Magnetic and Luminescent Properties of Isostructural 2D Coordination Polymers Based on 2-Pyrimidinecarboxylate and Lanthanide Ions

Abstract

A couple of isostructural coordination polymers with the general formula [Ln₄(pymca)₄(AcO)₈]_n have been obtained from reactions between pyrimidine-2-carboxylate (pymca) ligand and rare-earth ions (Ln = Dy (1), Nd (2)). These two-dimensional compounds have been characterized and the crystal structures have been solved by single-crystal X-ray diffraction technique, resulting in layers along the bc plane based on pymca and acetate anions that act as bridging ligands between metal atoms. Given that pymca and acetate anions possess carboxylate and hetero-nitrogen groups, it is possible to build a coordination polymer whose metal centers have a nine coordination. Furthermore, static and dynamic magnetic measurements of compound 1 reveal the lack of single molecule-magnet (SMM) behavior in this system due to the following two effects: (i) the ligand field does not stabilize magnetic ground states well separated from excited states, and (ii) anisotropy axes are not collinear, according to results with Magellan software. On another level, luminescent properties of compounds 1 and 2 are attributed to singlet π-π* transitions centered on pymca ligand as corroborated by time-dependent density functional theory (TD-DFT) calculations.

1. Introduction

Coordination polymers (CPs) are defined as infinite systems built up from metal ions and organic ligands linked via coordination bonds and other weak chemical bonds such as hydrogen bonds, π-π stacking, or van der Waals interactions [1]. These materials have gained considerable attention in recent decades, in particular, lanthanide-based coordination polymers, due to their varied structures [2,3], high coordination numbers, gas storage properties [4], biochemical and medical applications [5], as well as interesting luminescent [6] and magnetic properties [7,8,9]. The key factor for the synthesis of CPs is to choose appropriate metal centers and bridging ligands to obtain the desired topologies and properties, although other factors can also have an influence such as metal-to-ligand ratio, temperature, pH value, and solvents [10,11]. Among ligands, with the goal to achieve coordination frameworks based on f-block metal centers with local connectivity numbers larger than six, ligands containing both carboxylate groups and hetero-nitrogen chemical functions have been extensively used in the construction of these novel systems because both groups can coordinate to metal centers in different ways [12,13,14].

Ligands such as pyridylcarboxylate, imidazolylcarboxylate, pyrimidylcarboxylate, or indolylcarboxylate derivatives are used extensively in coordination chemistry based on lanthanide ions [15,16,17]. Specifically, we chose the ligand pyrimidine-2-carboxylate (pymca), which contains both oxygen and nitrogen coordination atoms (Scheme 1), because, to date, only a few structures based on rare-earth ions and this ligand have been reported, all of them by Wang et al. [8]. Inspired by all the above-mentioned ideas, we decided to react pymca and lanthanide ions with the objective of building coordination polymers with interesting luminescent and magnetic properties, due to the fact that, fifteen years ago, we were the pioneers to design and work with this novel ligand, which shows great ability to form two-dimensional and three-dimensional coordination polymer compounds. In this way, we were able to obtain a great variety of multidimensional coordination polymers with interesting magnetic and luminescent properties effectively explained by density functional theory (DFT)-type calculations [18,19,20,21].

In this paper, we reported the synthesis and characterization of two isostructural two-dimensional coordination polymers based on lanthanide ions, mixed pyrimidine-2-carboxylate, and acetate anions as ligands with the general formula [Ln₄(pymca)₄(AcO)₈]_n (Ln = Dy (1), Nd (2)). Their magnetic and luminescent properties were also investigated and corroborated by theoretical calculations.

2. Materials and Methods

2.1. Synthesis

All starting reagents were purchased from commercial sources (Sigma-Aldrich) and were used as received and without additional purification.

2.1.1. Synthesis of [Dy₄(pymca)₄(AcO)₈]_n (1)

Compound 1 was obtained following a solvothermal route through the following procedure: First, 0.020 g (0.19 mmoL) of 2-pyrimidinecarbonitrile, in situ precursor of pymca ligand, was dissolved in 1 mL of N,N-dimethylformamide (DMF). In a separate vial, 0.0215 g (0.063 mmoL) of Dy(AcO)₃ was dissolved in 0.5 mL of distilled water. Metal solution was added dropwise to the ligand solution. The resulting mixture was stirred and heated at 100 °C for 24 h. After this time, white crystals of 1 for single crystal X-ray diffraction were obtained during the heating process under autogenous pressure. Anal. Calcd. for C₁₈H₁₈Dy₂N₄O₁₂: C, 26.76; H, 2.25; N, 6.94%. Found: C, 26.54; H, 2.13; N, 7.02%. FT-IR: 1620 cm⁻¹ (m), 1568 cm⁻¹ (s), 1413 cm⁻¹ (s), 1373 cm⁻¹ (s).

2.1.2. Synthesis of [Nd₄(pymca)₄(AcO)₈]_n (2)

Compound 2 was synthesized following a similar procedure as follows: First, 0.015 g (0.142 mmoL) of 2-pyrimidinecarbonitrile ligand was dissolved in 0.5 mL of DMF. In a separate vial, 0.0155 g (0.047 mmoL) of Nd(AcO)₃·H₂O was also dissolved in 0.5 mL DMF. Metal solution was added to the ligand solution. The reaction solution was stirred and heated at 100 °C for 24 h. X-ray quality white crystals of 2 were obtained. Anal. Calcd. for C₁₈H₁₈Nd₂N₄O₁₂: C, 28.03; H, 2.35; N, 7.27%. Found: C, 27.96; H, 2.28; N, 7.33%. FT-IR: 1622 cm⁻¹ (m), 1560 cm⁻¹ (s), 1419 cm⁻¹ (s), 1373 cm⁻¹ (s).

2.2. Characterization Methods

2.2.1. Physico-Chemical Characterization

Elemental analyses (C, H, and N) were carried out at the “Centro de Instrumentación Científica” (University of Granada) on a THERMO SCIENTIFIC analyzer model Flash 2000. The FTIR spectra were recorded with a BRUKER TENSOR 27 FT-IR and OPUS data collection program.

2.2.2. Single-Crystal X-Ray Diffraction

Measured crystals were prepared under inert conditions immersed in perfluoropolyether as protecting oil for manipulation. Suitable crystals were mounted on MiTeGen Micromounts™, and these samples were used for data collection. Data for 1 and 2 were collected with a Bruker D8 Venture diffractometer with a Photon detector equipped with graphite monochromated MoKα radiation (λ = 0.71073 Å). The data were processed with APEX3 suite [22]. The data contain monoclinic pseudo-translational symmetry. This results in a pseudo C-centered cell (Bravais centering), but the data clearly show that the reflections with h + k = odd are weak but definitely not extinct, therefore, P-centered cell was chosen. Looking at the systematic absences, there is clearly a 2₁ screw axis along the crystallographic b-axis. Additionally, there are indications for the presence of a-, n-, and c-glide planes. There are some violations of reflection conditions for the glide planes. The structures were solved in space group P2₁ and refined as inversion twins. The packing consisted of two symmetry-independent coordination planes parallel to the bc plane. The structures were solved by Intrinsic Phasing using the ShelXT program [23], which revealed the position of all non-hydrogen atoms. These atoms were refined on F² by a full-matrix least-squares procedure using anisotropic displacement parameters [24]. All hydrogen atoms were located in difference Fourier maps and included as fixed contributions riding on attached atoms with isotropic thermal displacement parameters 1.2 or 1.5 times those of the respective atom. The OLEX2 software was used as a graphical interface [25]. As a consequence of the pseudo-symmetries, the least-squares refinements of the structures were not stable, and the use of restraints was required. Some ISOR and RIGU commands had to be used to obtain reasonable displacement parameters for selected non-hydrogen atoms. The Addsym routine implemented in the program PLATON [26] found a P2₁/c symmetry. The two symmetry-independent coordination planes were maintained in the higher symmetry, but refinement in P2₁/c was not successful. Clearly, the P2₁ solution describes the structure better than in the P2₁/c space group, being the inversion center a pseudo-symmetry operator. The crystallographic data for the reported structures were deposited with the Cambridge Crystallographic Data Center as supplementary publication no. CCDC 2006485 and 2006486. Additional crystal data are shown in

Table 1. Crystallographic data and structure refinement details of compounds 1 and 2.

Compound	1	2
Formula	C18H18N4O12Dy2	C18H18N4O12Nd2
Mr (g moL−1)	807.36	770.84
Crystal system	Monoclinic	Monoclinic
Space group	P21	P21
Temperature (K)	298(2)	100
a (Å)	15.0967(7)	15.1628(7)
b (Å)	9.3786(4)	9.5069(4)
c (Å)	16.4416(8)	16.6857(8)
α (°)	90	90
β (°)	103.8920(10)	104.791(2)
λ (°)	90	90
V (Å3)	2259.81(18)	2325.56(18)
Z	4	4
ρ (g cm−3)	2.373	2.202
µ (mm−1)	6.635	4.487
Unique reflections	10,291	7722
Rint	0.057	0.088
GoF a	1.119	1.023
R1 b/wR2 c [I > 2σ(I)]	0.0590/0.1481	0.0750/0.1740
R1 b/wR2 c [all data]	0.0823/0.1624	0.1279/02049

^aR₁ = S||F_o| − |F_c||/S|F_o|; ^b values in parentheses for reflections with I > 2s(I); ^c wR₂ = {S[w(F_o² − F_c²)²]/S[w(F_o²)²]}^½.

. Copies of the data can be obtained free of charge at http://www.ccdc.cam.ac.uk/products/csd/request.

2.2.3. Magnetic Measurements

Magnetic susceptibility measurements were performed on polycrystalline samples of the complexes with a Quantum Design SQUID MPMS-7T susceptometer at an applied magnetic field of 1000 Oe. The susceptibility data were corrected for the diamagnetism, the temperature-independent paramagnetism, and the magnetization of the sample holder. Ac measurements were performed on a Quantum Design Physical Property Measurement System Model 6000 magnetometer under a 3.5 G ac field and frequencies ranging from 60 to 10,000 Hz.

2.2.4. Luminescent Measurements

Fluorescence emission spectra were recorded at room temperature using polycristalline samples in a Varian Cary Eclipse fluorescence spectrofluorimeter equiped with a xenon discharge lamp (peak power equivalent to 75 kW), Czerny–Turner monochomators, and an R-928 photomultiplier tube. The emission spectra were recorded in the visible region (300–700 nm), the photomultiplier detector voltage was fixed at 550 V, and the excitation and emission slits were set both at 5 nm.

2.2.5. Computational Details

The computational strategy adopted to calculate the magnetic coupling constant (J_calc) values corresponds to the broken symmetry strategy and has been described and validated elsewhere [27]. One calculation was performed to determine the high-spin state and another to determine the low-spin broken symmetry state, using Gaussian 16 package [28]. The correctness of the latter state was ensured by means of its spin density distribution. These calculations were performed using density functional theory (DFT) with hybrid B3LYP functional and Gaussian-implemented 6-311G(d) basis set for all non-metallic atoms, whereas the corresponding LANL2DZ pseudopotentials were used for the metal atoms. Spin-density surfaces were plotted using GaussView 5 [29]. The anisotropy axis of the Dy³⁺ ions was calculated with Magellan software [30]. Regarding the calculations of the PL spectrum of pymca ligand, it was carried out by time-dependent density functional theory (TD-DFT) methodology with Gaussian 16 package, using equivalent model chemistries detailed for broken symmetry calculations. The 40 lowest excitation states were calculated by the TD-DFT method. Results were analyzed with GaussSum program package [31] and molecular orbitals (MOs) plotted using GaussView 5.

3. Results and Discussion

Pyrimidine-2-carboxylic acid ligand allowed the formation of two isostructural coordination polymers by varying the metal center. Thus, the solvothermal reaction of the pymca with dysprosium(III) and neodymium(III) acetate hydrate in DMF/H₂O mixture with a 3:1 molar ratio afforded two-dimensional metal-organic frameworks.

3.1. Crystal Structure of [Ln₄(pymca)₄(AcO)₈]_n (Ln = Dy (1), Nd (2))

Compounds 1 and 2 are isostructural with the monoclinic space group P2₁, therefore, just compound [Dy₄(pymca)₄(AcO)₈]_n (1) is described in detail.

The crystal structure of complex 1 consists of a two-dimensional coordination polymer with mixed acetate and pymca bridging ligands. The asymmetric unit is based on two units with two Dy³⁺ cations, two pymca ligands, and four acetate anions each one (Figure 1). Dy1, Dy3, and Dy4 atoms are coordinated with five oxygen atoms from four acetate anions and two nitrogen and two oxygen atoms belonging to two pymca ligands to complete a nine-coordination environment DyN₂O₇ that resembles a muffin polyhedron (see Figure S1 in Supplementary Materials). Instead, the coordination number drops to eight for the Dy2 atom despite its equivalent environment, hence showing a less distorted square antiprism, owing to the fact that one of the Dy–O_acetate bond distance is too large (ca. 2.98 Å) as to be considered a coordination bond (see Section S3 in Supplementary Materials).

The structure consists of two similar layers parallel to the bc plane. They are not equal, since each one contains two chemically equivalent but crystallographically different Dy atoms (one layer is composed of Dy1 and Dy2 atoms, meanwhile the other one is made up of Dy3 and Dy4 atoms) (Figure 2). Selected bond lengths and angles of coordination spheres are not the same in the two layers, so they are all included in Tables S1–S4 for compounds 1 and 2.

Within the structure, both pymca and acetate anions act as bridging ligands by joining metal atoms, resulting in two different distance ranges of 6.472(2)–6.480(2) and 3.952(2)–4.087(2) Å, respectively. The shorter range of distances corresponds to inequivalent metal bonds, e.g., Dy3–Dy4 bond, and the largest one belongs to bonds between equal metal atoms, e.g., Dy3–Dy3 bond. The anionic pymca ligands coordinate in a bisbidentate form to metal atoms. In contrast, the acetate anions show two different coordination modes. Those two anions that lie out perpendicular to the bc plane show a µ-κO:κO′, while the other two acetate anions arrange parallel to the bc plane and create a µ-κ²O,O′: κO coordination between two Dy³⁺ atoms [32].

This arrangement of the ligands produces mesh shaped two-dimensional layers in the bc plane whose holes are occupied by acetate ligands (Figure 3a). A summary of the structure refinements of compounds 1 and 2 is given in Table 1.

3.2. Magnetic Properties of Compound 1

3.2.1. Static Magnetic Measurements

The static magnetic properties of compound 1 were investigated by direct current (dc) magnetic susceptibility measurements in the 2–300 K temperature range under an applied field of 1000 Oe (Figure 4).

The χ_MT vs. T plot (Figure 4, black dots) reveals a value of 28.37 cm³ K moL⁻¹ at room temperature, which is in good agreement with the expected value of 28.34 cm³ K moL⁻¹ for two non-interacting Dy³⁺ ions in the free-ion approximation (4f⁹, J = 15/2, S = 5/2, L = 5, g = 4/3 ⁶H_15/2). Upon cooling down from room temperature, the χ_MT value remains almost unchanged until nearly 50 K, where it starts to fall abruptly reaching a minimum value of 16.04 cm³ K moL⁻¹ at 2.0 K. This drop could be associated with diverse effects. On the one hand, the thermal depopulation of the Stark sublevels resulting from crystal-field effects provoke this decrease in χ_MT but, on the other hand, there could also be weak antiferromagnetic interactions between the two Dy³⁺ ions provided by carboxylate bridges of the pymca ligands. The strong spin-orbit coupling present in the Dy³⁺ ions complicates the treatment of the data in order to obtain the coupling exchange constant J. However, with the aim of shedding some light regarding the nature of the metal–metal interactions, we opted to follow the next approach. First, we studied the magnetic data in the same temperature range by using the Curie–Weiss law, namely 1/χ = (T − θ)/C (Figure 4, blue dots and line). The fitting afforded C and θ values of 28.69 cm³ K moL⁻¹ and −2.9 K, respectively. The low negative value of θ indicates the presence of weak intramolecular antiferromagnetic interactions suggesting that the lowering of the χ_MT value is not solely influenced by the depopulation of the Stark sublevels. In addition, with the purpose of supporting these results and better estimating the nature of the magnetic exchange interactions, we conducted DFT calculations with the broken symmetry methodology [33]. In order to do that, two different dimeric fragments were considered separately with the aim of evaluating the two possible superexchange pathways found in the compound. The first selected dinuclear entity evaluates the interactions within the paddle-wheel shaped fragment involving four bridging acetate groups (inside it a short Ln⋯Ln distance of ca. 4.0 Å is established), whereas the second entity involves the bisdidentate pymca bridging ligand (this establishes a longer Ln⋯Ln distance about 6.5 Å). As mentioned, the strong spin-orbit coupling of the Dy³⁺ ions complicated the calculation, thus, the anisotropic ions were replaced by isotropic Gd³⁺ ones in order to have a good estimation of the J parameter. In both superexchange pathways, a negligible coupling constant was calculated (Figure 5). Therefore, we conclude that the decay in the χ_MT value arises mainly from the depopulation of the Stark sublevels and besides intramolecular interactions insignificantly contribute to it.

The M(H) plot was also recorded at 2 K, which is shown in the inset of Figure 4. The magnetization value rapidly increases with field strength up to 1 T, above it a gradual increase is observed reaching a maximum value of 11.54 N_µB, at 5 T. This gradual increase and the low “saturation” value, which is far lower than the expected value of 19.9 N_µB expected for two Dy³⁺ ions suggests significant magnetic anisotropy.

3.2.2. Dynamic Magnetic Properties

In view of the fact that highly anisotropic lanthanides such as Dy³⁺ are potential candidates to present slow relaxation of the magnetization [34], dynamic magnetic properties of 1 were studied by alternating current (ac) magnetic measurements. Firstly, measurements were carried out applying an oscillatory field of 3.5 Oe and under zero dc field. However, the out-of-phase component of the susceptibility (χ_M″) did not show any signal (Figure S1). The lack of single molecule-magnet (SMM) behavior in this case could be attributed to the fast quantum tunneling of the magnetization (QTM). It is well known [35] that the use of an external magnetic field could be a good technique in order to overcome this undesired phenomenon. Nonetheless, in this case the Dy³⁺ based compound did not show any χ_M″ signal even when applying an external magnetic field of 1 kOe. This fact indicates that either there is still a strong QTM process that the field is not able to quench, or that the system is actually a common paramagnet.

The dynamic magnetic properties of SMMs are very sensitive to the crystal field. In general, Ln³⁺ ions are divided into the following two main groups: ions with an oblate (Dy³⁺ and Tb³⁺, for example) or prolate (Er³⁺ and Yb³⁺, for example) electron density. The qualitative design principles proposed by Rinehart and Long [36] indicate that in order to enhance the single ion anisotropy of Dy³⁺, and thus the SMM behavior, the negatively charged ligands are preferred to be axially coordinated. This has been proved for a wide number of Dy³⁺ systems, such as for dysprosium metallocenes or systems with pentagonal-bipyramid geometry [34,37], where the strong electron donating groups are axially coordinated to the metal ions. We believe that the lack of SMM behavior in this compound arises from the inappropriate ligand field established around the metal ions, as well as the weak exchange coupling between them. On the one hand, as mentioned in the crystal structure description, the coordination environments around the four Dy³⁺ centers are quite distorted. Noteworthy, Dy2 displays the lowest SHAPE value (see SI, Tables S5 and S6) for a square antiprism polyhedron, which has been proved to be a suitable geometry in order to stabilize the highly magnetic m_J = ±15/2 states as ground state [21]. However, in this compound, Dy2 displays nearly nine coordination and, moreover, there are not remarkably short and negatively charged Dy–O bonds that provide axiality to the system. On the other hand, in a previously Dy³⁺ based MOF reported by us [38], it was shown that even in weakly exchange coupled systems, when anisotropy axes are parallel, a slower relaxation of the magnetization is enhanced. In our case, considering the dinuclear entity where both of the lanthanide ions possess the same coordination environment, the lack of an inversion center between both ions provokes a lack of parallelism between the axes (calculated by Magellan software, Figure 6). Hence, both of the mentioned effects probably account for the lack of SMM behavior in this system.

3.3. Luminescence Properties

As it is well-known, lanthanide-based emissions are characterized by very narrow and pure emission bands derived from the shielding of valence f shell, although, at the same time, their emission brightness tends to be low because of the low coefficients of their absorption bands [39]. Therefore, in order to obtain a luminescent compound, a ligand-to-metal charge transfer process (between ligands triplet state and excited inner states of the lanthanide(III)) must take place, which is denoted as antennae effect [40]. In this work, the luminescence properties of compounds 1 and 2 were investigated in solid state, at room temperature. Figure 7 shows the emission spectra of both complexes at λ_ex = 240 nm. Compounds 1 and 2 show similar emission spectra compared to that of free pymca ligand, with only one intense peak at 400 nm, which allows one unequivocally attributing the emission as ligand-centered in both compounds. The absence of any lanthanide(III)-based characteristic emission in the visible spectra, mainly for the Dy³⁺ atom in compound 1 means that the pymca ligand is not able to transfer the energy gained during the excitation process to the lanthanide atoms (i.e., there is no efficient antennae effect happening in these compounds), pymca ligand does not sensitize the lanthanide(III) ions in these compounds in such a way that only the ligand-centered emission is observed. Time-dependent DFT (TD-DFT) calculations conducted on a suitable model of the ligand reveals that the main transition occurs from the LUMO + 1 to the HOMO − 2, which can be regarded as a π ← π* transition in view of the shape of the molecular orbitals involved.

4. Conclusions

Two isostructural coordination polymers with the general formula [Ln₄(pymca)₄(AcO)₈]_n were obtained from reactions between pyrimidine-2-carboxylate (pymca) ligand and rare-earth ions (Ln = Dy (1), Nd (2)). These two-dimensional compounds were characterized resulting in layers along the bc plane based on pymca and acetate anions that act as bridging ligands between metal atoms. Regarding the magnetic properties, although highly anisotropic lanthanide ions such as Dy³⁺ were employed, we concluded that the ligand field provided by the carboxylate groups and nitrogen donor atoms was not suitable to stabilize high magnetic ground states well separated from excited states. Moreover, we find out that the anisotropy axes calculated by Magellan software are not collinear, a fact that even in weakly exchange coupled systems negatively affects the SMM behavior. Under UV excitation, both compounds exhibit a pymca ligand-centered emission based on a main π ← π* transition, confirming the null capacity of pymca to sensitize the selected lanthanide(III) ions in the present structures. New studies in these directions are currently being carried out in our laboratory focusing, for instance, on the possible use of 3s-4f materials in electronic devices.