《复合材料 Composites》课程教学资源（学习资料）第五章 陶瓷基复合材料_Experimental evidence for magnetic or electric field effects on

MIATERIALS ENE S ENGINEERING ELSEVIER Materials Science and Engineering A287(2000)213-218 www.elsevier.com/locate/msea Experimental evidence for magnetic or electric field effects on phase transformations C.C. Koch Department of Materials Science and Engineering, North Carolina State Unirersity, Raleigh, NC 27695-7907, USA abstract This review presents examples of the effects of magnetic or electric fields on phase transformations in a variety of materials Magnetic fields have been shown to influence phase stability and, for example, induce martensite in iron -base alloys. Application of a magnetic field during the growth of some phases can produce an alignment with the field and an optimized microstructure for certain properties. Electric fields can modify phase transformations by their enhancement of atomic diffusivity, as, for example, their influence on diffusion-controlled transformations such as eutectoidal decomposition of steel or age hardening of aluminum alloys. Electric fields can also bias phase transformations toward phases which have higher dielectric constants than the parent phase. Much more research is needed to fully exploit magnetic/electric fields as another tool to control structure, and therefore properties of materials. C 2000 Elsevier Science S.A. All rights reserved Keywords: Phase transformations; Magnetic field; Electric field; Martensite; Phase separation; Precipitation 1. Introduction This paper will give examples of experiments which illustrate the influence of magnetic fields and electric Phase transformations are determined by thermody- fields on phase transformations. Examples of phase namics. i. e. can the transformation occur? - and transformations in a wide variety of inorganic and kinetics -i.e. how fast can the transformation occur? organic materials will be given to show the breadth of In terms of the thermodynamics that limits phase sta- phase transformation phenomena that can be bility, the minimum Gibbs free energy specifies which phase is stable under particular conditions. The vari ables commonly applied to phase stability are tempera ure and pressure. However, magnetic field intensity 2. Effects of magnetic fields and electric field strength can also contribute to the free energy since phases may have, e. g. different magnetic 2. 1. Effect of magnetic field, H, on phase stability susceptibilities or different dielectric constants. Thus magnetic or electric fields can modify the free energy 2.1.1. Martensite start temperature, M, in Fe-base and therefore the conditions under which phases are aLlows stable Since the magnetization of the ferromagnetic bcc The kinetics of phase transformations can also be structure phase in Fe-based alloys is much greater than affected by magnetic or electric fields. Transformations that of the fcc paramagnetic parent phase, magnetic which require atomic transport can be modified by the fields can affect the Ms temperature of the fcc-bcc effect electric fields, for example martensitic transformation. An expression The growth morphology can also be controlled by derived for the change in equilibrium temperature be- magnetic field direction in transformations involving tween phases of unequal magnetizations subjected to an phases with differing magnetic properties applied magnetic field [1]. It is found that the tempera ture of equilibrium between the fcc and bcc phases is *Tel.:+1-919-5152377;fax:+1-919-5157724 increased by a few degrees K per Tesla of applied E-mail address: carl koch(@ncsu. edu(C C. Koch) magnetic field. The estimation of the equilibrium tem- 0921-5093/00/S- see front matter o 2000 Elsevier Science S.A. All rights reserved PI:S0921-509300)00778-4

Materials Science and Engineering A287 (2000) 213–218 Experimental evidence for magnetic or electric field effects on phase transformations C.C. Koch * Department of Materials Science and Engineering, North Carolina State Uni6ersity, Raleigh, NC 27695-7907, USA Abstract This review presents examples of the effects of magnetic or electric fields on phase transformations in a variety of materials. Magnetic fields have been shown to influence phase stability and, for example, induce martensite in iron–base alloys. Application of a magnetic field during the growth of some phases can produce an alignment with the field and an optimized microstructure for certain properties. Electric fields can modify phase transformations by their enhancement of atomic diffusivity, as, for example, their influence on diffusion-controlled transformations such as eutectoidal decomposition of steel or age hardening of aluminum alloys. Electric fields can also bias phase transformations toward phases which have higher dielectric constants than the parent phase. Much more research is needed to fully exploit magnetic/electric fields as another tool to control structure, and therefore properties of materials. © 2000 Elsevier Science S.A. All rights reserved. Keywords: Phase transformations; Magnetic field; Electric field; Martensite; Phase separation; Precipitation www.elsevier.com/locate/msea 1. Introduction Phase transformations are determined by thermody￾namics, i.e. can the transformation occur? — and kinetics — i.e. how fast can the transformation occur? In terms of the thermodynamics that limits phase sta￾bility, the minimum Gibbs free energy specifies which phase is stable under particular conditions. The vari￾ables commonly applied to phase stability are tempera￾ture and pressure. However, magnetic field intensity and electric field strength can also contribute to the free energy since phases may have, e.g. different magnetic susceptibilities or different dielectric constants. Thus magnetic or electric fields can modify the free energy and therefore the conditions under which phases are stable. The kinetics of phase transformations can also be affected by magnetic or electric fields. Transformations which require atomic transport can be modified by the effect electric fields, for example, have on diffusivity. The growth morphology can also be controlled by magnetic field direction in transformations involving phases with differing magnetic properties. This paper will give examples of experiments which illustrate the influence of magnetic fields and electric fields on phase transformations. Examples of phase transformations in a wide variety of inorganic and organic materials will be given to show the breadth of phase transformation phenomena that can be influenced. 2. Effects of magnetic fields 2.1. Effect of magnetic field, H, on phase stability 2.1.1. Martensite start temperature, Ms, in Fe–base alloys Since the magnetization of the ferromagnetic bcc structure phase in Fe–based alloys is much greater than that of the fcc paramagnetic parent phase, magnetic fields can affect the Ms temperature of the fccbcc martensitic transformation. An expression has been derived for the change in equilibrium temperature be￾tween phases of unequal magnetizations subjected to an applied magnetic field [1]. It is found that the tempera￾ture of equilibrium between the fcc and bcc phases is increased by a few degrees K per Tesla of applied magnetic field. The estimation of the equilibrium tem- * Tel.: +1-919-5152377; fax: +1-919-5157724. E-mail address: carl–koch@ncsu.edu (C.C. Koch) 0921-5093/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved. PII: S0921-5093(00)00778-4

214 C C. Koch/ Materials Science and Engineering 4287(2000 )213-218 perature is useful in accounting for experimental data field-induced martensitic transformation in Fe-Ni al on the promotion of athermal martensitic transforma- loys for Fe-299 at Ni, Fe-317 at Ni, and Fe- tions by magnetic fields. Experimental studies in the 32.5 at. Ni The slope of the temperature dependence 1960s [2] observed small increases in the M, tempera- of critical magnetic field strength to induce the marten- ture with the modest magnetic field used. Kakeshita et site becomes smaller in the order of the 31.7, 29.9, and al. [3]used pulsed magnetic fields up to 31.75 MA m 32.5 at. Ni alloys. That is, the magnetic field is most to induce larger increases in M,. The increase in M, effective in inducing martensite in the 31.7 at. Ni temperature, AT, versus magnetic field is shown in Fig alloy. For constant AT, the amount of magnetic-field I for a Fe-317 at Ni alloy. All the martensites in induced martensite is about constant with magnetic his alloy were lenticular with a mid-rib and no differ- field strength for the 29.9 and 31.7 at. Ni alloys but ences in martensite morphology were noted between the increases with field strength for the 32.5 at. Ni alloy hermally-induced and magnetically-induced marten- This was found to be due to both the formation of new sites. Subsequent studies of Kakeshita et al. [4] com- martensite plates and the growth of existing plates in pared the composition dependence of magnetic the 32.5 at. Ni alloy- that is, they are partly thermoelastic. The 29.9 and 31.7 at.% Ni martensites re nonthermoelastic Kakeshita et al. [5] have recently studied the effects of magnetic fields on both the isothermal and athermal E kinetics of martensite transformations in Fe-24 9 Ni- 39 Mn and Fe-314 Ni-05 Mn alloys, respectively For the Fe-24.9Ni-39 Mn alloy, a pulsed magnetic field above a critical value induced martensite instanta- neously. This suggests that the original isothermal ki netics of the martensitic transformat changed athermal behavior under high magnetic fields. At static of the isothermal martensitic transformation were ch of the T lower temperature and shorter incubation time than that for ied field as shown in Fig. 2. Models susceptibility, and forced volume magnetostriction were △T(=TMs)(K) able to explain the observed influence of magnetic field Watanabe and Sato [6] showed that the martensitic Fig. 1. Increase in M, temperature, Ar versus magnetic field (after transformation of 90-nm Fe particles in a Cu-1.5 wt% Ref.[]). Fe alloy involving the change from the antiferromag- netic to ferromagnetic state at 4.2 K was promoted by an applied magnetic field. The martensitic transforma- 0. 1% Transformation tion was induced by tensile straining the samples at 4.2 K at a strain rate of 1. 7x10-4s-I with or without an applied magnetic field of 5.6 T parallel to the tensile 200 axis. Without the magnetic field 29+1 vol. of martensite(a-Fe) was formed compared to about 50+ 2 vol. in the presence of the field 100 2. 1.2. Other examples of phase transformations affected by applied magnetic fields the pressure-induced phase transformation between the 81 and B31 phases of M 101010 minor constituent, is superparamagnetic and the b31 Holding Time to 1/sec phase goes through a paramagnetic to antife netic transition with hydrostatic pressure near 3 katm Fig. 2. TTT diagrams of the isothermal martensitic transformation in The application of a pulsed magnetic field of 100 kOe an Fe-24 9Ni-39Mn alloy with and without an applied magnetic causes an abrupt transformation from the pressure field(after Ref. D abilized B31 to the b8 ph

214 C.C. Koch / Materials Science and Engineering A287 (2000) 213–218 perature is useful in accounting for experimental data on the promotion of athermal martensitic transforma￾tions by magnetic fields. Experimental studies in the 1960s [2] observed small increases in the Ms tempera￾ture with the modest magnetic field used. Kakeshita et al. [3] used pulsed magnetic fields up to 31.75 MA m−1 to induce larger increases in Ms. The increase in Ms temperature, DT, versus magnetic field is shown in Fig. 1 for a Fe–31.7 at.% Ni alloy. All the martensites in this alloy were lenticular with a mid-rib and no differ￾ences in martensite morphology were noted between the thermally-induced and magnetically-induced marten￾sites. Subsequent studies of Kakeshita et al. [4] com￾pared the composition dependence of magnetic field-induced martensitic transformation in Fe–Ni al￾loys for Fe–29.9 at.% Ni, Fe–31.7 at.% Ni, and Fe– 32.5 at.% Ni. The slope of the temperature dependence of critical magnetic field strength to induce the marten￾site becomes smaller in the order of the 31.7, 29.9, and 32.5 at.% Ni alloys. That is, the magnetic field is most effective in inducing martensite in the 31.7 at.% Ni alloy. For constant DT, the amount of magnetic-field induced martensite is about constant with magnetic field strength for the 29.9 and 31.7 at.% Ni alloys but increases with field strength for the 32.5 at.% Ni alloy. This was found to be due to both the formation of new martensite plates and the growth of existing plates in the 32.5 at.% Ni alloy — that is, they are partly thermoelastic. The 29.9 and 31.7 at.% Ni martensites are nonthermoelastic. Kakeshita et al. [5] have recently studied the effects of magnetic fields on both the isothermal and athermal kinetics of martensite transformations in Fe–24.9 Ni– 39 Mn and Fe–31.4 Ni–0.5 Mn alloys, respectively. For the Fe–24.9Ni–3.9 Mn alloy, a pulsed magnetic field above a critical value induced martensite instanta￾neously. This suggests that the original isothermal ki￾netics of the martensitic transformation changed to athermal behavior under high magnetic fields. At static magnetic fields less than the critical value, the kinetics of the isothermal martensitic transformation were changed. The ‘nose’ of the TTT curve occurred at a lower temperature and shorter incubation time than that for no applied field as shown in Fig. 2. Models containing the effects of magneto static, high field susceptibility, and forced volume magnetostriction were able to explain the observed influence of magnetic field. Watanabe and Sato [6] showed that the martensitic transformation of 90-nm Fe particles in a Cu–1.5 wt% Fe alloy involving the change from the antiferromag￾netic to ferromagnetic state at 4.2 K was promoted by an applied magnetic field. The martensitic transforma￾tion was induced by tensile straining the samples at 4.2 K at a strain rate of 1.7×10−4 s−1 with or without an applied magnetic field of 5.6 T parallel to the tensile axis. Without the magnetic field 2991 vol.% of martensite (a-Fe) was formed compared to about 509 2 vol.% in the presence of the field. 2.1.2. Other examples of phase transformations affected by applied magnetic fields Galkin et al. [7] found that magnetic fields affected the pressure-induced phase transformation between the B8I and B31 phases of MnAs. The B8I phase, when the minor constituent, is superparamagnetic and the B31 phase goes through a paramagnetic to antiferromag￾netic transition with hydrostatic pressure near 3 katm. The application of a pulsed magnetic field of 100 kOe causes an abrupt transformation from the pressure stabilized B31 to the B8I phase. Fig. 1. Increase in Ms temperature, DT versus magnetic field (after Ref. [3]). Fig. 2. TTT diagrams of the isothermal martensitic transformation in an Fe–24.9Ni–3.9Mn alloy with and without an applied magnetic field (after Ref. [5])

CC. Koch/ Materials Science and Engineering 4287(2000)213- 21 Quench Medium throughout the specimen with consequent pronounced Silicone Oi Air145°c25°c ral oil results on the energy product of the magnet to make it a useful permanent magnet material [11] Another, more recent, example of magnetic field E=1 kV/cm nt is the atte the high Tc oxide superconductors to minimize the weak link' problem at grain boundaries to increase E=0 critical current density [12]. Fine, preferably, single crystal platelets of, e.g. YBa2Cu3O7_8, are dispersed in 4340 Steel a liquid medium such as heptane and a magnetic field E=1 kV/cm of 1-8 T applied. The anisotropy in the paramagnetic susceptibility of the high T superconducting oxides provides alignment with the c-axis parallel to the ap. 020406080100120140160 plied field direction. After the heptane is allowed to evaporate the compact is sintered in an ox atmo- Average Cooling Rate('C/s) sphere at 950-970oC and slowly cooled. A good (oon) Fig 3. Vickers hardness versus average cooling rate with and without texture is developed by this technique. Subsequently have been used which provide more complete alignment The tetragonal distortion of the A15 structure V3Si of the a and b as well as the c axes phase at low temperatures was predicted to change from a c/a ratio >1 to a c/a ratio I with the application of a strong magnetic field [8]. At 17. 4K the 3. Effects of electric fields field to accomplish this transformation was predicted to be 275 kOe and it is not known if this prediction was The application of an electric field to a variety of ever tested experimentally. phase transformations will be discussed in this section Cser et al. [9] used Mossbauer spectroscopy to study An electrical potential applied to an electrical conduc- the structural transformation between the B2 and DO3 magnetic field to be induced. The effects of electric phases in Fe3Al. They found that an applied magnetic field of 17 kOe stabilized a disordered metastable inter- currents and the induced magnetic fields on phase mediate phase that can appear during the transforma transformations will be covered by another paper in tion of B2 to DO3 this workshop. Only the effects of applied electric field Liu et al. [10] reported that a critical magnetic field without current flow will be dis (between 40 and 60 kOe) induced the tetragonal to monoclinic transformation in Zro, particles in 3. 1. Effects of electric fields on phase transformations MnZn-ferrite. The magnetic field did not directly affect in metallic materials the zro, particles, since they are paramagnetic, but created the stress-induced transformation due to the Cao et al. [13] studied the hardenability of two magnetostriction caused by the magnetic field in the commercial steels, namely 02(a high carbon steel)and nagnetic MnZn-ferrite matrix 4340(a medium carbon alloy steel) with and without an applied electric field of l kv cm. The steel specimens were made one electrode of an electrostatic circuit and 2. 2. Effect of magnetic field, H, on growth the field was applied during austenitizing an quenching at several cooling rates. The electric field produced an increase of hardness at intermediate cool- Besides the influence of magnetic fields on the free ing rates, with a larger effect for the 02 steel as illus- energy, and therefore stability, of given phases to bias trated in Fig 3. The influence of the electric field was the transformation in a given direction, magnetic fields much larger when applied only during the quench can also affect the morphology of the growing new compared to only during austenitizing. The effects are phase. The classic case for the influence of a magnetic rationalized by a change in the kinetics of the diffusion field on growth morphology is the Alnico magnet alloy. controlled transformations as reflected in the movement During the precipitation of the ferromagnetic phase of the pearlite and bainite'noses' of the TTt diagrams from the paramagnetic matrix, the presence of a mag- to longer times. This explained the changes in mi- netic field aligns the easy axis of these ferromagnetic crostructure and hardness for intermediate cooling rates precipitates along the field direction. In this way a which intersect the pearlite and bainite regions in the strongly preferred uniaxial anisotropy can be produced Ttt diagram for the 02 steel but have a smaller effect

C.C. Koch / Materials Science and Engineering A287 (2000) 213– 215 Fig. 3. Vickers hardness versus average cooling rate with and without an applied electric field for 02 steel and 4340 steel (after Ref. [13]). throughout the specimen with consequent pronounced results on the energy product of the magnet to make it a useful permanent magnet material [11]. Another, more recent, example of magnetic field alignment is the attempt to develop strong texture in the high Tc oxide superconductors to minimize the ‘weak link’ problem at grain boundaries to increase critical current density [12]. Fine, preferably, single￾crystal platelets of, e.g. YBa2Cu3O7−d, are dispersed in a liquid medium such as heptane and a magnetic field of 1–8 T applied. The anisotropy in the paramagnetic susceptibility of the high Tc superconducting oxides provides alignment with the c-axis parallel to the ap￾plied field direction. After the heptane is allowed to evaporate the compact is sintered in an oxygen atmo￾sphere at 950–970°C and slowly cooled. A good (001) texture is developed by this technique. Subsequently, better methods of grain alignment/texture development have been used which provide more complete alignment of the a and b as well as the c axes. 3. Effects of electric fields The application of an electric field to a variety of phase transformations will be discussed in this section. An electrical potential applied to an electrical conduc￾tor will cause current to flow in the conductor and a magnetic field to be induced. The effects of electric currents and the induced magnetic fields on phase transformations will be covered by another paper in this workshop. Only the effects of applied electric fields without current flow will be discussed here. 3.1. Effects of electric fields on phase transformations in metallic materials Cao et al. [13] studied the hardenability of two commercial steels, namely 02 (a high carbon steel) and 4340 (a medium carbon alloy steel) with and without an applied electric field of 1 kV cm−1 . The steel specimens were made one electrode of an electrostatic circuit and the field was applied during austenitizing and/or quenching at several cooling rates. The electric field produced an increase of hardness at intermediate cool￾ing rates, with a larger effect for the 02 steel as illus￾trated in Fig. 3. The influence of the electric field was much larger when applied only during the quench compared to only during austenitizing. The effects are rationalized by a change in the kinetics of the diffusion￾controlled transformations as reflected in the movement of the pearlite and bainite ‘noses’ of the TTT diagrams to longer times. This explained the changes in mi￾crostructure and hardness for intermediate cooling rates which intersect the pearlite and bainite regions in the TTT diagram for the 02 steel but have a smaller effect The tetragonal distortion of the A15 structure V3Si phase at low temperatures was predicted to change from a c/a ratio \1 to a c/a ratio B1 with the application of a strong magnetic field [8]. At 17.4 K the field to accomplish this transformation was predicted to be 275 kOe and it is not known if this prediction was ever tested experimentally. Cser et al. [9] used Mo¨ssbauer spectroscopy to study the structural transformation between the B2 and DO3 phases in Fe3Al. They found that an applied magnetic field of 17 kOe stabilized a disordered metastable inter￾mediate phase that can appear during the transforma￾tion of B2 to DO3. Liu et al. [10] reported that a critical magnetic field (between 40 and 60 kOe) induced the tetragonal to monoclinic transformation in ZrO2 particles in a MnZn-ferrite. The magnetic field did not directly affect the ZrO2 particles, since they are paramagnetic, but created the stress-induced transformation due to the magnetostriction caused by the magnetic field in the magnetic MnZn-ferrite matrix. 2.2. Effect of magnetic field, H, on growth morphologies Besides the influence of magnetic fields on the free energy, and therefore stability, of given phases to bias the transformation in a given direction, magnetic fields can also affect the morphology of the growing new phase. The classic case for the influence of a magnetic field on growth morphology is the Alnico magnet alloy. During the precipitation of the ferromagnetic phase from the paramagnetic matrix, the presence of a mag￾netic field aligns the easy axis of these ferromagnetic precipitates along the field direction. In this way a strongly preferred uniaxial anisotropy can be produced

216 C C. Koch/ Materials Science and Engineering 4287(2000)213-218 n the 4340 steel as shown in Fig. 4. These shifts are ransformation of austenite to either pearlite or banite believed due to the electric field reducing the diffusive- It was suggested that the electric field might reduce the ties of the carbon and or substitutional solutes for the diffusivity by reducing the concentration of vacancies at the surface of the specimen, or, if the field can exist throughout the specimen, reduce the vacancy concen- Silicon Oil(25°C) tration locally in the sample interior, or reduce the E=1 kV/cm carbon mobility Conrad et al. [14] studied the influence of an electric field on the quench aging of a low carbon steel. The electric field caused a reduction in hardness and a higher activation required to attain maximum hardness TEM observations showed that the electric field pro- duced an order of magnitude increase in the width of E=0WHN=420 kVIcm VHN=860 the precipitate-depleted zone at grain boundaries as 1010 102 10 101010 illustrated in Fig. 5. This was concluded to be the eason for the reduction in hardness since the depleted Time(s) zone is softer than the precipitate-containing region To explain the increased width of the depleted zone Fig. 4. Estima diagram without and with an electric field of I kv cm- fc eel(after Ref. [13]). produced by the electric field it was suggested that the electric charge at the surface reduces the chemical po- tential of vacancies there. The increase in this chemical potential results in an increase in the gradient in chem ical potential between the surface and the grain boundary with an increase in the flux of vacancies along the grain boundaries to the surface. This then leads to an increase in flux from the grain interior to he boundary which results in the width of the depleted zone. That is, the nucleation sites for carbide preci tates in a-Fe are individual or clusters of vacancies. If vacancies migrate to the grain boundaries the resulting deficiency of vacancies adjacent to the boundaries give rise to the depleted zones since there would be a lack of nucleation sites for the carbide precipitates. This would predict a smaller effect of the electric field on quench aging as the grain size increases and this is experimen ).5 observe Liu and coworkers [15-18] have studied the influence of an applied electric field on the phase transformation in an Al-Li alloy. They conclude that the electric field has two effects on phase transformations in Al-L alloys. First, the electric field accelerates the dissolution of an unstable phase during solution treatment. The homogenization of cast Al-Li alloys [17] in an electric field accelerated the dissolution of undissolved second- phase particles as well as eliminating dendritic segrega- tion. This enhanced the workability of the alloy Second, an electric field apparently retards the formation of a precipitate during aging treatments [18] The application of an electric field was found to sup- press the nucleation of the 8 phase during aging of 2091 Al-Li alloy. This was due to an increase in the activation energy for the aging treatment. The aging process was retarded and the peak yield strength was lower. The 8 precipitates were larger and Fig. 5. TEM micrographs of the precipitate ging at56°Cfor8Oh. fewer in number density for the samples aged in an (a)v=0;(b)V=7 kv(after Ref [14]) electric field

216 C.C. Koch / Materials Science and Engineering A287 (2000) 213–218 in the 4340 steel as shown in Fig. 4. These shifts are believed due to the electric field reducing the diffusivi￾ties of the carbon and/or substitutional solutes for the transformation of austenite to either pearlite or banite. It was suggested that the electric field might reduce the diffusivity by reducing the concentration of vacancies at the surface of the specimen, or, if the field can exist throughout the specimen, reduce the vacancy concen￾tration locally in the sample interior, or reduce the carbon mobility. Conrad et al. [14] studied the influence of an electric field on the quench aging of a low carbon steel. The electric field caused a reduction in hardness and a higher activation required to attain maximum hardness. TEM observations showed that the electric field pro￾duced an order of magnitude increase in the width of the precipitate-depleted zone at grain boundaries as illustrated in Fig. 5. This was concluded to be the reason for the reduction in hardness since the depleted zone is softer than the precipitate-containing regions. To explain the increased width of the depleted zone produced by the electric field it was suggested that the electric charge at the surface reduces the chemical po￾tential of vacancies there. The increase in this chemical potential results in an increase in the gradient in chem￾ical potential between the surface and the grain boundary with an increase in the flux of vacancies along the grain boundaries to the surface. This then leads to an increase in flux from the grain interior to the boundary which results in the width of the depleted zone. That is, the nucleation sites for carbide precipi￾tates in a-Fe are individual or clusters of vacancies. If vacancies migrate to the grain boundaries the resulting deficiency of vacancies adjacent to the boundaries give rise to the depleted zones since there would be a lack of nucleation sites for the carbide precipitates. This would predict a smaller effect of the electric field on quench aging as the grain size increases and this is experimen￾tally observed. Liu and coworkers [15–18] have studied the influence of an applied electric field on the phase transformation in an Al–Li alloy. They conclude that the electric field has two effects on phase transformations in Al–Li alloys. First, the electric field accelerates the dissolution of an unstable phase during solution treatment. The homogenization of cast Al–Li alloys [17] in an electric field accelerated the dissolution of undissolved second￾phase particles as well as eliminating dendritic segrega￾tion. This enhanced the workability of the alloy. Second, an electric field apparently retards the formation of a precipitate during aging treatments [18]. The application of an electric field was found to sup￾press the nucleation of the d% phase during aging of 2091 Al–Li alloy. This was due to an increase in the activation energy for the aging treatment. The aging process was retarded and the peak yield strength was lower. The d% precipitates were larger and fewer in number density for the samples aged in an electric field. Fig. 4. Estimated TTT diagram without and with an electric field of 1 kV cm−1 for 02 steel (after Ref. [13]). Fig. 5. TEM micrographs of the precipitate for aging at 56°C for 80h. (a) V=0; (b) V=7 kV (after Ref. [14])

C C. Koch/ Materials Science and Engineering 4287(2000)213-218 217 found to be a maximum of 0.78% and was attributed PZST 43/9/2 to the elastic softening associated with the transforma- 2°cmin,1kHz tion Electric fields have also been found to influence AFE phase transformations in ferroelectric polymers. Koga et al. [23] studied the electric field-induced phase trans- E110 formation in the P(vinylidene fluoride (VDF)-trifl- uoroethylene (TrFE)) copolymers. The structures of these copolymers depend on their composition. For VPF TrFE(-x with molar ratios in the range of 082<x<0.9 the copolymers crystallize from the melt FE into a mixture of the a-, B, and possibly y-phases. The a-phase consists of antiparallel chains and is a nono- lar, nonpiezoelectric crystal. The B-phase is a ferroelectric phase consisting of all-trans chains. Appli cation of a strong AC electric field transforms the mixed phase (o, B, y) crystals completely into the Bias Field kV/cm B-phase. The B-phase crystals are ferroelectric and ex hibit strong piezoelectric activity stable up to the melt- Fig. 6. Effect of DC electric field on the transformation temperatures Ing point. for the AFE to FE transformation in PZST (after Ref [22D 3.4. Electric field induced phase separation in oxide 3. 2. Effects of electric fields on phase transformations in liquid-crystalline polymers The enhanced phase separation kinetics due to The application of a sufficiently strong electric field applied electric field(1-2 kV, either AC(50 Hz)or DC) (or magnetic field) can induce a cholesteric-to-nematic was first reported for oxide glasses by de vekey and phase transformation in liquid crystalline polymers Majumdar [24]. The system studied was CaO-AlO [19-21]. The helices in the cholesteric phase of some Sio,-MgO-TiO,. The normal nucleation temperature liquid-crystal films can untwist into a homeotropic range for phase separation for this glass is 725-800oC (perpendicular) nematic configuration. The details of Glasses heated at 690 C showed little evidence for this and similar transformations affected by electric and phase separation without an electric field, but with the magnetic fields will be described in the paper by r.J. electric field, well formed nuclei of the phase separation Spontak in this worksho were observed. Heating in the normal temperature range for phase separation in the presence of an electric 3.3. Effects of electric fields on phase transformations field resulted in a coarser microstructure than without fe the field. That is, the growth of the phase separated fast field Yang and Payne studied the effect of an applied It was suggested that both these effects of electric field electric field on the antiferroelectric(aFe) to ferroelec may be attributed to the enhanced diffusivity the field tric (FE)phase transformation in tin-modified lead provides zirconate titanate [Pb(zr, Sn, Ti)O3, i.e. PZST[22]. The Subsequently, Liu et al. [25] systematically studied experimental results showed that the application of an the effect of an electric field on the phase separation of electric field tends to extend the ferroelectric phase glasses. They studied two systems: (1)Cao-Al2O3 region. This is illustrated in Fig. 6 where it is shown SiO2( CAS) and (2)CaO-B2O3-P2Os( CBP). They that an increase in the electric field strength increases found that increasing the electric field strength greatly the ferroelectric(FE)-to-antiferroelectric(AFE) trans- increased the effect of the electric field on phase separa formation temperature on both heating(TA)and cool- tion. In the CAs system, the electric field treatment g (TE). This was explained by the effect of the electric promotes the phase separation. However, in the CBP eld promoting the long-range interactions which re- glass system phase separation was inhibited by the duces the sublattice coupling and results in a field-in- electric field. A model was developed to explain the duced lattice softening. The AFE structure effect of the electric field on the phase separation of progressively becomes unstable and the effective force glasses. The free energy change for formation of a constant for the soft phonon mode decreases, resulting cluster included a term involving the electric field. That in lattice softening, and the transformation to the Fe is, the free energy of formation for a nucleus of critical phase. The electrically induced AFE-to-FE strain was size was given as

C.C. Koch / Materials Science and Engineering A287 (2000) 213–218 217 Fig. 6. Effect of DC electric field on the transformation temperatures for the AFE to FE transformation in PZST (after Ref. [22]). found to be a maximum of 0.78% and was attributed to the elastic softening associated with the transforma￾tion. Electric fields have also been found to influence phase transformations in ferroelectric polymers. Koga et al. [23] studied the electric field-induced phase trans￾formation in the P (vinylidene fluoride (VDF)-trifl- uoroethylene (TrFE)) copolymers. The structures of these copolymers depend on their composition. For VPFxTrFE(1−x) with molar ratios in the range of 0.825x50.9 the copolymers crystallize from the melt into a mixture of the a-, b-, and possibly g-phases. The a-phase consists of antiparallel chains and is a nonpo￾lar, nonpiezoelectric crystal. The b-phase is a ferroelectric phase consisting of all-trans chains. Appli￾cation of a strong AC electric field transforms the mixed phase (a, b, g) crystals completely into the b-phase. The b-phase crystals are ferroelectric and ex￾hibit strong piezoelectric activity stable up to the melt￾ing point. 3.4. Electric field induced phase separation in oxide glasses The enhanced phase separation kinetics due to an applied electric field (1–2 kV, either AC (50 Hz) or DC) was first reported for oxide glasses by deVekey and Majumdar [24]. The system studied was CaO–Al2O3 – SiO2 –MgO–TiO2. The normal nucleation temperature range for phase separation for this glass is 725–800°C. Glasses heated at 690°C showed little evidence for phase separation without an electric field, but with the electric field, well formed nuclei of the phase separation were observed. Heating in the normal temperature range for phase separation in the presence of an electric field resulted in a coarser microstructure than without the field. That is, the growth of the phase separated regions was faster with an applied electric field. It was suggested that both these effects of electric field may be attributed to the enhanced diffusivity the field provides. Subsequently, Liu et al. [25] systematically studied the effect of an electric field on the phase separation of glasses. They studied two systems: (1) CaO–Al2O3 – SiO2 (CAS) and (2) CaO–B2O3 –P2O5 (CBP). They found that increasing the electric field strength greatly increased the effect of the electric field on phase separa￾tion. In the CAS system, the electric field treatment promotes the phase separation. However, in the CBP glass system phase separation was inhibited by the electric field. A model was developed to explain the effect of the electric field on the phase separation of glasses. The free energy change for formation of a cluster included a term involving the electric field. That is, the free energy of formation for a nucleus of critical size was given as: 3.2. Effects of electric fields on phase transformations in liquid-crystalline polymers The application of a sufficiently strong electric field (or magnetic field) can induce a cholesteric-to-nematic phase transformation in liquid crystalline polymers [19–21]. The helices in the cholesteric phase of some liquid–crystal films can untwist into a homeotropic (perpendicular) nematic configuration. The details of this and similar transformations affected by electric and magnetic fields will be described in the paper by R.J. Spontak in this workshop. 3.3. Effects of electric fields on phase transformations in ferroelectric materials Yang and Payne studied the effect of an applied electric field on the antiferroelectric (AFE) to ferroelec￾tric (FE) phase transformation in tin-modified lead zirconate titanate [Pb(Zr,Sn,Ti)O3, i.e. PZST] [22]. The experimental results showed that the application of an electric field tends to extend the ferroelectric phase region. This is illustrated in Fig. 6 where it is shown that an increase in the electric field strength increases the ferroelectric (FE)-to-antiferroelectric (AFE) trans￾formation temperature on both heating (TA) and cool￾ing (TF). This was explained by the effect of the electric field promoting the long-range interactions which re￾duces the sublattice coupling and results in a field-in￾duced lattice softening. The AFE structure progressively becomes unstable and the effective force constant for the soft phonon mode decreases, resulting in lattice softening, and the transformation to the FE phase. The electrically induced AFE-to-FE strain was

21 C C. Koch/ Materials Science and Engineering 4287(2000 )213-218 △G。 References Metall.8(1964)502. where o, interfacial energy; AGv, driving free energy [M.A. Krivoglaz, V.D. Sadovsky, 22V.D. Sadovsky, L.V. Smirnov, kina. P.A. Malinen. I.P. E,, electric field, and &n and s are the dielectric Soroskin, Fizika Metal Metall. 24( 502 constants of the matrix and precipitating phase, 3 T. Kakeshita, K. Shimizu, T. Sakakibara, S. Funada, M. Date, respectively. If E2>E1, AGe is reduced and nucleus Scripta Metall. 17(1983)987 [4 T. Kakeshita, K. Shimizu, S. Funada, M. Date, Acta Metall. 33 formation or phase separation of a glass is pro- noted by the applied electric field. However, 5T.Kakeshita, K. Shimizu, T. Saburi, in: W.C. Johnson, JM if E2 <E, phase separation is inhibited. This model was Howe, D.E. Laughlin, w.A. Soffa (Eds ) Solid- Solid Phase consistent with the experiments on CAs and CBP Transformations, TMS. Warrendale, PA. 1994. pp. 817-822. glass 6 Y. Watanabe, A. Sato, Scripta Metall. 23(1989)359 [7 A.A. Galkin, E.A. Zavadskii, Y.I. Yalkov, Phys. Stat Sol.(b)46 (1971)K23 [8C.S. Ting, A.K. Ganguly, J. L. Birman, Phys. Rev. Lett. 30 (1973)1245 It is clear from the above examples that the effects of (10)Y.-C.Liu,DGau,P.Shen,J either magnetic or electric fields can have a signifi F.E. Luborsky, J D. Livingston, G.Y. Chin, in: R.w. Cahn, influence on phase transformations in different ways for Haasen (Eds ) Physical Metallurgy, Part Il, North-Holland, Amsterdam, 1983, p. 1689. a variety of materials. Since phase transformations [12]S Jin, JOM, March 1991,p. 7 determine the structure and often the microstructure, [13] w.D. Cao, X P. Lu, A.F. Sprecher, H. Conrad, Mater. Lett. 9 magnetic or electric fields are additional tools to control (1990)193 the structure to obtain the desired structure/property [14 H. Conrad, Y. Chen, H.A. Lu, The influence of an electric relationships and therefore material performance. Ex charge on the quench aging of a low carbon steel, in: P K. liav amples include the well established magnetic field align R. Viswanathan, K L. Murty, E.P. Simoson, D. Freas(Eds ) ment of precipitates to optimize the hard Microstructures and Mechanical Properties of Aging Material. ferromagnetic TMS rendall, PA, 1993, P. 279 properties of AINiCo magnets to very recent sugges- [15] w. Liu, K.M. Liang. Y K. Zheng, J.Z. Cui,J. Mater. Sci. Lett tions to tailor-make the microstructure of glass-ceram ics with applied electric fields. Since the attention given (17 W. Liu, K.M. Liang. Y.K. Zhong. .Z. Cui, J. Mater. Sci. Lett 15(1996)1918 for research remain [18 w. Liu, J.Z. Cui, J. Mater. Sci. Lett. 16 [19 J. Wysocki, J. Adams, w. Haas, Mol CI iquid Crystals 8 (1969)471 20C.M. Chen, F.C. MacKintosh, Eure t.30(1995)21 Acknowledgements 21]C G. Liu-Hendel, J. Appl. Phys. 53 22 P. Yang, D.A. Payne, J. Appl. Phys The author wishes to thank Professor Hans Conrad [23]KKoga, N Nakano, T. Hattori, H Ohigashi, J Appl. Phys. 67 suggesting this topic to review and Professor Con- (1990)965 rad and Dr Y. Fahmy for assistance in obtaining the 24R.C de Vekey, A.J. Majumdar, Nature 225(1970)172. 25w.Liu, K M. Liang, Y.K. Zheng, S.R. Gu, H. Chen, J. Phys. D relevant literature Appl.Phys30(1997)356

218 C.C. Koch / Materials Science and Engineering A287 (2000) 213–218 DGc= 16ps3 3[DGv+1 2E2 2 (o2−o1)]2 where s, interfacial energy; DGv, driving free energy; E2, electric field, and o1 and o2 are the dielectric constants of the matrix and precipitating phase, respectively. If o2\o1, DGc is reduced and nucleus formation or phase separation of a glass is pro￾moted by the applied electric field. However, if o2Bo1, phase separation is inhibited. This model was consistent with the experiments on CAS and CBP glasses. 4. Summary It is clear from the above examples that the effects of either magnetic or electric fields can have a significant influence on phase transformations in different ways for a variety of materials. Since phase transformations determine the structure and often the microstructure, magnetic or electric fields are additional tools to control the structure to obtain the desired structure/property relationships and therefore material performance. Ex￾amples include the well established magnetic field align￾ment of precipitates to optimize the hard ferromagnetic properties of AlNiCo magnets to very recent sugges￾tions to tailor-make the microstructure of glass-ceram￾ics with applied electric fields. Since the attention given to this topic has been limited, many new possibilities for research remain. Acknowledgements The author wishes to thank Professor Hans Conrad for suggesting this topic to review and Professor Con￾rad and Dr Y. Fahmy for assistance in obtaining the relevant literature. References [1] M.A. Krivoglaz, V.D. Sadovsky, Fizika Metall. 18 (1964) 502. [2] V.D. Sadovsky, L.V. Smirnov, Y.A. Fokina, P.A. Malinen, I.P. Soroskin, Fizika Metal Metall. 24 (1964) 502. [3] T. Kakeshita, K. Shimizu, T. Sakakibara, S. Funada, M. Date, Scripta Metall. 17 (1983) 987. [4] T. Kakeshita, K. Shimizu, S. Funada, M. Date, Acta Metall. 33 (1985) 1381. [5] T. Kakeshita, K. Shimizu, T. Saburi, in: W.C. Johnson, J.M. Howe, D.E. Laughlin, W.A. Soffa (Eds.), SolidSolid Phase Transformations, TMS, Warrendale, PA, 1994, pp. 817–822. [6] Y. Watanabe, A. Sato, Scripta Metall. 23 (1989) 359. [7] A.A. Galkin, E.A. Zavadskii, Y.I. Yalkov, Phys. Stat. Sol. (b) 46 (1971) K23. [8] C.S. Ting, A.K. Ganguly, J.L. Birman, Phys. Rev. Lett. 30 (1973) 1245. [9] L. Cser, Y.M. Ostanevich, L. Pal, Phys. Stat. Sol. 42 (1970) K147. [10] Y.-C. Liu, D. Gau, P. Shen, J. Am. Ceram. Soc. 79 (1996) 559. [11] F.E. Luborsky, J.D. Livingston, G.Y. Chin, in: R.W. Cahn, P. Haasen (Eds.), Physical Metallurgy, Part II, North-Holland, Amsterdam, 1983, p. 1689. [12] S. Jin, JOM, March 1991, p. 7. [13] W.D. Cao, X.P. Lu, A.F. Sprecher, H. Conrad, Mater. Lett. 9 (1990) 193. [14] H. Conrad, Y. Chen, H.A. Lu, The influence of an electric charge on the quench aging of a low carbon steel, in: P.K. Liaw, R. Viswanathan, K.L. Murty, E.P. Simoson, D. Freas (Eds.), Microstructures and Mechanical Properties of Aging Material, TMS, Warrendale, PA, 1993, p. 279. [15] W. Liu, K.M. Liang, Y.K. Zheng, J.Z. Cui, J. Mater. Sci. Lett. 15 (1996) 1327. [16] W. Liu, J.Z. Cui, Scripta Metall. Mater. 33 (1995) 623. [17] W. Liu, K.M. Liang, Y.K. Zhong, J.Z. Cui, J. Mater. Sci. Lett. 15 (1996) 1918. [18] W. Liu, J.Z. Cui, J. Mater. Sci. Lett. 16 (1997) 1410. [19] J. Wysocki, J. Adams, W. Haas, Mol. Crystals Liquid Crystals 8 (1969) 471. [20] C.-M. Chen, F.C. MacKintosh, Europhys. Lett. 30 (1995) 215. [21] C.G. Liu-Hendel, J. Appl. Phys. 53 (1982) 916. [22] P. Yang, D.A. Payne, J. Appl. Phys. 80 (1996) 4001. [23] K. Koga, N. Nakano, T. Hattori, H. Ohigashi, J. Appl. Phys. 67 (1990) 965. [24] R.C. deVekey, A.J. Majumdar, Nature 225 (1970) 172. [25] W. Liu, K.M. Liang, Y.K. Zheng, S.R. Gu, H. Chen, J. Phys. D Appl. Phys 30 (1997) 3366. .