Rotating components and their manufacturing methods

Abstract

A lightweight rotating component with higher tensile strength and a method for manufacturing the same are provided. The high-speed rotating component is a carbon fiber reinforced molded body using composite material 10 as the reinforcing fiber. In composite material 10, a structure 14 composed of a plurality of carbon nanotubes 17 is formed on the surface of each carbon fiber 11 of the carbon fiber bundle 12 constituting composite material 10. The carbon nanotubes 17 have a curved shape. The carbon nanotubes 17 are attached to the surface of the curved carbon fiber 11 in various postures to form the structure 14. A first sizing agent crosslinks the directly contacting carbon nanotubes 17 with each other through the structure of carbodiimide obtained by the reaction of functional groups derived from carbon nanotubes 17 and carbodiimide groups.

Description

Technical Field

[0001] This invention relates to a rotating component and a method for manufacturing the same. Prior Technology

[0002] As a high-speed rotating component, a component known is the rotor of a surface magnet type motor (see, for example, Patent Document 1). Regarding the rotor of the surface magnet type motor, a plurality of permanent magnets are assembled on its outer circumferential surface. To prevent these permanent magnets from peeling off and scattering from the rotor due to centrifugal force, the rotor is embedded in the hollow interior of the rotating component. Since a force from the permanent magnets acting outwards in the diametrical direction acts on the rotating component, high tensile strength is required in the circumferential direction of the rotating component. Furthermore, centrifugal force also acts on the mass of the rotating component itself. Therefore, as a rotating component, a rotating component made of carbon fiber reinforced plastic with carbon fiber as the reinforcing fiber is also known. In addition, as a high-speed rotating component, there is also the cylindrical rotating drum of a centrifuge used for centrifugal separation. [Existing technical documents] [Patent Literature]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2003-319581 Summary of the Invention

[0004] [The problem the invention aims to solve]

[0005] Rotating components made from carbon fiber reinforced plastics as described above are lightweight and have high tensile strength, but even higher tensile strength is desired to cope with the high-speed rotation of surface magnet motors and centrifuges.

[0006] The purpose of this invention is to provide a lightweight rotating component with higher tensile strength and a method for manufacturing the same. [Methods used to solve problems]

[0007] The rotating component of the present invention includes a carbon fiber wound in the circumferential direction, a matrix resin covering the carbon fiber, and a structure disposed on the surface of the carbon fiber. The structure is composed of a plurality of carbon nanotubes with curved portions and forms a mesh structure with contact portions in which the carbon nanotubes are in direct contact with each other.

[0008] The method for manufacturing a rotating component of the present invention includes: a structure forming step, wherein carbon fibers are immersed in a dispersion liquid subjected to ultrasonic vibration while dispersing carbon nanotubes with curved portions, so that a plurality of the aforementioned carbon nanotubes are attached to the aforementioned carbon fibers to form a structure on the surface of the aforementioned carbon fibers, the aforementioned structure having a mesh structure of contact portions in which the aforementioned carbon nanotubes are in direct contact with each other; and a molding step, wherein a matrix resin is coated on the aforementioned carbon fibers on which the aforementioned structure has been formed, the aforementioned carbon fibers coated with the aforementioned matrix resin are wound on a mandrel, and the aforementioned mandrel is pulled out after the aforementioned matrix resin has hardened. [Invention Benefits]

[0009] According to the present invention, since carbon fibers with a structure composed of carbon nanotubes are formed on the surface of the matrix resin coating, it is possible to provide a lightweight rotating component with higher tensile strength. Simple Explanation of the Diagram

[0010] Figure 1 is an explanatory diagram showing the main structure of a surface magnet type motor according to an embodiment. Figure 2 is a partial cross-sectional view showing the composite material in the rotating component. Figure 3 is an explanatory diagram illustrating the structure of the composite material. Figure 4 is an illustrative diagram showing the state of the first sizing agent adhering to CNTs. Figure 5 is an explanatory diagram showing the state of the first sizing agent adhering to the contact portion where CNTs are in contact with each other. Figure 6 is an illustrative diagram showing other states of the first sizing agent adhering to CNTs. Figure 7 is an illustrative diagram showing other states of the contact area where the first sizing agent adheres to the CNTs in contact with each other. Figure 8 is an explanatory diagram showing the main manufacturing sequence of the rotating components. Figure 9 is an illustrative diagram showing the structure of the attachment device for attaching CNTs to carbon fibers. Figure 10 is an illustrative diagram showing the unfolding of the carbon fiber bundle on the guide roller. Figure 11 is an illustrative diagram showing the positions through which carbon fibers pass in the dispersion. Figure 12 is a perspective view illustrating an example of a filament winder. Figure 13 is an explanatory diagram illustrating an example of a resin coating apparatus. Figure 14 is a graph illustrating an example of how the heating temperature changes in stages during the curing of the matrix resin. Figure 15 is an illustrative diagram showing the cross-linking state of carbon fibers. Figure 16 shows an image of the internal structure of a rod made of composite material, as revealed by computed tomography (CT). Figure 17 shows an X-ray CT image of the internal structure of a rod using carbon fiber precursor. Figure 18 is an illustrative diagram showing the state of the second sizing agent adhering to CNT. Figure 19 is an SEM image showing the bending state of the CNT material used in the embodiment. Figure 20 is a graph showing the results of the three-point bending test in Example 3. Figure 21 is a graph showing the results of the NOL ring test in Example 4. Figure 22 is a graph showing the fiber lengths cut using the fragmentation method in Example 5. Figure 23 is a SEM image showing the adhesion of the first sizing agent to the structure with a sizing agent mass ratio Rm of 0.8%. Figure 24 is a SEM image showing the adhesion of the first sizing agent to the structure with a sizing agent mass ratio Rm of 1.1%. Figure 25 is a SEM image showing the adhesion of the first sizing agent to the structure with a sizing agent mass ratio Rm of 1.5%. Figure 26 is a graph showing the results of the NOL ring test in Example 6 and Comparative Example 6. Figure 27 is a magnified graph showing the areas in the NOL ring test results of Example 6 and Comparative Example 6 where the fiber volume content is 60% or more and the tensile strength (index value) is 90 or more. Figure 28 shows a photograph of a ring-shaped test piece damaged under fault failure mode. Figure 29 shows a photograph of a ring-shaped test piece that failed under the interlaminar failure mode. Figure 30 is a SEM image showing a cross-section of the observation sample taken from the annular test piece in Example 8. Figure 31 is a SEM image showing a cross-section of the observation sample taken from the annular test piece in Comparative Example 8. Figure 32 is a further magnified FE-SEM image showing a cross-section of the observation sample taken from the annular test piece in Example 8. Figure 33 is a further magnified FE-SEM image showing a cross-section of the observation sample taken from the annular test piece in Comparative Example 8. Implementation

[0011] [The form in which the invention is carried out]

[0012] Figure 1 illustrates the rotating component 2 according to an embodiment. In this example, the rotating component 2 is configured as an anti-scattering component for the surface magnet type motor 3. The rotating component 2 is an annular body, and in this example, it has a cylindrical shape. The rotating component 2 is externally embedded in the rotor 4. That is, the rotating component 2 is assembled and fixed by pressing the rotor 4 into its hollow interior, and rotates integrally with the rotor 4. Therefore, the rotating component 2 rotates along its circumferential direction. A plurality of permanent magnets 5 are spaced apart at predetermined intervals along the circumferential direction of the rotor 4 and embedded in the outer circumferential surface of the rotor 4. The rotating component 2 can resist centrifugal force and suppress the permanent magnets 5, so that when the rotor 4 rotates at high speed, the permanent magnets 5 will not peel off and scatter to the outside of the rotor 4 in the diametrical direction due to the centrifugal force acting on them. Alternatively, the rotating component 2 can also be a narrow-width (short length in the axial direction) annular shape, etc. Furthermore, a plurality of narrow-width annular rotating components can also be used in combination.

[0013] As shown in the schematic diagram of Figure 2, the rotating component 2 is a carbon fiber reinforced molded body (carbon fiber reinforced plastic) having a composite material 10 and a matrix resin M containing carbon fibers 11 (see Figure 3) covering the composite material 10. The composite material 10 (carbon fibers 11) is wound along the circumferential direction of the rotating component 2, that is, the rotation direction (direction A in the figure), and a plurality of composite materials 10 form one or more fiber layers in the diametrical direction of the rotating component 2. The fiber layer is composed of composite materials 10 closely arranged along the axial direction.

[0014] The composite material 10 is wound along the circumferential direction of the rotating component 2. Wounding along the circumferential direction of the rotating component 2 means that the composite material 10 has a component along the circumferential direction of the rotating component 2, that is, wound at a predetermined tilt angle (≠0°) relative to the axial direction of the rotating component 2. Therefore, as will be described later, the fiber layer is formed by either annular winding (parallel winding) of the composite material 10 on the outer circumferential surface of the mandrel at a substantially perpendicular (tilt angle ≒90°) position to its axis, or by winding the composite material 10 at a tilt angle less than 90° relative to the axis (spiral winding). From the perspective of the required characteristics of the rotating component 2, the tilt angle is to prevent the collapse of the annular layer during assembly rings, etc., and it can be arbitrarily set. In the case of multiple fiber layers, the tilt angle can be determined for each fiber layer.

[0015] In addition, in Figure 2, for ease of explanation, the composite material 10 is depicted in an exaggerated manner, and each cross section of the composite material 10 is depicted as a circle to facilitate differentiation, but in reality, it is impossible to distinguish them in this way.

[0016] [Composite Materials] In Figure 3, the composite material 10 includes a carbon fiber bundle 12 that bundles a plurality of continuous carbon fibers 11. A structure 14 is formed on the surface of each carbon fiber 11, and a first sizing agent 15 is attached to the structure 14 (see Figure 4).

[0017] In the carbon fibers 11 constituting the carbon fiber bundle 12, the fiber axis directions of each carbon fiber 11 are aligned, and they are generally not intertwined. The fiber axis direction is the direction of the axis (extension direction) of the carbon fiber 11. In this example, the carbon fiber bundle 12 is composed of 12,000 carbon fibers 11. The number of carbon fibers 11 constituting the carbon fiber bundle 12 is not particularly limited, and can be set to a range of more than 10,000 and less than 100,000. In addition, in Figure 3, only a dozen or so carbon fibers 11 are depicted for ease of illustration. Furthermore, in this example, the composite material 10 is composed of a plurality of carbon fibers 11 having the above-mentioned structure 14 and the first sizing agent 15, but the composite material 10 may also be a single carbon fiber 11 having the above-mentioned structure 14 and the first sizing agent 15. In addition, in the following description, the fiber on which the structure 14 is formed on the surface (in this example, carbon fiber 11) and its structure 14 are sometimes referred to together as CNT composite fiber.

[0018] The degree of entanglement of the carbon fibers 11 in the carbon fiber bundle 12 can be assessed by observing the disorder of the carbon fibers 11. For example, the carbon fiber bundle 12 can be observed at a fixed magnification using a scanning electron microscope (SEM), and the length of a predetermined number (e.g., 10 fibers) of carbon fibers 11 within the observation range (a predetermined length range of the carbon fiber bundle 12) can be measured. The degree of disorder of the carbon fibers 11 can be assessed based on the length variation of the predetermined number of carbon fibers 11 obtained from this measurement, the difference between the maximum and minimum values, and the standard deviation. Alternatively, the degree of entanglement can be measured according to, for example, the entanglement measurement method of "Test Method for Chemical Fiber Filaments" (JIS L1013:2010) to determine that the carbon fibers 11 are generally not entangled. The smaller the measured degree of entanglement, the less entanglement there is between the carbon fibers 11 in the carbon fiber bundle 12.

[0019] The carbon fiber bundles 12, which are mostly unentangled or minimally entangled with each other, allow for easy and uniform unfolding of the carbon fibers 11. This facilitates the uniform attachment of CNT 17 to each of the original carbon fibers 11, and the uniform impregnation of the matrix resin M into the carbon fiber bundles 12 within the CNT composite fiber helps to enhance the strength of each carbon fiber 11.

[0020] The carbon fiber 11 is not particularly limited. It can be PAN-type or pitch-type carbon fibers obtained by calcining organic fibers derived from polyacrylic nitrile, rayon, pitch, or petroleum, coal, or coal tar; or carbon fibers obtained by calcining organic fibers derived from wood or plant fibers. Commercially available products are also acceptable. Furthermore, the diameter and length of the carbon fiber 11 are not particularly limited. It is preferable to use carbon fiber 11 with a diameter between approximately 5 μm and 20 μm, and even more preferably between 5 μm and 10 μm. Long strips of carbon fiber are preferred, with a length of 50 m or more, between 100 m and 100,000 m, and even more preferably between 100 m and 10,000 m. Additionally, the carbon fiber 11 can be cut shorter when forming prepregs or carbon fiber reinforced molded articles.

[0021] As described above, a structure 14 is formed on the surface of the carbon fiber 11. The structure 14 is formed by the intertwining of a plurality of carbon nanotubes (hereinafter referred to as CNTs) 17. The CNTs 17 constituting the structure 14 are uniformly dispersed and intertwined on almost the entire surface of the carbon fiber 11, forming a mesh structure in which a plurality of CNTs 17 are connected in an intertwined state. The connection referred to here includes physical connection (simple contact) and chemical connection. There are no media such as surfactants or adhesives between the CNTs 17, and the CNTs 17 are in direct contact with each other.

[0022] A portion of the CNTs 17 constituting the structure 14 is directly attached to and fixed to the surface of the carbon fiber 11. Thus, the structure 14 is directly attached to the surface of the carbon fiber 11. The phrase "CNTs 17 are directly attached to the surface of the carbon fiber 11" means that the CNTs 17 are directly attached to the carbon fiber 11 without any dispersant or adhesive such as surfactant between the CNTs 17 and the surface of the carbon fiber 11. This attachment (fixation) is a bond produced by Van der Waals forces. Because a portion of the CNTs 17 constituting the structure 14 is directly attached to the surface of the carbon fiber 11, and there is no dispersant or adhesive between them, the structure 14 is in direct contact with the surface of the carbon fiber 11.

[0023] Furthermore, among the CNTs 17 constituting structure 14, some do not directly contact the surface of the carbon fiber 11, but are instead wrapped around other CNTs 17 and thus fixed to the carbon fiber 11. Also, some are directly attached to the surface of the carbon fiber 11 and simultaneously wrapped around other CNTs 17 and thus fixed to the carbon fiber 11. Hereinafter, the aforementioned CNTs 17 fixed to the carbon fiber 11 will be collectively referred to as "attached to the carbon fiber 11" for explanation. Additionally, the state of CNTs 17 being wrapped or intertwined includes a state where some CNTs 17 are pressed down by other CNTs 17.

[0024] In the CNTs 17 constituting the structure 14, in addition to those directly attached to the surface of the carbon fiber 11 as described above, there are also some that are not in direct contact with the surface of the carbon fiber 11 but are wrapped around other CNTs 17 and thus fixed to the carbon fiber 11. Therefore, compared to conventional composite material structures which are composed only of CNTs directly attached to the surface of the carbon fiber, this example of structure 14 is composed of more CNTs 17. That is, the number of CNTs 17 attached to the carbon fiber 11 is greater than before.

[0025] As described above, since the plurality of CNTs 17 are interconnected without a medium on each other's surfaces to form the structure 14, the composite material 10 exhibits the electrical and thermal conductivity properties derived from CNTs. Furthermore, since the CNTs 17 are attached to the surface of the carbon fiber 11 without a medium, the CNTs 17 constituting the structure 14 are not easily peeled off from the surface of the carbon fiber 11, and the rotating component 2 containing the composite material 10 can improve mechanical strength, including tensile strength.

[0026] In the rotating component 2 as described above, a carbon fiber bundle 12 composed of a plurality of carbon fibers 11 (i.e., a plurality of CNT composite fibers) having a structure 14 is impregnated with a matrix resin M and then hardened. Since the matrix resin M impregnates the structure 14, the structure 14 of each carbon fiber 11 is fixed to the matrix resin M along with the surface of the carbon fiber 11. In this way, each carbon fiber 11 is firmly adhered to the matrix resin M, meaning the interfacial adhesion strength between the carbon fiber 11 and the matrix resin M is high, and the tensile strength of the rotating component 2 is increased.

[0027] As will be described later, the CNTs 17 are made to have a curved shape. Compared to the case of using CNTs with high straightness, there are more CNTs 17 attached to the carbon fibers 11, and while increasing the thickness of the structure 14, a structure 14 in which the CNTs 17 are woven like non-woven fabric fibers is formed. Around each carbon fiber 11 of the rotating component 2, a region (hereinafter referred to as the composite region) 18 is formed in which the matrix resin M is impregnated into the structure 14 and hardened (see Figure 14). By forming such a composite region 18, the interfacial adhesion strength between the carbon fibers 11 and the matrix resin M becomes higher, and the tensile strength of the rotating component 2 becomes higher. Furthermore, by inserting CNTs 17 into the resin portion between adjacent carbon fibers 11, the interaction between the carbon fibers becomes stronger, and the strength reduction caused by defects in the carbon fibers 11 is compensated by other adjacent carbon fibers 11, thus suppressing the strength reduction caused by defects.

[0028] In addition, by attaching CNT 17 to the surface of carbon fiber 11 and forming a thick nonwoven structure 14, the properties of the carbon fiber reinforced molded body (rotating component 2) are improved, including not only tensile strength, but also elastic modulus, vibration damping characteristics (damping properties), and durability against repeated bending.

[0029] The structures 14 formed on the plurality of carbon fibers 11 are independent of each other, wherein the structure 14 of one carbon fiber 11 and the structure 14 of another carbon fiber 11 do not share the same CNT 17. That is, the CNT 17 contained in the structure 14 set on one carbon fiber 11 is not included in the structure 14 set on another carbon fiber 11.

[0030] As shown in Figure 4, a first sizing agent 15 is added to CNT 17 in a manner that coats the contact portions of CNT 17 that are in direct contact with each other. The first sizing agent 15 has a structure containing functional groups present on the surface of CNT 17, such as hydrophilic groups derived from hydroxyl (-OH) and carboxyl groups (-COOH), and carbodiimide groups (-N=C=N-) of carbodiimide compounds. That is, the first sizing agent 15 crosslinks the CNT 17 that are in direct contact with each other through its carbodiimide-derived structure. In this way, the CNT 17 are fixed together by the first sizing agent 15 through their contact portions.

[0031] There is no particular limitation on the method of adding functional groups to the surface of CNT 17. The addition of functional groups can be achieved through various treatments performed after the manufacture of CNT 17, or it can be achieved through treatments that add functional groups. For example, anodic electrolytic oxidation or ozone oxidation can be used as wet methods. If functional groups are present on the surface of CNT 17 when the first sizing agent 15 is added to the structure 14 during the first sizing treatment, there is no particular limitation on the time point at which the functional groups are added to the surface of CNT 17.

[0032] A carbodiimide compound is a compound containing two or more of the structures shown in formula (1), where n is an integer greater than or equal to 1. R is, for example, a hydrocarbon. Examples of hydrocarbons include aliphatic hydrocarbons, alicyclic hydrocarbons, and aromatic hydrocarbons.

[0033] [Chemistry 1] ···(1)

[0034] As a carbodiimide compound, for example, as described in Japanese Patent Application Publication No. 2007-238753, compounds used as curing agents for water-based resins can be used, as well as commercially available compounds. Commercially available carbodiimide compounds include, for example, "CARBODILITE V-02" (trade name, manufactured by Nisshinbo Chemical Co., Ltd.). Carbodiimide compounds containing hydrophilic segments dissolve in water and function as water-based crosslinking agents that crosslink CNT 17 molecules.

[0035] As shown in FIG5, the CNTs 17 at the contact portion maintain direct contact with each other, and around the contact portion near the surface of each CNT 17, the CNTs 17 are cross-linked with each other by the first sizing agent 15.

[0036] As described above, the first sizing agent 15 acts as a crosslinking agent to crosslink the CNTs 17 that are in contact with each other that constitute the structure 14, thereby further strengthening the adhesion between the CNTs 17 and making the structure 14 less prone to collapse.

[0037] Furthermore, as shown in Figure 4, the first sizing agent 15 adheres to the carbon fiber 11 and the CNT 17 in a state of coating the contact portion of the carbon fiber 11 and the CNT 17 in direct contact with the carbon fiber 11. Similar to the relationship between the CNTs 17, the first sizing agent 15 crosslinks the carbon fiber and CNT 17 through the structure of the carbodiimide groups generated by the reaction of functional groups on the surfaces of the carbon fiber 11 and CNT 17. In this way, the first sizing agent 15 further strengthens the adhesion of the CNT 17 to the carbon fiber 11 by crosslinking the carbon fiber and CNT 17, making the structure 14 less likely to peel off from the carbon fiber 11.

[0038] By means of the first sizing agent 15, when manufacturing composite material 10 or rotating component 2, the detachment of structure 14 from carbon fiber 11 and CNT 17 from structure 14 is suppressed, thereby preventing the reduction of interfacial adhesion strength or even the reduction of tensile strength of rotating component 2, and obtaining uniform and better properties.

[0039] Furthermore, if the CNTs 17 are in direct contact with each other and are cross-linked around the contact area by the first sizing agent 15, then as described above, the CNTs 17 may be coated with the first sizing agent 15, or as shown in Figures 6 and 7, the CNTs 17 may not be coated. Similarly, if the carbon fiber 11 and the CNTs 17 are in direct contact and are cross-linked around the contact area by the first sizing agent 15, then as shown in Figure 6, the first sizing agent 15 may not coat the CNTs 17.

[0040] Furthermore, in the structure 14, a plurality of CNTs 17 form a void (mesh) 19 surrounded by the plurality of CNTs 17. To prevent the matrix resin M from impregnating the structure 14, it is preferable that the first sizing agent 15 does not clog the void 19. To prevent clogging of the void 19, the mass ratio of the first sizing agent 15 attached to the structure 14 to the mass of the carbon fiber 11, i.e., the sizing agent mass ratio Rm, is preferably in the range of 0.6% to 1.1%.

[0041] Depending on the diameter of the carbon fiber 11, the mass per unit length of the carbon fiber 11 increases or decreases, and the mass of the suitable first sizing agent 15 attached to the structure 14 also increases or decreases. However, within the diameter range of the carbon fiber 11 generally used for carbon fiber reinforced molded articles, the change in the suitable sizing agent mass ratio Rm relative to the diameter of the carbon fiber 11 is very small. Regardless of the diameter of the carbon fiber 11, as long as it is within the aforementioned sizing agent mass ratio Rm range, blockage of the voids 19 can be prevented. It has been confirmed that for carbon fibers 11 with a diameter between 4 μm and 8 μm, when the sizing agent mass ratio Rm is between 0.6% and 1.1%, the voids 19 are not blocked. The matrix resin M impregnates into the structure 14 and hardens, making the matrix resin M less likely to peel off from the structure 14 or even from the carbon fiber 11, which is beneficial to the improvement of mechanical strength.

[0042] The CNTs 17 attached to the carbon fiber 11 have a curved shape. This curved shape of the CNTs 17 is due to the presence of five-membered and seven-membered rings of carbon, resulting in curved portions (bent sections) in the graphite structure of the CNTs 17, and is evaluated using SEM observation to assess whether the CNTs 17 exhibit a curved or zigzag shape. For example, the curved shape of the CNTs 17 means that there is a bend at at least one location along the average length of the CNTs 17's usable range, as described later. Such curved CNTs 17 are attached to the surface of the curved carbon fiber 11 in various ways, even in cases of long lengths. Furthermore, spaces (gaps) can easily form between the curved CNTs 17 and the surface of the carbon fiber 11 to which the CNTs 17 are attached, and between the attached CNTs 17 themselves, allowing other CNTs 17 to be accommodated. Therefore, compared to using CNTs with a highly linear shape, by using CNTs with a curved shape, the number of CNTs 17 attached to the carbon fiber 11 (the number of CNTs 17 forming the structure 14) increases.

[0043] The length of CNT 17 is preferably between 0.1 μm and 10 μm. If the length of CNT 17 is 0.1 μm or more, the CNT 17 can intertwine with each other to more reliably form a structure 14 in direct contact or direct connection, and at the same time, it can more reliably form a space that can accommodate other CNT 17 as described above. Furthermore, if the length of CNT 17 is 10 μm or less, the situation where CNT 17 spans between carbon fibers 11 and is attached will not occur. That is, as described above, the situation where a CNT 17 contained in a structure 14 provided on one carbon fiber 11 is contained within a structure 14 provided on another carbon fiber 11 will not occur.

[0044] The length of CNT 17 is preferably between 0.2 μm and 5 μm. If the length of CNT 17 is 0.2 μm or more, the number of CNT 17 attached can be increased, resulting in a thicker structure 14. If it is 5 μm or less, when attaching CNT 17 to carbon fiber 11, CNT 17 is less likely to aggregate and becomes easier to disperse more uniformly. As a result, CNT 17 is more uniformly attached to carbon fiber 11.

[0045] Furthermore, as CNTs attached to carbon fiber 11, it is not excluded that highly linear CNTs or CNTs with lengths outside the range described above may be mixed in. Even if mixed, for example, the number of CNTs attached to carbon fiber 11 can be increased because highly linear CNTs fill the space formed by CNT 17.

[0046] The average diameter of CNT 17 is preferably between 0.5 nm and 30 nm, and more preferably between 3 nm and 10 nm. If the diameter of CNT 17 is less than 30 nm, it is flexible, easily adheres to the surface of carbon fiber 11, and easily intertwines with other CNT 17 and is fixed to the carbon fiber 11, thus forming the structure 14 more securely. Furthermore, if it is less than 10 nm, the bonding between the CNT 17 constituting the structure 14 becomes more stable. The diameter of CNT 17 is measured using a transmission electron microscope (TEM) image. Single-layer or multi-layer CNT 17 is acceptable, but multi-layer is preferred.

[0047] The number of CNTs 17 attached to the carbon fiber 11 can be evaluated by the thickness of the structure 14 (the radial length of the carbon fiber 11). For example, a portion of the structure 14 on the surface of the carbon fiber 11 can be adhered to transparent tape or the like and peeled off, and the thickness of each portion of the structure 14 can be obtained by calculating and measuring the cross-section of the structure 14 remaining on the surface of the carbon fiber 11 using SEM or the like. The thickness of the structure 14 is measured at 10 locations within a measurement range that almost uniformly covers a predetermined length along the fiber axis of the carbon fiber 11, and the average value is taken as the thickness of the structure 14. The length of the measurement range can, for example, be set to five times the upper limit of the aforementioned length range of CNTs 17.

[0048] The thickness (average) of the structure 14 obtained as described above is preferably between 10 nm and 300 nm, and more preferably between 15 nm and 200 nm, and even more preferably between 50 nm and 200 nm. If the thickness of the structure 14 is less than 200 nm, the resin impregnation between the carbon fibers 11 is better.

[0049] Furthermore, the CNT mass ratio Rc is the mass ratio of CNT 17 attached to carbon fiber 11 to CNT composite fiber. The state of CNT 17 attached to carbon fiber 11 can be evaluated using the CNT mass ratio Rc. When the mass of carbon fiber 11 of a predetermined length (hereinafter referred to as CF mass) is set as Wa and the mass of CNT 17 attached to this carbon fiber 11 (hereinafter referred to as CNT mass) is set as Wb, the CNT mass ratio Rc is calculated from "Rc=Wb / (Wa+Wb)".

[0050] It is preferable that CNT 17 is uniformly attached to carbon fiber 11, and preferably that it is attached to cover the surface of carbon fiber 11. The adhesion state, including the uniformity of CNT 17 relative to carbon fiber 11, can be evaluated by observing the obtained images using SEM. In this case, it is preferable to observe and evaluate multiple locations (e.g., 10 locations) covering a predetermined length (e.g., 1 cm, 10 cm, 1 m) of carbon fiber 11 almost uniformly along the fiber axis.

[0051] Furthermore, the uniformity of CNT 17 attachment to carbon fiber 11 can be evaluated using the aforementioned CNT mass ratio Rc. A CNT mass ratio Rc of 0.001 to 0.008 is preferred. When the CNT mass ratio Rc is 0.001 or higher, the enhanced properties resulting from the structure 14 as described above can be reliably obtained in the case of rotating component 2. When the CNT mass ratio Rc is 0.008 or lower, the resin of the matrix resin M can reliably impregnate into the structure 14. Furthermore, a CNT mass ratio Rc of 0.002 to 0.006 is preferred. When the CNT mass ratio Rc is 0.002 or higher, the structure 14 (CNT 17) functions more reliably among almost all carbon fibers 11. When the CNT mass ratio Rc is below 0.006, the matrix resin M can reliably impregnate the structure 14, and the structure 14 can function more reliably even when the proportion of matrix resin M in the rotating component 2 is low. Furthermore, even when the proportion of resin M is low, since CNT 17 exists in a high concentration in the resin between the carbon fibers, its reinforcing effect can improve toughness and strength.

[0052] Within a 1-m length of a carbon fiber 11 (hereinafter referred to as the evaluation range), the standard deviation s of the CNT mass ratio Rc at 10 measurement sites is preferably 0.0005 or less, and more preferably 0.0002 or less. Furthermore, the ratio of the standard deviation s to the average CNT mass ratio Rc is preferably 40% or less, and more preferably 15% or less. It is preferable to set the 10 measurement sites to almost uniformly cover the evaluation range. The standard deviation s is an indicator of the number (attachment amount) of CNTs 17 attached to the carbon fiber 11 and the variation in the thickness of the structure 14; the smaller the variation, the smaller the value. Therefore, the smaller this standard deviation s, the better. The number of attached CNTs 17 and the variation in the thickness of the structure 14 indicate the differences in the characteristics of the CNTs 17 from the composite material 10 and the rotating component 2 using the composite material 10. If the standard deviation s is less than 0.0005, the characteristics of the composite material 10 and the carbon fiber reinforced CNT 17 can be more accurately represented, and if it is less than 0.0002, the characteristics of the CNT 17 can be fully and accurately represented. Furthermore, the standard deviation s can be obtained using equation (2). In equation (2), n is the number of measurement sites (n=10 in this example), Ri is the CNT mass ratio Rc of the measurement sites, and Ra is the average of the CNT mass ratio Rc.

[0053] [Number 1] ···(2)

[0054] [Measurement of CNT mass ratio] For the measurement site where the CNT mass ratio Rc is to be determined, a carbon fiber bundle 12 (e.g., approximately 12,000 CNT composite fibers) is cut into approximately 1m sections to serve as a measurement sample, and the CNT mass ratio Rc is determined as described below. (1) Place the measurement sample into a liquid (hereinafter referred to as the measurement solution) that serves as the dispersion medium for CNT 17. For example, NMP (N-methyl-2-pyrrolidone, CAS Registry No.: 872-50-4) with a dispersant added can be used as the measurement solution. The dispersant may be added to the measurement solution to prevent CNT 17 from adhering to the carbon fiber 11, but it may not be added. The amount of measurement solution is, for example, 100 ml relative to 10 g of measurement sample. (2) Measure the difference between the mass of the measuring liquid before the measurement sample is placed and the mass of the measuring liquid containing the measurement sample after it is placed, and use it as the sum of the mass of the measurement sample (i.e., the CF mass Wa of carbon fiber 11) and the CNT mass Wb of the CNT 17 attached to this carbon fiber 11 (Wa+Wb). (3) Heating the measuring liquid containing the measurement sample completely separates the CNT 17 attached to the carbon fiber 11 from the carbon fiber 11, so that the CNT 17 is dispersed in the measuring liquid. (4) Measure the absorbance (transmittance) of the measurement solution containing CNT 17 using a spectrophotometer. Calculate the concentration of CNT 17 in the measurement solution (hereinafter referred to as CNT concentration) from the spectrophotometer measurement results and a pre-made calibration line. The CNT concentration is the mass percentage concentration obtained by setting the CNT concentration value as C, the mass of the measurement solution as W1, and the mass of CNT 17 contained in the measurement solution as W2, using the formula "C = W2 / (W1 + W2)". (5) The mass Wb of CNT 17 in the measuring solution is obtained from the CNT concentration and the mass of the measuring solution before the sample is placed in the measuring solution. (6) Calculate the CNT mass ratio Rc (=Wb / (Wa+Wb)) from the sum of the CF mass Wa and CNT mass Wb obtained in (2) and the CNT mass (Wb) of CNT 17.

[0055] In the aforementioned absorbance measurement, a spectrophotometer (e.g., the SolidSpec-3700 manufactured by Shimadzu Corporation) can be used, and a measurement wavelength such as 500 nm can be used. Furthermore, it is preferable to store the measurement solution in a quartz cell during measurement. Moreover, by using the absorbance of a dispersion medium free of impurities other than the dispersant as a reference, the concentration C of CNT 17 can be determined by the difference between the absorbance of the measurement solution containing CNT 17 and the reference. Additionally, in the measurement of the CNT mass ratio Rc, carbon fiber bundles 12 with the first sizing agent 15 removed are used. However, when the mass of the first sizing agent 15 is approximately 1 / 100 of that of the carbon fiber 11, the presence or absence of the first sizing agent 15 (i.e., the mass of the first sizing agent 15) will not substantially affect the optimal range of the CNT mass ratio Rc. Therefore, in this case, the mass of the CNT composite fiber with the first sizing agent 15 attached can be regarded as the sum of the CF mass Wa and the CNT mass Wb (Wa+Wb).

[0056] To assess uniformity based on the CNT mass ratio Rc, 10 measurement sites were set up to almost uniformly cover the assessment range (e.g., 1 m in length) of the carbon fiber bundle 12 to be evaluated. These 10 measurement sites were set at the two ends and 8 locations between the two ends of the assessment range, and the CNT mass ratio Rc was calculated for each measurement site according to the steps described above.

[0057] [Measurement of the mass ratio of sizing agent] In measuring the sizing agent mass ratio Rm, for example, three CNT composite fibers are cut from the carbon fiber bundle 12 produced by attaching the first sizing agent 15 to the structure 14 of each carbon fiber 11 as a measurement sample, and the sizing agent mass ratio Rm is determined as described below. The length of the CNT composite fibers cut as the measurement sample is, for example, 5 m. However, the number and length of the CNT composite fibers used as the measurement sample are not limited to this.

[0058] (1) Place the sample into the measuring solution of CNT 17. The conditions of the measuring solution and dispersion medium are the same as those for measuring the CNT mass ratio Rc described above. (2) Measure the difference between the mass of the measuring liquid before the measuring sample is placed and the mass of the measuring liquid containing the measuring sample after it is placed, and use it as the sum of the mass of the measuring sample (i.e., the CF mass Wa of carbon fiber 11), the CNT mass Wb of CNT 17 attached to carbon fiber 11 and the sizing agent mass Wc of the first sizing agent 15 attached to CNT 17 (Wa+Wb+Wc). (3) Heating the measuring liquid containing the measurement sample completely separates the CNT 17 attached to the carbon fiber 11 from the carbon fiber 11, so that the CNT 17 is dispersed in the measuring liquid. (4) Similar to the case of measuring the CNT mass ratio Rc, the absorbance of the measurement solution containing CNT 17 is measured using a spectrophotometer. The CNT concentration of CNT 17 in the measurement solution is determined based on this absorbance and the pre-made calibration line. The CNT mass Wb in the measurement solution is determined from the obtained CNT concentration and the mass of the measurement solution before the measurement sample is placed in it. (5) Determine the CF mass Wa based on the specifications of the carbon fiber 11 (raw yarn) used. (6) Calculate the difference between the mass of CF Wa obtained in (5) and the mass of CNT Wb obtained in (4) by subtracting the mass of the measured sample (Wa+Wb+Wc) from the mass of the measured sample, and use it as the mass of the first sizing agent Wc added to the measured sample. (7) Calculate the sizing agent mass ratio Rm (=(Wc / Wa)×100%) based on the CF mass Wa obtained in (5) and the sizing agent mass Wc obtained in (6).

[0059] Furthermore, in the aforementioned measurement of the sizing agent mass ratio Rm, the mass of the CF (cream fiber) Wa, determined based on the specification value of the carbon fiber 11 (raw yarn), is also determined when the sizing agent is not attached. The sizing agent described here is attached to the surface of the carbon fiber 11 (raw yarn) to prevent it from tangling, etc., and is different from the first sizing agent. However, the mass of the sizing agent used to prevent tangling is generally about 1 / 100 of the CF mass Wa of the carbon fiber 11. In this case, whether or not the sizing agent is attached does not substantially affect the optimal range of the sizing agent mass ratio Rm. Therefore, in this case, the mass of the carbon fiber 11 with the sizing agent attached can also be considered as the CF mass Wa at the time the sizing agent mass ratio Rm is obtained.

[0060] Furthermore, the CF mass Wa is not limited to being determined based on specification values. For example, the mass of carbon fiber 11 after separating CNT 17 can also be measured and used as the CF mass Wa. Alternatively, the CF mass Wa can be determined from the mass of carbon fiber 11 of the same type as that used in the CNT composite fiber as the measurement sample, without attached CNT 17.

[0061] As shown in Figure 8, the rotating component 2 is manufactured through a structure forming step ST1, a first sizing treatment step ST2, and a forming step ST3. In the structure forming step ST1, CNT 17 is attached to each carbon fiber 11 (raw yarn) of the carbon fiber bundle 12 to form the structure 14. For this purpose, the carbon fiber bundle 12 is immersed in a CNT monodisperse liquid (hereinafter referred to as the dispersion liquid) containing monodispersed CNT 17, and mechanical energy is applied to the dispersion liquid. Monodispersion means that CNT 17 is physically separated and dispersed in the dispersion medium without entanglement, which means that the proportion of aggregates of two or more CNT 17 is less than 10%. Here, if the proportion of aggregates is more than 10%, it will promote the aggregation of CNT 17 in the dispersion medium and hinder the attachment of CNT 17 to the carbon fiber 11.

[0062] As shown in Figure 9, in one example, the attachment device 21 consists of a CNT attachment tank 22, guide rollers 23-26, an ultrasonic generator 27, and a traveling mechanism (not shown in the figure) that causes the carbon fiber bundle 12 to travel at a fixed speed. The dispersion liquid 28 is contained within the CNT attachment tank 22. The ultrasonic generator 27 applies ultrasonic waves to the dispersion liquid 28 within the CNT attachment tank 22 from below.

[0063] Long, strip-shaped (e.g., approximately 100 m) carbon fiber bundles 12, before forming the structure 14, are continuously supplied to the attachment device 21. The supplied carbon fiber bundles 12 are sequentially wound onto guide rollers 23-26 and travel at a fixed speed by a traveling mechanism. Carbon fiber bundles 12 without any sizing agent adhering to each carbon fiber 11 are supplied to the attachment device 21. The sizing agent described here is to prevent the carbon fibers 11 from tangling, etc.

[0064] In the unfolded state, each carbon fiber bundle 12 is wound onto guide rollers 23-26. Due to the appropriate tension, the carbon fiber bundles 12 wound on guide rollers 23-26 can reduce concerns about entanglement with carbon fiber 11. The winding angle of the carbon fiber bundles 12 on guide rollers 24-26 is preferably small (below 90°).

[0065] Guide rollers 23-26 are all flat rollers. As shown in Figure 10, the roller length (axial length) L1 of guide roller 23 is sufficiently greater than the width WL of the unfolded carbon fiber bundle 12. For guide rollers 24-26, similar to guide roller 23, their roller length is sufficiently greater than the width WL of the unfolded carbon fiber bundle 12. For example, guide rollers 23-26 all have the same dimensions, with a roller length L1 of 100 mm and a roller diameter (outer diameter) of 50 mm. In the unfolded carbon fiber bundle 12, a plurality of carbon fibers 11 are arranged in the thickness direction (the diameter direction of the guide roller).

[0066] Guide rollers 24 and 25, among guide rollers 23-26, are positioned within the CNT attachment groove 22. In this way, the carbon fiber bundle 12 travels linearly in the dispersion liquid 28 at a fixed depth between guide rollers 24 and 25. The travel speed of the carbon fiber bundle 12 is preferably between 0.5 m / min and 100 m / min. A higher travel speed of the carbon fiber bundle 12 results in higher productivity, while a lower travel speed is more effective for uniform attachment of the CNT 17 and for suppressing entanglement of the carbon fibers 11. Furthermore, less entanglement of the carbon fibers 11 results in higher uniformity of CNT 17 attachment to the carbon fibers 11. If the travel speed of the carbon fiber bundle 12 is below 100 m / min, entanglement of the carbon fibers 11 can be suppressed more effectively, and the uniformity of CNT 17 attachment can be further improved. Furthermore, the travel speed of the carbon fiber bundle 12 is preferably between 5 m / min and 50 m / min.

[0067] The ultrasonic generator 27 applies ultrasonic vibrations as mechanical energy to the dispersion 28. This creates a reversible reaction state in the dispersion 28, alternating between a dispersed state and an aggregated state of CNT 17. When the carbon fiber bundle 12 passes through the dispersion 28 in this reversible reaction state, the CNT 17 adheres to each carbon fiber 11 by van der Waals forces as it transitions from a dispersed state to an aggregated state. The mass of the carbon fiber 11 is more than 100,000 times greater than that of the CNT 17, and the energy required to detach the attached CNT 17 becomes greater than the energy generated by the ultrasonic vibration. Therefore, once the CNT 17 is attached to the carbon fiber 11 in the dispersion 28, it will not peel off even by subsequent ultrasonic vibrations. Furthermore, since the masses of the CNT 17 are very small, they alternate between dispersed and aggregated states due to ultrasonic vibrations.

[0068] By repeatedly transforming from a dispersed state to an aggregated state, a large number of CNTs 17 are attached to each carbon fiber 11 to form a structure 14. As described above, by using CNTs with curved shapes as CNTs 17, other CNTs 17 enter the spaces formed between the CNTs 17 and the surfaces of the carbon fibers 11 to which the CNTs 17 are attached, or between the attached CNTs 17 and each other, thus more CNTs 17 are attached to the carbon fibers 11 to form a structure 14.

[0069] The frequency of ultrasonic vibration applied to the dispersion 28 is preferably between 40 kHz and 950 kHz. If the frequency is above 40 kHz, the entanglement of the carbon fibers 11 in the carbon fiber bundle 12 can be suppressed. Furthermore, if the frequency is below 950 kHz, the CNTs 17 adhere well to the carbon fibers 11. To further reduce the entanglement of the carbon fibers 11, the frequency of ultrasonic vibration is preferably above 100 kHz.

[0070] Furthermore, the inventors discovered that the number of CNTs 17 attached to the carbon fiber 11, by causing the CNTs 17 to change from a dispersed state to an aggregated state more than 100,000 times, effectively suppressed the entanglement between the carbon fibers 11 and ensured the uniformity of the thickness of the structure 14. In addition, the maximum number of attached CNTs varies depending on the CNT concentration of the dispersion liquid 28, increasing with the increase of the CNT concentration of the dispersion liquid 28. However, if the CNT concentration of the dispersion liquid 28 becomes too high, such that the CNTs 17 cannot remain dispersed when ultrasonic vibration is applied, then the CNTs 17 will become unable to attach to the carbon fiber 11.

[0071] Therefore, the traveling speed of the carbon fiber bundle 12, the distance the carbon fiber bundle 12 travels in the dispersion liquid 28 (the interval between the guide rollers 24 and 25), and the frequency of the ultrasonic vibration applied to the dispersion liquid 28 are determined so that the length of the period during which the carbon fiber bundle 12 travels in the dispersion liquid 28, that is, the time it travels between the guide rollers 24 and 25 (hereinafter referred to as the immersion time), is preferably 100,000 times or more the period of the ultrasonic vibration applied to the dispersion liquid 28. That is, when the frequency of the ultrasonic vibration is set to fs (Hz) and the immersion time is set to Ts (seconds), it is preferable to satisfy "Ts≥100000 / fs". For example, if the frequency of the ultrasonic vibration is 100kHz and the distance the carbon fiber bundle 12 travels in the dispersion liquid 28 is 0.1m, then the traveling speed of the carbon fiber bundle 12 can be 6m / min or less. Furthermore, even when the carbon fiber bundle 12 is immersed in the dispersion liquid 28 in multiple times, if the total immersion time is 100,000 times or more the period of ultrasonic vibration, the number of CNT 17 attached can be almost maximized.

[0072] As shown in the schematic diagram in Figure 11, due to the ultrasonic vibration applied from the ultrasonic generator 27, a standing wave with a fixed sound pressure (amplitude) distribution is generated in the dispersion liquid 28 within the CNT attachment tank 22. In this attachment device 21, the positions of the guide rollers 24 and 25 in the depth direction are adjusted so that the carbon fiber bundle 12 travels in the dispersion liquid 28 at the node of the ultrasonic vibration standing wave, that is, at a depth where the sound pressure is very low. Therefore, when the depth of the carbon fiber bundle 12 from the surface of the dispersion liquid 28 during its travel is set as D, the wavelength of the ultrasonic vibration standing wave generated in the dispersion liquid 28 is set as λ, and n is set to an integer greater than or equal to 1, the aforementioned depth is determined by satisfying "D = n·(λ / 2)". In addition, the wavelength λ of the standing wave can be obtained based on the sound velocity in the dispersion liquid 28 and the frequency of the ultrasonic vibration applied from the ultrasonic generator 27.

[0073] As described above, by adjusting the depth of the carbon fiber bundle 12 traveling in the dispersion liquid 28, the vibration of the carbon fiber 11 caused by sound pressure can be suppressed, and yarn tangling caused by yarn slack can be prevented. This, in turn, suppresses friction between the carbon fibers 11 and between the CNTs 14 attached to the surfaces of each carbon fiber 11, allowing for the formation of a thick structure 14. Furthermore, by suppressing friction, even with a large thickness of the structure 14, the variation in the CNT mass ratio Rc can be suppressed, and the aforementioned standard deviation s becomes smaller. Additionally, the depth of the carbon fiber bundle 12 traveling in the dispersion liquid 28 can be slightly deviated from the node of the standing wave. In this case, it is preferable to be within the range of n·λ / 2-λ / 8 to n·λ / 2+λ / 8 (n·λ / 2-λ / 8≤D≤n·λ / 2+λ / 8). In this way, yarn tangling caused by yarn slack of the carbon fiber 11 can be kept within an acceptable range.

[0074] For example, long strips of CNT (hereinafter referred to as material CNT) are added to a dispersion medium, and a homogenizer, shear force, or ultrasonic disperser is used to cut the material CNT into CNT 17 of the desired length, while simultaneously homogenizing the dispersion of CNT 17 to prepare dispersion 28.

[0075] As a dispersion medium, alcohols such as water, ethanol, methanol, and isopropanol, toluene, acetone, tetrahydrofuran (THF), methyl ethyl ketone (MEK), hexane, n-hexane, diethyl ether, xylene, methyl acetate, ethyl acetate, and other organic solvents, as well as mixtures of the above in any proportion, can be used. Dispersion 28 does not contain dispersants or binders.

[0076] The CNTs based on the curved CNT 17 described above have a curved shape. It is preferable that the diameter of each individual CNT is uniform. It is also preferable that the CNTs are of a length that allows for monodispersity even when cut. In this way, a dispersion 28 of CNT 17 that satisfies the aforementioned length requirements can be easily obtained.

[0077] In this example of composite material 10, as described above, CNTs with curved shapes are attached as CNT 17. Therefore, other CNTs 17 can be accommodated in the spaces formed between the CNT 17 and the surface of the carbon fiber 11 to which the CNT 17 is attached, and between the attached CNTs 17 themselves. In this way, more CNTs 17 are attached to the carbon fiber 11. Furthermore, since the CNTs 17 are firmly attached to the carbon fiber 11 to form the structure 14, the CNTs 17 are less likely to peel off from the carbon fiber 11. Moreover, the rotating component 2 made using this composite material 10 has higher CNT-derived properties.

[0078] The concentration of CNT 17 in dispersion 28 is preferably in the range of 0.003 wt% to 3 wt%. The concentration of CNT 17 in dispersion 28 is more preferably 0.005 wt% to 0.5 wt%.

[0079] The carbon fiber bundle 12 is extracted from the dispersion 28 and then dried. The first sizing agent 15 is applied to the structure 14 by subjecting the dried carbon fiber bundle 12 to a first sizing treatment.

[0080] The first sizing treatment step ST2 involves applying a first sizing solution to the dried carbon fiber bundle 12 (bringing it into contact with the solution) and then drying it. The first sizing solution can be prepared by dissolving the aforementioned carbodiimide compound in a solvent. Water, alcohols, ketones, and mixtures thereof can be used as the solvent for dissolving the carbodiimide compound.

[0081] The first sizing solution can be applied to the carbon fiber bundle 12 by any of the following methods: immersing the carbon fiber bundle 12 in a tank containing the first sizing solution, spraying the carbon fiber bundle 12 with the first sizing solution, or coating the carbon fiber bundle 12 with the first sizing solution. The first sizing solution is applied to the surface of the CNT 17 while maintaining direct contact between the CNTs 17. The lower the viscosity, the easier it is to accumulate near the contact area between the CNTs 17 and near the contact area between the carbon fiber 11 and the CNT 17.

[0082] In the first sizing treatment step ST2, the desired sizing agent mass ratio Rm can be obtained by adjusting the amount of the first sizing treatment liquid applied to the carbon fiber bundle 12 and the concentration of the carbodiimide compound in the first sizing treatment liquid.

[0083] In the drying step after coating with the first sizing solution, the solvent of the first sizing solution (water in this example) is evaporated. As a drying method, known drying methods can be used, such as placing the carbon fiber bundle 12 coated with the first sizing solution for drying, blowing air or other gases onto the carbon fiber bundle 12, or heating the carbon fiber bundle 12. Alternatively, either the method of placing for drying or blowing gas can be combined with heating.

[0084] In forming step ST3, the rotating component 2 is formed using the composite material 10 that has undergone the first sizing treatment step ST2 by filament winding. As shown in Figure 12, for example, multiple strands of composite material 10 are supplied from a creel (filament feeding device) 31, simultaneously adjusted to a predetermined tension, and the supplied composite material 10 is fed to the filament winder 33 via a resin application device 32. As the composite material 10 passes through the resin application device 32, an uncured liquid matrix resin M is applied to the surface of the CNT composite fibers. At this time, the resin application device 32 uses the matrix resin M to impregnate the structure 14 formed on the surface of the carbon fiber 11.

[0085] A mandrel 34 is rotatably disposed within a filament winder 33. By rotating the mandrel 34 using the filament winder 33, the composite material 10, to which the matrix resin M has been applied, is wound onto the mandrel 34 while a predetermined tension is applied. The position of the composite material 10 wound onto the mandrel 34 is determined by a coating head (not shown in the figure) disposed in the resin application device 32. By reciprocating the resin application device 32 in the axial direction of the mandrel 34 in sync with the rotation of the mandrel 34, the winding position of the composite material 10 relative to the mandrel 34 is displaced along the axial direction of the mandrel 34 during winding.

[0086] At this point, by adjusting the winding angle (tilt angle) of the composite material 10 relative to the axial direction of the mandrel 34 to, for example, almost 90°, the CNT composite fibers coated with the matrix resin M are wound circumferentially around the mandrel 34, that is, finely wound on the outer circumferential surface of the mandrel 34 in a direction almost perpendicular to its axis. Furthermore, by adjusting the winding angle of the composite material 10 relative to the axial direction of the mandrel 34 to an angle smaller than 90°, for example, 45°, the CNT composite fibers coated with the matrix resin M are wound helically around the mandrel 34, that is, finely wound on the outer circumferential surface of the mandrel 34 at an angle of 45° relative to its axis.

[0087] As described above, after forming a molded body of composite material 10 wound in a predetermined number of layers (more than one layer) on the outer peripheral surface of mandrel 34, mandrel 34 and molded body are removed from filament winder 33 together. The molded body and, for example, the removed mandrel 34 are heated together, and the matrix resin M applied to the CNT composite fiber is hardened. The molded body after the matrix resin M has hardened is removed from mandrel 34 and cut to the desired width (length in the axial direction) as required to form rotating part 2.

[0088] As the resin application device 32, for example, as shown in FIG13, a touch roll type device is used. In this resin application device 32, the lower part of the touch roll 35 is immersed in uncured liquid matrix resin M stored in the storage tank 36, and the composite material 10 is pressed against the upper outer peripheral surface of the touch roll 35 by a pair of guide rollers 35a. As the touch roll 35 rotates, the stored liquid matrix resin M is coated onto the CNT composite fibers through the outer peripheral surface of the touch roll 35. By adjusting the rotation speed of the touch roll 35, the pressing pressure generated by the pair of guide rollers 35a on the touch roll 35 of the composite material 10, etc., the amount of matrix resin M applied to the composite material 10, i.e., the carbon fiber 11 forming the structure 14, is adjusted, and the applied matrix resin M sufficiently impregnates the structure 14.

[0089] The matrix resin M is not particularly limited and can be any type of thermosetting or thermoplastic resin. Examples of thermosetting resins include epoxy resins, phenolic resins, melamine resins, urea-formaldehyde resins (urea resins), unsaturated polyesters, alkyd resins, thermosetting polyimides, cyanate ester resins, bismaleimide resins, and vinyl ester resins. Furthermore, examples of thermoplastic resins include general-purpose resins such as polyethylene, polypropylene, polyvinyl chloride, polystyrene, acrylonitrile / styrene (AS) resins, acrylonitrile / butadiene / styrene (ABS) resins, methacrylic resins (PMMA, etc.), and thermoplastic epoxy resins; engineering plastics such as polyamides, polyacetals, polyterephthalic acid, ultra-high molecular weight polyethylene, polycarbonate, and phenoxy resins; and super engineering plastics such as polyphenylene sulfide, polyetheretherketone, polyetherketoneketone, liquid crystal polymers, polytetrafluoroethylene, polyetherimide, polyarylate, and polyimide.

[0090] It has been confirmed that the tensile strength of the rotating component 2 is positively correlated with the fiber volume fraction (Vf) of the carbon fiber 11. From the viewpoint of improving tensile strength, a higher fiber volume fraction (Vf) of the carbon fiber 11 is preferable, and the fiber volume fraction of the carbon fiber 11 can be, for example, 75% or more, or even less than 75%. On the other hand, from the viewpoint of the brittleness resistance of the rotating component 2, it is preferable to set the fiber volume fraction to less than 75%. Furthermore, from the viewpoint of the ease of molding when manufacturing the rotating component 2, a fiber volume fraction of 50% or more of the carbon fiber 11 is preferable. Therefore, a fiber volume fraction of 50% or more but less than 75% is preferable for the rotating component 2. In addition, the fiber volume fraction of the carbon fiber 11 in the rotating component 2 can be changed, for example, by adjusting the amount of matrix resin M applied to the composite material 10 by the resin application device 32 and the tension of the composite material 10 when wound around the mandrel 34.

[0091] Alternatively, the fiber volume content of the carbon fiber 11 of the rotating component 2 can be obtained using, for example, Equation (3) as described below. In Equation (3), ρ is the specific gravity of the rotating component 2, ρf is the specific gravity of the carbon fiber 11, and ρm is the specific gravity of the matrix resin M. The values ​​measured by a measuring instrument (e.g., a high-precision electronic hydrometer SD-200L (manufactured by ALFA MIRAGE AG)) are used as the specific gravity ρ of the rotating component 2 and the specific gravity ρm of the matrix resin M. The specific gravity ρf of the carbon fiber 11 can be the same as that of the rotating component 2, measured by the measuring instrument, or a specification value (the nominal value of the manufacturer of the carbon fiber 11) can be used. In addition, since the specific gravity of CNT 17 and the first sizing agent 15 attached to the carbon fiber 11 is very small relative to the specific gravity of the carbon fiber 11, the specific gravity of the carbon fiber 11 can also be considered as the specific gravity ρf alone.

[0092] [Number 2] ···(3)

[0093] In the above-described molding step ST3, when the matrix resin M of the molded body made by winding the composite material 10 around the mandrel 34 is hardened, if the matrix resin M is a thermosetting resin such as epoxy resin, it is preferable to change the heating temperature in stages to improve the accuracy of the inner diameter of the rotating part 2. In the example shown in Figure 14, in the hardening step of hardening the matrix resin M of the molded body, the temperature is changed in two stages: the temperature is raised from room temperature to a first heating temperature T1 and maintained at the first heating temperature T1 for a predetermined time, and then the temperature is raised to a second heating temperature T2, which is higher than the first heating temperature T1, and maintained at the second heating temperature T2 for a predetermined time. After maintaining the second heating temperature T2 for the predetermined time, the molded body is allowed to cool naturally, and then the molded body is removed from the mandrel 34 to form the rotating part 2.

[0094] The first heating step, which heats the molded body at a first heating temperature T1, aims to reduce the error in the inner diameter of the manufactured rotating part 2. Its main purpose is to harden the molded body into a stable shape with minimal dimensional variation while suppressing thermal expansion or contraction of the mandrel 34 due to heating. Therefore, the first heating temperature T1 is set to suppress the thermal expansion of the mandrel 34. Furthermore, heating at the first heating temperature T1 continues until a stable shape of the molded body is obtained, that is, until it can be considered hardened even without achieving final strength. More specifically, the time for maintaining the first heating temperature T1 is, for example, until the storage modulus of the matrix resin M becomes almost constant. Additionally, it is preferable that the first heating step aims to suppress thermal expansion or contraction of the molded body to further reduce the error in the inner diameter of the rotating part 2. In this case, the first heating temperature T1 is the temperature at which the mandrel 34 has a small thermal expansion or contraction, and is set to suppress the thermal expansion or contraction of the molded body at a temperature below the glass transition point of the matrix resin M at the end of the first heating step (a fixed temperature below the glass transition point of the matrix resin M raised in the first heating step). This is also the case in this example.

[0095] The time for maintaining the first heating temperature T1 can also be the time until the rate of change (increase rate) of the storage modulus of the matrix resin M begins to decrease, the time until the loss modulus reaches its peak, or the time until it reaches its peak. Alternatively, a rheometer can be used to determine the changes in the storage modulus and loss modulus of the matrix resin M relative to each heating temperature over the heating time, allowing for a predetermined determination of the time to maintain the first heating temperature T1. Furthermore, the glass transition point of the matrix resin M can be known in advance.

[0096] In the second heating step, where the molded body is heated at a second heating temperature T2, heating the molded body at a second heating temperature T2, which is higher than the first heating temperature T1, promotes the hardening of the matrix resin M after the first heating step, thereby obtaining the final strength, elastic modulus, and heat resistance of the rotating part 2. When the matrix resin M is a cyanate ester resin, it is preferable, for example, that the first heating temperature T1 is between 100°C and 200°C and the second heating temperature T2 is between 200°C and 300°C.

[0097] By progressively changing the heating temperature as described above, the molded body is heated to a first heating temperature T1 in the first heating step, causing the base resin M to harden and form a stable shape. Then, in the second heating step, the hardening of the base resin M is further promoted, resulting in the molded body possessing final strength, elastic modulus, and heat resistance. The inner diameter of the rotating component 2 manufactured in this way is almost determined when the stable shape of the molded body is formed in the first heating step. Furthermore, because heating is performed at the first heating temperature T1 in the first heating step, the thermal expansion of the mandrel 34 and the thermal expansion or contraction of the base resin M are both small. In this way, a molded body with a small inner diameter error, namely the rotating component 2, can be obtained.

[0098] Furthermore, by heating the molded body at a second heating temperature T2 in the second heating step, although the mandrel 34 undergoes more thermal expansion than in the first heating step, causing the molded body with a stable shape to deform, most of this deformation is elastic deformation and will therefore return to its original shape after cooling. Therefore, the effect on the inner diameter of the molded body in the second heating step is very small.

[0099] When 50 rotating parts 2 are manufactured by progressively changing the heating temperature as described above and actually using S45C carbon steel for mechanical construction as the material for the mandrel 34, the inner diameter is 40mm ± 0.003mm. In contrast, when 50 rotating parts 2 are manufactured by heating the molded body and the mandrel 34 made of the same material together from room temperature to a second heating temperature T2 to harden them, the inner diameter is 40mm ± 0.01mm or more. Furthermore, in the former case, the temperature is increased from room temperature and maintained at the first heating temperature T1 for 90 minutes, then increased from the first heating temperature T1 and maintained at the second heating temperature T2 for 120 minutes. In the latter case, the temperature is increased from room temperature to the second heating temperature T2 and maintained at the second heating temperature T2 for 210 minutes. The first heating temperature T1 in the former case is 145°C, while the second heating temperature T2 in both cases is 200°C. Therefore, it can be seen that by gradually changing the heating temperature as described above to harden the matrix resin M of the molded body, it becomes easier to control the inner diameter of the manufactured rotating part 2, and it is possible to manufacture a rotating part 2 that suppresses errors relative to the desired inner diameter. Furthermore, this method can be used regardless of the presence or absence of the structure 14, the first sizing agent 15, etc., when manufacturing rotating parts using the filament winding method. Additionally, the heating temperature in the first heating step can be gradually changed in two or more stages, with each heating temperature heated for a predetermined time.

[0100] In the rotating component 2 using the aforementioned composite material 10, as shown in the schematic diagram of FIG15, a cross-linked structure is formed by the cross-linking portions CL of the composite regions 18 between the carbon fibers 11, which are mutually fixed to each other. As described above, the composite regions 18 are regions composed of the structure 14 and the matrix resin M impregnated with and hardened from the structure 14. The composite regions 18 have a higher hardness than the hardened matrix resin monomers, and at the same time have high elasticity, i.e., a large elastic limit. Furthermore, the composite regions 18 have higher wear resistance than the matrix resin. Through the bonding of these composite regions 18, the bonding between the carbon fibers 11 becomes stronger, and the tensile strength of the rotating component 2 using the composite material 10 is improved.

[0101] Since a cross-linked structure is formed when the carbon fibers 11 are close enough to contact each other, a greater thickness of the structure 14 is more advantageous in terms of increasing the number of cross-links. However, from the viewpoint of ensuring quality stability due to uniform thickness and preventing the carbon fibers 11 from detaching, a thickness of 300 nm or less is preferable. In particular, a thickness of 50 nm to 200 nm is preferable.

[0102] Furthermore, in the structure 14, since the plurality of CNTs 17 are intertwined with each other in a non-woven fabric-like manner with a certain thickness, the matrix resin M applied to the carbon fiber 11 is held within the structure 14 in an impregnated state. Therefore, in this carbon fiber reinforced molded body, the rotating component 2, regardless of its molding method, hardly any skewing of the matrix resin M occurs on the surface of the carbon fiber 11, and the spacing between the carbon fibers becomes uniform. As a result, the load generated between the carbon fibers due to the shear force of the matrix resin M is uniformly transmitted, effectively improving the tensile strength of the rotating component 2.

[0103] Figure 16 shows an X-ray CT image of the internal structure of a cylindrical rod made of composite material 10 as a carbon fiber reinforced molded body using pultrusion molding. It can be confirmed that in the rod made using composite material 10, the matrix resin M does not exhibit significant skewness or cavities, and the spacing between the carbon fibers is almost uniform throughout the rod. In contrast, the internal structure of a rod made in the same manner using carbon fibers (raw yarn) without attached CNTs, as shown in Figure 17, shows significant skewness of the matrix resin M and the formation of cavities.

[0104] In addition to a first sizing agent having a structure derived from carbodiimide, a second sizing agent can also be applied to the composite material as a coating agent covering the surface of the CNTs of the structure. As shown in Figure 18, the second sizing agent 37 is attached to CNT 17 to cover the surface of the CNT 17 of the structure 14. This second sizing agent 37 is composed of a cured or uncured product of a reaction-curing resin, a thermosetting resin, or a thermoplastic resin.

[0105] After the first sizing treatment by applying the first sizing agent 15, a second sizing agent 37 is applied to the composite material 10 by a second sizing treatment. The second sizing treatment can be carried out by a conventional method, in which a second sizing treatment solution containing the resin (polymer) as the second sizing agent 37 is dissolved in a dispersion medium, and the carbon fiber bundles 12 (carbon fibers 11) that have been expanded after the first sizing treatment are brought into contact with it, so that the carbon fiber bundles 12 are coated with the second sizing agent 37, and then the dispersion medium is evaporated, and the second sizing agent 37 is simultaneously hardened or semi-hardened. Alternatively, an emulsion-type second sizing treatment solution containing droplet-shaped resin as the second sizing agent can also be used.

[0106] The resin used as the second sizing agent 37 is not particularly limited, and various reactive curing resins, thermosetting resins, thermoplastic resins, etc., can be used. For example, as thermosetting resins, examples include epoxy resins, phenolic resins, melamine resins, urea-formaldehyde resins (urea resins), unsaturated polyesters, alkyd resins, thermosetting polyimides, cyanate ester resins, and resins with reactive groups. Furthermore, as thermoplastic resins, examples include general-purpose resins such as polyethylene, polypropylene, polystyrene, acrylonitrile / styrene (AS) resins, acrylonitrile / butadiene / styrene (ABS) resins, methacrylic resins (PMMA, etc.), vinyl chloride, and thermoplastic epoxy resins; engineering plastics such as polyamides, polyacetals, polyterephthalic acid, ultra-high molecular weight polyethylene, and polycarbonate; and super engineering plastics such as polyphenylene sulfide, polyetheretherketone, liquid crystal polymers, polytetrafluoroethylene, polyetherimide, polyarylate, and polyimide.

[0107] The second sizing agent 37 covers the surface of the CNT 17 containing the first sizing agent 15. In this second sizing agent 37, portions containing the first sizing agent 15 are cross-linked, while portions not containing the first sizing agent 15 are not cross-linked, thus exhibiting viscosity. This second sizing agent 37 enhances the cohesiveness of the fiber bundles formed by the plurality of carbon fibers 11 that constitute the structure 14. Similar to the first sizing agent 15, the second sizing agent 37 does not block the voids 19 of the structure 14.

[0108] While the above description pertains to rotating components used in surface magnet motors, the term "rotating component" is not limited to this. For example, it could be a component mounted on the rotor of a motor having the same structure as the aforementioned surface magnet motor, or a rotating drum used in a centrifuge, etc. Furthermore, it could be a structure where other components are integrally formed or assembled onto the inner or outer circumference of a cylindrical, annular, or similar rotating component. Therefore, a carbon fiber reinforcing layer can be formed as part of a component containing other components. [Example]

[0109] <Example 1> In Example 1, composite material 10 was prepared using the above steps, a peel test of CNT 17 was performed, and the effect of the first sizing agent 15 was confirmed. Dispersion 28 used in preparing composite material 10 was prepared using CNTs with a curved shape as described above. Figure 19 shows an SEM image of the CNTs used in preparing dispersion 28. This CNT material has multiple layers and a diameter between 3 nm and 10 nm. The CNTs were washed with a 3:1 mixture of sulfuric acid and nitric acid to remove catalyst residues, then filtered and dried. The CNTs were added to acetone, which served as the dispersion medium for dispersion 28, and the CNTs were cut using an ultrasonic homogenizer to obtain CNT 17. The length of CNT 17 in dispersion 28 was between 0.2 μm and 5 μm. Furthermore, the CNT 17 in dispersion 28 could be evaluated as having a curved shape. The concentration of CNT 17 in dispersion 28 was set to 0.12 wt% (= 1200 wt ppm). No dispersant or binder was added to dispersion 28.

[0110] The carbon fiber bundle 12 uses Trading Card (registered trademark) T700SC-12000 (manufactured by Toray Industries, Inc.). This carbon fiber bundle 12 contains 12,000 carbon fibers 11. The diameter of the carbon fiber 11 is approximately 7 μm and the length is approximately 500 m. In addition, before attaching CNT 17, the sizing agent used to prevent the carbon fiber 11 from tangling is removed from the surface of the carbon fiber bundle 12.

[0111] The carbon fiber bundle 12 is wound in an unwound state onto guide rollers 23-26 and travels in the dispersion liquid 28 within the CNT attachment tank 22. The travel speed of the carbon fiber bundle 12 is set to 1 m / min, and ultrasonic vibration at a frequency of 200 kHz is applied to the dispersion liquid 28 using an ultrasonic generator 27. The immersion time between guide rollers 24 and 25 is 6.25 seconds. This immersion time is equivalent to 1,250,000 cycles of ultrasonic vibration applied to the dispersion liquid 28. The carbon fiber bundle 12 travels in the dispersion liquid 28 at a depth D from the surface of the liquid, satisfying "D = n·(λ / 2)".

[0112] After drying the carbon fiber bundles 12 drawn from the dispersion 28, a first sizing treatment is performed, applying a first sizing agent 15 to the CNTs 17 constituting the structure 14. In the first sizing treatment, a first sizing solution is used, in which "CARBODILITE V-02" (trade name, manufactured by Nisshinbo Chemical Co., Ltd.), a carbodiimide compound, is dissolved in water. The concentration of the carbodiimide compound in the first sizing solution is adjusted so that the mass ratio Rm of the sizing agent is 1.0%. The carbon fiber bundles 12 that have undergone the first sizing treatment are dried to obtain the composite material 10.

[0113] SEM observations revealed that a plurality of CNTs 17 were uniformly dispersed and attached to a plurality of carbon fibers 11 (hereinafter referred to as sample fibers) obtained by cutting a portion of a carbon fiber bundle 12 that had undergone the first sizing treatment as described above. As a result, it was confirmed that CNTs 17 were uniformly attached and formed a structure 14 even within a narrow (local) and wide range along the fiber axis of the carbon fibers 11. Furthermore, it was confirmed that the structure 14 formed a nonwoven fabric with a three-dimensional mesh structure, i.e., gaps 19, composed of a large number of CNTs 17, and that the gaps 19 were almost not blocked by the first sizing agent 15.

[0114] In the CNT 17 peel test, the sample fibers were impregnated in a mixture of water and surfactant, and the mixture was subjected to ultrasonic vibration for 10 minutes using an ultrasonic generator. The length of the carbon fiber 11 impregnated in the mixture was 1 m, the concentration of the surfactant in the mixture was 0.2% by mass, and the frequency of the ultrasonic vibration applied to the mixture was 100 kHz.

[0115] As Comparative Example 1, carbon fibers obtained by first sizing treatment using an epoxy resin sizing solution containing a substituted carbodiimide compound dissolved in acetone were subjected to a CNT peeling test. In this Comparative Example 1, the conditions were the same as described above, except for the sizing solution. The concentration of epoxy resin in the sizing solution was adjusted to a sizing agent mass ratio Rm of 1.0%.

[0116] In the peel test of Comparative Example 1, the mixture turned black and cloudy, and upon confirmation after the peel test, it was found that more than 90% of the CNTs had detached from the carbon fibers. In contrast, in the peel test of Example 1, the mixture was slightly cloudy, and it was confirmed that almost no CNTs 17 detached from the carbon fibers 11. That is to say, it can be seen that by using the first sizing agent 15 having a structure derived from carbodiimide, the contact between CNTs 17 and between CNTs 17 and carbon fibers 11 becomes stable, effectively suppressing the collapse of the structure 14 and the peeling of the structure 14 from the carbon fibers 11.

[0117] <Example 2> In Example 2, a test piece A21, a carbon fiber reinforced molded body, was prepared using composite material 10 produced under the same conditions as in Example 1, and a Mode I interlaminar failure toughness test was performed. Test piece A21 with a length of 160 mm, a width of 23 mm, and a thickness of 3 mm was produced. In Example 2, a prepreg was prepared using composite material 10 that had undergone a first sizing treatment. Multiple prepreg layers were cut into rectangular (160 mm × 23 mm) sheets and laminated. The matrix resin was hardened under pressure and heat to produce test piece A21. Each prepreg was cut with its length direction aligned with the fiber axis direction of the carbon fibers 11, ensuring that the fiber axis direction of all carbon fibers 11 was aligned with the length direction of test piece A21.

[0118] The Mode I interlaminar failure toughness test was conducted on the aforementioned test specimen A21 using an Autograph precision universal testing machine AG5-5kNX (manufactured by Shimadzu Corporation) according to JIS K7086. The double cantilever beam interlaminar failure toughness test method (DCB method) was used. That is, an initial crack of 2-5 mm was created at the front end of the test specimen A21 using a sharp tool, and then the crack was allowed to propagate further. The test ended when the crack length reached 60 mm from the initial crack tip. The crosshead speed of the testing machine varied according to the crack propagation. Specifically, the crosshead speed was set to 0.5 mm / min until the crack propagation reached 20 mm. When the crack propagation exceeded 20 mm, the crosshead speed was set to 1 mm / min. The crack propagation length was measured from both sides of the test specimen A21 using a microscope, and the Mode I interlaminar failure toughness value (GIC) was obtained from the load-COD (Crack Opening Displacement) curve by measuring the load and crack opening displacement.

[0119] As Comparative Example 2, a composite material obtained by first sizing with an epoxy resin containing a substituted carbodiimide compound dissolved in acetone was used to prepare a carbon fiber reinforced molded body, namely, test piece B21, which uses this composite material as the reinforcing fiber, and test piece B22, which uses unattached carbon fibers (raw yarn) as the reinforcing fiber, as the carbon fiber reinforced molded body. Mode I interlaminar failure toughness tests were then performed. Test piece B22 was not sizing treated. In Comparative Example 2, all conditions were the same as those for test piece A21, except for the conditions described above.

[0120] The Mode I interlaminar fracture toughness (GIC) value, or Mode I interlaminar fracture toughness strength (GIR), when the crack propagation is 20 mm, is 0.425 kJ / m² for specimen A21 in Example 2, while it is 0.323 kJ / m² for specimen B21 and 0.231 kJ / m² for specimen B22 in Comparative Example 2. These results indicate that the Mode I interlaminar fracture toughness strength (GIR) of specimen A21 (carbon fiber reinforced molded body) in Example 2 is approximately 1.32 times higher than that of specimen B21 in Comparative Example 2, and approximately 1.84 times higher than that of specimen B22.

[0121] <Example 3> In Example 3, solid rods A31 and A32, which were carbon fiber reinforced molded bodies and were manufactured using composite material 10 as the reinforcing fiber in the same manner as in Example 1, were produced, and a three-point bending test was performed on rods A31 and A32. In Example 3, composite material 10 and matrix resin were directly formed into cylindrical molded bodies with a diameter of approximately 2.6 mm using pultrusion molding. These molded bodies were then centerless ground to produce rods A31 and A32 with a diameter of approximately 2.2 mm and a cross-section close to a perfect circle. Therefore, the axial direction of rods A31 and A32 is consistent with the fiber axis direction of carbon fiber 11. In addition, Trading Card T700SC (manufactured by Toray Industries, Inc.) was used as the carbon fiber 11 (raw yarn) for rod A31, and Trading Card T800SC (manufactured by Toray Industries, Inc.) was used as the carbon fiber 11 (raw yarn) for rod A32, respectively. Other manufacturing conditions for rods A31 and A32 were the same as in Example 1.

[0122] The aforementioned rods A31 and A32 were cut to produce multiple test pieces with a length of 120 mm. Three-point bending tests were then performed on these test pieces according to JIS K 7074:1988 "Bending Test Method for Carbon Fiber Reinforced Plastics". The distance between the support points was set to 80 mm. In the three-point bending test, the test was conducted at a relatively low indenter descent speed (hereinafter referred to as low-speed bending test) and at a high speed (hereinafter referred to as high-speed bending test), and the breaking load (N) was determined. In the low-speed bending test, the indenter descent speed was 5 mm / s, while in the high-speed bending test, the indenter descent speed was either 1 m / s or 5 m / s. The fiber volume fraction (Vf) converted to 60% was used as the breaking load (N) for each result of the low-speed bending test and the two high-speed bending tests. Furthermore, multiple test pieces were used to perform multiple tests on rods A31 and A32 in both the low-speed bending test and the two high-speed bending tests.

[0123] As Comparative Example 3, solid rods B31 and B32, which are carbon fiber reinforced molded bodies using unattached carbon fiber (raw yarn) as reinforcing fibers, were manufactured, and three-point bending tests were performed on rods B31 and B32. The carbon fiber of rod B31 is the same as that of rod A31, and the carbon fiber of rod B32 is the same as that of rod A32. Rods B31 and B32 were manufactured under the same conditions as rods A31 and A32, except that carbon fiber was used as the raw yarn. Furthermore, under the same conditions as rods A31 and A32, three-point bending tests were performed on rods B31 and B32, and the failure load (N) was determined.

[0124] The measurement results are shown in Figure 20. In the low-speed bending test, the breaking strength of rod A31 was 11% higher than that of rod B31, and the breaking strength of rod A32 was 23% higher than that of rod B32. Furthermore, in the high-speed bending test, the breaking strength of rod A31 was 14% higher than that of rod B31, and the breaking strength of rod A32 was 11% higher than that of rod B32.

[0125] <Example 4> In Example 4, ring-shaped test pieces A41 and A42, which were made in the same manner as in Example 1, as carbon fiber reinforced molded bodies, were used as rotating components, and NOL ring tests (according to ASTM D2290) were performed. That is, the circumferential tensile strength was determined using ring-shaped test pieces A41 and A42, which were formed by winding the composite material 10, which served as the fiber substrate, in parallel in one direction. As described above, the ring-shaped test pieces A41 and A42 were made using a filament winding method. That is, the composite material 10 (CNT composite fiber) after the first sizing treatment was impregnated with the matrix resin M and wound onto a cylindrical mandrel 34 equipped with a filament winder 33 to form a molded body. The molded body and the mandrel 34 were heated together to harden the matrix resin M. After the matrix resin M hardened, the molded body with the mandrel 34 removed was cut into predetermined widths to form ring-shaped test pieces A41 and A42. The matrix resin M was coated onto the composite material 10 after the first sizing treatment using a roller coater. Trading Card T1100 (manufactured by Toray Industries, Inc.) was used as carbon fiber 11 (raw yarn). Other manufacturing conditions for composite material 10 were the same as in Example 1.

[0126] The annular test pieces A41 and A42 have an inner diameter of 35 mm, an outer diameter of 40 mm, and a width of 3 mm. In the NOL ring test, a 5582 universal testing machine (manufactured by Instron) was used at a tensile speed of 2 mm / min. The test results are shown in Figure 21. In Figure 21, the horizontal axis represents the fiber volume fraction (Vf), and the vertical axis represents the relative tensile strength when the theoretical strength is set to 100% and the fiber volume fraction is 100%. Furthermore, the theoretical tensile strength values ​​for the same fiber volume fraction as annular test pieces A41 and A42 are also shown in the graph of Figure 21.

[0127] As Comparative Example 4, annular test pieces B41-B44, which were carbon fiber reinforced moldings using unattached carbon fiber (raw yarn) as the reinforcing fiber, were fabricated and subjected to NOL ring tests. The carbon fiber used in annular test pieces B41-B44 was the same as that used in annular test pieces A41 and A42, and the annular test pieces B41-B44 were fabricated under the same conditions as those in annular test pieces A41 and A42, except that carbon fiber was used as the raw yarn. The test results of annular test pieces B41-B44 of Comparative Example 4 are shown in Figure 21.

[0128] As shown in Figure 21, the ring test pieces A41 and A42, which use carbon fiber 11 with CNT 17 as the reinforcing fiber, have higher tensile strength than the ring test pieces B41~B44, which use carbon fiber (raw yarn) as the reinforcing fiber, and are close to the theoretical value of tensile strength.

[0129] [Example 5] In Example 5, the fragmentation method was used to evaluate the interfacial adhesion strength between carbon fiber 11 and the matrix resin. Simultaneously, the interfacial adhesion strength between carbon fiber 11 and the matrix resin was evaluated based on the difference in the mass ratio Rm of the sizing agent.

[0130] First, the concentration of the carbodiimide compound in the first sizing treatment solution was adjusted to prepare composite materials 10 with sizing agent mass ratios Rm of 0.6%, 0.8%, 1.0%, 1.1%, and 1.5%, respectively. For each composite material 10, a CNT composite fiber was taken from the carbon fiber bundle 12, and test pieces A51 to A55 were made by embedding this CNT composite fiber in soft epoxy resin. Three test pieces A51 to A55 were produced in each of the multiple production batches (lots) of composite material 10. A Trading Card T700SC (manufactured by Toray Industries, Inc.) was used as carbon fiber 11. Furthermore, the production conditions for the other composite materials 10 were the same as in Example 1.

[0131] For each test piece A51~A55, a tensile load was applied until the carbon fiber 11 could no longer be cut. Then, the length of each cut piece of carbon fiber 11 at a certain length was measured for each test piece. For each test piece A51~A55, the average length of the cut pieces (cut fiber length) of each production batch was calculated. The measurement results are shown in the graph of Figure 22.

[0132] As Comparative Example 5, a test piece B51 was fabricated in which a single carbon fiber (raw yarn) without CNTs was embedded in soft epoxy resin. Similar to test pieces A51-A55, a tensile load was applied to test piece B51, and the average length of the cut pieces (cut fiber length) for each production batch was determined. The measurement results are shown in the graph in Figure 22. Furthermore, the fabrication conditions for test piece B51 were the same as for test pieces A51-A55, except that the CNTs were not attached to the carbon fiber.

[0133] It can be seen that the cut fiber length of test pieces A51~A55 is shorter than that of test piece B51 in Comparative Example 5, and the interfacial adhesion strength between the CNT composite fiber and the matrix resin is higher than that of the carbon fiber (original yarn) without CNTs. Furthermore, it can be seen that, compared with test piece B51, the interfacial adhesion strength of test pieces A51~A54, which have a sizing agent mass ratio Rm in the range of 0.6% to 1.1%, is significantly higher.

[0134] Regarding the structures 14 in which the mass ratio of sizing agent Rm in the composite material 10 produced as described above is 0.8%, 1.1%, and 1.5%, the state of the first sizing agent 15 on the surface of the structures 14 observed using a field-emission scanning electron microscope (FE-SEM) is shown in Figures 23 to 25. It was confirmed that in the structures 14 with a sizing agent mass ratio Rm of 0.8% and 1.1%, the voids 19 formed by CNTs 17 were not blocked by the first sizing agent 15, but in the structures 14 with a sizing agent mass ratio Rm of 1.5%, many voids 19 were blocked by the first sizing agent 15.

[0135] [Example 6] In Example 6, composite material 10 was used as the reinforcing fiber to prepare multiple annular test pieces A6 with different fiber volume fractions (Vf) of carbon fiber as rotating parts 2, and their tensile strengths were measured by the NOL ring test (according to ASTM D2290). Furthermore, as Comparative Example 6, raw yarn (carbon fiber) was used as the reinforcing fiber to prepare multiple annular test pieces B6 with different fiber volume fractions (Vf) of carbon fiber, and their tensile strengths were also measured. In the NOL ring tests of Example 6 and Comparative Example 6, the aforementioned 5582 universal testing machine was used, and the tensile speed was 2 mm / min. Additionally, annular test pieces A6 and B6 were prepared in the same manner as in Example 4. However, both annular test pieces A6 and B6 used Trading Card T1100SC-12000 (manufactured by Toray Industries, Inc.) as carbon fiber 11.

[0136] The measurement results described above are shown in Figure 26. In Figure 26, ● represents the measurement results of Example 6, and ▲ represents the measurement results of Comparative Example 6. Furthermore, in Figure 26, the horizontal axis represents the fiber volume fraction (Vf) of the carbon fiber in the annular test piece, and the vertical axis represents the tensile strength (index value). The solid line represents an approximate straight line showing the relationship between the fiber volume fraction and tensile strength (index value) obtained by the least square method based on the measurement values ​​of Example 6, while the dashed line represents an approximate straight line showing the relationship between the fiber volume fraction and tensile strength (index value) obtained by the same method as in Example 6 based on the measurement values ​​of Comparative Example 6. The tensile strength (index value) is the value representing the tensile strength relative to the fiber volume fraction of the reinforcing fiber when the tensile strength is 70%, calculated based on the relationship (approximate straight line) between the fiber volume fraction and tensile strength (index value) of Comparative Example 6, and set to "100". Furthermore, Figure 27 shows a magnified view of the area in Figure 26 where the fiber volume content is above 60% and the tensile strength (index value) is above 90.

[0137] Figures 26 and 27 confirm that the increase in tensile strength is directly proportional to the fiber volume content. Even in the relatively low range of 60% to 70% fiber volume content, and even in the relatively high range of 70% to 75% fiber volume content, the tensile strength (index value) of each annular test piece A6 is plotted above the approximate straight line representing the relationship between fiber volume content and tensile strength (index value) in Comparative Example 6. In other words, it can be seen that each annular test piece A6 using CNT composite fiber as the reinforcing fiber has a higher tensile strength than the annular test piece using carbon fiber yarn with the same fiber volume content as the reinforcing fiber. Specifically, with the same fiber volume content, the tensile strength of the annular test piece using composite material 10 as the reinforcing fiber, as in Example 6, is approximately 10% higher than the tensile strength of the annular test piece using raw yarn as the reinforcing fiber, as in Comparative Example 6. Furthermore, to obtain the same tensile strength, the fiber volume content is reduced by approximately 10% when composite material 10 is used as the reinforcing fiber compared to when the original yarn is used as the reinforcing fiber. Approximately linearly, the tensile strength of the ring-shaped test piece with a fiber volume content of 74% when composite material 10 is used as the reinforcing fiber is equivalent to the tensile strength of the ring-shaped test piece with a fiber volume content of 81% when the original yarn is used as the reinforcing fiber.

[0138] Typically, when the fiber volume content exceeds 80%, the amount of matrix resin decreases, causing the carbon fiber reinforced molded body itself to become brittle. Therefore, there is a limit to increasing the fiber volume content in order to improve tensile strength. By using composite material 10 as the reinforcing fiber of the rotating component 2, even if the fiber volume content is relatively small, higher tensile strength can be obtained compared with the case of using raw yarn as the reinforcing fiber. Thus, it is possible to balance high tensile strength and high brittleness resistance.

[0139] [Example 7] As Example 7, the failure state of multiple ring-shaped test pieces with different fiber volume fractions (Vf) was evaluated. The multiple ring-shaped test pieces evaluated included the ring-shaped test pieces A6 for which tensile strength was measured in Example 6 above. The failure state was evaluated as follows. First, the multiple ring-shaped test pieces A6 to be evaluated were divided into a first group with a fiber volume fraction of 60% or more but less than 70%, and a second group with a fiber volume fraction of 70% or more. Then, they were stretched using a 5582 universal testing machine at a tensile speed of 2 mm / min, and the failure state of each ring-shaped test piece after failure was examined. It can be seen that the failure state of the ring-shaped test pieces can be roughly classified into two types: fault failure mode and interlaminar failure mode. Each group was classified into fault failure mode and interlaminar failure mode. The percentage of each type of failure mode in each group was calculated. The obtained percentages are shown in Table 1 as evaluation results. Furthermore, as Comparative Example 7, the failure state was evaluated in the same manner as in Example 7 for multiple ring-shaped test pieces with different fiber volume fractions (Vf) of carbon fibers and using the original yarn as the reinforcing fiber. The multiple ring-shaped test pieces evaluated included each ring-shaped test piece B6 for which tensile strength was measured in Comparative Example 6 above. The evaluation results are shown in Table 2.

[0140] As shown in Figure 28, the aforementioned tomographic failure mode is a failure pattern in which the multiple carbon fibers constituting the annular test piece are individually severed. Analysis of the X-ray CT images revealed that this tomographic failure mode obliquely penetrates each fiber layer, generating cracks and causing damage. Furthermore, it was confirmed that almost all or nearly all of the fiber layers remained intact after the failure. On the other hand, as shown in Figure 29, the interlayer failure mode is a failure mode in which cracks form in the circumferential direction, resulting in a collapsed shape. In the interlayer failure mode, analysis of the X-ray CT images confirmed traces of delamination between the fiber layers of the annular test piece. It was confirmed that almost all of the fiber layers had delaminated after the failure.

[0141] [Table 1] Example 7 60%≤Vf<70% 70%≤Vf Fault failure mode 79% 100% Interlayer failure mode twenty one% 0%

[0142] [Table 2] Comparative Example 7 60%≤Vf<70% 70%≤Vf Fault failure mode 0% 100% Interlayer failure mode 100% 0%

[0143] In Comparative Example 7, the second group (70% ≤ Vf) with relatively high tensile strength was entirely characterized by dissociative failure, while the first group (60% ≤ Vf < 70%) with relatively low tensile strength was entirely characterized by interlaminar failure. In contrast, in Example 7, the second group, like Comparative Example 7, was entirely characterized by dissociative failure, but in the first group, 21% were characterized by interlaminar failure and 79% by dissociative failure. This can be attributed to the fact that in Example 7, even with a fiber volume content of 60% or more but less than 70%, the interfacial adhesion strength between the matrix resin M and the carbon fiber 11 was significantly improved by forming the composite region 18, making interlaminar delamination less likely and reducing the number of interlaminar failure modes compared to Comparative Example 7.

[0144] [Example 8] In Example 8, a portion of the annular test piece A6, which broke in the tensile test of Example 6, was cut out as an observation sample for observing the cut surface. The cut surface of this observation sample was ground and observed using both scanning electron microscopy (SEM) and field-emission scanning electron microscopy (FE-SEM). Furthermore, in Comparative Example 8, a portion of the annular test piece B6, which broke in the tensile test of Comparative Example 6, was cut out to prepare an observation sample, and it was observed in the same manner as in Example 8. Figure 30 is a SEM photograph of Example 8, Figure 31 is a SEM photograph of Comparative Example 8, Figure 32 is a FE-SEM photograph of Example 8, and Figure 33 is a FE-SEM photograph of Comparative Example 8. Additionally, the fiber volume content of the annular test piece A6 obtained by cutting out the observation sample was 72.2%, while the fiber volume content of the annular test piece B6 was 72.5%.

[0145] In the annular test pieces A6 and B6, as shown by the areas enclosed by dashed lines in Figures 30 and 31, it was observed that both annular test pieces A6 and B6 had the densest filling areas with very dense carbon fibers, and no significant difference was found between Example 8 and Comparative Example 8. The densest filling areas were confirmed when the fiber volume content was approximately 72%. Therefore, if the fiber volume content is further increased, these areas will become brittle due to a reduction in the amount of matrix resin absorbing deformation. Therefore, there are limitations to improving tensile strength by increasing the fiber volume content.

[0146] As observed in Figure 32, in the annular test piece A6, at the interface between the carbon fiber 11 and the matrix resin M, a composite region 18 with a thickness of approximately 0.1 μm covers the surface of the carbon fiber 11, resulting in a uniform film thickness. Multiple CNTs 17 are intertwined in a non-woven fabric manner, meaning the matrix resin M enters the voids within the three-dimensional network structure 14. This matrix resin M becomes a three-dimensional network and interpenetrates with each other in a network structure. The CNTs 17 and the matrix resin M form a nanoscale interpenetrating network structure. It was confirmed that in the densest filling area, where adjacent carbon fibers 11 are close to each other, the composite region 18 is compressed to a thickness of less than 0.01 μm. It can be seen that these composite regions 18 on adjacent carbon fibers 11 are fixedly bonded together, and the aforementioned carbon fibers 11 form a cross-linked structure.

[0147] On the other hand, as shown in Figure 33, a clear boundary was observed at the interface between the carbon fibers and the matrix resin in the annular test piece B6. At the densest filling, adjacent carbon fibers are in contact with each other, or only a very thin layer of matrix resin lies between adjacent carbon fibers. Therefore, the bonding strength between two adjacent carbon fibers does not exceed the adhesive strength at the interface between the carbon fibers and the matrix resin.

[0148] In the annular test piece A6, the cross-linking structure makes the two adjacent carbon fibers 11 adhere more firmly than in the annular test piece B6. Therefore, it can be seen that even when a large tension is applied outward along the diameter of the annular test piece A6, the bond is not easily broken due to the cross-linking structure between adjacent carbon fibers 11, and it is found that the integrally formed multiple carbon fibers 11 have a higher tensile strength than the annular test piece B6.

[0149] 2: Rotating components 3: Surface magnet type motor 4: Rotor 5: Permanent magnets 10: Composite materials 11: Carbon fiber 12: Carbon fiber bundles 14: Construct 15: First sizing agent 17: Carbon nanotubes 18: Composite Area 19: Gap 21: Attachment device 22: CNT attachment groove 23, 24, 25, 26: Guide rollers 27: Ultrasonic generator 28: Dispersion 31: Wire feeding device 32: Resin application device 33: Wire Winder 34: Mandrel 35: Contact Roller 35a: Guide roller 36: Storage slot 37: Second sizing agent CL: Crosslinking section M: Matrix resin

Claims

1. A rotating component, wherein the rotating component rotating in a circumferential direction includes a carbon fiber wound in a circumferential direction, a matrix resin covering the carbon fiber, and a structure disposed on the surface of the carbon fiber, wherein the structure is composed of a plurality of carbon nanotubes having a curved shape with a bend, and forms a mesh structure having contact portions in which the carbon nanotubes are in direct contact with each other.

2. The rotating component as claimed in claim 1, wherein the thickness of the aforementioned structure is in the range of 50 nm to 200 nm.

3. The rotating component as described in claim 1 or 2, wherein the fiber volume content of the aforementioned carbon fiber is less than 75%.

4. A method for manufacturing a rotating component, comprising: The structure forming step involves immersing carbon fibers in a dispersion liquid subjected to ultrasonic vibration while dispersing carbon nanotubes with curved portions, thereby attaching a plurality of the aforementioned carbon nanotubes to the aforementioned carbon fibers to form a structure on the surface of the aforementioned carbon fibers, the aforementioned structure having a mesh structure with contact portions in which the aforementioned carbon nanotubes are in direct contact with each other; and the molding step involves coating a matrix resin onto the aforementioned carbon fibers on which the aforementioned structure has been formed, winding the aforementioned carbon fibers coated with the aforementioned matrix resin onto a mandrel, and pulling out the aforementioned mandrel after the aforementioned matrix resin has hardened.

5. A method for manufacturing a rotating component as claimed in claim 4, wherein after the aforementioned structure forming step, a first sizing step is included, in which the aforementioned carbon fibers are brought into contact with a first sizing solution containing a carbodiimide compound having a carbodiimide group, to provide a first sizing agent that crosslinks the aforementioned carbon nanotubes in direct contact with each other.

6. The method for manufacturing a rotating component as claimed in claim 5, wherein in the aforementioned first sizing step, the mass of the aforementioned first sizing agent is in the range of 0.6% to 1.1% relative to the mass of the aforementioned carbon fiber.

7. A method for manufacturing a rotating component as claimed in claim 5 or 6, wherein in the aforementioned structure forming step, a structure with a thickness between 50 nm and 200 nm is formed.

8. A method for manufacturing a rotating component as claimed in any one of claims 4 or 6, wherein in the aforementioned forming step, the aforementioned carbon fiber coated with the aforementioned matrix resin is wound onto a mandrel such that the fiber volume content of the carbon fiber in the rotating component is less than 75%.

Images