Formation of a diiron–(μ-η1:η1-CN) complex from acetonitrile solution

A diiron end-on μ2-η1:η1-CN-bridged complex is obtained from a crystallization experiment of an open-chain iron–NHC complex. The cyanide appears to originate from the acetonitrile (MeCN) solvent by C—C bond cleavage or through carbon–hydrogen oxidation.


General procedures and analytical methods
Complex 1 was synthesized according to a literature method (Schlachta et al., 2024).Solvents were purified, dried and degassed using standard methods (Armarego, 2017) or received from a solvent purification system by M. Braun.All other chemicals were obtained from commercial suppliers and were used without further purification.NMR spectra were recorded on a Bruker Avance Ultrashield AV400 (400.13MHz for 1 H NMR and 100.53MHz for 13 C NMR).The chemical shifts are given in � values in ppm (parts per million) relative to TMS (tetramethylsilane) and are reported relative to the residual deuterated solvent signal (Fulmer et al., 2010).Electrospray ionization mass spectrometry (ESI-MS) data were measured on a Thermo Fisher Ultimate 3000.FT-IR measurements were conducted on a PerkinElmer Frontier FT-IR spectrometer (ATR).The 'inVia Reflex Raman System' comprises a research grade optical microscope [Leica DM2700M, Magnification 5�, 20� and 50� (in this case, 50� was used)] coupled to a high-performance Raman spectrometer (Renishaw).A 633 nm wavelength laser was used (Renishaw RL633 Class 3B).

Crystallization of 2
Single crystals of 2 suitable for X-ray diffraction were obtained by slow evaporation of a solution of 1 in CD 3 CN over a period of six months at room temperature under an ambient atmosphere near a window with sunlight (see supporting information).
A solution of 1 (around 1-2 mg) in CD 3 CN (around 0.4 ml, dry and degassed) from an NMR tube (see supporting information) was placed in a 10 ml vial under an ambient atmosphere.A human hair was fixed with adhesive tape to the inside of the vial, reaching into the solution.Heterogeneous nucleation occurs more frequently than homogeneous nucleation (Sear, 2014;Pruppacher & Klett, 1997) and human hair has been used for the growth of nanoparticles or as catalyst-support material (Deng et al., 2016;Liu et al., 2015;Haveli et al., 2012;Walter et al., 2006).The vial was closed and the cap was punctured with a cannula.The vial was left for six months at room temperature under ambient conditions near a window with sunlight, allowing the solvent to evaporate slowly.Orange crystals suitable for SC-XRD analysis were obtained.(Sheldrick, 2015b), PLATON (Spek, 2020) and enCIFer (Allen et al., 2004).

Figure 1
The molecular structure of 1. H atoms and hexafluorophosphate anions have been omitted for clarity.Displacement ellipsoids are shown at the 50% probability level (Schlachta et al., 2024).

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 1.H atoms could not be located in difference Fourier maps and were calculated in ideal positions (riding model), with C-H = 0.98 A ˚and U iso (H) = 1.5U eq (C) for CH 3 groups, C-H = 0.99 A ˚and U iso (H) = 1.2U eq (C) for CH 2 groups, and C-H = 0.95 A ˚and U iso (H) = 1.2U eq (C) for CH groups.Split-layer position refinement was used for atoms P2, F7, F8, F9, F10, F11 and F12 (PF 6 À anion), as well as N8 and C28 (bridging cyanide).Restraints were applied to atoms N8 and C28 to ensure reasonable ellipsoids.CD 3 has been modelled as CH 3 as there is no appreciable difference in SC-XRD.

Results and discussion
When a solution of 1 (Scheme 1 and Fig. 1) in CD 3 CN was evaporated slowly over a period of six months under ambient conditions, a diiron end-on �-� 1 :� 1 -CN-bridged complex, [(MeCN)(NHC)Fe] 2 (�-� 1 :� 1 -CN)(PF 6 ) 3 (2) (Scheme 2 and Fig. 2), was obtained, as determined by X-ray diffraction.The two iron centres are bridged by a cyanide anion, hence three PF 6 À anions are present in the crystal structure.Under similar conditions, i.e.MeCN solution, room temperature and air, a dinuclear Cu II cryptate has been found to form a �-� 1 :� 1 -CNbridged complex by C-C bond cleavage of MeCN (Lu et al., 2004).A possible mechanism involving the activation of the sp-hybridized C atom of MeCN, bound to one Cu atom (MeCN-Cu), by the second Cu centre has been suggested.The increased electrophilicity of the methyl group would allow cleavage by H 2 O to form MeOH and the cyanide-bridged compound (Lu et al., 2004;Ahmad et al., 2020).Another possible mechanism for the formation of 2 might be the carbon-hydrogen oxidation of MeCN by iron complex 1 to form glycolonitrile, as observed previously for an iron(III) tetracarbene complex, and subsequent release of cyanide upon decay of glycolonitrile (Knapp et al., 2012;Dyckhoff et al., 2021;Lewis, 2008).Due to the stronger Me-CN bond (122 kcal mol À 1 ) compared to the H-CH 2 CN bond (93 kcal mol À 1 ) (Spentzos et al., 2020;Blanksby & Ellison, 2003;Goebbert et al., 2010;Miscione & Bottoni, 2014), the carbonhydrogen oxidation of MeCN seems to be more likely the origin of cyanide in this case.However, C-C bond cleavage of MeCN by UV irradiation is known (Grirrane et al., 2016) and, given the fact that the crystallization setup with 1 was also accessible for sunlight during the extensive period of six months, C-C bond cleavage of MeCN cannot be excluded.
The crystal structure of 2 reveals strongly bent equatorial NHC ligands.This finding is in stark contrast to 1, where the NHC ligand is largely planar (Fig. 1).This sandwich-like structure encapsulates the cyanide ion and is indicative of some noncovalent interactions between the equatorial ligands, likely contributing to the stability of 2. Interestingly, the pyridine units are bent less towards the centre compared to the NHC units, forming a Z-shape or diamond-shape, depending on the viewing angle of 2. The Fe-N-C angle is slightly bent (Table 2) in a trans fashion, resulting in a 'zigzag' vertical axis.Another interesting finding is the rotation of the NHC ligands towards each other in an anti conformation, resulting in a dihedral angle (CH 2 -Fe-Fe-CH 2 ) of 160.7 � (Scheme 2).The crystal structure can in principle also be solved as the diiron-(�-� 1 :� 1 -N 2 ) complex (Fig. 2), which is why we refrain from a detailed structural discussion at this point.However, there are several arguments against a diiron-(�-� 1 :� 1 -N 2 ) complex: (i) The main argument against a diiron-(�-� 1 :� 1 -N 2 ) complex is the fact that the crystal structure contains three counter-ions.As the crystallization was performed with 1 containing an iron(II) centre, bridging two Fe II atoms with a neutral N 2 ligand should lead to the presence of four counterions.Otherwise, three counter-ions would indicate that a redox process has occurred during the formation of 2, but the nature of a hypothetical reducing agent and the location of reduction are highly speculative.The main components of the

Figure 2
The molecular structure of 2. H atoms and hexafluorophosphate anions have been omitted for clarity.Displacement ellipsoids are shown at the 50% probability level.
crystallization experiment were 1 and CD 3 CN, as well as unreacted ligand precursor as a minor impurity (see supporting information).In a cyclic voltammetry study of 1, the first reduction event occurred at À 1.78 V (versus Fc/Fc + ).A preliminary experiment measuring 1 in cyclic voltammetry under an N 2 atmosphere did not show significant redox processes or electric current.Considering all these facts, the involvement of a redox process appears to be quite implausible.
(iii) A diiron-(�-� 1 :� 1 -N 2 ) complex should show a distinctive v NN absorption band in Raman spectroscopy and be IR inactive due to the centrosymmetric structure (Suess & Peters, 2013;McWilliams et al., 2018;Gu et al., 2018).No v NN band was detected in the crude material either by IR or Raman spectroscopy.However, no pronounced v CN stretch could be observed either and, interestingly, complex 1 also does not show a characteristic v CN band in IR, contrary to similar complexes (Raba et al., 2012), but signals attributable to axial MeCN are visible in the Raman spectrum (see supporting information).

Conclusion
A diiron end-on �-� 1 :� 1 -CN-bridged complex, 2, was obtained from a crystallization experiment with an open-chain iron NHC complex 1.The cyanide presumably originates from the MeCN solvent by C-C bond cleavage or through carbonhydrogen oxidation.The strongly bent NHC ligands are positioned in an anti conformation.

Special details
Experimental.Diffractometer operator Michael J. Sauer scanspeed 8 s per frame dx 52 mm 2745 frames measured in 9 data sets phi-scans with delta_phi = 0.5 omega-scans with delta_omega = 0.5 shutterless mode Geometry.All esds (except the esd in the dihedral angle between two l.s.planes) are estimated using the full covariance matrix.The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry.An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s.planes.Refinement.Refinement on F 2 for ALL reflections except those flagged by the user for potential systematic errors.Weighted R-factors wR and all goodnesses of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 .The observed criterion of F 2 > 2sigma(F 2 ) is used only for calculating -R-factor-obs etc. and is not relevant to the choice of reflections for refinement.R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.Hydrogen-bond geometry (Å, º)

Table 1
Experimental details.