Multisite Exchange-Enhanced Barocaloric Response in ${\mathrm{Mn}}_3\mathrm{NiN}$

We study the barocaloric effect (BCE) in the geometrically frustrated antiferromagnet ${\mathrm{Mn}}_3\mathrm{NiN}$ across the Néel transition temperature. Experimentally, we find a larger barocaloric entropy change by a factor of 1.6 than that recently discovered in the isostructural antiperovskite ${\mathrm{Mn}}_3\mathrm{GaN}$ despite significantly greater magnetovolume coupling in ${\mathrm{Mn}}_3\mathrm{GaN}$. By fitting experimental data to theory, we show that the larger BCE of ${\mathrm{Mn}}_3\mathrm{NiN}$ originates from multisite exchange interactions amongst the local Mn magnetic moments and their coupling with itinerant electron spins. Using this framework, we discuss the route to maximize the BCE in the wider ${\mathrm{Mn}}_{3}A\mathrm{N}$ family.

Popular Summary

Solid-state caloric cooling offers opportunities for more efficient refrigeration without the need for environmentally harmful chemicals. Magnetocalorics are the most studied alternative but rely on expensive permanent magnets. Far less studied are barocalorics, which control temperature with just an application of mechanical pressure. Recent experiments have shown that the material ${\mathrm{Mn}}_3\mathrm{GaN}$ has a giant barocaloric effect (BCE). Here we report an unexpectedly large BCE in the related material ${\mathrm{Mn}}_3\mathrm{NiN}$, exceeding ${\mathrm{Mn}}_3\mathrm{GaN}$ by almost 200%. By modeling the behavior of both materials, we provide a framework to understand these improved properties and also pave the way for an informed search for even larger BCEs in the wider family of Mn-antiperovskite materials.

The giant BCE of ${\mathrm{Mn}}_3\mathrm{GaN}$ arises from the underlying magnetic frustration, where individual magnetic interactions cannot all be satisfied simultaneously. This causes large volume and caloric changes at the magnetic transition. ${\mathrm{Mn}}_3\mathrm{NiN}$ is known to have a significantly smaller volume change at its magnetic transition, however, despite this, our results show that the associated caloric changes are larger. Our theoretical model reveals that magnetic interactions occur between multiple sites in the crystal structure, and these interactions increase the caloric change at the transition in ${\mathrm{Mn}}_3\mathrm{NiN}$, thus explaining the larger BCE.

Our theoretical framework offers a guide to finding yet-larger effects in the chemically flexible ${\mathrm{Mn}}_{3}A\mathrm{N}$ family of materials, where the $A$ site may take numerous different elements. Therefore, we anticipate that our results will stimulate further research into improving the promising BCE in these materials.

Figure 1

(a) ${\mathrm{\Gamma }}^{5g}$ and (b) ${\mathrm{\Gamma }}^{4g}$ magnetic structures of ${\mathrm{Mn}}_3\mathrm{NiN}$. ${\mathrm{\Gamma }}^{5g}$ is a fully compensated antiferromagnetic structure stable at low temperatures. ${\mathrm{\Gamma }}^{4g}$ is also an antiferromagnetic structure (purple arrows) but has a symmetry-allowed ferromagnetic component (gray arrows) causing canting of the spins out of the (111) plane. (c) Thermodiffractogram measured using neutron powder diffraction ($\lambda =1.544\AA$). (d) Unit-cell volume as a function of the temperature refined from the neutron-diffraction data. (e) The refined contributions of the ${\mathrm{\Gamma }}^{5g}$ and ${\mathrm{\Gamma }}^{4g}$ representations to the magnetic structure. (f) Bulk magnetometry data collected on the same sample under an applied field ${\mu }_{0}H=0.05\mathrm{T}$ after zero-field cooling. The blue highlighted region indicates the enlarged moment associated with the ferromagnetic contribution allowed in the ${\mathrm{\Gamma }}^{4g}$ representation. The transition region is (g) temperature-dependent heat flow at ambient pressure $[(dQ)/(dT)]{|}_{p=0}$, which is equal to the heat capacity outside the transition region and includes contributions that arise from latent heat within, and (h) is the total entropy change calculated from (g) for the same sample on cooling.

Figure 2

(a) Magnetization under an applied field of ${\mu }_{0}H=0.05\mathrm{T}$ as a function of the temperature and measured at different pressures after zero-field cooling. (b) Pressure dependence of $T_N$ taken from the derivative of the $M(T)$ VSM data and the peak of the HC data. Solid lines are linear fits to the data.

Figure 3

(a) Temperature-dependent heat flow under pressure $[(dQ)/(dT)]{|}_p$ on cooling through the transition at different values of increasing pressure $p$ after baseline subtraction. The full legend is the same as in (d) and (e). (b) Entropy changes cooling through $T_N$ obtained from the data in (a). (c) Isobaric entropy curves $S-S_{210\mathrm{K}}$ on cooling at 0- and 0.48-GPa applied pressure. The green arrow from 1 to 2 and orange arrow from 3 to 4 indicate the isothermal entropy and adiabatic temperature changes, respectively, upon removal of pressure. (d),(e) Isothermal entropy and adiabatic temperature changes obtained from the data in (a) as explained in (c).

Figure 4

The theoretical quantities (a) $T_N$ and (b) $[(d{T}_N)/(dp)]$ as a function of $a_2$ and $c_{\mathrm{MV}}$, respectively, calculated using our model described in the text. Dotted lines indicate the experimentally measured values for $A=\mathrm{Ni}$ and Ga. The parameters used are shown above each plot. (c),(d) The dependence of (c) $[(d{T}_N)/(dp)]$ and (d) $\mathrm{\Delta }S$ as a function of $a_4$ and $c_{\mathrm{MV}}$ with $a_2=36\mathrm{meV}$ for both nitride systems. The plot is effectively insensitive to the $a_2$ value for the pertinent range ($33<a_2<37\mathrm{meV}$). The red line indicates the critical curve separating first- and second-order behavior.