Abstract
In this investigation, three different consolidation ways for Inconel 718 MIM samples are compared. The conventional sintering in a furnace is compared to field assisted hot pressing (FAHP) and microwave sintering. By accelerating the sintering kinetics, the thermal behaviour may be modified. Hence, the behaviour of the Inconel 718 sintered by field assisted and microwave sintering has been investigated. The sintered samples were all injected from a feedstock composed of a fine particle Inconel powder and a binder principally composed of cellulose acetate butyrate and polyethylene glycol. The effects of the different process on the microstructure and the mechanical properties are then compared. There was no difference in distribution of pores between the conventional and the FAHP sintering. The microwave sintering of a metal injection moulding brought defects during the test and lower properties were obtained.
This paper is part of a special issue on the Advances in Materials and Processing Technologies (AMPT) 2015 and has subsequently been revised and extended before publication in Powder Metallurgy.
Introduction
Nickel chrome super alloys exhibit good mechanical strength and high resistance to creep at high temperatures in addition to a good level of corrosion resistance [1]. Thus, they can operate in combined conditions of strength and temperature under the influence of a corrosive media. This kind of material fulfils most of the requirements requested in aerospace and power generation industries. The Inconel 718 super alloy represents 70% of the world production of Nickel-based super alloys thanks to it good mechanical and chemical properties for a reasonable cost. The problems, using this kind of material, is the difficulty to machine due to its ductility, imposing dimensional limits and strong loss of material in order to obtain the desired component [2].
Metal injection moulding (MIM) relies on shaping metal particles, followed by a debinding and sintering process. This method combines the advantages of PM and the flexibility of the thermoplastic moulding process. The final product is nearly fully densified and with mechanical characteristics close to the machined product [3]. The high production quantities of complex shaped component coupled with low loss of material makes this method competitive with the other methods [4].
There are three steps in the MIM process: the formulation of a feedstock by mixing of a metallic powder with a polymeric binder, the moulding of that feedstock by injection into tooling with a determined pressure, and finally the solvent and thermal processing of the shaped sample. In the case of Inconel 718, the sintering is followed by an aging treatment in order to obtain the best performance of the alloy [5]. This method consumes a lot of time and power. In order to improve these parameters, new ways of sintering have been investigated.
The field assisted hot pressing (FAHP) is a method inspired in the conventional hot pressing. The powder is introduced into a graphite die (punches and matrices) and submitted under pressure and crossed by an electric current of high intensity. The use of a uni-axial pressure combined with the electric pulses accelerates the sintering kinetics, hence obtaining in a short period of time near fully dense components [6]. In this way, it is also limiting the grain growth and so, the development of fine microstructures bringing sometimes to better performances [7].
The microwave sintering has been used essentially on ceramics and composites materials since the beginning of the 70's. It was established that the heating of the sample was induced by the movement of the mobiles (ions, dipoles) inside the materials created by the microwaves [8]. The heating rate is dependent of the microwave electric field penetration. Janney et al. [9] sintered alumina by this method and showed an improvement of the kinetic of sintering and a reduction of the sintering temperature of 300°C.
The microwaves sintering is a process appreciated for industrial applications because it is allowing reducing the times of sintering and the energy used. Unfortunately, the metallic materials are reflecting most of the microwaves, that is why this set-up was investigated only recently [10,11]. In the case of a metallic powder, it was proven that the microwaves were able to penetrate the material by the porosities [12]. By increasing the heating rate, this method has also an effect on the grain size distribution and so the mechanical properties [13].
This study is based on the comparison between conventional FAHP and microwave sintering of Inconel 718 MIM samples. There is no information about the processing of Inconel 718 through these later two new methods. The first step is to optimise the sintering parameters of these two methods by finding the best density while keeping the advantages of the MIM process with a simple geometry. The microstructures in function of the sintering temperature will allow comparing the behaviour of the material at different heating rates. The effect of this parameter is directly impacting the grain size and so the mechanical properties, so the hardness of the samples is compared.
Experimental methods
Powder and binder
The Inconel 718 powder is provided by Sandvik Osprey Ltd and its chemical composition is presented in Table 1. The grain size distribution has been measured (Figure 1) and the diameter values corresponded to cumulative volume fractions D10, D50 and D90 are 3.53, 6.24 and 10.97 µm, respectively. The standard deviation is 0.86 µm. According to the literature [14], the structure of the Inconel 718 after sintering is composed by a γ matrix reinforced by a γ′ phase Ni3 (Al, Ti) and a γ″ phase Ni3 (Nb, Ti). These two intermetallic phases are the responsible of a rise of the hardness and the good performance off the alloy at high temperature. The presence of carbon, that can bring the formation of Niobium and Titanium carbides, must be limited as there is a risk of lowering the mechanical performances of the component.
Particles size distribution and MEB picture of the Inconel 718 powder. Chemical composition in mass percentage of the Inconel 718 powder provided by Sandvik Osprey.
The selected binder is composed of: cellulose acetate butyrate (CAB 30 K), Polyethylene Glycol (PEG 4 K and 20 K) Stearic Acid (SA) and Phenotiazine (PTZ). The compatibility of this binder with fine grain size powder is helping during the injection [15]. The mixing of the binder with the metallic powder was performed with a twin screw mixer Brabender Plastograph EC. The feedstock was then milled into order to make easier the next manipulations. The feedstock is injected into plates by using an injection press Arburg Allrounder 220S.
The green parts undergo a first debinding step. They are introduced into distilled water at room temperature for 48 h in order to get rid of the PEG. They are then dried in an oven at 50°C for 5 h. The CAB is eliminated by thermal debinding at 500°C for 3 hours under Argon atmosphere. A heating temperature rate of 2°C min−1 is applied during all the cycle in order to avoid overpressure inside the sample during the degradation of the polymer. These temperatures have been defined after analysing the thermos-gravimetric analysis (Figure 2). All the polymers at this temperature before or after the solvent debinding are eliminated.
STA of the feedstock Inconel 718+binder.
The MIM samples are sintered in a high vacuum furnace. A heating temperature rate of 5°C min−1 is applied during all the cycle. The sample is maintained at 1290°C during 2 h in order to reach the higher level of density [16]. The sample is then cooled down inside the oven after 2 h. The density of the samples is then measured by the use of Archimedes’ method. They are also etched in order to observe the microstructure obtained on a microscope Nikon Eclipse-LV150 and grains distribution. The Hardness is finally measured by a Vickers test on a Shimadzu HMV with a load of 1.0 HV.
FAHP sintering
In the case of the FAHP, the Inconel 718 powder is directly introduced into a graphite die. The graphite die is composed of main cylinder pierced with an intern diameter of 20 mm and of two punches of the same diameter. In order to protect the equipment and to improve the electric contacts between the different parts of the system, a foil is placed between the powder and the die. The foil is composed of pure tungsten, which will block the carbon diffusion inside the Inconel 718. If the powder is in contact with carbon during the test, an important diffusion of the carbon could weaken the characteristics of the sample [17].
The samples have been consolidated in a Gleeble 3800 equipment by the application of a strong electric current going through the punches under vacuum. The temperature of the samples is measured via two thermocouples. One is at the centre of one of the punches and the second one close to the powder inside the graphite die. The shrinkage of the sample is measured via the displacement of the punches.
The heating rate of the sample is fixed at 50°C min−1 until reaching the desired temperature. A pressure of 50 MPa is applied at 600°C. The temperature used are between 1100 and 1250°C. After a dwelling time of 15 min, the current is stopped and the sample is cooling naturally in around 45 min. In order to be able to sinter the injected samples, the process is performed with the minimum pressure needed in order to maintain the system between the punches but also without crushing the sample.
Microwave sintering
The injected samples are sintered inside a microwave furnace provided by Sairem. The equipment is presented in Figure 3. The sample is introduced inside the chamber composed of a quartz tube and positioned thanks to an alumina tube in front of the microwave source. The control of the atmosphere inside the chamber is done via a vacuum pump and different inlets of gas. In our case, Argon is going to be used thanks to its facility to be ionised [18]. The power of the microwaves can be set between 0.10 and 3.0 kW.
Microwave furnace built by Sairem.
The microwave source allows generating plasma inside the quartz tube which serves as a dielectric. The electric field of the microwaves is propagating between the tube and a waveguide providing an electric field absorbed by the plasma. This principle allows the creation and the hold up a plasma column of a certain length, depending of the pressure, the power and the nature of the gas. Depending of the gas used, a plasma flame can go to high temperature at a really fast rate [19]. This equipment was designed to hold up until around 2300°C. The heating rate and the temperature obtained during sintering are depending of the length of the flame and the distance from the sample.
In the case of the microwave sintering, the distribution of the porosity inside the material is really important [20]. In order to benefit from the empty spaces left after the polymer is degraded, the injected samples used during microwave sintering were just debinded into water. The thermal debinding is done at 500°C during 30 min under a power below 0.3 kW with a limited heating rate in order to avoid overpressure inside the sample. After the debinding steps, the power of the microwave source is increased step by step of 100°C every minute until around 1300°C during a measured time. The power presented in the result is the real power that was absorbed by the plasma flame and the sample. It is obtained by the subtraction of the power sent by the source and the measured power that is going back to the source.
Conventional sintering of Inconel 718 MIM sample
The evolution of the shrinkage in Figure 4 of the MIM cylinders during the sintering shows two growing of the volume, a first one at 650°C and another at 1025°C. The origin of the volume enlargement at 650 and 1025°C could be attributed to the phase transitions coming out during the temperature elevation. A study of Slama [13] showed that the dissolution of the γ′ phase occurs between 620 and 750°C and the dissolution of the γ phase after 950°C. These transitions were observable by dilatometry by an inflation of the sample. The temperatures of solidus and liquidus are respectively 1260 and 1335°C (from Stoloff) [21].
Dilatometric curve of Inconel 718 sintered at 1290°C during 2 h at 1°C min−1.
We can see that the microstructures in Figure 5. At 900°C the powder is just starting to form necks between the particles. Then the main grain size varies with the raise of temperature from 10 to 32 µm at 1200°C until reaching 45 µm at 1290°C. Some remaining porosity can be observed between the grains at high temperature sintering.
Evolution of the microstructure of Inconel 718 sintered in a furnace at (a) 900°C (b) 1200°C (c) 1290°C.
A temperature of 1290°C under conventional sintering conditions is necessary to reach near full density in the consolidation of Inconel 718 powder. The application of a pressure during FAHP sintering is bringing additional energy to the system, lowering the sintering temperature of the materials. Three different temperatures have been tested in order to obtain the higher density. The dimensional behaviour of the three different cycles is presented in Figure 6.
Shrinkage in function of the time during FAHP sintering of Inconel 718 powder.
The best density was obtained with a temperature of 1250°C, reaching 97% of the theoretical density. The shrinkage behaviour at the end of the dwelling is showing a limit at this temperature. At 1100 and 1200°C, the remaining porosity can be observed (Figure 7) especially in between the former particles grain boundaries. At 1250°C, the density obtained is of 98.8% but the former Inconel particles have been disappeared into a well-developed microstructure. At this temperature the best results are obtained.
Optical Microscopy images of the samples sintered by FAHP at (a)1100°C (b)1200°C and (c) 1250°C.
A first test was performed with the same processing conditions used with the direct consolidation from the powder. The pressure of 50 MPa is too high for the debinded sample and did not sustain the evolution of the process. The second test was performed with a minimum pressure of 5 MPa. At this pressure, the sample was successfully sintered while keeping its shape and did not present cracks at the surface (Figure 8).
Images of the sample (a) debinded, (b) sintered by FAHP at 50 MPa and (c) 5 MPa.
The microstructure observed by SEM (Figure 9) of the injected sample and sintered by FAHP is different than the obtained from the powder directly processed. The injected sample undergoes different heating rates which have an effect on the grain growth. The EDX analysis of the surface of the sample is not showing any carbon or tungsten diffusion and the chemical composition measured correspond to the ratio given in Table 1.
SEM images of the MIM sample sintered by FAHP at 5 MPa and EDX analysis.
The microstructures of the two samples presented in Figure 10 were obtained after a sintering time of 30 and 60 min. After 30 min, the grain size distribution of the bigger grains varies between 20 and 40 µm. The growth behaviour of the grains looks similar to the conventional sintering process at the lower temperature. However, the ratio between large and lower grains is not the same and could be explained by a phenomenon already observed with other materials, which is the formation of hot spots on some samples. These hot spots can reach locally and abruptly a critical temperature where the material starts to dwell, and so followed by an abnormal enlargement of the grains. A too strong power of the microwaves applied on the material explains this phenomenon [17]. After 1 h of sintering, the microstructure looks more homogenous and the former particles are no more observable. Some cracks can be perceived on the surface. They could be explained by the debinding step when the heating rate was too extreme and damaged the shape of the sample.
Optical Microscopy images of the samples sintered by microwave sintering during (a) 30 min and (b) 60 min.
During the microwave sintering, the evolution of the temperature and the power absorbed by the system were measured in Figure 11. When the sintering temperature was reached, the power needed to keep a constant temperature remained constant. With the time, more power was needed until reaching the limit of 3 kW of the equipment. At the end of the 1 h sintering time, it was not enough to stay at this temperature. This effect could be explained by the porosity closure with the time. Because of the lack of communication between the pores, it was harder for the microwaves to go through the sample and more power was needed to force the way [18,19].
Evolution of the temperature and the microwave power with the time during sintering during 1 h.
Comparison of the methods employed and the properties obtained.
Comparison of the methods employed and the properties obtained.
The behaviour of the sintering of Inconel 718 in the three different ways of sintering has been observed. The optimisation of the sintering of Inconel 718 powder by FAHP method is giving better results than the other two methods. The aim of the study was to obtain a MIM sample by these methods while keeping the shape given during the injection. It was possible to maintain the shape of the sample while improving the mechanic properties. By comparing the results between the different processes, the FAHP is giving the best results. On the other hand, it is also the equipment which is the harder to use in order to keep the shape of moulded samples. The microwave sintering is giving lower results than the conventional way of sintering, but by optimising the power parameters, it may be easier to obtain equivalent properties without having problems with the shape of the MIM sample.
Footnotes
Acknowledgements
The author is thankful to Marcos Angulo and Dr Andrea García-Junceda of IMDEA Materials Institute for their help on the utilisation of the Gleeble 3800 equipment.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes on contributors