Fabrication of Textured Porous Ti₃SiC₂ by Slip Casting under High Magnetic Field and Microstructural Evolution through High Temperature Deformation

Abstract

To clarify the effect of constraint conditions on the kink formation, fabrication process of the texture and porosity controlled Ti₃SiC₂ polycrystals was investigated and microstructural evolution during high temperature deformation was examined in it under high temperature uniaxial compression tests at 1200°C. Dense textured Ti₃SiC₂ sintered body was fabricated by slip casting in the high magnetic field of 12 T and following pressureless sintering at 1400°C for 1 h. The porosity of the textured Ti₃SiC₂ was controlled by dispersing polymethyl methacrylate (PMMA) particles into the textured Ti₃SiC₂ as a spacer media. The highly textured Ti₃SiC₂ polycrystals with porosity of 8.4 vol% and 16.7 vol%, respectively, were successfully fabricated by the slip casting in the high magnetic field. After the high temperature uniaxial compression perpendicular to the c-axis of the textured structure, both the porous and dense Ti₃SiC₂ showed kink formation, which is a common deformation mode for anisotropic layered materials. However, the average rotation angles of the kink boundaries were higher in the porous specimen than in the dense specimen. Since the crystal rotation is necessary for the kink formation, kink bands would be preferably developed in the porous area due to its weaker constraint than in the dense area. It can be concluded from the microstructural analysis that the constrain factor caused by the neighbor grains affects the crystalline rotation, resulting in the kink boundary formation with different rotation angles.

 

This Paper was Originally Published in Japanese in J. Japan Inst. Met. Mater. 85 (2021) 256–263.

Fig. 5 (a) IPF map of the sintered body and (b) its color-coded map. (c) PF of (0001) plane of (a), and (d) PF of (0001) plane of non-textured sintered body for comparison.

1. Introduction

Ti₃SiC₂ is one of the MAX phases—ternary carbides or nitrides expressed by the general formula M_n+1AX_n (M: transition metal, A: group A element, X: C and/or N, n = 1, 2, 3).¹⁾ They have attracted widespread attention because of their combination of metallic (good thermal and electrical conductivity^1–4) and excellent machinablity⁴⁾) and ceramic properties (low density¹⁾ and good thermal shock resistance²⁾ and oxidation resistance¹⁾). In particular, much work has been conducted on Ti₃SiC₂ following Barsoum and El-Raghy’s report of the synthesis of pure bulk Ti₃SiC₂ in 1996.³⁾

The MAX phases exhibit strong anisotropy because of their hexagonal layered crystal structure, where M_n+1X_n layers and A layers stack along the c-axis direction.¹⁾ Kink deformation is one of the deformation mechanisms in materials with strong anisotropy, such as MAX phases. Kink deformation is a folding deformation of a layered structure and occurs in various materials ranging from hexagonal close-packed (HCP) pure metals like Zn⁵⁾ to Mg-based HCP alloys⁶⁾ and micas.⁷⁾ However, kink deformation is not well understood compared with slip deformation by dislocation or twin deformation. In particular, the effect of external conditions, such as the strain ratio or constraint factor, on kink deformation has hardly been investigated.

One of the few studies about the effects of external constraints on kink deformation is Lei and Nakatani’s molecular dynamics simulation of layered ceramics under compression.⁸⁾ They evaluated the deformation behavior under uniaxial compression perpendicular to a layered structure in a ceramic. Specifically, they set the elastic constraint parallel to the layered structure (i.e., perpendicular to the loading axis) and varied the weakness of the constraint. They found that the constraint conditions changed both the deformation behavior and the final shape of the kinks. Experimental works involving Ti₃SiC₂ have also suggested that kink deformation is dependent on the constraint factor. Barsom and El-Raghy subjected a dense Ti₃SiC₂ sintered body to uniaxial compression and showed that kink deformation occurred first at the location with weaker constraint.⁹⁾ Sun et al. conducted cyclic compression and nanoindentation tests on a porous Ti₃SiC₂ sintered body and found that energy dissipation was larger in a porous specimen than in a dense one. They attributed the greater energy dissipation to the weak constraint in a porous sample promoting kink formation.¹⁰⁾

For a deeper understanding of the effect of the constraint factor on kink formation, imposing different constraint conditions and evaluating the deformation behavior is a valid approach. Grains near pores are weakly constrained by neighboring grains; however, grains in a dense area with fewer pores will be constrained strongly within the surrounding grains. Thus, if the internal pore density and pore structure can be controlled, then a different constraint situation can be achieved via the preparation of a porous sample. Therefore, porous Ti₃SiC₂ polycrystals with a controlled microstructure should be useful for clarifying the effect of constraint on kink deformation.

In addition, limiting deformation modes other than kink deformation will simplify the overall deformation and facilitate a discussion of the effect of constraint on kink deformation. For example, a single-crystal micropillar compression test on Ti₃SiC₂¹¹⁾ revealed that kink deformation occurred only when the compression axis was almost parallel to the basal plane; basal slip or fracture mainly occurred when it was not parallel to the basal plane. Hence, a simpler situation is possible if we use a polycrystalline sample in which the grain orientations are aligned (textured). Therefore, an ideal investigation of the effect of constraint factors on kink formation in polycrystals requires the fabrication of textured porous Ti₃SiC₂ with controlled porosity and texture.

In materials such as metals with sufficient ductility, texture control by recrystallization is possible through cold working and subsequent heat treatment. However, in brittle materials such as ceramics, such a process is not straightforward because, as with metals, achieving plasticity is difficult. Instead, a texturing process using a strong magnetic field has been developed for ceramic and metal powders.^12,13) The process texturizes grains by applying a strong magnetic field to a slurry in which powder particles with high crystallographic anisotropy are dispersed. A strong magnetic field applied during slip casting rotates the particles via the torque induced as a result of the magnetic anisotropy of the particles. As a result, a textured microstructure is formed. We herein refer to this method as “slip casting under a high magnetic field”. For Ti₃SiC₂ with strong crystallographic anisotropy, the fabrication of a textured sintered body has been reported.^14–16)

One of the common methods used to fabricate pore structures and porosity-controlled porous materials is to incorporate a pore-forming spacer powder into a green body in advance and then remove it before/after sintering. For slip casting under a magnetic field, polymethyl methacrylate (PMMA) as a spacer powder enabled the fabrication of textured β-alumina from porous α-alumina.¹⁷⁾

The objective of the present work is to clarify the effect of a constraint factor on kink formation in Ti₃SiC₂. First, Ti₃SiC₂ with a controlled porosity and pore structure is fabricated using PMMA as a spacer powder through slip casting under a high magnetic field. Second, the effect of the constraint condition on kink formation and development is investigated through high-temperature uniaxial compression of the obtained textured porous Ti₃SiC₂ sintered body using a spark plasma sintering (SPS) technique.

2. Experimental Procedure

2.1 Slurry preparation

As shown in Figs. 1(a) and 1(b), plate-like Ti₃SiC₂ powder with an average particle diameter of 5.2 µm (Maxthal (312), KANTHAL) and spherical PMMA powder with an average particle diameter of 10 µm (MX-1000, Soken Chemical & Engineering Co.) were used in this study. PMMA powder (10 vol%) was added as a spacer to the Ti₃SiC₂ powder. To obtain a suspension, ethanol was used as a solvent and polyethyleneimine (PEI) with an average molecular weight of 10000 (FUJIFILM Wako Pure Chemical Corp.) was used as a polymer dispersant. The amounts of the solvent, dispersant, and the raw powders were adjusted such that the powder content was 30 vol%. The PEI content was 1.5 mass% relative to the Ti₃SiC₂ and 1.0 mass% relative to the PMMA.

Fig. 1

FE-SEM images of the starting powders of (a) Ti₃SiC₂ and (b) PMMA.

To achieve a uniform dispersion of PMMA and Ti₃SiC₂, two slurries containing the individual powders were prepared and subsequently mixed. Thus, the weighed PMMA powder was first added to the ethanol and PEI and the resultant mixture was stirred using a planetary centrifugal mixer (ARE-310, THINKY Corp.). A slurry was similarly prepared by adding Ti₃SiC₂ powder to ethanol with added PEI and stirring the resultant mixture. The slurry of Ti₃SiC₂ was then added to and thoroughly mixed with the slurry of PMMA. In the last step, a uniformly distributed slurry of PMMA and Ti₃SiC₂ was obtained after a defoaming treatment was applied for 10 min using a stirrer, where the mixture was stirred under vacuum.

2.2 Fabrication of green body

A green body for sintering was obtained by slip casting of the adjusted slurry under a rotating magnetic field.¹⁴⁾ First, a porous alumina mold was covered with a membrane filter with a pore diameter of 0.2 µm; the slurry was then poured into an acrylic tube (diameter: 10 mm) placed onto the mold. For texturing of the grains, the mold was placed into a superconducting magnet (JMTD-12T1-NC5, JASTEC Co.) during slip casting, as shown in Fig. 2(a). The mold was rotated horizontally at 20 rpm during fabrication of a green body (Fig. 2(b)). A magnetic field B of 12 T was applied vertical to the rotating axis and the casting direction. Because the easily magnetizable axis of Ti₃SiC₂ is the a-axis, the magnetic field aligned the c-axis parallel to the casting direction. The raw powder of Ti₃SiC₂ exhibited a plate-like shape elongated vertical to the c-axis; thus, the Ti₃SiC₂ particles stacked along the casting direction (Fig. 2(c)).

Fig. 2

Schematic explanation of the slip casting under the magnetic field B. (a) Mold rotates under the high magnetic field and (b) green body is formed from the slurry containing Ti₃SiC₂ and PMMA. (c) Orientational relationship between green body, grain and crystal lattice.

To clarify the effect of the magnetic field on the texture and pores, a green body was fabricated from the same slurry via slip casting but without an applied magnetic field or rotation. Also, for comparison of the deformed microstructures after compression, a green body was fabricated by slip casting under a rotating magnetic field using a slurry without the PMMA dispersion but otherwise prepared in the same manner.

The green bodies were sintered using a spark plasma sintering (SPS) machine (SPS-510L, Fuji Electronic Industrial Co.). After the slip casting process, the green bodies were pressed by cold isostatic pressing (CIP) at 392 MPa for 10 min, followed by pressureless sintering. In general, the SPS technique applies a direct pulsed current to an electrically conductive die with a uniaxial pressing force; the mechanical pressure and the Joule heat at the die promote sintering.¹⁸⁾ In the present work, however, sintering was conducted without mechanical pressure to maintain the porosity of the sample. The pressureless sintering was carried out as follows: First, the sintering die was placed between punches with diameters larger than the inner diameter of the die, which resulted in the mechanical pressure being applied not to the green body itself but only to the die to maintain conductivity. Second, the sample was heated to 800°C at a heating rate of 500°C/min and maintained at 800°C for 1 h to burn out the PMMA. Third, it was heated again to 1400°C at a heating rate of 50°C/min and maintained at 1400°C for 1 h to complete the pressureless sintering.

2.3 Evaluation of sintered bodies

Figure 2(c) shows the definition of the “top” and “side” surfaces, where the former is the surface perpendicular to the casting direction and the latter is the surface parallel to the casting direction.

Identification of the phase and evaluation of the texture degree were conducted for the top and side surfaces using X-ray diffraction (XRD, Smartlab, Rigaku Corp.) with Cu(K_α1) radiation.

In addition, backscattered electron (BSE) images of side surfaces were collected by field-emission scanning electron microscopy (FE-SEM, JSM-7001FA, JEOL Ltd.) after they were mirror-polished with emery paper, diamond film, and diamond slurry. From the BSE images, the average pore diameter and the average porosity were obtained by the following procedure. First, the BSE images were binarized using image processing software (Image J, NIH); the areas of each pore and the area ratios per unit area were calculated on the basis of the number of black pixels derived from pores. Second, the equivalent circle diameters were calculated from the obtained number of pixels and pores with a diameter larger than 5 µm were regarded as pores formed by PMMA. Third, the average pore diameter was obtained by calculating the average value of more than 100 pore-defined diameters. The Archimedes method was also used to measure the porosity for comparison to the area ratio obtained from the BSE images.

The damaged layer formed by the machine polishing was removed using colloidal silica (particle diameter: 0.04 µm). Crystallographic analysis was then carried out using electron backscattered diffraction (EBSD) patterns obtained using a field-emission scanning electron microscope equipped with an orientation imaging microscopy data collection (OIM-DC, TSL Solutions K.K.).

2.4 Uniaxial compression and subsequent microstructure evaluation

To investigate the effect of constraint conditions on kink deformation in Ti₃SiC₂, we carried out uniaxial compression tests in the direction likely to cause kink deformation. First, the Ti₃SiC₂ sintered bodies were cut to 4 × 4 × 6 mm³ such that the compression axis was perpendicular to the casting direction (i.e., perpendicular to the c-axis of the texture). All the side surfaces were mirror-polished using emery paper, diamond film, and diamond slurry.

The SPS machine was used for the uniaxial compression test. The die was configured such that, inside the sintering die, the sample was placed between SiC plates and graphite punches. The sample was heated to the test temperature of 1200°C at a heating rate of 100°C/min. An initial force of 225 MPa was then applied to the sample, which was deformed to a compression strain of 6.8%. For comparison, the sample prepared from dense textured Ti₃SiC₂ was also compressed to a strain of 7.4%, similar to that for the porous Ti₃SiC₂. The sample was fabricated from the slurry without PMMA by slip casting under a high magnetic field, followed by CIP and pressureless sintering.

After deformation, the samples were cut parallel to the compression axis for evaluation of their deformed microstructure. They were then mirror polished and finished using colloidal silica, as described in section 2.3. The microstructure of the samples was then analyzed using BSE images, and their crystallographic properties were analyzed by the EBSD method. These analyses were carried out at the center of the compression specimen, where uniform deformation was expected to occur.

3. Results and Discussion

3.1 Evaluation of the sintered bodies

Figures 3(a) and 3(b) show the microstructure at the center and bottom area in the side surface of the sintered body. The sintered body consists of plate-like grains with a long side of approximately 10–20 µm and a narrow side of approximately 5–10 µm. Also, the plate-like crystal grains tend to be oriented parallel to the sheet.

Fig. 3

BSE images of the textured microstructure taken from (a) center and (b) bottom areas of the sintered body. White arrows show the pores formed by the PMMA dispersion.

In addition, the microstructures in both Figs. 3(a) and 3(b) have numerous pores larger than the grains, as indicated by the arrows, in addition to submicrometer smaller cavities. The average diameter of the large pores was 9.0 µm. The shapes and sizes of the large pores suggest that they were formed by PMMA particles with an average diameter of 10 µm. Also, the center part of the sintered body in Fig. 3(a) contains less sphere-like pores than the bottom area in Fig. 3(b). The porosities calculated from the binarized images in Figs. 3(a) and 3(b) are 7.4 and 17.9 vol%, respectively; thus, the porosity of the bottom area is greater than that of the center area. Also, the porosity of the sintered body measured by the Archimedes method was 16.7 vol%, which is closer to the porosity of the bottom area. One reason for this similarity is that the porosity measured by the Archimedes method includes small cavities in addition to large pores. Moreover, pore localization occurs not only in the bottom area but also in the side area of the sintered body. These observations suggest that pore localization occurred throughout the sintered body because of the difference in the specific weight between the Ti₃SiC₂ and PMMA powders: the rotation of the mold during the slip casting localized the PMMA particles at the bottom and side areas.

Figure 4 shows XRD patterns for the Ti₃SiC₂ powder and the sintered bodies. The reflection peaks related to the (000l) planes of Ti₃SiC₂ are denoted by the bold-italic Miller indices. The pattern in Fig. 4(a) shows that the powder before the sintering process included impurity phases such as TiC or TiSi₂ in addition to Ti₃SiC₂; these impurity phases remained after the sintering process, as shown in Figs. 4(b) and 4(c).

Fig. 4

X-ray diffraction patterns of (a) the Ti₃SiC₂ powder and the sintered body taken from (b) side surface and (c) top surface. Scans were made with Cu (K_α1) radiation.

The diffraction patterns for the side and top surfaces in Figs. 4(b) and 4(c) show different characteristics. The diffraction peaks from the side surface in Fig. 4(b) arise mainly from the $(10\bar{1}1)$, $(10\bar{1}4)$, $(10\bar{1}5)$, and $(10\bar{1}0)$ planes. The angle of each of these planes from the (0001) plane is 81.4°, 58.9°, 53.0°, and 90°, respectively, indicating that the peaks arise from planes far from (000l) planes. However, the peaks in the pattern for the top surfaces in Fig. 4(c) arise mainly from the (000l)-related planes. These two results mean that the normal vectors of the (000l) plane of most grains in the sintered body are perpendicular to the direction of the applied magnetic field.

For a more detailed discussion of the textured microstructure and its texture degree, the results of the crystallographic analysis conducted by the EBSD method are shown in Fig. 5. In Fig. 5, to specify each orientation relationship, the diameter direction of the cylinder sintered body is expressed as the x-axis, the casting direction is expressed as the y-axis, and the normal direction of the side surface is expressed as the z-axis. Notably, the x- and the z-axes distinguished here for convenience are equivalent to each other for the nature of the sintered body. Figure 5(a) shows an inverse pole figure (IPF) map of the z-axis direction colored by the color key on the standard stereographic triangle in Fig. 5(b). The black regions indicated by the arrows correspond to pores. The distribution of crystallographic orientations of grains near pores is the same as that of crystallographic orientations of grains in dense areas. This result suggests that the pores do not affect texturing. In addition, a few grains are displayed in red (the (0001) planes), which means the (0001) planes hardly appear on the observation plane, consistent with the XRD results in Fig. 4.

Fig. 5

(a) IPF map of the sintered body and (b) its color-coded map. (c) PF of (0001) plane of (a), and (d) PF of (0001) plane of non-textured sintered body for comparison.

The (0001) pole figure in Fig. 5(c) shows how the (0001) planes are oriented relative to the coordinate system of the sample. Figure 5(c) is an intensity distribution expressing the direction of (0001) planes for each grain in Fig. 5(a) relative to the coordinate system of the sample. In Fig. 5(c), the y-axis direction of the circle is represented in red, which means that the normal vectors of (0001) planes for each grain are concentrated in the y-axis direction (i.e., the casting direction). For comparison of the effect of the applied magnetic field on the degree of concentration (i.e., the texture degree), Fig. 5(d) shows a (0001) pole figure of the sintered sample fabricated in the absence of a magnetic field and without rotation.

In Fig. 5(d), a slight concentration of (0001) planes is observed for the sintered body without a magnetic field because of the slight texturing of plate-like grains during slip casting. However, no obvious concentration of (0001) planes is observed in the specific direction of the coordinate system of the sample, which means the crystallographic texture is almost random. The maximum intensities of the degrees of concentration (texture degrees) are 3.468 and 13.608 for the samples prepared in the absence and presence of a magnetic field, respectively. A greater maximum intensity indicates a higher concentration in a specific direction (higher texture degree). Therefore, the sample prepared under a magnetic field has a more highly textured microstructure than the sample prepared in the absence of a magnetic field. Thus, a highly textured microstructure where each c-axis of grains is parallel to the casting direction is formed when a magnetic field is applied; the textured microstructures near pores are the same as in dense areas. These XRD and EBSD results show that the rotation of Ti₃SiC₂ grains by the magnetic torque and the c-axis texture occur even when PMMA is added.

In conclusion, the BSE, XRD, and EBSD results indicate that a textured porous Ti₃SiC₂ sintered body with a controlled texture and porosity was successfully prepared when PMMA was added and a magnetic field was applied.

3.2 Microstructural evaluation after uniaxial compression

A uniaxial compression test at 1200°C was conducted on the aforementioned textured porous Ti₃SiC₂ sintered body to investigate the effect of pores (i.e., a constraint condition) on kink formation. For comparison, the same uniaxial compression test was conducted on a textured dense Ti₃SiC₂ sintered body with a porosity of 8.4 vol%. The tests were conducted under uniaxial compression stress. The compression specimens were cut from cylindrical sintered bodies in the direction corresponding to easy kink formation (i.e., where the compression axis is perpendicular to the c-axis).

The deformed sample exhibited no obvious crack formation, meaning that almost uniform plastic deformation occurred. Figures 6(b) and 6(c) show cross-sectional BSE images of samples after uniaxial compression. The loading direction is the vertical direction in these figures. Figure 6(b) shows the representative microstructure of the textured porous sample, and Fig. 6(c) shows that of the textured dense sample. The areas marked by the black arrows in Fig. 6(b) correspond to the pores from PMMA. The small cavities in Fig. 6(c) originate from incomplete densification through pressureless sintering or from deformation (at the grain boundaries). In the porous and dense samples shown in Figs. 6(b) and 6(c), the grains extended parallel to the compression axis have folded structures (kink bands) and kink boundaries (white arrows). These results indicate that kink deformation occurred in both the textured porous samples and the dense samples.

Fig. 6

(a) Schematic explanation of the sintered body and the compression specimen fabricated from the sintered body. (b) BSE images of the porous specimen and (c) the dense specimen after the uniaxial compression. Black arrows in (b) show the pores formed by the PMMA dispersion. White arrows in (b) and (c) show the kink boundaries.

The textured porous samples and the dense samples were analyzed by EBSD to investigate the difference in their kink structures in terms of crystallographic orientation. Figure 7 shows the results for the textured porous sample after compression. Figure 7(a) is an IPF map of the normal direction of the observed surface in the textured porous sample after compression; it is colored according to the key in Fig. 7(b). The observing surface is the cross-section of the sample, and the loading direction is vertical to this figure. The arrows show the uniform pores formed by PMMA in the sample. We here discuss the orientational relationships in detail using the grains in Fig. 7(c), which is an enlargement of the white rectangular area in Fig. 7(a). To visualize the crystallographic orientation, Fig. 7(c) shows the grains with hexagonal lattices at each point. The crystallographic orientation continuously changes with the rotation axis perpendicular to the [0001] direction. Such a change in orientation occurs because kink deformation is the deformation of local crystallographic rotation with a certain crystallographic orientation axis.¹⁹⁾ Thus, the investigation of the rotation angle and rotation axis of each kinked grain enables a geometric analysis of kink deformation.

Fig. 7

(a) IPF map of the porous specimen after compression and (b) its color-coded map. (c) Enlarged IPF map of the kinked grain surrounded by the white rectangular in (a) and the change of the misorientation angle of the grain.

The relationships between the rotation angle and axis were examined for the kinked structures after compression (Fig. 7). We here explain the measurement procedure using the kinked grain in Fig. 7(c) as an example. First, minor changes in orientation (rotation angle θ) and rotation axis are measured in the kinked grain in the right part of Fig. 7(c), at each pixel along the broken line from the upper point to the lower point. The result in the left part of Fig. 7(c) is obtained, where the distance along the broken line is represented on the vertical axis and the rotation angle θ is represented on the horizontal axis. From this figure, sharp peaks in the rotation angle exist locally at very narrow areas inside the kinked grain; the crystallographic rotation at the peak position forms the overall folded structure. These sharp peaks of the rotation axis occur similarly at the kink boundaries in Figs. 6(b) and 6(c). Thus, an evaluation of the orientational change inside kinked grains, such as those in Fig. 7(c), reveals the rotation angle and the rotation axis at each kink boundary. Next, differences in the rotation angle and the rotation axis for kink boundaries are investigated in the textured porous and dense specimen after deformation, and the effect of the constraint conditions on kink deformation is discussed.

First, we show the evaluation results for the rotation angle θ. The histogram in Fig. 8 shows the distribution of the rotation angle at 104 kink boundaries in the textured porous sample and 97 kink boundaries in the textured dense sample. The rotation angle θ ranges widely from a few degrees to 70° at maximum. From this distribution, the average rotation angle θ in the textured porous and dense samples was calculated. The calculated average rotation angle θ in the former is 16.4°, and that in the latter is 12.1°; that is, the rotation angle in the textured porous sample is larger than that in its dense counterpart. This result suggests that the rotation angle per kink boundary is greater in the more weakly constrained porous sample than in the more strongly constrained dense sample.

Fig. 8

Histogram of the misorientation angle at each kink boundaries of kinked grains observed in the porous and in the dense samples.

We next discuss the rotation axis. The calculation of the rotation axes at each kink boundary reveals that most of the rotation axes are perpendicular to the c-axis; these axes exhibit various directions at each boundary. This result is consistent with the conventional model of kink deformation⁵⁾ and with the observed rotation axes in the long-period stacking ordered (LPSO) phases of Mg–TM–RE (TM: transition metal, RE: rare earth)⁶⁾ alloys, in which dislocation motion on basal planes is considered to induce crystallographic rotation, resulting in the formation of kink bands. In addition, this tendency was similarly observed in the textured porous and dense samples. Therefore, a rotation axis at a kink boundary is unaffected by the weak constraint.

These considerations suggest that no obvious difference exists between the crystallographic rotation axis at kink boundaries in the textured porous sample with weak constraint and that in the textured dense sample with strong constraint even though their rotation angles differ somewhat. Therefore, we reasonably assume that crystal rotation by dislocation motion forms kink bands in both the porous and dense samples. The arrangement of dislocations in a row results in a sharp peak (localization) in the rotation angle θ. The lattice rotation in the porous sample is larger than in the dense sample even though the deformation degree in the former sample (6.8%) is smaller than that in the latter sample (7.4%). Thus, the porous sample must have a smaller constraint on lattice rotation, promoting kink formation. Previous investigations^8–10) have qualitatively shown that the constraint factor affects the morphology, priority area, and the energy dissipation of kink deformation. The present study quantitatively suggests that the constraint affects the lattice rotation, resulting in a difference in kink formation behavior.

4. Conclusion

To investigate the effect of constraint conditions on kink deformation in Ti₃SiC₂, we fabricated a textured porous Ti₃SiC₂ sintered body via slip casting under a strong magnetic field and carried out high-temperature uniaxial compression using SPS. From the results, we drew the following conclusions.

1. (1)    PMMA addition controls the porous microstructure of the sintered body, and the magnetic field controls the texture of the Ti₃SiC₂ grains. Thus, slip casting of a slurry with PMMA in a strong magnetic field enables the fabrication of a textured porous Ti₃SiC₂ sintered body.
2. (2)    The results of high-temperature uniaxial compression of the textured porous and textured dense Ti₃SiC₂ bodies sintered via SPS clarified that the average rotation angle at each kink boundary is greater in kinked grains of the porous sample than in kinked grains of the dense sample.
3. (3)    The larger average rotation angle suggests that the constraint of lattice rotation to form kink bands is weaker in the porous sample than in the dense sample, which enables the development of a kinked structure with a higher rotation angle than in the dense sample.

Acknowledgments

This work was supported by JSPS KAKENHI (Grant Numbers JP19H05115 and JP18H05482) and by the Kazuchika Okura Memorial Foundation. Parts of this work were also conducted at Hokkaido University, supported by the “Nanotechnology Platform Program” of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, and the IMS Joint Research Hub Program (ID2021-039). The authors are grateful to Mr. Yuji Shirakami, a graduate student in the Graduate School of Engineering, Hokkaido University, for his assistance with the experiments and contribution to the discussion.

REFERENCES

• 1)   M.W.  Barsoum: Prog. Solid State Chem. 28 (2000) 201–281. doi:10.1016/S0079-6786(00)00006-6
• 2)   M.W.  Barsoum,  D.  Brodkin and  T.  El-Raghy: Scr. Mater. 36 (1997) 535–541. doi:10.1016/S1359-6462(96)00418-6
• 3)   M.W.  Barsoum and  T.  El-Raghy: J. Am. Ceram. Soc. 79 (1996) 1953–1956. doi:10.1111/j.1151-2916.1996.tb08018.x
• 4)   Z.M.  Sun,  H.  Hashimoto,  Z.F.  Zhang,  S.L.  Yang and  S.  Tada: Mater. Trans. 47 (2006) 170–174. doi:10.2320/matertrans.47.170
• 5)   J.B.  Hess and  C.S.  Barrett: Metals Trans. 185 (1949) 599–606.
• 6)   T.  Matsumoto,  M.  Yamasaki,  K.  Hagihara and  Y.  Kawamura: Acta Mater. 151 (2018) 112–124. doi:10.1016/j.actamat.2018.03.034
• 7)   S.  Misra and  J.P.  Burg: Tectonophysics 548–549 (2012) 22–33. doi:10.1016/j.tecto.2012.04.014
• 8)   X.W.  Lei and  A.  Nakatani: J. Eur. Ceram. Soc. 36 (2016) 2311–2317. doi:10.1016/j.jeurceramsoc.2015.12.054
• 9)   M.W.  Barsoum and  T.  El-Raghy: Metall. Mater. Trans. A 30 (1999) 363–369. doi:10.1007/s11661-999-0325-0
• 10)   Z.M.  Sun,  A.  Murugaiah,  T.  Zhen,  A.  Zhou and  M.W.  Barsoum: Acta Mater. 53 (2005) 4359–4366. doi:10.1016/j.actamat.2005.05.034
• 11)   M.  Higashi,  S.  Momono,  K.  Kishida,  N.L.  Okamoto and  H.  Inui: Acta Mater. 161 (2018) 161–170. doi:10.1016/j.actamat.2018.09.024
• 12)   T.S.  Suzuki,  Y.  Sakka and  K.  Kitazawa: J. Ceram. Soc. Jpn. 109 (2001) 886–890. doi:10.2109/jcersj.109.1274_886
• 13)   J.  Akiyama,  H.  Asano,  K.  Iwai and  S.  Asai: J. Japan Inst. Metals 71 (2007) 108–112. doi:10.2320/jinstmet.71.108
• 14)   C.  Hu,  Y.  Sakka,  S.  Grasso,  T.S.  Suzuki and  H.  Tanaka: J. Am. Ceram. Soc. 94 (2011) 742–748. doi:10.1111/j.1551-2916.2010.04186.x
• 15)   K.  Sato,  H.  Hirano,  T.S.  Suzuki,  T.  Ishigaki and  Y.  Sakka: J. Soc. Powder Technol. Japan 51 (2014) 163–168. doi:10.4164/sptj.51.163
• 16)   Y.  Uchida,  K.  Morita,  T.S.  Suzuki,  T.  Nishimura,  K.  Fujimoto and  Y.  Sakka: Mater. Trans. 59 (2018) 829–834. doi:10.2320/matertrans.Y-M2018812
• 17)   Y.  Sakka,  A.  Honda,  T.S.  Suzuki and  Y.  Moriyoshi: Solid State Ionics 172 (2004) 341–347. doi:10.1016/j.ssi.2004.02.049
• 18)   K.  Kakegawa: J. Soc. Inorganic Mater. Japan 13 (2006) 157–164. doi:10.11451/mukimate2000.13.157
• 19)   M.  Yamasaki,  K.  Hagihara,  S.  Inoue,  J.P.  Hadorn and  Y.  Kawamura: Acta Mater. 61 (2013) 2065–2076. doi:10.1016/j.actamat.2012.12.026