Acicular Ferrite Formation on Ti-Rare Earth Metal-Zr Complex Oxides

Abstract

Acicular ferrite (AF) formation potency of Ti-Rare earth metal (REM)-Zr (TRZ) complex oxide has been investigated in the simulated heat affected zone of low carbon steel. The TRZ complex oxide shows higher AF formation potency than Ti and Al oxides. A TRZ oxide particle is composed of REM-rich and Zr-rich phases. AF crystals nucleate on the interface between austenite and the Zr-rich oxide phase, having a crystal structure that is face-centered cubic with lattice parameter of 0.44 nm. The Zr-rich phase and AF have an orientation relationship described by (011)_AF//(011)_Oxide, [100]_AF//[011]_Oxide (the AF-TRZ orientation relationship), which allows good lattice coherency. It is suggested that the formation of this orientation relationship promotes AF nucleation on TRZ complex oxides. The AF also satisfies the Kurdjumov-Sachs (K-S) orientation relationship with the austenite matrix. It is considered that the coexistence of the AF-TRZ and K-S “three phase” orientation relationships is caused by variant selection of AF in addition to the formation of a rational orientation relationship between the Zr-rich oxide phase and the austenite matrix during the HAZ thermal cycle.

1. Introduction

Thick plates for building construction, container ship production, etc. are required to combine high strength and toughness. In particular, it is of great importance to improve toughness of the heat affected zone (HAZ), which is the region adjacent to liquid weld metal. A coarse structure is formed in the HAZ due to reverse transformation and coarsening of austenite grains at high temperature during the welding process, and causes a decrease of HAZ toughness. For many years, enhancement of HAZ toughness has been accomplished through refinement of austenite grains by the pinning effect of fine precipitates, i.e. TiN.^1,2,3) However, nowadays high heat input welding processes are demanded from the viewpoint of welding efficiency. High heat input welding causes coarsening of HAZ microstructure and decrease of HAZ toughness through dissolution of TiN particles. So new techniques are required to achieve high HAZ toughness in high heat input welding.

Oxide metallurgy is a technique which is expected to enhance the HAZ toughness of high heat input welds. Oxides, which are generally considered to be more stable than TiN at high temperature, act as nucleation sites for fine AF microstructures in the HAZ of low carbon steels. AF has a three-dimensional inter-locked morphology and contributes to refinement of the HAZ microstructure and enhances toughness.^4,5,6) Many studies have been done to identify oxides that promote AF formation and to establish the mechanism of AF nucleation.^{7,8,9,10,11,12,13,14)} It is well known that Ti oxides have a positive influence on AF nucleation.^{10,12,13,15,16)} Researchers have pointed out that Ti₂O₃ incorporates Mn and forms a Mn depleted zone around the oxide, which increases the driving force for decomposition of the surrounding austenite matrix during cooling.^12,17) Additionally, it has been reported that AF nucleation on TiO is due to the good lattice coherency between TiO and AF.^10,18) According to Yamada et al., a TiO layer formed on the surface of an inclusion particle promotes AF nucleation. The AF satisfies the Baker-Nutting (B-N) orientation relationship with the TiO layer.¹⁹⁾ Related to this behavior, Kasai et al. compared interfacial energies between the AF and various B1 compounds by first principles calculations and found that the interfacial energy for TiO is lower than for TiN or MgO.²⁰⁾ Moreover, it is reported that some complex oxides also promote AF nucleation. It is revealed that MnTi₂O₄ and MnAlO₄ are good nucleation sites for AF from the viewpoint of lattice coherency.^10,21,22) However, AF nucleation on complex oxides is still incompletely understood. The purpose of this paper is to investigate the influence of complex oxides on the formation of AF in simulated HAZ microstructures of low carbon steel.

2. Experimental Procedure

Al-added, Ti-added and Ti-Rare earth metal (REM)-Zr-added (TRZ-added) steel alloy ingots were prepared by vacuum induction melting. The chemical compositions of the three ingots are shown in Table 1. REM was added by Ce-rich Misch metal. The P and S concentrations of each ingot are under 0.003 and 0.001 mass% respectively. N contents were kept below 0.002 mass% so as to decrease the amount of TiN, and thus minimize the influence of any TiN effects on AF nucleation.^2,11,23)

Table 1. Chemical composition of experimental ingots (mass%).

	C	Si	Mn	Ni	Al	Ca	B	Ti	REM*	Zr	O	N	Fe
Al-added	0.046	<0.01	1.54	0.34	0.019	0.0007	0.0015	–	–	–	0.0017	0.0016	bal.
Ti-added	0.042	<0.01	1.51	0.34	<0.003	0.0007	0.0014	0.009	–	–	0.0023	0.0013	bal.
TRZ-added	0.040	<0.01	1.53	0.34	<0.003	0.0005	0.0017	0.008	0.0014	0.0006	0.0018	0.0014	bal.

*  Cerium + Lanthanum

Cylindrical specimens 12 mm in height and 8 mm in diameter for heat treatment were cut from the ingots. The specimens were heated at 20 K/s, held at 1723 K for 5 s, and subsequently cooled to room temperature at 50 K/s using a Thermecmastor-Z (Fuji Electronic Industrial Co., Ltd.) simulator. According to Thermo-Calc software, it is expected that TiN should dissolve completely above 1620 K.

After heat treatment, the oxide sizes and microstructures of the specimens were characterized using light optical microscopy (OM). The cross-sections at mid-length of the long axis of the columnar specimens were polished and etched with 3% nital solution (3% nitric acid and 97% methanol) before observation. The AF formation rate P_AF, defined by the following equation, was measured at 400 times magnification.   

$$P_{AF}=N_{AF}/N_S\times 100$$

where N_AF is the number density of oxides on which AF nucleation is observed by OM and N_S is the total number of oxides per unit area. Both N_AF and N_S were measured in regard to oxides having a circle equivalent diameter (CED) over 1.0 μm. Oxide particle sizes were measured using ImageJ software. The total area for measurement was 0.74 mm² per specimen.

The compositions of the oxide particles were analyzed semi-quantitatively by energy dispersive X-ray spectroscopy (EDS) using a scanning electron microscope (JEOL JSM-7001F) equipped with an EDS analyzer (JEOL JED-2300F). Orientations of AF crystals were analyzed by electron backscatter diffraction patterns (EBSD) using a JEOL JSM-7000F. Selected area electron diffraction pattern analysis by transmission electron microscopy (TEM) was made to identify the crystal structure of the oxides which act as nucleation sites for AF. TEM observation was performed using a Philips CM200 operated at 200 kV. Specimens for TEM analysis were fabricated using a focused ion beam (FEI Helios Nanolab 600i). The orientation relationship between the oxide and the AF was evaluated using Kikuchi pattern analysis.²⁴⁾

3. Results

3.1. Microstructure and Oxide

The OM images of the alloys are shown in Fig. 1 after holding 5 s at 1723 K and cooling at 50 K/s. In Al-added alloy, a coarse bainite structure which originates mostly from prior austenite grain boundaries is observed. For Ti-added and TRZ-added alloys, a mixed bainite and AF microstructure is observed. However, the fraction of AF microstructure is greater in the TRZ-added alloy compared to the Ti-added alloy. Grain boundary ferrite was not found in any of the three alloys. Diameters of prior austenite grains of the Al-added, Ti-added, and TRZ-added alloys were 272, 229, and 347 μm respectively. The N_S, N_AF, and P_AF values for each alloy are displayed in Table 2. The table indicates that the TRZ-added alloy exhibits the highest AF formation rate. The size distributions of oxide particles in the three alloys are depicted in Fig. 2. All alloys show a peak at CED values ranging between 1.5–2.0 μm. However, the fraction of oxide particles with CED values less than 2.0 μm in the TRZ-added alloy is larger than those in the Al-added and Ti-added alloys. This result indicates that the TRZ-added alloy contains finer oxide particles than the other alloys.

Fig. 1.

Optical micrograph of the (a) Al-added, (b) Ti-added, and (c) TRZ-added alloys heated at 1723 K for 5 s and cooled at 50 K/s. “B” and “AF” indicate bainite and acicular ferrite microstructure respectively.

Table 2. Total number of oxides per unit area (N_S), number density of oxides which act as AF nucleation sites (N_AF), and AF formation rate (P_AF) with 95% confidence interval²⁵⁾ of the alloys heated at 1723 K for 5 s and cooled at 50 K/s.

	NS (mm–2)	NAF (mm–2)	PAF (%)
Al-added	80	8	10±8
Ti-added	72	14	19±11
TRZ-added	93	42	45±11

Fig. 2.

Size (CED) distribution of oxide particles in the three alloys.

Figure 3 shows backscattered electron images of oxides in the three alloys. The image contrast for the oxides in the Al-added and Ti-added alloys seem to be uniform, which implies that these oxides are composed of a single phase. In contrast, the oxide in the TRZ-added alloy is divided into gray and white zones. Figure 4 shows the chemical compositions of the oxides at points P1 to P4 described in Fig. 3. The concentration of each element is normalized excluding Fe and O. In the Al-added alloy, the oxide is composed primarily of Al and Ca, along with Mg which is considered to be an impurity element. In the Ti-added alloy, the oxide is composed mainly of Ti. The oxide in the TRZ-added alloy contains rare earth metals (Ce, La), along with Zr, Ti, Ca, Al, Mn, and Mg. In this study, we call this oxide a “Ti-REM-Zr (TRZ) complex oxide”. The REM concentration in the gray zone is much lower than that of the white zone. Conversely, the gray zone show a higher Zr concentration compared to the white zone. Hereafter, we focus on the mechanism of AF nucleation on the TRZ complex oxides, which showed the highest AF formation rate.

Fig. 3.

Backscattered electron images of oxides in the (a) Al-added, (b) Ti-added, and (c) TRZ-added alloys heated at 1723 K for 5 s and cooled at 50 K/s.

Fig. 4.

Normalized concentrations of selected elements at points P1–P4 indicated in Fig. 3.

3.2. Acicular Ferrite Orientation Relationships and Crystal Structure of the Ti-REM-Zr Complex Oxide

An inverse pole figure map of the AF microstructure in the TRZ-added alloy is displayed in Fig. 5. The black region located at the center of the map is an oxide particle. Black and red lines represent high-angle boundaries (over 15 degrees) and low-angle boundaries (2–15 degrees), respectively. There are two AF crystals (indicated as “AF1” and “AF2” in Fig. 5) which appear to originate on the surface of the oxide particle and “radiate” outward from the oxide. Figure 6(a) shows the (001)_α pole figure of the area corresponding to Fig. 5. Figure 6(b) illustrates (001)_α poles of points P1 and P2 indicated in Fig. 5 and 24 variants of the K-S orientation relationship²⁶⁾ ((110)_α//(111)_γ, [111]_α//[011]_γ) calculated based on the inferred crystal orientation of the austenite. The analyzed pole figure from the AF microstructure exhibits the general character expected for the K-S orientation relationship. The crystal orientations of both points P1 and P2 approximately satisfy the apparent K-S orientation relationship, representing two different variants.

Fig. 5.

Inverse pole figure map of α-Fe in the TRZ-added alloy heated at 1723 K for 5 s and cooled at 50 K/s.

Fig. 6.

(a) (001)_α pole figure of the area corresponding to Fig. 5. (b) (001)_α poles of points P1 and P2 indicated in Fig. 5, prior austenite, and 24 variants of the K-S orientation relationship calculated based on inferred crystal orientation of the austenite.

Figure 7 shows a TEM bright field image of the same oxide particle, and crystals AF1 and AF2 depicted in Fig. 5 (prepared using FIB lift out techniques) and a selected area electron diffraction (SAED) pattern of point O1 within the oxide. The oxide is longitudinally divided by a dashed line in Fig. 7 into two regions. The crystal structure at point O1 is indexed as face-centered cubic with a lattice parameter of 0.44 nm. The structure at point O2 is not cubic. Figure 8 shows EDS spectra from points O1 and O2 within the oxide. REM, Zr, Ti, Ca, Al, and Mn peaks are detected at both points. The Cu peak originates from the TEM holder. It is considered that the oxide phase at point O1 represents the “gray zone” of the TRZ complex oxide shown in Fig. 3 since the Zr peak is much higher than the REM peak. The oxide phase around point O2 is considered to correspond with the “white region” shown in Fig. 3 considering its higher Ce peak. While there might be another (unidentified) phase at the thin area shown in Fig. 7, an EDS spectrum was not obtained in that location.

Fig. 7.

Bright field image of oxide and AF crystals shown in Fig. 5 and SAED pattern of the oxide (point O1).

Fig. 8.

EDS spectra from points (a) O1 and (b) O2 indicated in Fig. 7.

AF1 and AF2 are in contact with the oxide phase near point O1. Thus, it is concluded that the Zr-rich phase, rather than the other phases in the TRZ complex oxide, promoted nucleation of AF1 and AF2. A stereographic projection including AF1, AF2 and the Zr-rich phase in the TRZ complex oxide (denoted as “oxide”) observed in Fig. 7 are shown in Fig. 9 (based on electron diffraction) including the (100) and (110) poles for both phases. It should be noted that the crystal orientation of the oxide in Fig. 9(a) is different from that in Fig. 9(b) because the specimen was tilted to a different zone axis during TEM analysis to obtain suitable orientation information. It is notable in Fig. 9(a) that the (110)_AF1 and (011)_Oxide poles, (001)_AF1 and (011)_Oxide poles are close to each other. A different variant of the same relationship is also observed between AF2 and the Zr-rich oxide phase, albeit with a misorientation from the ideal relationship that is slightly larger. These results imply that a rational orientation relationship is formed between AF and the Zr-rich phase in the TRZ complex oxide. Assuming that the habit planes and parallel directions are (011)_AF/(011)_Oxide and [100]_AF/[011]_Oxide respectively, the planar disregistry, which is defined by following equation,²⁷⁾ is 8.8%   

$$\delta =\sum _{i=1}^3\frac{\left|d_{Oxide}^{i}cos\theta -d_{AF}^i\right|}{3\times d_{AF}^i}\times 100$$

where δ is the planar disregistry, θ is the angle between the low-index directions of the TRZ complex oxide and AF crystals in the habit planes, d_Oxide and d_AF are the interatomic spacings along low-index directions of the TRZ complex oxide and AF crystals in the habit planes respectively. If the habit planes and parallel directions were (100)_AF/(011)_Oxide [011]_AF/[011]_Oxide respectively, the planar disregistry would be larger, 13.0%. Therefore, it is believed that the following orientation and habit plane relationship is formed between AF and the Zr-rich phase in the TRZ complex oxide:   

$${\left(011\right)}_{\text{AF}}//{\left(011\right)}_{\text{Oxide}},{\left[100\right]}_{\text{AF}}//[01\overline{1}{]}_{\text{Oxide}}$$

Hereafter we refer to this as the “AF-TRZ orientation relationship”. In this study, nine AF crystals and five oxide particles were investigated. The various orientation relationships between the AF crystals and adjacent oxide particles (with a deviation angle of less than 5 degrees in all instances) are shown in Table 3. Three of the five oxide particles are adjacent to AF having the AF-TRZ orientation relationship with AF. The K-S orientation relationship ((110)_AF//(111)_Oxide, [111]_AF//[011]_Oxide) was also observed between an AF crystal and one of the five oxides.

Fig. 9.

(100) and (110) poles of (a) AF1, (b) AF2 and the Zr-rich phase of TRZ complex oxide (point O1) shown in Fig. 7. The crystal orientation of the oxide in (a) is different from that in (b) because the specimen was tilted to a different zone axis during TEM analysis to obtain suitable orientation information.

Table 3. Observed orientation relationship between AF and oxides.

Oxide number	AF crystal number	Observed orientation relationship
1	1	AF-TRZ
	2	Irrational
2	3	AF-TRZ
3	4	AF-TRZ
	5	Irrational
	6	Irrational
4	7	Irrational
5	8	K-S
	9	Irrational

4. Discussion

4.1. Influence of Ti-REM-Zr Complex Oxides on AF Nucleation

The AF formation rate in the TRZ-added alloy was much higher than in the Al-added and Ti-added alloys. This result implies that the Ti-REM-Zr complex oxides can act as favorable nucleation sites for AF. However, as pointed out by Barbaro et al.,²⁸⁾ the sizes of prior austenite grains (PAG) and of the oxide particles also affect AF formation. Therefore, one should consider the effects of these constituent dimensions in order to evaluate the AF formation potency of the Ti-REM-Zr complex oxide.

First, we evaluate the influence of oxide particle size on AF formation behavior. It has been reported that lager particles promote AF nucleation and a certain minimum particle size is required for AF nucleation.²⁸⁾ Moreover, Morikage et al.²⁹⁾ reported that the energy barrier for AF nucleation rises with a decrease of the size of oxide particle from the viewpoint of the balance of interfacial energies associated with curvature effects. The average CEDs of all oxide particles (D_A) and of the specific oxide particles on which AF nucleation was observed by OM (D_AF) are displayed in Table 4. The D_A and D_AF values in each alloy are almost the same. According to Barbaro’s study,²⁸⁾ the influence of oxide particle size on AF formation is remarkably large when oxide particle size is less than approximately 0.7 μm. In addition, Morikage et al.²⁹⁾ concluded that the energy barrier for AF nucleation decreases rapidly with oxide particle size up to 1 μm, then remains approximately constant when the oxide particle size is over 1 μm. In this study, the CED of the analyzed oxide particles is over 1.0 μm. Hence, it is concluded that the size of oxide particles had little influence on the AF formation rate in the present study. It is also notable that the AF formation rate for the TRZ-added alloy was the greatest among the three alloys, despite its finer oxide particle size distribution compared to the other alloys, as shown in Fig. 2.

Table 4. Average circle equivalent diameter of all oxide particles (D_A) and of oxides which act as nucleation sites for AF (D_AF).

	DA (μm)	DAF (μm)
Al-added	1.9	1.9
Ti-added	2.0	2.1
TRZ-added	1.8	1.8

As for the size of PAGs, it has been reported that the influence of PAG size on AF formation is large when the PAG size is less than 100 μm.²⁸⁾ According to this criterion, the PAG size should not have a large influence in the present study, where PAG sizes are greater than 200 μm.

Thus, it is concluded that the effects of both PAG and oxide particle sizes on AF formation rate may be small in this study, and that TRZ complex oxides strongly promote AF formation.

4.2. Formation Mechanism of AF on Ti-REM-Zr Complex Oxide

The AF-TRZ orientation relationship was frequently observed between AF and TRZ complex oxide crystals in the present study. The planar disregistry associated with the habit plane of the crystallographic relationship described earlier is 8.8%. This value is larger than reported for the AF-TiO relationship.²⁰⁾ Bramfitt has reported that carbides or nitrides having disregistry less than 12% act as nucleation sites for solidification of liquid iron.²⁷⁾ Therefore, it is considered that TRZ complex oxides promote AF nucleation due to the low disregistry associated with the AF-TRZ orientation relationship. Since some AF crystals do not satisfy any clear rational orientation relationships with the oxide, it is presumed that oxide particles acted as inert substrates for heterogeneous nucleation¹⁴⁾ or AF nucleation occurred on oxides present outside of the plane of imaging.

In addition to the AF-TRZ orientation relationship related to the oxides, the K-S orientation relationship was observed between the AF crystals and the austenite matrix in this study. The coexistence of both the AF-TRZ and K-S orientation relationships should be considered carefully. Miyamoto et al. reported that the crystal orientation of intragranular ferrite depends on the transformation temperature.³⁰⁾ More specifically, the intragranular ferrite which nucleates on MnS+V(C,N) precipitate at high temperature (953 K) satisfies the B-N orientation relationship with the precipitate. In contrast, AF formed at lower temperature (823 K) does not maintain the B-N orientation relationship with the precipitate, but has the K-S orientation relationship with the austenite matrix. Dilatometry was conducted to investigate the AF transformation temperature. Figure 10 shows the temperature-dilatation curve of the TRZ-added alloy during cooling. It seems that two dilatations occur, one at 990–840 K and the other at 720–540 K. The dilatations at higher and lower temperatures presumably correspond to the AF and bainite transformations, respectively. The transformation start temperature for AF is higher than in Miyamoto’s study, where a Fe–0.20%C–2.0%Mn–0.3%V alloy was investigated, and the AF in the present work satisfies both the K-S and AF-TRZ orientation relationships.

Fig. 10.

Relationship between temperature and dilatation of the TRZ-added alloy during cooling at 50 K/s following high temperature heat treatment at 1723 K for 5 s.

One explanation for the coexistence of the AF-TRZ and K-S orientation relationships is that they are simply accidental, i.e. a random occurrence. However, the probability that an AF crystal satisfies both the AF-TRZ and K-S orientation relationships within 5 degrees is only 18%, as determined using Grong’s method.⁹⁾ Considering that three of nine AF crystals examined in this study exhibited the unique orientation relationships, it is difficult to envision that the coexistence of the AF-TRZ and K-S orientation relationships is accidental. It is likely instead that AF satisfies both the AF-TRZ and K-S orientation relationships simultaneously, through favorable variant selection. That is, the AF crystal selects a K-S variant (during nucleation) which has a near-AF-TRZ orientation relationship with the oxide, resulting in the coexistence of both orientation relationships.

Another possibility considered is the presence of a rational orientation relationship between the TRZ complex oxide and the austenite matrix. For example, if austenite grains nucleated on oxide particles with a rational orientation relationship. In this case, only a single oxide particle (“inoculant”) within an austenite grain would have a rational orientation relationship with the austenite grain. According to the PAG size (347 μm) and number density of oxide particles (93 mm^–2) of the TRZ-added alloy, there should be nine oxide particles within an austenite grain on average. Hence, it is expected that one of nine oxide particles would have a rational orientation relationship with the austenite matrix following this mechanism. Considering that both the AF-TRZ and K-S orientation relationship are observed in three of five oxide particles, it is hard to attribute the formation of a rational orientation relationship between the oxides and the austenite matrix to nucleation of austenite grains on oxide particles. One final mechanism that could explain a rational orientation relationship between austenite and TRZ complex oxide is as follows:

– TRZ complex oxide particle liquefies during the high temperature heat treatment.

– The Zr-rich phase forms a rational orientation relationship with the surrounding austenite matrix during solidification of the liquid oxide during cooling.

For this hypothesis to be true, the liquefaction temperature of the TRZ complex oxide would need to be lower than the highest temperature encountered during heat treatment (1723 K). While the melting points of pure Ti oxides, Al₂O₃, CaO, CeO₂, and ZrO₂ are much higher than 1723 K,^31,32,33) it is conceivable that the complex oxide has a much lower liquefaction temperature, such as via eutectic reaction as in the Al₂O₃–CaO³¹⁾ or MnO–TiO₂³⁴⁾ systems, for example. Thus, the orientation relationship between the Zr-rich phase in the TRZ complex oxide and the austenite matrix was further investigated based on analysis of the TEM results. The crystal orientation of the austenite matrix was reconstructed from the measured orientations of the ferrite crystals, assuming a K-S orientation relationship between ferrite and austenite. A stereographic projection for a crystal of the Zr-rich oxide phase and the austenite matrix reconstructed from ferrite orientations (Fig. 7) are depicted in Fig. 11. Note that (111)_Austenite and (011)_Oxide poles, and (101)_Austenite and (011)_Oxide poles are close to each other. This implies the formation of the following oxide/austenite orientation relationship (referred to as the OX-AU orientation relationship):   

$$\text{(11}\overline{1}{\text{)}}_{\text{Austenite}}{\text{//(011)}}_{\text{Oxide}}{\text{, [101]}}_{\text{Austenite}}\text{//[01}\overline{1}{\text{]}}_{\text{Oxide}}$$

This orientation relationship is consistent with the coexistence of the K-S (AF/austenite) and AF-TRZ orientation relationships, or a distinctive “three-phase” orientation relationship between austenite, AF, and TRZ complex oxide. These orientation relationships were considered further using O-lattice theory.^35,36) In this theory, the misfit between precipitate and matrix phases is represented by the det |(I-A^–1)| value, where I is identity matrix and A is a transformation matrix that accounts for the orientation relationship between the crystals. Figure 12 shows the relationship between the lattice parameter ratio and the det |(I-A^–1)| value for habit planes of a variety of possible orientation relationships. The lattice parameter ratio for the austenite/TRZ complex oxide is 0.83. Taking this value into account, the OX-AU orientation relationship has the lowest misfit in comparison with the K-S, Nishiyama–Wassermann, cube-cube, AF-TRZ, and B-N orientation relationships between the austenite and the TRZ complex oxide. This implies that the formation of the OX-AU orientation relationship is expected from the viewpoint of lattice coherency. It should be noted that the K-S (AF/austenite) and AF-TRZ orientation relationships are also expected to form, considering their lattice parameter ratios (austenite/AF=1.3 and oxide/AF=1.5). While good lattice coherency between oxide and the austenite matrix may be expected to have a negative effect on the AF nucleation (due to low surface energy), it is believed that the formation of the OX-AU orientation relationship promotes AF nucleation indirectly, through the coexistence of the special AF-TRZ and K-S three phase orientation relationships.³⁷⁾

Fig. 11.

(110) and (111) poles of the Zr-rich phase (for point O1 in Fig. 7) and the austenite matrix inferred from orientations of ferrite crystals shown in Fig. 7.

Fig. 12.

Relationship between lattice parameter ratio and lattice misfit (det |(I-A^–1)| value) in O-lattice theory. a_m and a_p represent lattice parameters of the matrix and precipitate phases respectively. Misfit values are calculated for habit planes of a variety of possible orientation relationships between the austenite, AF, and TRZ complex oxide (Zr-rich phase).

5. Conclusions

Ti-REM-Zr (TRZ) complex oxides act as favorable nucleation sites for AF. A rational orientation relationship (AF-TRZ orientation relationship) was observed between AF and the Zr-rich phase in the TRZ complex oxide. This orientation relationship shows good lattice coherency, which is suggested to promote AF nucleation. AF nucleated on a TRZ complex oxide also holds the K-S orientation relationship with the austenite matrix. It is considered that the coexistence of the AF-TRZ and K-S special orientation relationships is caused by variant selection of AF and formation of a rational orientation relationship between the TRZ complex oxide and the austenite matrix during the high temperature thermal cycle associated with a simulated weld HAZ.

References

• 1)   Y.  Kasamatsu,  S.  Takashima and  T.  Hosoda: Tetsu-to-Hagané, 65 (1979), 1232.
• 2)   S. F.  Medina,  M.  Chapa,  P.  Valles,  A.  Quispe and  M. I.  Vega: ISIJ Int., 39 (1999), 930.
• 3)   S.  Kanazawa,  A.  Nakashima,  K.  Okamoto and  K.  Kanaya: Trans. Iron Steel Inst. Jpn., 16 (1976), 486.
• 4)   K. M.  Wu,  Y.  Inagawa and  M.  Enomoto: Mater. Charact., 52 (2004), 121.
• 5)   K.  Yamamoto,  T.  Hasegawa and  J.  Takamura: ISIJ Int., 36 (1996), 80.
• 6)   H. K.  Sung,  S. Y.  Shin,  W.  Cha,  K.  Oh,  S.  Lee and  N.  Kim: Mater. Sci. Eng. A, 528 (2011), 3350.
• 7)   Y.  Horii,  K.  Ichikawa,  S.  Ohkita,  S.  Funaki and  N.  Yurioka: Q. J. Jpn. Weld. Soc., 13 (1995), 500.
• 8)   T.  Koseki and  G.  Thewlis: Mater. Sci. Technol., 21 (2005), 867.
• 9)   Ø.  Grong,  A. O.  Kluken,  H. K.  Nylund,  A. L.  Dons and  J.  Hjelen: Metall. Mater. Trans. A, 26A (1995), 526.
• 10)   A. R.  Mills,  G.  Thewlis and  J. A.  Whiteman: Mater. Sci. Technol., 3 (1987), 1051.
• 11)   Y.  Tomita,  N.  Saito,  T.  Tsuzuki,  Y.  Tokunaga and  K.  Okamoto: ISIJ Int., 34 (1994), 829.
• 12)   J. M.  Gregg and  H. K. D. H.  Bhadeshia: Acta Metall. Mater., 42 (1994), 3321.
• 13)   J. M.  Gregg and  H. K. D. H.  Bhadeshia: Acta Metall. Mater., 45 (1997), 739.
• 14)   J. M.  Dowling,  J. M.  Corbett and  H. W.  Kerr: Metall. Trans. A, 17A (1986), 1611.
• 15)   N.  Mori,  H.  Homma,  M.  Wakabayashi and  S.  Okita: J. Jpn. Weld. Soc., 50 (1981), 786.
• 16)   T.  Yamada,  H.  Terasaki and  Y.  Komizo: ISIJ Int., 49 (2009), 1059.
• 17)   J.-S.  Byun,  J.-H.  Shim,  Y. W.  Cho and  D. N.  Lee: Acta Mater., 51 (2003), 1593.
• 18)   N.  Mori,  H.  Homma,  S.  Okita and  M.  Wakabayashi: J. Jpn. Weld. Soc., 50 (1981), 174.
• 19)   T.  Yamada,  H.  Terasaki and  Y.  Komizo: Tetsu-to-Hagané, 96 (2010), 608.
• 20)   K.  Kasai,  C.  Lee,  S.  Nambu,  J.  Inoue and  T.  Koseki: Tetsu-to-Hagané, 96 (2010), 123.
• 21)   Y.  Okazaki,  H.  Ishida,  K.  Suenaga and  T.  Hidaka: Weld. Int., 26 (2012), 593.
• 22)   A. G.  Fox,  E. S. K.  Menon and  G. M.  Evans: Proc. of 28th MST Annual Conf., Microscopy Society of Thailand, Thailand, (2011), 23.
• 23)   Y.  Ito and  M.  Nakanishi: J. Jpn. Weld. Soc., 44 (1975), 815.
• 24)   S.  Zaefferer: J. Appl. Cryst., 33 (2000), 10.
• 25)   A.  Agresti and  B. A.  Coull: Am. Stat., 52 (1998), 119.
• 26)   H.  Kitahara,  R.  Ueji,  N.  Tsuji and  Y.  Minamino: Acta Mater., 54 (2006), 1279.
• 27)   B. L.  Bramfitt: Metall. Trans., 1 (1970), 1987.
• 28)   F. J.  Barbaro,  P.  Krauklis and  K. E.  Easterling: Mater. Sci. Technol., 5 (1989), 1057.
• 29)   Y.  Morikage,  K.  Oi,  F.  Kawabata and  K.  Amano: CAMP-ISIJ, 9 (1996), 1353.
• 30)   G.  Miyamoto,  T.  Shinyoshi,  J.  Yamaguchi,  T.  Furuhara,  T.  Maki and  R.  Uemori: Scr. Mater., 48 (2003), 371.
• 31)   B.  Philips and  A.  Muan: Slag Atlas, 2nd ed., Verein Deutsher Eisenhuttenleute, Düsseldorf, (1995), 39.
• 32)   P. G.  Wahlbeck and  P. W.  Gilles: J. Am. Ceram. Soc., 49 (1966), 18.
• 33)   E.  Tani,  M.  Yoshimura and  S.  Somiya: J. Am. Ceram. Soc., 66 (1983), 506.
• 34)   B.  Philips and  A.  Muan: Slag Atlas, 2nd ed., Verein Deutsher Eisenhuttenleute, Düsseldorf, (1995), 92.
• 35)   W.  Bollmann: Surf. Sci., 31 (1972), 1.
• 36)   S.  Hashimoto: Bull. Jpn. Inst. Met., 22 (1983), 151.
• 37)   H.  Nako,  H.  Hatano,  O.  Yoshitomi,  K.  Yamashita and  M.  Otsu: ISIJ Int., 54 (2014), 1690.