|NDT.net - February 2003, Vol. 8 No.2|
Ultrasonic velocity measurements have been carried out in Ni base superalloy Inconel 625 alloy in
The ultrasonic wave propagation is influenced by the microstructure of the material, through which it propagates. The velocity of the ultrasonic wave depends upon the elastic moduli and the density of the material, which in turn are mainly governed by the amount of various phases present and the damage to the material. Further, these microstructural features also govern the mechanical properties of the materials. Hence, a possibility exists for the assessment of the mechanical properties, through the ultrasonic measurements. Ultrasonic velocity in a precipitation hardened alloy is governed by the changes in the velocity in the matrix due to changed composition (due to solutes coming out of the matrix during precipitation), velocity in precipitates and velocity in precipitate-matrix interface. Hence the ultrasonic velocity monitored during precipitation hardening process would provide insight into the kinetics of precipitate formation and their growth. In Nimonic alloy PE 16, Jayakumar et al. (1, 2) correlated the ultrasonic shear and longitudinal velocities with the variations in density and elastic moduli due to occurrence of secondary phases. The velocities increased linearly with increase in the volume fraction of g' precipitates in this alloy. In 9Cr-1Mo ferritic steel, Anish Kumar et al. (3) have found that the ultrasonic velocity of the normalized and tempered alloy increased with the precipitation of fine Cr2N and Fe2Mo precipitates and decreased with their coarsening and dissolution. Ultrasonic parameters have been used extensively for the characterization of precipitation behaviour in aluminum alloys (4, 5). The variation in ultrasonic velocity with ageing time exhibited various peaks corresponding to the precipitation of different phases in 2219 aluminum alloy (4). The correlation of ultrasonic velocity with hardness exhibited an increase in hardness with increase in ultrasonic velocity followed by a decrease in hardness with further increase in the velocity in 2024 aluminum alloy (5). Precipitation of any phase should either lead to a monotonic increase or decrease of hardness with respect to the ultrasonic velocity. But in the case of 2024 aluminum alloy, the non-monotonic behavior may be attributed to the precipitation of different phases, which may have different effect on the elastic (ultrasonic velocity) and plastic (hardness) behaviour of the material. In none of the earlier studies, the effect of individual precipitate on the correlation of ultrasonic velocity with mechanical property has been studied. In the present work, we have studied the effect of various precipitates, generated by giving suitable heat treatments to the service exposed and re-solution annealed alloy, on the correlation of ultrasonic velocity with yield stress in Nickel base alloy Inconel 625.
Bars of size 9 mm x 10 mm x 60 mm were cut from the service exposed (~873 K for ~60000 h) Inconel 625 tube of 11 mm diameter and given suitable heat treatments to dissolve various types of precipitates preferentially and to precipitate out the new phases. Table 1 gives the details of the heat treatments employed. The bars from service exposed materials were heat treated at different temperatures (923 K, 1023 K and 1123 K) for durations upto 500 h. Also, one set of the bars was solution annealed to dissolve all the precipitates followed by thermal ageing at 923 and 1123 K to precipitate out different phases. Standard tensile specimens of 4 mm diameter and 25 mm gauge length were tested in air at room temperature (298 K) at a constant strain rate of 3.2x10-4 s-1. Detail of the microstructural characterization studies, carried out using optical microscope and Transmission Electron Microscope (TEM), is dealt elsewhere (6). Specimens, used for optical microscopy, were surface ground to obtain plane parallelism to an accuracy of ± 3 µm to enable precise ultrasonic velocity measurements. Fig. 1 shows the schematic of the experimental setup for ultrasonic velocity measurements. 100 MHz broad band pulser- receiver (M/s. Accutron, USA) and 500 MHz digitizing oscilloscope (Tektronix TDS524) were used for carrying out the ultrasonic measurements. For these measurements, the rf signals were digitized at 500 MHz and the gated backwall echoes (1024 ns duration) from the oscilloscope were transferred to the personal computer with the help of GPIB interfacing and LabVIEW software. Ultrasonic longitudinal wave velocity was measured using 15 MHz longitudinal beam transducer. Cross correlation technique has been used for precise velocity measurements. The accuracy in time of flight measurement was better than 1 ns and the maximum scatter in the ultrasonic longitudinal wave velocity was ± 2.5 m/s.
|Fig 1: Schematic of the experimental setup for ultrasonic velocity measurements.|
Table 1 also gives the details of the precipitates present after different heat treatment conditions. The service exposed alloy exhibited an extensive precipitation of intragranular intermetallic phases g'' and Ni2(Cr, Mo) and grain boundary carbides (6). Short duration aging of service exposed alloy at 923 K upto 10 h led to the dissolution of the Ni2(Cr,Mo) phase. Ageing the service exposed alloy to longer durations (500h) at 923 K led to the precipitation of the d (Ni3(Nb, Mo)) phase in the form of elongated needles. Ageing at intermediate temperature (1023 K) exhibited faster rate of dissolution of the g" and Ni2(Cr,Mo) precipitates as compared to that at 923 K. Precipitation of the d phase also occurred much earlier at 1023 K than that at 923K. Ageing the service-exposed alloy at 1123 K for 1 h resulted in the complete dissolution of both g'' and Ni2(Cr,Mo) precipitates.Precipitation of d phase occurred at 1123 K upon further ageing and its volume fraction increased with ageing time. Carbides were observed at the grain boundaries during all the temperatures of treatment. The re-solution annealing at 1423 K for 0.5 h dissolved not only the intermetallic precipitates but also the grain boundary carbides that were earlier observed in all the heat treated specimens. Ageing the re-solution annealed alloy led to the precipitation of g'' phase at 923 K and d phase at 1123 K.
|Heat treatment||Precipitates present||Yield stress|
|Ultrasonic longitudinal wave velocity (m/s)|
|Service exposed (SE)||Ni2(Cr, Mo), g" and grain boundary carbides||1013||5904|
|SE + 923 K, 1 h-200 h||g" and grain boundary carbides||842-803||5865-5858|
|SE + 1023 K, 1 h||Coarsened g" and grain boundary carbides||553||5849|
|SE + 1023 K, 100 h, 200 h, 500 h||Coarsened g", d and grain boundary carbides||603, 620, 584||5847, 5854, 5860|
|SE + 1123 K, 1 h||grain boundary carbides||410||5832|
|SE + 1123 K, 10 h- 500 h||grain boundary carbides + d||426-541||5836-5860|
|Re-solution annealed (RSA)||__||370||5833|
|RSA + 923 K/ 200 h, 500 h||g"||750, 766||5854|
|RSA + 1123 K/ 200 h, 500 h||d||440, 454||5854, 5864|
|Table 1: Details of the heat treatment employed to the service exposed alloy and corresponding precipitates present, yield stress and ultrasonic velocity.Heat treatment Precipitates present Yield stress.|
Figure 2 and Table 1 show the variation in yield stress (YS) and ultrasonic velocity (UV) with ageing time for the service-exposed alloy. It was observed that the service exposed alloy had a very high YS value of ~ 1013 MPa and UV of 5904 m/s as compared to that prior to service, which has YS value of ~ 375 MPa and UV of 5830 m/s. The high yield strength of the service exposed alloy is attributed mainly to the precipitation hardening derived from the intragranular g'' and Ni2(Cr, Mo) precipitates. Subjecting the service-exposed alloy to temperatures higher than the service temperature, led to a decrease in the YS value due to the dissolution of g'' and Ni2(Cr, Mo) precipitates. The extent of reduction in the YS value of the service exposed alloy increased with the increase in the ageing temperature. The largest decrease in the YS value (605 MPa decrease) during 1h ageing at 1123K resulted from the complete dissolution of both the intermetallic phases. Comparatively smaller decrease in YS (170 MPa decrease) after ageing at 923 K for 1 h was due to the dissolution of only the Ni2(Cr, Mo) phase. It can be seen that the YS decreased continuously with ageing at 923 K up to 200 h, indicating a progressive dissolution of Ni2 (Cr, Mo) precipitates. The precipitation of the d phase at longer durations caused increase in YS. While the d phase precipitated at 923 K only after 200 h ageing, it was found to precipitate much earlier at higher temperatures. The re-solution annealing at 1423 K for 0.5 h decreased the yield stress to 370 MPa (40 MPa less than that for the specimen heat treated at 1123 K for 1 h having only grain boundary precipitates). The study showed that the YS is primarily affected by intermetallic precipitates and less by the grain boundary carbides.
The variation in UV with ageing time also exhibited similar behaviour as that of YS. The service exposed alloy exhibited highest ultrasonic velocity, which is attributed to the presence of the intermetallic phases. Ultrasonic velocity decreased initially with ageing time at all the temperatures. The decrease in the velocity was found to be minimum at the lowest temperature of ageing (923 K) due to the dissolution of only Ni2(Cr, Mo) type of precipitates. Maximum decrease in ultrasonic velocity after 1h ageing at 1123 K (5832 m/s) resulted from the complete dissolution of both g'' and Ni2(Cr, Mo) precipitates. Even after the re-solution annealing at 1423 K for 0.5 h, ultrasonic velocity was similar to that in the specimen heat treated at 1123 K for 0.5 h. The only microstructural difference in these two heat treated conditions is the absence of grain boundary carbides in solution annealed specimen. This showed that the ultrasonic velocity is affected by the intermetallic precipitates only and not by the grain boundary carbides in Inconel 625 alloy.
|Fig 2: Variation in yield stress and ultrasonic longitudinal wave velocity with ageing time for service exposed (SE) alloy.|
Figure 3 and Table 2 show the variation in yield stress (YS) and ultrasonic velocity (UV) with ageing time for the re-solution annealed alloy. Both the YS and UV increased with ageing time at both the temperatures of ageing (923 K and 1123 K). Further, it can also be seen from Fig. 3 that the variation in UV in the RSA alloy with ageing at 923 K and 1123 K are almost similar, whereas YS exhibited much larger variation with ageing time at 923 K than at 1123 K. This is attributed to the different extents of effects of various precipitates on YS and UV. The precipitation of fine and coherent g'' at 923 K leads to higher increase in YS as compared to the precipitation of d at 1123 K. As the UV depends mainly on the Young’s modulus and density of the alloy, and not on the degree of coherency, fineness and distribution of the precipitates, this effect could not be seen in the variation of ultrasonic velocity. Hence it is clear that the type of precipitate plays a great role in the correlation of nondestructive UV and mechanical properties, such as YS. Therefore a single relationship may not be sufficient to show the correlation between destructively determined mechanical properties, such as YS and non-destructive ultrasonic velocity, and hence microstructural state must be accounted first before attempting a one to one correlation between these two parameters.
|Fig 3: Variation in ultrasonic velocity with ageing time for re-solution annealed (RSA) alloy.|
The present study revealed that the effect of different precipitates on the mechanical properties and ultrasonic velocity are quite different. Yield stress is primarily affected by intermetallic precipitates and less by the grain boundary carbides, whereas, ultrasonic velocity is affected by the intermetallic precipitates only and not by the grain boundary carbides in Inconel 625 alloy. Further, different intermetallic precipitates also affect the yield stress and ultrasonic velocity differently and hence the type of the precipitate plays an important role in correlating these two material properties.
We are thankful to Prof. K. K. Ray, Indian Institute of Technology, Kharagpur for useful discussions. We are also thankful to Dr. S.L.Mannan, Associate Director, MDG and Mr. P.Kalyanasundaram, Head, DPEND for their cooperation.
|© NDT.net - email@example.com|||Top||