The Selective Monobromination of a Highly Sterically Encumbered Corrole: Structural and Spectroscopic Properties of Fe(Cl)(2-Bromo-5,10,15-tris(triphenyl)phenyl corrole)

Abstract

The corrole ligand serves as a versatile tri-anionic, macrocyclic platform on which to model biological catalytic systems, as well as to effect mechanistically challenging chemical transformations. Here in we describe the synthesis, structure, and characterization of an isomerically pure corrole ligand, selectively mono-brominated at the β-carbon position adjacent to the corrole C-C bond (2-C) and produced in relatively high yields, as well as its iron chloride complex. Analysis of the iron metalated complex by cyclic voltammetry shows that the bromine being present on the ligand resulted in anodic shifts of +93 and +63 mV for first oxidation and first reduction of the complex respectively. The Mossbauer spectrum of the iron metalated complex shows negligible change relative to the non-brominated analog, indicating the presence of the halide substituent predominantly effects the orbitals of the ligand rather than the metal.

Graphical Abstract

INTRODUCTION

Halogenation of macrocycles has been a focus of continued interest for several decades, particularly for both porphyrins and corroles, with examples in both cases of perbrominated and perchlorinated compounds.^1-8 One area of interest concerns the effect of β-carbon halogenation on the redox potentials of corroles.^{2, 9-12} Compared to porphyrins, data published thus far indicates that corroles display a greater shift in their redox potentials when perbrominated.¹³ Another reason for interest in β-brominated porphyrins and corroles is their role as effective electrophilic coupling partners in cross-coupling reactions, i.e. partially brominated corroles may be used to form carbon-carbon bonds at select β-positions.^14-16 The practical synthesis of a site-specific brominated corrole could provide a useful synthon for preparing a diverse array of corrole derivatives. A selectively brominated corrole would also allow for the examination of the effects of site-specific halogenation on the physical, spectroscopic, electronic, and electrochemical characteristics of corroles and their metalated analogs.

Paolesse and co-workers synthesized a series of polybrominated corroles in 2007 by addition of Br₂ saturated CHCl₃ to a germanium 5,10,15-triphenyl-corrole complex. They were able to isolate and characterize two new germanium corrole species, which exhibited bromination at various β-carbon positions around the corrole ring without site specificity.¹⁷ Several more partial brominations of corrole systems have been reported, such as Chen and co-workers partially-brominated 5,10,15-tris(pentafluorophenyl)corrole complex,¹⁸ 2,3-dibromo-10-(4-methoxycarbonylphenyl)-5,15-bis(pentafluorophenyl)corrole reported by Nocera,⁴ and silver metalated 3,17-dibromo-5,10,15-tris(4-tert-butylphenyl)corrole from Paoloesse.¹⁶ All of these compounds were synthesized using the radical brominating agent N-bromosuccinimide (NBS). Another more recent example comes from Osuka and coworkers, who prepared a partially brominated corrole by first synthesizing a dibrominated bilane compound, and then oxidizing it to form the corrole C-C bond (8-12-dibromo-5,10-bis(pentafluorophenyl)corrole).¹⁹ None of these examples include an isolated and isomerically pure monobrominated corrole,¹⁸ though in 2010 Nardis and Paolesse reported the synthesis and isolation of two such monobrominated corrole isomers; 2-bromo-5,10,15-tritolylcorrole and 3-bromo-5,10,15-tritolylcorrole.¹³ These compounds were obtained as side products from the reaction of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) and CH₃CH₂MgBr with 5,10,15-tritolylcorrole, during the synthesis of meso-alkyl substituted isocorroles. Although these monobrominated isomers were successfully separated, their combined yields were less than 10%. More recently, Paolesse and co-workers improved the synthesis of these monobrominated corroles by adding HBr in acetic acid to the nitrocorrole dissolved in toluene. These reactions produced both di-brominated and mono-brominated corroles, however their combined yields remained lower than 40%.²⁰

A source of difficulty in preparing and isolating monobrominated corroles is their lower degree of symmetry relative to porphyrins, having only a single C2 rotation axis and mirror plane. This lower symmetry, while still having eight β-carbons that are potential nucleophilic sites, can lead to 4 potential isomers relative to 1 potential isomer for porphyrins,²¹ making it difficult to synthesize an isomerically pure monobrominated corrole. Much of the previous work involved halogenation of the metal-free corroles, but there are a few descriptions of the corresponding metalated species, with the iron chloride 2-bromo-tritolyl and 3-bromo-tritolyl corroles having been reported but not structurally charecterized.²² The UV-vis spectra for these complexes display a red-shift in their Soret bands in both the metal-free corrole ligands as well as for the metalated monobrominated tritolylcorrole system.^22-23

Our group previously published a series of papers employing the sterically encumbered tris(triphenyl)phenyl corrole (ttppc) as a platform for modeling high-valent intermediates in heme oxidation chemistry, including the protonated Compound-II intermediate in Cytochrome P450.^24-26 The metal-free corrole ttppcH₃ was originally synthesized by Chang and coworkers,²⁷ and it was later demonstrated that the β-carbon position of the pyrrole rings of the ligand would undergo a slow exchange with DCM solvent to afford a monochlorinated 2-Cl-ttppcH₃.²⁸ Inspired by this result we used methods previously described by Paolesse and others to produce the brominated version of the ttppc ligand in practical amounts.^{18, 29} Herein we demonstrate a site-selective, reasonably high yielding synthesis, purification, spectroscopic and structural characterization of the monobrominated 2-Br-ttppcH₃, as well as its associated iron chloride complex. Their spectroscopic and electrochemical features are compared to their unmodified ttppc analogs, providing information on the effect of a single halogen substituent at a specific β-carbon position.

RESULTS AND DISCUSSION

Synthesis of mono- and tetra-brominated tris(triphenyl)phenyl corrole.

The selective bromination of the corrole ttppcH₃ was carried out as shown in Scheme 1. The starting ttppcH₃ was prepared according to a previous report.²⁷ To produce the monobrominated corrole, ttppcH₃ was dissolved in CH₂Cl₂ and treated with one equivalent of NBS in CH₃CN dropwise over the course of 6 – 8 h. The slow addition of the NBS solution was necessary to favor a singly brominated product, and was used to minimize the formation of isomers, or multiply brominated complexes.¹⁸ Aqueous workup of this reaction, followed by analysis with TLC (hexanes/CH₂Cl₂ 1/1 v/v), shows one new spot (R_f = 0.7) along with trace amounts of unreacted starting material. The product was isolated by column chromatography on silica gel (hexanes/CH₂Cl₂ 1/1 v/v). The starting material R_f (0.6) is similar to the product under these conditions, but separation is observable under UV light (254 nm). The starting ttppcH₃ compound exhibits a bright pink fluorescence when illuminated by UV lamp, while the monobrominated product does not. The resulting dark green solid isolated by chromatography was pure by TLC analysis and was identified as a selectively monobrominated corrole by UV-vis and NMR spectroscopy, as well as by mass spectrometry (Figure S1).

Scheme 1.

Synthesis of 2-bromo-ttppcH₃ and 2,3,17,18-tetrabromo-ttppcH₃ (Ph3Ar = 2,4,6-triphenylphenyl)

We also attempted to synthesize the tetrabrominated complex (Br₄-ttppcH₃) by treating ttppcH₃ with 4.4 equivalents of NBS added dropwise over the course of 2 h. Aqueous workup to remove any excess NBS followed by TLC analysis (hexanes/CH₂Cl₂ 1/1 v/v) showed one new spot (R_f = 0.85) that was isolated by column chromatography using silica gel (hexanes/CH₂Cl₂ 1/1 v/v). The brominated species can be differentiated from any non-brominated material through the use of short-wave UV light (254 nm). The UV-vis spectrum of this mixture of perbrominated species isolated as a dark green solid after column chromatography is shown in Figure 1, together with the mono-brominated corrole 2-Br-ttppcH₃. Laser desorption ionization mass spectrometry (LDI-MS) reveals an intense isotopic cluster centered at 1525.8 m/z, corresponding to the tetrabrominated Br₄-ttppcH₃ (Figure S2). However, a second isotopic cluster of significant intensity centered at 1604.8 m/z is also observed, which corresponds to a pentabrominated species (Br₅-ttppcH₃). Analysis by ¹H NMR spectroscopy was consistent with the presence of more than one compound, and attempts to further purify Br₄-ttppcH₃ were unsuccessful. The ¹H NMR spectrum of the mixture of perbrominated compounds isolated from the reaction mixture and containing the intended tetrabominated compound (Br₄-ttppcH₃) displays two doublets at 8.05 and 7.90 ppm, chemical shifts typical for β-pyrrole H atoms (Figure S3). These peaks are consistent with a mirror plane of symmetry bisecting the corrole C-C bond following bromination and implies that the bromine substitutions occured at the 2, 3, 17, and 18 pyrrole carbon atoms.

Figure 1.

UV-vis spectrum of crude Br₄-ttppcH₃ (blue line) and 2-Br-ttppcH₃ (orange line) in CH₂Cl₂.

The UV-vis spectrum for the monobrominated corrole is shown in Figure 2 together with the starting ttppcH₃. The Soret band undergoes a 2 nm bathocromic shift upon bromination. This observation is in agreement with previous reports of other corroles that were halogenated on the β-carbon atoms, including the analogous monochlorinated Cl-ttppcH₃, which exhibited a 1 nm redshift in the Soret region.²⁸ The Q-band region exhibits a more complex absorbance, displaying additional long wavelength absorbance between 700 – 800 nm.

Figure 2.

UV-vis spectrum of 2-Br-ttppcH₃ (orange line) and ttppcH₃ (blue line) in CH₂Cl₂.

Monobromination of the corrole system led to an overall downfield shift in the ¹H NMR peaks corresponding to the protons at the β-positions of the macrocycle. The addition of the bromine substituent to the 2-carbon position also disrupts the symmetry present in the starting ttppcH₃ compound.²¹ Due to this decrease in symmetry the ¹H NMR spectrum of 2-Br-ttppcH₃ shows seven distinct peaks corresponding to the β-pyrrole protons: six doublets, two of which are partially overlapping, and one singlet assigned to the proton at the 3-carbon position adjacent to the bromine substituent (Figure 3). In contrast, the β-pyrrole protons of ttppcH₃ produce only 3 distinct peaks in the ¹H NMR.²⁴ The remaining resonances in the spectrum for 2-Br-ttppcH₃ (shown in Figure S4) correspond to the meso-aryl groups and show relatively minor shifts compared to ttppcH₃.²⁷

Figure 3.

¹H NMR spectrum (400 MHz, CDCl₃) of 2-Br-ttppcH₃ showing the peaks associated with the H atoms attached to the pyrrole β-carbon atoms.

Structure of 2-Br-ttppcH₃.

Crystallization of the monobrominated corrole was achieved through the slow evaporation of a CH₂Cl₂/hexane mixture containing 2-Br-ttppcH₃ following purification by column chromatography. Single crystals for analysis by X-ray diffraction (XRD) were isolated as dark green blocks. The structure revealed that the corrole ring was selectively mono-brominated at the 2(18)-carbon position (Figure 4), exhibiting the single regioisomer expected from the NMR and LDI-MS data. This preference for the C2 carbon position is not observed in other C-H substitutions of corroles by electrophilic substrates, with examples of chlorination, sulfonation, nitration, and hydroformylation all preferring to substitute at the C3 carbon.^{20, 30-31} However, the previous example of mono-bromination of the 5,10,15-tritolylcorrole by Paolesse and coworkers produced a mixture of isomers, with bromination occurring at either the C2 or C3 carbons.^{13, 20} We speculate that our system’s preference for substitution at the C2 position may be due to the high steric profile of the meso triphenylphenyl substituents crowding nearer to the C3 carbon. The same regioselectivity was observed in the incidental monochlorination of the ttppcH₃ corrole, and steric crowding was also invoked as the likely reason for C2 selectivity.²⁸

Figure 4.

Displacement ellipsoid plot (30% probability level) for 2-Br-ttppcH₃ at 110(2) K. H atoms (except for those attached to N1, N2, and N3) are omitted for clarity.

Due to rapid tautomerization, the three internal N-H protons are distributed over 4 positions, ^32-34 and were modeled as being 4 hydrogens with 0.75 occupancy. The new C-Br bond has a length of C2-Br1 = 1.880(2) Å, which is only slightly shorter than the C-Br bond in carbon tetrabromide (1.942 Å).³⁵ A distorted planar structure was revealed for the macrocycle, with a torsion angle of N1-C1-C19-N4 = 6.9(3)°.. A diagram of the corrole core, with all atom labels included, for all complexes discussed is given in Figure S5.

Structure of ttppcH₃ and comparisons with 2-Br-ttppcH₃.

To our knowledge, the structure of the non-brominated ttppcH₃ ligand has not yet been reported.^{24, 27} Crystallization of the complex was achieved by slow evaporation of a fluorobenzene/pentane solution containing both the ttppcH₃ compound and 1,3,5-triphenylbenzene as a co-crystallant. The asymmetric unit of the resulting crystals contained one molecule of 1,3,5-triphenylbenzene and fluorobenzene per asymmetric unit. The structure of the ttppcH₃ molecule is shown in Figure 5. Similar to the 2-Br-ttppcH₃ complex, this crystal structure shows a distorted planer geometry about the corrole macrocycle, with a torsion angle for the N1-C1-C19-N4 = 10.2(3)°, though the magnitude of this distortion is significantly greater. The C-C bonds around the corrole ring do not differ significantly between the two structures (Table S1).

Figure 5.

Displacement ellipsoid plot (30% probability level) for ttppcH₃ at 110(2) K. H atoms (except for those attached to N1, N2, and N3) are omitted for clarity.

Metalation of mono-brominated tris(triphenyl)phenyl corrole (Fe(Cl)(2-Br-ttppc)).

A modified version of the procedure used for metalation of the non-brominated ttppc ligand was followed (scheme 2), starting with the addition of FeCl₂ to a solution of 2-Br-ttppcH₃ in anhydrous DMF under Ar. This solution was heated to 130 °C for 12 h, at which time an apparent color change from green to dark brown was observed. Shorter reaction times lead to the same color change though afford a lower yield. Upon workup and purification by removal of the DMF in vacuo and column chromatography using silica gel and Et₂O as the eluent (R_f~0.90), the metalated product was isolated as a dark brown solid. Analysis by TLC indicated the product at this stage still contained trace impurities as well as unreacted ligand. However, running the reaction for longer times resulted in the formation of additional impurities that were difficult to remove by column chromatography. Further purification by silica gel with 4:1 CH₂Cl₂:hexanes as eluent (R_f ~0.60) afforded Fe(Cl)(2-Br-ttppc) as a dark brown solid, which appeared pure by both ¹H NMR spectroscopy and TLC.

Scheme 2.

Synthesis of Fe(Cl)(2-bromo-ttppc)

The UV-vis spectrum of Fe(Cl)(2-Br-ttppc) is shown in Figure 6. A 3 nm bathochromic shift of both Soret peaks and both Q-bands is observed in comparison to the reported UV-vis of Fe(Cl)(ttppc).²⁴ All other features of the UV spectra are similar to those observed for the Fe(Cl)(ttppc) complex, though with slight variation in extinction coefficient. Comparison between the 2-Br-ttppcH₃ and the metalated species in the UV-vis spectra shows a split Soret band for the iron complex.

Figure 6.

UV-vis spectrum of Fe(Cl)(2-Br-ttppc) in CH₂Cl₂.

The ¹H NMR spectrum of the metalated monobrominated corrole in CDCl₃ displays six distinct paramagnetic peaks between −5.1 and −7.7 ppm (Figure S6). There are no peaks further downfield than 9 ppm, with many of the peripheral C-H peaks from the corrole’s aryl groups appearing in the diamagnetic aromatic region (between 7 and 9 ppm). The analogous Fe(Cl)(ttppc) complex shows a distinct upfield peak at −35 ppm, which may correspond to the H atom attached to the 2-carbon position in ttppc, which has been replaced by Br in the 2-Br-ttppc ligand.

Structure of Fe(Cl)(2-Br-ttppc).

Crystallization of the metalated corrole was done by layering a solution of the complex dissolved in DCM with hexanes and allowing the layers to slowly mix over the course of several days. This mixture was then left partly open to atmosphere for the solution to slowly evaporate, depositing small, blocky crystals of the desired Fe(Cl)(2-Br-ttppc) complex, whose structure is shown in Figure 7. The C2-Br1 bond distance was determined to be 1.864(2) Å. The distance from the Fe to the mean plane of the four nitrogen atoms of the ring is 0.349 Å, similar to the distance of 0.341 Å found in the analogous Fe(Cl)(ttppc) complex. The Fe-Cl bond distance of 2.225(2) Å is also comparable, being only slightly contracted relative to the 2.2559(1) Å observed for the non-brominated corrole analog.²⁴ A more complete list of the bond lengths and angles for both the Fe(Cl)(ttppc) and Fe(Cl)(2-Br-ttppc) complexes is provided in the Supporting Information (Table S2).

Figure 7.

Displacement ellipsoid plot (30% probability level) for Fe(Cl)(2-Br-ttppc) at 110(2) K. H atoms are omitted for clarity.

Electrochemistry.

Cyclic voltammograms (CVs) were collected on the brominated corroles to examine the influence of a single bromine substituent at a β-carbon position on their electrochemistry. The CVs of a crystalline sample of metal-free 2-Br-ttppcH₃ acquired in CH₂Cl₂ is shown in Figure S7. The compound shows two quasi-reversible waves at E_1/2 = −720 and +170 mV vs ferrocene (Fc^+/0). The CV of the unsubstituted ttppcH₃ taken under the same conditions (Figure S8) displays a reductive wave at E_1/2 = −690 mV and an oxidative wave at +150 mV vs Fc^+/0. These data show that substitution with a single Br atom induces a −30 mV shift for the first reduction, while inducing a +20 mV shift for the first oxidation, as compared to the unsubstituted macrocycle ttppcH₃. These values can be compared to the singly chlorinated analog, 2-Cl-ttppcH₃. Although the CV for 2-Cl-ttppcH₃ is collected in a different solvent (DMF), it shows a first reduction potential that is the same as the unsubstituted ttppcH₃, whereas its first oxidation is shifted by +90 mV.²⁸ The first reductions for the halogenated compounds do not appear to shift as expected from simple electronegativity considerations, but the oxidation potentials are consistent with the electronegativity trend Cl > Br > H.^36-37

Iron complexes.

The CV of Fe(Cl)(2-Br-ttppc) is shown in Figure 8, and displays two quasi-reversible waves. The first oxidation event occurs at E_1/2 = +553 mV, and the first reduction can be seen at E_1/2 = −577 mV. The anodic and cathodic peak currents show a linear dependence on the square root of the scan rate, consistent with quasi-reversible redox processes (Figure S9).³⁸ Both of these redox events are shifted anodically relative to the unsubstituted Fe(Cl)(ttppc) analog, with a positive shift of +93 mV (E_1/2 = +460 mV) for the first oxidation and +63 mV (E_1/2 = −640 mV) for the first reduction.²⁴ This difference in shift for the two redox events indicates that the bromide has a greater influence on the energy of the HOMO than it does on the LUMO of the corrole iron complex.

Figure 8.

Cyclic voltammogram of Fe(Cl)(2-Br-ttppc) in CH₂Cl₂ containing 0.1 M TBAPF₆ electrolyte.

It should be noted that metallocorroles are often described as non-innocent, in which the π-electrons of the macrocycle can be delocalized on to the metal center. For an Fe(Cl) corrole, the non-innocent behavior results in ambiguity regarding the assignment of the formal iron oxidation state, with Fe^IV(Cl)(corrole) versus Fe^III(Cl)(corrole•+) as the two possible extremes. Determining the degree to which an iron corrole is non-innocent is difficult to achieve through experiment,³⁹ and therefore we do not assign a formal oxidation state to Fe(Cl)(2-Br-ttppc). It has been suggested that metalated corroles with more positive reduction potentials are considered to be less innocent,⁴⁰ which implies that the metal center in Fe(Cl)(2-Br-ttppc) may be closer to an Fe(III) oxidation state than the metal center in Fe(Cl)(ttppc). The HOMO-LUMO gap in porphyrinoid compounds can be related to the difference in the E_1/2 values between the first oxidation and reduction.⁴¹ The 2-Br-ttppcH₃ complex has a 50 mV larger HOMO-LUMO gap than its non-brominated analog, while the Fe metalated complex shows a smaller increase in the HOMO-LUMO gap of ~30 mV. This pattern is opposite to the results reported for β-nitro-tritolyl corrole versus unsubstituted tritolyl corrole, with the nitro-substituted metal-free compound having a HOMO-LUMO gap ~320 mV smaller than the unsubstituted analog, while the nitro-substituted Fe-Cl corrole exhibited no change in the HOMO-LUMO gap relative to the undecorated corrole.^42-43

The Mössbauer of the ⁵⁷Fe(Cl)(2-Br-ttppc) complex has a similar quadrupole splitting and isomer shift to ⁵⁷Fe(Cl)(ttppc) when measured at 4K in toluene (Figure 9). The isomer shift of 0.17 mm s⁻¹ for the 2-Br-ttppc derivative is nearly identical to that of the ttppc analog (0.18 mm s⁻¹), while the quadrupole splitting of 2.76 mm s⁻¹ for the brominated complex is slightly lower than the 2.86 mm s⁻¹ observed for the non-brominated analog.²⁴ These parameters are comparable to other formally Fe(IV) chloride corrole complexes.⁴⁴ The general similarity of the Mössbauer spectra, together with the significant changes in the electrochemical data, suggests that the main influence of monobromination on the iron complex involves the electronic distribution of ligand-based, as opposed to metal-based, orbitals.

Figure 9.

Mössbauer spectrum (80 K) of ⁵⁷Fe(Cl)(2-Br-ttppc) in toluene.

SUMMARY AND CONCLUSIONS

The syntheses and characterization of an isomerically pure, monobrominated metal-free corrole has been demonstrated. While previous attempts to synthesize monobrominated corroles using NBS have resulted in a mixture of isomers,^{4, 18} we successfully employed NBS for the selective halogenation of a sterically bulky, metal-free corrole at the 2-carbon position. A key factor in the synthesis was the controlled, very slow addition (6 – 8 h) of the brominating agent. A significantly higher yield (47%) and better selectivity was achieved with this methodology than could be obtained with haloacid and the related 5,10,15-tritolylcorrole,²⁰ and makes this method practical for routine ligand preparation, metal complexation, and possible further ligand derivatization.

The molecular structures of both 2-Br-ttppcH₃ and ttppcH₃ were obtained by single crystal XRD, as well as the iron chloride complex Fe(Cl)(2-Br-ttppc). The electrochemistry of the monobrominated corrole and related Fe(Cl) complex revealed significant shifts in redox potentials relative to their non-brominated analogs. These shifts reveal the influence of a single bromine atom attached at a β-pyrrole carbon position on the electronic character of the corrole macrocycle. The small change in quadrupole splitting, and negligible change in isomer shift, of the Mössbauer spectrum of Fe(Cl)(2-Br-ttppc) versus Fe(Cl)(ttppc), suggests that monobromination at the 2-carbon position of the corrole macrocycle has only a minor impact on the electronic distribution about the metal center.

EXPERIMENTAL

General Procedures.

All chemicals were purchased from commercial sources and used as received, except for N-bromosuccinimide (NBS), which was recrystallized from boiling water and then dried in vacuo before use. Syntheses requiring an inert atmosphere were performed using standard Schlenk techniques under Ar. Toluene and acetonitrile were purified via a Pure-Solv solvent purification system from Innovative Technologies, Inc. All deuterated solvents for NMR and ⁵⁷Fe metal were purchased from Cambridge Isotope Laboratories, Inc. (Tewksbury, MA). The electrolyte (Bu₄N)PF₆ was purchased from Sigma-Aldrich and recrystallized twice from ethanol prior to use.

Instrumentation.

All ¹H nuclear magnetic resonance (NMR) spectra were obtained using a Bruker Avance 400 MHz spectrometer. UV-vis measurements were collected using quartz cuvettes (path length = 1 cm) and a Hewlett-Packard Agilent 8453 diode-array spectrophotometer. Laser desorption ionization mass spectrometry (LDI-MS) was performed on a Bruker AutoFlex III Maldi-Tof/ToF instrument equipped with a nitrogen laser at 335 nm using an MTP 384 ground steel target plate. Before each set of runs, the instrument was calibrated against a peptide standard of known molecular weight. All cyclic voltammograms were recorded in CH₂Cl₂ at 23 °C under an Ar atmosphere using tetrabutylammonium hexafluorophosphate (TBAPF₆, 0.1 M) as the electrolyte. An EG&G Princeton Applied Research potentiostat/galvanostat model 263A and a three-electrode setup comprised of a glassy carbon working electrode, a platinum counter electrode, and an Ag/AgNO₃ reference electrode (0.01 M AgNO₃ in 0.1 M TBAPF₆/CH₂Cl₂) was used to obtain the scans, and the raw data were referenced against an external ferrocene standard. Mössbauer spectra were recorded on a SEE Co. (Edina, MN) spectrometer in constant acceleration mode with transmission geometry. All measurements were recorded at 80 K, maintained with liquid N₂ using a Janis SVT-400 cryostat from Janis Research Co. (Wilmington, MA). Isomer shifts were calibrated relative to the centroid of the spectrum of a α-Fe metallic foil, collected at room temperature. Data analysis was performed using the program WMOSS (www.wmoss.org) version F, and quadrupole doublets were fit to Lorentzian line shapes.

Single Crystal X-ray Crystallography.

For all structures, reflection intensities were measured at 110(2) K using a SuperNova diffractometer (equipped with Atlas detector) Cu Kα radiation (λ = 1.54178 Å) under the program CrysAlisPro (Version 1.171.39.29c, Rigaku OD, 2017). The same program was used to refine the cell dimensions and for data reduction. The structures were solved with the program SHELXS-2014/7 and refined on F² with SHELXL-2014/7.⁴⁵ The temperature of the data collection was controlled using the system Cryojet (manufactured by Oxford Instruments). An analytical numeric absorption correction method was used involving a multifaceted crystal model based on expressions derived elsewhere.⁴⁶

2-Br-ttppcH₃.

The hydrogen atoms attached to the pyrrole nitrogen atoms N1 and N3 were located from the difference map, those attached to N2 and N4 were placed at calculated positions using the instruction AFIX 43 and given isotropic displacement parameters having values 1.2 Ueq of the attached N atoms. All other H atoms were placed at calculated positions using the instructions AFIX 43 with isotropic displacement parameters having values 1.2 Ueq of the attached C atoms. Due to delocalization of the 3 H atoms between the 4 pyrrole N atoms, the 4 H positions were refined as having a 0.75 occupancy. The crystal lattice contains disordered and/or partially occupied lattice solvent molecules (CH₂Cl₂ and C₆H₁₄). Their contributions were removed from the final refinement using the SQUEEZE program.⁴⁷ The crystallographic data for 2-Br-ttppcH₃ are summarized in Table S3.

ttppcH₃.

The hydrogen atoms attached to the pyrrole nitrogen atoms were located from the difference map and given isotropic displacement parameters having values 1.2 Ueq of the attached N atoms. All other H atoms were placed at calculated positions using the instructions AFIX 43 with isotropic displacement parameters having values 1.2 Ueq of the attached C atoms. Due to delocalization of the 3 H atoms between the 4 pyrrole N atoms, the 4 H positions were refined as having a 0.75 occupancy. The crystallographic data for ttppcH₃ are summarized in Table S4.

Fe(Cl)(2-Br-ttppc).

All H atoms were placed at calculated positions using the instructions AFIX 43 with isotropic displacement parameters having values 1.2 Ueq of the attached C and N atoms. The crystal lattice contains disordered and/or partially occupied lattice solvent molecules (CH₂Cl₂ and C₅H₁₂). Their contributions were removed from the final refinement using the SQUEEZE program.⁴⁷ The crystallographic data for ttppcH₃ are summarized in Table S5.

Mössbauer spectroscopy.

Fe(Cl)(2-Br-ttppc) enriched in ⁵⁷Fe was used to collect the Mössbauer spectrum. The isotopically-labeled complex was prepared from ⁵⁷FeCl₂ (⁵⁷Fe, 95+% purity) following the same procedure as the unlabeled material. The ⁵⁷FeCl₂ was synthesized from ⁵⁷Fe metal and HCl (aq).⁴⁸ The Mössbauer sample (10 mM, 400 μL toluene) was prepared, transferred into a custom-made Delrin cup (10 mm x 12.4 mm OD), and frozen in liquid N₂ prior to data collection at 80 K.

Synthesis

2-Bromo-(5, 10, 15-tris(2, 4, 6-triphenylphenyl)corrole) (2-Br-ttppcH₃).

A solution of NBS (57 mg, 0.32 mmol) in CH₃CN (20 mL) was added dropwise to a solution of ttppcH₃ (350 mg, 0.289 mmol) in CH₂Cl₂ (40 mL) at 23 °C over the course of 6 – 8 h. After the addition was complete, the reaction mixture was stirred for an additional 1 h. The product mixture was then washed with deionized water, followed by a saturated NaHCO₃ (aq) solution and then with brine. The organic layer was dried with Na₂SO₄, and the volatiles were removed in vacuo. Pure product was obtained by column chromatography on silica gel with 1:1 CH₂Cl₂:hexanes as eluent. The first green band (R_f ~ 0.7) corresponds to the desired monobrominated corrole. Unreacted starting material (ttppcH₃) eluted in a second green band (R_f ~0.55), which was isolated in sufficient purity to be recycled. The first green band was dried under reduced pressure to yield the monobrominated corrole, 2-Br-ttppcH₃, as a dark green solid (175 mg, 47%). Crystals of the product suitable for X-ray diffraction were obtained by slow evaporation of a solution of the product in CH₂Cl₂ layered with hexanes. UV-vis (CH₂Cl₂): λ_max [nm] (ε x 10⁻³ M⁻¹ cm⁻¹): 432 (35.5), 573 (4.9), 655 (4.8). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 8.70 (d, J = 4.4 Hz, 1H), 8.45 (d, J = 4.8 Hz, 1H), 8.36 (d, J = 4.8 Hz, 1H), 8.28 (s, 1H), 8.23 (d, J = 4.8 Hz, 1H), 8.18 (d, J = 4.4 Hz, 1H), 8.16 (d, J = 4.4 Hz, 1H), 7.95 (m, 12H), 7.59 (t, 6H), 7.49 (t, 4H), 6.95 (t, 8H), 6.70 (d, 4H), 6.60 (t, 5H), 6.50 (m, 8H), 6.32 (t, 4H). LDI-MS (m/z): isotopic center clustered at 1291.05 (M+H)⁺ (calcd 1291.00, C₉₁H₆₂N₄Br₁).

Fe(Cl)(2-Br-ttppc).

The procedure for metalation of the monobrominated corrole with Fe^IICl₂ was modified from the procedure employed for the unsubstituted corrole ttppcH₃.²⁴ To a solution of 2-Br-ttppcH₃ (83.5 mg, 64.7 μM) in dry DMF (20 mL), an amount of Fe^IICl₂ (190 mg, 1.49 mM, 24 equiv) was added under Ar. The solution was heated to 130 °C for 12 h with stirring. A color change from green to dark brown was observed after 3 h. Analysis by TLC indicated the formation of a new brown product together with the presence of some unreacted starting material. Longer reaction times led to the formation of unidentified sideproducts. Upon completion of the reaction, the DMF was removed by vacuum distillation to give a crude brown solid. This crude product was dissolved in CH₂Cl₂ and purified by column chromatography on silica gel with Et₂O as eluent (Rf ~ 0.90). This initial column helped to remove any remaining metal salts. Further purification was then carried out by a second silica gel column with 4:1 CH₂Cl₂:hexanes as eluent (Rf ~0.60). The pure product, Fe(Cl)(2-Br-ttppc), was isolated as a dark brown solid (78 mg, 85%). Crystals suitable for XRD were grown by layering a solution of the pure product in CH₂Cl₂ with pentane followed by slow evaporation, giving dark brown blocks (66 mg, 75%). UV-vis (CH₂Cl₂): λ_max [nm] (ε x 10⁻³ M⁻¹ cm⁻¹): 372 (4.3), 410 (4.5), 523 (1.8), 642 (0.5). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 8.68 (s, br), 8.49 (s, br), 8.33 (m, br), 8.16 (m, br), 7.02 (s, br), 6.72 (s, br), 4.07 (s, br), −5.15 (s, br), −5.52 (s, br), −5.85 (s, br), −6.20 (s, br), −6.71 (s, br), −7.64 (s, br). LDI-MS (m/z): isotopic center clustered at 1378.98 (M+H)⁺ (calcd 1379.00, C₉₁H₅₉N₄Br₁Cl₁Fe₁).