Martensitic transformation,crystal structure and shape memory effect in Ni 55− x Mn 25 Ga 20 Co x alloys

Abstract

The effect of cobalt addition instead of nickel on the crystal structure, martensitic transformation behaviour and shape memory effect were investigated. Within the analysed range of chemical composition, a single non-modulated martensite was detected at ambient temperature. Cobalt addition modified the lattice parameters and, thus, affected the tetragonality of the martensite unit cell. The hysteresis of martensitic transformation was differently affected by the type of heat treatment applied. For furnace cooled samples, the hysteresis decreased from 50°C to 30°C; in the case of water quenched samples, the hysteresis sharply increased up to 60°C. The shape memory effect, measured as the recoverable strain upon annealing after compression tests, reached a fully recoverable deformation at 10 at.-% of cobalt.

This paper is part of a Thematic Issue on The Crystallographic Aspects of Metallic Alloys.

Keywords

Shape memory effect martensitic transformation Heusler alloys crystal structure

Introduction

The ternary Ni–Mn–Ga Heusler alloys attracted considerable attention due to their functional properties such as magnetic-field-induced strain (MFIS) effect [1 -5]. Materials which exhibit this effect could find practical use in actuators and sensor devices [6 -10]. The MFIS may be observed in the ferromagnetic state at temperatures below the martensitic transformation (M_T). The mechanism of MFIS is based on the rearrangement of crystallographic domains (twin variants) under an external magnetic field of about 1 T which varies depending on the twinning stress and the magneto-crystalline anisotropy of different crystal structures [11 -15].

The stoichiometric Ni₂MnGa alloy at room temperature possesses cubic L2₁ crystal structure and upon cooling undergoes reversible martensitic transformation from a parent phase to a low-temperature martensite phase. Three main types of martensite phases have been discovered in Ni–Mn–Ga alloys, i.e. 5-layer modulated (10M) and 7-layer modulated (14M) monoclinic structures and a non-modulated (2M) tetragonal structure. The ‘a’ and ‘c’ lattice parameters of the martensitic structure define the maximal theoretical MFIS described as 1−c/a. These maximum values of longitudinal strain induced by the external magnetic field were already reported in 10M and 14M martensite single crystals which equal about 7% and 11%, respectively [16,17]. However, for the 2M martensite of ternary Ni–Mn–Ga alloy, the MFIS effect has not been reported so far which is a consequence of a relatively large twinning stress (about 8–10 MPa) and insufficient magneto-crystalline anisotropy [18 -23]. The twinning stress may be reduced by modifying the unit cell in order to decrease its tetragonality (c/a) [24]. This effect was achieved by replacing Ni by Co [25]. Apart from the change of the unit cell parameters, the addition of other elements affects the martensitic transformation temperature which is very sensitive to the chemical composition of the parent phase [26 -29]. In contrast, the Curie temperature is not so much composition dependent [30]. Thus, by composition adjustment one may control the main functional properties and other factors important for industrial applications of Ni–Mn–Ga-based Heusler alloys. Ageing of Ni–Co–Mn–Ga alloys, as potential high temperature shape memory materials, has increased the Curie temperature. On the other hand, only a modest increase of the martensitic transformation temperature takes place after longer ageing times of ∼10⁴min which was attributed to defect annihilation or to an early stage of precipitation [31].

In this paper, the effect of Co addition on the temperature of martensite transformation, structure and shape memory effect of Ni_55−xMn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys was examined. Also, the influence of water quenching (WQ) and thus atomic ordering of the structure was investigated upon Co addition.

Materials and methods

Polycrystalline alloys of nominal composition Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) were prepared by repeated melting of high purity constituents (Ni – 99.9%, Mn – 99.95%, Ga – 99.99% and Co – 99.95%) in an arc furnace under argon atmosphere. Additional 5 wt-% of Mn content was added to compensate for evaporation losses. Subsequently, the buttons were sealed into quartz ampoules at vacuum condition and further annealed at 900°C for 48 h in order to obtain good chemical homogeneity and then were furnace cooled (FC) to ambient temperature. After homogenisation buttons were cut into two pieces. The second part was once more encapsulated in quartz ampoules and then annealed at 900°C for 30 min followed by WQ. The morphology of the samples was observed using a scanning electron microscope (SEM) in a backscattered electron mode. For SEM observations samples were electro-polished using electrolyte of 20% perchloric acid and 80% methanol and then etched in a solution of 60 ml HNO₃ + 20 ml HCl + 10 ml H₂O₂. The composition of the alloys was checked by an energy-dispersive spectrometer (EDS) attached to SEM. Also microstructure and crystal structure were examined by Tecnai G2 Transmission Electron Microscope (TEM). Thin foil samples for TEM observations were prepared with TenuPol-5 double jet electropolisher using an electrolyte of nitric acid (1/3) and methanol (2/3) at a temperature of about −28 °C. The substitution of Co instead of Ni decreases the valence electron concentration per atom ratio (e/a) which affects the martensitic transformation temperature, lattice parameters, etc. (Table 1). The crystal structure at ambient temperature was identified by X-ray diffraction (XRD) analysis using CoK _α radiation. The phase transformation temperatures were determined by differential scanning calorimetry (DSC) in the temperature range of 150–400°C with a heating and cooling rate of 10°C min⁻¹. Room temperature uni-axial compression tests with a strain rate of 10⁻³s⁻¹ were performed. The samples were of rectangular prisms with the height to width ratio of 3/2. After compression tests, the shape memory effect measured by a recoverable strain is obtained after annealing at 500°C for 15 min (i.e. above the finish temperature of the reverse transformation of martensite to the parent phase during heating, A_f).

Table 1

The chemical composition, e/a ratio and lattice parameters of Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x alloys.

						Furnace cooled			Water quenched
Alloy	Ni (at.-%)	Mn (at.-%)	Ga (at.-%)	Co (at.-%)	e/a	a (Å)	c (Å)	c/a	a (Å)	c (Å)	c/a
Co0	53.4 ± 1.1	27.3 ± 0.8	19.3 ± 0.8	–	7.83	5.40	6.69	1.24	5.40	6.69	1.24
Co5	48.7 ± 1.0	27.2 ± 0.8	19.6 ± 0.8	4.5 ± 0.3	7.77	5.43	6.61	1.22	5.43	6.62	1.22
Co10	44.2 ± 0.9	27.3 ± 0.8	19.5 ± 0.8	9.0 ± 0.4	7.72	5.44	6.58	1.21	5.43	6.59	1.21

Results and discussion

The SEM microstructural observations revealed a typical martensitic morphology of self-accommodated lamellar microstructure in all the investigated samples (Figure 1). The martensitic plates were of few micrometers in thickness with visible branching that occurs due to necessary shape accommodation taking place during martensitic transformation. The TEM investigations showed that individual martensitic plates were internally twinned and thus, a so-called twin within twin microstructure was created which is typically observed in Heusler alloys [32] (Figure 2(a)). The dark field TEM image revealed that the mean twin thickness spread throughout each plate was about 10 nm (Figure 2(b)).

Figure 1

SEM micrographs revealing platelike microstructure of FC Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x alloys: (a) x = 0, (b) x = 5, (c) x = 10.

Figure 2

Selected area diffraction pattern (a) with a corresponding dark field image and (b) of Ni₅₀Mn₂₅Ga₂₀Co₅ FC sample.

The EDS data of the chemical composition of the alloys are collected in Table 1 together with the calculated e/a ratio. To calculate the e/a ratio the number of valence electrons of Ni, Mn, Ga and Co were taken as 10, 7, 3 and 9, respectively. Thus, the substitution of Ni by Co in the Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x alloy decreases the e/a ratio from 7.83 down to 7.72 for Co0 and Co10 alloys, respectively.

The XRD profiles recorded at ambient temperature revealed that the investigated alloys were single phase of 2M martensite structure (Figure 3). The appearance of this type of martensite phase is independent of the amount of Co addition and the type of heat treatment applied (i.e. FC and WQ samples). Although the type of martensite was the same in each case, one can observe changes in the lattice parameters. The addition of Co modifies the lattice parameters of both the FC and WQ samples. The ‘a’ and ‘c’ lattice parameters change from 5.40 to 5.43 Å and from 6.69 to 6.58 Å, respectively. The tetragonality of the martensite unit cell then changes accordingly from 1.24 to 1.21.

Figure 3

XRD patterns of polycrystalline arc melted Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys. All of the reflection were indexed according to the 2M martensite structure.

The DSC curves of the Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys are shown in Figure 4. All of the samples possess a reversible martensitic transformation. The exothermic and endothermic peaks during full cycle of heating and subsequent cooling correspond to the reverse and forward martensitic transformation, respectively. The characteristic transformation temperatures (i.e. the start (M_s) and finish (M_f) temperatures of the forward transformation of the parent phase to the martensite phase during cooling and the start (A_s) and finish (A_f) temperatures of the reverse transformation of martensite to the parent phase during heating) were estimated by the two-tangent method (see the DSC curve of sample Co0 WQ in Figure 4). One can notice that generally the M_s is proportional to the e/a ratio (see Figure 5 and Table 1). For the FC samples, the M_s drops from 291°C to 213°C that corresponds to a decrease of e/a from 7.83 to 7.72, respectively. The applied WQ procedure affects the A_s increasing the reverse transformation temperature. Also, the hysteresis of martensitic transformation (ΔT) was significantly affected by the heat treatments. The hysteresis was defined as the temperature difference between the peak maximum of the forward and reverse martensitic transformations observed during DSC cooling/heating cycles. For the FC samples, the hysteresis narrows upon Co addition from 50°C to 28°C. Interestingly, the ΔT of WQ samples, turned out to be inversely proportional to the e/a ratio which has increased from 49°C to 63°C for Co0 WQ and Co10 WQ samples, respectively. This effect is a consequence of the strong influence of WQ on the A_s and A_f temperatures increasing them sharply in each case (Figure 4).

Figure 4

DSC curves of Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) that show the forward (on cooling) and reverse (heating) martensitic transformation.

Figure 5

Dependence of the martensitic start temperature M_s and thermal hysteresis ΔT on valence electron concentration per atom e/a in Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys.

From the uni-axial stress–strain compression characteristics, one can see that the maximum strain (ε_max), which was measured after unloading, decreases upon the Co addition from 7.17% to 6.68% and from 8.09% to 5.12% for FC and WQ samples, respectively (Table 2 and Figure 6). Although the ε_max has somewhat decreased, the recoverable strain (ε_SME) measured after annealing at 500°C (i.e. the shape memory effect) increases from 6.01% to 6.68% for Co0 and Co10, respectively. Interestingly, a fully recoverable shape memory effect was obtained for the FC and WQ samples that contained 10 at.-% of Co only. One can also see that the twinning stress (σ_T), measured at the half of ε_max, also decreases upon Co addition (Table 2).

Figure 6

Compressive stress vs. strain curves of samples Co0 and Co10 after both types of heat treatment. Note the difference of recoverable strain ε_SME upon Co addition.

Table 2

Mechanical properties of Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys measured from the uni-axial compression stress–strain curves performed at ambient temperature.

	Furnace cooled			Water quenched
Alloy	σ_T (MPa)	ε_max (%)	ε_SME (%)	σ_T (MPa)	ε_max (%)	ε_SME (%)
Co0	178	7.17	6.01	169	8.09	6.60
Co5	163	7.06	6.30	112	7.08	5.79
Co10	142	6.68	6.68	125	5.12	5.12

Conclusions

In this work, the type of crystal structure, martensitic transformation behaviour and compressive properties of polycrystalline Ni₅₅₋ _x Mn₂₅Ga₂₀Co _x (x = 0, 5, 10 at.-%) alloys after two types of heat treatments were studied. Based on the experimental results, the following conclusion can be drawn:

The substitution of Ni for Co does not change the type of martensite structure which turned out to be the 2M non-modulated phase. The Co addition, however, modifies the lattice parameters that led to a change of the tetragonality of the unit cell from 1.24 to 1.21 for Co0 and Co10 alloys, respectively.

The martensitic transformation temperature decreased when Ni was replaced by Co which was proportional to the change of the e/a ratio.

The hysteresis of martensitic transformation ΔT has a different composition dependence with respect to the type of the heat treatment. For FC samples, the ΔT narrows considerably from 50°C to 28°C upon Co addition. On the other hand, in the case of WQ samples, an increase of ΔT occurs that reached ∼63°°C for the Co10 sample.

The recoverable strain of FC samples induced by annealing at 500°C increased with Co addition and eventually reached a full recoverable shape memory effect at the highest amount of 10 at.-% of Co.

Footnotes

Acknowledgements

The experimental tests were performed in the Accredited Testing Laboratories at the Institute of Metallurgy and Materials Science of the Polish Academy of Sciences.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Ullakko

Huang

Kantner

Large magnetic-field-induced strains in Ni₂MnGa single crystals. Appl Phys Lett. 1996; 69:1966–1968. doi: 10.1063/1.117637

Meng

Giant magnetic-field-induced strains in Heusler alloy NiMnGa with modified composition. Appl Phys Lett. 1999; 75:2990–2992. doi: 10.1063/1.125211

Heczko

Sozinov

Ullakko

Giant field-induced reversible strain in magnetic shape memory NiMnGa alloy. IEEE Trans Magn. 2000; 36:3266–3268. doi: 10.1109/20.908764

Murray

Marioni

Allen

6% magnetic-field-induced strain by twin-boundary motion in ferromagnetic Ni–Mn–Ga. Appl Phys Lett. 2000; 77:886–888. doi: 10.1063/1.1306635

Sozinov

Likhachev

Lanska

Giant magnetic-field-induced strain in NiMnGa seven-layered martensitic phase. Appl Phys Lett. 2002; 80:1746–1748. doi: 10.1063/1.1458075

Ullakko

Magnetically controlled shape memory alloys: A new class of actuator materials. J Mater Eng Perform. 1996; 5:405–409. doi: 10.1007/BF02649344

Suorsa

Pagounis

Ullakko

Magnetic shape memory actuator performance. J Magn Magn Mater. 2004; 272–276:2029–2030. doi: 10.1016/j.jmmm.2003.12.1026

Müllner

Chernenko

Kostorz

Stress-induced twin rearrangement resulting in change of magnetization in a Ni-Mn-Ga ferromagnetic martensite. Scr Mater. 2003; 49:129–133. doi: 10.1016/S1359-6462(03)00219-7

Smith

Tellinen

Ullakko

Rapid actuator and response of Ni-Mn-Ga to magnetic-field-induced stress. Acta Mater. 2014; 80:373–379. doi: 10.1016/j.actamat.2014.06.054

10.

Hobza

Patrick

Ullakko

Sensing strain with Ni-Mn-Ga. Sens Actuators A Phys. 2018; 269:137–144. doi: 10.1016/j.sna.2017.11.002

11.

James

Wuttig

Magnetostriction of martensite. Philos Mag A. 1998; 77:1273–1299. doi: 10.1080/01418619808214252

12.

Likhachev

Ullakko

Magnetic-field-controlled twin boundaries motion and giant magneto-mechanical effects in Ni–Mn–Ga shape memory alloy. Phys Lett A. 2000; 275:142–151. doi: 10.1016/S0375-9601(00)00561-2

13.

Heczko

Straka

Novak

Magnetic anisotropy of nonmodulated Ni–Mn–Ga martensite revisited. J Appl Phys. 2010; 107:09A914. doi: 10.1063/1.3357409

14.

O'Handley

RC.

Model for strain and magnetization in magnetic shape-memory alloys. J Appl Phys. 1998; 83:3263–3270. doi: 10.1063/1.367094

15.

Karaca

Karaman

Basaran

Magnetic field and stress induced martensite reorientation in NiMnGa ferromagnetic shape memory alloy single crystals. Acta Mater. 2006; 54:233–245. doi: 10.1016/j.actamat.2005.09.004

16.

Pagounis

Chulist

Szczerba

Over 7% magnetic field-induced strain in a Ni-Mn-Ga five layered martensite. Appl Phys Lett. 2014; 105:052405. doi: 10.1063/1.4892633

17.

Pagounis

Szczerba

Chulist

Large magnetic field-induced work output in a NiMnGa seven-layered modulated martensite. Appl Phys Lett. 2015; 107:152407. doi: 10.1063/1.4933303

18.

Likhachev

Sozinov

Ullakko

Modeling the strain response, magneto-mechanical cycling under the external stress, work output and energy losses in Ni–Mn–Ga. Mech Mater. 2006; 38:551–563. doi: 10.1016/j.mechmat.2005.05.022

19.

Sozinov

Likhachev

Lanska

Stress-induced variant rearrangement in Ni-Mn-Ga single crystals with nonlayered tetragonal martensitic structure. J Phys IV France. 2004; 115:121–128. doi: 10.1051/jp4:2004115015

20.

Soolshenko

Lanska

Ullakko

Structure and twinning stress of martensites in non-stoichiometric Ni₂MnGa single crystal. J Phys IV France. 2003; 112:947–950. doi: 10.1051/jp4:20031037

21.

Söderberg

Straka

Novak

Tensile/compressive behaviour of non-layered tetragonal Ni_52.8Mn_25.7Ga_21.5 alloy. Mater Sci Eng A. 2004; 386:27–33. doi: 10.1016/j.msea.2004.07.045

22.

Sozinov

Likhachev

Ullakko

Magnetic and magnetomechanical properties of Ni-Mn-Ga alloys with easy axis and easy plane of magnetization. Proc SPIE. 2001; 4333:189–196. doi: 10.1117/12.432756

23.

Szczerba

Chulist

Detwinning of a non-modulated Ni–Mn–Ga martensite: from self-accommodated microstructure to single crystal. Acta Mater. 2015; 85:67–73. doi: 10.1016/j.actamat.2014.11.008

24.

Soroka

Sozinov

Lanska

Composition and temperature dependence of twinning stress in non-modulated martensite of Ni-Mn-Ga-Co-Cu magnetic shape memory alloys. Scripta Mater. 2018; 144:52–55. doi: 10.1016/j.scriptamat.2017.09.046

25.

Sozinov

Lanska

Soroka

12% magnetic field-induced strain in Ni-Mn-Ga-based non-modulated martensite. Appl Phys Lett. 2013; 102:021902. doi: 10.1063/1.4775677

26.

Vasil'ev

Bozhko

Khovailo

Structural and magnetic phase transitions in shape-memory alloys Ni_2+xMn_1-xGa. Phys Rev B. 1999; 59:1113–1120. doi: 10.1103/PhysRevB.59.1113

27.

Yang

ST.

Effect of composition on transformation temperatures of Ni–Mn–Ga shape memory alloys. Mater Lett. 2003; 57:4291–4296. doi: 10.1016/S0167-577X(03)00305-7

28.

Chen

Qin

ZX.

Martensitic transformation and microstructure of polycrystalline Ni–Mn–Ga and C-doped Ni–Mn–Ga Heusler alloys. Mater Sci Tech-Lond. 2009; 25(7):829–832. doi: 10.1179/174328408X369375

29.

Gao

Wang

Sui

Microstructural characterization and properties of quaternary Ni₅₀Mn₂₉Ga₂₀Gd₁ ferromagnetic shape memory alloy. Mater Sci Tech-Lond. 2013; 29(9):1095–1100. doi: 10.1179/1743284713Y.0000000308

30.

Chernenko

VA.

Compositional instability of β-phase in Ni-Mn-Ga alloys. Scr Mater. 1999; 40:523–527. doi: 10.1016/S1359-6462(98)00494-1

31.

Seguí

Cesari

Effect of ageing on the structural and magnetic transformations and the related entropy change in a Ni-Co-Mn-Ga ferromagnetic shape memory alloy. Intermetallics. 2011; 19:721–725. doi: 10.1016/j.intermet.2011.01.015

32.

Müllner

King

AH.

Deformation of hierarchically twinned martensite. Acta Mater. 2010; 58:5242–5261. doi: 10.1016/j.actamat.2010.05.048