Synthesis, structural characterization, and chemical properties of pentacoordinate model complexes for the active site of [Fe]-hydrogenase

Abstract

Based on the synthesis of two new hexa-coordinate N-heterocyclic carbene (NHC) substituted precursors FeI₂(CO)₃(NHC) with NHC = IPr (1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene) (2) and NHC = SIPr (1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylide) (3) and two known complexes FeI₂(CO)₃SIMes, SIMes = (1,3-bis(2,4,6-trimethylphenyl)-imidazol-2-ylide) (1) and FeI₂(CO)₃IMe IMe = 1,3-bismethylimidazol-2-ylidene (4), four new mononuclear [Fe]-hydrogenase model complexes Fe(CO)₂(NS)(NHC) (NS = aminothiophenol; 5, NHC = SIMes; 6, NHC = IPr; 7, NHC = SIPr; 8, NHC = IMe) were prepared by substitution of two iodine ligands with the help of dipotassium-aminothiophenol salt and subsequently the absence of a CO ligand. New complexes 2, 3 and 5–8 were fully structurally characterized by infrared spectroscopy (IR), elemental analysis, NMR spectroscopy, and X-ray crystallography. IR spectroscopy studies show that complexes 5–8 exhibit similar IR patterns and absorption wavelengths in terms of ν(CO) with the active site of [Fe]-hydrogenase. The facile protonation/deprotonation of the NS ligand of complexes 6 and 7 was disclosed with the assistance of IR spectroscopy. The NS ligand accepts a proton reversibly as an internal base, generating two protonated species, [Fe(CO)₂IPr(H-NS)]⁺ and [Fe(CO)₂SIPr(H-NS)]⁺, which play the same role with the intrinsic cysteine thiolate ligand in [Fe]-hydrogenase. DFT results showed that the N atom of the NS ligand is the thermodynamically active proton acceptor in acetone while the NS ligand is prone to be dually protonated first in the N atom and then the S atom in the gas phase. Complex 5 exhibited simultaneous protonation and the combination of CO, forming a new mer-tricarbonyl species in the presence of CO and HBF₄. It showed easy reversible deprotonation and deprivation of CO with the assistance of t-BuOK or Et₃N. Also, the electrochemical properties of these new pentacoordinate model complexes were explored through cyclic voltammetry, which enabled us to identify the contributions of different NHC ligands to the complexes' redox properties.

---

Introduction

Hydrogenases are enzymes found in some microorganisms that catalyze H₂ metabolism reversibly and are classified into three species according to the different metal cores inside: [FeFe]-hydrogenase, [NiFe]-hydrogenase and [Fe]-hydrogenase.^1–5 Different from the [FeFe]-hydrogenase and [NiFe]-hydrogenase that catalyze reversible proton reduction and hydrogen oxidation (H₂ = 2H⁺ + 2e⁻), the third class of hydrogenase, [Fe]-hydrogenase (Hmd, the H₂-forming methylenetetrahydromethopterin dehydrogenase) catalyzes the reversible reduction of N⁵,N¹⁰-methenyl-tetrahydromethanopterin (methenyl-H₄MPT⁺, or MPT⁺) with H₂ to N⁵,N¹⁰-methylene-tetrahydromethanopterin (methylene-H₄MPT, or HMPT) and a proton.^4,5

Recent studies of the active site structure of Hmd have revealed that a mono iron center has a pseudo-octahedral coordination with a cysteine sulfur atom, two cis-CO ligands, a bidentate pyridone ligand through its nitrogen and acyl carbon atoms, and an open site (or a H₂O) trans to the acyl group (Fig. 1).^6–11 The H₂ molecule is possible to bind with Hmd at the open site, and subsequent heterolytic H–H bond cleavage happens with the help of internal base, i.e. cysteine-S ligand.¹² Some models have proven that it is challenging for the cysteine-S ligand to activate H₂ molecule, but it hasa reversible protonation property, which means that the internal base works as a proton acceptor.^4,8,10,13,14 Song et al. developed some hexacoordinated mononuclear Fe model complexes with the NS ligand including 2-mercapto-pyridine and 8-mercapto quinoline to mimic the active site of Hmd.^15–18 Hu et al. did research with semisynthetic [Fe]-hydrogenases and they found the reconstituted complex containing 2-hydroxypyridine group could activate H₂ while the reconstituted stuff containing 2-methoxypyridine was inactive.¹⁹ The experimental results along with DFT computions demonstrate that a competent internal base is essential to the model complexes to activate H₂. Based on the finding they are the first to develop Fe-based functional mimic of the Hmd active site. The functional model with a pendant amine moiety was able to activate H₂ and hydrogenate an aldehyde.²⁰ Besides, the iron(II) hydride is also considered to be a key intermediate in the metabolism of CO₂ to methane. Rose et al. reported such a biomimetic Fe–H species, which showed some reactivities yet the structure and reactivity is somehow distinct from that found in the active site of Hmd.²¹ As the hydroxyl group of acylmethylpyridinol moiety may be responsible for H₂ activation as well, Hu et al. reported Hmd models containing acylmethylpyridinol ligands.^22–24 The research showed the 6-hydroxy group of the acylmethylpyridinol ligand may effect the activity.²³

Fig. 1 Structure of the active-site of Hmd.

It is worth mentioning that the active site is pentacoordinate without regard to the open site or the possible solvent molecule. Some synthetic structural analogues of Hmd²⁵ have been developed and some of them show interesting CO-exchange or CO-uptake behaviors^26–28 and spectroscopic properties similar to Hmd, among which few have the exact same direct ligating atoms and same ligancies.

Based on the developing models FeI₂(CO)₃L, (L = SIMes (1), IMe (4)) obtained in our previous work and two new analogues FeI₂(CO)₃L, (L = IPr (2), SIPr (3)),²⁹ four new corresponding pentacoordinate mono iron di-carbonyl complexes Fe(CO)₂L(NS) (L = SIMes (5), IPr (6), SIPr (7), IMe (8)) with nitrogen heterocyclic carbene (NHC) ligand and 2-aminothiophenolate dianion bidentate non-innocent ligand were reported in this work as models of Hmd active site. The series of model complexes possess exactly the same direct coordinated atoms and similar geometry as the Hmd active site. We also conducted elaborate research on different ligands and found that complex 5 exhibited CO-uptake behavior successfully while complexes 6 and 7 only got protonated species due to the hamper from two isopropyl groups. The reasons we are interested in studying the series of complexes are as follows: (I) as far as we know, very few five coordinated mono iron di-carbonyl complexes have been reported³⁰ (others are mostly six coordinated). (II) Several different kinds of NHCs have been introduced, which have become promising and potential ligands for homogeneous catalysis and organocatalysts due to their excellent σ-donating ability during the past 20 years.^31,32 (III) The internal base, i.e. 2-aminothiophenolate, is of interest and can help us explore more about the protonation process. Liu et al. introduced a diazadiphosphine ligand to explore the role of the pendant base in heterolytic H–H bond cleavage to resemble the active site of [FeFe]-hydrogenaze enzyme.³³ Corresponding to the proposed working mechanism of the active site of the [FeFe]-hydrogenase, they concluded that a Fe^II coordination center and the ligand of pendant amines are key to heterolytic H–H bond cleavage. Since the internal base plays a key role in the heterolytic cleavage of dihydrogen, possibly working as a proton accepter, we were inspired to prepare Hmd model complexes with different ligands to explore the protonation process. Different ligands result in different electronic and steric effects, which can help us to further understand the active site structure of Hmd. We hope to report the synthesis, characterization and chemical properties of new types of Hmd model complexes.

Results and discussion

Synthesis and characterization of FeI₂(CO)₃L (L = IPr (2), SIPr (3))

We previously synthesized FeI₂(CO)₃L (L = SIMes (1), IMe (4)) following the procedure in Fig. 2.²⁹ Similarly, we prepared FeI₂(CO)₃L(L = IPr (2), SIPr (3)) as the precursors of target compounds of Fe(CO)₂L(NS) (L = IPr (6), SIPr (7)). The freshly prepared NHC ligand in hexane was slowly added into a solution of FeI₂(CO)₄ at room temperature, giving rise to the hexa-coordinate FeI₂(CO)₃IPr and FeI₂(CO)₃SIPr in 50% and 53% yields, respectively. Like other precursors, precursor 2 and 3 are air-stable dark solids, which have been characterized by IR spectroscopy, NMR spectroscopy, X-ray diffraction and elemental analysis. According to the ν(CO) IR patterns in Fig. 3, the two compounds are of the meridional configuration, the same kind of geometrical configuration with configurations of reported complexes 1, 4, FeI₂(CO)₃IMes, FeI₂(CO)₃PCy₃, FeI₂(CO)₃PPh₃, etc.²⁹ As expected, the CO peak shifted to a lower wavenumber compared with the original CO peak of FeI₂(CO)₄ due to stronger donor abilities of the substituent NHC ligands. The CO frequencies of the two precursors (2087(w), 2038(s), 2020(m) for complex 2 and 2086(w), 2038(s), 2019(m) for complex 3) are close to those with NHC ligands 1 2082(w), 2039(s), 2019(m), 4 2082(w), 2034(s), 2023(m) and FeI₂(CO)₃IMes 2087(w), 2040(s), 2018(m) while higher than those with phosphine ligands, FeI₂(CO)₃PMe₃ 2093(s), 2047(s), 2022(s), FeI₂(CO)₃PPh₃ 2093(w), 2045(s), 2032(m), and FeI₂(CO)₃P(OEt)₃, 2104(w), 2053 (s, br) indicating IPr, SIPr and other NHC ligands have stronger electron-donating abilities compared with phosphine ligands.²⁵

Fig. 2 The synthetic route of complexes 5–8.

Fig. 3 Infrared spectra (ν(CO) region) of complexes 2 and 3 recorded in CH₂Cl₂.

We determined the crystal structures of precursors 2 and 3 by X-ray diffraction analysis. The ORTEP plots of 2 and 3 are shown in Fig. 4. Meanwhile, the selected bond lengths and angles are listed in Table 1. The structures of complexes 2 and 3 have much in common, except for the C–C bond lengths in the NHC five-membered rings. The C–C bond lengths of complex 3 are slightly longer than those of complex 2. The two structures are both of meridional geometry and the NHC rings are almost in the same plane with FeI(CO)₂. The CO ligand and trans iodine atom are almost perpendicular to the FeI(CO)₂ plane with the C1–Fe1–C2 93.5(3) and I1–Fe1–I2 91.77(4) for complex 2. The benzene rings flanking the NHC ring have an effect on the ipsilateral CO ligand, leading to a C1–Fe1–C3 angle of 162.7(3) and a C2–Fe1–C3 angle of 160.6(4) in complex 2.

Fig. 4 Molecular structures of 2 and 3 with 30% probability level ellipsoids. Hydrogen atoms are omitted for clarity, except the ones in the imidazole to distinguish saturated and unsaturated complexes.

Table 1 Selected bond lengths [Å] and angles [°] for 2 and 3

2		3
Fe1–I1	2.6907(12)	Fe1–I1	2.6914(15)
Fe1–C1	1.837(8)	Fe1–C1	1.758(11)
Fe1–C3	1.847(8)	Fe1–C3	1.861(11)
N1–C18	1.373(8)	N1–C4	1.447(12)
C5–C6	1.344(10)	C16–C17	1.488(15)
Fe1–I2	2.6306(13)	Fe1–I2	2.6255(15)
Fe1–C2	1.787(10)	Fe1–C2	1.850(11)
Fe1–C4	1.998(7)	Fe1–C18	2.006(9)
N1–C4	1.454(8)	N1–C18	1.350(11)
C1–Fe1–C2	93.5(3)	C1–Fe1–C2	95.7(4)
C2–Fe1–C18	92.3(3)	C2–Fe1–C18	97.5(4)
C4–N1–C5	112.5(6)	C16–N1–C4	117.5(7)
C1–Fe1–C3	162.7(3)	C1–Fe1–C3	94.5(4)
I1–Fe1–I2	91.77(4)	I1–Fe1–I2	92.00(4)
C4–N1–C19	127.6(6)	N1–C18–N2	107.9(8)

---

Synthesis and characterization of Fe(CO)₂L(NS) (L = SIMes (5), IPr (6), SIPr (7), IMe (8))

Synthesis of complexes 5–8 followed the latter procedure outlined in Fig. 2. Due to the excellent electron donor ability of NHC ligands and dianion 2-aminothiophenolate, complexes 5–8 were isolated as air-stable black crystalline solids in good yield with few byproducts detected by IR spectroscopy. Different from the phosphine (Fe(CO)₂PR₃(NS), R = PCy₃, PPh₃, P(OEt)₃) and cyanide (Fe(CO)₂CN(NS)⁻) derived complexes,³⁴ the target complexes 5–8 were rather stable and could be handled in the air and stored at room temperature for months. Furthermore, the solution forms of 5–8 in methanol (or ethanol) exhibited thermal stability (as high as 50 °C) and light stability, which leads to potential applications in catalytic hydrogenation or catalytic H₂ production reactions.

Fig. 5 displays the ν(CO) IR spectra (all are measured in CH₂Cl₂) of compounds 5–8 in terms of two clean carbonyl groups for their strong characteristics. The nearly equal intensities of the two bands suggest cis-dicarbonyls are at about 90° angles to those of the Hmd active site.⁶ The Hmd possesses two characteristic ν(CO) absorption bands at 2011 and 1944 cm⁻¹.^7,9

Fig. 5 Infrared spectra of Hmd model complexes 5–8 tested in CH₂Cl₂.

In agreement with our hypothesis, the four target compounds show similar carbonyl absorption bands with Hmd in terms of patterns and values, of which 7 exhibits the bands most similar to Hmd at 1995 and 1933 cm⁻¹. The ν(CO) band wavenumbers indicate that the strong electron-donating ability of NHC ligand is comparable to the PCy₃ ligand (derived complex Fe(CO)₂PCy₃(NS), measured in CH₂Cl₂, 1985 and 1927 cm⁻¹) as is common with these types of derivatives.³⁴ The CO bands of 5–8 shift to lower numbers in contrast with the respective six-coordinated precursors due to the substitution of iodine atoms by stronger electron-donating bidentate dianion 2-aminothiophenolate, resulting in excellent stability. Compared to 6, the corresponding saturated 7 exhibits a higher wavenumber at approximately 6 cm⁻¹, indicating that the unsaturated NHC ligand is more electron-donating than the corresponding saturated one consistent with the case of 5 and Fe(CO)₂IMes(NS)³⁵ reported in our previous work. The pair of IMes and SIMes ligands has a better capacity to donate electrons than the pair of IPr and SIPr ligands as the former pair has a lower average wavenumber by 6 cm⁻¹ compared to the latter.

The single crystals suitable for X-ray diffraction study were grown by slow evaporation of a diethyl ether solution of complexes 5–8 at room temperature. The molecular structures are presented in Fig. 6, and the selected bond lengths and angles are listed in Tables S3–S6.† The crystals of complexes 5–8 have similar geometries and share much in common and we take complex 6 as an example to elaborate. S1 and trans CO nearly form a straight angle with the angle S1–Fe1–C2 of 168.9° while N1–Fe1–C1 exhibits a greater curvature with an angle of 142.9° due to the steric influence of the 2,6-diisopropyl phenyl substituents flanking the carbene N. The Fe1–2-aminothiophenolate dianion bidentate ligand is parallel to the 2,6-diisopropyl phenyl substituent group. Compared with their unsaturated compounds, the saturated structures of 5 and 7 both show three longer bonds to a degree of 0.1 Å and two shorter bonds to a degree of 0.01 Å with the boundary of double N atoms within imidazole rings respectively. Of all the complexes, the structure of complex 8 is most unlike the other three structures because of the methyl groups possessing differences in terms of electronic effect and steric influence. In complex 8, the NHC ligand IMe has a trans-S arrangement instead of a vacancy. Both the 2-aminothiophenolate dianion bidentate ligand and imidazole ring have a twist of 90° in comparison to the three structures above. The Fe–CO distances in precursors 2, 3, in the range of 1.758(11)–1.861(11) Å (Table 1) are longer than 6, 7, in the range of 1.737(7)–1.774(7) Å (Tables S4 and S5†), on one hand due to the enhanced Fe–CO π-back-donation caused by strong electron-donor dianion 2-aminothiophenolate, which cannot be reflected by IR spectra.^36–38 Besides, the additional CO group also contributes to longer distances. Also, the Fe–CO bond wavelengths of 1 and 4 in the range of 1.794(3)–1.859(3) Å are respectively longer than those of 5 and 8 1.730(5)–1.783(5) Å.²⁹

Fig. 6 Molecular structures of 5–8 with 30% probability level ellipsoids. Hydrogen atoms are omitted for clarity, except the ones in the imidazole to distinguish saturated and unsaturated complexes.

Here, Addison's τ value is used as an index to study the geometry. The τ value of 0 means the geometry is a square pyramid and the τ value of 1 means a trigonal bipyramid.^34,39 Of all the compounds, the geometry of complex 8 is closest to a square pyramid with a τ value of 0.058 while compounds 5–7 are hybrids of trigonal bipyramidal and square pyramidal geometries. All τ values have been listed in Table 2.

Table 2 τ values for complexes 5–8

Compound	τ value
(5) Fe(CO)2SIMes(NS)	0.535
(6) Fe(CO)2IPr(NS)	0.433
(7) Fe(CO)2SIPr(NS)	0.402
(8) Fe(CO)2IMe(NS)	0.058

---

Electrochemical study on complexes 5–8

The electrochemical behaviors of complexes 5–8 were studied by cyclic voltammetry (CV) in CH₃CN in the presence of Bu₄NPF₆ as a supporting electrolyte, with the goal of evaluating its redox properties, as the Fe atom plays an important role in the formation of Fe–H intermediate and the following evolution of H₂ in Hmd. Many electrochemical studies about [FeFe]-hydrogenase model complexes have been reported,^40–42 but few investigations of [Fe]-hydrogenase have been reported, especially pentacoordinate complexes.^43,44 The anodic peak potential, E_pa and the cathodic E_pc value (referenced to Fc/Fc⁺ = 0.00 V) of the complexes are shown in Fig. 7 and Table 3. They were started from around rest potentials and went forward in the cathode way. All the complexes showed a quasi-reversible (5, 8) or irreversible (6, 7) reduction peak in the range of −1.84 to −1.86 V and, moreover, compounds 5–7 showed another irreversible reduction peak from −2.26 to −2.30 V, close to the peak value of Fe(CO)₂(NS)IMes.²⁰ The first reduction events are ascribed to a one-electron reduction Fe^II + e⁻ → Fe^I by comparison with the electrochemical studies of the [FeFe]-hydrogenase models.⁴⁵ The second reduction events of 5–7 are assigned to the couple Fe^I + e⁻ → Fe⁰. The second reduction peak of 8 was not observed within the scanned range. The two reduction peaks of 6 shift to a more negative potential with the value of 20–40 mV in contrast to the values of 7. The slight differences demonstrate that complex 7 is easier to be reduced, indicating that the saturated and unsaturated IPr groups have different effects on the electron density of Fe core of 6 and 7, the same situation as 5 and its unsaturated complex Fe(CO)₂(NS)IMes.²⁰ The IPr and IMes groups have a stronger donor property than their saturated forms SIPr and SIMes respectively consistent with the IR values.

Fig. 7 Cyclic voltammograms of complexes 5–8 (2.5 mM) in 0.1 M nBu₄NPF₆/MeCN at a scan rate of 50 mV s⁻¹.

Table 3 Redox potentials of complexes 5–8 in acetonitrile solution

Complexes	Epc/V vs. Fc/Fc^+ Fe^II− → Fe^I, Fe^I → Fe^0	Epa/V vs. Fc/Fc^+ Fe^0− → Fe^I, Fe^I → Fe^II
5	−1.86, −2.29	0.27, 0.50
6	−1.86, −2.30	0.17, 0.53
7	−1.84, −2.26	0.24
8	−1.85	0.00, 0.27

---

CO uptake and protonation properties

Inspired by the fact that Hmd could combine with an external CO molecule to generate a new facial compound with three carbonyls, we tested complexes 5, 6 and 7 at a CO atmosphere of 1 bar. It turned out that none of them could take up CO without further assistance. Like the iron dicarbonyl complexes with IMe, IMes and PCy₃ as the ligands reported in our previous work,^34,35 complex 5 showed the tendency to take up CO and gradually generated a new mer-tricarbonyl complex with the assistance of strong acid HBF₄ in acetone, along with the color change from dark brown to orange. The pattern of ν(CO) bands at 2102(w) cm⁻¹, 2052(s) cm⁻¹, 2030(m) cm⁻¹ (Fig. 8a), demonstrates the new tricarbonyl species is mer-[Fe(CO)₃SIMes(H-NS)]⁺. The deprotonation occurred with the addition of t-BuOK or Et₃N along with the drop of CO and the regeneration of complex 5.

Fig. 8 Protonation/deprotonation of complexes 5 (a), 6 (b) and 7 (c) in the presence of external CO in acetone. Three red IR spectra demonstrate the generation of new tricarbonyl compound [Fe(CO)₃SIMes(H-NS)]⁺ (a), protonated [Fe(CO)₂IPr(H-NS)]⁺ (b) and protonated [Fe(CO)₂SIPr(H-NS)]⁺ (c) respectively.

The situation was somewhat different when it came to complex 6 Fe(CO)₂IPr(NS) or 7 Fe(CO)₂SIPr(NS) with two isopropyl groups flanking the benzene ring instead of methyl groups. With the addition of HBF₄ under a CO atmosphere, we observed the gradual disappearance of previous IR spectra and the appearance of two new ν(CO) bands at 2027, 1973 cm⁻¹ to complex 6 (Fig. 8b) and 2027, 1974 cm⁻¹ to complex 7 (Fig. 8c) both with the color changing from dark red to light red. As far as we know, the steric influence from two isopropyl groups hampers the attack of external CO resulting with a product of protonated [Fe(CO)₂IPr(H-NS)]⁺ and [Fe(CO)₂SIPr(H-NS)]⁺ as shown by the following DFT calculation. The protonation of complex 5 makes it possible to bind CO due to the decreased electron concentration while the enhanced steric influence from 2,6-diisopropylphenyl groups probably hamper the attack of CO to complex 6 and 7. The deprotonation occurred with the addition of t-BuOK or Et₃N leading to the regeneration of complexes 6 and 7. It should be noted that, protonation of both 6 and 7 in the absence of CO will lead to decomposition of complexes. No obvious IR change was detected for complex 8, when protonated under CO atomsphere.

Theoretical computations of the protonation process

As stated above, complexes 6 and 7 can not bind CO in the presence of HBF₄ and CO atmosphere, producing protonated species reversibly. The IR process indicated that it was a one-step process and the two prominent peaks implied the formation of one new protonated compound without the appearance of transitional species, which could be single protonated 6-NH⁺, 6-SH⁺, 7-NH⁺, 7-SH⁺ or dual protonated 6-NH⁺-SH⁺ and 7-NH⁺-SH⁺. In order to figure out the favored protonated species of 6 and 7, DFT calculations were employed. We employed BP86 functional to calculate energies in the gaseous phase,^46,47 which works appropriately for the transition metal complex system. Different solvents have different effects on the stability of the solutes. We employ a common implicit solvent model, SMD, to show the effect of solvent on energy and electron structure. The combination of BP86 functional with the SMD solvent model gives the energy change of the whole system in the presence of the solvent. Geometry optimization of each system was done and the optimized system didn't show any sign of imaginary frequency, indicating the minimum point structure was obtained after optimization. Simple mathematic relations of energy are shown in ESI.†

With the process of geometry optimization, the most stable configurations were obtained for original complexes 6 and 7 and their corresponding different protonated species, as shown in Fig. S1 and S2† and detailed atomic coordinates in Tables S9–S16.† As shown in Tables S7 and S8,† complexes 6 and 7 give the same conclusions regardless of small differences in data. In the gas phase (Fig. S3, Tables S7 and S8†), the N atom is more prone to protonize than the S atom i.e. forming 6-NH⁺ and 7-NH⁺ because it releases more energy to get a more stable stage than the protonation of the S atom thermodynamically. Both N and S have lone electron pair and show electronegativity thus N and S are both prone to get protonated and release energy. While N atom has a higher electronegativity, leading to the priority to be protonated. The S protonation is prone to occur afterwards, generating 6-NH⁺-SH⁺ and 7-NH⁺-SH⁺, which means a dual protonation process occurred according to the calculation. However, when we took solvent effect into consideration, acetone in our case, we obtained a different conclusion as shown in Fig. 9. According to variation of free energy in solution phase, the protonation process only happens to the N atom forming 6-NH⁺ and 7-NH⁺, neither to S atom, nor to both atoms because of the solvent effect. The variation of free energy of 6-NH⁺ (−40.30 kJ mol⁻¹) and 7-NH⁺ (−35.06 kJ mol⁻¹) in acetone reflect they are thermodynamically favoured than S protonation and dual protonation. The released energy due to the combination of the proton and the S atom cannot make up the energy consumption of desolvation of protons. In the case of the N atom, the released energy of protonation can make up the energy consumption of desolvation of protons. Under either gas phase or acetone, the difference of protonation energy between complex 6 and 7 is quite small. The difference of IPr ligand and SIPr ligand doesn't have a significant impact on the process of protonation process. On one hand, the difference between saturated ligand and unsaturated ligand is small; on the other hand, the ligands are far from the protonated region.

Fig. 9 The free energy comparison between possible protonated products in acetone.

Conclusions

In summary, we have provided four novel pentacoordinate models 5–8 of Hmd active site and two corresponding precursor compounds 2 and 3 by substituting CO and subsequent iodine ligands with stronger electron-donating NHC and dianion 2-aminothiophenolate ligands successively. They not only reproduced the similar first coordination sphere, but also showed exactly the same coordination number and ligating atoms that directly coordinate with Fe. Dicarbonyl complexes 5–8 exhibited similar infrared ν(CO) patterns and absorption frequencies as the active site of Hmd, among which complex 7 has the most similar bonds with Hmd, at 2005 and 1944 cm⁻¹. X-ray diffraction study has shown that (i) complexes 5–7 are hybrids of trigonal bipyramidal and square pyramidal geometries, while the geometry of complex 8 is closest to a square pyramid as a comprehensive result of electronic and steric effects, (ii) the Fe–CO bond lengths of precursors 1–4 are longer than those of iodine-substituted complexes 5–8 demonstrating that strong σ-donor dianion 2-aminothiophenolate enhances the Fe–CO π-back-donation. The electrochemical studies show that the reduction potentials of complexes 5 and 7 are more positive than their corresponding unsaturated forms Fe(CO)₂(NS)IMes and complex 6 because unsaturated IPr and IMes groups have a stronger donor property. It is worth mentioning that complex 5 easily underwent simultaneous reversible protonation process and combination of CO with the assistance of HBF₄ and t-BuOK. In addition, complexes 6 and 7 showed interesting reversible protonation and deprotonation property, generating a protonated species [Fe(CO)₂IPr(H-NS)]⁺ and [Fe(CO)₂SIPr(H-NS)]⁺. DFT calculations indicated that the N atom of the NS ligand is the protonated point thermodynamically, forming 6-NH⁺ and 7-NH⁺ in acetone. As an internal base, the NS ligand is possible to work as a proton acceptor to accept the proton in the heterolytic cleavage of dihydrogen.

Experimental section

Materials and instruments

All reactions and operations were carried out on a double manifold Schlenk vacuum line under N₂ atmosphere with rigorous exclusion of light. Solvents were dried and distilled prior to use according to the standard methods. The purified solvents were stored with molecular sieves under N₂ for no more than 1 week before use. Complexes FeI₂(CO)₄, 1 and 4 were prepared according to literature procedures.²⁹ The following materials were of reagent grade and used as purchased from Sigma-Aldrich: 1,3-bis(2,4,6-trimethylphenyl)imidazolium chloride, potassium tert-butoxide, 2-aminothiophenol and nBu₄NPF₆. The Fe(CO)₅ was obtained as a gift from Jiangsu Tianyi Ultra-fine Metal Powder Co., Ltd (China).

The NMR spectra were measured on a Bruker AVANCE III 400 MHz NMR spectrometer. ¹H NMR shifts were referenced to residual solvent resonances, according to literature values. Solution IR spectra were recorded on a Shimadzu FTIR-8400 spectrometer using 0.1 mm KBr sealed cells. Elemental analysis was carried out on a Heraeus CHN-Rapid, fully automatic elemental analyzer with TCD detection, type: TMT CHN, BESTELL-NR 2215001.

Synthesis of FeI₂(CO)₃(IPr) (2)

The synthetic procedures of FeI₂(CO)₃IPr (2) and FeI₂(CO)₃SIPr (3) are similar with those of FeI₂(CO)₃IMes, FeI₂(CO)₃SIMes and FeI₂(CO)₃IMe.²⁹ 1,3-Bis(2,6-diisopropylphenyl)imidazolium chloride (424.5 mg, 1 mmol),⁴⁸ potassium tert-butoxide (224 mg, 2 mmol), and THF (25 mL) were added to a 50 mL flask and stirred at room temperature for 1 h. After the solvent was removed under reduced pressure, the residual solid was extracted with hexane (30 mL). The resulting suspension was filtered through celite and the filtrate of free carbene was added to a hexane (20 mL) solution of FeI₂(CO)₄ (421.6 mg, 1 mmol). The reaction mixture was stirred for 1 h at room temperature in the dark. The precipitated solid product was collected by filtration, washed with two 25 mL portions of hexane, and dried under vacuum to give a red-brown powder. Yield: 391 mg (50%). ¹H NMR (400 MHz, acetone-d₆): δ = 7.88 (s, 2H, NCH); 7.69 (t, 2H, p-Ar-H); 7.53 (d, 4H, m-Ar-H); 1.48 (d, 12H, CHMe₂), 1.26 (s, 1H, CHMe₂); 1.28 (s, 1H, CHMe₂); 1.32 (s, 1H, CHMe₂); 1.34 (s, 1H, CHMe₂); 1.13 (d, 12H, CHMe₂). IR (CH₂Cl₂, cm⁻¹): 2087(w), 2039(s), 2021 (sh). Elemental analysis, found (calculated)% for C₃₀H₃₆O₃N₂FeI₂: C, 45.48 (46.07); H, 4.02 (4.65); N, 2.98 (3.58).

Synthesis of FeI₂(CO)₃(SIPr) (3)

As described above, the 1,3-bis(2,6-diisopropylphenyl) imidazolinium chloride (426.5 mg, 1 mmol) salt⁴⁸ was deprotonated with potassium tert-butoxide (224 mg, 2 mmol), and the free carbene was added to a hexane (20 mL) solution of FeI₂(CO)₄ (421.6 mg, 1 mmol). Finally, a dark red powder was collected. Crude yield: 407 mg (52%). ¹H NMR (400 MHz, chloroform-d₆): δ = 7.85 (s, 4H, NCH); 7.70 (t, 2H, p-Ar-H); 7.52 (d, 4H, o-Ar-H); 1.48 (d, 12H, CHMe₂), 1.24 (s, 1H, CHMe₂); 1.26 (s, 1H, CHMe₂); 1.33 (s, 1H, CHMe₂); 1.35 (s, 1H, CHMe₂); 1.21 (d, 12H, CHMe₂). IR (CH₂Cl₂, cm⁻¹): 2086(w), 2038(s), 2019 (sh). Elemental analysis, found (calculated)% for C₃₀H₃₈O₃N₂FeI₂: C, 45.60 (45.96); H, 4.13 (4.89); N, 3.12 (3.57).

Synthesis of Fe(CO)₂(SIMes)(NS) (5)

2 equiv. of potassium tert-butoxide (112 mg, 1.0 mmol) in MeOH was added dropwise to a MeOH (10 mL) solution of 2-aminothiophenol (54 μL, 0. 5 mmol) under N₂ atmosphere, followed by vigorous stirring for an hour. The solvent was removed under vacuum and the residual dark solids were washed with diethyl ether (20 mL). The residue was dried under vacuum, dissolved in methanol (10 mL), and then added into a THF (10 mL) solution of FeI₂(CO)₃SIMes (280 mg, 0.4 mmol) under N₂ atmosphere. The reaction was monitored by IR spectroscopy to completion. The mixed solvent was removed under vacuum and the residual dark solids were extracted with diethyl ether (30 mL). The ether was removed under vacuum to give a dark red powder. Yield: 155 mg (71.5%). The single crystals suitable for X-ray diffraction analysis were grown by slow evaporation of a diethyl ether solution of complex 5 at 4 °C. ¹H NMR (400 MHz, CDCl₃): δ = 8.80 (s, 1H, FeNH), 7.65 (s, 2H, NCH₂), 7.00 (s, 2H, NCH₂), 6.75 (s, 4H, m-Mes), 2.44 (s, 6H, p-Mes), 2.22 (s, 12H, o-Mes). IR (CH₂Cl₂, cm⁻¹): 1996(s), 1937(s). Elemental analysis, found (calculated)% for C₂₉H₃₁O₂N₃SFe: C, 64.50 (64.33); H, 5.61 (5.73); N, 7.69 (7.76); S, 5.84 (5.91).

Synthesis of Fe(CO)₂(IPr)(NS) (6)

Similarly, when complex 2 (313 mg, 0.4 mmol) was employed in place of complex 1, Hmd model complex 6 (182 mg, 73%) was obtained as a dark brown solid. The crystals were obtained by the slow evaporation of a diethyl ether solution of complex 6 at 4 °C. ¹H NMR (400 MHz, CDCl₃): δ = 8.97 (s, 1H, FeNH), 7.65 (s, 1H, NCH), 7.32 (s, 1H, NCH), 6.34–7.0 (m, 10H, C₆H₃, C₆H₃, C₆H₄), 2.7–3.7 (m, 4H, CHMe₂), 1.32 (d, J = 6.8 Hz, 12H, CHMe₂), 1.10 (d, J = 6.8 Hz, 12H, CHMe₂). IR (CH₂Cl₂, cm⁻¹): 1997(s), 1937(s). Elemental analysis, found (calculated)% for C₃₅H₄₁O₂N₃SFe: C, 66.89 (67.42); H, 6.67 (6.58); N, 6.59 (6.74); S, 5.33 (5.14).

Synthesis of Fe(CO)₂(SIPr)(NS) (7)

Similarly, when complex 3 (314 mg, 0.4 mmol) was employed in place of complex 1, Hmd model complex 7 (173 mg, 69%) was obtained as a dark brown solid. The crystals were obtained by the slow evaporation of a diethyl ether solution of complex 7 at 4 °C. ¹H NMR (400 MHz, CDCl₃): δ = 9.01 (s, 1H, FeNH), 7.64 (s, 2H, NCH), 7.30 (s, 2H, NCH), 6.29–6.9 (m, 10H, C₆H₃, C₆H₃, C₆H₄), 3.3 (m, 4H, CHMe₂), 1.27 (d, J = 6.8 Hz, 12H, CHMe₂), 1.09 (d, J = 6.8 Hz, 12H, CHMe₂). IR (CH₂Cl₂, cm⁻¹): 2005(s), 1944(s). Elemental analysis, found (calculated)% for C₃₅H₄₃O₂N₃SFe: C, 66.97 (67.20); H, 7.01 (6.88); N, 6.63 (6.72); S, 5.29 (5.12).

Synthesis of Fe(CO)₂(IMe)(NS) (8)

Similarly, when complex 4 (196 mg, 0.4 mmol) was employed in place of complex 1, Hmd complex 8 (80 mg, 60%) was obtained as a dark brown solid. The crystals were obtained by the slow evaporation of a diethyl ether solution of complex 8 at 4 °C. ¹H NMR (400 MHz, CDCl₃): δ = 10.5 (s, 1H, FeNH), 8.11 (s, 1H, NCH), 7.92 (s, 1H, NCH), 6.52–7.07 (m, 4H, C₆H₄), 4.00 (s, 6H, NMe). IR (CH₂Cl₂, cm⁻¹): 2005(s), 1941(s). Elemental analysis, found (calculated)% for C₁₃H₁₃O₂N₃SFe: C, 46.89 (47.13); H, 4.03 (3.93); N, 12.47 (12.69); S, 9.15 (9.67).

X-ray structure determinations of complexes 2, 3, 5–8

The single-crystal X-ray diffraction data were collected with a Rigaku MM-007 diffractometer equipped with a Saturn 724CCD. Data were collected at 293 K or 113 K using a confocal monochromator with Mo-Kα radiation (λ = 0.71073 Å). Data collection, reduction, and absorption correction were performed with the CRYSTALCLEAR program 25. The structure was determined by direct methods using the Diamond 3.2 program. Details of crystal data, data collections and structure refinements are summarized in Tables S1–S6.†

Electrochemical experiments

Cyclic voltammograms were obtained in a three-electrode cell under N₂ using a CHI 660B electrochemical workstation. The working electrode was a glassy carbon disk (diameter 3 mm) polished with 3 and 1 μm diamond pastes and sonicated in ion-free water for 20 min prior to use. The reference electrode was a non-aqueous Ag/Ag+ (in a CH₃CN solution of 0.01 M AgNO₃/0.1 M n-Bu₄NPF₆) electrode and the counter electrode was platinum wire. A solution of 0.1 M n-Bu₄NPF₆ in CH₃CN was used as supporting electrolyte, which was degassed by bubbling with dry N₂ for 10 min before measurement. Ferrocene was used as an internal standard under the same measuring conditions and all potentials were referenced to the Cp₂Fe^+/0 couple at 0 V.

CO uptake of complex 5 in the presence of HBF₄

HBF₄ was dropwise added to an acetone (10 mL) solution of complex 5 (16 mg, 0.03 mmol) under a CO atmosphere. The reaction was monitored with IR spectroscopy to indicate the generation of [Fe(CO)₂IPr(H-NS)]⁺ (2102(w), 2052(s), 2030(m) cm⁻¹) along with color changing from dark brown to orange within half an hour. The addition of t-BuOK to [5(CO)-H]⁺BF₄⁻ led to the liberation of CO and regeneration of complex 5, as monitored by IR spectroscopy (see Fig. 8a).

CO uptake of complex 6 in the presence of HBF₄

HBF₄ was dropwise added to an acetone (10 mL) solution of complex 6 (19 mg, 0.03 mmol) under a CO atmosphere. The reaction was monitored with IR spectroscopy to indicate protonation and the generation of [6-H]⁺BF₄⁻ (2027, 1973 cm⁻¹) along with color changing from dark red to light red within half an hour. The addition of t-BuOK to [6-H]⁺BF₄⁻ led to the deprotonation process and regeneration of complex 6, as monitored by IR spectroscopy (see Fig. 8b).

CO uptake of complex 7 in the presence of HBF₄

HBF₄ was dropwise added to an acetone (10 mL) solution of complex 7 (19 mg, 0.03 mmol) under a CO atmosphere. The reaction was monitored with IR spectroscopy to indicate protonation and the generation of [7-H]⁺BF₄⁻ (2027, 1974 cm⁻¹) along with color changing from dark red to light red within half an hour. The addition of t-BuOK to [7-H]⁺BF₄⁻ led to the deprotonation process and regeneration of complex 7, as monitored by IR spectroscopy (see Fig. 8c).

Theoretical details

The pseudopotential base set of Hay and Wadt (LANL2DZ) was used on the iron atom.^49,50 and other atoms are calculated using the base set of 6-31G*.^51,52 Geometry optimization of each system was done and vibration analysis was used to obtain thermal correction to free energy in gas phase at 1 atm, 298.15 K. The electronic energy was calculated with high-quality def2-TZVP base set.⁵³ In order to explore protonation process in acetone, SMD solvent base set⁵⁴ was implemented to calculate solvation free energy of each system in acetone. All the calculations ware implemented with Gaussian 09.01.⁵⁵ To get the variation of free energy in the protonation process, corresponding one or two proton's free energy was deducted from the protonated species. Free energy of proton in the gas phase is −26.31 kJ mol⁻¹ at 1 atm, 298.15 K.⁵⁶ While the free energy in acetone doesn't have corresponding experimental value, so the theoretical value −1060.5 kJ mol⁻¹ (ref. 57) was used, plus free energy of proton in the gas phase, getting the free energy of proton in acetone −1086.81 kJ mol⁻¹.

Acknowledgements

We are grateful to the National Natural Science Foundation of China (21103121, 21276187), the Tianjin Municipal Natural Science Foundation (16JCYBJC20800) and the Tianjin science and technology innovation platform program (14TXGCCX00017) for financial support of this work.

Notes and references

1. J. C. Fontecilla-Camps, A. Volbeda, C. Cavazza and Y. Nicolet, Chem. Rev., 2007, 107, 4273–4303.
2. M. W. Adams and E. I. Stiefel, Science, 1998, 282, 1842–1843.
3. R. Cammack, Nature, 1999, 397, 214–215.
4. S. Dey, P. K. Das and A. Dey, Coord. Chem. Rev., 2013, 257, 42–63.
5. J. A. Wright, P. J. Turrell and C. J. Pickett, Organometallics, 2010, 29, 6146–6156.
6. S. Shima, O. Pilak, S. Vogt, M. Schick, M. S. Stagni, W. Meyer-Klaucke, E. Warkentin, R. K. Thauer and U. Ermler, Science, 2008, 321, 572–575.
7. S. Shima, E. J. Lyon, R. K. Thauer, B. Mienert and E. Bill, J. Am. Chem. Soc., 2005, 127, 10430–10435.
8. M. Salomone-Stagni, F. Stellato, C. M. Whaley, S. Vogt, S. Morante, S. Shima, T. B. Rauchfuss and W. Meyer-Klaucke, Dalton Trans., 2010, 39, 3057–3064.
9. E. J. Lyon, S. Shima, R. Boecher, R. K. Thauer, F. W. Grevels, E. Bill, W. Roseboom and S. P. Albracht, J. Am. Chem. Soc., 2004, 126, 14239–14248.
10. T. Hiromoto, K. Ataka, O. Pilak, S. Vogt, M. S. Stagni, W. Meyer-Klaucke, E. Warkentin, R. K. Thauer, S. Shima and U. Ermler, FEBS Lett., 2009, 583, 585–590.
11. T. Hiromoto, E. Warkentin, J. Moll, U. Ermler and S. Shima, Angew. Chem., Int. Ed., 2009, 48, 6457–6460.
12. T. Zhang, L. Sheng, Q. Yang, S. Jiang, Y. Wang, C. Jin and B. Li, Chin. J. Catal., 2015, 36, 2011–2019.
13. Y. Guo, H. Wang, Y. Xiao, S. Vogt, R. K. Thauer, S. Shima, P. I. Volkers, T. B. Rauchfuss, V. Pelmenschikov, D. A. Case, E. E. Alp, W. Sturhahn, Y. Yoda and S. P. Cramer, Inorg. Chem., 2008, 47, 3969–3977.
14. D. Chen, R. Scopelliti and X. Hu, J. Am. Chem. Soc., 2010, 132, 928–929.
15. L. C. Song, F. Q. Hu, G. Y. Zhao, J. W. Zhang and W. W. Zhang, Organometallics, 2014, 33, 6614–6622.
16. L. C. Song, M. Cao and Y. X. Wang, Dalton Trans., 2015, 44, 6797–6808.
17. L. C. Song, F. Q. Hu, M. M. Wang, Z. J. Xie, K. K. Xu and H. B. Song, Dalton Trans., 2014, 43, 8062–8071.
18. L. C. Song, K. K. Xu, X. F. Han and J. W. Zhang, Inorg. Chem., 2016, 55, 1258–1269.
19. S. Shima, D. Chen, T. Xu, M. D. Wodrich, T. Fujishiro, K. M. Schultz, J. Kahnt, K. Ataka and X. Hu, Nat. Chem., 2015, 7, 995–1002.
20. T. Xu, C. J. Yin, M. D. Wodrich, S. Mazza, K. M. Schultz, R. Scopelliti and X. Hu, J. Am. Chem. Soc., 2016, 138, 3270–3273.
21. G. Durgaprasad, Z. L. Xie and M. J. Rose, Inorg. Chem., 2016, 55, 386–389.
22. X. Yang and M. B. Hall, J. Am. Chem. Soc., 2009, 131, 10901–10908.
23. B. Hu, D. Chen and X. Hu, Chem.–Eur. J., 2014, 20, 1677–1682.
24. H. Tamura, M. Salomone-Stagni, T. Fujishiro, E. Warkentin, W. Meyer-Klaucke, U. Ermler and S. Shima, Angew. Chem., Int. Ed., 2013, 52, 9656–9659.
25. K. M. Schultz, D. Chen and X. Hu, Chem.–Asian J., 2013, 8, 1068–1075.
26. A. M. Royer, T. B. Rauchfuss and D. L. Gray, Organometallics, 2009, 28, 3618–3620.
27. P. J. Turrell, J. A. Wright, J. N. Peck, V. S. Oganesyan and C. J. Pickett, Angew. Chem., Int. Ed., 2010, 49, 7508–7511.
28. D. Chen, R. Scopelliti and X. Hu, Angew. Chem., Int. Ed., 2010, 49, 7512–7515.
29. B. Li, T. Liu, C. V. Popescu, A. Bilko and M. Y. Darensbourg, Inorg. Chem., 2009, 48, 11283–11289.
30. D. Chen, R. Scopelliti and X. Hu, Angew. Chem., Int. Ed., 2011, 50, 5671–5673.
31. A. P. Marchenko, H. N. Koidan, A. N. Hurieva, I. I. Pervak, S. V. Shishkina, O. V. Shishkin and A. N. Kostyuk, Eur. J. Org. Chem., 2012, 2012, 4018–4033.
32. M. J. Ingleson and R. A. Layfield, Chem. Commun., 2012, 48, 3579–3589.
33. T. Liu, S. Chen, M. J. O'Hagan, M. Rakowski DuBois, R. M. Bullock and D. L. DuBois, J. Am. Chem. Soc., 2012, 134, 6257–6272.
34. T. Liu, B. Li, C. V. Popescu, A. Bilko, L. M. Perez, M. B. Hall and M. Y. Darensbourg, Chem.–Eur. J., 2010, 16, 3083–3089.
35. S. Jiang, T. Zhang, X. Zhang, G. Zhang and B. Li, Dalton Trans., 2015, 44, 16708–16712.
36. Y. Wang, T. Zhang, B. Li, S. Jiang and L. Sheng, RSC Adv., 2015, 5, 29022–29031.
37. D. Morvan, J.-F. Capon, F. Gloaguen, A. Le Goff, M. Marchivie, F. Michaud, P. Schollhammer, J. Talarmin, J.-J. Yaouanc, R. Pichon and N. Kervarec, Organometallics, 2007, 26, 2042–2052.
38. P. Buchgraber, L. Toupet and V. Guerchais, Organometallics, 2003, 22, 5144–5147.
39. A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
40. G. A. Felton, C. A. Mebi, B. J. Petro, A. K. Vannucci, D. H. Evans, R. S. Glass and D. L. Lichtenberger, J. Organomet. Chem., 2009, 694, 2681–2699.
41. S. J. Borg, T. Behrsing, S. P. Best, M. Razavet, X. Liu and C. J. Pickett, J. Am. Chem. Soc., 2004, 126, 16988–16999.
42. H. H. Cui, N. N. Wu, J. Y. Wang, M. Q. Hu, H. M. Wen and C. N. Chen, J. Organomet. Chem., 2014, 767, 46–53.
43. P. J. Turrell, A. D. Hill, S. K. Ibrahim, J. A. Wright and C. J. Pickett, Dalton Trans., 2013, 42, 8140–8146.
44. S. Kaur Ghumaan, L. Schwartz, R. Lomoth, M. Stein and S. Ott, Angew. Chem., Int. Ed., 2010, 49, 8033–8036.
45. M. Wang, L. Chen and L. Sun, Energy Environ. Sci., 2012, 5, 6763–6778.
46. J. P. Perdew, Phys. Rev. B: Condens. Matter Mater. Phys., 1986, 33, 8822–8824.
47. A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100.
48. S. Zhu, R. Liang, L. Chen, C. Wang, Y. Ren and H. Jiang, Tetrahedron Lett., 2012, 53, 815–818.
49. P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
50. W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985, 82, 284–298.
51. P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213–222.
52. W. J. Hehre, J. Chem. Phys., 1972, 56, 2257–2261.
53. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
54. A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 6378–6396.
55. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2009.
56. J. J. Fifen, Z. Dhaouadi and M. Nsangou, J. Phys. Chem. A, 2014, 118, 11090–11097.
57. Z. Marković, J. Tošović, D. Milenković and S. Marković, Comput. Theor. Chem., 2016, 1077, 11–17.

---

Footnote

† Electronic supplementary information (ESI) available: Tables giving crystallographic data for complexes 2, 3, 5–8 and selected bond lengths and bond angles for 5–8. CCDC 1429929, 1429930, 1429932–1429935. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra18628a

---