William D. Jones, Principal Investigator
Phone: 716-275-5493

Contract No. DE-FG02-86ER13569

Project Period: December 1, 1995 - November 30, 1998
Total Award Amount: $ 379,251

Research Plans for the Period December 1, 1996 - November 30, 1997:

This coming year, we will focus our research on three of the items presented in our proposal. These include: (1) carbon-carbon bond cleavage reactions, (2) synthesis of new complexes for hydrogenolysis reactions, and (3) C-H bond cleavage reactions of tris-pyrazolylboraterhodium complexes. We have made progress in each of these areas over the past year, as described in our report, and will continue our studies in these areas.

Our carbon-carbon bond cleavage study is based upon the notion that metal-phenyl bonds are the strongest metal-carbon bonds. Cleavage of the C-C bonds in biaryl systems will therefore give two very strong metal-aryl bonds, and consequently offers the most thermodynamically preferred situation for observing C-C cleavage. While C-H activation is anticipated to compete kinetically, the interconversion with the h2-arene complexes should permit migration around the ring to ultimately allow insertion of the electron rich fragment into the C-C bond. There are currently no examples in the literature of insertion into unstrained, free carbon-carbon bonds, and this work could have an impact on C-C cleavage in aromatic networks in coal.

We have had some success in C-C cleavage with biphenylene, a molecule with a weaker C-C bond than biphenyl. The success has to do with the use of a new metal system for C-C cleavage, Pt(PEt3)3. This complex has a labile phosphine ligand and a metal center that is reactive towards oxidative addition. In addition, it is less sterically hindered than the [Cp*Rh(PMe3)] fragment that we studied earlier. This feature should be beneficial in attacking hindered biaryl C-C bonds where the rhodium system failed. We will also look at other derivatives in this class of compounds, including precursors for the intermediate [Pt(chelating phosphine)]. Control of the bite angle of the chelate has been found to be an important factor in tailoring the reactivity of the PtL2 fragment, and we will take advantage of knowledge of these factors.

The second area that we will become involved in is the synthesis of new catalysts for hydrogenolysis based upon what we believe to be occurring in the biphenylene system with Cp*Rh(PMe3)H2. Our proposal is that square planar d8 bis-olefin phosphine hydride complexes are the active intermediates, and we will target several of these for study. We have not yet had the opportunity to explore this avenue of work.

The third area for investigation involves a continuation of our studies with the tris-pyrazolylborate fragment [Tp'Rh(CNCH2CMe3)] and the examination of new derivatives of the type [Tp'Rh(L)]. Earlier work has allowed us to establish the relative stabilities of various hydrocarbon adducts of this fragment. During the past year, a breakthrough in synthetic methodology has allowed us to prepare generally a variety of long sought derivatives of the isocyanide species as well as to prepare derivatives in which L can be varied. We will first look at the analogous phosphine systems for comparison with the isocyanide system. We will also look at rearrangements of alkyl hydride complexes that we were unable to prepare until now.

Progress Report for the Period
December 1, 1995- November 30, 1996.

1. Tris-pyrazolylborate Studies.

During the first year of this grant cycle, we have developed methodology for the independent synthesis of a variety of new complexes that arise from C-H activation, (HBPz*3)Rh(CNCH2CMe3)(R)H. Many C-H activation products have been identified in earlier studies using a carbodiimide derivative which photodissociates with unit quantum efficiency, thereby serving as a useful precursor for the 16-electron Rh(I) fragment [(HBPz*3)Rh(CNCH2CMe3)] (= [Tp'Rh(CNR)]). This fragment activates (via oxidative addition) a wide range of aromatic and aliphatic C-H bonds to give the corresponding aryl or alkyl hydride complexes (Scheme I). The mechanism of activation of benzene by this fragment has been studied in some detail.

Scheme I:

Reaction of the [Tp'Rh(CNR)] fragment with a variety of aliphatic hydrocarbons has also been investigated. We have been able to establish the relative thermodynamic stability of the alkyl and aryl hydride complexes and have determined the relative metal-carbon bond strengths.

During the past year we have developed general synthetic methods for the preparation of alkyl chloride derivatives of the trispyrazolylborate complex and have learned how to convert these cleanly into alkyl hydride complexes. This methodology is important in that it allows us to prepare complexes that cannot be made using the photochemical route. The new method involves the use of Cp2ZrH2 as a hydride reducing agent, converting a chloro derivative into the hydride derivative and generating Cp2ZrCl2. The zirconium product is easily removed via chromatography on silica gel, if desired, or left in the reaction mixture as an innocuous bystander. This method offers distinct advantages over the previous methods using borohydride or aluminum hydride reagents in that no work-up is required, allowing the initial products formed to be examined immediately. The reaction of the rhodium chloro complex with Cp2ZrH2 is bimolecular, and can be made to proceed more rapidly by increasing the zirconium reagent concentration.

We have now been able to cleanly and rapidly synthesize the methyl deuteride complex Tp'Rh(CNR)(CH3)D and watch it interconvert with the isomer Tp'Rh(CNR)(CH2D)H prior to the elimination of methane. The complex Tp'Rh(CNR)(CD3)H has been prepared and behaves similarly. As these rearrangements occur in benzene solvent, they indicate that an alkane 's complex' is present along the reaction coordinate for reductive elimination. We have measured the rates of approach to equilibrium for each of these complexes also, and have done a complete kinetic analysis for obtaining the critical rate constant kfH for the formation of a methane s-complex from a methyl hydride complex. The analysis gives a value of 1.41 x 10-4 s-1 for kfH and a measure of the ratio krH/krD = 1.69. The reactions involved are shown below.

Another interesting observation was made with the methyl hydride system providing the first example of kinetic evidence for a methane s- complex. The elimination of CH4 from Tp'Rh(CNR)(CH3)H was found to be slower than the elimination of CD4 from Tp'Rh(CNR)(CD3)D, consistent with an inverse kinetic isotope effect as observed in many alkane reductive elimination reactions. However the elimination of CH4 from Tp'Rh(CNR)(CH3)H was found to be different in C6H6 than in C6D6 solvent, suggesting that the solvent was somehow involved in the transition state for methane elimination. This notion is contrary to the expected pathway of first order reductive elimination of methane to generate a s-complex which would then undergo unimolecular loss of methane.

This hypothesis was tested by varying the concentration of benzene in inert C6F6 solvent. The rate of methane formation was found to decrease as the concentration of benzene decreased. Examination of a plot of kobs vs [C6D6] showed a slight curvature, but a plot of 1/kobs vs 1/[C6D6] was found to be linear, suggesting a kinetic scheme involving a unimolecular equilibrium followed by a rate determining bimolecular step.

A scheme that is consistent with these observations is shown below. The reversible formation of the methane s-complex is proposed to form an h2-Tp' complex, since Rh(I) is d8and would favor a square planar geometry. Bimolecular substitution with benzene leads to an h2-C6D6 complex that would then activate the benzene C-H bond. This is the first example of kinetic evidence for a methane s-complex.

A second important question we have been able to address is the kinetic selectivity for activation of primary vs secondary C-H bonds in an alkane such as propane. Irradiation of the carbodiimide complex Tp'Rh(CNR)(PhN=C=NR) in liquid propane at -50 °C generated only the primary product Tp'Rh(CNR)(n-propyl)H. It was not possible to say, however, if the secondary activation product Tp'Rh(CNR)(i-propyl)H had also been produced but rearranged to the primary derivative under the reaction conditions. We have now shown that the primary activation product is indeed the only product formed by synthesizing the secondary complex and demonstrating that the rate of rearrangement to the primary complex is slow under the reaction conditions employed in the photolysis experiment.

The experiment performed involved the synthesis of the isopropyl chloride complex Tp'Rh(CNR)(i-propyl)Cl from Tp'Rh(CNR)Cl2 plus isopropyl Grignard. Treatment of this product with Cp2ZrH2 in benzene solvent leads to the clean formation of the i-propyl derivative, which then rearranges to the n-propyl derivative, which finally forms the more stable phenyl hydride product. In modeling the kinetic scheme, however, it was necessary to include a pathway for the direct conversion of the i-propyl hydride complex to the phenyl hydride complex. The reactions involved and the modeling of the distribution of species are indicated in the figure below.

We have also investigated a series of reactions of the reactive Tp'Rh(CNR) fragment with cyclopropane. Photochemical generation of this fragment in liquid cyclopropane using the carbodiimide precursor leads to the clean formation of the C-H activation product. Upon standing, this product rearranges to the C-C insertion product in benzene solution (see following scheme). The fact that the phenyl hydride complex is not produced indicates an intramolecular rearrangement, once again arguing for the intermediacy of a s-alkane complex. The C-H insertion product has been converted to the chloro derivative and characterized by X-ray diffraction. The C-C insertion product has also been characterized by X-ray diffraction.

In addition to this C-H and C-C activation chemistry, we have found that the Rh(I) fragment reacts with t-butylethylene and ethylene to give vinylic C-H activation products. The vinyl hydride complex rearranges intramolecularly to the h2-ethylene complex, but the t-butyl-ethylene complex loses the olefin to generate the phenyl hydride complex. Reaction of the Rh(I) fragment with propene and isobutylene leads to formation of the allylic C-H activation products. The former rearranges intramolecularly to the h2-propene complex, whereas the latter loses the olefin and activates benzene.

The activation of the vinylic C-H bond of t-butylethylene and its direct elimination in benzene to give the phenyl hydride complex allow this substrate to be used for the comparison of the C-H and Rh-C bond strengths. The kinetic preference of the Rh(I) fragment for benzene over olefin activation was independently measured as 10:1. In addition, the activation and elimination of isobutylene allow the inclusion of an allylic C-H bond in this analysis. The results are indicated in the plot below, which shows two important points: (1) the correlation is not linear, but curved, and (2) the slope of the best line through the data is not 1. This latter point means that equilbrium constants for different hydrocarbons will vary tremendously. The curvature of the line means that there will be a thermodynamic preference for one hydrocarbon relative to another if it lies above a reference line of slope 1.

We have also completed a project investigating the reaction of [(C5Me5)Rh(PMe3)] with a variety of 1,3-disubstituted benzenes. We have noted a trend in the activation parameters for the reductive elimination of arenes in the series of complexes (C5Me5)Rh(PMe3)(3,5-C6H3R2)H. While the origin of this trend is not yet fully understood and further mechanistic studies are underway to examine possible explanations. Alternative transition states and substitution mechanisms will be investigated.

2. C-C Bond Cleavage Studies

A new area for study that we have initiated is our work with C-C bond cleavage. This field has received little attention, and we believe that we have discovered a strategy for tackling this difficult problem. Our hypothesis is that metal complexes should prefer to add to the stronger C-C bonds found in biaryls, since two strong metal-aryl bonds will be formed in the product. Initial studies have been carried out with biphenylene, and we have discovered that the rhodium fragment [Cp*Rh(PMe3)] inserts first into the C-H bond, and then into the aryl-aryl C-C bond as indicated below.

We have also explored the use of the bis-olefin complexes Cp*M(C2H4)2 where M = Rh or Co for reactivity with biphenylene. Thermal reaction of the rhodium complex also gives a product in which C-C bond cleavage has occurred, and in which a second metal has coordinated to the metallacyclopentadiene. These complexes react with a variety of small molecules, including PMe3, CO, and H2. The latter results in the formation of biphenyl, thereby accomplishing a net hydrogenolysis of the C-C bond of biphenylene.

We have also found that these complexes can serve as catalysts for the carbonylation of C-C bonds in the presence of CO. Ketones are formed where C-C bonds once existed. While the reaction only has been demonstrated with biphenylene, it will be extended to other molecules as well.

We have recently initiated studies with a platinum(II) complex that can cleave the C-C bond of biphenylene. Reaction of Pt(PEt3)3 with biphenylene leads initially to the formation of the C-C insertion product Pt(PEt3)2(2,2'-biphenyl). Upon further heating, this complex goes on to a complex dinuclear product. In the presence of excess biphenylene, the complex is a catalyst for the C-C cleavage and coupling to generate an 8-membered ring product.

In summary, the cleavage of C-C bonds looks to be an exciting new area for discovery and development. We will extend our studies with biphenylene to other biaryl systems with stronger C-C bonds, with the ultimate target being the cleavage of biphenyl. New catalysts will be developed for hydrogenolysis, and reactions with other small molecules will be examined as a means of transforming the bis-aryl metal complexes that are formed.