Device having double tunnel structure

Abstract

Provided is a device comprising a tunnel junction which can maintain a state of continuing to flow to a wide gap semiconductor side without returning electrons accumulated in a hetero interface to a narrow gap semiconductor side, and which can suppress the diffusion of electrons after excitation without restricting the movement of electrons before excitation, wherein output power is increased by increasing an open-circuit voltage. A solar cell 100 and an infrared sensor 110 have an undoped compound semiconductor layer 4 sandwiched by either or both of n-type or p-type doped compound semiconductor layers (2, 3). The undoped compound semiconductor layer 4 is a heterojunction of a wide gap semiconductor and a narrow gap semiconductor, and has, on the narrow gap semiconductor side near a hetero-interface 6, a quantum dot region 7 and a three-layer structure (double tunnel structure) 1 in which a p+-type degenerate semiconductor layer is sandwiched between two n+-type degenerate semiconductor layers.

Description

Devices with double tunnel structure

[0001] The present invention relates to devices having a double tunnel structure, particularly to high-efficiency solar cells and infrared sensors.

[0002] Solar cells are not only a research and development field for reducing greenhouse gas emissions and providing a clean energy source, but also an industrial technology that holds promise for sustainable economic growth. Most conventional solar cells use silicon (Si), but they transmit light with wavelengths greater than those corresponding to the bandgap of Si (approximately 1.1 eV), which prevents the utilization of long-wavelength light (infrared light) contained in sunlight. To address this issue, the inventors have previously proposed a two-step photon upconversion solar cell consisting of a heterojunction of two semiconductors, a wide-gap semiconductor and a narrow-gap semiconductor (see Non-Patent Document 1). In this two-step photon upconversion solar cell, electrons generated in the narrow-gap semiconductor accumulate at the heterointerface between the wide-gap and narrow-gap semiconductors. The electrons accumulated at the heterointerface are excited (upconverted) by long-wavelength light (low-energy photons) in sunlight, thereby increasing current without reducing voltage. In other words, light that would otherwise be lost in a short-circuited single-junction solar cell is absorbed by the narrow-gap semiconductor, excited (first-stage excitation), and the electrons are accumulated at the heterointerface. These electrons are then excited by infrared light into the wide-gap semiconductor (second-stage excitation), allowing the light component that would otherwise be lost in a single-junction solar cell to contribute to power generation, achieving high conversion efficiency. In addition, by incorporating quantum dots into the heterointerface between the wide-gap and narrow-gap semiconductors, the optical transition selection rules between intra-band levels are relaxed, increasing the excitation efficiency of the accumulated electrons and making it possible to increase electrons more efficiently.

[0003] The inventors have demonstrated that this two-step photon upconversion solar cell can significantly increase the current, but the increase in voltage was smaller than expected. This is because the upconverted electrons return to the narrow-gap semiconductor in the open state where no current flows. For this reason, it is difficult to further improve the conversion efficiency with the current structure.

[0004] Furthermore, it has been shown that in two-stage photon upconversion solar cells, the second stage of excitation at the heterointerface enhances the short-circuit current and improves the maximum output power. However, the increase in open-circuit voltage, another factor that contributes to power generation efficiency, is only slight. Therefore, to further improve conversion efficiency, it is necessary to increase the open-circuit voltage and increase the output power.

[0005] That is, in order to maintain a state in which electrons do not return to the narrow-gap semiconductor side but continue to flow to the wide-gap semiconductor side even in an energy band state when the voltage is close to the open-circuit voltage, a barrier is installed on the narrow-gap semiconductor side of the heterointerface to restrict the movement of excited electrons toward the narrow-gap semiconductor side. However, simply installing a barrier will hinder the movement of electrons from the narrow-gap semiconductor side to the heterointerface, preventing them from reaching the heterointerface and resulting in a problem of electron accumulation.

[0006] For this reason, the inventors have investigated a tunnel structure that allows electrons to reach the heterointerface without restricting their movement before excitation and that can suppress the diffusion of electrons after excitation. A tunnel diode is a typical example of a tunnel structure. A tunnel diode is fabricated by doping p-type and n-type dopants at high concentrations with a film thickness of several tens of nanometers. Its characteristic I-V characteristics include a portion called negative resistance. This is a characteristic when a tunnel diode is forward biased, in which the current increases due to the tunnel current up to a certain voltage, but after reaching a peak, the current decreases, and then the current value increases again. Negative resistance is observed in the portion where this current decreases, and in this portion, the tunnel path enters the forbidden band, causing the n ＋ From inside the conduction band of the layer ＋ Tunneling into the valence band of the layer is not possible, resulting in a decrease in current. After the decrease, the current increases again, which is the same operation principle as that of a normal semiconductor after the threshold voltage. Furthermore, operation under negative bias can be explained by the same principle as that of a Zener diode, and as the voltage increases in the negative direction, the reverse current also increases.

[0007] In the field of solar cells, an application of the tunnel diode is known as a tunnel junction that connects each subcell in a tandem multi-junction solar cell (see Patent Documents 1 to 4). In the solar cell disclosed in Patent Document 1, p ＋ type (Al)GaInAs layer and n ＋ The tunnel junction layer is formed by joining the n-type InP layer to the n-type InP layer, and the resistance of the junction interface in the tunnel junction layer is extremely low. ＋ From the semiconductor layer side of the p ＋ This tunnel junction layer is a three-layer tunnel junction layer (p ＋ type GaAs layer, p ＋ type GaInAs layer, n ＋In the solar cell disclosed in Patent Document 2, a tunnel junction is formed between a cell composed of an n-type GaN layer, a light absorption layer, and a p-type GaN layer, and another cell composed of an n-type GaN layer, a light absorption layer, and a p-type GaN layer, and the tunnel junction layer is a three-layer tunnel junction layer (p ＋ type GaN layer, n ＋ In the solar cell disclosed in Patent Document 3, a four-layer tunnel junction layer (n-type GaN layer and n-type GaN layer) is formed between the GaAs bottom cell and the top cell. ＋ type AlInP layer, n ＋ type InGaP layer, p ＋ type InGaP layer, p ＋ A type AlInP layer is shown as an example.

[0008] JP 2015-46572 A JP 2017-85055 A JP 2001-102608 A

[0009] Shigeo Asahi1 et al., “Two-step photon up-conversion solar cells”, nature communications, DOI:10.1038 / ncomms14962.

[0010] As mentioned above, to further improve the conversion efficiency of solar cells, it is necessary to increase the open-circuit voltage and increase the output power. Therefore, in the energy band state when the voltage is close to the open-circuit voltage, it is necessary to keep the electrons accumulated at the heterointerface at the heterointerface without returning them to the narrow-gap semiconductor side, and to excite the electrons so that they continue to flow to the wide-gap semiconductor side. In addition, a tunnel junction is required that does not restrict the movement of electrons before excitation and can suppress the diffusion of electrons after excitation.

[0011] In view of this situation, an object of the present invention is to provide a device that has a tunnel junction that prevents electrons accumulated at a heterointerface from returning to the narrow-gap semiconductor side, keeps them flowing to the wide-gap semiconductor side, does not restrict the movement of electrons before excitation, and restricts the suppression of electrons after excitation, thereby increasing the open-circuit voltage and increasing output power.

[0012] In order to solve the above problems, the present invention provides a device that extracts carriers generated by sunlight as a photocurrent, comprising: ＋ Two layers of degenerate semiconductor layers ＋ a three-layer structure sandwiched between degenerate semiconductor layers, or ＋ Two layers of p ＋ The above-mentioned three-layer structure (also referred to as a double tunnel structure in this specification) functions as a barrier layer that unidirectionally controls the movement of carriers generated by sunlight components, and while allowing electrons (or holes) to move by the tunnel effect, it is possible to restrict the diffusion of electrons (or holes) photoexcited at the heterointerface, thereby increasing the open-circuit voltage and increasing the output power. The double tunnel structure is a structure in which n-type semiconductor layers are doped with highly concentrated impurities. ＋ p ＋ n ＋ It has a junction structure, and the middle p ＋ The barrier layer is formed by a degenerate n-type layer so that electrons can be transported toward the heterointerface by two interband tunnelings at energies near the Fermi level of the barrier layer. ＋ Two layers of p ＋ This structure sandwiches a layer between them. For example, in a device such as a solar cell in which a wide-gap semiconductor and a narrow-gap semiconductor are heterojunctioned, providing a double tunnel structure near the heterointerface can prevent electrons excited by long-wavelength components of sunlight (e.g., infrared light components) at the heterointerface from diffusing to the narrow-gap semiconductor. Here, sunlight components refer to components consisting of ultraviolet light, visible light, and infrared light that reach the Earth's surface.

[0013] The double tunnel structure allows carriers near the bottom of the conduction band or the top of the valence band to move by two interband tunnelings, while unidirectionally guiding carriers that are photoexcited at the heterointerface and have energy greater than the energy of the bottom of the conduction band or the top of the valence band. In other words, the double tunnel structure functions as a tunneling layer that generates two interband tunnelings and a barrier layer that unidirectionally guiding carrier movement.

[0014] The device of the present invention preferably has an undoped compound semiconductor layer sandwiched between doped compound semiconductor layers of either n-type or p-type, or both. The undoped compound semiconductor layer is a heterojunction between a wide-gap semiconductor and a narrow-gap semiconductor, and a quantum dot region and the above-described three-layer structure are provided near the heterointerface on the narrow-gap semiconductor side. In the quantum dot region, an energy level is formed lower than the bottom of the conduction band of the narrow-gap semiconductor. The Fermi level is the energy position at which there is a 50% probability that an electron will exist. If the energy level of the conduction band at the heterointerface is lower than the Fermi level, it becomes possible to accumulate many electrons at the heterointerface. Here, the Fermi level is uniquely determined by the system (the impurity concentration and thickness of the doped compound semiconductor layer, and the device structure, such as the composition and thickness of the undoped compound semiconductor layer sandwiched therebetween). The device of the present invention may also have an n-type or p-type doped compound semiconductor layer and an undoped compound semiconductor layer, the undoped compound semiconductor layer being a heterojunction between a wide-gap semiconductor and a narrow-gap semiconductor, and a quantum dot region and a three-layer structure being provided on the narrow-gap semiconductor side near this heterointerface. This makes it possible to utilize long-wavelength light (infrared light). However, conventional solar cells cannot utilize long-wavelength light (infrared light).

[0015] In the device of the present invention, quantum dots are located at the heterointerface of the undoped compound semiconductor layer, and by accumulating many electrons at this heterointerface, further improvement in conversion efficiency can be achieved. In an ideal two-dimensional heterointerface, optical transitions between intraband levels are prohibited. However, by providing quantum dots at the heterointerface, the optical transition selection rules are relaxed, thereby increasing the excitation efficiency of the accumulated electrons. The quantum dot region forms a conduction band energy level lower than the Fermi level, and electrons accumulated at the heterointerface are excited from the intraband level formed in the quantum dot region by sunlight components and extracted as photocurrent. Examples of heterogeneous compound semiconductors that constitute the heterojunction include II-VI group heterogeneous compound semiconductors, III-V group heterogeneous compound semiconductors, and Group IV semiconductors such as Si-based, Ge-based, and C-based.

[0016] Furthermore, the undoped compound semiconductor layer (hereinafter also referred to as an i-layer) has a structure sandwiched between either an n-type doped compound semiconductor layer (hereinafter also referred to as an n-layer) or a p-type doped compound semiconductor layer (hereinafter also referred to as a p-layer), i.e., has any of the following structures: n-layer-i-layer-p-layer, p-layer-i-layer-n-layer, n-layer-i-layer-n-layer, or p-layer-i-layer-p-layer.

[0017] The quantum dot region is preferably composed of a single layer or multiple layers and has a thickness of 50 nm or less. In conventional solar cells, the conversion efficiency could not be improved unless multiple quantum dot regions were stacked, but when the device of the present invention is applied to a solar cell, sufficient sensitivity can be obtained even if the quantum dot region is a single layer.

[0018] In the device of the present invention, the three-layer structure is preferably such that the end face of the upper layer of the three-layer structure is located at a distance of 3 to 50 nm from the heterointerface. The end face on the upper layer side is the end face located on the side that receives light. The closer the end face of the three-layer structure is to the heterointerface, the better. In order to form a quantum dot region near the heterointerface, the distance from the heterointerface is set to 3 nm or more. Furthermore, if the distance from the heterointerface exceeds 50 nm, the effect of preventing the diffusion of carriers excited by intraband transition becomes smaller.

[0019] In the device of the present invention, the heterojunction is preferably made of a III-V compound semiconductor, and the quantum dot region is preferably made of quantum dots of any of InAs, InSb, InGaAs, GaSb, Ge, and HgCdTe. The device of the present invention can be used as a solar cell or an infrared sensor for detecting infrared light.

[0020] The device of the present invention has the effect of preventing electrons accumulated at the heterointerface from returning to the narrow-gap semiconductor side and maintaining a state in which they continue to flow to the wide-gap semiconductor side, providing a tunnel junction that does not restrict the movement of electrons before excitation and can restrict the diffusion of electrons after excitation, thereby increasing the open-circuit voltage and increasing output power.When the device of the present invention is used as a solar cell, a highly efficient solar cell can be realized.Furthermore, when the device of the present invention is used as an infrared sensor, infrared light can be detected with high sensitivity.

[0021] Schematic diagram of a high-efficiency solar cell of Example 1. Explanatory diagram of the configuration of a high-efficiency solar cell of Example 1. Explanatory diagram of electron behavior (effect of functioning as a barrier layer). Explanatory diagram of electron behavior (electron transport by tunneling effect). Explanatory diagram of the band structure of a solar cell. Explanatory diagram of the band structure of a high-efficiency solar cell of Example 1. Graph showing the effect of increasing open-circuit voltage. Schematic diagram of a measurement system accompanying changes in infrared light intensity. Graph showing the effect of increasing short-circuit current density depending on the distance from the heterointerface of a double tunnel structure. Schematic diagram of a solar cell module. Explanatory diagram of the band structure of a high-efficiency solar cell of Example 2. Explanatory diagram of the configuration of a high-efficiency solar cell of Example 2. Outline diagram of a high-sensitivity infrared sensor of Example 3. Graph showing the irradiation intensity dependence of photocurrent and photovoltage upon infrared light irradiation.

[0022] An example of an embodiment of the present invention will be described in detail below with reference to the drawings. Note that the scope of the present invention is not limited to the following examples and illustrated examples, and many modifications and variations are possible.

[0023] First, the behavior of carriers in the three-layer structure (double tunnel structure) of the device of the present invention will be described. Here, the behavior of electrons as carriers will be described with reference to the figures, showing the effect of the double tunnel structure functioning as a barrier layer and the effect of the double tunnel structure functioning as a tunneling layer that causes two interband tunnelings. Figures 3 and 4 are explanatory diagrams of the behavior of electrons (the effect of functioning as a barrier layer) and the behavior of electrons (the effect of functioning as a tunneling layer), respectively.

[0024] The first function of the double tunnel structure is to prevent electrons excited intraband (also called intraband transition) by long-wavelength light (e.g., infrared light) contained in sunlight from diffusing toward the narrow-gap semiconductor by functioning as a barrier layer in the double tunnel structure. Because the barrier layer is approximately several tens of nanometers thick, it is believed that excited electrons cannot flow toward the narrow-gap semiconductor (NGS) by intraband tunneling. As shown in Figure 3 , when sunlight components are incident from the wide-gap semiconductor (WGS) side, interband transitions (12, 16) occur in the band gap portion of the energy corresponding to each solar spectrum, generating carrier electrons (13, 17). The electron 13 then drifts 14 due to the internal electric field, and electron 18 accumulates in the quantum dot region at the heterointerface through two interband tunneling events (described below). The electron 18 then undergoes intraband transition 19 due to infrared light (long-wavelength light). The electrons 20 undergoing the intraband transition 19 drift 21 toward the WGS side, but are prevented from diffusing 22 toward the NGS side by the tunnel barrier layer.

[0025] Next, focusing on the movement of electrons generated by band-to-band transition in GaAs, a narrow-gap semiconductor, the second operation of the double tunnel structure will be explained with reference to Figure 4. Most of the electrons generated in GaAs are generated in the i-layer, which is not doped with impurities. As shown in Figure 4, the generated electrons are transported by the internal electric field from the narrow-gap semiconductor (NGS) side of the tunnel barrier layer to the n-layer of the first layer. ＋ The electrons accumulate there, and n ＋ The quasi-Fermi level of electrons on the tunnel barrier layer rises, but ＋ Since the layers are also degenerated, electrons undergo band-to-band tunneling 23 from the conduction band (CB) to the valence band (VB). This corresponds to the operation of a tunnel diode up to the tunnel peak threshold voltage when forward biased. ＋ As the quasi-Fermi level of the electrons in the second layer also rises, ＋ Layer and p ＋As the difference in quasi-Fermi levels of electrons in the layers widens, band-to-band tunneling 24 occurs from the valence band (VB) to the conduction band (CB). This corresponds to the operation of a tunnel diode under reverse bias. Electrons 18 are then accumulated in the quantum dot region at the heterointerface, and long-wavelength light (e.g., infrared light) induces intraband transition 19. This electron behavior makes it possible to extract electrons without a tunnel barrier layer. Note that in existing tunnel junctions, generated electrons are transferred from the conduction band (CB) to the valence band (VB) by band-to-band tunneling, so the above-mentioned intraband transition 19 does not occur. The double tunnel structure in the device of the present invention is a structure that allows the barrier layer to suppress diffusion 22 of electrons 20 that have undergone intraband transition 19 while leaving electrons 18 in the conduction band.

[0026] FIG. 1 shows a schematic diagram of a solar cell according to one embodiment of the present invention. As shown in FIG. 1, a solar cell (also referred to as a cell) 100 of the present invention has a structure including an n-type doped compound semiconductor layer 2, a p-type doped compound semiconductor layer 3, an undoped compound semiconductor layer 4 sandwiched between them, and an electrode 5. The undoped compound semiconductor layer 4 is formed as a heterojunction of different semiconductors, with a quantum dot region 7 near the heterointerface 6 and a three-layer structure (double tunnel structure) 1 below it. A semiconductor substrate 8 is located below the p-type doped compound semiconductor layer 3, and a back electrode 52 is located below that. An electrode 51 is located above the n-type doped compound semiconductor layer 2, and the energy of sunlight incident from above is converted in the undoped compound semiconductor layer 4 to generate photocurrent. In the quantum dot region 7 of the undoped compound semiconductor layer 4, electrons excited by sunlight escape from the quantum dot wells, and many electrons accumulate at the heterointerface 6, which are then extracted as photocurrent. As shown in FIG. 1, a bias voltage is applied to the solar cell, and photocurrent generated when sunlight is incident is extracted.

[0027] A specific configuration of the solar cell of Example 1 will be described. As shown in FIG. 2, the solar cell of Example 1 has an nip structure, and an n-type doped compound semiconductor layer 2 is made of n-Al. 0.3 Ga 0.7As, p-type doped compound semiconductor layer 3 ＋ -GaAs, and as the undoped compound semiconductor layer 4, Al 0.3 Ga 0.7 A heterojunction between As and GaAs is formed, and a quantum dot region 7 of InAs quantum dots (QDs) is formed at the heterointerface 6. The undoped compound semiconductor layer 4 has a p-type doped compound semiconductor layer 3 side of the quantum dot region 7. ＋ -GaAs degenerate semiconductor layer (p ＋＋ -GaAs) in two layers ＋ -GaAs degenerate semiconductor layer (n ＋＋ The double tunnel structure 1 is formed by sandwiching the Al-GaAs layer between the Al-GaAs and the GaAs layers. The double tunnel structure 1 is a structure in which two tunnel barrier layers in a tunnel diode are stacked. In the solar cell of Example 1, the Al-GaAs layer is used as the wide-gap semiconductor. 0.3 Ga 0.7 The double tunnel structure 1 suppresses the diffusion of electrons at the heterointerface toward the narrow-gap semiconductor (GaAs), while ensuring the transport of electrons to the heterointerface by tunnel current. ＋ The quantum dot regions 7 are stacked in this order on a GaAs semiconductor substrate 8, and electrodes 5 are provided on the top and bottom surfaces thereof. In the first embodiment, the quantum dot region 7 is a single layer, but multiple quantum dot regions 7 may be stacked.

[0028] The doping concentration and thickness of the tunnel barrier layer in the double tunnel structure will be explained. Generally, the conditions for the tunnel junction part set in a tandem solar cell, etc., are a total thickness of 100 nm or less, a tunnel peak current density J ｐｅａｋ = 20 mA / cm² or more, but because it has the same structure as one layer of the implemented tunnel junction, two tunnel barrier layers were fabricated within a total thickness of 100 nm. As will be described later, the distance from the heterointerface to the tunnel barrier layer was set to 15 nm, and n ＋A 5 nm thick i-GaAs layer was grown from the heavily doped layer, quantum dots were grown thereon, and a 10 nm thick i-GaAs layer was further stacked on the grown quantum dots as a capping layer.

[0029] A method for manufacturing the solar cell of Example 1 will be described. The solar cell of Example 1 was fabricated by solid-source molecular beam epitaxy. ＋ The GaN layer was fabricated on a GaAs (001) substrate. ＋ A p-GaAs layer was grown on the substrate as a buffer layer, and a 150 nm thick p-GaAs layer was grown on top of that. During crystal growth, the substrate temperature was monitored using an infrared pyrometer. A 1000 nm thick GaAs layer was then deposited on top of that as an undoped compound semiconductor layer (i-layer), and a 26 nm thick n-GaAs layer was then grown on top of that. ＋ -GaAs degenerate semiconductor layer (n ＋＋ -GaAs), 28 nm thick p ＋ -GaAs degenerate semiconductor layer (p ＋＋ -GaAs), 26 nm thick n ＋ -GaAs degenerate semiconductor layer (n ＋＋ A three-layer structure (double tunnel structure) of InAs and GaAs was formed. A 5-nm-thick GaAs layer was deposited on the double tunnel structure, and an InAs quantum dot layer (QD layer) was formed on this. The nominal thickness of the InAs was 0.64 nm (2.1 monolayers). The typical height and width of the QD layer were 3 nm and 20 nm, respectively, and the quantum dot density was approximately 1.0 × 10 １０ cm ３ The substrate temperature before the deposition of InAs quantum dots was 550° C. GaAs was further deposited to a thickness of 10 nm, and Al was deposited to a thickness of 250 nm. 0.3 Ga 0.7 As is deposited, and Al is 0.3 Ga 0.7 An intrinsic carrier layer with a heterojunction compound semiconductor structure of As / GaAs was fabricated. An InAs QD layer followed by a 10 nm GaAs capping layer was grown at 490 °C. Finally, a 150 nm thick n-Al contact layer was grown. 0.3 Ga 0.7 As layer, 30 nm thick ＋ -Al0.3 Ga 0.7 As layer, 30 nm thick ＋ -Al 0.8 Ga 0.2 As layer, 200 nm thick ＋ -GaAs was grown on the substrate. Next, an Au-Ge / Au electrode was formed as the top electrode, and an Au-Zn / Au electrode was formed as the bottom electrode. Note that the above-described method for manufacturing a solar cell is merely an example, and the structure of the solar cell of the present invention is not limited to this. As long as the same structure can be obtained, there are no restrictions on the manufacturing method or materials. Furthermore, in order to obtain higher conversion efficiency, it is possible to optimize the thickness and doping concentration of each layer, introduce anti-reflection coating, etc.

[0030] Figure 5 is an explanatory diagram of the band structure of a solar cell. The band structure shown in the figure was created using simulation software (COMSOL Multiphysics), which was used to obtain the band structure in a short-circuit state before actually fabricating the device. COMSOL is software that can simultaneously apply multiple physical phenomena to a single device and simulate each stage, such as design, process, and device, and was used to simulate the performance characteristics of the solar cell. As shown in Figure 5, the band structure of the high-efficiency solar cell device of Example 1 in a short-circuit state, drawn using COMSOL, shows that a barrier layer is formed due to degeneration caused by high doping concentration near the heterointerface. Note that Figure 5(1) shows the band structure of a solar cell device without a double tunnel structure, and Figure 5(2) shows the band structure of the high-efficiency solar cell device of Example 1 with a double tunnel structure. As explained below, the band structure of the high-efficiency solar cell device of Example 1, which is heavily doped with impurities (degenerate), is ＋ p ＋ n ＋ In the double tunnel structure of the junction, degenerate n ＋ Two layers of p ＋ Since the layers are sandwiched, p ＋ At an energy level near the Fermi level of the barrier layer, electrons can be transported toward the heterointerface by two band-to-band tunnelings. ＋The barrier layer can prevent electrons excited by the infrared light component at the heterointerface from diffusing toward the narrow-gap semiconductor.

[0031] The band structure shown in FIG. 5 shows the case of a short circuit. Electrons excited by sunlight flow from the p-GaAs side to the n-Al 0.3 Ga 0.7 At the heterointerface (position = 660 nm), the Fermi level (E ｆ ) is the bottom level of the conduction band (E Ｃ ), which indicates that electrons are present at a high density. That is, the quantum dot region increases the Fermi level (E ｆ ) lower energy level of the conduction band (E Ｃ ) is formed, and a large amount of electrons accumulate in this region, increasing the amount of photocurrent and improving the conversion efficiency of the solar cell. However, in the band structure shown in Figure 5(1), when electrons accumulated at the heterointerface are excited and move toward the wide-gap semiconductor, some of the electrons return to the narrow-gap semiconductor and diffuse toward the narrow-gap semiconductor. This diffusion of electrons is particularly pronounced when a positive bias is applied to the solar cell. Therefore, in actual solar cells, the increase in photovoltage is small relative to the increase in photocurrent, and the conversion efficiency of the solar cell is not sufficiently improved. Therefore, as shown in the band structure shown in Figure 5(2), a barrier layer is provided with a double tunnel structure, so that when electrons accumulated at the heterointerface are excited and move toward the wide-gap semiconductor, they do not return to the narrow-gap semiconductor and continue to flow toward the wide-gap semiconductor.

[0032] Figure 6 shows the band structure of the high-efficiency solar cell of Example 1. As with the band structure of Figure 5, it was created using COMSOL Multiphysics. As shown in the figure, a highly doped n-type GaAs is located near the heterointerface (at 660 nm) between the wide-gap semiconductor (WGS) and the narrow-gap semiconductor (NGS). ＋ p ＋ n ＋A double tunnel structure consisting of a junction is provided. In the double tunnel structure, electrons are supplied to the hetero interface by two interband tunnelings, while the potential barrier prevents the upconverted electrons from diffusing to the NGS side. By incorporating the double tunnel structure, a voltage boost effect can be achieved.

[0033] Figure 7 is a graph showing the effect of increasing the open-circuit voltage in an actually fabricated solar cell (see Figure 2). In the measurement, the intensity of the first-stage interband excitation light that excites GaAs was fixed, and the increase in short-circuit current density and open-circuit voltage (ΔJ ＳＣ , ΔV ＯＣ ) was measured by varying the intensity of the added infrared light (i.e., the intensity of the intraband excitation light). Specifically, as shown in Figure 8, the measurement experiment was performed using two lasers: a 784 nm wavelength CW (Continuous Wave) laser 41 (manufactured by TOPTICA Photonics, model number iBeam-smart-785-S) and a 1319 nm wavelength IR (Infrared) laser 42 (manufactured by Changchun New Industries, model number MSL-FN-1319), and a tunable ND filter 44 to vary the laser intensities. The two lasers were used as excitation light sources for interband transitions in GaAs (784 nm) and intraband transitions at heterointerfaces (1319 nm), respectively. Diffracted light from the IR laser was cut by inserting a long-pass filter 46 between the lasers. By controlling the tunable ND filter 44, the intensity of the incident excitation light could be varied without changing the laser intensity. Each laser beam was passed through a condenser lens 47, then incident on an optical fiber 45, and condensed and irradiated onto the upper part of the solar cell 1, which was the sample. Among these, the increase in the short-circuit current density of the solar cell 1 (ΔJ ＳＣ ), the increase in open circuit voltage (ΔV ＯＣ ) was measured using a Source Meter 43 (Tektronix, Model No. 2401). All experiments were carried out at room temperature (291 K) and atmospheric pressure. The short-circuit current increase ΔJ observed in the experiment when the solar cell was a single-junction solar cell wasＳＣ The open circuit voltage increase amount ΔV′ estimated from ＯＣ The ΔV observed in the experiment was calculated. ＯＣ is ΔV' ＯＣ If it exceeds , it can be interpreted as a voltage increase. ＯＣ and ΔV' ＯＣ The results of comparing the voltage rise and the voltage drop are shown below. As a result, it was observed that the voltage rise increased as the irradiation intensity of the additional infrared light increased. This is the result of the double tunnel structure making the voltage rise more pronounced.

[0034] Next, as shown in the graph of FIG. 9, by changing the distance from the heterointerface of the double tunnel structure from 30 nm to 15 nm, the short-circuit current density (ΔJ ＳＣ In addition, it was observed that the increase in short-circuit current density (ΔJ ＳＣ ), the increase in open circuit voltage (ΔV ＯＣ ) may increase.

[0035] Since the output voltage of a single solar cell is low at a few volts, to obtain the required voltage, several solar cells are connected in series, and these are then further connected in series or parallel to form a solar cell module. The solar cell module is sealed to withstand the surrounding environment, and then fitted into an outer frame to provide overall strength. An example of the structure of the solar cell module is shown below, but the present invention is not limited to this, and known solar cell module structures can be applied to the high-efficiency solar cell with a double tunnel structure of the present invention.

[0036] FIG. 10 shows a schematic diagram of a solar cell module. FIG. 10(1) shows several solar cells (also simply referred to as cells) connected in series and then in parallel. FIG. 10(2) is a cross-sectional view taken along the line A-A in FIG. 10(1). The solar cell module 30 shown in FIG. 10(1) is composed of 60 cells 31 and fitted into an outer frame 32. As shown in FIG. 10(2), the solar cell module includes cells 31 sealed with a sealant 34 such as EVA (ethylene vinyl acetate copolymer) or PVB (polyvinyl butyral) resin, and a transparent layer 33 made of glass or other light-transmitting material and a protective layer 35 made of multiple layers of films such as PVF (polyvinyl fluoride) or PET (polyethylene terephthalate) on the top and bottom surfaces. Each cell 31 is electrically connected to another cell by a thin metal plate 36 made of a thin metal plate. The photocurrent from each cell 31 is accumulated and led to an external terminal (not shown). An outer frame 32 is provided on the periphery of the structure in which the plurality of cells 31 are sealed between the light-transmitting layer 33 and the protective layer 35 to support the structure.

[0037] Another embodiment of the high-efficiency solar cell of the present invention will now be described. In the band structure of the solar cell of Example 1 described above, electrons excited by interband transition in the narrow-gap semiconductor (NGS) are excited again at the heterointerface (intraband transition). However, the solar cell of this example has a structure that utilizes only the excitation at the heterointerface (intraband transition). That is, the quantum dot region and double tunnel structure near the heterointerface remain the same, but no narrow-gap semiconductor is provided on the side of the double tunnel structure in the direction of sunlight incidence. In this case, it is possible to utilize long-wavelength light (infrared rays) that cannot be used in conventional solar cells. It can also be used as an infrared sensor.

[0038] FIG. 11 shows the band structure of the high-efficiency solar cell of this example. In the band structure shown in the figure, there is nothing to the right of the double tunnel structure on the right side of the heterointerface (the side in the direction of sunlight incidence). As in the band structure of the solar cell of Example 1 (FIG. 5(2)), there is no narrow gap semiconductor (NGS). FIG. 12 shows the configuration of the high-efficiency solar cell of this example. As shown in the figure, it has an nn structure, and n ＋- A highly doped n-type semiconductor layer is placed between the GaAs semiconductor substrate and the i-type layer. ＋ p ＋ n ＋ The double tunnel structure 1 of the junction is formed. Specifically, the n-type doped compound semiconductor layer 2 is made of n-Al. 0.3 Ga 0.7 As, the undoped compound semiconductor layer 4 is Al 0.3 Ga 0.7 A heterojunction between As and GaAs is formed, and a quantum dot region 7 of InAs quantum dots (QD) is formed at the heterointerface 6. In the undoped compound semiconductor layer 4, p ＋ -GaAs degenerate semiconductor layer (p ＋＋ -GaAs) in two layers ＋ -GaAs degenerate semiconductor layer (n ＋＋ -GaAs) is sandwiched between the double tunnel structure 1, and ＋ A GaAs semiconductor substrate 8 is provided, and an electrode 5 is provided below that. The double tunnel structure 1 suppresses the diffusion of carriers at the heterointerface, while ensuring the transport of carriers to the heterointerface by tunnel current.

[0039] Fig. 13 is a schematic diagram of a high-sensitivity infrared sensor according to another embodiment of the present invention. As shown in Fig. 13, the infrared sensor 110 of the present invention has a structure similar to that of the solar cell shown in Fig. 1, including an n-type doped compound semiconductor layer 2, a p-type doped compound semiconductor layer 3, an undoped compound semiconductor layer 4 sandwiched between them, and an electrode 5. The undoped compound semiconductor layer 4 is formed as a heterojunction of different semiconductors, and includes a quantum dot region 7 near the heterointerface 6 and a three-layer structure (double tunnel structure) 1 below it. A semiconductor substrate 8 is located below the p-type doped compound semiconductor layer 3, a backside electrode 52 is located below the semiconductor substrate 8, and an electrode 51 is located above the n-type doped compound semiconductor layer 2. The energy of infrared light incident from above is converted in the undoped compound semiconductor layer 4 to generate a photocurrent. In the quantum dot region 7 of the undoped compound semiconductor layer 4, electrons excited by infrared light escape from the quantum dot wells, and many electrons are accumulated at the heterointerface 6, and these electrons are extracted as a photocurrent. A bias voltage is applied and the photocurrent generated when infrared light is incident is extracted.

[0040] The specific configuration of the infrared sensor of Example 3 is the same as that of Example 1 shown in Fig. 2. That is, it has an nip structure, and the n-type doped compound semiconductor layer 2 is made of n-Al. 0.3 Ga 0.7 As, p-type doped compound semiconductor layer 3 ＋ -GaAs, and as the undoped compound semiconductor layer 4, Al 0.3 Ga 0.7 A heterojunction between As and GaAs is formed, and a quantum dot region 7 of InAs quantum dots (QDs) is formed at the heterointerface 6 (see FIG. 2). In the undoped compound semiconductor layer 4, a p-type doped compound semiconductor layer 3 side of the quantum dot region 7 is formed. ＋ -GaAs degenerate semiconductor layer (p ＋＋ -GaAs) in two layers ＋ -GaAs degenerate semiconductor layer (n ＋＋ In the infrared sensor of Example 3, a double tunnel structure 1 is formed in which a wide gap semiconductor is sandwiched between Al and GaAs, as in Example 1. 0.3 Ga 0.7The double tunnel structure 1 suppresses the diffusion of electrons at the hetero interface toward the narrow-gap semiconductor (GaAs), while ensuring the transport of electrons to the hetero interface by tunnel current. ＋ The quantum dot regions 7 are stacked in this order on a GaAs semiconductor substrate 8, and electrodes 5 are provided on the top and bottom surfaces. In Example 3, the quantum dot region 7 is a single layer, but multiple quantum dot regions 7 may be stacked. The doping concentration of the tunnel barrier layer in the double tunnel structure, the thickness of the barrier layer, and the manufacturing method are the same as those of the solar cell in Example 1.

[0041] Next, the irradiation intensity dependence of the photocurrent and photovoltage upon infrared light irradiation in the infrared sensor of Example 3 will be described with reference to FIG. 14. The graph in FIG. 14 shows the results of measuring the response to 1.3 μm infrared light irradiation, and shows the irradiation intensity dependence of the photocurrent (FIG. 14(1)) and photovoltage (FIG. 14(2)) upon infrared light irradiation. When irradiated with 1.3 μm infrared light, the photocurrent and photovoltage increased to 100 mA / cm irradiated with infrared light. ２ It can be seen that the amount of generated light increases sharply from the double tunnel structure. On the other hand, the device without the double tunnel structure does not exhibit the responsiveness to 1.3 μm as shown in Figure 14. This result shows that the insertion of the double tunnel structure produces responsiveness to infrared light, and that the device can be used as a highly sensitive infrared sensor.

[0042] (Other Examples) (1) In the above examples, the three-layer structure (double tunnel structure) in the undoped compound semiconductor layer (i-layer) is p ＋ Two layers of degenerate semiconductor layers ＋ The structure is a three-layer structure sandwiched between two degenerate semiconductor layers, but the polarity is reversed, i.e., n ＋ Two layers of p ＋ In this case, the three-layer structure supplies holes to the heterointerface by two band-to-band tunneling events for carrier holes, while the potential barrier prevents the upconverted holes from diffusing to the NGS side.

[0043] (2) In Examples 1 and 3, an n-i-p structure is used in which an undoped compound semiconductor layer (i-layer) is sandwiched between an n-type doped compound semiconductor layer (n-layer) and a p-type doped compound semiconductor layer (p-layer). However, the present invention is not limited to this structure and may have a p-i-n structure in which a p-layer, an i-layer, and an n-layer are arranged in this order from the direction of incidence of sunlight or infrared light. Alternatively, a p-i-p structure in which an i-type layer is sandwiched between p-type doped compound semiconductor layers (p-layers) or an n-i-n structure in which an i-type layer is sandwiched between n-type doped compound semiconductor layers (n-layers) may also be used.

[0044] (3) In Examples 2 and 3, n ＋ - A structure in which an undoped compound semiconductor layer (i-layer) and an n-type doped compound semiconductor layer (n-layer) are stacked on a GaAs semiconductor substrate (nin structure), but is not limited to this. ＋ A GaAs semiconductor substrate may be used, and a p-type doped compound semiconductor layer (p layer) may be stacked on the i layer instead of an n layer (pip structure). Alternatively, an nip structure or pin structure may be used.

[0045] (4) In the above examples, the quantum dot region at the heterointerface is a single layer, but multiple quantum dot regions 7 can be stacked to form a multilayer. (5) In the above examples, quantum dots of InAs were formed at the heterointerface, but quantum dots of other narrow bandgap semiconductors such as InSb, InAsSb, GaSb, Ge, or HgCdTe can also be formed. (6) In the above examples, an AlGaAs / GaAs heterojunction was used as the undoped compound semiconductor layer (i-layer), but other III-V group semiconductors, II-VI group semiconductors, or group IV semiconductors such as Si-based, Ge-based, or C-based can also be used.

[0046] The present invention is useful for optical devices such as solar cells and infrared sensors.

[0047] REFERENCE SIGNS LIST 1 Three-layer structure (double tunnel structure) 2 n-type doped compound semiconductor layer 3 p-type doped compound semiconductor layer 4 Undoped compound semiconductor layer 5, 51, 52 Electrode 6 Heterointerface 7 Quantum dot region 8 Semiconductor substrate 12, 16 Interband transition 19 Intraband transition 23, 24 Interband tunneling 30 Solar cell module 31 Cell 32 Outer frame 33 Light-transmitting layer 34 Sealant 35 Protective layer 36 Thin metal plate 100 Solar cell (cell) 110 Infrared sensor

Claims

1. In a device that extracts carriers generated by sunlight as a photocurrent, ＋ The degenerate semiconductor layer is divided into two layers, ＋ or a three-layer structure in which a degenerate semiconductor layer is sandwiched between n ＋ The degenerate semiconductor layer is divided into two p ＋ A device comprising a three-layer structure in which a first degenerate semiconductor layer is sandwiched between the first degenerate semiconductor layer and the second degenerate semiconductor layer.

2. The device described in claim 1, characterized in that the three-layer structure functions as a tunneling layer that causes two interband tunnelings of carriers present near the bottom of the conduction band or the top of the valence band, and a barrier layer that ensures one-way movement of carriers having energy greater than the energy of the bottom of the conduction band or the top of the valence band.

3. The device according to claim 1, characterized in that the device has an undoped compound semiconductor layer sandwiched between n-type or p-type or both doped compound semiconductor layers, the undoped compound semiconductor layer being a heterojunction of a wide gap semiconductor and a narrow gap semiconductor, and a quantum dot region and the three-layer structure are provided on the narrow gap semiconductor side near the heterointerface.

4. The device according to claim 1, characterized in that the device has an n-type or p-type doped compound semiconductor layer and an undoped compound semiconductor layer, the undoped compound semiconductor layer being a heterojunction between a wide gap semiconductor and a narrow gap semiconductor, and a quantum dot region and the three-layer structure are provided on the narrow gap semiconductor side in the vicinity of the heterointerface.

5. The device according to claim 3 or 4, characterized in that the three-layer structure is provided such that one end face of the three-layer structure is located at a distance from the heterointerface within a range of 3 to 50 nm.

6. A device as described in claim 3 or 4, characterized in that the quantum dot region forms an energy level of a conduction band lower than the Fermi level, and electrons accumulated at the heterointerface are excited from the intra-band level formed in the quantum dot region by sunlight and extracted as a photocurrent.

7. The device according to claim 3 or 4, characterized in that the quantum dot region is composed of a single layer or multiple layers and has a thickness of 50 nm or less.

8. The device according to claim 3 or 4, characterized in that the heterojunction is composed of a III-V compound semiconductor, and the quantum dot region is formed of quantum dots of any one of InAs, InSb, InGaAs, GaSb, Ge, and HgCdTe.

9. A solar cell using the device according to any one of claims 1 to 4.

10. A solar cell module using the solar cell of claim 9.

11. An infrared sensor for detecting infrared light, using the device according to any one of claims 1 to 4.

Images