New Ferrocene-Based Metalloligand with Two Triazole Carboxamide Pendant Arms and Its Iron(II) Complex: Synthesis, Crystal Structure, 57 Fe Mössbauer Spectroscopy, Magnetic Properties and Theoretical Calculations

: The new ferrocene-based metalloligand bis ( N -4-[3,5-di-(2-pyridyl)-1,2,4-triazoyl])ferrocene carboxamide ( L ) was prepared through derivatization of 1,1 (cid:48) -ferrocenedicarboxylic acid with 4-amino-3,5-di(pyridyl)-4 H -1,2,4-triazole. The composition and purity of L in the solid state was determined with elemental analysis, FT-IR spectroscopy, and its crystal structure with single-crystal X-ray analysis, which revealed that the substituted cyclopentadienyl rings adopt the antiperipla-nar conformation and the crystal structure of L is stabilized by O–H ··· N and N–H ··· O hydrogen bonds. The molecular properties of L in solution were investigated with NMR and UV-VIS spectro-scopies, and cyclic voltammetry disclosed irreversible redox behavior providing one oxidation peak at E 1/2 = 1.133 V vs. SHE. Furthermore, the polymeric Fe II complex {Fe(L)(C(CN) 3 ) 2 } n ( 1 ) was prepared and characterized with elemental analysis, FT-IR spectroscopy, 57 Fe Mössbauer spectroscopy, and magnetic measurements. The last two methods conﬁrmed that a mixture of low-and high-spin species is present in 1 ; however, the spin crossover properties were absent. The presented study was also supported by theoretical calculations at the DFT/TD-DFT level of theory using TPSS and TPSSh functionals.


Introduction
Multifunctional materials represent a group of any materials that integrally combine two or more applicable properties. There is currently great interest in the study and design of multifunctional molecular materials that have spin-switching as one of the functions, not only for fundamental reasons but also in attempts to make innovative multifunctional devices. However, it is relatively hard to design a single molecular material capable of performing multiple functions. One of the synthetic ways involves sophisticated design of the ligands used. Traditional ligand functions include binding to a metal center and provision of steric hindrance or binding groups. In this context, the use of a metalloligand is a powerful synthetic strategy with considerable advantages. The metalloligand is a suitably designed complex which may act as a ligand capable of placing appended functional groups in limited directions, and, therefore, such functional groups can bind to secondary metal ions in a limited geometrical manner. The metalloligand approach presents a facile way to obtain multinuclear complexes with a specific combination of a primary metal ion (within the metalloligand) and a secondary metal ion (within the resultant complex). of a primary metal ion (within the metalloligand) and a secondary metal ion (within the resultant complex). Another advantage is that the structural rigidity of the metalloligand often causes the structural motif of the resultant complex to be highly predictable [1][2][3][4].
In this context, ferrocene derivatives bearing donor substituents are suitable for the metalloligand approach and thus are very useful for the development of specifically designed multinuclear metal complexes [5,6]. One of the well-known examples of a ferrocene-based metalloligand is 1,1′-bis-(phenylphosphino)ferrocene, which is often used as a ligand in catalysis and for the generation of novel coordination compounds with a wide range of coordination geometries and properties [7].
Spin crossover (SCO) complexes show dynamic switching between high-spin and low-spin states upon external stimuli, such as temperature and/or pressure changes, as well as light radiation treatment. This transition leads to drastic changes in electronic, magnetic, optical, and mechanical properties giving them a bistable character, which could be useful for the design of molecular devices for data storage or optical displays [8][9][10][11]. Multifunctional ligands in the field of SCO are very topical due to the fact that they provide the resulting SCO complex with their secondary function, resulting in further associated properties such as porosity, electrical conductivity, magnetic order, liquid crystal, and non-linear optical activity [8][9][10][11]. Coupling the ferrocenyl group, a part with welldefined redox properties, to ligands capable of inducing SCO affords the possibility to investigate synergies between SCO and the other properties and ultimately to find new physical phenomena and potential new applications.
In this context, ferrocenyl-containing pyridyl-triazole ligands were previously studied as suitable building blocks for the synthesis of SCO complexes with redox/electrontransfer as the second function [12]. Generally, triazole-based ligands are widely studied and used to build switchable coordination compounds. For example, complexes of iron(II) comprising the abpt ligand (4- [13,14]. Ferrocenyl and analogous cobaltocenyl-containing abpt ligands ( Figure 1) were synthesized, and complexation with Co II , Cu II , Zn II , and Cd II was studied [15,16].   Here, we report the synthesis and characterization of the new ferrocene-based metalloligand L with two triazole carboxamide pendant arms and its polymeric Fe II complex ({Fe(L)(C(CN) 3 ) 2 } n (1)).

Synthesis of Metalloligand L and Fe II Complex
The ligand L was prepared with a two-step synthesis. In the first step 1,1 -ferrocenedicarboxylic acid was converted to the corresponding dichloride, which was used in the following reaction with abpt to give the product. After purification through column chromatography, the ligand was isolated as an orange solid. The identity and purity of L was confirmed on the basis of elemental analysis and multinuclear ( 1 H, 13 C) NMR data. Single crystals suitable for X-ray structure analysis were prepared with slow evaporation of a chloroform-methanol solution.
The Fe II complex (1) was prepared through the reaction of equimolar amounts of L, FeCl 2 ·4H 2 O, and potassium tricyanomethanide in a mixture of methanol-DCM under an argon atmosphere. The identity of the complex was confirmed on the basis of elemental analysis, FTIR, and 57 Fe Mössbauer data. Unfortunately, as a result of the low solubility, all experiments to prepare crystals suitable for X-ray structure analysis through recrystallization from different solvent systems or using different temperature gradients, slow diffusion, or hydrothermal synthesis were unsuccessful. Bidentate ferrocene-based ligands containing flexible spacer, including amide group, are inclined to form oligomers or low -dimensional polymers [17][18][19][20][21]. Based on this fact and the results of analyses, we suppose a polymeric structure of the Fe II complex 1.

Description of the Crystal Structure of the Ligand (L)
The ligand L crystallizes in the monoclinic space group P2 1 /c with four molecules of the ligand in the asymmetric unit (Z = 4). The molecular structure of L is shown in Figure 2 (additional structural data are available in the Supporting Information, Table S1). Here, we report the synthesis and characterization of the new ferrocene-based metalloligand L with two triazole carboxamide pendant arms and its polymeric Fe II complex ({Fe(L)(C(CN)3)2}n (1)).

Synthesis of Metalloligand L and Fe II Complex
The ligand L was prepared with a two-step synthesis. In the first step 1,1′-ferrocenedicarboxylic acid was converted to the corresponding dichloride, which was used in the following reaction with abpt to give the product. After purification through column chromatography, the ligand was isolated as an orange solid. The identity and purity of L was confirmed on the basis of elemental analysis and multinuclear ( 1 H, 13 C) NMR data. Single crystals suitable for X-ray structure analysis were prepared with slow evaporation of a chloroform-methanol solution.
The Fe II complex (1) was prepared through the reaction of equimolar amounts of L, FeCl2·4H2O, and potassium tricyanomethanide in a mixture of methanol-DCM under an argon atmosphere. The identity of the complex was confirmed on the basis of elemental analysis, FTIR, and 57 Fe Mössbauer data. Unfortunately, as a result of the low solubility, all experiments to prepare crystals suitable for X-ray structure analysis through recrystallization from different solvent systems or using different temperature gradients, slow diffusion, or hydrothermal synthesis were unsuccessful. Bidentate ferrocene-based ligands containing flexible spacer, including amide group, are inclined to form oligomers or low -dimensional polymers [17][18][19][20][21]. Based on this fact and the results of analyses, we suppose a polymeric structure of the Fe II complex 1.

Description of the Crystal Structure of the Ligand (L)
The ligand L crystallizes in the monoclinic space group P21/c with four molecules of the ligand in the asymmetric unit (Z = 4). The molecular structure of L is shown in Figure  2 (additional structural data are available in the Supporting Information, Table S1). The average values of the Fe-C and C-C bond lengths in the ferrocenyl unit of L are 2.04 and 1.41 Å , respectively ( Table 1). The average C-C-C bond angles in the cyclopentadienyl (Cp) rings are 108.0°. These values agree with those of ferrocene reported elsewhere within the experimental error [22][23][24]. The Cp rings in the crystal structure of the ligand L are slightly tilted by an angle of 2.7°. The conformation of the disubstituted ferrocenyl ligands can be defined by the torsion angle τ, defined as the torsion angle of CA- The average values of the Fe-C and C-C bond lengths in the ferrocenyl unit of L are 2.04 and 1.41 Å, respectively ( Table 1). The average C-C-C bond angles in the cyclopentadienyl (Cp) rings are 108.0 • . These values agree with those of ferrocene reported elsewhere within the experimental error [22][23][24]. The Cp rings in the crystal structure of the ligand L are slightly tilted by an angle of 2.7 • . The conformation of the disubstituted ferrocenyl ligands can be defined by the torsion angle τ, defined as the torsion angle of C A -Cp A -Cp B -C B , where C A and C B are carbon atoms bonded to the substituents and Cp A and Cp B are the centroids of the Cp rings [5]. The Cp rings in the crystal structure of L adopt the antiperiplanar conformation (τ = 157.8 • ). The dihedral angles between the mean planes of the rings of cyclopentadienyl and triazole equal 88.0 • (for cyclopentadienyl and triazole defined by the atoms C14-C18, and N2-3, N5, C6-7, respectively), and 79.4 • (for C19-23 and N8-10, C30-31). The pyridine and triazole rings of the abpt arms of L deviate significantly from planarity. The dihedral angles between the mean planes of the triazole and pyridine rings are equal to 40.4 and 26.9 • for the first and 49.3 and 38.3 • for the second arm. The crystal structure of L is stabilized by O-H···N and N-H···O hydrogen bonds ( Figure 3a). The N-H···O hydrogen bonds, between the carboxamide groups of neighboring molecules (N6-H6···O6), self-assemble the molecules into chains parallel to the a axis. The chains are interconnected through molecules of water (O7) by hydrogen bonds (O7-H7A···N2 and N7-H7···O7), which link neighboring chains into a two-dimensional network parallel with the plane ac ( Figure S5a). An additional C-H···π interaction connects the layers into a three-dimensional network (Figure 3b, ESI Figure S5b).

Electrochemical Properties
The electrochemical properties of L were studied with cyclic voltammetry in CH 3 CN (1 × 10 −3 M) containing tetrabutylammonium perchlorate as the supporting electrolyte. The cyclic voltammogram of L shows irreversible redox behavior providing one oxidation peak at E 1/2 = 1.133 V vs. SHE (∆E p = 90 mV; Figure 4). The large anodic to cathodic peak current ratio (i a /i c = 3.62) is attributed to substrate deposition at the working electrode [27]. The observed value of E 1/2 is shifted to more positive potentials in comparison with ferrocene (E 1/2 = 0.619 V vs. SHE), owing to the substitution of the cyclopentadienyl rings with electron withdrawing carboxamide groups. A similar phenomenon was observed for several ferrocene derivatives containing a carboxamide group [28].

UV-VIS Absorption Spectroscopic Studies
The absorption spectrum of L obtained in dichloromethane ( Figure S6; c = 2.1 mmol/L) shows bands at 346 nm (444 L·mol -1 ·cm -1 ) and 448 nm (241 L·mol -1 ·cm -1 ), which correspond to the π-π* transitions from the ferrocenyl to the pyridyl-triazole moiety and to d-d transitions, probably mixed with d-π* transitions, respectively. Due to high intensity, further absorption bands in the UV region were studied at a lower concentration ( Figure S6; c = 43.2 µ mol/L). These bands at 228, 256, and 292 nm with molar absorption

Magnetic Properties
The magnetic properties of 1 were measured in the temperature range of 2 to 300 K and are displayed as the effective magnetic moment (µ eff ) in Figure 5. The µ eff is practically constant in the whole temperature range adopting the value of 2.1 µ B . current ratio (ia/ic = 3.62) is attributed to substrate deposition at the working electrode [27]. The observed value of E1/2 is shifted to more positive potentials in comparison with ferrocene (E1/2 = 0.619 V vs. SHE), owing to the substitution of the cyclopentadienyl rings with electron withdrawing carboxamide groups. A similar phenomenon was observed for several ferrocene derivatives containing a carboxamide group [28].

Magnetic Properties
The magnetic properties of 1 were measured in the temperature range of 2 to 300 K and are displayed as the effective magnetic moment (µeff) in Figure 5. The µeff is practically constant in the whole temperature range adopting the value of 2.1 µB. The theoretical value of Fe II in the high-spin (HS) state should span the interval between 4.90 µB (g = 2.00) and 5.39 µeff (g = 2.20), while the low-spin (LS) state of Fe II is diamagnetic (µB is Bohr magneton). Thus, the experimental magnetic data clearly say that Fe II coordinated to L and tcm anions in 1 is present both in the LS and HS spin states and evidently spin crossover is not induced by a temperature change. The average value of µeff of 1 was used to estimate the ratio of LS and HS according to Equation (1) where NA is the Avogadro constant, µ0 is the permeability of vacuum, k is the Boltzmann constant, xHS is the molar ratio of the HS species, and χHS and χLS are the molar susceptibilities of the HS and LS states, respectively. Linear regression of 1/χmol provided the value of the Weiss constant Θ = −5.0 K and the product value of HS • HS 2 = 0.743 ( Figure 5). Therefore, xHS should cover the interval between 19% and 15% as calculated with g = 2.00 and g = 2.20, respectively. The theoretical value of Fe II in the high-spin (HS) state should span the interval between 4.90 µ B (g = 2.00) and 5.39 µ eff (g = 2.20), while the low-spin (LS) state of Fe II is diamagnetic (µ B is Bohr magneton). Thus, the experimental magnetic data clearly say that Fe II coordinated to L and tcm anions in 1 is present both in the LS and HS spin states and evidently spin crossover is not induced by a temperature change. The average value of µ eff of 1 was used to estimate the ratio of LS and HS according to Equation (1) based on the Curie-Weiss law:

57 Fe Mö ssbauer Spectroscopy
where N A is the Avogadro constant, µ 0 is the permeability of vacuum, k is the Boltzmann constant, x HS is the molar ratio of the HS species, and χ HS and χ LS are the molar susceptibilities of the HS and LS states, respectively. Linear regression of 1/χ mol provided the value of the Weiss constant Θ = −5.0 K and the product value of x HS ·g 2 HS = 0.743 ( Figure 5). Therefore, x HS should cover the interval between 19% and 15% as calculated with g = 2.00 and g = 2.20, respectively.

57 Fe Mössbauer Spectroscopy
The Mössbauer spectroscopy of 57 Fe was applied to compound 1. First, the low temperature spectrum of 1 was acquired at 5 K, and it consists of three doublets (Figure 6a). The first doublet was fitted with the isomer shift δ = 0.53 mm·s −1 and the quadrupole splitting ∆E Q = 2.28 mm·s −1 . These isomer shift and quadrupole splitting values are similar with the values for ferrocene and its amide derivatives [29,30]; therefore, this signal was assigned to iron ions coordinated to cyclopentadienyl moieties labelled as {Fe(Cp)} in Table 2. The second doublet was fitted with the isomer shift δ = 0.41 mm·s −1 and the quadrupole splitting, ∆E Q = 0.48 mm·s −1 . The isomer shift value and the quadrupole splitting are similar with the values for octahedral low-spin Fe II complexes [31] and was assigned to iron ions coordinated to abpt moieties labelled as {Fe(abpt)} LS in Table 2 This ratio corresponds to the analysis of magnetic data. Next, the room temperature (298 K) spectrum was measured, and again it consists of three doublets (Figure 6b). Analogous analysis resulted in parameters listed in Table 2. There is a small decrease of the isomer shift and the quadrupole splitting values upon increasing the temperature, which is due to the second-order Doppler effect and due to the temperature-dependent Boltzmann population of the iron(II) d-orbitals split by low-symmetry ligand field, respectively [32]. The integrated areas of the reported doublets were found to be in the ratio 50:13:37 for {Fe(abpt)} LS :{Fe(abpt)} HS :{Fe(Cp)}. Evidently, the ratio of the three signals changed with the temperature, which would usually indicate spin crossover. However, as the magnetic measurements excluded this phenomenon, the change in the intensities can be assigned to different temperature dependence of Lamb-Mössbauer factors of these species. Moreover, the analyzed parameters of δ and ∆E Q are consistent with the parameters found in other Fe II complexes [33,34] with the abpt ligand as showed in Table 2, thus confirming the correct assignment in the case of compound 1. quadrupole splitting values upon increasing the temperature, which is due to the secondorder Doppler effect and due to the temperature-dependent Boltzmann population of the iron(II) d-orbitals split by low-symmetry ligand field, respectively [32]. The integrated areas of the reported doublets were found to be in the ratio 50:13:37 for {Fe(abpt)} LS :{Fe(abpt)} HS :{Fe(Cp)}. Evidently, the ratio of the three signals changed with the temperature, which would usually indicate spin crossover. However, as the magnetic measurements excluded this phenomenon, the change in the intensities can be assigned to different temperature dependence of Lamb-Mössbauer factors of these species. Moreover, the analyzed parameters of δ and ΔEQ are consistent with the parameters found in other Fe II complexes [33,34] with the abpt ligand as showed in Table 2, thus confirming the correct assignment in the case of compound 1.

Theoretical Calculations
First, the molecular and electronic structure of the metalloligand L was theoretically studied at the DFT/TD-DFT level of theory using ORCA 5.0 software. The molecular structure of the metalloligand L was optimized with TPSS functional upon application of the SMD solvation model for dichloromethane (Figure 7, Table S2). Next, the optimized geometry underwent TD-DFT calculations with TPSSh functional comprising three hundred excited states. The resulting absorption spectrum is shown in Figure 8. In order to analyze the calculated spectrum, Multiwfn software was utilized [35]. Herein, the intensities calculated from the TD-DFT oscillator strengths were transformed into the molar absorption coefficients as implemented in Multiwfn. The compound L was divided into five fragments as graphically depicted in Figure 7. The iron atom is labelled as M, two carboxamidefunctionalized cyclopentadienes are labelled as L2 and L3, and dipyridyl-triazole units are labelled as L4 and L5. This enabled us to calculate interfragment charge transfer during electron excitation (IFCT) [36] and analyze the individual contribution of the metal-centered states (MC), intra-ligand states (IL), metal-to-ligand charge transfer states (MLCT), ligand to-metal charge transfer states (LMCT), and ligand-to-ligand charge transfer states (LLCT) as showed in Figure 8. Evidently, the strongest absorption band located at~33,000 cm −1 is based on the dominant contributions of the intra-ligand excitation of the L4 and L5 fragments with a minor contribution of MLCT (Figure 8a). Much weaker absorption in the visible part of the spectrum (~21,800 cm −1 ) is mainly caused by the combinations of the MC and MLCT contributions (Figure 8b), where L2 and L3 fragments based on cyclopentadiene and also L4 and L5 fragments based on abpt are involved within the MLCT contributions. To conclude, the main features of the UV/VIS spectrum of L reproduced with TD-DFT and IFCT analyses helped us to understand the origin of these electron excitations.
L4 and L5 fragments with a minor contribution of MLCT (Figure 8a). Much weaker absorption in the visible part of the spectrum (~21,800 cm −1 ) is mainly caused by the combinations of the MC and MLCT contributions (Figure 8b), where L2 and L3 fragments based on cyclopentadiene and also L4 and L5 fragments based on abpt are involved within the MLCT contributions. To conclude, the main features of the UV/VIS spectrum of L reproduced with TD-DFT and IFCT analyses helped us to understand the origin of these electron excitations.  Next, we also analyzed the properties of 1, and with the aim to support the observation from 57 Fe Mössbauer spectroscopy, a part of the presumed polymeric structure of 1 labelled as 1′ was optimized both for the LS and HS states using TPSS functional together with SMD solvation model for water (Figure 9, Tables S3 and S4). Such functional was utilized for the geometry optimization in the computation study by Krewald et al. focused on the 57 Fe Mössbauer spectroscopy [37]. Next, we also analyzed the properties of 1, and with the aim to support the observation from 57 Fe Mössbauer spectroscopy, a part of the presumed polymeric structure of 1 labelled as 1 was optimized both for the LS and HS states using TPSS functional together with SMD solvation model for water (Figure 9, Tables S3 and S4). Such functional was utilized for the geometry optimization in the computation study by Krewald et al. focused on the 57 Fe Mössbauer spectroscopy [37].
Next, we also analyzed the properties of 1, and with the aim to support the observation from 57 Fe Mössbauer spectroscopy, a part of the presumed polymeric structure of 1 labelled as 1′ was optimized both for the LS and HS states using TPSS functional together with SMD solvation model for water (Figure 9, Tables S3 and S4). Such functional was utilized for the geometry optimization in the computation study by Krewald et al. focused on the 57 Fe Mössbauer spectroscopy [37]. Figure 9. The DFT optimized molecular structure of 1′ in the low-spin state used for the calculation of 57 Fe Mössbauer parameters. The atoms are colored as following: iron (orange), nitrogen (blue), oxygen (red), and carbon (dark gray). Hydrogen atoms were omitted for clarity.
Afterwards, TPSSh functional and ORCA 4.2.1 were used to calculate the quadrupole splitting (ΔEQ TPSSh ) and the electron density at the iron nucleus (ρ0 TPSSh )- Table S8. The Afterwards, TPSSh functional and ORCA 4.2.1 were used to calculate the quadrupole splitting (∆E Q TPSSh ) and the electron density at the iron nucleus (ρ 0 TPSSh )- Table S8. The methodology reported in [37] was then utilized to calculate final values of the isomer shift (δ) and the quadrupole splitting(∆E Q ) for all studied complexes. Herein, we employed the reported calibration equations for TPSSh functional: The results are summarized in Table 2. Here, the iron ions undergoing the change of the spin state are labelled as {Fe(abpt)}, and the iron ions coordinated to cyclopentadienyl moieties are labelled as {Fe(Cp)}. The isomers shift values are in very good agreement with those measured at 5 K (Table 2) as can be seen for DFT-calculated values 0.54 mm·s −1 and 1.14 mm·s −1 for LS and HS {Fe(abpt)} compared to the experimental data 0.41 mm·s −1 and 1.18 mm·s −1 . Furthermore, the quadrupole splitting parameters are also consistent: 0.63 mm·s −1 and 3.28 mm·s −1 for LS and HS {Fe(abpt)} agrees well with the experimental data 0.48 mm·s −1 and 3.15 mm·s −1 . Also, the values of δ and ∆E Q for the {Fe(Cp)} fragments are consistent with the experimental data. Therefore, it seems that the DFT-optimized molecular structures of 1 is appropriate. The same procedure was applied also to the above mentioned [Fe(DAPP)(abpt)] 2+ complex (Tables S5 and S6) (DAPP = [bis(3-aminopropyl)(2pyridylmethyl)amine), and in this case there is good agreement found for the isomer shifts, whereas larger discrepancies are observed for the quadrupole splitting, which can be most likely assigned to higher temperatures at which the experimental data were acquired- Table 2. Nevertheless, the calculated data are similar to 1 .

Synthesis of the Ligand (L)
In a round three-neck flask 1,1 -ferrocenedicarboxylic acid (548.7 mg, 2.00 mmol) and pyridine (161 µL, 1.99 mmol) were dissolved in dry CHCl 3 (10 mL) under a nitrogen atmosphere at room temperature. To the well stirred mixture, oxalyl chloride (560 µL, 6.62 mmol, 3.3 eq) was added dropwise. The reaction mixture was heated to 60 • C under a reflux condenser for 2 h, and then all volatile solvents were evaporated under vacuum to give a dark red solid of 1,1 -ferrocenyl dichloride. In the next step, 1,1 -ferrocenyl dichloride was used without further purification. It was dissolved in dry DCM (10 mL) under a nitrogen atmosphere, and a solution of 4-amino-3,5-di(pyridyl)-4H-1,2,4-triazole (973.6 mg; 4.09 mmol, 2 eq) and pyridine (330 µL, 4.08 mmol) in dry DCM (10 mL) was added. The reaction mixture was stirred over a period of 48 h at room temperature. The obtained dark orange suspension was washed with 0.01 M HCl (4 × 10 mL). The organic phase was dried over anhydrous sodium sulfate, filtered, and evaporated in vacuum. The crude product was purified with column chromatography using silica gel and a mixture of CHCl 3 , MeOH and ammonia (w = 25%) in a volume ratio of 15:4:0.5 as a mobile phase.

Analytical Methods
Elemental analysis (C, H, N) was performed on a Flash 2000 CHNO-S Analyzer (Thermo Scientific, Waltham, MA, USA). Infrared spectra (IR) were recorded on a Jasco FT/IR-4700 spectrometer (Jasco, Easton, MD, USA) using the ATR technique on a diamond plate in the range 400-4000 cm −1 . Electronic spectra were recorded on a Cintra 3030 (GBC Scientific Instruments, IL, USA) spectrometer with 10 mm path length quartz cuvettes in dichloromethane. The mass spectra (MS) were collected on a LCQ Fleet Ion Mass Trap mass spectrometer (Thermo Scientific, Waltham, MA, USA) equipped with an electrospray ion source and a three-dimensional ion-trap detector in the positive mode. Cyclic voltammetry (CV) measurements were carried out using an electrochemical analyzer CHI600C (CH Instrument, Austin, TX, USA) with a three-electrode-type cell. A glassy carbon working electrode, a platinum wire auxiliary electrode, and a Ag/Ag + reference electrode (0.01 M AgNO 3 in 0.1 M TBAP, CH 3 CN) were used during the measurements. The internal Fc/Fc + standard (E 1/2 = 0.077 V vs. reference electrode, E 1/2 = 0.624 V vs. SHE) was employed in order to obtain the final potential values referred to SHE [41]. The measurements were performed under an inert argon atmosphere in an acetonitrile solution containing tetrabutylammonium perchlorate (TBAP), as a supporting electrolyte. The 1 H and 13 C NMR spectra were recorded at 298 K on a Varian 400 MHz NMR spectrometer (Varian, Palo Alto, CA, USA) operating at 399.95 MHz ( 1 H) and 100.60 MHz ( 13 C). The signal assignments in 1 H and 13 C NMR spectra were based in part on two-dimensional COSY, HMBC and HMQC experiments. The multiplicity of the signals was indicated as follows: s-singlet, d-doublet, and t-triplet. The transmission 57 Fe Mössbauer spectrum of complex 1 was measured with laboratory Mössbauer spectrometer with a 57 Co(Rh) radiation source. The Mössbauer spectrum was fitted with the Lorentzian line shapes using MossWinn 4.0 program. The isomer shift values were referred to the 28 µm α-Fe foil (Ritverc). For the lowtemperature Mössbauer measurement (5-300 K), the sample was placed inside the closedcycle cryogen-free cryogenic system for Mössbauer spectroscopy (Cryostation, Montana Instruments). The magnetometry was performed using a low temperature vibrating sample magnetometer (Cryogenic Limited) in the temperature range 2-300 K in the magnetic field of 0.2 T. The experimental data were corrected for the diamagnetism of the sample and for the diamagnetism of the sample holder.

X-ray Crystallography
Single crystals of L for X-ray structure analysis were prepared with slow evaporation of a chloroform-methanol solution. Data collection for L was done using an XtaLAB Synergy-I diffractometer with a HyPix3000 hybrid pixel array detector and microfocused PhotonJet-I X-ray source (Cu Kα). The structure was solved using SHELXT [42] program and refined through the full matrix least-squares procedure with Olex2.refine [43] in OLEX2 (version 1.5) [44]. The multi-scan absorption corrections were applied using the program CrysAlisPro 1.171.40.82a [45]. Figures with detailed structure features were drawn using Diamond software [46]. Non-routine aspects of crystal structure determination and refinement are as follows: The isolated single-crystals were of a poor quality, and this affected the data collection. We collected complete a data set for diffractions only up to resolution 0.88 Å. One of the abpt moieties L exhibited positional disorder that was modeled as disorder over two positions (ratio of occupational factors: 0.55:0.45).
The crystallographic data and refinement data for L are as follows: C 36 Table 1.

Theoretical Calculations
The ORCA 4.2 or ORCA 5.0 software was used for all quantum chemical calculations [47,48]. The molecular geometries were optimized with ORCA 5.0 using the polarized triple-ζ quality basis set def2-TZVP for all atoms, except carbon and hydrogen atoms for which the def2-SVP basis set was applied [49]. The calculations utilized the Split-RI-J Coulomb approximation [50] with the auxiliary basis sets def2/J [51]. Increased integration grids (DEFGRID3) and tight SCF convergence criteria were used in all calculations. The meta-GGA functional TPSS was employed together with the atom-pairwise dispersion correction (D3BJ) [52,53]. The geometry optimization was not done in vacuum but using a SMD solvation model [54]. Moreover, the tight optimization criteria were required (TightOpt), and all convergence criteria must have been fulfilled (EnforceStrictConvergence was set to True). The vibrational analyses confirmed proper convergence for complexes at local energy minimum (no imaginary frequencies)- Table S7. The subsequent calculations utilized hybrid meta-GGA functional TPSSh [55,56] together with the chain-of-spheres (RIJCOSX) approximation to exact exchange [57,58] as implemented in ORCA.

Conclusions
In conclusion, we have designed, synthesized, and characterized with different spectral analyses a new ferrocene-based metalloligand with two triazole carboxamide pendant arms L and presumable polymeric Fe II complex 1. The structure of L was characterized through single crystal X-ray structure analysis. Magnetic data and 57 Fe Mössbauer spectra of 1 confirmed the presence of the mixture of low-and high-spin species. For better understanding of the electronic spectra of L, as well as magnetic properties and Mössbauer spectra of complex 1, theoretical calculations were performed.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/inorganics10110199/s1, Figure S1: 1 H-1 H g-COSY NMR spectrum of L; Figure S2: 1 H-13 C g-HMQC NMR spectrum of L; Figure S3: 1 H-13 C g-HMBC NMR spectrum of L; Figure S4: Comparison of FTIR spectra of studied ligand (L) and complex (1); Figure S5: (a) Representation of the two-dimensional hydrogen-bonding network in the crystal structure of L. The network of interconnected neighboring molecules of L is parallel with the plane ac. (b) Representation of the final three-dimensional network created through interconnection of layers with C-H···π interactions. The layers are colored for clarity (orange and green); Figure S6: The UV-Vis absorption spectrum of ligand L in DCM solution with molar concentrations c = 2.1 mmol/dm 3 (top) and c = 43.2 µmol/dm 3 (bottom); Table S1: Crystal data and structure refinements for L; Table S2: The XYZ coordinates of the molecular structure of L optimized with DFT; Table S3: The XYZ coordinates of the molecular structure of 1 in the low-spin state optimized with DFT; Table S4: The XYZ coordinates of the molecular structure of 1 in the high-spin state optimized with DFT; Table S5: The XYZ coordinates of the molecular structure of [Fe(DAPP)(abpt)] 2+ in the low-spin state optimized with DFT; Table S6: The XYZ coordinates of the molecular structure of [Fe(DAPP)(abpt)] 2+ in the high-spin state optimized with DFT; Table S7: The list of calculated frequencies for DFT optimized molecular structures in Tables S2-S7; Table S8

Acknowledgments: The authors acknowledge Lubomír
Havlíček for the measurement of the magnetic data for 1, Lukáš Kouřil for his help with the Mössbauer spectra, and Radka Křikavová for language corrections.

Conflicts of Interest:
The authors declare that they have no known competing financial interest or personal relationship that could have appeared to influence the work reported in this paper.