Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Thermolysis of unsymmetrically substituted vinylcyclopropane systems : site-selectivity in homo-[1,5]-sigmatropic… Maxwell, Anderson Richard 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1983_A1 M39.pdf [ 12.04MB ]
Metadata
JSON: 831-1.0059454.json
JSON-LD: 831-1.0059454-ld.json
RDF/XML (Pretty): 831-1.0059454-rdf.xml
RDF/JSON: 831-1.0059454-rdf.json
Turtle: 831-1.0059454-turtle.txt
N-Triples: 831-1.0059454-rdf-ntriples.txt
Original Record: 831-1.0059454-source.json
Full Text
831-1.0059454-fulltext.txt
Citation
831-1.0059454.ris

Full Text

THERMOLYSIS OF UNSYMMETRICALLY SUBSTITUTED VINYLCYCLOPROPANE SYSTEMS : SITE-SELECTIVITY IN HOMO-[l,5]-SIGMATROPIC HYDROGEN MIGRATIONS by ANDERSON RICHARD MAXWELL B.Sc, University of the West Indies, 1972 M.Sc, University of the West Indies, 1978 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1983 (c) Anderson Richard Maxwell In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of (-~H£f*\ iSTfcV The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6 (3/81) ABSTRACT This thesis describes the preparation and thermal re-arrangement, v i a homo-[l,5]-hydrogen migration, of a series of unsymmetrically substituted b i c y c l i c and t r i c y c l i c v i n y l c y c l o -propanes (65a)-(67a), (68b), (69), (70), (71a), (72a), (74)-(76) and (77a). In each of these compounds two hydrogen atoms one (H c) adjacent to the substituent and the other (Hj) more distant from i t - are suitably disposed to undergo homo-[1,5]-hydrogen migration. It was found that an oxygen substituent retarded the rate of homo-[l,5]-migration of the hydrogen atom (H^) adjacent to i t r e l a t i v e to that of the other more remote hydrogen atom (H T) The retarding e f f e c t appeared to be dependent on the structure of the oxy-substituted vinylcyclopropane as well as on the na-ture of the oxygen substituent. A methyl substituent exerted a n e g l i g i b l e , retarding effect on the rate of migration of while phenyl and t r i m e t h y l s i l y l groups accelerated the rate of migration of r e l a t i v e to that of Hj. The r e s u l t s are consistent with the proposal that in the t r a n s i t i o n state for the homo-[1,5]-hydrogen migration the car-bon centre bearing the migrating hydrogen atom develops a part-i a l negative charge. - i i i -- iv -TABLE OF CONTENTS Page TITLE PAGE i ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i LIST OF FIGURES v i i ACKNOWLEDGEMENTS v i i i INTRODUCTION I. General 1 II. [l,5]-Hydrogen S h i f t s - Scope and Mechanism .. 2 A. Dienyl [l,5]-Hydrogen S h i f t s 2 B. Homo-[l,5]-Hydrogen S h i f t s 5 III. Toposelectivity and [l,5]-Hydrogen S h i f t s . . 13 IV. [l,5]-Hydrogen S h i f t s in Synthesis 17 V. Substituent E f f e c t s in Sigmatropic Reactions . 22 A. General 22 B. Substituent E f f e c t s in [l,5]-Hydrogen Migrations 23 VI. The Problem 27 DISCUSSION I. Choice of Substrates and General Approaches to t h e i r Preparation . 34 II. Preparation of Starting Olefins 46 A. Preparation of Olefins (78) and '(.92) 46 - V -DISCUSSION (Cont'd.) B. Preparation of Ol e f i n (108) 48 C. Preparation of Compounds (84), (109) and (110) 50 D. Preparation of Compound (111) 53 E. Preparation of Compound (112) 56 F. Preparation of Compound (99) 59 III. Preparation of the Dibromocyclopropanes . . . 63 IV. Preparation of the Monobromocyclopropanes . . 82 V. Preparation of the Cyclopropyl Esters (89) and (96) and the Carboxylic Acid (151a) . . 94 VI. Preparation of the Cyclopropyl Esters (100) . 98 VII. Preparation of the Cyclopropyl Alcohols (90), (97), (103) and (153) 100 VIII. Preparation of the Cyclopropyl Aldehydes (91), (98), (104), (154), (156), (158), (160), (162), (164) 112 IX. Preparation of the Vinylcyclopropanes (65)-(77) 120 X. Thermal Rearrangement of the endo-Vinyl-cyclopropanes (65a)-(67a), (68b), (69), (70) (71a), (72a), (74)-(76), (77a) 143 XI. Discussion of Rearrangement Results 163 EXPERIMENTAL 190 BIBLIOGRAPHY 309 - v i -LIST OF TABLES Page TABLE I: [l,5]-Hydrogen S h i f t in some Acycl i c and C y c l i c Systems 4 TABLE II: Homo-[l,5]-Hydrogen Sh i f t in some V i n y l -cyclopropanes 8 TABLE I I I : Relative Rate of [l,5]-Hydrogen Migration in 7-Substituted Cycloheptatrienes 24 TABLE IV: Results of Homo-[l,5]-Rearrangement of the endo-Vinylcyclopropanes 164 - v i i -LIST OF FIGURES Page FIGURE 1: "Saddle" conformation of Bicyclo[6.1.0] non-2-ene (20) 11 FIGURE 2: The HOMO of the pentadienyl r a d i c a l 16 FIGURE 3: O r b i t a l picture of the "allowed" t r a n s i t i o n state for [l,5]-hydrogen migration in c i s -1, 3-pentadiene 17 FIGURE 4: a. The 400 MHz *H nmr spectrum of dibromo-cyclopropane (141) 78 b. The 400 MHz ^H nmr spectrum of (141) after i r r a d i a t i o n of the signal due to H^ , and H D . 79 c. The 400 MHz "^H nmr spectrum of (141) after i r r a d i a t i o n of the signal due to H„ . 80 FIGURE 5: a. The 400 MHz 1H nmr spectrum of (103a-b) in CDC13 104 b. The 400 MHz 1H nmr spectrum of (103a-b) in CDC13-D20 105 c. The 400 MHz 1H nmr spectrum of (103a-b) on i r r a d i a t i o n of the t r i p l e t (due to H.) at 60.72 106 d. The 400 MHz 1H nmr spectrum of (103a-b) on i r r a d i a t i o n of the multiplet at 63.74 -3.86 107 e. The 400 MHz % nmr spectrum of (103a-b) on i r r a d i a t i o n of the signal due to H R in the region 61.04-1.14 108 FIGURE 6: a. The o l e f i n i c region of the 80 MHz % nmr spectrum of (69) 137 b. The simulated "^H nmr spectrum of the o l e f i n i c protons of (69) using parameters obtained d i r e c t l y from the actual spectrum. . 138 c. The simulated ''"H nmr spectrum of the o l e f i n i c protons of (69) using calculated parameters 139 - v i i i -ACKNOWLEDGEMENTS During the past few years I have had the p r i v i l e g e to have worked in a very stimulating learning environment under the inspired guidance of Dr. Edward Piers. To him I wish to express my deepest appreciation. I would l i k e to thank a l l the many members of Dr. Pier s ' research group with whom I have had the pleasure of associat-ing for many helpful discussions and shared ideas. I owe a great debt of gratitude to my wife, Margaret, for her unstinted support and encouragement throughout the course of t h i s work and for the s k i l l , patience and love she displayed in the typing of t h i s manuscript. Many thanks are extended to my colleagues Miss Grace Jung and Messrs. Veranja Karunaratne, Peter Marrs and N e i l Moss for t h e i r c a r e f u l proofreading of t h i s thesis. F i n a n c i a l assistance from the University of B r i t i s h Columbia and MacMillan Bloedel Limited i s g r a t e f u l l y acknow-ledged. - i x -TO MARGARET -who gave so much so w i l l i n g l y . - x -"Obstacles are what you see when you take your eyes off the goal." Anon. - 1 -INTRODUCTION I. General Sigmatropic rearrangements constitute a class of re-actions that has been of great interest to both synthetic and t h e o r e t i c a l chemists for the past two (2) decades. These re-actions involve the concerted migration, under thermal or photo-chemical conditions, of a sigma bond from i t s i n i t i a l p o s ition in a molecular framework across a TT system to another loca t i o n . 1 2 3 3 The well known Claisen ' ' and Cope rearrangements ( i l l u s t r a t e d in equations 1 and 2, respectively) are formally designated [3,3]-sigmatropic s h i f t s . In these cases the mi-grating sigma bond i n i t i a l l y linked to atoms 1 and 1' i n the (1) (2) reactant molecule becomes bonded to atoms 3 and 3' in the pro-duct. Thus a [3,3]-shift has occurred in each case. Other types of sigmatropic reactions include the f a m i l i a r - 2 -[2,3]- and [ l , 3 ] - s h i f t s , examples of which are given i n equa-tions 3 4 and 4 5, and the much rarer [1,4]-, [3,4]-, [5,4]- and Q [5,5]-shifts, examples of which have been discussed elsewhere . Thermal [l,5]-sigmatropic hydrogen s h i f t s , of primary interest -OAc (4) in t h i s thesis, w i l l be discussed in the ensuing sections II . [l,5]-Hydrogen S h i f t s - Scope and Mechanism A. Dienyl [l,5]-Hydrogen S h i f t s Perhaps the f i r s t known example of a sigmatropic hydrogen s h i f t was the [l,7]-hydrogen migration involved i n the pre-c a l c i f e r o l (1) to c a l c i f e r o l (2) equilibrium . These [1,7]-hydrogen s h i f t s are, however, quite rare;hence [l,5]-hydrogen - 3 -g s h i f t s have since been more extensively studied . A l l the availa b l e r e s u l t s indicate that the [l,5]-hydrogen s h i f t i s a concerted, suprafacial process. Thus in the c y c l i c trans-i t i o n state of these reactions the migrating hydrogen i s simultaneously bonded across one face of the diene system to both of i t s termini. In a c y c l i c dienes t h i s requires a c i s re l a t i o n s h i p about one double bond between the a l k y l group bearing the migrating hydrogen and a v i n y l group. This c i s 9 re l a t i o n s h i p about the - bond (equation 6) in the s t a r t i n g compound i s translated into a c i s re l a t i o n s h i p about the newly formed - bond in the product. The example shown in equation 7^, in addition to again i l l u s t r a t i n g the above point, also demonstrates another stereochemical feature of these reactions. Where there i s an appreciable difference in s i z e between the substituents on , the t r a n s i t i o n state in which the larger substituent i s pseudo-equatorial i s favoured and t h i s arrangement leads to the predominant product. - 4 -Table I: [l,5]-Hydrogen S h i f t in Some Acy c l i c and C y c l i c  Systems. Temp. Range u # Studied, K AH , A S , (Reaction -Reaction Medium) kcal mol eu Ref (1) f| , f\ 475-518 31.5 -12 10 (g) (2) r v C D 2 C H 3 - n D 2 C H C H 2 458-478 (g) 35.4 - 4.9 11 (3) 279-298 ( cc i 4 ) 19.3 -10.8 12 R=-CH„ 402-499 32.3 - 4.9 14 • (g) R=-OMe 353-413 25.7 -15 15 (neat) 2 2 1 2 1 W , H 6 6 Thus the - C & double bond (equation 7) i s formed highly s t e r e o s e l e c t i v e l y . Table I l i s t s several examples of [l,5]-hydrogen s h i f t s in both c y c l i c and a c y c l i c systems along with the a c t i v a t i o n parameters. It w i l l be noted that [l,5]-hydrogen s h i f t s in the cyclopentadiene system are p a r t i c u l a r l y f a c i l e . The fact that the migrating hydrogen atom i s moving to a carbon atom adjacent to i t s o r i g i n a l l o cation in a system already con-strained into a r i n g must make attainment of the r e q u i s i t e t r a n s i t i o n state geometry comparatively easy. / ~ 3 ' ,5 (7) B. Homo-[l,5]-Hydrogen S h i f t s Cyclopropane rings can p a r t i c i p a t e in p e r i c y c l i c reaction i n place of double bonds. The well known Cope rearrangement of 3 1,2-divinylcyclopropanes as exemplified by the extremely 16 f a c i l e rearrangement of c i s -divinylcyclopropane (3) - 6 -i t s e l f demonstrates t h i s . The double bond-like character of (8) the cyclopropane r i n g i s further exemplified i n i t s a b i l i t y , when properly oriented, to s t a b i l i s e an adjacent p o s i t i v e charge and to enter into conjugation with adjacent multiple bonds. This has been explainedon the basis of the s i m i l a r i t y of bonding i n the carbon-carbon double bond and the cyclo-17 propane r i n g It i s no surprise, then, that the [l,5]-hydrogen s h i f t (properly c a l l e d the homo -[l,5]-hydrogen s h i f t ) in vinylcyclopropanes with a c i s r e l a t i o n s h i p between the v i n y l group and an a l k y l group on the cyclopropyl r i n g i s a well documented, r e l a t i v e l y f a c i l e reaction. The prototype of t h i s reaction, the rearrangement of c i s - l - m e t h y l - 2 - v i n y l -18 cyclopropane (5) has been studied i n d e t a i l by both Frey 19 and Roth . Both groups, on the basis of a consideration of C H : (9) - 7 -a c t i v a t i o n parameters, concluded that(5) must rearrange i n a concerted fashion to y i e l d the product, cis-1,4-hexadiene(6). Homo -[l,5]-hydrogen s h i f t s have since been reported in a great variety of vinylcyclopropane systems, c y c l i c and a c y c l i c . Table II l i s t s some representative examples along with corresponding a c t i v a t i o n parameters. Systems in 20 which a carbon-oxygen double bond replaces the v i n y l group (equation 10) and one example (equation 11) in which 21 the cyclopropane ri n g i s replaced by an epoxide have been studied. Recently, the f i r s t examples of the reaction i J (10) 7 8 ( I D 9 10 (12) 11 12 C H (13) H 13 14 - 8 -Table I I : Homo-[l,5l-Hydrogen S h i f t s in Some Vl n y l c y c l o -propanes. Temp. Range Studied, K # # (Reaction AHff _ ± AS # Reaction Medium) kcal mol eu Ref. (1) (2) CHO R=H R=-CH, 443-463 30.2 (g) 439-493 30.3 (g) 473-533 32.5 (g) 373-423 27.3 (cyclohexane) -11 -11.6 - 9.7 - 7.2 19 18 24 25 R=-H R=-OH R=-CH20H 473-513 27.2 33.0 28.6 -17.2 - 8.8 - 6.5 26 (4) . 424-443 \\ (neat) 31.4 - 6 27 - 9 -Table II (cont'd.) Temp. Range Studied, K # # (Reaction AH" _ ± AS Reaction Medium) kcal mol eu Ref, (5) R=-CH3 R=-OMe 424-443 (neat) 532-557 (neat) 31.4 36.1 - 6 -14 20 28 - 10 -22 involving cyclopropyl allenes were reported (equations 12 and 13). F i n a l l y , i n the only example of i t s kind to date, cis-2-ethynyl-l-methylcyclopropane (15) rearranged, a l b e i t at high temperature (350°C), to give, i n i t i a l l y , 1,2,5-hexa trien e (16) 2 3 . (14) The wealth of accumulated knowledge of homo-[l,5]-hydro-gen s h i f t s supports a s u p r a f a c i a l , concerted mechanism and points to r i g i d geometrical requirements in the c y c l i c t r a n s i -t i o n state for these reactions. For example, 1,l-dimethyl-2-vinylcyclopropane ( 1 7 ) rearranged to give only cis-2-methyl-1,4-hexadiene (18). This r e s u l t requires exclusive rearrange-ment v i a a chair type t r a n s i t i o n state since a boat t r a n s i -24 ti o n state would lead to trans-2-methyl-l,4-hexadiene (19) The geometrical requirements of the t r a n s i t i o n state could be quite demanding in some systems. For example, in order for the [1,5]-hydrogen migration to occur in b i c y c l o 27 [6.1.0]non-2-ene (20) , the eight-membered r i n g must adopt the somewhat unfavourable "saddle" conformation as - 11 -Scheme I / / i l l u s t r a t e d in F i g . l . The migrating hydrogen i s c l e a r l y trans r e l a t i v e to the cyclopropane methylene group ( F i g . l ) (15) and the reaction proceeds to give cis,cis-1,4-cyclononadiene(21) with complete s t e r e o s e l e c t i v i t y . 2 7 However, with a s u b s t i -tuent in the 9-endo p o s i t i o n , the saddle conformation i s - 12 -s u f f i c i e n t l y d e s t a b i l i s e d that the homo-[l,5]-hydrogen s h i f t i s not observed. ' . In a si m i l a r vein, the homo-[l,5]-hy-drogen s h i f t s i n bicyclo[4.1.0]hept-2-ene (22) and bi c y c l o [3.1.0]hex-2-ene (23) are not observed as the required con-formations are energet i c a l l y unattainable. As would be expected, substitution of an a l k y l group in the 7-endo pos i t i o n of the bicyclo[4.1.0]hept-2-ene system (equation 18) makes for f e a s i b l e homo-[l, 5]-hydrogen migration': The complete s t e r e o s e l e c t i v i t y of the transformations (equation 18) in the examples shown i s p a r t i c u l a r l y noteworthy. 26 [1,5]-H s h i f t (.16) (17) (18) a. R=H b. R=-0H c. R=-CH20H - 13 -It should be pointed out in closing t h i s b r i e f , s e l e c t i v e review that [l,5]-sigmatropic s h i f t s are by no means l i m i t e d to hydrogen atoms. Examples of [l,5]_migra-t i o n of a l k y l , methoxy, carbomethoxy and various organo-g m e t a l l i c groups, notably t r i m e t h y l s i l y l , are known Interestingly, the rate of [ l , 5 ] - t r i m e t h y l s i l y l migration in 5-trimethylsilylcyclopentadiene (26) has been estimated 6 31 to be 10 times that of [l,5]-hydrogen migration Hv/SiMe-6 ' 26 I I I . T oposelectivity and [1,5]-Hydrogen S h i f t s In theory, there are two (2) "t o p o l o g i c a l l y d i s t i n c t " paths that the migrating hydrogen can follow during the [1,5]-sigmatropic migration process. If the hydrogen migrates across one face of the diene, as in Scheme II, the process i s described as s u p r a f a c i a l . In the t r a n s i t i o n state the migrat-ing hydrogen i s simultaneously bonded to both termini of the diene system across one of i t s faces. The a l t e r n a t i v e a n t a r a f a c i a l migration process i s - 14 -Scheme II H i l l u s t r a t e d in Scheme I I I . In t h i s case, the migrating hydro-gen moves from one face at a given terminus of the diene sys-tem to the opposite face at the other. In the t r a n s i t i o n Scheme III state of t h i s kind of process,the migrating hydrogen i s simultaneously bonded to opposite faces of the diene system. 32 Roth and co-workers unambiguously proved that [1,5]-hydrogen migration was a su p r a f a c i a l , concerted process. They showed that the diene (27) rearranged to give only those two isomers (2j3) and (29) which would be expected i f hydrogen migration were indeed suprafacial (scheme IV). - 15 -Scheme IV H 27 27 The question of why these [l,5]-hydrogen s h i f t s and, indeed,other sigmatropic reactions are such highly toposelec-33 t i v e processes was f i r s t addressed by Woodward and Hoffmann They proposed that the lowest energy or "allowed" pathway in a concerted reaction i s one i n which o r b i t a l symmetry i s con-served in proceeding from reactant to product. This p r i n c i p l e of conservation of o r b i t a l symmetry has proved to be of great p r e d i c t i v e value and i s applicable to the whole spectrum of p e r i c y c l i c reactions. Its a p p l i c a t i o n to the [1,5]-sigmatropic - 16 -hydrogen s h i f t can be i l l u s t r a t e d in a stepwise manner as follows: 1. The reaction i s considered as proceeding through a t r a n s i t i o n state formed by a combination of the s p h e r i c a l l y symmetrical o r b i t a l of a hydrogen atom with the f i v e (5) electron pentadienyl r a d i c a l . + H (19) 2. The symmetry of the highest occupied o r b i t a l (HOMO) of the pentadienyl r a d i c a l i s determined. This o r b i t a l , shown in Fig.2, possesses two (2) nodes. t F i g . 2 3. Bonding of the hydrogen atom to the termini of the pentadienyl r a d i c a l , in the t r a n s i t i o n state, i s considered in terms of overlap between the o r b i t a l of the hydrogen atom and the HOMO of the pentadienyl r a d i c a l . It i s c l e a r that, since the lowest energy t r a n s i t i o n state w i l l r e s u l t from p o s i t i v e overlap, as i n Fig.3, supra-f a c i a l migration w i l l be the chosen pathway i n [l,5]-sigma-- 17 -t r o p i c s h i f t s . Fig.3 Similar application of the Woodward-Hoffmann p r i n c i p l e to other sigmatropic reactions has led to the understanding 33 of many previously puzzling features of these reactions IV. [1,5]-Hydrogen S h i f t s in Synthesis Unlike the Claisen and Cope rearrangements, [1,5]-hydrogen migrations have seen l i t t l e use in synthesis and have usually been considered as reactions of primary mechan-i s t i c or t h e o r e t i c a l i n t e r e s t . The fact i s , however, that these reactions do generate c i s double bonds substituted in an e n t i r e l y predictable manner. Obviously t h i s c a p a b i l i t y could be extremely useful in synthesis,especially of such natural products as insect j u v e n i l e hormones and pheromones. Corey" et a l . have reported such an application of the 34 [l , 5 ] - d i e n y l hydrogen s h i f t . The aldehyde (32) containing a t r i s u b s t i t u t e d double bond with (Z)-stereochemistry, was - 18 -required as an intermediate in connection with studies re-lated to the stereocontrolled synthesis of C 17 and C 18 OSiMe. OSIMG. 3 ,0 MG MG Me' MG (20) 30 31 32 Cecropia juvenile hormones. The desired double bond was formed with complete s t e r e o s e l e c t i v i t y in aldehyde precursor diene (30). It should be noted i n passing that hydrogen migration appears to have been completely s i t e - s e l e c t i v e in that only the product r e s u l t i n g from migration of a hydrogen atom adjacent to oxygen was formed. The authors, however, did not discuss whether the observed res u l t was due to kine-t i c or thermodynamic fa c t o r s . The homo-[1,5]-hydrogen s h i f t translates a c i s r e l a t i o n -ship between a v i n y l group and an a l k y l group on a cyclopropane r i n g into a c i s r e l a t i o n s h i p between a l k y l substituents on a double bond. This predictable t r a n s l a t i o n of stereochemistry can be exploited for the solution of synthetic problems. The aldehyde (34), a higher homolog of (^2) , was also required in connection with Corey's juvenile hormone studies referre d to e a r l i e r . In t h i s case the desired t r i s u b s t i t u t e d double bond with (Z)-stereochemistry was formed completely (31) v i a [1,5]-sigmatropic hydrogen rearrangement of the - 19 -ste r e o s e l e c t i v e l y on rearrangements of the alcohol (33_) (21) 33 34 Monti and co-workers sought to apply the homo-[1,5]-hydrogen s h i f t reaction to a more complex problem. They de-signed the system (3_5) in which homo-[l, 5]-hydrogen s h i f t i n -volving the cyclopropyl ketone moiety resulted in r i n g expan-sion and formation of the u s e f u l l y functionalized hydroazulene derivative (36). Elaboration of t h i s compound into pseudo-guiananes such as (±)-damsinic acid (3_7) and (± )-conf e r t i n (38) seems eminently f e a s i b l e s y n t h e t i c a l l y . (22) 37 38 - 20 -37 In connection with studies on the biosynthesis of the marasmane- and vellerane-type sesquiterpenes, i t was shown that i s o v e l l e r a l (39), possessing the marasmane skeleton, undergoes smooth rearrangement to the dienofuran (40). Since (40) seem-ed a reasonable precursor to various velleranes e.g. (£1), i t was suggested that the homo-[l,5]-hydrogen s h i f t b i o g e netically i n t e r r e l a t e s the marasmane and vellerane skeletons (Scheme V). Scheme V - 21 -F i n a l l y , the oft-synthesised cis-jasmone (45) was also prepared v i a an i n t e r e s t i n g application of the homo-38 [l,5]-hydrogen s h i f t . The vinylcyclopropane (43), pre-pared by heating the a l l y l i c alcohol (42) in the presence of dicyclohexylcarbodiimide, rearranged at 240°C to y i e l d the dienone (44) v i a a homo-[l,5]-hydrogen s h i f t . Subse-quent treatment of (44) with ( i ) methyllithium and ( i i ) chromic acid produced cis-jasmone (45) (Scheme VI). Scheme VI The above examples by no means define the l i m i t s o synthetic u t i l i t y of the [l,5]-hydrogen s h i f t reaction. 0 the contrary,they i l l u s t r a t e the synthetic p o t e n t i a l of th reaction in the hands of the creative chemist. - 22 -V. Substituent E f f e c t s in Sigmatropic Reactions A. General Over the years a tremendous amount of attention has been paid to substituent e f f e c t s in cycloaddition reactions -e s p e c i a l l y the Diels-Alder reaction because of i t s synthetic value. The substantial amount of data on substituent e f f e c t s on rate and s e l e c t i v i t y in these reactions i s impressively 39 explained by Frontier O r b i t a l Theory Very l i t t l e attention has been paid, on the other hand, to substituent e f f e c t s in sigmatropic reactions -with the possible exception of the Cope rearrangement. There i s thus a paucity of systematic data with which to test the emerging theories that treat substituent e f f e c t s 40 on sigmatropic reactions. Nevertheless, E p i o t i s and 41 Carpenter have applied t h e i r separate t h e o r e t i c a l t r e a t -ments to the available data with some promising r e s u l t s . The Linear Combination of Fragment Configurations (LCFC) approach of E p i o t i s has been shown to explain the trends in the data for [1,3]- and [3,3]-sigmatropic s h i f t s as well as for [1,5]-hydrogen s h i f t s in substituted cyclopentadienes and cycloheptatrienes. Migratory aptitudes in these two systems also seem understandable, for the most part, on the basis of the LCFC approach. Carpenter has applied his model to a wide range of p e r i c y c l i c reactions. He has demonstrated the a b i l i t y of the - 23 -model to predict substituent e f f e c t s on rate i n the Claisen 42 rearrangement and on the mechanism of alkoxide accelerated [l,3]-sigmatropic migration 4^. Detailed predictions 4"'" as to how substituents at various positions on the migrating frame-work should a f f e c t the rate of [l,5]-hydrogen migrations are yet to be tested, however. Thus, while the two (2) approaches mentioned above have had successes they s t i l l remain to be r i g o r o u s l y tested. Indeed, some "anomalies" point to the present need for r e f i n e -ment . B. Substituent E f f e c t s in [1,5]-Hydrogen Migrations There have been a few l i m i t e d studies on the e f f e c t of substituents on the rate of [l,5]-hydrogen migrations. A b r i e f review of the s i g n i f i c a n t r e s u l t s of these studies w i l l be presented here. 44 K l o o s t e r z i e l and co-workers studied a s e r i e s of 7-substituted cycloheptatrienes and generated the r e l a t i v e rate data l i s t e d i n Table I I I . In the 5-substituted cyclo-pentadienes i t has been shown that methyl s u b s t i t u t i o n 12 45 accelerates and halogen su b s t i t u t i o n retards the rate of [l,5]-hydrogen migration. One might be tempted to explain these findings i n terms of the development of electron de-45 f i c i e n t character ( p o s i t i v e charge) at the migrating centre - 24 -in the t r a n s i t i o n state of [l,5]-hydrogen migration. How-ever, the accelerating e f f e c t of the 7-cyano substituent in the cycloheptatriene series cautions strongly against such a simple picture of the t r a n s i t i o n state. Table III : Relative Rate of [l,5]-Hydrogen Migration 44 in 7-Substituted Cycloheptatrienes -Me 90 -SMe 40 -CN 60 -Ph 100 -OMe 100 -NMe0 700 a. The substituent in 7-substituted cyclohepta-triene . b. Relative rate based on one-half the rate i n cycloheptatriene. c. Determined in the range 100 - 140°C. - 25 -Acceleration of some sigmatropic reactions by both elec-tron-withdrawing and electron-donating substituents (non-Hammett 41 behaviour) has been predicted by Carpenter and he has pointed out that t h i s seems to be the case at least for the Claisen re-42 arrangement of 2-substituted v i n y l a l l y l ethers A study of substituent e f f e c t s on the homo-[1,5]-hydro-46 gen s h i f t reaction was car r i e d out by Meehan and co-workers on the vinylcyclopropane system (46). E l e c t r o n i c e f f e c t s close to the s i t e of a r r i v a l of the migrating hydrogen were systematically varied by changing the substituent X. The modest e f f e c t of substituents on the rate of t h i s reaction -(23) 46 47 (a) X = H- (e) X = 4-CN-(b) X = 4-OMe- (f) X = 4-N02-(c) X = 4-Me- (g) X = 3-C1-(d) X = 4-C1- (h) X = 3-CN-a factor of 2.5 rel a t e d the fastest (X=4-N02) and slowest (X=4-0Me) rates in the serie s - were nevertheless impressive since substituent e f f e c t s on rates in the related benzyl a l l y l - 26 -47 ether fragmentation reaction were p r a c t i c a l l y zero An acceptable c o r r e l a t i o n of the rate data was obtained with Hammett o values giving p = +0.38. The authors concluded that (1) the sign of p suggests the development of a p a r t i a l negative charge in the t r a n s i t i o n state and (2) the low p value and poor c o r r e l a t i o n of the rate data with a+ and a~ s u b s t i -tuent constants indicate that the charge developed i s at C 2 rather than of (46) (equation 23). The l a t t e r conclusion would seem to be in c o n f l i c t with that which follows from a consideration of the r e s u l t s report-48 ed by Jorgenson and Thacher (equation 24) 200°C (24) 48 49 (a) R=H, R'=-H (b) R=-C02Et, R'=-H (c) R=-C02Et, R'=-CH3 t, = t, = t, = 29 min 15 min 10 min The most ex c i t i n g development in the study of substituent e f f e c t s on the [l,5]-hydrogen s h i f t reaction was that recently reported by Paquette et a l . 4 9 . These workers found that [1,5]-hydrogen s h i f t s are subject to dramatic rate accelerations by - 27 -an alkoxide substituent i n much the same way as [3,3]- and [ l , 3]-sigmatropic s h i f t s are The rate enhancements were smaller than those reported f o r [3,3]-shifts but were, nevertheless, s i g n i f i c a n t being 10 - 10 at 25 C. For [ l , 5 ] -hydrogen migration i n the 2,4-cyclooctadienol system (50), the (.25) 50 51 (a) R=-0H (b) R=-0"K+ 5 o rate enhancement was 1.8 x 10 at 25 C. Unknown at the pre-sent time, however, i s the e f f e c t of alkoxide substitution on the rate of homo-[l,5]-migration of an adjacent hydrogen atom. It w i l l be clear from t h i s b r i e f review that substitutent e f f e c t s on [l,5]-hydrogen s h i f t s are far from being well under-stood. Further, more data i s a necessity for the informed development of the much needed t h e o r e t i c a l framework for ex-pl a i n i n g these substituent e f f e c t s . VI. The Problem The development i n our laboratory of a simple procedure 52 for the preparation of c y c l i c B-iodoenones coupled with r e l a t i v e l y easy access to lithium phenylthio(cyclopropyl) cup-- 28 -rate reagents has opened up a f a c i l e route to B-cyclopropyl-aB-unsaturated ketones. These compounds, suitably function-53-55 a l i z e d , are of notable synthetic u t i l i t y . For example, Cope rearrangement of the B-(2-vinylcyclopropyl)-aB-unsaturat-ed ketone (52) has been exploited as a key seven-membered ring 56 annulation step in an e f f i c i e n t synthesis of (±)-8-himacha-lene (54). The l e s s f u n c t i o n a l i z e d 3-(1-methylcyclopropyl) cyclohex-2-enone (55) on heating to 450°C rearranged by a [l,3]-sigmatro-p i c migration to give, a f t e r base treatment, the b i c y c l i c enone 57 (56), a key intermediate in a formal t o t a l synthesis of (±)-zizaene (5_7). The f e r t i l e imagination can conjure up many further 53 54 (26) (27) 55 56 57 synthetic a p p l i c a t i o n s of B-cyclopropyl-aB-unsaturated ketones involving the bond forming Cope and [l,3]-sigmatropic re-- 29 -arrangements. However, there lurks the p o t e n t i a l for yet another type of sigmatropic rearrangement - the homo-[l,5]-hydrogen s h i f t - along the framework of the 2-alkyl s u b s t i -tuted B-cyclopropyl-aB-unsaturated ketone system. This i s i l l u s t r a t e d by the hypothetical example shown below (equation 28) . The f i r s t compounds of t h i s type studied i n our lab-58 oratory were the t r i c y c l i c enone derivatives (58) . These compounds were shown to rearrange smoothly and i n high y i e l d to the B-substituted enones (59). Two aspects of these r e s u l t s sparked our immediate i n t e r e s t . F i r s t l y , the u s e f u l l y f u n c t i o n a l i z e d products (59) seemed as though they could be exploited i n natural product synthesis. For example, r i n g closure in compound (59a), - 30 -(29) 58 a n=l, 111=1, X=H b n=2, m=2, X=H c n=2, m=l, X=H 59 d n=l, m=2, X=H e n=2, m=2, X=-OMEM f n=l, m=2, X=-0MEM -MEM=-CH2OCH2CH2OCH3 obtained from rearrangement of 6-endo-(3-oxo-cyclopenten-l-yl) b i c y c l o [3.1.0]hexane (58a), could provide a f a c i l e entry into 2 6 the c i s , a n t i , c i s - t r i c y c l o [ 6 . 3 . 0 . 0 ' ] skeleton (see (60) ) 59 of the hirsutanes . Secondly, rearrangement of the compounds with the -OMEM substitutent (58e-f) occurred with s t r i k i n g s i t e - s e l e c t i v i t y - no products a r i s i n g from migration of the hydrogen adjacent to the oxygen substituent (H c in equation 29) were observed. This r e s u l t i s contrary to expectation based 44 15 on the observations of K l o o s t e r z i e l and Nozoe that a 7-me-thoxy substituent in cycloheptatriene accelerates the rate of the [1,5]-hydrogen s h i f t (Table III p. 24). Interestingly, 26 Ohloff had reported that rearrangement of the vinylcyclopro-pane (61) gave products (62) and (63) which could only have arisen from migration of H (Scheme VII). He also observed CL - 31 -(30) 58a 59a none of the product (64) that would have resulted from migra-ti o n of a hydrogen atom (H^) adjacent to the oxygen atom in (61) . 58 Apart from the i n i t i a l communication from our labora-26 tory and Ohloff's lone example , there seems to have been no other studies dealing with t h i s unusual e f f e c t of oxygen on the s i t e - s e l e c t i v i t y of homo-[1,5]-hydrogen s h i f t s . We were thus extremely keen to investigate t h i s phenomenon further. The work to be described in t h i s thesis was undertaken to provide answers to the following questions which arose from consideration of the rearrangements depicted i n equation 29. (1) Is the e f f e c t of the rather bulky and highly oxygenated -OMEM substitutent on the s i t e - s e l e c t i v i t y of the homo-[1,5]-- 32 -Scheme VII hydrogen s h i f t (equation 29) a general one exhibited by other oxy-substituents e.g. -OMe or -OH? (2) Does the conjugating keto-substituent in vinylcyclopro-panes such as (58e-f)influence the s i t e - s e l e c t i v i t y of the homo-[l,5]-hydrogen s h i f t reaction? (3) Is the orientation of the oxygen lone pairs with respect to the carbon-hydrogen bond being broken, a s i g n i f i c a n t factor in determining the s i t e - s e l e c t i v i t y of the rearrangement in the oxy-substituted cases? (4) How would substituents (-X in equation 29) other than oxygen a f f e c t the s i t e - s e l e c t i v i t y of the homo-[l,5]-hydrogen s h i f t reaction? - 33 -In the l i g h t of the dearth of information on the eff e c t of substituents on the rate of [l,5]-hydrogen migrations, i t appeared that some much needed r e l a t i v e rate data would res u l t from our studies. As well, with a better understanding of how substituents influence s i t e - s e l e c t i v i t y in the reaction, we hoped to be i n a better p o s i t i o n to achieve our long range goal - to exploit these homo-[l,5]-hydrogen s h i f t s in synthesis. - 34 -DISCUSSION I. Choice of Substrates and General Approaches to  t h e i r Preparation With the questions to be investigated in t h i s study rather c l e a r l y spelt out, the choice of substrates was general-ly quite straightforward. For example, compounds (65a),(66a) and (67a) v i r t u a l l y suggested themselves as appropriate sub-strates for probing questions (1) and (2). With respect to the selection of substrates (68b), (69) and (70), however, some b r i e f explanation i s in order. Com-pound (68b) i s s i m i l a r to compounds (65a), (66a) and (67a) in that in each case there are two hydrogen atoms - one adja-cent to an oxygen substituent and the other remote from i t -available for [l,5]-hydrogen migration. However, while these hydrogen atoms (Hj, H c) are appended to the f l e x i b l e methyl and methoxymethyl side chains in (68b), they are bonded d i r e c t -l y in (65a), (66a) and (67a) to a more r i g i d r i n g system. Thus, compound (68b) was an appropriate substrate for t e s t i n g the p o s s i b i l i t y that the degree of f l e x i b i l i t y of the groups bearing the migrating hydrogen atoms may be a factor in deter-mining the s i t e - s e l e c t i v i t y of the [1,5]-hydrogen s h i f t re-action in the oxygen substituted vinylcyclopropane systems. In both compounds (69) and (70) because an oxygen atom has been suitably incorporated into the molecular skeleton, - 35 -one can determine the r e l a t i v e orientation between the oxygen lone p a i r ( s ) and the carbon-hydrogen bond (C-H^) that has to be broken in the [1,5]-hydrogen migration reaction. This 6 0 stereoelectronic factor has been shown to be of c r i t i c a l importance i n a number of reactions (involving h e t e r o l y t i c and homolytic C-H bond cleavage),and we were interested in finding out whether i t was of any importance in [1,5]-sigmatropic hy-drogen migrations. 65a 66a R=-Me 68b 67a R=-MEM 69 70 An investigation of substituent e f f e c t s on the s i t e -s e l e c t i v i t y of the homo-[1,5]-sigmatropic s h i f t (question (4)) was ca r r i e d out with the set of compounds (71a), (72a), (74)-(76) and (77a) l i s t e d below. While any number of substituents - 36 -can be envisaged, an ide a l l i s t would include substituents covering a wide spectrum of e l e c t r o n i c properties. The l i s t of substrates examined in t h i s study, though f a r from ideal in the above sense, represented the best compromise among the need for a set of substituents with a wide range of elec t r o n i c properties, ease of preparation of the corresponding substrates and time constraints. There i s l i t t l e doubt that the r e s u l t s obtained with these substrates (71a), (72a), (74)-(76) and (77a) constitute a valuable, i n i t i a l contribution to the area of substituent e f f e c t s on s i t e - s e l e c t i v i t y in [l,5]-hydrogen migrations. Work continuing in our laboratory i s intended to extend the l i s t of substrates studied so as to provide a more complete picture of these substituent e f f e c t s . - 37 -In the previous study"" done i n our laboratory the desired vinylcyclopropanes, for example, (58c), were prepared by a synthetic scheme in which the corresponding dibromocyclo-propanes were key intermediates. Dibromocyclopropanes (of increasing importance in synthesis ) are p a r t i c u l a r l y attrac-t i v e intermediates i n the preparation of substituted cyclopro-panes. They are e a s i l y obtainable from the appropriate ole-~. 62, 63 , , , .. , . ,64,65,66 . ,, f i n s ' and can be s e l e c t i v e l y reduced ' ' to the corresponding monobromides which undergo f a c i l e lithium -bro-67 mine exchange to y i e l d the rather v e r s a t i l e l i t h i o c y c l o p r o -panes (Scheme VIII ) . The l a t t e r synthetic species can be ex-Scheme VIII p l o i t e d to obtain vinylcyclopropanes substituted in desired fashion on the v i n y l group (Scheme IX). For the preparation of the vinylcyclopropanes (65a) and (71a), i n which the v i n y l group i s incorporated into the cyclo-pentenone moiety, a preparative scheme very s i m i l a r to the one 58 already u t i l i z e d for the preparation of compounds (58a-f) - 38 -Scheme IX was envisaged (Schemes X and XI). Thus,dibromocarbene addi-ti o n to the o l e f i n s (78) and (84) should lead to the dibromo-cyclopropanes (79) and (85),respectively. Reduction of these 64—66 dibromocyclopropanes by any of the several methods ~ a v a i l -able should y i e l d the corresponding monobromide mixtures (80) and (86). Lithium-bromine exchange should then res u l t i n the formation of the lithiocyclopropanes (81) and (J37), which on CO treatment with phenylthiocopper followed by 3-iodo-2-cyclo-52 penten-l-one (83) would be expected to y i e l d the enones (65a-b) and (71a-d) v i a the corresponding phenylthiocuprates (82) and (88). Separation of the mixtures (65a-b) and (71a-d) would then y i e l d the desired endo-enones (65a) and (71a). - 39 -Scheme X 88 R=-Cu(SPh)Li 71a-d 71a - 40 -Scheme XII OR 81 R=-Me,R'=-Li 95 R=-MEM,R'=-Li 89 R=-Me,R'=-C02Me 96 R=-MEM,R"=-C02Me - 41 -For the preparation of the vinylcyclopropanes (66)-(70) and (72)-(77) (no substituent on the v i n y l group), i t appeared that the appropriate lithiocyclopropanes would again be s u i t -able intermediates. In the case of the pairs (65a) and (66a) and (71a) and (72a) u t i l i z i n g the same cyclopropyllithium intermediates to gain access to both compounds in one pair would be p a r t i c u l a r l y advantageous. Scheme XII i l l u s t r a t e s the synthetic approach envisaged for the preparation of the vinylcyclopropanes (66a) and (67a). Treatment of the appropriate lithiocyclopropanes, (81) or (95), obtained from the corresponding dibromocyclopropanes, (79) or (93), by reduction and lithium-bromine exchange, with methyl chloroformate would be expected to y i e l d the cyclopropyl esters, (89) or (96). Subjection of these esters, (89) or (96), to a sequence of reactions involving successively ( i ) reduction ( i i ) oxidation and ( i i i ) Wittig methylenation followed by separation of the o l e f i n mixture, (66a-b) or (67a-b), should then y i e l d the desired endo-vinylcyclopropane, (66a) or (67a). In considering the preparation of the vinylcyclopropane (68b), i t occurred to us that preparation of the esters (100) d i r e c t l y from the o l e f i n (99) by copper catalysed addition of ethyl diazoacetate ' would be a considerable improvement over the al t e r n a t i v e three step process shown in Scheme XIII. It was expected that conversion of the esters (100) into a mixture of the vinylcyclopropanes (68a-b), followed by separa-t i o n of the l a t t e r to obtain (68b), would follow the route - 42 -Scheme XIII OMG OMe 101 102 outlined i n Scheme XIV. It should be pointed out at t h i s juncture that ethyl diazoacetate addition to the o l e f i n s (78) and (92), and indeed to the other c y c l i c o l e f i n s that would be required as s t a r t i n g materials in t h i s study, would not constitute a p r a c t i c a l approach to the desired cyclopropyl esters. In these cases, unfortunately, the thermodynamically more stable exo-cyclopro-- 43 -Scheme XIV 68a-b 68b pyl esters l a r g e l y predominate in the products over the less stable endo-esters which would be required for eventual con-version to the desired endo-vinylcyclopropanes. In the p a r t i -cular case of the tetrasubstituted o l e f i n (99), i t appeared that diazoester addition would be quite useful since the de-s i r e d exo-cyclopropyl ester was expected to predominate s l i g h t -ly in the product. The reader w i l l have noted that the synthetic sequences outlined for the preparation of the vinylcyclopropanes (66a) and (67a), were quite lengthy (six steps not including the preparation of the s t a r t i n g o l e f i n s (78) and (92) ). On con-sideration of the p o s s i b i l i t i e s for shortening these sequences, 72 i t appeared that d i r e c t formylation of the cyclopropyllithium - 44 -intermediates would be a practicable way of achieving t h i s 73 goal. Thus , i t was indeed encouraging to f i n d that Kobrich had accomplished the transformation of the carbenoid (105) d i r e c t l y to the a-bromoaldehyde (106) using N-methylformani-l i d e as the formylating agent. The aldehyde (106) was in fact further converted into the vinylcyclopropane (107) (Scheme XV). Scheme XV Li ¥ e CHO , C H - C H 2 PhNCHO Br Ph3P=CH2 I 105 106 107 Scheme XVI outlines in generalised form the synthetic sequence that we hoped to u t i l i z e for the preparation of the vinylcyclopropanes (69), (70), (73)-(76) and (77a). This se-quence in practice quite e f f i c i e n t l y generated the required vinylcyclopropanes. It w i l l have been clear from the foregoing discussion that the projected syntheses of the vinylcyclopropanes (65)-(77) must begin with the corresponding c y c l i c o l e f i n s . These have been l i s t e d below and t h e i r preparation w i l l be described in the following section. The o l e f i n (108) was chosen as a - 45 -Scheme XVI convenient s t a r t i n g material that would lead v i a compound (73) to the desired vinylcyclopropanes (74) and (75) (Scheme XVII). Scheme XVII 108 73 74 75 - 46 -II. Preparation of Starting Olefins OMG 78 R=-Me 92 R=-MEM 84 X=-CH3 108 X=-0THP 109 X=-Ph 110 X=-SiMe. 99 111 112 74 58 A. Preparation of Olefins (78) and (92) Both o l e f i n s ( 7 8 ) 7 4 and (92)°° were obtained through a 75 common intermediate, 3-chlorocyclopentene (113) . This com-pound was prepared by bubbling dry hydrogen chloride gas into cold (-78°C) cyclopentadiene freshly obtained from the thermal cracking of dicyclopentadiene (see Scheme XVII). Following 58 - 47 -74 the method of Alder and Flock , hydrolysis of d i s t i l l e d aqueous s l u r r y of sodium bicarbonate produced 2-cyclopenten-74 l - o l (114) in 55% y i e l d . Also obtained was a substantial quantity of a higher b o i l i n g byproduct, presumably (115) which 74 reportedly was formed on treatment of crude hydrogen chloride containing 3-chlorocyclopentene (113) with sodium bicarbonate solution. The alcohol (114), previously prepared in our laboratory, exhibited the expected spectral properties including a band at 3300 cm-"1" in the i r spectrum. Treatment of the alcohol (114) Scheme XVII 3-chlorocyclopentene (113) by s t i r r i n g (at 0°C) with an 115 CI OH ONEM 113 114 92 (31) Ct OMG 113 78 - 48 -with B-methoxyethoxymethyl chloride in the presence of d i i s o -propylethylamine as described by Corey and co-workers y i e l d -58 ed the required 3-methoxyethoxymethyl (MEM) ether (92) in 80% y i e l d . Compound (92), also previously prepared in our laboratory, showed the expected i r and "^H nmr spectral charac-t e r i s t i c s . The chloride (113) was converted d i r e c t l y into the meth-74 oxy ether (78) , as reported by Alder and Flock, by s t i r r i n g with an i c e - c o l d s l u r r y of sodium bicarbonate in anhydrous methanol. The i r and "^H nmr spectral data (the l a t t e r to the best of our knowledge previously unreported) of compound (78) were consistent with expectations. Its **"H nmr spectrum, for example,showed a methoxy sin g l e t at 63.34, a one proton multi-plet at 64.35-4.65, assigned to H c, and a two proton o l e f i n i c multiplet at 65.83-6.15. B. Preparation of O l e f i n (108) Synthesis of compound (108) required the p r i o r prepara-77 t i o n of 2-cyclohexen-l-ol (116) . This material was pre-pared by reduction of 2-cyclohexen-l-one with diisobutylalumi-78 1 num hydride (Dibal) and gave a H nmr spectrum that was 79 i d e n t i c a l with the published spectrum . None of the saturated alcohol (117) could be detected by glc or nmr analysis of the alcohol (116). Tetrahydropyranylation of t h i s material accord-- 49 -80 ing to the method of Grieco afforded the unsaturated t e t r a -hydropyranyl (THP) ether (108) i n good y i e l d (94%). Compound (108) showed no 0-H stretch in i t s i r spectrum but exhibited a weak band at 1640 cm-''" and strong bands i n the 990 - 1160 cm-"'" region. Its "'"H nmr spectrum showed a l l the Scheme XVIII 117 signals expected for t h i s a l l y l i c tetrahydropyranyl ether showing,for example, a one proton multiplet at 64.50-5.00 assigned to the methine proton (Hg) on the tetrahydropyran rin g and a two proton multiplet i n the o l e f i n i c region at 65.60-6.02. - 50 -C. Preparation of Compounds ( 8 4 ) 8 1 , ( 1 0 9 ) 8 1 , 8 2 and ( I I P ) 8 3 The preparation of a l l three o l e f i n s proceeded from 3-bromocyclohexene (119) which was prepared by dibenzoyl peroxide 84 i n i t i a t e d a l l y l i c bromination of cyclohexene (118) NBS • • • (32) CC1 118 R l R l R9 Although both (84) and (109) ' had been prepared in reasonable y i e l d by the d i r e c t coupling of methyl and phenyl Grignard reagents, respectively, with 3-bromocyclohexene (119), i t was f e l t that the reaction could be greatly improved by the use of a copper c a t a l y s t . Indeed, addition of 3-bromocyclo-hexene to an ether s l u r r y of phenylmagnesium bromide and cup-85 o rous bromide - dimethyl sulphide (10 mole per cent) at 0 C resulted, a f t e r vigorous reaction, in the i s o l a t i o n of a 91% R l R9 y i e l d of 3-phenylcyclohexene (109) ' . This material showed the i r absorptions (3040, 1590, 755 —1 86 and 750 cm ) expected for a monosubstituted aromatic compound. Its ''"H nmr spectrum was consistent with i t s structure showing, for example, a one proton multiplet at 63.28-3.54 assigned to the a l l y l i c methine proton (H c), a two proton o l e f i n i c multiplet (65.62-6.06) and a f i v e proton multiplet in the aromatic region - 51 -(67.10-7.44). PhMgBr, CuBr.Me2S ether (33) 109 85 Cuprous bromide - dimethyl sulphide catalysed addition of methylmagnesium iodide to 3-bromocyclohexene (119) by a procedure s i m i l a r to that used in the preparation of (109) 81 yielded 3-raethylcyclohexene (j[4) i n a modest 63% y i e l d . This moderate y i e l d was probably due more to the v o l a t i l i t y of the compound (84),than to any shortcomings of the copper catalysed coupling reaction. (34) 119 84 The XH nmr spectrum of (84) which was i d e n t i c a l with 87 that reported , exhibited the required secondary methyl doub-l e t at 60.93 (J=7Hz) and a two proton o l e f i n i c multiplet at 65.28-5.73. - 52 -Preparation of compound (110) cl o s e l y followed the pro-83b 83a cedure reported by Salomon and by Eaborn . Thus, a THF solution of 3-bromocyclohexene (1 equiv.) was added slowly to a s t i r r e d , cooled (ice-bath) mixture of magnesium turnings (1.7 equiv.) and chlorotrimethylsilane (1.2 equiv.) in THF. Though Salomon reportedly obtained 50% y i e l d s of (110), in our hands the above procedure gave consistently poorer y i e l d s (30%) of t h i s material and substantial quantities (25%) of the 88 homo coupling product, 2,2'-bicyclohexenyl (121) . Clearly, trapping of the xn s i t u formed 2-cyclohexenylmagnesium bromide (120) by the 3-bromocyclohexene present to a small extent at each stage of the reaction, i s much more e f f i c i e n t than trapp-ing by chlorotrimethylsilane. Although t h i s i s a l i t t l e sur-p r i s i n g , since chlorotrimethylsilane i s i t s e l f quite a good e l e c t r o p h i l e , i t should be r e c a l l e d that formation of homo coupling products i s a major problem in the preparation of 89 a l l y l i c Grignard reagents by conventional methods . Usually , quite elaborate techniques are required to obtain s y n t h e t i c a l l y 89 90 useful y i e l d s ' of these valuable reagents. - 53 -(36) 120 119 121 Despite the poor y i e l d s of (110) working quantities could be obtained by the procedure described and so no attempt was made to u t i l i z e more complicated procedures i n the hope of im-proving the y i e l d s . Compound (.110) showed i r bands (1245, 835, 750 and 705 cm - 1) expected of a compound containing the t r i m e t h y l s i l y l 91 1 group . The H nmr spectral data obtained for (110) were v i r t u a l l y i d e n t i c a l with those reported. The nine proton sing-l e t due to the t r i m e t h y l s i l y l methyl groups occurred at 6-0,02 (60.00 in ref.82b) and the two proton o l e f i n i c s i n g l e t at 65.66 (65.60 i n re f . 82b). The known 2,2'-bicyclohexenyl ( 1 2 1 ) 8 8 gave mass spectral and nmr data consistent with i t s symmetri-c a l structure. D. Preparation of Compound (111) 92 The two step procedure of Cologne and Boisde which u t i -93 l i z e d 3-buten-l-ol (122) as s t a r t i n g material was followed f a i t h f u l l y in the preparation of compound (111) (Scheme XIX). - 54 -The required 3-buten-l-ol (122) was prepared57"* in 33% 95 y i e l d by the reaction of allylmagnesium bromide with para-formaldehyde i n r e f l u x i n g ether. The material i s o l a t e d for t h i s reaction showed the expected i r absorptions (3325, 3050 and 1630 cm~^) and i t s ^ "H nmr spectrum was i d e n t i c a l , except for the chemical s h i f t of the hydroxyl proton, with the publish-96 ed spectrum of 3-buten-l-ol (122). Treatment of (122) with s-trioxan i n the presence of dry hydrogen chloride resulted i n the formation of 4-chlorotetra-97 hydropyran (124) in 62% y i e l d . Since s-trioxan i s known to (37) Scheme XIX HCl(g) s-trioxan 122 124 111 decompose into formaldehyde in the presence of acid, the above reaction probably proceeds v i a chloromethylation of (122) to - 55 -y i e l d the y,6-unsaturated chloromethyl ether (123) which then c y c l i s e s to give (124) (Scheme XX). Scheme XX ^^-^NDH • C H 2 0 • HCl > ' J 122 123 124 The i r spectrum of (124) showed bands at 760 and 720 cm"1 which were assigned to carbon-chlorine stretching v i -99a brations and a s e r i e s of moderate to strong bands charac-t e r i s t i c of a tetrahydropyran r i n g in the 1000-1160 cm"1 99b 1 region . The 60 MHz H nmr spectrum showed a complex f i v e . proton multiplet at 63.27-4.40 which was assigned to the me-thylene protons on the carbons adjacent to the oxygen atom and the methine proton on the carbon adjacent to the chlorine atom. A four proton multiplet at 61.50-2.30 was assigned to the remaining methylene protons. Due to i t s symmetry no regiochemical ambiguity a r i s e s i n the dehydrochlorination reaction of compound (124). Thus r e f l u x i n g (124)with f i n e l y powdered potassium hydroxide in ethylene g y l c o l for 6h provided as the only product the desired 5,6-dihydro-2H-pyran ( 1 1 1 ) 1 0 0 in 67% y i e l d . The i d e n t i t y of t h i s compound was assured on consideration - 56 -of i t s previously unreported spectral properties. The i r spectrum showed bands (at 3010 and 1090 cm"^) i n d i c a t i v e of an oxygen containing o l e f i n . The "^E nmr spectrum showed a two proton multiplet at 61.90-2.27 assigned to the a l l y l i c protons H R and HR,; a two proton t r i p l e t (J=5Hz) at 63.75, undoubtedly due to the methylene protons H^ and H^,; a two proton multiplet further downfield at 64.00-4.20 due to the protons H A and H^, and a two proton o l e f i n i c multiplet at 65.43-5.83. E. Preparation of Compound (112)"*"""1" O l e f i n ( 1 1 2 w a s o n e 0 f those compounds prepared by Nicolaou et al. during t h e i r work on the preparation of oxygen heterocycles by the s e l e n o e t h e r i f i c a t i o n reaction. The Nicolaou procedure (Scheme XXI), which u t i l i z e d the re a d i l y available J.02 alcohol (126) as s t a r t i n g material and produced (112) a f t e r two high y i e l d i n g steps, was the method of choice for the pre-paration of the o l e f i n ( 1 1 2 ) 1 0 1 . - 57 -Alcohol (126)" L U i J was prepared i n 95% y i e l d by lithium aluminum hydride (LiAlH^) reduction of the commercially a v a i l -able 2-(2-cyclopentenyl)acetic acid (125). The i r spectrum of (126) showed absorptions as expected at 3325 and 1605 cm - 1 but (38) 125 126 no absorptions in the carbonyl region (1800-1600 cm - 1). The ''"H nmr spectrum showed, s i g n i f i c a n t l y , a two proton t r i p l e t (J=6.5Hz) at £3.73 assigned to the methylene protons adjacent to the hydroxyl group. Following Nicolaou's p r o c e d u r e 1 0 1 , the selenoether (127) could be consistently obtained from (126) in y i e l d s (79%) s i m i l a r to that reported. However, i n our hands the procedure reported for conversion of (127) into (112) gave rather low Scheme XXI 126 127 112 - 58 -y i e l d s Ov50%) of the o l e f i n . Fortunately, i t was found that an a l t e r n a t i v e procedure (Scheme XXII) described by Nicolaou" 1 0' in connection with the preparation of another o l e f i n , gave excellent o v e r a l l y i e l d s of o l e f i n (112). Thus the phenyl selenoether (127), prepared i n 95% y i e l d from (126) under the usual reaction conditions, was dissolved without p u r i f i c a t i o n i n dry methylene chloride and oxidized at -78°C with ozone. 103 Triethylamine was then added and the mixture was s t i r r e d at room temperature for 18h. Workup and d i s t i l l a t i o n gave the o l e f i n (112) i n 74% o v e r a l l y i e l d (based on (126). Although, a p r i o r i , the selenoxide intermediate (128) Scheme X X II 112 - 59 -can eliminate phenylselenenic acid to give r i s e to two o l e f i n s (112) and (129), only one o l e f i n (glc and t i c analysis) was obtained from t h i s reaction. The absence of any signal in the ''"H nmr spectrum of the is o l a t e d o l e f i n that could be assigned to of (129), the presence of a narrow one proton multiplet at 64.95-5.20, which could be c l e a r l y a ttributed to of (112) and the two proton multiplet in the o l e f i n i c region (65.51-6.00) made i t abundantly clear that the o l e f i n i s o l a t e d was in fact (112). This tendency for selenoxide eliminations to occur away 104 105 from s i t e s adjacent to oxygen has been well documented ' and indeed was r e l i e d upon in the above preparation to give exclusively o l e f i n (112) from selenoxide (128). F. Preparation of Compound (99) Preparation of o l e f i n (99) proceeded through a series of r e l a t i v e l y straightforward reactions from the r e a d i l y available cyclohexanone (Scheme XXIII, p. 62). The $-keto ester (130) was obtained in 85% y i e l d following the procedure reported by - 60 -Deslongchamps et a l 106 This material exhibited spectral features c h a r a c t e r i s t i c of a c y c l i c 8-keto ester. For i n -stance, i r bands at 1740 and 1710 cm - 1 as well as 1655 and 1615 cm - 1, the f i r s t two due to the keto form and the l a t t e r two to the enol form of the 8-keto ester, were evident. The nmr spectrum showed a mul t i p l e t at 6*3.27-3.54 (0.25H) and a sin g l e t at 69.18 (0.75H) due to (130a) and - (130b), respectively and the expected three proton si n g l e t at 63.75. n / H - 0 0 || (39) 130a 130b The unsaturated ester (132) had been prepared previously 107 by Weiler and Sum in t h e i r study of the addition of d i a l k y l -cuprates to enol phosphates of 8-keto esters. Following t h e i r published procedure, the enol phosphate (131) was prepared i n 94% y i e l d and was u t i l i z e d without p u r i f i c a t i o n in the coupl-ing reaction with lithium dimethylcuprate to y i e l d (132). The enol phosphate (131) exhibited spectral properties ( i r and "*"H nmr) in agreement with those reported by Weiler and Sum. The i r spectrum showed bands at 1295, 1150 and 1030 cm - 1 which, 108 according to Bellamy , are c h a r a c t e r i s t i c of the diethylphos-phoryloxy group. In the ''"H nmr spectrum the signals due to - 61 -the methylene and the methyl protons of the diethylphosphoryl-oxy group showed long range coupling with the magnetic phos-31 phorus-31 ( P) nucleus. Thus, the methyl protons appeared as a doublet of t r i p l e t s (J=l, 7Hz) at 61.36 instead of the usual t r i p l e t . The 1Hz coupling represents the four bond phosphorus-4 109 hydrogen coupling constant ( Jp_ H) • In the case of the 3 109 methylene protons the value of Jp ^ (7Hz) equalled, coin-2 c i d e n t a l l y , the v i c i n a l coupling constant ( ^) between the methyl and methylene protons i n the ethyl group. Thus, the methylene protons appeared as a quintet (J=7Hz) at 64.23. Treatment of the enol phosphate (131) with lithium d i -107 methylcuprate as described by Weiler and Sum yielded,after f l a s h chromatography of the crude i s o l a t e , a 76% y i e l d of the o l e f i n (132). Analysis by glc of the crude i s o l a t e of the re-actions we performed indicated that an impurity, believed to be ketone (134), was always present to the extent of ^10% des-134 p i t e the use of p u r i f i e d cuprous iodide"1""1"u or cuprous bromide-dimethyl sulphide for the formation of the dimethylcuprate. In any event, the material i s o l a t e d a f t e r chromatography was c l e a r l y the o l e f i n (132) as i t exhibited spectral features i n 107 agreement with those reported . For example, the i r spectrum _ 62 _ showed absorptions at 1705 and 1635 cm and the H nmr Scheme XX III 132 133 99 spectrum showed a broad v i n y l methyl singlet at 62.00 and a three proton si n g l e t due to the methyl ester at 63.73. Treatment of an ether solution of the unsaturated ester (132) with 2.2 equiv. of Dibal at -78°C resulted in a 92% y i e l d of the alcohol (133). The i r spectrum of t h i s material exhibited the required absorptions (3300 and 1660 cm - 1) and i t s nmr spectrum was e n t i r e l y consistent with i t s structure. The spectrum included a broad, one proton, DgO exchangeable s i n g l e t at 61.34 which was assigned to the hydroxyl proton, a broad s i n g l e t at 61.70 due to the v i n y l methyl group and a two proton s i n g l e t at 64.13 which was assigned to the i s o l a t e d methylene protons adjacent to the hydroxyl group. - 6 3 -Methylation of (133) was c a r r i e d out in THF using a modification of the procedure reported by Brown e_t al . 1 1 1 . The sodium alkoxide of (133) was generated by r e f l u x i n g a mixture of the alcohol with sodium hydride in THF and then was treated with methyl iodide. The r e s u l t i n g mixture was re-fluxed for a few hours or s t i r r e d overnight at room tempera-ture to y i e l d the methyl ether (99) in good y i e l d s (^90%). The o l e f i n (99) exhibited a 1670 cm"1 band in i t s i r spectrum and i t s 1H nmr spectrum c l e a r l y supported the assign-ed structure. The following key signals were displayed - a broad s i n g l e t at 61.69 due to the v i n y l methyl group, a three proton si n g l e t due to the methyl ether protons at 63.30 and at lower f i e l d (63.89) a two proton s i n g l e t assigned to the a l l y -l i c methylene protons of the methoxymethyl ( C H 9 0 M e ) group. I l l . Preparation of the Dibromocyclopropanes The o r i g i n a l procedure for the preparation of dibromo-62 cyclopropanes was developed by Doering and Hoffmann . They showed that dibromocarbene, generated in s i t u under s t r i c t l y anhydrous conditions from potassium tert-butoxide and bromo-form, added to cyclohexene to give 7,7-dibromonorcarane (135) i n 75% y i e l d . Their procedure constituted the major route to dibromocyclopropanes for close to twenty years. 6 3 More recently, Makosza has shown that dibromocyclo-_ 64 -pentane 0°C KOBu , CHBr 3 Br (40) 118 135 propanation of o l e f i n s could be e f f i c i e n t l y achieved i n a two 112 phase system in the presence of a phase transfer catalyst , with dibromocarbene generated i n s i t u from concentrated aqueous sodium hydroxide and bromoform. U t i l i z i n g t r i e t h y l b e n z y l -ammonium chloride (TEBA) as cata l y s t , they found that y i e l d s were optimum when a small amount of an alcohol (e.g. ethanol) was added. The sheer convenience of the Makosza method makes i t now the method of choice for the preparation of dibromocyclopro-panes which, perhaps because of t h e i r easy preparation, are 61 becoming increasingly important i n synthesis (41) - 65 -In the present study a l l but one of the dibromocyclo-propanes were prepared using Hakosza's phase transfer catalysed procedure. The reactions were carr i e d out at 40°C, as des-cribed in the experimental section of t h i s t h esis, and t h e i r progress could be monitored by glc analysis of aliquots of the reaction mixtures. In general, those o l e f i n s with a l l y l i c or homoallylic oxygen substituents required longer reaction times for completion of the dibromocyclopropanation reaction than did the o l e f i n s (84) and (109) which have a l l y l i c methyl and 113 phenyl substituents . Because i t was doubtful whether the t r i m e t h y l s i l y l group of o l e f i n (110) would survive the condi-tions of the Makosza reaction, the Doering procedure was u t i -l i z e d i n the dibromocyclopropanation of t h i s o l e f i n . In general, the crude dibromocyclopropanes i s o l a t e d from the dibromocyclopropanation reactions were p u r i f i e d by chroma-114 tography. Because of the well known thermal l a b i l i t y of these compounds,distillation was attempted only i n those cases where temperatures less than ^80°C could be u t i l i z e d . The p u r i f i e d dibromocyclopropanes a l l showed a strong band in the 730-760 cm - 1 region of t h e i r i r spectra. This band i s at %725 —1 62 cm i n the i r spectrum of 7,7-dibromonorcarane (135) and appears to be c h a r a c t e r i s t i c of dibromocyclopropanes. The o l e f i n (78), subjected to the Makosza conditions gave 115 afte r f l a s h chromatography and d i s t i l l a t i o n of the crude product, a 52% y i e l d of the trans-methoxydibromocyclopropane - 66 -(79a). A small quantity (^3% i s o l a t e d y i e l d ) of the more polar cis-methoxydibromocyclopropane (79b) was also obtained. Simi-l a r l y , the related o l e f i n (92) under e s s e n t i a l l y the same re-action conditions gave as the major product the previously re-58 ported trans-dibromocyclopropane (93a) and the more polar cis-dibromocyclopropane (93b) as the minor product. * (42) 78 79a vyo %95 : 5 Assignment of the trans r e l a t i v e stereochemistry to the major products from the dibromocyclopropanation of ( 7 8 ) and (92) was i n i t i a l l y based on s t e r i c approach considerations as well as on "^H nmr spectral evidence. For s t e r i c reasons, the rather bulky dibromocarbene species would be expected to ap-proach the double bond of either (78) or (92) p r e f e r e n t i a l l y * Throughout t h i s thesis the descriptors c i s and trans are used to indicate the r e l a t i v e stereochemistry of a substituent at C 2 of a bicyclo[n.1.0]compound. The descriptors endo and exo are reserved for the stereo-chemistry of a substituent on the cyclopropyl ring i . e at the (n+3) p o s i t i o n of the b i c y c l i c system. - 67 -: CBr, + (43) OMEM 93b ^95 from the face opposite to the substituent. Thus a trans r e l a t i o n s h i p , with respect to the five-membered ri n g , between the alkoxy substituent and the dibromocyclopropane moiety should characterise the major product in both cases. As far as the proton nmr spectra of the cis-and trans-dibromocyclopropanes are concerned,one can argue as follows. Since the dihedral angle between the proton and both H R and H n i s approximately 90° in both (79a) and (93a) then J D BC and would both be close to zero 116 This means that i n the trans-dibromocyclopropanes, (79a) and (93a), the proton H £ would couple appreciably ( c i s v i c i n a l coupling in the range 117 5-10Hz ) with only one proton (H_,) and should therefore appear e s s e n t i a l l y as a doublet in the 1H nmr spectra of (79a) and (93a). On the other hand, in the cis-dibromocyclopropanes (79b) and (93b) the H^ , proton would be expected to couple appreciably to both H R and ( c i s v i c i n a l coupling). More-over, d i s t o r t i o n s of the cis-dibromocyclopropane molecules to minimise the s t e r i c i n t e r a c t i o n between the cis-alkoxy s u b s t i -tuents and the dibromocyclopropane moieties would resu l t in - 68 -moderate coupling (^7Hz>J(-;E>OHz) between the trans- v i c i n a l protons and H^. Thus, at the very least, the signals due to the protons l a b e l l e d in the cis-dibromocyclopropanes (79b) and (93b) should be more complex (probably appearing as multiplets) than the signals (probably doublets) due to the protons of the trans-dibromocyclopropanes (79a) and (93a). The predictions l o g i c a l l y arrived at above were borne out exactly. In the nmr spectrum of the major compounds (79a) and (93a) the protons appeared as a doublet of doub-l e t s (J=6.5, 2Hz at 63.83) and a broad doublet («K6Hz at 64.25), respectively. In the nmr spectra of the minor products (79b) and (93b), by contrast, the H^ , protons appeared as multi-p l e t s which were at lower f i e l d ( 64.34-4.42 and 64.44-4.70, respectively) than the corresponding signals of the t r a n s - d i -bromocyclopropanes. An i n t e r e s t i n g point of difference between the nmr spectra of (93a) and (93b) i s the appearance of the signal due to the methylenedioxy protons of the -OMEM substituent. These protons appear as a si n g l e t at 64.80 in the 100MHz "*"H nmr spec-trum of the trans-dibromocyclopropane (93a) whereas in the nmr spectrum of the cis-dibromocyclopropane (93b) they appear as an AB type doublet of doublets (J=7Hz) at 64.85. Presumably, in the c i s compound (93b) the molecular asymmetry i s r e f l e c t e d in the environment of the two protons to a greater extent than in the trans compound (93a). The trans r e l a t i o n s h i p between the alkoxy substituent and - 69 -the cyclopropane ri n g in compounds (66a) and (67a) was such a c r u c i a l requirement in t h i s study that i t was important to ensure that the dibromocyclopropanes from which they were pre-pared possessed the proper r e l a t i v e stereochemistry. Thus i t was that despite a l l the evidence which indicated that the major products from the dibromocyclopropanation of the o l e f i n s (78) and (92) were the desired trans-dibromocyclopropanes (79a) and (93a), we resolved to prove the r e l a t i v e stereochemistry of the l a t t e r two compounds by independent synthesis. To t h i s end, the cis-MEM ether (137b) was prepared by an unambiguous two step route (Scheme XXIV). Zinc-copper couple mediated cyclopropanation of 2-cyclopenten-l-ol was expected, 118 according to ample l i t e r a t u r e precedent , to y i e l d the known cis-bicyclo[3.1.0]hexan-2-ol ( 1 3 6 ) 1 1 9 . Treatment of t h i s a l -cohol with diisopropylethylamine and B-methoxyethoxymethyl chloride then gave the cis-MEM ether (137b). Scheme XXIV 114 136 137b - 70 -The major product from the dibromocyclopropanation of the o l e f i n (92) was completely debrominated, as outlined in equation 44, to y i e l d the presumed trans-MEM ether (137a). (44) B r > - B r fx/ 1 ) B u ^ L i ; H + r 2)Bu tLi;H + HA O E M OMEM 93a 137a The two compounds (137a) and (137b) on the basis of comparisons of glc retention times and i r and "^H nmr spectra proved to be quite d i f f e r e n t . S i g n i f i c a n t l y , the major d i f -ferences between the "'"H nmr spectra of these two compounds were sim i l a r i n nature to those already discussed for (93a) and (93b). Thus, while H c of (137a) appeared as a broad doub-l e t (J=5Hz) at 64.20, the corresponding proton (H c) of (137b) appeared as a broad multiplet between 64.30-4.58. The methyl-enedioxy protons within the -OMEM group of (137a) appeared as a si n g l e t at 64.81 while the corresponding protons in (137b) appeared as an AB type doublet of doublets (J=7Hz) also at 64.81. At t h i s point the assignment of the trans r e l a t i v e stereochemistry to the major products from the dibromocyclo-propanation of the o l e f i n s (78) and (92) appeared to have been f u l l y j u s t i f i e d . - 71 -Dibromocyclopropanation of the o l e f i n (108) gave two products in a r a t i o 96:4. Following the precedent e s t a b l i s h -ed in the dibromocyclopropanation of the o l e f i n s (78) and (92), the major product was assigned the trans r e l a t i v e stereochemis-t r y (structure (138a) ) and the minor product the c i s stereo-chemistry ( i . e . structure (138b)). The mass spectra of both (138a) and (138b) showed the three peak c l u s t e r in the M+ region (m/e 356, 354, 352) 120 c h a r a c t e r i s t i c of a dibromide and in each case the base peak appeared at m/e 85 (due to the C-H-O ion). While the i r (45) 108 138a 138b ^96 : 4 and H nmr spectra of (138a) and (138b) were in accord with t h e i r structures as tetrahydropyranyloxy substituted dibromo-cyclopropanes, there was no clear cut nmr evidence to sup-port the assignment of the trans r e l a t i v e configuration to the - 72 -the major product. One point of difference between the nmr spectra of (138a) and (138b) was, however, reminiscent of one of the differences between the spectra of (93a) and (93b) and could be considered, by analogy, to be in support of our assignments. The protons of the two THP diastereomers of (138a) gave r i s e in the "*"H nmr spectrum to an unresolved multi-plet at 6 4.68-5.03. In the "^H nmr spectrum of (138b), however, the H D protons of the two THP diastereomers gave r i s e to two d i s t i n c t m u l t i p l e t s at 64.68-4.95 and at 65.00-5.18. A 6:1 mixture (by c a p i l l a r y glc analysis) of the dibromo-cyclopropanes (85a) and (85b) was obtained on subjecting the o l e f i n (84) to the Makosza dibromocyclopropanation conditions. These very nonpolar compounds could not be separated by the usual chromatographic methods and the p u r i f i e d mixture of (85a) and (85b) was u t i l i z e d in the succeeding reaction. Paquette 121 and co-workers , who had previously dibromocyclopropanated 62 the o l e f i n (84) under the Doering conditions , had also re-ported t h e i r i n a b i l i t y to separate the r e s u l t i n g dibromocyclo-propanes (85a-b). The mixture, (85a) and (85b), gave a s a t i s f a c t o r y element' a l analysis for CgH^2 B r2 a n ( * t n e mass spectra showed the three peak c l u s t e r i n the M+ region (m/e 270, 268, 266) characteris-t i c of a dibromo compound. In the 400 MHz "'"H nmr spectrum of the mixture, two secondary methyl doublets could be c l e a r l y seen. The major doublet,at 61.22 (J=7Hz), was assigned to the methyl group of (85a) and the minor doublet, at 61.26 _ 73 _ (46) 84 85a 85b ^6 : 1 (J=7Hz), to the methyl group of (85b). Assignment of the trans r e l a t i v e stereochemistry to the major dibromocyclopropane (85a) was based on the well established precedent (vide supra). The o l e f i n (109) yielded just one i d e n t i f i a b l e product, the dibromocyclopropane (139), when subjected to the usual re-action conditions. Chromatographic p u r i f i c a t i o n of the crude i s o l a t e from the reaction mixture yielded an o i l which s o l i d i -f i e d on storage in the cold (+5°C) to y i e l d a white s o l i d . R e c r y s t a l l i z a t i o n of a portion of t h i s material from petroleum ether gave white c r y s t a l s which melted sharply at 48.5-49.5°C. These c r y s t a l s gave a s a t i s f a c t o r y elemental analysis for <"13 H14 B r2 a n c* e x n i b i t e d mass spec t r a l , i r and "''H nmr character-i s t i c s consistent with expectations. For example, i r absorp-tions at 3035, 1595, 765 and 700 cm as well as a f i v e proton multiplet at 67.18-7.50 in the nmr spectrum confirmed the presence of the phenyl substituent in (139). The strong —1 62 + i r absorption band at 735 cm" and the M c l u s t e r (at m/e 332, 330 and 328) in i t s mass spectrum indicated the dibromo-- 74 -(47) 109 139 cyclopropane moiety of (139). 122 The dibromocyclopropane (140) had been prepared pre-viously from the o l e f i n (111) in low y i e l d (30%, 95% purity) 62 using the Doering dibromocyclopropanation conditions. In our hands the o l e f i n (111) subjected to the Makosza reaction conditions gave, afte r workup and d i s t i l l a t i o n , a 73% y i e l d of compound (140) as a colourless o i l . The material was chromatographically and spectroscopically pure and gave a white c r y s t a l l i n e s o l i d on cooling (^5°C). The """H nmr spectrum obtained for (140) was very s i m i l a r 122 to that reported by Taylor ejt a l . ^  for t h i s compound. How-ever, the greater dispersion obtained with the 400MHz spectro-meter available to us did reveal some in t e r e s t i n g d e t a i l s . For example, the two proton " t r i p l e t " at 63.98 in the reported spectrum turned out to be two doublet of doublets at 63.90 (H c, J B C=2Hz, J C D=12Hz) and 63.97 (Hp, J D B=6Hz, J C D=12Hz). One can argue on the basis of the magnitude of these coupling con-stants ( J R p and J n n ) that (140) exi s t s predominantly in the -75 -o Br (48) 111 140 ha l f - c h a i r conformation (14OB). In the h a l f - c h a i r conforma-ti o n (140A) the dihedral angles between Hg and H^.and H D and Hg are such that both Jg^ and J^g would be expected to be small (0-3Hz). The dihedral angles between Hg and H D (^0°) and between and Hg (close to 90°) in conformation (140B) would be expected to lead to the values of J D B and Jg^, such as those a c t u a l l y observed (6Hz and 2Hz respectively) . Similar 140A 140B _ 76 _ 122 arguments were employed by Taylor et_ a l . to suggest that some derivatives of (140) existed predominantly in the chair form corresponding to (140A). The dibromocyclopropane (141) was the only compound i s o -lated (48% y i e l d ) on subjection of the o l e f i n (112) to the 6 3 Makosza dibromocyclopropanation conditions. Compound (141) showed in i t s mass spectrum the c h a r a c t e r i s t i c c l u s t e r i n the M+ region"1"'20 (m/e 284, 282 and 280) and i t s i r spectrum exhi-bited a strong 740 cm~^~ band i n d i c a t i v e of the gem dibromocyclo-propane moiety. The 400 MHz "^H nmr spectrum of (141) was assigned in de-t a i l with the aid of proton decoupling experiments (Fig. 4a). The lowest f i e l d signal (doublet, J H C=7Hz) at 64.40 was c l e a r l y due to H^ ,. The fact that t h i s signal was a doublet was not un-expected considering that the H^-H^ dihedral angle i s approxi-mately 90°. On t h i s basis also,assignment of the doublet (^J^y ^ > *X I V ^ P r (50) 112 141 (J A B=7Hz) at 62.42 to H R seemed quite secure. The assignment of the multiplet at 63.70-3.88 to the protons H £ and H D and that at 62.70-2.82 to the methine, bridgehead proton H„ was - 77 -also quite straightforward. Assignment of the remaining signals was rather less straightforward. The signals due to and Hj were assigned on the basis of the following arguments. Since Ej and are approximately orthogonal (cf. Hg and H^), H^ was expected to be a doublet of doublets (coupling with Hg and Hj) in which J^g equals 7Hz. Hj was also expected to be a doublet of doublets with one large (geminal) coupling constant ( J J J ) and one v i c i n a l coupling constant ( J J J J ) expected to be in the v i -117 c m i t y of 5-10Hz . On the basis of the predictions above, the doublet of doublets at 62.24 with coupling constants 10 and 14.5Hz was assigned to Hj and the t r i p l e t (overlapped doublet of doublets, J=7Hz) at 6 2.33 was assigned to H^. The multiplets centred at 61.59, 1.76 and 1.97, so far unassigned, must be due to the remaining protons H.,, H_, and H T. r (j i Of these, the two due to H„ and H_, were i d e n t i f i e d by i r r a d i a t -ing the multiplet (63.70-3.88) due to H D and H^ (Fig. 4b). The multiplet centred at 61.59 collapsed into a doublet with J=12Hz 123 i n d i c a t i v e of geminal coupling whereas that at 61.97 became a doublet of doublets with coupling constants of 8 and 12Hz. This r e s u l t led to the assignment of the former multiplet (at 61.59) to H Q and that at 61.97 to H p since was expected to be v i r t u a l l y zero and the c i s v i c i n a l coupling constant, J F H i 117 was expected to be somewhere between 5 and 10Hz . The multi-plet at 61.76 was unaffected under the decoupling conditions and was thus assigned to Hj. Other decoupling experiments for _ 81 _ example, i r r a d i a t i o n of the multiplet due to H^, served to confirm the assignments made above ( F i g . 4c). As mentioned before, the o l e f i n (110) was dibromocyclo-62 propanated using the Doering conditions. Despite several attempts, the reaction never went to completion and i t proved impossible to separate the unreacted s t a r t i n g material from excess bromoform. Thus the 55% i s o l a t e d y i e l d of the dibromo-cyclopropanes (142a) and (142b) was considered quite respect-able indeed. C a p i l l a r y glc analysis of the i s o l a t e d mixture indicated that two products were formed i n a r a t i o of approxi-mately 94:6. As before, the trans r e l a t i v e stereochemistry was assigned to the major product. The mass, i r and "^H nmr (51) 110 142a 142b ^94 : 6 spectra of the mixture (142) were f u l l y in accord with expecta-tions. Thus the mass spectrum exhibited the three peak c l u s t e r in the M+ region (at m/e 328, 326 and 324) c h a r a c t e r i s t i c of a 120 dibromo compound . Other peaks in the mass spectrum at m/e 254, 252 and 250 (M +-H-SiMe 3) and at m/e 73 (base peak); absorp-tions bands at 1250, 840, 755 and 735 cm-^" in the i r spectrum and s i n g l e t s at 60.10 (the major) and 60.19 (the minor) i n the - 82 -H nmr spectrum of the mixture a l l provided overwhelming e v i -dence for the presence of the t r i m e t h y l s i l y l group. IV. Preparation of the Monobromocyclopropanes. As already mentioned, several methods exist for the con-t r o l l e d conversion of dibromocyclopropanes to monobromocyclo-propanes. S e y f e r t h 6 4 reported that 7,7-dibromobicyclo[4.1.0] heptane (135) was reduced with one equiv. of t r i - n - b u t y l t i n 124 hydride to give, in 82% y i e l d , a mixture of endo- and exo-monobromides, (143) and (144), in a r a t i o of 2.5 to 1. (52) 135 143 144 2.5 : 1 The a-bromocyclopropyl protons, and H R, both appeared as t r i p l e t s in the "^H nmr spectrum of the mixture of (143) and (144). The larger coupling constant (J . =8Hz) in the lower •• ™ C I S f i e l d t r i p l e t (63.58) was associated with the endo-bromo isomer (143) in which i s c i s to the other cyclopropyl protons. In the exo-bromide (144), E„ i s trans to the other cyclopropyl protons and thus the H R t r i p l e t (at 62.58) exhibits a smaller _ 83 _ coupling constant (J =3.7Hz). 65 Hofman et a l . had reported even e a r l i e r that z i n c -acetic acid reduction of dibromocyclopropanes yielded the corresponding monobromo compounds. This method was shown, in 58 a previous report from our laboratory, to give very good r a -t i o s of endo- to exo- monobromides (^10:1) in the reduction of 7,7-dibromobicyclo[4.1.0]heptanes. The corresponding r a t i o s in the reduction of 6,6-dibromobicyclo[3.1.0]hexanes were, how-ever, much poorer (>1:2) with both z i n c - a c e t i c acid and t r i - n -b u t y l t i n hydride reducing systems. Yet another method which has been u t i l i z e d for the mono-reduction of dihalocyclopropanes involves low temperature (<-78°C) lithium-halogen exchange, followed by protonation of the rather unstable a-halocyclopropyllithium intermediates * ^66 formed In the preparation of the monobromocyclopropanes for t h i s study a l l three of the methods described above were u t i l i z e d . Assignment of the endo- and exo-bromo configurations was based 64 (following Seyferth ) on the magnitude of the coupling constant in the t r i p l e t associated with the a-bromocyclopropyl proton. This proton in a l l cases occurred at higher f i e l d in the nmr spectra of the exo-bromo isomers than i t did in the spectra of of the corresponding endo-isomers. Because of the poor endo- to exo-monobromide r a t i o pre-58 viously obtained from the t r i - n - b u t y l t i n hydride and z i n c -acetic acid reductions of the dibromocyclopropane (93a), i t was - 84 -decided to investigate the low temperature bromine-lithium ex-change/protonation protocol for the reduction of t h i s compound. After some experimentation i t was found that, by slow addition of a THF solution of the dibromocyclopropane (93a) to a solu-t i o n of n-butyllithium (1.2 equiv.) at low temperature (-127°C) followed by careful quenching of the reaction mixture at the same temperature, a mixture of endo- and exo-monobromides (94a) and (94b) in an optimum r a t i o of 1.5 to 1 could be obtained. The method was quite tedious requiring s t r i c t control of temp-erature and addition rates and variable r a t i o s in the v i c i n i t y of 1:1 were usually obtained. Br, Br Br n l)Bu L i THF-ether -hexane t~^ T OMEM 2 ) H + OMEM + (53) OMEM 93a 94 a 94b Although the monobromides (94a) and (94b) had been separat-58 ed previously by column chromatography, i t was expedient i n t h i s study to u t i l i z e the mixture of these monobromides for the succeeding reactions. The mixture exhibited,in i t s i r and *H nmr spectra,features that were c h a r a c t e r i s t i c of the -OMEM group. The diagnostic t r i p l e t s due to the H R protons of (94a) and (94b) were also i d e n t i f i a b l e i n the "'"H nmr spectrum of the mixture. The t r i p l e t at 62.44 (with J=^2Hz) was c l e a r l y due - 85 -to of the exo-bromo isomer (94b). Although the t r i p l e t due to H K of the endo-bromide (94a) was a l l but obscured by the methoxy si n g l e t at 63.42, the chemical s h i f t (63.88) and coupl-ing constant (J=^7.5Hz) could be r e l i a b l y determined. The signals due to the protons of the endo- and exo-monobromides (94a) and (94b) were recognisable as two p a r t i a l l y overlapped doublets at 64.19 (J=5.5Hz) and 64.27 (J=5Hz)respectively. The dibromocyclopropane (79a) was reduced with t r i - n - b u -t y l t i n hydride to give in 92% y i e l d a mixture of the exo- and endo-monobromides, (80b) and (80a), in a r a t i o of 1.6:1, re-spectively. These monobromides were separated by f l a s h chroma-115 tography and i n d i v i d u a l l y characterized although, as in the case of (94a) and (94b), i t was convenient to u t i l i z e the mix-ture in succeeding reactions. The mass spectra of both (80a) and (80b) c l e a r l y i n d i c a t -120 ed the presence of one bromine atom , showing two equal i n -tens i t y M+ ions (at m/e 192 and 190) and a base peak at m/e 111 (M +-Br). In the "^H nmr spectra, the H R proton of the endo-OMe (54) 79a 80a 80b - 86 -bromide (80a) appeared as a p a r t i a l l y obscured t r i p l e t at 63.39 (J=7.5Hz) while the t r i p l e t due to H„ proton of the exo-bromide (80b) occurred at 62.43 with J=2Hz. The H c protons of the i s o -mers (80a) and (80b) occurred, as did the corresponding protons in (94a) and (94b), as doublets of s l i g h t l y d i f f e r e n t chemical s h i f t s with the proton of the exo-bromide (80b) occurring at lower f i e l d 63.84 (J=5Hz) than the H c proton (63.75, J=5Hz) of the endo-bromide (80a). Zinc-acetic acid reduction of the 6:1 mixture of c i s - and trans-dibromocyclopropanes (85a-b) yielded an inseparable mix-ture of the monobromocyclopropanes (86a-d) in which the r a t i o of the endo-monobromides (86a-b) to the exo-monobromides (86c-d) was t y p i c a l l y 9 to 1 (by both c a p i l l a r y glc and nmr a n a l y s i s ) . The mixture exhibited in i t s mass spectrum the two M+ peaks, at m/e 190 and 188, and the base peak at m/e 109 (M +-Br) expected of the monobromides (86). Its "^H nmr spectrum showed the expected t r i p l e t s due to the H K protons of the endo-exo-monobromides. Careful examination of the major (0.9H) 85a-b 86a-b 86c-d - 87 -t r i p l e t (63.28, J=8Hz) showed that i t actually consisted of two poorly resolved t r i p l e t s with one much larger than the other. This signal on the basis of the magnitude of the coupling constant (J=8Hz) was assigned to the H K proton of the ehdo-monobromides (86a-b). The t r i p l e t (0.1H) at 62.59 with J=3.5Hz was assigned to the H K proton of the exo-mono-bromides (86c-d). Reduction of the mixture ( 94:6 r a t i o ) of c i s - and trans-dibromocyclopropanes (142a) and (142b) with t r i - n - b u -115 t y l t i n hydride gave, afte r f l a s h chromatography of the crude reaction mixture, a 71% y i e l d of the monobromocyclopro-pane mixture (145a-d). The r a t i o of the four monobromides by c a p i l l a r y glc was 72:5:21:2 from which was deduced a r a t i o of endo- to exo-monobromides of %3.3 to 1. (56) 142a-b 145a 145b 145c 145d The i r and mass spectra of the monobromide mixture (145) exhibited the expected features. Thus,in the i r spectrum ab-sorptions at 1250, 840 and 750 cm"1, i n d i c a t i v e of the t r i -91 m e t h y l s i l y l group , were prominent. In the mass spectrum a l l - 8 8 -the c h a r a c t e r i s t i c peaks, for example at m/e 248 and 246 (M 1), 175 and 173 (M +-SiMe 3), 167 (M +-Br) and the base peak at 73 (due to the SiMe^ ion), were present. The "''H nmr spectrum of the mixture (145) showed a t r i -p let at 63.34 with J=7Hz which was assigned to the H K protons of the endo-monobromides (145a-b) and another at 62.60 with J=3.5Hz assigned to the H K protons of the exo-bromides (145c-d). Careful integration of these two t r i p l e t s led to the deter-mination of the r a t i o of endo- to exo-monobromides as 3.7 to 1. It proved impossible to completely separate the four monobromides (145a-d), however, afte r two chromatographic runs there was obtained a mixture which consisted of 88% (by c a p i l l -ary glc) of the trans, ehdo-monobromide (145a). This mixture,thus enriched in (145a), was u t i l i z e d in succeeding reactions. The dibromocyclopropane (138a) on reduction with t r i - n -b u t y l t i n hydride gave a mixture of endo- and exo-monobromides 115 in a r a t i o of 2.3 to 1 (by g l c ) . Flash chromatography of the crude mixture afforded (146a) and (146b) in 43% and 25% y i e l d s , respectively. In the ''"H nmr spectra of these compounds the a-bromocylo-propyl protons (H„) occurred as the usual t r i p l e t s . The t r i -l e t due to H K of the endo-monobromide (146a) appeared at 63.30 with J=7Hz and that due to of the exo-monobromide (146b) was found at 62.53 with J=3.8Hz. _ 89 _ (57) 138a 146a 146b 2.3 : 1 The i r and mass spectra of (146a) and (146b) were f u l l y consistent with the presence of the tetrahydropyranyloxy group and a single bromine atom. The mass spectra of both compounds for example, showed diagnostic peaks at m/e 276 and 274 (M +), 195 (M +-Br) and at 85 (base peak) due to the C^H^O fragment ion. Reduction of the dibromocyclopropane (139) with t r i - n -b u t y l t i n hydride gave a 3.4 :1 mixture of the endo- and exo-115 monobromides (147a) and (147b). Flash chromatography of the crude monobromide mixture afforded compounds (147a) and (147b) in i s o l a t e d y i e l d s of 44 and 13%, respectively. (58) 139 147a 147b - 90 -The "^H nmr spectra of (147a) and (147b) were f u l l y in accord with t h e i r structures. Thus, the H.. proton of the iv endo-monobromide (147a) gave r i s e to the expected t r i p l e t at 6 3.37 (J=8Hz) while the t r i p l e t due to the proton of the endo-monobromide (147b) occurred at higher f i e l d (62.76) with a smaller coupling constant (J=3.5Hz). The presence of the phenyl group was indicated by the f i v e proton multiplets in the region 67.20-7.50 of the 1H nmr spectra of both (147a) and (147b). The proton (H^ ,) adjacent to the phenyl group was also recognisable. The proton of the exo-monobromide (147b) occurred as a multiplet at lower f i e l d (62.75-3.02) than the H c proton (62.56-2.82) of the endo-monobromide (147a). Further support f o r the presence i n (147a) and (147b) of the phenyl group and the single bromine substituent was pro-vided by t h e i r mass spectra which both exhibited diagnostic peaks at m/e 252, 250 (M +), 171 (M +-Br) and 91 (C ?H 7). T r i - n - b u t y l t i n hydride reduction of compound (140) gave a 2.2:1 mixture of the endo- and exo-monobromides (148a) and (148b). Chromatographic separation of the monobromides obtain-ed on di r e c t d i s t i l l a t i o n of the reaction mixture afforded (148a) in 41% y i e l d and (148b) in 19% y i e l d . The exo-bromide (148b) had been prepared previously by 122 1 Taylor et a l . and i t s H nmr spectrum was i d e n t i c a l with that reported by these authors. Thus H R appeared as the usual t r i p l e t at 62.84 (J=3.5Hz), H n appeared as a doublet of doub-- 91 -l e t s at 6 3.73 (with J(-;D=12Hz and Jg£=3.5Hz) and gave r i s e to a broad doublet at 64.05 (with J^D=12Hz and Jg^OHz). 122 Taylor used arguments based on the magnitude of the coupl-ing constants Jg^, and Jg^ to propose that (148b) exist pre-dominantly in the h a l f - c h a i r conformation shown in (148E). (See discussion on p. 75). In the nmr spectrum of (148a) appeared as a doub-l e t of doublets at 64.06 with J C D=12Hz and JgD=6Hz while H c appeared as a doublet of doublets at 63.81 with J C D=12Hz and J^g=1.5Hz. The magnitude of these coupling constants, follow-122 ing Taylor's arguments ,would require that (148a) exist pre-dominantly in the h a l f - c h a i r conformation shown in (148C). Whatever the conformation of (148a) might be, the one proton t r i p l e t at 63.32 (J=7Hz), assigned to H K > c l e a r l y confirms the endo-configuration of the bromine atom in t h i s compound. It was i n i t i a l l y quite s u r p r i s i n g that the dibromide (140) and the ehdo-monobromide (148a) should p r e f e r e n t i a l l y ex-i s t i n the h a l f - c h a i r conformations i l l u s t r a t e d by (14OB) and - 92 -148F 148E (148C) rather than the a l t e r n a t i v e conformations represented by (140A) and (148D) since hydrogen-bromine non-bonded i n t e r -actions are c l e a r l y greater in the former conformation than in the l a t t e r . The fact that the exo-monobromide (148b) exists p r e f e r e n t i a l l y in the h a l f - c h a i r conformation l a b e l l e d (148E), however, led us to propose that a strong d e s t a b i l i s i n g i n t e r -action e x i s t s between the a x i a l l y directed oxygen lone pair and the endo-bromine atom in the h a l f - c h a i r conformations (140A) and (148D). Thus ,(140B) and (148C) are the preferred conforma-tions of (140) and (148a). In the exo-monobromide (148b), - 93 -t h i s d e s t a b i l i s i n g i n t e r a c t i o n i s non-existent and so the conformation (148E) with less non-bonded hydrogen-hydrogen interactions i s again preferred. On reduction of the dibromocyclopropane (141) with t r i - n - b u t y l t i n hydride an approximately 1:1 mixture of endo-and exo-monobromides, (149a) and (149b), was produced. The in d i v i d u a l monobromides were i s o l a t e d in f a i r l y low y i e l d (%35% t o t a l y i e l d ) a f t e r high pressure l i q u i d chromatography of the crude monobromide mixture. Both monobromides, (149a) and (149b), showed the expect-ed peaks at m/e 204 and 202 (M +) and 123 (M +-Br) in t h e i r mass spectra. In the "*"H nmr spectra, the H^ . proton of (149b) occurred as the expected t r i p l e t at 62.30 with J=2Hz while the H„ proton of the endo-bromo isomer (149a) occurred as a t r i p -le — — l e t at 63.37 with J=8Hz. The H p protons of (149a) and (149b) (60) 141 149a 149b 1 : 1 appeared as doublets (J=5Hz, c f . H c i n (141)) at 64.33 and 64.39, respectively. The geminal protons H D and H E, which in the "'"H nmr spectrum of the dibromo compound (141) occurred as a sym-_ 94 _ metrical, complex multip l e t centred at 63.79, appeared, in the nmr spectra of both ("149a) and (149b), as coincident 117 doublet of doublets (J=5, 8Hz) at about 63.85. V. Preparation of the Cyclopropyl Esters (89) and (96) and the Carboxylic Acid (151a) With the monobromide mixtures (80) and (94) in hand, preparation of the corresponding esters (89) and (96) pro-ceeded as outlined i n Scheme XXV. Thus, treatment of an ethereal solution of the monobromides (94) with 2.2 equiv. of t e r t - b u t y l l i t h i u m followed by treatment of the r e s u l t i n g solution of the lithiocyclopropanes (95) with excess methyl chloroformate yielded, on workup and d i s t i l l a t i o n , the cyclo-propyl esters (96) in 57% y i e l d . The trapping of the i n t e r -mediate lithiocyclopropanes (9_5) by methyl chlorof ormate did not appear to be very e f f i c i e n t . In several t r i a l s , variable quantities (up to 20%) of the cyclopropane (137a) (formed by protonation of (95))were detected by glc analysis of the crude material i s o l a t e d from the reaction mixture. As might be ex-pected, the r a t i o of endo- to exo-isomers i n the i s o l a t e d cyclopropyl esters C96) also varied. By contrast, the chloroformate trapping of the l i t h i o -cyclopropanes (81), which was performed i n the presence of hexamethylphosphoramide (1 equiv.), was far more e f f i c i e n t with - 95 -only a small amount (>6%) of the cyclopropane (150),result-Scheme XXV 80 R=-Me 81 R=-Me 89a R=-Me 89b R=-Me 94 R=-MEM 95 R=-MEM 96a R=-MEM 96b R=-MEM ing from protonation of (81) t being detected by glc analysis of the crude reaction i s o l a t e . A respectable 70% y i e l d of the cyclopropyl esters (89) could be consistently obtained. i OR 150 R=-Me 137a R=-MEM The "'"H nmr spectrum of the ester mixture (89) showed one signal for the methyl protons of the carbomethoxy groups (a s i n g l e t at 63.68) and one sing l e t (at 63.35) for the methyl ether protons. The "^H nmr spectrum of the ester mixture (96) s i m i l a r l y showed one singlet at 63.68, due to the methyl ester protons of both isomers, and another at 63.41 due to the meth-- 96 -oxy protons of the -OMEM groups. For both ester mixtures (89) and (96), however, the doublet due to the proton of the endo-isomer was distinguishable from that of the e'xo-isomer. Thus, the doublet of exo-ester (89b) occurred at 63.83 (J=5Hz) while the doublet of the ehdo-ester (89a) was loc a t -ed at 63.95 (J=5Hz). For the cyclopropyl ester mixture (96) the corresponding doublets (J=5Hz) for the exo- and endo-esters, (96a) and (96b), were at 64.26 and 64.36, respectively. The i r spectra of both ester mixtures (89) ( V Q = Q 1715 cm-^") and (96) ( V^ - Q 1725 cm"1) appeared to be in accord with expectations and t h e i r mass spectra showed fragment ions (e.g. M+-OCH3 and M+-C02Me/M+-C02Me-OMEM) in d i c a t i v e of the methyl ester f u n c t i o n a l i t y . Treatment of the mixture of monobromides (86a-d)(endo-to exo-bromides r a t i o ^9:1) with 2.2 equiv. of t e r t - b u t y l l i -thium and subsequent bubbling of dry carbon dioxide into the re s u l t i n g solution of lithiocyclopropanes (87_) gave, a f t e r workup, a 74% y i e l d of a mixture of the acids (151a-d). Of these only the major one, (151a), had the correct stereochemis-t r y required for the preparation of the vinylcyclopropane (72a). Our hope in preparing the s o l i d acids (151a-d) (as opposed to the corresponding esters which would probably be l i q u i d ) was that the desired acid (151a) would be separable from the mix-ture of acids (151a-d) by f r a c t i o n a l c r y s t a l l i z a t i o n . Indeed, af t e r two successive r e c r y s t a l l i z a t i o n s of the iso l a t e d mixture of acids from petroleum ether, there was obtained a crop of _ 97 _ white c r y s t a l s which melted sharply (57.5-58.5°C) and appeared to consist of just one compound (:151a) on analysis (by both 270 MHz "'"H nmr and c a p i l l a r y g l c ) . To our dismay, product ana-l y s i s ( e s p e c i a l l y c a p i l l a r y glc) a f t e r subsequent reactions with t h i s material indicated that i t contained 3-4% of an i s o -mer which was most probably (151b). Scheme XXVI 86a-d 87a-d 151a 151b 151c-d The r e c r y s t a l l i z e d material gave a s a t i s f a c t o r y elemental analysis for CgH^Og and i t s mass spectrum, with peaks due to M -CH3, M -HgO and M -COgH ions, was consistent with the pre-125 sence of methyl and carboxyl groups . The i r spectrum, obtained as a nujol mull, c l e a r l y indicated the presence of a carboxyl group (absorptions at 2660 and 1690 cm - 1). In the 270 MHz ^H nmr spectrum just one secondary methyl doublet at 61.10 (J=7Hz) was v i s i b l e . - 98 -VI, ,Preparation of the Cyclopropyl Esters (100) As explained e a r l i e r , the copper catalysed addition of 69 ethyl diazoacetate to the tetrasubstituted o l e f i n (99) seemed a p o t e n t i a l l y viable one-step approach to the cyclopro-pyl esters (100) and we were anxious to test i t s p r a c t i c a b i l i t y . After some experimentation i t was discovered that, under the optimum reaction conditions (120°C, freshly prepared copper bronze), the inseparable cyclopropyl esters (100) could only be i s o l a t e d in a disappointing 33% y i e l d . A substantial quan-t i t y of a less v o l a t i l e , higher molecular weight (M + at m/e 312) material was also obtained. The molecular weight of t h i s material suggested that i t had been formed by the incorporation of two carbethoxycarbene (:CHC02Et) units into the o l e f i n (99). It seems reasonable to suggest that the formation of the by-126 127 product might have occurred v i a i n s e r t i o n ' of the car-bethoxycarbene into one of the methylene carbon-hydrogen bonds of the methoxymethyl group of (99_), p r i o r to addition to the double bond, to give compounds (152). The "''H nmr spectrum of t h i s undesired material was quite complex but was not incon-sistent with the proposed structure (152). Despite the low y i e l d , the diazoester addition route to the esters (100) was s t i l l considered a p r a c t i c a l one. The al t e r n a t i v e procedure, involving the dibromocyclopropane (101) (see Scheme XIII) , consists of three steps and would require approximately 70% y i e l d from each step to achieve a 33% o v e r a l l - 99 -N 2CHC0 2Et E t0 2 C OMe 99 Cu(Zn) 120°C OMe 100a E t0 2 C (61) 152 y i e l d . The mixture of esters (100) exhibited the c h a r a c t e r i s t i c s -1. in i t s i r spectrum (e.g. absorption at 1715 cm ) and i t s mass spectrum (e.g. M +-CH 3 > M+-CH"3OH and M +-0C 2H 5 peaks) that were f u l l y consistent with i t s structure. The 400 MHz "*"H nmr spectrum showed several i n t e r e s t i n g features which w i l l be discussed. Two sets of signals were i d e n t i f i a b l e for a l l but the cyclopropyl proton and the cyclo-hexyl methylene protons. On the basis that the carbethoxy group should deshield the protons of the methyl and methoxy-128 methyl groups c i s to i t i n (100b), the major t e r t i a r y methyl singlet at 61.32 (the minor t e r t i a r y methyl si n g l e t appeared at 61.22) was assigned to the exo-ester (100b). S i m i l a r l y , of the two AB type doublet of doublets (at 63.34 and 63.69 with J=10Hz), the major one at lower f i e l d was - ioo -assigned to the methylene protons of the methoxymethyl group of (100b). It was observed that the methoxy singlet of (100b) (at 63.31) was not deshielded with respect to that of (100a) (at 63.33). This observation could be understood i f one con-siders that, for s t e r i c reasons, in the exo-ester (100b) the methoxy group of the methoxymethyl side chain would be pre-f e r e n t i a l l y oriented well away from the carbethoxy group. Deshielding of the methoxy protons by the carbethoxy group would not then be possible. The methyl protons of the carbe-thoxy group occurred as two overlapping t r i p l e t s (at 61.25 and 1.26, J=7Hz) while the methylene protons occurred as overlapp-ing quartets (at 64.09 and 4.12, J=7Hz). In each case the major signal was the one at s l i g h t l y higher f i e l d . A r a t i o of 1.8:1,(100b) to (100a),was obtained from the integration of one half of each AB doublet of doublets in the "*"H nmr spectrum of the mixture. VII. Preparation of the Cyclopropyl Alcohols (90), (97), (103) and (153) These alcohols were prepared in good y i e l d by lithium aluminum hydride reduction of the corresponding cyclopropyl esters (or carboxylic acid in the case of (153). Reduction of the ester mixture (9J3) resulted in a 70% y i e l d of the mixture of alcohols (97a) and (97b). The alcohol _ 101 _ mixture exhibited i r absorptions Qe.g. 3450 cm - 1) and mass spectral c h a r a c t e r i s t i c s (e.g. M+-0CH3 and M+-0MEM-H peaks) expected for compounds containing hydroxy and -OMEM groups. The ''"H nmr spectral features were as expected except that the H c protons of (97a) and (97b) occurred as a single + \\\ ^ci H^l QvEM OMEM OMEM 96a-b 97a 97b (62) broad doublet at 6 4.24 (J=5Hz). This was contrary to what was observed in the case of the ester mixture (96a-b) where sepa-rate doublets for the H^ , proton of each isomer were obtained. The methylene protons of the hydroxymethyl groups gave r i s e to signals that were obscured by the multiplet due to the four methyleneoxy protons (-OCHgCHgO-) of the -OMEM group. The alcohol mixture (90a-b) is o l a t e d (70% yi e l d ) on re-duction of the esters (89) also exhibited a single doublet (63.79, J=5Hz) for the H"c protons of (90a) and (90b). However, the methylene protons of the hydroxymethyl group of (90a) and (90b) occurred as separate doublets (J=7Hz) at 6 3.45 (major) and fi3.61 (minor). Since the r a t i o of the s t a r t i n g esters (89a-b) was 1.7:1 i n favour of the exo-ester (89b), the lower f i e l d doublet was assigned to the endo-alcohol (90a) and the - 102 -higher f i e l d doublet to the exo-alcohol (90b). Based on the integrated i n t e n s i t i e s of these two doublets, the r a t i o of (90b) to (90a) in the i s o l a t e d mixture was found to be 1.5:1. (63) 89a-b 90a 90b The i r absorptions (e.g. 3320 cm"x) and mass spectral c h a r a c t e r i s t i c s (e.g. M+-H20 and M+-0CHg peaks) were f u l l y consistent with expectations for the alcohol mixture (90). Reduction of the mixture of esters (100) provided an i n -separable mixture of the alcohols (103a) and (103b) in 81% y i e l d . Both the i r and mass spectra of t h i s mixture were i n -terpreted in terms of the assigned structures. The 400 MHz *H nmr spectrum of the mixture was i n t e r p r e t -ed in some d e t a i l with the aid of proton decoupling experiments (Fig. 5a-e). The s i n g l e t s at 61.09 (major) and 61.17 (minor) were c l e a r l y due to the t e r t i a r y methyl protons of the two isomers. The s i n g l e t s at 63.33 (minor) and 63.36 (major) were also e a s i l y assignable to the methoxy protons. Since the r a t i o of the alcohols (103a) to (103b) was not expected to d i f -f e r s i g n i f i c a n t l y from the r a t i o of the esters (100a) to (100b), - 1 0 3 -the major signals at 5.1.09 and S3.36 were assigned to the exo-alcohol (103b). An AB type doublet of doublets (J=10Hz) at 8 (64) lOOa-b 103a 103b 63.31 and a s i n g l e t at 63.47 were assigned to the methylene protons at of the endo- and exo-alcohols, respectively. This assignment was f a c i l a t e d by the fact that addition of D 20 to the "^H nmr sample of (103a) and (103b) caused the doublet of doublets at 63.31 to collapse into a singlet (Fig. 5b). It then became clear that the singlet at 63.47 was the major of the two signals due to the methylene protons at C^. The doublet of doublets at 63.51 (J=10, 12Hz) and the multiplet at 63.74-3.86 (Fig. 5b) were without doubt due to the Cg methylene protons of the two isomers, (103a) and (103b). However, i t was quite unclear how these signals were to be * While we haye no convincing explanation of t h i s unusual observation, i t should be pointed out that changes in solvent can cause quite dramatic differences in the appearance of ~*"H nmr spectra owing to changes in chemi-129 c a l s h i f t s and coupling constants F i g . 5b. The 400 MHz H^ nmr spectrum of (103a-b) i n CDCl„-D o0. Fig. 5e. The 400 MHz H nmr spectrum of (lQ_3_a-b) on i r r a d i a t i o n of the signal due to H n in the region 61.04-1.14. - 109 -assigned u n t i l the proton decoupling experiments (Fig. 5c-e), about to be described, were performed. I r r a d i a t i o n of the t r i p l e t (J=7Hz) at 50.72, due most probably to R"A of (103a), caused the u p f i e l d portion of the multiplet at 6 3.74-3.86 to simplify into an AB type doublet of doublets at 6 3.77 with J=10Hz. The downfield portion of t h i s multiplet was apparent-l y unchanged and became recognisable as a doublet of doublets centred at 63.82 with J=6, 12Hz (Fig. 5c). Ir r a d i a t i o n of the entire multiplet at 6 3.74-3.86, as expected, caused collapse of the t r i p l e t (J=7Hz) at 60.72 into a s i n g l e t . In addition, the doublet of doublets at 63.51 collapsed, e s s e n t i a l -l y , into a doublet (J=10Hz). It was obvious, too, that a mul-t i p l e t (apparently a doublet of doublets), a l l but obscured by the methyl group at 61.09, narrowed by ^6Hz to become a doublet with J=10Hz (Fig. 5d). Irra d i a t i o n of t h i s obscured si g n a l , in the region 61.04-1.14, resulted in the doublet of doublets at 63.51 collapsing into a doublet (J=12Hz) and the lower f i e l d portion ( e a r l i e r recognised as a doublet of doub-l e t s with J=6, 12Hz) of the multiplet at 63.74-3.86 colla p s i n g into a doublet with J=12Hz (Fig. 5e). Based on the assignment of the doublet of doublets (J=6, 10Hz) at 61.10 to the proton Hg of (103b) and the t r i p l e t at 0.72 (J=7Hz) to the proton H^ of (103a) ( t h i s assignment i s supported by the r e l a t i v e i n t e n s i t i e s of these signals) the information from the decoupling experiments described above leads to the following conclusions: - HQ ~ (1) The higher f i e l d portion of the multiplet at 63.74-3.86 was due to the methylene protons H"x and H y ) at Cg of (103a), and was a c t u a l l y the A B part of an A B X system (H Y, H x and of (103a)) for which J X Y=10Hz and j A X = J A Y = 7 H Z -( A B d of d of d) 63.77 J x y=10Hz J A X = 7Hz J A Y = 7Hz 61.10—^ U u (d of d) H B O L N / 1 ^ 63.82 (d of d) 60.72 (t) 103a OMe 103b 63.51 (d of d) J x y=12Hz J B X = 6Hz J B y=10Hz (2) The lower f i e l d portion of the multiplet at 63.74-3.86 consisted of a doublet of doublets (J=6, 12Hz) at 63.82 due to one of the methylene protons (say H x) at Cg of (103b). The remaining proton (H Y) at C g of (103b) must be assigned to the doublet of doublets (J=10, 12Hz) at 63.51. Reduction of the carboxylic acid (151a) with lithium aluminum hydride in ether was slow at 0°C but proceeded r a p i d l y at room temperature. C a p i l l a r y glc analysis of the material i s o l a t e d indicated that i t consisted almost e n t i r e l y (%97%) of the alcohol (153a). A small amount of another component (^3%), t e n t a t i v e l y i d e n t i f i e d as (153b) was also present. In the 270 MHz "'"H nmr spectrum of the alcohols (153) only one secondary methyl doublet (due to (153a)) at 61.11 (J=7Hz) - I l l -(65) 153b was apparent. The doublet at 53.73 (J=7Hz) was assigned to the methylene protons of the hydroxymethyl group of (153a) Another doublet (small but c l e a r l y v i s i b l e ) at 53.64 (J=7Hz) was assigned to the corresponding protons of (153b). Ir ab-sorptions (e.g. 3300 cm - 1) and mass spectral c h a r a c t e r i s t i c s (e.g. M+ and M^I^O peaks) were consistent with expectations for the mixture of (153a) and (153b). - 112 -- 113 -In the presence of anhydrous sodium acetate the cyclo-propyl alcohols ( 9 0 ) , (97), CIO3) and (153) were smoothly 130 oxidized with pyridinium chlorochromate (PCC) to the corresponding aldehydes, (91), '(98). (104) and (154), in good 130 y i e l d s . Following Corey's procedure , these oxidations were performed at room temperature in dry methylene chloride with a,1.3 equiv. of pyridinium chlorochromate and, in each case, reaction was complete after 2h. D i s t i l l a t i o n of the crude i s o l a t e d material gave aldehyde pure enough (>95% by glc) for immediate use in the next synthetic step. The material i s o l a t e d in 78% y i e l d on PCC oxidation of the alcohol mixture (90) was shown to consist almost e n t i r e l y (^99% by glc) of the aldehydes (91a) and (91b). This material (66) • t t OMe OMe OMe 90a-b 91a 91b showed the diagnostic aldehyde peaks at 2805, 2700 and 1700 cm - 1 in i t s i r spectrum. Two low f i e l d doublets, the major at 69.12 and the minor at 69-45 (J=5.5Hz in each case), in the "''H nmr spectrum of the mixture ("9.1 a-b), were assigned to the alde-hyde protons of the exo- and endo-aldehydes.respectively. Integration of these two doublets indicated a 3:2 r a t i o of (91b) - 114 -to (91a). The aldehyde mixture (98), obtained by oxidation of (97), was i s o l a t e d in 79% y i e l d and exhibited i r absorptions (e.g. 2790, 2700 and 169Q cm"1) and 1H nmr c h a r a c t e r i s t i c s (e.g. doublets of 69.11 and 69.45 with J=5.5Hz) in accord with ex-pectations. (67) i i ; O M E M O M E M OMEM 97a-b 98a 98b A f a i r l y low y i e l d (55%) of the aldehydes (154a-b) was obtained from the oxidation of the alcohols (153) (containing 97% (153a) and 3% (153b)). This might have been the res u l t of polymerisation since a substantial quantity of a viscous o i l was l e f t behind on d i s t i l l a t i o n of the aldehydes. These alde-hydes were u t i l i z e d immediately aft e r d i s t i l l a t i o n , but a small sample did show the expected i r (e.g. 1690 cm-"1' absorp-tion) and "^H nmr (e.g. doublet at 69.63 with J=6.5Hz) features. The aldehyde mixture (104a-b) was obtained in 80% y i e l d on oxidation of alcohol mixture (103a-b). The mixture (104a-b) exhibited the t y p i c a l aldehyde absorptions at 2840, 2805, 1715 and 1685 cm"1 in i t s i r spectrum. In the nmr spectrum,the aldehyde protons of (104b) and (104a) gave the expected doublets - 115 " (68) 154b at 69.60 (major) and 69.72, respectively. Integration of these signals yielded a value of 7:3 for the exo- to endo-aldehyde r a t i o . (69) 103a-b 104a 104b - 1 1 6 -. -The aldehydes (.156), (158), (160), (162) and (164) were prepared by dir e c t formylation of the corresponding cyclopro-72a p y l l i t h i u m intermediates . Although several supposedly 72b more e f f i c a c i o u s formylating reagents have been reported , N,N-dimethylformamide was the reagent of choice for us because of the great convenience associated with i t s use and the sat-i s f a c t o r y y i e l d s obtained with i t . Solutions of the required lithiocyclopropanes were pre-pared as usual at -78°C, by addition of ethereal solutions of the corresponding monobromocyclopropanes to t e r t - b u t y l l i t h i u m (2.4 equiv.) in ether-pentane. Excess N,N-dimethylformamide was then added and after an appropriate reaction time (usually 5-7h), the reaction mixtures were worked up according to the 72a procedure of Evans . The desired aldehydes were obtained in 90-98% purity (as determined by glc a n a l y s i s ) , contaminated only with the material (2-10%) r e s u l t i n g from protonation of the corresponding lithiocyclopropane derivative. After i r and, in most cases, nmr spectra were recorded the aldehydes were then u t i l i z e d d i r e c t l y without further p u r i f i c a t i o n in the Wittig methylenation reaction. Compound (156) was prepared, as described, from the mono-bromide (146a) (Scheme XXVII). The aldehyde was >90% pure by glc analysis and showed the required bands at 2840, 2700 and 1690 cm - 1 i n i t s i r spectrum. This material was u t i l i z e d d i r e c t l y for the preparation of the vinylcyclopropane (73). - 1 1 7 -Scheme XXVII 146a 155 156 Preparation of the aldehyde (158) from the monobromide (147a) proceeded smoothly to y i e l d material of acceptable purity (94% by g l c ) . The i r spectrum of (158) showed the ab-sorptions i n d i c a t i v e of both the aldehydic and phenyl groups with bands at 2850, 2725, 1675, 755 and 700 cm - 1. The 1H nmr spectrum also c l e a r l y indicated the presence of these two Scheme XXVIII 147a 157 158 groups. Thus, a one proton doublet (J=5.5Hz) at 69.84 and a f i v e proton multiplet at 67.19-7.38 were in evidence. A very small doublet (J=5.5Hz) at 69.03 was interpreted as in d i c a t i n g the presence of a few percent of the exo-isomer of (158). - 1 1 8 -Application of the formylation sequence to the monobro-mide (148) gave an excellent y i e l d (83%) of the aldehyde (160) in o08% purity (by g l c ) . Typical aldehyde absorptions at 2830 and 1685 cm - 1 were observed in the i r spectrum of (160). The *H nmr spectrum showed the required aldehyde doublet (J=6.5Hz) at 69.90 and the remainder of the spectrum was as expected for Scheme XXIX B ^ L i , ether OHC H 148 159 160 structure (160). The mass spectrum of (160), with peaks due to M+, M+-H and M+-CH0 ions, served to confirm the structure of t h i s aldehyde. The mixture of monobromides (145a-d) , containing 88% of (145a),led to the mixture of aldehydes (162a-d) in which one - 119 -isomer, presumably (162a), predominated. The crude mixture of aldehydes (162a-d) showed bands at 2820, 2675 and 1680 cm -1 Scheme XXX 145a-d Bu t L i , ether -78°C SiMe-DMF; OHC 161 OHC SiMe- 162a in i t s i r spectrum. Its H nmr spectrum, with a doublet (J=6Hz) at 69.71 and s i n g l e t s at 60.00 (minor) and 60.02 (major) con-firmed the presence of the aldehyde and t r i m e t h y l s i l y l groups. Aldehyde (164) was obtained ^ 94% pure (by glc analysis) on subjecting the endo-monobromide (149a) to the sequence in Scheme XXXI. The i r (absorptions at 2840, 2740 and 1685 cm - 1) and "^H nmr spectra (e.g. doublet at 69.43 with J=6.5Hz) of the aldehyde (164) were f u l l y consistent with i t s structure. - 120 -Scheme XXXI 164 IX. Preparation of the Vinylcyclopropanes (65)-(77) Preparation of compounds (65) and (71) proceeded accord-ing to the synthetic plans outlined in Schemes X and XI, respectively (see p. 39). Thus, treatment of a cold (-78°C) ethereal solution of the monobromides (8()) or (86) with t e r t -b u t y l l i t h i u m (2.2 equiv.), conversion of the r e s u l t i n g solution of the lithiocyclopropanes (81) or (87) to the corresponding phenylthiocuprates (82_) or (8j8) followed by addition of 52 3-iodo-2-cyclopenten-l-one (83_) resulted in the formation of the mixture of enones (65a-b) or (71a-d). The "'"H nmr spectrum of the crude mixture of enones (65a) and (65b) (1:1.66 by glc analysis) c l e a r l y showed two signals for the methoxy protons and for the protons H c and E^. After separation of the isomers (65a) and (65b) by s i l i c a gel prepa-r a t i v e layer chromatography (40% ethyl acetate-petroleum ether), p o s i t i v e i d e n t i f i c a t i o n of each was based on the chemical - 121 -(70) 80 65a 65b 58 s h i f t s of the cyclopentenyl protons (H z). Previous work in our laboratory had shown that the o l e f i n i c cyclopentenyl pro-ton occurred at lower f i e l d in the "'"H nmr spectra of the endo-enones, for example, (58a) than i t did in the "'"H nmr spectra of the corresponding exo- derivatives, for example, (58g). On th i s basis the more mobile f r a c t i o n (with H„ at 65.96-6.04) ZJ turned out to be the endo-enone (65a). The exo-enone (65b) 58a 58g (H z at 65.78-5.88) made up the more polar f r a c t i o n . The s l i g h t l y impure endo-enone (65a) obtained from preparative t i c of the crude enone mixture (6_5) was p u r i f i e d by column chromatography to y i e l d the pure compound (65a). The uv (X =237nm, e=16;000) and i r (absorptions at 1700, 1675 - 122 -and 1610 cm""") spectra were consistent with the presence of an a8-unsaturated ketone in (65a). The "*"H nmr spectrum showed a methoxy sin g l e t at 63.35, a doublet (J=5Hz) due to at 63.75 and a one proton o l e f i n i c s i g n a l due to H^ , at 65.96-6.04. The exo-enone (65b) also exhibited uv ( X m Q v 246nm, _ _ _ _ _ I T l c L X e=16,000) and i r (1690 and 1665 cm - 1) spectral c h a r a c t e r i s t i c s i n d i c a t i v e of an aB-unsaturated ketone. In the nmr spectrum, , as expected, was at higher f i e l d (65.78-5.88) than H^  of (65a) and H^ occurred as the expected doublet (J=5Hz) at 63.83. The mixture of enones (71a-d) was f i r s t separated into two f r a c t i o n s by preparative layer s i l i c a gel chromatography (30% ether-petroleum ether). On the basis of the o l e f i n i c signals, the less polar f r a c t i o n (H^ multiplets at 66.00 and 66.11) was deemed to be a mixture of the endo-enones (71a-b). The more polar f r a c t i o n (overlapping H^  m ultiplets at 65.78) consisted mainly of the exo-enones (71c-d) but contained a substantial amount (^20% by glc) of another compound. A sample of t h i s l a t t e r material, obtained by preparative glc of the crude lower f r a c t i o n , exhibited i r , nmr and mass spectral features which i d e n t i f i e d i t as the B-phenylthioenone (165). The exo-enone mixture (71c-d) was the f i r s t eluted f r a c t i o n from the preparative glc mentioned above. Its uv (A 253nm, in 3.x e=25,200) and i r (.1690 and 1660 cm - 1) spectra were in accord with expectations (cf. exo-enone (65b)). The "'"H nmr spectrum f u l l y supported the s t r u c t u r a l assignment. Thus, the one pro-- 1 2 3 -2) PhSCu 3) 83 l)Bu L i 0^  C71) 8 6 71a 71b 71c-d ton (H^) multiplet at 65.70-5.85 indicated the presence of the cyclopentenone ri n g and two methyl doublets (J=6Hz) at 6 0.98 (minor) and 61.10 (major) indicated the two isomers, (71c) and (71d), of the mixture. The trans-methyl-endo-enone (71a) was obtained as pure, white c r y s t a l s on f r a c t i o n a l c r y s t a l l i z a t i o n of the mixture (71a-b) from petroleum ether. The c r y s t a l s melted sharply at 7 4 ° C and gave a s a t i s f a c t o r y elemental analysis for C - . O H - . Q O . 1680 and 1615 cm~~L) spectra indicated an aB-unsaturated enone. The *H nmr spectrum was in accord with the structure (71a) showing, s i g n i f i c a n t l y , a single methyl doublet (J=6Hz) at 61.11 and a one proton o l e f i n i c multiplet, c l e a r l y due to H z, at 65.94-6.04. 165 The uv (A max 242nm, e=14,000) and i r (absorptions at 1700, - 124 -The aldehyde mixtures (91), (98) and (104) reacted with 131 methylenetriphenylphosphorane to give the vinylcyclopropane mixtures ( 6 6 ) , (67) and (68). In each case the endo- and exo-isomers were separated by preparative t i c on s i l v e r n i t r a t e 132 impregnated s i l i c a gel using low wavelength u l t r a - v i o l e t i r r a d i a t i o n as the v i s u a l i z i n g agent. As was anticipated, the endo-alkene was the chromatographically more mobile isomer for each of the mixtures (66)-(68) since the less s t e r i c a l l y encumbered double bond of the exo-isomer would be more strongly bound by the adsorbent. The vinylcyclopropanes (66a-b) were obtained in 95% y i e l d . The minor, more mobile f r a c t i o n i s o l a t e d a f t e r s i l v e r n i t r a t e -132 s i l i c a gel preparative t i c of the mixture was thought to be the endo-alkene (66a) both on the grounds of i t s chromatogra-phic mobility and because in the s t a r t i n g aldehydes (9_1) the endo-isomer (91a) was the minor component ( r a t i o of endo- to exo-aldehydes ^1:1.3). We sought corroboration of our assignments by comparing the r e l a t i v e chemical s h i f t s of the and protons of the endo- and exo-isomers (65a) and (65b) with the r e l a t i v e chemi-c a l s h i f t s of the corresponding protons in (66a) and (66b). It was found that, by analogy with the observations for the endo- and exo-isomers (65a) and (65b), the proton of endo-isomer (66a) occurred at higher f i e l d (doublet at 63.70 with J=6Hz) than the H c proton (doublet at 63.83, J=5Hz) of the exo-isomer (66b). - 125 -Ph 3P=CH 2 H D ' ' + OMe (72) 66a 66b The si g n a l due to the proton H z of (66a) was also expect-ed, on the basis of observations with (65a) and (65b),to be at lower f i e l d than that of (66b). Assignment of the o l e f i n i c protons in the "*"H nmr spectra of (66a) and (66b),based on the m u l t i p l i c i t y and coupling constants of t h e i r signals, did con-firm our expectation as the following discussion shows. Assuming the chemical s h i f t difference between H v and H„. 133 i s s u f f i c i e n t l y large ( A V y Z / Jy Z>10) , one can predict that the signal due to would be a doublet of doublet of doublets with two small coupling constants, J y Z a n d ^\yz (koth ^ n 134 135a range 0-2Hz) ' and one large trans coupling constant J x z ( i n the range 10-20Hz) . The si g n a l due to H v should be quite s i m i l a r showing two small coupling constants, J y Z and J ^ y , and a much larger one J^y* However, the major coup-l i n g constant, J ^ y , i s a c i s coupling constant and should be 135 less than In the 400 MHz "*"H nmr spectrum of (66a) (where the con-d i t i o n A V y Z / J y Z >10 holds) the doublet of doublet of doublets - 126 -.82 66a 66b at ^5.07 (J=1.5, 2 and 10Hz) was assigned to H v and that at 65.21 (J=1.5, 2, 17Hz), with the larger (trans) coupling con-stant, was c l e a r l y due to H^. A t h i r d o l e f i n i c doublet of doublet of doublets centred at 65.53 (J=7, 10, 17Hz), in which the coupling constants J ^ , JXY' a n c * JZX w e r e i d e n t i f i a b l e , was obviously due to E^. In the 80 MHz "'"H nmr spectrum of (66b) the o l e f i n i c signals were, naturally, not as well dispersed as in the 400 MHz spectrum of (66a). It was not surpri s i n g , then, that H z and H y occurred as two p a r t i a l l y overlapping AB type doublet - 127 -of doublets CAvzy / J Z Y <10) at 64.73-5.13 with J z y=2.5Hz. A seven l i n e multiplet to s l i g h t l y lower f i e l d (65.20^5.70) was assigned to H x. It was s t i l l possible to dist i n g u i s h between the Hy and H z portions of the two AB subspectra on the basis of the r e l a t i v e magnitudes of the coupling constants J X Y and J x z - Thus, the higher f i e l d AB doublet of doublets was due to H v and the lower f i e l d doublet of doublets to H^ .. A l l the evidence that could be gleaned from the 1H nmr spectra suggested that we had co r r e c t l y i d e n t i f i e d the i n d i v i -dual vinylcyclopropanes (66a) and (66b). Later, homo-[l,5]-hydrogen migration in the isomer i d e n t i f i e d as the endo-vinyl-cyclopropane (66a) provided i r r e f u t a b l e proof of the correct-ness of our assignments. The i r spectra of (66a) and (66b) were quite s i m i l a r (e.g. absorptions at 1620 and 1630 cm - 1, respectively) to each other and in accord with expectations. The mass spectra of these two compounds gave c h a r a c t e r i s t i c peaks, for example, M+, M+-CHg and M+-CHgOH , supporting the proposed structures. The mixture of vinylcyclopropanes (67a-b) was isola t e d in 87% y i e l d and was separated by s i l v e r n i t r a t e - s i l i c a gel 132 preparative t i c into the endo- and exo- compounds (67a) and (67b). The i r spectra of these two compounds showed the ex-pected absorptions for the v i n y l substituent and the -OMEM moiety. The mass spectrum of each isomer showed no molecular ion (M +) peak but other diagnostic peaks (e.g. M+-CH20 and M+-0MEM) were i d e n t i f i a b l e . The differences between the 1H - 128 -nmr spectra of the endo- and exo-alkenes (67a) and (67b) were sim i l a r in nature to those already discussed for the.isomer (73) 98a-b 67a 67b pairs (65a), (65b) and (66a) and (66b). Thus, the doublet (J=5Hz) due to the K c proton of (67a) occurred at somewhat higher f i e l d (54.16) than that due to the proton (doublet at 64.24, J=5Hz) of (67b). In addition, the signals due to the H z and Hy protons of (67a) (multiplet at 65.02-5.36) were at lower f i e l d than those of (67b) (multiplet at 64.73-5.13). The i n d i v i d u a l vinylcyclopropanes (68a) and (68b) were 132 obtained by s i l v e r n i t r a t e - s i l i c a gel preparative t i c of the mixture (68a-b). The i r spectra of the two isomers were quite s i m i l a r , as would be expected, and showed carbon-carbon double bond stretching absorptions at 1630 and 1620 cm - 1, respectively. The mass spectra of (68a) and (68b) were also quite s i m i l a r , showing the expected peaks (e.g. M+, M+-CHg, M+-0CH3 and M +-0C 2H 5) in each case. The nmr spectra of both vinylcyclopropanes (68a) and (68b) were f u l l y in accord with expectations. Interestingly, while the methylene protons - 129 -(74) 104a-b 68a 68b of the methoxymethyl group of (68b) appeared as a clean sing-l e t at 63.39, these protons in (68a) gave r i s e to an AB type doublet of doublets (J=10Hz) at 63.33. The aldehyde (154a) (containing a small quantity of (154b)) was allowed to react with methylenetriphenylphosphor-131 ane to produce the o l e f i n (72a) (containing 4% of (72b)) in a modest 41% y i e l d . The low y i e l d may be attributed, at least in part, to the v o l a t i l i t y of the product alkenes. The mass spectrum of the vinylcyclopropanes (72a-b) showed a series of M - c n H 2 n + i fragment ions ,the base peak be-ing due to the M+-C^Hg ion.The i r spectrum of the mixture d i s -played the expected absorptions at 3050 and 1620 cm - 1. The % nmr spectrum (400 MHz) was f u l l y consistent with that expected for the major component. Thus, a secondary methyl doublet (J=6.5Hz) was i d e n t i f i a b l e at 61.10 and the signals due to the o l e f i n i c protons (H v, H^ and H^) c l o s e l y resembled,in chemical s h i f t s and coupling constants,those due to the corresponding protons in the endo-vinylcyclopropane (66a) already described. - 130 -(75) H v and H z appeared as doublet of doublets at 65.07 (J=2.5, 10Hz) and 65.19 (J=2.5, 17Hz), respectively and H x occurred as a doublet of t r i p l e t s at 65.68 (J x v=J x w=10Hz, J x z=17Hz). The aldehydes (156), (158), (160), (162) and (164) were obtained by trapping of the corresponding lithiocyclopropanes (155), (157), (159), (161) and (163) with DMF, as described e a r l i e r , and were u t i l i z e d d i r e c t l y without p u r i f i c a t i o n in 131 the Wittig reaction with methylenetriphenylphosphorane The crude products i s o l a t e d from these reactions were p u r i f i e d by chromatography and/or d i s t i l l a t i o n to y i e l d the desired vinylcyclopropanes, (69), (70_) and (72-77). The vinylcyclopropane (73), prepared as described above from the aldehyde (156), was not i t s e l f a substrate for re-arrangement in t h i s study but was to be converted into the desired compounds (74) and (75). In practice i t was, therefore, convenient to convert the crude tetrahydropyranyl ether (73) into the alcohol (74) and then to pu r i f y the l a t t e r . However, in order to characterise compound (73) a pure sample was - 131 -(76) OTHP OTHP 156 73 obtained by preparative t i c (5% ether - petroleum ether) of a small amount of the crude material. The mass spectrum, i r and "'"H nmr spectra of t h i s compound were f u l l y in accord with i t s structure. The conversion of the THP ether (73) into the alcohol (74) was accomplished by s t i r r i n g an ethanol solution of the 80 former with ^0.1 equiv. of pyridinium p_-toluenesulphonate at 50°C for 2h. Flash chromatography 1 1 5 of the crude material i s o l a t e d from t h i s reaction afforded the pure (99% by glc) alcohol (74),in 40% o v e r a l l y i e l d from the monobromide (146a). Compound (74) showed the expected features in i t s i r spectrum (absorptions at 3300, 3050 and 1620 cm - 1) and in i t s + + 1 mass spectrum (e.g. M -H and M -Ho0 peaks). The H nmr spectrum - 132 -(80 MHz) emphasised the absence of the THP moiety. Thus the (77) multiplet at 64.63-4.88 due to the protons H^  in the diastereo-meric mixture (73) was absent from the "'"H nmr spectrum of (74). The broad three proton multiplet (due to H c, H g and HK>) found at 63.33-4.13 in the nmr spectrum of (73_) was replaced by a much narrower one proton signal due to H^ at 63.72-4.00 in the spectrum of (74). Treatment of a THF solution of the alcohol (74) with sodium hydride followed by addition of methyl iodide to the r e s u l t i n g sodium alkoxide yielded, after workup, the ether (75) in 73% y i e l d . Compound (75) showed no 0-H stretch in i t s i r spectrum but retained the absorptions at 3050 and 1620 cm - 1 found in the i r spectrum of (74). The mass spectrum showed peaks due to M+, M+-CHg, and M+-0CHg ions and was consistent with the structure of (75). The most remarkable feature of the 270 MHz ^H nmr spectrum of (75) was that the H^ proton was s h i f t e d u p f i e l d compared to i t s location (6 3.72-4.00) in the "*"H nmr spectrum of (74). - 133 -H 'W (78) 74 75 While some u p f i e l d s h i f t 137 of the signal due to the H_, proton was expected in going from (74) to (7_5) the signal due to a methine proton adjacent to a methoxy group would normally be expected to occur at lower f i e l d than the methoxy singlet (cf. compounds (65), (66) and (67)). In the present case, however, the methoxy singlet occurred at 63.38 while the multiplet was found at 63.21-3.29. It would seem, then, that the pro-ton i s better shielded by the v i n y l group in compound (75) than are the corresponding protons (H^) in (65a), (66a) and (67a). This can be explained by the fact that in the preferred conformation (75A) the proton i s pseudoaxial and thus i s placed more d i r e c t l y within the shielding cone of the double bond in t h i s compound than the corresponding protons on the f i v e membered rings of (65a), (66a) and (67a). The o l e f i n i c protons Hy, and gave the usual pattern of signals occurring as doublet of doublet of doublets at 65.09 (J=VL.5, 2.5, 10Hz), 65.22 (J=^1.5, 2.5, 17Hz) and 65.58 (J=^9.5 10, 17Hz), respectively. - 134 -75B The crude material i s o l a t e d from the Wittig reaction of the cyclopropyl aldehyde (158) with methylenetriphenylphosphor-131 ane was p u r i f i e d by s i l v e r n i t r a t e - s i l i c a gel preparative 132 t i c to afford the pure vinylcyclopropane (76) (100% by c a p i l l a r y glc) in 41% o v e r a l l y i e l d from the monobromide (147a). 147a Compound (76) exhibited i r (3050, 1620, 1595, 755 and 700 -1 + + cm ) and mass spectral c h a r a c t e r i s t i c s (e.g. M and C^H^ peaks) that c l e a r l y supported i t s structure. In the 270 MHz *H nmr spectrum of (76) the proton H c appeared as a doublet of t r i p l e t s ( J B C = J C E =3Hz, J C D=10.5Hz) at 62.61. By contrast, i n the bicyclo[3.1.0]vinylcyclopropanes such as (65a), (66a) and (67a) - 135 " the protons are v i r t u a l l y orthogonal to the trans v i c i n a l protons ( l a b e l l e d Hg and H^) and they thus appear as simple doublets. The o l e f i n i c protons H v, H z and H x of compound (76) (79) gave the usual, predictable pattern of doublet of doublet of doublets while the aromatic protons separated, s u r p r i s i n g l y , into a one proton multiplet at 67.14-7.26, presumably due to H R and a four proton multiplet at 67.27-7.35. The vinylcyclopropane (69) was i s o l a t e d in 71% y i e l d by simple d i s t i l l a t i o n of the crude product obtained from the re-action of the aldehyde (160) with methylenetriphenylphosphor-131 ane Compound (69) displayed in i t s i r spectrum the absorp-tions (3050 and 1625 cm - 1) expected for an o l e f i n and i t s nmr spectrum was c l e a r l y i n accord with i t s structure. It was in t e r e s t i n g that while for the aldehyde (160) (as well as for the monobromides (148a) and (148b)) the signal due to the pro-ton H c (doublet with J=12Hz at 64.26) was c l e a r l y separated from and downfield of the signal due to H n (doublet of doublets - 136 -148a 148b with J=3.5, 12Hz at 63.98), for the vinylcyclopropane (69) these signals are v i r t u a l l y coincident occurring as an unre-solved multiplet at 63.85-4.10. This observation was i n t e r -preted as supporting a preferred conformation for (69) in which H c i s a x i a l (see (69A)). The preferred conformation of 122 (160), deduced using arguments presented e a r l i e r , i s shown for comparison. 69A 160A The protons H Y, R"z and R"x gave r i s e to a pattern of signals that was t y p i c a l of the v i n y l protons of the v i n y l -cyclopropanes prepared in t h i s study. For t h i s reason the o l e f i n i c region of the 80 MHz 1H nmr spectrum of (69) seemed appropriate for the analysis to be described. U t i l i z i n g the Chemical s h i f t s and coupling  constants used: Chemical s h i f t s (Hz from TMS) v„ = 406.5Hz v„ = 420.0Hz ZJ v x = 482.2Hz Coupling Constants (Hz) JYZ = 2 . 5Hz JXY = 6 .0Hz JXZ = 12 .5Hz Jxw = 9 .0Hz CO 00 P P M Fig. 6b. The simulated H^ nmr spectrum of the o l e f i n i c protons of (69) using parameters obtained d i r e c t l y from the actual spectrum. Chemical S h i f t s and Coupling constants calculated using an  ABX analysis. Chemical s h i f t s (Hz from TMS): v y = 409.2Hz v z = 417.8Hz v x = 482.2Hz Coupling Constants (Hz) J y z = 2.5Hz J x y = 9.33Hz J x z = 16.67Hz J x w = 9.0Hz M W CO Fig. 6c 7 6 p p M 5 The simulated 1H nmr spectrum of the o l e f i n i c protons of (69) using calculated parameters. - 140 -chemical s h i f t s , v z, V y , and v x > and coupling constants J y Z > J z x , and (both in hertz) derived d i r e c t l y from the l i n e positions in the 80 MHz "'"H nmr spectrum of (69_), the computer simulated ^H nmr spectrum of the three spin v i n y l system (Hy, H z and H x) was generated. The simulated spectrum bore l i t t l e resemblance to the actual spectrum (cf. F ig 6a-b), starkly demonstrating that a f i r s t order approach to the analysis of t h i s system was quite inappropriate. However, using the same values of v^, J y Z and used in the previous (80) 160 69 simulation but substituting values of V2 , VY'^XZ a n c * ^XY c a l c u ~ 138 lated on the assumption that Hy, H z and H x constituted an ABX system, produced a simulated spectrum that was a reasonable f a c s i m i l i e of the actual one (Fig.6c). The simulation analysis c l e a r l y j u s t i f i e d our treatment of the signals (in the 80MHz spectra) due to the terminal o l e f i n i c protons (Hy,Hz) of the vinylcyclopropanes in t h i s study as AB spectra. Such spectra would y i e l d v a l i d J y Z values but the values of J X Y and J y Z ob-tained d i r e c t l y from these spectra would be incorrect. It should be pointed out that the greater dispersion of signals obtained in - 141 -270 and 400 MHz H nmr spectra would j u s t i f y a f i r s t order analysis of the three proton o l e f i n i c system of these v i n y l -cyclopropanes. The vinylcyclopropane (70) gave the c h a r a c t e r i s t i c ab-sorptions (3050, 1620 cm - 1) in i t s i r spectrum and i t s *H nmr spectrum (270 MHz) c l e a r l y supported the proposed structure. The o l e f i n i c protons,H Y, H z and Reappeared as doublet of doub-le t of doublets at 65.09 (J=^1.5, 2.5, 10Hz), 65.23 (J=^1.5, 2.5, 16Hz) and 65.45 (J=8, 10, 16Hz), respectively. It was (81) 164 70 observed that that both H c (doublet at 64.27, J=6.5Hz) and H R (multiplet at 62.36-2.50) were s h i f t e d u p f i e l d r e l a t i v e to t h e i r positions (64.40 and 62.70-2.82, respectively) in the dibromocyclopropane (141). The shi e l d i n g e f f e c t of the double - 142 -bond in.(70) can again be invoked to explain t h i s observation. This s h i e l d i n g of H^ ,, as well as i t s occurrence as a doublet, which implies JgQ=0, was considered firm evidence that was c i s to the vinylcyclopropane moiety as desired. The crude mixture of o l e f i n s (77a-d) obtained by Wittig methylenation of the aldehydes (162a-d) was,as expected,highly enriched in one isomer (77a). S i l v e r n i t r a t e - s i l i c a gel pre-132 parative t i c (2% ether - petroleum ether) of t h i s crude mixture resulted in the i s o l a t i o n , from the more mobile fraction, of material that consisted very lar g e l y (95%) of the v i n y l -cyclopropane (77a). The remaining 5% (by c a p i l l a r y glc) con-s i s t e d of three other components presumably the diastereomers (77b-d). The i r spectrum of t h i s material showed absorptions (3050, 162a-d 77a 77b 77c-d 1620, 1255 amd 835 cm"1) expected for both a double bond and a t r i m e t h y l s i l y l group and the mass spectrum was dominated by fragment ions (e.g. M +-SiMe 3) that indicated the presence of a t r i m e t h y l s i l y l group. The 400 MHz *H nmr spectrum showed a nine proton singlet - 143 -at 6 0.02 due to the methyl groups attached to s i l i c o n and a one proton multiplet at 6 0.51-0.59 due to the methine proton (H c) adjacent to the t r i m e t h y l s i l y l group. The remainder of the spectrum, notably the o l e f i n i c region, was in f u l l accord with expectations. X. Thermal Rearrangement of the endo-Vinylcyclopropanes (65a)-(67a),(68b), (69), (70), (71a),(72a),(74)-(76),(77a) 66a R=-Me 67a R=-MEM OMe 68b 69 H C : 71a He i 72a X=-CH, 74 X=-OH 76 X=-Ph 75 X=-0Me 77a X=-SiMe3 - 144 -A l l but one of the thermolyses were performed by heat-ing benzene solutions of the l i q u i d endo-vinylcyclopropanes in sealed,base-washed, s i l y l a t e d glass tubes (see experimental section) at 230-240°C for periods of 6-8h. In the case of (71a), the compound was heated neat at 215°C for 15 min. Glc analysis and, in a few instances, "*"H nmr analysis were u t i l i z e d to determine the composition of the crude thermolysate in each case and the product(s) was (were) then i s o l a t e d by chromato-graphy and/or d i s t i l l a t i o n . A l l the endo-vinylcyclopropanes were found to undergo clean homo-[1,5]-sigmatropic hydrogen migration under the con-d i t i o n s described above to y i e l d the expected diene products in excellent y i e l d s (80-100%). There was l i t t l e doubt that the corresponding exo-isomers would not rearrange under the same conditions since i t was demonstrated that the exo-vinyleyelo-propane (67b) remained v i r t u a l l y unchanged after being subject-OMEM 67b ed to the same thermal treatment that resulted in smooth re-arrangement of i t s endo-isomer (67a). The r e s u l t s obtained on rearrangement of the v i n y l c y c l o -propanes (65a), (66a), (67a) and (75) were reminscent of those - 145 -obtained with the compounds (58e-f) . The hydrogen migration process was highly s i t e - s e l e c t i v e in that only those products, (166) , (168), (170) and (172), that resulted from migration of the protons l a b e l l e d Hj could be i d e n t i f i e d . While the minor components (< 6%) detected on glc analysis of the crude thermo-lysates of (65a), (66a), (67a) and (75) were not p o s i t i v e l y i d e n t i f i e d , i t was clear that they were not the enol ethers, (167) , (169), CI71) and (173), that would have arisen from migration of the protons l a b e l l e d H^. Thus, in the case of the rearrangement of compound (75), c a p i l l a r y glc analysis of the thermolysed material indicated the presence of one major product (94%) and two minor compo-nents ( ^ 6%) a l l three of which were shown to be isomeric (M+=m/e 152) by GC-MS analysis. In order to determine i f 167 169 CO I "e OMEM 171 - 146 -- 147 -either of the minor components might be the enol ether (173), the 400 MHz ^H nmr spectrum of the crude thermolysate was c a r e f u l l y examined i n the region 63.40-5.00. In t h i s region, of course, any weak signals due to the methoxy protons and the o l e f i n i c proton H R of (173) would be v i s i b l e . At up to s i x -teen times the normal spectrum amplification, however, neither of these quite diagnostic signals could be discerned. Thus, i t was concluded that no enol ether (173) was formed on re-arrangement of vinylcyclopropane (7_5). In a s i m i l a r manner the enol ethers (.167), (169) and (171) were discounted as pro-ducts in the rearrangement of (65a), (66a) and (67a). The i d e n t i t y of compound (166) was c l e a r l y borne out by i t s spectral (uv, i r , ~*"H nmr) c h a r a c t e r i s t i c s . Its uv spectrum (X 227.5nm, e=18,100) was consistent with the presence of max an ag-unsaturated ketone. The fact that in the parent cyclo-propyl enone (65a) the A„ Q_, (237nm) was some lOnm to longer wavelength could be at t r i b u t e d to the conjugation between the cyclopropyl r i n g and the enone system being l o s t in going from (65a) to (166). The i r spectrum of (166) (3045, 1700, 1670, 1610 cm"1) again c l e a r l y indicated the presence of the ctB-un-saturated ketone moiety. The 400 MHz ^H nmr spectrum indicated two new o l e f i n i c protons, H^ and Hj, which occurred as two one proton multiplets at 65.60-5.64 and 65.71-5.76. The cyclo-pentenone moiety was indicated by the signal at 66.02-6.07 which c l e a r l y was due to H^. The presence of a conspicuous doublet of t r i p l e t s (J r, R=J p n=4Hz, J p i ?=7Hz) at 63.70 c l e a r l y - 148 -indicated that the proton (H^) adjacent to the methoxy group was intact in compound (166). The 400 MHz "''H nmr spectrum of compound (172) provided a l l the important s t r u c t u r a l information required. Thus the presence of the expected (Z)-propenyl side chain was indicated by the occurrence of a v i n y l methyl doublet of doublets (J=2.5, 7Hz) at 61.69, a doublet of doublet of quartets (J= 1,7,10.5Hz) at 65.55, obviously due to H x, and a doublet of doublet of quartets (appearing as a t r i p l e t of quartets with J=2.5, 10.5Hz) at 65.24 due to H^. It should be pointed out that the (Z)-stereochemistry of the propenyl double bond was expected, owing to the concerted nature of the homo-[l,5]-hydrogen s h i f t re-action. Furthermore, the magnitude of J " x w (10.5Hz) was con-sidered reasonable proof of the c i s rel a t i o n s h i p of H x and H w, although in rare cases trans coupling constants may be of si m i l a r magnitude. Two other o l e f i n i c signals - a doublet of multiplets (J=10Hz) at 65.36 and a multiplet at 65.64-5.71 -were assigned to the protons H^ and Hp respectively. F i n a l l y , i t was clear that H^ was part of a two proton multiplet (along with Hg) at 63.13-3.22. The 400 MHz ^H nmr spectrum of compound (170) showed, as well, a v i n y l methyl doublet of doublets (J=2.5, 7Hz) at 61.71 and a doublet of doublet of quartets (appearing as a t r i p l e t of quartets with J=2.5, 10Hz) due to H w at 65.20. However, the signal due to H x was almost f u l l y obscured owing to over-lap with the multiplet (65.47-5. 78) due to H A. The lowest f i e l d - 149 -si g n a l , as in (172), was that due to Hj (multiplet at 65.69-5.74). The proton (H c) adjacent to the -OMEM substituent appeared as a doublet of t r i p l e t s ( r e a l l y a. doublet of doub-l e t of doublets) at 64.15 (cf. H c of (166)) with J C D=J C B=5Hz and J C E=7Hz. Irra d i a t i o n of the signal due to H R (multiplet at 63.61-3.67) caused t h i s signal to collapse into a doublet of doublets with JgQ=5 and J^E=7Hz. In the 100 MHz spectrum of (168) the o l e f i n i c signals were not well enough dispersed to be separately assigned but four o l e f i n i c protons were nevertheless indicated from the integration of the signals. The presence of a v i n y l methyl doublet of doublets (J=2, 7Hz) at 61.72, a methoxy singlet at 63.36, and a two proton multiplet at 63.50-3.90, due to the protons H^ , and H R, a l l strongly supported the assigned struc-ture (168). In the l i g h t of the r e s u l t s just discussed, the r e s u l t s obtained on rearrangement of the vinylcyclopropane (74) were somewhat unexpected. Two products (174) and (175),in close to quantitative y i e l d (98%),were obtained in a r a t i o (glc) of 12:88. The products were separated by preparative t i c (30% ether - petroleum ether) and were i n d i v i d u a l l y character-ised . The minor, less polar (174) exhibited no 0-H absorptions in i t s i r spectrum but showed a strong absorption at 1700 cm - 1 Its 400 MHz "^H nmr spectrum showed the expected v i n y l methyl doublet of doublets (J=2, 7Hz) at 61.55, a one proton t r i -- 150 -plet of multiplets ( r e a l l y a doublet of doublet of multiplets with J A W=J x w=10Hz) at 65.19 due to and a one proton doublet of doublet of quartets (J=1.5, 7, 10Hz) at 65.36 due to H~x. S i g n i f i c a n t l y , no other o l e f i n i c signals were present. Absent as well was any signal corresponding to the proton H of the parent vinylcyclopropane (74). (87) The major rearrangement product (175) showed a strong absorption (0-H stretching vibration) i n i t s i r spectrum at 3350 cm - 1. Its "'"H nmr spectrum (400 MHz) f u l l y confirmed the structure (175). The presence of the proton H^ was confirmed by the one proton multiplet at 63.52-3.60. The (Z)-propenyl side chain was indicated by the presence of a v i n y l methyl doublet of doublets (J=2, 7Hz) at 61.73 and a t r i p l e t of multi-p l e t s ( r e a l l y a doublet of doublet of multiplets with J x w=J B W=10Hz) at 65.25 due to H^. The expected doublet of quartets due to H x appeared to be part of a two proton multiplet (with Hj as the other proton) at 65.67-5.80. The remaining o l e f i n i c s i g n a l , the one proton doublet of multiplets (J=10Hz) 13 at 65.36, was c l e a r l y due to H A. The C nmr proton decoupled - 151 -spectrum c l e a r l y showed the expected four o l e f i n i c signals as well as the signal due to the carbon atom attached to the hy-droxy group. A l l the other signals i n t h i s spectrum could be assigned to carbon atoms of (175). The vinylcyclopropane (68b) rearranged to give two pro-ducts (176) and (177) in 94% y i e l d . These compounds were poor-ly resolved by glc which yielded the unreliable value of 55:45 for the r a t i o (176):(177). The 400 MHz 1H nmr spectrum of the mixture c l e a r l y indicated that (177) was the major component and gave a r a t i o of 1:^1.8 (for (176):(177)) based on the i n t e -grated i n t e n s i t i e s of the methoxy signals of the two compounds. Preparative t i c (5% ether - petroleum ether) of the mixture enabled characterisation of the i n d i v i d u a l compounds (176) and (177). The less polar product (176), in i t s 80 MHz H^ nmr spec-trum, showed a t e r t i a r y methyl singlet at 61.14 and a broad one proton singlet at low f i e l d (65.85) which was c l e a r l y due to H^. While these two signals c l e a r l y distinguished between (176) and (177), they could be interpreted in terms of either structure (176) or i t s stereoisomer (178). The rather serious OMe 176A 178A - 152 -s t e r i c interactions that would exist between the methoxy group and the t e r t i a r y methyl group in the t r a n s i t i o n state (178A) leading to (178) makes i t reasonable to assume that compound (176) i s the enol ether a c t u a l l y formed v i a t r a n s i t i o n state (176A). The remainder of the 80MHz 1H nmr spectrum of (176) was consistent with i t s structure. Thus the methoxy sin g l e t at 63.58 was, as expected for an enol ether, at lower f i e l d than that i n the parent vinylcyclopropane (68b). The signals of the (Z)-propenyl side chain appeared as a two proton multi-plet at 65 . 26 - 5 . 60 (due to R"x and H~w) and a doublet (J=6Hz) at 61.64 due to the v i n y l methyl protons. The 80 MHz ^E nmr spectrum of compound (177) enabled (88) 68b 176 177 178 - 153 -i d e n t i f i c a t i o n of the important s t r u c t u r a l features of the com-pound. The presence of the methoxymethyl side chain intact was indicated by a methylene singlet at 6 3.56 and a methoxy si n g l e t at 6 3.38. The presence of the (Z)-propenyl side chain was indicated by a v i n y l methyl doublet of doublets (J=1.5,7Hz) at 61.65, a doublet of multiplets (J=11.5Hz) at 65.25 due to and a doublet of quartets (J=7, 11.5Hz) at 65.60 due to H x. F i n a l l y , the terminal methylene protons H^, H R appeared as a doublet (J-2Hz) and a broad s i n g l e t at 64.76 and 64.86, respect-i v e l y . The assignment of the signal at higher f i e l d to was based primarily on the fact that "trans" a l l y l i c coupling con-134 stants are generally larger than the c i s type . It i s probable, too, that i s shielded to some extent by the double bond of the (Z)-propenyl side chain. Glc analysis of the thermolysate from compound (69) i n -dicated that two compounds, in a r a t i o 1:^1.9, had been formed in quantitative y i e l d . I d e n t i f i c a t i o n of the two compounds as (179) and (180), respectively, was possible on the basis of the 400 MHz ^H nmr spectrum of the mixture. Nevertheless, the i n -divi d u a l compounds were obtained by preparative glc and f u l l y characterized. The i r spectrum of the minor, more v o l a t i l e product (179) showed absorptions (e.g. 3020 and 1635 cm - 1) consistent with i t s structure and i t s 400 MHz ^H nmr spectrum displayed a l l the expected signals. Thus, the doublet of doublets (J A B=4Hz, J n R=6.5Hz) due to H R was observed at 64.53 which i s a t y p i c a l - 154 -chemical s h i f t for the g - o l e f i n i c proton of an enol ether. The expected low f i e l d signal due to H^, the proton adjacent to the oxygen atom of the enol ether, occurred as a doublet of doub-l e t s (J=2.5, 6.5Hz) at 66.39. The (Z)-propenyl side chain of (179) was indicated by the v i n y l methyl doublet of doublets (J=2, 7Hz) at 61.65, the doublet of doublet of quartets (appear-ing as a t r i p l e t of quartets with J=2, 10Hz) at 65.26 due to H"w and the doublet of doublet of quartets (J=^1.5, 7, 10Hz) at . 65.45 due to H v. (89) 69 179 180 The structure of (180) was f u l l y elucidated by i t s 400 MHz nmr spectrum. Thus the four protons adjacent to the oxygen atom were c l e a r l y indicated. The two doublet of doub-l e t s at 63.40 (J=7.5, 11Hz) and 63.87 (J= 5.5, 11Hz) could be assigned to H^-, and H^, respectively on the assumption that the larger of the two coupling constants, J ^ B and J D B , i s l i k e l y to be the pseudo-axial - a x i a l one, The fact that in (180) the proton H^ i s expected to be a x i a l c l e a r l y sup-139 ports t h i s assignment . A two proton multiplet at lower f i e l d , 64.10-4.16, was obviously due to Ev and H p since these - 155 -two protons are at the same time a l l y l i c and adjacent to oxy-gen. Four o l e f i n i c protons were also indicated. Of these, the two due to and H^ gave r i s e to a doublet of doublet of quar-te t s (appearing as a t r i p l e t of quartets with J=2, 10Hz) at 65.24 and a doublet of doublet of quartets (J=,vl.5, 7, 10Hz) at 65.57. The v i n y l methyl signal associated with these pro-tons in the (Z)-propenyl side chain appeared as the expected doublet of doublets (J=2, 7Hz) at 61.68. The remaining two o l e f i n i c protons, H A and Hj, appeared as two doublet of multi-p l e t s at 65.68 and 65.76 (J=10Hz for each), respectively. It was quite evident both from c a p i l l a r y glc analysis of the benzene solution of the thermolysate of compound (70) and la t e r analysis of the nmr spectrum of the i s o l a t e d material (89% y i e l d ) that a single compound (181) had been formed on rearrangement of the vinylcyclopropane (7JD). The 400 MHz ^R nmr spectrum of (181) l a i d bare i t s structure and t o t a l l y ex-cluded compound (182). Thus, the doublet (J=6.5Hz) at 64.28 c l e a r l y indicated the presence of the proton H^ and the two proton o l e f i n i c multiplet at 65.53-5.64 c l e a r l y due to H^ and Hj was consistent only with the structure (181). The proton H H, a l l y l i c in (181), was s h i f t e d downfield r e l a t i v e to i t s location (62.36-2.50) in (70) and appeared as a multiplet at 63.38-3.46. The doubly a l l y l i c proton Hg, which appeared as a broad doublet CJ==J-0Hz) at 63.59, was shown by decoupling experiments to couple largely with the o l e f i n i c proton HT„ and only to a small extent (small, unresolved coupling) with H p. - 156 -(90) 70 181 182 This observation i s in accord with the fact that i t s e l f appears only as a doublet ( J^JJ=6.5HZ) and emphasises that H^, i s on the same face of the b i c y c l i c system (181) as H J J but 117 i s on the face opposite to H R . Cle a r l y , the proton must have been on the same face of (70) as the vinylcyclopro-pane moiety and thus was in p o s i t i o n to undergo [l,5]-hydro-gen s h i f t . The t o t a l s i t e - s e l e c t i v i t y observed in the re-arrangement of (70) cannot, therefore, be ascribed to incorrect assignment of the r e l a t i v e stereochemistry of and the v i n y l -cyclopropane unit in t h i s compound. Each of the methyl substituted vinylcyclopropanes (71a) - 157 -and (72a) underwent [ l,5]-sigmatropic hydrogen rearrangement to give a mixture of two products in a r a t i o of approximately 1:1.3-1.4. The minor product in each case was the one a r i s i n g from migration of the hydrogen atom adjacent to the methyl substituent. Compounds (183) and (184), obtained in 96% y i e l d from the rearrangement of (71a), were separated by s i l v e r n i t r a t e -132 s i l i c a gel preparative t i c (30% ether-petroleum ether) and were f u l l y characterized. The more mobile, minor component (183) exhibited uv (A 218nm, e=16,400) and i r (1705, 1675, 1615 cm-1") c h a r a c t e r i s t i c s expected for an aB-unsaturated ke-tone. Its nmr spectrum showed the required v i n y l methyl signal as a broad singlet at 61.64. Two o l e f i n i c signals, each a broad s i n g l e t , were also in evidence. The signal at lower f i e l d (55.96) was obviously due to H z while that at 65.22 was assigned to H R. The uv (A 219nm, £=19,000) and i r (1705, 1675, 1615 cm - 1) spectral features of compound (184) were very s i m i l a r to those observed for i t s isomer (183) and were in accord with 0 71a 183 184 1 1 , 3 - 158 -expectations. The H nmr spectrum of (184) showed the expect-ed secondary methyl doublet (J=6Hz) at 5 0.90 and indicated the presence of three o l e f i n i c protons. The broad s i n g l e t at 55.97 was assigned to the cyclopentenyl proton (H z) while the two proton multiplet at 65.33-5.82 was assigned to Hj and H^. Rearrangement of the rather v o l a t i l e (72a) was performed in deuterobenzene so that the ^~R nmr spectrum of the rearrange-ment products could be obtained without the need to i s o l a t e the mixture. C a p i l l a r y glc and ^R nmr (400 MHz) analysis of the thermolysis mixture indicated that two products (185) and (186) had been formed in a r a t i o of 1:^1.4. A small amount (4.5%) of a t h i r d component (187) (formed by rearrangement of (72b) o r i g i n a l l y present to the extent of 4-5% i n (72a)) was also present. The major component (186) was obtained pure (100% by c a p i l l a r y glc) by preparative glc of the rearrangement mixture. Its 400 MHz ^R nmr spectrum was f u l l y consistent with the assigned structure. Thus, the expected secondary methyl doub-l e t (J=7Hz) was found at 60.92. The (Z)-propenyl side chain gave r i s e to the usual signals - a v i n y l methyl doublet of doublets (J=2, 7Hz) at 61.64, a doublet of doublet of multiplets (appearing as a t r i p l e t of multiplets with J=10.5Hz) at 65.14 due to H^ and a doublet of quartets (J=7, 10Hz) at 65.51 due to H x« The o l e f i n i c protons H^ and Hj., l i k e s i m i l a r protons in the compounds already described, appeared as a doublet of multiplets (J=10Hz) at 65.36 due to H A and a lower f i e l d - 159 -multiplet at 6 5.65-5.73 due to H H, (92) 72a 185 186 Although the minor rearrangement product (185) could not be obtained 100% pure, the l a t e r eluted preparative glc f r a c -t i o n was s u f f i c i e n t l y enriched (%90%) in t h i s component to enable assignment of the signals in i t s 400 MHz "*"H nmr spec-trum. Thus, the H 0 proton gave r i s e to a broad singlet at 65.19. The other o l e f i n i c protons, H^ and H x > gave r i s e to the usual doublet of doublet of quartets (appearing as a t r i p -l e t of quartets with J=2.5, 10.5Hz) at 65.25 and doublet of doublet of quartets (J= 1, 7, 10Hz) at 65.41. The usual doub-l e t of doublets for the v i n y l methyl of the (Z)-propenyl side chain was obscured due to overlapping in the region 61.62-1.68 with the signal (a broad s i n g l e t ) due to the other v i n y l methyl group in the molecule. 187 - 160 -The phenyl substituted vinylcyclopropane (76) on re-arrangement gave, in 87% y i e l d , a mixture of two products (188) and (189) in a r a t i o (glc) of 18:82. Pure samples of each of the two products were obtained by preparative glc of the mix-ture . The f i r s t eluted, minor product (188) gave i r absorptions (e.g. 3025, 1590, 755 and 750 cm - 1) t y p i c a l of a monosubstitut-86 1 ed benzene . Its 400 MHz H nmr spectrum indicated the H^ proton as a doublet of t r i p l e t s (J=^4.5, 10.5Hz) at 62.47. Clearly H^ i s a x i a l i n the preferred conformation of (188) and J D = J ==10. 5Hz. The four o l e f i n i c protons occurred in the expected pattern. Thus H^ and H x gave r i s e , respectively, to a doublet of doublets of quartets (appearing as a t r i p l e t of quartets with J=2.5, 10Hz) at 65.13 and a doublet of doublet of quartets (J=%1.5, 7, 10Hz) at 65.32. The pair H^ and Hj were the usual doublet of multiplets (J=10Hz) at 65.51 and a multiplet at 65.74-5.82. The v i n y l methyl doublet of doublets (J=2, 7Hz) was s h i f t e d well u p f i e l d of i t s usual position (%61.70) and appeared at 61.24. The most l i k e l y explanation for t h i s observation i s that the (Z)-propenyl side chain in (188) i s aligned such that the methyl group i s held in the shielding zone of the benzene r i n g . The f i v e protons of the l a t t e r group occurred as two multiplets, integrating for three and two protons respectively, between 67.12 and 7.33. The major rearrangement product (189) also gave a c h a r a c t e r i s t i c i r spectrum with absorptions at 3060, 1590, 755 - 161 -(93) and 700 cm - 1. Its 400 MHz "'"H nmr spectrum displayed a l l the features expected of (189). Thus only three o l e f i n i c protons were indicated. The proton H D was expected to be the most deshielded and occurred as a broad si n g l e t at 65.87. The pro-tons H^ and H^ appeared, respectively, as a broad t r i p l e t (J A W=J w x=10.5Hz) at 65.32 and a doublet of quartets (J=7, 10.5 Hz) at 65.46. The v i n y l methyl doublet of doublets (J=2,7Hz) was in i t s usual position at 61.70. As in the isomer (188), the aromatic protons of (189) separated into a three proton and a two proton multiplet in the region 67.17-7.41. The sample of the t r i m e t h y l s i l y l vinylcyclopropane (77a) that was prepared was known to contain a small amount (^5%) of isomeric impurities (77b-d). Rearrangement of t h i s material resulted i n a product that was 97% pure by c a p i l l a r y glc ana-l y s i s . It was noted that one of the isomers, presumably (77c), accounting for ^2% of the mixture remained completely un-changed. These observations led to the conclusion that a single product had been formed i n the rearrangement of (77a). - 162 -77b 77c 77d The 400 MHz "^H nmr spectrum of the d i s t i l l e d rearrange-ment product confirmed t h i s interpretation and revealed the id e n t i t y of t h i s product to be (190). Thus, the one proton multiplet (due to the proton adjacent to the t r i m e t h y l s i l y l group) which appeared at 60.51-0.59 in the ^ H nmr spectrum of (77a) was notably absent. In the o l e f i n i c region only three protons were indicated. The doublet of doublet of quartets 77a 190 (appearing as a t r i p l e t of quartets with J=2, 10Hz) at 65.26 and the doublet of doublet of quartets (J=1.5, 7, 10Hz) at 65.44 were c l e a r l y due to H^ and H x, respectively , The remaining one proton multiplet at 65.70-5.75 was assigned to - 163 -H„. The v i n y l methyl group of the (Z)-propenyl side chain occurred as the usual doublet of doublets (J=2,7Hz) a t 61.66. F i n a l l y , the t r i m e t h y l s i l y l protons appeared as the expected s i n g l e t at 60.04. The i r (1600, 1250 and 855 cm - 1) and mass spectral feat-ures of (190) were also f u l l y in accord with the assigned structure. XI. Discussion of the Rearrangement Results The reader w i l l r e c a l l that at the beginning of the pre-sent work we posed some quite s p e c i f i c questions concerning the s i t e - s e l e c t i v i t y of the homo-[l,5]-hydrogen migration re-action in some unsymmetrically substituted b i c y c l i c and t r i -c y c l i c vinylcyclopropane systems e.g. (58e-f), (65a) and (76). The r e s u l t s we have obtained in t h i s study (see Table IV) do in large measure answer these questions and further allow us to advance a tentative proposal as to the nature of the trans-i t i o n state i n these homo-[1,5]-hydrogen s h i f t processes. In terms of the f i r s t two questions (p.31 ), referred to above, i t i s clear, from the most casual perusal of Table IV, that there i s a quite general oxy-substituent e f f e c t on the s i t e - s e l e c t i v i t y of the homo-[l,5]-hydrogen migration. The tendency of an oxy-substituent i s to disfavour the product that would resu l t from the [1,5]-migration of an adjacent - 164 -Table.IV : Results of Homo-[l,5]-Rearrangement of the endo-Vinylcyclopropanes. Vinylcyclopropane Possible Products (Ratio obtained) b. Isolated Y i e l d (%) 65a OMe 167 (0:100) 90 OR 66a 67a R=-Me R=-MEM OR 169 171 (0 (0 100) 100)' 168 170 90' 93 177 94 69 179 C34:66) 180 100 - 165 -Table,IV (Cont'd.) Vinylcyclopropane Possible Products (Ratio obtained) Isolated Y i e l d (%) 70 182 (0:100) 181 89 C O >^ ( i 9 ] > y 0 1 / 71a 183 (43:57) 184 96* '* r' y X X 1 . A X ^ 72a X=-CH3 185 (40:60) 186 _ f 75 X=-0Me 173 ( 0:100)c" 172 80 76 X=-Ph 189 ( 82:18) 188 87 77a X=-SiMe„ 190 (100: 0) 191 90 74 X=-0H 174 (12:88) 98 - 166 -~ Rearrangements were performed in benzene in sealed, base washed, s i l y l a t e d glass tubes by heating at 230-240°C for 6-8h. Ratios were determined by glc and/or nmr. This r a t i o was determined by "*"H nmr so i s actually in the range 0-2:98-100. Rearrangement as in (a) both neat and in benzene solution. Y i e l d quoted for neat rearrangement. The s o l i d vinylcyclopropane was rearranged by heating under a nitrogen atmosphere for 15 min at 215°C. Rearrangement performed in deuterobenzene. The y i e l d was not determined but the " h f nmr spectrum of the thermo-lysed solution indicated clean rearrangement into two products. - 167 -hydrogen atom (H^) with respect to the product that would resul t from migration of the more remote hydrogen atom (Hj). The presence of a conjugating keto-substituent on the v i n y l moiety of the vinylcyclopropane does not noticeably influence t h i s unfavourable oxy-substituent e f f e c t . It appears, though, that the e f f e c t could be attenuated somewhat by s t r u c t u r a l v a r i a t i o n s , as evidenced by the d i f f e r e n t product r a t i o s ob-tained on rearrangement of (66a) and (68b), and by changes in the nature of the oxy-substituent (cf. product r a t i o s from rearrangement of (74) and (75) in Table IV). With respect to the influence of substituents other than oxygen on the s i t e - s e l e c t i v i t y of the rearrangement, our study was by no means comprehensive being l i m i t e d to three such sub-st i t u e n t s . It i s i n t e r e s t i n g , though, that two groups as d i f -ferent in t h e i r e l e c t r i c a l properties as phenyl and trimethyl-s i l y l both favour the rearrangement product r e s u l t i n g from migration of the hydrogen atom (H^) adjacent to the substitu-ent while the methyl group, which i s not unlike the trimethyl-s i l y l group ( i n a Hammett sense), s l i g h t l y disfavours forma-ti o n of t h i s product. At t h i s juncture, before entering into detailed discus-sion of the r e s u l t s in Table IV, i t i s appropriate to point out that the product r a t i o s obtained r e f l e c t k i n e t i c factors at work and are not determined by product s t a b i l i t y (thermo-dynamic c o n t r o l ) . If one assumes that of the two products - 168 -possible on rearrangement of a p a r t i c u l a r vinylcyclopro-pane the more stable i s the one with the more stable double bonds then i t follows that in a l l cases (except perhaps for rearrangement of (70)) the more stable product would be the one produced by migration of the hydrogen atom (H^ ,) adjacent to the substituent. Some idea of the differences in s t a b i l i t y in the cases we have studied i s obtained from the double bond s t a b i l i s a t i o n energies of 5.2, 4.9 and 3.2 kcal mole - 1 obtain-140 ed by Hine for the methoxy, phenyl and methyl substituents, respectively. Clearly, our r e s u l t s , in p a r t i c u l a r those obtained from rearrangement of compound (74), indicate that factors related to product s t a b i l i t y do not determine the product r a t i o s of 181 182 While the enol ether double bond of (182) should be st a -b i l i s e d , to the extent of about 5 kcal mole - 1 ^ 4 0 , with respect to the endocyclic double bond of (181) i t i s a l -so severely d e s t a b i l i s e d owing to i t s location at the bridgehead of a bic y c l o [3.3.0] system. Thus, i t i s by no means certain which of the two rearrangement products of (70) i s the thermodynamically more stable. - 169 -the homo-[1,5]-hydrogen migration reaction in the oxy-substi-tuted (except perhaps for (70_)) and methyl substituted cases. This conclusion probably applies to the rearrangement of (76) and (77a) (and possibly (70)) as well, even though the more stable of the two products possible predominates in these cases. The fact that the product r a t i o s in the homo-[l,5]-hydro-gen s h i f t s studied are not determined by product s t a b i l i t y leads to two important conclusions about the reaction mechan-ism. F i r s t l y , the rearrangement reaction must be e s s e n t i a l l y i r r e v e r s i b l e under the conditions u t i l i z e d . While i t could be argued that the product r a t i o s in the rearrangement of (68b), (69), (71a), (72a) and (74) might have arisen from incomplete e q u i l i b r a t i o n of products under the rearrangement co n d i t i o n s . i t seems highly u n l i k e l y that r e v e r s i b l e rearrangement of (65a), (66a), (67a) and (75) would have resulted in v i r t u a l l y none of the more stable product. A far more plausible explanation i s simply that the rearrangements we have studied are i r r e v e r s i b l e . The r e v e r s i b i l i t y of the homo-[l,5]-hydrogen s h i f t reaction 19 20 27 has been demonstrated in some systems ' ' so that, while the reaction may in fact be i r r e v e r s i b l e in the majority of cases, the issue of r e v e r s i b i l i t y i s always in doubt in i n d i -vidual instances. The second conclusion to be drawn from the fact that the rearrangement product r a t i o s are not determined by product - 170 -s t a b i l i t y pertains to the p o s i t i o n of the t r a n s i t i o n state a-long the reaction coordinate. A l a t e t r a n s i t i o n state i s apparently ruled out by the r e s u l t s since such a t r a n s i t i o n 141 state, according to Hammond's postulate , would have pro-duct-like character. This would have meant that in the re-arrangement of each vinylcyclopropane the t r a n s i t i o n state leading to the more stable of the two possible rearrangement products would have been of lower energy than that leading to the less stable product. Consequently, the more stable pro-duct would have predominated in every case. The two conclusions stated above are s t r i c t l y v a l i d only for the oxy-substituted (except perhaps for (70)) and the me-t h y l substituted vinylcyclopropanes for which the less stable rearrangement product predominated. However, these conclusions should apply as well to the rearrangement of compounds (76), and (77a) (and probably (70)) for which the more stable pro-duct was actually the predominant one, provided there i s no "mechanistic d i s c o n t i n u i t y " i n the series of compounds studied. This idea of v a r i a t i o n in mechanism (more s p e c i f i c a l l y i n the nature of the t r a n s i t i o n state) of a concerted reaction with substituents i s not new, of course, having been very thought-142 143 f u l l y and convincingly argued for by Gajewski and others in r e l a t i o n to the Cope rearrangement. In the present case, though, there seems to be no overriding reason why, for ex-ample, (69) and (70) should rearrange v i a d i f f e r e n t mechanis-- 171 -t i c pathways or indeed why (76) and (77a) should show d i f -ferent mechanistic behaviour from (75). If one accepts the conclusion that only k i n e t i c factors determine the product r a t i o in the vinylcyclopropane rearrange-ments we have studied, then the r a t i o of the product a r i s i n g from migration of H c to that a r i s i n g from migration of Hj, can be taken in each case, as a measure of the r e l a t i v e rate ( k c / kj) of migration of these two hydrogen atoms. It follows that oxygen substituents must be exerting a d e s t a b i l i s i n g e f f e c t on the t r a n s i t i o n state for migration of R"c r e l a t i v e to the t r a n s i t i o n state for Hj migration with the re s u l t that the rate ( k c) of migration of H"c becomes appreciably smaller than the rate (kj) of migration of Hj. S i m i l a r l y , the methyl sub-stituent must be exerting a small retarding e f f e c t on k^ r e l a -t i v e to k j i n (71a) and (72a) and the phenyl and trimethyl-s i l y l groups both accelerate the rate of migration of r e l a -t i v e to Hj i n (76) and (77a), respectively. These substituent e f f e c t s on the rate of [l,5]-migration of the adjacent hydrogen atom (H c) are quite i n t r i g u i n g especi-a l l y when compared with known substituent e f f e c t s in dienyl [1,5]-hydrogen s h i f t s . Thus both methyl and methoxy substitu-ents accelerate the rate of [1,5]-hydrogen migration in the 44 7-substituted cycloheptatriene system . In the 5-substituted cyclopentadiene system a methyl group has an appreciable, 1 2 accelerating effect on the rate of the [l,5]-hydrogen migra-- 172 -tio n as well but hydrogen migrates at about the same rate in 31 5-trimethylsilylcyclopentadiene (26) as i n 1, 2, 3, 4, 5-144 pentadeuterocyclopentadiene . One can only speculate in the l i g h t of the quite d i f f e r e n t substituent e f f e c t s observed in the dienyl and homodienyl [1,5]-hydrogen s h i f t s that there e x i s t s some difference in mechanistic d e t a i l between the two reactions. A t r a n s i t i o n state in which there i s p o s i t i v e charge developed at the carbon bearing the migrating hydrogen has 45 been suggested to explain the e f f e c t s of some substituents on the rate of [1,5]-hydrogen migration in 7-substituted cyclo-heptatrienes and 5-substituted cyclopentadienes. Our r e s u l t s , e s p e c i a l l y considering the series (72a), (74)-(76) and (77a), rule out such a t r a n s i t i o n state. An oxygen substituent (as in (74) or (75)) would have s t a b i l i s e d the t r a n s i t i o n state - 173 -for migration of r e l a t i v e to that for migration of Hj and one would have expected the enol ether product (173) and the ketone (174) to predominate in the rearrangement of (7_5) and (74), respectively. In the rearrangement of (72a), since a methyl group s t a b i l i s e s an adjacent p o s i t i v e charge, the pro-duct (185) should have predominated over (186). There i s 174 175 strong evidence that the t r i m e t h y l s i l y l group actually de-145 s t a b i l i s e s an adjacent p o s i t i v e charge . Thus, i f p o s i t i v e charge developed at the carbon atom bearing the migrating hy-drogen (H^ ,) during the rearrangement of (77a), c l e a r l y com-pound (190) would not be the only product observed. As for the phenyl substituted case, i n the t r a n s i t i o n state the - 174 -benzene r i n g i s not expected to be suitably oriented (co-' planar with the saturated six-membered ring) to s t a b i l i s e a 190 191 developing p o s i t i v e charge adjacent to i t and would l i k e l y mildly d e s t a b i l i s e such a charge as a res u l t of a s l i g h t l y dominating electron withdrawing e f f e c t . It i s clear, then, that in the case of the rearrangement of (7^6), as for rearrange-ment of (72a), (75) and (77a), the actual r e s u l t s are not con-sis t e n t with the development of p o s i t i v e charge at the carbon atom bearing the migrating hydrogen. Using arguments very s i m i l a r to those employed above, one arrives at the conclusion that a t r a n s i t i o n -state in which r a d i c a l character i s developed at that same carbon centre i s - 1 7 5 -not supported by our r e s u l t s - with the possible exception 146 of the t r i m e t h y l s i l y l substituted case The r e s u l t s we have accumulated so far do, however, ap-pear to be consistent with the development of a p a r t i a l nega-t i v e charge at the s i t e of o r i g i n of the migrating hydrogen atom. Thus even without the c a p a b i l i t y to s t a b i l i s e by con-jugative d e l o c a l i s a t i o n an adjacent negative charge the phenyl substituent can provide some s t a b i l i s a t i o n by inductive elec-tron withdrawal. Of the two pathways (Scheme XXXII) leading to products from (76) that v i a t r a n s i t i o n state (192) would be of lower energy and lead to a predominance of product (189) Scheme XXXII - 176 -over (188) as observed. The a b i l i t y of a t r i m e t h y l s i l y l group to s t a b i l i s e adjacent negative charge i s well document-145 ed . Clearly, the exclusive formation of product (190) i n the rearrangement of (77a) can be neatly r a t i o n a l i z e d in terms of a t r a n s i t i o n state in which there i s negative charge deve-lopment at the carbon atom bearing the migrating hydrogen. The r e s u l t s obtained in the case of the methyl substituted vinylcyclopropanes (71a) and (72a) are also q u a l i t a t i v e l y understandable. A methyl substituent i s expected to have only a small d e s t a b i l i s i n g e f f e c t on a developing p a r t i a l nega-t i v e charge adjacent to i t and the t r a n s i t i o n states leading to migration of Hj and would not be expected to d i f f e r s i g -n i f i c a n t l y in energy from each other. A r a t i o of products i n each case, (71a) and (72a), of close to 1:1 i s therefore quite reasonable. What of the r e s u l t s i n the oxy-substituted cases? One would naively expect on the basis of the electron withdrawing a b i l i t y of oxygen (a^ for methoxy i s 0.27 while for phenyl 147 i s 0.10) that oxygen substituents should favour migration of r e l a t i v e to Hj in the rearrangement of the oxygen s u b s t i -tuted vinylcyclopropanes. These rearrangements should show even greater preference for H^ , migration than that observed for the phenyl substituted compound (76). How then does one ex-p l a i n the fact that the actual r e s u l t s are absolutely contrary to predictions? - 177 -It appears to have been well established that there i s a repulsive d e s t a b i l i s i n g i n t e r a c t i o n between the lone pairs on an oxygen atom and a developing adjacent negative charge which retards the rate of reactions in which such a charge i s to be placed alpha to the oxygen atom in the product. This destabi-l i s i n g i n t e r a c t i o n makes i t s e l f f e l t i n equilibrium s i t u a t i o n s as well. Thus,the retarding e f f e c t of a-methoxy substituents on the rate of base catalysed deuterium exchange in a series 148 of methyl esters could be explained on the basis of repul-sive i n t e r a c t i o n between the oxygen lone pairs and the develop-ing carbanion in the t r a n s i t i o n state. 149 Very recently Hirsch reported that the oxaketone (194), under conditions of k i n e t i c deprotonation (followed by enolate trapping by t r i a l k y l s i l y l c h l o r i d e ) , yielded as the predominant product that r e s u l t i n g from deprotonation at the position re-mote from the oxygen atom. Under thermodynamic conditions despite the known s t a b i l i s i n g e f f e c t of oxygen on a double bond (^5 kcal m o l e - 1 " ) 1 4 0 , the isomer (195) was s t i l l predominant. 194 195 196 Ki n e t i c Thermodynamic 75-88% 89% 12-25% 11% - 178 -The lone pair - carbanion interaction must be sub s t a n t i a l l y d e s t a b i l i s i n g (at least in t h i s case) es p e c i a l l y when one con-siders that in the enolate anion giving r i s e to (196) only a small percentage of the negative charge i s actually located on the carbon atom alpha to the ring oxygen. 150 The r e s u l t s of a study by Garst on the nitrogen ana-logues of oxaketone (194) lend credence to the lone pair - car-banion repulsion theory. Thus under k i n e t i c conditions the product of y-deprotonation predominates with the N-ethyl com-pound (197) but in the urethane (200), in which the nitrogen lone pair i s presumably well delocalised, alpha deprotonation predominates. 1 3.5 - 179 -. , By way of an example which does not involve carbonyl 151 compounds Bordwell's study on the e f f e c t of substituents on the pK values of 9-substituted fluorenes should be quoted. a. It was found that the pK a values of 9-oxy-substituted f l u o r -enes (203) and (204) (MeO- and PhO- substituents) calculated on 147 the basis of the polar e f f e c t s (as measured by a^) of the substituents were consistently lower (1-2 pK units) than those ac t u a l l y observed. These r e s u l t s were explained in terms of the d e s t a b i l i s i n g e f f e c t s of the oxy-substituents on the car-banions as a r e s u l t of the lone pair interactions. It i s X 203 X=-OMe 205 X=-SMe 204 X=-OPh 206 X=-SPh in t e r e s t i n g to note that for the corresponding sulphur s u b s t i -tuents (205) and (206) the calculated pK values were higher (2-4 pK units) than those actually obtained, suggesting s t a b i -l i s a t i o n of the anions over and above that expected from the polar e f f e c t (of -SR) of greater than 6 kcal mole - 1. It would appear then that our r e s u l t s for the oxy-substi-tuted vinylcyclopropanes are not af t e r a l l inconsistent with the development of negative charge at the s i t e of o r i g i n of the migrating hydrogen. The s t a b i l i s a t i o n of the t r a n s i t i o n state - 180 -for migration of by the electron withdrawing e f f e c t of oxygen i s apparently more than overcome by the d e s t a b i l i s i n g interaction between the oxygen lone pairs and the p a r t i a l negative charge developed in the t r a n s i t i o n state. The r e s u l t i s that the product r e s u l t i n g from migration of the hydrogen atom Hj i s highly favoured. Within the series of oxygen substituted compounds re-arranged the v a r i a t i o n of the product r a t i o s i s quite f a s c i n a t -ing. Those vinylcyclopropanes having an oxy-substituent attach-ed externally to the ri n g system bearing the migrating hydrogen (H^) i . e . (65a)-(67a) and (75_) rearrange to give no detectable amounts of the product that would have resulted from migration of H^ (see Table IV). On the other hand, compound (68b), in which the H^ , hydrogen atom i s attached to the conformationally f l e x i b l e methoxymethyl side chain, and compound (69), in which the oxygen atom alpha to H^ i s actually incorporated into the b i c y c l i c r i n g system, both give substantial amounts of product derived from migration of H c. Curiously, these compounds, (68b) and (69), give remarkably s i m i l a r product r a t i o s (kj/k(-,%2:1) as well. P a r t i c u l a r l y i n t e r e s t i n g i s the fact that compound (70), which ( l i k e (69)) incorporates an oxygen atom into i t s c y c l i c framework in a position adjacent to a hydrogen atom (H^) that can undergo [1,5]-migration, gave (unlike (69)) no de-tectable amounts of the product (182) that would have resulted from migration of that hydrogen atom (H r). - 181 -_Considering the s t r u c t u r a l differences among the oxy-substituted vinylcyclopropanes i t appeared worthwhile to pur-sue the idea that the var i a t i o n s in t h e i r product r a t i o s might be due to conformational e f f e c t s . Examination of models of (69) and (70) make i t clear that the orientation of the lone pairs on oxygen with respect to the adjacent carbon-hydrogen (C-H^) bond that i s broken in the [l,5]-hydrogen migration re-action i s quite d i f f e r e n t in these two compounds. As might be expected, since the oxygen atom in (70) i s part of a f i v e -membered rin g , one of the oxygen lone p a i r s i s approximately eclipsed with the C-H^ , bond while the other i s directed well away (110°) from t h i s bond. The limi t e d f l e x i b i l i t y of the sk e l e t a l system of (70) ensures that t h i s orientation remains r e l a t i v e l y unvarying and a si m i l a r o r i e n t a t i o n of the lone pairs r e l a t i v e to the C-H^ bond would be expected in the trans-i t i o n state leading to product (182) from (70). Thus in the charged t r a n s i t i o n state (207) we envisage for the homo-[l,5]-hydrogen migration one oxygen lone p a i r w i l l be in close proximity to (eclipsed with) the p a r t i a l negative charge de-He 207 208 - 182 -veloped at and strong repulsive d e s t a b i l i s a t i o n of (207) with respect to (2_08) should r e s u l t . If one compares the orientation of the oxygen lone pairs in (69) with respect to the C-H^ , bond i t transpires that, in the conformation required to allow migration of H^, one oxygen lone pair i s oriented antiperiplanar to the C-H^ -, bond and the other i s gauche to i t . In the t r a n s i t i o n state for migration of H_, the oxygen lone p a i r s would be oriented with respect 209 to the negative charge at such that t h i s t r a n s i t i o n state (209) i s d e s t a b i l i s e d with respect to the t r a n s i t i o n state for migration of Hj but to a lesser extent than would be expected in the case of rearrangement of (70). On t h i s basis one can r a t i o n a l i z e in a q u a l i t a t i v e way why the r a t i o k^ik^ i s so much smaller for rearrangement of (69) than for rearrangement of (70). It i s a rather complicated task to apply arguments s i m i l -ar to those outlined above to r a t i o n a l i z e the d i f f e r e n t pro-duct r a t i o s obtained on rearrangement of (75) and (68b). In each case rotation about the C-OMe bond gives r i s e to an - 183 -i n f i n i t e number of orientations of the -OMe group with d i f -f e r i n g stereorelationships between the two oxygen lone pairs and the C-Hc bond. If one considers the more stable con-formations of the -OMe group in (68b) i t would appear, from examination of models, that the conformation in which the oxygen lone p a i r s are antiperiplanar and gauche to the C-H^ bond (oriented as i t must be in order to migrate) i s most stable. On the assumption that t h i s most stable -OMe rotamer i s in fact preferred in the t r a n s i t i o n state (210) for mi-gration of H^-,, one could explain why the r e s u l t s of the re-arrangement of (68b) are so s i m i l a r to those obtained for (69). 210 In the case of compound (7_5) the r e s u l t s could be explained i f i t i s assumed that in the t r a n s i t i o n state (211) leading to migration of the -OMe group also adopts i t s minimum energy conformation i n which both oxygen lone pairs are gauche to the breaking C-Hc bond. This orientation would be expected to lead to d e s t a b i l i s a t i o n of the t r a n s i t i o n state for migration of r e l a t i v e to that for migration of Hj of at least the same magnitude as in the case of (70) and thus - 184 -Hex a s i m i l a r r a t i o of products was obtained. Much of the preceeding discussion has been based on a proposal that in the t r a n s i t i o n state for the homo-[l,5]-hy-drogen s h i f t in a series of b i c y c l i c and t r i c y c l i c v i n y l c y c l o -propanes a p a r t i a l negative charge develops at the carbon centre bearing the migrating hydrogen. The reader, understand-ably, may be quite s c e p t i c a l of such an idea in r e l a t i o n to a reaction which i s presumably concerted. Thus i t would be appropriate at t h i s point to attempt to resolve the apparent c o n f l i c t inherent i n the proposal of a charged t r a n s i t i o n state f o r a concerted reaction. It i s understood that in a concert-ed reaction the bond forming and bond breaking processes are i n e x t r i c a b l y coupled. This i s why, of course, a c t i v a t i o n energies of concerted reactions are always lower than the to-t a l energy required for bond breaking alone. However, in the t r a n s i t i o n state of a concerted reaction the extent of com-pleteness (or incompleteness) of the bond forming and bond breaking at various bonding s i t e s of the t r a n s i t i o n state - 185 -could vary s i g n i f i c a n t l y . Put another way, bond breaking and bond making though not consecutive processes may not be completely synchronous processes either. In t h i s s i t u a t i o n p a r t i a l charges may develop at p a r t i c u l a r s i t e s in the trans-i t i o n state. S p e c i f i c a l l y , a p a r t i a l negative charge would develop at the carbon centre bearing the migrating hydrogen, in the t r a n s i t i o n state (212) of the homo-[1,5]-hydrogen mi-grations we have been discussing, i f bond forming between the hydrogen atom (migrating as a "proton") and the terminal car-212 bon of the v i n y l group i s well in advance of double bond form-ation between and Cg. Such a picture of the t r a n s i t i o n state, i n c i d e n t a l l y , explains quite neatly why s t a b i l i t y of the double bonds formed in the reaction appears to have so l i t t l e influence on the product r a t i o . While our proposal about the nature of the t r a n s i t i o n state, in the homo-[l,5]-hydrogen migrations we studied, ap-pears to be f u l l y consistent with the r e s u l t s we obtained i t should be noted that the l i s t of substituents in the series of substrates was li m i t e d to -CH^, -Ph, -SiMe Q and -OR. - 186 -Further studies with c a r e f u l l y chosen substituents at the s i t e of o r i g i n of the migrating hydrogen atom would seem to be c a l l e d for in order to f u l l y test our proposal. Some very clear cut predictions can be made about the way various substituents should a f f e c t the rate of homo-[1,5]-migration of an adjacent hydrogen atom (H^) and these can be the basis for the design of future experimental work. An alkoxide substituent should be far more d e s t a b i l i s i n g to an adjacent negative charge than a methoxy (or hydroxy) substituent. Therefore, the rate of homo-[1,5]-migration of an adjacent hydrogen atom should be appreciably slower with an alkoxide substituent than with a methoxy (or hydroxy) substituent. This prediction i s of some interest since alk-50 oxide substitution in some well known sigmatropic reactions induces dramatic rate accelerations. Of p a r t i c u l a r s i g n i f i -5 6 cance in t h i s context are rate accelerations of 10 -10 ob-49 served by Paquette for the dienyl [1,5]-hydrogen s h i f t s 50a R=-H 50b R=-0~K+ in (50b) compared to (50a). - 187 -In contrast to an alkoxide substituent, the t r i f l u o r o -methoxy (-OCFg) group, which possesses lower electron densi-ty on the oxygen atom than the methoxy and hydroxy s u b s t i t u -ents, would be expected to accelerate the rate of homo-[l,5]-migration of an adjacent hydrogen atom compared with those two substituents. Thus a study of the rearrangement of a ser i e s of vinylcyclopropanes having, appropriately positioned, the substituents -OCFg, -OCHg and 0~M+ would constitute a c r i t i c a l test of the proposal that lone pairs on oxy-substitu-ents d e s t a b i l i s e the t r a n s i t i o n state leading to homo-[im-migration of an adjacent hydrogen atom (H c). In order to properly assess the r e s u l t s of the rearrange-ments suggested by the preceeding discussion i t i s clear that a f u l l y quantitative k i n e t i c study w i l l be required. It w i l l be desirable as well to modify the vinylcyclopropane system (exemplified by (75)) previously studied so that only the hydrogen atom CHC) adjacent to the substituent can migrate. By t h i s means, any complications which may a r i s e in following the rate of migration of due to the presence of the product of more rapid migration of another hydrogen atom w i l l be avoided. Compounds (213),(214) and (215) would seem to be a suitable s e r i e s for study. In view of the demonstrated a b i l i t y of sulphur (as op-151 posed to oxygen) to s t a b i l i s e adjacent carbanions the e f f e c t of the thiomethyl group (-SMe) on the homo-[1,5]-hy-drogen rearrangement i s predicted to be diametrically opposite - 188 -217 to that of a methoxy group. The observed rate of migration of in compound (216) should be su b s t a n t i a l l y greater than that observed in (213). It follows as well that a dramatic difference in s i t e - s e l e c t i v i t y should be observed for re-arrangement of (75) and (217). Perhaps the best means of probing the nature and the ex-tent of charge development at Cg, in the t r a n s i t i o n state for migration of H^ ,, would be to undertake a k i n e t i c study of homo-[l,5]-hydrogen rearrangement of compounds (218) to (223). 152 A Hammett plot of the rate data (log k c vs a) should y i e l d a l i n e of p o s i t i v e slope ( p ) i f negative charge i s in fact b u i l t - 189 -up at- C 2 in the t r a n s i t i o n state for H c migration. The magnitude of p should also be quite i n s t r u c t i v e as to the extent of charge b u i l d up at C 9. H 218 X=-H 221 X=-4-Br 219 X=-4-CF3 222 X=-4-0Me 220 X=-4-N0o 223 X=-4-Me The r e s u l t s obtained from the rearrangement studies suggested above, whether they confirm or refute our proposal about the nature of the t r a n s i t i o n i n the homo-[l,5]-hydrogen migration reaction, would at the very least contribute to a better understanding of t h i s i n t e r e s t i n g reaction. - 190 -EXPERIMENTAL General Information Melting points, determined using a Fisher-Johns melting point apparatus, and b o i l i n g points are uncorrected. Air-bath temperatures are quoted for short path (Kugelrohr) d i s t i l l a -t i o ns. U l t r a v i o l e t (uv) spectra were run on a Cary 15 spectro-photometer i n ethanol or hexanes (spectrophotometric grade). Infrared ( i r ) spectra were recorded, using Perkin-Elmer model 710 or 710B spectrophotometers, as l i q u i d films, in chloroform solution or as nujol mulls, and were cal i b r a t e d with the 1601 cm " band of the polystyrene f i l m . Proton and carbon-13 nu-1 13 clear magnetic resonance ( H nmr and C nmr) spectra were done in deuterochloroform solution (except where stated other-wise) with tetramethylsilane (TMS) as int e r n a l standard. In those cases where the t r i m e t h y l s i l y l group was present no TMS was added. The nmr spectra were recorded using Varian Asso-ciat e s models T-60, HA-100 or XL-100 and Bruker models WP-80 and WH-400 spectrometers. The 270 MHz spectra were recorded on a home made unit consisting of an Oxford Instruments 63.4 KG superconducting magnet and a Nicolet 16K computer attached to 13 a Bruker TT-23 console. C nmr spectra were recorded on the Bruker model WH-400 instrument. Signal positions are in parts per m i l l i o n (6) r e l a t i v e to the TMS signal except for those compounds containing the t r i m e t h y l s i l y l group. In those cases 15 signal positions are r e l a t i v e to the chloroform signal (67.27). - 191 -The m u l t i p l i c i t y , number of protons, coupling constants (where possible) and assignments a l l follow the signal positions in parentheses. The usual abbreviations are used to indicate the m u l t i p l i c i t i e s of the ''"H nmr signals i . e. s=singlet; d=doublet; t = t r i p l e t ; q=quartet; qn=quintet; m=multiplet. Low resolution mass spectra were recorded with a Varian/MAT CH4B mass spectro-meter. Glc- mass spectroscopic analyses were performed using a 6ft x 0.125in column packed with 5% OV-17 on chromosorb W(HP) (80-100 mesh) in a system comprised of a Pye Unicam Series 104 gas chromatograph interfaced with a VG Micromass 12 mass spec-trometer. High resolution mass spectra were recorded with Kratos/AEI MS50 or MS902 mass spectrometers. Microanalyses were performed by Mr. P. Borda, Microanalytical Laboratory, University of B r i t i s h Columbia. Packed column gas l i q u i d chromatographic (glc) analyses were performed using a Hewlett-Packard model HP5832A gas chro-matograph and the following columns: (A) 6ft x 0.125in, 5% OV-17 on chromosorb W (HP) (80-100 mesh); (B) 6ft x 0.125in, 5% OV-210 on chromosorb W (HP) (80-100 mesh). C a p i l l a r y glc analyses were performed using a Hewlett Packard model 5880 gas chromatograph equipped with a 25m SE-54 c a p i l l a r y column. Preparative glc was performed using a Varian Aerograph model 90-P gas chromatograph on the following column: (C) 10ft x 0.25in, 5% OV-17 on chromosorb W (60-80 mesh). Thin layer chromatography ( t i c ) was c a r r i e d out using - 192 -20 x 5cm glass plates coated with 0.3mm of s i l i c a gel GF 2g 4 (E. Merck, s i l i c a gel 60) or on commercial s i l i c a gel plates (Eastman Chromagram Sheet Type 13181 and E. Merck t i c sheets No. 5554). Preparative t i c was car r i e d out on 20 x 20cm glass plates coated with a 0.9mm layer of s i l i c a gel (E.Merck, s i l i c a gel 60 G-F^^). Normal (gravity) column chromatography 115 and f l a s h chromatography were performed using s i l i c a gel (E. Merck) of 70-230 and 230-400 mesh, respectively. F l o r i s i l (J.T. Baker Chemical Co., 100-200 mesh) was routinely used for f i l t r a t i o n purposes during workup of reactions. Preparative high pressure l i q u i d chromatography (HPLC) was performed on a Waters Associates Prep LC/System 500 instrument using the Waters Associates Prep Pak-500 s i l i c a gel columns. A l l reactions involving oxygen and/or moisture se n s i t i v e reagents were carr i e d out under an inert atmosphere (nitrogen or argon) using glassware flame-dried under nitrogen (or argon) flow. Solvents and Reagents Solutions of methyllithium, n-butyllithium and t e r t - b u t y l -lithium were obtained from Al d r i c h Chemical Company, Inc. and 154 were standardised according to the Gilman procedure Commercially obtained cuprous iodide was p u r i f i e d 1 1 0 by diss o l v i n g i t in a hot solution of potassium iodide in water, decolourising the r e s u l t i n g solution with charcoal followed by - 193 -f i l t r a t i o n and d i l u t i o n of the f i l t r a t e with water to p r e c i p i -tate the pure cuprous iodide. This material was then c o l l e c t -ed by f i l t r a t i o n and dried under high vacuum. Commercial cuprous bromide was p u r i f i e d by formation of, 85 i t s dimethyl sulphide complex. The crude cuprous bromide, in a sintered glass funnel, was washed with methanol to remove 155 much of the copper(II) impurities and then dissolved in a minimum quantity of cooled (0°C) dimethyl sulphide. The r e s u l t -ing solution was di l u t e d with %10 times i t s volume of petro-leum ether to p r e c i p i t a t e the white cuprous bromide - dimethyl sulphide complex. The s o l i d was co l l e c t e d by suction f i l t r a -t i o n and transferred, while s t i l l moist with solvent, into a round-bottomed f l a s k i n which i t was dried while protected from l i g h t by pumping at high vacuum overnight (/vl8h). 68 Phenylthiocopper was prepared by ref l u x i n g a well s t i r r e d mixture of cuprous oxide (42.9g, 0.3 mol) and t h i o -phenol (77g, 0.70 mol) in 1500 mL of ethanol for eight (8) days. Suction f i l t r a t i o n of the r e s u l t i n g yellow s l u r r y , thorough washing of the yellow s o l i d obtained with ethanol, followed by drying for several days under high vacuum gave phenylthiocopper in e s s e n t i a l l y quantitative y i e l d . Zinc-copper couple was freshly prepared before use 156 according to the following procedure . To a well s t i r r e d , hot solution of copper acetate monohydrate (0.12g, 0.6mmol) in 4 mL of g l a c i a l acetic acid was added 2.1g (32 mg-atom) of - 194 -z i n c d u s t . After about 0.5 min., the reaction mixture was allowed to s e t t l e for 1 min. after which most of the acetic acid was c a r e f u l l y decanted o f f . The grey couple was washed with 4 mL of g l a c i a l a c e tic acid and then with anhydrous ether (4x15 mL). The moist couple was pumped under high vacuum for %0.5h at room temperature and was ready for use. 157 Copper bronze was freshly prepared as follows . To a s t i r r e d solution of copper sulphate pentahydrate (5g, 20mmol) in 30 mL of water was added, i n small portions, l g (15.3mmol) of zinc dust. The r e s u l t i n g p r e c i p i t a t e was washed sequenti-a l l y with water, 95% ethanol and anhydrous ether. After i t had been pumped under high vacuum for ^0.5h, the copper bronze was ready for use. Ethyl diazoacetate was prepared according to the follow-69 ing procedure. To a solution of 27.9g (0.2 mol) of glycine ethyl ester hydrochloride and 0.2g (2.4 mmol) of sodium ace-tate in 55 mL of water was added 80 mL of petroleum ether and the mixture was cooled to -15°C. Then was added, i n small portions with s t i r r i n g , a cold (0°C) solution of 22g (0.32mol) of sodium n i t r i t e and 7 mL of 1M aqueous sulphuric acid in 37 mL of water. The mixture was s t i r r e d for 0.5h after com-pl e t i o n of the addition (%10 min) and was then extracted with petroleum ether. The combined organic extracts were washed with saturated aqueous sodium bicarbonate and brine and dried over anhydrous magnesium sulphate. Evaporation of the solvent and f l a s h d i s t i l l a t i o n (30°C, 0.4 Torr) of the residue into a - 195 -cooled (-78°C) receiver gave spectroscopically (nmr,ir) pure ethyl diazoacetate ( 18g, 79%). 124 T r i - n - b u t y l t i n hydride was prepared by lithium aluminium hydride (2g,52.7 mmol) reduction of t r i - n - b u t y l t i n chloride (32.5g, 99.8 mmol) in tetrahydrofuran at room temp-erature. After a 12h reaction time the reaction mixture was cooled (0°C) and quenched by the slow addition of cold water. The mixture was then extracted with ether and the combined organic extracts were washed with brine and dried over anhy-drous magnesium sulphate. Evaporation of the solvent and rapid d i s t i l l a t i o n of the residue (air-bath temperature o.85°C, 0.4 Torr) gave 24.3g (84%) of t r i - n - b u t y l t i n hydride. 52 3-Iodo-2-cyclopenten -1-one was prepared by the reaction ( i n a c e t o n i t r i l e ) of 1,3-cyclopentanedione with triphenylphos-phine diiodide i n the presence of triethylamine. Chlorotrimethylsilane was d i s t i l l e d just p r i o r to use from N,N-dimethylaniline. Methyl chloroformate (Aldrich) was pretreated with anhy-drous sodium carbonate and then stored over 4A molecular sieves. Commercial bromoform and iodomethane were passed through a short column of A c t i v i t y I basic alumina just p r i o r to use. Sodium hydride was obtained as a 50% dispersion in mine-r a l o i l from the A l d r i c h Chemical Company, Inc. and was rout-inely washed free of o i l (with ether or tetrahydrofuran as app-ropriate) in the reaction f l a s k i n which i t was to be used. - 196 -• - Tetrahydrofuran, d i e t h y l ether and diglyme were freshly d i s t i l l e d from sodium benzophenone ketyl under argon. Triethylamine, benzene and pentane were d i s t i l l e d from o calcium hydride and stored over 4A molecular sieves. Hexamethylphosphoramide was d i s t i l l e d from barium oxide and stored over 13X molecular sieves. Dichloromethane and a c e t o n i t r i l e were both d i s t i l l e d from phosphorous pentoxide and a c e t o n i t r i l e was stored over 4A sieves. N,N-Dimethylformamide was d i s t i l l e d under reduced press-ure (^20 Torr) from calcium hydride and was stored over 4A molecular sieves. A l l other reagents were commercially available and were u t i l i z e d without further p u r i f i c a t i o n . Petroleum ether ref e r s to the low b o i l i n g (35-60°C) petroleum f r a c t i o n . Diethyl ether i s referred to simply as - 197 -Preparation of 2-Cyclopenten-l-ol (114) CI OH 113 114 Dry hydrogen chloride gas (59.8g, 1.64 mol) was bubbled o * into 108g (1.65 mol) of cold (-78 C), neat cyclopentadiene contained i n a three-necked f l a s k protected by a calcium chlo-ride drying tube. The contents of the flask were d i s t i l l e d d i r e c t l y under reduced pressure and the l i q u i d d i s t i l l i n g at ^19°C at 5 Torr was c o l l e c t e d (receiver cooled to -78°C) to y i e l d 132g (73%) of the colourless 3-chlorocyclopentene (113) 75 o ( l i t . bp 18-25 C, 5 Torr) which was used immediately for the next synthetic step. A portion (57.5g, 0.56 mol) of the neat chloride (113) was added dropwise over a period of l h to a vigorously s t i r r e d i c e - c o l d s l u r r y of sodium bicarbonate (53g, 0.63 mol) in 200 mL of water. The resultant mixture was s t i r r e d for a further 2 h at 0°C, then i t was saturated with sodium chloride and extracted twice with ether (250 mL). The combined ether extracts were dried over anhydrous sodium s u l -phate and the solvent was evaporated under reduced pressure to afford 35g of crude l i q u i d . F ractional d i s t i l l a t i o n of t h i s residue yielded 26g (38% based on cyclopentadiene) of 2-cyclo-* Prepared by the thermal cracking of dicyclopentadiene as described in reference 75. - 198 -penten-l-ol (114) (bp 51-52°C, 14 Torr; l i t . 7 4 bp 52°C, 12 Torr). This material was one component by glc analysis and exhibited i r ( f i l m ) : 3300, 3045, 1615, 1455, 1355, 1320, 1115, 1050, 975, 920, 880, 740 cm - 1; % nmr (100 MHz)6 : 1.50-1.90 (m, 2H, -OH, one of -CHgCHOH), 2.01-2.80 (m, 3H, -CH2CH=CH-, one of -CH2CH0H), 4.72-4.98 (m, 1H, -CHOH), 5.72-6.06 ( m, 2H, -CH=CH-). Preparation of the g-Methoxyethoxymethyl (MEM) Ether (92) OMEM -MEM=-CH2OCH2 CH 20CH 3 To a cooled (0°C) solution of the alcohol (114) (6.0g, 71.4 mmol) and diisopropylethylamine (15.42g, 119.3 mmol) in 60 mL of dry methylene chloride was added with s t i r r i n g 12.23 76 mL (107 mmol) of B-methoxyethoxymethyl chloride over a period of ^ l h . After completion of the addition, s t i r r i n g was continued for a further 5h at room temperature. The re-action mixture was then washed twice with 50 mL portions of 2% hydrochloric acid, twice with 3% aqueous sodium carbonate (2x30 mL), once with water and f i n a l l y dried over anhydrous magnesium sulphate. The solvent was evaporated under reduced pressure and the residue was d i s t i l l e d (bp 42-43°C, 0.2 Torr; - 199 -C O 0 l i t . a i r bath temperature 35-45 C, 0.1 Torr) to y i e l d 9.7g (80%) of the MEM ether (92). This material was one.component by glc analysis and exhibited i r ( f i l m ) : 3060, 1620, 1460, 1370, 1250, 1210, 1185, 1160, 1120, 1045, 990, 945, 925, 885, 860, 740 cm - 1; 1H nmr (100 MHz)6 : 1.60-2.50 (m, 4H, =CHCH2CH2-), 3.35 (s, 3H, -OCHg), 3.46-3.76 (m, 4H, -OCHgCHgO-), 4.62-4.86 (hidden multiplet, 1H, -CHOMEM), 4.74 (s, 2H, -OCH20-), 5.74-6.06 (m, 2H, -CH=CH-). 74 Preparation of 3-Methoxycyclopentene (78) OMG To an ice- c o l d , vigorously s t i r r e d s l u r r y of sodium b i -carbonate (36g, 0.43 mol) i n spectral grade methanol (150 mL) was slowly added 40g (0.39 mol) of 3-chlorocyclopentene (113) (prepared as described previously) over a period of about 2h. After a further 2h of s t i r r i n g (0°C), the reaction mixture was f i l t e r e d and the f i l t r a t e was poured into %200 mL of water. The layers were separated and the aqueous layer was saturated with sodium chloride. Allowing t h i s mixture to stand resulted in separation from the aqueous layer of some more organic l i -quid. This was removed and the aqueous layer was twice ex-- 200 -tracted with ether (2x75 mL). The combined organic layers were then dried over anhydrous magnesium sulphate overnight. Subsequent ca r e f u l f r a c t i o n a l d i s t i l l a t i o n at atmospheric pressure yielded 20g (52%) of colourless l i q u i d (bp 106-108°C; 74 o l i t . bp 108 C, 760 Torr) . This material was one component by glc analysis and exhibited i r ( f i l m ) ; 3020, 1610, 1455, 1365, 1200, 1120, 1090, 1035, 960, 935, 890, 740 cm - 1; 1H nmr (80 MHz)<5: 1.55-2.58 (m, 4H), 3.34 (s, 3H, -OCHg), 4.35-4.65 (broad m, IH, -CHOMe), 5.83-6.15 (m, 2H, -CH=CH-). Preparation of 2-Cyclohexen-l-ol (116). OH To a cooled (-78°C) solution of 2-cyclohexene-l-one (9.6g, 100 mmol) in 100 mL of dry ether under an argon atmos-78 phere was added a solution of diisobutylaluminum hydride in hexane (115 mL, 115 mmol). The reaction mixture was s t i r r -ed for l h at -78°C, l h at 0°C and then quenched by the slow addition of saturated aqueous ammonium chloride (75 mL). S t i r r i n g for about 20 min and warming to room temperature resulted in the p r e c i p i t a t i o n of a copious quantity of alu -minum s a l t s . Ether (100 mL) was then added and the mixture was f i l t e r e d , with suction, through a sintered glass funnel. - 2 0 1 -The aluminum s a l t s were washed thoroughly with ether and the combined ether f i l t r a t e s were washed once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent, followed by d i s t i l l a t i o n of the residue ( a i r bath temperature ^35°C, 0.85 Torr; l i t . 7 7 bp 65°C, 7 Torr) yielded 9.2g (92%) 77 of 2-cyclohexen-l-ol (116). This material was pure by glc and t i c ( s i l i c a gel, 30% ether- petroleum ether) analyses and exhibited i r ( f i l m ) : 3300, 3000, 1640, 1445, 1430, 1385, 1285", 1160, 1135, 1060, 1005, 960, 900, 810, 730, 675 cm - 1; "hi nmr (80 MHz)6 : 1.38-2.20 (m, 7H), 4.00-4.38 (broad m, IH, -CH-OH), 5.58-6.00 (m, 2H, -CH=CH-). Preparation of 3-(2-Tetrahydropyranyloxy)cyclohexene (108) OTHP A solution of 4,4g (44.84 mmol) of 2-cyclohexen-l-ol (116), 5.66g (67.3 mmol) of 5,6-dihydropyran and 1.12g (4.48 80 mmol) of pyridinium p_-tolUenesulphonate in 100 mL of dry methylene chloride was s t i r r e d for 3h at room temperature. The reaction mixture was then d i l u t e d with ether (100 mL), washed once with half-saturated brine and once with brine and the organic layer was dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure, followed by - 202 -d i s t i l l a t i o n of the residue (air-bath temperature ^70°C, 0.4 Torr) yielded 7.65g (94%) of the tetrahydropyranyl- ether (108) which as ^98% pure by glc analysis. The material exhibited i r ( f i l m ) : 3000, 1640, 1450, 1440, 1380, 1375, 1350, 1340, 1315, 1260, 1200, 1135, 1120, 1080, 1070, 1035, 1025, 1015, 995, 960, 915, 875, 815, 730 cm - 1; 1H nmr (80 MHz)6 : 1.25-2.15 (m, 12H), 3.30-4.35 ( m, 3H, -OCHg-, -0CHCH=), 4.50-5.00 (m, IH, -0CH0-), 5.60-6.02 (m, 2H, -CH=CH-). Exact Mass calcd. for C 1 1 H 1 Q 0 o : 182.1287; found: 182.1297. Preparation of 3-Phenylcyclohexene (109) Ph fe A solution of 11.7g (74.52 mmol) of bromobenzene i n 20 mL of dry ether was added, dropwise with s t i r r i n g , to a mix-ture of magnesium turnings (1.8g, 74.1 mmol) and ether (75 mL) under argon. The rate of addition was such that gentle r e f l u x of the ether was maintained. On completion of the addition, the reaction mixture was refluxed for 20 min and then cooled o 85 to 0 C in an ice bath. Cuprous bromide-dimethyl sulphide (1.53g, 7.5 mmol) was added and the mixture was s t i r r e d at o 8' 0 C for about 10 min. A solution of 3-bromocyclohexene (119) - 203 -(8g,49.7 mmol) in 10 mL of dry ether was then added, v i a a syringe, and s t i r r i n g was continued for 2h at 0°C. The re-action was quenched, at 0°C, by the slow addition of saturated ammonium chloride solution (50 mL) and the mixture was vigor-ously s t i r r e d for ^15 min. The ether layer was separated and washed twice with a 1:1 mixture of saturated ammonium chloride and ammonium hydroxide solutions. After extraction of the aqueous layers once with ether, the combined ether layers were washed with water (1x25 mL) and then dried over anhydrous mag-nesium sulphate. Evaporation of the solvent and d i s t i l l a t i o n (air-bath temperature 'v65°C, 0.85 Torr; l i t . 8 2 bp 70-71°C, 2 Torr) of the residue yielded 6.8g (91%) of 3-phenylcyclohexene 81 (109) which was ^98% pure by glc analysis. This material exhibited i r ( f i l m ) : 3040, 3000, 1590, 1485, 1445, 1430, 880, 755, 725, 700, 675 cm - 1; "Si nmr (100 MHz)6 : 1.44-2.24 (m, 6H),3.28-3.54 (broad m, 1H, W^=13Hz, -CHPh), 5.62-6.06 (m, 2H, -CH=CH-), 7.10-7.44 (m, 5H, aromatic protons). 81 Preparation of 3-Methylcyclohexene (84) C H 3 The procedure c l o s e l y followed that already described for the preparation of 3-phenylcyclohexene (109). Thus, to a cooled (0°C) solution of methylmagnesium iodide - prepared - 204 -from,3.6g (148 mmol) of magnesium turnings and 9.3 mL (150 mmol) of methyl iodide i n V100 mL of dry ether - was added 85 2g (9.80 mmol) of copper bromide-dimethyl sulphide and the mixture was s t i r r e d at 0°C for ^ lOmin. A solution of 3-bromo-84 cyclohexene (119) (18g, 112 mmol) i n 20 mL of dry ether was then added, v i a syringe, and s t i r r i n g was continued for 2h at 0°C. The reaction mixture was allowed to warm to room temp-erature over an hour, then recooled to 0°C and worked up as described already for the preparation of 3-phenylcyclohexene (109). The solvent used for extraction was d i s t i l l e d o f f at atmospheric pressure and the residue was d i s t i l l e d , also at atmospheric pressure, (bp 100°C; l i t . 8 1 bp 103-104°C, 760 Torr) to y i e l d 6.8g (63%) of material which was one component by glc analysis and exhibited i r ( f i l m ) : 3000, 1450, 1370, 1145, 1120, 985, 970, 860, 720, 670 cm - 1; 1H nmr (60 MHz)6: 0.93 (d, 3H, J=7Hz, -CHCHg), 0.8-2.30 (m, 7H), 5.28-5.73 (m, 2H, -CH=CH-). 83 Preparation of 3-(Trimethylsilyl)cyclohexene (110) 5__£ 6 5 SMG 3 1 1 0 A s t i r r e d mixture of chlorotrimethylsilane (9.46 mL, 74.53 mmol), magnesium turnings (3.0g, 124.2 mmol) and 100 mL - 205 -of dry tetrahydrofuran under argon was cooled to ^5°C i n an 84 ice-bath. A solution of 3-bromocyclohexene (119) -(10g, 62.1 mmol) in 20 mL of dry tetrahydrofuran was added slowly, v i a dropping funnel, over a period of l h . The reaction mix-ture was allowed to warm to room temperature and s t i r r e d over-night (^15h). To the r e s u l t i n g white s l u r r y was added ^60 mL of water followed by 100 mL of ether. The layers were sepa-rated and the organic layer was washed three times with water (3x25 mL). The combined aqueous layers were saturated with sodium chloride and extracted with ether (^75 mL). The com-bined ether layers were then washed once with a c i d i f i e d brine (30 mL), twice with brine (2x30 mL) and dried over anhydrous magnesium sulphate. Removal of the solvent gave a l i q u i d r e s i -due, consisting of an approximately 1:1 mixture of two compon-ents (glc an a l y s i s ) , which was f r a c t i o n a l l y d i s t i l l e d . The more v o l a t i l e component d i s t i l l e d (bp 62-64°C, 20 Torr; l i t 8 3 b bp 69-72°C, 10 Torr) to y i e l d as a colourless o i l , 2.6g (27%) 83 of 3-(trimethylsilyl)cyclohexene (110) This material gave one peak on glc analysis and exhibited i r ( f i l m ) : 2995, 1430, 1245, 895, 865, 860, 835, 750, 705 cm - 1; 1H nmr (100 MHz)6 : -0.02 (s, 9H, S i ( C H 3 ) 3 ), 1.32-2.20 (m, 7H), 5.66 (broad s, 2H, -CH=CH-). The higher b o i l i n g component d i s t i l l e d (bp 110-112°C, 20 Torr; l i t . 8 8 b bp 110°C, 15 Torr) to y i e l d , as a colourless o i l , 2.3g (24%) of the known, undesired coupling product 2,2^-bicy-88 1 clohexenyl C.121). This material exhibited H nmr (270 MHz)6: - 206 -1.24-1.86 (in, 8H, protons at C5,C6,C5',C6'. ), 1.91-2.04 (m, 4H, a l l y l i c methylene protons), 2.04-2.22 (m, 2H, a l l y l i c methine protons), 5.51-5.79 (m, 4H, o l e f i n i c protons). 9°. Preparation of 3-Buten-l-ol (122) To 17g (0.7 g-atom) of dried magnesium i n a one l i t r e , three-necked f l a s k equipped with a s e l f - e q u a l i z i n g dropping funnel, r e f l u x condenser and nitrogen i n l e t , 400 mL of dry ether was added. The reaction f l a s k was cooled i n an ice bath and a solution of freshly d i s t i l l e d a l l y l bromide (72.6g, 0.6 mol) in 60 mL of dry ether was added at a rate such that 94 95 gentle reflux of the ether was maintained. ' On completion of the addition (0.5h), 21g (0.7 mol) of paraformaldehyde (dried overnight under vacuum with P 2°5) w a s added. The re-action mixture was refluxed for 20h and then was poured into an ice-water mixture containing 70 mL of 30% sulphuric acid. The r e s u l t i n g l i q u i d was steam d i s t i l l e d u n t i l only water seemed to be d i s t i l l i n g (^lh). The d i s t i l l a t e was saturated with sodium chloride and the organic layer was separated. After extraction of the aqueous layer with ether (2x75 mL), the organic layers were combined and dried over anhydrous - 207 -magnesium sulphate. Removal of the solvent by d i s t i l l a t i o n at atmospheric pressure, followed by f r a c t i o n a l d i s t i l l a t i o n of the residue also at atmospheric pressure, gave 14g (33%) of 3-buten-l-ol (122)(bp 110-112°C, atmospheric pressure; 93 o l i t . bp 111-113 C, 760 Torr). This material exhibited i r ( f i l m ) : 3325, 3050, 1630, 1050, 915 cm - 1; 1H nmr (60 MHz) : 2.25 (q, 2H, J=7Hz, CH2=CH-CH2~), 2.30 (s, IH, -OH), 3.60 ( t , 2H, J=7Hz, -CH20H), 4.80-5.23 (m, 2H, -CH=CH2), 5.40-6.10 (m, IH, CH2=CH). 92 97 Preparation of 4-Chlorotetrahydropyran (124) ' A s t i r r e d mixture of 3-buten-l-ol (122) (9.65g, 134 mmol) and s-trioxan (4g, 44.4 mmol) cooled to 0°C, was treated with dry hydrogen chloride gas (-6.4g, 175 mmol). The reaction mix-ture, semi-solid at f i r s t , became a clear solution and was warmed to room temperature and s t i r r e d overnight. Ether (100 mL) was then added and the layers were separated. The ether layer was washed once with brine and dried over anhydrous mag-nesium sulphate. Removal of the solvent by d i s t i l l a t i o n at atmospheric pressure, followed by d i s t i l l a t i o n of the residue (39-41°C, 13 Torr; l i t . 9 2 bp 56-59°C, 19 Torr) yielded lOg (62%) of 4-chlorotetrahydropyran (124) as a colourless o i l . - 208 -This material gave one peak on glc analysis and exhibited i r ( f i l m ) : 2930, 1440, 1300, 1260, 1225, 1140, 1090, 1070, 1020, 1005, 835, 760, 720 cm"1; 1H nmr (60 MHz)5 : 1.50-2.30 (m, 4H, -CH2CHC1 CH 2-), 3.27-4.40 (complex m, 5H, -CH^OCHg-, -CHC1-). Preparation of 5,6-Dihydro-2H-pyran (111) A mixture of 4-chlorotetrahydropyran (124) (2.8g, 23.2 mmol) and powdered 85% potassium hydroxide (4.6g, 69.7 mmol) 92 in 40 mL of ethylene g l y c o l was refluxed for 6h. Direct d i s t i l l a t i o n of the reaction mixture at atmospheric pressure o 92 yielded a colourless l i q u i d d i s t i l l i n g at 90-92 C ( l i t . 93.5°C, 760 Torr). This material (1.3g, 67%) gave a single peak on glc analysis and exhibited i r (film) : 3010, 1420, 1375, 1230, 1180, 1090, 1075, 1035, 975, 915, 890, 840, 770, 650 cm - 1; XH nmr (60 MHz) 6: 1.90-2.27 (m, 2H, -CH2CH=CH-), 3.75 ( t , 2H, J=5Hz, -OCH2CH2-), 4.00-4.20 (m, 2H, -OCH2CH=CH-), 5.43-5.87 (m, 2H, -CH=CH-). - 209 -Preparation of 2-(2-Cyclopehteny1)ethanol (126) To a cooled (0°C) s l u r r y of 6.Og (158 mmol) of lithium aluminum hydride in 100 mL of dry ether was added slowly a solution of 20g (158.7 mmol) of 2-(2-cyclopentenyl)acetic acid in 20 mL of dry ether. On completion of the addition, s t i r r i n g was continued overnight at room temperature. The reaction mix-ture was quenched, under an argon atmosphere, by the slow addi-t i o n of cold water (50 mL) to the ice-cooled mixture. Ether (100 mL) was then added and the ether layer was separated from the aqueous s l u r r y of aluminum s a l t s . The l a t t e r was extracted with a further 100 mL of ether and the combined ether layers• were washed twice with brine and then dried over anhydrous mag-nesium sulphate. Removal of the solvent followed by d i s t i l l a -t i o n of the residue (air-bath temperature 55-60°C, 1 Torr; 102 l i t . bp not quoted) gave 16.9g (95%) of the alcohol (126). This material was one component by glc and t i c ( s i l i c a gel, 30% ether- petroleum ether) analyses and exhibited i r ( f i l m ) : 3325, 3025, 1605, 1430, 1355, 1060, 1010, 980, 915, 880, 720 cm - 1; """H nmr (80 MHz)6 : 1.17-2.53 (m, 6H), 2.53-3.00 (m, 2H, -OH, -CH-CH=CH-), 3.73 ( t , IH, J=6.5Hz, -CHgOH), 5.58-5.88 (m, 2H, -CH=CH-). - 210 -Preparation of 2-Oxabicyclo[3.3.0]oct-7-ene (112) To a s t i r r e d solution of the alcohol (126) (9.0g, 80.4 mmol) in 75 mL of dry methylene chloride at -78°C, under argon, was added s o l i d phenylselenenyl chloride (16.92g, 88.4 mmol). S t i r r i n g was continued at -78°C u n t i l the s o l i d had dissolved and t i c analysis ( s i l i c a gel, 20% ether- petroleum ether) i n -dicated that the reaction was complete ( 0.5h). The reaction mixture was then warmed to room temperature, washed once with saturated aqueous sodium bicarbonate (25 mL), once with water (20 mL) and the organic layer was dried over anhydrous magne-sium sulphate. Evaporation of the solvent under reduced pres-sure gave 20.3g (95%) of the phenyl selenoether ( 1 2 7 ) 1 0 1 as a yellow o i l containing traces of diphenyl diselenide. This material was dissolved in 100 mL of dry methylene chloride, the solution was cooled to -78°C and ozone was bubbled in u n t i l the pale yellow solution turned a l i g h t but persistent green. Excess ozone was removed by bubbling f i r s t oxygen and then ar-gon into the cooled (-78°C) reaction mixture. Then dry t r i -ethylamine (21.3 mL, 76 mmol) was a d d e d 1 0 1 , 1 0 3 and the mixture was s t i r r e d overnight at room temperature. The reaction mixture was washed twice with a c i d i f i e d brine, once with saturated - 211 -sodium bicarbonate and three times with brine. F i n a l l y , the organic layer was dried over anhydrous magnesium sulphate. Removal of the solvent by d i s t i l l a t i o n at atmospheric pressure, followed by d i s t i l l a t i o n of the residue (air-bath temperature ^55°C, 20 Torr) gave 6.5g (74% based on s t a r t i n g alcohol) of the alkene (112) as a colourless l i q u i d . This material was one component by glc and t i c ( s i l i c a gel, 10% ether- petroleum ether) analyses and exhibited i r ( f i l m ) : 3020, 1605, 1440, 1350, 1230, 1060, 1030, 990, 930, 900, 740, 715 cm"1; ^ nmr (80 MHz)6 : 1.35-3.10 (m, 5H), 3.38-3.92 (m, 2H, -OCHg-), 4.95-5.20 (m, IH, -OCH-CH=CH-), 5.51-6.00 (m, 2H, -CH=CH-). 106 Preparation of Methyl 2-Oxocyclohexanecarboxylate (130) 0 02Me To a dried 500 mL three-necked round-bottomed fla s k , equipped with a reflux condenser, pressure-equalized dropping funnel and nitrogen i n l e t , was added 22.5g (0.47 mol) of a 50% dispersion of sodium hydride in o i l . The o i l was removed by washing the mixture with dry tetrahydrofuran (3x30 mL). Dry tetrahydrofuran (120 mL) and 33.8g (0.38 mol) of dimethyl carbonate were added to the sodium hydride and the r e s u l t i n g mixture was heated to reflux with brisk s t i r r i n g . A solution - 212 -of 14.7g (0.15 mol) of cyclohexanone in 40 mL of dry tetrahy-drofuran was added dropwise v i a the addition funnel.. After about two minutes of t h i s addition, ^100 mg of potassium hy-dride (24.5% in o i l ) was introduced, as a s l u r r y i n tetrahy-drofuran, to i n i t i a t e the reaction. Reflux of the reaction mixture was maintained u n t i l a l l the cyclohexanone had been added and was continued for a further 0.5h thereafter. The reaction mixture was then cooled in an ice-bath and quenched by the slow addition of 155 mL of 5% aqueous hydrochloric acid followed by 100 mL of brine. The layers were separated and the aqueous phase was extracted with chloroform (2x100 mL). The combined organic layers were washed once with water (40 mL) and once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure and d i s t i l l a t i o n of the residue (air-bath temperature 45°C, 0.4 Torr; l i t . 1 0 6 bp 68°C, 0.8 Torr) afforded 20g (85%) of the B-keto ester (130) as a colourless l i q u i d which was one component by glc analysis and exhibited i r ( f i l m ) : 1740, 1710, 1655, 1615, 1445, 1385, 1365, 1305, 1270, 1225, 1180, 1090, 840 cm - 1; 1H nmr (80 MHz)5: 1.40-2.60 (m, 8H), 3.27-3.54 (m, ^0.25H, - 8 c H C 0 2Me), 3.75 (s, 3H, -C0oMe), 9.186 (s, ^0.75H, H0-C=C-C0QMe). - 213 -Preparation of Methyl 2-CDiethylphosphoryloxy)cyclohexene-107 carboxylate (131) To a s t i r r e d , cooled suspension of 1.52g (63.3 mmol) of o i l - f r e e sodium hydride i n 80 mL of dry ether under a nitrogen atmosphere was added dropwise, a solution of the B-keto ester (130) (9g, 57.7 mmol) in 10 mL of dry ether. After the almost s o l i d reaction mixture had been s t i r r e d f o r ^20 min at 0°C, 9.2 mL (63.3 mmol) of die t h y l phosphorochloridate was added dropwise (at 0°C). On completion of the addition, s t i r r i n g was continued for 3h at room temperature. Saturated aqueous ammonium chloride (>30 mL) was then cautiously added to the cooled, s t i r r e d reaction mixture. The layers were separated and the aqueous layer was extracted once with ether. The com-bined ether layers were washed once with water, once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure yielded 15.9g (94%) of the 107 enol phosphate (131) as a clear l i q u i d which was -v98% pure on glc analysis and was used d i r e c t l y in the next reaction (coupling with lithiu m dimethylcuprate). A sample of the material exhibited i r ( f i l m ) : 1715, 1650, 1435, 1370, 1295, 1245, 1200, 1150, 1120, 1080, 1060, 1040, 990, 970, 930, 850 cm - 1; -'•H nmr (100 MHz) 6 : 1.36 (d of t, 6H, J=l, 7Hz, - 214 --PO(OCH 2CH 3) 2) , 1.52-1.86 (m, 4H), 2.20-2.60 (m, 4E), 3.74 (s, 3H, -C02Me), 4.23 (qn, 4H, J=7Hz, -PO(OCH 2CH 3) g). Preparation of Methyl 2-Methylcyclohexenecarboxylate (132) 107 To a cooled (0°C) s t i r r e d suspension of p u r i f i e d cuprous iodide 1 1 0 (15.6g, 81.7 mmol) in 170 mL of dry ether under an atmosphere of nitrogen was added an ethereal solution of methyllithium (163.4 mmol). A solution of the enol phosphate (131) (15.8g, 54.1 mmol) in 20 mL of dry ether was added drop-wise to t h i s freshly prepared solution of lithium dimethyl cuprate at 0°C. The purple reaction mixture was s t i r r e d at t h i s temperature for 2h and for an additional 2h afte r remov-a l of the cooling bath. The reaction mixture was then quenched (at 0°C) by the addition of a 5:1 mixture of saturated aqueous ammonium chloride and concentrated ammonium hydroxide solu-tions (60 mL). The layers were separated and the aqueous lay-er was extracted with ether (2x100 mL). The combined ether extracts were washed once with water then with brine and dried over anhydrous magnesium sulphate. The solvent was removed under reduced pressure to y i e l d %8.2g of a l i q u i d residue which on glc analysis contained -v-10% of an impurity. Flash chromatography (lOOg s i l i c a gel, 15% ether- petroleum ether) - 215 -of t h i s residue i n three batches followed by d i s t i l l a t i o n of the material obtained on concentrating the appropriate f r a c -tions yielded 6.4g (76%) of the sweet smelling, colourless a,3-unsaturated ester ( 1 3 2 ) 1 0 7 (air-bath temperature 90°C, ^20 Torr; l i t . 1 0 7 air-bath temperature 85-88°C, 20 Torr). This material was one component on glc analysis and exhibited i r ( f i l m ) : 1705, 1635, 1435, 1280, 1230, 1210, 1085, 1055, 765 cm - 1; XH nmr (80 MHz)5: 1.43-1.80 (m, 4H), 1.90-2.40 (m, 4H), 2.00 (broad s, 3H, v i n y l methyl), 3.73 (s, 3H, -COgMe). Reduction of Methyl 2-Methylcyclohexenecarboxylate 133 To a cooled (-78 C) solution of the unsaturated ester (132) (4g, 26.0 mmol) in 50 mL of dry ether, under an argon atmosphere, was added a solution of diisobutylaluminum hydride in hexane (57 mL, 57 mmol). The reaction mixture was s t i r r e d for l h at -78°C, l h at 0°C and then was quenched by the slow addition of saturated aqueous ammonium chloride (50 mL). S t i r r i n g the mixture for about 30 min. while allowing i t to warm to room temperature resulted i n the p r e c i p i t a t i o n of a copious quantity of aluminum s a l t s . Ether (75 ml) was added and the mixture was f i l t e r e d , with suction, through a short - 216 -column of F l o r i s i l (^50g). The aluminum s a l t s were washed thoroughly with ether and the combined ether f i l t r a t e s were washed once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure followed by d i s t i l l a t i o n of the residue (air-bath temperature 60-70°C, 0.4 Torr) yielded 3.0g (92%) of the unsaturated a l -cohol (133). This material gave once peak on glc analysis, one spot on t i c analysis ( s i l i c a gel, 40% ether- petroleum ether) and exhibited i r ( f i l m ) ; 3300, 1660, 1450, 1440, 1145, 1025, 1000 cm - 1; XH nmr (80 MHz)6 : 1.34 (broad s, 1H, -OH), 1.43-1.78 (m, 4H), 1.70 (broad s, 3H, v i n y l methyl), 1.80-2.30 (m, 4H). 4.13 (s, 2H, -CHgOH). Exact Mass calcd. for C RH 1 40 : 126.1045; found 126.1034. Preparation of l-Methoxymethyl-2-methylcyclohexene (99). To a s t i r r e d s l u r r y of o i l - f r e e sodium hydride (1.54g, 64.3 mmol) in 30 mL of dry tetrahydrofuran, under an argon at-mosphere, was slowly added a solution of 2.95g (23.4 mmol) of the alcohol (133) in 10 mL of dry t e t r a h y d r o f u r a n . 1 1 1 The mixture was refluxed for l h with constant s t i r r i n g , cooled to room temperature and 7.4 mL (119 mmol) of methyl iodide was - 217 -added dropwise. The reaction mixture was then s t i r r e d over-night at room temperature. Saturated aqueous ammonium chlo-ride (20 mL) was subsequently added to the cooled (0°C) re-action mixture and the layers were separated. The aqueous layer was extracted once with ether and the combined ether layers were washed once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure, followed by d i s t i l l a t i o n of the residue (air-bath temperature 35°C, 0.4 Torr) yielded 3.0g (90%) of the un-saturated methyl ether (99). This material, one component on glc analysis exhibited i r ( f i l m ) : 1670, 1460, 1450, 1385, 1190, 1140, 1100, 960, 920 cm - 1; 1H nmr (100 MHz)5: 1.46-1.74 (m, 4H), 1.69 (broad s, 3H, v i n y l methyl), 1.84-2.15 (m, 4H), 3.30 (s, 3H, -0CH 3), 3.89 (s, 2H, -CHgOMe). Exact Mass calcd. for C „ H 1 c O : 140.1201 ; found : 140.1186. 9 16 General Procedure A : Preparation of Dibromocyclopropanes.~" To a vigorously s t i r r e d solution of the appropriate a l -kene, bromoform (1.5-3.0 equiv.), 95% ethanol (0.1-0.2 mL) and triethylbenzylammonium chloride ('vO.Ol equiv.) in methylene chloride (25-100 mL) was added dropwise 50% aqueous sodium hy-63 droxide solution (/\>9 equiv). The mixture was then heated at 40°C ( o i l bath) for 4-50h with e f f i c i e n t s t i r r i n g . The r e s u l t -ing dark, viscous reaction mixture was d i l u t e d with methylene chloride (or another desired solvent) and f i l t e r e d through a - 218 -column^ of F l o r i s i l ( 3 0 - 7 5 g t h e column was eluted with further quantities of methylene chl'oride and the combined eluate was washed once with water, once with brine and dried over anhydrous magnesium sulphate. Evaporation of the solvent under reduced pressure yielded an o i l containing bromoform. The bromoform was generally removed by heating the mixture at 40°C (0.4 Torr) and the dibromocyclopropane(s) was (were) i s o l a t e d by chroma-tography and/or d i s t i l l a t i o n . Preparation of trans and cis-6,6-Dibromo-2-methoxybicyclo [3.1.0lhexane (79a) and (79b) OMe OMe 79a 79b Following the general procedure A for the preparation of dibromocyclopropanes, a two phase mixture of the alkene (78), (5.0g, 51.0 mmol), bromoform (6.7 mL, 76.3 mmol), t r i e t h y l b e n z y l -ammonium chloride (116 mg, 0.51 mmol), 50% sodium hydroxide (26 mL),0.1 mL of 95% ethanol and 50 mL of methylene chloride was vigorously s t i r r e d at 40°C for 24h. The usual workup yielded ^17g of crude l i q u i d which was subjected to f l a s h chromatogra-115 phy (150g s i l i c a gel, 5% methylene chloride - 5% ether - 90% petroleum ether) in three batches. A t o t a l of 7.1g (52%) of the trans-2-methoxy-dibromocyclopropane C79a) w a s obtained after d i s -t i l l a t i o n (air-bath temperature 65°C,0.4 Torr) of the f i r s t eluted - 219 -compound. This material was one component by glc and t i c ana-lyses and exhibited i r ( f i l m ) : 3025, 1460. 1440, 1370, 1350, 1290, 1275, 1210, 1200, 1190, 1110, 1100, 1065, 1030, 1000, 935, 865, 750 cm"1; 2H nmr (400 MHz) 6: 1.73-1.90 (m, 2H), 2.00-2.12 (m, 1H), 2.15-2.26 (m, 1H), 2.36 ( t , 1H, J=7Hz, H^), 2.43 (d, lH,J=7Hz, Hg), 3.37 (s, 3H, -OCHg), 3.83 (d of d), 1H, J=2, 6.5Hz, -CHOMe). Exact Mass calcd. for C 7 H 1 Q 7 9 B r O (M +- 8 1Br) : 188.9915; found: 188.9913 The l a t e r eluted cis-2-methoxydibromocyclopropane (79b) ( 400 mg, 2.9%) was homogeneous by glc and t i c analyses and exhibited i r ( f i l m ) ; 3000, 1450, 1430, 1350, 1300, 1280, 1210, 1200, 1120, 1100, 1075, 1040, 1010, 995, 980, 950, 815, 760 cm - 1; 1H nmr (400 MHz) <5 : 1.77-1.89 (m, 1H), 2.00-2.11 (m, 3H), 2.23-2.30 (m, 1H), 2.36 (d of d, 1H, J=6,7Hz, H^), 3.47 (s, 3H, -0CH 3), 4.34-4.42 (m, 1H, -CHOCHg). Exact Mass calcd. for 7 9 81 C^ H., n Br O J-BrO : 269.9078; found 269.9078. 58 Preparation of the Dibromocyclopropanes (93a) and (93b) Br. Br OMEM 93a -MEM=-CH2OCH2CH2OCH3 - 220 -Following the general procedure A for the preparation of dibromocyclopropanes, a two-phase mixture of the alkene (92) (2g, 11.63 mmol), bromoform (1.7 mL, 19.37 mmol), t r i e t h y l -benzylammonium chloride (^27mg, 0.12 mmol), 0.1 mL of 95% ethanol, 50% aqueous sodium hydroxide solution (6 mL) and 30 mL of methylene chloride was vigorously s t i r r e d at 45-50°C ( o i l bath) for 26h. The usual workup yielded a dark l i q u i d con-tai n i n g bromoform which was removed by heating at 45-55°C at 0.4 Torr. The residual l i q u i d , which on t i c analysis ( s i l i c a gel, 30% ether - petroleum ether) showed two spots, was sub-115 jected to f l a s h chromatography (lOOg s i l i c a gel, 30% ether -petroleum ether). The less polar, major product, the trans-dibromocyclopropane (93a) was obtained as a colourless l i q u i d (3.0g, 75%) which exhibited i r ( f i l m ) : 2965, 1450, 1365, 1285, 1200, 1180, 1160, 1130, 1110, 1055, 1030, 995, 855, 745 cm - 1; XH nmr (100 MHz) 6: 1.70-2.58 (m, 6H), 3.43 ( s, 3H, -OCHg) 3.50-3.84 (m, 4H, -OCHgCHgO-), 4.25 (broad d, IH, J=6Hz, -CHOMEM), 4.80 (s, 2H, -OCHgO-). The more polar, minor product (^150 mg, ^4%) was the c i s -dibromocyclopropane (93b). This material exhibited i r ( f i l m ) : 2965, 1460, 1360, 1305, 1285, 1190, 1155, 1115, 1100, 1085, 1055, 985, 955, 850, 765 cm"1; "'"H nmr (100 MHz) 6: 1.84-2.32 (m, 5H), 2.56 (d of d, IH, J=5.5, 7Hz, H R), 3.44 (s, 3H, -OCHg), 3.56-4.12 (m, 4H, -OCHgCHgO-), 4.44-4.70 (broad m, IH, -CHOMEM), 4.85 (an AB type d of d, 2H, J=7Hz, -OCHgO-). Exact Mass calcd. 79 81 + for C gH 7 "'Br^Br (M -CjHgOg) : 238.8894; found 238.8894. - 221 -119 Preparation of c l s - B i c y c l o [3.1.0]hexan-2-ol (136) To a ref l u x i n g , s t i r r e d mixture of anhydrous ether (5 mL) 156 and 0.21g (3.08 mg-atom) of zinc-copper couple was added a mixture of 2-cyclopenten-l-ol (114) (lOOmg, 1.2 mmol) and d i -iodomethane (0.5g, 1.86 mmol). After 3h of reflux, the re-action mixture was cooled to room temperature and quenched by the slow addition of saturated ammonium chloride solution (3 mL). The mixture was extracted with ether (3x10 mL) and the combined ethereal extracts were washed once with saturated aqueous sodium bicarbonate (5 mL), once with water (5 mL), twice with brine and, f i n a l l y , dried over anhydrous magnesium sulphate. Removal of the solvent followed by column chromato-graphy (5g s i l i c a gel, 15% ether- petroleum ether then 50% ether- petroleum ether) of the residue and f i n a l l y d i s t i l l a -t i o n (air-bath temperature 20-23°C, 0.4 Torr; l i t . 1 1 9 bp 76°C, 17 Torr) yielded 70 mg (60%) of the ci s - a l c o h o l ( 1 3 6 ) 1 1 9 This material was one component by glc analysis and exhibited i r ( f i l m ) : 3330, 3050, 3005, 1470, 1450, 1415, 1335, 1120, 1070, 1050, 1020, 980, 910, 830, 820, 760 cm"1; "Si nmr (100 MHz) : 0.22-0.70 (m, 2H, two methylene cyclopropyl protons), 0.80-2.06 (m, 7H), 4.42-4.80 (broad m, 1H, -CHOH). Exact Mass calcd. for CgH^O : 98.0732; found: 98.0728. - 222 -Preparation of the cis-MEM Ether (137b) To a s t i r r e d , cooled (0 C) solution of the alcohol (136) (28.3 mg, 0.29 mmol) and diisopropylethylamine (86 yL, 0.49 mmol) in 5 mL of dry methylene chloride was added drop-7 6 wise 50 yL (0.44 mmol) of $-methoxyethoxymethyl chloride. After <v0.5h, the cooling bath was removed and s t i r r i n g was continued for 12h at room temperature. The reaction mixture was d i l u t e d with methylene chloride (%15 mL) and washed once with a c i d i f i e d brine, once with saturated aqueous sodium b i -carbonate solution, once with water and f i n a l l y , once with brine. The organic phase was then dried over anhydrous mag-nesium sulphate and the solvent was removed under reduced pressure. The residue was chromatographed (^2g s i l i c a gel, 20% ether- petroleum ether) to y i e l d after d i s t i l l a t i o n ( a i r -bath temperature 60-70°C, 0.4 Torr) of the crude material thus obtained, 43 mg (80%) of the MEM ether (137b). This material, one component by glc analysis, exhibited i r ( f i l m ) : 3050, 3010, 1470, 1455, 1365, 1200, 1160, 1140, 1120, 1055,1030, 990, 860, 835, 770 cm"1; 1H nmr (100 MHz)6: 0.24-0.65 (m, 2H, two methylene cyclopropyl protons), 1.00-2.00 (m, 6H), 3.42 (s, 3H, -0CH 3), 3.49-3.84 (m, 4H, -OCHgCHgO-), 4.30-4.58 (broad m, IH, - 223 --CHOMEM), 4.81 (AB type doublet of doublets, 2H, J=7Hz, -OCH20-). Exact Mass calcd. for C g H 1 5 0 2 (M+-OCH"3) : 155.1072; found: 155.1075. Preparation of the trans-MEM Ether (137a) A solution of the dibromocyclopropane (9_3a) (lOOmg, 0.29 mmol) in a mixture of dry tetrahydrofuran (4 mL) and dry ether (0.8 mL) under nitrogen was cooled to -127°C (n-propanol- l i q . nitrogen) and 0.2 mL of n-butyllithium i n hexane (0.34 mmol) was added over a period of n,2 min. After the solution had been s t i r r e d for 5 min at -127°C, 1 mL of a 1:1 mixture of meth-anol and ether was added slowly and the reaction mixture was warmed to room temperature. Ether (^15mL) was then added, the layers were separated and the organic layer was washed with water (3x3 mL), with brine (1x5 mL) and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure yielded 62mg of a crude mixture of the monobromocyclo-propanes (94), which was further reduced as follows. A solu-tion of t h i s crude material (62mg, 0.23 mmol) i n 5 mL of dry ether under nitrogen was cooled to -78°C and 0.43 mL (0.46 mmol) of t e r t - b u t y l l i t h i u m in pentane was added dropwise. OMEM -MEM=-CH2OCH2CH2OCH3 - 224 -After the reaction mixture had been s t i r r e d for 2h at -78°C i t was treated with 1 mL of a 1:1 mixture of methanol and ether. Workup,via a procedure i d e n t i c a l to that described above, and d i s t i l l a t i o n (air-bath temperature 60-70°C, 0.4 Torr) of the crude product yielded 33mg of the trans-MEM ether (137a). This material, one component by glc analysis, exhibited i r ( f i l m ) : 3055, 3025, 1460, 1445, 1465, 1360, 1195, 1180, 1160, 1135, 1115, 1095, 1055, 995, 975, 925, 875, 850, 795 cm - 1; 1H nmr (100 MHz)<$ : 0.28-0.58 (m, IH, one cyclopro-pyl proton), 1.12-2.10 (m, 7H), 3.42 (s, 3H, -OCHg), 3.50-3.84 (m, 4H, -OCH^CHgO-), 4.20 (broad d, IH, J=5Hz, -CHOMEM), 4.81 (s, 2H, -0CH-0-). Exact Mass calcd. for C^H,_0o (M +-0CH„): 155.1072; found: 155.1068. Preparation of trans- and cis-7,7-Dibromo-2-(2-tetrahydro-pyranyloxy)bicyclo[4.1.0]heptane (138a) and (138b). 138a 138b Following the general procedure A for the preparation of dibromocyclopropanes, a two-phase mixture of the alkene (108) (8g, 43.9 mmol), bromoform (11.6 mL, 131.7 mmol), t r i -ethylbenzylammonium chloride (100 mg, 0.44 mmol), 95% ethanol - 225 -(0.2 mL), 50% sodium hydroxide (22 mL) and 50 mL of methylene chloride was s t i r r e d vigorously at 40°C for 72h. The usual workup resulted in the i s o l a t i o n of lOg of crude material which on t i c analysis ( s i l i c a gel, 15% ether- petroleum ether) exhibited two spots. High pressure l i q u i d chromatography ( s i l i c a gel, 9% ethyl acetate - petroleum ether, 2 runs) of t h i s material yielded 7g (45%) of the major, more mobile com-ponent, the trans-dibromocyclopropane (138a) which was ^97% pure by glc analysis and exhibited i r ( f i l m ) : 3000, 1450, 1435, 1350, 1320, 1200, 1185, 1170, 1155, 1130, 1115, 1080, 1060, 1030, 1020, 1010, 985, 910, 875, 820, 730 cm"1; XH nmr (80 MHz) 6: 1.00-2.25 (m, 14H), 3.30-4.15 (m, 3H, -OCHg-, -OCH-), 4.68-5.03 (m, 1H, -0CH0-); ms m/e 356/354/352 (M +), 275/273 (M +-Br), 173/171 (M +-C 5H g0 2H-Br), 85 (C 5H g0 +, base peak). Exact Mass calcd. for C 1 2 H l g 7 9 B r 8 1 B r 0 2 : 353.9653 ; found:353.9667. The impure material obtained from the l a t e r eluted HPLC fr a c t i o n s was p u r i f i e d by preparative layer chromatography (15% ether- petroleum ether) to y i e l d 350 mg (2%) of the more polar component, the cis-dibromocyclopropane (138b). This material exhibited i r ( f i l m ) : 2975, 1455, 1440, 1375, 1340, 1255, 1200, 1130, 1115, 1075, 1055, 1025, 980, 900, 865, 815 735, 700 cm"1, 1H nmr (80 MHz) 6: 1.10-2.40 (m, 14H), 3.30-4.42 (m, 3H, -0CH2-, -0CH-), 4.68-4.95, 5.00-5.18 (two multi-p l e t s , 1H, -0CH0-, due to the diastereomers); ms m/e 356/354/ • 352 (M +), 275/273 (M +-Br), 173/171 (M +-C 5H g0 2H-Br), 85 (C 5H g0 +, - 226 -7 9 81 base peak). Exact Mass calcd. for C^2 H18 r 0 2 : 3 5 3 * 9 6 5 3 ' found: 353.9646 Preparation of trans- and cis-7,7-Dibromo-2-methylbicyclo [4.1.Olheptane (85a) and ( 8 5 b ) 1 2 1 . 85a 85b Following the general procedure A for the preparation of dibromocyclopropanes, a two phase mixture of the alkene (84) (6g, 62.5 mmol), bromoform (8.4 mL, 95.7 mmol), t r i e t h y l b e n z y l -ammonium chloride (150 mg, 0.67 mmol), 0.2 mL of 95% ethanol, 50% aqueous sodium hydroxide (31.3 mL) and 50 mL of methylene chloride was vigorously s t i r r e d at 40°C for 6h. In the usual manner, the reaction mixture was f i l t e r e d through a column of s i l i c a gel (^75g) and the column was eluted with 20% ether -petroleum ether. The combined eluate was washed once with wa-ter, once with brine and dried over anhydrous magnesium sulph-ate. Evaporation of the solvent yielded ^18.8g of crude material which contained bromoform and appeared to consist of two other components on t i c analysis ( s i l i c a g el, petroleum - 227 -115 ether).. This material was subjected to f l a s h chromatography"1""""" (180g s i l i c a g e l , petroleum ether) in three batches to y i e l d a f t e r d i s t i l l a t i o n (air-bath temperature 60-65°C, 0.3 Torr) lOg (60%) of a ^6:1 mixture (by c a p i l l a r y glc) of the trans-121 and cis-dibromocyclopropanes (85a) and (85b) , respectively. This mixture was c l e a r l y one spot on t i c analysis ( s i l i c a gel, petroleum ether) and exhibited i r ( f i l m ) : 3000, 1460, 1375, 1340, 1145, 1130, 1100, 1050, 1030, 735 cm - 1; XH nmr (400 MHz)6 0.77-0.89 (in, IH), 1.22, 1.26 (two doublets, 3H, J=7Hz for each -CHCHg of (85a) and (85b), r e s p e c t i v e l y ) , 1.28-1.43, 1.45-1.63, 1.67-1.90 (multiplets, 2H, 3H and 3H); ms m/e 270/268/266 (M +). 7 9 81 Exact Mass calcd. for C g H 1 2 Br Br : 267.9285; found: 267.9286 Anal. calcd. for C g H i 2 B r 2 : C 35.83 , H 4.51, Br 59.66: found: C 36.13, H 4.41, Br 59.48. I d e n t i f i c a t i o n of the more polar compound is o l a t e d (^500 mg) was not pursued. Preparation of 7,7-Dibromo-trans-2-phenylbicyclo [4.1.0] heptane (139). Ph - 228 -^Following the general procedure A l o r the preparation of dibromocyclopropanes, a two phase mixture of the alkene (109) (5g, 31.65 mmol), bromoform (5.6 mL, 63.30 mmol), t r i e t h y l -benzylammonium chloride (150 mg, 0.66 mmol), 0.2 mL of 95% ethanol, 50% aqueous sodium hydroxide (16 mL) and 50 mL of methylene chloride was vigorously s t i r r e d at 40°C for 4h. The usual workup yielded ^9g of crude material consisting of bromo-form and e s s e n t i a l l y one product by glc and t i c ( s i l i c a gel, 115 petroleum ether) analyses. Flash chromatography (180g s i l i c a gel, petroleum ether) of t h i s material yielded 5.8g (56%) of the dibromocyclopropane (139) which was one component by glc and t i c analyses and s o l i d i f i e d on storage at ^ 5°C. R e c r y s t a l l i z a t i o n (petroleum ether) of a portion of t h i s s o l i d gave white c r y s t a l s (mp 48.5-49.5°C) which exhibited i r (Nujol mull): 3035, 3005, 1595, 1490, 1455, 1410, 1375, 1030, 960, 765, 735, 700 cm"1; 1H nmr (100 MHz) 6: 1.12-2.08 (m, 8H), 2.54-2.82 (broad m, IH, W^=19Hz, -CHPh), 7.18-7.50 (m, 5H, aromatic protons); ms m/e 332/330/328 (M +), 171 (M +-Br), and + 7Q R1 91 (C 7H ?). Exact Mass calcd for Br Br: 329.9442; found: 329.9443. Anal.calcd. for C 1 3 H 1 4 B r 2 : C 47.31, H 4.28, Br 48.42; found: C 47.10, H 4.30, Br 48.21. - 229 -122 Preparation of 7,7-Dibromo-3-oxabicyclo[4.1.0]heptane (140) Following the general procedure A for the preparation of dibromocyclopropanes a two phase mixture of 5,6-dihydro-2H-py-ran (111) (1.3g, 15.48 mmol), bromoform (1.8 mL, 20.64 mmol), triethylbenzylammonium chloride (^50mg, 0.22 mmol), 0.1 mL of 95% ethanol, 50% aqueous sodium hydroxide solution (7.8 mL) and 20 mL of methylene chloride was vigorously s t i r r e d at 40°C for 20h. The usual workup yielded a yellow o i l containing bromoform which was removed by heating at 40°C at 0.4 Torr. The residual o i l was d i s t i l l e d (air-bath temperature 65-70°C, 0.4 Torr; l i t . 1 2 2 bp 95°C, 4 Torr) to give 2.9g (73%) of a colourless o i l . This material was homogeneous by t i c analysis ( s i l i c a gel, 5% ether- petroleum ether) and exhibited i r ( f i l m ) ; 3000, 1460, 1435, 1375, 1345, 1250, 1135, 1115, 1080, 1035, 1015, 985, 855, 760, 750, 700 cm"1; % nmr (400 MHz) 6: 1.69-1.82 (m, 2H), 1.97-2.14 (m, 2H), 3.15 (d of d of d, 1H, J=5, 10, 11Hz, Hj,), 3.59 Cd of d of d, 1H, J=3.5, 6, 11Hz, H £), 3.90 (d of d, 1H, J=2, 12Hz, H c), 3.97 (d of d, 1H, J=6,12Hz, - 230 -Preparation of 9,9-Dlbromo-3-oxatricyclo[6.1.0.0 ' ]nonane (141) Following the general procedure A for the preparation of dibromocyclopropanes, a two phase mixture of the alkene (112) (4g, 36.4 mmol), bromoform (6.4 mL, 72.8 mmol), t r i e t h y l b e n z y l -ammonium chloride ( lOOmg, 0.44 mmol), 0.1 mL of 95% ethanol, 50% aqueous sodium hydroxide solution (18.2 mL) and 50 mL of methylene chloride was vigorously s t i r r e d at 40°C for 45h. The usual workup gave ^10g of crude material free of s t a r t i n g material, according to glc analysis, and containing bromoform and product. This material was subjected to f l a s h chromato-115 graphy (200g s i l i c a gel, 8% ethyl acetate- petroleum ether) to y i e l d 4.9g (48%) of the colourless, o i l y dibromocyclopropane (141). This compound was one component by t i c analysis ( s i l i c a gel, 10% ether- petroleum ether) and exhibited i r ( f i l m ) : 3000, 1440, 1345, 1310, 1290, 1260, 1230, 1170, 1070, 1060, 1025,980, 925, 900, 870, 850, 740 cm"1; 1H nmr (400 MHz) 6 : 1.53-1.65 (m, 1H, H Q), 1.71-1.82 (m, 1H, Hj), 1.91-2.04 (m, 1H, E^), 2.24 (d of d, 1H, J=10, 14.5Hz, H T), 2.33 ( t , 1H, J=7Hz, H ) , *J A 2.42 (d, 1H, J=7Hz, H R), 2.70-2.82 (m, 1H, H J , 3.70-3.88 (m, - 231 -2H, Hg., Hp), 4.40 Cd, IH, J=^7Hz, H c), Exact Mass calcd. for 7Q R1 C gH 1 0' i ?Br o- LBro : 281.9078; found: 281.9094. The above assignments of the protons in the *H nmr spec-trum of dibromocyclopropane (141) were secured by proton de-coupling experiments at 400 MHz. Preparation of trans- and cis-7,7-Dibromo-2-(trimethylsilyl) bicyclo[4.1.0]heptane (142a) and (142b). 142a 142b A mixture of 3-(trimethylsilyl)cyclohexene (110) (1.3g, 8.44 mmol) and potassium tert-butoxide (1.5g, 12.66 mmol) in 25 mL of dry pentane was cooled to -20°C under argon. To the well s t i r r e d , cooled suspension was slowly added, v i a syringe, a solution of bromoform (0.94 mL, 10.71 mmol) in 5 mL of dry 62 pentane . After completion of the addition, the reaction mixture was allowed to warm to room temperature and s t i r r e d for 3h. Water (20 mL) was added and the layers were separated. The aqueous layer was twice extracted with petroleum ether - 232 -(2x25 mL) and the combined organic layers were washed once with water (10 mL), once with brine (20 mL) and then dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure gave a l i q u i d residue (^3.6g) which was shown by glc analysis to consist of bromoform, s t a r t i n g materi-115 a l and product. Flash chromatography (180g s i l i c a g e l , petroleum ether) of t h i s material f a i l e d to separate bromoform and s t a r t i n g material from the product. However, heating the i s o l a t e d mixture at 40-50°C at 0.4 Torr removed the more vola-t i l e components and l e f t 1.5g (55%) of the colourless, l i q u i d dibromocyclopropanes (142) which appeared homogeneous by t i c analysis ( s i l i c a gel, petroleum ether). C a p i l l a r y glc analysis of t h i s l i q u i d showed that i t consisted of two components, thought to be the trans- and cis-isomers (142a) and (142b), in a r a t i o of 94:6, respectively. A s i m i l a r r a t i o (93:7) was i n -dicated by nmr analysis. The dibromide mixture exhibited i r ( f i l m ) : 2975, 1450, 1440, 1430, 1260, 1250, 1210, 1185, 1070, 1030, 975, 955, 875, 865, 840, 755, 735, 710, 690, 630 cm - 1; 1H nmr (80 MHz) 6: 0.11, 0.20 (two s i n g l e t s , 9H, due to me-t h y l s of trans-Si(CHg) 3 and c i s - S i ( C H 3 ) 3 , respectively),0.35-1.95 (m, 9H); ms m/e 328/326/324 (M +), 254/252/250 (M +-H-SiMe 3) and 73 (Me 3Si +, base peak). Exact Mass calcd. for C ^ H ^ 0 7 9 B r 8 1 B r 2 8 S i : 325.9524: found 325.9529. - 233 -General Procedure B: Reduction of Dibromocyclopropanes with T r i - n - b u t y l t i n Hydride 0*' x . To neat, w e l l - s t i r r e d dibromocyclopropane was added, 124 dropwise, t r i - n - b u t y l t i n hydride (jyl equiv.)- A cold water bath (^10°C) was employed to ensure that the reaction tempera-ture did not r i s e above 35°C°^. After completion of the addi-t i o n , the reaction mixture was s t i r r e d for about 20 min at room temperature. The products (monobromides) were isolated , in some cases, by dir e c t d i s t i l l a t i o n or by dir e c t chromato-graphy of the reaction mixture. However, in the majority of cases the i s o l a t i o n procedure was the following. An ethereal solution of the crude reaction mixture was s t i r r e d with aqueous 158 potassium f l u o r i d e at room temperature for 5-10 min. The dense, white s o l i d that p r e c i p i t a t e d was separated from the l i q u i d phase by suction f i l t r a t i o n and was washed well with ether. The ether layer, separated from the combined f i l t r a t e s , was washed once with brine and dried over anhydrous magnesium sulphate. After removal of the solvent, the residual l i q u i d was subjected to column chromatography, thus affording the diastereomeric (endo- and exo-) monobromides. - 234 -Reduction of the Dibromocyclopropane (93a). Br-94 -MEM=-CH2OCH2CH2OCH3 OMEM A solution of n-butyllithium (6.16 mL in hexane, 10.47 mmol) i n a mixture of dry tetrahydrofuran (18 mL) and dry ether (6 mL) under an atmosphere of nitrogen was cooled to -127°C (n-propanol - l i q . nitrogen) and a solution of 3.0g (8.72 mmol) of the dibromide (93a) in 1 mL of dry tetrahydro-furan was added, dropwise, over a period of 0.5h. After com-ple t i o n of the addition the reaction mixture was s t i r r e d for about 2 min at -127°C and was then quenched, at the same temperature, by the slow addition of a 1:1 mixture of methanol ether (2 mL). The reaction mixture was warmed to room temperature and the solvent was removed under reduced pressure. The residue was f i l t e r e d through a short column of F l o r i s i l (o,5g) and the column was eluted with petroleum ether. After the combined eluate had been dried over anhydrous magnesium sulphate, the solvent was removed to y i e l d , following b r i e f heating at 40-50°C at 0.4 Torr and pumping under vacuum for 58 'vlh, 1.6g (71%) of the monobromides (94). This material was of acceptable purity (^97%) by glc analysis and gave r i s e to two poorly resolved spots on t i c analysis ( s i l i c a gel, 25% ether-petroleum ether). Analysis of the 100 MHz nmr spec-trum of the above mixture indicated a r a t i o of exo- to endo-- 235 -monobromides (based on the integrated i n t e n s i t i e s of the H^ , doublets) of 1:1. This mixture which was u t i l i z e d as such in the following sequence, exhibited i r ( f i l m ) : 3000, 1445, 1360, 1290, 1250, 1200, 1170, 1155, 1130, 1105, 1050, 1010, 970, 920, 860, 850, 705 cm - 1, 1H nmr (100 MHz)6 : 1.12-2.32 (m, 6H), 2.44 ( t , ^ 0.5H, J=2Hz, -CHBr of the exo-monobromide), 3.38 (obscured t, -v0.5H, J=^7.5Hz, -CHBr of the endo-monobro-mide), 3.42 (s, 3H, -OCHg), 3.50-3.84 (m, 4H, -0CH 2CH 20-), 4.19, 4.27 (two doublets, 1H, J=5.5 and 5Hz,respectively, -CHOMEM of the endo- and exo-monobromides,resp.), 4.78, 4.79 (two p a r t i a l l y overlapping s i n g l e t s , 2H, -0CH o0-). Reduction of the Dibromocyclopropane (79a). The dibromocyclopropane (79a) (2.7g, 10 mmol) was re-duced with t r i - n - b u t y l t i n hydride (2.63 mL, 10 mmol) as describ-ed in the general procedure B. Analysis (by glc) of the re-action mixture at the end of the reaction indicated two pro-ducts i n a r a t i o of ^ 1.6:1 i n addition to the t r i - n - b u t y l t i n bromide byproduct. Careful d i s t i l l a t i o n (air-bath temperature - 236 -45-55°C, 0.4 Torr) of the reaction mixture yielded 1.76g (92%) of the monobromides (80a) and (80b) e s s e n t i a l l y free of t r i - n -b u t y l t i n bromide (^5%). Analysis of the 400 MHz "*"H nmr spec-trum of t h i s mixture conformed that the r a t i o of monobromides was 1.6:1. The monobromide mixture was conveniently u t i l i z e d as such i n succeeding reactions. However, the i n d i v i d u a l 115 monobromides could be obtained by f l a s h chromatography (180g s i l i c a gel, 10% ether- petroleum ether) of the mixture ( l g ) . The f i r s t eluted material on d i s t i l l a t i o n (air-bath temperature 35-40°C, 0.4 Torr) yielded 507 mg (47%) of the exo-monobromide (80b) which was one component on glc analysis and exhibited i r ( f i l m ) : 3025, 1455, 1435, 1365, 1340, 1290, 1280, 1195, 1160, 1100, 1055, 960, 920, 865, 825, 745, 705 cm"1; AH nmr (80 MHz)6 : 1.45-2.05 (m, 6H), 2.43 ( t , IH, J=^2Hz, -CHBr), 3.35 (s, 3H, -OCHg), 3.84 (d, IH, J=5Hz, -CHOMe). Exact Mass calcd. for C ^ ^ B r O : 189.9994; found: 189.9989. The l a t e r eluted f r a c t i o n yielded on d i s t i l l a t i o n 324 mg (30%) of the endo-monobromide (80a) which was one component on glc analysis and exhibited i r ( f i l m ) : 3000, 1460, 1450, 1430, 1370, 1350, 1260, 1210, 1200, 1110 1100,1060, 1030, 975, 930, 875, 810, 740 685 cm"1; 1H nmr (80 MHz)6: 1.60-2.35 (m, 6H), 3.35 (s, 3H, -OCHg), 3.39 ( p a r t i a l l y obscured t, IH, J=^7.5Hz, -CHBr), 3.75 (broad d, IH, J=5Hz, -CHOMe). Exact Mass calcd. for C,H., 7 9 BrO : 189.9994; found : 190.0000. - 237 -Reduction of the Dibrombcyclopropane mixture (85a-b) 86a-b 86c-d To a s t i r r e d solution of 8g (29.9 mmol) of a mixture of the dibromocyclopropanes (85a) and (85b) (/\,6 :1, respectively) and 10.24 mL (.179 mmol) of g l a c i a l acetic acid in 60 mL of ether was added, i n small portions over a period of 0.5h, 65 13.29g (203.3 mmol) of zinc dust. After the reaction mix-ture had been s t i r r e d for an additional 1.5h at 20°C, i t was poured onto 30 mL of water. The layers were separated and the aqueous layer was extracted twice with ether (40 mL). The combined organic layers were washed three times with satu-rated aqueous sodium bicarbonate (3x15 mL) then twice with brine and dried over anhydrous magnesium sulphate. Evaporation of the solvent under reduced pressure yielded a residue which on glc analysis was shown to consist of desired product (endo-: exo-monobromides ^9:1 by c a p i l l a r y glc analysis) as well as a more v o l a t i l e component. The l a t t e r was removed by heating the mixture at 55-60°C at 20 Torr and the residue was d i s t i l l e d (air-bath temperature ^24°C, 0.4 Torr) into a receiver cooled to -78°C to y i e l d 4g (71%) of the four monobromocyclopropanes (86a-d). C a p i l l a r y glc analysis indicated the mixture consist-ed nearly e n t i r e l y (>97%) of the endo- and exo-monobromides - 238 -(86a-b) and (86c-d) in a r a t i o of -v9:l. This mixture e x h i b i t -ed i r ( f i l m ) : 2950, 1445, 1430, 1260, 1245, 750 725 cm"1; -""H nmr (80 MHz) 6 : 0.60-2.13 (m, 9H), 1.13, 1.15 (two v i s i b l e doublets, 3H, J=6Hz each, -CHCHg, due to exo- and endo-mono-bromides ,resp. ), 2.59 ( t , i<0.1H, J=3.5Hz, -CHBr, due to exo-monobromides), 3.28 ( t , MD.9H, J=8Hz, -CHBr, due to endo-mono-bromides); ms m/e 190/188 (M +), 109 (M +-Br, base peak). Exact Mass calcd. for C^H-„Br: 188.0201; found: 188.0191. Reduction of the Dibromocyclopropane mixture (142a-b). 145a-d 145a The dibromocyclopropane mixture (142a-b) (2.8g, 8.59 mmol) was reduced with t r i - n - b u t y l t i n hydride (2.27 mL, 8.59 mmol) according to the general procedure B previously described. Analysis of the reaction mixture by glc at the end of the re-action indicated the presence of monobromides (which were poorly resolved) and t r i - n - b u t y l t i n bromide. Flash chroma-115 tography of t h i s mixture (180g s i l i c a gel, petroleum ether) yielded 1.5g (71%) of material which on c a p i l l a r y glc analysis - 239 -was shown to consist of four compounds in a r a t i o 72:21:5:2. Analysis of the "*"H nmr spectrum of the mixture gave a t o t a l endo- to exo-monobromides r a t i o of ^ 3.7:1. This mixture ex-hi b i t e d i r ( f i l m ) : 3010, 2960, 1440, 1260, 1250, 1230, 875, 870, 840, 750, 710, 690, 625 cm - 1; % nmr (80 MHz) 6 : 0.00 (s, 9H, - S i ( C H 3 ) 3 ) , 0.38-2.00 (diffuse multiplet, 9H), 2.60 ( t , 0.20H, J=3.5Hz, -CHBr of exp-monobromides), 3.34 ( t , 0. 8H, J=7Hz, -CHBr of endo-monobromides); ms m/e 248/246 (M +), 175/173 (M +-SiMe 3), 167 (M +-Br) and 73 (Me 3Si +, base peak). 7 9 28 Exact Mass calcd. for C 1 0 H 1 9 Br S i : 246.0439; found: 246.9438. Complete separation of the four monobromides was not possi-ble but afte r two f l a s h chromatographic runs on the material (1.5g) i s o l a t e d above, 700 mg of a mixture of a l l four mono-bromides containing ^88% ( c a p i l l a r y glc) of the major endo-bromide (145a) was obtained a f t e r d i s t i l l a t i o n (air-bath temp-erature 65°C, 0.4 Torr). An additional 600 mg of d i s t i l l e d material (air-bath temperature 65°C, 0.4 Torr) consisting of a ^1:1 mixture of endo- and exo-monobromides (145a-b) and (145c-d) was obtained from l a t e r eluted f r a c t i o n s . - 240 -Reduction Of the Dibromocyclopropane (138a) 146a 146b The dibromocyclopropane (138a) (3.6g, 10.17 mmol) was reduced with t r i - n - b u t y l t i n hydride (2.67 mL, 10.17 mmol) as described in the general procedure B. Analysis of the crude reaction mixture by glc showed, i n addition to the t r i - n - b u -t y l t i n bromide byproduct, two products in a r a t i o of %2.3:1. Treatment of t h i s crude mixture with aqueous potassium f l u o r i d e as described in the general procedure B, followed by the usual workup, yielded 2.5g of a mixture of the monobromides (146a) and (146b) which s t i l l contained some t r i - n - b u t y l t i n bromide. High pressure l i q u i d chromatography of t h i s material ( s i l i c a gel, 8% ether- petroleum ether, one run with one recycle) en-abled i s o l a t i o n of the i n d i v i d u a l monobromides. The more mobile exo-monobromide (146b) (700 mg, 25%) gave one peak on glc analysis and exhibited i r ( f i l m ) : 2975, 1450, 1440, 1380, 1350, 1260, 1220, 1200, 1130, 1120, 1075, 1035, 1020, 1000, 980, 910, 870, 815, 770 680 cm"1; "*"H nmr (80 MHz) 6: 0.75-2.10 (m, 14H), 2.53 ( t , 1H, J=3.8Hz, -CHBr), 3.25-4.18 (m, 3H, -OCH2-, -0CH-), 4.63-4.90 (m, 1H, -OCHO-); ms m/e 276/274 - 241 -(M +), 195 CM+-Br), 93 ("M+-C5Hg02H-Br ) and 85 (C 5H gO +, base 79 peak). Exact Mass calcd. for C 1 2 H i g Br0 2 : 274.0568; found: 274.0568. The more polar, endo-monobromide (146a) (1.2g, 43%) was homogeneous by t i c analysis and exhibited i r ( f i l m ) : 1420, 1335, 1240, 1185, 1120, 1100, 1060, 1045, 1025, 1010, 995,970, 900, 860, 805 cm - 1; -hi nmr (80 MHz) 6 : 1.10-2.10 (m, 14H), 3.30 (broad t, 1H, J=7Hz, -CHBr), 3.40-4.13 (m, 3H, -OCH-, -0CH 2-), 4.63-4.88 (m, 1H, -OCHO-): ms m/e 276/274 (M +), 195 (M +-Br), 93 (M +-C 5H g0 2H-Br) and 85 (C 5H gO +, base peak). 79 Exact Mass calcd. for C 1 9 H 1 Q Br0 9 : 274.0568;found:274.0572. Reduction of the Dibromocyclopropane (139) 147a 147b The dibromocyclopropane (139) ( l g , 3.03 mmol) was re-duced with t r i - n - b u t y l t i n hydride (0.8 mL, 3.03 mmol), as described i n the general procedure B. Analysis (by glc) of the reaction mixture at the end of the reaction indicated two products in a r a t i o of 3.4:1,in addition to t r i - n - b u t y l t i n - 242 -bromide, the l a t t e r was removed by s t i r r i n g an ether solution of the reaction mixture with aqueous potassium f l u o r i d e for ^5 min. The usual workup yielded ^0.7g of material which was 115 subjected to f l a s h chromatography (180g s i l i c a gel, petro-leum ether). The f i r s t eluted material, the endo-bromocyclo-propane C147a) (330 mg, 44%), was one component by glc analysis and exhibited i r ( f i l m ) : 3030, 3000, 1590, 1485, 1445, 1255, 790, 750, 740, 700, 635 cm - 1; "'"H nmr (100 MHz) 6 : 1.22-1.90 (m, 8H), 2.56-2.82 (broad m, 1H, W^19Hz, -CHPh), 3.37 ( t , 1H, J=8Hz, -CHBr), 7.20-7.50 (m, 5H aromatic protons); ms m/e 252/250 (M +), 171 (M +-Br) and 91 (C ?H 7). Exact Mass calcd. 79 for C,0H ' Br : 250.0357: found: 250.0347. 13 l o The l a t e r eluted exo-monobromide (147b) (100 mg, 13%) was one component on glc analysis and exhibited i r ( f i l m ) : 3060, 3035, 3000, 1595, 1490, 1445, 1220, 1010, 755, 705, 685 cm - 1; 1H nmr C100 MHz) 6 : 0.90-2.04 Cm, 8H), 2.76 ( t , 1H, J=3.5Hz, -CHBr), 2.75-3.02 ( p a r t i a l l y obscured m, 1H, -CHPh), 7.20-7.50 (m, 5H, aromatic protons); ms m/e 252/250 (M +), 171 (M +-Br) and 91 (C ?H 7). Exact Mass calcd. for C 1 3 H 1 5 ? 9 B r : 250.0357; found: 250.0359. A further 100 mg (13%) of a mixture of both monobromides (147a) and (147b) was also i s o l a t e d . - 243 -Reduction of the Dibromocyclopropane (140) 148a 148b The dibromocyclopropane (140) (2.0g, 7.8 mmol) was re-duced with t r i - n - b u t y l t i n hydride (2.06 mL, 7.8 mmol) as described in the general procedure B. Analysis (by glc) of the reaction mixture at the end of the reaction indicated two products in a r a t i o of ^ 2.2:1, in addition to the t r i - n - b u t y l -t i n bromide byproduct. Bulb to bulb d i s t i l l a t i o n of the re-action mixture (air-bath temperature 60°C, 0.4 Torr), gave 1.25g (91%) of the monobromide mixture containing traces of t r i - n - b u t y l t i n bromide. This material was subjected to f l a s h 115 chromatography (180g s i l i c a g e l , 20% ether - petroleum ether). The f i r s t eluted component on d i s t i l l a t i o n (air-bath temperature 45-50°C, 0.4 Torr, l i t . 1 2 2 bp 94°C, 30 Torr) gave 1 22 260 mg (19%) of the exo-monobromide (148b) which exhibited i r ( f i l m ) : 3000, 1460, 1435, 1380, 1350, 1295, 1245, 1235, 1200, 1125, 1105, 1080, 1065, 1000, 975, 945, 860, 780, 705 655 cm"1; 1H nmr (80 MHz) <S : 1.20-2.03 (m, 4H ), 2.84 ( t , IH, J=^3.5Hz, H R), 2.93-3.63 (m, 2H, H^ ,, H^,), 3.73 - 244 -(d of d, 1H, J=^3.5, 12Hz, H^), 4.05 (broad d, 1H, J=12Hz, H c). The l a t e r eluted component on d i s t i l l a t i o n (air-bath temperature 45-50°C, 0.4 Torr) yielded 560 mg (41%) of the endo-bromide (148a) which exhibited i r ( f i l m ) : 2995, 1455, 1430, 1380, 1350, 1260, 1250, 1130, 1110, 1085, 965, 950, 890, 840, 720, 630 cm - 1; 1H nmr (400 MHz) 6 : 1.15-1.23, 1.44-1.53 (two multiplets, 2H, Hg and H A , r e s p e c t i v e l y ) , 1.64-1.75, 1.93-2.03 (two multiplets, 2H, Hfi and H^, respectively),3.23 (d of d of d, 1H, J=^5.5, 9, 10.5Hz, Hj,), 3.32 ( t , 1H, J=7.0Hz, H R), 3.59 (d of d of d, 1H, J=4, 6, 10.5Hz, H £), 3.81 (d of d, 1H, J=^1.5, 12Hz, H c), 4.06 (d of d, J= 6, 12Hz, H^). Exact Mass 7 9 calcd. for C_H ' BrO : 175.9837; found: 175.9836. Reduction of the Dibromocyclopropane 1141) 149a 149b The dibromocyclopropane (141) (3.0g, 10.64 mmol), was reduced with t r i - n - b u t y l t i n hydride (.2.8 mL, 10.64 mmol) as - 245 -described i n the general procedure B. Analysis (by glc) of the reaction mixture at the end of the reaction indicated two products i n a r a t i o of ^ 45:55, i n addition to t r i - n - b u t y l t i n bromide. The l a t t e r was largely removed by s t i r r i n g an ether solution of the reaction mixture with aqueous potassium f l u o -ride for ^ 5 min. Following the usual workup, %1.5g of a mix-ture of monobromides (149a) and (149b) which s t i l l contained some t r i - n - b u t y l t i n bromide, was obtained. This mixture on t i c analysis (.15% ether- petroleum ether) gave e s s e n t i a l l y an elongated spot and was subjected to high pressure l i q u i d chromatography ( s i l i c a gel, 15% ether- petroleum ether, 1 re-cyc l e ) . The f i r s t eluted component the exo-monobromide (149b) (200 mg,9%) was 98% pure by glc analysis and exhibited i r ( f i l m ) : 3010, 1445, 1365, 1295, 1275, 1210, 1075, 1045, 1030, 980, 945, 900, 870, 850, 830, 710 cm - 1; 1H nmr (80 MHz)6: 1.40-2.40 (m, 7H), 2.30 ( t , IH, J=2Hz, CHBr), 3.86(two c o i n c i -dent d of d, 2H, J=5, 8Hz, -OCHg-), 4.39 (d, IH, J=5Hz, -OCH-) 79 Exact Mass calcd. for C g H 1 1 BrO : 201.9993; found;201.9995. The l a t e r eluted component, the endo-monobromide (149a) (400 mg, 19%) was ^ 99% pure by glc analysis and exhibited i r ( f i l m ) : 2980, 1440, 1350, 1325, 1250, 1235, 1180, 1070, 1045, 975, 935, 915, 890 cm"1; 1H nmr (80 MHz)6: 1.45-2.35 (m,6H), 2.50-2.93 (broad m, IH, H^), 3.37 ( t , IH, J=^8Hz, -CHBr), 3.85 (two coincident d of d, 2H, J=^5, 8Hz, -OCHg-), 4.33 (d, IH, J=6Hz, -OCH-). Exact Mass calcd. for C gH 1 ; L 7 9BrO : 201.9993; found: 201.9992. - 246 --A further 100 mg (5%) of a mixture of both monobromides, in which the exo-compound (149b) predominated, was obtained. Preparation of the Cyclopropyl Esters (96) Li. 95 Cfc€M OMEM -MEM=-CH2OCH2CH2OCR"3 To a cooled (-78°C) solution of 1.7g (6.42 mmol) of the monobromides (94) (endo- to exo- r a t i o ^1:1) in 10 mL of dry ether under an atmosphere of nitrogen was added 9.2 mL (11.6 mmol) of a pentane solution of t e r t - b u t y l l i t h i u m . S t i r r i n g was continued for 2h at -78°C. To the r e s u l t i n g solution of the lithiocyclopropanes (95) (a white p r e c i p i t a t e was also present) was added 2.5 mL (32.4 mmol) of methyl chloroformate. After a further 2h at -78°C, the reaction mixture was allowed to warm to room temperature and then was poured into cold water (^10 mL). The layers were separated and the organic layer was washed once with water, once with brine and dried over anhy-drous magnesium sulphate. Analysis (glc) of the crude materi-a l obtained a f t e r evaporation of the solvent , showed that i t - 247 -consisted of a mixture of the esters (96) (.endo- to exo-r a t i o VI.5:1) and a more v o l a t i l e component presumably a r i s -ing from protonation of the intermediate lithiocyclopropanes (95). Heating at 60°C at 0.4 Torr, to remove t h i s impurity, followed by d i s t i l l a t i o n of the residue (air-bath temperature 120°C, 0.4 Torr) yielded 900 mg (57%) of the esters (96) which were not resolved by t i c ( s i l i c a gel, 40% ethyl acetate- petro-leum ether). The mixture exhibited i r ( f i l m ) : 3030, 1725, 1435, 1265, 1195, 1170, 1105, 1065, 850 cm - 1; % nmr (100 MHz) : <$ 1.10-1.36 and 1.54-2.18 (two multiplets, 7H), 3.41 (s, 3H, -0CH 3), 3.50-3.82 (m, 4H, -OCHgCHgO-), 3.68 (s, 3H, -C0 2CH 3), 4.26, 4.36 (two doublets, IH, J=5Hz for each, -CHOMEM, exo-and endo-isomers ,resp.), 4.80 (s, 2H, -OCHgO-); m s m / e 244 (M +), 213 (M +-0CH 3), 139 (M+-0MEM), 89 ( C 4 H 9 0 2 + ) , 79 (M+-0MEM-C02Me), 59 (C 3H 70 +, base peak). Exact Mass calcd. for C 1 2 H 2 Q 0 5 : 244.1311; found : 244.1309. Preparation of the Cyclopropyl Esters (89) OMe OMe 89 81 - 248 -To a cooled (-78°C) solution of 800 mg (4.19 mmol) of a mixture of the monobromides (80b) and (80a) Cin a r a t i o of ^1.6:1,respectively) i n 8 mL of dry ether under a nitrogen atmosphere was added 3.51 mL (7.55 mmol) of a pentane solution of t e r t - b u t y l l i t h i u m . S t i r r i n g was continued for 2h at -78°C. To the r e s u l t i n g solution of the lithiocyclopropanes (81) was added 1.6 mL (20.94 mmol) of methyl chloroformate followed by 0.72 mL (4.19 mmol) of hexamethylphosphoramide. After being s t i r r e d f o r 4h at -78°C, the reaction mixture was poured into cold water (/vlO mL) and the layers were separated. The organ-i c layer was washed twice with half-saturated copper sulphate solution (2x5 mL), once with water and once with brine and then dried over anhydrous magnesium sulphate. Removal of the solvent and d i s t i l l a t i o n (air-bath temperature 60°C, 0.4 Torr) of the residue gave 500 mg (70%) of the esters (89) i n a r a t i o (glc) of ^ 1.7:1 (exo-:endo-). This material was 98% pure by glc analysis. An a n a l y t i c a l sample, obtained by preparative glc of a portion of the above material, exhibited i r ( f i l m ) ; 3020, 1715, 1455, 1435, 1405, 1355, 1330, 1310, 1280, 1265, 1190, 1170, 1145, 1100, 1060, 965, 925, 885, 870, 850, 805, 715 cm"1; XH nmr (80 MHz) 5: 1.18-1.35 and 1.50-2.22 (two multiplets, 7H), 3.35 (s, 3H, -OCHg), 3.68 (s, 3H, -COgCHg) 3.83 (d, -V0.64H, J=5Hz, -CHOMe of exo-ester), 3.95 (broad d, ^0.36H, J=5Hz, -CHOMe of endo-ester); ms m/e 170 (M +), 139 (M +-0CH 3), 111 (M+-C02Me, base peak). Exact Mass calcd. for C 9 H 1 4 0 3 : 170.0943; found: 170.0936. - 249 -Preparation of trans -2-Methylbicyclo[4.1.0]heptane-endo-7- carboxylic acid (151a) 87 151a 151b 151c-d To a cold (-78°C) solution of t e r t - b u t y l l i t h i u m (11.64 mL of a pentane solution, 23.3 mmol) in 30 mL of dry ether was added a solution of a mixture of the monobromides (86a-d) (2g, 10.58 mmol, endo-:exo-monobromides r a t i o of ^ 9:1) i n 5 mL of dry ether. The mixture was then s t i r r e d under an argon atmos-phere for 2h at -78°C. Into the r e s u l t i n g solution of the lithiocyclopropanes (87) was bubbled dry carbon dioxide for 0.5h at -78°C and for a further 1.5h as the reaction mixture was warmed to 0°C. A dense, white p r e c i p i t a t e was obvious at t h i s point. The reaction mixture was s t i r r e d for a further l h while i t was allowed to warm to room temperature. Water (^25 mL) was then added and, a f t e r the mixture was s t i r r e d for a few minutes, the layers were separated. The aqueous layer was c a r e f u l l y a c i d i f i e d by the dropwise addition of concentrated hydrochloric acid and then was extracted thoroughly with ether (3x50 mL). The combined ether extracts were washed with brine - 250 -and dried over anhydrous magnesium sulphate. The solvent was evaporated under reduced pressure to y i e l d a rather viscous, colourless l i q u i d . Analysis (by glc) of t h i s l i q u i d revealed, in addition to product, the presence of an impurity of appreci-ably shorter retention time. This impurity was removed by heat-ing the mixture at 120°C,30 Torr. There was thus obtained 1.2g (74%) of a mixture of the acids (151a-d) which on c a p i l l -ary glc analysis showed three peaks, the major one due to the desired acid (151a), making up 84% of the material. Two re-c r y s t a l l i z a t i o n s (petroleum ether) of the i s o l a t e d mixture of acids (151a-d) yielded 400 mg (25%) of the trans-methyl endo-acid (151a) as white c r y s t a l s (mp 57.5-58.5°C) which gave one peak on c a p i l l a r y glc analysis and exhibited i r (nujol mull): 2660, 1690, 1245, 1220, 1180, 975, 955, 935, 810 cm - 1; % nmr (270 MHz) : 0.77-0.95 (m, IH), 1.10 (d, 3H, J=7Hz, -CHCHg), 1.11-1.22 (m, IH), 1.35-1.84 (m, 8H); ms m/e 154 (M +), 153 ( M + - l ) , 139 (M +-CH 3), 136 (M +-H 20), 109 (M +-C0 2H). Exact  Mass calcd. for C 9H 1 40 2: 154.0994; found : 154.0994. Anal, calcd. for CQE14p2: C 70.1, H 9.15: found C 69.95, H 9.21. - 251 -Preparation of the Cyclopropyl Esters (100) Et02C OMe To a well s t i r r e d , heated (120°C) mixture of the o l e f i n (99) (2.5g, 17.9 mmol), and 150 mg of freshly prepared copper 157 bronze , in a 10 mL round-bottomed, three-necked f l a s k equipped with a reflux condenser and septum cap, was added, 69 v i a syringe pump, 7g (61.4 mmol) of ethyl diazoacetate over a period of ^4h. The reaction mixture was s t i r r e d at 120°C for 0.5h a f t e r completion of the addition. A solution of the reaction mixture in 150 mL of acetone was then s t i r r e d with excess potassium permanganate (^15g) for 0.75h to remove the ethyl maleate and fumarate byproducts. The r e s u l t i n g mixture, containing p r e c i p i t a t e d manganese dioxide and excess potassium permanganate as well as product, was poured c a r e f u l l y with s t i r r i n g and cooling into excess aqueous sodium b i s u l p h i t e . Thorough extraction of the colourless mixture thus obtained with ethyl acetate (3x100 mL), followed by drying of the com-bined extracts with magnesium sulphate and evaporation of the solvent gave a thick, o i l y residue. Analysis (by glc) of t h i s residue showed that i t consisted e s s e n t i a l l y of two components. - 252 -The more v o l a t i l e material was separated from the mixture by d i s t i l l a t i o n (air-bath temperature 90-100°C, 0.4 Torr) and the material thus obtained (^2.1g) was p u r i f i e d by f l a s h chromatography (^100g s i l i c a gel, 8% ethyl acetate - petroleum ether) to y i e l d 1.3g (33%) of a mixture of esters (100) in which the r a t i o of exo- to endo-isomers was 1.8:1 (based on integration of the methylene signals of the methoxymethyl group). This mixture gave one spot on t i c analysis and one peak on glc analysis. A d i s t i l l e d sample (air-bath tempera-ture 65°C, 0.4 Torr) exhibited i r Cfilm) : 1715, 1450, 1380, 1190, 1160, 1125, 1105, 1095, 1040 cm"1; 1H nmr (400 MHz) 6 : 1.03-2.05 (m, 9H), 1.22, 1.32 (s, s, 3H, t e r t i a r y methyls of endo- and exo-esters, resp.), 1.25, 1.26 ( t , t, 2H, J=7Hz for each, -OCHgCHg, due to the exo- and endo-esters, resp.), 3.31, 3.33 (s,s, 3H, -OCHg, due to the exo- and endo-esters,resp.), 3.34, 3.69 (two AB type d of d's, 2H, J=10Hz for each, -CHgOMe due to the endo- and exo-esters,resp.), 4.09, 4.12 ( p a r t i a l l y overlapped q's, 2H, J=7Hz for each, (-COgCHgCHg), due to exo-and endo-esters, resp.); ms m/e 226 (M +), 211 (M +-CH 3), 194 (M+-CH30H), 181 (M +-0CH 2H c, base peak). Exact Mass calcd. for C 1 C } H o o 0 „ : 226.1569; found: 226.1573. - 253 -Reduction of the Cyclopropyl Esters (9(3) H0H2C-97 OMEM -MEM=-CH2OCH2CH2OCH3 To a s t i r r e d , cooled (0°C) s l u r r y of lithium aluminum hydride (40 mg, 1.16 mmol) i n 5 mL of dry ether was added a solution of 379 mg (1.55 mmol) of the esters (96) (exo-:endo-r a t i o ^2:3) in 2 mL of dry ether. The reaction mixture was s t i r r e d for 2h during which time i t was allowed to warm to room temperature. Ether (^15 mL) was then added and the re-action mixture was quenched (at 0°C) by the slow addition of f i n e l y ground sodium sulphate decahydrate u n t i l gas (hydrogen) evolution ceased. S t i r r i n g for a further 10 min at room temp-erature resulted in a white s l u r r y which was f i l t e r e d ( e l uting with ether) through a column of F l o r i s i l (^10g). The solvent was evaporated from the eluate and the residue was d i s t i l l e d (air-bath temperature ^120°C, 0.4 Torr) to y i e l d 235 mg (70%) of a mixture (endo- to exo- r a t i o ^3:2) of the alcohols (97), greater than 99% pure by glc analysis. This material exhibited i r ( f i l m ) : 3420, 3010, 1450, 1360, 1200, 1175, 1155, 1120, 1105, 1045, 1025, 1000, 980, 920, 850 cm"1; 1H nmr (100 MHz) 6 : 0.68-2.20 (m, 8H), 3.42 (s, 3H, -OCHg), 3.47-3.84 (m, 6H, - 254 --OCHgCHgO, -CHgOH), 4.24 (d, IH, J= 5Hz, -CHOMEM), 4.80 (s, 2H, -OCH20-) ; ms m/e 215 (M + - l ) , 185 (M+-OCH3), 110 (M+-0MEM-H), 89 ( C 4 H g 0 2 + ) . Exact Mass calcd. for C--H-Q0. ( M + - l ) : 215.1283; found: 215.1276. Preparation of the Cyclopropyl Alcohols (90) H 0 H 2 C OMe To a s t i r r e d , cooled (0 C) s l u r r y of lithium aluminum hydride (117 mg, 3.09 mmol) in 10 mL of dry ether was added a solution of 700 mg (4.12 mmol) of the esters (89) (exo:endo r a t i o ^1.7:1) in 3 mL of dry ether. The reaction mixture was s t i r r e d for 2.5h at 0°C. Ether (20 mL) was then added and the reaction mixture was quenched (at 0°C) by the slow addition of f i n e l y ground sodium sulphate decahydrate u n t i l gas (hydrogen) evolution ceased. S t i r r i n g for a further 10 min at room temp-erature resulted in a white s l u r r y which was f i l t e r e d (eluting with ether) through a column of F l o r i s i l (^20g). The solvent was evaporated from the eluate and the residue was d i s t i l l e d (air-bath temperature V70°C, 0.4 Torr) to y i e l d 530 mg (91%) - 255 -of the alcohols J90). This mixture (exo : endo r a t i o %1.5:1 by nmr) was greater than 98% pure by glc analysis and exhibited i r ( f i l m ) : 3320, 1460, 1445, 1360, 1210, 1130, 1105, 1085, 1035, 1010, 990, 955, 930, 890, 875, 820, 755 cm"1; % nmr (80 MHz) 6 : 1.03-2.13 (m, 8H), 3.35 (s, 3H, -OCHg), 3.45 (d, 1.18H, J=7Hz, -CH20H due to exo-isomer), 3.61 (d, 0.82H, J=7Hz, -CHpOH due to endo-isomer), 3.79 (d, 1H, J=5Hz, -CHOMe); ms m/e 124 (M +-H 20), 111 (M +-0CH 3). Exact Mass calcd. for CgH^O (M +-H 90) : 124.0888; found:124.0872. Preparation of the Cyclopropyl Alcohols (103) H O H o C OMG To a s t i r r e d , cooled (0°C) s l u r r y of lithium aluminum hydride (140 mg, 3.69 mmol) in 10 mL of dry ether was added a solution of 0.86g(3.80 mmol) of the esters (100) in 2 mL of dry ether. The reaction mixture was s t i r r e d for 2h while be-ing allowed to warm to room temperature. Ether (20 mL) was then added and the reaction mixture was quenched (at 0°C) by the slow addition of f i n e l y ground sodium sulphate decahydrate - 256 -u n t i l gas (.hydrogen) evolution ceased. S t i r r i n g for a further 10 min at room temperature resulted in a white s l u r r y which was f i l t e r e d (eluting with ether) through a column of F l o r i s i l (^10g). Evaporation of the solvent from the combined eluate followed by d i s t i l l a t i o n of the residue (air-bath temperature ^80°C, 0.4 Torr) yielded 0.57g (81%) of the alcohols (103). These were poorly resolved on t i c analysis ( s i l i c a g el, 30% ethyl acetate - petroleum ether), gave r i s e to a singl e peak on glc analysis and exhibited i r ( f i l m ) : 3395, 1450, 1380, 1200, 1110, 1090, 1010 cm - 1; 1H nmr (400 MHz) 6: 0.72, 1.03-1.98 ( t , J=7Hz and m, 10H), 1.09, 1.17 (s, s, 3H, t e r t i a r y methyls of the two isomers), 3.31, 3.47 (AB type d of d, J=10Hz and s, 2H, -CH2OMe), 3.33, 3.36 (s, s, 3H, -OCHg), 3.51 and 3.74-3.86 (d of d, J=10, 12Hz and m, 2H, -CH20H). Exact Mass calcd. for C 1 1H 1gO(M +-H 20) : 166.1357; found:166.1351. Careful analysis of the r e s u l t s of proton decoupling en-abled c l a r i f i c a t i o n of some of the more complex signals of the spectrum, e s p e c i a l l y the multiplet at 63.74-3.86. Preparation of the Cyclopropyl Alcohol (153a) 153a 153b - 257 -To a well s t i r r e d , cooled (0°C) s l u r r y of lithium a l u -minum hydride (100 mg, 2.64 mmol) in 15 mL of dry ether was added a solution of 392 mg (2.55 mmol) of the acid (151a) in 5 mL of dry ether. The reaction mixture was s t i r r e d for 3h at 0°C and a further 2.5h at room temperature. Ether (^20 mL) was then added and the reaction was quenched (at 0°C), under an argon atmosphere, by the slow addition of cold water (^3 mL). After s t i r r i n g the mixture for a few minutes at room tempera-ture, the ether layer was separated from the aqueous sl u r r y of aluminum s a l t s . The l a t t e r was extracted twice with ether (%100 mL) and the combined ether layers were washed once with brine and dried over anhydrous magnesium sulphate. Removal of the solvent under reduced pressure, followed by d i s t i l l a t i o n of the residue (air-bath temperature ^60°C, 0.4 Torr) afforded 266 mg (75%) of a colourless o i l , 97% (by c a p i l l a r y glc analy-s i s ) of which was the alcohol (153a). Analysis of the 270 MHz proton nmr spectrum of t h i s o i l suggested that the minor com-ponent ^3% (by c a p i l l a r y glc analysis) was a diastereomer (153b) of alcohol (153a). The material i s o l a t e d exhibited i r ( f i l m ) : 3300, 2975, 1445, 1015, 735 cm - 1; % nmr (270 MHz) 6 : 0.67-0.77 (m, IH), 0.79-1.07 (m, 4H), 1.11 (d, 3H, J=7Hz, -CHCHg), 1.33-1.66, 1.69-1.86 (two multiplets,4H and IH, resp.), 3.64 (weak doublet, J=7Hz), 3.73 (d, n,2H, J=7Hz, -CHgOH); ms m/e 140 (M +), 122 (M +-H 20). Exact Mass calcd. for C g H 1 4 (M +-H 20): 122.1096; found: 122.1100. - 258 -Preparation of the Cyclopropyl Aldehydes (91) To a s t i r r e d s l u r r y of pyridinium chlorochromate - 1" 0" (0.73g, 3.38 mmol) and anhydrous sodium acetate (54 mg,0.63 mmol) in 8 mL of dry methylene chloride under a nitrogen at-mosphere was added a solution of 300 mg (.2.11 mmol) of the alcohols (90) (exo:endo r a t i o ^1.5:1) in 2 mL of dry methylene chloride. The reaction mixture, which turned dark almost immediately, was s t i r r e d for 2h at room temperature. Anhydrous ether (^30 mL) and anhydrous magnesium sulphate were added and, after being s t i r r e d for a few minutes, the mixture was f i l -tered through a column of F l o r i s i l (^15g) and the column was eluted with ether. The solvent was removed from the eluate and the residue was rapidly d i s t i l l e d (air-bath temperature ^60°C, 0.4 Torr) to y i e l d 229 mg (78%) of the aldehydes (91) which were ^98% pure on glc analysis and were u t i l i z e d d i r e c t -l y in the next reaction. The mixture exhibited i r ( f i l m ) : 3020, 2805, 2700, 1700, 1460, 1440, 1400, 1360, 1205, 1195, 1170, 1140, 1110, 1095, 1045, 1015, 990, 960, 940, 930, 885, 865, 835 cm"1; "'"H nmr (80 MHz) <5: 1.45-2.35 Cm, 7H), 3.35, - 259 -3.38 (two p a r t i a l l y overlapping s i n g l e t s , 3H, -OCHg), 3.86 (broad d, ^ 0.57H, J=5Hz, -CHOMe due to exo-isomer), 4.05 (broad d, 0.43H, J=5Hz, -CHOMe due to endo-isomer), 9.12,9.45 (two doublets, 1H, J=5.5Hz for each, -CHO). To a s t i r r e d s l u r r y of pyridinium chlorochromate •L'"5U (0.37g, 1.7 mmol) and anhydrous sodium acetate (41 mg, 0.51 mmol) i n 5 mL of dry methylene chloride under a nitrogen at-mosphere was added a solution of 229 mg (1.06 mmol) of the a l -cohols (97) (exo: endo r a t i o ^2.3) in 2 mL of dry methylene chloride. The reaction mixture, which turned dark almost immediately, was s t i r r e d for 2h at room temperature. Anhydrous ether (20 mL) and anhydrous magnesium sulphate (^lg) were added and a f t e r i t had been s t i r r e d for a few minutes, the mixture was f i l t e r e d through a column of F l o r i s i l (VlOg) and the column was eluted with ether. The solvent was removed from Preparation of the Cyclopropyl Aldehydes (98) OMEM - 260 -the combined eluate and the residue was rapidly d i s t i l l e d (air-bath temperature ^130°C, 0.4 Torr) to y i e l d 182 mg (79%) of the aldehydes (98). The mixture which exhibited the charac-t e r i s t i c peaks at 2790, 2700 and 1690 cm"1 in the in f r a r e d spectrum, was used d i r e c t l y in the following reaction (Wittig methylenation). Preparation of the Cyclopropyl Aldehydes (104) OHC-OMe To a s t i r r e d s l u r r y of pyridinium chlorochromate" L O U (0.46g, 2.13 mmol) and anhydrous sodium acetate (20 mg, 0.24 mmol) in 5 mL of dry methylene chloride under a nitrogen at-mosphere was added a solution of 0.26g (1.41 mmol) of the a l -cohols (103) in 2 mL of dry methylene chloride. The i n i t i a l l y orange s l u r r y turned dark almost immediately and was s t i r r e d for 2h at room temperature. Anhydrous ether (20 mL) and an-hydrous magnesium sulphate were then added and, after being s t i r r e d f o r a few minutes, the mixture was f i l t e r e d through a column of F l o r i s i l (^10g) and the column was eluted with ether. - 261 -Evaporation of the solvent from the eluate and d i s t i l l a t i o n of the residue (air-bath temperature ,v60°C, 0.4 Torr), gave 206 mg (80%) of the aldehydes (104) which were u t i l i z e d d i r e c t l y in the following reaction. A small sample of the aldehydes gave one peak on glc analysis and exhibited i r (film) : 2840, 2805, 2700, 1715, 1685, 1445, 1195, 1110 cm"1; XH nmr (100 MHz) 6 : 1.04-2.10 (m, 9H), 1.27, 1.44 ( s, s, 3H, t e r t i a r y methyls of the endo- and exo-aldehydes, resp.), 3.34, 3.36 (s, s, 3H, -OCHg , due to the endo- and exo-aldehydes, resp.), 3.35, 3.73 (two AB type d of d's, 2H, J=10Hz for each, -CHgOCHg, due to the endo- and exo-aldehydes, resp.), 9.60, 9.72 (d, d, 0.7H and 0.3H, resp., J=5.5 and 6.5Hz,resp., -CHO of exo- and endo-aldehydes, resp.) Preparation of the Cyclopropyl Aldehyde (154a) 154a 154b 130 To a s t i r r e d s l u r r y of pyridinium chlorochromate (0.55g, 2.57 mmol) and anhydrous sodium acetate (42 mg, 0.51 mmol) i n 10 mL of dry methylene chloride, under an argon at-- 262 -mosphere, was added a solution of 240 mg(1.71 mmol) of the alcohol (153a) (97% as described above) in 2 mL of dry methy-lene chloride. The reaction mixture turned dark almost immedi-ately and was s t i r r e d for 3h at room temperature. Anhydrous ether (30 mL) and anhydrous magnesium sulphate were added and, aft e r s t i r r i n g for a few minutes, the mixture was f i l t e r e d through a column of F l o r i s i l (VLOg) and the column was eluted with ether. Removal of the solvent from the eluate and rapid d i s t i l l a t i o n of the residue (air-bath temperature 45-55°C, 0.4 Torr) yielded 130 mg (55%) of the s e n s i t i v e aldehyde (154a) (96% by c a p i l l a r y glc) and i t s presumed diastereomer (154b). An appreciable quantity of viscous, o i l y l i q u i d residue was also obtained. The d i s t i l l e d material was u t i l i z e d immediately in the next reaction (Wittig methylenation). However, a small sample of the material exhibited i r ( f i l m ) : 1690, 1450, 1170, 955 cm - 1; """H nmr (80 MHz) 6 : 0.78-2.15 (m, 10H), 1.13, 1.14 (minor d, major d, J=6.5Hz for each, 3H, -CHCHg due to the minor and major aldehydes), 9.63 (d, IH, J=6.5Hz, -CHO). General Procedure C : Preparation of endo-Cyclopropyl Aldehydes: 72a N,N-Dimethylformamide Method To a cold (-78°C) solution of t e r t - b u t y l l i t h i u m (2.4 equiv. of a pentane solution) in 5-10 mL of dry ether was added an ethereal solution (2-5 mL) of the endo-cyclopropyl - 263 -bromide CI equiv,) . The mixture was s t i r r e d under an argon atmosphere for l-2h at -78°C. To the r e s u l t i n g solution of endo-lithiocyclopropane was added N,N-dimethylformamide C5 equiv.) and the reaction mixture was allowed to warm to room temperature with s t i r r i n g continued for a t o t a l of 5-7h. 72a Saturated aqueous ammonium chloride was then added and the mixture was s t i r r e d vigorously for 20 min. The layers were separated and the aqueous layer was twice extracted with ether. The combined ether layers were washed once with brine and dried over anhydrous magnesium sulphate. Evaporation of the solvent gave the desired endo-cyclopropyl aldehyde. The aldehydes obtained by the procedure just described were 90-98% pure by glc analysis and contained just one other component (2-10%), the product r e s u l t i n g from protonation of the intermediate l i t h i o d e rivative. These aldehydes were u t i l i z e d without further p u r i f i c a t i o n in the succeeding reaction (Wittig me-thylenation). Preparation of 3-Oxabicyclo[4.1.0]heptane-endo-7-carboxalde-hyde O60) Li OHC 'c 159 160 - 264 -Following the general procedure C outlined previously, a solution of the endo-lithiocyclopropane (159) was prepared by s t i r r i n g a solution of 200 mg (1.13 mmol) of the endo-mono-bromide (148a) and t e r t - b u t y l l i t h i u m (1.31 mL of a pentane solution, 2.49 mmol) in 10 mL of dry ether at -78°C for 40 min. A solution of 0.5 mL (6.5 mmol) of N,N-dimethylformamide i n 3 mL of dry ether was then added and the reaction mixture was allowed to warm to room temperature with s t i r r i n g continued for a t o t a l of 7.5h. The usual workup followed by rapid d i s -t i l l a t i o n (air-bath temperature 60°C, 0.4 Torr) of the crude product gave 118 mg (83%) of the aldehyde (160) as a colourless o i l which was ^98% pure by glc analysis. This compound exhi-b i t e d i r ( f i l m ) : 3000, 2830, 1685, 1445, 1400, 1380, 1355, 1310, 1240, 1215, 1140, 1105, 1075, 1040, 945, 910, 870, 840, 800, 745 cm"1; % nmr (100 MHz) 6 : 1.40-2.42 (m, 5 H), 3.20-3.52 (m, 1H), 3.83-4.01 (m, 1H), 3.98 (d of d, 1H, J=3.5,12Hz, H D), 4.26 (broad d, 1H, J=12Hz, H c), 9.90 (d, 1H, J=6.5Hz, -CHO); ms m/e 126 (M +), 125 (M+-H), 97 (M+-CH0). Exact Mass calcd. for C ? H 1 0 0 2 : 126.0682; found: 126.0682. - 265 -Preparation Of trans-2-Methyl-endo-7--(3-oxo-cyclopenten-l-yl) 71c-d 165 To a cooled (-78°C) solution of 500 mg (2.65 mmol) of a mixture of the monobromides (86a-b) and (86c-d) ( r a t i o 9:1, respectively) in 5 mL of dry ether, under a nitrogen atmosphere, was added 3.8 mL (4.8 mmol) of a pentane solution of t e r t -b u tyllithium. S t i r r i n g was continued for 2h at -78°C. To the re s u l t i n g solution of the lithiocyclopropanes (8J7) was added 15 mL of dry tetrahydrofuran followed by 470 mg (2.72 mmol) of 68 s o l i d phenylthiocopper and the reaction mixture was warmed to -20°C. S t i r r i n g for 0.5h at t h i s temperature resulted in a l i g h t brown solution of the cyclopropylcuprate reagents (88). This solution was cooled to -78°C and a solution of 550 mg - 266 -(2.64 mmol) of 3-iodo-2-cyclopenten-l-one (83) in 1 mL of dry tetrahydrofuran was added. The reaction mixture was warmed to -20°C and s t i r r e d at t h i s temperature for 2h a f t e r which i t was s t i r r e d at 0°C for a further 2h. Methanol (1.5 mL), ether (^25 mL) and magnesium sulphate (^2g) were added and the mixture was s t i r r e d at room temperature for 10 min. during which time the phenylthiocopper p r e c i p i t a t e d . After f i l t r a t i o n of the above mixture through a column of F l o r i s i l ('V/20g) and elut i o n of the column with ether (a.90 mL), the s o l -vent was removed from the combined eluate to y i e l d a crude s o l i d residue. This material when subjected to preparative t i c ( s i l i c a gel, 30% ether- petroleum ether) afforded two f r a c t i o n s . The less polar f r a c t i o n (330 mg, 66%) consisted of a s o l i d mixture of the endo-enones (71a-b) in which the r a t i o of trans-methyl to cis-methyl isomers was ^5:1 by both glc and nmr analyses. The more polar f r a c t i o n , on d i s -t i l l a t i o n (air-bath temperature ^130°C, 0.4 Torr), gave 35 mg of material which consisted of the exo-enones (71c-d) (80% by glc) as well as an impurity (^20% by glc) which appeared to be the g-phenylthioenone (165). An a n a l y t i c a l sample of the exo-enones (71c-d) was obtained as the f i r s t eluted f r a c t i o n on preparative glc of the mixture and exhibited uv (95% EtOH) 1 o v : 253 nm (e= 25,200); i r ( f i l m ) : 2975, 1690, 1660, 1590, 1430, 1330, 1285, 1170, 840 cm - 1; 1 H nmr (80 MHz) 6: 0.70-2.00 (m, 10H), 0.98, 1.10 (two doublets, 3H, J=6Hz for each, due to - 267 -the c i s - and the trans-methyl exo-enones, resp.), 2.20-2.58 (m, 4H, -CH 2CH 2C-), 5.70-5.85 Cm, 1H, -^-CH=C-). Exact Mass calcd. for C^H^gO : 190.1358; found: 190.1357. The trans-methyl endo-enone C71a) was obtained, free of the cis-methyl compound, as white c r y s t a l s (.200 mg, 40%) on r e c r y s t a l l i z a t i o n of the mixture of endo-enones C71a-b) from petroleum ether. These c r y s t a l s melted sharply C74°C), con-s i s t e d of one component Cglc) and exhibited uv C95% EtOH) Xmax : 242 nm Ce=14,000): i r Cnujol mull): 3075, 1700, 1680, 1615, 1280, 1245, 1200, 900, 850 cm"1; -^R nmr C100 MHz) :6 0.70-1.96 Cm, 10H), 1.11 C d i s t o r t e d d, 3H, J=6Hz, -CHCH„) Q 9 " 3 2.30-2.72 Cm, 4H, -CCH 2CH 2-), 5.94-6.04 Cm, 1H, -CCH=C-). Exact Mass calcd. for C 1 3 H l g 0 : 190.1358; found 190.1358. Anal, calcd. for C^H^O : C 82.06, H 9.53; found: C 81.80, H 9.67. Preparation of trans-2-Methoxy-endo-6-C3-oxocyclopenten-l-yl) bicyclo[3.1.0]hexane C65a) 81 R=-Li 65a 65b 82 R=-CuCSPh)Li - 268 -To a cooled C-78°C) solution of 241 mg (1.26 mmol) of a mixture of the monobromides (,80b) and (80a) (exo:endo = 1.6:1) in 5 mL of dry ether under nitrogen was added 1.2 mL (2.27 mmol) of a pentane solution of t e r t - b u t y l l i t h i u m . S t i r r i n g was continued for 2h at -78°C. To the r e s u l t i n g solution of the lithiocyclopropanes (81) was added 10 mL of dry tetrahydro-68 furan followed by 220 mg (1.28 mmol) of s o l i d phenylthiocopper and the reaction mixture was warmed to -20°C. S t i r r i n g for 20 min at t h i s temperature resulted in a s l i g h t l y yellow so-l u t i o n of the cyclopropylcuprate reagents (82). This solution was cooled to -78°C and a solution of 230 mg (1.10 mmol) of 58 3-iodo-2-cyclopenten-l-one (83) in 2 mL of dry tetrahydro-furan was added. The reaction mixture was warmed to -20°C and s t i r r e d at t h i s temperature for 2h after which i t was s t i r r e d at 0°C for a further 2h. The reaction was then quench-ed by the addition of methanol (1 mL). Ether (20 mL) was added and the mixture was s t i r r e d at room temperature for ^ 10 min during which time the phenylthiocopper p r e c i p i t a t e d . The e l u -ate obtained af t e r f i l t r a t i o n of t h i s mixture through a column of F l o r i s i l (^20g) and eluti o n of the column with ether (^80 mL) was dried over anhydrous magnesium sulphate. Removal of the solvent gave 160 mg of an o i l consisting of a mixture of endo- (65a) and exo-enones (65b)(in a r a t i o of 1:1.66 by glc analysis) as well as un i d e n t i f i e d byproducts. Preparative t i c ( s i l i c a g el, 40% ethyl acetate - petroleum ether) gave 40 mg (19%) of the less polar endo-enone (65a) ( s l i g h t l y impure by - 269 -glc analysis) and a f t e r d i s t i l l a t i o n (air-bath temperature 110-120°C, 0.4 Torr) 67 mg (32%) of the less mobile exo-enone (65b). Column chromatography(1.6g s i l i c a gel, 40% ethyl ace-tate - petroleum ether) of the impure endo-enone (65a) follow-ed by d i s t i l l a t i o n (air-bath temperature 110-120°C, 0.4 Torr) yielded 30 mg (14%) of the endo-enone (65a) which was one com-ponent by glc and t i c analyses and exhibited uv (95% EtoH) X : 237 (e=16,000); i r ( f i l m ) : 3020, 1700, 1675, 1610, 1460, 1440, 1410, 1370, 1270, 1240, 1210, 1175, 1110, 1095, 935, 885, 870, 850, 805 cm"1; 1H nmr (80 MHz) 6: 1.50-2.08 (m, 7H), 2.25-2.70 (m, 4H - 8cH0CH0-),3.35 (s, 3H, -0CH„), 9 3.75 (d, IH, J=5Hz, -CHOMe), 5.93-6.04 (m, IH, -CCH=C-). Exact Mass calcd. for C 1 2 E 1 6 ° 2 : 1 9 2 * 1 1 5 0 ' f o u n d : 192.1165. The exo-enone (65b) was also one component (by glc and t i c analyses) and exhibited uv (95% EtOH) X : 246 nm max (£=16,000): i r ( f i l m ) : 3020, 1690, 1665, 1595, 1460, 1435, 1405, 1350, 1285, 1200, 1180, 1160, 1100, 1055, 1020, 990, 925, 880, 875, 840, 830, 820, 790 cm _ 1 ; 1H nmr (80 MHz) 6: 1.10-9 2.15 (m, 7H), 2.20-2.55 (m, 4H, -CCH 0CH 0-), 3.31 (s, 3H, -0CH„) ~~z - z 0 ~ 3.83 (d, IH, J=5Hz, -CHOMe), 5.75-5.88 (m, IH, -CCH=C-). Exact Mass calcd. for C. 0H._0 o: 192.1150; found: 192.1159. - 270 -General Procedure D : Preparation of Vinylcyclopropanes To a s t i r r e d suspension of methyltriphenylphosphonium bromide (1.3 equiv.) in 5-10 mL of dry tetrahydrofuran under an argon atmosphere, was added a solution of n-butyllithium (1.3 equiv.) in hexane. S t i r r i n g was continued for 20 min at room temperature. The r e s u l t i n g yellow solution of methylene-triphenylphosphorane 1 3 1 (1.3 equiv.) was cooled to -78°C and a solution of one equiv. of the cyclopropyl aldehyde (or mix-ture of aldehydes) in 2-5 mL of dry tetrahydrofuran was added. The cooling bath was removed a f t e r a few minutes and the re-action mixture was s t i r r e d for 2-3h at room temperature. Petroleum ether (20-50 mL) was then added and the r e s u l t i n g suspension was f i l t e r e d through a column of F l o r i s i l (15-30g). The column was eluted with further quantities of petroleum ether (60-100 mL). Evaporation of the solvent from the com-bined eluate yielded a l i q u i d residue which was processed by chromatography and/or d i s t i l l a t i o n to y i e l d the desired, pure vinylcyclopropanes. - 271 -Preparation of the Vinylcyclopropanes (66a) and (66b) 66a 66b 131 A solution of methylenetriphenylphosphorane (4.30 mmol) in 10 mL of tetrahydrofuran was prepared, as described in gener-a l procedure D, from 1.54g(4.30 mmol) of methyltriphenylphos-phonium bromide and n-butyllithium (2.26 mL of a hexane solu-t i o n , 4.29 mmol) and cooled to -78°C. A solution containing 376 mg (2.69 mmol) of the aldehydes (91) (exo:endo r a t i o 1.3:1) in 3 mL of dry tetrahydrofuran was added and the reaction mix-ture was s t i r r e d for 2h at room temperature. The usual workup followed by d i s t i l l a t i o n (air-bath temperature rv30°C, 0.4 Torr, receiver cooled to -78°C) of the crude product yielded 350 mg (95%) of a mixture of the alkenes (66a) and (66b). Preparative t i c (12.5% s i l v e r n i t r a t e - s i l i c a gel, 10% ether - petroleum ether, two developments) of the mixture resulted in the i s o l a -t i o n of the in d i v i d u a l alkenes. The less polar f r a c t i o n on d i s t i l l a t i o n (air-bath temperature ^30°C, 0.4 Torr) gave 98 mg (27%) of the endo-alkene (66a), which was one peak by glc ana-l y s i s and exhibited i r (film) : 3050, 3005, 1620, 1445,1435, 1375, 1355, 1205, 1190, 1110, 1090, 990, 970, 930, 910, 875, - 272 -800 cm - 1; "^H nmr (400 (MHz) 6: 1.46-1.74 (m, 5H), 1.85 (d of d, IH, J = l l , 14Hz, H £), 2.02-2.13 (m, IH, H^), 3.32 (s, 3H, -OCHg), 3.70 (d, IH, J=6Hz, -CHOMe), 5.07 (d of d of d, IH, J=^1.5,v2, 10Hz, H y), 5.21 (d of d of d, IH, J=^1.5, V2.5, 17Hz, H z), 5.53 (d of d of d, IH, J=7, 10, 17Hz, H x); ms m/e 138 (M +), 123, (M +-CH 3), 106 (M+-CH30H). Exact Mass calcd. for C gH 1 40 : 138.1045; found: 138.1049. The more polar f r a c t i o n , on d i s t i l l a t i o n (air-bath temp-erature ^30°C, 0.4 Torr), yielded 147 mg (40%) of the exo-alk-ene (66b) which gave one peak on glc analysis and exhibited i r ( f i l m ) : 3050, 3000, 1630, 1455, 1435, 1380, 1350, 1280, 1200, 1190, 1145, 1100, 1090, 1080, 1050, 1025, 985, 960, 925, 900, 885, 860, 835, 800,750 cm - 1; 1H nmr (80 MHz)6: 0.95-2.15 (m, 7H), 3.35 (s, 3H, -OCHg), 3.83 (d, IH, J=5Hz, -CHOMe), 4.73-4.91, 4.85-5.13 (two p a r t i a l l y overlapping AB type d of d, 2H, J Y Z=2.5Hz, H z, H y), 5.20-5.70 (7 l i n e multiplet, IH, H x); ms m/e 138 (M +), 123 (M +-CH 3), 107 (M +-0CH 3), 106 (M+-CH30H). Exact Mass calcd. for CQH.,,,0 : 138.1045; found : 138.1051. - 273 -Preparation of the Vihylcyclopropanes (67a) and (67b) 67a t 67b OMEM OMEM MEM=-CH2OCH2 CH2OCH3 A solution of methylenetriphenylphosphorane 131 (1.10 mmol) in 5 mL of dry tetrahydrofuran was prepared, as describ-ed in general procedure D, from 0.394g ('vl.lO mmol) of methyl-triphenylphosphonium bromide and n-butyllithium (0.69 mL in aldehydes (98) (170 mg, 0.8 mmol) in 2 mL of dry tetrahydro-furan was added and the reaction mixture was s t i r r e d for 2h at room temperature. After the usual workup, there was obtain-ed 147 mg (87%) of a clear o i l which, according to glc analysis, consisted very largely (>97%) of the two alkenes (67a) and (67b),in a r a t i o of ^ 3:2. Preparative t i c (12.5% s i l v e r n i t r a t e - s i l i c a gel, 40% ether - petroleum ether, two develop-ments) of the mixture resulted in the i s o l a t i o n of the i n d i v i -dual alkenes. The less polar f r a c t i o n on d i s t i l l a t i o n ( a i r -bath temperature VL20°, 0.4 Torr) afforded 70 mg (42%) of the endo-alkene (67a), which appeared to be one component on glc analysis and exhibited i r (film) : 3060, 3005, 1627, 1455, 1365, 1205, 1185, 1160, 1130, 1100, 1050, 1020, 990, 925, 915, 875, hexane, 1.10 mmol) and cooled to -78°C. A solution of the - 274 -860, 800 cm"1; 1H nmr (100 MHz) 6: 1.46-1.88 (m, 7H), 3.41 (s, 3H, -OCH3), 3.50-3.84 (m, 4H, -OCH2CH20-), 4.16 (d, 1H, J=5Hz, -CHOMEM), 4.80 (s, 2H, -0CH 20-), 5.02-5.36 (m, 2H, H y, H z), 5.40-5.80 (7 l i n e multiplet, 1H, H x); ms m/e 182 (M +-CH 20), 107 (M+-0MEM), 106 (M+-0MEM-H), 89 ( C 4 H g 0 2 + ) . Exact Mass calcd. for C gH 1 : LCM +-C 4H g0 3) : 107.0860; found :107. 0880. The more polar component on d i s t i l l a t i o n (air-bath temp-erature 120°C, 0.4 Torr) yielded 30 mg (18%) of the exo-alk-ene (67b), which was one peak by glc analysis and exhibited i r (film) : 3065, 3010, 1635, 1455, 1360, 1200, 1180, 1135, 1110, 1095, 1060, 1040, 1025, 990, 900, 890, 865 cm"1; % nmr (80 MHz) <S : 1.00-1.97 (m, 7H), 3.41 (s, 3H, -OCHg), 3.48-3.83 (m, 4H, -0CH 2CH 20-), 4.24 (d, 1H, J=5Hz, -CHOMEM), 4.73-5.13 ( p a r t i a l l y obscured multiplet, 2H, Hy, H z), 4.80 (s, 2H, -0CH 20-), 5.18-5.65 ( 6 l i n e multiplet, 1H, H x); ms m/e 182 (M +-CH 20), 107 (M +- OMEM), 106 (M+-OMEM-H) , 89 ( C 4 H g 0 2 + ) . Exact Mass calcd. for CgH^ (M +-C 4H g0 3) : 107. 0860; found : 107.0864. - 275 -Preparation of the Vinylcyclopropane (72a) 72 a 72b A solution of methylenetriphenylphosphorane 131 (1.17 mmol) in 5 mL of dry tetrahydrofuran was prepared, as describ-ed in general procedure D, from 0.32g (1.17 mmol) of methyl-triphenylphosphonium bromide and n-butyllithium (0.72 mL of a hexane solution, 1.17 mmol) and cooled to -78°C. A solu-t i o n of 124 mg (0.90 mmol) of the aldehyde (154a) (obtained as previously described) in 2 mL of dry tetrahydrofuran was added and the reaction mixture was s t i r r e d for 3h at room temperature. The usual workup was followed by removal of the solvent by careful d i s t i l l a t i o n at atmospheric pressure. The residue obtained was d i s t i l l e d (air-bath temperature ^45°C, 20 Torr) to y i e l d 50 mg (41%) of a colourless, v o l a t i l e l i q u i d . C a p i l l a r y glc analysis showed that t h i s material was 96% alkene (72a) and 4% of i t s diastereomer (72b). The i s o l a t e d material exhibited i r ( f i l m ) ; 3050, 2970, 1620, 1445, 1185, 980, 900, 790, 735 cm"1; "hi nmr (400 MHz) 6: 0.79-0.96 (m, 2H),0.99-1.16 ( p a r t i a l l y obscured m, 2H), 1.10 (d, 3H, J=6.5Hz, -CHCH3), 1.39-1.59,1.60-1.69, 1.72-1.84 (three multiplets, 4H, - 276 -1H, IH), 5.07 Cd of d, IH, J=v2.5, 10Hz, Hy), 5.19 (d of d, IH, J=^2.5, 17Hz, H z), 5.68 (d of t, J=10, 17Hz, tt^): ms m/e 136 (M +), 121 (M +-CH 3), 107 (M +-C 2H 5), 93 CM +-C 3H ?), 79 (M +-C 4H g). Exact Mass calcd. for C ^ H ^ : 136.1252; found: 136.1242. Preparation of the Vinylcyclopropanes (68a) and (68b) OMe OMe 68a 68b 131 A solution of methylenetriphenylphosphorane (1.65 mmol) in 5 mL of dry tetrahydrofuran was prepared,as described in general procedure D,from 0.59g (1.65 mmol) of methyltri-phenylphosphonium bromide and n-butyllithium (1.08 mL of a hexane solution, 1.65 mmol) and cooled to -78°C. A solution of 200 mg (1.1 mmol) of the aldehydes (104), obtained as pre-viously described, i n 3 mL of dry tetrahydrofuran was added and the reaction mixture was s t i r r e d for 2h at room temperature, The usual workup followed by d i s t i l l a t i o n (air-bath temperature ^60°C, 0.4 Torr) of the crude product yielded 170 mg (86%) of a mixture of the alkenes (68a) and (68b). Preparative t i c - 277 -(12.5% s i l v e r n i t r a t e - s i l i c a g e l , 20% ether - petroleum ether, two developments) of the mixture enabled i s o l a t i o n of the i n d i v i d u a l alkenes. The less polar f r a c t i o n on d i s t i l l a -t i o n (air-bath temperature 45-50°C, 0.4 Torr) gave 30 mg (15%) of the endo-alkene (68a), which gave one peak on glc analy-s i s and exhibited i r ( f i l m ) : 3060, 1630, 1450, 1385, 1380, 1200, 1120, 1110, 1095, 900 cm"1; 1R nmr (80 MHz) 6: 1.03-2.10 (m, 9H), 1.17 (s, 3H, t e r t i a r y methyl), 3.31 (s, 3H, -0CH 3), 3.33 ( p a r t i a l l y obscured AB type d of d, 2H, J=10Hz, -CH20Me), 4.90-5.09 (m, IH, H y), 5.18 (AB type d of d, IH, J=^3Hz, H z), 5.53-6.05 (m, IH, H x); ms m/e 180 (M +), 165 (M +-CH 3), 149 (M +-0CH 3), 135 (M +-0C 2H 5). Exact Mass calcd. for C 1 2H 2 Q0 : 180.1514; found: 180.1517. The more polar f r a c t i o n on d i s t i l l a t i o n (air-bath temp-erature 45-50°C, 0.4 Torr) yielded 65 mg (33%) of the exo-alkene (68b). This material gave one peak on glc analysis and exhibited i r ( f i l m ) : 3055, 1620, 1450, 1375, 1195, 1110, 1095, 900 cm - 1; 1H nmr (80 MHz) 6: 1.13 (s, 3H, t e r t i a r y methyl), 1.10-2.10 (m, 9H), 3.31 (s, 3H, -OCHg), 3.39 (s, 2H, -CHgOMe) 4.96 (AB type d of d, IH, J~3Hz, H y), 5.51 (AB type d of d, IH, J=3Hz, H z), 5.69 (d of t, IH, J=10, 17Hz, H x). ms m/e 180 (M +), 165 (M +-CH 3), 149 (M +-0CH 3), 135 (M +-0C 2H 5, base peak). Exact Mass calcd.for C^2 H20° • 180.1514; found : 180.1516. - 278 -Preparation of trans-2-(2-Tetrahydropyranyloxy)-endo-7-vinyl-bicycloT4.1.0] heptane (73) _ Following the general procedure C outlined previously, a solution of the endo-lithiocyclopropane (155) was prepared by s t i r r i n g 1.Og (3.64 mmol) of the monobromide (146a) and te r t - b u t y l l i t h i u m (4.3 mL in pentane, 8.74 mmol) in 15 mL of dry ether at -78°C for l h . N,N-Dimethylformamide (1.41 mL, 18.20 mmol) was then added and the reaction mixture was warmed to room temperature and s t i r r e d for 5h. The usual workup afforded 654 mg of the crude aldehyde (156). This material, which exhibited a strong 1690 cm - 1 band as well as bands at 2840 and 2700 cm - 1 in the i n f r a r e d spectrum, was used d i r e c t -l y in the following methylenation reaction. A solution of methylenetriphenylphosphorane (3.77 mmol) in 10 mL of dry tetrahydrofuran was prepared, as described in general procedure D, from 1.35g (3.77 mmol) of methyltriphenyl-phosphonium bromide and n-butyllithium (2.31 mL in hexane, 3.77 mmol) and cooled to -78°C. A solution of the crude alde-hyde (156)(650 mg, 2.90 mmol)j obtained as described above, i n •H 155 156 73 - 279 -5 mL of dry tetrahydrofuran was then added and the reaction mixture was s t i r r e d for 3h at room temperature. After the usual workup, 560 mg of the alkene (7_3) was obtained. It proved convenient to convert t h i s material into the hydroxy alkene (74) before f i n a l p u r i f i c a t i o n . However, preparative t i c ( s i l i c a gel, 5% ether - petroleum ether) of a small amount (V20 mg) of the crude material followed by d i s t i l l a t i o n ( a i r -bath temperature 100°C, 0.85 Torr) gave a pure sample of the THP ether (73), which exhibited i r ( f i l m ) : 3050, 1620, 1440, 1350, 1200, 1130, 1115, 1080, 1035, 1020, 1000, 905, 870, 820 cm - 1; -^H nmr (80 MHz) <5: 0.88-2.00 (m, 15H), 3.33-4.13 (m, 3H, -0CH2-, -0CH-), 4.63-4.88 (m, IH, -0CH0-), 5.00-5.93 (m, 3H, o l e f i n i c protons), Exact Mass calcd. for C^4 H22°2' 222.1620; found : 222.1618. Preparation of trans-2-Hydroxy-endo-7-vinylbicyclof4.1.0]hep-tane (74) To a solution of the crude alkene (73) obtained as des-cribed above (540 mg, 2.43 mmol) in 25 mL of ethanol, was 80 added 61 mg (0.24 mmol) of pyridinium p_-toluenesulphonate - 280 -The mixture was s t i r r e d at 50°C Coil bath) for 2h at which point t i c analysis ( s i l i c a gel, 20% ethyl acetate - petroleum ether) indicated that the reaction was complete. The ethanol was evaporated under reduced pressure and the residue was d i -luted with ether (V15 mL) and f i l t e r e d through a short column of F l o r i s i l (^5g). The column was eluted with a further 60 mL of ether and the combined eluate was dried over anhydrous mag-nesium sulphate. The residue obtained on evaporation of the 115 solvent was subjected to f l a s h chromatography ( s i l i c a gel, 20% ethyl acetate - petroleum ether) to afford on d i s t i l l a t i o n (air-bath temperature 60°C, 0.85 Torr) 200 mg (40% based on monobromide (146a)) of the hydroxy alkene (74). This material was %99% pure by glc analysis and exhibited i r ( f i l m ) : 3300, 3050, 2980, 1620, 1435, 1360, 1065, 1055, 990, 980, 900 cm - 1; •^H nmr (80 MHz) 6: 0.95-1.95 (m, 10H), 3.72-4.00 (m, 1H,-CH0H), 5.00-5.90 (m, 3H, o l e f i n i c protons); ms m/e 138 (M +), 137 (M+-H), 120 (M +-H 20). Exact Mass calcd. for CgH^O : 138.1044; found: 138.1048. - 281 -Preparation of trans-2-Methoxy-endo-7-vinylbicyclo[4.1.ol heptane (75) To a s t i r r e d s l u r r y of o i l - f r e e sodium hydride (40 mg, 1.67 mmol) in 10 mL of anhydrous tetrahydrofuran under an argon atmosphere was slowly added a solution of 100 mg (0.72 mmol) of the hydroxy alkene (74) in 2 mL of dry tetrahydro-furan. The mixture was refluxed for l h , then cooled to room temperature, and 0.2 mL (3.2 mmol) of methyl iodide was added. After the r e s u l t i n g solution had been refluxed for 4h, t i c analysis ( s i l i c a gel, 30% ether - petroleum ether) of an a l i -quot showed that the reaction was complete. The reaction mix-ture was cooled to 0°C and treated with saturated aqueous ammonium chloride (5 mL). The layers were separated and the aqueous layer was extracted twice with ether. The combined ether layers were then washed with brine and dried over anhy-drous magnesium sulphate. Careful removal of the solvent under reduced pressure, followed by d i s t i l l a t i o n of the residue ( a i r -bath temperature 40°C, 0.4 Torr) into a well cooled (-78°C) receiver afforded 80 mg (73%) of the methoxy alkene (75), which was one component by glc analysis. This material exhibited i r - 282 -(f i l m ) ; 3050, 2975, 1620, 1460, 1450, 1370, 1145, 1100, 1090, 990, 900 cm"1; 1H nmr (270 MHz) <5 : 0.99-1.14, 1.19-1.41, 1.47-1.70, 1.74-1.90 (multiplets, 9H), 3.21-3.29 (m, IH, -CHOMe), 3.38 (s, 3H, -OCH3 ), 5.09 (d of d of d, IH, J=vl.5, 2.5,10Hz, H v), 5.22 (d of d of d, IH, J=VL.5, 2.5, 17Hz, H z), 5.58 (d of d of d, IH, J=^9.5, 10, 17Hz, H x); ms m/e 152 (M +) 137 (M +-CH 3), 121 (M +-0CH 3). Exact Mass calcd. for C 1 0H l gO : 152.1201 ; found: 152.1223. Preparation of trans-2-Phenyl-endo-7-vinylbicyclo[4.1.0]heptane 76 Following the general procedure D outlined previously, a solution of the lithiocyclopropane (157) was prepared by s t i r r i n g 246 mg (0.98 mmol) of the endo-monobromide (147a) and t e r t - b u t y l l i t h i u m (1.18 mL of a pentane solution, 2.36 mmol) in 8 mL of dry ether at -78°C for l h . N,N-dimethylforma-mide (0.38 mL, 4.90 mmol) was then added and the reaction mix-- 283 -ture was warmed to room temperature with s t i r r i n g continued for a t o t a l of ^7h. The usual workup yielded 133mg of the aldehyde (158) which was 94% pure by glc analysis. This material, which was u t i l i z e d without further p u r i f i c a t i o n i n the next reaction, exhibited i r ( f i l m ) : 3020, 2850, 2725, 1675, 1440, 1375, 1145, 1085, 965, 755, 740, 700 cm"1; XE nmr (400 MHz) 6: 1.43-1.58 (m, 2H), 1.75-1.92 (m, 5H), 2.02-2.11 (m, 2H), 3.01-3.11 (m 1H, Wj=20Hz, H c), 7.19-7.38 (m, 5H, aromatic protons), 9.84 (d, 1H, J=5.5Hz, -CHO).* 131 A solution of methylenetriphenylphosphorane (0.86 mmol) was prepared, as described in general procedure D, from 0.31g (0.86 mmol) of methylenetriphenylphosphonium bromide and n-butyllithium (0.53 mL of a hexane solution, 0.86 mmol) in 5 mL of dry tetrahydrofuran and cooled to -78°C. A solution of the aldehyde (158), obtained as described above, (128 mg, 0.64 mmol) in 3 mL of dry tetrahydrofuran was then added and the reaction mixture was s t i r r e d for ^3h at room temperature. The material i s o l a t e d a f t e r the usual workup was subjected to preparative t i c (12.5% s i l v e r n i t r a t e - s i l i c a gel, 5% ether -petroleum ether, two developments) to y i e l d , a f t e r d i s t i l l a t i o n (air-bath temperature 75-80°C, 0.4 Torr) of the crude i s o l a t e d A weak doublet (J=5.5Hz) at 69.03 hinted at the presence of a few percent of the exo-isomer of (158). - 284 -material, 80 mg (41% based on s t a r t i n g monobromide (147a)) of the vinylcyclopropane (76), which exhibited one component on c a p i l l a r y g l c analysis. This material exhibited i r ( f i l m ) ; 3050, 3000, 1620, 1595, 1490, 1450, 1175, 990, 900, 800, 755, 735, 700 cm"1; 1H nmr (270 MHz) 6: 1.07-1.45 (m, 4H),1.51-2.00 (m, 5H), 2.61 (d of t, IH, J=^3, 10.5Hz, -CHPh), 5.13 (d of d of d, IH, M1.5A2.5, 10Hz, H y), 5.26 (d of d of d, IH, J= VI. 5, ^2.5, 16.5Hz, H z), 5.82 (d of d of d, IH, J=10, 11, 16.5Hz, H x), 7.14-7.26 (m, IH, aromatic proton), 7.29-7.35 (m, 4H, aromatic protons); ms m/e 198 (M ), 91 (C 7H 7), 77 (CgH 5). Exact Mass calcd. for C 1 5 H l g : 198.1408; found: 198.1408. Preparation of endo-7-Vinyl-3-oxabicyclo[4.1.0]heptane (69) A solution of methylenetriphenylphosphorane (0.98 mmol) prepared as previously described from methyltriphenylphospho-nium bromide (0.35g, 0.98 mmol) and n-butyllithium (0.61 mL of a hexane solution, 0.98 mmol) in 5 mL of dry tetrahydrofuran was cooled to -78°C. A solution of the aldehyde (160) (95 mg, - 285 -0.75 mmol) in 2 mL of dry tetrahydrofuran was added and the reaction mixture was s t i r r e d for 3h at room temperature. Workup, as previously described, followed by d i s t i l l a t i o n (air-bath temperature 60°C, 15 Torr) of the crude product, gave 66 mg (71%) of the vinylcyclopropane (69) as a colourless l i q u i d . This compound gave one peak on glc analysis and ex-hi b i t e d i r ( f i l m ) ; 3055, 2985, 1625, 1460, 1440, 1380, 1355,. 1295, 1245, 1165, 1130, 1110, 1085, 1055, 1005, 945, 915, 905, 865, 790, 750, 710, 680 cm - 1; 1H nmr (80 MHz)6 : 0.75-2.00 (m, 5H), 2.95-3.35, 3.55-3.85 ( two multiplets, 2H, H £, Hj.), 3.85-4.10 (m, 2H, H c, Hp), 5.00-5.15, 5.15-5.38 (two AB type doub-l e t of doublets, 2H, J=^2.5Hz, H y, H z), 5.75-6.30 (6 l i n e m, 1H, H x). Exact Mass calcd. for CgH^O : 124.0888; found: 124.0890. Preparation of the Vinylcyclopropane (70) 163 164 70 Following the general procedure C outlined previously, a solution of the lithiocyclopropane (163) was prepared by s t i r r i n g a solution of 240 mg (1.18 mmol) of the endo-monobro-- 286 -mide (149a ) and t e r t - b u t y l l i t h i u m (1.42 mL of a pentane solu-t i o n , 2.84 mmol) in 9mL of dry ether at -78°C for 1 h. Excess N,N-dimethylformamide (0.46 mL, 5.94 mmol) was then added and the reaction mixture was allowed to warm to room temperature with s t i r r i n g continued for a t o t a l of -v7h. The usual workup yielded 127 mg of the aldehyde (164) which was 94% pure by glc analysis. This material was u t i l i z e d d i r e c t l y without further p u r i f i c a t i o n i n the next reaction as described below. However, the following diagnostic spectral features were ex-hi b i t e d : i r ( f i l m ) : 2840, 2740, 1685 cm"1; XE nmr (80 MHz) 6: 9.43 (d, J^e.SHz, -CHO). A solution of methylenetriphenylphosphorane (1.10 mmol)was prepared, as described i n general procedure D, from me t h y l t r i -phenylphosphonium bromide (0.39g, 1.10 mmol) and n-butyllithium (0.67 mL of a hexane solution, 1.10 mmol) in 5 mL of dry tetrahydrofuran, and cooled to -78°C. A solution of the alde-hyde (164), obtained as described above, (120 mg, 0.79 mmol), in 5 mL of dry tetrahydrofuran was then added and the reaction mixture was s t i r r e d for ^3h at room temperature. Workup, as previously described, followed by d i s t i l l a t i o n (air-bath temp-erature 40°C, 0.4 Torr) yielded 83 mg (47% based on s t a r t i n g monobromide (149a)) of the vinylcyclopropane (70) as a colour-less o i l which was ^99% pure by glc analysis. This material exhibited i r ( f i l m ) : 3050, 3000, 1620, 1440, 1085, 1065, 1035, 975, 915, 890, 790 cm"1; 1H nmr (270 MHz) 6: 1.50-1.79 (m, 5H), - 287 -1.94-2.10 (m, 2H), 2.36-2.50 (m, IH, E^), 3.73-3.88 (m, 2H, H D, H E), 4.27 (d, IH, J=6.5Hz, H c) 5.09 (d of d of d, 1H, J= VI.5, V2.5, 10Hz, H y), 5.23 (d of d of d, IH, J=^1.5, ^ 2.5, 16Hz, H z), 5.45 (d of d of d, IH, J=8, 10, 16Hz, H x); ms m/e 194 (M +), 179 (M +-CH 3), 121 (M +-SiMe 3), 73 CMe 3Si +, base peak). Exact Mass calcd. for : 150.1045; found: 150.1044. Preparation of trans-2-(Trimethylsilyl)-endo-7-vinylbicyclo [4.1.0]heptane (77a) 161 162 77a Following the general procedure C outlined previously, a solution of the lithiocyclopropanes (161) was prepared by s t i r r i n g a solution of 680 mg (2.75 mmol) of the endo- enriched monobromide mixture (145a-d) (containing 88% of endo-monobro-mide (145a)) and t e r t - b u t y l l i t h i u m (3.30 mL of a pentane solu-t i o n , 6.60 mmol) i n 15 mL of dry ether at -78°C f o r 2h. Excess N,N-dimethylformamide (1.07, 13.75 mmol) was then added and the reaction mixture was allowed to warm to room temperature - 288 -with s t i r r i n g continued for a t o t a l of 7h. The usual workup yielded 390 mg of crude aldehydes (162) which was u t i l i z e d d i r e c t l y without p u r i f i c a t i o n in the next reaction as describ-ed below. However, the crude material exhibited the following key spectral features: i r ( f i l m ) : 2820, 2675, 1680, 1245, 1145, 850, 830, 745 cm"1; 1H nmr (80 MHz) <S: 0.00, 0.02 (poorly r e-solved s i n g l e t s , 9H, - S i ( C H 3 ) 3 ) , 0.70-2.18 (dif f u s e m, 10H), 9.71 (d, 1H, J=6Hz, -CHO). A solution of methylenetriphenylphosphorane (2.5 mmol) was prepared, as described i n general procedure D, from methyl-triphenylyphosphonium bromide (0.89g, 2.5 mmol) and n-butyl-lithium (1.53 mL of a hexane solution, 2.5 mmol) in 10 mL of dry tetrahydrofuran, and cooled to -78°C. A solution of the crude aldehydes (162) (377 mg, 1.92 mmol) obtained as described above in 5 mL of dry tetrahydrofuran was then added and the reaction mixture was s t i r r e d for 3h at room temperature. Workup, as previously described, and d i s t i l l a t i o n (air-bath temperature ^120°C, 30 Torr) gave 260 mg of a mixture of the alkenes (77a-d) which was ^ 90% pure by glc analysis. Prepara-t i v e layer chromatography on s i l v e r n i t r a t e impregnated s i l i c a 132 gel (12.5% s i l v e r n i t r a t e - s i l i c a g e l , 2% ether - petroleum ether, 2 developments) separated t h i s mixture into two f r a c -tions. The less polar f r a c t i o n gave afte r d i s t i l l a t i o n ( a i r -bath temperature VL10°C, 30 Torr) 190 mg (35% based on s t a r t i n g monobromides (145 )) of material which was, by c a p i l l a r y glc - 289 -analysis, 95% trans-2-(trimethylsilyl)-endo-7-vinylbicyclo [4.1.0]heptane (77a). The remaining ^5% consisted of a mix-ture of the other isomers of t h i s compound. The material ex-h i b i t e d i r ( f i l m ) ; 3050, 1620, 1455, 1440, 1255, 1245, 1190, 985, 900, 870, 860, 835, 790, 745, 730, 690 cm"1; "Si nmr (400 MHz) 6 : 0.02 (s, 9H, - S i ( C H 3 ) 3 ) , 0.51-0.59 (m, IH, -CHSiMe 3), 0.96-1.11 (m, 4H), 1.38-1.59 (m, 3H), 1.64-1.73 (m, IH), 1.76-1.87 (m, IH), 5.08 (d of m, IH, J=llHz, Hy), 5.19 (d of m, IH, J=16Hz, H z), 5.66-5.77 (7 l i n e m, IH, H x). Exact Mass calcd. for C 1 2 H 2 2 2 8 S i : 194.1491; found 194.1489. The more polar f r a c t i o n on d i s t i l l a t i o n , as above gave a mixture of isomeric alkenes (77a-d) (21 mg, 4% based on s t a r t -ing monobromides) which was not further u t i l i z e d . General Procedure E : Thermolysis of endo-Vinylcyclopropanes A solution of the endo-vinylcyclopropane i n an appro-p r i a t e quantity of dry benzene (0.3-1.5 mL) in a thick-walled, base washed, s i l y l a t e d glass tube was completely degassed by subjecting i t to three freeze-pump-thaw cycles. The tube was then sealed under vacumm (/x.0.4 Torr) and placed i n a sand-packed s t e e l bomb which had been pre-heated to %230°C in an oven. The bomb and contents were then quickly replaced in the heating oven and heated for 6-8h at 230-240°C. The r e s u l t i n g clear solution was f i r s t analysed by glc and then the benzene - 290 -was removed. Chromatography and/or d i s t i l l a t i o n of the residue yielded the rearranged productCs). The glass tubes p r i o r to t h e i r use for thermolysis were subjected to the following procedures to remove any p o s s i b i l i t y of acid c a t a l y s i s of the rearrangements. Each tube was : (1) washed inside with very d i l u t e detergent solution and then rinsed with l o t s of water to remove traces of grease and dust; (2) washed inside with one percent aqueous potassium hydroxide and allowed to stand for 15 min f i l l e d with t h i s solution; (3) rinsed with four successive methanol washes; then (4) rinsed with two successive toluene washes; and (5) f i l l e d with a f i v e percent solution of dichlorodimethyl-silane in toluene and allowed to stand for 15 min; (6) emptied and rinsed immediately with toluene; (7) rinsed well with methanol and dried in an oven at 'v-130°C; (8) cooled in a dessicator or in a stream of inert gas and stored,with the open end capped, u n t i l ready for use. - 291 -Thermolysis of the endo-Enone (65a) OMe" 166 Following the general procedure E just described, 20 mg (0.10 mmol) of the endo-enone (65a) i n 0.5 mL of dry benzene was heated for 7h at 230-240°C. Analysis (by glc) of the r e s u l t i n g benzene solution indicated complete rearrangement to a single product. Removal of the benzene and d i s t i l l a t i o n (air-bath temperature 90-95°C, 0.4 Torr) of the residue gave 18 mg (90%) of the dienone (166) which was one component by glc analysis and exhibited uv (95% EtOH) X : 228 nm (e= 18,100); i r (film) : 3045, 1700, 1670, 1610, 1435, 1410, 1355, 1285, 1235, 1190, 1110, 1100, 845, 715 cm - 1; 1H nmr (400 MHz) 6 : 2.33-2.70 (m, 8H), 2.93-3.01 (m, IH, -CH=CH-CH-), 3.32 (s, 3H, -0CH 3), 3.70 (d of t, IH, J=4, 7Hz, -CHOMe), 5.60-5.64, 5.71-5.76 (two multiplets, 2H, -CH=CH-), 6.02-6.07 (m, IH, 9 -CCH=C-) . Exact Mass calcd. for C 1 2 H 1 6 ° 2 : 1 * 1 1 5 0 ' f o u n d : 192.1152. - 292 -Thermolysis of the Vinylcyclopropane (66a) Following the general procedure E already described, a solution of 30 mg (0.22 mmol) of the vinylcyclopropane (66a) in ^0.3 mL of dry benzene was heated for 7h at 230-240°C. The r e s u l t i n g benzene solution on glc analysis gave a single peak. Removal of the benzene by very c a r e f u l d i s t i l l a t i o n ( a i r -bath temperature 90-100°C) at atmospheric pressure followed by d i s t i l l a t i o n of the residue (air-bath temperature ^140°C), also at atmospheric pressure, yielded ^20 mg (67%) of the very v o l a t i l e diene (168). This material was e s s e n t i a l l y one component (/v99%) on glc analysis and exhibited i r (film) : 3025, 1630, 1440, 1400, 1350, 1230, 1195, 1110, 1095, 990, 950, 920, 855, 840, 725, 705 cm"1; 1H nmr (100 MHz) 6 : 1.72 (d, of d, 3H, J=2, 7Hz, -CH=CH-CH3), 2.15-2.87 (m, 2H, -CH2CH=CH-), 3.36 (s, 3H, -0CH 3), 3.50-3.90 (m, 2H, -CHOMe,-CH-CH=CH-), 5.03-5.80 (m, 4H, o l e f i n i c protons). Exact Mass calcd. for C 9 H 1 4 ° : 1 3 8 ' 1 0 4 5 ' f o u n d : 138.1043. It proved possible to rearrange the vinylcyclopropane (66a) by heating without solvent under the same conditions as above. Accordingly, 33 mg (0.24 mmol) of (66a) was heated - 293 -neat and the s l i g h t l y dark l i q u i d obtained was d i s t i l l e d ( a i r -bath temperature ^140°C) at atmospheric pressure to y i e l d ^30 mg (90%) of a colourless l i q u i d . This material was i d e n t i c a l by glc analysis and i n i t s proton nmr spectrum with that ob-tained from the thermolysis i n benzene. Thermolysis of the Vinylcyclopropane (67a) 170 Following the general procedure E already described, 40 mg (0.19 mmol) of the vinylcyclopropane (67a) in 1 mL of dry benzene was heated for 5.5h at 230-240°C. Analysis of the r e s u l t i n g benzene solution (by glc) indicated that e s s e n t i a l l y one product had been formed (^5% of another component was pre-sent). Removal of the solvent and d i s t i l l a t i o n (air-bath temperature ^125°C, 0.4 Torr) of the residue yielded 37 mg (93%) of a colourless l i q u i d . Careful analysis of the 400 MHz proton spectrum of t h i s material f a i l e d to confirm the i d e n t i t y of the minor component. However, the major product was un-- 294 -equivocally the diene (170) and exhibited i r (film) : 3020, 2975, 1440, 1360, 1200, 1175, 1150, 1130, 1110, 1025, 1020, 980, 850, 720 cm - 1; % nmr (400 MHz) s: 1.71 (d of d, 3H, J=2.5, 7Hz, -CH=CHCH3), 2.36 (d of m, 1H, J=16.5Hz, Hp), 2.70 (d of d of m, 1H, J=7, 16.5Hz, H £), 3.40 (s, 3H, -OCHg), 3.57 (t, 2H, J=\,5Hz, -CH 20CH 3), 3.61-3.67 (d of m, 1H, J= 10Hz, Hg), 3.68-3.74 (m, 2H, -OCH_2CH2OCH3), 4.15 (d of t, 1H, J=5, 7Hz, H c), 4.77 (AB type d of d, 2H, J=6Hz, -OCHgO-), 5.-20 (d of d of q, 1H, J1=2.5Hz, J 2=Jg W=J w x=10Hz, H w), 5.47-5.57 (m, 2H, H^, H x), 5.69-5.74 (m, 1H, Hj). Exact Mass calcd. for CgH 1 ; L(M +-C 4H g0 3) ; 107. 0860; found : 107.0876. Proton decoupling experiments at 400 MHz aided in the assignments of the signals. Thermolysis of the Vinylcyclopropane (75) 172 A solution of 30 mg (0.2 mmol) of the vinylcyclopropane (75) i n 0.5 mL of dry benzene was thermolysed, following the general procedure E described previously, at 230-240°C for - 295 -8.5h. Examination of the r e s u l t i n g benzene solution, by c a p i l l a r y g l c , indicated that i t consisted of one predominant product (/v94%) and two minor components (^6%). Removal of the solvent by careful d i s t i l l a t i o n at atmospheric pressure, followed by d i s t i l l a t i o n of the residue (air-bath temperature %70°C, 20 Torr) afforded 24 mg (80%) of a colourless, v o l a t i l e l i q u i d consisting of the same three components in the same proportions as above (by c a p i l l a r y glc a n a l y s i s ) . A l l three of these components on GC-MS analysis gave molecular ion peaks of m/e 152. Careful analysis of the 400 MHz proton nmr spec-trum of the thermolysate, at high magnification, did not re-veal the i d e n t i t y of the two minor components. However, no signal that might be due to either the methoxy protons or the o l e f i n i c proton H_ of the enol ether (173) ( i n the region 63.50 to 5.10) could be discerned. The major product was un-equivocally the diene (172) and exhibited i r (film) : 3000, 1440, 1180, 1100, 740, 710 cm - 1; 1H nmr (400 MHz) 6 : 1.60-1.68 (m, 1H), 1.69, (d of d, 3H, J=^2.5, 7Hz, -CH=CH-CH3), 1.90-2.22 (m, 3H), 3.13-3.23 (m, 2H, H c, Hg), 3.39 (s, 3H, -OCHg), 5.24 (d of d of q, 1H, J1=2.5Hz, J 2 = JBW = JWX = 1 0' 5 HZ' HW)' 5 , 3 6 ( d ° f m, 1H, J=10Hz, H A), 5.55 (d of d of q, 1H, J=^l, 7, 10.5Hz, H x) 5.64-5.71 (m, 1H, Hj). Exact Mass calcd. for C 1 0 H 1 6 ° : 1 5 2•1202; found: 152.1208. - 296 -Thermolysis of the Vinylcyclopropane (74) 174 175 Following the general procedure E already described, 110 mg (0.80 mmol) of the vinylcyclopropane (74) in 1.5 mL of dry benzene was heated at 230-240°C for 7.5h. Analysis (by glc) of the r e s u l t i n g benzene solution indicated clean rearrange-ment into two products i n a r a t i o of 88:12. Removal of the benzene and d i s t i l l a t i o n (air-bath temperature 50°C, 0.85 Torr) yielded 108 mg (98%) of the mixture of products i n the same r a t i o (88:12 by glc) as indicated above. Preparative layer chromatography ( s i l i c a g e l , 30% ether - petroleum ether) of t h i s mixture enabled characterisation of each component. The less polar and minor component, the ketone (174), gave one peak on glc analysis and exhibited i r (film) : 2975, 1700, 1440, 1345, 1310, 1220, 1210, 1180, 950, 710 cm - 1; ^ nmr (400 MHz) 6: 1.35-1.47 (m, IH), 1.55 (d of d, 3H, J=^2.5, 7Hz, -CH=CH-CH3), 1.59-1.80 (m, 2H), 1.95-2.10, 2.15-2.35 (two multiplets, 5H), 2.67-2.80 (m, IH, H^), 5.19 ( d of d of m, IH, J ^ J ^ l O H z , H w), 5.36 (d of d of q, IH, J=1.5, 7, 10Hz, H Y). Exact Mass calcd. for C QH 1 40 : 138.1045;found: - 297 -138.1048. The more polar major component, the hydroxy diene (175), was d i s t i l l e d (air-bath temperature 40°C, 0.85 Torr) to y i e l d material that gave one peak on g l c analysis and exhibited i r (film) : 3350, 3000, 1435, 1400, 1065, 1035, 960, 930, 915, 885, 745, 710, 685 cm - 1; h nmr (400 MHz) 6 : 1.59-1.71 (m, 2H), 1.73 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 1.96-2.05 (m, 1H), 2.11-2.27 (m, 2H, H^ ,, H Q), 3.06-3.15 (m, 1H, Hg), 3.52-3.60 (m,. 1H, H c), 5.25 (d of d of m, 1H, Jm= J ^ l O H z , H w), 5.36 (d of m, 1H, J=10Hz, H A), 5.67-5.80 (m, 2H, H x, Hj); 1 3 C nmr (100.8 MHz, broad band proton decoupled) 6 : 13.18 (CH=CH-CH3), 24.50 (-CH=CH-CH2-), 29.31 (-CH2CH0H), 43.21 (doubly a l l y l i c carbon) 71.86 (-CH0H), 126.93, 127.16, 128.03, 131.85 ( o l e f i n i c carbons). Exact Mass calcd. for CgH^O : 138.1045 ; found: 138.1045. The hydroxy alkene ( 74 ) (15 mg) was also thermolysed in dry diglyme (0.5 mL) under conditions s i m i l a r to those described above ( i . e . 230-240°C, 7.25h ). Analysis of the re-s u l t i n g solution by glc indicated that the same products (175) and ( 174) had been formed in the same r a t i o (88:12) as had been the case when the rearrangement was c a r r i e d out in benzene. - 298 -Thermolysis of the Vinylcyclopropane (68b) 176 177 Following the general procedure E already described, a solution of 51.9 mg (0.29 mmol) of the vinylcyclopropane (68b) in 2 mL of dry benzene was heated for 9h at 230-240°C. Analy-s i s of the r e s u l t i n g benzene solution by glc indicated clean rearrangement into two poorly resolved products i n a r a t i o (45:55) which was not considered r e l i a b l e . Removal of the benzene under reduced pressure and d i s t i l l a t i o n of the residue (air-bath temperature 40°C, 0.4 Torr) yielded 49 mg (94%) of the mixture of dienes (176) and (177). The r a t i o of (176) to (177) was found to be ^1:1.8 by integration of the signals due to the methoxy protons and the protons in the 400 MHz "*"H nmr spectrum of the mixture. Separation of the thermolysis mixture into i t s components was effected by preparative t i c ( s i l i c a g el, 5% ether - petroleum ether). The less polar f r a c -tions on d i s t i l l a t i o n (air-bath temperature 35-40°C, 0.4 Torr) gave the enol ether (176) as a colourless l i q u i d which was one component on g l c analysis. This material exhibited i r ( f i l m ) : 1660, 1450, 1220, 1140, 1110, 1095, 715 cm - 1; 1H nmr (80 MHz) - 299 -6 : 1.14 (s, 3H, t e r t i a r y methyl), 1.23-2.00 (m, 7H), 1.64 (d, 3H, J=6Hz, -CH=CH-CH3), 2.43-2.78 Cm, IH), 3.58 (s, 3H, -OCH3), 5.25-5.60 (m, 2H, H"x, H w), 5.85 (broad s, IH, H^). Exact Mass calcd. for C 1 2 H 2 Q 0 : 180.1514; found: 180.1509. The more polar f r a c t i o n on d i s t i l l a t i o n (air-bath tempera-ture 40-45°C, 0.4 Torr) yielded the diene (177) as a colour-l e s s l i q u i d which was one component on glc analysis. The material exhibited i r (film) : 3055, 1630, 1440, 1375, 1200, 1190, 1145, 1110, 970, 890, 725, 695 cm"1; aH nmr (80 MHz) 6 : 1.25-1.93 (m, 6H), 1.65 (d of d, 3H, J=1.5, 7Hz, -CH=CH-CH3), 2.03-2.35 (m, 2H), 3.38 (s, 3H, -OCHg), 3.50 (s, 2H, -CHgOMe), 4.76 (d, IH, J=2H, H^), 4.86 (broad s, IH, H R), 5.25 (d of m, IH, J=11.5Hz, H w), 5.60 (d of q, IH, J=7, 11.5Hz, H x). Exact Mass calcd. for C-j^H^O : 180.1514; found: 180.1511. 179 180 - 300 -Following the general procedure E already described, 60 mg (0.48 mmol) of the vinylcyclopropane (69) in 1.5 mL of dry benzene was heated at 230-240°C for 7.75h. Analysis of the r e s u l t i n g benzene solution by glc indicated the clean formation of two products in a r a t i o of 66:34. Removal of the benzene by careful d i s t i l l a t i o n (air-bath temperature 100°C, atmospheric pressure) followed by d i s t i l l a t i o n of the residue (air-bath temperature 50°C, 15 Torr) gave a quantitative y i e l d of a mix-ture of the dienes (180) and (179) i n the same proportion (66:34, respectively) as above (by glc a n a l y s i s ) . Preparative glc enabled i s o l a t i o n of each of these compounds. The minor product (179) (of lower retention time) exhibited i r ( f i l m ) : 3020, 2975, 1635, 1240, 1125, 1065, 885, 850, 740, 720 cm - 1; 1H nmr (400 MHz) 6: 1.65 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 1.59-1.70 (m, IH, Hj) , 1.89-1.98 (m, IH, Hj ) , 3.12-3.21 (broad m, IH, W^=16Hz, H A), 3.93-4.08 (m, 2H, Hp, H £), 4.53 (d of d, IH, J=4, 6.5Hz, H R), 5.26 (d of d of q, IH, J±=2Hz, J 2 = JAW = JWX = 10Hz, H w), 5.45 ( d of d of q, IH, J=^l.5, 7, 10Hz, H x), 6.39 (d of d, IH, J= 2.5, 6.5Hz, H^). Exact Mass calcd. f o r CgH^O: 124.0888; found: 124.0882. The major product (180) exhibited i r ( f i l m ) : 3010, 3000, 1455, 1445, 1400, 1380, 1310,1225, 1190, 1115, 1095, 970, 890, 850, 825, 780, 725, 685 cm - 1 ; 1H nmr (400 MHz) 6: 1.68 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 3.25-3.34 (broad m, IH, ¥ 4=22Hz, H R), 3.40 (d of d, IH, J=7.5, 11Hz, H c), 3.87 (d of d, - 301 -1H, J=5.5, 11Hz, Hp), 4.10-4.16 (ra, 2H, Hj., Hj,), 5.24 (d of d of q,lH, J1=2Hz, J 2=J B W.=J w x=10Hz, H w), 5.57 (d of d of q, 1H, J=^1.5, 7, 10Hz, H x), 5.68 (d of m, 1H, J=10Hz, H^), 5.76 (d of m, 1H, J=10Hz, Hj). Exact Mass calcd. for CgH^O: 124.0888; found: 124.0887. Thermolysis of the Vinylcyclopropane (70) 181 Following the general procedure E already described, 40 mg (0.27 mmol) of the vinylcyclopropane (70) in 0.5 mL of dry benzene was heated at 230-240°C for 7.25h. C a p i l l a r y glc ana-l y s i s of the r e s u l t i n g benzene solution indicated clean re-arrangement to a single product. Removal of the benzene under reduced pressure and d i s t i l l a t i o n of the residue (air-bath temperature ^80°C, 20 Torr) yielded 35.6 mg (89%) of the diene (181). This material gave one peak on c a p i l l a r y glc analysis and exhibited i r (film) : 3020, 2970, 1600, 1445, 1400, 1360, 1200, 1090, 1070, 1045, 930, 915, 845, 730, 710 cm - 1 ; 1H nmr (400 MHz) 6 : 1.69-1.80 (m, 1H, H Q), 1.72 (d of d, 3H, J=^2.5, - 302 -7Hz, rCH=CH-CH3), 1.93-2.05 Cm, 1H, Hj.) , 3.38-3.46 Cm, 1H, H^), 3.59 (broad d, 1H, J=10Hz, Hg), 3.64-3.72, 3.75-3.82 Ctwo multiplets, 2H, Hp, H £), 4.28 (d, 1H, J=6.5Hz, H c), 5.09 (d of d of q, 1H, J1=^2.5Hz, J 2 = JBW = JWX = 1 0 H z' V ' 5 - 4 8 ( d o f d o f q ' 1H, J=%1.5, 7, 10Hz, H x), 5.53-5.64 Cm, 2H, Hj , H^). Exact  Mass calcd. for C 1 0 H 1 4 0 : 150.1045; found: 150.1041. The ambiguities in the spectral assignments, e s p e c i a l l y d i s t i n g u i s h i n g between the multiplets Hg and H^, were c l a r i f i e d with the aid of decoupling experiments. Thermolysis of the ehdo-Enone (,71a) 183 184 A 5 mL round-bottomed f l a s k , equipped with a Kugelrohr d i s t i l l a t i o n bulb and containing 120 mg (0.63 mmol) of the trans-methyl endo-enone (71a), was evacuated and r e f i l l e d with nitrogen twice. The fla s k was then heated in an air-bath at 210-215°C f o r 15 min under a s l i g h t p o s i t i v e pressure of nitrogen. After being allowed to cool somewhat, the l i q u i d in the fl a s k was d i s t i l l e d (air-bath temperature ^120°C, 0.4 Torr) to y i e l d 115 mg (96%) of a colourless, odourless l i q u i d . Analysis (by glc) indicated that the rearrangement product - 303 -consisted of two components (not completely resolved under the conditions) in a r a t i o of * v l : l . Very careful analysis of the 100 MHz proton spectrum of the mixture, however, suggested that the r a t i o of (183) to (184) was Vl:1.3. Preparative t i c 132 on s i l v e r n i t r a t e impregnated s i l i c a gel (12.5% s i l v e r n i -t r a t e - s i l i c a g e l , 30% ether - petroleum ether, two develop-ments) separated the two components. The more mobile material the enone (183), aft e r d i s t i l l a t i o n (air-bath temperature rollO°C, 0.4 Torr) was one component by glc analysis and ex-hi b i t e d uv (hexane) X : 218 nm (e=16,400): i r (film) : max 1705, 1675, 1615, 1440, 1285, 1190 cm - 1; ^ nmr (80 MHz) 6 : 1.00-2.05 (m,4H), 1.64 (broad s, 3H, -CH=C-CH„), 2.12-2.70 (m, O a 9H, -CCH2CH2-, and the f i v e a l l y l i c protons), 5.22 (broad s, IH, W,=6Hz, CH0-C=CH-), 5.96 (broad s, IH, W,=4Hz, -?~CH=C-). 2 «J 2 — Exact Mass calcd. for C^H^O : 190.1358; found: 190.1364. The less mobile enone (184) after d i s t i l l a t i o n (air-bath ^110°C, 0.4 Torr), was >95% pure by glc analysis and exhibited uv (hexane) X : 219 nm (e=19,000); i r (film) : 3015, 1705, 1675, 1615, 1435, 1285, 1195, 1185, 690 cm - 1; % nmr (100 MHz) 6 : 0.96 (d, 3H, J=6Hz, secondary methyl), 1.14-1.82 (m, 3H), 0 1.86-2.24 (m, 3H), 2.24-2.72 (m, 6H, -CCHgCHg and two a l l y l i c protons), 5.33-5.82 (m, 2H, -CH=CH-), 5.97 (broad s, IH, 4Hz, -?-CH=C-). Exact Mass calcd. for C 1 3 H l g 0 : 190.1358; found: 190.1361. - 304 -Thermolysis of the Vinylcyclopropane (72a) 185 186 187 Following the general procedure E already described, a solution of 40 mg (0.29 mmol) of the vinylcyclopropane (72a) (96% plus 4% of a diastereomer) i n 0.5 mL of dry deuteroben-zene was heated at 230-240°C f o r 7.7h. C a p i l l a r y glc analysis of the r e s u l t i n g deuterobenzene solution indicated that the vinylcyclopropane (72a) had rearranged into two products ( i n a r a t i o of ^ 1.4:1) which made up 95.5% of the thermolysis mixture. Further, the hitherto u n i d e n t i f i e d diastereomer had also rearranged into a compound, very l i k e l y the diene (187), which made up ^ 4.5% of the thermolysis mixture and was not c l e a r l y separated ( c a p i l l a r y glc) from the minor of the other two products. Analysis of the 400 MHz proton nmr spectrum of the thermolysis mixture enabled the i d e n t i f i c a t i o n of signals appropriate to each of the compounds (185), (186) and (187) and yielded a r a t i o of (186) to (185) of VL.4:1. GC-MS analy-s i s confirmed the isomeric nature of the components of the thermolysis mixture giving molecular masses of m/e 136 f o r both (186) and the unresolved mixture of (185) and (187). - 305 -Jit was not possible to cleanly separate these very vola-t i l e rearrangement products. However, careful preparative glc gave as the f i r s t eluted f r a c t i o n the diene (186) which was one component by c a p i l l a r y glc analysis and exhibited ^ H nmr (400 MHz) 6 : 0.92 (d, 3H, J=7Hz, -CHCHg), 1.24-1.42 (m, 2H, H D, H £), 1.64 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 1.68-1.76 (m, 1H, H c), 1.99-2.09 (m, 2H, Hp, H Q), 2.63-2.73 (m, 1H, Hg), 5.14 (d of d of m, 1H, Jg W=J w x=10.5Hz, H f f), 5.36(d of m, 1H, J=10Hz, H A), 5.51 (d of q, 1H, J=7, 10.5Hz, H x) , 5.65-5.73 (m, 1H, Hj) Exact Mass calcd. for C 1 ( )H 1 6 ; 136.1252; found: 136.1245. The second f r a c t i o n c o l l e c t e d consisted mainly of the diene (185) (^90% by c a p i l l a r y g l c analysis) but also contain-ed dienes (186) and (187). The 400 MHz proton nmr spectrum of t h i s material thus showed weak signals a t t r i b u t a b l e to the l a t t e r two dienes. The major component, the diene (185), ex-hi b i t e d h nmr (400 MHz) 6: 1.19-1.31 (m, 1H), 1.51-1.62 (m, 1H), 1.62-1.68 (m - due to overlapping d of d and s-,6H, -CH=CH-CH3, -CH=C-CH3), 1.68-1.81 ( p a r t i a l l y obscured m,2H), 1.83-1.98 (m, 2 H , a l l y l i c methylene protons), 3.02-3.12 (m, 1H, H A), 5.19 (broad s, 1H, W^=8Hz, Hg), 5.25 (d of d of q, 1H, J1=^2.5Hz, J 2= J ^ J ^ l O . S H z , H w), 5.41 (d of d of q, 1H, J=lHz, 7Hz, 10.5Hz, H x). - 306 -Following the general procedure E already described, a solution of 115 mg (0.58 mmol) of the vinylcyclopropane (76) in 1.5 mL of dry benzene was heated at 230-240°C for 7.5h. Analysis (by glc) of the r e s u l t i n g benzene solution indicated clean rearrangement into two products in a r a t i o of 82:18. Removal of the benzene under reduced pressure and d i s t i l l a t i o n (air-bath temperature ^80°C, 0.4 Torr) of the residue yielded 100 mg (87%) of the dienes (189) and (188) which were again in a r a t i o 82:18 by c a p i l l a r y glc analysis. The two thermoly-s i s products were separated by preparative g l c . The minor product, the diene (188), exhibited i r ( f i l m ) : 3025, 2995, 1590, 1485, 1445, 755, 730, 700, 680 cm"1; 1H nmr (400 MHz) 5: 1.24 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 1.81-1.94 (m, 2H, H £, H D), 2.08-2.21 (m, 2H, H p, H Q) , 2.47 (d of t, IH, J=^4.5, 10.5Hz, H c), 3.20-3.29 (m, IH, W^=22Hz, H R), 5.13 (d of d of q, IH, J ^ ^ . S H z , J 2=J R W=J w x=10Hz, H w), 5.32 (d of d of q, IH, J=VL.5, 7, 10Hz, H x), 5.51 (d of m, IH, J=10Hz, H^), 5.74-5.82 (m, IH, H T), 7.12-7.19, 7.21-7.33 (two multiplets, 3H, - 307 -2H, aromatic protons). Exact Mass calcd. for C^gH^g: 198.1409; found: 198.1409. The major product (of longer retention time), the diene (189), exhibited i r ( f i l m ) : 3060, 3025, 2990, 1590, 1490, 1440, 1430, 755, 720, 700 cm"1; h nmr (400 MHz)6 : 1.32-1.43, 1.64-1.77, 1.78-1.98 (three multiplets, 1H, 1H, and 2H, respectively) 1.70 (d of d, 3H, J=2, 7Hz, -CH=CH-CH3), 2.34-2.44 (m, 2H, H^ ,, H D), 3.22-3.32 (m, 1H, W^=20Hz, H^), 5.32 (broad d of d, 1H, JAW = JWX = 1 0* 5 H z' V ' 5 ' 4 6 ( b r o a d d o f <1> 1 E> J = 7 » 10.5Hz, H x), 5.87 (broad s, 1H, Hg), 7.17-7.32, 7.35-7.41 (two multiplets, 3H, 2H, aromatic protons). Exact Mass calcd. for C^^H^g : 198.1049; found: 198.1405. Thermolysis of the Vinylcyclopropane (77a) SiMe3 190 Following the general procedure E already described, 30 mg (0.15 mmol) of the vinylcyclopropane (77a) (containing ^ 5% of isomeric impurities) in ^ 1 mL of dry benzene was heated at - 308 -230-240°C for 7h. C a p i l l a r y g l c analysis of the r e s u l t i n g benzene solution indicated the formation of a single product. Removal of the benzene under reduced pressure and d i s t i l l a t i o n of the residue (air-bath temperature 110°C, 30 Torr) yielded 27 mg (90%) of the diene (190) which was ^97% pure by c a p i l l -ary glc analysis. This material exhibited i r ( f i l m ) : 2980, 1600, 1440, 1400, 1250, 1245, 1070, 985, 945, 915, 880, 860, 855, 835, 825 750, 715, 690 cm - 1; 1H nmr (400 MHz)6 : 0.04 (s, 9H, - S i ( C H 3 ) 3 ) , 1.25-1.36 (m, IH), 1.48-1.59 (m, IH), 1.66 (d of d, 3H, J=V2, 7Hz, -CH=CH-CH3), 1.70-1.81 (m, 2H), 1.94-2.08 (m, 2H, H £, Hp), 3.04-3.14 (m, IH, H^), 5.26 ( d of d of q, IH, J±= 2Hz, J 2=J A W=J w x=10Hz, H w), 5.44 (d of d of q, IH, J=^1.5, 7, 10Hz, H x), 5.70-5.75 (narrow m, IH, H R). Exact Mass calcd. for C 1 9 H 9 9 S i : 194.1491; found: 194.1504. - 309 -BIBLIOGRAPHY 1. G.B. Bennett, Synthesis, 1977, 589. 2. F.E. Ziegler, Acc. Chem. Res. 1977, 10, 227 3. S.J. Rhoads and N.R. Raulins, Org. Reactions, 1975, 22,1. 4. D.R. Rayner, E.G. M i l l e r , P. Bickart, A.J. Gordon and K. Mislow, J. Amer. Chem. Soc., 1966, 88, 3138. 5. J.A. Berson and G.L. Nelson, J. Amer. Chem. S o c , 1967, 89, 5503. 6. T.L G i l c h r i s t and R. C. Storr, "Organic Reactions and Or-b i t a l Symmetry," 2nd Ed i t i o n , Cambridge University Press, 1979. pp. 287-288. 7. L. Velluz, G. Amiard and B. Goffinet, Bull.Soc. Chem.Fr.. 1955, 1341. 8. C.W. Spangler, Chem. Rev., 1976, 76, 187. 9. H.M. Frey and R.J. E l l i s , J. Chem. Soc., 1965, 4770. 10. H.M. Frey and B.M. Pope, J. Chem. Soc. A, 1966, 1701. 11. W.R.Roth and J. Konig, Justus Liebigs Ann. Chem., 1966, 699, 24. 12. S. McLean, C.J. Webster and R.J.D. Rutherford, Can. J. Chem., 1969, 47, 1555. 13. E.N. Marvel, G. Caple, B. Schatz and W. Pippin, Tetrahe-dron, 1973, 29, 3781. 14. K.W. Egger, J. Amer. Chem. Soc, 1967, 89, 3688. 15. T. Nozoe and K. Takahashi, Bull. Chem. Soc. Jpn., 1965, 38, 665. 16. J.M. Brown, B.T. Golding and J.J. Stafko J r . , J. Chem. Soc. Chem. Comm., 1973, 319. 17. A. de Meijere, Angew. Chem. Int. Edn. Engl., 1979, 18, 809. 18. R.J. E l l i s and H.M. Frey, J. Chem. Soc. Suppl.I, 1964,5578. - 310 -19. W.R. Roth and J. Konig, Justus Llebigs Ann. Chem. 1965, 688, 28. 20. R.M. Roberts, R.G. Landolt, R.N. Greene and E.W. Heyer, J. Amer. Chem. S o c , 1967, 89, 1404. 21. J.K. Crandall and R.J. Watkins, Tetrahedron Lett., 1967, 1717. 22. D.E. Minter and G.J. Fonken, Tetrahedron Lett.,1977, 4149. 23. V. Dalacker and H. Hopf, Tetrahedron Lett., 1974, 15. 24. H.M. Frey and R.K. So l l y , Int. J. Chem. Kinet., 1969, 1, 473. 25. M. Schakel and G.W. Klump, Rec. Trav. Chim. Pays-Bas, 1973, 92, 605. 26. G. Ohloff, Tetrahedron Lett.,1965, 3795. D.S. Glass, R.S. Boikess and S. Winstein, Tetrahedron Lett., 1966, 999. 27 28 D.E. McGreer and N.W.K. Chiu, Can. J. Chem., 1968, 46, 2217. 29. M.S. Baird, G.S. Lindsay and C.B. Reese, J. Chem.  Chem. Comm. 1968, 784. 30. D.L. Garin, J. Amer. Chem.Soc.. 1970, 92, 5254. 31. A. J.Asche I I I , J. Amer. Chem. Soc., 1970, 92, 1233. 32. W.R. Roth, J. Konig and K. Stein, Chem. Ber., 1970, 103, 426. 33. R.B. Woodward and R. Hoffmann, "The Conservation of Or-b i t a l Symmetry," Academic Press, 1970. 34. E.J. Corey and D.K. Herron, Tetrahedron Lett., 1971,1641. 35. E.J. Corey, H. Yamamoto, D.K. Herron, K. Achiwa, J. Amer.  Chem. S o c , 1970/ 92, 6635. 36. S.A. Monti and T.W. McAninch, Tetrahedron Lett., 1974, 3239. 37. J. Froborg and G. Magnusson, Tetrahedron, 1978, 34, 2027. - 311 -38. y. Bahurel, L. Cotti e r and G. Descotes, Synthesis, 1974, 118. 39. I. Fleming, "Frontier Orbitals and Organic Chemical Re-actions," John Wiley and Sons Ltd., 1976. 40. N.D. E p i o t i s and S. Shaik, J. Amer. Chem. Soc., 1977, 99, 4936. 41. B.K. Carpenter, Tetrahedron, 1978, 34, 1877. 42. C.J. Burrows and B.K. Carpenter, J. Amer. Chem. Soc., 1981 103, 6983; 6984. 43. M.T. Zoeckler and B.K. Carpenter, J. Amer. Chem. S o c , 1981, 103, 7661. 44. A.P. ter Borg, E. Razenberg and H. K l o o s t e r z i e l , Reel. Trav. Chim. Pays-Bas, 1965, 84, 1230. 45. R. Breslow, J.H. Hoffman and C. Perchonock, Tetrahedron  Lett., 1973, 3723. 46. G.V. Meehan, G.G. Pegg and J. Taylor, J. Chem. Soc. Chem.  Comm., 1979, 709. 47. H. Kwart, S.F. Sarner and J. Slutsky, J. Amer. Chem. Soc. 1973, 96, 5234. 48. M.J. Jorgenson and A.F. Thacher, Tetrahedron Lett., 1969, 4651. 49. L.A. Paquette, G.D. Crouse, A.K. Sharma, J. Amer. Chem. Soc., 1980, 102, 3972. 50. D. Evans and A.M. Golob, J. Amer. Chem. Soc., 1975, 97, 4765. 51. B. Franzus, M.L. Scheinbaum, D.L. Waters H.B. Bowlin, J. Amer. Chem. Soc., 1976, 98, 1241. 52. E. Piers and I. Nagakura, Synth. Comm. 1975, 5, 193 and E. Piers, J.R. Grieson, C.K. Lau and I. Nagakura, Can.J. Chem., 1982, 60, 210. 53. a.E. Piers, C.K. Lau and I.Nagakura, Can. J. Chem., 1983, 61, 288. b.E. Piers, K.F. Cheng and I. Nagakura, Can. J. Chem.,1982, 60, 1256. - 312 -54. a.J.P. Marino and L.J. Browne, Tetrahedron Lett., 1976, 3245. b.J.P. Marino and M.P. Ferro, J. Org. Chem., 1981, 46, 1912. 55. P.A. Wender and M.P. F i l o s a , J. Org. Chem, 1976, 41, 3490. 56. E. Piers and E. Ruediger, J. Chem. Soc. Chem. Comm., 1979,166. 57. a.E. Piers and J. Banvi l l e , J. Chem. Soc. Chem. Comm., 1979, 1138. b.E. Pi e r s , J. Ban v i l l e , C.K. Lau and I. Nagakura, Can. J.  Chem., 1982, 60, 2965. 58. E. Piers, I. Nagakura and J.E. Shaw, J. Org. Chem., 1978, 43, 3431. 59. For a compilation of these sesquiterpanes see: R.D. L i t t l e A. Bukhari, M.G. Venegas, Tetrahedron Lett., 1979, 305. 60. a.P. Deslongchamps, Tetrahedron, 1975, _31, 2463; Hetero-cycles, 1977, 7, 1271. b.V. Malatesta and J.C. Scaiano, J. Org. Chem.,1982, 47, 1455 and references therein. 61. a.See, for example, (a) T. Hirao, Y. Harano, Y. Yamana, Y. Ohshiro and T. Agawa, Tetrahedron Lett., 1983, 1255. b. R. Danheiser, J.M. Morin J r . , M. Yu, and A. Basak, Tetra-hedron Lett., 1981, 4205. c. K. K i t a t a n i , T. Hiyama and H. Nozaki, B u l l . Chem. Soc. Jpn., 1977, 50, 1600; 3288. 62. W. von Doering and A.K. Hoffmann, J. Amer. Chem. S o c , 1954, 76, 6162. 63. M. Makosza and M. Fedorynski, Synth. Comm.,1973, 305. 64. D. Seyferth, H. Yamazaki and D.L. Alleston, J. Org. Chem. 1963, 28, 703. 65. K. Hofmann, S. Orochena, S.M. Sax and G. A. Jeffrey, J. Amer. Chem. Soc., 1959, 81, 992. - 313 -66. G. Kobrich and W. Goyert, Tetrahedron, 1968, 24, 4327. 67. a.H.M. Walborsky, F.J. Impasto and A.E. Young, J. Amer. Chem. Soc., 1964, 86, 3283. b.B.J. Wakefield, "The Chemistry of Organolithium Compounds," Pergamon Press Ltd., 1974. Chapter 4. 68. G.H. Posner, D.J. Brunelle and L. Sinoway, Synthesis, 1974, 622. 69. For a comprehensive review on the addition of diazoacetic esters to unsaturated compounds see: V. Dave and E.W. Warnhoff, Org. Reactions, 1970, 18, 217. 70. For a discussion of the e f f i c a c y of other catalysts e.g. Rhodium (II) carboxylates for diazoester addition to o l e -f i n s see: M.P. Doyle, D. Van Leusen and W.H. Tamblyn, Synthesis, 1981, 787 and references therein. 71. See: W.R. Moser, J. Amer. Chem. Soc., 1969, 91, 1135 as well as references 69 and 70. 72. a.E.A. Evans, J. Chem. Soc., 1956, 4691. b.For a very recent report on formylation of alk y l l i t h i u m s using a variety of formylating reagents see: W. Amaratunga and J.M.J. Frechet, Tetradedron Lett., 1983, 1143. 73. A. Schmidt and G. Kobrich, Tetrahderon Lett., 1974, 2561. 74. K. Alder and F.H. Flock, Chem. Ber., 1956, 89, 1732. 75. R.B. Moffett, Org. Syn. C o l l . Vol. IV, 1963, 238. 76. E.J. Corey, J-L. Gras and P. U l r i c h , Tetrahedron Lett., 1976, 809. 77. P. Bedos, B u l l . Soc. Chim. Fr., 1926, 39, 292. 78. K.E. Wilson, R.T. Seider and S. Masamune, J. Chem. Soc.  Chem. Comm., 1970, 213. 79. J. Pouchet and J.R. Campbell, "The A l d r i c h Library of NMR Spectra," Vol. I, No. 113A. 80. M. Miyashita, A. Yoshikoshi and P.A. Grieco, J. Org. Chem., 1977, 42, 3772. 81. A. Berlande, B u l l . Soc. Chim. Fr., 1942, 9, 642. - 314 -82. E-.R. Alexander and A. Mudrak, J. Amer. Chem. S o c , 1950, 72, 1810, 83. a.C. Eaborn, R. Jackson and R. Pearce, J. Chem. Soc. Perkln I, 1974, 2055. ~ b.J.M. Reuter, A. Sinha and R.G. Salomon, J. Org. Chem., 1978, 43, 2438. 84. B.S. Furniss, A.J. Hannaford, V. Rogers, P.W.G. Smith and A.R. T a t c h e l l , "Vogel's Textbook of P r a c t i c a l Organic Chemistry,** 4th E d i t i o n , Longman 1978; pp.401-402. 85. H.O. House, C.Y. Chu, J.M. Wilkins and M.J. Umen, J. Org.  Chem., 1975, 40, 1460. 86. L.J. Bellamy, "The Infra-red Spectra of Complex Molecules" 3rd E d i t i o n , Chapman and H a l l , London, 1975; p. 86. 87. J. Pouchet and J.R. Campbell, "The A l d r i c h Library of NMR Spectra", Vol. I, No. 39D. 88. a.A. Berlande, B u l l . Soc. Chim. Fr., 1942, 9, 641. b.For spectral data for t h i s compound see: D.L.J. C l i v e , P.C. Anderson, N. Moss and A. Singh, J . Org. Chem., 1982, 47, 1641. For a review see: Y-H. L a i , Synthesis, 1981, 585. For a very recent report on the e f f i c i e n t preparation of a l l y l i c Grignard reagents using s l u r r i e s of precondensed magnesium see: W. Oppolzer, E.P. Kiindig, P.M. Bishop and C. Perret, Tetrahedron Lett., 1982, 3901. 91. Ref. 86,pp.376-377. 92. J. Cologne and P. Boisde, B u l l Soc. Chim. Fr., 1956, 524. 93. H. P a r i s e l l e , Ann. Chim., 1911, 24, 315. 94. H.O. House, W.C. Liang and P.D. Weeks, J. Org. Chem., 1974, 39, 3102 and references therein. 95. 0. Grummitt, E.P. Budwitz and C.C. Chudd, Org. Syn. C o l l . Vol. IV, 1963, 748. 96. J. Pouchet and J.R. Campbell, "The A l d r i c h Library of NMR Spectra," Vol. I, No. 101B. 97. O. Heuberger and L.N.Owen, J . Chem. S o c , 1952, 910. 89, 90 - 315 -98. MThe Merck Index," 9th E d i t i o n , 1976. p. 9405. 99. a. Ref. 86,pp. 370-371 b. Ref. 86, p. 135. 100. R. Paul and S.Tchelitcheff, Compt. Rend., 1947, 224, 1722. 101. K.C. Nicolaou, R.L. Magolda, W.J. Sipio, W.E. Barnette, Z. Lysenko and M.M. J o u i l l e , J. Amer. Chem. Soc., 1980, 102, 3784. 102. D.L. Garin, J. Org. Chem., 1969, 34, 2355. 103. For previous examples of the use of amines in selenoxide eliminations see: a. H.J. Reich, J. M. Renga and I.L. Reich, J. Amer. Chem. Soc., 1975, 97, 5434. b. D. Labar, L. Hevesi, W. Dumont and A. K r i e f , Tetrahedron  Lett., 1978, 1141. 104. a.K.B. Sharpless and R.F. Lauer, J. Amer. Chem. Soc. 1973, 95, 2697. b. K.C. Nicolaou and W.E. Barnette, J. Chem. Soc. Chem.Comm. 1977, 331 . c. D.L.J. C l i v e , Tetrahedron, 1978, 34, 1049. 105. For an explanation of t h i s i n t e r e s t i n g s e l e c t i v i t y in the related sulfoxide elimination reaction see: B.M. Trost, T.M. Salzmann and K. H i r o i , J. Amer. Chem. S o c , 1976,98, 4887. 106. L. Ruest, G. Blouin and P. Deslongchamps, Synth. Comm., 1976, 6, 169. 107. F.W. Sum and L. Weiler, Can. J. Chem. 1979, 57, 1431. 108. Ref. 86,p.348. 109. D.H. Williams and I. Fleming, "Spectroscopic Methods in Organic Chemistry," 3rd E d i t i o n , Mc Graw H i l l Book Com-pany (UK) Ltd., 1980, p.146. 110. G. Kauffman and L.A. Teter, Inorg. Syn. 1963, 7, 10. 111. C.A. Brown, D. Barton and S. Sivaram, Synthesis, 1974,434. - 316 -112. .For an excellent review of the p r i n c i p l e s of phase transfer c a t a l y s i s and i t s application i n organic chemis-t r y see: W.P. Weber and G.W. Gokel, "Phase Transfer Catalysis in Organic Synthesis," Springer-Verlag, 1977 113. See;E. Dehmlov and M. L i s e l l , Chem. Ber.,1978, 111, 3873. 114. See: D.G. Lindsay and C.B. Reese, Tetrahedron, 1965, 21, 1673 and references therein. 115. W.C. S t i l l , M. Kahn and A. Mitra, J. Org. Chem., 1978, 43, 2923. 116. L.M. Jackson and S. Sternhell, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry," 2nd E d i t i o n , Pergamon Press, 1969. p. 281. 117. Ref. 116, pp. 287-288 118. For a f u l l review of zinc-copper couple/methylene iodide cyclopropanation of o l e f i n s including a discussion of the d i r e c t i n g e f f e c t s of hydroxy groups on the stereo-chemistry of the reaction see: H.E. Simmons, T.L. Cairns, S.A. Vladuchick and CM. Hoiness, Org. Re-actions, 1973, 20, 1. 119. M. Hanack and H. Allmendinger, Chem. Ber., 1964, 97, 1669. 120. See,for example, V.M. Parikh, "Absorption Spectroscopy of Organic Molecules," Addison-Wesley Publishing Co.,Inc., 1974. p.173. 121. L.A. Paquette, S.E. Wilson, R.P. Henzel and G.R. Allen J r . J. Amer. Chem. Soc., 1972, 94, 7761. 122. K.G. Taylor, W.E. Hobbs and M. Saquet, J. Org. Chem,1971 36, 369. 123. Ref. 116, p. 275. 124. For a procedure for i t s preparation see: H.G. K u i v i l a , Synthesis, 1970, 499. 125. Ref. 120, p. 170. 126. Ref. 69,pp. 230-231. 127. M. Alonso and M. Gomez, Tetrahedron Lett., 1979, 2763. - 317 -128. -See examples i n Ref. 116, p. 228. 129. G.H. Gunther, "NMR Spectroscopy - An Introduction," Translated by R.W. Gleason, John Wiley and Sons Ltd., 1980. p. 51 and pp. 93-94. 130. E.J. Corey and J.W. Suggs, Tetrahedron Lett., 1975, 2647. 131. G. Wittig and U. Schoellkopf, Org. Syn., 1960, 40, 66. 132. For some applications and references r e l a t i n g to t h i s technique see: J.G. Kirchner in "Techniques of Chemis-t r y , " Vol XIV; "Thin Layer Chromatography," Ed. E.S. Perry. pp. 47,80 and 888-889. 133. Ref. 129, p. 41. 134. See Ref. 116, p. 316 for a discussion of a l l y l i c coupl-ing in lH nmr. 135. a. See Ref. 116, p. 227 re geminal coupling of protons of a terminal methylene group. b. See Ref. 116, pp.301-304 for a discussion of v i c i n a l interproton coupling across double bonds. 136. Ref. 120, p. 160. 137. Ref. 116, p. 164. 138. Ref. 129, p. 162. 139. Ref. 129,p. 72. 140. J. Hine, "Structural E f f e c t s on E q u i l i b r i a in Organic Chemistry," John Wiley and Sons, Inc., N.Y. 1975; p.273. 141. G.S. Hammond, J. Amer. Chem. S o c , 1955, 77,334. 142. J . J . Gajewski, Acc. Chem. Res., 1980, 13, 142. 143. R. Wehrli, D. Bellus , H-J. Hansen and H. Schmid, Chimia, 1976, 30, 416; Helv. Chim. Acta, 1977, 60, 1325. 144. W.R. Roth, Tetrahedron Lett., 1964, 1009. 145. E.W. Colvin, Chem. S o c Rev., 1978, 7, 15. 146. J.W. Wilt and P.M. Aznavoorian, J. Org. Chem.,1978, 43, 1978. - 318 -147. -Ref. 140, pp. 97-99. 148. J. Hine, L.G. Mahone and C L . L i o t t a , J. Amer. Chem. Soc., 1967, 89, 5911. 149. J.A. Hirsch and X.L. Wang, Synth. Comm., 1982, 12, 333. 150. M.E. Garst, J.N. Bo n f i g l i o , D.A. Grudoski and J. Marks, J. Org. Chem., 1980, 45, 2307. 151. F.G. Bordwell, M. Van Der Puy, N.R. Vanier, J. Org.Chem., 1976, 41, 1885. 152. See Ref. 140, chapter 3, section 3-3. 153. Ref. 129, p. 47. 154. H. Gilman and F.K. Cartledge, J. Organomet. Chem., 1964. 2, 447. 155. P.G.M. Wuts, Synth. Comm., 1981, 11, 139. 156. E. Le Geoff, J. Org. Chem., 1964, 29, 2048. 157. J.E. Hodgkins and R.J. Flores, J. Org. Chem., 1963, 28, 3356. 158. J.E. Leibner and J. Jacobus, J. Org. Chem., 1979, 44, 449. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059454/manifest

Comment

Related Items