Robust evidence for the stabilization of the premartensite phase in Ni-Mn-In magnetic shape memory alloys by chemical pressure

The thermodynamic stability of the premartensite (PM) phase has been a topic of extensive investigation in shape memory alloys as it affects the main martensite phase transition and the related physical properties. In general, the PM phase is stable over a rather narrow temperature-composition range. We present here evidence for chemical pressure induced suppression of the main martensite transition and stabilization of the PM phase over a very wide temperature range from 300 to $\sim 5\mathrm{K}$ in a magnetic shape memory alloy (MSMA), ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{16}$, using magnetic susceptibility, synchrotron x-ray powder diffraction (SXRPD) studies, and first-principles calculations. The ac-susceptibility studies show a highly skewed and smeared peak around 300 K without any further transition up to the lowest temperature of our measurement (5 K) for $\sim 5\%$ Al substitution. The temperature evolution of the SXRPD patterns confirms the appearance of the PM phase related satellite peaks at $T\le 300\mathrm{K}$ without any splitting of the main austenite (220) peak showing preserved cubic symmetry. This is in marked contrast to the temperature evolution of the SXRPD patterns of the martensite phase of the Al free as well as $\sim 3\%$ Al substituted compositions where the austenite (220) peak shows a clear splitting due to Bain distortion signalling symmetry breaking transition. Our theoretical calculations support the experimental findings and reveal that the substitution at the In site by a smaller size atom, like Al, can stabilize the PM phase with preserved cubic symmetry. Our results demonstrate that Al-substituted Ni-Mn-In MSMAs provide an ideal platform for investigating the physics of various phenomena related to the PM state.

Figure 1

The temperature dependent real part of ac-susceptibility for (a) ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{15.5}{\mathrm{Al}}_{0.5}$ and (b) ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{15.2}{\mathrm{Al}}_{0.8}$ MSMAs. The insets are enlarged view around 300 K for the field cooled protocol. $T_M$, $T_{\mathrm{PM}}$, $T_C^M,$ and $T_C$ represent the martensite transition temperature, premartensite transition temperature, Curie temperature of the martensite phase, and Curie temperature of the austenite phase, respectively. ZFCW, FC, and FCW correspond to measurements performed during warming on the zero field cooled sample, during field cooling, and during warming on the field cooled sample, respectively.

Figure 2

Typical SXRPD patterns of ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{15.5}{\mathrm{Al}}_{0.5}$ MSMA in the (a) austenite, (b) premartensite, and (c) martensite phases. An enlarged view around the most intense (220) Bragg peak for the austenite and the premartensite (PM) phases, given in inset (i) of (a) and (b), respectively, reveal the appearance of the satellite peaks [indicated by “PM” in the inset (i) of (b)] due to $3M$ like modulation in the PM phase. Untruncated view of the (220) cubic peak for the austenite and PM phases, given in inset (ii) of (a) and (b), respectively, reveal the absence of Bain distortion in the PM phase. The inset of (c) depicts the splitting of the most intense (220) cubic peak and appearance of the satellite peaks due to Bain distortion and $3M$ like modulation of the martensite (M) phase. The peaks related to the martensite phase are marked as “M” in the inset of (c).

Figure 3

The SXRPD patterns of ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{15.2}{\mathrm{Al}}_{0.8}$ are shown in (a) at (i) 400 K, (ii) 220 K, and (iii) 100 K. The insets show an enlarged view around the most intense Bragg peak to reveal the satellite peaks of the premartensite (PM) phase. Enlarged view around the most intense cubic peak (220) at various temperature in the range 400–100 K are given in (b) and (c). The arrows in (c) indicate the temperature dependent shifts of the PM satellite peak positions. Note the gradual sharpening of the satellite peaks in (c) on lowering the temperature. (d) An enlarged view around the most intense (220) cubic peak at selected temperatures reveal the appearance of the most intense satellite peak of the PM phase at $T\sim 300\mathrm{K}$, indicated by an arrow. (e) Untruncated SXRPD profiles of the (220) cubic Bragg peak is depicted in the 400–100-K range while its inset depicts the XRD profiles in the 300–13-K range recorded on a rotating anode-based diffractometer.

Figure 4

The observed (dark black dots), calculated (continuous red line), and difference patterns (continuous green line) obtained after LeBail refinement using SXRPD pattern of ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{34}{\mathrm{In}}_{15.2}{\mathrm{Al}}_{0.8}$ MSMA for the (a) cubic austenite phase at 400 K and (b) $3M$ modulated premartensite (PM) phase at 100 K in the $Fm\overline{3}m$ and $P2/m$ space groups, respectively. The vertical tick marks above the difference profile represent the Bragg peak positions in (a) and (b). The inset of (a) shows the presence of (111) and (200) Bragg reflections characteristic of the $L{2}_1$ ordering in the cubic austenite phase. The inset of (b) shows an enlarged view of the fit around the most intense Bragg peak and satellite reflections (marked as “PM” with their indices) related to the $3M$ modulated premartensite phase. The temperature dependence of the dc magnetization, measured on zero field cooled sample during warming cycle, is shown in (c) for different magnetic fields. The enlarged view of (c) around the FM $T_C$, shown in (d), reveals a skewed diffuse peak due to the PM transition. The variation of the PM transition temperature ($T_{\mathrm{PM}}$) with magnetic field is shown in (e). An enlarged view of SXRPD pattern around the most intense Bragg peak collected at 294 K under zero field (black dots connected with continuous line) and an external magnetic field of 2500 Oe (red squares connected with continuous line) is shown in (f). Note the disappearance of the satellite peaks related to the premartensite (PM) phase under the magnetic field.

Figure 5

Calculated total energies as a function of lattice constant in the austenite and modulated premartensite phases of (a) ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{33.3}{\mathrm{In}}_{12.5}{\mathrm{Al}}_{4.2}$ and (b) ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{33.3}{\mathrm{Al}}_{16.7}$. Supercells with relaxed atomic positions for the austenite and the modulated premartensite phases of ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{33.3}{\mathrm{In}}_{12.5}{\mathrm{Al}}_{4.2}$ and ${\mathrm{Ni}}_{50}{\mathrm{Mn}}_{33.3}{\mathrm{Al}}_{16.7}$ are shown in insets (ii), (iii) of (a) and (b), respectively. ${\mathrm{Mn}}_{\mathrm{In}}$ (${\mathrm{Mn}}_{\mathrm{Al}}$) indicate excess Mn atoms (i.e., above 25% Mn content) in our simulation cells occupying In (Al) sublattice. The Al atoms are behind the In atoms in the insets (ii) and (iii) of (a).