JP2018075580A - Manufacturing method of brazing joint and brazing joint - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a manufacturing method which is suitable for improvement of sulfuric acid resistance of a brazing joint which is prepared by joining metal members with a brazing foil containing B, and to provide the brazing joint using the manufacturing method.SOLUTION: When metal members made of austenite type stainless steel are joined with each other by using a brazing foil of Ni-based or Fe-based amorphous alloy containing B of 0.30 to 6.0 mass%, a retention temperature and a retention time of brazing heat treatment are selected in such a range that J value obtained by J=∫(T(t).1/(t-t0))dt in an interval from t0 to t1 satisfies 1/J≤1.0×10(K.log(s))when T(t) is a temperature of a metal member after t times from starting of heating, Ts is solid phase line temperature of the brazing foil, t0 is a time when T(t) reaches Ts and t1 is a time when T(t) again reaches Ts after exceeding Ts. In the braze joint manufactured according to the method, λ value related to a distribution figure and a size of Cr type boride on a diffusion region satisfies λ≥10 μm.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a brazing joint manufacturing method and a brazing joint, for example, a brazing joint manufacturing method in which two or more metal members made of austenitic stainless steel are joined using a brazing foil containing boron (B), The present invention also relates to a brazed joint in which two or more metal members made of austenitic stainless steel are joined via a brazing layer containing B.

  For example, in a heat exchanger (EGR cooler) of an exhaust gas recirculation system (exhaust gas recirculation system) of an automobile or the like, a metal member (joined material) made of austenitic stainless steel is joined using a brazing material ( A brazed joint formed by brazing is often used. As a brazing material used for such a brazed joint, a Ni-based brazing material is often used. In such a Ni-based brazing member, B may be added to Ni as the main component for the purpose of reducing the brazing temperature by lowering the melting point. For example, BNi-2 standardized by JIS and AWS is a typical Ni-based brazing material containing B of 2.75% by mass or more and 3.50% by mass or less.

  In addition to the above-mentioned BNi-2, for example, Patent Document 1 discloses Ni (nickel) containing 1.1 mass% or more and 1.3 mass% or less of B which is useful for brazing a metal member made of stainless steel. ) -Cr (chromium) -Si (silicon) -B-based foil-like brazing material (brazing foil) is disclosed, and Patent Document 2 discloses Ni containing 0.1 mass% or more and 3.8 mass% or less of B. -Cr-Si-B-P-based foil-like brazing material (brazing foil) is disclosed, and Patent Document 3 discloses Ni-Cr-Si-B- containing 0 mass% or more and 5.4 mass% or less of B. P-Fe (iron) -based foil-like brazing material (brazing foil) is disclosed, and Patent Document 4 includes Ni-Cr-Si- containing 0.8 mass% or more and 1.1 mass% or less of B. A BP-Fe-Mo (molybdenum) -Cu (copper) -based foil-like brazing material (brazing foil) is disclosed.

U.S. Pat. No. 4,543,135 Special table 2011-501700 gazette US Pat. No. 4,302,515 Special table 2009-545451

  A brazed joint formed by joining a metal member (bonded material) made of austenitic stainless steel with good corrosion resistance using a brazing material (brazing foil) containing B is formed by melting and solidifying the brazing material. The metal member has a diffusion region containing a Cr-based boride adjacent to the brazing layer. This is because the B contained in the brazing material diffuses into the interior of the metal member due to heating during brazing and a diffusion region is formed, and the diffused B is combined with Cr contained in the metal member to form a Cr-based boride. By being formed. When such a Cr-based boride is included, the surrounding Cr is consumed and a Cr-deficient region is formed. Even austenitic stainless steel with good corrosion resistance is corroded when exposed to, for example, a sulfuric acid atmosphere, but for example, the sulfuric acid corrosion resistance of a brazed joint having a structure morphology in which such Cr-deficient regions are continuously distributed from the surface to the inside is In particular, it decreases.

  The present invention aims to improve sulfuric acid corrosion resistance of a brazed joint in which a metal member made of austenitic stainless steel is joined using a brazing foil containing B, and a method for producing a brazed joint suitable for the purpose, and A brazed joint obtained by using the method is provided.

  The present inventor paid attention to the distribution form of the Cr-based boride in the diffusion region in the process of examining the solution to the above-described problem. And by controlling the distribution of the Cr-based boride in the diffusion region adjacent to the brazing layer in the diffusion region containing the Cr-based boride adjacent to the brazing layer of the metal member to a specific form, The inventors have found that the sulfuric acid corrosion resistance of the brazed joint subjected to such control is improved, and reached the present invention.

That is, in the method for producing a brazed joint of the present invention, the first metal member made of austenitic stainless steel and the second metal member made of austenitic stainless steel are 0.30% by mass or more 6 A method for manufacturing a brazed joint formed by joining using a brazing foil made of a Ni-based amorphous alloy or Fe-based amorphous alloy containing boron (B) of 0.0 mass% or less, wherein the brazing foil is heated and melted A heat treatment step including a heating step for brazing and a cooling step for cooling and solidifying the molten wax, and the heating temperature and holding time in the heat treatment step are set to T (t), and the heating of the brazing foil is started. T is the solidus temperature of the brazing foil, t0 is the time when T (t) reaches Ts, and t1 is greater than Ts. After And the time when reaching Ts, the J value obtained by J = ∫ [T (t) · 1 / (t−t0)] dt in the interval from t0 to t1 is 1 / J ≦ 1.0 × 10 ⁻⁴ [K · log (seconds)] This is a method of selecting within a range satisfying ⁻¹ .

  A brazed joint excellent in sulfuric acid corrosion resistance can be obtained by using the above-described method for producing a brazed joint of the present invention. That is, in the brazed joint of the present invention, the first metal member made of austenitic stainless steel and the second metal member made of austenitic stainless steel are interposed via a brazing layer containing boron (B). And in a diffusion region including a Cr-based boride adjacent to the brazing layer of the metal member, the brazing joint being in a direction substantially parallel to the brazing layer, A region having a predetermined width is set as a measurement range so as to include a brazing layer adjacent diffusion region in which the Cr-based boride adjacent to the brazing layer has a higher concentration with the direction to be the longitudinal direction, and L is The length of the line segment LSj (j = 1 to m, m is an integer of 1 or more), which is a line segment connecting the base point and the end point in the longitudinal direction of the measurement range, and is set with uniformity in the measurement range, Li The length of the i-th Cr-based boride on the line segment LDj from the base point on the first line segment LSj, and n (an integer of 1 or more) is the number of Cr-based borides counted on the j-th line segment LSj. In this case, the λ value that is an average value of j λj values obtained by λj = (L−ΣLi) / n satisfies λ ≧ 10 μm.

  ADVANTAGE OF THE INVENTION According to this invention, the sulfuric acid corrosion resistance of the brazing joint in which the metal member using austenitic stainless steel is joined using the brazing foil containing B can be improved.

It is a figure which shows an example of the heat processing test data for calculating | requiring J value (brazing parameter) of a brazing joint. It is a figure which shows the relationship between (lambda) value which concerns on the distribution form of Cr type boride in the spreading | diffusion area | region of a brazing joint, and the maximum erosion depth D value by sulfuric acid corrosion of a brazing joint. It is a figure which shows the relationship between 1 / J value which concerns on selection of the heat processing conditions (holding temperature, holding time) which braze, and the maximum erosion depth D value by sulfuric acid corrosion of a brazing joint. It is a figure which shows typically the intensity | strength distribution of Cr by EPMA of the part containing the brazing layer adjacent diffusion area | region where the Cr-type boride which adjoins the brazing layer of a brazing joint tends to become a high concentration. It is the intensity distribution (200 times) which performed ternary emphasis based on the EPMA measurement result of the test body A of Inventive Example 1, wherein (a) shows Cr and (b) shows B. It is the intensity distribution (1800 times) which performed ternary emphasis based on the EPMA measurement result of the test body A of Invention Example 1, wherein (a) shows Cr and (b) shows B. It is the intensity distribution (200 times) which performed ternary emphasis based on the EPMA measurement result of the test body B of Example 2 of the present invention, where (a) shows Cr and (b) shows B. It is the intensity distribution (1800 times) which performed the ternary emphasis based on the EPMA measurement result of the test body B of Invention Example 2, wherein (a) shows Cr and (b) shows B. It is the intensity distribution (200 times) which performed ternary emphasis based on the EPMA measurement result of the test body C of a comparative example, (a) is Cr, (b) is a figure which shows B. It is the intensity distribution (1800 times) which performed ternary emphasis based on the EPMA measurement result of the test body C of a comparative example, (a) is Cr, (b) is a figure which shows B. (A) is a trial example of a brazed joint, (b) is a figure which shows the prototype of the test body used for the sulfuric acid corrosion test.

  In the method for producing a brazed joint of the present invention, a first metal member made of austenitic stainless steel and a second metal member made of austenitic stainless steel are 0.30% by mass or more and 6.0. This is a method of joining by using a brazing foil made of a Ni-based amorphous alloy or Fe-based amorphous alloy containing B (boron) in an amount of mass% or less.

  The austenitic stainless steel to be joined has higher corrosion resistance than the ferritic stainless steel frequently used as the joined material for the brazed joint. Moreover, the brazing foil made of Ni-based amorphous alloy or Fe-based amorphous alloy containing B of 0.30 mass% or more and 6.0 mass% or less is a manufacturing method in which a molten metal is injected onto the surface of a cooled roll and cooled at high speed. It is easy to form continuously by a general ultra-quenching method called single roll method, liquid quenching method, melt span method, melt spinning method, etc. It becomes the foil which has.

  Ni and Fe are cheaper than Co and are generally used as the base of amorphous alloys. A case where Ni is contained in an amount of 50% by mass or more is referred to as a Ni group, and a case where Ni is contained in an amount of 50% by mass or more is referred to as an Fe group. . If a brazing foil made of an amorphous alloy can be formed even if B is less than 0.30% by mass, there is a possibility that a brazed joint with good sulfuric acid corrosion resistance can be obtained by applying the present invention. If B exceeds 6.0% by mass, the formation of excess Cr-based borides is predicted, so that the sulfuric acid corrosion resistance of the brazed joint may be deteriorated. Further, for example, a brazing foil having a thickness of 10 μm or more and 30 μm or less is easy to handle and can be easily processed into a shape suitable for the bonding surface of the material to be bonded by, for example, punching or bending. . In general, when the brazing foil containing B does not contain B in the range of 0.30 mass% or more and 6.0 mass% or less, the formation of the amorphal structure is affected, and thus the brittle foil tends to become brittle. Formation by the above-mentioned general super rapid cooling method becomes difficult.

In the method for producing a brazed joint of the present invention, brazing is performed by a heat treatment step including a heating step for heating and brazing the brazing foil and a cooling step for cooling and solidifying the brazing brazing as in the conventional method. . The holding temperature and holding time in this heat treatment step have been selected empirically in the past, but in the present invention, they are selected according to a predetermined criterion. Specifically, the holding temperature and holding time in the heat treatment step, T (t) is the temperature of the metal member after t hours from the start of the heating of the brazing foil, Ts is the solidus temperature of the brazing foil, When t0 is the time when T (t) reaches Ts and t1 is the time when T (t) reaches Ts again after T (t) exceeds Ts, in the interval from t0 to t1, J = ∫ [ T (t) · 1 / (t−t0)] The J value obtained by dt is selected within a range satisfying 1 / J ≦ 1.0 × 10 ⁻⁴ [K · log (seconds)] ^{− 1} . In calculating the J value using the actual measurement value described later, if it can be determined that the holding time is sufficiently long and there is no substantial influence of the time required for temperature increase and decrease, in the above formula for calculating the J value, The heat pattern between t0 and t1 can be approximated to a rectangular line (rectangular pattern), and the calculation can be simplified with t0 to t1 as the integration range. In addition, since [log (second)] is dimensionless, the unit of the J value is [K] (Kelvin).

An important feature of the present invention is that when selecting the heat treatment conditions (holding temperature, holding time) for brazing, a J value (brazing parameter) is introduced, and the inverse 1 / J value is adopted as a criterion. That is. By selecting the brazing heat treatment conditions (holding temperature, holding time) within a range satisfying 1 / J ≦ 1.0 × 10 ⁻⁴ [K · log (second)] ⁻¹ , the material to be joined is obtained. In the diffusion region including the Cr-based boride adjacent to the brazing layer of the metal member, particularly in the brazing layer adjacent diffusion region in which the Cr-based boride adjacent to the brazing layer tends to have a higher concentration, the distribution pattern of the Cr-based boride is Optimized. Specifically, the distribution pattern of the Cr-based boride in the brazing layer adjacent diffusion region of the brazing joint is represented by j λj values obtained by λj = (L−ΣLi) / n on the j-th line segment LSj described later. The average value λ is formed so as to satisfy λ ≧ 10 μm.

  As described above, when a metal member made of austenitic stainless steel is brazed using a brazing foil containing B, B diffuses into the metal member due to melting of the brazing foil. At this time, in a high-B brazing foil (for example, a brazing foil containing 4.0% by mass to 6.0% by mass of B), the behavior is such that B is equalized to a predetermined concentration inside the metal member, The diffusion region of B tends to be wider. Therefore, just because it is a high-B brazing foil, B does not always remain in the brazing layer adjacent diffusion region of the brazing joint and the concentration of the Cr boride does not increase. Considering such characteristics, by appropriately controlling the heat treatment conditions for brazing as described above, it is possible to optimize the distribution form of the Cr-based boride in the brazing layer adjacent diffusion region of the brazing joint. Further, in a low B brazing foil (for example, a brazing foil containing 0.30% by mass to 2.0% by mass of B), B is equalized to a predetermined concentration inside the metal member as in the high B brazing foil. However, the diffusion region of B tends to be small due to the small amount of B that can be supplied. Even in this low-B brazing foil, the distribution form of the Cr-based boride in the brazing layer adjacent diffusion region of the brazing joint can be optimized by appropriately controlling the heat treatment conditions for brazing.

  From such a viewpoint, for example, in a brazing foil containing 2.0% by mass to 4.0% by mass of B, the behavior of B diffusing into the metal member is unstable due to the strong influence of the heat treatment conditions of brazing. It seems to be easy to become. Therefore, for example, in a brazing foil containing 2.0% by mass to 4.0% by mass of B, it is considered that the distribution form of Cr boride strongly influenced by the diffusion of B is highly sensitive to the heat treatment conditions of brazing. . From such a viewpoint, when 2.0% by mass to 4.0% by mass of B is contained in the brazing foil, for example, it reacts sensitively under the influence of the heat treatment conditions of brazing, and enters the inside of the metal member. It is considered that the behavior of diffusing B tends to be unstable.

In the present invention, the J value (brazing parameter) is a value that serves as an index for selecting the holding temperature and holding time in heat treatment important for obtaining a brazed joint with good sulfuric acid corrosion resistance. In the present invention, the 1 / J value that is the reciprocal of the J value is adopted as a criterion (threshold) for selecting the holding temperature and holding time, and 1 / J ≦ 1.0 × 10 ⁻⁴ [K · log. (Seconds)] Select within a range satisfying ^-1 . The reason why the 1 / J value is selected within the range satisfying the above is that, in a brazed joint having a 1 / J value exceeding 1.0 × 10 ⁻⁴ [K · log (second)] ⁻¹ , Cr-based boron in the diffusion region is selected. It was confirmed that the λ value related to the distribution form of the chemical compound was reduced to less than 10 μm, and it was confirmed that the maximum erosion depth due to sulfuric acid corrosion of the brazed joint exceeded 0.10 mm due to the decrease in the λ value. is there. This point will be described in detail later. The reason why the 1 / J value is used is that, as is the case with the reciprocal of the absolute temperature, linearity with respect to heat input is maintained, so that it is simpler than using the J value.

Hereinafter, the present invention will be described in detail including a method for determining the holding temperature and holding time in the heat treatment.
In the present invention, the 1 / J value is set, for example, in the range of Ts (solidus temperature) + 50 ° C. to 150 ° C. exceeding the melting point of the brazing foil used for brazing, and the holding time at that time is sufficient. For example, a heat treatment test with a heat pattern set in the range of 3 h to 5 h is performed, or the holding temperature is set in a range of Ts + 150 ° C. to 450 ° C., for example, which sufficiently exceeds the melting point of the wax foil, and the holding time at that time Can be obtained by a method of performing a heat treatment test with a heat pattern set in a range of 0.1 minutes to 60 minutes, for example, and obtaining a J value from the test data. The holding time suitable for the desired holding temperature can be determined using the 1 / J value obtained by this method as a criterion.

For example, in the heat treatment test, a brazing foil is sandwiched between two metal members to be bonded (test body) and a dummy material having the same size as the test body is placed in a heat treatment furnace and vacuum is applied. A desired heat pattern can be set in an atmosphere (about 2 × 10 ⁻³ Pa to about 5 × 10 ⁻³ Pa). In addition, you may substitute the actual temperature of a test body with the actual temperature of the dummy material measured, for example by attaching a sheath thermocouple. For example, when the melting point of the wax foil (in the present invention, the solidus temperature is adopted) is 1000 ° C., the holding temperature is set to, for example, 1280 ° C., and the heating rate and the cooling rate are set to, for example, 80 ° C./min. (About 1.3 ° C./second), for example, 270 minutes (16200 seconds) so that the holding time after raising the temperature to a predetermined holding temperature has a sufficient margin, and in the subsequent temperature lowering process, when it reaches about 400 ° C. It can be set to perform blast cooling with nitrogen gas.

  In the heat treatment test described above, the 1 / J value in the present invention records the temperature change of the dummy material in increments of 20 seconds, for example, in increments of 20 seconds, obtains the J value based on the test data, and reciprocates the J value. It can be determined by the conversion method. For example, FIG. 1 is an example of test data obtained for the J value based on the heat treatment test described above. The change pattern of the temperature T (t) ° C. of the dummy material shown in FIG. 1 starts the temperature rise (heating) when t = 0 seconds and T (0) = 20 ° C. (room temperature), and t = t 0 seconds. T (t0) = Ts (solidus temperature), and the temperature is further raised, and during t = t0 seconds + α0 to t1 seconds−α1, T (t) = T (t1) = Th (holding temperature) Then, the temperature is lowered, and T (t1) = Ts is reached again at t = t1 seconds, and then the temperature is lowered toward room temperature. When obtaining the J value, the holding time is significantly increased (about 20000 seconds in the pattern shown in FIG. 1), and the holding time T (t) = Th and again exceeds Ts (t0 seconds) and Ts ( The substantial influence of the difference from the required time to reach (t1 seconds) is suppressed, and the above α0 and α1 can be regarded as 0 seconds.

  Therefore, in obtaining the J value in the present invention, the change of the temperature T (t) between Ts ° C. to Th ° C. to Ts ° C. of the dummy material from t = t0 seconds to t1 seconds is represented by the integration interval from t0 seconds to t1. According to J = ∫T (t) × 1 / (t−t0) × dt [K · log (seconds)]. At this time, as shown in FIG. 1, by replacing with a rectangular pattern and simplifying, the J value can be easily obtained by J = (Th−Ts) / (t1−t0) [K · log (seconds)]. Can be sought.

In determining the 1 / J value threshold (1.0 × 10 ⁻⁴ [K · log (seconds) ⁻¹ ) related to the selection of the heat treatment conditions (holding temperature and holding time) for performing the brazing described above, austenite A first metal member made of a stainless steel and a second metal member made of an austenitic stainless steel are subjected to heat treatment (brazing joint) using a brazing material containing B, and several solders A joint (test body) was prepared, and the relationship between the λ value related to the distribution of Cr boride in the diffusion region of the brazed joint and the maximum erosion depth D value due to sulfuric acid corrosion of the brazed joint (see FIG. 2) In addition, the relationship between the 1 / J value and the D value determined based on the heat treatment (brazing joining) (see FIG. 3) was examined.

  When determining the 1 / J value, the first metal member and the second metal member used as the brazing joint material are made of SUS304, which is a typical austenitic stainless steel, which is easily available. A rectangular parallelepiped having a total length of about 25 mm, a vertical length of the joining surface of about 35 mm, and a horizontal length of about 10 mm was used. The joining surface of the metal member was subjected to acetone ultrasonic cleaning after paper polishing (# 800). In addition, the brazing material used for the brazing joint has an average thickness in the width direction of 26 μm to 42 μm produced by a melt spinning method, B is 0.02 mass% to 3.12 mass%, and the solidus temperature Ts is A rectangular sheet (wax foil) that was selected from Ni-based amorphous alloy ribbons having a composition of 877 ° C. to 992 ° C. and cut out from the ribbon to a size corresponding to the joint surface of the metal member was used. The sizes of the metal member and the brazing foil are examples, and are not limited thereto.

  Table 1 shows the chemical composition (mass%), the solidus temperature Ts (° C), the liquidus temperature Tl (° C), and the average thickness in the width direction of the brazing foil (Nos. 1 to 6) used for the trial production of the brazed joint. (Μm). In addition, Ts and Tl shown in Table 1 are graphs of the temperature values of the respective solidus lines and liquidus lines obtained by changing the rate of temperature rise in 3 steps of 5 ° C, 10 ° C, and 20 ° C per minute. The value corresponding to the intercept in the graph is obtained. Further, Ni shown in Table 1 includes impurities (trace elements) that are inevitably included. Further, Ts and Tl were measured in an argon gas atmosphere by DTA (Differential Thermal Analysis). In addition, wax foil No. 5 is out of the scope of the present invention with B being 0.02% by mass, but since it could be formed as a brazing foil made of an amorphous alloy, the effects of the present invention may be obtained. You may apply ideas.

When producing the brazed joint, a temporary assembly was carried out by sandwiching the brazing foil between the joint surface of the first metal member and the joint surface of the second metal member described above, thereby producing a temporary assembly of the braze joint. The temporary assembly of the brazed joint is placed in a heat treatment furnace (degreasing furnace), and degreasing is performed in an argon gas atmosphere at a heating rate of 80 ° C./min, a holding temperature of 500 ° C., and a holding time of 1 h. went. Note that the degreasing treatment may be performed according to the degree of cleanliness of the metal member or the brazing material, and is not necessarily performed. Subsequently, the temporary assembly of the brazed joint after the degreasing treatment was placed in a heat treatment furnace (brazing furnace), and heat treatment for brazing (brazing joining) was performed. The heat treatment is performed in a vacuum atmosphere (about 2 × 10 ⁻³ Pa to about 5 × 10 ⁻³ Pa), and Ts shown in Table 1 is 948 ° C. to 992 ° C. Therefore, the holding temperature is set to 1280 ° C. The cooling rate is 80 ° C./min, the holding time is 270 minutes (16200 seconds), and the subsequent cooling process is performed in a heat pattern (see FIG. 1) in which blast cooling is performed with nitrogen gas when the temperature reaches about 400 ° C. It was. As a result, the brazing length shown in FIG. 11A is about 50 mm (about 25 mm + the thickness of the brazing foil + about 25 mm), the brazing joint surface has a vertical length of about 35 mm, and a lateral length of about 10 mm. The joint 10 can be obtained.

  In the sulfuric acid corrosion test, the corrosive liquid was special grade sulfuric acid, the sulfuric acid concentration was 10% by mass, the liquid temperature was 60 ° C., and the immersion time was 3 hours. The specimen 20 used in the sulfuric acid corrosion test is a square having a thickness of about 1 mm and a length of about 10 mm as shown in FIG. 11B, and the brazing layer 3 is located in a stripe shape at the approximate center of the square. A flat plate having a juxtaposed structure in which the metal members 1 and 2 are located on both sides thereof. Further, as shown by a dotted line in FIG. 11 (a), the front and back surfaces of the flat plate (test body 20) follow the full length method of the brazed joint 10, and the thickness direction is a vertical length of about 35 mm of the brazed joint surface. It cut out from the substantially center part of the brazing joint 10 so that the direction might be followed. Moreover, the measurement of the maximum erosion depth (D value) of the test body 20 is performed on a test body after the sulfuric acid corrosion test (hereinafter referred to as “test body 21” in order to distinguish from the test body 20 before the sulfuric acid corrosion test). What was subjected to ultrasonic cleaning with water and dried at 80 ° C. for 2 hours was used.

  In addition, for observation and measurement of the above-described λ value and D value, EPMA (Electron Probe Microscope Analysis), SEM (Scanning Electron Microscope), laser microscope, and the like can be used. The specimen 20 before the sulfuric acid corrosion test was used for the observation and measurement of the Cr boride for obtaining the λ value. A method for obtaining the λ value will be described later. Further, the D value is determined by using a laser microscope, and the observation surface of the test body 21 (the narrow side surface in the plate thickness direction of the plate of the parallel structure described above) faces the reference surface of the laser microscope, and the test body 21. The brazing layer 3, the diffusion region 4 (see FIG. 4) including the λ measurement range 5 to be described later, and the metal members 1 and 2, for example, have a square shape of about 0.66 mm at a magnification of 20 times. A measurement range is set, and the surface of the specimen 21 having the smallest erosion amount in the measurement range is used as a reference for the depth (reference surface), or the corrosion amount of the metal members 1 and 2 and the brazing layer 3 is substantially equal. If it is negligibly small, the surfaces of the metal members 1 and 2 and the brazing layer 3 are used as reference depths (reference planes), and a plurality of profiles across the erosion part observed in the measurement range are measured. It can be obtained by a method in which the largest (deep) value is the D value. Note that the entire observation surface of the specimen 21 may be divided into a plurality of regions and observed, and the maximum depth of the plurality of divided regions may be set as the D value. Moreover, when the area | region containing the location where the erosion depth is the largest can be specified clearly, the maximum value of the erosion depth calculated | required by observing the measurement range specified clearly can also be made into D value.

  As shown in FIG. 2, the D value greatly fluctuates to about 0.03 mm or about 0.14 mm when the λ value is around 3 μm. In addition, it includes 3.12% by mass B indicated by a black circle in FIG. 2, which is almost the median value of the range of B (0.30% by mass to 6.0% by mass) of the wax foil that is the subject of the present invention. The D value seen on the regression line (the line segment shown in FIG. 2) when brazing foil is used is about 0.08 mm when the λ value is around 5 μm, and about 0.07 mm when the λ value is around 10 μm. Is about 0.05 mm around 45 μm. Therefore, if the tendency of the D value to decrease as the λ value increases, the characteristics of the approximate line, and the reliability and reproducibility of the measurement are taken into account, the λ value satisfies λ ≧ 10 μm (preferably 20 μm). It can be expected that a brazed joint having a value of 0.10 mm or less, further 0.05 μm or less is obtained.

From the viewpoint of the relationship between the λ value and the D value described above, the threshold value of the 1 / J value can be determined by considering the relationship between the 1 / J value and the λ value shown in FIG. In addition, in FIG. 3, the case where the brazing foil containing 3.12 mass% B is used is shown by the black circle similarly to FIG. 2, and the regression line is also written together. As shown in FIG. 3, the 1 / J value has a point where the λ value is less than 10 μm in the range exceeding 1.1 × 10 ⁻⁴ [K · log (seconds) ⁻¹ , 10 ⁻⁴ [K · log (seconds)] In the range of ⁻¹ or less, there is no point at which the λ value is less than 10 μm. In other words, in the range where the 1 / J value satisfies 1 / J ≦ 1.1 × 10 ⁻⁴ [K · log (seconds)] ^{− 1} , it can be seen that the λ value satisfies λ ≧ 10 μm. Further, it can be expected that the D value is 1.0 mm or less, further 0.05 μm or less.

Therefore, from the viewpoint of the relationship with the λ value, 1.1 × 10 ⁻⁴ [K · log (seconds)] ⁻¹ may be considered as a 1 / J value threshold, but the 1 / J value is large. Then, in consideration of the tendency that the λ value becomes smaller and the reliability and reproducibility of measurement, in the present invention, 1 × J value in which 1.0 × 10 ⁻⁴ [K · log (seconds)] ⁻¹ is a threshold value. It was decided that. From the viewpoint of productivity, it is preferable to set the holding temperature for brazing heat treatment lower and set the holding time shorter. When emphasizing such productivity, the holding temperature and holding time satisfy 1 / J ≦ 1.0 × 10 ⁻⁴ and the λ value of the brazed joint is sufficiently large from the regression line shown in FIG. in a range of 1 / J values which can be expected to reach a large 300μm as a totaling 8.0 × ^{10 -5} around satisfies ^{8.0 × 10 -5 ≦ 1 / J} ≦ 1.0 × 10 -4 It is preferable.

  As described above, the brazed joint of the present invention has excellent sulfuric acid corrosion resistance that can be expected to have a D value of 0.10 mm or less, further 0.05 μm or less. The brazed joint of the present invention can be produced by using the above-described method for producing a brazed joint of the present invention, and includes a first metal member made of austenitic stainless steel and a first metallic member made of austenitic stainless steel. Two metal members are joined to each other through a brazing layer containing B. The region adjacent to the brazing layer of the first metal member and the region adjacent to the brazing layer of the second metal member have diffusion regions containing Cr-based borides. In the brazing joint of the present invention, the λ value obtained by a predetermined method described later satisfies λ ≧ 10 μm in the brazing layer adjacent diffusion region closest to the brazing layer of the diffusion region containing Cr boride.

  The sulfuric acid corrosion resistance of metal members made of austenitic stainless steel, which is a material for brazing joints, is generally good, but these metal members are joined together using a brazing material (brazing foil or brazing powder) containing B. The sulfuric acid corrosion resistance of brazed joints is not always as good as that of the material. In the brazed joint with reduced sulfuric acid corrosion resistance, B diffused from the brazing filler metal side by heating during brazing in the boundary region between the metal member and the brazing layer, particularly in the region closest to the brazing layer (the brazing layer adjacent diffusion region). In some cases, a structure (Cr-deficient layer) having a low Cr concentration generated by consumption of surrounding Cr when bonded to Cr on the metal member side to generate a Cr-based boride may be confirmed. When a Cr-depleted layer is connected in a brazed joint and a larger Cr-depleted layer is formed, or when a wider network of Cr-depleted layers is formed, the brazed joint cannot be expected to have sulfuric acid corrosion resistance.

  The λ value represents the average spacing of Cr-based borides in the brazing layer adjacent diffusion region containing the Cr-based boride closest to the brazing layer of the metal member as described above. It can be used as an index representing the size and spread of Cr-based boride (in other words, Cr-deficient layer) in the diffusion region. For example, if the λ value in a brazed joint is 10 μm or more (preferably 20 μm or more), the sulfuric acid corrosion resistance is excellent and the D value is small (see FIG. 2), and if it is less than 10 μm, the D value is large. There is. Therefore, the brazed joint of the present invention satisfying the λ value of λ ≧ 10 μm has good sulfuric acid corrosion resistance.

  The λ value in the brazed joint will be described with reference to the model shown in FIG. The range in which the λ value is measured (λ measurement range 5) is a direction that is substantially parallel to the brazing layer 3 in the diffusion region 4 including the Cr-based boride, and the parallel direction is defined as follows. As a longitudinal direction, a region having a predetermined width (for example, 5 μm) including a region (a brazing layer adjacent diffusion region) in which Cr-based boride close to the brazing layer 3 is likely to have a higher concentration can be set. The λ value is a line segment connecting a base point and an end point in the longitudinal direction of the λ measurement range 5, and the line segment LSj (j = 1 to m, m is a length of 1 or more), and Li is the length on the line segment LSj of the i-th Cr-based boride from the base point PSj (the portion where the i-th Cr-based boride overlaps the line segment LSj from the base point PSj) Length), and n (N is an integer equal to or greater than 1) is the number of Cr-based borides counted on the line LSj, j λj values obtained by λj = (L−ΣLi) / n It can be calculated as an average value. The length L of the line segment LSj coincides with the length in the longitudinal direction of the brazing layer of the specimen used in the sulfuric acid corrosion test described above (a square flat plate having a stripe-shaped brazing layer with four sides having a length of about 10 mm). By doing so, the length of the j line segments LSj can be made uniform.

  In addition, the clear setting of the boundary between the brazing layer 3 and the metal member 1 is difficult because a method suitable for practical use for uniquely identifying the boundary is not known. Therefore, in the present invention, the boundary line (straight line) of the brazing layer 3 is used to analyze the brazing layer 3, the diffusion region 4, Cr, B, and other elements (for example, Ni) of the brazing layer 3, the diffusion region 4, and the metal member 1. A rational boundary line (probably a curve) is set in consideration of the distribution form, and a rational boundary line (straight line) is set in consideration of the boundary line (curve). And since the direction of the boundary line (straight line) cannot be uniquely specified as to whether it is truly parallel to the brazing layer 3, it is referred to as the “substantially parallel direction”. In the model shown in FIG. 4, for the sake of simplicity, a region including a line segment LSj sandwiched between two dotted lines is shown as a λ measurement range 5, and the λ measurement range 5 has a high concentration of Cr-based boride. It is shown as being equivalent to the “region of a predetermined width” including the diffusion region near the soldering layer.

  In addition, the λ value in the brazed joint is a digital that represents a strength (concentration) distribution of at least Cr and B by EPMA, for example, in a portion including a brazing layer adjacent diffusion region that tends to have a higher concentration of Cr-based boride. An image was obtained, and the λj value in the j-th line segment LSj was obtained by λj = (L−ΣLi) / n using the text data array of Cr intensity distribution with reference to B, and obtained. It can be obtained as an average value of m λj (λ1 to λm) values. In addition, as a method for discriminating a region where the Cr-based boride has a higher concentration, for example, the intensity distribution of Cr by EPMA is displayed in black color using, for example, an intensity distribution emphasized by ternarization described later. For example, a method of setting the area as a higher density area can be employed.

  In the above-described measurement range of the λ value of the digital image, j is obtained when there is one pixel (cell) in a direction orthogonal to the line segment LSj having the length L (Y direction shown in FIG. 4). = 1, and in the case of m, j = m. If there is a reasonable reason such as the uniformity of Cr-based boride concentration distribution in the λ measurement range 5, the λ value is determined as j = 1, thereby producing a product for determining whether the sulfuric acid corrosion resistance of brazed joints is good or bad. Increased efficiency can be expected. Further, the length Lj (j = 1 to m) of the line segment LSj is preferably matched with the length of the brazing layer in the longitudinal direction as described above when j is 2 or more. When j is 2 or more, since m λ values (λj values, j = 1 to m) are obtained for m line segments LSj as described above, an average of the m λj values is obtained. Therefore, the reliability of the λ value can be expected to be improved. From this point of view, the λ value can be regarded as an average value of the size of the Cr-based boride on the line segment LSj, or can be regarded as an average value of the size of the Cr-based boride in the λ value measurement range. You can also

  As an example, referring to FIG. 4 schematically showing the strength distribution of Cr by EPMA in a portion including the brazing layer adjacent diffusion region closest to the brazing layer of the brazing joint, one line segment LSj (j = m = 1) ) And setting one λj value (j = m = 1) as the λ value will be specifically described. FIG. 4 is obtained by setting the magnification so as to include the brazing layer adjacent diffusion region in the diffusion region 4 of the brazing joint having the brazing layer 3 between the metal member 1 and the metal member 2 (not shown). 2 schematically shows a digital image of Cr intensity distribution by EPMA. In addition, the rough division of the brazing layer 3, the diffusion region 4, and the brazing layer adjacent diffusion region (substantially overlapping with the λ measurement range 5) is indicated by a dotted line.

  In such a digital image (actually, text data array), a line segment LSj (j = 1) of length L connecting the base point PSj (j = 1) and the end point PEj (j = 1) in the X direction of the λ measurement range 5 Set. Subsequently, on the line segment LSj, the length L1 (i = 1) on the line segment LSj of the first Cr boride C1 closest to the base point PSj is obtained, and then the second Cr boride C1 of the second Cr boride C1 is obtained. Find the length L2 (i = 2) on the line segment LSj, ... (omitted) ..., find the length Li on the line segment LSj of the i-th Cr-based boride Ci, ... (omitted) ..., finally By determining the length Ln (i = n) of the line segment LSj of the n-th Cr-based boride Cn closest to the end point PEj, n Li (i = n) on the line segment LSj (j = 1) Can be requested. Using these n Li (i = n), the λ value when j = 1 can be obtained from λ = (L−ΣLi) / n.

  The magnification of the above-mentioned EPMA can be adjusted, for example, in the range of 200 times to 1800 times so that a plurality of Cr-based borides exist in the λ measurement range 5 of the specimen (brazing joint). The intensity distribution (digital image) of Cr and B by EPMA may be a CPS value (Counts Per Second) or an atomic% (at%) value. The text data array by EPMA may be in CSV (Comma Separated Values) format or the like. Further, when obtaining the λ value, the text data array of Cr and B intensity distributions (digital image) by EPMA can be processed according to a predetermined rule and used as a more emphasized intensity distribution (digital image). For example, in a text data array in which intensity values are arranged, a data group having a large numerical value (e.g., equivalent to 5% of the total number of data) among all data is associated with black color display so that a data group having the next largest numerical value ( For example, 20% of all data is associated with gray color display, and the remaining numerical values (for example, 75% of all data) are associated with white color display so as to be ternary (black). (Gray / white) intensity distribution may be used.

  In addition, it is difficult to uniquely determine the boundary of the above-described determination of the λ measurement range 5 even when the structure of the brazed joint may be different from one another. Therefore, the determination of the λ measurement range 5 takes into account the strength distribution of Cr and B by EPMA, and also refers to the strength distribution of other elements as necessary, from the metal member side 1 and 2 to the brazing layer 3. On the other hand, a method of determining the boundary where the distribution form of the Cr boride changes from coarse to dense can also be adopted. In that case, by performing an operation of emphasizing the intensity distribution of Cr and B by the above-described method such as ternarization, the change in roughness of the distribution form of the Cr boride is further clarified, so that the λ measurement range 5 can be easily determined. Furthermore, a threshold value for determining a change in density of the Cr-based boride is set based on a pixel number ratio (area ratio) occupied by Cr and / or B with respect to all pixels, and this threshold value is used as a determination criterion. According to the method, the determination of the λ measurement range 5 can be performed more easily. In addition to these methods, from the viewpoint of practicality, from the above-mentioned digital image of the strength distribution of Cr and B, for example, within 2 μm from the brazing layer side that is prone to sulfuric acid corrosion in brazing joints, in order to improve reliability A method of determining within 5 μm and within 10 μm as the predetermined width of the λ measurement range 5 is preferable for further improvement of reliability. From the viewpoint of achieving both reliability and production efficiency, the predetermined width of the λ measurement range 5 is preferably 5 μm.

(Invention Example 1)
A brazed joint having the configuration shown in FIG. 11A (hereinafter referred to as “brazed joint A”) was produced in consideration of the 1 / J value. The first metal member and the second metal member that are the materials of the brazed joint A were made of SUS304 and had a rectangular parallelepiped shape having a total length of 25 mm and a joint surface size of 35 mm × 10 mm. The joining surface of the metal member was subjected to acetone ultrasonic cleaning after paper polishing (# 800). The brazing material used for brazing is No. 1 shown in Table 1. No. 6 brazing foil (B is 1.16% by mass, Ts is 948 ° C., average thickness is 31 μm), and a rectangular sheet (wax foil) cut into a size corresponding to the joining surface of the metal member is used. It was. And the brazing foil was pinched | interposed with the joining surface of the 1st metal member, and the joining surface of the 2nd metal member, the temporary assembly was performed, and the raw material of the brazing joint A was produced.

Subsequently, the shape of the brazed joint A is degreased (argon gas atmosphere, temperature rising and cooling rate is 80 ° C./min, holding temperature is 500 ° C., holding time is 1 h), and 1 / J value is taken into consideration A brazing heat treatment was performed. The brazing heat treatment is started in a vacuum atmosphere (about 2 × 10 ⁻³ Pa to about 5 × 10 ⁻³ Pa) at room temperature (20 ° C.), the temperature rising rate is 80 ° C./min, and the holding temperature is 1 min. As a holding time at which 1 / J value is 19.4 ° C. and 9.4 × 10 ⁻⁵ [K · log (seconds) ⁻¹ within the specified range of the manufacturing method of the present invention with respect to the holding temperature. When the temperature decreasing rate reached 80 ° C./min and 400 ° C. in the temperature decreasing process, blast cooling with nitrogen gas was started. The J value here is obtained by simplifying the rectangular pattern as described above.

  Next, a flat plate having the configuration shown in FIG. 11B was cut out from the brazed joint A produced by the above-described method, and a test body used for the sulfuric acid corrosion test (hereinafter referred to as “test body A”) was produced. Subsequently, the Cr and B strength distribution of the specimen A was measured using EPMA. Then, using the EPMA (digital image) text data array, a text data array in which the intensity distribution of Cr and B by the above-described ternarization (black / gray / white) is emphasized was created. FIG. 5 shows a digital image (200 times) in which the intensity distribution of Cr and B is emphasized, and FIG. 6 shows a digital image (1800 times) obtained by enlarging them. Next, using the ternary Cr intensity distribution (digital image), a boundary line (straight line) of the brazing layer 3 which is the λ measurement range 5 of the specimen A is set, and the width of the λ measurement range 5 is The λ value was determined according to the above-described method for determining the λ value, with 5 μm and j = 1. The boundary line (straight line) of the brazing layer 3 was determined based on the distribution form of Cr and B shown in FIGS. 5 and 6 (particularly focusing on the right end side). As a result, the λ value of the specimen A, that is, the brazed joint A, was 24.9 μm.

  Next, a sulfuric acid corrosion test was performed using the specimen A from which the λ value was obtained. In the sulfuric acid corrosion test, the corrosive liquid was special grade sulfuric acid, the sulfuric acid concentration was 10 mass%, the liquid temperature was 60 ° C., and the immersion time of the specimen A was 3 h. And the test body after a sulfuric acid corrosion test (henceforth "the test body Aa") was ultrasonically washed with water, and dried for 2 h at 80 degreeC. According to the method for obtaining the maximum erosion depth (D value) using the specimen Aa after the sulfuric acid corrosion test, a plurality of profiles crossing the eroded portion of the specimen Aa are measured using a laser microscope, The largest (deep) value of was determined as the D value. As a result, the D value of the specimen Aa was 0.04 mm which is 0.10 mm or less.

From this result, a brazing joint is produced by brazing (brazing) a first metal member made of SUS304 and a second metal member using a brazing foil containing 1.16% by mass of B. In the present invention, the holding temperature and holding time of the brazing heat treatment are selected within a range where the 1 / J value satisfies 1 / J ≦ 1.0 × 10 ⁻⁴ ) [K · log (seconds)] ^−1. By applying the method, it was confirmed that the brazed joint of the present invention having a λ value satisfying λ ≧ 10 μm, a small D value and good sulfuric acid corrosion resistance was obtained.

(Invention Example 2)
A brazed joint (hereinafter referred to as “brazed joint B”) was produced in the same manner as in the first example of the present invention in consideration of the 1 / J value. For the brazing material used for brazing, No. 1 shown in Table 1 was used. No. 1 wax foil (B: 3.12% by mass, Ts: 976 ° C., average thickness: 42 μm) was used. The brazing heat treatment is started at room temperature (20 ° C.) in a vacuum atmosphere (about 2 × 10 ⁻³ Pa to about 5 × 10 ⁻³ Pa), and the temperature rising rate is maintained at 80 ° C./min. 270 minutes as the holding time when the temperature is 1200 ° C. and the 1 / J value is 9.0 × 10 ⁻⁵ [K · log (seconds) ⁻¹ within the specified range of the manufacturing method of the present invention with respect to the holding temperature. When the temperature decreasing rate reached 80 ° C./min and 400 ° C. in the temperature decreasing process, blast cooling with nitrogen gas was started. Other than the above, for example, a sulfuric acid corrosion test, a method for preparing a test specimen used therefor, a method for obtaining a λ value and a D value, and the like are the same as those in the first example of the present invention, and thus description thereof is omitted.

  Using the text data array of EPMA (digital image) measured for the specimen B cut out from the brazed joint B in the same manner as in Example 1 of the present invention, the strength of Cr and B by the above-described ternarization (black / gray / white) A text data array with an emphasized distribution was created. FIG. 7 shows a digital image (200 times) in which the intensity distribution of Cr and B is emphasized, and FIG. 8 shows a digital image (1800 times) obtained by enlarging them. Using the ternary Cr intensity distribution (digital image), the λ measurement range 5 of the specimen B (within 5 μm from the brazing layer 3) is the same as that of the specimen A according to the method for obtaining the λ value described above. As a result, the λ value of the specimen B, that is, the brazed joint B, was 291 μm. Moreover, as a result of obtaining the D value of the test specimen (test specimen Ba) after the sulfuric acid corrosion test in the same manner as in the case of the specimen Aa according to the method for obtaining the D value described above, the D value of the specimen Ba is 0.10 mm or less. 0.03 mm.

From this result, when a brazing joint is produced by brazing (brazing) a first metal member made of SUS304 and a second metal member using a brazing foil containing 3.12% by mass of B. In the present invention, the holding temperature and holding time of the brazing heat treatment are selected within a range where the 1 / J value satisfies 1 / J ≦ 1.0 × 10 ⁻⁴ ) [K · log (seconds)] ^−1. By applying the method, it was confirmed that the brazed joint of the present invention having a λ value satisfying λ ≧ 10 μm, a small D value and good sulfuric acid corrosion resistance was obtained.

(Comparative example)
A brazed joint (hereinafter referred to as “brazed joint C”) was produced in the same manner as in the above-described Invention Example 2 in consideration of the 1 / J value. In addition, the brazing material used for brazing is similar to Example 2 of the present invention in No. 1 shown in Table 1. No. 1 wax foil (B: 3.12% by mass, Ts: 976 ° C., average thickness: 42 μm) was used. The brazing heat treatment is started at room temperature (20 ° C.) in a vacuum atmosphere (about 2 × 10 ⁻³ Pa to about 5 × 10 ⁻³ Pa), and the temperature rising rate is maintained at 80 ° C./min. The holding time when the temperature is 1200 ° C. and the 1 / J value is 1.1 × 10 ⁻⁴ [K · log (seconds)] ⁻¹ outside the specified range of the manufacturing method of the present invention is 0. 1 minute, the temperature drop rate was 80 ° C./min. When the temperature drop process reached 400 ° C., blast cooling with nitrogen gas was started, and the heat pattern was used. Other than the above, for example, a sulfuric acid corrosion test, a method for preparing a test body used therefor, a method for obtaining a λ value and a D value, and the like are the same as in the examples of the present invention, and thus description thereof is omitted.

  The intensity distribution of Cr and B by ternarization (black / gray / white) described above using the text data array of EPMA (digital image) measured for the specimen C cut out from the brazed joint C as in the present invention example A text data array with emphasis was created. FIG. 9 shows a digital image (200 times) in which the intensity distribution of Cr and B is emphasized, and FIG. 10 shows a digital image (1800 times) obtained by enlarging them. Using the ternary Cr intensity distribution (digital image), the λ measurement range 5 (within 5 μm from the brazing layer 3) of the test specimen C is the same as the test specimens A and B according to the method for obtaining the λ value described above. As a result, the λ value of the specimen C, that is, the brazed joint C, was 2.9 μm. In addition, as a result of obtaining the D value of the test specimen after the sulfuric acid corrosion test (test specimen Ca) in the same manner as the specimens A and B according to the method for obtaining the D value described above, the D value of the specimen Ca is 0.10 mm. It was more than 0.14 mm.

From this result, as in Example 2 of the present invention, the brazing joint was brazed using a brazing foil containing 3.12 mass% B of the first metal member and the second metal member made of SUS304 ( In the case where the 1 / J value does not satisfy 1 / J ≦ 1.0 × 10 ⁻⁴ ) [K · log (seconds)] ^{− 1} , the brazing heat treatment holding temperature and holding are maintained. By applying a manufacturing method that selects time, it was confirmed that a brazed joint having a λ value not satisfying λ ≧ 10 μm and a large D value and poor sulfuric acid corrosion resistance was obtained. In addition, when a brazing foil containing 3.12% by mass of B is used, even if the brazing foil contains the same amount of B due to the difference in heat treatment conditions for brazing, the λ value of the brazed joint, that is, the Cr-based boride It has been found that the distribution form of changes. Therefore, when the brazing foil contains, for example, 2.0% by mass to 4.0% by mass of B, it is strongly affected by the heat treatment condition of brazing and reacts sensitively and diffuses into the metal member. It turns out that there is a risk that the behavior of B becomes unstable.

1. Metal member, 2. Metal member, 3. 3. brazing layer; 4. diffusion region, λ measurement range, 10. Brazed joint, 20. Specimen

Claims (2)

A first metal member made of austenitic stainless steel and a second metal member made of austenitic stainless steel are each composed of a Ni-based amorphous material containing 0.30 mass% or more and 6.0 mass% or less of boron. A method for producing a brazed joint formed by joining using a brazing foil made of an alloy or an Fe-based amorphous alloy, wherein the brazing foil is heated to be melted, and the molten brazing is cooled and solidified. A holding step and a holding time in the heat treatment step, T (t) is defined as a temperature of the metal member after t hours from the start of the heating of the brazing foil, and Ts Is the solidus temperature of the brazing foil, t0 is the time when T (t) reaches Ts, and t1 is the time when T (t) exceeds Ts and again reaches Ts. , T0 In the section of al t1 J = ∫ [T (t) · 1 / ( t-t0) ] J value obtained in dt is ^{1 / J ≦ 1.0 × 10 -4} [K · log ( ^{sec) -1} A method for manufacturing a brazed joint, which is selected within a range that satisfies the above.   A brazing joint in which a first metal member made of austenitic stainless steel and a second metal member made of austenitic stainless steel are joined through a brazing layer containing boron, In the diffusion region containing the Cr-based boride adjacent to the brazing layer of the metal member, the brazing layer has a direction substantially parallel to the brazing layer, and the parallel direction is a longitudinal direction of the brazing layer. A region having a predetermined width is set as a measurement range so as to include a brazing layer adjacent diffusion region having a higher concentration of the adjacent Cr-based boride, and L is a base point and an end point in the longitudinal direction of the measurement range , And the length of a line segment LSj (j = 1 to m, m is an integer of 1 or more) set with uniformity in the measurement range, and Li is a base point on the jth line segment LSj Or When the length of the i-th Cr-based boride on the line segment LSj is assumed to be n (an integer of 1 or more), the number of Cr-based borides counted on the j-th line segment LSj, λj = (L A brazed joint in which a λ value that is an average value of j λj values obtained by ΣLi) / n satisfies λ ≧ 10 μm.