Comparison of local structure of ${\mathrm{CrCl}}_3$ bulk and nanocrystals above and below the structural phase transition

At least since the discovery of graphene and the subsequent finding of a plethora of other 2D materials, it is well anticipated that the dimensionality of a material may constitute a functional parameter. In this paper, we discuss zero-field ${}_{53}\mathrm{Cr}$ nuclear magnetic resonance (NMR) measured in the magnetically ordered state and ${}_{35}\mathrm{Cl}$ nuclear quadruple resonance (NQR) data derived in the paramagnetic state of the two-dimensional van der Waals material ${\text{CrCl}}_3$, comparing the results for a bulk single crystal and a nanocrystal. In particular, we apply these spectroscopic methods to monitor the evolution of local environments in the single crystal across the structural phase transition and compare the structural and magnetic properties of a bulk single crystal and nanocrystal sample at low temperatures. The actual structural transition is reported to be of first order, where a certain hysteresis is to be expected. However, we see that both the high- and low-temperature phases coexist in both sample types across the full temperature range (300 K–1.5 K) albeit with different phase fractions. This coexistence of phases in different sample types originates in a kinetic arrest where the arrested structural domains are related to defects and stacking faults. Such defects are to a large part found in the nanocrystal but to a smaller extent in the bulk single crystal. These frozen-in phases have further consequences: The critical exponent $\beta$, derived by fitting the ${}_{53}\mathrm{Cr}$ NMR data, is considered to denote the dimensionality of magnetic interactions. Here, the values differ considerably for both sample types. Probably, the difference in $\beta$ arises from the specific domain structure of kinetically arrested phases and, in turn, from an altered interlayer magnetic exchange mediated by magnetic moments related to frozen-in domains that are related to defects and stacking faults,with their number being much higher in the nanocrystal. These findings may, in part, explain the different magnetic properties reported for different samples as their individual defect landscape determines the kinetically arrested phase fraction in ${\mathrm{CrCl}}_3$. Hence, the structure-property relation in ${\mathrm{CrCl}}_3$ is even more complex than anticipated.