Chiral Resolution of Spin‐Crossover Active Iron(II) [2x2] Grid Complexes

Abstract Chiral magnetic materials are proposed for applications in second‐order non‐linear optics, magneto‐chiral dichroism, among others. Recently, we have reported a set of tetra‐nuclear Fe(II) grid complex conformers with general formula C/S‐[Fe4L4]8+ (L: 2,6‐bis(6‐(pyrazol‐1‐yl)pyridin‐2‐yl)‐1,5‐dihydrobenzo[1,2‐d : 4,5‐d′]diimidazole). In the grid complexes, isomerism emerges from tautomerism and conformational isomerism of the ligand L, and the S‐type grid complex is chiral, which originates from different non‐centrosymmetric spatial organization of the trans type ligand around the Fe(II) center. However, the selective preparation of an enantiomerically pure grid complex in a controlled manner is difficult due to spontaneous self‐assembly. To achieve the pre‐synthesis programmable resolution of Fe(II) grid complexes, we designed and synthesized two novel intrinsically chiral ligands by appending chiral moieties to the parent ligand. The complexation of these chiral ligands with Fe(II) salt resulted in the formation of enantiomerically pure Fe(II) grid complexes, as unambiguously elucidated by CD and XRD studies. The enantiomeric complexes exhibited similar gradual and half‐complete thermal and photo‐induced SCO characteristics. The good agreement between the experimentally obtained and calculated CD spectra further supports the enantiomeric purity of the complexes and even the magnetic studies. The chiral resolution of Fe(II)‐ [2×2] grid complexes reported in this study, for the first time, might enable the fabrication of magneto‐chiral molecular devices.


6-(pyrazol-1-yl)picolinic acid (A)
Pyrazole (6.2 g, 91.1 mmol, 2.1 eq) was dissolved in 1,4-dioxane (200 mL) in a 500 mL threenecked round-bottomed flask. The mixture was bubbled with argon for 30 min. Later potassium tertiary butoxide (10.7 g, 95.4 mmol, 2.2 eq) was added and stirred at room temperature under argon for 30 min. Later ethyl 6-bromopyridine-2-carboxylate (10.0 g, 43.5 mmol, 1 eq) was added and the mixture was stirred for 48 hours at 115 °C under Argon atmosphere. The reaction was stopped and cooled to room temperature. The solvent was evaporated under reduced pressure. The crude product was suspended and refluxed for 3 hours in EtOH (100 mL) and NaOH solution (2M, 50 mL). Ethanol was removed under reduced pressure and 50 mL water was added to the mixture. The pH was adjusted to 3 with aq. HCl, which resulted in precipitation. The colorless precipitate was collected by filtration and dried at 80 °C in an oven.

2,6-bis(6-(pyrazol-1-yl)pyridin-2-yl)-1,5-dihydrobenzo[1,2-d:4,5-d']diimidazole (L)
A (6-(pyrazol-1-yl)picolinic acid, 2.41 g, 12.75 mmol, 2.1 eq), 1,2,4,5-benzenetetramine tetra hydrochloride (1.72 g, 6.05 mmol, 1 eq) and polyphosphoric acid (20 mL) were taken in a roundbottomed flask. The reaction mixture was gently heated at 130 °C until the polyphosphoric acid got viscous enough to allow the stirring with a magnetic stirrer bar. Then the temperature was set to 200 °C and stirred for 4 hours. The reaction was cooled to 100 °C and was added into crushed ice. The aqueous suspensions were combined, and the precipitate was filtered off. The solid obtained was suspended in water, and the pH was adjusted to 10 using aq. Na2CO3 solution (2 M). After the suspension was stirred for 2 hours, it was collected by filtration. The precipitate was re-suspended in water, pH was set to 4 using HCl (1 M), and stirred overnight. The precipitate was filtered off by washing with water (250 mL) and the light brown color product obtained was dried in the oven at 120 °C.

Modified Ligand LS
L (0.895 g, 2.0 mmol, 1 eq) and Cs2CO3 (2.64 g, 8 mmol, 4 eq) were taken in a three-necked roundbottom flask and dried under vacuum at 100 °C for three hours. The flask was flushed with argon, and DMSO (30 mL) was added. After 30 min, the suspension was allowed to cool to ambient temperature. The (S)-(+)-1-bromo-2-methylbutane (0.838 g, 6 mmol, 3 eq, V=0.566 mL) was added using a syringe and continued to stir at 80 °C under Argon. The reaction was stopped after 24 h. The reaction mixture was taken into a mixture of CHCl3/EtOAc (1:1) and H2O. After phase separation, the aqueous layer was extracted twice with CHCl3. The organic layer was dried over Na2SO4, filtered, and evaporated under reduced pressure. Then the products were purified by silica gel column chromatography (1:24 MeOH/CHCl3 v/v) using CHCl3 as an initial eluent. (Length of the column = 35 cm, and diameter of the column used = 3 cm)

Modified Ligand LR
L (0.895 g, 2.0 mmol, 1 eq) and Cs2CO3 (2.64 g, 8 mmol, 4 eq) were taken in a three-necked roundbottomed flask and dried under vacuum at 100 °C for three hours. The flask was flushed with argon, and DMSO (30 mL) was added. After 30 min, the suspension was allowed to cool to ambient temperature. The (R)-(-)-1-bromo-2-methylbutane (0.838 g, 6 mmol, 3 eq, V=0.566 mL) was added using a syringe and continued to stir at 80 °C under argon. The reaction was stopped after 24 h. The reaction mixture was taken into a mixture of CHCl3/EtOAc (1:1) and H2O. After phase separation, the aqueous layer was extracted twice with CHCl3. The organic layer was dried over Na2SO4, filtered, and evaporated under reduced pressure. Then the products were purified by silica gel column chromatography (1:24 MeOH/CHCl3 v/v) using CHCl3 as an initial eluent. (Length of the column = 35 cm, and diameter of the column used = 3 cm)

Synthesis of [2x2]-iron(II) grid complexes
Synthesis of PS LS (0.20 g, 0.35 mmol, 1 eq.) was suspended in dry acetonitrile (60 mL) sonicated for 30 min and then degassed with argon for 30 min. Afterwards Fe(CF3SO3)2 (anhydrous) (0.12 g, 0.35 mmol, 1 eq.) was added. The reaction was stirred at 60 °C for 14 hours under Ar atmosphere. The reaction mixture was cooled to room temperature and filtered. The filtrate was reduced to about 5 mL. The product was precipitated by the addition of diethyl ether (30 mL) and collected by centrifugation (4000 Hz, 10 min). The ether was removed using a pipette. The dark maroon coloured precipitate was dried under vacuum. Yield: 0.22 g (69 %)

Synthesis of PR
LR (0.20 g, 0.35 mmol, 1 eq.) was suspended in dry acetonitrile (60 mL), sonicated for 30 min and then degassed with argon for 30 min. Afterwards Fe(CF3SO3)2 (anhydrous) (0.12 g, 0.35 mmol, 1 eq.) was added. The reaction was stirred for 14 hours at 60 °C under Ar atmosphere. It was cooled to room temperature and filtered. The filtrate was reduced to about 5 mL. The product was precipitated by the addition of diethyl-ether (30 mL) and collected by centrifugation (4000 Hz, 10 min). The ether was removed using a pipette. The dark maroon colored precipitate was dried at vacuum.

Synthesis of 57 FeTf2 (anhydrous):
57 Fe powder (200 mg, 3.51 mmol) was taken in dry acetonitrile (3 mL) in a 10 mL round bottomed flask. Triflic acid (0.7 mL) was added slowly dropwise, as the reaction is highly exothermic. After the initial effervescence subsided, the mixture was heated to 60 °C and maintained under Ar. Once the effervescence had stopped, the mixture was cooled to RT and unreacted Fe powder was filtered through celite pad. The filtrate was concentrated under reduced pressure and stored at -20 °C over night which resulted in colourless crystals. The crystals were filtered and by washing with DEE and dried under vacuum overnight to remove solvent.

ESI-MS and IMMS:
Experimental Instrument Parameters: About 10 µg/mL concentrated solution in acetonitrile was directly electrosprayed. A Waters Synapt G2S HDMS high-resolution mass spectrometer coupled with ion mobility separation cell was used for this purpose. The optimized parameters are as follows:

Circular dichroism (CD) of grid complexes
A Jasco J-815 CD Spectrometer (Jasco, Inc.) was used with Spectra Manager software (Jasco Inc.) to record CD spectra. Solutions were prepared by dissolving the samples in acetonitrile to a concentration of 10 -6 M. All spectra were recorded at 25 °C between 700 and 200 nm at a rate of 50 nm/min and were averaged to 2 accumulations. A 1.0 mm path length cuvette was used to hold solutions, and all spectra were background-subtracted from those of the mobile phases.

Deconvolution of diastereomeric grid complexes
Diastereomers in each sample were separated by performing recrystallization several times. In first step of crystallization, 150 mg of sample was dissolved in ca. 15 mL of acetonitrile, and THF was added slowly via vapor diffusion. This yielded in the crystalline sample. The crystals were dried, powdered and again dissolved in acetonitrile. THF was slowly diffused again to yield the second recrystallized sample. In each step, the finely powdered sample was used for measuring CD. The powder sample was used instead of crystals to maintain the uniformity. After four steps, the saturation of the CD signal in each case was observed.

X-ray diffraction analysis
The single-crystal X-ray diffraction data of the complexes S and R were collected using STOE Stadi Vari 25 diffractometer with graded multilayer mirrored monochromatic Ga-K radiation (1.34143 Å).
The structures were solved using direct methods and were refined by full-matrix least square methods and using SHELX-2014 inbuilt in Olex2. Data were collected using  and  scans chosen to complete the asymmetric unit. All non-hydrogen atoms were refined anisotropically, whereas the hydrogen atoms were calculated geometrically, riding on their parent atoms. Since the crystals were weakly diffracting, and thus due to poor quality of the data, only four of eight anions could be located. Also, all well diffracting crystals were found to be the pure diastereomers. We expect the other crystals could contain various ratio of diastereomers, which could account for the poor diffraction profile. Our attempts to determine the X-ray structure of the mixed crystals were not successful due to the inferior quality of the crystals.
Crystallographic data for the structures reported in this paper have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication no. CCDC-2064448 and 2064449. Copies of the data can be obtained free of charge from www.ccdc.cam.ac.uk/conts/retrieving.html.

TDDFT Calculations
TDDFT calculations on the S-complex were performed in order to study its CD spectrum using the escf module [1,2,3] of the Turbomole package (Version 7.4.1) [4] . For this, we used the PBE0 functional [5,6] together with the def2-SV(P) [7] basis set for all atoms except for iron for which the def2-TZVP [7] basis set was chosen. The exchange-correlation functional was evaluated using a gridsize of 5, as implemented in Turbomole. [8] Coordinates of optimized structure of S (HS complex)

Differences in ECD spectra between ΛS and ΛR complexes
In order to verify the circular dichroism and the helicity of the grid complexes, the ECD spectra of both diastereomers of the complex with Λ helicity are calculated. If the chirality of these molecules was mainly caused by the chirality of the ligands, the ΛR and ΛS complexes should exhibit ECD spectra which only differ by a sign of Δε. TDDFT calculations on both diastereomers have been carried out using the PBE0 functional and a def2-SV(P) basis set for all atoms except iron for which a def2-TZVP was chosen as described in Section 7 of this SI. Only the Low-Spin complexes have been considered.
As shown in figure SI-9, both diastereomers exhibit ECD spectra which are virtually identical. This implies that the circular dichroism of these molecules is primarily caused by the helicity of the system and not the chiral sites of the ligands. Circular dichroism can thus be used in order to verify the helicity of these systems, confirming that the experimentally analyzed structures all belong to complexes with Λ helicity.

Figure SI-9:
ECD spectra of the Low-Spin ΛR (black) and ΛS (red) complexes as calculated using TDDFT (PBE0 functional). The diastereomers exhibit ECD spectra that are virtually identical, strongly implying that the circular dichroism of these systems is primarily caused by the helicity of the grid complex, not the chiral sites in the ligands.

Magnetic measurements
All herein reported magnetic measurements were performed on an MPMS-XL7 and MPMS-XL5 magnetometers (Quantum Design). For standard magnetic as well as for photo-magnetic experiments, the temperature-dependent magnetization was recorded at BDC = 0.1 T as an external magnetic field. The temperature sweeping rate was 3 K/min (standard measurements in the dark) or 0.3 K/min (photo-magnetic experiments) and it was the same for cooling and for heating modes. Each temperature data point was stabilized for 1 minute before the measurement. Gelatin capsule (standard measurements in the dark) was used as sample holders in the temperature range 1.8 ↔ 385 K. The very small diamagnetic contribution of the gelatin capsule was negligible to the overall magnetization, which was dominated by sample. The diamagnetic corrections of the molar magnetic susceptibilities were applied using Pascal's constants. The photo-magnetic measurements were performed by using a diode-pumped solidstate lasers (DPSS) Kvant (λ= 637 nm, 300 mW) coupled through an optical fiber to the cavity of an MPMS SQUID XL7 and the power on the sample surface was adjusted to 10 mW/cm 2 . For the photo-magnetic experiments, the small amount of sample (≈1 mg) was dispersed in melted eicosane, introduced into the plastic straw and congealed. The exact weight of samples was obtained by weighting and verified by comparison of thermal χT vs. T curve with a more accurately weighed sample of the same compound. After the cooling to 5 K, the sample was irradiated and the change in magnetization was followed. When the saturation point had been reached, the light was switched off, the temperature was increased at a rate of 0.3 K/min, and the magnetization was measured at 1 K intervals. We observed the most intense increase of magnetic moment under the red light irradiation (637 nm). T(LIESST) value was determined from the minimum of the ∂(χT)/∂T vs. T curve for the relaxation process.

Mössbauer spectroscopy
57 Fe Mössbauer Spectroscopy was performed in transmission geometry employing a constant acceleration spectrometer which was used in conjunction with a 512-channel analyzer in timescale mode (WissEl GmbH). The 14.4 keV γ-radiation was delivered by a radioactive source containing 57 Co which was diffused in Rh. Source activity was 1.4 GBq. The calibration of the spectrometer was carried out with α-iron at room temperature (RT). Variable temperature measurements between 77 K and RT were conducted in a continuous flow cryostat (OptistatDN, Oxford Instruments). The spectral data were transferred from the multi-channel analyzer to a PC utilizing the public domain program Vinda running on an Excel 2003® platform. The analysis was performed with least-squares fits using Lorentzian line shapes which reveal the parameters isomer shift δ, the quadrupole splitting ΔEQ and the line width at half maximum Γ. [9]