RSC Advances PAPER Single molecule magnets of cobalt and zinc homo- Cite this: RSC Adv., 2020, 10, 45090 and heterometallic coordination polymers prepared by a one-step synthetic procedure† Núria Portolés-Gil, a Silvia Gómez-Coca, b Oriol Vallcorba, c Gregorio Marbán, d Núria Aliaga-Alcalde,ae Ana López-Periago,a José A. Ayllón *f and Concepción Domingo *a The synthesis of 1D cobalt and zinc monometallic and heterometallic coordination polymers (CPs) was carried out applying one-pot synthetic methods by using either supercritical carbon dioxide or ethanol as the solvent. A collection of four 1D CPs were thus obtained by the combination of a metal (or a mixture of metals) with the linker 1,4-bis(4-pyridylmethyl)benzene. The used metallic complexes were zinc and cobalt hexafluoroacetylacetonate, which can easily incorporate pyridine ligands in the coordination sphere of the metal centre. Independently of the used solvent, the precipitated phases involving Zn(II), i.e., homometallic CP of Zn(II) and bimetallic CP of Zn(II)/Co(II), were isostructural. Contrarily, homometallic CPs of Co(II) were precipitated as an isostructural phase of Zn(II) or with Received 26th October 2020 Accepted 7th December 2020 a different structure, depending on the used solvent. All the structures were resolved by XRD using synchrotron radiation. In addition, the magnetic properties of the new CPs involving Co(II) were studied. DOI: 10.1039/d0ra09132d Remarkably, at low temperatures with the application of an external field, they acted as field-induced rsc.li/rsc-advances single molecule magnets. Introduction a Instituto de Ciencia de Materiales de Barcelona (CSIC), Campus UAB, 08193 Bellaterra, Spain. E-mail: conchi@icmab.es Coordination polymers (CPs) or metal–organic frameworks b Departament de Quı́mica Inorgànica i Orgànica, Institut de Recerca de Quı́mica (MOFs) comprise metal nodes, usually transition metals or Teòrica i Computacional, Universitat de Barcelona, Diagonal 645, 08028 Barcelona, metallic clusters, connected in a crystalline network by multi- Spain podal organic linkers and coordination bonds. One of the c ALBA Synchrotron Light Source, 08290 Cerdanyola del Vallés, Spain unique features of these extended structures is the tunability in d Instituto de Ciencia y Tecnologı́a del Carbono (INCAR-CSIC), 33011 Oviedo, Spain e the design, based on the concept of the modication of the ICREA, Institució Catalana de Recerca i Estudis Avançats, Passeig Lluis Companys 23, composition and geometry of the repeating unit. This includes 08010 Barcelona, Spain f Universidad Autónoma de Barcelona, Dept. Quı́mica, Campus UAB, 08193 Bellaterra, not only the use of different organic linkers or metal clusters, Spain. E-mail: JoseAntonio.Ayllon@uab.cat but also the preparation of multivariate CPs with more than one † Electronic supplementary information (ESI) available: Crystallographic S-SXRD linker and/or inorganic building unit in the design.1 New data for Zn(1), Zn/Co(4) and Co(2); and Co(3); Cartesian coordinates of the functionalities and applications are described for these hybrid molecular models used to study de magnetic properties of Co(3), Co(2) and polymers, particularly in heterogeneous catalysis by designing Zn/Co(4); XPS results and discussion; Results from the t of the Cole–Cole plot, systems with more than one active center.2 One of the earliest for compounds Co(2), Co(3) and Zn/Co(4) with an applied dc eld of 1500 Oe, to a Debye model using the CCt package; calculated rst excitation energies synthetic routes described to introduce heterogeneity in CPs with (dE1) and without (DE1) spin–orbit coupling (cm1) and ZFS parameters (D was through the organic part, by pre-mixing in the solvothermal and E in cm1) and g-factors of the ground and rst excited states for the synthesis more than one bridging unit. This design is used for studied molecules at CASSCF level with OpenMolcas and Orca soware; line compounds involving exible organic linkers, which are diffi- scan performed by EDS in a single crystal of the Zn/Co(Et) sample; cult to crystallize in networks with high symmetry factors. The representation of molar magnetization vs. eld for Co(3) and Zn/Co(4); representation of frequency dependence of real and imaginary parts of the addition of an auxiliary linker, typically a bipyridine or dicar- magnetic susceptibility at 2.8 K and for different applied static dc elds for boxylic acid, increases the dimensionality and rigidity of the compound Co(2), Co(3) and Zn/Co(4); Representation of Cole–Cole plot at an framework, oen providing permanent porosity to the end applied static dc eld of 1500 Oe and for different temperatures for compound product.3 A second option to introduce heterogeneity in CPs is Co(2), Co(3) and Zn/Co(4); comparison of the susceptibility and magnetization to use a mixture of metals. Representative heterometallic curves obtained from the CASPT2 calculations and experimental data. CCDC 1950459–1950462. For ESI and crystallographic data in CIF or other electronic frameworks are schematized in Fig. 1. Pioneering studies in format see DOI: 10.1039/d0ra09132d bimetallic networks concern compounds with several inorganic 45090 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances Fig. 1 Typical heterometallic frameworks combined with the linker L. Compounds in A and B are built with pre-formed homometallic clusters (c(M1) and c(M2)) or heteronuclear clusters (c(M1M2)), respectively. Compounds in C and D are core–shell structures in which the two homometallic components have lattice match. In C, crystals are precipitated by soaking seed particles of the most stable CP [M1L–M1L–M1L]n in a solution of the second metal M2 and the ligand L. In D, sophisticated core–shell MOFs are prepared by co-precipitation of M1/L and M2/L pairs with different nucleation rate, in which the faster precipitating compound (M1L) forms the core and the second precipitates latter as the shell (M2L). Contrarily, in E the statistic co-precipitation of both phases in small domains occurs by using M1/L and M2/L pairs with similar nucleation rate. F and G describe post-synthetic transmetalation, in which crystals of the CP with the lowest stability [M2L–M2L–M2L]n are soaked in a solution of the second metal M1 in the absence of ligand. Core–shell structures are obtained at short processing times (F), while almost total replacements can be obtained at high running time (G). clusters of different geometries.4–6 Next, innite networks have study is to demonstrate the feasibility of using one-pot co- also been built with heteronuclear nodes involving two metals precipitation methods, performed in either green supercritical with different oxidation state.7–9 Finally, multilayer core–shell or CO2 (scCO2) or ethanol (EtOH), to develope heterometallic multidomain single crystals of heterometallic CPs are described systems. The system chosen for study involves a ditopic bipyr- for compounds with lattice match.10–13 Indeed, the latter has idyl derivative, the 1,4-bis(4-pyridylmethyl)benzene (bpymb) been extended to systems with 10 different divalent metals, linker, and Zn(II) and/or Co(II) hexauoro acetylacetonates as possessing a wide range of ionic radii sizes, and a structurally nodes. Zn(II) and Co(II) are cations with similar radius, valence single linker.14 Core–shell structures can also be obtained by and coordination geometry.20–24 As a consequence, this two post-synthetic transmetalation,15 a process in which an almost metals have been deeply study to form heterometallic MOFs, in complete exchange of metals ions would be possible by particular those with a zeolitic imidazolate framework, such as increasing sufficiently the processing time. ZIF-8 and ZIF-67.25–28 In addition, Co(II) cations are very inter- Transmetalation is proposed for the preparation of iso- esting from the magnetic point of view. In the expected octa- structural networks that are difficult to obtain by a direct hedral coordination environment, they have large anisotropy synthetic strategy, like those involving Ti(III), V(II), Cr(II) and and can behave as single molecule magnets (SMM) or single Fe(II) cations.16,17 It is worth mentioning that this is a single chain magnets (SCM) in one dimensional systems.29–34 In this crystal-to-single crystal process kinetically viable only for porous work, four new CPs were crystallized: one involving Zn(II), two compounds with enhanced internal diffusion.18,19 involving Co(II) and one involving the mixture of Zn(II) and Multivariate CPs are usually solvothermally prepared, either Co(II). The structures of all of them were elucidated by in routes of one-pot synthesis or by post-synthetic exchange or synchrotron X-ray diffraction. For the compounds involving cationic substitution. In general, the later can only be used for Co(II), static and dynamic magnetic studies were performed in metals that occupy chemically identical lattice positions, while order to provide insight on their magnetic behavior. Remark- the former can also be used for systems in which the position of ably, homometallic compounds involving Co(II) and hetero- the different cations is not interchangeable. The interest of this metallic Zn(II)/Co(II) act as eld-induced single molecule This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45091 RSC Advances Paper magnets (SMMs), expanding this family of appealing systems Characterization devoted to the development of high-density information storage Routine powder X-ray diffraction (PXRD) patterns of the materials, quantum computing and molecular spintronic.35–37 precipitated compounds were recorded in a Siemens D5000 diffractometer, using the Cu Ka incident radiation in the 2q Experimental section range of 5 to 30 . Synchrotron single crystal X-ray diffraction (S- SCXRD) experiments for three of the new compounds, namely Materials Zn(1), Co(3) and Zn/Co(4), were performed in the XALOC 1,4-Bis(4-pyridylmethyl)benzene (bpymb) was selected as the beamline at the ALBA synchrotron (Spain).38 Data were collected organic linker and obtained from Cymit. Zinc hexa- at 100 K with a 0.72931  A wavelength using the Dectris Pilatus uoroacetylacetonate dihydrate (Zn(hfacac)2$H2O) and cobalt 6M detector placed at 120 mm from the sample. The f-scans hexauoroacetylacetonate hydrate (Co(hfacac)2$xH2O) were were performed from 0 to 360 in steps of 0.5 and at a collec- chosen as metal complexes (Scheme 1). These reagents, ethanol tion time of 0.15 s step-1. The scan was repeated at three (EtOH) and chloroform (CHCl3) were all purchased from Sigma different k angles (0, 45 , 90 ) and merged aerwards to Aldrich. Compressed CO2 (99.95%) was supplied by Carburos increase the completeness and redundancy when possible. Data Metálicos S.A. were indexed, integrated and scaled using the XDS soware.39 The crystal structures were solved by intrinsic phasing Equipment and methods (SHELXT) and rened with SHELXL (version 2014/7)40 using Crystallization runs were carried out in two different solvents, Olex2 as graphical interface.41 For the homometallic Co(II) CP scCO2 and EtOH. Reactions in scCO2 were performed in a 5 mL precipitated in scCO2 (compound Co(2)) structure elucidation Pyrex vial placed into a 100 mL high pressure autoclave (Thar- was achieved by synchrotron powder X-ray diffraction (S-PXRD) Design). The vial was charged with ca. 100 mg of either Zn(II) or carried out on a high resolution powder diffraction end station Co(II) metal complex and 1 equivalent of bpymb, together with of the MSPD beamline (BL04) at the ALBA synchrotron. Data a small magnetic bar. Aer capping the vial with cellulose were collected at 100 and 298 K using the Si-microstrip paper, scCO2 was added up to a pressure of 15 MPa at 333 K, and MYTHEN-II detector (6 modules, 1280 channels per module, the system was stirred at 500 rpm. For the precipitation of 50 mm per channel, sample-to-detector distance 550 mm) at an heterometallic Zn(II)/Co(II) CPs, similar conditions as for the energy of 25 keV (wavelength ¼ 0.49587  A determined with the single metal compound were used, but with a 0.50 : 0.50 molar NIST Si standard). Crystallographic data for all compounds are ratio in the mixed metal source. Aer a running period of 3 h, summarized in Tables S1 and S2 in the ESI,† and the corre- the product was washed with fresh scCO2. The autoclave was sponding CIF les have been deposited in the Cambridge then depressurized and cooled down to room temperature Crystallographic Data Centre (CCDC 1950459–1950462). before recovering a dry powdered sample. For the crystallization Samples chemical composition was estimated by elemental tests completed in EtOH, each reagent (ca. 100 mg of metal analysis (E.A., Thermo Carlo Erba Flash 2000). The percentage complex and 1 equivalent of bpymb) was separately dissolved in of each metal in mixed metal structures was determined by 20 mL of alcohol, then mixed in a 100 mL ask and stirred at inductively coupled plasma mass spectrometry (ICP-MS, Agilent 500 rpm for 3 h at room temperature. In all cases, the recovered 7700x) aer digesting the solid in hydrochloric and nitric acids. samples were ltered, rinsed with fresh EtOH and air dried. In Metal distribution was assessed by energy dispersive spectros- some cases, the use of a layering technique was needed to copy (EDS, FEI Magellan 400L) in a scanning electron micros- obtain crystals suitable for the single crystal XRD study. The copy (SEM, Ultim Extreme Oxford Inst.) with high resolution used solutions were obtained by dissolving 4.5 mg of bpymb in (0.8–0.9 nm), placing the sample on a holder of silicon wafer 12 mL of EtOH : CHCl3 (25 : 75 v%) and 9 mg of metal complex and metalizing with Pt. Ex situ X ray photoelectron spectroscopy in 2 mL of EtOH. The layering was settled by placing the bpymb (XPS) was carried out in a XPS Spectrometer Kratos AXIS Supra solution at the bottom of a vial and slowly adding the metal using Al-Ka monochromatic (20 eV) radiation emitted at 225 W solution on the top. The vial was then capped and le at room (15 mA/15 kV). temperature for 1 week to allow the slow liquid–liquid diffusion Magnetic measurements were carried out in a Quantum of one solution into another. Crystals were recovered at the Design SQUID MPMS magnetometer. The analysis was per- liquid/liquid boundary area and located on the walls. formed on polycrystalline samples of the Co(II) compounds. Scheme 1 Chemical structures of metal complex M(hfacac)2 (M ¼ Zn(II) or Co(II)), and the organic linker bypmb. 45092 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances Each sample was grinded and introduced in a gelatine capsule that was mounted in a capillary tube made of polyimide for the measurement. Variable temperature magnetic susceptibility measurements between 2 and 300 K were collected under an applied eld of 3000 Oe. Reduced magnetization measurements were performed between 1.8 and 6.8 K and applied elds up to 5 T. Dynamic susceptibility measurements under different applied static elds were performed employing an oscillating ac eld of 4 Oe in the frequencies range between 1 and 1500 Hz. Diamagnetic corrections of the constituent atoms were esti- mated using Pascal's tables and the susceptibility was also corrected for the sample holder. In the case of the hetero- metallic sample Zn(II)/Co(II), the acquired data was corrected to have the same cT value at 300 K than the Co(II) isostructural sample precipitated in EtOH, giving a Zn(II) : Co(II) atomic ratio of 0.43 : 0.57. Computational details Multireference calculations were performed using two different soware packages, OpenMolcas (along with Single Aniso)42 and Orca43 Neese. The studied systems are CPs, so molecular models, including one Co(II) ion and the coordinated ligands, were created for each compound from the crystal structure to understand the magnetic properties of the subunits at a molecular level. Employed Cartesian coordinates are on Tables S3–S5.† OpenMolcas version 18.09 and Orca version 4.1.0 version were employed. With the former, the energy of the states was calculated at CASSCF level and then the SO-RASSI (Restricted Active Space State Interaction) approach was employed to mix them and obtain the nal energy states.44 All Fig. 2 PXRD patterns of: (a) precipitated samples, and (b) simulated from single-crystal data. electron ANO-RCC basis set were employed: Co (6s 5p 4d 2f), Zn (5s 4p 3d 1f), N (4s 3p 2d 1f), C (3s 2p) and H (2s).45–47 In addi- tion, the matrix elements of the transition magnetic moments have been calculated to have an estimation about the proba- Hess) procedure. The def2-TZVPP basis set was employed, bility of transition between two different states of the mole- including the auxiliary basis sets for correlation and Coulomb cules.48 Such matrix elements are calculated as proposed by the tting, for all the atoms.49,50 The employed active space in both golden Fermi rule, as the integral between the two involved cases includes seven electrons in the ve 3d orbitals of the Co(II) states using a magnetic moment operator. For Orca version ion, CAS (7,5). All states for the quartet states arising from the 4.1.0, CASSCF and NEVPT2 calculations were performed. In this 4F and 4P terms of Co(II), and all the 40 states for the doublet case spin–orbit effects were included using the quasi- state coming from the 2P, 2D (x2), 2F, 2G and 2H terms of the degenerate perturbation theory (QDPT) and scalar relativistic Co(II) ion have been included. effects were taken into account using the DKH (Douglas–Kroll– Table 1 Solvents of synthesis and composition of the different precipitated compounds Sample Solvent Compounda Crystal structure E.A. (C, H, N) [%wt] Zn (sc) scCO2 cis-[Zn(hfaac)2(bpymb)]n, Zn(1) A f. 45.52, 2.43, 3.88 Zn (Et) EtOH c. 45.46, 2.45, 3.79 Co(sc) scCO2 trans-[Co(hfaac)2(bpymb)]n, Co(2) B f. 45.71, 2.36, 3.94 c. 45.86, 2.47, 3.82 Co(Et) EtOH cis-[Co(hfaac)2(bpymb)]n, Co(3) A f. 45.69, 2.36, 3.91 c. 45.86, 2.47, 3.82 Zn/Co(sc) scCO2 cis-[Zn0.5Co0.5(hfaac)2(bpymb)], Zn/Co(4) A f. 45.7, 2.36, 3.94 Zn/Co(Et) EtOH c. 45.66, 2.46, 3.80 a The cis/trans prex make reference to the relative position of the N pyridine atoms around the metallic center. This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45093 RSC Advances Paper Results and discussion The structures of these compounds are described grouped in the two isostructural frameworks (structures A and B in Fig. 3). Structural description Structure A. Zn(1), Co(3) and Zn/Co(4) crystallize in the Table 1 describes the synthetized mono and bimetallic CPs as  space group. The crystallographic angles were very triclinic P1 a function of processing conditions. Precipitation experiments similar for the three products, with values between a ¼ 104.4– were performed in scCO2 (samples Zn(sc), Co(sc) and Zn/Co(sc)) 104.7 , b ¼ 96.0–96.4 and g ¼ 97.4–97.5 . For this structure, the and EtOH (samples Zn(Et), Co(Et) and Zn/Co(Et)) solvents. Both asymmetric unit involves an hexacoordinated metal linked to reagents, M(hfacac)2 and bpymb, were highly soluble in these two bpymb (half) molecules and two hfacac groups (Fig. 3a), solvents. Hence, precipitation experiments were carried out thus giving the stoichiometry [M(hfacac)2(bpymb)]n. The two from solution in both tested solvents. Crystal structures were bpymb linkers are located in relative cis positions, forming rst assessed by routine PXRD (Fig. 2a). The patterns of samples N–M–N angles of 93.7, 92.8 and 93.1 for Zn(1), Co(3) and Zn/ Zn(sc), Zn(Et), Co(Et), Zn/Co(sc) and Zn/Co(Et) were all similar. Co(4), respectively. As a consequence, the structure extends in On the contrary, the different PXRD pattern displayed by the zig-zag chains. For all compounds, F/H–Car and F/F weak Co(sc) sample pointed to a dissimilar structure. Differences interactions are established among chains, resulting in stable were also reected in the observed color of the precipitated 3D crystals (Fig. 3b). Despite the larger covalent radii of Co(II) samples. Any compound involving only Zn(II) exhibits a white (1.26 A) vs. Zn(II) (1.22  A),51 the unit cell volumes of the Co(3) color, while the products with pure Co(II) reect either an and Zn(1) compounds were the same (1464  A3). Contrarily, the orange color for Co(sc) or orange-brown for Co(Et). This last heterometallic compound Zn/Co(4) displayed a signicant color was also observed for the Zn/Co(sc) and Zn/Co(Et) larger value (1471  A3) than the homometallic counterparts, samples. By using the layering method in EtOH, single crys- likely originated by the slim distortion introduced in the tals were successfully produced for the S-SCXRD structure network by the presence of two metals with distinct radii. elucidation of samples Zn(Et) (and, by extension, Zn(sc)), Co(Et)  space group. Structure B. Co(2) crystallizes in the triclinic P1 and Zn/Co(Et) (and, by extension, Zn/Co(sc)), denoted as As occurs for the three isostructural compounds with frame- compounds Zn(1), Co(3) and Zn/Co(4). The unidentied struc- work A, the asymmetric unit involves hexacoordinated Co(II) ture of Co(sc) was elucidated from the powder by S-PXRD linked to two bpymb ligands and two hfacac groups (Fig. 3c), techniques (compound Co(2)). with a stoichiometry of [M(hfacac)2(bpymb)]n. However, in this Fig. 3 Schematic representation of structures: (a and b) A (Zn(1), Co(3), Zn/Co(4)), and (c and d) B, Co(2). Color legend: Zn or Co, blue; C, grey; O, red; N, violet; F, violet; H are omitted for clarity. 45094 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances case the two bpymb linkers are located in a trans positions, with Zn(II) : Co(II) of 0.2 : 0.8. The product was an heterometallic an N–Co–N angle of 180.0 . The polymer extends in 1D chains compound with stoichiometry [Zn0.15Co0.85(hfacac)2bpymb]n that interact through F/F intermolecular weak bonds (Fig. 3d). and A-type framework. Hence, even a small amount of Zn(II) Comparing Co(2) and Co(3) crystal structures with frame- present in the Co(II) scCO2 solution directs the crystallization to works B and A, respectively, it becomes evident that the struc- the structure A instead of B, which is a framework exclusively tural changes are mainly originated on the large differences of found for homometallic Co(II) in this solvent. It should be taken the values of the angle N–Co–N, of 180.0 for Co(2) and almost into account that only the mixture of Zn(1) and Co(3), both with half of this value (92.8 ) for Co(3). The rest of the structure is framework A, produces a precipitate with lattice match, not similar, since in both cases the bpymb linker adopts a Z possible for the mixture of Zn(1) and Co(2) with frameworks A conguration and the coordination of the Co(II) is completed and B, respectively. One additional advantage of the co- with two hfacac units. This change in the N–Co(II)–N angle also precipitation method is that the metal content of the resulting induces a noticeable variation of the density of the crystal, heterometallic compound can be controlled through variations larger (1.788 g cm3) in Co(2) than in Co(3) (1.664 g cm3). In in the molar ratio of the precursor mixture. both structures the shortest metal–metal distances are the EDS general spectra with nanometric resolution were intermolecular (ca. 9  A), while an intra-chain metal–metal acquired at different spots of the SEM images of the hetero- distances of ca. 16  A. metallic compound to investigate metal distribution. Elemental mapping for the Zn/Co(sc) sample are here described as an Composition and spatial arrangement of Zn(II) and Co(II) in illustrative example of the Zn/Co(4) compound (Fig. 4a). For mixed metal CPs comparison, a physical mixture was obtained by integrating powder of the isostructural Zn(sc) and Co(Et) samples (Fig. 4b). Phase purity of the different precipitated compounds was The mapping of Zn/Co(sc) shows an identical distribution of conrmed by comparison of routine PXRD patterns and simu- Zn(II) and Co(II) in the analyzed sections, which suggests that lated proles calculated from data of the single-crystal struc- both ions were evenly spread throughout the particles. The EDS tures (Fig. 2a and b). E.A. data also corroborate a high degree of gives an approximated atomic ratio of Zn(II) : Co(II) of purity in the precipitated samples, since the C, N, H values 0.51 : 0.49. Contrarily, the physical mixture displayed Zn(II) and calculated for the stoichiometry [M(hfacac)2bpymb]n match the Co(II) segregated in spots representing the different homo- experimentally found values (Table 1). Quantication of the two metallic specimens. Additionally, the EDS spectrum obtained metals in the heterometallic compound was performed by ICP- for one individual crystal of the sample Zn/Co(sc) indicates MS in several samples of the Zn/Co(4) phase prepared in either again equal and homogeneous distribution of both metals, EtOH or scCO2. The starting molar ratio of Zn(II) : Co(II) metals showing proper intermixing of Zn(II) and Co(II) atoms (Fig. S1†). was always 0.50 : 0.50. The measured molar (or equivalent It is worth pointing that EDS has a nanometric resolution. atomic) percentages were in the order of 0.50  0.10 for both Therefore, this technique is only indicating that the Zn(II) and Zn(II) and Co(II), i.e., the molar ratio values were close to the Co(II) atoms were evenly scattered around the sample at added amount of each metal. a nanometric level. However, they can be distributed in the The preparation of the heterometallic CP produces at all framework either as intercalated metals or segregated into times the Zn/Co(4) phase with framework A, indicating that this small (smaller than the EDS resolution) domains of homo- structure is the thermodynamically most stable, even the pha- metallic CPs. This behavior has been theoretically demon- ses involving pure Co(II) might crystallize in different poly- strated for other heterometallic MOFs involving Mg/Ni and Mg/ morphic crystal lattices. This fact was conrmed in an extra Cd.52 experiment performed in scCO2 with an initial molar ratio Fig. 4 EDS mapping of: (a) Zn/Co(sc) sample, and (b) physical mixture of Zn(sc) and Co(Et) in a molar ratio 0.50 : 0.50. This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45095 RSC Advances Paper Surface analysis Table 2 Results from the fitting of compounds Co(2), Co(3) and Zn/ Co(4), including D, E, gx, gy, gz and zJ parameters XPS is a powerful tool to collate information on the chemical compositions of materials surface, their coordination chemistry Condence and degradation behavior. Moreover, the technique has also Co(2) Co(3) Zn/Co(4) interval been proposed to help on the detection of the spatial position of D (cm1) 46.3 48.0 55.1 0.4 the different metals in heterometallic CPs. For compounds with E (cm1) 9.4 11.4 14.0 0.2 an even distribution of metals, the coordination environment of gx 1.73 1.86 2.09 0.02 each cation would be inuenced by the presence of adjacent gy 2.60 2.58 2.51 0.01 metals of different characteristics. Theoretically, signicant gz 3.13 3.00 2.91 0.01 shis in the binding energies would be observed in the mixed zJ (cm1) 0.012 0.017 0.019 0.001 metal compound compared to the corresponding isostructural homometallic products. However, for the samples studied in this work no shi in the binding energy of the metals in Zn/ highlighting the similar environment and isolation of the Co(II) Co(4) was observed with respect to Zn(1) and Co(3) (Fig. S2†). units in the systems. At room temperature, the cT product is The clearest spectrum was the one obtained for the Zn(sc) close to 3 cm3 K mol1 in all of the cases, larger than the ex- sample, in which the low binding energy peak in the Zn2p3/2 pected for a S ¼ 3/2 orbitally non-degenerate ground state (1.875 region was placed at 1021.9 eV. The same peak for the Zn/ cm3 K mol1). This indicates that the systems have a large Co(sc) sample appeared at 1022.1 eV, i.e., with a negligible unquenched orbital contribution, which is characteristic of binding energy shi. The deconvolutions peaks for the Co 2p3/2 Co(II) ions in pseudo-octahedral coordination.59,60,62–67 sub-region were also similar in the Co(Et) and Zn/Co(sc) sample. When lowering temperature, the cT product slowly The conclusion is that the interaction between the two different decreases to the minimum of 1.64 cm3 K mol1 observed at 2 K, metals was not signicant for XPS detection. Again, this tech- which is mainly due to spin orbit coupling (SOC). For Co(II) ions nique cannot discern between structures obtained by metals in a distorted octahedral geometry the rst order SOC would be intercalation or homometallic domains. The analysis of the important and the experimental data have been sometimes total amount of Zn(II) and Co(II) in the Zn/Co(sc) sample indi- cated a surface rich in Co(II), with a Zn(II) : Co(II) molar ratio of 0.38 : 0.62. In scCO2, this fact can reect the faster nucleation of the Zn(II) phase with respect to the isostructural Co(II) with framework A, since in this uid the most stable phase for homometallic Co(II) solutions was the one with framework B. For the compounds synthetized in this work, an evident change of color was observed as a function of the XPS measurement time, which was as clear indication of the degradation process provoked by the X-ray radiation.53 A quantitative analysis performed on the basis of the percentage of the different atoms in each sample (Table S6†) allows to estimate that the percentage of degradation was higher than 50 mol% for the three studied samples Zn(sc), Co(Et) and Zn/ Co(sc). This fact makes difficult to mine extra information from XPS data. Magnetic properties CPs formed by Co(II) subunits are good candidates as single chain magnets (SCMs)54,55 or single molecule magnets (SMMs)56,57 due to the intrinsic magnetic anisotropy of the Co(II) in various geometries, including the pseudo-octahedral,58–61 and the nature of the resulting exchange interaction through their linkers. The studied systems have a small distortion with respect to the ideal octahedral geometry as showed by their shape measurement, 0.059, 0.465 and 0.520 for Co(2), Co(3) and Zn/Co(4) respectively. Herein, the magnetic properties of the reported new CPs are explored. The variable temperature susceptibility and magnetization Fig. 5 (a) cT vs. temperature curve for compounds Co(2), Co(3) and curves of Co(2), Co(3) and Zn/Co(4) are similar and depicted in Zn/Co(4) at an applied field of 3000 Oe, and (b) M vs. field curves at Fig. 5, S3 and S4.† Despite the structural differences among different temperatures for Co(2). The lines are the results of the fit with them, the magnetic properties are almost identical, PHI package. 45096 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances tted with the corresponding Hamiltonian. However, the large Co(II) complexes in pseudo-octahedral geometry and also in energy difference between the states implies that only the lowest other Co(II) CPs.77–82 Kramers doublets are populated and, phenomenologically, the Dynamic susceptibility measurements were performed to zero-eld splitting (ZFS) Hamiltonian can be used to study the study the possibility of slow relaxation of the magnetization. magnetic properties of this compounds.67–71 Herein, we Without an applied static dc eld, no c0 signal was observed; employed the later approach for sake of simplicity and to allow however, when an external dc eld was used, a dependence of c0 further comparison with published systems.58,60–67 This way, the appeared. Therefore, the dependence with the eld for the ZFS Hamiltonian, as implemented in the PHI package (eqn (1)), different compounds was studied (Fig. S5–S7†). Under the latest was employed for the simultaneous t of the experimental conditions, a maximum in c0 appeared around 500 Hz for the susceptibility and magnetization curves. three different compounds indicating that they all behave as D ^0  eld induced mononuclear SMMs. The maximum slightly H¼ ^2 2 þ mB gx S^x Bx þ gy S^y By þ gz S^z Bz O þ EO (1) moves to higher frequencies when the applied dc eld was 3 2 increased. A common static dc eld of 1500 Oe was selected for where S is the spin (and their spin operators), D is the axial the study of the temperature dependence of the dynamic anisotropy, E the rhombic anisotropy, mB is the Bohr magneton, susceptibility for the three compounds (Fig. 6). Similar B is the magnetic eld vector and Ôkq is the equivalent Stevens frequency dependence of the out of phase component of the operator. The rst two terms relate to the crystal-eld Hamil- susceptibility was observed for both Co(2) and Co(3) despite tonian (following the operator equivalent technique), and the their different geometry. However, for the magnetically diluted last term is connected to the Zeeman Hamiltonian. In addition, sample Zn/Co(4) the maximum in c0 moves to lower frequen- due to the polymeric character of the compounds, the possible cies, which might be due to the decrease of dipolar interactions. antiferromagnetic interactions between the subunits should be This result indicates that, although the distance between considered despite the long size of the linker. This possible paramagnetic centers was already large, and the magnetic effect has been included in the t of the magnetic data by dilution stoichiometry still leaves several nearest neighbors applying the van Vleck equation in the mean eld approxima- (assuming a random distribution of the metal ions), part of the tion as implemented in the PHI package.72 dipolar interactions between Co(II) centers are being reduced. The results of the t, Table 2 and Fig. 5, show that this kind The Cole–Cole plots were tted to a generalized Debye model of compounds has easy plane anisotropy with a positive and using CCt package (Fig. S8–S10 and Tables S7–S9†). Temper- relatively large, around 50 cm1, D values and with two gi atures from 1.8 to 9 K were analyzed, however for temperatures components larger than ge. Besides the 1D nature of the larger than 4.6 K less than half semicircles were observed in the compounds the interaction between units is very small with no Cole–Cole plot and consequently that data was not included in signicant difference between the magnetically diluted sample the analysis. The obtained a values were always smaller than Zn/Co(4) and the others. This indicates that the linker effec- 0.16, indicating a narrow distribution of relaxation times. tively separates the magnetic subunits and no SCM behavior Several different relaxation processes are usually proposed would be expected, only SMM.73–76 The observed behavior and for mononuclear SMMs.60,83 The Arrhenius expression (eqn (2)) anisotropic values are similar to other published mononuclear was employed to t the high temperature region of the analyzed data to account for a thermally activated process, the so-called Fig. 6 Frequency dependence of: (a) real and (b) imaginary parts of the magnetic susceptibility at an applied static dc field of 1500 Oe and for different temperatures for compounds Co(2), Co(3) and Zn/Co(4). The lines are a guide for the eye. This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45097 RSC Advances Paper Table 3 Fitting parameters obtained from the fit of the relaxation time using eqn (3) In all three cases, the QTM is not responsible for the relax- ation process because of the applied static dc eld. The direct Co(2) Co(3) Zn/Co(4) process leads the spin relaxation at low temperatures while the A (s1 K1) 9.57  105 1.35  106 6.67  105 Raman process does it at higher temperatures. Despite the B (s1) 8.8  1010 0 0 similitude between Co(2) and Co(3) curves, the results of the t C (s1 Kn) 124.5 12.8 14.4 are different, especially in the Raman process; the pre- N 2.83 4.42 4.34 exponential term decreases one order of magnitude from Co(2) to Co(3), while the exponent increases. The exponent for Kramers ions should be 9, however values between 1 and 6 are Orbach process. The obtained energy barriers are similar for the usually observed and accepted in this type of compounds and studied compounds and they are found in the range of 5.9 and attributed to the presence of acoustic and optical phonons.84 11.3 cm1, with s0 values in the order of 105 to 106. It should The differences between Co(2) and Co(3) curves may relate to be noticed that the energy barrier should correspond to the the different structure and packing. The main differences energy difference between the ground and rst excited doublet between Co(3) and Zn/Co(4) are observed at low temperature, states. That energy difference can be estimated from the D where the direct process is the predominant. The smaller A values obtained in the simultaneous t of the susceptibility and value for Zn/Co(4) can be attributed to the magnetic dilution magnetization (vide supra) and from ab initio calculations (vide and the decrease of dipolar interactions. Overall, the systems infra). In both cases the energy difference between those states have similar magnetic behaviors; the discrepancy found in is larger than the obtained here. It indicates that the relaxation some of the parameters may relate to the magnetic dilution and of the spin is not occurring via an Orbach process and other the lower concentration of the Co(II) ions due to the employed process or processes are the main responsible. ratio and despite the non-homogeneous distribution indicated by XPS of the Zn/Co ions in the samples. s1 ¼ s01eUeff/kBT (2) To gain insight on the magnetic properties of the studied In this regard, the principal spin relaxation processes are the compounds, multireference calculations using OpenMolcas quantum tunneling of the magnetization (QTM), direct and and ORCA soware packages (see Computational details) were Raman processes. All of them have been considered for the t of carried out. A molecular model was employed for all the the dependence of the relaxation time with temperature in eqn compounds and the geometries obtained from the crystal (3), Table 3 and Fig. 7. The rst term accounts for the direct structures were used, where the bridging ligand bpymb was term, with a linear dependence with temperature, the second modelled as 4-methylpyridine. Fig. 8 shows the employed corresponds to the temperature independent QTM term and the models for compounds Co(2) and Co(3). For Co(II) complexes, last one describes the Raman term, with a temperature n the dynamic correlation can be relevant so CASPT2 results are exponent that can have different values depending on the shown in the manuscript while CASSCF are in the ESI (Table system and the available phonons.60,83 S10†). In the past, some of us demonstrated that octahedral Co(II) s1 ¼ AH4T + B + CTn (3) compounds can present positive or negative large D values depending on the splitting of the d orbitals by the crystal eld.66 Our predictions were based on the analysis of the Dii compo- nents of the tensor. The qualitative model shows that the D value depends on the orbitals involved in the excitation and the Fig. 7 Temperature dependence of the relaxation time for Fig. 8 Graphical representation of the employed models in the mul- compounds Co(2), Co(3) and Zn/Co(4). The solid lines are the result of tireference calculations, compounds Co(2) and Co(3), left and right, the fit with eqn (3). respectively. 45098 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances Table 4 Calculated first excitation energies with (dE1) and without (DE1) spin–orbit coupling (cm1) and ZFS parameters (D and E in cm1) and g- factors of the ground and first excited states for the studied molecules at CASPT2 (NEVPT2) level Co(2) Co(3) Zn/Co(4) DE1 (cm1) 302 (374) 680 (591) 699 (531) dE1 (cm1) 215 (204) 137 (165) 141 (180) D (cm1) 106.2 (99.3) 65.4 (75.8) 66.8 (81.0) |E| (cm1) 10.3 (12.9) 11.4 (18.6) 12.8 (22.4) KD1 gx 5.48 (2.60) 2.23 (1.99) 2.15 (1.93) gy 4.31 (3.74) 3.50 (2.91) 3.37 (2.66) gz 3.03 (6.31) 6.44 (7.13) 6.58 (7.41) KD2 gx 0.40 (0.98) 1.23 (1.76) 1.33 (1.91) gy 0.52 (1.17) 1.33 (1.89) 1.45 (2.15) gz 5.40 (5.39) 5.87 (5.48) 5.79 (5.28) Fig. 9 Calculated states energies as function of their average projected magnetic moment for the first and second KDs and transition proba- bilities between states. Values above 0.1 indicate an efficient spin relaxation mechanism. The dashed green arrows correspond to the tunneling process, dashed purple arrow shows the hypothetical Orbach relaxation process and the red arrows indicate the transition between the ground and excited Kramer's doublets (solid) and the relaxation pathway to the ground state with the reversed spin (dashed). energy different between them. This way, small energy differ- obtained from the CASPT2 calculations with the experimental ences results into large D values. If the orbitals involved in the ones matches well (Fig. S11†). rst excitation have the same |ml| value, then D will be negative, The different transition probabilities have been computed and as the largest contribution will be Dzz. On the contrary, when the are shown in Fig. 9. In all the cases, the transition probabilities of orbitals involved in the rst excitation have a ml change of 1, both tunneling and Orbach processes are large. The large then the largest contributions will be Dxx and Dyy and D will be tunneling probabilities rationalize the absence of SMM behavior at positive.66,85 The main calculated parameters are collected in zero eld in this family of compounds. Once an external static eld Table 4. It can be observed that the rst excitation energy before is applied, an energy barrier of more than 140 cm1 is expected SOC is small, which gives to a large energy difference between due to the also large transition probabilities obtained. However, the ground and rst excited KDs. In addition, the ground state gi the experimental values do not present such pattern. This is factors indicates an easy plane character that, in some cases, is common in mononuclear complexes due to the presence other closer to an intermediate situation between easy plane and easy relaxation processes not included in the calculations. Those indi- axes, providing three gi factor values larger than 2 and ranging cate that computational results agree with experimental data, between 2.1 and 6.6. The large D and E values and the gi factors rationalizing the anisotropy of these CPs. The easy plane anisot- of the ground state allow us to describe the system as rhombic ropy is conrmed; however, as it is seen from the t of the dynamic anisotropy. Although the excited state in all the cases is axial. susceptibility measurements, additional relaxation processes, When comparing with the experimental values obtained from such as Raman and direct terms, are the responsible of the the simultaneous t of the susceptibility and magnetization observed spin relaxation at the lowest temperatures. data can be observed that the anisotropy of this compounds with a relatively large D and E values is well described. However, the D and gi values are overestimated, probably due to the Conclusions intrinsic limitations of the calculations and the created molecular model. Although it is worth mentioning that the Multivariate CPs involving two different metals (i.e., Zn(II) and comparison of the susceptibility and magnetization curves Co(II), with similar radius and equal valence) and a ditopic bipyridyl derivative linker were prepared by a one-pot co- This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45099 RSC Advances Paper precipitation method performed in either scCO2 or EtOH heterogeneous catalysts, Catal. Sci. Technol., 2016, 6, 5238– solvent. Four new CPs were thus crystallized: one homometallic 5261, DOI: 10.1039/c6cy00695g. involving Zn(II), two homometallic involving Co(II) and one 3 D. N. Bunck and W. R. Dichtel, Mixed Linker Strategies for heterometallic involving the mixture of Zn(II) and Co(II). Organic Framework Functionalization, Chem.–Eur. J., 2013, Synchrotron XRD characterization was used to elucidate the 19, 818–827, DOI: 10.1002/chem.201203145. crystal structures, but could not discriminate between both 4 B. Tu, Q. Pang, E. Ning, W. Yan, Y. Qi, D. Wu and Q. Li, metals in the framework. EDS and XPS analysis indicated that Heterogeneity within a Mesoporous Metal–Organic the metals were evenly mixed along the heterometallic sample. Framework with Three Distinct Metal-Containing Building Static and dynamic magnetic studies were performed in order to Units, J. Am. Chem. Soc., 2015, 137, 13456–13459, DOI: provide insight on the magnetic behavior of Co(II) compounds 10.1021/jacs.5b07687. linked to metal distribution in the structure. For the three CPs 5 D. Ghoshal, A. K. Ghosh, T. K. Maji, J. Ribas, G. Mostafa, involving Co(II), despite the structural differences, the magnetic E. Zangrando and N. Ray Chaudhuri, Different topologies properties were almost identical, highlighting the similar in heterometallic frameworks of copper(II) with bridging environment and isolation of the Co(II) units in each system. ligand: Syntheses, crystal structures, thermal and magnetic The used linker effectively separates the magnetic subunits and properties, Inorg. Chim. Acta, 2006, 359, 593–602, DOI: no single chain magnet behavior was observed. However, 10.1016/j.ica.2005.09.056. homometallic compounds with Co(II) and heterometallic Zn(II)/ 6 S. Yuan, J.-S. Qin, J. Li, L. Huang, L. Feng, Y. Fang, C. Lollar, Co(II) act as eld-induced single molecular magnets. J. Pang, L. Zhang, D. Sun, A. Alsalme, T. Cagin and H.-C. Zhou, Retrosynthesis of multi-component metal– Conflicts of interest organic frameworks, Nat. Commun., 2018, 9, 808, DOI: 10.1038/s41467-018-03102-5. There are no conicts to declare. 7 Y. Wang, B. Bredenkötter, B. Rieger and D. Volkmer, Two- dimensional metal–organic frameworks (MOFs) Author contributions constructed from heterotrinuclear coordination units and 4,40 -biphenyldicarboxylate ligands, Dalton Trans., 2007, Conceptualization: C. Domingo, J. A. Ayllón; funding acquisi- 689–696, DOI: 10.1039/B609733B. tion: C. Domingo, A. López-Periago, N. Aliaga; Investigation: N. 8 J.-P. Zhang, S. K. Ghosh, J.-B. Lin and S. Kitagawa, New Portolés, A. M. López; formal analysis: O. Valcorba, S. Gómez- Heterometallic Carboxylate Frameworks: Synthesis, Coca, G. Marbán; project administration: C. Domingo, N. Structure, Robustness, Flexibility, and Porosity, Inorg. Aliaga-Alcalde; writing-review & editing: C. Domingo, J. A. Chem., 2009, 48, 7970–7976, DOI: 10.1021/ic900919w. Ayllón, N. Aliaga-Alcalde, S. Gómez-Coca. 9 Y. Zhang, B. Chen, F. R. Fronczek and A. W. Maverick, A nanoporous Ag–Fe mixed-metal–organic framework Funding sources exhibiting single-crystal-to-single-crystal transformations upon guest exchange, Inorg. Chem., 2008, 47, 4433–4435, Ministerio de Ciencia, Innovación y Universidades; Generalitat DOI: 10.1021/ic800183v. de Catalunya. 10 S. Furukawa, K. Hirai, K. Nakagawa, Y. Takashima, R. Matsuda, T. Tsuruoka, M. Kondo, R. Haruki, D. Tanaka, Acknowledgements H. Sakamoto, S. Shimomura, O. Sakata and S. Kitagawa, Heterogeneously Hybridized Porous Coordination Polymer The research reported here was supported by the Spanish Crystals: Fabrication of Heterometallic Core–Shell Single Ministerio de Ciencia, Innovación y Universidades (grants Crystals with an In-Plane Rotational Epitaxial Relationship, CTQ2017-83632, PGC2018-093863-B-C21, MAT2016-77852-C2-1- Angew. Chem., Int. Ed., 2009, 48, 1766–1770, DOI: 10.1002/ R). C. D., N. A. A., A. L. P., an N. P. G. acknowledge the nancial anie.200804836. support through the “Severo Ochoa” program for Centres of 11 H. J. Lee, Y. J. Cho, W. Cho and M. Oh, Controlled Isotropic Excellence in R&D (SEV-2015-0496). S. G. C. thanks the Gen- or Anisotropic Nanoscale Growth of Coordination Polymers: eralitat de Catalunya for a Beatriu de Pinòs grant (2017 BP Formation of Hybrid Coordination Polymer Particles, ACS 00080). N. A. A. thanks the Generalitat de Catalunya for the Nano, 2013, 7, 491–499, DOI: 10.1021/nn304597h. grant 2017SGR1277. 12 R. Sibille, T. Mazet, B. Malaman, Q. Wang, E. Didelot and M. François, Site-Dependent Substitutions in Mixed-Metal References Metal–Organic Frameworks: A Case Study and Guidelines for Analogous Systems, Chem. Mater., 2015, 27, 133–140, 1 H. Furukawa, U. Müller and O. M. Yaghi, “Heterogeneity DOI: 10.1021/cm503570f. within Order” in Metal–Organic Frameworks, Angew. 13 T. Tanasaro, K. Adpakpang, S. Ittisanronnachai, Chem., Int. Ed., 2015, 54, 3417–3430, DOI: 10.1002/ K. Faungnawakij, T. Butburee, S. Wannapaiboon, anie.201410252. M. Ogawa and S. Bureekaew, Control of Polymorphism of 2 A. Dhakshinamoorthy, A. M. Asiri and H. Garcia, Mixed- Metal–Organic Frameworks Using Mixed-Metal Approach, metal or mixed-linker metal organic frameworks as 45100 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances Cryst. Growth Des., 2018, 18, 16–21, DOI: 10.1021/ Am. Chem. Soc., 2015, 137, 1572–1580, DOI: 10.1021/ acs.cgd.7b01193. ja511539a. 14 L. J. Wang, H. Deng, H. Furukawa, F. Gándara, K. E. Cordova, 26 Y.-Z. Chen, C. Wang, Z.-Y. Wu, Y. Xiong, Q. Xu, S.-H. Yu and D. Peri and O. M. Yaghi, Synthesis and Characterization of H.-L. Jiang, From Bimetallic Metal–Organic Framework to Metal–Organic Framework-74 Containing 2, 4, 6, 8, and 10 Porous Carbon: High Surface Area and Multicomponent Different Metals, Inorg. Chem., 2014, 53, 5881–5883, DOI: Active Dopants for Excellent Electrocatalysis, Adv. Mater., 10.1021/ic500434a. 2015, 27, 5010–5016, DOI: 10.1002/adma.201502315. 15 C. G. Carson, J. Ward, X. T. Liu, J. Schwartz, R. A. Gerhardt 27 D. Saliba, M. Ammar, M. Rammal, M. Al-Ghoul and and R. Tannenbaum, Dopant-Controlled Crystallization in M. Hmadeh, Crystal Growth of ZIF-8, ZIF-67, and Their Metal–Organic Frameworks: The Role of Copper(II) in Zinc Mixed-Metal Derivatives, J. Am. Chem. Soc., 2018, 140, 1,4-Benzenedicarboxylate, J. Phys. Chem. C, 2012, 116, 1812–1823, DOI: 10.1021/jacs.7b11589. 15322–15328, DOI: 10.1021/jp2117032. 28 F. Hillman, J. M. Zimmerman, S.-M. Paek, M. R. A. Hamid, 16 M. Kim, J. F. Cahill, H. Fei, K. A. Prather and S. M. Cohen, W. T. Lim and H.-K. Jeong, Rapid microwave-assisted Postsynthetic Ligand and Cation Exchange in Robust synthesis of hybrid zeolitic–imidazolate frameworks with Metal–Organic Frameworks, J. Am. Chem. Soc., 2012, 134, mixed metals and mixed linkers, J. Mater. Chem. A, 2017, 5, 18082–18088, DOI: 10.1021/ja3079219. 6090–6099, DOI: 10.1039/C6TA11170J. 17 C. K. Brozek and M. Dincă, Ti 3+ -, V2+/+3, Cr 2+/3, Mn 2+ , 29 L. Bogani, A. Vindigni, R. Sessoli and D. Gatteschi, Single and Fe 2+ -Substituted MOF-5 and Redox Reactivity in Cr- chain magnets: where to from here?, J. Mater. Chem., 2008, and Fe-MOF-5, J. Am. Chem. Soc., 2013, 135, 12886–12891, 18, 4750–4758, DOI: 10.1039/b807824f. DOI: 10.1021/ja4064475. 30 A. Caneschi, D. Gatteschi, N. Lalioti, C. Sangregorio, 18 S. Das, H. Kim and K. Kim, Metathesis in Single Crystal: R. Sessoli, G. Venturi, A. Vindigni, A. Rettori, M. G. Pini Complete and Reversible Exchange of Metal Ions and M. A. Novak, Cobalt(II)-Nitronyl Nitroxide Chains as Constituting the Frameworks of Metal–Organic Molecular Magnetic Nanowires, Angew. Chem., Int. Ed., Frameworks, J. Am. Chem. Soc., 2009, 131, 3814–3815, DOI: 2001, 40, 1760–1763, DOI: 10.1002/1521-3773(20010504) 10.1021/ja808995d. 40:9<1760::AID-ANIE17600>3.0.CO;2-U. 19 C. Hon Lau, R. Babarao and M. R. Hill, A route to drastic 31 X. Liu, L. Sun, H. Zhou, P. Cen, X. Jin, G. Xie, S. Chen and increase of CO2 uptake in Zr metal organic framework Q. Hu, Single-Ion-Magnet Behavior in a Two-Dimensional UiO-66, Chem. Commun., 2013, 49, 3634–3636, DOI: Coordination Polymer Constructed from Co II Nodes and 10.1039/c3cc40470f. a Pyridylhydrazone Derivative, Inorg. Chem., 2015, 54, 20 X. Song, T. K. Kim, H. Kim, D. Kim, S. Jeong, H. R. Moon and 8884–8886, DOI: 10.1021/acs.inorgchem.5b01651. M. S. Lah, Post-Synthetic Modications of Framework Metal 32 Y.-Y. Zhu, M.-S. Zhu, T.-T. Yin, Y.-S. Meng, Z.-Q. Wu, Ions in Isostructural Metal–Organic Frameworks: Core–Shell Y.-Q. Zhang and S. Gao, Cobalt(II) Coordination Polymer Heterostructures via Selective Transmetalations, Chem. Exhibiting Single-Ion-Magnet-Type Field-Induced Slow Mater., 2012, 24, 3065–3073, DOI: 10.1021/cm301605w. Relaxation Behavior, Inorg. Chem., 2015, 54, 3716–3718, 21 X. Song, S. Jeong, D. Kim and M. S. Lah, Transmetalations in DOI: 10.1021/ic5029299. two metal–organic frameworks with different framework 33 A. Y. Chahine, W. Phonsri, K. S. Murray, D. R. Turner and exibilities: Kinetics and core–shell heterostructure, S. R. Batten, Coordination polymers of a bis-isophthalate CrystEngComm, 2012, 14, 5753–5756, DOI: 10.1039/ bridging ligand with single molecule magnet behaviour of c2ce26115d. the CoII analogue, Dalton Trans., 2020, 49, 5241–5249, 22 W. R. Lee, D. W. Ryu, W. J. Phang, J. H. Park and C. S. Hong, DOI: 10.1039/c9dt04606b. Charge effect of foreign metal ions and the crystal growth 34 P. Konieczny, A. B. Gonzalez-Guillén, K. Luberda-Durnaś, process in hybridized metal–organic frameworks, Chem. E. Čižmár, R. Pełka, M. Oszajca and W. Łasocha, 1D Commun., 2012, 48, 10847–10849, DOI: 10.1039/c2cc35526d. coordination polymer (OPD) 2 Co II SO 4 showing SMM 23 H. Fei, J. F. Cahill, K. A. Prather and S. M. Cohen, Tandem behaviour and multiple relaxation modes, Dalton Trans., Postsynthetic Metal Ion and Ligand Exchange in Zeolitic 2019, 48, 7560–7570, DOI: 10.1039/C9DT00624A. Imidazolate Frameworks, Inorg. Chem., 2013, 52, 4011– 35 D. Gatteschi, R. Sessoli and J. Villain, Molecular 4016, DOI: 10.1021/ic400048g. Nanomagnets, Oxford University PressEditor, 2006, DOI: 24 D. Sun, F. Sun, X. Deng and Z. Li, Mixed-Metal Strategy on 10.1093/acprof:oso/9780198567530.001.0001. Metal–Organic Frameworks (MOFs) for Functionalities 36 M. Atzori and R. Sessoli, The Second Quantum Revolution: Expansion: Co Substitution Induces Aerobic Oxidation of Role and Challenges of Molecular Chemistry, J. Am. Chem. Cyclohexene over Inactive Ni-MOF-74, Inorg. Chem., 2015, Soc., 2019, 141, 11339–11352, DOI: 10.1021/jacs.9b00984. 54, 8639–8643, DOI: 10.1021/acs.inorgchem.5b01278. 37 E. Coronado, Molecular magnetism: from chemical design 25 J. Tang, R. R. Salunkhe, J. Liu, N. L. Torad, M. Imura, to spin control in molecules, materials and devices, Nat. S. Furukawa and Y. Yamauchi, Thermal Conversion of Rev. Mater., 2020, 5, 87–104, DOI: 10.1038/s41578-019-0146- Core–Shell Metal–Organic Frameworks: A New Method for 8. Selectively Functionalized Nanoporous Hybrid Carbon, J. 38 J. Juanhuix, F. Gil-Ortiz, G. Cunı́, C. Colldelram, J. Nicolás, J. Lidón, E. Boter, C. Ruget, S. Ferrer and J. Benach, This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45101 RSC Advances Paper Developments in optics and performance at BL13-XALOC, 50 F. Weigend, Accurate Coulomb-tting basis sets for H to Rn, the macromolecular crystallography beamline at the Alba Phys. Chem. Chem. Phys., 2006, 8, 1057–1065, DOI: 10.1039/ Synchrotron, J. Synchrotron Radiat., 2014, 21, 679–689, DOI: b515623h. 10.1107/S160057751400825X. 51 B. Cordero, V. Gómez, A. E. Platero-Prats, M. Revés, 39 W. Kabsch, XDS, Acta Crystallogr., Sect. D: Biol. Crystallogr., J. Echeverrı́a, E. Cremades, F. Barragán and S. Alvarez, 2010, 66, 125–132, DOI: 10.1107/S0907444909047337. Covalent radii revisited, Dalton Trans., 2008, 2832–2838, 40 G. M. Sheldrick, Crystal structure renement with SHELXL, DOI: 10.1039/b801115j. Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3–8, DOI: 52 J. D. Howe, C. R. Morelock, Y. Jiao, K. W. Chapman, 10.1107/S2053229614024218. K. S. Walton and D. S. Sholl, Understanding Structure, 41 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. K. Howard and Metal Distribution, and Water Adsorption in Mixed-Metal H. Puschmann, OLEX2: a complete structure solution, MOF-74, J. Phys. Chem. C, 2017, 121, 627–635, DOI: renement and analysis program, J. Appl. Crystallogr., 10.1021/acs.jpcc.6b11719. 2009, 42, 339–341, DOI: 10.1107/S0021889808042726. 53 A. Cano, L. Lartundo-Rojas, A. Shchukarev and E. Reguera, 42 I. F. Galván, M. Vacher, A. Alavi, C. Angeli, F. Aquilante, Contribution to the coordination chemistry of transition J. Autschbach, J. J. Bao, S. I. Bokarev, N. A. Bogdanov, metal nitroprussides: a cryo-XPS study, New J. Chem., 2019, R. K. Carlson, L. F. Chibotaru, J. Creutzberg, N. Dattani, 43, 4835–4848, DOI: 10.1039/C9NJ00141G. M. G. Delcey, S. S. Dong, A. Dreuw, L. Freitag, L. M. Frutos, 54 M. Wang, X. Gou, W. Shi and P. Cheng, Single-chain L. Gagliardi, F. Gendron, A. Giussani, L. González, G. Grell, magnets assembled in cobalt(II) metal–organic M. Guo, C. E. Hoyer, M. Johansson, S. Keller, S. Knecht, frameworks, Chem. Commun., 2019, 55, 11000–11012, DOI: G. Kovačević, E. Källman, G. Li Manni, M. Lundberg, 10.1039/c9cc03781k. Y. Ma, S. Mai, J. P. Malhado, P. Å. Malmqvist, 55 B. Sieklucka, D. Pinkowicz, K. S. Pedersen, A. Vindigni, P. Marquetand, S. A. Mewes, J. Norell, M. Olivucci, R. Sessoli, C. Coulon and R. Clérac, Molecular Magnetic M. Oppel, Q. M. Phung, K. Pierloot, F. Plasser, M. Reiher, Materials Molecular Multiferroics, ed. B. Sieklucka and A. M. Sand, I. Schapiro, P. Sharma, C. J. Stein, D. Pinkowicz, 2016, ch. 6, DOI: 10.1002/9783527694228. L. K. Sørensen, D. G. Truhlar, M. Ugandi, L. Ungur, 56 K. S. Pedersen, J. Bendix and R. Clérac, Single-molecule A. Valentini, S. Vancoillie, V. Veryazov, O. Weser, magnet engineering: building-block approaches, Chem. T. A. Wesołowski, P.-O. Widmark, S. Wouters, A. Zech, Commun., 2014, 50, 4396–4415, DOI: 10.1039/C4CC00339J. J. P. Zobel and R. Lindh, OpenMolcas: From Source Code 57 N. T. Madhu, J.-K. Tang, I. J. Hewitt, R. Clérac, to Insight, J. Chem. Theory Comput., 2019, 15, 5925–5964, W. Wernsdorfer, J. Van Slageren, C. E. Anson and DOI: 10.1021/acs.jctc.9b00532. A. K. Powell, What makes a single molecule magnet?, 43 F. Neese, The ORCA program system, Wiley Interdiscip. Rev.: Polyhedron, 2005, 24, 2864–2869, DOI: 10.1016/ Comput. Mol. Sci., 2012, 2, 73–78, DOI: 10.1002/wcms.81. j.poly.2005.03.015. 44 P. Å. Malmqvist, B. O. Roos and B. Schimmelpfennig, The 58 L. Chen, J. Song, W. Zhao, G. Yi, Z. Zhou, A. Yuan, Y. Song, restricted active space (RAS) state interaction approach Z. Wang and Z.-W. Ouyang, A mononuclear ve-coordinate with spin–orbit coupling, Chem. Phys. Lett., 2002, 357, 230– Co(II) single molecule magnet with a spin crossover 240, DOI: 10.1016/S0009-2614(02)00498-0. between the S ¼ 1/2 and 3/2 states, Dalton Trans., 2018, 47, 45 B. O. Roos, R. Lindh, P.-Å. Malmqvist, V. Veryazov, 16596–16602, DOI: 10.1039/C8DT03783C. P.-O. Widmark and A. C. Borin, New Relativistic Atomic 59 F. El-Khatib, B. Cahier, F. Shao, M. López-Jordà, R. Guillot, Natural Orbital Basis Sets for Lanthanide Atoms with E. Rivière, H. Hafez, Z. Saad, J.-J. Girerd, N. Guihéry and Applications to the Ce Diatom and LuF 3, J. Phys. Chem. A, T. Mallah, Design and Magnetic Properties of 2008, 112, 11431–11435, DOI: 10.1021/jp803213j. a Mononuclear Co(II) Single Molecule Magnet and Its 46 B. O. Roos, R. Lindh, P. Å. Malmqvist, V. Veryazov and Antiferromagnetically Coupled Binuclear Derivative, Inorg. P. O. Widmark, Main Group Atoms and Dimers Studied Chem., 2017, 56, 4601–4608, DOI: 10.1021/ with a New Relativistic ANO Basis Set, J. Phys. Chem. A, acs.inorgchem.7b00205. 2004, 108, 2851–2858, DOI: 10.1021/jp031064+. 60 R. Dı́az-Torres, M. Menelaou, O. Roubeau, A. Sorrenti, 47 P. O. Widmark, P. Å. Malmqvist and B. O. Roos, Density G. Brandariz-de-Pedro, E. C. Sañudo, S. J. Teat, J. Fraxedas, matrix averaged atomic natural orbital (ANO) basis sets for E. Ruiz and N. Aliaga-Alcalde, Multiscale study of correlated molecular wave functions, Theor. Chim. Acta, mononuclear Co II SMMs based on curcuminoid ligands, 1990, 77, 291–306, DOI: 10.1007/BF01120130. Chem. Sci., 2016, 7, 2793–2803, DOI: 10.1039/C5SC03298A. 48 L. Ungur and L. F. Chibotaru, Strategies toward High- 61 M. R. Saber and K. R. Dunbar, Ligands effects on the Temperature Lanthanide-Based Single-Molecule Magnets, magnetic anisotropy of tetrahedral cobalt complexes, Inorg. Chem., 2016, 55, 10043–10056, DOI: 10.1021/ Chem. Commun., 2014, 50, 12266–12269, DOI: 10.1039/ acs.inorgchem.6b01353. c4cc05724d. 49 A. Schaefer, C. Huber and R. Ahlrichs, Fully optimized 62 Y.-F. Deng, M. K. Singh, D. Gan, T. Xiao, Y. Wang, S. Liu, contracted Gaussian basis sets of triple zeta valence quality Z. Wang, Z. Ouyang, Y.-Z. Zhang and K. R. Dunbar, for atoms Li to Kr, J. Chem. Phys., 1994, 100, 5829–5835, Probing the Axial Distortion Effect on the Magnetic DOI: 10.1063/1.467146. 45102 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances Anisotropy of Octahedral Co(II) Complexes, Inorg. Chem., polynuclear d - and f -block complexes, J. Comput. Chem., 2020, 59, 7622–7630, DOI: 10.1021/acs.inorgchem.0c00531. 2013, 34, 1164–1175, DOI: 10.1002/jcc.23234. 63 L. Rigamonti, N. Bridonneau, G. Poneti, L. Tesi, L. Sorace, 73 J.-Y. Wang, Y. Shi, D.-L. Tao, G.-Y. Yin and Q.-B. Bo, 2D chain D. Pinkowicz, J. Jover, E. Ruiz, R. Sessoli and A. Cornia, A layer versus 1D chain: rigid aromatic benzoate disassembling Pseudo-Octahedral Cobalt(II) Complex with exible alicyclic dicarboxylate-based lanthanide Bispyrazolylpyridine Ligands Acting as a Zero-Field Single- coordination polymers with enhanced photoluminescence Molecule Magnet with Easy Axis Anisotropy, Chem.–Eur. J., and characteristic single-molecule magnet behavior, 2018, 24, 8857–8868, DOI: 10.1002/chem.201801026. CrystEngComm, 2020, 22, 4449–4467, DOI: 10.1039/ 64 L. Váhovská, S. Vitushkina, I. Potočňák, Z. Trávnı́ček and D0CE00583E. R. Herchel, Effect of linear and non-linear pseudohalides 74 M. Orts-Arroyo, I. Castro, F. Lloret and J. Martı́nez-Lillo, on the structural and magnetic properties of Co(II) Field-induced slow relaxation of magnetisation in two one- hexacoordinate single-molecule magnets, Dalton Trans., dimensional homometallic dysprosium(III) complexes 2018, 47, 1498–1512, DOI: 10.1039/C7DT04256F. based on alpha- and beta-amino acids, Dalton Trans., 2020, 65 H.-W. Wei, Q.-F. Yang, X.-Y. Lai, X.-Z. Wang, T.-L. Yang, 49, 9155–9163, DOI: 10.1039/d0dt01126f. Q. Hou and X.-Y. Liu, Field-induced slow relaxation of 75 A. Cornia, A.-L. Barra, V. Bulicanu, R. Clérac, M. Cortijo, magnetization in a distorted octahedral mononuclear E. A. Hillard, R. Galavotti, A. Lunghi, A. Nicolini, high-spin Co(II) complex, CrystEngComm, 2018, 20, 962– M. Rouzières, L. Sorace and F. Totti, The Origin of 968, DOI: 10.1039/C7CE01981E. Magnetic Anisotropy and Single-Molecule Magnet Behavior 66 S. Gomez-Coca, E. Cremades, N. Aliaga-Alcalde and E. Ruiz, in Chromium(II)-Based Extended Metal Atom Chains, Inorg. Mononuclear Single-Molecule Magnets: Tailoring the Chem., 2020, 59, 1763–1777, DOI: 10.1021/ Magnetic Anisotropy of First-Row Transition-Metal acs.inorgchem.9b02994. Complexes, J. Am. Chem. Soc., 2013, 135, 7010–7018, DOI: 76 A. K. Jassal, N. Aliaga-Alcalde, M. Corbella, D. Aravena, 10.1021/ja4015138. E. Ruiz and G. Hundal, Neodymium 1D systems: targeting 67 S. Roy, I. Oyarzabal, J. Vallejo, J. Cano, E. Colacio, A. Bauza, new sources for eld-induced slow magnetization A. Frontera, A. M. Kirillov, M. G. B. Drew and S. Das, Two relaxation, Dalton Trans., 2015, 44, 15774–15778, DOI: Polymorphic Forms of a Six-Coordinate Mononuclear 10.1039/C5DT02533H. Cobalt(II) Complex with Easy-Plane Anisotropy: Structural 77 P. Mondal, B. Dey, S. Roy, S. P. Bera, R. Nasani, A. Santra and Features, Theoretical Calculations, and Field-Induced Slow S. Konar, Field-Induced Slow Magnetic Relaxation and Relaxation of the Magnetization, Inorg. Chem., 2016, 55, Anion/Solvent Dependent Proton Conduction in Cobalt(II) 8502–8513, DOI: 10.1021/acs.inorgchem.6b01087. Coordination Polymers, Cryst. Growth Des., 2018, 18, 6211– 68 M. A. Palacios, J. Nehrkorn, E. A. Suturina, E. Ruiz, S. Gómez- 6220, DOI: 10.1021/acs.cgd.8b01080. Coca, K. Holldack, A. Schnegg, J. Krzystek, J. M. Moreno and 78 R. Dı́az-Torres, M. Menelaou, A. González-Campo, S. Teat, E. Colacio, Analysis of Magnetic Anisotropy and the Role of E. Sañudo, M. Soler and N. Aliaga-Alcalde, Comparative Magnetic Dilution in Triggering Single-Molecule Magnet Magnetic Studies in the Solid State and Solution of Two (SMM) Behavior in a Family of Co II Y III Dinuclear Isostructural 1D Coordination Polymers Containing CoII/ Complexes with Easy-Plane Anisotropy, Chem.–Eur. J., 2017, NiII-Curcuminoid Moieties, Magnetochemistry, 2016, 2, 29, 23, 11649–11661, DOI: 10.1002/chem.201702099. DOI: 10.3390/magnetochemistry2030029. 69 F. Lloret, M. Julve, J. Cano, R. Ruiz-Garcı́a and E. Pardo, 79 Y. Wu, D. Tian, J. Ferrando-Soria, J. Cano, L. Yin, Z. Ouyang, Magnetic properties of six-coordinated high-spin cobalt(II) Z. Wang, S. Luo, X. Liu and E. Pardo, Modulation of the complexes: Theoretical background and its application, magnetic anisotropy of octahedral cobalt single-ion Inorg. Chim. Acta, 2008, 361, 3432–3445, DOI: 10.1016/ magnets by ne-tuning the axial coordination j.ica.2008.03.114. microenvironment, Inorg. Chem. Front., 2019, 6, 848–856, 70 A. Masegosa, M. A. Palacios, E. Ruiz, S. Gómez-Coca, DOI: 10.1039/C8QI01373J. J. Krzystek, J. M. Moreno and E. Colacio, Dinuclear Co II Y 80 J. Palion-Gazda, K. Choroba, B. Machura, A. Świtlicka, III vs. tetranuclear CoII2YIII2 complexes: the effect of R. Kruszynski, J. Cano, F. Lloret and M. Julve, Inuence of increasing molecular size on magnetic anisotropy and the pyrazine substituent on the structure and magnetic relaxation dynamics, Dalton Trans., 2019, 48, 14873–14884, properties of dicyanamide-bridged cobalt(II) complexes, DOI: 10.1039/C9DT02969A. Dalton Trans., 2019, 48, 17266–17280, DOI: 10.1039/ 71 J. Vallejo, I. Castro, R. Ruiz-Garcı́a, J. Cano, M. Julve, C9DT02976A. F. Lloret, G. De Munno, W. Wernsdorfer and E. Pardo, 81 A. Świtlicka, J. Palion-Gazda, B. Machura, J. Cano, F. Lloret Field-Induced Slow Magnetic Relaxation in a Six- and M. Julve, Field-induced slow magnetic relaxation in Coordinate Mononuclear Cobalt(II) Complex with a Positive pseudooctahedral cobalt(II) complexes with positive axial Anisotropy, J. Am. Chem. Soc., 2012, 134, 15704–15707, and large rhombic anisotropy, Dalton Trans., 2019, 48, DOI: 10.1021/ja3075314. 1404–1417, DOI: 10.1039/c8dt03965h. 72 N. F. Chilton, R. P. Anderson, L. D. Turner, A. Soncini and 82 J. Vallejo, M. Viciano-Chumillas, F. Lloret, M. Julve, I. Castro, K. S. Murray, PHI: A powerful new program for the analysis J. Krzystek, M. Ozerov, D. Armentano, G. De Munno and of anisotropic monomeric and exchange-coupled J. Cano, Coligand Effects on the Field-Induced Double This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 45090–45104 | 45103 RSC Advances Paper Slow Magnetic Relaxation in Six-Coordinate Cobalt(II) Single- 84 P. L. Scott and C. D. Jeffries, Spin-Lattice Relaxation in Some Ion Magnets (SIMs) with Positive Magnetic Anisotropy, Inorg. Rare-Earth Salts at Helium Temperatures; Observation of the Chem., 2019, 58, 15726–15740, DOI: 10.1021/ Phonon Bottleneck, Phys. Rev., 1962, 127, 32–51, DOI: acs.inorgchem.9b01719. 10.1103/PhysRev.127.32. 83 D. Aravena and E. Ruiz, Spin dynamics in single-molecule 85 S. Gómez-Coca, D. Aravena, R. Morales and E. Ruiz, Large magnets and molecular qubits, Dalton Trans., 2020, 49, magnetic anisotropy in mononuclear metal complexes, 9916–9928, DOI: 10.1039/D0DT01414A. Coord. Chem. Rev., 2015, 289–290, 379–392, DOI: 10.1016/ j.ccr.2015.01.021. 45104 | RSC Adv., 2020, 10, 45090–45104 This journal is © The Royal Society of Chemistry 2020