JP5574181B2 - Inorganic photovoltaic cell - Google Patents

Description

The present invention relates to a solid photovoltaic cell comprising three inorganic solid materials, each comprising:
A transparent n-type semiconductor compound porous layer (2),
A layer of transparent P-type semiconductor compound (3) at least partially filled into the plurality of pores of the porous layer, and an aggregate between the two semiconductor compounds of the n-type and p-type layers ( 4) an inorganic absorber compound which forms

  Such photovoltaic devices are well known and are described in particular in the following patent documents (1-3), which are incorporated herein by reference.

  The principle of operation of this device is as follows. The radiation (photons) is absorbed only by the absorber material; electrons are then transferred to the transparent (or non-absorbing) n-type semiconductor, during which positive charges are transferred to the p-type semiconductor. According to this principle, the movement of charge carriers of different signs takes place in two different phases. This theoretically reduces a significant amount of performance loss due to charge recombination. In order to provide a sufficient internal area to include a sufficient amount of absorber in the film relative to the projected area, a porous nanocrystalline or microcrystalline film must be used.

There is a report that a photovoltaic conversion efficiency of 2.3% was obtained by a ZnO / CdSe / CuSCN cell in the 0.36 sun (see Non-Patent Document 1). There is also a report that a photovoltaic conversion efficiency of 3.4% was obtained in the area of 0.03 cm ² of a ZnO / In ₂ S ₃ / CuSCN cell in one sun (see Non-Patent Document 2). .

Photovoltaic conversion efficiency of 1.3% is obtained in a 0.54 cm ² TiO ₂ / CdS / CuSCN cell with a highly internal nanocrystalline TiO ₂ and a CdS coating of ~ 5 nm-10 nm thick in one sun (See Non-Patent Document 3). Subsequently, a French patent (Patent Document 4) discloses that a TiO ₂ / Sb ₂ S ₃ / CuSCN cell with a band gap of ˜1.65 eV (˜750 nm) can provide an efficiency of 3.4%. ing. The best external quantum efficiency (EQE) is up to 80%. This value is the maximum value obtained in consideration of at least ˜15% loss due to absorption and reflection by the TCO glass substrate. In recent years, a similar Sb ₂ S ₃ nanoporous solar cell has obtained a conversion efficiency of 3.37% in a TiO ₂ / In—OH—S / Sb ₂ S ₃ / (KSCN) CuSCN structure with an area of 0.15 cm ^2. There was a report. In this report, the authors have also confirmed that the thiocyanate pretreatment prior to CuSCN impregnation previously reported in Non-Patent Document 4 has a beneficial effect.

French Patent No. 2 881 880 (French Patent No. 05 01 114) French Patent No. 2 881 881 (French Patent No. 05 01 113) French Patent No. 2 899 381 (French Patent No. 06 02791) French Patent No. 2899381

Levy-Clement, C .; Tena-Zaera, R .; Ryan, M.A .; Katty, A .; Hodes, G. Adv. Mater. 2005, 17, 1512 Belaidi, A .; Dittrich, T .; Kieven, D .; Tornow, J .; Schwarzburg, K .; Lux-Steiner, M.Ch. Phys. Status Solidi, Rapid Res. Lett. 2008, 2, 172 Larramona, G .; Chone, C .; Jacob, A .; Sakakura, D .; Delatouche, B .; Pere, D .; Cieren, X .; Nagino, M .; Bayon, R. Chem. Mater. 2006, 18, 1688) Larramona, G et al. Chem. Mater. 2006, 18, 1688

  The stability and reproducibility of the photovoltaic performance of these cells is not sufficient for practical use. When the cell is sealed to avoid oxidative effects due to contact with ambient air, the cell performance is drastically reduced.

  The problems of the stability and reproducibility of the cell's photovoltaic performance described above can be caused by changing the gas atmosphere in the box in which the cell is sealed, or by the initial, unsealed cell light. It also remains when adding reactants such as alkali metal thiocyanate MSCN which has a beneficial effect on the electromotive force performance. The reason and cause of such performance degradation are not understood from the teachings of the prior art.

The object of the present invention is to provide a new photovoltaic device exhibiting improved performance, in particular the stability and reproducibility of the photovoltaic performance of the cell after it has been sealed for practical use. Is to provide a further improved photovoltaic device. The object of the present invention is to achieve satisfactory photovoltaic performance, especially improved photovoltaic performance, short circuit photocurrent value with an irradiance of 1000 W using inexpensive, stable and highly toxic and environmentally friendly materials. Photovoltaic performance greater than 2 mA / cm ² under / m ² and even greater than 5 mA / cm ² and high photoelectric conversion efficiency after cell sealing, especially from 1% To obtain a conversion efficiency which is large and even greater than 2%.

  As described below, several experiments were conducted to clarify the cause of the above problem. The lack of reproducibility of photovoltaic performance appears to involve uncontrollable factors such as the concentration of oxygen and water in the cell's atmosphere and the initial adsorbed oxygen and water content inside the cell. In order to obtain high photovoltaic performance, it has been found that the cell needs to come into contact with oxygen in an intermediate step of the manufacturing stage in order to create limited oxidation of various chemical species in the structure of the cell.

More specifically, it has been found that the photovoltaic performance is gradually improved when visible light illumination is started in the presence of a gas containing oxygen (hereinafter, this process is called optical soaking process). . This indicated that some sort of photocatalytic oxidation reaction occurred inside the cell containing oxygen, water (in the case of ambient air atmosphere) and other residues. This reaction is a reaction with a cell composition such as CuSCN or Sb ₂ S ₃ to improve the charge transfer mechanism at the interface. Once the optical soaking is performed for a period of time, the photovoltaic performance is enhanced while the series resistance remains small even after the atmosphere becomes dry nitrogen gas. However, it has been found that such improvements require partial and limited oxidation of the absorber. In fact, if this optical soaking process continues for a long time, after the initial improvement, a decrease in cell photovoltaic performance is induced. Long illumination in the presence of oxygen can lead to cell degradation due to additional photocatalytic reactions, such as the oxidation of the absorber as long as the absorber is gradually photooxidized in the presence of oxygen. It is.

To more accurately solve the above problems, the present invention relates to a solid state photovoltaic cell comprising three inorganic and solid materials, each comprising:
A porous layer of a transparent n-type semiconductor compound,
A layer of a transparent P-type semiconductor compound that fills at least part of the pores of the porous layer,
An inorganic absorber compound that forms an aggregate between the two semiconductor compounds of the n-type and p-type layers, at least partly not in contact;
Further, an oxidation product different from the semiconductor compound and the absorber are included, and the oxidation product includes a chemical species resulting from partial oxidation of the absorber compound and / or the p-type semiconductor compound, It is disposed between the two during and / or the absorber aggregation and p-type semiconductor compound layer of two n-type and semiconductor compounds of p-type respectively, to form a discontinuous portion of the interfacial layer, the P Type semiconductor compound is CuSCN, the absorber is a thermally annealed antimony sulfide antimony compound, the cell further contains MSCN of thiocyanate compound, in which M represents an alkali metal, the interfacial layer is characterized by (SCN) ₂ and / or polymer (SCN) _n der Rukoto.

  The discontinuous portion of the interface layer reflects the limited oxidation of the absorber, and thus only the portion. The presence of such oxidizing species can be detected by infrared spectroscopy.

In one preferred embodiment of the invention, the P-type semiconductor compound is CuSCN and the absorber is a thermally annealed antimony sulfide-based antimony compound such as Sb ₂ S ₃ .

M is preferably lithium, sodium or potassium .

The interface layer is formed at the interface between Sb ₂ S ₃ and CuSCN. This consists of oxidation products such as thionocyanate SCN ₂ or (SCN) _n resulting from the oxidation of SCN ^- derivatives present in the cell, including the P-type semiconductor CuSCN and / or doped MSCN. If no alkali metal thiocyanate MSCN is added to the cell, the interface layer is also formed from the photocatalytic oxidation of CuSCN. However, the addition of MSCN is preferable for improving the photovoltaic performance.

  Improved performance was observed even when MSCN was not used. However, the presence of MSCN promoted the improvement of photovoltaic performance by the light soaking process.

  More specifically, the weight ratio of MSCN to CuSCN is 0.001 to 2. The optimal ratio will vary depending on the type of alkali metal in MSCN and the type of absorber.

The improvement in photovoltaic performance can be explained as follows. Naturally, the contact between Sb ₂ S ₃ and CuSCN for efficient charge transfer does not work because of the gap between them. Thus, the light guide holes inside Sb ₂ S ₃ are not moved to CuSCN, resulting in insufficient cell performance. However, in the gap between Sb ₂ S ₃ and CuSCN, M ⁺ (Li ⁺ , Na ⁺ , or K ⁺ ) and SCN ⁻ are melted with a small amount of water due to the hygroscopic nature of MSCN. Sb ₂ S ₃ is known as a photocatalyst. When this is irradiated with light, holes are formed and electrons are formed. This electron is transferred to TiO ₂ or reduces the valence of the chemical species on the surface of Sb ₂ S ₃ by giving the electron. If oxygen and H ⁺ are present, H ₂ O or H ₂ O ₂ is formed by these electrons. The pores created by light in Sb ₂ S ₃ oxidize the surface species. Sb ₂ S ₃ itself is also likely to be oxidized by the pores to form S and Sb ³⁺ . SCN ^- is oxidized to (SCN) ₂ or (SCN) _x polymer, known as a conductive polymer. Thereby, the gap at the interface between the CuSCN and Sb ₂ S ₃ is filled, and the charge transfer of the solar cell can be facilitated. Even if not if added MSCN is, dipropyl sulfide remaining, this is used as a solvent for depositing a CuSCN layer, a small amount of SCN from ^- is obtained. What is oxidized is determined by the mechanism of each oxidation. The generated electrons reduce oxygen while the pores formed in Sb ₂ S ₃ oxidize MSCN to form thiocyan (SCN) ₂ or polymerized thiocyan (SCN) _x . Thus, in the absence of oxygen, the electrons recombine with the holes, resulting in non-oxidation of MSCN.

Cell Le is sealed into the wall, the wall of that is at least partially transparent, the cell may comprise a Drying gas atmosphere. Immediately after the cell is sealed, oxygen and moisture can be prevented from entering the cell, and further deterioration of the cell due to oxidation can be prevented or minimized.

More specifically, the transparent n-type semiconductor material is a metal oxide such as TiO ₂ , ZnO or SnO ₂ , preferably TiO ₂ .

  An optical soaking process in a flat load cell was tested. However, no improved high photovoltaic performance was observed compared to the interpenetrating cell.

  The cell may exhibit a through structure. A porous substrate of an n-type semiconductor compound has a pore diameter of 10 mm to 100 mm, and the inner surface of the pore is at least partially coated with an aggregate of the composition of the absorber, and the pore is at least 10% by volume, preferably 15 And filled with a cover layer made of the other p-type solid semiconductor compound, and the interface layer portion of the oxidation product has a thickness of 0.1 nm to 5 nm, preferably from 0.1 nm. 1 nm.

According to another particularly advantageous feature of the cell of the invention,
The roughness coefficient of the porous substrate before depositing the absorber is greater than 50, preferably greater than 100;
The porous substrate is made of particles or crystals, the average size of the particles or crystals is between 30 nm and 50 nm, and the average pore size is between 20 nm and 50 nm,
The n-type semiconductor compound is present as a metal oxide porous substrate, and the cover layer is made of the p-type solid semiconductor compound;
The n-type semiconductor metal oxide of the porous substrate is TiO ₂ with an average particle size between 30 nm and 50 nm and an average pore size between 20 nm and 50 nm; and
The cover layer also forms an upper layer of at least 10 nm thickness and is disposed on one surface of the porous substrate.

Other inorganic absorbers, such as CdS, have been tested and shown that the method of the invention works well and can be applied effectively to other absorbers combined with TiO ₂ and CuSCN.

The expression “antimony sulfide group” here means that said compound is at least mostly made preferably by antimony sulfide Sb ₂ S ₃ . Such antimony sulfide compounds may contain in some percent more than one of Sb or S elements, and / or oxide ions, and / or hydroxide ions as a partial substitution of sulfide ions.

  The expression “transparent” here means that the material does not absorb much, ie an absorption ratio of less than ˜35% for radiation at wavelengths between 400 nm and 1200 nm, preferably between 450 nm and 1200 nm. Means an absorption ratio of less than 10% for radiation at a wavelength of.

  The expression “absorber” here means that the material absorbs at an absorption rate of at least 50% for radiation at wavelengths between 400 nm and an absorption starting point located between 600 nm and 1200 nm. To do.

Preferably, the thermal annealing treatment is performed after the absorber material is deposited on the n- or p-type semiconductor film at a heat treatment temperature of about 300 ^° C., preferably in a nitrogen gas atmosphere.

  The photovoltaic device of the present invention is based on three solid compounds (two transparent semiconductors n and p-type and one absorber) and has the advantage of exhibiting satisfactory photovoltaic performance at low manufacturing costs And even superior performance for embodiments of the present invention.

  The expression “hole diameter” here means the opening diameter of the hole, in other words, the average dimension in the longitudinal section, not the longitudinal direction or depth direction through the thickness of the film.

  The expression “roughness factor” here means the ratio between the inner surface of the porous film (this is the same as the inner surface of the holes present in the film) and the projected area of the film on the substrate. This porosity is due to the fact that the film is made of particles or crystals. These agglomerates create pores in the space of the rift that remains between such particles and crystals during film production. And / or it may be due to the creation of pores directly into a single material that does not contain particle boundaries. Such roughness coefficient is estimated by the approximate value method (pore diameter is 2 nm) by measuring the porosity (and pore size distribution) using nitrogen adsorption / desorption technology that can be understood by calculating the BET active surface. For larger and nearly spherical shapes).

  The porous substrate is formed of particles or crystals. This means that the inner surface of the hole constitutes the surface of the particle or crystal.

Preferably, the n-type semiconductor metal oxide is made of TiO ₂ having an average particle size between 30 nm and 50 nm and an average pore size between 20 nm and 50 nm. In the case of TiO ₂ , the particle size corresponds to the BET surface of the porous substrate before the absorber is deposited and is greater than 25 m ² / g, more preferably between 25 m ² / g and 50 m ² / g. is there.

  Advantageously, the cell of the present invention comprises a first compact layer of non-transparent semiconductor compound (non-contained) interposed between the porous semiconductor film and a transparent conductive substrate called “pre-contact”. Porous). Such a thin layer (called a barrier layer) is made of the same type (n or p) material from the same member or porous film material, and a transparent conductive material called “pre-contact” with the semiconductor filling the hole. This makes it possible to avoid a short circuit with the conductive substrate. The transparent conductive substrate or “pre-contact” may be a commercially available transparent conductive glass.

  As a further advantageous aspect of the cell of the present invention, the cell is formed of a semiconductor compound disposed between the semiconductor porous substrate and a layer of the conductive substrate referred to as “post contact” of at least 10 nm. Includes a layer of thickness. This thin upper layer is made of the same semiconductor material, or another semiconductor material of the same type as the one that fills the holes, and a layer of conductive substrate called “post contact” with the semiconductor porous substrate. Allow avoidance of short circuit between. This conductive substrate or “post-contact” may be made of another conductive material such as carbon or metal.

More specifically, the cells of the present invention are characterized as follows:
-The first surface of the porous substrate is a transparent barrier layer of a molded body (non-porous), preferably made of an n-type semiconductor metal oxide, and is conductive and transparent like conductive glass. Deposited on a "pre-contact" first substrate, and the porous substrate constitutes a film with a thickness greater than 1 μm, preferably between 2 μm and 10 μm, and
The absorber composition forms aggregates of 10 nm-200 nm in size, and the cover layer of P-type semiconductor compound fills at least 10% of the pore volume, preferably 15% or more, And the second surface of the porous substrate is covered by a second conductive substrate “post-contact”, and preferably, the cover layer is also “post-contact” with the porous substrate Constituting an upper layer of at least 10 nm thickness placed between the conductive layer of the second conductive substrate “post-contact” and
The porous substrate, its absorber and cover layer are housed between the two conductive substrates that act as pre-contact and post-contact. The second conductive substrate in post contact may or may not be transparent, and is preferably made of metal or carbon.

Deposition of nanocrystalline porous films based on semiconductor metal oxides such as TiO ₂ is a doctor blade (or tape casting) colloidal dispersion agent for the metal oxide such as TiO ₂ or for example the TiO ₂ It is performed by a low-cost manufacturing technique such as screen printing of a paste based on a metal oxide.

  The absorber layer is deposited by low cost manufacturing techniques such as chemical bath deposition or electrochemical deposition.

  Covering the free volume of the holes with a second “transparent” semiconductor is done with low cost manufacturing techniques. For example, impregnation with a solution of a dissolved material and subsequent solvent evaporation, impregnation with a material precursor solution using a spin coating method, or electrochemical deposition is performed.

  The cells of the present invention are made according to various controlled oxidation processes, such as heat treatment.

However, in certain preferred embodiments, the present invention further provides a process for making a cell according to the present invention. In the process, the oxidation product of the interface layer described above is obtained by a so-called optical soaking process including visible light irradiation in a gaseous atmosphere containing oxygen in an apparatus including the following composition members of the cell.
The porous layer of a transparent n-type semiconductor compound,
The transparent layer of p-type semiconductor compound filling at least a portion of the pores of the porous layer, and at least a portion of the two layers of p-type and n-type semiconductor compounds, respectively, that are not in contact with each other An inorganic absorber compound presenting an aggregate located in the region.

Many atmospheres were tested, including dry O ₂ , dry synthetic air without moisture, ambient air (room air), pure N ₂ gas, and humidified N ₂ . It was concluded that visible light radiation in a gas containing oxygen, preferably ambient air, improved photovoltaic performance. Certainly, good results can be obtained if the ambient air appropriately contains moisture. However, too much moisture reduces the photovoltaic performance of the cell. Even in the N ₂ atmosphere, the performance progress rate was low, and a low light soaking effect was observed. In this case, however, it is considered that the oxygen adsorbed in the cell in the initial stage reacts under visible light illumination. In accordance with this discovery, it was observed that the light soaking effect under N ₂ was reduced if the cell was heat treated to remove the adsorbed oxygen and water before light irradiation.

  “Visible light” means a wavelength of 400 nm to 1200 nm.

Preferably, during the light soaking process, the gas atmosphere containing oxygen is ambient air, preferably air containing a moisture percentage lower than the supersaturated moisture percentage of said ambient air at ambient temperature. In fact, it shows that water is involved in charge transfer, and the photovoltaic performance of the cell is linked to the humidity level of the ambient air. Water is most likely generated by the reduction of O _{2 in} the presence of H ⁺ during light soaking. However, according to the present invention, it has been observed that very little water, such as water normally present in ambient air below supersaturation, is sufficiently useful and lowers photovoltaic performance at high water retention.

According to another particular advantage of the process of the invention, the optical soaking process comprises irradiating the cell with a visible light power of 0.1 kW / m ² to 5 kW / m ² for 5 minutes to 20 hours. It is out.

  More advantageously, the optical soaking process described above includes illuminating the cell in an open circuit condition.

  The open circuit state means that no load is connected to the solar cell, the plus electrode and the minus electrode are open, and no current flows. Since electrons cannot get out of the cell, the highest optical soaking effect and photovoltaic performance is observed. This is because the cell is in an open circuit state. The opposite state is a short circuit state. In this case, plus and minus are directly connected without load, and electrons go out of the cell.

  According to yet another specific advantageous feature of the process of the invention, parameters such as short circuit photocurrent Jsc and open circuit voltage Voc are monitored and optical soaking dependent on the series resistance and photovoltaic conversion efficiency. The process includes illuminating the cell as follows.

Illuminate as long as the photovoltaic conversion efficiency is increased, and stop the illumination when the conversion efficiency is highest, and / or
Illuminate as long as the series resistance of the cell is reduced, and stop the illumination when the resistance is at its lowest value.

  The photovoltaic conversion efficiency is related to the short circuit current Jsc, the open circuit voltage Voc, and the filling factor ff. Typically, all these parameters Jsc, Voc, ff improve as an interfacial layer of oxidizing species occurs when light soaking is performed. The relationship between the series resistance and the above parameters will be described in detail below.

  In the first embodiment of the present invention, the optical soaking process is performed in an unsealed cell, and the cell is sealed immediately after the optical soaking process is performed. The optical soaking treatment is preferably performed under a dry gas atmosphere, more preferably under dry nitrogen gas or dry synthetic air.

  The expression “sealed cell” means that the cell is confined in a wall.

More specifically, the device contains an MSCN compound and is fabricated through the following sequential steps.
・ Vapor is deposited on the porous layer of n-type semiconductor,
Introducing an alkali metal thiocyanate M SCN (6) (M is preferably Li, Na or K) by casting into a porous substrate covered by the absorber;
Depositing the cover layer of the P-type semiconductor compound containing CuSCN on a porous substrate by casting the solution, preferably by casting a solution of dipropyl sulfide, and optical soaking treatment on the resulting device Applied.

The advantages of the novel photovoltaic device according to the invention are that it is low in production cost and provides high photovoltaic performance when stored in a sealed state. This is because materials that do not require high purity can be used and these materials can be manufactured with low-cost manufacturing techniques (especially these materials require high vacuum technology as used in silicon-based devices. Because it does not.)

The photovoltaic cell of the present invention exhibits better photovoltaic performance compared to previous similar concepts. In particular, it shows a high photocurrent and higher photovoltaic conversion efficiency in the short circuit. To be more precise, the present invention shows that the short circuit photocurrent is 10 mA / cm ² or more at 1000 W / m ² and the open circuit voltage Voc (at I = 0) is 0.50 V. It provides a cell that is larger and exhibits a conversion efficiency greater than 3% and that the stability of photovoltaic performance remains stable for at least more than 300 days.

  In addition to this penetration type, the three compounds satisfy certain conditions, in particular the agreement of the respective energy bands. If the band model is taken for a semiconductor, as a first approximation, the conduction band of the absorber must be a negative value that is even smaller than the conduction band of the n-type semiconductor (when most commonly used, the energy level is Negative value, starting from zero, or empty level). And the valence band of the absorber must be more negative than the valence band of the p-type semiconductor. And all this is to allow injection of electrons and holes respectively.

  The fact that the three compounds are inorganic solid members gives the superior potential of longer durability for outdoor applications, and more particularly for application on the roof of a house.

  Further advantages and features of the present invention will become apparent from the detailed description of various examples of the following examples and the results described in the following drawings.

TiO ₂ / Sb ₂ S ₃ / (LiSCN) CuSCN cell illuminated by _{one sun (1006 W / m} ² ) _{measured at different times t after the start of optical soaking in the open air} The photocurrent density (mA / cm ² ) curve for voltage (V) (IV curve) (cell active area is 0.54 cm ² ), (a) t = 0; (b) t = 10 min (c) t = 20 minutes; (d) t = 100 minutes. ˜1 solar TiO ₂ / Sb ₂ S ₃ in an open circuit state during light soaking, initially in dry N ₂ (A), then dry air (B), and finally in a dry N ₂ atmosphere (C) / (LiSCN) represents the curve evolution (development) of the short-circuit current density Jsc over time (t = 0 to 3000 minutes) of the CuSCN cell. It shows three different cell types, (○), the light soaking before type to (stored in 15 hours N ₂ glove box at room temperature) are not thermal treatment; (●) is 80 ^o The type heated in the N ₂ glove box for 15 hours at C, and (▲) indicates the type heated in the synthetic dry air glove box at 80 ^° C. for 15 hours. ˜1 solar TiO ₂ / Sb ₂ S ₃ in an open circuit state during light soaking, initially in dry N ₂ (A), then dry air (B), and finally in a dry N ₂ atmosphere (C) / (LiSCN) represents the curve evolution (development) of the open circuit voltage Voc over time (t = 0 to 3000 minutes) in the CuSCN cell. It shows three different cell types, (○), the light soaking before type to (stored in 15 hours N ₂ glove box at room temperature) are not thermal treatment; (●) is 80 ^o The type heated in the N ₂ glove box for 15 hours at C, and (▲) indicates the type heated in the synthetic dry air glove box at 80 ^° C. for 15 hours. ˜1 solar TiO ₂ / Sb ₂ S ₃ in an open circuit state during light soaking, initially in dry N ₂ (A), then dry air (B), and finally in a dry N ₂ atmosphere (C) / (LiSCN) represents the curve evolution (development) of the filling factor ff over time (t = 0 to 3000 minutes) of the CuSCN cell. It shows three different cell types, (○), the light soaking before type to (stored in 15 hours N ₂ glove box at room temperature) are not thermal treatment; (●) is 80 ^o The type heated in the N ₂ glove box for 15 hours at C, and (▲) indicates the type heated in the synthetic dry air glove box at 80 ^° C. for 15 hours. ˜1 solar TiO ₂ / Sb ₂ S ₃ in an open circuit state during light soaking, initially in dry N ₂ (A), then dry air (B), and finally in a dry N ₂ atmosphere (C) / (LiSCN) represents the curve evolution (development) of the efficiency η (%) with time change (t = 0 to 3000 minutes) of the CuSCN cell. It shows three different cell types, (○), the light soaking before type that is not the thermal treatment to (stored in 15 hours N ₂ glove box at room temperature); (●) in 80oC The type heated in the N ₂ glove box for 15 hours, and (▲) indicates the type heated in the synthetic dry air glove box at 80 ° C. for 15 hours. ˜1 solar TiO ₂ / Sb ₂ S ₃ in an open circuit state during light soaking, initially in dry N ₂ (A), then dry air (B), and finally in a dry N ₂ atmosphere (C) / (LiSCN) represents the curve evolution (development) of the series resistance Rs (Ω) over time (t = 0 to 3000 minutes) of the CuSCN cell. It shows three different cell types, (○), the light soaking before type that is not the thermal treatment to (stored in 15 hours N ₂ glove box at room temperature); (●) in 80oC The type heated in the N ₂ glove box for 15 hours, and (▲) indicates the type heated in the synthetic dry air glove box at 80 ° C. for 15 hours. Curve evolution of photocurrent Jsc (mA / cm ² ) over time (t = 0 to 400 days) of four performance parameters normalized to one sun (shown below) is shown. Three cell types TiO ₂ / Sb ₂ S ₃ / (LiSCN) CuSCN ˜1 With continuous irradiation in the sun, curve evolution under controlled atmospheric air flow in three different conditions Show. These three flows are (◆) N ₂ flow, open circuit (OC) state; (△) N ₂ flow, short circuit (SC) state; (●) synthetic air flow, open circuit (OC) state. The curve development of the open circuit voltage Voc (V) in the time change (t = 0 to 400 days) of four performance parameters normalized to one sun (shown below) is shown. Three cell types TiO ₂ / Sb ₂ S ₃ / (LiSCN) CuSCN ˜1 With continuous irradiation in the sun, curve evolution under controlled atmospheric air flow in three different conditions Show. These three flows are (◆) N ₂ flow, open circuit (OC) state; (△) N ₂ flow, short circuit (SC) state; (●) synthetic air flow, open circuit (OC) state. It shows the curve evolution of efficiency (%) over time (t = 0 to 400 days) of four performance parameters normalized to one sun (shown below). Three cell types TiO ₂ / Sb ₂ S ₃ / (LiSCN) CuSCN ˜1 With continuous irradiation in the sun, curve evolution under controlled atmospheric air flow in three different conditions Show. These three flows are (◆) N ₂ flow, open circuit (OC) state; (△) N2 flow, short circuit (SC) state; (●) synthetic air flow, open circuit (OC) state. The curve evolution of the filling factor over time (t = 0 to 400 days) of four performance parameters normalized to one sun (shown below) is shown. Three cell types TiO ₂ / Sb ₂ S ₃ / (LiSCN) CuSCN ˜1 With continuous irradiation in the sun, curve evolution under controlled atmospheric air flow in three different conditions Show. These three flows are (◆) N ₂ flow, open circuit (OC) state; (△) N ₂ flow, short circuit (SC) state; (●) synthetic air flow, open circuit (OC) state. Between light soaking, in an open circuit condition, the initial dry N ₂ gas (A), then dry air (B), and finally dried N ₂ TiO ₂ / Sb at ~ 1 sun in a gas atmosphere (C) ₂ shows the evolution of the efficiency η curve over time from t = 0 to 3000 minutes for the ₂ S ₃ / (alkali metal thiocyanate) CuSCN cell. These show four different thiocyanate treatments and three different heat treatments for each thiocyanate treatment, (●) for non-salt treatment, (▲) for LiSCN treatment, (■) for KSCN treatment, (◆) Is the NaSCN treatment on the non-heat treated cell, and the corresponding white symbol indicates the N ₂ heat treated cell. Gray symbols indicate cells that have been heat-treated with dry air. Between light soaking, in an open circuit condition, the initial dry N ₂ gas (A), then dry air (B), and finally dried N ₂ TiO ₂ / Sb at ~ 1 sun in a gas atmosphere (C) ₂ S ₃ / (alkali metal thiocyanate) shows the curve evolution of the filling factor ff of the CuSCN cell over time from t = 0 to 3000 minutes. These show four different thiocyanate treatments and three different heat treatments for each thiocyanate treatment, (●) for non-salt treatment, (▲) for LiSCN treatment, (■) for KSCN treatment, (◆) Is the NaSCN treatment on the non-heat treated cell, and the corresponding white symbol indicates the N ₂ heat treated cell. Gray symbols indicate cells that have been heat-treated with dry air. Between light soaking, in an open circuit condition, the initial dry N ₂ gas (A), then dry air (B), and finally dried N ₂ TiO ₂ / Sb at ~ 1 sun in a gas atmosphere (C) ₂ S ₃ / (alkali metal thiocyanate) shows the curve evolution of the short circuit current density Jsc over time from t = 0 to 3000 minutes for the CuSCN cell. These show four different thiocyanate treatments and three different heat treatments for each thiocyanate treatment, (●) for non-salt treatment, (▲) for LiSCN treatment, (■) for KSCN treatment, (◆) Is the NaSCN treatment on the non-heat treated cell, and the corresponding white symbol indicates the N ₂ heat treated cell. Gray symbols indicate cells that have been heat-treated with dry air. Between light soaking, in an open circuit condition, the initial dry N ₂ gas (A), then dry air (B), and finally dried N ₂ TiO ₂ / Sb at ~ 1 sun in a gas atmosphere (C) ₂ S ₃ / (alkali metal thiocyanate) shows the curve evolution of open circuit voltage Voc over time from t = 0 to 3000 minutes for a CuSCN cell. These show four different thiocyanate treatments and three different heat treatments for each thiocyanate treatment, (●) for non-salt treatment, (▲) for LiSCN treatment, (■) for KSCN treatment, (◆) Is the NaSCN treatment on the non-heat treated cell, and the corresponding white symbol indicates the N ₂ heat treated cell. Gray symbols indicate cells that have been heat-treated with dry air. Between light soaking, in an open circuit condition, the initial dry N _{2 (A),} then dry air (B) and finally dried N ₂ atmosphere (C) in to 1 sun TiO ₂ / Sb ₂ S ₃ / (Alkali metal thiocyanate) The curve progression of the cell series resistance of CuSCN cell over time from t = 0 to 3000 minutes is shown. These show four different thiocyanate treatments and three different heat treatments for each thiocyanate treatment, (●) for non-salt treatment, (▲) for LiSCN treatment, (■) for KSCN treatment, (◆) Is the NaSCN treatment on the non-heat treated cell, and the corresponding white symbol indicates the N ₂ heat treated cell. Gray symbols indicate cells that have been heat-treated with dry air. Photovoltaic density (mA / cm ² ) curves for TiO ₂ / Sb ₂ S ₃ / (KSCN) CuSCN cells versus voltage (V) (IV curve) at different concentrations of KSCN immersion solution in one sun . 1 schematically shows a structure observed by a transmission electron microscope (TEM) analysis of an internal structure of a cell having an oxidized interface layer according to the present invention.

(1-Experiment section)
<1.1. Preparation of TiO2 compact and porous layer>
F-doped SnO ₂ transparent conductive oxide (TCO) glass from Asahi Glass was used as “pre-contact”. A hole barrier compact TiO ₂ layer of thickness ˜20 nm (confirmed by SEM and TEM) was deposited by spray pyrolysis ¹ . A porous nanocrystalline TiO ₂ film with a thickness of ˜3 μm, an average particle size of ˜40 nm-50 nm and a roughness coefficient of 150-300 was prepared as described in the prior art.
<1.2. Fabrication of TiO ₂ / Sb ₂ S ₃ / (Alkali Metal Thiocyanate) CuSCN Cell>
Antimony sulfide was deposited on the porous nanocrystalline TiO _{2 by} chemical bath deposition (CBD) ^3-9 as described below. This CBD is prepared by first preparing 1M of SbCl ₃ solution in acetone so that the final concentrations are ˜0.025M Sb ³⁺ and ˜0.25M S ₂ O ₃ ^2- , respectively. Was diluted with ˜1 M Na ₂ S ₂ O ₃ cold aqueous solution and cold water. Samples were immersed perpendicular to the bath and placed in a 5-10 ^° C refrigerator. The CBD reaction was typically left for 2 hours. The sample is then immersed in each of two beakers containing distilled water, shaken for about 5 seconds, and blown with nitrogen gas horizontally on the surface of the sample for several seconds to dry in the two beakers. Washed twice. The sample was then annealed in a nitrogen glove box at a temperature of 300-320 ^° C. for about 15-30 minutes.

  The sample was then immersed for 30 seconds in a 0.5 M aqueous solution of either LiSCN, KSCN, or NaSCN salt. Non-salt treated samples were made for comparison as well. For KSCN salt, other concentrations were prepared for comparison. The sample was then dried by blowing nitrogen gas over the surface for several seconds.

After this process, CuSCN filling was performed by impregnation and evaporation of CuSCN solution in dipropyl sulfide. And immediately after CuSCN filling, a “post-contact” of about ˜40 nm thick gold layer was evaporated by thermal evaporation using an Edward 306 evaporator.
<1.3 Cell drying by heat treatment>
Some of the completed cells were stored until they were characterized in three different conditions:

1. 15 hours at room temperature in a nitrogen glove box (hereinafter referred to as non-heat-treated cell)
2. 15 hours on an 80 ° C heating plate placed in a nitrogen glove box (hereinafter referred to as N ₂ heat treatment cell)
3. 15 hours on an 80 ° C heating plate placed in a synthetic dry air glove box (hereinafter referred to as dry air heat treatment cell)
These post-treatments after post-contact electrode deposition, ie the final fabrication process of the cell, were tested to check the drying effect of the cell (eg removal of possible traces of dipropyl sulfide used in CuSCN filling). Effect of moisture remaining in the cell).

Current-voltage IV curve measurements and quantum efficiencies (also known as QE, IPCE) were recorded in reported setup ² .
<1.4 Optical soaking process>
The cell was light soaked in various gas atmospheres to observe the effect of gas and heat treatment. This optical soaking process is to irradiate ˜1 sun (= approximately 1000 W / m ² ) in an open circuit state in a controlled gas atmosphere for 20 to 1200 minutes. The cell was exposed to a solar simulator under continuous illumination with illumination close to one sun. The cell temperature was not controlled, but was in the range of ~ 30-35 ° C.

Optical soaking is performed under the following three conditions (a) ambient air (room air), or (b) a controlled atmosphere of synthesized dry air, or
(C) N ₂ gas flow,
Made in

  During light soaking, I-V curves were measured regularly.

Prior to light soaking, the cell was sealed as follows. The cell is in a cell container made of a metal box with a glass window for light illumination and a gas inlet and gas outlet that allows a controlled gas atmosphere under one of the following controlled gas flows: It was put.
(a) dry N ₂ or
(b) Dry synthetic air.

  And the measurement of the sealed cell in the atmosphere controlled in that way was implemented.

The cell was initially light soaked in dry synthetic air for several hours. The gas atmosphere was changed to N ₂ or remained dry air and a slight flow of such gas continued. The cell was either an open circuit or a short circuit, and performance was measured at appropriate times. The evolution of performance parameters for the three cells under different conditions was monitored over the course of almost a year in these numerous stability tests.
<1.5. Measurement>
Similar to the progress of Jsc (mA / cm ² ), Voc (V), ff, series resistance Rs (Ω), and efficiency η (%), IV curves were also measured regularly.

It is to create the derivative of the IV curve (at single exposure) scanned to a sufficiently high voltage forward bias to create a flat on the (-dV / dI) vs. V plot. Gives the value of Rs. SEM images were obtained with a field emission scanning electron microscope Hitachi S-4700. This microscope was equipped with EDX (energy dispersive X-ray microanalysis) Nolan system SIX. TEM analysis was performed with a JEOL 2100F FEG-200kV microscope, which was equipped with a STEM accessory and an integrated JEOL JED-2300T EDX analyzer.
(2. Examples)
Examples 1 to 4 below were performed on the materials according to the methods disclosed in the first paragraph above. The results are explained in detail in the third paragraph.
[Comparative Example (1)]
An untreated sealed cell TCO / nanocrystal TiO ₂ / Sb ₂ S ₃ / CuSCN / Au was made for use and photovoltaic performance was evaluated in an N ₂ atmosphere. The energy conversion efficiency was less than 0.1%.
[Example (2)]
The same cell as Comparative Example (1) was treated by the optical soaking of the present invention and characterized under ambient air (room air). Due to the formation of the interfacial layer of oxidizing species, the efficiency increased to 1.2% 15 hours after the start of illumination. However, if changing the atmosphere N _2, the insufficient formation of the interface layer between the composition of the cell, the performance dropped to 0.4%.
[Comparative Example (3)]
TCO / nanocrystal-TiO ₂ / Sb ₂ S ₃ / LiSCN, CuSCN / Au cells were fabricated and characterized for photovoltaic performance under N ₂ atmosphere in the illumination process. Prior to CuSCN deposition, LiSCN was introduced into the cell by immersing the sample in 0.05M LiSCN aqueous solution. The initial efficiency at the beginning of the illumination was ˜0.2%. After that, when the cell was left open for 5 days in an illuminated state, it increased to 1.5%. The slow progress in performance is due to the lack of oxygen to form the interface layer. It appears that only oxygen that is initially adsorbed and present in the cell functions. The formation of the interface layer is not evident due to the limited amount of oxygen.
[Embodiment (4) of the present invention]
The same cell as Comparative Example (3) was illuminated in dry air and measured. The initial efficiency was ˜0.2%, and then increased to 3.8% after 2 hours in an open circuit under illumination. The rapid progress in performance is due to the formation of an interface layer in the presence of oxygen and illumination. After the light soaking process, changing to the N ₂ atmosphere, the efficiency higher than 1.7% lasted more than 300 days.
(3. Results and discussion)
<Cell performance in optical soaking in ambient air>
The cell showed an important light soaking effect that when exposed to ambient air during light soaking and storage, the cell performance typically increased from an initial low efficiency to a higher one. Oxygen has been shown to be related to this effect. The best cell efficiency of the cell was 3.7% with ~ 1 sun for the cell treated with LiSCN. The cell was shown to operate for more than 300 days, maintaining an efficiency higher than 1.5% at ~ 1 sun.

The cell was stored overnight in a N ₂ glove box prior to measurement. High cell efficiency was obtained after the cell was left in open circuit for 20 to 200 minutes in ambient air at 1 sun in cell light soaking condition. Prior to light soaking, the cell typically showed less than 1% (<1%) efficiency in its first IV measurement. Figure 1 shows the evolution of the IV curve of one of the best cells during optical soaking. As seen in this figure, the performance improves with time, the highest cell efficiency η is 3.7% at ˜1 sun (1006 W / m ² ), t = 100 minutes after starting the illumination. This shows a photocurrent Jsc normalized to 11.6 mA / cm ² per sun, an open circuit voltage Voc of 0.56 V, and a filling factor ff of 0.58. In contrast, the initial efficiency before starting illumination (t = 0) was only 0.8%. The time to maximum efficiency varied from cell to cell. In a systematic experiment consisting of 36 cells made and measured under the same conditions as the LiSCN treatment, the cell efficiency is in the range of 1.5-3.7% (average 3.9%) and the Jsc is 6. 1-13.2mA / cm ² (average 10.5mA / cm ² ), Voc is 0.43-0.58V (average 0.53V), and ff is 0.44-0.60 (average) 0.52).

Despite the introduction of LiSCN treatment for cells with TiO ₂ / Sb ₂ S ₃ / CuSCN structure due to its high performance, unsalt-treated cells are also initially less than 0.5% at 1 sun The efficiency was found to show a light soaking effect from 1 to 1.5% final efficiency (Jsc -6-9 mA / cm ² , Voc ˜0.4, ff ˜0.4). These performance evolutions during light soaking were slower than LiSCN cells, and the time to maximum performance was typically -10-20 hours. High series resistance and low voltage Voc are the characteristics of the unsalted cell. Different salt treatment cells were also investigated using different thiocyanates such as KSCN and NaSCN. These behaved slightly differently, but the general trend was similar to that of LiSCN. The difference in operation seems to be related to the difference in hygroscopicity or the difference in cation type or volume. A detailed report is given below.

This light soaking effect was also generated by radiation excluding the UV portion (100% less than 500 nm was cut with a UV cut-off filter). Thus, contrary to the UV effect reported using other nanocrystalline TiO ₂ cells, such a light soaking effect is not limited to UV alone ¹⁰ .

Following the above results, stability evaluation of sealed cells made in N ₂ atmosphere was started. Then, it seemed that a similar light soaking effect was not seen in cells exposed to ambient air. This suggests that oxygen and / or moisture are involved in the observed light soaking effect. To clarify these effects, experiments were conducted under the following controlled atmosphere.
<3.2. Effects of gas and heat in optical soaking>
Three different types of cells subjected to heat treatment or non-heat treatment, as disclosed in paragraph 1.3, were light soaked. This treatment is performed under different gas atmospheres to observe the effect of gas and heat treatment, light soaking is ~ 1 sun (cell temperature maintained at 40 ° C.), first dry N ₂ in 500 minutes (8 hours), 1000 minutes following the dry air (16.5 hours), last took place 1200Min (to 20 hours) at once dry _{N 2.} IV curves were measured regularly during optical soaking in an open circuit condition. FIG. 2 shows the progress of Jsc (mA / cm ² ), Voc (V), ff, series resistance Rs (Ω), and efficiency η (%) during optical soaking. It can be seen that for all three different types of cells, there is a clear difference in all parameters between the first dry N ₂ and the light soaking of the next dry air. It is clear that oxygen plays an important role in improving cell performance.

In the first N ₂ optical soaking seen in FIG. 2, the N ₂ heat treated cell showed the lowest η among the three cell types with the lowest Jsc and Voc. Even for this N ₂ heat-treated cell, there was a little progress in performance parameters such as Voc, ff and Rs, but η was a much lower constant value. Note that the initial series resistance of the N ₂ heat treated cell is very high compared to the other cells. For the non-heat treated cell, as shown in the figure, a clear progress of η was observed in comparison with this N ₂ heat treated cell. The initial series resistance value was the lowest among the three types of cells. It can be seen that the dry air heat treatment cell shows the highest Jsc, Voc, and η among the three cells at the beginning and a much lower Rs than the N ₂ heat treatment cell. Then, the Jsc, Voc, and η of the dry air heat treatment cell were lower than those of the non-heat treatment cell within 30 minutes of starting light soaking. These movements in the first dry N ₂ light soaking are interpreted as follows. Non-heat treated cells contain the most residual water of all cells and have the lowest Rs before light soaking. Furthermore, this non-heat treated cell contains oxygen adsorbed during cell manufacture and enhances the optical soaking effect associated with oxygen. Residual water originated from hygroscopic LiSCN involved in the wet chemical process of cell fabrication. The relationship between series resistance and moisture content is based on results obtained in separate experiments. In that experiment, the series resistance is greatly reduced by the humidity in the surrounding gas. The N ₂ heat treatment cell contains almost no water or adsorbed oxygen in the cell. This results in the highest Rs, and in the sense of performance improvement, there is almost no light soaking effect associated with oxygen. However, for this cell you will notice a surprising decrease in Rs during light soaking in the figure. This is interpreted as a result of the cell still containing a small amount of oxygen that causes this change in illumination. Possible the following reaction to involve oxygen and illumination can be considered: dipropyl sulfide solution organic impurities photocatalytic oxidation and / or SCN, such as ^- with (SCN) ₂ is formed by the photocatalytic oxidation, or OH Addition of reactive oxygen species such as radicals and CuSCN. The dry air heat treatment cell is believed to contain less water than the non-heat treatment cell and contain more adsorbed oxygen than the N ₂ heat treatment cell. Thus, a high oxygen content lowers the initial Rs even immediately after light soaking has begun, even though it contained less water prior to light soaking. Another possible interpretation is that photosoaking occurs in N ₂ without adsorbed oxygen, but the increase in photocurrent occurs over a longer time span (several days). In that case, this process may be due to the removal of organic impurities in the gas flow or photolysis. Maybe in this case there may be assistance of photoactive antimony sulfide. However, as explained below, an advantageous oxygen effect on Voc and series resistance cannot be obtained under N ₂ . And the final efficiency in the N ₂ atmosphere after several days of continuous lighting will be lower than in the dry air atmosphere.

  In the next dry air light soaking, a rapid evolution of Jsc, Voc and η can be seen in FIG. It should be noted here that immediately after changing to dry air light soaking in the figure, there is an initial decrease in ff in all three cell types. Such a decrease corresponds to the increase in Jsc seen in the figure. The decrease in ff suggests that cell performance evolution is not necessarily related only to series resistance. The inventor interprets this behavior as follows. At the start of dry air light soaking, the photocurrent is increased by some sort of mechanism. However, the series resistance is not low enough to maintain a high ff value with the photocurrent increased in the early stages of dry air photosoaking, and therefore ff will decrease. However, the decrease in Rs, along with the process, slowly recovers ff after reaching the minimum value. The increase in Voc with dry air also contributed to the recovery of ff. This is particularly evident in dry nitrogen gas heat treatment cells. From the above discussion, it can be concluded that oxygen has two independent roles: the first one is to reduce Rs, the second is to make Jsc and Voc independent of its Rs. To increase. Although it has been found that water is particularly oxygen, which promotes cell degradation, the role of residual water present in the cell is not yet clear at this point.

In the last dry N ₂ light soaking following dry air light soaking, the Jsc of all three cell types increases for the first few hours and then decreases slowly. As seen in FIG. 2, Voc begins to decrease immediately after changing to dry N ₂ . In the N ₂ heat treatment cell, a decrease in ff and an increase in Rs are seen in FIG. However, such changes are not seen in FIG. 2 for other cells. As seen in the cell stability test in the next paragraph, the initial decrease in cell performance after changing to the N ₂ atmosphere has stabilized with time, and the efficiency has exceeded 300 days in open circuit after light soaking. , Maintaining a value greater than 1.5%. This is one of the most important characteristics of LiSCN processing cell. In the case of non-salt treatment cells that have reached an efficiency of about 1.2% by dry air light soaking, the efficiency is typically less than 0.5% in ~ 15 hours after changing to an N ₂ atmosphere. (This plot is not shown in FIG. 2).
<3.3. Cell stability test>
The cell was made according to the standard procedure described above with LiSCN treatment. These cells were initially light soaked in dry synthetic air for several hours. The gas atmosphere was either changed to N ₂ or maintained as dry air, and such gases were continuously flowed in small quantities. The cell was in either an open circuit or a short circuit, and performance measurements were taken from time to time. FIG. 3 shows the evolution of the performance parameters of the three cells in these stability tests over almost a year under different conditions. The cells under N ₂ in the short circuit are stable, but the other two cells in the open circuit (N ₂ or air) have degraded performance in the first 100 to 150 days. After that, it tends to stabilize. The efficiency after one year is over 1.5% with three cells. The Jsc at 1 sun is maintained within 7.5-10 mA / cm ² . The vertical shift (movement) observed in the plot is due to the irradiance drift of the old lamp used in these stability tests, no matter how well calibrated. It may also be due to cell temperature changes between measurement points. Compared to the stability test in ambient air, ambient moisture also seems to be the cause of cell degradation, which must be avoided.

Water is effective for the activation of the initial MSCN or CuSCN oxidation, but after reaching the highest efficiency, water can also cause or promote further absorber oxidation, resulting in cell Causes deterioration.
＜ 3.4. TiO ₂ / Sb ₂ S ₃ / (Alkali metal thiocyanate) CuSCN cell>
As already confirmed above, treatment with LiSCN in cell fabrication and ambient gas in measurement greatly affects cell performance, but is treated with other thiocyanates such as KSCN, NaSCN and non-salts. Several different types of cells were characterized by IV curve measurements to gain the essence of thiocyanate and ambient gas effects. For each thiocyanate-treated cell and non-salt-treated cell, three different conditions for sample storage were used to create the three cell types (see experimental section) described below after cell fabrication was completed. Was adopted. The three cell types are an N ₂ heat treatment cell, a dry air heat treatment cell, and a non-heat treatment cell. These three different types of cells were then optically soaked to observe the effects of gas and heat treatment under different gas atmospheres. Light soaking was done at ~ 1 sun (maintaining cell temperature at 40 ° C.), initially 500 minutes with dry N ₂ , then 1000 minutes with dry air, and finally again with dry N ₂ for 1200 minutes. . Light soaking is an open circuit state with ~ 1 sun of visible light illumination. During this light soaking, IV curves were measured regularly. Figure 1 shows the development of energy conversion efficiency η (%), short circuit current density Jsc (mA / cm ² ), open circuit voltage Voc (V), and filling factor ff, for each type treated with thiocyanate. Shown during light soaking of 12 different cells consisting of non-salt, LiSCN, NaSCN and KSCN treated cells in 3 different sample storage conditions.

Initially, the non-thiocyanate treatment cell was focused. The highest efficiency was ˜1.2%, which was a type of cell with dry N ₂ light soaking t = ˜1450 minutes. As seen in FIG. 4, apart from this experiment, the highest efficiency currently recorded is 1.5%.

During light soaking, all cells show similar characteristics in the evolution of performance parameters. Note in FIG. 4 that the initial efficiency, Voc, and Jsc are the highest among the three different heat treatments when the dry air heat treated cell is the beginning of light soaking starting in dry N ₂ and ff is It was the lowest. Then, η, Voc and Jsc decreased in dry N ₂ with time and ff increased. In the subsequent dry air atmosphere, as seen in FIG. 4, all three cells responded quickly to the introduction of air, η, Voc, Jsc increased and ff decreased. At the end of the dry air light soaking, the non-heat treated cell showed the highest efficiency η. As can be seen, it should be noted that this ff has begun to recover at the midpoint of dry air light soaking. In subsequent introduction of the second N ₂ at t = ˜1500 min, η, Voc, and Jsc began to drop gradually. However, a temporary increase in Jsc during the initial period was observed in the non-heat treated cell. The results of these non-heat treated cells clearly show that oxygen is involved in the photovoltaic process. It is considered that the heat treatment affects the adsorption of oxygen and water, which causes developmental differences between light soaking.

The next focus was on the thiocyanate treatment cell. The efficiency of the nine thiocyanate treated cells of FIG. 4A was 1.8% to 3%, with the cell treated with dry air heat treatment with KSCN treatment showing the highest efficiency. However, the observed differences in cell characteristics may be due to experimental variation due to thiocyanate differences or other unknown factors. Because, in separate isolated experiments, a similar degree of variability between 1.5% and 3.5% (average: 2.9%) is seen for 36 identical cells treated with LiSCN Because. It was also found that moisture in the measurement atmosphere affects the progress in cell performance during light soaking. These facts are because the heat treatment was performed in the dry gas state in this study. The following general rules were extracted by repeating the same experiment several times. The thiocyanate treated cell consistently performs better than the non-salt treated cell after dry air light soaking. The dry N ₂ light soaked KSCN treatment cell shows the highest η, Jsc, and Voc of the four different thiocyanate treatments. For the KSCN treatment cell, the dry air heat treatment cell shows the highest cell performance among the three different heat treatments. For LiSCN-treated cells, non-heat treated cells consistently show the highest cell performance. The ff of the KSCN treatment cell is the highest, and that of LiSCN (ff) is the lowest. During the introduction of the second dry N ₂ followed by light soaking under dry air, the slow decrease begins after the first temporary Jsc increase, A decrease is also observed.

  Data not shown, but light soaking has always been performed in ambient air (room air) for non-heat treated cells containing thiocyanate, especially before the effects of oxygen and moisture were confirmed . The following general characteristics were confirmed under such conditions. In the case of KSCN and NaSCN treatment cells, the initial light soaking progress is not reproducibly dependent on the water content. They are very sensitive to water coming from ambient air and water that was originally present in the cell. A high water content results in very low performance. In LiSCN treated cells, even at high water content, the progress is observed with an initial, high Jsc, low ff, and distorted S-shaped IV curve. Cell characteristics typically vary from cell to cell and over time. Even if the performance progress is performed at an appropriate humidity during the optical soaking in the ambient air condition (in the room air condition), the performance starts to decline slowly after reaching the maximum performance by the initial optical soaking. Typically, the degree of humidity drops to less than 1% efficiency in a few days. If the performance of these cells begins to damp in moisture, they will not recover even after drying. This suggests that irreversible changes occur with water. On the other hand, non-salt treatment cells do not suffer from such moisture effects on a scale of tens of hours. Typically, these efficiencies begin to increase upon introduction of moist air by improving ff and reducing cell series resistance.

FIG. 5 shows the cell series resistance of the 12 different cells, which were measured several times during various times during light soaking. At t = 0 (actually several tens of seconds to several minutes later than the time to start lighting due to sequential data acquisition), all of the four N ₂ heat treatment cells of KSCN, LiSCN, NaSCN, and non-salt As can be seen, it had a higher series resistance compared to the others. It should be noted here that in the figure there is a clear difference between the thiocyanate-treated and non-salt-treated cells, the former showing low resistance. From the fact that heat treatment makes a difference in series resistance, it can be assumed that the initial series resistance at t = 0 is related to the amount of initial adsorbed oxygen and water in the cell. Thus, the results shown in FIG. 5 can be interpreted as the result of a photocatalytic reaction that occurs during photosoaking involving oxygen, thiocyanate, water, and other residual impurities. However, it should be remembered here that the series resistance measured in a dry N ₂ atmosphere is not high enough to suppress Jsc sufficiently. That is, in FIG. 1, the series resistance itself is not the main factor with low η and Jsc during dry N ₂ optical soaking. Thus, the improvement in cell performance observed when oxygen is introduced can be interpreted as a result of improved injection of photo-generated carriers for some reason.
<3.5. Effect of solution concentration used for KSCN treatment>
FIG. 6 shows the IV curve at 1 sun of a cell that was KSCN treated with various concentrations of KSCN immersion solution. The IV curves for the highest performance obtained during ambient air (room air) light soaking are shown in the figure for each KSCN soak solution concentration. Very surprisingly, even cells treated with a 5M dipping solution show considerable performance. This is because it was believed that thiocyanate was deposited on the absorber-buried TiO ₂ layer and was present only at the boundary between them and the CuSCN layer. This is because the effect of thiocyanate on the TiO ₂ / CdS / CuSCN cell was first discovered. To clarify this initial consideration, a TEM analysis was performed. TEM analysis was performed with STEM-HAADF imaging at the center of the cross section of a 5M KSCN-treated cell and an EDX spectrum in the region corresponding to the inset of the image. The estimated atomic ratio of K / Cu was ~ 1.7 for this sample. However, the values were scattered depending on the position and were between 0.3 and 2. Several TEM and STEM imaging in different modes showed that the CuSCN and KSCN layers were made of nanocrystalline dots with a size of ˜5 nm. Each material was indistinguishable from the TEM image. However, data indicate that KSCN is distributed with CuSCN through the porous TiO _2. A similar analysis was performed on a 0.5M KSCN treated cell. The estimated K / Cu atomic ratio was 0.03 to 0.07. Similarly, KSCN was distributed with CuSCN.
<3.6. Structural analysis>
Analysis using TEM, STEM / HAADF imaging and STEM / EDX mapping shows that the morphology of the cell structure according to the present invention is schematically represented as in FIG. 7, in which cell (1) Includes:
A porous substrate (2) formed of TiO ₂ nanocrystals with a thickness of more than 1 μm, preferably 10 to 100 nm constituting a film with a thickness of 2 to 3 μm,
The first surface of the porous substrate is made of an n-type semiconductor metal oxide, preferably conductive and transparent, and more compact (non-porous) and transparent, like conductive glass. Coated with a barrier layer (7a), deposited on the first substrate of the pre-contact (7), and the absorber forms an aggregate (4) of size 10nm-200nm, said absorption Body aggregates are embedded in TiO ₂ nanocrystals, with smaller CuSCN aggregates. And the absorber does not cover TiO ₂ uniformly or completely in the Sb ₂ S ₃ deposition process, and • CuSCN formed by well-crystallized ~ 3-5 nm nano-domains (3a) The cover layer (3) of at least 10%, preferably more than 15%, of the volume of the pores (2a) of the porous TiO ₂ substrate, and MSCN fine particles (6) are Sb ₂ S ₃ It is placed in the CuSCN cover layer (3) in the same way as the porous TiO ₂ structure (2) embedded with the absorber (4). And MSCN (6) and CuSCN agglomerates (3a) are mixed and evenly distributed within the porous structure and into the upper layer of CuSCN (3), and • part of the interface layer of oxidizing species (5) Is formed between several agglomerates of the absorber and the TiO ₂ particles (2b) or CuSCN particles (3), and the interface layer portion has a thickness of 0.1 nm to 5 nm. ,
The second surface of the porous substrate (2) is covered by a second conductive substrate in post-contact (8), and the porous substrate is also at least 10 nm thick 2 and an upper layer disposed between the conductive layers of the back contact (8a) of the conductive substrate, and the porous substrate, its absorber and cover layers (1, 2, 3) are in front contact ( 7) being confined between the two conductive substrates and the back contact (8).

A metal box with a glass window was used in the above experiment to conveniently seal these cells. However, the actual cell will be covered to seal the cell with the wall 9 by a transparent glass or resin film. The electrode 9b passes through the wall 9 so that the front contact set and the rear contact set 8 can come into contact with each other.
<Refer to bibliography:>
(1) Kavan, L .; Gratzel, M. Electrochim. Acta 1995, 40, 643.
(2) Larramona, G .; Chone, C .; Jacob, A .; Sakakura, D .; Delatouche, B .; Pere, D .; Cieren, X .; Nagino, M .; Bayon, R. Chem. Mater 2006, 18, 1688.
(3) Mandal, KC; Mondal, AJ Phys. Chem. Solids 1990, 51, 1339.
(4) Deshmukh, LP; Holikatti, SG; Rane, BP; More, BM; Hankare, PPJ Electrochem. Soc. 1994, 141, 1779.
(5) Lokhande, CD; Sankapal, BR; Mane, RS; Pathan, HM; Muller, M .; Giersig, M .; Ganesan, V. Appl.Surface Sci. 2002, 193, 1.
(6) Grozdanov, IT Semicond. Sci. Technol. 1994, 9, 1234.
(7) Nair, MTS; Pena, Y .; Campos, J .; Garcia, VM; Nair, PKJ Electrochem. Soc. 1998, 145, 2113.
(8) Salem, AM; Selim, MSJ Phys. D: Appl. Phys .. 2001, 34, 12.
(9) Yesugade, NS; Lokhande, CD; Bhosale, CH Thin Solid Films 1995, 263, 145.
(10) O'Regan, B .; Schwartz, DT, Chem. Mater. 1998, 10, 1501.

Claims (12)

A solid photovoltaic cell (1) of the type comprising three inorganic and solid materials, each comprising:
A porous layer of transparent n-type semiconductor compound (2),
A transparent p-type semiconductor compound layer (3) that fills at least some of the pores of the porous layer, and two semiconductor compounds of the n-type and p-type layers that are not at least partially in contact An inorganic absorber compound which forms an aggregate (4) between,
Further, the semiconductor compound and the agglomerate include different oxidation products (5), and the oxidation products contain chemical species caused by partial oxidation of the absorber compound and / or the p-type semiconductor compound. Including a discontinuous portion of the interface layer disposed between the two layers of the n-type and each of the p-type semiconductor compound and / or between the aggregate of the absorber and the layer of the p-type semiconductor compound. And
The p-type semiconductor compound is CuSCN, and the absorber is a thermally annealed antimony sulfide-based antimony compound;
Said cell further comprises a thiocyanate compound MSCN (6), M is an alkali metal, wherein the interfacial layer (SCN) ₂ and / or polymer (SCN) solid photovoltaic cell, which comprises an _n . Weight ratio CuSCN of MSCN a solid photovoltaic cell of claim 1 which is 2 0.001. The solid photovoltaic cell according to claim 1 or 2 , wherein the cell is sealed within a wall (9), the wall being at least partially transparent, the cell comprising a dry gas atmosphere. The transparent n-type semiconductor compounds, solid photovoltaic cell according to any one of claims 1 to 3 is a metal oxide containing TiO _2, ZnO or SnO _2. The cell has a shape penetrating each other, and the porous substrate (2) of the n-type semiconductor compound has a hole (2a) having a hole diameter of 10 nm to 100 nm, and an inner surface of the hole. Is covered with the aggregate (4) of the absorber compound, and the pores are filled with a cover layer (3) made of at least 10% by volume of the other p-type solid semiconductor compound. The solid photovoltaic cell according to any one of claims 1 to 4 , wherein the interface layer portion of the oxidation product (5) has a thickness of 0.1 nm to 5 nm. A process for producing a cell according to any one of claims 1 to 5 , wherein the oxidation product (5) of the interfacial layer is obtained by oxidation treatment by so-called light soaking treatment, the light soaking. Processing involves irradiation with visible light in an oxygen-containing gas atmosphere of a device comprising the following components of the cell:
The porous layer (2) of a transparent n-type semiconductor compound,
The layer (3) of a transparent p-type semiconductor compound filling at least part of the pores of the porous layer; and
An inorganic absorber compound present as an aggregate (4) located between two layers of the n-type and p-type semiconductor compounds that do not contact at least partly. Said oxygen-containing gas atmosphere, the process according to claim 6 including the lower moisture ratio than the supersaturation moisture percentage of the ambient air in the ambient air der is, ambient temperature. The process according to claim 6 or 7 , wherein the optical soaking process includes irradiating the cell with visible light power of 0.1 kW / m ² to 5 kW / m ² for 5 minutes to 20 hours. 9. A process according to any one of claims 6 to 8 , wherein the light soaking process comprises illuminating the cell in an open circuit state. The series resistor, and the voltage of the photocurrent (Jsc) and the open circuit of the photovoltaic conversion efficiency that depend short fault circuit (Voc) is observed, light soaking process, illuminating as follows the cell including,
Illuminate as long as the photoelectric conversion efficiency is increased, and stop the irradiation when the conversion efficiency is maximized, and / or illuminate as long as the series resistance of the cell is reduced, and the resistance Stop the irradiation when becomes the minimum,
A process according to any one of claims 6 to 9 . The process according to any one of claims 6 to 10 , wherein the optical soaking process is performed on an unsealed cell, and the cell is sealed under a dry gas atmosphere immediately after the optical soaking process. . The cells are fabricated by stepping the following steps:,
The absorber compound is deposited on the porous layer of an n-type semiconductor, and the alkali metal thiocyanate MSCN salt is introduced and injected onto the porous substrate covered with the absorber compound and the said filler layer of · p-type semiconductor compound comprises CuSCN said porous substrate containing said absorbent compound, wherein the alkali metal thiocyanate MSCN is deposited by applying a CuSCN solvent liquid, and, The optical soaking process is performed on the device thus obtained.
12. Process according to any one of claims 6 to 11 .

Images