IJI‘. .II .2... {mm} .4.va . a? .m. him . .0. .34...» Iii 0.: an. fl . J «41 .. Ac... .3 . am... . Agra. . .1: $3. a... at a; .3 - fiflpfiwu v3 :721. ":1.va u .1:- I h.‘ 3% . “VII. £4 r I V. 3.. .» : .. 1.74;... ~ . .k. .. r... _ z... : . _ . . .. A, :. I: . ..l.. . . . u... .. :4 . . t .121... afiqéflwafi: in. .u .~ ,n.‘..n.m..,.h_..wz..a.fi,m, a? fimfféfmfigggé .. Q. .....m , . ..,.ue.-u.;‘u.u.1.vtavyéiaé‘xfl ‘3.“ or! . 2 an??? m . Mina? . LIBRARY Michigan Sta re University ABSTRACT REACTIONS OF ENOLATE ANIONS By Andreas Lindert The formation of polyalkylated products is a major source of dif- fulty in the alkylation of ketone enolates. In order to reduce this side reaction, trialkylborates and boranes were added to ketone enolates prior to alkylation. Both triethanolamineborate and triethylborane were effective additives for controlling polyalkylation. The organic borons function by coordinating with the ketone enolates to furnish a new anion which possesses greater selectivity for alkylation. Thus cyclohexanone, 2-methylcyclohexanone, cyclopentanone, and 3-pentanone were monomethylated in 80-95% yield (0.0% to 1.0% dimethylation), and cyclohexanone was monobutylated with n-butyl iodide in 64% yield (1% to 7% dibutylation). Especially noteworthy were the results obtained in the methylation of 2-methylcyclohexanone. The thermodynamically controlled enolate distribution gave predominately 2,2-dimethylcyclohexanone upon methylation; while the kinetically controlled distribution of enolates yielded mainly the 2,6-dimethylated product. In part II, stable lithium ester enolates were prepared by the reaction of lithium N-isopropylcyclohexylamide (LiICA) with the ethyl or tert-butyl esters of acetic acid and mono- or di-alkyl acetic acids. Andreas Lindert .+ R u . R\\ o-Ll H -8-o-R LiICA A =c’ 72% THF/ -78°C 7 Ri/c \OR" These ester enolates were alkylated at 0°C with methyl, n-butyl, n-octyl, isobutyl, isoamyl, and isopropyl iodide and allyl and benzyl bromide by the addition of the ester enolate to a DMSO solution of the Organic halide. The best yields of the desired alkylated products were obtained when the less sterically hindered and the more reactive orQanic halides were employed. In part III. thelithium anions of esters, dialkyl amide, carboxylic acids, nitriles and ketones were dimerized with either copper (II) bromide or copper (II) valerate. Li-—#-—X + 2Cu(Y)2—————€>X-%-¢-—X + ZCuY + LiY X -8-0R, -g-0H, -CEN, -8-R CH CH or Br' -0-8-CH2CH2 2 3 .< II The ester enolates and their derivatives rapidly dimerized and either the CuBr2 or Cu(V)2 oxidant could be successfully used in di- merization. In the coupling of ketone enolates, only copper (II) valerate gave satisfactory yields of the l,4-diketone. With all compounds studied. the best yields were afforded when sterically unhindered enolate anions were coupled (63-l00%). Bulky substituted carbanions gave lower yields of the desired dimer (20-50%) and greater yields of side products. REACTIONS OF ENOLATE ANIONS By Andreas Lindert A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1972 Zu meiner Familie und meiner lieber Frau, Gudrun ii ACKNOULEDGMENTS The author wishes to extend his appreciation to Dr. Michael N. Rathke for his guidance and assistance throughout the course of this investigation. Thanks are also given to Dr. William H. Reusch for his many helpful suggestions in the preparation of this thesis. The financial assistance provided by Michigan State University in the form of teaching assistantships from September, 1968 to June, 1972 and by the Petroleum Research Fund in Fall, 1971 is gratefully acknowledged. Finally, the author wishes to thank his wife, Gudrun for her understanding and encouragement as well as for the typing of this thesis. iii TABLE OF CONTENTS PART I THE ALKYLATION OF KETONE ENOLATES HISTORICAL INTRODUCTION ..................... Generation of a Specific Ketone Enolate ........... Reduction of Side-Reactions in Alkylation .......... INTRODUCTION .......................... RESULTS ............................. Trialkyl Borates ...................... Trialkyl Boranes ...................... Butylation of Cyclohexanone ................. Methylation of 2-Methy1cyclohexanone ............ Summary ........................... DISCUSSION SECTION ....................... EXPERIMENTAL SECTION ...................... I. General .......................... Spectra ........................... Gas Chromatography ..................... Melting Points ....................... Refractive Index ..................... . II. Materials ......................... Tri-t-butylborate ...................... Triisopropylborate ..................... Page 2 5 6 TO 15 20 25 26 26 TABLE OF CONTENTS - Continued Page Tri-n-butylborate ..................... 26 Trimethylborate ...................... 27 Triethanolamineborate ................... 27 Triethylborane ...................... 27 Tricyclopentylborane ................... 27 Ketones .......................... 27 Alkyl Halides ....................... 28 III. Bases .......................... 28 Preparation of Lithium bis (trimethylsilyl)amide (LiHMDS) , 28 Preparation of Lithium N-Isopropylcyclohexaylamide (LiICA). 28 Preparation of Sodium bis (trimethylsilyl)amide (NaHMDS). . 30 IV. The Preparation and Reaction of Ketone Enolates ..... 30 Preparation of Cyclohexanone Enolate ........... 31 Recovery of Cyclohexanone ................. 31 Reaction of Ketone Enolates with Alkyl Halides ...... 31 The Use of Alkyl Borates and Boranes ........... 32 The Use of Triethanolamine Borate (TEAB) ......... 32 The Use of Triethylborane ................. 32 Methylation of 2-Methylcyclohexanone ........... 33 PART II THE ALKYLATION OF ESTER ENOLATES ACID and ESTER ENOLATES in ALKYLATION REACTIONS ........ 35 Ester Enolates ...................... 37 TABLE OF CONTENTS - Continued Page RESULTS AND DISCUSSION .................... 41 EXPERIMENTAL SECTION ..................... 48 I. Materials ........................ 48 Alkyl Halides ....................... 48 Tosylates ......................... 48 Esters .......................... 49 II. Preparation and Reactions of Ester Enolates ....... 49 Preparation of Lithium tert-Butyl Acetate ......... 49 Recovery and Deuteration of the Esters .......... 49 Reaction of the Ester Enolate with Alkyl Halide ...... 50 Product Analysis ..................... 50 tert-Butyl Hexanoate .................. 52 tert-Butyl 4-Methylpentanoate ............. 52 tert-Butyl 3-Methy1butanoate .............. 52 tert-Butyl 2-Methylhexanoate .............. 52 tert—Butyl 2,2-Dimethylpropanoate ........... 52 Ethyl 2-Isopr0pylhexanoate . . . ............ 52 Ethyl 2-Methy1hexanoate . . .............. 53 PART III THE COUPLING 0F ENOLATE ANIONS with COPPER COMPOUNDS INTRODUCTION AND HISTORY ................... 55 RESULTS ............................ 60 Acids ........................... 62 Amides .......................... 63 TABLE OF CONTENTS — Continued DISCUSSION Nitriles ......................... Benzylonitrile ...................... Ketones ......................... Copper (1) Enolate Complexes ............... Copper (11) Salts The Reactivity and Nature of the Anion .......... EXPERIMENTAL SECTION ..................... I. II. Materials ....................... 1-Bromo-3,3-Dimethyl-2-Butanone ............. Cuprous Iodide ...................... Ketones ......................... Preparation of Copper (II) Valerate ........... Preparation of N, N-Dibutyl Acetamide .......... Preparation of N, N-Diethyl Isobutrate .......... Coupling of Enolate Anions ............... Esters .......................... Dimerization Dimerization Dimerization Dimerization Dimerization Dimerization of Ethyl Propionate ........... Ethyl Isovalerate ........... Ethyl Hexanoate ............ Ethyl Isobutyrate ........... Ethyl Phenylacetate .......... of T-Butyl Acetate ............ Amides .......................... Dimerization Dimerization of of N, N—Di-n-Butylacetamide ....... N, N-Diethylisobutyramide ....... vii Page 64 65 65 73 75 76 78 82 82 82 82 83 83 83 84 84 84 84 85 85 85 86 86 86 86 87 TABLE OF CONTENTS — Continued Page Carboxylic Acids ..................... 87 Dimerization of Octanoic Acid ............. 87 Dimerization of 3-Methy1propionoic Acid ........ 88 Dimerization of Isobutyronitrile ............. 88 Benzylnitrile ....................... 89 Ketones .......................... 90 Dimerization of 3,3-Dimethyl-2-Butanone ........ 90 Dimerization of Cyclohexanone ............. 90 Dimerization of 4-Methyl-2-Pentanone .......... 91 Dimerization of 2,4—Dimethyl-3-Pentanone ........ 91 Dimerization of 3-Pentanone .............. 91 Dimerization of 2-Hexanone ............... 92 Dimerization of Acetophenone .............. 92 Aldol Condensation Reaction of Dicyclopropyl Ketone . . 92 III. Reactions of Ketone Enolate-Copper (I) Complexes . . . . 93 Cyclohexanone Enolate ................... 93 3,3-Dimethy1-2-Butanone Enolate .............. 93 BIBLIOGRAPHY ......................... 95 viii Table 10. 11. 12. 13. 14. 15. 16. 17. LIST OF TABLES Methylation of Cyclohexanone Enolate ........... Methylation of Cyclohexanone Metal Enolate with TEAB in THF/DMSO (1:1 ratio) ................ Methylation of the Lithium Ketone Enolates Using TEAB in Monoalkylation ................... Methylation of Cyclohexanone Metal Enolates Using TricyclOpentylborane ................. Methylation of Cyclohexanone Metal Enolate Using Triethylborane ..................... n-Butylation of Cyclohexanone Metal Enolates ...... Butylation of Sodium Cyclohexanone Enolate Using Triethylborane ..................... Butylation of Metal Enolates of Cyclohexanone with TEAB in THF/DMSO (1:1 ratio). . . . . ........ Methylation of 2-Methylcyclohexanone with TEAB in a THF/DMSO solvent (1:1 ratio) ............. Results of Deuterium Oxide Quenching EXperiments of Ester Enolate Solutions Generated with LiLCA ...... n-Butylation of Lithium tert-Butyl Acetate with n-Butyl Iodide. . . . . . . . ................. Alkylation of Lithium Ester Enolates with Organic Halides. . Product Constants Compared with Literature Values . . . . Dimerization of Esters Using Copper (II) Salts ...... The Dimerization of Ketones with Copper (II) Valerate . The Dimerization of Ketones with Copper (11) Bromide. . . . Coupling of 3,3-Dimethy1-2-Butanone ........... ix Page 12 12 14 14 16 16 17 17 19 42 42 45 51 61 68 7O 72 LIST OF FIGURES Figure Page 1. Reaction Apparatus ..................... 29 2. The Coupling of Anions with Copper Compounds ........ 56 PART I THE ALKYLATION OF KETONE ENOLATES HISTORICAL INTRODUCTION The base-promoted alkylation of ketones provides the chemist with a useful method for the formation of a new carbon to carbon bond. Gen- eration of intermediate ketone enolate anions has been accomplished both quantitatively by the action of strong bases or in equilibrium amounts by using weaker bases such as the sodium alkoxides. The sub- sequent alkylation of the enolates provides a useful method of syn- thesizing substituted ketones (l). -Na+ 0 0 NaNH CH2=CHCH2C1.CH2CH=CH2 2 \\ .(CH2 CH= CH 2)2 Ether 5 reflux reflux 54-62% (Ref. 2) Alpha-alkylation of ketones is often beset with undesirable side reactions which decrease its usefulness in synthesis, and ingeneous methods have been developed to overcome or circumvent these diffi- culties (1). Normal monoalkylation of ketones has always been plagued by di- and polyalkylation (3), self-condensation (4) to aldol-type side products and, to a much lesser extent, O-alkylation (1). The importance of these side reactions is largely dependent on the nature of the ketone, and the reaction conditions employed. In performing the alpha alkylation of ketones, it is always impor- tant to choose conditions that will minimize side reactions. In the past, the chemist has often effected the alkylations of ketones under 2 unfavorable conditions. This is particularly true when heterogeneous bases (NaH) or weak bases such as metal alkoxides (l) are used to gen- erate the ketone enolates, since these bases generate the enolates slowly or in equilibrium concentrations. In order to reduce self-condensation, it is therefore necessary to generate the enolate in the presence of excess alkylating agent. This method of running the reaction leads to increased yields of dialkylation, and the degree to which self-conden- sation under these conditions takes place varies greatly from ketone to ketone. With cyclohexanone self-condensation is relatively unimportant, while with cyclobutanone and especially cyclopentanone, self-condensation is of major importance (3). Self-condensation of the enolate with starting material can be largely eliminated by employing strong, homogenous bases. A wide variety of strong bases has been used in the quantitative generation of ketone enolates and a comprehensive list is presented on page 547 of ref. one. Of the bases listed, the sodium and lithium dialkylamides are the most useful since they can be easily prepared and are of sufficient strength to generate the enolates quantitatively (5). It is desirable to employ metal dialkylamides such as the sodium or lithium salts of diisopropylamine (6,7), hexamethyldisilazane (8), or N-isopropylcyclo- hexylamine (9), since these amines possess bulky alkyl groups. The sterically hindered bases therefore preferentially abstract an alpha proton from a ketone and no side-products resulting from attack upon the carbonyl are observed: .+ . 0-L1 [(CH ) -CH] -N’L1+ \\ . 32 2 + [(CH3)2-CH]2-N-H -78°C/DME (Ref. 7b) V/ 4 A further advantage achieved by the use of bulky alkyl substituents is the increased solubility of the basic reagent. Most metal dialkylamides have been shown to be soluble in a wide range of solvents from non-polar hydrocarbons, and benzene derivatives to the polar ethers, THF, DMSO, HMPA and DMF. These soluble, homogeneous bases react rapidly with ketones to generate ketone enolates with little self-condensation being observed. Furthermore, the bases are easily separated from the final reaction products by either simple distillation or extraction with an acid solution. If an equivalent amount of a strong base is employed in the mono- alkylation of ketones, the amount of di- and polyalkylation can be re— duced but not eliminated. This polyalkylation is due to a proton exchange of alkylated products with the starting enolate (reaction K2 and K3) illustrated in the following reaction with cyclohexanone: 0 0°C Base DME .CH3 56% Ether CH3 I :2//:. ‘&\\\\\:J C" ”t; £6" 6‘“ 13% (Ref. 7b) If reactions K2, K3, and K4 can compete with the initial alkylation re- action K1 considerable dialkylation will be observed (2). Generation of a Specific Ketone Enolate. Often in the course of a synthesis the need arises to introduce an alkyl group selectively at one of the two alpha positions of an symmet- rical ketone. Many imaginative methods have been developed to accom- plish specific alkylation. Among the most common and useful procedures for the formation of the desired alkylated ketone has been the reaction of the ketone with activating or blocking groups or the preparation of enamine derivatives of the ketone prior to alkylation (1). In recent years, the alkylation of a specific enolate anion under conditions where equilibrium among the possible structural isomers cannot occur has been increasingly used (1). The reduction of a, B- unsaturated ketones (10) or a-halo-ketones (11), the reaction of enol esters (l) and enol silyl ethers with organometallic (1,6,7a) reagents, and the direct reaction of ketones with bases under suitable conditions have proven to be the most useful methods for the generation of specific ketone enolates. The reaction of a ketone with base provides the only direct method of generating a specific ketone enolate from the corresponding ketone. Two differing procedures can be used to generate enolates. When weak bases are used, or the reaction is run under conditions which allow the formation of an equilibrium mixture (i.e. with an excess of ketone), the more highly substituted enolate usually predominates (1). In the second method, the enolate anion is generated under kinetic- allycontrolled condition. For this purpose strong hindered bases, such as the metal dialkylamides are most satisfactory and yield the less highly substituted ketone enolate as the predominant conjugate base. n-C4H9CH2COCH3 [(CH3)2CH]2NLi -78°CD:6 25° (1'4 QQUIV) .+ . O—L.‘ 8"..1 n-C4H9CH==C-CH3 + n-C4H9CH2- -CH2 16% mixture 84% mixture (Ref. 7b) Subsequent alkylation of the enolates produces the desired product along with some di- and polyalkylation. Reduction of Side-Reactions in Alkylation. To limit self-condensation, O-alkylation and di- and polyalkylation in the alkylating of ketone enolates a variety of cations and complexing agents have been utilized. For example, by increasing the covalent character of the oxygen to metal bond, it has been possible to decrease side reactions in some instances (1). Also complexation of the ketone enolate with tributyl tin chloride or triethylaluminum prior to alkylation has proved useful in reducing side reactions (12). .+ Li .+ - 1 - L lEt 0 o/A 3 HMPA 0 o 0 Et Al 3 \ \ CH3I CH (CH3) CH3 H3 7 7\ DME 2 hrs./r.t. . . . In the latter case a lower rate of alkylation necessitated the addition of hexamethylphosphoramide in order to achieve reaction in a reasonable time. Triethylaluminum is much more useful as an additive with lithium enolates than tributyl tin chloride, however it possesses the disadvantage of being extremely flammable and is therefore incon- venient to use. A more tractable reagent that would complex with ketone enolates and reduce the side reactions that normally accompany monoalkylation would be useful to the organic chemist. In order to find such a con- venient complexing reagent for ketone alkylation, a study to determine the effectiveness of employing trisubstituted boron compounds in reducing the reactivity of ketone enolates was undertaken. INTRODUCTION A systematic study of the effect of borane and borate compounds on ketone enolates in a-alkylation reactions has been made. In most reac- tions, cyclohexanones were employed since condensation is seldom encoun- tered with cyclohexanone derivatives. Consequently the effect of boron compounds on di- and polyalkylation could be determined with less re- gard to other side reactions. The boron compounds that were used in these alkylation reactions were tri-t-butyl borate, triisopropyl borate, trimethyl borate, triethanolamine borate, triethyl borane, and tri- cyclopentyl borane. It was hoped that a boron compound could be found that would form a ketone enolate complex (I): which would alkylate (Ra) much faster than undergoing condensation (RC), and exchange (Re). The results presented in the following section demonstrate that some boron compounds are indeed useful in maximizing the alkylation reactions of ketones. RESULTS Two methods were employed to generate metal dialkylamides used in this study. In the first, 1ithiunHi-Isopropylcyclohexylamide (LiICA) and lithiwnbis(trimethylsilyl)amide (LiHMDS) were easily prepared by the reaction of the corresponding amine at 0°C with n—butyllithium in hexane. R\ 0°C R /N-H + n-butylLi 3>\N—Li + n-butane R Hexane R/ The sodium salts of the amines, on the other hand, are more difficult to generate and were prepared by refluxing a heterogeneous mixture of the amine with sodium amide in dry benzene (13). Sodium amide can be suc- cessfully used only in the preparation of sodium bis(trimethylsilyl)- amide, and is not of sufficient reactivity to generate the sodium salt of N—Isopropylcyclohexylamine. The sodium or lithium ketone enolate employed in the alkylation re- actions was generated by addition of the ketone to the corresponding metal dialkylamides in THF solution. 10 11 The ketone enolate thus produced is stable in THF over a prolonged pe- riod and the parent ketone can be recovered in 89-95% yield by quenching with diluted acid and analysis by GLC. The metal cyclohexanone enolate thus generated was alkylated, and the extent of polyalkylation was determined as a function of both cation and solvent system. The results are shown in Table l and all future work concerned with preventing dialkylation was compared with the results of Table 1. All reactions were performed at 0°C and were complete within 15 minutes. As expected, the least amount of dialkylation was obtained with the more covalent lithium cation and the greatest dialkylation re- sulted when the THF/DMSO solvent system was utilized. Trialkyl Borates. In an attempt to prevent dialkylation, organic borates were used to reduce the reactivity of the enolate. The sodium cyclohexanone enolate was complexed with the borate prior to methylation. N + “3+ ‘ (0R) 0' a (T8 3 o I illlll )7 illIII 7' THF r.t. Whereas the sodium and lithium enolates react instantly the trimethyl borate-enolate complex did not give products even after 2 hours at 0°C. Increasing the reaction to 3 1/2 days yielded 5.6% product while refluxing the reaction for 4 hrs. yielded 11% of the 2-methylcyclohexanone. The borate was changed from triisopropyl borate to tri-n-butyl borate, to 12 Table 1. Methylation of Cyclohexanone Enolate. Cation Solvent Cyclo- Methylcyclo- Di-Methyl- hexanone hexanone cyclohexanones Li THF 9.0% 74.0% 5.5% Na THF ------- 50.0% 17.0% Na THF/DMSO 25.0% 43.0% 23.5% Table 2. Methylation of Cyclohexanone Metal Enolate with TEAB in THF/DMSO (1:1 ratio). Cation Monomethylated Di-Methylated Cyclo- TEAB Order of Cyclohexanone Cyclohexanones hexanone Addition*’ Li 71.5% 6.8% 1.0% '-- ' Na 73.1% 12.0% 15.0% -" - Li 79.0% 0.0% 1.7% yes + Na 78.7% 3.5% 4.6% yes + Na 66.8% 26.6% 10.0% yes — * + means that DMSO is added before the alkyl halide. - means that alkyl halide and DMSO are added to the enolate solution all at the same time. 13 tri-t-butyl borate but the yield of 2-methy1cyclohexanone only increased from 15 to 25%. In all cases, large amounts of cyclohexanone were iso- lated and little or no dialkylation was observed. Triethanolamine borate (TEAB) proved to be the only borate investig- ated which increased the yield of 2-methylcyclohexanone and at the same time reduced dialkylation. As shown in Table 2, the yield of 2-methy1- cyclohexanone is high, ranging from 71 to 79%. TEAB also proved useful in the methylation of 3-pentanone and cyclopentanone (see Table 3). It is necessary to use a THF/DMSO solvent system when employing TEAB due to the insolubility of TEAB in THF even in the presence of the enolate. The aprotic solvent must be added prior to the addition of the alkylating agent with sufficient time being allowed (usually 5 minutes) for the reaction mixture to become homogeneous, otherwise the borate- enolate complex does not form and considerable dialkylation is observed (see Table 2). Trialkyl Boranes. Two trialkyl boranes were evaluated to determine their usefulness in reducing dialkylation in the methylation of ketone enolates. Tri- cyclopentylborane was useful in reducing dimethylation, however the yield of monoalkylated cyclohexanone was low (see Table 4). Increasing the reaction time improved this yield slightly, however the dialkylation side product also increased. Triethylborane proved to be the most successful reagent employed in limiting the side reactions in ketone alkylations. Using a THF solvent system 76 to 80% of the monoalkylation product can be obtained with no corresponding dialkylation being detected. The reaction in THF is slow, Table 3. Methylation of the Lithium Ketone Enolates Using TEAB in Monalkylation. Ketone TEAB Monoalkylated Dialkylated Alkylating Ketone Ketones Agent 3-Pentanone No 90% 10% CH3-I 3-Pentanone Yes 92% 1% CH3-I Cyclo- pentanone No 85% 12% CH3-I Cyclo- pentanone Yes 95% 0% CH3-I Table 4. Methylation of Cyclohexanone Metal Enolates Using Tricyclopentylborane. Cation RX'N Solvent Cyclo- Methylcyclo- Di—Methyl- Time hexanone hexanone cyclohexanones Li 1 hrs. THF 18.4% 55.5% None Li 2 hrs. THF 19.5% 55.0% 3.3% Li 18 hrs. THF 13.0% 56.0% 6.0% Na 2 hrs. THF 17.0% 44.0% None Na 11 hrs. THF 17.0% 45.0% 8.7% 15 requiring two hours for at least 97% of the reaction to go to completion. The reaction time can be shortened to 1/2 hours and the yield increased to 90% by using a THF/DMSO solvent system in 1:1 ratio (see Table 5). Butylation of Cyclohexanone. The reactivity of the alkylating agent is of utmost importance in a—alkylation reactions of ketones. When the alkylating agent is unreac- tive, as in the case of long chain alkyl halides, self-condensation and exchange can compete favorably with the alkylation reaction. Thus the butylation of cyclohexanone proceeds slowly, and considerable dialkylation is observed. Changing the cation from lithium to sodium or using DMSO/ THF solvent has little effect in reducing dialkylation (see Table 6). Both triethanolamine borate and triethylborane were used to decrease dialkylation. The butylation of the triethylborane-enolate complex is extremely slow at room temperature (see Table 7) and refluxing the reac- tion only results in greater yields of dialkylation products. The highest yield of monoalkylation and smallest yield of dialkylation products are obtained by using a 1:1 ratio of DMSO/THF solvent in conjunction with triethylborane. The use of DMSO or HMPA proved supe- rior to DMF as a solvent. An attempt to run the reaction in 100% DMSO proved futile since it was impossible to generate the enolate from lithium N-isoproplycyclo- hexylamide in a DMSO solvent. TEAB also proved useful in the butylation of cyclohexanone and yielded as much as 60-69% of the monoalkylated product (Table 8). The amount of dialkylation was reduced, but not to the extent observed with triethylborane. 16 Table 5. Methylation of Cyclohexanone Metal Enolate Using Triethylborane. Cation Reaction Solvent Cyclo- Methylcyclo- Di-Methylcyclo- Time System hexanone hexanone hexanone Li 2 hrs. THF 12.6% 77.8% None Na 2 hrs. THF 10.3% 80.8% None Na 1 hrs. DMSO/ Table 5. n-Butylation of Cyclohexanone Metal Enolates Cation Reaction Temp. Solvent Cyclo- Butylcyclo- Di-Butylcyclo- Time System hexanone hexanone hexanone Na 1 hrs. r.t. THF 82.0% 18.6% None Na 2 hrs. r.t. THF 47.7% 51.5% 10.0% Na 8 hrs. r.t. THF 15.8% 66.0% 25.0% Na 12 hrs. reflux THF 3.7% 39.5% 23.7% Li 20 hrs. r.t. THF 14.3% 55.6% 18.6% Li 1 hrs. r.t. THEM??{1) ---- 56.5% 23.2% Na 1 hrs. r.t. “”50/ 20.6% 63.3% 21.0% THF (1:1) Table 7. Butylation of Sodium Cyclohexanone Enolate Using Triethylborane. 17 Solvent Reaction Solvent Cyclo- Butylcyclo- Di-Butylcyclo- Ratio Time System hexanone hexanone hexanone --— 2.5 hrs. THF 89.0% 8.9% None --- 8.0 hrs. THF 64.0% 22.3% 6.1% . DMSO/ 0 1.1 .5 hrs. THF 19.7% 70.0% .5% 1:1 2.0 hrs. DMSD/ 10.9% 76.3% 1.4% THF 1:5 20.0 hrs. onag/ 22.0% 50.5% 11.8% 1:5 20.0 hrs. D¥fig/ 28.0% 56.3% 17.0% 1:5 20.0 hrs. D¥fi2/ 25.4% 46.3% 25.3% Table 8. Butylation of Metal Enolates of Cyclohexanone with TEAB in THF/DMSO (1:1 ratio). Cation Monobutylated Di-Butylated Cyclo- TEAB Order of* Cyclohexanone Cyclohexanones hexanone Addition Li 56.5% 23.2% ---- ---- + Na 63.3% 21.0% 20.6% ---- + Na 60.8% 24.9% 13.8% ---- - Li 63.5% 7.3% ---- yes + Na 58.5% 15.6% 16.0% yes + Na 69.0% 15.6% 12.7% yes - * + means that DMSO is added before the alkyl halide. - means that alkyl halide and DMSO are added to the enolate solution all at the same time. 18 Methylation of 2-Methylcyclohexanone. By reacting 2-Methy1cyclohexanone with NaHMDS under both equilibrium and kinetically controlled conditions, it was possible to produce the enolate predominately at either the two or six position in the ketone. (my 11)-” (kinetic) R .+ .+ L1 L1 .4. 0' 0’ 0E1 Li+ \ CH / CH 0" 3 3 d. "3 maJor m1n0r 99% 1% TEAB CH I TEAB CH I 3 3 0 0 CH ) CH CH 3 2 3 CH3CH3)2 (see Table 9) 19 5m. Rm.wp &_.5 NF.m Rm.~o mm» .CTE om .p.2\u.o Rom. &¢.mp fio.m xm.~ fio.wm was .:we mm .p.c\ooo Re. 55.np so.m~ 5P.NN 5m._fi mm» ----- .o.s\uom~- Rm. 5m.m 5N.mm gm.mm 55.5 mm» ----- .p.c\uoom 5N. No.“ 56.5m &o.o¢ &~.o mm» ----- .H.s\u.o 5m.~ 50.5 5m.mm 5m.oe fim.m --- ----- .6.2\Uoo o o . corpmcnw cowpummc a N32 _ 5 m1 _ max: 5 m: _ 5 £38‘ m :33. 5:282“. . . o . . F mpmrocu Aswan; Fury pca>F6m omzo\azc 6 :2 mRCH2=CH2 + Hé}-c-o Na+ T ”ax R O'Na RCH 4) P.L. Creger (6) has successfully extended the usefulness of acid enolates to aromatic systems by producing and alkylating the acid enolates of toluic acids and dimethylbenzoic acids. 8-0-“ + fi-O-Li+ -0H 2 L1 N—[CH(CH3)2]g> 1) RX 2_x 2) HT 7 CH3 H2’L1 HZ-R 90% S) M.M. Larcheveque (33) has obtained lithio acid enolates by the reduction of a, B-unsaturated acids with lithium metal in HMPA. Methylation of the lithuim acid enolate resulted in moderate yields of methylated products. Ester Enolates. The increased solubility of ester enolates in organic solvents as well as the ease of handling and purification of the ester products, make the ester enolates desirable intermediates in organic synthesis. In some instances, as in the carboxylation of ester enolates to pro- duce mixed malonic acid derivatives (34), the ester enolates provide the most direct method of synthesis. The generation and alkylation of ester enolates has proved less successful than the acid enolates due to side reactions. 38 The following bases have been employed in generating and alkylating ester enolates. NaH (35; Na or LiC03 (36 Li or Na[Si(CH3)3]2 (8a,c) Na,Ll'-N-R2 (5) K,Na-NH2 (37) With the weaker bases or in poor alkylation solvents, self—condensation of the ester seriously reduces the yield of expected product, pro- ducing instead a B-keto ester. CH3-8-OR + CH2=<::i——--—§'CHB-Q-CHz-E-OR When strong metal amide bases are used, self-condensation can be re- duced; however, the formation of amides becomes a competing side-reaction as illustrated below with sodium amide (39). R-X NaCH2-8-0-0(CH3)3-———9RCH2-fi-0-C(CH3)3 NaNH2 ///;“ CH3-g-0-C(CH3)3 CHB-E-NH2 + Na-O-C(CH3)3 When the t-butyl or t-amyl esters were alkylated from 30% to 87% of the desired product was isolated (37). The increased steric hin- drance afforded by the use of the triethylcarbinyl ester of dialkyl- acetic acids enabled Hauser to increase the yield of the alkylated product to 70-80% (38). The alkylation reaction is most successful with the esters of acetic acid or dialkylacetic acids and least 39 successful with the esters of straight-chain acids giving only poor to fair yields of the desired product. The use of sterically hindered metal dialkyl amides greatly reduces both amide formation and self-condensation. The first attempts by Hauser to use a solution of lithium diisopropylamide in ether to generate the lithium ester enolate of ethyl isobutyrate proved unsuccessful (37). Kruger and Rochow, however, were able to produce the sodium ester eno- late in ether by employing the more reactive sodium salt of hexamethyl- disilazane. Since the sodium enolates are less stable than their lithium counterparts, they pose difficulties as synthetic reagents. Both our work (40) and that of others (5,34) has shown that the lithiwndialkyl amides are very useful reagents for the generation of ester enolates when the polar aprotic solvents THF, DMSO, or HMPA are employed. The reaction conditions for alkylation using these bases will be discussed in the following experimental section. To decrease self-condensation, the sodium ester enolate has also been generated on the surface of a polystyrene polymer (41). OEC'Lfi + 8H2-8-0-[p]__5’£13r‘_9§=({mzp] ——-T 0- 1) R'X 2) Na0H 2 HO-g-CHR + H-0-8-CH2R \R' 20% 80% 40 The alkylated acid was isolated in only 20% after hydrolysis of the ester. The low yield obtained is most likely a result of the unfavor— able solvent system employed in the alkylation reaction. This method of generating ester enolates on a polymer surface has however proved use- ful for the synthesis of ketones by acylation and decarboxylation of an ester enolate (36b) as well as for specific Dieckman ring for- mation (42). RESULTS AND DISCUSSION 'Hnalithiwnester enolates were prepared by the slow addition of an ester to lithium bis(trimethylsilyl)amide (LiHMDS) or lithium N- isopropylcyclohexylamide (LiICA) in a THF solution. 1 -78°C + . - {‘II'T + H-ccoz-C(CH3)3-—————-—9» L1 C02C(CH3)3 :‘III‘ N’L°+ H-N (CH3)2CH/ 1 \CH(CH3)2 The LiHMDS base is useful in the preparation of the less hindered and more reactive acetate esters. Unfortunately, attempts to generate the lithium conjugate base of ethyl hexanoate at -78°C resulted in slow condensation of the ester over a one hour period. The use of the more sterically hindered t-butyl hexanoate gave similar results; however, the rate of self-condensation was much reduced. The LiICA base proved to be useful for the preparation of all the lithio ester enolates studied. The generation of the lithium ester enolates by LiICA was studied by reacting the base with the ester at either -78°C or 0°C. The reaction was quenched with deuterium oxide and the product examined by GLC for recovered starting material and by NMR for deuterium incorporation (see Table 10). 41 42 Table 10. Results of Deuterium Oxide Quenching Experiments of Ester Enolate Solutions Generated with LiLCA. Ester Recovered Deuterium Temp. Ester, % Incorp.,% Ethyl propionate 90 50 ~78°C 60 50 0°C __ Ethyl hexanoate 100 55 -78°C P 70 50 0°C tert-Butyl hexanoate 97 45 -78°C Ethyl nonanoate 100 50 -78°C _ Ethyl isobutyrate 97 75 -78°C 5 92 70 0°C 4: Ethyl isovalerate 97 60 -78°C Ethyl cyclohexanecarboxylate 95 70 -78°C Ethyl phenylacetate 98 60 -78°C Table 11. n-Butylation of Lithium tert-Butyl Acetate with n-Butyl Iodide. Temp. Solvent Alkylated Order of Product* Addition ~78°C THF .2% Halide to Enolate -45° THF 3.9% " " " —22°C THF 4.9% " " " 0°C THF 53.0% " " ” r.t. THF 55.0% " " ” r.t. DMSO/THF 71.0% " " " r.t. DMSO/THF 85.0% Enolate to Halide -45°C HMPA/THF 64.0% Halide to Enolate * All reaction times 1 hour. 43 When the LiICA base was used to generate lithium ester enolates at -78°C, the ester was 90-100% recovered on quenching, demonstrating that self-condensation does not occur. When the lithium ester enolate was prepared at 0°C, 60-92% of the ester was recovered indicating that some side reaction is taking place at the higher temperature. The deuterium incorporation was never quantitative and varied from 50 to 75%. The low deuterium incorporation cannot represent incomplete formation of the ester enolate (as in an equilibrium), since extensive condensation would have resulted. This was demonstrated by the addition of a slight excess of ethyl hexanoate to the enolate at -78°C. Quenching of this reaction after 5 minutes showed almost complete disappearance of ester. These results are consistant only with quantitative and irreversible formation of the ester enolate, consequently, the incom- plete deuterium incorporation must be a result of an unusual protonation mechanism involving the N-isopropylcyclohexylamine. 3‘ ?, 020 H., “Li 4/ H-é-g-OEt + DNC H(CH3)2 Although the ester enolates are more stable at low temperatures; alkylation in THF does not occur readily at temperatures below 0°C (see Table 11). The alkylation reaction can be enhanced either by raising the temperature or by using HMPA or DMSO as co-solvents. 44 Thus in THF lithium tert-butyl acetate reactions at -45°C with excess butyl iodide to yield only 3.9% of alkylated products, while addition of either DMSO or HMPA co-solvent at the same temperature yields 71% of the desired product. The yield of this reaction can be further improved to 85% by adding the ester enolate to a solution of butyl iodide in DMSO at room temperature. The steric hindrance and reactivity of the alkylating agent plays an important role in these alkylation reactions. The use of more reactive alkyl iodides gives superior yields, as demonstrated by the alkylation of t-butyl acetate with BuCl, BuBr and BuI, which gives respectively 5%, 75%, or 85% yields of t-butyl hexanoate (see Table 12). It has also been demonstrated that increasing steric hindrance in the alkyl halide reduces the yield of alkylated product. Thus when lithiwnt-butyl acetate was alkylated with the following alkylating agents the yield of product decreased in the designated order: isoamyl iodide > isobutyl iodide > isopropyl iodide. Since the alkyl iodides, of all the halides gave the best yields of alkylated product, an attempt was made to use alkyl tosylates as the alkylating agents. n-Butyl, n—benzyl, alkyl, octyl and n-propyl tosylate were synthesized and reacted with lithium t-butyl acetate under a wide variety of reaction conditions. No alkylated products or side products were ever obtained. In some cases, the starting ester was isolated but never in quantitative amounts. A.I. Meyers observed that the reaction of the lithium dihydro-l,3-oxazine anion with n-butyl tosylate resulted in the formation of the sulfone (43). 45 --- 666 6I66I6666666I63 666666 66666: 66666666666 66666 666 666 6I66I6666666I63I6I6666I636I6 666666 666666666 666666666 66666 --- 666 6I66I66666I6I6666I636I6 666666 666662 666666666 66666 --- 666 666I6366666I6I66 66I636I6 666666 66666: 666666666 66666-6666 --- 666 6A6I636666666I63 666666 666662 66666666666 66666-6666 66 666 666I366666I6I6N 66I63 666666 666666666 6666666 66666-6666 666 666 666I636666666I636 6666666 666666 6666666 66666-6666 666 6666 666I636666666I66I666I6 6666666 66666 6666666 66666-6666 666 666 666I636666666I63I6666I66 666666 6666666 6666666 66666-6666 666 666 66 :I636 66- 66 N35.. 666I63 666666 66666666 6666666 66666-6666 666 6666 666I a3 : :I63 I6 666666 66666-6 6666666 66666-6666 66 66 6; 36 M666 66 I636 I6 66666666 66666-6 6666666 66666-6666 66 666 666I a3 : NI3 6I6 6666666 66666-6 6666666 66666-6666 666 666 666I iv :6 3I 666666 66666-6 6666666 66666-6666 6:; 6:696 6I6 66626 66666666 6666666666 66666 666:6: 3:695 5.63 63265 .6366 65.26: 6o 65.666362 .~_. 023. 46 NJTCHZ + n-BuOTs ——9Af\i , _ .+ ,6 / N CH2-l Q CH3 + n-BuO L1 ’ 1 Li It is possible that the lithiumester enolate reacts in a similar fashion with tosylates, attacking the sulfur atom, however, this prod- uct was not isolated from the reaction mixture. The lithium enolates of acetate esters are particularly prone to self—condensation, and it is advantageous to employ the more hindered t-butyl ester in synthesis in order to reduce this unfavorable side reaction. The substituted ester acetates are less susceptible to the con- densation reaction; consequently, they give better yields of alkylated products, and even the ethyl ester can be satisfactorily alkylated with only a minimum of side reaction. The main by-product is unreacted starting material, with only a little B-keto ester being formed. Polyalkylation, which is so troublesome hithe alkylation of ketones, was not observed in the alkylation oflithium ester enolates. Thus the equilibrium reaction of thelithium ester enolate with al kylated product must be slow in relation to the rate of alkylation. 47 .+ CH3-fi—O-C(CH3)3 -L1 CH =8-0-C(CH ) + Bu-CH -fi-0-C(CH ) _§19!_3> + 2 3 3 2 3 3 .+ fast g'L‘ BuI Bu-CH= -o-C(CH3)3 \/ Bu-CHZ-g-O-C(CH + Lil 3)3 In general, alkylation oflithiwnester enolates is most successful when unbranched alkyl iodides are used as the alkylating agent and a THF/DMSO solvent system is employed. The alkylation of lithio ester enolates provides a convenient method of synthesizing a large variety of alkylated esters from readily available starting materials. EXPERIMENTAL SECTION 1. Materials Alkyl Halides. Commercial n-butyl bromide, n-butyl chloride and n-octyl iodide were used without further purification. Isobutyl iodide (31-32°C/38mm). isoamyl iodide (62°C/55mm) and isopropyl iodide (BB-89°C) were shaken repeatedly with a dilute solution of sodium thiosulfate dried over anhydrous calcium chloride and distilled prior to use. Allyl bromide (70°C) and benzyl bromide (110°C/l3mm) were purified by distillation and stored over a few drops of mercury in an amber bottle. Tosylates. The tosylates were prepared either by the method of Schleyer (44) from the alcohol and p-toluenesulfonyl chloride in a pyridine solution or by the method of Drahowzal and Klamann (45) from the alcohol and tosyl chloride in a 25% sodium hydroxide solution. The following tosylates were synthesized. n-Butyl tosylate 32-34°C/10mm; n30 1.5004, [lit. n30 1.5046] (45): n-propyl tosylate n30 1.5042 [lit. n30 1.5080] 20 (45); n—octyl tosylate nD 1.4931 [lit. 1120 D tosylate m.p. 54-56°C [lit. 56°C] (45); and allyl tosylate b.p. 137- 140°C/2mm, n30 1.5174 [lit. n30 1.5209] (45). 1.4946] (45); benzyl 48 49 Esters. All esters were commercially available except tert-butyl hexanoate and ethyl cyclohexanecarboxylate which were synthesized from readily available starting materials. All esters were dried over molecular sieves prior to use. 11. Preparation and Reactions of Ester Enolates The ester enolates were prepared and treated in a manner similar to that illustrated for tert-butyl acetate. All analyses were per- formed using a 1/4 inch by 6 ft. SE-30 column. All liquids were accurately measured and transferred with syringes. Preparation ofLithiwntert-Butyl Acetate. A 50 ml flask equipped as in Figure l was flame dried under a nitrogen steam and then cooled to -78°C in a dry-ice acetone bath. To this flask was added 5 mmoles of lithium N-isopr0pylcyclohexylamide in 5 ml of tetrahydrofuran, followed by .67 m1 (5 mmoles) of tert-butyl acetate. After 30 minutes of stirring at -78°C, the formation of the lithio ester enolate was considered complete. Recovery and Deuteration of the Esters. The ester enolate (10 mmoles), prepared as above either at -78°C or 0°C, was quenched with deuterium oxide. Extraction by 25 m1 of pentane, gave after drying over anhydrous magnesium sulfate a solution of the ester which was analyzed by GLC, using an inert internal standard. The amount of deuterium incorporation was determined by NMR examination of the recovered ester isolated by preparative GLC from the pentane 50 solution. In some cases the ester was examined in a crude state by simple evaporation of the solvent with identical results. Reaction of the Ester Enolate with Alkyl Halide. Two methods were used to react the enolate with an alkylating agent, as demonstrated below for the butylation ofliififiLm1tert-butyl acetate. (A). The ester enolate was prepared on a 5 mmole scale as described above. When desired, 2 m1 of DMSO was added before rapid addition of .855 ml (7.5 mmoles) of butyl iodide. The reaction mixture was warmed to room temperature, stirred for one hour and then quenched with 6N hydrochloric acid. Following the addition of 30 ml of pentane, and an internal standard, the organic layer was dried over anhydrous po- tassium carbonate and analyzed by GLC. (B). The ester enolate prepared at -78°C as previously described was added to a solution of .855 ml (7.5 mmoles) of butyl iodide in 5 ml of DMSO/THF solvent (3:2 ratio) at either 0°C or room temperature. The reaction was stirred for one hour and quenched with 6N hydrochloric acid, followed by the addition of 30 ml of pentane and an internal standard. The organic layer was then dried and analyzed by GLC. Product Analysis. The physical constants of five esters synthesized are compared with published values for these compounds in Table 13. All compounds in this Table were also analyzed by NMR and IR spectra and were in agreement with the proposed structure. The spectral data and physical constants for the remaining esters synthesized are also presented in this section. 51 0660060060 6663 6666.6 6m: 666.666-666 6666.6 6m: 6.666-666 -66666666-6.6 66I66 o 0 0660660060 666.663 6666.6 666 606666666 6666._ 666 66666666-66 -666666-6 66666-6666 66663 6666.6 6m: 666666.66-66 6666.6 6%: EE6666.66.66 6666666666-6 66666-6666 mpmocmxw; 66663 ---------- 666666.66-66 ---------- 666666.66-66 -666666-6 66666-6666 66663 6666.6 6m: 6666666666 6666.6 6%: 606.66.66-66 666666666 66666-6666 . .0.6 x0006 .0.0 66000066 666 66666 0>66006603 0>66066603 0606060666 6660056600xm .60666> 0606660666 :66: 0066050o 660066000 6060066 .mF 06006 52 tert-Butx] Hexanoate. This compound was compared with authentic materia1. I.r. 1750 cm- 1165 cm". 56-58°C/20mm. tert-Buty1 4-Methxlpentanoate. NMR (CCI 7.861(t, 2H), J=6.5cps; 8.431(m, 1H), one -CH2- 4)‘ from 8.501 to 8.701; 8.591(5, 9H); 9.121(d, 6H), J=5.Scps. I.r. 1 1 1730 cm- , 1155 cm- . tert-Buty1 3-Methy1butanoate. NMR (€014): 7.97r(d, 2H), J=1.75cps; -CH- 8.551(5, 9H); 9.071 (d, 6H), J=3.5cps. tert-Buty1 2-Methy1hexanoate. NMR (neat): 7.781(m, 1H), J=3.5cps; -CH20H20H2- from 8.301 to 8.931; 8.58r(s, 9H); 8.951(d, 3H). I.r. 1725 cm", 1160 cm". B.p. 84-86°C/20mm. Refractive index n30 1.4078. tert-Buty1 2,2-Dimethy1propanoate. NMR (neat): 7.101(5. 9H); 7.701(5, 9H); I.r. 1735 cm". 20 B.p. 130°C. Refractive index nD 1.3921. Ethy1 2-Isopropy1hexanoate. NMR (CC14): 5.931(q, 2H); 1.971(m, 1H); 8.701(t, 3H); -CH- and -CH2CHZCH2- from 8.33: to 9.001; three methy1 groups from: 9.00r 1 1 to 9.33:. I.r. 1730 cm' , 1165 cm' . B.p. 89-95°C/9mm. Mass spectrum m/e 172. Refractive index n30 1.4073. B.p. 1 53 Ethy1 Z-Methxjhexanoate. NMR (neat): 5.921(q, 2H), J=3.Scps; 7.681(m, 1H), J=3.Scps; - from 8.301 to 8.941; 8.801(t, 3H); 8.911(d, 3H); 9.121 1 1, 1150 cm". -CH2CH2CH2 (t, 3H), J=3.0cps. I.r. 1750 cm' , 1185 cm' B.p. 73- 74°C/18mm. Refractive Index n30 1.4078. PART III THE COUPLING 0F ENOLATE ANIONS with COPPER COMPOUNDS 54 INTRODUCTION AND HISTORY Over a hundred years ago, oxidative coup1ing became a useful synth- etic reaction with the discovery of the copper coup1ing of pheny1acet- y1ene by C. G1aser (48). .‘CsC-H 1) cu+, 11114011 > “C—CEC 2) air The oxidative coup1ing of acetylenes had been extensiveTy studied in the 1ast 30 years and is wide1y used in the synthesis of symmetrica1 and unsymmetrica1 di- and po1yacety1enes (49). The oxidative coup1ing reaction has by no means been restricted to acety1ene derivatives. and a search of the 1iterature revea1s a wea1th of organic anions that yie1d usefu1 dimerization products with cupric sa1ts, as exemp1ified by the 1ist in Figure 2. The 1ast of these reactions, the oxidative coup1ing of the pheny1 methy1 ketone eno1ate by cupric ha1ides gives on1y a 10w yie1d of dimeric product. CuC1 91-8-0112“ Li” 2 > 0-fi-0H20H2-fi-0 15% 55 56 ”x CH CH CH CH— (50) g-CH (51) Q 3 2 2 2 0/ 7 H2- (52) 70 U! (I) (53) (57).»3-(21‘12— (51) CH2= CH— (54) C=N— (55) - 1 -CH CH3'" (56) © 2— (51) Figure 2. The Coup1ing of Anions with Copper Compounds. 57 The coupling of enolate anions has apparently been sparsely in- vestigated. The lack of suitable methods for the quantitative generation of ketone enolates as well as the knowledge that ketones give a-halo products with the cupric halides normally used in coupling reactions (58), has limited the interest in such reactions. 0 l 1 x zcux2 + § . + 2CuX Nevertheless, w. Brackman and H.C. Volger discovered that ketone enolates could be oxidatively coupled by a solution of cupric nitrate in methanol,pyridine and triphenylphosphine (59). g NaOCH3 W8 EH3 CH3- -CH2-CH3 40°C 7\ CH3 -iH H- ”8 -3CH H3 The coupled products of the above reaction were isolated in low yield based on starting ketone and since sodium methoxide generates an equilibrium mixture of conjugate bases, this reaction is generally restricted to the production of 1,4-diketone products resulting from the more stable ketone enolates. These 1,4-diketone products are useful intermediates in the synthesis of symmetrical furans (60) and in the preparation of cyclopentenone derivatives (61). 58 j-CH R / ‘0H \ R-CH RCHz-g-CHZCH The chemical literature abounds with unique synthetic methods for the preparation of 1,4—dicarbonyl compounds. Only two methods exist that effect such a synthesis by direct a-coupling of carbonyl compounds: \/ 1 .' Di acetyl Peroxide However, the yield in these reactions usually is very low (62, 63). \w’ A general coupling reaction of enolate anions would provide the chemist with a useful synthetic method for the preparation of 1,4- dicarbonyl compounds. Since the ketone, acid, ester and amide enolates 59 were readily available to us by the use of hindered lithium dialkylamide bases, an investigation of the feasibility of oxidative coupling of these intermediates was undertaken. Although metal ions other than cupric salts are capable of one- electron oxidative coupling, the great success in the use of cupric salts in coupling reactions and the unique properties of cupric salts (49b) prompted an initial study of this metal salt in enolate coupling reactions. RESULTS A study of the cupric salt promoted oxidative coupling reactions of ester, acid, amide, ketone and nitrile anions has been conducted. All anions were prepared by the slow addition of the substrate to a THF solution of lithium N-isopropylcyclohexylamide at -78°C. Addition of cupric salts and subsequent warming of the reaction to room tem- perature resulted in a facile reaction with the copper (11) compounds, which in most cases, provide a satisfactory yield of the corresponding dimer. 1 CH(CH ) THF CH(CH ) H-c-x + Li-N’ 3 2 —————————) Li+ 'C-x + H- 3 2 ' ' o X= -g-OR, -g-OH, -CEN, -8-R 2Li-%-X + 2Cu(v)2 ) x-é-%-x + 2CuY + LiY 0 Y: 08-CH2CHZCHZCH3, or Br Esters. The results obtained in the dimerization of ester enolates with both copper (II) bromide and copper (II) valerate are illustrated in 60 61 60660606 00 6066: 060 30 0006560600 0603 60606» 660 000 .6.6 60 600; 0:0 003 0266 00660006 .06000066 0 6602066066066 60 0606x6e 0 60 00060600 666006: 0 66 66 06666006666606666.6.~ 6666066 0666006666066 .6666 om om 0600600666660660661601m.~ 6666066 0606060>0m6 Fxs6m ON 66 06666006666660266606 6666066 06666666066 66666 -- 66 0666600666666666-66.6.~ 6666066 0660:6x06 66666 06 66 06666006666660666-6.N _6;6066 0666066066 66666 66 66 066660066 666:66m-66 0666006 666:66m 66666060630 N63 06 0606» 06000066 60666 .6660m AHHV 606600 6:66: 6606mm 60 00660~660ewo .66 0—666 62 Table 14. with the exception of di-t-butyl succinate, the succinic esters were obtained as a mixture of racemic and meso isomers. The most favorable experimental conditions for the ester enolate dimerization reaction were determined by varying the reaction conditions and reagent concentrations in the coupling oflitflfi1m1t-butyl acetate. The addition of only one-half molar equivalent of copper (II) valerate resulted in reduced (50%) yield of the di-t-butyl succinate. Execution of the reaction entirely at -78°C or room temperature gave a 45% yield of succinate product. An inverse addition of'lithiunrt—butyl acetate to a THF solution of copper (II) valerate was also detrimental to the dimerization reaction (44% yield). Best results (95%) were obtained when one molar equivalent of the copper compound was added at -78°C to the enolate and the reaction subsequently warmed to room temperature. Increased amounts of unreacted starting material and a-bromoester were obtained from the copper (II) bromide coupling reactions as the steric bulk of the enolate increased. When copper (II) valerate was employed as the oxidant, the yield of coupled product decreased in most cases; however, no side products could be isolated which would account for this decrease in yield. Acids. The feasibility of preparing succinic acids by the coupling of the lithium acid enolates was investigated with octanoic and isobutyric acid substrates. Difficulties in separating the racemic and meso pro- ducts from each other as well as from the starting acid and cuprous and cupric salts, reduced the usefulness of this coupling reaction. Thus, from octanoic acid only 40% of crude product was isolated and after 63 repeated recrystalizations small quantities of the racemic and meso isomers were separated. The more hindered isobutyric acid enolate did not dimerize with CuBrz, and the starting material was isolated in 91% yield. Amides. Since primary and secondary amides couple at the nitrogen (52), only N,N-dialkyl amides are capable of successful dimerization to succinamide products. Moreover, the increased solubility of amides and succinamides in water hampers their isolation from the reaction mixture. In an attempt to synthesize the succinamide dimer of N,N- dimethylacetamide by the coupling of the enolate with copper (II) bromide, for example, neither starting material nor product could be isolated. It is clear then that facile isolation of the coupled pro- ducts necessitates the use of N,N-dialkylamides with sufficient solu- bility in non-polar solvents to facilitate their isolation. Thus N,N- di-n-butylacetamide readily coupled to yield dimerization products which were isolated in 75% yield by extraction of the aquous solution with ether. Recovered starting material (19%) was also obtained. These sterically hindered N,N-dialkylamides gave slightly lower yields than their corresponding esters. This was amply demonstrated in the coupling of N,N-diethylisobutyramide which gave no succinamide product; only starting material (86%) and N,N-diethyl methylacrylamide were isolated (12%) from the reaction mixture. In general, amides can be successfully coupled. However, the yields are slightly lower than those obtained with esters, and non- polar N,N-dialkylamides must be employed in order to facilitate separa- tion of the products from the reaction mixture. 64 Nitriles. In the copper (II) promoted coupling of ester, acid and amide enolates, the yield of dimerized product significantly decreases with increased steric bulk in thelitfifitm1enolate. Since the nitrile functional group is smaller than the previously noted carboxylic acid derivatives, coupling reactions of isobutyronitrile anion were inves- tigated in order to determine whether nitrile coup1ing might offer an alternative means of preparing highly substituted succinic acid der- ivatives. Two products were obtained, as shown in the following equation: CuBr2 + Li-C(CH3)2-C2N —————————]/ 1‘ ‘T H3 /CH3 H3 fH3 + NEC- -——-C£C2N + =C=N- -CsN H /H20 H3 H3 H3 \CH3 "head to head" H-C CH -g- -C CH -CEN (313(3)2 "head to tail” These products were difficult to separate, both from the copper salts and from the solvent. The "head-to-head" dimer was isolated in 21% yield and the "head-to-tail" dimer was isolated in 10% yield. Due to the low yields and difficulties in separating the products, nitrile coupling does not provide a useful synthesis of succinic acid derivatives. 65 Benzylonitrile. The dilithium dianion of benzylonitrile was prepared from butyl lithium in hexane (64). L1 D—CHz-Cm + n-Butyl-Li—T—HE—é H—é—CEN + n-butane Li This dianion was subsequently oxidized by the addition of c0pper (II) valerate in the presence of cyclohexene. It was hoped that reduction of the dianion which, in the presence of cyclohexene, would result in the formation of a cyclopropyl derivative. Li 1? .BL--¢-—-—CEN x's CU(V)2 663 é>=cul / L1 NE n-Bu' O C:N \C/ Q-CH-CEN . n- U However, only starting materials and l-cyano-l-phenyl-n-pentane (33%) were isolated from the reaction mixture. Ketones. Ketone enolates were quantitatively prepared by slow addition of the ketone to a THF solution oflithiumThisopropylcyclohexylamide at -78°. 66 THF ‘ '1 0 L1 .. - - \ - (CH3)3 C CH3 , (CH3)3 6%” 2 -78°C LiICA This method did not work well with dicyclopropyl ketone. An aldol condensation product was the major component of the reaction mixture. A THF/HMPA solvent system resulted in even greater yields of the aldol condensation side-product. >84 - .312; > >430 1) [>—8<] 2) NaHC03/ H20 DBQ + byxflfl <—J THF: 10.8% 44.0% HMPA/THF 10.5% 64.0% A slow rate of enolization of dicyclopropyl ketone is probably responsible for the formation of the aldol product. This repression of enolization in the ketone has been attributed to angle strain in the delocalization of the anion in the transition state (65). All other ketones studied were successfully enolized and were coupled with either CuBr2 or Cu(V)2. In the case of 3-methyl—2-butanone and 2—hexanone, a kinetically controlled mixture of enolates was obtained (3). 67 - Li+ - Li (CH3)2CH-8-CH3 L‘ICA (CH 3)2CH- -CH= -8- CH3 + (CH 3)2CH- -CH2 -2= -CH2 THF 6% 94% - Li+ - Li LiICA - CH3-( (CH H) -g- CH 3-———-f> CH3 CH2 CH2 CH= mg + CH3(CH2)3-8-CH2 30% or less 70% or more Reaction of these enolate mixtures with Cu(V)2 resulted in formation of the 1,4-diketone from coupling of the most abundant enolate and from cross coupling of the enolates. No product was isolated which would have required coupling of the least abundant enolate (see Table 15). From Table 15 it can be seen that the highest yields of 1,4- diketone were obtained from dimerization of sterically unhindered ketone enolates. A decrease in enolate reactivity and/or an increase in the steric bulk of the enolate resulted in lower yields of dimers, increased recovery of starting material and small amounts of unsaturated 1,4-diketones and a,B-unsaturated ketones. The results obtained in the coupling of cyclohexanone and 3,3- dimethyl-Z-butanone enolates with copper (II) bromide are presented in Table 16. When 3,3-dimethy1-2-butanone, was coupled considerable amounts of a-bromoketone and unsaturated 1,4-diketone were obtained. The dimerization of the more hindered cyclohexanone enolate resulted in none of the desired 1,4-diketone and even the side-products were obtained in lower yields. However, the coupling of the above ketone 68 66.66 I66:6-@-A66606uu666666-®-666666 66 66 6 I -®-666666 ----- 6:6666-®-666666 6: ----- - - -- . . - -6 6 - - . 6 - - - -- - - 66 66 6 w :6 66 w 6 6666 660 66 m 6 66.66 666-®-.m-666-®-666660-666 66.66 6 6I6-®-666600666 66.6 660-®-666666660 666-666666 66.6 6:6-@-6 -666-®-666I6-666666 66.66 6 666-®-666I6-666666 66.6 6:6-w-666:6-666660 6666666 -------------- 6666 6 6I6-@-0-666666 66.6 6:0-@-6-666666 66060066 60660 6060066 60660 0:060x 0666666m 60666066> AHHV 606600 :66: 6060606 60 006666660260 066 .m6 06066 69 .606606600 066 00 606666066 60 0066660600 0603 60606» 06066 6 .006066606 066360660 660666 06606666 66660666 66 06666 000 60 0066560600 663 0606» 6600606 066 .6606 6 606 0066666 066 .6.6 06 006666 60660606 .0006- 66 0666060 066 06 00006 663 NA>V60 066 6 -------------- 6666.660 66.66 _ o 6. 66» 66.66 66666066-®-666606u666 66.66 6 .1..60-®-66-666660 66.66 66666066-®-66-666660 :66 6 66060066 60660 6060066 60660 060606 06666660 .6.6_66600 66 66666 70 .06606666 66660666 66 06666 000 60 0066260600 0603 606066 666 .6606 6 LOV UwLqum Ucm .H.L Op flmmeL COwwummL 600”“! um mum—.OCw MS“ Ou. UflUUm mm? NLmDU m..:. k. 666 66 66oz 66 L“ . o o o o 666 666 666 66 66-666-®-666660 666660-6-®-66666-®-6-666660 6 -666-®-6-666660 666-®-6-666660 060606-0606010 0606066016.6 0066666666: 60260 060606 6.6666666 6660 666666 6666 6666666 66 666666666666 666 .66 66666 7l enolates with copper (II) valerate was more successful in obtaining the desired l,4-diketone product. When the 3,3-dimethyl-2-butanone enolate was coupled a 100% yield of the l,4-diketone was obtained and when the cyclohexanone enolate was coupled 70% of the desired product was isolated. Table l7 shows the results from the coupling of the lithium 3,3- dimethyl-Z-butanone enolate with copper (II) valerate under varying reaction conditions. It is of particular interest to note that l00% excess copper (II) valerate was not detrimental to the dimerization reaction. Since Cu(V)2 is a good radical scavanger, considerable quantities of a radical—copper (II) complex are probably formed when excess copper (II) valerate is employed as the oxidant (66). /N Cu \v 0 / (CH3)3-8-CH2- + Cu (V)2———9 (0193-ch 2 However, the high yield of l,4-diketone (98%) obtained indicates that radical coupling in this reaction is much faster than ligand or electron transfer, although the former reaction (shown below) has been observed in other systems (67). F?°CU(V)2——————%> R-O-g-n-Bu + CuIV Not unexpectedly, the ketone enolates did not couple at -78°C and therefore are less reactive than ester enolates to dimerization. 72 Table l7. Coupling of 3,3-Dimethyl—2-Butanone.* + 9 Cum2 E'L‘ (CH3)3—C- -CH2 2 (CH3)3- -C-CH2 Temp. moles moles 100% .0050 .005 -78°C to r.t. 98.0% .0100 .005 -78°C to r.t. 49.0% .0025 .005 -78°C to r.t. 2.3% .0050 .005 -78°C * Reaction run for one hour. 73 Copper (I) Enolate Complexes. Cyclohexanone and 3,3-dimethyl-2-butanone enolate-copper (I) complexes were prepared by the reaction of the enolate with copper (I) iodide. Li \L G ‘\ THF + CuI red solution It is known that alkyl copper (I) compounds will react with iodo- benzenes to yield phenyl alkanes (68). However, an attempt to prepare 2-phenylcyclohexanone by the reaction of the enolate-copper (I) complex. with iodobenzene proved fruitless and only starting material was iso- lated. The enolate-copper (1) complex did react with methyl iodide and gave results similiar to those obtained with the alkali metal enolates, although the yields were lower. uI CHI ' l 0 0 ‘\ 3 R.T. 14% 48% 2.3% The reaction of dialkyl cuprates with a-bromoketones has been shown to be useful in the preparation of alkyl ketone products (69). However, the reaction of enolate-copper (I) complexes with a-bromo- ketones gave only low yields of l,4-diketone products. 74 + - Li CuI (CH ) -C- =CH + (CH ) -C-g-CH BrT—HF—) (CH ) -c-9-CH CH -fi-C-(CH ) 3 3 2 3 3 2 r t 3 3 2 2 3 3 15% The same reaction carried out in the absence of copper (I) iodide gave ll% yield of l,4-diketone product, indicating that neither method is useful in the preparation of l,4-diketones. DISCUSSION Although the dimerization of carbanions has become a familiar reaction in organic synthesis, no great effort has been exerted in determining the detailed reaction mechanism. Nevertheless, sufficient examples can be found in the literature to give an insight into the reaction mechanism. A rationale based on available knowledge and experimental results is presented in this section. The initial step in the dimerization of enolate anions and car- banions is probably the substitution of a solvent ligand by the anion. -l S V n V Li+ 'l‘%-COR + Sn-Cd ——> \Cu/ + S 2 \V ROZC-?// \\V S n-l \\Cu } Roz-C- - + CuV + V" /\ ROZC-C V Carbanions are exceedingly soft (highly polarizable) ligands and the metal-ligand complex produced is expected to be very short lived and extremely unstable. A rapid one-electron redox reaction should take place resulting in the formation of a radical species and a copper (I) compound. 75 76 Subsequent dimerization of the radical will then give the desired dimeric product. - -- \ ..-..- - 2R20C% ,RZOHCOZR The side products normally obtained are the result of radical reactions with various reagents present in the reaction mixture before dimerization can take place. The experimental results suggest that in order to obtain a high yield of the desired dimer, the redox reaction must be of sufficient rapidity to produce high concentrations of the radical. The factors which effect the redox reaction and which would subsequently influence the yield of the reaction are: l) The reac- tivity and nature of the copper compound. 2) The reactivity of the enolate. 3) Steric hindrance. Copper (II) Salts. The polarizability of a ligand should greatly affect the reac- tivity of the metal ion in a redox reaction. It is expected that polarizable or "soft" ligands would reduce the reactivity of copper (II) ions while "hard" ligands would increase the reactivity of the metal ion in the redox reaction. Consequently, of the copper com- pounds employed in this work, copper (II) valerate was the most reactive and effective. The practicability of copper (II) bromide in enolate coupling was limited to acids and their derivatives, since its use in the coupling of ketones led to considerable side product for- mation. This failure is particularly stricking in the case of 3,3- dimethyl-Z-butanone. 77 - Li+ (CH3)3C- =CH2 + CuBr2 THF \/ (CH3)3C-8-CH2CH2-E-C(CH3)3 + (CH3)3-C-8—CH2-Br + (CH3)3-C-8-CH-8-C(CH3)3 52% 29% 10% The formation of a-bromoketone results from the reaction of the radical produced in the redox step with copper (II) bromide by a ligand trans- fer reaction (66,70). 0 \ - - - _ , (CH3)3 C C CH Br + CuBr + CuBr 2 (CH3)3-C-g-CH2- 2 The formation of both l,4-diketone and a-bromoketone in the above reaction gives support to the following radical mechanism for the a-halogenation of ketones and aldehydes. CuX R'-CH2-g-R $105 > R'-CH-8-R + HX + CuX CuX .ma > 6.6.8... . This mechanism has recently been supported by Walling (7l), how- ever alternate mechanismsikn‘this reaction have been proposed by Kosower (72) and Kochi (70). The formation of the unsaturated l,4-diketone can be most satis- factorily explained by a radical process. 78 CH 3 CH3>¢-8-CH2- + (CH3)3-C-8-CH2- —-9(CH3)3-c-g-CHzéH-E-c(CH3)3 CH 2 3 CuBr2 CH 683?. H—g-C CH (CH3)3-C-fi-CH=CH-E-C(CH3)3 + CuBr 49””’( 3)3 [EuBr ( 3)3 2 The above reaction has precedent in the work of Kochi who has demonstrated that alkyl radicals can be readily converted to alkenes by an electron-transfer oxidation with a carboxylatocopper (II) species (66). H + CuI('20c-C H9) C H - + Cu”(‘20C-C4H9)2——)CH CH CH=CH + c H co 4 4 9 3 2 2 4 9 2 The formation of unsaturated l,4-diketone in the coupling of ketones with CuBr2 indicates that the reduction with CuBr2 is slow and that both CuBr2 and product are present in solution while the radical is still being generated. The Reactivity and Nature of the Anion. The more reactive ester enolates and their derivatives are reduced at a faster rate, consequently, the less reactive copper (II) bromide oxidant can be used in this coupling reaction. With ketone enolates only copper (II) valerate gives adequate results, since the slowness of the redox reaction with CuBrzleads to considerable side-product formation. The use of copper (II) valerate in the dimerization of 79 ester enolates gives lower yields of dimer, however, the side-products from the use of this reagent have not been elucidated. Steric Hindrance in the Anion. In all the coupling reactions studied, the yield of coupled pro- duct decreases as the steric hindrance in the carbanion increases. This decrease in dimerized product is accompanied by a corresponding increase in recovered starting material, plus small amounts of a,8-unsaturated ketone or acid derivatives, and in the cases where copper bromide was employed, an increase in the a-bromo-derivative. The c.8-unsatured product formed in the dimerization reaction of hindered reactants is most likely the result of a radical disproportion- ation reaction. ~CH-8-CéégH2 H3 (CH3)2 2 (CH ) -CH-8-C‘/CH3 32 \CH 3 (CH3)2-CH-8-CH(CH3)2 The alternative formation of a,B-unsaturated products from the reaction of a radical intermediate with copper (II) salts seems unlikely due to the high activation energy expected for this reaction (59b). lil‘llillllll’l‘ll‘ 80 H CH Cu(V) éfiH 3 2 2 (CH3)2-CH-@-< > (CH3)2-CH-8— \c—écw2 H3 H3 8 W 66 (CH3)2- - + CuV + n-Bu-COZH CH3 This disporportionation reaction, however, cannot account for the large quantities of starting material isolated from the reaction mixture. Three explanations that can account for the recovery of starting material are: l) The recovered starting material may be the result of incomplete reaction. 2) The recovery of starting material could result from carbon-oxygen coupling of the ketone enolate as proposed by Brackman (59b). If carbon to oxygen radical coupling occurs, the subsequent hydrolysis of this unstable intermediate would result in the formation of starting material and a a-hydroxy product. The fact that a-hydroxy ketones were never isolated from the reaction mixture argues against this explanation, however. (CH -CH-g-C-0H(CH 3)2 3)2 (CH 3)2 + \l/ -CH-g-C(CH3)2-——o-E;Eé:H§)2 + 3 2 2 (CH3)2-CH-g-CH(CH3)2 81 3) Due to the slower reduction of hindered carbanions, the radicals formed should have sufficient time to react with solvent resulting in the formation of starting material. C /CH3 C 8 H ) + ' (CH3)2-CH- -c-\ + ——-9 (CH3)2- H- -CH(C 3 2 Z S. o 0 CH3 Products The THF radical formed by this solvent reaction would also consume cupric reagents and result in the depletion of the cupric compound and in unreacted enolate. In summary, cupric compounds are useful reagents for the formation of new carbon-carbon bonds. Highest yields are obtained when reactive and unhindered enolates are employed. The more reactive the enolate, the less side products are generated in the coupling reaction. The nature of the cupric reagent is also vital. EXPERIMENTAL SECTION 1. Materials l-Bromo-3,3-Dimethyl-2-Butanone. 3,3-Dimethyl-2-butanone (50 ml or 40 mmoles) was dissolved in 100 ml of ether and placed in a three necked flask equipped with a reflux condenser, addition funnel and magnetic stirring apparatus. To this solution, 21 ml (40 mmoles) of bromine was added over a 30 minute period. The reaction was quenched with ice-water, the ether washed twice with 50 ml of water and dried over anhydrous calcium sulfate. The solvent was evaporated and the product distilled at b.p. 69-7l°C/l3mm yielding 51% of the desired product (73) GLC analysis revealed only a trace of dibromination. Cuprous Iodide. Impure cuprous iodide (8.77 gms or .138 moles) was dissolved in a solution of 130 gms of potassium iodide in 100 ml of water. The solution was diluted with distilled water to precipitate the pure cuprous iodide; the product was collected on a sintered glass filtering funnel and was washed successively with ethyl alcohol, acetone and dry ether. The puri- fied white cuprous iodide was dried in vacuum over night and stored in an amber bottle (74). 82 83 Ketones. All ketones were readily available commercially and were dried over molecular sieves and used without further purification. Di- cyclopropyl ketone was impure and distilled (b.p. 66-67°C/l8mm) prior to use. Preparation of Copper (II) Valerate. This compound was prepared by the method of Lieben and Rossi (75). In a typical preparation, a solution of 23 ml (200 mmoles) of valeric acid in 100 ml of water was slowly added to a solution of 8.0 gms(200 mmoles) of sodium dydroxide in 100 ml of water. The sodium valerate solution thus prepared was added to a water solution of copper (II) sulfate over a fifteen minute period. An insoluble green precipitate was rapidly formed. The product was collected on a sintered filtering funnel, washed repeatedly with distilled water and dried under vacuum in a desiccator untilno further weight loss was observed (24 hours). The turquoise product weighed 23.9 gms (86.5%), and yields as high as 98% were occasionally obtained. Preparation of N,N-Dibutyl Acetamide. The amide was prepared by the slow addition of 35.5 ml (.5 moles) of acetyl chloride to an ether solution of 168 ml (1 mole) of dibutyl- amine kept at 0°C. The reaction mixture was warmed to 50-55°C for 30 minutes to complete the reaction. The amine hydrochloride which had precipitated was then filtered off, the ether layer washed with hydrochloric acid, and sodium bicarbonate and the organic layer was 84 dried over anhydrous potassium carbonate. After the evaporation of the ether solvent, the product was distilled at 66°C/2mm to yield 65.2 gms (76.2%) of very pure product. NMR (neat): 6.701(t, 4H), J=3.5cps; 8.031(5, 3H); 8.601(m, 8H); 9.08r(t, 6H). Preparation of N,N-Diethyl Isobutrate. The product was prepared by an identical procedure employed in the synthesis of N,N-dibutylacetamide. The reaction was run on a one-half mole scale and 33.417 gms (63%) of the desired product was isolated in pure form after distillation at 46.5-48°C/1mm. NMR (neat): 6.631 (q, 4H), J=4.0cps; 7.201(m, 1H), J=3.25cps; 8.831(t, 6H); 8.951(d, 6H), J=2.5cps. II. Coupling of Enolate Anions Esters. The ester enolates were coupled in a manner analogous to the dimerization of ethyl pr0pionate presented below. Dimerization of Ethyl Pr0pionate. A dry flask was equipped with a magnetic stirrer, septum inlet and mercury bubbler. The flask was flushed with nitrogen and 25 m1 of a 1.0M solution of lithium N-isopropylcyclohexylamide in tetra- hydrofuran was injected with a syringe. The flask was immersed in a dry-ice acetone bath and 2.6 gms(25 mmoles) of ethyl propionate was added dropwise. After 15 minutes, CuBr2 (6.7 gms, 36 mmoles) was added all at once through a powder funnel. The solution was stirred 85 for 15 minutes and then allowed to reach room temperature and stirred for an additional hour. Hydrochloric acid (15 ml or a 10% solution) was added together with 30 ml of pentane. The separated organic layer was dried over anhydrous magnesium sulfate and subjected to vacuum distillation to obtain 2.1 gms (75%) of a racemic and meso mixture of diethyl-2,3-dimethylsuccinate. B.p. 106-109°C/15mm [lit. 20 b.p. 108°C/15mm] (76), nD - 1.4262, NMR (CC14): 5.90r(q, 4H), J=3.5cps; 7.70r and 7.73T(V, 2H), 8.771(t, 6H), 8.911(d, 6H), J=3.5cps. Dimerization of Ethyl Isovalerate. Diethyl-2,3-isopropylsuccinate was isolated in 20% yield and MR showed a mixture of racemic and meso isomers. B.p. 79-83°C/.15mm; 20_ 1 1 l nD - 1.4371; i.r. 1725 cm- , 1158 cm- , 1185 cm' . Dimerization of Ethyl Hexanoate. Diethyl-2,3-di—n-butylsuccinate was obtained as a racemic and meso mixture as indicated by NMR (CC14): 5.901 and 5.951(q, 4H); 7.491(m, 2H); and 8.77r and 8.801(t, 6H), J=3.0cps; and 12H under triplet peak (-CH2CH CH 1 2 2-), 9.]3T(t, 6H), J=3.5cps. I.r. vc=o 1740 cm", 1178 cm' ; b.p. 90-95°C/.1mm [lit. 149°C/7mm] (77). Dimerization of Ethyl Isobutyrate. Diethyl tetramethylsuccinate was isolated in 25% yield. B.p. 115-117°C/12mm [lit. 122°C/15mm] (36a). n3°= 1.4357 [lit. n30 - 1.439] (78). NMR (CC14): 5.921(q, 4H), J=3.5cps; 8.761(t, 6H); 8.811(5, 12H). Ethyl 2-bromo-2-methylpropionate was isolated from the above reaction mixture and purified by GLC. NMR (CC14): 5.771(q, 2H), J=3.5cps; 8.101(5, 6H); 8.68r(t, 3H). 86 Dimerization of Ethyl Phenylacetate. Diethyl—2,3-diphenylsuccinate was isolated as a mixture of racemic and meso isomers. Repeated recrystallization from ethanol (95%) yielded the meso form M.P. l39-l4l°C [lit. 141°C] (79). NMR (CC14): 2.621(m, 10H); 5.751(5, 2H); 6.181(q, 4H), J=3.5cps; 9.13T(t, 5H). Dimerization of T-Butyl Acetate. Di-t-butyl succinate was obtained in yields as high as 95%. B.p. 100°C/1011m [lit. 115°C/14mm] (80). M.P. 34-35°C [lit. 36°C] (36a). NMR 1, 1115 cm“. (CC14): 7.6lr(s, 4H); 8.581(5, 18H). I.r. 1730 cm- Amides. The following lithium amide enolates were coupled with copper (II) bromide. Dimerization of N,N-Di-n-Butylacetamide. The enolate was prepared from 1.713 gms(10 mmoles) of acetamide in the usual manner, then brought to room temperature for 15 minutes before again being cooled to -78°C and 2.234 gms(10 mmoles) of copper (II) bromide added. The reaction was brought to room temperature, stirred 30 minutes, quenched with an ammonium chloride saturated ammonia solution and 25 m1 of pentane added. The pentane was dried over anhydrous potas- sium carbonate, treated with Dowex A-1, and evaporated under vacuum yielding 1.5991 gms (94%) of product. Analysis by NMR showed that the reaction mixture contained 79% of the desired coupled product and 14% starting material. NMR (CDC13): 6.671(t, 8H), J=3.5cps; 7.351(s,4H); 8.571(m, 6H); 9.03r(t, 12H), J=3.0cps. 87 Dimerization of N, N-Diethylisobutyramide. This reaction was run as above on a 10 mmole scale and yielded 1.4122 gms (98.5%) of recovered material which consisted of 86% starting material identified by NMR and 12% N, N-di-n-butyl methylacrylamide as indicated by vinyl peaks at 4.971 and 5.151 in the NMR. Carboxylic Acids. Two carboxylic acid enolates were coupled with copper (II) bromide and the results are presented below. Dimerization of Octanoic Acid. To a 150 m1 flask equipped as in Figure 1 was added a solution of 55 mmoles of LICA in 50 ml of THF and cooled to 0°C. The dilithium dianion of octanoic acid was then prepared by the slow addition of 3.96 ml (25 mmoles) of octanoic acid to the above solution of base and sub- sequently heated at 50-55°C for 1.5 hrs. A white jelly solution was formed when the reaction mixture was cooled to -78°C. Addition of copper (II) bromide (5.5843 gms or 25 mmoles) decomposed this white gel and resulted in the formation of a dark green solution. The reaction was then warmed to room temperature and stirred for .5 hrs. during which time a brown precipitate formed. The reaction was then quenched with 6N hydrochloric acid and the aquous layer extracted three times with 100 m1 of ether. The cuprous and cupric salts were difficult to separate from the products, however, best results were obtained by reduction of the cupric salts with sodium thiosulfate. Subsequent addition of a dilute solution of sodium thiocyanate resulted in the formation of a cuprous thiocyanate precipitate which was filtered off. The ether layer was dried over anhydrous sodium sulfate, and the 88 solvent evaporated yielding 1.45 gms (40%) of crude product. Repeated recrystallization from pentane resulted in the formation of .6047 gms (17%) of the diacid M.P. 144.5-145°C [lit. 144-144.5°C] (81). The pentane mother liquid was evaporated and the residue recrystallized from chloroform yielding .3110 gms (8.6%) of the lower melting isomer M.P. 91-96°C [lit. 95-96°C] (81). Examination of the reaction mixture by GLC gave three peaks. The first peak was octanoic acid identified by the retention time of an authentic sample and the second peak was 2-bromooctanoic acid. NMR (CDC13): -.531(s, 1H); 5.831(t, 1H), J=3.5 cps; 7.831(m, 2H); -CH2CH2CH2CH2- 1 from 8.331 to 9.001; 9.101(t, 3H), J=3.0cps. I.r. 3300 cm' to 2500 cm'](s), 1710 cm'1(s), 1425 cm-](m), and 1280 cm-](m). The third peak was identified as 2,3-di-n-hexyl- 1, 1225 cm“(m), succinic anhydride from the i.r. 1860 cm'](m), 1780 cm- and resulted from the dehydration of the diacid on the SE—30 chromato- graphy column. Dimerization of 3-Methy1propionoic Acid. The above procedure was repeated using 2.2035 gms(25 mmoles) of isobutyric acid. All other reagents and conditions were kept the same. After evaporation of the solvent and filtration of .0346 gms of an unidentified copper compound, only starting material (1.9582 gms or 91%) was isolated, NMR (CDC13): .731(s, 1H); 7.501(m, 1H), J=3.5cps; 8.901(d, 6H). Dimerization of Isobutyronitrile. In a 50 m1 flask equipped as in Figure 1, 15 mmoles of lithium N-isopropylcyclohexylamide in THF (15 ml) was prepared in the usual 89 manner. The reaction was cooled to -78°C and 1.34 ml (15 mmoles) of isobutyronitrile was slowly added. The reaction was brought to room temperature for 15 minutes and then cooled to -78°C before the rapid addition of 3.56 gms (16 mmoles) of copper (II) bromide. The reaction mixture was again brought to room temperature, stirred for one hour, quenched with GM hydrochloric acid and 30 m1 of ether was added. The organic layer was washed with an ammonium chloride saturated ammonia solution and dried over anhydrous sodium sulfate. Evaporation of the solvent yielded .3689 gms (35%) of a mixture of products. Addition of pentane resulted in the formation of a precipitate identified as (2-cyanoisopropyl)-isobutryamide. M.P. 101—103°C NMR (CDC13): 4.601 (s, 1H); 7.631(m, 1H); 8.301(5, 6H); 8.871(d, 6H), J=3.25cps. I.r. 3280 cm"(m), 2220 cm"(w), 1640 cm'](m), 1530 cm"(m). Evaporation of the pentane solvent resulted in the slow precipitation of 2,3-dicyano-2,3-dimethy1butane (.220 gms or 21%). M.P. 166-167°C [lit. l67-167.5°C] (82). NMR (CDC13): 8.471-singlet. Mass spectrum m/e = 136, 121, 69, 68. Benzylnitrile. ThelitflfiLmidianion of this nitrile was prepared by the addition of 6.75 ml (11.25 mmoles) of butyl lithium to a solution of .577 m1 (5 mmoles) of benzylonitrile and .51 m1 (5 mmoles) of cyclohexene in 7 m1 of THF at -78°C (64). The reaction was stirred for one hour at —78°C. When copper valerate (7.5 mmoles) was added, the reaction mixture turned black. The reaction was warmed to room temperature, stirred for one hour, quenched with 6N hydrochloric acid, and 30 m1 of pentane was added. The pentane layer was washed with an ammonium chloride 90 saturated ammonia solution and dried over anhydrous calcium sulfate. Evaporation of the solvent yielded .550 gms of product. NMR analysis of the product mixture showed 66% of starting material and a 33% yield of 1-cyano-lphenyl-n-pentane. Mass spectra m/e = 173. NMR (CC14): 2.671(5, 5H); 6.331(t, 1H), J=3.5cps; 2H multiplet from 7.921 to 8.381, 4H multiplet from 8.381 to 8.871; 9.101(t, 3H), J=3.0cps. Ketones. The following ketones were coupled with either Cu(V)2 or Cu(Br)2 in a manner similar to the coupling of ester enolates. A11 yields were obtained by GLC using an internal standard. Dimerization of 3,3—Dimethy1-2-Butanone. 2,2,7,7-Tetramethy1-3,6-octandione was obtained in 100% yield by enolate coupling with c0pper (II) valerate and in 52% yield by enolate coupling with copper (II) bromide. B.p. 58-64°C/.5mm. n30: §°= 1.4400] (62). M.P. 23-25°C [111. 22-25.5°C] (62), NMR (CC14) 8.871(s, 18H); 7.321(5, 4H). I.r. 1701 cm". 1.4396 [lit. n Dimerization of Cyclohexanone. 2.2'—Dicyclohexan-1,1'-dione was obtained in 70% yield from Cu(V)2 dimerization of the ketone enolate. M.P. 71.5-73°C [lit. 70-71°C] (62). NMR (CC14): the spectra consists of three broad peaks: from 6.901 to 7.501(2H); from 7.501 to 7.931(4H); and from 7.931 to 9.101(12H). From the coupling reaction of the ketone enolate with CuBr2 two products were obtained: 2-bromocyclohexanone (18%) NMR (C014): 91 6.301(m, 1H); two broad peaks from 6.871 to 7.931(4H) and from 7.931 to 9.001(4H). 2,2'—dicyclo-2-hexen-1,1'-dione (6%) NMR (CC14): broad peaks from 7.331 to 8.001(4H) and at 8.206(4H). I.r. 1810 cm", 1501 cm-1. Dimerization of 4-Methy1-2-Pentanone. The ketone enolate was dimerized with Cu(V)2 and the following product was obtained: 2,9-dimethyl-4,7-decadione in 66% yield (isolated 60%). B.p. 85-87°C/.5mm. NMR (CC14): 7.451(5, 4H); 7.731(4H), 7.901 (m, 2H); 9.101(d, 12H), J=3.5cps. I.r. 1705 cm“. Dimerization of 2,4-Dimethyl-3-Pentanone. 2,4,4,5,5,7-Hexamethy1-3,6-octandione was obtained in 17% yield from ketone enolate coupling with Cu(V)2. B.p. 93-98°C/lmm. NMR (CC14): 6.901(m, 2H), J=3.5cps; 8.801(s, 12H); 9.001(d, 12H). I.r. 1590 cm’1. 2,4-Dimethy1-1-penten-3-one was identified in the above reaction mixture. NMR showed H2C=Cfi protons at 4.071(s) and 4.231(s) and ‘I=C-CH3 methyl group at 8.131. Dimerization of 3-Pentanone. Two products were isolated from the enolate coupling reaction with Cu(V)2; 4,5-dimethy1-3,6-dione NMR (CC14): 7.131 and 9.171(q, 2H); 7.541(q, 4H); 9.051(t, 6H), J=3.75cps; 9.981(d, 6H), J=2.75cps. I.r. 1705 cm-]. This product was isolated as a mixture of racemic and meso isomers. 4,5-Dimethy1-4-octendi-3,6-one NMR (CC14): 7.481(q, 4H), 1 1 J=3.8cps; 8.181(t, 6H); 8.851(t, 6H). I.r. 1700 cm- , 1640 cm- . 92 Dimerization of 2-Hexanone. Two compounds were obtained from the Cu(V)2ketone enolate coupling reaction: 5,8-dodecandione (40.6%) M.P. 48-49°C. NMR (CC14): 7.431 (S, 4H); 7.621(t, 8H), J=3.5cp5; -CH2CH2- from 8.171 to 8.931; 9.101 (t, 6H), J=3.0cps. Mass spectrum m/e = 198. 3-Propyl-2,S-octandione (17.5%). GLC retention time and NMR were consistent with structure. Mass spectrum m/e = 198. Dimerization of Acetophenone. The following compound was obtained by Cu(V)2 coupling of the ketone enolate: diphenyl-l,4-butandione (27%) M.P. l46-7°C [lit. 145-147°C] (83). NMR (CDC13): aromatic protons (10H) from 1.731 to 2.731, 6.531(5, 4H). Aldol Condensation Reaction of Dicyclopropyl Ketone. In an attempt to prepare the lithium enolate of dicyclopropyl ketone the following reactions were performed. Dicyclopropyl ketone (.57 ml or 5 mmoles) was added to 5 mmoles of N-isopropylcyclohexylamide at -78°C in either 5 ml of THF or in 10 ml of a HMPA/THF (ratio 1:5) solution. The reaction was quenched with a saturated sodium bicarbonate solution, then 30 ml of pentane and .473 ml (2.5 mmoles) of diisopropylbenzene were added. The organic layer was dried and the solutions analyzed by GLC. When THF was used as the reaction solvent 10.8% of the starting ketone and 44.0% of the aldol product were obtained while when HMPA/THF solvent system was used 10.5% of the starting ketone and 64.0% of the aldol product were obtained. NMR (CC14): 6.021(s, 1H), and multiple 93 l 1 , 3000 cm- , 1 peaks from 8.051 to 9.971(19H). I.r. 3410 cm’ , 3080 cm- 1550 cm", Mass spectra m/e = 220, 179, 111, 110, 109, 59, 41. III. Reactions of Ketone Enolate-Copper (I) Complexes Cyclohexanone Enolate. The lithium cyclohexanone enolate was prepared by the addition of .52 m1 (5 mmoles) of cyclohexanone to a THF solution of LiICA (5 mmoles). Copper (I) iodide (.9600 gms or 5 mmoles) was added and the reaction brought to room temperature for 15 minutes during which time the capper (I) iodide dissolved producing a deep red solution. Three reactions were performed using the enolate-copper (1) complex: 1) To the reaction mixture was added 5 ml of dilute hydrochloric acid and 30 m1 of pentane. The organic layer was then separated and dried over anhydrous calcium chloride. Analysis by GLC showed 61% recovery of the starting ketone. 2) Addition of .34 m1 of methyl iodide to the enolate-copper (1) complex and reaction of the solution at room temperature for 1 hour resulted in the formation of 48% 2-methylcyclohexanone and 2.3% di- alkylated product. 3) The copper (I)-enolate complex was reacted with 1.02 gms(5 mmoles) of iodobenzene at room temperature for 6 hours. No 2-pheny1cyclohexanone formation was observed and only starting materials were isolated from the reaction mixture. 3,3-Dimethy1-2-Butanone Enolate. The lithium 3,3-dimethy1-2—butanone enolate was prepared as above for cyclohexanone from .623 m1 (5 mmoles) of the ketone at -78°C.and 94 Copper (I) iodide (.96 gms or 5 mmoles) was added to the THF solution of the ketone enolate and the reaction was brought to room temperature. l-Bromo-3,3—dimethyl-2—butanone (68 ml or 5 mmoles) was added to this solution all at once and the solution changed color from red to light green over a one hour period. Examination of the reaction mixture by GLC after the usual work-up, showed 15% yield of 2,2,7,7-tetramethyl- 3,6-octandione. The above reaction was repeated except the copper (I) iodide was excluded from the reaction mixture. The yield of l,4-diketone by GLC revealed a 10% yield of the desired product. BIBLIOGRAPHY 10. BIBLIOGRAPHY H. 0. House, "Modern Synthetic Reactions, " 2nd edition, W. A. Benjamin, Inc., New York, New York, 1972, chapter 9. C. A. Vanderwerf and L. F. Lemmerman, Organic Synthesis, Coll. Vol. 3, 44 (1955). H. 0. House, Rec. Chem. Progr., 55, 99 (1967). J. M. Conia, Rec. Chem. Progr., 53, 43 (1963). 82. H) Normant aid Th. Cuvigny, Organometal. Chem. Syn., _1_, 233, 1971 5. T). Cuvigny and H. Normant, Orggnometal. Chem. Syn., 1, 237, 1971 . L. Creger, J. Amer. Chem. Soc. , 55, 2500 (1967). L. Creger, J. Amer. Chem. Soc., 92, 1396, 1397 (1970). . L. Creger, Organic Synthesis, 5Q, 58 (1970). . Wittig and A. Hesse, Orggnicg§ynthesis, 59, 66 (1970). ancrm CDT??? I .0. House, L. J. Czuba, M. earl, and H. D. Olmstead, J. Org. Chem. , 55, 2324 (1969). 5. HL 0. House, M. Gall and H. D. Olmstead, J. Org, Chem, 55, 2361 1971 ?. CL R. Kruger and E. G. Rochow, J. Orggnometal. Chem., 1, 476 1964 b. D. H. Barton, R. H. Hesse, G. Tarzia and M. M. Pechet, Chem. Commun. ,1497 (1969L c. M. W. Rathke, J. Amer. Chem. Soc., 55, 3222 (1970). J. Deitch, M. S. Thesis, Michigan State University, East Lansing, Michigan, 1971. M. Larcheveque, Ann. Chim. (Paris), 5, 129 (1970). G. Stork, P. Rosen, N. Goldman, R. V. Coombs and J. Tsuji, Amer. Chem. Soc. , 87, 275 (1965). H. A. Smith, B. J. _L. Huff, W. J. Powers, III and D. Caine, . Org. Chem. 32, 285 (1967). L. E. Hightafier, L. R. Glasgow, R. M. Stone, 0. A. Albertson and A. Smith, J. Org_. Chem. 35, 188 (1970). ?. R). M. Coates and R. L. Sowerby, J. Amer. Chem. Soc., 55, 1027 1971 10.90 CaU'OJ 95 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 96 a. M. J. Weiss, R. E. Schaub, G. R. A11en, Jr., J. F. Po11etto, C. Pidacks, R. B. Conrow, and C. J. Coscia,'Tetrahedron, 59, 357 (1964), and references sighted therein. b. E. R. H. Jones and P. A. WiTSon, J. Chem. Soc., 2933 (1965). c. T. A. Spencer, R. W. Britton and D. 5. Watt, J. Amer. Chem. Soc., 52, 5727 (1967). P. A. Tarde11a, Tetrahedron Lett., 1117 (1969). U. Wannagat and H. Niederprfim, Chem. Ber., 99, 1540 (1961). S. W. Jacob, E. E. Rosenbaum and D. C. Wood, eds., "Dimethy1 Su1foxide," Marce1 Dekker, Inc., New York, N. Y., 1971. pg. 1. H. C. Brown and E. A. F1etcher, J. Amer. Chem. Soc., Z5, 2808 (1951). "Handbook of Chemistry and Physics," Forty-Ninth edition, The Rubber Pub1ishing Company, C1eve1and. L. F. Fieser and M. Fieser, "Reagents for Organic Synthesis," John Wi1ey and Sons, Inc., New York, N. Y., pg. 17. R. L. Shriner, Org, Reactiog§,, 1, 20 (1942). S. Reformatsky, Chem. Ber., 59, 1210 (1887). J. B. Conant and A. H. B1att, J. Amer. Chem, 509., 51, 1227 (1929). ?. D) Ivanov and A. Spassov, Bu11. Soc. Chim. Fr. [4], 99, 19 1931 . b. F. F. B1icke and H. Roffe1son, J. Chem. Soc., 19, 1730 (1952). M. A.)Spie1man and M. T. Schmidt, J. Amer. Chem. 599,,59, 2009 1937 . F. C. Frostick and C. R. Hauser, J. Amer. Chem. 500-: 21» 1350 (1949). 9. Ivgnov and A. Spassov, Bu11. Soc. Chim. Fr. [4], 99, 371, 375 1931 . D. Ivanov and N. I. Nico1ov, Bu11. Soc. Chim. Fr. [4], 51, 1331 (1932). a. H. L. Cohen and G. F. Wright, J. 0r . Chem., 18, 432 (1953). b D. Ivanov and A. Spassov, Bu11. Soc. Chim. FrT—I4], 99, 377 (1931). ?. Iv?nov and N. I. Noco1ov, Bu11. Soc. Chim. Fr.. [4], 51, 1325 1932 . E. F.)B1icke and H. Zinnes, J. Amer. Chem. Soc., 11, 4849, 5168 1955 . 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 97 P. E. Pfeffer and L. S. Si1bert, J. Org. Chem., 55, 262 (1970). ?. C) R. Hauser and W. J. Chambers, J. Amer. Chem. SOC-. Z§, 4942 1956 . b. A. J. Birch, J. Chem. SOC., 1551 (1950). ?. D) O. DePree and R. D. C1osson, J. Amer. Chem. Soc., 59, 2311 1958 . b. D. O. DePree, J. Amer. Chem. Soc., 55, 721 (1960). H. Hopff and H. Diethe1m, Justus Liebigs Ann. Chem., 591, 61 (1966). a W. H. G1aze, Organometa1. Chem. Rev., Section B, 45 (1970). b. M. M. Larcheueque, C. R.—Acad}. Sci., Paris,15er. C, 268, 640 (1969). S ( a. R. E. Pincock and J. H. Ro1ston, J. Org, Chem., 59, 2990 (1965). 5. F; W. Swamer and C. R. Hauser, J.7Amer. Chem. Soc., 55, 2647 1946 . a. B. E. Hudson, Jr. and C. R. Hauser, J. Amer. Chem. Soc., 55, 3156 (1941). 5. A5 Patchornik and M. A. Kraus, J. Amer. Chem. Soc., 95, 7587 1970 . ?. C5 R. Hauser, W. J. Chambers, J. Amer. Chem. Soc., 55, 3837 1956 . b. K. Sisido, Y. Kazama, H. Kodama, H. Nozaki, J. Amer. Chem. Soc., 51, 5817 (1959). %. R.)Hauser and W. J. Chambers, J. Amer. Chem. Soc., 15, 3837 1956 . C. R. Hauser, R. Levine and R. F. KibTer, J. Amer. Chem. Soc., 55, 26 (1946). M. W. Rathke and A. Lindert, J. Amer. Chem. Soc., 95, 2318 (1971). F. Camps, J. Caste11s, M. J. Ferrando and J. Font, Tetrahedron Lett., pg. 1713 (1971). J. I. Crow1ey and H. Papoport, J. Amer. Chem. Soc., 95, 6363 (1970). A. I. Meyers, A. Nabeya, H. W. Adickes and I. R. Po1itzer, J. Amer. Chem. Soc., 91, 763 (1969). Reference 17, pg. 1180. F. Drahowza1 and D. K1amann, Monatsh. Chem., 55, 452 (1951). 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 98 C. R. Hauser, B. F. Hudson, Jr., B. Abramovitch and J. C. Shivers, Organic Synthesis. C011. Vo1. 5,144 (1955). W. voyOoering and R. H. Haines, J. Amer. Chem. Soc. , 76, 485 1954 a. C. G1aser, Chem Ber. , 2, 422 (1869). b. C. G1aser, Ann. Cfiem. ,‘137, 154 (1870). a. H. G. Viehe, "Chemistry of Acety1enes," p. 597 (Marve1 Dekker, Inc., 1969). b. G. Eg1inton and W. McCrae, Adv. Org. Chem., 9, 225 (1963). E. Sake11arius and Th. Kyrimis, Chem. Ber., 55, 322 (1924). T. Kauffman and D. Berger, Chem. Ber., 101, 3022 (1968). T. Kauffman, G. Beissner, E. Koppe1man, D. Kuh1man, A. Schott and H. Schrecken, Angew.Chem. Internat. Edit., 7,131 (1968). E. J. Core and D. Seebach, Aggew. Chem. Internat. Edit., 9, 1075, 1077 (1965 . G. Wittig and G. Lehrman, Chem. Ber., 99, 875 (1957). T. Kauffman, J. A1brecht, D. Berger and J. Leg1er, Angew. Chem. Internat. Edit., 5, 633 (1967). H. Gi1man and H. H. Parker, 91‘Amer. Chem. Soc., 95, 2823 (1924). P. E. Fanta, Chem. Rev., 55, 139 (1946). Reference 17, pg. 161. ?. Brgckman and H. C. Vo1ger, Rec. Trav. Chim. Payg- Bas, 85, 446 1966 G. Now1in, J. Amer. Chem. Soc., 25, 5754 (1950). J. E. McMurry and T. E. G1ass, Tetrahedron Lett., 2575 (1971). M. S. Karasch, H. C. McBay and W. H. Urry, J. Amer. Chem. Soc. , 70,1269 (1948). S. G. P. P1ant, J. Chem. 595,, 1595 (1930). G. A. Gornowicz and R. West, J. Amer. Chem. Soc., 95, 1714 (1971). a. H. W. Amburn, K. C. Kauffman, and H. Shechter, J. Amer. Chem. Soc., 91,530 (1969). b. R. E. Dess ,Y. Okuzumi and A. Chen, J. Amer. Chem. Soc., 59, 2899 (1962. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 99 J. K.)Kochi and R. V. Subramanian, J. Amer. Chem. Soc., 51, 1508 1965 . J. K. Kochi, Tetrahedron, 15, 483 (1962). . E.)J. Corey and G. H. Posner, J. Amer. Chem. Soc., 52, 3911 1967 . E3 J. Corey and G. H. Posner, J. Amer. Chem. Soc., 55, 5615 1968 . . E.)Dubrois, C. Lion, and C. Mou1ineau, Tetrahedron Lett.177 1971 . . K. Kochi, J. Amer. Chem. Soc., 11, 5274 (1955). > C..- AL. AUAD’ . Lorenzini and C. Wa11ing, J. Org, Chem., 55, 4008 (1967). . M. Kosower, W. J. Co1e, G. S. Wu, 0. E. Cardy and G. Meister, . Org, Chem., 55, 630, 633 (1963). Co1onge and J. Grenet, Bu11. Soc. Chim. Fr., 1304 (1954). B. Kauffman and L. A. Teter, Inorg, Syn., 1, 9 (1963). Lieben and A. Rossi, Justus Liebigs Ann. Chem., 152, 66 (1871). ZDOQLM . Hucke1 and H. Mu11er, Chem. Ber., 51, 1981 (1931). m . B1aise and L. Picard, Ann. Chim. (Paris), [8], 55, 280 (1912). 7< . V. Auwers and B. Ottens, Chem. Ber., 57B, 446 (1924). D. Biquard, Ann. Chim. (Paris), [10], 19-20, 119 (1933). A. L. McC1oskey, G. S. Fonken, R. W. K1uiber and W. S. Johnson, Orggnic Synthesis, 51, 26 (1954). Barry)and Twomey, Pr. Irish Acad., 515, 148 (1947), C. A. 4454 1947 . A. F. Bicke1 and W. A. Wa1ters, Rec. Trav. Chim, Pays-Bas,.55, 1490 (1950). a. A. J. Ficini and J. P. Genet, Tetrahedron Lett., 1565 (1971). b. B. S. Majeti, Tetrahedron Lett., 2523 (1971). MMMMM 1|1ifi|111111111111[1111111111119111111111s