Excellent mechanical properties of taenite in meteoric iron

Meteoric iron is the metal that humans first obtained and used in the earliest stage of metal culture. Advances in metallographic analysis techniques have revealed that meteoric iron largely comprises kamacite, taenite, and cohenite, which correspond to ferrite, austenite, and cementite in artificial steel, respectively. Although the mechanical properties of meteoric irons were measured previously to understand their origin and history, the genuine mechanical properties of meteoric iron remain unknown because of its complex microstructure and the pre-existing cracks in cohenite. Using micro-tensile tests to analyse the single-crystalline constituents of the Canyon Diablo meteorite, herein, we show that the taenite matrix exhibits excellent balance between yield strength and ductility superior to that of the kamacite matrix. We found that taenite is rich in nitrogen despite containing a large amount of nickel, which decreases the nitrogen solubility, suggesting that solid-solution strengthening via nitrogen is highly effective for the Fe–Ni system. Our findings not only provide insights for developing advanced high-strength steel but also help understand the mysterious relationship between nitrogen and nickel contents in steel. Like ancient peoples believed that meteoric iron was a gift from the heavens, the findings herein imply that this thought continues even now.


Scientific Reports
| (2021) 11:4750 | https://doi.org/10.1038/s41598-021-83792-y www.nature.com/scientificreports/ meteoric irons 2,3 . According to the compositional mapping of alloying elements by EPMA shown in Fig. 1c, there was no segregation of alloying elements in the boxed area in Fig. 1a. Quantitative analysis of the alloying elements (Extended Data Table 1) in the area shown in Fig. 1c revealed that the α-phase could be regarded as ferritic iron containing approximately 6 at.% Ni, i.e. kamacite. A small fraction of taenite (determined by γ-iron in the EBSD analysis) was observed as shown in Fig. 1d,e. The compositional mapping of alloying elements in taenite revealed that the iron content decreased and nickel content increased upon approaching the taenite/ kamacite boundary (Fig. 1f). Beyond the area near the taenite/kamacite boundary, taenite was composed of 1.61C, 5.18 N, 0.03P, 72.60Fe, 0.23Co, and 20.35Ni in at.% (Extended Data Table 1). Additionally, cohenite (determined by cementite in the EBSD analysis) was frequently observed, as shown in Fig. 1g,h. Cohenite was composed of 20.66C, 0.53 N, 0.001P, 77.48Fe, 0.17Co, and 1.16Ni in at.%, corresponding to Fe 3 C cementite (Extended Data Table 1). Furthermore, numerous cracks and small areas of taenite were observed in cohenite ( Fig. 1g-i). A micro-tensile test of single-crystalline cohenite revealed that the cleavage fracture occurred on the (011) plane without plastic deformation at the maximum tensile strength of 2.4 GPa (Extended Data Fig. 1). We expect that these are why the previous measurements of the tensile properties of meteoric iron 8,11 exhibited low strength levels, because the brittle materials are sensitive to defects. Subsequently, based on the microstructural determination by EBSD and EPMA analyses, we performed micro-tensile tests for kamacite and taenite of the Canyon Diablo meteorite sample.
Stress-strain curves. Figure 2a,b show an example of preparing a micro-sized tensile specimen for taenite.
As illustrated in the inverse pole figure map of taenite (Fig. 2a), a micro-tensile specimen with a gauge section of 11 × 20 × 50 μm 3 was fabricated using a focused ion beam (Fig. 2b) so that the gauge part was completely single crystalline. We chose this length scale because no significant size effect will appear down to this speci-   [21][22][23] . Thus, we can evaluate the mechanical properties of the microstructural constituents, which are equivalent to those of the bulk specimen. A single-crystalline specimen of kamacite was prepared in the same way. Figure 2c shows the stress-strain curves obtained by micro-tensile testing of the single-crystalline taenite and kamacite specimens with their loading directions (LDs) nearly parallel to the [123] direction. For kamacite, the yield strength and elongation-to-failure were 350 MPa and 19%, respectively, which are in good agreement with the previous estimated yield strength of 335 MPa and elongation-to-failure of 19%, obtained by the tensile testing of a Gibeon meteorite 7 with a Widmanstätten structure and coarse kamacite widths. Meanwhile, the yield strength and elongation-to-failure of taenite were 935 MPa and 65%, respectively. Although there have been no reports on high-nitrogen alloying in nickel-rich austenitic steels because nickel decreases the nitrogen solubility 24 , introducing high levels of nitrogen into Fe-Cr-Mn alloys 25,26 is well known as a remarkable strengthening method via solid-solution strengthening. For example, the yield strength of the Fe-24Cr-10Mn-1.43 N alloy (mass%), which contains an amount of nitrogen comparable to that in the taenite in this study (1.35 N mass%), was measured as 830 MPa 26 . Therefore, the high yield strength of taenite was presumably attributable to a solid-solution-strengthening mechanism via the interaction between dislocations and interstitial nitrogen atoms 27 .
Micro-tensile behaviours. Figure 3 shows the deformation and fracture morphology of the kamacite micro-tensile specimen. Slip steps formed at inclination angles of 80° and 54° with respect to the LD after the onset of yielding (Fig. 3a,b), and they propagated while maintaining a constant stress level (Supplementary Video 1). Finally, chisel-edge-type fracture with significant necking occurred (Fig. 3c). Figure 3d shows a stereographic projection of the kamacite specimen based on the EBSD analysis, which indicates that the observed slip steps correspond to the primary slip systems (110) [11 1] and (211 ) [111], with Schmid factors of 0.47 and 0.49, respectively. This suggested that cross slips occurred on the favourable slip planes with the same slip direction and that the kamacite matrix deformed based on Schmid's law. Figure 4 shows the deformation and fracture morphology of the taenite micro-tensile specimen. In the taenite specimen, a linear slip step formed at an incli- www.nature.com/scientificreports/ nation angle of 62° with respect to the LD at the onset of yielding (Fig. 4a), and thereafter the deformation band extended throughout the gauge part of the specimen (Fig. 4b), like the Lüders deformation (Supplementary Video 2). Finally, the specimen fractured with significant necking (Fig. 4c). The linear slip step observed in the taenite specimen corresponds to the primary slip system of (11 1) [011], with a Schmid factor of 0.47 (Fig. 4a,d).
It should be noted that the observed slip steps in the kamacite and taenite specimens exhibited wavy and linear morphologies, respectively. Active slip systems in BCC crystals are in the <111> direction on the {110}, {112}, and {123} planes, whereas those in FCC crystals are in the <110> direction on the {111} planes. Put simply, kamacite (BCC) has more active slip systems than taenite (FCC). Therefore, the difference in slip behaviour was attributable to the difference in the number of slip systems in kamacite and taenite. The stress-strain curve of taenite (Fig. 2c) revealed a high yield strength without a loss in the good ductility. Figure 5 shows the relationships between the true stress, σ T , and strain hardening rate, dσ T /dε T , plotted against www.nature.com/scientificreports/ the true strain, ε T , for the taenite specimen. Generally, single-crystalline FCC metals exhibit a transition from the easy-glide stage to linear-hardening stage owing to the interaction of multiple slip systems. In the taenite specimen, indeed, single slip gliding proceeds throughout the gauge part of the specimen (Fig. 4a,b), followed by the strain hardening concurrent with the activation of the secondary slip system (Fig. 5). The onset of plastic instability can be determined by the Considѐre's condition: dσ T /dε T ≤ σ T . In the taenite specimen having high yield strength, the strain hardening stage is short because the stress level exceeds the strain hardening rate at an early stage of strain hardening (indicated by the arrow in Fig. 5). In austenitic stainless steels, nitrogen increases not only the strength but also the strain hardening rate 25 . This indicates that the occurrence of local necking is suppressed in the easy-glide stage, which enables uniform deformation up to a high-strain region. Micro-tensile tests of the Canyon Diablo meteorite specimens revealed that the nickel-rich austenite with an ultra-high nitrogen content exhibited an excellent relationship between yield strength and ductility. Generally, advanced high-strength austenitic steels, which combine a high strength and moderate ductility via transformation-induced plasticity and twin-induced plasticity, exhibit tensile strengths exceeding ca. 800 MPa. It is a major challenge to increase the yield strength of advanced high-strength austenitic steels because of a risk of hydrogen embrittlement under severe external conditions 28 . Therefore, we anticipate that our findings will aid

Methods
Material and microstructural characterisation. The material used in this study was a Canyon Diablo meteorite (type IAB, coarse octahedrite). Small-cut samples were polished with emery paper and a colloidal SiO 2 paste. Compositional mappings in the constituents of the meteorite were conducted using a JEOL (JCM-5700) scanning electron microscope (SEM) equipped with an electron probe micro-analyser (EPMA, Shimadzu 1720), and operated at a beam current of 0.1 μA and accelerating voltage of 15 kV. All the measurements were obtained using the Kα signal of C, N, P, Fe, Co, and Ni. Quantitative 12-point analyses were performed in each characteristic phase using a focused beam at a beam current of 0.05 μA and accelerating voltage of 15 kV. The ZAF correction method was applied for the quantitative analysis of chemical compositions using reference samples of Cr 3 C 2 , AlN, GaP, Fe, Co, and Ni for each element (C, N, P, Fe, Co, and Ni, respectively). The methods of Duncumb-Reed, Philibert, and Reed were used for correcting the effects of atomic number (Z), absorption (A), and fluorescence (F), respectively. The average values were calculated from the results of the 12-point analyses for each element in phases to estimate the chemical compositions. After the EPMA analysis, the crystal orientation was determined at a scanning step size of 0.4 μm using an SEM instrument equipped with an EBSD detector and orientation imaging microscopy software (TSL OIM v.7.1.0). A clean-up procedure was applied to all the EBSD images to adjust single points having misorientations greater than 5° in comparison with their neighbours. Additionally, points with a confidence index lower than 0.1 were excluded from the analysis based on Field's study 29 .
Micro-tensile tests. The samples were thinned to a thickness of approximately 20 μm using emery paper, and both the surfaces were then mirror-finished with a colloidal SiO 2 paste. Micro-tensile specimens with a gauge section of 11 × 20 × 50 μm 3 , 18 × 20 × 50 μm 3 , and 22 × 20 × 50 μm 3 were fabricated using a focused ion beam for taenite, kamacite, and cohenite, respectively. In the regime of this length scale, no difference was observed in the strength 22,23 due to the sample size effect which occurs in micropillar compression 21 . Single-crystalline specimens were prepared with their LDs approximately parallel to the [123] direction for kamacite and taenite, and to the [011] direction for cohenite. Tensile tests were performed at room temperature under laboratory atmospheric conditions with a displacement rate of 0.1 μm s -1 , corresponding to a strain rate of 2 × 10 -3 s -1 . The set-up has been described in more detail in a previous study 30 . The gauge section of the tensile specimen was monitored during tensile testing using an optical microscope to dynamically measure the strain as a function of time.

Data availability
All data generated or analysed during this study are included in the published article and Supplementary Information and are available from the corresponding author upon reasonable request.

Figure 5.
Relationship between true stress and strain hardening rate plotted against true strain for the taenite specimen. σ T and ε T were calculated using the following equations: σ T = σ (1 + ε) and ε T = ln (1 + ε). Since these equations can be applied in the case of isotropic shrinkage due to uniaxial deformation under the constantvolume deformation assumption, the calculations are invalid during local deformation regime and beyond the plastic instability. License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.