Article pubs.acs.org/Organometallics
Reactivity of a (Bis-NHC)tricarbonylruthenium(0) Complex with Methyl Triflate and Methyl Iodide. Formation of Methyl- and Acetylruthenium(II) Derivatives: Experimental Results and Mechanistic DFT Calculations Javier A. Cabeza,*,† Marina Damonte,† Pablo García-Á lvarez,† and Enrique Pérez-Carreño‡ †
Departamento de Química Orgánica e Inorgánica-IUQOEM, Universidad de Oviedo-CSIC, E-33071 Oviedo, Spain Departamento de Química Física y Analítica, Universidad de Oviedo, E-33071 Oviedo, Spain
‡
S Supporting Information *
ABSTRACT: The ruthenium(0) complex [Ru{κ 2 C 2 -MeIm(CH2)3ImMe}(CO)3] (1), MeIm(CH2)3ImMe = 1,3-bis(3-methylimidazol-2-yliden-1-yl)propane, which contains a chelating bis(N-heterocyclic carbene) ligand, reacts with MeOTf at room temperature to give the ionic ruthenium(II) methyl derivative [RuMe{κ2C2-MeIm(CH2)3ImMe}(CO)3]OTf ([2]OTf), whereas an analogous reaction of 1 with MeI renders the neutral ruthenium(II) acetyl derivative [RuI{C(O)Me}{κ2C2MeIm(CH2)3ImMe}(CO)2] (3), in which the iodide and acetyl ligands occupy mutually trans coordination sites. The fact that [2]OTf reacts with [Et4N]I to give 3 evidences the participation of the cationic species 2+ in the synthesis of 3. The mechanisms of these reactions in THF solution have been modeled by DFT calculations. They have shown that 2+ can be made from complex 1 and either MeOTf or MeI. In both cases, two plausible reaction pathways have been identified. They start with a rate-determining SN2 substitution process in which the metal atom of 1 attacks the C atom of MeI or MeOTf, displacing the corresponding anion to give, depending on how compound 1 approaches MeI or MeOTf, 2+ or a less stable isomeric species 2′+ that is easily transformed into 2+. A subsequent CO migratory insertion in 2+ leads to an unsaturated (pentacoordinated) acetyl intermediate that easily adds the iodide anion to end in 3. DFT calculations have also shown that the reaction of 1 with MeOTf to give [2]OTf is thermodynamically more favorable than that of 1 with MeI to give [2]I due to the resonance stabilization and greater solvation energy of the triflate anion. These two effects are also responsible for the fact that the incorporation of the triflate anion with 2+ to give a putative triflate complex analogous to 3 is thermodynamically disfavored, whereas the incorporation of the iodide anion with 2+ to give 3 is thermodynamically favored.
■
[Fe(κC-R2Im)(CO)4] (R = Me, Mes, 2,6-C6H3iPr2) and [Fe(κC-Mes2Im)2(CO)3], have just been reported,12 but analogous osmium(0) NHC complexes still remain unknown. As all these mononuclear NHC-containing group-8 metal(0) complexes have been described very recently, they have hitherto been implicated in very few reactivity studies. Wang’s group13 has reported that, as a consequence of the high basicity of the NHC ligands, the complexes [Ru(κC-RMeIm)(CO)4] (R = Ph, tBu) are prone to undergo easy intramolecular C−H bond-activation processes that lead to cyclometalated ruthenium(II) derivatives;14 Whittlesey’s group7 has shown that [Ru(κC-R2Im)2(CO)3] (R = iPr; R2Im = Et2ImMe2, i Pr2ImMe2) complexes react easily with oxygen to give the corresponding ruthenium(II) carbonato derivatives [Ru(CO3)(κC-R2Im)2(CO)2]; Royo’s group12 has reported that [Fe(κCR2Im)(CO)4] (R = Me, Mes, 2,6-C6H3iPr2) complexes are useful catalyst precursors for the hydrosilylation of benzaldehyde; and we have recently reported the easy abstraction of a hydride anion from an alkyl C−H bond of the trimethylene-
INTRODUCTION The intensive research activity carried out in the past decade on the coordination chemistry of N-heterocyclic carbenes (NHCs)1−3 has also included transition metal carbonyl cluster complexes.3 In this field, it has been shown that the reactions of [Ru3(CO)12] with imidazole-derived NHCs (R2Im) lead only to simple trinuclear monosubstituted derivatives of the type [Ru3(κC-R2Im)(CO)11] (R = Me,4 Mes,4 tBu,5 Ad5) when they involve equimolar amounts of the reagents (the products may contain the R2Im ligand coordinated in normal4 or abnormal5 mode, depending on the volume of the R group). In these reactions, the use of greater amounts of the NHC reagent leads to mixtures of products or to mononuclear ruthenium(0) derivatives of the types [Ru(κC-R2Im)(CO)4] (R = Mes, 2,6C6H3iPr2)6 or [Ru(κC-R2Im)2(CO)3] (R = iPr; R2Im = Et2ImMe2, iPr2ImMe2),7 depending on the ratio of the reactants.7 Related complexes containing chelating bidentate ligands of the type [Ru(κ2-NHC−X)(CO)3] (X = phosphine fragment,8,9 NHC fragment10,11) have been prepared by treating [Ru3(CO)12] with the appropriate NHC−X ligands using a 1:3 ratio of the reagents. Regarding the remaining group-8 metals, the first mononuclear iron(0) NHC derivatives, © XXXX American Chemical Society
Received: June 10, 2013
A
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
linked bis(NHC) ligand of [RuMe{κ2C2-MeIm(CH2)3ImMe}(CO)3] (1; MeIm(CH2)3ImMe = 1,3-bis(3-methylimidazol-2yliden-1-yl)propane).11 We now report that the reactions of compound 1 with methyl triflate and methyl iodide selectively lead to methyl- and acetylruthenium(II) derivatives, respectively, and we complement these experimental results with theoretical DFT studies that have shed light on the mechanisms of these reactions. We chose complex 1 as a convenient stating material of the type [Ru(κC-R2Im)2(CO)3] because (a) it can be easily made from [Ru3(CO)12] and MeIm(CH2)3ImMe (Scheme 1),11 (b) its
Scheme 2
Scheme 1
The solid-state structures of [2]OTf and 3·CH2Cl2 were determined by X-ray crystallography. Both complexes are octahedral ruthenium(II) complexes, but, while 2+ (Figure 1, strong σ-donating bis(NHC) ligand enhances the nucleophilic character of its ruthenium(0) atom,11 and (c) its chelating bis(NHC) ligand prevents the formation of mixtures containing cis and trans isomeric products, which have been observed in related ruthenium(0) complexes containing two monodentate NHC ligands.7 It is noteworthy that no reaction of alkyl halides or pseudohalides with ruthenium(0) complexes of the type [Ru(κ2-L2)(CO)3] (L2 = chelating ligand) has been previously described, and although some reactions of methyl iodide with ruthenium(0) derivatives of the types [Ru(κP-PR3)(CO)4],15 trans-[Ru(κP-PR3)2(CO)3],15 and [Ru(κ3P3-triphos)(CO)2] (triphos = tridentate tripod16 or pincer17 P-donor ligand) have been reported, their mechanisms have not been studied by theoretical methods. The oxidative addition of methyl iodide to transition metal complexes is a fundamental reaction in organometallic chemistry that plays a key role in important catalytic processes, such as the Monsanto and Cativa industrial syntheses of acetic acid by carbonylation of methanol.18
Figure 1. XRD structure of cation 2+ in [2]OTf. Only one of the two disordered positions in which some atoms were found in the crystal is shown (see text). Only the H atoms of methyl groups are displayed.
Table 1) is a tricarbonyl cation that has three CO ligands in a fac arrangement and a methyl group cis to both NHC fragments, compound 3 (Figure 2, Table 1) is a neutral cisdicarbonyl derivative that has an acetyl group and a iodide ligand trans to each other. In both complexes, the planes of their NHC rings are not coplanar (they form a dihedral angle of 129.9° in 2+ and 124.7° in 3) in order to minimize conformational strain within the trimethylene linker of their bis(NHC) ligands while maintaining cis coordination.19 In CD2Cl2 solution at room temperature, both 2+ and 3 maintain the asymmetry observed in the solid state, since their 1 H and 13C NMR spectra contain the resonances of two inequivalent NHC fragments and three (in 2+) or two (in 3) inequivalent CO ligands. While the 1H and 13C resonances of the Ru-bound methyl of 2+ appear at 0.06 and −15.4 ppm, respectively, in their corresponding spectra, those of the acetyl methyl group of 3 are observed at 2.40 and 39.7 ppm, respectively. The higher oxidation number of the ruthenium atoms of 2+ and 3, in comparison with that of the starting complex 1, is markedly manifested by their solution (THF) IR spectra, which contain the carbonyl νCO absorptions at higher wavenumbers (2+: 2093, 2025, and 1984 cm−1; 3: 2017 and
■
RESULTS AND DISCUSSION Reactions and Product Characterizations. The reaction of compound 1 with methyl triflate in THF led to the ionic ruthenium(II) methyl derivative [RuMe{κ 2 C 2 -MeIm(CH2)3ImMe}(CO)3][OTf] ([2]OTf), which was isolated as an air-stable, yellow solid. In contrast, an analogous treatment of complex 1 with methyl iodide led to the air-stable, neutral ruthenium(II) acetyl complex [RuI{C(O)Me}{κ2C2-MeIm(CH2)3ImMe}(CO)2] (3). Both reactions were quantitative and proceeded at room temperature with excess or equimolar amounts of the appropriate methylating reagent (Scheme 2). In an attempt to prepare a neutral acetyl complex analogous to 3 but having a triflato instead of an iodido ligand, namely, [Ru(OTf){C(O)Me}{κ2C2-MeIm(CH2)3ImMe}(CO)2] (4), we heated a THF solution of [2]OTf at reflux temperature for 1 h, but no reaction was observed. However, [2]OTf reacted instantaneously at room temperature with [Et4N]I to give compound 3 quantitatively (Scheme 2). Therefore, the cationic complex 2+ does not react with the hard triflate anion, but it reacts easily with the softer iodide anion to give 3. B
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
tridentate tripod16 or pincer17 P-donor ligand) with methyl iodide do not give acetyl derivatives either, but ionic methyl derivatives of the type [Ru(Me)(κ3P3-triphos)(CO)2]I.16,17 Therefore, as all previously reported reactions of methyl iodide with ruthenium(0) carbonyl complexes end in Ru-methyl derivatives and as all of these complexes contain less σ-donating ligands than MeIm(CH2)3ImMe, the strong σ-donating character of this ligand has to be claimed as responsible for the formation of an acetyl derivative (3) as the final product of the reaction of 1 with methyl iodide. Theoretical Calculations. The above-described experimental results prompted us to undertake a DFT theoretical study in order to get an explanation for the isolation of different reaction products from the reactions of methyl triflate and methyl iodide with complex 1, namely, the cationic Ru-methyl complex 2+ in the case of methyl triflate but the neutral Ru− acetyl derivative 3 in the case of methyl iodide. As pointed out in various theoretical studies that examined the oxidative addition of methyl iodide to unsaturated rhodium(I) complexes (relevant to the Monsanto process), gas-phase quantum chemistry calculations are discouraged to model reactions that are carried out in solution and that involve ionic ligand exchange processes in which charged and neutral species coexist.20,21 Solvent effects become crucial, for instance, in stabilizing the iodide leaving group, which otherwise is a highly unstable species in the gas phase.20 On the other hand, it has also been previously shown that the use of diffuse basis sets is especially important for calculations on systems containing noncovalently bound ionic species.22 Consequently, our DFT calculations have been performed with basis sets augmented with diffusion and d,p-polarization functions for iodine and all remaining nonmetallic atoms. Although the following discussion only refers to Gibbs energies in THF solution (ΔGTHF), for comparison, both gas-phase and solution energies are given in the associated figures. Pentacoordinated ruthenium(0) complexes are saturated 18electron species that frequently act as metallic nucleophiles.11,15−17 Therefore, we computed the HOMO of compound 1 with the aim of obtaining information about its behavior as a nucleophile. Figure 3 shows that this orbital has
Table 1. Selected XRD Interatomic Distances (Å) in [2]OTf and 3·CH2Cl2 bond C1−C15 C1−Ru1 C2−N1 C2−N2 C2−Ru1 C3−N1 C4−C5 C4−N1 C5−N2 C6−C7 C6−N2 C7−C8 C8−N3 C9−C10 C9−N3 C10−N4 C11−N4 C12−N4 C12−Ru1 C13−Ru1 C14−Ru1 C15−O15 C15−Ru1 I1−Ru1 a
[2]OTf
3·CH2Cl2 1.522(7)
2.206(6)a 1.363(4) 1.363(4) 2.144(3) 1.461(5) 1.333(5) 1.385(5) 1.391(4) 1.516(5) 1.464(4) 1.510(5) 1.463(4) 1.343(5) 1.387(4) 1.379(4) 1.470(4) 1.354(4) 2.161(3) 1.935(7)a 1.928(4) 1.926(4)
1.373(6) 1.357(6) 2.160(4) 1.463(6) 1.346(7) 1.372(6) 1.382(6) 1.510(7) 1.471(6) 1.533(7) 1.466(6) 1.342(7) 1.378(6) 1.374(6) 1.470(6) 1.358(6) 2.176(4) 1.898(5) 1.904(5) 1.200(7) 2.059(4) 2.8957(4)
Average value (two disordered positions).
Figure 2. XRD structure of compound 3 in 3·CH2Cl2. Only the H atoms of methyl groups are displayed.
1952 cm−1) than those of 1 (1986, 1974, and 1852 cm−1),11 indicating a decrease of π back-donation (Ru→CO) on going from 1 to 2+ and 3. The acetyl νCO absorption of 3 is observed as a medium band at 1624 cm−1 (THF). Interestingly, while contradictory reports on the reaction of methyl iodide with trans-[Ru(κP-PPh3)2(CO)3] have been published (cis,trans,cis-[RuI(Me)(κP-PPh3)2(CO)2]15a and mer,trans-[RuI3(κP-PPh3)2(CO)]15b have been described as reaction products), it has been shown that the reactions of methyl iodide with the ruthenium(0) complexes [Ru(κPPMe3)(CO)4] and trans-[Ru(κP-PMe3)2(CO)3] do not afford acetyl derivatives but the neutral iodido-methyl products fac[RuI(Me)(κP-PMe3)(CO)3] and cis,trans,cis-[RuI(Me)(κPPMe3)2(CO)2].15c Similar reactions of the dicarbonyl ruthenium(0) complexes [Ru(κ3P3-triphos)(CO)2] (triphos =
Figure 3. Side (left) and front (right) views of the HOMO of compound 1, shown at an isosurface of 0.05.
mostly metallic character and that it is asymmetrically located in the equatorial plane of the trigonal bipyramid defined by the coordinated atoms of the ligands. Therefore, when 1 acts as a nucleophile, the electrophile can approach the metal atom laterally (between the equatorial NHC fragment and the C15O15 carbonyl ligand) or frontally (between the two equatorial CO ligands). For the mechanistic studies, we considered the two reactants separated by an infinite distance as the starting stage. For both methylating reagents, two higher-energy (due to a negative C
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
Figure 4. DFT-calculated mechanisms for the side (left drawing) and front (right drawing) attacks of 1 over MeOTf. Relative Gibbs energies in THF solution (electronic gas-phase energies in parentheses) are given in kcal mol−1.
Figure 5. DFT-calculated mechanisms for the side (left drawing) and front (right drawing) attacks of 1 over MeI. Relative Gibbs energies in THF solution (electronic gas-phase energies in parentheses) are given in kcal mol−1.
entropic effect) prereaction complexes, arising from a lateral (1·MeI and 1·MeOTf) or a frontal (1·MeI′ and 1·MeOTf’′)
approach of 1 to the methylating reagent, were found (Figures 4 and 5). The four systems evolve through an initial SN2 D
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
2·OTf → 2+ + OTf− (ΔGTHF = −14.4 kcal mol−1) and 2′·OTf → 2′+ + OTf− (ΔGTHF = −14.6 kcal mol−1), which imply the separation of the solvated tight ion-pairs into the corresponding solvent-separated ion-pairs, release ca. 7 kcal mol−1 more than those involving iodide, 2·I → 2+ + I− (ΔGTHF = −7.3 kcal mol−1) and 2′·I → 2′+ + I− (ΔGTHF = −7.5 kcal mol−1). These apparently innocent observations have a key influence on the final outcome of the reactions of complex 1 with MeOTf and MeI (see below). Investigating the formation of acyl derivatives from 2+ and + 2′ , we found that these two cations readily interconvert via unsaturated acetyl intermediates (Figure 6). Thus, the migratory insertions of the C15O15 carbonyl ligand of 2+ and 2′+ into the corresponding Ru1−C1 bond lead to pentacoordinated acetyl intermediates (2ac+ and 2ac′+, respectively) that differ only in the orientation of the acetyl group and they are connected through a low-energy transition state (TS4+). The barrier to the transformation of 2′+ into 2+ is only 12.5 kcal mol−1 (TS5+), 2+ being 2.7 kcal mol−1 more stable than 2′+. We finally investigated the reactions of the cationic acetyl intermediates 2ac+ and 2ac′+ with triflate and iodide anions. We found that both acetyl intermediates lead to compound 4 when they react with triflate and to compound 3 when they react with iodide, the activation barriers being very low in all cases. As representative examples, the reaction pathways that lead to compounds 4 and 3 via 2ac+ from 2+ and the corresponding anion are depicted in Figures 7 and 8, respectively. The large values of the electronic energies of the points that imply separated ions (cation + anion) are due to the high instability of the separated ions in the gas phase.20 It is noteworthy that, as a consequence of the greater stability of the separated ion-pairs of the triflate anion (with respect to those of the iodide anion), the formation of the tight ionic-pair 2ac·OTf from 2ac+ and OTf− costs 7.1 kcal mol−1 more than the formation of 2ac·I from 2ac+ and I−. Once the tight ionic-pairs are formed, their transformations into the final products 4 and 3 are almost barrierless (ΔG < 0.1 kcal mol−1) because 2ac+ is a
substitution process in which the metal atom of 1 attacks the C atom of MeI or MeOTf, displacing the corresponding anion from the methylating reagent. The lateral approach of 1 to MeI or MeOTf gives 2+, which has the methyl group trans to a CO ligand, whereas the front approach of 1 to MeI or MeOTf gives 2′+, which has the methyl group trans to an NHC fragment and is 2.7 kcal mol−1 less stable than 2+. Relevant data of transition states TS1, TS1′, TS2, and TS2′, in which the CH3 group is roughly planar and the Ru1, C1, and O1 or I1 atoms are almost colinear, are collected in Table 2, where the given activation Table 2. Selected Data of Transition States TS1, TS1′, TS2, and TS2′ ΔG‡THF, kcal mol−1 ΔE⧧, kcal mol−1 Ru1−C1, Å C1−I1, Å C1−O1, Å Ru1−C1−I1, deg Ru1−C1−O1, deg
TS1
TS1′
TS2
TS2′
28.3 16.9 3.018
29.5 19.7 3.024
20.9 15.3 2.809 2.746
21.7 18.5 2.813 2.794
1.892
1.933 166.3
167.0
168.5
175.1
energies (ΔG⧧THF) correspond to the difference between the energies of the separated reactants and the transition state. As, for each methylating reagent, the activation energies of the lateral and front attacks are very similar, both processes should occur simultaneously. In other words, both 2+ and 2′+ should initially be formed at similar rates. Mechanistically, these SN2 substitution processes are reminiscent of the first step of the oxidative addition of methyl iodide to square planar d8 metal complexes, which has been characterized by DFT methods in several instances.20,21,23 It is interesting to note that the reaction of 1 with MeOTf to give [2]OTf is thermodynamically more favorable (ΔGTHF = −26.7 kcal mol−1) than that of 1 with MeI to give [2]I (ΔGTHF = −10.4 kcal mol−1). An additional noteworthy point observable in Figures 4 and 5 is that the transformations
Figure 6. DFT-calculated mechanism for the isomerization of 2+ into 2′+. Relative Gibbs energies in THF solution (electronic gas-phase energies in parentheses) are given in kcal mol−1 and are relative to that of 2+. E
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
Figure 7. DFT-calculated mechanism for the transformation 2+ + OTf− → 4. Gibbs energies in THF solution (electronic gas-phase energies in parentheses) are given in kcal mol−1 and are relative to that of 1 + MeOTf.
Figure 8. DFT-calculated mechanism for the transformation 2+ + I− → 3. Gibbs energies in THF solution (electronic gas-phase energies in parentheses) are given in kcal mol−1 and are relative to that of 1 + MeI.
coordinatively unsaturated pentacoordinated ruthenium(II) species, but the formation of 4 from 2ac·OTf releases 5.7 kcal mol−1 less than the formation of 3 from 2ac·I. A comparison between the S−O bond distances and NBO atomic charges of the free triflate anion with those of the same anion in transition state TS1 and in the tight ion-pair 2·OTf revealed that the resonance-equilibrated situation of the free triflate, which has three equal S−O distances (1.482 Å) and O
atom charges (−1.001 e), is strongly distorted in TS1 (S−O distances: 1.536, 1.477, and 1.462 Å; O atom charges: −0.872, −0.980, and −0.923 e) but only slightly different from that in the ion-pair 2·OTf (S−O distances: 1.489, 1.482, and 1.484 Å; O atom charges: −0.997, −1.025, and −1.007 e). Therefore, the resonance stabilization of the triflate anion should importantly contribute to the great stabilization of 2·OTf with respect to TS1 (ΔGTHF = −40.0 kcal mol−1), because that F
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
of 2·I with respect to TS2 is much smaller (ΔGTHF = −24.0 kcal mol−1). However, the discrepancy in the differences between the energies of the solvent-separated ion-pairs with respect to those of the corresponding solvated tight ion-pairs of the iodide and triflate systems should be mainly due to the different solvation energies of the iodide and triflate anions. Also of key importance in the outcome of the reactions of 1 with MeI and MeOTf is that the energy released in the final step (transformation of the corresponding ionic-pair, 2ac·I or 2ac·OTf, into the final product, 3 or 4, respectively) is 5.7 kcal mol−1 greater for the iodide system than for the triflate system. This fact not only reflects the preference of a soft Lewis acid (the carbonyl metal complex) for a soft Lewis base (the large iodide anion) over a hard Lewis base (the O-donor triflate anion) but is also affected by the decrease of resonance stability that results from the formation of an O-coordinated triflate from a free triflate. Therefore, the resonance stabilization and the greater solvation energy of the triflate anion are responsible for the fact that, in THF solution, the solvent-separated ions of the salt [2]OTf are 7.0 kcal mol−1 more stable than compound 4. Hence, as the transformation 2+ + OTf− → 4 is thermodynamically disfavored, the experimental reaction of compound 1 with methyl triflate in THF solution does not end in 4 but in 2+ + OTf− (ΔGTHF = −26.7 kcal mol−1). On the other hand, as the iodido-acetyl derivative 3 is 5.8 kcal mol−1 more stable than 2+ + I−, the final product of the reaction of compound 1 with methyl iodide is the neutral iodido-acetyl derivative 3 (ΔGTHF = −16.2 kcal mol−1).
on silica gel. Compound 1 was prepared as described previously.11 All remaining reagents were purchased from commercial sources. All reaction products were vacuum-dried for several hours prior to being weighed and analyzed. IR spectra were recorded in solution on a Perkin-Elmer Paragon 1000 FT spectrophotometer. NMR spectra were run on a Bruker DPX-300 instrument. Microanalyses were obtained from the University of Oviedo Analytical Service. FAB mass spectra were obtained from the University of A Coruña Mass Spectrometric Service; data given refer to the most abundant molecular ion isotopomer. [RuMe{κ2C2-MeIm(CH2)3ImMe}(CO)3][OTf] ([2]OTf). Methyl triflate (6 μL, 0.24 mol) was injected into a solution of compound 1 (94 mg, 0.24 mmol) in THF (25 mL). The reaction was instantaneous (IR monitoring), although no color change was observed (yellow). The solvent was removed under reduced pressure, and the resulting yellow solid was washed with hexane (3 × 20 mL) and dried under vacuum (119 mg, 89%). Anal. Calcd for C16H19F3N4O6RuS (553.5): 34.72; H, 3.46; N, 10.12. Found: C, 34.83; H, 3.49; N, 10.07. (+)-FAB MS: 405 [2+]. IR (THF, cm−1): νCO 2093 (s), 2025 (vs), 1984 (m). 1 H NMR (CD2Cl2, 298.2 K, 300.13 MHz, ppm): δ 7.32 (d, J = 2.2 Hz, 1 H, NCH), 7.31 (d, J = 2.2 Hz, 1 H, NCH), 7.24 (d, J = 1.8 Hz, 1 H, NCH), 7.23 (d, J = 1.8 Hz, 1 H, NCH), 4.20−3.94 (m, 2 H, NCH2), 4.07 (s, 3 H, NCH3), 4.00 (s, 3 H, NCH3), 3.61−3.40 (m, 2 H, NCH2), 1.92−1.83 (m, 2 H, CH2), 0.06 (s, 3 H, RuCH3). 13C{1H} and DEPT-135 NMR (CD2Cl2, 75.48 MHz, 298.2 K): δ 195.1 (CO), 194.2 (CO), 187.2 (CO), 170.8 (CNHC), 170.5 (s, CNHC), 126.9 (s, NCH), 126.2 (s, NCH), 124.1 (s, NCH), 123.1 (s, NCH), 46.9 (s, NCH2), 45.9 (s, NCH2), 40.9 (s, NCH3), 40.3 (s, NCH3), 34.2 (s, CH2), −15.4 (s, RuCH3). [RuI{C(O)Me}{κ2C2-MeIm(CH2)3ImMe}(CO)2] (3). Methyl iodide (15 μL, 0.24 mol) was injected into a solution of compound 1 (94 mg, 0.24 mmol) in THF (25 mL). The reaction was instantaneous (IR monitoring), although no color change was observed (yellow). The solvent was removed under reduced pressure, and the resulting yellow solid was washed with hexane (3 × 20 mL) and dried under vacuum (117 mg, 92%). Anal. Calcd for C15H19IN4O3Ru (531.3): 33.91; H, 3.60; N, 10.55. Found: C, 33.96; H, 3.64; N, 10.45. (+)-FAB MS: 532 [M+]. IR (THF, cm−1): νCO 2017 (vs), 1952 (vs), 1624 (m). 1H NMR (CD2Cl2, 298.2 K, 300.13 MHz, ppm): δ 7.05 (d, J = 1.8 Hz, 1 H, NCH), 7.01 (d, J = 1.8 Hz, 1 H, NCH), 6.89 (d, J = 1.6 Hz, 1 H, NCH), 6.88 (d, J = 1.6 Hz, 1 H, NCH), 4.63−4.44 (m, 2 H, NCH2), 4.26 (s, 3 H, NCH3), 3.85 (s, 3 H, NCH3), 3.79−3.69 (m, 2 H, NCH2), 2.40 (s, 3 H, C(O)CH3), 1.77−1.66 (m, 2 H, CH2). 13C{1H} and DEPT-135 NMR (CD2Cl2, 75.47 MHz, 296.6 K): δ 199.6 (CO), 198.8 (CO), 179.2 (CNHC), 177.3 (s, CNHC), 124.5 (s, NCH), 124.2 (s, NCH), 121.4 (s, NCH), 120.7 (s, NCH), 50.7 (s, NCH2), 47.7 (s, NCH2), 45.4 (s, NCH3), 42.3 (s, NCH3), 39.7 (s, C(O)CH3), 34.4 (s, CH2) (the acetyl CO carbon atom was not observed). Reaction of [2]OTf with [NEt4]I. Solid [NEt4]I (20 mg, 0.780 mmol) was added to a solution of [2]OTf (40 mg, 0.763 mmol) in THF (10 mL). After stirring for 5 min, an IR spectrum of the solution revealed the complete transformation of cation 2+ into complex 3. X-ray Diffraction Analyses. Crystals of [2]OTf and 3·CH2Cl2 were analyzed by XRD methods. A selection of crystal, measurement, and refinement data is given in Table 3. Diffraction data were collected on an Oxford Diffraction Xcalibur Onyx Nova single-crystal diffractometer, using Cu Kα radiation. Empirical absorption corrections were applied using the SCALE3 ABSPACK algorithm as implemented in the program CrysAlisPro RED.24 The structures were solved using the program SIR-97.25 Isotropic and full matrix anisotropic least-squares refinements were carried out using SHELXL.26 In [2]OTf, the C1 methyl group and its trans carbonyl ligand C13O13 were disordered over two positions with a 50% occupancy. This disorder consists of an exchange of their coordination sites. Additionally, the carbonyl O14 and O15 oxygen atoms of [2]OTf are also disordered over two nearby positions (61:39 and 66:34 occupancy ratios, respectively). The CH2 group of the solvent molecule of 3·CH2Cl2 was found disordered over two positions (63:37 occupancy ratio). All non-H atoms of 3·CH2Cl2 were refined anisotropically, while all H atoms were set in calculated positions
■
CONCLUDING REMARKS The reactions of compound 1 with methyl triflate and methyl iodide lead to the cationic methyl derivative 2+ (as triflate salt) and to the neutral iodido-acetyl complex 3, respectively, as final products. No doubt, the chelating and strong σ-donating bis(NHC) ligand of complex 1 not only minimizes the number of possible isomeric products but also enhances the nucleophilic character of the metal atom, facilitating its reactions with methyl triflate and methyl iodide, which occur readily at room temperature. DFT calculations not only have shed light on the mechanisms of these reactions but have also explained the different chemoselectivity observed for methyl triflate and methyl iodide. For each methylating reagent, two alternative rate-determining SN2 substitution processes, which differ on how the metal atom approaches the corresponding methylating reagent, have been identified. However, all these processes lead to the cationic methyl derivative 2+. The resonance stabilization and greater hardness and solvation energy of the triflate anion, with respect to that of the iodide anion, are responsible for the fact that the reaction of complex 1 with methyl triflate ends in 2+ (as triflate salt), while the iodide anion reacts readily with 2+ to give the neutral iodido-acetyl complex 3 as final product. This work is the first to use DFT calculations to study the mechanisms of oxidative addition reactions of alkyl halides and pseudohalides to carbonyl ruthenium(0) complexes.
■
EXPERIMENTAL SECTION
General Procedures. Solvents were dried over sodium diphenyl ketyl (hexane, toluene, THF) and distilled under nitrogen before use. The reactions were carried out under nitrogen, at room temperature, using Schlenk-vacuum line techniques, and were monitored by solution IR spectroscopy (carbonyl stretching region) and spot TLC G
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
■
Table 3. Crystal, Measurement, and Refinement Data for the Compounds Studied by XRD formula fw cryst syst space group a, Å b, Å c, Å a, b, g, deg V, Å3 Z F(000) Dcalcd, g cm−3 μ (Cu Kα), mm−1 cryst size, mm T, K θ range, deg min./max. h, k, l no. collected reflns no. unique reflns no. reflns with I > 2σ(I) no. params/ restraints GOF on F2 R1 (on F, I > 2σ(I)) wR2 (on F2, all data) min./max. Δρ, e Å−3
[2]OTf
3·CH2Cl2
C15H19N4O3Ru·CF3O3S 553.49 monoclinic C2/c 24.4685(8) 6.5425(2) 25.8504(8) 90, 93.070(3), 90 4132.3(2) 8 2224 1.779 7.730
C15H19IN4O3Ru·CH2Cl2 616.24 monoclinic P21/c 10.6317(2) 8.7517(2) 22.9820(3) 90, 96.620(2), 90 2124.11(7) 4 1200 1.927 19.912
0.08 × 0.07 × 0.07 123(2) 3.42 to 72.44 −22/30, −7/8, −28/31 7779
0.07 × 0.04 × 0.02 123(2) 3.87 to 70.00 −9/12, −10/10, −24/27 7370
3956
3968
3673
3465
313/0
257/0
1.088 0.034
1.032 0.032
0.094
0.085
−1.060/0.591
−0.785/1.089
Article
ASSOCIATED CONTENT
S Supporting Information *
ORTEP plots of the XRD structures, 1H and 13C NMR spectra, atomic coordinates of the DFT-optimized structures, and crystallographic data in CIF format. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank the European Union (Marie Curie action FP7-2010RG-268329) and the Spanish Government (MICINN-FEDER grants CTQ2010-14933, MAT2010-15094, and DELACIERVA-09-05) for funding this work. This work is dedicated to Prof. Antonio Laguna on the occasion of his 65th birthday.
■
REFERENCES
(1) For recent leading references on NHC-metal chemistry, see: (a) Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47, 3122. (b) Schuster, O.; Yang, L.; Raubenheimer, H. G.; Albrecht, M. Chem. Rev. 2009, 109, 3445. (c) Poyatos, M.; Mata, J. A.; Peris, E. Chem. Rev. 2009, 109, 3677. (d) Melaimi, M.; Soleihavoup, M.; Bertrand, G. Angew. Chem., Int. Ed. 2010, 49, 8810. (2) For relevant reviews on catalytic applications of NHC complexes, see: (a) Normand, A. T.; Cavell, K. J. Eur. J. Inorg. Chem. 2008, 2781. (b) Díez-González, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612−3676. (c) Velázquez, H. D.; Verpoort, F. Chem. Soc. Rev. 2012, 41, 7032. (3) For a review on NHC transition metal cluster chemistry, see: Cabeza, J. A.; García-Á lvarez, P. Chem. Soc. Rev. 2011, 40, 5389. (4) Cabeza, J. A.; del Río, I.; Miguel, D.; Pérez-Carreño, E.; SánchezVega, M. G. Organometallics 2008, 27, 211. (5) (a) Ellul, C. E.; Mahon, M. F.; Saker, O.; Whittlesey, M. K. Angew. Chem., Int. Ed. 2007, 46, 6343. (b) Critall, M. R.; Ellul, C. E.; Mahon, M. F.; Saker, O.; Whittlesey, M. K. Dalton Trans. 2008, 4209. (6) Bruce, M. I.; Cole, M. L.; Fung, R. S. C.; Forsyth, C. M.; Hilder, M.; Junk, P. C.; Konstas, K. Dalton Trans. 2008, 4118. (7) Ellul, C. E.; Saker, O.; Mahon, M. F.; Apperley, D. C.; Whittlesey, M. K. Organometallics 2008, 27, 100. (8) Cabeza, J. A.; Damonte, M.; García-Á lvarez, P.; Kennedy, A. R.; Pérez-Carreño, E. Organometallics 2011, 30, 826. (9) Cabeza, J. A.; Damonte, M.; García-Á lvarez, P.; Hernández-Cruz, M. G.; Kennedy, A. R. Organometallics 2012, 31, 327. (10) Cabeza, J. A.; Damonte, M.; Pérez-Carreño, E. Organometallics 2012, 31, 8355. (11) Cabeza, J. A.; Damonte, M.; García-Á lvarez, P.; Pérez-Carreño, E. Chem. Commun. 2013, 49, 2813. (12) Warratz, S.; Postigo, L.; Royo, B. Organometallics 2013, 32, 893. (13) Zhang, C.; Li, B.; Song, H.; Xu, S.; Wang, B. Organometallics 2011, 30, 3029. (14) Related intramolecular C−H bond-activation processes have also been observed for triruthenium NHC complexes. See, for example, references 5, 13, and (a) Zhang, C.; Luo, F.; Cheng, B.; Li, B.; Song, H.; Xu, S.; Wang, B. Dalton Trans. 2009, 7230. (b) Cabeza, J. A.; del Río, I.; Miguel, D.; Sánchez-Vega, M. G. Angew. Chem., Int. Ed. 2008, 47, 1920. (c) Cabeza, J. A.; del Río, I.; Pérez-Carreño, E.; Sánchez-Vega, M. G.; Vázquez-García, D. Angew. Chem., Int. Ed. 2009, 48, 555. (d) Cabeza, J. A.; del Río, I.; Pérez-Carreño, E.; Sánchez-Vega, M. G.; Vázquez-García, D. Organometallics 2010, 29, 4464. (15) (a) Collmann, J. P.; Roper, W. R. Adv. Organomet. Chem. 1968, 7, 53. (b) Jeffery, J.; Mawby, R. J. J. Organomet. Chem. 1972, 40, C42.
and refined riding on their parent atoms. The WINGX program system was used throughout the structure determinations.27 CCDC deposition numbers: 939178 ([2]OTf) and 939179 (3·CH2Cl2). Computational Details. DFT calculations were carried out using Becke’s three-parameter hybrid exchange−correlation functional28 and the hybrid B3LYP nonlocal gradient correction.29 The LanL2DZ basis set,30 with relativistic effective core potentials, was used for the Ru and I atoms. The standard LanL2DZ basis set of the iodine atom was augmented with a set of p and d polarization functions (with 0.0308 and 0.294 exponents, respectively)31 and one SP diffuse function (with an exponent of 0.035).32 The basis set used for the remaining atoms was the diffuse and polarized 6-31+G(d,p).33 No simplified model compounds were used for the calculations. All stationary points of the mechanistic studies were fully optimized in the gas phase and confirmed as energy minima (reactants, products, and intermediates; all positive eigenvalues) or transition states (one imaginary eigenvalue) by analytical calculation of frequencies. These calculations were also used to determine the gas-phase Gibbs energies (G). IRC calculations were used to verify that the transition states found were correct saddle points connecting the proposed minima. For each stationary point, the effect of the solvent was estimated by computing the CPCM energy,34 ECPCM, in a single-point calculation (εTHF = 7.4257). Solvent effectcorrected Gibbs energies, Gsol, were calculated by using the equation Gsol = ECPCM + (G − E), where E is the potential (electronic) energy.35 Molecular orbital data were obtained from NBO analysis of the data.36 All calculations were carried out without symmetry constraints employing the Gaussian 09 package.37 Cartesian coordinates for the atoms of all optimized structures are given in the Supporting Information. H
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX
Organometallics
Article
(c) Bellachioma, G.; Cardaci, G.; Macchioni, A.; Madami, A. Inorg. Chem. 1993, 32, 554. (16) Hommeltoft, S. I.; Baird, M. C. Organometallics 1986, 5, 190. (17) (a) Jaunky, P.; Schmalle, H. W.; Alfonso, M.; Fox, T.; Berke, H. J. Organomet. Chem. 2004, 689, 801. (b) Jaunky, P.; Schmalle, H. W.; Blacque, O.; Fox, T.; Berke, H. J. Organomet. Chem. 2005, 690, 1429. (18) Van Leeuwen, P. W. N. M. Homogeneous Catalysis: Understanding the Art; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2004. (19) An analogous situation has been observed in other complexes containing cis trimethylene-bridged bis(NHC) ligands. See, for example: (a) Mata, J. A.; Chianese, A. R.; Miecznikowski, J. R.; Poyatos, M.; Peris, E.; Faller, J. W.; Crabtree, R. H. Organometallics 2004, 23, 1253. (b) Poyatos, M.; Mas-Marzá, E.; Sanaú, M.; Peris, E. Inorg. Chem. 2004, 43, 1793. (c) Ahrens, S.; Zeller, A.; Taige, M.; Strassner, T. Organometallics 2006, 25, 5409. (d) Viciano, M.; Poyatos, M.; Sanaú, M.; Peris, E.; Rossin, A.; Ujaque, G.; Lledós, A. Organometallics 2006, 25, 1120. (e) Viciano, M.; Mas-Marzá, E.; Sanaú, M.; Peris, E. Organometallics 2006, 25, 3063. (f) Cheng, Y.; Lu, X.-Y.; Xu, H.-J.; Li, Y.-Z.; Chen, X.-T.; Xue, Z.-L. Inorg. Chim. Acta 2010, 363, 430. (g) Krüger, A.; Neels, A.; Albrecht, M. Chem. Commun. 2010, 46, 315. (20) Feliz, M.; Freixa, Z.; van Leeuwen, P. W. N. M.; Bo, C. Organometallics 2005, 24, 5718. (21) (a) Amovili, C.; Barone, V.; Cammi, R.; Cances, E.; Cossi, M.; Mennuci, B.; Pomelli, C. S.; Tomasi, J. Adv. Quantum Chem. 1999, 32, 227. (b) Ivanova, E. A.; Gisdakis, P.; Nasluzov, V. A.; Rubailo, A. V.; Rö sch, N. Organometallics 2001, 20, 1161. (c) Kinnunen, T.; Laasonen, K. J. Mol. Struct. (THEOCHEM) 2001, 540, 91. (d) Kinnunen, T.; Laasonen, K. J. Mol. Struct. (THEOCHEM) 2001, 542, 273. (e) Kinnunen, T.; Laasonen, K. J. Organomet. Chem. 2001, 628, 222. (f) Cheong, M.; Ziegler, T. Organometallics 2005, 24, 3053. (g) Conradie, M. M.; Conradie, J. J. Organomet. Chem. 2010, 695, 2126. (h) Erasmus, J. J. C.; Conradie, J. Inorg. Chim. Acta 2011, 375, 128. (i) Erasmus, J. J. C.; Conradie, M. M.; Conradie, J. React. Kinet. Mech. Catal. 2012, 105, 233. (22) Papajak, E.; Truhlar, D. G. J. Chem. Theory Comput. 2010, 6, 597. (23) (a) Hill, G. S.; Puddephatt, R. J. Organometallics 1998, 17, 1478. (b) Bartlett, K. L.; Goldberg, K. I.; Borden, W. T. Organometallics 2001, 20, 2669. (c) Gonzales, J. M.; Musaev, D. G.; Morokuma, K. Organometallics 2005, 24, 4908. (d) Fazaeli, R.; Ariafard, A.; Jamshidi, S.; Tabatabaie, E. S.; Pishro, K. A. J. Organomet. Chem. 2007, 692, 3984. (e) Hosseini, F. N.; Ariafard, A.; Rashidi, M.; Azimi, G.; Nabavizadeh, S. M. J. Organomet. Chem. 2011, 696, 3351. (24) CrysAlisPro RED, version 1.171.34.36; Oxford Diffraction Ltd.: Oxford, UK, 2010. (25) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. C.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115. (26) SHELXL: Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112. (27) WINGX, version 1.80.05 (2009): Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837. (28) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (29) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. (30) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299. (31) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T. M.; Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111. (32) Ardura, D.; López, R.; Sordo, T. L. J. Phys. Chem. B 2005, 109, 23618. (33) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213. (34) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995. (35) (a) Balcells, D.; Ujaque, G.; Fernández, I.; Khiar, N.; Maseras, F. J. Org. Chem. 2006, 71, 6388. (b) Braga, A. A. C.; Ujaque, G.; Maseras, F. Organometallics 2006, 25, 3647. (36) (a) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735. (b) Reed, A. E.; Curtis, L. A.; Weinhold, F. J. Chem. Rev. 1988, 88, 899.
(37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.01; Gaussian, Inc.: Wallingford, CT, 2009.
I
dx.doi.org/10.1021/om400534n | Organometallics XXXX, XXX, XXX−XXX