A family of fourteen soluble stable macrocyclic [Ni^II₃Ln^III] heterometallic 3d–4f complexes

Abstract

A family of fourteen tetranuclear, 3d–4f heterometallic nickel(II)–lanthanide(III) complexes of the hexaimine macrocycle (L^Pr)⁶⁻, with general formula Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents (Ln^III = La^III, Ce^III, Pr^III, Nd^III, Sm^III, Eu^III, Gd^III, Tb^III, Dy^III, Ho^III, Er^III, Tm^III, Yb^III or Lu^III), were prepared in a one-pot synthesis using a 3 : 1 : 3 : 3 reaction of nickel(II) acetate, the appropriate lanthanide(III) nitrate, the dialdehyde 1,4-diformyl-2,3-dihydroxybenzene (H₂L^Ald) and 1,3-diaminopropane. In addition, three tetranuclear heterometallic nickel(II)–lanthanide(III) complexes of H₂L^Ald, with general formula Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents, were deliberately prepared (Ln^III = La^III, Dy^III or Yb^III) as in effect they represent intermediates en route to the above macrocyclic complexes. Whilst single crystals of the macrocyclic complexes were not forthcoming, X-ray crystal structure determinations on Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents (Ln^III = Dy^III or Yb^III) confirmed that the large ten-coordinate lanthanide(III) ion is bound in the central O₆ pocket while the smaller six-coordinate nickel(II) ions are bound in the outer O₄ pockets. In all fourteen cases, addition of the diamine to this intermediate (all in one pot) gives the tetrametallic [3 + 3] macrocyclic product. The magnetic properties of all fourteen macrocyclic complexes were measured down to 1.8 K to check for Single-Molecule Magnet behaviour, but no slow dynamics of magnetisation was observed.

---

Introduction

Single-Molecule Magnets (SMMs) are coordination complexes in which the individual molecules behave like traditional ‘bulk’ magnets.¹ As such they can exhibit slow relaxation of magnetisation after removal of the magnetising field, which may make them useful in nano-scale data storage and processing. This is usually the result of a non-zero spin ground state and significant uniaxial anisotropy of that state which induces an energy barrier to reversing the magnetisation. However, due to the small energy barrier between these two states (in comparison to the temperature) and/or quantum tunnelling of the magnetization, a hysteresis effect in magnetisation (M) vs. field (H) plots is not typically seen above 5 K, with the record currently at 14 K.² Clearly, for any practical device application of SMMs, the barrier must be raised.

A variety of complexes have been shown to display SMM behaviour, including systems based on transition metal,³ lanthanide,^4–6 actinide⁷ and mixed-metal^6,8 systems. Combining transition metal and lanthanide ions is also common. However, the majority of SMMs are prepared with polydentate acyclic ligands, which can make characterisation somewhat dependent on growing single crystals, as the structure can be difficult to predict with any certainty. Moreover, small modifications to one aspect of the system may make a dramatic change to the structure of the resultant complex, which can make it impossible to tune the magnetic properties.

An alternative approach is to use a macrocyclic ligand, which allows predictable binding in discrete binding pockets. A macrocyclic ligand also confers stability and helps to prevent undesirable inter-molecule interactions. The advantage of using a macrocyclic ligand to prepare SMMs was first illustrated by Ishikawa, who has demonstrated that a single lanthanide ion can show SMM behaviour when sandwiched between two phthalocyanine macrocycles in an N₈ donor environment.⁹ More examples of macrocyclic SMMs containing a single lanthanide ion have been reported in the literature since then,¹⁰ including organometallic species where the single lanthanide ion is sandwiched by the π systems of two cyclic, anionic ligands.⁵

Recently, we expanded the macrocyclic approach to prepare, in a controlled and predictable manner, mixed 3d–4f [M^II₃Ln^III] SMMs, using the [3 + 3] macrocycle (L^Pr)⁶⁻ which features distinct binding pockets tailored for binding to M^II (transition metal) vs. Ln^III (lanthanide) ions (Fig. 1).^11–14 Nabeshima, Kajiwara and co-workers have also utilised this approach using the macrocycle (L^Ph)⁶⁻ (Fig. 1).¹⁵ More recently we have gone on to demonstrate that linking the axial sites of the robust and soluble SMM building block [Cu^II₃Tb^III(L^Pr)]³⁺ generates, as anticipated, magnetically interesting chains.¹⁶ Pleasingly, we have also found that the template method used to prepare (L^Pr)⁶⁻ has proven to be remarkably robust, likewise working for the related, larger macrocycle (L^Bu)⁶⁻ (Fig. 1).¹⁷ We now further demonstrate the versatility of this system by reporting the synthesis of the nickel(II) series, [Ni^II₃Ln^III(L^Pr)], and selected intermediates, [Ni^II₃Ln^III(L^Ald)₃], along with the magnetic properties of all 14 macrocyclic complexes.

Fig. 1 Schematic representation of the [3 + 3] macrocycles (L^Pr)⁶⁻, (L^Bu)⁶⁻ and (L^Ph)⁶⁻, all of which are obtained as metal complexes.

Experimental

General

All chemicals were of A.R. Grade and used as received. A working synthesis of H₂L^Ald was developed at Otago University, based on the outline reported by Nabeshima and co-workers¹⁸ and with advice from Professor M. J. MacLachlan (UBC). This is practically identical to the detailed method that was subsequently published by Nabeshima and co-workers¹⁹ so is not repeated here. IR spectra were recorded as solids on a Bruker Alpha FT-ATR IR spectrometer with a diamond anvil Alpha-P module between 400 and 4000 cm⁻¹. Microanalysis was carried out by the Campbell Microanalytical Laboratory at the University of Otago. Metal analysis was carried out by the Centre for Trace Metal Analysis at the University of Otago. Conductivity measurements were carried out at 25 °C on a Mettler Toledo FE20 FiveEasy Conductivity Meter. Thermogravimetric analysis (TGA) was performed on a TA Instruments Q50 TGA, using nitrogen as a flow gas.

Single crystal X-ray measurements were carried out on a Bruker SMART Apex diffractometer using graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å). Structure solution and full-matrix least-squares refinement were carried out using the SHELXTL software package.²⁰ All non-H atoms were refined anisotropically except for the disordered atoms in [Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(MeOH)_4.75(H₂O)_1.25](NO₃). H atoms bonded to C atoms were placed in calculated positions whilst those on O atoms were placed in calculated positions that gave the best hydrogen bonds. Due to a non-coordinated nitrate anion being very badly disordered in both structures, SQUEEZE²¹ was used to account for it. For [Ni^II₃Dy^III(L^Ald)₃(NO₃)₂(MeOH)₆](NO₃) and [Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(H₂O)_1.25(MeOH)_4.75](NO₃) the voids contained 187 and 184 electrons per cell, respectively, both of which are consistent with one nitrate anion and one water molecule per asymmetric unit (expect 164 electrons per cell). For further details see the cif files.

Crystal data for [Ni^II₃Dy^III(L^Ald)₃(NO₃)₂(MeOH)₆]⁺ (NO₃⁻ counter ion accounted for by SQUEEZE). Monoclinic, P2₁/n, red block, a = 14.0050(11) Å, b = 13.0327(11) Å, c = 24.2807(19) Å, β = 90.976(4)°, V = 4431.1(6) ų, Z = 4, T = 92(2) K. The structure was solved by direct methods (SHELXS-97)²⁰ and refined against all 9052 unique F² data (SHELXL)²⁰ to wR² = 0.1020, GooF = 1.161; for the 7869 data with F > 4σ(F), R¹ = 0.0448. CCDC 913629.

Crystal data for [Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(H₂O)_1.25(MeOH)_4.75]⁺ (NO₃⁻ counter ion accounted for by SQUEEZE). Monoclinic, P2₁/n, red block, a = 13.9687(6) Å, b = 13.0219(6) Å, c = 24.2144(11) Å, β = 90.8930(10)°, V = 4404(3) ų, Z = 4, T = 92(2) K. The structure was solved by direct methods (SHELXS-97)²⁰ and refined against all 9019 unique F² data (SHELXL)²⁰ to wR² = 0.1524, GooF = 1.169; for the 8511 data with F > 4σ(F), R¹ = 0.0560. CCDC 913628.

The magnetic susceptibility measurements were obtained with the use of MPMS-XL Quantum Design SQUID magnetometer. The magnetometer operates between 1.8 and 400 K for dc applied fields of up to 70 kOe. Measurements were performed on polycrystalline samples of 13.27, 14.95, 13.54, 17.16, 14.35, 18.06, 8.34 (and 12.25), 12.81, 13.05 (and 19.71), 12.34, 17.92, 11.10, 14.65 and 8.81 (and 9.87) mg respectively for La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb and Lu introduced in a sealed polyethylene bag (3 × 0.5 × 0.02 cm). The ac susceptibility measurements were measured with an oscillating ac field of 3 Oe with frequency between 1 to 1500 Hz (MPMS). The magnetic data were corrected for the sample holder and the diamagnetic contributions (the latter estimated as −0.5 × 10⁻⁶ × M_r then adjusted in the fit).

Synthesis

General synthesis of Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents. To a solution of H₂L^Ald (0.030 g, 0.18 mmol) in MeOH (5 mL) was added a solution of Ni^II(OAc)₂·4H₂O (0.044 g, 0.18 mmol) and Ln^III(NO₃)₃·xH₂O (0.06 mmol, Ln^III = La^III, Dy^III or Yb^III) in MeOH (10 mL), resulting in a dark red solution. The solution was stirred for one hour. Vapour diffusion of Et₂O into the reaction solution yielded crystals which were filtered and washed with Et₂O (2 × 5 mL) giving Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents as red powder after drying in air.

Ni^II₃La^III(L^Ald)₃(NO₃)₃·3H₂O·5MeOH. Red powder (0.037 g, 51%). Found: C, 28.65; H, 2.98; N, 3.43. Calculated for Ni^II₃La^IIIC₂₉H₃₈O₂₉N₃ (1213.65 g mol⁻¹): C, 28.84; H, 3.17; N, 3.48. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1614 (s); 1489 (m); 1447 (s); 1413 (m); 1366 (m); 1329 (m); 1245 (s); 1159 (s); 1104 (s); 1030 (m); 834 (s); 796 (w); 770 (w); 736 (w); 703 (s); 596 (s); 561 (s); 521 (m).

[Ni^II₃Dy^III(L^Ald)₃(NO₃)₂(H₂O)₆](NO₃)·H₂O. Red powder (0.031, 45%). Found: C, 25.38; H, 2.32; N, 3.53. Calculated for Ni^II₃Dy^IIIC₂₄H₂₆O₂₈N₃ (1149.09 g mol⁻¹): C, 25.22; H, 2.29; N, 3.68. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1616 (s); 1486 (m); 1445 (s); 1418 (m); 1370 (m); 1332 (m); 1246 (s); 1160 (s); 1112 (s); 1032 (m); 834 (s); 783 (w); 768 (w); 741 (w); 700 (s); 609 (s); 560 (s); 526 (m).

[Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(H₂O)₆](NO₃)·3H₂O. Red powder (0.034, 50%). Found: C, 24.52; H, 2.37; N, 3.33. Calculated for Ni^II₃Yb^IIIC₂₄H₃₀O₃₀N₃ (1195.66 g mol⁻¹): C, 24.23; H, 2.54; N, 3.53. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1618 (s); 1491 (m); 1449 (s); 1414 (m); 1369 (m); 1333 (m); 1291 (m); 1248 (s); 1209 (m); 1159 (s); 1111 (s); 1033 (m); 835 (s); 796 (w); 767 (w); 703 (s); 597 (s); 560 (s); 527 (m).

General synthetic method for Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents. To a solution of H₂L^Ald (0.050 g, 0.3 mmol) in methanol (5 mL) was added Ni^II(OAc)₂·4H₂O (0.045 g, 0.18 mmol) suspended in a solution of Ln^III(NO₃)₃·xH₂O (0.1 mmol, Ln^III = La^III, Ce^III, Pr^III, Nd^III, Sm^III, Eu^III, Gd^III, Tb^III, Dy^III, Ho^III, Er^III, Tm^III or Yb^III or Lu^III) in MeOH (10 mL), resulting in a red solution. After stirring this solution for one hour, 1,3-diaminopropane (3 mL of 0.1 mol L⁻¹ MeOH solution, 0.3 mmol) was added dropwise, resulting in no change. This solution was stirred for a further thirty minutes before being left to stand overnight. Diethyl ether was then vapour diffused into the solution. The resulting solid was filtered and washed with Et₂O (2 × 5 mL), giving Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents as red powder after drying in air.

Ni^II₃La^III(L^Pr)(NO₃)₃·11H₂O. Further recrystallised from Et₂O/MeOH. Red powder (0.029 g, 37%). Found: C, 30.03; H, 3.72; N, 9.48. Calculated for Ni^II₃La^IIIC₃₃H₅₂N₉O₂₆ (1305.80 g mol⁻¹): C, 30.35; H, 4.01; N, 9.65. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1624 (m); 1507 (w); 1446 (m); 1411 (m); 1321 (s); 1230 (m); 1168 (m); 1084 (w); 1027 (w); 832 (w) 770 (m); 718 (m); 657 (w).

Ni^II₃Ce^III(L^Pr)(NO₃)₃·7.5H₂O·4MeOH. Dark red powder (0.055 g, 40%). Found: C, 32.34; H, 4.48; N, 9.15. Calculated for Ni^II₃Ce^IIIC₃₇H₆₁N₉O_26.5 (1372.91 g mol⁻¹): C, 32.39; H, 4.48; N, 9.19. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1617 (m); 1518 (w); 1453 (m); 1410 (m); 1315 (s); 1240 (m); 1194 (m); 1171 (m); 1089 (w); 1055 (m); 1035 (w); 965 (m); 834 (w) 780 (m); 750 (m); 733 (m); 658 (m).

Ni^II₃Pr^III(L^Pr)(NO₃)₃·9H₂O·2.5MeOH. Dark red powder (0.058 g, 43%). Found: C, 31.46; H, 4.20; N, 9.18. Calculated for Ni^II₃Pr^IIIC_35.5H₅₈N₉O_26.5 (1351.87 g mol⁻¹): C, 31.54; H, 4.32; N, 9.32. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1624 (m); 1507 (w); 1449 (m); 1407 (m); 1323 (s); 1230 (m); 1170 (m); 1126 (m); 1077 (w); 1021 (w); 828 (w) 769 (m); 718 (m); 657 (w).

Ni^II₃Nd^III(L^Pr)(NO₃)₃·8H₂O·3MeOH. Red powder (0.037 g, 27%). Found: C, 31.86; H, 4.41; N, 9.21. Calculated for Ni^II₃Nd^IIIC₃₆H₅₈N₉O₂₆ (1353.21 g mol⁻¹): C, 31.95; H, 4.32; N, 9.32. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1624 (m); 1508 (w); 1449 (m); 1406 (m); 1323 (s); 1230 (m); 1170 (m); 1126 (m); 1076 (w); 1021 (w); 832 (w) 769 (m); 720 (m); 656 (w).

Ni^II₃Sm^III(L^Pr)(NO₃)₃·6.5H₂O·2.5MeOH. Red powder (0.067 g, 51%). Found: C, 32.44; H, 4.17; N, 9.67. Calculated for Ni^II₃Sm^IIIC_35.5H₅₃N₉O₂₄ (1316.29 g mol⁻¹): C, 32.39; H, 4.06; N, 9.58. Λ_m (1 mM in DMF) = 111 mol⁻¹ cm² Ω⁻¹. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1622 (m); 1507 (w); 1451 (m); 1404 (m); 1321 (s); 1230 (m); 1172 (m); 1125 (m); 1075 (w); 1023 (w); 835 (w) 765 (m); 723 (m); 655 (w).

Ni^II₃Eu^III(L^Pr)(NO₃)₃·8.5H₂O·2MeOH. Red powder (0.058 g, 43%). Found: C, 31.35; H, 4.15; N, 9.43. Calculated for Ni^II₃Eu^IIIC₃₅H₅₅N₉O₂₅.₅ (1337.84 g mol⁻¹): C, 31.42; H, 4.14; N, 9.42. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1621 (m); 1508 (w); 1449 (m); 1411 (m); 1323 (s); 1231 (m); 1168 (m); 1085 (w); 1027 (w); 833 (w) 771 (m); 719 (m); 659 (w).

Ni^II₃Gd^III(L^Pr)(NO₃)₃·7.5H₂O. Prepared on a 0.1 g scale of H₂L^Ald, to allow for additional characterisation. Further recrystallised from Et₂O/MeOH. Red powder (0.145 g, 57%). Found: C, 31.00; H, 3.74; N, 10.07; Ni, 13.0; Gd, 11.4. Calculated for Ni^II₃Gd^IIIC₃₃H₄₅N₉O_22.5 (1261.09 g mol⁻¹): C, 31.43; H, 3.60; N, 10.00; Ni, 13.9; Gd, 12.4. TGA found weight loss: 9.7%; calculated for loss of 7.5H₂O: 10.7%. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1624 (m); 1508 (w); 1449 (m); 1400 (m); 1321 (s); 1218 (m); 1169 (m); 1125 (m); 1074 (w); 1027 (w); 833 (w); 765 (m); 720 (m); 654 (w).

Ni^II₃Tb^III(L^Pr)(NO₃)₃·7H₂O·2MeOH. Further recrystallised from Et₂O/MeOH. Red powder (0.052 g, 39%). Found: C, 31.66; H, 3.91; N, 9.18. Calculated for Ni^II₃Tb^IIIC₃₅H₅₂N₉O₂₄ (1317.84 g mol⁻¹): C, 31.90; H, 3.98; N, 9.57. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1621 (m); 1510 (w); 1450 (m); 1407 (m); 1322 (s); 1230 (m); 1170 (m); 1085 (w); 1028 (w); 832 (w) 769 (m); 719 (m); 657 (w).

Ni^II₃Dy^III(L^Pr)(NO₃)₃·6H₂O·3MeOH. Further recrystallised from Et₂O/MeOH. Red powder (0.040 g, 30%). Found: C, 32.02; H, 3.84; N, 9.19. Calculated for Ni^II₃Dy^IIIC₃₆H₅₄N₉O₂₄ (1335.44 g mol⁻¹): C, 32.38; H, 4.08; N, 9.44. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1622 (m); 1508 (w); 1450 (m); 1407 (m); 1324 (s); 1229 (m); 1171 (m); 1131 (m); 1085 (w); 1025 (w); 893 (w); 833 (w); 770 (m); 721 (m); 657 (w).

Ni^II₃Ho^III(L^Pr)(NO₃)₃·8.5H₂O·0.5MeOH. Red powder (0.043 g, 33%). Found: C, 30.98; H, 4.08; N, 10.00. Calculated for Ni^II₃Ho^IIIC_33.5H₄₉N₉O₂₄ (1302.80 g mol⁻¹): C, 30.88; H, 3.79; N, 9.68. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1617 (m); 1508 (w); 1449 (m); 1408 (m); 1320 (s); 1218 (m); 1170 (m); 1122 (m); 1071 (w); 1026 (w); 898 (w); 827 (w); 769 (m); 714 (m); 669 (w).

Ni^II₃Er^III(L^Pr)(NO₃)₃·7.5H₂O·3.5MeOH. Red powder (0.074 g, 53%). Found: C, 31.67; H, 4.18; N, 9.02. Calculated for Ni^II₃Er^IIIC_36.5H₅₉N₉O₂₆ (1383.25 g mol⁻¹): C, 31.69; H, 4.30; N, 9.11. Λ_m (1 mM in DMF) = 102 mol⁻¹ cm² Ω⁻¹. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1623 (m); 1508 (w); 1451 (m); 1406 (m); 1324 (s); 1229 (m); 1171 (m); 1126 (m); 1085 (w); 1026 (w); 893 (w); 834 (w); 770 (m); 721 (m); 659 (w).

Ni^II₃Tm^III(L^Pr)(NO₃)₃·10H₂O·MeOH. Red powder (0.0315 g, 52%). Found: C, 30.00; H, 4.21; N, 9.31; Ni, 11.8; Tm, 12.2. Calculated for Ni^II₃Tm^IIIC₃₄H₅₄N₉O₂₆ (1349.84 g mol⁻¹): C, 30.25; H, 4.03; N, 9.34; Ni, 13.0; Tm, 12.5. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1621 (m); 1508 (w); 1450 (s); 1323 (s); 1253 (m); 1228 (s); 1173 (m); 1085 (w); 1021 (w); 846 (w); 827 (w); 813 (w); 772 (w); 720 (m).

Ni^II₃Yb^III(L^Pr)(NO₃)₃·10H₂O. Red powder (0.037 g, 28%) Found: C, 29.89; H, 4.02; N, 9.84. Calculated for Ni^II₃Yb^IIIC₃₃H₅₀N₉O₂₅ (1321.92 g mol⁻¹): C, 29.98; H, 3.81; N, 9.53. Λ_m (1 mM in DMF) = 121 mol⁻¹ cm² Ω⁻¹. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1626 (m); 1511 (w); 1455 (m); 1404 (m); 1324 (s); 1230 (m); 1173 (m); 1127 (m); 1086 (w); 1027 (w); 895 (w); 827 (w); 771 (m); 721 (m); 659 (w).

Ni^II₃Lu^III(L^Pr)(NO₃)₃·7H₂O. Prepared on a 0.1 g scale of H₂L^Ald, to allow for additional characterisation. Further recrystallised from Et₂O/MeOH. Red powder (0.09 g, 36%). Found: C, 30.71; H, 3.53; N, 9.87; Ni, 14.0; Lu, 13.8. Calculated for Ni^II₃Lu^IIIC₃₃H₄₄N₉O₂₂ (1269.80 g mol⁻¹): C, 31.21; H, 3.49; N, 9.93; Ni, 13.9; Lu, 13.8. TGA found weight loss: 9.9%; calculated for loss of 7H₂O: 9.9%. IR (FT-ATR diamond anvil) [small nu, Greek, macron]/cm⁻¹ = 1624 (m); 1526 (w); 1457 (s); 1323 (s); 1229 (s); 1176 (m); 1086 (w); 1025 (w); 846 (w); 827 (w); 772 (w); 720 (m).

Results and discussion

Synthesis of the complexes

The fourteen macrocyclic complexes are all prepared in the same manner. Three equivalents of Ni^II(OAc)₂·4H₂O and one equivalent of Ln^III(NO₃)₃·xH₂O (Ln = La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb or Lu) in methanol are added to a room temperature solution of three equivalents of H₂L^Ald in methanol, forming the ‘intermediate’ complex in situ (Fig. 2). After stirring for one hour, three equivalents of 1,3-diaminopropane in methanol are added dropwise to this solution. After stirring overnight, the solution is diffused with diethyl ether vapour, resulting in the precipitation of a red solid. This solid was isolated, washed with diethyl ether and dried in air, giving complexes of general formula Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents, referred to hereon by the abbreviated form [Ni^II₃Ln^III(L^Pr)].

Fig. 2 Schematic representation of formation of the macrocyclic complex via the ‘intermediate’ complex.

The formation of the desired macrocyclic complexes was confirmed by elemental analysis, TGA and IR spectroscopy. In all cases the C=N stretch of the macrocycle occurred at 1621 cm⁻¹, while no absorptions for unreacted H₂L^Ald (1660 cm⁻¹) or 1,3-diaminopropane (3360 cm⁻¹ and 1601 cm⁻¹) were observed. The presence of nitrate anions was supported by the pronounced band at around 1315–1325 cm⁻¹. The CHN elemental analysis data matched the calculated values expected for complexes of the formula Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents and were corroborated by nickel and lanthanide metal analysis on three representative samples (Ln^III = Gd^III, Tm^III and Lu^III) which confirmed the presence of one lanthanide(III) ion and three nickel(II) ions in all cases. Similarly, thermogravimetric analysis on two representative samples (Ln^III = Gd^III and Lu^III) confirmed the presence of the solvent content determined from the microanalysis data.

Conductivity values on three representative samples in DMF solution were between 102–120 mol⁻¹ cm² Ω⁻¹, which is slightly below the expected range²⁷ (130–170 mol⁻¹ cm² Ω⁻¹) for 2 : 1 electrolytes (note: literature ranges for 1 : 1 and 3 : 1 electrolytes in DMF are 65–90 and 200–240 mol⁻¹ cm² Ω⁻¹, respectively). This is likely due to the low mobility of the large [Ni^II₃Ln^III(L^Pr)(NO₃)_x]^(3−x)+ cations in solution (e.g. as observed for salts of the bulky tetraphenylborate anion²⁷). ESI-MS was not helpful in establishing that the desired complexes had formed, due to extensive fragmentation.

The very broadened peaks seen in the ¹H NMR spectrum of Ni^II₃La^III(L^Pr)(NO₃)₃·11H₂O were consistent with paramagnetic octahedral nickel(II) ions in DMF-d₇ solution. Although nickel(II) is typically diamagnetic with a square-planar geometry when in the N₂O₂ pocket of salen- or salpn-like ligands,²² an octahedral geometry (paramagnetic) can be observed if the phenolate oxygen atoms are also coordinated to a second metal ion, such as zinc(II), due to reducing the electron density on these oxygen donor atoms.²³ Here, the trivalent lanthanide(III) ion removes electron density from the catecholate oxygen atoms, increasing the tendency of the nickel(II) ion to coordinate axial donors (consistent with the magnetic data, see later). In the related Robson macrocyclic system,^24–26 salpn-like pockets bind nickel(II) in an octahedral fashion,²⁴ whereas the smaller macrocycles with tighter salen-like pockets provide a strong enough field to always generate square-planar nickel(II), even when dissolved in pyridine.²⁵ It should be noted that the ¹H NMR spectrum of Ni^II₃La^III(L^Ald)₃(NO₃)₃·3H₂O·5MeOH showed a similar paramagnetically broadened spectrum, which precluded determining if the intermediate complex is present in solution.

These macrocyclic complexes are soluble and stable so can be recrystallised from MeOH (or DMF) by vapour diffusion of diethyl ether, but despite numerous attempts no single crystals were forthcoming. The intermediate complexes (Fig. 2), Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents, can be isolated by not adding the 1,3-diaminopropane and diffusing diethyl ether vapour into the reaction solution as usual. Three such complexes (Ln^III = La^III, Dy^III or Yb^III) were prepared, and the structures of two of them were determined.

Crystal structures

Single crystals of [Ni^II₃Dy^III(L^Ald)₃(NO₃)₂(MeOH)₆](NO₃) were grown by vapour diffusion of diethyl ether into a solution of the complex in methanol (Fig. 3, Table 1).

Fig. 3 Perspective view of the crystal structure of the cation [Ni^II₃Dy^III(L^Ald)₃(NO₃)₂(MeOH)₆]⁺. For clarity, non-acidic hydrogen atoms have been omitted. Selected bond lengths (Å): Dy(1)–O(2), 2.500(4); Dy(1)–O(3), 2.487(3); Dy(1)–O(6), 2.478(3); Dy(1)–O(7), 2.486(4); Dy(1)–O(10), 2.486(4); Dy(1)–O(11), 2.496(4); Dy(1)–O(30), 2.478(5); Dy(1)–O(31), 2.514(5); Dy(1)–O(40), 2.460(5); Dy(1)–O(41), 2.459(5); Ni(1)–O(1), 2.008(4); Ni(1)–O(2), 1.979(4); Ni(1)–O(11), 1.975(4); Ni(1)–O(12), 2.006(4); Ni(1)–O(60), 2.050(5); Ni(1)–O(70), 2.085(5); Ni(2)–O(3), 1.968(4); Ni(2)–O(5), 2.014(4); Ni(2)–O(6), 1.988(4); Ni(2)–O(80), 2.067(5); Ni(2)–O(90), 2.098(6); Ni(3)–O(7), 1.970(4); Ni(3)–O(8), 2.004(4); Ni(3)–O(9), 2.011(4); Ni(3)–O(10), 1.961(4); Ni(3)–O(100), 2.063(5); Ni(3)–O(110), 2.091(5). Selected angles (°): Ni(1)–O(2)–Dy(1), 109.81(17); Ni(1)–O(11)–Dy(1), 110.05(15); Ni(2)–O(3)–Dy(1), 110.51(15); Ni(2)–O(6)–Dy(1), 110.17(15); Ni(3)–O(7)–Dy(1), 110.62(16); Ni(3)–O(10)–Dy(1), 110.94(16).

Table 1 Comparison of selected structural parameters of the Ni^II₃Ln^III(L^Ald)₃(NO₃)₃·xsolvents complexes, Ln^III = Dy^III or Yb^III

Bond length (Å), angle (°) or distance (Å)	[Ni^II3Dy^III(L^Ald)3(NO3)2(MeOH)6]^+	[Ni^II3Yb^III(L^Ald)3(NO3)2(H2O)1.25(MeOH)4.75]^+
Ln^III–Ocatecholate range (average)	2.479–2.501 (2.489)	2.449–2.480 (2.464)
Ln^III–Onitrate range (average)	2.455–2.511(2.475)	2.405–2.458 (2.426)
Ni^II–Ocatecholate range (average)	1.961–1.988 (1.974)	1.959–1.984 (1.972)
Ni^II–Oformyl range (average)	2.007–2.019 (2.011)	2.003–2.011(2.007)
Ni^II–Omethanol range (average)	2.054–2.107 (2.078)	2.057–2.203 (2.107)
Ni^II–Owater	—	1.971–2.061 (2.029)
Ni^II–O–Ln^III range (average)	110.88–109.71 (110.3)	110.2–111.1 (110.8)
Ni^II⋯Ni^II range (average)	6.355–6.371 (6.364)	6.328–6.353 (6.342)
Ni^II⋯Ln^III range (average)	3.673–3.767 (3.675)	3.656–3.667 (3.662)
Shortest Ln^III⋯Ln^III distance in lattice	9.912	9.900

---

As expected, the large dysprosium(III) ion is bound to six catecholate oxygen atoms, while the nickel(II) ions are bound to two catecholate and two formyl oxygen atoms each. The dysprosium(III) ion is ten-coordinate with the remaining sites occupied by two nitrate anions binding in a bidentate η²-fashion. The three nickel(II) ions are six-coordinate with methanol molecules in both axial positions. The dysprosium(III) ion is a somewhat distorted tetradecahedron (ChSM (ref. 28) = 1.21) while the nickel(II) ions are regular octahedra (Σ (ref. 29)/ChSM for Ni(1) = 45.3°/0.43; Ni(2) = 34.5°/0.46; Ni(3) = 30.4°/0.41). The complex is quite planar; the angles of the plane made by an arbitrarily chosen catecholate ring to the other two ring planes are 10.7° and 14.6°.

The position of the third, non-coordinated nitrate anion could not be accurately determined due to extensive disorder in that region which resulted in the use of SQUEEZE.²¹ The resulting electron count for that region was consistent with the presence of one nitrate anion and one water molecule per complex. Furthermore, the microanalysis data are entirely consistent with the presence of three nitrate anions overall: this is particularly clear-cut as in these intermediate complexes the nitrate anions are the only possible source of nitrogen.

Single crystals of [Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(H₂O)_1.25(MeOH)_4.75](NO₃) were grown by vapour diffusion of diethyl ether into a solution of the complex in methanol. This structure was also treated with SQUEEZE. Overall, this structure (Fig. S1, ESI†) is very similar to that of the Dy^III analogue, but with disorder around the axially-coordinated ligands of the nickel(II) ions.

Magnetic properties

Having discovered SMM behaviour in the complexes [Cu^II₃Tb^III(L^Pr)(NO₃)₂(H₂O)₂(MeOH)](NO₃)₃·3H₂O,¹³ Zn^II₃Er^III(L^Pr)(NO₃)₃·3H₂O·2MeOH,¹² Zn^II₃Yb^III(L^Pr)(NO₃)₃·7H₂O¹² and [Zn^II₃Dy^III(L^Pr)(NO₃)₃(MeOH)₃]·4H₂O,^11,12 we were particularly interested to investigate the magnetic properties of the corresponding nickel(II) analogues, [Ni^II₃Ln^III(L^Pr)], with Ln^III = Tb^III, Er^III, Yb^III or Dy^III. However, for completeness (and to rule out the chance that an unexpected member of the family displays SMM behaviour) we have magnetically characterised all fourteen of these [Ni^II₃Ln^III(L^Pr)] complexes (Fig. 4 and S2–S15†).

Fig. 4 Temperature dependence of the χT product (with χ defined as the molar magnetic susceptibility at 1000 Oe and equal to M/H per mole of complex) for complexes Ni^II₃Ln^III(L^Pr)(NO₃)₃·(solvents), where Ln^III = La^III, Ce^III, Pr^III, Nd^III, Sm^III, Eu^III, Gd^III, Tb^III, Dy^III, Ho^III, Er^III, Tm^III, Yb^III and Lu^III.

Replacing Cu^II by Ni^II will increase the number of unpaired electrons on the 3d ion from 1 to 2, providing the nickel(II) ion is octahedral and high spin (S = 1), not square planar and diamagnetic (i.e. not S = 0 low spin). At room temperature, the χT product for each complex is in agreement with the expected value (Table 2) for three octahedral nickel(II) ions (S = 1, and thus the Curie constant per Ni^II is equal to 1 cm³ K mol⁻¹ when g = 2; calculated from C = g²S(S + 1)/8) plus the particular lanthanide(III) ion present (the Curie constant is expected to be C = g²J(J + 1)/8, e.g. for Dy^IIIJ = 15/2 and g = 4/3 so C ≈ 14.2 cm³ K mol⁻¹).

Table 2 Magnetic properties of the fourteen Ni^II₃Ln^III(L^Pr)(NO₃)₃·xsolvents complexes. No g value is given for La^III, Eu^III or Lu^III because J = 0

Complex	g of Ln^III ion	Measured/expected (with gNi = 2) χT at 300 K (cm^3 K mol^−1)	χT at 1.8 K (cm^3 K mol^−1)	M at 1.8 K and 70 kOe (μB)
Ni^II3La^III(L^Pr)(NO3)3·11H2O	—	3.4/3.00	1.44	4.8
Ni^II3Ce^III(L^Pr)(NO3)3·7.5H2O·4MeOH	6/7	4.0/3.80	1.58	5.6
Ni^II3Pr^III(L^Pr)(NO3)3·9H2O·2.5MeOH	4/5	4.6/4.60	1.39	5.1
Ni^II3Nd^III(L^Pr)(NO3)3·8H2O·3MeOH	8/11	4.4/4.64	1.01	4.5
Ni^II3Sm^III(L^Pr)(NO3)3·6.5H2O·2.5MeOH	2/7	3.4/3.09	1.96	5.0
Ni^II3Eu^III(L^Pr)(NO3)3·8.5H2O·2MeOH	—	4.8/3.00	1.03	3.8
Ni^II3Gd^III(L^Pr)(NO3)3·7.5H2O	2	11.1/10.88	17.5	12.2
Ni^II3Tb^III(L^Pr)(NO3)3·7H2O·2MeOH	3/2	15.0/14.81	17.0	9.4
Ni^II3Dy^III(L^Pr)(NO3)3·6H2O·3MeOH	4/3	18.3 /17.17	19.2	11.0
Ni^II3Ho^III(L^Pr)(NO3)3·8.5H2O·0.5MeOH	5/4	15.9/17.06	10.5	9.7
Ni^II3Er^III(L^Pr)(NO3)3·7.5H2O·3.5MeOH	6/5	14.0/14.47	9.76	9.8
Ni^II3Tm^III(L^Pr)(NO3)3·10H2O·MeOH	7/6	10.0/10.15	4.13	7.9
Ni^II3Yb^III(L^Pr)(NO3)3·10H2O	8/7	5.2/5.57	3.19	6.2
Ni^II3Lu^III(L^Pr)(NO3)3·7H2O	—	3.4/3.00	1.38	4.5

---

The temperature dependence of the χT product measured for [Ni^II₃La^III(L^Pr)] was used to investigate the nickel(II)–nickel(II) interaction without the complications arising from the presence of a paramagnetic lanthanide(III) ion (Fig. 4). Upon lowering the temperature for [Ni^II₃La^III(L^Pr)], the χT product is constant until around 50 K, at which point the value decreases to a minimum of 1.4 cm³ K mol⁻¹. This decrease at low temperature suggests that the interaction between the paramagnetic nickel(II) ions is weakly antiferromagnetic. To quantify the nature and magnitude of the nickel(II)–nickel(II) interaction, the experimental magnetic properties of [Ni^II₃La^III(L^Pr)] were modelled.

As a first approximation, a simple approach was developed that treats all three paramagnetic nickel(II) ions as equivalent and assumes one interaction, J₁, between adjacent metal ion centres (Fig. 5). This model is similar to that used for [Cu^II₃La^III(L^Pr)],¹³ except with nickel(II) ions in place of the copper(II) ions. From this structural motif, the resulting Heisenberg spin Hamiltonian used to calculate the magnetic susceptibility is:

Fig. 5 Scheme of the spin and exchange interaction topologies in [Ni^II₃La^III(L^Pr)] and [Ni^II₃Lu^III(L^Pr)].

From this model, the application of the van Vleck equation³⁰ allows a determination of the low field (μ_BH/k_BT ≪ 1) analytical expression of the magnetic susceptibility.³¹ The temperature dependence of the χT product was reproduced well with weak and equal antiferromagnetic coupling between the nickel(II) ions (J₁/k_B = −0.50(2) K and g = 2.13(5), Fig. 6). As an aside, [Ni^II₃Lu^III(L^Pr)], which also features a diamagnetic lanthanide ion (lutetium), could also have been used for this analysis (Fig. 6; J₁/k_B = −0.45(2) K and g = 2.15(5)). The good fit of the experimental data suggests that the simple approximation shown in Fig. 5 is accurate enough for the purposes of the analysis.

Fig. 6 Temperature dependence of the χT product (where χ is the molar magnetic susceptibility that equals M/H per mole of complex) collected in an applied dc magnetic field of 1000 Oe for [Ni^II₃La^III(L^Pr)] and [Ni^II₃Lu^III(L^Pr)]. The solid lines are the best fit to the model described in the text.

The nickel(II)–nickel(II) exchange interactions are similar (and weak) enough that only one coupling parameter is required (rather than J_Ni1–Ni2, J_Ni2–Ni3 and J_Ni3–Ni1) to model the experimental data above 1.8 K. It appears the subtle geometric differences between each Ni^II–catecholate–Ni^II moiety are not significant magnetically. The weakness of the interaction is probably because the interaction is via the multi-atom delocalised π system of (L^Pr)⁶⁻ rather than via a single atom bridge.

For nine of the complexes, [Ni^II₃Ce^III(L^Pr)], [Ni^II₃Pr^III(L^Pr)], [Ni^II₃Nd^III(L^Pr)], [Ni^II₃Sm^III(L^Pr)], [Ni^II₃Eu^III(L^Pr)], [Ni^II₃Ho^III(L^Pr)], [Ni^II₃Er^III(L^Pr)], [Ni^II₃Tm^III(L^Pr)] and [Ni^II₃Yb^III(L^Pr)], the χT product is either almost constant down to low temperatures or shows only a slight decrease due to thermal depopulation of excited sublevels (Fig. 4, S3–S7 and S11–S14, ESI†). Below about 50 K, there is a more pronounced decrease of the χT product on further decreasing the temperature, consistent with weak antiferromagnetic nickel(II)–nickel(II) coupling, as observed for [Ni^II₃La^III(L^Pr)] and [Ni^II₃Lu^III(L^Pr)], and only weak nickel(II)–lanthanide(III) coupling (otherwise the χT product would be expected to have stronger temperature dependence).

Due to the isotropic nature of the gadolinium(III) ion, the temperature dependence of the χT product of [Ni^II₃Gd^III(L^Pr)] above 5 K is only representative of the interactions between the metal ions (Fig. 7).

Fig. 7 Temperature dependence of the χT product (where χ is the molar magnetic susceptibility that equals M/H per mole of complex) collected in an applied dc magnetic field of 1000 Oe for [Ni^II₃Gd^III(L^Pr)] (empty circles). The solid red line is the best fit to the model described in the text. Insert: semi-logarithm plot of the main figure.

The increase in the χT product from a minimum at room temperature to a maximum at 3.5 K is consistent with a dominant ferromagnetic nickel(II)–gadolinium(III) interaction (over the antiferromagnetic nickel(II)–nickel(II) interaction that was identified in [Ni^II₃La^III(L^Pr)]), with the subsequent small decrease in the χT product from 3.5 to 1.9 K probably due to the intermolecular antiferromagnetic interactions. As a result of too large a number of adjustable parameters (i.e. overparametrization) attempts to fit both the χT versus T and M versus H experimental data with the same model failed to give a self-consistent set of magnetic parameters even when including intra-complex magnetic interactions and the magnetic anisotropy of the nickel(II) ions. Therefore, only the χT versus T data have been fitted with the following Heisenberg spin Hamiltonian:

The model only produces an acceptable fit to the experimental data when inter-complex magnetic interactions in the frame of the mean field approximation are introduced, keeping in mind that this additional parameter might also include phenomenologically the magnetic anisotropy of the Ni^II sites. Therefore the susceptibility is given by:

An excellent simulation of the experimental χT vs. T data is obtained with this model, with J_Ni–Ni/k_B = −0.02(2) K, J_Ni–Gd/k_B = +2.6(1) K, g = 2.00(5) and zJ′/k_B = −0.021(5) K, down to 1.8 K (Fig. 7). This result is also consistent with the presence of weak antiferromagnetic coupling between the well separated S = 1 Ni^II spins while ferromagnetic interactions dominate between the Ni^II spins and the S = 7/2 Gd^III spins within the [Ni^II₃Gd^III(L^Pr)] complex, mediated by the single oxygen atom bridge between them, leading to an S_T = 13/2 spin ground state (the S = 11/2 excited state lies 5 K above it in energy).

In the case of [Ni^II₃Tb^III(L^Pr)] and [Ni^II₃Dy^III(L^Pr)], the temperature dependence of the χT product is very similar (Fig. 4, 8, S9, S10 and S16†). As the temperature is lowered, the value remains fairly constant until around 50 K, at which point it increases to a maximum at very low temperature. The χT product then decreases slightly until the temperature limit of the measurement (1.8 K). It is usually difficult to determine the nature of the exchange interactions in 3d–4f complexes because of thermal depopulation of excited sublevels of the ^2S+1L_J ground state of the lanthanide(III) ion. In our case, comparing the magnetic data with that of the appropriate analogues [Ni^II₃La^III(L^Pr)] and [Zn^II₃Ln^III(L^Pr)]¹² allows a qualitative estimation of the 3d–4f interactions by removing the contributions of the 3d–3d interactions and the intrinsic nickel(II) and lanthanide(III) paramagnetism. As shown in Fig. 8 and S16,† the resulting χT product difference plot exhibits a minimum at room temperature before increasing to a maximum at 1.8 K, suggesting that the nickel(II)–terbium(III) and nickel(II)–dysprosium(III) interaction are ferromagnetic.

Fig. 8 Temperature dependence of the χT product (where χ is the molar magnetic susceptibility that equals M/H per mole of complex) collected in an applied dc magnetic field of 1000 Oe for [Ni^II₃La^III(L^Pr)], [Zn^II₃Tb^III(L^Pr)], [Ni^II₃Tb^III(L^Pr)] and the remainder after subtraction of the data for [Ni^II₃La^III(L^Pr)] and [Zn^II₃Tb^III(L^Pr)] from [Ni^II₃Tb^III(L^Pr)].

The field dependence of magnetisation was measured for all fourteen complexes to check for hysteresis of magnetisation, as this is the best evidence of SMM behaviour. However, even at 1.8 K, there is no sign of hysteresis behaviour for any of these complexes when scanning at 100–200 Oe min⁻¹ (Fig. S17–S30, ESI†).

The presence of significant magnetic anisotropy in all of the complexes is confirmed by the field dependence of reduced magnetisation plots (M vs. H/T). In all cases, these show non super-imposable data on a single master curve (Fig. S17–S30, ESI†) and, as expected for significantly anisotropic systems, there is no saturation of the magnetisation at 1.8 K and 70 kOe. Interestingly, the M vs. H/T plots for [Ni^II₃Ln^III(L^Pr)] with Ln^III = diamagnetic La^III/Lu^III or isotropic Gd^III do not show super-imposable data on a single master curve indicating the presence of low lying excited states and likely a minor contribution from the magnetic anisotropy of the nickel(II) ions (Fig. S17, S23 and S30, ESI†).

The ac susceptibility of all fourteen complexes was measured to check for possible slow relaxation of the magnetisation. At 1.8 K, there was no out-of-phase component for any of the complexes, which indicates there is no appreciable energy barrier to spin reversal (Fig. S21–S44, ESI†).

Conclusions

The template procedure we reported for the generation of large families of [Zn^II₃Ln^III] and [Cu^II₃Ln^III] macrocyclic complexes has been extended to the preparation of a family of [Ni^II₃Ln^III] macrocyclic complexes. The resulting fourteen complexes vary in the identity of the lanthanide(III) ion present. Although none of the complexes displayed SMM behaviour, this study further illustrates the robustness and versatility of this system in the preparation of a wide range of analogous complexes. Our attention is now focussed on extending this study from the late transition metal ions towards the mid-transition metal ions with a view to elucidating the factors which induce slow relaxation of the magnetisation in 3d/4f blends. Furthermore, it will be of interest to incorporate trivalent 3d ions such as manganese(III) which should provide anisotropy as well as a larger spin.

Acknowledgements

This work was supported by grants from the MacDiarmid Institute for Advanced Materials and Nanotechnology (including a PhD scholarship to SD), the University of Otago (including a PhD scholarship to HLCF) as well as the award of a Julius von Haast Fellowship (RSNZ) to AKP, hosted by SB. We also thank the University of Bordeaux, the CNRS, the Région Aquitaine and the Dumont d'Urville NZ-France Science & Technology Support Programme which supported research exchange visits between Bordeaux and Otago (no. 23793PH).

Notes and references

1. D. Gatteschi, R. Sessoli and F. Villain, Molecular Nanomagnets, Oxford University Press, 2006.
2. J. D. Rinehart, M. Fang, W. J. Evans and J. R. Long, J. Am. Chem. Soc., 2011, 133, 14236.
3. See for example: G. Aromí and E. K. Brechin, Struct. Bonding, 2006, 122, 1; M. Murrie, Chem. Soc. Rev., 2010, 39, 1986; T. Glaser, Chem. Commun., 2011, 47, 116; P. Happ, C. Plenk and E. Rentschler, Coord. Chem. Rev., 2015, 289, 238; T. Glaser, V. Hoeke, K. Gieb, J. Schnack, C. Schröder and P. Müller, Coord. Chem. Rev., 2015, 289, 261.
4. See for example: R. Sessoli and A. K. Powell, Coord. Chem. Rev., 2009, 253, 2328; J. D. Rinehart and J. R. Long, Chem. Sci., 2011, 2, 2078; J. Luzona and R. Sessoli, Dalton Trans., 2012, 41, 13556; D. N. Woodruff, R. E. P. Winpenny and R. A. Layfield, Chem. Rev., 2013, 113, 5110; P. Zhang, L. Zhang and J. Tang, Dalton Trans., 2015, 44, 3923.
5. See for example: R. A. Layfield, Organometallics, 2014, 33, 1084.
6. See for example: H. L. C. Feltham and S. Brooker, Coord. Chem. Rev., 2014, 276, 1.
7. See for example: K. R. Meihaus and J. R. Long, Dalton Trans., 2015, 44, 2517.
8. See for example: M. Andruh, J.-P. Costes, C. Diaz and S. Gao, Inorg. Chem., 2009, 48, 3342; K. Liu, W. Shi and P. Cheng, Coord. Chem. Rev., 2014, 289–290, 74; K. S. Pedersen, J. Bendix and R. Clérac, Chem. Commun., 2014, 50, 4396; S. Dhers, H. L. C. Feltham and S. Brooker, Coord. Chem. Rev., 2015, 296, 24.
9. N. Ishikawa, Polyhedron, 2007, 26, 2147.
10. M. Gonidec, E. S. Davies, J. McMaster, D. B. Amabilino and J. Veciana, J. Am. Chem. Soc., 2010, 132, 1756; M. Gonidec, F. Luis, À. Vílchez, J. Esquena, D. B. Amabilino and J. Veciana, Angew. Chem., Int. Ed., 2010, 49, 1623; P.-E. Car, M. Perfetti, M. Mannini, A. Favre, A. Caneschi and R. Sessoli, Chem. Commun., 2011, 47, 3751; M. Gonidec, D. B. Amabilino and J. Veciana, Dalton Trans., 2012, 41, 13632; G. Cucinotta, M. Perfetti, J. Luzon, M. Etienne, P.-E. Car, A. Caneschi, G. Calvez, K. Bernot and R. Sessoli, Angew. Chem., Int. Ed., 2012, 51, 1606; U. J. Williams, B. D. Mahoney, P. T. DeGregorio, P. J. Carroll, E. Nakamaru-Ogiso, J. M. Kikkawa and E. J. Schelter, Chem. Commun., 2012, 48, 5593; H. Wang, K. Wang, J. Tao and J. Jiang, Chem. Commun., 2012, 48, 2973; M.-E. Boulon, G. Cucinotta, J. Luzon, C. Degl'Innocenti, M. Perfetti, K. Bernot, G. Calvez, A. Caneschi and R. Sessoli, Angew. Chem., Int. Ed., 2013, 52, 350; C. R. Ganivet, B. Ballesteros, G. d. l. Torre, J. M. Clemente-Juan, E. Coronado and T. Torres, Chem. – Eur. J., 2013, 19, 1457.
11. H. L. C. Feltham, Y. Lan, F. Klöwer, L. Ungur, L. F. Chibotaru, A. K. Powell and S. Brooker, Chem. – Eur. J., 2011, 17, 4362.
12. H. L. C. Feltham, F. Klöwer, S. A. Cameron, D. S. Larsen, Y. Lan, M. Tropiano, S. Faulkner, A. K. Powell and S. Brooker, Dalton Trans., 2011, 40, 11425.
13. H. L. C. Feltham, R. Clérac, A. K. Powell and S. Brooker, Inorg. Chem., 2011, 50, 4232.
14. H. L. C. Feltham, R. Clérac, L. Ungur, V. Vieru, L. F. Chibotaru, A. K. Powell and S. Brooker, Inorg. Chem., 2012, 51, 10603.
15. A. Yamashita, A. Watanabe, S. Akine, T. Nabeshima, M. Nakano, T. Yamamura and T. Kajiwara, Angew. Chem., Int. Ed., 2011, 50, 4016.
16. S. Dhers, H. L. C. Feltham, R. Clérac and S. Brooker, Inorg. Chem., 2013, 52, 13685.
17. H. L. C. Feltham, R. Clérac, L. Ungur, L. F. Chibotaru, A. K. Powell and S. Brooker, Inorg. Chem., 2013, 52, 3236.
18. S. Akine, T. Taniguchi and T. Nabeshima, Tetrahedron Lett., 2001, 42, 8861.
19. S. Akine, T. Taniguchi and T. Nabeshima, J. Am. Chem. Soc., 2006, 128, 15765.
20. G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008, 64, 112.
21. P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 1990, 46, 194.
22. S. Yamada, E. Ohno, Y. Kuge, A. Takeuchi, K. Yamanouchi and K. Iwasaki, Coord. Chem. Rev., 1968, 3, 247; M. G. B. Drew, R. N. Prasad and R. P. Sharma, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1985, 41, 1755; Y. Elerman, M. Kabak and O. Atakol, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1993, 49, 1905.
23. C. Arici, I. Svoboda, M. Sari, O. Atakol and H. Fuess, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2001, 57, 31; O. Atakol, H. Nazir, C. Arici, S. Durmus, I. Svoboda and H. Fuess, Inorg. Chim. Acta, 2003, 342, 295.
24. N. H. Pilkington and R. Robson, Aust. J. Chem., 1970, 23, 2225.
25. H. Okawa and S. Kida, Bull. Chem. Soc. Jpn., 1972, 45, 1759.
26. A. J. Atkins, D. Black, A. J. Blake, A. Marin-Becerra, S. Parsons, L. Ruiz-Ramirez and M. Schroder, Chem. Commun., 1996, 457; S. Brooker, Coord. Chem. Rev., 2001, 222, 33.
27. W. J. Geary, Coord. Chem. Rev., 1971, 7, 81.
28. M. Llunell, D. Casanova, J. Cirera, P. Alemany and S. Alvarez, in SHAPE: Program for the Stereochemical Analysis of Molecular Fragments by Means of Continuous Shape Measures and Associated Tools, Barcelona, 2013.
29. M. G. B. Drew, C. J. Harding, V. McKee, G. G. Morgan and J. Nelson, J. Chem. Soc., Chem. Commun., 1995, 1035; F. A. Deeney, J. H. Charles, G. G. Morgan, V. McKee, J. Nelson, S. J. Teat and W. Clegg, J. Chem. Soc., Dalton Trans., 1998, 1837; P. Guionneau, C. Brigouleix, Y. Barrans, A. E. Goeta, J.-F. Létard, J. A. K. Howard, J. Gaultier and D. Chasseau, C. R. Acad. Sci., Ser. IIc: Chim., 2001, 4, 161.
30. J. H. van Vleck, The Theory of Electric and Magnetic Susceptibility, Oxford University Press, 1932.
31. M. Murugesu, R. Clérac, C. E. Anson and A. K. Powell, J. Phys. Chem. Solids, 2004, 65, 667.

---

Footnote

† Electronic supplementary information (ESI) available: Additional dc and ac magnetic data and selected crystallographic details for [Ni^II₃Yb^III(L^Ald)₃(NO₃)₂(MeOH)_4.75(H₂O)_1.25](NO₃). CCDC 913628 and 913629. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5qi00130g

---