New Litho Carbanions and Their Reactions with Carbonyl Compounds

Home , 860s, Dimethoxymethane, Estrone methyl ether

INFORMATION TO USERS

This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material subm itted.

The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.

1.The sign or “target” for pages apparently lacking from the document photographed is “Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.

2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photo graphed the photographer has followed a definite method in “sectioning” the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.

4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.

University Microfilms International 300 N. ZEEB ROAD, ANN ARBOR. Ml 48106 18 BEDFORD ROW. LONDON WC1R 4EJ, ENGLAND 8100236

R o y , G l e n n M ic h a e l * *

NEW LITHIO CARBANIONS AND THEIR REACTIONS WITH CARBONYL COMPOUNDS

The Ohio Slate University PH.D. 1980

University Microfilms International 300 N. Zeeb Road, Ann Arbor, MI 48106 NEW LITHIO CARBANIONS AND THEIR REACTIONS

WITH CARBONYL COMPOUNDS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Glenn M. Roy, B.A., M.S.

* * # # #

The Ohio State University

1980

Reading Committee: Approved By

Dr. Jack Hine

Dr. Matthew Platz

Philip dL MagnusU Dr. Larry Robertson Department of Chemistry To LeAnn and Brooks Michael ACKNOWLEDGEMENTS

I would like to thank the NIH and NSP for financial'

support of my research program. Mr. Dick Weisenberger is thanked for mass spectral support and Dr. Charles

Cottrell for his NMR work. The members of the P. D.

Magnus research group, 1975-1980, are thanked for their close association and camaraderie. Professor P. D.

Magnus is kindly thanked for his constant patience, support and guidance during my degree program. I personally give thanks to my wife, Lee Ann, for typing the rough details of this manuscript and then giving birth to our son, Brooks.

iii VITA

December 5, 1953...... Born, East Paterson, New Jersey

1975...... B.A., St. Michael's College, Winooski, Vermont

1975-1977...... Teaching Assistant, Chemistry Department, The Ohio State University, Columbus, Ohio

1977...... M.S., The Ohio State University Columbus, Ohio

1977-1980...... Research Associate, Chemistry Department, The Ohio State University, Columbus, Ohio

PUBLICATIONS

Selected Cleavage of 1,4-Cieole, G.L. Grady and G. Roy, Cosmetics and Perfumery, Vol. 90, 1975, p. ^9.

A Short Synthesis of Prontalin and Latia Luciferin, P.D. Magnus and G. Roy, J. Chem. Soc. Chem. Comm., 1978, 297.

Methoxymethyltrimethylsilane, A New Reagent for Reductive Nucleophilic Acylation, P.D. Magnus and G. Roy, J. Chem. Soc. Chem. Comm., 1979, 822. a-Lithiohexamethylphosphoramide-Formation and Reaction with Carbonyl Compounds, P.D. Magnus and G. Roy, Synthesis, 1980, 575.

FIELDS OF STUDY

Organic Chemistry. Professor P.D. Magnus

Chemistry. Professor G.L. Grady

iv TABLE OP CONTENTS

Page DEDICATION...... ii

ACKNOWLEDGEMENTS...... ill

VITA...... iv

ABBREVIATIONS...... vi

PART I * INTRODUCTION...... 1

RESULTS...... 6

EXPERIMENTAL...... 70

PART II INTRODUCTION...... 143

RESULTS...... 152

EXPERIMENTAL...... l8l

PART III INTRODUCTION...... 228

RESULTS...... 232

EXPERIMENTAL...... 237

REFERENCES TO PART 1 ...... 245

REFERENCES TO PART II...... 251

REFERENCES TO PART III...... 255

v ABBREVIATIONS

n-BuLi n-butyllithium s-BuLi s-butyllithium t-BuLi t-butyllithium

THE tetrahydrofuran

MgSOi* magnesium sulfate

Na2S0i» sodium sulfate

BP3Et20 boron trifluoride etherate complex

EtOAc ethyl acetate

Et20 diethyl ether

DMSO dimethyl sulfoxide

TMEDA N,N,N’,N'-tetramethylethylenediamine

LDA lithium diisopropylamide

TLC, PLC thin layer chromatography, preparative layer chromatography mm millimeters of mercury mmol millimole (s) br.s, t, t, m broad singlet, doublet, triplet, multiplet (NMR) br. s, m, w broad strong, moderate, weak (IR)

-TMS trimethylsilyl group

-OMe, OAc, Ac methoxyl group, acetoxy group, acyl group

Me, Et methyl group, ethyl group

vi R.T. room temperature

PCC pyridinium chlorochromate

LAH lithium aluminum hydride min minutes h hour petrol high boiling petroleum ether (60°-90°C)

CF3CO2H trifluroacetic acid

MeOH methanol

CaH2 calcium hydride

NHi»Cl ammonium chloride

NaHC03 sodium bicarbonate

vii GENERAL INTRODUCTION OF EACH PART

PART I The first part of this dissertation deals with homologation of aldehydes and ketones to their respective aldehydes or ketones. This trans formation is carried out using a method that was developed in our laboratory. It was our intention to explore the scope and limitations of the reagents, chloromethyltrimethylsilane and a-chloroethyltrimethylsilane as they are used to prepare new epoxysilane intermediates of important synthetic potential.

PART II The second part describes the preparation of a new reagent, methoxymethyltrimethylsilane. The lithiation of this silane had been attempted in other laboratories without success. We were successful in adding this lithiated species to several carbonyl compounds thus illustrating the use of a new homologation reagent.

PART III The third part merely describes an unprecedented observation that a common solvent, HMPA, can be lithiated and added to carbonyl compounds. This observation now serves as a caution to those synthetic chemists who use this solvent in alkyl lithium media since it not only acts as a solvent but also as a reagent that can prepare new intermediates never before studied.

viii PART PART I

INTRODUCTION

The recent surge in the development of organosilicon chemistry into the more general areas of organic synthesis has produced many interesting reactions that demonstrate the synthetically useful properties of silicon.1 Apart from the Peterson reaction2 and associated procedures3 little synthetic use is made of organosilicon chemistry since silicon reagents to a large extent are comparatively inaccessible to the non-specialist.

The use of silicon reagents in modifying the reactivity of an organic substrate has presented excellent potential for development. Silicon functionalized epoxides have been shown to be useful in the formation of ketones,11 >5 olefinsf’7 vinyl ethers and halogen compounds.8 The studies of a, 3-epoxysilanes by Hudrlik0 illustrate reactions with nucleophiles under acidic conditions giving vinyl bromides, enolacetates, enolethers and enamides. The preparation of the substrate silane epoxide was usually accomplished by epoxidation of vinyl silanes9 which them selves are not readily procurred, or by hydrosilylation with MeCl2SiH catalyzed by chloroplatinic acid, reaction

1 with MeMgX and epoxidation.10 A new method for generation

of a,3-epoxysilanes was recently developed1* that circum

vented the laborious procedures known. Many adducts have

been prepared and a few limitations on the preparation

of a,3-epoxysilanes by this methods have been described.12

The first major synthetic use of a-silyl carbanions was

described by Peterson.2>13

Deprotonation of readily available a-chloromethyl-

trimethylsilane or a-chloroethyltrimethylsilane1 *»15

with sec-butyllithium in THF at -78°C with TMEDA gave

lithio reagents 1 (CTC) and 2 (MCTC) that add cleanly •V *v to carbonyl compounds to give a,3-epoxysilanes. Several

new, unreported adducts and their pertinent use in

synthesis are to be described here.

The reagents illustrate the process of reductive

nucleophilic acylation.16 While colleagues have prepared

other adducts as well, to the present time no incorpora

tion of the method has appeared in any natural product

synthetic strategy by other authors.

However, we reported that Prontalin 3 an aggregation

pheromone, and Latia Luciferin 4 a bioluminescent substrate

can be prepared17 using this organosilicon chemistry.

In the last few years there has been a prolific number of publications describing the use of organo

silicon chemistry in synthesis. Nearly every journal

that synthetic organic chemists read contains papers 3

Me3SiCH2CI s-BuLi Me-aSiCHCI RCOR1 3 I w IMP*" Li R2 SiMe3 m EPA Cl I R1 Me,SiCH - Me R 0 Me ° I »" BuLi Me3SiC-Me 2 I b ° Cl 7H P Li R2 TtA&DA r 2 Si Me3

Me^SiO, Me HO. Me Me^SiO Me Me,Si , Li C')03 HO {\ + X (ii) N0BH4 Me Cl CHO

Me Si Me: Me SiMe:

. OCHO OCHO

HC0 2H Me OH Lotia Luciferin 4 4 V»H. *• devoted to the use of silicon based chemistry for the

construction of organic molecules. Recently two very

substantial and authoritative reviews have appeared18

that cover all the main aspects of organosilicon chemistry

as applied to synthesis. Since the carbon-carbon bond

is a focal-point of organic synthesis, it is important

to have many and varied ways to construct this bond in

a predictable fashion. There are two main ways in which

the chemistry of silicon can be used to form carbon-carbon

bonds', they are the addition of carbon electrophiles

to unsaturated organosilicon compounds, and the addition

of carbon electrophiles to a-metallosilanes. There are,

of course, other ways, but the two mentioned constitute

the overwhelming majority.

In the first and second part of this dissertation

the use of the a-silylcarbanion for the construction of

C-C bonds is discussed. Even though silicon is more

electropositive than carbon it is able to sustain

adjacent carbon-metal bonds. Various reasons have been

advanced to account for this stabilization, the most popu-lar being overlap of the carbon-metal bond with

the empty d-orbitals on silicon. Alternative explanations

have been advanced but will not be pursued here.18

Apparently the degree of stabilization is not as great as that imparted by sulfur or phosphorus, but, unfortunately, there is no quantitative data to support this qualitative observation. During the past three years a number of references

to the reagents described here have been reported.19

We have prepared many homologated aldehydes and ketones

from new epoxysilane precursors. These new adducts are

to be reported with details concerning their hydrolysis.

The fact that these reagents add to so many aldehydes

and ketones to give epoxysilanes in good yields is an

indication of their relatively high nucleophilicity and

low basic character. These synthetic products illustrate new uses of organosilicon reagents and offer an improve ment over existing methods, in that a,3-epoxysilanes

are now readily available and easily hydrolyzed with

formic acid or known procedures, to give aldehydes or ketones.

Some new epoxysilanes prepared with CTC are now to be described. These adducts along with many others represent a masked aldehyde moiety. The routine hydrolysis of the epoxysilanes with formic acid, in some

cases, affords not only the homologate aldehyde but the

carboxylic acid as well. Formic acid is just too strong an oxidizing media for some of these hydrolysis steps.

Therefore, the usual mineral acid procedures11* are used and specified in the experimental section. The "over” oxidation was no problem in the hydrolysis of MCTC adducts to their methyl ketones and will be described later. RESULTS

While exploring the nucleophlilicity of CTC some easily enolizable ketones were used as substrates. When distilled 2-methylcyclohexanone 5 was treated with

1.5 eq of CTC, the epoxysilanes 6 were obtained in 77$ distilled yield. Perhaps about 10-15$ enolization occurred as judged on TLC. The pure epoxysilanes were treated with formic acid at R.T. and gave a mixture

(111) of the aldehydes 7 contaminated by about HQ% carboxylic acid (overall 70$ isolated yield). Perchloric acid hydrolysis of 6 gave the aldehydes cleanly in 869o yield.

H 6 7 H

5 CHO

6 Cycloheptanone 8 was treated with 1.5 eq CTC and gave the pure epoxysilane in 60$ distilled yield. Some enolization of the flexible skeleton is not unexpected accounting for a lower yield. Hydrolysis of the adduct

9 with perchloric acid in 20$ aqueous THP gave the aldehyde 10 in 88$ isolated yield. Formic acid

9 10

g hydrolysis gave the aldehyde plus an appreciable amount of carboxylic acid.

The IR spectrum of all these hydrolysis procedures is extremely informative in that the loss of silicon absorption (1250, 850 cm-1) and growth of carbonyl absorption (1700-1725 cm"1) can be monitored. When norcamphor 11 was treated with 1.2 eq of CTC the epimeric epoxysilanes were obtained in a 76$ distilled yield. The product was one homogeneous spot on TLC and gave excellent mass spectral data. Camphor •

49 a and nopinone 49 b gave 20-40$ yields of epoxysilanes.

When 12 was treated with perchloric acid in 20$ aqueous

THF the aldehydes 13 were isolated as a 3.6:1 epimeric mixture in 72% distilled yield. The aldehyde protons

CHO 12 + 13

appeared as singlets unsplit by the a-hydrogens. The

IR data was strongly indicative of an aldehyde functional group. When 12 was treated, formic acid at R.T. no aldehyde was formed. Curiously enough no carboxylic acid was detected either. The IR and NMR data gave indication of a saturated ketone . A hindered, enolizable enone was chosen to react * * with CTC. When (d)-Carvone 16 was treated with 1.5 eq

CTC the epoxysilane 17 was obtained in 76% distilled yield. The NMR data indicated that one epimer was produced (singlet appeared at 6 0.15). The hydrolysis of 17 with perchloric acid in aqueous THF gave a 7Q% distilled yield of two aldehydes evident by IR and NMR data. No traces of carboxlic acid were present. The tetrasubstituted enal 18 was the major product with about 30% of the 3 ,Y unsaturated aldehyde present.

TIMS 17

HCI04 THF- 10

The hindered enolizable enal 19 was treated with

1.5 eq of CTC and the epoxysilanes 20 were obtained in

95$ distilled yield. The planar array of the enal appears to ease the sterically encumbering approach of the bulky nucleophilic CTC. The product was a 211 ratio of epimers about the carbon bearing silicon as evidenced by the two signals in the NMR.

Hydrolysis of 20 with perchloric acid in aqueous

THF gave exclusively the 3,y unsaturated aldehyde 21

(9^% crude.) as evidenced by the triplet at 6 9.7 (J=6Hz)

/CHO

in the NMR spectrum. This product was pure by TLC and

IR. None of the conjugated aldehyde was detected.

This should be compared to the mixture obtained with the hydrolysis of 17. The aldehyde 21 did not withstand distillation and decomposition product were obtained

(loss of triplet in NMR on distilled material). 11

In our laboratories a rather unique spiro furanone 22 was prepared.51 Since our purpose was to explore the utility of CTC as a homologation reagent we chose this compound as another test for the nucleophilicity of CTC.

CTC is 7j0

22 23

24 12

In this dissertation only aldehydes, ketones and diones have been reacted with CTC or MCTC. When 22 was treated with 1.5 eq of CTC the epoxysilanes 23 were isolated in only 15# yield by PLC as two fast Rf products with two silicon signals in the NMR. The remainder of the material balance was about 15# spiro furanone and about 70%> crystalline aldol condensation product 24. The

NMR, IR-, mass spec, and combustion analysis data were consistent with that structure proposed for 24. The adducts 23 were treated with perchloric acid in aqueous

THF and gave after 5 h at R.T. a product that was not the desired aldehyde, 25.

- * H > HrHMS * 2 3 25 13

Estrone methyl ether 26 was added slowly to 3 eq w v of CTC at -78°C. The usual workup gave cleanly a 40%

Isolated yield of the pure epoxysilane 27 in a 311 isomer

ratio. The remainder of the material balance was pure

starting material 26. The quality of s-BuLi in these

reactions is very critical since we believe optimized

yields of 27 and other steroid additions of CTC or MCTC

can be improved.

The epoxysilanes 27 were cleanly hydrolyzed with

formic acid to the C-20 formyl steroid 28 in 98% yield with the 8 -configuration at C-17. No carboxylic acid was detected in the NMR or mass spectrum.

SiMe

MeO MeO

CHO 14

The androstadienone 29 was added slowly to' 3 eq of v w v CTC at -78°C and provided a 35% isolated yield of epoxy silanes 30. Formic acid hydrolysis did not provide 31.

Therefore, the usual mineral acid procedure11* (HC1, aq

MeOH, or HClOi*, aq THF) was employed and this also did not provide any of the steroid 31.

TMS CTC

MeO

29 HCI/ a MeOH

CHO 15

Just after the Magnus paper came out on CTC we

received a letter from Dr. A. deMeijere at Universitat

Hamburg. He asked us to react CTC with a C2 .2H para-

cyclophane-l,9-dione. The premise was that its reactivity

with CTC cannot be compared to the a-tetralones. A

brief understanding of the cyclophane nucleus is necessary.

While D. J. Cram32 was studying paracyclophanes the first

compound of this class was reported by another group.33

Notable among those containing parabridges are cyclic

diesters,34 ketones,33 ethers,36 and ketols.37 They

are found to be non planar crystalline structures in

which the carbon atoms of the ring bearing CH2 are out

of plane by 0.013$ and the angle of distortion is 11°.

It is clear that repulsion exists between the two

benzene rings in this compound, increasing the bond

angle between the methylene bridges and the benzene rings

(0x) from what would otherwise be 90° to 115°, thus only

6° of strain in this angle.38 The remaining strain

incurred by repulsion between the two rings is distributed

almost equally between the other bond angles 02 and 03.

These configurations illustrate the principle that the

distribution of strain between several bond angles is more economical in terms of energy than concentrating

the same number of degrees of strain in only one bond

angle. 16

When the cyclophanes 32 we received were reacted with 2.5 eq of CTC a clean bis-epoxysilane product was isolated in 89% yield. Unfortunately this represented a mixture of both (1,9) and (1,10) bis-epoxysilane isomers since the 60140 mixture of diones was originally supplied to us as such. Also unfortunate was the fact that hydrolysis of these epoxysilanes did not provide the desired homologate bis-aldehydes 34. It is believed by spectral evidence that the cyclophane nucleus has ring expanded to give a mixture of ketones, 35-38.

It is surprising to see the rapidity with which CTC reacted with the dione. A clean reaction was evident by TLC. The mass spectral and NMR data were in agreement with the desired bis-epoxysilane products 33.

+ 32

0 4 0 % bO'/ 3.09X. 1.S5&.2.83 X. 1.S5&.2.83

CHO

35-38 Since the diones 33 had reacted well with CTC but the adducts did not provide the homologous aldehyde another substrate was studied. The most available was anthraquinone 39 and when treated with 2.5 eq of CTC provided a 88$ recrystallized yield of 40. A repeated treatment with 4.5 eq of CTC did not provide the bis- epoxysilane adduct. It would appear that the remaining carbonyl in 40 is less electrophilic than the quinone.

This may be due to the resistance of the ir clouds of the aromatic rings, and carbonyl benzylic carbon to support a positive charge. When !lQ was treated with formic acid at R.T. no aldehyde or carboxylic acid was indicated by NMR or IR data. Only decomposition products of the anthraquinone system were isolated.

J-L T M S

40 seq. 0 0 39

no HCI MeOH 19

When 40 was treated with 4 eq of boron-trlfluoride etherate complex (distilled from CaH2) in aqueous methanol at 0°C only starting material remained. The

Y positive charge that must develop is unfavorable.

Warming to 25°C and 50°C did not produce the desired dimethylacetal. The resistance to these conditions is interesting. The recovered material 40 (homogeneous by TLC and NMR) was treated with 3N HC1 in aqueous THP or ION HC1 for quick conversion to two products that were slightly soluble in all available solvents.

IR data indicated the total loss of both silicon absorption (1240 and 850 cm-1) and carbonyl absorption

(1695 cm-1). The mass spectrum of the major product 41, by TLC, gave an observed molecular weight v w of 208 g/m. MCTC did not give any isolated mono- or bis-epoxysilane upon reaction with 39. The premise that was described for the behavior of 32 with CTC suggested that an a-tetralone 42 could not serve as a model for preparations of 33. This premise was tested and showed that CTC was not very nucleophilic towards 42. Enolization similar to that of 46 (in this case, 62), is observed with many polar products evident by TLC, The labile epoxysilane

CTC

42 MS

HCIO,

0 ^ 0 «

CHO ' 43 (about 20%) was hydrolyzed during PLC purification on silica to give the aldehyde in 26% isolated yield based on 43 (4% from 42). The remaining polar products gave IR data consistent with aldol condensation products.

The aldehyde 4j4 gave good NMR and IR data that were consistent with the desired product. 6-Tetralone 45 reacted with CTC only to give 45 plus aldol condensation products. Enolization to 46 is energetically easier in the 0-case 45 than the a-case 42 because the carbonyl a-hydrogens are then benzylic.

46 62

CT C 22

A recent report18 was studying the homologation of 1-pyrene carboxaldehyde. As will be seen later,

MCTC provided the desired methyl ketone contrary to the claim that MCTC was a poor reagent. When the aldehyde was treated with 3 eq of CTC, and worked up as usual, it provided a 81 % isolated yield of epimeric epoxysilanes as a liquid. Formic acid and perchloric acid hydrolysis of ^7 did not provide ^8 as judged by NMR and IR data. Perhaps this aldehyde is labile to the conditions employed and subsequently aldol condenses with itself.

HCIO

4 7 23

These and other examples11* of the reactions of CTC

with aldehydes^ and ketones illustrate the wide range of

carbonyl substrates which can be converted to a,(3-epoxy-

silanes by a-lithio chloromethyltrimethyl silane. The

limitations of the reaction are the poor yields of epoxy-

silane obtained from sterically hindered (e.g. camphor) 49a and readily enolisable (nopinone) 49b substrates. In the

latter case the competing reaction of CTC as a base becomes

important. This is probably because CTC is a bulky nucleo

phile and aldo quite a strong base.

4 9 a

The utility of the reaction is enhanced because of

the further transformations which a,(3-epoxysilanes may

undergo. These have already been described in other

areas.8 >9>12 2H

The overall transformation of a carbonyl group to an homologated aldehyde is described as reductive nucleophilic formylation, CTC reacting as a formyl anion equivalent e (HC — 0) .

The mechanism of the hydrolysis of epoxysilanes has not been investigated. Stork, who first described the reaction,10 having prepared the a,(3-epoxysilane by epoxidation of a vinyl silane, has suggested that nucleophilic attack occurs at silicon under acidic conditions to give the enol as the initial product, which then rapidly tautomerizes to the keto form.

In this case the oxygen atom of the original carbonyl function becomes the oxygen atom of the aldehyde group.

However, the diol A has recently been reported as the initial product of hydrolysis of a,j3-epoxysilanes using extremely mild acidic conditions.5 This suggests that the enol is formed by (3-elimination of trimethylsilanol from the diol A.

e = bf3 , h 25 The new unreported MCTC adducts are now described.

Cycloheptanone ^8 reacted cleanly with MCTt to give the epoxysilane 50 in 75% yield. The hydrolysis

(formic acid) to the methyl ketone 51 was quick and clean, however, normal evaporation isolation was difficult owing to the volatility of the ketone. An aqeuous workup and extraction into diethyl ether gave a isolated yield of the methyl ketone.

HC0?H

The ketone ^-methylcyclohexanone Jj. is a good substrate to test the nucleophilicity of MCTC since such a ketone is very apt to enolize rather than add a hindered nucleophile. However, MCTC reacted cleanly to give the epoxysilane 52 as isomeric mixtures in 82% isolated yield. Formic acid hydrolysis of this mix ture gave a single methyl ketone 53 in 56% yield. 26

M e ^ T M S 52

Norcamphor 11, an enolizable ketone, was treated with

2 eq of MCTC and provided a 68% distilled yield of 54 w w as a mixture of enantiomers as one homogeneous spot

on TLC. It should be remembered that camphor 9a and nopinone 49b reacted poorly with CTC. The steric hind rance has shown a dramatic effect. The epoxysilanes 54

H C Q ^

TMS CH. 54 0 55 The enone (d)-Carvone 16 reacted cleanly with MCTC to give, after workup, the pure isomeric epoxysilanes 56 in 86% crude yield (pure by TLC). Formic acid hydrolysis gave the methyl ketone 57 as a mixture of conjugated IMMK. * and unconjugated isomers in 70% distilled yield.

56 + 57

Myrtenal 19 was treated with 1.5 eq of MCTC and the epoxysilanes 58 were Isolated as diastereomers in

84$ distilled yield.

Hydrolysis of 58 with formic acid gave a mixture of the geometric E,Z-isomers as well as the 3,Y unsaturated (deconjugated) methyl ketone. This was evident in the NMR data obtained on crude material.

Treatment of 58 with HC104 in 20$ aqueous THP gave exclusively the deconjugated isomer.

(V*

The ketoanthracene 42 suffered enolization bv MCTC as expected. The epoxysilane 60 was obtained in a 40$ yield. Hydrolysis gave the pure methyl ketone in 65$ isolated yield. Some enolization to the styryl enolate

62 caused a lower yield of the desired adduct 60. 29

(XXH

J l L

A recent report10 claimed that the methyl ketone ■

64 from 1-pyrene carboxaldehyde was difficult to prepare by the variety of known homologation procedures as well as using MCTC. We were able to prepare the epoxysilane 53 as a mixture of isomers in 80# yield.

Formic acid hydrolysis was quick (10 min) and clean to give the required methyl ketone 64 in 82# yield. No problems were encountered in the usual MCTC reaction, contrary to the claim in the literature.18 30

o o HCQH oo •TMS

O O o o

The furanone 22 prepared by Gange and Magnus51 was an interesting starting material for epoxysilanes from MCTC. Unfortunately, no methyl ketone or epoxy silane could be prepared. Enolization occurred and aldol condensation results. The dimer 24 was crystal- •WV line and gave good spectral data in agreement with the designated product.

MCTC

0 22 24 With the plethora of new epoxysilanes prepared by our methods, it was decided to utilize these intermediates for the synthesis of some natural products, namely steroid

derivatives.

Estrone methyl ether 26 provided the best starting material for a short synthesis of 19-nor-progesterone.

The requisite epoxysilane was prepared in k0% isolated yield with 3 eq of MCTC reagent.15 The use of 5 eq or more does not increase the reaction yield. Possibly the alkoxide concentration of the sec-butyllithium (in some cases as high as 30$) has inflicted an appreciable amount of enolization so frequently encountered with the

17-keto position.

Some experiments were attempted to improve the yield of the CTC and MCTC reactions with estrone methyl ether.

Initially, it was chosen to form a zinc complex between

CTC and ZnCl2 • A complex such as this promoted

Li i ZnCL ZnCI i Me3SiCHCI Me3SiCHCI + Li Cl 32

nucleophilic attack of silicon based anions at the 17- keto position of certain steroids to provide adducts of

the type 65. When CTC was prepared as usual (-78°C ■*

-40°C -* -78°C) and ZnCl2 was added with subsequent

stirring to -40°C (solubility is noted) the steroid was

added. Stirring at -40°C then warming to R.T.

MeO OH

MeO M e O 66 followed by the usual workup did not give any of the

expected epoxysilane as per TLC. The one product that

was isolated in about 25$ yield was found to be 3,17.6-

estradiol 3-methyl ether 66, the result of reduction

(mpt. 120°-121°C mpt. lit. ll8°-119°C)?0 Since ZnCl2

and CTC did not provide a more powerfully nucleophilic

species to react with estrone methyl ether it was chosen

to try to increase the electrophilicity of the 17-keto

carbonyl. One such method would be to prepare a solution

of the ketone with 5 to 10 mole percent of anhydrous

LiClOi*. When this type of solution was added slowly to

a stirred solution of the anion only a small amount of

epoxysilane was detected on TLC. Even at 20 mole percent

of LiClCU no increase in the yield of epoxysilane was

noted. In fact, as many as six products were observed

on TLC making this method inferior to the normal CTC

and TMEDA conditions that provide the epoxysilane. No

•experiments with MCTC and ZnCl2 or LiClOu were carried

out. It was thought perhaps that an inverse addition

(anion to steroid) might provide an increased yield of

epoxysilane. However, this attempt gave about seven

products on TLC and once again proved to be inferior to

the normal mode of procedure. 34 In the present preparations of the epoxysilane 67 only a 35$ yield could be obtained. This proved to be the limiting step in the strategy for norprogesterone 69 involving Birch reduction of the estrone epoxysilane followed by hydrolysis as outlined briefly below.

M e , Si Si Me- Me

MeO MeO

0 H

MeO 69 35

The Birch reduction of the epoxysilane 67 was carried out similar to that of Dryden**9 and gave a pro duct that could not be converted to norprogesterone 69 or its bis-DNP derivative. Presumably the Birch reduction procedure gave an isomeric silane that is not converted to the methyl ketone either because the epoxysilane is reduced or rearranged. The obvious alternative was to prepare the ketal of the available methyl ketone and then do the Birch reduction. This ketal required harsh conditions and long reaction times for an overall 60% isolated yield. The oxidative pathway with steroid 68a would offer higher overall yields for norprogesterone according to the literature.26 36

It was found that a,B-epoxysilanes were conveniently converted to their ketals.12

Ri" 0. ,H XH XH H *XCH2>„ TsOH . 0H, a n = 2,3 K = S,0

HO C)H 67 68 and no 7 0 TsOH* PhH 37

e<3 Si Me MeSiO. 0 ^ 3 , V ' M e L]/NH3

f t "'H •67

Li/NH ><

SiMe H Me

The isomeric silane just mentioned could be accounted

for by the scheme above. Reduction of the epoxide could give ring opening to an intermediate which would then lose

Me3SiOH. The final olefinic product (isolated in 56% yield) depicted here gave mass spectral and NMR data to support this structure. IR and NMR data confirmed the presence of the enone from reduction of the A ring but no absorption (1720 cm"1) for the C-20 ketone was observed. Since attempts to take the a-methyl substituted epoxysilane 67 through Birch conditions were unsuccessful it was necessary to prepare the C-20 ketal. Pure steroid epoxysilane was treated with ethylene glycol and catalytic p-TsOH in benzene at 80°C for 3 days. Only a slow conversion to the methyl ketone 68 was observed by TLC, mass spec, and IR data. Therefore, the reaction was repeated in toluene. In these reactions also, the silanol residues depress the rate of ketal formation.

The key to success for a short synthesis of 19-norprogesterone 69 would be the ability of an epoxysilane to withstand liquid ammonia dissolving metal reduction conditions. In effect, this would carry a masked carbonyl moeity through a crucial step normally reserved for ketals or other protecting groups.

Several experiments were conducted using lithium in liquid ammonia with solvent combinations of THF, t-butyl alcohol and ether. It Is believed that 18 eq of lithium wire (Dryden's conditions) completely cleave or reduce the epoxysilane in such a manner that subsequent hydrolysis does not provide the 20-keto carbonyl. Also,

4 or 8 eq of lithium is unsuccessful in preparing 19-norprogesterone. When two equivalents is used incomplete reduction of the aromatic ring occurs. Some of the reduced product was isolated and hydrolyzed once again not revealing any isolable nor-progesterone. It may be 39 that reduction of the epoxysilane under these conditions is faster than or even competes with the reduction of the aromatic A ring. Although the infrared data did not indicate a strong OH absorption it was thought perhaps we had isolated A14-19-nor-pregnen-20-ol-3-one 68a (after hydrolysis to the enone). If such were the case, procedures are known to oxidize this position and give 19-nor- pregesterone. However, both pyridinium chlorochromate and chrominium trioxide-acetic acid conditions did not provide NMR or IR data consistent with the desired steroid.

At this point it is not known with certainty what the

fate of the epoxysilane was nor what steroid had been prepared.

With NMR data of the enol ether obtained from a

Birch reduction using 8 eq of lithium wire it was assumed that the epoxysilane was intact. Two clean singlets

(6 0.0, 0.1) seemed to indicate that no harm had been done to the epoxysilane. However, data from hydrolysis procedures did not afford evidence of a 17-acetyl substi tuent, meaning in fact that these signals may merely be rearranged or reduced structures of a silicon containing substituent. This observation was indeed disappointing since routes around this problem lengthen the synthesis and fail to add any novelty to the strategy.. Two possible alternatives were possible, both of which are known procedures. This in effect then makes our epoxysilane a precursor to the total synthesis of 19-nor-progesterone.

The first is to hydrolyze the epoxysilane to the 20-methyl

ketone 68 and ketalize, then carry out the Birch reduction,

or secondlyj the C-20 methyl ketone can be Birch reduced

and subsequently oxidized to orovide 19-nor-oroeesterone.

M e Q T 68 a 70 a OHCI ' MeOH 2)CrO*

19-Norprogesterone was the first 19-norsteroid in

which removal of an angular methyl group was shown to be

accompanied by a remarkable increase in biological

activity. In 19^, M. Ehrenstien23 first prepared an

amorphous 19-norprogesterone from strophanthidine (Scheme

1) and found it at least as active as progesterone.2u

This substance was obtained in crystalline form and shown

to possess the l^-iso-17-iso orientation with the Degradation of Stropnantmdine (Enrenstein, 1944)

CO,H v3 COCH

19-nor-140-17

S c h e m e i 42

10-3-configuration.25 The crystalline isomer exhibited eight times the biological activity.26 Then 19-norpro gesterone with all the correct stereochemical centers was described27 thus the mode of synthesis automatically established the "normal" configuration at all asymmetric centers with exception of C-10. Fortunately this comes out during hydrolysis to give the trans junction. This prompted the synthesis of a large number of 19-nor analogs of steroid hormones and the first to find clinical appli cation has been 19-nor-17a-ethinyl testosterone (Norlutin), a substance readily available from estrone.2ea» 281:3

The ethinylation of estrone methyl ether proceeds in 90$ yield.2 9a

69 The problem was to develop conditions for deacetoxy-

lation without producing D-homo rearrangement.2913 The

Glaxo group290 developed an elegant condition (calcium/

liquid ammonia) for removal of the acetoxy function in

ring C ketols of the steroidal sapogenin series but

mechanistically this should be equally applicable to

other ketol acetates.

Ca N H ^

Progestational activity is extremely specific and

is limited to the natural hormone, progesterone, and a

few of its dehydro derivatives.3oa Biological tests

show that 19-norprogesterone is potent30*3 while 68 is

devoid of progestational activity.300 The question of

the importance of 19-nor methyl groups in cortical hormones has also been investigated. 44

The androstadienone 29 was a good material to prepare progesterone. The requisite isomeric epoxysilanes 71 were obtained in 35% isolated yield. Hydrolysis with 6N HC1 in ether-tetrahydrofuran gave a clean product in 82$ yield.

The purified steroid was homogeneous on TLC and identical to an authentic sample.

TMS Ha MeOK

72 Now that CTC and MCTC have been discussed in some length, some other studies in organosilicon chemistry were undertaken.

A variety of a-heterosubstituted epoxides of the type below have been found to undergo stereospecific a-lithiation by use of such bases as n-BuLi, t-BuLi, or

LDA in solvent combinations of hexane with THF, Et20

or TMEDA at temperatures of -78°C to -110°C.20»21

R ^ S i P h j

S O j F h Ph COiEt CNF Such a route to a-metalated. epoxides bearing a variety of heterosubstituted group is valuable as a source of nucleophilic epoxide synthons. The conclusions of many authors state such epoxide anions are often only fleeting intermediates whose ultimate fate is cis-trans isomerization, reprotonation or carbenoid formation.

For successful alkylation their composite studies indicate that the epoxide must be rapidly metallated and the intermediate must be sufficiently stable to react subse quently with the electrophile. In order to further test the nucleophi-licity of CTC it was decided to try reactions on substrates containing two functional groups. With this study it was found that CTC is a rather indiscriminate nucleophile. No conditions could be found whereby a selective mono-adduct epoxysilane was isolated. With 1.2-3 eq of CTC reagent both mono- and bis-epoxysilanes were detected by mass spectral data. The IR data never showed total consumption of either carbonyl absorption. The following is repre sentative of those compounds that could not be selectively reacted with CTC.

CHO The most promising stabilizing groups for inter mediate lithio epoxides seem to be those containing

Second Row Elements (Si, P or S).

Since CTC provided good yields of epoxysilanes their lithiation was studied. Described earlier22 were epoxy silanes which also were shown to metallate with t-BuLi or s-BuLi in THP/hexane. While extending the methodology of CTC1U to chloromethyldimethylphenylsilane 76 the corresponding epoxide could be prepared. The products could not be easily hydrolyzed to the aldehydes because the silyl residues were difficult to remove.

Also, the dimethylphenylsilyl epoxide 77 did not alkylate as judged by recovered starting material. Several alkylation attempts were unusccessful and it was thought

s-BuLi PhMe2SiCH2CJ » PhMe,SCHCI c i 76 Li

77 > ~ -2)RX 48 that the lithio species (if formed) is so stabilized by the phenyl group through Si 3d orbitals that its reactivity is poor with alkyl halides to provide 78. No attempts to decompose the epoxide was made.

As mentioned earlier in the introduction MCTC was a useful reagent for the preparation of Prontalin, an aggregation pheromone. It was thought that other alkyl substituted carbanions stabilized by silicon could be prepared. Simple alkylations of CTC do not provide good yields of the desired reagents.

If the ethyl analog could be prepared in good yields one would have a new tool in synthesis for the preparation of homologate ethyl ketones from carbonyl compounds.

Such a reagent would be useful for the synthesis of another pheromone, multistriatin.uu

U R r=m?,e+ I O V J > MejSiCHCI —— Me3SiCHCI X= I,B r 1

Frontalin Multistriatin Normal alkylations of CTC would not be adequate for scale up. A study was done to provide the desired reagent in situ by alkyllithium additions to a vinyl silane. This is a known concept21* whereby telomerization of silyllithium reagents is a severe problem.

For n-BuLi addition only 10$ is achieved and the rest is telomer product.1*1 For t-BuLi reactions in ether only 25$ addition resulted.1*2 Methyllithium with the chlorovinylsilane 29 at -50°C in THF with TMEDA did not add to the vinylsilane rather 1,2 addition to give the carbinol

80 was observed with no silicon incorporation (8l) in the product. The same reaction in diethyl ether at -78°C to 50°C without TMEDA gave the same 1,2 addition product.

• U

it 50

If the reaction conditions are at -78°C to -20°C or initially at -20°C in THP with TMEDA and methyllithium the silylalkynyl carbinol 80 is obtained in good yield.

Presumably as the temperature rises to -20°C the elimination to silylacetylene is faster than formation of the desired reagent or even telomer. The mechanism proposed is purely speculative, yet the initial addition of methyllithium did evolve gases for a time. All data indicate the pro ducts of this effort were not the desired epoxide.

Similar reactions of alkylvinylsilanes1*x and -phos- phines140 were shown to be more complex in that the initial

Michael adducts subsequently underwent telomerization with additional vinyl compound.

z c h =c h 2 ZCH=CH2+RLi > ZCHLiCH2R ------> ZCHCH2R > etc.

CH2CHLiZ

Z=(n-C4H9)2P, (CH3)3Si

Trialkylvinylsilanes are effectively activated by the replacement of an alkyl group by an electronegative chloro or alkoxy substituent."3 Such activation is so effective, in fact, that various Grignard reagents competively add to the S-carbon atom of these vinyl- silanes rather than causing displacement of the leaving groups. The effect of electron withdrawing groups on the ability of silicon to stabilize an adjacent negative charge in a S-addition is remarkably similar to the 51 of these substituents on the metalation of methylsilanes of the type (CH3 )3SiX*, i.e., these metalations, which are facilitated relative to the metalation of tetra- methylsilane, occur competitively with displacement.

R2SiCH=CH2 + R'MgCl ------> RgSi-CHCHgR' X X MgCl

(X = C l , OR)

Olofson19^ has reported that treatment of chloromethyltrimethylsilane with lithio-2 ,2 ,6 ,6-tetramethyl- piperidine ("Harpoon base") in the presence of a carbenoid trap such as cyclohexene gives the cyclopropyl- silane 83 (ca. 25%) as shown below.

Me3SiCH2CI 'X^P^S-Me 3

8 3

_, s-BuLi TMS. M^Si*CH2CI — * o §1 TMS The thermal carbenoid formation from the anion generated with s-BuLi and TMEDA was attempted under a variety of conditions with no success. Merely dimerization of the reagent ’’carbene" was observed. Curiously the Rf on TLC with silica gel was identical to the silylcyclopropane, prepared by Olofson. Thus the supposed carbenoid CTC was effectively trapped by itself to give the bis-trimethylsilylethylene 84. CTC generated from

In our experimentation the olefin was added at -60°C followed by reflux to 50°C immediately. The major product was bis-trimethylsilylethylene. The next reaction in pure olefin-free dry hexane at reflux gave, however, the same ethylene as the major product, and not the silyl cyclopropane. The development of these two identical

Rf products on silica gel was different. The silyl cyclopropane was purple while the ethylene was grey on I2 treatment, H2SO4 spray then heating. Overnight heating, overhead mixing, and temperature changes gave no change in reaction products. A different reacting species appears to be operating in the two cases or the dimerization of CTC in s-BuLi media is faster with itself than cyclohexene. 53

The Peterson Reaction

The first reported a-metallosilane was prepared by

Whitmore and Sommer in 19^6.11 Treatment of chloromethyltrimethylsilane with magnesium in ether gave the stable Grignard species. Surprisingly this reagent did not reappear until 1968 when Peterson developed an alkene synthesis based upon its reaction with aldehydes and ketones.2 The Grignard reacts with aldehydes or ketones to give after mild acid work-up 3-hydroxysilanes 8 5 .

rf MgjBCHpl lvte£iCH2M9Cr Frequently when R 1 and R2 are part of a ring the adduct is not stable, and eliminates trimethylsilanol to give an alkene 86. The overall result is the replacement of the carbonyl group by a methylene group, and as such the Peterson reaction constitutes a silicon version of the Wittig reaction. The cuprate of this Grignard reagent has not been reported. A colleague was unable to prepare it with Cul or copper (I) n-propylacetylide and could not isolate any 1,4 addition products, among the many products observed on TLC. Such a 1,4 adduct could provide a route to 1,6 dienes when fragmented.

With this in mind we sought once again to prepare this elusive cuprate. Undoubtedly the first example of a copper catalyzed Grignard conjugate addition actually occurred when such reactions were normally performed in copper reaction vessels. 55

The 1,^-Addition of CTC Cuprate

The reaction of CTC with 2-cyclohexenone had been carried out.11* The substrate enone was completely consumed when treated with 1.5 equivalents of CTC at -60°C followed by warming to R.T. before work-up.

The crude product was distilled to give a 52$ yield of epoxysilane. A considerable amount of non-volatile residue remained in the distillation flask although the reaction was quite clean by TLC. The major fraction analysed as 95$ pure by VPC. Mass spectral data was in agreement with the assigned structure and NMR and IR spectral data confirmed the presence of the double bond (3015, 16^5 cm-1). No products of 1,^-addition were isolated at that time.

Having established that CTC adds exclusively in a

1,2-mode to enones, an attempt to devise a reagent which would Michael-add to an a,J3-unsaturated ketone was made. 1-kxTMS f TMS X

0 H TMS H 57

If a good leaving group (X) is incorporated into the

reagent, then intramolecular nucleophilic displacement of

X should occur to give the silylated cyclopropane A.

The mode of addition of alkyl lithiums to a , (3-

unsaturated ketones can be changed to give 1,4-addition products exclusively by the preparation of an Intermediate

cuprate. If 0.5 equivalents of copper (I) iodide are added to a solution of an alkyl lithium, a species is formed, according to the stoichiometry shown, which is thought to be the dialkyl cuprate B . 1,5

2 RLi + Cul ^ R2CuLf + Lil B The new reagent thus formed Michael-adds because of a change in the mechanism. The cuprate is thought to react via an initial one electron transfer accompanied by oxida tion of copper (I) to copper (II), followed by coupling of the resulting radical anion C with an alkyl radical generated from the unstable copper (II) species.

C + R2C u 1L RCu1 53

Our colleagues’ attempts to prepare a cuprate from

CTC and to react it with 2-cyclohexenone were unsuccessful.

The color changes observed on adding copper (I) iodide to a solution of the anion at -70°C were consistent with those expected1*5 and the substrate was consumed, but none of the expected product A was observed. Silicon-containing decomposition products were observed as non-polar residues in all cases. Variation of the temperature and time for the formation of the cuprate, and addition of TMEDA and dimethyl sulphide, which coordinates with copper (I) rendering it more soluble in THP, all failed to give the required product.

The difficulties encountered in preparing this reagent led to an investigation of the alternative mixed cuprate reagent. Corey has successfully used mixed cuprates, e.g. D , to selectively transfer an alkyl group R in a

1,4-addition to an enone.1*6

e rRCu-c=c-c3iv Cu~ C=C~CjH7 D E Acetylene ligands are tightly bound to copper (I) ir. aprotic media at low temperature. Thus the copper acetvlile formed after the transfer of the R group is stable under the reaction conditions, only decomposing when the reaction is worked up. This reagent has two advantages. A value:It 59

R group is not wasted as when the dialkyl cuprate is used,

and no highly reactive anionic species remain in the

solution after the desired reaction has occurred.

The copper (I) n-propylacetylide D was found most

convenient1*6 because it gives pent-l-yne after work-up,

easily separable from the product because of its volatility.

Copper (I) n-propylacetylideD was prepared as a

yellow, amorphous solid according to literature procedure.1*7

A solution of the acetylide in hexamethylphosphorous

triamide, one of the few solvents for this insoluble material, was added to a solution of CTC in THF at -70°C

to give a bright yellow solution. When 2-cyclohexenone was added to the reagent none of the required product A

could be isolated, even after prolonged reaction.

Corey reported a very rapid transfer of simple alkyl groups to occur to this substrate under these conditions.1*6

It was concluded that CTC is not amenable to cuprate forma tion and that it would not be possible to achieve the desired 1,^-addition to an enone by this method. In our laboratory, the cuprate of Me3SiCH2MgCl was

successfully prepared with 10 mole percent of CuCN in

THF and found to undergo 1,4 addition in 88% distilled yield to 2-cyclohexenone. Other enones that were tried worked poorly and showed that the developed reagent was

a very bulky nucleophile and unreactive to many enones that other cuprates add easily. At temperatures above

15°C it was observed by IR data that 1,2 addition began to predominate.

MS 'TMS

'IMS 90

91 ^ 893; (0S0 Me) 2 61

With one good example to work with no new chemistry was found. The intended fragmentations were entirely unsuccessful on the OPNB derivative 89, the mesylate derivative 89a as well as the inital keto silane 87. elimination of the leaving group is believed to be the major reaction pathway under the lewis acid conditions employed. The mass spectrum of 90 was consistent with that shown.

Or perhaps the mass spectrometer conditions of 70 eV gives the M+ 168 as the parent ion. This is unlikely in view of the fact that 89 gave excellent NMR,

IR and mass spectral data. The parent M+ 335 was seen to have M+-15 which is characteristic of organosilanes.

The M/e included a fragment for M+ 168. Because the

TMS group loses a methyl group initially, then M+ 168 is not due to 90. No further characterization was done. During the following reactions the progress was monitored by VPC referenced to an authentic sample of

91. At 80°-100°C column temperature an injection of

89 in ether gave a product with the identical retention «s^s^v time as 91. When the column was cooled to 50°C no 91

T M S C H 2 MqCI/THF io-2 0 mol% CuCN ^ . -15° or RT TMS

The cuprate also did not add to 1-acetyl-cyclohexene

X > 63

There was no evidence that 91 had been formed during the treatment of 89 with cesium fluoride at

140°C, or boron trifluoride etherate complex (5 eq) in dichloromethane or 37% aqueous hydrofluoric acid. The starting material 89 was recovered in each case. The resistance of this OPNB derivative to such media is as remarkable as the next example which presents what was thought to be a better case for a fragmentation to d

1,6 diene since better orbital overlap is -involved.

The cuprate just described was found to undergo 1,4 addition in 33% isolated yield to A 1>*-2-octalone. The remainder of the reaction products looked to be that resulting from 1,2 addition. Once again with another example to work with it was decided to try the fragmen tation. The requisite starting material 93 was obtained as an oil that could not be crystallized. By NMR data

(only one broad singlet for 1H at 6 4.5) it was assumed that the equatorial OPNB derivative was the major product we desired. Unfortunately, the same series of cesium fluoride, fluoroboric acid, and Lewis acid induced media gave only recovered starting material.

There was no evidence of 94. As depicted below it was assumed that the 1,4 addition occurred to give the cis ring juncture as this would be the least encumbering approach for the cuprate species. 6H

1)NdBH4 l'MeOH TM: 2 ) ^ q P h C 0 C f

rr- > 70

OPhNC^f

O^OflhNO^f

)(

21 This unsuccessful fragmentation study was prompted

by an earlier study in our laboratory. It was found that

adducts 95-97 with one less carbon in the framework between oxygen and silicon underwent fragmentation. The hydroxy silanes were prepared via a Grignard reaction

on the epoxide or cuprate addition to the epoxide and then fragmented.1*0 65

H d b i > 65 f

95 TMSCH2MgCI f\S\S^ TMS THF > r ^ y - 0 ^ 75 '/» or M S

(tm sch 2 )2 Gul; 96

90% TMS

9 7 It was necessary to prepare the mesylates, tosylates and OPNB derivatives to study the fragmentations since a push-pull effect is believed to operate.

The OMs derivative 96a prepared below was studied although little data is known. Initially the alcohol was treated with methane sulfonyl fluoride in pyridine at -78°C to R.T. and no norcarane 98 was identified by

VPC comparisons. In concentrated sulfuric acid in ether at 0°C for 15 min vinylcyclopentane 99 was identified.

When the mesylate was treated with 50$ fluoroboric aci in aqueous dioxane at R.T. for 12 h the major product was identified as norcarane 98 by VPC comparison. The cyclopentyl system 97 under similar conditions gave many products not identified. No data was available on studies with 95.

However, it was reported that the OPNB derivative

96b, 97b of this system were unreactive toward fluoride mediated treatments."8 67

r^\>0H MeSO,F

96 n rr0H H^S

^ £M & so./. olh&rs CV-TMS HBF^" 96d 4

^ y s O P N B KF or '?i(r rx/7.

96b DMSOor ^ c h 3c n

100° 20 h. 68

GENERAL EXPERIMENTAL INFORMATION

All XH NMR spectra were obtained on either a Varian

A-60, Varian EM-36O, or Varian EM-390 spectrometer in

CCli» or CDCI3 as reference standard. The solvent ds-DMSO was used to aid solubility in some cases. Melting points were taken on a Thomas Hoover melting point apparatus and are uncorrected, as are the boiling points. Infra red (IR) spectra were run on a Perkin-Elmer 267 Grating infrared Spectrophotometer under the conditions specified.

Elemental analysis were done by M-H-W Laboratories in

Phoenix, Arizona. Mass spectral data were obtained on a Double Focusing Consolidated Electronic MS-9 mass spectrometer. All reported M/e were clean above the designated M+. Gas chromatographic analyses were done on a Perkin-Elmer 3920B instrument with 10% OV101 on chrom WHP (80-100) or 1035 SE-30 on Chrom DAW (100-120).

Preparative layer chromatography was done in petroleum ether (bpt. 60°-90°C) and ethyl acetate (^.’1) unless otherwise specified. Column chromatography was done with silica gel (Nominal Grade 923) in the same pair ratio.

TMEDA was distilled from CaH2 (bpt. 120°C). THF was distilled from LAH (bpt. 65°C). CTC (bpt. 97°C) and MCTC (bpt. 120°C) were distilled and stored over molecular

Sieves 4 % away from light. s-BuLi (1.1 M) and (1.3 M) were purchased from Foote Mineral Company, Exton, Pa.

The contentrations of (1.4 M) and (1.2 M) were purchased

from Aldrich Chemical Company, and were usually contam inated with up to 40# alkali other than s-BuLi. All transfer of liquids were done under positive pressure of argon with glass syringes and needles to ovendried

(160°C) vessels. EXPERIMENTAL TO PART I EXPERIMENTAL TO PART I

Preparation of 7-Trimethylsilyl-2-methylcyclohexylidene oxide 6

To a dry flask under argon at -78°C was added dry

THE (40 ml from LAH) and 3.7 ml, 26.7 mmol, 1.5 eq) of chloromethyltrimethylsilane followed by (4.03 ml, 1.5 eq) of TMEDA (from CaH2). Slowly s-BuLi (20.5 ml, 1.5 eq) (1.3 M) was added. The contents were stirred to

-40°C over 1.5 h, and recooled to -78°C. Distilled

2-methylcyclohexanone (2 g, 2.1 ml, 17.8 mmol, 1 eq) was added dropwise. The contents were stirred to R.T. over 1.5 h, and poured into saturated NH«C1 (100 ml) and estracted with diethyl ether (3X30 ml). Drying (MgSO*) and evaporation gave a yellow oil that was distilled at 0.5 mm to give 2.7 g, (77%), bpt. 65°C. NMR (CDC13)

6: 0.10 (s, 9H), 0.15 (s, 9H), 0.8-1.2 (br.m, 3H),

1.5 (br.s, 9H), 2.0 (s, 1H), 2.2 (s, 1H). IR (film):

2930s, 2860s, l470w, 1440m, I4l0w, 1370w, 1260m, 1250s,

1220w, 1120w, 920w, 890m, 860m, 840s, 800w, 760w, 750m, cm” 1. Mass spec. Calc, for CnHazSiO 198.144, Obs.

198.144. M/e 199 (1.3), 198 (6.3),,197 (1)., 185 (1.5),

184 (33.3), 170 (1.2), 169 (4.7), 167 (1.5), 157 (1.2),

156 (1.2), 155 (4.3), 141 (1.5), 139 (1.9), 137 (1.9),

70 71

125 (1.2), 112 (3.1), 108 (10.7), 100 (6.3), 95 (14.3),

93 (43), 85 (64.3), 75 (28.6), 73 (100).

Hydrolysis of 6. Preparation of cis and trans

2-Methylcyclohexanecarboxaldehyde J

The epoxysilane (900 mg, 4.5 mmol) was stirred with

88# HCOaH (with 0.4# AcOH) (3 ml) over 0.5 h at R.T. whereby a distinct orange color developed. After 1 h, the solution was evaporated carefully and washed into a separatory funnel with cold aqueous NaHC03 and extracted with diethyl ether. Drying (MgSOu and Na2SCU) gave after careful evaporation a labile liquid 44 mg, (77%) crude. IR showed some silicon impurities and was retreated with formic acid. Distillation at 0.5 mm gave a liquid (bpt. < 80° ) . NMR (CDC13) 5! 0.1 (s, 9H), residual SM. 0.9 (d, 3H) J=3Hz, 1.1 (s, 1H), 1.4-1.9

(br.s, 8H), 2.2-2.5 (br.s, 1H), 9.6 (d', 1H) J=3Hz, 9.8

(s, 1H). IR (film): (weak OH) 2920s, 2850m, 2700w,

1720s, l460w, 1450m, 1380 (doublet), 1250w, 1190m, 1050w,

840w, cm-1. Mass spec. Calc, for C0Hi«O 126.104, Obs.

126.105. M/e 143 (11.7), 142 (70.5), 127 (14.6), 126

(100), 125 (26), 124 (20.5), 112 (29), 111 (26), 110

(91), 109 (38), 108 (17.6), 98 (35-3), 97 (2 6), 96 (41),

86 (20.5), 85 (58.8), 72 (64.7), 60 (17.6), 59 (32.3),

58 (29). 72

Preparation of 8-Trimethylsil.yl Cyclohept.ylidene

Oxide 9

To a dry flask under argon at -78°C was added dry

THF (30 ml from LAH) and (2.76 ml, 19.95 mmol, 1.5 eq) of chloromethyltrimethylsilane and (3 ml, 1.5 eq) of

TMEDA (from CaH2). Slowly s-BuLi (15.4 ml, 1.5 eq) (1.3

M) was added and the contents of the flask were stirred to -40°C over 1.5 h, then recooled to -78°C. Distilled cycloheptanone (1.57 g, 13.3 mmol, 1 eq) was added dropwise and the contents were stirred to R.T. over 1.5 h, then poured into saturated NHUC1 (100 ml) and extracted with diethyl ether (3X30 ml). Drying (MgSOu) and evapor ation gave a liquid that was distilled at 0.9 mm to give

1.5 g (58$) of a clear liquid, (bpt. 60°C.) NMR (CDC13)

61 0.2 (s, 9H), 1.7 (br.s, 12H), 2.1 (s, 1H). IR (film):

2920s, 2850m, 1460m, 1440m, l400w, 1350w, 1330w, 1250s,

1020w, 920w, 860m, 840s, 760w, 750m, cm-1. Mass spec.

Calc, for C n H a2SiO 198.144, Obs. 198.144. M/e 199 (6.9),

198 (46.5), 184 (9.3), 183 (53,5), 182 (13-9), 169 (6.9),

157 (11.6), 156 (16.3), 155 (51), 130 (4.6), 129 (13.9),

108 (30.2), 107 (11.6), 75 (100), 73 (100). 73

Hydrolysis of g. Preparation of Cycloheptanecarboxaldehyde

The epoxysilane (0.5 g, 2.5 mmol) was stirred at

R.T. with 90# HC02H (3 ml) for 1 h. The contents were then quenched with cold saturated MaHC03 (20 ml) and extracted into diethyl ether (3X10 ml). Drying (MgSOu) and careful evaporation gave crude product (310 mg,

98#). The Infrared spectrum gave the maximums for two carbonyls. Presumably the bands (1700s, 1725s, cm-1) are due to the carboxylic acid. Microdistillation at

0.5 mm gave a liquid (280 mg, 88#). NMR (CDC13) <$ •

1.5 (br.s, 12H), 2.3 (br.s, 1H), 9.7 (d, 1H) J=2Hz.

IR (film): 3450m, 2920s, 2850s, 2700m, 1725s, 1700s,

1450m, 1440m, I4l0w, 1375w, 1340w, 1250m, 1120w, 1050w,

940w , 830m, 750w, cm- 1 . 74

Preparation of (Syn and Anti)-8-Trimethylsilyl-bicyclo-

[2.2.l]heptane-3-methylene Oxide 12

To a dry 50 ml three necked flask under argon at

-78°C was added dry THF (15 ml from LAH) and (1.5 ml,

10.9 mmol, 1.2 eq) of chloromethyltrimethylsilane followed by (1.6 ml, 1.2 eq) of dry TMEDA (from CaH2). Slowly

(7.7 ml, 1.2 eq) of sec-butyllithium (1.4 M) was added dropwise. The yellow color was consumed and the contents were stirred to -50°C over lh, then recooled to -78°C.

Norcamphor (1 g, 9 mmol, 1 eq) was added in dry THP

(3 ml from LAH) stirred to R.T. over 1 h. The contents were poured into saturated NHi»Cl (100 ml) and extracted with, diethyl ether (3X30 ml).Drying (MgSCU) and rotary evaporation gave a crude oil that was distilled at

<60°C at 0.05 mm to give 1.3 g (76%) pure epoxysilane.

NMR (CDCls) 6: 0.1 (s, 9H), 0.2 (s, 9H), 0.25 (s, 9H),

1.0-1.9 (m, 10H), 2.2 (s, 1H), 2.5 (s, 1H). IR (film):

2960s, 2900w, 2860m, l450w, 1410s, 1300w, 1250s, 1190w, ll40w, lllOw, 1060w, lOlOw, 940m, 890m, 870m, 845s,

790w, 750m, cm-1. Mass spec. Calc, for C n H 22SiO 196.128,

Obs. 196.128. M/e 197 (6.8), 196 (36.3), 182 (4.5), l8l

(22.7), 180 (2.2), 169 (6.8), 168 (27.2), 167 (100),

93 (100), 85 (100), 75 (100), 73 (100). 75 '

Hydrolysis of 12. Preparation of (Exo, Endo)-3-formyl-

(bicyclo [2. 2 . l] heptane ) 13

The epoxysilane 12 (500 mg, 2.5 mmol) was stirred at R.T. with 20% aqueous THf (10 ml) and

60% perchloric acid (20 drops) overnight. The con tents were poured into cold saturated NaHC03 (20 ml) and extracted with diethyl ether(3X20 ml). Drying

(MgSOu) and evaporation gave a liquid that was distilled at 0.5 mm (bpt. < 80°C) to give 230 mg (73*) of the aldehyde. NMR (CC14) 61 1.1-2.0 (br.m, 6H),

2.1-2. 4 (br.m, 2H), 2.6 (br.s, 2H), 3.6 (br. triplet,

1H) J=6Hz, 9.5 (s, 1H), 9.6 (s, < 1H), exo-endo ratio 3 .6 H . IR (film)*. 2940s, 2860s, 2800w, 2700w,

1715s, 1450m, 1310w, 1240w, ll60w, 1040w, 830m, 750w cm-1. Mass spec. Calc, for CBHi20 124.088, Obs.

124.088. M/e 124 (16), 96 (32), 95 (100), 81 (5.8),

80 (26.4), 79 (11.7), 68 (11.7), 67 (64.7), 66 (26.4). Hydrolysis of 12. Preparation of Bicyclo £3. 2 . l]]octan-

3-one, and Bicyclo ("3. 2. lloctan-2-one * ______

The epoxysilane (500 mg, 2.5 mmol) was stirred at R.T. with 90$ formic acid (2 ml) for 0.5 h. Evapora tion of excess formic acid and aqueous bicarbonate workup gave a low melting white solid that had no aldehyde signal in the NMR spectrum 6 0.9 (br.s, 2H),

1.2 (s, 6H), 2.0-2.3 (br.m, 4H). IR (film): 3450m,

2980s, 2920s, 2860m, 2850m, 1725s, 1450m, 1370w,

1300‘w, 1260W, 1250w, 1180m, 1070m, 1020m, 950w, 940m,

800w.Mass spec. Calc, for CQHi20 77

Preparation of (syn and anti)-2-Methyl-5(l-methylethenyl)-

7-trimethylsilyl-2-cyclo-hexene-l-methylene _____ ^____ Oxide 17 To a dry flask under argon at -78°C was added dry

THF (30 ml) and (2.7 ml, 19.95 mmol, 1.5 eq) of chloromethyltrimethylsilane followed by (3 ml, 1.5 eq) of

TMEDA. Slowly s-BuLi (18 ml, 1.5 eq) (1.1 M) was added dropwise. After stirring to -40°C over 1.5 h, the contents were recooled to -78°C and distilled Carvone

(2 g, 13.3 mmol, 1 eq) was added slowly dropwise. After stirring to R.T. over 1.5 h, the contents were poured into saturated NHi»Cl (100 ml) and extracted with ethyl ether (3X50 ml). Drying (MgSCU) and evaporation gave a crude liquid that was distilled at 0.4 mm to give (76%) two fractions (bpt. 75°C and 83°C). The data of both fractions was identical. NMR (CCIj*) 0.15 (s, 9H),

1.4 (br.s, 3H), 1.55 (s, 3H), 1.8-2.5 (br.m, 6H), 4.6

(s, 2H), 5.6 (br.s, 1H). IR (film): 3080w, 2940s,

2920s, 2840w, 1640m, 1440m, 1370m, 1300m, 1260m, 1250s,

1220w, 1170m, 1050w, lOOOw, 960m, 890m, 860s, 840s,

8l0w, 790w, 760m, 700m cm-1. Mass spec. Calc, for

Ci*Ha*SiO 236.159, Obs. 236.160. M/e 237 (5-5), 236

(19.4), 221 (11.1), 220 (8.3), 196 (9.4), 195 (100),

181 (5.5), 179 (8.3), 147 (33.3). 78

Hydrolysis of 1J. Preparation of (cis and trans)-l- formyl-2-methyl-5-(1-methylethenyl)-2-cyclohexene and the Major Product 1-Formyl- 2-nethyl-5-(1-ncthylcthonyl)-

1-cyclohexene 13

The epoxysilane (500 mg, 2.1 mmol) was stirred

at R.T. with 20% aqueous THP (6 ml) and 60% perchloric

acid (20 drops) for 2 h. The contents were poured into

saturated NaHC03 (15 ml) and extracted with diethyl

ether (3X10 ml). Drying (MgSOu) and evaporation gave

a liquid that was distilled at 0.4 mm to give 270 mg

(78%) of the aldehydes bpt. 85°C. NMR (CC1*):

1.7 (s, 6H), 1.9 (s, 2H), 2.1 (br.s, 5H), 4.6 (d, 2H)

(J=5Hz), 5.6 (br.s, < 1H), 10.0 (s, 1H). About 35# of

the deconjugated aldehyde was determined by NMR inte

gration. IR (film): 3060w, 2920s, 2860s, 2720w, 1710

(very weak), 1660s, 1635s, 1580w, 1440s, 1370s, 1300w,

1230s, ll40w, llOOw, 890s, 8l0s, 790s, 760s, cm . Mass

spec. Calc, for CaiHlsO 164.120, Obs. 164.120. M/e:

165 (8.6), 164 (54.5), 150 (13.6), 140 (27-3), 136 (27.3),

135 (18.2), 123 (34), 121 (54.5), 109 (35-7), 107 (63.6),

105 (35-7), 93 (100), 91 (72.7). 79 Preparation of eis and trans 2-r2-(6,6-Dimethylbicyclo-

(■}. 1. l]hept-2-ene)B-trimethylsilylJethylene Oxide 20

To a dry flask under argon at -J8°C was added

dry THF (25 ml) and (1.4 ml, 9*98 mmol, 1.5 eq) of

chloromethyltrimethylsilane followed by (1.5 ml,

1.5 eq) of TMEDA. Slowly s-BuLi (9 ml, 1.5 eq) (1.1 M)

was added dropwise. The contents were stirred to -40°C

over 1.5 h, then recooled to -78°C. Distilled 19

myrtenal (Aldrich) (1 g, 6.65 mmol, 1 eq) was added in

dry THF (2 ml). The contents were then stirred to 0°C

over 1.5 h, and poured into saturated NH4CI (80 ml)

and extracted with diethyl ether (3X30 ml). Drying

(MgSOii) and evaporation gave a liquid that was distilled

at 0.3 mm to give 1.5 g (95%) (bpt. 73°C). The product

was pure by TLC and no starting material evident.

NMR (CClu) 6 1 0.1 (d, 9H) J=2Hz, 0.9 (m, 3H), 1.0-1.2

(m, 2H), 1.3 (s, 3H), 2.0-2.4 (br.m, 5H), 3.1 (br.s,

1H), 5-4 (br.s, 1H). IR (film): 2920s, 2860s, 2820w,

l460w, 1450w, 1 3 8 0 w , 1360m, 1250s, 1200w, 1050w, 950w,

880m, 860m, 840s, 800w, 750w, 700w, cm-1. Mass. spec.

Calc, for CmHauSiO 236.159, Obs. 236.157. M/e 236

(8.5), 221 (25.5), 201 (11.9), 181 (13.6), 164 (54.5),

149 (100), 147 (100). 80

Hydrolysis of 20. Preparation of 2-(6,6-Dimethylbicyclo-

T3.1.llhept-2-ene Acetaldehyde 21

The epoxysilane (250 mg, 1.06 mmol) was stirred at

R.T. with 20$ aqueous THP (5 ml) and 60$ HClOi* (20 drops) for 2 h. The contents were then poured into saturated

NaHCOs (15 ml) and extracted with diethyl ether (3X10 ml).

Drying (MgSCU) and evaporation gave 180 mg (9^$) of a pure liquid. The material did not withstand distil lation at 0.^ mm. NMR (CCln) crude I 6 0.8 (s, 3H),

1.0 (s, 2H), 1.3 (s, 3H), 2.0-2.4 (br.m, 6H), 6.6

(br.m, 1H), 9-7 (t, 1H) J=6Hz . IR (film): 1700s, l6l0m, cm-1. No absorption for the a,3-unsaturated aldehyde

(1660 cm-1) was observed. Mass spec. Calc, for C n H i S0

164 81

Since the furanone 22 was apparently enolized by CTC and MCTC the unsubstituted furanone was reacted with CTC.

This treatment gave only very polar products on TLC with no incorporation of silicon in the NMR spectrum. This was not surprising since two positions of enolization are available. It was assumed that the aldol condensation proceeded as usual but no characterization of the products was made. 82

Preparation of the Epoxysilane 23

To a dry flask under argon at -78°C was added dry

THP (30 ml from LAH) and (2.76 ml, 19.95 mmol, 1.5 eq)

of chloromethyltrimethylsilane and (3 ml, 1.5 eq) of

TMEDA (from CaH2). Slowly s-BuLi (15*4 ml, 1.5 eq) (1.3

M) was added and the contents were stirred to -40°C over 1.5 h, then recooled to -78°C. The spiro furanone

22 (1.85 g, 13.33 mmol, 1 eq) was added dropwise. The contents were stirred to R.T. over 1.5 h, and poured into saturated NH4CI (100 ml) and extracted with diethyl ether

(3X30 ml). Drying (MgSOu) and evaporation gave a liquid.

PLC isolation gave 15$ yield of the isomeric epoxysilanes and about 15$ recovered furanone and the remainer 65-70$ was the aldol dimer that is seen as the major product of the furanone and MCTC 24. NMR (CDC13) 6 : 0.1 (d, 9H)

J-2Hz, 1.1-1.8 (br.m, 8H ) , 2.0 (s, 1H), 2.1-2.4 (m, 2H),

3.6-4,0 (m, 2H). IR (film): 2940s, 2860m, 1735 very weak, l460w, 1430m, l4l0m, 1340w, 1250s, 1180m, ll60w,

1060m, 1040s, 980w, 870m, 840s, 750m, 690w, cm-1.

Mass spec. Calc, for Ci2H 22Si0 2 226.139, Obs. 226.139.

M/e 226 (5-3), 225 (3.6), 211 (1.8), 210 (< 1), 199

(5.3), 198 (17.9), 197 (100), 183 (16), 181 (5-3), 170

(5.3), 169 (19.7), 155 (5.3), 137 (12.5), 136 (17.9),

124 (28.6). 83

Preparation of the Aldol Condensation Product 24 from

- 22 and.MCTC

To a dry three necked flask under argon was added

3 ml of THP (from LAH) and a-chloroethyltrimethylsilane

(0.33 ml, 0.002 mmol, 1.5 eq). sec-Butyllithium (1.4 M

in cyclohexane) (1.5 ml, 0.002 mmol, 1.5 eq) was added

slowly followed by TMEDA (0.31 ml, 0.002 mmol, 1.5 eq).

The contents were stirred at -78°C and then warmed to

-55°C over 1.5 h. After recooling to -78°C the furanone 22

was added (0.2 g, 0.0014 mmol, 1 eq). The reaction

mixture was poured into saturated NH4C1 (20 ml) and

extracted with ethyl acetate (2X20 ml). The organic

layer was washed with water (20 ml) and brine (10 ml).

Drying (MgSOu) and rotary evaporation gave 150 mg of

a viscous liquid. PLC on silica gel gave 100 mg pure

product, mpt. 71°-73°C. NMR (CDC13) 6: 1.75 (br.s,

17H), 2.3-2.8 (br.m, 3H), 3-6-3.8 (br.m, 2H), 4.2 (dd,

J=4Hz and 8 Hz, 2H). Anal, for Ci6H240i». Calc.

68.59/SC, 8.63$H. Found 68.325&C, 8.49#H. IR (film):

3480s, 2960s, 2880s, 1750s, l640w, 1460m, 1450m, 1430m,

1410m, 1370s, 1350w, 1330w, 1240m, 1190w, 1120m, 1060s,

1030m, 990m, 940w, 930m, 840w, 780w, 730s, cm"1. Mass

spec. Calc, for Ci6H240u 280.167, Obs. 280.168. M/e

280 (0.7), 262 (1.7), 234 (1.6), 205 (1.15), 197 (8.8),

196 (26), 178 (3.5), 169 (6), 150 (9-4), 140 (18.3),

139 (24.8), 125 (43.7), 116 (30.7), 114 (100). 84 Hydrolysis of 23. Attempts to Prepare the Aldehyde 25 ______. iW W

The epoxysilane 23 (110 mg) was stirred at R.T. with IWW 20# aqueous THF (2 ml) and 50# HC10* (20 drops). After

0.5 h no starting material remained-on TLC. The con tents were poured into saturated NaHC03 (10 ml) and extracted with diethyl ether (3X10 ml). Drying (MgSCU) and evaporation afforded an oil that darkened on isolation.

NMR (CC1 **) 6: 1.1 (br.s, 1H), 1.5 (s, 4H), 1.7 (s, 2H),

2.0-2.8 (br.m, 4H), 2.8-3.1 (br.m, 1H), 3-8-4.1 (br.m,

2H), 5-6 (weak m, < 1H), 7-1 (weak m, < 1H). No aldehyde proton was detected. IR (CCli,) 3400m, 2940m, 2920s,

2840m, 1720m, 1460m, 1440m, 1370m, 1250w, ll60w, 1030m,

950w, 910s, 840w, 730s cm-1. No accurate mass measurement for C9H i i*02 (M+ 154) was obtained.M/el obs. 273 (20),

272 (85.7), 256 (5.7), 255 (22.8), 254 (8.6), 243 (20),

242 (11.4), 241 (34.3), 227 (62.8), 226 (40), 211 (54.3),

205 (65.7), 198 (28.6), 197 (100). Starting material has (M+ 226). No structure was assigned for this product. 85

,50 Preparation of Estrone Methyl Ether 2i>

Estrone (19 mmol, 5.0 g), potassium carbonate (95

mmol, 13.2 g), and methyl iodide (0.48 mmol, 68 g, 30 ml)

were combined in acetone (125 ml), and the resulting

reaction mixture was heated at 50°C for 30 h. The reaction

was quenched in saturated aqueous ammonium chloride

layered with chloroform. The aqueous phase extracted,

and the combined organics were washed with aqueous

ammonium chloride and water. Removal of the dried

(MgSOn) solvent gave 5.04 g (97%) of pure estrone methyl

ether 26. IR (Nujol): 2920s, 1740s, 1510s cm-1. NMR

(CDCI3) 6 : 6.90 (mult, 3H, aromatics), 3.75 (s, 3H

-OMe), 3.2-1.3 (envelope, 15H, aliphatics), 0.88 (s,

3H, -CH3). Mass spec. M/e 285 (24*), 284 (100*). mpt. 169°-171°C. 86

Preparation of the Epoxysilane 27 from Estrone Methyl

Ether

To a dry (50 ml) three necked flask was added 8 ml of THF (from LAH) and chloromethyltrimethylsilane (freshly distilled bpt. 97°C atm. press.) (0.76 ml, 5.5 mmol,

3 eq). sec-butyllithium (1.4 M in cyclohexane) (4.2 ml,

5.88 mmol, 3.2 eq) was added slowly via syringe. TMEDA

(0.89 ml, 5.88 mmol, 3.2 eq) from CaHz was added and the contents of the flask were stirred at -78°C to -50°C over 1 h. The mixture was then recooled to -78°C and estrone methyl ether (0.5 g, 1.84 mmol, 1.0 eq) was added slowly as a solid from a shaker tube. After stirring at -78°C for 0.5 h, slow warming to R.T. was done over 1.5 h. At R.T. the contents were poured into saturated NHUC1 (50 ml) and extracted with ethyl acetate

(2X50 ml). The organic layer was washed with water

(30 ml) and brine (20 ml) and dried (MgS04). Rotary evaporation gave 0.6l g (92$). PLC isolation gave 240 mg (36%) of the epoxysilane. mpt. 65°-70°C. IR (Nujol)I

2920s, 2940s, 1690 very weak, l6l0s, 1575w, 1500s, 1450s,

1380m, 1310w, 1280m, 1250s, ll85w, 1160m, HOOw, 1035s,

970w, 900m, 860s, 840s. cm-1. NMR (CC1„) 6: 0.05

(s, 9H), 0.10 (s, 9H) 311, 1.8 (s, 3H), 1.3-2.2 (br.m,

14 H), 2.7 (br.s, 2H), 3.6 (s, 3H),6.5 (d, 2H) J=4Hz , 87

6.9 (d, 1H) J=4Hz. Mass spec. Calc. 370,233, Obs. 370.233.

Anal, for C23H 3„Si02 Calc. 74.6l$C, 9.24$H. Found: 74.30$C,

9.30$h. M/e 371 (10.7), 370 (34.1), 356 (22.7), 355 (81.8),

280 (8.2), 268 (22.7), 265 (22.7), 252 (7.9), 237 (7-9),

223 (7.9), 211 (7.9), 195 (7.9), l8l (54.5), 174 (36.4),

173 (84.1), 171 (36.4), 160 (36.4), 147 (100), 73 (100).

Hydrolysis of 2J. Preparation of 3-Methoxy-17-formyl-

1,3,5-estratriene 28

The epoxysilane (50 mg) was stirred at R.T. for

0.5 h with 90% formic acid (2 ml). The reaction mixture became lilac in color. Rotary evaporation and evapora tion to 0.05 mm gave a pasty solid. PLC (4:1 petrol/EtOAC) of the crude product gave 38 mg (98$) of a white solid, mpt. 110°-112°C. IR (Nujol): 2920s, 2700m, 1710s,

1610s, 1575m, 1400s, 1450s, 1375m, 1275m, 1250s, 1235s,

1150m, 1035m, 840m, cm"1. NMR (CDC13) 61 0.7 (s, 3H),

1.1-2.4 (br.m, 14H), 2.7 (br.s, 2H), 3.6 (s, 3H),

6.2-7.9 (m, 3H), 9.6 (s, 1H). Mass spec. Calc. 298.193,

Obs. 298.194, Anal, for C2OH 2602. Calc. 80.53$C,

8.72$H. Found 80.42$C, 8.96$H. M/e 298 (6 5 .6 ), 278

(8), 240 (11), 227 (28), 213 (10), 200 (12.5), 199 (45),

187 (25), 186 (100), 174 (28), 173 (67), 171 (25), 161

(18), 160 (53), 159 (33), 158 (15.6), 148 (15.6), 147

(53), 175 (15.6), 134 (15.6), 129 (15.6), 128 (18), 122

(12.5), 121 (25), 117 (12.5), 115 (28) metastable. 88

Preparation of the Epoxysilane 30

To a dry flask under argon at -7d°C was added dry

THP (20 ml) and (1.4 ml, 9.9 mmol, 3 eq) of chloromethyltrimethylsilane and (1.5 ml, 3 eq) of TMEDA.

Slowly s-BuLi (1.27 M) (7.8 ml, 3 eq) was added dropwise. The contents were stirred to -40°C over 1.5 h and then recooled to -78°C. The steroid 29 (1 g, 3-3 mmol, 3 eq) was added as a solid slowly from a shaker tube. The contents were stirred at -78°C for 1 h, then warmed to R.T. over 1.5 h and poured into a saturated

NHuCl (50 ml) and extracted with ethyl acetate (3X50 ml).

Drying (MgSOu) and evaporation gave a solid that was vacuum filtered with ether (30 ml) to give a pasty solid from the filtrate. PLC purification gave 520 mg,

(Hl%) of a solid, mpt. 58°-62°C. 89

Atter.pts to Isolate 2_1-Ncrprogesterone 31

The epcxysilane 30 was obtained in fairly pure form by column or preparative layer chromatography. The plan was to hydrolyze both the aiencl ether and epcxysilane in one step. However, aqueous formic acid gave very polar products on TLC and no crystalline aldehyde. 31 was, isolated. Then dilute mineral acid (HC1) in aqueous

THF was employed and only polar products were seen on

TLC. Ho aldehyde was evident in the IR spectrum or

HIIR spectrum. No mass measurement could be made on the crude product. Since the epoxysilane 30 was produced in lew yield and the hydrolysis procedures used were not providing ?1, then it seemed that the synthesis of

21-norprogest.erone was also not working as planned. In the case where 12 was produced from 71 no problems were encountered. Although formic acid hydrolysis did not provide 12 it was found that aqueous mineral acid in THF provided the clean conversion to 12 as desired. Through out this dissertation the hydrolysis of epoxysilanes prepared from CTC has posed a problem in some cases.

Overall, the epoxysilanes prepared from MCTC were far easier to hydrolyze providing the methyl ketone cleanly. 90

Preparation of the bis-Adducts 3J from (1,9) and

(1,10) [2.2f]para Cyclophane Dione 32

To a dry 25 ml three-necked flask under nitrogen

at -78°C was added dry THF (4 ml from LAH) and (0.12 ml,

0.85 mmol, 4 eq) of chloromethyltrimethylsilane followed

by (0.63 ml, 0.89 mmol, 4.2 eq) of sec-butyllithium

(1.4 M in cyclohexane). No TMEDA was used. After

stirring to -50°C over 0.5 h the solution was cooled to

-78°C and (50 mg, 0.21 mmol, 1.0 eq) of the cyclophane

diones 32 in dry THF (1 ml) was added. An immediate

green color developed. After total addition the contents

of the flask were orange. After stirring to R.T. (still

orange) over 1 h the contents were poured into saturated

NHuCl (10 ml) and extracted with ethyl acetate (2X10 ml).

Drying (MgSO*) and evaporation gave 77 mg (89%) of pure product containing no starting material. IR (Nujol):

3080w, 3060w, 2960s, 2920s, 2850s, 1670m, 1605m, 1570w,

1510w, 1460s, I4l0w, 1380s, 1250s, ll80w, HOOw, 1020w,

900w, 840s, 760w, 720w cm-1. NMR (CDC13) <5: 0.0 (s, 9H),

0.3 (s, 9H), 3 H , 0.8 (br.s, 4H), 1.2 (s, 6H), 2.5 (m,

1H), 3.0 (m, 1H), 6 .1-6.7 (br.m, 8H). Mass spec. Calc,

for C24H 32Si202 408.194 , Found 408.195. M/e 410 (7),

409 (13.7), 408 (31), 393 (13.7), 38l (1 0 ), 380 (31),

364 (13.8), 362 (27.6), 360 (13.8), 358 (21), 356 (21),

309 (17), 308 (65.5), 261 (17), 232 (45), 231 (58), 220

(35), 221 (21), 219 (52), 218 (62), 205 (70), 204 (72),

203 (86), 202 (100). 91

Attempts to Prepare the [[2 ,2jParacyclophane (1,9) and

(1,10) Dialdehydes 34

The bis adducts 33 (20 mg, 0.05 mmol) was stirred at ■ v w R.T. with 90$ formic acid for 1 h. Evaporation to 0.1 mm gave a pasty solid that was recrystallized from

acetone/hexanes to give 11 mg of a yellow solid mpt.

66°-68°C. The NMR and IR data did not give any indication

of an aldehyde functional group. When 33 (10 mg) was

stirred at R.T. with aqueous methanol and dilute sulphuric

acid only intractable tars were isolated. From the

formic acid hydrolysis it was assumed that a complex mixture of the possible diketones 35-38 was formed.

NMR (CDC13) weak singlet for cyclophane aromatic protons

visible at 6 7*6. The protons on the epoxides were

not evident. An unusually large singlet appeared at

6 1.2. This data does not correlate with the mixture

of diketones described. IR (Nujol) 3040w, 2920s, 2900s,

2820s, 1710s, 1670s, 1590m, 1500w, 1460s, l400w, 1360m,

1230w, 1150s, llOOw, lOlOw, 980w, 830m cm-1, mpt. 94°-

95°C. No mass measurement could be made on the desired

products. 92

Preparation of 9-0xo-10-oxamethylene 11-trimethylsilyl

Anthracene 40

To a dry 25 ml three-necked flask under argon at

-78°C was added (0.73 ml, 0.005 mmol, 2.2 eq) of chloromethyltrimethylsilane and dry THF (10 ml from LAH). slowly(4.07 ml, 0.005 mmol, 2.2 eq) of sec-butyllithium

(1.3 M) was added followed by (0.79 ml, 2.2 eq) of TMEDA

(from CaH2). After stirring to -50°C over 1 h the solu tion was cooled to -78°C and (0.5 g } 0.0024 mmol, 1.0 eq) of pure anthraquinone 39 was added through a solid shaker tube over 10 min. Then THF (5 ml) was added to aid solubility. After 0.5 h at -78°C the contents of the flask were dark brown after a n initial red color. After

1 h at -78°C the contents were dark green and lightened to R.T. The contents were poured into saturated NHuCl

(50 ml) and extracted with chloroform (2X20 ml). Drying

(MgS0i») and evaporation to 0.10 mm gave 740mg (89%) of a pure homogeneous product on TLC. mpt. 113•5°-115°C.

Recrystallization from CH2Cl2/hexanes gave 400 mg (57%) of a pale yellow solid, mpt. 111°-112°C. NMR (CDCI3)

6: 0.0 (s, 9H), 2.5 (s, 1H), 7.4-7.6 (m, 6H), 8.1-8.4

(m, 2H). IR (Nujol): 3040w, 3020w, 2940s, 2920s, 2840m,

1710s, 1600s, 1590s, 1470m, 1460m, 1440m, 1370w, 1330w,

1300s, 1275s, 1260w, 1250s, 1170w, 1150w, ll40w, 1060w,

1030w, 930m, 900w, 890m, 850s, 840s, 8l0s, 760s, 750s,

700w, cm-1. Mass spec. Calc, for CieHi8Si02 294.198, Obs. 294.198. Anal. Calc. 73.40*0, 6.12%H. Pound

73.55$C, 6.24*H. M/e 296 (7.5), 295 (25), 294 (100),

293 (27.5), 279 (12.5), 266 '5 ), 251 (3-7), 235 (3.7),

220 (3.7), 204 (10), 191 (15), 130 (100). 13C NMR 61

-1.4, 58.9, 71.0, 75.4, 77.9, 78.6, 121.9, 124.0,

127.3, 127.8, 128.3, 132.4, 133.2, 133.6, 133.9, 140.8,

143.7, 183.8 .

A repeated reaction with 4.4 eq of CTC did not give the bis-epoxysilane of the anthraquinone. 94

Hydrolysis of 40

The epoxysilane (100 mg, 0.34 mmol, 1 eq) was slurried in 10% aqueous methanol (3 ml) at 0°C. Boron trifluoride etherate (0.1 ml, 0.8 mmol, 2.5 eq) was added and the contents were stirred to R.T. over 1 h.

No change was noted by TLC and another equivalent of

BF3*Et2O was added. No change was evident. After workup with water and dichloromethane the NMR of the product was unchanged starting material. When this material was dissolved in 20% aqueous THF (3 ml) and stirred with 15 drops of 10 N HC1 a polar spot

(Rf =~0.4) on TLC was immediately evident. The red spot (H+ , A) was isolated by PLC and gave a yellow solid mpt. 2l8°-220°C. Insoluble in all available

NMR solvents. IR (Nujol)I 3060w, 2900s, 2840s, 1650s,'

1580s, 1450s, 1370w, 1300s, 1280s, 1250w, 1170w,

1040w, 930m, 830w, 8l0w, 780w, 730w cm-1. M/e obs. for M+ 210 (4.5), 209 (27), 208 (77), 207 (13.7),

195 (3), 194 (19.7), 193 (12), 182 (15.2), 181 (100),

166 (3), 165 (19.7), 153 (13.7), 152 (8 5 ), 151 (38),

150 (18.3). Preparation of (eis and trans)-ll-Trimethylsilyl-l- oxamethylene-1,2,3,455,6,7,8-octahydroanthracene 43 • f s r * This isolation procedure gave l-Formyl-1,2,3,4,5,6,7,8- octahydroanthracene 44

To a dry flask under argon at -7d°C was added dry

THF (30 ml) and (2.76 ml, 19*95 mmol, 1.5 eq) of chloromethyltrimethylsilane followed by TMEDA (3 ml, 1.5 eq).

The contents of the flask were stirred at -78°C and slowly s-BuLi (15*35 ml, 1.5 eq) (1*3 M) was added dropwise, then stirred to -40°C over 1.5 h. The con tents were recooled to -78°C and the distilled keto- octahydroanthracene 42 (2 g, 13*3 mmol, 1 eq) was adde in dry THF (3 ml). After stirring at -78°C for 1 h, the contents were slowly warmed to R.T. over 1.5 h, then poured into saturated NHUC1 (100 ml) and extracted with ethylacetate (3X50 ml). Drying (MgSOi*) and evaporation gave a liquid of 2.2 g. TLC showed an appreciable amount of polar materials (probably aldol products) plus starting material 42. PLC purification of 43 gave 44 by hydrolysis on the plate. NMR (CDC13)

6: 1.6-2.1 (br.m, 8H), 2.6-3*0 (br.m, 6h), 3*5-3*7

(br.m, 1H), 6.9 (s, 2H), 9*7 (d, 1H) J=3Hz). No mass spec. Calc, for Ci5H i 80 (MW 214) was obtained. IR (film)

1710s, cm-1. 96

Preparation of 1-(1-Pyrenyl)-2-trimethylsilyl Ethylene

Oxide 47

To a dry flask under argon at -78°C was added dry THF (10 ml) and (0.90 ml, 6.5 mmol, 3 eq) of chloromethyltrimethylsilane followed by (0.98 ml,

3 eq) of TMEDA. Slowly s-BuLi (1.3 M), (5.1 ml, 3 eq) was added dropwise. The contents were stirred to -40°C over 1 h, and then recooled to -78°C. i-Pyrene carboxaldehyde (0.5 g, 2.1 mmol, 1 eq) in dry THF (5 ml) was added dropwise slowly. The resulting green solution was stirred to R.T. over 1.5 h, and then poured into saturated NHUC1 (50 ml) and extracted with diethyl ether (3X20 ml). Drying (MgSOi,) and evaporation gave a liquid that was purified by PLC to give 0.53 g,

Ql%. Mass spec. Calc, for

CaiHaoSiO 316.128, Found 316.127. M/e 318 (12.3),-

317 (30.7), 316 (100), 301 (6.2), 290 (168). IR

(Nujol): 2920s, 2860s, 1460m, 1450m, 1370m, 1240m, ll80w cm-1. o erlu ehr nldd ihn h pout The product. the within included ether petroleum to ire a 6 . ws o acutd for. accounted not was 2.8 6 at sinriet protons field high The The low field protons (3*7, 2H) are split similar to those those to similar split are 2H) (3*7, protons field low The t o ., H ad per o e et o heteroatom. a to next be to appear and 2H1 1.8, (oat =H, H, . (, H, . (rm 2) 28 s 1H), (s, 2.8 2H), (br.m, 1.8 1H), (s, 1.2 2H), J=6Hz, bopin a eiet N as esrmn ws taken. was measurement mass No evident. was absorption spectrum. A structure for the product was net determined determined net was product the for structure A spectrum. rm h floig aa HR CC3) : . (br.t, 0.9 ) 6: (CDC13 HKR data. following the from IR the in evident was stretch C=0 no and NKR the in and rave only black tars. No aldehyde signal was evident evident was signal aldehyde No tars. black only rave and n qeu mnrl cd fri ai o eclrc acid perchloric or acid formic acid, mineral aqueous in isolate the aldehyde 48. The epcxysilane 47 was hydrolyzed hydrolyzed was 47 epcxysilane The 48. aldehyde the isolate O.) ^8 Acetaldehyde Atterr.pts 1-Pyrenyl Isolate to 7 s 2H) 7c83 b., H. R fl) n carbonyl no (film): IR 9H). (br.m, ), 7.c-8.3 H 2 (s, .7 As v:ill e en n at I w wr nt be to able not were we II, Part in seen be (6 . ad 12 my e due be may 1.2) 6 and 0.9 97 98

Preparation of 8-Trimethylsilyl-8-methyl Cycloheptylidene

Oxide 5J2

To a dry 50 ml three necked flask under argon at

-78°C was added THF (20 ml from LAH) and (2.06 ml,

13.37 mmol, 1.5 eq) of a-chloroethyltrimethylsilane.

Slowly (11.14 ml, 1.5 eq) of sec-butyllithium (1.2 M) was added, followed immediately by (2.02 ml, 1.5 eq) of dry TMEDA (from CaH2). After stirring to -55°C over

1 h, the contents of the flask were cooled to -78°C and (1.05 ml, 8.9 mmol, 1 eq) of distilled cycloheptanone was added. After stirring to R.T. over 1.5 h, the contents were poured into saturated NHUC1 (50 ml) and extracted with diethyl ether (2X30 ml). Drying (MgS0«) and evaporation gave a liquid that was distilled to give

1.4 g (75%) of a pure liquid, bpt. 93°C at 1.5 mm.

NMR (CDCI3) 6: 0.15 (s, 9H), 1.3 (s, 3H), 1.5-1.9

(br.s, 12H). IR (film): 2920s, 2850m, l460w, l440w,

1250s, 1040w, 920w, 860s, 840s, 750m, cm-1. Mass spec.

Calc, for Ci2H2uSiO 212.160, Obs. 212.160. M/e 213

(13.8), 212 (65-5), 198 (10.3), 197 (51-7), 183 (13-8),

170 (12.7), 169 (100), 157 (13.8), 155 (13.8), 143 (38),

130 (24), 113 (65.5), 112 (31). 99

Preparation of Acetyl Cycloheptane 51

The epcxysilane (600 mg, 2.8 mmol) was stirred at

R.T. with 8 ml of 90% formic acid for 0.5 h. Rotary

evaporation followed by careful evaporation to 0.15 mm

gave 400 mg of liquid (100$). Distillation at 40°C at 0.15 mm gave 385 mg (98$) of the pure methyl ketone.

NMR (CDCls) 6: 1.5 (br.s, 12H), 2.1 (s, 3H), 2.3-2.6

(br.s, 1H). IR (film): 2920s, 2850s, 1720s, 1460m,

1445m, 1370m, 1350m, 1170m, 950w, cm-1. Mass spec.

Calc, for C9H i60 140.120, Obs. 140.120. M/e 140 (28.6),

125 (14.3), 97 (100), 82 (83), 71 (75), 55 (100). 100

Preparation of cis and trans 7-Methyl-7-trimethylsilyl-

2-methyl Cyclohexylidene Oxide £2

To a dry flask under argon at -78°C was added dry

THF (40 ml from LAH) and (4.1 ml, 26.7 mmol, 1.5 eq) of a-chloroethyl trimethylsilane followed by (4.03 ml,

1.5 eq) of TMEDA. Slowly s-BuLi (20.5 ml, 1.5 eq)

(1.3 M) was added. The contents were stirred to -40°C over 1.5 h, and then recooled to -78°C. Distilled

2-methylcyclohexanone (2 g, 2.1 ml, 17.8 mmol, 1 eq) was added dropwise. The contents were stirred to R.T. over 1.5 h, and then poured into saturated NHi*Cl (100 ml) and extracted with diethyl ether (3X30 ml). Drying

(MgSOi*) and evaporation gave a liquid that was distilled at 0.1 mm to give 3.13 g (82%) of a liquid (bpt. 48°-51°C).

NMR (CDC13) SI 0.1 (s, 9H), 1.0 (s, 3H), 1.15 (s, 3H),

1.2 (s, 3H), 1.4-2.0 (br.m, 9H). IR (film): weat 3420,

2960s, 2920s, 2860m, 1470m, 1440m, l4l0w, 1370w, 1300w,

1260w, 1250s, 1050w, 980w, 910w, 870w, 860s, 840s,

790w, 760m. cm-1. Mass spec Calc, for Cx2H 2i*SiO 212.160,

Obs. 212.160. M/e 213 (2.4), 212 (12.7), 199 (4.8), 198

(16.6 ), 197 (58), 184 (2.9), 183 (6 .8 ), 171 (1), 170

(4), 169 (11.7), 157 (2.9), 156 (2.9), 155 (7.8), 144

(1.9), 143 (4.8), 142 (1.9), l4l (3 .8 ), 130 (2.9), 129

(1.9), 127 (2.9), 123 (2.9), 122 (8.7), 121 (2.9), 117 101

(5.8), 115 (2.9), 113 (2.9), 112 (l6.9), 197 (16.9), 101

(2.9), 97 (4.8), 95 (5.8), 93 (8.7), 91 (4.8), 85 (12.7),

81 (5.8), 79 (9.7), 75 (38.7), 73 (100).

Hydrolysis of 52. Preparation of trans l-Acetyl-2- methyl Cyclohexane 53

The epoxysilane (900 mg, 4.24 mmol) was stirred with 90$ HC02H (3 ml) over 0.5 h, at R.T. where by a distinct lilac color developed. After 1 h the solution was evaporated carefully and distilled to give 330 mg,

(56$) of a minty volatile liquid (bpt. 38°C at 0.4 mm).

Some product was lost to the conditions accounting for a lower yield. NMR (CDCls) 61 0.8 (d, 3H) J=6Hz, 1.3-

1.8 (br.m, 9H), 2.1 (s, 3H), 2.3-2.6 (br.m, 1H). IR

(film): 2920s, 2850m, 1710s, l460w, 1450m, 1380w,

1370w, 1350m, 1310w, 1270w, 1250w, 1210w, 1170m, ll60w,

1040w, 980w, 950w, 850w, cm-1. Mass spec. Calc, for

C9H160 140.120, Obs. 140.120. M/e l4l (3-4), 143 (23),

126 (1.3), 125 (12.3), 112 (1.3), 111 (8.2), 99 (2.7),

98 (5.4), 97 (47), 96 (4.1), 86 (6.8), 86 (20.5), 84

(10.9), 83 (20.5), 82 (8.2), 72 (8.2), 71 (24.6), 69

(8.2), 68 (2.7), 67 (10.9), 56 (5.4), 55 (100). 102 Preparation of (syn and anti)-8-Methyl-8-trimethylsilyl- bicyclo[2 .2.J heptane-3-methylene Oxide

To a dry flask under argon at -78°C was added dry

THF (30 ml from LAH) and (2.8 ml, 18.1 mmol, 2 eq) of a-chloroethyltrimethylsilane and (2.7 ml, 2 eq) of

TMEDA (from CaH2). Slowly s-BuLi (13-9 ml, 2 eq) (1.3 M) was added and the contents were stirred to -40°C over

1.5 h, then recooled to -78°C. Norcamphor (1 g, 9*07 mmol, 1 eq) was added in dry THF (3 ml) and the contents were stirred to R.T. over 1.5 h. The usual workup gave after isolation a crude liquid that was distilled at

1.5 mm to give 1.3 g, (6Q%) of a liquid, bpt. 75°C.

NMR (CDCI3) 61 0.1s and 0.15s (9H ca. Ill), 0.5 (d,

1H) J=3Hz, 0.8 (br.s, 1H), 1.1s and 1.2s (3H ca. Ill),

1.4-2.0 (br.m, 6h ), 2.3 (br.s, 2H). IR (film): 2950s,

2860m, 1450m, l400w, 1370w, 1310w, 1250s, 1150w, lOOOw,

860s, 840s, 750m, cm-1. Mass spec. Calc, for Ci2H22SiO

210.144, Obs. 210.145. M/e 2.2 (1.6), 211 (5-5), 210

(19), 196 (3.2), 195 (19), 194 (3.2), 183 (8), 182 (2 2 .6 ),

181 (100), 169 (7), 167 (9.5), 165 (8), 144 (5.5), 143

(5.5), 129 (10.3), 127 (6.3), 111 (11.8), 93 (15.8), 91

(15.8), 75 (64.5), 73 (100). 103

Hydrolysis of 54. Preparation of (Exo, Endo )-3-Acyl-

(bieyclo [2.2 .1) heptane) 5Jj

The epoxysilane (150 mg, 0.7 mmol) was stirred with 20$ aqueous THF (3 ml) and 20 drops of 3N HC1 were added. Stirring at R.T. for 1 h showed consumption of starting material by TLC. The contents were poured into water (20 ml) and extracted with dichloromethane (3X10 ml).

Drying (MgSOi,) and evaporation gave a liquid that was distilled at 1.5 mm to give 85 mg (87$) of a clean liquid (bpt < 80°C). NMR (CDC13) 61 0.8-1.6 (br.m,

10H), 2.1 (d, 3H) J=2Hz, 2.2-2.5 (br.m, 1H). IR (film):

3500w, 2950s, 2860m, 1710s, 1450m, 1360m, 1310m, 1250m,

1200w, 1170m, 1070w, 1020w, lOOOw, 870w, 840m, 760w cm-1. 104

Preparation of (syn and anti)-2-Methyl-5-(l-methylethenyl)-

7-methyl-7-trimethylsilyl-2-cyclohexene-l-methylene Oxide §6

To a dry (50 ml) three necked flask under Nz was added (20 ml) of THF (from LAH) and a-chloromethyltrimethylsilane (2.66 ml, 0.0173 mmol, 1.3 eq). The contents were cooled to -78°C and sec-butyllithium

(1.4 M in cyclohexane) (12.35 ml, 0.0173 mmol, 1.3 eq) was added slowly with rapid stirring. A pale yellow cloudy solution resulted. TMEDA (from CaH2) (2.6 ml,

0.0173 mmol, 1.3 eq) was added and the contents of the flask were stirred to -55°C over 1 h, then recooled to

-78°C. Distilled (d)-Carvone (bpt. 45°-46°C at 0.03 mm)

(2.1 ml, 0.0133 mmol, 1.0 eq) was added slowly and a pale blue solution resulted. Stirring was continued a

-78°C then slow warming to R.T. was done over 1.5 h.

At -20°C to 0°C the blue color faded and a yellow solu tion resulted. At R.T. the contents of the flask were poured into saturated NHuCl (50 ml) and extracted with diethyl ether (2X20 ml). The ether layers were washed with water (50 ml) and brine (20 ml). After drying

(MgSOi*) and rotary evaporation (2.88 g) crude product was obtained (86$). Distillation at (0.03 mm) gave two fractions bpt. 6l°-63°C (1.65 g, 50$) and bpt. 68°-73°C

(0.18 g, 16$). IR (film): 3080m, 2960s, 2920s, l680w,

1645s, 1450s, 1375s, 1 3 3 0 w , 1250s, ll60w, 1110m, 1050m, 980w, 900s, 850s, 800m, 760m, 700w, cm-1. NMR (CC1*)

6: 0.1 (d, 9H), 1.1-1.2 (d, 3H), 1.6 (d, 6h ), 2.0-2.2

(br.m, 5H), 4.6 (s, 2H), 5.6-6.0 (m, 1H).Mass spec, calc 250.175, Obs. 250.176. M/e 251 (5-7), 250 (26.6),

236 (3.8), 235 (23.3), 210 (20), 209 (100), 195 (9-5),

193 (7.6), 182 (7.6), l8l (5.7), 167 (5.7), 165 (7.6),

145 (26.6). 106

Hydrolysis of 56. Preparation of (cis and trans)-Acyl-

2-methyl-5-(l-methylethenyl)-2-cyclohexene and the Major

Product l-Acyl-2-methyl-5(l-methylethenyl)-l-cyclo-

hexene 57

The epoxysilane (200 mg) was stirred at R.T. with

1 ml (90$) formic acid for 2 h. Rotary evaporation then

evaporation to 0.05 mm removed excess formic acid and

hexamethyldisiloxane to give a yellow oil. No starting material was evident by TLC. Distillation at 0.05 mm

gave a pure enone product (bpt. < 60°C) in 70# yield.

The homogeneous product on TLC gave IR (film): 3080w,

2980m, 2920s, 1710s, 1680s, 1645m, 1450m, 1440m, 1375m,

1360m, 1250w, 1210m, 1170m, 1150m, HOOw, 970w, 930w,

840m, cm"1. NMR (CDC13) 6: 1.2 (d, 1H) J=3Hz, 1.5-1.8

(br.m, ca. 13H), 1.9 (s, 3H), 2.2 (s, 3H), 4.7 (s, 2H),

5.7 (weak br.s, 1H). Mass spec. Calc, for C i2H i 80

178.136, Pound 178.136. M/e 179 (2.2), 178 (31.7), 163

(29), 150 (20), 136 (13.6), 135 (86), 134 (20), 121

(15.8), 119 (20), 107 (6 1 ), 93 (100). 107

Preparation of 2-£2-(6,6-Dimethylbicyclo[3.1.l]hept-2-ene)-

P-methyl-O-trimethylsilylJ Ethylene Oxide 58

To a dry flask under argon at -78°C was added dry THF (20 ml) and (1.55 ml, 9.98 mmol, 1.5 eq) of a-chloroethyltrimethylsilane followed by (1.5 ml,

1.5 eq) of TMEDA. Slowly s-BuLi (1.1 M) (9 ml, 5 eq) was added dropwise.The contents of the flask were stirred to -40°C over 1.5 h, then recooled to -78°C.

Distilled myrental (Aldrich) (1 g, 6.65 mmol, 1 eq) was added in dry THF (2 ml) and the contents warmed to R.T. over 1.5 h, then poured into saturated NH«C1

(80 ml) and extracted with ether (3X30 ml). Drying

(MgSOi,) and evaporation gave a liquid that was distilled at 0.5 mm to give 1.5 g (89$) bpt. 95°C.

NMR (CC14) 61 0.25 (s), 0.35 (s) 111 (9H), 1.0 (s, 3H),

1.3 (s, 1H), 1.4 (s, 1H), 1.5 (s, 3H), 1.6 (s, 2H),

2.1-2.6 (br.m, 5H), 3.1 (br.s, 1H), 5.4 (br.s, 1H).

IR (film)! 2920s, 2900s, 2840m, l460w, 1440m, 1370w,

1360m, 1245s, ll60w, 1040w, 950w, 840s, 750m. Mass spec. Calc, for Ci5H26SiO 250.175, obs. 250.175.

M/e 251 (1.5), 250 (6.9), 236 (2.9), 235 (6.9), 222

(4.1), 221 (3.9), 209 (4.1), 207 (4.1), 181 (5.4), 178

(5.4), 149 (18.6 ), 148 (32.5), 147 (100). 108

Preparation of 2-(6,6-Dimethylbicyclo Q3 .,1. l]hept-2-ene )-

Acetone

The epoxysilane (500 mg, 2.0 mmol) was stirred at R.T.

with 88# formic acid (4 ml) for 3 h whereby a distinct red

color developed. Evaporation to 0.4 mm followed by

distillation at 0.4 mm gave 230 mg (66#) of a liquid

(bpt. < 90°C) as three spots on TLC. This mixture of the

conjugated enone E,Z isomers and deconjugated ketone gave

a complex NMR spectrum. NMR (CCli*) 0.65 (s, 3H), 0.75

(d, 3H) J=3Hz, 1.0 (s, 2H), 1.2 (d, 3H) J=3Hz, 1.35

(s, 1H), 1.6 (s, 1H), 2.0 (s, 3H), 2.1 (s, 3H) ratio

2.1, 2.2 (br.m, 2H), 5.4 (br.d, 1H), 5.8 (d, 1H) J=12Hz.

IR (film): 2920s, 2860s, 1710s, 1680m, 1600s, 1450m,

1430m, 1370m, 1360m, 1290w, 1160s, 1080w, 870w, 830w cm"1.

Mass spec. Calc, for C i2H i80 178.135, obs. 178.136.

M/e 179 (8.7), 178 (43.4), 163 (39), 161 (26), 153 (26),

147 (26), 135 (95.6), 134 (43.4), 119 (doublet, 100),

117 (doublet, 100). Impurity 198 (8.7) and 196 (21.7)

are not seen in the mass spectrum of the starting

material 58. 109

Preparation of (cis and trans)-ll-Methyl-ll-trimethylsilyl-

1-oxamethylene-1,2,3,4,5,6,7>8-octahydroanthracene 60

To a dry (50 ml) three necked flask under N 2 was

added (20 ml) of THF (from LAH) and a-chloroethyltri-

methylsilane (2.66 ml, 0.0173 mmol, 1.3 eq). The

contents were cooled to -78°C and sec-butyllithium

(1.4 M in cyclohexane) (12.35 ml, 0.0173 mmol, 1.3 eq)

was added slowly with rapid stirring. A pale yellow

solution resulted. TMEDA (from CaH2 ) (2.6 ml, 0.0173

mmol, 1.3 eq) was added and the contents of the flask

were stirred to -55°C over 1 h, then recooled to -78°C.

Distilled Keto-octahydroanthracene (bpt. 115°-120°C

at 0.025 mm) (2 g, 0.0133 mmol, 1 eq) was dissolved

in (5 ml) dry THF and added dropwise slowly. No color

change was observed and stirring was continued at -78°C

then slow warming to R.T. was done over 1.5 h. At R.T.

the contents were poured into saturated NHUC1 (50 ml)

and extracted with diethyl ether (2X20 ml). The ether

layers were washed with water (50 ml) and brine (20 ml).

Drying (MgSOi*) and rotary evaporation gave (2.2 g)

crude product. PLC of the crude product gave pure

isomeric epoxysilanes in 40-50% isolated yields. Approxi mately 30-40% of enolization by the reagent resulted in much recovered starting material. NMR (CDC13) 60.1

(s, 9H), 1.3 (d, 3H), J=3Hz, 1.7-2.0 (br.m, 6H), 5.6-5.8 110

(br.m, 1H), 6.8-7.0 (d, 1H) J=4Hz. IR (film): 2920s,

2860m, 2820w, l680w, 1500m, 1460m, 1440m, 1370m, I350w,

1340w, 1300w, 1250s, 1070w, 1040m, 990w, 930w, 860s,

840s, 760m cm-1. Mass spec. Calc, for Ci9H2aOSi 300.191,

Pound 300.191. M/e 301 (5-7), 300 (21.2), 286 (1.7),

285 (8), 271, 257, 24l (< 1), 229 (1.6), 228 (9), 201

(1.6), 200 (5), 176 (1 6 .3), 175 (100).

Hydrolysis of 60.- Preparation of l-Acetyl-1,2,3,4,5,6,7,8- octahydro Anthracene 61

The epoxysilane (135 mg, 0.45 mmol) was stirred at R.T. with 90$ formic acid (1 ml) for 2 h. The solution developed a green then aqua color which faded after 2 h.

Rotary evaporation then evaporation to 0.05 mm gave a yellow oil that was purified by PLC to give 70 mg of a liquid (65$). 90M Hz NMR (CDC13) 0.8-1.0 (br.s, 2H),

1.2 (br.s, 2H), 1.6-1.9 (b4.m, 6H), 2.0 (s, 3H), 2.4-

2.7 (br.s, 4H), 5.6 (m, 1H), 6.6 (s, 1H), 6.8 (s, 1H).

IR (Film): 3010w, 2980w, 2840w, 1700s, l680s, 1610m,

1600m, 1560w, 1500m, 1450m, 1430s, 1370m, 1350s, 1330w,

1270w, 1240m, 1150m, 920m, 840m, 790s, 760s, 720w, 700w,

Mass spec. Calc, for C i6H2oO 228.151, Obs 228.152.

M/e 229 (1.7), 228 (7.75), 186 (15.5), 185 (100), 153

(24), l4l (10). Ill

Preparation of cis and trans l-(2-Pyrenyl )-2-methyl-2- trimethylsilyl Ethylene Oxide 63

To a dry (50 ml) three necked flask under N2 was added 8 ml THP (from LAH) and a-chloroethyltrimethylsilane (2.7 ml, 0.017 mmol, 5 eq). The contents of the flask were cooled to -78°C and sec-butyllithium (1.4M in cyclohexane) was added slowly via syringe. TMEDA

(2.6 ml, 0.017 mmol, 5 eq) from CaH2 was added and the contents were stirred at -78°C for 0.5 h. After warming to "55°C over 1 h, the contents were recooled to -78°C and (1 g, 0.004 mmol, 1 eq) of pyrene carboxaldehyde in 2 ml dry THP was added slowly. The reaction mixture changed from red to orange. While warming to -20°C the color of the reaction mixture became orange. After warming to R.T. the contents were poured into saturated

NHuCl (50 ml) and extracted with ethyl acetate (2X30 ml).

The organic layer was washed with water (2X30 ml) and brine (20 ml). Drying (MgSOu) and rotary evaporation gave 1.3 g, crude product. PLC on silica gel (4:i petrol/EtOAc) gave pure product in 78% yield, mpt.

100°-102°C. NMR (CDC-la) 5: .0.1 (s, 9H), 0.7 (s, 3H),

1.4 (s, 1H), 4.3 (s, 1H), 1.6-7.9 (m, 9H). IR (Nujol):

3040w, 2920s, 2850s, l600w, 1450m, l4l0w, 1370m, 1310w,

1250s, 1180m, 1060w, 980w, 960w, 870m, 850s, 840s, 830s,

770w, 750m, 710s, cm” 1. Anal Calc. 80.035SC, 6.71*H.

Pound 79.88#C, 6.79$H. Mass spec. Calc, for C22H22SiO 112 330.144, Pound 330.145. M/e 332 (4.8), 331 (26.3), 330

(100), 315 (6), 314 (1.5), 301 (1.0), 258 (8), 257 (42),

256 (7.3), 242 (7.9), 240 (13), 266 (7-3), 215 (14),

214 (95.7), 213 (14), 206 (4.8).

19a. Hydrolysis of 63. Preparation of 2-Pyrenyl Acetone 64 AAAI

The epoxysilane (300 mg) was stirred at R.T. with

3 ml of 90$ formic acid over 1 h. The reaction mixture became brown on stirring. Immediate hydrolysis was noted by TLC in 10 min. Rotary evaporation and then evaporation to 0.05 mm gave a pasty liquid. PLC of the crude product gave the pure methyl ketone 200 mg (82$). mpt. 83°-84°C. mpt.lit 85°-86°C. NMR (CDC13) 6! 2.1

(s, 3H), 4.3 (s, 2H), 7.6-8.0 (m, 9H). IR (Nujol):

3030w, 2920s, 2840s', 1705s, 1600m, 1580w, 1460s, 1440m,

1420w, 137.0m, 1350m, 1270w, 1230s, 1170m, ll60w, ll40w,

1130w, 960w, 850s, 840s, 830s, 820w, 760w, 730s, 710s, cm-1. M/e 258 (23), 215 (18.4), 215 (100), 32 (4.6),

28 (44), 18 (65), 17 (39-5). 113

Preparation of cis and trans Epoxysilane 67

To a dry (50 ml) three necked flask under N2 was

added 8 ml THP (from LAH) and a-chloroethyltrimethyl-

silane (9.84 ml, 5*5 mmol, 3 eq). The contents were

cooled to -78°C and sec-butyllithium (1.4 M in cyclo-

hexane) (4.2 ml, 5.88 mmol, 3-2 eq) was added via syringe.

TMEDA (0.89 ml, 5-88 mmol, 3-2 eq) from CaH2 was added

and the mixture was stirred to -55°C over 1.5 h. After

recooling to -78°C estrone methyl ether (0.5 g, 1.84

mmol, 1.0 eq) was added as a solid via a shaker tube.

After stirring at -78°C slow warming to R.T. was done

over 2 h. The contents were poured into saturated NH4CI

(50 ml) and extracted with ethyl acetate (2X50 ml).

The organic layer was washed with water (2X30 ml) and brine (20 ml). Drying (MgSOu) and rotary evaporation

gave O .58 g, (85%) of crude product. PLC gave the pure epoxysilane in 40# yield as a homogeneous spot on

TLC. mpt. 128°C. IR (Nujol): 2920s, 2850m, 1610m,

1500m, 1460m, 1380w, 1280m, 1250s, 1240m, ll60w, lllOw,

1040w, 860w, 840m cm"1. NMR (CC14) <5: 0.1 (s, 9H),

0.2 (s, 9H), 3 :1 , 0.8 (s, 3H), 0.9 (s, 3H), 1.0 (s, 3H),

1.2 (s, 3H) 3:i, 1.3 (s, 3H), 1.5-2.3 (m, 14H), 2.7-

2.9 (m, 2H), 3-6 (s, 3H), 6.3-7.1 (m, 3H). Mass spec

Calc. 384.248, Obs. 284.249. Anal, for C2uH36Si02 Calc. 114

75• 00$C, 9.37/8H, Found 74.90$C, 9.44#H. M/e 385 (8.4),

384 (23), 370 (38.5), 369 (100), 355 (16.7), 312 (8.4),

295 (5), 268 (21.7), 195 (31.8), 173 (36.8 ), 163 (28.4),

149 (54), 148 (84.6), 147 (100). 115

Preparation of 3-Methoxy-17-acetyl-l,3,5-estratriene 6830C

The epoxysilane (50 mg) was stirred at R.T. with

90$ formic acid (3 ml) for 1 h. The reaction mixture developed a cherry red color. Rotary evaporation and then evaporation to 0.05 mm gave a pasty solid. PLC on silica gel (411 petrol/EtOAc) gave 38 mg (93$) of a white solid, mpt. 132°-134°C. mpt.llt* 134°-136°C.

IR (Nujol): 2920s, 2860s, 1700s, 1605s, 1500s, 1470s,

1370m, 1360m, 1310w, 1290w, 1250m, 1230s, 1190w, ll80w,

1170w, ll60w, 1040s, 910m, 870w, 830m, 790w cm"1.

90M Hz NMR (CDC13) 6 : 0.6 (s, 3H), 1.1-2.6 (br.m, 14H),

2.1 (s, 3H), 2.6-3.0 (br.m, 2H), 3.7 (s, 3H), 6.4 (s,

1H), 6.6 (d, 1H) J=3HZ, 7.0 (d, 1H) J=9Hz Mass spec. Calc 312.209, Obs. 312.209. Anal, for C2J.H2b02

Calc. 80.7 9 % C , 9.04$H, Found 80.83%C, 9-05$H. M/e

3.3 (2 1 ), 312 (100), 298 (8.4), 294 (10.5), 269 (4.2),

268 (9), 277 (5 .2 ), 227 (17-5), 199 (14.7), 187 (3-5),

186 (6.3), 174 (10.5), 173 (21), 171 (11.9), 160 (26),

147 (25). 116

The Birch Reduction of 67

The Birch reduction of the estrone epoxysilane was

carried out similar to that of Dryden.1*9 The epoxy

silane (180 mg, 0.47 mmol) was added in dry THP (5 ml

from LAH) to a solution of double-distilled ammonia

(20 ml) and t-butyl alcohol (from Na) (8 ml). Overhead

stirring is done and Dewar condensers contain the

ammonia vapors. The vessel is packed in vermiculite.

Lithium wire (oil-free) was added in pieces (50 mg,

8.6 mmol, 18 eq) and a blue solution resulted which

lasted briefly (10-15 min). After 30 min, the contents

were poured into saturated NH*C1 (50 ml) and extracted

with ethyl acetate (3X30 ml) and brine (20 ml). Drying

(MgSOn) and rotary evaporation then evaporation to

0.03 mm gave an oil (170 mg). No starting material was present by TLC and no UV activity was noted. NMR

(CDCls) <5: 0.1 (s), 0.15 (s, 9H), 1.0 (s, 3H), 1.1-2.2

(br.m, 18 H), 2.7 (br.s, 4h), 3.6 (s, 3H), 4.7 (m, 1H).

Mass spec. Calc, for C2UH 3eSi02 366.264, Pound 386.264.

M/e 388 (10.5), 386 (47.4), 385 (14.4), 374 (30.2),

370 (34.2), 368 (63), 356 (34), 341 (14.4), 296 (100),

294 (60), 284 (85.5), 269 (50), 255 (26.3), 242 (73-7),

227 (59), 207 (81.5), 186 (84). 117

The oily enol ether of the above reaction (130 mg,

0.33 mmol) was heated to 60°C with 20% aqueous methanol

(5 ml) and 6N HC1 (10 drops) for 1 h. The contents were then poured into cold saturated NaHC03 (10 ml) and extracted with ethyl acetate (3X10 ml). Drying

(MgSCU) and rotary evaporation, then evaporation to

0.03 mm gave a pasty solid (104 mg) which was purified by PLC to give one major product (38 mg, 37$) based on

MW 300 of a UV active product. Recrystallization from acetone/hexanes gave a white solid mpt. 86°-88°C.

This was not norprogesterone (mpt»lit l4l°-l42°C).

NMR (CDCIb)I no acyl methyl group at 2.1 6. The enone was evidently the singlet at 5.85 5- IR (Nujol): weak C-20 carbonyl at 1720 cm-1. The enone was evident by a strong absorption at 1680, 1620 cm-1. Mass spec,

for this material was identical to that described in the following experiments (M+ 284). M/e I 285 (21.7)»

284 (86.9), 270 (13), 269 (52), 256 (21.7), 255 (100). 118

Another reaction of the oily enol ether (140 mg)

was stirred at R.T. for 1 h with 90% HC02H whereby a

faint pink color developed. Rotary evaporation then

evaporation to 0.05 mm gave a pasty solid that was

identical to the previous solid on TLC. Recrystalli

zation from acetone/hexane gave 70°-72°C. Once again

this was not nor-progesterone. A bis-DNP could not be prepared. NMR (CDC13) <5: 0.8 (t, 3H), 1.0 (s, 3H),

1.3 (5, 2H), 1.6-2.8 (m, 15H), 5.8 (s, 1H), (no acyl methyl group present). IR (film)! 3010w, 2960s, 2920s,

2850s, 1725 (very weak), 1670s, 1660s, 1600s, 1450s, l4l0w, 1375m, 1360m, 1330m, 1325w, 1290w, 1260s, 1235w,

1210s, ll60w, 1120w, 990w, 970m, 940w, 900m, 850w,

760w cm-1. Anal, for C20H 2002 MW 300. Calc! 79.95$C,

9.39/&H*, Pound: 81.12&C, 9.69#H. Mass. spec. M/e

285 (1.7), 284 (6.8), 270 (<1%), 269 (3), 256 (20.7),

255 (100), 254 (6.8), 237, 228 (< 136), 213 (1.5), 199

(2), 185 (< 1%), 173 (3.5), l6l (8.2), 159 (7). Since it was believed that the epoxysilane was being cleaved by the excess lithium metal, another Birch reduction was carried out. This procedure involved adding small amounts (10 mg) of lithium wire over 20 minute intervals and following the consumption of the starting material by TLC. After 2 eq of lithium had been added over 1 h it was observed that a non polar product appeared with an appreciable amount of 6j remaining. At that time, the normal workup gave a product that was immediately hydrolyzed and isolated by TLC. This product was identical to that obtained previously and showed that the Birch conditions cleaved or reduced the epoxysilane faster than the aromatic A ring was reduced. Thus, no conditions could be found to reduce the A ring without destroying the epoxysilane. 120

Ketalization of 3-Methoxy-17-acetyl-l,3,5-Estratriene 68. 3 oc

Initially it was attempted to prepare this ketal

in situ from the methyl substituted epoxysilane. It is

known that hydrogen substituted epoxysilanes can be both

homo- and hetero-atom ketalized easily.

The epoxysilane 67 (120 mg, 0.31 mmol) was dissolved

in dry benzene (5 ml from Na) and TsOH (50 mg) with

(0.2 ml, 8 eq) of ethylene glycol under argon and heated

to 90°C for 3 days. TLC aliquots of the mixture showed

two non-polar spots (possibly epimeric ketals at C-17)

appearing slowly. Workup and isolation showed that the

epoxysilane gave 3-methoxy-17-acetyl-l,3,5-estratriene

(60%) 68 and a small amount (30%) of the desired ketal

70. Thus Ts0H»2H20 in benzene with ethylene glycol

hydrolyzes the epoxysilane 67 to the methyl ketone 68 but we believe the silanol residues may depress the rate

of ketal formation. 121

Preparation of 70 from 68

The methyl ketone (440 mg, 1.4 mmol) was dissolved in dry benzene (50 ml) under argon. Ethylene glycol

(0.2 ml, 2.5 eq) and Ts0H»H20 (100 mg) was added and the contents were heated to 100°-110°C for 3 days with azeotropic removal of water. The reaction mixture was then cooled and poured into saturated NaHC03 (100 ml) and extracted with ethyl acetate (4X30 ml). Drying

(MgSOu) and rotary evaporation then evaporation to 0.05 mm gave a pasty solid 300 mg (60$) of the ketal recrys tallized from acetone/hexanes mpt. NMR (CDC13)

6: 0.7 (s, < 3H), '0.8 (s, 3H), ca 3:i, 1.0 (br.m, 2H),

1.2-1.5 (br.s, 9H), 1.8 (br.s, 3H), 2.1 (weak s, < 3H),

2.2 (br.m, 2H), 2.4 (s, 1H), 2.8 (br.m, 3H), 3.7 (s, 3H),

3.9 (br.s, 4h ), 6.5 (s, 1H), 6.6 (s, 1H), 7-0 (s),

7.1 (s, 1H). IR (film): 3040w, 2920s, 2860m, 1600m,

1570w, 1490m, 1450m, 1360s, 1310w, 1290w, 1240m, 1190s,

1180s, 1040m, 1020m, 920s, 8l0s, 770m, 740m cm-1. Mass. spec. Calc, for C23H3203 356.235, Found 356.236. M/e!

356 (2), 312 (100), 269 (2.2), 24l (2.7), 227 (19.0). 122 Preparation of ZX from 3-Methoxyandrost-3,5-diene-17- one 23

To a dry 50 ml three necked flask under argon at

-78°C was added 15 ml of dry THF (from LAH) and (1.53 ml, 9.9 mmol, 3 eq) of a-chloroethyltrimethylsilane.

(1.5 ml, 3 eq) of TMEDA (from CaH2) was added. Slowly

(7.79 ml, 9.9 mmol, 3 eq) of sec-butyllithium (1.27 M) was added with rapid stirring. The contents of the flask were stirred to -50°C over 1 h. The pale yellow solution was recooled to -78°C and (1 g, 3.3 mmol,

1 eq) of the steroid 29 added as a solid. Extra THF

(10 ml) was added. The normal workup and PLC isolation gave 71 in 38# yield, mpt. NMR (CDC13) 61 *WS» 0.0 (s, 9H, ca. i:i, 0.1 (s, 9H), 0.8 (s, 3H), 0.9 (s,

3H), 1.0-2.0 (br.m, 15H), 3-5 (s, 3H), 5.0-5.3 (br.m,

2H). Mass spec. Calc, for CasHi,oSi02 400.280, Found

400.280. M/e 401 (12.7), 400 (3.5), 385 (7.0), 258

(4.9), 328 (5.3), 323 (5 -6 ), 301 (2 2.6 ), 300 (100),

299 (6.8), 285 (15). IR 123

Preparation of Progesterone 72 from the Epoxysilane 71

The epoxysilane 71 (500 mg, 1.25 mmol) was stirred * at R.T. with diethyl ether (5 ml) and 6N HC1 (3 ml) for

1 h. TLC monitoring of the reaction against an authentic sample of progesterone showed the loss of 71 and clean appearance of a spot that had identical Rf and developing properties as the authentic sample. Formic acid hydrolysis of 71 gave the same product but less cleanly. The contents of the above clean hydrolysis were poured into saturated NaHC03 (10 ml) and extracted with ethyl acetate

(3X30 ml). Drying (MgSCU) and evaporation gave a pasty solid that was purified by PLC to give 320 mg (81$) of a solid mpt. 122°-123°C (from EtOH) mpt-lit of the 0 form 121°C mpt. of the a form 128°C. The NMR was identical to that of the authentic sample. IR (CCli*)

3^50w, 2940s, 2860m, 1735s, 1675s, l620w, 1540w, l460w,

1370m, 1250m, 1230w, HOOw, 1050w, lOlOw, 850w, 790w,

760w cm"1 . 12*4

Reaction of CTC with the Weiland-Mischler Ketone £3

(a Ketone and Enone Functionality).

To a dry flask under argon at -78°C was added

dry THF (15 ml) and (1.4 ml, 9-9 mmol, 3 eq) of chloro

methyltrimethylsilane followed by (1.5 ml, 3 eq) of

TMEDA. Slowly s-BuLi (7.6 ml, 3 eq) (1.3 M) was added

dropwise. The contents were stirred to -40°C over

15 h and then recooled to -78°C. 9-Methyl-A5(10)-

octalone-1,6-dione (0.58 g, 3-3 mmol, 1 eq) was added

slowly as a solid and an aqua solution developed.

The contents were stirred at -78°C for 1.5 h whereby

the solution became brown. After stirring to R.T.

over 1 h the solution was poured into saturated NHuCl

(50 ml) and extracted with ethyl acetate (3X30 ml).

Drying (MgSOu) and evaporation gave 1 g of a crude

liquid. By TLC it was evident that many products had been formed. PLC isolation of the non polar products

showed that they were bis-adducts (MW 350) to both

carbonyls. IR and NMR data indicated a massive amount

of silicon. One product isolated in 2035 yield by PLC

(Rf=~0.5) was shown to be the adduct of CTC to the C-l (enone

in IR) carbonyl (MW 264). Mass spec. Calc, for Ci5H2u Si02

264.155, Found 264.155. M/e 265 (2), 264 (11.4), 251

(6.2), 250 (21.7), 249 (100), 236 (7.2), 235 (6.2),

222 (7.2), 221 (21.7), 219 (6.2), 208 (6.2), 193 (7.2) 125

175 (6.2), 163, 161 (10.3), 1^7 (17.5). IR (film):

Both carbonyls were evident in the spectrums and massive silicon absorptions observed. A reaction of 1.5 eq of CTC gave a small amount of products with a lot of starting material remaining. The bis-adducts were stirred with 90% HC02N whereby a purple color developed.

Evaporation to 0.1 mm after 1 h gave a brown oil that solidified with hexanes to give a solid mpt. 80°-85°C.

NMR (CDC13) weak signals at 6! 9*5 and 9*9. No vinyl proton was seen in either spectrum. Mass. spec. Calc, on the bis-aldehyde .(MW 206) and the mono-aldehyde accurate (MW 192) was too small for a mass measure ment to be made. M/e 222, 220, 208, 206 (1), 192 (10).

The mono adduct gave the same NMR and mass spectral data. The solid from hexanes gave mpt. 1^0o-l45oC.

Treatment of 7jJ or 75 with 1.2-2.0 eq of CTC gave less than 20# of the desired epoxysilanes. 126

Preparation of "-Dimethylphenylsilylcyclohexylidene

Oxide 7J

To a dr;/ ml three necked flask under argon was added dry TH7 '15 ml from LAH) and (2.20 ml, 12.2 mmol,

1.2 eq) of ohloromethyldir.ethylphenylsilane 76 (P.C.R.) and cooled to -7-'l. Slow addition via syringe of (8.73 ml,

12.22 mmol, 1.2 eq) of sec-butyllithium (1.4 M) gave a straw colored solution. The yellow color was consumed rapidly. After 15 min, to -60°C or -55°C the solution was cooled to -78°C and quenched with (1.05 ml, 10.18 mmol, 1.0 eq/ of distilled cyclohexanone was added. The contents were stirred to 0°C and then to R.T. and poured into saturated NJUC1 (2X30 ml) and diethyl ether (2X30 ml). The ethereal layers were washed with water (2X20 ml) and brine (20 ml). Drying (MgSO*) and rotary evaporation gave 2.8 g, (93%) crude. Fractional distil lation gave 1.1 g, (50$) of pure product (bpt. 90°-92°C at 0.05 mm). NMR (CC1«) 6: 0.25 (s, 6H), 1.0 (s, < 1H) impurity), 1.35 (s, 10H), 1.9 (s, 1H), 7.05-7.40 (m, 5H)

IR (heat)'. 3060w, 3040w, 2920s, 2050m, 1590w, 1450s, lH30s, 1410s, 1250s, 1115s, lOOOw, 970m, 890s, 840s,

820s, cm*"1. Mass spec. Calc, for CisHa2SiO 246.144,

Obs. 246.144. M/e 246 (21.7), 231 (94.3), 221 (4.3),

168 (5.2), 153 (13), 147 (6.9), 137 (52), 135 (100),

121 (14.7), 97 (4.7), 75 (26). 127

Hydrolysis of 77

The dimethylphenylsilyl epoxide 64 (50 mg) was stirred at R.T. with 1 ml of 90% formic acid. Rotary evaporation after 1 h gave a non-homogeneous product.

The crude liquid was taken up in saturated NHnCl (20 ml) and diethyl ether (20 ml). Drying (MgSOu) and evaporation gave a liquid that did correlate to an authentic sample of the desired product by TLC. The product gave IR

(film): 3510w, 3420w, 3060w, 3040w, 3020w, 3000m, 2960s,

2940s, 2850s, 2700w, 1900w, 1950w, l880w, l825w, 1725s,

1700w, l680w, l6l0w, 1590m, l490w, 1450s, 1430s, l4l0w,

1250s, 1170m, 1120s, 1050s, lOOOw, 840s, 800s, cm-1.

This silyl epoxide does not easily hydrolize by this method. Removal of the silanol residue is not as versa tile as the trimethyl silyl version due to its decreased volatility. The epoxide (100 mg) was stirred and heated with (0.1 M) DNP Reagent in Me0H/H2S0u to give a solid mpt. 164°-165°C. NMR (CDCl3-d6 DMSO) 81 1.8 (s, 10H),

2.4-2.5 (br.m, 1H), 2.7-2.9 (br.m, 1H), 7.3 (s, 1H),

8.5 (d, 1H) J=lHz, 8.70 (d, 1H). M/e 276 (14), 275 (100),

274 (26), 260 (1.8), 259 (2.2), 247 (14), 246 (64), 245

(4), 235 (2.4), 234 (10.4), 233 (16), 229 (5.6), 228

(5.2), 221 (15.2), 220 (7.2), 201 (5-6), 200 (8.8), 188

(3.2), 187 (4.8), 183 (6.4), 182 (10.4), 155 (9.2), 154 (15.6 ), 152 (4.8), 151 (150), 139 (3.2), 138 (7 .2 ',

(4.8), 126 (3.2), 125 (2.4). The DNP derivative did not correlate with the known product, (mpt.llt'

Attempts to lithiate 77 with n-BuLi, s-BuLi and t-ruLI in THF, hexanes at -78°C to -20°C, were unsuccessful as judged by the inability to isolate any of the product I 78, where R = methyl. 129

Attempted Metallation of a-Chloro Vinyltrimethylsilane 79

To a dry flask under nitrogen at -78°C was added dry THF (5 ml from LAH) and 0.3 g, 2.2 mmol, 1 eq) of a-chloro-a-trimethylsilyl ethylene. Slowly (1.88 ml,

2.45 mmol, 1.1 eq) of methyllithium (1.3 M in Et20) was added. No color was observed. (0.37 ml, 2.45 mmol, 1.1 eq) of TMEDA (from CaH2) was added and the contents were stirred to -50°C over 0.5 h. (0.2 ml,

2 mmol, 0.9 eq) of distilled cyclohexane was added.

After stirring to R.T. the contents were poured into saturated NHuCl (20 ml) and extracted with diethyl ether (2X20 ml). Drying (MgSOi*) and evaporation gave

200 mg of a clear liquid that proved to be the carbinol

80 IR (3400s) resulting from addition of methyl lithium to cyclohexanone. No silicon incorporation was noted (8l). A second reaction done in diethyl ether instead of THF and stirring of the methyllithium with the reagent silane for 1 h gave the carbinol also. M/e 114. Methyl lithium does not add to vinyl silane under these conditions.

Another reaction at -20°C with identical workup gave a semisolid product. Addition of methyllithium at that temperature evolved a gas. The product was identified as the silylynol 82. IR (film): 3400m, 2150s, 1700w,

960m cm-1. No silicon incorporation. NMR (CDCla) 6! 0.4 (s, 9H), 1.60-1.80 (br.m, 10H), 2.3 (br.s, 1H).

Mass spec. Calc. 196.128, Obs. 196.129. M/e 197

(1.2), 196 (6.8), 195 (1.2), 182 (18.7 ), 171 (100),

180 (18.7), 167 (12.5), 165 (18.7), 163 (37.5), 154

(18.7), 153 (100), 140 (10.2), 135 (8.5), 122 (25-5),

99 (32.4), 98 (30.6). A :reaction stirred from -78°C to -20°C gave the same product. 131

Attempts to Prepare 7-Trimethylsilylnorcarane 83 from

CTC and Cyclohexene

To a dry 50 ml three necked flask under argon at

-78°C was added dry THF (30 ml from LAH) and (2.25 ml,

16.3 mmol, 1.0 eq) of chloromethyltrlmethylsilane.

Slowly (13.0 ml, 19*5 mmol, 1.2 eq) of sec-butyllithium

(1.5 M in cyclohexane) was added. (2.95 ml, 19.5 mmol,

1.0 eq) of dry TMEDA (from CaH2) was added and the solution stirred to -50°C over 0.5 h, (8.25 ml, 81.5 mmol, 5 eq) of distilled cyclohexene was added and the contents were poured into saturated NHUC1 (100 m), and extracted with diethyl ether (2X50 ml). The organic layers were washed with water (2X20 ml) and brine (20 ml). Careful rotary evaporation gave 1.1 g of an amber liquid. The liquid was distilled at 25 mm to give a liquid bpt. ca. 60°C. No fraction had a product identical to an authentic sample by TLC. Mass spec, indicated the product to be the bis-trimethylsilylethylene 84. The

Rf of the product was identical to the silylcyclopropane but did not develop on TLC in the same manner at all.

The same reaction was repeated with dry olefin free hexane (30 ml distilled from sodium metal that was decanted from concentrated sulfuric acid). After over night reflux and the usual workup it gave the same product as above. NMR (CDC13) 6: massive Si at 0.0, 6.6 (s, 1H), 6.8 (s, 1H). IR (film): 1610m, 1250s, 850s, cm-1.

M/e 173 (2), 172 (.215), 159 (3), 158 (5), 157 (25),

147 (1), 141 (1), 131, 129 (less than 1J6), 115 (1.8)

100 (1.8), 99 (11.5), 98 (1.8), 84 (2.5), 83 (7.5),

(10), 73 (100). 133

Preparation of 3-Trimethylsilylmethylcyclohexanone.87

To a dry 100 ml three-necked flask under argon and

fitted with a reflux condenser was added (5*5 ml, 40 mmol, 1 eq) and THP (40 ml from LAH) along with (972 mg,

40 mmol, 1 eq) of Mg metal turnings. A very small chip

of iodine was added and the contents were stirred and brought to reflux for 1 h (60°C) or until all the Mg was gone. The solution was cooled to R.T. and then to

-15°C and (358 mg, 4 mmol, 0.1 eq) of CuCN was added as the grey salt. Stirring was done to solubilize the salt over 0.5 h. With rapid stirring (3.1 ml, 32 mmol, 0.8 eq) of distilled cyclohexenone was added slowly. As each drop hits the surface a yellow color is rapidly consumed by the grey solution. TLC after 0.5 h at -15°C showed no enone residue. At R.T. a pale yellow solution results. Poured the -15°C solution into saturated

NH«C1 (100 ml) and the blue solution was extracted with ethyl acetate (2X50 ml). The organic layer was washed with water (2X30 ml) and brine (20 ml). Drying (MgSCU) and rotary evaporatin gave 6 g (100%) of a liquid.

Distillation gave 5*2 g (88%) of a pure liquid bpt.

53°C at 0.15 mm. NMR (CDC13) 6 ’. 0.0 (s, 9H), 0.7 (d,

2H) J=5Hz, 1.2-2.6 (br.m, 9H). IR (film): 2980w,

2890s, 2860s, 1710s, l460w, 1450m, 1420s, 1350w, 1320w, 1280w, 1260m, 1250s, 1230m, 1170m, ll40w, 1070w, 1060m,

1020w, 900w, 860s, 840s, 780m, 760m, 690s, cm-1. Mass spec. Calc, for Ci0H20SiO 184.128, Obs. 184.129. M/e

184 (24), 169 (53), 156 (metastable), l4l (12), 130

(70), 127 (17), 115 (100), 75 (100), 73 (100). 135 Preparation of (cis and trans) 3-Trimethylsilylmethyl

Cyclohexan-l-ol 88

. To a dry flask under argon at 0°C was added methanol

(30 ml from Na) and (4 g, 21.7 mmol, 1 eq) of pure 6 keto silane (y silylketone). (940 mg, 23.9 mmol, 1.1 eq) of sodium borohydride was added slowly. Stirring was done at 0°C for 1-2 h whereby H2 evolution stops.

The contents were poured into saturated NHttCl (100 ml) and extracted with diethyl ether (3X30 ml). After washing with water (2X20 ml) and brine (2X10 ml), drying

(Na2S0u) overnight followed by filtration and evaporation gave 4 g (98$) of a colorless liquid as two homogeneous spots on TLC. NMR (CDC13) S ’. 0.1 (s, 9H), 0.6 (dd, 2H)

J=Hz, 1.0-2.0 (m, 9H), 2.4 (s, 1H), 3.5 (s, 1H). IR

(film): 3350s, 2900s, 2890m, 2850m, 1450m, l420w, 1360w,

1300w, 1260w, 1250s, 1220w, llOOw, 1050s, 1020w, 980w,

930w, 900w, 880m, 860s, 840s, 760w cm-1. Mass spec.

Calc, for Ci0H22SiO 171.121, Found 171.121. M/e 184

(0.98), 171 (1^.4), 169 (6), 168 (15-6), 153 (19.2),

(143 (10.8), 125 (16.7), 115 (6.4), 114 (6.9), 96 (28.8),

97 (26.4), 91 (30), 81 (100), 75 (100), 73 (100). 136

Preparation of the Mesylate of §8

The mixture of two alcohols (380 mg, 2.0 mmol) was dissolved in pyridine (4 ml) at 0°C under argon. Mesyl chloride (0.19 ml, 2.45 mmol, 1.2 eq) was added slowly.

After 15 min a precipitate formed and the contents were stirred at R.T. for 2 h longer. The brown solution was then poured into saturated NaHC03 (20 ml) and extracted with diethyl ether (3X20 ml). Drying (MgSOi,) and evapora tion gave a pure yellow oil (390 mg, 74$). NMR (CDC13)

6.* 0.1 (s, 9H), 0.6 (dd, 2H) J=4Hz , 1.0-2.0 (m, 9H),

2.1 (s, 1H), 3.0 (s, 3H). IR (film): 2940s, 2890w,

2860m, l460w, l450w, l4l0w, 1350s, 1330s, 1260w, 1250s,

1180s, 970m, 940s, 930s, 900m, 870w, 860s, 840s, 750w cm-1. Mass spec. Calc, for C n H 2uSi303 no M+ for 264

M+-95 for CioHaoSi. M/e 169 (16.2 ), 168 (6.1), 155

(6.1), 154 (6.1), 153 (64.8), 125 (6.1), 114 (7.7), 96

(7.7), 94 (4.8), 81 (16.2), 73 (100). 137

Preparation of the para-Nitro Benzoate Derivative of 88

The mixture of two alcohols (1 g, 5.4 mmol, 1 eq) in pyridine (6 ml) was stirred at R.T. with (1.1 g,

5.9 mmol, 1.1 eq) of p-nitrobenzoylchloride (from recrystallized benzene and S0C12) for 10 h. TLC showed one alcohol would not be entirely consumed. Poured into water (150 ml) and extracted with dichloromethane

(3X30 ml). The organic layers were washed with 5% NaOH

(50 ml) and 10$ HC1 (50 ml) then water (50 ml) and brine

(30 ml). Drying (MgSCU) and evaporation gave 2 g ofa liquid.Three products were separated by PLC. Fast Rf

860 mg (50$) had mpt. 6l°-63°C. NMR (CDC13) 5*. 0.1

(s, 9H), 0.6 (d, J=5Hz, 2H), 1.2-1.3 (br.m, 9H), 5.1

(br.s, 1H), 8.3 (s, 1H). IR (Nujol): 3100w, 3040w,

2980s, 2920s, 2880s, 1715s, l675w, l605w, 1590w, 1520w, l460w, l450w, m o w , 1350s, 1320s, 1300m, 1290s, 1260m,

1250m, 1200w, 1120s, 1100s, 1020m, 980m, 930w, 870s,

840s, 800m, 760w, 720s, cm"1. Mass spec. Calc, for

Cx7H25SiOi*N (M+-15) 320.132, Obs. 320.133. M/e 335

(9 -2 ), 320 (0 .6 ), 305 (3 .2 ), 240 (13-6 ), 224 (100), 208

)2 .1 ), 195 (4.3), 178 (7.9), 168 (13-6 ), 153 (27.3),

151 (31.8). 138

Attempts to Fragment 89 and 89a

The benzoate or mesylate (200 mg) was treated with cesium fluoride (140 mg, 1 eq) in dimethylsulfoxide

(3 ml) under argon at 150°-l60°C. These conditions provided data consistent with 90 and not 91. NMR

(CDC13 ) silicon functionality was retained while the loss of the CH3 group at 6 (3.0s, 3H) was observed.

Mass spec, for Ci0H 20Si MW 168. M/e 169 (2.3), 168

(13.5), 153 (11.8 ), 125 (1 1 .8 ), 114 (20.3), 94 (11.8 ),

74 (13.5), 73 (100). The y olefin or 6 olefin are possible. None of the desired 1,6-diene 91 was observed. 139

Preparation of cis-9-Trimethylsilyl Methyl-2-octalone 92

To a dry flask under argon was added Mg turnings

(324 mg, 13-3 mmol, 2 eq) and chloromethyltrimethylsilane

(I.85 ml, 2 eq) and dry THP (20 ml from LAH). One chip of I2 was added and the contents were refluxed at 60°C for 1 h until all magnesium had been consumed. The flask was cooled to R.T. and then to -15°C. A 1^9^-2~ octalone (1 g, 6.6 mmol, 1 eq) was added dropwise while the reaction vessel was kept at -15°C for 2 h, then warmed to R.T. The solution had developed from a yellow to a pea green. TLC showed a small amount of starting material was remaining so the contents were poured into saturated NH4C1 (50 ml) and extracted with ethyl acetate

(3X30 ml). Drying (MgSOu) and evaporation gave an oil

(1.2 g, 76%). PLC isolation gave a clear yellow oil

(520 mg, 33$) with the remainder being the 1,2 addition product. The desired product of 1,4 addition gave NMR

(CDCls) 6 : 0.1 (s, 9H), 0.7 (s, 2H), 1.1-2.0 (br.m,

11H), 2.1-2.8 (br.m, 4h). IR (film): 2920s, 2860m,

1720s, l460w, l-50m, l420w, 1320w, 1260w, 1250s, 1230w,

1210w, 920w, 910w, 860s, 840s, 760w, 730m cm"1. Mass spec Calc, for Ci*H26SiO 238.174, Found 238.176. M/e

238 (15.15), 224 (15.15), 223 (8 1 .8 ), 210 (48.5), 209

(15-15), 195 (27.3 ), 183 (2 1 .2 ), 182 (21 .2 ), l8l (2 1 .2 ),

165 (12 .12), 148 (69.7), 143 (42.4), 133 (39.4 ), 130

(45.4), 115 (100), 108 (42.4), 91 (33-3), 75 (100), 73

(100). 140

Preparation of 9-Trimethylsilylmethyl-2-octanol

The 6 keto silyl octalone (230 mg, 0.96 mmol) was dissolved in methanol (6 ml) at 0°C and sodium borohydride (44 mg, 1.15 mmol, 1.2 eq) was added. After 2 h, the contents were poured into saturated NHUC1 (20 ml) and extracted with dichloromethane (3X10 ml). Drying

(Na2S0i*) and evaporation gave an oil that was two (50/50) homogeneous spots on TLC (210 mg, 91$). NMR (CDC13) 5:

0.1 (s, 9H), 0.8-1.1 (dd, 2H), J=3Hz, 1.0-2.4 (br.m,

16H), 3.9 (br.s, 1H). IR (film): 3350 (br.s), 2920s,

2860s, 1450m, l4l0w, 1370w, 1260m, 1250s, ll60w, 1055m,

1035m, 960w , 865m, 855m, 840s, 780w, 760w cm” 1. Mass spec. Calc, for Cn,H2BSiO 240.191, Found 240.190.

M/e 241 (3), 240 (16.6), 225 (10.7), 223 (10.7), 207

(8.7), 183 (2), l8l (1.4), 179 (2)2, 168 (4.8), 150

(16.6), 149 (11.2), 135 (57), 122 (23.8), 121 (47.6),

109 (23.8), 108 (69), 107 (31), 95 (31), 94 (43), 93

(71), 75 (100), 73 (100). l4l

Preparation of the para-Nitrobenzoate Derivative froin the Alcohol of £2

The 6 hydroxy silyl octalone (210 mg, 0.87 mmol) was stirred at R.T. in pyridine (3 ml) and recrystallized p-nitrobenzoylchloride (194 mg, 1.05 mmol, 1.2 eq) was added. Stirred to R.T. over 2 h and rotary evaporation was used to remove the pyridine. PLC isolation gave

260 mg (77%) of a viscous yellow liquid that would not crystallize. NMR (CDC13) *: 0.2 (d, 9H), J=1H, 1.411 ratio 1.0 (s, < 1H), 1.1 (s, < 1H), 1.3 (s, 1H),

1.4-2.2 (br.m, l4H), 5-3 (br.s, 1H), 8.3 (s, 4H). IR

(film): 3400w, 3100w, 2920s, 2860m, 1725s, 1605m, 1530s,

1470w , 1450w , 1410w , 1350s, 1320m, 1280s, 1250m, 1170w,

1120s, 1100s, 1015w, 980w , 950w , 910w , 860w, 850m, 840s,

780m, 760w, 720s cm-1. Mass spec. Calc, for CaiHsiNO^Si

389.202, Found 389.201. M/e 391 (less 1%), 390 (1.2),

389 (3.6), 376 (less 256), 375 (1.7), 374 (6.1), 361 (less

1%), 360 (1), 359 (3), 242 (1), 24l (2.2), 240 (8.3),

225 (16.6 ), 224 (80 ), 223 (9), 222 (33.3), 208 (3), 207

(18), 195 (18), 194 (6), 179 (9), 178 (9), 168 (7.5),

135 (46.6), 134 (100). 1U2

Attempts to Fragment 93 to 9^ •wvw < v w

All conditions failed. Treatment with BF3Et20 in dichloromethane at 40°C} cesium fluoride in dimethyl- sulfcxide at 120°C, 31% fluoroboric acid and benzyl dimethylammonium fluoride at 60°C gave no evidence for the loss of the OPNB derivative as seen with 89 or the formation of 9j4 . The starting material 93 was recovered in each case. PART II

INTRODUCTION

A number of silicon containing ylides1 and crgano- metallics2 have been described in which electron delocali zation into silicon 3d orbitals may be important. Carey and Ccurt3 have prepared the phosphonate, Me3SiCHLiFO'OEt)2 and the sulfide, K e 3SiCHLiS?h. Their reactivity with carbonyl compounds has been studied. Petersen* made the discovery that olefination from carbonyls was possible with Me3SiCHLiSMe and Me3SiCHLiSPn via the loss of Me3SiOLi to give the hetero-substituted olefins.

The formation of vinyl phosphonates rather than vinyl silanes (loss of (0Et)2F0zLi) is consistent with the current thinking regarding reactions of phosphonate carbanions.5 The S-hydroxy phosphonates resulting, from addition of phosphonate carbanions to carbonyls lose diethyl phosphonate only when the carbon bearing phosphorous carries an additional electron withdrawing substituent, while base catalysed elimination of the B-hydroxysilanes occurs re.adily.2 0 »14 The reagent, Me3SiCHLiOMe would be useful for extension of carbonyl groups to homologous enol ethers. Attempts to prepare the required organolithium derivative by proton abstraction from Me3SiCK20Me

1*»3 m n

had not been previously successful.6 When n-BuLi was

used the silane was cleaved to yield n-butyltrimethyl-

silane (attack at silicon) with cleavage of the -CH20Me

group as the only identifiable product after quenching with H 20. This cleavage is analogous to that which

occurs in Schollkpof's procedure for preparation of

MeOCH2Li.7

MeOCH2SnR3 + n-BuLi — » MeOCH2Li + BuSnR3

\ s - BuU Me3 SiCH2OMe ------► Me3SiCHOMe

Li i n - BuLi 1-BuLi

LiCH2Me2SiCH20Me Me3SiBun + LiCHgOMe etc.

To minimize nucleophilic attack at silicon, t-BuLi was used as the base and was observed to abstract a proton from the Si-methyl group to give the kinetic product MeOCH2Si(CH3)2CH2Li rather than MeOCHLiSiMe3.

In another experiment methyl iodide was added to the organometallic and MeOCH2SiMe2CH2CH3 was isolated in 50% yield. 145

It was of particular interest to us that the above reported attempts to- deprotonate MeOCH2SiMe3 were largely unsuccessful. If such a reactive species could be prepared and added to a carbonyl compound it would be the same adduct that arises from nucleophilic attack of methanol. a,3-Epoxysilanes are opened with Me0H/BP3Et20 or MeOH/CF3C02H to give exclusively the a-methoxy adduct.8*9

HucHik

The overall effect would be an additional source of the species LiCH20Me recently explored by W.C. Still.9

Alkyl substituted a-alkoxy organolithium reagents have also been described.10 Activation by the benzoate carbonyl has been shown to facilitate direct metalative preparation of primary and secondary organolithiums adjacent to oxygen, sulfur and nitrogen.11 While such organometallics can be considered dipole-stabilized carbanions, further information on that point, as well as development of the synthetic potential of other species, is being studied. 146

The heteroatom-substituted organometallic compounds exhibit variations in inherent stabilities that are a function of the particular heteroatom substituent. The limiting extreme of instability is exemplified by the carbanions obtained from metalations of certain activated oxy-ethers. These compounds have a propensity for undergoing rearrangement to the isomeric alkoxides.

Olefins and aldehydes have also been obtained from decompositions of various RCHMOR1. Phenylcarbene has been implicated as an intermediate in the decomposition of CeHsCHLiOCgHs.2u Because of the tendency of RCHMOR' to rearrange and because of the considerable interest in the mechanisms of the rearrangement, little attention has been directed toward their isolation and utilization.

A detailed discussion of rearrangements of RCHMOR', which was first recognized by Wittig in 1942, is clearly beyond this review. Interested readers are referred to the comprehensive reviews of The Wittig Rearrangement

Reaction by Dalrymple et al.,22 Schollkopf,25 and Cram,23 and references cited therein. 147

RCH20R’ + R"M — > RCHMOR’ + R"H I RCH(R')OM (R=C6H5> CH2=CH, plus other electron stabilizing groups)

The differences in stabilities that exist between

those compounds mentioned above (RCHMOR') and the simple

oxy-substituted methylmetal.compounds, ROCH2M, are only matters of degree. The oxy-substituted Grignards, for example, can be prepared and utilized in both tetrahydrofuran and methylal, but tend to decompose quite rapidly above 0°C.

It is tempting, though not strictly proper, to rationalize the position of the equilibrium in terms of the stabilities of the organolithium reagents involved.

Although it might be expected that intramolecular chelation by, for example, an ethoxyethyl protecting group might stabilize organolithium reagents of this type relative to n-BuLi, Still finds that the a-methoxy stannane also gives at least 98% exchange of the stannyl group with n-BuLi at -70°C.

Other important factors affecting a-alkoxy organo lithium stabilization may include intermolecular (aggregate) chelation and inductive effects of the type that Still suggested for sp3 hybridized oxygen substituted carbanions.

I PLEASE NOTE:

This page not included with original material. Filmed as received.

University Microfilms International 149

The replacement of ketonic oxygen by hydrogen and

formyl, R 2CO -*■ R 2CHCHO, is an Important and commonly

used tactic in synthesis for which several different

standard reagents are available.26 The most common reagent, methoxymethylenetriphenylphosphorane A, often

fails to provide the intermediate methyl enol ethers in acceptable yield, and although the analogous phosphine oxide reagent B can be more effective, its use is also limited.27 Anions derived from diphenyl alkyl phos- phines have been described,20 but their utility remained largely unexplored. Phosphine C was prepared in nearly quantitative yield by treatment of lithiodiphenyl- phosphide with chloromethyl methyl ether.29

CHjOCHPOPh2 CH3O C H =PPh3 Li A B

Phosphine C can be metallated efficiently to D by treatment with s-BuLi in THF. This reagent was successful

in preparing the homologous methyl enol ether of an aphidicolin intermediate (see p.. ). In further

attempts to establish its utility, the reaction with a

series of hindered ketones was examined29 and the homologate aldehydes after hydrolysis were obtained in their respec

tive yields from the requisite ketone. Ph2PCl * PhjPLi ■ * Ph2PCH2OCH3 ■» Ph2PCHLiOCH C D

CHO CHO 91% 96% 73% 79% ca. 10%

Corey points out that the reagent D should be considered the reagent of choice for methoxymethylenation of hindered ketones above all other procedures.26 As will be seen shortly, lithio methoxymethyltrimethylsilane gave comparable results with only 1.5 eq strength and no special precautions for toxic materials. Corey uses

5 eq or more of D and the reaction must be conducted in a fume hood since carcinogenic materials make up the preparation of reagent D. 151

A similar homologation procedure was recently reported14 whereby a 3-hydroxy-a-trimethylsilyl ether was obtained. In the synthesis of (+)-aphidicolin one hindered carbonyl had to be converted to its homo logate aldehyde. None of the current methods were adequate.15 Our method, however, was not referenced or used. In this method outlined below a long sequence is employed. Corey claims this new method (5 steps) should be useful as a general solution to the problem of attaching carbon to very hindered ketonic groups. Such hindered carbonyl groups as menthone and others were conveniently homologated by our methods in good yields.

4eq. TMSCN

o.95eq e Reagent B27 was TMSU -35 ineffective whereas HMFft reagent D29 was effective for this homologation. 80% RESULTS

Experimentally, methoxymethyltrimethylsilane could

be lithiated with s-BuLi in THF at -25°C to -20°C to give

the species Me3SiCHLiOMe I_ which added well to aldehydes

and ketones.13

When these adducts were treated with cesium fluoride

in dimethylsulfoxide at 80°C for 0.5 h with the objective

of eliminating H0SiMe3 under essentially neutral conditions,31*

surprisingly the -SiMe3 group was cleanly removed to give the desilylated product. The adamantanone example is.

represented here.

This product 4 was prepared by another route to prove the regiochemical array of the ethers. The epoxide of adamantanone 5 was conveniently obtained and opened under basic conditions to give the same (3-hydroxy ether *1 as from the desilylation reaction of 3.

4— “OH 5

152 153

Having chemical proof that this product is the result of protodesilylation other adducts were subjected to the identical conditions. The adducts resulting from the addition of this new reagent with aldehydes could be oxidized with pyridinium chlorochromate to provide a-methoxy ketones after chromatography. Because the original adduct could be hydrolyzed to the homologated aldehyde, methoxymethyltrimethylsilane illustrates another new reagent for the process of reductive nucleophilic acylation. Additionally, this procedure provides a novel entry into regiospecific 3-hydroxy ethers and a-methoxy carbonyl moieties. The following is represent ative of those compounds prepared with this reagent.

'Me SiMe. liMe.

SiMe SiMe Me 154

Si Me. IQOVTs®! * 24 /3W 5 v

SiMe. Me 30 Me The utility of these intermediates lies in their

conversion to the corresponding aldehydes, enol ethers,

and 3-hydroxy-a-methoxy methanes. For those adducts

derived from carbonyl compounds these transformations

proceed generally in the yields listed.

'HO 6 0 - 9 0 %

! OHOMe 70-90 7. Me S i M e ^ K CsF

i V ~ \ Me Perhaps the most promising of these adducts are

those derived from the aldehydes since they also provide

a secondary alcohol that can be conveniently oxidized

and chromatographed to give a-methoxy ketones in good yields. 156

Wissner16 has reported a method by which acid

chlorides can be converted to their a-methoxy carbonyl

counterparts. The mild conditions complement this

method.

The mechanism proposed involves that shown below.

^ R ' > = < z JL, ,'HF 2 2 TMSCI ,HF { bSiRj . c o 2r

l £ R- ,CG,R Z=0H,0M<2,0Fh,SMe

n o ■ T W S O < p z H v S R- 'IMS """Cl A R blMS

HMDS H OCH 2C02H ^^OTMS'JHMDSLj-

pyr T l v l S 0 0

H .OTMS -/ ■\ TM OTMS

4 157

Each adduct prepared with lithio-methoxymethyl- triraethylsllane was transformed into the available functional groups just described.

Adamantanone 2 was treated with 1.1 eq of 1 and provided the adduct 3 in 89# yield. The enol ether 7 was obtained in 87# yield from 3 with KH in THP. Formic acid hydrolysis of 3 gave the unstable aldehyde 6 in

85# yield. The 3-hydroxy-a-methoxy methane k was obtained in 89# yield from either 3 (or 5) as just described. This preparation of the regiospecific ether 4 prompted another study of protodesilylation to be discussed later.

The adduct 9 was prepared in 73# isolated yield from cyclohexanone 8. The hydroxy ether 12 was difficult to isolate due to water solubility. The mass spectra data showed (M+-H20) but the M+ was too small to mass measure. The desired enol ether 11 was not isolated either due to its volatility. However, the crude IR spectrum for 11 showed the lossI of the silicon absorption and the loss of absorption in the OH region. Likewise the ether 12 was not isolated but the consumption of starting material and the IR spectral data (loss of silicon, no loss of OH) convinced us that this functional array 12 was the major product. The aldehyde 10 was S W i DOd a W mo

H = ^ ‘("h9})- =,d *• iioz 6(81 v* ¥ h o > = ^ a • Ti 91 si fi ^H0> = 2a;a *• Si'ii'S'S

<2Ih 6o > = ?a‘a i V 2'§? a ‘okaV aiAlO a V H O I v £awis. $ \. 3 .a f <5 H a wO H 159

characterized upon hydrolysis with formic acid by NMR,

TLC and IR data.

Since enolization of carbonyl compounds with this

reagent 1 appeared not to be a critical factor, other

adducts were prepared to prove the outcome of all sub

sequent transformations. The volatility and water

solubility of some of the above compounds was a problem.

We overcame this by preparing higher molecular weight

products with more hydrocarbon character to aid

isolation of the desired products. To further illustrate

the nucleophilicity of 1, we chose another enolizable

carbonyl compound, cycloheptanone 13. This ketone gave

a 65% distilled yield of 14 when treated with 1.5 eq < w v of 1. Some decomposition was noted on distillation.

The aldehyde 15 was obtained by hydrolysis with formic

acid and characterized by its NMR, IR and mass spectral

data. The enol ether 16 was obtained in 79% distilled

yield by KH in THF treatment of lh. The 0-hydroxy

ether 17 was obtained in 60% yield by several extractions

with ethyl acetate from the aqueous dimethylsulfoxide media. The vinyl ether 16 was also evident by NMR data.

The adduct 18 was obtained in 60% distilled yield with 1.5 eq of 1. Hydrolysis with formic acid gave

cyclohexylacetaldehyde 19 in 79% yield and was charac

terized by its NMR, IR and mass spectral data. The 160 cesium fluoride and dimethylsulfoxide treatment of 18 at

60°C gave 21 in 69$ yield with good mass spectral data for M+ and M+-H20. The enol ether 20 was obtained in quantitative yield and characterized by its NMR, and mass spectral data.

The important transformation to the a-methoxy ketone 22 proceeded in 56$ yield and was the first example for the two step process of oxidative nucleophilic methoxy methylenation. This coined phrase implies the original aldehyde carbon has been attacked nucleo- 0 philically by a source of CH20Me with subsequent oxidation of that carbon to provide 22. Other illustra tions of this process are to be described briefly.

Recently, A. S. Kende at Rochester used reagent

1 for the total synthesis of the antifeedant (+) isotadeonal.21 This is the first example of the use of our reagent in synthesis. 161 s Since such recent interest29 lies in the homologation of hindered enolizable carbonyl compounds a few examples were reacted with this new homologation reagent. We chose

OU-menthone 29 (prepared-by chromic acid oxidation of

(I)-menthol), to illustrate the non-basic, nucleophilic character of the bulky nucleophile. When 1.5 eq of lithio-methoxymethyltrimethylsilane was quenched at -25°C with ($,)—menthone the usual workup gave a 89$ isolated yield of the desired diastereomeric products. Hydrolysis with dilute aqueous perchloric acid in THP gave the epimeric aldehydes. Formic acid hydrolysis gave an appreciable amount of carboxylic acid. The diastereomeric adducts were treated with potassium hydride in THF and gave a 86% isolated yield of a single enol ether 32. The stereochemistry could not be assigned.

Treatment of the adducts 30 with cesium fluoride in dimethylsulfoxide gave the desired epimeric B-hydroxy-a- methoxy ethers, 33 in 8l% yield. 162

+ ,CH0 ’TMS H 30

KH Me THF 32

29 T A j U V

. 9 H CsF r ^ r ^ o M e DM SO T 3 3 163

A sterlcally crowded a,3**unsaturated aldehyde was used to illustrate another homologation to an a-methoxy ketone. Myrtenal 23 was added to 1.5 eq of the reagent at -25°C and the usual workup gave a 95$ isolated yield of the desired diastereomeric products, 24. Oxidation in dichloromethane with PCC followed by chromatography gave the a-methoxy enone 25 directly in 60$ isolated yield.

The enol ethers 26 were prepared easily with potassium hydride in THF. The E,Z-isomers were seen as a complex pattern in the NMR spectrum and obtained in 86$ yield.

Formic acid hydrolysis gave the 3 ,y-unsaturated aldehyde 27 as the only product. Relief of strain in the [^3* 1 • llbicyclic system could be probable cause for shift of the double bond into conjugation. The adducts

24 were treated with cesium fluoride in dimethylsulfoxide did not provide the ethers 28. A moderate carbonyl developed in the IR and even at 70°-80°C not all of the silicon absorption was gone. i) PCC

2) Si gel

KH THF

H f 165

Another available aldehyde was chosen to illustrate

the homologation to an a-methoxy ketone by our reagent.

2-Pyrenecarboxaldehyde 3j} was treated with 3 eq of I at -30°C and gave two polar products by TLC which had no aldehyde IR absorption (2700 cm-1) or NMR signal

for the starting material. The red oils would not solidify and PLC purification of the adducts 35 was possible for each product to give an overall 55$ yield.

About 39$ of a non polar material was isolated but not characterized. The methoxyl signal ( §3.^) was observed in the NMR and the silicon region was very broad.

The purified products 35 were treated with KH in

THF at 60°C to provide the E-methyl enol ether 37 in 70$ isolated yield. The yellow liquid would not solidify but gave excellent NMR, IR and mass spectral data. The hydrolysis of 35 or 37 did not provide the aldehyde as a stable solid. Perchloric acid, formic acid and hydrochloric acid did not provide 38 as per NMR and

IR data. When 35 was oxidized with PCG at 0°C in dichloromethane, a quick transformation to the intermediate silylated form of 36 was observed in the IR spectrum 166

0 DFCC 35 36

2) Si 9 2 P/l" ^ 0 Me

H HO KH OMe

THF H or

C S F ^ DM SO 37

H + 37 * p / r ^ N D H O .38 167

(1250 and 845 cm” 1). Isolation by silica gel column chromatography gave a 70% yield of the a-methoxy ketone

36 as a yellow solid characterized by NMR and IR data. * / w s » Unfortunately the adducts 35 did not undergo protodesilylation upon treatment with cesium fluoride in dimethylsulfoxide. By TLC comparison and IR data it was assumed that elimination of trimethylsilanol occurred to give the enol ether in poor yield. The hydroxy absorption in the IR spectrum 37 was entirely removed. This reaction, unlike other examples, was an unclean conver sion to undesired product. It is not surprising that this pyrenyl alcohol eliminates trimethylsilanol.

The intermediate positive change 3 to silicon is stabilized by donation from the tt cloud system which subsequently collapses to the vinyl ether. Perhaps temperature studies on the rates of reactions that occur in this example are necessary. Many products were evident at 60°-80°C. Sluggish reactions occurred at

R.T. This example 35 was a good illustration of the » w v limits put on protodesilylation procedures. 168

The [2,2~3 para-cyclophane (1,9) and (1,10) diones

32 that were discussed in Part I were reacted with lithio-

methoxymethyltrimethyl silane to give clean bis-adducts

in 78$ yield. IR data indicated the loss of the starting

material (1700s cm-1). NMR data could not be collected

on the bis-adducts since they were totally insoluble in

all available NMR solvents. Unfortunately, the bis-

aldehyde could not be isolated (no proton in the NMR) because of the hydrolysis conditions employed. Prom IR

data it appears to be a mixture of di-ketones. It might be added that these reactions were carried out on the smallest scale in this entire dissertation (20 mg). If more material had been supplied then other studies may have been carried out.

At the time of publication of our new reagent and the difficulties encountered with this cyclophane adduct we received a letter from Robert Dipardo at Merck,

West Point, Pennsylvania. He suggested that the 3- hydroxy-a-methoxytrimethylsilane be converted to the tosylate and subsequently desilylate with tetra-n-butyl- ammonium fluoride to provide the labile bis-vinyl ether 42.

The tosylate, however, may undergo spontaneous elimination to give the vinyl trimethylsilylenolether, which would then be desilylated. 169

The path of hydrolysis of these adducts may be quite

complex. From the observed IR and TLC data on isolated products it was evident that many products existed.

Mechanisms as those outlined below are only speculative but may account for the observed IR data (1715w, 1675s,

1605s cm-1).

0 IR 1688cm'. 0 = 3 )— 0

IR1674,1701cm"'. IR lesscm"1

0 IR 1688cm1 f 1>9J .. .r>M» CuoJ CD CL + CL CO ^ x (V [ x © O T <\> ■><) rAl 171

M e 40

r42 v A A n

Some other carbonyl compounds did not react with up to 3 or 5 eq of the reagent in THP at -20°C. 172

Extending this methodology further, we explored the

preparation of other alkoxy silyl reagents.

The trimethylsilyloxytrimethylsilylmethane 5^ would

provide the reagent needed to prepare an intermediate

that E. J. Corey obtained by a more difficult route in

his synthesis of aphidocolin.1 u However, the isolation

and purification of some of these ethers proved to be

difficult. The THP ether 52 could be distilled from the

reaction mixture easily affording a pure liquid. Subse

quent reactions with carbonyl compounds were satisfactory

although diastereomeric products were obtained.

*M e 3S,CH2OTHPj52 fyj€jSiCH2OH » Me3SiCH2OE£ .53 TMSCHOR

^ Me'jSiCHpSiMSjSAj /R,CO

TMS /

regiospecific ethers of choice. OH OR

* 173

The ethoxy ethyl 53 and trimethylsilyloxy 54 ethers

were di-fficult to isolate fro.m the reaction mixture in

pure form since the usual literature20 conditions require

an amine to act as the base during alkylation or silyl-

lation. In these cases it was extremely difficult to

remove the residual amine since it always codistilled

with the desired product. When the reaction mixture was

first washed with IN HC1 to remove the amine then the

ethereal layers only afforded a degraded form of the

desired reagent. The lability of these reagents to dilute mineral acid is not unexpected. The OMEM ether

(not listed) did not give correct mass spectral data either. At the present time only the OTHP ether looked promising.

The ether 56 was prepared in 60% isolated yield from the allylic alcohol 55 while suffering from desilylation (in situ) from trimethylsilyliodide or iodide ion to give 55 a as a side product. KH in THF with l8-crown-6 gave a 111 mixture of 56 and 55a.

Two studies were done to prepare the homoallylic alcohol from 56. It was surmised from proto desilylation34 just described and fragmentation literature to be discussed shortly that fluoride induced desilylation of the substrates would provide neutral media to carry out the same transformations (B -*■ C) as Still's alkyllithium

(basic media) exchange with a novel tin reagent. 174

However, these efforts were fruitless. The pure starting material 56 gave excellent NMR, IR, and mass spectral data. No evidence was found for the desired homoallylic alcohol 57. Incredibly the only product < y w v obtained from these reactions was the original allylic alcohol 55- No Lewis acid conditions were employed. Another area in this lithiation concept was studied.

It would be attractive to provide an Intramolecular attack of a lithio species upon an allylic ether. The overall effort gives the homoallylic alcohol that is

1,4 transposed in oxygen functionality [B Cj. Attempts to lithiate a to silicon 56a were met with failure with s-BuLi. The NMR spectrum indicated the loss of silicon functionality. The clean product of lower Rf than

56 gave (M+ ) 250 and 238 that were too small to mass measure. No evidence for starting material 56 or product

58 or 59 was obtained. n-BuLi gave the identical TLC and mass spectral data. 176

The silyl ether 56 was treated with 0.1 to 2.1 eq of cesium fluoride in dimethylsulfoxide and THP (211') at 80°C. The progress of the reaction was monitored by

TLC and IR spectra. After 0.5 h an intense hydroxyl absorption developed while the silicon absorption disappeared. The mass spectrum showed parent ions at

M+ 152 for the alcohol 55 and evidence to small to mass measure of the possible (M+-H20) 150 for 57 with no parent •SAM ion (M+ ) 166 for 57. A treatment in neat dimethylsulfoxide (no THP) gave no trace of M+ 150. A cleaner conversion to 55 was realized. No conversion to 57 was seen with cesium fluoride in THF at 60°C overnight and the starting material 56 was recovered. Potassium fluoride dihydrate in DMSO or THP with l8-crown-6 ether did not give any evidence of reaction by TLC and IR spectra. 177

The Cuprate of Lithio-methoxymethyltrimethyl Silane

Prom various studies of the preparations and use

of lithium organocuprate reagents it has become apparent

that two practical problems complicating the general

use of these reagents are inadvertent thermal decom

position and inadvertent oxidation leading to coupling

of the organic residues.30 Both of these initial side

reactions often.lead to the formation of Cu(0), which

usually appears as a black colloidal suspension in the

reaction mixture and is believed to catalyze the decom position of still more copper reagent.31 This fact led

us to examine other possible Cu(I) derivatives that might offer the advantages of both ether solubility and easy purification to separate unwanted Cu(I) impurities.

In earlier work,32 it was noted that solubility advantages are offered by several Cu(I) halide complexes such as n-Bu3PCuI, (MeO)3PCuI, (MeO)3PCuBr, and especially, the liquid complexes (n-BuaS)aCuI and (n-BuaS)aCuBr.

The use of these complexes in synthetic work is made less attractive by the relatively high boiling points of the ligands, n-BuaS (bpt 189°C), n-Bu3P (bpt 150°C, 50 mm), and (MeO)3P (bpt 112°C) that complicate their removal from reaction products and by the persistent disagreeable odor associated with phosphine and phosphite ligands.

We were attracted by reports indicating that complexes of certain Cu(I) salts with the ligand Me2S (bpt 37°C) 178 were both soluble in ether and could be obtained as

crystalline solids.33

OJvie 60 •TMS Y 46-60* ■ o YCSF *

J d ~ or

The cuprate (MeOCH)2CuMe2S 'Me TMS 60 66 179

Analogous to the 1,4 addition of CTC cuprate

discussed in Part I we tried to effect formation of the

cuprate of lithio-methoxymethyltrimethylsilane 1.

Initially CuCN was tried and no 1,4 addition product

could be isolated. The insoluble saltsolution in THP

and HMPA became bright orange and never fadedwhen

quenched with cyclohexenone. This color in the CTC

cuprate reaction was rapidly consumed at that point. A

repeated reaction of lithio-methoxymethyltrimethylsilane

and Me2SCuBr in Et20 easily overcame the problem associated

with insoluble CuCN derivatives.19

MeO^ MeO CH-4 CuMe2S 60 £ H — Li 1 / 2 — / TMS TMS

When a solution of 60 (26 mole percent of Me2SCuBr) was quenched with cyclohexenone at -78°C, a green solution was formed. A red solution developed upon slow warming to R.T. Workup gave a 60% yield of two polar diastereomeric 1,4 addition products. Unfortunately, this was the only enone that our reagent would add in 1,4 fashion.

This reagent is apparently a bulky nulceophile. However, it was not possible to desilylate the intermediate 63 to 66 under conditions that were described earlier to remove a silicon substituent on the 0-hydroxy-a-methoxy methane.

Similar treatment of 62 did not provide 65. The enol 180 ether 64 was never detected. The Lewis acid, BF3»Etz0 did not effect the reaction of 62 to 64 or 63 to 64. ■«w -wv vwv These adducts 62, 63 did not behave like the 3-hydroxy a-methoxytrimethylsilyl methanes discussed earlier in

Part II.

With the excellent reviews on fragmentations mediated by fluoride and the like12 it was indeed disappointing to see this study as a failure. EXPERIMENTAL TO PART II 181 6 Preparation of Methoxymethyltrimethylsilane 1 .

Sodium metal (2.6 g, 0.113 mmol) was washed with diethyl ether and added to 50 ml of dry methanol (distilled from sodium metal) in a 100 ml round-bottom flask equipped with N2 inlet and reflux condenser fitted with a drying tube. After dissolution of the metal, chloromethyltrimethylsilane (freshly distilled bpt. 97°C atm. press.)

(13.8 ml, 0.1 mmol) was added and the contents of the flask were heated to 60°C for 15 h. Distillation with a Vigreux column gave 15 • g of a clear liquid bpt. 59°-59.5°C.

The liquid was taken up with 20 ml of water and shaken.

The upper layer was separated and washed with water again.

The upper layer was dried (Na2S0u), filtered and distilled with a Vigreux column to afford 5.8 g 50% (lit. yield

75$) of a liquid (bpt. 83°C atm. press.) NMR (CCluK 6

0.2 (s, 9H), 3.1 (s, 2H), 3-4 (s, 3H). vpc analysis showed less the 8% of starting material present. EXPERIMENTAL TO PART II

Preparation of 2-Hydroxy-ll-trimethylsilyl-ll-methoxy

Adamantylidene 3

To a dry 25 ml three necked flask under argon was added dry THP (6 ml from LAH) and methoxymethyltrimethyl silane (0.66 ml, 4.23 mmol, 1.1 eq). The consents were cooled to -78°C and slowly sec-butyllithium (1.4 M in cyclohexane) (3-0 ml, 4.23 mmol, 1.1 eq) was added via syringe. The reaction was warmed to -25°C and held there for 0.5 h. After recooling to -35°C to the pale yellow solution was added solid adamantanone (0.57 g, 3*8 mmol,

1.0 eq). The solution became milk white. Slowly warming to R.T. over 1.5 h resulted in a clear, colorless solu tion. The contents were poured into saturated NHUC1

(30 ml) and extracted with diethyl ether (2X30 ml). The organic layers were washed with water (2X20 ml) and brine (10 ml). Drying (MgS0«*) and rotary evaporation gave a white solid 0.91 g (89$) with no evidence of starting material. Preparative layer chromatography on silica gel (411 Petrol/EtOAC) of the crude product gave

0.65 g (64%) of pure product as a homogeneous epot on

TLC. mpt. 65°-67°C. IR (Nujol) 3500s, 2900s, 2850m,

1450s, 1375w, 1320w, 1250s, 1170m, 1070s, 1050m, 990s,

182 183

930m, 910m, 870m, 840s. cm"1. NMR (CDC13) 6 0.1 (s,

9H), 1.65 (br.s, 10H), 1.7 (br.s, 4H), 2.2 (br.s, 1H),

3.4 (s, 3H). M/e (M+-H20) Calc, for CnHzeSiOa 250

(87.5), 236 (60), 223 (40), 205 (70), 178 (100), 163

(17.5), 151 (60), 131 (50), 121 (100).

Protodesilylation of 3- Preparation of 2-Hydroxy-ll- methoxy Adamantylidene 4

The methoxysilane (50 mg, 0.186 mmol, 1 eq) was added to DMSO (3 ml) with anhydrous cesium fluoride

(0.04 g, 0.2 mmol, 1.2 eq) under argon. After heating for 1 h at 80°C the contents were poured into water (10 ml) and extracted with ether (2X10 ml). Drying (MgSCU) and rotary evaporation gave 36 mg (98%) of a pure solid mpt. 86°-87°C (sublimes at 32°C at 0.1 mm). The same reaction conditions with KP.2H20 does not give the product.

IR (Nujol) 3520m, 2900m, l485w, l470w, 1460m, 1350m,

1340w , 1210m, 1195m, 1170, 1100s, 1050m, 1035m, 965m,

955m, 935m, 875w, 880w, 8l0w. cm-1. No silicon retained.

NMR (CDC13) 6 1.2 (s, 4H), 1.7 (s, 10H), 2.1-2.3 (br.s,

2H), 3.4 (d, 3H) J=6Hz. Anal, for C23H 2o02 Calc. 73-46%C,

10.20%H. Pound 73.66%C, 10.38%H. M/e 196 (0.26), 178

(3), 164 (2), 152 (10), 151 (100), 150 (3.6), 148 (3-1),

135 (2), 109 (2.3), 107 (5), 105 (2.9), 95 (3), 93 (6.3),

91 (13.3), 81 (7), 79 (7.7). 184 6 Preparation of Adamantylidene Oxide 5.

Sodium hydride (300 mg, 0.012 mmol) and dry DMSO

(20 ml) was heated under argon at 60°-70°C for 2 h in

a 50 ml three necked flask fitted with a reflux condenser.

A milky solution resulted. The contents were cooled to

0°C and dry THF (5 ml from LAH) was added. Trimethyl-

sulfonium iodide (2.45 g, 0.012 mmol) was added and

stirring done for 10 min at 0°C. Solid admantanone

(1.5 gj 0.01 mmol) was added. The contents were stirred

at 0°C for 1 h, then at R.T. for 2 h. The mixture was poured into water (100 ml) and extracted with diethyl ether (2X50 ml). The organic layers were washed with brine (20 ml). Drying (MgSO*) and rotary evaporation gave a white fluffy solid 1.6 g (.97%) mpt. ca. 40°C.

The solid is volatile on evaporation to 0.1 mm. NMR

(CDCla)8' 1.4 (s, 2H), 1.9 (br.s, 12H), 2.6 (s, 2H),

Mass spec. Calc. 164.120, Obs. 164.120 M/e 164

(100), 148 (25), 135 (44), 122 (44), 106 (19). 185

Proof of Structure of S-Hydroxy-a-methoxymethane 4 from 5

Anhydrous potassium hydroxide (65 mg, 0.8 mmol,

1.2 eq) was dissolved in dry methanol (1 ml from sodium) under argon. At R.T., the epoxide5 (120 mg, 0.7 mmol,

1.0 eq) in dry methanol (2 ml) was added. After four days of heating at 55°C under argon (methanol was added periodically) the starting material was consumed as evidenced on TLC. The contents were poured into saturated

NHUC1 (10 ml) and extracted with diethyl ether (3X10 ml).

The organic layers were dried (MgSOi,) and rotary evapora tion gave 116 mg (85$) of a white solid, mpt. 86°-87°C.

NMR (CDC13) 6 1.4-2.2 (br.m, l4H), 3.A (s, 3H), 3-5

(s, 2H). IR (Nujol): 3520w, 2900m, l485w, l470w, 1460m,

1350m, 1340w, 1210m, 1195m, 1170w, 1100s, 1050m, 1035m,

965m, 955m, 935m, 880w, 875w, 840w, 8l0w cm-1. Mass spec.

Calc, for Ci2H20O2 196.146, Obs. 196.147. M/e 196

(0.13), 179 (0.3), 178 (1.9), 164 (0.96), 152 (12.5),

151 (100), 150 (3 .8 ), 135 (1.3), 133 (3 .8 ), 121 (O.9 6 ),

109 (2.8), 107 (6.25), 105 (3), 95 (3.8), 93 (5.7), 91

(11.5), 81 (7.7), 78 (7.7). 186 Preparation of the Dimethyl Acetal of 2-Adamantane

Carboxaldehyde

The epoxysilane (230 mg, 0.97 mmol) obtained by the reaction of CTC and adamantanone1 4 was dissolved in 2 ml of methanol at 0°C and three drops of trifluoro- acetic acid was added. The contents were stirred for

1 h and no change in the Rf of an aliquot was noted on silica. Overnight stirring produced no change. The contents of the flask were poured into saturated NaHC03

(30 ml) and extracted with diethyl ether (2X30 ml). Drying

(MgSOi*) and rotary evaporation gave an oil which would not crystallize. PLC isolation afforded a homogeneous oil. NMR (CClu) <5 ! no silicon 1.6-2.0 (br, 3, l4H),

3.15 (s, < 1H), 3.40 (s, 3H), 3.55 (s, 1H), 3.70 (s, 1H).

IR (film): no OH 2900s, 2850s, 2650w, 1720m, 1450s,

1385w , 1360w , 1350s , 1310w , I260w, 1250w, 1190w, 1135s,

1100s, 1065s, 1045s, 1040s, 985m, 970m, 960s, 940w,

910m, 880w, 840w, 820w cm-1. M/e on C 13H 2202. 210 (1.6)

194 (1 .6 ), 179 (38), 178 (100), 165 (89), 164 (20.4),

150 (19-5), 149 (12.2), 135 (31), 121 (31), 105 (20.4),

104 (17.1), 93 (24.5), 91 (48.3), 75 (100). This hydrolysis did not give the expected intermediate 3. 187

Hydrolysis of 3* Preparation of 2-Adamantane Carbox- aldehyde 6.17

The 3-hydroxy a-methoxysilane (50 mg, 0.186 mmol,

1.0 eq) was stirred with 90% formic acid (1 ml) at 0°C to R.T. over 0.5 h. Rotary evaporation and then evapor ation to 0.1 mm afforded ca. 30 mg (89#) of a pasty solid, mpt. 99°-102°C. NMR (CDC13) 6 : 1.75-2.00 (br.d,

10H), 2.10s, 1H), 2.20 (s, 2H), 2.45 (br.s, 1H) , 9-7

(s, 1H), (no C02H signal in NMR). IR (Nujol)! 2900s,

2850s, 2700w, 1725s, 1450m, 1100m, 935m, 875w, cm-1.

Mass spec. Calc, for CnHisO 164.120, Obs. 164.120.

M/el 180 (24), 164 (62), 162 (31), 150 (13.8), 136

(34.5), 135 (100), 134 (55). The sample was easily air oxidized to the carboxylic acid. It is known that this aldehyde is unstablei7 188

Preparation of 2-Adamantylidene Methyl Enol Ether 7

The 3-hydroxy a-methoxysilane (50 mg, 0.18 mmol)

was added to a slurry of washed KH dispersion (20$ in oil

washed 3X with dry pentane and decanted, 0.2 g, 1.2 eq,

0.8 mmol) in dry THP (4 ml from LAH) under argon. After

heating to 60°C for 1 h, the contents were poured into

saturated NH4CI (10 ml) and extracted with diethyl ether

(2X20 ml). The organic layers were washed with water

(2X10 ml) and brine (10 ml). Drying (MgSOn) and rotary evaporation gave an oil 28mg (87%) of pure product with no evidence of starting material. The oil solidified at refrigerator temperature. IR (film): 2920s, 2860s,

1460s, 1370s, 1300w, 1200w, 1150m, 1120m, 1030w, 970w,

930w, cm-1. NMR (CDC13) 6 no silicon, 1.8-2.2 (br.d,

14h), 3.6 (s, 3H), 5.8 (s, 1H). Mass spec. Calc, for

CxaHiaO 178.136, Obs. 178.136. M/e 179 (13.5), 178 (100),

163 (10.8), 151 (2.7), 135 (4.8 ), 131 (5•7), 121 (24.3). 189

Preparation of l-Hydroxy-7-trimethylsilyl-7-methoxy

Cyclohexylidene 9

To a dry 25 ml flask under argon at -30°C was added dry THF (5 ml from LAH)and methoxymethyltrimethylsilane

(bpt. 83°C atm. press.) (0.66 ml, 4.2 mmol, 1.1 eq).

Slowly via syringe sec-butyllithium (1.4 M in cyclohexane) was added (3.0 ml, 4.2 mmol, 1.1 eq). No yellow color was consumed. The mixture was warmed 0.5 h, to -25°C or -30°C, and at -20°C to -25°C color is consumed to give a pale yellow solution. The mixture was cooled to -35°C and distilled cyclohexanone was added (0.4 ml, 3.8 mmol,

1.0 eq). After warming to R.T. over 1.5 h., the reaction became clear .and colorless. The contents were poured into saturated NHnCl (20 ml) and extracted with diethyl ether (2X30 ml). The organic layer was washed with water

(2X10 ml) and brine (10 ml). Drying (MgSOu) and rotary evaporation gave 0.6 g (73$) of a pure product containing no starting material. The product was distilled at < 60°C at 0.05 mm. NMR (CDCI3) *. 6 0.0 (d, 9H) 311 0.8 (br. m, 2H), 1.35 (br.s, 10H), 1.70 (br.m, 2H), 2.15 (br.m,

2H), 2.45 (s, 1H), 3.30 (s, 3H). IR (film): 3450br.s,.

2940s, 2850s, 2820m, 1700w, 1450m, 1375w, 1350w, 1310w,

1300w, 1260m, 1250s, 1175w, 1150w, 1090s, 1050w, 980w,

960w , 940w , 890w, 870m, 840s, cm-1. Mass spec. Calc, for

Ci iH2 n Si02 216.1545, Obs. 216.1549, M/e I 216 (0.3), 201

(1.5), 184 (3.5), 183 (2.7), 178 (2.5), 171 (19), 169 190 (5.2), 163 (4.2), 126 (33.3), 123 (12.5), 111 (7-3), 103

(11.4), 99 (100).

Hydrolysis of 9. Preparation of Cyclohexanecarboxaldehyde

The adduct 9 (200 mg) was stirred at R.T. with

90$ formic acid (2 ml) for 1 h. .The contents were poured into saturated NaHC03 (10 ml) and extracted with diethyl ether (3X10 ml). Drying (MgS04) and evaporation gave a liquid that was identical to an authentic sample of 10 on TLC and VPC. NMR (CC1U) 5 1.5 (br.s,

10H), 2.3 (br.m, 1H), 9-7 (d, 1H) J=2Hz. IR (film):

2920s, 2850s, 2700m, 1725s, 1450m, 1440m, l4l0w, 1375w,

1340w , 1250m, 1120w, 1050w, 940w, 830m, 750w cm"1. 191

Preparation of Cyclohexylidene Methyl Enol Ether 11

The 0-hydroxy a-methoxysilane (150 mg, 0.69 mmol,

1.0 eq) was added to a slurry of washed KH dispersion

(20% in oil washed 3X with dry pentane and decanted,

0.2 g, 1.2 eq, 0.0 mmol) in dry THP (3 ml from LAH) under argon. After refluxing for 1 h the contents were poured into saturated NHUC1 (10 ml) and extracted with diethyl ether (2X20 ml). The organic layer was washed with water

(2X20 ml) and brine (10 ml). Drying (MgSOu) and rotary evaporation gave ca. 100 mg of a volatile liquid with no evidence of starting material. Due to the volatility of this ether isolation and characterization proved difficult. The next adduct in the series gave good spectral data. 192

Desilyation of 9* Preparation of 1-Methoxymethyl-l-

cyclohexanol 12

The adduct 9 (150 mg) was stirred at 80°C for 1 h

with cesium fluoride (117 mg) and dimethyl sulfoxide

(2 ml). The contents were poured into water (10 ml)

and extracted with ether (4X10 ml). Drying (MgSOu)

and evaporation gave 70 mg (69%) of an oil that was

contaminated by DMSO. The desired product could not

be distilled for purification. The crude NMR spectrum

indicated that some of the vinyl ether 11 was present

(broad singlet at 6 5 .7 ) as well as some of the desired product (singlet at 6 3.3 (-CH20CH3)). This was an

example of the loss of Me3SiOH and should be compared

to the preparation of 4 where this is not observed. 193

Preparation of l-Hydroxy-8-trimethylsilyl-8-methoxy

Cycloheptylidene 14

To a dry flask under argon at -30°C was added dry

THF (20 ml from LAH) and (2.1 ml, 13*37 mmol, 1.5 eq) of methoxymethyltrimethylsilane. Slowly s-BuLi (10 ml,

1.5 eq) (1.3 M) was added and the contents were stirred to — 15°C over 1 h, then recooled to -30°C. Distilled cycloheptanone (1 ml, 8.9 mmol, 1 eq) was added dropwise.

The contents stirred at -30°C for 0.5 h, then were slowly warmed to 0°C to R.T. over 1 h. The solution was poured into saturated NH4C1 (50 ml) and extracted with diethyl ether (3X20 ml). Drying (MgS04) and evaporation gave a liquid that was distilled at 2.2 mm to give 1.3 g (65%) bpt. 103°C. Some decomposition was noted on distillation.

NMR (CDC13)I 6 0.1 (s, 9H), 1.5 (br.s, 12H), 2.6 (s,

1H), 3*2 (s, 1H), 3*4 (s, 3H). IR (film): 3450s, 2920s,

2840m, 2800w, 1460m, l440w, 1250s, 1080m, 1050w, 1020w,

920w, 900w, 860m, 840s, 750s, cm-1. 194

Hydrolysis of 14. Preparation of Cycloheptane- carboxaldehyde 15

When the adduct 14 was treated with either formic acid, hydrochloric acid or perchloric acid at R.T. only a small amount of the desired aldehyde 15 was detected in the NMR spectrum. (CC14) 6 1,5 (br.d, 12H), 2.4

(br.s, ca. 3H), 8.0 (br.s, < 1H), 9*6 (weak d, J=2H,

< 1H). IR (film)! weak 1720 cm-1. The mass spectrum contained high molecular weight peaks not of the desired product. This aldehyde is a known compound and a stable distillable liquid. 195

Preparation of Cycloheptylidene Methyl Enol Ether 16

The silane (250 mg, 1.08 mmol) was added to a slurry of (260 mg, of 23.5% KH in oil) (64.8 mg, 1.62 mmol,

1.5 eq) in THP (10 ml) and heated at reflux for 10 h.

The contents were poured into saturated NaHC03 (20 ml) and extracted with ethylether (3X10 ml). Drying (MgSOu) and evaporation then distillation at 1.5 mm gave a liquid 120 mg {19%). bpt. < 80°C. NMR (CDC13) 6 :

1.5 (br.s, 8H ), 2.0-2.4 (br.m, 4H), 3.6 (s, 3H) , 5.8

(m, 1H) J=2Hz. IR (film): 3450s, 2920s, 2850s, 1670s,

1460s, 1440s, 1370w, 1340w, 1370w , 1250m, 1210s, 1190s,

1150s, 1120s, 1050s, lOOOw, 960w, 890w, 830m, BOOw,

750w, cm-1. 196

Desilylation of 14. Preparation of 17*

Results identical to the preparation of 12 were obtained. Only a 60$ yield of product was isolated which contained both the enol ether 16 and desilylated product 17. This was obtained by several extractions

3H), 5-7 (s, 1H). This rate of elimination to the enol ether 16 is fast at 80°C and should be compared to when 30 (adduct of 1 with (I )-menthone was treated at

60°C for 12 h. No reaction took place except when raised to 70°C for 24 h. Even at 60°-70°C some of the enol ether comes from the simple cases of 9, 14, and *%» 18 at 1.1 eq of cesium fluoride. 197

Preparation of threo and erythro 1-Hydroxy-l-cyclohexyl-

2-methoxy-2-trimethylsilyl Ethane 18

To a dry 25 ml three necked flask under argon was a-ded dry THP (from 8 ml LAH) and methoxymethyltrimethylsilane (1.0 ml, 6.68 mmol, 1.5 eq) and the contents were cooled to -78°C. Slowly sec-butyllithium (1.4 M in cyclohexane) (4.77 ml, 6.68 mmol, 1.5 eq) was added.

The reaction mixture was warmed to -25°C and held there for 0.5 h. After recooling to -35°C to the pale yellow solution was added distilled cyclohexanecarboxaldehyde

(0.54 ml, 4.4 mmol, 1.0 eq) whereby the solution became white. Slow warming to R.T. over 1.5 h resulted in a clear, colorless reaction mixture. The contents were poured into saturated NH*C1 (30 ml) and extracted with diethylether (2X30 ml). The organic layers were washe with water (2X20 ml) and brine (10 ml). Drying (MgSOu) and rotary evaporation gave 0.8 g (80$) of a liquid which was two spots on TLC with no evidence of starting material. The product was distilled at 69°C at 0.2 mm.

IR (film): 3450m, 2920s, 2860s, 1450m, 1250s, 1100m,

1080m, 1060m, 1040m, 980m, 940w, 890w, 845s, 760w cm"1.

NMR (CDC13): 6 0.2 (s, 9H), 1.1-1.9 (br.d, ca. 10H),

2.4 (s, 1H), 3.1-3.2 (d, 1H J=2.5Hz), 3.5-3-6 (d, 3H

3:1 J=2.5Hz) no mass ion could be obtained from this secondary alcohol. M/e 215 (4.4), 199 (3.7), 198 (6),

197 (7.4), 183 (12.6), 173 (2.5), 167 (23), 155 (15-6),

147 (23), l4l (15.6), 140 (100), 109 (46), 108 (100). 198

Hydrolysis of 18. Preparation of Cyclohexyl Acetaldehyde 19

The 3-hydroxy a-methoxysilane (50 mg) was stirred with 90% formic acid (1 ml) at R.T. for 0.5 h. Rotary evaporation and then evaporation to 0.1 mm gave 21 mg

(7650 of an oil. IR (film): 2720s, 1710s, cm-1. NMR

(CDCla): 6 0.8-1.8 (br.d, 11H), 2.0-2.2 (d, 2H), 9-6

(t, 1H). Reisolation from NMR sample gave a low melting solid, presumably the carboxlic acid. 199

Preparation of els and trans l-Methoxy-2-cyclohexyl

Ethylene 20

The 3-hydroxy a-methoxysilane (0.5 g, 2.1 mmol, 1

eq) was added to a slurry of KH dispersion (1 g, 2 eq,

20$ in oil, washed 3X with dry pentane and decanted) in

dry THP (8 ml from LAH) under argon. After refluxing for

3 h the contents were poured into saturated NHuCl (20 ml)

and extracted with diethyl ether (2X30 ml) and brine

(20 ml).' Drying (MgSO*) and rotary evaporation gave

0.29 g (100$) of a pure product with no evidence of

starting material. The sweet smelling oil gave IR (film)!

2920s, 2840s, 1720w, l660s, 1650s, 1440s, 1375m, 1280w,

1250m, 1225m, 1200m, ll60w, 1100s, 1080w, 970w, 940m,

880m, 830w. cm"1. NMR (CDC13) 6 no silicon, 1.1-1.9

(br.m, 10H), 2.1 (s, 1H), 3-6 (d, 3H), 5-8-5.9 (d, 1H)

J=6Hz , 6.2-6.5 (d, 1H) J=12Hz. mixture of E,Z isomers.

Mass spec. Calc, for C9H 160 140.12010, Obs. 140.12056

M/el 140 (26.8), 119 (5-5), 108 (33-3), 97 (100), 67

(66.6), 41 (77), 32 (7 0), 28 (100). 200 Protodesilylation of 18. Preparation of (+)-Hydroxy-

l-cyclohexyl-2-methoxy ethane 21

To a dry flask under argon was added (150 mg, .70 mmol) of the starting material (two spots on TLC) and treated with (117 mg, 0.77 mmol, 1.1 eq) of anhydrous

CsF. Dimethylsulfoxide (5 ml) was added and the contents were heated at 80°C for 1 h. Consumption of the starting material was evident by IR whereby the silicon 1250,

850 cm-1 absorption was lost. NMR (CDC13)I 6 1.1-1.9

(br.s, 11H), 2.4 (br.s, 1H), 3-1-3.2 (m, 1H), 3-5-3-6

(d, 3H) J=3Hz, 3-7 (d, 2H) J=3Hz. IR (film): 3^50m,

2920s, 2860s, 1450m, 1100m, 1080m, 1060m, 1040m, 980m,

940w, 890w, 8l0w, 760w, cm-1. M+-H20 Calc, for C9Hi60

140.120, Obs. 140.120. M/e: 158 (0.5), 140 (5),

130 (1.6), 113 (20.7), 112 (9-1), 108 (5), 197 (6.6),

96 (5-8), 95 (100). Only a 60% yield of this product was observed. The lower yield may be attributable to the partial solubility of 21 in water. 201

Oxidation of 18 or 21. Preparation of 8-Methoxy Acetyl

Cyclohexane 22

Pyridinium chlorochromate (PPC)1U (70mg, 0.325 mmol,

1.5 eq) was slurried in 1 ml dry dichloromethane at 0°C.

The 0-hydroxy a-methoxysilane (50 mg, 0.217 mmol, 1 eq) was added with 1 ml dry dichloromethane. The orange solution became black after 1 h and the contents of the reaction were diluted with diethyl ether (15 ml). The murky mixture was filtered twice through a bed of Celite and rotary evaporation afford an oil. PLC gave 31 mg

(50%) of a clear oil. IR (film): 2920s, 2860s, 1725s,

1450s, 1375w, 1310w, 1250w, 1100s, 890w, 845w, 750w cm"1.

NMR (CC1„): 6 1.0-1.8 (br.m, 10H), 2.1 (br.s, 1H),

3.3 (s, 3H), 3.8 (s, 2H). Mass spec. Calc, for Ci2H2USi02

228.1545, Found. 228.5320. M/e 228 (2.3), 212 (1.75),

198 (7), 197 (6), 183 (7 .6 ), 182 (4.6), 167 (8), 165

(4), 155 (7.6), 143 (9.3), 140 (33.3), 139 (5.8), 108

(5 0 ), 97 (63.3), 89 (41.7), 83 (35-7), 79 (42.5), 75

(55.5), 73 (100). 202

Preparation of threo and erythro 1-Hydroxy-l-(7,7-

dime thy lbicyclo £3 .1 »21 . ljhept-l-ene )-2-methoxy-2-

trimethylsilyl Ethane 24

To a dry flask under argon at -30°C was added

dry THE (15 ml) and (1.55 ml, 9-9 mmol, 1.5 eq) of methoxymethyltrimethylsilane. Slowly s-BuLi (7.86 ml,

1.5 eq) (1.3 M) was added dropwise. The contents were

stirred to -15°C over 1 h, then recooled to -30°C.

Distilled Myrtenal 23 (crude material is commercially

70$ pure) (1 g, 6.6 mmol, 1 eq) was added in THP (1 ml) and the contents were stirred at -30°C for 1 h and to

R.T. over 1.5 h, then poured into saturated NHUC1 (50 ml) and extracted with ethyl ether (3X30 ml). Drying

(MgS04) and rotary evaporation gave an oil of 1.5 g

(85$) crude IR data indicated no starting material was left unconsumed (no 1700 cm-1). TLC showed a homo geneous product of identical Rf as the endl but no UV activity was noted. NMR (CClu) <5 1 0.1 (d, 9H) J= < 1Hz,

0.8 (d, 6H) J-3Hz, 1.3 (s, 4h), 2.3 (br.m, 3H), 2.8

(triplet of doublets, 1H) J=3Hz, 3.3 (d, 3H) J=3H),

3.85 (m, 1H), 5.4 (br.s, 1H). Mass spec. calc, for

CiSH2eSiO (M+-H20) 250.175, obs. 250.175. M/e 250 (2.6),

222 (5.1), 221 (14.3), 178 (14.3), 177 (10.2), 164 (17.9),

163 (12.8), 149 (4.6), 147 (100). 203

2-Acyl-2’-methoxy 2— (6,6-dimethylbicyclo £$.1.l3hept-2- ene 25

The adduct 24 (250 mg, 0.93 mmol) was stirred at

0°C under argon with dichloromethane (6 ml) and (301 mg,

1.4 mmol, 1.5 eq) of PCC. After stirring to R.T. for 1 h the black contents were evaporated and the residue then diluted with diethyl ether (20 ml). Elution through a short column of silica gel layered with Celite gave

110 mg (60%) of an oil that was one major spot on TLC.

NMR (CC14) crude: 6 0.8 (br.d, J=6Hz , 3H), 1.3 (br.d,

J=6Hz, 3Hz), 1.6-2.5 (br.m, 8H), 3-3 (br.s, 3H), 6.5

(br.s, 1H). IR (film) 3010w, 2920s, 2860s, 2820w,

1670s, 1610m, 1460m, 1370m, 1360m, 1240m, HOOw, 1050m,

830m, 780w, 750w, 700w cm-1. The olefinic.(C=CH) stretch in the IR spectrum was evident. Unfortunately, no mass ion for C i2H i 802 was obtained (M+ 194). M/e obs. 295 (< 1), 252 (2.5), 223 (3.3), 222 (4.7), 221

(20.7), 179 (1.3), 178 (2.0), 177 (10.0), 149 (17.2),

148 (17.2), 147 (100). 204

(E,Z)-l-Methoxy-2-(6,6-dimethylbicyclo[j.1.l]hept-2-ene)

Ethylene 26

To a dry flask under argon equipped with a reflux

condenser was added dry THF (6 ml) and (250 mg, 1.17

mmol, 1.5 eq) of KH (23-5% in oil) (47 mg KH). The

adduct 24 (210 mg, 0.78 mmol) was added dropwise with

THP (1 ml). An immediate red solution developed and

vigorous bubbling was noted. The contents were heated

at 60°C for 1 h then cooled and poured into saturated

NaHC03 (20 ml) and extracted with diethyl ether (3X20 ml).

Drying (MgSOi,) and evaporation gave an oil that was

distilled at 1 mm (bpt. < 100°C). The NMR data gave

evidence of a mixture of E,Z isomers. NMR (CC1«) 6

0.9 (br.s, 3H), 1.1 (br.s, 2H), 1.3 (br.s, 3H), 1.7

(br.m, 2H), 2.0-2.6 (br.m, 2H), 3-3 (s), 3.6 (s) ratio of 1.611 for 3H, 5.2-5.8 (br.m of overlapping quartets plus singlet for overall 3H. IR (film): 3030w, 2920s,

2880s, 2840s, 1670s, 1640m, l600w, 1460s, 1375m, 1360m,

1270w, 1190w, 1100m, 1050m, 960w, 890w, 830m, 800w,

780w cm-1. Mass spec. Calc, for C i 2H i 80 178.135, obs. 178.136. M/e 179 (4.2), 178 (18.2), 177 (6.3),

163 (9.1), 149 (22.7), 148 (18.2 ), 147 (18.2 ), 135

(31.8), 133 (18.2), 131 (18.2), 121 (22.7), 119 (27.3),

105 (91), 93 (50), 91 (100). Impurity! 204 (4.2), and 207 (4.2). 205 2-(6,6-Dimethylbicyclo£3•1.l]hept-2-ene Acetaldehyde 27

Hydrolysis of 24 under the usual formic acid conditions did not provide »wv 27. This may be due to the lability of the allylic alcohol to strongly acidic media.

When the enol ethers 26 were stirred with 6n HC1 in methanol-tetrahydrofuran at R.T. over 1 h two major products were evident by TLC. These products were identified in the NMR spectrum as the conjugated aldehyde (6 9-3, d,

1H) and deconjugated aldehyde (5 9.7, t, 1H). 206

Preparation of 1-(1-Methoxy-l-trimethylsIlyl)methyl-5- methyl-2-(1-methylethyl)cyclohexanol 3£

To a dry flask under argon at -30°C was added dry THF (15 ml) and (1.55 ml, 9.9 mmol, 1.5 eq) of methoxymethyltrimethylsilane. Slowly s-BuLi (1.3 M),

(7.8 ml, 1.5 eq) was added slowly dropwise. The contents were stirred to -15°C over 1.h and then recooled to -30°C. Distilled (I )-menthone 29 (1 ml, "/W 6.65 mmol, 1 eq) was added in dry THF (2 ml) and the contents were stirred to R.T. over 1.5 h, then poured into saturated NH«C1 (50 ml) and extracted with diethyl ether (3X20 ml). Drying (MgSOu) and evaporation gave a liquid that was distilled at 1.5 mm to give 1.6 g

(89%) of a liquid (bpt. 83°C). NMR (CC1„) 61 0.1

(d, 9H) J= < 1Hz, 0.6-0.9 (dd, 9H), J-3Hz, 1.1-1.9

(br.m, 9H), 3-0 (s, 1H), 3-2 (d, 1H) J=3Hz, 3.4 (s, 3H).

IR (film): 3500w, 2940s, 2920s, 2040m, 1450m, 1380m,

1360m, 1320w, 1260w, 1250s, 1175m, 1150m, 1080s,

1070m, lOOOw, 980w, 940w , 930w , 910w , 880w , 860m,

840s, 780w, 760w, 750m cm”1. Mass spec. Calc, for

Ci5H 30SIO (M+-H20) 253.972, obs. 253.972. M/e 254 (1.5),

240 (1.7), 239 (1.7), 228 (1.7), 227 (7-7), 211 (1.7),

198 (3.4 ), 197 (28.2 ), 173 (1.7), 172 (10.2 ), 169

(3 .8 ), 155 (17.9), 140 (1 0 .2 ), 139 (100). 207

Hydrolysis of 30. Preparation of (cis and trans)-l-Formyl-

5-methy1-2-(1-methylethyl)cyclohexane 31

The 0-hydroxy-a-methoxytrimethylsilyl methane 30

(25-0 mg, 0.9 mmol) was stirred with 88% formic acid

(2 ml) at R.T. for 1 h. The orange solution was then poured carefully into saturated NaHC03 (20 ml) until basic and extracted with diethyl ether (3X10 ml).

Drying (MgSCU) and evaporation gave a pale yellow minty liquid that was distilled at 0.4 mm to give liquid 150 mg (98%) of a liquid (bpt. < 80°C). NMR

(CCU) <5: 0.7-1.0 (br.d, J=6Hz, 9H), 1.0-2.2 (br.m,

10H), 3.1 (s, 1H), 9.3 (d, J=4Hz , 1H), 9.7 (s, 1H).

IR (film): 2960s, 2920s, 2880s, 2700w, 1720s, 1450m,

1380w, 1370m, 1250w, 1220w, 1180m, lllOw, 1050w, 1030w,

930w, 840w, 780w, 750w, 700w cm-1. Mass spec. Calc, for CnHzoO 168.151, obs. 168.150. M/e 168 (17.8),

139 (35.7 ), 112 (100).

t 208

Preparation of 5-Methyl-2-(l-methylethyl)cyclohexylidene

Methyl Enol Ether 32

The P-hydroxy-a-methoxy silane 30 (400 mg, 1.h7 mmol) < W S f was heated at 60°C for 1 h in dry THF (10 ml) and (440 mg,

1.5 eq) of KH (23-5 % in oil) (88 mg KH). The orange solution was cooled then poured into saturated NaHC03

(20 ml) and extracted with diethyl ether (3X30 ml). Drying

(MgSOi,) and evaporation gave an oil that was distilled at 0.4 mm to give 230 mg (86$) of a clear colorless liquid (bpt. < 95°C). NMR (CC1*) 61 0.8 (d, 3H) J=3Hz,

0.9 (d, 6H) J=3Hz, 1.2 (s, 2H), 1.5-1.9 (br.m, 5H),

2.1-2.4 (br.m, 2H), 3.5 (s, 3H), 5.7 (s, 1H). IR (film):

2940s, 2920s, 2860s, 1670m, 1460s, 1450s, 1370m, 1360m,

1280w, 1250m, 1230m, 1220m, 1200s, ll80w, 1120s, 1110s,

1060m, 1050m, 1010m, 980w, 960w, 890w, 870w, 830s,

750w cm-1. Mass spec. Calc, for C12H 22O 182.167, obs.

182.167. M/e: 183 (2.5), 182 (10.5), 177 (2.8), 155

(2.6), 152 (2.6), 149 (2.6), 147 (2.6), 140 (10.5),

139 (100), 109 (23.7), 107 (36.8). Starting material was evident as impurity 272, 258, 229 (less than 1$),

228 (1.8), 227 (10.5). This reaction did not go to completion at 60°C for 1 h. Other authors21 claim a hindered adduct like 30 underwent clean reaction at 0°C for 0.5 h. We have observed that none of our adducts give the vinyl ether with KH in THF at 0°C for 0.5 h.

* 209

Preparation of (cis and trans)-5-Methy1-2-(1-methylethyl)

1-methoxymethyl Cyclohexanol 33

The adduct 30 (200 mg, 0.74 mmol) was stirred under

argon at 70°C for 24 h with cesium fluoride (123 mg,

1.1 eq) and dimethyl sulfoxide (3 ml). The contents

were poured into water (20 ml) and extracted with

diethyl ether (4X10 ml). Drying (MgSCU) and evaporation

gave an oil 100 mg (67#) which was contaminated by

DMSO (NMR singlet at 6 2.5). The oil was essentially pure 33 with none of 32 detected in the NMR spectrum.

This should be compared to the attempts to prepare 12

and 17 at 80°C bath temperature. This specific reaction

did not work at 60°C but went to completion at 70°C.

NMR (CClu) S: 0.6 (s, 3H), 0.8 (d, 6h ), 1.0-1.8

(br.m, 9H), 3.0 (d, 2H J=2Hz, possibly a doublet because

of an epimeric mixture), 3.1 (s, 3H). IR (film) 3400s,

2940s, 2920s, 2860s, 1460m, 1450m, 1430m, l400w, 1370w,

1360w, 1250w, 1190m, ll60m, 1110s, 1050s, 1030s, 970w,

950m, 910w, 880w, 840w, 800w, 750w, 700w, cm-1. Mass

spec. calc, for Ci2H2tt02 200.177, obs. 200.177. M/e 200

(2.7), 182 (5.4), 156 (10.8), 155 (100), 139 (54), 137

(40.5). 210

Preparation of threo and erythro 1-Hydroxy-l-pyrenyl-

2-methoxy-2-trimethylsilyl Ethane 35

To a dry flask under argon at -30°C was added dry

THF (20 ml from LAH) and 2.1 ml, 13.37 mmol, 3 eq) of methoxymethyltrimethylsilane. Slowly s-BuLi (10 ml,

3 eq) (1.3 M) was added. The contents were stirred to

-15°C over 1 h and then recoiled to -30°C. Slowly (1 g,

4.3^ mmol, 1 eq) of 1-pyrenecarboxaldehyde (in 5 ml of dry THF) was added. As the green solution of starting material adds to the stirred anion a deep orange-red

color developed and then stirred at -30°C for 0.5 h, then warmed to R.T. over 1 h. The contents were poured into saturated NH4C1 (100 ml) and extracted with ethyl acetate

(3X30 ml). Drying (MgS04) and evaporation to 0.5 mm gave a crude oil that was purified by PLC to give two polar products (55%) that would not crystallize. NMR

(CDCI3) 6 : 0.1 (s, 9H), 1.1 (br.d, 1H), 3-5 (s, 3H),

3.8 (br.m, 1.H), 6.2 (d, 1H) J=6Hz, 8 .0-8.6 (br.m, 9H).

IR (film): 3400s, 3040m, 2960s, 2920s, 2890s, 2800w,

1600m, 1590m, 1580m, 1450m, 1410m, 1370m, 1250s, ll80m,

1080s , 1060s , 1040s, 920s, 860s, 840s, 750m, 720m, cm-1.

Mass spec. Calc, for C22H 24Si02 348.154, Obs. 348.154.

M/e 349 (2.3), 348 (8), 332 (2), 331 (2), 330 (4.6),

317 (8), 316 (25 .8 ), 303 (7.5), 301 (4), 286 (2 ), 285

(5.3), 259 (4.6), 258 (5.3), 244 (7.3), 232 (19.3), 231

(100), 230 (25.8), 215 (22.5), 203 (22.5), 202 (22.5). 211

Oxidation of 35. Preparation of a-Methoxyacetyl-2- pyrene 36

The 3-hydroxy-3-pyrenyl methoxymethyltrimethylsilane

(130 mg, 0.37 mmol) was dissolved in dry dichloromethane

(5 ml from K2C03) at 0°C under argon and stirred. PCC

(120 mg, 0.56 mmol, 1.5 eq) was added under a blanket or argon and stirring was continued to R.T. over 1 h.

The black-grey solution was diluted with ether (25 ml) and filtered through a bed of Celite (3X). The ether layers were dried (MgSO.,) and evaporated to give the crude silylated a-methoxy ketone. IR (film): moderate on 3*100, 3040m, 2920s, 2850s, 1690m, 1675s, l620w, 1590m,

1580w, 1500w, 1460m, 1450m, 1370m, 1260s, 1230w, 1200w,

1150w, 1060w, 900w, 850s, 820w , 780w , 740s, 710s, cm-1.

The crude material was taken up in ether and eluted down a column of silica gel (nominal Grade 923) with ether and concentrated to give the pure a-methoxy ketone

(48 mg, 47$) mpt. - 45°C as one red spot on TLC.

NMR (CDCla) 6 : 3.70 (s, 3H), 4.35 (s, 2H), 7.9-8.3

(m, 9H). IR (film): no OH absorption, 3040w, 2920s,

2840m, 1670s, weak 1720, 1590m, 1580m, 1500w, 1460m,

1370m, 1240m, 1200m, ll80w, 1060w, 900m, 870w, 840s,

820m, 780w, 770w, 750m, 710w, cm-1. Mass spec. Calc, for

Ci9H 11*0 2 274.099, obs. 274.100. M/e 274 (11.2), 260 (4.5),

231 (31.5), 230 (100), 229 (58), 216 (7.8), 215 (31.5),

203 (12.3 ), 202 (47.3), 201 (6 3 ), 200 (36 .8). 212

Preparation of 2-Pyrenenyl Methyl Enol Ether 37

To a flask under argon was added 150 mg of 23.5%

KH in oil (ca. 35 mg of KH, 0.86 mmol, 3 eq). To the

slurry in dry THP (5 ml) was added (100 mg, 0.28 mmol,

1 eq) of the 3-hydroxy-a-methoxysilane in dry THF (5 ml).

The contents were heated at 60°C overnight and then

carefully quenched with saturated NaHC03 (20 ml) and extracted with ethyl acetate (3X10 ml). Drying (MgSO*) and evaporation gave a liquid that was purified by PLC.

The yellow liquid obtained 78 mg, (70%). NMR (CDC13)

<5* 3.9 (s, 3H) AB quartet centered at 7.05 (2H J=l4Hz,

8.0-8.5 (br.m, 9H). IR (film): 3040s, 2940m, 2920s,

2850s, l630s, l600w, 1580w, l480w, 1460m, I4l0w, 1330m,

1310w, 1250w, 1240w, 1220s, 1210s, ll80w, 1160m, 1150s,

1130m, 1090w, 1050w, lOOOw, 930m, 910m, 840s, 820w, 800w,

780w, 740m, 710s, cm-1. Mass spec. Calc, for CisHnO

258.104, Obs. 258.104. M/e 259 (18.7), 258 (100), 244

(12.5), 243 (56.2), 242 (12.5), 216 (18.7), 215 (81.2 ),

213 (25), 205 (31.2), 202 (69), 149 (100), 145 (69).

Also seen in the mass spec. 273 (1.4), 272 (6.2). 213

Attempts to Isolate 1-Pyrenyl Acetaldehyde 38

The same problems that were encountered with the

isolation of ^8 in Part I were seen here. No hydrolysis

procedure was adequate for converting 35 to 38. Only

tars were obtained that refused to solidify in a variety

of solvents. While no problems were encountered with the preparation of the homologous methyl ketone (6^, Part I)

it seems that the aldehyde 38 (or ^8, Part I) can not be prepared by our methods. The IR and NMR data did not give evidence for the formation of the aldehyde. The only proposed explanation is that this labile aldehyde undergoes condensation with itself under the acidic conditions employed. Subsequent dehydration provides product that can add water in Michael fashion and further dehydrate to an easily polymerisable substance

(tars ). 214

Reaction of 1 with the (1,9) and (1,10) Diones 39

The cyclophane (200 mg) was obtained from A. deMeijere and not further purified. To a dry flask under argon at

-20°C was added dry THF (3 ml from LAH) and methoxymethyltrimethylsilane (0.15 ml, 0.847 mmol, 4 eq). Slowly sec-butyllithium (0.63 ml, 0.889 mmol, 4 eq) (1.4 M in cyclohexane) was added. After stirring for 0.5 h, at -20°C the cyclophane (45 mg, 0.19 mmol, 1.0 eq) was added as a solid. An immediate green color developed and then changed to amber red after total addition. Stirring to

R.T. over 1 h, gave a red solution which was poured into saturated NHUC1 (20 ml) and extracted with chloroform

(2X20 ml). Drying (MgSOn) and evaporation gave a white solid 70 mg, (78$) with no starting material by IR data.

IR (Nujol): 3480br.m, 3l40br.m, 3040 spike, 2920s,

2860s, 1675s, 1605s, 1570w, 1510m, 1460s, 1415s, 1380m,

1280s, 1250m, 1185m, lllOw, 1.020w, 990w, 845s, cm"1.

The mass spec, for this compound was very complex. No

M+-H20 or M+-2H20 was observed and the only discernible peaks in the spectrum gave M/e 427, 412, 409, 408, 397,

396, 380 (all less than 1$), 368 (1.2), 355, 353, 339

(less than 1%), 338 (1), 324 (1), 323 (1.6), 311, 309

(less than 1$), 295 (1.6), 279, 271 (less than 1$), 265

(1.6), 255 (11.5), 254 (100). No MW 244 for starting material, mpt. 170°-175°C decomp. Si

215

Hydrolysis of 40. Attempts to Prepare the bis-Aldehyde 4l

Hydrolysis of 40 (20 mg) with formic acid at R.T.

gave a semisolid that could not be crystallized for a melting point. NMR (CDC13) 6: 1.1 (br.s), 2.5 (br.s),

2.8-(br.s), 7.3 (br.m.) very small silicon residues were evident. IR (film): broad OH still present, 2980s,

2920s, 2860s, 1730s, 1680s, 1610s, 1520m, 1460s, l420w,

1370w, 1270m, 1190m, 1050m, 890m, 850w, 800w cm-1. The tar could not be crystallized. Other procedures with mineral acids (CHI, HC1CU and H 2S04) on 10 mg did not give IR or

NMR data to support an aldehyde product.

The NMR data of these compounds was very peculiar in that a large singlet at 6 1.1 was always present. The only reason that 4j3-4j5 are proposed as possible products is the IR data which gave maximum absorptions for all C=0 belonging to the known compounds. No aldehyde moiety could be inferred from the IR, NMR or mass spectral data. 216

?v:c attempts to prepare the vinyl ethei" 42 were dene.

Treatment of 4 0 with KH in THF at reflux gave a product as an oil. This product was contaminated with mineral oil so no accurate spectral data was obtained. The KH had been washed several times with olefin-free dry pentane but still oil was seen in the product. The second attempt was a treatment of 40 with cesium fluoride in DMSO at 80°C.

No product could be isolated to provide data in agreement with the structure 42. The strain present within the para-cyclophane skeleton would permit a variety of reaction pathways to occur. If this desilylation had gone as planned the e n d ether as product •■ould be extremely labile. Once again, this was disappointing in light of the fact that so many of these reactions proceeded well on other substrates. At this point, it is certain that the bis aldehyde 41 remains as an elusive synthetic target by our methods. 217

Preparation of Tetrahydropyranyloxytrimethylsilyl Methane 52

Trimethylsilylmethanol (3 g, 28.8 mmol, 1 eq) was dissolved In diethyl ether (5 ml) and 5.2 ml, 60.6 mmol,

3 eq) of dihydropyran under argon. (100 mg) of para- toluenesulfonic acid was added. The contents of the flask was heated for 20 h, at 60°C, and then distilled at atmospheric pressure. After excess DHP (bpt. 85°C) and trimethylsilylmethanol (67°C) were removed, distillation was continued at house vacuum pressure (ca. 10 mm) to give a clear, colorless liquid (bpt. 80°-85°C). NMR

(CDC13) 6: 0.1 (s, 9H), 1.4-1.8 (br.s, 6H), 2.1 (s,

1H), 2.8 (s, 1H), 3.5 (s, 2H), 3.6-3.8 (m, 1H). M+-15

Calc, for C0HiuSiO2 173.100, Found. 173.100. M/e 188

(too small to mass measure, < 1%), 174 (2.2), 173 (12.5)}

157 (2.1), 156 (2.1), 155 (< 1), 145 (2.1), 143 (1),

131 (2.4), 130 (1), 129 (2.1), 117 (2.7), 116 (1.6),

115 (10), 98 (7 .5 ), 86 (100). 218

Lithiation of Tetrahydropyranyloxy-trimethylsilyl

Methane 52

Only 52 when treated with 1.25 eq of sec-BuLi

(_25°c to -15°C then recooling to -25°C) in THF gave indication of that a nucleophilic species had been formed. The electrophile cyclohexanone (1.0 eq) was added and the usual workup gave a semi-solid (85%) of diastereomeric products. NMR (CDC13)S0.0 (d, J=3Hz) overall 9H, 1.0-2.0 (br.s, ca. 16H), 2.2 (br.s, ca.

1H), 2.9 (t, J=6H z , 2H) , 3.5 (br.t, J=6Hz, < 2H),

4.4 (br.s, ca. 1H) possibly the OH signal. IR (crude film): appreciable amount of ketone remained (1710 cm"1), 3420s, 2920s, 2840s, 2710 very weak, 2640w,

1710s, 1610m (in situ eliminations), 1440s, 1380m,

1350m, 1310w, 1250s, 1240s, 1200m, 1150m, 1120s, 1110s,

1070s, 1040m, 1020s, 990w, 960s, 930w, 900m, 890w,

860s, 830s, 8l0w, 790w, 750m cm-1. No mass measurement was taken.

Another reaction of the lithio species of 52 with adamantanone 2 (non-enolizable ketone) gave a cleaner reaction by IR and NMR data. When 1.25 eq of the reagent

1 was quenched with adamantanone the crude semi-solid showed a fast running spot was the major product by TLC.

Two minor products (one UV active and one not) were also 219 evident. No purification by PLC was done due to the lability of the adducts. NMR (crude/CDCl3) 6: 0.1

(d, J=2Hz, 1:1 ratio) overall less than 9H, 1.8

(br.s, 20H), 2.3 (br.s, residual adamantanone), 2.5

(br.s, 2H), 2.7 (br.m, < 1H), 3.6 (weak t, J=6Hz),

4.1 (s), 4.3 (s) overall ca. 1H), 4.5 (br.s, ~ 1H).

IR 3540 sharp, 3420brs, 2900s, 2840s, 2710w, 2640m, * 1710m, l660w, I6l0w, 1440s, 1430m, 1380m, 1340m,

1240s, 1200m, 1160m, i070m, 1020m, 970w, 950w, 920w,

890w, 850s, 830s, 800w, 730m cm” 1. No mass spec, was calculated for Ci9H3USi03 (M+ 338). M/e observed

258 (2.3), 222 (5.7), 205 (3), 204 (12.5), 177 (1.5),

176 (10), 175 (3), l6l (7.5), 152 (42.5), 151 (100),

150 (55), 135 (10), 134 (47.5), 133 (10). 220

Preparation of 1-Ethoxy Ethyloxytrimethylsilylmethane 53

To a flask under argon at 0°C was added trimethyl

silylmethanol (1.05 ml, 10 mmol, 1 eq) and dichloromethane (25 ml from K2C03) and N,N-dimethylaniline. At

0°C a-chloroethyl ethyl ether (1.7 ml, 15 mmol, 1.5 eq) was added and the reaction was followed by IR and TLC.

At R.T. over 1 h a nonpolar product appeared on TLC.

IR data showed the loss of the OH absorption. The con tents were poured into cold IN HC1 (30 ml) and extracted with diethyl ether (3X30 ml). Drying (Na2S04) and evaporation gave a liquid that was distilled at 15 mm to give a clear liquid (bpt. 88°C). M/e 268, 252, 217.

None of the desired product CBH2OSi02 (MW 176) or M+-15 was observed in the mass spectrum. NMR (CDC13)

0.1 (s, 9H), 1.25 (t, 3H) J=6Hz , 3.5 (d, 3H) J=8H z no quartet was observed for the methylene next to methyl or the hydrogen next to oxygen. 3.6 (s, 2H), 3.7

(s, 3H). IR (film): No OH absorption, 2960s, 2920s,

2860m, 2800w, 1740w (unaccounted for), l690m, 1600s,

1500s (N,N-dimethylaniline), 1460m, 1370w, 1360w, 1340m,

1260m, 1250s, 1230w, ll’90w, 1170w, 1060m, 1030m, 990w,

940w, 860s, 840s, 800w, 750s cm"1. This material was treated with s-BuLi at -78°C to -20°C, then quenched with cyclohexanone. No desired addition products were observed. M/el 269 (4.7), 278 (21.3), 256 (4.7), 229 221

(2.4), 228 (19), 21 (7), 221 (38), 202 (9.5),.177 (9.5),

149 (54.5), 127 (23.7), 212 (54.5), 120 (100), 110 (81.8),

107 (72.7 ), 106 (100).

Trimethylsilyloxy-trimethylsilyl Methane £4

The ether 54 could not be isolated in pure form.

40°C. No distillable liquid of the desired structure was obtained. The chlorotrimethylsilane and chloro-^ butyl- dimethylsilane (electrophiles) were previously distilled to clear (neutral) liquids over molecular sieves. In closing this topic for now, it still seems this area deserves more experimentation. Even though the subse quent desired lithiation may result in Wittig rearrange ment to a non-nucleophilic species (i.e. basic) this concept should be pursued further. 222

Preparation of the Allylic Alcohol, Perilla Alcohol 55

To a dry flask at 0°C under argon was added (2.5 g,

16.6 mmol, 1 eq) of perilla aldehyde (FD & 0) and dry methanol (30 ml from Na). Slowly (1.26 g, 33 mmol, 2 eq) of sodium borohydride was added. Evolution of gas ceased after 1 h at 0°C and the contents were poured into saturated NHUC1 (100 ml) and extracted with diethyl ether

(3X30 ml). Drying (MgSOu) and rotary evaporation gave

2.55 g (1002) of a liquid which was distilled at 0.1 mm,

(bpt. 61.5°-62.5°C) 2.1 g (80$). Mass spec. Calc, for

C ioH i60 152.120, Obs. 152.120. 223

Preparation of 4-(1-Methylethenyl)-l-cyclohexene-l-

carbinoltrimethylsilyl Methyl Ether 56

To a dry flask under argon was added the allylic

alcohol (1.8 g, 118 mmol, 1 eq) and dry THP (12 ml from

LAH) and sodium hydride (370 mg, 15 mmol, 1.2 eq). The

contents were refluxed at 55°C for 2 h. Iodomethyltri- methylsilane (3.2 g, 15 mmol, 1.2 eq) was added and the

contents were stirred at reflux (55°C) overnight with

adequate cold water cooling. By TLC two major spots developed and the allylic alcohol was totally consumed.

The contents were poured into saturated NaHC03 (100 ml) and extracted with diethyl ether (2X30 ml). Drying

(MgSOi*) and rotary evaporation gave a liquid 2.86 g

(100%). PLC isolation of the two fast Rf products gave

M+ 238 Rf, and M+ 166 Rf 2. The fastest spot was Rfi identified as the desired major product. NMR (CDC13) <5.'

0.1 (s, 9H), 1.0 (s, 2H), 1.8 (s, 3H), 2.0-2.2 (br.s,

7H), 4.0 (t, J=10Hz, 2H), 4.7 (s, 2H), 5-7 (s, 1H).

IR (film) 3080w, 2960s, 2920s, 2840s, 1640m, 1450m,

1430m, 1370m, 1360m, 1250s, 1190w, 1150m, 1070s, 1020w,

960w, 920w , 880m, 860s, 840s, 780m, 700w cm”1. Mass spec. Calc, for Ci*HasSiO 238.175, Obs. 238.176.

M/e 239 (2.2), 238 (15.4), 224 (9-9), 223 (4.5), 209

(12.7), 195 (10.4), 171 (9.9), 169 (9), 155 (9.9), 135

(25.6 ), 134 (71.7), 133 (17.9), 119 (58.9), 103 (56.4),

93 (35.8), 91 (35.8), 89 (35.8), 87 (35.8), 75 (100), 224

73 (100). The-product Rf2 was identified as CnHiaO

Mass spec. Calc. 166.136, Obs. 166.136 and the methoxyl

signal in the NMR 5 3.3 (s, 3H).

Attempts to Prepare 56a by Lithiation of 56

Treatment of 56 (100 mg) with s-BuLi (1.1 eq) identical to the preparation of 1 gave after the usual workup at

R.T. an oil (83 mg, 83$) of lower Rf than 56 on TLC. which was not the desired product 58. Mass spec. •vw for CiuH86SiO M/e 250, 238, 235, 223, 208, 195, 190,

164, 151. Treatment of 56 with n-BuLi at -78°C to -20°C »N*N

Attempts to Prepare 57 from the Intramolecular (Fluoride

Induced) Rearrangement of 56

Treatment of 56 (250 mg) with cesium fluoride in dimethylsulfoxide at 85°-l40°C gave a liquid Identical to 55 by TLC and NMR data. M/e 152, 135, 121, 107, 93.

IR (film)I intense OH absorption (3^50 cm” 1) and loss of silicon (1250, 850 cm"1). NMR (CDC13) did not give evidence for the extra olefinic signal. Treatment of

56 with KF»2H2o in DMSO at 110°C gave only recovered starting material. 225 The Addition of 60 to Cyclohexenone. Preparation of

3*(l-Methoxy-l-trimethylsilyl)methyl Cyclohexan-l-one 62

To a dry double flask addition apparatus under argon was added to the top flask at -30°C dry THP

(10 ml) and (1.2 ml, 7.8 mmol, 1.5 eq) of methoxymethyltrimethylsilane. Slowly s-BuLi (6.5 ml, 1.5 eq) (1.2 M) was added dropwise and the contents were stirred to

-15°C over 1 h, then recooled to -30°C. The bottom flask was prepared as follows. Me2SCuBr (600 mg,

2.9 mmoo, 0.37 eq, 25 mole $), dimethyl sulfide (5 ml) and ether (7 ml) were combined and cooled to -78°C, then the upper flask was emptied slowly Into the lower flask with rapid stirring. The green solution that developed was stirred at -78°C for 40 min and then distilled cyclohexenone (0.5 ml, 5.2 mmol, 1.0 eq) was added. Upon warming to -20°C over 1.5 h a red solution developed. At 0°C the contents were poured into saturated NH4C1 (30 ml) and extracted with ethyl acetate (3X30 ml). Drying (MgS04) and rotary evaporation gave a crude liquid (1.2 g). PLC isolation of two polar products gave 520 mg (46$). Another identical reaction gave crude product that was distilled at 0.17 mm to give 1.4 g (60$) of a liquid (bpt. 120°C). NMR (CDC13)

5: 0.1 (s, 9H), 1.1-1.4 (d of d,J=4Hz, 1H), 1.6-2.4

(br.m, 9H), 3-3 (d, J=2Hz, 3H). IR (film): 3450m, 226

2960s, 2920s, 2860m, 2820w, 1710s, 1450m, l430w, 1350w,

1320w, 1250s, 1230w, 1170w, ll60w, 1090m, 840s, 750w,

700w. Mass spec. Calc, for CiOHi9Si02 (M+-15) 199.115, obs. 199.116. No M+ for 214 observed. M/e 199 (64.3),

185 (7), 169 (100).

Reduction of 62 to 63 and Attempts to Prepare 64 or 65

The methoxymethyltrimethylsilyl ketone (130 mg,

0.6 mmol) was stirred under argon at 0°C in methanol

(5 ml). Sodium borohydride (35 mg, 0.9 mmol) was added and the contents were stirred to R.T. over 1 h, then poured in saturated NH<*C1 (10 ml) and extracted with ethyl acetate (3X20 ml). Drying (Na2S0«) and evaporation gave a viscous liquid (130 mg, 98%) that nearly solidified at 0°C. NMR (CDCls) 6 0.1 (d, 9H), 1.0-2.0

(br.m, < 9H), 3.1 (d, 1H) J=4Hz, 3.4 (s, 3H), 3-5 (br.s,

1H), 4.2 (br.s, 1H). IR (film): 3400m, 2920s, 2840m,

1450m, l400w, 1330w, 1250s, 1070m, 1030w, 960w, 910s,

850w, 840s, 730s. No accurate mass measurement of

CiiH2«Si02 (MW 216) was obtained for this labile inter mediate. The ketone 62 did not desilylate with fluoride ion so this intermediate 63 was treated with cesium fluoride in DMSO at 120°C. 227

Preparation of l-methoxy-l-trimethylsilyl-3-butene

To a dry flask under argon at -20°C was added methoxymethyltrimethylsilane (2.66 ml, 16.9 mmol, 1.2 eq) and dry THP (20 ml from LAH). Slowly (10 ml,

14 mmol, 1.1 eq) of sec-butyllithium (1.4 M) was added.

After stirring at -15°C for 0.5 h, the contents of the flask were quenched with (3.7 ml, 3 eq) of distilled allyl bromide at -15°C. The contents became white and were stirred at R.T. over 1 h, then they were poured into saturated NHuCl (20 ml) and extracted with diethyl ether (3X20 ml). After drying (MgSOi,) and distillation the reagent was obtained as a clear liquid 1.55 g (50%) bpt. 125°C.

The distilled reagent was treated with s-BuLi in

THF at -20°C and stirred with hopes of forming the allyl substituted methoxymethyltrimethyl silane carbanion.

No products were observed that indicated that the desired adduct had been formed. If this adduct had been prepared it would have illustrated the use of another new annellation reagent. PART III

INTRODUCTION

HMPA (hexamethylphosphoroustriamide) is a useful solvent of high polarity and low nucleophilic character in alkyllithium reactions with electrophiles. Its use in organic reactions has been mainly as a ligand in carbanion formations. Few examples on the synthetic use of HMPA as a reagent rather than merely a solvent are known.

Eliel1 has deprotonated it with n-BuLi, s-BuLi, and PhLi when 2.'1 with THF at -30°C over many hours to give the homologous N-methyl amines in 50-75$ yields.

Me I z)z *RH + LiCH2NP(NMe2 )2 It

Li I * RCHaNHMe H'

At that time Leroux2 reported addition of a phenyl substituted lithio phosphoramide to carbonyl compounds as an illustration of the balance between high nucleophilicity and low basic character of their reagent.

While studying the reactivity of the P-N bond it was found that I and II behaved quite differently when treated with

228 229

RLi. In THP at -70°C either n-BuLi or MeLi gave two compounds plus starting material.

(EtO)2PNMeCH2Ph . N(Me2)2PNMeCH2Ph 8 8 i ii BuLi > (EtO)2PNMeCHPh + BuH ----* '6 Li1

MeN=CHPh + (EtO)2PLi II III 0

I + MeNHCHPh + III + (EtO)2PH I II Bu 0

This is comparable to the basic arylsulfonamides already known.3 Leroux added the anion to carbonyls and workup with water gave reasonable yields of adducts shown below. These results come from reactivity of the anion stabilized by the phenyl group. Unsubstituted HMPA anions are unknown.

0 0 (Me2N)2PNMeCHPh 75-80# (Me 2 N)2 PNMe CHPh 40-80# I I HCOH RCOH I I R R

Also noteworthy is the finding that nucleophilic aminomethylation can be effected by the use of the appropriate "masking group".11 For this purpose, the diphenylmethylidine group has been used to advantage as 230 evidenced by the modernteyields of isolated aminomethylated products from reactions such as the one shown below.

(C6H s )2C=NCH3 .+ (i-C3H7)2NLi ^££^ .Et2? -> (C6HB ) gC=NCH2Li

(C6H5)2C=NCH2Li + (C6H5)2C 0 --» (C6H5)2C=NCH2C(OH)(C6H5)2 (42%)

aq. I HCI H2NCH2C(0H)(C6H5)2 (7658)

A nucleophilic methylaminomethylation has been realized indirectly.12

?' (CH3)NCH2Li + CeHsCHO-S^2-^ (CH3 )2N—N- CH2CH(0H)C6H5 +

CuCl/pyridine >'

H H 0+ CH3NCH2CH(0H)C6H5 — C6H5'

N CH, 3 Although not well documented, it is logical that, in addition to enjoying comparable reactivities, the heteroatom-substituted organometallic compounds and the

"classical organometallic compounds" suffer similar occasional deficiencies. Most important in this respect is the tendency of organometallic compounds to function as bases in effecting unwanted proton removal at the expense of the desired nucleophilic reaction. Peterson has also prepared lithio-trimethylamine and added it to aldehydes in good yield.13

Me2NCH2Li + PhCHO - ^ - > M e 2N-/s^ ^ Ph

OH RESULTS

A large number of silicon containing ylides and organometallics have been described over many years in which electron delocalization into silicon 3d orbitals may be important.4 To prepare derivatives such as these without second row element stabilization is difficult.

The LiCH2OMe reagent has been shown by Peterson to be a poor synthetic tool. As of yet no one has prepared

LiCH(OMe)2 , or the lithio carbanion of the commonly used solvent methylal. Efforts to deprotonate this solvent under many conditions were unsuccessful. However, one unprecedented observation was made. In the reaction of methylal with s-BuLi and HMPA in THF at -70°C solutions of a-lithiohexamethylphosphoroustriamide 1 were formed over 1 h as judged by the results.’ No evidence for the a-lithio-a,a-dimethoxymethane was found.

Quenching the above solutions of 1 with the listed carbonyl compounds gave the 111 adducts 2, 3, 4 and 5. a-Lithioalkylphosphoramides have been implicated as inter mediates in reaction involving strong bases and alkyl- phosphoramides, but adducts of the type 2-5 have not been observed.6 Treatment of 2 with 3N hydrochloric acid or

232 233

0 SBuLi || (M82 N)3 P = 0 ------LiCHaNPCNMflpJo — CH2- NMe CH2(0Me)2 ^ + LiOP(NMe2 )2 1

OH o & (80%) mp. I09-II0.5°C. 2

« 0 ^ S ^ C H O (83%)mp. 30°C. • (J^P(NM%’23

OH q a° (80%)b.p. l45°/05mm. 4

°H ~ 0

(50%)mp. 140- I42°C cor <0Xj !»' p m ‘^ 5

I 234

3N sulfuric acid in methanol,7 heating to reflux, and quenching with sodium carbonate gave the cyclic urethane

6, mp. 191°-192.5°C. Surprisingly when 5 was subjected to these conditions only piperonal was isolated.

While methods have been developed for the preparation of a-lithiomethyleneamino compounds8 none is based upon the deprotonation of hexamethylphosphoroustriamide. It should be added as a caution that presumably 1 might be present in reactions involving other alkyllithiums, hexamethylphosphoramide and electrophilesj and as can be seen from our results, hexamethylphosphoramide is more than a solvent, it can be a reagent.

During the course of isolation and characterization of these HMPA adducts it was thought that TMEDA had been lithiated. This observation has been made during another transmetallation.1u Tributyltinlithium, an organometallic compound considerably less reactive than n-BuLi failed to react with the diamino tin compound at R.T. in THP to give lithiated TMEDA.

The far fetched idea to prepare LiCH(OMe)2 in methylal was indeed not well though out since MeOCH2Li is more easily prepared in hydrocarbon solvents rather than in ethers where they have only limited stability.12 Methylal, however, is a useful solvent for alkoxymethyl Grignards.15 235

Several attempts to cleave the N-P bond were

unsuccessful. Owing to the water solubility of the

expected amino alcohols as products, the adamantyl system

was used so as to increase their solubility in organic

solvents. Treatment of 2 with KH in THF overnight gave

no products in the organic extracts. Mild acidification

of the aqueous layer with hydrochloric acid also gave no

products then neutralization with 20# sodium hydroxide

gave a solid when extracted with ethyl acetate. This

product was identical to a reaction of 2 with aqueous

sodium hydroxide and formaldehyde and was identified as

7. Treatment of 2 with sodium metal in liquid ammonia

in THF and t-Butyl alcohol gave only starting material.

Heating of 2 in xylene at l40°C for 3 h gave some

starting material and an appreciable amount of 7* Only

the starting material was isolated from treatment of 2

with LAH in THF at reflux (60°C). Attempts to silylate

2 gave only very high molecular weight products of no

discernible structure. The spiro urethane 6 may have

pharmacological importance since compounds such as these

have been studied.9 Shanzer has prepared substituted

2-oxa-zolidones via a-amino-3-hydroxy acids.10 c o 2h i)LDA (TMS)2 NCH,CG,TMS •+ R2co EtOH 236

Na no rxn. n h 3

U\H 11 A >S. THF

A S.M . + 7

O H Me

^ N M e ,

N a O H ? ^ > e z 37/. C H j O

DKH ' II 2SHCI KOH

MeOH i) HCl

4 0 ’/ . EXPERIMENTAL TO PART III EXPERIMENTAL TO PART III

Preparation of the Hydroxy Phosphoramide of Adamantanone 2

To a dry 50 ml three-necked flask under argon at

-78°C was added dry methylal (35 ml from CaH2) and dry

HMPA (4.3 ml from Mol. Sieves 48). Slowly (17.4 ml,

24.3 mmol, 2 eq) of s-BuLi (1.4 M in cyclohexane) was added and the mixture stirred for 1.5 h. At -78°C

(1.86 g, 12.38 mmol, 1 eq) of solid adamantanone was added whereby the contents of the flask became milk white.

After stirring to R.T. over 1.5 h, the contents were poured into saturated ammonium chloride (100 ml) and extracted with ethyl acetate (2X50 ml). The organic layers were washed with water (2X20 ml) and brine (20 ml).

Drying (MgSOu) and rotary evaporation gave 3.2 g, (80$) of a white solid. Recrystallization from acetone/petrol gave 3 g, of mpt. 109°-110.5°C. NMR (*CDC13) 5: I .85

(br.s, 14H), 2.70 (d, 2H) J=4.5Hz, 2.80 (d, 12H J=4.5Hz,

3.30 (d, 2H) J=6Hz, 5.35 (br.s, 1H). IR (Nujol)*. 3310m,

3240m, 2920s, 2820s, 1450s, l4l0w, 1370w, 1360w, 1325m,

1290s, 1270m, 1225w, 1180m, 1160s, 1130s, 1060s, 1130s,

1060m, 1020w, 980s, 970s, 930w, cm"1. Mass spec. Calc, for Ci«H32N302P 329.223, Obs. 329.224. M/e 330 (1.9),

237 238

329 (5.8), 211 (1.7), 285 (2.4), 284 (2.4), 241 (0.75),

224 (0.33), 190 (0.55), 180 (9.2), 179 (100), 178 (4),

164 (1.4), 151 (4), 150 (9-2), 136 (77), 135 (100).

The Hydroxyphosphoramide of Cyclohexanecarboxaldehyde 3

This procedure was identical to the preparation of 2. At -78°C the substrate (distilled cyclohexane carboxaldehyde, 1.45 ml, 12.9 mmol, 0.5 eq) was added.

After stirring to R.T. over 1.5 h the usual workup gave a pure semisolid of 3.1 g (83$) mpt. ca. 30°C.

IR (film): 3300s, 2980m, 2900s, 2800s, 2650w, l640w,

1450s, 1300s, 1175s, 1200s, 1100m, 1060m, 1040m,

970s, 890m, 860w, 840w, cm-1. NMR (CDC13 ) 6: 1.2-1.65

(br.s, 10H), 2.5 (s, 12H), 2.7 (s, 3H), 3.0 (d, 1H)

J=6Hz, 3.2 (d, 1H) J=6Hz, 3.6 (t, 2H) J=8Hz . Mass spec. Calc, for CI3H 3oN3P02 291.208, Obs. 291.208.

M/e 292 (1.5), 291 (3), 274 (1), 273 (5.5), 247 (2),

230 (1), 209 (2.5), 208 (5-7), 180 (3.7), 179 (69.2),

178 (15.8), 163 (18.7), 151 (1 8 ), 136 (52.5), 135 (100). 239

The Hydroxyphosphoramide of Cyclohexanone 4

This procedure was identical to the preparation of 2. At -78°C distilled cyclohexanone (1.21 ml,

12.2 mmol, 1 eq) was added. After stirring to R.T. over 1.5 h the contents were worked up as usual to give

1.8 g (80$) of a liquid, bpt. l45°C at 0.05 mm.

IR (film): 3350s, 2900s, 2820s, 2780m, l630w, 1440s,

1290s, 1200s, 1060w, 980s, 940m, 830w, 790w, 760m,

730s cm” 1. NMR (CDC13) 61 1.4 (br.s, 10H), 1.8

(br.s, 2H), 2.4 (s, 12H), 2.6 (s, 3H), 3.2 (s, 1H).

Mass spec. Calc. 259-181, Obs. 259.182. M/e (M-H20)

259 (100), 242 (13), 232 (30), 214 (22.5), 199 (17.6),

187 (10), 180 (17.5), 179 (13), 171 (27.5), 154 (22.5),

135 (100), 124 (100). 240

The Oxaphosphoramide of Piperonal 5

To a dry 100 ml three necked flask under argon

at -78°C was added dry methylal (60 ml from CaH2) and

HMPA (8.6 ml) and stirred while slowly (34.8 ml) of

s-BuLi (1.4 M) was added. The contents were stirred

for 1.5 h at -78°C and then piperonal (3.72 g, 24.8

mmol, 1 eq) was added in a shaker tube. The content

became straw colored and were stirred to R.T. over

1.5 h. The usual workup gave 6 g of a thick oil.

Recrystallization from light petroleum ether/acetone

gave 4.6 g (58%) of a white solid, mpt. 140°-142°C.

NMR (CDC)s) 61 1.2 (t, J-6Hz , 1H), 2.30 (s, 3H), 2.55

)d, J=6Hz , 12H), 4.2 (br.s, 1H), 4.9 (br.d, J=6Hz , 2H)

5.9 (s, 2H), 6.8-7.0 (m, 3H). IR (Nujol): 3220s,

2920s, 2850s, l840w, 1750w, l600w, 1500m, 1480m,

1460s, . 1440s, 1370s, 1360m, 1290s, 1240s, 1170s, 1120w,

1090m, 1060m, 1050m, 1030s, 980s, 940m, 920s, 880m,

820m cm-1. Anal, for Ci2Hi7P0ttN2. Calc. 50.70%C,

6.98%H, 9.45%N. Found 50.15%C, 6.71%H, 9.66%N. The

data seems to indicate that the closed form of the product as shown below is actually the predominate product. 0 — -!^0(NMe2) 241

Preparation of 3-Methyl-5,5(2-adamantyl)oxazolldin-

2-one 6

The hydroxyphosphoramide 2 (400 mg, 1.2 mmol) was stirred with methanol (2 ml) and 6n HC1 (10 ml) at

70°C for 1 h under a water cooled condenser. The contents were cooled and neutralized with saturated

Na2C03 (ca. 20 ml) to ca. Ph 7. The mixture was the extracted with EtOAc (2X30 ml) and washed with water

(2X20 ml) and brine (10 ml). Drying (MgSCU) and rotary evaporation gave 105 nig (40$) of a white solid mpt. 191°-192°C. IR (Nujol): 2900s, 2650w, 1755s,

1490s, 1460s, 1440s, 1400m, 1370m, 1340m, 1320w, 1290s,

1270m, 1200w, ll80w, 1110s, 1090s, 1060w, 1030s,

1000s, 920s, 8l0s, 790w, 740s, cm"1. NMR (CDC13) 6:

1.85 (br.s, 14H), 2.25 (s, 1H), 2.65 (s, 1H), 2.95

(s, 3H). Mass spec. Calc, for C13Hi9N02 221.142,

Obs. 221.142. M/e 221 (34), 206 (4), 193 (4), 177

(100), 165 (8), 162 (16), 151 (20), 150 (44), 141-143

(metastable), 136 (28), 134 (28). l3C NMR 6'. 26.3,

26.5, 27-3, 31.0, 32.5, 32.8, 34.2, 34.8, 36.8, 37.2,

37.8, 38.1, 56.5, 137.8, 154.4. 242

Treatment of 2 with Sodium in Liquid Ammonia

To a Birch reduction vessel with overhead stirring was added (30 ml) double distilled ammonia and (200 mg,

0.6 mmol) of the starting material. (5 ml) THP and

(5 ml) t-butanol was added. (200 mg) washed sodium metal was added whereby the blue color lasted 15 min.

(50 mg) Sodium was added again. After 20 min, the contents were poured into methanol (10 ml) and saturated

NHi*Cl (100 ml). Extraction with ethylacetate (3X20 ml) gave after drying (MgS0«) and evaporation 85 mg of a white solid, mpt. 104°-107°C. Sublimed at 90° at

0.1 mm. mpt. 109°-110.5°C. Same as starting material.

Desired product may have been lost in aqueous layers.

M+ 329 IR (Nujol): 3240s, 2900s, 1360w, 1350w,

1330m, 1320m, 1280m, 1190m, 1180m, 1160s, 1090w, 1060m,

1020m, 960s, 940m, 920w, 860w, 760s, 730s cm"1. NMR was identical to 2. 243

Reaction of 2 wit.h LAH

To a dry flask under argon at R.T. was added

(50 mg, 0.15 mmol) of the starting material (recovered) and (3 ml) THF (from LAH). LAH (2 eq) was added to give a vigorous effervescence. The slurry was heated for

2 min to 70°C, then cooled and stirred. The contents were poured into saturated NHi»Cl (20 ml) and extracted with ethyl acetate (2X20 ml) to give after drying

(MgSOu) and evaporation 27 mg, of the starting material, mpt. 105°-108°C. The NMR was identical to 2.

Preparation of 2 +. 7

To a dry flask under argon was added (100 mg) of the phosphoramide and xylene (3 ml). The mixture was stirred and heated to l40°C for 3 h. A clear colorless solution resulted. Rotary evaporation and further evaporation to 0.1 mm gave a white solid of 40 mg mpt. 80°-8l°C. NMR (CDC13) (d6, DMSO) 1.80 (s, 14H),

2.7, 3.9 (d, 6H) J=9Hz, 2.75, 2.9 (d, 3H) J=9Hz, 3.25,

3.45 (d, 2H) J=llHz. Mass spec. Calc, for C1MH23O2N 2P

284.165, obs. 284.166. 244

Preparation of 7 with Dilute Alkali

The phosphoramide (25 mg, 0.076 mmol) was stirred at R.T. with 10$ NaOH (5 ml) and 37$ formaldehyde solution (5 ml). Some change on TLC was noted and

CH20 solution (5 ml) was added. After 0.5 h, the contents were again poured into water (20 ml) and extracted with ethyl acetate (2X20 ml). Drying (MgS0«) and evaporation gave 17 mg (78$) of a white solid mpt. 86°-88°C. Mass spec. Calc, for Cii»H25N202P

M/e 285 (3), 284 (10.5), 179 (100), 151 (21), 150

(18), 149 (21), 136 (71), 135 (100). No M+ 329 or 311 for starting material was seen.

Preparation of 7 with KH in THF

To a dry flask under argon was added 1 g, 2 eq) of 23.5$ KH in oil which was washed with (3X8 ml) dry olefin free pentane (settled and decanted). THF

(10 ml from LAH) was added and (6.4 g, 1.5 mmol, 1 eq) of the starting material. No evolution of gas was noted. Stirred at R.T. overnight, then the contents were poured into water (20 ml) and acidified with sulphuric acid to pH 3. No product was isolated.

Neutralized with 20$ sodium hydroxide and shaken.

Extraction (2X30 ml) gave 80 mg of a white solid.

M/e 285, 284, 221, 205, 190, 177, 175, 151. NMR

(CDCla/ds DMSO) 61 1.8 (s, 14H), 2.5 (s, 6H), 2.8

(s, 2H). REFERENCES TO PART I

1. For reviews see - Washburne, S. j. , J. Organometallic Chem., 83, 155 (1974); Cunico, R. F., Ibid. , 10g, I (T976); Fleming, I., Chem. and Ind., 449 (1975)•

2. Peterson, D. J., J. Org. Chem., 3J, 780 (1968); Chan, T. H., Chang, E. and Vinokur, E., Tet. Lett., 1137 (1970).

3. Carey, F. A. and Court, A. S., J . Org. Chem., 37, 939 (1972); Gilman, H. and Tomasi, R., Ibid., 27, 3647 (1962). ~~

4. Stork, G. and Jung, M. E., J. Amer. Chem. Soc., 96, 3682 (1974).

5. Robbins, C. M. and Whitham, G. H., J. Chem. Soc., Chem. Comm., 697 (1976).

6. Chan, T. H. and Chang, E., J. Org.Chem., 39, 3264 (1974).

7. Carey, F. A. and Hernandez, D., J. Org. Chem., 38, 2670 (1973).

8. Hudrlik, P. F., Hudrlik, A. M., Rona, R. J., Misra, R. N., and Withers, G.P., J. Amer. Chem. Soc., 99, 1993 (1977).

9. Hudrlik, P. F., Peterson, D. and Rona, R. J., J. Org. Chem., 40, 2263 (1975).

10. Stork, G. and Colvin, E., J. Amer. Chem. Soc., 93, 2080 (1971).

11. Whitmore, F. C. and Sommer, L. H . , J. Amer. Chem. Soc., 68, 481, 488, 1083 (1946); Ibid., 70, 2869 TT948) .

245 2 46 12. M.S. Thesis, 1977. While exploring the scope of these reactions some difficulties were noted: a) The s-BuLi of whatever molarity must be alkoxide free since that impurity may deprotect a trimethylsilyloxy alcohol derivative or simply enolize the very carbonyl to which the CTC or MCTC is to add. b) For the phenylacetaldehyde substrates appreciable enolization was noted for reactions conducted at 1.2-2.0 eq strengths of CTC thus accounting for low yields of epoxysilanes. As this was observed the reactions with MCTC were not done. For the substrate with a proximate acetoxy group i the epoxysilane is not formed as it is assumed a mode of acetate exchange occurs. This observation may be a good illustration of the effect an acetate group may have on the efficiency of CTC to form epoxy silanes. At the present time, this appears as the only limitation on the functionality permitted in other portions of the substrate molecule (except CTC is an indiscriminate nucleophile and will add to other ketone functionalities present in the molecule), c) The overall result of the reaction from an approach to nor-3 gave a product that can not be converted to the epoxysilane with sodium hydride in THF and ultimately results in the loss of silicon functionaliity during this treatment. The product ii^ is isolated as the rchlor6-acetate" in the reactions with or without TMEDA. The assumption that the major product is the exchanged acetate can only be proposed at this tiem. The polar products gave infrared data and mass spectral data in agreement with the product shown, d) While acidic conditions tend to react cleanly with epoxysilanes to provide a wide array of functional groups it was found that basic conditions used to prepare the same functional groups do not work cleanly. No one had reported on this behavior of epoxysilanes under basic nucleophilic conditions.

13. See reference 11, note 2.

1*1. a-Chlorotrimethylsilylcarbanion (CTC): Burford, C. S., Cooke, F., Ehlinger, E. and Magnus, P. D., J. Amer. Chem. Soc., 99, *1536 (1977).

15. a-Methyl-a-chlorotrimethylsilylcarbanjon (MCTC): Cooke, F. and Magnus, P. D., J . Chem. Soc., Chem. Comm., 513 (1977). 21)7

-TMS

d p

16 . This term implies that CTC serves as a nucleophilic acylating species (©CHO synthon) where reduction has occurred at the original electrophilic carbonyl group. See - Lever, Jr., 0. W. , Tetrahedron Report, No. 19, 1976, p. 1965.

17. Magnus, P. D. and Roy, G., J. Chem. Soc. Chem. Comm., 297 (1978)*

18. Colvin, E. W . , "Silicon in Organic Synthesis", Chem. Soc. Rev., 15 (1978); Fleming, I., "Organic Silicon Chemistry", Comprehensive Organic Chemistry Vol. 3D, D. H. R. Barton and W. D. Ollis, Eds., Pergamon Press, New York, 1979*

19- a) Lyga, J. W. and Secrist III, J., J. Org. Chem., 44, 2941 (1979); b) Olofson, R. A., Hoskin, D. H. and Lotts, K. D., Tet. Lett., 1677 (1978); c) Barton, T. J. and Hockman, S. K., J. Amer. Chem. Soc., 102, 1584 (1980).

20. Eisch, J. J. and Galle, J. E., J. Organometallic Chem., 121, C10-C14 (1976).

21. Bamford, W. R. and Pant, B. C., J. Chem. Soc, Chem. Comm., 1470 (1967).

22. M.S. Thesis, 1977* The trimethylsilylepoxides were conveniently ilthiated and alkylated.

23. Ehrenstein, M. , J. Org. Chem., g, 435 (1944).

24. Allen, W. M. and Ehrenstein, M., Science, 100, 251 (1944).

25. Barber, G. W. and Ehrenstein, M., Ann., 6 0 3 , 89 (1957); Djerassi, C. D. C., Ehrenstein, M. and Barber, G. W. , Ibid. , 612, 93 (1958).

26. Djerassi, C., Miramontes, L. and Rosenkranz, G., J. Amer. Chem. Soc., ££, 4440 (1953).

27. Djerassi, C., Riniker, R. and Riniker, B., Ibid., 18, 6377 (1956). 249

28. a) Djerassi, C., Miramontes, L., Rosenkranz, G. and Sondheimer, F., J. Amer. Chem. Soc., 7 6 , 4089 (1954); b) Ringold, H. J., Rosenkranz, G., and Sondheimer, F., J. Amer. Chem. Soc., 78, 2477 (1956); c) Mills, J. S., Ringold, H. J. and Djerassi, C., J. Amer. Chem. Soc., 80, 6118 (1958).

29. a) Inhoffen, H. H., Logemann, W., Hohlweg, W. and Serini, A., Ber., 71, 1024 (1938); b) Turner, R. B., Perchman, M., and Park, K. T., J. Amer. Chem. Soc., 1%, 1108 (1957); c) Chapman, J. H., Elks, J., Phillips, G. H. and Nyman, L. J., J. Chem. Soc., 4344 (1956); Birch, A. J. and Smith, H., Quart. Rev., 12, 17 (1958).

30. a) Ehrenstein, M., Chem. Revs., 42, 457 (1948); b) Tuller, W. W. and Hertz, R., J. Clin. Endo. and Metab., 12, 915 (1952); c) Djerassi, C., Iriarte, J. , Romo, J. and Berlins, J., J. Amer. Chem. Soc., 73, 1523 (1951).

31. Sandoval, A., Miramontes, L., Sondheimer, F. and Djerassi, C., J. Amer. Chem. Soc., 75, 4117 (1953).

32. Cram, D. J. and Steinberg, H., J. Amer. Chem. Soc., 5691 (1951).

33. Brown, C. J. and Fartling, A. C., Nature, 164, 915 (1949).

34. Spanagel, E. W. and Carothers, W. A., J. Amer. Chem. Soc. , 935 (1935).

35. Fourneau, E. and Baranger, P. M., Bull. Soc. Chem., Fr., 4g, 1161 (1931); Ruzicka, L., Buijs, J. B. and Stoll, M., Helv. Chim. Acta, ljj, 1220 (1932).

36. Littringhaus, A., Ann., 528, 185 (1973); Ziegler, K. and Luttringhaus, A., Ibid., 511, 1 (1934).

37. Kelly, R., MacDonald, D. M. and Weisner, K., Nature, 1 66, 225 (1950).

38. That this angle is greater than the normal bond angle of 109° is taken as evidence that repulsive forces are thrusting the rings away from one another.

39. Zh. Obshch. Khim. , , 199 (1965). 250

40. Peterson, D. J., Organometal. Chem. Rev., A., 7 295-358 (1972).

41. Cason, L. F. and Brooks, H. G., J. Org. Chem., 19, 1278 (1954).

42. Mulvaney, J. E. and Gardlund, Z. G., J . Org. Chem.,

w30, w " 917 (1965). 43. Buele, G. R., Corrlu, R., Guerin, C. and Spialte, L., J. Amer. Chem. Soc., 92, 7424 (1970)-

44. This material has been prepared many times. Mori, K., Tet., 32, 1979 (1967); Ibid., 32, 1101 (1976) and references cited therein.

45. Posner, G. H., Org. Reactions, 19, 1 (1972).

46. Corey, E. J. and Beames, D. J., J. Amer. Chem. Soc., £4, 7210 (1972).

47. Castro, C. E., Gaughan, E. J. and Owsley, D. C., J. Org. Chem., 31, 4071 (1966).

48. Moerck, R. E., unpublished results from our laboratory.

49- Dryden, Jr., H. L., Webber, G. M. , Burtner, R. R. and Celia, J. A., J. Org. Chem., 26, 3237 (1961).

50. a) Butenandt, A. and Georgeus, C., Z. Physiol. Chem., gl8, 136 (1937)J b) Huffmann, M. N. and Darby, H. H., J. Amer. Chem. Soc., , 150 (1944); c) Huffmann, M. N. and Loft, M. H., J. Amer. Chem. Soc., £9, 1835 (1947); d) Winterstein, D., J. Amer. Chem. Soc., 53, 765 (1937)', e) Weltstein, A., Helv. Chim. Acta., 22, 250 (1939).

51. Gange, D. and Magnus, P., J. Amer. Chem. Soc., 100, 7746 (1978). REFERENCES TO PART II

1. a) Miller, N, E., J. Amer. Chem. Soc., 87, 390, (1965); b) Miller, N. E., Inorg. Chem., 4, 1458, (1964); c) Miller, N. E. and Mathiason, D. R., Ibid., 7, 709 (1968); d) Schmidbaur, H., and Malisch^ W. , Chem. B e r .., 103, 3448 (1970), and previous papers in this series; e) Seyferth, D., and Singh, G., J. Amer. Chem. Soc., 8 j , 4156 (1965); f) Gilman, H., and Tomasi, R. A., J. Org. Chem., 2J, 3647 (1962).

2. a) Peterson, D. J., J. Organometal. Chem.,

3. Carey, F. A. and Court, A. S., J. Org. Chem., 37, 939 (1972).

4. Peterson, D. J., J. Org. Chem., 33, 780 (1968).

5. Corey, E. J. and Kwiat Kowski, G. T., J. Amer. Chem. Soc., 88, 5654 (1966); Pudovik, A. N. and Yastrebova, G. E., Russ. Chem. Rev., 39, 542 (1970).

6. Speier, J. L., J . Amer. Chem. Soc., 70, 4142 (1948); Speier, J. L., Daubert, B. F., and McGregor, R. R., Ibid., * 70, 1117 (1948). 7. Schollkpof, U., in E. Muller, Ed., "Methoden der Organischen Chemie," Vol. 13, Georg Thieme Verlag., Stuttgart, 1970, pp. 87, 253.

8. Hudrlik et al, J. Amer. Chem. Soc., 99, 1993 (1977); Robbins, C. M. and Whitham, G. H., J. Chem. Soc., Chem. Comm., 697 (1976).

9. Still, W. C., J. Amer. Chem. Soc., 100, 1482 (1978).

251 252

10. Evans, D. A., Andrews, G. C. and Buchwalter, B., J. Amer. Chem. Soc., 96 , 5560 (1974); Still, W. C. , and MacDonald, T. L., Ibid., 96, 5561 (1974); Burke, S. D., Tetrahedron Lett., 1285 (1980); Beak, P., et al., J. Org. Chem., 437 4256 (1978).

11. Beak, P. and McKinnie. B. G., J. Amer. Chem. Soc., 99, 5213 (1977); Reitz, D. B., Beak, P., Farney, R. F. and Heimlck, L. S., Ibid., 100, 5428 (1978); Beak, P., McKinnie, B. G., and Reitz, D. B., Tet. Lett., 1839 (1977); Beak, P. and Reitz, D. B., Chem. Rev.3 78 , 275 (1978); Schecker, R., Seebach, D., and Lubosch, W., Helv. Chim. Acta, 6l, 512 (19780.

12. Grob, C. A. and Schiess, P. W., Angew. Chem. Int. Ed., 6, 1 (1967) and ref. 34.

13. Magnus, P. D. and Roy, G., J. Chem. Soc., Chem. Comm., 822 (1979).

14. Corey, E. J. , Titus, M. A. and Das, J. , J. Amer. Chem. Soc., 102, 1742 (1980).

15. The following reagents failed to add satisfactorily to the carobnyl group (a)methoxymethylenetriphenylphosphorane (Wittig, G,, Boll, W. , Kurck, K.-H., Chem. Ber., 95, 2519 (1962)); Lithiomehthoxymethyl- diphenylphosphine oxide (Earnshaw, C., Wallis, C. J., Warren, S., J. Chem. Soc., Chem. Comm., 314 (1977)); Phenylthiomethyllithium (Corey, E. J. , Seebach, D., J. Org. Chem., 31, 4097 (1966)); Methylselenomethyl- lithium in ether; and Dimethylsulfonium methylide.

16. Wissner, A., J. Org. Chem., 44, 4617 (1979); Tet. Lett., 2749 (1978)7

17. Farcasin, D., Synthesis, 6l4 (1972).

18. House, H. 0., Chu, C. Y., Wilkins, J. M. , and Umen, M. J., J. Org. Chem., 40, 1460, 1975.

19. Gorlier, J. P., Hamon, L., Levisalles, J., and Wagnon, J., J. Chem. Soc., Chem. Comm., 88 (1973).

20. Corey, E. J., and Venlateswarlu, A. J. , J. Amer. Chem., 94, 6190 (1972); Chaudhary, S. K., and Hernandez, Tet. Lett., 99 (1979). 253

21. Kende, A. S. and Blacklock, T. J. , Tet. Lett., 1980, 3119.

22. Dalrymple, D. L., Kruger, T. L., and White, W. N. in S. Patai, Ed., The Chemistry of the Ether Linkage, Wiley, New York, 1967, p. 6l8.

23. Cram, D. J., Fundamentals of Carbanion Chemistry, Academic Press, New York, 1965, p. 230.

24. Schollkopf, U. and Eisert, M., Ann. Chem., 664, 76 (1963).

25. Schollkopf, U., in E. Muller, Ed., Methoden der Organischen Chemie, Vol. 13, Pt. 1, Gerog Thieme Verlag, Stuttgart, 1970, pp. 87-253.

26. a) Vig, 0. P., Singh, A. and Raj, I., J. Ind. Chem. Soc., 40, 114 (1963); b) Matteson, D. S., Moody, R. J. and Jesthi, P. K., J. Amer. Chem. Soc., 2LIf 5608 (1975); c) Traas, P. C., Boelens, H. and Takken, H. J., Tet. Lett., 2287 (1976); d) Reutrakul, V. and Kanghae, W., Ibid., 1377 (1977); e) Burford, C., Cooke, F., Ehlinger, E. and Magnus, P., J. Amer. Chem. Soc., 5L9, 4536 (1977); f) Martin, S. F., Synthesis, 633 (1979); g) Magnus, P. and Roy, G., J . Chem. Soc., Chem. Comm., 822 (1979); h) Reich, H. J., Chow, F., and Shah, S. K., J. Amer. Chem. Soc., 101, 6638 (1979). 1

27. Wittig, G., Boll, W. and Kruck, K.-H., Chem.Ber., 95, 25i9 (1962); Pettit, G. R., Green, B., Dunn, 57 L. and Sunder-Plassman, D., J . O r g . Chem., 35, 1385.(1970); Ferwanah, A., Pressler, W. and Reichardt, C., Tet. Lett., 3979 (1973); Earnshaw, C., Wallis, C. J. and Warren, S., J. Amer. Chem. Soc., Chem. Comm., 314 (1977).

28. Peterson, D. J., and Hays, H. R., J. Org. Chem., 30, 1939 (1965); Peterson, D. J., Orgmet. Chem. Sev., A7, 295 (1972), and references cited therein.

29. Corey, E. J. and Tius, M. A., private communication; Aguiar, A. M., Hansen, K. C. and Mague, J. T., J. Org. Chem., 32, 2383 (1967); Clark, P. W., Org. Prep. Proc. Int., 11, 103 (1979). 254

30. Whitesides, G. M . , San Filippo, Jr., J., Casey, C. P., and Panek, E. J., J. Amer. Chem. Soc., 83, 5302, (1967); Whitesides, G. M. , Fischer, Jr., W. F., San Filippo, Jr., J., Bashe, R. W. and House, H. 0., Ibid., gl, 4871 (1969); Smith, J. G., and Wikman, R. T. , Synth. Reac. Inorgomet. Org. Chem., 239 (1974).

31. Wada^- K., Tamura, M. , and Kochi, J., J. Amer. Chem. Soc. , g2, 6615 (1970); Tamura, M., and Kochi, J., J. Organomet. Chem., 42, 205 (1972); Kochi, J. K., Acc. Chem. Res., 7 , 351 (1974).

32. House, H. 0. and Umen, M. J., J. Org. Chem., J8> 3893 (1973); Smith, J. G. and Wikman, R. T., Synth. React. Inorg. Met. Org. Chem., 4, 239 (1974); House, H. 0., Respess, W. L. and Whitesides, G. M., J. Org. Chem., 31, 3128 (1966); House, H. 0. and Fischer, Jr., W. F., Ibid., 33, 949 (1968); House, H. 0., Fischer, Jr., W. F . , Gall, M. , McLaughlin, T. E. and Peet, N. P., Ibid. , 316, 3429 (1971).

33. Clark, R. D.. and Heathcock, C. H., Tet. Lett. , 1713 (1974); Kamidate, T. , Yotsuyanagi, T., and Aomura, K., Nippon Kaguki Kock, 2087 (1972); Chem. Abstr. 18, 34498U (1973).

34. Hosomi, A., Shirahata, A. and Sakurai, H., Tet. Lett., 3043 (1978); Vedjs, E. and Martinex, G. R., J. Amer. Chem. Soc., 6452 (1979) and references cited therein. REFERENCES TO PART III

1. Eliel, E., J. Org. Chem., 39, 3042 (1974).

2. Savignac, P. and Leroux, Y., J. Organomet. Chem., 57, C47 (1973); Leroux, Y., Tet. Lett., 591, 593

3. Paterson, W. and Proctor, G. H., Proc. Chem. Soc., 248 (1961); Speckamp, W. M., de Konlg, H., Pandit, V. K. and Huisman, H. D., Tetrahedron, 21, 2517 (1965); Negishi, E. and Day, A. R. , J. Org. Chem., 30, 43 (1965). ------

4. See references 1 and 2 in Part II.

5. Magnus, P. D. and Roy, G., Synthesis, in press.

6 . Abatjoglou, A. G. and Eliel, E. L., J . Org. Chem., 39, 3042 (1974); Cleavage of HMPA see - Kaiser, E. M . , Petty, J. D. and Solter, L. F., J . Organomet. Chem., 6l, Cl (1973); Normant, H., Cuvigny, T. and Martin, G. J., Bull. Soc. Chim., Fr., 1605 (1969).

7. Corbel, B., Paugam, J.-P., Dreux, M., and Savignac, P., Tet. Lett., 835 (1976).

8 . Seebach, D. and Enders, D., Chem. Ber., 108, 1293 (1975); Peterson, D. J., Organometallic Chemistry Reviews, Elsevier, A., 7, 295 (1972); Peterson, D. J., J. Amer. Chem. Soc., £3, 4072 (1971); Beak, P. and Brown, R. A., J. Org. Chem., 44, 4463 (1979). ~~

9. Zon, G., Tet. Lett., 3139 (1975); Takamizawa, A., J. Med. Chem., 17, 1237 (1974), and references cited therein; Eganland, W., and Zon, G., Tet. Lett., 813 (1976); Hall, Jr., H. K. and Zbinden, R. J., J. Amer. Chem. Soc., 85, 6420 (1958).

255 256

10. Shanzer, A., J. Org. Chem., 44, 3967 (1979) and also in Int. J. Peptide Protein Research, 12, l'30 (1978). ~

11. Kaufmann, T., Koppelmann, E. and Berg, H., Angew. Chem. Int. Ed., Engl., 9, 163 (1970).

12. Peterson, D. J., unpublished results.

13. Peterson, D. J., J. Amer. Chem. Soc., 91, 4027 (1971).

14. Peterson, D. J. and Ward, J. P., unpublished results.

15. Taeger, E., Fiedler, C., Chian, A. and Berndt, H. P., J. Prakt. Chem., 28, 1 (1965).