Influence of Current Density upon Hydrogenation on the Shape Memory Effect of Binary TiNi Alloy Single Crystals

Abstract

Some results concerning the hydrogen effect at electrolytic saturation at a current density of j = 1500 and 3500 A/m² for 3 h at room temperature on the temperature dependence of the yield stress σ_0.1(T) and the shape memory effect (SME) under tension of the [011]-oriented Ti-50.55%Ni (at.%) alloy single crystals are presented. It was shown that hydrogen is in a solid solution and forms particles of titanium hydride TiH₂ after hydrogenation at j = 1500 and 3500 A/m², respectively. Both hydrogen in the solid solution and TiH₂ particles led to a decrease in the M_s temperature of the onset of the forward martensitic transformation (MT) upon cooling and the M_d temperature (M_d is the temperature at which the stresses for the onset of the stress-induced MT are equal to the stresses for the onset of plastic flow of the high-temperature B2 phase), and increased the yield stress σ_0.1 of the B2 phase at the M_d temperature compared to hydrogen-free crystals. It was found that the SME under stress depends on the tensile stress level and current density. The maximum SME ε_SME = 10 ± 0.2% at σ_ex = 200 MPa and ε_SME = 10.5 ± 0.2% at σ_ex = 300 MPa was observed in the hydrogen-free crystals and after hydrogenation at j = 1500 A/m², respectively, which exceeded the theoretical value of lattice deformation ε₀ = 8.95% for the B2-B19′ MT in [011] orientation under tension. At j = 1500 A/m², the physical reason for the excess of the SME of the theoretical ε₀ value was due to the increase in the plasticity of B19′ martensite upon hydrogenation. At j = 3500 A/m², ε_SME = 8.0 ± 0.2%, and it was less than ε₀ = 8.95% for B2-B19′ MT in [011] orientation under tension. The decrease in SME after hydrogenation at j = 3500 A/m² was associated with the interaction of two types of B19′-martensite: oriented under stress and non-oriented, formed near TiH₂ particles. It was shown that the redistribution of hydrogen in the bulk of the crystals during long-term holding for 168 h at 263 K after hydrogenation at j = 1500 A/m² increases the SME relative to crystals without long-term holding: 3.5 times at 50 MPa and 1.8 times at 100–150 MPa. After long-term holding, ε_SME = 9.5 ± 0.2% at 150 MPa, which exceeds the theoretical value ε₀ = 8.95% for B2-B19′ MT in [011] orientation under tension.

1. Introduction

It is well known that interstitial atoms are the most effective in solid-solution hardening of alloys compared to substitutional atoms [1,2,3,4]. Among the interstitial atoms, such as nitrogen and carbon, hydrogen can be distinguished in a special row. Firstly, hydrogen is “the smallest and lightest atom”, which, as a hardener, can be located both in octahedral and tetrahedral interstices. However, hydrogen strengthens the solid solution much weaker than nitrogen and carbon atoms, due to the small elastic distortion of the crystal lattice. Therefore, to increase the yield stress of alloys with a face-centered cubic lattice (fcc) and a body-centered cubic lattice, a high concentration is necessary [1]. It is important to note that at high concentrations above 400 wppm (wppm is weight-part per million; ~100 wppm = 0.01 wt.%~0.5 at.% [5]), hydrogen in TiNi alloys forms metal hydride particles, which, as well as hydrogen in a solid solution, lead to strengthening of the metal and alloy [1,6,7,8]. Secondly, saturation with hydrogen, in contrast to alloying with nitrogen and carbon atoms, can be carried out electrolytically at room temperature due to the small size and high diffusion mobility of hydrogen atoms [1,7,8,9,10,11,12]. This is important in terms of technology, since to improve the properties of the sample, for example, superplasticity, hydrogen atoms can be quickly introduced into the samples to the full depth [1]. Thirdly, due to high mobility, hydrogen can be exchanged with the environment and redistributed in the volume of the sample during plastic deformation at normal strain rates of 10⁻⁴ s⁻¹, which can have a significant effect on the properties of alloys. In the electrolytic method of hydrogenation, hydrogen, as a rule, is concentrated in a narrow near-surface layer, which is significantly hardened and leads to the appearance of hydrogen embrittlement [1,12,13]. Long-term holding of electrolytically hydrogenated samples at room temperature leads to a diffusion redistribution of hydrogen over the sample volume. This, for instance, in austenitic steels increased plasticity and suppressed hydrogen embrittlement [1].

The hydrogen effect on thermoelastic martensitic transformations (MT) has been most thoroughly studied in polycrystalline TiNi alloys. These studies have shown that hydrogen changes the M_s temperature of the onset of the forward MT upon cooling [7,8,14,15], which leads to a decrease in the stability of the crystal lattice [8,16], promotes the formation of oriented martensite and the appearance of a two-way shape memory effect (SME) during hydrogenation [17], leads to the formation of titanium hydride TiH₂ particles, and affects the strength properties of martensite and austenite [1,7,17]. On a high-nickel Ti-50.9 Ni alloy with B2-R-B19′ MT and a nanocrystalline structure, it has shown that long-term holding for 6 months at room temperature suppresses R-B19′ MT and does not affect the B2–R transition. As a matter of fact, the B2–R transition has been observed at the same temperature after a long-term holding. However, in this case, it was characterized by lower entropy ΔH values compared to hydrogenated crystals before a long-term holding at room temperature [10]. It can be assumed that a long-term holding after hydrogenation will also affect the SME.

The purpose of this work was to study the effect of different current densities (1500 and 3500 A/m²) during hydrogenation for 3 h at room temperature on the temperature dependence of yield stress σ_0.1(T) within the temperature range of 77 to 373 K, the SME and the effect of long-term holding for 168 h at temperature of 263 K after hydrogenation at 1500 A/m² on the SME in [011]-oriented Ti-50.55Ni (at.%) alloy single crystals under tension. At 296 K, hydrogenation occurs in the B2 phase, since the Ti-50.55Ni (at.%) alloy at 296 K is in the high-temperature B2 phase, due to the low M_s temperature for the start of forward B2-B19′ MT upon cooling due to the high Ni concentration [18,19,20]. It was assumed that in the [011]-oriented Ti-50.55Ni single crystals after hydrogenation, by analogy with previous studies [4,7,8,17], hydrogen atoms will be in a solid solution at a current density of j = 1500 A/m² and form particles of titanium hydride TiH₂ at j = 3500 A/m². The choice of the [011] orientation is due to the large theoretical value of the lattice deformation ε₀ = 8.95%, which includes the strain of the formation of twinned B19′ martensite ε_CVP = 5.2% (CVPs are the correspondence variant pairs) and the detwinning strain of B19′ martensite ε_det = 3.75% [18,21]. This choice of orientation will make it possible to elucidate the effect of different current densities during hydrogenation (or hydrogen concentration) on the mobility of twin boundaries and the SME during isobaric tensile deformation. Secondly, the [011]-oriented crystals are characterized by large values of the Schmid factor m_sl = 0.35 for <100>{110} slip systems in the high-temperature B2 phase under tension [22,23]. This means that the B2 phase is easily deformed by slip and, at high levels of external tensile stresses σ_ex~200–300 MPa, B2–B19′ MT under stress in the temperature range of the stress-induced MT (SIM) can be accompanied by local plastic deformation of the B2 phase [19,22,23]. This can lead to a deviation from the linear dependence σ_0.1 on the temperature dependence of σ_0.1(T) in the temperature range of SIM, the appearance of irreversible deformation ε_irr, and SME degradation in isobaric experiments in “transformation strain-temperature” cycles under tensile stresses. Thus, it was assumed that hydrogen, being in a solid solution, and TiH₂ particles will lead to an increase in σ_0.1 stresses of the B2 phase, suppress local plastic deformation of the B2 phase during the development of the stress-induced MT in the temperature range of SIM on the σ_0.1(T) dependence, and improve the SME compared to the hydrogen-free crystals.

2. Materials and Methods

Ti-50.55Ni alloy (at.%) single crystals were grown by the Bridgman method in a helium atmosphere with graphite crucibles on a Russian-made Redmet installation (Firm “Kristallooptika”, Tomsk, Russia). To determine the [011] orientation of the crystals, the diffractometric method was used by means of a DRON-3M X-ray diffractometer (Bourevestnik, St.-Petersburg, Russia) with monochromatic Fe Kα radiation. Dog-bone-shaped tension samples with a gauge length of 12 mm and a cross section of 2 × 1.5 mm² were cut using wire electrical discharge machining ARTA-5.9 (DELTA-TEST, Fryazino, Moscow region, Russia). The damaged layer after cutting was removed by chemical etching in a 3H₂O + 2HNO₃ + 1HF solution. All samples were initially homogenized at 1220 K for 12 h in a quartz tube, in a helium atmosphere, followed by water quenching. After quenching, the samples were subjected to mechanical grinding and electrolytic polishing in 490 mL of an CH₃COOH + 10 mL of an HClO₄ electrolyte at 263 K, with 20 V applied voltage. The chemical composition of the single crystals after quenching was determined using the X-ray fluorescence method, by means of a wavelength dispersive X-ray fluorescence XRF-1800 spectrometer (SHIMADZU, Kyoto, Japan), giving the atomistic percentages Ti = 49.45% and Ni = 50.55% (at.%). The martensitic transformation (MT) was monitored by the temperature dependence of the electrical resistance ρ(T) (a Russian-manufactured installation (Firm “Kristallooptika”, Tomsk, Russia), with a heating/cooling rate of 10 K/min within a temperature range of 77 to 623 K). Starting M_s and finishing M_f temperatures of the forward B2-B19′ MT during cooling and starting A_s and finishing A_f temperatures of the reverse B19′-B2 MT during heating were determined by the intersection of tangents on the ρ(T)- curves. Mechanical tests were determined using an Instron 5969 universal testing machine (Instron, Norwood, MA, USA), at a strain rate of 4 × 10⁻⁴ s⁻¹. SME under isobaric tensile deformation was studied on home-made dilatometer (Firm “Kristallooptika”, Tomsk, Russia) during cooling and heating in the temperature range from 77 to 400 K at a constant stress in the cycle, with a heating/cooling rate of 10 K/min.

The samples were subjected to electrolytic hydrogenation in a thermostated cell in a 0.9% NaCl solution at a current density of 1500 and 3500 A/m². The saturation time was 3 h at room temperature, which was chosen to obtain a high hydrogen concentration in order to obtain strengthening of the B2 phase. To obtain the same hydrogen concentration in the samples under the same hydrogenation regime, several samples were placed simultaneously in the cell. The anodes were plane-parallel stainless steel plates, between which the samples were placed. The samples were the cathodes. The hydrogen concentration was measured on a LECO RHEN 602 gas analyzer (LECO, St. Joseph, MI, USA). To safekeep hydrogen in the samples after hydrogenation, the samples were placed in a vessel with liquid nitrogen. Long-term holding of samples after hydrogenation at j = 1500 A/m² for 3 h was carried out at a temperature of 263 K for 168 h in a freezer. It was assumed that long-term holding at 263 K for 168 h would slow down the release of hydrogen from the sample surface into the environment, and almost all hydrogen introduced by the electrolytic method would be redistributed over the sample volume compared to long-term holding at 296 K.

To study the phase composition and microstructure of the samples after hydrogenation, a transmission electron microscope (TEM), a JEOL-2010 (JEOL, Tokyo, Japan), with an accelerating voltage of 200 kV was used. The thin foils were prepared using double-jet electropolishing (TenuPol-5; “Struers”, Ballerup, Denmark), with an electrolyte containing 20% sulfuric acid in methyl alcohol at room temperature, with 12.5 V applied voltage. In hydrogen-free crystals, thin foils were cut from the body of the sample. Thin foils of hydrogenated samples were cut from directly at the edge of the samples that contained the hydrogenation product. The fracture surfaces were investigated using a TESCAN VEGA3 scanning electron microscope (SEM) (TESCAN, Brno, Czech).

3. Results and Discussion

3.1. Temperature Dependence of Yield Stress

The temperatures of B2-B19′ MT of the [011]-oriented Ti-50.55Ni alloy single crystals after hydrogenation at different current densities j are presented in

Table 1. Martensitic transformation temperatures depend on the current density of hydrogenation.

State	Concentration H2, wppm	Concentration H2, at.%	Ms, K	Mf, K	As, K	Af, K	Th1 = Af − Ms, K	Th2 = As − Mf, K
Without H2	0	0	224	182	222	258	34	40
With H2, j = 1500 A/m2	100 ± 20	~0.5	205	170	215	255	40	45
With H2, j = 3500 A/m2	500 ± 20	~2.5	213	170	215	259	46	45

. The [011]-oriented Ti-50.55Ni single crystals after hydrogenation are characterized by a one-stage B2-B19′ MT, regardless of the current density j [8,19]. Analysis of the data presented in Table 1 shows that hydrogenation at a current density of j = 1500 and 3500 A/m² leads to a decrease in the M_s temperature by 19 and 11 K, respectively, relative to hydrogen-free crystals. Previously, a decrease in the M_s temperature under similar hydrogenation regimes was observed on Ti-50.3Ni and Ti-50.7Ni (at.%) single crystals [8,14,15]. The temperature hysteresis after hydrogenation is symmetrical, since T_h¹ = A_f − M_s and T_h² = A_s − M_f are equal to each other, but it becomes larger in magnitude than in the hydrogen-free crystals.

TEM studies showed that after hydrogenation at j = 1500 A/m² at 300 K, only reflections from the ordered B2 phase were observed in microdiffraction patterns with the (110) zone axis. Reflections from the R phase and particles of secondary phases were not detected (Figure 1a). Consequently, hydrogen atoms after hydrogenation at a current density of j = 1500 A/m² for 3 h of [011]-oriented Ti-50.55Ni alloy single crystals are in solid solution, similarly to the Ti–50.3Ni alloy crystals after hydrogenation in this regime [8]. With regard to hydrogenation at j = 3500 A/m², the [011]-oriented Ti-50.55Ni alloy single crystals were characterized by a two-phase state at 300 K. On the bright-field images, particles of a non-equiaxed shape were observed in the B2 phase (Figure 1b). Diffraction analysis showed that these particles are titanium hydride TiH₂, which have an fcc lattice and non-coherent conjugation with the B2 phase (Figure 1b). Titanium hydride TiH₂ particles of non-equiaxed shape had a length of 1500 nm and a width of 400 nm on average (Figure 1b). The volume fraction of TiH₂ particles according to TEM data was 7%. The formation of TiH₂ particles at j = 3500 A/m² leads, firstly, to an increase in the nickel concentration in the B2 phase. As a result of this factor, the M_s temperature should decrease relative to the M_s temperature of hydrogenated crystals at 1500 A/m². Secondly, the particles are sources of internal stresses and this factor should lead to an increase in the M_s temperature compared to hydrogenated crystals at j = 1500 A/m². The result of these two factors is an increase in the M_s temperature by 11 K relative to the M_s temperature of hydrogenated crystals at j = 1500 A/m² (Table 1). A more important role in the increase in M_s temperature is played by internal stresses from the TiH₂ particles.

The temperature dependence of yield stress σ_0.1(T) during tensile deformation of [011]-oriented Ti-50.55Ni alloy single crystals without and with hydrogen within a wide temperature range of 170 to 570 K is shown in Figure 2. It can be seen that the σ_0.1(T) dependence exhibits stages characteristic of alloys experiencing MT under stress [18,19,24,25]. The minimum stresses σ_0.1 on the σ_0.1(T) dependence were observed at the M_s temperature coinciding with the M_s obtained in the study of the temperature dependence of the electrical resistance ρ(T) (Table 1). At T < M_s, a low-temperature stage was observed, at which σ_0.1 increased with decreasing test temperature. This is typical of alloys in the martensitic state and is associated with the thermally activated motion of twinning and interphase boundaries. The maximum stresses σ_0.1 on the σ_0.1(T) dependence correspond to the M_d temperature. At M_d temperature, the stresses for the onset of stress-induced B2-B19′ MT are equal to the stresses for the onset of plastic deformation of the high-temperature B2 phase. In the temperature range M_s < T < M_d, the σ_0.1 increased linearly with increasing test temperature. This stage of the σ_0.1(T) dependence is associated with stress-induced B19′-martensite and is described by the Clausius–Clapeyron equation [18,19,24,25]:

$$\frac{d{\sigma }_{0.1}}{dT}=-\frac{\Delta H}{{\epsilon }_0{T}_0}$$

(1)

where ΔH is the change in enthalpy at B2–B19′ MT, T₀ is the phase equilibrium temperature, and ε₀ is the lattice deformation at B2–B19′ MT. At T > M_d, a stage was observed that was related with plastic deformation of the high-temperature B2 phase.

Analysis of the data presented in Figure 2 led to the following findings. Firstly, hydrogenation leads to a decrease in the M_s temperature by 25 and 15 K at j = 1500 and 3500 A/m², respectively (Figure 2, Inset). Secondly, in the temperature range M_s < T < M_d, the σ_0.1(T) dependence in the hydrogen-free crystals has two sections with different values of α = dσ_0.1/dT: the first section in the temperature range of 225 to 300 K with α = dσ_0.1/dT = 5.1 MPa/K and the second section from 300 K to M_d temperature with α = dσ_0.1/dT = 1.6 MPa/K. After hydrogenation at j = 1500 A/m², the σ_0.1(T) dependence has only one section with α = dσ_0.1/dT = 5.0 MPa/K in the temperature range M_s < T < M_d. As for the hydrogenation at j = 3500 A/m², two sections again appear on the σ_0.1(T) dependence in the temperature range M_s < T < M_d: the first in the temperature range of 213 to 290 K with α = dσ_0.1/dT = 6.5 MPa/K and the second from 290 K to M_d with α = dσ_0.1/dT = 2.1 MPa/K. Thirdly, hydrogen lowers the M_d temperature and increases the σ_0.1(M_d) of the high-temperature B2 phase. Relative to the hydrogen-free crystals, the M_d temperature decreases by 20 and 40 K, and σ_0.1(M_d) increases by 55 and 75 MPa at j = 1500 and 3500 A/m², respectively, (Figure 2). Hence, both the hydrogen in the solid solution and TiH₂ particles strengthen the B2 phase. A similar strengthening of the initial phase upon hydrogenation by the electrolytic method was previously observed in single crystals of the Ti–50.7Ni (at %) alloy and austenitic stainless steel, when hydrogen was in a solid solution and formed titanium hydride particles [3,6,7,8].

The TiH₂ particles are non-coherent and interact with dislocations by the rounding mechanism [26,27,28]. According to [26,27,28], the yield stress σ_0.1 in alloys with non-coherent particles during slip deformation is determined as

$${\sigma }_{0.1}={\sigma }_f\left(b\right)+\frac{G·b}{L}$$

(2)

Here, σ_f(b) is the friction stress experienced by the perfect dislocation a/2<001> moving in the B2 phase; G = 41,000 MPa is the shear modulus of the TiNi alloys in the B2 phase [18]; L is the distance between the TiH₂ particles, which from the TEM data was equal to 700 nm on average; and b = 0.15 nm is the modulus of the Burgers vector of the perfect dislocation a/2<100>. As a result, the estimate of the contribution (Gb/L) from the TiH₂ particles to σ_0.1 gives a value of the order of 10 MPa for L = 700 nm. This value turned out to be close to the stress difference Δσ_0.1 = 20 MPa at the M_d temperature between the crystals after hydrogenation at j = 1500 and 3500 A/m² (Figure 2). Consequently, a low volume fraction of TiH₂ particles does not make a significant contribution to the strengthening of the B2 phase during hydrogenation at j = 3500 A/m² relative to the hydrogenation at j = 1500 A/m², or does not compensate for the softening of the matrix due to the reduction of Ti and H₂ atoms during their formation.

The difference in the phase state of [011]-oriented Ti-50.55Ni alloy single crystals upon hydrogenation at a current density of j = 1500 and 3500 A/m², when hydrogen atoms, H₂, are in solid solution at j = 1500 A/m² (single-phase state) or form the TiH₂ particles at j = 3500 A/m² (two-phase state: B2 phase + TiH₂ particles), may be the reason for the appearance of one or two stages on the σ_0.1(T) dependence in the SIM temperature range.

On the σ_0.1(T) dependence in the SIM temperature range, the transition from the stage with high α = dσ_0.1/dT to the stage with low α = dσ_0.1/dT indicates that the development of stress-induced MT proceeds simultaneously with the local plastic deformation of the B2 phase [18,19,22,23]. In this case, the plastic deformation that accompanies the stress-induced MT blocks the increase in the transformation strain ε_tr during cooling/heating under stress, leads to the appearance of irreversible deformation ε_irr when heated under stress, and reduces the SME [19,22]. The appearance of a stage with low α = dσ_0.1/dT in the SIM temperature range with a one-stage B2-B19′ MT is controlled by the stress level of the high-temperature B2 phase [18,19,22,23]. Indeed, in the [011]-oriented Ti-50.55Ni alloy crystals after hydrogenation at j = 1500 A/m², when hydrogen was in solid solution and increased σ_0.1(M_d), only one stage of the linear σ_0.1(T) dependence took place in the SIM temperature range. As for hydrogenation at j = 3500 A/m², when TiH₂ particles appeared, the B2 phase stresses at the M_d temperature also increased, as in the case of hydrogenation at j = 1500 A/m². Nevertheless, the σ_0.1(T) dependence in the SIM temperature range showed a deviation from the linear dependence of α = dσ_0.1/dT (Figure 2); namely, there were two sections with different α = dσ_0.1/dT values. In the case of hydrogenation at j = 3500 A/m², the transition from the stage with high α = dσ_0.1/dT to the stage with low α = dσ_0.1/dT on the σ_0.1(T) dependence in the SIM temperature range can be associated with the plastic deformation of the TiH₂ particles themselves and local deformation of the B2 phase directly near the particles during the formation of stress-induced B19′-martensite [18,19]. The TiH₂ particles have an fcc structure (Figure 1b) and, like other fcc alloys that do not experience MT, they can be deformed by slip or twinning [1]. However, in the [011]-oriented crystals containing TiH₂ particles, the transition from the stage with high α = dσ_0.1/dT to the stage with low α = dσ_0.1/dT occurred at higher σ_0.1 = 550 MPa than in the [011]-oriented hydrogen-free crystals due to the higher stress level of σ_0.1(M_d) (Figure 2).

It is important to note that at the stage of linear σ_0.1(T) dependence in the SIM range from 50 to 450–500 MPa, α = dσ_0.1/dT had close values of 5–5.1 MPa/K in the hydrogen-free crystals and after hydrogenation at j = 1500 A/m², and it increased to 6.5 MPa/K after hydrogenation at j = 3500 A/m². According to equation (1), the α = dσ_0.1/dT value is proportional to 1/ε₀ [18,19,22,23]. Therefore, in accordance with equation (1), we should expect close SME for the hydrogen-free crystals and after hydrogenation at j = 1500 A/m² and its decrease at j = 3500 A/m², which will be presented below.

3.2. Shape Memory Effect

The results of studying the transformation strain ε_tr during cooling/heating at different tensile stress levels σ_ex from 50 to 300 MPa for the [011]-oriented Ti-50.55Ni alloy single crystals without and with hydrogen are presented in Figure 3 and Figure 4. It can be seen from Figure 3 that in the [011]-oriented Ti-50.55Ni single crystals without and with hydrogen, the B2-B19′ MT under a tensile stress developed almost according to explosive kinetics with large values of dε_tr/dT [22]. The M_s temperature increased with the growth of σ_ex (Figure 3 and Figure 5) and the value α = dσ/dM_s was described by equation (1). The value of reversible ε_rev (or SME) and irreversible ε_irr deformation depended on the tensile stress level σ_ex and the current density during hydrogenation (Figure 3 and Figure 4). The transformation strain ε_tr was the total strain that occurred in the “cooling-heating” cycle under a tensile stress σ_ex during cooling. The value of reversible deformation ε_rev or SME ε_SME was determined in the “cooling-heating” cycle under σ_ex after heating. If after heating in the “cooling-heating” cycle under a tensile stress σ_ex there was no irreversible deformation, ε_irr = 0, then ε_tr = ε_SME and when ε_irr ≠ 0, then ε_SME = ε_tr – ε_irr and ε_tr > ε_SME. All types of deformations, namely, transformation deformation during cooling under stress ε_tr, reversible deformation ε_rev, and irreversible deformation ε_irr during heating under stress, are shown in Figure 3.

As can be seen from Figure 3 and Figure 4, in the [011]-oriented Ti-50.55Ni single crystals with and without hydrogen, the ε_tr(T) curves in the “cooling-heating” cycle under stress had a closed form at tensile stresses of σ_ex = 50–150 MPa. Hence, B2-B19′ MT, which was realized under stress during cooling, was completely reversible when heated, and, therefore, SME was realized [8,18,19,22,23,24]. At σ_ex = 50 MPa, the SME in hydrogen-free crystals was 2–3 times higher than after hydrogenation at j = 1500 and 3500 A/m². With a successive increase in σ_ex, the ε_tr and SME values in hydrogen-free crystals were always higher than in crystals after hydrogenation. Consequently, both hydrogen in solid solution and TiH₂ particles increased the resistance to the movement of interphase boundaries [7,8]. The irreversible deformation ε_irr in the “cooling-heating” cycle under tensile stress σ_ex appeared at σ_ex ≥ 150 MPa and then increased with increasing σ_ex. The maximum reversible deformation ε_rev or SME of 10.0 ± 0.2% was realized in the hydrogen-free [011]-oriented Ti-50.55Ni single crystals at σ_ex = 200 MPa. This SME value, ε_SME = 10.0 ± 0.2%, in hydrogen-free crystals turned out to be 1.0 ± 0.2% larger than the theoretical value of the lattice deformation ε₀ = ε_CVP + ε_det = 8.95% for the [011] orientation to B2-B19′ MT under tension [21]. After hydrogenation, the maximum SME was 10.5 ± 0.2% at σ_ex = 300 MPa and 8.0 ± 0.2% at σ_ex = 200 MPa, respectively, at j = 1500 and 3500 A/m² (Figure 3 and Figure 4). When hydrogen was in solid solution at j = 1500 A/m², the maximum ε_SME = 10.5 ± 0.2% turned out to be 1.55 ± 0.2% higher than the theoretical value of the lattice deformation ε₀ = ε_CVP + ε_det = 8.95% for the [011] orientation to B2-B19′ MT under tension [21]. But, in the case of hydrogenation at j = 3500 A/m², when hydrogen formed particles, the maximum ε_SME = 8.0 ± 0.2% was somewhat less than the theoretical value ε₀. At σ_ex > 200 and 300 MPa, the [011]-oriented Ti-50.55Ni crystals after hydrogenation at j = 3500 and 1500 A/m², respectively, were destroyed from the beginning of the development of B2-B19′ MT under stress during cooling.

A similar result, when the SME exceeded the theoretical ε₀ value, was obtained on Ti-40Ni-10Cu single crystals for the B2-B19-B19’ MT [22]. In the case of the [011]-oriented Ti-40Ni-10Cu crystals, the physical reason for the excess of the reversible strain under tensile stresses of the theoretical ε₀ value was associated with the simultaneous development under stress of mechanical {110}_B19 twinning in B19 martensite and {001}_B19′ twinning in B19’ martensite. These twins were completely reversible in the “cooling-heating” cycle under stress at σ_ex ≤ 200 MPa. By analogy with the [011]-oriented Ti-40Ni-10Cu crystals, in our case, in the [011]-oriented Ti-50.55Ni crystals, the excess SME of the theoretical ε₀ value can also be associated with the development of {001}_B19′ twins in B19′-martensite. In the [011]-oriented Ti-50.55Ni crystals in the “cooling-heating” cycle, the irreversible deformation was 2.0 ± 0.2% at σ_ex = 250 MPa in hydrogen-free crystals and 1.3 ± 0.2% at σ_ex = 300 MPa in crystals after hydrogenation at j = 1500 A/m². TEM studies of the microstructure in both cases revealed only dislocations that were in the one system (Figure 6a–c). The dislocation density was higher in hydrogen-free crystals than in crystals after hydrogenation at j = 1500 A/m², which was due to the difference in the magnitude of irreversible deformation. Residual B19’ martensite with {001}_B19′ twins could not be detected. To elucidate the physical reason for the excess of the SME of the theoretical ε₀ value in these crystals, additional in situ studies of the MT under stress in the column of a transmission electron microscope are required. It is important to note that after hydrogenation at j = 1500 A/m² the [011]-oriented Ti-50.55Ni crystals showed SME at σ_ex = 300 MPa, while hydrogen-free crystals at these stresses were already destroyed from the beginning of the development of an MT under stress. Consequently, hydrogen, being in a solid solution, not only increased the σ_0.1 stresses of the B2 phase, but also increased the plasticity of B19’ martensite in the [011]-oriented Ti-50.55Ni crystals.

As for the [011]-oriented Ti-50.55Ni crystals, hydrogenated at j = 3500 A/m², in a TEM study of the microstructure after the “cooling-heating” cycle at σ_ex = 200 MPa, when irreversible deformation ε_irr = 0.5 ± 0.2%, residual B19’ martensite was found directly near the TiH₂ particles (Figure 6d–f). From Figure 6d,f, it can be seen that the particles generate their own variant of B19′martensite, which is not oriented with respect to the external tensile stress. Under stress, as is known, oriented B19’ martensite is formed [19,21], which interacts with non-oriented B19′ martensite formed by TiH₂ particles. The interaction of two types of B19′ martensite (oriented and non-oriented) hinders the development of oriented B19′-martensite upon cooling under stress and its reverse movement upon heating under stress. In addition, the TiH₂ particles themselves do not experience MT, but can be deformed only by slip or twinning and are brittle [1]. Both of these factors contribute to the appearance of irreversible deformation during heating under stress at σ_ex > 150 MPa and reduce the SME.

At σ_ex ≥ 250 MPa, all the studied crystals demonstrated a quasi-brittle fracture (Figure 7). However, the fracture surface had differences at σ_ex ≥ 250 MPa. In the case of hydrogen-free crystals, dimples and cleavage facets were observed on the fracture surface, which are characteristic of a quasi-brittle fracture (Figure 7a) [24]. In hydrogenated crystals at j = 1500 A/m², traces of localized deformation in one system were visible on the fracture surface (Figure 7b), which was previously observed in [8], when hydrogen was in a solid solution. As for hydrogenation at j = 3500 A/m², no localization bands were observed and the crack developed in several systems (Figure 7c). Such a crack propagation mechanism is typical for the presence of hydrides [29] and can occur along the “particle-matrix” interfaces or at the intersections of two types of B19′-martensite (oriented and non-oriented formed by titanium hydride TiH₂ particles). Thus, the observed differences on the fracture surface of crystals after hydrogenation are direct evidence of the different nature of the presence of hydrogen in the volume of the sample: in the solid solution after hydrogenation at j = 1500 A/m² and as TiH₂ hydride at j = 3500 A/m².

The effect of hydrogen on the SME (Figure 3 and Figure 4) was studied on samples with a relatively large thickness (h = 1.5 mm). As was mentioned in the introduction and in [10,30], when saturated with hydrogen by the electrolytic method at 300 K, hydrogen in thick samples did not pass through the entire thickness of the sample, but only surface saturation occurred to a depth of about 30–50 μm. This means that, after hydrogenation, the sample is a composite consisting of a hydrogenated surface layer and the inner part of the sample that does not contain hydrogen.

The SME obtained on the samples immediately after hydrogenation at j = 1500 A/m² and after long-term holding for 168 h at 263 K at the same tensile stress level (50, 100 and 150 MPa) is shown in Figure 8. An analysis of the data presented in Figure 8 showed that the long-term holding of the samples after hydrogenation leads, firstly, to an increase in the M_s temperature under stress at the same level of σ_ex. In fact, the M_s temperature under stress increased by 30 K at σ_ex = 50 MPa and by 10–15 K at σ_ex = 100–150 MPa (Figure 8). Therefore, long-term holding of the samples after hydrogenation at j = 1500 A/m² for 168 h at 263 K leads to an increase in the M_s temperature in the free state. Secondly, an increase in the SME was found. The SME increased by 3.5 times at σ_ex = 50 MPa and by 1.8 times at σ_ex = 100–150 MPa, relative to hydrogenated samples without long-term holding. It is important to note that the SME after long-term holding at the same σ_ex level became larger than in hydrogen-free crystals (Figure 4 and Figure 8). In addition, after long-term holding, the ε_SME = 8.3 ± 0.2% at σ_ex = 100 MPa and ε_SME = 9.5 ± 0.2% at σ_ex = 150 MPa, which was almost equal to and greater, respectively, than the theoretical lattice deformation ε₀ = ε_CVP + ε_det = 8.95% for B2-B19′ MT in [011] orientation under tension [21]. It is important to note that after long-term holding, the SME, almost equal to ε₀, was achieved at lower σ_ex = 100 MPa than in the hydrogen-free crystals and after hydrogenation at j = 1500 A/m² without long-term holding. This is due to an increase in the plasticity of B19′ martensite upon the redistribution of hydrogen in the bulk of the sample.

4. Conclusions

The above studies of the development of stress-induced B2-B19′ MT and SME in the [011]-oriented Ti-50.55Ni alloy single crystals with varying current density j = 1500 and 3500 A/m² during electrolytic hydrogenation for 3 h at room temperature allow us to draw the following findings.

• Variation of the current density j = 1500 and 3500 A/m² during electrolytic hydrogenation for 3 h at room temperature did not change the type of B2-B19′ MT and did not lead to the formation of the R phase. When hydrogenated at a current density of 1500 A/m² for 3 h at room temperature, hydrogen was in solid solution and lowered the M_s temperature by 19 K relative to the hydrogen-free crystals with M_s = 224 K. In the case of hydrogenation at a current density of 3500 A/m² for 3 h at room temperature, particles of titanium hydride TiH₂ were formed and the alloy was in a two-phase state at room temperature (B2 phase + TiH₂ particles) with a temperature of M_s = 213 K, which was at 11 K below the M_s temperature in the hydrogen-free crystals. Both hydrogen in solid solution and TiH₂ particles led to an increase in the yield stress σ_0.1 of the B2 phase at the M_d temperature by 55 and 75 MPa, respectively, relative to the hydrogen-free crystals with σ_0.1(M_d) = 700 MPa.
• The SME value depended on the level of the external tensile stresses σ_ex and current density. The maximum SME was ε_SME = 10.0 ± 0.2% at σ_ex = 200 MPa in the hydrogen-free crystals; ε_SME = 10.5 ± 0.2% at σ_ex = 300 MPa in crystals with hydrogen after hydrogenation at j = 1500 A/m²; and ε_SME = 8.0 ± 0.2% at σ_ex = 200 MPa in crystals with hydrogen after hydrogenation at j = 3500 A/m². At j = 1500 A/m², when hydrogen was in solid solution, the maximum ε_SME = 10.5 ± 0.2% exceeded the theoretical value of lattice deformation ε₀ = ε_CVP + ε_det = 8.95% for B2-B19′ MT in [011] orientation under tension, which was associated with an increase in the plasticity of B19′ martensite upon hydrogenation. In the case of formation of TiH₂ particles after hydrogenation at j = 3500 A/m², the maximum ε_SME = 8.0 ± 0.2% was less than ε₀. The physical reason for the decrease in the SME was due to the interaction of two types of B19’ martensite (oriented and non-oriented martensite), which inhibited the development of oriented B19′ martensite under stress during cooling and its reverse movement upon heating.
• The long-term holding of samples for 168 h at 263 K after hydrogenation at j = 1500 A/m² increases the SME relative to hydrogenated samples without long-term holding. The SME was 1.5 ± 0.2%, 4.5 ± 0.2% and 5.9 ± 0.2% without long-term holding, respectively, at σ_ex = 50, 100 and 150 MPa. After long-term holding, the SME became 5.4 ± 0.2%, 8.3 ± 0.2% and 9.5 ± 0.2%, respectively, at σ_ex = 50, 100 and 150 MPa. The physical reason for the increase in the SME after long-term holding was related to the redistribution of hydrogen in the bulk of the crystal and an increase in the plasticity of B19′ martensite.