OVERDUE FINES: 25¢ per day per item \ RETURNING LIBRARY MATERIALS: xmx§ I :1; {MIN} 1‘.- ' ~J~ 1"” , 1 Place in book return to remve . Q‘ ‘\ 1/ L" charge from circulation records PART I PREPARATION AND APPLICATION OF POLYMER-SUPPORTED CATALYSTS PART II THERMAL DECOMPOSITION OF TRANSITION METAL ALKYLS By Biau-Hung Chang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1979 ABSTRACT PART I PREPARATION AND APPLICATION OF POLYMER-SUPPORTED CATALYSTS PART II THERMAL DECOMPOSITION 0F TRANSITION METAL ALKYLS By Biau-Hung Chang The industrial application of homogeneous catalysts has been expanded greatly in recent years. However, homogeneous catalysts are less widely used than heterogeneous catalysts in the chemical industry mainly due to the difficult separation of the very expensive catalysts from the reaction products. Therefore, heterogenizing homogeneous catalysts by attaching homogeneous catalysts to polymer supports is a significant step in improving their utility and versatility. In this research, the attachment of transition metal complexes of Ti, Zr, Hf, Nb, Ta. Co, and Rh to 20% cross-linked macroreticular polystyrene-divinylbenzene copolymer beads by cyclopentadienyl ligands has been developed. The organometallic polymer beads have been tested for a variety of catalytic activity. Biau-Hung Chang The polymer-attached zirconium and hafnium complexes were effective in hydrogenation of olefins and acetylenes, isomerization of allylbenzene and l.5-cyclooctadiene. and epoxidation of cyclohexene. The zirconium beads have also been used in hydrozirconation to produce terminal aldehydes. Further observations showed that supported metal- locene derivatives were more effective in catalysis than the related compounds in homogeneous solutions and their activities were in the decreasing order, Ti>Zr>Hf. Also. polymer-attached methylene-bridged titanocene dichloride beads were prepared and used for the study of nitrogen fixation. The pentavalent derivatives of niobium and tantalum containing beads were prepared by the reaction of their pentachlorides with the beads containing tin alkyls. They were active catalysts in hydro- genation of diphenylacetylene. and isomerization of allylbenzene. The tantalum beads were also an active catalyst in dimerization of ethylene. Mononuclear complexes of cyclopentadienyl cobalt and rhodium dicarbonyls covalently attached to a polystyrene-divinylbenzene copolymer support were prepared from their carbonyl derivatives and cyclopentadiene-substituted beads. The beads have been tested for a number of catalytic reactions. The rhodium containing beads were effective in hydrogenation of olefins, aldehydes and ketones, isomeri- zation of olefins. disproportionation of l,4-cyclohexadiene and cyclohexene. cyclotrimerization of ethyl propiolate, and hydroformy- lation of l-pentene and l-hexene. Decomposition of the rhodium catalysts occurs except in hydroformylation. although only slight Biau-Hung Chang loss of the carbonyl groups and catalytic activity was observed in cyclotrimerization. The rhodium catalyst appears to be the first example of cyclopentadienyl coordination compound of Rh active in hydroformylation catalysis. The cobalt containing beads have proven to be inactive except in case of the cyclotrimerization of ethyl propiolate. The second part of this research was a study of the thermal decom- position of some early transition metal alkyl complexes such as CpZTiR2 (R=n-alkyl, neopentyl, neohexyl). and CpZVR, szTaCle, szMoR2 and szwR2 (R- neohexyl). Two new mechanisms of the thermoly- sis, y- and 6-hydrogen eliminations, which involve metallacycles as the intermediates have been proposed and discussed. To My Wife and My Family. ACKNOWLEDGMENTS I would like to express my sincere gratitude to my research preceptors, Professors Carl H. Brubaker, Jr., and Robert H. Grubbs, for their guidance, patience, support and inspiration during the course of this study. I would also like to thank Dr. Akira Miyashita, Dr. Larry W. Shive, Dr. Devinder Gill and my fellow graduate students for all the help, discussions and friendship. Also, I am very grateful to my parents, my brothers, my sisters, my parents-in-law and my sisters-in-law for their love. encouragement and support. Finally, I wish to express my deepest appreciation to my wife. Yue-Yeh, for her unlimited love, encouragement and understanding. ii TABLE OF CONTENTS Chapter Page LIST OF TABLES. O O O O O O O O O O O O O O O O O O O O O O 0 1V LISTOFFIGURES.00000000000000000000o '1 PART I. PREPARATION AND APPLICATION OF POLYMER-SUPPORTED CATAL YST S O O O O O O O O O O O I O O O O O O C O O O 1 Introduction . . . . . . . . . . . . . . . . . . . . 1 Results and Discussion . . . . . . . . . . . . . . . ll Experimental . . . . . . . . . . . . . . . . . . . . 76 PART II. THERMAL DECOMPOSITION OF TRANSITION METAL ALKYLS . . 98 Introduction . . . . . . . . . . . . . . . . . . . . 98 Results and Discussion . . . . . . . . . . . . . . . l09 Experimental ...... . . . . . . . . . . . . . . l22 REFERECES ......... O O O O O O O O O O O O O O O O O 130 iii Table 10 ll 12 13 LIST OF TABLES Page Far infrared bands (cm']) of polymer-attached and unattached CpZZrCl2 . . . . . . . . . . . . . . . 22 Far infrared bands (cm'1) of polymer-attached and unattached szHfClz . . . . . . . . . . . . . . . 23 Catalytic epoxidation of cyclohexene. . . . . . . . . 34 Synthesis of aldehydes from hydrozirconation of alkenes with polymer-attached CpZZrClz. . . . . . . . . . . . 38 van Tamelin nitrogen fixation (flow system) . . . . . 45 Vol'Pin-Shur nitrogen fixations (at lSDO psi of N2) . 46 Hydrogenation rates for polymer-attached epRh(c0)2 beads at 25° and l atm. . . . . . . . . . . 64 Hydrogenation of ketones and aldehydes with polymer-attached CpRh(CO)2. . . . . . . . . . . . . . 65 The isomerization of allylbenzene l,5-cyclooctadiene. and cis-stilbene by polymer-attached CpRh(CO)2. . . . 67 The disproportionation of l.4-cyclohexadiene and cyclohexene by polymer-attached CpRh(CO)2 . . . . . . 68 Cyclotrimerization of ethyl propiolate with polymer-attached CpCo(CO)2 and CpRh(CO)2 . . . . . . 7O Hydroformylations of l-pentene and l-hexene catalyzed by polymer-attached CpRh(CO)2. . . . . . . . . . . . . 72 Minor products from thermal decomposition of n-alkyl transition metals . . . . . . . . . . . . . . . . . . lll iv Table Page 14 Metal-carbenes trapped by cyclohexene. . . . . . . . ll2 l5 Decomposition products of neohexyl metal complexes other than a-hydrogen elimination . . . . . . . . . l2l Figure 10 LIST OF FIGURES Page Functionalization of polystyrene with ligands. . . . 7 Scheme for the attachment of szTiClz to a polymer. . . . . . . . . . . . . . . . . . . . . . . 9 Scheme for the attachment of cyclopentadienyl anion to the polystyrene copolymer . . . . . . . . . l3 Scheme for the preparation of supported CpZZrCl2 by the reaction of CerCl3 with lithium cyclopenta- dienide-substituted polymer . . . . . . . . . . . . 14 Scheme for the attachment of szMCl2 to polystyrene through the exchange reaction of Uandvcyclopenta- dienyl ligands . . . . . . . . . . . . . . . . . . . l5 Scheme for the attachment of szMCl2 to polystyrene by the modified method . . . . . . . . . . . . . . . 16 Scheme for the attachment of CpMCl3 to the polystyrene copolymer . . . . . . . . . . . . . . . . . . . . . l8 Scheme for the attachment of Cp3MCl to the polystyrene copolymer 19 Scheme for the preparation of polymer-attached methylene-bridged CpZZrCl2 . . . . . . . . . . . . . 2l Far infrared spectra of polymer-attached and unattached CpZZrCl . . . . . . . . . . . . . . . . . 24 vi Figure Page ll Far infrared spectra of polymer-attached and unattached szHfClz. . . . . . . . . . . . . . . . . 25 12 Hydrogenation of diphenylacetylene catalyzed by supported CpZZrClz . . . . . . . . . . . . . . . . . 27 l3 Hydrogenation of diphenylacetylene catalyzed by supported szHfClz . . . . . . . . . . . . . . . . . 27 I4 Isomerizaiton of allylbenzene catalyzed by supported CpZZrCl2 . . . . . . . . . . . . . . . . . 29 15 Isomerization of allylbenzene catalyzed by supported CpZHfCl2 . . . . . . . . . . . . . . . . . 29 16 Isomerization of allylbenzene catalyzed by supported CerCl3 . . . . . . . . . . . . . . . . . 30 l7 Isomerization of allylbenzene catalyzed by supported Cp3ZrCl . . . . . . . . . . . . . . . . . 3O l8 Isomerization of l,5-cyclooctadiene catalyzed by supported CpZZrClz . . . . . . . . . . . . . . . . . 3l l9 Isomerization of l,5-cyclooctadiene catalyzed by supported CpZHfCl2 . . . . . . . . . . . . . . . . . 3l 20 Isomerization of l,5-cyclooctadiene catalyzed by supported CerCl3 . . . . . . . . . . ... . . . . . 32 2l Isomerization of l,5-cyclooctadiene catalyzed by supported CprCl3 . . . . . . . . . . . . . . . . . 32 22 Scheme for the synthesis of aldehydes from hydro- zirconation of alkenes . . . . . . . . . . . . . . . 3S vii Figure 23 24 25 26 27 28 29 3O 31 32 33 34 The ESR spectrum of the reduced species of polymer-attached CpZZrCl2 beads under argon. . . . . The ESR spectrum of the reduced species of polymer-attached CpZZrCl2 beads under hydrogen . . . Scheme for the preparation of polymer-attached methylene-bridged szTiClz via chloromethylation . . Scheme for the preparation of polymer-attached methylene-bridged szTiCl2 via lithiation . . . . . Scheme for the preparation of polymer-attached CpZTi(CO)2 . . . . . . . . . . . . . . . . . . . . IR spectrum of polymer-attached szTi(CO)2 . . . . . Scheme for the preparation of polymer-attached CpMCl4 (M=Nb, Ta) . . . . . . . . . . . . . . . . . Scheme for the preparation of polymer-attached (NI-PhCH2)MCl4 (M-Nb and Ta) . . . . . . . . . . . . Scheme for the dimerization of ethylene catalyzed by polymer-attached CpTaCl4 . . . . . . . . . . . . . . Isomerization of allylbenzene catalyzed by reduced species of polymer-attached CpMCl4 (MaNb and Ta) . . Scheme for the preparation of polymer-attached CpCo(CO)2 . . . . . . . . . . . . . . . . . . . . . Scheme for the preparation of polymer-attached CpRh(C0)2 . . . . . . . . . . . . . . . ... . . . . viii Page 41 42 47 48 50 51 53 54 56 58 60 61 Figure 35 36 37 38 39 4O 41 42 43 44 45 46 47 Page CO stretching spectra of polymer-attached CpCo(CO)2. . . . . . . . . . ... . . . . . . . . 62 CO stretching spectra of polymer-attached CpRh(C0)2. . . . . . . . . . . . . . . . . . . . 62 Scheme for the mechanism of hydroformylation reaction catalyzed by polymer-attached CpRh(CO)Z. . . . . 74 Decomposition of di-n-butylbis(triphenylphosphine)- platinum(II) . . . . . . . . . . . . . . . . . . lOl Formation of the tantalum neopentylidene complex. . l02 Decomposition of l,4-tetramethylenebis(cyclo- pentadienylltitanium(IV) . . . . . . . . . . . . l04 Conversion of olefins to cyclopentanones by reaction with titanocene equivalents . . . . . . lO4 Decomposition of l.4-tetramethylenebis(tri-n- butyl-phosphine)platinum(II) . . . . . . . . . . l06 Decomposition of nickelacyclopentanes. . . . . . lO7 Possible scheme for the formation of ethylene via 6-hydrogen elimination in the decomposition of szTiBuz . . . . . . . . . . . . . . . . . . . . ll3 Scheme for'y-elimination in organoaluminum and silicon compounds. . . . . . . . . . . . . . . . ll4 Decomposition of di-neopentylbis(triphenylphosphine)- nickel(II) . . . . . . . . . . . . . . . . . . . llS The scheme for the proposed new mechanisms of thermal decomposition of transition metal alkyls . . . . 120 ix Figure Page 48 Infrared spectrum of trichlorobis(cyclopentadienyl )- tafltalum (cp2T3c13 o o o o o o o o o o o o o o o o 129 PART I PREPARATION AND APPLICATION OF POLYMER-SUPPORTED CATALYSTS INTRODUCTION In an energy conscious world, the chemists have long been inter- ested in ways of minimizing the energy requirements of chemical reac- tions and to this end have diligently sought out suitable catalysts. The use of both soluble and insoluble catalysts in chemical reactions is almost as old as chemistry itself. Acid catalysis is a typical example of homogeneous catalysis which has been known and applied for a long time. On the other hand, the hydrogenation of unsaturated compounds catalyzed by transition metals such as platinum, palladium, or Raney nickel, a representative example of heterogeneous catalysis, was not developed until the early nineties. In the last thirty years, the industrial application of processes catalyzed by soluble transition metal compounds has become very signi- ficant. Despite this impressive growth. homogeneous catalysts are less widely used than heterogeneous catalysts in the chemical industry and have little application in petroleum refining due to the difficult separation of the very expensive catalysts from the products at the end of the reaction. Nevertheless, homogeneous catalysts demonstrate great selectivity and economic efficiency. Recently. considerable research effort has been expended in developing catalysts that combine as many advantages as possible. There have been two major approaches. l. Fluid-bed catalysts:l The use of fluid-bed catalysts, in which control of the catalyst movement is maintained by the flow of the reactants, represents an attempt to homogenize heterogeneous catalysts. This technique is widely used for the catalytic cracking of petroleum and much other technology. 2. Supported complex catalysts. These catalysts essentially invo- lve a metal complex covalently bound to some form of support. They have also been referred to as hybrid phase catalysts.2 3 heterogenized 4 or solid- homogeneous catalysts, homoactive heterogeneous catalysts phase synthesis.5 This technique represents an attempt to heterogenize homogeneous catalysts. The supported homogeneous catalysts are most often insoluble in any solvent, and in this sense. they are indeed quite heterogeneous. However. they are prepared by the same reactions as those used for the homogeneous complexes. The transition metals retain the ligands around them even during the catalytic processes. Moreover, these catalysts function in operating conditions comparable to those used for conven- tional homogeneous catalysts. for instance at relatively low tempera- tures. Therefore, these supported homogeneous complex catalysts have several advantages such as overcoming the problem of catalyst separation from the reaction products, retaining the advantages of homogeneous catalysts, having higher selectivity than the conventional heterogeneous catalysts and reducing the chance of poisoning. Heterogeneous supported catalysts have been developed in order to get a better utilization of the potential catalytic activity by dispersing the metal on an inorganic oxide carrier. Similar effects have also been observed when anchoring a metal complex on a polymer, the supported catalyst being more active than the homogeneous complex from which it is derived. It is generally agreed that homogeneous catalysis practically always involves at least one intermediate in which the metal atom is coordinatively unsaturated. Such an unsatura- ted intermediate very often tends to dimerize, which is detrimental to the catalytic activity. It is then conceivable that anchoring the catalyst on a sufficiently rigid support can prevent the association of the isolated and unsaturated catalytic centers, thus preserving all of the potential catalytic activity. Haag and Hhitehurst6 reported the carbonylation of allyl chloride catalyzed by either soluble or supported palladium amine complexes. It has been shown the catalytic activity of the supported catalyst increases linearly with the amount of palladium used. 0n the other hand, the activity of the homogeneous catalyst soon reaches a maximum when its concentration is raised. This corresponds to a parallel increase of the mutual interactions between palladium atoms, leading to aggregation and formation of catalytically inactive oligomeric species. Such an aggregation is not found when the palladium complexes are firmly anchored on a rigid polymer. Another example of activation of a homogeneous complex by anchor- ing on a support, reported by Grubbs, Brubaker and coworkersZ'Il is the hydrogenation of olefins catalyzed by titanocene-supported analogues. The attachment of titanocene species to a rigid polymer resulted in an increase in activity for olefin hydrogenation. Since titanocene species undergo deactivation by polymerization processes, it was proposed that the increase in activity resulted from site isolation on the polymer. The catalyst activity varies with the loading of the catalyst on the polymer. The relationship between loading and rate is best explained by assuming the material is attached to an almost rigid, immobile surface. Besides the increasing activity, one can expect also some peculiar selectivity effects from the supported homogeneous catalysts. The use of a polymer raises the question of the diffusion of the substrate from the bulk solution to the catalytic centers, through the pore channels of the polymer. Diffusion barriers are associated with substrate size, polymer crosslinking, swelling power of the solvent used, and relative substrate and polymer polarities. Rhodium(I) hydrogenation catalysts attached to 2% crosslinking polystyrene show decreased reduction rates as olefin size increases or as solvent swelling power decreases.]2’ ‘3 These effects are mostly associated with diffusion limitations and become more important as catalyst activity increases.14 All the aims which have been discussed up to now are more or less directed towards the applied aspects of the supported homogeneous catalysis. In fact, the supported homogeneous catalysis may be used as a tool for a new approach of the heterogeneous catalysis. Supported catalysis is particularly pertinent to the enviroment of the catalytic sites, and of the interactions between these sites. One can hape in this way, to analyze the factors which influence the activities and the selectivities of heterogeneous catalysts. The choice of a supporting solid depends on the particular catalyst system. Generally speaking, the first property is that the polymer must bear functional groups which can be used as ligands by the metal, Other parameter, from a chemical standpoint, is inertness to reagents. Concerning the physical properties of the polymer, the desirable characteristics of a support include: good mechanical and thermal stability, porosity, surface area and heat-transfer properties. A number of materials have been used as the basic supports. It is convenient to describe them as either organic or inorganic supports, although there is a considerable region of overlap. Among these materials, polystyrene and silica have received most attention as organic and inorganic supports,and the techniques used to make these suitable as supports are generally typical of those required for organic and inorganic supports,respectively. Crosslinked polystyrene is available with a wide range of crosslink densities, surface areas and porosities. The basic polymer backbone is chemically inert15 and the polar properties can be modified by controlled functionaliza- tion or by the preparation of appropriate copolymers. These supports, however, have poor mechanical and thermal stabilities and poor heat- transfer properties. In this thesis research, 2% and 20% crosslinked polystyrene-divinylbenzene copolymer beads have been used as supports. A homogeneous catalyst can be heterogenized in a variety of ways. The common methods are to attach the metal complex to a solid support 15 l7, l8 or either by deposition or by adsorption, or by an ionic a covalent chemical bond. Less common methods involve the polymerization of a complex, bringing a polymerizable function to such a high mole- cular weight that it becomes insoluble in the medium in which it will be used or by trapping it in a gel or other porous medium.]9"23 If the complex is electrically neutral, a covalent or a coordina- tive bond must be used to link it to a support. Thus, polymers have to be functionalized with linking ligands used in homogeneous catalysis such as arene, cyclopentadiene, phosphine, pyridine, cyano, amine, etc.. Numerous techniques have been developed for functionalization of poly- styrene with suitable ligands as shown in Figure 1.13 These modes allow linking agents to be attached through either nucleophilic or electrophilic reactions and provide great flexibility in the synthetic design. Any transition metal complex can be attached to the polymer by appropriate choice of linking ligands. According to the linking ligands, the supported homogeneous catalysts can be divided into two groups: (1) complexes that require mobile, labile supporting ligands, and (2) complexes that exhibit a high degree of unsaturation and are aided by immobile supporting ligands. Good examples of the first class are the group VIII phos- phine catalysts,and the early transition element metallocenes provide examples of the second classification. The attachment of transition metal complexes to solid support by phosphine ligands has been studied extensively. The phosphines in these complexes exchange readily and in general the complexes are easily lost from the supports during the catalytic reactions. Also, the phosphines have high reactivity toward electrophiles,oxygen,and oxidizing agents. The polymer-attached metal complexes through relatively inert cyclopentadienyl ligands have only recently been studied. Grubbs, 7-11.24-27 Brubaker, and coworkers reported that titanocene dichloride, c m a * P v w c m & a m >1 P F o a a N F P c O P U c 3 k F m g a m P o u H Mg» ! IZNm/ _o NICO Orzo ... ICWZ w . o szTiClz, could be attached to the polystyrene resin by reacting first the cyclopentadienyl groups with methyllithium and then with cyclo- pentadienyl trichlorotitanium. Upon reduction of the polymeric, “matrix isolated” titanocene dichloride with butyllithium or sodium naphthalide, a gray polymer was obtained that readily catalyzed hydro- genation of olefins and acetylenes (Figure 2). The reaction of the sodium cyclopentadienide-substituted copoly- mer with TiCl4 led to a supported CpTiCl3 species. Similarly, the polymer-attached CpMCln species (MsZr, Hf, Mo, H, and Nb) can be synthesized by the analogous reactions of polymer-attached cyclo- pentadienyl anions with anhydrous transition element halides. The reduced supported metallocene halides exhibit a high activity for the hydrogenation and isomerization of unsaturated compounds.26"28 A number of polymers containing the ferrocene group have been prepared in recent years due to the thermal stability, moisture stabi- lity and air stability of this group. The products obtained during the thermal treatment of ferrocene polymers exhibited catalytic activity in the oxidation and dehydration of alcohols.29 l 30'3] reported the preparation of mononuclear Brintzinger et a cyclopentadienyl cobalt, chromium, iron, molybdenum, rhodium and tungsten carbonyl derivatives covalently linked to macroreticular polystyrene and silica gel supports. The cobalt and rhodium deriva- tives were found to be active catalysts for olefin hydrogenation and hydroformylation reactions. (I) CH3L: momma + Coordinatively III I) Unsaturated Figure 2. Scheme for the attachment of szTiCl2 to a polymer. TO The research work reported here consists of (l) the preparation and characterization of polymer-attached zirconocene and hafnocene species, and the testing of their catalytic activities towards hydro- genation of unsaturated compounds, isomerization of allylbenzene and l,5-cyclooctadiene, epoxidation of olefins, and hydrozirconation of olefins to produce terminal aldehydes, (2) the preparation of polymer- attached methylene-bridged titanocene compounds and their activity toward nitrogen fixation, (3) the preparation and characterization of polymer-attached niobium and tantalum complexes and examination of their catalytic activities, and (4) the synthesis and characterization of polymer-attached cyclOpentadienyl cobalt and rhodium dicarbonyls, and study of their applications to the catalytic hydrogenation of olefins, aldehydes and ketones; hydroformylation of olefins; isomeriza- tion of allylbenzene and l,5-cyclooctadiene; disproportionation of cyclohexene and cyclohexadiene; and cyclotrimerization of acetylene derivatives. RESULTS AND DISCUSSIONS Homogeneous catalysts have several advantages over heterogeneous catalysts in several aspects such as efficiency, reproducibility, specificity, controllability and selectivity. However, the major disadvantage of the homogeneous catalysts is the need to seperate the reaction products and to recover the catalyst, which is often somewhat more expensive. So it's less useful in industrial applications than a "classical" heterogeneous catalyst. Hybrid catalysts prepared by reaction of organometallic compounds with polymer or metal oxide supports, may combine the advantages of both homogeneous and hetero- geneous catalysts. These developments should open new industrial applications. It has been demonstrated that the attachment of titanocene- related catalysts to polymers resulted in an increase in activity for the olefin hydrogenation.9’]0'28 Since titanocene species undergo deactivation by dimerization or polymerization process, it was proposed that this increase in activity resulted from site isolation on the ll polymer. Polymer-attached titanocene has also been proved to be excellent catalysts for the isomerization and epoxidation of olefins and oligomerization of acetylene derivatives.26’32 For the analogous study, zirconocene and hafnocene derivatives, which are isoelectronic and most probably isostructural with the titanium homologs, have been attached to the polystyrene copolymers and their catalytic activities are examined in this thesis research. ll 12 Crosslinked polystyrene, available with a wide range of crosslink densities, surface areas, and porosities. has received the most atten- tion as an organic support. Collman and his coworkers33 have found that 2% crosslinked polystyrene-divinylbenzene copolymers are mobile enough to allow ligands attached to the polymer beads to act as chelates. Consequently, this polymer is not rigid enough to prevent dimerization of attached unstable species. For the research discussed here, 20% crosslinked macroreticular polystyrene-divinyl benzene copo- lymer beads, ranging in size from 30 to 35 mesh (600 A average pore size)34 were used as the supports. The 20% crosslinked copolymers are rigid, insoluble and have a large surface area. The cyclopentadienyl anion has been used as good ligand for a 35-37 It has also variety of potentially useful metal complexes. provided a remarkably effective means of binding transition metal complexes to the polystyrene-divinylbenzene copolymer and its penta- hepto bonding to the metal ensures the formation of a strong, covalent w-bond between the polymer and the metal. The attachment of cyclopentadienyl groups to polystyrene-divinyl benzene copolymer can be achieved by the reaction of sodium cyclo- pentadienide with the chloromethylated capolymer beads, which obtained by following the chloromethylation method of Pepper et al.38 The cyclopentadiene is coverted into cyclopentadienide anion by the treatment with methyllithium or butyllithium. The procedures are outlines in Figure 3. A variety of transition metal complexes can be attached to the polymer through the cyclopentadienyl ligands. l3 ' CICH -O-C2H I SNZI4 s T’ H zCl Polystyrene- Divinylbenzene Copolymer N§® CH3LI ‘3? BuLi Z Z Li+ Figure 3. Scheme for the attachment of cyclOpentadienyl anion to the polystyrene copolymer. Preparation of Polymer-Attached Bis(cyclopentadienyl)zirconium and Hafnium Dichlorides ( P-CpZMClZ; M=Zr, Hf) Three methods for the preparation of polymer-attached zirconocene and hafnocene dichlorides have been studied. They are (a) reaction of CpMCl3 with polymer-attached lithium cyclopentadienide,39 (b) reaction of szMCl2 with polymer-attached lithium cyclopentadienide followed by treatment with excess hydrogen chloride gas, 40’4] and (c) l4 reaction of MCI4 with polymer-attached lithium cyclopentadienide followed by the addition of stoichiometric amout of sodium cyclo- pentadienide, and then by the treatment with small amount of hydrogen chloride gas. The reaction schemes are shown in Figure 4,5 and 6. METHOD a : “49' NF §£::7' .1EE!2...4L “AQKKQSFASEZIE +12 Xme Figure 4. Scheme for the preparation of supported CpZZrCl2 by the reaction of CerCl3 with lithium cyclopentadienide- substituted polymer. 15 METHOD b CPzMFC'z 4 § dd 17:0 HCI(9)(Dcess) § fil’HF GE: /C' ~CWGE /CI M @M\CI @fb (IJI==ZEr;|'Nf) Figure 5. Scheme for the attachment of szMCl2 to polystyrene through the exchange reaction of o and n cyclopenta- dienyl ligands. HCKQ) a H2 (SMALL AMT.) M/C| - Figure 6. Scheme for the attachment of CpZNCl2 to polystyrene by the modified method. 17 Due to the difficult preparation of CpMCl3, the first procedure is less useful. For the second procedure, the remaining unreacted cyclopentadienyl and vinyl groups on the polymer take up hydrogen chloride when excess hydrogen chloride is used. Therefore, the higher ratio of chloride to metal from elemental analyses than the expected value is observed. After the exchange reaction of nI-CSH5 and n.5-C5H5 rings in Cp3MCl followed by the treatment with hydrogen chlo- ride gas, the loading of szMClz on the polymer is only two-thirds of the original concentration of Cp3MCl. In order to overcome these problems, the third procedure has been developed to produce higher loading of szHCl2 on the polymer. The addition of stoichiometric amount of sodium cyclopentadienide to polymer-attached CpMCl3 mostly gives szMClz. Although small amount of Cp3HCl and Cp4H species can be conceivably produced, they can be easily converted into szMClz by the treatment with small amount of hydrogen chloride gas. Therefore, the loading of szMCl2 on the polymer is almost the same as the original concentration of CpMCl3 and elemental analyses show the ratio of chloride to metal is very close to the expected value. Preparation of Polymer-Attached Monocyclopentadienyl Zirconium and Hafnium Trichlorides ( P-CpMClB; M-Zr, Hf) Brubaker and Chandrasekaran27’28 synthesized polymer-attached monocyclopentadienyl zirconium and hafnium trichlorides from the reaction of the lithium cyclapentadienide-substituted capolymer with zirconium tetrachloride and hafnium tetrachloride pyridine adducts 18 respectively, due to the low solubility of the pure tetrachlorides. In the study of supported monocyclopentadienyl zirconium and hafnium trichlorides, it was found that these supported complexes could be prepared by the direct reaction of the lithium cyclopentadienide- substituted polymer and pure tetrachlorides for longer reaction time. 42 reported that the reaction between dicyclopentadienyl- Renaut et al magnesium and hafnium tetrachloride in decalin gave CprCl3.2THF adduct after removal of szHfClz with THF. Therefore, the species on the polymer after washing with THF is probably CpHCl3.2THF adduct. MCI4 Benzene or THE' 0 /Cl M CM=Zn Hf) Cl/ \CI Figure 7. Scheme for the attachment of CpMCl3 to the polystyrene copolymer. 19 Preparation of Polymer-Attached Tris(cyclopentadienyl)zirconium and Hafnium Monochlorides ( P6Cp3MCl; M-Zr, Hf) Recently the reactivities of Cp22r(R)c152'54 with a variety of reagents have been studied extensively. In order to compare the reactivity of supported Cp3HCl with those of the supported CpMCl3 and szMCl2 species, the supported Cp3HCl (M=Zr, Hf) species was prepared. The attachment of Cp3MCl to the polymer has been achieved by the reaction of szHClz with polymer-attached lithium cyclopenta- dienide. The reaction scheme is shown below (Figure 8). Figure 8. Scheme for the attachment of Cp3MCl to the polystyrene copolymer. 20 Preparation of Polymer-Attached Methylene-Bridged Bis(cyclgpenta- dienyl)zirconium Dichloride The attachment of CpZZrCl2 to a polymer has been described above. The higher loading of CpZZrCl2 on the polymer can be achieved by using methylene-bridged cyclopentadiene-substituted polymer which contains higher concentration of cyclopentadienyl groups. The con- centration of metal complexes attached by this way usually can be increased to twice that of non-bridged cyclOpentadiene-substituted polymer. The reaction scheme is shown in Figure 9. Far Infrared Studies An understanding of the factors which control the change in catalytic activity on polymer attachment has been difficult to determine due to the lack of good methods for the analysis of the structure of the catalyst on the polymer. Several techniques have been found useful for this analysis in our laboratory. Elemental analysis gives the loading of the catalyst and the ratio of metal to 27.28 chloride. Electron microprobe analysis allows the rapid deter- mination of the distribution of the catalyst inside of a polymer bead. 31P NMR spectroscopy has been used for the analysis of low 43,44 Recently the cross-linked phosphinated polystyrenes. The IR analysis, although successful in some cases, is normally hindered by high background absorption of the polymer and the low loadings used. 21 (no—1202 moo/TH? 6 m (mag 1w..- (DCPZZI'CIZ ‘EHCI egego 69+ LI+ 0/ \Cl Cl Figure 9. Scheme for the preparation of polymer-attached methylene-bridged CpZZrClz. 22 The study of far infrared spectra of metallocene chlorides allows assignments to be made for the metal-ring and metal-ligand stretching modes. Since the far infrared spectra for the titanocene systems show a good correlation between the polymer-attached and unattached 27’28 a similar investigation of the zirconocene and hafnocene species, systems was undertaken as a further means of identifying the species on the polymer. It is indicated that there is also a good correlation between the attached and unattached zirconocene and hafnocene chlo- rides (Figure l0 and ll). These absorption bands are in substantial agreement with those reported in the literature.45’46 The band assignments are deduced from these reported data and are listed in Table l and 2. Table 1. Far infrared bands (cm']) of polymer-attached and unattached CpZZrClz. Assignments CpZZrClz(lit.)45'46 szzmz P—szzmz Obs. Calc. . vs(Zr-Cp) 358 350 36l 355 va(Zr-Cp) 358 368 36l 355 vs(Zr-Cl) 333 329 332 330 va(Zr-Cl) 333 332 332 330 tilt 3l0 ... 310 306 tilt 266 ... 265 269 23 Table 2. Far infrared bands (cm'1) of polymer-attached and unattached szHfCl 2. Assignments CpZch12(11t.)45'46 szHfClz P-Cp2HfCl2 Obs. Calc. vs(Hf-Cp) 360 349 350 360(sh) va(Hf-Cp) 350 338 350 335 vs(Hf-Cl) 310 314 312 308 va(Hf-Cl) 310 310 312 308 tilt 254 ... 265 265 tilt 284 ... 285 285(sh) Hydrogenation Studies Titanocene and its hydrides are useful catalysts for the hydroge- nation of unsaturated compounds. However, these complexes readily dimerize or polymerize to form catalytically inactive materials.47’ 48 To prevent the dimerization, it has been achieved by attaching the titanocene precursor to a rigid polymer support. 0n reduction, a catalyst is produced whose hydrogenation efficiency is greater than a corresponding nonattached species.9’ ‘0 Zirconocene and hafnocene are also found to be active catalysts for the hydrogenation of alkenes. Their activities in hydrogenation have been increased after attachment to the polymer. For example, the rates of hydrogen uptake for l-hexene catalyzed by reduced nonat- tached and attached CpZZrCl2 are 0.l2 mL and 0.95 mL/m mmol of metal 24 PI .NPQLNNQU vogumupacs can vogumuumueosxpog we ecaumam vogmcmcv Lem 8m 83 8” SN . q . 1e 2 . 4 . mmn on» op" \r.me~ \ ) I) I _ /\..I \ ffu / ~ I , c . > . . . Z _ : . _ . . x / _ _ L . I _ . u . 2. _ _ . . . : r 7 N N .. ---- ....7. ,1; ‘ 8:815 a J c a . Z (r \ NpucNNaula II. _ \c a _. ./ \ ex , . C e . .o. beamed E A 25 .N —uEu com ooe can com d u q u N 2m) 3m \ /mm~ mew I J) (’ ~8§~8 + 81.. ---- ... magma; ....I r .. 1111.. pumzuqu voguouuucz use vogucuueueosapoa Go ucaoonm vmcecmcv and ..p me:m.u 3v 26 at room temperature respectively. Previous hydrogenations reported with homogeneous cyclopentadienyl zirconium catalysts were carried out only at high temperature and high pressure.49 Supported zirconocene and hafnocene complexes are also active hydrogenation catalysts for alkynes. The rates of hydrogen uptake for diphenylacetylene catalyzed by supported zirconocene and hafnocene catalysts are 3.6 mL and 2.7 mL/m mmol of metal at l00°, respectively. It is shown that diphenyl- acetylene is hydrogenated to give trans-stilbene and further to give l,2-diphenylethane (Figure 12 and I3). Some catalytic hydrogenation of unsaturated compounds in the presence of supported titanocene catalysts is also examined and indicates a reduced activity for the heavier metals. Isomerization Allylbenzene and l,5-cyclooctadiene, when treated with reduction product of polymer-attached zirconocene and hafnocene chlorides, appear to undergo a rapid double bond migration. The exact mechanism by which they act, whether of the addition-elimination or n-allyl type, is still not certain. Allylbenzene can be isomerized to form the more stable conjugated system, cis- and trans-propenylbenzene (Figure l4-l7). Similarly, l,5-cyclooctadiene is effectively isomerized to l,3-cyclo- octadiene with l,4-cyclooctadiene as the intermediate product (Figure l8-2l). The catalysts produced by reducing the supported metallocene chlorides under hydrogen are more active than those generated under argon. For example, the catalyst produced, by reducing the supported CpZZrCl2 under argon, can only effect about 50% isomerization of 27 100 Substrate : Zr = 1.5 : 0.03 80 - 60 - A , X Diphenylacetylene A Trans-stilbene “ o l ,2-Diphenyl ethane Hydrogenation (%) 4o 60 ' 80 'fig'oo Time (h) Figure l2. Hydrogenation of diphenylacetylene catalyzed by supported CpZZrClZ. Substrate : Hf 8 l.5 : 0.09 100 80 33 60 X Diphenylacetylene § 4 Trans-stilbene E 40 o l,2-Diphenylethane 0 'U 3:1 2o 40 80 100 Time (h) Figure 13. Hydrogenation of diphenylacetylene catalyzed by supported szHfClz. 28 allylbenzene in 5 h, while the catalyst obtained under hydrogen brings the isomerization to an extent of 90% under the same conditions. The reduction of metallocene chlorides under hydrogen probably generates metallocene hydrides such as CpZZrHZ and szHfHZ and the isomerization most probably proceeds through the addition-elimination mechanism. A comparison of isomerization activity is examined, it is shown that CpMC13 is the most efficient catalyst, followed by CpZMClz, with Cp3MCl being least active under the conditions employed (Figure l4, l6 and 17; Figure 18 and 20; Figure 19 and 21). A reduced activity for the heavier metal is also indicated (Figure 14 and 15; Figure 18 and 19; Figure 20 and 21). The catalyst can be repeatly used without much reduced activity (Figure 14). Other experiments indicate that polymer-attached zirconocene and hafnocene catalysts are also good catalysts for cis-trans isomerization. Cis-stilbene, when treated with these catalysts, is isomerized rapidly to trans-stilbene. But no appreciable amount of isomerization products was observed for l,5-hexadiyne, B-pinene and l-octene. The isomerization of allylbenzene and l,5-cyclooctadiene catalyzed by the reduction product of the supported titanocene chlorides has been studied by C. P. Lau.26' 32 To compare its activity with those of zirconocene and hafnocene catalysts, it is found that their activities are in the decreasing order, Ti>Zr>Hf. 29 Substrate : Zr = 15.1 : 0.09 X Al lyl benzene 6 C i s-propenyl benzene o Trans-propenyl benzene ---- A second reaction on the same catalyst. Isomerization (%) 7.21%th- IXLI . fiufi 13E 20 25 30 Time (h) Figure 14. Isomerization of allylbenzene catalyzed by supported CpZZrClZ. ‘00 Substrate : Hf = 15.1 : 0.09 SE 80 r c 5 . 1g, 50 - x Allylbenzene ”E 8 A Cis-propenylbenzene g 40_ @ Trans-propenylbenzene In 20L 4S‘\\\‘~‘. Time (h) Figure 15. Isomerization of allylbenzene catalyzed by supported CpZHfClZ. Isomerization (%) Isomerization (%) 3O 100 Substrate : Zr = 15.1 : 0.09 -—e r I‘D— 80 - 60 X Al 1 yl benzene A Ci s-propenyl benzene 40 . o Trans-propenyl benzene 20 a l I + + 10 15 20 25 30 Time (h) Figure 16. Isomerization of allylbenzene catalyzed by supported CerC13. Substrate : Zr 8 15.1 : 0.09 100 80 60 )( Allylbenzene A Ci s-propenyl benzene ‘0 O Trans-propenyl benzene 20 XF—~ 2 a h :1".__:.g_ 5 10 15 20 25 30 Time (h) Figure 17. Isomerization of allylbenzene catalyzed by supported Cp3ZrCl. 31 100 Substrate : Zr = 15.1 : 0.09 x l ,5-Cyclooctadiene A l ,4-Cycl ooctadi ene O 1,3-Cyclooctadiene Isomerization (%) 20 25 30 Time (h) Figure 18. Isomerization of l.5-cyclooctadiene catalyzed by supported CpZZrClz. 100 \\\\\\‘ Substrate : Hf = 15.1 : 0.09 :3 80 L Ill/’1”’—-—-.cr————. EE 60 . ' ° ‘5 X l.5-Cyclooctadiene ‘5 (’1 A l,4-Cyclooctadiene § 40 ' G 1,3-Cyclooctadiene 20. \\ l l 1 L j 5 10 15 20 25 30 Time (h) Figure 19. Isomerization of l,5-cyclooctadiene catalyzed by supported szHfClz. 100 80 E C :2 60 +3 3 '5 40 E 20 Figure 20 100 ,~ 80 23 5 :3 60 m N 'E g 40 :2 20 Figure 21 32 Substrate : Zr 3 15.1 : 0.09 // X l ,5-Cyc1 ooctadiene A l ,4-Cyclooctadiene O l ,3-Cyclooctadiene ‘T‘a-th“—--“““*‘---.. 16 15 20 25 30 Time (h) . Isomerization of l.5-cyclooctadiene catalyzed by supported CerC13. Substrate : Hf = 15.1 : 0.09 x 1 ,S-Cyclooctadiene A l ,4-Cyclooctadieoe O 1 ,3-Cyclooctadiene X\x 3"4F’ “‘Ir—~ ___‘:::::::::r. j j j 5 1O 15 20 25 30 Time (h) . Isomerization of l,5-cyclooctadiene catalyzed by supported CprCl3. 33 Epoxidation It is knownso’ 5] that compounds of certain transition metals. notably Ho, H, Ti and V catalyze the liquid-phase epoxidation of olefins with alkyl hydroperoxides. ROOH + >=< “7' —» >2$< + ROH 50’ 5] have been Both homogeneous and heterogeneous catalysts described. Brubaker and Lau26’ 32 have demonstrated that polymer- attached titanocene dichloride and trichloride can be used as catalysts for the epoxidation of cyclohexene and cyclooctene with t-butyl hydro- peroxide. Similarly, the epoxidation of cyclohexene with t-butyl hydroperoxide in the presence of polymer-attached zirconocene and hafnocene catalysts has been studied. Polymer-attached zirconocene and hafnocene chlorides are active catalysts for the epoxidation, but their activities are relatively lower than that of the supported titanocene catalysts. The results are given in Table 3. The yields of major product, epoxycyclohexane, are ranged from 9% to 22%. Some other by-products,5] possibly 3-t-butylperoxy-l-cyclohexene, 2-cyclohexen-1-ol, 2-cyclohexen-l-one, trans-cyclohexane-l,2-diol, 2-hydroxycyclohexanone and 2,3-epoxy-cyclohexan-l-ol, were not examined. The continuous decrease in the activity of the catalysts, possibly due to the autoretardation by the co-product t-butanol, has been obse- rved. However, the catalysts can be regenerated by passing anhydrous 34 hydrogen chloride gas into the beads suspended in THF. The regenerated catalysts are reused for the epoxidation without much loss of activity. The detailed mechanism of epoxidation catalyzed by these supported catalysts was not determined. Table 3. Catalytic epoxidation of cyclohexene. Catalysts(wt.) Solvent Time(h) Product(%) (loading) (epoxycyclohexane) P -CpZZrC12(0.l444g) benzene 18 16 (0.63 mol Zr/g) P -CpZZrC12(O.14489) cyclohexane 18 19 (0.63 mmol Zr/g) P -Cp22r012(0.l4489) cyclohexane 18 15 (recycled)a P -CpZZrC12(O.l46Zg) benzene 48 20 (0.63 mmol Zr/g) P -CerC13 (0.27569) cyclohexane 18 22 (0.33 mmol Zr/g) P -Cp2HfC12(0.2007g) cyclohexane 48_ 9 (0.45 mmol Hf/O) P -CprCl3 (0.30759) cyclohexane 48 11 (0.29 mmol Hf/g) a Catalyst used was recycled from previous reaction. 35 Hydrozirconation of Alkenes Schwartz _e_t__a_l_52'54 have demonstrated that CpZZrC1H will react with olefins and alkynes to produce terminal alkyl and alkenyl zirconium derivatives. These zirconocene alkyls and alkenyls will react with a number of reagents to produce terminally functionalized alkanes and alkenes in high yield. The most interesting reaction is the production of terminal aldehydes from olefins (Figure 22). The development of hydrozirconation illustrates that the introduction of new procedures will continue to be an exciting aspect of organic synthesis because of the broad scope and high reactivity of organometallic complexes. CPEZ /Cl VITRIDE a C9 /C1 3m 0.: ‘c (:1 (NaAle(OR)2)CP 2' H 1 Ce Cl CP/Zr HCI C" zr/Cl co CP/ zr\g/\/\/\ HQW O 0 Figure 22. Scheme for the synthesis of aldehydes from hydrozirconation £13: of alkenes. 36 If one can attach CpZZrClz to a polymer support, zirconocene dichloride can be recovered by filtration and re-used for hydrozirco- nation after hydrolysis of an acyl zirconium complex with dilute aqueous hydrochloric acid solution. Attempt to use the polymer-attached CpZZrClz reagents for the hydrozirconation of olefins to produce terminal aldehydes did not meet with much success. The results are given in Table 4. The zirconium hydride, Cp22r(H)Cl, was first prepared by Hailes M493 55 from CpZZrClz and LiAlH4 or LiA1(O-t-Bu)3H. Schwartz ggLafléz reported that it could be easily prepared by treatment of zirconocene dichloride in THF with a stoichiometric amount of NaA1H2(OR)2 (Vitride). Both procedures give high yield of zirconocene hydridochloride. The preparation of polymer-attached Cp22r(H)C1 from Vitride and supported CpZZrClz, which was prepared from Method b or c, has been difficult to achieve, possibly due to the remaining unreacted cyclopentadienyl groups on the polymer taking up some alumium hydride reagents and the optimum amount of Vitride is difficult to determine. For example, with a CpZZrClz-substituted polymer with a loading of 0.32 mmol of CpZZrCl2 per g of polymer, there are 1.58 mmol of unreac- ted cyclopentadienyl groups per g of polymer remaining on the polymer after attachment of CpZZrCl2 to the polymer. The remaining unreacted cyclOpentadienyl groups on the polymer will react with Vitride to give cyclopentadienide. Attempts to increase the formation of supported CpZZr(H)Cl by varying the molar ratio of Vitride to zircono- cene dichloride from one half to two are not successful. The polymer- attached methylene-bridged Cp22r012 beads with higher loading of zirconium were also used, but no improvement in the production of 37 aldehydes was observed. To solve the above problems, the polymer-attached CpZZrCl2 beads which prepared from the reaction of CerCl3 and lithium cyclopentadi- enide-substituted copolymer beads (Method a) were used. The remaining cyclopentadienide groups instead of cyclopentadienyl groups on the polymer will not take up the aluminum hydride reagents. Therefore, polymer-attached CpZZr(H)Cl can be easily prepared by treatment of supported CpZZrCl2 with a stoichiometric amount of Vitride, and then used in hydrozirconation to produce aldehydes with high yields. However, the recovered beads from previous reaction can not be reused in hydrozirconation because the cyclopentadienide groups remaining on the polymer are converted into cyclopentadienyl groups after hydro- lysis with dilute aqueous HCl. Hailes gt_al§5 reported that the IR spectrum of szZr(H)Cl showed one broad Zr-H stretching band at 1390 cm']. Attempts to determine the species on the polymer after the reaction of supported CpZZrC12 with Vitride are not successful due to the background absorption of the polymer and the low loadings used. 38 I III Amoev .eeeoeee-e cop eeexee-m .~.~ .e.e eANe.ov a m.o_ ANNAV Peeeeeee-e om eeexee-p em.. AA.N eaue.cv a e.e ANNFV .eeeeeee-e co. eeexe;-_ o.ep m.m 8A.».ov a m.°_ ARFNV .eeeeeee-e co. eeexee-F o.~p m.w eAFm.cv a o... c on eeexee-. e.m m.e eAPm.ov m e.w c om eeeeeee-F o.e m.m Ase.ov a o.m “we. .eeexee-e om eeoueee-_ o.e m.m ase.cv a o.m a on eeexee-P c.m m.m Ahe.ov a o.m o me eeexee-. A._ m.m Aee.ov a c.m o ow eeeoeee-. o.~ w.. A~m.ov m A.m o om eeeoeee-_ o.. 3., As~.ov a F.A e co eeexoe-_ o.F o.~ .Nm.cv a «.8 ARV “.mevou to eepeo_> .eee 2N _eee Am\e~ _eee .me.eee_v umocxgoup< assumes; mceempo com: macmuuuom coeapoa mo .»3 .NpueNNau cocumuuaucmsapon new: mmcoxpu mo :o_uecooew~ocu»; scam mmca;oupa mo mwmmgacam .e opnap 39 .covuouoc maom>wca mg» soc» uopozoom .u .vom: we: Au voguozv eosxpoa nouauwumnamuovwcowu tuucoqopuzu Ezwguv— use m—QENQu we cowuuemc use seem voemaoca sows: NPQLNNau vogunuuuteoexpoa .u .voma we: NpucNmnu ummuwea ocopxzuoe vogoeuumucosapoe .n .muwcopsovv mcooocoucwu co women use nope?» web .a .eeeeov peeeeeee-e so. eeexee-. e..~ m~.e came.ov a ~.o. ARFmv Peeeeeee-e cc. eeexee-_ ep.~ m~.e efiue.cv a ~.o. ARV A_mev cu ce eepeop> peas eN .eee Am\e~ .eee .me.eee_v emou5;oup< oezmmmea mcpmm—o vow: muceuoeam cesspoa we .u: A.u.ueeov e epeee 40 Reduction of Carbon Monoxide to Alcohols by Polymer-Attached CEZZrCl2 Catalysts Schwartz et al56 reported that diisobutyl aluminum hydride (DIBAH) in the presence of CpZZrClz as a catalyst could reduce carbon monoxide at room temperature to give, on hydrolysis, a mixture of linear alipha- tic alcohols. Reduction of C0 by DIBAH in the presence of polymer- attached CpZZrC12 catalyst has been studied. The reaction mixture after hydrolysis gave the indication of methanol and propanol as shown by GLC analysis. Unfortunately, the amount is too small to be isolated for further analysis. ESR Studies The ESR spectra of the reduced polymer-attached szTiC12 species have been reported by C. P. Lau.26 It has been claimed that the species formed by reduction of polymer-supported szTiC12 under argon is the supported titanocene, while that generated under hydrogen is the supported titanocene monohydride. The ESR spectra of reduced species formed by the reaction of polymer-attached CpZZrCl2 with n-butyllithium give a singlet signal regardless of the reduction carried out under argon or hydrogen (Figure 23 and 24). The siglet is possibly due to supported zirconocene formed by reduction. 41 9:2.0055 I l l l j a l l l__. I 3320 3360 3400 3440 3480 Figure 23. The ESR spectrum of the reduced species of polymer-attached Cp22r012 beads under argon. 42 g-2.0061 I I J I I I I I l 3320 3360 3400 3440 3480 Figure 24. The ESR spectrum of the reduced species of polymer-attached Cp22r012 beads under hydrogen. 43 Nitrogen Fixation Studies 57 Since the first report of an uncharacterized but isolable dicyclopentadienyltitanium dinitrogen complex, homogeneous titanocene has been intensively studied and used for nitrogen fixation.58'67 The titanocene dinitrogen complexes can be generated by the reaction of titanocene dichloride with a variety of reducing agents in a nitro- gen atmosphere. At times considerable confusion has arisen over the nature of the dinitrogen complexes. But it is believed that a dinuclear titanium complex is involved in the process of titanocene nitrogen fixation. Attempts to fix nitrogen were made by Kroll68 and Chandrasekaran27 using crosslinked copolymer-supported titanocene. No significant amount of ammonia was produced by the polymer-supported titanocene, possibly due to the titanocene centers being separated too far to fix nitrogen as a dinuclear Ti complex. This leads to the expectation that if the titanocene centers are brought closer by methylene-bridged ligands, then some nitrogen fixation should be observed. This was found out to be the case. The results of nitrogen fixation by using polymer- supported methylene-bridged titanocene are given in Table 5 and 6. Two approaches for the preparation of polymer-supported methylene- bridged titanocene dichloride have been developed. First approach starts with cyclopentadienide-substituted copolymer beads which prepa- red from the reaction of chloromethylated copolymer beads with sodium cyclOpentadienide, followed by reacting with alkyl lithium reagents. The reaction of cyclopentadienide-substituted polymer with methylene chloridegfollowed by reacting with sodium cyclopentadienide, gives 44 methylene-bridged cyclopentadiene-substituted polymer. The methylene- bridged cyclopentadienyl groups are converted into cyclopentadienide, then reacted with CpTiC13 to give supported methylene-bridged titanocene dichloride. The reaction scheme is shown in Figure 25. Chloromethyl ethers are suspected carcinogens and are used in the step of chloromethylation. In order to overcome this problem, the second approach starting with polystyrene beads was developed. Direct lithiation113'115 of cross-linked polystyrenes with butyllithium- tetramethylethylenediamine (BuLi.TMEDA), followed by reacting with carbon tetrachloride, gives trichloromethylated beads. Further reac- tion of trichloromethylated polymer with sodium cyclopentadienide and then removal of the chlorine atom gives methylene-bridged cyclopenta- diene-substituted polymer. The methylene-bridged cyclopentadiene- substituted polymer beads are then treated with butyllithium and CpTiCl3 to give supported methylene-bridged titanocene dichloride. The reaction scheme is shown in Figure 26. 45 Table 5. van Tamelin nitrogen fixations (flow system). Titanocene Reaction Reactants Used Products Produced Dichloride(g) Temp.(°C) mmol Ti mmol NaNp mmol NH3 mmol NH,/mmol Ti Homogeneous -40 0.676 6.8 0.04 0.06 (0.1684 g) Homogeneous -40 0.915 17.0 0.06 0.07 (0.2279 g) Homogeneous 25 1.213 22.5 0.01 0.01 (0.3020 g) 20% supported Methylene-Bridged -40 0.366 3.7 0.00 0.00 (0.523 g) 2% Supported Methylene-Bridged -40 1.869 34.7 0.00 0.00 (1.201 g) 46 Table 6. Vol'Pin-Shur nitrogen fixations (at 1500 psi of N2). Titanocene Reactants Used Products Produced Dichloride(g) mmol Ti mmol NaNp mmol NH3 mmol NH,/mmol Ti Homogeneous 1.11 17.0 0.50 0.45 (0.28 g) Homogeneous 1.61 40.0 0.89 0.55 (0.40 g) 20% Sup orted 2.30 38.9 0.02 0.01 (5.11 g 20% Supported Methylene-Bridged 2.45 60.0 0.15 0.06 (2.30 g) 20% Supported Methylene-Bridged 0.89 40.0 0.10 0.11 (0.84 g) 20% Supported Dimethylene-Bridged 1.83 35.2 0.16 0.15 (2.13 g) 20% Supported Trimeth lene-Bridged 1.50 39.5 0.14 0.16 (1.36 g 20% Supported . Tetramethylene-Bridged 1.97 40.6 0.11 0.13 (2.19 g) 2% Supported Methylene-Bridged 1.93 50.0 0.19 0.10 (1.24 g) 2% Supported Methylene-Bridged 2.21 100.0 0.30 0.13 (1.42 g) 47 (2)NaC17/THF—. D Cl'hln’ 1 (DBuLi/THF (2) CPTIC13 Figure 25. Scheme for the preparation of polymer-attached methylene-bridged szTi012 via chloromethylation. 48 Li 0, NaCP I N (DVIrnmE/THF § ‘— .JH" H (2)Meotj/Hzo O 8 0”» (DBuLVTHF Cl Cl Cl Cl Figure 26. Scheme for the preparation of polymer-attached methylene-bridged szTi012 via lithiation. 49 Preparation and Study of Polymer-Attached Titanocene Dicarboqyl ( P -C23Ti(CO)3) Titanocene dicarbonyl has been reported as an active catalyst 69'7] Huffman et al72 for the hydrogenation of diphenylacetylene. reported a homogeneous hydrogenation of carbon monoxide to methane using szTi(CO)2. The synthesis of szTi(CO)2, all starting from titanium(IV) derivatives,73'76 e.g. CpZTiRz (R8CH2Ph or Me) has been reported. Recently, Floriani et_gl?7 reported a convinient synthetic route gave high yield of szTi(CO)2 from the widely available and well-characterized CpZTi(8H4). Polymer-attached CpZTi(CO)2 has been prepared by Lau26 from the reaction of supported titanocene dibutyl beads with carbon monoxide. The preparation of supported CpZTi(CO)2 in this research is from the carbonylation of supported titanocene tetrahydroborate in the presence of tridthylamine. The reaction scheme is shown in Figure 27. The I.R. Spectrum of the supported szTi(C0)2 beads shows two strong CO stret- ching bands at 1880 and 1965 cm"1 (Figure 28). The C0 stretching bands are much more intense than those of dicarbonyl beads prepared from the polymer-attached szTiC12 with butyllithium under CO atmosphere. The supported titanocene dicarbonyl species is an active catalyst for the hydrogenation of diphenylacetylene to give 1,2-diphenylethy1ene and 1,2-diphenylethane under 10 atm of hydrogen and at 100°. However, this supported dicarbonyl is not active catalyst for the hydrogenation of olefins at room temperature or for the isomerization of allylbenzene at 145°. 50 Irradiation of dicarbonyl beads with U.V. light showed disappea- rance of two strong CO stretching bands and the color changed from brown to greyish brown. After stirred under 10 atm of C0, the greyish brown beads showed two weak C0 stretching bands at 1880 and 1965 cm". The study of hydrogenation of carbon monoxide to methane using supported szTi(CO)2 was not successful, possibly due to the active metal centers isolated on the rigid polymer. Figure 27. Scheme for the preparation of polymer-attached CpZTi(C0)z. 51 . y-“ ’ 1965cm'l .1 .. 1880011 1 L 1 I I I _ 2000 1800 cm' Figure 28. IR spectrum of polymer-attached szTi(C0)2. 52 Polymer-Attached Niobium and Tantalum Complexes Recently Schrock et‘al78 reported in a series of papers some new chemistry of niobium and tantalum. The complexes of niobium and 5 tantalum containing a single'" -cyclopentadienyl ligand are, in cont- rast to the bis(nS-cyclopentadienyl) complexes, not easily available.79 The complexes of niobium such as Cpr013 and szNb012 have been atta- ched to the polystyrene beads and proved to be active catalysts in the hydrogenation of olefins and isomerization of allylbenzene by Lau,26 but attempt to attach CprCl4 to the polymer was not successful. The synthesis of CpMCl4 (M=Nb and Ta) has only recently been describedzg’80 Therefore, stimulated by those reports, the attachment of CpMC14 (M=Nb and Ta) to the polymer was developed and their catalytic activities were also examined. Preparation of Polymer-AttachedeMCl4 (MsNb and Te) The complexes CpMCl4 (M-Nb and Ta) were prepared recently from the reaction of MC15 with CpSnR3 (RsCH3 or n-Bu)79’ 8° 79 or Mng2. The attachment of CpMCl4 species to the polymer can be achieved by the reaction scheme shown in Figure 29. Treatment of the cyclopentadiene-substituted copolymer beads with n-butyllithium followed by the addition of R3SnCl (R-Me or n-Bu) gave pale yellow polymer-attached CpSnR3 beads. Then the direct reaction of supported CpSnR3 with 11015 gave supported CpMCl4 beads. These species were confirmed by the comparison of their colors with unattached species and by ehmental analyses. This method is good for attaching CpMCln species to polymers. 53 ‘ (0801/1141? mason OMQS HZ-(M “Nb Ta) 7- (R=Me, 11-80) H2 /R GEM/0 Figure 29. Scheme for the preparation of polymer-attached CpMCl4 (M-Ib, Ta). Preparation of Polymer-Attached (“I-Benzy1)Niobium and Tantalum Chlorides ( P -Ph-CH,MC1 , M=Nb and Ta) Transition metal complexes attached to polymers by means of phosphine, nitrogen of n-bonded ligands have been studied extensively. However, polymers where the transition metal complex is attached to the matrix by a carbon-to-metal a-bond have rarely been studied.8]"85 Pittman et_el§2'84 found the use of metal carbonyl anions provided convenient routes to incorporate transition metal carbonyl moieties into polymers. In this research, polymer-bound (n‘-benzyl)niobium and tantalum chlorides have been prepared by treating polymer-bound 54 (n‘-benzy1)tin alkyls with anhydrous niobium and tantalum pentachlo- rides respectively. Polymer-bound (n1-benzy1)trimethy1tin can be prepared in good yield by treating chloromethylated polystyrene beads with trimethylstannyllithium which was prepared by treatment of a tetrahydrofuran solution of hexamethyldistannane with methyllithium.86 The pr0posed reaction may be illustrated in Figure 30. This method provides convenient routes for attaching transition metal halides to 1 polymers through a n -benzy1 ligand. Figure 30. Scheme for the preparation of polymer-attached (nI-Pncuzmm4 (MsNb and Ta) 55 Olefin Dimerization Catalyzed by Polymer-Attached Tantalum Complexes Recently Schrock et a187’ 88 reported that a neopentylidene complex, Tan(CHCMe3)C12, reacts with ethylene to give a tantallo- cyclopentane complex, CpClZTLCHZCHzCHZCHZ. They also demonstrated that the tantallocyclopentane complex could be intermediates in selective dimerization of ethylene to give 1-butene. The dimerization of ethylene catalyzed by polymer-attached CpTaCl4 was carried out as follows: The polymer-attached CpTaC14 beads were treated with two equiva- lents of neopentyllithium in pentane. The color of beads changed from yellow to brown. The brown beads were washed with pentane. the pressure bottle was then pressurized with 60 psi of ethylene and the mixture was stirred for 2 h. The solvent and ethylene were then removed by means of vacuum. Mesitylene was added and the mixture was stirred at 80 psi of ethylene at 100°. From the GLC analysis of the reaction mixture, it indicated that 1-butene was formed selectively and catalytically with a turnover of 18. When the brown beads were treated with 60 psi of ethylene and then treated with 90 psi of CD. a dicarbonyl complex, supported CpTaC12(C0)287 was formed (vco: 2040 and 1960 cm"). The reaction scheme is shown in Figure 31. 56 CO Ta ' 1- Burma Figure 31. Scheme for the dimerization of ethylene catalyzed by polymer-attached CpTaCl4. 57 Hydrpgenation Catalyzed by Polymer-Attached Niobium and Tantalum Complexes Lau26 reported the reduction product of polymer-attached CprCl3 was an active catalyst in the hydrogenation of olefins. Similarly, the catalytic activities of polymer-attached CprCl4, CpTaCl4 and onI-benzy1)niobium and tantalum chlorides were examined in this research. It was found that no hydrogen uptake was observed for the hydrogenation of olefins at room temperature using these reduced beads. However, the reduction products of these beads are active catalysts in the hydrogenation of diphenylacetylene at 1000 and at 1 atm of H2. For example, diphenylacetylene (1.5 mmol) was hydrogenated to give 1,2-dipheny1ethane via trans-stilbene as an intermediate within 23 h, when 0.63 g of supported CpTaC14 beads (0.79 mmol Ta/g of beads) was used. The reduced polymer-attached CprCl4 beads were much less reactive and brought the hydrogenation to an extent of 20% under the same conditions. The hydrogenations catalyzed by reduction products of polymer-bound (“I-benzyl)niobium and tantalum chlorides were not examined in detail in this case. Isomerization Catalyzed by Polymer-Attached Niobium and Tantalum Complexes The polymer-attached CpTaCl4 beads were treated with excess of n-BuLi in hexane under atmospheric pressure of hydrogen. An active isomerization catalyst, red beads, was formed. The same procedures I were carried out for polymer-attached CprCl4 beads. The generated 58 reduced species, black beads, was much less reactive than that of supported CpTaCl4. The isomerizations of allylbenzene catalyzed by these reduced beads were shown in Figure 32. The reduction products of polymer-bound (“I-benzy1)niobium and tantalum chlorides are also active isomerization catalysts and are as reactive as those of supported CprCl4 and CpTaC14 in the isomerization of allylbenzene respectively. Substrate : H - 144 : 1 100 \ ‘49h.~_ _45—~ 80 ‘~~~-*‘ E “Nu-x- S 60 I; :3 E 40 :2 20 15 ' 20 25 30 Time(h) X Allylbenzene -- Reaction catalyzed by Ta species. A Cis-propenyl benzene ---- Reaction catalyzed o Trans-propenyl benzene by Nb Species. Figure 32. Isomerization of allylbenzene catalyzed by reduced species of polymer-attached CpMCl4 (M-Nb and Ta). 59 Polymer-Attached Cyclopentadienyl Cobalt and Rhodium Dicarbonyls Cobalt and rhodium have been the focus of most research and are metals of choice for commercial hydroformylation of olefins and related synthetic reactions. Many low oxidation state cobalt and rhodium catalysts have been supported on a polymer network through phosphine ligands.]3’ 89'98 Interestingly, the catalytic properties of polymer-supported cyclopentadienyl cobalt and rhodium carbonyl derivatives are rarely studied,99 although compounds of this kind are easily synthesized and characterized. The use of cobalt and rhodium complexes attached to polymers through the reasonably inert cyclopen- tadienyl groups is reported here. It is shown that polystyrene- attached cyclopentadienyl cobalt and rhodium dicarbonyl complexes are active hydrogenation, isomerization, disproportionation, cyclo- trimerization and hydroformylation catalysts. Preparation of Polymer-Attached Cyclppentadienyldicarbonylcobalt Two methods for the synthesis of polymer-attached cyclopentadi- enylcarbonylcobalt, photochemical and thermal techniques, have been developed. In a photochemical reaction, a mixture of cyclopentadiene- substituted polystyrene-divinylbenzene copolymer beads and 002(00)8 in benzene was irradiated with UV light under argon. After being washed and dried in vacuo, the beads showed two strong CO stretching bands at 2020 and 1960 cm'1 (Figure 35) Similar procedures were carried out in a thermal method, except 60 a steam bath was used to reflux the reaction mixture. The beads showed the same C0 stretches in the IR. The reaction scheme is illustrated in Figure 33. COZLCOJB , Benzene, he I 4 OR COzECOIB, CH2C|2,F?EFLLD( Hz . H 2 /co \co Figure 33. Scheme for the preparation of polymer-attached CpCo(C0)2. Preparation of Polymer-Attached Cyclopentadienyldicarbopylrhodium Polymer-attached CpRh(C0)2 beads were prepared from the reaction of the {Rh(CO)2C1}2 dimer and cyclopentadienide ion-substituted copolymer, which was obtained by treating the cyc10pentadiene- substituted copolymer beads with n-BuLi in THF or hexane. The copo- lymer beads show two strong C0 stretching peaks at 2040 cm"1 and 1980 cm"1 (Figure 36). The reaction scheme is shown in Figure 34. 61 N-BuL; A. 08110012050 THF THF Figure 34. Scheme for the preparation of polymer-attached CpRh(C0)2. 62 T T T ' (%)- (%) . b L u . _ a 2040 U 2020 ~ 980 . figsc) . l l l L .| 1_|__L J I ZKJCXD IENDCJICNTI 12IXDC) lggcxplcujiI Figure 35. CO stretching spectra Figure 36. C0 stretching of polymer-attached spectra of polymer- CpCo(CO)2. attached CpRh(C0)2. 63 Hydrogenation of Olefins Rhodium complexes coordinatively bonded to polymer-supported cyclopentadiene ligands are active olefin hydrogenation catalysts. During these reactions the color of the catalyst changed to blaCk. The two IR CO stretching peaks of the catalyst recovered from the reactions show the same decrease. This result should probably be interpreted in terms of the formation rhodium metal. The results are given in Table 7. Some or perhaps all of the catalysis may be due to Rh(0). Polymer-attached CpCo(C0)2 is not an active hydrogenation catalyst under these conditions. Hydrogenation of Ketones and Aldehydes Polymer-attached CpRh(C0)2 will hydrogenate aldehydes and some ketones. Acetone is most easily hydrogenated. Hhen acetophenone is the substrate, the aromatic ring is also hydrogenated, as has been 102 and would be the metal of choice for observed for rhodium compounds total reduction. Other ketones were not hydrogenated. Aldehydes are hydrogenated somewhat selectively by polymer-attached CpRh(C0)2. The black catalyst recovered from the reactions exhibited two C0 stretches that were considerably diminished and suggests that Rh metal has formed and accounted for the catalytic activity. The results are given in Table 8. Table 7. Hydrogenation rates for polymer-attached CpRh(CO)2 beads at 25° and 1 atm. Olefin (1.2 M in hexane) Hydrogenation rate (mL Hzlm/mmol) 1-hexene 23.0 l-octene 20.0 2-octene 17.0 1-dodecene 5.2 cyclohexene 19.2 l-methylcyclohexene 2.9 1.3-cyclooctadiene 0.3 1,5-cyclooctadiene 2.9 a-pinene 1.3 phenylacetylene (cyclotrimerization) styrene 20.1 allylbenzene 19.6 allyl alcohol ' 18.8 methylvinyl ketone 5.7 65 Table 8. Hydrogenation of ketones and aldehydes with polymer-attached CpRh(C0)2. Temperature 100°; Pressure Hz; 7.8 atm. Substrate Molar Ratio Reaction Product Yie1d(%) Note Substrate/Rh Time(h) Acetone 313 24 No Reaction Temperature 25° Acetone 313 12 2-Propanol 100 3-Pentanone 153 24 No Reaction Cyclohexanone 160 24 No Reaction Cyclohexanone 152 24 Cyclohexanol 1 Temperature 150° Pressure 150 psi Acetophenone 221 7 l-Phenylethanol 36 Acetophenone 221 15 1-Phenylethanol 75 1-Cyclohexy1- ethanol 25 Benzaldehyde 161 24 Benzyl Alcohol 8 Benzaldehyde 158 36 Benzyl Alcohol 17 Temperature 150° n-Butanal 170 48 n-Butanol 51 3-Pheny1propana1 219 24 3-Pheny1propanol33 n-Hexanal 159 24 n-Hexanol 50 n-Hexanal 159 48 n-Hexanol 65 66 Isomerization Allylbenzene can be isomerized to form the more stable cis and trans-propenylbenzene by using a polymer-attached CpRh(C0)2 catalyst. The catalyst was reused repeatedly without loss of activity. The isomerization of l,5-cyclooctadiene gives a mixture of 1,3- and 1,4-cyclooctadiene. Comparison of the results shows that the rate of isomerization of 1,5-cyclooctadiene is much slower than that of allylbenzene. The two CO stretches of the catalysts are much less intense in the spectra of resins recovered from the isomerization reaction of 1,5-cyclooctadiene than those of the catalysts recovered from the isomerization reaction of allylbenzene and might be due, in part, to the formation of rhodium diene complex during the reaction. The results for the isomerization of allylbenzene, l,5-cyclooctadiene and cis-stilbene are shown in Table 9. During the reaction, the catalyst changed to black. The two CO stretches of the recovered black catalysts were much weaker than those of the fresh beads. Thermal C0 elimination is indicated. Rhodium metal may be the active catalyst. Polymer-attached CpCo(CO)2 was not an active isomerization catalyst under these conditions. 67 .co.uouoe msoa>oca soc» vopuauoc we: vow: umapaunue ll w.eo ~.m me. man mm «.mm m.oe me— com on econ—.umumceea ocoap.umtm.u ocoappamumvo e.on m.v p.m~ mep com ea «.mm o.o m.em me. com me o.¢m —.op m.mm me— com mp «cove oco.u ocovu ocovc touooopozuum.— lagoon—ozone.— -ouooapoaoum.p tauuoopoAuum.p “.mm m.m m.m map occm N o.—m o.m o mep com m o.mw ~.m ~.w mcp ocm N mwuwneee awwweee c.mm mN men on upxcoaocaumcucu apxcoaocgumvu. mco~coapxppn oconcon—appa Auev e¢\eeeeene=m .gv we.» “acme can opoev muoaeoee oesuaconeo» ovuax capo: covuoaox .Nficuveaeo eeeeeeoe-eeea_ee >3 econpvumum'o use .mcowvcauooFUAuum.p .mcm~:mapappa mo copuuNVLoEOm. mg» .m epoch Disproportionation The disproportionation of 1,4-cyclohexadiene catalyzed by polymer-attached CpRh(CO)2 initially gave benzene, cyclohexene, and cyclohexane and finally only benzene and cyclohexane. Cyclohexene also disproportioned to give benzene and cyclohexane within an hour. Thus, polymer-attached CoRh(CO)2 is also a good catalyst for olefin disproportionation. The black beads recovered from the reactions still contain CO ligands according to IR spectra but the CO intensities were much less intense than those of the initial material. Again Rh(O) is indicated as the active catalyst. The results are given in Table 10. Table 10. The disproportionation of 1,4-cyclohexadiene and cyclohexene by polymer-attached CpRh(CO)2. Temperature 145°. Reaction Molar Ratio Products (mole per cent) Time (h) Substrate/Rh 1,4-Cyc10- Benzene Cyclohexene Cyclohexane hexadiene 1,4-Cyclohexadiene 5 1000 8.7 49.9 35.3 6.1 16 1000 2.3 62.4 12.0 23.3 30 1000 0 66.9 0 33.1 Cyclohexene 0.5 930 31.1 7.5 61.5 1 930 33.2 0 66.8 69 Cyclotrimerization of Ethyl Propiolate Homogeneous CpCo(C0)2 and CpRh(CO)2 serve as active catalysts for cyclotrimerization of a wide variety of acetylenic compounds.103 Ethyl propiolate is one of the most reactive substrates. Reactions were carried out by stirring a suspension of the catalyst beads in the presence of ethyl propiolate in benzene, under argon. The reaction mixture was heated for periods of 6 to 48 h at 80°. Hark-up of the benzene reaction mixtures with polymer-attached CpRh(C0)2 beads led to a yield of 17.5% 1,3,5-tricarbethoxybenzene and 52.5% of l,2,4-tricarbethoxybenzene by weight based on the amount of ethyl propiolate used. Ethyl propiolate was only converted in 20.5% yield to a mixture consisting of 2.4% 1,3,5-tricarbethoxybenzene and 18.1% of 1,2,4-trisubstituted isomer by using polymer-attached CpCo(C0)2. Same loss of intensity in the CO stretches was noted. The results are given in Table 11. The dark brown catalyst recovered from the reaction mixture was active in a second reaction converting only a slightly lower percentage of ethyl propiolate. These dark brown beads contain coordinated CO ligands and exhibit intense ester CO absorptions in IR. 70 Table 11. Cyclotrimerizationsa of ethyl propiolate with polymer- attached CpCo(CO)2 and CpRh(C0)2. Polymer-Attached Molar Ratio Reaction Conver- Products(%)b catalyst (g) Substrate/M Time siona 1,3,5-Tri- 1,2,4-Tri- carbethoxy- carbethoxy- (MsCo or Rh) (h) benzene benzene CpCo(C0)2 435 48 20.5 2.4 18.1 (0.14) ecycledc 435 48 17.5 1.7 15.9 {0.14) CpRh(CO)2 812 6 38.0 14.0 24.0 (0.24) recycledc 812 5 30.0 10.0 20.0 (0.24) CpRh(C0)2 661 32 70.0 17.5 52.5 (0.30) aPer cent by weight converted to cyclized products. bHeight per cent based on the amount of ethyl propiolate used. cPolymer supported reagent used was recycled under argon from preceding reaction. 71 Hydroformylation It is well known that Group VIII transition metal complexes, such as those of Rh and Co, are good catalysts for hydroformylation of olefins. Recently, much attention has been directed towards the polymeric phosphine-attached rhodium complexes for hydroformylation of olefins. But the use of rhodium complexes coordinatively bonded to polymeric cyclopentadiene ligands has not been examined exten- sively. The results of hydroformylation with polymer-attached CpRh(CO)2 catalysts are presented in Table 12. No hydroformylation was observed for polymer-attached CpCo(CO)2 catalyst even at 110°, under 120 atm pressures of CO and H2. The selectivity of linear/branched aldehydes was about 1 when the H2/CO ratio was 1. At equal pressures, the selectivity increased as temperature increased. Similarly, the selectivity increased as pressure increased at constant temperature. The selectivity varied slightly by the addition of triphenyl- phosphine. For example, the selectivity increased from 1.24 to 2.06 when the P/Rh ratio increased from O to 20. But when the P/Rh ratio was raised to 50, the selectivity dropped to 1.23. The selectivity also varied slightly when the H2/C0 ratio was changed. The dark brown beads recovered from the hydroformylation reactions showed two CO stretches as strong as those of the fresh beads and were reused without loss of activity. 72 Table 12. Hydroformylations of l-pentene and 1-hexene catalyzed by polymer-attached CpRh(CO)2. Height of Substrate Molar Ratio Time Temp Pressurea Yieldd Selecti:— vity Catalyst(g) Substrate/Rh (h) (0C) (psig) (%) (Linear/ __. Branched) 0.109 1-Pentene 430 20 20 1500 50.7 0.83 0.119 394 9 40 1000 53.5 1.00 0.071 550 12 80 500 93.9 0.95 0.071e 550 12 80 500 88.2 0.90 0.093 504 8 80 500b 89.1 1.35 0.090 520 9 80 500c 85.2 0.83 0.101 454 3 80 1000 89.2 1.34 0.130 350 24 110 100 25.0 0.92 0.181 259 4 110 1000 100 1.18 0.112 418 3 110 1500 100 1.24 0.108 434 3 110 1500b 97.0 1.51 0.132 355 3 110 1500c 83.1 0.85 0.143f 328 5 110 1500 99.0 1.27 0.1119 422 5 110 1500 91.1 2.05 0.144h 325 5 110 1500 100 1.23 0.099 l-Hexene 414 10 110 1500 98.3 1.59 aH2/CO - 1/1. bHZICO -1/3. c112/00 . 3/1. inelds are based on olefins consumed. Only small amount of pentane (~5%) formed during the reactjions. eCatalyst used was recycled from preceding reaction. fTriphenylphosphine was added, P/Rh - 5/1. gP/Rh . 20/1. hP/Rh - 50/1. 73 Cyclopentadienyldicarbonyl rhodium, (05H5)Rh(00)2, is well known to form dinuclear species, (05H5)3Rh3(C0)3, at elevated temperatures or under photolytic conditions.104"106 These rhodium carbonyl clusters appear to be quite inert due to a substantially increased metal-metal bond strength. If the cyclopentadienylrhodium dicarbonyl species is attached to a rigid polymer support, then the coordinative- ly unsaturated intermediates, generated by thermal C0 elimination, chould be isolated from each other. These isolated intermediates provide an efficient pathway for the formation of a reactive hydride as a necessary prerequisite for a subsequent olefin hydroformylation reaction. The proposed mechanism is shown in Figure 37. Mechanisms The mechanisms of the above catalytic processes are presumed to be analogous to those proposed for the homogeneous rhodium catalysts. but detailed mechanistic studies have not been made. Either dissoci- ation of CO, or a reduction of the cyclopentadienyl coordination below 115 must precede olefin binding and (in hydrogenation or hydroformyla- tion) hydrogen addition.107 Since olefin isomerization and disproportionation are carried out at 145°, it is possible that decomposition, possibly to rhodium metal, occurred and the rhodium metal has been reSponsible for the isomeri- zation and disproportionation. Several samples that had been used in the disproportionation of cyclohexene for 24 and 48 h were tested for their effectiveness as catalysts in cyclotrimerization of ethyl propiolate. The fresh catalyst gave an 81% yield of cyclic trimers 74 .Nfiouvemeu vogoouuaseoipg an amen—53 .33er coZeEEoeoeuP we 22:28... 85 com oeosom .mm 8.53... 28%8891 oo 4 1.82:8 8558..-. @303: BTW m .mnwumv..prm+.mmmwwmv AI._nu ... 8/ A8Y~$¢®v Monroe ...onol... 8\:~T®1m 75 after 24 h, while the one that had been used in disproportionation for 24 h gave only 62% and the 48 h sample only 35%. Thus it appears that decomposition of the catalyst had indeed occurred and may account for its ability to catalyze the isomerization and disproportionation processes. It is clear that polymer-attached CpRh(CO)2 is an excellent hydroformylation catalyst and, under the conditions described, does not decompose. Decomposition does occur under the conditions given for hydrogenation, isomerization, and disproportionation of olefins, and for cyclotrimerization of alkynes. It is possible that the decomposition product (Rh(0)?) is responsible for the catalysis. Gubitosa and Brintzinger99'101 reported that supported CpRh(CO)2 served as a hydrogenation catalyst and did not note decomposition of the catalyst. Gubitosa and Brintzinger also found that polymer-attached CpCo(CO)2 was an effective hydrogenation and hydroformylation catalysts. In our hands no catalytic activity was found, but it was not possible for us to duplicate their experiments exactly. EXPERIMENTAL General Manipulations involving air-sensitive materials were performed under argon in Schlenk-type vessels. Hhere necessary, transfers were make in an argon-filled glove box. NMR spectra were obtained by use of a Varian T-60 spectrometer. Electron spin resonance (ESR) spectra were obtained by use of a varian E-4 Spectrometer. IR spectra were recorded on Perkin-Elmer 457 or 2378 spectrophotometers. Samples were prepared by crushing the polystyrene beads in a ball mill under anaerobic conditions and mulling the powder with dry nujol in a dry glovebox. Gas chromatography, GLC, analyses were performed by use of Varian model 1400 analytical gas chromatograph and a model 920 gas chromatograph. All solvents used were A. C. S. reagent grade. Tetrahydrofuran (THF), hexane, cyclohexane, toluene, xylene and benzene were distilled from sodium-benzophenone under argon. All alkenes used for hydrogena- tions were at least 95% pure, and were further purified by distillation from sodium under argon. Allyl benzene and l.5-cyclooctadiene were dried through an activated alumina column and distilled under vacuum. N,N,N',N'-tetramethy1ethylenediamine (TMEDA) was dried over molecular sieves (4A) for 4 h before distilling under argon. Analytical grade cis-stilbene, aldehydes and ketones were used without further purifi- cation. Ethyl propiolate was distilled under vacuum and a center cut collected. The reagents, n-butyl lithium, allylbenzene, l-hexene, 76 77 cyclohexene, 1-octene, 3-hexene, l,5-cyclooctadiene, ethyl propiolate, trimethyltin chloride, tri-n-butyltin chloride, hexamethyldistannane and chloromethyl ethyl ether were obtained from Aldrich Chemical Co. 1-Pentene was obtained from J.T. Baker Chemical Co. Dicobalt octa- carbonyl, chlorodicarbonylrhodium dimer and cyclopentadienyltitanium trichloride were obtained from Strem Chemicals Inc. Zirconium tetra- chloride, hafnium tetrachloride, niobium pentachloride, tantalum pentachloride, bis(cyclopentadieny1)zirconium dichloride and bis(cyclo- pentadieny1)hafnium dichloride were obtained from Alfa Chemical Co. Sodium bis(2-methoxyethoxy)aluminum hydride (Vitride) was obtained from Eastmen Kodak Co. The 20% cross-linked polystyrene-divinylbenzene copolymer beads were a gift from the Dow Chemical Co. and were washed to remove impurities before use. They were washed with 10% HCl (aq.), 10% NaOH (aq.), H20, HZO-CH3OH (1:1), CH30H, CH30H-CH2012 and benzene and then dried under vacuum. Analytical Methods Chloride in the chloromethylated beads was removed from the copolymer beads with boiling pyridine and determined by Volhard technique.108 Similarly, chloride from the supported metallocene chlorides beads was removed from the copolymer with boiling 2 N KOH solution for 24 h and determined by the same technique. Titanium, zirconium and hafnium were determined by ignition of the metal- containing polymer at 10000 for 24 h and weighed as the oxides.109 78 Other elemental analyses were performed by Schwarzkopf Microanalytical Laboratory, Hoodside, N.Y. Preparation of Cyclopentadiene-Substituted Copolymer Typically, following the chloromethylation method of Pepper, et_ el_38, 50 g of washed and dried capolymer beads was taken into a 1 L, three-necked, flask with a drying tube and an overhead stirrer. About 180 mL of chloromethyl ethyl ether (chloromethyl ethyl ether is a “cancer suspect agent" because the related compound (a possible contaminant), bis-(chloromethy1)ether, is a carcinogen) was added and the mixture was stirred for 2 h. A solution prepared by cautiously adding 8.5 mL of SnC14 to 80 mL of ice-chilled chloromethyl ethyl ether was then introduced slowly through a dropping funnel. After the reaction mixture was stirred vigorously at room temperature for 18 h, the ether was removed by suction and a dispersion tube. The beads were washed with four 250 mL portions of 50% aqueous dioxane, aqueous dioxane containing 10% HCl (v/v) and finally with dry dioxane until the washings were chloride free.- The chloride analysis of the chloromethylated beads after it had been dried for 2 days jg_!eggg_ yielded 1.45 mmol of 01/9 of beads or 16.7% chloromethylation of the styrene rings. The chloromethylated beads obtained above were suspended in 100 mL dry, air-free THF and 100 mL of 2.0 M sodium cyclopentadienide in THF. After the mixture was stirred for 6 days at room temperature, excess sodium cyclopentadienide and THF were removed by filtration. The beads were washed with ethanol/THF (1/1) and THF until the 79 washings were chloride free and then were dried under vacuum. The beads contain 0.04 mmol of Cllg of beads. Thus the cyclopentadiene content was 1.41 mmol C5H5/g of beads determined by the difference between the Cl contents before and after reaction with sodium cyclo- pentadienide. Preparation of Polymer-Attached Biscyclopentadienyl Zirconium and Hafnium Dichlorides ( P-CpZMClZ; M-Zr, Hf) Method a: Magnesium cyc10pentadienide,]°° Mg(C5H5)2, may be prepared by the direct reaction of magnesium metal with cyclopentadiene vapor at 500° to 6000 under argon. The pure white crystalline Mg(C5H5)2 was then obtained by sublimation in a vacuum at 600 and identified by its melt- ing point and IR spectrum. Monocyclopentadienyl zirconium trichloride was also prepared by the reported procedure.“2 Treatment of ZrCl4 in xylene with Mg(CSH5)2 (0.5 mol. equiv.) gave CerC13. The pure CerCl3 was obtained by sublimation under vacuum (10'4 mmHg) at 90°. In a typical reaction, 2 g of cycl0pentadiene-substituted c0po- lymer beads (0.5 mmol Cp/g of beads) was treated with excess butyl- lithium in THF and the mixture was stirred for 2 days under argon. The excess of n-BuLi was removed and the beads were washed with THF. They were then suspended in 30 mL of THF, and CerCl3 (1.3 mmol) was added. The reaction mixture was stirred for 2 days. The solution was then removed and the beads were washed with THF in a soxhlet 80 extractor for 3 days. The product, yellowesh beads, was dried under vacuum. Anal. Calcd: Cl/Zr, 2.00. Found: Zr, 0.420 mmol/g of beads: 01, 0.827 mmol/g of beads: Cl/Zr, 1.97. Method b: In a typical reaction, 2 g of cyclopentadiene-substituted copo- lymer beads (1.41 mmol Cp/g of beads) was treated with excess butyl- lithium in THF and the mixture was stirred for 2 days under argon. The excess of n-BuLi was removed and the beads were washed with THF. They were then suspended in 20 mL of THF, and 2 g of szMC12(M=Zr, Hf) was added. The reaction mixture was stirred for 3 days, then treated with excess anhydrous hydrogen chloride gas for 4 h. The solution was removed and the beads were washed with THF in a soxhlet extractor for 5 days. The product, brownish yellow beads, was dried under vacuum. Anal. Calcd. for supported zirconocene dichloride: C1/Zr, 2.00. Found: Zr, 0.142 mmol/g of beads: Cl, 0.328 mmol/g of beads; C1/Zr, 2.31. Anal. Calcd. for supported hafnocene dichloride: Cl/Hf, 2.00. Found: Hf, 0.126 mmol/g‘of beads; C1, 0.302 mmol/g of beads; Cl/Hf, 2.40. Method c: Typically, 10 g of polymer-attached CerCl3 (0.636 mmol Zr/g of beads) or CprCl3 (0.463 mmol Hf/g of beads) were suspended in 10 mL of THF and 3.1 mL (2.3 mL for CprCl3) of 2.2 M sodium cyclopenta- dienide in THF was added. After the reaction mixture was stirred for 5 days, the beads were washed with THF several times. Then a small 81 amount of hydrogen chloride gas was introduced and allowed to stir for 24 h. After being washed with THF in a soxhlet extractor for 3 days, the yellow beads were dried jg_!eegg_for 2 days. Anal. Calcd. for supported zirconocene dichloride: Cl/Zr, 2.00. Found: Zr, 0.627 mmol/g of beads: Cl, 1.223 mmol/g of beads: Cler, 1.95. Anal. Calcd. for supported hafnocene dichloride: Cl/Hf, 2.00. Found: Hf, 0.450 mmol/g of beads; Cl, 0.869 mmol/g of beads; C1/Hf, 1.93. Preparation of Polymer-Attached Monocyclopentadienyl Zirconium and Hafnium Trichloride ( P-CpMC13; M=Zr, Hf) About 5 g of polymer-attached lithium cyclopentadienide beads (1.41 mmol Cp/g of beads) was suspended in 40 mL of benzene and 2 g of MC14 (M=Zr, Hf) was added. The reaction mixture was stirred for 7 days. The solution was then removed. After being washed with THF in a soxhlet extractor for 4 days, the product, cream colored beads, was dried. Anal. Calcd. for supported CerCl3: C1/Zr, 3.00. Found: Zr, 0.338 mmol/g of beads; Cl, 0.953 mmol/g of beads; Cl/Zr, 2.82. Anal. Calcd. for supported CprCl3: Cl/Hf, 3.00. Found: Hf, 0.428 mmol/g of beads; Cl, 1.198 mmol/g of beads; Cl/Hf, 2.79. Preparation of Polymer-Attached Tricyclopentadienyl Zirconium and Hafnium Monochloride ( P-Cp3M01; M=Zr, Hf) The polymer-attached lithium cyclopentadienide beads were sus- pended in THF and a two-fold excess of Cp22r01 was added. The reaction mixture was stirred for 3 days. The excess of CpZZrCl2 was 82 removed and the beads were then washed with THF in a soxhlet extractor until excess chloride had been removed. The product, yellow beads, was dried. Anal. Calcd. for supported Cp3Zr01: 1.00. Found: Zr, 0.302 mmol/g of beads; C1, 0.297 mmol/g of beads: C1/Zr, 0.98. Anal. Calcd. for supported Cp3HfCl: 1.00. Found: Hf, 0.345 mmol/g of beads; C1, 0.328 mmol/g of beads; C1/Hf, 0.95. Far Infrared Studies The far infrared spectra in the 100-500 cm'] region were obtained by use of a Digilab Model FTS-16 fourier transform spectrophotometer. The samples were prepared by crushing the polystyrene-divinylbenzene copolymer beads in a ball mill under anaerobic conditions and mulling the powder with dry nujol in a glove box. The spectra were recorded with the samples mounted between polyethylene plates in a dry nitrogen atmosphere. The spectra were obtained by scanning the samples 1 (nujol mull) 1000 times with a resolution of 4 cm' and a sampling interval of 8 microns. Hydrogenation Studies The polymer-attached catalysts were weighed into a 100 mL round- bottomed flask with a side arm. The beads were suspended in 5 mL n-hexane and treated with 3 mL of 2.0 M n-butyllithium in n-hexane for 2 days. The excess of n-butyllithium was removed and the beads were washed with n-hexane several times. Then 8.5 mL hexane was injected into the flask and 1.5 mL 1-hexene was added. The reaction mixture was stirred at a constant stirring rate, while the temperature 83 was maintained at 25°. The rate of hydrogen uptake was measured by using an 100 mL gas buret. For the hydrogenation of diphenylacetylene, the catalyst was suspended in 5 mL toluene, then 0.267 g (1.5 mmol) of diphenylacety- lene in 5 mL toluene was added into the flask. The rate of hydrogen was measured at 1000 and at normal atmospheric pressure. A small amount of reaction mixture was drawn out periodically and then chromatographed by GLC on a 8 ft x 1/8 in 15% SE-30/Chromosorb P column. The relative amounts of diphenylacetylene, trans-stilbene and 1,2-dipheny1ethane were measured by integration. Isomerization The reduction of polymer-attached zirconocene and hafnocene chlo- ride beads with excess n-butyllithium in hexane was carried out under hydrogen at room temperature for 2 days. The excess of n-butyllithium was removed and the beads were washed with hexane several times and dried under vacuum. Allylbenzene or 1,5-cyclooctadiene was then added and allowed to stir under argon at 145°. A small amount of reaction mixture was drawn out periodically and chromatographed by GLC on a 8 ft x 1/8 in 15%SE-30/Chromosorb P column. The relative amounts of allylbenzene, cis-propenylbenzene and trans-propenylbenzene were measured by integration. The relative amounts of 1,3-, 1,4- and l,5-cyclooctadiene were measured by GLC integration and NMR integra% tion. 84 Epoxidation Reactions were carried out under an atmosphere of argon. A quantity of 5 mL benzene or cyclohexane, 3 mL cyclohexene (30 mmol) and the catalyst (0.09 mmol Zr of Hf) was warmed to 800 and 8 mL of 80% tert-butyl hydrOperoxide (60 mmol) was added. Aliquots were removed at time intervals and analyzed by GLC on 16 ft x 1/8 in 20% Carbowax/Chromosorb H and 8 ft x 1/8 in 15% SE-30/Chromosorb P columns. Chlorobenzene was used as an internal standard. Hydrozirconation of Olefins Typically, the reduction of 5 g of polymer-attached CpZZrC12 beads (3.3 mmol Zr) with Vitride (1.7 mmol) in THF was carried out in a 200 mL pressure bottle under argon for 2 days. After the beads were washed with THF and benzene several times, a quantity of 2 mL of l-hexene and 10 mL of benzene was added and allowed to stir for 2 days. The argon was then removed and carbon monoxide was introduced (150 psi). The reaction mixture was allowed to stir for 2 days at room temperature and then hydrolyzed with dilute aqueous hydrochloric acid. The solution was taken out and extracted with ether. The extract was concentrated and then analyzed for aldehydes by GLC on a 16 ft x 1/8 in 20% Carbowax/Chromosorb H column. 85 Preparation of Polymer-Attached Methylene Bridged-bi5(eyclppentadienyl) Zirconium Dichloride About 30 g of cyclopentadienide ion-substituted capolymer beads (2.2 mmol Cp/g of beads) was suspended in 80 mL of THF and the reaction flask was cooled in an ice bath. Then, 30 mL of freshly distilled methylene chloride was added slowly. The dark red beads changed to yellow immediately. After stirred at room temperature for one day, the beads were washed with THF in a soxhlet extractor and dried under a vacuum. The chloride analysis gave 2.0 mmol Cl/g of beads. To the yellowish beads suspended in 80 mL of THF, three fold excess of sodium cyclopentadienide in THF was added. After stirred for 5 days, the beads were dried under vacuum. The product, yellowish methylene bridged cyclopentadiene-substituted copolymer beads, contains 0.05 mmol Cl/g of beads. Following the method c in the preparation of polymer-attached bis(cyclopentadienyl)zirconium dichloride, the polymer-attached methylene-bridged Cp22r012 beads were prepared from the methylene bridged cyclopentadiene-substituted copolymer beads and ZrCl4. The elemental analysis gave 0.81 mmol Zr/g of beads. Reduction of Carbon Monoxide to Alcohols by Polymer-Attached szlrglz Catalysts Typically, about 8 g of supported CpZZrCl2 beads (0.67 mmol Zr/ 9 of beads) in benzene in the presence of 3 equivalents of diiso- butylaluminum hydride (DIBAH) was stirred in a 200 mL pressure bottle 86 under 60 psi of C0 atmosphere for 3 days. The reaction mixture was then hydrolyzed with dilute aqueous sulfuric acid. The product was analyzed by GLC on 10 ft x 1/8 in 19% FFAP/Chromosorb G and 12 ft x 1/8 in Carbowaz lSOO/Chromosorb 0 columns. ESR Studies A sample of 0.3 g polymer-attached zirconocene dichloride (0.112 mmol Zr/g of beads) was charged into a reaction flask and the beads were suspended in 5 mL hexane. The supported zirconocene dichloride was then reduced by the addition of 1 mL 1.6 M butyllithium in hexane and the reaction mixture was stirred for 12 h under hydrogen. The solution was removed and the beads were washed several times with hexane. The beads were dried under vacuum and transferred into the ESR tube. The ESR spectrum of the reduction product was recorded. The same experiment was repeated under argon. Preparation of Polymer-Attached Methylene-Bridged szTiClz Method 1. The procedures for the preparation of methylene-bridged cyclo- pentadiene-substituted copolymer beads is the same as those described in the preparation of polymer-attached methylene-bridged zirconocene dichloride. Typically, 10 g of methylene-bridged cyclopentadiene-substituted copolymer beads (2.2 mmol Cp/g of beads) were suspended in 20 mL THF, and two fold excess of butyllithium in hexane was added. The beads 87 changed from yellow to dark red. After being stirred at room tem- perature for 2 days, the solution was removed and the beads were washed several times with benzene. To the dark red beads wuspended in 50 mL benzene, 9 g of CpTiCl3 was then added and the reaction mixture was allowed to stir for 4 days. The solution was removed and the beads were washed with benzene in a soxhlet extractor. The product, red beads, was dried under a vacuum. Metal analysis was performed. Anal. Found for supported methylene-bridged CpZTiClZ: Ti, 0.700 mmol/g of beads (1st batch); 1.066 mmol/g of beads (2nd batch). Anal. Found for supported dimethylene-bridged szTiClz: Ti, 0.859 mmol/g of beads. Anal. Found for supported trimethylene-bridged szTiC12: Ti, 1.103 mmol/g of beads. Anal. Found for supported tetramethylene-bridged CpZTi01z: Ti, 0.899 mmol/g of beads. Method 2. About 20 g of 2% cross-linked polystyrene-divinylbenzene copo- lymer beads was suspended in 60 mL of freshly distilled cyclohexane. The mixture was refluxed for 24 h, and then 80 mL of 1.6 M butyl- lithium in hexane and 25 mL of N,N,N',N'-tetramethylenediamine (TMEDA) (150 mL) were added by means of a syringe.113’115 The reaction mix- ture was heated for 3 days. After cooling, the polymer beads were washed several times with cyclohexane and THF. To the dark brown lithiated polymer beads suspended in 60 mL of THF and cooled at 0°. 30 mL of CCl4 was then added slowly. After being stirred at room temperature for 24 h, the solution was removed and the beads were washed with THF in a soxhlet extractor. The dried 88 beads gave 3.45 mmol Cl/g of beads. The trichloromethylated beads were suspended in 80 mL of THF and 80 mmol of Nan was added. After being stirred for 5 days, the beads were washed with THF. The dry beads contain 0.9 mmol Cl/g of beads. The beads were then suspended in 80 mL of THF and 35 mmol of Vitride in THF was added. The mixture was refluxed for 24 h. Then, 10 mL of MeOH/HZO (1:1) was added slowly and the beads were washed with THF in a soxhlet extractor. The dried beads, methylene-bridged cyclopenta- diene-substituted copolymer, contain 0.014 mmol Cl/g of beads. The procedures for the attachment of szTiC12 to the 2% cross-linked methylene-bridged cyclopentadiene-substituted copolymer beads are the same as those described in Method 1. Metal analysis gave 1.556 mmol Ti/g of beads. Nitrpgen Fixation Studies van Tamelin Nitrogen Fixation. Typically, 0.343 g of supported methylene-bridged szTi012 beads (1.066 mmol Ti/g of beads) was suspended in 20 mL THF. The reaction flask was cooled at -400 and 3.7 mmol sodium naphthalide in THF was added slowly. Nitrogen was then bubbled through the solution into a gas washing tower filled with 60 mL 20% H2504 for one day. The ammonia produced was analyzed by Kjeldahl method. Nessler's Reagent"6 was also used for the presence of ammonia. No ammonia was detected. 89 Vol'pin-Shur Nitrogen Fixation A11 fixations were done by using 1500 psi of nitrogen in an autoclave previously flushed with nitrogen, at ambient temperatures with continual stirring during the pressurized period of the reaction. In a typical reaction, 2.30 g of supported methylene-bridged CpZTiC12 beads (1.066 mmol Ti/g of beads) was transferred into an autoclave, 60 mmol of sodium naphthalide in THF and 100 mL of diethyl ether. The system was pressurized and stirred for 2 days, then depressurized and acidified with 20 mL of absolute methanol and 20 mL of 20% H2504. After stirring the resulting solution for 3 h in the autoclave, the solution was removed and ammonia was analyzed by Kjeldahl method. Preparation of Polymer-Attached CpaTi(00)3 About 1 g of polymer-attached szTiC12 beads (1.07 mmol Ti/g of beads) was treated with excess LiBH4 in 20 mL of diethyl ether for 2 days. The color of beads changed from red to greyish blue. The beads were then washed with ether and dried under vacuum. Treatment of the greyish blue beads, in 10 mL of Et3N with CO at atmospheric pressure for 2 weeks, gave brown beads. The beads were washed with hexane and dried under vacuum. The IR spectrum of brown beads showed two strong C0 stretching bands of similar intensity at 1880 and 1955 en". The synthesis of polymer-attached CpZZr(C0)2 was not successful when a similar reaction in Et3N with polymer-attached CpZZr(BH4)2 under 150 atm of CD for a week was carried out. 90 Photolysis of Polymer-Attached szTi(CO)a About 0.2 g of polymer-attached szTi(CO)2 beads suspended in 5 mL hexane or toluene was irradiated under argon. The color of beads changed from brown to greyish brown. The IR spectrum of greyish brown beads shows no CO stretching bands. After stirred under 1 atm of CD for 2 days, the greyish brown beads gave two very weak c0 stretching bands at 1880 and 1955 en". Hydrogenation of Carbon Monoxide by Polymer-Attached szTi(C0)a About 5 g of polymer-attached szTi(CO)2 beads (0.7 mmol Ti/g of beads) in 20 mL of toluene was treated with 200 psi of H2 and CO mixture (3:1 molar ratio) at 160°. No methane was observed after the reaction mixture was stirred for 5 days. Trace of methane was observed for another reaction in which 8.5 g of 2% polymer-attached methylene-bridged szTi(CO)2 beads (1.5 mmol Ti/g of beads) was used. Methane was analyzed by GLC on a 20 ft x 1/8 in Durak column and a 6 ft x 1/4 in Molecular Sieve 5A column. Preparation of Polymer-Attached CpMCl4_(M=Nb and Ta) In a typical reaction, 1 g of cycl0pentadiene-substituted copo- lymer beads (1.41 mmol Cp/g of beads) was treated with excess n-BuLi in THF for 2 days under argon. The excess n-BuLi was removed and the beads were washed with THF, then dried under vacuum. The dried beads were suspended in benzene and R3Sn01 (R=n-Bu, or Me) was added. 91 The reaction mixture was stirred for 3 days. The color of beads changed from red to pale yellow. The beads were then washed with benzene and dried. Treatment of pale yellow beads in toluene with MCl5 (MaNb, or Ta) gave red supported CprCl4 beads and yellow supported CpTaCl4 beads, respectively. Anal. Calcd for supported CprCl4: C1/Nb, 4.00. Found: Nb, 0.810 mmol/g of beads: Cl, 3.081 mmol/g of beads; Cl/Nb, 3.80. Anal. Calcd for supported CpTaCl4: C1/Ta, 4.00. Found: Ta, 0.792 mmol/g of beads: Cl, 3.052 mmol/g of beads; Cl/Ta, 3.85. Preperation of Polymer-Attached (PhCHz)M01q (M-Nb and Ta) Typically, 2 g of chloromethylated polystyrene beads (1.57 mmol Cl/g of beads) were suspended in 10 mL of THF and the reaction flask was cooled in an ice bath. A two-fold excess of trimethylstannyl- lithium in THF, which was prepared by treatment of a tetrahydrofuran solution of hexamethyldistannane with methyllithium, was added slowly. After being stirred at room temperature for 36 h, the solution was removed and the beads were washed several times with THF, then dried. The polymer-attached benzyltrimethyltin beads were suspended in 20 mL toluene and treated with excess TaC15. After being stirred for 5 days, the beads were washed in a soxhlet extractor with toluene for 3 days. The product, cream beads, was then dried under vacuum and gave chloride contents of 3.61 mmol/g of beads. 92 Olefin Dimerization Catalyzed by Polymer-Attached Tantalum Complexes Typically, 0.50 g polymer-attached CpTaCl4 beads (0.792 mmol/g of beads) was placed into a pressure bottle and suspended in 15 mL pentane. Two equivalents of neopentyl lithium in pentane (0.80 mmol) was added slowly. The beads changed from yellow to brown. After being stirred for 2 days, the brown beads were washed several times with pentane. The brown beads were then suspended in 15 mL pentane. The pressure bottle was pressurized with 60 psi ethylene and the reaction mixture was stirred for 2 h. The brown color of the beads lightens to pale yellow. The solvent and ethylene were removed and the beads were dried under vacuum. After being st rred at 100° under 80 psi of ethylene for 24 h; the gas sample was analyzed on a 20 ft x 1/8 in Durapak column. When the pale yellow beads suspended in ether were treated with 90 psi of CO far 30 h, a supported dicarbonyl complex was formed. The IR spectrum shows two CO stretching bands at 2020 and 1940 cm'I. Hydrogenation of Olefins and DiphenylaCetylene Catalyzed by Polymer-Attached Niobium and Tantalum Complexes The procedures carried out for hydrogenation of olefins and diphenylacetylene are the same as those for the hydrogenations catalyzed by polymer-attached zirconocene and hafnocene catalysts. 93 Isomerization of Allylbenzene Catalyzed by Polymer-Attached Niobium and Tantalum Complexes In a typical reaction, 0.25 g polymer-supported CpTaC14 beads (0.21 mmol Ta/g of beads) was treated with excess BuLi in hexane under atmospheric pressure of hydrogen for 36 h. The beads changed from light yellow to brown. The brown beads were washed several times with hexane and dried in a vacuum. Allylbenzene (1 mL) was then injected into the reaction flask and allowed to stir under argon at 145°. A small amount of reaction mixture was drawn out periodically and chromatographed by GLC on a 8 ft x 1/8 in 15% SE-30/Chromosorb P column. The relative amounts of allylbenzene, cis-propenylbenzene and trans-propenylbenzene were measured by integration. Preparation of Polymer-Attached CpCo(C0), Photochemical Reaction. A mixture of 3 g C02(CO)8 in 30 mL benzene and 5 g cyclopentadiene- substituted copolymer beads (1.41 mmol °5H5/9 of beads) in a quartz apparatus was irradiated under argon for 2 days with a 140 H Hanovia type 30620 UV lamp. The excess Coz(CO)8 and benzene were then removed and the beads were washed with benzene until the washings were clear. The product, greyish brown CpCo(CO)2-substituted copolymer beads, was then dried under vacuum and yielded a substance containing 0.26 mmol Co/g of beads. 94 Thermal Reaction.119 Five 9 of cyclopentadiene-substituted copolymer beads (1.41 mmol C5H5/g of beads), 50 mL dried methylene chloride and 3 g °°2(°°)8 were placed in a 100 mL flask, which was fitted with a reflux condenser. The system was flushed with argon and was covered with aluminum foil to exclude light. The contents were heated to reflux on a steam bath for 3 days, then the methylene chloride was removed and the beads were washed with methylene chloride until the washings were clear. The greyish green beads were then dried under vacuum and stored under argon. The cobalt content was 0.32 mmol Co/g of beads. Preperation of Polymer-Attached CpRh(CO)212° In a typical reaction, 3 g cyclopentadiene-substituted copolymer beads (0.36 mmol C5H5/g of beads) was treated with a two-fold excess of butyllithium in THF or hexane and the mixture was stirred for 3 days under argon. The excess n-BuLi and THF or hexane were removed and the beads were washed with THF. They were suspended in THF, the reaction flask was cooled in an ice bath, and a solution containing 0.5 mmol of (Rh(CO)201)2 dissolved in THF was then introduced. The orange color of the solution disappeared within 20 min. The mixture was stirred for 2 days at room temperature. The color of beads changed from deep purple-red to dark brown. The solvent was removed and the beads were washed with THF and then dried under vacuum. The rhodium content was 0.29 mmol Rh/g of beads. Other solvents, petroleum ether and hexane, were also used. It was found that THF was the most appro- 95 priate solvent. Hydrogenation of Olefins The hydrogenations were carried out by using gas burets of 100 mL volume. The hydrogen uptake was measured at normal atmospheric pressure and at 25° 3 0.5°. All reactions were carried out in a 100 mL of round-bottomed flask with a side arm. The catalyst (0.29 mmol Rh/g of beads) was weighed into the reaction flask, placed under an atmosphere of hydrogen and suspended in 8-9 mL hexane. The appropriate olefin (1-2 mL) was then added. The rate of hydrogen uptake was measured by using the buret. Hydrogenation of Ketones and Aldehydes A mixture of 0.1-0.3 g of catalyst (0.29 mmol Rh/g of beads), 1 mL aldehyde or ketone and 5 mL benzene were placed in a 100 mL of pressure bottle. The bottle was flushed with hydrogen, pressurized to 7.8 atm with H2 and heated and stirred at 1000 for 7-48 h. After being cooled to room temperature, the reaction mixture was analyzed by GLC and NMR. Isomerization and Disproportionation About 2-4 mL substrates and 0.05-0.2 9 catalyst (0.29 mmol Rh/g of beads) were placed in a 5 mL pressure bottle under argon. The vessel was then heated to 145°. A small amount of reaction mixture was drawn out periodically and then chromatographed by GLC. The 96 relative amounts of reaction products were measured by GLC and NMR integration. Cyclotrimerization of Ethyl Propiolate Typically, 0.3 g of polymer-attached CpRh(CO)2 (0.10 mmol Rh/g of beads) beads was introduced into a side-armed flask containing 5 mL benzene under argon. After the mixture was stirred for about 10 min, 2.0 mL ethyl propiolate (1.91 g; 19.5 mmol) was injected and the reaction mixture was then heated at 80° for 32 h. The color of solu- tion gradually turned dark brown. After it was cooled to room temper- ature, 50 mL benzene was introduced into the flask and after being stirred for several minutes, the solution was drawn out by means of a syringe. The viscous oil, obtained by concentrating the brown solution under vacuum, was then extracted with 200 mL of a refluxing CCl4/petroleum ether (1/4) mixture. The yellow extract was filtered and the solvents were removed under vacuum to give 70% products based on the weight of ethyl propiolate used. The relative amounts of 1,3,5-tricarbethoxybenzene and l,2,4-tricarbethoxybenzene were measured by NMR integration. Hydroformylation In a typical reaction, 0.1 g polymer-attached CpRh(CO)2 catalyst (0.39 mmol Rh/g of beads) and 10 mL dry benzene were introduced into a 250 mL stainless steel autoclave. Following the addition of 2.0 mL 1-pentene (18.3 mmol) and the attachment of the head, the system was 97 flushed three times with hydrogen and twice with carbon monoxide. The system was then pressurized to 14.6 atm with CO and heated to 110° in an oil bath. After 40 min., during which the system achieved physical equilibrium, carbon monoxide and then hydrogen (1:1) were rapidly added to 69 atm and the autoclave was heated and stirred at 110° for 3 h. During this period there was a 6.1 atm pressure drop. After being cooled to room temperature, the vessel was vented and the clear golden solution was drawn out by using a syringe. A sample of the product was injected into the gas chromatograph with the column temperature at 40° in order to separate the pentane reduction product from any unreacted l-pentene. The column temperature was then increased to 90° and the linear and branched hydroformylation products were separated. A column temperature of 130° was used to verify that no higher boiling products-alcohols were obtained. PART II THERMAL DECOMPOSITION 0F TRANSITION METAL ALKYLS INTRODUCTION Since Frankland's discovery‘21 in 1849 of the first organometallic compound, the spontaneously inflammable ZnEtz, metal alkyls have been of considerable interest. Attempts to synthesize transition metal alkyls were not successful until the isolation of the trimethylplatinum iodide tetramer, {(CH3)3PtIl4. a stable product obtained by Pope and Peachy122 in 1907. Unsuccessful attempts to prepare simple transition metal derivatives led to the generalization that they are much less 123 and that transition 124 stable than their main group element analogues metal-carbon bonds are weak, a view supported by calculations. The scope of relevant problems is immense, ranging from theoretical or structural studies to organic or biological chemistry. Many metal alkyls have a key role as synthetic reagents, notably the Grignard reagents and alkyls of Li, Zr, Hg, A1, Sn and Cu. Some main group metal alkyls feature as important materials for industrial use: PbEt4 IV compounds as stabilizers for as an antiknock gasoline additive; Sn p01y(viny1 chloride) and as fungicides or germicides; A1 derivatives in the manufacture of isoprene or long chain alcohols: and Me25i012 in silicones as rubbers, resins, or paper finishes. Vitamin B12 coenzyme provides a metal alkyl as an essential biological material. Transition metal alkyls are key intermediates in many catalytic reactions of olefins or acetylenes, e.g., carbonylation, dimerization, 98 99 polymerization, isomerization and hydrogenation. These intermediates are well characterized in homogeneous catalysis but only recently has it become apparent that similar alkyl species exist on the surfaces of heterogeneous catalysts. Some transition metal alkyls are important in petrochemical, polymer, or heavy organic chemical industry, e.g.. alkyls of Ti in Ziegler-Natta polymerizations and of Co in hydroformy- lation by Coz(CO)B. Metallacycles are also of current interest due to the potential importance of metallacyclic intermediates. Small-ring metallacycles have been implicated as intermediates in a large number of transition metal catalyzed reactions of olefins and acetylenes. Thus, metallacy- clobutanes are postulated to play a key role in the catalysis of the olefin metathesis reactions.]2°'133 Metallacycles are also implicated in various ring openings and rearrangements,134'136 {2 + 2} cycloaddi- 137-141 tions and cyclooligomerization of olefins, and polymerization of silacyclobutanes.”2'143 Transition metal alkyls are often thermally unstable and their thermal instability is frequently the characteristic that makes them catalytically important. Metal-carbon bond breaking may formally be uni- or bimolecular. A unimolecular process involves either (a) mig- ration of a substituent from the alkyl group to the metal (a, 8, etc.. elimination) or (b) homolytic cleavage of metal-carbon bond. The pathways of higher molecularity result in disproportionation or for- mation of clusters. Much mechanistic information derives from studies of platinum alkyls, which are experimentally convenient because of stability to air, but the results probably apply equally well to early 100 transition metal alkyls. The most important pathways for the decom- position 0f transition metal alkyls will now be discussed. Beta-hydrogen elimination dominates the thermal decomposition of metal alkyls and may be represented by the equation as follows: LnM-O-b-Cl-bR :2 LwM-H +CHZ=CHR The migration of a a-bydrogen from carbon to the metal (M) with for- mation of an alkene probably proceeds through a concerted reaction. The olefin may remain within the coordination sphere of the metal. The best-studied example is the thermal decomposition of di-n-butyl- bis(triphenylphosphine)platinum(II).144 The products on thermolysis are n-butane, l-butene and a complex of platinum(O). The decomposition has been proposed to take place by intramolecular a-hydrogen elimi- nation process (Figure 38). Beta-hydrogen elimination is reversible, and indeed both metal- carbon bond-making and bond-breaking by this route are involved in many catalytic reactions such as olefin or acetylene hydrogenation, hydroformylation, hydroboration, hydrosilylation, isomerization and olefin oligomerization and polymerization.145 A e-hydrogen may also be abstracted from an aryl ligand,1°°'148 where evidence for the formation of a coordinated benzyne, CpZTi(06H4), was provided and 149'151 Finally, inter- confirmed by deuterium-labeling experiments. molecular abstraction of hydrogen is also possible in a-hydrogen elimination. 101 HISP\P /C4H9 / —-—P P Phsp C4H9 P113 —Pt\QH9 I .....12. HZC =CHCsz 9H9 Pep P10391372 P '17P“— Fit-H PH». N-BUTANE l "BUTENE Figure 38. Decomposition of di-n-butylbis(triphenylphosphine)- platinum(II). Alpha-hydrogen elimination involves migration of a hydrogen from theta-carbon to the metal with formation of a carbenic fragment which may remain coordinated to the metal. A substantial amount of infor- mation indicates this is also a significant decomposition pathway. 102 LNM—CHR’R” ._.._. LNM(H)(CRF1’) The above process is more likely to occur in complexes with less than 18 valence electrons since the number must increase by two; the rela- tionship to formal metal oxidation state is less clear. Alpha-hydrogen elimination is generally less well established than e-hydrogen elimi- nation. The best example of o-hydrogen elimination is neopentyl tantalum complex (Figure 39). The neopentylidene ligand forms by abstraction of a neopentyl a-proton by a neighboring neopentyl group in the sterically crowded penta(neopenty1) intermediate. Ta (O-izCMeslsClz +2 uo—boweSJTa(o-50\4e,1g’ + 2 L10 ‘ ’ ,H Ta(Ci-|20\4e3)s—'(Me3CG~lz)3Ta=-'C\ + CMe4 CM93 Figure 39. Formation of the tantalum neOpentylidene complex. 103 Reductive elimination is the reverse of oxidative addition and provides a route to cleavage of metal-carbon bonds. The process is shown below. Both the coordination number and the oxidation state x LNM: s LNM + X-Y Y of the metal are reduced by two. This mode of thermal decomposition plays a role in several organic syntheses where transition metal species are involved in the formation of carbon-carbon bonds. Decomposition by homolytic metal-carbon bond cleavage often gives a complex mixture of products, the composition of which is dependent on reaction conditions. This pathway formally involves oxidation of the ligand (R‘+R°) with corresponding decrease in the oxidation state of the metal. Other modes for thermal decomposition such as ligand hydrogen abstraction, binuclear elimination, etc.. have also been discussed recently.]°3'155 _ In addition to B-hydrogen elimination and reductive elimination, a different elimination which involves carbon-carbon bond cleavage is observed in the thermal decomposition of metallacycles. 1,4-Tetra- methylenebis(cyclopentadienyl)titanium(IV) has been reported to decom- pose to produce ethylene in good yieldv‘m']$9 (Figure 40). No reduc- tive elimination (formation of cyclobutane) was reported while B-hydro- gen elimination (formation of l-butene) occurred to only eight percent. 104 :>l'i ——+ H2C=CHZ +// \ + szT.' 92% 8% Figure 40. Decomposition of 1,4-tetramethy1enebis(cyclopentadieny1)- titanium(IV). It is believed that titanacyclopentane decomposes by reversible carbon- carbon bond cleavage to produce an intermediate bis(ethylene) complex. That is, when bis(cyclopentadienyl)titanium dichloride is reduced to titanocene in the presence of ethylene, a titanacyclopentane is formed. Carbonylation of the titanacycle with carbon monoxide yields cyclopen- tanone in 17% yield (Figure 41). CPZTi ” Git-1%» szT. 0:0 . CO 05210 Figure 41. Conversion of olefins to cyclopentanones by reaction CPzTiCb‘—’ 2e \\’P">7 with titanocene equivalents. 105 The thermal decomposition of Pt(IV) metallacyclopentanes was investigated by Puddephatt et_gll°° The decomposition of 1,4-tetra- methylenebis(dimethylphosphine)diiodoplatinum(IV) gives exclusively l-butene generated by'e-hydrogen elimination. It is interesting to note that no cyclobutane is formed in the decomposition of any platinum(IV) metallacyclopentanes. Puddephatt and coworkers believe the lack of cyclobutane production from a reductive elimination process is due to the higher activation energy needed to close the strained cyclobutane ring. Hhitesides et_gl}°1'1°3 reported the thermal decomposition of a series of platinum(II) metallacycles and compared rates of decomposi- tion to their corresponding alicyclic platinum complexes. It was found that the thermal decomposition of 1,4-tetramethylenebis(tri-n-buty1pho- 5phine)p1atinum(ll) dichloromethane and dibromomethane solutions generated both cyclobutane and butenes derived from the metallacyclic moiety and cyclopentane and pentenes formed by incorporation of a solvent-derived methylene group. Thermal decomposition of platina- cyclopentane in hydrocarbon or ethereal solutions yielded only butenes having the same number of carbon as the metallacyclic group (Figure 42). 106 ETHERS OR Maj +\IL‘+D 3031’ / W150 \O-tcb aj +\|L\+D +> +j+. ’ 2'2. Figure 42. Decomposition of 1,4-tetramethylenebis(tri-n-butyl- phosphine)p1 ati num(I I). 107 However, Grubbs, Miyashita and coworkers that thermolysis of phosphine nickelmetallacyclopentanes produced ethylene, cyclobutane and butenes and the coordination number of the nickel played an important role on the decomposition pathway. Three coordinated nickelacyclopentanes decompose by 3-hydrogen elimination to give 1-butene. Four coordinated nickelacyclopentanes decomposed by reductive elimination to give cyclobutane while five coordinated nickelacyclopentanes decomposed by reversible carbon-carbon bond cleavage to give ethylene. Figure 43 summarizes the effects of added phosphine on the decomposition modes of nickelacyclopentanes. . L L LNO .— L2N1::) 7: L900 1 I 1: fl 2 CH; =CI—b L: TRIALKYL OR ARYL PHOSPHINE. Figure 43. Decomposition of nickelacyclopentanes. 108 Thermal decomposition of metal alkyls often affords a complex mixture of products. It is not always possible to ascertain all the product-forming reactions. Although a number of thermal decomposition modes have been discussed and studied specifically,""3'155 no unified theory of decomposition has resulted. Knowledge of the detailed mechanisms of thermal decompositions is important both in the appli- cation of known catalytic reactions and in the development of new ones. In this research, new decomposition pathways, y- and a-hydrogen elimi- nation, are proposed and discussed for the thermal decomposition of transition metal alkyls. RESULTS AND DISCUSSIONS Alkyllithium, -magnesium and -aluminum reagents]°7"172 have been reported to react rapidly with transition metal salts, yielding complex mixtures of hydrocarbons which are believed to be derived from the thermal decomposition of intermediate transition metal alkyls. A number of mechanisms for the thermal decomposition of transition metal alkyls have been mentioned, but the detailed ones have not been well established. Similarly, the detailed mechanisms for the thermal decomposition of alkyl- and aryl-bis(cyclopentadieny1)titanium compou- nds are still uncertain. Razuvaev et al173 proposed a homolytic clea- vage of the Ti-C o-bond as the main pathway in the thermal decomposi- tion of szTiRz (R= alkyl and aryl) in various solvents. Dvorak et_el}49 suggested that benzene was formed by abstracting a hydrogen atom from another phenyl group with formation of a phenylenetitanium complex in the decomposition of CpZTi(C6H5)2. Teuben et_el}°1 found that the compounds, szTiR2 (R8 aryl and benzyl), decomposed with quantitative formation of R-H via intramolecular abstraction of a hydrogen atom either from a cyclopentadienyl ring or from the other coordinated group R. In our study, some unexpected hydrocarbons such as methane and ethylene from the reaction of dichlorobis(cyclopentadienyl)titanium and n-butyllithium were observed. The results outlined above prompted us to prepare dialkyltitanocene complexes and investigate their detailed mechanisms of thermal decomposition. 109 110 Preparation and Thermal Decomposition of Di-n-butyl-bis(cyclopentadi- egyl)titanium(IV) The preparation of di-n-butylbis(cyclopentadienyl)titanium (szTiBuz) can be achieved by the reaction of a suspension of dichlo- robis(cyclopentadieny1)titanium (Cp211012) with n-butyllighium in diethyl ether at -78°. The orange product, szTiBuz, was purified by column chromatography over alumina at -78° under argon by using n-pentane as eluent. Since its thermal instability precluded tradi- tional analytical procedures, it was characterized by some chemical reactions. Treatment of orange solution with hydrogen chloride gas in pentane gave CpZTiC12 and n-butane in molar ratio of 1:1.98; treatment with concentrated 02504 or gaseous 0C1 in n-pentane gave 1-deutero- butane; treatment with bromine in the same solvent gave szTiBrz and 1-bromobutane in molar ratio of 1:1.88. Di-n-butylbis(cyclopentadi- enyl)titanium decomposes at -40°, therefore it is stored at -78° or below. Hhitesides et_el}57 reported that thermal decomposition of szTiBuz at 250° by injecting samples directly into the gas chromato- graph gave the 1:1 mixture of n-butane and butenes. In this research, the thermal decomposition was carried out by stirring a toluene solu- tion of szTiBu2 at -60° to -780 for 5 h, then warming up to 60°. Interestingly, small amounts (“1%) of methane and ethylene were generated besides the major products of n-butane and butenes produced from initial B-hydrogen elimination followed by reductive elimination of n-butane. The production of light hydrocarbons other than B-hydro- gen elimination products was also observed in the thermal 111 decomposition of some early transition metal n-alkyls. The results are given in Table 13. The formation of methane and ethylene is apparently not consistent with the frequently discussed mechanisms of thermal decomposition for transition metal alkyls and will be discussed first. Table 13. Minor products from thermal decomposition of n-alkyl transition metals. Compounds Decomposition Products other than B-Hydrogen Elimination Products(%) .fEEL _f§flEL Other Hydrocarbons szTiPr2 1.9 1.1 C6 (5.2%) szTiBu2 0.2 0.6 03 (0.1%), 08(not examined) szTiAm2 0.8 0.6 C3 (0.9%), C4 (0.3%) 010 (not examined) CpZZrBuz trace trace szHfBu2 trace trace szNbBu2 trace trace szTaBu2 0.6 0.2 Grubbs and Miyashita174 have reported that the carbenoid intermediate is formed and provides a route to the formation of methane in the thermal decomposition of nickel and titanium metallacyclohexanes. The 175 to produce metal-carbene complexes have also been found to dimerize ethylene. Therefore, di-n-alkylbis(cyclopentadienyl)titanium was allowed to decompose in the presence of cyclohexene. Indeed, about 112 1-3% yields of norcarane were isolated. The results are given in Table 14. The generation of metal carbene is not certain in these themal decompositions, but the methane and ethylene may be formed by metal-carbene complexes according to the following equations. H frironoertm Absmcnow $ CH4 LNM=CHZH DIHERIZATION 4 CHzr-CHZ Table 14. Metal-carbenes trapped by cyclohexene. Compounds Product---Norcarane (yield) szTiPr2 2.93% szTiBu2 - 1.51% szTiAm2 1.10% 113 However, the production of ethylene in the decomposition of CpZTiBu2 possibly results from s-carbon-carbon bond cleavage of titanacyclopentane formed by c-hydrogen elimination with reductive elimination of one molecule of n-butane as shown below (Figure 44). S-Hvoeoeeu (%CWCHZCH” ELIMINATION /CF°‘SIHZ szTI i. T1 \O-bGJeo-bCHg‘N‘BWE 0° \CHZ/CI-b. t1 , (Sheff? " '1- £2 (:EIZ:==(:I12 4,...— Figure 44. Possible scheme for the formation of ethylene via 6-hydrogen elimination in the decomposition of szTiBuZ. Delta-elimination as well as y-elimination are not well established in organometallic chemistry. In principle, they are similar to ‘53 involving y-elimination are B-elimination. Two good examples found in organoaluminum and silicon compounds and illustrated in Figure 45. Diisobutyl(3-ethoxypropyl)aluminum, derived from diiso- butylaluminum hydride and allyl ethyl ether, decompose to give cyclo- propane and diisobutylaluminum ethoxide. Similarly, diethylmethyl- 114 (chloropropyl)si1ane thermally decomposes to give cyclopropane and diethylmethylsilyl chloride. 0'12 , / CH (I""BU)2A|KP m—A-PG-BukAIOEt + / \2 ASH: C Hz—CHZ Et C /”2\ A .4 gig—.444... . at 01/ Cit-Cit Figure 45. Scheme for y-elimination in organoaluminum and silicon compounds. 17° prepared nickelacyclobutane Recently, Grubbs and Miyashita compounds from di-neopentylbis(triphenylphosphine)nickel. The forma- tion of nickelacyclobutane involves y-hydrogen elimination with loss of one molecule of neopentane. The nickelacyclobutane complex contai- ning odd number of carbons decomposes by cleavage of an a,B-carbon- carbon bond to produce carbenoid complexes which dimerize to give ethylene complexes or react with olefins to produce substituted 115 cyclopropanes. The scheme for the decomposition of dineopentylbis- (triphenylphosphine)nickel is illustrated in Figure 46. To further examine y- and c-hydrogen elimination, the complexes of dineopentyl- and dineohexylbis(cyclopentadienyl)titanium were prepared and their thermal decompositions were studied. (1.) (F1392 Ni (C2H4) + I ice-13 + Q-tchHBH O-lzCHs CH3 Figure 46. Decomposition of di-neopentylbis(triphenylphosphine)- nickel(II). 116 Preparation and Thermal Decomposition of Dineopentyl- and Dineohegylbis(cyclopentadienyl)titanium Dineopentylbis(cyclopentadienyl)titanium was prepared by the reaction of CpZTiC12 and two equivalents of neopentyllithium in diethyl ether or pentane at ~78°. After the solvent was removed, the residue was chromatographed on an alumina column at -78° under argon using n-pentane as eluent. The product was characterized by treatment of the orange pentane solution with hydrogen chloride gas and bromine. Similarly, the dineohexyl compound was prepared by the reaction of CpZTiClz and two equivalents of neohexyllithium and characterized by the same methods as those for dineopentylbis(cyclo- pentadienyl)titanium. Transition metal alkyls in which the alkyl contains e-hydrogen atoms often decompose fairly readily by a-hydrogen elimination as seen in the decomposition of dibutylbis(cyclopentadienyl)titanium. This decomposition pathway is blocked if no a-hydrogen atoms are available. For example, the high stability associated with methyl, neopentyl, benzyl and aryl derivatives of transition metals is thus attributable to the absence of B-hydrogens which can be eliminated as metal hydride. The thermal decomposition of szTiR2 (R- neohexyl) gave mostly neohexane and small amounts (1.3%) of ethylene and isobutylene. The neohexane is probably produced from initial a-hydro- gen elimination followed by reductive elimination of neohexane or from intramolecular abstraction of a hydrogen atom from a cyclopenta- dienyl ring formulated as in the following equation. 117 CPzTI'R * it «CIOHBTI’, + 2 RH Thermal decomposition of szTi(CH3)2 has shown that some methane arises from an intramolecular a-hydrogen elimination and some abstrac- tion of hydrogen atoms from the cyclopentadienyl ring."9 These two processes in the decomposition of CpZTiR2 (Ra neohexyl) could be distinguished by experiments with a deuterated compound such as (0505)2TiR2 (R= neohexyl), but were not examined in this research. He are most interested in the formation of ethylene and isobutylene. Norcarane was also produced with 1.84% yield when thermal decomposition of dineohexylbis(cyclopentadienyl)titanium was carried out in the presence of cyclohexane. The production of a carbenoid complex proba- 174 of intermedi- b1 y resulted from the a,B-carbon-carbon bond cleavage ate dimethylsubstituted metallacyclopentane, which is formed by initial dehydrogen elimination followed by reductive elimination of neohexane. Therefore, the formation of ethylene can occur either by dimerization of metal-carbene complexes or by’Bgy-carbon-carbon bond cleavage of resulting intermediate substituted metallacyclopentane. Similarly, the isobutylene also probably resulted from s ,y -carbon-carbon bond cleavage of intermediate substituted metallacyclopentane or from further cl,5-carbon-carbon bond cleavage of intermediate carbenoid metallacyclobutane. The organometallic product of thermal decomposi- tion of szTiR2 (R8 neohexyl) was possibly an unstable compound, "titanocene“, which gave a dimeric titanium hydride at decomposition temperature. The proposed mechanism is shown in Figure 47 (Reaction b). 118 Similarly, the neopentyl complex, szTiR2 (R- neopentyl), decom- poses to give small amount of isobutylene (1.2%) possibly produced from 8,8-carbon-carbon bond cleavage of intermediate dimethyl-substi- tuted metallacyclobutane which formed by y-hydrogen elimination with reductive elimination of neopentane. A small amount of ethylene (0.98%) was also found besides the major product, neopentane. in the decomposition of szTiR2 (R= neopentyl). The formation of ethylene probably resulted from the dimerization of carbenoid complexes which are produced from a,B-carbon-carbon bond cleavage of intermediate metallacyclobutane and have been trapped to give norcarane (1.71%), when the dineopentyl complex was allowed to decompose in the presence of cyclohexene. A 1°C NMR signal at 359.07 ppm was observed for dineopentylbis(cyclopentadienyl)titanium in toluene at -50°. This signal is probably due to resonance, characteristic of carbene-type a-carbon atom bound to transition metal absorbing at the low field. The mechanism for the formation of ethylene and isobutylene in the thermal decomposition of dineopentyl titanium complex is proposed as shown in Figure 47 (Reaction c). ' At present, the production of isobutylene in the thermal decom- position of szTiR2 (R= neohexyl and neopentyl) can only be increased to 2.3% by stirring a toluene solution of CpZTiR2 at -60° to -78° for a week, then allowing it to decompose at 80°. No significant amount of ethylene and isobutylene was observed when a toluene solution of szTiR2 (Rsneohexyl and neopentyl) was irradiated with UV light at -78°. Hhen the thermal decomposition of CpZTiR2 was carried out in toluene-d8, no significant amount of deuterium was found in the 119 products, ethylene and isobutylene, and so participation of the solvent in the decomposition was unlikely. Similarly, a preliminary examination showed that the reaction of some metallocene chlorides with neohexyllithium in toluene yielded methane, ethylene and isobu- tylene. The results are given in Table 15. Possibly, the methane is formed by hydrogen abstraction of metal-carbene complexes and the formation of ethylene and isobutylene resulted from the same pathway as in the thermal decomposition of dineohexyl titanium complex. The C4 compounds (1-2%) such as butane and butenes are probably formed by dimerization of ethylene.157 In an experiment, the amount of C4 was doubled, when szTiR2 (R= neohexyl) decomposed in an atmosphere of ethylene and argon. An attempt to isolate intermediate metallacycles was not successful due to their instability and low yields. Although there have been several studies of the thermal decom- position of transition metal alkyl complexes, there still is uncerta- inty concerning the mechanism of the thermolysis. From this research, it appears that the mechanisms proposed (Figure 47), y- and a-hydrogen eliminations, are one type of process for thermal decomposition of transition metal alkyls and at least can serve as a starting point for further experiments and discussions. This new mechanism can also provide models for a number of homogeneous and heterogeneous reactions of hydrocarbons. Foams coruwmcmcu mo cowuvmoaeouou pussoxu mo mamvcmguos 1 nee/:3} fouwzovg mo" :2... +MM/uflumr 4 22... + WONNIU Pox +3 Imoxu 4....uxu..11.um.._ fi.mvfiux\ur,% ~60? .mpzxpm 38: vomoaoga on» com osogum ugh .Ne oesmpm powfo/ a) 22.. Q 552.23 \ 58551» #01 ea mnemfo 1425981 , z 292sz v. ._ as 58891..” womb. ~28 ..a $3.3. “.55... new} G 2922.sz ,swmxumxupTTAw nWfiumrnmuInva 121 .888: 8 884 cast 88 8881 88 8883—88 :8 88888888 88888 88m 88 88: 8888888858888 8 N8.N 88.N 8N.8 88.N 88.8 8_m.88 N8=N88 NN.N 88.N 88.8 88.N 8N.“ 888.8. N888N88 88.. 84.N 88.8 8N._ 84.. 888.88 N88848N88 88.. 88.8 84.8 8N.” 88.N 888.88 8>N88 88.N 88.. 88.8 88.N 88.8 888.88 4N888N88 NN.. 88.. N_.8 8N._ 88.8 8N8.88 N888N88 «:8, N a m 8 N a N4.8.848 : 8 8 z 8 =8 Apaxeeeez 488 Amy 88888885888 8888885148 8888 88:88 8888888858888 8828888 88 88888888 Apossv 88x8—8588 Faxogooz ’l MIL-11)) I) A 88888885888 8888885818 8888 88888 mmxmpasou 88885 pxxogooc 88 88888888 8888888858888 .m— 88888 EXPERIMENTAL All reactions and manipulations were performed under welding- grade argon purified before use by a BASF deoxygenation catalyst and molecular sieves (Linde 4A). All solvents were purified as described in the EXPERIMENTAL section of Part I. Neopentyl chloride and neohe- xanol (t-amyl carbinol) were obtained from ICN pharmaceuticals, Inc. Toluene-d8 (>99%) was purchased from Aldrich Chemical 00. Analytical GLC was performed on a Varian series 1400 F10 chroma- tograph with a 20 ft x 1/8 in Durapak column or a 13 ft Paraffin wax/ 5% AgNO3 on A1203 column. Product yields were determined by response relative to an internal standard. Mass spectra were recorded on a Hitachi Perkin-Elmer RMU-6 mass spectrometer and Finnigan 4000 GC mass spectrometer. 13C NMR spectra were measured on a Varian CFT-20 spectrometer with TMSas internal standard. Preparation of Alkyllithium Reagents Except for the solution of n-butyllithium in hexane which is commercially available, the solutions of n-alkyllithium reagents in n-pentane or diethyl ether were prepared from the corresponding chlorides or bromides by previously reported procedures.178 Neopentyllithium in n-pentane was prepared from neopentyl chlo- ride. A mixture of 30 g neopentyl chloride and a two-fold excess of finely chopped lithium wire in 500 mL n-pentane was stirred and refluxed under argon for 10 days. The excess Li metal and LiCl were removed by filtration and the ne0pentyllithium was isolated from the 122 123 filtrate by cooling and reducing the volume in vacuo and yielded 90% of white, crystalline LiCHZCMe3. The white solid was then dissolved in freshly distilled n-pentane, diethyl ether or toluene ane store at -10°. Neohexyllithium (LiCHZCMezEt) was prepared from neohexyl bromide which was obtained from the reaction of neohexyl alcohol and 18] A sample of 16 g (0.16 mole) dry neohexyl alcohol P(n-C4H9)3Br2. (t-amylcarbinol) was mixed with 43 mL (0.17 mole) of freshly distilled tri-n-butylphosphine in 85 mL of dry DMF in an argon atmosphere. About 8 mL of bromine was added slowly while the reaction flask was cooled in an ice bath. After being stirred for 6 h, all volatile material was then removed by vacuum distillation into a receiver cooled in a dry ice bath. Cold water was added and neohexyl bromide was separated. Neohexyl bromide (yield 90%) was then dried and distilled at 55°l40 torr. A 12 g neohexylbromide and a two-fold excess of Li wire were placed in 150 mL n-pentane under argon and refluxed for 12 days. The mixture was filtered and the neohexylli- thium (>89%) was isolated from the filtrate. It was dissolved in n-pentane, diethyl ether or toluene and stored at -10°. Preparation of Di-n-butylbis(cyclopentadienyl)titanium A sample of 2 g (8 mmol) szTi012 was suspended and stirred in 25 mL diethyl ether at -78°. An ethereal solution of n-butyllithium (16 mmol) was added slowly by a syringe. The reaction mixture was stirred at -78° for 35 h. During this period, the mixture turned dark red. About 100 0L methanol was then injected to destroy any 124 unreacted n-butyllithium. The solvent was removed at -78° to give orange-red residue. The residue was chromatographed on an alumina column which has been thoroughly dried and flushed with argon. This column was provided with a cooling jacket, and maintained at -78°. The column was eluted with n-pentane and an orange band was collected as it eluted. The orange pentane solution was then concentrated at -78° in vacuo and the product was stored below -78°. Di-n-butylbis(cyclopentadienyl)titanium was characterized by its chemical reactions. Treatment of orange pentane solution with anhyd- rous HCl gas yielded n-butane and szTiClz in a molar ratio of 1.98 : 1. Similar treatment with 001 gas or concentrated 02504 gave l-deuterobutane with isotopic yield >99%. Hhen the orange solution was treated with bromine, the dark red CpZTiClz and 1-bromobutane were formed. GLC analysis of l-bromobutane and weighing szTiBrz showed that the molar ratio of szTiBrz to l-bromobutane was 1 to 1.88. Thermal Decomposition of Di-n-albylbis(cyclppentadienyl)titanium A pentane or toluene solution of szTiBu2 was stirred at -60° to -78° for 5 h, then warmed up to 60°. The gases produced were injected into the CC for analysis, cyclopropane was the internal standard. Subsequent analyses gave the same results, indicating complete decom- position. After complete decomposition, the products produced were collected by distillation into a liquid nitrogen trap. On being warmed to room temperature, the product gases were analyzed by GLC in the vapor phase as well as the liquid phase and their composition was determined. 125 Other thermal decompositions of di-n-alkylbis(cyc1opentadienyl)- titanium were carried out as follows: A sample of szTi012 was suspended in n-pentane or diethyl ether at -78°. Two equivalents of alkyllithium in n-pentane or diethyl ether were added slowly. After the reaction mixture was stirred at -78° for 36 h, small amount of methanol was injected into the flask to destroy the excess alkyllithium reagent. The solvent was removed in vacuo at -78° and toluene was added. After being stirred at -78° for 5 h, the toluene solution was warmed to 60°. The products were analyzed by GLC as above. The composition of products was unaffected by chromatographing the solu- tion through alumina. Preparation and Decomposition of Di-n-neohexyl and Di-neopentylbis- (eyclopentadienyl)titanium A sample of l g (4 mmol) szTiClz was reacted with neohexyl- or neopentyllithium (8 mmol) in 30 mL diethyl ether or n-pentane at -78° for 2 days. About 100 uL of methanol was injected to destroy any unreacted lithium reagent. Removal of the solvent at -78° left an orange sludge. The product was chromatographed by a procedure ana- logous to that used for di-n-butylbis(cyclopentadienyl)titanium with n-pentane as the eluting solvent. A clear orange pentane solution was obtained and concentrated. Treatment of the orange pentane solution with anhydrous HCl gave neohexane or neopentane and CpZTiCl2 with the molar ratio of 1.93 to 1. Similar treatment with bromine yielded neohexyl bromide 126 and CpZTiBrZ. The molar ratio of neohexyl bromide to CpZTiBrz was 1.90 to l. The characterization of di-neopentylbis(cyclopentadienyl)- titanium was the same as that used for CpZTiR2 (Rsneohexyl). The procedures for the thermal decompositions at 80° are similar to those used for CpZTiR2 (R-n-butyl). Hhen the decomposition was carried out in deuterated solvent toluene-d8, no significant amounts of deuterated products was found. Hhen a toluene solution of CpZTiR2 (R=neohexyl) (1 mmol) and 20 mL of ethylene was stirred at -78° for 5 h, then decomposition at 800 gave 2.5% butenes and butane. Metal-Carbene Trapping Experiment Typically, a toluene solution of szTiBuz (1.5 mmol) and 1 mL of freshly distilled cyclohexene were stirred at -78° for 3 h, then warmed to 60°. After being stirred at 60° for 36 h, the product was collected by distillation into a liquid nitrogen trap. On being warmed to room temperature, the product was analyzed by GLC on a 32 ft x 1/8 in 10% Carbowax 20M/Chromosorb H column and a 13 ft xl/8 in Parafin wax 15% AgNO3 on A1203 column, compared with authentic sample and the yield of norcarane was 1.51%. Preparation of Metallocene Chlorides 180 181-183 Vanadocene monochloride and the metallocene dichlorides of Nb, Ta, Mo and H were prepared as previously reported. 127 The new compound of trichlorobis(cyclopentadienyl)tantalum (szTaC13) was prepared and characterized as follows: A sample of 5.4 g (15 mmol) tantalum pentachloride (TaCls) was reacted with 15 mmol of Mg(C5H5)2 in 100 mL of toluene. After being stirred at room temperature for 24 h, the toluene was removed 1p_ypppp and the yellow residue remained. A yellow crystalline compound, CpZTaC13, was obtained by sublimation of the yellow residue at 260°/0.01 mmHg. Characterization was made by IR spectroscopy that showed a band pattern characteristic of biscyclopentadienyl complexes’g (Figure 48) and by mass spectrometry. The characteristic IR peaks are 1440 (s), 1340 (m), 1125 (vw), 1015 (m) and 855 (s) cm". The principal ions in mass spectra are szTaCl3 (M1, m/e 417-421), szTa012 (m/e 381-385) and CpTaCl2 (m/e 316-320). Preparation and Thermal Decomposition of Neohexyl Metallocene Derivatives A sample of metallocene chloride was treated with the required amount of newhexyllithium reagent in diethyl ether or n-pentane at -78° for 2 days. Methanol (100 0L) was injected to destroy any unreacted lithium reagent. The solvent was removed at -78° and toluene was then added to dissolve the residue. After being stirred at -78° for 5 h, the toluene solution was warmed to 80°. The produce gases were injected into the CC for analysis. Subsequent analyses gave the same results, indicating complete decomposition. Some results are given in Table 15. The decomposition of CpZZrRz, szHfR2 and szNbR2 (R-neohexyl) only 128 gave trace amounts (<0.01%) of ethylene and isobutylene. 129 o~ o - ~ot- .- o~n --oo-coo-uno-IIo'-00-.°".'--u-woooo.-.. -.-...~ao-qoo-nocnado-uum . ~‘----O---~O-o--ooo -.a.w-oo ...-.0-3- a. I. a... . .... .... .....o........-...........—....--..... ' ... ... .o ..... ... o-o -. . e .....-..8.. . . -I.'IOIID-Io'o-II ....I. ..I.I.OO.' - 4...... - u. ......0-000-0- ”*R”..Cm.*“Hm-O-—.m.w:‘ a a - 4"IODOOI:C'OOODOO ...-......o—oc onto-eon c.-o-.~o-...I. o O...".'. .IIAletll I-o‘noooo-oooOo-c-A 1 .<;' u ‘OUOOOQOOIIOU¢<:’.Q..~.... 0.. ......O...’......-O...Il.—.....”...o¢ - .e . ...-.0... ...-9....- .I-ou—---o...-o..1 n-u - p ...... (:’ oo- o... A dot-'Oooooluco—ou- .mOOOOQI... ..... . ... a!) II v .......<=’ ......aeo.~....-.. 800 ,...:.......V. .......I.......N ....-.....-- ..... rogue-......lifltloy ouc-o...a..o¢~.e ....ueooouo. ...-oxuovo'o-ae: . a . coo-ooc voo-noo-ooQ-ooouuooo aII.....O..II....D..IOI2......IOOCIuooo - v-0 0 no. I . o o o e o - O v o ....o-. ......... o...- ... ...-.0... ....l..Il-'a-I o .w :--ooo.....o o-oaoloo-ououo- o 8.. 00.0.0... ...-.8... .-...... e.|..o~.-n~-.-venoon-corona. ..... ......-........ ,,.__,.... ......oganu-uunnunnongoncuugeuorg-no- ......Q-OOI-ooI-Ooono-o- A ‘0... on... I...‘ on. can... ------------ --- ... ...... g .. .... ' . ......a-ae. CrIU-O1..‘I...I-.-.-I-D.-IO-IIIO-Ic-O o ‘-Q.oo...---.-.--- . a neg-ao..-bees-coounooonoAuO-ooonoooo ' - .- ooooo . .~ 0 . u... -e.lo.l.oilu'oI-ogoo--'-~I-..ool:..nn..- -;.......,.. ........ A p ...... ...--I...-.c.o.ln.~o-..l-noo.-....n- ...... . u...-'o..o ....ap. coco-00......usalu-.-.-n..-oo--e~n-a. .- . . . 1000 WAVE NUMBER (CM ') I g . . .....o'a.-.'.. .. u -......eau--'.o.-'-..-g..e 8.4.8.8...o... O I-u-OO.I”O”.~”.~OOI;OOO'l-Oo .. - a... I.."I..|.. coco-8.80.0...-o.oz-.-opoo--gc ....8..t........- I O u-~...o.u-o IIIJOIII.OOOPI.......IOII..O ...."....C.I..t~...~...DIC.O ......o.-eo-I no. uOla-CI...pow-......o-ouyovuo on-...... ....‘I.- ......I...I~II.I...' .I-OO—I ...-.lcollooon..lnn v v v ......u. {Ion-'O-I|.I0I.'-Ol~' . .A... , . A. . o - . -o-o-oooov-.noo.o-........oeuu-g-uoo.~-. ...--..”... cocoo-oowo-om-m ..o ..e a. .0 A 0.90-.4- 'l "‘.Io-nIA.CO.-Ovr.-oou.on .........-.8-... . A... I ........ Q a g o n-A-oooO-o..6-o--ooo-.OO.~¢.~ .4. 00". .. o aaaaaaa o A ~l o 0 a --o.-onocv¢o- .......-..-- ......g...—- 8.. u 0 u I a... so.— -0- ...-098-8- u-oooooaoo-cuoguooo..8...---- ... .5 . .. equo-o-ooopoo-on..........e.-o.n.—..—-.———o.....—...9oo.o--—-......—-—-. .-.—.-. g I. '8': 1'3? l......g ...... :u. . ~- 9 . ..., ..... o ..' o-- o- 'oo-~;o~.-o-o~-oo--;o----I-v. . . . up on o..>..... ......a..l...- - . ... .ltooouuno. -0...“ .-- ...-o 8...... (£2 I : .A g .. t . ...... ...:. ..A .... - .|.. uo-aa.,-.-ooo::.A:—. -.:o a... . -o-. o g, 4 . .. . . .. p ....... . .... . ...-4 - Io'Io-o'ooooo ..l..... n '(3 O > I ..ptr.‘ -a- o -;- | at. - .o-oeo 00-. n I g . .. ...... .c CYI- o- oooooooooo n K e p ...... up. - p 8 agree ‘0 u w to ?" . ..g. a... .- U '1' TI: '73 IX: ' '... ' 7: 72:37.. 313’: "3 1.3.3123: ' :: ‘0'-..’....:... ~00- eoA-oo-ooo :9.- no... .“Aum—OOQOM --.—......- D ' ...t. . ... ..A. .........-. o 'l ........ toenou-ecA88...88-.. ......... . ........ .- ' ...e. . , . .... ... ....0- . g .. {oto-o-o-..-.ooooua.'..... ..-... 8 J I ._,, ., . ,,:,, I. A .. .. g ....o.. ...-....'!..'.'-...'..... ..'.. 1...,1-8 .0 y . no ... . canoe ...oee-o-cv-o-uoe..... ....n..........-........... . '. ,'.". ....-. .: :. ...“: ----- o-e ooooooo .-. '8... ......ggg'....p.. ... a. ‘ A . p , o .'. . ...un-Io- o. 9.. Izooo-oo-oo usage-.....8. .. ,.... .- ......... 0.. o o 0 --0-’ o 0 can. ..I.... o. n - ' .- .. J~ .....°.:;;::.':.' .. ..... .. .;:;.;; ....:;;;:..-.:.:...° ':::.;::L;; "' ' ""° ' . -' ' ' ....A- will re - ... ...-a l'l-Il‘.‘ ooooo ..OI‘-.l‘-l"cooo n-II.O-...-OC av... . . no... .4 o. ...... -o-aa.-.. o... ooc.z-.o-pooo-p-oo-Aoa-oio.-.,........ one . . ..... II o .0 .o A..... u -I-. I'aoo tee-Aoo-wooueeo-neo .8.......... - . 4' o A In ..-.I.-a- a ..... I a ‘:'0.l'.'.ll'¢. n‘n-<:' ........ A ~ 88.00... loco... ....TE"~... I v - 2000 lltantalum (CpZTaC13). REMARKS leny vclooentad J I: l .!.. 1 O I “)-AT°. . I. o. II'uA-n‘r ”1.0: on --g.... . . I . loo-a t a I a II. ’00 g 5.. "":"::"°"':" A O:I.Oleo':l.n.ooo..o.u goo 1 .'. I' I 2212.. "iI.'IIZZL'(L§I.I...l'...l.:l..1.1...".iln.IZZZLI..ZZITIT1:IIZ.II... '3"""" '.'?°f'2. .. . '72 ."IT IITZFTI'II'T'I'ITT'....,ITI...... .......I ......'..'."."..'.If'..".‘ jyurr):ln'gvu .N :3-unrmhmnxmrmxflmrmx 4 3:2 ;--:-|.t-: o'- . I --! . --’-4-0: 0! tDOOI-O non-zn- go Ion-Ao-no ole-00:0... eon-o» I u”. out»... .0'..I so. u so. nooogluooihcl- .00....“ .0 .‘. .e ::.- 00' - a o .n... nnnnn - I... coo -.-g on. ecu-Opoou .... a a. I. . .‘l:lneoI gel :09. I ...A.. A on: w - -- nAaoeo’ooao::~-o 80.8.8... 8. t .I A - 8.. o 4 .. .‘.... ...-... ..... ,‘ .'_ x . .-0::. .:.8..:.::. .::':_T:: ..,,',_. ...._._:_. 8!;;.-:::.opooo::IgnoA-.:.:.:'. t...g.o..tlo.:::7.o:::f::o:::: . . .rf::::.... :T.::Tftftttjftttt. 4’ Imlfi’l .7..§.'T'| .."'. . ...... I. I)...” ........ ...... ........ ‘ . ... .A. .- .-. o n... a ., Ion I . . 1' o 4 uI.. ... l- - A I... no . o - --ooeo-~u ......... a. . """IT‘L‘I' f§'°'7;'7'31"21{37" "° ......i?"°?.’°?‘.:??."..’?2.’f 1. ' ::':'.':. .';:ft':. :1. ..... .I?::.:’.' ; :"':;;'.'::::: o .. ...... g I g '. . ...... o---.--..A ... . CD J gnaw: Itxz'4' .U. Jhfijh E'HLJ: : a i :Lfl:tr:r'*;f:..rr '"'.3;;h:128332 - o. Aooo-Ae I o--4--o.bo- ope-0‘ . :::::::T:I:tf‘f-OOIDOT-oooo. .: :Jr'wrtg"p:4: ,'“ """ .... . .:x. : . ZJL.!L :3”.:NL. n: ....... :an:.uxurtxxru ”EVFr'T17337F‘)..N ... :xEEkaxjmKEE J: gw'u(:”n:nrzxn.xt;n' ::;:'.L o- o'oooo DODC'IGIO ...-....- ..’.'o..¢ - A:‘:::;";:.;1t'::;:::I:;.' 33 33.23). ,.i;i.:Ii; 32.035139: Ill? E'-17"E'="==~‘ 1 I. ........ .. ':;!::.’ .‘ 11.: t' .: .f.: I 1: '.I ' ';I;'::. ::.I - A p- a: An--.o a o ---|o I I o ape....o-- . L....'1...1...Ll. f}..;;;.‘.;.......“.9.;:-L;.:;;:::9" I. . a . N . p . .... 1!..“I. .: . (..1....:.....l.:.'!...:, .I .2 '1, 0232‘.zzr:1“'1“;h':". ....H.“. .:::':‘etta::.o:::. .::. .;~ 0 00...: ai....:o.:o:...-:.ooei 1... ...” ' .. .1...,... ..... , ... ., . (CM '1 mmm-}MJOLm_mqn-dmuuu_flw CONCENTRAIION -_ - CELL PATH --- nrrCRrHCE - -8--.- Infrared Spectrum of Trichlorobis(c Figure 48. REFERENCES 10. 11. 12. i3. 14. 15. REFERENCES H. H. Flood and B. S. Lee, Sci. Amer., £39§(Ju1y). 212, 94. R. H. Grubbs, Dept. of Chemistry, Caiifornia Institute of Technolo- gy. Private Communication. J. C. Baiiar, Jr.. Cat. Rev.-Sci. Eng., 1313, 19, 17. A. Bjorkman, Chem. Eng. News, 12z3£238t~l31' p.5. R. B. Merrifieid, J. Amer. Chem. Soc.. 1863, §§, 2149. H. 0. Haag and D. D. Hhitehurst. Second North American Meeting_gj. the Catalxsis Soc.. Houston, Tex.. l311° R. H. Grubbs and L. C. Kroii, Abstracts of the 162nd ACS National fleeting, Washington, D.C.. Sept. 12-17, ng1, Poiymer Paper 68. C. H. Brubaker, Jr.. W. D. Bonds. Jr.. and E. S. Chandrasekaran, Abstracts of the 166th ACS Meeting, Symposium on Homogeneous Cataixsis, Chicago, 111., Aug. 27-31. lgzg. Paper 31. R. H. Grubbs, C. Gibbons. L. C. Kroii, H. D. Bonds. Jr., and C. H. Brubaker. Jr., J. Amer. Chem. Soc., 1223, gg, 2373. H. D. Bonds. Jr.. C. H. Brubaker, Jr., E. S. Chandrasekaran. C. Gibbons, R. H. Grubbs and L. C. Kroll. J. Amer. Che. Soc.. ng3 21, 2128. R. H. Grubbs. C. P. Lau, R. Cukier and C. H. Brubaker, Jr., J. Amer. Chem. Soc., lgzz.‘gg, 4517. R. H. Grubbs and L. C. Kroli, J. Amer. Chem. Soc.. 1211,‘2§, 3062. R. H. Grubbs. Chemtech, Aug.. ngz. 512; and references therein. E. M. Sweet, Ph.D. Thesis, Michigan State University, ngL. T. M. Fyiesand and C. C. Leznoff. Can J. Chem.. lgzg,‘§fi, 935. 130 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 131 D. C. Neckers, D. A. Kooistra and G. H. Green, J. Amer. Chem. Soc., 1972, 94, 9285. m— H. 0. Haag and D. D. Hhitehurst. Be1gian Patent, 1968,721,686. P. Legzdins, G. L. Rempe1 and G. Hi1kinson. J. Chem. Soc.. Chem. Comm., 1969, 825. —W C. U. Pittman, Jr., Chem. Tech., 1971, 1, 416. m C. U. Pittman, Jr.. Chem. Abstr.. 1971, 15, 141190. C. U. Pittman, Jr., P. L. Grube, 0. E. Ayers, S. P. McManus, M. D. Rausch and G. A. Moser, J. Polym. Sci.._Po1ym. Chem. Ed., 1972, 10, 379. M— N. Fujita and K. Sonogashira. J. Polym. Sci.1 Po1ym. Chem. Ed., 1974, 12, 2845. Nit/Vi"- F. R. Hart1ey and P. N. Vezey, Adv. 0rganometa1. Chem., 1977. 1_, 189; and references therein. C. H. Brubaker, Jr., Encyc1. Polym. Sci. Techno1.. Suppl. No. 1, 1376, 116; and references therein. C. H. Brubaker, Jr.. Cataiysis in Organic Synthesis, Academic Press, N. Y., 1378, p25. C. P. Lau, Ph.D. Thesis, Michigan State University, 1977. E. S. Chandrasekaran, Ph.D. Thesis, Michigan State University, 1375. E. S. Chandrasekaran, R. H. Grubbs and C. H. Brubaker. Jr., J. Organometa1. Chem., 1976, 129, 49. J. M. Paushkin. T. P. Vyshnyakova, T. A. Soko1inkaya. L. M. Burova, A. D. Gornstein and I. S. Panidi, Abstr. ORPL 82, 162nd Amer. Chem. Soc. Nationa1 Meeting, Washington. 1971. W 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 132 G. Gubitosa, M. Bo1dt and H. H. Brintzinger, J. Amer. Chem. Soc.. m, 39;, 5174. (a). G. Gubitosa and H. H. Brintainger, J. Organometa1. Chem.. 1371. 149, 187. (b). F. R. H. P. Hi1d, G. Gubitosa and H. H. Brintzinger, J. 0rganometa1. Chem., 1378, 158, 73. C. P. Lau and C. H. Brubaker, Jr.. Petro1eum Preprints, Nat. AC5 Meeting, Chicago, 1371, p1184. J. P. C-11man, L. S. Hegedus, M. P. Cooke, J. R. Norton, G. Do1cetti and D. N. Marquardt, J. Amer. Chem. Soc.. 1312. 24, 1789. A gift obtained from The Dow Chemica1 Co.. Mid1and, Michigan. G. Hi1kinson and F. A. Cotton, Progress in Inorg. Chem.. 1222. 1, 1. F. A. Cotton and G. Hi1kinson, “Advanced Inorganic Chemistry", 3rd Ed.. Interscience, New York, N. Y.. 1212, Chapter 26. E. 0. Fischer and A. Treiber. Chem. Ber., 1261,‘2g, 2193. K. H. Pepper, H. M. Pais1ey and M. A. Young, J. Chem. Soc.. 1331, 4097. ‘ T. M. O'Brien, M. S. Thesis. Michigan State University, 1276. J. G. S. Lee and C. H. Brubaker, Jr., J. Organometa1. Chem.. 1217, 135, 115. J. G. S. Lee, Ph. D. Thesis. Michigan State University, 1911, P. Tenaut. G. Tainturier and 8. Gautheron, J. 0rganometa1. Chem.. 197;, 121, C35. R. H. Grubbs and S. C. H. Su. 0rganometa1. Polym.. Ed. C. E. Carra- hen, J.E. Sheats and C.U. Pittman. Academic Press, N.Y., 1218, p129. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 133 . C. H. Su, Ph. D. Thesis, Michigan State University,,13;§. . Samue1, R. Ferner and M. Bigorgne. Inorg. Chem...1333, 12, 881. . MasToweky, Jr. and K. Nakamoto, Appl. Spectrosc., 1211, 22, 187. . Davison and S. S. Hreford. J. Amer. Chem. Soc..‘1313, 22, 3017. ca>rnmm . E. Bercaw. R. H. Marvich. L. G. Be11 and H. H. Brintzinger, J. Amer. Chem. Soc.. 1263’ 22, 1219; and references therein. P. C. Hai1es, H. Heigo1d and A. P. Be11. J. 0rganometa1. Chem., 1972, 43, C32. m— R. Hiatt, Oxidation, Techniques and,Apincation in Organic Synthe- 212, Ed.. R. L. Augustine, Dekker, N. Y. 1971, V01. 2, pp113. R. A. She1don and J. A. van Doorn, J. Cata1., 1973, 21, 427; and references therein. 0. H. Hart and J. Schwartz, J. Amer. Chem. Soc., 1874, 22, 8115. C. A. Berte1o and J. Schwartz. J. Amer. Chem. Soc., 1375, 21, 228. J. Schwartz and J. A. Labinger, Angew. Chem. Int. Ed. En 1.. 1216, 15, 333; and references therein. P. C. Hai1es and H. Heigo1d, J. 0rganometa1. Chem., 1gz0, 21, 405. L. I. Shoer and J. Schwartz. J. Amer. Chem. Soc., 1211' 22, 5831. E. E. van Tamelen, R. B. Fechter, S. H. Schne11er, G. Boche, R. H. Gres1ey and B. Akermart. J. Amer. Chem. Soc.. 1222. 21, 1551. A. E. Shi1ov, A. K. Shi1ova, E. F. Kvashina and T. A. Borontsova. J. Chem. Soc.._Chem. Comm., 12Z1. 1590. Yu.G. Borodko, I. N. Iv1eva. L. M. Kachapina, E. F. Kvashina, A. K. Shi1ova and A. E. Shi1ov, J. Chem. Soc..,Chem. Comm., lgzg. 169. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 134 Yu.G. Borodko. I. N. Iv1eva. L. M. Kachapina. S. J. Sa1ienko, A. K. ShiTova and A. E. Shi1ov, J. Chem. Soc., Chem. Comm., 1972, W 1178. F. H. van der Heij, J. H. Teuben and H. SchoTtens, J. 0rganometa1. Chem., 1216, 122, 223. F. H. van der Heij, J. H. Teuben and H. SchoTtens, J. 0rganometa1. Chem., 1977, 127. 299. —— WW — J. H. Teuben and H. J. de Liefde Meyer, J. 0rganometa1. Chem., 1972, 46, 313. «MA; — J. H. Teuben, J. Dgganometa1. Chem.. m3, _5_7_, 159. J. E. Bercaw. R. H. Marvich, L. G. Be11 and H. H. Brintzinger, J. Amer. Chem. Soc., 1972. 94, 1219. MA!» — G. P. Pez and S. C. Kwan, J. Amer. Chem Soc., 1976, 98, 8079. M — For a recent review, see J. Chatt, J. R. Di1worth and R. L. Rich- ards, Chem. Rev., 1978, 78, 589. _________. .Vvv» ... L. C. Kro11 Ph. D. Thesis, Michigan State University. 4243' K. Sonogashira and N. Hagihara. Bu11. Chem. Soc.,_Japan..1966, 22, 1178. . Fachinetti and C. Floriani, J. Chem. Soc.. Chem. Comm.. 1213.66. . L. Thomas and K. T. Brown, J. 0rganometa1. Chem., 1976, 111, 297. . C. Huffman, J. G. Stone, H. C. Kruse11 and K. G. Cau1ton, . Amer. Chem. Soc.. 1977, 99, 5829. ’VVV» "" mecca . Fachinetti and C. F1oriani, J. Chem. Soc..,Chem. Comm.. 12Z2, 654. H. A1t and M. D. Rausch, J. Amer. Chem Soc.. 1212. gg, 5936. B. Demerseman. G. Bouquet and M. Bigorgne, J. 0rganometa1. Chem. 1975, 93, 199. -- -- 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 135 B. Demerseman, G. Bouquet and M. Bigorgne, J. 0rganometa1. Chem.. 1274.191. m G. Fachinetti, G. Fochi and C. F1oriani, J. Chem. Soc.,gChem. 29mm,. 4248' 230. R. R. Schrock, Accts. Chem. Res.. 1212. 12, 98; and references therein. R. J. Burt, J. Chatt, G. J. Leigh, J. H. Teuben and A. Hesterhof, J. 0rganometa1. Chem.. 4244’11229 C33. M. J. Bunder, A. De Cian and M. L. H. Green, J. Chem. Soc., Chem. 2999,. 1211, 59. H. Beck, R. Hoefer, J. Erbe, H. Menze1, U. Nage1 and G. P1atzen, Z. Naturforschg.. 1224. 222, 567. C. U. Pittman, 0. E. Ayers and S. P. McManus, J. Macromo1. Sci. _C_hgn_.. 127/3. 11], 1563. C. U. Pittman and G. 0. Evans. Chem. Techno1., 1212. 2, 560. C. U. Pittman and R. F. Fe1is, J. 0rganometa1. Chem.. 1214.,12, 389. S. Harwe11 and P. Buschmeyer, Angew. Chem. Int. Ed. Eng]., 4248’ 11, 131. H. C. Sti11, J. Amer. Chem. Soc.. 1211, 22, 4836. S. J. Mc1ain, C. 0. Hood and R. R. Schrock, J. Amer. Chem. Soc., 1211, 9_9_, 3519. S. J. McLain and R. R. Schrock, J. Amer. Chem. Soc., 1218,‘122, 1315. C. H. Brubaker, Jr.. Encyci. Po1ym. Sci. TechnoT.,gSuppJement No 1, 1226, 116; and references therein. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 136 P. R. Rong, J. M01. Cata1.. 1222422. 1, 13. J. Conan, M. Barthoiin and A. Guyot. J. M01. Cata1., 1222422. 1, 375. K. Koch1oef1 and H. Liebe1t. J. Chem. Soc.. Chem. Comm.. 1222. 510. H. Arai. J. Cata1.. 12Z§4.§l: 135. T. Masuda and J. K. Sti11e. J. Amer. Chem. Soc.. 1222. 122, 268; and references therein. J. Reed, P. Eisenberger, B. K. Teo and B. H. Kincaid, 21_fimgg, Chem. Soc., 1222,‘122, 2375; and references therein. M. S. Jarre11, B. C. Gates and E. D. Nicho1son, J. Amer. Chem. .So_c.. 1222, 1_(10_,_ 5727. F. Pinna, C. Candi1era, G. Struku1, M. Bonivento and M. Graziani, J. 0rganometa1. Chem., 1222. 122, 91; and references therein. H. 0. Haag and D. D. Hhitehurst, US Patent, 1222, 4,098,727. G. Gubitosa, M. Bo1dt and H. H. Brintzinger, J. Amer. Chem. Soc.. m, 23, 5174. F. R. H. P. Hi1d, G. Gubitosa and H. H. Brintzinger. J. 0rganometa1 Chem., 1222. 142, 73. G. Gubitosa and H. H. Brintzinger, Proc. Inter. Conf. Homog,_ Cata1zsis, Lyon, 1222. H. A. Smith and R. G. Thompson, “Advanced in Cata1ysis', Academic Press, New York, 1222. K. P. C. Bo11hardt, Accts. Chem. Rev., 1222.‘12, 1; and references therein. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 137 E. F. Pau1us, E. 0. Fischner, H. P. Fritz and H. Schuster-Ho1dan, J. 0rganometa1. Chem., 1222, 12, P3. E. F. Pau1us, Acta Crysta11ogr., 1222, 222, 2206. H. A. Herrmann, C. Krfiger, R. Goddard and I. Berna1, J. 0rganometa1 223m,, 1222,2122, 73. H. D. Covey and T. L. Brown, Inorg. Chem., 1222, 12, 2820. A. I. Voge, “A Textbook of Quantitative Inorganic Ana1ysis", Lowe and Brydone (Printers) Ltd., London, 1222. R. Pribi1 and M. Ma1at, "Gravimetric Analysis", Part 11, L. Erdey, Ed.. Pergamon Press, New York, N.Y., 1222, PP. 540. H. A. Barber, Inorg. Synth., 1960, 2, 11. G. Hi1kinson, F. A. Cotton and J. M. Birmingham, J. Inorg. Nuc1. Elfiflhutlfififib £9 95- A. F. Reid and P. C. Hai1es, J. 0rganometa1. Chem., 1222, 2, 329. D. C. Evans, L. Phi11ips, J. A. Barrie and M. H. George, 21_291x, Sci. Poly, Lett. Ed.. 1222, 12, 199. T. M. Fy1esand and C. C. Leznoff, Can. J. Chem., 1222, 22, 935. R. H. Grubbs and S. C. H. Su, J. 0rganometa1. Chem., 1222.‘122, 161. ' R. C. Heast, Ed.. "The Handbook of Chemistry and Physics“, V01. 51, Chemica1 Rubber Co.. C1eve1and, Ohio, 1222, P D-90. M. E. Vo1'pin and V. B. Shur in '0rganometa11ic Reactions”, 291, 1, E. 1. Becker, M. Tsutsui, Ed.. Hi1ey-Interscience, New York, N.Y., 1222, pp 102. R. B. King, '0rganometa11ic Synthesis“, Academic Press. New York, N.Y., mg. 1,115. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. D 0 F. J. . A H H an I '7! 2 “U m 3 O 2538. C. P. 3808. T. J. 2519. R. H. 138 Rausch and R. A. Genetti, J. 0:9. Chem., 1222, 22, 3888. Fischer and K. Bitter, Z. Naturforschg.. 1221. 122, 225. Frank1and, J. Chem. Soc., 122222, 2, 263. Pope and S. J. Peachey, Proc. Chem. Soc., 1222.‘22, 86. Cotton, Chem. Rev., 1222,‘22, 551. Jaffe and G. 0. Doak, J. Chem. Phys., 1222, 21, 196. Grubbs and T. K. Brunck, J. Amer. Chem. Soc., 1222.‘22, Casey and T. J. Brukhardt, J. Amer. Chem. Soc., 1222. 22, Katz and R. Rothchde, J. Amer. Chem. Soc., 1222, 98, Grubbs, D. D. Carr, C. Hoppin and P. L. Burk, J. Amer. Chem. P. G. Soc., 1976, 98, 3478. m— Gassman and T. H. Johnson, J. Amer. Chem. Soc., 1222, 22, 5055. H. 9. M. T. Green1ee and M. F. Farona, Inorg. Chem., 1222, 12, 2129. Moce11a, R. Rovner and B. L. Muetterties, J. Amer. Chem. 225,. 1222, 22, 4089. M. Ephretikhine, J. Levisa11es, H. Rudder and D. Hi11emin, J. 0rganometa1. Chem.,ML 124, C1. R. H. Grubbs, Prog, Inorg. Chem., 1278, 22, 1; and references therein. L. Cassar, P. E. Eaton and J. Ha1pern, J. Amer. Chem. Soc., 1970, 93, 3515. 139 135. P. G. Gassman and J. A. Nikora, J. 0rganometa1. Chem., 1222, 22, B1. 136. K. C. Bishop, Chem. Rev.. 1222.212, 461. 137. T. J. Katz and N. Acton, Tetrahedron Lett., 1221, 21, 2601. 138. T. J. Katz and S. Cerefice, J. Amer. Chem. Soc., 1222. 21, 2405. 139. N. Acton, R. J. Roth, T. J. Katz, J. K. Frank, C. A. Maier and I. C. Pau1, J. Amer. Chem. Soc., 1222,'22, 5446. 140. A. R. Fraser, P. H. Bird, 5. A. Bezman, J. R. Shap1ey, R. Hhite J. A. Osborn, J. Amer. Chem. Soc., 1212, 22, 597. 141. M. J. Doy1e, J. McMeeking and P. Binger, J. Chem. Soc., Chem. Comm., 1222, 376. 142. C. S. Cundy, C. Eaborn and M. F. Lappert, J. 0ranometa1. Chem., mg, 55, 291. 143. C. S. Cundy and M. F. Lappert, J. Chem. Soc., Chem. Comm., 1222, 445. 144. G. M. Hhitesides, J. F. Gaasch and E.'R. Stedronsky, 21_flmgg, Chem. Soc., 1222,‘22, 5258. 145. M. L. H. Green in '0rganometa11ic Compounds“, 2211_11, 3rd ed., G. E. Coates, M. L. H. Green and K. Hade, Ed.. London, 1222. 146. P. S. Braterman and R. J. Cross, Chem. Soc. Rev.. 1212, 2, 271. 147. F. Ca1derazzo, Pure App1. Chem., 1212,‘22, 453. 148. M. C. Baird, J. 0rganometa1. Chem., 1212,‘22, 289. 149. J. Dvorak, R. J. O'Brien and H. Santo, J. Chem. Soc., Chem. Comm., 1222, 411. 150. V. B. Shur, E. G. Berkovitch, L. B. Vasi1jeva, R. V. Kudryavtsev and M. E. Vo1'pin, J. 0rganometa1. Chem., 1222,'22, 127. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 140 C. P. Boeke1, J. H. Teuben and H. J. de Liefde Meijer, J. 0rganometa1. Chem., 1212, 21, 371. R. R. Schrock, Acc. Chem. Res., 1212, 12, 98; and references therein. P. J. Davidson, M. F. Lappert and R. Pearch, Acc. Chem. Res.. 1212, Z, 209; and references therein. P. J. Davidson, M. F. Lappert and R. Pearce, Chem. Rev.. 1212, 12, 219; and references therein. R. R. Schrock and G. H. Parsha11, Chem. Rev.. 1212, 12, 243; and references therein. J. K. McDermott and G. M. Hhitesides, J. Amer. Chem. Soc., 1222, 22, 947. J. X. McDermott, M. E. Hi1son and G. M. Hhitesides, J. Amer. Chem. _S_0_c_.. 1222, _9_8, 6529. R. H. Grubbs and A. Miyashita, J. Chem. Soc.,_Chem. Comm., 1211, 864. R. H. Grubbs and A. Miyashita, J. Amer. Chem. Soc., 1212. 122, 1300. M. P. Brown, A. Ho11ings, K. J. Houston, R. J. Puddephatt and M. Rashidi, J. Chem. Soc., Da1ton, 1212, 786. J. X. Dermott, J. F. Hhite and G. M. Hhitesides, J. Amer. Chem. 229.. W2, 22, 4451. J. X. Dermott, J. F. Hhite and G. M. Hhitesides, J. Amer. Chem. 222., 1222, 22, 6521. G. 8. Young and G. M. Hhitesides, J. Amer. Chem. Soc., 1212. 122, 5808. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 141 R. H. Grubbs, A. Miyashita, M. Liu and P. Burk, J. Amer. Chem. §_o_c_.. 1222, 22, 3863. R. H. Grubbs, A. Miyashita, H. Liu and P. Burk, J. Amer. Chem. 295.. 1222, 122, 2418. R. H. Grubbs and A. Miyashita, J. Amer. Chem. Soc., 1222. 122, 7416. H. Gi1man, R. G. Jones and L. A. Hood, J. Amer. Chem. Soc., 1222, 12, 3615. M. S. Kharasch, J. H. Hancock, H. Nudenberg and P. O. Tawney, J. Org. Chem., 1222, 21, 322. H. B. Smith, J. Org. Chem., 1221, 22, 5187. F. H. Frey, Jr.. J. Org. Chem., 12219.29: 4206. M. 1. Prince and K. Heiss, J. 0rganometa1. Chem., 1222, 2, 166. M. H. Abraham and M. J. Hogarth, J. 0rganometa1. Chem., 1222. 12, 1. G. A. Razuvev, V. N. Latyaeva and L. I. Vyshinskaya, Dok1. Akad. Nauk SSSR, 1222, 122, 383. R. H. Grubbs and A. Miyashita, J. Amer. Chem. Soc., 1222,2122, 7418. i R. H. Grubbs and C. R. Hoppin, J. Chem. Soc., Chem. Comm.,1222, 634; and references therein. R. H. Grubbs and A. Miyashita, Abst. of 6th North American Meeting of the Cata1ysis Soc., Chicago, 111., 1222, C1. G. J. Erskine, J. Hartgerink, E. L. Heinberg and J. 0. McCowan, J. 0rganometa1. Chem., 1222,‘1Z2, 51; and references therein. 142 178. D. E. App1equist and D. F. O'Brien, J. Amer. Chem. Soc., 1226’1§§9 743; and references therein. 179. G. A. Hi1ey, R. L. Hershkowitz, B. M. Rein and B. C. Chung, J. Amer. Chem. Soc., W, 86, 964. 180. E. O. Fischer, S. Vigoureus and P. Kuze1, Chem. Ber.. 1269,.43, 701. 181. F. H. Siegert and H. J. de Liefde Meijer, J. 0rganometa1. Chem., 1212: g, 177. 182. A. V. Baa1en, C. J. Groenenboom and H. J. de Liefde Heijer, J. 0rganometa1. Chem., 1213, 13, 245. 183. R. L. Cooper and M. L. H. Green, J. Chem. Soc.(A), 1361, 1155. "1111111111111111“