Three Novel Thiazole-Arm Containing 1,3,4-Oxadiazole-Based [HS-HS] Fe(II) Dinuclear Complexes

Novel synthesis of 2,5-bis[(1,3-thiazol-2-ylmethyl)amino]methyl-1,3,4-oxadiazole (LThiazole) is reported, along with the preparation of three new dinuclear Fe(II) complexes with different counterions [Fe2(μ2-L)2](BF4)4·2CH3CN (1), [Fe2(μ2-L)2](ClO4)4 (2) and [Fe2(μ2L)2](CF3SO3)4·2CH3CN (3). The obtained complexes were characterized by single-crystal X-ray crystallography, SQUID magnetometry and IR spectroscopy. The structure of the crystalline material was determined at 120 K for 1 and 3. The magnetic properties of all three complexes (1–3) were measured between 2–300 K and clearly show that all three complexes stay in the high-spin state over the measured temperature range.


Introduction
The chemistry of first-row transition metals is fascinating because of their plethora of very interesting properties. The most interesting of all is possessing partially populated d-orbitals, which lead to their different oxidation and spin states [1,2]. The ambiguity around the spin states of the metals is not as thoroughly understood as the oxidation state. In theory, controlling the spin state of the transition metal is straightforward through ligand design. Molecular complexes containing d 4 to d 7 electronic configuration can have two different spin states which can be switched by applying an external stimulus. This is still fascinating and has potential for applications in sensors, displays and data storage devices [3][4][5][6][7]. In addition to this, spin state plays a very important role in metalloenzymatic reactions, biological activities and in catalysis involving organometallic compounds where different reactions take place via different spin states [8][9][10][11]. The phenomenon by which the compounds can switch their spin state from low spin (LS) to high spin (HS) with an external stimulus such as heat [12,13], pressure [14][15][16], light [17,18] or magnetic field [19] is called spin crossover (SCO). This phenomenon was first discovered in Fe(III) [20,21] but more commonly studied in Fe(II) [22][23][24][25] because the spin-state transition is between a diamagnetic low-spin (LS) state (S = 0) to a paramagnetic high-spin (HS) state (S = 2). This phenomenon is also rarely studied in Co(II) [26,27] and more rarely studied in Mn(III) [28][29][30][31] and Cr(II) [32][33][34]. In the case of the Fe(II) SCO complexes, the N-6 ligand donor set is commonly preferred [35]. The SCO phenomenon in solid state is sensitive to the effect of crystal packing [36][37][38], choice of counterion [39,40], presence of lattice solvents [41][42][43], and the choice of coligands [44,45]. In this quest of finding the right ligand field strength for the central metal ion to exhibit bistability of its spin state, various ligands have been explored and the ligands were designed to fine-tune the switching temperature.
Complexes based on 1,3,4 oxadiazole are seldom studied in SCO literature. Our recent advances have shown that the ligand design using (PMOD = 2,5-bis{[(2-pyridylmethyl)amino]methyl}-1,3,4-oxadiazole) [46] together with Fe(II) leads to the formation of [Fe 2 (µ 2 -PMOD) 2 ](X) 4 -type complexes where X = BF 4 , ClO 4 and CF 3 SO 3 [46]. The complex with the BF 4 − counterion stays in HS over the temperature range measured, while the complexes with the ClO 4 − and CF 3   show SCO behavior with a T 1/2 = 150 K [46]. So far, to our best of our knowledge these are the only two 1,3,4-oxadiazole complexes reported that show SCO behavior. To explore this further, the ligand designed is altered by changing the axially coordinated six-membered pyridine rings with a five-membered thiazole ring with a sulfur as a heteroatom, to learn its influences on the SCO behavior of the dimeric Fe(II) complexes, which from the literature is known to reduce the ligand field. It is worthy to note that Fe(II) dimeric complexes with a very similar ligand based on 1,2,4-thiadiazole, reported earlier by some of us, was also SCO active [47]. Since the SCO phenomenon in solid state is highly dependent on the counterion, three different commonly used counterions (BF 4 − , ClO 4 − and CF 3 SO 3 − ) have been selected and the results are shown here.

Materials and Methods
All chemicals were used without further purification and were purchased from Alfa Aesar, Acros Organics and Fisher Chemicals. The C, H and N elemental analyses were carried out on a Foss Heraeus Vario EL (Elementar Analysensysteme GmbH, Langenselbold, Germany) at the microanalytical laboratories of the department of chemistry at Johannes Gutenberg University Mainz. The infrared absorption spectra were collected at room temperature in a range of 3000-450 cm −1 on a Thermo Fisher NICOLET Nexus FT/IR-5700 spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) equipped with Smart Orbit ATR Diamond cell (Thermo Fisher Scientific, Waltham, MA, USA). X-ray diffraction data for the structure analysis were collected from suitable single crystals at 120 K with a STOE IPDS 2T at Johannes Gutenberg University Mainz. The structures were solved with ShelXT [48][49][50] and refined with ShelXL [49][50][51] implemented in the program Olex2 [52]. Magnetic data of a microcrystalline sample were collected using a Quantum Design SQUID magnetometer MPMSXL in a temperature range between 2-300 K with an applied field of 1 kOe (0.1 T). The magnetic contribution from the holder was experimentally determined and subtracted from the measured susceptibility data. The contribution from underlying diamagnetism was corrected using the estimation in equation χdia = kM·10 −6 cm 3 mol −1 [53]. In this equation, M is the molecular weight of the compound and k is equal to 0.5.

Complex Synthesis
L Thiazole (32 mg, 0.1 mmol, 1 eq) was dissolved in 2 mL of acetonitrile. To this solution, Fe(BF 4 ) 2 ·6H 2 O (34 mg, 0.1 mmol, 1 eq), dissolved in 2 mL acetonitrile, was added dropwise and stirred for an hour. The reaction mixture was set for vapor diffusion against diethyl ether and kept undisturbed at room temperature. After three days, colorless block-shaped crystals suitable for X-ray structure analysis were obtained. The crystals were collected by filtration in moderate yields (23.79 mg, 0.022 mmol, 22%), washed with diethyl ether and air-dried. FT-IR: [Fe 2 (µ 2 -L Thiazole ) 2 ](CF 3 SO 3 ) 4 · 2CH 3 CN L Thiazole (32 mg, 0.1 mmol, 1 eq) was dissolved in 2 mL of acetonitrile. To this solution, Fe(CF 3 SO 3 ) 2 (35 mg, 0.1 mmol, 1 eq), dissolved in 2 mL acetonitrile, was added dropwise and stirred for an hour. The reaction mixture was set for vapor diffusion against diethyl ether and kept undisturbed at room temperature. After one week, colorless block-shaped crystals suitable for X-ray analysis were obtained. The crystals were collected by filtration in low yields (22.54 mg, 0.017 mmol, 17%), washed with diethyl ether and air-dried. FT-IR:

Crystallographic Discussion of 1 and 3
The X-ray crystal data collected for 1 and 3 at 120 K show that the two complexes are isostructural to each other and crystallize in a triclinic P space group (Appendix A). The crystallographic data and the selected bond lengths are tabulated in Tables 1 and 2. The two iron(II) ions are bridged by the 1,3,4-oxadiazole moieties of two opposing ligands ( Figure 1). The secondary amine functions of both ligands are also coordinating in the equatorial position, thus saturating the horizontal vertices of both octahedral coordination spheres. The axial position is filled by the thiazole moiety using the nitrogen donor. Both ligands coordinate in a cis-axial manner (Figure 1, left).  The asymmetric unit consists of one ligand coordinated to an iron(II) center, two perchlorate anions and one acetonitrile in the case of 1 (Figure 2). The complex cation is located on top of a center of inversion in the middle of both ab-planes of the unit cell. Thus, half of the complex cation points into the unit cell while the other half formally lies outside of the unit cell. One tetrafluoroborate and its symmetry-generated pendant lie between the two halves of the two complex molecules. The acetonitrile molecules are located between the thiazole rings above an oxadiazole unit ( Figure S11). This BF 4 − anion is involved in hydrogen bonding, connecting the complex cations in a one-dimensional way ( Figure S13). The amine function of one ligand and a fluorine atom of the tetrafluoroborate shows that N1-H1· · · F1 distance is equal to 2.01 Å, with an N1-H1-F1 angle of 165.5 • . Two other fluorine atoms of the same BF 4 − anion form additional hydrogen bonds to a neighboring complex cation. The N4-H4· · · F4 distance is found to be 1.95 Å with an enclosed N4-H4-F4 angle of 165.5 • , while the N4-H4· · · F3 distance is equal to 2.65 Å. The N4-H4-F3 angle appears to be 129.9 • . Due to the bridging nature of the BF 4 − counter ion, a one-dimensional chain of complex cation is formed via hydrogen bonds ( Figure S13). Since the complex cation is inversesymmetrical, the coordination spheres of both iron(II) centers are equivalent (Figure 2, right). Thus, only the coordination environment of Fe1 will be discussed. The mean distance between Fe1 and a secondary amine function [(Fe-N Amine ) = (d(Fe1-N1) + d(Fe1-N4))/2] is found to be 2.34 Å, which represents the longest Fe-N bond type. The next smaller average bond length can be found between the iron(II) ion and the oxadiazole nitrogen atom [(Fe-N ODA ) = (d(Fe1-N2) + d(Fe1-N3))/2 = 2.14 Å]. With an average value of (Fe-N TA ) = (d(Fe1-N5) + d(Fe1-N6))/2 = 2.13 Å, the bonds between the metal center and the thiazole ring are the shortest. While the amine function only offers σ-bonding, the metal ligand bonds to the oxadiazol and thiazole moieties can have further π-back bonding character [54]. Therefore, the bond distances between Fe1 and the heteroaromatic compounds are shorter than those to the amine groups.  Table 1. It is known by literature that high-spin iron(II) complexes have a larger distortion than low-spin complexes. Thus, the octahedral distortion parameter for LS complexes is usually less than 80 • , while it is greater than 100 • for HS complexes [54]. This can be explained by the asymmetric electron configuration of the iron(II) ion in the high-spin state, compared to the symmetric population of the t 2g -orbitals of the low-spin species. In the case of complex 3, the asymmetric unit consists of one L Thaizole coordinated to an iron(II) ion, one acetonitrile and two trifluoromethanesulfonate anions ( Figure S15). Thus, one half of the unit cell is symmetry generated. The distance d(O1-N7) between the oxadiazole (O1) and the acetonitrile (N7) equals to 2.90 Å. On the other side of the oxadiazole motif a triflate anion is located with d(O1-S3) = 3.60 Å ( Figure S15). Two distinguishable hydrogen bonds between an amine function of the ligand and a triflate anion are found in complex 3. The first one can be found between N3-H3· · · O4 with a distance of 1.97 Å and an N3-H3-O4 angle of 161 • . Next to that, a second hydrogen bond N4-H4· · · O7 (2.05 Å) is formed, showing an N4-H4-O7 angle of 153 • ( Figure S16).
The bond lengths of the complex cation look very similar to 1, where the longest average Fe-N bond type is displayed by (Fe-N Amine ) (2.36 Å), followed by (Fe-N ODA ) (2.14 Å) and (Fe-N TA ) (2.11 Å). These bond distances result in a mean Fe-N bond distance of 2.21 Å. The octahedral distortion parameter ((FeN 6 ) = 128 • ) shows a larger distortion of the coordination sphere than found for 1, which is confirmed by continuous shape measurement (CShM(O h ) = 4.57).The bond lengths indicate that both iron centers stay in [HS-HS] state at the given temperature of 120 K. Selected bond lengths are tabulated in Table 1.

Infrared Spectroscopy
IR spectra of the isolated complexes are shown in Figure 4. All three compounds 1-3 show a blueshift of the symmetric N-H stretching frequency [ν s (N-H)] compared to the free ligand L Thiazole (3277 cm −1 ). Here, the biggest difference is found in 1 (3333 cm −1 ), followed by 2 (3307 cm −1 ) and 3 (3278 cm −1 ). In most cases, the coordination of a NHR 2 group to transition metals results in a redshift of the symmetric N-H stretching frequency [56]. However, there are also reported cases of a shift towards higher wavenumbers [57]. The different shifts of the symmetric N-H stretching frequency can be caused by varying hydrogen bonds to the secondary NHR 2 group of the ligand. The coordination of the oxadiazole-N results in a red shift of the corresponding C=N stretching mode, moving to lower wave numbers from a strong ν(C=N) stretching at 1667 cm −1 for the free ligand to less pronounced weak stretching around 1640 cm −1 as expected due to the weakening of the C=N bond when binding to the transition metal [58]. In addition to the ν s (N-H) band, the counter ion-specific IR bands are also clearly visible. Hence, 1 shows very pronounced asymmetric B-F stretching bands ν as (BF 4 ) at 1084 cm −1 , 1052 cm −1 and 1036 cm −1 . The IR band at 762 cm −1 can be assigned to the symmetric stretching mode ν s (BF 4 ), while the bands at 522 cm −1 , 506 cm −1 and 500 cm −1 display deformation vibrations δ(BF 4 ) [58].

Magnetic Measurement of 1-3
Magnetic measurements of 1-3 were carried out in a temperature range from 300-2 K at 0.1 T. Therefore, the magnetic susceptibility was determined and plotted as χ M T product versus the temperature (T) for each complex ( Figure 5). [Fe 2 (µ 2 -L Thiazole ) 2 ](BF 4 ) 4 · 2CH 3 CN (1) and [Fe 2 (µ 2 -L Thiazole ) 2 ](CF 3 SO 3 ) 4 ·2CH 3 CN (3) were measured as colorless block-shaped crystals, while [Fe 2 (µ 2 -L Thiazole ) 2 ](ClO 4 ) 4 (2) was isolated and studied as microcrystalline powder. The χ M T value of 2 at 300 K is equal to 7.28 cm 3 mol −1 K. This value is higher than the spin-only value, which is 6 cm 3 mol −1 K for two S = 2 spin centers with g = 2 ( Figure 5). Below 50 K, the χ M T product for 2 drops significantly to 0.62 cm 3 mol −1 K at 2 K. The decline of χ M T from high temperatures towards 2 K is caused by antiferromagnetic coupling between the oxadiazole-bridged iron(II) ions. Simulation with the spin-exchange Hamiltonian H = −2J ij ∑Si · Sj leads to a weak antiferromagnetic-exchange interaction parameter J = −0.66 (2) cm −1 with g = 2.217 (4), ( Figure 5, blue solid line). The data were fitted with the PHI program [59]. High crystallographic anisotropy can lead to pronounced magnetic anisotropy, which in turn can cause the orientation of the crystals within the applied magnetic field. Since the crystals were measured in a nonblocked way, an alignment of the crystals with the magnetic field can occur. The resulting measured magnetic moment can therefore be formally higher than expected. For 1 and 3, in contrast, the χ M T product increases with decreasing temperature, which is due to orientation effects of the single crystals. Both complexes show a constant χ M T value of approx. 7 cm 3 mol −1 K in the range of 300-150 K. Regardless of the magnetic coupling of both iron(II) centers, complexes 1-3 do not exhibit SCO activity. Thus, all three complexes 1-3 remain in the [HS-HS] state. The change of the six-membered pyridine group in the sidearms to the five-membered thiazole did not lead to the desired SCO-active dinuclear iron(II) compounds.

Conclusions
We report the synthesis and magnetic properties of three novel dinuclear Fe(II) complexes based on a new bridging 1,3,4-oxadiazole ligand bearing coordinating thiadiazole side arms. All three compounds are locked in the high-spin state [HS-HS]. The reason for this can be assigned to a combination of geometrical and electronic effects. From our earlier work where ligand field strength in imidazole-arm-containing thiadiazole-based [Fe 2 (µ 2 -I 2 MTD) 2 ](X) 4 complexes (I 2 MTD = 2,5-bis[(1H-imidazol-2-ylmethyl)amino]methyl1,3,4thiadiazole) is lower than pyridyl-arm-containing thiadiazole-based complexes of type [Fe 2 (µ 2 -PMTD) 2 ](X) 4 (PMTD = 2,5-bis[(2-pyridyl-methyl)amino]methyl-1,3,4-thiadiazole), (X = BF4 − , ClO 4 − and CF 3 SO 3 − ) [46]. While the [Fe 2 (µ 2 -I 2 MTD)2](X) 4 -type complexes remain in the [HS-HS] state between 2-300 K, the compounds of type [Fe 2 (µ 2 -PMTD) 2 ](X) 4 stabilize the diamagnetic [LS-LS] state, showing only the start of an incomplete SCO above room temperatures. This is because of the different angles between the heteroaromatic ring and the adjacent methylene group. The angle α, formed by the pyridine ring, is approximately 120 • , whereas the angle β, formed by a five-membered ring, is approximately 125 • . Due to the steric effect of such ligand systems, the bigger angle β results in longer Fe-N-bonds favoring the high-spin state. The same arguments seem to apply for the analogous oxadiazole-based ligands L Thiazole and PMOD. In addition to that, thiazole is considered to generate a weaker ligand field than pyridine, which we understand from the basic ligand-field arguments and literature.
Interestingly, the coordination of a 5-membered thiazole heteroaromatic ring in the apical positions (1-3) does not lead to longer axial bond lengths compared to the [Fe 2 (µ 2 -PMOD) 2 ] (X) 4 -type complexes [46]. In contrast, all three complexes show smaller average Fe-N bond lengths to the axially coordinated aromatic heterocycle than the analogous iron(II) complexes of PMOD. It appears that the increased angle β (Figure 6) has more effect on the Fe-N Amine bonds. The fact that complexes 1-3 have shorter axial bond lengths and significantly longer Fe-N Amine distances than the [Fe 2 (µ 2 -PMOD) 2 ] (X) 4 complexes clearly shows the modification in the steric effect of the ligand. Most likely, stronger π-back bonding to the thiazole ring in 1-3 still cannot compensate for the sterically unfavorable constraint of the five-membered ring in the ligand side arm. Thus, the increased steric tension and the lower ligand-field strength of L Thiazole lead to dinuclear iron(II) [HS-HS] complexes without any SCO behavior in all three cases. in CDCl 3 . S3. FD Mass spectroscopy: Figure S10: FD mass spectrum of 2,5-Bis[(1,3-thiazol-2ylmethyl)amino]methyl-1,3,4-oxadiazole (L Thiazole ). S4. Crystallography: Figure S11: Asymmetric Unit of 1 with front and side view; Figure S12: Unique hydrogen bonds in 1 between complex cations and tetrafluoroborate anions; Figure S13: One-dimensional network of complex cations via hydrogen bonds in 1; Figure S14: Unit cell of 1 with view along the bisector of γ and perspective; Figure S15: Asymmetric Unit of 3 with front and side view; Figure S16: Unique hydrogen bonds in 3 between complex cations and tetrafluoroborate anions; Figure S17: Unit cell of 3 with view along the a-axis and perspective; Table S1: Selected angles of complex 1 and 3.
Author Contributions: Synthesis, investigations, experimental, J.E.; methodology, J.E. and E.R.; spectroscopy, spectrometry and magnetic data analysis, J.E. and S.S.; crystal structure analysis, L.M.C.; writing-original draft preparation, S.S. and J.E.; writing-review and editing, S.S. and E.R.; supervision, E.R.; funding acquisition, E.R. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement:
The magnetic data presented in this study are available on request from the corresponding author.