Constrained DFT+$U$ approach for understanding the magnetic behaviour of ACr$_{2}$O$_{4}$ (A=Zn, Mg, Cd and Hg) compounds

Abstract: In this work, we try to understand the inconsistency reported by [Yaresko, Phys. Rev. B. {\bf 77}, 115106 (2008)] in the theoretically estimated nature and the variation of magnitude of nearest neighbour exchange coupling constant ($\arrowvert${\it J${1}$}$\arrowvert$) with increasing $U$ in ACr${2}$O${4}$ (A=Zn, Cd, Mg and Hg) compounds by using density functional theory. In unconstrained calculations, the nature and variation of $\arrowvert${\it J${1}$}$\arrowvert$ as a function of $U$ in the present study are not consistent with the experimental data and not according to the relation, {\it J${1}$}$\propto$$\frac{t^{2}}{U}$ especially for CdCr${2}$O${4}$ and HgCr${2}$O${4}$ for $U >$3 eV and U=2-6 eV, respectively. Such an inconsistent behavior of $\arrowvert${\it J${1}$}$\arrowvert$ is almost similar to that of Yaresko for these two compounds for $U$=2-4 eV. For ZnCr${2}$O${4}$ and MgCr${2}$O${4}$, the nature and the variation of $\arrowvert${\it J${1}$}$\arrowvert$ in the present work are in accordance with the experimental data and above mentioned relation for $U$=2-6 eV and are similar to that of Yaresko for ZnCr${2}$O${4}$ for $U$=2-4 eV. However, in constrained calculations the nature and variation of $\arrowvert${\it J${1}$}${c}$$\arrowvert$ in the present work are according to experimental data and above above mentioned relation for all four compounds. Hence, the present study shows the importance of constrained calculations in understanding the magnetic behaviour of these spinels. The values of magnitude of Curie-Weiss temperature [$\arrowvert$($\varTheta$${CW}$)${c}$$\arrowvert$] for ZnCr${2}$O${4}$$>$MgCr${2}$O${4}$$>$CdCr${2}$O${4}$$>$HgCr${2}$O$_{4}$ for $U$=2-5 eV, which are according to the order of experimentally observed values for these spinels.