theoretical limit efficiency is being studied. .

  One of the new methods studied so far is a solar cell using semiconductor quantum dots (hereinafter referred to as “quantum dot solar cell”). A quantum dot used in a quantum dot solar cell is, for example, a semiconductor nanocrystal having a size of about 10 nm. Electrons and holes generated by light irradiation (hereinafter, these may be collectively referred to as “carriers”). .) Can be confined in three dimensions. By confining electrons in quantum dots, it becomes possible to use the properties of electrons as quantum mechanical waves, and it is possible to absorb the solar spectrum in a band that could not be absorbed by conventional solar cells. Become. Furthermore, according to the quantum dot solar cell, it is possible to reduce energy lost as heat. Therefore, according to the quantum dot solar cell, it is considered that the theoretical limit efficiency can be improved to 60% or more.

As a technique related to such a solar cell, for example, Patent Document 1 discloses that a substrate 1, a p-type or n-type semiconductor layer 2 formed on the substrate 1 with a first compound semiconductor material, and the semiconductor layer 2 The formed i-type semiconductor layer 3 and the n-type or p-type semiconductor layer 4 formed on the i-type semiconductor layer 3 with the first compound semiconductor material, and the i-type semiconductor layer 3 is the first compound. A base layer 3c made of a semiconductor material; a quantum dot layer 3a formed on the base layer 3c with a second compound semiconductor material; and a cap layer 3b formed on the quantum dot layer 3a with a first compound semiconductor material; A solar battery cell 10 formed as a structural unit 30 is disclosed. Further, in Patent Document 2, in a nitride semiconductor device having an n-side nitride semiconductor layer, an active layer, and a p-side nitride semiconductor layer on a substrate, the n-side nitride semiconductor layer includes an n-type impurity. A buffer layer made of Ga _d Al _1-d N (0 <d ≦ 1), which has a contact layer, the n-side contact layer is formed on the undoped GaN layer, and the undoped GaN layer is grown at a low temperature. A forming technique is disclosed. Patent Document 3 includes a first semiconductor layer and a second semiconductor layer made of carbon, and a quantum well portion made of carbon formed between the first and second semiconductor layers. The quantum well portion includes a wall layer formed of a carbon semiconductor thin film, a plurality of quantum dots embedded in the wall layer, and an sp ² bond prevention layer provided in a peripheral portion of the quantum dot, A solar cell is disclosed in which the sp ² bond prevention layer is configured by bonding hydrogen to carbon atoms in a portion of the wall layer adjacent to the quantum dots. Patent Document 4 also discloses a first conductivity type first InP layer provided on a semi-insulating or first conductivity type InP substrate, and between the InP substrate and the first InP layer. Provided between the first InP layer having the first conductivity type, the third InP layer having the second conductivity type region, the first InP layer, and the third InP layer, and having a pn junction therein. And a light receiving layer, wherein the carrier concentration of the second InP layer is greater than the carrier concentration of the first InP layer.

Japanese Patent No. 3753605 Japanese Patent Laid-Open No. 2000-244072 JP 2006-332540 A JP 2006-147670 A

  According to the technique disclosed in Patent Document 1, since quantum dots are used, carriers can be confined three-dimensionally. As a result, it is considered possible to reduce energy loss of carriers. However, there is a problem that it is difficult to sufficiently reduce the size of the quantum dots and it is difficult to improve the photoelectric conversion efficiency only by using the technique disclosed in Patent Document 1. Such a problem has been difficult to solve even when the techniques disclosed in Patent Documents 1 to 4 are combined.

  Then, this invention makes it a subject to provide the photoelectric conversion element which can improve a photoelectric conversion efficiency, and its manufacturing method.

In order to solve the above problems, the present invention takes the following means. That is,
A first aspect of the present invention includes a first gallium nitride layer produced by growing at a low temperature, and an indium nitride quantum dot formed on the surface of the first gallium nitride layer. It is a photoelectric conversion element.

  Here, in the present invention, “low temperature” is lower than 800 ° C. to 900 ° C., which is a general temperature when epitaxially growing gallium nitride (hereinafter sometimes referred to as “GaN”). It refers to a temperature of 600 ° C. or lower. The temperature is preferably 360 ° C. or higher and 470 ° C. or lower. In the present invention, “growth” refers to a method by MOCVD or MBE, for example.

  In the first aspect of the present invention, the first gallium nitride layer may be a p-type semiconductor layer.

  Here, in the present invention, the “p-type semiconductor layer” refers to a layer that functions as a p-type semiconductor.

  In the first aspect of the present invention, it is preferable that a second gallium nitride layer is disposed on the surface of the indium nitride quantum dots.

  In the first aspect of the present invention in which the second gallium nitride layer is disposed on the surface of the indium nitride quantum dots, the first gallium nitride layer and the second gallium nitride layer may contain hydrogen. preferable.

  Here, “the first gallium nitride layer and the second gallium nitride layer contain hydrogen” means that the first gallium nitride layer and the second gallium nitride layer are activated by plasma or heat. In other words, the layer is grown under an environment where hydrogen is actively added to the first gallium nitride layer and the second gallium nitride layer.

  In the first aspect of the present invention in which the second gallium nitride layer is disposed on the surface of the indium nitride quantum dots, the doping concentration of the second gallium nitride layer is higher than the doping concentration of the first gallium nitride layer. High is preferred.

  Here, the “dope concentration” is generally proportional to the hole concentration or the electron concentration (hereinafter referred to as “carrier concentration”). Therefore, in order to control the carrier concentration, the doping concentration may be controlled.

  In the first aspect of the present invention in which the second gallium nitride layer is disposed on the surface of the indium nitride quantum dots, the second gallium nitride layer may be a p-type semiconductor layer.

  In the first aspect of the present invention, the first gallium nitride layer is preferably disposed on the surface of the third gallium nitride layer.

  In the first aspect of the present invention in which the first gallium nitride layer is disposed on the surface of the third gallium nitride layer, it is preferable that the third gallium nitride layer does not contain hydrogen.

  “The third gallium nitride layer does not contain hydrogen” means that hydrogen is not actively added to the third gallium nitride layer. In the present invention, hydrogen as an unavoidable impurity is the third. The form contained in the gallium nitride layer is acceptable.

  In the first aspect of the present invention in which the first gallium nitride layer is disposed on the surface of the third gallium nitride layer, the third gallium nitride layer may be an n-type semiconductor layer.

  Here, in the present invention, the “n-type semiconductor layer” refers to a layer that functions as an n-type semiconductor.

  The second aspect of the present invention includes a first step of forming a gallium nitride layer grown at a low temperature, and a second step of forming indium nitride quantum dots on the surface of the gallium nitride layer formed by the first step. And a method for producing a photoelectric conversion element.

  In the second aspect of the present invention, the gallium nitride layer formed in the first step may be a p-type semiconductor layer.

  The second aspect of the present invention preferably further includes a third step of forming a gallium nitride layer grown at a low temperature on the surface of the indium nitride quantum dots formed in the second step.

  In the second aspect of the present invention having the first to third steps, it is preferable that the first step and the third step are steps of forming a gallium nitride layer while adding hydrogen.

  Further, in the second aspect of the present invention having the first to third steps, the doping concentration of the gallium nitride layer formed in the third step is higher than the doping concentration of the gallium nitride layer formed in the first step. Is preferably high.

  In the second aspect of the present invention having the first to third steps, the gallium nitride layer formed in the third step may be a p-type semiconductor layer.

  The second aspect of the present invention preferably includes a fourth step of forming a gallium nitride layer on the surface of which the gallium nitride layer to be formed in the first step.

  In the second aspect of the present invention having the first to fourth steps, the fourth step is preferably a step of forming a gallium nitride layer without adding hydrogen.

  In the second aspect of the present invention having the first to fourth steps, the gallium nitride layer formed in the fourth step may be an n-type semiconductor layer.

  In the first aspect of the present invention, first gallium nitride in which indium nitride quantum dots (hereinafter, indium nitride is sometimes referred to as “InN” and indium nitride quantum dots are sometimes referred to as “InN quantum dots”) is formed. The layer is made by growing at low temperature. When gallium nitride is produced by epitaxial growth at a low temperature, gallium nitride having many crystal defects is formed, so that it is possible to form many minute irregularities on the surface of the gallium nitride layer. When InN quantum dots are formed on the surface of the gallium nitride layer having such a large number of minute irregularities, the movement of InN quantum dots on the surface of the gallium nitride layer can be suppressed, and a plurality of InN quantum dots can be suppressed. As a result of suppressing the bonding of dots, an InN quantum dot having a smaller size than the conventional one can be produced. When the size of the InN quantum dots is reduced, it is easy to reduce energy loss due to the phonon bottleneck effect. Therefore, according to the present invention including the configuration in which InN quantum dots are formed on the surface of the first gallium nitride layer produced by low temperature growth, it is possible to provide a photoelectric conversion element capable of improving the photoelectric conversion efficiency. it can.

  In the first aspect of the present invention, even if the first gallium nitride layer is a p-type semiconductor layer, the photoelectric conversion efficiency can be improved.

  In the first aspect of the present invention, since the second gallium nitride layer is disposed on the surface of the InN quantum dot, GaN has good heat resistance, and thus the shape of the InN quantum dot is protected. Is possible.

  In the first aspect of the present invention, the first gallium nitride layer and the second gallium nitride layer contain hydrogen so that carriers trapped by lattice defects can be reduced. Therefore, it becomes easy to improve the photoelectric conversion efficiency.

  In the first aspect of the present invention, since the doping concentration of the second gallium nitride layer is higher than the doping concentration of the first gallium nitride layer, the carriers in the InN quantum dots can be easily taken out. It becomes easy to improve.

  In the first aspect of the present invention, even if the second gallium nitride layer is a p-type semiconductor layer, the photoelectric conversion efficiency can be improved.

  In the first aspect of the present invention, since the first gallium nitride layer is disposed on the surface of the third gallium nitride layer, the carriers in the InN quantum dots are shaped like blades in the third gallium nitride layer. Since it becomes possible to take out via a dislocation | rearrangement, it becomes easy to improve a photoelectric conversion efficiency.

  Further, in the first aspect of the present invention, since the third gallium nitride layer does not contain hydrogen, dangling bonds of edge dislocations in the third gallium nitride layer are inactivated by hydrogen. Since it can avoid, it becomes easy to improve a photoelectric conversion efficiency.

  In the first aspect of the present invention, even if the third gallium nitride layer is an n-type semiconductor layer, the photoelectric conversion efficiency can be improved.

  According to the 2nd aspect of this invention, it becomes possible to manufacture the photoelectric conversion element concerning the 1st aspect of this invention. Therefore, according to the 2nd aspect of this invention, the manufacturing method of the photoelectric conversion element which can manufacture the photoelectric conversion element which can improve a photoelectric conversion efficiency can be provided.

It is sectional drawing explaining the photoelectric conversion element 10 of this invention. 3 is a flowchart for explaining a manufacturing process of the photoelectric conversion element 10. It is a band figure explaining the photoelectric conversion element 10 of this invention. It is a figure which shows the atomic force microscope observation photograph of an InN quantum dot. It is a figure which shows the atomic force microscope observation photograph of an InN quantum dot. It is a figure which shows the atomic force microscope observation photograph of an InN quantum dot. It is a figure which shows the atomic force microscope observation photograph of an InN quantum dot. It is a figure which shows the atomic force microscope observation photograph which emphasized the unevenness | corrugation of a GaN substrate. It is sectional drawing which shows the unevenness | corrugation of a GaN substrate. It is a figure which shows the atomic force microscope observation photograph which emphasized the unevenness | corrugation of a GaN substrate. It is sectional drawing which shows the unevenness | corrugation of a GaN substrate. It is a band diagram of GaN and InN. It is sectional drawing explaining the laminated structure of GaN and an InN quantum dot. It is a band diagram of dislocations in GaN. FIG. 14A is a band diagram of edge dislocations in GaN, and FIG. 14B is a band diagram of screw dislocations in GaN.

  When materials having different lattice constants are thinly epitaxially grown, growth called a Stranki-Krastanov mode (hereinafter referred to as “SK mode”) in which quantum dots grow in an island shape in order to minimize strain energy due to the difference in lattice constants. May happen. When InN is epitaxially grown on GaN, InN quantum dots can be produced by the SK mode. However, even if an InN quantum dot is simply produced by the SK mode, the diameter becomes about 70 to 255 nm, and there is a problem that a sufficient quantum effect cannot be obtained. Although the lattice constant of InN and the lattice constant of GaN are different by 11%, the diameter of the quantum dot increases because the manufactured InN quantum dot moves on the GaN surface and adjacent quantum dots merge. This is thought to be due to (coalescence). As a result of intensive studies, the present inventors have made it possible to suppress the movement of quantum dots by roughening the surface of GaN on which InN quantum dots are produced, and as a result, producing small quantum dots. I made a hypothesis that would be possible, and investigated the truth.

  FIGS. 4 to 7 show the SK mode after cooling a GaN substrate having InN quantum dots formed on the surface thereof, and then taking it out from a molecular beam epitaxy (hereinafter referred to as “MBE”) apparatus. , Referred to as “AFM”).

  FIG. 4 is an AFM observation photograph of Sample A in which InN quantum dots were produced by irradiating indium equivalent to five molecular layers in terms of InN under the condition that the temperature of the GaN substrate was set at 460 ° C. As shown in FIG. 4, in sample A, a plurality of quantum dots having a large diameter were produced. Since a dent can be confirmed at the top of the InN quantum dot, the InN quantum dot shown in FIG. 4 is considered to be a combination of a plurality of InN quantum dots.

  FIG. 5 is an AFM observation photograph of Sample B in which an InN quantum dot was produced by irradiating indium equivalent to a 3.55 molecular layer in terms of InN under the condition where the temperature of the GaN substrate was set to 430 ° C. As shown in FIG. 5, the InN quantum dots could be reduced in size by lowering the substrate temperature by 30 ° C. and reducing the amount of indium irradiation. However, it is considered that if the substrate temperature is simply lowered, many defects are formed in the manufactured InN quantum dots and it is difficult to improve the photoelectric conversion efficiency. Therefore, after growing a thin GaN layer on the GaN substrate at a low temperature, an attempt was made to grow InN quantum dots on the surface of the thin GaN layer.

  FIG. 6 shows a case in which GaN having a thickness of 10 nm is epitaxially grown on a GaN substrate adjusted to a temperature of 470 ° C., and then the temperature of the GaN having a thickness of 10 nm is maintained at 470 ° C. It is an AFM observation photograph of the sample C which produced InN quantum dot by irradiating considerable indium. As shown in FIGS. 4 to 6, the temperature of the GaN layer on which the InN quantum dots are formed is higher than that of the sample A, and the amount of indium irradiated is larger than that of the sample A. In Sample C, InN quantum dots could be made smaller than Sample A and Sample B.

  FIG. 7 shows that after 10-nm-thick GaN is epitaxially grown on a GaN substrate whose temperature is adjusted to 430 ° C., the temperature of the 10-nm-thick GaN is maintained at 430 ° C. and 3.55 in terms of InN. It is the AFM observation photograph of the sample D which produced the InN quantum dot by irradiating indium equivalent to a molecular layer. As shown in FIG. 5 and FIG. 7, the InN quantum dots can be made smaller in the sample D as compared with the sample B in which the GaN temperature and the indium irradiation amount at the time of producing the InN quantum dots are similar. did it. Also, as shown in FIGS. 4 to 7, it was sample C that was able to make InN quantum dots the smallest. This is considered to be because, in sample C, the temperature of GaN at which the InN quantum dots were produced was higher than that in sample D, and thus InN, which is the quantum dot material, was re-evaporated into small particles. .

  FIG. 8 is an AFM observation photograph that emphasizes the unevenness of the underlying layer (GaN substrate) of Sample A, and FIG. 9 is a cross-sectional view showing some of the unevenness of FIG. FIG. 10 is an AFM observation photograph that emphasizes the unevenness of the base layer (GaN layer grown at 470 ° C.) of Sample C, and FIG. 11 is a cross-sectional view showing some unevenness of FIG. is there. As shown in FIGS. 9 and 11, sample A in which large InN quantum dots are fabricated has relatively few irregularities on the underlayer, but sample C in which small InN quantum dots are fabricated has many irregularities on the foundation layer. The surface was rough.

  As demonstrated above, the present inventors have found that the InN quantum dots can be reduced in size by producing InN quantum dots in SK mode on the roughened GaN (in the following, This knowledge is referred to as “Knowledge 1”). This mechanism is not clear, but is estimated as follows.

  In general, as a mechanism for generating quantum dots by depositing a material that constitutes quantum dots (hereinafter referred to as “dot material”), a lattice mismatch between a base material and a dot material (a crystal lattice). In order to minimize the strain energy due to the constant mismatch, it is said that the dot material gathers in islands and becomes quantum dots (SK mode). For example, in a system in which InAs quantum dots are produced on GaAs, which is actively studied, there is a lattice mismatch of 7% between the base material (GaAs) and the dot material (InAs), and quantum dots with a diameter of several tens of nanometers are generated. . In a system in which InN quantum dots are formed on the underlying GaN, the lattice mismatch between the two is 11%, and it is considered that quantum dots are generated by the same mechanism. However, since the lattice mismatch between GaN and InN is too large, the familiarity between the quantum dots and the substrate is poor, and the InN quantum dots move around the surface of the GaN, and the quantum dots that have moved around fuse together (coalescence) and become enormous. Presumed to be.

  In order to prevent coalescence, it is considered that the movement of quantum dots may be suppressed or prevented by some method. As a method of suppressing and preventing the movement of the quantum dots, (1) improving the familiarity between the ground and the quantum dots using a small amount of impurities, (2) pinning the quantum dots by lattice defects on the ground surface, etc. It is possible. In “L.Zhou et. Al., Appl.Phys.Lett. 88, p.231906, (2006)” (hereinafter referred to as “Non-Patent Document 1”), quantum dots are generated on threading dislocations. It is reported that it is easy (corresponding to (2) above). As a method for increasing the lattice defects, it is conceivable to grow the base at a low temperature. Therefore, the present inventors presume that it is possible to reduce the size of InN quantum dots by producing InN quantum dots on a substrate (GaN) grown at a low temperature. did.

  FIG. 12 is a band diagram of bonded GaN and InN. The upper side of FIG. 12 has higher electron energy, and the lower side has higher hole energy. “●” in FIG. 12 represents an electron, and the broken line in FIG. 12 is an expected position of an energy level of a crystal defect existing in GaN (hereinafter referred to as “defect level”). Since InN has a large electron affinity, as shown in FIG. 12, the upper end of the forbidden band is at a position lower than the upper end of the forbidden band of GaN. Therefore, if InN and GaN are bonded, it is considered that electrons existing in the conduction band in InN can be captured at the defect level of GaN without losing energy. There are various types of crystal defects, but it is considered that electrons can move along dislocations if they are dislocations that are linear defects. Based on the above theoretical prediction, the present inventors devised the structure shown in FIG. When light is applied, InN quantum dots absorb light and generate photoexcited carriers. If the diameter of the quantum dot is several tens of nanometers or less, the electron energy is kept high by the phonon bottleneck effect. The electrons kept in a high energy state in this way are captured by the defect level of GaN with a certain probability, and if an appropriate electric field is applied here, the electrons can move along the dislocations. Regarding the movement of electrons through defect levels, “L. Ivanova et. Al., Appl. Phys. Lett. Vol. 93, p. 192110 (2008)” (hereinafter referred to as “Non-patent Document 2”). ). In Non-Patent Document 2, the tunnel current when a metal short needle is brought close to a plane parallel to the dislocation in the GaN sample is measured, and the result is analyzed to move electrons through the defect level. Prove that you can.

There are two types of dislocations, edge dislocations and screw dislocations. GaN dislocations are reported in Non-Patent Document 1. FIG. 14 shows a band diagram of edge dislocations of GaN and a band diagram of screw dislocations of GaN reported in Non-Patent Document 1. FIG. 14A is a band diagram of edge dislocations in GaN, and FIG. 14B is a band diagram of screw dislocations in GaN. As shown in FIG. 14A, the edge dislocations in GaN have only one or two defect levels. Therefore, only carriers whose energy matches the defect level can selectively enter the edge dislocations, and the carriers that have entered the edge dislocations can be moved along the dislocations. Therefore, by moving the electrons in the InN quantum dots along the edge dislocations of GaN, it becomes possible to extract only the carriers of specific energy while confining the electrons in the quantum dots. It is thought that it becomes easy to improve the photoelectric conversion efficiency of a solar cell. On the other hand, as shown in FIG. 14B, many defect levels are generated in the screw dislocations in GaN. Therefore, when electrons in the InN quantum dots are moved along the screw dislocations of GaN, the effect of confining the electrons in the quantum dots is weakened. As a result, it is difficult to improve the photoelectric conversion efficiency of the quantum dot solar cell. From the above study, the present inventors have found that it is possible to improve the photoelectric conversion efficiency of the quantum dot solar cell by taking out the electrons in the InN quantum dot through the edge dislocation of GaN. (Hereinafter, this knowledge is referred to as “knowledge 2”.) GaN produced by a known technique includes 10 ⁸ to 10 ¹⁰ cm ⁻² dislocations, most of which are edge dislocations. Therefore, it is considered that a structure in which InN quantum dots and GaN edge dislocations are connected can be realized by forming InN quantum dots on the surface of GaN produced by a known technique.

  As described above, the InN quantum dots can be reduced in size by producing InN quantum dots on the base (GaN) whose lattice defects are increased by growing at a low temperature. However, it is estimated that various lattice defects generated by such a method are included, and many lattice defects harmful to electrical characteristics are also included. As a result of intensive studies, the inventors of the present invention have disclosed a technique (in Japanese Patent No. 4126812) in which hydrogen contains GaN to inactivate dangling bonds of lattice defects and improve electrical characteristics. Technology (hereinafter referred to as “prior patent technology”). However, if this prior art is simply used, edge dislocations in GaN are inactivated, and the mechanism of Knowledge 2 may not function. Therefore, as a result of intensive studies, the inventors of the present invention have made the form in which carriers can move by the tunnel effect, so that the layer (including the base) that is in direct contact with the InN quantum dots is made thin to contain hydrogen. On the other hand, by not containing hydrogen in the layer that is not in direct contact with the InN quantum dots, it becomes possible to take out carriers from the quantum dots while inactivating lattice defects, and improve the photoelectric conversion efficiency of the quantum dot solar cell. (Hereinafter, this knowledge is referred to as “Knowledge 3”).

  This invention is made | formed based on the said knowledge 1-3. The main gist of the present invention is to provide a photoelectric conversion element capable of improving the photoelectric conversion efficiency.

  Hereinafter, the case where this invention is applied to a quantum dot solar cell is demonstrated, referring drawings. In addition, the form shown below is an illustration of this invention and this invention is not limited to the form shown below. In the following description, A to B may be expressed as A to B.

FIG. 1 is a cross-sectional view showing the form of a photoelectric conversion element 10 of the present invention (hereinafter sometimes referred to as “solar cell 10”). In FIG. 1, the description of some symbols is omitted. As shown in FIG. 1, the solar cell 10 of the present invention includes a sapphire substrate 1, a fifth GaN layer 2 formed on the upper surface of the sapphire substrate 1, and a fourth GaN layer formed on the upper surface of the fifth GaN layer 2. 3, a third GaN layer 4 formed on the upper surface of the fourth GaN layer 3, a first GaN layer 5 formed on the upper surface of the third GaN layer 4, and an InN quantum formed on the upper surface of the first GaN layer 5 A layer including dots 6, 6,... (Hereinafter referred to as “InN quantum dot 6-containing layer”), a second GaN layer 7 formed on the upper surface of the InN quantum dot 6-containing layer, and the second GaN layer 7 The fourth GaN layer 3 in contact with the sixth GaN layer 8 formed on the upper surface, the positive electrode 9 formed on the upper surface of the sixth GaN layer 8, and the hole 11 reaching the fourth GaN layer 3 from the positive electrode 9 side. In contact with the negative electrode 12 and the positive electrode 9 Has an electrode 13 of the comb shape formed on so that, the. In the solar cell 10, the third GaN layer 4 is produced by epitaxial growth without mixing hydrogen in an environment in which the temperature of the sapphire substrate 1 is maintained at around 800 ° C. The first GaN layer 5 is made of sapphire. In an environment where the temperature of the substrate 1 is maintained at 360 ° C. to 470 ° C., the substrate 1 is produced by epitaxial growth by mixing hydrogen while p-doping. The InN quantum dot 6-containing layer is produced by epitaxial growth in an environment where the temperature of the sapphire substrate 1 is maintained at 350 ° C. to 470 ° C., and the temperature of the sapphire substrate 1 is 360 °. In an environment maintained at 470 to 470 ° C., it is fabricated by epitaxial growth with hydrogen mixed while p-doping. That is, the InN quantum dot 6-containing layer is formed on the upper surface of the first GaN layer produced by epitaxial growth at a low temperature in a low temperature environment. Further, since the first GaN layer 5 and the second GaN layer 7 contain hydrogen, lattice defects in the first GaN layer 5 and the second GaN layer 7 are inactivated. On the other hand, since hydrogen is not positively added to the third GaN layer 4, lattice defects of the third GaN layer 4 are not inactivated. In the solar cell 10, the fourth GaN layer 3 is an n ⁺ semiconductor layer, the third GaN layer 4 is an n ⁻ semiconductor layer, the first GaN layer 5 and the second GaN layer 7 are p ⁻ semiconductor layers, and the sixth GaN layer 8 is a p ⁺ semiconductor layer. is there.

  FIG. 2 is a flowchart for explaining a manufacturing process of the photoelectric conversion element 10. As shown in FIG. 2, the photoelectric conversion element 10 includes a substrate manufacturing step (S1), a fifth GaN layer manufacturing step (S2), a fourth GaN layer manufacturing step (S3), a fourth step (S4), and a first step (S5). ), Second step (S6), third step (S7), sixth GaN layer preparation step (S8), positive electrode preparation step (S9), hole formation step (S10), negative electrode preparation step (S11), and electrode preparation It is manufactured through the step (S12).

  The substrate manufacturing step S1 can be a step of manufacturing the sapphire substrate 1 by a known method such as vapor deposition. The thickness of the sapphire substrate 1 produced by S1 can be set to 0.3 to 0.5 mm, for example.

  In the fifth GaN layer manufacturing step S2, a gallium molecular beam and a plasma-activated nitrogen beam are formed by a known method such as the MBE method while keeping the temperature of the sapphire substrate 1 manufactured in S1 in the vicinity of 600 to 800 ° C. Is applied to the upper surface of the sapphire substrate 1 to epitaxially grow GaN, thereby forming the fifth GaN layer 2. The thickness of the fifth GaN layer 2 produced by S2 can be set to 1 μm, for example.

In the fourth GaN layer manufacturing step S3, while maintaining the temperature of the sapphire substrate 1 in the vicinity of 600 to 800 ° C., a gallium molecular beam and a nitrogen beam and a silicon beam activated by plasma are obtained by a known method such as the MBE method. By irradiating the upper surface of the fifth GaN layer 2 produced in S2 and epitaxially growing strongly n-doped GaN, the fourth GaN layer 3 can be produced. The thickness of the fourth GaN layer 3 produced by S3 is set to 500 nm to 1000 nm from the viewpoint of obtaining sufficient electric conduction and enabling the hole 11 to be easily formed by etching. The carrier concentration of the fourth GaN layer 3 can be set to 10 ¹⁹ to 10 ²⁰ cm ⁻³ , for example. Here, the intensity of the gallium molecular beam is a value obtained by exposing the beam to a vacuum gauge, and can be 6.67 to 8.00 × 10 ⁻⁵ Pa. The nitrogen beam can be activated by plasma with a plasma excitation microwave intensity of 270 W, and the supply amount of nitrogen gas can be 0.7 cm ³ / min at an atmospheric pressure of 1.013 hPa. The speed of epitaxial growth of the fourth GaN layer 3 can be set to a speed at which the thickness becomes 50 nm in 3 minutes, for example.

In the fourth step S4, while maintaining the temperature of the sapphire substrate 1 at around 600 to 800 ° C., a gallium molecular beam and a plasma-activated nitrogen beam were produced in S3 by a known method such as the MBE method. By irradiating the upper surface of the fourth GaN layer 3 and epitaxially growing n-doped GaN, the fifth GaN layer 4 can be produced. The thickness of the third GaN layer 4 produced by S4 is set to 50 nm to 200 nm from the viewpoint of obtaining sufficient electric conduction and enabling the hole 11 to be easily formed by etching. In S4, the third GaN layer 4 is formed so that the n-doping concentration changes sharply at the boundary between the fourth GaN layer 3 and the third GaN layer 4. A steep change in the n-doping concentration can be realized by closing the shutter of the K cell that supplied the silicon beam in S3. The carrier concentration of the third GaN layer 4 can be set to 10 ¹⁶ to 10 ¹⁷ cm ⁻³ , for example. Since the crystal defects that occur naturally in GaN serve as donors, the carrier concentration of the third GaN layer 4 may be the above concentration even when the doping concentration is zero. The fifth GaN layer 2 produced in S2, the fourth GaN layer 3 produced in S3, and the third GaN layer 4 produced in S4 have threading dislocations having a density of about 10 ⁸ to 10 ¹⁰ cm ^−2. Many of them are edge dislocations.

In the first step S5, while maintaining the temperature of the sapphire substrate 1 at 360 ° C. to 470 ° C., a gallium molecular beam and a nitrogen beam and a hydrogen beam activated by plasma are converted into the above S4 by a known method such as MBE method. While irradiating the upper surface of the third GaN layer 4 produced in step 1, the p-doped GaN (e.g., doped with magnesium) is lightly (10 ¹⁵ to 10 ¹⁷ cm ⁻³ as the carrier concentration) to epitaxially grow the p-doped GaN. Thus, the first GaN layer 5 can be formed on the upper surface of the third GaN layer 4. The thickness of the first GaN layer 5 produced by S5 is 3 nm to 20 nm from the viewpoint of allowing the electrons present in the quantum dots 6, 6,... To pass through the first GaN layer 5 by the tunnel effect. . In order to promptly move to the second step S6, the temperature of the sapphire substrate 1 in S5 is preferably the same as the temperature of the sapphire substrate 1 in the second step S6. In order to keep the quality of InN quantum dots good, it is preferable that the temperature of the sapphire substrate 1 is high, and it is difficult to obtain a good crystal at a temperature lower by 150 ° C. or more than the thermal decomposition temperature. Since the thermal decomposition temperature of InN is about 500 ° C., the lower limit temperature of the sapphire substrate 1 is set to about 360 ° C. in S5. On the other hand, from the viewpoint of obtaining dots with good reproducibility, it is preferable to lower the temperature of the sapphire substrate 1, and it has been confirmed through experiments that 470 ° C. is the upper limit temperature.

In the second step S6, the surface of the sapphire substrate 1 is activated by an indium molecular beam and plasma on the upper surface of the first GaN layer 5 prepared in S5 by a known method such as MBE while keeping the temperature of the sapphire substrate 1 at 350 ° C. to 470 ° C. By supplying the nitrogen beam thus formed, it is possible to prepare a layer containing InN quantum dots 6 in the SK mode on the upper surface of the first GaN layer 5. Here, the nitrogen beam can be activated by plasma with a plasma-excited microwave intensity of 300 W, and the supply amount of nitrogen gas can be 0.7 cm ³ / min at an atmospheric pressure of 1.013 hPa. The indium molecular beam can be irradiated after the nitrogen beam irradiation. The intensity of the indium molecular beam is a value obtained by exposing the beam to a vacuum gauge, which can be 1.33 × 10 ⁻⁵ Pa, and the temperature of the K cell as the indium molecular beam source is 755 ° C. The irradiation time of the indium molecular beam can be 70 seconds. In order to keep the quality of InN quantum dots good, it is preferable that the temperature of the sapphire substrate 1 is high, and it is difficult to obtain a good crystal at a temperature lower by 150 ° C. or more than the thermal decomposition temperature. Since the thermal decomposition temperature of InN is about 500 ° C., the lower limit temperature of the sapphire substrate 1 is set to about 350 ° C. in S6. On the other hand, from the viewpoint of obtaining dots with good reproducibility, it is preferable to lower the temperature of the sapphire substrate 1, and it has been confirmed through experiments that 470 ° C. is the upper limit temperature.

In the third step S7, the strength generated in the K cell at 980 ° C. is 8.00 × 10 ⁻⁵ Pa by a known method such as MBE method while keeping the temperature of the sapphire substrate 1 at 360 ° C. to 470 ° C. In Step S6, a gallium molecular beam, a magnesium molecular beam having a strength of 4.00 × 10 ⁻⁵ Pa generated in a K cell at 290 ° C., and a nitrogen beam and a hydrogen beam activated by plasma are produced in S6. By irradiating the upper surface of the InN quantum dot 6-containing layer for 220 seconds, the second GaN layer 7 having a thickness of 20 nm can be formed on the upper surface of the InN quantum dot 6-containing layer. The thickness of the 2nd GaN layer 7 produced by S7 shall be 10 nm-50 nm from a viewpoint made into the thickness which can move a hole, protecting the shape of quantum dot 6,6, ... produced by said S6. As described above, since the dot changes its shape as it moves around, it is necessary to move to S7 promptly. Therefore, it is preferable that the temperature of the sapphire substrate 1 in S7 is the same as that in S6.

The sixth GaN layer manufacturing step S8 was activated by a gallium molecular beam and plasma by a known method such as the MBE method while keeping the temperature of the sapphire substrate 1 at around 680 ° C. in order to facilitate mixing of magnesium. While irradiating the upper surface of the second GaN layer 7 produced in S7 with a nitrogen beam and a hydrogen beam, magnesium was heavily doped (carrier concentration was 10 ¹⁷ to 10 ¹⁹ cm ⁻³ ) and strongly p-doped. By epitaxially growing GaN, the sixth GaN layer 8 can be formed on the upper surface of the second GaN layer 7. The thickness of the sixth GaN layer 8 produced by S8 is set to 10 nm to 100 nm from the viewpoint of securing a thickness for stably bonding to the electrode. In addition, although the form in which hydrogen was mixed in the 6th GaN layer 8 was illustrated here, hydrogen does not need to be mixed. However, from the viewpoint of terminating dangling of defects that become donors and inhibit p-type formation, it is preferable that the sixth GaN layer 8 also contain hydrogen.

  In the positive electrode manufacturing step S9, a positive electrode 9 made of a known conductive material that can be used as a positive electrode material of a solar cell such as metal is formed on the upper surface of the sixth GaN layer 8 manufactured in S8 by a known method such as vapor deposition. It can be set as the process to do. The positive electrode 9 produced by S9 can be formed by depositing, for example, 10 mm of nickel and 20 mm of gold.

  In the hole forming step S10, the fifth GaN layer 2 to the positive electrode 9 are sequentially removed from the MBE apparatus by lowering the temperature of the fifth GaN layer 2 to the positive electrode 9 by the above S1 to S9 and then taken out from the MBE device. The step of forming the hole 11 reaching 3 can be performed.

  In the negative electrode preparation step S11, a negative electrode 12 made of a known metal that can be used as a negative electrode material for a solar cell is formed by a known method such as vapor deposition so as to come into contact with the fourth GaN layer 3 of the hole 11 produced in S10. It can be a manufacturing process. The negative electrode 12 produced by S11 can be formed by laminating titanium / aluminum / nickel / gold, which are generally well-known, by several millimeters.

  The electrode production step S12 is a step of producing an electrode 13 made of a known conductive material that can be used as an electrode of a solar cell, such as metal, on the upper surface of the positive electrode 9 produced in S9 by a known method such as vapor deposition. can do. The electrode 13 produced by S12 can be gold with a thickness of 50 mm, for example.

FIG. 3 is a band diagram for explaining the operation mode of the solar cell 10. In FIG. 3, “●” represents an electron, “◯” represents a hole, and an arrow represents the direction of carrier movement. As shown in FIG. 3, the solar cell 10 includes a first GaN layer 5, a second GaN layer 7, a sixth GaN layer 8, which are p-type semiconductor layers (including a p ⁻ semiconductor layer and a p ⁺ semiconductor layer), and An internal electric field is generated by the fourth GaN layer 3 and the third GaN layer 4 which are n-type semiconductor layers (including an n ⁻ semiconductor layer and an n ⁺ semiconductor layer).

  Carriers generated in the InN quantum dot 6-containing layer and carriers dropped from the first GaN layer 5 and the second GaN layer 7 to the InN quantum dot 6-containing layer are maintained at a high energy level by the phonon bottleneck effect. Electrons (including photoexcited electrons) present in the InN quantum dot 6 containing layer move to the defect level of the third GaN layer 4 by moving the thin first GaN layer 5 having a thickness of 20 nm or less by the tunnel effect. Be captured. The first GaN layer 5 produced by epitaxial growth at a low temperature contains many lattice defects, and the energy level of the defect level is disturbed because many lattice defects are entangled or large in strain. It is expected that this is unsuitable for moving electrons. However, since the lattice defects are inactivated by adding hydrogen to the first GaN layer 5, the electrons can be moved without being trapped by the defects.

  The electrons captured in the defect level of the third GaN layer 4 in this way move toward the fourth GaN layer 3 by an internal electric field generated by the pn junction. At the boundary between the third GaN layer 4 and the fourth GaN layer 3, as shown in FIG. 3, the band is bent due to the abrupt change in the n-doping concentration. Therefore, the electrons at the defect level of the third GaN layer 4 move to the conduction band of the fourth GaN layer 3 by the tunnel effect, and then move to the negative electrode 12. Here, since the defect level of the fourth GaN layer 3 is already strongly n-doped, it is filled with electrons, and it is considered that the electrons that are photoexcited carriers do not fall here.

  On the other hand, holes (including photoexcited holes) existing in the InN quantum dot 6 containing layer are contained in the InN quantum dot 6 due to the electric field generated by the difference in the p-doping level between the second GaN layer 7 and the sixth GaN layer 8. It escapes from the layer and moves to the positive electrode 9 through the second GaN layer 7 and the sixth GaN layer 8.

In the above description related to the present invention, the fourth GaN layer 3 containing no In is exemplified, but the present invention is not limited to this form. In the present invention, it is preferable that the fourth GaN layer is composed of strongly n-doped In _x Ga _1-x N. If x is a value around 0.5, the energy at the lower end of the conduction band of the fourth GaN layer 3 is determined. Since the level can be brought close to the defect level of the third GaN layer 4, it is more preferable. By bringing the energy level at the lower end of the conduction band of the fourth GaN layer 3 close to the defect level of the third GaN layer 4, electrons are transferred from the third GaN layer 4 to the fourth GaN layer 3 without bending the band of the fourth GaN layer 3. It can be moved. By adopting such a configuration, it becomes possible to reduce the restriction on the accuracy of the n-doping concentration of the third GaN layer 4 and the fourth GaN layer 3, so that the solar cell can be easily manufactured.

  Moreover, in the said description regarding this invention, although the form in which only one InN quantum dot 6 content layer was contained was illustrated, this invention is not limited to the said form, InN quantum dot 6 of two or more layers It is also possible to adopt a form in which the inclusion layer is laminated with the GaN layer interposed therebetween.

  As mentioned above, although the case where this invention was applied to the quantum dot solar cell was demonstrated, this invention is not limited to the said form, It applies also to the photoelectric conversion element of other forms represented by the photodetection element etc. It is also possible.

  The photoelectric conversion element of this invention can be utilized for the power source of an electric vehicle, a solar power generation system, etc.

DESCRIPTION OF SYMBOLS 1 ... Substrate 2 ... 5th GaN layer 3 ... 4th GaN layer 4 ... 3rd GaN layer 5 ... 1st GaN layer 6 ... InN quantum dot 7 ... 2nd GaN layer 8 ... 6th GaN layer 9 ... Positive electrode 10 ... Photoelectric conversion element 11 ... Hole 12 ... Negative electrode 13 ... Electrode

Claims (18)

A photoelectric conversion element comprising: a first gallium nitride layer manufactured by growing at a low temperature; and an indium nitride quantum dot formed on a surface of the first gallium nitride layer. The photoelectric conversion element according to claim 1, wherein the first gallium nitride layer is a p-type semiconductor layer. The photoelectric conversion element according to claim 1, wherein a second gallium nitride layer is disposed on a surface of the indium nitride quantum dots. The photoelectric conversion element according to claim 3, wherein hydrogen is contained in the first gallium nitride layer and the second gallium nitride layer. 5. The photoelectric conversion element according to claim 3, wherein a doping concentration of the second gallium nitride layer is higher than a doping concentration of the first gallium nitride layer. The photoelectric conversion element according to claim 3, wherein the second gallium nitride layer is a p-type semiconductor layer. The photoelectric conversion element according to claim 1, wherein the first gallium nitride layer is disposed on a surface of the third gallium nitride layer. The photoelectric conversion element according to claim 7, wherein hydrogen is not contained in the third gallium nitride layer. The photoelectric conversion element according to claim 7, wherein the third gallium nitride layer is an n-type semiconductor layer. A first step of forming a gallium nitride layer grown at a low temperature; and a second step of forming indium nitride quantum dots on the surface of the gallium nitride layer formed by the first step. The manufacturing method of a photoelectric conversion element. The method for manufacturing a photoelectric conversion element according to claim 10, wherein the gallium nitride layer formed in the first step is a p-type semiconductor layer. 12. The method according to claim 10, further comprising a third step of forming a gallium nitride layer grown at a low temperature on the surface of the indium nitride quantum dots formed in the second step. A method for producing a photoelectric conversion element. 13. The method for manufacturing a photoelectric conversion element according to claim 12, wherein the first step and the third step are steps of forming the gallium nitride layer while adding hydrogen. 14. The photoelectric device according to claim 12, wherein a doping concentration of the gallium nitride layer formed in the third step is higher than a doping concentration of the gallium nitride layer formed in the first step. A method for manufacturing a conversion element. The method for manufacturing a photoelectric conversion element according to any one of claims 12 to 14, wherein the gallium nitride layer formed in the third step is a p-type semiconductor layer. Furthermore, it has a 4th process of forming the gallium nitride layer by which the gallium nitride layer which should be formed at the said 1st process is arrange | positioned on the surface, The any one of Claims 10-15 characterized by the above-mentioned. Manufacturing method of the photoelectric conversion element. The method of manufacturing a photoelectric conversion element according to claim 16, wherein the fourth step is a step of forming the gallium nitride layer without adding hydrogen. 18. The method for manufacturing a photoelectric conversion element according to claim 16, wherein the gallium nitride layer formed in the fourth step is an n-type semiconductor layer.