Dimensionality Selection in a Molecule-Based Magnet

Gaining control of the building blocks of magnetic materials and thereby achieving particular characteristics will make possible the design and growth of bespoke magnetic devices. While progress in the synthesis of molecular materials, and especially coordination polymers, represents a significant step towards this goal, the ability to tune the magnetic interactions within a particular framework remains in its infancy. Here we demonstrate a chemical method which achieves dimensionality selection via preferential inhibition of the magnetic exchange in an $S=1/2$ antiferromagnet along one crystal direction, switching the system from being quasi-two- to quasi-one-dimensional while effectively maintaining the nearest-neighbor coupling strength.

Figure 1

(a) View of the crystal structure of the planar material $[\mathrm{Cu}(\mathrm{pyz}{)}_2(\mathrm{pyO}{)}_2]({\mathrm{PF}}_6{)}_2$ determined using x-ray diffraction showing the 2D Cu—pyrazine network in the $ab$ plane. (b) A projection of the same structure along the $\mathbf{a}$ axis highlighting the shift between adjacent Cu—pyrazine layers and the arrangement of the ${\mathrm{PF}}_6$ counterions. (c) Crystal structure of the chainlike material $[\mathrm{Cu}(\mathrm{pyz})(\mathrm{pyO}{)}_2({\mathrm{H}}_2\mathrm{O}{)}_2]({\mathrm{PF}}_6{)}_2$ showing the 1D Cu—pyrazine chains in the $ab$ plane, and (d) a projection along the chains showing the arrangement of the pyO and ${\mathrm{H}}_2\mathrm{O}$ ligands and the ${\mathrm{PF}}_6$ counterions. The dashed lines in (c) and (d) enclose one unit cell. Cu, brown; C, grey; N, blue; O, red; P, orange; F, green. Print edition: In each diagram, the largest sphere denotes the spin-carrying Cu ion [21]. Hydrogen atoms other than those in the water molecules have been omitted for clarity.

Figure 2

(a) Example muon-spin relaxation (${\mu }^{+}\mathrm{SR}$) spectra measured on planar $[\mathrm{Cu}(\mathrm{pyz}{)}_2(\mathrm{pyO}{)}_2]({\mathrm{PF}}_6{)}_2$. Data at different temperatures are offset for clarity. Inset: Evolution of the precession frequency $\nu$ with temperature. Long-range magnetic order is observed below 1.71 K. (b) ${\mu }^{+}\mathrm{SR}$ spectra measured on chainlike $[\mathrm{Cu}(\mathrm{pyz})(\mathrm{pyO}{)}_2({\mathrm{H}}_2\mathrm{O}{)}_2]({\mathrm{PF}}_6{)}_2$. Insets: The evolution of the amplitude $A_{\parallel }$ (left) and relaxation rate $\lambda$ (right) with temperature indicates the onset of long-range magnetic order below about 0.27 K. Red lines are fits to functions described in the text and 21.

Figure 3

(a) Normalized pulsed-field magnetization of planar $[\mathrm{Cu}(\mathrm{pyz}{)}_2(\mathrm{pyO}{)}_2]({\mathrm{PF}}_6{)}_2$ at $T=1.5\mathrm{K}$, and chainlike $[\mathrm{Cu}(\mathrm{pyz})(\mathrm{pyO}{)}_2({\mathrm{H}}_2\mathrm{O}{)}_2]({\mathrm{PF}}_6{)}_2$ at $T=0.5\mathrm{K}$. (b) The results of QMC simulations of the low-temperature magnetization for 3D ($J_{\perp }/J=1$), 2D ($J_{\perp }/J=0$), and 1D ($J_{\perp }/J=0.02$) antiferromagnets. The lines are the pulsed-field data scaled by the saturation field ($B_c$). (c) The relation between the exchange anisotropy and the ratio of critical temperature and primary exchange energy in Q1D and Q2D antiferromagnets deduced from QMC simulations [28]. The circles indicate the materials reported here. (d) Anisotropy of the $g$ factor in the planar and (e) chainlike compounds measured using ESR at 10 K and 1.5 K, respectively. Red lines are fits to $g(\theta )=(g_{xy}^2{sin}^2\theta +g_z^2{cos}^2\theta {)}^{1/2}$.