Effects of Ga Concentration, Heat Treatment and Deformation Temperatureon Pseudoelasticity of Fe­Ga Polycrystals

Hiroyuki Y. Yasuda and Takenori Maruyama+

Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, Suita 565-0871, Japan

The pseudoelastic behavior of Fe­Ga polycrystals was examined focusing on the effects of Ga concentration, heat treatment anddeformation temperature. Fe­Ga polycrystals solutionized at 800°C exhibited pseudoelasticity at room temperature based on reversible motionof 1/4h111i superpartial dislocations dragging an antiphase boundary (APB) in the D03 superlattice structure. In particular, at and above25 at%Ga, paired 1/4h111i superpartials dragging the next-nearest-neighbor APB were found to play an important role in the pseudoelasticity.The amount of strain recovery at room temperature was highest at 25 at%Ga. Moreover, heat treatment strongly influenced the pseudoelasticbehavior of Fe­Ga alloys. For instance, the L12 and ½-like phases precipitating at higher Ga concentrations led to a decrease in strain recoverysince the precipitates suppressed the reversible motion of 1/4h111i superpartials. On the other hand, pseudoelasticity based on twinning anduntwinning of 2.2T-type pseudo-twins appeared at low temperatures, especially at ¹180°C. The energy of {211} pseudo-twins formed in theFe­Ga alloys was so high that the twins tended to disappear during unloading resulting in the pseudoelasticity. However, the amount of strainrecovery decreased if pseudoelasticities based on dislocation motion and twinning concurrently took place between ¹150 and ¹50°C.[doi:10.2320/matertrans.M2012266]

(Received August 1, 2012; Accepted October 12, 2012; Published December 25, 2012)

Keywords: intermetallic compounds, shape memory alloys, Fe3Ga, pseudoelasticity, dislocations, pseudo-twins

1. Introduction

Fe­Ga alloys have attracted much attention due to theirlarge magnetostriction and high strength.1) The phase equi-librium and microstructure in Fe­Ga alloys have beenextensively examined since the magnetostriction behaviordepended strongly on the microstructure.2­4) On the otherhand, we first reported that Fe3Ga single crystals with the D03superlattice structure showed large pseudoelasticity.5­11) It isalso interesting to note that three types of pseudoelasticitybased on dislocation motion, twinning and martensitictransformation appeared in Fe3Ga single crystals.8) In orderto understand the mechanism of the pseudoelasticity clearly,the effect of microstructure on the pseudoelastic behaviorin Fe3Ga single crystals was investigated focusing on Gaconcentration and heat treatment.6,8) However, Fe­Ga singlecrystals with high Ga contents above 25 at% are difficultto obtain because of complicated phase constituents inthe equilibrium phase diagram, as shown in Fig. 1(a).2)

In contrast, as shown in the metastable phase diagram(Fig. 1(b)), the D03 single-phase can be obtained in the wideGa concentration range since the formation rate of theequilibrium L12 phase was slow.2) Therefore, the pseudoe-lastic behavior of Fe­Ga alloys at high Ga concentrationscan be examined using the polycrystals even though thesingle crystals could not be produced. In this article, thepseudoelastic behavior of Fe­Ga polycrystals with 20­30 at%Ga was investigated focusing on Ga concentration,heat treatment and deformation temperature.

2. Mechanism of Pseudoelasticity in Fe3Ga SingleCrystals with the D03 Structure

In Fe3Ga single crystals with the D03 structure, there arethree types of pseudoelasticity based on dislocation motion,

twinning and martensitic transformation.8) Figure 2 showspossible dislocation configurations for h111i superlatticedislocation in the D03 structure which is based on the bccstructure (Fig. 2(a)).12,13) In general, h111i superdislocationin the structure is dissociated into four 1/4h111i superpartialdislocations bound by the nearest-neighbor and next-nearest-neighbor APBs (NNAPB and NNNAPB), which is desig-nated as a type III configuration (Fig. 2(d)). On the otherhand, a 1/4h111i superpartial dislocation moves independ-ently dragging the NNAPB in the type I configuration(Fig. 2(b)). In the type II configuration, the NNNAPBfollows paired 1/4h111i superpartials (Fig. 2(c)). In Fe3Gasingle crystals, both types I and II configurations wereselected during loading.5,6) During unloading, the APBspulled back the superpartials due to their tension resulting inthe pseudoelasticity. This type of pseudoelasticity is called“APB pseudoelasticity”. Note that D03-ordered Fe3Al singlecrystals also exhibited the APB pseudoelasticity.14­16)

Figure 3 shows the atomic arrangement near the twinsformed in Fe3Ga crystals with the D03 structure. If twin shearis introduced in the D03 crystals parallel to h111i direction,the positions of Ga atoms in the D03 matrix and the twins arenot mirror-symmetric about {211} twin plane as shown inFig. 3(a). Such a twin is called a 2.2T-type pseudo-twin.17)

Note that the unit cell of the pseudo-twin becomes theorthorhombic structure (Fig. 3(b)). Thus, the twinning can beregarded as a kind of displacive transformation. The energyof the pseudo-twins was so high that the twins disappearedduring unloading resulting in the pseudoelasticity.8­10) Here-after, we call this phenomenon “twinning pseudoelasticity”.

It is well known that pseudoelasticity is generally based onmartensitic transformation.18) During loading, a thermoelasticmartensite is stress-induced while the reverse transformationtakes place during unloading, resulting in transformationpseudoelasticity. In solutionized Fe­23.8 at%Ga single crys-tals, stress-induced and the reverse transformation of themartensites with the 14M structure led to the transformation+Graduate Student, Osaka University

Materials Transactions, Vol. 54, No. 1 (2013) pp. 36 to 42©2012 The Japan Institute of Metals

pseudoelasticity. These details are well described in aprevious paper.8,10)

3. Experimental Procedure

Master ingots of Fe­Ga alloys containing 20, 23, 25, 28and 30 at%Ga were prepared by melting high purity Fe andGa using a plasma arc furnace. These ingots were hot rolledat 1100°C followed by homogenization at 1100°C. Com-pression specimens (2.5 © 2.5 © 5.5mm) were cut from therolled samples by spark machining and then enclosed inquartz tubes filled with argon gas. The specimens weresolution-treated at 800°C for 1 h and subsequently quenchedin iced water. The solutionized polycrystals are designated as

ST throughout this paper. It is noted that the quartz tubeswere not crushed in iced water to avoid a thermal shock.After the solutionization, these crystals were annealed for1 h between 300 and 600°C to examine the effect of heattreatment on the pseudoelasticity.

Compression tests were performed in air from ¹180°C toroom temperature. The specimens were compressed to aplastic strain (¾p) of 2.5% at a constant cross-head speedcorresponding to an initial strain rate of 1.7 © 10¹4 s¹1.During unloading, the specimens were unloaded at the samecross-head speed. The amount of strain recovery wasevaluated using the recovery ratio (R), which is defined asfollows:

R ¼ ¾p � ¾r

¾p� 100 ð1Þ

where ¾r is the residual strain after complete unloading.The surface of the deformed samples was observed by an

optical microscope equipped with Nomarski contrast. Themicrostructure of the samples was observed by a transmissionelectron microscope (TEM) operated at 300 kV. Note that thecrystal planes, directions and reciprocal vectors in the Fe­Gaalloys were indexed with respect to the D03 lattice.

4. Results

4.1 Microstructure of Fe­Ga alloysFigure 4 shows selected area electron diffraction patterns

(SAEDP) of ST. Faint spots of 111 and 020 reflections thatare typical in the D03 phase are visible even at 20 at%Ga

Fig. 2 Unit cell of the D03 structure (a) and dislocation configuration in the D03 crystals. (b) Type I, (c) type II and (d) type III.

Fig. 3 Atomic arrangement of 2.2T-type pseudo-twins in Fe3Ga with theD03 structure. (a) Atomic arrangement of the pseudo-twin, (b) unit cell ofthe pseudo-twin.

Fig. 1 Equilibrium (a) and metastable (b) Fe­Ga phase diagrams.2)

Effects of Ga Concentration, Heat Treatment and Deformation Temperature on Pseudoelasticity of Fe­Ga Polycrystals 37

(Fig. 4(a)). In addition, 111 and 020 reflections can be clearlyseen at and above 23 at%Ga (Figs. 4(b)­4(e)). This meansthat the D03 ordering in Fe­Ga alloys is so fast that the phasedeveloped even during the quenching process. In fact, atypical ordered domain structure develops in the D03 phase inthe Fe­25 at%Ga ST as shown in Fig. 5. In contrast, no L12phase which is the equilibrium phase at low temperatures(Fig. 1(a)) is observed suggesting that the formation of theL12 phase is slow. It is also noted there are some diffuse spotsaround the 020 and 222 reflections, that cannot be indexedwith respect to the D03 structure (Figs. 4(a)­4(f )). Theadditional reflections come from the oxide passive layerformed during electropolishing, which is well described in aprevious paper.6)

If Fe­25­30 at%Ga polycrystals are annealed after thesolutionization, the equilibrium L12 and the ½-like phases areformed as shown in Fig. 6. For instance, the L12 phasecontaining the twin boundary can be seen in Fe­25 at%Gaannealed at 600°C for 1 h (Figs. 6(a) and 6(b)). On the otherhand, there are fine ½-like precipitates in Fe­30 at%Ga alloysannealed at 500°C for 1 h as shown in Fig. 6(c). From theSAEDP (Figs. 6(d) and 6(e)), the ½-like phase is composedof two crystallographic variants, which is consistent with a

previous paper.19) Table 1 summarizes the phase constituentsof Fe­Ga alloys annealed from 300 to 600°C aftersolutionization. The D03 ordering is so fast that the D03phase precipitates from the bcc matrix during the quenchingprocess even at 20 at%Ga, though the quartz tubes enclosingthe specimens were not crashed in iced water. The equi-librium L12 phase tends to be formed in Fe­25­30 at%Gaalloys at and above 500°C. On the other hand, numerousprecipitates of the ½-like phase are observed in Fe­30 at%Gaalloys annealed at and above 300°C.

4.2 Pseudoelastic behavior at room temperatureFigure 7 shows stress­strain (S­S) curves of ST com-

pressed at room temperature. Pseudoelasticity obviously takesplace, especially at and above 23 at%Ga. At 20 at%Ga, aserrated flow can be seen during loading while there are fewserrations during unloading. This means that strain recoveryat 20 at%Ga and at room temperature is independent oftwinning pseudoelasticity. On the other hand, no serration isobserved in the S­S curves at 23 at%Ga and 25 at%Ga. Itshould also be noted that the nicked S­S curves at 28 at%Gaand 30 at%Ga reflect the formation of small cracks alonggrain boundaries, irrelevant to twinning. These stronglysuggest that the APB pseudoelasticity based on the dis-location motion takes place in Fe­Ga polycrystals at roomtemperature. However, the amount of strain recovery in thepolycrystals is much smaller than that of the single crystals.5,6)

R of Fe­Ga polycrystals compressed at room temperaturedepends strongly on Ga concentration and annealing

Fig. 4 SAEDP of ST (a) 20 at%Ga, (b) 23 at%Ga, (c) 25 at%Ga, (d) 28 at%Ga and (e) 30 at%Ga and (f ) a schematic illustration of thepattern.

Fig. 5 A typical ordered domain structure in the D03 phase in Fe­25 at%Ga ST; a reflection vector (g) of 111.

Table 1 Phase constituent of Fe­Ga alloys annealed at different temper-atures after solutionization at 800°C.

Ga content(at%)

20 23 25 28 30

ST bcc + D03 D03 D03 D03 D03annealed 300°C bcc + D03 D03 D03 D03 D03 + ½

annealed 500°C bcc + D03 D03 D03 + L12 D03 + L12 D03 + L12 + ½

annealed 600°C bcc + D03 D03 D03 + L12 D03 + L12 D03 + L12 + ½

H. Y. Yasuda and T. Maruyama38

condition as shown in Fig. 8. R of ST shows a maximum at25 at%Ga and decreases with increasing deviation from25 at%Ga. Moreover, at 20 at%Ga and 23 at%Ga where noL12 and ½-like precipitates are formed, R is insensitive to theheat treatment at temperatures from 300 to 600°C. Incontrast, annealing between 300 and 600°C causes a decreasein R at and above 25 at%Ga. In particular, Fe­28 at%Gaannealed at 500 or 600°C and Fe­30 at%Ga at 300°C exhibitlittle strain recovery. Furthermore, Fe­30 at%Ga annealedat 500 or 600°C were so brittle that the polycrystals easilybroke during loading. At and above 25 at%Ga, the equi-librium L12 phase and/or the ½-like phase precipitated bythe heat treatment (Fig. 6), which is closely related to thepseudoelastic behavior.

Figures 9(a) and 9(b) show optical micrographs of Fe­20 at%Ga and Fe­25 at%Ga ST compressed at room temper-ature. At 20 at%Ga, coarse {211} twin bands are predom-inantly observed, as shown in Fig. 9(a), which is consistentwith the serration observed in the S­S curve shown in Fig. 7.It is also mentioned that the twins hardly disappear during

Fig. 6 TEM micrographs (a), (c) and SAEDP (b), (d) of Fe­Ga polycrystals. (a), (b) the equilibrium L12 phase in Fe­25 at%Ga annealedat 500°C for 1 h. (c), (d) the ½-like phase in Fe­30 at%Ga annealed at 300°C for 1 h. The schematic illustration for (d) is also shownin (e).

Fig. 7 S­S curves of Fe­20­30 at%Ga ST compressed to ¾p = 2.5% at room temperature.

Fig. 8 Variation in R with Ga content in Fe­20­30 at%Ga polycrystalsannealed at different temperatures and then compressed to ¾p = 2.5% atroom temperature.

Effects of Ga Concentration, Heat Treatment and Deformation Temperature on Pseudoelasticity of Fe­Ga Polycrystals 39

unloading suggesting that the twinning pseudoelasticityrarely contributes to the strain recovery in Fe­20 at%Gapolycrystals at room temperature. In contrast, there exists fine{101} slip traces at 25 at%Ga (Fig. 9(b)). In addition, notwins were observed in Fe­23­30 at%Ga ST compressed atroom temperature. Furthermore, paired 1/4h111i superpartialdislocations showing a type II configuration (Fig. 2(c)) isobserved in Fe­28 at%Ga ST (Fig. 9(c)). This suggests thatthe APB pseudoelasticity based on the dislocation motionappears in ST at room temperature.

4.3 Pseudoelastic behavior at low temperaturesIn contrast to the APB pseudoelasticity at room temper-

ature, Fe­20­25 at%Ga ST compressed at low temperaturesdemonstrate the twinning pseudoelasticity as shown inFig. 10. A serrated flow can be seen in the S­S curves at¹180°C both during loading and unloading. Note that theformation of the twins accompanies a clicking sound. A largestrain recovery occurs during unloading due to the twinningpseudoelasticity. However, the pseudoelastic behavior of Fe­28 at%Ga and Fe­30 at%Ga ST at low temperatures could notbe examined since the crystals broke during loading due totheir brittleness at grain boundaries. Figure 11 shows opticalmicrographs of Fe­20 at%Ga and Fe­25 at%Ga ST com-pressed at ¹180°C. Since microstructural observation wasdone at room temperature after compression at low temper-atures, some defects possibly have disappeared duringunloading. However, the contrast of the {211} twin bandsat ¹180°C (Fig. 11) is fainter than that at room temperature(Fig. 9(a)), associated with the untwinning process. More-

over, no slip trace caused by dislocation motion is observed,suggesting that the twinning pseudoelasticity predominantlyappears at ¹180°C. R of Fe­20­25 at%Ga ST is plottedagainst deformation temperature in Fig. 12. At 25 at%Ga,R shows a high value at ¹180°C due to the twinningpseudoelasticity. However, R decreases at ¹150°C whilefurther increase in deformation temperature results in anincrease in R again. R of Fe­23 at%Ga ST shows a tendencysimilar to that of Fe­25 at%Ga, though a minimum R isobtained at ¹100°C. On the other hand, Fe­20 at%Ga STexhibit high R only at ¹180°C.

Fig. 9 Optical (a), (b) and TEM (c) micrographs of Fe­Ga ST compressed to ¾p = 2.5% at room temperature. (a) 20 at%Ga, (b) 25 at%Gaand (c) weak-beam dark-field image, 28 at%Ga, g = 202, g/4g condition.

Fig. 10 S­S curves of Fe­20­25 at%Ga ST compressed to ¾p = 2.5% at¹180°C.

Fig. 11 Optical micrographs showing the {211} pseudo-twins in Fe­Ga STcompressed to ¾p = 2.5% at ¹180°C; (a) 20 at%Ga and (b) 25 at%Ga.

Fig. 12 Variation in R with deformation temperature in Fe­20­25 at%GaST compressed to ¾p = 2.5%.

H. Y. Yasuda and T. Maruyama40

5. Discussion

Fe­20­30 at%Ga polycrystals exhibited pseudoelasticitybased on the motion of 1/4h111i superpartials and 2.2T-typepseudo-twins depending on Ga concentration, heat treatmentand deformation temperature. In particular, the APB pseudo-elasticity appeared even in polycrystalline Fe­28 at%Ga andFe­30 at%Ga alloys of which single crystals are difficultto produce. In contrast, the transformation pseudoelasticitybased on the 14M martensite has never been observed inFe­Ga polycrystals in the present study. Since the trans-formation pseudoelasticity in Fe­Ga alloys took place at thelimited Ga concentration around 24 at%,8) the pseudoelas-ticity was not observed. Therefore, the behaviors of the APBand twinning pseudoelasticities in Fe­20­30 at%Ga poly-crystals were discussed focusing on Ga concentration, heattreatment and deformation temperature.

At room temperature, the APB pseudoelasticity based onthe dislocation motion preferentially took place in Fe­Ga STas shown in Fig. 7. When the APB pseudoelasticity appears,a critical resolved shear stress (¸y) and a recovery-start shearstress (¸r) can be written as:5,6,14­16)

¸y ¼ ¸0 þ ¸b ð2Þ¸r ¼ ¸b � ¸0 ð3Þ

where ¸0 is the frictional stress of 1/4h111i superpartials and¸b is the backward stress due to the APB. In general, lower ¸0and higher ¸b lead to higher recovery ratio. Both ¸0 and ¸b canbe calculated by solving the simultaneous eqs. (2) and (3). ¸yand ¸r were calculated from the yield stress and recovery-startstress assuming that the Taylor factor is 2.13) ¸b of ST wascalculated from eqs. (2) and (3) and is plotted against Gaconcentration in Fig. 13. ¸b increases with increasing Gaconcentration and reaches a maximum at 25 at%Ga. Furtherincrease in Ga content leads to a decrease in ¸b. Thedependence of ¸b on Ga concentration is in good agreementwith R; higher ¸b results in higher R as shown in Fig. 8. Next,let us consider the dislocation configuration in Fe­Ga alloys.In the case of the type I configuration, uncoupled 1/4h111isuperpartials move individually dragging the NNAPB(Fig. 2(b)). In the configuration, ¸b is mainly caused by thetension of the NNAPB (¸NN), that is given by,13)

¸NN ¼ 2ffiffiffi

2p

ba20f4V1S

21 þ V2ðS2

2 � 4S21Þg ð4Þ

where a0 is the lattice constant, b is the magnitude of1/4[111] superpartials, V1 and V2 are the first- and second-nearest ordering energies and S1 and S2 are the first- andsecond-nearest degrees of order, respectively. In the type IIconfiguration (Fig. 2(c)) in which paired 1/4h111i super-partials glide dragging the NNNAPB, the back stress due tothe NNNAPB (¸NNN) is written as:13)

¸NNN ¼ffiffiffi

2p

ba20f2V2S

22g: ð5Þ

If the maximum degree of order is achieved in Fe­Ga alloys,the S1 and S2 depend on Ga concentration (x) as follows:13)

S1 ¼ 2x ð0 � x � 0:5Þ ð6ÞS2 ¼ 4x ð0 � x � 0:25Þ ð7Þ

S2 ¼ 2ð1� 2xÞ ð0:25 � x � 0:5Þ ð8ÞThis assumption is reasonable since the D03 ordering is veryfast as described in a previous paper.5,6) ¸NN and ¸NNN werecalculated from eqs. (4)­(8) and are also plotted in Fig. 13. ¸bis in quantitatively good agreement with ¸NNN, especially atand above 25 at%Ga. This means that the type II config-uration is more frequently selected than the type I in the Gaconcentration range. Actually, the type II configurationwas observed to be chosen at 28 at%Ga (Fig. 9(c)). Thus,it is concluded that the dislocation motion with thetype II configuration demonstrated the APB pseudoelasticity,especially at and above 25 at%Ga. In contrast, the APBpseudoelasticity in Fe3Al single crystals is based on thetype I configuration in which 1/4h111i superpartial disloca-tions move individually dragging the NNAPB.14­16) ¸NNN ofFe3Ga alloys was much higher than that of Fe3Al alloys,5)

which led to the APB pseudoelasticity based on the type IIconfiguration in Fe3Ga alloys. However, R of Fe­Gapolycrystals was much smaller than that of the singlecrystals.5,6) Grain boundaries in Fe­Ga polycrystals sup-pressed the reversible motion of 1/4h111i superpartialsresulting in a decrease in R.

R at and below 23 at%Ga was insensitive to the heattreatment while those at and above 25 at%Ga dependedstrongly on the annealing condition as shown in Fig. 8.At and above 25 at%Ga, the equilibrium L12 or the ½-likephase precipitated by the heat treatment (Fig. 6, Table 1).There is no driving force for the APB pseudoelasticityin these phases. Moreover, the L12 and the ½-like phasessuppressed the to-and-fro motion of 1/4h111i superpartialsin the D03 phase, which resulted in a decrease in R (Fig. 8).For instance, little strain recovery was observed in Fe­28 at%Ga heat-treated at 500 and 600°C, associated withthe formation of the L12 phase. In addition, Fe­30 at%Gaannealed at and above 500°C easily broke due to the L12and ½-like phases.

Below room temperature, the twinning pseudoelasticitybased on 2.2T-type pseudo-twins appeared at and below25 at%Ga, especially at ¹180°C. The S­S curves showednumerous serrations both during and unloading, associated

Fig. 13 Variation of ¸b, ¸NN and ¸NNN as a function of Ga concentration.The ¸b is for Fe­Ga ST compressed to ¾p = 2.5% at room temperature.

Effects of Ga Concentration, Heat Treatment and Deformation Temperature on Pseudoelasticity of Fe­Ga Polycrystals 41

with the formation and annihilation of the pseudo-twins,respectively (Fig. 10). However, R of Fe­23 at%Ga andFe­25 at%Ga ST showed low values between ¹150 to¹50°C where both the APB and twinning pseudoelasticitiesconcurrently took place, which is well documented in aprevious paper.9) An intersection between the dislocationsand twins suppressed their motion resulting in a decreasein R at ¹150 to ¹50°C. In other words, at ¹180°C androom temperature, the twinning and APB pseudoelasticitiesseparately appeared at 23­25 at%Ga, respectively, which ledto high R at the temperatures. In contrast, Fe­20 at%GaST exhibited high R at ¹180°C due to the twinningpseudoelasticity while R showed low values at and above¹150°C. Figure 14 shows TEM micrographs of the pseudo-twins in ST compressed at ¹180 or ¹50°C. At ¹180°C(Fig. 14(a)), dislocations are rarely observed near the pseudo-twin while numerous dislocations are introduced near thetwin at ¹50°C (Fig. 14(b)). The dislocations which accom-modate the strain caused by the twin formation may stabilizethe twin boundaries, resulting in a decrease in R at and above¹150°C.

6. Conclusions

Pseudoelastic behavior of Fe­Ga polycrystals was exam-ined focusing on Ga concentration, heat treatment anddeformation temperature. The following conclusions werereached.(1) At room temperature, the APB pseudoelasticity based

on the reversible motion of 1/4h111i superpartialsappeared in ST. R showed a maximum at 25 at%Ga anddecreased with increasing deviation from 25 at%Ga.

(2) The APB pseudoelasticity at and above 25 at%Garesulted from the reversible motion of paired1/4h111i superpartials dragging the NNNAPB. Duringunloading, the NNNAPB pulled back the superpartialsresulting in the APB pseudoelasticity.

(3) At and above 25 at%Ga, the equilibrium L12 and the½-like phases were precipitated by the appropriate heattreatment, which was harmful for the APB pseudoelas-ticity at room temperature.

(4) Below room temperature, the twinning pseudoelasticity

based on twinning and untwinning of the 2.2T-typepseudo-twins appeared at 20­25 at%Ga, especially at¹180°C.

(5) If the APB and twinning pseudoelasticities solely tookplace at room temperature and ¹180°C, respectively,large strain recovery could be obtained at 23 at%Ga and25 at%Ga. In contrast, the interaction between thedislocations and the pseudo-twins associated with theAPB and twinning pseudoelasticities, respectively, ledto a decrease in R at ¹150 to ¹50°C. An accommo-dation of twin boundaries by dislocations at 20 at%Gaalso resulted in a decrease in R.

Acknowledgments

This work was supported by a Grant-in-Aid for ScientificResearch (B) from the Ministry of Education, Culture,Sports, Science and Technology of Japan. This work was alsosupported by “Priority Assistance of the Formation ofWorldwide Renowned Centers of Research®The GlobalCOE Program (Project: Center of Excellence for AdvancedStructural and Functional Materials Design)” from theMinistry of Education, Culture, Sports, Science and Tech-nology of Japan.

REFERENCES

1) A. E. Clark, J. B. Restorff, M. Wun-Fogle, T. A. Lograsso and D. L.Schlagel: IEEE Trans. Magn. 36 (2000) 3238­3240.

2) O. Ikeda, R. Kainuma, I. Ohmura, K. Fukamichi and K. Ishida:J. Alloy. Compd. 347 (2002) 198­205.

3) Q. Xing, Y. Du, R. J. McQueeney and T. A. Lograsso: Acta Mater. 56(2008) 4536­4546.

4) S. Bhattacharyya, J. R. Jinschek, J. F. Li and D. Viehland: J. Alloy.Compd. 501 (2010) 148­153.

5) H. Y. Yasuda, M. Aoki, Y. Umakoshi and A. Takaoka: Scr. Mater. 53(2005) 253­257.

6) H. Y. Yasuda, M. Aoki and Y. Umakoshi: Acta Mater. 55 (2007)2407­2415.

7) H. Y. Yasuda, K. Fukushima, M. Aoki and Y. Umakoshi: ISIJ Int. 48(2008) 1014­1019.

8) H. Y. Yasuda, Y. Oda, M. Aoki, K. Fukushima and Y. Umakoshi:Intermetallics 16 (2008) 1298­1304.

9) H. Y. Yasuda, T. Kishimoto and Y. Umakoshi: Mater. Trans. 51 (2010)2196­2200.

10) H. Y. Yasuda and Y. Umakoshi: Proc. Intermetallic-Based Alloys forStructural and Functional Applications, 2010 MRS, ed. by M. Palm,B. P. Bewlay, K. S. Kumar, K. Yoshimi, (Cambridge University Press,Cambridge, 2011) pp. 21­32.

11) H. Y. Yasuda, T. Furuta and T. Maruyama: Mater. Sci. Forum 706­709(2011) 2032­2037.

12) M. J. Marcinkowski and N. Brown: Acta Metall. 9 (1961) 764­786.13) H. J. Leamy and F. X. Kayser: Phys. Status Solidi 34 (1969) 765­780.14) L. P. Kubin, A. Fourdeux, J. Y. Guedou and J. Rieu: Philos. Mag. A 46

(1982) 357­378.15) H. Y. Yasuda, K. Nakano, T. Nakajima, M. Ueda and Y. Umakoshi:

Acta Mater. 51 (2003) 5101­5112.16) H. Y. Yasuda, T. Nakajima, K. Nakano, K. Yamaoka, M. Ueda and Y.

Umakoshi: Acta Mater. 53 (2005) 5343­5351.17) J. W. Christian and D. E. Laughlin: Acta Metall. 36 (1988) 1617­1642.18) K. Otsuka and C. M. Wayman (ed.): Shape Memory Materials,

(Cambridge University Press, Cambridge, 1998) p. 1.19) M. Nakagawa, S. Matsumura, N. Kuwano and K. Oki: Scr. Metall. 21

(1987) 461­464.

Fig. 14 TEM micrographs of ST compressed to ¾p = 2.5%. (a) Fe­20 at%Ga compressed at ¹180°C, (b) Fe­25 at%Ga compressed at¹50°C. The dislocation tangles are indicated by “dis.”.

H. Y. Yasuda and T. Maruyama42