Ruthenium metal alkylidene complexes coordinated with triazolylidene ligands that exhibit high olefin metathesis activity

[0003] This application claims the benefit of U.S. Provisional Patent Application Ser. No. 60/127,469, filed Mar. 31, 1999, the contents of which are herein incorporated by reference in its entirety.

[0005] Metathesis catalysts have been previously described by for example, U.S. Pat. Nos. 5,312,940, 5,342,909, 5,728,917, 5,750,815, 5,710,298, and 5,831,108 and PCT Publications WO 97/20865 and WO 97/29135 which are all incorporated herein by reference. These publications describe well-defined single component ruthenium or osmium catalysts that possess several advantageous properties. For example, these catalysts are tolerant to a variety of functional groups and are more active than previously known metathesis catalysts. Olefin metathesis is a carbon-carbon bond breaking/bond making process in which there is an overall exchange of double bond moieties between two olefins. The. three main ways that olefin metathesis can be applied are illustrated in Scheme 1. Ring-opening metathesis polymerization (ROMP) involves the formation of polyolefins from strained cyclic olefins; ring-closing metathesis (RCM) involves the intramolecular transformation of an alpha, omega-diene to a cyclic olefin; and acyclic diene metathesis (ADMET) involves the intermolecular exchange of olefins.

[0006] Olefin metathesis can be mediated by a number of transition metals, but the two most widely used catalysts are the Schrock molybdenum alkylidene (1) and the Grubbs ruthenium alkylidene (2). FIGS. 1A and 1B show examples of these two catalysts.

[0007] The commercial availability and high activity of these well-defined, single-component catalysts has led to the development of olefin metathesis as a standard synthetic method. In particular, RCM has been applied to a diverse array of problems, ranging from the total synthesis of natural products to the synthesis of catenanes. As one review author recently commented, ring-closing metathesis “has come of age as a synthetic technique. It is no longer a novelty, to be included in the title of every paper, it is a synthetic tool available to every practicing organic synthetic chemist.”

[0008] Yet, there is still considerable room for improvement. Neither 1 nor 2 is a “perfect” catalyst; each has significant problems associated with it. Although the Schrock alkylidene (1) has the greater overall activity, it suffers from extreme air and moisture sensitivity, and it lacks tolerance for many functional groups (e.g. alcohols, aldehydes, and carboxylic acids). On the other hand, the Grubbs alkylidene (2) is easier and less expensive to make, is air stable as a solid and has a much wider functional group tolerance, but its activity is limited to at least two orders of magnitude less than 1. Additionally, neither 1 nor 2 provides stereo-selective control over the metathesis products.

[0009] Because of these problems, the design of metathesis catalysts with better activity, stability, and selectivity is an area of active investigation. Recently, several modifications of complex 2 have been reported, including a heterobimetallic complex (3), a bidentate Schiff base supported complex (4), and a bis (N-heterocyclic carbene) substituted complex (5). FIGS. 2A, 2B and 2C show examples of each of these liganids.

[0010] Complex 3 is approximately 80 times more active than 2 for the ROMP of 1,5-cyclooctadiene, and so it can be used as an alternative in RCM reactions that would proceed too slowly with 2 to be practical. However, at the same time, 3 is also more unstable than 2 and decomposes more rapidly. Complex 4 is more active at elevated temperatures than 2 for RCM, and it has the advantage of remaining active in polar protic media. Finally, complex 5 displays ROMP and RCM activity at elevated temperatures that is comparable to the activity of 2 at room temperature.

[0011] Thus, there is a need for a stable, more active metathesis catalyst. The invention address this need by providing for mono-substituted derivatives as more active metathesis catalysts than those previously examined. In addition, the invention provides a method of attaching N-heterocyclic carbene ligands to a metal center and methods of using the same.

[0012] FIG. 1A shows an example of a Schrock molybdenum catalyst;

[0013] FIG. 1B shows an example of a Grubbs ruthenium alkylidene catalyst; and

[0014] FIGS. 2A, 2B and 2C show examples of ligand modifications of the Grnbbs ruthenium alkylidene catalyst.

[0015] The present invention relates to novel metathesis catalysts with a tliazolylidene-based ligand and methods for preparing and using the same. Preferred embodiments of the catalysts of the present invention are of the general formula

[0016] wherein:

[0017] X and X¹are each independently an anionic ligand;

[0018] L is a neutral electron donor ligand;

[0019] R, R¹R⁶and R⁷are each independently hydrogen or a substituent selected from the group consisting of C₁-C₂₀alkyl, C₂-C₂₀alkenyl, C₂-C₂₀alkynyl, aryl, C₁-C₂₀carboxylate, C₁-C₂₀alkoxy, C₂-C₂₀alkenyloxy, C₂-C₂₀alkynyloxy, aryloxy, C₂-C₂₀alkoxycarbonyl, C₁-C₂₀alkylthio, C₁-C₂₀alkylsulfonyl and C₁-C₂₀alkylsulfinyl. Optionally, each of the R, R¹, R⁶and R⁷substituent groups may be independently substituted with one or more moieties selected from the group consisting of C₁-C₁₀alkyl, C₁-C₁₀alkoxy, and aryl which in turn may each be further substituted with one or more groups selected from a halogen, a C₁-C₅alkyl, C₁-C₅alkoxy, and phenyl. Moreover, any of the catalyst ligands may further include one or more functional groups. Examples of suitable functional groups include but are not limited to: hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiimide, carboalkoxy, carbamate, and halogen. The inclusion of a triazolylidene ligand to the previously described ruthenium or osmium catalysts has been found to dramatically improve the properties of these complexes. In particular, the new catalyst has been shown to be remarkably more active at elevated temperatures for metathesis reactions such as the ring-opening metathesis polymerization of cyclic monomers and the ring closing metathesis of dienes.

[0020] The present invention generally relates to ruthenium carbene catalysts for use in olefin metathesis reactions. More particularly, the present invention relates to triazolylidene-based ruthenium carbene catalysts and methods for making the same.

[0021] Unmodified ruthenium and osmium carbene complexes have been described in U.S. Pat. Nos. 5,312,940, 5,342,909, 5,728,917, 5,750,815, and 5,710,298, all of which are incorporated herein by reference. The ruthenium and osmium carbene complexes disclosed in these patents all possess metal centers that are formally in the +2 oxidation state, have an electron count of 16, and are penta-coordinated. These catalysts are of the general formula

[0022] wherein:

[0023] M is ruthenium or osmium;

[0024] X and X¹are each independently any anionic ligand;

[0025] L and L¹are each independently any neutral electron donor ligand;

[0026] R and R¹are each independently hydrogen or a substituent selected from the group consisting of C₁-C₂₀alkyl, C₂-C₂₀alkenyl, C₂-C₂₀alkynyl, aryl, C₁-C₂₀carboxylate, C₁-C₂₀alkoxy, C₂-C₂₀alkenyloxy, C₂-C₂₀alkynyloxy, aryloxy, C₂-C₂₀alkoxycarbonyl, C₁-C₂₀alkylthio, C₁-C₂₀alkylsulfonyl and C₁-C₂₀alkylsulfinyl. Optionally, each of the R or R¹substituent group may be substituted with one or more moieties selected from the group consisting of C₁-C₁₀alkyl, C₁-C₁₀alkoxy, and aryl which in turn may each be further substituted with one or more groups selected from a halogen, a C₁-C₅alkyl, C₁-C₅alkoxy, and phenyl. Moreover, any of the catalyst ligands may further include one or more functional groups. Examples of suitable functional groups include but are not limited to: hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiimide, carboalkoxy, carbamate, and halogen.

[0027] In preferred embodiments of these catalysts, the R subistituenlt is hydrogen and the R¹substituent is selected from the group consisting of C₁-C₂₀alkyl, C₂-C₂₀alkenyl, and aryl. In even more preferred embodiments, the R¹substituent is phenyl or vinyl, optionally substituted with one or more moieties selected fiom the group consisting of C₁-C₅alkyl, C₁-C₅alkoxy, phenyl, and a functional group. In especially preferred embodiments, R¹is phenyl or vinyl substituted with one or more moieties selected from the group consisting of chloride, bromide, iodide, fluoride, —NO₂, —NMe₂, methyl, methoxy and phenyl. In the most preferred embodiments, the R¹substituent is phenyl or —C═C(CH₃)₂.

[0028] In preferred embodiments of these catalysts, L and L¹are each independently selected from the group consisting of phosphine, sulfonated phosphine, phosphite, phosphinite, phosphonite, arsine, stibine, ether, amine, amide, imine, sulfoxide, carboxyl, nitrosyl, pyridine, and thioether. In more preferred embodiments, L and L¹are each a phosphine of the formula PR³R⁴R⁵, where R³, R⁴, and R⁵are each independently aryl or C₁-C₁₀alkyl, particularly primary alkyl, secondary alkyl or cycloalkyl. In the most preferred embodiments, L and L¹ligands are each selected from the group consisting of -P(cyclohexyl)₃, -P(cyclopentyl)₃, -P(isopropyl)₃, and -P(phenyl)₃.

[0029] In preferred embodiments of these catalysts, X and X¹are each independently hydrogen, halide, or one of the following groups: C₁-C₂₀alkyl, aryl, C₁-C₂₀alkoxide, aryloxide, C₃-C₂₀alkyldiketonate, aryldiketonate, C₁-C₂₀carboxylate, arylsulfonate, C₁-C₂₀alkylsulfonate, C₁-C₂₀alkylthio, C₁-C₂₀alkylsulfonyl, or C₁-C₂₀alkylsulfinyl. Optionally, X and X¹may be substituted with one or more moieties selected from the group consisting of C₁-C₁₀alkyl, C₁-C₁₀alkoxy, and aryl which in turn may each be further substituted with one or more groups selected from halogen, C₁-C₅alkyl, C₁-C₅alkoxy, and phenyl. In more preferred embodiments, X and X¹are halide, benzoate, C₁-C₅carboxylate, C₁-C₅alkyl, phenoxy, C₁-C₅alkoxy, C₁-C₅alkylthio, aryl, and C₁-C₅alkyl sulfonate. In even more preferred embodiments, X and X¹are each halide, CF₃CO₂, CH₃CO₂, CFH₂CO₂, (CH₃)₃CO, (CF₃)₂(CH₃)CO, (CF₃)(CH₃)₂CO, PhO, MeO, EtO, tosylate, mesylate, or trifluoromethanesulfonate. In the most preferred embodiments, X and X¹are each chloride.

[0030] The catalysts of the present invention are as described above except that L¹is an unsubstituted or substituted triazolylidene of the general formula:

[0031] wherein:

[0032] R⁶and R⁷are each independently hydrogen or a substituent selected from the group consisting of C₁-C₂₀alkyl, C₂-C₂₀alkenyl, C₂-C₂₀alkynyl, aryl, C₁-C₂₀carboxylate, C₁-C₂₀alkoxy, C₂-C₂₀alkenyloxy, C₂-C₂₀alkynyloxy, aryloxy, C₂-C₂₀alkoxycarbonyl, C₁-C₂₀alkylthio, C₁-C₂₀alkylsulfonyl and C₁-C₂₀alkylsulfinyl. In addition, each of the R⁶and R⁷substituent groups may be substituted with one or more moieties selected from the group consisting of C₁-C₁₀alkyl, C₁-C₁₀alkoxy, and aryl which in turn may each be further substituted with one or more groups selected from a halogen, a C₁-C₅alkyl, C₁-C₅alkoxy, and phenyl. Moreover, each of the R⁶and R⁷substituent groups may further include one or more functional groups. Examples of suitable functional groups include but are not limited to: hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiimide, carboalkoxy, carbamate, and halogen.

[0033] Optionally, R⁶and R⁷together may form a cycloalkyl or an aryl moiety. In preferred embodiments, R⁶and R⁷are both hydrogen or phenyl, or R⁶and R⁷together form a cycloalkyl group.

[0034] Without being bound by theory, it is believed that bulkier R⁶and R⁷groups result in catalysts with improved characteristics such as thermal stability. In even more preferred embodiments, R⁶and R⁷are the same and each is independently of the formula

[0035] wherein:

[0036] R⁸, R⁹, and R¹⁰are each independently hydrogen, substituted or unsubstituted C₁-C₁₀alkyl, C₁-C₁₀alkoxy, aryl, or a functional group selected from hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiumide, carboalkoxy, carbamate, and halogen. In the most preferred embodiments, R⁸, R⁹, and R¹⁰are the same and are each methyl.

[0037] As stated above, R¹may be hydrogen or a substituent selected from the group consisting of C₁-C₂₀alkyl, C₂-C₂₀alkenyl, C₂-C₂₀alkynyl, aryl, C₁-C₂₀carboxylate, C₁-C₂₀alkoxy, C₂-C₂₀alkenyloxy, C₂-C₂₀alkynyloxy, aryloxy, C₂-C₂₀alkoxycarbonyl, C₁-C₂₀alkylthio, C₁-C₂₀alkylsulfonyl and C₁-C₂₀alkylsulfinyl. Optionally, the R¹substituent group may be substituted with one or more moieties selected from the group consisting of C₁-C₁₀alkyl, C₁-C₁₀alkoxy, and aryl which in turn may each be further substituted with one or more groups selected from a halogen, a C₁-C₅alkyl, C₁-C₅alkoxy, and phenyl. Moreover, any of the catalyst ligands may further include one or more functional groups. Examples of suitable functional groups include but are not limited to: hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiimide, carboalkoxy, carbamate, and halogen. In preferred embodiments, R¹is either substituted or unsubstituted aryl.

[0038] A preferred embodiment of the catalyst is of the general formula:

[0039] Another embodiment of the invention includes both L and L¹being an unsubstituted or substituted triazolylidene. In this embodiment, R, R¹, R⁶, R⁷, X and X¹are as described above.

[0040] Metathesis Reactions

[0041] The catalysts of the invention may be used for any metathesis reaction (i.e. ring opening metathesis polymerization, ring closing metathesis, cross metathesis, etc.) by contacting the inventive catalysts with an appropriate olefin. The RCM activity of an exemplary catalyst complex 14 in comparison to the Schrock (1) and Grubbs (2) catalysts is summarized in Table 1. Specifically, Table 1 shows the percent conversion in the RCM of substrates with 1, 2, and 14. This series of substrates was chosen in order to probe the ability of these new complexes to initiate ling-lo sing metathesis (RCM), and also to determine if they have any effect on the stereoselectivity of the RCM reaction. As seen, E=CO₂Et where the percent conversion and E:Z ratios were determined by ¹H NMR Integration. The conditions under which the reactions took place for catalyst 1 are as follows: 5 mol % catalyst, 0.10 in C₂D₆, 65° C. The conditions under which the reactions took place for catalyst 1 are as follows: 5 mol % catalyst, 0.05M CD₂Cl₂, 40° C.

[0042] In the case of di ethyl diallylmalonate, the first substrate, initial screening revealed that the activity of complex 14 is relatively slow at room temperature (25° C.). For example, with a catalyst loading of 5 mol % at a concentration of 0.05M in CD₂Cl₂, the Grubbs catalyst (2) takes less than 30 minutes to complete the cyclization, while complex 14 takes 8 hours to accomplish the same conversion. However. the activity of 14 increased dramatically at elevated temperatures. At ˜40° C., it is able to cyclize diethyl diallylmalonate within 30 minutes.

[0043] Complex 14 showed superior activity for even more sterically challenging substrates. While complex 2 is not able to ring-close 2-t-butyl diethyl diallylmalonate, 4,4′-dicarbethoxy-2,6-dimethyl-1,6-heptadiene, or 4,4′-dicarbethoxy-2,7-dimethyl-1,7-octadiene, 14 was active for this transformation (Table 1). The conversions at ˜40° C. ranged from moderate (53%) to very good (85%) for these substrates. This activity is significant because, up until now, the only metathesis catalyst able to ring-close highly substituted dienes to yield tri- and tetra-substituted olefins has been the Schrock alkylidene (1). In fact, the new catalyst (14) shows improved activity over the Schrock molybdenum alkylidene in the RCM of 4,4′-dicarbethoxy-2,7-dimethyl-1.7-octadiene (82% conversion for 14 vs. 61% for 1).

[0044] The RCM of triethylene glycol diallyl ether was tested in order to determine if this new catalyst had any effect on cis,trans-stereoselectivity. Complex 14 produced a product mixture with an E:Z ratio of approximately 2.3:1, which is only somewhat greater than the ratio of 1.6:1 observed for complex 2.¹⁷Interestingly, when the reaction with catalyst 14 was allowed to proceed at room temperature (44%. conversion after 20 hours), the observed E:Z ratio was 5:1. The relatively high proportion of trans product is due to the reaction reaching equilibrium during the extended reaction time.

[0045] In addition to RCM activity, the novel catalyst (e.g., complex 14) is able to catalyze the ROMP of cyclic olefins. Because these new ruthenium-based olefin metathesis catalysts coordinated with N-heterocyclic carbene ligands display such high activity, they are capable of mediating the ROMP of more challenging monomers, such as those that are less-strained and sterically bulky. For example, these new catalysts can ROMP trisubstituted cyclooctene derivatives and cyclooctatetraene derivatives, toward which the parent bis(phosphine) catalyst 2 is inactive. In addition, these new catalysts are capable of polymerizing the more highly strained and sterically less bulky monomers traditionally polymerized by 2, such as cyclopropenes, cyclobutenes, benzocyclobutenes, cyclopentenes, cyclopentadiene oligomers, cyclohexenes, cycloheptenes, cyclooctenes, cycooctadienes, norbomenes, norbornadienes, [2.2.1]bicycloheptenes, [2.2.2]bicyclooctenes, cyclohexenylnorbornenes, and norbomene dicarboxylic anhydrides. Preferably, the novel catalyst is able to catalyze the ROMP of dicyclopentadiene. When just a fractional mol % of catalyst is added to vigorously stirred, warmed (˜40° C.) dicyclopentadiene, the polymerization reaction occurs within minutes to yield a solid mass of yellow material. For example, Scheme 2 shows the use of 0.1 mol % of complex 14 to ROMP dicyclopentadiene (DCPD).

[0046] Further, complex 14 was more active in situ for ROMP than the general Grubbs complex 2 with L=PPh₃. For example, the PPh₃derivative of complex 2 is not active enough to initiate the ROMP of DCPD. However, as shown in Scheme 3, when the PPh₃derivative of complex 2 was combined with the triazolylidene alkoxide adduct 12 in DCPD and warmed to 60-70° C., a more active catalyst formed in situ that was able to perform the ROMP reaction.

[0047] This type of complex has been shown to be remarkably more active at elevated temperatures for metathesis than the parent complex (PCy₃)₂Cl₂Ru(CHPh) (2). For example, complex 14 (L=1,3,4-triphenyl-4,5-dihydro-1H-triazol-5-ylidene) is able to ring-close highly substituted dienes to give tetra-substituted cyclic olefins in good yields. To date, this reaction has not been possible with complex 2. In addition to enhanced activity, complex 14 demonstrates the analogous stability towards oxygen and moisture that is exhibited by complex 2. Complex 14 is also able to initiate the ring-opening metathesis polymerization of dicyclopentadiene. Thus, it has been possible to combine the many desirable properties of the ruthenium alkylidene (2) with the higher activity of the Schrock molybdenum alkylidene (1) in a new catalyst.

[0048] Synthesis

[0049] In general, the catalysts of the present invention are prepared by substituting one of the L groups in the previously described ruthenium catalysts (wherein L=L¹), LLXX¹M=CRR¹, with a tliazolylidene ligand. Scheme 4 shows the currently accepted mechanism of olefin metathesis by ruthenium alkylidenes.

[0050] The proposed mechanism in Scheme 4 has been determined through extensive kinetic and reactivity studies, and it corroborates well with recently determined thermochemical data. As shown, the main steps in the reaction involve: 1) olefin coordination cis to the alkylidene, ii) alkylidene rotation and dissociation of a phosphine, iii) formation of a metallacyclobutane intermediate, and finally, iv) formal retrocycloaddition to complete the metathesis reaction.

[0051] The most important aspect about the above mechanism that is pertinent to this discussion is that the most active species is a mono (phosphine) intermediate, which requires Ru—P bond cleavage. Consistent with this mechanism, it has been found that both more sterically bulky and more electron donating phosphines increase the catalyst activity, since these properties facilitate phosphine dissociation and/or stabilization of the mono (phosphine) intermediate. In addition, it has been proposed that more electron donating phosphines also favor the formation of the metallocyclobutane intermediate. Using this concept, the ruthenium alkylidene complex with an N-heterocyclic ligand of the invention would be a more active metathesis catalyst because of the ligand's strong electron donating character. Furthermore, it is proposed that the more liabile phospine would dissociate, giving a mono (N-heterocyclic carbene) intermediate as the active species.

[0052] As ligands, the N-heterocyclic carbenes are characterized as pure sigma-donors. As a result, the N-heterocyclic carbenes are able to form complexes with all metals, from transition elements in both high and low oxidation states to the main group metals and the lanthanides. Several studies have focused on the applications and advantages of these ligands in comparison to conventional donor ligands, such as tertiary phosphines. In a number of cases, the metal-carbene bond has been found to be remarkably robust. For example, the palladium complex L₂PdI₂where L=1,3-dimethyl-2,3-dihydro-1H-imidazol-2-ylidene, remained intact within the presence of oxygen in boiling THF for several days.

[0053] Thus, the N-heterocyclic carbene ligands incorporate several characteristics that would be advantageous in making an improved olefin metathesis catalyst. Most importantly, these ligands are stronger electron donors than phosphines, which according to the accepted mechanism of olefin metathesis, should make a mono-substituted analog of the type shown in FIG. 3 even more active than the parent complex (2). Previous work has also indicated that metal-carbene bonds of this kind are thermally stable and resistant to dissociation. Finally, the synthesis of N-heterocyclic carbene derivatives is relatively straightforward, so it would be possible to rapidly investigate the effects of steric and electronic changes to the ligand on catalyst activity.

[0054] Thus, considering the electron donating characteristics of these ligands, the invention provides a method for attaching a N-heterocyclic carbene ligand to the metal center of a ruthenium metathesis catalyst. The preferred N-heterocyclic carbene ligand is triazolylidene. In the first step, a salt is prepared from an N-mono(substituted) amide as follows:

[0055] Next, and as shown in Scheme 6, the salt may be converted to a substituted or unsubstituted primary alkyl oxide, for example, a C₁-C₃alkyl oxide. Moreover, the primary alkyl oxide may include one or more functional groups. Examples of suitable functional groups include but are not limited to: hydroxyl, thiol, thioether, ketone, aldehyde, ester, ether, amine, imine, amide, nitro, carboxylic acid, disulfide, carbonate, isocyanate, carbodiimide, carboalkoxy, carbamate, and halogen. Preferably, the salt is converted to a methoxide. In the preferred embodiment, the methoxide is then reacted with reacted with LLXX¹M=CRR¹(previously described ruthenium metathesis catalyst) to yield a triazolylidene based ruthenium carbene catalyst.

[0056] Alternatively, the product may be made directly from the salt by generating the carbene in situ as shown in Scheme 7.

[0057] Finally, the free carbene can be reacted with LLXX¹M=CRR¹(previously described ruthenium metathesis catalyst) to yield a triazolylidene based ruthenium carbene catalyst as shown in Scheme 8.

[0058] Examples of other N-heterocyclic carbenes that may be used with the invention are:

[0059] For the purposes of clarity, the specific details of the invention will be illustrated with reference to especially preferred embodiments. However, it should be appreciated that these embodiments and examples are for the purposes of illustration only and are not intended to limit the scope of the invention.

[0060] Synthesis of (PCy₃)(L)Cl₂Ru(CHPh)[L=1,3,4-triphenyl-4,5-dihydro-1H-triazol-5-ylidene]. The target was chosen to be a mono-substituted complex containing a triazolylidene ligand (14). The synthesis of the 1,3,4-triphenyl substituted tuiazolylidene (13) has been previously reported. As shown in Scheme 9, the route to the triazolium salt involves the preparation of N-phenylbenzimidoyl chloride (9) from benzanilide (8), followed by reaction with phenylhydrazine to give a mixture of N-phenylbenzamide phenylhydrazone (10) and N-amino-N,N-diphenyl benzamide. The N-phenylbenzamide phenylhydrazone is then converted to the triazolium salt by cyclocondensation with formic acid. The perchlorate salt (11) is obtained by treatment with perchloric acid. Finally, the triazoliuni salt can then be readily converted to the methoxide (12), as shown in Scheme 10, which eliminates methanol upon heating under reduced pressure to give the free carbene (13).

[0061] One equivalent of the isolated carbene reacted with (PCy₃)(L)CI₂Ru(CHPh) (2) in toluene at room temperature to give the mono-substituted complex 14 (Scheme 10). The ligand exchange occurred remarkably quickly (within 5 minutes). Alternatively, it was found that the product could be made directly from the reaction of the methoxide (12) and 2 in toluene at 80° C. for 5-10 minutes. This route was used to prepare complex 14 on a 0.6 g scale in 59% yield. The free PCy₃byproduct could be readily separated by crystallization of the product from pentane at −78° C. Furthermore, the product could also be made directly from the triazolium salt (11) by generating the carbene in situ as shown in Scheme 10.

[0062] The product of these reactions (14) is actually a mixture of two species. By ¹H NMR, two resonances for alkylidene alpha-protons were observed, one at 20.04 ppm (doublet, ³J_HP=9.6 Hz % 1,=9.6 Hz) and the other at 19.84 ppm (doublet, ³J_HP=7.5 Hz). These two resonances appeared consistently in a 1.5 to 1 ratio. The fact that the two alkylidene moieties exhibit different coupling constants is consistent with the products being conformational isomers, in which only the orientations of the triazolylidene ligand and the alkylidene moiety are different. Likewise, the ³¹P NMR. showed two resonances, one at 24.14 and the other at 23.04 ppm (both singlets). The composition of the products was further supported by high resolution mass spectrometry data, which showed only one product molecular ion peak. Attempts have not yet been made to separate the isomers. Like complex 2, this product is also air stable as a solid.

[0063] General Considerations. All manipulations were carried out using standard Schlenk and vacuum line techniques or in a nitrogen filled drybox. Diethyl diallylmalonate was obtained from Aldrich and degassed before use. Additional RCM substrates were obtained from other group members. Dicyclopentadiene (98%) was obtained from Aldrich and used as received. (PCy₃)₂Cl₂Ru(CHPh) (2) was obtained from a group supply. 1,3,4-Triphenyl-4,5-dihydro-1H-triazol-5-ylidene (13) and its intermediates were prepared as previously reported. ¹⁵NMR solvents were used as received from Cambridge Isotopes. NMR experiments were performed in J. Young gas NMR tubes. ¹H NMR spectra were recorded on a General Electric QE-300 spectrometer (300.1 MHz for ¹H). ³¹P spectra were recorded on a JEOL JNM-GX400 spectrometer (161.9 MHz ³¹P). Chemical shifts are reported in parts per million (ppm) downfield from tetramethylsilane with reference to internal solvent for proton spectra. Phosphoric acid was employed as the internal solvent for phosphorus spectra. Mass spectral analysis was performed at the Southern California Mass Spectrometry Facility (University of California at Riverside).

[0064] Synthesis of (PCy₃)(L)Cl₂Ru(CHPh)[L=1,3,4-Triphenyl-4,5-dihydro-1H-triazol-5-ylidene] (14). In the glovebox, a Schlenk was charged with 0.500 g (0.608 mmol) of (PCy₃)₂Cl₂Ru(CHPh) (2) and 0.195 g (0.592 mmol) of triazolylidene methoxide (12), and then 17 mL toluene were added. The reaction was stirred at room temperature for 20 minutes and at 80° C. for 20 minutes. The brown solution was pumped down under vacuum. Then 100 mL dry pentane was added, and the mixture was gently wanned to dissolve as much material as possible. Upon cooling this solution to −78° C., a tan colored precipitate formed. The supernatant was filtered off via cannula and the product dried under vacuum. Yield=0.293 g (59%). ¹H NMR (C₆D₆) 20.04 (d, ³J_HF=9.6 Hz), 19.84 (d, ³J_HF=7.5 Hz); ³¹P NMR (C₆D₆) 24.14, 23.04; HRMS (FAB) C₄₅H₅₄Cl₂N₃PRu [M] 839.2476, found 839.2450.

[0065] Typical RCM experiment. In the glovebox, 5 mol % of complex 14 dissolved in CD₂Cl₂was added to the appropriate amount of a 0.05M solution of substrate in CD₂Cl₂. This solution was mixed and 0.75 mL of it was transferred to an NMR tube, which was then sealed. The reaction was either heated at ˜40° C. for 30 minutes or left at room temperature, and it was monitored periodically by ¹H NMR.

[0066] Typical ROMP experiment. Approximately 2-3 mgs of complex 14 dissolved in a minimal amount of toluene (0.1-0.2 mL) was added to a small beaker containing 25 mL of vigorously stirred, warmed (40-50° C.) dicyclopentadiene. Reaction occurted within minutes to yield a solid block of yellow polymer.