JP7660869B2 - Packed bed and flow method using same - Google Patents

Description

The present invention relates to a packed bed and a flow method using the same.

As material functions such as catalytic reactions become more sophisticated, nanomaterials such as gold nanoparticle catalysts, metal organic frameworks (MOFs), and graphene oxide, as well as nanoparticles carrying peptides, proteins, enzymes, etc., have been developed.

Nanomaterials are often actually provided in the form of fine powder, and it is believed that nanometer-sized primary particles form micrometer-sized secondary particles through aggregation, etc.

When nanomaterials are used in catalytic reactions, etc., there is a method in which fine powders as catalyst particles are dispersed in a liquid or gas to carry out the reaction. For example, there is a method in which two types of fine powders as catalyst particles are dispersed in a liquid, stirred, and optically resolved as a batch method (Non-Patent Document 1, Non-Patent Document 2, etc.).

When nanomaterials are used in flow-type flow catalytic reactions, etc., fine powders are filled into a reaction tube as catalyst particles, and reaction fluids such as water and gas are passed through the reaction tube. However, filling a reaction tube with fine powders as catalyst particles narrows the passageway for the reaction fluid, resulting in large pressure losses (hereinafter also referred to as "pressure loss"). This makes it difficult to pass a sufficient amount of reaction fluid through the reaction tube.

Conventionally, as a method for reducing pressure loss, a method is known in which a reaction tube filled with fine powder as catalyst particles is mixed with quartz sand as a gap filler for the fine powder, and a reaction fluid is passed through the reaction tube (Non-Patent Document 3, Non-Patent Document 4, etc.). In addition, a method is known in which catalyst particles are supported on a monolith with a hierarchical porous structure and low pressure loss (Non-Patent Document 5). Furthermore, a method is known in which fine powder as catalyst particles is supported on macroporous particles with macropores in the micron range (Patent Document 1).

Special Publication No. 2006-506224

Shinji Kawanishi, Shinya Oki, Dhiman Kundu, and Shuji Akai, “Lipase/Oxovanadium Co-Catalyzed Dynamic Kinetic Resolution of Propargyl Alcohols: Competition between Racemization and Rearrangement”, Org. Lett. 2019, 21, pp 2978-2982. Koji Sugiyama, Yasuhiro Oki, Shinji Kawanishi, Katsuya Kato, Takashi Ikawa, Masahiro Egi and Shuji Akai, “Spatial effects of oxovanadium-immobilized mesoporous silica on racemization of alcohols and application in lipase-catalyzed dynamic kinetic resolution”, Catal. Sci. Technol., 2016, 6, pp. 5023-5030. T.F. Narbeshuber, A, Brait, K. Seshan, and J.A. Lercher, "The influence of extraframework aluminum on H-FAU catalyzed cracking of light alkanes", Applied Catalysis A: General, 1996, 146 (1), pp 119-129. Vincenzo Palma, Marco Martino, Domenico Pisano, and Paolo Ciambelli "Catalytic Activities of Bimetallic Catalysts for Low Temperature Water Gas Shift Reaction", Chemical Engineering Transactions, 2016, 52, pp 481-486 Ping He, Stephen J. Haswell, Paul D. I. Fletcher, Stephen M. Kelly and Andrew Mansfield, “Scaling up of continuous-flow, microwave-assisted, organic reactions by varying the size of Pd-functionalized catalytic monoliths” Beilstein J. Org. Chem. 2011, 7, 1150-1157.

However, in the flow method using a packed column in which fine powder catalyst particles and gap filler are uniformly mixed, the catalyst particles made of fine powder are not fixed to the gap filler, and when a fluid (liquid, gas, etc.) is passed through, the fine powder catalyst particles are carried by the fluid and flow out of the column outlet. Therefore, it is difficult to uniformly fill the reaction tube with the mixture of fine powder catalyst particles and gap filler, resulting in the problem that catalytic activity is not exerted.

In addition, the fine powder that flows out becomes concentrated and clogs the outlet of the packed column, causing an increase in pressure, which creates the problem of not being able to fully reduce pressure loss.

Furthermore, if an organic polymer, inorganic binder, or the like is used as an adhesive to fix the fine powder as catalyst particles to the gap filling material, the surface of the fine powder as catalyst particles becomes buried in the adhesive, resulting in the problem that catalytic activity is not exerted.

The present invention has been made in consideration of the above circumstances. That is, the object of the present invention is to provide a packed bed that can suppress the outflow of fine powder without using an adhesive and maintain catalytic activity, and a flow method using the same.

As a result of intensive research conducted by the inventors to achieve the above object, they discovered that the above object can be achieved by using a packed bed filled with a mixture of porous particles having through holes, the size and particle size of the through holes being controlled within an appropriate range, and a fine powder having an average primary particle size controlled within an appropriate range, and thus completed the present invention.

The present invention provides the inventions described in the following sections.
Item 1.
A packed bed formed of a mixture of porous particles and fine powder,
(1) The porous particle has a skeleton, the skeleton has through holes, the through holes have a diameter of 0.1 to 100 μm, the through holes have a most frequent pore size of 0.5 μm or more and 50 μm or less in a pore size distribution, and the porous particle has a particle size of 20 μm or more and 3000 μm or less,
(2) the average primary particle size of the fine powder is 1 nm or more and is 80% or less of the most frequent pore size in the pore size distribution of the through holes;
(3) The bulk density of the porous particles is 0.05 to 7.5 (mg/mm ³ );
A packed bed characterized by:
Item 2 .
Item 2. The packed bed according to item 1 , wherein the fine powder has a bulk density of 0.02 to 7.5 (mg/mm ³ ).
Item 3 .
Item 3. The packed bed according to item 1 or 2 , wherein the packed bed has a bulk density of 0.05 to 7.5 (mg/mm ³ ).
Item 4 .
The filling bed according to any one of items 1 to 3 ,
A packed bed, in which a mixing ratio of the porous particles and the fine powder in the packed bed changes stepwise.
Item 5 .
The packed bed according to any one of Items 1 to 4 , wherein the mixture further contains enzyme particles.
Item 6 .
Item 6. The packed bed according to item 5 , wherein the enzyme particles are lipase.
Item 7 .
Item 5 or 6, the packed bed according to item 5 or 6 ,
A packed bed, in which a mixing ratio of the fine powder and the enzyme particles in the packed bed changes stepwise.
Item 8 .
A method for passing a gas or liquid through a pipe provided with the packed bed according to any one of items 1 to 7 .

The packed bed of the present invention is formed from a mixture of porous particles having through holes, the size and particle size of the through holes being controlled within an appropriate range, and fine powder having an average primary particle size controlled within an appropriate range, so that the outflow of the fine powder can be suppressed without using an adhesive, and therefore catalytic activity can be maintained.

1A is a scanning electron microscope (SEM) photograph of particle 2 obtained in Example 1. FIG. 1B is a graph showing the through hole diameter distribution of particle 2 obtained in Example 1, in which the horizontal axis indicates the through hole diameter (μm) and the vertical axis indicates the frequency (number) of the analyzed through holes. 1 is a graph showing the pore size distribution of particles 2 obtained in Example 1, in which the horizontal axis represents pore size (nm) and the vertical axis represents log differential pore volume (cm ³ /g). 1A is an SEM photograph of particle 2 obtained in Example 3. FIG. 1B is a through hole diameter distribution graph of particle 2 obtained in Example 3, in which the horizontal axis indicates the through hole diameter (μm) and the vertical axis indicates the frequency (number) of the analyzed through holes. 1 is a graph showing the measurement results of pressure loss when He gas was passed through the quartz tubes equipped with the packed beds obtained in Examples 1 and 2 and Comparative Examples 1 to 3. The horizontal axis shows the He gas flow rate (mL/min), and the vertical axis shows the pressure loss (kPa). 1 is a graph showing the measurement results of pressure loss when He gas was passed through the quartz tubes equipped with the packed bed obtained in Example 3 and Comparative Example 4. The horizontal axis shows the He gas flow rate (mL/min), and the vertical axis shows the pressure loss (kPa). FIG. 2 is a diagram showing the first region, the second region, and the third region in the catalytic reaction columns equipped with the packed beds obtained in Examples 15 to 24.

In this specification, a numerical range indicated using "~" indicates a range that includes the numerical values before and after "~" as the minimum and maximum values, respectively.

In this specification, the term "through holes" refers to holes that are also called macropores or macropores, and are submicron to micron sized, typically with a diameter of 0.1 to 100 μm.

In this specification, "pores" are also called mesopores, and typically have a diameter of 1 to 200 nm and are smaller than through-holes.

In this specification, the terms "integral porous body" and "monolith-type porous body" are used interchangeably and refer to a porous body having a characteristic structure in which one of the three sides of the structure is 1 mm or longer and a mesh-like skeleton of the order of micrometers is connected. Etymologically, it means a single piece of stone, and it looks like a single piece of chalk, and when observed under an electron microscope, it has a structure in which a jungle gym-like skeleton is connected. There are countless micron-scale pores called through holes that surround the gaps in the skeleton, and there may also be nano-scale pores within the skeleton. The through holes and pores are connected without being blocked, and it is said to boast a high porosity, with pores typically accounting for about 85% of the total volume. Due to the presence of nano-scale pores, the porous body has a high specific surface area.

The porous particles used in the present invention are crushed particles obtained by crushing and classifying a monolithic porous body, and have a high porosity and specific surface area.

In this specification, the "surface of the skeleton" refers to the surface of the skeleton exposed toward the through holes. When a porous particle has multiple pores in addition to the through holes, the "surface of the skeleton" refers to the surface of the skeleton exposed toward the through holes, and does not include the inner wall surfaces of the pores. When a porous particle has multiple pores in addition to the through holes, the total surface of the skeleton, which is the combination of the "surface of the skeleton" and the inner wall surfaces of the pores, means the "surface of the porous particle."

In this specification, Me in the structural formula means a methyl group, and Ac means an acetyl group.

The present invention includes the following embodiments. Preferred embodiments of the present invention are described in detail below. The components described below may be explained based on representative embodiments and specific examples, but the present invention is not limited to such embodiments.

1. Packing layer
The present invention relates to a packed bed formed of a mixture of porous particles and fine powder. That is, the present invention relates to a packed bed packed with a mixture of porous particles and fine powder. The packed bed of the present invention may be simply referred to as "the present invention".

In the present invention, (1) the porous particle has a skeleton, the skeleton has through holes, the most frequent pore size in the pore size distribution of the through holes is 0.1 μm or more and 100 μm or less, and the particle size of the porous particle is 20 μm or more and 3000 μm or less. That is, in the present invention, (1) the porous particle has a skeleton having through holes, the most frequent pore size in the pore size distribution of the through holes is 0.1 μm or more and 100 μm or less, and the particle size of the porous particle is 20 μm or more and 3000 μm or less.

In the present invention, (2) the average primary particle size of the fine powder is 1 nm or more, and the average primary particle size of the fine powder is 80% or less of the most frequent pore size in the pore size distribution of the through holes.

By satisfying the above-mentioned requirements, the present invention allows the fine powder to disperse through the through holes inside the porous particles, increasing the adhesion of the fine powder to the porous particles. Therefore, it is possible to prevent the outflow of the fine powder from the packed layer without using an adhesive.

In the present invention, the average primary particle size of the fine powder is 1 nm or more, and is 80% or less of the most frequent pore size in the pore size distribution of the through holes of the porous particles, so it is believed that the fine powder can enter the through holes of the porous particles.

The present invention also includes an embodiment in which the aggregates of the fine powder form coarse particles, and the coarse particles are caught in the gaps between the projections and recesses of the through-holes of the porous particles. In this case, electrostatic forces or the like act between the gaps between the projections and recesses of the through-holes of the porous particles and the coarse particles, so that the coarse particles that are larger than the through-holes of the porous particles do not flow out of the packed bed.

The present invention relates to a packed bed formed of a mixture of porous particles and fine powder,
(1) The porous particle has a skeleton, through holes formed in gaps in the skeleton, and a plurality of pores extending from the surface of the skeleton toward the inside,
the most frequent pore size in the pore size distribution of the through holes is 0.1 μm or more and 100 μm or less, and the particle size of the porous particles is 20 μm or more and 3000 μm or less,
the most frequent pore size in the pore size distribution of the pores is 1 nm or more and 200 nm or less, and is 20% or less of the most frequent pore size in the pore size distribution of the through holes;
(2) It is preferable that the average primary particle diameter of the fine powder is 1 nm or more and is 80% or less of the most frequent pore diameter in the pore diameter distribution of the through holes of the porous particles.

By satisfying the above-mentioned requirements, the present invention is able to suppress the outflow of fine powder, maintain catalytic activity, and achieve an excellent effect of reducing pressure loss. In addition, the excellent effect of reducing pressure loss is achieved because the filling of the fine powder near the porous particles is disturbed, and therefore the fluid flows effectively through the through holes without clogging the through holes with fine powder.

In the present invention, the most frequent pore size (μm) of the pore size distribution of the through holes of the porous particles is 0.1 μm or more and 100 μm or less. If the most frequent pore size of the pore size distribution of the through holes is less than 0.1 μm, when the fine powder diffuses into the through holes, the electrostatic interaction between the through hole surface and the fine powder is strengthened, causing the fine powder to aggregate at the through hole entrance and cause clogging. Therefore, no fluid flows through the through holes, and pressure loss cannot be reduced, making this impractical. If the most frequent pore size of the pore size distribution of the through holes exceeds 100 μm, it is difficult to manufacture the porous particles themselves.

The most frequent pore size (μm) of the pore size distribution of the through holes of the porous particles is preferably 0.5 μm or more and 50 μm or less, more preferably 0.75 μm or more and 20 μm or less, and even more preferably 1 μm or more and 10 μm or less.

The most frequent pore size (nm) of the pore size distribution of the porous particles is preferably 1 nm or more and 200 nm or less, more preferably 1.5 nm or more and 150 nm or less, even more preferably 2 nm or more and 100 nm or less, and particularly preferably 5 nm or more and 20 nm or less.

The most frequent pore size (nm) of the pore size distribution of the pores in the porous particles is preferably 20% or less, more preferably 15% or less, and even more preferably 10% or less of the most frequent pore size (μm) of the through holes in the porous particles.

The most frequent pore size (nm) of the pore size distribution of the pores in the porous particles is preferably 0.01% or more, more preferably 0.05% or more, and even more preferably 0.1% or more of the most frequent pore size (μm) of the through holes in the porous particles.

The most frequent pore size (nm) of the pore size distribution of the pores of the porous particles is 1.5 nm or more and 150 nm or less, and more preferably 0.01% or more and 15% or less of the most frequent pore size (μm) of the through holes of the porous particles. The most frequent pore size (nm) of the pore size distribution of the pores of the porous particles is 2 nm or more and 100 nm or less, and more preferably 0.05% or more and 10% or less of the most frequent pore size (μm) of the through holes of the porous particles. The most frequent pore size (nm) of the pore size distribution of the pores of the porous particles is 5 nm or more and 20 nm or less, and particularly preferably 0.1% or more and 5% or less of the most frequent pore size (μm) of the through holes of the porous particles.

(Method of generating a pore size distribution of through holes in a porous particle)
In the present invention, the most frequent pore size in the pore size distribution of the through holes in the porous particles means the mode value of the pore size distribution of the through holes observed by SEM. Specifically, the most frequent pore size distribution of the through holes observed by SEM is calculated according to the following procedure. First, the surface of the porous particles is photographed so that 20 or more through holes are displayed in the image. From the SEM photograph of the photographed porous particles, the circle equivalent diameters of 20 through holes are measured, and the integer value rounded off is taken as the through hole diameter. Then, a through hole diameter-number graph is created, and the diameter of the through hole with the largest number is taken as the most frequent pore size. Note that, since the overall through hole diameter distribution is constant, the calculation is performed under the assumption that the diameter of the through holes continuous within one porous particle is uniform throughout, including other particles.

In the present invention, the most frequent pore size in the pore size distribution of the porous particles means the most frequent value (mode value) of the pore size distribution calculated by the BJH method using the adsorption isotherm by the nitrogen adsorption method. Specifically, the pore size distribution of the porous particles of the present invention was analyzed by the BJH method by nitrogen adsorption measurement using a nitrogen adsorption/desorption device. The BJH method is a calculation method commonly used for pore analysis, and was proposed by Barrett, Joyner, Halenda, et al. (E.P. Barrett, L.G. Joyner and P. Halenda: J. Am. Chem. Soc., 73, 373 (1951)). As a specific measurement method, porous particles are sealed in a glass tube of a certain size, depressurized under vacuum, cooled to liquid nitrogen temperature, and nitrogen gas is gradually introduced and the pressure inside the glass tube containing the sample is measured, and this cycle is repeated until the nitrogen gas is adsorbed onto the sample to normal pressure. Next, the glass tube is gradually depressurized and the amount of nitrogen gas desorbed from the sample and the pressure inside the glass tube are measured, and this cycle is repeated until the relative pressure reaches 0.3 atmospheres. The resulting adsorption and desorption amounts of nitrogen gas, as well as the adsorption and desorption isotherms versus relative pressure, are then measured, and the pore size distribution of the pores in the porous particles is calculated using the BJH method, assuming that the pores are cylindrical models.

In the present invention, the porous particles are produced by pulverizing and granulating a monolithic porous body having through-holes synthesized by the sol-gel method described in detail below. Since the particle sizes of the porous particles immediately after pulverization are a mixture of large and small, the particles are classified by sieving to obtain porous particles in the desired particle size range.

Next, a method for producing porous particles will be described. The method for producing porous particles can be broadly divided into a synthesis process of a monolithic porous body that serves as the raw material for the porous particles, and a subsequent granulation process. Examples of the monolithic porous body include a monolithic inorganic porous body made of silica gel or silica glass having through-holes.

First, we will explain the synthesis process by the sol-gel method for a monolithic porous body made of silica gel or silica glass with through-holes. The synthesis process is further divided into a sol preparation process, a gelation process, and a removal process.

In the sol preparation process, a silica precursor, which is the raw material for silica gel or silica glass, and a coexisting substance that induces sol-gel transition and phase separation in parallel are added to an acid or alkaline aqueous solution, and the mixture is stirred at a low temperature, for example 5°C or less, at which the sol-gel transition is unlikely to proceed, to cause a hydrolysis reaction and prepare a uniform precursor sol.

As the main component of the silica precursor, water glass (aqueous solution of sodium silicate) or an inorganic or organic silane compound can be used. Examples of inorganic silane compounds include tetraalkoxysilane compounds such as tetramethoxysilane, tetraethoxysilane, tetra-isopropoxysilane, tetra-n-butoxysilane, and tetra-t-butoxysilane. Examples of organic silane compounds include trialkoxysilane compounds such as trimethoxysilane, triethoxysilane, triisopropoxysilane, and triphenoxysilane having substituents such as methyl, ethyl, propyl, butyl, hexyl, octyl, decyl, hexadecyl, octadecyl, dodecyl, phenyl, vinyl, hydroxyl, ether, epoxy, aldehyde, carboxyl, ester, thionyl, thio, and amino; dialkoxysilane compounds such as methyldiethoxysilane, methyldimethoxysilane, ethyldiethoxysilane, and ethyldimethoxysilane; and monoalkoxysilane compounds such as dimethylethoxysilane and dimethylmethoxysilane. In addition, alkoxysilicates containing crosslinking reaction rate control group substitutes such as monoalkyl, dialkyl, and phenyltriethoxy, as well as oligomers such as disilane, which is a dimer, and trisilane, which is a trimer, are also considered as silica precursors. Various compounds of the above-mentioned hydrolyzable silanes are commercially available, making them easy and inexpensive to obtain, and it is also easy to control the sol-gel reaction that forms a three-dimensional crosslinked body consisting of silicon-oxygen bonds.

The acid or alkaline aqueous solution is an aqueous solution in which an acid or base that functions as a catalyst for promoting the hydrolysis reaction of the silica precursor is dissolved in a solvent. The solvent may be water or an alcohol such as methanol or ethanol, with water being preferred. Specific examples of the acid include acetic acid, hydrochloric acid, sulfuric acid, nitric acid, formic acid, oxalic acid, and citric acid. Specific examples of the base include amines such as sodium hydroxide, potassium hydroxide, aqueous ammonia, sodium carbonate, sodium bicarbonate, and trimethylammonium; ammonium hydroxides such as tert-butylammonium hydroxide; and alkali metal alkoxides such as sodium methoxide.

Specific examples of the above coexisting substances include block copolymers such as polyethylene oxide, polypropylene oxide, polyacrylic acid, and polyethylene oxide-polypropylene oxide block copolymers; cationic surfactants such as cetyltrimethylammonium chloride; anionic surfactants such as sodium dodecyl sulfate; and nonionic surfactants such as polyoxyethylene alkyl ethers.

In the gelation process, the precursor sol prepared in the sol preparation process is poured into a gelation vessel and gelled at a temperature at which the sol-gel transition is likely to proceed, for example, at about 40°C. Here, the precursor sol contains the above-mentioned coexisting substance that acts to induce the sol-gel transition and phase separation in parallel, so that spinodal decomposition is induced and a porous structure of a silica hydrogel (wet gel) phase and a solvent phase is gradually formed.

In the gelation process, even after the silica hydrogel layer is formed, the polycondensation reaction of the wet gel proceeds slowly, causing the gel to shrink. Therefore, as a post-gelation process, the porous structure of the silica hydrogel phase and the solvent phase formed in the pores of the sol container in the gelation process is immersed in a basic aqueous solution such as ammonia water and heat-treated in a pressurized container. This allows the hydrolysis reaction, polycondensation reaction, and dissolution-reprecipitation reaction of the silica hydrogel phase to proceed further, making it possible to make the skeletal structure of the silica hydrogel phase stronger. The "porous structure" referred to here means (1) a co-continuous structure of the silica hydrogel phase and the solvent phase (a mesh-like porous structure in which the skeleton forming the through holes has a three-dimensionally continuous space, also called a "three-dimensional continuous mesh structure"), or (2) a particle structure of the silica hydrogel phase. The post-gelation process may be performed as necessary. The heat treatment does not necessarily have to be performed in a pressurized container or a sealed container. Heat treatment may produce or volatilize ammonia and other components, so it is preferable to carry out the treatment in a sealed container or a pressure-resistant pressurized container.

The pore size formed in the skeleton of the silica hydrogel phase is enlarged by the progress of the dissolution and reprecipitation reaction of the silica fine particles. Furthermore, the dissolution and reprecipitation reaction can be repeated by hydrothermal treatment, making it possible to control the pore size to be further enlarged. The pore size can also be controlled by adding urea to the precursor sol in addition to the catalyst and coexisting substances. Urea is hydrolyzed at a temperature of 60°C or higher to generate ammonia, and the pore size of the pores formed in the skeleton of the wet gel synthesized in the gelation process is enlarged by the ammonia, so that the addition of urea makes it possible to control the pore size. On the other hand, the structure and pore size of the through holes can be controlled by adjusting the amount of water or silica precursor added to the precursor sol in the sol preparation process, or the composition and amount of the coexisting substance added.

In the subsequent removal process, the wet gel is washed and dried or dried only to remove the solvent phase containing additives, unreacted materials, etc. The space left after the removal of the solvent phase becomes a through hole. By washing, the surface tension caused by additives and unreacted materials remaining in the solvent phase during drying is eliminated, and distortion and cracking of the gel during drying can be suppressed. The cleaning liquid is preferably a liquid such as an organic solvent or an aqueous solution. A liquid in which an organic compound or an inorganic compound is dissolved can also be used. Furthermore, even if a solution with a pH different from the isoelectric point of the gel, such as an acid or alkali, is used as the cleaning liquid, the additives remaining in the gel can be easily removed. Specifically, acids such as hydrochloric acid, sulfuric acid, nitric acid, hydrofluoric acid, acetic acid, formic acid, carbonic acid, citric acid, phosphoric acid, etc.; bases such as sodium hydroxide, potassium hydroxide, ammonia, water-soluble amines, sodium carbonate, and sodium bicarbonate can be used. The wet gel may be dried naturally, or, in order to eliminate distortion or cracks that may occur when drying the wet gel, the solvent in the wet gel may be replaced with a low-surface tension solvent such as isopropanol, acetone, hexane, or hydrofluorocarbon, which has a lower surface tension than water, or it may be dried by freeze sublimation. It is also preferable to use supercritical drying, which is performed in a surface tension-free state after replacing the solvent in the wet gel with carbon dioxide in a supercritical state.

The resulting dried gel can then be sintered by firing to turn it into silica glass. Note that if the firing temperature is lower than the glass transition temperature of silica (approximately 1000°C), it will not turn into silica glass.

Through the above sol preparation process, gelation process, and removal process, a monolithic porous body made of silica gel or silica glass with a three-dimensional continuous network structure is obtained.

The granulation process is a process in which the monolith-type porous body obtained through the above-mentioned sol preparation process, gelation process, and removal process is crushed and granulated. The crushing process in the granulation process may be performed manually, or may use a mortar or the like, or may use a crushing device such as a ball mill. Furthermore, when the dried gel obtained in the above-mentioned removal process is sintered, the granulation process may be performed either before or after the sintering.

The granulated monolithic porous body after the granulation process is classified by sieving through sieves with mesh sizes of X μm and Y μm (where D0≦X<Y≦1000) to recover porous particles with a particle diameter Dp within the desired particle diameter range (D0 μm or more and 4000 μm or less). However, the lower limit D0 (μm) of the desired particle diameter range is the larger of 1 (μm) or 5 times the most frequent pore diameter (μm) of the pore size distribution of the through holes.

In the present invention, the pore size, through hole size, and particle size can be controlled independently as described above. The most frequent pore size (μm) in the pore size distribution of the through holes is specified as 5 times or more the most frequent pore size (nm) in the pore size distribution of the fine pores, and the particle size (μm) is specified as 5 times or more the most frequent pore size (μm) in the pore size distribution of the through holes. This is based on the fact that it has been empirically understood that in order for the skeleton of each particle of the porous particle to maintain a skeleton structure having through holes even after granulation, a dimensional ratio of at least about 5 times is required between the pore size and the through hole size, and between the through hole size and the particle size.

The porous particles used in each example are all porous particles of silica gel (also called "silica porous particles") produced by the above-mentioned production method, that is, through a synthesis process and a granulation process using the sol-gel method of a monolithic porous body.

More specifically, the silica monolith (manufactured by SNG Corporation, lot number: E087), which is a monolith-type porous body used in each example, was prepared as follows. 0.9 g of polyethylene glycol (molecular weight 100,000) as a coexisting substance was dissolved in 9 mL (milliliters, cm ³ ) of 1 mol/L nitric acid aqueous solution, 7 mL of tetraethoxysilane (TEOS, silica precursor) was added, and the solution was stirred to obtain a uniform solution, which was then gelled at 40° C. The gel was then immersed in 0.1 M ammonia water and heated at 100° C. for 24 hours in a closed container, and then sintered at 600° C. for 5 hours. The obtained silica monolith was crushed in a mortar and classified using a JIS standard sieve to have a particle size within a predetermined range for each example, to obtain silica porous particles. In each example, the through-hole diameter was controlled by increasing or decreasing the amount of polyethylene glycol added, and the pore diameter was controlled by adjusting the temperature and time of heating with 0.1 M ammonia water.

In the present invention, the skeletal body of the porous particle is preferably composed of an organic compound or an inorganic compound. In the present invention, from the viewpoint of improving the strength, heat resistance, and chemical resistance of the porous particle, it is more preferable that the skeletal body of the porous particle is composed of an inorganic compound.

When the skeleton of the porous particle is composed of an inorganic compound, examples of the inorganic compound include oxides and nitrides containing typical metal elements or transition metal elements such as silica (silica gel _or silica glass), titania ( _TiO2 ), alumina ( _Al2O3 ), zirconia ( _ZrO2 ), hafnia ( _HfO2 ), germanium oxide ( _GeO2 ), silicon nitride (SiN), alumina nitride (AlN), and gallium nitride (GaN); oxides and nitrides containing alkali metal elements and alkaline earth metal elements such as lithium (Li), sodium (Na), potassium (K), rubidium (Rb), cesium (Cs), beryllium (Be), magnesium (Mg), calcium (Ca), strontium (Sr), and barium (Ba); and complexes of these containing typical elements such as boron (B), carbon (C), phosphorus (P), and sulfur (P). Other inorganic compounds may also be organic-inorganic hybrid compounds such as silicone represented by the chemical formula (OSi(CH ₃ ) ₂ ) _n , and it is preferable that the inorganic compound constituting the skeleton of the porous particles is not decomposed by the reaction.

Among the above inorganic compounds, silica, titania, alumina, zirconia and hafnia are preferred from the viewpoints of chemical resistance, heat resistance, pressure resistance, etc., and silica and titania are more preferred.

In the present invention, the framework of the porous particles is more preferably silica and titania, and particularly preferably silica. That is, in the present invention, the porous particles are more preferably silica porous particles and titania porous particles, and particularly preferably silica porous particles.

When the skeleton of the porous particles is composed of an organic compound, examples of the organic compound include polystyrene, polyurethane, polyacrylic acid, polymethacrylic acid, polyacrylonitrile, polyacrylic, polystyrene-divinylbenzene copolymer, resorcinol-formaldehyde copolymer, graphite, and carbon materials made from carbon nanotubes.

Among the above organic compounds, polystyrene-divinylbenzene copolymers and polyacrylics are preferred from the viewpoints of improving the strength of the porous particles and surface processability.

In the present invention, the particle size of the porous particles is preferably 20 μm or more and 720 μm or less, more preferably 20 μm or more and 7500 μm or less, even more preferably 20 μm or more and 300 μm or less, and particularly preferably 20 μm or more and 100 μm or less, from the viewpoints of ease of packing into a powder-packed column, gas permeability, etc.

The particle size of the porous particles can be measured using an optical microscope, a scanning electron microscope, a sieve with openings standardized by ISO3310 or JIS Z8801-1, a particle size distribution meter, etc.

The bulk density of the porous particles is appropriately determined depending on the material and porosity of the porous particles, but from the viewpoint of ease of mixing with fine powder, it is preferably 0.05 to 7.5 (mg/mm ³ ), more preferably 0.1 to 6 (mg/mm ³ ), even more preferably 0.15 to 5 (mg/mm ³ ), and particularly preferably 0.2 to 3 (mg/mm ³ ).

When the skeleton of the porous particle is composed of silica (SiO ₂ ), the bulk density of the porous particle is preferably 0.05 to 1.65 (mg/mm ³ ), more preferably 0.1 to 1.65 (mg/mm ³ ), and even more preferably 0.15 to 1.65 (mg/mm ³ ). Here, the value 1.65 (mg/mm ³ ) is uniquely calculated as 2.2 × (1 - 0.25) = 1.65, since the true density of silica (SiO ₂ ) is 2.2 (mg/mm ³ ) and the porosity in closest packing is 25%.

When the skeleton of the porous particles is made of titania (TiO ₂ ), the bulk density of the porous particles is preferably 0.05 to 3.15 (mg/mm ³ ), more preferably 0.1 to 3.15 (mg/mm ³ ), and even more preferably 0.15 to 3.15 (mg/mm ³ ). Here, the value 3.15 (mg/mm ³ ) is calculated uniquely as 4.2×(1−0.25)=3.15, since the true density of titania (TiO ₂ ) is 4.2 (mg/mm ³ , rutile type) and the porosity in closest packing is 25%.

When the skeleton of the porous particles is made of hafnia (HfO ₂ ), the bulk density of the porous particles is preferably 0.05 to 7.26 (mg/mm ³ ), more preferably 0.1 to 7.26 (mg/mm ³ ), and even more preferably 0.15 to 7.26 (mg/mm ³ ). The value 7.26 (mg/mm ³ ) is calculated uniquely as 9.68×(1−0.25)=7.26, since the true density of hafnia (HfO ₂ ) is 9.68 (mg/mm ³ ) and the porosity in closest packing is 25%.

(Method of tapping porous particles)
The bulk density of porous particles means the hardened bulk density obtained after mechanically tapping a container (e.g., a quartz tube filled with quartz wool) containing the porous particles. The tapping method is performed in accordance with the method specified in the 16th Revised Japanese Pharmacopoeia, 3.01, Bulk Density and Tapped Density Measurement Method. The bulk density of porous particles is obtained by mechanically tapping a container containing the porous particles. Specifically, after measuring the initial mass of the porous particles placed in a container, the container is mechanically tapped and the mass is read until almost no mass change is observed. The value obtained by dividing the mass at this time by the volume of the container is defined as the bulk density of the porous particles.

The average primary particle size (nm) of the fine powder is 1 nm or more, and is 80% or less of the most frequent pore size (μm) of the pore size distribution of the through holes. If the average primary particle size of the fine powder is less than 1 nm, it is difficult to produce the fine powder.
When the average primary particle diameter of the fine powder exceeds 80% of the most frequent pore diameter in the pore diameter distribution of the through holes, the fine powder does not enter the through holes and the through holes are blocked, causing clogging of the fine powder in the through holes, preventing fluid from flowing in the through holes and making it impossible to reduce pressure loss. Also, when the primary particles of the fine powder aggregate to form secondary particles, aggregated particles, and agglomerated particles, there is no problem in using the above average primary particle diameter as the diameter of the secondary particles, aggregated particles, and agglomerated particles.

The average primary particle size of the fine powder is preferably 0.07% to 10% of the most frequent pore size in the pore size distribution of the through holes, and more preferably 0.1% to 1%.

The average primary particle size of the fine powder is preferably 3 nm or more and 75 nm or less, and more preferably 5 nm or more and 30 nm or less.

The average primary particle diameter of the fine powder is preferably 3 nm or more and 75 nm or less, and the average primary particle diameter of the fine powder is preferably 0.07% or more and 10% or less of the most frequent pore diameter in the pore diameter distribution of the through holes. It is more preferable that the average primary particle diameter of the fine powder is 5 nm or more and 30 nm or less, and the average primary particle diameter of the fine powder is more preferably 0.1% or more and 1% or less of the most frequent pore diameter in the pore diameter distribution of the through holes.

In the present invention, a commercially available fine powder is used, and the average primary particle diameter of the fine powder is the guaranteed value provided by the manufacturer. If the guaranteed value is not clear, a transmission electron microscope (TEM) is used to directly measure the distribution of the diameters of the primary particles of the fine powder. Specifically, an image is taken so that 20 or more fine powder particles are displayed in the image. From the transmission electron microscope photograph of the fine powder taken, the circular equivalent diameters of the 20 primary particles are measured, and the arithmetic average is taken as the diameter. The calculation is performed assuming that the overall particle diameter is uniform throughout.

Examples of fine powders include powders of nanoparticle metal oxides, nanocarbons, nanoparticle catalysts, MOFs, metal complexes, low molecular weight organic compounds, enzymes, proteins, peptides, nucleic acids, etc., or complexes of these.

Suitable examples of nanoparticle metal oxides include titanium oxide, silicon oxide (silica), aluminum oxide (alumina), zinc oxide, strontium titanate, iron oxide, cerium oxide, and zirconium oxide.

As nanocarbons, for example, graphene, graphene oxide, fullerene, carbon nanotubes, etc. are preferably used.

As a nanoparticle catalyst, for example, a catalyst in which the above-mentioned nanoparticle oxide is supported with nanoparticles of a precious metal such as gold, platinum, or palladium is preferably used.

As MOF powder, for example, MOF-5, MOF-177, HKUST-1, etc. are preferably used.

Metal complexes that are suitable for use include, for example, scandium, titanium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, zinc, molybdenum, tungsten, ruthenium, osmium, rhodium, iridium, palladium, platinum, silver, gold, cadmium, mercury, and lanthanides, with ammine, carbonyl, phosphine, and thiol complexes as ligands.

Suitable low molecular weight organic compounds include, for example, proline and its derivatives, MacMillan catalyst, asymmetric phase transfer catalyst, thiourea, and N-heterocyclic carbene.

Suitable enzymes include, for example, pepsin, trypsin, chymotrypsin, and catalase.

Suitable peptides include, for example, D-proline-tyrosine-phenylalanine, proline-D-proline-2-aminoisobutyric acid-tryptophan-tryptophan-polyleucine acid, etc.

As nucleic acids, for example, DNA, RNA, etc. are preferably used.

The bulk density of the fine powder is appropriately determined depending on the material and porosity of the fine powder, but is preferably 0.02 to 7.5 (mg/mm ³ ), more preferably 0.03 to 6 (mg/mm ³ ), even more preferably 0.04 to 4.5 (mg/mm ³ ), and particularly preferably 0.05 to 3 (mg/mm ³ ).

When the fine powder is titanium oxide (TiO ₂ ), the bulk density of the fine powder is preferably 0.02 to 3.15 (mg/mm ³ ), more preferably 0.03 to 3.15 (mg/mm ³ ), further preferably 0.04 to 3.15 (mg/mm ³ ), and particularly preferably 0.05 to 3.15 (mg/mm ³ ). Here, the value of 3.15 (mg/mm ³ ) is calculated uniquely as 4.2×(1−0.25)=3.15, since the true density of titanium oxide (TiO ₂ ) is 4.2 (mg/mm ³ , rutile type) and the void fraction in closest packing is 25%.

When the fine powder is silica (SiO ₂ ), the bulk density of the fine powder is preferably 0.02 to 1.65 (mg/mm ³ ), more preferably 0.03 to 1.65 (mg/mm ³ ), further preferably 0.04 to 1.65 (mg/mm ³ ), and particularly preferably 0.05 to 1.65 (mg/mm ³ ). Here, the value of 1.65 (mg/mm ³ ) is calculated uniquely as 2.2×(1−0.25)=1.65, since the true density of silica (SiO ₂ ) is 2.2 (mg/mm ³ ) and the porosity in closest packing is 25%.

When the fine powder is hafnia (HfO ₂ ), the bulk density of the fine powder is preferably 0.02 to 7.26 (mg/mm ³ ), more preferably 0.03 to 7.26 (mg/mm ³ ), further preferably 0.04 to 7.26 (mg/mm ³ ), and particularly preferably 0.05 to 7.26 (mg/mm ³ ). Here, the value of 7.26 (mg/mm ³ ) is calculated uniquely as 9.68×(1−0.25)=7.26, since the true density of hafnia (HfO ₂ ) is 9.68 (mg/mm ³ ) and the void fraction in closest packing is 25%.

(Method of tapping fine powder)
The bulk density of a fine powder means the hardened bulk density obtained after mechanically tapping a container (e.g., a quartz tube filled with quartz wool) containing the fine powder. The tapping method is performed in accordance with the method specified in the 16th Revised Japanese Pharmacopoeia, 3.01, Bulk Density and Tapped Density Measurement Method. The bulk density of a fine powder is obtained by mechanically tapping a container containing the fine powder. Specifically, after measuring the initial mass of the fine powder placed in a container, the container is mechanically tapped and the mass is read until almost no mass change is observed. The value obtained by dividing the mass at this time by the volume of the container is the bulk density of the fine powder.

In the present invention, since the fine powder enters the through-holes of the porous particles or forms a catch on the surface of the through-holes, it is believed that the porous particles and the fine powder can be mixed uniformly without separation when mixed. As a result, the filling of the fine powder near the porous particles is disturbed, and the fluid flows effectively through the through-holes without clogging the through-holes with the fine powder. Therefore, it is believed that when gas or liquid is passed through a pipe having a packed layer formed from a mixture of the porous particles and the fine powder, the pressure loss can be significantly reduced.

The filling layer of the present invention is formed from a mixture of porous particles and fine powder. The method for preparing the mixture by mixing the porous particles and fine powder can be a known method, such as a dry mixing method or a wet mixing method.

For dry mixing, any method, whether manual or mechanical, can be used, such as shaking, shaking, rotating, stirring, etc., as long as it does not destroy the porous structure of the porous particles.

If wet mixing is used, various similar mixing methods can be used after adding the solvent.

From the viewpoint of uniformly mixing the porous particles and the fine powder, the mixing ratio of the porous particles to the fine powder in the packed bed of the present invention is preferably 20/80 to 80/20 by weight, more preferably 25/75 to 75/25 by weight, even more preferably 30/70 to 70/30 by weight, and particularly preferably 35/65 to 65/35 by weight.

The bulk density of the packed bed of the present invention is uniquely determined depending on the mixed state of the porous particles and the fine powder. From the viewpoint of ensuring good contact of the fluid with the fine powder, the bulk density is preferably in the same range as that of the porous particles and the fine powder, i.e., 0.05 to 7.5 (mg/mm ³ ), more preferably 0.1 to 6 (mg/mm ³ ), even more preferably 0.15 to 3 (mg/mm ³ ), and particularly preferably 0.2 to 2 (mg/mm ³ ).

(Method of tapping a mixture of porous particles and fine powder)
The bulk density of the packed bed means the hardened bulk density obtained after mechanically tapping a container (e.g., a quartz tube filled with quartz wool) containing a mixture of porous particles and fine powder. The tapping method is performed in accordance with the method specified in the 16th Revised Japanese Pharmacopoeia, 3.01, Bulk Density and Tapped Density Measurement Method. The bulk density of the packed bed is obtained by mechanically tapping a container containing a mixture of porous particles and fine powder. Specifically, after measuring the initial mass of the mixture of porous particles and fine powder placed in the container, the container is mechanically tapped and the mass is read until almost no mass change is observed. The value obtained by dividing the mass at this time by the volume of the container is the bulk density of the packed bed.

The packed bed of the present invention is preferably used as a filler packed into a tube.

For example, quartz tubes, glass tubes, resin tubes, metal tubes, etc. can be used as tubes.

The packed bed of the present invention can be used, for example, by packing it into a column, as a catalytic reactor, photocatalytic reactor, adsorption column, chromatography column, etc.

The packed bed of the present invention is also widely used as a filter (e.g., exhaust gas purification filter, air purification filter, water purification filter, etc.).

The packed bed of the present invention can be used for gas-phase catalytic reactions, liquid-phase catalytic reactions, gas-phase photocatalytic reactions, liquid-phase photocatalytic reactions, adsorption separation, adsorption removal, etc.

In the present invention, the mixture in the packed bed preferably further contains enzyme particles. That is, the packed bed of the present invention is preferably formed from a mixture of porous particles, fine powder, and enzyme particles. By having such a configuration, the present invention (1) not only maintains catalytic activity, but also (2) enables dynamic optical resolution with high efficiency, high yield, and high optical purity.

In the present invention, it is preferable that the mixture ratio of the porous particles and the fine powder in the packed bed formed from a mixture of the porous particles and the fine powder changes stepwise.

In the present invention, it is preferable that the mixture ratio of the fine powder and the enzyme particles in the packed layer formed from a mixture of the porous particles, the fine powder, and the enzyme particles changes stepwise.

In the present invention, a packed bed formed of a mixture of porous particles and fine powder is divided into a first region, a second region, and a third region in this order from the upstream side to the downstream side in the flow direction of the liquid,
(1) the first region is filled with fine powder and porous particles such that a mixing ratio of the fine powder to the porous particles is [fine powder]/[porous particles]=1/1 to 1/17 (weight ratio);
(2) the second region is filled with fine powder and porous particles such that the mixing ratio of the fine powder to the porous particles is [fine powder]/[porous particles]=1/0.5 to 1/10 (weight ratio);
(3) It is preferable that the third region is filled with fine powder and porous particles such that the mixing ratio of the fine powder to the porous particles is [fine powder]/[porous particles]=1/0.25 to 1/10 (weight ratio).

In the present invention, a packed bed formed of a mixture of porous particles, fine powder, and enzyme particles is divided into a first region, a second region, and a third region from the upstream side to the downstream side in the flow direction of the liquid,
(1) The first region is filled with fine powder and enzyme particles such that the mixing ratio of the fine powder to the enzyme particles is [fine powder]/[enzyme particles]=1/3 to 1/34 (weight ratio);
(2) the second region is filled with fine powder and enzyme particles such that the mixing ratio of the fine powder to the enzyme particles is [fine powder]/[enzyme particles]=1/2 to 1/20 (weight ratio);
(3) It is preferable that the third region is filled with fine powder and enzyme particles so that the mixing ratio of the fine powder to the enzyme particles is [fine powder]/[enzyme particles] = 1/1.4 to 1/20 (weight ratio).

In the present invention, for example, the enzyme particles are preferably lipase.

The present invention relates to a packed bed formed of a mixture of porous particles, fine powder and lipase,
(1) The porous particle has a skeleton, through holes formed in gaps in the skeleton, and a plurality of pores extending from the surface of the skeleton toward the inside,
the most frequent pore size in the pore size distribution of the through holes is 0.1 μm or more and 100 μm or less, and the particle size of the porous particles is 20 μm or more and 3000 μm or less,
the most frequent pore size in the pore size distribution of the pores is 1 nm or more and 200 nm or less, and is 20% or less of the most frequent pore size in the pore size distribution of the through holes;
(2) It is preferable that the average primary particle diameter of the fine powder is 1 nm or more and is 80% or less of the most frequent pore diameter in the pore diameter distribution of the through holes of the porous particles.

2. Distribution method
The present invention is a method for passing a gas or liquid through a pipe provided with the above-mentioned packed bed.

Examples of gases include nitrogen, oxygen, helium, argon, hydrogen, carbon monoxide, carbon dioxide, air, gaseous compounds, and mixtures of these.

Examples of liquids include water, organic solvents, ionic liquids, and mixtures of both.

For example, natural water, purified water, distilled water, ion-exchanged water, pure water, tap water, river water, pond water, sewage water, seawater, etc. can be used as water. Among these waters, distilled water and ion-exchanged water are preferred.

As the organic solvent, a conventionally known organic solvent can be used. Examples of such organic solvents include alcohols, ketones, esters, hydrocarbons, and ethers. Examples of alcohols include methanol, ethanol, and isopropyl alcohol. Examples of ketones include acetone and methyl ethyl ketone. Examples of esters include ethyl acetate and isopropyl acetate. Examples of hydrocarbons include cyclohexane, n-hexane, and n-octane. Examples of ethers include diethyl ether and dimethyl ether. The organic solvent is not limited to these organic solvents. In addition, it may be used as a mixed solvent with a polar organic solvent that has an affinity with water. As a mixed solvent of water and an organic solvent, for example, a highly viscous organic substance may be used as an aqueous solution, and examples of such a mixed solvent include an aqueous glycerin solution and a 70% aqueous ethanol solution.

The amount of gas supplied when circulating gas through a tube equipped with a packed bed is defined as the space velocity (SV), and is preferably 100 to 100,000/hr. The SV here can be calculated by dividing the amount of gas supplied per hour by the void volume in the column.

The amount of liquid supplied when circulating the liquid through a tube equipped with a packed bed is defined as the contact time (CT), and is preferably 0.1 to 3600 seconds. The CT here can be calculated by (void volume in the column) ÷ (amount of liquid supplied per second), or (column length) ÷ (linear velocity of the liquid per second).

Applications of the packed bed of the present invention include catalytic columns.

The catalytic column is preferably a catalytic column for optical resolution.

The present invention will be specifically explained below with reference to examples and comparative examples. However, the present invention is not limited to the examples.

(Comparative Example 1)
A quartz tube (having claws for holding the quartz wool) with an outer diameter of 8 mm, an inner diameter of 6 mm, and a length of 340 mm was filled with 20 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) as a support. 50 mg of fine powder of titanium oxide (manufactured by Nippon Aerosil Co., Ltd., product name: AEROXIDE (registered trademark) TiO ₂ P25, average primary particle diameter by transmission electron microscope observation: 21 nm) was filled on the quartz wool as particles 1. Then, tapping was performed according to the above-mentioned fine powder tapping method. The tapping was performed by dropping the quartz tube filled with particles 1 from a height of 10 mm at 60 times/min 200 times, until the length of the packed bed did not change after each 200 times. As a result, a packed bed of particles 1 with a length of 15 mm was obtained by tapping.

The quartz tube with the packed bed was attached to a gas distribution device (product name: TP-5000, manufactured by Hemmy Slide Rule Co., Ltd. (formerly Okura Riken Co., Ltd.)) and He gas was circulated at a flow rate range of 0 to 100 mL/min in increments of 10 mL/min using a mass flow controller, while the gas pressure upstream of the packed bed was measured with a pressure gauge (electronic pressure transmitter PT3000, manufactured by Okura Electric Co., Ltd.). The measurement range of the pressure gauge was 0 to 980 kPa (gauge pressure, display resolution 1 V, accuracy ±0.25% F.S.). Since the quartz tube outlet was open to the atmosphere, the indication value (kPa) of the pressure gauge was defined as the pressure loss (kPa) across the packed bed. The pressure loss due to the quartz wool alone was ignored, as the indication value (kPa) of the pressure gauge remained 0 kPa even when 100 mL/min of He gas was circulated. When the He gas flow rate was increased from 0 mL/min to 100 mL/min in increments of 10 mL/min, the pressure loss increased to 18 kPa. Next, the He gas flow rate was decreased to 0 mL/min in increments of 10 mL/min, and the cycle of increasing and decreasing the He gas flow rate was repeated a total of three times, and the changes in pressure loss in the second and third cycles were consistent.

The change in pressure loss with respect to the He gas flow rate is shown in Figure 4. As shown in Figure 4, hysteresis was observed in which the pressure was higher when the He gas flow rate was decreased (dotted line) than when the He gas flow rate was increased (solid line). After the third cycle was completed, the length of the packed bed of particles 1 was measured to be 13 mm. This length of the packed bed was designated as L1. Furthermore, the bulk density of particles 1 was calculated to be 0.14 (mg/ ^mm3 ) from the following formula.
Bulk density of particle 1 (mg/mm ³ )=weight of particle 1 (mg)/[length of packed bed L1 (mm)×cross-sectional area of quartz tube (mm ² )]

In the following Examples 1 and 2 and Comparative Examples 2 and 3, the weight of the titanium oxide fine powder (particle 1) was adjusted to 50 mg, the bulk density was adjusted to 0.14 (mg/mm ³ ), and the length (L1) of the packed bed of particle 1 was adjusted to 13 mm in all cases.

Example 1
As in Comparative Example 1, 20 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) was packed as a support in a quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm). Next, silica monolith (manufactured by SNG Corporation, lot number: E087) was pulverized using a mortar to obtain silica porous particles as particles 2. Thereafter, the particle size of particles 2 was uniformed in the range of 500 to 710 μm by sieving using a stainless steel sieve (with two types of mesh openings, 510 μm and 710 μm) conforming to JIS Z8801-1. The surface of particles 2 with the uniform particle size was photographed using a scanning electron microscope (manufactured by JEOL, product name: JSM-6510). From the SEM photograph of the photographed particles 2 (FIG. 1(a)), a pore size distribution of the through holes of particles 2 was created according to the above-mentioned "Method for creating a pore size distribution of through holes in porous particles" (FIG. 1(b)). As shown in Figure 1 (b), the most frequent pore size in the pore size distribution of the through holes of particle 2 was 10 μm. For particles 2 with uniform particle size, a nitrogen adsorption/desorption device (manufactured by Japan Microtrack BEL, product name: BELSORP-mini II) was used to create a pore size distribution of the pores derived by the BJH method through nitrogen adsorption measurement (Figure 2). As shown in Figure 2, the most frequent pore size in the pore size distribution of particle 2 was 6 nm.

Next, as a preliminary step, the amount of particles 2 with a uniform particle size in the range of 500 to 710 μm was determined as follows. First, 200 mg of quartz wool was packed as a support in a quartz tube, as in Comparative Example 1. Then, particles 2 with a uniform particle size were packed on the quartz wool, and tapping was performed while adjusting the amount of particles 2 packed according to the above-mentioned tapping method for porous particles. The tapping was performed by dropping the quartz tube filled with particles 2 from a height of 10 mm at 60 times/min 200 times, and was performed until the length of the packed bed did not change every 200 times. As a result, a packed bed of particles 2 with a length of 13 mm was obtained by tapping. The length of this packed bed was designated as L2. Since the weight of particles 2 at this time was 72 mg, the amount of particles 2 used was determined to be 72 mg. Furthermore, the bulk density of particles 2 was calculated to be 0.20 (mg/mm ³ ) from the following calculation formula.
Bulk density of particles 2 (mg/mm ³ )=weight of particles 2 (mg)/[length of packed bed L2 (mm)×cross-sectional area of quartz tube (mm ² )]

50 mg of titanium oxide fine powder (manufactured by Nippon Aerosil Co., Ltd., product name: AEROXIDE (registered trademark) TiO ₂ P25, average primary particle size: 21 nm) as particle 1 and 72 mg of silica porous particles (most frequent pore size distribution of through holes: 10 μm, most frequent pore size distribution of fine pores: 6 nm, particle size: 500 to 710 μm) as particle 2 were placed in a 6 mL screw tube bottle (manufactured by AS ONE Corporation, No. 2) and the bottle was lidded and mixed by shaking well by hand to obtain a mixture of particles 1 and 2. Next, 20 mg of quartz wool was packed into a quartz tube, and the mixture of particles 1 and 2 was filled on the quartz wool, and tapping was performed according to the tapping method for the mixture of porous particles and fine powder described above. Tapping was performed by dropping a quartz tube filled with a mixture of particles 1 and 2 from a height of 10 mm at 60 times/min 200 times, until the length of the packed bed did not change after each 200 times. As a result, a packed bed consisting of a mixture of particles 1 and 2 and having a length of 22 mm was obtained by tapping. The length of this packed bed was designated as L. Thereafter, the quartz tube with the packed bed was attached to a gas flow device (manufactured by Henmi Slide Rule Co., Ltd., product name: TP-5000), and He gas was passed through the quartz tube under the same conditions as in Comparative Example 1.

As shown in Fig. 4, even with a He gas flow rate of 100 mL/min, the pressure loss was 6 kPa. Although the length of the packed bed was about twice as long as that of Comparative Example 1, the pressure loss was greatly reduced. In addition, there was no difference in pressure change between the first and second cycles of increasing and decreasing the He gas flow rate, and no hysteresis was observed between the increase in the He gas flow rate (solid line) and the decrease in the He gas flow rate (dotted line). After the second cycle was completed, the length of the packed bed was measured to be 22 mm, which was the same as the length of the packed bed before the He gas flow (L: 22 mm). Furthermore, the bulk density of the packed bed was calculated to be 0.20 (mg/mm ³ ) from the following formula.
Bulk density of packed bed (mg/mm ³ )=[weight of particle 1 (mg)+weight of particle 2 (mg)]/[length of packed bed L (mm)×cross-sectional area of quartz tube (mm ² )]

Example 2
Except for adjusting the particle size of the silica porous particles to 212 to 500 μm, 50 mg of titanium oxide fine powder (manufactured by Nippon Aerosil Co., Ltd., product name: AEROXIDE (registered trademark) TiO ₂ P25, average primary particle size: 21 nm) as particle 1 and 73 mg of silica porous particles (mode diameter in pore size distribution of through holes: 10 μm, mode diameter in pore size distribution of fine pores: 6 nm, particle size: 212 to 500 μm, bulk density: 0.20 mg/mm ³ ) as particle 2 were mixed in the same manner as in Example 1 to obtain a mixture of particles 1 and particles 2.

A quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm) was stuffed with 20 mg of quartz wool (fiber thickness 2-6 μm, manufactured by Tosoh Corporation), and a mixture of particles 1 and 2 was packed onto the quartz wool. A packed layer was obtained by tapping in the same manner as in Example 1, according to the tapping method for the mixture of porous particles and fine powder described above. The length (L) of the packed layer was measured to be 22 mm. The quartz tube with the packed layer was attached to a gas distribution device (manufactured by Henmi Slide Rule Co., Ltd., product name: TP-5000), and He gas was passed through it under the same conditions as in Comparative Example 1.

As shown in Fig. 4, the pressure loss was 3 kPa when the He gas flow rate was 100 mL/min, and it was confirmed that the pressure loss was lower than that of Example 1. The length of the packed bed after the pressure loss measurement was 22 mm, which was the same as the length of the packed bed before the He gas flow (L = 22 mm). Furthermore, the bulk density of the packed bed was calculated to be 0.20 (mg/ ^mm3 ) using the same calculation formula as in Example 1.

(Comparative Example 2)
Quartz sand having no through holes or pores (manufactured by FUJIFILM Wako Pure Chemical Industries, Ltd., product number: 172-00015, product particle size: 600 to 850 μm at least 50%) was pulverized using a mortar to obtain quartz sand as particles 2. Thereafter, the particle size of particles 2 was uniformed to the range of 500 to 710 μm by sieving using a JIS Z8801-1 stainless steel sieve (two types of mesh sizes: 510 μm and 710 μm).

Next, as a preliminary step, a quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm) packed with 200 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) was used, and tapping was performed so that the length (L2) of the packed bed was 13 mm in the same manner as in Example 1, thereby determining the amount of particles 2 with a uniform particle size to be used to be 508 mg. Furthermore, using the same calculation formula as in Example 1, the bulk density of particles 2 was calculated to be 1.38 (mg/mm ³ ).

50 mg of fine powder of titanium oxide (manufactured by Nippon Aerosil Co., Ltd., trade name: AEROXIDE (registered trademark) TiO ₂ P25, average primary particle diameter: 21 nm) as particle 1 and 508 mg of particle 2 with a uniform particle size in the range of 500 to 710 μm were mixed in the same manner as in Example 1 to obtain a mixture of particle 1 and particle 2. Next, 20 mg of quartz wool was packed into a quartz tube, and the mixture of particle 1 and particle 2 was filled on the quartz wool, and a packed layer was obtained by tapping in the same manner as in Example 1. The length (L) of the packed layer was measured to be 23 mm. Thereafter, the quartz tube with the packed layer was attached to a gas distribution device (manufactured by Henmi Slide Rule Co., Ltd., trade name: TP-5000), and He gas was passed through under the same conditions as in Comparative Example 1.

As shown in Fig. 4, the pressure loss was 13 kPa when the He gas flow rate was 100 mL/min. From this, it was confirmed that the pressure loss in Comparative Example 2 was lower than that in Comparative Example 1, and did not decrease to less than half of the pressure loss in Comparative Example 1. The length of the packed bed after the pressure loss measurement was 23 mm, which was the same as the length of the packed bed before the He gas flow (L = 23 mm). Furthermore, the bulk density of the packed bed was calculated to be 0.86 (mg/ ^mm3 ) using the same calculation formula as in Example 1.

(Comparative Example 3)
Silica gel having no through holes but fine pores (manufactured by Fuji Silysia Chemical Ltd., product name: CARiACT (registered trademark) Q-15, particle size: 500 to 710 μm) was pulverized using a mortar to obtain silica gel as particles 2. Thereafter, the particle size of particles 2 was uniformed to the range of 500 to 710 μm by sieving using a stainless steel sieve conforming to JIS Z8801-1 (two types of mesh sizes: 510 μm and 710 μm).

Next, as a preliminary step, a quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm) packed with 200 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) was used, and tapping was performed in the same manner as in Example 1 so that the length (L2) of the packed bed was 13 mm, thereby determining the amount of particles 2 with a uniform particle size to be used to be 133 mg. Furthermore, using the same calculation formula as in Example 1, the bulk density of particles 2 was calculated to be 0.36 (mg/mm ³ ).

50 mg of fine powder of titanium oxide (manufactured by Nippon Aerosil Co., Ltd., product name: aeroxide P25, average primary particle diameter: 20 nm) as particle 1 and 508 mg of particle 2 with a uniform particle size in the range of 500 to 710 μm were mixed in the same manner as in Example 1 to obtain a mixture of particle 1 and particle 2. Next, 20 mg of quartz wool was packed into a quartz tube, and the mixture of particle 1 and particle 2 was packed on the quartz wool, and a packed layer was obtained by tapping in the same manner as in Example 1. The length (L) of the packed layer was measured to be 23 mm. Then, the quartz tube with the packed layer was attached to a gas distribution device (manufactured by Hemmy Slide Rule Co., Ltd., product name: TP-5000), and He gas was distributed under the same conditions as in Comparative Example 1.

As shown in Fig. 4, the pressure loss was 11 kPa when the He gas flow rate was 100 mL/min. From this, it was confirmed that the pressure loss in Comparative Example 3 and that in Comparative Example 2 were comparable. It was also confirmed that the pressure loss in Comparative Example 3 did not decrease to less than half of the pressure loss in Comparative Example 1, in which only titanium oxide fine powder was used. The length of the packed bed after the pressure loss measurement was 23 mm, which was the same as the length of the packed bed before the He gas flow (L = 23 mm). Furthermore, the bulk density of the packed bed was calculated to be 0.28 (mg/ ^mm3 ) using the same calculation formula as in Example 1.

The relationship between particles 1, particles 2, and the packed layer in Examples 1-2 and Comparative Examples 1-3 is shown in Table 1 below.

From FIG. 4, it is believed that in the porous silica particles of Examples 1 and 2, the silica porous particles with through holes are mixed into the packed layer of titanium oxide fine powder, which causes a geometric effect that disrupts the packed state of the titanium oxide fine powder, resulting in large gaps in the interparticle voids of particle 2. More specifically, it is believed that the presence of through holes in the porous silica particles causes the titanium oxide fine powder to disperse into the through holes of the porous silica particles, and particle 1, which was present in the gaps between particle 2, moves into the gaps between particle 2, causing the gaps between particle 2 to become larger. It is presumed that for these reasons, Examples 1 and 2 showed a greater reduction in pressure loss than Comparative Examples 1 to 3.

(Comparative Example 4)
A quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm) was filled with 20 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) as a support. 50 mg of silica fine powder (manufactured by Nippon Aerosil Co., Ltd., product name: AEROSIL (registered trademark) 200, average primary particle diameter observed by transmission electron microscope: 12 nm) was filled on the quartz wool as particle 1. Thereafter, tapping was performed in the same manner as in Comparative Example 1 to obtain a packed bed with a length of 32 mm.

The quartz tube with the packed layer was attached to a gas flow device (manufactured by Henmi Slide Rule Co., Ltd., product name: TP-5000), and He gas was passed through it under the same conditions as in Comparative Example 1 to measure the pressure loss. The cycle of increasing the He gas flow rate and decreasing the He gas flow rate was repeated a total of three times, and the changes in pressure loss between the second and third cycles were consistent. The changes in pressure loss versus He gas flow rate at this time are shown in Figure 5.

As shown in Fig. 5, the pressure loss was 52 kPa when the He gas flow rate was 100 mL/min. As shown in Fig. 5, hysteresis was observed in which the pressure was higher when the He gas flow rate was decreased (dotted line) than when the He gas flow rate was increased (solid line).
After the third cycle, the length of the packed bed of the silica fine powder was measured and found to be 24 mm. This length of the packed bed was designated as L1. Furthermore, the bulk density of Particle 1 was calculated to be 0.07 (mg/ ^mm3 ) using the same calculation formula as in Comparative Example 1.

In the following Example 3, the weight of the silica fine powder (particles 1) was adjusted to 50 mg, the bulk density was adjusted to 0.07 (mg/mm ³ ), and the length (L1) of the packed bed of particles 1 was adjusted to 24 mm.

Example 3
As in Comparative Example 4, 20 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) was packed as a support into a quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm). Next, titania monolith (manufactured by SNG Corporation, lot number: T5071) was pulverized using a mortar to obtain titania porous particles as particles 2. Thereafter, the particle size of particles 2 was uniformed in the range of 125 to 300 μm by sieving using a stainless steel sieve (with two types of mesh openings, 125 μm and 300 μm) conforming to JIS Z8801-1. The surface of particles 2 with uniform particle size was imaged using a scanning electron microscope (manufactured by JEOL, product name: JSM-6510). From the captured SEM photograph of particle 2 ( FIG. 3( a )), a pore size distribution of the through holes of particle 2 was created ( FIG. 3( b )) according to the above-mentioned “Method for creating a pore size distribution of through holes in a porous particle”. As shown in FIG. 3( b ), the most frequent pore size in the pore size distribution of the through holes of particle 2 was 5 μm. For particles 2 with a uniform particle size, the most frequent pore size in the pore size distribution of the pores was 6 nm.

Next, as a preliminary step, the amount of particles 2 with a uniform particle size in the range of 125 to 300 μm was determined as follows. First, 200 mg of quartz wool was packed as a support in a quartz tube in the same manner as in Example 1. Then, particles 2 with a uniform particle size were packed on the quartz wool, and tapping was performed until the length of the packed bed did not change in the same manner as in Example 1. Tapping was repeated while adjusting the amount of particles 2 packed so that the length (L2) of the packed bed became 24 mm. Since the weight at this time was 430 mg, the amount of particles 2 used was determined to be 430 mg. Furthermore, the bulk density of particles 2 was calculated to be 0.63 (mg/mm ³ ) from the following formula.
Bulk density of particles 2 (mg/mm ³ )=weight of particles 2 (mg)/[length of packed bed L2 (mm)×cross-sectional area of quartz tube (mm ² )]

50 mg of silica fine powder (manufactured by Nippon Aerosil Co., Ltd., product name: AEROSIL (registered trademark) 200, average primary particle diameter observed by transmission electron microscope: 12 nm) as particle 1 and 430 mg of titania porous particles (most frequent pore diameter in through-hole diameter distribution: 5 μm, most frequent pore diameter in fine pore diameter distribution: 6 nm, particle size: 125-300 μm) as particle 2 were placed in a 6 mL screw tube bottle (manufactured by AS ONE Corporation, No. 2) and the bottle was shook well by hand to mix and obtain a mixture of particles 1 and 2. Next, 20 mg of quartz wool was packed into a quartz tube, and the mixture of particles 1 and 2 was filled on the quartz wool, and tapping was performed in the same manner as in Comparative Example 1 to obtain a packed layer. The length (L) of the packed layer was measured to be 46 mm. The quartz tube with the packed bed was then attached to a gas flow device (manufactured by Henmi Slide Rule Co., Ltd., product name: TP-5000), and He gas was passed through it under the same conditions as in Comparative Example 1.

As shown in Fig. 5, even at a flow rate of 100 mL/min, the pressure loss was 8 kPa, and the pressure loss was greatly reduced, even though the length of the packed bed was about twice that of the packed bed in Comparative Example 4. Furthermore, there was no difference in pressure change between the first and second cycles of increasing and decreasing the He gas flow rate, and no hysteresis was observed between the increase in the He gas flow rate (solid line) and the decrease in the He gas flow rate (dotted line). Furthermore, after the second cycle was completed, the length of the packed bed was measured to be 46 mm, which was the same as the length of the packed bed before the He gas flow (L: 46 mm). Furthermore, the bulk density of the packed bed was calculated to be 0.37 (mg/mm ³ ) from the following calculation formula.
Bulk density of packed bed (mg/mm ³ )=[weight of particle 1 (mg)+weight of particle 2 (mg)]/[length of packed bed L (mm)×cross-sectional area of quartz tube (mm ² )]

The relationship between particles 1, particles 2, and the packed layer in Example 3 and Comparative Example 4 is shown in Table 2 below.

(Comparative Example 5)
A quartz tube (outer diameter 8 mm, inner diameter 6 mm, length 340 mm) was packed with 20 mg of quartz wool (fiber thickness 2 to 6 μm, manufactured by Tosoh Corporation) as a support. 332 mg of titanium oxide fine powder (manufactured by Sigma-Aldrich, product name: P25, average primary particle size: 25 nm) was packed on the quartz wool as particles 1. Thereafter, tapping was performed in the same manner as in Comparative Example 1 to obtain a packed layer with a length of 13 mm.

Example 4
332 mg of titanium oxide fine powder (manufactured by Sigma-Aldrich, product name: P25, average primary particle size: 25 nm) as particle 1 and 60 mg of silica porous particles (most frequent pore size distribution of through holes: 10 μm, most frequent pore size distribution of fine pores: 6 nm, particle size: 1000 to 3000 μm, manufactured by SNG Corporation, lot number: E093) as particle 2 were placed in a 6 mL screw capped vial (manufactured by AS ONE Corporation, No. 2) and mixed by shaking well by hand to obtain a mixture of particles 1 and 2. Next, 20 mg of quartz wool was packed into a quartz tube, and the mixture of particles 1 and 2 was filled on the quartz wool, and tapping was performed according to the tapping method for the mixture of porous particles and fine powder described above. Tapping was performed by dropping a quartz tube filled with a mixture of particles 1 and 2 from a height of 10 mm at 60 times per minute 200 times until the length of the packed bed did not change after each 200 times. As a result, a packed bed consisting of a mixture of particles 1 and 2 with a length of 20 mm was obtained by tapping.

Example 5
A packed bed was prepared in the same manner as in Example 4, except that the particle size of the silica porous particles as the particles 2 was 710 to 1000 μm. As a result, a packed bed consisting of a mixture of the particles 1 and the particles 2 and having a length of 21 mm was obtained by tapping.

Example 6
A packed bed was prepared in the same manner as in Example 4, except that the particle size of the silica porous particles as the particles 2 was 212 to 710 μm. As a result, a packed bed consisting of a mixture of the particles 1 and the particles 2 and having a length of 20 mm was obtained by tapping.

(Example 7)
A packed bed was prepared in the same manner as in Example 4, except that the particle size of the silica porous particles as the particles 2 was 20 to 212 μm. As a result, a packed bed consisting of a mixture of the particles 1 and 2 and having a length of 19 mm was obtained by tapping.

A gas flow test was performed on the quartz tubes with the packed layers obtained in Comparative Example 5 and Examples 4 to 7 in the same manner as in Comparative Example 1. However, argon gas was used instead of helium gas, and the pressure was measured when the argon gas flow rate was 100 mL/min. The measurement was repeated until no change in pressure loss was observed. The length of the packed layers obtained in Comparative Example 5 and Examples 4 to 7 before and after the argon gas flow was unchanged and was the same.

The relationships between particle 1, particle 2, and the packed bed in Examples 4 to 7 and Comparative Example 5 are shown in Table 1 below. In addition, the bulk density of particle 1, the bulk density of particle 2, and the bulk density of the packed bed for Examples 4 to 7 and Comparative Example 5 were calculated using the above-mentioned formula, and the results are shown in Table 3.

The length of the packed bed in Comparative Example 5 is different from that in Examples 4 to 7, and the pressure loss is linearly proportional to the length of the packed bed. Therefore, the pressure loss normalized to a packed bed length of 10 mm was used to compare the pressure loss between Comparative Example 5 and Examples 4 to 7. As shown in Table 3, the pressure loss normalized to a packed bed length of 10 mm was 83 kPa in Comparative Example 5, 33 kPa in Example 4, 15 kPa in Example 5, 26 kPa in Example 6, and 64 kPa in Example 7. From this, it was confirmed that, even though the length of the packed bed in Examples 4 to 7 is about 1.5 times that of the packed bed in Comparative Example 5, the pressure loss can be reduced more than that in Comparative Example 5 by mixing Particles 1 and 2.

As shown in Table 3, the pressure loss was most reduced for Example 5 with a particle size of 710 to 1000 μm, followed by Example 6 with a particle size of 212 to 710 μm, Example 4 with a particle size of 1000 to 3000 μm, and Example 7 with a particle size of 20 to 212 μm. As a result, when the particle size range of the porous particles (particle 2) was 20 to 3000 μm, the pressure loss was reduced more than in Comparative Example 5.

(Example 8)
A packed bed was prepared in the same manner as in Example 4, except that 330 mg of a fine powder of titanium oxide colored with brilliant blue (manufactured by Sigma-Aldrich Corporation, product name: P25, average primary particle size: 25 nm) was used as particle 1, and 80 mg of a porous silica particle (mode diameter of through-hole size distribution: 0.1 μm, mode diameter of fine pore size distribution: 6 nm, particle size: 200 to 700 μm, manufactured by SNG Corporation, lot number: PM105) was used as particle 2.

(Example 9)
A packed bed was prepared in the same manner as in Example 4, except that 330 mg of a fine powder of titanium oxide colored with brilliant blue (manufactured by Sigma-Aldrich Corporation, product name: P25, average primary particle size: 25 nm) was used as particle 1, and 80 mg of a porous silica particle (mode pore size in the pore size distribution of through holes: 1 μm, mode pore size in the pore size distribution of fine pores: 6 nm, particle size: 200 to 700 μm, manufactured by SNG Corporation, lot number: PM123) was used as particle 2.

(Example 10)
A packed bed was prepared in the same manner as in Example 4, except that 330 mg of a fine powder of titanium oxide colored with brilliant blue (manufactured by Sigma-Aldrich Corporation, product name: P25, average primary particle size: 25 nm) was used as particle 1, and 80 mg of a porous silica particle (mode pore size in the pore size distribution of through holes: 20 μm, mode pore size in the pore size distribution of fine pores: 6 nm, particle size: 200 to 700 μm, manufactured by SNG Corporation, lot number: E040) was used as particle 2.

(Comparative Example 6)
A packed bed was prepared in the same manner as in Example 4, except that 330 mg of fine powder of titanium oxide colored with brilliant blue (manufactured by Sigma-Aldrich Corporation, product name: P25, average primary particle size: 25 nm) was used as particle 1, and 540 mg of sea sand having no through holes or pores (manufactured by Kishida Chemical Co., Ltd., particle size: 370 to 840 μm) was used as particle 2.

For the quartz tubes with packed layers obtained in Examples 8 to 10 and Comparative Example 6, gas flow tests were carried out in the same manner as in Comparative Example 1. However, argon gas was used instead of helium gas, and the pressure was measured when the argon gas flow rate was 100 mL/min. The measurements were repeated until no change in pressure loss was observed.

The relationship between particle 1, particle 2, and the packed bed in Examples 8 to 10 and Comparative Example 6 is shown in Table 4 below. In addition, the bulk density of particle 1, the bulk density of particle 2, and the bulk density of the packed bed for Examples 8 to 10 and Comparative Example 6 were calculated using the above-mentioned formula, and the results are shown in Table 4.

In Examples 8 to 10 and Comparative Example 6, before each test, the blue color of the titanium oxide fine particles colored with brilliant blue was uniformly dispersed in the packed bed. After the tests of Examples 8 to 10, the entire column was uniformly blue, confirming that the outflow of the titanium oxide fine powder was suppressed. On the other hand, after the test of Comparative Example 6, the blue color on the inlet side of the column had disappeared and become lighter, and blue fine particles had accumulated on the outlet side, so it was visually confirmed that the titanium oxide fine powder had moved due to the flow of argon gas. Furthermore, in Comparative Example 6, the pressure loss normalized to 10 mm increased from the initial value of 19 kPa to 167 kPa after 10 minutes of continuous measurement due to the gas flow, and it is believed that the pressure loss increased due to clogging on the outlet side as the fine powder flowed out. In other words, after the test of Comparative Example 6, it was confirmed that the outflow of the titanium oxide fine powder could not be suppressed.

Below is an example of an optical resolution column using the packed bed of the present invention. Specifically, an example of a catalytic column for dynamic optical resolution of racemized alcohol is shown.

In this dynamic optical resolution method, when a solution containing racemic alcohol is passed through a column packed with a solid phase consisting of a mixture of vanadium oxide-immobilized mesoporous silica nanoparticles (V-MPS4) and the hydrolytic enzyme lipase (e.g., CAL-B), the R or S optical isomer of the racemic alcohol is eluted from the column outlet. The reaction mechanism is that one optical isomer of the racemic alcohol is selectively esterified by optical resolution using lipase, while the other optical isomer of the racemic alcohol is racemized by the vanadium oxide contained in V-MPS4. These two reactions proceed simultaneously, converting all of the racemic alcohol into a single optically active ester.

When V-MPS4 and CAL-B are mixed with particles (gap filling material) such as celite or silica gel and packed into a column, the particle sizes of V-MPS4 and CAL-B are different, so the fine particles of V-MPS4, which have a smaller particle size, move to the outlet side along with the fluid. This causes the initial state in which the two types of catalysts are uniformly distributed to collapse, resulting in a problem that the catalytic activity cannot be stably exerted. Therefore, this problem was solved by mixing silica porous particles as a gap filling material to fix the two types of catalyst particles inside the column.

Hereinafter, Particle 1 (V-MPS4) corresponds to the fine powder in the present invention, Particle 2 plays the role of a gap filler, and lipase (CAL-B) plays the role of an enzyme. Note that Particle 2 (silica porous particles) used in Examples 11 to 25 corresponds to the porous particles in the present invention.

Example 11
20 mg of V-MPS4 (manufactured by FUJIFILM Wako Pure Chemical Industries, Ltd., average primary particle size 1 μm, pore size 4 nm, vanadium content 0.2 mmol/g) as particle 1, 140 mg of silica porous particles (manufactured by SNG Corporation, lot number: PM043, most frequent pore size in the pore size distribution of through holes: 2 μm, most frequent pore size in the pore size distribution of pores: 6 nm, particle size: 20 to 63 μm) as particle 2, and 300 mg of CAL-B (Candida antarctica lipase B, manufactured by Roche, particle size: 300 to 700 μm) as lipase were placed in a glass container. Then, the mixture was mixed by shaking well by hand under an argon atmosphere, which is an inert gas, to obtain a mixture of particle 1, particle 2, and lipase.

Next, the mixture was poured into an Omnifit EZ empty column (Isis Co., Ltd., inner diameter 6.6 mm) using a medicine spoon and uniformly distributed to obtain a packed layer. After that, the packed layer was compressed and fixed by tightening the movable plug of the Omnifit EZ empty column without tapping, and a column for catalytic reaction was prepared. The column was placed in a thermostatic device at 35°C. Next, raw material compound (1a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to raw material compound (1a) (concentration of 1a: 0.1 M) were pumped into the prepared column for catalytic reaction at a flow rate of 0.03 mL/min. After three hours of continuous liquid flow, the raw material compound (1a), target compound (1b) and by-product (1c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (%ee) of the raw material compound (1a) and target compound (1b), as well as the yield (%) of the by-product (1c) were calculated. The by-product (1c) was a dimeric ether of the raw material compound (1a).

The structural formula of the starting compound (1a) is as follows.

The structural formula of the target compound (1b) is as follows.

(Comparative Example 7)
The test was conducted in the same manner as in Example 11, except that 280 mg of single-pore silica gel particles (manufactured by Kanto Chemical Co., Ltd., particle size: 40 to 50 μm, pore size: 6 nm) having no through-holes but having fine pores were used as particles 2.

(Comparative Example 8)
The same test as in Example 11 was conducted except that 500 mg of Celite particles (Kishida Chemical Co., Ltd., particle size: 20 to 100 μm) having neither through holes nor fine pores was used as particle 2.

The results of Example 11 and Comparative Examples 7 and 8 are shown in Table 5 below.

As shown in Table 5, in Example 11, 96% of the raw material compound (1a) was reacted (= reaction rate 96%), and the target compound (1b) was produced with a yield of 87% and an optical purity of 98% ee. The reaction rate in Example 11 was higher than that in Comparative Example 7 (= reaction rate 78%) and Comparative Example 8 (= reaction rate 72%). This leads to a high yield of the target compound (1b) even under the same reaction conditions. In addition, the optical purity of the raw material compound (1a) recovered in Example 11 was lower than that in Comparative Examples 7 to 8, indicating that racemization proceeded rapidly in the case of Example 11. This is an extremely important result in terms of efficiently performing dynamic optical resolution. Furthermore, in Example 11, although the weight of the particle 2 (gap filling material) was less than half that of Comparative Examples 7 and 8, dynamic optical resolution with high efficiency, high yield, and high optical purity was possible.

In Example 11 and Comparative Examples 7 to 8, before each test, the beige color of V-MPS4 was uniformly dispersed in the packed layer. After the test in Example 11, the entire column was uniformly beige, confirming that the movement of the fine powder V-MPS4 had been suppressed. On the other hand, after the tests in Comparative Examples 7 to 8, the beige color on the inlet side of the column had disappeared and turned white, and the beige color on the outlet side had become darker, confirming with the naked eye that the fine powder V-MPS4 had moved due to the liquid being pumped. In other words, after the tests in Comparative Examples 7 to 8, it was confirmed that the movement of the fine powder V-MPS4 could not be suppressed.

The raw material compound (1a), target compound (1b), and by-product (1c) in the following Examples 12 to 19 are the same as the raw material compound (1a), target compound (1b), and by-product (1c) in Example 11.

Example 12
A catalytic reaction column was prepared under the same conditions as in Example 11, except that 61 mg of V-MPS4 was used as particle 1 and 118 mg of silica porous particles were used as particle 2, and the column was placed in a thermostatic device at 35° C. Then, raw material compound (1a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to raw material compound (1a) (concentration of 1a: 0.1 M) were fed to the prepared catalytic reaction column at a flow rate of 0.03 mL/min. After continuous feeding for 3 hours, raw material compound (1a), target compound (1b) and by-product (1c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of raw material compound (1a) and target compound (1b), as well as the yield (%) of by-product (1c) were calculated.

Example 13
A column for catalytic reaction was prepared under the same conditions as in Example 11, except that 30 mg of V-MPS4 was used as particle 1 and 135 mg of porous silica particles was used as particle 2, and the column was placed in a thermostatic device at 35° C. Then, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

(Example 14)
A column for catalytic reaction was prepared under the same conditions as in Example 11, except that 15 mg of V-MPS4 was used as particle 1 and 140 mg of porous silica particles was used as particle 2, and the column was placed in a thermostatic device at 35° C. Then, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

(Example 15)
61 mg of V-MPS4 was used as particle 1, and 118 mg of silica porous particles were used as particle 2. The packed bed was divided into a first region, a second region, and a third region in the flow direction of the liquid sent from the upstream side to the downstream side in that order, and the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set as follows: first region = 1: 4.1, second region = 1: 2.4, third region = 1: 1, and the mixing ratio (weight ratio) of particle 1 to lipase was set as follows: first region = 1: 9.1, second region = 1: 5.9, third region = 1: 3.03. A column for catalytic reaction was prepared under the same conditions as in Example 11, and the column was placed in a thermostatic device at 35 ° C. Subsequently, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated. The lengths of the first, second and third regions are the lengths of the packed bed divided into three equal parts. That is, the length of the packed bed is the sum of the length of the first, second and third regions. The same applies to the following Examples 16 to 19.

(Example 16)
A column for catalytic reaction was prepared under the same conditions as in Example 11, except that 30 mg of V-MPS4 was used as particle 1, 135 mg of silica porous particles was used as particle 2, and the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to 1:7.8 in the first region, 1:4.5 in the second region, and 1:3.1 in the third region, and the mixing ratio (weight ratio) of particle 1 to lipase was set to 1:16.7 in the first region, 1:10 in the second region, and 1:7.1 in the third region. The column was then placed in a thermostatic device at 35° C. Subsequently, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

(Example 17)
22 mg of V-MPS4 was used as particle 1, and 139 mg of silica porous particles were used as particle 2. The mixing ratio (weight ratio) of particle 1 to particle 2 in each region was 1:9.6 in the first region, 1:6.6 in the second region, and 1:4.5 in the third region, and the mixing ratio (weight ratio) of particle 1 to lipase was 1:20 in the first region, 1:14.3 in the second region, and 1:10 in the third region. Except for this, a column for catalytic reaction was prepared under the same conditions as in Example 11, and the column was placed in a thermostatic device at 35° C. Subsequently, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

(Example 18)
A column for catalytic reaction was prepared under the same conditions as in Example 11 except that, when 15 mg of V-MPS4 was used as particle 1 and 143 mg of porous silica particles were used as particle 2, the packed bed was divided into a first region, a second region, and a third region in the order from the upstream side to the downstream side in the flow direction of the liquid in the packed bed, and the ratio of particle 1 to particle 2 was changed, the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to the first region = 1:16.3, the second region = 1:9.6, and the third region = 1:6.6, and the mixing ratio (weight ratio) of particle 1 to lipase was set to the first region = 1:33.3, the second region = 1:20, and the third region = 1:14.3, and the column was placed in a thermostatic device at 35°C. Subsequently, the test was carried out under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the starting compound (1a), the target compound (1b) and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

(Example 19)
A column for catalytic reaction was prepared under the same conditions as in Example 11, except that 22 mg of V-MPS4 was used as particle 1, 139 mg of silica porous particles was used as particle 2, and the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to 1:4.5 in the first region, 1:6.6 in the second region, and 1:9.6 in the third region, and the mixing ratio (weight ratio) of particle 1 to lipase was set to 1:10 in the first region, 1:14.3 in the second region, and 1:20 in the third region. The column was then placed in a thermostatic device at 35° C. Subsequently, a test was performed under the same conditions as in Example 12, and the yield (%) and optical purity (% ee) of the raw material compound (1a), the target compound (1b), and the by-product (1c), as well as the yield (%) of the by-product (1c) were calculated.

The results for Examples 12 to 19 are shown in Table 6 below.

As shown in Table 6, compared to the case where the mixing ratio of particle 1 and particle 2 was not changed from the upstream side to the downstream side, i.e., the mixing ratio of particle 1 and lipase was not changed (Examples 12 to 14), the yield of by-product (1c) was reduced and the yield of target compound (1b) was improved when the mixing ratio of particle 1 and particle 2 was changed by dividing into the first region, the second region, and the third region, i.e., when the mixing ratio of particle 1 and lipase was changed (Examples 15 to 18). In all of Examples 15 to 18, the optical purity of target compound (1b) was 97% ee or more. In addition, Examples 15 to 18, in which the weight of particle 1 was increased in the order of the first region, the second region, and the third region, and the weight of particle 2 was decreased, while the weight of lipase was kept constant, were able to reduce the yield of by-product (1c) and improve the yield of target compound (1b) more than Example 19, in which the weight of particle 1 was decreased in the order of the first region, the second region, and the third region, and the weight of particle 2 was increased.

(Example 20)
A column for catalytic reaction was prepared under the same conditions as in Example 11 except that, when 100 mg of V-MPS4 was used as particle 1 and 98 mg of porous silica particles were used as particle 2, the packed bed was divided into a first region, a second region, and a third region in the order from the upstream side to the downstream side in the flow direction of the liquid in the packed bed, and the mixing ratio of particle 1 to particle 2 was changed, the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to the first region=1:2.4, the second region=1:1, and the third region=1:0.5, and the mixing ratio (weight ratio) of particle 1 to lipase was set to the first region=1:5.9, the second region=1:3.03, and the third region=1:2, and the column was placed in a thermostatic device at 35°C. Subsequently, the raw material compound (2a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to the raw material compound (2a) (concentration of 2a: 0.1 M) were fed to the prepared catalytic reaction column at a flow rate of 0.03 mL/min. After continuous feeding for 3 hours, the raw material compound (2a), the target compound (2b) and the by-product (2c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of the raw material compound (2a) and the target compound (2b), as well as the yield (%) of the by-product (2c) were calculated. The by-product (2c) was a dimeric ether of the raw material compound (2a).

The structural formula of the starting compound (2a) is as follows.

The structural formula of the target compound (2b) is as follows.

(Example 21)
A column for catalytic reaction was prepared under the same conditions as in Example 11 except that, when 43 mg of V-MPS4 was used as particle 1 and 129 mg of porous silica particles were used as particle 2, the packed bed was divided into a first region, a second region, and a third region in the flow direction of the liquid in the packed bed from the upstream side to the downstream side in that order, the mixing ratio of particle 1 to particle 2 was changed, and the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to the first region=1:6.7, the second region=1:3.1, and the third region=1:1.8, and the mixing ratio (weight ratio) of particle 1 to lipase was set to the first region=1:14.3, the second region=1:7.1, and the third region=1:4.5. The column was then placed in a thermostatic device at 35°C. Subsequently, the raw material compound (3a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to the raw material compound (3a) (concentration of 3a: 0.1 M) were fed to the prepared catalytic reaction column at a flow rate of 0.03 mL/min. After continuous feeding for 3 hours, the raw material compound (3a), the target compound (3b) and the by-product (3c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of the raw material compound (3a) and the target compound (3b), as well as the yield (%) of the by-product (3c) were calculated. The by-product (3c) was a dimeric ether of the raw material compound (3a).

The structural formula of the starting compound (3a) is as follows.

The structural formula of the target compound (3b) is as follows.

(Example 22)
A column for catalytic reaction was prepared under the same conditions as in Example 21, and the column was placed in a thermostatic device at 35 ° C. Then, the raw material compound (4a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to the raw material compound (4a) (concentration of 4a: 0.1 M) were fed to the prepared column for catalytic reaction at a flow rate of 0.03 mL / min. After continuous feeding for 3 hours, the raw material compound (4a), the target compound (4b) and the by-product (4c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of the raw material compound (4a) and the target compound (4b), as well as the yield (%) of the by-product (4c) were calculated. The by-product (4c) was a dimer ether of the raw material compound (4a).

The structural formula of the starting compound (4a) is as follows.

The structural formula of the target compound (4b) is as follows.

(Example 23)
A column for catalytic reaction was prepared under the same conditions as in Example 21, and the column was placed in a thermostatic device at 35°C. Then, the raw material compound (5a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to the raw material compound (5a) (concentration of 5a: 0.1 M) were fed to the prepared column for catalytic reaction at a flow rate of 0.015 mL/min. After continuous feeding for 6 hours, the raw material compound (5a), the target compound (5b) and the by-product (5c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (%ee) of the raw material compound (5a) and the target compound (5b), as well as the yield (%) of the by-product (5c) were calculated. The by-product (5c) was a dimer ether of the raw material compound (5a).

The structural formula of the starting compound (5a) is as follows.

The structural formula of the target compound (5b) is as follows.

(Example 24)
A column for catalytic reaction was prepared under the same conditions as in Example 11 except that, when 150 mg of V-MPS4 was used as particle 1 and 75 mg of porous silica particles were used as particle 2, the packed bed was divided into a first region, a second region, and a third region in the order from the upstream side to the downstream side in the flow direction of the liquid in the packed bed, and the mixing ratio of particle 1 to particle 2 was changed, the mixing ratio (weight ratio) of particle 1 to particle 2 in each region was set to the first region=1:1, the second region=1:0.5, and the third region=1:0.25, and the mixing ratio (weight ratio) of particle 1 to lipase was set to the first region=1:3.03, the second region=1:2, and the third region=1:1.49, and the column was placed in a thermostatic device at 35°C. Subsequently, the raw material compound (6a) and an acetonitrile solution containing 4 molar equivalents of vinyl butyrate relative to the raw material compound (6a) (concentration of 6a: 0.1 M) were fed to the prepared catalytic reaction column at a flow rate of 0.03 mL/min. After continuous feeding for 3 hours, the raw material compound (6a), the target compound (6b) and the by-product (6c) present in the solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of the raw material compound (6a) and the target compound (6b), as well as the yield (%) of the by-product (6c) were calculated. The by-product (6c) was a dimeric ether of the raw material compound (6a).

The structural formula of the starting compound (6a) is as follows.

The structural formula of the target compound (6b) is as follows.

The results of Example 16 and Examples 20 to 24 are shown in Table 7 below.

In Examples 20 to 24, similar to Example 11, each target compound was obtained in a yield of 90% or more and with an optical purity of 96% ee or more, demonstrating that dynamic optical resolution proceeds quickly even for a wide range of compounds with different substituents.

The raw material compound (2a), target compound (2b), and by-product (2c) in Example 25 are the same as the raw material compound (2a), target compound (2b), and by-product (2c) in Example 20.

(Example 25)
A catalytic reaction column was prepared under the same conditions as in Example 11, except that 100 mg of V-MPS4 was used as particle 1, 100 mg of silica porous particles as particle 2, and 300 mg of CAL-B was used as lipase, and the column was placed in a thermostatic device at 35°C. Then, raw material compound (2a) and an acetonitrile solution containing 4 molar equivalents of vinyl acetate relative to raw material compound (2a) (concentration of 2a: 0.1 M) were fed to the prepared catalytic reaction column at a flow rate of 0.03 mL/min. After every 6 hours of continuous feeding, the solution obtained from the column outlet was collected, and continuous feeding was performed for a total of 78 hours. The starting compound (2a), the target compound (2b), and the by-product (2c) present in each solution obtained from the column outlet were analyzed by 1H-NMR and HPLC, and the yield (%) and optical purity (% ee) of the starting compound (2a) and the target compound (2b), as well as the yield (%) of the by-product (2c) were calculated.

The results of Example 25 are shown in Table 8 below.

From the results of Example 25, even with continuous liquid transfer for 78 hours, the yield of the target product (2b) was stable at around 90%, and the optical purity was not reduced at 99% ee, indicating that the catalyst can be used continuously without deterioration.

Claims (8)

A packed bed formed of a mixture of porous particles and fine powder,
(1) The porous particle has a skeleton, the skeleton has through holes, the through holes have a diameter of 0.1 to 100 μm, the through holes have a most frequent pore size of 0.5 μm or more and 50 μm or less in a pore size distribution, and the porous particle has a particle size of 20 μm or more and 3000 μm or less,
(2) the average primary particle size of the fine powder is 1 nm or more and is 80% or less of the most frequent pore size in the pore size distribution of the through holes;
(3) The bulk density of the porous particles is 0.05 to 7.5 (mg/mm ³ );
A packed bed characterized by: 2. The packed bed according to claim 1 , wherein the fine powder has a bulk density of 0.02 to 7.5 (mg/mm ³ ). 3. The packed bed according to claim 1 , wherein the packed bed has a bulk density of 0.05 to 7.5 (mg/mm ³ ). The packed bed according to any one of claims 1 to 3 ,
A packed bed, in which a mixing ratio of the porous particles and the fine powder in the packed bed changes stepwise. The packed bed according to any one of claims 1 to 4 , wherein the mixture further comprises enzyme particles. 6. The packed bed of claim 5 , wherein the enzyme particles are lipase. The packed bed according to claim 5 or 6 ,
A packed bed, in which a mixing ratio of the fine powder and the enzyme particles in the packed bed changes stepwise. A method for passing a gas or liquid through a pipe provided with the packed bed according to any one of claims 1 to 7 .