EP4726102A1

EP4726102A1 - Sizing agent-containing carbon fiber bundle, carbon fiber-reinforced composite material, and pressure container

Info

Publication number: EP4726102A1
Application number: EP24848996.5A
Authority: EP
Inventors: Masahiro Matsumoto; Jun Watanabe
Original assignee: Toray Industries Inc
Current assignee: Toray Industries Inc
Priority date: 2023-07-31
Filing date: 2024-07-23
Publication date: 2026-04-15
Also published as: JPWO2025028341A1; WO2025028341A1; KR20260042350A; CN121443796A

Abstract

The main object of the invention is to provide a carbon fiber bundle that allows the tensile strength of the carbon fiber to be exhibited efficiently when used for producing a carbon fiber reinforced material. It is a carbon fiber bundle containing a sizing agent having a strand strength (strand strength A') of 5.9 GPa or more as measured according to JIS R7608 (2007), a strand strength (strand strength B') of 5.7 GPa or more as measured according to JIS R7608 (2007) wherein the resin formulation specified therein is replaced with Araldite (registered trademark) LY1564 SP Cl and Baxxodur (registered trademark) EC331, a percentage of 90% or more as the proportion of the strand strength B' to the strand strength A', a single-fiber diameter of 5.1 µm or more and less than 6.0 µm, and an interfacial shear strength of 16 MPa or less.

Description

TECHNICAL FIELD

The present invention relates to a carbon fiber bundle containing a sizing agent that can be used suitably for manufacturing aircraft members, automobile members, and ship members, as well as sporting goods such as golf shafts and fishing rods and other general industrial applications and particularly relates to a carbon fiber bundle containing a sizing agent that exhibits excellent strength development when used for forming carbon fiber reinforced materials and also to a carbon fiber reinforced material and a pressure vessel.

BACKGROUND ART

Being higher in specific strength and specific modulus than other fibers, polyacrylonitrile (hereinafter, occasionally abbreviated as PAN) based carbon fibers have been used widely as reinforcing fiber for composite materials in conventional sporting goods, aviation and aerospace products, automotive materials, civil engineering and construction materials, and other general industrial products such as pressure vessels and windmill blades, and now there is a strong demand for such fibers with further improved performance. In particular, for compressed hydrogen gas containers capable of storing hydrogen gas for fuel cells, there is an increasing demand for materials with higher strength because these containers are intended for use under higher pressures than conventional containers for compressed natural gas.
If carbon fibers with enhanced mechanical properties are developed, they will contribute to weight reduction of components of pressure vessels and so forth, and it is important therefore to improve their mechanical properties such as tensile strength and tensile modulus. In particular, there is a stronger demand for materials with even higher strength that can be applied to pressure vessels to be used in automobiles. To produce CFRP pressure vessels with higher strength, carbon fiber strands with higher tensile strength should be adopted and in addition, it is important to allow the carbon fibers used as reinforcing fibers in the composite materials to exhibit their characteristic mechanical properties as efficiently as possible.
To date, several proposals have been made with the aim of providing carbon fibers that can exhibit increased strength when used in carbon fiber reinforced materials. Patent documents 1 to 4 have made proposals that are intended to prevent bundle splitting, increase the fiber openability of carbon fiber bundles, and improve the impregnating property of resins during a production process of carbon fiber reinforced materials, which will serve to improve the moldability during the production of carbon fiber reinforced materials and also serve to realize enhanced strength as a result of eliminating portions that suffer from insufficient or excessive resin impregnation. Patent document 1 discloses a technique that is intended to allow a sizing agent to adhere into the inner layer portions of a carbon fiber bundle and then blow gas to remove the sizing agent from the outer layer portions of the bundle, thereby serving to achieve high abrasion resistance and high yarn break resistance as well as good resin-impregnating property at the same time.
Patent document 2 discloses a technique designed to impart inter-filament entanglement within a precursor fiber bundles for carbon fiber to reduce the occurrence of bundle splitting during the step for fiber opening in forming composite materials, thereby serving to improve the moldability during the composite material production process. Patent document 3 discloses a technique designed for reducing the fiber diameter of single-fibers in carbon fiber bundles and forming grooves with specific depth running in the fiber axis direction on the surface of each single-fiber, which serves to improve the resin-impregnating property. In addition, Patent document 4 discloses a technique that uses carbon fiber bundles of a flattened shape to increase the fiber openability during the molding step, which serves to obtain homogeneous molding.
Patent documents 5 to 7 put forth proposals for improving the mechanical properties of carbon fiber bundles. Patent document 5 discloses a technique designed for controlling the oxidation and carbonization conditions during the oxidation and carbonization process so as to maintain the crystal structure in a desirable state within the carbon fibers, thereby enabling the carbon fiber reinforced material to easily develop 0° tensile strength. Patent documents 6 and 7 disclose a technique designed for forming a carbon fiber bundle from a precursor fiber bundle having an enhanced density, which is realized due to dry-jet wet spinning, and then performing electrolytic oxidation of its surface layer in an aqueous electrolyte solution that contains the nitrate ion as an essential component.
Furthermore, Patent document 8 focuses on the effect of suppressing the decrease in mechanical properties that occurs during the period between the step of processing a carbon fiber bundle into a prepreg structure and the step of processing the prepreg into a final molded article and proposes, as a technique to realize it, the implementation of hydrolysis of the excess epoxy groups contained in the sizing agent before or after, or both before and after, the application of the sizing agent to the carbon fiber bundle.
Patent document 9 discloses a technique designed for improving production efficiency through the use of an increased filament count while performing uniform heat-treatment, enhancing the fracture toughness of single-fibers, and controlling inter-filament entanglement within the fiber bundle, thereby serving to provide a carbon fiber bundle that is high in stability when in the form of a yarn during the composite material production step and useful to produce a carbon fiber reinforced material having high tensile strength.

PRIOR ART DOCUMENTS

PATENT DOCUMENTS

Patent document 1: Japanese Patent No. 6543309
Patent document 2: Japanese Unexamined Patent Publication (Kokai) No. 2020-059937
Patent document 3: Japanese Patent No. 4533518
Patent document 4: Japanese Unexamined Patent Publication (Kokai) No. 2002-294568
Patent document 5: Japanese Patent No. 7239401
Patent document 6: Japanese Examined Patent Publication (Kokoku) No. HEI 5-4463
Patent document 7: Japanese Examined Patent Publication (Kokoku) No. HEI 4-9227
Patent document 8: Japanese Patent No. 6051987
Patent document 9: Japanese Patent No. 6575696

SUMMARY OF INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

However, the techniques disclosed in Patent documents 1 to 4 are essentially focused on improving the moldability of carbon fiber bundles. When good moldability has already been established, therefore, no further improvement in mechanical properties can be expected. Although a technique for causing the sizing agent to adhere to inner portions of the fiber bundle and a technique for decreasing the adhesion unevenness between the inner and outer portions of the bundle are proposed in Patent document 1, none of the cited patent documents addresses the optimization of the interfacial adhesion between carbon fibers and resin. Patent document 4 demonstrates that if the carbon fiber bundle is flattened after being impregnated with resin, it serves to suppress the decreasing rate of ring tensile strength relative to the tensile strength of resin-impregnated strands. However, this is intended merely to realize improved performance by forming homogeneous molding, and there is no description about the optimization of the interfacial adhesion between carbon fibers and resin. Thus, it cannot serve to ensure a strength of a satisfactorily high level.
In addition, although it is emphasized in Patent document 5 that the technique disclosed therein serves to produce a carbon fiber reinforced material having a high 0° tensile strength relative to the tensile strength of resin-impregnated strands of the carbon fiber bundle, it fails to realize a sufficient strength in terms of an absolute strength value. In the case of the techniques proposed in Patent documents 6 and 7, improvement in strand strength is realized in both resin materials examined before and after nitric acid treatment. However, carbon fibers having small single-fiber fineness need to be subjected to post-treatments including nitric acid treatment, drying, and inactivation in a nitrogen atmosphere at 700°C for several minutes, and as a result, there are problems such as quality degradation due to fuzz generation and reduction in productivity and cost due to the necessity of post-treatment.
The technique proposed in Patent document 8 is intended to suppress the decrease in strength that occurs during the period between the step of forming prepreg and the step of forming a final molded article, and it is not focused on realizing an increase in strength.
The technique proposed in Patent document 9 can realize an improvement in the fracture toughness of carbon fiber bundles and their moldability, which successfully leads to a carbon fiber reinforced material with an increased 0° tensile strength. However, it gives no description about interfacial adhesion between carbon fibers and resin and fails to achieve a further enhanced strength that can be realized through the optimization of the adhesiveness at the interface between the carbon fibers and the resin in addition to the optimization of the adhesiveness of the carbon fibers alone.
It should be noted here that as a technique for evaluation of the tensile strength of resin-impregnated strands, the evaluation method based on the resin composition and procedure specified in JIS R7608 (2007) is advantageous in terms of simple specimen preparation and acquisition of stable test results. In many cases, however, resins used have fewer functional groups and lower adhesiveness to carbon fiber bundles in comparison with practical resins used in actual carbon fiber reinforced materials. Therefore, a discrepancy may arise between the tensile strength of resin-impregnated strands specified in JIS R7608 (2007) and the 0° tensile strength of a carbon fiber reinforced material prepared using a practical resin.
Since the pressure resistance of a pressure vessel depends predominantly on the 0° tensile strength of the carbon fiber reinforced material, a discrepancy between the tensile strength of resin-impregnated strands specified in JIS R7608 (2007) and the 0° tensile strength of a carbon fiber reinforced material prepared using a practical resin can pose a problem in designing a pressure vessel.
On the other hand, even if the resin composition specified in JIS R7608 (2007) is simply replaced with a practical resin composition, resin impregnation of the carbon fiber bundle may not be achieved sufficiently in the case, for example, where the viscosity of the resin itself is high or where the resin is highly reactive and tends to thicken easily during impregnation, which may lead to large variation in evaluation results and thereby hinder accurate evaluation.
Thus, the main object of the present invention is to identify an evaluation method that employs a resin composition capable of reflecting the development of strength in producing a carbon fiber reinforced material and to provide a carbon fiber bundle that can efficiently develop a high tensile strength when used for producing a carbon fiber reinforced material.

MEANS OF SOLVING THE PROBLEMS

To meet the above object, the present invention is configured as described below.

(1) A carbon fiber bundle provided with a sizing agent having a strand strength (strand strength A') of 5.9 GPa or more as measured according to JIS R7608 (2007) using a resin formulation A as specified below, a strand strength (strand strength B') of 5.7 GPa or more as measured according to JIS R7608 (2007) using a resin formulation B as specified below, a percentage of 90% or more as the proportion of the strand strength B' to the strand strength A', a single-fiber diameter of 5.1 µm or more and less than 6.0 µm, and an interfacial shear strength of 16 MPa or less, and also relates to a carbon fiber bundle containing a sizing agent: resin formulation A: using Celloxide (registered trademark) 2021P, boron trifluoride monoethylamine, and acetone mixed at 100/3/4 (by parts by mass), and using ordinary pressure, a temperature of 125°C, and a time period of 30 minutes as curing conditions, and resin formulation B: using an Araldite (registered trademark) LY1564 SP Cl and Baxxodur (registered trademark) EC331 mixed at 100/35 (by parts by mass), and using ordinary pressure, a temperature of 80°C, and a time period of 120 minutes as curing conditions, followed by using a temperature of 110°C and a time period of 240 minutes.
(2) A carbon fiber bundle as set forth in the paragraph (1), wherein the filament count in the carbon fiber bundle is 24,000 or more.
(3) A carbon fiber bundle as set forth in either the paragraph (1) or (2), wherein the strand strength B' is 6.0 GPa or more.
(4) A carbon fiber bundle as set forth in the paragraph (1) or (2), wherein the sizing agent contains at least one component having an epoxy group and wherein the relationship expressed by the formula 1 is satisfied in which X (meq/g) is the epoxy value of the sizing agent extracted by immersing the carbon fiber bundle in an N,N-dimethylformamide solvent and subjecting it to ultrasonic treatment and Y (mass%) is the percent amount of the sizing agent applied to the carbon fiber bundle:
$0.1 meq / g \leq X \times Y \leq 0.68 meq / g$
(5) A carbon fiber bundle as set forth in either the paragraph (1) or (2), wherein the drape value is 18 cm or less.
(6) A carbon fiber bundle as set forth in the paragraph (5), wherein the drape value is 4 cm or more and 10 cm or less.
(7) A carbon fiber reinforced material comprising a carbon fiber bundle as set forth in either the paragraph (1) or (2).
(8) A pressure vessel comprising a carbon fiber bundle as set forth in either the paragraph (1) or (2).

ADVANTAGEOUS EFFECTS OF THE INVENTION

The use of the carbon fiber bundle containing a sizing agent according to the present invention makes it possible to efficiently improve the tensile strength of a carbon fiber reinforced material.

BRIEF DESCRIPTION OF THE DRAWINGS

[Fig. 1] This is a diagram illustrating the method employed for preparing a sample used for measurement of the drape value.
[Fig. 2] This is a diagram illustrating the method employed for measuring the drape value.
[Fig. 3] This is a schematic diagram illustrating the equipment employed for performing the pressure resistance test of a pressure vessel.

DESCRIPTION OF PREFERRED EMBODIMENTS

For the carbon fiber bundle containing a sizing agent according to the present invention, the strand strength that is determined using the resin formulation A described below, which prescribes a combination of resins as specified in JIS R7608 (2007), (hereinafter, such a strand strength determined using this resin formulation A will be occasionally referred to as strand strength A') is 5.9 GPa or more, preferably 6.3 GPa or more and 8.5 GPa or less as measured according to the tensile testing method for resin-impregnated strands specified in JIS R7608 (2007).
The resin formulation A: A resin mixture consisting of Celloxide (registered trademark) 2021P, boron trifluoride monoethylamine, and acetone mixed at a ratio of 100/3/4 (by parts by mass) is employed, and curing conditions including the use of an oven operating under ordinary pressure at a temperature of 125°C for a time period of 30 minutes are adopted.
Here, Celloxide (registered trademark) 2021P is an epoxy resin that contains (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexyl carboxylate in an amount of 97 mass% or more and has an epoxy equivalent weight of 130 g/eq and a viscosity of 240 mPa·s at 25°C. An equivalent material may be used instead.
The strand strength A' is an indicator of the resistance of carbon fiber to fracture that is exhibited when a load is applied, and it represents the strength of the carbon fiber bundle containing a sizing agent in an isolated state. If the strand strength A' is 5.9 GPa or more, it serves for easier production of a carbon fiber reinforced material that has an increased strength. It is preferable for the strand strength A' to be as high as possible, but actually, a strand strength A' of 6.2 GPa or more generally serves to produce a practically usable carbon fiber reinforced material, and 8.5 GPa is sufficient in many cases.
The strand strength A' will be controllable if the carbon fiber production method described later is used.
For the carbon fiber bundle containing a sizing agent according to the present invention, the strand strength that is determined according to the tensile test for resin-impregnated strands specified in JIS R7608 (2007) using the resin formulation B described below (hereinafter, such a strand strength determined using the resin formulation B will be occasionally referred to as strand strength B') is 5.7 GPa or more, preferably 6.0 GPa or more and 7.8 GPa or less, and more preferably 6.5 GPa or more and 7.8 GPa or less.
The resin formulation B: A resin mixture consisting of Araldite (registered trademark) LY1564 SP Cl and Baxxodur (registered trademark) EC331 mixed at a ratio of 100/35 (by parts by mass) is employed, and curing conditions including the use of ordinary pressure, a temperature of 80°C, and a time period of 120 minutes, followed by the use of ordinary pressure, a temperature of 110°C, and a time period of 240 minutes are adopted.
Here, instead of Araldite (registered trademark) LY1564 SP Cl, an epoxy resin containing 2,2'-[(1-methylethylidene)bis(4,1-phenylenoxymethylene)]bisoxirane and 1,4-bis(2,3-epoxypropoxy)butane in an amount of 80 mass% and 20 mass%, respectively, may be used. Furthermore, instead of Baxxodur (registered trademark) EC331, 2,2'-dimethyl-4,4'-methylenebis(cyclohexylamine) may be used.
The resin formulation B provides a practical resin useful for filament winding and has a viscosity suitable for preparing strand specimens by the impregnation method described in JIS R7608 (2007). Furthermore, resins prepared according to the resin formulation B have a larger number of functional groups and higher adhesiveness to carbon fibers than those prepared according to the resin composition described in JIS R7608 (2007), and it was found that accordingly, the strand strength B' of a resin prepared according to the resin formulation B can work as a model to give an index that represents the strength of the carbon fiber reinforced material to be formed from a carbon fiber bundle.
If the strand strength B' is 5.7 GPa or more, it serves practically to produce a carbon fiber reinforced material having an increased strength. If the strand strength B' is 6.0 GPa or more, furthermore, it can easily serve to produce a practically usable carbon fiber reinforced material, and if it is 6.5 GPa or more, it can be expected to allow effective reduction in the amount of carbon fiber required when producing carbon fiber reinforced materials, accordingly contributing to providing carbon fiber reinforced materials with reduced weight. If it is 7.8 GPa, it will suffice in most cases.
The strand strength B' is controllable when a carbon fiber bundle containing a sizing agent is produced according to the production method for a carbon fiber bundle containing a sizing agent that will be described later.
For the carbon fiber bundle containing a sizing agent according to the present invention, the ratio of the strand strength B' to the strand strength A' (hereinafter also referred to as strength utilization rate) is 90% or more, preferably 93% or more, and more preferably 95% or more and 99% or less. The strength utilization rate is an index that indicates the degree of strength development that is expected to occur when a carbon fiber bundle is processed into a carbon fiber reinforced material. If the strength utilization rate is 90% or more, a high degree of strength development can be achieved easily in a practical process of forming a carbon fiber reinforced material, and if it is 95% or more, it allows the mechanical properties characteristic of the carbon fiber in use can be effectively exhibited in the resulting carbon fiber reinforced material. Although the strength utilization rate is preferably as close to 100% as possible, 99% is sufficient in most carbon fiber reinforced materials.
The carbon fiber bundle containing a sizing agent according to the present invention has an interfacial shear strength of 16 MPa or less, preferably 10 MPa or more and 16 MPa or less, and more preferably 12 MPa or more and 16 MPa or less. When tensile load is applied to a carbon fiber reinforced material, the shear force that occurs at the interface works to transmit the load to the fibers, and therefore, it is important to appropriately control the interfacial shear strength in order to enhance the tensile strength of the carbon fiber reinforced material. When producing a carbon fiber reinforced material, the interfacial shear strength serves as an index that represents the adhesiveness between the carbon fibers and the resin present in the carbon fiber bundle. If the adhesiveness between the carbon fibers and the resin present is too low, interfacial delamination may occur between the carbon fibers and the resin, possibly leading to fracture of the entire carbon fiber reinforced material. If the interfacial shear strength is 10 MPa or more when producing a carbon fiber reinforced material, it can be expected that it serves to practically prevent such interfacial delamination from occurring between the carbon fibers and the resin. If the adhesiveness between the carbon fibers and the resin is too high, it can easily allow stress concentration to occur at fracture origins during the production of a carbon fiber reinforced material, which may result in fracture of the entire carbon fiber reinforced material. Thus, it is preferably 16 MPa or less.
The single-fiber diameter of the carbon fiber bundle containing a sizing agent according to the present invention is 5.1 µm or more and less than 6.0 µm, preferably 5.1 or more and 5.8 µm or less, and more preferably 5.2 or more and 5.8 µm or less. The fracture load per single-fiber depends on the strand strength and the cross-sectional area of the single-fiber, and accordingly, the single-fiber diameter affects the fracture load per single-fiber. In addition, the single-fiber diameter also affects the fiber quality grade because fuzzing caused by abrasion during the production process decreases with an increasing single-fiber diameter. If the single-fiber diameter is 5.2 µm or more, it tends to lead to a higher fiber quality grade when producing carbon fiber or when processing it into a carbon fiber reinforced material. If the single-fiber diameter is too large, it can lead to uneven reactions occurring in a single-fiber during the oxidation and carbonization process. Thus, it is preferably less than 6.0 µm.
The carbon fiber bundle containing a sizing agent according to the present invention is expected to provide a carbon fiber reinforced material with an increased tensile strength when all of the above parameters, i.e., strand strength A' for the resin formulation A, strand strength B' for the resin formulation B, the ratio of the strand strength B' to the strand strength A', single-fiber diameter, and interfacial shear strength, fall within the ranges specified above.
For the carbon fiber bundle containing a sizing agent according to the present invention, it is preferable that at least one component having an epoxy group be contained in the sizing agent, and it is also preferable that the relationship expressed by the formula 1 given below be satisfied in which X (meq/g) is the epoxy value of the sizing agent extracted by immersing the carbon fiber bundle containing the sizing agent in an N,N-dimethylformamide solvent and subjecting it to ultrasonic treatment and Y (mass%) is the content of the sizing agent in the carbon fiber bundle containing the sizing agent. Furthermore, it is more preferable that the relationship expressed by the formula 2 given below be also satisfied. $0.10 meq / g \leq X \times Y \leq 0.68 meq / g$ $0.10 meq / g \leq X \times Y \leq 0.40 meq / g$
The epoxy value is an index for indicating the amount of epoxy groups contained per gram of the sizing agent, and a larger epoxy value means a larger number of epoxy groups contained per gram of the sizing agent. The percent content of the sizing agent is expressed as the percentage (mass%) of the total mass of the sizing agent extracted by immersing the carbon fiber bundle containing the sizing agent in N,N-dimethylformamide and subjecting it to ultrasonic treatment relative to the mass of the carbon fiber bundle containing the sizing agent, which accounts for 100 mass%.
Accordingly, the term X×Y in the formulae 1 and 2 is an index representing the amount of epoxy groups that are present in the carbon fiber bundle containing a sizing agent. As the term X×Y increases, it works more strongly to enhance the adhesiveness at the interface between the carbon fibers present in the carbon fiber bundle containing a sizing agent and the resin present in the carbon fiber reinforced material. If the adhesiveness between the carbon fibers and the resin is too low, it is not preferable because it can lead to delamination occurring at the interface between the carbon fibers and the resin, possibly resulting in fracture of the entire carbon fiber reinforced material. On the other hand, If the adhesiveness between the carbon fibers and the resin is too high, it can easily allow stress concentration to occur at fracture origins during the production of a carbon fiber reinforced material, which may result in fracture of the entire carbon fiber reinforced material. Therefore, it is not preferable.
If the term X×Y is controlled appropriately within the range specified by the formula 1, it serves to easily realize a high strength when actually producing a carbon fiber reinforced material. If the term X×Y is controlled appropriately within the range specified by the formula 2, it serves to easily produce a carbon fiber reinforced material that is durable in practical use.
The carbon fiber bundle containing a sizing agent according to the present invention preferably has a drape value of 18 cm or less, more preferably 4 cm or more and 16 cm or less, and still more preferably 4 cm or more and 10 cm or less. The drape value is an index representing the stiffness of a carbon fiber bundle, and a larger drape value indicates a stiffer carbon fiber bundle.
If the drape value is less than 4 cm, it indicates that the fiber bundle is too soft, and therefore, when yarns are withdrawn from bobbins and aligned during a production process of a composite material, they may be easily bent or twisted on a guide parts of combs etc. designed for their alignment. Such bending or twisting occurs, it hinders fiber opening in the affected part and causes uneven fiber opening, and therefore, it is not preferable. If the drape value is more than 18 cm, it means that the fiber bundle is very stiff, and fuzzing may occur easily in some cases when it deforms upon contact with guide parts of combs etc. If the drape value is 4 cm or more and 10 cm or less, it is preferable because it serves to suppress both uneven fiber opening and fuzzing upon contact with guide parts.
For the carbon fiber bundle containing a sizing agent according to the present invention, it is preferable that the filament count be 24,000 or more, more preferably 36,000 or more. When producing a composite material by the filament winding method, the productivity depends on the yarn speed and filament count, and therefore, the use of a larger filament count makes it possible to produce a composite material more efficiently. If the filament count is 24,000 or more, it is satisfactory from the viewpoint of productivity. The upper limit of the filament count is not particularly restricted, but a larger filament count tends to lead to a larger variation in the mass per unit area due to a larger unevenness in the stretching ratio between the inner and outer portions of the yarn bundle during the stretching step. Thus, the filament count is preferably 48,000 or less.
Next, described below is a production method for a precursor fiber bundle for carbon fiber that is suitable for producing the carbon fiber bundle containing a sizing agent according to the present invention.
A precursor fiber bundle for carbon fiber can be produced by spinning a spinning dope solution of a polyacrylonitrile copolymer. Examples of such a polyacrylonitrile copolymer include not only homopolymers formed from acrylonitrile alone, but also copolymers formed from combinations of other monomers along with acrylonitrile as main component. Specifically, it is preferable that in the polyacrylonitrile copolymer, 90 to 100 mass% be accounted for by acrylonitrile while less than 10% is accounted for by a copolymerizable monomer.
Useful monomers that are copolymerizable with acrylonitrile include, for example, acrylic acid, methacrylic acid, itaconic acid, alkali metal salts thereof, ammonium salts thereof, lower alkyl esters thereof, acrylamide, derivatives thereof, allyl sulfonic acid, methallyl sulfonic acid, salts thereof, and alkyl esters thereof.
A spinning dope solution is prepared by dissolving a polyacrylonitrile copolymer as described above in a solvent that can dissolve the polyacrylonitrile copolymer, such as dimethyl sulfoxide, dimethyl formamide, dimethyl acetamide, nitric acid, aqueous solutions of zinc chloride, and aqueous solutions of sodium rhodanide. In the case where the solution polymerization technique is used for preparing a polyacrylonitrile copolymer, it is preferable to use the same solvent for both polymerization and spinning because it eliminates the necessity of steps for separating the resulting polyacrylonitrile copolymer and re-dissolving it in the spinning solvent.
For the production of a precursor fiber bundle for carbon fiber, either the dry-jet wet spinning technique or the wet spinning technique may be used for the spinning. The spinning process includes a coagulation process in which spinning is performed by discharging a spinning dope solution from a spinneret into a coagulation bath, a washing process in water in which the coagulated fiber resulting from the coagulation process is rinsed in a water bath, a water bath stretching process in which the fiber resulting from the washing process in water is stretched in a water bath, and a drying and heat-treatment process in which the fiber resulting from the water bath stretching process is subjected to drying and heat-treatment. If required, a steam stretching process may also be included in which the fiber resulting from the drying and heat-treatment process is stretched in steam. If required, furthermore, a combining step is preferably included to make the number of fibers to be equal to the filament count of the intended carbon fiber bundle. Here, the order of the processes may be rearranged as appropriate.
The spinning dope solution is prepared by dissolving the aforementioned polyacrylonitrile based copolymer in a solvent that can dissolve polyacrylonitrile, such as dimethyl sulfoxide, dimethyl formamide, and dimethyl acetamide.
It is preferable for the coagulation bath to contain the same solvent as used in the spinning dope solution, such as dimethyl sulfoxide, dimethyl formamide, and dimethyl acetamide, along with a coagulant. The coagulant to use may be one that is not able to dissolve the aforementioned polyacrylonitrile based copolymer and is compatible with the solvent present in the spinning dope solution. Specifically, it is preferable to adopt water as the coagulant.
In the washing process in water, the fiber resulting from the spinning process is introduced into a washing bath with the aim of further removing the organic solvent from the coagulated fiber bundle. To improve the passability for the fiber traveling through the washing process in water, the fiber may be stretched in the washing process in water.
Commonly, the water bath stretching process can be carried out in a single or a plurality of water baths that are controlled in the temperature range of 30°C to 98°C. It is preferable for the stretching ratio in the water bath stretching process to be set at 2 to 6.
The oil agent application process is performed after the process for water bath stretching in order to apply an oil agent with the aim of preventing adhesion between fibers. It is preferable that the oil agent to use in this process contain silicone as primary component. If the oil agent in use does not contain silicone, it will not work effectively in preventing adhesion between fibers in the oxidation process, resulting in a decrease in strand strength. Furthermore, it is preferable for the silicone based oil agent to contain a modified silicone such as amino-modified silicone that is high in heat resistance. Other good silicone based oil agents include modified silicones such as epoxy-modified ones and alkylene oxide-modified ones.
For the drying process, a generally known method may be used. In addition, from the viewpoint of improving the productivity and improving the orientation parameter of crystallites, it is preferable to carry out stretching in a heated heat medium after the drying process. Heat mediums that can be used suitably include, for example, compressed steam and superheated steam, which are preferred from the viewpoint of operation stability and cost.
Here, a dry heat stretching process, a steam stretching process, etc., may be carried out additionally after the drying process.
Next, described below is the production method for a carbon fiber bundle containing a sizing agent according to the present invention. The production method for the carbon fiber bundle containing a sizing agent according to the present invention is intended to produce a carbon fiber bundle by, for example, carrying out an oxidation process designed so that a precursor fiber bundle for carbon fiber prepared by the aforementioned method is oxidized in an oxidizing atmosphere at a temperature of 200°C to 300°C, a pre-carbonization process designed for performing pre-carbonization treatment in an inert atmosphere at a maximum temperature of 500°C to 1,200°C, and a subsequent carbonization process designed for performing carbonization treatment in an inert atmosphere at a maximum temperature of 1,200°C to 2,000°C.
In order to obtain carbon fiber with an increased strength, it is preferable that the oxidation process be controlled in such a manner that the resulting oxidized fiber gives an infrared spectrum in which the ratio of the peak intensity at 1,453 cm^-1 to the peak intensity at 1,370 cm^-1 is in the range of 0.60 to 0.75 while at the same time the ratio of the peak intensity at 1,254 cm^-1 to the peak intensity at 1,370 cm^-1 in the infrared spectrum is in the range of 0.50 to 0.65. The peak at 1,453 cm^-1 in the infrared spectrum is attributed to alkene and its intensity decreases with the progress of the oxidation. The peak at 1,370 cm^-1 and the peak at 1,254 cm^-1 are attributed to the oxidized structure and they increase with the progress of the oxidation. This peak intensity ratio of the peak at 1,254 cm^-1 to the peak at 1,370 cm^-1 decreases with the progress of the oxidation, and in particular, a large decrease occurs in the initial stage. Depending on the oxidation conditions, however, the peak intensity ratio may not decrease down to 0.65 or less even if the oxidation is performed for a prolonged time.
To allow both of these peak intensity ratios to be in the desired ranges, basically it suffices that the treatment conditions are set up focusing mainly on the requirements that the copolymer components contained should account for only a small proportion in the polyacrylonitrile based copolymer present in the precursor fiber, that the precursor fiber should have a high orientation parameter of crystallites, that the precursor fiber should have a decreased single-fiber fineness, and that the temperature used in the oxidation process should be increased in the latter half of the process. More specifically, it is preferable that the oxidation process be carried out in the following two steps: the first oxidation step in which oxidation is continued for 8 to 40 minutes so that the ratio of the peak intensity at 1,453 cm^-1 to the peak intensity at 1,370 cm^-1 in the infrared spectrum comes in the range of 0.80 to 1.05 and the second oxidation step in which the fiber resulting from the first oxidation step is subjected to additional oxidation performed for 5 to 40 minutes, preferably 5 to 30 minutes, so that the ratio of the peak intensity at 1,453 cm^-1 to the peak intensity at 1,370 cm^-1 in the infrared spectrum comes in the range of 0.60 to 0.75 while at the same time the ratio of the peak intensity at 1,254 cm^-1 to the peak intensity at 1,370 cm^-1 in the infrared spectrum comes in the range of 0.50 to 0.65.
It is preferable that the oxidation temperature used in the first oxidation step be controlled in the range of 200°C to 250°C, more preferably 230°C to 250°C, in order to ensure that the peak intensity ratios in the infrared spectrum come in the rages specified above.
The oxidation in the second oxidation step is preferably performed at a higher oxidation temperature than that in the first oxidation step. To shorten the oxidation time in the second oxidation step, it may be effective to adjust the oxidation temperature to a higher value, although the optimum oxidation temperature depends on the properties of the precursor fiber. In order to control the peak intensity ratios in the aforementioned ranges in the infrared spectrum, it is preferable for the oxidation temperature to be adjusted to 250°C to 300°C, more preferably 260°C to 290°C. The oxidation temperature does not need to be constant and may be set in multiple temperature stages. To obtain carbon fiber having an increased strand strength, it is preferable to adopt a higher oxidation temperature and a shorter oxidation time.
It should be noted that for the present invention, the oxidation process refers to a process in which precursor fiber is subjected to heat-treatment in an oxygen-containing atmosphere at 200°C to 310°C.
The oxidation time referred to herein means the time period over which the fiber stays in the oxidation furnace. Oxidized fiber refers to the fiber that has resulted from the oxidation process and is not yet sent to the pre-carbonization process. In addition, the peak intensity referred to herein means the absorbance at a particular wavelength determined by sampling a small amount of oxidized fiber, subjecting it to infrared spectrum analysis, and applying baseline correction to the spectrum obtained, with the spectrum remaining unamended by peak splitting etc. Here, the sample is diluted with KBr to a concentration of 0.67 mass% before being subjected to infrared spectrum measurement. Each time the oxidation conditions are to be changed, infrared spectrum analysis should be newly performed in this way so that new conditions can be set up for the preferred production method that will be described later. If the infrared spectrum peak intensity ratios of oxidized fiber are controlled appropriately, it serves to enable the production of a carbon fiber that has an appropriately adjusted strand strength.
Air is desirably adopted as the oxidizing atmosphere used for the oxidation treatment. For the present invention, pre-carbonization treatment and carbonization treatment are performed in an inert atmosphere. Good gases for the inert atmosphere include, for example, nitrogen, argon, and xenon, of which nitrogen is preferred from an economical point of view.
The resulting carbon fiber bundle may be subjected to electrolytic treatment in order to modify its surface. Electrolytic treatment is effective because it can ensure optimized adhesiveness to the matrix for carbon fiber in the fiber reinforced composite material to be obtained.
After finishing the electrolytic treatment, sizing treatment is performed in order to impart convergency to the resulting carbon fiber bundle. Depending on the type of resin in use, a sizing agent that is highly compatible with the matrix resin may be selected appropriately for use as the aforementioned sizing agent. However, it is important to control the adhesiveness appropriately in order to realize both suppression of interfacial delamination between the carbon fibers and the resin and alleviation of stress concentration in the resulting carbon fiber reinforced material, thereby ensuring development of better mechanical properties.
To this end, it is preferable to take measures such as optimization of the electrolytic treatment conditions including the type and concentration of the electrolyte and the amount of the sizing agent to be applied, and control of the number of functional groups contained in the sizing agent. In the case where the sizing agent contains epoxy groups, it is preferable to hydrolyze the epoxy groups in advance before applying the sizing agent or subjecting the carbon fiber bundle having the sizing agent applied to its surface to aging treatment in a constant-temperature and constant-humidity atmosphere, thereby hydrolyzing the epoxy groups to ensure an appropriate number of epoxy groups.
The carbon fiber bundle containing a sizing agent according to the present invention is used suitably as reinforcing fiber of carbon fiber reinforced materials. Members formed from such carbon fiber reinforced materials can be applied to a variety of fields including aerospace, automobiles, railroad vehicles, ships, civil engineering construction materials, and sporting goods. In particular, they can be suitably used as parts of hollow vessels such as pressure vessels or as cylindrical members.
Generally known methods can be used for the production of carbon fiber reinforced materials from the carbon fiber bundle containing a sizing agent according to the present invention. To produce hollow vessels such as pressure vessels or cylindrical products, in particular, it is preferable to employ a molding method using the filament winding technique or tow prepreg formed by impregnating fiber with resin.

EXAMPLES

The methods for the measurement of various physical properties used in the examples of the present invention are described below. For the comparative examples in which the carbon fiber bundle does not contain a sizing agent, the term "carbon fiber bundle containing a sizing agent" used in measurement method-related items and in the tables of results should be read as "carbon fiber bundle".

An oxidized fiber sample to be examined was freeze-crushed, and 2 mg thereof was accurately weighed, mixed well with 300 mg of KBr, placed in a molding jig fixture, and pressed under 40 MPa for 2 minutes in a pressing machine to prepare a pellet for measurement. This pellet was placed in a Fourier transform infrared spectrometer and subjected to spectral analysis in the range of 1,000 to 2,000 cm^-1. Here, background correction was performed by subtracting the minimum value in the range of 1,700 to 2,000 cm^-1 from all intensity values so that the minimum value becomes zero after the correction. The Fourier transform infrared spectrometer used above was Paragon 1000, manufactured by PerkinElmer. The peak intensity ratio was calculated from peak intensities determined in this way at different wave numbers.

From the carbon fiber bundle containing a sizing agent under examination, a sample with a length of 10 m was taken, dried absolutely at 120°C for 2 hours, and then weighed (in grams), followed by dividing the measured mass by 10 to determine the overall fineness (in g/m), which represents the mass per meter.

From the carbon fiber bundle containing a sizing agent under examination, a sample with a length of 1 m was taken and measurement was performed by the Archimedes method using o-dichloroethylene as specific gravity liquid. Measurements were taken from three specimens. Their average value was adopted as the measured specific gravity.

<Single-fiber diameter>

The single-fiber diameter was calculated from the overall fineness (g/m) determined by the aforementioned method, the specific gravity, and the filament count of the carbon fiber bundle used for measurement. Here, the measured single-fiber diameter was expressed in micrometers (µm).

- Strand strength

The strand strength of a carbon fiber bundle containing a sizing agent was determined according to the test method for tensile strength of resin-impregnated strands specified in JIS-R-7608 (2007) using either of the two resin formulations (resin formulation A and resin formulation B). The curing conditions adopted here were as specified in each resin formulation. Ten strands formed from carbon fiber bundles containing a sizing agent were subjected to measurement, and the average value of the measurements (in GPa, rounded to the first decimal place) was adopted as the strand strength A' or B'.

- Strength utilization rate

The strength utilization rate was determined by the following formula from the strand strengths A' and B' calculated above. $Strength utilization rate (%) = (strand strength B' / strand strength A') \times 100$
Here, the strength utilization rate was determined to two significant figures.

A carbon fiber bundle containing 2.0 ± 0.5 g of a sizing agent was weighed (W₁) (to the fourth decimal place) and left to stand for 15 minutes in an electric furnace (capacity 120 cm³) that was set to a temperature of 450°C and placed in a nitrogen flow of 50 milliliters/min, thereby ensuring complete thermal decomposition of the sizing agent. Then, the carbon fiber bundle was transferred to a container placed in a dry nitrogen flow of 20 liters/min and cooled for 15 minutes, and subsequently the weight (W₂) of the carbon fiber bundle was measured (to the fourth decimal place), followed by calculating the weight of the sizing agent present in the carbon fiber bundle containing a sizing agent as W₁ - W₂. This calculated weight of the sizing agent contained was divided by W₁ to provide a percent value in mass% (rounded off to the second decimal place) relative to the entire carbon fiber bundle containing a sizing agent, which accounted for 100 mass%, thereby giving the percent content Y of the sizing agent (in mass%). This measurement procedure was performed twice and the average of the two measurements was adopted to represent the percent content Y of the sizing agent (in mass%).

Thirty (30) grams of the carbon fiber bundle containing a sizing agent was immersed in 150 ml of N,N-dimethylformamide solvent maintained at a temperature of 25°C and subjected to ultrasonic treatment at 40 kHz, which was performed three times for 30 minutes each. The resulting solution that contained the sizing agent extracted from the fiber was analyzed to determine the epoxy value. Here, after the extraction of the sizing agent from the carbon fiber bundle containing the sizing agent, the sizing agent still remaining in the carbon fiber accounted for 0.20 mass% or less. The solution containing the extracted sizing agent was treated with hydrochloric acid to open the ring in the epoxy group and subjected to acid-base titration to determine the epoxy value X (meq/g).

As shown in Figure 1, a 50 cm long carbon fiber bundle 2 containing a sizing agent was suspended in an atmosphere at a temperature of 25°C with a weight 3 attached to apply a load of 0.0375 [g/tex] and left for 30 minutes or more to remove twist set. A 30 cm long portion was taken from the central part and as shown in Figure 2, it was placed on a rectangular horizontal platform 4 with a 90° edge corner in such a manner that the carbon fiber bundle 2 containing a sizing agent extended 25 cm beyond the edge while supporting the carbon fiber bundle 2 to prevent it from bending sharply. Then, the carbon fiber bundle 2 containing a sizing agent was fixed with tape on the horizontal platform 4. Subsequently, the support for the overhanging portion of the carbon fiber bundle 2 containing a sizing agent extending from the horizontal platform 4 was removed to allow it to hang down, and the horizontal distance L from the supporting point was measured one second later. The measurement was performed five times, and the arithmetic average value of the measurements was adopted to represent the drape value.

The interfacial shear strength is carried out according to the steps (1) to (4) described below.

(1) Preparation of resin

First, 100 parts by mass of a bisphenol A type epoxy compound jER (registered trademark) 828 (manufactured by Mitsubishi Chemical Corporation) and 14.5 parts by mass of meta-phenylene diamine (manufactured by Sigma-Aldrich Japan) were put in separate containers. Then, they were heated at 75°C for 15 minutes to ensure a decrease in the viscosity of jER 828 and complete dissolution of meta-phenylene diamine. Subsequently, they were mixed thoroughly and degassed in a vacuum at a temperature of 80°C for approximately 15 minutes.

(2) Fixing of single carbon fiber in dedicated mold

A single-fiber was taken out from a carbon fiber bundle containing a sizing agent, placed on a dumbbell shaped mold along the longitudinal direction, and fixed at both ends with adhesive under a constant tension of 4.0 ± 2.0 g. Subsequently, it was vacuum-dried at a temperature of 80°C for 30 minutes or more to remove moisture from both the single carbon fiber and the mold. The dumbbell shaped mold used was made of silicone rubber and had a casting section shaped with central portion width and length of 5 mm and 25 mm, respectively, an end portion width of 10 mm, and a total length of 150 mm.

(3) From casting to curing of resin

A resin material prepared according to the procedure described in the above paragraph (1) was poured in the mold that had been vacuum-dried as described in the above paragraph (2). Then, it was placed in an oven, heated to a temperature of 75°C at a temperature ramp rate of 1.5°C/min, maintained there for 2 hours, further heated to a temperature of 125°C at a temperature ramp rate of 1.5°C/min, maintained there for 2 hours, and then cooled to a temperature of 30°C at a temperature ramp rate of 2.5°C/min. Subsequently, it was demolded to provide a test piece.

(4) Measurement of interfacial shear strength

To the test specimen prepared according to the procedure described in the above paragraph (3), a tensile load was applied in the fiber axial direction (longitudinal direction) of the single carbon fiber to cause a strain of 12%. Then, the number N (number) of fractured portions in the single-fiber within a 22 mm region around the center of the test piece was measured under a polarizing microscope. Next, the average fractured fiber length la was calculated by the following equation: la (µm) = 22 × 1,000 (µm) / N (number). Then, the critical fiber length Ic was calculated from the average fractured fiber length la by the following equation: Ic (µm) = (4/3) × la (µm). Using the strand strength A' and single-fiber diameter d of the carbon fiber measured above, the interfacial shear strength, which is an index that represents the adhesive strength at the interface between the carbon fibers and the resin, was calculated by the equation given below. The measurement was performed five times, and the arithmetic average value of the measurements was adopted as test result. $Interfacial shear strength (Mpa) = strand strength A' (MPa) \times d (μm) / (2 \times lc) (μm)$

A case in which the filament count is 24,000 is described below as an example. Cases for other filament counts, in which each filament count is represented as N, will be described later.
Two metal bars (stainless steel) with a diameter of 50 mm and a surface roughness Rmax of 0.3 µm were fixed at different vertical heights with a space of 150 mm between them in such a manner that a carbon fiber bundle is allowed to travel while in contact with each metal bar at an angle of 0.3925 π (rad) ± 0.04 π (rad), i.e., a total angle of 0.785 π (rad). Then, a carbon fiber bundle containing a sizing agent and having a filament count of 24,000 was stretched over the metal bars. The unwinding tension from the package was set to 1,600 g and a driving roller was operated to pull the carbon fiber bundle to allow it to travel on the metal bars at a speed of 4 m/min. The number of fuzz defects that occurred during a one minute period after passing the second metal bar was measured to represent the processability. Number of fuzz defects detected during processing of carbon fiber bundle (number/m) = number of fuzz defects (number) /4 (m)

A: The number of fuzz defects detected during processing is 4.9 or less per meter.
B: The number of fuzz defects detected during processing is more than 4.9 per meter and 7.0 or less per meter.
C: The number of fuzz defects detected during processing is more than 7.0 per meter.

In the case where the filament count of the carbon fiber bundle to be examined was N (N is a natural number), the unwinding tension was set to 1,600 × N / 24,000 g and the number of fuzz defects detected during processing was divided by √(N/24,000) so as to standardize the evaluation criteria. The processability was determined in this way. For example, in the case where the filament count of the carbon fiber bundle to be examined was 36,000, the unwinding tension was set to 2,400 g and the number of fuzz defects detected during the processing was divided by √1.5, thereby enabling the calculation of the processability. It should be noted that except for the unwinding tension, all measurement conditions were common to those for the case in which the filament count was 24,000. The ratings of A and B are preferable for the present invention, with A being more preferable.

A pressure vessel to be examined as test specimen was placed in a hydrostatic burst test apparatus for pressure vessels as shown in Fig. 3, and water was supplied by a hydraulic pump to apply pressure. The burst pressure at which the vessel ruptured was measured. The measurement was performed three times, and the arithmetic average value of the measurements was adopted for evaluation of pressure resistance, which was performed according to the following criteria. The ratings of S, A, B, and C are preferable for the present invention, with S, A, and B being more preferable. In the tables, "-" represents a missing value.

S: The burst pressure is 107 MPa or more.
A: The burst pressure is 99 MPa or more and less than 107 MPa.
B: The burst pressure is 91 MPa or more and less than 99MPa.
C: The burst pressure is 86 MPa or more and less than 91MPa.
D: The burst pressure is less than 86 MPa.

[Reference example 1-1] Carbon fiber bundle production method I

A copolymer formed from acrylonitrile and itaconic acid was dissolved in dimethyl sulfoxide to prepare a spinning dope solution. Coagulated fibers were produced through a dry-jet wet spinning process in which the spinning dope solution prepared above was extruded into air through a spinneret having 6,000 holes and introduced into a coagulation bath containing an aqueous solution of dimethyl sulfoxide. The resulting fiber bundle was washed with water at 30°C to 98°C by a conventional method, followed by stretching. Then, an amino-modified silicone based silicone oil agent was applied to this fiber bundle resulting from the stretching in a water bath and subjected to dry densification treatment using heated rollers maintained at 160°C. The filament count was adjusted to 12,000 and then stretched 3.7 times in pressurized steam to achieve a total stretching ratio of 13 over the entire spinning process to provide a polyacrylonitrile based precursor fiber bundle with a single-fiber fineness of 0.7 dtex.
Next, such polyacrylonitrile based precursor fiber bundles prepared above were combined as required to ensure a filament count as specified in Table 1 and treated in an oxidation and carbonization process as described below to provide a carbon fiber bundle.
Specifically, the first oxidation step was carried out under the conditions of an oxidation temperature of 240°C and an oxidation time of 36 minutes and then the second oxidation step was carried out under the conditions of an oxidation temperature of 250°C and an oxidation time of 37 minutes. In this way, the precursor fiber bundles of carbon fiber were treated by oxidation while being stretched at a stretching ratio of 1 in an oven having an air atmosphere, thereby forming oxidized fiber.
When the fiber resulting from the first oxidation step was examined by infrared spectroscopy, the ratio of the peak intensity at 1,453 cm^-1 to the peak intensity at 1,370 cm^-1 was 0.87. When the fiber resulting from the second oxidation step was examined by infrared spectroscopy, the ratio of the peak intensity at 1,453 cm^-1 to the peak intensity at 1,370 cm^-1 was 0.63, and the ratio of the peak intensity at 1,254 cm^-1 to the peak intensity at 1,370 cm^-1 was 0.60.
The resulting oxidized fiber was subjected to pre-carbonization treatment performed at a stretching ratio of 0.96 in a nitrogen atmosphere at a maximum temperature of 900°C to provide pre-carbonized fiber. The resulting pre-carbonized fiber was subjected to carbonization treatment performed at a stretching ratio of 0.950 in a nitrogen atmosphere at a maximum temperature of 1,500°C.
Following this, electrochemical treatment of the fiber surface was performed using an aqueous sulfuric acid solution as electrolyte, followed by washing with water and drying, thereby providing a carbon fiber bundle that contained no sizing agent. For convenience, a carbon fiber bundle formed by the carbon fiber bundle production method I is designated as carbon fiber I in the tables.

[Reference example 1-2] Carbon fiber bundle production methods II to IV

Carbon fiber bundles containing no sizing agent were prepared according to the carbon fiber bundle production method I, with the conditions for oxidation, pre-carbonization, and carbonization being altered as shown in Table 1. The different carbon fiber bundle production methods are designated as production methods II to IV and, as in the case of the production method I, the resulting carbon fiber bundles are designated, for convenience, as carbon fibers II to IV.

[Table 1]

	precursor fiber bundle for carbon fiber	oxidation								pre-carbonization		carbonization		carbon fiber bundle
	single-fiber fineness (dtex)	first oxidation		second oxidation		after first oxidation	after second oxidation		stretching ratio (-)	maximum temperature (°C)	stretching ratio (-)	maximum temperature (°C)	stretching ratio (-)	filament count (number)
	single-fiber fineness (dtex)	temperature (°C)	time (min)	temperature (°C)	time (min)	1,453 cm^-1/ 1,370 cm^-1	1,453 cm^-1/ 1,370 cm^-1	1,254 cm^-1/ 1,370 cm^-1	stretching ratio (-)	maximum temperature (°C)	stretching ratio (-)	maximum temperature (°C)	stretching ratio (-)	filament count (number)
carbon fiber bundle I	0.7	240	36	250	37	0.87	0.63	0.60	1.00	900	0.96	1500	0.95	36,000
carbon fiber bundle II	0.7	240	17	269	28	1.01	0.64	0.59	1.00	900	1.03	1500	0.95	36,000
carbon fiber bundle III	0.7	250	17	267	20	0.92	0.68	0.60	1.00	900	1.10	1400	0.96	24,000
carbon fiber bundle IV	0.7	250	11	280	6	1.04	0.70	0.61	1.00	900	1.20	1400	0.95	24,000

[Reference example 2-1] Production of sizing agents i to iii

Ethylene oxide was added to bisphenol A in the presence of a potassium hydroxide catalyst, followed by removal of the catalyst to obtain an adduct of bisphenol A with ethylene oxide (2 moles). In a 1 L flask equipped with a stirring device, stirring blades, a nitrogen gas seal device, and a distillation column, 2 moles of the above adduct of bisphenol A with ethylene oxide, 1.5 moles of maleic acid, and 0.5 mole of sebacic acid were fed, and heating and stirring were performed at 160°C for 4 hours while bubbling nitrogen through the liquid, followed by removing the generated water by distillation, thereby producing a condensation product of an unsaturated dibasic acid and an adduct of bisphenol with an alkylene oxide. A water-dispersed emulsion was prepared by mixing 2 parts by mass of the above condensation product, 2 parts by mass of jER (registered trademark) 828 (manufactured by Mitsubishi Chemical Corporation), 1 part by mass of polyoxyethylene (70 mol) styrenated (5 mol) cumylphenol, and 95 parts by mass of distilled water, thereby providing a sizing agent i. Sizing agents ii and iii were prepared by subjecting the sizing agent i to heat-treatment under the conditions shown in Table 2.

[Reference example 2-2] Sizing agents iv

A 10 mole amount of an adduct of bisphenol A with ethylene oxide as represented by the formula given below was used as a sizing agent iv.

[Table 2]

	temperature (°C)	time (hours)
sizing agent i	-	-
sizing agent ii	55	140
sizing agent iii	55	445

[Examples 1 to 14 and Comparative examples 1 to 7]

The carbon fiber bundles I to IV containing no sizing agent that were prepared in Reference example 1 were impregnated with the sizing agents i to iv prepared in Reference example 2, wherein they were combined appropriately and the concentrations of the aqueous sizing agent solutions were adjusted appropriately to realize the sizing agent contents shown in Table 3. Then, they were subjected to heat-treatment at 210°C for 75 seconds to provide carbon fiber bundles each containing a sizing agent.
Physical properties of the resulting carbon fiber bundles are summarized in Table 3. In Comparative example 1, in which the fiber bundles prepared contained no sizing agent, the strand strength A' was 5.7 GPa. This strength is lower than that exhibited in the case of the test condition where the carbon fiber bundle I provided with a sizing agent was used, and this resulted from fuzzing and alignment irregularities that occurred during measurement of the strand strength. Furthermore, measurement of the epoxy value was omitted in Comparative example 1, in which a carbon fiber bundle containing no sizing agent was adopted, and in Examples 5, 8, 11, and 14, in which the sizing agent iv that had no epoxy group was used.
Next, Araldite (registered trademark) LY1564 SP Cl and Baxxodur (registered trademark) EC331 were mixed at 100:35 (by parts by mass) to prepare a resin composition B according to the resin formulation B.
In a filament winding molding apparatus, a polyethylene liner having a capacity of 7.5 L was installed, and the resin composition B, uniformly mixed beforehand at 25°C, and the aforementioned carbon fiber bundle provided with a sizing agent were fed while impregnating the latter with the former in such a manner that the latter accounted for 22 to 28 mass% of the total mass of the carbon fiber bundle and the resin composition B.
First, hoop layers having an angle of +89° to the liner axis and hoop layers having an angle of -89° to the liner axis were stacked alternately and they were wound to a thickness of 1.4 mm to form a first layer.
Then, helical layers having angles of ±20° to the liner axis were wound to a thickness of 2.2 mm to form a second layer.
In addition, hoop layers having an angle of +89° to the liner axis and hoop layers having an angle of -89° to the liner axis were stacked alternately and they were wound to a thickness of 0.6 mm to form a third layer. Then, an intermediate formed of the first to third layers were produced.
The intermediate that contained a carbon fiber bundle containing a sizing agent, impregnated with the resin composition B, and wound around a 7.5 L polyethylene liner as described above was maintained while rotating at 7 rpm for 15 minutes in an atmosphere at 20°C. The aforementioned intermediate was heated at 80°C for 120 minutes and then at 110°C for 240 minutes to cure the resin composition B, thereby providing a pressure vessel for pressure resistance testing.
In comparison with the carbon fiber bundles prepared in Comparative examples 2 and 3, those prepared in Examples 4 and 5 were higher in the strand strength B' and gave pressure vessels with higher pressure resistance although they were formed of carbon fiber bundles produced by the same carbon fiber bundle production method I and exhibited comparable values of the strand strength A'. This demonstrates that the optimization of the interfacial adhesion between the carbon fibers and the resin through the use of a sizing agent serves to allow the tensile strength of the carbon fiber to be exhibited efficiently, thereby enabling the production of pressure vessels with higher level pressure resistance.
In comparison with the carbon fiber bundle prepared in Comparative example 5, the one prepared in Example 8 was higher in the strand strength B' and gave a pressure vessel with higher pressure resistance although they were formed of carbon fiber bundles produced by the same carbon fiber bundle production method II and exhibited comparable values of the strand strength A'. This demonstrates that the optimization of the interfacial adhesion between the carbon fibers and the resin through the use of a sizing agent serves to allow the tensile strength of the carbon fiber to be exhibited efficiently, thereby enabling the production of a pressure vessel with higher level pressure resistance.
In comparison with the carbon fiber bundle prepared in Comparative example 6, the one prepared in Example 11 was higher in the strand strength B' and gave a pressure vessel with higher pressure resistance although they were formed of carbon fiber bundles produced by the same carbon fiber bundle production method III and exhibited comparable values of the strand strength A'. This demonstrates that the optimization of the interfacial adhesion between the carbon fibers and the resin through the use of a sizing agent serves to allow the tensile strength of the carbon fiber to be exhibited efficiently, thereby enabling the production of a pressure vessel with higher level pressure resistance.
In comparison with the carbon fiber bundle prepared in Comparative example 7, the one prepared in Example 14 was higher in the strand strength B' and gave a pressure vessel with higher pressure resistance although they were formed of carbon fiber bundles produced by the same carbon fiber bundle production method IV and exhibited comparable values of the strand strength A'. This demonstrates that the optimization of the interfacial adhesion between the carbon fibers and the resin through the use of a sizing agent serves to allow the tensile strength of the carbon fiber to be exhibited efficiently, thereby enabling the production of a pressure vessel with higher level pressure resistance.

[Examples 15 to 18]

The carbon fiber bundles prepared in Comparative examples 3, 5, 6, and 7 were subjected to aging treatment by leaving them to stand for 200 hours in a constant-temperature and constant-humidity environment maintained at a temperature 60°C and a relative humidity of 90% to provide carbon fiber bundles for Examples 15 to 18. Then, pressure vessels for pressure resistance testing were produced in the same way as in Comparative examples 3, 5, 6, and 7. Physical properties of the resulting carbon fiber bundles are summarized in Table 4. In comparison with the non-aged ones, the above ones were found to be higher in the strand strength B' and able to produce pressure vessels with improved pressure resistance. Thus, it is demonstrated that the optimization of the interfacial adhesion between the carbon fibers and the resin through the use of a sizing agent serves to allow the tensile strength of the carbon fiber to be exhibited efficiently, thereby enabling the production of a pressure vessel with higher level pressure resistance.

[Table 3]

	carbon fiber bundle	sizing agent	carbon fiber bundle containing sizing agent										pressure vessel
	carbon fiber bundle	sizing agent	diameter (µm)	epoxy value X (meq. /g)	Sz content Y (mass%)	formula 1 (X×Y)	IFSS (MPa)	strand strength A' (GPa)	strand strength B' (GPa)	strength utilization rate (%) strand strength B'/ strand strength A' ×100	processability	drape value (cm)	pressure resistance
Comparative example 1	I	-	5.7	-	-	-	8	5.7	-	-	C	-	-
Example 1		i	5.7	0.7	0.11	0.08	10	6.0	5.8	97	C	13	-
Example 2		i	5.8	0.7	0.28	0.20	13	5.9	5.7	97	B	15	-
Comparative example 2		i	5.7	0.7	0.51	0.36	15	6.0	5.5	92	A	16	D
Comparative example 3		i	5.7	0.7	1.00	0.70	17	5.9	5.2	88	B	20	D
Example 3		ii	5.7	0.6	0.99	0.59	16	59	5.7	97	A	18	-
Example 4		iii	5.6	0.4	1.04	0.42	15	5.9	5.8	98	A	16	C
Comparative example 4		i	5.6	0.7	1.68	1.18	19	6.0	5.0	83	C	21	-
Example 5		iv	5.7	-	0.18	-	14	6.0	5.8	97	A	9	C
Comparative example 5	II	i	5.5	0.7	1.02	0.71	17	6.3	5.5	87	B	20	D
Example 6		ii	5.6	0.6	1.01	0.61	16	6.3	59	94	A	17	-
Example 7		iii	5.6	0.4	0.98	0.39	14	6.4	6.2	97	A	16	-
Example 8		iv	5.7	-	0.22	-	15	6.2	6.1	98	A	9	B
Comparative example 6	III	i	5.6	0.7	1.03	0.72	17	7.0	6.2	89	B	19	B
Example 9		ii	5.5	0.6	1.03	0.62	16	6.9	6.6	96	A	16	-
Example 10		iii	5.4	0.4	1.02	0.41	15	7.0	6.8	97	A	16	-
Example 11		iv	5.5	-	0.18	-	14	7.0	6.9	99	A	8	A
Comparative example 7	IV	i	5.4	0.7	1.05	0.74	17	7.8	6.8	87	B	19	A
Example 12		ii	5.3	0.6	1.02	0.61	16	7.8	7.5	96	A	17	-
Example 13		iii	5.3	0.4	0.95	0.38	15	7.7	7.6	99	A	15	-
Example 14		iv	5.2	-	0.23	-	15	8.0	7.8	98	A	7	S

[Table 4]

	carbon fiber bundle	sizing agent	carbon fiber bundle containing sizing agent										pressure vessel
	carbon fiber bundle	sizing agent	diameter (µm)	epoxy value X (meq/g)	Sz content Y (mass%)	formula 1 (X×Y)	IFSS (MPa)	strand strength A' (GPa)	strand strength B' (GPa)	strength utilization rate (%) strand strength B'/ strand strength A' ×100	processability	drape value (cm)	pressure resistance
Example 15	Comparative example 3	i	5.7	0.4	1	0.4	13	5.9	5.7	0.97	B	16	C
Example 16	Comparative example 5	i	5.5	0.4	1.02	0.4	14	6.3	6.1	0.97	B	16	B
Example 17	Comparative example 6	i	5.6	0.4	1.03	0.4	13	7.0	6.9	0.99	B	15	A
Example 18	Comparative example 7	i	5.4	0.4	1.05	0.4	15	7.8	7.5	0.96	B	15	S

EXPLANATION OF NUMERALS

1: fixed bar
2: carbon fiber bundle
3: weight
4: horizontal table
5: compressed air line
6: pressure intensifier
7: hydraulic pump
8: liquid feed hose
9: fixing mechanism
10: pressure vessel
11: safety cover
12: data logger

Claims

A carbon fiber bundle containing a sizing agent having a strand strength (strand strength A') of 5.9 GPa or more as measured according to JIS R7608 (2007) using a resin formulation A as specified below, a strand strength (strand strength B') of 5.7 GPa or more as measured according to JIS R7608 (2007) using a resin formulation B as specified below, a percentage of 90% or more as the proportion of the strand strength B' to the strand strength A', a single-fiber diameter of 5.1 µm or more and less than 6.0 µm, and an interfacial shear strength of 16 MPa or less:
resin formulation A: using Celloxide (registered trademark) 2021P, boron trifluoride monoethylamine, and acetone mixed at 100/3/4 (by parts by mass), and using ordinary pressure, a temperature of 125°C, and a time period of 30 minutes as curing conditions, and resin formulation B: using Araldite (registered trademark) LY1564 SP Cl and Baxxodur (registered trademark) EC331 mixed at 100/35 (by parts by mass), and using ordinary pressure, a temperature of 80°C, and a time period of 120 minutes as curing conditions, followed by using a temperature of 110°C and a time period of 240 minutes.
A carbon fiber bundle containing a sizing agent as set forth in claim 1, wherein the filament count of the carbon fiber bundle containing a sizing agent is 24,000 or more.
A carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2, wherein the strand strength B' is 6.0 GPa or more.
A carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2, wherein the sizing agent contains at least one component having an epoxy group and wherein the relationship between X and Y expressed by the formula 1 is satisfied in which X (meq/g) is the epoxy value of the sizing agent extracted by immersing the carbon fiber bundle containing a sizing agent in an N,N-dimethylformamide solvent and subjecting it to ultrasonic treatment and Y (mass%) is the percent amount of the sizing agent contained in the carbon fiber bundle containing a sizing agent: $0.1 meq / g \leq X \times Y \leq 0.68 meq / g$
A carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2, wherein the the drape value is 18 cm or less.
A carbon fiber bundle containing a sizing agent as set forth in claim 5, wherein the drape value is 4 cm or more and 10 cm or less.
A carbon fiber reinforced material produced from a carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2.
A pressure vessel produced from a carbon fiber bundle containing a sizing agent as set forth in claim 1 or 2.
A carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2, wherein the filament count of the carbon fiber bundle containing a sizing agent is 36,000 or more.
A carbon fiber bundle containing a sizing agent as set forth in either claim 1 or 2, wherein the single-fiber diameter of the carbon fiber bundle containing a sizing agent is 5.8 µm or less.