Micro Resistance Spot Welding Optimization for Ultra-Thin Mo-Re Sheet Used in TWT

<inline-formula> <tex-math notation="LaTeX">$30~\mu $ </tex-math></inline-formula>m ultra-thin Mo-Re sheet had been joined by micro resistance spot welding (MRSW) and successfully applied to the low heating power cathode of traveling wave tube (TWT). The effects of different welding parameters, surface state and protection conditions on the joint’s appearance, micro-structure and lap shear strength (LSS) were analyzed in this paper. The results showed that a 63.5 N LSS of the MRSW joint was obtained under the following conditions: electrode force <inline-formula> <tex-math notation="LaTeX">$F=9.0$ </tex-math></inline-formula> N, welding current <inline-formula> <tex-math notation="LaTeX">$I=1000\text{A}$ </tex-math></inline-formula>, ramping time <inline-formula> <tex-math notation="LaTeX">$t_{1}=5$ </tex-math></inline-formula>ms, welding time <inline-formula> <tex-math notation="LaTeX">$t_{2}=2$ </tex-math></inline-formula>ms and dip in the alcohol for protection. The fracture occurs in the heat affected zone (HAZ). Using metallurgical microscopy observed the nugget, metallurgical fusion could be found, the diameter of nugget was about <inline-formula> <tex-math notation="LaTeX">$\Phi 0.3$ </tex-math></inline-formula> mm, without gas pores. Scanning electron microscopy (SEM) was used to observe the fracture surface. It could be found some “whisker” shape diverged from the joint center to the surroundings in HAZ. The electrode tip was observed by energy dispersive spectroscopy (EDS) and optical profiler. It was found that the tip was uneven and adhered the base metal obviously after spark. Additional work about electrode sticking will be required. The thermal test data showed that the use of ultra-thin Mo-Re sheet can reduce a space TWT cathode heating power (CHP) by 0.57 W, which will effectively save satellite energy and improve the expected working life of vacuum electronic devices.


I. INTRODUCTION
Molybdenum rhenium (Mo-Re) alloy has excellent hightemperature mechanical properties, high recrystallization temperature and resistivity for chemical corrosion, commonly used as vacuum electronic devices (VEDs) cathode supporting materials [1], superconducting thin films [2], [3], [4], advanced aero motor propulsion, nuclear fusion reactor [5], [6] and another extreme environment. Many kinds of materials will inevitably be customized with different shapes, connected with themselves or other materials during application. As an efficient and stable connection mode, different welding methods have been proposed and applied to the joint of Mo-Re sheet or wire, such as resistance welding [7], [8], [9], electron beam welding, laser beam The associate editor coordinating the review of this manuscript and approving it for publication was Mauro Gaggero .
welding [10] and vacuum brazing [11], and many reference data have been obtained. The welding difficulties of ultra-thin refractory alloy materials have already been summarized as follows: First, the smaller thickness of the material is, the less thermal inertia of the welding nugget area, which is harmful to the repeated heating of the material or the heating for a long time. It is necessary to accurately control the extremely short power on time. Second, it is difficult to form a sufficient volume nugget at the interface without damaging the outer surface, because refractory alloy has a high melting point, the base metal has small temperature difference between the inner and outer surface. Too small heat input will lead to nonwelding, and too large heat input will lead to sticking between the base metal and the electrode [10]. This sticking will lead to tearing of the previous weld nugget, or even tearing of the ultra-thin sheet. Third, molybdenum (Mo) is easy to recrystallize and hot brittle. Although a certain proportion of VOLUME 11, 2023 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ rhenium (Re) is added to improve the weldability, excessive heat input will still lead to significant reduction of material strength in the heat affected zone (HAZ). Fourth, the material surface is easy to form oxide film with high melting point. After the first spot is formed, it will have adverse effects on other adjacent and subsequent spots. In this article, the joining process with 30 µm thickness Mo-47.5Re sheet was carried out. With the increasingly strict requirements of space TWT on cathode heating power (CHP) [13], [14], [15], the thickness of Mo-Re sheet cylinder has also been reduced to 30-50 µm in order to improve thermal resistance and increase thermal insulation performance, while the traditional thickness is about 70-100 µm. When the cathode temperature is 1050 • C, thermodynamic analysis and test data showed that the CHP can be reduced about 0.5 W by using 0.03 mm in place of 0.05 mm, and more 0.5 W by using 0.01 mm in place of 0.03 mm. The mechanical strength of all these thicknesses is sufficient, but insufficient welding nugget strength will lead to changes in the relative positions of the cathode and the focusing electrode (FE), affect the stability of the electronic optical system (EOS), and even cause a short circuit between the cathode and the FE.
Micro resistance spot welding (MRSW) is one of the most commonly used methods in the manufacturing process for joining sheet metal together [16], [17], but reports of ultra-thin Mo-Re sheet are very rare. Further research is needed. Considering these difficulties from structural design and material characteristics, referring to the welding specifications of slightly thicker Mo-Re sheet, aluminum alloy, magnesium alloy, etc. [18], [19], [20], it is necessary to adopt high current, short time and low pressure, which is called hard specifications, for spot welding of ultra-thin Mo-Re sheet. The functions of the above three parameters are as follows: large current makes small resistivity generate enough heat, short time to prevent material grain coarsening caused by long time heating, low pressure keeps the contact resistance in a large range.
In this study, an ultra-thin Mo-47.5Re sheet MRSW method design is given in Section II. Section III shows the optimization process of MRSW parameters, and the nugget characteristics obtained by different parameters are analyzed. The MRSW reliability is verified by TWT electron gun vibration test and micro deformation measurement in Section IV, some follow-up research plans also listed in this Section. Finally, the last Section is the summary of the full text.

II. MRSW METHOD DESIGN
The experimental equipment adopts UNITEK DC25 MRSW machine (Energy storage welding machine, the welding time control accuracy is 0.01ms, the maximum welding current is 4000A), UNITEK THIN-LINE 80&88 pneumatic welding head (the welding pressure control range is 2.2-89N), and double self-made tungsten electrodes. The diameter of the negative electrode supported at the bottom is larger, and it is placed horizontally. For each spot welding, rotate a certain angle according to the direction marked in the Figure 1 for  avoiding too much sticking occur in the near area. Two layers of Mo-47.5Re sheet are stacked on it. The positive electrode rises and falls with the welding head in the vertical direction. The effects of different parameters and surface conditions on the appearance of the spot weld, the microstructure of the joint and the lap shear strength (LSS) properties were analyzed through comparative tests. Determined according to DIN 53283, the LSS is then measured with a tensile machine at a pull rate of 1 mm/min and expressed in MPa on an average of five specimens [21]. The design of the test samples was shown in Figure 2. The microcomputer controlled mechanical testing machine REGER RGM-4200 is used to LSS test. The tension sensor is CELREON STC-50kg.
At the beginning of the experiment, cut the Mo-Re sheet into small squares from the whole sheet. Then, immerse them in acetone solution to remove oil by ultrasonic wave for 30 minutes, and remove the surface oxide film in hydrogen furnace. The heat treatment process parameter is 600 • C hold for 10 min. The contact surfaces of two sheet shall be polished with a rasp or 240 # sandpaper to increase the contact resistance.
The mass fraction of each element of Mo-47.5Re is shown in Table 1. The main physical properties of Mo-47.5Re and other common metal materials are shown in Table 2. ''MRSW solderability W=100 * ρ/(λT)'' in the table is still an empirical formula, which aims to explain that Mo-47.5Re material has a not too bad weldability. The ultra-thin structure and high melting point are the direct difficulties facing at present. The welding method designed later will mainly solve these problems. The default welding parameters are shown in Table 3.

III. MRSW PARAMETERS OPTIMIZATION AND DISCUSS
The current, voltage, resistance and power curves obtained by the default parameters are shown in Figure 2.
According to the current curve, the ramp time only spent 3-5 ms, and the target value can be reached within the ramp    time. The climbing track of the output current is consistent with the preset curve, indicating that the output capacity of the welding machine meets the welding requirements of large current and short time.
With the increase of current on time, the base metal temperature increases, and the resistivity increases. For example, when the temperature of the base metal rises to 1400 • C, the electrical resistivity of Mo-Re is 84.0 µ ·cm, increase by 3 times of the normal temperature state. The resistance curve shows a steep rise trend in the early stage.
When the temperature rises to the softening temperature of Mo-Re, under the pressure of the electrodes, the junction area expands rapidly and the resistance decreases rapidly. When the temperature of the material increases further, metallurgical nuggets begin to appear at the interface, and the resistance decreases further. With the growth of nugget, the conductive area increases and the resistance keeps decreasing. When the nugget cross-section is equivalent to the diameter of electrode tip, it will stop growing, and the resistance will drop to a relatively stable range, which will not continue to decline with the extension of power on time. Based on the above theories and physical phenomena, optimize the MRSW parameters one by one.

A. EFFECT OF MRSW MAIN PARAMETERS
Based on the default parameters in Table 3, first study the effect of different welding currents on the LSS. The results are shown in Figure 4 (a). It can be seen from the figure that the LSS reaches the maximum (68.5 N) at 1100A. When the welding current is greater than 1100A, sticking was very easy to occur, as shown in Figure 4 (b). The electrode and the base metal are severely sticking, some formed weld nuggets would be torn, leaving a hole between the two layers of base metal, and the LSS also decreases significantly. In order to improve the welding stability and electrodes life, we continue to use 1000A for the next test. The LSS (63.5 N, 92.7% of the VOLUME 11, 2023 Second, the comparative test results of climbing time are shown in Figure 4 (c). When the climbing time within 5 ms, the longer the ramp time is, the stronger the welding nugget is. However, when the ramp time was set greater than 5 ms, the LSS decreased significantly. Through analysis, it can be considered that with the extension of ramp time, the contact surface softened, the interface resistance decreased significantly, and the large contact resistance at the beginning was not fully utilized to release enough heat, which ultimately leads to insufficient heat to melt the base metal.
Third, the comparative test results of welding time were shown in Figure 4 (d). We found 2ms was the best parameter. When it is less than 2ms, the heat input was insufficient, while more than 2ms, sticking was very easy to occur. Moreover, as shown in Figure 5, the samples after LSS test showed that the fracture occurred in HAZ, the material have defects due to excessive heat input, so the welding current should not be increased.
Last, the comparative test results of electrode force are shown in Figure 4 (e). According to the previous test results, the electrode sticking occurred only after being used dozens of times, which was a sign of insufficient pressure. If the pressure was reduced, sticking would be more likely to occur, but after the pressure was increased, although sticking was not easy to occur, LSS had significantly decreased. Therefore, according to the standard of MRSW, the default parameters, proposed in this paper, was an insufficient pressure process specification. These parameters would lead to excessive contact resistance between electrodes and base metal, which will damage the electrode while forming nuggets on the base metal contact surface. Additional work about electrode sticking will be required to extend electrode life.

B. INFLUENCE OF SURFACE STATE
Use TH2513 DC low resistance tester (DCLRT) to test the contact resistance. The pressure of the welding head was set to 9.0 N. When the two electrodes are directly short circuited, the DCLRT was set to zero. Place two clean Mo-Re sheet between the two electrodes, and the resistance was 25.6 m . After polishing the contact surface of two sheet skins with 240 # sandpaper, the resistance become to 40 m . Continue to use the rasp to grind out large scratches, and the resistance changed to 43 m .
It can be seen that the contact resistance of rough surfaces significantly increased. 240 # sandpaper or rasp grinding has no significant difference. According to the heating formula of resistance welding the contact area of two layers of Mo-Re sheet will generate more heat. The welding and LSS tests using the default parameters in Table 3 showed that the LSS of the sample without grinding was only 50.3 N, as shown in Figure 4(f), which was 20.8% lower than that of the sample after grinding.

C. CHARACTERIZATION MICROSTRUCTURE ANALYSIS
The formula for making metallographic corrosion solution is: 5% potassium ferricyanide (K3[Fe(CN)6]) solution and 10% sodium hydroxide (NaOH) solution, which should be prepared separately, mixed when using, cleaned and dried after wiping for 3-5 seconds. Observe the nugget obtained with welding current of 600A, 1000A and 1200A by an upright metallographic microscope Nikon ECLIPSE LV100. As shown in Figure 6. The nugget corresponding to 600A still has the original interface between two layers of base metal. Metallurgical fusion exists in the weld nugget corresponding to 1000A, and the diameter of the weld nugget is about 0.3 mm, the nugget height runs through two layers of base metal, and there are no gas pores in the nugget. While in [22] and [23], gas pores are inevitable. The shear strength σ c of the welding nugget is reached 75.8% of the base metal tensile strength (1180 MPa). [10] reported by using electron beam welding, an Mo-44.5 wt%Re alloy was welded successfully, with the welds having an ultimate strength of 821. 8-885.3 MPa, similar to the results in this paper. The strength of the weld was enhanced from 100.0 N to 113.0 N in [22], no more than 2 times of this paper (63.5N), while the thickness of the test specimen in [22] is 0.127 mm, 4 times of this paper (0.03 mm). Because the original grain size is equivalent to the thickness of the base metal, it is difficult to evaluate whether the grains in the nugget grow up. There is obvious electrode residue on the side near the upper electrode. The nugget VOLUME 11, 2023    corresponding to 1200A has severe recrystallization, and the nugget is clustered, with obvious cracks with the base metal.
JEOL JSM-6510 scanning electron microscope (SEM) was used to detect the torn welding nuggets, and it can be found that the appearance of ''whiskers'' spread around the center of the spot, as shown in Figure 7. Use the energy dispersive X-ray spectrometer (EDS) APOLLO PRIME added to the scanning electron microscope to analyze the metal components adhered to the electrode tip, and the results are shown in Figure 8. Use KEYENCE VHX-5000 optical profiler to conduct 3D imaging analysis the upper electrode tips, and the results are shown in Figure 9.
The mass fraction of Re was more than 5% in WC20 electrodes, which was not originally included. The analysis of surface morphology shows that the height difference of electrode tip reaches 316.19 µm. On the electrode tip, the base metal is adhered, and a new protrusion is formed. The    protrusion is about 0.25 mm, equivalent to the nugget diameter. We believe that this protrusion is growing with each spot, which will lead to the electrode tip becoming more and more irregular, smaller contact area between base metal and electrode, and larger resistance between them. Finally serious sticking will occur, tearing the base metal. Therefore, the use times of an upper electrode should not exceed 15 times.

IV. MRSW APPLICATION AND VERIFICATION IN ELECTRON GUN
As shown in Figure 10, the Mo-Re sheet was prepared into a cylinder and welded with the cathode and other supporting structures to complete the fabrication of TWT electron gun.
TWT has to face several mechanical vibrations loads during satellite launching and upper stage separation. Therefore, it is important to study the effect of vibration on electron gun [24], [25]. Then, place the electron gun on the vibration test instrument, shown in Figure 11, and set the target spectrum of random vibration to 26.4 grms, lasting for 3 min in axial and radial directions respectively. The target spectrum table parameters were shown in Table 4 and Figure 12. Before and after the vibration, an optical measuring machine (OMM) Nikon VMA2520 was used to detect the maximum change in horizontal direction of the cathode. The test results, as shown in Figure 13 and Table 5, showed the average of CHP reduced 0.57W, close to the ANSYS simulation result in Figure 14, and the deformation of both thicknesses is about 3-4 µm. This showed that the thermodynamics application verification test was passed after changing the material thickness.

V. CONCLUSION
For the MRSW of ultra-thin refractory alloy, a set of parameters was proposed in this paper. Under 1000 A welding current, 5 ms climbing time, 2 ms welding time, 9.0 N electrode force, 150 ms preloading time, 999 ms holding time after welding, and dig in the alcohol, an ideal joint without gas pores was obtained. The average of LSS was 63.5N. The above process parameters have the disadvantage of sacrificing electrode life, because the electrode tip was easy contaminated by the base metal, resulting in reduced contact area and excessive contact resistance, which was the main reason for sticking.
The new technology and new materials had been successfully applied to TWT electron gun, and the CHP had been reduced 0.57 W. In the future, it is still of great significance to continue exploring the rolling and welding of 0.02mm, even 0.01mm Mo-Re sheet. This has been proved to be an effective way to reduce CHP. VOLUME 11, 2023