JP3671183B2 - Method for producing dye-sensitized solar cell - Google Patents

Description

  The present invention relates to a method for producing a dye-sensitized solar cell in which a sensitizing dye is adsorbed on a porous film of metal oxide fine particles formed on the surface of a conductive substrate.

  For the production of a conductive porous titanium oxide thick film used as an electrode material for a dye-sensitized solar cell, a transparent conductive glass substrate is prepared by mixing a fine particle of titanium oxide with a surfactant as a dispersant. The most common method (Non-Patent Document 1) devised by O'Regan and Gratzel et al. Is to apply it uniformly and heat it at 400 to 500 ° C. to burn and remove organic components. It is used.

  In many cases, other oxide fine particle films such as zinc oxide and tin oxide are produced by the same technique. In addition to the electrolytic synthesis method and the sol-gel method, a method using electrophoresis (Patent Document 1) is also known. In each case, heat treatment at a high temperature of 400 ° C. or higher is performed to form a crystal phase or sinter particles. Is necessary for. Therefore, there is a problem that a material that can be used as the conductive substrate is limited, and a light and inexpensive material such as a plastic film cannot be used.

For photocatalysts, a coating solution that can be coated at room temperature and dried to produce a titanium oxide film has also been commercialized, but since there is no necking between particles, there is no electrical conductivity and the dye-sensitized solar cell It cannot be used as an electrode material.
In addition, it has been reported that a conductive thin film can be obtained by coating an aqueous suspension of titanium oxide containing no organic matter on a conductive substrate and baking it at a low temperature (Non-patent Document 2). The problem is that you cannot earn thickness.

  In order to solve the above problems, a film forming method (Non-Patent Document 3) is devised by applying a semiconductor fine particle paste not containing an organic dispersion material to a conductive substrate and then pressing it at a high pressure. Although it has succeeded in producing electrode materials without any problems, there is a problem in the microstructure of the film obtained, and although it has high performance in weak light, when the light intensity is as strong as sunlight, the transport of the electrolyte has reached its peak. Thus, the current does not follow the light intensity. In addition, in such a physical method, it is not possible to improve the characteristics by changing the film forming conditions. For example, even if a porous film produced by a press method is subjected to a heat treatment, no improvement in the performance of solar cells using this is observed (Non-patent Document 4).

JP-A-11-310898 B. O’Regan and M. Gratzel, Nature, 353, 737 (1991). F. Pichot, S. Ferrere, R.J.Pitts and B.A.Gregg, Langmuir, 16, 5626 (1999). H. Lindstrom, A. Holmberg, E. Magnusson, S.-E. Lindquist, L. Malmqvist, A. Hagfeldt, NanoLett. 2001, 1, 97. H. Lindstrom, E. Magnusson, A. Holmberg, S. Sodergren, S.-E. Lindquist, A. Hagfeldt, Sol.Energ. Mater. Sol. Cells 2002, 73, 91

  The present invention has been made in view of the above-described conventional problems, and provides a method for producing a dye-sensitized solar cell capable of producing a light-weight and inexpensive dye-sensitized solar cell with high production efficiency. It is what.

The present invention provides a paste by kneading metal oxide fine particles and a metal alkoxide for bonding the metal oxide fine particles, applying the paste to a conductive substrate, and hydrolyzing the metal alkoxide. Forming a porous film of the metal oxide fine particles on the surface of the conductive substrate,
Next, the porous film is irradiated with ultraviolet rays in an oxygen atmosphere in which oxygen gas is introduced to generate ozone, so-called UV ozone treatment is performed,
Next, a method for producing a dye-sensitized solar cell is characterized in that a sensitizing dye is adsorbed on the porous film (claim 1).

In the manufacturing method, the porous film can be formed without any heat treatment because the porous film is formed by applying the paste to a conductive substrate and hydrolyzing the metal alkoxide. . Therefore, the material used for the conductive substrate is not limited. That is, it is not necessary to use a particularly heat resistant material, and a light and inexpensive material such as a plastic film can be used.
Therefore, a lightweight and inexpensive dye-sensitized solar cell can be obtained.

  In addition, since the formation of the porous film does not require much time for the application of the paste and the hydrolysis of the metal alkoxide, the dye-sensitized solar cell can be produced with high production efficiency.

  As described above, according to the present invention, it is possible to provide a method for producing a dye-sensitized solar cell, which can produce a light-weight and inexpensive dye-sensitized solar cell with high production efficiency.

In the present invention (Invention 1), the metal oxide fine particles include, for example, fine particles of titanium oxide, zinc oxide, or tin oxide.
Examples of the metal alkoxide include Ti alkoxide, Zr alkoxide, Nb alkoxide, and the like.

The metal alkoxide has a film thickness of 0.05 to 1.3 nm when the metal oxide film produced by hydrolysis of the metal alkoxide uniformly covers the surface of the metal oxide fine particles. Thus, it is preferable to mix with the metal oxide fine particles.
In this case, it is possible to obtain a porous film in which the metal oxide fine particles are securely adhered and the sensitizing dye can be sufficiently adsorbed.
When the film thickness is less than 0.05 nm, it may be difficult to sufficiently bond the metal oxide fine particles or the metal oxide fine particles and the conductive substrate. On the other hand, when the film thickness exceeds 1.3 nm, it may be difficult to sufficiently adsorb the sensitizing dye to the porous film.

The said film thickness is a value calculated | required by calculation as follows.
That is, (1) all metal oxide fine particles have the same size and the same particle size, (2) metal alkoxide is uniformly distributed on the surface of metal oxide fine particles, (3) metal alkoxide is completely hydrolyzed Under the assumption that the metal oxide is changed to a crystalline metal oxide, the film thickness (layer thickness) of the metal oxide film derived from the metal alkoxide formed on the surface of the metal oxide fine particles is calculated.

Specifically, it is obtained by the following calculation formula (1).
r = ^{_{[{(d p CyM × 10 -3}} ) / (xd 0) +1} 1/3 -1 ] _{× (r 0/2) ···} (1)
Here, r [nm] is the layer thickness, x [g] is the powder amount of the metal oxide fine particles, d _p [gcm ^-3 ] is the true density of the metal oxide fine particles, and C [moll ^-1 ] is the metal alkoxide. Concentration, M [gmol ^-1 ] is the molecular weight of the metal oxide, y [ml] is the weight of the metal alkoxide, d ₀ [gcm ^-3 ] is the density of the metal oxide, r ₀ [nm] is the particle of the metal oxide fine particles Is the diameter.

Also, after the porous film, before the adsorption of the sensitizing dye, it intends row ultraviolet radiation to the porous membrane.
As a result , the organic content remaining in the paste can be removed, the amount of the sensitizing dye adsorbed to the porous film can be improved, and a dye-sensitized solar cell excellent in conversion efficiency can be obtained. it can.

Further, the ultraviolet irradiation, intends row in an oxygen atmosphere to generate ozone by introducing oxygen gas.
As a result , so-called UV ozone treatment is performed, and the organic component remaining in the paste is removed, so that more of the sensitizing dye can be adsorbed to the porous film, and the dye sensitization with higher conversion efficiency can be achieved. A sensitive solar cell can be obtained.

Also, after the porous film, before the adsorption of the sensitizing dye, it is preferable to dry at 80 to 200 ° C. the porous membrane (claim 3).
In this case, the organic content remaining in the paste is removed, the amount of the sensitizing dye adsorbed on the porous film is increased, and a dye-sensitized solar cell excellent in conversion efficiency can be obtained.
When the drying temperature is less than 80 ° C., it is difficult to sufficiently increase the conversion efficiency of the dye-sensitized solar cell. When the drying temperature exceeds 200 ° C., it is used as the conductive substrate. There is a possibility that the material that can be used is limited.

The metal oxide fine particles are preferably heat-treated at 300 to 500 ° C. before kneading (Claim 4 ).
In this case, a dye-sensitized solar cell excellent in conversion efficiency can be obtained by removing organic substances and moisture mixed in the metal oxide fine particles.
When the temperature of the heat treatment is less than 300 ° C., it is difficult to sufficiently remove organic substances and moisture, and it may be difficult to improve the conversion efficiency of the dye-sensitized solar cell. On the other hand, when the temperature of the heat treatment exceeds 500 ° C., the metal oxide fine particles are sintered and it may be difficult to form the porous film.

The metal oxide fine particles are preferably irradiated with ultraviolet rays before kneading (Claim 5 ).
In this case, the organic component contained in the metal oxide fine particles is removed to increase the amount of the sensitizing dye adsorbed on the porous film, thereby obtaining a dye-sensitized solar cell excellent in conversion efficiency. it can.

Also, the metal oxide fine particles is preferably a particle diameter 5 to 100 nm (claim 6).
In this case, the resulting porous membrane maintains sufficient porosity and has a large surface area, so that the amount of dye adsorbed can be increased and the electrolyte can be transported sufficiently.
When the particle size is less than 5 nm, the metal oxide fine particles may aggregate to reduce the surface area of the porous membrane, or the pores of the porous membrane may become too small to hinder electrolyte migration. is there. On the other hand, when the particle diameter exceeds 100 nm, the surface area of the porous film is decreased, and the amount of dye adsorbed may be insufficient.

A method for producing a dye-sensitized solar cell according to an example of the present invention will be described below.
In this production method, first, metal oxide fine particles and metal alkoxide for adhering the metal oxide fine particles are kneaded to obtain a paste.
Next, the paste is applied to a conductive substrate to hydrolyze the metal alkoxide.
Thereby, the porous film of the metal oxide fine particles is formed on the surface of the conductive substrate.
Next, a sensitizing dye is adsorbed on the porous film.
The photoelectrode of the said dye-sensitized solar cell is produced by the above.

Hereinafter, the method for producing the dye-sensitized solar cell will be described in detail.
[Raw material for forming porous film]
As the metal oxide fine particles, those having a particle diameter of about 5 to 100 nm are used, but the particle diameters are not necessarily uniform. Here, titanium oxide, zinc oxide and tin oxide were used.
The crystal form of titanium oxide may be any of rutile type, anatase type, and a mixture thereof. Here, P25 powder manufactured by Degussa with 30% rutile, 70% anatase and an average particle size of 25 nm was used.
As the zinc oxide, MZ-300 of Tayca having an average particle size of 35 nm was used. As the tin oxide, a product manufactured by Johnson Matthey having a primary particle size of 15 nm was used.
In addition, these can also be mixed and used suitably.

As the metal alkoxide that serves to adhere the metal oxide fine particles, various metal alkoxides that are stable in alcohol and easily hydrolyzed in the atmosphere can be used. Titanium (IV) tetraisopropoxide (TTIP) is used as the titanium material, zirconium (IV) tetra-n-propoxide is used as the zirconium material, and niobium (V) pentaethoxide (all manufactured by Aldrich) is used as the niobium material. It was.
These were appropriately diluted with alcohol. Any alcohol can be used as long as it dissolves them, but ethanol (manufactured by Nacalai Tesque) was used here. These metal alkoxide solutions can be stably stored for more than half a year if sealed and refrigerated.

[Pretreatment of metal oxide fine particle powder]
Depending on the preparation method and storage method, the surface of the metal oxide fine particles as a raw material is often contaminated with organic substances adsorbed from the atmosphere. Further, in many cases, moisture is adsorbed on the surface. When this is excessive, hydrolysis of the metal alkoxide may proceed at the stage of mixing with the metal alkoxide solution.
Therefore, the surface organic substances and moisture were removed by heat-treating the metal oxide fine particles in advance.

  In the case of titanium oxide fine particles, they were heated in an oven at 450 ° C. in the atmosphere for 30 minutes, but the temperature and time are appropriately selected according to the type of metal oxide and the size of the particles. Further, when the particle size is a nano size of 10 nm or less, the sintering may proceed even by heat treatment at a low temperature. In this case, the same effect can be obtained by decomposing organic matter by UV ozone treatment and vacuum drying.

[Measurement of water content in metal oxide fine particles]
In relation to the above, the amount of water contained in the metal oxide fine particles was measured. It was quantified by Karl Fischer titration of the weight loss in thermogravimetry and the amount of moisture that is desorbed when the powder is heated to 300 ° C. Titanium oxide (P25) powder stored in an environment of temperature 26 ° C. and humidity 72% was used for the measurement.

In thermogravimetric measurement, 18.27 g of the above titanium oxide powder (P25) was sampled, and the change in weight when the temperature was raised from room temperature to 500 ° C. at 10 ° C./min was recorded. The result is as shown in FIG.
In Karl Fischer titration, 0.1033 g of the above titanium oxide powder (P25) was sampled, sealed in a sealed moisture vaporizer, and the amount of moisture desorbed when heated to 300 ° C. was measured using a coulometric Karl Fischer titrator. It was measured.

[Preparation method of paste]
As a method of applying the paste to the conductive substrate, a doctor blade method, a screen printing method, a spray coating method, or the like can be used, and an appropriate paste viscosity is appropriately selected depending on the application method. Here, a method of applying simply with a glass rod (similar to the doctor blade method) was used. In this case, the concentration of the metal oxide fine particles that give an appropriate paste viscosity is generally in the range of 5 to 30 wt%.

The molar concentration ratio between the metal oxide fine particles and the metal alkoxide is appropriately determined according to the size of the metal oxide fine particles used so that the amorphous layer generated by hydrolysis of the metal alkoxide is not excessively thick and is sufficient for adhesion between the particles. Adjust. For example, when mixing titanium oxide (P25, particle size 25 nm) and TTIP, 3.55 g of 0.1M TTIP solution was mixed with 1 g of titanium oxide fine particles. At this time, the density | concentration in the paste of P25 microparticles | fine-particles will be about 22 wt%, and will become a viscosity suitable for application | coating. At this time, the weight ratio of P25, TTIP, and ethanol is 1: 0.127: 3.42 by weight, and 1: 0.0356 by mole.
5.92.

  In this composition, if all of TTIP in the paste is hydrolyzed and a uniform crystalline titanium oxide layer is formed on the surface of the P25 fine particles, the thickness of the P25 fine particles is considered to be a perfect sphere having a diameter of 25 nm. The thickness is about 0.14 nm. Since the titanium oxide crystal lattice size is about 0.3 to 0.9 nm (different between anatase and rutile), the formed titanium oxide is monolayer or less.

  However, in fact, as will be shown later, titanium oxide formed by hydrolysis of TTIP in room temperature atmosphere is amorphous. In addition, since the metal alkoxide is expected to be concentrated at the fine particle interface due to surface tension during the process of air-drying the paste, it is considered that the uniform distribution does not actually occur.

The layer thickness required for the calculation of the amorphous layer formed by hydrolysis of the metal alkoxide is a scale for determining an appropriate concentration ratio between the metal oxide fine particles and the metal alkoxide even when the metal oxide fine particle size used is different. Become.
Table 1 shows the estimated values of the paste composition and the layer thickness of the amorphous layer formed on the surface of the metal oxide fine particles by the decomposition of the metal alkoxide after air drying in this example, by the same calculation as the above assumption. (See equation (1) above).
In Table 1, three types of titanium oxide, zinc oxide, and tin oxide are used as the metal oxide fine particles to be mixed, and three types of Ti alkoxide, Zr alkoxide, and Nb alkoxide are used as the metal alkoxide. The values for the paste obtained by the combination of are listed.

All the metal alkoxides were used as 0.1 M ethanol solutions.
Further, “mixing ratio” in Table 1 means the weight ratio of the amount of metal alkoxide solution added to the metal oxide fine particles 1.
The “molar ratio of metal alkoxide” means the molar ratio of metal alkoxide to metal oxide fine particles, and the “weight ratio of metal alkoxide” means the weight ratio of metal alkoxide to metal oxide fine particles.

Further, “layer thickness” in Table 1 means that TiO ₂ , ZnO, and SnO _{2 of} metal oxide fine particles are spheres having diameters of 25, 35, and 15 nm, respectively, and hydrolysis, respectively, of Ti, Zr, and Nb alkoxides. Assuming that crystalline rutile TiO ₂ , ZrO ₂ , Nb ₂ O ₅ are produced, and using these bulk body densities of 3.90, 5.85, 4.50 g / cm ³ , these homogeneous When the film is formed on the surface of the metal oxide fine particles, it means the layer thickness estimated from the paste composition (see the above formula (1)).

  A paste was also prepared so that the fine particle concentration of the mixed paste of titanium oxide fine particles and alkoxide other than TTIP was 22 wt%. In the paste using zinc oxide and tin oxide fine particles, the content was 16 wt%. In this case, the metal alkoxide solution was mixed at a ratio of 5.25 g to 1 g of the metal oxide fine particles.

In this embodiment, the layer thickness of the amorphous metal oxide produced by the decomposition of the metal alkoxide is in the range of about 0.1 to 0.6 nm. A range of about 0.05 to 1.3 nm is appropriate for room temperature film formation by this method.
When the amount of the metal alkoxide added is smaller than this, the adhesion between the metal oxide fine particles and the adhesion of the porous film to the substrate are insufficient, the mechanical strength of the porous film is insufficient, and it is easy to peel off. When the amount of metal alkoxide added is larger than this, film formation is possible, but a thick amorphous layer is formed on the surface of the metal oxide fine particles, the film becomes a non-porous structure, and the charge transport property deteriorates as an electrode. Therefore, the film needs to be hydrothermally treated.

  As a method of mixing the metal fine particles and the metal alkoxide solution, there is a method of obtaining a uniform paste by stirring for 2 hours with a magnetic stirrer in a sealed container. In addition, it is possible to knead using a ball mill or an ultrasonic crusher, but it is important not to expose the metal alkoxide in the paste excessively for a long time so that hydrolysis of the metal alkoxide does not proceed excessively. .

[Application of paste on conductive substrate and air drying treatment]
Two scotch tapes that serve as spacers on a polyethylene terephthalate (PET) film substrate with tin-doped indium oxide (ITO) conductive film (20Ω / sq.) Or a glass substrate with F-doped SnO ₂ (FTO) conductive film (10Ω / sq.) Each paste prepared according to the above-described method was uniformly applied to a piece adhered in parallel at regular intervals using a glass rod.

  In the case of this method, the concentration of the metal oxide fine particles in the paste becomes an appropriate viscosity at about 15 to 25 wt%. However, when forming a film by spray coating or screen printing, the molar concentration of the metal oxide fine particles and the metal alkoxide If the ratio is in the above range, the viscosity of the paste may be adjusted appropriately.

The film after application to the conductive substrate is air-dried in the atmosphere at room temperature for about 2 minutes. In this process, the metal alkoxide in the paste is hydrolyzed by moisture in the atmosphere, and amorphous titanium oxide, zirconium oxide, and niobium oxide are formed from Ti alkoxide, Zr alkoxide, and Nb alkoxide, respectively.
Since the produced amorphous metal oxide plays a role of adhering metal oxide fine particles to each other and the film and the substrate, a porous film excellent in mechanical strength and adhesion can be obtained only by air drying.

[UV ozone treatment]
The porous film obtained by the above process can be used as it is as a photoelectrode for a solar cell, but its performance is further improved by removing the organic components remaining in the film with a UV ozone cleaner.
NL-UV253 UV ozone cleaner manufactured by Nippon Laser Electronics was used for UV ozone treatment. The UV light source is equipped with three 4.5 W mercury lamps with emission lines at 185 nm and 254 nm, and the sample is placed horizontally at a distance of about 6.5 cm from the light source. Ozone is generated by introducing an oxygen stream into the chamber. In this example, this UV ozone treatment was performed for 2 hours. In addition, the electroconductive fall of the ITO film | membrane by this UV ozone treatment was not seen at all.

[Product identification and film thickness measurement, surface morphology observation]
The crystal state of the product was examined with an X-ray diffractometer. The result is as shown in FIG.
Moreover, the film thickness of the porous film was measured with a stylus type surface roughness meter. The results are as shown in Table 3 to be described later.

[Sensitizing dye adsorption]
The most commonly used ruthenium complex, cis-bis (4,4'-dicarboxy-2,2'-bipyridine) bis (thiocyanato) ruthenium (II) (Kojima Chemical), is used as the sensitizing dye. A 5 mM ethanol solution was prepared. In this example, the porous film produced by the above process was dried in an oven at 100 ° C. for 1 hour, then immersed in a sensitizing dye solution, and allowed to stand overnight at room temperature to adsorb the sensitizing dye on the titanium oxide surface. . The sample after adsorption of the sensitizing dye was washed with ethanol and air-dried. This drying treatment is not always necessary depending on the sensitizing dye to be used, such as when the sensitizing dye is adsorbed from an aqueous solution.

[Solar cell prototype and battery characteristics evaluation]
A sandwich type solar cell is obtained by using a conductive substrate on which a porous film after adsorption of a sensitizing dye is formed as a photoelectrode, and an ITO / PET film or an FTO / glass counter electrode in which platinum fine particles are modified by sputtering. Prototype. The effective area of the photoelectrode was about 0.2 cm ² . 0.5 M for electrolyte solution
Using 3-methoxypropionitrile containing LiI, 0.05 M I2, 0.5 M t-buthylpyridine, it was introduced into the gap between both electrodes by capillary action.

Battery performance is evaluated using a fixed number of photons (10 ¹⁶
cm ^-2 ) Photocurrent action spectrum measurement under irradiation and IV measurement under AM1.5 simulated sunlight (100 mW cm ^-2 ) irradiation. A CEP-2000 type spectral sensitivity measuring device manufactured by Spectrometer Co., Ltd. was used for these measurements.
The results are as shown in FIGS. 9 and 10 described later.

[Evaluation of charge transport properties in porous membranes]
Light intensity is periodically in the range of 0.2 to 100 Hz by using a high-intensity green LED with an emission wavelength of 530 nm as a light source and modulating the applied voltage ± 8% sine wave using an S-5720C type frequency analyzer of NF electrons. The IMPS and IMVS plots were prepared by plotting the phase shift of the short-circuit photocurrent and open circuit voltage response of the battery and the amount of change synchronized with the frequency on the complex plane.
The result is as shown in FIG.
Here, IMPS stands for Intensity Modulated Photocurrent Spectroscopy (Intensity Modulated Photocurrent Spectroscopy).
Spectroscopy) and IMVS is the intensity modulated photovoltaic spectroscopy (Intensity Modulated)
Photovoltage Spectroscopy).

A Toho Giken 2000 type potentiostat was used to detect current and voltage. The light intensity was 2.6 mWcm ^-2 at maximum, and the light intensity was adjusted to various intensities with an ND filter. The frequencies (f _min ) giving the minimum values appearing in the third quadrant of the IMPS and IMVS plots are the electron transition time (τ _D ), the electron lifetime (τ _n ), τ _D = 1 / 2πf _min , and τ _n = These values can be determined because of the relationship of 1 / 2πf _min .

[Results obtained in Examples]
(1) Moisture content in titanium oxide powder and removal of attached organic matter It is known that many metal oxide surfaces adsorb water molecules, and fine particles with a large specific surface area contain a considerable amount of moisture. It is expected that Therefore, the moisture desorption temperature was estimated by thermogravimetry, and the amount of adsorbed moisture was examined by Karl Fischer titration.
FIG. 1 shows the measured thermogravimetric change curve. As can be seen from the figure, the weight gradually decreases as the temperature rises, and the weight change almost stops at around 300 ° C. Although weight reduction due to volatile components other than moisture is also conceivable, it is clear that the removal of moisture is completed at about 300 ° C. or less.

When the amount of moisture desorbed when the sample was heated at 300 ° C. was quantified by Karl Fischer titration, 0.253 mg of water was contained in 0.1033 g of P25 powder. That is, the titanium oxide powder contains about 2.5 wt% of water.
Although the influence of adsorbed water is not clear, when mixed with a metal alkoxide solution, hydrolysis of the metal alkoxide may proceed in the paste.
Therefore, the metal oxide fine particle powder was heat-treated in an oven at 450 ° C. for 30 minutes before mixing with the metal alkoxide, and after cooling, stored in a desiccator.

In addition, when the titanium oxide powder was heat-treated, it was observed that when heated at about 200 ° C., the powder colored brown, and when heated at a temperature of 300 ° C. or higher, it turned white again. This is considered to be based on the fact that the organic substance adsorbed on the titanium oxide surface is carbonized and remains when heated at about 200 ° C., but is burned and removed at a temperature of 300 ° C. or higher.
Since the amount of adsorbed organic molecules is extremely small, it was impossible to directly analyze them by XPS (X-ray photoelectron spectroscopy) or the like. However, if the metal oxide surface is contaminated with organic substances, It is conceivable that the adsorption of water is hindered and charge recombination is likely to occur.

(2) Relationship between Paste Concentration and Film Form It was found that the adjustment of the paste concentration is important for enabling room temperature synthesis of a titanium oxide porous thick film electrode. FIG. 2 (a) shows an electron of a film prepared by applying a paste having a high TTIP content described in D. Zhang, T. Yoshida and H. Minoura, Adv. Mater., 15, 814-817 (2003). A photomicrograph is shown, and FIG. 2 (b) shows an electron micrograph of the porous film produced with the paste concentration optimized this time.

In the conventional membrane shown in FIG. 2 (a), the molar ratio of P25: TTIP is 1: 0.356, whereas in the porous membrane of the present invention shown in FIG. 2 (b), P25: TTIP The molar ratio is 1: 0.0356.
Assuming that crystalline titanium oxide is uniformly formed on the surface of the P25 fine particles by hydrolysis of TTIP, the calculated layer thickness is 1.43 nm for the conventional film, and is 0. 0 for the porous film of the present invention. 143 nm (Table 1).

  In the conventional film (FIG. 2A), the fine particles derived from P25 are not isolated but have a non-porous structure in which they are covered with an amorphous substance. In contrast, in the porous membrane of the present invention (FIG. 2B), P25 particles having a particle size of around 25 nm are clearly confirmed, and the membrane has a porous structure.

As will be described later, amorphous titanium oxide is formed by hydrolysis of TTIP at room temperature. When the density of the amorphous titanium oxide to be generated is smaller than the bulk density of the rutile type titanium oxide crystal used in the calculation (the above formula (1)), and the amount of TTIP to be added is too much compared with the amount of the titanium oxide fine particles, A non-porous structure as shown in FIG. In this case, the function as a photoelectrode is low only by air drying. In this case, D. Zhang, T. Yoshida and H. Minoura, Adv. Mater., 15, 814-817 (2003)
It becomes necessary to hydrothermally treat the film by the method described in 1). When hydrothermal treatment is performed, amorphous titanium oxide changes to crystalline titanium oxide, and the film has a porous structure.

  In the case of the porous membrane of the present invention (FIG. 2B), it has already a porous structure and can be used as it is as a photoelectrode as will be shown later. However, if the amount of TTIP is too small, adhesion between the titanium oxide fine particles becomes insufficient, and a strong film cannot be formed.

(3) Production of Metal Oxide Composite Porous Membrane Composite Porous Membrane and Titanium Oxide Fine Particle Obtained from Paste Prepared by Mixing Titanium Oxide Fine Particles and Zirconium Alkoxide in FIGS. 3 (a) and 3 (b) Electron micrographs of a composite porous membrane obtained from a paste prepared by mixing bismuth and niobium alkoxide are shown. By maintaining the amount of metal alkoxide to be added within an appropriate range, a porous film could be formed in the same manner.

(4) X-ray diffraction measurement FIG. 4 shows an X-ray diffraction pattern of the obtained material. In order to examine the presence or absence of the generation of crystal phase, an ethanol solution of TTIP not containing metal oxide fine particles was applied to an FTO glass substrate and dried at room temperature (a), and this was subjected to UV ozone treatment for 2 hours (b) , P25 and TTIP mixed paste applied to the substrate and UV ozone treated (c) was measured.

  As shown in FIG. 4, the pattern a shows only the diffraction peak derived from the FTO substrate, and it can be seen that the substance produced by hydrolysis at room temperature is amorphous. It can be seen that the pattern b after the UV ozone treatment is not changed and crystallization does not occur. When this is heat-treated at 400 ° C. or higher, it changes to anatase-type titanium oxide, but formation of titanium oxide crystals that show X-ray diffraction is not observed at room temperature.

  The X-ray diffraction pattern c of the film obtained from the mixed paste of P25 and TTIP shows the diffraction peaks of the titanium oxide crystals belonging to the rutile type and the anatase type, which are derived from the P25 fine particles, The added TTIP appears to form amorphous titanium oxide. For other metal alkoxide solutions, an amorphous substance is formed without giving a crystalline metal oxide by hydrolysis at room temperature in the atmosphere.

(5) Changes in film surface morphology associated with UV ozone treatment FIGS. 5 (a), 5 (b) and 5 (c) show changes in the membrane surface morphology of the porous film before and after UV ozone treatment. .
An ITO-coated PET film is used for the conductive substrate. The paste composition is P25: TTIP = 1: 0.0356 (molar ratio).
The porous film (FIG. 5 (a)) that is only dried after the paste is applied has substantially the same form as FIG. 2 (b) described above, and has a porous structure in which individual P25 fine particles are observed. Even in the porous film after the UV ozone treatment (FIG. 5B), this does not substantially change, but the voids between the particles slightly increase and the porosity increases.

  According to the same low-magnification image of the porous film (FIG. 5C), it can be seen that the film is uniform as a whole, and no cracks or the like are generated with the UV ozone treatment. For this reason, the said porous film is excellent in the mechanical strength, and is closely_contact | adhered to the electroconductive board | substrate. Moreover, the change of the film thickness by UV ozone treatment was not seen.

(6) Removal of organic components by UV ozone treatment As shown in FIGS. 6 and 7, changes in the porous film accompanying the UV ozone treatment were examined by X-ray photoelectron spectroscopy (XPS). P25 + TTIP mixed paste applied to a conductive substrate and dried at room temperature in the atmosphere (a), treated with UV ozone for 2 hours (b), and (a) heat treated at 450 ° C. in the atmosphere (c) Compared.

In the wide spectrum of FIG. 6, peaks of Ti, O, and C are confirmed for all samples.
In FIG. 7, when the C 1s spectra are compared, in (a), a clear shoulder appears in the vicinity of 283.5 eV in addition to the peak in the vicinity of the binding energy of 285 eV. The former is carbon derived from contamination in the measurement system, while the latter is hydrocarbon carbon derived from TTIP, indicating the presence of residual organic components that cannot be removed only by coating and drying. This shoulder is reduced by the UV ozone treatment (b), indicating that the UV ozone treatment is effective in removing organic components.
When the sample is heat-treated at 450 ° C. (c), it can be confirmed that this shoulder originates from the residual organic component contained in the sample because this shoulder disappears completely.

(7) Solar cell characteristics In accordance with the above method, a sandwich cell (dye-sensitized solar cell) using a porous titanium oxide thick film produced on an FTO / glass substrate and an ITO / PET film substrate as a photoelectrode was constructed (sample) 1-8). Various characteristics (IPCE (photoelectric conversion efficiency per incident monochromatic light), I _sc (short-circuit photocurrent value), V _oc (open circuit voltage), FF (fill factor, η (conversion efficiency)) Evaluated.
The produced dye-sensitized solar cell has a porous film thickness of about 10 μm and an effective electrode area of 0.2 cm ² . A CEP-2000 type AM1.5 solar simulator (light intensity: 100 mW cm ⁻² ) manufactured by Spectrometer Co., Ltd. was used as a light source.
The obtained output characteristic values are summarized in Table 2.

  In Table 2, the columns of “UV ozone”, “UV”, and “dry” indicate the presence / absence of UV ozone treatment, UV irradiation treatment, and drying treatment after formation of the porous film and before sensitizing dye adsorption, respectively. The processed one is “◯”, and the unprocessed one is “×”. The UV ozone treatment was performed for 2 hours. Further, the UV irradiation treatment was performed for 2 hours using the same UV light source, blocking oxygen, and performing UV irradiation only for 2 hours. Moreover, the drying process was performed for 1 hour in 100 degreeC oven just before pigment | dye adsorption | suction.

Sample 5 was a sample in which the pretreatment of the titanium oxide fine particles before preparing the paste was not performed, and the other samples (samples 1 to 4 and 6 to 8) were heat-treated in a 450 ° C. oven for 30 minutes. It is used after.
Sample 6 represents a sample using a paste having a high TTIP concentration (P25: TTIP molar ratio is 1: 0.356). The other samples (samples 1 to 5, 7, and 8) all use a paste of P25: TTIP = 1: 0.0356.

From the results shown in Table 2, the following can be understood.
(7-1) Effect of paste composition First, D. Zhang, T. Yoshida
and H. Minoura, Adv. Mater., 15, 814-817 (2003), coated with a paste containing a relatively large amount of TTIP, and prepared using a film prepared by applying UV ozone treatment thereto. The battery curve of the battery (Sample 6 in Table 2) is shown in FIG. From the figure, it can be seen that the short-circuit photocurrent is very low although it operates as a battery.
Here, those plotted with ★ represent values under light irradiation, and those represented with broken lines represent values under dark conditions.

In the case of Sample 6, the short-circuit photocurrent value _Isc was very small, and the conversion efficiency η was only about 0.4% (Table 2). When the same material is hydrothermally treated, the conversion efficiency is improved to 2.3% (D. Zhang, T. Yoshida and H. Minoura, Adv. Mater., 15, 814-817 (2003)). From this, it can be seen that hydrothermal treatment is essential when a paste having a high TTIP content is used. When the hydrothermal treatment is not performed, the amount of dye adsorbed is small because the porosity of the film is low as shown in FIG. 2A (sample 6 in Table 2).

  On the other hand, when the film was prepared with a paste having a low TTIP content, the conversion efficiency η reached 2.06% even when the dye was adsorbed immediately after air drying at room temperature for 2 minutes after coating (sample 4). (Table 2). Furthermore, when the dye adsorption was performed after drying in an oven at 100 ° C. for 1 hour before the dye adsorption as in Sample 3, the dye adsorption amount increased and the conversion efficiency was improved to 2.68%.

(7-2) Pretreatment of titanium oxide fine particles Except for the sample 5 shown in Table 2, the titanium oxide fine particles were heat-treated at 450 ° C. for 30 minutes prior to paste preparation to remove excess water and adsorb to the surface. When organic substances were removed but this pretreatment was not performed (sample 5), the battery performance was greatly reduced to a conversion efficiency of 1.62%. There was no particular problem in the preparation of the porous membrane itself, and the strength and adhesion of the membrane were good, but the dye adsorption amount was lower than others.

  There was no difference between the sample 5 and other samples in the porosity of the surface judged from the electron micrograph. This is probably because the adsorptive water and organic components remain on the surface, which reduces the adsorptivity of the dye. Since the battery performance was also improved by subjecting the titanium oxide fine particles to UV ozone treatment in advance, the effect of removing organic substances was particularly great.

(7-3) Effect of UV and UV ozone treatment When UV ozone treatment is performed after applying and drying the paste (Sample 1), the short-circuit photocurrent value (I _sc ) is greatly improved and the open circuit voltage is improved. (V _oc ) also improved slightly, and the conversion efficiency η was 4.00%.
Even when the supply of oxygen gas to the UV ozone cleaner is shut off and only the UV light irradiation process is performed (sample 2), the performance of the untreated sample 3 is improved, but the conversion efficiency η is about 3%. Therefore, it can be seen that the combined use of ozone is effective.

When ITO / PET film substrate is used as the conductive substrate (samples 7 and 8), the tendency is similar, and the conversion efficiency η is 2.55% when only dried at 100 ° C. after coating (sample 8). On the other hand, when UV ozone treatment was performed (sample 7), the conversion efficiency η was improved to 3.27%.
Changes in the battery curve and the photocurrent action spectrum with and without UV ozone treatment are shown in FIGS. 9 and 10, respectively. As can be seen from these figures, it can be seen that the UV ozone treatment mainly increases the photocurrent and improves the conversion efficiency.

FIG. 9 (A) shows a battery curve for a conductive substrate using FTO / glass. FIG. 9 (B) shows a battery curve for a conductive substrate using ITO / PET.
FIG. 10A shows a photocurrent action spectrum of a dye-sensitized solar cell using FTO / glass as a conductive substrate. FIG. 10B shows a photocurrent action spectrum of a dye-sensitized solar cell using ITO / PET as a conductive substrate. 9 and 10, a curve a is for a film prepared by applying a P25 + TTIP mixed paste and drying at room temperature, and a curve b is for a film subjected to UV ozone treatment for 2 hours.

  In this example, the time required for the preparation of the photoelectrode material excluding the dye adsorption treatment is 2 hours for the UV ozone treatment and 1 hour for the drying treatment, so that the time can be greatly shortened compared to the hydrothermal reaction requiring 12 hours. Is possible. By optimizing each reaction condition, further speeding up of the process can be expected. Furthermore, since a sample that has only been dried at room temperature has an output of about half that of the best case, the process can be further simplified and speeded up in balance with the battery performance.

(8) Solar cell characteristics using metal oxide composite porous thick film electrode Metal oxide composite porous film prepared from a mixed paste other than the above-mentioned titanium oxide fine particles and TTIP is also a photoelectrode of a dye-sensitized solar cell. Was confirmed to function as. For this comparison, pretreatment (heat treatment or UV ozone treatment) of metal oxide fine particles and UV ozone treatment after paste application were not performed.

  11 to 13 show the IV curves measured for each of the nine types of batteries under irradiation with AM1.5 simulated sunlight, and the evaluation values are summarized in Table 3. As metal oxide particles, those using titanium oxide fine particles are shown in FIG. 11, those using zinc oxide fine particles are shown in FIG. 12, and those using tin oxide fine particles are shown in FIG.

As can be seen from Table 3, the one prepared from a mixed paste of titanium oxide fine particles and TTIP is the best, but solar cells can also be produced using other metal oxide fine particles, metal alkoxides, and mixtures thereof. is there. Moreover, it is possible to improve these performances by optimizing conditions, UV ozone treatment, or the like.
As a light source, a YSS-150A type AM1.5 solar simulator (light intensity: 100 mW cm ⁻² ) manufactured by Yamashita Denso was used.

(9) Evaluation of charge transport characteristics In order to clarify the reason for the improvement of conversion efficiency by UV ozone treatment, the electron transition time and the electron lifetime were determined for the samples before and after UV ozone treatment by IMPS and IMVS measurements, respectively (FIG. 14). , FIG. 15).
The electron lifetime is a half-life in which electrons injected from the photoexcited dye into titanium oxide are extinguished by recombination with the dye oxidant or iodine in the electrolyte solution. The electron transition time is the time required for the electrons photogenerated in the film at a certain moment to diffuse in the film and half of them reach the FTO electrode.

FIG. 14 shows IMPS and IMVS plots in which the phase difference and intensity of the periodic response of the photocurrent at the time of short circuit and the photovoltaic power at the time of open circuit are plotted on the complex plane. In the figure, ○ and ● are IMVS complex plane plots of the dye-sensitized solar cell. Also, ☆ and ★ are IMPS complex plane plots. White (◯, ☆) is a film prepared only by applying and drying a paste, and black (●, ★) represents a UV ozone-treated film. The measurement frequency was 0.2 to 100 Hz, and the frequency at the minimum value of each plot is shown in the figure. The light intensity is 2.6 mWcm ⁻² . The substrate is FTO glass.
From FIG. 14, it can be seen that in the IMPS plot, the frequency at which the minimum value is obtained does not change much before and after the UV ozone treatment, whereas in the IMVS plot, it varies to the lower frequency side after the UV ozone treatment.

  The same measurement was performed for various light intensities, and the relationship between the electron lifetime and the electron transition time and the light intensity is summarized in FIG. In the figure, the electron transition time determined from IMPS measurement is represented by ☆ and ★, and the electron lifetime determined from IMVS measurement is represented by ○ and ●. The substrate is FTO glass. White (◯, ☆) indicates a film prepared only by applying and drying a paste, and black (●, ★) indicates a product obtained by UV ozone treatment.

  Both the electron lifetime and the electron transition time vary depending on the light intensity. Increasing the light intensity increases the electron concentration in the conduction band and shortens the electron lifetime. On the other hand, since the trapping order of electrons generated at the crystal grain boundaries is satisfied, the electrons move easily, and the electron transition time Is shortened with increasing light intensity.

As can be seen from FIG. 15, although the electron transition time is slightly shortened by the UV ozone treatment, it does not change so much. On the other hand, the electron lifetime is longer after processing. As a result, the difference between the electron lifetime and the electron transition time becomes larger in the sample after the UV ozone treatment, so that more electrons can reach the electrode and the current value increases.
In other words, UV ozone treatment does not contribute much to reducing the density of electron traps that are thought to occur mainly at the grain boundary, but it is interpreted that it has the effect of suppressing electron recombination and increasing electron lifetime. it can. Since the organic component remaining in the film acts as an electron recombination center, it seems to be an effect of removing this by UV ozone treatment.

The diagram showing the thermogravimetric measurement result of the titanium oxide powder in an Example. The electron micrograph which shows the change of the film | membrane surface form accompanying the difference in TTIP content in an Example (magnification 100,000 times). (A) A film obtained by applying a mixed paste of P25: TTIP = 1: 0.358 (molar ratio). (B) A film obtained by applying a mixed paste of P25: TTIP = 1: 0.0358 (molar ratio). Electron micrographs (magnification: 50,000 times) of (a) titanium oxide + zirconium oxide and (b) titanium oxide + niobium oxide composite thin film in Examples. The X-ray diffraction pattern in an Example. (A) A 0.1M TTIP ethanol solution coated on an FTO glass substrate and dried at room temperature. (B) The film of (a) treated with UV ozone for 2 hours. (C) A P25 + TTIP mixed paste applied, dried and then UV ozone treated for 2 hours. The dotted line is the diffraction pattern of the FTO glass substrate only. The electron micrograph of the film | membrane surface form before and behind UV ozone treatment in an Example. (A) A P25 + TTIP mixed paste was applied and dried in the air at room temperature (magnification 100,000 times). (B) The film of (a) treated with UV ozone for 2 hours (magnification 100,000 times). (C) Low magnification image of (b) (magnification 20,000 times). The XPS (wide spectrum) of the titanium oxide porous thick film in an Example. (A) A P25 + TTIP mixed paste applied and dried at room temperature. (B) The film of (a) treated with UV ozone for 2 hours. (C) A film obtained by heat treating the film of (a) at 450 ° C. for 30 minutes. The XPS (C 1s spectrum) of the titanium oxide porous thick film in an Example. (A) A P25 + TTIP mixed paste applied and dried at room temperature. (B) The film of (a) treated with UV ozone for 2 hours. (C) A film obtained by heat treating the film of (a) at 450 ° C. for 30 minutes. The solar cell output characteristic produced by the coating of the paste (P25: TTIP = 1: 0.358) of the high TTIP content in an Example. The solar cell characteristic of the prototyped cell in an Example. (1) uses FTO / glass for the conductive substrate, and (2) uses ITO / PET. The photocurrent action spectrum of the prototyped cell in an Example. (1) uses FTO / glass for the conductive substrate, and (2) uses ITO / PET. The IV curve of the solar cell using the metal oxide composite porous thick film prepared from the titanium oxide microparticles | fine-particles + each metal alkoxide mixed paste in an Example. The IV curve of the solar cell using the metal oxide composite porous thick film prepared from the zinc oxide microparticles | fine-particles + each metal alkoxide mixed paste in an Example. The IV curve of the solar cell using the metal oxide composite porous thick film prepared from the tin oxide fine particle + each metal alkoxide mixed paste in an Example. IMVS (*, *) and IMPS ((circle), ●) complex plane plot of the produced solar cell in an Example. The diagram which shows the relationship between the electron transition time (*, *) and electron lifetime ((circle), *)) which were determined from the IMPS measurement and the IMVS measurement, and light intensity in an Example.

Claims (6)

A metal oxide fine particle and a metal alkoxide for adhering the metal oxide fine particle are kneaded to form a paste, the paste is applied to a conductive substrate, and the metal alkoxide is hydrolyzed to thereby obtain the conductive property. Forming a porous film of the metal oxide fine particles on the surface of the substrate;
Next, the porous film is irradiated with ultraviolet rays in an oxygen atmosphere in which oxygen gas is introduced to generate ozone, so-called UV ozone treatment is performed,
Then, the manufacturing method of the dye-sensitized solar cell characterized by making a sensitizing dye adsorb | suck to this porous film.   2. The metal alkoxide according to claim 1, wherein the metal oxide film formed by hydrolysis of the metal alkoxide has a film thickness of 0.05 to 1 when the surface of the metal oxide fine particles is uniformly coated. A method for producing a dye-sensitized solar cell, which is mixed with the metal oxide fine particles so as to have a thickness of 3 nm. 3. The method for producing a dye-sensitized solar cell according to claim 1, wherein the porous film is dried at 80 to 200 ° C. after the formation of the porous film and before the adsorption of the sensitizing dye. The method for producing a dye-sensitized solar cell according to any one of claims 1 to 3 , wherein the metal oxide fine particles are subjected to heat treatment at 300 to 500 ° C before kneading. The method for producing a dye-sensitized solar cell according to any one of claims 1 to 3 , wherein the metal oxide fine particles are irradiated with ultraviolet rays before kneading. The method for producing a dye-sensitized solar cell according to any one of claims 1 to 5 , wherein the metal oxide fine particles have a particle size of 5 to 100 nm.

Images