Effect of pore structure on the impact toughness of copper-infiltrated sintered steel

Copper-infiltrated sintered steel is a prominent area of research in powder metallurgy, with a particular focus on enhancing impact toughness. In this study, sintered steels with varying pore structures were prepared using iron powders and infiltrated with copper to investigate their impact toughness. The results indicate a significant improvement in the impact toughness of the sintered steels with concentrated iron particle sizes. The density of the samples increased from 6.85 g cm−3 to 7.55 g cm−3 through copper infiltration. The large copper-phase sample with the particle size of 150 μm exhibits an impact toughness as high as 41.09 J cm−2, and its fracture morphology mainly shows transgranular fractures of large iron particles. It is 1.5 times that of the sample made from commercial iron powders which measured 27.46 J cm−2. This enhancement is primarily attributed to the precipitation of γ-Fe within the large copper phase, thereby enhancing the copper phase and transferring the load into large iron particles. A brief mechanism of γ-Fe precipitation in the large copper-phase has been provided.


Introduction
Iron-based powder metallurgy products, also known as sintered steels, are the largest category within the industry and are known for their high precision, excellent material utilization, high performance and cost efficiency.However, compared to conventional cast and forged steels, sintered steels have a lower density, primarily due to the presence of pores [1][2][3], which limits the dynamic performance of sintered steel, especially its impact toughness, thus limiting its further application.Utilized as a cost-effective and widely used process, copper infiltration significantly improves the impact toughness of sintered steels.For instance, sintered steel typically featuring the FC0208 material has an impact value of less than 10 J cm −2 , which can be increased to more than 20 J cm −2 after copper infiltration [4].Furthermore, copper infiltration enhances the density, strength, wear resistance, thermal conductivity and fatigue resistance of sintered steels.
Currently, research on copper-infiltrated sintered steels mainly focuses on the composition and content of copper infiltrant.It also involves optimizing the copper infiltration process, which includes parameters such as copper infiltration temperature, sintering atmosphere, infiltration time, pre-sintering, and heat treatment [5][6][7][8][9][10][11][12][13][14][15].Limited research exists on the influence of the pore structure of the sintered steel skeleton on the copper infiltration process.Nevertheless, it has been demonstrated that the pore structure plays a crucial role in the permeability of liquids and the mechanical properties of the infiltrated materials [16][17][18][19][20][21][22][23][24].For example, Johannes et al [22] experimented by modifying the proportion of fine powder added to the initial powder of H11 tool steel within the range of 0 wt% to 15 wt%.When comparing the sintered H11 tool steel to the modified materials, they observed that after copper infiltration, the modified material exhibited an increase in thermal conductivity by 1.84 times (with 15 wt% fine powder in the skeleton) and 2.67 times (with 0 wt% fine powder in the skeleton).
Liu et al [23] discovered that, with an increase in the particle size of the skeleton material, the pressure along the infiltration direction decreases.This reduction in pressure leads to a decrease in residual pores at the same location, thereby improving the mechanical properties of the composite.Montoya et al [24] used three different particle size distributions of SiC powders to prepare skeletons for infiltration and produce infiltrated materials.The results showed that with an increasing particle size distribution, the composites exhibited a parabolic distribution pattern in surface hardness and fracture toughness.In conclusion, different pore structures can affect the copper infiltration process.Currently, there is no clear indication of which pore structure is more advantageous for the impact toughness of sintered steel after copper infiltration.To study the impact toughness of copper-infiltrated sintered steel, a thorough analysis of the pore structure of the skeleton is essential.Despite this, the influence of the skeletal pore structure on the copper infiltration strengthening effect remains unclear.This study aims to systematically investigate the influence of different pore structures before and after copper infiltration in sintered steel, aiming to provide theoretical and technical support for high-performance ironbased copper infiltration components.
In this study, water-atomized iron powders with different particle size distributions were used to prepare sintered steel skeletons with different pore structures.The study's primary aim was to improve the impact toughness of sintered steel and examine the effect of these pore structures on its toughness, both before and after copper infiltration.Additionally, the study aimed to analyze and optimize the preparation process of copperinfiltrated sintered steel by linking the impact toughness to the changes in microstructure during the material fracture process.

Materials
In this experiment, the average particle size (D50) was used to quantitatively characterize the particle size of powders.Commercial water-atomized iron powder (Shandong Luyin New Material Technology Co. Ltd) with a particle size of 100 μm was screened using 100 mesh, 150 mesh, 200 mesh, 270 mesh and 400 mesh screens.Consequently, powders with D50 values of 150 μm, 80 μm and 40 μm were obtained and used to prepare impact samples.The particle size of the powder was tested using a laser particle size analyzer (Bettersize2600; Dandong Baite; China).Additionally, figure 1 displays the scanning electron microscope (SEM) morphologies of the four iron powders with different particle sizes, and the D50 values of the iron powders are listed in table 1.

Sample preparation
The four iron powders were separately mixed with 2 wt% of electrolytic copper powder (GRIPM Advanced Materials Co. Ltd), 0.8 wt% of graphite (Qingdao Tianheda Graphite Co. Ltd), and 0.5 wt% of zinc stearate (Jiangxi Hongyuan Chemical Co.Ltd) in a three-dimensional mixer (SBH-10; Jinan Feichi; China) for 1 h at 15 r min −1 (optimized in various studies) [17,25].The powders were pressed using zinc stearate as a lubricant.The electrolytic copper powder had a D50 value of 38 μm, while graphite had a D50 value of 6 μm.After the mixing, the powders were pressed into unnotched specimens for impact tests in one direction with a pressing pressure of 500∼700 Mpa, following the ISO 26843:2015 standard (55 mm × 10 mm × 10 mm).The green compacts were sintered in a tubular sintering furnace (SK-G08123K; TIANJIN ZHONGHUAN; China) to obtain skeletons.Infiltrated samples were produced by placing the copper infiltrant (GRIPM Advanced Materials Co. Ltd) at the bottom of the skeletons.The chemical composition of the copper infiltrant is shown in table 2. The sintering and infiltration were carried out at 1120 °C for 60 min under an atmosphere of 75%N 2 -25%H 2 .The green compacts were held at 750 °C for 30 min to remove the lubricant.

Morphologies and mechanical properties
The density was measured by the Archimedes method according to ISO 2738:1999.The Charpy impact test was performed on the specimens at room temperature in the Charpy impact testing machine (JB-300B; JINAN KOOHEI; China) according to ISO 5754:2017.The samples were ground, polished and then etched in 4% nitric acid alcohol.The microstructures of the samples were examined with an optical microscope (OM.Axio Vert A1; ZEISS; Germany), a scanning electron microscope (SEM.EVO 15; ZEISS; Germany), and a transmission electron microscope (TEM.TECNAI G2 F20; FEI; USA) with an energy dispersive x-ray spectrometer (EDS).The TEM data were analyzed using Digital Micrograph 3.5 software.A Zeiss microscope image analysis system was used to perform statistical analysis of 500× metallographic images and obtain information on pore size and pore morphology.For each sample, ten measurements were conducted.

Pore structure
The characteristics of the structure formed during particle packing are reflected in the pore morphology.After sintering, the skeletons were obtained, and which density was measured at 6.85 g/cm 3 .From table 3, it is evident that the porosity of the four skeletons exhibits minimal distinction.Figure 2 shows the pore morphology of skeletons for four samples, accompanied by statistics on pore size and pore area.Notably, significant differences can be seen in both pore size and morphology.Sample 0# utilizes iron powder with a wide distribution of particle sizes and a broad range of pore sizes, resulting in an uneven distribution of pores in the samples after compaction.In sample 1#, there are localized large-sized pores where the boundaries of the original particles are visible, and some particles have not fully combined.Sample 2# demonstrates a uniform distribution of pore sizes.Following the formation of the skeleton framework, sample 3# contains numerous small pores and exhibits a sparse distribution of large-sized pores.The raw iron powder of sample 0# has a wide particle size distribution, and the pore size and morphology of the sample exhibit a high degree of heterogeneity.Among the pores, 90% of the pore sizes are smaller than 31 μm, with a maximum pore size not exceeding 45 μm.In sample 1#, 90% of the pore sizes are smaller than 53 μm, with a maximum pore size not exceeding 72 μm.Among the pores, the area with a size larger than 40 μm accounts for 20.55% of the total pore area.Some particles are not completely fused, clearly revealing the boundaries of the original particles in sample 1#.In sample 2#, 90% of the pore sizes are smaller than 27 μm, with a maximum pore size not exceeding 38 μm.As for sample 3#, 90% of the pore sizes are smaller than 25 μm, with a maximum pore size not exceeding 35 μm.In conclusion, larger iron particles will produce larger pores.Conversely, when the iron powder particles are smaller, the pore size significantly decreases due to the increased contact area between the particles.

Impact toughness
The impact toughness of the samples is shown in figure 3. Before copper infiltration, the impact toughness of the screened samples exceeds that of the unscreened sample 0#, and the impact toughness gradually increases as the particle size decreases.In particular, sample 3#, which is made of the smallest particles, achieves the highest impact toughness of 21.41 J cm −2 , followed by sample 2# with 14.01 J cm −2 , sample 1# with 11.71 J cm −2 , and sample 0# with 7.92 J cm −2 .This shows that the particle size of the raw material influences the impact toughness to some extent.
After copper infiltration, the density increased to 7.55 g cm −3 while maintaining the same copper infiltration content.The impact toughness of the samples was significantly improved.This can be attributed to two key factors: the solidification of the copper by the solution and the reduction in stress concentration associated with open pores.Sample 1# achieved the highest impact toughness of 41.09 J cm −2 , followed by sample 3# (38.21 J cm −2 ), sample 2# (33.97 J cm −2 ), and sample 0# (27.46 J cm −2 ) sequentially, which represents a 49.64% improvement compared to conventional copper-infiltrated sintered steel.Compared to the 250.90% increase in impact toughness of sample 1#, sample 3# showed only a modest 78.47% increase after copper infiltration, while sample 0# increased by 246.71% and sample 2# increased by 142.50%, respectively.

Microstructure
Figure 4 shows the microstructure of the four samples after copper infiltration.All the samples contain matrix (ferrite, pearlite), copper phase and pores.Sample 0# has a lamellar perlite structure and a relatively uniform size distribution of the copper phase.Samples 1#, sample 2#, and sample 3# show a gradual change in the copper morphologies, transitioning from larger and aggregated structures to elongated and dispersed structures in a sequential manner, which can be attributed to the preservation of the original sample's pore characteristics by the copper phase.Due to a higher percentage of large-size pores before copper infiltration compared to the other samples, sample 1# exhibits a greater degree of copper agglomeration after infiltration.In particular, there appear to be suspected precipitates in the area in figure 4(d), and the location is in the central area of the copper phase.Labelling is performed on the regions of suspected precipitate and non-precipitate copper phases, spotted as 1 and 2, respectively.EDS scanned the two spots and the results are shown in figure 4(d).The location of spot 1 has a higher Fe content than that of spot 2. It should be highlighted that similar precipitates were rarely observed in sample 0#, sample 2# and sample 3#.One possible reason is that only sample 1# had pore sizes exceeding 45 μm before copper infiltration.

Fracture morphology
Figure 5 shows the impact fractured surfaces of the four samples after copper infiltration.Characteristic ductile dimples, cleavage surfaces and pores can be seen in all four samples.As shown in figure 5(a), sample 0# displays a ductile fracture with cleavage surfaces and pores.The fracture surface of sample 1# shows a massive pearlite fracture due to cleavage and some ductile dimples.As the particle size decreases, there is a reduction in the dimensions of the cleavage surfaces and dimples seen in samples 2# and 3#.
Figure 6 shows the microstructure near the fracture surfaces of the four samples.In sample 0#, the fracture surface appears relatively flat, with pores and plastic deformation of the copper near the crack path.In sample 1#, a strong twist is observed in the pearlite structure along the crack path, indicating significant cracking and plastic deformation of the material.In sample 2#, the crack runs mainly along the iron-copper interface and the  particle shapes are visible.In sample 3#, due to the small initial iron particles and a higher degree of microstructural homogeneity, the failure is concentrated in the ductile copper region.The above-described difference in pore structure between the four samples is the key factor explaining the impact behavior after copper infiltration.

Discussion
The experimental results show that the pore structure affects the impact toughness, microstructure and microstructure before and after copper infiltration.

Impact toughness
The presence of pores with different morphology can lead to a reduction of the load-bearing area compared to the theoretical cross-section.Therefore, it is insufficient to accurately evaluate the mechanical properties of materials by porosity [26].To quantify the load-bearing area of specimens, an empirical method proposed by Molinari [27] can be used.In this method, mechanical properties are predicted based on pore morphology parameters, Φ represents the fraction of the load-bearing area, which is calculated using equation (1) and (2).Here ε stands for the fractional porosity and f circle describes the pore morphologies.
Where A and P are the area and perimeter of the pores in the metallographic image, respectively.As the value of f circle approaches 1, the morphology of the pore increasingly resembles that of a circular shape.By calculating the load-bearing cross-section of the samples before copper infiltration, the data for f circle and Φ were obtained.The results in table 4 show that as the average particle size of the screened samples decreases, the load-bearing portion of the material gradually increases.This is consistent with the trend shown in figure 3 for the samples before copper infiltration, as the synergistic effect of near-circular pore morphology and particle size refinement significantly increases the toughness of the samples.The impact toughness results of the sintered steel after copper infiltration indicate that the particle size distribution of the iron powder greatly affects the toughness.The infiltration behavior of copper, which is influenced by various pore structures, in turn affects the toughness of the material.The relationship between the microstructure and fracture behavior of sintered steel is of great significance.

Microstructure of the large particle sample after copper infiltration
A further analysis using TEM was carried out on sample 1#.The TEM images are shown in figure 7(a).The results of the elemental analysis of sample 1# are shown in figures 7(a1) and (a2).Numerous iron-rich precipitates can be seen in the copper phase region.The TEM micrograph in figure 7(b) reveals a coffee bean-like shape with an approximate particle size of 50 nm of a single iron precipitate, which has a beam orientation in the [110] Cu direction.The dashed region in figure 7(b) was analyzed using the Fast Fourier Transform (FFT).Figure 7(c) shows the observation of two groups of diffraction spots with the same orientation.According to reference [28], the lattice constant of Cu (FCC) is 0.3615 nm, while the lattice constant of γ-Fe is 0.3562 nm.The lattice mismatch between Cu (FCC) and γ-Fe is only 0.007.Thus, the conclusion can be drawn that the precipitates in sample 1# are γ-Fe particles, which are uniformly distributed in the central region of the copper phase at the nanoscale.Figure 8 shows a schematic representation of the solidification process to better explain this phenomenon.When the copper phase is larger, the solidification rate varies due to the high thermal conductivity, leading to the absorption of the majority of the latent heat generated during solidification by copper.However, when the surface area of the copper phase is reduced, the solidification speed exhibits little variation, hence decreasing the likelihood of γ-Fe precipitates in the molten state.The solidification process of the copper phase in the center area exhibits a slower rate, thereby offering an extended duration for the precipitation of γ-Fe particles.Consequently, the presence of precipitation is apparent inside the center area of the copper phase, whereas its absence can be observed at the edge.

Impact fracture
Under impact loads, copper-infiltrated steel exhibits three fracture modes: the fracture of the copper phase, the transgranular fracture of the iron particle, and the intergranular fracture between the iron particle and copper phase.The observed differences in impact toughness after copper infiltration can be explained by analyzing these fracture modes.When subjected to impact, the copper phase and the iron particles jointly carry the external load.The presence of larger particles results in the creation of larger pore sizes, which in turn are transformed into copper phases of larger dimensions after copper infiltration.This transformation enables the transfer of the load to larger iron particles.The large and concentrated copper phases, along with the large perlite blocks, have a favorable influence on the impact toughness.Additionally, the precipitation of γ-Fe particles within the locally aggregated copper phase may further enhance the material's toughness.Compared to other samples, sample 1# contains large perlite blocks and concentrated copper phases, respectively.Therefore, it primarily undergoes transgranular fracture through large iron particles during the fracture process, resulting in the highest energy absorption.Figure 6 demonstrates that the fracture mode of sample 1# is primarily characterized by the tearing of larger particles and the fracture along the copper infiltration phase.In contrast, the fracture modes of sample 2# and sample 3# mainly entail fractures along the copper infiltration phase.Due to its larger specific surface area, sample 3# exhibits a longer crack propagation path along the copper infiltration phase, contributing to its greater toughness compared to sample 2#.

Conclusion
In this study, the pore structure of the skeleton was regulated by varying the particle size of iron powder used, and the impact toughness of copper-infiltrated sintered steels was investigated.The results of the investigation can be summarized as follows: • The impact toughness of skeletons prepared with commercial iron powder was found to be 7.92 J cm −2 .As the particle size distribution narrows, the impact toughness increases.The impact toughness was significantly enhanced by reducing the particle size of iron powder from 150 μm to 40 μm, increasing from 11.71 J cm −2 to 21.41 J cm −2 .
• After copper infiltration, the samples with concentrated iron particle sizes show a significant improvement in impact toughness.The large copper-phase sample with the particle size of 150 μm achieved the highest impact toughness of 41.09 J cm −2 , which is 1.5 times higher than that of the sample made from commercial iron powders, which has an impact toughness of 27.46 J cm −2 .
• The precipitation of γ-Fe particles inside the central region of the large copper-phase becomes evident, and a simple model was established to explain the mechanism of γ-Fe precipitation.• The large copper-phase sample with the particle size of 150 μm primarily exhibits a fracture mode characterized by transgranular fracture through large iron particles, accompanied by intergranular and copper phase fracture.This improvement is mainly attributed to the precipitation of γ-Fe in the large copper-phase, which enhances the copper phase and transfers the load into large iron particles.

Table 3 .
Porosity of the four skeletons.

Table 4 .
Φ and f circle of samples before copper infiltration.