Sizing agent-attached carbon fiber bundle and method for producing same, carbon fiber-reinforced composite material, and pressure vessel

Abstract

The purpose of the present invention is to provide a carbon fiber bundle having stable quality when formed into a carbon fiber-reinforced composite material, by optimally designing a method for producing a carbon fiber bundle, the shape stability and scratch resistance of a sizing agent-attached carbon fiber bundle, and the resin impregnation of the carbon fiber bundle.　A sizing agent-attached carbon fiber bundle according to the present invention is a carbon fiber bundle to which a sizing agent is applied and which has 6,000 filaments to 50,000 filaments, while having a thickness of 0.10 mm to 0.20 mm, a thickness variation coefficient of 6.5% or less, and a drape value of 6 cm to 9 cm. The sizing agent-attached carbon fiber bundle has a strand tensile strength (strand tensile strength A') of 4.9 GPa to 6.7 GPa as evaluated on the basis of JIS R7608 (2007) using a resin formulation A specified in claim 1.

Description

Sizing agent-adhered carbon fiber bundle and its manufacturing method, carbon fiber-reinforced composite material, and pressure vessel

[0001] This invention relates to carbon fiber bundles that are suitable for use in aircraft components, automobile components, and ship components, as well as in sports applications such as golf shafts and fishing rods, and other general industrial applications.

[0002] Carbon fiber possesses excellent mechanical properties, particularly high specific strength and specific modulus. Therefore, it is widely used in aerospace, leisure, and general industrial applications such as automobiles, and various molding methods have been developed. Among these, filament winding (FW) molding is becoming widely applied to carbon fiber due to its excellent moldability and the properties of the resulting composite materials. In particular, fuel containers for natural gas vehicles, which have attracted attention in recent years, are beginning to use FW-molded containers with carbon fiber as a reinforcing fiber to achieve lightweight and high-performance properties. Furthermore, in recent years, there has been a growing demand for carbon fiber suitable for FW molding in applications using higher pressures than before, such as compressed hydrogen gas containers intended for filling with hydrogen gas for fuel cells. For example, compressed hydrogen gas containers operate at pressures of 50-100 MPa, which is higher than the approximately 20-30 MPa used for conventional compressed natural gas containers. Especially for compressed hydrogen gas containers for automobiles, weight reduction is desired to increase the driving range of the vehicle. Therefore, weight reduction is achieved by using high-strength carbon fiber reinforced composite materials (hereinafter sometimes referred to as CFRP) to reduce the amount used. In the FW molding method, carbon fibers are opened and impregnated with resin by passing them through multiple guide members such as rollers and fixed bars (hereinafter referred to as fixed bars). In particular, when passing through the fixed bars, the carbon fiber bundles rub against the fixed bars in contact, which can cause abrasion fuzz on the carbon fiber bundles, and insufficient impregnation (void formation) may occur due to thickness variations caused by fiber opening and differences in resin impregnation properties caused by sizing agents. If voids are formed in the resin-impregnated material, it will ultimately lead to a decrease in the quality and grade of the molded product, so the application of carbon fiber bundles with uniform thickness that have abrasion resistance and resin impregnation properties is desired. Thus, it is desired that the mechanical properties of carbon fibers be efficiently expressed in CFRP used in such applications. Furthermore, it is desired to improve the uniformity of CFRP and improve quality stability by making the carbon fibers suitable for the FW molding method processability.

[0003] Several proposals have been made to improve the strength development when using CFRP. Patent documents 1 to 4 propose improving the moldability of CFRP by preventing fiber splitting and improving the fiber opening properties and resin impregnation of the carbon fiber bundles during molding, thereby improving strength development by eliminating unimpregnated and excess resin areas. Patent document 1 proposes a technique that achieves both abrasion resistance, fiber breakage resistance, and resin impregnation by applying a sizing agent to the inner layer of the carbon fiber bundle and then blowing gas to remove the sizing agent from the outer layer of the carbon fiber bundle. Patent document 2 proposes a technique that improves fiber opening properties during molding and produces a homogeneous molded body by flattening the shape of the carbon fiber bundle. Patent documents 3 and 4 propose improving the abrasion resistance of carbon fiber bundles by devising sizing agents applied mainly from the viewpoint of improving the convergence properties of carbon fiber bundles and controlling the mechanical properties of the molded product. Patent Document 3 discloses a technique to reduce the coefficient of friction with a fixed bar and suppress abrasion fuzz by adding a higher fatty acid ester to a sizing agent mainly composed of epoxy resin. Patent Document 4 describes how using an alkylene oxide adduct of bisphenol A as a sizing agent reduces the coefficient of friction with a fixed bar and reduces abrasion fuzz. Patent Documents 5 and 6 propose ways to improve the mechanical properties of carbon fiber bundles. Patent Document 5 describes how the crystalline structure in carbon fibers is controlled to a desirable state by adjusting the firing conditions in the flame-retardant and carbonization process, making it easier to achieve 0° tensile strength in CFRP. Patent Document 6 proposes a technique in which the surface layer of a carbon fiber bundle obtained from a precursor fiber bundle whose density has been increased by wet-dry spinning is electrolytically oxidized in an aqueous electrolyte solution containing nitrate ions as an essential component.

[0004] Japanese Patent Publication No. 2018-9280, Japanese Patent Publication No. 2002-294568, Japanese Patent Publication No. 2002-317382, Japanese Patent Publication No. Hei 07-009444, Japanese Patent Publication No. 2020-73737, Japanese Patent Publication No. Hei 2-104767

[0005] However, the technologies described in Patent Documents 1 and 2 are merely techniques to improve the moldability of carbon fiber bundles, and further improvement in mechanical properties cannot be expected if good moldability has already been established. Patent Document 1 proposes a technique to improve uneven adhesion inside and outside the bundle by adhering the sizing agent to the inside of the yarn bundle, but it does not mention controlling the thickness of the yarn bundle or the impregnation of the resin into the carbon fiber bundle by the properties of the sizing agent. Patent Document 2 proposes a technique to create a homogeneous molded body by improving the fiber-opening properties of the carbon fiber bundle, but it does not mention controlling the thickness of the yarn bundle or the impregnation of the resin into the carbon fiber bundle by the properties of the sizing agent. While Patent Documents 3 and 4 certainly show an effect of suppressing fraying, they do not mention controlling the thickness of the yarn bundle or the impregnation of the resin into the carbon fiber bundle by the properties of the sizing agent or the method of applying the sizing agent, or giving the carbon fiber bundle uniform bundleness to improve the quality stability of the molded product. Furthermore, while the technology described in Patent Document 5 shows that the 0° tensile strength of CFRP is high compared to the resin-impregnated strand strength of carbon fiber bundles, it does not mention controlling the thickness of the yarn bundle or the impregnation of the resin into the carbon fiber bundle by the properties of the sizing agent, as is the case with Patent Documents 1 to 4. In the technology described in Patent Document 6, although an improvement in strand strength is observed when measured under either of the two resin formulations before and after nitric acid treatment, post-treatment such as nitric acid treatment, drying, and inactivation at 700°C in a nitrogen atmosphere for several minutes is required. Therefore, especially for carbon fibers with small single fiber fineness, there are issues in terms of quality reduction due to the generation of fluff, and productivity and cost due to the post-treatment.

[0006] Therefore, the present invention aims to provide carbon fiber bundles that have stable quality when used as carbon fiber reinforced composite materials by optimally designing the manufacturing method of carbon fiber bundles, the morphological stability of carbon fiber bundles with sizing agents attached, abrasion resistance, and resin impregnation properties of carbon fiber bundles.

[0007] To achieve the above objective, the present invention comprises the following configuration: (1) A carbon fiber bundle to which a sizing agent has been applied, having 6,000 to 50,000 filaments, a thickness of 0.10 mm to 0.20 mm, a coefficient of variation of thickness of 6.5% or less, a drape value of 6 cm to 9 cm, and a strand tensile strength (strand tensile strength A') evaluated according to JIS R7608 (2007) using the following resin formulation A, which is 4.9 GPa to 6.7 GPa. Resin formulation A: An epoxy resin containing 97% by mass or more of (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexylcarboxylate, with an epoxy equivalent of 130 g / eq and a viscosity of 240 mPa·s at 25°C, with a composition of boron trifluoride monoethylamine / acetone = 100 / 3 / 4 (parts by mass), and curing conditions of atmospheric pressure, temperature of 125°C, and time of 30 minutes. (2) A carbon fiber bundle with sizing agent attached as described in (1) above, wherein the strand tensile strength A' is 5.6 GPa or more and 6.7 GPa or less, and the strand tensile strength (strand tensile strength B') evaluated according to JIS R7608 (2007) using the following resin formulation B is 5.3 GPa or more and 6.4 GPa or less.

[0008] Resin formulation B: The main component is 80% by mass of 2,2'-[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bisoxirane and 20% by mass of 1,4-bis(2,3-epoxypropoxy)butane, and 2,2'-dimethyl-4,4'-methylenebis(cyclohexylamine) is used as a curing agent, with a ratio of 100 / 35 (parts by mass) of the main component to the curing agent. The curing conditions are atmospheric pressure, temperature 80°C, time 120 minutes, followed by atmospheric pressure, temperature 110°C, time 240 minutes. (3) A carbon fiber bundle with sizing agent attached as described in (2) above, wherein the strand tensile strength A' is 5.9 GPa or more and 6.7 GPa or less, and the strand tensile strength (strand tensile strength B') evaluated using resin formulation B in accordance with JIS R7608 (2007) is 5.6 GPa or more and 6.4 GPa or less. (4) A carbon fiber bundle with sizing agent attached as described in any of (1) to (3) above, wherein the resin incorporation ratio is 0.45 or more and 0.64 or less. (5) A carbon fiber bundle with sizing agent attached as described in (1) to (4) above, wherein the single fiber diameter is 6.6 μm or more and 7.3 μm or less. (6) A carbon fiber bundle with sizing agent attached as described in (3) above, wherein the strand tensile strength A' is 6.3 GPa or more and 6.7 GPa or less, and the strand tensile strength B' is 6.0 GPa or more and 6.4 GPa or less. (7) A carbon fiber bundle with a sizing agent attached, as described in any of (1) to (6) above, wherein the strand tensile modulus, as evaluated in accordance with JIS R7608 (2007) using resin formulation A, is 240 GPa or more and 275 GPa or less. (8) A carbon fiber bundle with a sizing agent attached, as described in any of (1) to (7) above, wherein the amount of sizing agent attached is 0.7% by mass or more and 1.0% by mass or less per 100% by mass of the carbon fiber bundle with the sizing agent attached. (9) A carbon fiber bundle with a sizing agent attached, as described in any of (1) to (8) above, wherein the sizing agent does not contain a cyclic oxygen-containing compound. (10) A carbon fiber bundle with a sizing agent attached, as described in any of (1) to (9) above, wherein the sizing agent has a viscosity of 500 mPa·s or more and 1,400 mPa·s or less at 30°C. (11) A carbon fiber bundle with a sizing agent attached, as described in any of (1) to (10) above, wherein the coefficient of variation of thickness is 4.3% or less and the drape value is 6.5 cm or more and 7.5 cm or less.(12) A carbon fiber reinforced composite material using a carbon fiber bundle with a sizing agent attached as described in any of (1) to (11) above. (13) A pressure vessel using a carbon fiber bundle with a sizing agent attached as described in any of (1) to (12) above. (14) A method for producing carbon fiber bundles with a sizing agent attached, comprising the steps of: flame-retardant treatment, pre-carbonization, and carbonization of a polyacrylonitrile-based carbon fiber precursor fiber bundle having 6,000 to 50,000 filaments; oxidation treatment of the fiber bundle obtained in the carbonization step to obtain a carbon fiber bundle; and immersion of the carbon fiber bundle in a sizing agent-containing liquid to set the amount of sizing agent attached to 100% by mass of the carbon fiber bundle with the sizing agent attached to 0.6% by mass or more and 1.1% by mass or less, wherein the sizing agent has a proportion of cyclic oxygen-containing compound components of 50% by mass or less in 100% by mass of the total amount of sizing agent, and a viscosity of 200 mPa·s or more and 2,000 mPa·s or less at 30°C, and is substantially untwisted. (15) A polyacrylonitrile carbon fiber precursor fiber bundle having a single fiber fineness of 0.58 dtex or more and 1.20 dtex or less, and a number of filaments of 18,000 or more and 42,000 or less, is subjected to flame retardation in two stages: a first flame retardation step and a second flame retardation step, and in the second flame retardation step, the infrared spectrum is 1,370 cm. －１ 1,453 cm⁻¹ relative to the peak intensity －１ The ratio of peak intensities is in the range of 0.70 to 0.75, and the infrared spectrum is 1,370 cm⁻¹. －１ 1,254 cm⁻¹ relative to the peak intensity －１The method for producing a carbon fiber bundle with a sizing agent attached, as described in (14) above, wherein the fiber bundle is flame-retardant until the ratio of the peak intensities is in the range of 0.50 to 0.65, and comprises a pre-carbonization step in which the fiber bundle obtained in the second flame-retardant step is stretched to a ratio of 1.00 to 1.15 in an inert atmosphere with a maximum temperature of 600 to 800°C, and a carbonization step in which the fiber bundle obtained in the pre-carbonization step is carbonized in an inert atmosphere with a maximum temperature of 1000 to 1600°C at a heating rate of 0.4 to 1.1°C / second. (16) The method for producing a carbon fiber bundle with a sizing agent attached, as described in (15) above, wherein in the step of immersing the carbon fiber bundle in a sizing agent-containing liquid, the amount of sizing agent attached is 0.7% by mass or more and 1.0% by mass or less, and the sizing agent does not contain cyclic oxygen-containing compounds and has a viscosity of 500 mPa·s or more and 1,400 mPa·s or less at 30°C.

[0009] According to the present invention, the carbon fiber bundle has good morphological stability, abrasion resistance, and resin impregnation properties, making it possible to stably maintain a high tensile strength in the carbon fiber reinforced composite material.

[0010] Figure 1 shows how to set up the sample used for drape value measurement. Figure 2 shows how to measure the drape value.

[0011] The thickness of the carbon fiber bundle with sizing agent attached in this invention is 0.20 mm or less. The thickness of the carbon fiber bundle with sizing agent attached is evaluated by the method described in the Examples section. When impregnating with resin using the filament winding molding method, the resin impregnation depends on the thickness, so a thinner thickness allows for the efficient production of composite materials. If the thickness of the carbon fiber bundle with sizing agent attached is 0.20 mm or less, the quality when made into CFRP tends to be good. The thickness of the carbon fiber bundle with sizing agent attached is preferably 0.19 mm or less, and more preferably 0.18 mm or less. Thus, from the viewpoint of quality, the smaller the thickness of the carbon fiber bundle, the better. However, if the thickness is less than 0.10 mm, the thickness of the carbon fiber bundle with sizing agent attached becomes too small, and in the FW molding process, the fiber opening becomes uneven, gaps are created within the carbon fiber bundle, and the resin impregnation decreases, which can lead to a decrease in quality and mechanical properties when made into CFRP. Therefore, in this invention, the thickness of the carbon fiber bundle with sizing agent attached is controlled to be 0.10 mm or more. In other words, the thickness of the carbon fiber bundle with sizing agent attached is 0.10 mm or more and 0.20 mm or less, preferably 0.10 mm or more and 0.19 mm or less, and more preferably 0.10 mm or more and 0.18 mm or less. The thickness of the carbon fiber bundle with sizing agent attached can be controlled by the number of filaments in the carbon fiber precursor bundle, the properties of the sizing agent, and the amount of sizing agent attached. The thickness of the carbon fiber bundle with sizing agent attached in this invention can be evaluated by the method described in the Examples section.

[0012] The carbon fiber bundles with sizing agent attached according to the present invention have a coefficient of variation of thickness of 6.5% or less, preferably 6.0% or less, more preferably 5.5% or less, and even more preferably 4.3% or less. While there is no particular lower limit, it should be 0% or greater. A small coefficient of variation of the carbon fiber bundle thickness allows for the production of smooth CFRP, resulting in a greater effect in suppressing deformation of the component. If the coefficient of variation of the carbon fiber bundle thickness exceeds 6.5%, even with a small carbon fiber bundle thickness, the gaps between carbon fiber bundles when they are laminated increase, significantly reducing the smoothness of the CFRP. In other words, in addition to a small carbon fiber bundle thickness, it is important that the coefficient of variation of the carbon fiber bundle thickness is below a certain level. The coefficient of variation of the carbon fiber bundle thickness can be controlled not only by reducing the thickness of the carbon fiber bundle itself, but also by the surface temperature of the hot roller used to attach the sizing agent, its processing time, the properties of the sizing agent, and the amount of sizing agent attached. The thickness of the carbon fiber bundles with sizing agent attached according to the present invention can be evaluated by the method described in the Examples section.

[0013] The carbon fiber bundle with sizing agent attached according to the present invention has a drape value of 6 cm or more and 9 cm or less, preferably 6.5 cm or more at the lower limit, preferably 8.5 cm or less at the upper limit, and more preferably 7.5 cm or less. The drape value is an index representing the stiffness of the carbon fiber bundle; the larger the drape value, the stiffer the carbon fiber bundle. The drape value can be evaluated by the method described in the Examples section.

[0014] When the drape value of a carbon fiber bundle is less than 6 cm, there is often little or no crosslinking between the single filaments due to the sizing agent, and even if there is crosslinking, it is often weak. As a result, the fiber bundle is soft, making it prone to bending and twisting when the bundle is pulled from the bobbin and used as a guide, such as a comb, during the manufacturing of the composite material. When bending and twisting occur, those parts become difficult to open, resulting in uneven opening, which is undesirable. Also, when the drape value of a carbon fiber bundle is greater than 9 cm, the fiber bundle is stiff, which can easily cause fuzzing when it deforms in contact with a guide such as a comb, and can also result in poor resin impregnation. A drape value of 6 cm to 9 cm is preferable because it suppresses both uneven opening and fuzzing when in contact with a guide, and also allows for good resin impregnation.

[0015] The carbon fiber bundle of the present invention preferably has a resin incorporation ratio of 0.45 to 0.64. More preferably, the lower limit is 0.47 or higher, more preferably 0.60 or lower, and even more preferably 0.58 or lower. The resin incorporation ratio is an index representing the resin impregnation properties of the carbon fiber bundle during filament winding, and the higher the resin incorporation ratio, the better the resin impregnation properties of the carbon fiber bundle during filament winding. The resin incorporation ratio can be evaluated by the method described in the Examples section.

[0016] When the resin incorporation ratio of carbon fiber bundles is 0.45 or higher, the carbon fiber bundles tend to spread widely or uniformly, making it less likely for voids to be present in the pressure vessel when it is manufactured. Reducing the void content helps prevent a decrease in the strength of the pressure vessel. Furthermore, when the resin incorporation ratio of carbon fiber bundles is 0.64 or lower, it becomes less likely for localized areas with high resin content to be present between layers in the pressure vessel. Suppressing localized areas of high resin content helps prevent a decrease in the strength of the pressure vessel. In other words, a resin incorporation ratio of 0.45 or higher and 0.64 or lower is preferable because it suppresses both the occurrence of voids and areas of high resin content, resulting in good strength for the pressure vessel.

[0017] The sizing agent-attached carbon fiber bundle of the present invention has a filament count of 6,000 to 50,000. Preferably, it is 18,000 to 42,000, more preferably 24,000 to 36,000, and even more preferably 30,000 or more. The filament count refers to the number of single fibers contained in one sizing agent-attached carbon fiber bundle. If the filament count is 18,000 or more, the number of cracks occurring between the sizing agent-attached carbon fiber bundles will be reduced when it is made into CFRP. If the filament count is 50,000 or less, the thickness of the carbon fiber bundle can be kept small, and a CFRP with a small thickness can be obtained. The filament count can be arbitrarily determined during the manufacturing process of the polyacrylonitrile-based carbon fiber precursor fiber bundle.

[0018] The carbon fiber bundles with sizing agent attached according to the present invention are preferably 4.9 GPa or more and 6.7 GPa or less, more preferably 5.6 GPa or more and 6.7 GPa or less, even more preferably 5.9 GPa or more and 6.6 GPa or less, and most preferably 6.3 GPa or more and 6.6 GPa or less, when evaluated using the resin formulation A described below, which is a combination of resins described in JIS R7608 (2007), according to the tensile test method for resin-impregnated strands described in JIS R7608 (2007).

[0019] Resin formulation A: Contains 97% by mass or more of (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexylcarboxylate, and has an epoxy equivalent of 130 g / eq and a viscosity of 240 mPa·s at 25°C. The composition is 100 / 3 / 4 (parts by mass) of epoxy resin / boron trifluoride monoethylamine / acetone, and is cured under normal pressure and temperature of 125°C in an oven for 30 minutes.

[0020] Furthermore, for the epoxy resin used in resin formulation A (hereinafter referred to as resin composition A), which contains 97% by mass or more of (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexylcarboxylate, and has an epoxy equivalent of 130 g / eq and a viscosity of 240 mPa·s at 25°C, "Celoxide®" 2021P may be used.

[0021] Strand tensile strength A' is an index indicating the resistance of carbon fibers to breakage when a load is applied, and represents the strength of the carbon fiber bundle itself. If the strand tensile strength A' is 4.9 GPa or higher, it is easier to increase the strength when it is made into CFRP. The strand tensile strength A' can be controlled by manufacturing according to the carbon fiber manufacturing method described later.

[0022] The carbon fiber bundles with sizing agent attached according to the present invention preferably have a strand tensile strength (hereinafter, the strand tensile strength in this resin formulation B may be referred to as "strand tensile strength B'") of 5.3 GPa or more and 6.4 GPa or less, as evaluated according to the tensile test for resin-impregnated strands described in JIS R7608 (2007) using the resin formulation B described below, and more preferably 5.6 GPa or more and 6.3 GPa or less. In particular, carbon fiber bundles having a strand tensile strength A' of 5.6 GPa or more and a strand tensile strength B' within the above range are preferably used. Another preferred embodiment is a carbon fiber bundle having a strand tensile strength A' of 6.3 GPa or more and 6.7 GPa or less and a strand tensile strength B' of 6.0 GPa or more and 6.4 GPa or less.

[0023] Resin formulation B: The main component is 80% by mass of 2,2'-[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bisoxirane and 20% by mass of 1,4-bis(2,3-epoxypropoxy)butane, and 2,2'-dimethyl-4,4'-methylenebis(cyclohexylamine) is used as a curing agent, with a ratio of 100 / 35 (parts by mass) of the main component to the curing agent. The curing conditions are as follows: curing in an oven set to atmospheric pressure and temperature 80°C for 120 minutes, followed by curing in an oven set to temperature 110°C for 240 minutes.

[0024] Incidentally, "Araldite (registered trademark)" LY1564 SP CI may be used as the above main agent in the resin formulation B. Also, "Baxxodur (registered trademark)" EC331 may be used as the above curing agent.

[0025] The resin composition used in the resin formulation B (hereinafter referred to as resin composition B) is a practical resin for filament winding by the wet method, has a viscosity suitable for producing strand test pieces by the impregnation method described in JIS R7608 (2007), and the resin composition B has more functional groups and higher adhesiveness to the carbon fiber bundle compared to the resin composition A which is a combination of resins described in JIS R7608 (2007). Therefore, the strand tensile strength B' can be an index for model - representing the strength when a certain carbon fiber bundle is made into CFRP. If the strand tensile strength B' is 5.3 GPa or more, it is easy to increase the strength when actually made into CFRP. Also, if the strand tensile strength B' is 6.0 GPa or more, it is likely to become a CFRP that can withstand practical use and is often sufficient.

[0026] The strand tensile strength B' can efficiently control the physical properties by improving the strand tensile strength A' of the carbon fiber bundle according to the carbon fiber manufacturing method described later and adjusting the properties of the sizing agent existing between the carbon fiber and the resin.

[0027] The single - fiber diameter of the sizing - agent - attached carbon fiber bundle of the present invention is preferably 5.2 μm or more. The single - fiber diameter of the carbon fiber bundle can be evaluated by the method described in the section of the examples.

[0028] When manufacturing a prepreg, since the impregnability of the matrix resin to the carbon fiber bundle depends on the single - fiber diameter, an increase in the single - fiber diameter improves the impregnability of the resin, and a composite material can be efficiently manufactured. Also, since the breaking load per single fiber is determined from the strand tensile strength and the single - fiber cross - sectional area, the single - fiber diameter affects the breaking load per single fiber. Also, as the single - fiber diameter increases, the tendency for fuzzing due to rubbing during the process decreases, which affects the quality.

[0029] If the single fiber diameter is 5.2 μm or larger, the quality tends to be good when producing carbon fiber bundles or CFRP. The single fiber diameter is preferably 6.3 μm or larger, and more preferably 6.6 μm or larger. If the single fiber diameter is too large, the reaction within the single fiber may become non-uniform during the firing process, which can reduce the strand tensile strength and strand tensile modulus. Therefore, it should be controlled to be 7.3 μm or less. In other words, the single fiber diameter is preferably 5.2 μm or larger and 7.3 μm or smaller, more preferably 6.3 μm or larger and 7.2 μm or smaller, and even more preferably 6.6 μm or larger and 7.1 μm or smaller.

[0030] The diameter of a single fiber can be controlled by factors such as the amount of carbon fiber precursor fiber bundle extruded from the spinneret during spinning and the draw ratio during each process.

[0031] In the present invention, if the strand tensile modulus of the carbon fiber bundle with sizing agent attached is too high when evaluated using resin formulation A in accordance with JIS R7608 (2007), it becomes difficult to improve the strand tensile strength using the same resin formulation A. Therefore, it is preferable to set the strand tensile modulus to 230 GPa or more and 290 GPa or less, more preferably 235 GPa or more and 280 GPa or less, and even more preferably 240 GPa or more and 275 GPa or less. This strand tensile modulus can be controlled by adjusting the firing conditions according to the carbon fiber manufacturing method described later. Hereafter, unless otherwise specified, strand tensile strength and strand tensile modulus refer to the properties obtained by evaluating them using resin formulation A in accordance with JIS R7608 (2007).

[0032] The sizing agent attached to the carbon fiber bundle with the sizing agent attached in this invention has a viscosity at 30°C of 200 mPa·s or more and 2,000 mPa·s or less. Viscosity in this invention refers to the value obtained by measurement using a B-type viscometer. Details of the conditions can be found in the Examples section. By having a viscosity of 200 mPa·s or more and 2,000 mPa·s or less of the sizing agent at 30°C, good resin impregnation properties can be imparted to the carbon fiber bundle. The viscosity of the sizing agent at 30°C is preferably 500 mPa·s or more and 1,400 mPa·s or less, more preferably 600 mPa·s or more and 1,300 mPa·s or less, and the lower limit is even more preferably 800 mPa·s or more. If the viscosity of the sizing agent at 30°C is 500 mPa·s or more, sufficient bundling properties can be imparted to the carbon fiber bundle with the sizing agent attached, and the thickness of the carbon fiber bundle with the sizing agent attached and the coefficient of variation of the thickness can be reduced. If the viscosity of the sizing agent at 30°C becomes too high, the resin impregnation into the carbon fiber bundles to which the sizing agent is attached may become uneven during the FW molding process, potentially degrading the quality of the CFRP. Therefore, it is preferable to control the viscosity to 2000 mPa·s or less. The viscosity of the sizing agent can be measured using a B-type viscometer if the sizing agent is available. Alternatively, the sizing agent can be extracted from the carbon fiber bundles to which it is attached using a solvent, and in some cases, separated using a known method, after which it can be measured using a B-type viscometer. If the sizing agent is a mixture, its viscosity (η) is calculated additively by taking the respective viscosities (ηa, ηb) of compound a and compound b according to their respective mass ratios (Ra, Rb).

[0033] In the present invention, the sizing agent attached to the carbon fiber bundle is preferably such that the total amount of cyclic oxygen-containing compounds in 100% by mass of the total sizing agent is 50% by mass or less. There are no particular restrictions on the above cyclic oxygen-containing compounds, but compounds containing a cyclic oxygen-containing structure that can react with carboxyl groups or hydroxyl groups include epoxy group-containing compounds and oxazoline group-containing compounds. Sizing agents containing the above cyclic oxygen-containing compounds tend to have a high intermolecular crosslink density due to the reaction of the compounds, which increases viscosity, reduces the impregnation of the resin, and the layer formed by the sizing agent becomes brittle, leading to a decrease in the tensile strength of the CFRP. The total amount of compounds containing a cyclic oxygen-containing structure in 100% by mass of the total sizing agent is preferably 30% by mass or less, and it is more preferable that the sizing agent does not contain any compounds containing a cyclic oxygen-containing structure.

[0034] The proportion and presence or absence of cyclic oxygen-containing compounds in 100% by mass of the total sizing agent can be determined from the structural formula if the structural formula of the compound and the composition of the sizing agent are known. If they are unknown, the sizing agent can be extracted from a carbon fiber bundle containing the sizing agent, and its structure can be identified by known methods such as proton NMR, carbon NMR, mass spectrometry, and TOF-SIMS.

[0035] In the present invention, the amount of sizing agent attached to the carbon fiber bundle is preferably in the range of 0.6% by mass or more and 1.1% by mass or less per 100% by mass of the carbon fiber bundle attached with the sizing agent. The amount of sizing agent attached in the present invention is determined by taking 2.0 ± 0.5 g of the carbon fiber bundle attached with the sizing agent and measuring the change in mass before and after heat treatment at 450°C in a nitrogen atmosphere for 15 minutes, and dividing the change in mass by the mass before heat treatment (by mass%). When the sizing agent content is 0.6% by mass or more, sufficient bundling properties can be imparted to the carbon fiber bundle attached with the sizing agent, the accumulation of fuzz can be suppressed and the thickness of the carbon fiber bundle attached with the sizing agent can be reduced, and the coefficient of variation of thickness can also be reduced. When the sizing agent content is 1.1% by mass or less, the resin impregnation during higher-order processing is excellent and the development of strand tensile strength is stable. If the amount of sizing agent attached is too high, the resin impregnation properties decrease and the development of strand tensile strength becomes unstable, and as a result the carbon fiber bundle attached with the sizing agent may have poor quality stability. Therefore, the sizing agent content is preferably 0.7% by mass or more and 1.0% by mass or less, and more preferably 0.8% by mass or more and 0.9% by mass or less. The amount of sizing agent attached to the carbon fiber bundle with the sizing agent attached in the present invention can be evaluated by the method described in the Examples section.

[0036] In the present invention, it is preferable to process the carbon fiber bundles impregnated with a sizing agent using a hot roller with a surface temperature of 100°C to 180°C. If the surface temperature of the hot roller is 100°C or higher, the sizing agent applied to the carbon fiber bundles will dry sufficiently, and as a result, the thickness of the carbon fiber bundles will not increase easily. If the surface temperature of the hot roller is 180°C or lower, the sizing agent applied to the carbon fiber bundles will not decompose and the bundle-forming properties will not deteriorate easily, the quality of the carbon fiber bundles themselves can be maintained, and the occurrence of fuzzy wrapping around the rollers during carbon fiber bundle manufacturing will not increase, so the number of untangled fuzzy fibers will not increase and the coefficient of variation of the thickness of the carbon fiber bundles will not increase easily. It is preferable to process the carbon fiber bundles using a hot roller surface temperature corresponding to the viscosity of the sizing agent. If the viscosity of the sizing agent at 30°C is 500 mPa·s or higher and 2000 mPa·s or lower, it is preferable to process them using a hot roller with a temperature of 150°C to 170°C. Because sizing agents tend to flow easily and unevenly, leading to cracking after drying, it is preferable to fix the sizing agent before it becomes uneven. A temperature of 150°C or higher is preferable because it can sufficiently suppress unevenness, and a temperature of 170°C or lower is preferable because it can suppress the decomposition of low-viscosity sizing agents. If the viscosity of the sizing agent at 30°C exceeds 2000 mPa·s, it is preferable to process it with a hot roller at a temperature of 120°C or higher and below 150°C. It is preferable to dry at a low temperature to allow time for penetration into the interior, as this facilitates film formation during drying. A temperature of 120°C or higher is preferable because it can sufficiently fix the sizing agent before it becomes thick, and a temperature of 150°C or lower is preferable because it can sufficiently penetrate into the interior.

[0037] The carbon fiber bundles with sizing agent attached according to the present invention are preferably substantially untwisted. A carbon fiber bundle being substantially untwisted means that it is either completely untwisted, or, if twisted, has 0.5 turns or less per meter. When carbon fiber bundles are untwisted, they exhibit excellent spreading properties when used as reinforcing fibers for CFRP, often resulting in superior physical properties and quality of the CFRP.

[0038] Next, the method for producing carbon fiber bundles with sizing agent attached according to the present invention will be described.

[0039] In the production of carbon fiber bundles, a polyacrylonitrile-based precursor fiber bundle is first obtained. It is preferable to use a polyacrylonitrile copolymer as the raw material for producing the polyacrylonitrile-based precursor fiber bundle. In this invention, a polyacrylonitrile copolymer refers to one in which acrylonitrile is the main component of the copolymer. The main component usually refers to a component that accounts for 90% to 100% by mass of the polymer. As monomers that can be used as copolymer components other than the main component, monomers containing one or more carboxylic acid groups or amide groups are preferably used from the viewpoint of promoting flame resistance. For example, monomers containing carboxylic acid groups include acrylic acid, methacrylic acid, itaconic acid and their alkali metal salts, and ammonium salts. A monomer containing an amide group is acrylamide.

[0040] In the production of polyacrylonitrile precursor fiber bundles, the method for producing the polyacrylonitrile copolymer can be selected from known polymerization methods.

[0041] In producing polyacrylonitrile precursor fiber bundles, either a wet-dry spinning method or a wet spinning method may be used for spinning, but it is preferable to use a wet-dry spinning method which is advantageous for the strand tensile strength of the resulting carbon fiber bundles. The spinning process consists of a spinning step in which a spinning solution is discharged from a spinneret into a coagulation bath and spun; a washing step in which the fibers obtained in the spinning step are washed in a water bath; a water bath stretching step in which the fiber bundles obtained in the washing step are stretched in a water bath; and a drying heat treatment step in which the fiber bundles obtained in the water bath stretching step are dried and heat treated. If necessary, a steam stretching step is included in which the fiber bundles obtained in the drying heat treatment step are steam stretched. The order of each step can be changed as appropriate.

[0042] The spinning solution is obtained by dissolving the polyacrylonitrile copolymer described above in an organic solvent such as dimethyl sulfoxide, dimethylformamide, or dimethylacetamide, or in an aqueous solution such as nitric acid, zinc chloride, or sodium rhodane, in which the polyacrylonitrile copolymer is soluble.

[0043] The above-mentioned coagulation bath preferably contains a solvent such as dimethyl sulfoxide, dimethylformamide, or dimethylacetamide, which was used as the solvent for the spinning stock solution, and a coagulation-promoting component. As the coagulation-promoting component, a component that does not dissolve the above-mentioned polyacrylonitrile copolymer and is compatible with the solvent used in the spinning solution can be used. Specifically, it is preferable to use water as the coagulation-promoting component.

[0044] In the washing process described above, it is preferable to use a washing bath consisting of multiple stages with a temperature of 30°C to 98°C. Furthermore, it is preferable that the stretching ratio in the water bath stretching process be between 2 and 6 times.

[0045] After the water bath stretching process, it is preferable to apply an oil made of silicone or the like to the fiber bundle to prevent the individual fibers from fusing together. It is preferable to use a modified silicone oil, and more preferably one that contains highly heat-resistant amino-modified silicone.

[0046] The drying heat treatment process can utilize known methods. For example, a drying temperature of 100 to 200°C is exemplified.

[0047] After the above-described water washing, water bath stretching, oil application, and drying heat treatment steps, a polyacrylonitrile-based precursor fiber bundle suitable for obtaining the carbon fiber bundle of the present invention can be obtained by steam stretching as needed. Steam stretching is preferably performed in pressurized steam with a stretching ratio of 2 to 6 times.

[0048] The single fiber fineness of the polyacrylonitrile-based precursor fiber bundle is preferably 0.58 dtex or more and 1.20 dtex or less, from the viewpoint of increasing the strand tensile strength of the carbon fiber bundle. Such single fiber fineness is preferably 0.70 dtex or more and 1.15 dtex or less, and more preferably 1.05 dtex or more and 1.10 dtex or less.

[0049] In a method for producing carbon fiber bundles, carbon fiber bundles are obtained by subjecting a polyacrylonitrile-based precursor fiber bundle to a flame-retardant step, a pre-carbonization step, and a carbonization step.

[0050] In the present invention, the flame resistance improvement step refers to heat-treating a polyacrylonitrile-based precursor fiber bundle at 200°C or higher and 300°C or lower in an atmosphere containing oxygen.

[0051] The flame resistance improvement of polyacrylonitrile-based precursor fibers is preferably controlled such that the density of the flame-resistant fiber bundle falls within the range of 1.25 g / cm ３ or more and 1.35 g / cm ３ or less. Specifically, the flame resistance improvement temperature is preferably carried out in air at 200°C or higher and 300°C or lower, more preferably in air at 210°C or higher and 285°C or lower, and even more preferably in air at 230°C or higher and 280°C or lower. The preferred ranges of the treatment time and temperature for these flame resistance improvements vary depending on the characteristics of the polyacrylonitrile-based precursor fiber bundle and the copolymer composition of the polyacrylonitrile copolymer. The densities of the flame-resistant fiber bundle and the carbon fiber bundle are evaluated by the method described in the Examples section.

[0052] In order to increase the strand tensile strength of the carbon fiber, in the flame resistance improvement step, the obtained flame-resistant fiber preferably has a ratio of the peak intensity at 1,453 cm －１ in the infrared spectrum to the peak intensity at 1,370 cm －１ in the range of 0.60 or more and 0.75 or less, and a ratio of the peak intensity at 1,254 cm －１ in the infrared spectrum to the peak intensity at 1,370 cm －１ in the range of 0.50 or more and 0.65 or less. The peak at 1,453 cm －１ in the infrared spectrum is derived from alkenes and decreases as the flame resistance improvement progresses. The peaks at 1,370 cm －１ and 1,254 cm －１ in the infrared spectrum are peaks derived from the flame-resistant structure and increase as the flame resistance improvement progresses. A standard flame-resistant fiber with a specific gravity of 1.35 has a ratio of the peak intensity at 1,453 cm －１ to the peak intensity at 1,370 cm －１ in the range of about 0.63 or more and 0.69 or less. In the flame resistance improvement step of the present invention, it is preferable that the obtained flame-resistant fiber contains more alkene-derived structures. 1,370 cm－１ 1,254 cm⁻¹ relative to the peak intensity －１ The ratio of peak intensities decreases as flame resistance progresses, with a particularly large decrease in the initial stages. However, depending on the flame resistance conditions, the peak intensity ratio may not fall below 0.65 even with increased time, so it is advisable to adjust the conditions.

[0053] To achieve a balance between these two peak intensity ratios within the desired range, the basic principle is to set the conditions by focusing on the following: a low amount of copolymer components in the polyacrylonitrile polymer constituting the precursor fiber, a high degree of crystal orientation in the precursor fiber, a low single-fiber fineness in the precursor fiber, and a high flame-retardant temperature in the later stages. Specifically, flame-retardant treatment is performed in two stages: a first flame-retardant treatment and a second flame-retardant treatment. In the second flame-retardant treatment, the fiber obtained in the first flame-retardant treatment is subjected to infrared spectroscopy at 1,370 cm⁻¹. －１ 1,453 cm⁻¹ relative to the peak intensity －１ The ratio of peak intensities is in the range of 0.60 to 0.75, and the infrared spectrum is 1,370 cm⁻¹. －１ 1,254 cm⁻¹ relative to the peak intensity －１ The flame-retardant treatment is performed until the peak intensity ratio is in the range of 0.50 to 0.65. In addition, in the first flame-retardant treatment process, the infrared spectrum at 1,370 cm⁻¹ is used. －１ 1,453 cm⁻¹ relative to the peak intensity －１ It is preferable to make the material flame-resistant until the ratio of the peak intensities is in the range of 0.98 to 1.10.

[0054] In the first flame-retardant treatment step, the flame-retardant temperature is preferably 200°C to 250°C, more preferably 230°C to 250°C, in order to control the infrared spectrum within the range described above.

[0055] The flame-retardant temperature in the second flame-retardant process is usually higher than that of the first flame-retardant process. To shorten the flame-retardant time in the second flame-retardant process, the flame-retardant temperature should be adjusted to a higher temperature, but the appropriate flame-retardant temperature depends on the properties of the precursor fiber. It is preferable to set the flame-retardant temperature to 265°C to 295°C, more preferably 265°C to 285°C, and even more preferably 270°C to 275°C in order to control the infrared spectrum within the range described above. The flame-retardant temperature does not need to be constant, and a multi-stage temperature setting is also acceptable. To increase the strand tensile strength of the resulting carbon fiber, it is preferable to set a high flame-retardant temperature and a short flame-retardant time.

[0056] In this invention, flame-retardant treatment refers to heat treatment of the precursor fiber in an oxygen-containing atmosphere at a temperature of 200°C to 310°C.

[0057] The flame-retardant time referred to here means the time the fibers remain in the flame-retardant furnace. Flame-retardant fibers refer to fibers after the flame-retardant process but before the pre-carbonization process. The peak intensity referred to here is the absorbance at each wavelength after baseline correction of the spectrum obtained by measuring the infrared spectrum of a small sample of flame-retardant fibers; no peak splitting is performed. The sample concentration for infrared spectrum measurement is diluted with potassium bromide (KBr) to 0.67 mass%. In this way, the infrared spectrum should be measured each time the flame-retardant conditions are changed, and the conditions should be investigated according to the preferred manufacturing method described later. By appropriately controlling the infrared spectral peak intensity ratio of the flame-retardant fibers, the strand tensile strength of the resulting carbon fibers can be controlled.

[0058] In the pre-carbonization step, which pre-carbonizes the fiber bundles obtained in the flame-retardant step, the obtained flame-retardant fibers are usually heat-treated in an inert atmosphere at a maximum temperature of 500°C to 1,000°C until the specific gravity is preferably 1.5 to 1.8. If the maximum temperature of the pre-carbonization step is 500°C or higher, the pre-carbonized fiber bundles can be carbonized in the subsequent carbonization step without breaking due to thermal decomposition. There is no particular upper limit to the maximum temperature of the pre-carbonization step, but it is preferable to be 1,000°C or lower in order to keep it below the carbonization temperature in the subsequent carbonization step. The stretch ratio in the pre-carbonization step is preferably 1.00 to 1.15. More preferably 1.02 to 1.13. If the stretch ratio in the pre-carbonization step is 1.00 or higher, the strand tensile modulus tends to increase, and the strand tensile strength tends to increase. If the stretch ratio in the pre-carbonization step is 1.15 or lower, the strand tensile modulus tends to be suppressed to 280 GPa or lower.

[0059] In the carbonization process of pre-carbonized fibers, it is preferable to carbonize in an inert atmosphere at a maximum temperature of 1,000°C to 1,600°C. More preferably, the maximum temperature is 1,250°C to 1,550°C, and even more preferably 1,300°C to 1,500°C. From the viewpoint of increasing the strength of the resulting carbon fibers, a lower maximum temperature in the carbonization process is preferable, but if it is too low, the strand tensile strength may decrease, so it is preferable to set the temperature taking both into consideration. A carbonization temperature of 1,000°C is sufficient to allow carbonization to proceed sufficiently and increase the crystallite size, while a temperature of 1,600°C or lower is sufficient to maintain a balance between the strand tensile strength and the strand tensile modulus of the carbon fiber bundle.

[0060] The heating rate in the carbonization process is preferably 0.2°C / second or more and 1.1°C / second or less, more preferably 0.3°C / second or more and 1.0°C / second or less, and even more preferably 0.4°C / second or more and 0.6°C / second or less. The heating rate in the carbonization process affects the desorption rate of the decomposition gas and therefore affects the strand tensile strength. In the present invention, the heating rate is defined as the average temperature increase per second when the fiber passes through a region where the temperature of a section exceeds 1,000°C, as the fiber passes through a plurality of sections in the carbonization furnace, where the temperature is controlled to increase in stages. Specifically, for example, if the fiber passes through the section with a temperature of 1,000°C in the carbonization furnace and then through to the next section with a temperature of 1,100°C in 100 seconds, the heating rate will be 1.0°C / second. In another example, if the fibers pass through the 950°C section of the carbonization furnace and then reach the next 1,150°C section in 200 seconds, the heating rate is 1.0°C / second. Furthermore, if the maximum temperature of the carbonization process is set below 1,100°C, the heating rate up to the maximum temperature is used. That is, if the maximum temperature is 1,050°C, and the fibers pass through the 1,000°C section of the carbonization furnace and then reach the next 1,050°C section in 50 seconds, the heating rate is 1.0°C / second. The temperature of the first section in the carbonization furnace is preferably 1,000°C or lower. A heating rate of 0.2°C / second or higher makes it easier to obtain a stable strand tensile modulus. A heating rate of 1.1°C / second or less makes it easier to suppress the decrease in strand tensile strength.

[0061] The carbon fiber bundle obtained as described above is preferably subjected to an oxidation treatment on its surface. For example, oxygen-containing functional groups can be introduced to the surface of the carbon fibers by electrolytic surface treatment. For the electrolytic surface treatment of the present invention, gas-phase oxidation, liquid-phase oxidation, and liquid-phase electrolytic oxidation can be used, but liquid-phase electrolytic oxidation is preferably used from the viewpoint of high productivity and uniform treatment. In the present invention, there are no particular restrictions on the method of liquid-phase electrolytic oxidation, and any known method may be used.

[0062] In the present invention, examples of electrolytes used in liquid-phase electrolytic oxidation include acidic electrolytes and alkaline electrolytes. Examples of acidic electrolytes include inorganic acids such as sulfuric acid, nitric acid, hydrochloric acid, phosphoric acid, boric acid, and carbonic acid; organic acids such as acetic acid, butyric acid, oxalic acid, acrylic acid, and maleic acid; or salts such as ammonium sulfate and ammonium bisulfate.

[0063] Examples of alkaline electrolytes include aqueous solutions of hydroxides such as sodium hydroxide, potassium hydroxide, magnesium hydroxide, calcium hydroxide, and barium hydroxide; aqueous solutions of carbonates such as sodium carbonate, potassium carbonate, magnesium carbonate, calcium carbonate, barium carbonate, and ammonium carbonate; aqueous solutions of bicarbonates such as sodium bicarbonate, potassium bicarbonate, magnesium bicarbonate, calcium bicarbonate, barium bicarbonate, and ammonium bicarbonate; and aqueous solutions of ammonia, tetraalkylammonium hydroxide, and hydrazine.

[0064] Next, the means for applying the sizing agent constituting the present invention to the carbon fiber bundle will be described. In the present invention, it is preferable to use the sizing agent as a sizing agent-containing solution obtained by diluting the sizing agent component with a solvent. Examples of solvents include water, methanol, ethanol, isopropanol, acetone, methyl ethyl ketone, dimethylformamide, and dimethylacetamide, but among these, an aqueous solution or an aqueous dispersion emulsified with a surfactant is preferably used because it is easy to handle and advantageous from the viewpoint of safety.

[0065] Methods for applying the sizing agent to the carbon fiber bundle include, for example, immersing the carbon fibers in a sizing agent-containing liquid via a roller (immersion method), bringing the carbon fibers into contact with a roller to which the sizing agent-containing liquid has adhered, and spraying the sizing agent-containing liquid onto the carbon fibers in a mist. However, in manufacturing the sizing agent-containing carbon fiber bundle of the present invention, the immersion method is preferably used.

[0066] Furthermore, the sizing agent can be applied in either a batch or continuous manner, but a continuous method is preferred because it offers better productivity and reduces variability. In this case, it is preferable to control the concentration of the sizing agent solution, temperature, and yarn tension so that the sizing agent active ingredient is uniformly applied to the carbon fibers within an appropriate range. It is also preferable to vibrate the carbon fibers with ultrasound when applying the sizing agent.

[0067] Furthermore, in the present invention, it is preferable to apply the sizing agent to the carbon fiber bundle and then pass it through a heated roller as a pre-drying step. The heating temperature in the pre-drying step is preferably 125°C or higher, which is the temperature at which the water is sufficiently dried, and the drying time should be as short as possible. If this step is considered the first drying step, a second drying step may be added, which involves further heat treatment. In this step, it is preferable to perform the heat treatment at a temperature range of 100 to 260°C for 10 to 600 seconds. In addition, the heat treatment in the second drying step can also be performed by microwave irradiation and / or infrared irradiation.

[0068] By using the sizing agent-adhered carbon fiber bundles of the present invention as described above, the quality and grade of carbon fiber reinforced composite materials can be improved when combined with a matrix resin. In particular, the quality and grade of pressure vessels manufactured using methods such as filament winding molding, where carbon fiber bundles are wound at high speed and the accumulation of fuzz during the process tends to be a problem, can be improved. The carbon fiber reinforced composite material described above preferably contains sizing agent-adhered carbon fiber bundles and a matrix resin, and may contain other components as needed. Specifically, it can be manufactured by impregnating sizing agent-adhered carbon fiber bundles with a matrix resin, or by curing the same, and the same applies to the pressure vessels described above.

[0069] The upper and lower limits of the numerical ranges described above can be combined in any way. Furthermore, the present invention will be specifically described below with reference to examples.

[0070] The methods for measuring various physical properties used in this invention are as follows.

[0071] <Strand Tensile Test of Carbon Fiber Bundles> <Strand Tensile Strength and Strand Tensile Modulus of Carbon Fiber Bundles> The strand tensile strength and strand tensile modulus of carbon fiber bundles were determined according to the resin-impregnated strand strength test method of JIS-R-7608 (2007), using each resin formulation (strand tensile strength was determined using resin formulation A, resin formulation B, and strand tensile modulus was determined using resin formulation A). The resin compositions A and B used in each resin formulation are described below. The curing conditions performed using resin composition A or resin composition B are as shown for each of the resin formulations A and B described above. Ten strands of carbon fiber bundles were measured, and their arithmetic mean values ​​were taken as the strand tensile strength and strand tensile modulus. The strand tensile modulus was measured in the strain range of 0.1 to 0.6%.

[0072] [Resin Composition A] The compounds (i) to (iii) were used to make resin composition A in the following proportions: (i) (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexylcarboxylate (100 parts by mass) (ii) boron trifluoride monoethylamine (3 parts by mass) (iii) acetone (4 parts by mass).

[0073] As compound (i), Celoxide P2021P (manufactured by Daicel Corporation) was used.

[0074] [Resin Composition B] The compounds (i) to (iii) were used to make resin composition B in the following proportions: (i) 2,2'-[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bisoxirane (80 parts by mass) (ii) 1,4-bis(2,3-epoxypropoxy)butane (20 parts by mass) (iii) 2,2'-dimethyl-4,4'-methylenebis(cyclohexylamine) (35 parts by mass) "Araldite®" LY1564 SP CI was used as the mixture of (i) and (ii), and "Baxxodur®" EC331 (35 parts by mass) was used as the compound for (iii).

[0075] <Resin Incorporation Ratio> The gap between the blade and roller (100 mm in diameter) of the filament winder's kiss roller is set to 0.13 mm. A resin with a viscosity of 450 mPa·s is placed in the kiss roller bath. The carbon fiber bundle is fed out at a speed of 60 m / min and the creel tension is set to 10 N. The carbon fiber bundle is run using the filament winder, passes through the kiss roller bath to coat it with resin, and then 20 m of the carbon fiber bundle is wound onto a weighed paper tube. After winding, the amount of resin incorporated into the carbon fiber bundle (hereinafter referred to as "resin incorporation amount") is weighed. The weight of the carbon fiber bundle before resin impregnation of 20 m is calculated from the basis weight (g / m) of the carbon fiber bundle, and the resin incorporation ratio is calculated using the following formula. The basis weight (g / m) of the carbon fiber bundle is measured by weighing a carbon fiber bundle cut to 1 m. Three samples are used for measurement, and the arithmetic mean is used. Resin incorporation ratio = Amount of resin incorporated (g) / Weight of 20m carbon fiber bundle before resin impregnation (g) <Single fiber diameter of carbon fiber bundle> For a carbon fiber bundle consisting of many carbon filaments to be measured, the mass per unit length A ｆ (g / m) and density B ｆ (g / cm ３ The density is determined by the density measurement method described below. The number of filaments in the carbon fiber bundle to be measured is C. ｆ Assuming the cross-section of the carbon fiber is a perfect circle, the single fiber diameter (μm) of the carbon fiber is calculated using the following formula: Single fiber diameter (μm) of carbon fiber = ((A ｆ / B ｆ / C ｆ ) / π) （１／２） ×2 × 10 ３ .

[0076] <Density measurement of flame-resistant yarn fiber bundles and carbon fiber bundles> Density B of flame-resistant yarn fiber bundles and carbon fiber bundles ｆ (g / cm ３ The specific gravity is calculated using the Archimedes method with o-dichlorobenzene as the specific gravity solution. Three samples are used for the measurement, and the arithmetic mean is used.

[0077] <Method for Measuring Viscosity> The compounds were placed in plastic bottles and measured using a Type B viscometer (Tokyo Keiki Seisakusho, Model: BL) while the bottles were heated to 30°C in a water bath. Plastic bottles of 160cc size were used for compounds with a viscosity lower than 10,000 mPa·s, and 100cc size were used for compounds with a viscosity of 10,000 mPa·s or more. For the Type B viscometer, the rotor and rotation speed were set to No. 1 and 30 rpm for compounds predicted to have a viscosity of less than 1,000 mPa·s, No. 3 and 30 rpm for compounds predicted to have a viscosity between 1,000 mPa·s and less than 10,000 mPa·s, and No. 4 and 6 rpm for compounds predicted to have a viscosity of 10,000 mPa·s or more.

[0078] <Amount of sizing agent applied> Take 2.0 ± 0.5 g of carbon fiber bundle coated with sizing agent and weigh it (W １ (Read to the fourth decimal place) Then, an electric furnace (capacity 120 cm³) set to a temperature of 450°C in a nitrogen gas stream of 50 ml / min. ３ The sizing agent was left in the air for 15 minutes to completely decompose. Then, the carbon fiber bundle was transferred to a container in a 20 liter / min dry nitrogen stream and cooled for 15 minutes before being weighed (W ２ ) (Read to the fourth decimal place) and W １ -W ２ The heat loss was determined by the following method. This heat loss was converted to a mass percentage (rounded to the third decimal place) with the carbon fiber bundle coated with the sizing agent set to 100% by mass. This value was then defined as the amount of sizing agent applied (mass %). The measurement was performed twice, and the arithmetic mean was used as the amount of sizing agent applied.

[0079] <Thickness of carbon fiber bundle with sizing agent attached and its coefficient of variation> The carbon fiber bundle with sizing agent attached is oriented so that the horizontal direction is the longer side, and is 0.9 × 10 －４The carbon fiber bundles with sizing agent attached are aligned by applying a load of mN / dtex, and the thickness is measured using a micrometer along a line segment that passes through the center of the circumscribed circle in the cross-section of the bundle and is perpendicular to the long side within that cross-section. The thickness of the above line segment is measured at 50 points at 0.2m intervals in the direction of the long side, and the arithmetic mean of these 50 points is used as the thickness of the carbon fiber bundle with sizing agent attached, and the standard deviation of these 50 points is used as the standard deviation of the thickness of the carbon fiber bundle with sizing agent attached. The coefficient of variation of the thickness of the carbon fiber bundle with sizing agent attached is the ratio of the thickness of the carbon fiber bundle with sizing agent attached measured above to the standard deviation of the thickness, expressed as a percentage ([Standard Deviation] / [Arithmetic Mean] × 100).

[0080] <Measurement of Drape Value> As shown in Figure 1, a 50 cm carbon fiber bundle 2 was suspended vertically from a fixing bar 1, and a weight 3 was attached with a load of 0.0375 [g / tex] in an atmosphere of 25°C. It was left for more than 30 minutes to release the twist. A 30 cm length of the carbon fiber bundle was sampled from the center of its longitudinal direction, extending 15 cm above and below. As shown in Figure 2, the carbon fiber bundle 2 was placed on a rectangular horizontal stand 4, supporting it to prevent it from bending, so that 25 cm of the carbon fiber bundle extended beyond the edge of the stand. The carbon fiber bundle 2 on the horizontal stand 4 was then secured with tape. After that, the support for the carbon fiber bundle 2 extending beyond the horizontal stand 4 was removed, and the horizontal distance L from the fulcrum was measured after 1 second. The measurement was performed once for each of the five carbon fiber bundles 2, for a total of five measurements, and the arithmetic mean was taken as the drape value.

[0081] <Process Passability of Carbon Fiber Bundles> Two metal bars (stainless steel) with a diameter of 50 mm and a surface roughness Rmax of 0.3 μm are placed horizontally at a 150 mm interval, with a height difference between them, so that when the carbon fiber bundle is viewed from the side, it passes through the metal bars while in contact with them at an angle of 0.3925π (rad) ± 0.04π (rad) or -0.3925π (rad) ± 0.04π (rad) with respect to the horizontal plane of each metal bar. The carbon fiber bundle is then passed from the metal bar positioned higher to the metal bar positioned lower. When a carbon fiber bundle with 24,000 filaments is stretched across the metal bars, the unwinding tension from the package is set to 1600 gf, and the metal bars are pulled at 4 m / min by a drive roll for 1 minute. The number of fluff generated on the carbon fiber bundle during the 1 minute after passing through the second metal bar is counted, and the number of fluffs in the process is calculated using the following formula as an indicator of process passability.

[0082] Process fuzz count (fuzz / m) of carbon fiber bundle = Fuzz count (fuzz) / 4 (m) S: Process fuzz count is 1.0 fuzz / m or less A: Process fuzz count exceeds 1.0 fuzz / m and is 1.5 fuzz / m or less B: Process fuzz count exceeds 1.5 fuzz / m and is 5.0 fuzz / m or less C: Process fuzz count exceeds 5.0 fuzz / m When a carbon fiber bundle with 36,000 filaments is stretched across a metal bar, the above release tension was set to 2,400 gf, and the process fuzz count obtained by counting and using the above formula was divided by √1.5 (1.5 to the power of 1 / 2) to standardize the evaluation criteria and determine the passability through the process. A and B are preferred results in the present invention, and A is a more preferred result.

[0083] <Uniformity Index of Carbon Fiber Bundles in Pressure Vessels> A 7.5 L polyethylene liner was placed in a filament winding molding apparatus, and resin composition B, which had been uniformly mixed in advance at 25°C, was fed into the carbon fiber bundles that had been treated with a sizing agent, impregnating them with the resin composition B at a rate of 22 to 28% by mass relative to the total mass of the carbon fiber bundles and resin composition B. By feeding the yarn in the above manner, a hoop layer forming an angle of +89° with respect to the axial direction of the liner and a hoop layer forming an angle of -89° with respect to the axial direction of the liner were wound as the first layer, with a thickness of 1.4 mm. Next, a helical layer forming an angle of +20° with respect to the axial direction of the liner and a helical layer forming an angle of -20° with respect to the axial direction of the liner were fed and wound as the second layer, with a thickness of 2.2 mm. Furthermore, as a third layer, a hoop layer forming an angle of +89° with respect to the axial direction of the liner and a hoop layer forming an angle of -89° with respect to the axial direction of the liner were fed and wound to a thickness of 0.6 mm to obtain an intermediate. After winding the carbon fiber bundle, the intermediate was rotated at a speed of 7 rpm and held in a 20°C environment for 15 minutes. The intermediate was cured under the conditions of resin formulation B to obtain a pressure vessel for pressure resistance testing.

[0084] The pressure vessel was cut perpendicular to the hoop layer, and the cross-section of the carbon fiber bundle was observed to evaluate the stability of the carbon fiber bundle (uniformity of thickness, presence or absence of voids).

[0085] Observations were performed using an optical microscope on 20 fiber bundles in the hoop layer. The thickness was defined as the line segment passing through the center of the cross-section of the carbon fiber bundle and perpendicular to the long side of the cross-section. The standard deviation of the thicknesses at these 20 points was calculated, and the coefficient of variation of the thickness of the carbon fiber bundles in CFRP was determined as a percentage by dividing the thickness by the standard deviation of the thickness ([standard deviation] / [arithmetic mean] × 100).

[0086] A coefficient of variation in thickness of 10% or less was judged as good uniformity, and a coefficient of variation between 10% and 20% was judged as acceptable uniformity. In addition, the void content was confirmed by observing the innermost layer in a cross section perpendicular to the hoop layer of the pressure vessel, within a range of 3 mm in the thickness direction and 30 mm in the radial direction, using an optical microscope, and the uniformity was evaluated according to the following criteria and recorded in the table "Uniformity of carbon fiber bundles in pressure vessel". A and B are preferred results in the present invention, with A being a more preferred result. S: Void content: less than 2.5%, good uniformity A: Void content: less than 4.0%, good uniformity B: Void content: less than 4.0%, acceptable uniformity C: Void content: less than 4.0% D: Void content: 4.0% or more, the present invention will be further explained in detail by examples.

[0087] The compounds used in each example and comparative example are as follows. Viscosities are summarized in Table 1.

[0088]

[0089] a1: Bisphenol A ethylene oxide adduct (average number of moles of ethylene oxide added: 10) a2: Polyethylene glycol (Mn = 600) a3: Propylene oxide / ethylene oxide block copolymer (Mn = 3,750) b1: Polyglycerol polyglycidyl ether (Ex-521) b2: Bisphenol A type epoxy (jER834) c1: 2-ethylhexyl stearate (Example 1) A polyacrylonitrile copolymer copolymerized with itaconic acid was polymerized by solution polymerization using dimethyl sulfoxide as the solvent to produce a polyacrylonitrile copolymer. The spinning solution obtained from the produced polyacrylonitrile copolymer was extruded into the air from a spinneret and introduced into a coagulation bath consisting of an aqueous solution of dimethyl sulfoxide to produce a coagulated yarn by a wet-dry spinning method. After washing this coagulated yarn with water by a conventional method, it was stretched 3.5 times in a hot water bath. Next, an amino-modified silicone-based silicone oil was applied to the fiber bundle after water bath stretching, and a drying and densification treatment was performed using a heated roller at 160°C. The bundle was then stretched 3.7 times in pressurized steam to achieve a total stretch ratio of 13 times. Following this, an entanglement treatment was performed to obtain a polyacrylonitrile-based precursor fiber bundle with a crystal orientation of 93%, a single fiber fineness of 1.11 dtex, and 24,000 single fibers. Next, the flame-retardant temperature was set from 235°C to 267°C, resulting in a flame-retardant fiber with a density of 1.30 g / cm³. ３ The flame-retardant treatment time was adjusted to obtain flame-retardant fiber bundles by stretching polyacrylonitrile-based precursor fiber bundles at a stretch ratio of 1 in an oven with an air atmosphere. The obtained flame-retardant fiber bundles were then run in a nitrogen atmosphere at a temperature of 300 to 800°C, and the stretch ratio was controlled to obtain pre-carbonized fiber bundles. The obtained pre-carbonized fiber bundles were then subjected to carbonization treatment in a nitrogen atmosphere, with the maximum temperature and heating rate controlled. The flame-retardant treatment conditions, pre-carbonization conditions, and carbonization conditions are summarized in Table 2.

[0090]

[0091] The obtained carbon fiber bundles were subjected to liquid-phase electrolytic oxidation and sizing agent coating to obtain the final carbon fiber bundles with sizing agent attached. For liquid-phase electrolytic oxidation, an aqueous sulfuric acid solution was used as the electrolyte, and the electrolytic surface treatment was performed at an electric charge of 10 coulombs per gram of carbon fiber. The carbon fiber bundles subjected to this liquid-phase electrolytic oxidation were then washed with water and dried in heated air at 150°C to remove water, thereby obtaining the carbon fiber bundles.

[0092] The composition of A1 in Table 1 (100% by mass of a1) was used as a sizing agent in an aqueous solution. The sizing agent was applied to the carbon fiber bundles by immersion, and as a first drying step, the bundles were pre-dried for 15 seconds using a hot roller at a surface temperature of 160°C. Subsequently, as a second drying step, the bundles were heat-treated in heated air at 230°C for 35 seconds to obtain carbon fiber bundles with the sizing agent attached. The amount of sizing agent attached was adjusted to 0.9% by mass relative to 100% by mass of the total amount of surface-treated carbon fiber bundles with the sizing agent attached.

[0093] The carbon fiber bundles with the sizing agent attached were evaluated according to the strand tensile tests for each resin formulation described above, as well as the evaluation methods for abrasion fuzz, drape value, thickness, resin incorporation ratio, and uniformity of yarn width after resin impregnation. As summarized in Table 2, carbon fiber bundles with high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundle after FW molding were obtained.

[0094] (Example 2) Except for changing the hot roller temperature in the sizing agent process, carbon fiber bundles were obtained in the same manner as in Example 1 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0095] (Example 3) Except for changing the sizing agent applied in the sizing agent step to A3, a carbon fiber bundle was obtained in the same manner as in Example 1, and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0096] (Example 4) Except for changing the heating rate in the carbonization process as shown in Table 2, carbon fiber bundles were obtained in the same manner as in Example 1 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0097] (Example 5) Carbon fiber bundles were obtained in the same manner as in Example 1, except that the sizing agent applied in the sizing agent step was changed to A4, and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles with stable uniformity of the yarn bundles after FW molding and high mechanical properties were obtained.

[0098] (Example 6) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the amount of sizing agent applied in the sizing agent process was changed as shown in Table 2, and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0099] (Example 7) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the sizing agent applied in the sizing agent step was changed to A5, and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0100] (Example 8) Except for changing the sizing agent applied in the sizing agent step to A3, a carbon fiber bundle was obtained in the same manner as in Example 4, and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0101] (Example 9) Except for changing the heating rate in the carbonization process as shown in Table 2, carbon fiber bundles were obtained in the same manner as in Example 1 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had good mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0102] (Example 10) Except for changing the heating rate in the carbonization process as shown in Table 2, carbon fiber bundles were obtained in the same manner as in Example 1 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had good mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0103] (Example 11) Except for changing the stretching ratio in the preliminary carbonization process and the maximum temperature in the carbonization process as shown in Table 2, carbon fiber bundles were obtained in the same manner as in Example 1 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0104] (Example 12) Except for changing the maximum temperature in the carbonization process as shown in Table 2, and changing the sizing agent applied in the sizing agent process to A3, a carbon fiber bundle was obtained in the same manner as in Example 11, and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, good abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0105] (Example 13) Except for changing the maximum temperature in the carbonization process as shown in Table 2, carbon fiber bundles were obtained in the same manner as in Example 11 and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had good mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundles after FW molding.

[0106] (Example 14) Except for changing the single fiber diameter, number of filaments, flame-retardant temperature, and IR peak intensity ratio of the flame-retardant fiber bundle as shown in Table 2, a carbon fiber bundle was obtained in the same manner as in Example 4 and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, good abrasion resistance, and good uniformity of the yarn bundle after FW molding.

[0107] (Example 15) Carbon fiber bundles were obtained in the same manner as in Example 14, except that the single fiber diameter was changed as shown in Table 2, and various evaluations were performed. The results are summarized in Table 2, and carbon fiber bundles were obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0108] (Example 16) Except for changing the heating rate in the carbonization process as shown in Table 2, a carbon fiber bundle was obtained in the same manner as in Example 15, and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0109] (Example 17) Except for changing the stretching ratio in the preliminary carbonization process, the maximum temperature in the carbonization process, and the heating rate in the carbonization process as shown in Table 2, a carbon fiber bundle was obtained in the same manner as in Example 16, and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0110] (Example 18) Except for changing the maximum temperature in the carbonization process as shown in Table 2 and changing the sizing agent applied in the sizing agent process to A3, a carbon fiber bundle was obtained in the same manner as in Example 17 and various evaluations were performed. The results are summarized in Table 2, and a carbon fiber bundle was obtained that had high mechanical properties, high abrasion resistance, and stable uniformity of the yarn bundle after FW molding.

[0111] (Example 19) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the flame-retardant temperature, the IR peak intensity ratio of the flame-retardant fiber bundles, and the single fiber diameter were changed as shown in Table 2, and various evaluations were performed. The results are summarized in Table 2, and the obtained carbon fiber bundles showed good abrasion resistance and uniformity of the yarn bundles after FW molding. (Example 20) Carbon fiber bundles were obtained in the same manner as in Example 1, except that the temperature in the flame-retardant process, the IR peak ratio of the flame-retardant fiber bundles, the stretching ratio in the pre-carbonization process, and the heating rate in the carbonization process were changed as shown in Table 2, and various evaluations were performed. The results are summarized in Table 2, and the obtained carbon fiber bundles showed high abrasion resistance and uniformity of the yarn bundles after FW molding.

[0112] (Comparative Example 1) Carbon fiber bundles were obtained in the same manner as in Example 3, except that the hot roller temperature in the sizing agent process was changed, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high and the abrasion resistance was good, but the void content in the yarn bundle after FW molding was high.

[0113]

[0114] (Comparative Example 2) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the amount of sizing agent applied in the sizing agent process was changed as shown in Table 3, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high, but the abrasion resistance and the uniformity of the yarn bundle after FW molding were poor.

[0115] (Comparative Example 3) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the amount of sizing agent applied in the sizing agent process was changed as shown in Table 3, and various evaluations were performed. The results are summarized in Table 3. Although the abrasion resistance and mechanical properties were good, the uniformity of the yarn bundle after FW molding was poor.

[0116] (Comparative Example 4) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the sizing agent applied in the sizing agent process was changed to A2, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high, but the abrasion resistance and the uniformity of the yarn bundle after FW molding were poor.

[0117] (Comparative Example 5) Carbon fiber bundles were obtained in the same manner as in Example 4, except that the hot roller temperature in the sizing agent process was changed and the sizing agent applied in the sizing agent process was changed to A6, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties and uniformity of the yarn bundle after FW molding were good, but the abrasion resistance was poor.

[0118] (Comparative Example 6) Carbon fiber bundles were obtained in the same manner as in Example 3, except that the sizing agent applied in the sizing agent process was changed to A7, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high and the abrasion resistance was good, but the void content of the yarn bundle after FW molding was high.

[0119] (Comparative Example 7) Carbon fiber bundles were obtained in the same manner as in Comparative Example 6, except that the hot roller temperature in the sizing agent process was changed, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high and the abrasion resistance was good, but the void content in the yarn bundle after FW molding was high.

[0120] (Comparative Example 8) Carbon fiber bundles were obtained in the same manner as in Example 1, except that the stretching ratio and single fiber diameter in the preliminary carbonization process were changed as shown in Table 3, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high, but the abrasion resistance and the uniformity of the yarn bundle after FW molding were poor.

[0121] (Comparative Example 9) Carbon fiber bundles were obtained in the same manner as in Example 15, except that the amount of sizing agent applied in the sizing agent process was changed as shown in Table 3, and various evaluations were performed. The results are summarized in Table 3. The mechanical properties were high, but the abrasion resistance and the uniformity of the yarn bundle after FW molding were poor.

[0122] 1. Fixing bar 2. Carbon fiber bundle 3. Weight 4. Horizontal platform

Claims

1. A carbon fiber bundle with a sizing agent applied, having 6,000 to 50,000 filaments, a thickness of 0.10 mm to 0.20 mm, a coefficient of variation of thickness of 6.5% or less, a drape value of 6 cm to 9 cm, and a strand tensile strength (strand tensile strength A') of 4.9 GPa to 6.7 GPa, evaluated according to JIS R7608 (2007) using the following resin formulation A. Resin formulation A: Contains 97% by mass or more of (3',4'-epoxycyclohexane)methyl-3,4-epoxycyclohexylcarboxylate, and has an epoxy equivalent of 130 g / eq and a viscosity of 240 mPa·s at 25°C. The composition is 100 / 3 / 4 (parts by mass) of epoxy resin / boron trifluoride monoethylamine / acetone, and the curing conditions are atmospheric pressure, temperature of 125°C, and time of 30 minutes.

2. The sizing agent-adhered carbon fiber bundle according to claim 1, wherein the strand tensile strength A' is 5.6 GPa or more and 6.7 GPa or less, and the strand tensile strength (strand tensile strength B') evaluated according to JIS R7608 (2007) using the following resin formulation B is 5.3 GPa or more and 6.4 GPa or less. Resin formulation B: The main component is 80% by mass of 2,2'-[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bisoxirane and 20% by mass of 1,4-bis(2,3-epoxypropoxy)butane, and 2,2'-dimethyl-4,4'-methylenebis(cyclohexylamine) is used as a curing agent, with a ratio of 100 / 35 (parts by mass) of the main component to the curing agent. The curing conditions are atmospheric pressure, temperature 80°C, time 120 minutes, followed by atmospheric pressure, temperature 110°C, time 240 minutes.

3. The sizing agent-adhered carbon fiber bundle according to claim 2, wherein the strand tensile strength A' is 5.9 GPa or more and 6.7 GPa or less, and the strand tensile strength (strand tensile strength B') evaluated according to JIS R7608 (2007) using resin formulation B is 5.6 GPa or more and 6.4 GPa or less.

4. A sizing agent-adhered carbon fiber bundle according to claim 1 or 2, wherein the resin incorporation ratio is 0.45 or more and 0.64 or less.

5. A sizing agent-adhered carbon fiber bundle according to claim 1 or 2, wherein the single fiber diameter is 6.6 μm or more and 7.3 μm or less.

6. The sizing agent-adhered carbon fiber bundle according to claim 3, wherein the strand tensile strength A' is 6.3 GPa or more and 6.7 GPa or less, and the strand tensile strength B' is 6.0 GPa or more and 6.4 GPa or less.

7. A carbon fiber bundle with a sizing agent attached, according to claim 1 or 2, wherein the strand tensile modulus, as evaluated using resin formulation A in accordance with JIS R7608 (2007), is 240 GPa or more and 275 GPa or less.

8. The carbon fiber bundle with sizing agent attached according to claim 1 or 2, wherein the amount of sizing agent attached is 0.7% by mass or more and 1.0% by mass or less of the carbon fiber bundle with sizing agent attached, relative to 100% by mass of the carbon fiber bundle with sizing agent attached.

9. The sizing agent-attached carbon fiber bundle according to claim 1 or 2, wherein the sizing agent does not contain a cyclic oxygen-containing compound.

10. The sizing agent is a carbon fiber bundle to which a sizing agent is attached, wherein the sizing agent has a viscosity of 500 mPa·s or more and 1,400 mPa·s or less at 30°C, according to claim 1 or 2.

11. The carbon fiber bundle with sizing agent attached according to claim 1, wherein the coefficient of variation of thickness is 4.3% or less, and the drape value is 6.5 cm or more and 7.5 cm or less.

12. A carbon fiber reinforced composite material using a sizing agent-adhered carbon fiber bundle according to claim 1 or 2.

13. A pressure vessel using a carbon fiber bundle with a sizing agent attached, as described in claim 10.

14. A method for producing carbon fiber bundles with a sizing agent attached, comprising the steps of: flame-retardant treatment, pre-carbonization, and carbonization of a polyacrylonitrile-based carbon fiber precursor fiber bundle having 6,000 to 50,000 filaments; oxidation treatment of the fiber bundle obtained in the carbonization step to obtain a carbon fiber bundle; and immersion of the carbon fiber bundle in a sizing agent-containing liquid to set the amount of sizing agent attached to 100% by mass of the carbon fiber bundle with the sizing agent attached to 0.6% by mass or more and 1.1% by mass or less, wherein the sizing agent has a proportion of cyclic oxygen-containing compound components of 50% by mass or less in the total amount of sizing agent, and a viscosity of 200 mPa·s or more and 2,000 mPa·s or less at 30°C, and is substantially untwisted.

15. A polyacrylonitrile-based carbon fiber precursor fiber bundle having a single fiber fineness of 0.58 dtex or more and 1.20 dtex or less, and a filament count of 6,000 to 42,000, is subjected to flame retardation in two stages: a first flame retardation stage and a second flame retardation stage. In the second flame retardation stage, the infrared spectrum at 1,370 cm⁻¹ is measured. －１ 1,453 cm⁻¹ relative to the peak intensity －１ The ratio of peak intensities is in the range of 0.70 to 0.75, and the infrared spectrum is at 1,370 cm⁻¹. －１ 1,254 cm⁻¹ relative to the peak intensity －１ The method for producing a carbon fiber bundle with a sizing agent attached, according to claim 14, comprising: flame retardation until the ratio of peak intensities is in the range of 0.50 to 0.65; a preliminary carbonization step in which the fiber bundle obtained in the second flame retardation step is subjected to a stretch ratio of 1.00 to 1.15 in an inert atmosphere with a maximum temperature of 600°C to 800°C; and a carbonization step in which the fiber bundle obtained in the preliminary carbonization step is carbonized in an inert atmosphere with a maximum temperature of 1000°C to 1600°C at a heating rate of 0.4°C / second to 1.1°C / second.

16. A method for producing a carbon fiber bundle with a sizing agent attached, according to claim 15, wherein in the step of immersing the carbon fiber bundle in a sizing agent-containing liquid, the amount of sizing agent attached is 0.7% by mass or more and 1.0% by mass or less, the sizing agent does not contain cyclic oxygen-containing compounds, has a viscosity of 500 mPa·s or more and 1,400 mPa·s or less at 30°C, and is processed with a hot roller with a surface temperature of 150°C or more and 170°C or less.

Images