1 Introduction

Numerous researchers have reported on attempts to fabricate 17-4 PH stainless steel components via powder metallurgy techniques [1,2,3,4,5]. Challenges associated with fabricating 17-4 PH components with desired properties via powder metallurgy processes such as the conventional press and sinter (PS) and additive manufacturing (AM) techniques have been documented [2, 5,6,7]. These challenges include low powder compressibility for PS [4, 6] and the undue residual stress accumulation from martensitic transformation during deposition for AM [2, 5, 7]. More recent reports have however widely reported on the attractiveness of metal injection molding (MIM) in the fabrication of complex, near-net shaped 17-4 PH stainless steel components. For example, Wu et al. have reported MIM of 17-4 PH stainless steel and the results showed that better mechanical properties of this alloy can be achieved through MIM [8].

MIM is an attractive powder metallurgy technique that, in recent years, has developed to be a cost-effective manufacturing process for high-volume production of intricately shaped parts where high-dimensional accuracy is required [9,10,11]. It is a multistep process by its nature; involving the preparation of suitable feedstocks, injection molding, solvent/catalytic and thermal debinding, high-temperature sintering, and post-sintering processing if required. The mechanical properties of final MIM components are highly dependent on several factors such as the starting powder material and feedstock characteristics, as well as the fabrication process itself [12]. Even though the effects of the starting powder material characteristics and MIM processing (viz. debinding and sintering) on the final mechanical properties of 17-4 PH stainless steel parts are well understood [13,14,15,16,17,18,19], the effects of feedstock flow stability, on the other hand, appear to not be equally documented.

This can be attributed to the existence of numerous and seemingly unrelated rheological parameters where each parameter often gives incomplete or conflicting insights into the formulated feedstocks’ qualities [20, 21]. For this study, two general parameters are sought—the feedstock flow stability and the moldability index (αstv). The feedstock flow stability measures the consistency of the viscosity. The moldability index is a more generalized parameter commonly used as a comparative indicator of feedstock qualities (superiority/inferiority) with respect to its rheological behavior and moldability characteristics [21,22,23,24,25]. The moldability index exists in the literature in various forms, for this work, it is defined as in Eq. (1) below after the Weir model of polymers [26]:

$$ {\boldsymbol{\alpha}}_{\boldsymbol{stv}}=\frac{\mathbf{1}}{{\boldsymbol{\eta}}_{\mathbf{0}}}\frac{\mathbf{1}-\boldsymbol{n}}{\left(\raisebox{1ex}{$\mathbf{E}$}\!\left/ \!\raisebox{-1ex}{$\boldsymbol{R}$}\right.\right)} $$
(1)

where the subscripts s, t, and v represent the index’s sensitivity to the shear rate, temperature, and viscosity variations, respectively, whereas η0, n, E, and R indicate viscosity at a reference temperature, power-law flow index behavior, activation energy for the Arrhenius temperature dependence of the viscosity, and the gas constant, respectively [20, 21, 23, 24].

Our preliminary studies investigated on the effect of powder particle size on mechanical the properties of 17-4 PH stainless steel (using the – 5-, − 15-, and – 45-μm unimodal feedstocks) demonstrated that the sintered − 5-μm feedstock achieved the highest final density as well as superior moldability during injection molding [12]. Further studies on the rheological properties of 17-4 PH stainless steel MIM feedstocks concluded that at high shear rates, feedstocks of bimodal powder distributions exhibit higher moldability index values which exemplify better rheological behavior during injection molding [27]. This current report is intended to evaluate the influence of the feedstock moldability index and flow stability of bimodal MIM feedstocks on the physical and mechanical properties of as-sintered 17-4 PH stainless steel.

2 Materials and experimental methods

Figure 1 shows the scanning electron micrographs of the unimodal − 5-μm, − 15-μm, and – 45-μm 17-4 PH stainless steel powder materials used for this study. The – 5-μm and – 15-μm powder materials were supplied by Epson Atmix Corporation (JP) and the – 45-μm powder material by Praxair Surface Technologies.

Fig. 1
figure 1

Scanning electron micrograph (SEM) of – 5-μm (a), − 15-μm (b), and – 45-μm (c) 17-4 PH stainless steel powder. Adapted from [27], with permission

Full size image

Using a proprietary polymer wax–based binder system and the said powder materials, five 60 vol% ready-to-inject feedstocks were formulated; a summary of the feedstocks is shown in Table 1. The 25:75 bimodal powder blending follows Joo Won Oh [28].

Table 1 Feedstock formulations used in this study
Full size table

The feedstock rheology was characterized using a Ceast SmartRheo capillary rheometer (Pianezza TO, Italy). The capillary used has a length-to-diameter ratio of 5:1 and shear rates ranging from 1000 to 10,000 s−1 were applied. All prepared feedstocks were tested at 110, 120, 130, and 140 °C as reported prior in reference [27].

The molding of the “dog-bone” tensile specimen, specified according to the ISO 2740 standard, at prior determined optimum injection molding parameters given in Table 2. The dimensions of the dog-bone tensile specimens were 88 mm (overall length), 40 mm (gauge length), 14.8 mm (grip section diameter), and 5 mm (gauge section diameter), see Fig. 2. Debinding of green specimens was achieved in two steps—first in n-heptane (Merck Millipore) at 60 °C for 24 h and thermally at 600 °C for 1 h in a Carbolite tube furnace. A pre-sintering step of the brown specimens was carried out at 1000 °C for 1 h in a Carbolite tube furnace under an argon gas flowing (at 1.0 L/min) environment.

Table 2 Optimum injection molding parameters [29]
Full size table
Fig. 2
figure 2

Shrinkage variations in the specimens at 60 vol%. a As-molded. b Solvent debound. c As-sintered

Full size image

Sintering was performed at temperatures of up to 1300 °C at 10 °C/min for 3 h in a Xerion vacuum furnace. Sintered samples were evaluated for shrinkage, apparent density, microstructure, pore sizes and distribution, hardness, and tensile properties. The densities were evaluated using a precision analytical balance (OHAUS) based on the Archimedes’ technique according to ASTM B962-08 and Spierings et.al [30, 31].

A Leica DMI500 M optical microscope was used to observe the as-sintered microstructures. Apparent hardness measurements were carried using a Vickers microhardness tester (FM-700) with a 500 g load applied for 10 s as per ASTM-E384 standard. Tensile testing was done on an instrumented INSTRON™ Servo Hydraulic 1342 test apparatus at a strain rate of 0.5 mm/min. The linear shrinkages of the sintered samples were measured against the as-molded green parts using a Vernier caliper.

Figure 2 illustrates the visual differences between the as-molded, solvent debound, and sintered (PS-45/5 for example) specimens. The appearance of specimens in Fig. 3 indicates the level of shrinkage that occurred during processing. Specimen (b) underwent solvent debinding and no shrinkage difference is observed relative to the as-molded part, whereas, specimen (c) underwent sintering and shows remarkable shrinkage relative to the as-molded part.

Fig. 3
figure 3

a, b A summary of the rheological properties of feedstocks tested. From [27], with permission

Full size image

3 Results and discussion

3.1 Rheological properties

The results presented in this section endeavor to extend the findings from our previous studies by exploring the influence of the feedstock flow stability and moldability index parameters on the mechanical properties of the MIM materials produced from the feedstocks studied prior.

Figure 3 summarizes the findings from a recent report on the effects of the powder size distribution on numerous rheological properties of 17-4 PH stainless steel MIM feedstocks [27]. Figure 3 a shows the variation of the measured viscosity of the feedstocks with the shear rate at the minimum injection temperature. Figure 3 b shows the effects of bimodal powder distributions on feedstock stability (flow activation energy) and moldability index.

Subdued variations of the measured viscosity of the feedstocks are an indication of the “stability.” Feedstock instability has been attributed to the size-segregation of powder particles and local loading variations in the feedstock during flow [32, 33]. Bimodal feedstocks and agglomerated fine powder-based feedstocks exhibited apparent flow instability, especially at lower shear rates. According to Khakbiz et al. [34], at relatively low shear rates, such feedstocks instability tentatively manifest small density or pore distribution variations, especially in the bimodal feedstocks. However, the results show that the feedstocks become stable at higher shear rates (typical injection molding conditions) as the breakup of agglomerates and clusters lead to increased particle-particle interaction to aid flow in accordance [35].

Higher moldability index (αstv) values infer superior rheological properties and desirable moldability qualities [22, 24]. Figure 3 b indicates that the FS-15 and FS-5/15 MIM feedstocks give significantly high moldability index values at high shear rates, where the injection molding is intended. The feedstocks formulated from the coarser powder (FS-45 and FS-15/45 in this case) consistently exhibit inferior moldability along with the FS-5 unimodal feedstock.

Particle size analysis and the SEM micrographs of the starting powders − 5 μm, as seen in Fig. 1a, exhibit a considerable degree of agglomeration compared with the coarser − 45-μm powder material seen in Fig. 1c. Particle size analysis and the SEM micrograph of the − 15 μm powder materials shows a hint of a multimodal distribution.

According to references [16, 23, 36,37,38], at high shear rates, the FS-15 and FS-5/15 feedstocks could be ideal for MIM requiring very complex geometries/features or microMIM.

3.2 Microstructure and physical properties

The optical micrographs demonstrating the resulting microstructures of the sintered specimens are shown in Fig. 4a–e while Fig. 4 f shows the image analysis of the representative micrographs of the sintered feedstocks as exhibited in Fig 4a–e. The relative density and shrinkage behavior of the sintered materials are displayed in Fig. 5.

Fig. 4
figure 4

Optical micrographs showing the microstructures of the sintered materials, i.e., PS-45 (a), PS-45/15 (b), PS-15 (c), PS-15/5 (d), and PS-5 (e). f Representative pore size distributions obtained by image analysis of the micrographs (shown in Fig. 3(a–e))

Full size image
Fig. 5
figure 5

Representation of the shrinkage and densification behavior of the feedstocks

Full size image

The coarse PS-45 and fine PS-5 unimodal feedstocks as well as the PS-45/15 bimodal-based feedstocks demonstrated lower sintered densities. The lower sintered densities in the PS-45 unimodal and PS-45/15 bimodal feedstocks can easily be attributed to the fact that larger particles tend to have lower sintering rates thus less shrinkage [38,39,40]. As a testament, the pore size distributions obtained by image analysis of the micrographs also show abundant, relatively larger, and characteristically irregular pore structures. On the other hand, the low densities in the PS-5 unimodal feedstock can be attributed to the inefficient particle packing associated with agglomeration and fine powder surface chemistry [39, 41]. The PS-5 unimodal feedstock possibly shows the most shrinkage which indicates that the powder had a lower starting packing density.

Conversely, the as-sintered PS-5 and PS-P15 microstructures exhibit relatively uniformly distributed fine and spherical pore structures—an indication that the finer starting powder materials resulted in comparatively better sinterability (compared with the relatively coarse unimodal feedstocks). Prior reports by Kearns and co-workers [40] further collaborate with this observation.

Generally, higher porosity is expected in conventional PM parts (when compared with wrought materials) [1, 42]. However, if MIM is to find industrial applications, MIM parts are expected to be “fully dense and comparable in properties with their wrought equivalents” [43]. In this work, perhaps one of the most significant differences between the micrographs of the unimodal and bimodal as-sintered materials lies in the density and distribution of pores as exhibited in Fig. 4f. Finer starting powder particle sizes, in general, result in finer pores. In addition, and as also demonstrated in computer simulations and in practice [18, 38, 39, 44], improving gap-filling and the propensity for particle rearrangement in the bimodal feedstocks typically leads to improved sinterability and even finer and narrower pore distributions as well as enhanced mechanical properties.

Feedstocks that exhibit superior rheological behavior of significantly higher moldability index values in Fig. 3b, i.e., formulated using the – 15 and 5/15-μm powder, yielded favorable densification [28] and more uniform sintered microstructures in the case of the 5/15-μm powder. The as-sintered densities are comparable with the minimum specifications for 17-4 PH MIM materials as specified by various suppliers (powder and MIM feedstock), as according to prior studies available in the literature and the respective Metal Powder Industries Federation (MPIF) and American Society for Testing and Materials (ASTM) standards [45, 46].

3.3 Mechanical properties

Table 3 gives a summary of the “as-sintered” mechanical properties of the test samples used in this study. Tensile properties generally improve with bimodal powder distributions. This observation can be attributed to the increasing densification and grain boundary volume as the particles become finer [28, 39]. Tang and Pistorius [47] indicated that defects including porosity control fatigue life. Fotovvati et al. [48] summarized these defects as residual stresses, surface roughness, internal defects, and microstructural inhomogeneities. These defects are typically the origin of fatigue cracks; thus, abundantly larger irregular pores are subject to acting as crack initiation sites and therefore reducing fatigue life. Fatigue and tensile behaviors are monotonic; hence, poor tensile properties of feedstocks FS-45 and FS-45/15 are a direct indication of inferior fatigue life of the samples and this is attributed to larger porosity (Fig. 4 a and b). Even though the PS-5 sintered material exhibited desirable tensile strength and ductility and hardness properties, the occurrence of fine pore structures appears to have led to sintered densities below the minimum specifications for 17-4 PH MIM products.

Table 3 Properties of as-sintered parts 17-4 PH stainless steel against MPIF standards
Full size table

As-sintered mechanical properties of 17-4 PH stainless steel products are largely determined by microstructures, including the porosity. This is supported by Gülsoy, Özbek, and Baykara [1], Kearns and co-workers [40], and German [49]. The bimodal feedstocks, and PS-15/5 in particular, exhibit the best combination of low cumulative pore volumes and mechanical properties exceeding the minimum property values as set-out in the MPIF Standard 35 for MIM products [49]. By observation, SEM micrograph of the – 15-μm powder appears bimodally distributed, see Fig. 1a.

Superior rheological properties of feedstocks generally lead to favorable as-sintered mechanical properties. Significantly higher moldability indices, for the − 15 and 5/15-μm feedstocks in particular, possibly indicated the breakup of powder agglomerates and clusters to aid flow [35] and also lead to increased densification and grain boundary volume as the particles become finer [28, 39]. The same argument holds when the as-sintered mechanical properties of the FS-45/15 bimodal feedstock are compared with that of the FS-45 feedstock.

4 Conclusions

This current report is an extension of prior published work. This work has evaluated the influence of the feedstock moldability index and flow stability of bimodal MIM feedstocks on the physical and mechanical properties of as-sintered 17-4 PH stainless steel materials.

A review of the rheological properties of 17-4 PH stainless steel MIM feedstocks and the physical and mechanical properties of as-sintered 17-4 PH stainless steel materials were presented.

The results show that coarser particle sizes (unimodal or bimodal) exhibited relatively lower shrinkage levels which is desirable for dimensional control but inferior “as-sintered” mechanical properties (viz. tensile and fatigue behavior). While the finest unimodal feedstock used in this work exhibited inferior rheological feedstock properties and mechanical properties, bimodal feedstocks formulations exhibited higher moldability index values which exemplify better rheological behavior during injection molding [27] and comparatively improved as-sintered mechanical properties. Tentatively, a higher feedstock moldability index is an indicator of favorable as-sintered mechanical properties.