Martensitic steel achieving both high fatigue limit and high strength and method of producing the same

Abstract

To provide a method of producing a martensitic steel achieving both high fatigue limit and high strength.SOLUTION: By performing pre-fatigue deformation on a martensitic steel with a tensile strength of 1.5 GPa or higher obtained by quenching steel materials having a composition containing C: 0.15 to 0.6 mass%, Mn: 0.2 to 5 mass%, and the remainder being Fe and unavoidable impurities, after austenitization under conditions that prevent crack initiation and thereby strengthening areas susceptible to cracking, crack initiation is suppressed and the fatigue limit is improved. For pre-fatigue deformation, pre-fatigue deformation limited to the smaller of either 101 cycles or more and the number of less than the crack initiation cycle under each stress condition, or 107 cycles or less is performed with an initial maximum stress of 1 MPa or more and 99% or less of the tensile strength, and a stress ratio of -1 or more and 0.99 or less. For the second and subsequent pre-fatigue deformations, the maximum stress is increased from the previous deformation, and as long as there is no cracking, the pre-fatigue deformation and maximum stress increase are repeated.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a martensitic steel that combines a high fatigue limit with high strength, and a method for producing the same. [Background technology]

[0002] The fatigue limit of most steel materials corresponds to the crack non-propagation limit. Therefore, the conventionally accepted upper limit of the fatigue limit for high tensile strength can also be said to be the upper limit of the crack non-propagation limit. Figure 6 shows the maximum non-propagating crack length √area required to achieve a fatigue limit of a tensile strength ratio of 0.5 in the case of fully reversed stress (stress ratio -1) based on the conventional theory (Non-Patent Document 1). According to Non-Patent Document 1 and Non-Patent Document 2, the fatigue limit σ w and tensile strength σ B is given by the following equation:

number

[0003] The curves in Figure 6 show that for tensile strengths above 1.5 GPa, surface cracks must be stopped at a size of approximately 2 μm or less. However, because early fatigue cracks formed along slip bands or grain boundaries are typically several grains in size (several μm to several tens of μm), it is natural that an upper limit for high tensile strength appears in the fatigue limit corresponding to the crack non-propagation limit. Therefore, in order to improve the fatigue limit of high-strength steels with tensile strengths above 1.5 GPa and overcome the upper limit for high tensile strength, it is believed that it is more effective to improve the crack initiation limit rather than crack non-propagation limit.

[0004] Shot peening is a widely used method for improving the crack initiation limit. Shot peening imparts residual compressive stress to the material surface, suppressing crack initiation. However, shot peening may not be effective in martensitic steel, the base material for low- and medium-carbon high-strength steels. It has been found that shot peening of as-quenched martensitic steels only shifts the crack initiation point from the surface to the interior, with almost no effect on improving the fatigue limit (Non-Patent Document 3). Therefore, there is a need to develop steel materials that achieve both a high fatigue limit and high strength by controlling the structure of the entire material, not just the surface.

[0005] Patent Document 1 also proposes a method for improving the fatigue limit by using impacts with an ultrasonic vibration terminal. However, like shot peening, this method introduces compressive residual stress into the material surface and does not modify the structure of the entire material, suggesting that it is not very effective for as-quenched martensitic steel, etc. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 4374306 [Non-patent literature]

[0007] [Non-Patent Document 1] Takayoshi Murakami, The influence of small defects and non-metallic inclusions on fatigue strength and their quantitative evaluation method, Iron and Steel, Vol. 75 (1989), pp. 1267-1277 [Non-patent document 2] Norihiko Hasegawa, Junichi Arai, Michishichi Tanaka, Correlation between static strength parameters, Materials, Vol. 39 (1990), pp. 859-863 [Non-patent document 3] N. Iwata et al., Materials Science and Technology, 18(2002), 629 [Non-patent document 4] K. Okada et al., International Journal of Fatigue, 143(2021), 105921 Summary of the Invention [Problem to be solved by the invention]

[0008] As mentioned above, there is still room for further development in the development of steel materials that achieve both a high fatigue limit and high strength by controlling the structure of the entire material.

[0009] The present invention aims to solve these problems, and its first objective is to provide a material with a high fatigue limit compared to untreated materials by controlling the structure of the entire material, not just the surface, in high-strength steel with a tensile strength of over 1.5 GPa, thereby improving the crack initiation limit rather than the crack non-propagation limit. A second object of the present invention is to provide a high strength steel having a fatigue limit of about 0.42 times the tensile strength (stress ratio 0), ie, a fatigue limit of more than 0.63 GPa, among high strength steels having a tensile strength of more than 1.5 GPa. [Means for solving the problem]

[0010] Martensitic steel is often tempered before use, and it is known that the fatigue limit of tempered martensitic steel is the crack non-propagation limit. However, the inventors discovered that the fatigue limit of as-quenched martensitic steel corresponds to the crack initiation limit, which led to the invention. In other words, it has been conventionally believed that the fatigue limit of most steel materials corresponds to the crack non-propagation limit. However, contrary to the conventional belief, the inventors have discovered that the fatigue limit of as-quenched martensitic steel (smooth material) corresponds to the crack initiation limit. In other words, as-quenched martensitic steel is the optimal material for the present invention, which aims to improve the crack initiation limit rather than the crack non-propagation limit. In general, metallic materials deform non-uniformly in accordance with their microstructure, and the deformation concentrates in areas prone to localized deformation, resulting in crack initiation. Therefore, by pre-fatigue deformation under conditions that prevent crack initiation and strengthening areas prone to deformation (where cracks may occur) in advance, it is expected that crack initiation can be suppressed and the crack initiation limit (fatigue limit) can be improved. The present inventors therefore believed that the crack initiation limit of as-quenched martensitic steel could be improved by applying pre-fatigue deformation to as-quenched martensitic steel, and thus conceived the present invention.

[0011] [1] The martensitic steel of the present invention, which achieves both a high fatigue limit and high strength, achieves the first object, and contains 0.15 to 0.6 mass% C, 0.2 to 5 mass% Mn, and the balance being Fe and unavoidable impurities. The martensitic steel has a tensile strength of 1.5 GPa or more, which is obtained by austenitizing a steel material of the above composition and then quenching it. The coefficient of variation in a micro-Vickers hardness test is 3.2 × 10 -2 The ratio of the average nano hardness of the large angle boundary where the crystal orientation difference is 15 degrees or more measured by the nanoindentation test to the average measured value of the micro Vickers hardness test is 1.55 × 10 -2 The above is the first objective achievement criterion. Here, the coefficient of variation is the value obtained by dividing the standard deviation by the mean value. [2] In the martensitic steel [1] of the present invention which achieves both a high fatigue limit and high strength, preferably, the first and second objects are achieved, and the coefficient of variation in the micro Vickers hardness test is 2.5 × 10 -2 or less (criterion for achieving the second objective), and the ratio of the average nanohardness of the large-angle boundary where the crystal orientation difference is 15 degrees or more measured by the nanoindentation test to the average measured value of the micro Vickers hardness test is 1.62 × 10 -2 It is desirable that the above (criteria for achieving the second objective) be met.

[0012] [3] In the martensitic steel [1] according to the present invention that combines a high fatigue limit and high strength, the micro-Vickers hardness test is preferably in accordance with the provisions of JIS Z 2244 (Vickers hardness test: ISO 6507-1), and the measured values ​​of the micro-Vickers hardness test are preferably measured at 1000 or more points under a test load of 490 mN. [4] In the martensitic steel [1] of the present invention that combines a high fatigue limit and high strength, the nanoindentation test is preferably in accordance with the provisions of JIS Z 2255 (ultra-micro load hardness test method: ISO 14577), and the measurement value of the nanoindentation test is preferably obtained using a Berkovich indenter. [5] In the martensitic steel of the present invention [4] that combines high fatigue limit and high strength, the nanohardness of the high-angle boundary where the crystal orientation misorientation is 15 degrees or more is preferably the average value of measurements taken at 300 or more points under a maximum load of 1000 μN.

[0013] [6] The method for producing a martensitic steel having both a high fatigue limit and high strength according to the present invention achieves the first object, and includes the steps of: preparing a martensitic steel having a tensile strength of 1.5 GPa or more by austenitizing a steel material having the above composition, containing 0.15 to 0.6 mass% of C, 0.2 to 5 mass% of Mn, and the balance being Fe and unavoidable impurities, as shown in FIG. 1 (S100); For the martensitic steel material after heat treatment quenched after austenitization, The lower limit of the number of cycles for pre-fatigue deformation is set to 10 1 The upper limit is the maximum stress (σ max ) the number of cycles for crack initiation (N pre ) or less than 10 7 cycles, whichever is smaller:

number

[0014] [7] In the method for manufacturing a martensitic steel having both a high fatigue limit and high strength according to the present invention [6], preferably, the lower limit of the number of cycles of the pre-fatigue deformation is 10 2 The upper limit of the maximum stress of the pre-fatigue deformation is preferably 98% or less of the tensile strength above the elastic limit, and the stress ratio of the pre-fatigue deformation is preferably 0 or more and 0.97 or less. [8] In the manufacturing method of the martensitic steel having both a high fatigue limit and high strength according to the present invention [7], more preferably, the lower limit of the number of cycles of the pre-fatigue deformation is 10 3 The upper limit of the maximum stress of the pre-fatigue deformation is preferably 70% or more of the yield stress or 0.2% proof stress and 97% or less of the tensile strength, and the stress ratio of the pre-fatigue deformation is preferably 0 or more and 0.2 or less. [9] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength [6], it is preferable to further perform grinding after the completion of the pre-fatigue deformation to remove irregularities on the material surface introduced by the pre-fatigue deformation.

[10] In the method for manufacturing a martensitic steel that achieves both a high fatigue limit and high strength according to the present invention [9], preferably, the grinding removes a range of 1 μm to 100 μm from the surface of the martensitic steel material.

[0015]

[11] In the method for manufacturing a martensitic steel [6] that achieves both a high fatigue limit and high strength of the present invention, preferably, the first and second objects are achieved by carrying out the pre-fatigue deformation described in the method for manufacturing a martensitic steel [6] as the first pre-fatigue deformation, and further, the maximum stress in the pre-fatigue deformation from the second pre-fatigue deformation onwards (nth pre-fatigue deformation (n=2, 3, 4, . . .)) is increased by a predetermined value from the maximum stress in the most recent pre-fatigue deformation (n-1th pre-fatigue deformation), and the maximum stress after the increase satisfies the maximum stress given by the following formula:

number

[0016]

[12] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength

[11] , preferably, the second and subsequent (nth) pre-fatigue deformations are repeated up to a maximum stress that is within a range in which no cracks have been generated in the most recent (n-1th) pre-fatigue deformation and no cracks have been generated in the nth pre-fatigue deformation.

[13] In the manufacturing method

[11] of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength, it is preferable that in the first and each subsequent (nth) pre-fatigue deformation, the maximum value of the stress applied in each pre-fatigue deformation is constant relative to the initial maximum stress in the current (nth) pre-fatigue deformation.

[14] In the manufacturing method

[11] of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength, preferably, in the first and each subsequent (nth) pre-fatigue deformation, the maximum value of the stress applied in each pre-fatigue deformation is varied within a range from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value.

[15] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength

[14] , preferably, the maximum value of the maximum stress in the current (nth) pre-fatigue deformation is gradually increased as the number of cycles of the current pre-fatigue deformation increases, within a range from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value.

[16] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength

[11] , it is preferable that the value by which the maximum stress in the previous pre-fatigue deformation is increased by a predetermined value is 200 MPa or less.

[17] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength

[16] , it is preferable that the value by which the maximum stress in the previous pre-fatigue deformation is increased by a predetermined value is 100 MPa or less.

[18] In the manufacturing method of the present invention for a martensitic steel that achieves both a high fatigue limit and high strength

[11] , it is preferable to further perform grinding after the second or subsequent (nth) pre-fatigue deformation to remove any irregularities on the material surface that have been introduced by the first nth pre-fatigue deformation.

[19] In the method for manufacturing a martensitic steel having both a high fatigue limit and high strength according to the present invention

[18] , the method is preferably carried out after subjecting the martensitic steel to pre-fatigue deformation for a number of cycles that gives the martensitic steel a fatigue limit of at least 0.40 times the tensile strength.

[20] In the method for manufacturing a martensitic steel that achieves both a high fatigue limit and high strength according to the present invention

[19] , preferably, the grinding removes a range of 1 μm to 100 μm from the surface of the martensitic steel material.

[21] In any of the manufacturing methods [5] to

[20] of the present invention for producing a martensitic steel that achieves both a high fatigue limit and high strength, the frequency of the pre-fatigue deformation is preferably 0.01 Hz or more and 300 Hz or less (0.6 CPM or more and 18,000 CPM (cycles per minute) or less).

[0017] <Reasons for limiting steel composition> C is an effective element for strengthening steel, but if it is less than 0.15 mass%, sufficient strength cannot be obtained. On the other hand, if it is added in excess, lens / butterfly / thin plate martensite may appear, and a single-phase lath martensite structure will not be obtained, so the upper limit of the amount of C added is set at 0.6 mass%. Mn is an important element for strengthening steel and ensuring hardenability, but a sufficient effect cannot be obtained at 0.2 mass% or less. On the other hand, if added in excess, retained austenite may be produced, preventing the formation of a single-phase lath martensite structure, so the upper limit for its addition is set at 5 mass%. Other steel components are unavoidable impurities and are not particularly limited.

[0018] <Reasons for limiting the conditions for pre-fatigue deformation> Since the present invention aims to improve the fatigue limit by suppressing the occurrence of cracks through pre-fatigue deformation, all pre-fatigue deformation must be performed under conditions (stress and number of cycles) that do not cause cracks to occur. The first objective of this invention is to provide a material with a higher fatigue limit than untreated materials by controlling the structure of the entire material, not just the surface, to improve the crack initiation limit rather than the crack non-propagation limit, thereby achieving a higher fatigue limit than untreated materials. Even a single pre-fatigue deformation is effective. Applying a stress greater than the tensile strength during this single pre-fatigue deformation will cause the material to fracture, so the maximum stress is set to 99% or less of the tensile strength. Because cracks will form in the material if the load during the initial pre-fatigue deformation is too large, the initial pre-fatigue deformation is performed on the as-heat-treated material (undeformed material) within a range that satisfies equations (3a) and (3b). Furthermore, in order to achieve the second objective of this invention, which is to provide high-strength steel with a tensile strength exceeding 1.5 GPa, which is approximately 0.42 times the tensile strength (stress ratio 0), i.e., a fatigue limit exceeding 0.63 GPa, it is necessary to apply pre-fatigue deformation from the second onwards (nth deformation (n = 2, 3, 4, ...)). If the stress increase or final load reached in these pre-fatigue deformations from the second onwards is too large, cracks will form in the material. Therefore, the maximum stress increase between each deformation is set to 200 MPa or less, and the maximum stress in the final pre-fatigue deformation is set to be equal to or less than the upper limit defined by equation (4). Although there is no particular restriction on the lower limit of the maximum stress for pre-fatigue deformation, it is set to 1 MPa or more because a load must be applied. Furthermore, as shown in the examples, pre-fatigue deformation with a larger maximum stress and a larger stress amplitude is more effective in improving the fatigue limit.

[0019] The fewer the number of cycles of pre-fatigue deformation, the shorter the process can be, but if the number of cycles is too small, sufficient effect cannot be obtained. Therefore, the lower limit of the number of cycles of pre-fatigue deformation is preferably 10 1 10 cycles or more, more preferably 10 2 cycles or more, most preferably 10 3 The upper limit of the number of cycles for pre-fatigue deformation is preferably less than the number of cycles for crack initiation under each stress condition (defined by formulas (3a) and (3b)) or 10 7 The fatigue deformation treatment is carried out with the smaller of the following cycles as the upper limit: Fatigue cracks often originate on the surface of the material. If the unevenness introduced into the material surface by plastic deformation during pre-fatigue deformation is left behind, it can become a local stress concentration site and reduce the fatigue limit. Therefore, it is a good idea to grind the surface after pre-fatigue deformation. When grinding the surface after pre-fatigue deformation, if the amount of grinding is too small, it will not be effective in removing surface unevenness, so it is recommended to grind at least 1 μm from the surface of the steel. There is no particular upper limit on the grinding thickness, but grinding beyond 100 μm will increase the working time and wear on the grinding equipment, resulting in higher costs. [Effects of the Invention]

[0020] According to the martensitic steel of the present invention as set forth in claim 1 and the manufacturing method thereof as set forth in claim 6 and following, in high-strength steel with a tensile strength of more than 1.5 GPa, by controlling the structure of the entire material, not just the surface, it is possible to improve the crack initiation limit rather than the crack non-propagation limit, and provide a material with a higher fatigue limit than untreated material. According to the martensitic steel of the present invention as set forth in claim 2 and the manufacturing method thereof as set forth in claim 11 and following, a high-strength steel having a fatigue limit of about 0.42 times the tensile strength (stress ratio 0), i.e., a fatigue limit of more than 0.63 GPa, can be provided, thereby breaking the upper limit of the fatigue limit of high tensile strength steels that existed in conventional martensitic steels. Therefore, it is expected that the use of the martensitic steel after the pre-fatigue deformation treatment of the present invention for automobile suspension parts and the like will promote the thinning and weight reduction of steel members that require high fatigue properties, leading to a reduction in the environmental load. [Brief explanation of the drawings]

[0021] [Figure 1] 1 shows a flow chart of a testing procedure illustrating one embodiment of the present invention. [Figure 2] The figures show the shapes of the test specimens used for the undeformed material and pre-fatigued material shown in Figure 1, with (A) being a front view and (B) being a side view. [Figure 3] The graph shows fatigue characteristics (relationship between maximum stress and number of cycles to fracture) (stress ratio 0.1). [Figure 4] FIG. 1 is a diagram showing the correlation between the tensile strength and fatigue limit of steel materials (converted to a stress ratio of 0). [Figure 5] FIG. 1 shows the correlation between the fatigue limit and (a) the average micro-Vickers hardness value, (b) the standard deviation of the micro-Vickers hardness value, and (c) the average nano-hardness value of the large-angle boundary. [Figure 6] Based on the conventional theory, the maximum non-propagating crack length required to achieve the fatigue limit at a tensile strength ratio of 0.5 in a fully alternating fatigue test (stress ratio -1) is shown. DETAILED DESCRIPTION OF THE INVENTION

[0022] The best mode for carrying out the present invention will be described in detail below. FIG. 1 shows a flow chart of a test procedure illustrating one embodiment of the present invention. Figure 2 shows the shape of the test specimens used for the undeformed material and pre-fatigued material shown in Figure 1, with (A) being a front view and (B) being a side view. The shape of the test specimen conforms to the plate-shaped test specimens in JIS Z2241, "Methods for Tensile Testing of Metallic Materials," with the plate-shaped test specimen having a thickness of 1.0 mm before the start of the test, a parallel section width of 4.0 mm, a parallel section length of 8.0 mm, a total length of 59.0 mm, a gripping section width of 11.5 mm, a gripping section length of 15.6 mm, and a shoulder radius of 15 mm.

[0023] In the method for producing a martensitic steel of the present invention that achieves both a high fatigue limit and high strength, as shown in FIG. 1, for example, a steel material having the above composition contains 0.15 to 0.6 mass% C, 0.2 to 5 mass% Mn, and the balance being Fe and unavoidable impurities. The steel material is austenitized and then quenched to obtain a martensitic structure and to prepare a martensitic steel having a tensile strength of 1.5 GPa or more (S100). To remove surface irregularities from the undeformed martensitic steel specimen, the specimen surface is polished and ground (S110). For this polishing, emery paper #80 to #4000 is used, for example. Emery paper is abrasive paper made by mixing alumina or silicon carbide abrasive grains with a binder and applying it to paper or cloth, then solidifying it. It is classified by the size of the abrasive grains, with the higher the number, the finer the abrasive grains. Alternatively, surface grinding, rotary grinding, cylindrical grinding, internal grinding, lapping, buffing, electrolytic polishing, or chemical polishing may be used instead of emery paper.

[0024] Next, the as-quenched martensitic steel was subjected to 10 1 10 cycles or more and less than the number of cycles required for crack initiation under each stress condition 7 The pre-fatigue deformation is defined as the maximum stress (σ max The lower limit of the number of cycles of pre-fatigue deformation is more preferably 10 2cycles or more, most preferably 10 3 It is recommended to use more than one cycle. σ of the first pre-fatigue deformation max It is preferable that the stress is 1 MPa or more and 99% or less of the tensile strength, more preferably from the elastic limit to 98% or less of the tensile strength, and most preferably from 70% or more of the yield stress or 0.2% proof stress to 97% or less of the tensile strength. The stress ratio of the pre-fatigue deformation is preferably -1 or more and 0.99 or less, more preferably 0 or more and 0.97 or less, and most preferably 0 or more and 0.2 or less.

[0025] The test frequency is not particularly limited, but at low frequencies below 0.01 Hz, the pre-fatigue deformation process takes too long and becomes costly, while at high frequencies above 300 Hz, the mechanical operating speed of the fatigue testing machine performing the pre-fatigue deformation process becomes too fast, making the fatigue testing machine costly and potentially altering the structure due to heat generated during the pre-fatigue deformation, so these are not appropriate. The number of cycles to crack initiation under each stress condition varies depending on the stress ratio of the pre-fatigue deformation. For example, if we assume that the number of cycles to crack initiation is half the fatigue life, we can roughly calculate the number of cycles to crack initiation by experimentally obtaining fatigue life diagrams for each stress ratio (see equations (6a), (6b), (7a), and (7b) below). Although it is more accurate to predict the number of cycles to crack initiation based on experimental data for individual materials and test conditions, it is also useful in design to be able to roughly estimate the number of cycles to crack initiation under any test condition using empirical formulas (see equations (3a) and (3b)).

[0026] Next, the surface of this martensitic steel pre-fatigue test piece is polished and ground to remove any surface irregularities (S130). This polishing is performed using, for example, emery paper #600 to #4000, followed by electrolytic polishing. This polishing and grinding of the sample surface preferably removes at least approximately 25 μm from the surface (front and back, left and right sides), resulting in a pre-fatigue test piece with a thickness of 0.95 mm. Then, a fatigue test is performed under stress control using a flat, smooth test piece (S140). The stress in the fatigue test is the stress for determining the fatigue limit, for example, 1300 MPa, 7 The fatigue deformation is carried out in cycles, and if it breaks, the number of times it breaks is taken as the fatigue life.

[0027] On the other hand, if the martensitic steel specimen does not fracture in the fatigue test of S140, the specimen surface is polished and ground to remove surface irregularities (S130). This polishing is performed using, for example, emery paper #600 to #4000, followed by electrolytic polishing. This polishing and grinding of the specimen surface preferably removes a thickness of 1 μm to 100 μm from the surface, and more preferably removes at least approximately 25 μm from the surface. When grinding to remove 25 μm from the surface, the thickness of the second pre-fatigue specimen is 0.90 mm. In the second pre-fatigue deformation, the maximum stress is increased (by 200 MPa or less) based on the maximum stress of the first fatigue test. Then, a fatigue test is performed under stress control using a flat, smooth test piece (S140). The stress in the fatigue test is the stress for determining the fatigue limit, for example, 1400 MPa, and the stress is increased by 10 7 The fatigue deformation is carried out in cycles, and if it breaks, the number of times it breaks is taken as the fatigue life.

[0028] On the other hand, if no fracture occurs in the second fatigue test in S140, the sample surface is polished and ground to remove any irregularities on the surface of the pre-fatigue material test piece made of martensitic steel (S130). In the third and subsequent pre-fatigue deformations, the maximum stress is increased (by an increase of 200 MPa or less) based on the maximum stress of the most recent fatigue test. Then, a fatigue test is performed under stress control using a flat, smooth test piece (S140). In addition, the number of pre-fatigue cycles at each step is set to 10. 5 In the case of the number of cycles or less, polishing of the specimen surface after pre-fatigue deformation is sufficient only after the final pre-fatigue deformation. In this way, for example, when the stress ratio is 0.1, it is preferable to repeat the process up to as high a stress as possible while still satisfying the following condition: The following conditional formula (4) was obtained empirically, and N pre is the upper limit of the number of repeated cycles of pre-fatigue deformation at each step when the maximum stress of pre-fatigue deformation is increased in stages. pre When an arbitrary value is set for , σ exceeds the following condition. max If the temperature is increased to , the possibility of crack initiation is high.

number

[0029] As-quenched martensitic steel (lath martensite single phase structure) with a tensile strength of 1.6 GPa was used as the high-strength steel with a tensile strength of over 1.5 GPa. The test specimen was an Fe-3Mn-0.2C (mass%) alloy. Table 1 shows the detailed composition of the test specimen. [Table 1]

[0030] The as-received material was austenitized at 900°C for 30 minutes, then quenched in ice-water brine and subzero-treated in liquid nitrogen for 110 minutes to obtain a lath martensite single-phase structure. This as-quenched martensitic steel is referred to as the undeformed material. As shown in Table 2, the tensile strength of the undeformed material was 1615 MPa, and the 0.2% yield strength was 1118 MPa. Using a flat, smooth test piece as shown in Figure 2, a fatigue test was carried out under stress control, a frequency of 50 Hz, and a stress ratio of 0.1 (uniaxial tension). Figure 3 is a graph showing the relationship between maximum stress and the number of cycles to fracture. The fatigue limit (maximum stress) of the undeformed material was 675 MPa. The number of cycles to crack initiation (N int ) is half of the fatigue life, the fatigue life diagram (approximate dashed line in Figure 3, undeformed material) shows that σ max Maximum number of pre-fatigue cycles (N pre) is given by the following equations (5a) and (5b): σ B is 1615 MPa.

number

[0031] Next, pre-deformation processing was performed on the undeformed material, and the details of the pre-fatigue processing for the planned load material and pre-fatigued material (A to F) are shown in Table 2. Of these, the planned load material and pre-fatigued material (C) correspond to Comparative Examples 1 and 2, and the pre-fatigued materials (A, B, D to F) correspond to Examples 1 to 5. [Table 2]

[0032] The undeformed material was subjected to a preliminary fatigue deformation of 1000 MPa maximum stress, 50 MPa stress amplitude (equivalent to a stress ratio of 0.9), and 50 Hz frequency. 7 This was designated as the pre-fatigued material (A: Example 1). For comparison with the pre-fatigued material, the undeformed material was subjected to the same time (2 × 10 5 s) A material was prepared by applying a constant stress load (stress amplitude 0 MPa, stress ratio 1) at the same maximum stress (1000 MPa). This is the planned load material (Comparative Example 1).

[0033] After pre-fatigue deformation, the specimens were re-polished and subjected to fatigue tests under stress control, a frequency of 50 Hz, and a stress ratio of 0.1 (uniaxial tension). As shown in FIG. 3, the pre-fatigued material (A: Example 1) was 10 7 The cycle fatigue limit (maximum stress) is 1025 MPa, which is the same as the undeformed 10 7 This is a value that is 350 MPa higher than the cycle fatigue limit (maximum stress). 7 The cycle fatigue limit (maximum stress) is 725 MPa, which is 10 times that of the undeformed material. 7 The cycle fatigue limit (maximum stress) was improved by only 50 MPa, which shows that pre-fatigue deformation was effective in improving the fatigue limit. As shown by the black solid line in Fig. 3, the pre-fatigued material (B: Example 2) was subjected to a pressure of 1050 MPa for 10 minutes, compared to the pre-fatigued material (A). 7 After cycle loading, the surface was polished and 7 The maximum stress increase per cycle (100 MPa) was repeated twice up to 1250 MPa. The pre-fatigued material (B) was subjected to a maximum stress of 1300 MPa for 10 minutes. 7 The specimen did not break even after fatigue deformation for 1,350 MPa. 6 It broke during the cycle. In contrast, the pre-fatigued material (C: Comparative Example 1) was subjected to a fatigue limit stress of 675 MPa for 10 minutes. 7 After the cycle loading, the surface was polished, and the maximum stress was increased by 100 MPa to 775 MPa for 10 minutes, as shown by the black dotted line in Figure 3. 7 After the cycle load, the surface was polished again. The pre-fatigued material (C) was subjected to a maximum stress of 825 MPa for 10 7 Although no fracture occurred even after fatigue deformation of 10 cycles, the maximum stress of 875 MPa was 9.2 × 10 6 It broke during the cycle.

[0034] Same σ max As can be seen from the fact that the fatigue life at σ is significantly longer for the pre-fatigued material (A) compared to the undeformed material, the fatigue crack initiation life is extended by pre-fatigued deformation, resulting in a larger σ max In other words, it was confirmed that pre-fatigue deformation not only improves the fatigue limit but also enables pre-fatigue deformation under more suitable conditions.

[0035] Number of cycles to crack initiation (N int ) is assumed to be half of the fatigue life (approximate dashed line in Figure 3, pre-fatigued material (A)). When the stress ratio is 0.1, the maximum number of pre-fatigued cycles (N pre ) is given by the following equations (6a) and (6b): σ B is 1619 MPa.

number

[0036] The maximum number of pre-fatigue cycles (N pre ) prediction formula was calculated experimentally based on the fatigue life curve for a stress ratio of 0.1, but the tensile strength (σ B ) and the fatigue limit load and corresponding maximum stress (σ W-max ) can be used to predict the upper limit of σ by the following equations (3a) and (3b). W-max (For example, σ W-max =0.99σ B ), tens to hundreds of cycles of fatigue deformation are required for crack initiation, and in most high-strength steels, the fatigue crack initiation rate is 2 × 10 5 If the specimen is not broken after 10 cycles of fatigue deformation, 7 This condition was empirically determined, taking into consideration that no breakage occurs after cycling.

number

[0037] Furthermore, if the fatigue limit at a certain stress ratio is known, the fatigue limit at any stress ratio can be calculated using the modified Goodman diagram according to the following equation (7): B is the tensile strength, σ W is the fatigue limit when the stress ratio is -1, σ amp and σ mid are the stress amplitude and mean stress at a given stress ratio, respectively.

number

[0038] An improvement in the fatigue limit was also confirmed for the pre-fatigued specimen (C), but the final fatigue limit (maximum stress) was 825 MPa, which was not as great an improvement as that observed for the pre-fatigued specimens (A and B). Therefore, it is considered desirable to perform pre-fatigue deformation at as high a stress as possible. When the surfaces of the test pieces after fatigue testing were observed using an optical microscope and a scanning electron microscope, no cracks were observed on the surface of any of the unfractured specimens. Therefore, it is considered that the fatigue limit of the as-quenched martensitic steel in this invention corresponds to the crack initiation limit, and that the improvement in the fatigue limit was due to the improvement in the crack initiation limit caused by pre-fatigue deformation.

[0039] Figure 4 is a graph showing the correlation between tensile strength and fatigue limit. The results for the example (stress ratio 0.1) are shown converted to a stress ratio equivalent to 0 using a modified Goodman diagram. While the fatigue limit of conventional steel materials (quoted from a NIMS data sheet held by the applicant) peaks out at a tensile strength of about 1.4 GPa, the pre-fatigued material (B) of this example achieves a fatigue limit 0.43 times the tensile strength at tensile strengths exceeding 1.5 GPa. By improving the crack initiation limit using pre-fatigue deformation, the upper limit of the fatigue limit for high tensile strength was successfully achieved.

[0040] <Verification of the number of cycles of pre-fatigue deformation> The undeformed material was subjected to a pre-fatigue deformation of 1000 MPa maximum stress, 50 MPa stress amplitude (equivalent to a stress ratio of 0.9), and 50 Hz frequency. 5 This specimen was further subjected to pre-fatigue deformation at a maximum stress of 1000 MPa, a stress ratio of 0.1, and a frequency of 50 Hz, and then subjected to a 10 5 The maximum stress was increased by 100 MPa for each cycle, and continued up to 1300 MPa. After this pre-fatigue deformation, the sample surface was re-polished again, and this was designated as the pre-fatigued material (D: Example 3). The undeformed material was subjected to a pre-fatigue deformation of 1000 MPa maximum stress, 50 MPa stress amplitude (equivalent to a stress ratio of 0.9), and 50 Hz frequency. 3 This specimen was further subjected to pre-fatigue deformation at a maximum stress of 1000 MPa, a stress ratio of 0.1, and a frequency of 50 Hz, and then subjected to a 10 3The maximum stress was increased by 100 MPa for each cycle up to 1400 MPa. Finally, the maximum stress was set to 1450 MPa, and a pre-fatigue deformation was performed at a stress ratio of 0.1 and a frequency of 50 Hz for 10 cycles. 3 After this pre-fatigue deformation, the surface of the sample was polished again, and the result was used as a pre-fatigue material (E: Example 4).

[0041] According to equation (4), when the stress ratio is 0.1, N pre is 10 5 Or 10 3 When cycle, σ max If the stresses are not below 1367 MPa and 1468 MPa, respectively, there is a high possibility that cracks will occur during pre-fatigue deformation. max is increased stepwise within that range. According to equations (5a) and (5b), for an undeformed material, when the stress ratio is 0.1, σ max When is 1000 MPa, N pre If the fatigue life is not set to 6680 cycles or less, there is a high possibility of cracks occurring. 5 It is not possible to perform pre-fatigue deformation of the cycle immediately. On the other hand, in the above process, pre-fatigue deformation is first performed at a stress ratio of 0.9 to forge a material equivalent to pre-fatigued material (A). Therefore, according to equations (6a) and (6b), when the stress ratio is 0.1, σ max When the pressure is 1000 MPa, 10 5 Pre-fatigue deformation of the cycle is possible. Furthermore, according to equations (6a) and (6b), when the stress ratio is 0.1, σ max When is 1100 MPa, N pre In fact, in the pre-fatigue material (A), the stress ratio is 0.1, σ max The fatigue life is 56,765 cycles when the stress is 1100 MPa. However, the fatigue life is 56,765 cycles when the stress is 1100 MPa. max When increasing the stress ratio to 0.1, σ max is 10 at 1100 MPa 5 No cracks occur even after pre-fatigue deformation for 10 cycles. This is because the stress ratio is 0.1 and σ maxis 1000MPa, 10 5 N by pre-fatigue deformation of cycle pre This is thought to be due to the fact that the upper limit of σ has been further improved from equations (6a) and (6b). max This demonstrates the effectiveness of increasing

[0042] Fatigue tests were conducted using flat, smooth specimens under stress control, with a frequency of 50 Hz and a stress ratio of 0.1 (uniaxial tension). Both pre-fatigued specimens (D) and pre-fatigued specimens (E) were subjected to a maximum stress of 1300 MPa for 10 minutes. 7 Therefore, the effect of pre-fatigue deformation on improving the fatigue limit was 3 It was demonstrated that 10 cycles was sufficient for the pre-fatigue test. 3 The cycle load was applied seven times with increasing stress levels, for a total of 7 x 10 3 The pre-fatigue deformation was 100 cycles (approximately 2 min at 50 Hz). The results are summarized in Figure 4 and Table 2. The pre-fatigued material (D) and the pre-fatigued material (E) achieved fatigue limits (equivalent to a stress ratio of 0) of 0.41 and 0.40 times the tensile strength, respectively. [Table 3]

[0043] <Verification of maximum stress of pre-fatigue deformation> The undeformed material was subjected to a pre-fatigue deformation of 1500 MPa maximum stress, 50 MPa stress amplitude (equivalent to a stress ratio of 0.93), and 50 Hz frequency for 10 minutes. 7 This was designated as the pre-fatigued material (F: Example 5). The pre-fatigued material (E) was re-ground and subjected to fatigue testing under stress control, a frequency of 50 Hz, and a stress ratio of 0.1 (uniaxial tension). The fatigue limit (maximum stress) was 1050 MPa, confirming an improvement in the fatigue limit from the undeformed material, but this fatigue limit was not significantly different from that of the pre-fatigued material (A). Therefore, the effect of pre-fatigued deformation was σ max It is clear that it is not necessarily the case that only a large σ is good. Considering the results of the pre-fatigued material (D) and the pre-fatigued material (E), a larger σmax Furthermore, it is believed that the effect of pre-fatigue deformation in improving the fatigue limit is maximized by using a larger stress amplitude (smaller stress ratio).

[0044] Generally, metallic materials deform unevenly according to their microstructure, and cracks occur when deformation concentrates in areas prone to local deformation. Therefore, suppressing localized deformation concentration is expected to improve the fatigue limit. It has been revealed that in as-quenched martensitic steel, fatigue cracks initiate from intrusions and extrusions caused by strain mismatch at large-angle boundaries (see Non-Patent Document 4). Therefore, to improve the crack initiation limit of as-quenched martensitic steel, it is necessary to suppress the fatigue crack initiation mechanism at large-angle boundaries.

[0045] Micro Vickers hardness tests were conducted at 1,681 points and nano indentation tests were conducted at 1,200 points on each sample: undeformed material, scheduled load material, pre-fatigued material (A), and pre-fatigued material (B). Micro Vickers hardness tests conformed to the provisions of JIS Z 2244 (Vickers hardness test: ISO 6507-1). Nano indentation tests conformed to the provisions of JIS Z 2255 (ultra-micro load hardness test method: ISO 14577). The micro-Vickers test was performed with a test load of 490.3 mN and a loading time of 10 s. The nano-indentation test was performed using a Berkovich indenter with a loading / unloading rate of 50 μN / s and a maximum load of 1000 μN (held for 20 s). The results are summarized in Table 3 and Figure 5. As shown in Figure 5(a), the fatigue limit and the average micro-Vickers hardness (Hv ave ) was not found to have a clear correlation with the standard deviation of the micro Vickers hardness distribution (Hv dev ) became smaller, the fatigue limit became higher (Fig. 5(b)).

[0046] The coefficient of variation (Hv) of the micro Vickers hardness of the pre-fatigued material (A) and the pre-fatigued material (B) with significantly improved fatigue limits dev / Hv ave) were reduced by 34.9% and 47.1%, respectively, from the undeformed material. In addition, after the nanoindentation test, crystal orientation analysis was performed on the measurement area using electron backscatter diffraction with a scanning electron microscope to examine the correspondence between the indentation and the microstructure. The nanohardness of the measurement point where the indentation overlapped even slightly with a high-angle boundary with a crystal orientation difference of 15 degrees or more was evaluated as the nanohardness of the high-angle boundary. The number of indentations that overlapped with high-angle boundaries varied slightly depending on the sample, but was between 380 and 420. As shown in Figure 5(c), the average nanohardness of the high-angle boundary (Hn GB ) increased, the fatigue limit increased. The ratio of the average nanohardness to the average Vickers hardness of the large-angle boundary (Hn GB / Hv ave ) increased by 6.0% and 11.3%, respectively, from the undeformed material in the pre-fatigued material (A) and pre-fatigued material (B), where the fatigue limit was significantly improved. Therefore, it is thought that crack initiation was suppressed by the fact that the pre-fatigued deformation made the hardness distribution uniform, suppressing the concentration of local stress and local strain on the material surface, and also by the improvement of the nanohardness at the large-angle boundary, which is the fracture initiation point.

[0047] Table 3 lists the coefficient of variation in the micro Vickers hardness test for each sample: undeformed material, pre-loaded material, pre-fatigued material (A), and pre-fatigued material (B), as well as the calculated ratio of the average nano hardness value of high-angle boundaries with a crystal orientation difference of 15 degrees or more measured by nano indentation testing to the average measured value in the micro Vickers hardness test. That is, based on the experimental value of the pre-fatigued material (A) that was subjected to a single pre-fatigued deformation treatment on the undeformed material, the coefficient of variation in the micro Vickers hardness test for the martensitic steel that was subjected to a single pre-fatigued deformation treatment according to the conditions of the present invention was 3.2 × 10 -2 The ratio of the average nano hardness of the large angle boundary with a crystal orientation difference of 15 degrees or more measured by nanoindentation testing to the average measured value of the micro Vickers hardness test is 1.55 x 10 -2 This can be said to be the above (criteria for achieving the first objective). Furthermore, based on the experimental value of the pre-fatigue deformation material (B) in which the undeformed material was subjected to pre-fatigue deformation treatment twice or more, and the fatigue limit was 641 MPa, which corresponds to approximately 0.42 times the tensile strength (stress ratio 0), the coefficient of variation in the micro-Vickers hardness test for martensitic steel that was pre-fatigue deformed twice or more according to the conditions of the present invention was 2.5 × 10 -2 The ratio of the average nano hardness of the large angle boundary with a crystal orientation difference of 15 degrees or more measured by nanoindentation testing to the average measured value of the micro Vickers hardness test is 1.62 x 10 -2 This can be said to be the above (criteria for achieving the second objective).

[0048] The present invention is not limited to the above examples, and various parameters may be selected as appropriate within the scope obvious to those skilled in the art. For example, the test load, loading time, and number of measurement points for the micro Vickers hardness test may be selected as appropriate in accordance with JIS Z 2244. Furthermore, the type of indenter used in the nanoindentation test, the loading / unloading rate, maximum load, and maximum load holding time may be selected as appropriate in accordance with JIS Z 2255. Furthermore, the numerical range of pre-fatigue deformation for heat-treated martensitic steel that has been austenitized and then quenched may be selected arbitrarily as long as the stress ratio is within the range of -1 or more and 0.99 or less, and the number of cycles per unit time for pre-fatigue deformation may be selected arbitrarily as long as it is within the range of 0.6 CPM or more and 18,000 CPM (cycles per minute) or less (0.01 Hz or more and 300 Hz or less). In the above examples, the maximum stress value in each pre-fatigue deformation, from the first pre-fatigue deformation and the second (nth) pre-fatigue deformation onward, is a constant value that is increased by a predetermined value from the maximum stress in the most recent (n-1th) pre-fatigue deformation. However, the present invention is not limited to this. For example, the maximum stress value applied in each pre-fatigue deformation, from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value, may be varied. The maximum stress value in the current (nth) pre-fatigue deformation may be gradually increased as the number of pre-fatigue deformation cycles increases, for example, from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value. [Industrial Applicability]

[0049] As described above in detail, according to the martensitic steel of the present invention as set forth in claim 1 and the manufacturing method thereof as set forth in claim 6 and following claims, a single pre-fatigue deformation according to the conditions of the present invention improves the fatigue limit not only to the crack non-propagation limit but also to the crack initiation limit. Thus, by performing a single pre-fatigue deformation according to the conditions of the present invention, it is possible to provide a material having a higher fatigue limit than an untreated material. According to the martensitic steel of the present invention as set forth in claim 2 and the manufacturing method thereof as set forth in claim 11 and following, by performing a second or subsequent (nth (n=2, 3, 4, etc.)) pre-fatigue deformation in addition to a first pre-fatigue deformation according to the conditions of the present invention, a martensitic steel having a fatigue limit of about 0.42 times the tensile strength (stress ratio 0) can be obtained, thereby achieving both a high fatigue limit and high strength. Therefore, martensitic steel that has been subjected to a second or subsequent (nth (n=2, 3, 4, etc.)) pre-fatigue deformation in addition to a first pre-fatigue deformation according to the conditions of the present invention is suitable as a structural material for transportation vehicles, etc.

Claims

1. Contains C: 0.15 to 0.6 mass%, Mn: 0.2 to 5 mass%, and the balance being Fe and unavoidable impurities; A martensitic steel having a tensile strength of 1.5 GPa or more obtained by austenitizing a steel material of the above composition and then quenching it, The coefficient of variation in the micro Vickers hardness test is 3.2 x 10 －２ is as follows: The ratio of the average nanohardness of the large angle boundary having a crystal orientation misorientation of 15 degrees or more measured by the nanoindentation test to the average measured value of the micro Vickers hardness test is 1.55 × 10 －２ That's all. Martensitic steel that combines high fatigue limit and high strength.

2. The coefficient of variation in the micro Vickers hardness test is 2.5 x 10 －２ is as follows: The ratio of the average nanohardness of the large angle boundary having a crystal orientation misorientation of 15 degrees or more measured by the nanoindentation test to the average measured value of the micro Vickers hardness test is 1.62 × 10 －２ That's all. The martensitic steel according to claim 1, which has both a high fatigue limit and high strength.

3. 2. The martensitic steel according to claim 1, wherein the micro Vickers hardness test conforms to the provisions of JIS Z 2244 (Vickers hardness test), and the measured values ​​of the micro Vickers hardness test are obtained by measuring 1,000 or more points at a test load of 490 mN.

4. 2. The martensitic steel according to claim 1, wherein the nanoindentation test conforms to the provisions of JIS Z 2255 (ultra-micro load hardness test method), and the measurement values ​​of the nanoindentation test are obtained using a Berkovich indenter.

5. The martensitic steel according to claim 3, wherein the average nano-hardness of the high-angle boundaries having a crystal orientation difference of 15 degrees or more is an average value of measurements taken at 300 or more points under a maximum load of 1000 μN.

6. Contains C: 0.15 to 0.6 mass%, Mn: 0.2 to 5 mass%, and the balance being Fe and unavoidable impurities; A martensitic steel having a tensile strength of 1.5 GPa or more obtained by austenitizing a steel material of the above composition and then quenching it, For the martensitic steel material after heat treatment quenched after austenitization, The lower limit of the number of cycles of pre-fatigue deformation is set to 10 １ The upper limit is the maximum stress (σ) applied in the pre-fatigue deformation defined by the following equation 1: ｍａｘ Number of cycles for crack initiation (N ｐｒｅ ) less than or 10 ７ cycles, whichever is smaller: [Equation 1] (N ｐｒｅ : Maximum number of pre-fatigue cycles, σ Ｂ : tensile strength [MPa], σ ｍａｘ : Maximum stress of pre-fatigue deformation [MPa], σ Ｗ－ｍａｘ : Maximum stress [MPa] corresponding to the fatigue limit load at the stress ratio in the pre-fatigue deformation The maximum stress of the pre-fatigue deformation is set to 1 MPa or more and 99% or less of the tensile strength, and the stress ratio is set to -1 or more and 0.99 or less. How martensitic steel is manufactured.

7. The lower limit of the number of cycles of the pre-fatigue deformation is 10 ２ Cycles or more, The upper and lower limits of the maximum stress of the pre-fatigue deformation are set to be equal to or greater than the elastic limit and equal to or less than 98% of the tensile strength, The stress ratio of the pre-fatigue deformation is 0 or more and 0.97 or less. The method for producing a martensitic steel according to claim 6.

8. The lower limit of the number of cycles of the pre-fatigue deformation is 10 ３ Cycles or more, The upper and lower limits of the maximum stress of the pre-fatigue deformation are set to 70% or more of the yield stress or 0.2% proof stress and 97% or less of the tensile strength, The stress ratio of the pre-fatigue deformation is 0 or more and 0.2 or less. The method for producing a martensitic steel according to claim 7.

9. Furthermore, after the completion of the pre-fatigue deformation, grinding is performed to remove the irregularities on the material surface introduced by the first pre-fatigue deformation. The method for producing a martensitic steel according to claim 6.

10. The grinding removes a range of 1 μm to 100 μm from the surface of the martensitic steel material. The method for producing a martensitic steel according to claim 9.

11. The pre-fatigue deformation according to claim 6 is performed for the first time, Furthermore, the maximum stress in the second or subsequent (nth (n=2, 3, 4, ...)) pre-fatigue deformation is increased by a predetermined value from the maximum stress in the most recent (n-1th) pre-fatigue deformation, the maximum stress after the increase satisfies the maximum stress given by equation (2); [Equation 2] The lower limit of the number of cycles of pre-fatigue deformation for each of the second and subsequent cycles is 10 １ The upper limit is the maximum stress (σ ｍａｘ Number of cycles for crack initiation (N ｐｒｅ ) and satisfying the conditions of the formulas (1a) and (1b) of claim 6, ｍａｘ ) The second and subsequent pre-fatigue deformations are performed until How martensitic steel is manufactured.

12. The second and subsequent (n-th) pre-fatigue deformations are repeated up to a maximum stress within a range in which no cracks are generated in the most recent (n-1) pre-fatigue deformation and no cracks are generated in the n-th pre-fatigue deformation. The method for producing a martensitic steel according to claim 11.

13. In the first and each subsequent (nth) pre-fatigue deformation, the maximum value of the stress applied in each pre-fatigue deformation is constant relative to the initial maximum stress in the current (nth) pre-fatigue deformation. The method for producing a martensitic steel according to claim 11.

14. In the first and each subsequent (nth) pre-fatigue deformation, the maximum value of the stress applied in each pre-fatigue deformation is varied within a range from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value. The method for producing a martensitic steel according to claim 11.

15. The maximum value of the maximum stress in the current (nth) pre-fatigue deformation is gradually increased as the number of cycles of the current pre-fatigue deformation increases, within a range from the initial maximum stress in the current (nth) pre-fatigue deformation to a value that increases the maximum stress in the next (n+1th) pre-fatigue deformation by a predetermined value. The method for producing a martensitic steel according to claim 14.

16. The value by which the maximum stress in the previous pre-fatigue deformation is increased by a predetermined value is 200 MPa or less. The method for producing a martensitic steel according to claim 11.

17. The value by which the maximum stress in the previous pre-fatigue deformation is increased by a predetermined value is set to 100 MPa or less. The method for producing a martensitic steel according to claim 16.

18. Furthermore, after the second or subsequent (nth) pre-fatigue deformations are completed, grinding is performed to remove irregularities on the material surface introduced by the pre-fatigue deformations up to the nth. The method for producing a martensitic steel according to claim 11.

19. The grinding is performed after subjecting the martensitic steel to pre-fatigue deformation a number of times that gives the martensitic steel a fatigue limit of 0.40 times or more of its tensile strength. The method for producing a martensitic steel according to claim 18.

20. The grinding removes a range of 1 μm to 100 μm from the surface of the martensitic steel material. The method for producing a martensitic steel according to claim 19.

21. The frequency of the pre-fatigue deformation is 0.01 Hz or more and 300 Hz or less. A method for producing a martensitic steel according to any one of claims 6 to 11.

Images