2-2- TLP bonding process experiments
The TLP process was carried out in a vacuum furnace of 3?10?5 Torr. In order to restrict localized liquation of the substrate and subsequent solidification during holding times, the bonding temperature was selected 1070 °C which is lower than eutectic temperature of the binary NiB alloy (10801140 °C) [10,11]. The bonding time was varied from 5 to 640 min for evaluation the effect of time on the quality of the joints. The heating rate was 15 °C min?1, and the specimens were finally furnace cooled to room temperature in vacuum.
2-3- Microstructural characterization
Bonded specimens were sectioned perpendicular to the bond using EDM and then they were polished using SiC paper followed with a diamond abrasive step (9, 6, and 1 ?m). Final polishing was carried out using basic colloidal silica slurries followed by a brief vibratory polish. For microstructural examinations, the specimens were etched using 10 mL HNO3 + 10 mL C2H4O2 + 15 mL HCl. Microstructural observations and Semi-quantitative chemical analyses of phases were made on polished and etched specimens using a combination of scanning electron microscopy (SEM, KEYENCE VE-8800) using an angle selective backscattered electron detector in Ultra-55 (Carl Zeiss) and EPMA (Shimadzu EPMA-1720). The EPMA system was equipped with an ultra-thin window wavelength dispersive X-ray spectrometer (WDX) in order to detect light elements such as boron and carbon accurately. Hitachi SEM equipped with EBSD also was used for phase identification. An image analysis system (Clemex, Vision Pro. Ver. 3.5.025) was used to determine the quantitative data on the microstructures.
2-4- Microhardness and creep testing
Vickers Microhardness test was used to determine the hardness profile across the joints area, using a 50 g load and according to ASTM: E384 . The hardness value reported for each point was an average of at least five measurements. The stress rupture tests were conducted in air at 816 °C and 110 MPa. All specimens were cut out of the bonded samples using EDM according to the configuration shown in Fig. 2.
3- Results and discussion
Fig. 2 shows the microstructure of TLP bonded hastelloy X using Ni-Cr-B-Si-Fe interlayer at 1070 °C for 5, 80, 320, and 640 min. Three distinct microstructural zones of ISZ, ASZ, and DAZ can be discriminated in the joint area of the samples bonded for 5 and 80 min. The joints were solidified isothermally at the holding time of 320 and 640 min and no ASZ were detected. High magnification section of the ASZ and DAZ are separated below each image.
EPMA microanalysis revealed that the DAZ comprised (Mo, Cr)-rich borides. As shown in fig. 2, the density of borides decreases and they show different morphologies at different holding times. ISZ composed of ? solid solution phase which is formed as a result of isothermal solidification at the solid/liquid interface towards the center of the joint. According to SEM micrographs of Figs. 2 (a) and (b) and using the spot analysis of EPMA, ASZ composed of a Ni-Si eutectic phase, Ni3B and Ni2Si. These results are confirmed by the XRD pattern of the sample bonded for 5 min extracted from the fracture surface of the stress rupture test (Fig. 3). Ni solid solution, Ni3B and Ni2Si peaks are detected in XRD pattern which are in agreement with the EPMA results and other researchers findings [13,14]. If the holding time is not enough to complete isothermal solidification, solidification would be completed by cooling and residual liquid phase transforms into eutectic components in the centerline of the joint. As shown in high magnification SEM of the ASZ in Fig. 2(b), there are a number of cracks close to these hard and brittle phases. These shrinkage type cracks are a consequence of athermally solidification.