6. Understanding of Processing, Microstructure and Property Correlations During Different Sintering Treatments of TRIP-Matrix-Composites

Powder metallurgy (PM) is an attractive production route for conventional metal matrix composites (MMCs) because it offers a wide range of reinforcing particle size amounts and distributions in the matrix material. The additional advantages of PM processes are that the grain growth is lower than that from the ingot metallurgy route, and the reaction between the matrix and reinforcing particles can be minimized by using solid-phase sintering processes. This route is most commonly used in the case of composite materials with steel or other high-performance materials as a matrix that cannot be produced by means of the conventional ingot metallurgy route due to the high processing temperatures, different melting points or different densities of the composite components. Conventional PM production normally consists of three steps: blending the metal and ceramic powders, compacting and sintering. In some cases, a sizing step follows the sintering step. During the processing, such when the blending and compacting of the powder particles and pores inside the green compact are adjusted, consolidation processes, such as conventional or conductive sintering, hot pressing (HP) or hot isostatic pressing (HIP), complete the final compaction of the composite material. They differ in the type of heating that occurs (radiative or conductive heating) and whether additional consolidation acceleration occurs due to the presence an external compressive load (conductive sintering, HP and HIP).

In the past, the described advantages of PM instigated many investigations into MMCs that dealt with the optimization of the sintering processes. In several studies, conventionally sintered composite materials, such as functionally gradient materials (FGMs), that were based on stainless steel with a ceramic component, namely, ZrO₂, were investigated [1‐4]. The results showed significant residual stresses in the consolidated composite material that resulted in a high concentration of microcracks after the completion of all manufacturing steps. These stresses can be attributed to a difference in the thermal expansion coefficients between the ceramic and steel and the volume expansion during phase transformation. However, an optimal selection of sintering conditions that considers the layer structure leads to good interface quality. For example, a 4-layer FGM extruded in the form of a tube, consisting of a 316L steel matrix and yttrium partially stabilized ZrO₂ (PSZ), was produced after sintering at 1350 °C for 1.5 h with a good layer interface quality [5]. The good interface formation in composite materials is based on a chemical reaction at the phase interface, which can cause destabilization of the tetragonal(t) phase of ZrO₂ [6]. A variation of the sintering atmosphere showed that while nitrogen positively influenced the mechanical properties of the sintered composite materials, good corrosion properties were achieved by sintering in a vacuum or hydrogen [7].

The consolidation of powders by conductive sintering can be based on different technologies, such as spark plasma sintering (SPS), pulsed electric current sintering (PECS), pulse discharge sintering (PDS), plasma activated sintering (PAS), and field activated sintering technique (FAST). The resistance sintering used in this research is characterized by a direct conversion of electrical energy into heat in the sintered material by applying an electrical voltage to the sample. In [8], a sintering progress was investigated on chromium powder ΠX1M in the temperature range from 1300 to 1500 °C and for sintering times that reached 5 min. The results showed that at the beginning of the isothermal holding time, the effect of the electric current dominated, after which small pores were distributed in the microstructure. The increasing holding time at constant temperature caused low coagulation and enlargement of the pores as well as an increase in the microscopic stresses, which was explained by the action of Laplace forces. Similar investigations of iron powder showed that the densification of a sample to 95% of its relative density when it had dimensions of \( \varnothing 12.8\, {\text{mm}} \times 3.6\,{\text{mm}} \) was possible by heating at 600 K/s to a sintering temperature of 800 °C and with a sintering time of 6 min [9]. The tests were carried out with a relatively low voltage of 3–10 V and with high current values above 10 kA. The authors indicated that fast densification of the material occurred during the heating phase since at this time, the elevated resistance caused a high Joule heating effect. Furthermore, a theoretical prediction of the influence of the composition of the conductive and nonconductive phases on the physical-mechanical properties of the composite materials was carried out [10]. Theoretical modeling that included experimental validation showed that the limit of percolation (the limiting value of the conductor content at which the conductivity becomes zero) depends on the ratio of the particle sizes of the conducting and the insulating phases. As long as the particle size of the conducting phase decreases compared to that of the insulating phase, the percolation limit decreases (conductivity can also be achieved with a small content of the conducting phase). In other cases, the percolation limit increases (a small content of the insulating phase leads to the composite material becoming insulating).

Many investigations and broad experience from industrial practice have proven that an externally applied pressure accelerates the sintering process of a porous body. The influence of the pressure on the sintering kinetics at different pressures and temperatures results from the different deformation mechanisms in the porous sintered material, which have been described in detail in Geguzin [11]. The advantages of processes that apply pressure include the manufacturing of dense materials with lower sintering temperatures and shorter sintering times than those of conventional sintering processes; this results in sintered parts with an increased strength that quite often do not require sintering aids. While a substantial amount of knowledge has already been gained regarding the HP of ceramics, metals and cermets, there are only a few papers regarding HP of steel-ZrO₂ composites. For example, the properties of a ZrO₂(2Y)/transformation-induced plasticity (TRIP)-steel composite produced by HP were investigated depending on the value of applied stress [12]. The powder mixtures were subjected to HP at 1250 °C at 20 MPa for 30 min. The results indicated that the tensile strength and the modulus of elasticity of the composite material decreased under static loading with increasing ZrO₂ content up to 35 vol% due to weak bonding at the ZrO₂–ZrO₂ and ZrO₂-TRIP steel phase interfaces. The dynamic yield strength of the composite was greater than that of the static composite due to the induced martensite transformation, which provided improved plasticity and overall formability. The effect of the ZrO₂ content on the tensile strength of the composite material at the beginning of dynamic deformation was not evident. The rheological stresses increased with increasing ZrO₂ content and deformation in the ZrO₂(2Y)/TRIP-steel composites. Nevertheless, the formability and dynamic strength of the composites decreased with decreasing ZrO₂ content. Furthermore, manufacturing of FGMs based on tetragonal zirconia polycrystal (TZP) and stainless steel or Ni in the form of SUS304-TZP or Ni-TZP, respectively, could be achieved by HP at 1250 °C for 1 h in a nitrogen atmosphere with 5 K/min heating and cooling rates [13‐15]. Finite element modeling of the process indicated that the coefficient of thermal expansion of the composite components was a decisive factor in the production of FGMs free of residual stresses and cracks caused by cooling. Thus, Ni-TZP composites were demonstrated to have better thermal insulating properties than SUS304-TZP.

The postdensification of sintered parts can be achieved by HIP. Unfortunately, there are no data available in the literature about the postdensification of TRIP steel-ZrO₂ composites, while sufficient information is available on the HIP behavior of stainless steel powders. The HIP investigations showed that the relative density of AISI 316L samples was more influenced by the HIP temperature than they were by the HIP pressure. The maximum density of 97% was achieved by HIP at 1250 °C and 120 MPa [16]. Furthermore, the compression behavior of AISI 304 steel-based components at various manufacturing parameters was investigated in [17], whereby the manufacturing route consisted of compacting by selective laser sintering (SLS) with simultaneous cold isostatic pressing (CIP) in the range between 300 and 600 MPa and then sintering in the temperature range between 1250 and 1350 °C for 1 h under a vacuum of 10⁻³ Pa. Finally, the components were hot isostatically pressed at 1200 °C at 120 MPa for 1 h. The component manufactured at the maximum CIP pressure and sintering temperature achieved a maximum density of 97% after HIP.

The aim of the present work is the investigation of the densification behavior and final mechanical properties of TRIP-matrix composites depending on the different fabrication processes.

Both stainless steel and PSZ powders were used for the investigations. The gas-atomized steel powder was purchased commercially as grade AISI 304 stainless steel (grade 1.4301 stainless steel). Three batches were available, and each was used for one of the three consolidation routes. The batches differed slightly in chemical composition and the resulting delta ferrite fraction (Table 6.1). The particle size varied minimally from batch to batch and was in the range of approximately d₁₀ = 10 µm, d₅₀ = 25 µm and d₉₀ = 45 µm.

The solidus temperatures of the stainless steel powder batches calculated with the help of the FactSage^® program based on the present chemical analysis were 1426 °C (conventional sintering), 1361 °C (resistance sintering) and 1405 °C (HP).

ZrO₂ powder was used for the investigations in the partially stabilized ZrO₂ (PSZ) form. MgO was applied as a stabilizer. The chemical composition of the powder is shown in Table 6.2. The ceramic particles had an angular shape with a approximate particle size in the range of d₁₀ = 0.2 µm, d₅₀ = 3 µm and d₉₀ = 22 µm. The phases were distributed among the monoclinic (m), t and cubic (c) phases at nearly equal amounts of 35:32:33%.

The fabrication of MMCs was carried out by conventional and resistant sintering as well as by HP with partial application of the HIP process. The preparation of the powder mixture for all consolidation routes was carried out either by blending for conventional sintering and HP with partial application of the HIP process or by mechanical alloying of the powder mixtures for resistance sintering. Before blending, the volume fraction of ZrO₂ was adjusted from 0 to 30 vol%. The inorganic powder mixture was then blended by means of ZrO₂ balls stabilized with yttrium for approximately 30 min in a ball mill. Especially for the conventional sintering process, approximately 2.5 mass% of organic binder was added.

The powder mixture was prepared for resistance sintering using a mechanical alloying technique in a vibrating mill with high energy output (attritor) [18]. The grinding duration was approximately 90 s for each 250 g of the initial powder mixture. At the end of the process, the powder mixture was fabricated with an activated surface.

Compacting the green bodies for conventional sintering was carried out by extrusion of the prepared powder mixture. Extruded rods 20 mm in diameter were fabricated using a single-screw extruder with a vacuum chamber in combination with a kneading machine. After the extrusion process, the green compacts were cut into segments of the required length. Afterwards, the water added to the powder mixture before kneading was removed by an air dryer. The relative density of the extruded green bodies was approximately 60%. The subsequent removal of the organic binder occurred in the air at 350 °C for a duration of 5 h. The heating took place up to 100 °C at a rate of 2 K/min and at a rate of 0.2 K/min above 100 °C. The cooling took place at the same rates.

Conventional sintering was performed on extruded and debindered green compacts in an Astro 1100-60100-M1 model refractory metal hot-zone laboratory furnace (Thermal Technology LLC_TM). The accuracy of the temperature measurement was ±4 K over the applied temperature range. According to the furnace manufacturer, the typical temperature homogeneity of the furnace at 1700 °C is ±3 K for 35% and ±6 K for 50% of the furnace chamber volume. The sintering process took place in argon and nitrogen atmospheres (class 5.0) as well as in a vacuum of 10⁻⁶ mbar. The heating rate was approximately 5 K/min, while the cooling rate was approximately 10 K/min.

A multifunctional GLEEBLE HDS-V40 simulation system (Dynamic Systems Inc.) was used for the resistance sintering. The parts of the specially designed tool for resistance sintering in the GLEEBLE are shown in Fig. 6.1 and consisted of water-cooled base plates, a bottom panel, a ceramic die (Si₃N₄) with conical inward and outward design features, and a steel press sleeve to reinforce the ceramic die and a punch. To reduce the heat dissipation from the face of the sample in the front through the bottom panel, additional plates of carbon-fiber-reinforced carbon (with a thermal conductivity of 3 Wm⁻¹ K⁻¹ and specific electrical resistance of 26 µΩ) were interposed for thermal insulation. Furthermore, a higher electrical conductivity than that of steel was achieved by using punches fabricated from a material based on a molybdenum alloy with the addition of Ti and Zr. The base plates had current connections but were electrically insulated from the other parts of the GLEEBLE by isolation plates. The powder mixture to be sintered filled the ceramic die before resistance sintering, where the mass of the powder mixture to be sintered was always 250 g. The dimensions of the sintered samples were \( \varnothing 50\, {\text{mm}} \times 18\,{\text{mm}} \). The ceramic die and tool surfaces were lubricated by hexagonal boron nitride spray before each test to improve the removal of the sintered sample from the tool after the process. To avoid welding the powder mixture to the bottom panel and the punch as well as carburization of the sample by the carbon-fiber-reinforced carbon plates, a tantalum foil was placed in those areas. After closing the ceramic die with the punch and applying a critical presintering pressure of approximately 50 MPa, the resistance sintering was carried out according to a specified temperature-time cycle with an almost constant sintering pressure of approximately 20 MPa. The temperature of the sample during the sintering process was continuously controlled and adjusted with the help of a thermocouple that was located in the bottom panel and carbon-fiber-reinforced carbon plate so that the heating rate was approximately 50 K/min. Another thermocouple was used for the temperature measurement of the punch and was placed near its working surface. The sample resistance controlled the sintering temperature depending on the ZrO₂ content, the presintering pressure and the sintering pressure as well as the temperature profile. In the course of the resistance sintering tests, the voltage and current values were measured and recorded with an accuracy of ±0.01 V and ±1 A using an ALMEMO 2690-8 measuring instrument. The peak values of the voltage and current were approximately 4 V and 6 kA, respectively. Based on the determined values, the electrical resistance of the samples R was calculated according to the following equation:

where I is an electrical current and U is an electrical voltage.

After the tests, the samples were cooled in the sintering tool before they were removed.

Preliminary tests that directly hot pressed the powder mixtures in the molds (at 1250 °C for 2 h at 30 MPa) resulted in destruction of the graphite molds as a result of the penetration of the mixture into the mold gaps and its interaction with the mold material. For this reason, two methods of mixture precompaction into billets were tested: die cold pressing (CP) of the powder mixture and CIP followed by short-term presintering in a hydrogen atmosphere (Table 6.3). Significant advantages of the first method include the simplicity of the process, high productivity and possibility of mechanization. However, a significant heterogeneity of the billets in terms of the density and inability to produce parts with a complex shape did not allow us to consider this as a universal method. The second method overcomes the disadvantages of the die CP step, but the size inaccuracy of fabricated green compacts and their high surface roughness require the application of additional labor-intensive machining. To avoid the appearance of cracks and stratifications in the billets after CIP, the addition of short-term sintering in a hydrogen atmosphere was carried out, and a relative density of approximately 85% was achieved.

The dimensions of the HP samples were \( \varnothing 80\, {\text{mm}} \times 20\,{\text{mm}} \). To remove the residual porosity in the specimens after HP, HIP in an ABRA press apparatus (ABRA Fluid AG) was added and its impact on the microstructure and property evolution was investigated.

The density of the sintered samples was determined in accordance with DIN EN 623-2. The samples for metallographic analysis and mechanical testing were cut from the machined sintered workpieces. Tensile tests were carried out on an AG-100kNG universal testing machine at room temperature according to DIN EN ISO 6892-1 on cylindrical specimens with dimensions of \( \varnothing 4\, {\text{mm}} \times 20\,{\text{mm}} \). The impact toughness of the 55 mm × 10 mm × 10 mm samples with a V-shaped stress concentration was tested on a RKP450 impact testing machine at room temperature according to DIN EN ISO 148-1. In all cases, samples were obtained in the axial direction for conventionally sintered samples and in the transverse direction for the resistance sintered or hot-pressed samples. The hardness HV10 was determined according to DIN EN ISO 6507-1 on a ZHU250 instrument. The microstructure and fracture surface analysis of the tested samples were carried out by light and scanning electron microscopy. A quantitative evaluation of the micrographs was conducted to determine the pore shape factor f according to Equation:where A is a grain surface area and C is a circumference of a grain.

The size of the pores and grains was determined with the help of the image processing program Archive4Images. The phase analysis of the ceramics after different consolidation routes was conducted with X-ray diffraction (URD 6 X-ray diffractometer) with an accelerating voltage of 40 kV and current of 30 mA; energy-dispersive X-ray spectroscopy was also conducted.

Based on the experimental results, it was possible to model the sintering behavior with kinetics calculations equations [19] as follows:where \( \rho_{\text{rel}} \) is a relative density, t is a time, ϑ is a temperature, Z is a ZrO₂ content, z₁ and z₁ are constants to be determined from the available data. Furthermore,

Here, parameters occurring in the equation, such as the Avrami exponent and activation energy, were determined using experimental results and the mathematical-numerical method of nonlinear least square approximation. It was assumed that heating did not have a significant influence on the sintering process, i.e. the assumption of isothermal conditions was used. The modeling approach was sufficient to allow interpolation of the measured values after their successful validation for the purpose of design of optimum technological parameters and thus find an understanding of the changes taking place in the material.

In this work, the densification behavior and final mechanical properties of TRIP-matrix composites were investigated. The composites were based on 1.4301 stainless steel powder, which was available in the form of three batches with slight differences in their chemical composition and contained up to 30% MgO-PSZ. The materials were fabricated using three different PM consolidation processes: conventional sintering, resistance sintering and HP. In the case of resistance sintering, a special tool was designed and optimized to achieve low heat dissipation from the specimen. The microstructure evolution and material properties along each consolidation route were compared to each other. The experimental analysis of the densification kinetics was based on electrical resistance measurements and microstructure evaluation. It was shown that the chemical composition of the steel matrix had a significant influence on the optimum sintering temperature among the investigated sintering technologies.

Experimental investigations of resistance sintering have shown that increasing the presintering and sintering pressures and decreasing the heating rate result in an elevated density of the material. The sintering residence time for composites with a high ZrO₂ content should be extended compared to those with a low ZrO₂ content since the ZrO₂ content delays the consolidation process.

The investigations on HP showed an almost optimal production route, namely, by CIP, subsequent presintering at 1100 °C for 1 h and then HP. The HIP treatment has proven to be ineffective for the postcompaction of hot-pressed samples.

During conventional sintering, the dependence of the composite density on the sintering temperature and sintering time as well as on the ZrO₂ content was quantified by means of a physical-mathematical approach and experimentally supported modeling. It was shown that the densification rate in the investigated temperature range had a low dependency on the ZrO₂ content. The temperature distribution and evolution in the sintered material during resistance sintering was determined based on the physical-mathematical equations as a function of the electrical current value, the sample geometry and the ZrO₂ content. The modeling provided the basis for the densification kinetics and the local character in the composite specimen volume.

A fine austenite microstructure and higher density and tensile strength was achieved with the optimized HP route (2nd precompaction method) and presented a better combination of properties than those of conventional and resistance sintering. Furthermore, low sintering temperatures and short sintering times made the optimized process favorable. The optimally determined HP parameters are 1250 °C with a holding time of 30 min to 1 h and pressure of 30 MPa.

Furthermore, it was shown that the mechanical properties of the consolidated composites primarily depended on the relative density of the material and the ceramic content in the composite. Here, a linear correlation of the tensile strength and an exponential correlation of the total elongation with the relative density of the material was observed. There was no significant influence of the consolidation technique on these mechanical properties. The insignificant deviations could be explained by differences in the initial chemical composition of the powder mixture. With a particle content of approximately 10–30% ZrO₂, strength increases of up to approximately 40% were found. An increase in the tensile strength was accompanied by a decrease in the total elongation. The obtained dependences partially exceeded the mechanical properties of the particle-reinforced composites with a light metal matrix [31]. However, an increasing ZrO₂ content up to 20% caused early failure of the composite under tensile loading due to the formation of ceramic agglomerates with poor bonding to the ZrO₂ particles. Therefore, the best mechanical properties require sufficient fineness of the ceramic particles (<10 µm) and a uniform distribution as well as good bonding into the steel matrix and low destabilization of the t-ZrO₂ crystallographic modification.

Finally, it was shown that the content of m-ZrO₂ increased with increasing sintering temperature. The increase in the impurities in the ceramic powder caused a greater initial destabilization of the metastable t-ZrO₂ crystallographic phase. In contrast, the influence of the PM consolidation processes investigated in this research on the m-ZrO₂ content had a secondary effect.