Homogeneous Model Complexes for Supported Rhenia Metathesis

Article pubs.acs.org/Organometallics

Homogeneous Model Complexes for Supported Rhenia Metathesis Catalysts Yu-Ying Lai, Marc Bornand, and Peter Chen* Laboratorium für Organische Chemie, Eidgenössische Technische Hochschule Zürich (ETH Zürich), Wolfgang-Pauli-Strasse 10, CH-8093 Zürich, Switzerland S Supporting Information *

ABSTRACT: A series of rhenium trioxo complexes (L− ReO3) was synthesized, characterized, and demonstrated to be active in olefin metathesis. The relationship between perrhenate (ReO4−), perrhenyl (ReO3+), and metathesis-active rhenium complexes (L−ReO3) was elucidated. Their chemical behavior can be tuned through the Lewis acid−base interaction. DFT calculations were performed for the metathesis reaction of L−ReO3 with norbornene, which demonstrates that electron-withdrawing substituents or ligands are beneficial for olefin metathesis activity. Moreover, a rhenium− alkylidene complex was synthesized, which can be activated by B(C6F5)3 to afford an active metathesis catalyst. This system can be regarded as a homogeneous model of the hypothetical species present in the heterogeneous catalytic systems. The findings from the present study are consistent with our previous gas-phase studies and constitute the elucidation of the working principles for the metathesis-active rhenium species on the surface.

■

INTRODUCTION The system Re2O7/Al2O3 and its related system, CH3ReO3 (MTO)/Al2O3−SiO2, are effective heterogeneous catalysts for olefin metathesis at ambient temperature, even though the putative rhenium carbene active species is produced only in situ. Understanding of the mechanism on a molecular level for heterogeneous catalysis remains a challenge. Therefore, it is essential to synthesize homogeneous models of the hypothetical surface species present in the rhenium-based heterogeneous catalysts. These models can serve as a bridge between the realms of homogeneous and heterogeneous catalysis and help to elucidate the working principles for those heterogeneous catalysts. Our earlier experimental and computational studies on high-valent rhenium trioxo complexes in the gas phase1 inspired the pseudo-Wittig mechanism2 and have proven essential for elucidation of the structural requirements for a self-activating homogeneous metathesis catalyst. As is evident from Scheme 1, a reaction sequence from left to right would provide a route for a rhenium−oxo complex to react with an olefin to form a rhenium carbene by addition, rearrangement, and dissociation. As suggested by qualitative molecular orbital arguments, as well as high-level quantum chemical calculations,1c this requires that (i) the rhenium center is four-coordinate, (ii) the supporting ligand is a σ-donor, and (iii) the σ-donor ligand is electron-deficient. In other words, the electronic property of L in L−ReO3 plays an important role in metathesis activity. From a structural point of view, Re2O7 can be divided into two fragments: a perrhenate anion (ReO4−) and a perrhenyl cation (ReO3+). Cleavage of Re2O7 can be achieved in the © XXXX American Chemical Society

presence of electrophiles or nucleophiles to give monomeric compounds of the general formula L−ReO3.3 It was reported that Re2O7 dissolves in polar solvents such as THF, CH3CN, and pyridine to form solvent adducts.4 This unsymmetrical coordination enhances the reactivity of Re2O7 toward alkylation, arylation, and analogous reactions.4d,5 One can thus envision that, when Re2O7 is supported on the surface, cleavage is the primary reaction. Therefore, we focus on the use of perrhenate (ReO4−) and perrhenyl (ReO3+) as building blocks to furnish the target molecules (L−ReO3), whose structural and catalytic properties are investigated to give a comprehensive view of possible surface species. Moreover, a rhenium−alkylidene complex was synthesized, which is activated by a Lewis acid through the Lewis acid−base interaction to afford an active metathesis catalyst. The system can be regarded as a homogeneous model of the hypothetical species present in the heterogeneous catalytic systems.

■

RESULTS AND DISCUSSION Perrhenate Complexes. The perrhenate anion is a weakly coordinating anion. In comparison to commonly used anions, it is a much weaker base than Cl− but stronger than ClO4−, SO3CF3−, and BF4− (basicity: Cl− ≫ ReO4− > ClO4− > SO3CF3− ≫ BF4−).6 Perrhenates exhibit lower reactivity than structurally analogous permanganates and perchlorates in a variety of reactions. On the basis of our previous gas-phase and theoretical studies,1 substitution or complexation of one oxo Received: September 5, 2012

A

dx.doi.org/10.1021/om300852s | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 1. Pseudo-Wittig Mechanism for a Rhenium Trioxo Complex Reacting with an Olefin

moiety on perrhenate to introduce an electron-deficient ligand should render the resultant perrhenate complex metathesis active. Of the various metal oxide supporting agents for Re2O7, niobia was found to be a very effective support in promoting olefin metathesis, which may be attributed to its strong Lewis acidity.7 Thus, it is reasonable to choose niobia as the supporting ligand in making a homogeneous rhenium-based catalyst. The sterically hindered (salophen)NbV oxychloride complex 1 was prepared by reaction of (CH3CN)2Nb(O)Cl3 with the salophen ligand in benzene, followed by evaporation of solvent and recrystallization from CH2Cl2/benzene/hexane. Replacement of the chloride in 1 by BArF4 or perrhenate was achieved by treatment with NaBAr F 4 (Ar F = 3,5-bis(trifluoromethyl)phenyl), to afford 2, or AgReO4, to furnish 3, followed by filtration and recrystallization from CH2Cl2/ benzene and CH2Cl2/hexane, respectively (Chart 1).

Table 1. Conditions and Yields of Norbornene Polymerizationa by Perrhenate-Based Complexes b

1 2 3 4 5c

cat.

loading, %

T

t

yield, %

complex 3 complex 2 + MTO AgReO4 + AlCl3 AgReO4 + NbCl5 CReO4 + AlCl3

10 10 1 1 10

160 °C rt rt rt rt

24 h 2 days 2h 2h 2h

40 70 quantitative 20 none

a

In a typical polymerization experiment, a solution of 100 mg of norbornene in 5 mL of solvent, either CH2Cl2 or C2H4Cl2, was treated with 1% or 10% catalyst. The resulting polymer solution was poured into 100 mL of MeOH. After 1 h the coagulated polymer was collected, washed with MeOH, and dried under vacuum to yield polynorbornene. rt = room temperature. bThe reaction was carried out in a sealed tube. cCReO4 = [sodium(benzo-15-crown-5)]perrhenate.

metathesis catalyst when it is supported on an appropriate metal oxide.7a,b,8,9 The related heterogeneous systems of metal oxide supported MTO also generally showed higher catalytic activity than their perrhenate analogues.10 The results indicate that the metathesis activity of complex 3 might be attributed to the electron-withdrawing properties of the niobium-containing ligand, which activates perrhenate through adduct formation. However, the conditions employed here were harsher than those for common homogeneous catalysts; this might be due to the fact that the Lewis acidity of the niobium center is diminished by coordination of the Lewis basic nitrogens. ESI-MS spectrometry reveals that the chemical bonding between [(salophen)Nb(oxo)]+ and ReO4− is not robust and the perrhenate anion is described as a good leaving group in this case.11 It could be envisaged that a more electrondeficient ligand would strengthen the bonding between the ligand and perrhenate and further enhance the catalyst performance in olefin metathesis. With the intention of strengthening the bonding between perrhenate and the supporting ligand, we decided to use AlCl3 as the auxiliary activating reagent. Cl2Al−OReO3 is expected to be the primary product of the reaction between AgReO4 and AlCl3, the formation of AgCl being the driving force for its formation. As expected, an equimolar mixture of AgReO4 and AlCl3 is active in olefin metathesis, exhibiting quantitatively higher activity than complex 3 (entry 3 in Table 1). A combination of AgReO4 and NbCl5 also proved active in olefin metathesis, albeit with lower efficiency (entry 4 in Table 1). [Sodium(benzo-15crown-5)]perrhenate was synthesized in order to ascertain that the chloride substitution of AlCl3 by ReO4 is essential for rendering perrhenate metathesis active. No polymer was detected when norbornene was treated with an equimolar mixture of [sodium(benzo-15-crown-5)]perrhenate and AlCl3 (entry 5 in Table 1). This control experiment demonstrated that in the absence of chloride abstraction by Ag+ perrhenate leads to no ROMP activity, which indicates that a fourcoordinate aluminum is not sufficient, by itself, to activate perrhenate. A similar observation was noted by Commereuc for

Chart 1

The structures of complexes 1−3 were determined by X-ray crystallography (see the Supporting Information). All three complexes share a dimeric structure with a rhomboidal Nb2O2 core. Notable in 2 is the available coordination site on each niobium center, which in 1 is occupied by the chloride. The BArF4 anion in 2 shows no close contact to niobium. By comparison, each perrhenate moiety in 3 is bound through one oxygen atom to a seven-coordinate niobium, with Nb−O and Re−O bond lengths of 2.151 and 1.771 Å, respectively, consistent with the expected values for metal−oxygen single bonds. Each rhenium center is close to tetrahedral. Complex 3 was tested for metathesis activity (entry 1 in Table 1). Catalytic amounts (10 mol %) of 3 were found to promote the ring-opening metathesis polymerization (ROMP) of norbornene, the polymer being unambiguously identified as the ROMP product by NMR. Complexes 1 and 2 and perrhenate salts alone produced no detectable ROMP products under comparable conditions. Notably, an equimolar mixture of MTO and complex 2 also acts as a ROMP catalyst, even exhibiting qualitatively higher activity than 3 (entry 2 in Table 1). Control experiments showed that neither MTO nor complex 2 catalyzes the ROMP reaction individually. This observation is consistent with reports that MTO becomes a B


Organometallics

Article

Scheme 2. Proposed Mechanism for Aromatic Transfer from Boron to Rhenium

50% yield. On the basis of the structural requirements for a rhenium-based metathesis catalyst employed in our previous studies, ReO3[3,5-(CF3)2C6H3] in Scheme 2 is the most likely candidate for catalyzing this polymerization. ReO3[3,5(CF3)2C6H3] was synthesized in order to test the conjecture. Halogen−metal exchange between 3,5-bis(trifluoromethyl)bromobenzene and n-BuLi occurred readily at −78 °C to give [3,5-bis(trifluoromethyl)phenyl]lithium, which underwent a nucleophilic substitution reaction with ZnCl2 to furnish Zn[3,5-(CF3)2C6H3]2. ReO3[3,5-(CF3)2C6H3] was then obtained through transmetalation of 3,5-bis(trifluoromethyl)phenyl from Zn[3,5-(CF3)2C6H3]2 to Re2O7. The structures of Zn[3,5-(CF3)2C6H3]2 and ReO3[3,5-(CF3)2C6H3] have both been confirmed via X-ray crystallography (see the Supporting Information). However, as shown in Table 2 (entry 1), ReO3[3,5-(CF3)2C6H3] alone gave only trace amounts of polymer after 18 h at room temperature. The difference between the reactivity of the independently prepared ReO3[3,5(CF3)2C6H3] and the in situ prepared formulation could

the catalytic system in which a three-coordinate, highly electron-deficient aluminum is believed to render analogous rhenium complexes metathesis active.12 These results suggest that a strongly Lewis acidic supporting ligand on a high-valent four-coordinate rhenium center is crucial for the metathesis reactivity, which is in agreement with the conclusions obtained from the qualitative molecular calculations. Perrhenyl Complexes. Perrhenyl may be regarded as an extreme product of the perrhenate anion reacting with an exceptionally strong Lewis acid by the formal or actual abstraction of O2−. Most of the attention in the literature has been focused on exploring the properties and abilities of perrhenate,13 and studies of perrhenyl ions are relatively scarce.14 We attempted to synthesize the perrhenyl cation with a weakly coordinating anion, such as BArF4, which we expected to exhibit the character of perrhenyl without significant contribution from the counterion. Re2O7 was mixed with TMSCl in the presence of THF to give ReO3Cl(THF)2, which was further treated with NaBArF4. The mixture was filtered under Ar and then concentrated in vacuo. Unexpectedly, recrystallization afforded the colorless tris[3,5-bis(trifluoromethyl)phenyl]borane·H 2O·2THF complex, the structure of which was determined by X-ray crystallography (see the Supporting Information). However, the rheniumcontaining product could not be identified. A mechanism for the transformation above is proposed in Scheme 2. After chloride is replaced by BArF4−, ReO3(THF)2+ is presumably formed as a transient species which abstracts one aromatic group from BArF4−. In the absence of THF, ReO3+ is more reactive and abstracts two aromatic groups from BArF4− to give bis[3,5-bis(trifluoromethyl)phenyl]borinic acid, the identity of which was established by comparison with a sample of the authentic compound. Accordingly, perrhenyl itself can be regarded as an exceedingly strong Lewis acid, since it abstracts one or even two aromatic groups from the otherwise stable and inert counterion BArF4−. When norbornene was introduced into the reaction mixture of ReO3Cl and NaBArF4, polynorbornene could be isolated in

Table 2. Conditions and Yields of Norbornene Polymerizationa by Perrhenyl-Based Complexes cat. 1 2 3 4 5 6 7

ReO3[3,5-(CF3)2C6H3] ReO3[3,5-(CF3)2C6H3] + B(C6F5)3 ReO3(C6F5)THF ReO3Ph MTO MTO + B(C6F5)3 MTO + BPh3

loading, % T, °C

t, h

yield, %

5 5

rt rt

18 1

trace quantitative

5 5 5 1 1

rt 100 100 rt rt

16 16 16 1 19

quantitative 66 (C6F5)ReO3, indicating that the reaction energy barrier decreases as the Lewis acidity of the corresponding substituent L increases (Table 3). In this regard, the data from the DFT calculations are consistent with the experiments. DFT calculations were also performed for the adduct of MTO and B(C6F5)3, which demonstrate that for the pseudo-Wittig mechanism the energy barrier for formation of a rhenium carbene and an aldehyde from MTO decreases significantly in the presence of B(C6F5)3 (Table 3). The experimental observations can be rationalized with the help of the calculated results. Rhenium Activity Scale. Considering the experimental and theoretical data within the present study, the relation among the perrhenate anion, Re−oxo metathesis catalysts, and the perrhenyl cation is clarified and proposed as the rhenium E


Organometallics

Article

Figure 3. The rhenium activity scale.

B(C6F5)3. It can be envisioned that the coordination should also occur between ReO2(CH2tBu)(CHtBu) and B(C6F5)3 and the resulting adduct may show superior metathesis activity. Therefore, we set out to examine the metathesis activity for ReO2(CH2tBu)(CHtBu). It was tested in the ROMP reaction of norbornene. One percent of ReO2(CH2tBu)(CHtBu) does catalyze the ROMP reaction of norbornene. However, no crossmetathesis products could be observed when ReO2(CH2tBu)(CHtBu) reacted with acyclic terminal olefins at room temperature. An equimolar mixture of ReO2(CH2tBu)(CHtBu) and B(C6F5)3 was also tested in the ROMP reaction of norbornene. The combination of the Re complex and the Lewis acid was found to exhibit higher reactivity than ReO2(CH2tBu)(CHtBu) alone (Table 4). Remarkably, the cross-metathesis

activity scale (Figure 3). Since the slow initiation and much faster propagation is characteristic of the Re−oxo-based catalytic systems (see the last paragraph of Results and Discussion), the rhenium activity is considered to correlate with the activation energy for the initiation step in the catalytic cycle: i.e., the formation of the initial carbene. A proxy measure of the scale is the calculated Re−O stretching frequency at the B3PW91/SDD, D95(d) level of theory, which is an indication of the Re−O bond strength. On the right end of the scale, the perrhenate anion is a weak Lewis base and often acts as a weakly coordinating anion. It can be rendered metathesis active through attachment to an aluminum or niobium electrondeficient center. The metathesis activity of the resulting complexes depends on the electron deficiency of the supporting Lewis acidic ligands. On the left end of the scale, the perrhenyl cation is seen as a strong Lewis acid and it can abstract one or two aromatic groups from BArF4. However, when an aryl group such as perfluorophenyl is connected to perrhenyl, the resulting Re complex was found to be an active metathesis catalyst as well. In the middle of the scale, Re complexes bearing more strongly electron donating ligands, such as phenyl and methyl, were found to be stable and exhibit nearly no metathesis activity at room temperature. Nevertheless, with the aid of a Lewis acidic additive, such as B(C6F5)3, the aforementioned Re complexes can be rendered metathesis active. In brief, for rhenium-based complexes, the chemical behavior can be manipulated through the Lewis acid−base interaction. Rhenium−Oxo Alkylidene Complex. The rhenium−oxo alkylidene species are considered to be the active species not only for the Re2O7-based catalysts but also for the MTO-based catalysts. Hoffman and co-workers synthesized and characterized the first example for a rhenium oxo−alkylidene complex, ReO2(CH2tBu)(CHtBu).15 However, it was reported that it exhibited no metathesis activity on treatment with olefins.16 Our experimental studies along with the DFT calculations revealed that the initiation of metathesis activity of L−ReO3 could be enhanced significantly by the presence of

Table 4. Conditions and Yields of Norbornene Polymerization by ReO2(CH2-t-Bu)(CH-t-Bu)a cat. 1 2 a

ReO2(CH2-t-Bu)(CH-t-Bu) ReO2(CH2-t-Bu)(CH-t-Bu) + B(C6F5)3

loading, %

T

t, h

yield, %

1 1

rt rt

0.5 0.5

50 quantitative

The reaction solvent is CH2Cl2. rt = room temperature.

reaction of 1-hexene could be achieved at room temperature in presence of a catalytic amount of ReO2(CH2tBu)(CHtBu) and B(C6F5)3. Lewis acid additives appear to accelerate both the formation of rhenium carbenes and the subsequent metathesis steps. In order to obtain an insight into the mechanism of the above transformation, the reaction was monitored by 1H NMR spectroscopy. The peak of the rhenium−alkylidene proton was observed to shift downfield from 12.04 to 12.64 ppm after the addition of B(C 6 F 5 ) 3 to the CD 2 Cl 2 solution of ReO2(CH2tBu)(CHtBu), thus confirming unambiguously the occurrence of complexation between ReO2(CH2tBu)(CHtBu) and B(C6F5)3. Subsequently, 1-hexene was introduced into the F


Organometallics

Article

Scheme 4. 1H NMR Monitoring of ReO2(CH2tBu)(CHtBu) Reacting with 1-Hexene

Scheme 5. Comparison of ReO2(CH2tBu)(CHtBu) and ReO3CH3 in Terms of Metathesis Activity

mixture and three types of rhenium alkylidene signals were observed. On the basis of the chemical shift and coupling pattern, the two signals were assigned individually as the rhenium neopentylidene and pentylidene complexes; the third signal may be the Re methylidene complex (see the Supporting Information). The signals for those rhenium carbene species and 1-hexene decreased over time. After 1 h, the rhenium carbene peaks almost vanished and E/Z 5-decene and neohexene were observed as the major products (Scheme 4). The NMR experiments confirm the occurrence of the metathesis reaction between ReO2(CH2tBu)(CHtBu) and 1hexene, thereby lending further support to the hypothesis that rhenium oxo−alkylidene species are likely to be the active species of the rhenium-based heterogeneous catalysts. The low metathesis activity of ReO2(CH2tBu)(CHtBu) is instructive. Herrmann and co-workers used IR and Raman spectroscopy to demonstrate that π-donor ligands lower the Re−O bond order of L−ReO3 as a result of their electrondonating abilities, and more electron-deficient complexes, L− ReO3, in which L is a σ-donor, has a higher Re−O bond order.17 For instance, the average bond order of CH3ReO3 is 2.67 (one Re−O double and two pseudo Re−O triple bonds) and the rhenium center of CH3ReO3 thus gains more electron density than one would expect. On the basis of the same statement, ReO2(CH2tBu)(CHtBu) would have two pseudo Re−O triple bonds, which would be expected to reduce the metathesis activity of this complex. In the presence of B(C6F5)3, the triple-bond character of the Re−O bond in ReO2(CH2tBu)(CHtBu) is reduced by the coordination of the oxo ligand to B(C6F5)3, which makes the rhenium center more electron deficient and, in the meantime, enhances the metathesis activity of ReO2(CH2tBu)(CHtBu). From L−ReO3 to L−ReO2(CHR). It is well accepted that a metallocarbene is the important intermediate in olefin metathesis. The mechanism for generation of rhenium alkylidene species from rhenium oxo complexes on solid supports has not been established with any certainty. A rationale for the difficulty of detecting or trapping rhenium alkylidene species for those rhenium-based heterogeneous catalysts can be illustrated through comparison of ReO3CH3 and ReO2(CH2tBu)(CHtBu) in terms of metathesis activity. As illustrated in Scheme 5, ReO3CH3 and ReO2(CH2tBu)(CHtBu) exhibit dramatically different reactivities for both the ROMP reaction of norbornene and the cross-metathesis of 1-hexene. If

the methyl and neopentyl groups are regarded as approximately equivalent spectator ligands, ReO3CH3 could be considered as a formal “precursor” of ReO2(CH2tBu)(CHtBu) obtained from the pseudo-Wittig reaction of ReO3(CHtBu) and di-tertbutylethylene. Specifically, the reaction between ReO3CH3 and an alkene can be viewed as an initiation step and the reaction between ReO2(CH2-t-Bu)(CH-t-Bu) and an alkene as a propagation step. The observed reactivity order clearly indicates that the initiation process should be much slower than propagation, which, if generalized to the structural class, further indicates that observation of the propagating species in situ should be exceedingly challenging. Furthermore, for the metathesis reactions catalyzed by L−ReO3, we find that most of the precatalyst remains unreacted after full conversion of alkenes into the metathesis products, indicating that the active species are low in concentration but efficient in reactivity, which further supports the inefficient initiation and rapid propagation of the activation mechanism.

■

CONCLUSIONS Computational studies and gas-phase mass spectrometric experiments suggested structural requirements according to which a rhenium trioxo complex would react with an olefin to form a metathesis-active catalyst. A set of rhenium trioxo G


Organometallics

Article

Angew. Chem., Int. Ed. 1968, 7, 295−296. (f) Roesky, H. W.; Hesse, D.; Noltemeyer, M. Eur. J. Solid State Inorg. Chem. 1991, 28, 809−814. (5) (a) Herrmann, W. A.; Kuchler, J. G.; Felixberger, J. K.; Herdtweck, E.; Wagner, W. Angew. Chem. 1988, 100, 420−422. (b) Herrmann, W. A.; Rost, A. M. J.; Mitterpleininger, J. K. M.; Szesni, N.; Sturm, S.; Fischer, R. W.; Kühn, F. E. Angew. Chem., Int. Ed. 2007, 46, 7301−7303. (c) de Bellefon, C. M.; Herrmann, W. A.; Kiprof, P.; Whitaker, C. R. Organometallics 1992, 11, 1072−1081. (6) Heidrich, J.; Loderer, D.; Beck, W. J. Organomet. Chem. 1986, 312, 329−333. (7) (a) Buffon, R.; Choplin, A.; Leconte, M.; Basset, J.-M.; Touroude, R.; Herrmann, W. A. J. Mol. Catal. 1992, 72, L7−L10. (b) Buffon, R.; Auroux, A.; Lefebvre, F.; Leconte, M.; Choplin, A.; Basset, J.-M.; Herrmann, W. A. J. Mol. Catal. 1992, 76, 287−295. (c) Buffon, R.; Jannini, M. J. D.M.; Abras, A. J. Mol. Catal. A: Chem. 1997, 115, 173− 181. (8) (a) Salameh, A.; Joubert, J.; Baudouin, A.; Lukens, W.; Delbecq, F.; Sautet, P.; Basset, J.-M.; Copéret, C. Angew. Chem., Int. Ed. 2007, 46, 3870−3873. (b) Salameh, A.; Baudouin, A.; Basset, J.-M.; Copéret, C. Angew. Chem., Int. Ed. 2008, 47, 2117−2120. (9) (a) Moses, A. W.; Ramsahye, N. A.; Raab, C.; Leifeste, H. D.; Chattopadhyay, S.; Chmelka, B. F.; Eckert, J.; Scott, S. L. Organometallics 2006, 25, 2157−2165. (b) Moses, A. W.; Raab, C.; Nelson, R. C.; Leifeste, H. D.; Ramsahye, N. A.; Chattopadhyay, S.; Eckert, J.; Chmelka, B. F.; Scott, S. L. J. Am. Chem. Soc. 2007, 129, 8912−8920. (10) (a) Kühn, F. E.; Scherbaum, A.; Herrmann, W. A. J. Organomet. Chem. 2004, 689, 4149−4164. (b) Rost, A. M. J.; Schneider, H.; Zoller, J. P.; Herrmann, W. A.; Kühn, F. E. J. Organomet. Chem. 2005, 690, 4712−4718. (11) Lai, Y.-Y. Rhenium-based Olefin Metathesis Catalysts and Investigation of Ligand Properties; Dissertation ETH No. 19946, ETH Zürich, 2011. (12) Commereuc, D. J. Chem. Soc., Chem. Commun. 1995, 791−792. (13) (a) Ishihara, K.; Furuya, Y.; Yamamoto, H. Angew. Chem., Int. Ed. 2002, 41, 2983−2986. (b) Paraskevopoulou, P.; Psaroudakis, N.; Koinis, S.; Stavropoulos, P.; Mertis, K. J. Mol. Catal. A.: Chem 2005, 240, 27−32. (c) Sakakura, A.; Katsukawa, M.; Ishihara, K. Angew. Chem., Int. Ed. 2007, 46, 1423−1426. (d) Bellemin-Laponnaz, S. ChemCatChem 2009, 1, 357−362. (e) Ahmad, I.; Chapman, G.; Nicholas, K. M. Organometallics 2011, 30, 2810−2818. (f) Nikonova, O. A.; Capron, M.; Fang, G.; Faye, J.; Mamede, A.-S.; JalowieckiDuhamel, L.; Dumeignil, F.; Seisenbaeva, G. A. J. Catal. 2011, 279, 310−318. (g) Vicente, B. C.; Nelson, R. C.; Moses, A. W.; Chattopadhyay, S.; Scott, S. L. J. Phys. Chem. C 2001, 115, 9012−9024. (14) (a) Beyer, M. K.; Berg, C. B.; Bondybey, V. E. Phys. Chem. Chem. Phys. 2001, 3, 1840−1847. (b) Gisdakis, P.; Rösch, N. J. Am. Chem. Soc. 2001, 123, 697−701. (c) Fialko, E. F.; Kikhtenko, A. V.; Goncharov, V. B.; Zamaraev, K. I. Catal. Lett. 1996, 41, 7−11. (15) Cai, S.; Hoffman, D. M.; Wierda, D. A. Organometallics 1995, 15, 1023−1032. (16) Toreki, R.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1992, 114, 3367−3380. (17) (a) Herrmann, W. A.; Kiprof, P.; Rypdal, K.; Tremmel, J.; Blom, R.; Alberto, R.; Behm, J.; Albach, R. W.; Bock, H.; Solouki, B.; Mink, J.; Lichtenberger, D.; Gruhn, N. E. J. Am. Chem. Soc. 1991, 113, 6527− 6537. (b) Mink, J.; Keresztury, G.; Stirling, A.; Herrmann, W. A. Spectrochim. Acta 1994, 50A, 2039−2057. (c) Mealli, C.; López, J. A.; Calhorda, M. J.; Romão, C. C.; Herrmann, W. A. Inorg. Chem. 1994, 33, 1139−1143.

complexes was synthesized and demonstrated to be active in olefin metathesis. Systematic investigations of ligands and, in particular, Lewis acidic additives revealed important reaction parameters, indicating that electron-withdrawing substituents or ligands are beneficial for olefin metathesis. The relationship among perrhenate, perrhenyl, and metathesis-active rhenium complexes was elucidated, and it provides constructive information for further improvement of rhenium-based metathesis catalysts. The studies of the reactivity in olefin metathesis highly suggested that inefficient initiation and rapid propagation is characteristic of the rhenium-based catalysts, which complicates detecting or trapping rhenium alkylidene species for the rhenium−oxo-based olefin metathesis reactions. The Re alkylidene disclosed herein was found to undergo complexation with B(C6F5)3, thereby furnishing an active olefin metathesis catalyst. This catalytic system may be regarded as a homogeneous model of the hypothesized active species present of the heterogeneous catalysts. The findings from the present study are consistent with our previous gas-phase studies and constitute the elucidation of the working principles for the metathesis-active rhenium species on the surface.

■

ASSOCIATED CONTENT

S Supporting Information *

Text, tables, figures, and CIF files giving additional synthetic experiments, X-ray structures, crystallographic data, detection of the rhenium alkylidene complexes by 1H NMR spectroscopy, 1 H and 19F NMR spectra of MTO, B(C6F5)3, and the adduct [CH3ReO2{OB(C6F5)3}], and computational details. 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 acknowledge support from the ETH Zürich and the Swiss National Science Foundation. We also acknowledge Dr. J. W. H. (Han) Peeters for FAB-MS and EI-MS measurements.

■

REFERENCES

(1) (a) Chen, X.; Zhang, X.; Chen, P. Angew. Chem., Int. Ed. 2003, 42, 3798−3801. (b) Zhang, X.; Chen, X.; Chen, P. Organometallics 2004, 23, 3437−3447. (c) Narancic, S.; Chen, P. Organometallics 2005, 24, 10−12. (d) Zhang, X.; Narancic, S.; Chen, P. Organometallics 2005, 24, 3040−3042. (2) Salameh, A.; Copéret, C.; Basset, J.-M.; Böhm, V. P. W.; Röper, M. Adv. Synth. Catal. 2007, 349, 238−242. (3) (a) Towne, T. B.; McDonald, F. E. J. Am. Chem. Soc. 1997, 119, 6022−6028. (b) Middleditch, M.; Anderson, J. C.; Blake, A. J.; Wilson, C. Inorg. Chem. 2007, 46, 2797−2804. (c) Rost, A. M. J.; Herrmann, W. A.; Kühn, F. E. Tetrahedron Lett. 2007, 48, 1775−1779. (4) (a) Johnson, J. W.; Brody, J. F.; Ansell, G. B.; Zentz., S. Inorg. Chem. 1984, 23, 2415−2418. (b) Herrmann, W. A.; Roesky, P. W.; Kühn, F. E.; Elison, M.; Artus, G.; Scherer, W.; Romão, C. C.; Lopes, A.; Basset, J.-M. Inorg. Chem. 1995, 34, 4701−4707. (c) Herrmann, W. A.; Thiel, W. R.; Kühn, F. E.; Fischer, R. W.; Kleine, M.; Herdtweck, E.; Scherer, W.; Mink, J. Inorg. Chem. 1993, 32, 5188−5194. (d) Herrmann, W. A.; Kühn, F. E.; Romão, C. C. J. Organomet. Chem. 1995, 495, 209−213. (e) Beyer, H.; Glemser, O.; Krebs, B. H


Homogeneous Model Complexes for Supported Rhenia Metathesis

Recommend Documents