JP6847356B2 - Method for manufacturing hydrogen separation membrane and hydrogen separation membrane - Google Patents

Description

The present invention relates to a method for producing a hydrogen separation membrane and a hydrogen separation membrane.

As a hydrogen separation membrane, palladium (Pd), niobium (Nb), vanadium (V), tantalum (Ta), which are metals capable of selectively permeating hydrogen, and their alloys are thin-filmed by rolling or plating. The one processed into is used.

For example, in Patent Document 1, a metal film forming step of laminating a metal layer containing no palladium on one side of a palladium-containing metal thin film containing palladium to form a metal film serving as a precursor of a hydrogen separation film. For electroless plating, the osmotic pressure of the solvent in which the solute arranged on the metal layer side of the metal film is dissolved and the catalyst solution for electroless plating arranged on the palladium-containing metal thin film side of the metal film is used. A catalyst application step for electroless plating in which a catalyst solution is moved to the metal layer side of the metal film through a defect and a catalyst for electroless plating is applied to a defect portion on the metal layer side of the metal film, and a catalyst application step for electroless plating. It does not contain a solvent in which at least a reducing agent arranged on the metal layer side of the metal film to which a catalyst is applied is dissolved as a solute, and a reducing agent containing metal ions arranged on the palladium-containing metal thin film side of the metal film. A metal ion reduction / precipitation step in which the plating solution is moved to the metal layer side of the metal film by the osmotic pressure with the plating solution, and the metal ions are reduced / precipitated on the defective portion on the metal layer side of the metal film. A method for producing a hydrogen separation film comprising the above is described.

In order to purify high-purity hydrogen, hydrogen separation membranes are required to have high mechanical strength, hydrogen permeability and durability in a high-temperature hydrogen gas environment, and production of hydrogen separation membranes having such properties. A method is required.
Specifically, a hydrogen separation membrane having a thickness of several to several hundreds of microns is subject to a long-term hydrogen pressure load at a high temperature, so that it is required to have performance as a heat-resistant structural material while being a functional material, and therefore for a long time. It is a big issue to improve the durability (crack resistance, etc.) that can withstand the stable operation of.

Here, it has long been known that vanadium (V), niobium (Nb) and tantalum (Ta) become brittle in a hydrogen atmosphere and violently break into brittleness, and this problem can be solved in other countries. However, the development of non-Pd hydrogen separation membranes has hardly been carried out.
On the other hand, in Japan, the development of non-Pd hydrogen separation membranes is being continuously and enthusiastically promoted. This metal membrane is usually used at a high temperature of 400 ° C. or higher, but at a low temperature, hydrogen embrittlement occurs due to the formation of hydrides. It tended to be understood that it could not be used to do so.

Patent Document 2 describes a method of applying shear strain to a hydrogen storage alloy in order to produce a hydrogen storage alloy that is inexpensive, does not require high-temperature activation treatment, and can repeatedly absorb / release hydrogen. ing.
That is, Patent Document 2 includes a hydrogen transfer path forming step using TiFe as a material, extending from the surface of the material to the inside of the material, and forming a hydrogen transfer path having a lower hydrogen transfer resistance than the surrounding material. The hydrogen transfer path forming step is characterized in that a line defect or a surface defect extending from the material surface of the TiFe to the inside of the material is formed as the hydrogen transfer path by applying a shear strain of 7 or more to TiFe. , A method for producing a TiFe hydrogen storage alloy is described.

However, both of the above patent documents both describe and suggest that high mechanical strength, hydrogen permeability, and long-term durability can be obtained by applying a large shear strain during the production of a hydrogen separation membrane. It has not been.

We have described the hydrogen pressure-composition-isothermal line (PCT curve) of metal membranes, the chemical potential gradient of solid-dissolved hydrogen acting between the membranes, and the hydrogen concentration (DBTC) at which the metal membrane transitions from ductile to brittle. From this point of view, we have independently developed a design guideline that can accurately calculate the optimum chemical composition of the alloy film.
Many studies so far have focused on grasping the hydrogen permeability coefficient and the film strength after a short hydrogen pressure load, and the high temperature of the surface coating Pd as a hydrogen dissociation catalyst and the underlying V, Nb, and Ta. It is also known that the hydrogen permeability decreases with time due to mutual diffusion in the water, and low-temperature working membranes have been studied.

Accordingly, the present inventors have found that in manufacturing the hydrogen separation membrane, epsilon = 10 or more by performing strong distortion processing which imparts enormous strain, vanadium Group 5 metal, even at a low temperature of 300 ° C. 10 ^- We have found that a hydrogen permeability coefficient Φ in the range of ⁷ (mol H ₂ m ^-1 s ^-1 Pa ^{−0.5) can be obtained, and have completed the present invention.}

Here, S. A. Steward: According to Lawrence Livermore National Laboratory Report, UCRL-53441, (1983), iron (Fe) also has niobium (Nb) and tantalum (Nb), as shown in FIG. 1, which shows the temperature dependence of the hydrogen permeability coefficient of various metals. It has been suggested that it has hydrogen permeability similar to Ta), palladium (Pd) and the like, and is the subject of the present invention.

That is, the means for solving the problem of the present invention is as follows.

First, the hydrogen separation membrane is characterized by performing a strong strain process for imparting a strain of ε = 10 or more, preferably 30 or more to a thin film metal capable of selectively permeating hydrogen. Production method.
Here, in the strong strain processing in which a strain of ε = 10 or more is applied, a strong compressive force and a shearing force are applied to the material, and a large shear strain is applied to make the crystal grains ultrafine, resulting in high mechanical strength and hydrogen. It refers to a processing means performed for the purpose of obtaining permeability.
The strong strain processing includes high-pressure torsion (HPT) processing, high-pressure sliding (HPS) processing, repeated shear deformation processing (ECAP) method, repeated lap joint rolling (ARB) process, and multi-directional forging (MDF) process. Shot peening processing, drill processing, ball milling, etc. can be adopted.

The strain ε is expressed by the following equation, for example, in the case of HPT processing.
Here, ε is the equivalent strain, r is the distance from the center of the sample, N is the rotation speed, and t is the thickness of the disk sample.

The strain ε during HPS processing is expressed by the following equation.
Here, ε is the equivalent strain, X is the slide amount, and t is the plate thickness.

Second, the metal is niobium (Nb), vanadium (V), tantalum (Ta) or al selected, characterized in that it is a one or more metals, according to the first A method for producing a hydrogen separation membrane.
Third, the method for producing a hydrogen separation membrane according to the first or second method, wherein the strong strain processing is a high-pressure torsion (HPT) processing.
Fourth, a hydrogen separation membrane which is a metal membrane capable of selectively permeating hydrogen , belongs to Group 5 in the periodic table, and has a particle size number N of 18 or more.
Here, the particle size number N is obtained ^{by two-dimensionally measuring the number n of crystal grains in 1 mm 2} and rounding up the value of the following equation to indicate an integer value.

Fifth, the hydrogen separation membrane according to the fourth aspect, wherein the metal membrane is a vanadium membrane having a particle size number N = 20 and an average particle diameter d = 390 nm.

According to the present invention, the following effects can be obtained.

When producing a hydrogen separation membrane by performing a strong strain process to apply a huge strain to a thin film metal capable of selectively permeating hydrogen, for example, pure vanadium is twisted with high pressure (HPT: High Pressure Torsion). ) By performing strong strain processing by processing, ultrafine crystal grains are produced, but a vanadium film having a particle size number N = 20 and an average particle size d = 390 nm can be obtained.
The vanadium film has 2.5 to 5 times higher hydrogen permeation performance than ordinary molten materials and rolled materials even at a low temperature of 300 ° C. without coarsening (grain growth) of crystal grains. It was confirmed that (see FIG. 21).
Further, it was confirmed that the vanadium film could sufficiently suppress the film deformation caused by the loaded hydrogen pressure and the solid solution hydrogen (see FIGS. 22 and 23).

Furthermore, when the same process was applied to Ni, which had hardly seen a hydrogen permeation phenomenon so far, it was confirmed that the hydrogen permeation phenomenon was exhibited at the same temperature.

Therefore, a solid solution that ensures a balance between hydrogen permeation performance, hydrogen embrittlement resistance, and durability by performing alloying and metallographic control on Group 5 metals, Group 10 metals, or alloys thereof by various strong strain processing. It becomes possible to manufacture a hydrogen separation membrane of a single-phase alloy.

Here, it is expected to exhibit hydrogen separation performance that is an order of magnitude higher than that of molten materials, rolled materials, conventional Pd-Ag alloys and V-15Ni alloys.

Further, from the viewpoint of grain coarsening, operation at 400 ° C. or lower is preferable, but by performing strong strain processing after alloying to raise the PCT curve, a hydrogen separation membrane that can be operated at a lower temperature can be produced. It will be possible.

By producing a hydrogen separation membrane that establishes a balance between hydrogen permeation performance, hydrogen embrittlement resistance, and durability by the process found by the inventors, it is possible to produce a hydrogen separation membrane that can operate at a low temperature close to room temperature. It becomes.

It is a figure which shows the temperature dependence of the hydrogen permeability coefficient of various metals. It is the schematic of the HPT processing apparatus which concerns on this invention. It is a figure which shows the cut-out shape of the sample piece used for a tensile test. It is a nominal stress-strain diagram by a tensile test. It is a figure which shows the Vickers hardness at each heat treatment temperature. It is a photograph which shows the EBSD orientation image of the crystal structure before and after the heat treatment. It is a photograph which shows the optical microscope image of the comparative product 1. It is a photograph which shows the EBSD orientation image of the crystal structure heat-treated at two different temperatures. It is the schematic of the piping system of the SP test apparatus. It is the schematic which shows the internal structure of the SP test apparatus together with the SP jig. It is an SP curve in a vacuum at 300 degreeC. It is an SP curve in a vacuum at 400 ° C. It is an SP curve in a vacuum at 500 ° C. It is a figure which shows SP absorption energy at each temperature collectively. It is a photograph which shows the appearance and SEM image of the fracture surface of each sample. It is a figure which shows the result of the hydrogen atmosphere in-situ SP test. It is a DBTC curve. It is a figure which shows the appearance and the SEM image of the fracture surface of each sample after a hydrogen atmosphere in-situ SP test. It is the schematic of the piping system for pressure control in a hydrogen permeation apparatus. It is the schematic of the membrane sample fixing part in a hydrogen permeation apparatus. It is a figure which shows the hydrogen permeation test result at 300 degreeC. It is a figure which shows the appearance of each separation membrane after hydrogen permeation. It is a figure which shows the SEM image of the surface of each separation membrane after hydrogen permeation. It is the graph which compared the transmission coefficient at each temperature. It is a figure which shows the EBSD orientation image of the sample before and after the permeation test at 500 degreeC. It is a graph which shows the PCT curve of each measured sample. It is a graph which shows the result of the hydrogen permeation test in a nickel sample. It is a photograph which shows the EBSD orientation image before and after the hydrogen permeation test in a nickel sample.

(1) HPT processing device As shown in FIG. 2, the HPT processing device 1 for performing strong strain processing used in this embodiment places a sample in a flat recess in the center of the anvil and forcibly sandwiches it, resulting in a very large pressure. By rotating the lower anvil with (usually 1 GPa or more) applied, a large amount of shear strain is introduced into the sample, and the crystal grains are finely divided to the submicron level to obtain fine grains. It is a thing.

(2) Production of Hydrogen Separation Membrane Sample by HPT Processing In this example, a vanadium drawing material having a purity of 99.9% is heat-treated at 1100 ° C. for 1 hour using the HPT processing device, and then at room temperature. A sample of the hydrogen separation membrane, which is the product 1 of the present invention, was produced by performing HPT processing under the conditions of an applied pressure of 1.5 GPa, a rotation speed of 0.25 rpm, and a rotation speed of 5 rotations.
Here, according to the formula expressing the strain during the HPT processing, the strain ε was about 32.6 to the maximum value of 169.
At this time, for the region where hydrogen is permeated in the sample obtained by cutting out the 10 mm disc from the 20 mm disc, r indicating the distance from the sample center is a value in the range of 1.8 mm to 9.3 mm, and thus is strained. ε was also calculated as a lower to upper limit value.
Then, when the mold groove depth was 0.7 mm in total, a disk-shaped hydrogen separation membrane sample having a diameter of 20 mm and a thickness of 0.8 mm, which was the same as the mold hole diameter used for HPT processing, was obtained.
Here, for comparison, a vanadium annealed material having a purity of 99.9% and not subjected to HPT processing was formed into the same disk shape as the product 1 of the present invention to be used as the comparative product 1.

(3) Preparation of sample From the hydrogen separation membrane sample according to the disc-shaped product 1 of the present invention or the comparative product 1 obtained above, a wire-cut electric discharge machine (Wire cut Electrical Discharge Machine; WEDM) is used for each test described later. A sample for each test was prepared by cutting into a desired shape.

(4) Calculation of particle size number The particle size number N was calculated as a method for evaluating the crystal grain size subjected to HPT processing.
First, the average number of crystal grains n present in ^{1 mm 2 was secondarily measured.}
Here, the particle size number N is obtained ^{by two-dimensionally measuring the average number of crystal grains n of the crystal grains in 1 mm 2 of the} cross section of the test piece, and rounding up the value of the following equation and showing it as an integer value.

The number of crystal grains n per unit area was obtained by approximating the shape of the crystal grains as a circle whose diameter is the average particle size d calculated from the EBSD orientation image.

(5) Evaluation by particle size number The larger the particle size number, the finer the crystal grains, and the crystal grain size can be evaluated by calculating the particle size number.
Regarding the crystal grain size, the relationship between the particle size number N and the average crystal grain size (nm) is shown in the following table.

(6) Tensile test (6-1) Tensile test equipment For the tensile test, an AGS-H type universal testing machine manufactured by Shimadzu Corporation was used.
The testing machine is capable of applying only a load in the vertical direction to the gauge portion of the test piece by moving the tension jig connected to the crosshead along the linear guide.

(6-2) Sample Preparation The sample used for the tensile test is cut out from a disk-shaped sample by a wire-cut electric discharge machine (WEDM) into the shape shown in FIG. 3, and then emery wet polishing up to # 1000 and an average particle size. It was obtained by buffing with a 1 μm Al ₂ O ₃ suspension to adjust the final thickness to 0.4 mm.
The unit of the numbers in FIG. 3 is mm.
In order to investigate the effect of annealing, after the same procedure, the two invention products 1 were heat-treated at 500 ° C. and 700 ° C. for 1 hour, respectively (hereinafter abbreviated as HPT-V + 500 ° C. and HPT-V + 700 ° C.). ) And a sample piece of Comparative Product 1 (unprocessed V) cut out to the same shape.

(6-3) Test conditions In this tensile test, the gauge part length l is 1 mm, the crosshead moving speed is 1.0 mm / min, and the strain rate is 1.7 × 10 ^-2 s ^{-1 at room temperature.} It was carried out.
Then, the nominal stress σ was calculated from the tensile load f, and the nominal strain ε was calculated from the jig movement amount x, and a nominal stress-strain diagram was created and shown in FIG.
The nominal stress σ and strain ε were calculated by the following equations.

(6-4) Consideration of test results As shown in FIG. 4, the maximum tensile strength of HPT-V, which is the product 1 of the present invention, is about 1020 MPa, which is the highest strength among the samples used in this test. It was.
On the other hand, since the maximum tensile strength of virgin V, which is the comparative product 1, is about 170 MPa, it was confirmed that the strength increased 6 times by performing the HPT processing treatment.
Further, it was confirmed that the HPT-V + 500 ° C. material maintained a sufficient strength of 890 MPa, which is the maximum tensile strength, although there was a slight decrease in strength.
In the HPT-V + 700 ° C. material, a sharp decrease in maximum tensile strength was observed at 340 MPa.

(7) Hardness test (7-1) Test conditions For the product 1 of the present invention and the comparative product 1, argon for 1 hour in a temperature range from room temperature to 900 ° C. in order to quantitatively evaluate the change in Vickers hardness due to the heat treatment temperature. Heat treatment was performed in the atmosphere, and then the Vickers hardness at each heat treatment temperature was measured and shown in FIG.
In addition, virgin V in FIG. 5 represents pure V before HPT processing, and the hardness thereof is shown by a dotted line.

(7-2) Consideration of test results It was confirmed that the Vickers hardness increased up to 225 Hv by performing HPT processing.
Further, when heat treatment at 500 ° C. is performed after HPT processing, the Vickers hardness gradually decreases to about 200 Hv, whereas the Vickers hardness becomes about 100 Hv and sharply decreases in the heat treatment at 700 ° C. It was confirmed.

It is considered that this is because the heat treatment at 500 ° C. reduces the hardness due to the recovery of dislocations, whereas the heat treatment at 700 ° C. also acts on recrystallization, resulting in a significant decrease in hardness.
That is, the hardness of the product 1 of the present invention begins to decrease by heat treatment at 500 ° C., but since it is also a refractory metal, no appearance of coarsening such as recrystallization was observed, but the decrease in hardness is the recovery of dislocations. It was confirmed that the initial performance can be maintained even if the heat treatment temperature rises to about 700 ° C.
From the above results, it was confirmed that there was a correlation between the tensile strength and the hardness, and it was confirmed that the strength significantly increased by the HPT processing was maintained up to the heat treatment temperature of around 500 ° C. even after the processing.

(8) Crystal structure observation (8-1) Heat treatment test In order to clarify the structural changes in the crystal structure of the product 1 of the present invention before the heat treatment of HPT-V and after the heat treatment at 500 ° C. and 700 ° C. Crystal grains were observed by the electron backscatter diffraction (EBSD) method.
The EBSD orientation image is shown in FIG.
Further, FIG. 7 shows a photograph of an optical microscope image of Comparative Product 1 which is an annealed material before HPT processing.
Furthermore, the average particle size and particle size number of each sample are shown in the following table.

(8-2) Heat Treatment at Two Different Temperatures Crystal grains of Product 1 of the present invention were observed by the EBSD method in order to clarify the structural changes in the structure when heat-treated at 400 ° C. and 500 ° C. The EBSD orientation image is shown in FIG.

(8-3) Consideration of results The annealed material before HPT processing had large crystal grains with an average crystal grain size on the order of mm, whereas the average grain size after HPT processing was 380 nm, forming submicron crystal grains. Was there.
Further, the average particle size was 390 nm and the crystal grain size was hardly changed even when the heat treatment was performed at 500 ° C., but the average particle size when the heat treatment was performed at 700 ° C. was 1.35 μm, and the crystal grains were coarse. Confirmed the conversion.
Further, the average crystal grain size when heat-treated at 400 ° C. is 360 nm, whereas the average crystal grain size when treated at 500 ° C. is 390 nm, and even when treated at 500 ° C., the crystal grains. Was kept fine, and it was confirmed that no coarsening of crystal grains was observed.
That is, it is presumed that the product 1 of the present invention functions as a hydrogen separation membrane for a long time even at 500 ° C. and does not deteriorate in function due to coarsening.
From these results, it can be seen that the structure is recrystallized at 700 ° C.
Therefore, recrystallization is considered to occur in the temperature range of 500 to 700 ° C.

(9) Small Punch (SP) Test (9-1) SP Test Equipment The SP test has a pressure control piping system outlined in FIG. 9 and an SP jig shown in FIG. ..

In the pressure control piping system, since hydrogen used in the test is a flammable gas, an airtight structure that does not leak hydrogen is required.
Further, when the SP test is carried out in a vacuum, the temperature of the SP jig and its surroundings becomes as high as 500 ° C. Therefore, a piping system having excellent airtightness and heat resistance is required. Therefore, a valve manufactured by Swagelok Co., Ltd. -Piping was performed as shown in FIG. 9 using fittings and various valves.
Here, a hydrogen introduction port was connected to the primary side (P1) of the SP test apparatus, and a hydrogen discharge port was connected to the secondary side (P2).
This makes it possible to properly operate the valves in the piping system and perform destructive tests under various hydrogen pressures.

The SP jig is for performing the SP test on the spot after the hydrogen permeation test, SUS304 is used as the material of the main part, SUS316L is used as the material of the heat-resistant heater block, and the heater is manufactured by WATLOW. And a cartridge heater manufactured by Misumi Co., Ltd. was used so that the temperature of the SP jig could be raised to 600 ° C.
In addition, a K thermocouple is also built in the lower part of the SP jig for temperature measurement and temperature control.
Here, since it is necessary to withstand high-temperature and high-pressure hydrogen gas, a molded bellows capable of withstanding the set test conditions was separately designed and manufactured.
Then, in order to prevent hydrogen leakage to the outside of the test apparatus, the flanges and bellows flanges of the upper and lower dies were sealed with an ICF70 conflat flange and a copper gasket for high vacuum.
Further, an airtight structure was formed by fixing the upper die and the upper flange of the molded bellows with bolts.

(9-2) Sample Preparation After cutting a disk-shaped sample into a circular shape with a diameter of 7 mm using a wire-cut electric discharge machine (WEDM), the surface is emery wet-polished up to # 1000 and Al ₂ O _{3 with an average particle size of 1 μm.} The surface was mirror-polished by buffing with the suspension, and the final thickness was adjusted to 0.5 ± 0.01 mm.
Further, in order to impart hydrogen dissociation catalytic property, both sides of the sample were coated with Pd having a thickness of about 200 nm using a DC magnetron sputtering apparatus.
The same steps were applied to the sample for the SP test in vacuum in order to make the conditions the same as those for the hydrogen atmosphere in-situ SP test.
In the in-situ SP test so far, ^{a sample of 10 × 10 × 0.5 mm 3} was used, but as a result of the actual test, the punched ball touched the ground at the sample part contributing to deformation / destruction. Since the diameter was about 5 mm centered on the point, the SP test was performed with the above sample size in this test.

(9-3) Small Punch (SP) Test Method The small punch (SP) test method is a steel ball with a diameter of 2.4 mm on a small plate-shaped sample (in this test, Si _{3 from the viewpoint of hydrogen embrittlement resistance and maintenance of high temperature strength.} Place the use) N ₄ balls, a punching destructive test method for measuring the load and indenter movement amount generated when pressed against the steel balls sample puncher.
Here, the area under the load-indenter movement amount curve until the membrane sample breaks is defined as SP absorbed energy, and is used as a reference when comparing the energy required for breaking the membrane sample.
Comparing this test method with the Charpy impact test method, in the SP test, the test piece size is small enough to be obtained from one end of the Charpy impact test piece, and the change in absorbed energy evaluated in the test is in a narrow temperature range. It is known that a clear ductile-brittle transition behavior appears in.
Therefore, the SP test is often used for evaluation of irradiation damage of reactor core materials.
In the in-situ SP test, a sample with a diameter of 7 mm and a thickness of 0.5 mm was set between the upper jig and the lower die, and fixed with a uniform torque (9.8 Nm) using four hexagon socket head cap screws. did.
Next, vacuum exhaust is performed by an oil rotary vacuum pump and a turbo molecular pump, heated to a desired temperature, maintained for 1.0 hour, and then hydrogen gas under each test condition shown in the following (9-4) is introduced. So, the SP test was conducted.
The in situ moving speed in the in-situ SP test was constant at 0.5 mm / min.

(9-4) Test Conditions First, in order to grasp the film strength of the HPT-V material at each temperature, an SP test in vacuum was performed under three temperature conditions of 300 ° C., 400 ° C., and 500 ° C.
Next, in order to evaluate the effect of the amount of solid solution hydrogen on the mechanical properties of the material, a hydrogen atmosphere in-situ SP test was conducted under various solid solution hydrogen concentration conditions.
The temperature condition was 500 ° C.
The target hydrogen loading pressure was determined by reading from the PCT curve measured using a Sieberts-type hydrogen storage amount measuring device (PCT measuring device) described later.
Then, after applying the target hydrogen pressure and maintaining the state for 2.5 hours, the hydrogen atmosphere in-situ SP test was carried out.
Here, when hydrogen is not dissolved to a hydrogen concentration equilibrated with each hydrogen pressure, the hydrogen dissolution reaction in the membrane sample proceeds.
Then, when the introduction valve of the supplied hydrogen gas is closed, the hydrogen pressure in the test apparatus decreases.
Therefore, in this test, when the pressure drop in the test equipment due to the suspension of hydrogen gas introduction does not occur, it is judged that the hydrogen concentration in the membrane sample has reached the concentration equilibrated with the hydrogen pressure, and immediately after that, it is determined. An in-situ SP test was performed.

(9-5) Fracture surface observation In order to confirm the fracture morphology of the membrane material after the hydrogen atmosphere in-situ SP test and after the vacuum SP test, the fracture surface was observed using a scanning electron microscope (SEM).
The acceleration voltage of the electron gun was set to 20 kV, and two types of SEM images, an overall image of the film material and an enlarged image of the fracture surface, were taken, and a comprehensive evaluation of the characteristics of the fracture surface and the fracture morphology was performed.
The SP curves in vacuum at each temperature of HPT-V and virgin V are shown in FIGS. 11, 12, and 13, respectively.

In virgin V, the SP curve similar to that of a typical ductile material was obtained under all temperature conditions, whereas in HPT-V, the plastic film elongation step, which is the third step of the SP curve, was observed at all temperatures. However, at 300 ° C and 400 ° C, the behavior that the load was temporarily reduced at around 0.7 mm of indenter movement, then increased again, and broke through the fourth stage of plastic instability was confirmed. Was done.
Further, the maximum load at 500 ° C. is larger than that at 300 ° C. and 400 ° C., and is similar to the SP curve seen in a typical ductile material except that the plastic film elongation step is not observed.

The reason why the plastic film elongation stage was not observed is that the crystal grains are refined by the HPT processing, the necking that is seen in ordinary ductile materials does not occur, and the structure of the fractured part becomes sticky as a whole and causes relative deformation. It is probable that it was caused to break.
Further, the behavior observed at 300 ° C. and 400 ° C. occurs in the vicinity of the indenter movement amount of 0.7 mm at both temperatures, and is therefore considered to be a phenomenon in the same process.
Here, since the behavior was not observed at 500 ° C., the recovery of dislocations was not sufficient at a temperature of 400 ° C. or lower, and the remaining dislocations acted as immobile dislocations in the stress field, resulting in early cracking. It is thought that it has been released and destroyed.
Then, as shown in FIG. 5, it is considered that the heat treatment up to 400 ° C. did not sufficiently reduce the hardness and the dislocations were not sufficiently recovered.
Therefore, it can be said that brittle fracture occurred at a temperature of 400 ° C. or lower.

Further, FIG. 14 shows a summary of SP absorbed energy ( _{ESP) at each temperature.}
As the temperature rises, the SP absorption energy of virgin V decreases, whereas that of HPT-V rapidly increases at 500 ° C.

In addition, FIG. 15 shows the appearance of each sample and the SEM image of the fracture surface obtained by SEM observation of the HPT-V material of FIGS. 11 to 13 after the SP test in vacuum.
In FIG. 15, (a), (b) and (c) are external appearances, (a)'(b)'(c)'are enlarged images of the fracture surface, and (a), (a)' are 300 ° C., (b). ) (B)'refers to the condition of 400 ° C., and (c) (c)' indicates the condition of 500 ° C.
Circumferential fracture occurred under all temperature conditions, and dimples found on ductile fracture surfaces were confirmed at 300 ° C and 400 ° C.
At 500 ° C., dimples were not as remarkable as in other temperature conditions, but since the SP curve reached the fourth stage, it is considered to be ductile fracture.

From the above consideration, the temperature condition of the in-situ SP test in a hydrogen atmosphere is relatively close to the SP curve of a typical ductile material, and the recovery of dislocations is considered to be sufficient at 500 ° C. I decided to do it.

(9-6) Measurement of pressure-concentration-isothermal line (PCT curve) in hydrogen atmosphere In this test, pressure-in a hydrogen atmosphere was used in order to quantitatively investigate the hydrogen dissolution characteristics of HPT-V using a Sieberts-type PCT measuring device. The concentration-isothermal line (PCT curve) was measured.

The measurement procedure is shown below.
The sample is attached to the device, evacuated at room temperature for 2 hours, loaded with 5 MPa hydrogen at 500 ° C. and held for 0.5 hours.
After that, the temperature is lowered to room temperature and kept as it is until the pressure stabilizes.
When the pressure stabilizes, raise the temperature to 500 ° C again and evacuate for 0.5 hours.
Then, a hydrogen load of 5 MPa is applied again, and the same procedure is repeated.
The activation process is performed by repeating this operation four times.
When the activation treatment is sufficiently completed, vacuuming is performed for 2 hours to extract the residual hydrogen, and the measurement is started at the measurement temperature.
Evacuate for 2 hours each time before measurement at each temperature.

In this test, measurements were taken in the order of 500 ° C, 400 ° C, and 300 ° C.
For comparison, PCT measurement was also performed in virgin V by the same procedure.

(9-7) Hydrogen Atmosphere In-Situ SP Test The results of the hydrogen atmosphere in-situ SP test are shown in FIG.
HPT-V shows a ductile SP curve even at a solid hydrogen concentration of 0.3 H / M, and although brittle fracture occurs at a solid hydrogen concentration of 0.4 H / M, it exceeds the elastic range. It can be seen that the fracture occurs after sufficient plastic deformation.
17, the vertical axis _{E SP} obtained from the results of FIG. 16 also shows DBTC curve taking solid solution hydrogen concentration on the horizontal axis.
While virgin V transitions from ductile to brittle fracture in the range of 0.2 to 0.3 H / M, HPT-V retains sufficient strength even at 0.4 H / M. It can be said that.
From these results, HPT processing is very useful for improving the hydrogen embrittlement resistance of V, and if it has the hydrogen permeation performance of V as it is, it can be used as a guideline for designing a high-performance hydrogen separation membrane. It can be said that HPT processing has great potential.
In addition, FIG. 18 shows the appearance of the sample and the SEM image of the fracture surface after the hydrogen atmosphere in-situ SP test.
In FIG. 18, (a) and (b) are appearances, (a)'(a) "(b)'(b)" are enlarged images of the fracture surface, and (a) (a)'(a) "are 0.3H / M, (b) (b)'(b) "refers to those under the condition of 0.4H / M.

At a solid solution hydrogen concentration of 0.3 H / M, many dimples were found on the ductile fracture surface, but on the other hand, there were some places where the river pattern was found on the brittle fracture surface.
At 0.4 H / M, the area where dimples were seen decreased and the area where the river pattern was seen increased.
This can be said to be a reasonable result because the region leading to brittle fracture increases as the hydrogen solid solution concentration increases.

(10) Hydrogen permeation test (10-1) Hydrogen permeation test device The hydrogen permeation test device used in this test was originally developed by the inventor, and the pressure control piping system outlined in FIG. 19 and FIG. 20 It has a membrane sample fixing portion schematically shown in 1.
The pressure control piping system is the same as the pressure control piping system used in the SP test, and is piped using valve fittings manufactured by Swagelok and various valves.
Further, a hydrogen introduction _{port is connected to the primary side (P 1} ) of the VCR surface seal joint, and a hydrogen discharge port is connected to the _{secondary side (P 2).}
This enables a hydrogen permeation test under stable hydrogen pressure if the valves in the piping system are properly operated.
Further, the permeated flux can be measured on the spot by using the mass fromer incorporated in the secondary side of the pipe.
The membrane sample fixing portion is incorporated in a hydrogen permeation test apparatus.
The hydrogen permeable film sample is sealed by sandwiching the sample with the tip of the VCR surface seal joint so that the sample is placed in the center of the VCR surface seal joint by fitting it into a ring made of SUS316 stainless steel. After fixing this between the joints, the membrane sample was heated to the set test temperature using a tubular electric furnace, and then a hydrogen permeation test was performed.

(10-2) Sample Preparation A disk-shaped sample of the product 1 of the present invention is cut into a circle with a diameter of 8.9 mm by a wire-cut electric discharge machine (WEDM), and then heat-treated at 500 ° C. for 1 hour to make the surface #. It was mirror-polished by emery wet polishing up to 1000 and _{buffing containing an Al 2} O ₃ suspension having an average particle size of 1 μm, and the final thickness was adjusted to about 0.5 mm.
Further, in order to impart hydrogen dissociation catalytic property, both sides of the sample were coated with Pd-25% Ag having a thickness of about 200 nm using an RF sputtering apparatus.
Although pure Pd can also be coated, even in the set temperature range (300 ° C to 500 ° C), there is a concern that the function will deteriorate due to hydride formation during hydrogen permeation at low temperatures, so the limit temperature for hydride formation will be lowered. Was coated with Pd-25% Ag.
For comparison, a sample was also prepared with virgin V, which is a comparative product 1 which was heat-treated at 1100 ° C. for 1 hour in the same procedure.

(10-3) Method of hydrogen permeation test In the hydrogen permeation test, hydrogen pressure is applied to each surface of the sample so that the primary side hydrogen pressure is equal to or higher than the secondary side hydrogen pressure, and the hydrogen flowing out to the low pressure side due to the pressure difference is This is a test to evaluate the performance as a hydrogen permeable film by measuring the volume flow rate.
The procedure for measuring hydrogen permeation performance is shown below.

The membrane sample is attached inside the VCR surface seal joint and evacuated.
After that, the membrane sample fixing part of the VCR surface seal joint is heated to the set temperature (300 ° C to 500 ° C) in a tubular electric furnace, the hydrogen pressure on the primary side is raised to 5 kPa, and the hydrogen pressure remains as it is until hydrogen permeation is performed. Hold in state.
After confirming that hydrogen had permeated, it was held for 1 hour, and the volumetric flow rate Q [sccm] in the secondary side pipe was measured with a mass flow meter installed in the pipeline to calculate the hydrogen permeation flux.
Then, the pressure on the primary side was changed to 10 kPa, 15 kPa, 10 kPa, and 5 kPa, and each was held for 1 hour to calculate the hydrogen permeation flux.

(10-4) Hydrogen permeation flux measurement method In this hydrogen permeation test, the amount of hydrogen permeation can be calculated from the change in the pressure inside the pipe with the passage of time of the hydrogen permeation reaction.
The details of the evaluation method will be described below.

P _1, P ₂ and P ₃ in the piping diagram of FIG. 19, pressure transducer du - Sa - (Strain gate - di-type pressure sensor) shows the respective pipe pressure, which is divided into three cells, respectively Can be measured.
Here, P ₁ is the pressure on the primary side (high pressure side) _{separating the membrane sample, P 2} is the pressure on the secondary side (low pressure side) separating the membrane sample, and P ₃ is the pressure in the low pressure container (reserve cell). It is a pressure sensor for measuring the pressure of.
By operating each valve connected to partition between these cells, the pressure inside the cells can be adjusted independently.

When the inside of the reserve cell is evacuated and the high-pressure cell and the low-pressure cell are _{filled with hydrogen gas so that P 1} > P ₂ , the pressure difference causes a difference in hydrogen concentration on both sides of the membrane sample, which is caused by diffusion in the metal. Hydrogen permeates.
When hydrogen begins to permeate, the pressure inside the low-pressure cell rises, so by opening the needle valve and allowing the hydrogen gas in the low-pressure cell to leak into the reserve cell, the pressure at P ₂ becomes constant. Adjust the opening / closing amount of the needle valve.
At the same time, the pressure in the high pressure cell is also adjusted to the target value by the pressure regulator.
As a result, the low pressure cell and the high pressure cell are always kept at a constant pressure according to any measurement condition, and only the pressure in the reserve cell rises with the passage of time.

The hydrogen permeation performance was evaluated by calculating the hydrogen permeation rate based on the result of the pressure change in the reserve cell or the mass flow meter reading, and then determining the hydrogen permeation flux.

(10-5) Results of hydrogen permeation test The results of the hydrogen permeation test at 300 ° C. are shown in FIG.
The graph shown in FIG. 21 is a graph comparing the effects of the pressure difference of the hydrogen permeability coefficient in the hydrogen permeation test. The hydrogen permeation coefficient is on the vertical axis and the primary hydrogen pressure is on the horizontal axis. ^{A certain HPT-V is 10-7} [mol H ₂ m ^-1 s ^-1 Pa- ^0, which is 2.5 to 5 times the hydrogen permeability coefficient of virgin V, which is the comparative product 1, even in a low temperature range of 300 ° C. ^It shows a high hydrogen permeability coefficient of about .5].
That is, the product 1 of the present invention shows a better value as a whole than the comparative product 1, shows a good value of one digit or more when the inlet value is 5, and 2.5 times as the inlet value becomes higher. It shows a good value of about 5 times.
However, in the hydrogen atmosphere in-situ SP test, it was clarified that there was no change in hydrogen solubility due to HPT processing, and the improvement in hydrogen permeation performance by HPT processing was contributed by the increase in diffusion coefficient depending on atomic diffusion. It is thought that it is.
Therefore, it is considered that the factor that increased the diffusion coefficient was that the grain boundaries increased by the HPT processing functioned as a high-speed diffusion path for hydrogen.

FIG. 22 is a diagram showing the appearance of each separation membrane after hydrogen permeation.
FIG. 23 is a diagram showing an SEM image of the surface of each separation membrane after hydrogen permeation. The upper row shows the surface of the separation membrane which is the comparative product 1, and the lower row shows the surface of the separation membrane which is the product 1 of the present invention.

In virgin V, which is the comparative product 1, wrinkles are generated due to film deformation due to expansion due to hydrogen addition pressure and solidified hydrogen, and the surface is cracked, whereas the product 1 of the present invention is used. It can be confirmed that the film deformation is suppressed in HPT-V, and wrinkles and cracks have not occurred.
That is, in the annealed material of Comparative Product 1, it was confirmed that slip lines and microscopic cracks were generated due to deformation due to hydrogen load, whereas in Product 1 of the present invention, slip lines and microscopic cracks were not generated. Not confirmed.
This indicates that the HPT processing suppressed the deformation in hydrogen, increased the strength, had a high deformation resistance in the stress field, and obtained appropriate hydrogen permeation and pressure resistance performance.

FIG. 24 is a graph comparing the permeability coefficients at each temperature, with the hydrogen permeability coefficient on the vertical axis and the reciprocal of the temperature on the horizontal axis.
Here, the maximum value of the primary hydrogen pressure was 15 kPa at 300 ° C., 20 kPa at 350 ° C., 60 kPa at 400 ° C., and 100 kPa at 450 ° C. and 500 ° C., and the hydrogen permeability coefficient was calculated.
HPT-V exhibits a hydrogen permeability coefficient higher than that of virgin V under temperature conditions of 400 ° C. or lower.
It can be seen that even at 500 ° C., hydrogen permeation is carried out, albeit slightly, as compared with the temperature condition of 400 ° C. or lower.

FIG. 25 is an EBSD orientation image of the sample before and after the permeation test at 500 ° C., (a) shows after the permeation test, and (b) shows before the permeation test.
The average particle size did not change before and after the test and was 390 nm, confirming that the microstructure was maintained.
From this result, at 500 ° C., the hydrogen permeation performance is lower than that at the temperature range of 400 ° C. or lower, but it functions as a hydrogen separation membrane for a long time while maintaining fine crystals even in a hydrogen atmosphere, and the function is also deteriorated due to coarsening. It can be thought that it will not happen.
That is, the product 1 of the present invention after the hydrogen permeation test at 500 ° C. for 3.6 ks for 1 hour has an average crystal grain size of 390 nm, maintains a fine structure, functions as a hydrogen separation membrane even at 500 ° C. for a long time, and is coarse. It was confirmed that the functional deterioration due to the change does not occur.

The average particle size and particle size number of the sample before and after the hydrogen permeation test are shown in the following table.

Both samples have a particle size number N of 20, and from the results of crystal structure observation, it is considered that they have fine particles having a particle size number of 20 or more even in a temperature range of 500 ° C. or less. As a guide, it is considered that the particles have a particle size number of 20 or more.

(11) PCT measurement The PCT curve of each measured sample is shown in FIG.
The PCT curves of HPT-V and virgin V overlap on the same line, and no effect of HPT processing on the PCT curve was observed.
Therefore, there is no change in hydrogen solubility in the sample before and after HPT processing, and it is considered that the contribution of HPT processing to the PCT curve is small.
From this result, it is considered that the influence of the grain boundary is small in the bcc metal such as V because the hydrogen solubility in the crystal lattice in the crystal grain is high.
In addition, Iwaoka et al. Reported that the hydrogen solubility of Pd, which is an fcc metal, is improved by HPT processing, but since Pd is an fcc metal, the hydrogen solubility in the crystal grains is small and the hydrogen solubility at the grain boundaries is relative. It is thought that the number is increasing.

(12) Production of Hydrogen Separation Membrane Sample Using Nickel Using the HPT processing device, a nickel drawing material having a purity of 99.9% is heat-treated at 1100 ° C. for 1 hour and then brought to room temperature. A sample of the hydrogen separation membrane, which is the product 2 of the present invention, was produced by performing HPT processing under the conditions of an applied pressure of 1.5 GPa, a rotation speed of 0.25 rpm, and a rotation speed of 5 rotations.
According to the equation expressing the strain at the time of HPT processing, the strain ε was about 32.6 to the maximum value of 169.
That is, for the region where hydrogen is permeated in the sample obtained by cutting out the 10 mm disc from the 20 mm disc, r indicating the distance from the sample center is a value in the range of 1.8 mm to 9.3 mm, so that the strain ε Was also calculated as a lower to upper limit value.
Then, when the mold groove depth is 0.7 mm in total, a disk-shaped nickel-based hydrogen separation membrane sample having a diameter of 20 mm and a thickness of 0.8 mm, which is the same as the mold hole diameter used for HPT processing, is obtained. It was.
Here, for comparison, an annealed material made of nickel having a purity of 99.9% and not subjected to HPT processing was formed into the same disk shape as the product 2 of the present invention to be used as the comparative product 2.

(13) Preparation of sample with nickel From the disc-shaped hydrogen separation membrane sample obtained from the product 2 of the present invention or the comparative product 2 obtained above, a wire-cut electric discharge machine (WEDM) was used to cut out into a desired shape for each test. A sample for use was prepared.

(14) Hydrogen permeation test For the product 2 of the present invention and the comparative product 2, the secondary hydrogen pressure was unified to 10 kPa for each of the three types of 600 kPa, 400 kPa, and 200 kPa under the temperature condition of 600 ° C. Then, a hydrogen permeation test was conducted.
The result is shown in FIG.
In FIG. 27, the left column shows the product 2 of the present invention, and the right column shows the comparative product 2.

(15) Consideration of test results Pure Ni, which belongs to the same group 10 as Pd, does not show a hydrogen permeation phenomenon in the annealed material according to Comparative Product 2, but the product 2 of the present invention subjected to HPT processing has a primary hydrogen pressure. Under the condition of 600 kPa, a hydrogen permeability coefficient Φ = 2 × 10 ^-9 (mol H ₂ m ^-1 s ^-1 Pa ^−0.5 ) was exhibited, and hydrogen permeation was confirmed.
From this result, it was confirmed that the hydrogen permeation performance of Ni, which is an fcc metal, is improved by performing HPT processing.
The reasons for this are that Ni is an fcc metal like Pd, and it is considered that the hydrogen solubility is improved by HPT processing, and that the grain boundaries of Ni also function as a high-speed diffusion path for hydrogen.

(16) EBSD orientation image The EBSD orientation image of the sample surface before hydrogen permeation (left) and after hydrogen permeation (right) is shown in FIG. 28.
The average particle size before hydrogen permeation was 320 nm, which was ultrafine crystal grains, but the average particle size after hydrogen permeation was 3.84 μm, confirming the coarsening of the crystal grains.

The average particle size and particle size number of the sample before and after the hydrogen permeation test are shown in the following table.

It is considered that the hydrogen permeation performance of the sample after hydrogen permeation is deteriorated because the crystal grains are coarsened.
However, since it was confirmed that the hydrogen permeation performance of the fcc metal was improved by the HPT processing, it was confirmed that the HPT processing is effective as a design guideline for a high-performance hydrogen separation membrane.
Further, also in the fcc metal, the hydrogen permeation performance is improved in the sample having the particle size number 21, and one measure for improving the hydrogen permeation performance is to have fine particles having a particle size number of 20 or more.

(17) Summary (17-1) By performing HPT processing, a submicron crystal grain structure can be obtained, and the ultrafine crystal grains maintain a fine crystal grain size even in a temperature environment of 500 ° C. ..
In addition, recrystallization was started in the temperature range of 500 to 700 ° C., and coarsening of crystal grains was confirmed at 700 ° C.
(17-2) By performing HPT processing, the crystal grains are refined and the tensile strength is increased to 1.0 GPa.
(17-3) In the annealing after HPT processing of V, the crystal grains become coarse after exceeding 500 ° C., and the hardness and tensile strength sharply decrease.
(17-4) High-density dislocations due to HPT processing were caused by a decrease in toughness in a temperature environment of 400 ° C. or lower, but dislocations are restored at 500 ° C., and have characteristics as a ductile material. Was there.
(17-5) In bcc metals such as V, the effect of HPT processing on the amount of hydrogen solid solution is small, and the grain boundaries function not as hydrogen trap sites but as diffusion paths for hydrogen atoms. ..
(17-6) By applying HPT processing, toughness is maintained even in the hydrogen solid solution region where brittle fracture occurred in virgin V in a hydrogen atmosphere, and ductile fracture is exhibited even at 0.4 H / M.
(17-7) It was confirmed that the hydrogen permeation performance of Ni, which is an fcc metal, was improved by performing HPT processing.
(17-8) It was confirmed that the effect on the hydrogen permeation performance by HPT processing can be obtained when the particle size number N = 20 or more.

1 HPT processing equipment

JP-A-2015-147208 Japanese Unexamined Patent Publication No. 2014-181344

Claims (4)

A method for producing a hydrogen separation membrane, which comprises performing a strong strain process for imparting a strain of ε = 10 or more to a thin film vanadium capable of selectively permeating hydrogen. The method for producing a hydrogen separation membrane according to claim 1, wherein the strong strain processing is a high-pressure twisting process. It is a vanadium film capable of selectively permeating hydrogen, and the number n of crystal grains in 1 mm ^{2 is measured two-dimensionally.}

A hydrogen separation membrane, characterized in that the particle size number N indicated by a rounded-up integer value is 18 or more. The hydrogen separation membrane according to claim 3, wherein the vanadium membrane is a vanadium membrane having a particle size number N = 20 and an average particle diameter d = 390 nm.

Images