Magnetization steps in the diluted Heisenberg layer materials(CH3NH3)2MnxCd1−xCl4: Equilibrium data at 0.6 K

Abstract

The magnetization $M$ of ${({\mathrm{C}\mathrm{H}}_{3}{\mathrm{N}\mathrm{H}}_{3})}_{2}{\mathrm{Mn}}_{x}{\mathrm{Cd}}_{1\ensuremath{-}x}{\mathrm{Cl}}_{4}$, with $x$ from 0.025 up to 0.265, was measured at 0.6 K in a slowly varying magnetic field $B$ up to 17 T. The exchange interaction in these strongly diluted planar magnetic materials is antiferromagnetic. The in-plane cation structure is well approximated by a square lattice. The observed qualitative features, listed in the order that they appear in increasing $B$, are as follows: a fast rise of $M$, starting at $B=0$; a magnetization plateau (plateau of ``apparent saturation''); a large magnetization step (MST), attributed to nearest-neighbor (NN) pairs; a second magnetization plateau; another large MST from NN pairs; and a third plateau that is not completed below the highest available $B$. These features are expected from the NN cluster model presented in the preceding paper. The magnetic fields at the two MST's give ${J}_{1}∕{k}_{\mathrm{B}}=(\ensuremath{-}4.39\ifmmode\pm\else\textpm\fi{}0.10)\phantom{\rule{0.3p}{0ex}}\mathrm{K}$ for the NN exchange constant. This value is slightly lower than reported for the undiluted $(x=1)$ member of this series, ${({\mathrm{C}\mathrm{H}}_{3}{\mathrm{N}\mathrm{H}}_{3})}_{2}\mathrm{Mn}{\mathrm{Cl}}_{4}$. A smaller ${J}_{1}$ when $x\ensuremath{\leqslant}0.265$ may be the result of an in-plane expansion with decreasing $x$, caused by the slightly larger ${\mathrm{Cd}}^{2+}$ ion compared to ${\mathrm{Mn}}^{2+}$. Analysis of the initial rise of $M$ at low $B$ indicates the presence of weak interactions that are not included in the NN cluster model. This conclusion is consistent with the observation (to be reported later) of a weak exchange interaction with a neighbor that is more distant than a NN. The apparent saturation value ${M}_{s}$, at the first magnetization plateau, was determined for all seven samples. There is a fair agreement with the values expected from a random distribution of the Mn ions over all cation sites. The largest deviation is for samples with $x\ensuremath{\geqslant}0.15$, where the measured ${M}_{s}$ is somewhat higher. In the same samples the magnetization jump $\ensuremath{\Delta}M$ at the MST's from NN pairs is somewhat smaller than for a random Mn distribution. A proposed explanation of the discrepancies for $x\ensuremath{\geqslant}0.15$ postulates that the probability that a cation site is occupied by a ${\mathrm{Mn}}^{2+}$ is lowered by the presence of other ${\mathrm{Mn}}^{2+}$ ions at one or more NN cation sites.