Microwave assisted synthesis of heterometallic 3d-4f M 4 Ln complexes

In this paper we describe the synthesis and magnetic properties of a series of 3d-4f complexes of general formula [M4Ln(OH)2(chp)4(SALOH)5(H2O)(MeCN)(Solv)] (solv = MeOH, MeCN, H2O; chp stands for deprotonated 6-chloro-2-hydroxypyridine (C5H3ClNO), SALOH stands for monodeprotonated 3,5-ditert-butylsalicylic acid (C15H21O3)) obtained by a solvent-free microwave assisted synthesis method. The Ni(II) complexes (Ni4Gd, Solv = MeOH; Ni4Dy, Solv = MeCN) are not SMMs in the absence of an appied dc field. The replacement of Ni(II) by Co(II) (Co4La, Solv = MeOH; Co4Gd, Solv = H2O; Co4Gd-MeCN, Solv = MeCN; Co4Tb, Solv = MeOH; Co4Dy, Solv = H2O) results in improved SMM properties. b), 501 (s, b); where strong (s), medium (m), weak (w), broad (b). XPS spectra show that the precipitate contains a small impurity of NaCl and Co 2 (OH) 3 Cl. were performed at Servei de Microanàlisi in CSIC (Consell Superior d’Investigacions Científiques). Infrared spectra were collected on KBr pellets on an AVATAR 330 FT-IR at Departament de Química Inorgànica, Universitat de Barcelona. XPS experiments were performed in a PHI 5500 Multitechnique System (from Physical Electronics) with a monochromatic X-ray source (Aluminium K-alfa line of 1486.6 eV energy and 350 W), placed perpendicular to the analyzer axis and calibrated using the 3d5/2 line of Ag with a full width at half maximum (FWHM) of 0.8 eV. The analysed area was a circle of 0.8 mm diameter, and the selected resolution for the spectra was 187.85 eV of Pass Energy and 0.8 eV/step for the general spectra and 23.5 eV of Pass Energy and 0.1 eV/step for the spectra of the different elements. A low energy electron gun (less than 10 eV) was used in order to discharge the surface when necessary. All Measurements were made in a ultra high vacuum (UHV) chamber pressure between 5x10-9 and 2x10-8 torr. Magnetic measurements were performed at the Unitat de Mesures Magnètiques of the Universitat de Barcelona on a Quantum Design SQUID MPMS-XL magnetometer equipped with a 5 T magnet. Diamagnetic corrections for the sample holder and for the sample using Pascal’s constants were applied. Hysteresis measurements were performed with an array of micro-SQUIDs. This magnetometer works in the temperature range of 0.04 to 5 K and in fields up to 1.4 T with sweeping rates as high as 0.28


Introduction
Interest in molecular nanomagnets due to their potential applications in highdensity information storage, molecular spintronics, quantum computing 1,2,3 or magnetic coolers 4 has grown in the last 30 years. Since the discovery of the first single-molecule magnet (SMM) Mn 12 Ac in the 90's by Christou 5 and Gatteschi, 6 the synthesis of high nuclearity complexes with SMM properties is a great and challenging target for coordination chemists. An SMM must have a large spin ground state and easy axis anisotropy, leading to slow relaxation of the magnetization below its blocking temperature. Thus, an SMM is able to retain the magnetization and it behaves like a magnet at the molecular level. 7,8 SMMs were also obtained using other 3d transition metals including manganese, iron, nickel, cobalt or vanadium, 9 but in these cases higher working temperatures have been elusive. Thus, the anisotropy barrier for the reversal of the magnetization in transition metal SMMs depends on two properties: the total spin of the molecule S and the Ising-type anisotropy. This knowledge has been used to design improved SMMs based on two strategies: rising S or increasing the anisotropy of the molecule. Increasing S by introducing stronger ferromagnetic coupling has been achieved in several examples: Mn 18 , 10 Mn 21 , 11 Mn 84 12 or Mn 19 . 13 However, higher nuclearity structures with a large S value is no guarantee of a large molecular anisotropy. A great example is Mn 19 : it possesses the record spin of 83/2 for a molecular cluster, but it lacks anisotropy and thus it is not an SMM. Focus is set since the early 2000's in increasing the magnetic anisotropy of the prepared complexes in order to improve SMM properties, using metal ions with strong spin-orbit coupling as Co(II) or the lanthanides. Heterometallic 3d-4f compounds as well as pure 4f systems are seen as the route to better SMMs. Rare earths have long been used in magnetism due to their strong magnetic anisotropy. 4f complexes usually have high energy barriers compared with 3d metals SMMs but their hysteresis loops are usually closed due to fast QTM (Quantum Tunnelling of the Magnetization) and alternative relaxation pathways. 14,15 Since the first heterometallic 3d-4f SMM was reported in 2004, many groups have devoted much effort to these 3d-4f complexes. 16 Still, in 2018 high operational temperatures for 4f or 3d-4f SMMs remain elusive. In 2018, after his previous work on organometallic lanthanide SMMs, 17,18,19 Layfield reported a nearly linear Dy metallocenium that displayed hysteresis at 80 K. 20 Reta and Chilton offered theoretical insight on the high temperature hysteresis, relating it once again to the coupling with vibrational phonons in the structure. 21 The last results greatly improved the archetypical lanthanide SMMs, the phthalocyanine lanthanide sandwich complexes first studied as SMMs by Ishikawa et al. 22,23,24 The synthetic methodologies are crucial for obtaining homo-or heteronuclear complexes with the desired properties, but most are still based on serendipitous assembly. 25 The interest on finding new synthetic methodologies is a challenge in chemistry. 26 In the last few years, microwave assisted synthesis has been useful for organic chemists not only to synthesize products that are otherwise unattainable, also to obtain in pure form products that in conventional conditions appear in a mixture that requires taxing purification methods. In the same way, this synthetic method is useful for the synthesis of high nuclearity coordination compounds. 27 In general, with microwave assisted synthesis, formation of one species is favoured. 28 Microwave assisted synthesis is clean, quick and it is included in green chemistry, because of its partial or in some cases total absence of organic solvents. 29 The microwave reactor offers a unique environment, which allows high temperatures and high pressure, generated by the heating of the molecules by the microwave radiation. Microwave energy is delivered directly to materials through molecular interaction with the electromagnetic field. 30 Only those substances with a dipolar moment will be excited by microwaves. This energy transfer to some of the species in the reaction mixture is very efficient and the heating rate is homogeneous. 31 The technique has been used to synthesize polynuclear coordination compounds like Ni 8 and Ni 9 32 or Mn 3 , 33 MOF's (Metal-Organic Frameworks) 34 and nanoparticles. 35 Solventfree synthetic methods are often used in coordination chemistry, in a melt state [36][37][38][39][40] or by sublimation of the product. 41 Of particular relevance to spintronics is the archetypical mononuclear single molecule magnet (SMM) TbPc 2 , 22 synthesized by conventional heating over long periods of time and that often produce mixtures of products and require tedious purification methods. 42,43 In 2017 we used solvent-free microwave assisted synthesis to obtain [Ni 4 Tb(OH) 2 (chp) 4 (SALOH) 5 (H 2 O)(MeCN)(MeOH)] (Ni 4 Tb). and its La analogue, Ni 4 La. 44 Clearly, the prepared complexes were very interesting to us and we wanted to extend the synthetic method to other lanthanide ions and other transition metals. We herein exploit this synthetic method to prepare a family of M 4 Ln complexes of general formula [M 4 Ln(OH) 2 (chp) 4 (SALOH) 5  O) and its by-products. The change of Ni(II) for Co(II) leads to an improvement on the SMM properties of the prepared complexes, as the cobalt complexes display SMM properties in the absence of an applied dc field.

Results and Discussion
Following our work on microwave assisted synthesis in coordination chemistry we decided to apply the method to solvent-free systems. This would provide a clean, cost and time efficient method to obtain heterometallic coordination complexes. The clear limitation of the method is only that the ligands chosen must have easily attainable melting points in order to provide a good molten media for ion diffusion.
Microwave assisted synthesis was used to obtain new 3d-4f molecular nanomagnets with two versatile ligands (2-hydroxy-6-chloropyridine, Hchp, melting point 128-130°C and 3,5-ditertbutylsalycilic acid, SALOH 2 , (melting point 157-162°C)) chosen due to their many possible coordination modes and their low melting points. The tert-butyl groups also improve solubility, processability of SMMs and selforganization on a metal surface. [45][46][47] The metal salts used are the metal hydroxides, freshly prepared, since they provide useful OH-counterions that are desired bridges in the final products. In 2017 we reported complex Ni 4 Tb [Ni 4 Tb(OH) 2 (chp) 4 (SALOH) 5  Following this work on homometallic and heterometallic reactions with a versatile ligand system we decided to extend this chemistry to other transition metals and to study the reaction system in different reaction conditions. Three different reaction methods were studied: (a) solvent-free microwave assisted synthesis, (b) microwave assisted synthesis in solvent and (c) bench-top stirring reactions at room temperature.
The solvent free microwave assisted reaction (method a) was studied with microwave pulses between 100 W and 300 W at the melting point of the ligands. The 300 W pulse was chosen since it consistently produced the best yield of products. Once the reaction was cooled a solid was obtained and characterized by IR spectroscopy: the results were very similar to those of the products after recrystallization. The comparison of the IR of the melt with the IR of the crystals and that of free SALOH 2 (see ESI Figure S05) clearly shows that the organic ligands are already coordinated to the M(II) and Ln(III) ions, a fact that was supported by the formation of acetic acid during the reaction by protonation of the acetate groups from the lanthanide acetate reagent. The solid was extracted with the minimum amount of MeOH/MeCN mixture (1:1 in volume) and the green or pink-purple solution was left undisturbed. Crystals suitable for single crystal X-Ray diffraction were obtained after 15-20 days. For complex Co 4 Gd-MeCN, crystals of the product were obtained using only MeCN as crystallization solvent.  were only obtained from microwave assisted reactions. However, method b in the case of the Co/Gd and Co/Dy systems do not afford a pure product but a mixture of species. In fact, for the Co(II)/Ln(III) system we here observe that the microwave assisted reaction with solvent (method b in Scheme 1) affords a mixture of the products from the solvent-free microwave assisted reaction (method a in Scheme 1) and the bench-top reaction (method c in Scheme 1). To our knowledge, this Co/Ln system is unusual for microwave assisted synthesis which is often used to avoid by-products or mixtures of products.
The solvent-free microwave assisted reaction has been applied successfully to 3d-4f heterometallic coordination complexes in a carefully chosen ligand system. We believe this method can be extended to other metal-ligand systems with good results.

Description of Crystal structure
Crystallographic and data collection details for Ni 4 Dy and Co 4 La, Co 4 Gd, Co 4 Gd-MeCN and Co 4 Dy are presented in Table 1. For Ni 4 Tb-b only the unit cell was checked and it coincided with that of Ni 4 Tb. 44 Crystals were often very small and diffracted poorly due to the free-rotation of tert-butyl groups, thus data for complex Co 4 Gd were collected using synchrotron radiation. Data for Ni 4 Gd and Co 4 Tb had low resolution so only unit cells are reported. The unit cell parameters are as expected, very similar to those of the other M 4 Ln complexes and can be found in SI Table S-C1. All complexes crystallize in the monoclinic space group P21/c except Ni4Dy. They all have the same M4Ln(OH)2 core with small differences only in the coordinated solvent molecules. Ni4Dy [Ni4Dy(OH)2(chp)4(SALOH)5(H2O)(MeCN)2] crystallizes in the orthorhombic space group Fdd2. To avoid repetition, a general description of the common features of all M4Ln complexes will be given. The asymmetric unit contains the whole molecule and disordered non-coordinated solvents, this is also true for complex Ni4Dy. The crystal structure and the core of Co4Gd-MeCN are shown in Figure 1.   Figures 2 and 3, respectively. The data were collected in the 2-300 K temperature range at an applied field of 3000 Oe and a field of 300 Oe below 30 K. The χT product values at 300 K are collected in Table S01. The susceptibility of the M4Ln complexes can be understood as the addition of a lanthanide ion weakly coupled to the more strongly coupled tetranuclear transition metal M4 unit. The couplings between the transition metals in the M4 unit are easily derived from the La derivative, for the cobalt series, that is Co4La. For the Ni(II) series, the ferromagnetic coupling between the Ni(II) ions in Ni4La was described in our 2017 paper as well as the ferromagnetic Ni-Tb coupling in Ni4Tb, confirmed by XMCD. 44 For the Co(II) analogue there is some extent of antiferromagnetic coupling between the Co(II) centres of Co4Ln. DFT calculations have been used to estimate the exchange coupling between Co(II) ions and between Co(II) and the lanthanide ion.  Magnetization vs. field data at 2 K for Ni 4 Ln and Co 4 Ln complexes are shown in Figure S01. For the family of Ni 4 Ln complexes saturation is almost reached at 5 T in agreement with the ferromagnetic interactions that lead to the spin ground state. This is not the case of the Co 4 Ln complexes. The fact that the magnetization value does not saturate is consistent with magnetic anisotropy associated with strong spin orbit coupling and with a combination of ferromagnetic and antiferromagnetic interactions as calculated using theoretical approaches. Given the field-induced SMM nature of Ni 4 Tb 44 and the large spin ground states observed and the anisotropy associated to lanthanide complexes, the dynamic magnetic properties of the new Ni 4 Ln and the Co 4 Ln complexes were studied. Ac magnetic susceptibility data were collected between 20 K and 1.8 K. The data showed the absence of out-of-phase signals for Ni 4 Gd and Ni 4 Dy with or without an applied dc field. Ni 4 Tb is the only complex of the nickel family that shows clear field-induced SMM behaviour. 44 Clearly, the Ni 4 unit, even though it has a large spin does not have enough anisotropy to trigger SMM behaviour. The change of the transition metal from Ni(II) to Co(II) has a clear effect in the magnetic anisotropy of the complex, thus affecting the dynamic magnetic properties of the species. Ruiz and co-workers showed for octahedral Co(II) monomers that D values are larger than for Ni(II) but sometimes positive. In polynuclear systems, an effective negative D for the molecule can be obtained by the appropriate orientation of the local ZFS tensors with respect to the molecular easy axis. 56 Gd(III), with a large S = 7/2 is ferromagnetically coupled to the Co 4 unit in Co 4 Gd, the complex has a large magnetic moment. Co 4 Gd is an SMM and it displays an out-of-phase ac magnetic susceptibility peak at 5 K, as shown in Figure 4 (top). The relaxation data was fitted to Arrenhius' equation with τ o = 6.95e-12 s -1 and U eff = 86 K, values typical of a transition metal SMM ( Figure  S03) as good as the archetypical Mn 12 Ac. 9,57 The Argand plots for Co 4 Gd at different temperatures between 3 K and 6 K are shown in ESI Figures S05-06. The data were fitted to a Debye model for a distribution of relaxation processes with τ o = 1.06e-6-1.53e-2 s -1 and α = 0.56-0.69 between 3 and 6 K. The values for each temperature can be found in the ESI Table S01. Co 4 Tb and Co 4 Dy only show tails of out-of-phase ac susceptibility peaks if a dc field is applied. Thus, a static dc magnetic field was applied and ac magnetic susceptibility data were collected for Co 4 Tb and Co 4 Dy. The data are shown in Figure 4 (centre) with an applied dc field of 2000 Oe. A peak appears below 6 K for Co 4 Tb but the maximum is only observed at the highest frequencies (750 to 1500 Hz). If we compare and Co 4 Tb to Ni 4 Tb, the ac peak appears at higher temperatures so it is clear that using Co(II) instead of Ni(II) leads to higher energy barriers for the relaxation of the magnetization. The ac data for Co 4 Dy with an applied field of 2000 Oe shows two relaxation processes, one centred at 11 K and the other one below 3 K. The hightemperature process can be analysed using the Arrhenius' equation with τ o = 3.43e-7 s -1 and U ef f= 66 K (see ESI Figure S03). The low temperature peak is related to the application of a dc field and is important at low temperatures and low frequencies, as discussed for Ni(II), Co(II) and Cu(II) SMMs by Titis and Boca. 58,59 The ac susceptibility for Co 4 Dy was also studied as a function of frequency at several temperatures, (ESI Figure S04-05). The data were fitted to a Debye model for a distribution of relaxation processes with τ o = 8e-5-9.4e0-3 s -1 and α = 0.04-0.31 between 5 and 13 K. The values for each temperature can be found in the ESI Table S01.
Clearly the combination of a Co 4 SMM with an anisotropic lanthanide ion like Dy(III) or Tb(III) results in this case in worst properties than the combination of the Co 4 unit with Gd(III). The controversy between the importance of spin and anisotropy is clear for this family of Co 4 Ln complexes: if anisotropy is good enough, increasing spin wins over increasing magnetic anisotropy in a heterometallic complex and the Co/Gd analogue Co 4 Gd is the SMM with ac out-of-phase peaks at higher temperatures in the absence of an applied dc field, however, QTM effects are very relevant in all Co 4 Ln complexes. Theoretical approach to magnetic properties. For Co 4 La the magnetic properties were modelled using a theoretical approach that shows that there is a combination of ferromagnetic and antiferromagnetic interactions between the Co(II) ions. In order to study the intramolecular exchange interactions of the Co 4 La system, electronic structure calculations at density functional level have been performed using the Heisenberg Hamiltonian (see Computational Details). The computed exchange constants as well as the involved atoms can be found in Table 2.
The overall magnetic behaviour of Co 4 La is a combination of ferromagnetic and antiferromagnetic contributions. The ferromagnetic interaction J 3 takes place via two chp ligands coordinated by both the O and N donor atoms (see Figure 1). This interaction is responsible of the observed T value at room temperature. The only antiferromagnetic interaction observed in Co 4 La corresponds to J 1 , and involves the  3 -OH-bridging ligand, the O-donor atom from the chp ligand (coordination mode IV in Scheme 2) and a syn,syn-carboxylato group form SALOH (coordination mode I in Scheme 2). The syn,syn-carboxylato bridging mode is well known to favour antiferromagnetic coupling. Additional spin  In order to assess the Co(II)-Ln(III) coupling the magnetic data for complex Co 4 Gd, the magnetic properties of Co 4 Gd have also been calculated using theoretical approach. The magnetic properties of complex Co 4 Dy have been obtained by rescaling by a factor 5/7 and the anisotropy of Dy(III) studied. Introducing one lanthanide ion with a magnetic moment increases the complexity of the model. DFT calculations have been performed to explore the Co(II)-Co(II) and Co(II)-Gd(III) magnetic exchange interactions in Co 4 Gd. To estimate four different exchange coupling constants, five spin configurations were chosen (See ESI for details). The exchange coupling constants (Table 3) are estimated to be J 1 (Co-Co) = -2.43, J 2 (Co-Co) = 0.29, J 3 (Co-Co) = 0.76 and J 4 (Co-Gd) = 0.71 cm -1 . The Co-Co coupling constants (J 1 and J 2 ) are similar to those obtained for Co 4 La. Ab initio calculation predicts that axial zero field splitting parameter of each Co is positive. The anisotropy axis of the Co(II) and Dy(III) in Co 4 Dy are shown in Figure 5.  63 The values in Table 4 imply that D value increases with increasing deviation from the octahedral symmetry except Co(3) (see Table  S07). The magnitude of E/D value decreases with increasing deviation from the octahedral symmetry except for Co (2). To validate the exchange parameters obtained by DFT calculation we attempted to fit the experimental susceptibility data for Co 4 La, Co 4 Gd and Co 4 Dy. The susceptibility data of Co 4 La and Co 4 Gd were fitted using PHI 64 , with the ab initio computed D and E tensors of Co(II) ions along with DFT estimated exchange coupling constants as a starting point. The best fits are shown in Figure 3 as solid lines. For complex Co 4 La, the J 1 and J 3 exchanges were varied to get the best fit. The best fit was obtained with the value of -0.75 cm-1 for J 1 65,66 Although the zero field splitting parameter of the individual Co(II) centres are positive the anisotropy axis of the individual Co(II) centres are not collinear to each other ( Figure 5). The overall magnetic behaviour is dominated by the ferromagnetic exchange (J 2 , J 3 and J 4 ). This leads to S = 19/2 ground state for Co 4 Gd (which is also supported by the experimental value of temperature dependent susceptibility at high temperature) leading to a large barrier for magnetisation reversal in zero field. Figure 6 shows the population of each spin state calculated using PHI. For complex Co 4 Dy, simulations were performed using POLY_ANISO routine. The DFT computed exchange coupling constants (re-scaled to Dy(III)) were used for the simulation with zJ' = 0.001 cm -1 . This set already yields an impressive match to the experimental data highlighting the importance of parameter-free approach to the estimation of such cumbersome coupling constants.

Magnetic relaxation mechanism for the Dy(III) centre in Co 4 Dy.
A qualitative mechanism relaxation developed based on the ab initio calculations for the Dy(III) centre is shown in Figure 7. The very large anisotropy of the Dy centre dominates over the magnetic exchange and anisotropy of the Co centres. Therefore, the overall magnetic anisotropy of the Co 4 Dy originate from the single ion Dy(III) centre. The relaxation mechanism of the Dy centre implies very large QTM in the ground state and suggest no SMM characteristic (zero-field) should arise due to Dy(III) centre (Figure 7). Continuous Shape Measure analysis around Dy(III) centre reveals that the ion resides in a trigonal-prismatic geometry (see Table S07). The large QTM is due to the very large deviation from the trigonal-prismatic symmetry. The computed g tensors (Table S08) reveal a large transverse anisotropy in the ground state signifying strong tunnelling and absence of SMM characteristics in zero field. The ground and first excited Kramer's doublet (KD) energy gap is estimated to be very small (31 cm -1 ) due to the very low symmetry. The large QTM is also supported from the crystal field analysis where non-axial crystal field parameters are larger than the axial crystal field parameters (Table S10). The probability of Orbach process (2.22 µ B ) is very large which reinforces it to relax via the first excited KD. The energy barrier for ground and first excited states is 31 cm -1 which is underestimated compared to the experimental blocking barrier of 46 cm -1 . The good agreement between the experimental susceptibility and that calculated for Co 4 Dy using the DFT computed values re-scaled for Dy(III) highlights the importance of parameter-free approach to the estimation of such cumbersome Dy-Co coupling constants. However due to the weak nature of Co-Dy exchange, multiple low lying excited states are available, so that complex Co 4 Dy cannot be expected to be a good zero-field SMM. The application of field might quench the tunnelling to some extent leading to the observation of SMM under applied field conditions since the probability of transition from ground to first excited KD is very large, as experimentally observed.

Experimental Section
All chemicals and solvents were purchased from commercial sources and used as received. Microwave assisted reactions were performed in a CEM Discover microwave reactor. Chp stands for deprotonated 6-chloro-2-hydroxypyridine (C5H3ClNO). SALOH stands for monodeprotonated 3,5-ditert-butylsalicylic acid (C15H21O3). The syntheses for complexes Ni4La and Ni4Tb are reported in our previous paper. 44      Electronics) with a monochromatic X-ray source (Aluminium K-alfa line of 1486.6 eV energy and 350 W), placed perpendicular to the analyzer axis and calibrated using the 3d5/2 line of Ag with a full width at half maximum (FWHM) of 0.8 eV. The analysed area was a circle of 0.8 mm diameter, and the selected resolution for the spectra was 187.85 eV of Pass Energy and 0.8 eV/step for the general spectra and 23.5 eV of Pass Energy and 0.1 eV/step for the spectra of the different elements. A low energy electron gun (less than 10 eV) was used in order to discharge the surface when necessary. All Measurements were made in a ultra high vacuum (UHV) chamber pressure between 5x10-9 and 2x10-8 torr. Magnetic measurements were performed at the Unitat de Mesures Magnètiques of the Universitat de Barcelona on a Quantum Design SQUID MPMS-XL magnetometer equipped with a 5 T magnet. Diamagnetic corrections for the sample holder and for the sample using Pascal's constants were applied. Hysteresis measurements were performed with an array of micro-SQUIDs. This magnetometer works in the temperature range of 0.04 to 5 K and in fields up to 1.4 T with sweeping rates as high as 0.28

Conclusions
The solvent-free microwave assisted synthesis method can be easily extended to different ligand/metal systems, in this paper we show how we can systematically change the metals in the system using this synthetic approach. The requirements are simple: ligands with low melting points that can serve as a molten reaction media on the microwave reactor. The molten ligand facilitates ion diffusion in the reaction and is key for the formation of a coordination complex. The solvent free reaction drastically reduces the use of organic solvents, being a clean method that is both cost and time efficient. In the particular case reported here, the solvent free microwave assisted synthesis has been applied to the preparation of a family of heterometallic 3d-4f M4Ln complexes, with the possibility of changing both the transition metal (M = Co(II), Ni(II)) and the lanthanide ion (Ln = La(III), Gd(III), Dy(III), Tb(III)). This produces a family of complexes with tuneable magnetic moment and magnetic anisotropy. The static and dynamic magnetic properties of the complexes reported have been studied showing that the substitution of Ni(II) for Co(II) consistently leads to better SMMs. In the Co 4 Ln series, the largest magnetic moment for Co 4 Gd as well as the lack of the spin-orbit coupling in the lanthanide ion are key for obtaining high energy barriers for the relaxation of the magnetization and to avoid relaxation by QTM. The analogues of more anisotropic lanthanide ions like Dy(III) or Tb(III) have relaxation mechanisms dominated by QTM that must be quenched in order to obtain better SMM properties.

Conflicts of interest
In accordance with our policy on Conflicts of interest please ensure that a conflicts of interest statement is included in your manuscript here. Please note that this statement is required for all submitted manuscripts. If no conflicts exist, please state that "There are no conflicts to declare".