Understanding of dynamical responses and mechanical characteristics of metals and alloys at high strain rates holds significant importance in fundamental physics and optimizing the performance capabilities of materials. During high-speed impact scenarios, materials may be subjected to high pressure and plastic deformation, which have the potential to modulate their mechanical attributes. In this study, high-speed planar impact experiments were conducted to investigate the progressive alterations in the microstructures and mechanical properties in coarse-grain body-centered cubic (bcc) iron subjected to high-strain-rate (approximately 2.60–3.89 × 106 s−1) impact reaching approximately 15  GPa in a one-stage light-gas gun. The nanoindentation tests show that the nano-hardness of the post-shock iron improves 1.5 times from approximately 1.75–2.70 GPa. Microscopic analyses of the post-shock bcc-iron show no significant grain refinement but a noticeable increase in the twin boundaries (TBs) and low angle grain boundaries (LAGBs) proportion with increasing shock pressure. Therefore, the interaction between TBs, LAGBs, and dislocations in post-shock iron grains plays an important role in mediating its mechanical properties. Our findings serve as possible guidance for exploring the mechanical properties of single-crystalline and poly-crystalline iron-based materials, such as steel, with optimized mechanical performance.

Severe plastic deformation can induce lattice distortion, which subsequently causes alterations in the microstructures of metals. These modifications encompass the processes of grain refinement and the incorporation of structural imperfections, such as twins, textures, and dislocations.1–3 These refined grains and defects possess the capability to modify the hardness, corrosion resistance, and other exceptional mechanical properties of metals.4,5 As a result, grain refinement, surface defects between two adjacent grains, and twins can play a crucial role in influencing the plastic deformation behavior and mechanical characteristics of materials.6–8 Numerous studies have indicated that metals can be strengthened by grain refinement.9 Nevertheless, the effects of defects in metals, including TBs, LAGBs, and dislocations, on their mechanical behavior remain limited and constrained, thereby impeding a comprehensive understanding of the microstructure–property relationship in metals.

For bcc metals and alloys, the deformation twin is known as an extreme deformation mode that occurs when dislocations are not enough to carry the plastic strain.10–12 Twinning is a typical mode of strain energy relaxation with a unique displacive manner as compared with dislocation slip. Many dislocations can lead to a reduction in the static strength of metals. Meanwhile, TBs can hinder dislocation motion and, thus, act as stable interfaces for strengthening metals.13,14 As an important deformation mode that competes with dislocation slip, TBs in metals can also enhance the properties of metals, such as strength, hardness, and ductility, due to their minimal grain boundary energy.14–17 Therefore, we chose a typical bcc transition metal, iron (Fe), to investigate how interacts between its twins, grain boundaries, and dislocations in the grains during deformation without grain refinement under high strain rates.18 

Iron, the most abundant transition metal on the Earth and a primary constituent in the cores of terrestrial planets19,20 holds considerable significance in industrial applications due to its exceptional ductility, high strength, ease of alloying, and magnetism.21–23 Extensive studies have been conducted to investigate the physical and mechanical properties of iron under extreme pressure–temperature conditions.24–26 One of the important investigations focuses on comprehending the relationships between microstructures and bulk mechanical properties of iron subjected to severe plastic deformation, which holds great significance in the development of new advanced materials.

Because of the high stacking fault energy, it is challenging to generate twins in bulk bcc iron via the growth methods or mechanical deformation strategies.27 Most experimental investigations on deformation twins have primarily focused on alloys and nanoscale bcc metals.28–30 The studies on twins in pure bcc metals, however, have mainly emphasized theoretical approaches.12,28,31 Dieter32 reported that the density of deformation twins increases in shocked iron as the strain rate intensifies. Then, dynamic deformation twinning is observed in shock-deformed bcc-iron,33 where the volume fraction of twins monotonically increases from 0 to about 4%. It is interesting to note that plasticity is not primarily influenced by twinning when the shock pressure is below 25 GPa.34 High strain rate impact experiments have been proven to be an effective technique for generating twinning in bcc metals. However, the contribution of twins and LAGBs to the mechanical properties of bcc metals remains highly mysterious.

In this study, a high-speed impact method was employed to establish experimental conditions conducive to high strain rates. We conducted a series of shock recovery experiments on large-grain pure iron with an average grain size exceeding 1000 μm. The changes in the microstructures and mechanical properties of the post-shock iron were then analyzed. Our results show that many twinning and low-angle grain boundaries (hybrid low-energy interface architectures) were introduced into shocked iron. Meanwhile, the post-shock iron from 15 GPa shows an approximately 54% increase in nano-hardness. Our findings indicate that the increase in twins and LAGBs could be a primary contributor to elevating the hardness of post-iron. This study underscores the significance of deformation twinning and LAGBs in bcc-iron as a mechanism for enhancing mechanical properties, emphasizing its effectiveness comparable to traditional grain refinement techniques.

A coarse-grained polycrystalline iron sample, with a chemical purity exceeding 99.9%, was chosen as the starting material for the specimen. The grains in this material possessed sizes greater than 1000 μm. Shock compression experiments were carried out utilizing a one-stage light-gas gun with 24-mm diameter at the Institute of Atomic and Molecular Physics, Sichuan University. A schematic representation of the shock procedure is depicted in Fig. 1(a), and the post-shock samples were recovered. Prior to the experiments, the iron specimens were precision-cut into discs measuring approximately 12 mm in diameter and 2 mm in thickness. Both surfaces of the initial specimens were meticulously polished to achieve a mirror-like finish. Additionally, pure iron served as the flyer plate, characterized by dimensions of approximately 23 mm in diameter and 2 mm in thickness. The impact velocity of the flyer plate was accurately measured using an electromagnetic method with an uncertainty of approximately 0.5%.36,37

FIG. 1.

(a) Schematic diagram of shock experiments by planar impact; (b) the SEM map of the initial sample; (c) XRD patterns of the starting and recovered iron from shock compression; (d) typically particle velocity profile measured at the free surface of iron under planar impact of approximately 613 m/s by Fe flyer, where the final pressure could be reached in ∼30 ns. Hugoniot elastic limit (HEL) of the shock-compressed iron was observed at ∼50 m/s.35 XRD peaks of the initial and recovered iron could all be indexed with a bcc phase structure.

FIG. 1.

(a) Schematic diagram of shock experiments by planar impact; (b) the SEM map of the initial sample; (c) XRD patterns of the starting and recovered iron from shock compression; (d) typically particle velocity profile measured at the free surface of iron under planar impact of approximately 613 m/s by Fe flyer, where the final pressure could be reached in ∼30 ns. Hugoniot elastic limit (HEL) of the shock-compressed iron was observed at ∼50 m/s.35 XRD peaks of the initial and recovered iron could all be indexed with a bcc phase structure.

Close modal

The shock pressure (PH) was determined using the impedance matching method,38,39 specifically: PH = ρ0usup = ρ0(C0 + λus)up, where ρ0 represents the sample's initial density, us denotes the shock wave velocity, up is the particle velocity, and C0 and λ represent the fitted factors. The C0 and λ for bcc iron was determined to be 4.63 km/s and 1.33, respectively, so the Hugoniot parameters used for iron are us = 4.63 + 1.33up (km/s) with a density of 7.875 g/cm3.35 In this experiment, the flyer plate was identical to the sample, and the experimental setup followed a symmetrical impact configuration, resulting in up being half of the impact velocity (w). The specific impact conditions are detailed in Table I, where the iron was subjected to shock pressures and strain rates ranging from 10 to 15 GPa and 2.60 to 3.89 × 106 s−1, respectively.

TABLE I.

Planar impact conditions and experimental results for polycrystalline iron under shock compression. Note: P and T are the Hugoniot pressure and temperature in the iron sample under single shock compression, respectively; ɛ is the estimated strain rate of iron under shock compression in this study.40  Hd is the hardness of the recovered samples; Rs is the residual stress of the recovered samples.

No.Sample/flyerw (km/s)PH (GPa)T (K)ɛ (×106 s−1)Hd (GPa)Rs (GPa)
1(initial) … 300 1.75(3) − 
Fe/Fe 0.512(1) 10.0(2) 329(7) 2.60 1.91(8) −0.077(3) 
Fe/Fe 0.594(1) 11.8(2) 335(8) 3.05 2.17(7) −0.182(6) 
Fe/Fe 0.683(1) 13.7(3) 342(10) 3.53 2.61(6) −0.387(9) 
Fe/Fe 0.746(2) 15.1(3) 347(10) 3.89 2.70(5) −0.763(14) 
No.Sample/flyerw (km/s)PH (GPa)T (K)ɛ (×106 s−1)Hd (GPa)Rs (GPa)
1(initial) … 300 1.75(3) − 
Fe/Fe 0.512(1) 10.0(2) 329(7) 2.60 1.91(8) −0.077(3) 
Fe/Fe 0.594(1) 11.8(2) 335(8) 3.05 2.17(7) −0.182(6) 
Fe/Fe 0.683(1) 13.7(3) 342(10) 3.53 2.61(6) −0.387(9) 
Fe/Fe 0.746(2) 15.1(3) 347(10) 3.89 2.70(5) −0.763(14) 

The post-shock samples were sectioned into cross section (1 mm in size) parallel to the direction of impact. Nanoindentation tests were conducted employing a Nano Test Vantage system from Micro Material. Three random points were selected on the cross section of each sample for nanoindentation measurements, which were then averaged. A maximal force of 50 mN was applied, with a loading-unloading cycle lasting 20 s and a 2-s pause at maximum load. It is noteworthy that the spacing between indentations (40 μm) exceeded three times the characteristic length of residual imprints (<10 μm), which prevents perturbations arising from the mechanically affected zone of adjacent tests. Subsequently, the microstructures, grain size, and dislocation density of the post-shock iron were comprehensively examined using electron backscatter diffraction (EBSD) with two distinct indexation step sizes. A step size of 3 μm was used for broader-scale observations, whereas a finer step size of 0.5 μm was employed to detect finer grains.

In addition, x-ray diffraction (XRD), energy dispersive spectrometer (EDS), and scanning electron microscope (SEM) measurements were carried out to ascertain the crystal structure, elemental component, and uniformity of initial samples, respectively. The SEM map verified the homogeneity of the initial pure iron sample [Fig. 1(b)]. As shown in Fig. 1(c), XRD peaks of starting and post-shock samples are in good agreement with the standard peaks of the bcc-iron. The intensity of the (110) peak of compressed irons is significantly reduced compared to that of the starting iron, likely due to the introduction of many dislocations under high strain rates.41 

Considering the limitations imposed by the sample size, nanoindentation measurements have emerged as a preferred technique for ascertaining the nano-hardness of materials on a microscale. Figure 2(a) illustrates typical load/unload-depth curves of post-shock irons. A comparison of load/unload-depth curves in starting and post-shock irons reveals that the post-shock irons exhibit shallower profiles [Fig. 2(b)]. Moreover, the loading depth gradually decreases with the increasing strain rate. Consequently, the nanoindentation tests indicate that the nano-hardness of post-shock iron rises with pressure, peaking at 2.70 GPa [Fig. 2(c)], recovered from the highest shock pressure of 15.1 GPa. In addition, the residual stress of the post-shock samples can be derived from the nanoindentation measurement [Fig. 2(d)], which suggests that the samples subjected to shock show an improved capacity to resist deformation.

FIG. 2.

Nanoindentation tests of the post-shock irons recovered from varied pressures: (a) the typical loading and unloading profile curves for three random points conducted on a cross-sectional sample, characterized by a shock pressure of 15.1 GPa, (b) the average data of the loading/unloading-depth profile curves, (c) hardness distribution, and (d) residual stress of post-shock irons. It is evident that the nano-hardness and residual stress values exhibit varying for the samples processed at different strain rates.

FIG. 2.

Nanoindentation tests of the post-shock irons recovered from varied pressures: (a) the typical loading and unloading profile curves for three random points conducted on a cross-sectional sample, characterized by a shock pressure of 15.1 GPa, (b) the average data of the loading/unloading-depth profile curves, (c) hardness distribution, and (d) residual stress of post-shock irons. It is evident that the nano-hardness and residual stress values exhibit varying for the samples processed at different strain rates.

Close modal

Previous studies have shown that the hardness of pure iron at room temperature ranges from 170 HV (1.7 GPa) to 260 HV (2.5 GPa) for grain sizes ranging from 1500 to 200 nm.42–44 Meanwhile, in this study, the nano-hardness of post-shock iron reached 2.7 GPa. Our findings indicate that the effect of the interaction between twinning and dislocations on the mechanical behavior of polycrystalline materials is similar in importance to that of the grain refinement mechanism. The improvement of mechanical properties of post-shock iron may be linked to the augmentation of twinning, LAGBs, and dislocation density, achieved by integrating TBs and LAGBs statistics and dislocation density. The increase in residual stress in the sample is likely due to the development of compressive stresses that can resist impact deformation after the sample has undergone plastic deformation.

To quantitatively study the dislocation density changes of post-shock iron, the Kernel Average Misorientation (KAM) method was adopted to determine the local misorientation based on the EBSD orientation data.45,46 The geometrically necessary dislocations (GND) can be obtained from the measured local crystal orientations on the specimen surface. The KAM could present the average misorientation between a given point and its nearest neighbors in a grain. Any local misorientation angle calculated greater than 3° was excluded in this analysis. Therefore, the KAM histogram was used to evaluate the plastic strain in the post-shock samples. The local misorientation at a region measuring 300 × 300 nm2 was then determined by analyzing its 24 surrounding points,
Δ θ i = 1 n j = 1 n | θ j sur θ i | ,
(1)
where θi shows the local misorientation at the point “i” and θ j sur represents the misorientation at its neighboring point “j.” To infer the GND density information, a simple method based on the strain gradient theory was adopted,47,
ρ GND = 2 Δ θ i u b = B Δ θ i ,
(2)
where ρGND represents the GND density at the point of interest; Δθi is the local misorientation; u shows the unit length of the point (300 nm); and b represents the Burgers vector (0.248 nm for iron).48  B = 2.688 × 1016 m−2 is a constant for iron. The distribution of GND density for each post-shock sample is expected (Fig. 3).
FIG. 3.

The geometrically necessary dislocations density maps of (a) starting iron, and post-shock irons from (b) 10.0, and (c) 15.1 GPa. Color scale indicates the dislocation densities in the starting and post-shock iron. Global GND density distribution based on corresponding mapping results in (d)–(f). The dislocation density of recovered samples is proportional to pressure, resulting in hardening of pure irons after deformation.

FIG. 3.

The geometrically necessary dislocations density maps of (a) starting iron, and post-shock irons from (b) 10.0, and (c) 15.1 GPa. Color scale indicates the dislocation densities in the starting and post-shock iron. Global GND density distribution based on corresponding mapping results in (d)–(f). The dislocation density of recovered samples is proportional to pressure, resulting in hardening of pure irons after deformation.

Close modal

The KAM mappings and their corresponding KAM values vs frequency for all samples are depicted in Figs. 3(a)3(f), respectively. The observed high kernel values are found in post-shock samples rather than in the initial sample, and the GND density variation increases with the degree of compression. This phenomenon can be classified as a genuine deformation, as opposed to a mere measurement error.49,50 The histograms reveal that the initial sample spans a broad range with a relatively low concentration of GNDs. As a result, the dislocation density within the post-shock samples is significantly elevated. Notably, the average GND density increases to 1.67 × 1016 m−2 in post-shock iron from the shock pressure of 15.1 GPa, which signifies a 65% enhancement relative to the initial material's density.

The source of dislocations is activated by stress concentration of stacked dislocations at grain boundaries.51,52 Due to the presence of a low dislocation density in the initial sample, the samples underwent a dislocation surge and slip phenomenon during the deformation process. Lattice deformation is expected to occur in the microstructure of iron during shock compression, resulting in the increase of dislocation density. Beyond a critical point, interactions among the dislocations can hinder their mobility, making the material deformation more difficult. During deformation, dislocations can intersect, forming new dislocation loops. These loops then interact with other dislocations, triggering a reorganization of the existing dislocation network. This complex process of recombination ultimately may lead to the annihilation of certain dislocations.53,54 Furthermore, the material exhibits a cellular substructure, which is characterized by the presence of high-density dislocation walls. Indeed, the formation of sub-grain boundaries with lower orientation angles is attributed to the recombination of high-density dislocations.

The formation and evolution of grain/sub-grain boundaries in post-shock irons were investigated based on EBSD analyses. The grain/sub-grain boundary (GSB) includes high angle grain boundaries (HAGBs) between 15° and 62.8° and LAGB between 2° and 15°, as well as TBs [Figs. 4(a)4(c)]. The GSB maps show that the grain size of the post-shock iron was not significantly changed. However, a conspicuous abundance of densely populated LAGBs and TBs was observed within all grains after the impact-induced deformation [Figs. 4(b) and 4(c)]. These LAGBs can be attributed to the rearrangement and migration of accumulated dislocations.

FIG. 4.

Grain/sub-grain boundary maps (upper) and histograms (below) of the cross section of post-shock iron from different pressures: (a) and (d) 0; (b) and (e) 10.0; (c) and (f) 15.1 GPa. Black, green, and red lines in maps indicate high angle (θ ≥ 15°), low angle (2° ≤ θ < 15°), and near-twin boundaries in terms of the Brandon criterion,55 respectively; The black arrow indicates the direction of impact. The number of low angle grain boundaries and twin boundaries is proportional to pressure, and no refined grains are found.

FIG. 4.

Grain/sub-grain boundary maps (upper) and histograms (below) of the cross section of post-shock iron from different pressures: (a) and (d) 0; (b) and (e) 10.0; (c) and (f) 15.1 GPa. Black, green, and red lines in maps indicate high angle (θ ≥ 15°), low angle (2° ≤ θ < 15°), and near-twin boundaries in terms of the Brandon criterion,55 respectively; The black arrow indicates the direction of impact. The number of low angle grain boundaries and twin boundaries is proportional to pressure, and no refined grains are found.

Close modal

Grain boundaries, which serve as origins of dislocations, could produce slip bands and, thus, contribute to the occurrence of LAGBs along these boundaries. This, in turn, leads to an associated rise in stress levels, generating areas of strain concentration at these grain boundaries. LAGBs tend to initially appear near HAGBs, which is closely related to the motion of dislocations and slip systems.56 Therefore, the substructures were likely formed in grains that possess a higher quantity of GNDs.

The distribution histograms of grain/sub-grain boundary for post-shock irons from different pressures are illustrated in Figs. 4(d)4(f), where the term “Correlated” signifies the misorientation calculated employing adjacent data points, the “Uncorrelated” lines indicate misorientation calculated based on random data points from the scan, and the “Random” lines represent the theoretically calculated random misorientation distribution of a purely random texture.57 The misorientation angles in the “Correlated” data in the initial sample are distributed evenly across the range from 2° to 62° [Fig. 4(d)]. Nevertheless, the highly deformed samples exist many low misorientation angles, which were lower than 5° [Figs. 4(e) and 4(f)], similar findings were reported in other studies.58,59 The variation tendency of LAGBs is consistent with maps. Meanwhile, the deviation between the “Uncorrected” line and “Random” curve indicates the presence of preferred orientations in the examined samples.

Figure 5 illustrates the process of texture change with increasing shock pressure via {100} pole figures (PF) and inverse pole figures (IPF). As evident from the presented figures, a clear reduction in the intensity of texture orientation was observed as the strain rate rises. The initial sample exhibits a weak cubic texture characterized by the presence of { 111 } 011 orientation and 101 fiber orientation [Fig. 5(a)], and it exhibits the highest 223 fiber orientation strength. However, the fiber with an index of 223 underwent transformation in the deformed samples, which gradually transfers toward the 001 orientation ( 223 fiber →  113 fiber →  114 fiber). The emergence of the new { 001 } 100 orientation and 111 fiber orientation is observed in samples subjected to high pressures. Additionally, the maximum intensity of the { 001 } 100 orientation decreases as the shock pressure and strain rate increases.

FIG. 5.

The {100} pole figures (left) and inverse pole figures (right) of post-shock bcc-iron from different pressures: (a) and (b) 0; (c) and (d) 10.0; (e) and (f) 15.1 GPa. Color scale indicates the texture strength in the starting and post-shock iron. The maximum intensity of pole figures and inverse pole figures exhibits a gradual decrease from 24.7 and 5.3 to 21.4 and 3.3, respectively, as the pressure increases. Additionally, there is a gradual increase in the variety of textures observed.

FIG. 5.

The {100} pole figures (left) and inverse pole figures (right) of post-shock bcc-iron from different pressures: (a) and (b) 0; (c) and (d) 10.0; (e) and (f) 15.1 GPa. Color scale indicates the texture strength in the starting and post-shock iron. The maximum intensity of pole figures and inverse pole figures exhibits a gradual decrease from 24.7 and 5.3 to 21.4 and 3.3, respectively, as the pressure increases. Additionally, there is a gradual increase in the variety of textures observed.

Close modal

The three-dimensional [Figs. 6(a)6(c)] and the ɸ2 = 45° sections [Figs. 6(d)6(f)] of the orientation distribution function (ODF) are presented to investigate the potential changes in textures before and after deformation of iron. Despite the textures in iron are relatively weak, the changes between the initial and shocked samples can still be observed. Specifically, the intensity of texture orientation noticeably decreases as pressure increases, which aligns with the findings of PFs and IPFs. Furthermore, the initial sample no longer exhibits the rolling texture, and instead, the {111} texture is found in the sample recovered from the shock pressure of 10.0 GPa. The decrease in intensity of texture suggests a shift toward a more random orientation of the shocked material, thereby facilitating the initiation of multiple slip systems and subsequent plastic deformation.60,61

FIG. 6.

The three-dimensional (a)–(c) and ɸ2 = 45° (d)–(f) of orientation distribution function sections for bcc iron, at (a) and (d) 0, (c) and (e) 10.0, (c) and (f) 15.1 GPa. An increase in the diversity of texture types coupled with a significant decrease in strength.

FIG. 6.

The three-dimensional (a)–(c) and ɸ2 = 45° (d)–(f) of orientation distribution function sections for bcc iron, at (a) and (d) 0, (c) and (e) 10.0, (c) and (f) 15.1 GPa. An increase in the diversity of texture types coupled with a significant decrease in strength.

Close modal

When a dislocation moves along a specific direction on the slip plane, it leaves a trace, called a slip trace. This slip trace can be used to identify the slip system of the material. Traditional slip theory posits that the slip system with the highest Schmid Factor (SF) is the initial one to undergo slip. The determination of the activation state of the slip system necessitates the computation of the SFs linked to different slip systems. The activation of slip systems in polycrystalline materials is governed by the maximum shear stress and the interplay among adjacent grains. Therefore, in situations where the SFs of two slip systems are high and converging during the deformation process, conventional approaches prove inadequate in accurately determining the slip system that is initially activated.

As depicted in Figs. 7(c) and 7(d), grain A exhibits an orientation characterized by two crystal planes, namely, (110) and ( 1 ¯ 10 ). We have observed that the theoretical trace of the (110) plane in grain A is parallel to L2, while the theoretical trace of the ( 1 ¯ 10 ) plane is parallel to L1. The slip system of grain B was also verified utilizing the identical method, as shown in Figs. 7(e) and 7(f). It can be inferred that the formation of slip bands L1 and L2 is attributed to the displacement of the slip system along the ( 1 ¯ 10 ) and (110) slip planes, respectively. It is noteworthy that the SFs within the grains demonstrate a consistent behavior across the crystal planes that belong to the same family. This similarity can be attributed to the close proximity of the loading direction of the shock wave to L1 and L2. Therefore, we analyzed the strength of only one slip plane, where Figs. 7(g)7(j) illustrates the theoretical traces of the slip planes (112) and (123) in grain A and B, respectively.

FIG. 7.

The band slope images of post-shock irons recovered from the shock pressure of 10.0 (left) and 15.1 GPa (right). Theoretical traces of two slip planes of grain A and B: (a) 10.0; (b) 15.1; (c) and (e) (110); (d) and (f) ( 1 ¯ 10 ), (g) and (i) (112); (h) and (j) (123). The slip band L1 and L2 belong to grain A, and the slip band L3 and L4 are in grain B.

FIG. 7.

The band slope images of post-shock irons recovered from the shock pressure of 10.0 (left) and 15.1 GPa (right). Theoretical traces of two slip planes of grain A and B: (a) 10.0; (b) 15.1; (c) and (e) (110); (d) and (f) ( 1 ¯ 10 ), (g) and (i) (112); (h) and (j) (123). The slip band L1 and L2 belong to grain A, and the slip band L3 and L4 are in grain B.

Close modal

Our results indicate that three slip systems of bcc-iron were activated during high-speed impact, necessitating an assessment of their SFs. The average SFs for these slip planes, corresponding to the 111 crystallographic direction under varying pressures, are presented in Table II. By comparing the three slip systems, it was observed that the activation of the { 123 } 111 slip system serves as the primary mechanism during high-speed impact in the initial sample.

TABLE II.

Mean Schmid factors of the three slip planes at the 111 directionof the starting and post-shock irons.

Slip plane0 GPa10.0 GPa15.1 GPa
{110} 0.422 0.438 0.458 
{112} 0.430 0.446 0.476 
{123} 0.438 0.459 0.475 
Slip plane0 GPa10.0 GPa15.1 GPa
{110} 0.422 0.438 0.458 
{112} 0.430 0.446 0.476 
{123} 0.438 0.459 0.475 

The occurrence of twinning was not observed in previous studies on the microplastic deformation and low strain rate behavior of polycrystalline iron.62–65 Twinning in single-crystalline iron is likely to occur under the conditions of exceptionally high strain rates and low temperatures, preceding the plastic deformation.66 In this study, a significant number of deformation twins and LAGBs were observed in the post-shock irons [Figs. 4(b) and 4(c)]. The statistical analyses reveal that the length fraction of the TBs in iron recovered from 10.0 and 15.1 GPa is 9.79% and 21.30%, respectively. Additionally, a weaker 111 fiber texture was observed in the impacted samples, which could be attributed to the occurrence of abundant twinning along the migration front of the abnormally growing grains (Fig. 8).67 

FIG. 8.

The texture component maps (left) and inverse pole figures (right) of post-shock irons recovered from different shock pressures: (a) and (b) 10.0, (c) and (d) 15.1 GPa. Note: The red arrow in the left images points toward the texture inside the twin, which is consistent with the direction shown in the right images.

FIG. 8.

The texture component maps (left) and inverse pole figures (right) of post-shock irons recovered from different shock pressures: (a) and (b) 10.0, (c) and (d) 15.1 GPa. Note: The red arrow in the left images points toward the texture inside the twin, which is consistent with the direction shown in the right images.

Close modal

The formation of twins can lead to a diminished strength in the crystallographic texture of compressed samples, while simultaneously mitigating stress concentrations that arise from the complexities of dislocation slip. Twins are more likely to form near grain boundaries due to grain deformation, which causes relative slip between grains and subsequent slip of dislocation lines or planes and results in the formation of twinning structures.68,69 Figure 9 illustrates the notable emergence of LAGBs and twin structures within the deformed iron. Regions without grain boundaries primarily underwent deformation through the formation of these small-angle boundaries upon a single impact, while twin was mainly formed in the vicinity of grain boundaries.

FIG. 9.

Schematic diagram of the formation of low angle grain boundaries and twins: (a) In regions that are remote from grain boundaries, (b) the distribution of defects at location (a) after impact compression, (c) in areas proximal to grain boundaries, (d) the distribution of defects at location (b) after impact compression. In the figures, black, gray, green, and red lines represent grain boundaries, dislocations, LAGBs, and twins, respectively.

FIG. 9.

Schematic diagram of the formation of low angle grain boundaries and twins: (a) In regions that are remote from grain boundaries, (b) the distribution of defects at location (a) after impact compression, (c) in areas proximal to grain boundaries, (d) the distribution of defects at location (b) after impact compression. In the figures, black, gray, green, and red lines represent grain boundaries, dislocations, LAGBs, and twins, respectively.

Close modal

A significant number of 60° { 112 } 111 deformation twins were observed with an average size of 1.5 μm in the post-shock iron [Figs. 10(c) and 10(d)]. Through careful analysis, three distinct microstructural features have been observed in relation to twin–twin interaction [Fig. 10(a)]. These features include the quilted twin structure (T1 and T2), the “apparent crossing” twin structure (T3), and the double twin structure (T4 and T5). The quilted twin structure is normally formed through the transmission and interruption of multiple twin mutations. A crossing twin structure occurs when one twin intersects another by following a secondary twinning path within the crystal lattice. The double twin structure refers to the occurrence of a secondary twin formation within a primary twin domain. Twin knots play a distinctive role in twinning and detwinning during deformation, influencing local stress fields and playing a crucial part in their development. These factors inevitably impact the mechanical characteristics of materials, including strain hardening, crack initiation, and resistance to failure.70,71

FIG. 10.

Example of twin type in post-shock iron recovered from 15.1 GPa: (a) Diagram of Euler angle2 of a locally enlarged view in the sample; T1 and T2 represent the quilted twin structure; T3 expresses the “apparent crossing” twin structure; T4 and T5 show the double twin structure. (b) The geometrically necessary dislocations density maps, the red arrow refers to the highest point of dislocation density, which is located at the intersection of twins. (c) The misorientation of { 112 } 111 deformation twins (No. 1–14) boundaries of the dashed line in (a), and the misorientation of twins is 60°. (d) The sizes of { 112 } 111 deformation twins of No. 1–14 in (c), and the average size of twins are 1.5 μm.

FIG. 10.

Example of twin type in post-shock iron recovered from 15.1 GPa: (a) Diagram of Euler angle2 of a locally enlarged view in the sample; T1 and T2 represent the quilted twin structure; T3 expresses the “apparent crossing” twin structure; T4 and T5 show the double twin structure. (b) The geometrically necessary dislocations density maps, the red arrow refers to the highest point of dislocation density, which is located at the intersection of twins. (c) The misorientation of { 112 } 111 deformation twins (No. 1–14) boundaries of the dashed line in (a), and the misorientation of twins is 60°. (d) The sizes of { 112 } 111 deformation twins of No. 1–14 in (c), and the average size of twins are 1.5 μm.

Close modal

During the dynamic process of impact compression, the material experiences a proliferation of dislocations, which accumulate and coalesce, leading to the formation of LAGBs. In contrast to HAGBs, twin boundaries exhibit a higher degree of coherence and are less likely to cause significant lattice distortion. This inherent structural compatibility makes twin boundaries particularly effective in promoting the transfer of dislocation plasticity, thereby enhancing the material's capacity to withstand deformation under stress.72 The dislocation density is found to be significantly higher near the TBs and at the interactions of twins [Fig. 10(b)]. This phenomenon occurs due to the interaction of twinning, which leads to the formation of twin–twin boundaries. These boundaries have a significant impact on the twin and its growth process, which could enhance the dislocation strength by impeding twin reproduction and growth.73–75 

In this paper, the deformation behavior of large-grain pure iron was studied under high strain rates (∼106 s−1), with shock pressures reaching approximately 15 GPa. The main conclusions are summarized as follows:

  1. The nano-hardness of post-shock iron increases by 1.5 times, rising from around 1.75 to 2.70GPa.

  2. As strain rates increase, the proportion of low-angle grain boundaries rises in shocked-iron, a phenomenon closely related to dislocation movement within grains. Our analyses suggest that the entanglement and stacking resulting from dislocation interactions lead to the formation of low-angle grain boundaries, which, in turn, could hinder the movement of dislocations.

  3. The density of geometrically necessary dislocations is the highest around twin boundaries in post-shock iron, especially at the intersections of twins. As a result, a significant number of twin boundaries impede dislocation movement.

  4. The post-shock iron retains its initial bcc configuration with grain sizes minimal change. Consequently, the increase in its nano-hardness primarily hinges on the existence of twins and low-angle grain boundaries after shock.

We thank L. Y. Zhou, Y. F. Zhang, W. H. Song, J. Wu, and Q. M. Wang for the shock experiments. This work was supported by the National Natural Science Foundation of China (No. 42074298), the National Key Laboratory of Shock Wave and Detonation Physics (No. 2022JCJQLB05701), the Sichuan Science and Technology Program (No. 2023NSFSC1910), and the Institutional Research Fund from Sichuan University (No. 2022SCUNL102).

The authors have no conflicts to disclose.

Canlian Tang: Writing – original draft (equal). Bo Gan: Writing – review & editing (equal). Yukai Zhuang: Writing – review & editing (equal). Zhipeng Gao: Writing – review & editing (equal). Youjun Zhang: Writing – review & editing (equal).

The data that support the findings of this study are available within the article.

1.
X. D.
Zhang
,
N.
Hansen
,
Y. K.
Gao
, and
X. X.
Huang
,
Acta Mater.
60
,
5933
(
2012
).
2.
I.
Brooks
,
P.
Lin
,
G.
Palumbo
,
G. D.
Hibbard
, and
U.
Erb
,
Mater. Sci. Eng. A
491
,
412
(
2008
).
3.
B.
Sułkowski
and
R.
Chulist
,
Mater. Sci. Eng. A
749
,
89
(
2019
).
4.
X. L.
Zhou
,
Z. Q.
Feng
,
L. L.
Zhu
,
J. N.
Xu
,
L.
Miyagi
,
H. L.
Dong
,
H. W.
Sheng
,
Y. J.
Wang
,
Q.
Li
, and
Y. M.
Ma
,
Nature
579
,
67
(
2020
).
5.
Y. Q.
Li
,
P.
Shen
,
H. M.
Zhang
,
K.
Dong
,
Y.
Deng
,
X. W.
Chen
, and
Z. S.
Cui
,
Mater. Des.
210
,
110057
(
2021
).
6.
J.
Jeong
,
M.
Alfreider
,
R.
Konetschnik
,
D.
Kiener
, and
S. H.
Oh
,
Acta Mater.
158
,
407
(
2018
).
7.
Y.
Zhang
,
N. R.
Tao
, and
K.
Lu
,
Scr. Mater.
60
,
211
(
2009
).
8.
Y. D.
Wei
,
H. C.
Liao
, and
S. H.
Huo
,
Int. J. Fatigue
166
,
107293
(
2023
).
9.
A.
Chokshi
,
A.
Rosen
,
J.
Karch
, and
H.
Gleiter
,
Scr. Metall.
23
,
1679
(
1989
).
10.
X. Z.
Liao
,
Y. H.
Zhao
,
S. G.
Srinivasan
,
Y. T.
Zhu
,
R. Z.
Valiev
, and
D. V.
Gunderov
,
Appl. Phys. Lett.
84
,
592
(
2004
).
11.
Y. T.
Zhu
,
X. Z.
Liao
, and
X. L.
Wu
,
Prog. Mater. Sci.
57
,
1
(
2012
).
12.
C.
Huang
,
X. H.
Peng
,
T.
Fu
,
X.
Chen
,
H. G.
Xiang
,
Q. B.
Li
, and
N.
Hu
,
Mater. Sci. Eng. A
700
,
609
(
2017
).
13.
Z.
Cheng
,
H. F.
Zhou
,
Q. H.
Lu
,
H. J.
Gao
, and
L.
Lu
,
Science
362
,
eaau1925
(
2018
).
14.
L.
Lu
,
Y. F.
Shen
,
X. H.
Chen
,
L. H.
Qian
, and
K.
Lu
,
Science
304
,
422
(
2004
).
15.
Y. J.
Wei
,
H. J.
Gao
,
X. Y.
Li
,
L.
Lu
, and
K.
Lu
,
Nature
464
,
877
(
2010
).
16.
L.
Lu
,
X.
Chen
,
X.
Huang
, and
K.
Lu
,
Science
323
,
607
(
2009
).
17.
Y. J.
Tian
,
B.
Xu
,
D. L.
Yu
,
Y. M.
Ma
,
Y. B.
Wang
,
Y. B.
Jiang
,
W. T.
Hu
,
C. C.
Tang
,
Y. F.
Gao
,
K.
Luo
,
Z. S.
Zhao
,
L. M.
Wang
,
B.
Wen
,
J. L.
He
, and
Z. Y.
Liu
,
Nature
493
,
385
(
2013
).
18.
D. L.
Zou
,
L.
Zhen
,
Y.
Zhu
,
C. Y.
Xu
,
W. Z.
Shao
, and
B. J.
Pang
,
Mater. Sci. Eng. A
527
,
3323
(
2010
).
19.
T.
Sakai
,
T.
Kondo
,
E.
Ohtani
,
H.
Terasaki
,
N.
Endo
,
T.
Kuba
,
T.
Suzuki
, and
T.
Kikegawa
,
Geophys. Res. Lett.
33
, L15317, https://doi.org/10.1029/2006GL026868 (
2006
).
20.
L.
Dubrovinsky
,
N.
Dubrovinskaia
,
F.
Langenhorst
,
D.
Dobson
,
D.
Rubie
,
C.
Geßmann
,
I. A.
Abrikosov
,
B.
Johansson
,
V. I.
Baykov
, and
L.
Vitos
,
Nature
422
,
58
(
2003
).
21.
R. Z.
Valiev
and
I. V.
Alexandrov
,
Nanostruct. Mater.
12
,
35
(
1999
).
22.
I. V.
Alexandrov
and
R. Z.
Valiev
,
Scr. Mater.
44
,
1605
(
2001
).
23.
R. Z.
Valiev
,
Y. V.
Ivanisenko
,
E. F.
Rauch
, and
B.
Baudelet
,
Acta Mater.
44
,
4705
(
1996
).
25.
Y.
Ivanisenko
,
R. Z.
Valiev
, and
H.-J.
Fecht
,
Mater. Sci. Eng. A
390
,
159
(
2005
).
26.
C. T.
Seagle
,
E.
Cottrell
,
Y. W.
Fei
,
D. R.
Hummer
, and
V. B.
Prakapenka
,
Geophys. Res. Lett.
40
,
5377
, https://doi.org/10.1002/2013GL057930 (
2013
).
27.
S.
Ogata
,
J.
Li
, and
S.
Yip
,
Phys. Rev. B
71
,
224102
(
2005
).
28.
J. W.
Xiao
,
L. Y.
Zhu
,
R.
Wang
,
C.
Deng
,
Z. X.
Wu
, and
Y. T.
Zhu
,
Mater. Today
65
,
90
(
2023
).
29.
D.
Ma
,
H. J.
Xu
,
J. L.
Ding
,
Y.
Shen
, and
T.
Zhang
,
Mater. Sci. Eng. A
862
,
144489
(
2023
).
30.
L. G.
Zhao
,
G. X. J.
Chen
,
H.
Zheng
,
S. F.
Jia
,
K. X.
Li
,
R. H.
Jiang
,
L.
Li
,
Y.
Zhang
,
H. Y.
Peng
,
P. L.
Zhao
,
Z. Y.
Huang
, and
J. B.
Wang
,
J. Mater. Sci. Technol.
144
,
235
(
2023
).
31.
G.
Sainath
and
A.
Nagesha
,
Comput. Mater. Sci.
210
,
111449
(
2022
).
32.
G. E.
Dieter
,
Strengthening Mechanisms in Solids
(
ASM
,
Metals Park, OH
,
1962
), Vol.
279
.
33.
J. N.
Johnson
and
R. W.
Rohde
,
J. Appl. Phys.
42
,
4171
(
1971
).
34.
C.
Wehrenberg
,
D.
McGonegle
,
C.
Bolme
,
A.
Higginbotham
,
A.
Lazicki
,
H. J.
Lee
,
B.
Nagler
,
H.-S.
Park
,
B. A.
Remington
, and
R. E.
Rudd
,
Nature
550
,
496
(
2017
).
35.
L. M.
Barker
and
R. E.
Hollenbach
,
J. Appl. Phys.
45
,
4872
(
1974
).
36.
B.
Gan
,
Y. J.
Zhang
,
Y. Q.
Huang
,
X. H.
Li
,
Q. M.
Wang
,
J.
Li
,
Y. K.
Zhuang
,
Y.
Liu
, and
G.
Jiang
,
Geophys. Res. Lett.
48
,
e2021GL094446
, https://doi.org/10.1029/2021GL094446 (
2021
).
37.
Y. Q.
Huang
,
M. Q.
Hou
,
B.
Gan
,
X. H.
Li
,
D. W.
He
,
G.
Jiang
,
Y. J.
Zhang
, and
Y.
Liu
,
J. Geophys. Res.
127
,
e2021JB023645, https://doi.org/10.1029/2021JB023645
(
2022
).
38.
S. S.
Penner
,
J. Quant. Spectrosc. Radiat. Transfer
76
,
235
(
2003
).
39.
Y. B.
Zel Dovich
and
P.
Raizer Yu
, Physics of Shock Waves and High-Temperature Hydrodynamic Phenomena (Courier Corporation, 2002).
40.
R. F.
Smith
,
R. W.
Minich
,
R. E.
Rudd
,
J. H.
Eggert
,
C. A.
Bolme
,
S. L.
Brygoo
,
A. M.
Jones
, and
G. W.
Collins
,
Phys. Rev. B
86
,
245204
(
2012
).
41.
Y.
Yang
,
Z.
Jiang
,
C.
Wang
,
H. B.
Hu
,
T. G.
Tang
,
H. S.
Zhang
, and
Y. N.
Fu
,
Mater. Sci. Eng. A
731
,
385
(
2018
).
42.
B. Q.
Han
,
E. J.
Lavernia
, and
F. A.
Mohamed
,
Metall. Mater. Trans. A
34
,
71
(
2003
).
43.
Y.
Yan
,
Y.
Qi
,
Q.-W.
Jiang
, and
X.-W.
Li
,
Acta Metall. Sin. (Engl. Lett.)
28
,
531
(
2015
).
44.
Y.
Ding
,
J. H.
Jiang
, and
A. D.
Shan
,
Mater. Sci. Eng. A
509
,
76
(
2009
).
45.
M.
Calcagnotto
,
D.
Ponge
,
E.
Demir
, and
D.
Raabe
,
Mater. Sci. Eng. A
527
,
2738
(
2010
).
46.
P.
Chen
,
S. C.
Mao
,
Y.
Liu
,
F.
Wang
,
Y. F.
Zhang
,
Z.
Zhang
, and
X. D.
Han
,
Mater. Sci. Eng. A
580
,
114
(
2013
).
47.
L. P.
Kubin
and
A.
Mortensen
,
Scr. Mater.
48
,
119
(
2003
).
48.
Q.
Liu
,
L. M.
Fang
,
Z. W.
Xiong
,
J.
Yang
,
Y.
Tan
,
Y.
Liu
,
Y. J.
Zhang
,
Q.
Tan
,
C. C.
Hao
,
L. H.
Cao
,
J.
Li
, and
Z. P.
Gao
,
Mater. Sci. Eng. A
822
,
141704
(
2021
).
49.
A.
Kundu
,
D. P.
Field
, and
P.
Chandra Chakraborti
,
Mater. Sci. Eng. A
773
,
138854
(
2020
).
51.
V.
Moorthy
,
T.
Jayakumar
, and
B.
Raj
,
Int. J. Pressure Vessels Piping
64
,
161
(
1995
).
52.
R. C.
Bill
,
J. R.
Frederick
, and
D. K.
Felbeck
,
J. Mater. Sci.
14
,
25
(
1979
).
53.
E. I.
Galindo-Nava
,
J.
Sietsma
, and
P. E. J.
Rivera-Díaz-del-Castillo
,
Acta Mater.
60
,
2615
(
2012
).
54.
L. H.
Wang
,
Z.
Zhang
,
E.
Ma
, and
X. D.
Han
,
Appl. Phys. Lett.
98
, 051905 (
2011
).
55.
A. H.
King
and
S.
Shekhar
,
J. Mater. Sci.
41
,
7675
(
2006
).
56.
N.
Saeidi
,
F.
Ashrafizadeh
,
B.
Niroumand
, and
F.
Barlat
,
Mater. Des.
87
,
130
(
2015
).
57.
Z. G.
Liu
,
P. J.
Li
,
L. T.
Xiong
,
T. Y.
Liu
, and
L. J.
He
,
Mater. Sci. Eng. A
680
,
259
(
2017
).
58.
L.
Tan
,
X. Y.
Zhang
,
Q.
Sun
,
J. P.
Yu
,
G. J.
Huang
, and
Q.
Liu
,
Mater. Sci. Eng. A
699
,
247
(
2017
).
59.
Z. F.
Yan
,
D. H.
Wang
,
X. L.
He
,
W. X.
Wang
,
H. X.
Zhang
,
P.
Dong
,
C. H.
Li
,
Y. L.
Li
,
J.
Zhou
,
Z.
Liu
, and
L. Y.
Sun
,
Mater. Sci. Eng. A
723
,
212
(
2018
).
60.
T.
Walde
and
H.
Riedel
,
Mater. Sci. Eng. A
443
,
277
(
2007
).
61.
A.
Styczynski
,
C.
Hartig
,
J.
Bohlen
, and
D.
Letzig
,
Scr. Mater.
50
,
943
(
2004
).
62.
Q. Q.
Tian
,
H. Y.
Luo
,
R.
Yi
,
X. F.
Fan
,
Y.
Ma
,
D. Q.
Shi
, and
J. J.
Gao
,
Mater. Sci. Eng. A
771
,
138645
(
2020
).
63.
S.
Descartes
,
C.
Desrayaud
, and
E. F.
Rauch
,
Mater. Sci. Eng. A
528
,
3666
(
2011
).
64.
D.
Tumbajoy-Spinel
,
X.
Maeder
,
G.
Guillonneau
,
S.
Sao-Joao
,
S.
Descartes
,
J.-M.
Bergheau
,
C.
Langlade
,
J.
Michler
, and
G.
Kermouche
,
Mater. Des.
147
,
56
(
2018
).
65.
D.
Tumbajoy-Spinel
,
S.
Descartes
,
J.-M.
Bergheau
,
V.
Lacaille
,
G.
Guillonneau
,
J.
Michler
, and
G.
Kermouche
,
Mater. Sci. Eng. A
667
,
189
(
2016
).
66.
M. J.
Tang
and
J.
Marian
,
Acta Mater.
70
,
123
(
2014
).
67.
P.
Cizek
,
A.
Sankaran
,
E. F.
Rauch
, and
M. R.
Barnett
,
Metall. Mater. Trans. A
47
,
6655
(
2016
).
68.
X. Y.
Li
,
Z.
Zhang
, and
J. W.
Wang
,
Prog. Mater. Sci.
139
,
101160
(
2023
).
69.
X. Y.
Li
,
Q.
Zeng
, and
J. W.
Wang
,
Scr. Mater.
220
,
114930
(
2022
).
70.
P. A.
Juan
,
C.
Pradalier
,
S.
Berbenni
,
R. J.
McCabe
,
C. N.
Tomé
, and
L.
Capolungo
,
Acta Mater.
95
,
399
(
2015
).
71.
C.
Liu
,
F.
Roters
, and
D.
Raabe
,
Acta Mater.
242
,
118444
(
2023
).
72.
Y. J.
Song
,
T. X.
Li
,
X. Q.
Fu
,
Z. J.
Zhang
,
G.
Sheng
,
Y. H.
Zhu
,
Y. P.
Lu
, and
Q.
Yu
,
J. Alloys Compd.
947
,
169522
(
2023
).
73.
Q.
Yu
,
Y. Y.
Jiang
, and
J.
Wang
,
Scr. Mater.
96
,
41
(
2015
).
74.
M.
Koyama
,
E.
Akiyama
,
K.
Tsuzaki
, and
D.
Raabe
,
Acta Mater.
61
,
4607
(
2013
).
75.
C. A.
Stein
,
A.
Cerrone
,
T.
Ozturk
,
S.
Lee
,
P.
Kenesei
,
H.
Tucker
,
R.
Pokharel
,
J.
Lind
,
C.
Hefferan
,
R. M.
Suter
,
A. R.
Ingraffea
, and
A. D.
Rollett
,
Curr. Opin. Solid State Mater. Sci.
18
,
244
(
2014
).