Imidazol-2-ylidene-Based NCCN Ligands for Chiral-at-Iron Catalysis

Abstract

A new class of achiral tetradentate NCCN ligands containing two imidazol-2-ylidene moieties tethered by a tetraphenylene linker is reported. Upon coordination to iron­(II), these ligands enforce a C ₂-symmetric cis-α coordination geometry, where the helical twist of the tetradentate ligand induces metal-centered chirality. The racemic complexes were resolved into their individual enantiomers using a chiral auxiliary approach. Notably, the tetradentate framework provides significantly enhanced configurational stability compared to previously reported bis-bidentate analogues, showing negligible racemization in the presence of air and water. Finally, the catalytic utility of these chiral-at-iron complexes was demonstrated across three distinct asymmetric transformations, namely an intramolecular Cannizzaro reaction (up to 82% ee), an enantioselective C–H amidation (up to 81% ee), and a hetero-Diels–Alder reaction (up to 94% ee under open air conditions). Collectively, these results demonstrate the versatility of the new tetradentate NCCN architecture in stabilizing iron-based catalysts and facilitating efficient asymmetric induction through purely metal-centered chirality.

Introduction

The inherent versatility of chiral transition metal complexes stems from a vast array of available metals, oxidation states, and coordination environments which renders them indispensable tools for asymmetric catalysis. Central to their success is ligand design, which dictates reactivity, modes of activation and the efficiency of asymmetric induction. Acyclic tetradentate ligands with four sequential donor atoms (often referred as linear tetradentate ligands), typically in combination with two ancillary monodentate ligands, have emerged as a popular choice for generating chiral octahedral catalysts. A key advantage of symmetrical linear tetradentate ligands is their ability to enforce C ₂ symmetry through cis-α coordination, in which the terminal donor atoms are arranged trans to each other, rather than through cis-β or trans coordination (Figure ). This coordination mode effectively minimizes the number of diastereomeric transition states during catalysis. Furthermore, cis-α coordination induces a helical topology with Λ (left-handed helix) or Δ configuration (right-handed helix) which governs the stereochemical outcome of the catalyzed reaction. ,

1.

Octahedral metal complexes with symmetrical linear tetradentate ligands. Previous work and this study: NCCN linear tetradentate ligands with triazol-5-ylidenes or imidazol-2-ylidenes as C-donor ligands adopting a cis-α topology.

In the context of sustainable catalysis with iron, linear tetradentate ligands containing four nitrogen donor sites (N₄ ligands), such as chiral bis­(iminomethyl)­diamines, have been extensively utilized because the chelate effect compensates for the kinetic lability of iron­(II) centers. To further modulate the electronic properties of multidentate ligands for iron, researchers have turned to N-heterocyclic carbenes (NHCs). − Early work by Danopoulos and Hahn established the foundation for iron-NHC pincers and bidentate systems. Subsequent developments by Chen and Kühn explored tetradentate NCCN frameworks. However, these ligands typically resulted in square-planar geometries or, in the case of longer propylene linkers, sawhorse-type cis-β coordination (terminal donors cis to each other). While Glorius and co-workers recently utilized a CNNC ligand to achieve a distorted octahedral iron complex, a C ₂-symmetric cis-α NCCN iron catalyst has remained a synthetic challenge and we were recently the first to introduce such complexes based on mesoionic carbene ligands.

In this work, we introduce a new class of linear NCCN ligands incorporating two imidazol-2-ylidene type NHC donor moieties. We demonstrate that these ligands coordinate to iron in a cis-α fashion to yield configurationally stable, C ₂-symmetric complexes. Significantly, even though the ligand framework itself is achiral, it induces metal-centered chirality, providing chiral-at-iron catalysts which are active in asymmetric transformations as demonstrated for a Cannizzaro reaction, a C–H amidation and a hetero-Diels–Alder reaction.

Results and Discussion

NCCN Ligand Design

We recently reported a novel class of chiral-at-iron catalysts wherein overall chirality originates solely from the stereogenic metal center, utilizing exclusively achiral ligands. − In our initial design (FeNHC, Figure ), the iron­(II) center is cis-coordinated by two bidentate N-(2-pyridyl)-substituted N-heterocyclic carbene (PyNHC) ligands and two labile acetonitrile molecules, with the dicationic complex balanced by hexafluorophosphate counterions. The absolute configuration (Λ or Δ) is determined by the helical twist of the PyNHC ligands. Crucially, while the PyNHC framework remains constitutionally and configurationally inert, the labile acetonitrile sites enable asymmetric transition metal catalysis without requiring chiral ligand precursors. Within the FeNHC architecture, the two mesityl substituents engage in interligand π-stacking, which positions two of their methyl groups in close proximity. Molecular modeling indicated that replacing these methyl groups with a biphenyl moiety would create an effective linker, yielding a new family of tetradentate ligands. To streamline the scaffold, the remaining mesityl methyl substituents were removed, resulting in a structure where two NHC moieties are tethered by a linker composed of four phenyl moieties. This design is analogous to our previous reports on an NCCN ligand containing 1,2,3-triazol-5-ylidene moieties.

2.

Linker design. Previous chiral-at-iron catalyst containing two pyridyl-substituted NHC ligands (FeNHC) (left) and corresponding iron complexes containing an NCCN ligand framework (FeNCCN) (right).

Ligand Synthesis and Iron Coordination

Starting from commercially available 2-bromoaniline (1), the imidazole 2 was synthesized via a Debus-Radziszewski reaction using formaldehyde, glyoxal, and ammonium acetate (91%) (Scheme ). The resulting bromo-phenyl-substituted imidazole 2 was then subjected to a double Suzuki cross-coupling with biphenyldiborane 3, providing the coupling product 4 in 92% yield. The final ligands were then obtained via nucleophilic aromatic substitutions with the corresponding 2-bromo-5-pyridine derivatives 5-7, and subsequent anion exchange furnished the hexafluorophosphate salts 8–10 in 73–94% yields.

1. Ligand Synthesis and Coordination Chemistry to Generate Racemic FeNCCN Complexes .

^a DIPEA = diisopropylethylamine.

The iron complexes were obtained by reacting the tetradentate bis-imidazolium salts with FeCl₂ in the presence of the base diisopropylethylamine (DIPEA) thereby avoiding the use of expensive silver salts. Increasing the reaction temperature to 90 °C reduced the reaction time to 16 h and afforded the iron­(II) complexes rac-FeNCCN1 and rac-FeNCCN2 in isolated yields of 41 and 43%, respectively. A significantly higher yield of 56% was obtained for rac-FeNCCN3, which lacks a substituent on the pyridine ring.

The racemic complexes are air-stable and show no signs of decomposition when stored in acetonitrile under ambient conditions for several weeks (see Supporting Information for details). Importantly, even in the noncoordinating solvent dichloromethane, rac-FeNCCN1 and rac-FeNCCN3 remained intact for up to 1 week, while only rac-FeNCCN2 began to decompose after 2 days. Nevertheless, these FeNCCN complexes are constitutionally more robust than related chiral-at-iron complexes with bis-bidentate ligands.

For structural elucidation, single crystals suitable for X-ray diffraction were obtained. The crystal structure of rac-FeNCCN1 confirms the expected configurational motif, featuring a cis-α coordination geometry of the tetradentate bis-NHC ligand (Figure ). As anticipated, the acetonitrile ligands occupy the equatorial plane together with the NHC moieties, while the pyridyl donors are positioned axially, resulting in a C ₂-symmetric cis-α topology. Compared to our previously reported complex bearing a tetradentate MIC-based ligand, the NHC-based rac-FeNCCN1 exhibits a similar degree of distortion, most notably reflected in the twisting of the tetraphenylene linker.

3.

Crystal structure of the racemic iron complex rac -FeNCCN1. Counter ions and the hydrogen atoms are not shown. Displacement ellipsoids are shown at a 50% probability level at 100 K.

Resolution of Enantiomers

The racemic complexes rac-FeNCCN1-3 were resolved using our established chiral auxiliary method (Scheme ). Accordingly, treatment of the racemic mixtures with the chiral salicyloxazoline derivative (R)-Salox under basic conditions afforded a mixture of diastereomers Λ- and Δ-(R)-FeAux1-3. Subsequent purification via silica gel chromatography provided the individual diastereomers Λ-(R)-FeAux1-3 and Δ-(R)-FeAux1-3 in isolated yields of 32–35% (64–70% of theory). In contrast to previously reported MIC-based NCCN analogues, both diastereomers were each formed in almost equal ratios under identical conditions. This is somewhat surprising, as the Λ-(R)-diastereomers exhibit intramolecular π–π stacking between the oxazoline phenyl substituent and a pyridyl ligand (see the X-ray structures of Λ- and Δ-(R)-FeAux3 in Scheme ), a feature that has been shown to enhance the stability of related auxiliary complexes. − Thus, it can be concluded that the imidazolylidene framework possesses improved stability relative to its recently reported triazolylidene analogue, thereby minimizing the decomposition of the thermodynamically less stable diastereomers during chromatographic purification.

2. Chiral-Auxiliary-Mediated Synthesis of the Nonracemic Iron Complexes Λ- and Δ-FeNCCN1–3 .

^a Crystal structures of Λ- and Δ-(R)-FeAux3: Counter ions and the hydrogen atoms are not shown. Displacement ellipsoids are shown at 50% probability level at 100 K.

It is worth noting that the synthetic protocol for the auxiliary complexes can be streamlined by reacting the crude racemic iron complexes rac-FeNCCN1-3 directly with (R)-Salox after a simple filtration step (Scheme ). This approach results in higher overall yields for the synthesis of Λ- and Δ-(R)-FeAux1-2, although not Λ- and Δ-(R)-FeAux3, and it circumvents a more time-consuming rigorous purification of the racemic iron precursor complexes and thus bypasses one chromatography step.

3. One-Pot Method for the Preparation of the Auxiliary Complexes without Isolation of the Racemic Complexes and Direct Addition of the Chiral Auxiliary .

^a Shown are isolated yields based on the ligand. Note that theoretical yields are a factor of two higher.

The preparation of the enantiopure complexes was completed by the acidic cleavage of the auxiliary ligands, affording the individual enantiomers in excellent yields Λ- and Δ-FeNCCN1 (93% each), Λ- and Δ-FeNCCN2 (97% each), and Λ- and Δ-FeNCCN3 (94 and 93%, respectively). The mirror-image relationship of these complexes was confirmed by circular dichroism (CD) spectroscopy (see Supporting Information).

Configurational Stability of the Chiral-at-Iron Complexes

Next, the configurational stability of the newly developed chiral-at-iron complexes was evaluated by exposing enantiomerically pure samples to air and water (1% v/v) in a noncoordinating solvent (CD₂Cl₂). Racemization was assessed by reacting the mixture with a fluorine-containing chiral bidentate ligand, thereby converting the enantiomers into a set of diastereomers, which were subsequently analyzed by ¹⁹F NMR spectroscopy. This method has been previously established and discussed (see Supporting Information for details). As a result, starting from an initial ee of 97.2% for Δ-FeNCCN1, no evidence of racemization was observed after 24 or 48 h at room temperature (Table , entry 1). Similarly, Δ-FeNCCN2 exhibited no detectable racemization under identical conditions (entry 2). In contrast, the complex Δ-FeNCCN3, which lacks a substituent at the 5-position of the pyridine ring, showed slight racemization of 0.5% after both 24 and 48 h (entry 3). This latter observation is consistent with the stability of related bis­(pyridyl)-NHC complexes, where the presence of electron-withdrawing CF₃ groups on the pyridine moiety enhances the constitutional and configurational robustness of the system. However, overall the new tetradentate complexes Δ-FeNCCN1-3 display markedly enhanced configurational stability compared to the bis-bidentate system. Under identical conditions, Δ-FeNHC underwent substantial racemization (entry 4), further highlighting the beneficial effect of the newly introduced tetraphenylene linker.

1. Configurational Stability of FeNCCN Complexes .

entry	Fe-catalyst	initial ee (%)	ee after 24 h (%)	ee after 48 h (%)
1	Δ-FeNCCN1	97.2	97.1	97.1
2	Δ-FeNCCN2	97.1	97.0	97.0
3	Δ-FeNCCN3	99.1	98.6	98.6
4	Δ-FeNHC	98.5	86.0	58.0

^aRacemization study: The single enantiomers of the iron complexes were exposed to air for various lengths of time in CD₂Cl₂ with the addition of H₂O (1% v/v). After this incubation time, a chiral F-containing ligand was coordinated and analyzed by ¹⁹F-NMR.

Asymmetric Catalysis

Following the successful synthesis of the nonracemic complexes, their catalytic performance was evaluated. Initially, the new chiral-at-iron complexes were tested in the intramolecular Cannizzaro reaction of phenylglyoxal monohydrate (11) to mandelate ester (12). As summarized in Table , Δ-FeNCCN2 and Δ-FeNCCN3 afforded only moderate yields (73%) and enantioselectivities (45% ee) at room temperature (entries 3 and 5). Cooling the reaction to 4 °C for these catalysts resulted in a decreased yield (64%) with only marginal improvements in enantioselectivity (entries 4 and 6). The best results were obtained with Δ-FeNCCN1 (R = CF₃), which provided an 84% NMR yield and 78% ee at room temperature (entry 1). A similar trend was observed at 4 °C for this complex, where the ee increased to 82% at the expense of the yield (72%, entry 2).

2. Intramolecular Cannizzaro Reaction .

entry	Fe-catalyst	T	NMR-yield (%)	ee (%)
1	Δ-FeNCCN1	rt	84	78
2	Δ-FeNCCN1	4 °C	72	82
3	Δ-FeNCCN2	rt	73	45
4	Δ-FeNCCN2	4 °C	64	45
5	Δ-FeNCCN3	rt	71	41
6	Δ-FeNCCN3	4 °C	64	44

^aReaction conditions: Substrate 11 (0.05 mmol), Δ-FeNCCN1–3 (5 mol %), and 3 Å MS were placed in CH₂Cl₂ (0.05 M) under N₂. Then iPrOH was added, and stirring was carried out for the indicated temperature.

^bDetermined by ¹H NMR analysis of the crude product with 1,3,5-trimethoxybenzene as internal standard.

^cEnantiomeric excess of the crude product determined by HPLC analysis on a chiral stationary phase.

The complexes Δ-FeNCCN1-3 were subsequently evaluated in the nitrene-mediated ring-closing C­(sp³)–H amidation of benzoyloxyurea 13 into 2-imidazolidinone 14 (Table ). Accordingly, at 4 °C, Δ-FeNCCN1 provided 80% conversion (73% yield) with 71% ee (entry 1), while Δ-FeNCCN2 and Δ-FeNCCN3 achieved nearly quantitative conversion of the standard substrate (98 and 90% yield, respectively; entries 2 and 3). Notably, Δ-FeNCCN3, bearing an unsubstituted pyridyl moiety, exhibited no asymmetric induction (1% ee), highlighting the critical role of the pyridine substituent in providing enantioselectivity. Thus, FeNCCN1 emerged as the catalyst of choice for this ring-closing amination. However, further lowering the temperature of the Δ-FeNCCN1 system to −15 °C resulted in only a modest improvement in ee to 81%, while substantially decreasing both conversion and yield to 36% (entry 4). For Δ-FeNCCN2, the yield decreased to 88% at this temperature while the ee remained unchanged at 61% (entry 5). The lack of improvement in enantioselectivity prompted an investigation into substrate scope using the most efficient catalyst, Δ-FeNCCN2. Incorporating a bulky tBu group into the leaving group yielded similar results (91% yield, 60% ee; entry 6), suggesting the leaving group plays a minor role in the enantiodetermining step. Consequently, we examined steric effects on the phenyl ring via methyl substitution. The para-substituted substrate showed a slight improvement in enantioselectivity (74% ee, 85% yield; entry 7), whereas the meta- and ortho-derivatives provided 59 and 66% ee, respectively (entries 8 and 9). Interestingly, the lower yield observed for the para-substituted substrate contrasts with our previously reported findings.

3. C­(sp³)–H-Amidation of Benzoyloxyurea Derivatives to 2-Imidazolidinones .

entry	Fe-catalyst	substr.	T	conversion (%)	NMR-yield (%)	ee (%)
1	Δ-FeNCCN1	13a	4 °C	80	73	71 (R)
2	Δ-FeNCCN2	13a	4 °C	100	98	61 (R)
3	Δ-FeNCCN3	13a	4 °C	100	90	1 (R)
4	Δ-FeNCCN1	13a	–15 °C	36	36	81 (R)
5	Δ-FeNCCN2	13a	–15 °C	100	88	61 (R)
6	Δ-FeNCCN2	13b	4 °C	100	91	60 (R)
7	Δ-FeNCCN2	13c	4 °C	100	85	74 (R)
8	Δ-FeNCCN2	13d	4 °C	100	91	59 (R)
9	Δ-FeNCCN2	13e	4 °C	100	87	66 (R)
10	Λ-FeNCCN1	13a	4 °C	80	74	71 (S)

^aReaction conditions: Substrate 13a–e (0.03 mmol), Δ-FeNCCN1-3 (5 mol %), and K₂CO₃ (0.10 mmol) were placed in CH₂Cl₂ (0.1 M) under N₂. Then, stirring was carried out at the indicated temperature.

^bDetermined by ¹H NMR analysis of the crude product with 1,3,5-trimethoxybenzene as internal standard.

^cEnantiomeric excess of the crude product determined by HPLC analysis on a chiral stationary phase.

While all of these catalytic reactions were carried out using iron catalysts with a metal-centered Δ-configuration, employing a corresponding Λ-configured iron complex instead afforded the expected enantiomeric product (compare entry 1 with entry 10).

To further evaluate their Lewis-acid catalytic activity, Δ-FeNCCN1-3 were applied to the hetero-Diels–Alder (HDA) reaction of β,γ-unsaturated α-ketoester 15 with 2,3-dihydrofuran 16 to afford bicyclic dihydropyran 17 (Table ). ,, Under standard inert conditions, Δ-FeNCCN1–3 provided high yields and enantioselectivities while maintaining excellent diastereocontrol. Modest differences were observed in enantioselectivity, with Δ-FeNCCN1 yielding 86% ee (entry 1), compared to 94% ee for Δ-FeNCCN2 (entry 2) and 92% ee for Δ-FeNCCN3 (entry 3). To assess the robustness of the tetradentate NCCN-ligand architecture, the reactions were repeated under “open-flask” conditions in the presence of air and water (1% v/v). This resulted in only marginal decreases in yield (94–97%, entries 4–6), while both enantio- and diastereoselectivity remained virtually unchanged. These findings highlight the superior configurational stability of these NCCN-based complexes relative to their bis-bidentate analogs, demonstrating their utility for asymmetric catalysis in noninert environments.

4. Inverse Electron Demand Hetero-Diels-Alder Reaction .

entry	Fe-catalyst	conditions	NMR-yield (%)	dr	ee (%)
1	Δ-FeNCCN1	N2 + dry	99	99:1	86
2	Δ-FeNCCN2	N2 + dry	98	99:1	94
3	Δ-FeNCCN3	N2 + dry	98	99:1	92
4	Δ-FeNCCN1	air + 0.1% H2O	97	99:1	85
5	Δ-FeNCCN2	air + 0.1% H2O	94	99:1	94
6	Δ-FeNCCN3	air + 0.1% H2O	94	99:1	93

^aReaction conditions: Ketoester 15 (0.05 mmol), Δ-FeNCCN1–3 (3 mol %), and dihydrofuran 16 (0.80 mmol) were placed in distilled CH₂Cl₂ (0.05 M) under N₂. Then, stirring was carried out at room temperature for 24 h.

^bDetermined by ¹H NMR analysis of the crude product with 1,3,5-trimethoxybenzene as internal standard.

^cDiastereomeric ratios were determined by ¹H NMR.

^dEnantiomeric excess of the crude product determined by HPLC analysis on a chiral stationary phase.

Conclusions

In summary, we synthesized a series of achiral tetradentate NCCN ligands containing two imidazol-2-ylidene moieties that enforce C ₂-symmetric cis-α coordination to provide chiral-at-iron complexes. By utilizing a tetraphenylene linker, these systems achieve superior configurational stability over bis-bidentate analogues, resisting racemization even when exposed to air and moisture. The catalysts demonstrate promising performance across diverse asymmetric transformations, including an intramolecular Cannizzaro reaction (up to 82% ee), C­(sp³)–H amidation (up to 71% ee), and hetero-Diels–Alder (up to 94% ee). Notably, the high efficiency of the hetero-Diels–Alder reaction maintained under open-flask conditions underscores the practical utility of this catalyst scaffold. This work complements our recent work on NCCN ligands containing two 1,2,3-triazol-5-ylidenes and establishes the tetradentate NCCN scaffold as a powerful ligand toolbox for designing configurationally inert chiral-at-iron catalysts. Ongoing efforts in our laboratory are directed toward extending this ligand design to other earth-abundant metals.

Experimental Section

General Methods and Materials

All reactions were carried out under a nitrogen atmosphere in oven-dried glassware unless noted otherwise. Solvents for sensitive reactions were dried according to standard purification methods using calcium hydride (MeCN, CH₂Cl₂) and distilled under an atmosphere of nitrogen. The chemicals used are all from commercial sources and were used without further purification unless stated otherwise. For purification by column chromatography, Macherey-Nagel silica gel 60 M (irregularly shaped, 230–400 mesh, pH 6.8, pore volume: 0.81 mL/g, mean pore size: 66 Å, specific surface: 492 m²/g, particle size distribution: 0.5% < 25 μm and 1.7% > 71 μm, water content: 1.6%) was used as the stationary phase. ¹H NMR, ¹³C­{¹H} NMR and ¹⁹F­{¹H} NMR spectra were recorded on a Bruker AV III HD 300 MHz, AV II 300 MHz, AV III HD 500 MHz, AV III 500 MHz, NEO 300 MHz, or NEO 600 MHz spectrometer at ambient temperature. All ¹³C and ¹⁹F NMR spectra were recorded as proton-decoupled spectra. The chemical shift δ is reported in parts per million (ppm) with the residual proton signal of the deuterated solvent as reference. ¹⁹F­{¹H} NMR spectra were calibrated to trichlorofluoromethane (CFCl₃, δ = 0 ppm) as external standard. All infrared spectroscopy was performed on a Bruker Alpha FT–IR spectrometer. Chiral HPLC was performed on an AGILENT 1200 or AGILENT 1260 HPLC system with DAICEL columns as chiral stationary phase with a size of 4.6 × 250 mm and a particle size of 5 μm. CD-spectra were recorded on a Jasco J-810 CD spectropolarimeter from 600–200 nm, 1 nm bandwidth, 2 s response time, 50 nm/min scanning speed, and an accumulation of 3 scans. High-resolution mass spectrometry was performed on a Finnigan LTQ-FT Ultra mass spectrometer (Thermo Fischer Scientific) using Electrospray ionization (ESI) as ionization source. Melting points (MPs) were determined on a Mettler Toledo MP70 using one end closed capillary tubes.

The following experimental section presents representative synthetic procedures for the ligand and the complex bearing a CF₃-group at the 5-position of the pyridine ring. Syntheses and characterization of the related ligands and complexes are described within the Supporting Information.

General Procedure for the Synthesis of the Tetradentate Ligands

The bisimidazole 4 (1.00 equiv) (see Supporting Information for the synthesis of 4) was mixed with the 5-substituted 2-bromopyridine 5–7 (2.10 equiv) and placed in a Schlenk tube under an inert atmosphere. The tube was sealed and heated at 170 °C for 48 h in an oil bath. After cooling to room temperature, the residue was transferred with CH₂Cl₂ into a round-bottom flask, and the solvent was removed under reduced pressure to afford the bromide salt. The bromide salt was then dissolved together with NH₄PF₆ (5.00 equiv) in MeCN (0.16 M based on the imidazole) and stirred for 1.5 h under air. Afterward, the solvent was removed under reduced pressure, and the residue was taken up in CH₂Cl₂ to precipitate the ammonium salts and filtered again over Celite. The resulting filtrate of the crude hexafluorophosphate salts was dried under reduced pressure and purified by flash column chromatography (silica gel, CH₂Cl₂/MeOH, 98:2 → 90:10 → 80:20) to afford the corresponding bisimidazolium salts 8–10 as orange-brown solids.

Ligand 8

Following the general procedure, the bisimidazolium salt 8 (1.80 g, 1.77 mmol, 94%) was obtained as a light brown solid starting from the precursor 4 (0.81 g, 1.86 mmol). ¹H NMR: (500 MHz, CD₃CN) δ (ppm) = 9.57 (t, J = 1.7 Hz, 1H), 8.91 (dt, J = 2.6, 0.9 Hz, 1H), 8.48–8.43 (m, 1H), 8.16–8.12 (m, 1H), 7.95–7.90 (m, 1H), 7.83–7.71 (m, 4H), 7.65–7.55 (m, 2H), 7.52 (dd, J = 2.2, 1.7 Hz, 1H), 7.44 (t, J = 7.7 Hz, 1H), 7.25 (ddd, J = 7.7, 1.8, 1.1 Hz, 1H). ¹³C NMR: (126 MHz, CD₃CN) δ (ppm) = 149.5, 147.8, 147.7, 141.8, 139.3, 139.2, 138.3, 137.7, 136.2, 133.2, 132.8, 132.7, 132.7, 130.5, 130.5, 128.9, 128.6, 128.5, 128.3, 128.2, 127.5, 126.6, 120.4, 115.7. ¹⁹F-NMR: (282 MHz, CD₃CN) δ (ppm) = −62.94 (s, 3F), −72.82 (d, ¹ J _PF = 706.5 Hz, 6F). HRMS ESI­(+); m/z calculated for C₄₂H₂₈F₁₂N₆P [M–PF₆]⁺: 875.19, found: 875.1910 [M–PF₆]⁺. IR: ν̃ (cm^–1) = 3159 (w), 1604 (w), 1542 (w), 1489 (w), 1412 (w), 1328 (m), 1261 (w), 1231 (w), 1173 (w), 1134 (m), 1078 (w), 1064 (w), 1018 (w), 970 (w), 956 (w), 821 (s), 760 (w), 740 (w), 708 (w), 677 (w), 657 (w), 614 (w), 555 (s), 514 (w), 490 (w), 468 (w), 427 (w).

General Procedure for the Synthesis of Racemic Iron Complexes

Following a modified literature procedure, the bisimidazolium salts 8, 9, or 10 (1.00 equiv) (see Supporting Information for the synthesis of 8–10), FeCl₂ (1.00 equiv) and molecular sieve (3 Å, 1 g per 1.0 mmol ligand) were suspended in dry and degassed MeCN (0.025 M based on the imidazolium salt) under an atmosphere of nitrogen. After stirring for 5 min, DIPEA (2.50 equiv) was added and the mixture was heated to 90 °C for 16 h. The red-colored solution was allowed to cool to room temperature, diluted with MeCN, filtered over Celite, and the solvent was removed under reduced pressure. The residue was purified by flash column chromatography (silica gel, CH₂Cl₂/MeCN) with a NH₄PF₆ pad on top of the column to ensure complete elution of the desired complex. After removing the solvent under reduced pressure, the crude product was dissolved in CH₂Cl₂/MeCN (10:1) and washed three times with H₂O to remove excess of PF₆ ^–-salts. The combined organic phases were then dried over MgSO₄, filtered, and the solvent was removed under reduced pressure to afford the desired complexes rac-FeNCCN1–3.

rac-FeNCCN1

Following the general procedure, rac-FeNCCN1 (188 mg, 0.16 mmol, 40%) was obtained as a red solid starting from ligand 8 (400 mg, 0.39 mmol). Purification condition: CH₂Cl₂/MeCN = 10:1 → 4:1. ¹H NMR: (500 MHz, CD₂Cl₂) δ (ppm) = 8.88 (d, J = 2.0 Hz, 1H), 7.88 (dd, J = 8.7, 2.1 Hz, 1H), 7.70 7.65 (m, 2H), 7.59 (t, J = 7.7 Hz, 1H), 7.52 (dd, J = 7.7, 1.4 Hz, 1H), 7.40 (d, J = 1.4 Hz, 1H), 7.34–7.27 (m, 3H), 7.23 (dd, J = 7.8, 1.4 Hz, 1H), 6.29 (d, J = 2.3 Hz, 1H), 6.20 (d, J = 1.9 Hz, 1H), 2.58 (s, 3H). ¹³C NMR: (126 MHz, CD₂Cl₂) δ (ppm) = 196.9, 157.0, 151.8, 139.3, 137.7, 136.2, 134.7, 132.9, 130.6, 130.4, 129.8, 129.2, 128.8, 128.0, 126.8, 125.8, 118.2, 110.8, 4.5. ¹⁹F-NMR: (282 MHz, CD₂Cl₂) δ (ppm) = −62.18 (s, 3 F), −72.81 (d, ¹ J _PF = 711.0 Hz, 6 F). HRMS: ESI­(+); m/z calculated for C₄₂H₂₆F₆N₈FeF [M–(PF₆)₂ – (MeCN)₂ + F]⁺: 803.14, found: 803.1442 [M–(PF₆)₂ – (MeCN)₂ + F]⁺. IR: ν̃ (cm^–1) = 3149 (w), 2920 (w), 2851 (w), 1621 (w), 1508 (w), 1474 (w), 1425 (w), 1326 (m), 1306 (w), 1270 (w), 1253 (w), 1175 (w), 1135 (w), 1074 (m), 1042 (w), 947 (w), 824 (s), 772 (w), 763 (w), 748 (w), 705 (w), 680 (w), 665 (w), 620 (w), 600 (w), 555 (s), 508 (w), 469 (w), 435 (w).

General Procedure for the Synthesis of Auxiliary Complexes

The racemic iron complexes (1.00 equiv) and the chiral auxiliary (R)-Salox (1.05 equiv) were dissolved in dry CH₂Cl₂ (0.04 M based on the iron complex) under an atmosphere of nitrogen. Et₃N (1.50 equiv) was added, and the reaction mixture was stirred at 40 °C for 16 h. The solvent was then removed under reduced pressure, and the residue was purified by flash column chromatography (silica gel, CH₂Cl₂/MeCN, 100:1 → 75:1 → 50:1 → 25:1) to obtain both diastereomers of the auxiliary complexes.

Λ-(R)-FeAux1 and Δ-(R)-FeAux1. Following the general procedure, Λ-(R)-FeAux1 (25.4 mg, 21.4 μmol, 33%) was obtained as a purple solid from the corresponding complex rac-FeNCCN1 (75.0 mg, 65 μmol). The other diastereomer Δ-(R)-FeAux1 (26.4 mg, 22.3 μmol, 34%) was obtained as a purple solid.

Analytical Data for Λ-(R)-FeAux1

¹H NMR: (600 MHz, CD₂Cl₂) δ (ppm)= 8.63 (s, 1H), 8.42 (s, 1H), 7.98 (d, J = 7.8 Hz, 1H), 7.71 (d, J = 8.6 Hz, 1H), 7.66 (d, J = 7.9 Hz, 1H), 7.62–7.50 (m, 5H), 7.41–7.20 (m, 10H), 7.18–7.08 (m, 3H), 6.93 (t, J = 7.5 Hz, 1H), 6.81 (s, 2H), 6.69 (d, J = 8.6 Hz, 1H), 6.37 (d, J = 8.5 Hz, 1H), 6.31 (d, J = 8.8 Hz, 3H), 6.27 (s, 1H), 6.21 (s, 1H), 6.15 (t, J = 10.1 Hz, 1H), 5.99 (s, 1H), 5.08–5.02 (m, 1H), 4.45 (dd, J = 9.4, 3.5 Hz, 1H). ¹³C NMR: (151 MHz, CD₂Cl₂) δ (ppm)= 159.1, 157.2, 151.7, 151.7, 151.4, 140.2, 139.9, 139.7, 137.3, 136.9, 136.7, 135.9, 135.8, 135.6, 135.0, 133.7, 131.5, 130.9, 130.8, 130.6, 130.5, 130.1, 129.9, 129.8, 129.5, 129.2, 129.0, 128.9, 128.9, 128.8, 127.6, 127.4, 126.6, 126.0, 126.0, 125.8, 124.1, 123.9, 123.9, 123.4, 123.1, 122.1, 122.0, 117.7, 117.4, 110.1, 109.1, 76.1, 73.8. ¹⁹F-NMR: (282 MHz, CD₂Cl₂) δ (ppm) = −62.31 (s, 3F), – 62.47 (s, 3F), −73.07 (d, ¹ J _PF = 710.9 Hz, 6F), −106.07 (s, 1F). HRMS: ESI­(+); m/z calculated for C₅₇H₃₇F₇FeN₇O₂ [M–(PF₆)]⁺: 1040.22, found: 1040.2252 [M–(PF₆)]⁺. IR: ν̃ (cm^–1) = 3144 (w), 3085 (w), 1615 (m), 1577 (m), 1531 (w), 1505 (m), 1474 (w), 1447 (m), 1420 (w), 1387 (w), 1326 (s), 1312 (w), 1301 (w), 1256 (w), 1244 (w), 1233 (w), 1172 (w), 1133 (m), 1099 (w), 1072 (w), 1042 (w), 1004 (w), 991 (w), 954 (w), 944 (m), 924 (w), 832 (s), 795 (w), 771 (w), 762 (w), 747 (m), 702 (m), 677 (w), 661 (w), 620 (w), 600 (w), 582 (w), 557 (s), 534 (w), 510 (w), 462 (w).

Analytical Data for Δ-(R)-FeAux1

¹H NMR: (500 MHz, CD₂Cl₂) δ (ppm) = 8.84 (s, 1H), 8.70 (s, 1H), 8.08 (d, J = 7.7 Hz, 1H), 7.90 (dd, J = 8.6, 2.1 Hz, 1H), 7.85 (d, J = 2.4 Hz, 1H), 7.68–7.59 (m, 3H), 7.59–7.51 (m, 5H), 7.52–7.44 (m, 2H), 7.37 (dtd, J = 7.6, 3.7, 1.4 Hz, 2H), 7.32–7.23 (m, 2H), 7.29–7.24 (m, 4H), 7.19–7.16 (m, 1H), 7.03 (td, J = 7.6, 1.2 Hz, 1H), 6.92 (dt, J = 18.0, 8.5 Hz, 2H), 6.41 (d, J = 1.9 Hz, 1H), 6.33 (td, J = 7.8, 1.6 Hz, 2H), 6.24 (d, J = 2.2 Hz, 1H), 6.18 (t, J = 9.6 Hz, 1H), 6.13 (d, J = 2.3 Hz, 1H), 5.97 (d, J = 1.9 Hz, 1H), 4.64 (dd, J = 9.3, 3.4 Hz, 1H), 4.46 (d, J = 7.7 Hz, 1H), 4.13 (t, J = 9.4 Hz, 1H), 3.46 (d, J = 9.4 Hz, 1H).

¹³C NMR: (126 MHz, CD₂Cl₂) δ (ppm) = 165.8, 163.2, 161.2, 158.7, 157.8, 153.4, 153.3, 151.5, 141.7, 139.9, 139.8, 137.5, 136.8, 136.1, 135.8, 135.7, 135.6, 135.6, 135.5, 135.1, 135.0, 133.4, 133.3, 131.4, 130.7, 130.6, 130.5, 130.3, 130.2, 130.0, 129.9, 129.5, 128.9, 128.8, 128.6, 128.5, 128.1, 128.0, 127.8, 126.2, 125.9, 125.5, 124.6, 124.3, 124.1, 123.2, 123.0, 121.9, 117.5, 117.5, 109.6, 109.0, 100.9, 100.7, 77.1, 69.1. ¹⁹F-NMR: (282 MHz, CD₂Cl₂) δ (ppm) = −61.97 (s, 3F), −62.48 (s, 3F), – 72.90 (d, ¹ J _PF = 711.0 Hz, 6F), −110.35 (s, 1F). HRMS: ESI­(+); m/z calculated for C₅₇H₃₇F₇FeN₇O₂ [M – (PF₆)]⁺: 1040.22, found: 1040.2217 [M–(PF₆)]⁺. IR: ν̃ (cm^–1) = 3144 (w), 3077 (w), 1619 (m), 1588 (w), 1536 (w), 1504 (m), 1475 (w), 1447 (m), 1419 (m), 1386 (w), 1356 (w), 1326 (s), 1311 (w), 1283 (w), 1262 (w), 1233 (m), 1171 (w), 1136 (m), 1099 (w), 1085 (w), 1072 (w), 1038 (m), 1006 (w), 992 (w), 944 (m), 913 (w), 838 (s), 796 (w), 772 (w), 765 (m), 704 (m), 676 (w), 663 (w), 621 (w), 599 (w), 586 (w), 557 (s), 533 (w), 512 (w), 471 (w).

General Procedure for the Auxiliary Cleavage

Following a modified procedure from the literature. The single diastereomers of the auxiliary complexes (1.00 equiv) were dissolved in dry MeCN (0.02 M based on the auxiliary complex), and an excess of NH₄PF₆ (10.00 equiv) was added, and the mixture was stirred at 40 °C for 16 h. Afterward, the solvent was removed under reduced pressure and Et₂O was added. The precipitated solids were transferred to a Celite pad and washed several times with Et₂O to remove any residual free auxiliary. The complex was then eluted with CH₂Cl₂/MeCN, 30:1 to afford the corresponding Λ- or Δ-iron complexes after the solvent had been removed under reduced pressure.

Λ-FeNCCN1

Following the general procedure, Λ-FeNCCN1 (55 mg, 47.6 μmol, 94%) was obtained as a red solid from the corresponding complex Λ-(R)-FeAux1 (60 mg, 51.0 μmol). The spectroscopic data of the enantiopure complex were in accordance with rac-FeNCCN1. CD (MeCN, 0.25 mM): Δ, nm (Δε, M^–1 cm^–1) 234 (+98), 259 (−99), 282 (+60), 302 (−13), 330 (+4), 355 (+2), 390 (+16), 442 (−23), 517 (+4).

Δ-FeNCCN1

Following the general procedure, Δ-FeNCCN1 (66 mg, 57.1 μmol, 94%) was obtained as a red solid from the corresponding complex Δ-(R)-FeAux1 (72 mg, 61.0 μmol). The spectroscopic data of the enantiopure complex were in accordance with rac-FeNCCN1. CD (MeCN, 0.25 mM): Λ, nm (Δε, M^–1 cm^–1) 234 (−106), 259 (+119), 282 (−71), 302 (+16), 330 (−5), 355 (−3), 390 (−16), 442 (+25), 517 (−4).

General Procedure for the Intramolecular Cannizzaro Reaction

A modified literature procedure was followed. The iron catalyst (5 mol %), molecular sieve (3 Å, 5 mg per 0.01 mmol substrate) and phenylglyoxal (11) (7.6 mg, 0.05 mmol, 1.00 equiv) were suspended in dry and degassed CH₂Cl₂ (0.05 M) under an atmosphere of nitrogen in a Schlenk tube. Then, iPrOH (37.5 μL, 0.50 mmol, 10.0 equiv) was added and the mixture was stirred for 16 h at the indicated temperature. Subsequently, n-pentane (5 mL) was added, and the reaction mixture was filtered over a short Celite plug to remove the catalyst and rinsed with n-pentane/CH₂Cl₂ (4:1). The solvent was removed under reduced pressure and the yield was determined by ¹H NMR analysis of the crude product with TMB as internal standard. The enantiomeric excess of the crude product was determined by HPLC analysis on a chiral stationary phase. The spectroscopic data are in accordance with the literature.¹⁵ The absolute configuration of the product was determined by comparison of the HPLC traces with those reported in the literature. TLC: R _f = 0.41 (n-pentane/EtOAc 3:1). ¹H NMR: (300 MHz, CDCl₃) δ (ppm) = 7.44–7.30 (m, 5H), 5.14–5.01 (m, 2H), 3.47 (d, J = 5.8 Hz, 1H), 1.28 (d, J = 6.3 Hz, 3H), 1.11 (d, J = 6.2 Hz, 3H) HPLC: Daicel Chiralcel OD-H column, 250 × 4.6 mm, absorbance at 210 nm, n-hexane/iPrOH 95:5, isocratic flow, flow rate 1.0 mL/min, 25 °C, t _r (minor) = 7.67 min, t _r (major) = 14.72 min.

General Procedure for the C­(sp³)–H-Amidation of Benzoyloxyurea Derivatives

Following a slightly modified procedure from the literature. The iron catalyst (5 mol %), K₂CO₃ (0.10 mmol, 3.00 equiv) and the urea substrate 13a–e (0.03 mmol, 1.00 equiv) were placed in a Schlenk tube under an atmosphere of nitrogen. Dry and degassed CH₂Cl₂ (0.1 M) was added, and the reaction mixture was stirred for 24 h at the indicated temperature. According to that, n-pentane (5 mL) was added, and the reaction mixture was filtered over a short Celite plug to remove the catalyst and rinsed with n-pentane/CH₂Cl₂ (4:1). The solvent was removed under reduced pressure and the yield was determined by ¹H NMR analysis of the crude product with TMB as internal standard. The enantiomeric excess of the crude product was determined by HPLC analysis on a chiral stationary phase. The spectroscopic data are in accordance with the literature. The absolute configuration of the product was determined by comparison of the HPLC traces with those reported in the literature.

14a. TLC: R _f = 0.17 (n-pentane/EtOAc 2:1). ¹H NMR: (300 MHz, CDCl₃) δ (ppm) = 7.38–7.28 (m, 5H), 5.41 (s, 1H), 4.76 (t, J = 8.1 Hz, 1H), 3.79 (t, 1H), 3.24 (t, J = 8.0 Hz, 1H), 2.83 (s, 3H). HPLC: Daicel Chiralcel IA column, 250 × 4.6 mm, absorbance at 220 nm, n-hexane/iPrOH 95:5, isocratic flow, flow rate 1.0 mL/min, 30 °C, t _r (minor) = 28.60 min, t _r (major) = 22.86 min.

General Procedure for the Hetero-Diels–Alder Reaction

Following a slightly modified procedure from the literature. The iron catalyst (3 mol %) and ketoester 15 (0.05 mmol, 1.00 equiv) were dissolved in dry and degassed CH₂Cl₂ (0.05 M) under an atmosphere of nitrogen in a Schlenk tube. Then, dihydrofuran 16 (0.80 mmol, 15.0 equiv) was added, and the mixture was stirred at room temperature for 24 h. According to that, n-pentane (5 mL) was added, and the reaction mixture was filtered over a short Celite plug to remove the catalyst and rinsed with n-pentane/CH₂Cl₂ (4:1). The solvent was removed under reduced pressure and the yield was determined by ¹H NMR analysis of the crude product with TMB as internal standard. The enantiomeric excess of the crude product was determined by HPLC analysis on a chiral stationary phase. The spectroscopic data are in accordance with the literature.¹⁶ The absolute configuration of the product was determined by comparison of the HPLC traces with those reported in the literature. TLC: R _f = 0.32 (n-pentane/EtOAc 5:1). ¹H NMR: (300 MHz, CDCl₃) δ (ppm) = 7.37–7.20 (m, 5H), 6.20 (dd, J = 1.3 Hz, 1H), 5.64 (d, J = 3.5 Hz, 1H), 4.20–4.13 (m, 2H), 3.90–3.86 (m, 1H), 3.83 (s, 3H), 2.71–2.59 (m, 1H), 1.71 (tt, J = 12.3 Hz, 1H), 1.37–1.31 (m, 1H). HPLC: Daicel Chiralcel IG column, 250 × 4.6 mm, absorbance at 254 nm, n-hexane/iPrOH 95:5, isocratic flow, flow rate 1.0 mL/min, 25 °C, t _r (minor) = 32.25 min, t _r (major) = 38.36 min.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.organomet.6c00132.

• Detailed experimental procedures, analytical data, NMR spectra, CD spectra, HPLC traces, and crystallographic data (PDF)

The authors declare no competing financial interest.