RU2586960C1 - Method of measuring diffusion of hydrogen in titanium - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: invention relates to measurement equipment and can be used for determination of hydrogen diffusivity in various structural materials based on titanium, used in aerospace and nuclear engineering, in articles subjected to hydrogenation during operation. Analysed titanium sample is saturated with hydrogen inside electrolytic cell. One side of the sample is in contact with electrolyte and the second side is tightly pressed to Eddy current sensor of magnetic spectrometer. During sample saturation hydrogen diffuses to its opposite side. As a result readings of Eddy current sensor change. Using results of readings measurement time of Eddy current sensor calculate hydrogen diffusivity in titanium.

EFFECT: invention provides possibility of determining hydrogen diffusivity in titanium under production conditions in places inaccessible for analysis.

2 cl, 5 dwg

Description

The invention relates to the field of measurement technology and can be used to determine the diffusion coefficients of hydrogen in various structural materials, in space technology, nuclear energy, in products subjected to hydrogenation during operation. Measurement of the diffusion coefficients of hydrogen is of particular importance for the technology of newly created materials.

The following methods are known for measuring the diffusion coefficients of hydrogen in metals.

The diffusion coefficient is determined by the change in the moments of gravity arising in a hydrogen-saturated metal plate, which can rotate around an axis passing through the center of gravity. The plate is saturated with hydrogen on one side and balanced. Hydrogen extends along the plate from one end to the other. As a result, the center of gravity of the plate is shifted, in which hydrogen diffuses. The plate is rotated, and the diffusion process is judged by measuring the angle of rotation of the plate [USSR Author's Certificate No. 1636729, publ. 03/23/1991]. The method is characterized by low sensitivity.

The diffusion coefficient is determined by measuring the flow of diffusing hydrogen through a metal membrane with gas burettes, one of the surfaces of which (inlet) is in contact with the electrolyte solution [Mindyuk A.K., Svist E.I., Babey Yu.I. Installation for electrochemical studies of hydrogenation and hydrogen permeability of metals. - Protection of metals. - 1973. - T. 9. - No. 4. - S. 499-500]. The time for hydrogen to escape from the membrane depends on the sensitivity of the gas burette and is not fixed for small amounts of diffused hydrogen.

The diffusion coefficient of hydrogen is determined by the change in electrical resistance of steel [Beloglazoe S.M. Hydrogenation of steel during electrochemical processes. L., Publishing House Leningrad. University, 1975 .-- p. 27-36.]. The disadvantage of this method is that the diffusion coefficient is determined by the change in the total resistance of the hydrogenated sample, which, depending on the time of hydrogenation and the passage of hydrogen through the membrane, varies ambiguously, in a complicated way, and does not allow to accurately determine the time it takes to establish a steady flow of hydrogen through the membrane. There is conflicting information on this effect. For example, hydrogenation does not affect the electrical resistance of steel at all [Karpenko GV, Kripyakevich RI The effect of hydrogen on the properties of steel. M., Metallurgical Publishing House. 1962. - p. 198]. In the case of hydrogenation of palladium, the dependence of the electrical resistance on the hydrogen content has a complex form with a maximum and, therefore, does not allow us to uniquely identify the moment of saturation of the sample with hydrogen [Geld P.V., Ryabov R.A., Mokhracheva L.P. Hydrogen and physical properties of metals and alloys: Transition metal hydrides. - M .: Science. 1985. - p. 157].

The closest analogue selected as a prototype is the method described in [Grabovetskaya GP, Nikitenkov NN, Mishin IP, Dushkin IV, et al. Hydrogen diffusion in submicrocrystalline titanium // News of Tomsk Polytechnic University. - 2013. - T. 322. - No. 2. - S. 56-59]. In this method, the diffusion coefficient is determined by measuring the yield of hydrogen from a titanium membrane (sample) hydrogenated in an electrolytic cell using a mass spectrometer. The diffusion coefficients are calculated by the time-lag method according to the Barrera formula

$${\mathrm{<figure-callout\; id="3"\; label="t"\; filenames="00000005.png"\; state="\{\{state\}\}">t</figure-callout>}}_3={\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="00000005.png,00000007.png"\; state="\{\{state\}\}">h</figure-callout>}}^2/6D,\text{(one)}$$

where t ₃ is the time of establishment of a stationary flow of hydrogen through a metal membrane, h is the thickness of the metal membrane, D is the diffusion coefficient.

The main disadvantage of the prototype is that this method does not have sufficient sensitivity, requires the use of a vacuum system, bulky devices for measuring hydrogen concentration both at the entrance to the membrane (sample) and at the exit from the membrane (sample). The device used (mass spectrometer) is intended for scientific laboratory research, requires specially prepared devices, creates a vacuum and is unsuitable for operation in industrial conditions, for mass operational production control of hydrogen diffusion in parts that are in operation and exposed to hydrogen damage. In addition, during the diffusion of hydrogen in a metal, its uneven distribution over the volume of the metal occurs, caused by the heterogeneity of the metal structure [Nechaev Yu.S. Characteristics of hydride-like segregations of hydrogen at palladium dislocations // UFN, 2001. - T. 171 - No. 11. - S. 1252]. At high concentrations of hydrogen, metal hydrides are formed [Geld P.V., Ryabov R.A., Kodes E.S. Hydrogen and imperfections in the structure of metal. - M, Metallurgy, 1979. - S. 85-121]. The diffusion coefficient is affected by grain boundaries, material porosity, stresses, and dynamic loads [A.V. Gapontsev, V.V. Kondratyev. Hydrogen diffusion in disordered metals and alloys // UFN, 2003. - V. 173. - No. 10. - S. 1107-1129].

The task is to create a method for determining the diffusion coefficient of hydrogen in titanium under production conditions and places inaccessible for the analysis of hydrogen diffusing through a titanium sample.

To solve this problem in a method for measuring the diffusion coefficients of hydrogen in titanium, which includes, like the prototype, measuring the thickness of the test sample, saturating it with hydrogen in the electrolytic cell, fixing the start of the time of electrolysis and the time that hydrogen leaves the sample to the maximum mode, determining the difference between these values and the calculation of the diffusion coefficient of hydrogen in titanium by the Barrera formula, in contrast to the prototype, by the value of the resistivity and thickness of the sample, the eddy current frequency of the sensors is calculated a magnetic spectrometer, select the frequency of the eddy current sensor of the magnetic spectrometer so that the penetration depth of the eddy current exceeds the measured thickness of the sample by 3-4 times, install the eddy current sensor of the magnetic spectrometer on the sample membrane, and the time for hydrogen to exit from the sample to the maximum mode is fixed by voltage on the eddy current sensor.

It is advisable to determine the maximum voltage at the eddy current sensor during the hydrogenation of the sample, rotate the sensor around its axis by 180 degrees relative to the sample and fix the voltage every 7-8 degrees.

Theoretically, the problem of the distribution of eddy currents over the depth of a metal is described in [Lammeraner I., Stafl M. Eddy currents. Translation from Czech V.I. Dmitrieva. - M. - L., Energy. - 1967. - 208 s]. The eddy current penetrates the metal into a layer of thickness 8. The penetration depth 5 of the eddy current into the metal is a function of the electrical conductivity of the metal a, the eddy current frequency ω, the magnetic constant µ ₀ and the magnetic permeability of the metal µ.

$$\delta =\frac{\mathrm{one}}{\sqrt{\omega \mu {\mu }_0\sigma /2}}\text{(2)}$$

To determine the concentration of hydrogen in metals, eddy currents of various frequencies are used [Kalinin NP, Ostapenko V.D. Control of gas-saturated layers of titanium alloys by eddy currents of increased frequency // Defectoscopy. 1983. No. 5. S. 15-21 .; Leader A.M., Larionov V.V., Garanin G.V. The method for determining the hydrogen content in titanium. Patent No. 2498282 C1, IPC. G01N 27/02 - (2006.1) Bulletin No. 31 dated 11/10/2013. - S. 557-558.].

A change in the electrical conductivity of the metal leads to a change in the voltage at the eddy current sensor of the magnetic spectrometer. The magnitude of the electrical conductivity depends on the number of hydrogen atoms per titanium atom, on the direction of current propagation through the sample (membrane). This is caused by the appearance of defects during the hydrogenation. During diffusion through the sample (membrane), the appearance of hydrogen in its surface layer will be detected by the sensor if the frequency is chosen so that the penetration depth of the eddy current into the metal corresponds to the thickness of the sample. Here, the accumulation of hydrogen in the sample (membrane) during diffusion changes the readings of the sensor of the magnetic spectrometer. Thus, it is possible to record the time of voltage change at the eddy current sensor of the magnetic spectrometer, associated with the appearance of hydrogen as a result of diffusion through the sample.

In the inventive method, the eddy current sensor of the magnetic spectrometer is installed on the surface of the test sample of titanium, which is a cathode membrane located together with the anode in the electrolytic cell. The other side of the sample is in contact with the electrolyte solution. The thickness of the sample is measured and the frequency of the eddy current sensor is calculated by the formula (2) and its value is set on the sensor. Turn on the eddy current sensor and, turning it around the vertical axis, find the maximum voltage value. The voltage on the electrolytic cell is turned on and the on time is recorded. Hydrogen accumulates on the surface of the sample (membrane) from the side of the electrolytic cell during electrolysis. It diffuses through the sample to another surface of the sample. As soon as hydrogen reaches the surface layer from the side of the sample opposite to the cell, the eddy current sensor changes to stop. The period of time during which a change in the readings of the eddy current sensor of the magnetic spectrometer was recorded. Due to the fact that during the diffusion process defects are formed in the sample that affect the propagation of the eddy current, the sensor is rotated around the axis by 180 degrees with a frequency of once in 5-7 minutes. The value of the time interval for which the sensor readings change is substituted into the Barrera formula (1) and the coefficient of hydrogen diffusion through the sample membrane is calculated.

In FIG. 1 shows a diagram for measuring the diffusion coefficient of hydrogen through a titanium sample 1 (membrane) using an eddy current sensor 2 and a voltage meter 3 on an eddy current sensor 2.

In FIG. Figure 2 shows the voltage dependence of the eddy current sensor on the time of saturation of the sample with hydrogen.

In FIG. 3 shows the angular location of the eddy current device sensor relative to the titanium sample, 1 - sample, 2 - eddy current sensor of the device.

In FIG. 4 shows microcracks (a) and craters (b) on the surface of a titanium sample during hydrogenation.

In FIG. 5 shows a diffractogram of a hydrogenated sample. The numbers show the time (min) the saturation of titanium with hydrogen.

Sample 1 is a membrane - part of the wall of the electrolytic cell 4, in which electrolysis proceeds. The electrolytic cell 4, contains an electrolyte 5, in which are located connected to the power source 6 cells, anode 7 and sample 1 - the cathode.

The measurement of the diffusion coefficient of hydrogen through a sample 1 of titanium is carried out according to the following algorithm: measure the thickness h of sample 1, using the table value of the resistivity of titanium and the value of h, find the frequency of the eddy current. This frequency value is adjusted downward so that the eddy current penetration depth δ is 3-4 times greater than the measured value of h. The adjusted frequency value is set on a magnetic spectrometer and a magnetic spectrometer is turned on. The eddy current sensor 2 is rotated 180 degrees around its axis relative to sample 1 every 10-12 degrees (Fig. 3), the voltage on the eddy current sensor 2 is measured and its maximum value is found. Next, a solution of 0.1 M sulfuric acid is poured into the electrolytic cell 4, the power supply 6 of the electrolytic cell 4 is turned on (Fig. 1). Set the electrolysis current density equal to 1 A · cm ^-2 . The time of switching on the electrolysis current t _{1 is} recorded. Hydrogen enters the titanium sample 1 (membrane), which is the cathode, as a result of electrolysis, which diffuses through sample 1. The eddy current sensor 2 of the magnetic spectrometer 3 is constantly measured in time, turning the sensor around its axis by an angle φ equal to 5-7 ° to 180 degrees every 5-7 minutes. The time t _{2 is} recorded at which the eddy-current voltage values reach the maximum mode (Fig. 2, marked with a dash).

The figure 2 presents the dependence of the readings of the eddy current sensor 2 from the time of electrolysis and, where dashes indicate time points. Find the difference t ₂ -t ₁ = t ₃ and calculate the diffusion coefficient D of hydrogen according to the Barrera formula (1) D = h ² / 6t ₃ , where t ₃ is the time of establishment of a steady flow of hydrogen through titanium sample 1, h is the thickness of titanium sample 1 .

A specific example of the implementation of the method for determining the diffusion coefficient of hydrogen in titanium.

м²с^-1. Погрешность определения коэффициента диффузии зависит от точности фиксации времени достижения стационарного режима показаний вихретокового датчика и колеблется от 3% до 5%.Sample 1 was cut out of titanium foil with a thickness of 48 μm. Using the formula (2), the frequency of the eddy current sensor of the magnetic spectrometer was calculated. As a magnetic spectrometer, a 3MA device was used (manufactured in Germany, Saarbrücken). It is equal to 49 MHz. The frequency should be such that the penetration depth of the eddy current δ exceeds the membrane thickness by at least 3 times (according to the physical meaning of the magnitude of the eddy current penetration into the metal). For example, for VT1-0 titanium, with its specific resistance of 46 · 10 ^-8 Ohm · m and a penetration depth of 250 μm (this is 3 times the sample used here), the frequency of the eddy current installed on the sensor will be 1 MHz. A 0.1 M solution of sulfuric acid was poured into the electrolytic cell 4. Set the calculated 1 MHz frequency on the sensor. One side of sample 1 was in contact with the electrolyte, and the eddy current sensor of magnetic spectrometer 2 was pressed tightly to the second. The eddy current sensor of magnetic spectrometer 2 was turned on, the sensor was rotated around its axis by 180 ° degrees and every 10-12 degrees the maximum initial voltage value of the eddy current sensor was fixed . A constant voltage was applied to anode 7 and cathode – sample 1 (membrane) from a power supply 9 DC SUPPLY HY 3002 and a current density of 1 A cm ^{– 2 was set} . The time of turning on the current of the electrolytic cell t _{1 was} recorded and the eddy current sensor was monitored by turning it around the axis every 5-7 minutes. The moment of time was fixed when the reading of the magnetic spectrometer sensor became constant and maximum in angle and time t ₂ = 300 min and was recorded in the table. The difference in values of t ₂ -t ₁ = t _{3 was} substituted into the formula (1). t ₃ = t ₂ -t ₁ = 300 min. $$D=\frac{{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="00000005.png,00000007.png"\; state="\{\{state\}\}">h</figure-callout>}}^2}{6{\mathrm{<figure-callout\; id="3"\; label="t"\; filenames="00000005.png"\; state="\{\{state\}\}">t</figure-callout>}}_3}=2.13\cdot {10}^{-\mathrm{fourteen}}$$

m ² s ^-1 . The error in determining the diffusion coefficient depends on the accuracy of fixing the time to reach the stationary mode of eddy current sensor readings and ranges from 3% to 5%.

, где Δh и Δt - погрешность определения толщины образца и времени диффузии водорода через образец. Автоматизированная фиксация времени резко увеличивает точность измерения времени t₂.Error calculation is carried out according to the formula $$\Delta D/D=\sqrt{\mathrm{four}{\left(\frac{\Delta h}{h}\right)}^2+{\left(\frac{\Delta t}{t}\right)}^2}$$

, where Δh and Δt are the error in determining the thickness of the sample and the time of hydrogen diffusion through the sample. Automated time recording dramatically increases the accuracy of measuring time t ₂ .

The eddy current sensor of the magnetic spectrometer is calibrated using well-known ARMI standards (IARM 178 V standard: Ti-6Al-6V-2Sn / UNS R56620), as well as samples from reference copper.

A control experiment was carried out: when the sample was weighed after 340 min, stabilization of the sample mass was observed, i.e. a further increase in electrolysis time does not lead to an increase in the mass of hydrogen of the incoming sample. Then, hydrogen begins to leave the sample (there is a slight decrease in the mass of the sample with a simultaneous decrease in the voltage at the magnetic spectrometer sensor). This indicates the accuracy of the eddy current measurement.

The readings of the eddy current sensor depend on its position relative to the sample; therefore, it must be rotated, as shown in FIG. 3. The need for this action is due to the presence (Fig. 4) of a pronounced heterogeneity of the location of microcracks in titanium in shape, direction and depth caused by hydrogenation of titanium. The heterogeneity is greater, the greater the degree of hydrogenation of titanium. In hydrogenated titanium, titanium hydrides TiH _{n are formed} with a different number of hydrogen atoms n in the hydride molecule depending on the amount of hydrogen present in the metal, which leads to a change in the electrical conductivity along the depth of titanium. This is confirmed by studies of diffraction patterns on a PDIFF Beamline diffractometer. An example of a diffraction pattern is given in FIG. 5, which shows the formation of TiH ₂ hydrides. The numbers on each curve indicate the time in minutes of saturation of titanium with hydrogen. The diffraction pattern shows that with an increase in the time of movement (diffusion) of hydrogen through the sample from titanium, the peaks corresponding to the content of titanium hydrides in the sample increase. This indicates the formation of defects that differ from each other in the direction of location relative to the sample, the size of the defect, both in length and in its transverse dimensions.

The invention can be used to determine the diffusion coefficients of hydrogen in various structural materials based on titanium, used in space and nuclear technology, in products subjected to hydrogenation during operation. In addition, the eddy current sensor of the magnetic spectrometer can be removed from the measuring equipment at large safe distances for the operating personnel and located in hard-to-reach places.

Claims (2)

1. A method of measuring the diffusion coefficients of hydrogen in titanium, including measuring the thickness of the test sample, saturating it with hydrogen in the electrolytic cell, fixing the start of the time of electrolysis and the time that hydrogen leaves the sample to the maximum mode, determining the difference between these values and calculating the diffusion coefficient of hydrogen in titanium using the formula Barrera, characterized in that the eddy current frequency of the sensor of the magnetic spectrometer is calculated by the value of the resistivity and thickness of the sample, the vortex frequency is selected a retokovy sensor of a magnetic spectrometer such that the penetration depth of the eddy current exceeds the measured thickness of the sample by 3-4 times, install the eddy current sensor of the magnetic spectrometer on the sample, and the time of hydrogen leaving the sample to the maximum mode is fixed by the magnitude of the voltage at the eddy current sensor. 2. The method according to p. 1, characterized in that during the hydrogenation of the sample to determine the maximum voltage at the eddy current sensor, the sensor is rotated around its axis by 180 degrees relative to the sample and the voltage value is fixed every 7-8 degrees.

Images