COPPER(I) CHLORIDE-MEDIATED INTRAMOLECULAR CONJUGATE ADDITIONS OF VINYLTRIMETHYLSTANNANE FUNCTIONS TO a ,P-ALKYNIC ESTERS by EVA-MARIA BOEHRINGER B. Sc., The University of Waterloo, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1996 © Eva-Maria Boehringer, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Chun's' The University of British Columbia Vancouver, Canada Date %. 0 14, eq 5).6 12 13 14 A final example of intramolecular Michael additions involving primary alkyl functions is presented by Danheiser et al.1 In a general method, organozinc iodides, prepared by oxidative addition of activated zinc metal to the corresponding organo iodides, are added in a 1,4-manner to a,P-unsaturated ketones and a,p-alkynic esters, producing five-and six-membered ring products (e.g. 15 -> 16, eq 6). 5 2. Copper(I) chloride-mediated intramolecular Michael additions: Background Recent work in our laboratory led to the discovery that intramolecular cross-coupling of vmyltrimethylstannane and alkenyl halide functions can be achieved in an efficient, stereospecific manner by treatment of appropriate substrates with copper(I) chloride (e.g. 17 —> 18, eq 7).8a CuCl (2.2 equiv) DMF, 62°C 80% C0 2Et 17 18 Control experiments have shown that in order to maximize yields and minimize reaction times, at least two equivalents of copper(I) chloride are required for this coupling process. One equivalent is not sufficient to allow the reactions to go to completion, whereas 1.5 equivalents produce good yields, but greatly extend reaction times.8a As a result of mechanistic studies,8b it is believed that a transmetalation between the vinylstannane function and the copper(I) salt initiates the coupling process, leading to the production of a vinylcopper(I) species as an intermediate with trimethylstannyl chloride (Me3SnCl) as a co-product (eq 8). l| + CuCl j | + Me3SnCl (8) R 6 Evidence supporting the formation of this type of organocopper(I) intermediate has been presented by Liebeskind et al. in a recent paper.9 By monitoring the reactions of vinyl- and phenytributyltin with copper(I) iodide with the aid of 119Sn-nmr, the consumption of the organostannanes, as well as the formation of a new product, identified as tributylstannyl iodide ( B u 3 S n I ) , was observed. Based on these experimental findings, it was proposed that a second product, namely the organocopper(I) derivative, was in all likelihood also produced in this reaction. These observations, together with information in a recent paper by Torii et al. concerning a copper(I) chloride-mediated intermolecular Michael-type addition of alkenyl-tributylstannanes to allenecarboxylates,10 provided the impetus for work in a new area: intramolecular copper(I) chloride-mediated conjugate additions of vmyltrimethylstannane functions to a,P-unsaturated ketones. Recent reports from our laboratory have shown that these types of conjugate additions are indeed feasible, proceeding with good to excellent yields.11 Several substituted bicycles have thus far been produced (e.g. 19 —> 20, eq 9) and further work is ongoing to extend this methodology to the syntheses of tri- and polycyclic ring systems.12 7 3. Proposal to study copperfl) chloride-mediated conjugate additions of vmyltrimetfayl-stannane functions to a.P-alkyriic esters In view of the limited number of reports disclosing the use of non-stabilized carbanionic functions in intramolecular Michael additions, as well as the fact that studies in our laboratory have shown copper(I) chloride-mediated intramolecular conjugate additions to a,p~unsaturated ketones to be quite successful, it seemed reasonable to extend this new methodology to encompass copper(I) chloride-mediated additions of vmyltrimethylstannane functions to triple bonds, in particular to a,P-alkynic esters. To test the validity of the proposed method, we initially envisaged preparing a number of structurally related vmyltiimethylstannane substrates, in which only the chain length between the vinylstannane function and the a.^-alkynic ester group was varied (compounds 21-23). C0 2Me CQ 2Et Me0 2C Et0 2C SnMe3 SnMe3 21 22 C0 2Me Me0 2C SnMe3 23 8 Upon treatment with copper(I) chloride, these substrates should undergo the proposed transmetalation-intramolecular 1,4-addition to form symmetric monocycles of varying ring sizes (diesters 24-26). C0 2Me C0 2Et C0 2Me 24 25 26 Since it is well established that in the addition of an organocopper(I) reagent to an alkynic triple bond the carbon and copper atoms of the reagent add in a stricdy cis fashion,13 it seemed likely that the stereochemistry of the products 24-26 should be as shown above (see Scheme 1). C0 2 R + Me3SnCl R0 2 C R0 2 C 30 work-up R0 2 C R0 2 C 29 S C H E M E 1 9 Initial copper-tin transmetalation of compounds of general structure 27 to give the vinylcopper intermediates 28, followed by intramolecular cis addition of the alkenylcopper(I) function across the triple bond, should lead to the stereoselective formation of compounds 29. Upon protonative work-up, the symmetric diesters of general structure 30 would be produced. If the copper(I) chloride-mediated intramolecular additions proposed in Scheme 1 proved successful, we planned to extend the method further to include structurally more complex systems. In particular, we were interested in preparing highly functionalized and/or sterically hindered substrates to test the limitations of the proposed methodology. It was thought that attempting the cyclization of compounds 31-33 would effectively demonstrate the effects of functionalization on the intramolecular addition and would, as well, provide a convenient route to strained mono- and bicycles (esters 34-36, respectively). y ^ C 0 2 M e Cn ^ C 0 2 M e OH OH ^ C 0 2 M e O H 34 35 36 10 To study the effects of steric hindrance on the conjugate addition, we envisaged the preparation of compound 37, similar in structure to 36, but containing a quaternary centre adjacent to the allcenyltrimethylstannane function. Cyclization of this somewhat more sterically congested system would produce the interesting bicycle 38, presumably as a mixture of diastereomers. We believed that the successful preparation of the above products (24-26, 34-36 and 38) from their alkenylstannane precursors (23-25, 31-33 and 37, respectively) via the proposed route of a copper(I) chloride-mediated intramolecular conjugate addition would not only show the viability of the method, but also demonstrate its general utility. In the following section of this thesis, the preparation of the required substrates will be discussed. In addition, the results of the copper(I) chloride-mediated conjugate additions of these substrates will be given and a discussion of these results, as well as of any limitations encountered, will be provided. 37 38 11 n. RESULTS AND DISCUSSION 1. Preparation of the aJJcenyltrimethylstannanes 21. 22 and 23 We decided to prepare the requisite alkenyltrimethylstannanes of general structure 39 from the corresponding dialkyl dialkynedioates (general structure 40). These could, in turn, be readily produced from commercially available diyne precursors (general structure 41), as discussed in Section 1.1. A retrosynthetic strategy is provided in Scheme 2. CO2R 41 S C H E M E 2 1.1 Preparation of the dialkyl dialkynedioates 42. 43 and 44 Initially, syntheses of the dialkyl dialkynedioates 43 and 44 were required. A previously prepared sample14 of compound 42, synthesized by a procedure15 analogous to that employed for the diesters 43 and 44, was utilized for the present work. 12 42 43 The diester 43 was readily prepared via literature procedures153 (eq 10) from commercially available 1,6-heptadiyne (45). Product yields, as well as spectral data derived from compound 43, were found to match literature values.153 1) MeLi, THF, H _ _ ^ 7 ^ H - T P O - M ^ E t f V : _ _ _ ^ H I S - C C y 3 , (10) 2) EtOC(0)Cl, 45 -20°Ctort 43 40% When preparing compound 44 from commercially available 1,7-octadiyne (46), it was discovered that minor changes in the established reaction conditions156 improved previously reported yields (eq 11). For detailed reaction conditions, the reader is referred to the experi-mental section, p 63. Thus, product 44 was produced in a 71% yield compared to a 30% yield H 1) MeLi, THF, /ft -78°Cto-20°C 2)MeOC(0)Cl, Me0 2C Z Z -20°C to rt 4 6 71% 4 4 13 reported in the literature.156 The melting point of 44 and the derived spectral data compared favorably with literature values.15b 1.2 Addition of lithium (trimethylstannyD(cyano)cuprate to compounds 42. 43 and 44 Earlier work in our laboratory has demonstrated that lithium(trimethylstannyl)(cyano)-cuprate16 [Me3SnC.uCNl.Li (48) is readily prepared by reaction of trrmemylstarmylUthium17 (47) with copper(I) cyanide (eq 12). THF, -48°C Me3SnLi + CuCN • [Me3SnCuCN]Li (12) 47 48 Recent investigations have demonstrated18 that the addition of the cuprate 48 to a,P~ acetylenic esters is highly regio- and stereoselective and that the stereoselectively can be controlled effectively by the judicious choice of reaction conditions. Thus, reaction of 48 (~1 eq), with, for instance, ethyl 2-pentynoate (49) at -48°C in THF for a short time period (~1 h), followed by warming to 0°C and work-up, produces ethyl (Z)-3-trimethylstannyl-2-pentenoate (50) in good yield (eq 13).18 However, when the reaction of ethyl 2-pentynoate (49) with 48 (1.3 1) [Me3SnCuCN]Li (48), THF, -48°C to 0°C . H N Z Z CQ 2Et — W (13) 2) NH4CI -NH4OH Me3Sn C0 2Et 49 72% 50 14 eq) is carried out at low temperature (-78°C) in the presence of a proton source (EtOH), ethyl (F)-3-trimethylstannyl-2-pentenoate (51) is obtained (eq 14).18 V — = — C 0 2 E t 49 1) [Me3SnCuCN]Li (48), EtOH, THF, -78°C » 2) NH4CI-NH4OH 74% C0 2Et (14) Me3Sn H 51 A reaction scheme that rationalizes the formation of both the (E)-isomer (51) and the (Z)-isomer (50), is shown in Scheme 3.19 V — [Me3SnCuCN]Li C0 2Et 49 48 n Li C0 2 Et Me3Sn Cu(CN) 52 H work-up Me3Sn C0 2Et 50 Me^ Sn >=•={ OCu(CN) OEt 53 SCHEME 3 C0 2Et Me3Sn H 51 Thus, it has been proposed19 that cis addition of the cuprate 48 to the ester 49 leads to a vinylcopper(I) intermediate of type 52. At low temperatures (-78°C) in the presence of a proton source (EtOH), it is believed that the rearrangement of this intermediate to the corresponding 15 allenoate 53 is slow and that 52 will consequently be protonated with retention of configuration to give rise to the (£)-isomer 51 almost exclusively. Based on experimental results, it has also been proposed19 that, at higher temperatures (-48° to 0°C), rearrangement of the vinylcopper(I) species 52 to the allenoate 53 occurs more readily. Thus, if the addition reaction of 48 to 49 is conducted at a higher temperature, the intermediate 52 rearranges to the allenoate 53. An aqueous work-up, causing protonation of the allenoate on the face opposite the bulky Me3Sn group, leads to the stereoselective formation of the (Z)-isomer 50. Since the (^-configuration was the desired stereochemistry for the alkenyltrimethyl-stannanes 21-23, the additions of the cuprate 48 (1.0-1.3 equiv) to the dialkyl dialkynedioates of general structure 40 were carried out at low temperature (-78°C, THF) in the presence of a proton source ROH (methanol for methyl esters, ethanol for ethyl esters) (eq 15). R0 2 C Me3Sn H R 0 2 C - = 1) [Me3SnCuCN]Li, THF, ROH, -78°C 39 C0 2 R + (15) 2) NH4CI-NH4OH 40 H Me3Sn SnMe3 H 54 The results of these addition reactions are summarized in Table I. After some initial 16 experiments, it was found that reactions run with an excess of cuprate (-1.3 equiv) frequently produced a significant amount (-25%) of the bis-adduct (general structure 54) in addition to the desired product. This side reaction was successfully rriinimized by using an excess of starting material (1.3-1.5 equiv; entries 1 and 2, Table I). Table I Addition of [Me3SnCuCN]Li to the diesters 42,43 and 44 entry substrate (general structure 40) n R equivof cuprate equivof substrate reaction time(h) product Ojeneral structure 39) % yield" % diester recovered* 1 42 1 Me 1.0 1.5 1.5 21c 87 15 (33) 2 43 2 Et 1.0 1.3 2.0 22c 51 35 (23) 3 44 3 Me 1.3 1.0 1.25 23d 48 32 (0) "Yield of purified product. ^Parentheses indicate the theoretical amounts of starting material to be recovered. The crude products contained minor amounts (-2-8%) of bis-addition products (general structure 54). Removal of these side products was accomplished by flash chromatography. dThe crude product contained -18% of bis-addition product (general structure 54), which was removed by flash chromatography. Running the addition reactions with excess substrate necessitated the recovery of any surplus starting material, as well as a comparison of the expected (theoretical) amount with the actual amount isolated. In the first addition reaction (entry 1, Table I), yields of the mono-adduct 21 were high and the amount of recovered starting material was somewhat lower than expected. The reactivity of substrate 43 proved to be quite different from that of 42 and produced much lower yields of the mono-adduct 22. Furthermore, more starting material than predicted was recovered (entry 2, Table I). Use of excess substrate in the addition of the cuprate 48 to the diester 44 demonstrated an even greater decrease in substrate reactivity. A high percentage of 17 compound 44 was recovered (-51-76%) and only low yields of the desired product 23 (-24-35%) were isolated. This trend seems to indicate that an increase in the carbon chain length connecting the two acetylenic esters functions causes a concomitant decrease in the reactivity of the dialkyl dialkynedioates toward reaction with the cuprate reagent 48. Reasons for these observed changes in the reaction patterns are not immediately obvious, although it could be argued that an increase in the distance between the two a,P-acetylenic ester functions decreases the electrophilicity of both. In any case, this decreased reactivity made it necessary to conduct the last addition reaction with an excess of the cuprate 48 to obtain even moderate yields of the mono-adduct 23 (entry 3, Table I). A concurrent increase in the amount of bis-adduct produced could not be avoided. The presence of the expected functional groups in the addition products 21-23 was ascertained by ir spectroscopy. Ir spectra of the products displayed prominent triple bond (C=C) stretches in the vicinity of 2239 cm"1, weak double bond (C=C) stretches in the area of 1596 cm"1 and strong carbonyl (C=0) stretches at -1718 cm"1. Structural assignments were also based on ^ and 1 3 C nmr spectroscopic data. Each X H nmr spectrum exhibited the expected olefinic proton signal (=CH) at 8 « 6.0, a signal due to the trimethylstannyl moiety at 8 ~ 0.2 and two distinct methoxy (-OCH3) or ethoxy (-OCH2CH3) peaks in the regions of 8 ~ 3.7 and 8 ~ 4.1, respectively, accounting for the nonequivalent ester functions. Notable in the 1 3 C nmr spectra were the signals due to the acetylenic carbons (C^C) which resonated at 8 = 73 and 8 ~ 89. Determination of the configuration of the carbon-carbon double bonds in compounds 21-23 relied heavily on the value of the coupling constant ( 37sn-H) between the a-olefinic proton (Ha) 18 and the tin atom of the trimethylstannyl (MesSn) function (Figure 1). It is known20 that when a trialkylstannyl moiety and a proton are vicinal on a C=C bond, the 37SD-H value is larger when these moieties are trans (-120 Hz) than when they are cis (-70 Hz). Since the values of the tin-proton coupling constants for the addition products 21-23 fell within the expected limits (69-73 Hz) for a system in which the MesSn function and the olefrnic oc-proton are cis, the configurations could be assigned with confidence. R0 2C = 69-73 Hz Figure 1 3/sn-H coupling constants As it is frequently not possible to observe molecular ions for trimethylstannyl compounds in electron impact mass spectra,21 high resolution mass spectrometric measurements confirming the expected molecular formulae were taken on the (M + - Me) peaks for products 21-23. 19 2. Stereoselective syntheses of (E.E)- 1.2-bis(alkoxycarbonylmethylidene)cycloalkanes of general structure 55 2.1 Copperd) chloride-mediated cyclization of compounds 21. 22 and 23 Having successfully completed the synthesis of compounds 21, 22 and 23, the copper(I) chloride-mediated cyclization of these substrates was investigated (eq 16). The results of these experiments are provided in Table II. C 0 2 R 39 55 Table H Preparation of (Z?,E)-l,2-bis(alkoxycarbonylinethylidene)cycloalkanes 24,25 and 26 entry substrate (general struc-ture 39) n R equivof CuCl reaction time (min) product (general struc-ture 55) % yield" 1 21 1 Me 2.5 5 24 95 2 22 2 Et 2.5 10 25 94 3 23 3 Me 2.5 60 26 74 "Yield of purified product Treatment of a cold (0°C) solution of dimethyl (F)-3-trimethylstannyloct-2-en-6-ynedioate 21 in anhydrous DMF with copper(I) chloride (2.5 equiv) resulted in a rapid, clean reaction that 20 consumed the starting material in 5 min. After a work-up with aqueous NH4CI-NH4OH (pH = 8), chromatography of the crude product on silica gel and recrystallization of the acquired solid, (E,E)-l,2-bis(memoxycarbonylmethyUdene)cyclobutane 24 was isolated in a 95% yield (entry 1, Table II). The structure of 24 was confirmed by 'H and 1 3 C nmr spectroscopy. The X H nmr spectrum of 24 displayed the expected signals due to the methoxy (-OCH3), methylene (-CH2) and olefinic (=CH) protons, which appeared as clean singlets due to the symmetry of the molecule. The symmetrical nature of the structure of 24 was also shown by the presence of only 5 non-equivalent carbon atoms in the 1 3 C nmr spectrum. Although this spectral data precludes the possibility that the isolated compound was the non-symmetrical (£,Z)-isomer 24a, the possibility that the symmetric (Z,Z)-isomer 24b was formed must be addressed. CuO H H 56 57 58 H H Me0 2C H 24a 24b 21 Examination of the proposed reaction mechanism (Scheme 1, p 8) shows that formation of structure 24b would necessitate not only the isomerization of the transmetalation intermediate 56 from the (E)- to the (Z)-configuration, but also the isomerization of the vinylcopper(I) intermediate 57 to the allenoate 58, followed by stereoselective protonation of the more open face of 58. The fact that the vinylcopper(I) intermediate, formed as a result of the intramolecular 1,4-addition, will isomerize at least partially to the allenoate under appropriate reaction conditions was demonstrated when the cyclization of substrate 21 was conducted with copper(I) cyanide in DMSO at 60°C (eq 17). Two products were obtained as a partially separable mixture which were identified as compound 24 (major product) and compound 24a (minor product) based on ^ nmr spectroscopic data. Thus, formation of the allenoate and subsequent stereoselective protonation is indeed a possibility; however, to obtain compound 24b, isomerization of the double bond in the transmetalation intermediate 56 must also occur. Since recent experimental evidence from our laboratory22 suggests very strongly that intermediates of type 56 are configurationally stable, this type of configurational change seems highly unlikely. It may consequently be concluded that the 22 (Z,Z)-isomer 24b is not formed in the copper(I) chloride-mediated cyclization of 21 and that the structure of the product derived from this process is 24. The ir spectrum of 24 displayed a strong C=0 stretch at 1705 cm"1, as well as a C=C stretch at 1652 cm"1. The high resolution mass spectrum confirmed that compound 24 has a molecular formula of C10H12O4. Further investigation demonstrated that the yield of the cyclization reaction was affected significantly by the reaction time. If, for example, after all the starting material had been consumed, the reaction mixture was stirred for an additional 15 min, the isolated yield of the pure product 24 was only 40%. Stirring the mixture for 30 min resulted in a 13% yield. Careful purification of these mixtures by flash chromatography provided no identifiable side products to account for the low yields. However, in each case, a considerable amount of polar baseline material was detected on the chromatography column. When this baseline material was removed from the column by elution with methanol and concentrated, X H nmr spectroscopy of the isolated material revealed a complex mixture containing no identifiable products. The decreased yields observed with longer reaction time may reflect an inherent instability of the vinylcopper(I) intermediate formed in this reaction, decomposition of which could lead to the observed baseline material. Cyclization of diethyl (E)-3-trimethylstannylnon-2-en-7-ynedioate 22 under conditions analogous to those employed for the ring closure of 21 also proceeded quickly and efficiently and produced (£,E)-l,2-bis(ethoxycarbonylmethylidene)cyclopentane 25 in a 94% yield (entry 2, Table II). 23 The reaction time required (10 min) for complete conversion of 22 into 25 was longer than that needed for the cyclization of 21 to 24. Again, it was noted that continued stirring of the reaction mixture after disappearance of the starting material adversely affected the yield and produced chromatographically polar material similar to that obtained in the previous cyclization reaction. *H and 1 3 C nmr spectra confirmed the structural assignment for compound 25. In the XH nmr spectrum, the central methylene group (-CH2CH.2CH2-) on the five-membered ring appeared as a quintet with a coupling constant of 7.5 Hz, while the allylic methylene protons (=C-CH2) appeared as multiplets and the olefinic protons as singlets. The 1 3 C nmr spectrum showed 7 non-equivalent carbon signals, which served to confirm the symmetrical nature of the product 25. In the ir spectrum of 25, the C=0 and C=C stretches were found at 1713 cm"1 and 1649 cm _ 1, respectively. Mass spectrometric measurements confirmed that 25 has a molecular formula ofCi3H1 804. The copper(I) chloride-mediated cyclization of 23 to produce (E,E)-1,2-bis(methoxycarbonylmethylidene)cyclohexane 26 was quite sluggish. A full 60 min was needed for complete disappearance of the starting material. Nevertheless, 26 was produced in an acceptable yield of 74% after purification of the crude product. 2 4 Notable in this reaction is that the yield does not appear to be reduced significandy by the increased reaction time. Purification on silica gel of the crude sample after a reaction time of 60 23 26 min produced only a small amount of polar baseline material on the silica gel column. Drawing on the earlier proposal that the chromatographically polar material is a result of the decomposition of the vinylcopper(I) intermediates, the results of this ring-closure reaction may indicate an increased stability of this intermediate in the six-membered ring case. The structure of compound 26 was confirmed in a manner similar to that described for compounds 24 and 25. Both the X H and the 1 3 C nmr spectra of 26 demonstrated the symmetry of the structure, while the ir spectrum of 26 displayed the expected C=0 and C=C stretches at 1719 cm"1 and 1635 cm"1. The molecular formula of 26 (Ci 2Hi 60 4) was confirmed by high resolution mass spectrometry. When the reaction time of the CuCl-mediated cyclization of the substrate 23 was decreased to 7 min, a mixture of two products was isolated in a ratio of -2:1 (glc). This mixture, which could be partially separated by reversed-phase23 column chromatography, consisted of the expected 1,4-addition product 26 (minor product) and compound 61 (major product) (Figure 2). The structure of 61 was confirmed by lH nmr spectroscopy and nOe difference experiments. The *H nmr spectrum of 61 showed a signal due to a trimethylstannyl (MesSn) 25 group (a 9-proton singlet at 5 0.14, 27SN-H = 55 Hz), as well as two distinct methoxy (-OCH3) signals (two 3-proton singlets at 8 3.68 and 3.69) and one olefinic (=CH) signal (a 1-proton signal ^ S n M e s 5 = 0.14 8 = 5.69 J Figure 2 nOe difference experiments on compound 61 at 8 5.69). The C nmr spectrum of compound 61 showed 13 distinct carbon signals, including a signal at 8 -7.1 due to the methyl groups of the trimethylstannyl function. In nOe difference experiments (Figure 2) irradiation of the olefinic proton resonance at 8 5.69 produced an enhancement of the trimethylstannyl signal at 8 0.14. Conversely, irradiation of the trimethylstannyl resonance caused an enhancement of the olefinic proton signal. Significant in the ir spectrum of compound 61 were the C=0 stretch at 1719 cm"1 and two C=C stretches at 1636 and 1600 cm"1. A high resolution mass spectrometric measurement on the (M + - Me) peak of compound 61 confirmed the molecular formula of C14H21O4. A possible pathway leading to the formation of 61 is shown in Scheme 4. It is proposed that the transmetalation intermediate 59 is involved in two competing reactions. One such reaction is the expected intramolecular conjugate addition to form the vinylcopper(I) intermediate 60. Upon work-up, this intermediate is protonated to give the desired product 26. It is 26 tentatively proposed that the other reaction involves the expected conjugate addition to the triple bond, followed by a copper-tin-transmetalation to produce compound 61 and regenerate CuCl. 60 61 | work-up COoMe C0 2 M e 26 SCHEME 4 27 The experimental observation that an extended reaction time leads to the complete disappearance of 61 may be explained by reaction of 61 with the remaining CuCl to produce the vinylcopper(I) species 60 which is subsequently protonated. To determine more precisely the mechanistic details of this cyclization, further investigations are necessary; however, these were not carried out due to time constraints. Future studies should include exposure of a purified sample of 61 to copper(I) chloride (2.5 equiv) under standard reaction conditions (DMF, 0°C) to determine whether or not 61 is transformed efficiently into 26. 28 3. Preparation of the alkenyltrimethylstannanes 31. 32, 33 and 37 Having successfully completed the copper(I) chloride-mediated intramolecular cyclization reactions of substrates 21-23, our attention turned to the preparation of the structurally more complex systems 31-33 and 37. The individual steps of these syntheses are presented in a retrosynthetic manner in Schemes 5-7. To prepare the functionalized alkenyltrimethylstannane 31, we envisaged reacting methyl 3-lithiopropynoate24 (62) with the aldehyde 63. It was thought that 63 could be produced readily by applying standard transformations, including the oxidation of the alcohol 64 and the reduction of the a,P-unsaturated ester 50 (Scheme 5). To stereoselectively synthesize the ester 50 from commercially available ethyl 2-pentynoate (49), the methodology18 discussed previously (p 13) would be applied. 49 50 S C H E M E 5 29 The synthesis of substrates 32 and 33 (general structure 65) can be retrosynthetically analyzed in a fashion similar to that presented above. To produce 65, methyl 3-lithiopropynoate24 (62) would be added to the aldehyde of general structure 66. As in the previous case, it was believed that the aldehyde could be derived from the corresponding alcohol 67 which, in turn, would result from the reduction of the ester of general structure 68 (Scheme 6). To prepare 68 from commercially available starting material (69) the application of literature procedures was planned.25 69 68 S C H E M E 6 The retrosynthesis of compound 37 (Scheme 7) in principle parallels that of substrates 32 and 33. However, it includes one additional step, namely the preparation of compound 72. To 30 synthesize 72, it was decided to attempt the deconjugation-alkylation of the vmyltrimethylstannyl ester 68b by applying a procedure developed in our laboratory.8b 69b 68b SCHEME 7 In the following sections, a more detailed discussion of the preparation of the precursors leading to substrates 31-33 and 37 will be presented. 3.1 Preparation of the (3-trimemylstannyl a.ft-unsaturated esters 68a. 68b and 50 and the [3-u-imethylstannyl ft.y-unsaturated ester 72 , SnMe3 SnMe3 - ^ ^ x SnMe3 68a 68b 72 50 31 To synthesize the esters 68a and 68b, a procedure developed in our laboratory22'25 was applied with minor modifications (Scheme 8). O 69 c o R 1) KH, THF, 0°C 2 2)PhNTf 2 ,0°Ctort a n = 1, R = Me b n = 2, R = Et TfO y^^ C02R n 73 [Me3SnCu(SPh)]Li (74) THF-DMPU, -20°C 68a, 74% 68b, 69% SnMe3 ^ ^ C 0 2 R n 68 SCHEME 8 The vinyl triflates 73a and 73b were prepared from commercially available 2-methoxy-carbonylcyclopentanone (69a) and 2-ethoxycarbonylcyclohexanone (69b), respectively, by treat-ment of the keto esters with potassium hydride and A -^phenyltrifluoromethanesulfonimide (PhNTf2). Reaction of these triflates with lithium (phenyltMo)(trimethylstannyl)cuprate16 (74) provided, after a non-aqueous work-up and purification of the crude materials by chromatography on silica gel, the desired esters 68a and 68b. Isolated yields and spectral data derived from the purified products were in accordance with values reported in the literature.25 32 The preparation of compound 72 involved the deconjugation-alkylation of ethyl 2-trimethylstannyl-l-cyclohexenecarboxylate (68b). Studies by Rathke26a and Schlessinger26b have shown that the alkylation of dienolate anions, generated by treatment of various ot^ -unsaturated esters with LDA, occurs preferentially in the a-position. Based on molecular orbital calculations,26b this observed preference has been attributed to a maximization of the negative charge on the oc-carbon of these dianions. This type of selectivity was also observed in our case. A solution of the ester 68b in THF was added to a solution of an LDA-DMPU complex (1.5 equiv) in THF and the mixture was stirred (-78°C, 30 min; 0°C, 50 min) affording a yellow solution of the corresponding Uthium dienolate anion (75). Cooling the resulting mixture to -20°C, followed by the addition of iodomethane (-20°C, 50 min) produced compound 72 in an 84% yield after purification of the crude product by chromatography on silica gel (eq 18). ( 1 8 ) 68b 75 OEt Mel, -20°C 84% 72 The final substrate, ethyl (Z)-3-trimethylstannyl-2-pentenoate 50, was prepared readily according to literature procedures18 by reaction of commercially available ethyl 2-pentynoate 49 with the cuprate reagent 48 under conditions appropriate for the formation of the (Z)-isomer 33 (THF, -42°C -> 0°C). The geometry of the carbon-carbon double bond was confirmed by the large value of the tin-proton coupling constant (37SN-H =120 Hz) between the alkenyl proton and the tin atom. This value lies within the expected range for a system where the a-olefinic proton and the MejSn group are trans. The yield of this reaction and spectral data derived from 50 matched values cited in the literature16'18 (eq 19). 1) [Me3SnCuCN]Li (48), v THF,-42°C to 0°C V N = C0 2Et • ) = \ (19) 2) NH4CI-NH4OH M e 3 S n C ° 2 E t 49 50 3.2 Reduction of the esters 68a. 68b, 72 and 50. Preparation of the alcohols 67a. 67b, 71 and 64 67a 67b 71 64 Reduction of the esters 68a, 68b, 72 and 50 to the respective alcohols was achieved in a straightforward manner by utilizing DJBAL (2.5 equiv) in either THF (-78°C, 60 min; room temperature, 30 min) or E t 2 0 (0°C, 10-15 min). The results of these reactions are summarized in Table HI. 34 The esters 68a and 68b were reduced smoothly with DIBAL in THF to give the corresponding alcohols in excellent yields (entries 1 and 2, Table III). For the reductions of the esters 72 and 50, it was found that changing the solvent from THF to Et20 reduced reaction times (entries 3 and 4, Table HI). Table m Reduction of the esters 68a, 68b, 72 and 50 to give the alcohols 67a, 67b, 71 and 64 entry substrate procedure" product % yield6 1 68a A 67a 91 2 68b A 67b 88 3 72 B 71 94 4 50 B 64 72 "Procedure A: DIBAL (2.5 equiv, 1.0 M solution in hexanes), THF, -78°C, 60 min; rt, 30 min. Procedure B: DIBAL (2.5 equiv, 1.0 M solution in hexanes), Et20, 0°C, 10-15 min. *Yield of purified product. Structural assignments for the alcohols 67a, 67b, 71 and 64 were confirmed by ir and X H nmr spectrometric data. For example, alcohol 67a displayed a broad -OH stretch at 3365 cm"1 as well as a medium C=C stretch at 1616 cm"1 in the ir spectrum. The lH nmr spectrum included a signal attributed to the Me3Sn group (a 9-proton singlet, 27SN-H = 54 Hz, at 8 0.12), a peak assigned to the -OH group (a 1-proton triplet, 7 = 5 Hz, at 8 1.48) and a signal due to the hydroxymethyl protons (a 2-proton doublet, 7=5 Hz, at 8 4.19). The spectral data derived from the remaining alcohols 67b, 71 and 64 were analogous to those described above for compound 67a. In addition to the diagnostic Me3Sn and -CH 2OH signals in the X H nmr spectrum of alcohol 71, peaks attributed to the Me group (a 3-proton singlet 35 at 8 0.95) and the olefinic proton (a 1-proton triplet, J = 3.5 Hz, at 8 5.97) were also observed. Notable in the *H nmr spectrum of compound 64 was a signal due to the olefinic proton (a 1-proton triplet, at 8 6.20, J = 6.5 Hz, 37SN-H = 138 Hz). The molecular formula of each alcohol was confirmed by high resolution mass spectrometric measurements on the (M + - Me) fragments. 3.3 Oxidation of the alcohols 67a. 67b. 71 and 64. Preparation of the aldehydes 66a, 66b, 70 and 63 66a 66b 70 63 Oxidation of the alcohol 67a to the corresponding aldehyde 66a was accomplished through standard Swern27 conditions (Procedure A, Table IV), while the remaining alcohols were oxidized efficiendy with tetra-n-propylammonium perruthenate (TPAP, procedure B, Table IV).28 Results of the oxidation reactions are summarized in Table IV. Although the alcohol 66a responded well to Swern oxidation conditions (entry 1, Table IV), it was discovered that identical reaction conditions produced only 52% of the desired oxidation product when alcohol 66b was used as a substrate. In this case, ~35% of the destannylated side product 76 was also isolated. 36 Table IV Preparation of the aldehydes 66a, 66b, 70 and 63 entry substrate procedure" product % yield* 1 67a A 66a 80% 2 67b B 66b 80% 3 71 B 70 88% 4 64 B 63 74% "Procedure A: substrate added to a solution of oxalyl chloride (1.5 equiv) and DMSO (2.4 equiv) in CH2C12) -78°C; stirred 15 min; Et3N added, warmed to rt, H 20 added; stirred 10 min. Procedure B: substrate added to s suspension of NMO (1.5 equiv) and 4 A activated molecular sieves in CH2C12, followed by TPAP (0.1 equiv); stirred at rt for 15-35 min. *Yield of purified product. O 76 The difference in behaviour of the substrates 66a and 66b under identical reaction conditions is probably a result of relative product stability. Control experiments have shown29 that an increase in ring size causes a vmyltrimethylstannyl function to become more labile. Thus, in the presence of small amounts of acid, as may be produced by the decomposition of oxalyl chloride, protodestannylation30 of the trimethylstannyl function occurs more readily with the 6-membered ring aldehyde 66b than in the case of the 5-membered ring system 66a. Fortunately, this problem was readily overcome by switching to the mild conditions provided by a TPAP oxidation. The remaining alcohols were all oxidized quickly and efficiently under these conditions to produce the desired aldehydes in good to excellent yields (entries 2-4, Table IV). 37 Structural assignments for compounds 66a, 66b, 70 and 63 were confirmed by X H nmr and ir spectrometric data. Aldehyde 66a, for instance, exhibited a strong C=0 stretch at 1727 cm"1 and a weak C=C stretch at 1673 cm"1 in the ir spectrum. The X H nmr spectrum displayed the distinct Me3Sn signal at 8 0.27 (a 9-proton singlet, 27SN-H = 56 Hz), as well as a signal at 8 9.75 (a 1-proton singlet) attributed to the aldehyde function. Similar diagnostic spectral characteristics were observed for each of the remaining three aldehydes 66b, 70 and 63. As in previous cases, the molecular formulae of the substrates were confirmed by high resolution mass spectrometric measurements on the (M + - Me) fragments. 3.4 Preparation of methyl 3-lithiopropynoate (62)24a The lithium acetylide of commercially available methyl propynoate (77) was prepared readily by treatment of 77 with LDA (2 equiv) and DMPU (2 equiv) in THF at low temperature (-78°C, 10 min) (eq 20).24a LDA, THF, M e 0 2 C — = - H • M e 0 2 C — = — L i (20) DMPU, -78°C 77 62 Despite reports246 that temperatures of <-100°C must be maintained to achieve adequate addition of 62 to aldehyde functions, anion 62 proved to be stable at -78°C and reacted efficiently with the aldehyde substrates. 38 3.5 Addition of methyl 3-lithiopropynoate to the aldehydes 66a. 66b. 70 and 63 32 33 Addition of methyl 3-lithioproynoate 62 to the aldehyde substrates 66a, 66b, 70 and 63 proceeded efficiently to provide the desired products in good yields (Table V). Table V Synthesis of the alkenyltrimethylstannanes 32,33,37 and 31 entry substrate product % yield0 1 66a 32 86 2 66b 33 85 3 70 37* 65c(20)rf 4 63 31 85 "Yield of purified product. * Addition of 62 to the aldehyde 70 produced two diastereomeric products (37a and 37b) in a ratio of ~ 3:1 respectively (based on isolated yields). These compounds could be separated by radial chromatography on silica gel. c Yield of major isomer (37a). ''Yield of minor isomer (37b). When, for example, 2-trimethylstannyl-l-cyclopentenecarbaldehyde 66a was added to a 39 cold (-78°C) solution of 62 (2 equiv) in THF, the starting material was consumed in 1.5 h to give, upon work-up and purification of the crude material by chromatography on silica gel, 86% of the desired product 32 (entry 1, Table V). Addition of 62 to the aldehyde 70 produced a mixture of diastereomers in a ratio of -3:1 (entry 3, Table V), which could be separated by radial chromatography on silica gel. The structures of compounds 31-33 and 37 were confirmed by ir, J H nmr and 1 3 C nmr spectrometric data. For instance, the ir spectrum of the alkenyltrimethylstannane 32 showed strong -OH (3452 cm"1), C^C (2235 cm"1) and C=0 (1719 cm"1) stretches, as well as a weak C=C stretch at 1619 cm"1. The lH nmr spectrum indicated the presence of the MesSn group (a 9-proton singlet, 27S„-H = 54 Hz, at 8 0.16) and of the -OH function (a 1-proton doublet, 7=5 Hz, at 8 2.00). Also observed were signals due to the ester group (a 3-proton singlet at 8 3.76) and the carbinol proton (a 1-proton doublet, 7=5 Hz, at 8 5.15). Notable in the 1 3 C nmr spectrum of 32 were the signals derived from to the acetylenic carbons at 8 76.5 and 8 86.5. The molecular formula of compound 32 was confirmed by a high resolution mass spectrometric measurement of the (M + - Me) fragment. Structural assignments for the remaining alkenyltrimethylstannanes were also based on spectrometric data and were conducted in a similar manner. High resolution mass spectrometric measurements of the (M + - Me) fragment on each compound confirmed the expected molecular formula. Specific configurations for the diastereomers 37a and 37b were not determined, but a general structure was assigned with the aid of J H nmr, 1 3 C nmr and ir spectral data. 40 4. CopperfD chloride-mediated intramolecular conjugate additions of the alkenyltrimethyl-stannanes 31. 32 and 33 4.1 CopperfT) chloride-mediated cyclization of methyl (ZV4-hydroxy-6-trimethylstannyl-5-octen-2-ynoate 31 34 Preparation of the highly functionalized cyclobutene 34 was accomphshed by treatment of a solution of compound 31 in cold (0°C) DMF with CuCl (2.5 equiv). Stirring of the yellow mixture (0°C, 2 min), followed by work-up and purification of the crude material by chromatography on silica gel, produced the product 34 as a clear liquid, as well as a second compound 78 as a colourless solid (eq 21). Separation of the two products was achieved by flash column chromatography on silica gel and resulted in the isolation of 75% of ester 34 and 16% of compound 78. The expectation that the minor product possessed the structure proposed above was confirmed with the aid of ir and nmr spectroscopy. The ir spectrometric data for the two products was quite similar, but differed in the position of the -OH stretches (3490 cm"1 for 34; 3389 cm"1 for 78) and the C=0 stretches (1695 cm"1 for 34; 1719 cm"1 for 78). The difference in these vibrational frequencies provided the first clue to the structural differences of the two 41 products. In the cyclobutene 34, the occurrence of intramolecular hydrogen bonding is likely to affect the stretching frequencies of the functional groups involved (Figure 3). When hydrogen bonded, the C=0 stretching frequency, in particular, is expected to move to a lower wave number, since less energy is required to achieve bond vibration. As seen from the experimental data, such a shift in wave number is indeed observed for compound 34. Since this type of hydrogen bonding is not possible in 78, the carbonyl stretching vibration for this substance occurs at a higher wave number. Variations were also observed in the X H nmr spectra of the two products. For example, the -OH signal for compound 34 appeared as a broad 1-proton singlet at 8 4.02, while the carbinol proton (Hd) displayed a 1-proton singlet at 8 5.14. In contrast, the -OH signal in the spectrum of product 78 was seen as a 1-proton doublet at 8 1.79 (J = 9.5 Hz) which was coupled to the carbinol proton (a 1-proton doublet, J = 9.5 Hz, at 8 4.81). The observation that the signal for the -OH group of compound 34 is further downfield than the corresponding signal in compound 78 is significant, since it is known31 that hydrogen bonding moves the proton absorption of the hydroxylic proton to lower field by decreasing the electron density around the Figure 3 Intramolecular hydrogen bonding in compound 34 42 proton. Thus, the X H nmr spectrometric data provides further evidence for the proposed configurations of compounds 34 and 78. In order to firmly establish the structure of the major product 34, two more experiments were carried out, namely a nOe difference experiment (Figure 4) and an HMQC experiment. Irradiation of the signal due to the olefinic proton (Ha) in compound 34 led to an enhancement of the resonance derived from the -CH2b protons of the ethyl group. A reverse experiment showed that irradiation of the - C H 2 b protons enhanced the signal due to the olefinic proton H a . These observations are consistent with the structure proposed for compound 34. The HMQC experiment confirmed the carbon-hydrogen correlations expected for compound 34. The correlation of the proton signal at 8 6.87 to the carbon signal at 8 144.5 was of particular interest, since it confirmed that this signal was due to the olefinic proton Hf. The complete experimental data for the HMQC and the nOe difference experiments can be found in Table VUI (p 96). To verify the molecular formulae of products 34 and 78, high resolution mass spectra were obtained which confirmed that each compound has a molecular formula of C9H12O3. Although the configuration of the exocyclic double bond of the minor product 78 was not H 3 C OH Figure 4 nOe difference experiments on compound 34 43 confirmed by nOe experiments, it is clear from the spectral data that this material is a geometric isomer of 34. Mechanistically, the formation of this product can be rationalized as indicated in Scheme 9. MeO C0 2Me 34 78 34 SCHEME 9 As discussed earlier, isomerization of the vinylcopper(I) intermediates to the corresponding allenoates was not observed with standard conditions (2.5 equiv CuCl, DMF, 0°C), but appeared to occur under more vigorous conditions (CuCN, DMSO, 60°C). However, in the case of substrate 31, the proximity of the -CH2CH3 group and the Cu atom in the intermediate 79 leads to steric congestion, which can be relieved if the species rearranges to the allenoate 80. Thus, it is possible that, in this case, steric encumbrances cause the initially formed intermediate to isomerize at least partially to the allenoate 80. Since the faces of the allenoate are sterically very 44 similar, it is proposed that protonation of 80 would proceed non-stereoselectively to give the cyclized product as a mixture of isomers (34 and 78). 4.2 Preparation of the bicyclic esters 35 and 36 SnMe3 CuCl, DMF y ^ C 0 2 M e C0 2Me (22) 32,35; n = 1 33, 36; n = 2 OH OH 65 81 The copper(I) chloride-mediated cyclization of the alkenyltrimethylstannane 32 proceeded rapidly to produce the bicycle 35 in a 73% yield (eq 22). The reaction involved treatment of a cold (0°C) solution of substrate 32 in DMF with CuCl (2.5 equiv), followed by stirring of the yellow mixture (0°C, 2 min) and subsequent work-up with an aqueous solution of NH4CI-NH4OH (pH = 8). The crude product was purified by flash column chromatography on silica gel to yield a colourless solid which was moderately stable at -5°C, but demonstrated a significant degree of instability when dissolved in EtzO, CDCI3 or other solvents. This presented some difficulties during the collection of spectral data, since the sample partially decomposed during extended acquisition times (e.g. 1 3 C nmr, nOe difference experiments). Nevertheless, spectra adequate for the structural assignment of 35 were obtained. The ir spectrum of 35 exhibited peaks attributed to the -OH group (a broad stretch at 3428 cm"1), the C=0 function (a strong stretch at 1718 cm"1) and a C=C bond (a medium stretch at 1604 cm"1). 45 The presence of the C0 2Me and hydroxyl functions was confirmed by the *H nmr spectrum of 35 which displayed a 3-proton singlet at 8 3.67, derived from the methoxy group and a 1-proton singlet at 8 3.70 assigned to the -OH group. Also observed were signals due to the carbinol proton (a 1-proton singlet at 8 5.01) and the olefinic proton (a 1-proton singlet at 8 5.40). In nOe difference experiments (Figure 5), irradiation of the allylic methylene group (-CH2 a, 8 2.26) produced an enhancement of the olefinic proton (=CHb) signal, as well as of the resonance due to the adjacent methylene group (-CH2°). Conversely, irradiation of the olefinic proton produced an enhancement in the signal derived from the allylic methylene group. Figure 5 nOe difference experiments on compound 35 Although the instabihty of compound 35 at room temperature precluded the acquisition of a satisfactory elemental analysis, high resolution mass spectrometric measurements confirmed the molecular formula of CioHi203. Cyclization of the allcenyltrimethylstannane 33 to give the bicyclic product 36 proceeded rapidly under reaction conditions analogous to those employed for the ring-closure of compound 32. The bicyclic ester 36 was isolated in a 75% yield after purification of the crude product by 46 chromatography on silica gel. The colourless solid displayed greater stability at room temperature and in solution than its 5-membered ring analogue. Structural assignments for 36 were based on ir and lH nmr spectral data, as well as on nOe difference experiments performed on the benzoyloxy derivative 82. A nOe difference experiment on the non-derivatized compound 36 was not possible due to the proximity of the carbinol proton signal (a 1-proton singlet at 8 5.21) to the olefinic proton signal (a 1-proton singlet at 8 5.23) which made selective irradiations difficult. Significant peaks in the J H nmr spectrum of 36 included the aforementioned carbinol and olefinic proton signals, as well as a peak due to the ester function (a 3-proton singlet at 8 3.71) and a signal attributed to the -OH group (a broad 1-proton singlet at 8 4.13). The ir spectrum confirmed the presence of an -OH function (a broad stretch at 3472 cm"1) and a carbonyl group (a medium stretch at 1699 cm"1). Also observed was a stretch due to a double bond at 1601 cm"1. In addition to the high resolution mass spectrum, which confirmed that the bicycle 36 has a molecular formula of C 1 1 H 1 4 O 3 , a successful elemental analysis was obtained. The benzoyloxy derivative of 36 was obtained by treatment of compound 36 with DMAP (0.1 equiv) and benzoyl chloride (1.1 equiv) in CH2CI2 (0°C, lh; room temperature, 1.5 h) (eq 23). (23) 36 30% 82 47 Spectral data for compound 82 were similar to those of 36, but did include additional stretches in the ir spectrum attributed to the benzoyl function (1200-750 cm"1), as well as signals in the aromatic regions (8 7.41-8.01) of the *H nmr and 1 3 C nmr spectra (8 128.3-132.9). In a nOe difference experiment (Figure 6), irradiation of the allylic methylene group (-CH 2\ 8 2.16) produced an enhancement of the resonances due to the adjacent methylene groups (-CH 2 b-CH 2 c, 8 1.62-1.86) and of the olefinic proton signal (=CH, 8 5.32). Unfortunately, reverse nOe difference experiments proved inconclusive, since the enhancement of the allylic methylene group signal which was observed on irradiation of the olefinic proton was too minor to serve as structural confirmation. Figure 6 nOe difference experiment on compound 82 4 8 5.0 A limitation of the copper (D chloride-mediated conjugate addition method 5.1 Attempted cyclization of the alkenyltrimethylstannanes 37a and 37b SnMe3 C0 2Me y ^ C 0 2 M e (24) OH 37 38 When the alkenyltrimethylstannane 37a was exposed to standard cyclization conditions (2.5 equiv CuCl, DMF, 0°C), a dc analysis taken on an aliquot of the reaction mixture, which had been quenched with aqueous NH4CI-NH4OH (pH = 8) and extracted with Et20, revealed that no detectable reaction had occurred after a period of 20 min (entry 1, Table VI). In an attempt to force the reaction to proceed, one additional equivalent of CuCl was added to the mixture; however, after 5 more minutes at 0°C, still no detectable reaction had occurred. The reaction mixture was thus allowed to warm to room temperature and stirred for 5 min, at which point all the starting material had disappeared. After work-up with an aqueous solution of NH4CI-NH4OH (pH = 8) and purification of the crude material by chromatography on silica, 10.2 mg of what was believed to be the product was isolated. A *H nmr spectrum showed that the isolated material consisted of a mixture of two products in a ratio of -1:1. Closer examination of the spectral data revealed the disappearance of the signal due to the trimethylstannyl function and the appearance of three new signals in the olefinic region. One of these signals (a 1-proton triplet at 8 5.94) coincided very closely with the 49 signal attributed to the olefinic proton in the starting material 37a (a 1-proton triplet at 8 5.93) Due to the splitting pattern of this signal, initial proposals that one of the compounds in the mixture was the protodestannylation product were discarded and it was suggested that chlorodestannylation had occurred instead to produce compound 83. A low resolution mass spectrum confirmed the presence of a compound with the molecular formula C12H15O3CI (M/z = 242). It was further proposed that the mixture also contained some of the desired 1,4-addition product 38, based on the positions and splitting patterns of the two remaining olefinic signals (a 1-proton triplet at 8 5.96 and a 1-proton singlet at 8 5.84), as well as the appearance of several new signals, including a 1-proton singlet at 8 4.71 attributed to the -CHOH group and a 1-proton singlet at 8 3.22 assigned to the -OH function. Further evidence in favour of compound 38 was provided by the low resolution mass spectrum which indicated the presence of a compound with molecular formula Ci 2 Hi 6 0 3 (M/z = 208). In an attempt to avoid the chlorodestannylation reaction observed with CuCl, the reaction was repeated with copper(I) cyanide (CuCN) (entry 2, Table VI). Due to the limited solubility of CuCN in most solvents, it was necessary to run the reaction in DMSO at 60°C. After a reaction 83 50 time of 30 min, the starting material had disappeared entirely; however, a tic analysis of the resulting product mixture showed only polar baseline material. The suspicion that no viable products had been formed was confirmed by a *H nmr spectrum on the purified material, which revealed only a complex mixture. Table VI. Reaction conditions for the attempted cyclizations of compounds 37a and 37b entry substrate copper(I) source equiv of copper(I) solvent temperature reaction time result 1 37a CuCl 2.5 + 1.0" DMF 0°C to rt 25 min at 0°C 5 min at rt" two products 2 37a CuCN 2.5 DMSO 60°C 30 min complex mixture 3 37b CuCl 2.5 DMF 0°C 20 min n. r. 4 37b CuCl 2.5 DMF rt l h n. r. 5 37b CuCN 2.5 DMSO 60°C 3h n. r. The substrate was stirred with 2.5 equiv CuCl at 0°C for 20 min; an additional 1.0 equiv of CuCl was added; the reaction mixture was stirred at 0°C for 5 min, then warmed to rt and stirred for 5 min. It was therefore decided to abandon the cyclization attempts of compound 37a and focus instead on the copper(I)-chloride mediated 1,4-addition of the second diastereomer, 37b. Unfortunately, it soon became apparent that substrate 37b seemed to be completely resistant to cyclization under all of the standard conditions (entries 3-5, Table VI). For example, when substrate 37b (30 mg) was stirred with CuCl (2.5 equiv) in DMF at 0°C for 20 min, 79% (23.6 mg) of the starting material was recovered after work-up and purification. Even after extended heating of 37b with CuCN in DMSO, no observable reaction occurred and only starting material (71%) was recovered. 51 It is clear from the results discussed above that the intramolecular conjugate additions of substrates 37a and 37b do not occur readily, if at all. The difficulties encountered when attempting to cyclize these substrates can be attributed to a number of factors. For instance, the proximity of the methyl group to the MesSn function results in a sterically hindered system in which the Cu-Sn transmetalation is slowed down significantly. In addition, steric congestion resulting from the adjacency of the methyl group and the secondary hydroxyl group may make it difficult for the molecules to adopt a conformation in which cis addition to the double bond can occur. However, without further investigations, it is difficult to provide a more in depth explanation for the failure of substrates 37a and 37b to undergo cyclization. Future work in this area should include replacement of the methyl group on the alkenyltrimethylstannanes by a smaller group (e.g. hydrogen) in order to determine whether the substituent size does indeed affect the rate of cyclization. Despite the fact that the desired 1,4-additions of substrates 37a and 37b could not be accomplished, the results obtained from these attempted cyclizations are nevertheless valuable, since they provide a clearer picture of the limitations which may be encountered when applying the methodology presented in this thesis. 52 m. CONCLUSIONS 1. Stereoselective synthesis of compounds containing both vinyltrimethylstannane and a,P-alkynic ester functions The preparation of a number of specifically functionalized and stereochemically homogeneous alkenyltrimethylstannanes (structures 21-23, 31-33 and 37) has been carried out. The alkenyltrimethylstannanes 21-23 were synthesized in good to moderate yields by the 53 stereoselective addition of lithium (trimethylstannyl)(cyano)cuprate 48 to the corresponding diaikyi dialkynedioates 42-44 (eq 15, p 15). 43 [Me3SnCuCN]Li 48 The alkenyltrimethylstannanes 31-33 were prepared by the application of a series of established procedures. For compound 31 these included stereoselective addition of lithium (trimethylstannyl)(cyano)cuprate 48 to ethyl 2-pentynoate (49, eq 19, p 33),18 reduction and oxidation. The product was obtained by treatment of the oxidation product, aldehyde 63, with methyl 3-hthiopropynoate (62). = - C 0 2 E t M e 0 2 C ^ = ^ L i ^ } ^ 49 62 63 To synthesize compounds 32 and 33, vinyl triflates of general structure 73, derived from commercially available starting materials, were converted into the P-trimethylstannyl a,P-unsaturated esters of general structure 68 (Scheme 8, p 31). Reduction, oxidation and addition of the anion 62 to the resulting aldehydes of general structure 66 produced the desired alkenyl-trimethylstannanes 32 and 33 in good yields. 54 TfO SnMe3 SnMe3 CHO 73 68 66 a n = 1, R = Me b n = 2, R = Et The preparation of compound 37 was accomplished by a deconjugation-alkylation sequence on the ester 68b to produce compound 72, followed by reduction, oxidation and addition of the anion 62 to the resulting aldehyde 70. 2. Copper(I) chloride-mediated intramolecular 1,4-additions of vinyltrimethylstannyl functions to a,P-alkynic esters Treatment of each of the six alkenyltrimethylstannanes (21-23 and 31-33) with copper(I) chloride successfully induced the desired intramolecular conjugate addition of the vmyltrimethyl-stannane functions to the a,p%alkynic esters, thus demonstrating that the methodology proposed in this thesis is viable. In addition, this methodology presents an efficient strategy allowing for the facile synthesis of highly functionalized ring systems, as is shown by the preparation of a variety of O 68b 72 70 55 substituted mono- and bicycles (24-26 and 34-36). However, the limitations of the methodology were also shown by the inability to successfully cyclize the more sterically demanding substrate 37. C 0 2 M e C 0 2 E t C 0 2 M e C 0 2 M e C 0 2 E t C 0 2 M e 24 25 26 C0 2Me O OH OH O y ^ C 0 2 M e OH 34 35 36 56 IV. EXPERIMENTAL SECTION 1. General 1.1 Data acquisition and presentation Melting points were determined on a Fisher-Johns melting point apparatus and are uncorrected. Infrared (ir) spectra were obtained on liquid films (sodium chloride plates) or solid pellets (infrared grade potassium bromide) employing a Perkin-Elmer model 1710 Fourier transform spectrometer with internal calibration. Proton nuclear magnetic resonance (*H nmr) spectra were recorded on deuterio-chloroform (CDCI3) solutions using a Bruker model WH-400 (400 MHz) spectrometer. Signal positions (8 values) are given in parts per million and were measured relative to that of chloroform (8 7.24).31 The multiplicity, number of protons, coupling constants (/ values, given in Hz) and assignments (where possible) are indicated in parentheses. The tin-proton coupling constants (7SII-H) are given as an average of the 1 1 7Sn and n 9 Sn values. In some cases, the proton assignments were supported by nOe difference and/or HMQC (^-Detected Heteronuclear Multiple Quantum Coherence) experiments. These experiments were carried out using Brucker models WH-400 or AMX-500 spectrometers. Carbon nuclear magnetic resonance (1 3C nmr) spectra were run on Bruker models AC-200E (50.3 MHz) or AMX-500 (125.8 MHz) spectrometers or on a Varian model XL-300 (75.5 MHz) instrument using deuteriochloroform or acetone-d6 as solvent. Signal positions (8 values) 57 are given in parts per million and were measured relative to those of deuteriochloroform (8 77.0) or acetone-d6 (8 29.8).31 Low and high resolution mass spectra were recorded on a Kratos Concept II HQ or on a Kratos MS 80 mass spectrometer. For compounds containing the trimethylstannyl (MesSn) function, high resolution mass spectrometric measurements are based on 1 2 0Sn and were made on the M + - CH3 peak.21 Microanalyses were performed on a Fisons EA elemental analyzer, model 1108, by the Microanalytical Laboratory, University of British Columbia. Gas-liquid chromatography (glc) was performed with Hewlett-Packard models 5880A or 5890 capillary gas chromatographs, both using flame ionization detectors and fused silica columns. The former instrument contained a -20 m x 0.21 mm column, while the latter chromatograph utilized a -25 m x 0.20 mm column. Both columns were coated with HP-5 (Crosslinked 5% Ph Me silicone). Thin-layer chromatography (tic) was performed with commercially available, aluminum-backed sheets, precoated with silica gel 60 to a thickness of 0.2 mm (E. Merck, type 5554). Visualization of the chromatograms was accomplished with an ultraviolet light and/or with iodine, followed by heating of the chromatogram after staining with a solution of 1% w/v Ceric sulfate and 2% w/v molybdic acid in 10% aqueous H2SO4. Flash chromatography32 was performed with 230-400 mesh silica gel (E. Merck, silica gel 60). Radial chromatography33 was performed on a Chromatotron® Model 7924 using 1, 2 or 4 mm thick radial plates (silica gel 60, PF254, with Gypsum, E. Merck #7749). Cold temperatures were maintained using the following baths: 0°C, ice-water; -10°C, ice-acetone; -20°C, aqueous calcium chloride-dry ice (27 g CaCVlOO raL H 20); 3 4 -42°C, aceto-nitrile-dry ice; -63°C, chloroform-dry ice; -78°C, acetone-dry ice. 58 All reactions were carried out under an atmosphere of dry argon in flame and/or oven (~140°C) dried glassware. Glass syringes, needles and cannulae for handling anhydrous solvents and reagents were oven dried while plastic syringes were flushed with dry Argon before use. Microsyringes were placed under vacuum for 15 minutes and were flushed with dry Argon prior to use. Concentration, evaporation or removal of the solvent under reduced pressure (water aspirator) refers to solvent removal via a Buchi rotary evaporator at -20 Torr. 1.2 Solvents and Reagents All solvents used were dried and distilled using standard procedures.35 Dichloromethane (CH2CI2) was distilled from calcium hydride, while diethyl ether (Et20) and tetrahydrofuran (THF) were distilled from sodium benzophenone ketyl. These solvents were used immediately subsequent to distillation. Diisopropylamine, A^AT-dimethylformamide (DMF), dimethyl sulfoxide (DMSO), dimethylpropyleneurea (DMPU) and triethylamine were distilled from calcium hydride. Methanol (MeOH) and ethanol (EtOH) were stirred over activated 3 A molecular sieves overnight and then distilled. Copper(I) chloride (99.995%+ or 99%) and copper(I) cyanide (99%) were purchased from Aldrich Chemical Co. Inc. and were used without further purification. Deuteriochloroform, methyl chloroformate, iodomethane and benzoyl chloride were passed through a short column of basic alumina activity I, which had been dried in an oven (~140°C ) overnight and then allowed to 59 cool in a desiccator prior to use. Hexamethylditin was obtained from Organometallics, Inc., and was distilled prior to use. Solutions of memymthium in diethyl ether, n-butyllithium in hexanes and diisobutylaluminum hydride (DIBAL) in hexanes were purchased from Aldrich Chemical Co. Inc. and the former two reagents were standardized using diphenylacetic acid as primary standard.36 All other reagents are commercially available and were used without further purification. Aqueous ammonium chloride-ammonium hydroxide (NH4CI-NH4OH, pH = 8) was prepared by the addition of ~50 mL of aqueous ammonium hydroxide (28-30%) to -950 mL of saturated aqueous ammonium chloride. 2.0 Preparation of lithium (trimethylstannyD(cyano)cuprate16 [Me3SnCuCN]Li 48 To a cold (-42°C), stirred solution of hexamethylditin (1 equiv) in dry THF (-10 mL per mmol of hexamethylditin) was added a solution of MeLi (1 equiv) in E t 2 0 . The pale yellow solution of trimemylstannylhthium17 was warmed to 0°C, stirred for 20 min, and subsequently cooled to -42°C. Solid CuCN (1 equiv) was added in one portion and mixture was stirred at -42°C for 20 min to produce an orange solution of lithium (trimethylstannyl)(cyano)cuprate (48). 60 3. Addition of lithium (trimemylstannyD(cyano)cuprate to dialkyl dialkynedioates of general structure 40 General Procedure 1 R Q 2 C - = — ^ 40 = - C 0 2 R RC^C n = 1, 2, 3 R = Me, Et GOzR To a cold (-78°C) stirred solution of [Me3SnCuCN]Li (48) (1.0-1.3 equiv) in dry THF (~5 mL per mmol of the cuprate) was added dry MeOH (1.0-1.5 equiv for methyl ester substrates) or dry EtOH (1.3-1.5 equiv for ethyl ester substrates). After 5 min, a solution of the substrate diester of general structure 40 (1.0-1.5 equiv) in dry THF (~3 mL per mmol of the diester substrate) was added dropwise over a period of 2 min and the mixture was stirred at -78°C for 1-2 h. Aqueous NH4CI-NH4OH, pH = 8 (one-half the volume of the total volume of the reaction mixture) was added. The mixture was opened to the atmosphere and stirred vigorously until the aqueous phase became deep blue. The phases were separated and the aqueous phase was extracted three times with Et20. The combined organic extracts were washed with brine, dried (MgS04) and concentrated. The crude product was purified by flash chromatography on silica gel. 61 Preparation of dimethyl (iT)-3-trimethylstannyloct-2-en-6-ynedioate (21) ^ ^ ^ C 0 2 M e Me0 2C / ' SnMe3 21 Following general procedure 1 (p 60), dimethyl 2,6-octadiynedioate1415b (42) was converted into dimethyl (£)-3-trimethylstannyloct-2-en-6-ynedioate (21) with the following amounts of reagents and solvents: [Me3SnCuCN]Li (48), 1.7 mmol, in THF, 10 mL; MeOH, 69.2 |iL (1.7 mmol); dimethyl 2,6-octadiynedioate, 500 mg (2.6 mmol), in THF, 10 mL. In this experiment, the reaction time was 1.5 h. Flash chromatography (45 g silica gel, 4:1 petroleum ether-Et20) of the crude product gave 535 mg (87%) of the diester 21 as a clear oil, which exhibited ir (neat): 2239, 1718, 1596, 1261, 1170, 774 cm"1; J H nmr (CDC13, 400 MHz): 5 0.21 (s, 9H, Sn(CH3)3, 27Sn.H= 54 Hz), 2.47 (t, 2H, =C-CH2, J = 7.5 Hz), 3.09 (t, 2H, =C-CH2, J = 7.5 Hz, 37s„-H = 57 Hz), 3.69 (s, 3H, -OCH3), 3.73 (s, 3H, -OCH3), 6.04 (s, IH, =CH, 3JSn-H = 69 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -9.0, 18.7, 32.1, 51.0, 52.5, 73.2, 88.8, 129.4, 154.1, 164.2, 170.0. Exact mass calcd. for Ci 2H 1 70 4Sn (M + - Me): 345.0141; found: 345.0149. Anal, calcd. for Ci 3H 2 0O 4Sn: C 43.50, H 5.62; found: C 43.79, H 5.66. 62 Preparation of diethyl (ffi-3-trimethylstannylnon-2-en-7:ynedioate (22) Following general procedure 1 (p 60), diethyl 2,7-nonadiynedioate a (43) was converted into diethyl (£)-3-trimethylstannylnon-2-en-7-ynedioate (22) with the following amounts of reagents and solvents: [Me3SnCuCN]Li (48), 0.8 mmol, in THF, 7 mL; EtOH, 45.7 [ih (0.8 mmol); diethyl 2,6-nonadiynedioate, 238 mg (1.0 mmol), in THF, 3 mL. In this experiment, the reaction time was 2 h. Flash chromatography (20 g silica gel, 6:1 petroleum ether-Et20) of the crude product gave 155 mg (51%) of diester 22 as a colorless oil which exhibited ir (neat): 2237, 1713, 1598, 1251, 1186, 773 cm"1; lH nmr (CDC13, 400 MHz): 5 0.20 (s, 9H, Sn(CH3)3, 27SN-H= 53.5 Hz), 1.25-1.30 (m, 6H, -OCH 2CH 3), 1.69 (quintet, 2H, CH2-CH2-CH2, J = 7.5 Hz), 2.36 (t, 2H, =C-CH2, J = 7.5 Hz), 2.94 (t, 2H, =C-CH2, J = 7.5 Hz, 3/Sn-H= 60 Hz), 4.13 (q, 2H, -OCH 2 , J = 7 Hz), 4.19 (q, 2H, -OCH 2CH 3 , 7 = 7 Hz), 5.98 (s, IH, =CH, V S n - H = 72 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -9.2, 14.0, 14.3, 18.6, 27.4, 33.9, 59.7, 61.7, 73.5, 88.8, 128.7, 153.7, 164.1, 171.0. Exact mass calcd. for Ci 5H 2 30 4Sn (M + - Me): 387.0618; found: 387.0617. Anal, calcd. for Ci 6H 2 60 4Sn: C 47.92, H 6.53; found C 47.60; H 6.54. 63 Preparation of dimethyl 2.8-decadiynedioate (44) Me0 2 C C0 2Me 44 To a cold (-78°C), stirred solution of commercially available 1,7-octadiyne (100 mg, 0.9 mmol) in dry THF (9 mL) was added a solution of methyllithium (2.2 mmol) in Et20. After the solution was stirred at -78°C for 10 min and at -20°C for 30 min, methyl chloroformate (0.2 mL, 2.2 mmol) was added and the solution was stirred at -20°C for 30 min. ^The solution was treated with saturated aqueous sodium bicarbonate (5 mL) and the phases were separated. The aqueous phase was extracted with Et20 (3 x 10 mL), dried (MgS04), and concentrated. Flash chromatography (15 g silica gel, 7:3 petroleum ether-Et20, followed by 3:1 petroleum ether-Et20) of the crude product provided 148 mg (71%) of the diester 44 as a colourless solid, which exhibited m.p. 33°-34° C (literature15" m.p. 34°C); ir (KBr): 2239, 1714, 1439, 1261, 722 cm"1; *H nmr (CDC13, 400 MHz): 8 1.66-1.72 (m, 4H, -CH 2CH 2-), 2.36 (t, 4H, EEC-CH 2 , 7 = 7 Hz), 3.74 (s, 6H, -OCH3); 1 3 C nmr (CDC13, 50.3 MHz): 8 18.1, 26.4, 52.6, 73.4, 88.5, 154.1. Exact mass calcd. for Ci 2 Hi 4 0 4 : 223.0970; found 223.0967. Anal, calcd. for C 1 2 H i 4 0 4 : C 64.85, H 6.35; found: C 64.89, H 6.25. Preparation of dimethyl (g)-3-trimemylstannyldec-2-en-8-ynedioate (23) 64 Following general procedure 1 (p 60), dimethyl 2,8-decadiynedioate (44) was converted into dimethyl (£)-3-trimethylstannyldec-2-en-8-ynedioate (23) with the following amounts of reagents and solvents: [Me3SnCuCN]Li (48), 2.83 mmol, in THF, 15 mL; MeOH, 0.1 mL (2.8 mmol); dimethyl 2,8-decadiynedioate (44), 484 mg (2.1 mmol), in THF, 7 mL. In this experiment, the reaction time was 1.3 h. Flash chromatography (60 g silica gel, 6:1 petroleum ether-Et20) of the crude product gave 403 mg (48%) of the diester 23 as a colourless oil. In addition, 157 mg (32%) of starting material, compound 44, were recovered. The product 23 exhibited ir (neat): 2237, 1718, 1595, 1257, 1195, 773 cm1; *H nmr (CDC13, 400 MHz): 5 0.19 (s, 9H, Sn(CH3)3, 27 S d-H= 54 Hz), 1.46-1.64 (m, 4H, -CH 2CH 2-), 2.35 (t, 2H, E=C-CH 2, 7 = 7 Hz), 2.89 (t, 2H, =C-CH2, 7 = 7.5 Hz, 3 7 S n - H = 61 Hz), 3.67 (s, 3H, -OCH3), 3.73 (s, 3H, OCH3), 5.97 (s, IH, =CH, 37sn-H = 72.5 Hz) ; 1 3 C nmr (CDC13, 75.5 MHz): 5 -9.1, 18.4, 27.1, 28.5, 33.8, 50.9, 52.5, 73.1, 89.5, 127.5, 154.2, 164.6, 172.9. Exact mass calcd. for Ci 4H 2 10 4Sn (M + - Me): 373.0462; found: 373.0464. Anal, calcd. for Ci 5H 2 40 4Sn: C 46.55, H 6.25; found: C46.27, H 6.45. 65 4.0 Preparation of (ff.iT)-l,2-bis(alkoxycarbonylmethyhdene)c^^ General Procedure 2 COzR CQzR SnMe3 n = 1, 2, 3 R = Me, Et CO2R 39 55 To a cold (0°C), stirred solution of the appropriate diester of general structure 39 (1 equiv) in dry DMF (-10 mL per mmol of the diester substrate) was added CuCl (2.5 equiv) in one portion and the mixture was stirred at 0°C for reaction times ranging between 5 min and 1 h. Aqueous NH4CI-NH4OH, pH = 8 (one-half the volume of the total volume of the reaction mixture) was added. The mixture was opened to the atmosphere and stirred vigorously until the aqueous phase became deep blue. The phases were separated and the aqueous phase was extracted three times with Et20. The combined organic extracts were washed with brine, dried (MgS04) and concentrated. The crude product was purified by flash chromatography on silica gel and/or recrystallization. 66 Preparation of (E.F)-l,2-bis(memoxycarb^ (24) C0 2Me C0 2Me 24 Following general procedure 2 (p 65), dimethyl (Z -^3-trimethylstannyloct-2-en-6-ynedioate (21) was converted into (£,^-l,2-bis(memoxycarbonylmethylidene)cyclobutane (24) with the following amounts of reagents and solvents: dimethyl (Z -^3-trimethylstannyloct-2-en-6-ynedioate (21), 187 mg (0.5 mmol), in DMF, 5.2 mL; CuCl, 129 mg (1.3 mmol). In this experiment, the reaction time was 5 min. Flash chromatography (7 g silica gel, 4:1 pentane-Et20) of the crude product yielded 102 mg (95%) of the diester 24 as a colourless solid which, after recrystallization from hexane, exhibited m.p. 108-109°C; ir (KBr): 1705, 1652, 1220, 1172, 860 cm1; lU nmr (CDC13, 400 MHz): 5 3.10 (s, 4H, =C-CH2), 3.72 (s, 6H, -OCH3), 6.02 (s, 2H, =CH); 1 3 C nmr (CDC13, 75.5 MHz): 8 30.6, 51.5, 111.8, 158.8, 166.4. Exact mass calcd. for C10H12O4: 196.0737; found: 196.0736. Anal, calcd. for Ci 0 Hi 2 O 4 : C 61.20, H 6.17; found: C 61.19, H 6.00. 6 7 Preparation of (E.E)- 1.2-bis(emoxycarbonylmethylidene)cyclopentane (25) C0 2Et 25 Following general procedure 2 (p 65), diethyl (£)-3-trimethylstannylnon-2-en-7-ynedioate (22) was converted into (E,E)-l,2-bis(emoxycarbonylmethylidene)cyclopentane (25) with the following amounts of reagents and solvents: diethyl (£)-3-trimethylstannylnon-2-en-7-yndioate (21), 234 mg (0.6 mmol), in DMF, 6.1 mL; CuCl, 144 mg (1.5 mmol). In this experiment, the reaction time was 10 min. Flash chromatography (10 g silica gel, 4:1 pentane-Et20) of the crude product yielded 131 mg (94%) of the diester 25 as a colourless solid which, after recrystallization from pentane, exhibited m.p. 33-34°C; ir (KBr): 1713, 1649, 1276, 1158, 812 cm1; X H nmr (CDC13, 400 MHz): 8 1.27 (t, 6H, -OCH 2CH 2 , 7 = 7 Hz), 1.80 (quintet, 2H , CH 2 -CH 2 -CH 2 , 7 = 7.5 Hz), 2.90-2.94 (m, 4H, =C-CH2), 4.16 (q, 4H, -OCH 2CH 3 , 7 = 7 Hz), 6.27 (s, 2H, =CH); 1 3 C nmr (CDCI3, 75.5 MHz): 8 14.3, 23.8, 32.1, 60.1, 112.2, 159.6, 166.5. Exact mass calcd. for Ci 3 Hi 8 0 4 : 238.1205; found 238.1208. Anal, calcd. for Ci 3 H 1 8 0 4 : C 65.53, H 7.61; found: C 65.25, H 7.71. 68 Preparation of (£.F)-l,2-bis(memoxycarb^ (26) C0 2Me C0 2Me 26 Following general procedure 2 (p 65), dimethyl (£)-3-trimethylstannyldec-2-en-8-ynedioate (23) was converted into (E,E)-l,2-bis(methoxycarbonylmethylidene)cyclohexane (26) with the following amounts of reagents and solvents: dimethyl (E)-3-trimethylstannyldec-2-en-8-ynedioate (23), 215 mg (0.6 mmol), in DMF, 5.6 mL; CuCl, 138 mg (1.4 mmol). In this experiment, the reaction time was 1 h. Flash chromatography (9 g silica gel, 5:1 pentane-Et20) of the crude product yielded 92 mg (74%) of the diester 26 as a colourless solid, which exhibited m.p. 32-33° C; ir (KBr): 1719, 1635, 1435, 1173, 870 cm1; X H nmr (CDC13, 400 MHz): 5 1.69-1.73 (m, 4H, -CH2-CH2CH2-CH2-), 2.92-2.99 (m, 4H, =C-CH2), 3.69 (s, 6H, -OCH3), 5.81 (s, 2H, =CH); 1 3 C nmr (CDC13, 50.3 MHz): 8 25.7, 30.1, 21.2, 114.6, 160.5, 166.6. Exact mass calcd. for d 2 H i 6 0 4 : 224.1049; found 224.1058. Anal, calcd. for d 2 H i 6 0 4 : C 64.27, H 7.19; found: C 64.12, H 7.15. 69 Preparation of (1 Z. 2 iT)-l-((trimethylstannyl)(mem^ carbonylmethylidene)cyclohexane (61) C0 2Me C0 2Me 26 C0 2Me r ^ V ^ S n M e 3 b C0 2Me 61 Following general procedure 2 (p 65), dimethyl (jB)-3-trimethylstannyldec-2-en-8-ynedioate (23) was converted into a mixture of (1 Z, 2 £)-l-((trimethylstannyl)(methoxy-carbonyl)memylidene)-2-(memoxycarbonylmethyUdene)cyclohexane (61) and (£,£)-1,2-bis-(methoxycarbonylmethylidene)cyclohexane (26) with the following amounts of reagents and solvents: dimethyl (£)-3-trimethylstannyldec-2-en-8-ynedioate (23), 53 mg (0.2 mmol), in DMF, 1.4 mL; CuCl, 34 mg (0.3 mmol). In this experiment, the reaction time was 3 min, at which time the ratio of ester 61 to ester 26 was ~ 2:1, as shown by glc analysis. Partial separation of the two diesters was accomplished by reversed-phase23 flash chromatography (4 g C-18 bonded silica, 7:3 EtOH-H20); however, complete separation of the two compounds was not achieved and the "purified" sample of diester 61 retained minor amounts of diester 26. The ratio of 61:26 in this sample was ~ 6:1 CH nmr analysis). 70 Compound 61, containing -17% of 26, exhibited ir (neat): 1719, 1636, 1600, 1434,1360, 1283, 1171, 775 cm'1; 'H nmr (CDC13, 400 MHz): 8 0.14 (s, 9H, Sn(CH 3) 3, 2/S N-H= 5 5 Hz), 1.69-1.85 (m, 4H, -CH 2 c-CH 2 d), 2.29-2.53 (m, 2H, -CH2), 2.77-3.00 (m, 2H, -CH2), 3.68 (s, 3H, -OCH3), 3.69 (s, 3H, -OCH3), 5.69 (s, IH, =CHa). In nOe difference experiments, irradiation at 8 5.69 (=CHa) produced an enhancement of the signal at 8 0.14 (SnMe3b); conversely, irradiation at 8 0.14 produced an enhancement of the resonance at 8 5.69. 1 3 C nmr (CDC13, 75.5 MHz): 8 -7.1, 27.3, 27.4, 31.6, 35.5, 51.2, 51.4, 114.22, 132.1, 157.9, 164.0, 166.6, 170.7. Exact mass calcd. for C i 4 H 2 1 0 4 ( M + - Me): 373.0462; found: 373.0475. 71 5.0 Deconjugation and methylation of ethyl 2-trimetfaylstannyl-1 -cyclohexenecarboxylate (68b). Preparation of ethyl l-methyl-2-trimethylstannyl-2-cyclohexenecarboxylate (72) To a cold (-78°C), stirred solution of LDA (4.73 mmol) in dry THF (17 mL) was added DMPU (0.57 mL, 4.7 mmol) and the solution was stirred at -78°C for 10 min. A solution of ethyl 2-trimethylstannyl-l-cyclohexenecarboxylate25 (68b) (750 mg, 2.4 mmol) in 7 mL of dry THF was added dropwise over a period of 5 min. The resulting yellow solution was stirred at -78°C for 30 min and at 0°C for 50 min. The orange mixture was cooled to -20°C, iodomethane (0.22 mL, 3.6 mmol) was added in one portion and the solution was stirred at -20°C for 50 min. Saturated aqueous sodium bicarbonate (12 mL) was added, the mixture was opened to the atmosphere and the stirred mixture was allowed to warm to room temperature. The phases were separated and the aqueous phase was extracted with Et20 (3 x 15 mL). The combined organic extracts were washed with brine (2 x 20 mL), dried (MgSCu) and concentrated. Flash chromatography (55 g silica gel, 20:1 petroleum ether-Et20) of the crude product provided 656 mg (84%) of the ester 72 as a colourless oil, which exhibited ir (neat): 1722, 1600, 1262, 1181, 768 cm"1; X H nmr (CDC13, 400 MHz): 8 0.08 (s, 9H, Sn(CH3)3, 2 / S N -H= 52.5 Hz), 1.23 (t, 3H, 68b 72 72 -CH 2 CH 2 , 7 = 7 Hz), 1.25 (s, 3H, -CH3), 1.53-1.66 (m, 3H), 1.96-2.11 (ra, 3H), 4.10 (q, 2H, -CH 2 CH 3 , 7 = 7 Hz), 5.88 (t, IH, =CH, 7 = 3.5 Hz, 37Sn-H= 73 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -7.6, 14.2, 18.9, 26.7, 27.0, 34.6, 47.2, 60.6, 137.9, 145.6, 177.6. Exact mass calcd. for Ci 2H 2i0 2Sn (M + - Me): 317.0564; found: 317.0563. Anal, calcd. for C 1 3H 2 40 2Sn: C 47.17, H 7.31; found: C 47.47, H 7.26. 73 6.0 Reduction of (3-tTimethylstannyl ct.ft- and p.y-unsaturated esters. Preparation of the alcohols 67a. 67b. 71 and 64 General Procedure 3 To a cold, stirred solution of the appropriate P-trimethylstannyl a,P- or P,y-unsaturated ester (1 equiv) in dry THF (-78°C) or dry Et 20 (0°C) (-10 mL per mmol of the ester substrate) was added, dropwise, a 1.0 M solution of DIBAL in hexanes (2.5 equiv) over a period of -3 min. If the solvent was THF, the resulting clear solution was stirred at -78°C for 1 h, warmed to room temperature and stirred for 30 min. Alternatively, if the reaction was run in Et 20, the solution was stirred at 0°C for 10-15 min. Saturated aqueous sodium potassium tartrate was added (one-half the volume of the total volume of the reaction mixture), the cloudy mixture was opened to the 7 4 atmosphere and was allowed to stir at room temperature for 1 h. The phases were separated and the aqueous phase was extracted three times with Et 20. The combined organic extracts were washed with brine, dried (MgS04) and concentrated. The crude product was purified by flash chromatography on silica gel. Preparation of l-hydroxymemyl-2-trimethylstannylcyclopentene (67a) Following general procedure 3 (p 73), methyl 2-(trimethylstannyl)-1-cyclopentene-carboxylate25 (68a) was converted into l-hydroxymemyl-2-trimethylstannylcyclopentene (67a) with the following amounts of reagents and solvents: methyl 2-(trimethylstannyl)-cyclopentenecarboxylate, 1.2 g (4.2 mmol) in THF, 42 mL; DJJ3AL, 10.5 mL (10.5 mmol). Flash chromatography (75 g silica gel, 5:1 pentane-Et20) of the crude product provided 992 mg (91%) of the alcohol 67a as a colourless liquid which displayed ir (neat): 3365 (br), 1616, 1235, 768 cm"1; *H nmr (CDC13,400 MHz): 8 0.12 (s, 9H, Sn(CH3)3, 27SN-H= 54 Hz), 1.48 (t, IH, -OH, 7=5 Hz), 1.84 (q, 2H, -CH 2 CH 2 CH 2 , 7 = 7 Hz), 2.35-2.50 (m, 4H), 4.19 (d, 2H, -CH 2OH, 7 = 5 Hz); 68a 67a 75 1 3 C nmr (CDC13, 75.5 MHz): 8 -8.8, 24.1, 34.8, 39.7, 63.3, 139.7, 152.1. Exact mass calcd. for CgHisOSn (M + - Me): 247.0145; found: 247.0141. Anal, calcd. for C9Hi8OSn: C 41.43, H 6.95; found: C 41.14, H 6.88. Preparation of l-hydroxymemyl-2-trimethylstannylcyclohexene (67b) Following general procedure 3 (p 73), ethyl 2-trimethylstannyl-l-cyclohexene-carboxylate25 (68b) was converted into l-hydroxymethyl-2-trimethylstannylcyclohexene (67b) with the following amounts of reagents and solvents: ethyl 2-(trimethylstannyl)-l-cyclohexenecarboxylate, 3.4 g (10.6 mmol) in THF, 106 mL; DIBAL, 26.5 mL (26.5 mmol). Rash chromatography (205 g silica gel, 5:1 pentane-Et20) of the crude product provided 2.6 g (88%) of the alcohol 67b as a colourless liquid which displayed ir (neat): 3347 (br), 1623, 1272, 1187, 1067, 767 cm'1; J H nmr (CDC13, 400 MHz): 8 0.12 (s, 9H, Sn(CH3)3,27SN-H= 52 Hz), 1.15 (t, IH, -OH, 7 = 6 Hz), 1.52-1.66 (m, 4H, -CH 2 a-CH 2 b), 2.10-2.13 (m, 2H), 2.16-2.19 (m, 2H), 3.96 (d, 2H, -ClfcOH, 7 = 6 Hz); 1 3 C nmr (CDC13, 50.3 MHz): 8 -8.2, 22.7, 23.5, 28.5, 32.4, 68b 67b 76 69.2, 138.0, 144.9. Exact mass calcd. for C 9H 1 7OSn (M + - Me): 261.0302; found 261.0303. Anal, calcd. for C 1 0H 2 0OSn: C 43.68, H 7.33; found: C 43.40, H 7.33. Preparation of 3-hydroxymethyl-3-memyl-2-trimethylstannylcy (71) Following general procedure 3 (p 73), ethyl l-memyl-2-trimethylstannyl-2-cyclo-hexenecarboxylate (72) was converted into 3-hydroxymethyl-3-methyl-2-trimethylstannyl-cyclohexene (71) with the following amounts of reagents and solvents: ethyl l-methyl-2-trimethylstannyl-2-cyclohexenecarboxylate, 600 mg (1.8 mmol), in Et 20, 18.1 mL; DJJBAL, 4.5 mL (4.5 mmol). In this experiment, the reaction time was 15 min. Flash chromatography (36 g silica gel, 3:1 petroleum ether-Et20) of the crude product yielded 491 mg (94%) of the alcohol 71 as a colourless oil, which displayed ir (neat): 3385, 1600, 1034, 767 cm"1; ! H nmr (CDC13, 400 MHz): 5 0.11 (s, 9H, Sn(CH3)3, VS n-H= 51.5 Hz), 0.80-0.85 (m, IH, -OH), 0.95 (s, 3H, -CH3), 1.25-1.28 (m, IH), 1.34-1.39 (m, IH), 1.57-1.75 (m, 2H), 1.98-2.10 (m, 2H), 3.26-3.42 (m, 2H, 72 71 77 -CH2OH), 5.97 (t, 1H, =CH, J = 3.5 Hz, 3/Sn.H= 78 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -7.5, 18.7, 25.1, 27.5, 33.0, 41.7, 71.6, 140.2, 146.8. Exact mass calcd. for Ci 0Hi 9OSn (M + - Me): 275.0458; found: 275.0455. Anal, calcd. for CnH2 2OSn: C 45.72, H 7.67; found: C 45.82, H 7.81. Preparation of (Z)-3-trimethylstannyl-2-penten-l-ol (64) Following general procedure 3 (p 73), ethyl (Z)-3-trimethylstannyl-2-pentenoate (50) was converted into (Z)-3-trimethylstannyl-2-penten-l-ol (64) with the following amounts of reagents and solvents: ethyl (Z)-3-trimethylstannyl-2-pentenoate (50), 1.0 g (3.5 mmol), in Et 20, 35 mL; DIBAL, 8.8 mL (8.8 mmol). In this experiment, the reaction time was 10 min. Flash chromatography (60 g silica gel, 3:1 petroleum ether-Et20) of the crude product afforded 625 mg (72%) of the alcohol 50 as a colourless oil, which displayed ir (neat): 3339 (br), 1625, 1190, 1087,770 cm *H nmr (CDC13,400 MHz): 8 0.17 (s, 9H, Sn(CH3)3,27Sn.H= 53 Hz), 0.96 (t, 3H, 50 64 78 -CH 2CH 2 , J = 7.5 Hz), 1.15 (t, IH, -OH, J = 5.5 Hz), 2.23 (q, 2H, -CH 2 CH 3 , J = 7.5 Hz, 3 / S N - H = 57 Hz), 4.08 (dd, 2H, -OCH 2 , / = 6.5 Hz, J = 5.5 Hz), 6.20 (t, IH, =CH, J = 6.5 Hz, 37S„-H = 138 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 6 -8.0, 14.8, 33.1, 64.5, 136.9, 151.3. Exact mass calcd. for C7Hi5OSn (M + - Me): 235.0145; found: 235.0154. Anal, calcd. for C8Hi8OSn: C 38.61, H 7.29; found: C 38.86, H 7.31. 79 7.0 Oxidation of the alcohols 67a, 67b. 71 and 64 Preparation of 2-trimethvlstannvl-l-cvcloDentenecarbaldehvde (66a) To a cold (-78°C), stirred solution of oxalyl chloride (0.4 mL, 4.5 mmol) in dry CH2C12 (15 mL) was added DMSO (0.7 mL, 9.74 mmol) and the solution was stirred at -78°C for 15 min. A solution of l-hydroxymemyl-2-trimethylstannylcyclopentene (67a) (529 mg, 2.0 mmol) in dry CH2C12 (5 mL) was added dropwise over a period of 5 min, and the resulting cloudy white mixture was stirred for 15 min at -78°C. Triethylamine (2.8 mL, 20.3 mmol) was added dropwise over a period of 1 min. The clear solution was warmed to room temperature, water (10 mL) was added and the mixture was stirred for 10 min. The phases were separated, the aqueous phase was extracted with CH2C12 (3 x 10 mL), the combined organic phases were dried (MgS04) and concentrated. Flash chromatography (35 g of silica gel, 3:1 pentane-Et20) of the crude product provided 419 mg (80%) of the aldehyde 66a, a colourless liquid which exhibited ir (neat): 1727, 1673, 1125, 774 cm-1; *H nmr (CDC13, 400 MHz): 5 0.27 (s, 9H, Sn(CH3)3, VS n.H= 56 Hz), 1.89 (quintet, 2H, -CH 2 CH 2 CH 2 , J = 7.5 Hz), 2.57 (t, 2H, J = 7.5 Hz), 2.70 (t, 2H, J = 7.5 Hz), 9.75 80 (s, IH, -CHO); 1 3 C nmr (CDC13, 75.5 MHz): 8 -8.5, 23.6, 30.6, 41.8, 155.0, 175.8, 190.8. Exact mass calcd. for C8Hi 3OSn (M + - Me): 244.9988; found: 244.9987. Anal, calcd. for C 9H 1 6OSn: C 41.70, H 6.23; found: C 41.43, H 6.10. Preparation of 2-trimethylstannyl-l-cyclohexenecarbaldehyde (66b) To a stirred suspension of N-memylmorpholine N-oxide (NMO) (120.6 mg, 1.0 mmol) and activated, crushed 4 A molecular sieves (500 mg/mmol substrate) in dry CH2Q2 (4 mL) at room temperature was added a solution of l-hydroxymethyl-2-trimethylstannylcyclohexene (67b) (188.1 mg, 0.7 mmol) in dry CH2Q2 (2.8 mL) and the suspension was stirred for 5 min. Tetra-n-propylammonium perruthenate (TPAP) (24.0 mg, 0.1 mmol) was added as a solid in one portion and the black suspension was stirred at room temperature for 35 min. The mixture was filtered through silica gel (5 g, moistened with CH2CI2) and the collected material was washed with CH2C12 (30 mL). Concentration of the eluate, followed by flash chromatography (13 g silica gel, 3:1 petroleum ether-Et20) of the derived liquid afforded 150 mg (80%) of the aldehyde 66b, a colourless oil, which displayed ir (neat): 1678, 1572, 1141, 867, 722 cm"1; X H nmr (CDC13, 400 MHz): 8 0.23 (s, 9H, Sn(CH3)3, 2 / S N -H= 54 Hz), 1.60-1.64 (m, 4H, -CH 2-CH2CH 2-CH 2), 2.27-O 66b 81 2.29 (m, 2H), 2.47-2.50 (m, 2H), 9.42 (s, IH, -CHO); 1 3 C nmr (CDC13,125.8 MHz): 8 -7.8, 21.5, 23.3, 24.6, 34.6, 146.5, 172.4, 194.2. Exact mass calcd. for C 9H 1 5OSn (M + - Me): 259.0145; found: 259.0146. Anal, calcd. for Ci 0H 1 8OSn: C 44.06, H 6.65; found: C 44.32, H 6.72. Preparation of l-memyl-2-trimemylstarmyl-2-cyclohexenecarbald (70) To a stirred suspension of NMO (264.8 mg, 2.3 mmol) and activated, crushed 4 A molecular sieves (500 mg/mmol substrate) in dry CH2C12 (10 mL) at room temperature was added a solution of 3-hydroxymethyl-3-methyl-2-trimethylstannylcyclohexene (71) (434 mg, 1.5 mmol) in dry CH2C12 (5 mL) and the suspension was stirred for 5 min. TPAP (52.8 mg, 0.2 mmol) was added as a solid in one portion and the black suspension was stirred at room temperature for 30 min. The mixture was filtered through silica gel (15 g, moistened with CH2C12) and the collected material was washed with CH2C12 (90 mL). Concentration of the eluate, followed by flash chromatography (30 g silica gel, 5:1 petroleum ether-Et20) of the aquired liquid, afforded 377 mg (88%) of the aldehyde 70, a colourless oil, which displayed ir (neat): 1723, 1599, 1188, 956, 769 O 70 8 2 cm/1; lH nmr (CDC13, 400 MHz): 8 0.09 (s, 9H, Sn(CH3)3, VS n.H= 53 Hz), 1.14 (s, 3H, -CH3), 1.42-1.49 (m ,1H), 1.61-1.67 (m, 2H), 1.81-1.88 (m, IH), 2.04-2.11 (m, 2H), 6.08 (t, IH, =CH, J = 3.5 Hz, 3 J S n-H= 73 Hz), 9.42 (s, IH, -CHO); 1 3 C nmr (CDC13, 75.5 MHz): 8 -7.9, 18.3, 22.9, 27.1, 31.4, 51.9, 141.1, 204.1. Exact mass calcd. for Ci0H,7SnO (M + - Me): 273.0302; found: 273.0304. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean J H and 1 3 C nmr spectra. Preparation of (Z)-3-trimethylstannyl-2-pentenal (63) To a stirred suspension of NMO (409.3 mg, 3.5 mmol) and activated, crushed 4 A molecular sieves (500 mg/mmol substrate) in dry CH2Q2 (18 mL) at room temperature was added a solution of (Z)-3-trimethylstannyl-2-penten-l-ol (64) (578 mg, 2.3 mmol) in dry CH2C12 (5 mL) and the suspension was stirred for 5 min. TPAP (81.6 mg, 0.2 mmol) was added as a solid in one portion and the black suspension was stirred at room temperature for 15 min. The mixture was filtered through silica gel (20 g, moistened with CH2C12) and the collected material was washed with CH2Ci2 (100 mL). Concentration of the eluate, followed by flash chromatography (40 g silica gel, 5:1 pentane-Et20) of the derived liquid, afforded 425 mg (74 %) of the aldehyde 6 3 , a H 6 3 83 colourless oil which displayed ir (neat): 1683, 1565, 1156, 773 cm"1; *H nmr (CDC13, 400 MHz): 8 0.25 (s, 9H, Sn(CH3)3, 2/Sn.H= 54 Hz), 1.04 (t, 3H, =CCH2CH,, J = 7.5 Hz), 2.50 (q, 2H, =C-CH2CH3, / = 7.5 Hz, 37 S„.H = 41 Hz), 6.64 (d, IH, =CH, J = 5.5 H, 3JSn-H = 116 Hz), 9.57 (d, IH, -CHO, J = 5.5 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -7.5, 13.3, 34.0, 137.8, 183.5, 192.9. Exact mass calcd. for C7Hi3OSn (M+- Me): 232.9988; found: 232.9880. The ^stability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean J H and 1 3 C nmr spectra. 84 8.0 Addition of methyl 3-lithiopropynoate to the aldehydes 66a. 66b, 70 and 63. Preparation of the alkenyltrimethylstannanes 32. 33. 37 and 31 General Procedure 4 66a 66b 70 63 C0 2Me CQ 2Me To a cold (-78°C), stirred solution of LDA (2 equiv) in dry THF (~4 mL per mmol of LDA) were added sequentially DMPU (2 equiv) and methyl propynoate (2 equiv) and the resulting yellow solution was stirred for 10 min at -78°C. A solution of the appropriate aldehyde substrate (1 equiv) in dry THF (~3 mL per mmol of substrate) was added dropwise over a period of 5 min and the orange mixture was stirred at -78°C for 30 - 90 min. Water or aqueous NH4CI-85 NELOH, pH = 8 (one-half the volume of the total volume of the reaction mixture) was added, the mixture was opened to the atmosphere and warmed to room temperature. The phases were separated and the aqueous layer was extracted three time with Et 20. The combined organic extracts were washed with brine, dried (MgS04) and concentrated. The crude product was purified by flash chromatography on silica gel or flash chromatography followed by radial chromatography on silica gel. Preparation of methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclopenten-l-yl)-2-butynoate (32) Following general procedure 4 (p 84), methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclopenten-l-yl)-2-butynoate (32) was prepared from 2-trimethylstannyl-l-cyclopentenecarb-aldehyde (66a) with the following quantities of reagents and solvents: LDA, 4.0 mmol, in THF, 15 mL; DMPU, 0.5 mL (4.0 mmol); methyl propynoate, 0.36 mL (4.0 mmol); 2-trimethylstannyl-1-cyclopentenecarbaldehyde (66a), 519 mg (2.0 mmol), in THF, 5 mL. In this experiment, the reaction time was 1.5 h. Flash chromatography (48 g silica gel, 5:1 pentane-Et20) of the crude product provided 593 mg (86%) of the alcohol 32, a pale yellow liquid which exhibited ir (neat): 3452 (br), 2235, 1719, 1619, 1253, 771 cm"1; *H nmr (CDC13, 400 MHz): 8 0.16 (s, 9H, OH 32 86 Sn(CH 3) 3, 2/s„-H= 54 Hz), 1.88 (quintet, 2H, -CH 2 CH 2 CH 2 , J = 7.5 Hz), 2.00 (d, IH, -OH, J=5 Hz), 2.43-2.61 (m, 4H), 3.76 (s, 3H, -OCH3), 5.15 (d, IH, -CHOH, 7=5 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -8.7, 23.8, 33.0, 39.9, 52.8, 62.6, 76.5, 86.5, 143.7, 148.4, 153.7. Exact mass calculated for Ci 2Hi 70 3Sn (M + - Me): 329.0200; found: 329.0203. Anal, calcd. for Ci 3H 2 0O 3Sn: C 45.52, H 5.88; found: C 45.65, H 5.82. Preparation of methyl 4-hydroxy-4-(2-trimemylstarmyl-l-cyclohexen-l-yD-2-butynoate (33) Following general procedure 4 (p 84), methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclo-penten-l-yl)-2-butynoate (33) was prepared from 2-trimethylstannyl-l-cyclo-hexenecarbaldehyde (66b) with the following quantities of reagents and solvents: LDA, 11.0 mmol, in THF, 41 mL; DMPU, 1.3 mL (11.0 mmol); methyl propynoate, 1.0 mL (11.0 mmol); 2-trimethylstannyl-l-cyclohexenecarbaldehyde (66b), 1.5 g (5.5 mmol), in THF, 14 mL. In this experiment, the reaction time was 1.5 h. Flash chromatography (140 g silica gel, 5:1 pentane-Et20) of the crude product provided 1.67 g (85%) of the alcohol 33 as an orange hquid which exhibited ir (neat): 3420 (br), 2234, 1719, 1615, 1255, 1147, 770 cm"1; lU nmr (CDC13, 400 MHz): 8 0.15 (s, 9H, Sn(CH3)3, Vs„-H= 53 Hz), 1.47-1.72 (m, 4H, -CHzCH^CHzCH^, 2.05 (br s, IH, -OH), 2.10-2.34 OH 33 87 (m, 4H), 3.75 (s, 3H, -OCH3), 4.84 (d, IH, -CHOH, 7 = 4 Hz); 1 3 C nmr (CDC13, 50.3 MHz): 8 -8.0, 20.7, 22.4, 25.9, 32.7, 52.8, 67.1, 76.5, 87.1, 141.5, 142.3, 155.8.' Exact mass calcd. for Ci 3Hi 90 3Sn (M + - Me): 343.0356; found 343.0348. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean *H and 1 3 C nmr spectra. Preparation of methyl 4-hydroxy-4-(l-memyl-2-trimemylstannyl-2-cyclohexen-l-yl)-2-buW £37} CQ2Me 37 Following general procedure 4 (p 84), 4-hydroxy-4-(l-methyl-2-trimethylstannyl-2-cyclohexen-l-yl)-2-butynoate (37) was prepared from l-methyl-2-trimethylstannyl-2-cyclo-hexenecarbaldehyde (70) with the following quantities of reagents and solvents: LDA, 2.3 mmol, in THF, 8 mL; DMPU, 0.3 mL (2.3 mmol); methyl propynoate (77), 0.2 mL (2.3 mmol); 1-methyl-2-trimethylstannyl-2-cyclohexenecarbaldehyde (70), 1.5 g (5.5 mmol), in THF, 3.5 mL. In this experiment, the reaction time was 30 min. Flash chromatography (30 g silica gel, 5:1 pentane-Et20) of the crude product, followed by radial chromatography (4 mm silica gel plate, 5:1 pentane-Et20) of the derived liquid provided two diastereomeric esters, isomers 37a and 37b. 88 Isomer 37a (278 mg, 65%) was isolated as a colourless oil which exhibited ir (neat): 3486, 2235, 1719, 1599, 1252, 1141, 1038, 768 cm1; X H nmr (CDC13, 400 MHz): 8 0.09 (s, 9H, Sn(CH3)3,27SN-H= 52 Hz), 1.11 (s, 3H, -CH3), 1.57-1.72 (m, 4H), 1.92 (d, IH, -OH, J = 6.5 Hz), 2.02-2.06 (m, 2H), 3.76 (s, 3H, -OCH3), 4.22 (d, IH, -CHOH, J= 6.5 Hz), 5.93 (t, IH, =CH, J = 3.5 Hz, 37Sn-H= 76.5 Hz); 1 3 C nmr (CDC13, 75.5 MHz): 8 -6.7, 18.3, 22.7, 27.6, 33.0, 44.2, 52.8, 71.0, 77.9, 87.0, 139.3, 145.7, 153.7. Exact mass calcd. for Ci 4H 2i0 3Sn (M + - Me): 357.0513; found: 357.0516. Anal, calcd. for Ci 5H 2 40 3Sn: C 48.56, H 6.52; found: C 48.74, H 6.64. Isomer 37b (86 mg, 20%) was isolated as a colourless solid, which displayed m.p. 33°-34°C; ir (KBr): 3448, 2235, 1703, 1600, 1255, 1134, 1032, 767 cm"1; lH nmr (CDC13, 400 MHz): 8 (s, 9H, Sn(CH3)3, 2/Sn-H= 52 Hz), 1.09 (s, 3H, -CH3), 1.52-1.65 (m, 2H), 1.70-1.79 (m, IH), 1.86-1.93 (m, 2H), 2.02-2.06 (m„ 2H), 3.76 (s, 3H, -OCH3), 4.26 (br s, IH, -CHOH), 6.07 (t, IH, =CH, J = 3.5 Hz, 37Sn.H= 77 Hz); 1 3 C nmr (CDC13, 50.3 MHz): 8 -7.3, 18.6, 25.6, 27.4, 30.9, 45.4, 52.8, 70.4, 77.7, 86.5, 142.1, 145.3, 153.7. Exact mass calcd. for Ci 4H 2,0 3Sn (M + - Me): 357.0513; found: 357.0507. Anal, calcd. for Ci 5H 2 40 3Sn: C 48.56, H 6.52; found C 48.83, H 6.70. 89 Preparation of methyl (Z)-4-hydroxy-6-trimemylstannyl-5-octen-2-y (31) C0 2Me Following general procedure 4 (p 84), methyl (Z)-4-hydroxy-6-trimethylstannyl-5-octen-2-ynoate (31) was prepared from (Z)-3-trimethylstannyl-2-pentenal (63) with the following quantities of reagents and solvents: LDA, 0.7 mmol, in THF, 2.7 mL; DMPU, 0.9 mL (0.7 mmol); methyl propynoate, 1.0 mL (0.7 mmol); (Z)-3-trimethylstannyl-2-pentenal (63), 90 mg (0.4 mmol), in THF, 1 mL. In this experiment, the reaction time was 1.3 h. Flash chromatography (8 g silica gel, 5:1 pentane-Et20) of the crude product provided 102 mg (86%) of the alcohol 31 as a pale orange liquid, which exhibited ir (neat): 3420 (br), 2236, 1719, 1562, 1252, 1141, 771 cm"1; *H nmr (CDC13, 400 MHz): 8 0.21 (s, 9H, Sn(CH3)3, 27S„-H= 53 Hz), 0.98 (t, 3H, -CH 2 CH 2 , 7 = 7.5 Hz), 1.86 (d, IH, -OH, 7 = 5 Hz), 2.25 (q, 2H, -CH 2 CH 3 , 7 = 7.5 Hz, 37SN-H = 52 Hz), 3.76 (s, 3H, -OCH3), 4.86 (dd, IH, -CHOH, 7 = 7.5 Hz, 7 = 5 Hz), 6.10 (d, IH, =CH, 7 = 7.5 Hz, 37S„-H= 127 Hz); 1 3 C nmr (acetone-*/*, 75.5 MHz): 8 -7.4, 15.0, 33.5, 52.9, 63.0, 76.3, 88.8, 136.5, 151.1, 154.3. Exact mass calcd. for CnHi 70 3Sn (M + - Me): 317.0299; found: 317.0191. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean *H and 1 3 C nmr spectra. 90 9.0 Copperd) chloride-mediated cyclization of the aUcenyltrimemylstannan.es of general structure 65. Preparation of the bicyclic esters of general formula 81 General Procedure 5 To a cold (0°C), stirred solution of the alkenyltrimethylstannane of general structure 65 (1 equiv) in dry DMF (~10 mL per mmol of the substrate) was added CuCl (2.5 equiv) in one portion and the mixture was stirred at 0°C for 2 min. Aqueous NH4CI-NH4OH, pH = 8 (one-half the volume of the total volume of the reaction mixture) was added. The mixture was stirred vigorously, open to the atmosphere, until the aqueous phase became deep blue. The phases were separated and the aqueous phase was extracted three times with E t 2 0 . The combined organic extracts were washed with brine, dried (MgS04) and concentrated. The crude product was purified by flash chromatography on silica gel. 91 Preparation of 7-(YZ)-meuhoxycarbonylmethylidene)bic^ ^ (35) Following general procedure 5 (p 90), 7-((Z)-memoxycarbonylmethylidene)bicyclo-[3.2.0]hept-l(5)-en-6-ol (32) was prepared from methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclopenten-l-yl)-2-butynoate (35) with the following quantities of reagents and solvents: methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclopenten-l-yl)-2-butynoate (32), 156 mg (0.5 mmol) in DMF, 4.5 mL; CuCl, 112.2 mg (1.1 mmol). Purification of the crude product by flash column chromatography (5 g silica gel, 1:1 pentane-Et20) yielded 60 mg (73%) of the bicyclic ester 35 as a colourless solid, which, after recrystallization from pentane, exhibited m.p. 58-60°C; ir (KBr): 3428 (br), 1718, 1604, 1325, 1263, 734 cm"1; *H nmr (CDC13, 400 MHz): 8 2.22-2.28 (m, 2H, -CH 2 b ), 2.46-2.54 (m, 2H, -CH/), 2.62-2.69 (m, 2H, -CH2C_), 3.67 (s, 3H, -OCH3), 3.70 (s, IH, -OH), 5.01 (s, IH, -CHOH), 5.40 (s, IH, =CH). In a nOe difference experiment, irradiation at 8 2.50 (-CH2a) produced an enhancement of the resonance at 8 5.40 (=CH). 1 3 C nmr (CDC13, 75.5 MHz): 8 27.5, 29.1, 30.0, 51.1, 72.8, 99.4, 157.8, 160.6, 167.7, 174.8. Exact mass calcd. for CioHi 20 3: 180.0786; found 180.0789. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided a clean *H nmr spectrum. 35 92 Preparation of 8-((Zymethoxycarbonylmethylidene)bfc^ (36) Following general procedure 5 (p 90), 8-((Z)-raethoxycarbonylmethylidene)-bicyclo[4.2.0]oct-l(6)-en-7-ol (36) was prepared from methyl 4-hyckoxy-4-(2-trimethylstannyl-l-cyclohexen-l-yl)-2-butynoate (33) with the following quantities of reagents and solvents: methyl 4-hydroxy-4-(2-trimethylstannyl-l-cyclohexen-l-yl)-2-butynoate (33), 182 mg (0.5 mmol), in DMF, 5 mL; CuCl, 126 mg (1.3 mmol). Purification of the crude product by flash chromatography (7 g silica gel, 1:1 pentane-Et20) provided 74 mg (75%) of the bicyclic ester 36 as colourless needles, which displayed m.p. 27-28°C; ir (KBr): 3472 (br), 1699, 1601, 1337, 1267, 1117, 841, 718 cm"1; J H nmr (CDC13, 400 MHz): 8 1.54-1.83 (m, 2H, -CH 2 b CH 2 c ) , 1.95-2.10 (m, 2H, -CH 2 a), 2.27-2.44 (m, 2H, -CH 2 d), 3.71 (s, 3H, -OCH3), 4.13 (br s, IH, -OH), 5.21 (s, IH, -CHOH), 5.23 (s, IH, =CH); 1 3 C nmr (CDC13, 50.3 MHz): 8 20.9, 21.8, 22.0, 23.6, 51.3, 75.0, 96.9, 147.7, 163.8, 164.1, 168.9. Exact mass calcd. for CnHi 4 0 3 : 194.0943; found: 194.0945. Anal, calcd. for Ci ,H 1 4 0 3 : C 68.02, H 7.27; found: C 68.06, H 7.19. 36 93 10.0 Preparation of 7-benzoyloxy-8-((Z)-memoxycarbonylmethylide ene (82) To a cold (0°C), stirred solution of 8-((Z)-methoxycarbonylmethylidene)bicyclo[4.2.0]-oct-l(6)-en-7-ol (36) (141 mg, 0.7 mmol) in dry CH2C12 (7.3 mL) were added sequentially pyridine (147 fxL, 1.8 mmol), DMAP (8.7 mg, 0.1 mmol) and benzoyl chloride (136 itL, 0.8 mmol). The solution was stirred at 0°C for 1 h, warmed to room temperature and stirred for 1.5 h. Saturated aqueous NFLCl (4 mL) was added, the phases were separated and the aqueous phase was extracted with Et 20 (3x5 mL). The combined organic extracts were washed with brine (2x4 mL), dried (MgSO*) and concentrated. Flash chromatography (15 g silica gel, 1:1 pentane-Et20) of the crude product provided 63 mg (30%) of the diester 82 as a colourless liquid, as well as 80 mg (57%) of starting material 36. The product diester exhibited ir (neat): 1714, 1612, 1451, 1357, 1320, 1270, 1110, 711, 688 cm"1; *H nmr (CDC13, 400 MHz): 8 1.62-1.86 (m, 4H, -CH 2 B CH2 C ) , 2.09-2.24 (m, 2H. -CH 2 a), 2.29-2.37 (m, 2H, -CH 2 d), 3.55 (s, 3H, -OCH3), 5.32 (s, IH, =CHe), 6.39 (s, IH, -CHOH), 7.41 (dd, 2H, =C-Hm, J = 7.5 Hz, J = 7 Hz), 7.53 (t, IH, 94 =C-HP, J = 7.5 Hz), 8.01 (d, 2H, =C-H0, 7 = 7 Hz). In a nOe difference experiment, irradiation at 8 2.16 (-CH/) produced an enhancement of the signals at 8 1.62-1.86 (.-CH2bCUzc) and at 8 5.32 (=CHe). 1 3 C nmr (CDC13, 50.3 MHz): 8 21.0, 21.6, 22.1, 23.9, 51.2, 76.4, 100.0, 128.3, 129.7, 130.2, 132.9, 151.6, 157.2, 162.0, 166.4, 166.5. Exact mass calcd. for C 1 8 H 1 8 0 4 : 298.1205; found: 298.1212. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean *H and 1 3 C nmr spectra. 95 11. Preparation of 3-emyl-4-((Z)-methoxycarbonylmemyliden^ (34) and 3-ethyl-4- ((£^-methoxycarbonylmethylidene)-2-cyclobuten-1 -ol ( 7 8 ) To a cold (0°C), stirred solution of methyl (Z)-4-hydroxy-6-trimethylstannyl-5-octen-2-ynoate ( 3 1 ) (176 mg, 0.5 mmol) in DMF, 5.3 mL, was added CuCl (131.5 mg, 1.3 mmol) and the yellow solution was stirred at 0°C for 2 min. Aqueous NH4CI-NH4OH, pH = 8 (3 mL) was added. The mixture was stirred vigorously, open to the atmosphere, until the aqueous phase became deep blue. The phases were separated and the aqueous phase was extracted with Et 20 (3 x 4 mL). The combined organic extracts were washed with brine (2x4 mL), dried (MgS04) and concentrated. Flash chromatography (6 g silica gel, 3:1 petroleum ether-Et20) of the crude product provided 63 mg (70%) of ester 3 4 as a colourless oil and 14 mg (16%) of the ester 7 8 as a colourless solid. The former exhibited ir (neat): 3490 (br), 1695, 1582, 1337, 1248, 1178, 838 cm"1; XH nmr (CDC13, 400 MHz): 8 1.09 (t, 3H, -CH 2 CH 2 , J = 7.5 Hz), 2.05-2.22 (m, 2H, -CH2CH3), 3.73 (s, 3H, -OCH3), 4.02 (br s, IH, -OH), 5.14 (s, IH, =CH£ ), 5.38 (s, IH, -CHaOH), 6.87 (s, IH, =CHf); 1 3 C nmr (CDC13, 50.3 MHz): 8 10.3, 19.8, 51.5, 71.4, 99.5, 144.5, 154.9, 164.1, 168.3. For the results of nOe difference and HMQC experiments, the reader is referred to Table VJJ. Exact mass calcd. for C 9 Hi 2 0 3 : 168.0786; found 168.0788. The instabihty 3 4 7 8 96 of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided clean *H and 1 3 C nmr spectra. Table V H . Results of H M Q C and nOe experiments on compound 3 4 *H nmr (CDC13, 400 MHz): 8 (multiplicity, number of protons, 7) Assignment H x nOe correlations to Hx:° HMQC correlation to H X ( 1 3 C nmr, 125.8 MHz):6 8 1.09 (t, 3H, 7=7.5 Hz) H a 10.3 2.05-2.22 (m, 2H) H b H e , Hf 19.8 3.73 (s, 3H) He 51.6 5.14 (s, IH) H d 71.4 5.38 (s, IH) He H b 99.5 6.87 (s, IH) H f 144.5 "Entries in this column refer to proton signals which showed an enhancement on irradiation of H x . If no data is entered, a nOe difference experiment was not carried out on H x . ^Entries in this column refer to one bond C-H x correlations as determined by the H M Q C experiment. The ester 7 8 displayed m.p. 65°C; ir (KBr): 3389 (br), 1719,1579, 1364, 1284, 1200, 844 cm"1; *H nmr (CDC13) 400 MHz): 8 1.07 (t, 3H, -CH 2 CH 2 , 7 = 7 Hz), 1.79 (d, IH, -OH, 7 = 9.5 Hz), 2.48 (m, 2H, -CH2.CH3), 3.70 (s, 3H, -OCH3), 4.81 (d, IH, -CH(OH), 7 = 9.5 Hz), 5.50 (s, IH, =CH(C02Me)), 6.88 (s, IH, =CH(CHOH)); 1 3 C nmr (CDC13, 75.5 MHz): 5 10.8, 22.6, 51.2, 97 72.0, 103.5, 116.1, 158.3, 161.6, 166.9. Exact mass calcd. for C 9 Hi 2 0 3 : 168.0786; found 168.0785. The instability of this compound precluded the acquisition of satisfactory elemental (C, H) analyses. However, a freshly purified sample provided a clean *H nmr spectrum. 98 References 1. a) Perlmutter, P. Conjugate Addition Reactions in Organic Synthesis; Pergamon Press: Oxford, U. K., 1992. b) Jung, M. E. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford U. K., 1991; Vol 4, Semmelhack, M. E., Ed., pp. 1-67. 2. a) Kozlowski, J. A. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford U. K., 1991; Vol 4, Semmelhack M. E., Ed.; pp. 169-198. b) Lipshutz, B. H.; Sengupta, S. Org. React. 1992, 41, 135. c) Posner, G. H. Org. React. 1972,19, 1. 3. a) Wender, P. A.; Eck, S. L. Tetrahedron Lett. 1977,14, 1245. b) Wender, P. A.; White, A. W. J. Am. Chem. Soc. 1988,110, 2218. 4. Cooke, M. P.; Widener, R. K. J. Org. Chem. 1987, 52, 1381. 5. Curran, D. P.; Wolin, R. L. Synlett 1991, 317. 6. Kocovsky, P.; Srogl, J. J. Org. Chem. 1992, 57, 4565. 7. Bronk, B. S.; Lippard, S. J.; Danheiser, R. L. Organometallics 1993,12, 3340. 8. a) Piers, E.; Wong, T. J. Org. Chem. 1993, 58, 3609. b) Wong, T., Ph. D. Thesis, University of British Columbia, Vancouver, B. C , 1993. 9. Farina, V.; Kapadia, S.; Krishnan, B.; Wang, C ; Liebeskind, L. S. J. Org. Chem. 1994, 59, 5905. 10. Tanaka, H.; Kameyana, Y.; Sumida, S.; Torii, S. Tetrahedron Lett. 1992, 33, 7029. 11. Piers, E.; McEachern, E. J.; Burns, P. A. J. Org. Chem. 1995, 60, 2322. 12. Piers, E.; McEachern, E. J., unpublished results. 13. a) Corey, E. J.; Katzenellenbogen, J. A. J. Am. Chem. Soc. 1969, 91, 1851. b) Siddall, J. B.; Biskup, M.; Fried, J. H. J. Am. Chem. Soc. 1969, 91, 1853. c) Klein, J.; Turkel, R. M. J. Am. Chem. Soc. 1969, 91, 6168. d) Klein, J.; Aminadav, N. J. Chem. Soc.Chem., Commun. 1970, 1380. e) Naf, F.; Degen, P. Helv. Chim. Acta 1971,54, 1939. f) Corey, E. J.; Kim, C. U.; Chen, R. H. K.; Takeda, M. J. Am. Chem. Soc. 1972, 94,4395. g) Bowlus, S. B.; Katzenellenbogen, J. A. Tetrahedron Lett. 1973,1277. h) Corey, E. J.; Chen, R. H. K Tetrahedron Lett. 1973,1611. i) Klein, J.; Levene, R. J. Chem. Soc, Perkin Trans. II1973,1971. j) Anderson, R. J.; Corbin, V. L.; Cotterrell, G.; Cox, G. R.; Henrick, C. A ; Schaub, F.; Siddall, J. B. J. Am. Chem. Soc. 99 1975, 97,1197. k) Marino, J. P.; Browne, L. J. J. Org. Chem. 1976,41, 3629. 1) Roush, W. R; Peseckis, S. M. J. Am. Chem. Soc. 1981,103, 6196. m) Marino, J. P.; Linderman, R. J. J. Org. Chem. 1981, 46, 3696. n) Nishiyama, H.; Sasaki, M.; Itoh, K. Chem. Lett. 1981, 905, 1363. o) Poulter, C. D.; Wiggins, P. L.; Plummer, T. L. / . Org. Chem. 1981, 46,1532. p) Marino, J. P.; Linderman, R. J. J. Org. Chem. 1983, 48, 4621. q) Walba, D. M.; Stoudt, G. S. J. Org. Chem. 1983, 48, 5404. r) Oppolzer, W.; Mirza, S. Helv. Chim. Acta 1984, 67,730. s) Crimmins, M. T.; Mascaralla, S. W.; DeLoach, J. E. J. Org. Chem. 1984, 49, 3033. t) Lewis, D. E.; Rigby, H. L. Tetrahedron Lett. 1985, 26, 3437. u) Cooper, J.; Knight, D. W.; Gallagher, P. T. J. Chem. Soc.Chem., Commun. 1987, 1220. 14. A sample of this compound was prepared by Dr. Timothy Wong. 15. a) Skerlj, R. T., Ph.D. Thesis, University of British Columbia, Vancouver, B. C , 1988, p. 185. b) Piers, E.; Skerlj, R. T. Can. J. Chem. 1994, 72, 2468. 16. Piers, E.; Morton, H. E.; Chong, J. M. Can. J. Chem. 1987, 65, 78. 17. Still, W. C. J. Am. Chem. Soc. 1977, 99, 4836. 18. Piers, E.; Wong, T.; Ellis, K. A. Can. J. Chem. 1992, 70, 2058. 19. a) Piers, E.; Chong, J. M.; Morton, H. E. Tetrahedron 1989, 45, 363. b) Also see Ref. 9 for a discussion of vinylcopper(I) intermediates. 20. Leusink, A. J.; Budding, H. A.; Marsman, J. W. J. Organometal. Chem. 1967, 9, 285. 21. Occolowitz, J. L. Tetrahedron Lett. 1966, 5291. 22. Piers, E.; McEachern, E. J.; Romero, M. A. Tetrahedron Lett. 1996, 37, 1173. 23. Kiihler, T. C ; Lindsten, G. R. J. Org. Chem. 1983, 48, 3589. 24. a) Yamamato, K.; Miyaura, N.; Itoh, M.; Suzuki, A. Synthesis 1977, 679. b) Midland, M. M.; Tramontano, A.; Cable, J. R. J. Org. Chem. 1980,45, 28. 25. Piers, E.; Tse, H. L. A. Can. J. Chem. 1993, 71, 983 26. a) Rathke, M. W.; Sullivan, D. Tetrahdron Lett. 1972,4249. c) Hermann, J. L.; Kieczykowski, G. R.; Schlessinger, R. H. Tetrahedron Let. 1973, 2433. 27. Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651. 28. Griffith, W. P.; Ley, S. V. Aldrichim. Acta 1990,23, 13. 100 29. Piers, E.; Romero, M. A., unpublished results. 30. a) Alvanipour, A.; Eaborn, C ; Walton, D. R. M. J. Organometal. Chem. 1980, 201, 233. b) Peryre, M.; Quintard, J.-P.; Rahm, A. Tin in Organic Synthesis; Butterworths and Co. Ltd.: London, U. K., 1986, pp. 130-131. 31. Silverstein, R. M.; Bassler, G. C ; Morrill, T. C. Spectrometry Identification of Organic Compounds, John Wiley & Sons: New York, USA, 1981, pp. 194-195, 237, 288. 32. Still, W. C ; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923. 33. Harrison, I. T. Instruction Manual, Chromatotron; Harrison Research: 1985. 34. Bryan, W. P.; Byrne, R. H. J. Chem. Ed. 1970,47, 361. 35. Perrin, D. D.; Armarego, W. L.; Perrin, D. R. Purification of Laboratory Chemicals, 3rd Ed.; Pergamon Press: Oxford, U.K., 1988. 36. Kofron, W. G.; Baclawski, L. H. / . Org. Chem. 1976,41,1879.