JP7456085B2 - Method for separating or purifying a cyclic compound, method for producing a cyclic compound, separation material, and separation device - Google Patents

Description

The present invention relates to a method for separating or purifying a cyclic compound, a method for producing a cyclic compound, a separation material, and a separation device.

In recent years, cyclic compounds have attracted attention because they exhibit specific properties different from linear compounds and can be applied to various uses. The shape of a compound is one of the most important factors that influences the physical properties and functions of a compound, but since cyclic compounds do not have terminals, they are more difficult to form than linear compounds with approximately the same molecular weight and composition. They are known to exhibit various unique properties such as low viscosity, small hydrodynamic volume, and high glass transition temperature.

Generally, a cyclic compound is synthesized by a cyclization reaction of a linear compound. Here, in the cyclization reaction, not all linear compounds become cyclic compounds, but linear compounds that are unreacted substances and linear by-products generated during this reaction also exist. Therefore, after the cyclization reaction, the cyclic compound and the linear compound are in a mixture state, so in order to extract only the cyclic compound, it is necessary to separate the cyclic compound and the linear compound.

Conventionally, it has also been considered to separate or purify cyclic compounds and linear compounds by reprecipitation methods (for example, see Non-Patent Document 1), separation methods using gel permeation chromatography (GPC), and the like. The reprecipitation method is a method of separating by collecting a precipitate using a poor solvent. In the separation method using GPC, the size of the cyclic compound is slightly smaller than that of the linear compound, so this difference in size is used to separate the cyclic compound and the linear compound.

J. Cooke, K. Viras, G. E. Yu, T. Sun, T. Yonemitsu, A. J. Ryan, C. Price and C. Booth, Macromolecules, 1998, 31, 3030-3039.

However, in the reprecipitation method, the separation efficiency is low and it is necessary to repeat it many times. Furthermore, in the separation method using GPC, the larger the size of the cyclic compound and the linear compound, the smaller the difference in size between the cyclic compound and the linear compound, making separation even more difficult. Therefore, at present, no method has yet been developed for separating or purifying a cyclic compound with high separation or purification efficiency and independent of the size of the cyclic compound.

The present invention has been made to solve the above problems. That is, a method for separating or purifying a cyclic compound that has high separation or purification efficiency and can separate or purify a cyclic compound without depending on the size of the cyclic compound, a method for producing a cyclic compound using this separation method, a separation material, and to provide a separation device equipped with the same.

[1] A separation or purification method for separating or purifying the cyclic compound from a mixture containing the cyclic compound and the linear compound, the separation or purification method comprising bringing the mixture into contact with a porous channel material.

[2] The separation or purification method described in [1] above, in which the ratio of the pore size of the porous channel material to the maximum thickness of the linear compound is 0.5 or more and 2.0 or less.

[3] The separation or purification method according to [1] or [2] above, wherein the porous channel material is a porous metal complex, a covalent organic structure, or an organic cage compound.

[4] The separation or purification method according to any one of [1] to [3] above, wherein the cyclic compound and the linear compound are each a polymer compound.

一般式（１）中、Ｘ^１はポリマー主鎖を表し、
前記線状化合物が、一般式（２）で表されるポリマー化合物であり、

一般式（２）中、Ｘ^２はポリマー主鎖を表し、Ｙ^１およびＹ^２は互いに同一または互いに異なっていてもよい末端基を表し、
Ｘ^１およびＸ^２が同一である、上記[１]ないし［４］のいずれか一項に記載の分離または精製方法。 [5] The cyclic compound is a polymer compound represented by general formula (1),

In general formula (1), X ¹ represents a polymer main chain,
The linear compound is a polymer compound represented by general formula (2),

In general formula (2), X ² represents a polymer main chain, Y ¹ and Y ² represent terminal groups that may be the same or different from each other,
The separation or purification method according to any one of [1] to [4] above, wherein X ¹ and X ² are the same.

[6] The separation or purification method according to [5] above, wherein X ¹ and X ² each consist of a repeating unit.

[7] A method for producing a cyclic compound, comprising the step of cyclizing a linear compound to obtain a mixture containing the cyclic compound and the linear compound, and the method described in any one of [1] to [6] above. A method for producing a cyclic compound, comprising: separating or purifying the cyclic compound from the mixture by a separation method.

[8] A separation material for separating the cyclic compound from a mixture containing the cyclic compound and the linear compound, the separation material comprising a porous channel material.

[9] The separation material according to [8] above, wherein the ratio of the pore diameter of the porous channel material to the maximum thickness of the linear compound is 0.5 or more and 2.0 or less.

[10] The separation material according to [8] or [9] above, wherein the porous channel material is a porous metal complex, a covalent organic structure, or an organic cage compound.

[11] The separation material according to any one of [8] to [10] above, wherein the cyclic compound and the linear compound are each a polymer compound.

[12] Any one of [8] to [11] above, wherein the cyclic compound and the linear compound each contain a repeating unit, and the repeating unit of the cyclic compound and the repeating unit of the linear compound are the same. Separation material according to paragraph 1.

[13] A separation device for separating the cyclic compound from a mixture containing the cyclic compound and the linear compound, comprising a casing having an inlet and an outlet, and the casing filled with the above [8] to [12]. ] A separation device comprising the separation material according to any one of the above.

According to the method for separating or purifying a cyclic compound, the method for producing a cyclic compound, the separation material, and the separation device of the present invention, the separation or purification efficiency is high, and the cyclic compound can be separated or purified without depending on the size of the cyclic compound. Can be purified.

FIG. 1 is a graph showing the analysis results of the crude product used in Example 1 by gel permeation chromatography (GPC). 2(A) and FIG. 2(B) are diagrams showing the structure of the MOF 1 used in Example 1. FIG. 3 is a graph showing the analysis results of MOF1 synthesized in Example 1 by X-ray powder diffraction (XRPD). FIG. 4(A) is a graph showing the analysis results of a composite of linear PEG and MOF1 by differential scanning calorimetry (DSC), and FIG. 4(B) is a graph showing the differential scanning calorimetry of a composite of cyclic PEG and MOF1. It is a graph showing the analysis results by a digital camera (DSC). FIG. 5(A) is a diagram showing a simulation of introducing linear PEG into MOF1, and FIG. 5(B) is a diagram showing a simulation of introducing circular PEG into MOF1. FIG. 6 is a graph showing the analysis results of linear PEG and cyclic PEG with molecular weights of 1000, 2000, 3000, and 10000 by high performance liquid chromatography (HPLC) using a column packed with MOF1. FIG. 7 is a graph representing a gradient elution program. 8(A) is a graph showing the analysis results of the components of fraction 1 by carbon-13 nuclear magnetic resonance ( ^13C -NMR) spectroscopy, and FIG. 8(B) is a graph showing the analysis results of the components of fraction 2 by carbon-13 nuclear magnetic resonance (13C-NMR). 1 is a graph showing analysis results by resonance ( ¹³ C-NMR) spectroscopy. FIG. 9 is a graph showing the analysis results of the crude product, components of fraction 1, and components of fraction 2 by gel permeation chromatography (GPC). FIG. 10(A) is a graph showing the analysis results of the components of fraction 1 by matrix-assisted laser desorption ionization time-of-flight mass spectrometer (MALDI-TOF-MS), and FIG. It is an enlarged view of the vicinity of 2000 m/z in A), and FIG. 10(C) is an enlarged view of the vicinity of 4000 m/z in FIG. 10(A). FIG. 11(A) is a graph showing the analysis results of the components of fraction 2 by matrix-assisted laser desorption ionization time-of-flight mass spectrometer (MALDI-TOF-MS), and FIG. It is an enlarged view of the vicinity of 2000 m/z in A). FIG. 12 is a graph showing the analysis results of the compound A obtained in Comparative Example 1 by gel permeation chromatography (GPC). FIG. 13 is a graph showing the analysis results of Compound A by carbon-13 nuclear magnetic resonance ( ¹³ C-NMR) spectroscopy.

Hereinafter, a method for separating or purifying a cyclic compound, a method for producing a cyclic compound, a separation material, and a separation device according to an embodiment of the present invention will be described. The term "cyclic compound" as used herein means a compound having a cyclic structure, and the term "linear compound" herein refers to a compound having a linear structure.

<<Separation or purification method and manufacturing method of cyclic compound>>
First, a mixture of a cyclic compound and a linear compound is obtained. As described above, when it is attempted to obtain a cyclic compound by the cyclization reaction of a linear compound, unreacted linear compounds and linear by-products are also present in addition to the cyclic compound. Therefore, mixtures of cyclic and linear compounds can be obtained by cyclization reactions of linear compounds.

For example, when cyclic polyethylene glycol (hereinafter, polyethylene glycol may be referred to as "PEG") is to be obtained, cyclic PEG can be synthesized by the cyclization reaction described in Non-Patent Document 1 using linear PEG as a raw material, and a mixture of cyclic PEG and linear PEG can be obtained by this method. In this method, the hydroxy group at the end of linear PEG is tosylated, and when the alkoxide at the PEG end attacks the tosylated carbon atom with a nucleophilic action, -OTs is eliminated by an S _N 2 reaction, and cyclic PEG is synthesized. Tosylated linear PEG is susceptible to nucleophilic substitution reaction with oxygen atoms six units away from the end, and as a result, 1,4-dioxane and cyclic PEG shortened by two ethylene oxide units are produced from the linear PEG.

The cyclic compound may be either a cyclic low-molecular compound or a high-molecular compound, and the linear compound may be a linear low-molecular compound or a high-molecular compound. However, in the separation method using GPC, as the size increases, the difference in size becomes smaller and separation becomes difficult, so the method of the present invention is particularly effective for high molecular compounds. In this specification, the term "low molecular compound" refers to a compound with a molecular weight of less than 600, and the term "high molecular compound" refers to a compound with a molecular weight of 600 or more. In this specification, unless otherwise specified, "molecular weight" shall mean weight average molecular weight. The weight average molecular weight is a polyethylene glycol equivalent value measured by gel permeation chromatography (GPC).

When both the cyclic compound and the linear compound are low molecular weight compounds, the lower limit of the molecular weight is not particularly limited, but from the viewpoint of improving separation efficiency, it is 100 or more, 125 or more, 150 or more, 175 or more, 200 or more, 225 or more. , or 250 or more.

When both the cyclic compound and the linear compound are polymeric compounds, the lower limit of the molecular weight is not particularly limited, but from the viewpoint of improving separation efficiency, it is 1250 or more, 1500 or more, 1750 or more, 2000 or more, 2250 or more, 2500 or more. , 2750 or more, 3000 or more, 3250 or more, 3500 or more, 3750 or more, 4000 or more, 5000 or more, or 6000 or more. Further, the upper limit of the molecular weight is not particularly limited, but may be 8,000,000 or less, 4,000,000 or less, 200,000 or less, 50,000 or less, 20,000 or less, 10,000 or less, or 5,000 or less from the viewpoint of improving separation efficiency and easy availability.

In a cyclic compound, there are always two linear portions constituting the cyclic compound in the radial direction, so even if the diameter of the cyclic compound is not uniform, for example, the diameter of the linear compound is 2, which is the maximum thickness described below. It is preferable that it is more than double. The diameter of the cyclic compound is determined, for example, by dividing the total length of the cyclic compound by the circumference.

The maximum thickness of the linear compound is not particularly limited, but may be, for example, 2.0 Å (0.20 nm) or more and 20 Å (2.0 nm) or less. The lower limit of the maximum thickness of the linear compound may be 2.5 Å (0.25 nm) or more, or 3.0 Å (0.30 nm) or more, and the upper limit is 15.0 Å (1.5 nm) or less. , 10.0 Å (1.0 nm) or less, or 5.0 Å (0.5 nm) or less. The maximum thickness of a linear compound is calculated by the conjugate gradient method using a universal force field using commercially available molecular simulation software (for example, product name "BIOVIA Material Studio", manufactured by Dassault Systèmes Biovia). It is obtained by performing structural optimization calculations on the structural model and calculating the van der Waals radius of the linear compound whose structure has been optimized.

The length of the linear compound is not particularly limited, but may be, for example, 0.8 nm or more and 40 μm or less. The lower limit of the length of the linear compound may be 1 nm or more, 2 nm or more, 4 nm or more, 6 nm or more, or 8 nm or more, and the upper limit may be 20 μm or less, 1 μm or less, or 100 nm or less. The length of a linear compound can be calculated theoretically from the molecular weight and chemical structure of the linear compound, taking into account bond distances and bond angles.

Groups with small terminal groups or groups with low polarity (for example, halogen atoms, H, CH ₃ , OCH ₃ , OCOCH ₃ , COOCH _3, etc.) of the linear compound easily enter the pores.

一般式（１）中、Ｘ^１はポリマー主鎖を表す。Ｘ^１が炭素原子を含む場合、Ｘ^１の炭素数の下限は、特に限定されないが、例えば、６以上、８以上、または１２以上であってもよく、またＸ^１の炭素数の上限は、５５００００以下、１００００以下、または３００以下であってもよい。 The cyclic compound may be a polymer compound represented by general formula (1).

In general formula (1), X ¹ represents a polymer main chain. When X ¹ contains a carbon atom, the lower limit of the number of carbon atoms in X ¹ is not particularly limited, but may be, for example, 6 or more, 8 or more, or 12 or more, and the upper limit of the number of carbon atoms in X ¹ is: It may be 550,000 or less, 10,000 or less, or 300 or less.

一般式（２）中、Ｘ^２はポリマー主鎖を表し、Ｙ^１およびＹ^２は互いに同一または互いに異なっていてもよい末端基を表す。Ｘ^２が炭素原子を含む場合、Ｘ^２の炭素数の下限は、特に限定されないが、例えば、６以上、８以上、または１２以上であってもよく、またＸ^２の炭素数の上限は、５５００００以下、１００００以下、または３００以下であってもよい。 The linear compound may be a polymer compound represented by general formula (2).

In general formula (2), X ² represents a polymer main chain, and Y ¹ and Y ² represent terminal groups that may be the same or different from each other. When X ² contains a carbon atom, the lower limit of the number of carbon atoms in X ² is not particularly limited, but may be, for example, 6 or more, 8 or more, or 12 or more, and the upper limit of the number of carbon atoms in X ² is: It may be 550,000 or less, 10,000 or less, or 300 or less.

When a cyclic compound is obtained by a cyclization reaction of a linear compound, X ¹ of the cyclic compound in the mixture and X ² of the linear compound are the same. In conventional methods, separation and purification are difficult when X ¹ and X ² are the same, so the separation or purification method of the present invention is particularly effective when X ¹ and X ² are the same.

When X ¹ and X ² are the same, each of X ¹ and X ² may consist of a repeating unit. Specifically, a cyclic compound containing a repeating unit is represented by the following general formula (3), and a linear compound containing a repeating unit is represented by the following general formula (4).

一般式（３）中、Ａ^１は繰返し単位を表し、ｍは２以上の整数を表す。ｍは、特に限定されないが、例えば、２以上２００００以下、または４以上１５０００以下であってもよい。

In general formula (3), A ¹ represents a repeating unit, and m represents an integer of 2 or more. m is not particularly limited, and may be, for example, 2 or more and 20,000 or less, or 4 or more and 15,000 or less.

一般式（４）中、Ａ^２は繰返し単位を表し、ｎは２以上の整数を表し、ｎは、特に限定されないが、例えば、２以上２００００以下、４以上１５０００以下であってもよい。なお、環状化合物の繰り返し単位の数ｍと、線状化合物の繰り返し単位の数ｎは、同一である必要はない。

In general formula (4), A ² represents a repeating unit, n represents an integer of 2 or more, and n is not particularly limited, but may be, for example, 2 or more and 20,000 or less, or 4 or more and 15,000 or less. Note that the number m of repeating units of the cyclic compound and the number n of repeating units of the linear compound do not need to be the same.

（式中、R^bはアデニン、グアニン、シトシン、チミン又はウラシルを示し、Ｚは、Ｈ又はＯＨ基を示す。）などが挙げられるが、これに限定されない。ここで、「-6-β-D-Glup-1-」は、β-D-グルコピラノースの６位と１位で結合する繰り返し単位、「-4-β-D-Glup-1-」は、β-D-グルコピラノースの４位と１位で結合する繰り返し単位、「-6-β-D-Galp-1-」は、β-D-ガラクトピラノースの６位と１位で結合する繰り返し単位、「-4-β-D-Galp-1-」は、β-D-ガラクトピラノースの４位と１位で結合する繰り返し単位、「-6-β-D-Manp-1-」は、β-D-マンノピラノースの６位と１位で結合する繰り返し単位、「-4-β-D-Manp-1-」は、β-D-マンノピラノースの４位と１位で結合する繰り返し単位、「-6-β-D-GlupNAc-1-」は、β-D-N-アセチルグルコサミンの６位と１位で結合する繰り返し単位、「-4-β-D-GlupNAc-1-」は、β-D-N-アセチルグルコサミンの４位と１位で結合する繰り返し単位、「-6-β-D-GalpNAc-1-」は、β-D-N-アセチルガラクトサミンの６位と１位で結合する繰り返し単位、「-4-β-D-GalpNAc-1-」は、β-D-N-アセチルガラクトサミンの４位と１位で結合する繰り返し単位、「-6-β-D-Fluf-1-」は、β-D-フルクトフラノースの６位と1位で結合する繰り返し単位、「-4-β-D-Fluf-1-」は、β-D-フルクトフラノースの４位と１位で結合する繰り返し単位を各々示す。 ^{The above -} _{mentioned X} ¹ _{, ​ ​ ​ ​} CH ₂ CH ₂ O)－, -(CH ₂ CH(CH ₃ )O)－, -OCH(CH ₃ )CO－, -OCH ₂ CO－, -(OCH ₂ CH ₂ CH ₂ CH ₂ OCOCH ₂ CH ₂ CO)－,－(OCH ₂ CH ₂ CH ₂ CH ₂ OCOCH ₂ CH ₂ CH ₂ CH ₂ CO)－,－(OCH ₂ CH ₂ OCOC ₆ H ₄ CO)－,－(OCH ₂ CH ₂ CH ₂ CH ₂ OCOC ₆ H ₄ CO)－, -(OCH ₂ CH ₂ OCOCH ₂ CH ₂ CO)－, -(OCH ₂ CH ₂ OCOCH ₂ CH ₂ CH ₂ CH ₂ CO)－, -(NHCH ₂ CH ₂ CO)－, -(NHCH ₂ CH ₂ CH ₂ CO)-, -(NHCH ₂ CH ₂ CH ₂ CH ₂ CH ₂ CO)-, -(CO(CH ₂ ) ₄ CONH(CH ₂ ) ₆ NH)-, -(CO( CH ₂ ) ₈ CONH(CH ₂ ) ₆ NH)-, -(NH(CH ₂ ) ₁₀ CO)-, -(NH(CH ₂ ) ₁₁ CO)-, -(COC ₆ H ₄ CONHC ₆ H ₄ NH) -, -(CH ₂ CH ₂ )-, -(CH ₂ CH ₂ CH ₂ )-, -(CH ₂ CH(CH ₃ ))-, -(CH2CH(C6H5))-, -(CH ₂ C(CH ₃ )(COOCH ₃ )-, -(CH ₂ CH(OCOCH ₃ )-, -(CH2CH(OH))-, -(OSi(CH ₃ ) ₂ )-, -CH=CH-, -CH=N- , -N=N-, -C≡C-, 1,4-phenylene group, 1,4-cyclohexylene group, -CO-CHR ^a -NH- (R ^a is the side chain of 20 amino acids ), -6-β-D-Glup-1-, -4-β-D-Glup-1-, -6-β-D-Galp-1-, -4-β-D-Galp-1-, -6-β-D-Manp-1-, -4-β-D-Manp-1-, -6-β-D-GlupNAc-1-, -4-β-D-GlupNAc-1-, -6 -β-D-GalpNAc-1-, -4-β-D-GalpNAc-1-, -6-β-D-Fluf-1-, -4-β-D-Fluf-1-,

(In the formula, R ^b represents adenine, guanine, cytosine, thymine, or uracil, and Z represents H or OH group.) Examples include, but are not limited to, the following. Here, "-6-β-D-Glup-1-" is a repeating unit that binds at the 6th and 1st positions of β-D-glucopyranose, and "-4-β-D-Glup-1-" is , "-6-β-D-Galp-1-" is a repeating unit that is bonded at the 4th and 1st positions of β-D-glucopyranose, and "-6-β-D-Galp-1-" is a repeating unit that is bonded at the 6th and 1st positions of β-D-galactopyranose. The unit "-4-β-D-Galp-1-" is a repeating unit that bonds at the 4th and 1st positions of β-D-galactopyranose, and "-6-β-D-Manp-1-" is The repeating unit "-4-β-D-Manp-1-" that binds at the 6th and 1st positions of β-D-mannopyranose binds at the 4th and 1st positions of β-D-mannopyranose. The repeating unit "-6-β-D-GlupNAc-1-" is a repeating unit that bonds at the 6th and 1st positions of β-DN-acetylglucosamine, "-4-β-D-GlupNAc-1-" , "-6-β-D-GalpNAc-1-" is a repeating unit that binds to the 4th and 1st positions of β-DN-acetylglucosamine, and "-6-β-D-GalpNAc-1-" is a repeating unit that binds to the 6th and 1st positions of β-DN-acetylgalactosamine. The unit "-4-β-D-GalpNAc-1-" is a repeating unit that bonds at the 4th and 1st positions of β-DN-acetylgalactosamine, and "-6-β-D-Fluf-1-" is The repeating unit "-4-β-D-Fluf-1-" which binds at the 6th and 1st positions of β-D-fructofuranose binds at the 4th and 1st positions of β-D-fructofuranose. Each repeating unit is indicated.

Y ¹ and Y ² represent a hydrogen atom, a monovalent hydrocarbon group, or a monovalent heteroatom-containing group. Examples of the monovalent hydrocarbon group include a linear or branched alkyl group, a linear or branched alkenyl group, a linear or branched alkynyl group, and the like. The number of carbon atoms in the alkyl group, alkenyl group, or alkyl group is not particularly limited, and may be, for example, 1 or more and 150 or less.

Specific examples of the linear alkyl group include methyl group, ethyl group, n-propyl group, n-butyl group, n-pentyl group, n-hexyl group, and the like. Specific examples of the branched alkyl group include methylpropyl group, methylbutyl group, methylpentyl group, dimethylbutyl group, ethylpropyl group, and ethylbutyl group.

The alkenyl group is not particularly limited as long as it has one or more carbon-carbon double bonds in the structure of the linear alkyl group and branched alkyl group, but specific examples include: Examples include ethenyl group, propenyl group, butenyl group, pentenyl group, hexenyl group and the like.

The alkynyl group is not particularly limited as long as it has one or more carbon-carbon triple bonds in the structure of the above-mentioned linear alkyl group and branched alkyl group, but specific examples include enityl Examples include groups.

Examples of the monovalent heteroatom-containing group include OH group, SH group, NH ₂ group, NO ₂ group, COOH group, N ₃ group, halogen atom, alkoxy group, alkylthio group, alkynyloxy group, and aryloxy group. , aralkyloxy group, aralkylthio group, succinimidyloxy group, maleimidyloxy group, monoalkylamino group, dialkylamino group, protected amino group, alkoxycarbonyl group, aryloxycarbonyl group, aralkyloxycarbonyl group, alkoxy Carbonyloxy group, aryloxycarbonyloxy group, aralkyloxycarbonyloxy group, acryloyloxy group, methacryloyloxy group, alkoxycarbonylamino group, aryloxycarbonylamino group, aralkyloxycarbonylamino group, carbamoyl group, monoalkylcarbamoyl group, dialkyl group Examples include carbamoyl group. When the heteroatom-containing group contains a carbon atom in addition to the heteroatom, the number of carbon atoms is not particularly limited, and may be, for example, 1 or more and 150 or less.

Examples of cyclic compounds and linear compounds include, but are not limited to, alkylene glycols (polyethylene glycol, polypropylene glycol, polybutylene glycol, block copolymers of ethylene glycol and propylene glycol, block copolymers of ethylene glycol and butylene glycol) ), polyglycerin, polyvinyl alcohol, polyvinyl ester (polyvinyl acetate, etc.), polyester (polylactic acid, polyglycolic acid, polybutylene succinate, polybutylene succinate adipate, polybutylene adipate terephthalate, polyethylene terephthalate succinate, polycaprolactone) , polyethylene terephthalate, polybutylene terephthalate, etc.), aromatic vinyl (polystyrene, etc.), acrylic resin (polymethyl methacrylate, etc.), polyamide (polyglycine, poly(β-alanine), polyamide 4, polyamide 6, polyamide 66, polyamide 610, polyamide 11, polyamide 12, aromatic polyamide), polysilane (dimethylpolysilane, methylphenylpolysilane, diphenylpolysilane, etc.), polysiloxane (polydimethylsiloxane, polymethylphenylsiloxane, polydiphenylsiloxane, etc.), polyurethane, oligopeptide, polypeptide Alternatively, examples include proteins, nucleic acids (DNA, RNA), sugar chains, and the like.

This mixture is then contacted with a separation material consisting of a porous channel material for separating or purifying cyclic compounds from the mixture. The contact between the above mixture and the porous channel material may be carried out without a solvent, but when carried out without a solvent, the linear compound is adsorbed onto the porous channel material, and then the cyclic compound existing outside the porous channel material is removed. It needs to be washed out and recovered in some way. In that case, there is a possibility that even some linear compounds may flow out from the porous channel material, which may result in a decrease in separation performance or a decrease in recovery efficiency. For this reason, contact between the mixture and the porous channel material is preferably carried out in a solvent.

When the separation or purification step is carried out in a solvent, examples of the solvent include water, lower alcohols such as methanol and ethanol, ketones such as acetonitrile, acetone, methyl ethyl ketone and methyl isobutyl ketone, ethers such as tetrahydrofuran and dioxane, Examples include, but are not limited to, esters such as ethyl acetate, chlorinated hydrocarbons such as methylene chloride, chloroform, and 1,2-dichloroethane, toluene, xylene, N,N-dimethylformamide, and dimethyl sulfoxide. These solvents may be used alone or in combination of two or more.

The adsorption power of linear compounds to porous channel materials differs depending on the type of solvent. For example, when N,N-dimethylformamide or chloroform is used as a solvent, the adsorption power of the linear compound to the porous channel material is low, so the linear compound is adsorbed to the porous channel material, but is then eluted. Therefore, in this case, when the mixture is contacted with the porous channel material, the linear compound elutes from the porous channel material after the cyclic compound elutes from the porous channel material.

Moreover, when ethanol is used as a solvent, the adsorption power of the linear compound to the porous channel material is high, so that the linear compound does not elute after being adsorbed to the porous channel material. Therefore, in this case, when the mixture is contacted with the porous channel material, the cyclic compounds elute from the porous channel material, but the linear compounds elute less from the porous channel material.

Therefore, when it is desired to elute a linear compound from a porous channel material, N,N-dimethylformamide or chloroform is preferable, and when it is desired to separate a linear compound using a porous channel material, ethanol is preferable. If the linear compound is adsorbed to the pores of the porous channel material and does not elute, the porous channel material may be destroyed or washed with N,N-dimethylformamide or chloroform. By doing so, linear compounds can be recovered. The porous channel material can be reused if the linear compounds can be recovered by washing.

When separating or purifying a cyclic compound using a solvent, the temperature of the separation or purification step may be a temperature above room temperature (23° C.) and below the boiling point of the solvent. The upper limit of the temperature in this case depends on the solvent used, but from the viewpoint of easy separation or purification, it is, for example, 220° C. or lower, preferably 150° C. or lower, and more preferably 80° C. or lower. When this separation or purification step is performed by liquid chromatography, it is also preferable to perform the separation or purification step at a temperature of 40° C. or higher and 80° C. or lower, from the viewpoint that the peak shape appears clearly.

When separating or purifying a cyclic compound without a solvent, the temperature of the separation or purification step is a temperature at which the porous channel material does not decompose, such as 500°C or lower, preferably 300°C or lower, more preferably 250°C or lower, and even more preferably 200°C or lower. below ℃. The higher the temperature in the separation or purification process, the lower the viscosity, which has the advantage of allowing separation to be carried out in a shorter time.

When a cyclic compound is to be separated or purified without a solvent, and the cyclic compound and linear compound are liquid at room temperature, they can be separated or purified using a porous channel material by mixing them. If the cyclic compound and linear compound to be separated or purified without a solvent are solid at room temperature, the cyclic compound can be separated or purified using a porous channel material by melting the cyclic compound by heating.

When using a solvent to separate or purify cyclic compounds, it is preferable to use 0.1 parts by mass or more of the solvent per 100 parts by mass of the mixture. If the amount of the solvent is 0.1 parts by mass or more, a decrease in separation performance or a decrease in recovery efficiency can be suppressed. The upper limit of the amount of the solvent used may be 20 parts by mass or less from the viewpoint of reducing separation costs.

Separation performance depends on the composition of the mixture. For example, if the mixture contains a large amount of linear compounds and a small amount of cyclic compounds (e.g., when the weight ratio of cyclic compounds in the mixture is 0.2 or less), the mixture may be used as a porous channel material since there are many linear compounds. If the porous channel material is brought into contact with the linear compound, the pores of the porous channel material may become clogged with the linear compound, which may reduce the separation performance. Therefore, in such cases, in order to improve the separation performance, for example, the concentration of the mixture can be diluted to 100 mg/mL or less with a solvent, or the amount of porous channel material can be increased to improve the separation performance. It is preferable that the weight ratio of the linear compound (linear compound/porous channel material) is 0.4 or less.

Preferably, the plurality of pores in the porous channel material are regularly arranged. Preferably, the pores are linear in order for the linear compound to enter the interior along the length direction. It is also preferable that the porous channel material can be designed to appropriately control properties such as pore size, shape, and internal functional groups.

The ratio of the pore diameter of the porous channel material to the maximum thickness of the linear compound (pore diameter of porous channel material/maximum thickness of the linear compound) is preferably 0.5 or more and 2.0 or less. If this ratio is 0.5 or more, linear compounds will easily enter the pores of the porous channel material, and if this ratio is 2.0 or less, cyclic compounds will more easily enter the pores. It's hard to do. Here, in this specification, "the linear compound enters into the pores of the porous channel material" means not only that the entire linear compound enters into the pores, but also that only a part of the linear compound enters the pores. It also includes entering into the pores. Therefore, even if the largest thickness part of the linear compound has not entered the pore, it is determined that the linear compound has entered the pore if a part other than the largest thickness has entered the pore. . Therefore, even if the pore diameter of the porous channel material is smaller than the maximum thickness of the linear compound, the linear compound may enter the pores. The lower limit of the above ratio is more preferably 0.7 or more, 0.9 or more, or 1.1 or more. The upper limit of this ratio is more preferably 1.8 or less, 1.6 or less, or 1.4 or less.

Some porous channel materials have pore sizes that do not change substantially, while others have pore sizes that change in relation to the linear compound. The pore size of such a porous channel material with variable pore size is widened by the intrusion of the linear compound. Therefore, in porous channel materials whose pore diameters change, "pore diameter" refers to the pore diameter of the porous channel material before the linear compound enters and the pore diameter of the porous channel material after the linear compound enters. shall mean the average value of pore diameter. An example of such a flexible porous channel material whose pore diameter changes is an interdigitated porous metal complex.

The pore diameter of the porous channel material can be determined, for example, as follows. First, based on the structural information (single-crystal X-ray structural analysis data) of porous channel materials reported in past papers, we use molecular simulation software (e.g., product name "BIOVIA Material Studio", Dassault Systèmes Biovia). The pore surface structure is calculated from the van der Waals radius of the atoms forming one pore skeleton using the Connoly Surface method. Then, the pore diameter of the pore cross section is determined from the obtained pore surface structure. In addition, in the case of a porous channel material in which the pore size changes, according to the above method, crystal structure information before the pore size changes and crystal structure information after the pore size changes can be obtained separately. Therefore, the pore diameter before the pore diameter changes and the pore diameter after the pore diameter change can be obtained separately, and the average value of the pore diameters can be determined from these.

The ratio of the pore diameter of the porous channel material to the diameter of the cyclic compound (pore diameter of the porous channel material/diameter of the cyclic compound) is preferably less than 1.0. If this ratio is 1.0 or less, it is more difficult for the cyclic compound to enter the pores. The upper limit of this ratio is preferably 0.9 or less, more preferably 0.8 or less.

The pore diameter of the porous channel material is not particularly limited as it depends on the maximum thickness of the linear compound that is desired to enter the pores, but may be 2 Å (0.2 nm) or more and 100 Å (10 nm) or less. . The lower limit of the pore diameter may be 2.5 Å (0.25 nm) or more, 3.0 Å (0.3 nm) or more, or 3.5 Å (0.35 nm) or more, and the upper limit of the pore diameter is 80 Å (8 nm) or more. ) or less, 50 Å (5 nm) or less, 30 Å (3 nm) or less, or 25 Å (2.5 nm) or less.

The shape of the porous channel material is not particularly limited, but from the viewpoint of improving separation efficiency, it is preferably powder-like. In this case, the average particle diameter is preferably 1 μm or more and 50 μm or less. If the average particle diameter is 1 μm or more, it is possible to prevent the pressure required to flow the mixture from becoming too high when a column is packed with a porous channel material, and if it is 50 μm or less, separation performance can be reduced. The decline can be suppressed. The lower limit of the average particle diameter is more preferably 2 μm or more, 3 μm or more, or 4 μm or more, and the upper limit is more preferably 40 μm or less, 30 μm or less, or 20 μm or less. The average particle diameter can be measured using a particle size distribution analyzer.

The weight ratio of the mixture to the porous channel material (mixture/porous channel material) in the separation or purification step may be 0.05 or more and 300 or less, or 0.1 or more and 100 or less.

Porous channel materials include porous metal complexes, covalent organic frameworks (COFs), and organic cage compounds. Porous channel materials have multiple pores that are large enough for linear compounds to enter but not cyclic compounds to enter.

The porous metal complex is a metal complex that has a porous three-dimensional structure due to transition metal ions and organic bridging ligands connecting them, and is preferably composed of a transition metal cation and a first organic bridging ligand. Two-dimensional sheets form layers, and a second organic bridging ligand capable of bidentate coordination coordinates with the transition metal cation present in each layer to connect adjacent sheets, forming pores between them. Coordination polymers having a structure shown in FIG. Porous metal complexes broadly encompass porous metal complexes known as MOFs, PCPs, and the like. The pores of the porous metal complex may be one-dimensional pores, two-dimensional pores, or three-dimensional pores, but one-dimensional pores are preferred from the viewpoint of improving separation performance.

Separation or purification of a cyclic compound and a linear compound using a porous channel material is performed by the interaction of the cyclic compound and the linear compound with the pores of the porous channel material. More specifically, the mixture is separated into linear compounds that enter the pores of the porous channel material and cyclic compounds that cannot enter the pores of the porous channel material.

The separation or purification of cyclic compounds is important in terms of the relationship between the size (bulk) of the cyclic compounds, the maximum thickness of the linear compounds, and the pore size of the porous channel material, but by modifying the inside of the pores with appropriate functional groups, it is possible to adjust the ease and speed at which the linear compounds enter the pores. For example, in the case of MOFs, modification of the inside of the pores can be achieved by introducing appropriate substituents to the ligands, and in the case of porous channel materials other than MOFs, it can be similarly achieved by introducing substituents to the components.

Since hydrophilic or hydrophobic interactions between the linear compound and the porous channel material also contribute to the separation or purification of the cyclic compound, such interactions also influence the selection/combination of the porous channel material and the linear compound. This is one of the deciding factors.

The porous metal complex may include a monodentate organic ligand along with the first organic bridging ligand. The size of the complex crystals can be adjusted by adding the monodentate organic ligand.

The transition metal ions constituting the porous metal complex include metal ions of metals belonging to Groups 1 to 12 of the periodic table, specifically gold, platinum, silver, copper, ruthenium, tin, palladium, rhodium, iridium, Ions include osmium, nickel, cobalt, zinc, iron, yttrium, magnesium, manganese, titanium, zirconium, hafnium, calcium, cadmium, vanadium, chromium, molybdenum, scandium, and magnesium, calcium, manganese, iron, ruthenium, Ions such as cobalt, rhodium, nickel, palladium, copper, zinc, cadmium, titanium, vanadium, chromium, manganese, platinum, molybdenum, zirconium, and scandium are preferred, and manganese, iron, cobalt, nickel, copper, zinc, silver, and platinum are preferred. , palladium, ruthenium, rhodium, and other metal ions.

Among the organic bridging ligands, the first organic bridging ligands include benzene, naphthalene, anthracene, phenanthrene, fluorene, indane, indene, pyrene, 1,4-dihydronaphthalene, tetralin, biphenylene, triphenylene, acenaphthylene, and acenaphthene. A compound in which 2, 3 or 4 carboxyl groups are bonded to an aromatic ring such as an acylamino group, a cyano group, a hydroxyl group, a methylenedioxy group, an ethylenedioxy group, a methoxy group, an alkoxy group having 1 to 4 carbon atoms having a linear or branched chain such as an ethoxy group, a methyl group, an ethyl group, a propyl group, Straight-chain or branched alkyl groups having 1 to 4 carbon atoms such as tert-butyl group and isobutyl group, SH group, trifluoromethyl group, sulfonic acid group, carbamoyl group, alkylamino group such as methylamino group, dimethylamino group cyclic saturated aliphatic polycarboxylic acid compounds having 5 to 12 carbon atoms (for example, 1,2-cis-cyclo Propanedicarboxylic acid, 1,2-trans-cyclopropanedicarboxylic acid, 1,3-cis-cyclobutanedicarboxylic acid, 1,3-trans-cyclobutanedicarboxylic acid, 1,4-cis-cyclohexanedicarboxylic acid, 1,4-trans -cyclohexanedicarboxylic acid and 1,3-adamantanedicarboxylic acid), (1α,2α,4α)-1,2,4-cyclohexanetricarboxylic acid, fumaric acid, maleic acid, citraconic acid, itaconic acid, etc. Examples include carboxylic acids, preferably isophthalic acid, 5-methoxyisophthalic acid, 5-methylisophthalic acid, 5-fluoroisophthalic acid, 5-chloroisophthalic acid, 5-bromoisophthalic acid, 5-iodoisophthalic acid, 5-iodoisophthalic acid, -nitroisophthalic acid and 5-cyanoisophthalic acid, terephthalic acid (tp), 2-methylterephthalic acid, 2-methoxyterephthalic acid, 2-nitroterephthalic acid, dihydrocyclobuta[1,2-b]terephthalic acid, 4, 4'-dicarboxydiphenyl sulfone, 2,6-naphthalenedicarboxylic acid, 9,10-anthracenedicarboxylic acid, 2,3-pyrazinedicarboxylic acid (pzdc), tetrafluoroterephthalic acid, 4,4'-bibenzoic acid, octa Fluoro-4,4'-bibenzoic acid, 4,4'-biphenyldicarboxylic acid, 2,7-fluorenedicarboxylic acid, 2,7-pyrenedicarboxylic acid, 4,5,9,10-tetrahydropyrene-2,7 -Dicarboxylic acids such as dicarboxylic acid, succinic acid, maleic acid, fumaric acid, and acetylene digarboxylic acid.

The complex may further include a monodentate organic ligand, such as a monocarboxylic acid, in combination with the first organic bridging ligand. Examples of monocarboxylic acids include formic acid, acetic acid, trifluoroacetic acid, propionic acid, lactic acid, pyruvic acid, butanoic acid, pentanoic acid, hexanoic acid, and cyclohexanecarboxylic acid. By using monodentate ligands, the size of the porous metal complex can be reduced and the length of the pores can be shortened.

Among the organic bridging ligands, examples of the second organic bridging ligand include pyrazine, trans-1,2-bis(4-pyridyl)ethylene, 1,4-dicyanobenzene, and 4,4'-dicyanobiphenyl. , 1,2-dicyanoethylene, 1,4-bis(4-pyridyl)benzene, triethylenediamine (ted), 4,4'-bipyridyl (bpy), diazapyrene, 2,5-dimethylpyrazine, 2,2'- Dimethyl-4,4'-bipyridine, 1,2-bis(4-pyridyl)ethyne, 1,4-bis(4-pyridyl)butadiine, 1,4-bis(4-pyridyl)benzene, 3,6-di (4-pyridyl)-1,2,4,5-tetrazine, 2,2'-bi-1,6-naphthyridine, phenazine, 2,6-di(4-pyridyl)-benzo[1,2-c: 4,5-c']dipyrrole-1,3,5,7(2H,6H)-tetrone, N,N'-di(4-pyridyl)-1,4,5,8-naphthalenetetracarboxydiimide, trans -1,2-bis(4-pyridyl)ethene, 4,4'-azopyridine, 1,2-bis(4-pyridyl)ethane, 4,4'-dipyridyl sulfide, 1,3-bis(4-pyridyl) Examples include, but are not limited to, propane, 1,2-bis(4-pyridyl)-glycol, N-(4-pyridyl)isonicotinamide, and the like.

Porous metal complexes are described, for example, in the following literature and reviews (Angew. Chem. Int. Ed. 2004, 43, 2334-2375.; Angew. Chem. Int. Ed. 2008, 47, 2-14.; Chem. Soc Rev., 2008, 37, 191-214.; PNAS, 2006, 103, 10186-10191.; Chem.Rev., 2011, 111, 688-764.; Nature, 2003, 423, 705-714.) etc. However, the present invention is not limited to these, and a wide variety of known porous metal complexes or porous metal complexes that can be produced in the future can be used. Examples of porous metal complexes include, but are not limited to, CD-MOF-1, CD-MOF-2, CD-MOF-3, CPM-13, FJI-1, FMOF-1, HKUST-1, IRMOF- 1, IRMOF-2, IRMOF-3, IRMOF-6, IRMOF-8, IRMOF-9, IRMOF-13, IRMOF-20, JUC-48, JUC-62, MIL-101, MIL-100, MIL-125, MIL-53, MIL-88 (including MIL-88A, MIL-88B, MIL-88C, MIL-88D series), MOF-5, MOF-74, MOF-177, MOF-210, MOF-200, MOF- 205, MOF-505, MOROF-2, MOROF-1, NOT-100, NOT-101, NOT-102, NOT-103, NOT-105, NOT-106, NOT-107, NOT-109, NOT-110, NOTT-111, NOT-112, NOT-113, NOT-114, NOT-140, NU-100, rho-ZMOF, PCN-6, PCN-6', PCN9, PCN10, PCN12, PCN12', PCN14, PCN16, PCN-17, PCN-21, PCN46, PCN66, PCN68, PMOF-2(Cu), PMOF-3, SNU-5, SNU-15', SNU-21S, SNU-21H, SNU-50, SNU-77H, UiO-66, UiO-67, soc-MOF, sod-ZMOF, TUDMOF-1, UMCM-2, UMCM-150, UTSA-20, ZIF-2, ZIF-3, ZIF-4, ZIF-8, ZIF- 9, ZIF-10, ZIF-11, ZIF-12, ZIF-14, ZIF-20, ZIF-21, ZIF-23, ZIF-60, ZIF-61, ZIF-62, ZIF-64, ZIF-65, ZIF-67, ZIF-68, ZIF-69, ZIF-70, ZIF-71, ZIF-72, ZIF-73, ZIF-74, ZIF-75, ZIF-76, ZIF-77 or ZIF-90 etc. It will be done.

Examples of porous metal complexes and their pore diameters and pore cross-sectional areas are shown below.
IRMOF-1: Zn ₄ O(BDC) ₃ , BDC = 1,4-benzenedicarboxylate, pore diameter = 15.0 Å, pore cross-sectional area = 15.0 × 15.0 Å ²
MOF-69C: Zn ₃ (OH) ₂ (BDC) ₂ , pore diameter = 6.5 Å, pore cross-sectional area = 6.5 × 6.5 Å ² ,
MOF-74: _M2 (DOBDC), DOBDC=2,5-dihydroxyterephthalate, M=Zn, Co, Ni, Mg, pore diameter=11.0Å, pore cross-sectional area=11.0× ^11.0Å2 ,
HKUST-1: Cu ₃ (BTC) ₂ , BTC=1,3,5-benzenetricarboxylate, pore diameter = 9.0 Å, pore cross-sectional area = 9.0×9.0 Å ² ,
MOF-177: Zn ₄ O(BTB) ₂ , BTB=4,4',4"-benzene-1,3,5-triyl-tribenzoate, pore diameter=11.8Å, pore cross-sectional area=11.8×11.8Å ² ,
MOF-508: Zn(BDC)(bipy) _0.5 , pore diameter = 4.0Å, pore cross section = 4.0×4.0Å ² ,
Zn-BDC-DABCO: Zn ₂ (BDC) ₂ (DABCO), DABCO=1,4-diazabicyclo[2.2.2]-octane, pore diameter = 7.5 Å, pore cross-sectional area = 7.5 × 7.5 Å ² ,
Cr-MIL-101: Cr ₃ F(H ₂ O) ₂ O(BDC) ₃ , pore diameter = 29.0 Å, pore cross-sectional area = 29.0 × 29.0 Å ² ,
Al-MIL-110: Al ₈ (OH) ₁₂ [(OH) ₃ (H ₂ O) ₃ ][BTC] ₃ , pore diameter = 16.0 Å, pore cross-sectional area = 16.0 × 16.0 Å ² ,
MIL-103: M(BTB), M=light rare-earth elements [La-Ho], pore diameter=10.7Å, pore cross-sectional area=10.7×10.7Å ² ,
Al-MIL-53: Al(OH)[BDC], pore diameter = 8.5 Å, pore cross-sectional area = 8.5 × 8.5 Å ² ,
ZIF-8: Zn(MeIM) ₂ , MeIM=2-methylimidazole, pore diameter = 12.0 Å, pore cross section = 12.0×12.0 Å ² ,
MIL-88B: Cr ₃ OF(O ₂ CC ₆ H ₄ -CO ₂ ) ₃ , pore diameter = 15.6 Å, pore cross-sectional area = 15.6×15.6 Å ² ,
MIL-88C: Fe ₃ O(O ₂ CC ₁₀ H ₆ -CO ₂ ) ₃ , pore diameter = 18.7 Å, pore cross section = 18.7×18.7 Å ² ,
MIL-88D: _Cr3OF ( _{O2CC12H8 - CO2} ) _{3 ,} pore diameter=20.5Å, pore cross-sectional area=20.5×20.5Å2 ^,
CID-1[Zn ₂ (ip) ₂ (bpy) ₂ ] _n , ip=isophthalic acid, bpy=4,4'-bipyridine, pore diameter=5.0 Å, pore cross-sectional area=5.0×6.0 Å ² ,
[ZrO(bpdc)] _n , bpdc = 4,4'-biphenyl dicarboxylate, pore diameter = 6.4 Å, pore cross-sectional area = 6.4 × 6.4 Å ² ,
[Zn ₂ (ndc) ₂ (dabco)] _n , ndc=1,4-naphthalene dicarboxylate, pore diameter=5.7Å, pore cross-sectional area=5.7×5.7Å ²
[Al(OH)(ndc)] _n , ndc=2,6-naphthalene dicarboxylate, pore diameter=8.5Å, pore cross-sectional area=8.5×8.5Å ²

Examples of covalent organic frameworks (COFs) include those obtained by condensing derivatives such as diboronic acid, hexahydroxytriphenylene, dicyanobenzene, C ₉ H ₄ BO ₂ , and benzene-1,4- Diboronic acid (BDBA), 2,3,6,7,10,11-hexahydroxyltriphenylene (HHTP), tetrakis(4-bromophenyl)methane, tetrakis(4-ethynylphenyl)methane (TEPM), 1,3, Examples include those obtained from 5,7-tetrakis(4-ethynylphenyl)adamantine (TEPA) and 1,3,5,7-tetrakis(4-bromophenyl)adamantine (TBPA). In a preferred embodiment of the present disclosure, the covalent organic framework (COF) includes COF-1 (pore diameter = 9.0 Å, pore cross-sectional area = 9.0 × 9.0 Å ² ), COF-2, COF-5 ( Pore diameter = 27.0 Å, pore cross-sectional area = 27.0 × 27.0 Å ² ), COF-6 (pore diameter = 9.0 Å, pore cross-sectional area = 9.0 × 9.0 Å ² ), COF-8 (pore diameter = 16.0 Å, Pore cross-sectional area = 16.0 × 16.0 Å ² ), COF-10 (pore diameter = 32.0 Å, pore cross-sectional area = 32.0 × 32.0 Å ² ), COF-11 (pore diameter = 11.0 Å, pore cross-sectional area = 11.0 ×11.0 Å ² ), COF-14 (pore diameter = 14.0 Å, pore cross-sectional area = 14.0 × 14.0 Å ² ), COF-16 (pore diameter = 16.0 Å, pore cross-sectional area = 16.0 × 16.0 Å ² ), COF-18 (pore diameter = 18.0 Å, pore cross-sectional area = 18.0 × 18.0 Å ² ), COF-42 (pore diameter = 23.0 Å, pore cross-sectional area = 23.0 × 23.0 Å ² ), COF-43 (pore diameter = 38.0 Å, pore cross-sectional area = 38.0 × 38.0 Å ² ), COF-66 (pore diameter = 23.0 Å, pore cross-sectional area = 23.0 × 23.0 Å ² ), COF-102 (pore diameter = 12.0 Å, pore Cross-sectional area = 12.0 × 12.0 Å ² ), COF-103 (pore diameter = 12.0 Å, pore cross-sectional area = 12.0 × 12.0 Å ² ), COF-105 (pore diameter = 10.3 Å, pore cross-sectional area = 10.3 × 10.3 Å ² ), COF-108 (pore diameter = 15.5 Å, pore cross-sectional area = 15.5 × 15.5 Å ² ), COF-202 (pore diameter = 11.0 Å, pore cross-sectional area = 11.0 × 11.0 Å ² ), COF- Examples include ^, but are not limited to ^: .

Examples of organic cage compounds include condensation reaction products of 1,3,5-triformylbenzene and various diamine compounds described in Cooper et al., J. Am. Chem. Soc., 134, 588 (2012), and solid state compounds. Cage-like imines exhibiting porosity (Tozawa, T. et al., Porous organic cages. Nat. Mater. 8, 973-978 (2009)), porous materials that are liquid at room temperature (James, S. L. and coworker, Chem. Sci. 2012, 3, 2153), porous organic molecular crystals (Jones, J. T. A. et al., Nature 474, 367-371 (2011)), and compounds using crown ethers with closed loops as substituents (Giri, N. et al., Nature 2015, 527, 216), but are not limited to these.

The separation material may be used by being filled into a separation device. The separation device can be used by being incorporated into a separation device.

<Separation equipment and separation device>
The separation device includes a casing having an inlet and an outlet, and the separation material filled in the casing. The inlet of the casing may be provided at one end of the casing in the longitudinal direction, and the outlet of the casing may be provided at the other end of the casing in the longitudinal direction.

When separating a cyclic compound using a separation device, the mixture is introduced into the housing through the inlet, and the cyclic compound discharged from the outlet is collected. Cyclic compounds cannot penetrate into the porous channel material and therefore elute faster than linear compounds. Examples of separation instruments include columns, and examples of separation devices include liquid chromatographs.

Upon contacting the mixture with the porous channel material, the linear compound enters the pores of the porous channel material and is adsorbed. On the other hand, since cyclic compounds are bulkier than linear compounds, they are less likely to enter the pores of a porous channel material than linear compounds. This is thought to be because, in the case of a linear compound, only one compound has to enter the pore, but in the case of a cyclic compound, two compounds have to enter at the same time. Therefore, the cyclic compound does not enter the pores of the porous channel material, but exists outside the porous channel material. Therefore, for example, when the above separation device is used, cyclic compounds are eluted earlier than linear compounds. Thereby, the cyclic compound can be separated or purified from the mixture. The pore diameter of the porous channel material can be freely designed by changing the length of the organic crosslinking ligand, etc., so the pore diameter can be adjusted such that linear compounds can enter, but cyclic compounds cannot or have difficulty entering. By adjusting the size, various cyclic compounds can be separated or purified regardless of their size. Thereby, the cyclic compound can be separated or purified with high separation efficiency and independent of the size of the cyclic compound.

The reprecipitation method requires complicated steps. On the other hand, the separation or purification method of the present invention is a much simpler method than the reprecipitation method because it is sufficient to prepare a porous channel material.

EXAMPLES In order to explain the present invention in detail, Examples will be given and explained below, but the present invention is not limited to these descriptions. FIG. 1 is a graph showing the analysis results of the crude product used in Example 1 by gel permeation chromatography (GPC), and FIG. FIG. 3 is a graph showing the analysis results of MOF1 synthesized in Example 1 by powder X-ray diffraction (XRPD). FIG. 4(A) is a graph showing the analysis results of a composite of linear PEG and MOF1 by differential scanning calorimetry (DSC), and FIG. 4(B) is a graph showing the differential scanning calorimetry of a composite of cyclic PEG and MOF1. FIG. 5(A) is a graph showing a simulation of introducing linear PEG into MOF1, and FIG. 5(B) is a graph showing a simulation of introducing cyclic PEG into MOF1. It is a diagram. FIG. 6 is a graph showing the analysis results of linear PEG and cyclic PEG with molecular weights of 1000, 2000, 3000, and 10000 by high performance liquid chromatography (HPLC) using a column packed with MOF1. FIG. 7 is a graph representing a gradient elution program. FIG. 8(A) is a graph showing the analysis results of the components of fraction 1 by carbon-13 nuclear magnetic resonance ( ¹³ C-NMR), and FIG. 8(B) is a graph showing the analysis results of the components of fraction 2 by carbon-13 nuclear magnetic resonance ( 13 C-NMR). 13 is a graph showing the analysis results by ¹³ C-NMR). FIG. 9 is a graph showing the analysis results of the crude product, fraction 1 components, and fraction 2 components by gel permeation chromatography (GPC), and FIG. 10(B) is an enlarged view of the vicinity of 2000 m/z in FIG. 10(A), and FIG. C) is an enlarged view around 4000 m/z in FIG. 10(A). FIG. 11(A) is a graph showing the analysis results of the components of fraction 2 by matrix-assisted laser desorption ionization time-of-flight mass spectrometer (MALDI-TOF-MS), and FIG. It is an enlarged view of the vicinity of 2000 m/z in A). FIG. 12 is a graph showing the analysis results of Compound A obtained in Comparative Example 1 by gel permeation chromatography (GPC), and FIG. 13 is a graph showing the analysis results of Compound A by carbon-13 nuclear magnetic resonance (13C-NMR). This is a graph representing

<Example 1>
First, cyclic PEG and MOF were synthesized as follows.
(Synthesis of cyclic PEG)
First, according to the method described in J. Cooke, K. Viras, GE Yu, T. Sun, T. Yonemitsu, AJ Ryan, C. Price and C. Booth, Macromolecules, 1998, 31, 3030-3039. A cyclization reaction was performed using linear PEG as a raw material to synthesize cyclic PEG.

Specifically, first, a round-bottomed flask containing a stirring bar was sealed with a septum, and the flask was filled with nitrogen while removing air from the flask using a vacuum pump. In this flask, well-pulverized potassium hydroxide (2.2 g, 33.3 mmol) was dispersed in 100 mL of a solvent consisting of a mixture of tetrahydrofuran and heptane at a volume ratio of 3:1, and the mixture was stirred at 40° C. under a nitrogen atmosphere. . Heptane, which is a poor solvent, was added to bring the ends of the linear PEG molecules closer together to facilitate ring formation. Linear PEG with a molecular weight of 2050 (6.83 g, 3.33 mmol) and para-toluenesulfonyl chloride (635 mg, 3.33 mmol) were dissolved in 100 mL of tetrahydrofuran in a separate flask. PEG had a maximum thickness of 3.7 Å. The maximum thickness of PEG was determined by calculation based on the most stable structure obtained by computational chemistry. Specifically, first, the structure of the molecular structure model of PEG was calculated by the conjugate gradient method using a universal force field using molecular simulation software (product name "BIOVIA Material Studio", manufactured by Dassault Systèmes Biovia). Optimization calculations were performed. Then, the van der Waals radius was calculated for the molecular structure of one repeating unit including atoms present at the ends of the structure-optimized PEG, and its maximum cross-sectional diameter (maximum thickness) was determined.

This solution was added dropwise to the potassium hydroxide dispersion using a syringe pump over 48 hours. After stirring for an additional 24 hours, the mixture was filtered and the solvent was removed by rotary evaporation and vacuum pump.

The resulting mixture was dissolved in about 100 mL of water, and excess potassium hydroxide was neutralized with hydrochloric acid. The solution was concentrated to about 50 mL using a rotary evaporator, 50 mL of dichloromethane was added, and impurities such as PEG and salt were separated using a separatory funnel. When the dichloromethane solution was added dropwise to diethyl ether using a syringe, PEG precipitated. The solvent was removed by filtration and a vacuum pump to obtain a crude product. Analysis by gel permeation chromatography (GPC) confirmed that this crude product was a mixture of unreacted linear PEG, a linear high molecular weight compound in which linear PEG was condensed in series, and cyclic PEG. (Figure 1). Since the maximum thickness of the linear PEG is 3.7 Å, the diameter of the circular PEG is estimated to be 7.4 Å or more.

(Synthesis of MOF1)
We synthesized [Zn ₂ (ndc) ₂ (dabco)] _n (hereinafter, this compound is referred to as MOF1). The structure of MOF1 is shown in Figure 2. This MOF1 was synthesized based on previous literature. The reagents used are as follows.
zinc nitrate tetrahydrate 1,4-naphthalenedicarboxylic acid 1,4-diazabicyclo[2.2.2]octane (DABCO)
N,N-dimethylformamide (DMF)

Zinc nitrate tetrahydrate, DABCO, and DMF were purchased from Fuji Film Wako Pure Chemical Industries, Ltd., and 1,4-naphthalene dicarboxylic acid (1,4-ndc) was purchased from Tokyo Chemical Industry Co., Ltd.

First, synthesis method 1, which is a normal synthesis method, will be described. Dissolve 4 mmol of zinc nitrate tetrahydrate, 4 mmol of 1,4-naphthalene dicarboxylic acid, and 2 mmol of DABCO in 40 mL of DMF, transfer to a 100 mL Teflon (registered trademark) heat-resistant container, attach a stainless steel jacket to the heat-resistant container, and heat the mixture in an oven. The solution was heated to 120°C for 48 hours. After the reaction, white powder was collected by centrifugation and washed several times with DMF. After washing, the white powder was vacuum dried for 12 hours while heating at 120° C. to obtain MOF1.

In producing the MOF column described below, a large amount of MOF with smaller and more uniform particle size was required. Therefore, mass synthesis was performed using the following method (synthesis method 2). In a 1 L glass heat-resistant container, 40 mmol of zinc nitrate tetrahydrate, 40 mmol of 1,4-naphthalene dicarboxylic acid, and 20 mmol of DABCO were dissolved in 600 mL of DMF. Then, this solution was heated in an oil bath at 120° C. for 48 hours while stirring. The subsequent treatments were performed in the same manner as in Synthesis Method 1. The particle size of the obtained MOF1 was about 3 to 10 μm.

The X-ray powder diffraction (XRPD) pattern of MOF1 synthesized by Synthesis Method 1 and Synthesis Method 2 matched the simulation pattern based on the previously reported single crystal structure, indicating that the desired structure was obtained. This was confirmed (see Figure 3).

When the pore diameter and pore cross-sectional area of MOF1 were determined, the pore diameter was 5.7 Å, and the pore cross-sectional area was 5.7×5.7 Å ² . The pore diameter and pore cross-sectional area were determined by calculations based on the unit crystal structure. Specifically, based on the structural information of MOF1 (single crystal X-ray structural analysis data) reported in past papers, we used molecular simulation software (product name "BIOVIA Material Studio", manufactured by Dassault Systèmes Biovia). The pore surface structure was calculated using the Connoly Surface method from the van der Waals radius of atoms forming one pore skeleton. Then, the pore diameter and pore cross-sectional area of the pore cross section were determined from the obtained pore surface structure.

Then, three types of tests were conducted using the above MOF1.
(1) PEG introduction test into MOF1 It was verified whether linear PEG and cyclic PEG were adsorbed into the pores of MOF1. Complexes of MOF1 and linear PEG and cyclic PEG with molecular weights of 1000 and 2000 were created according to the method published in a previous paper (Le Ouay, B. et al. Nat Communs. 9, 3635 (2018).). Specifically, MOF1 and PEG were suspended and dissolved in 2 mL of acetonitrile, and the acetonitrile was evaporated under reduced pressure (0.3 kPa) using a vacuum pump. Thereafter, it was heated under reduced pressure at 70°C for 12 hours. The masses of MOF and PEG used are shown in Table 1. The amount of PEG was adjusted to about 6 to 10% by mass based on MOF1. Note that MW in Table 1 is molecular weight.

Regarding the prepared MOF1-PEG complex, whether or not PEG existed outside the MOF was examined using a differential scanning calorimeter (DSC, product name "DSC7020", manufactured by Hitachi High-Tech Science Co., Ltd.). Usually, when DSC measurement is performed on PEG, an endothermic peak is observed at the melting point of PEG. However, it is known that when PEG is adsorbed into the pores of MOF1, the PEG chains are trapped in the pores in a state close to a single molecule, resulting in a change in the melting point (Uemura, T. et al. Nat Commun. 1, 83 (2010)). Utilizing this phenomenon, it is possible to determine whether PEG has been introduced into the pores of MOF1 by DSC measurement of the composite. In this measurement, DSC measurements of the MOF1-PEG complex were performed, focusing on the temperature range of 15°C to 55°C.

If PEG is not adsorbed into the pores of MOF1, PEG will be left outside the MOF crystal. Therefore, when the DSC of the composite is measured, an endothermic peak is observed near the melting point of bulk PEG (original PEG). On the other hand, when PEG is completely adsorbed into the pores of MOF1, the melting point peak moves significantly to lower temperatures, so no melting point is observed in the current temperature range.

When DSC was measured for a complex of linear PEG with a molecular weight of 2000 and MOF1, as shown in FIG. 4(A), a melting point peak derived from bulk PEG like Reference 1 was not observed. This indicates that the linear PEG was completely introduced into the pores and adsorbed. On the other hand, in the complex with cyclic PEG, a melting point peak is observed as shown in Figure 4(B), indicating that a part of the cyclic PEG remains outside the MOF1 crystal (that is, is not adsorbed). It has been shown. That is, it has been revealed that when MOF1 is used, linear PEG is selectively adsorbed into the pores, so that the remaining cyclic PEG can be separated or purified by recovering it by a washing operation.

In addition, molecular dynamics simulation was performed to verify whether PEG was incorporated into the MOF1 pores. FIG. 5 shows the results of simulating the introduction process into MOF1 using linear PEG and cyclic PEG with a molecular weight of 1000 as models. In linear PEG, the diffusion is three-dimensional, frequently penetrating the narrow pores between the pores, whereas in the case of circular PEG, the diffusion is one-dimensional, only moving along the pores. It was confirmed that linear PEG diffused into the pores at a faster rate. That is, it has become clear that linear and cyclic PEGs have different rates of penetration into pores and adsorption.

(2) Separation of cyclic PEG by MOF column chromatography (creation of HPLC column packed with MOF)
0.8 g of the synthesized MOF1 powder was packed into an empty stainless steel column (inner diameter 4 mm, length 150 mm) manufactured by GL Science Co., Ltd. by the tapping method to produce a column filled with MOF1.

HPLC Analysis of Linear and Cyclic PEGs
The column prepared above was connected to a Prominence HPLC system manufactured by Shimadzu Corporation, and high-purity linear PEG and cyclic PEG (molecular weights of 1000, 2000, 3000, and 10000) were analyzed as standards under the following analytical conditions.
[HPLC analysis conditions]
・Column stationary phase: 2
Eluent: DMF
Column temperature: 80°C
Column pressure: 2.6 MPa
Sample concentration: 5.0 mg/mL
Flow rate: 1.0 mL/min Injection volume: 1.0 μL

As shown in FIG. 6, cyclic PEG was hardly retained on the column regardless of its molecular weight, and eluted in 1.4 minutes. On the other hand, linear PEG was observed to be retained depending on its molecular weight, and eluted at 2.0 minutes (molecular weight 1000), 2.6 minutes (molecular weight 2000), and 4.2 minutes (molecular weight 3000), respectively. Linear PEG with a molecular weight of 10,000 was strongly retained in the column and was not eluted. In the case of linear PEG, there was a tendency for the retention time in the column to become longer and the peak shape to become broader as the molecular weight increased. This is thought to be because as the molecular weight increases, the adsorption to the stationary phase becomes stronger. From this result, it was confirmed that cyclic PEG and linear PEG could be separated by using a column packed with MOF1.

(3) Separation of cyclic PEG using medium pressure preparative column First, a medium pressure preparative column was prepared. Specifically, MOF1 powder synthesized in large quantities while stirring and immersed in DMF was wet-packed into a glass column (inner diameter 20 mm, length 150 mm) manufactured by KYOSHIN Co., Ltd. The weight of the packing (MOF1 and DMF) was 54 g, and the column volume was 47 mL. Wet filling was performed by flowing DMF at a flow rate of 15 mL/min until the filler was tightly packed.

Then, we attempted to purify the crude product using the medium pressure preparative column prepared above. For medium pressure preparative column chromatography, we used Biotage's Isolera flash automatic purification system. The flow rate was 3.0 mL/min.

A solution of 0.050 g of the crude product in 2 mL of DMF was injected into the top of the column using a syringe, and gradient elution was performed using the program shown in FIG. This program first elutes cyclic PEG by flowing 3 column volumes (CV) of DMF, then switches the eluent to chloroform over 3 CV, and finally flows 3 CV of chloroform to elute the cyclic PEG. It was set to wash out the linear PEG.

The eluted solutions were collected as fraction 1, in which the component eluted with 100% DMF, and fraction 2, the component eluted with 100% chloroform. The solvent of each fraction is removed using a rotary evaporator, and the resulting compounds are measured by GPC, ¹³ C-NMR, and matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS). The ingredients were analyzed using this method.

Furthermore, the results of ¹³ C-NMR measurement are shown in FIG. As shown in Figure 8 (A), only one peak derived from the methylene carbon of the PEG main chain was observed at 70.6 ppm from the components of fraction 1, so only cyclic PEG was present in fraction 1. It was confirmed that On the other hand, as shown in FIG. 8(B), from the components of fraction 2, peaks originating from the terminal carbon characteristic of linear PEG were confirmed around 62 ppm and around 73 ppm. It was confirmed that PEG was present.

In order to analyze the components contained in fraction 1 and fraction 2, each was analyzed by GPC. The results are shown in FIG. As shown in Figure 9, multiple peaks are observed in the GPC chromatogram of fraction 1, and the retention time of the largest peak (20 minutes) is the same as that of the standard sample of cyclic PEG (molecular weight 2000). This compound can be identified as cyclic PEG. Furthermore, peaks of cyclic dimers and cyclic trimers with larger molecular weights were also observed.

On the other hand, as shown in FIG. 9, fraction 2 contained a large amount of high molecular weight components. The component with the lowest molecular weight eluting at 19.6 minutes had the same retention time as linear PEG (molecular weight 2000), so it was identified as the raw material linear PEG. High molecular weight components such as molecular weights of 4000, 6000, etc., which are the linear condensation of the raw material linear PEG, were also observed.

Furthermore, the results of MALDI-TOF-MS measurement of the components of fraction 1 are shown in FIG. As shown in FIG. 10(A), a distribution of signals was observed in fraction 1 around 2000 m/z, 4000 m/z, and 6000 m/z. When magnifying around 2000 m/z and 4000 m/z, it can be seen that signals are observed every 44 m/z, which corresponds to the molecular weight of the repeating unit of PEG (-CH ₂ CH ₂ O-) (Figure 10 (B) and FIG. 10(C)). One of the signals was 2004.8 m/z (position marked ▼ in FIG. 10(B)). Based on this calculation, it is considered that this peak corresponds to 44.0×n (degree of polymerization of PEG=45)+23(Na+adduct ion)=2003. If a terminus is present, all peaks should shift by a molecular weight of 18, which corresponds to H ₂ O. Therefore, from this measurement result, it became clear that only cyclic PEG was present in fraction 1. Furthermore, signals of cyclized dimer and cyclized trimer were also observed at 4000 m/z and 6000 m/z on the high molecular weight side. That is, it can be concluded that only cyclic PEG was selected by the MOF1 column regardless of its molecular weight and eluted in fraction 1.

On the other hand, the results of MALDI-TOF-MS measurement of the components of fraction 2 are shown in FIG. In addition, the peak at the position ^▼ on the left side of FIG. /z (n=45, Na ⁺ adduct ion derived from linear PEG). As shown in FIG. 11(A), a group of peaks corresponding to 44×n+23+18 were observed in fraction 2, centered around the molecular weight of 2000. These are the peaks corresponding to linear PEG. In fraction 2, a small amount of peak derived from cyclic PEG was also detected (Fig. 11(B)), so it is thought that a small amount of cyclized product is mixed. Since no peak derived from the cyclic monomer is observed in 9), it can be understood that the amount of cyclic PEG mixed into fraction 2 is extremely small.

<Comparative example 1>
In Comparative Example 1, an attempt was made to separate and purify only cyclic PEG from the crude product by a reprecipitation method. In this method, taking advantage of the fact that cyclic PEG has slightly higher solubility in organic solvents than linear PEG, the crude product is added to a mixed solution of toluene (a good solvent for PEG) and heptane (a poor solvent for PEG). was heated and dissolved, and during the cooling process, the precipitated substances (mainly consisting of linear PEG) were removed and separated. The resulting remaining solution contains a large amount of cyclic PEG.

First, the crude product (mixture of cyclic PEG and linear PEG) obtained by the method described in Example 1 was dissolved in 250 mL of toluene. While stirring the solution at 25° C., heptane was slowly added until the solution became cloudy. This was heated until the turbidity disappeared, and slowly cooled to 25° C. while stirring gently. The resulting precipitate was removed by centrifugation, and the solvent was removed from the remaining supernatant solution using a rotary evaporator and a vacuum pump to obtain a small amount of solid (Compound A). The weight of the obtained solid was 10% of the weight of the crude product used for purification.

The results of GPC (FIG. 12) and ¹³ C-NMR (FIG. 13) of the obtained solid compound (compound A) are shown below. In the GPC chromatogram, two types of peaks were observed; the compound that eluted earlier corresponded to linear PEG (unreacted product), and the compound that eluted later corresponded to cyclic PEG. That is, it was confirmed that the solid obtained by purification by reprecipitation was a mixture of both. Although high molecular weight linear by-products could be completely removed by this method, unreacted linear PEG and cyclic PEG could not be separated.

Furthermore, when ¹³ C-NMR of Compound A was measured, signals derived from terminal carbons were observed at 61.8 ppm and 72.5 ppm, confirming that Compound A contains linear PEG (Fig. 13). Note that the peak around 67.5 ppm is considered to be due to the by-product 1,4-dioxane. From these results, it can be understood that it is extremely difficult to completely separate cyclic PEG and linear PEG using the method according to Comparative Example 1.

Claims (11)

A separation or purification method for separating or purifying the cyclic compound from a mixture containing a cyclic compound and a linear compound, the method comprising:
contacting the mixture with a porous channel material;
A separation or purification method, wherein the porous channel material is a porous metal complex, a covalent organic framework, or an organic cage compound having a transition metal ion and an organic bridging ligand linked to the transition metal. The separation or purification method according to claim 1, wherein the ratio of the pore diameter of the porous channel material to the maximum thickness of the linear compound is 0.5 or more and 2.0 or less. The separation or purification method according to claim 1 or 2 , wherein the cyclic compound and the linear compound are each a polymer compound. The cyclic compound is a polymer compound represented by general formula (1),

In general formula (1), X ¹ represents a polymer main chain,
The linear compound is a polymer compound represented by general formula (2),

In the general formula (2), X ² represents a polymer main chain, Y ¹ and Y ² represent terminal groups that may be the same or different from each other,
The separation or purification method according to any one of claims 1 to 3 , wherein X ¹ and X ² are the same. The separation or purification method according to claim 4 , wherein X ¹ and X ² each consist of a repeating unit. A method for producing a cyclic compound, the method comprising:
cyclizing a linear compound to obtain a mixture containing a cyclic compound and the linear compound;
A method for producing a cyclic compound, comprising: separating or purifying the cyclic compound from the mixture by the separation or purification method according to any one of claims 1 to 5 . A separation material that separates the cyclic compound from a mixture containing the cyclic compound and the linear compound,
the separating material comprises a porous channel material;
A separation material, wherein the porous channel material is a porous metal complex, a covalent organic framework, or an organic cage compound having a transition metal ion and an organic bridging ligand linked to the transition metal. The separation material according to claim 7 , wherein the ratio of the pore diameter of the porous channel material to the maximum thickness of the linear compound is 0.5 or more and 2.0 or less. The separation material according to claim 7 or 8 , wherein the cyclic compound and the linear compound are each a polymer compound. The separation according to any one of claims 7 to 9 , wherein the cyclic compound and the linear compound each contain a repeating unit, and the repeating unit of the cyclic compound and the repeating unit of the linear compound are the same. Material. A separation device for separating the cyclic compound from a mixture containing the cyclic compound and the linear compound,
a casing having an inlet and an outlet;
A separation device, comprising: the separation material according to any one of claims 7 to 10 filled in the housing.

Images