A Dy₂ Complex Constructed by TCNQ^·− Radical Anions with Slow Magnetic Relaxation Behavior ^†

Abstract

A centrosymmetric dinuclear complex, [Dy₂(H₂dapp)₂(μ-OH)₂(H₂O)₂]·4TCNQ·2CH₃OH, was synthesized using the TCNQ^·− radical anion (TCNQ = 7,7,8,8-tetracyanoquino-dimethane) and pentadentate nitrogen-containing Schiff base ligand (H₂dapp = 2,6-diacetylpyridine)-bis(2-pyridylhydrazone). In the Dy₂ dimer, the two Dy^III ions adopt eight-coordinated geometries intermediate between D_4d and D_2d symmetries, linked by two OH⁻ groups, with ferromagnetic Dy-Dy interactions. The TCNQ^·− radical anions are uncoordinated, and they pack tightly into antiparamagnetic dimers to balance the system charge. Under zero field, weak magnetic relaxation was observed, with an approximate Δ_eff = 2.82 K and τ₀ = 6.88 × 10⁻⁶ s. This might be attributed to the short intermolecular Dy···Dy distance of 7.97 Å, which could enhance intermolecular dipolar interactions and quantum tunneling of magnetization (QTM).

1. Introduction

Single-molecule magnets (SMMs), a class of nanoscale magnets that exhibit magnetic bistability at the molecular level [1], offer unprecedented opportunities for miniaturizing data storage devices and advancing quantum technologies due to their ability to retain magnetization at the single-molecule level [2,3,4]. Among the high-performance SMMs, rare-earth ions (especially Dy^III) have emerged as leading candidates due to their strong spin-orbit coupling and large magnetic anisotropy, which are critical for achieving high energy barriers (U_eff) and blocking temperatures (T_B) [5,6,7]. In particular, significant progress in dysprosium metallocene SMMs has been demonstrated by [Dy(Cp^ttt)₂][B(C₆F₅)₄] achieving a blocking temperature of 60 K [8], followed by its improved heteroligand derivative [(Cp^iPr5)Dy(Cp*)][B(C₆F₅)₄] reaching 80 K [9]. More recently, a dysprosium bis(amide)–alkene complex of [Dy{N(SiⁱPr₃)[Si(ⁱPr)₂C(CH₃)=CHCH₃]}{N(SiⁱPr₃)(SiⁱPr₂Et)}][Al{OC(CF₃)₃}₄] has been reported to exhibit an effective energy barrier to magnetic reversal of 1843(11) cm⁻¹ and show slow closing of soft magnetic hysteresis loops up to 100 K, owing to the charge-dense amide ligands and pendant alkenes that structurally enforce a large N–Dy–N angle leading to weak equatorial interactions [10]. These discoveries offer new avenues for achieving magnetic memory functionality at higher temperatures.

While for most Dy^III-based SMMs, a limitation of their U_eff and T_B lies in the existence of QTM, which can induce rapid magnetization relaxation and loss of bistability under zero external field [11]. To quench QTM, symmetry control—by placing Dy^III ions in crystal ligands with high symmetry (e.g., D_4d, D_5h, D_6d, C_∞v)—can effectively suppress transverse crystal field components which facilitate QTM [12,13,14]. Alternatively, strong intramolecular magnetic interactions provided by 3d/4f metals or 2p organic radicals can also quench the QTM of Dy^III systems to some extent [15,16,17]. Radicals prove to be particularly effective due to their diffuse orbitals penetrating 4f electron clouds, creating a “giant spin” effect [18,19]. This approach has spurred the development of diverse 2p–4f heterospin systems, incorporating radicals such as nitronyl nitroxide [20,21], TCNQ^·− [22,23] and N₂³⁻ [24], etc. Among these, TCNQ^·− radical anions and its derivatives stand out due to their multiple coordination sites, strong coordination ability, and extended π-conjugation [22,25,26]. They can not only effectively modulate the ligand field environment around Ln^III ions but also facilitate strong magnetic coupling by coordinating to Ln^III ions without other bridges [27]. This unique electronic characteristic endows TCNQ^·−-based systems with intrinsic semiconducting properties, effectively bridging molecular magnetism and electronic conductivity—a particularly valuable feature in the design of SMMs that offers significant multifunctional potential [28]. The Dunabr group has synthesized a hetero-tri-spin complex [(valpn)CuTb(TCNQF)₂(H₂O)₄][TCNQF]·CH₃OH·6H₂O, where the coordinated TCNQF⁻ radical anions link the SMMs to single-chain magnetism via supramolecular π-stacking [23]. Subsequently, their further work of Dy(TPMA)(μ-TCNQ)(μ-OH)₂·CH₃CN with partially charged TCNQ^·− radical anions was prepared, which exhibited bifunctionality of SMMs performance (5.0–8.0 K) and semiconducting performance (180–350 K), advancing the development of radical-based multifunctional materials [29].

Auxiliary ligands critically influence SMM behavior by controlling metal ion coordination symmetry and magnetic anisotropy. The recently reported zero-dimensional dinuclear complex Dy₂(B(OMe)₄)(Hdapp)₂(tBu-DDTP)₂·3MeOH [30], with Dy^III centers in a distorted triangular dodecahedral (D_2d) symmetry, exhibits slow magnetic relaxation. Subsequently, the dinuclear complexes [Dy(HDAPP)(^MeDDTP)₂]·EtOH and Dy(H₂DAPP)(^MeDDTP)₂·2EtOH·H₂O [31] adopt distorted pentagonal bipyramidal (D_5h-like) geometries, where H₂dapp/Hdapp⁻ acts as an N₅ equatorial pocket, exhibiting SMMs behavior. Guided by these findings and in pursuit of more TCNQ^·−-Dy^III SMMs with interesting properties, in this study, the pentadentate nitrogen-rich Schiff base ligand 2,6-diacetylpyridine-bis(2-pyridylhydrazone) (H₂dapp) was utilized as the ancillary ligand, which might provide a weak equatorial coordination environment and leave the axial coordination sites available for TCNQ^·− radical anions (TCNQ = 7,7,8,8-tetracyanoquinodimethane). The structure of the ligands are illustrated in Scheme 1. We aimed to establish a high-symmetry D_5h configuration and develop bifunctional materials with both SMM properties and conductivity. The dinuclear Dy^III complex [Dy₂(H₂dapp)₂(μ-OH)₂(H₂O)₂]·4TCNQ·2CH₃OH (1) was successfully obtained. Although the TCNQ^·− radical anions do not coordinate with Dy^III ions, they form antiferromagnetic dimers via strong π–π interactions. These dimers are further connected into an extended chain structure through additional intermolecular π–π interactions, and the complex still exhibits slow magnetic relaxation behavior under a zero dc field.

2. Results

2.1. Crystal Structure Description

Single-crystal X-ray diffraction analysis reveals that complex 1 is a centrosymmetric dinuclear Dy^III complex, which crystallizes in the triclinic crystal system with space group P-1. The molecular structure consists of a [Dy₂(H₂dapp)₂(μ-OH)₂]⁴⁺ group, four uncoordinated TCNQ^·− radical anions, and two free methanol molecules (Figure 1). In the [Dy₂(H₂dapp)₂(μ-OH)₂]⁴⁺ unit, two Dy^III ions are bridged by oxygen atoms (O1 and O1a) from hydroxido groups. Each Dy^III ion is eight-coordinated by five nitrogen atoms from the H₂dapp ligand, two bridging oxygen atoms from hydroxido groups and one oxygen atom from a water molecule. The CShMs analysis (listed in Table S1) using SHAPE 2.1 software [32] reveals that the coordination geometry of Dy^III ions lies between square antiprismatic (D_4d) and triangular-dodecahedral (D_2d) symmetries. The Dy–N bond lengths range from 2.497(3) Å to 2.561(3) Å. The Dy–O bonds vary from 2.211(3) Å to 2.359(3) Å. The Dy···Dy distance bridged by μ-OH in Dy₂ dimer is 3.6603(3) Å, and the Dy–O–Dy bond angle is 110.34(11)°. The O–Dy–O and O–Dy–N angles are within 69.66(11)–154.73(10)° and 73.62(10)–142.17(10)°, respectively.

Based on the electrical neutrality of complex 1, the approximate charge on each TCNQ^·− radical anions were calculated using Kistenmacher’s empirical [22,33] (Equation (1)), where b, c, and d represent the relative average bond lengths of TCNQ^·− radical anions. Different TCNQ^·− radical anions were marked as A and B. The results indicate that all TCNQ^·− radical anions in the system carry a −1.0 charge (Table S3). The TCNQ^·− radical anions in complex 1 are all uncoordinated, serving only to balance the charge of the system. They form strongly packed antiparamagnetic dimers [29,34] (but do not form a continuous chain structure) through strong π–π interactions, with an interplanar centroid distance of 3.2241 Å between benzene rings (As in Figure 2c). Although these TCNQ^·− dimers themselves do not form a continuous infinite one-dimensional stack, intermolecular π–π interactions exist between the TCNQ^·− radical anions and the pyridine ring of adjacent [Dy₂(H₂dapp)₂(μ-OH)₂]⁴⁺ units (as in Figure 2b) with distances of 3.4049 Å (C26···C10) and 3.4430 Å (C24···C9). In addition, a strong π–π interaction is present between the cyano group of TCNQ (C45 and N15) and the pyridine ring of the H₂dapp ligand involving N3, with a distance of approximately 3.1567 Å. These interactions collectively contribute to the formation of one-dimensional supramolecular chains. Hydrogen bonds exist between TCNQ^·− radical anions and coordinated water molecules (Table S2). Along the a-axis, the [Dy₂(H₂dapp)₂(μ-OH)₂]⁴⁺ units are effectively isolated through staggered lateral stacking of TCNQ^·− radical anions with an intermolecular Dy···Dy separation of 13.8636(3) Å (Figure S1). However, along the b-axis (Figure 2a), the TCNQ^·− dimers fail to isolate intermolecular Dy^III ions effectively, with a relatively shorter Dy···Dy distance of 7.9667(4) Å.

$$\delta =-41.67[c/(b+d\left)\right]+19.83$$

(1)

2.2. Magnetic Properties

Variable-temperature magnetic susceptibility measurements were performed on complex 1 at a dc (direct current) field of 1000 Oe and in the temperature range of 300–2 K. The phase purity of the synthesized crystals was confirmed by the consistency between the experimental PXRD data of complex 1 and the simulated pattern as in Figure S2. At 300 K, the χ_MT value is 27.87 cm³·K·mol⁻¹ (Figure 3), which is close to the calculated value of 28.34 cm³·K·mol⁻¹ for two independent Dy^III ions (⁶H_15/2, g = 4/3), with no contribution from the spin of the TCNQ^·− radical anions. As evident from the structure, the strong π–π interactions between the TCNQ^·− radical anions lead to the formation of dimers with orbital overlap, causing strong antiferromagnetic coupling between spins and resulting in diamagnetic dimers that do not contribute to magnetic properties of the system [34]. Upon cooling, the χ_MT value gradually increases to a maximum value of 29.51 cm³·K·mol⁻¹ at 70 K, suggesting an intramolecular ferromagnetic interaction between Dy^III ions [35,36,37,38]. The magnetic susceptibility data of 1/χ_M versus T above 50 K (inset of Figure 3) were analyzed using the Curie–Weiss law [39,40] (Equation (2)). The fitting results show that the Curie constant C and the Weiss constant θ are 27.88 cm³·K·mol⁻¹ and 5.01 K, respectively, and θ > 0 clearly proves the presence of ferromagnetic interactions between Dy^III ions [36]. However, below 70 K, the χ_MT value decreases rapidly and reaches a minimum value of 20.16 cm³·K·mol⁻¹ at 2 K, which might be due to the depopulation of Dy^III ions Stark sublevels and the intermolecular antiferromagnetic interactions between Dy^III ions [41,42].

$$1/{\chi }_M={\displaystyle \frac{T-\theta }{C}}$$

(2)

The isothermal field-dependent magnetization (M vs. H plots in Figure 4) was measured at different temperatures from 0 to 70 kOe. It can be observed that the M values show a sharp rise as the magnetic field increases, reaching a maximum value of 13.80 μ_B under 70 kOe at 1.9 K. This maximum value is significantly lower than the expected value of 20 μ_B for two independent Dy^III ions (g_J × J = 4/3 × 15/2 = 10 μ_B), and the M vs. H/T curves do not overlap at various temperatures, indicating the presence of magnetic anisotropy and/or low-lying excited states.

To study the dynamic magnetic properties of complex 1, ac (alternating current) magnetic susceptibility measurements were conducted without an applied dc field. As shown in Figure 5, the χ″ signals exhibit clear frequency dependence in ac susceptibility measurements below 10 K, demonstrating slow magnetic relaxation behavior of complex 1. Unfortunately, no distinct peaks are observed, likely due to QTM effects. Then, the energy barrier (Δ_eff) and relaxation time (τ₀) cannot be determined through standard Arrhenius fitting. Assume that there is only one magnetization relaxation process, the ac susceptibility data can be analyzed using Equation (3) to approximate the Δ_eff and τ₀ [43,44]. The analysis yields an effective energy barrier Δ_eff = 2.82 K with a corresponding pre-exponential factor τ₀ = 6.88 × 10⁻⁶ s. The observed τ₀ values fall within the characteristic range for SMMs [45]. To suppress QTM, field-dependent χ″ measurements at 1.9 K and 997 Hz revealed an optimal dc field of 400 Oe for complex 1 (Figure S3a). Although χ″ signals were detected under this field, their minimal intensity variation and absence of distinct peaks (Figure S3b) indicate that quantum tunneling was not effectively quenched despite the applied field for complex 1 [46,47,48].

$$\ln ({\chi }^{″}/{\chi }^{\prime })=\ln (\omega {\tau }_0)+{\mathsf{\Delta }}_{\mathrm{eff}}/k_{B}T$$

(3)

The magnetic anisotropy axes of the Dy^III ions in complex 1 was calculated using the electrostatic model by the Magellan program [49] (Figure 6). The magnetic axis is oriented toward the μ-OH oxygen donor (O1). The shortest Dy1–O1 bond exhibits only a minor deviation of 8.504(139)° from the anisotropy axis. This near-parallel alignment likely enhances the axial electron density. The possible presence of D_2d symmetric geometries, combined with insufficient magnetic isolation due to the relatively shorter intermolecular Dy···Dy distance of 7.9667(4) Å (which enhances dipolar coupling), results in QTM and ultimately leads to poor SMM behavior. The comparison of the SMMs’ behavior with that of other reported dinuclear Dy^III complexes is shown in

Table 1. Some reported Dy₂ complexes exhibiting SMM behavior.

Dy2 Complex	Point Group	Shortest Intermolecular Dy···Dy Distance	Magnetic Coupling	The Slow Relaxation Process
				Field (Oe)	Ueff (K)	τ0 (s)
[Dy2(dmpd)6(bmp)]	D4d	10.487 Å	AF (J = −0.524 cm−1)	0	87.29	5.48 × 10−7
[Dy2(dbm)2(LH3)2]·H2O	D4d	9.423 Å	F (J = 15.00 cm−1)	0	76.18	2.54 × 10−8
[Dy(hfac)3]2(μ-HMq)2	D4d	10.028 Å	F (J = 1.00 cm−1)	0	9.2	1.7 × 10−5
[Dy2(L5)(tfac)4]	D4d	9.214 Å	F (J = 2.72 cm−1)	0	79.0	3.3 × 10−8
[Dy2(hfac)4(L3)2]	D2d	10.172 Å	F (/)	0	6.8	9.12 × 10−6
[Dy2(bfac)4(L4)2]·C7H16	D2d	9.502 Å	AF (/)	0	25.7	1.64 × 10−6
[Dy2(H2dapp)2(μ-OH)2(H2O)2]·4TCNQ·2CH3OH (this work)	D4d	7.967 Å	F (/)	0	2.82	6.88 × 10−6

[50,51,52,53,54].

The Dy-Dy magnetic interactions within the Dy₂ dimer can also be proved by the simplified Equation (4) [55,56], where θ represents the angle between the magnetic axes and the line connecting the two spin centers. Ferromagnetic coupling occurs when θ is below the threshold angle of 54.75°, whereas antiferromagnetic coupling prevails at larger angles. The θ value of 43.59° for 1 proves again the presence of ferromagnetic Dy-Dy coupling in the dimer.

$$E_{dip}=-{\displaystyle \frac{{\mu }_0}{4\pi }}{\displaystyle \frac{{\mu }^2}{r^3}}[3{\cos }^2\theta -1]$$

(4)

3. Materials and Methods

3.1. Materials and Characterizations

The starting materials H₂dapp [57] and LiTCNQ [58] were prepared following reported literature methods. All chemical reagents were purchased from commercial suppliers and utilized directly without additional purification steps, except for distilled water and methanol, which were deoxygenated prior to use. Elemental analyses were conducted on a Vario EL cube (Langenselbold, Germany). IR spectra data was on a Nicolet iS10 ATR-FTIR instrument (Thermo, Waltham, MA, USA), and the measured wavenumber range was 4000–400 cm⁻¹. Magnetic moment measurements were performed using a Quantum Design MPMS-7 SQUID magnetometer (Santa Clara, CA, USA). Powder X-ray diffraction (PXRD) analysis was conducted using a Rigaku Ultima IV diffractometer (Rigaku Corporation, Tokyo, Japan) equipped with a Cu Kα radiation. Diamagnetic contributions in the dc susceptibility measurements were corrected using Pascal’s constants [59,60,61].

Single-crystal X-ray diffraction data for complex 1 were collected at 120.01(10) K using an XtaLAB Synergy R diffractometer (Rigaku, Tokyo, Japan) equipped with a HyPix detector, with CuKα radiation (λ = 1.54184 Å) employed. The crystal structure was solved via the direct method, with analytical refinement carried out using the software package including SHELXS-2014 [62], SHELXL-2014 [63], and Olex2–1.5. Anisotropic displacement parameters were used for refining all non-hydrogen atoms. Some important bond lengths and angles for complex 1 are listed in Table S5. The detailed crystallographic parameters of complex 1 are summarized in

Table 2. Crystallographic data of complex 1.

	Complex 1
Empirical formula	C88H68Dy2N30O6
Formula weight	1966.70
Temperature/K	120.01(10)
Crystal system	triclinic
Space group	Pī
a [Å]	10.41860(10)
b [Å]	13.5613(2)
c [Å]	15.3557(2)
a [deg]	75.1470(10)
b [deg]	81.0730(10)
g [deg]	85.4510(10)
Volume [Å3]	2069.93(5)
Z	1
ρcalc [g/cm−3]	1.578
μ [mm−1]	10.163
F(000)	986
Reflections collected	25526
Unique/parameters	8352/577
R(int)	0.0584
Goodness-of-fit on F2	1.161
R1, wR2 [I > 2σ(I)]	0.0419, 0.1132
R1, wR2 (all data)	0.0431, 0.1140

. Crystallographic data have been deposited at the Cambridge Crystallographic Data Center (CCDC: 2465080).

3.2. Synthesis of [Dy₂(H₂dapp)₂(μ-OH)₂(H₂O)₂] 4TCNQ 2CH₃OH

A quantity of 0.05 mmol (22.8 mg) of Dy(NO₃)₃·6H₂O and 0.05 mmol (17.0 mg) of H₂dapp were dissolved in 2 mL of a MeOH/H₂O mixture (v:v = 1:1) and stirred for 30 min at 70 °C. After stirring, the solution was allowed to stand at room temperature. A solution of 0.05 mmol (10.5 mg) LiTCNQ in 2 mL of MeOH/H₂O (v:v = 1:1) was carefully layered on top of the above solution. Dark blue crystals formed after a few days. Yield: 4 mg (16%, based on LiTCNQ). Anal. Calcd. for C₈₈H₆₆Dy₂N₃₀O₆: C, 31.53; H, 1.16; N, 2.94%. Found: C, 31.50; H, 1.19; N, 2.89%. IR (KBr cm⁻¹): 2175(vs, br), 2154(vs, br), 2036(s), 1614(s), 1574(vs), 1543(m), 1524(m), 1487(vs), 1088(s), 1004(s), 979(m), 797(m), 768(m), 717(m), 647(w), 572(w).

4. Conclusions

In pursuit of more radical-Ln^III complexes, a dinuclear Dy^III complex was synthesized using TCNQ^·− radical anions and a pentadentate nitrogen-rich Schiff base ligand. The Dy^III centers exhibit a geometry between D_4d and D_2d symmetries and demonstrate intramolecular ferromagnetic coupling. Magnetic studies reveal weak slow magnetic relaxation behavior at zero field with an approximate Δ_eff = 2.82 K and τ₀ = 6.88 × 10⁻⁶ s. This might be due to the fact that antiferromagnetic TCNQ^·− dimers do not effectively isolate the molecules (with the shortest Dy···Dy distances < 8 Å), causing dipolar Dy-Dy interactions that favor QTM and further limit magnetic performance. Further studies focus on designing more suitable co-ligands to develop TCNQ^·−-coordinated Dy^III complexes that effectively suppress QTM while optimizing TCNQ^·− packing, potentially enabling multifunctional materials combining conductivity and magnetic properties.