JP7730766B2 - Light-absorbing material, solar cell, and method for producing light-absorbing material - Google Patents

Description

Article 30, paragraph 2 of the Patent Act applies. - Proceedings of the 68th Spring Meeting of the Japan Society of Applied Physics, published on February 26, 2021 (presentation number 17p-Z15-5) - Oral presentation at the 68th Spring Meeting of the Japan Society of Applied Physics, published on March 17, 2021 (presentation number 17p-Z15-5) - Isamu Jonoshita's personal website, published on January 10, 2021 (https://regionforsolarcell.wordpress.com/)

The present invention relates to light-absorbing materials, solar cells, and methods for producing light-absorbing materials. More specifically, it deals with light absorption by nanometer-sized regions with different electronic states from the surrounding area, which are created by ion implantation of dopants into a semiconductor material with a crystalline structure.

Figure 1 shows the energy band diagram of a commonly used Si single-crystal solar cell today, as well as a conceptual diagram of the power generation method. Light incident on the solar cell excites electrons in the valence band on the p-type side (left side in Figure 1) of the pn junction, which can be called the light absorption layer, to the conduction band. The excited electrons then pass through the n-type side (right side in Figure 1) and flow to the load, becoming the output current of the solar cell. After passing through the load, this output (electron) current returns to the valence band on the p-type side of the solar cell. The output of the solar cell is the current consumed by the load at this time multiplied by the band gap (the potential difference between the conduction band and valence bands).

Isamu Jonoshita, "Creation of Quantum Dot Structures Using Ion Implantation Technology," Solar Energy, Japan Solar Energy Society, May 2014, Vol. 40, No. 3, pp. 87-96 (especially p. 94) Takashi Kita (ed.), "Solar Cell Conversion Efficiency," Corona Publishing, October 2012, p. 40 Isamu Jonoshita, "Orbital Calculation of Channeling Ions," [online], September 19, 2016, Internet <URL: http://www7b.biglobe.ne.jp/~elbow/> (especially p. 10, Fig. 32) Toshiharu Saiki and Yasunori Toda, co-authors, "Nanoscale Optical Properties," Ohmsha, August 2004, pp. 8-10 Masaaki Nakayama, "Optical Properties of Semiconductors," Corona Publishing, August 2013, p. 51 "Dressed Photon" by Genichi Otsu, Asakura Publishing, March 2013, pp. 10-32

JP 2016-025178 A Japanese Patent Application Laid-Open No. 2020-202225

As described in Non-Patent Document 2 and the following Reference 1 (especially page 5, Appendix 7 sheet "conventional"), when electrons in the valence band on the p-type side (left side in Figure 1) are excited to the conduction band by incident light, the energy of the incident light that exceeds the band gap is immediately consumed as heat loss and cannot be extracted as solar cell output, resulting in heat loss. Furthermore, light that cannot excite electrons to an energy level above the conduction band cannot be absorbed by electrons in the valence band and is transmitted as a loss. Due to these losses, assuming the spectrum of a blackbody surface at 6000°C as the incident sunlight, the absorption efficiency of light that can be converted into electricity is essentially only 44.3%.
[Reference 1] Details of the research related to the present invention are published on the following webpage.
Isamu Jonoshita, "Electron energies in nm-scale impurity region and their application for photovoltaics," [online], January 10, 2021, Internet <URL: https://regionforsolarcell.wordpress.com>

The problem that the present invention aims to solve is to provide a light-absorbing material that has improved (enhanced) the above-mentioned light absorption efficiency value, a solar cell equipped with such a light-absorbing material, and a method for manufacturing such a light-absorbing material.

A means for solving the above problem is to create a nanometer-scale dopant-rich region (hereinafter sometimes referred to as the DR region) by injecting dopant ions into a light-absorbing material that is applied to the p-type side (left side in Figure 1) of the pn junction that constitutes the light-absorbing layer of a Si single crystal solar cell.
The following describes a light-absorbing material having a DR region and a method for manufacturing such a light-absorbing material.

First, an embodiment of the present invention will be described.
A light-absorbing material according to an embodiment of the present invention includes a semiconductor crystal and a dopant-rich region regularly arranged in the semiconductor crystal, the dopant-rich region containing a plurality of dopant ions and having a plurality of resonant absorption wavelengths in the visible range.

Here, it is preferable that the light-absorbing material be capable of absorbing light in a continuous energy range in the visible range.

The probability density function of light absorption of the light-absorbing material preferably has a maximum value at an energy level that is ±10% of the energy at which the solar light intensity is at its maximum.

The semiconductor crystal is a silicon crystal, the dopant is phosphorus, the dopant-rich region has a diameter of 5 nm to 7 nm, contains 4 to 10 dopant ions with an average of 7, and the distance between the dopant ions is 2 nm or more.

A solar cell according to an embodiment of the present invention may include the light-absorbing material according to an embodiment of the present invention as a light-absorbing layer.

In the method for producing a light-absorbing material according to an embodiment of the present invention, when producing the light-absorbing material according to an embodiment of the present invention, the diameter of the dopant-rich region, the number of dopant ions contained therein, and the distance between the dopant ions are determined using a molecular orbital method using a variational method so as to obtain the desired resonance absorption energy.

In this manufacturing method, a crystal film having a lattice constant different from that of the semiconductor crystal is formed on the surface of the semiconductor crystal, and the dopant ions are implanted into the semiconductor crystal through the crystal film to form the dopant-rich region.

Furthermore, the diameter of the dopant-rich region, the number of dopant ions contained, and the distance between the dopant ions can be controlled by adjusting the ion implantation dose, ion energy, and heat treatment conditions after ion implantation.

Here, as a detailed manufacturing method, it is preferable to apply the same method as that described in detail in Patent Document 2, as summarized below. Here, the semiconductor crystal is not limited to a Si single crystal, and the dopant is not limited to phosphorus ions.
In a method for producing a light-absorbing material, it is preferable to carry out the following steps: a first step of producing, on a surface of a semiconductor single crystal, a crystal film having a lattice constant different from that of the semiconductor single crystal, and forming a structure in which lattice matching points, at which channels of the semiconductor single crystal in which atomic nuclei do not exist on a straight line in the semiconductor single crystal and channels of the crystal film in which atomic nuclei do not exist on a straight line in the crystal film, overlap, are regularly arranged; and a second step of injecting ions that will become dopants in the semiconductor single crystal into the semiconductor single crystal via the crystal film, so that a structure in which DR regions are regularly distributed at a period corresponding to the period of the lattice matching points is produced by the ions injected into the surface of the crystal film traveling through the channels of the crystal film, passing through the lattice matching points, and traveling through the channels of the semiconductor single crystal to the interior of the semiconductor single crystal where they stop.

The ions preferably act as n-type dopants in the semiconductor single crystal. In this case, the ions are preferably phosphorus ions. The semiconductor single crystal is preferably made of a p-type semiconductor. The semiconductor single crystal is preferably a Si single crystal, and the crystal film is preferably a Ge crystal film.

In the second step, the depth at which the DR region is formed in the semiconductor single crystal and the size of the DR region can be controlled by adjusting the kinetic energy of the ions to be implanted and the temperature and time of annealing after ion implantation, and the doping density in the DR region can be controlled by adjusting the amount of ions implanted.

The light-absorbing material obtained in this manner has a semiconductor single crystal made of one of a p-type semiconductor, an i-type semiconductor, or an n-type semiconductor, and a DR region made of any other of these semiconductors. The DR region is a nanometer-scale region in which the semiconductor single crystal is doped with a dopant and is regularly arranged within the semiconductor single crystal.

The light-absorbing material obtained as described above can also be suitably used to construct solar cells. In such cases, the solar cell may have a light-absorbing layer made of a light-absorbing material in which the semiconductor single crystal is p-type and the DR region is n-type.

Here, a brief description will be given of the light absorption that can be achieved by a light-absorbing material having a DR region.
Figure 2 shows an example of the distribution of phosphorus ions in a Si crystal film for the light-absorbing material having the DR region described above. By performing appropriate annealing after ion implantation, these phosphorus ions are replaced with silicon atoms in the silicon crystal and activated. At this time, the energy of the electron orbitals in the DR region can be calculated using the Ritz variational method. According to the calculation results, under certain conditions, the distribution of the potential barriers in these DR regions exhibits a distribution range similar to the energy distribution of photons in visible light. This potential barrier allows the DR region to confine electrons within the region. Electrons confined in the DR region have a specifically high optical absorption for light with the same energy as the potential barrier. In other words, each DR region has its own unique potential barrier and specifically absorbs light with the same energy as its potential energy.

The light-absorbing material according to the embodiment of the present invention is a light-absorbing material having an improved (enhanced) light absorption efficiency value compared to conventional materials.
The optical absorption coefficient is an index that indicates the ease with which a substance absorbs light. Assuming a probability density function of 0.5, the value of the optical absorption coefficient of a light absorption layer containing a DR region is shown in Figure 3. When the light energy is approximately 3.3 eV or less, absorption by the DR region is dominant over absorption by the band gap. Furthermore, as shown in Figure 4, the optical absorption by the Si layer containing a DR region according to the present invention differs from the conventional absorption by the band gap; light is absorbed only when the light energy is the same as the potential barrier of the DR region. When light is absorbed under the above conditions, neither heat loss nor transmission loss occurs, resulting in essentially loss-free absorption.

1 is an energy band diagram of a silicon single crystal solar cell that is currently in common use, and a conceptual diagram of the power generation method. 1 is a diagram showing an example of the distribution of phosphorus atoms (or phosphorus ions) in a Si crystal film produced by a method according to an embodiment of the present invention. FIG. 10 is a diagram showing changes in the optical absorption coefficient for absorption by the DR region and absorption by the Si crystal band gap. FIG. 1 is a diagram illustrating the concept of light absorption by a Si layer including a DR region. 10A shows the change in the potential barrier with respect to the number of phosphorus ions under each condition, where (a) shows the case where the domain size is changed, and (b) shows the case where the distance between phosphorus ions is changed. FIG. 1 is a diagram showing the relationship between the energy state of electron orbitals in the DR region and the potential barrier. FIG. 10 is a diagram showing the relationship between the potential barrier and the probability density function in the DR region. FIG. 10 is a graph showing a relationship curve between light energy and light absorption coefficient α.

The light-absorbing material, solar cell, and method for manufacturing the light-absorbing material according to embodiments of the present invention are described in detail below.

The inventors have been researching optimal materials and manufacturing methods for improving the light absorption efficiency of solar cells. Patent Document 1 discloses a method for creating very small (nanometer-sized) regions richly doped with ions in a crystalline material. By implanting oxygen ions into crystalline silicon from four directions using a ZnTe crystal film as a filter, nanometer-sized semiconductor quantum dots (insulating regions of _SiO2 ) can be created regularly, densely, and uniformly.

Furthermore, Patent Document 2 shows that the amount of ion implantation can be reduced by changing the implanted ions in Patent Document 1 from oxygen ions to phosphorus ions (P ⁺ ), which are dopants, and by changing the filter film to a germanium crystal film. Note that the size of the ion implantation region is about 1.5 times larger than that in Patent Document 1, and the distance between the ion implantation regions is also longer by the same ratio.

In this invention, based on Patent Documents 1 and 2, we have conducted a detailed study of the behavior of electrons in ion-implanted regions in Si crystals, with the aim of proposing a method for more efficient light absorption, and have devised a method for improving the unavoidable losses in conventional photovoltaic power generation.

Figure 2 shows a conceptual diagram of the proposed DR region.
When the implanted ions are phosphorus ions (P ⁺ ), the P ⁺ loses its charge and becomes a phosphorus atom. After implantation, if the P+ is properly annealed, the phosphorus atom replaces a Si atom in the crystal structure and is incorporated into the crystal structure as a dopant, becoming a phosphorus ion. The free electrons in the Si crystal region around the phosphorus ion then trace orbits with specific energies. These specific energies can be calculated as an approximation using the molecular orbital method using the Ritz variational method.

FIG. 2 shows an example in which phosphorus ions are randomly arranged in a Si crystal under the following conditions, which are obtained as an appropriate set of parameters through trials:
Condition 1) The number of phosphorus ions in the DR region is 4 to 10 (average value is 7).
Condition 2) The size within the DR region (diameter when approximated to a circle) is set to approximately 5 nm, 6 nm, and 7 nm.
Condition 3) In the DR region, the distance between individual phosphorus ions is 2 nm or more.
These conditions can be achieved by properly controlling the ion implantation amount and ion energy, and by properly controlling the temperature and time of the post-ion implantation heat treatment (annealing).

<Calculation method>
As mentioned above, the "Ritz variational method" is used to approximately calculate the energies of the electronic levels in the DR region according to the following procedure.

(Step 1) Calculation of the effective Bohr radius and effective Rydberg energy In a silicon crystal, phosphorus atoms become phosphorus ions, which generate an electric field around them similar to that of a hydrogen nucleus in a vacuum. However, in the case of phosphorus ions in a silicon crystal, the equivalent of the vacuum is the silicon crystal structure, and in this case the effective Bohr radius _aB and effective Rydberg energy _ER are calculated using the following formulas.
In the above calculation formulas (1) and (2),
a ₀ : Bohr radius (0.0529 nm),
E ₀ : Rydberg energy (13.6 eV = 2.18 × 10 ⁻¹⁸ J),
m ₀ : electron mass,
m _e : effective electron mass (here, m _e /m ₀ =0.2 for Si),
ε _e : relative dielectric constant (here, 11.9 for Si).
As a result of calculation, a _B =3.15 nm and E _R =0.019 eV=3.07×10 ⁻²¹ eV.

(Step 2) Calculation of overlap integral, Coulomb integral, and resonance integral In the Ritz variational method, it is necessary to solve the following determinant (3).
Here, E _φ is an eigenvalue, and S _ij is called an "overlap integral" and is calculated by equation (4).
H _ij has two meanings, and when i=j, it is called the "Coulomb integral" and is calculated by equation (5).
where:
ε ₀ : dielectric constant of vacuum (8.854×10 ⁻¹² F/m),
R: distance between phosphorus ions,
R _a : indicates the distance between the relevant phosphorus ion and the average central position of other phosphorus ions in the DR region. In other words, multiple phosphorus ions other than the relevant phosphorus ion in the DR region are approximated by a single particle with a central charge, i.e., a point charge, and the distance from that point charge to the relevant phosphorus ion is defined as R _a . In this case, the charge n of the point charge is the charge in the space with a radius of R _a centered on the point charge, i.e., the number of phosphorus ions "actually present" in the space with a radius of R _a centered on the point charge.

When i≠j, H _ij is called the "resonance integral" and is calculated by equation (6).
where:
R _b : the center-to-center distance between the individual implanted phosphorus ions.

(Step 3) The eigenvalue E _φ is found.
The value of the determinant |H _ij -E _φ S _ij | is calculated for each energy value. This means that the value of the determinant |H _ij -E _φ S _ij | draws a curve as a function of "E _φ " in the x-y coordinate space. In this case, x is "E _φ " and y is the value of |H _ij -E _φ S _ij |. Furthermore, when considering a linear combination, this curve will have several intersections with the x-axis. The number of intersections indicates the number of electronic energy states in the DR region, which is equal to the number of phosphorus ions in the region. The energy of the electronic orbital in the region can then be calculated as x. As detailed in Reference 1, the minimum value of the electronic energy (maximum value in absolute terms) and the potential barrier in the DR region of each DR region in Figure 2 were calculated.

<Calculation results>
The above calculation results revealed the following:
Result 1) The more phosphorus ions there are in each DR region, the smaller the potential barrier becomes (the larger the absolute value becomes, assuming the energy of a free electron is 0).
Result 2) The smaller the size of each DR region, the smaller the potential barrier becomes (the absolute value becomes larger, assuming the energy of a free electron is 0).
Result 3) The potential barrier values of the individual DR regions range from −6 eV to −0.2 eV.
Result 4) Within each DR region, there are several electron energy states near 0 eV.

Figure 5(a) shows the relationship between the size of the DR region and the potential barrier when the region size is 4 nm, 5 nm, 6 nm, and 7 nm. Figure 5(b) shows this relationship when the distance between phosphorus ions is 2 nm or more, 2.5 nm or more, and 3 nm or more. Results 1) to 3) are clear from Figure 5. These graphs show the change in the potential barrier under each condition.

Result 4) will be explained in detail using Figure 6. Calculations using the Ritz variational method show that there are several electron energy states near 0 eV, and a resonance energy can be defined for each of these. This means that when photons that match the resonance energy come near the DR region, these photons are absorbed by electrons in the DR region, and there are multiple states in which the electrons in the DR region are pulled up to a state of around 0 eV. The existence of multiple such states means that the probability of the existence of a "photon that matches the resonance energy" increases. This is because electrons whose energy difference with that of a free electron (0 eV) is less than room temperature energy (25.8 meV, 300 K) can easily leave the DR region.

Next, we will consider the distribution range of the potential barrier for each DR region. Even if all of the above conditions 1) to 3) are the same, the potential barrier value will have a certain distribution range, but in addition to this distribution range, the distribution range of the potential barrier can also be expanded by changing conditions 1) to 3). DR regions with different sizes and charge densities are allowed to coexist, and the coexistence of these DR regions expands the distribution range of the potential barrier. If this distribution range is made wider than the energy distribution range of visible light, it will be possible to absorb the energy of all visible light. Therefore, the distribution range of the potential barrier is extremely important in this invention.

To illustrate this, the relationship between the potential barrier and the probability density function is shown in Figure 7. (For the data of this graph, see Appendix 4 of Reference 1.)
7 shows the potential barrier (horizontal axis) and probability density function (vertical axis) under the conditions of "an average number of phosphorus ions in a region of 7, a region size of approximately 5 nm, and a center-to-center distance between phosphorus ions of 2.5 nm or more." The dashed curve shows the condition of "an average number of phosphorus ions of 7, a region size of approximately 6 nm, and a center-to-center distance between phosphorus ions of 2.5 nm or more," and the dotted curve shows the condition of "7, approximately 7 nm, and 2 nm or more."

Figure 7 also shows the AM1.5 energy spectrum as part of the energy spectrum of sunlight on the surface of Japan. For each of the three probability density function curves, it can be seen that the distribution range of the potential barrier in the DR region roughly covers the AM1.5 energy range.

(For details of the formulas (7) to (11), see Non-Patent Documents 4 and 5.)
Furthermore, we will consider the optical absorption coefficient of the light-absorbing material. The optical absorption coefficient is a measure of how easily a material absorbs photons, and the value of this optical absorption coefficient is extremely important. Even if the resonance energy of the potential barrier in the DR region spans the entire range of visible light, if the optical absorption coefficient due to the band gap is dominant over the optical absorption coefficient due to the DR region, the optical absorption due to the band gap will be large, resulting in loss.

The value of the light absorption coefficient α of this light absorption system can be calculated by equation (7).
Here, ω is the frequency of light, C is the speed of light, and n ₂ is the value of the imaginary part of the refractive index of light, and n ₂ is calculated by the following equation (8).
Here, ε ₁ is the real part of the dielectric function, and ε ₂ is the imaginary part, which can be calculated using the following equations (9) and (10).
where N ₀ is the number of DR regions in a unit volume, and γ is the reciprocal of the relaxation time of the excitation of the DR region, where γ represents the full width at half maximum (FWHM) of ε ₂ at the resonance frequency ω _i .

Under this condition, if the DR region is considered as an electric dipole, the resonant frequency can be calculated by the following equation (11).

Table 1 shows an example of calculations using equations (7) to (11). Figure 8 (Fig. 5 in Reference 1) shows the relationship between light energy and the optical absorption coefficient α. These relationships indicate what type of optical absorption occurs. The curve in Figure 8 reveals that the value of α becomes particularly large when the light energy reaches the resonance condition (ω _i = ω), and is nearly zero otherwise. This means that only light of a specific energy is absorbed in a specific region that resonates with that light with a particularly large optical absorption coefficient α. In other words, only loss-free optical absorption in the DR region is possible. Because the potential barrier value in the DR region is approximately -6 eV to -0.2 eV, light in this region can be absorbed with high efficiency and without loss.

Next, we will consider the resonance condition (ω _i = ω). The full width at half maximum (FWHM) is usually used to express the resonance range. From equations (7) and (8), we can see that the important value that determines the value of α is ε ₂ , and the full width at half maximum of ε ₂ is expressed by γ. If the relaxation time is set to 1 ns (= 1 × 10 ^-9 [s]), the full width at half maximum can be converted into energy units using equation (12).
As a result, it is found that the half-width of ε ₂ is only 1.05×10 ⁻²⁵ [J] (=6.6×10 ⁻⁷ [eV]).

On the other hand, as shown in FIG. 7, if the probability density function when the potential barrier is near the resonance energy is approximately 50%, the probability that the potential barrier in a certain region is within the half-width of the resonance region of a certain light is only 0.000033%. Because of this very low probability, the value of _N0 in equation (10) becomes small. As a result, the value of _ε2 also becomes small, and the value of α also becomes small. With this small value of α, as shown in Table 1, the thickness of the light absorption layer required to absorb 90% of the number of photons is on the order of 100 μm. This is not thin compared to absorption by a normal band gap, and it cannot be said that light absorption by the DR region is advantageous.

However, there are the following factors that can improve (thinneren) this required thickness, in other words, increase the effective value of α:

1) There are multiple electron energy states near 0 eV. Some of the electron energy states in the DR region are near 0 eV. This means that there are several resonant states, each with a slightly different resonant energy from the other states. Taking this into account, the value of _N in equation (10) can be replaced with N _eff calculated by equation (13).
where f denotes the probability that the potential barrier of a certain DR region (given in Table 1) is within the half-width, and N is the number of energy states around 0 eV.

2) Influence of adjacent DR regions (See Non-Patent Document 6 for Equations (14) to (19) below)
Next, we consider the influence of adjacent DR regions. If the distance between the DR regions is sufficiently short, energy transfer occurs via so-called "near-field light" (dressed photons). The exchange energy is calculated using the following equation (14). As shown in Fig. 6 and Appendix 5 of Reference 1, this value can be as large as 1 meV.
m _pol : Effective mass of polariton (tentatively set to 0.2 _{m 0} )
m _α : Effective mass of DR region α (tentatively set to 0.5 m ₀ )
r: distance between the centers of the DR regions

As can be seen from the following equation (17) (see Non-Patent Document 6), which is the basis of equation (14), most of the energy transfer is concentrated when the resonance energy of the DR region is near the energy of the light.
where:
E(k): Energy of polariton Eα: Resonance energy of DR region α k: Wave number of light Only the integral term of equation (17) is extracted and expressed as the following equation (19).
From equation (19), we can see that the value of v is maximum when the energy of the incident light is slightly smaller than the resonance energy ( _Es or _Ep ) of the DR region, and is minimum (negative value, maximum as an absolute value) when it is slightly larger.

When the energy of the incident light is slightly greater than the resonance energy, a negative value indicates the emission of energy from the DR region. In other words, if the energy of the incident light is slightly greater than the resonance energy of the DR region, the energy required to excite electrons in the DR region decreases by emitting energy to the outside. However, the value of this emitted energy becomes zero when the energy required for electron excitation equals the resonance energy. In other words, the energy required for electron excitation is automatically adjusted to the resonance energy. (See graph 2 in Appendix 5 of Reference 1.)

As mentioned above, the resonance region is very narrow, and when the relaxation time is 1 ns, the resonance region is only about 6.5×10 ⁻⁷ [eV]. However, if energy transfer is a factor in adjusting the resonance conditions, the resonance region becomes about 1 meV (1.0×10 ⁻³ [eV]). This directly affects the thickness required for light absorption in solar cells, and the thickness can be reduced to 1/1000.
The calculations of the effects of items 1) and 2) above and the results are shown in the right column of Table 1 and in Appendix 6 of Reference 1.

[Conclusion]
A light-absorbing material having a DR region containing multiple dopants such as phosphorus ions in a semiconductor crystal such as a silicon crystal has multiple resonant absorption wavelengths in the visible range, and can also absorb light in a continuous energy range in the visible range.

Furthermore, we investigated the optimal configuration for a photovoltaic power generation method using the DR region using the method described herein. From this investigation, we reached the following conclusions. As shown in Figure 7, when comparing the AM1.5 light spectrum with the solid curve (the curve for conditions of "average number of phosphorus ions: 7, DR region size: approximately 5 nm, center-to-center distance between phosphorus ions: 2.5 nm or more"), it is easy to see that the peaks are located at approximately the same position on the x-axis. Light with the frequency that provides the maximum energy corresponding to this peak can be absorbed most efficiently. This also indicates that light energy can be efficiently absorbed by using a p-type Si crystal layer containing a DR region under these conditions as a light absorption layer. The specific parameters of the DR region (e.g., size of the DR region, number of phosphorus ions contained, distance between phosphorus ions) can be set so that the maximum value of the light absorption probability density function is located within an energy range of ±10% of the energy at which the solar light intensity is at its maximum. If even higher efficiency is desired, a multilayer structure can be considered, in which a light absorber having a DR region of a light absorption layer designed under different conditions is stacked, such as those shown by the dashed or dotted lines in Figure 7. Whether a single-layer or multilayer structure is adopted, and the structural conditions of each layer (average number of phosphorus ions, size of the DR region, center-to-center distance between phosphorus ions) can be determined depending on the circumstances, such as the desired light absorption efficiency.

To actually manufacture a light-absorbing material with a designed DR region, a method similar to that disclosed in Patent Document 2 can be used. That is, a crystal film with a different lattice constant than the semiconductor crystal is created on the surface of a semiconductor crystal such as Si, and dopant ions such as phosphorus are implanted into the semiconductor crystal through this crystal film to form the DR region. In this process, parameters such as the size of the dopant-rich region, the number of dopant ions contained, and the distance between dopant ions can be controlled by adjusting the ion implantation amount and ion energy when implanting the dopant ions, as well as the conditions for heat treatment after ion implantation.

The current limit to the conversion efficiency of solar cells using the band gap of Si crystal is said to be 29%. Photon energy is converted into electron energy by the band gap of the p-n junction of the Si crystal, but if the photon energy is greater than the band gap, that amount of energy is lost as heat energy. Also, if the photon energy is smaller than the band gap, the photon passes through without being absorbed. However, as can be seen by comparing Figures 1 and 4, the mechanism by which photon energy is absorbed by the DR region is completely different. Most photons are absorbed by electrons in the DR region, which has a potential barrier that matches their energy. Therefore, because the energy of the light and the energy given to the electrons are the same, theoretically neither the heat loss nor the transmission loss shown in Figure 1 occurs.

Finally, the basic difference between a solar cell having a light absorbing layer made of a light absorbing material including a DR region according to an embodiment of the present invention and a so-called quantum dot solar cell will be described.
Both quantum dot solar cells and solar cells using light-absorbing materials with DR regions excite the electrons inside them (in the process of extracting them from the quantum dots or DR regions) and absorb the energy of light, but the methods for extracting the electrons and transporting them outside the solar cell differ.

In quantum dot solar cells, electrons are extracted to the outside by the seepage of the electron wave function due to the tunneling effect, which makes this wave function continuous between quantum dots, resulting in electrons being extracted from the light-absorbing layer containing the quantum dots. To match the wave functions between each quantum dot, the quantum dots must be of uniform size. Furthermore, because energy is absorbed according to the wave function, heat loss and transmission loss occur, similar to absorption due to band gaps. However, because the energy of the wave function can be changed by the size of the quantum dots, stacking light-absorbing layers of matched size for each layer makes it possible to create solar cells with low heat loss and transmission loss, similar to compound solar cells with multiple band gaps.

On the other hand, electron extraction by the DR region occurs when the potential barrier and the absorbed energy match by absorbing light with an energy matching the potential barrier. By appropriately configuring the DR region conditions (average number of phosphorus ions, region size, and center-to-center distance between phosphorus ions), multiple energy levels can be established near 0 eV. Furthermore, energy transfer between adjacent regions broadens the range of absorbable light energy. As a result, visible light with a wide range of energies can be resonantly absorbed. In theory, no heat loss or transmission loss occurs during this resonant absorption process. Furthermore, the transport of extracted electrons occurs through the diffusion of extracted electrons and the electric field created by the pn junction, just like in conventional crystalline silicon solar cells, and is essentially unaffected by the size of the DR region, the number of dopants (phosphorus ions) within the region, or the distance between dopants.

Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present invention.

Claims (7)

a semiconductor crystal that is a silicon crystal ;
and dopant -rich regions regularly arranged in the semiconductor crystal,
The dopant-rich region is
a plurality of dopant ions, the dopant ions being phosphorus ions ;
The diameter is 5 nm or more and 7 nm or less,
the number of the dopant ions contained is 4 or more and 10 or less,
the distance between the dopant ions is 2 nm or more;
The light-absorbing material has a plurality of resonant absorption wavelengths in the visible range and is capable of absorbing light in a continuous energy range in the visible range with only a single layer. 2. The light-absorbing material according to claim 1 , wherein a probability density function of light absorption of the light-absorbing material has a maximum value at an energy of ±10% of the energy at which the intensity of sunlight is maximum. 3. The light-absorbing material according to claim 1 , wherein the dopant-rich region contains an average of seven dopant ions. A solar cell comprising the light-absorbing material according to claim 1 in a single layer as a light-absorbing layer. A method for producing the light-absorbing material according to any one of claims 1 to 3 , wherein the diameter of the dopant-rich region, the number of dopant ions contained therein, and the distance between the dopant ions are determined by a molecular orbital method using a variational method so as to obtain a desired resonance absorption energy. forming a crystal film having a lattice constant different from that of the semiconductor crystal on the surface of the semiconductor crystal;
The method for producing a light-absorbing material according to claim 5 , wherein the dopant-rich region is formed by implanting the dopant ions into the semiconductor crystal through the crystal film. 7. The method for producing a light absorbing material according to claim 6, wherein the diameter of the dopant-rich region, the number of dopant ions contained therein, and the distance between the dopant ions are controlled by adjusting the ion implantation amount and ion energy when implanting the dopant ions, and the conditions of heat treatment after ion implantation.