Reserach Express@NCKU - Articles Digest

Single-molecule magnets (SMMs) have attracted considerable interest because they represent nanoscale magnetic particles with a well-defined size and potential applications in data storage and quantum computation. The major goal is to synthesis new SMMs containing a large effective anisotropy energy barrier U_eff ≈ |D|S², which results from the combination of a large-spin ground state (S) and an Ising-type magnetic anisotropy (negative zero-field splitting parameter D). The first reported SMM [ Mn₁₂O₁₂(O₂CCH₃)₁₆(H₂O)₄ ] (1) has been extensively studied. It possesses a large energy barrier due to its high spin ground state S = 10 and large negative zero-field splitting of -0.5 cm⁻¹. Few high nuclearity manganese SMMs with spin ground states between S = 6 and 83/2 have been reported, however the values of U_eff are relatively small. Thus, many current research objectives for manganese(III) SMMs have been to develop new synthetic approaches, and to achieve the associated large single-ion anisotropy. The end-on bridging azide ligand often mediates ferromagnetic exchange between paramagnetic centres and this property has recently been exploited in the preparation of few SMMs.

Figure 1. ORTEP representation of anion of 2 at 30% probability level.

A new SMM ( NEt₄ )₃[Mn₅O(salox)₃(N₃)₆Cl₂] (2) contains a trigonal bipyramid Mn^II₂Mn^III₃ structure with a large magnetic anisotropy approaching a Ueff of complex 1, even so complex 2 is much smaller than 1. Complex 2 showed AC susceptibility out-of-phase (χM″) peaks in the 2-6 K range and large hysteresis loops below 3.0 K. The compound 2 crystallizes in rhombohedral space group R3c. Crystal structure of anion of complex 2 is shown in Figure 1. The molecular geometry of Mn₅ in 2 is a trigonal bipyramid in which four-coordinate Mn^II ions occupy the apical positions and six-coordinate Mn^III ions reside in the equatorial triangular plane with a capping μ3-O^2- ion. The C₃ axis is perpendicular to the Mn^III3 plane and passes through Mn^II ions and the central oxygen. Each Mn^II ion is linked by three end-on azide bridges to Mn^III centers and a terminal Cl- completes tetrahedral ligation, while the octahedral ligation of each MnIII ions is completed by a bridging η¹:η¹:η¹:μ-salox^2- group, whose phenol ring is bound terminally to a Mn. Bond valence sum (BVS) calculations, and the presence of Mn^III Jahn-Teller (JT) elongation axes establish the oxidation states of manganese and the protonation level of O^2- and salox^2- O atoms. In addition, the JT axes (N2-Mn1-N5) of Mn^III ions are almost parallel to each other as well as to crystallographic C₃ axis. The shortest intermolecular Mn…Mn distances is 8.65 Å.

Figure 2. χMT vs T plot for 2 at 100 G. The solid line represents a least-squares fit of the data. Inset: Plot of reduced magnetization vs H/T between 2 and 4 K. Solid lines represent least-squares fit of the data.

The variable temperature DC magnetic susceptibility data of compound 2 was shown in Figure 2. The value of χMT increase steadily from 18.83 cm³ mol^-1 K at 300 K as the temperature is lowered, to reach a maximum of 47.82 cm³ mol^-1 K at 6.0 K, and then decrease to 46.91 cm³ mol^-1 K at 2.0 K. The χMT value at 300 K is significantly larger than 17.75 cm³ mol^-1 K, the value expected for a Mn^II2Mn^III3 complex with noninteracting metal centers with g = 2.0. This behaviour clearly indicates the ferromagnetic coupling within 2 and the small decreasing in χMT at low temperature is likely the result of Zeeman effect, intermolecular interactions or zero-field splitting in the ground state. In order to describe the coupling within the cluster, the magnetic susceptibility data were fit to the appropriate χM vs T using a Mn^II₂Mn^III₃ Heisenberg-van Vleck model. The fitting result of DC data in 100 G gave the best fit parameters of g = 1.98, J1(Mn^III−Mn^II)) = 0.23 cm^-1, J2(Mn^III−Mn^III) = 2.41 cm^-1. This set of parameters lead to a ground state S_T = 11 and a first excited state S = 10 being closely by at 3.2 cm^-1. To identify the ground state, magnetization (M) data were collected in the 2.0-4.0 K and 10-70 kG ranges (Inset of Figure 2). The best and equally good fits to the data are obtained with S = 11, g = 1.90, and D = -0.22 cm^-1 ( -0.32 K), E = -0.071, and thus, the calculated energy barrier to relaxation (|D|S_z²) is 26.6 cm^-1 (38.3 K).

Figure 3. Plots of out-of-phase (χM″) AC susceptibility versus T in 3.5 G AC field oscillating at indicated frequencies for 2. Inset: Arrhenius law fit of the combined AC and DC data.

To investigate whether 2 might be a SMM, AC susceptibility measurements were performed in a 3.5 G AC field oscillating at 250–1500 Hz and with a zero applied DC field. The frequency dependent amplitude of the in-phase (χM'T) signal, increased as the temperature was lowered, reached a maximum value at 4.0–5.0 K, and finally approached zero. The out-of-phase (χM″) signals showed clear frequency and temperature dependences (Figure 3). As the frequency of the AC field was changed from 1500 to 250 Hz, the χM″ peak shifted from 4.1 to 3.5 K. This frequency dependence of the AC signals suggests that complex 2 is a SMM. Addition relaxation vs time measurements were obtained at temperature below 3.2 K by the DC magnetization decay vs time measurements. This gave a set of relaxation times, which were combined with the AC data and used to construct an Arrhenius plot. Good fits of the combined AC and DC data allowed us to obtain 0 = 2.6 x 10^-7 s and U_eff = 40.3 K (Inset of Figure 3).

In order to probe the anisotropy and quantum tunneling magnetization (QTM) of complex 2, single crystal hysteresis loops and relaxation measurements were performed by using a micro-SQUID setup. Figure 4 presents magnetization (M) vs H measurements. The hysteresis loops were strongly sweep rate and temperature dependent, and showed steps indicative of QTM. The hysteresis loops become temperature independent below 0.8 K establishing tunneling between ground state levels. The anisotropy of the spin ground state is obtained from H between the zero-field step and the first step at 1.1 T yielding a |D|/g value of 0.51 cm^-1 . This extreme large anisotropy |D|/g = 0.51 cm^-1 is coming from the almost parallel arrangement of JT axes of Mn^III ions in the equatorial triangular plane. At temperatures above about 1 and 1.5 K, other steps appears at 1.2 and 0.6 T, respectively, which are probably due to tunneling via the first excited state being rather close to the spin ground state. The steps at 0.6 and 1.2 T suggest |D|/g = 0.28 cm^-1 for the first excited spin state. Because the latter is very close to the spin ground state, the maximal barrier of |D|S² is strongly reduced to U_eff = 40.3 K.

In conclusion, complex 2 represents a new example of SMMs, with the anisotropy energy U_eff of 40.3 K. QTM was observed and allowed us to estimate the anisotropy parameter of the two lowest spin states.

Figure 4. Magnetization hysteresis loops for a single crystal of 2; (top) from 3.0 to 0.04 K at 0.14 Ts^-1 scan rate; (bottom) for different scan rates at T = 0.04 K.