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 .T h e sign or "target” fo r pages apparently lacking from the docum ent 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 o f com plete 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. If copyrighted materials were deleted you will find a target note listing the pages 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 docum ent may have indistinct print. In all cases we have filmed the best available copy.

University Microfilms International 300 N. ZEEB RD., ANN ARBOR, Ml 48106

8222190

Temansky, Robert John

THE SYNTHESIS OF DODECAHEDRANE AND ITS MONOMETHYL DERIVATIVE

The Ohio State University P h D . 1982

University Microfilms

International 300 N. Zeeb Road, Ann Arbor, M I 48106

THE SYNTHESIS OF DODECAHEDRANE AND

ITS MONOMETHYL DERIVATIVE

DISSERTATION

Presented in Partial Fulfillment of the Requirements

for the degree of Doctor of Philosophy in the

Graduate School of The Ohio State University

By

Robert John Ternansky, B.S.

*****

The Ohio State University

1982

Reading Committee: Approved By

Dr. Leo A. Paquette

Dr. John S. Swenton

Dr. Harold Shechter Adviser V Department of Chemistry To my Parents

ii ACKNOWLEDGMENTS

I would like to acknowledge the members of the Paquette group who, throughout the years, have dedicated their time and efforts to the dodecahedrane project, especially Drs. Douglas W. Balogh, William

J. Begley, and Martin G. Banwell who have contributed in varying de­ grees to the present approach. 1 would especially like to thank

Professor Leo A. Paquette for his support throughout my years at

O.S.U. and congratulate him for achieving the goal which has long been an important part of his research efforts.

Thanks are also extended to Mr. Dick Weisenberger for obtaining mass spectral data and to Dr. Richard J. Laub and members of his research group as well as Dr. Christopher W. Doecke for help with the capillary gas chromatographic analyses. A special word of thanks goes to Dr. Ole Mols for his interest in the project and for his invaluable assistance in obtaining the high field NMR data in particularly ex­ pedient fashion.

I wish to thank the many members of the Paquette group, past and present, especially Dr. Francis R. Kearney, who have provided friendship, encouragement, and enlightening discussions. Also, thanks to Mrs.

Donna Rothe for typing this manuscript.

Finally, and most importantly, I would like to thank my family for fheir constant love, support, and encouragement throughout my academic experience. iii VITA \ January 4, 1955...... Born - Cleveland, Ohio

1977 ...... B.S., Summa Cum Laude, John Carroll University, Cleveland, Ohio

1977-1978...... University Fellow, The Ohio State University, Columbus, Ohio

1978-198 2...... Research Associate, Department of Chemistry, The Ohio State University, Columbus, Ohio

PUBLICATIONS

"Dodecahedrane," Ternansky, R. J.; Balogh, D. W.; Paquette, L. A., submitted for publication (1982).

"A Strategy for the Synthesis of Monosubstituted Dodecahedranes. An Example of Unprecedented Electrophilic Attack on Unfunctionalized Saturated Carbon with Inversion of Configuration," Paquette, L. A.; Ternansky, R. J.; Balogh, D. W., submitted for publication (1982).

"Dissolving Metal Reduction of a Hemispherical Dichloro Diester. Tests of Relative Enolate Nucieophilicity Within a Mixed Dianion and Oxidative Behavior of a Trisecododecahedryl Alcohol," Paquette, L. A.; Balogh, D. W.; Ternansky, R. J.; Begley, W. J.; Banwell, M. G., submitted for publication (1982).

"Electronic and Molecular Structure of Simple Bicyclopropyls. Photo­ electron Spectroscopy and Model Calculations," Gleiter, R.; Spanget- Larsen, J.; Gubernator, K.; Ternansky, R. J.; Paquette, L. A. J. Org. Chem. 1982, 47, 0000. 'w a a , —

FIELD OF STUDY

Major Field: Organic Chemistry

iv TABLE OF CONTENTS Page DEDICATION...... ii

ACKNOWLEDGMENTS ...... iii

VITA...... iv

LIST OF TABLES...... vi

LIST OF FIGURES...... vii

INTRODUCTION...... 1

Part

I. An Approach to the Synthesis ofCxs-Hexaquinacene. . . . 11

II. The Dissolving Metal Reduction of a Hemispherical Dichloro Diester. Preparation and Oxidative Behavior of a Trisecododecahedryl Alcohol ...... 21

III. The Synthesis of Monomethyldodecahedrane...... 35

IV. The Synthesis of Dodecahedrane...... 44

EXPERIMENTAL SECTION...... 51

APPENDIX ...... 84

REFERENCES AND NOTES...... 92

v LIST OF TABLES Page

Table 1 Attempted Rearrangement of 35...... 19 'Vb

vi LIST OF FIGURES

Figure Page

1. ORTEP Perspective Drawing of 81...... 41 W j 2. 200 MHz *H NMR Spectrum of 81...... 85

3. 200 MHz *H NMR Spectrum of 83...... 86

4. 200 MHz *H NMR Spectrum of 84...... 87 'Vb

5. 200 MHz lH NMR Spectrum of 99...... 88 ' V b

6 . 200 MHz *H NMR Spectrum of Dodecahedrane 4 ...... 89

7. Broadband Decoupled l9C NMR Spectrum of Dodeca­ hedrane 4 90 'X. 8 . Off-Resonance Decoupled l9C NMRSpectrum of Dodecahedrane 4...... 91 *1/

vii INTRODUCTION

Greek mathematicians in the 5th to 4th centuries B.C. defined five uniform, convex polyhedra as the tetrahedron, the cube, the octahedron,

the icosahedron, and the dodecahedron. The concept of these simple

Top: tetrahedron, cube, octahedron

Bottom: icosahedron, dodecahedron geometrical solids served to enlighten and inspire the scientists and philosophers of the time. In Timasus, Plato's discourse on cosmology and natural science, these simple shapes were envisioned as being the corpuscles or particles of the basic elements of nature.

Four of the five solids can be recognized as being completely con­ structed from two types of simple triangles. These four - the tetrahedron, the cube, the octahedron, and the icosahedron - were respectively designated as the basic particles of fire, earth, air, and water. The fifth solid, which cannot be constructed from either of the elementary triangles, had a special place in Plato's cosmology. This solid, the dodecahedron, was considered to be used by God in i delineating and adorning the universe.

Scientists in the 20th century have taken Plato's ideas and re­ duced them to the molecular level. The result of this has been an at­ tempt to combine atoms in such a way so as to build molecules having the shapes of Plato's elementary particles. For the organic chemist, this translates into the challenge of placing carbon atoms at each of the vertices of the geometrical solids. Because of the limita- tations imposed by the bonding capabilities of carbon, only the shapes representing earth, fire, and the universe can have suitable hydro­ carbon equivalents.

Responding to this challenge, the molecular equivalent of the cube, cubane, was synthesized in 1964 by P. E. Eaton and coworkers 2 at the University of Chicago. Also, the synthesis of two derivatives of tetrahedrane have been reported. Maier and coworkers successfully 3 synthesized tetra-tert-butyltetrahedrane in 1978. In the same year,

the synthesis of tetralithiotetrahedrane 2 was reported by the Schleyer a * group. To date, the unsubstituted tetrahedrane 3 has defied syn- thesis.

Li

L i

1 2 3 'Xi 'U

The last and most symmetric of the organic polyhedranes, the do­

decahedrane 4 , has been the focus of intense synthetic effort during 'V 3

9 the last two decades. Throughout this period, several creative ap­ proaches to the synthesis of 4 have been pursued. Only very recently, however, has a derivative of dodecahedrane been prepared. The syn­ thesis of 1,16-dimethyldodecahedrane by Paquette and coworkers represents the first successful construction of the dodecahedrane

6 framework. The ultimate objective, the synthesis of the parent hydro­ carbon 4, still remained.

Dodecahedrane 4 belongs to the highest possible symmetry point group (Ijj, icosahedral). Its high symmetry makes it unique among or­ ganic molecules and provides an aesthetic dimension to the challenge of its synthesis. Another novel feature of dodecahedrane is the presence of a central cavity. This space may be large enough to en­ capsulate a small species such as a hydrogen atom or proton. Although the challenges associated with such a process are great, it remains intriguing to consider the existence of this internal void in the dodecahedrane molecule.

Relative to cubane and tetrahedrane, the strain energy associated with the dodecahedrane molecule is very small. The strain that does exist"is the result of torsional strain due to the precise eclipsing of all the C-H bonds. Angle strain in dodecahedrane i6 minimal since the angle at which the edges of the polyhedrane meet (108°) is very 4

close to the idealized tetrahedral angle for a C-C bond (109o28').

The strain energy for dodecahedrane has been calculated to be between 7 1.4 and 2.9 kcal/mole per C-C bond. This amount is trivial when com­

pared to the calculated strain energies of tetrahedrane and cubane a (24 and 14 kcal/mole per C-C bond, respectively).

The challenge presented by the synthesis of such a deceptively

simple molecule is of large proportions. The task involves the com­

bination of twenty methine units utilizing thirty carbon-carbon bonds

in such a manner that twelve mutually fused five-membered rings are

produced. Furthermore, the ring fusions must all be cis-syn placing

the twenty protons on the outside of the molecule. Thus, as the mole­

cule is built-up, the synthetic manipulations must be carefully con­

trolled in order to assure that the larger atom (carbon) is always directed to the inside of the developing sphere. This task becomes

increasingly difficult as the molecule is constructed since non-bonded

interactions resulting from steric crowding in the intermediates in­

crease as additional cis-syn rings are built-up. Although dodeca­ hedrane can be considered to be relatively strain-free, the progenitor molecules, in general, are not. The frustrating result has often been

that transformations of these molecules do not follow the expected course but proceed in such a manner so as to relieve strain, often­ times producing an undesired product which cannot be utilized further.

Because of the special challenges associated with the synthesis of dodecahedrane, several creative approaches have been developed in at­ tempting to construct this "unnatural" product. Unlike natural 5 product synthesis, the target molecule in this case is only Imagined

to be capable of existence. Its actual presence in the physical vorld can only be assured through its synthesis. Some of the approaches aimed at developing dodecahedrane into a physical reality are presented in

the following discussion.

In 1964, Woodward and coworkers synthesized triquinacene 5 with f\j the hope that two such molecules could be brought together to form 9 dodecahedrane. Although much effort was made to promote the desired dimerization, success was not realized. The unfavorable steric and entropic factors involved in the formation of the six requisite carbon- carbon bonds give the process a low probability for success.

----

5 4 'b

In an effort to minimize these problems, various triquinacene derivatives were prepared and joined by one or more sigma bonds, as i o illustrated by Paquette's synthesis of bivalvane Although closer to the target molecule, bivalvane and other coupled triquinacenes have 11 not been successfully elaborated to the dodecahedrane molecule. 6

McKervey and coworkers at Cork, Ireland have recently reported the synthesis of a C20-polyquinane 7 which may serve as a dodecahedrane 12 ^ precursor. The transformation of 7 into dodechaedrane, which requires 'Xj the formation of five carbon-carbon bonds, is presently under active 1 3 investigation.

7 'V

Since dodecahedrane is likely to be the most stable polycyclic

C2oH20 hydrocarbon, it was felt that molecular rearrangement of a strained C20H20 hydrocarbon m a y yield 4. An example of this approach 1 4 is the isomerization of 8 attempted by Schleyer and, independently, 1 5 by LeGoff. Unfortunately, no products containing the dodecahedrane skeleton have been observed to be formed by such a process.

isomerize

'X,8 £'Xj

Another approach presently being pursued by the Eaton group in- “ 16 volves the synthesis of a polyquinane network called peristylane

Capping of peristylane with a cyclopentane unit conceptually com­ pletes the synthesis. Although the synthesis of a highly functionalized 7

9 peristylane derivative has been reported, its conversion into

dodecahedrane has not been achieved.

Paquette's synthesis of Cx6-hexaquinacene 11 represents yet another r\f\j 17 approach to the construction of the dodecahedrane molecule. The ob­

vious relationship between 11 and h has stimulated Intense research in- vu % xe to the chemistry of this molecule as well as its derivatives.

Equally intriguing as a potential dodecahedrane precursor is the next

higher order hexaquinane, Cxe-hexaquinacene 12. This polyquinane, lack-

ing only two of the twenty carbons required for dodecahedrane, appeared

to be an attractive precursor to the target molecule. Furthermore, the

physical characteristics of such a compound would serve to shed addi- i • tlonal light on the question of neutral homoaromaticity. For these

reasons, efforts have been made to synthesize CXB-hexaquinacene. A

detailed discussion of these efforts will be presented in a later

portion of this dissertation. Further attempts to synthesize dodecahedrane by Paquette and co­ workers have led to the investigation of the chemistry of molecules containing all of the twenty carbon atoms required in dodecahedrane.

These molecules in turn have been prepared through the stepwise elabo­ ration of 13, resulting from the domino Diels-Alder addition of 9,10- r\ A j 20 dihydrofulvalene to dimethyl acetylenedicarboxylate. This approach, which will be explored in detail, has led to the first successful synthesis of the dodecahedrane molecule.

MeOOC-CsC-COOMe

ICOOCH COOCH,'

VL

The special interest in dodecahedrane shown by synthetic chemists has been shared by the theoreticians who, throughout the years, have made several predictions with respect to the physical properties of this unique molecule. Now that dodecahedrane has become a physical reality, these predictions can be tested and a better insight into the relationship between theory and reality can be obtained. Some of 9

the physical characteristics which have been calculated for dodeca- 7a,>i 7a,aa hedrane include its heat of formation, strain energy, size 3 2 la,33 of the central cavity, spectral properties, and orbital ener- «,2 ia gies. Furthermore, species derived from dodecahedrane will be of significant interest. The dodecahedryl cation 14 should prove to be a rather stable entity providing valuable information re­ garding the characteristics of the existing carbonium ion. The homododecahedral cation 15 will be an equally informative species. Do- <\A/ decahedrene 16 represents a molecule in which the strained double bond 'Vb will necessarily assume a bent geometry. This contrasts with the nor­ mal tendency of strained double bonds to twist in response to distorting influences.

to V i

Finally, the pharmacological significance of dodecahedrane will be borne out by the synthesis of derivatives such as aminododecahedrane.

Such compounds having spherical hydrocarbon portions have been shown

26 to exhibit biological activity, as is the case with 1-adamantyl- amine hydrochloride, a compound marketed by the duPont Company for treatment of Asian Flu and Parkinson's disease.

This dissertation will describe the first successful synthesis of the dodecahedrane molecule as well as its monomethyl derivative. These results should serve as a foundation for the continued study of dodecahedrane and related molecules. PART I: An Approach to the Synthesis of

Cja-Hexaquinacene

As previously discussed, the potential of CiB-hexaquinacene 12 to serve as a precursor to dodecahedrane ^ as well as its inherent physi­ cal properties make it a worthy objective for synthesis. Of the several approaches which could be devised for the synthesis of L2 , one based on the coupling of a suitable nortriquinacene derivative seemed best suited to our needs. This strategy becomes apparent if one views CiB-hexaquinacene as a molecule composed of two triquinacene portions fused via a common ethylene unit.

Furthermore, it was reasoned that the fused [3.3.0]bicyclic system around which the molecule is built could be constructed by way of a pinacolic coupling-rearrangement sequence. The synthesis of bicyclo-

[3.3.0]oct-l(5)-ene 17 from cyclobutanone serves to illustrate the 25 strategy

11 12 □I ^ 0 - 0 O O 0 Cs 6 OH OH a

Na, moist ether C D O p OH

17 'Vb

Applying these Ideas in a retrosynthetic fashion, 12 is viewed as 'VIi being prepared from .18 by selective reduction of the tetrasubstituted double bond. In turn, 18 is seen as arising from a pinacolic coupling- 'Vb rearrangement sequence beginning with nortriquinacen-2-one 3^. Thus,

12 OA/

18

COaH

& became our first synthetic objective. 13

Compound 19 appeared to be accessible through oxidative decar- AA/ boxylation of the known carboxylic acid This substance was pre­ pared by ring contraction of 2,3-dihydrotriquinacen-2-one which, 26 in turn, was synthesized by the sequence outlined below.

/ r ~ r \ NaNs 1. hv

H 2S 0 ^ 0 * ^ ^ ^ 2. Hs0+

20 AA/

2. LDA, C1P (0C2Hs)a C

tCr03.Pyr H 0 ^ C = = A THPO ROH H

24 23 AA/ AA/

Regiospecific dehydration of prepared from Thiele's acid 20 27 by the method of Deslongchamps, was accomplished indirectly. Con­ version of 21 to its tetrahydropyranyl ether followed by formation of AAj the enol phosphate yielded 22. Dissolving metal reduction of 22 and deprotection gave the alcohol 23 which was oxidized to the desired AA; ketone 24. AA/ Attempts had been made to directly dehydrate 21 using conven- AA tional techniques such as acid catalysis, pyrolysis of the derived acetate, and base promoted elimination of suitable derivatives. In these cases, the product obtained was not 24 but rather the isomeric AA/ 14

ae ketone It was discovered during the course of the present in-

-Ha0

OH 21 25 vu vestigation, however, that elimination of the exo tosylate with neutral alumina yielded the desired product 24 in 98% yield. Important- r\f\j ly, this two-step dehydration procedure is serviceable on a large scale.29

A1 2O3

O T S C H a C la

26 24 Suitable ring contraction of 24 required the preparation of the

29 hv dioxane, TsNa -j- n 2 Ha0 >H«C (c2h5)2nh 0<

27 28 r\/\j

HsCOaC 30 u 15 followed by treatment of 27 with j>-toluenesulfonyl azlde and diethyl- w amine. The photochemically Induced Wolff rearrangement of 28 gave ester 30 was o/v w obtained by Irradiating 28 in methanol solution. r\ A j Oxidative decarboxylation of 29 was first attempted utilizing 3 O the method of Wasserman and Lipshutz. In this procedure, the di­ anion of a carboxylic acid is treated with oxygen to form the o- hydroperoxide 31 which is directly decomposed with N,N-dimethylforma- w mide dimethylacetal to yield the decarboxylated product. Treatment

\©-co© L °» - >C°°H PMF » Vo ' 2. H 0+ X C0aH acetal /

31 esters, pre­ pared from the corresponding a-hydroperoxy compounds 31, are converted r 'XAj 3 1 into the desired carbonyl species by a two-step process. Reduction of the a-hydroxy acid or ester followed by oxidative cleavage of the resulting diol 32 gives rise to the degraded ketone. Several attempts #\A/

\ / ° H LAH \ r/ ° H NalO* V- CN OH * / C - 0 C0aR

& 16 were made to form the a-hydroxy derivatives of 29 and 30. In these V b v b cases, the corresponding anions were generated with lithium diiso- propylamide and quenched with molecular oxygen. Treatment with sodium sulfite or triethyl phosphite (in situ) followed by work-up led only to the recovery of unreacted starting material. The use of carefully dried oxygen did not alter the results. Furthermore, it was found that when attempts were made to generate the enolate of 30 by adding lithium V b diisopropylamide to a solution of 30 at 0 °C (inverse addition), an V b intermolecular condensation product 33 was isolated in 6 8 % yield. oa. The formation of this novel compound indicated not only that the anion was being formed but that it behaved as a highly reactive species.

C02 CHs

33 Vb Consideration of the mechanism involved in anion oxygenation may provide an explanation for the difficulties encountered in preparing the a-hydroxy derivatives of and The mechanism is thought to involve initial electron transfer from the anion to oxygen followed a 2 by radical recombination. This requirement stems from orbital con- 93 siderations of ground state oxygen and the spin conversion rule.

Since, the radical recombination is expected to be quite rapid, the initial electron transfer leading to the high energy cyclobutane radical is most likely the problematic step. 17

In the continued search for a useful oxidative decarboxylation procedure, our interest was attracted by the method developed by

Trost. In this three-step process, the dlanion of a carboxylic acid is quenched with dimethyl disulfide to form the o-sulfenylated acid. Subsequent treatment of this a-substituted acid with N- chlorosuccinimide in an alcohol solvent leads, via the mechanism shown below, to the protected carbonyl compound. Deprotection with acid yields the desired product.

CH3 SSCH3 V >SCHa NCS CHs V c© C' V*"/ © / x cc,® X X co © /

\ o// 0R ROH \ ^SCHs \ c=o HsO r / ^ O R

In contrast to our experiences with oxygen quernchines, deprotona­ tion of 29 with lithium diisopropylamide followed by treatment with di-

1. NCS, CH,OH

tered in the next step of the sequence. Although the desired ketone

19 was observed for the first time, the yield was low (32%). Efforts !\f\j to Improve the yield of including an attempt to electrochemically 95 degrade the a-sulfenylated acid 34, were to no avail. Although 'XAj quantities of nortriquinacen-2-one 19 were consequently very limited, *w the decision was made to proceed with the synthesis as planned.

The next step in the projected synthesis of CXB-hexaquinacene in­

volved pinacolic coupling of nortriquinacen-2-one T9. Initial at- 9 6 tempts utilizing amalgamated magnesium and trimethylchlorosilane were unsuccessful. Success was realized when 19 was reacted with AA. amalgamated magnesium and titanium tetrachloride, according to the

97 method developed by Corey. Although the desired diol 35 was obtained, i\Aj the yields were once again disappointingly low (2 0 %).

Mg(Hg)x TiCl*

19 35 'Vb AA/ With the small amount of pinacol product 35 available, an in- rV\j vestigation was made to determine suitable conditions for its re­ arrangement to the spirocyclic compound 36. The reactions were run on w a milligram or less of material and were monitored by TLC. In all cases examined (Table I) either the starting material was left un­ touched or extensive decomposition occurred. 19

Table I. Attempted Rearrangement of 35. < \A i

(a) Reagent Solvent Time (h) Temp, °C Result

© © EtsNS0aNC02Et38 CH3CN 1.25 25 N.R. © © Et3NS0aNC02Et c h 3cn 15 25 N.R. © © EtaNSOaNCOaEt c h 3cn 2.75 60 N.R.

(COaH)a Toluene 14 1 1 0 N.R.

BF 3 •0Et2 Ether 1.5 25 dec.

BF3 *0Et2 CHaCla 4 40 N.R.

BF3 »0Et2 Toluene 6 0 N.R.

H2 SO<, (con) -- 0.25 0 dec.

HaSO* (1 N) THF 2 0 25 N.R.

HC10* (20%) THF 48 25 N.R.

£-Tos-OH Toluene 16 25 N.R. j>-Tos-0H Toluene 3.5 1 1 0 ucu •

HBr (48-50%) THF 1 2 25 N.R.

HBr (48-50%) THF 1.5 90 dec.

P0C13 Pyridine 2 25 N.R.

P0C13 Pyridine 8 80 dec.

SnCl<, CHaCla 3.5 25 dec.

£-TosCl/DMAP(b) Pyridine 1 2 0 25 N.R.

(a)-N.R. = no reaction; dec. = decomposition was observed.

(b) DMAP «= 4-dimethylaminopyridine. 20

The low yields obtained in the latter stages of the synthetic sequence coupled with our inability to find suitable conditions to effect the crucial pinacolic rearrangement caused us to put aside our attempts to synthesize CXB-hexaquinacene. Our continued interest in obtaining dodecahedrane, however, prompted us to shift our attention

to a more promising approach. PART II; The Dissolving Metal Reduction of a Hemispherical Dlchloro

Diester. Preparation and Oxidative Behavior of a

Trisecododecahedryl Alcohol

Our strategy for the construction of the dodecahedrane framework evolved from an investigation into the chemical reactivity of the pre­ ss viously prepared dichloro diester This compound is derived from

37 OA, the cyclopentadienide anion 38, an inexpensive source of cyclopentane units. Oxidative coupling of 38 to 9,10-dihydrofulvalene, followed rU\j by a domino Diels-Alder reaction with dimethyl acetylenedicarboxylate yielded, after ester hydrolysis, the adducts ^39 and AO. The desired adduct readily separable from the unwanted isomer AO, was obtained in yields of 10 to 15%. Although the isolated yields are low, the mole­ cular complexity achieved in this single operation far outweighs this disadvantage. Close inspection of 39 reveals the presence of four VI; cis,syn fused five-membered rings resulting in six cis-locked methine hydrogens. Furthermore, 39 contains fourteen of the twenty carbons

21 22 required for dodecahedrane. Its conversion into diketo diester ^ sets the stage for the introduction of the remaining six carbon atoms.

MeOOC-CsC- COOMe

w

CH3OOC CHiOOC I

COOCH, C i " 'TT a / ICOOCH, ^VcOOCH COOCHj I OH- \ 0H‘

H0 _ 0

2 0 0 H COOH

$ 23

Iodolactonization of 39 followed by cleavage of 41 to the iodo- V b Vb hydrin 42 and Jones oxidation delivers 43. Reductive elimination of a , Vb the iodine atoms with zinc and ammonium chloride in methanol efficiently

completes the synthesis of 44 (80% overall from 39).

(NaOCHs) i S b CHsOH )h-0 ICOOH■COOH OOCH COOH COOCH3 '

39 41 42 Vb Vb Vb N3jCr207f II2 sc*

- ICOOOH3 C0 0 CH3

44 43 Vb Vb

Introduction of the six remaining carbon atoms is accomplished by

bisspiroannulation of 44 with diphenylcyclopropylsulfonium ylide. Vb Baeyer-Villiger oxidation of 45, accomplished by the action of hydrogen V b peroxide, is followed by isomerization of the resultant bis(spirolac- tone) 45 under strongly acidic conditions to deliver 47. Catalytic Vb Vb hydrogenation of ^47 serves to position four additional methine hydro­ gens on the exterior of the developing sphere and completes the con­ struction of two additional cis,syn fused cyclopentane units. Con­

trolled reduction of 48 with sodium cyanoborohydride affords the di- Vb 24

lactone 49 which is converted into the dichloro diester 37 by treatment rV\j

I ^-SPh2

NaCNBHa

P d -C

48

/•

37 /\A, Early investigations into the chemical reactivity of 37 revealed that generation of carbanionic centers at the originally chlorinated carbon atoms did not result in cyclization to the proximate ester functionalities. Instead, simple reductive dechlorination was ob- «o served-.

Recently, we were prompted to explore an alternate electronic process in which the ester functionalities of ^7 were made to experience 25 reduction first. An electrochemical analysis indicated that the reduc­ tion of ^ should follow a well defined course having its basis in the half-wave potentials of the component functional groups.

While the carbonyl groups of esters are less readily reduced than those of ketones and aldehydes, the first waves leading to radical anions (or to the protonated forms thereof) appear almost always at more <>i positive potentials than those of alkyl chlorides. It follows, there­ fore, that electron transfer to 37 should occur more rapidly to the r \A > more electropositive carbomethoxy group generating radical anion W .

The open question was whether this chemoselectivity would materialize cleanly when reduction was effected with an alkali metal in liquid ammonia at reduced temperatures. As will be shown, this extrapolation gives every indication of being proper.

Once 50 is generated, two chemical transformations are, in prin- 'W ciple, accessible to the radical anion. The first is based upon 26 recognition of the fact that 37 is a 1,4-dicarbonyl system and conse- r\ A j U2 quently subject to cleavage of its highly strained central bond.

However, the structural precondition for this reaction is proper stereo- *2e,<>9 electronic alignment of the carbonyl pir orbitals with the a bond. * In 50, this requires that both oxygenated carbons be oriented in that r\J\, manner which projects their pendant groups into space already occupied by the endo protons of the proximal transannular methylene groups.

For this reason, suitable stereoelectronic overlap of the negatively charged carbon with the ester carbonyl group is expected to be suf­ ficiently sterically disfavored to allow the second option to become kinetically dominant. More favorably, the intramolecular Sjj2 displace­ ment of the chloride ion which is illustrated requires only that one of the congested centers be properly oriented. Furthermore, molecular models of 50 reveal that the alignment required of the nucleophilic r\j\j uu carbon for backside attack at C-Cl is relatively low in steric demand.

Following construction of a new framework bond in this manner, addition of a second electron and loss of methoxide ion is anticipated with formation of 51. Continued reduction of this intermediate should, <\A> again from known typical half-wave reduction potentials, involve elec­ tron transfer to the ketone carbonyl. At this stage, central bond cleavage is inevitable. Since the resulting ester enolate is geome­ trically unable to displace the second halogen, this chlorine becomes subject to independent reductive cleavage.

From this analysis of the reducibility of ^37, we see that a single cyclization can result and that solutions of dianion ^ 2 are thereby made available. In agreement with this hypothesis, Balogh 27 observed that treatment of ^ with an excess of lithium in liquid am­ monia followed by addition of an excess of methyl iodide afforded mix- tures of and This result served as the basis for the syn-

6 thesis of 1,16-dimethyldodecahedrane.

53 W

a, R = H; b, R = CH3 % ^

In order to apply the dissolving metal reduction of 37^ to the syn­ thesis of dodecahedrane, it seemed appropriate to quench the dianion 52 r\f\j formed by this reaction with a proton source. This seemingly modest goal was complicated by the fact that the only product observed in ini- tial quenching experiments was the hydroxy ester 54. This product r\f\j presumably arises from the geometrically favorable formation of a transannular bond. An examination of molecular models reveals that *6 a this irreversible process serves to relieve non-bonded steric inter­ actions which exist along the "open" portion of the developing sphere.

CH30 0 c h 3Pns*° Cl- 1. Li, NH3

2. CaH 90H

37 54 vo w 28

It was reasoned that this undesirable result could be circumven­ ted if the ester functionality were to be reduced to a non-enolizable entity before transannular bond formation had a chance to materialize.

After much experimentation, it was found that addition of a proton source to a liquid ammonia solution of the dianion ^ 2 in the presence of a large excess of lithium metal served to adequately suppress the undesirable reaction. Strict control of the reaction conditions led to a 53% yield of hydroxy ketone 55. Along with 55, variable amounts of the overre- 'Vb 'Vb duced product 56 as well as the transannular product 54 were obtained. >Vb 'Vb

Li, NH9 C3K3OH 54 w

52 55 56 'Vb 'Vb 'Vb

Having successfully taken advantage of the unique reactivity of dichloro uiester 37 in preparing the tetresecododecahedrane derivative 'Vb ^5, we turned out attention to the photochemical behavior of this car­ bonyl-containing compound. 6 From earlier studies on similar systems, it was learned that photo­ excitation of such carbonyl groups leads to an efficient "homo-Norrish" cyclization resulting in the formation of a cyclopentanol unit. This rather unusual behavior is a direct consequence of the geometrical constraints imposed by these rigidly held hydrocarbon networks. In these systems, the familiar y-hydrogen abstraction which occurs from a slx-membered transition state is not possible. Instead, the proximate 29

6 -hydrogen is abstracted followed by radical recombination to provide a five-membered ring.

In accordance with these studies, irradiation of in benzene-tert- butyl alcohol (4>1) solution resulted in carbon-carbon bond formation and production of the diol 57 (76%). This triseco derivative was se- OAj lectively dehydrated to the hydroxy olefin 58 which was in turn reduced 'Wi with diimide to the Co-symmetric alcohol 59 (95% from 57). — OA. AAi

HO. HO. HO-

(TsOH)

HO !°2

$ 1 %fl

The use of diimide for the reduction of olefinic linkages in these systems results from the observation that catalytic hydrogenation is 6 ineffective in accomplishing this transformation. This phenomenon may reflect the fact that these strained olefins exhibit a very high formation constant for the metal-alkene bond, making the transition state for this process very energy demanding.

With arrival at 59, the installation of an additional framework bond became our next objective. In order to make use of the previously successful homo-Norrish process, the oxygenated carbon had to be chemi­ cally modified to make it responsive to photochemical activation.

Consequently, aldehyde 60 became the next target objective. At the 'VU outset, this appeared to be a modest aim. However, a great deal of experimentation has subsequently shown that we had grossly misjudged 30 the actual state o£ affairs in this most sterically congested^ molecule.

The first hint of difficulty appeared when treatment of 59 with pyri- *V\j *4 7 dinium chlorochromate (FCC) provided no Instead, a mixture of a,0- unsaturated aldehyde 62 and olefin 63 was obtained. The situation was OAi f\Ai not at all improved when recourse was made to PCC under buffered (NaOAc) 49 conditions, pyridinium dichromate, activated manganese dioxide, or

90 MnOa on carbon. The latter group of reagents delivered either 6>2 alone or a mixture of 62 and 63. In an effort to frustrate both overoxidation l\ f \ j < \A t 9 X and dehydration, we examined the suitability of the RuCl2 (PPh3)s and 92 KH-PtCla(PEt3)a reagent systems for our purposes. Whereas the first transition metal complex gave only 62, the latter generated a plethora of '\Aj products. As anticipated, those oxidants which act at the mechanistic

H O ^ H CHO

60 61 62 'VV r\ A j

63 64 'W, 'XA, level by activating the hydroxyl group via conversion to a good leaving

99 94 group, e.g., the Pfitzner-Moffatt and Doering-Parikh reagents, transformed 59 uniquely into 63. An exception to this behavior was V O ^ A. 99 observed with the Corey-Kim reagent which afforded only the 31 a,0-unsaturated aldehyde. The la££er course of even£s was also fol- 9C lowed when 59 was exposed £o N-phenyltr±azolinedione or the Swern vu 3 7 reagent. In a mechanistically relevant discovery, was found to be smoothly converted to norketone 64 under the conditions of Jones oxi- v b datlon.

In an attempt to explain these frustrating results, it can be assumed that the first-formed product in the oxidation of 59 is the r \ A j desired saturated aldehyde ^ . However, the high degree of steric crowding experienced by the aldehydo carbonyl serves to promote enolization to 61. Since the susceptibility of 61 to oxidation can 'Wi V b be intuitively assumed to be greater than, or at least comparable to that of its saturated precursor, the steady state concentration of the enol likely never increases to the point where its spectroscopic ob­ servation becomes possible. Also, the possibility that 61 undergoes air V b oxidation upon workup cannot be ruled out. In the case of the Jones oxidation, the heightened reactivity of this reagent is believed to allow for continued oxidation of 61 with ultimate extrusion of the ori- w ginal carbinol carbon.

The lack of success in achieving the conversion of 59 to 60 led V b V b us to investigate the oxidative behavior of alcohol Jifi. Examination of a molecular model of 58 revealed that the double bond causes dis- V b tortion of the carbocyclic framework resulting in the release of an appreciable level of steric congestion about the CH2 0H group. It was felt that this structural feature might alter the reactivity of the alcohol moiety under oxidizing conditions. With PCC, conversion to mixtures of 65 (two isomers) and the dehydration product (diene) was 32

observed in all too reminiscent fashion. Only 65 was formed when sil- V b ver carbonate on Celite was utilized. Jones oxidation produced the

norketone, in agreement with earlier precedent. Strikingly, however,

manganese dioxide in dichloromethane solution furnished chromatographi-

cally separable mixtures of 65 (42%) and 6 6 (45%). Zn contrast to the ^ w appreciable lability of a,8 -unsaturated aldehydes 62 and 65, 6 6 proved V b V b V b CHO

& entirely stable and, in fact, surprisingly resistant to air oxidation.

These findings served to solidify our position that the abnormal re­

sponse of 5 9 to oxidation was due chiefly, if not totally, to its ex- V b pecially congested environment.

Unfortunately, the geometrical features of 58 which make the iso- V b lation of aldehyde 6 6 possible would else appear to effect i t s photo- f\f\j chemical behavior, impeding the formation of a homo-Norrish product.

In accord with this prediction, the irradiation of 6 6 failed to produce V b any recognizable products. Furthermore, attempted diimide reduction *6 b of 6 6 did not produce the desired aldehyde 60. V b ^ The importance ascribed to the preparation of 60 prompted us to V b further explore the chemistry of alcohol 59. This led us to consider V b the possibility that oxidation of 59 to the aldehyde level might be - V b achieved indirectly. If the alcohol were first functionalized and this derivative subsequently degraded in the absence of an oxidizing agent, 33

It was deemed feasible that 60 might survive and prove capable of vb isolation. Very few schemes fall into this category. In the Barton

56 procedure, reaction of an alcohol with phosgene in the presence of

quinoline is followed by treatment with dry dimethyl sulfoxide and tri-

ethylamine. Dimethyl sulfoxide displaces the chlorine atom of the

chloroformate intermediate and serves as the indirect oxidant (pro­

vided that elimination is not competitive). In the case of 59, con- Vb version to the chloroformate proceeded in quantitative yield. However,

following exposure of this product to dimethyl sulfoxide and triethyl-

amine, there was isolated only a,B-unsaturated aldehyde 62 and starting V b alcohol.

The persistency of these problems led us to consider a scheme

which, in principle, would generate 60 photochemically under conditions w i where its homo-Norrish cyclization would be equally favorable. This

tandem excited state oxidation-cyclization, if successful, would obviate

the necessity of isolating and handling 60. Binkley has described Vb a photochemical method for the conversion of alcohols to ketones and s«,to aldehydes which involves the irradiation of pyruvate esters.

This precedent was followed and 59 was uneventfully condensed with Vb pyruvyl chloride. Upon direct irradiation of this pyrnvate ester in J»j60 benzene through pyrex as prescribed, mixtures of numerous compo­

nents (separable on preparative TLC) were formed. Aldehyde ^), or

its homo-Norrish successor, were clearly not present.

In a final attempt to generate the desired aldehyde the reduction

of a ,B-unsaturated aldehyde 62 was examined. It was found that 62 could V b Vb be reduced under dissolving metal conditions (lithium/liquid ammonia/ 34 tert-butyl alcohol) to yield, upon workup, the saturated aldehyde in its enol form 61. Unfortunately, 61 proved to be extremely susceptible < \A i V U to air oxidation (in contrast to 6 6 ), being completely converted to 62. rV \ j r\f\j Furthermore, photolysis of 61 in deoxygenated toluene-ethanol (9:1) solu-

On the basis of these results, it became evident that a modified approach was required in order to arrive successfully at target compound

4. PART III: The Synthesis of Monomethyldodecahedrane

A re-evaluation of our strategy for the construction of dodeca-

hedrane A led us to the conclusion that a suitable triseco aldehyde

derivative could be prepared only if the a position were fully substi­

tuted. This requirement would be met if monoalkylation of 52, derived rJ\t from the dissolving metal reduction of dichloro diester 37, could be w, accomplished. Furthermore, such monosubstituted derivatives would po­

tentially serve as precursors to a variety of monosubstituted dodeca-

hedranes. For these reasons, we decided to investigate the monoalkyla­

tion of ^ in order to assess the relative nucleophilicities of the

two anionic centers. (Little information is currently available on the 61 relative nucleophilicities of disconnected ester and ketone enolates. )

When consideration is given to the monoalkylation of 52, it be- a/b comes imperative for our purposes that electrophilic capture be achieved

selectively a to the carbomethoxy group to give 67. This monoanion was r\ f \ j viewed to be an intermediate whose propensity for subsequent intra­ molecular aldol condensation would be extensively reduced because of

the severe contortion demanded of the asterisked carbon in 69. In

6 8 , rapid ring closure to 70 would be expected since configurational ,w Inversion a to the carbomethoxy group now serves to project this

substituent to a sterically less crowded environment. Our expectation

35 36 that ester enolates should be more susceptible to electrophllic cap­ ture than ketone enolates encouraged us to explore the monoalkylation behavior of 52. 'Vb

52 C K , l / <\A, C113I

67

69 70 'Vb 'Vb

The addition of 37 to a solution of six equivalents of lithium In 'Vb liquid ammonia at -78 °C led to complete consumption of the metal as evidenced by the dissipation of the deep blue color. Addition of one equivalent of methyl iodide to this solution followed by neutralization with solid ammonium chloride led to a chromatographically separable mixture of ^ (46%) and (24%). This satisfying result confirmed our predictions and provided a convenient method for the Introduction of a variety of substituents onto the developing dodecahedryl framework.

c h 3

37

Having prepared the tetraseco derivative the decision was made to utilize this system as a model and to pursue the synthesis of mono- methyldodecahedrane.

Irradiation of 71 in deoxygenated benzene-tert-butyl alcohol (4:1) <\A> Bolution proceeded as anticipated with formation of the homo-Norrish product 72 (100%). Dehydration of 72 with p-toluenesulfonic acid in r\j\j rV\j refluxing benzene followed by diimide reduction of the resulting olefin yielded the triseco ester 74 (92%). The symmetric nature of 74 was < W 'VA/ confirmed by a simplified 19C NMR spectrum consisting of only fourteen lines. 38

With construction of the front portion of the developing sphere

completed, our attention was focused on altering the oxidation level

of the carbomethoxy group. Reduction with diisobutylaluminum hydride

occurred cleanly to give rise to the primary alcohol 75 (94%). Since Vb the carbon a to the carbinol group is fully substituted, the problem of

overoxidation which was encountered in the unsubstituted derivative

should be circumvented. As anticipated, oxidation of 75 proceeded V b cleanly to the aldehyde 76 (100%). V b

c h 2 o h

(i-Bu)aAlH PCC C6H6

V t & I k Although the fully substituted nature of the a-carbon in 7 6 made

its preparation possible, it was also expected to have a detrimental

effect in the ensuing homo-Norrish process. This expectation is based

on the known propensity of such systems to undergo facile decarbonyla-

62 tion. It was hoped, however, that the desired photocyclization would

compete to a favorable degree. Even though decarbonylation was diffi­

cult to suppress, partial success was achieved by performing the ir­ radiation in a 9:1 toluene-ethanol solvent system at -78 °C. These

conditions provided the desired homo-Norrish alcohol 77^ in 25% yield.

Although the major by-products proved to be the result of decarbonyla­

tion, some photoreduction to 75 was also noted. V b Oxidation of 77^ to the diseco ketone 78 proceeded in high yield and

provided the opportunity to once again make use of the homo-Norrish

process in forming a carbon-carbon bond. Irradiation of 7B^ under con­

ditions used previously in the conversion of 71 to 72 provided the seco -w, on, alcohol Acid-catalyzed dehydration of 79 resulted in the formation

of seco olefin (84% from 78)-

OH

hv (TsOH) 77 PCC 'W » CH2C1: CoH6, A

78 /9 oun «

In earlier work, a dimethyl seco olefin related to ffy was found to

be subject to rapid acid-catalyzed isomerization in the presence of

trifluoromethanesulfonic acid at room temperature, the multistep re­ ft arrangement culminating in formation of 1,16-dimethyldodecahedrane.

Not yet elucidated is the timing of the methyl shift vis-a-vis closure

of the final framework bond. In any event, the obvious complexity

of this cyclization process was further manifested when 80 was ana- *V \ j logously treated. Not only did a myriad of products result, but the most prevalent of these was the beautifully crystalline "isododecahedrane" 40

81 (55-60% of the volatile constituents). The unusual structure of w this novel polyqulnane was confirmed through x-ray crystal analysis

• 3 (Figure 1).

CF9SO3H

CH2CI2

80 Cl r\f\, \Aj noted to project the associated internal hydrogens well beyond intra­ molecular contact range. Evidently, the very severe nonbonded interac­ tion between the same hydrogen atoms in 80 provides the necessary ste- OA> 64 ric driving force for carbonium ion 82 to undergo unprecedented trans- AAi annular electrophilic attack at the unactivated methine carbon with inversion of configuration!

The unexpected behavior exhibited by 80 under the influence of r\r\j strong acid, although interesting in its own right, did not serve to accomplish our ultimate goal. Since the installation of the final framework carbon-carbon bond via acid-catalyzed rearrangement of an olefinic precursor did not appear to be generally applicable, a more 41

Figure 1. ORTEP Perspective Drawing of 81. 42

universal approach was sought. It was hoped that such an approach could

be readily adapted to the synthetic scheme which had been developed and

that it would be applicable to the synthesis of a variety of substituted

dodecahedranes.

With these goals in mind, our attention was directed to an approach

in which the final carbon-carbon bond would be formed by the removal of

a molecule of hydrogen from the two opposing methylene groups of a

seco derivative. Such a process would be expected to be energetically

favorable, since a great deal of strain which results from the serious non-bonded interactions between the opposing methylene groups would be relieved.

The use of catalytic dehydrogenation for the formation of carbon-

63 carbon bonds is a'well-established process. Although generally used for the formation of olefins, some examples of its use in sigma 63,66 bond-forming reactions are known. It was our hope that this pro­ cess could be applied to the task at hand.

In order to prepare a suitable substrate for the dehydrogenation reaction, olefin 80 was reduced with dilmide to the Beco compound 83 i\Aj r\J\, (100%). When an intimate mixture of &3 and 10% palladium on carbon was heated in a sealed stainless steel reactor at 250 °C for 30 minutes, a reaction was observed to occur as analyzed by capillary gas chroma­ tography. Among the products formed were compounds which appeared to contain sites of unsaturation. Presumedly, dehydrogenation had occurred by removal of hydrogen from the methine units. Repeating the reaction under an argon atmosphere did not alter the results. 43

It occurred to us at this point that it might be possible to

suppress, or reverse, the olefin-forming process by performing the

dehydrogenation under a hydrogen atmosphere. Success was realized

upon heating at intimate mixture of ,83 with fifty times its weight of

10% palladium on carbon (previously exposed to 50 psi of H3) at 250 °C

for 7 hours in a sealed stainless steel chamber. Although capillary

gas chromatography indicated that a variety of products had formed, the olefinic byproducts previously observed were insignificant. From

*H NMR analysis of the mixture, monomethyldodecahedrane 84 was present r\J\j to the extent of 35-40% and could be separated from the reaction mixture by recrystallization from benzene. In this way, a 28% yield of ^ was realized. An x-ray crystal structure analysis of monomethyldodeca-

69 hedrane is currently in progress.

H 2NNH, 10% Pd-C * HaOs 250°C

80 83 84 'V b

Having successfully developed an important protocol for the in­ troduction of the final carbon-carbon bond into the dodecahedrane framework, we were now prepared to pursue our ultimate objective 4. FART IV: The Synthesis of Dodecahedrane

In developing a suitable route to dodecahedrane 4, we hoped to

be able to apply the knowledge obtained from the synthesis of its monomethyl derivative. Specifically, we planned to follow the same

reaction sequence replacing the methyl group with a substituent that would effectively deter enolization of the carboxaldehyde functionality

at the triseco level, yet prove readily removable at a later stage.

Such a blocking group would necessarily be introduced via Sjj2 methodo­

logy at the tetraseco level and must survive the steps involved in

construction of the framework bond on the opposite molecular surface

as well as the process involved with adjusting the oxidation level

of the carbomethoxy group. Most importantly, this blocking group must not promote added photodecarbonylation of the triseco aldehyde, nor engage in capture of this photoexcited carbonyl group.

After much consideration, the decision was made to utilize a phenoxyjnethyl sidechain for these purposes. This sidechain could be

44 Introduced with the known electrophile, chloromethyl phenyl ether.

Furthermore, its removal at a later Btage seemed feasible through

Birch reduction of the phenyl ring and subsequent decarbonylation of the oxidized carbinol group. With these plans in mind, we set out to determine the suitability of such an approach.

Dissolving metal reduction of dichloro diester 37 followed by quenching of the resultant dianion with one equivalent of chloromethyl phenyl ether produced the tetraseco keto ester derivative 85 (48%). r\Aj As expected, a small amount (22%) of transannular hydroxy ester 8 6 was '\AJ also formed.

PhOCH

2. PhOCHaCl

PhOCH2

& &

Photocyclization of 85 proceeded without interference froqi the phenoxymethyl sidechain to yield the triseco alcohol 187 (90%). De­ hydration to 8 8 followed by diimide reduction gave the triseco ester

89 as anticipated (85% overall). Adjustment of the oxidation level of i\Aj the carbomethoxy group was accomplished as previously described in the monomethyl case. Thus, reduction to 90 and subsequent oxidation

PhOCH^ ? 0^-*0 PhOCHj ^ P h O C H ^

hv (TsOH) C«H«, A*

85 87 'UX,

HaNNHa, H aOa

PhOCH2 PhOCH2 /, CH2OH PhOCH^ ^

PCC (IBu)aAlH «------CHaCla c 6h 6

90 89 & OA/ 'W.

The suitability of the phenoxymethyl sidechain as a blocking group was further confirmed upon irradiation of 91. Not only did the phenoxy-

tion was not enhanced by this a substituent as evidenced by the rela­

tively good yield of diseco alcohol 92 which was obtained (36%). As r\y\j before, some photoreduction to 90 was also observed. AAi

HO PhOCH2 >**CHgOPh

hv

91 92 ' V b

It was now appropriate to direct our attention to the removal of the phenoxymethyl sidechain which had served us so well. Birch reduc­ tion of 92 delivered a dihydrobenzene product aqueous hydrolysis of which furnished 94 in 99% yield. This diol underwent efficient

HO '^CHgOPb ^ h 20 - O + Li, NKS H 30 ------i CaHsOH

92 93 94 'XA j 'XA j 'XA j / PCC, /PCHaCla / C

KOH C 2Hs0H

96 95 'XA j

Photocyclization of 96 proceeded without complication and gave rise to the seco alcohol which was dehydrated to olefin Compound represents an isomer of dodecahedrane which could potentially rearrange

to this species under suitable conditions. As in the case of the mono­

methyl olefin, however, treatment of 98 with trifluoromethanesulfonic

neither dodecahedrane nor any other single product was isolated from

the reaction mixture. Attempts to effect the desired isomerization with other agents (BF3 »0Et2; heat; ClPh(PPh3)3; RhCl3 *3H20; HF; H d O * ;

and FSO3H) were uniformly unsuccessful.

The known tendency of photoexicted olefins to undergo rearrange-

68 meiit led us to speculate that dodecahedrane might well be formed by

such a process. Furthermore, the ultraviolet absorption spectrum of

(t = 2 1 0 nm) suggested that photoexcitation of this strained olefin would be possible. In actuality, irradiation of 98 in hexane i\j\i solution through quartz with a 450W Hanovia lamp led to rapid formation of a product mixture consisting of at least two components (gas chromatographic analysis). Although these components remain as yet un­

identified, dodecahedrane was clearly not present (NMR analysis).

Having failed in effecting the isomerization of 98 to dodecahedrane,

it was decided to make use of the methodology developed in the synthesis of monomethyldodecahedrane to install the final bond. Diimide reduction of 98 led to secododecahedrane 99 (65% overall from 96). Upon

$ 250 °C for 4.5 hours or more, 99 was transformed with 40-50% efficiency f\A j into dodecahedrane 4. The success of this reaction was clearly evident from the *H NMR spectrum of the product mixture which exhibited a sharp singlet as its predominant feature. Dodecahedrane had become a physi­ cal reality for the first time!

Analysis of the product mixture by capillary gas chromatography showed the presence of several components. As in the monomethyl case, dodecahedrane was readily separated from the reaction mixture by re­ crystallization from benzene in 34% yield.

99

As fully expected, the and 19C NMR spectra of 4 (in CDC1S) are characterized by singlets, the former at 6 3.38 and the latter at 66.93 ppm. The 1 SC-H coupling constant of 134.9 Hz is somewhat larger than 69 the value earlier calculated by Mislow (128.1), but entirely com- 50

parable to those of the dimethyl derivative (131.2, 135.0). The vi­ brational frequencies exhibited by this 1 ^ symmetric molecule ( 1 2 0 identity operations) agree fully with a highly rigid network of inter­ linked methine units. Three infrared active bands are observed at

2945, 1298, and 728 cm-1; its eight Raman active frequencies occur at

2954, 2938, 1324, 1164, 1092, 840, 676, and 480 cm'1.70In general, these

7 1 findings compare reasonably well with values calculated by Ermer. The hydrocarbon gives no visible evidence of melting at temperatures up to

72 450 °C.

The successful synthesis of dodecahedrane 4 represents the culmi­ nation of several years of intense research. It also marks the beginning of an investigational era designed to probe the properties of this topo­ logically unique structure. These investigations, which appear to be as limitless as Plato's universe, should contribute significantly to a better understanding of the chemical systems represented by EXPERIMENTAL SECTION

Proton magnetic resonance spectra were obtained with Varian T-60,

EM-390, and Bruker WP-200 spectrometers) apparent splittings are given in all cases. 13C NMR spectra were recorded on Bruker WP-80, WP-200, and WM-300 spectrometers. Infrared spectra were determined on a Perkin-

Elmer Model 467 or a Nicolet Model 7199 instrument. Mass spectra were recorded on an AEI-MS9 spectrometer at an ionization potential of 70 eV.

Elemental analyses were performed by the Scandinavian Microanalytical

Laboratory, Herlev, Denmark. Melting points are uncorrected.

2-(Methylthio)tricyclo[4.2.1.03 ,9 ]nona-4,7-diene-2-carboxylic Acid (34 )

A solution of dry diisopropylamine

(0.26 mL, 1.85 mmol) in 3 mL of dry

tetrahydrofuran under an argon atmos­

phere was cooled to 0 °C with stirring. HOaC n-Butyllithium (1.32 mL of 1.4 M in hexane, 1.85 mmol) was added and the resultant mixture was stirred at

0 °C for 30 min. A solution of 29 (100 mg, 0.62 mmol) in 2 mL of dry tetrahydrofuran was introduced dropwise from a syringe. The mixture was then warmed to 50-60 °C and stirred for 2.5 h. After re-cooling to

51 52

0 °C, dimethyl disulfide (0.1 mL, 0.1 mmol) was introduced and the mixture was allowed to warm to room temperature over 2 h and poured into ice water. The layers were separated and the aqueous phase was extracted with ether (2x). The combined organic layers were washed with water and saturated sodium bicarbonate solution. The aqueous portions were acidi­ fied with 3 M hydrochloric acid and extracted with dichloromethane (3x).

The combined organic extracts were washed with water, dried, concentra­ ted, and applied to a preparative TLC plate (silica gel). Elution with

25% acetone in hexane furnished 130 mg of ^ (100%); mp 140-152 °C dec;

IR (KBr, cm"1) 3000, 2920, 1690, 1305, 1285, and 750;NMR (6 , CDC1S)

11.0 (br s, 1H), 6.2-5.9 (m, 2 H), 5.85-5.6 (m, 2 H), 4.1-3.2 (series of m, 4 H), 2.2 and 1.95 (s, 3 H); m/e calcd (M+) 208.0558, obsd

208.0562.

Tricyclo[4.2.1.09 ,9 ]nona-4,7-dien-2-one (19). /\AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/WVAA/W\Ai A mixture of 34 (433 mg, 2.08 mmol)

and anhydrous sodium bicarbonate (524 mg,

6.24 mmol) in dry methanol (30 mL) was

stirred for 30 min and treated with N-

chlorosuccinimide (694 mg, 5.2 mmol) in portions over 15 min. The resulting mixture was stirred at 25 °C under argon for 18 h, poured into 1 N aqueous sodium sulfite solution, and extracted with ether (3x). The combined organic layers were concentra­ ted to- a volume of 5-10 mL and treated with an equal volume of 5% aqueous hydrochloric acid. The two-phase system was stirred at 25 °C for 6 h, the layers were separated, and the aqueous layer was extracted 53 with ether (2x). The combined organic layers were dried, filtered, and concentrated to yield an oil which was subjected to preparative TLC

(silica gel). Elution with 30% ether in hexane yielded 21 mg of

(32%).

Pure 19 was obtained by preparative VPC (5% SE-30, 90 °C), mp 27.5- rV \ j 28 °C (sealed tube); IR (neat, cm”1) 3060, 3000, 2940, 1770, 1190, 1040,

999, 810, and 730; XH NMR (6 , CDC13) 6 .1-5.9 (m, 2 H), 5.6-5.4 (m, 2 H), and 4.4-3.7 (series of m, 4 H); m/e^ calcd (M+) 132.0575, obsd 132.0579.

Anal. Calcd for C«Ha0: C, 81.79; H, 6.10. Found; C, 81.65; H, 6.20.

Intermolecular Coupling of 'VWVWWWWWWWWWVWWWU A solution of dry diisopropyl­

amine (0.28 mL, 2.0 mmol) in 4 mL

of dry tetrahydrofuran under an

argon atmosphere was cooled to 0 °C

C02 CH3 with stirring. n-Butyllithium

(1.18 mL of 1.7 M in hexane, 2.0 mmol) was added and the resultant mixture was stirred at 0 °C for 30 min. A solution of 30 (35 mg, 0.20 OA, mmol) in 4 mL of dry tetrahydrofuran was cooled to 0 °C and treated in a dropwise fashion with the previously prepared solution of lithium diisopropylamide (0.6 mL, 0.2 mmol). The resulting solution was stirred at 0 °C for 2.5 h. Water was added and the aqueous portion was extracted with ether (3x). The combined organic layers were washed with brine (2x), dried, and concentrated to leave a yellow oil which was filtered through a short silica gel column (50% ether in hexane elution) to give 24 mg of 33 (6 8 %); IR (neat, cm-1) 3060, 2980, 1745, 54

1700, and 740; *H NMR (6 , CDC1S) 6.1-5.9 (m, 4 H), 5.8-5.4 (m, 4 H), 4.0-

3.2 (series of m, 6 H), 3.60 (s, 3 H), and 3.59 (s, 3 H); m/je calcd (M*)

320.1412, obsd 320.1419.

Reductive Coupling of 19^ 'VAAAAAAAAAAAAAAAAAAAAAAAAi To a solution of mercuric chlo­

ride (28.2 mg, 0.104 mmol) in 1.5 mL

of dry tetrahydrofuran was added 70-

OH OH 80 mesh magnesium (95 mg, 3.9 mmol),

and the resulting mixture was stirred at room temperature under argon for 0.5 h. The supernatant was removed by syringe and the remaining amalgam was washed with three 2-mL por­ tions of tetrahydrofuran. The amalgam was treated with 3 mL of tetra­ hydrofuran and the mixture was cooled to - 8 °C. Titanium tetrachloride

(214.3 yJi, 370.5 mg, 1.95 mmol) was added dropwise, followed by a solu­ tion of 19 (170mg, 1.30 mmol) in 1 mL of tetrahydrofuran. The mixture rV \ j was warmed to 0 °C and stirred for 2 n prior to quenching with 1 mL of a saturated potassium carbonate solution. After being stirred for an additional 15 min at 0 °C, the reaction mixture was filtered through

Celite and the blue residue was washed several times with dichlorome- thane. The filtrate was washed with brine, dried, and concentrated to afford 160 mg of a yellow oil. Purification by preparative TLC on silica gel (20% ethyl acetate in hexane elution) provided 35 mg of 35 (20%), mp 157.5-161 °C (from hexane); IR (KBr, cm"1) 3540, 3480, 3050, 3030,

2930, 2910, 1275, 1250, 1120, 840, 750, and 735; *H NMR (6 , CDC1S) 6.3-

6.1 (m, 4 H), 5.7-5.5 (m, 4 H), 4.2-3.0 (m, 8 H), and 2.4 (br s, 2 H); 55

**C NMR (ppm, CDCI3 ) 141.14, 129.07, 72.44, 59.34, 51.45, and 41.56;

m/e calcd (M+) 266.1307, obsd 266.1312.

Octadecahydro-7-(hydroxymethyl)-3H-cyclopenta[3,4]pentaleno[2,1,6- 'VA/WWWWV/WWWWWVWWWVWWWWWWWWWWWWWWWVWV/Wb gha] pentaleno [ 1,2,3-cd]pentalen-3-one (55), 0ctadecahydro-3-hydroxy-lH- ^Am/wwv/wwvAMAAAAn/wvwwwwvwwwi^Wwwwwwwwwwvwwvwm cyclopenta[3,4]pentaleno[2,1,6 -gha]pentaleno[1,2,3-cd]pentalene-7- /VWVWWWVAAAAAAAAnAAAAnAAAAA/WWVWWWVnAAAAAnA/WWWWVWWWU methanol (36), and Methyl Hexadecahydro-9-hydroxy-l,4,8-methenodipenta- 'VnAAAAAnAAAAAAAAAnAAn/WVWV/WWWWWWWWVnnAAAAAAA/WWWWVAnAAnnAAAA, leno[1,2,3-cd: 1 ’,2',31 -gh]pentalene-4(1JH)-carboxylate (54) . WWWWIAWUWUWVWWW\AAAA/W\AA/WWVW\/WWW\/WWWWV

A solution of lithium (125 mg, 18.1 mg-at) in 60 mL of liquid ammonia

(freshly distilled from sodium) was cooled to -80 °C (dry ice/ether) under argon. A solution of 37 (500 mg, 1.18 mmol) in tetrahydrofuran (5 mL) was r\T\j added dropwise over 2 min. After 8 min, a solution of 25% tetrahydrofuran in ethanol (.625 mL) was added rapidly and the resulting mixture was stir­ red for an additional 2 min at -80 °C. Solid ammonium chloride was added rapidly to discharge the blue color and the ammonia was evaporated. The residue was added to water (100 mL) and extracted with dichloromethane

(3 x 50 mL). The combined organic extracts were washed with water (50 mL), dried, and filtered. Evaporation of the solvent in vacuo gave a clear oil which was subjected to preparative TLC on silica gel (10% ether in dichloromethane elution). At Rf - 0.5, hydroxy ketone 55 was obtained: 186 mg (53%). Recrystallization from ethyl acetate gave pure fc, mp 165-170 °C; IR (KBr, cm”1) 3480, 2940, and 1710; XH NMR (6 , 56

CDClg) 4.0 (d, * 7.5 Hz, 2 H) and 3.8-1.1 (series of m, 24 H); 13C

NMR (ppm, C sD3N) 225.23, 66.69, 61.38, 60.94, 59.10, 57.15, 55.56,

54.49, 53.73, 53.16, 52.59, 51.33, 51.20, 50.70, 48.17, 35.46, 31.92,

30.60, 29.96, and 28.13; m/e calcd (tf4-) 298.1933, obsd 298.1927.

Anal. Calcd for C 2 0 H 2 6 O 2 : C, 80.50; H, 8.78. Found: C, 80.45;

H, 8.85.

At Rf = 0.45, diol 56 was obtained (amounts varied from run to run).

Recrystallization from ethyl acetate gave the analytically pure material, mp 230-232 °C; IR (KBr, cm"1) 3600, 3430, 3350, 2930, and 1020; XH NMR

(6, CSDSN) 5.6 (s, 1 H), 4.3 (m, 2 H), and 3.4-1.0 (series of m, 25 H);

13C NMR (ppm, CsD5N) 79.47, 66.66, 61.75, 61.03, 59.72 (2C), 58.94,

58.36, 53.74, 52.82, 52.48, 51.80, 51.17, 50.54, 48.06, 34.71, 31.80,

30.58, 28.84, and 24.56; m/£ no M 4" observed, 282 (M+-H20).

Anal. Calcd for C 2 0 H 2 BO2 : C, 79.95; H, 9.39. Found; C, 79.77;

H, 9.36.

At Rf = 0.8, hydroxy ester ^ was obtained (minor amounts which varied from run to run). Recrystallization from ethyl acetate gave the analytically pure material, mp 75-76 °C; IR (CDC1S, cm“x) 3450, 2940, and

1690; *H NMR (6, CDC1S) 5.1 (s, 1 H), 3.7 (s, 3 H), and 2.9-1.1 (series of m, 22 H); 13C NMR (ppm, CDC19) 178.37, 80.11, 60.54, 60.20, 59.96,

56.80, 56.17, 52.58, 51.80 (2C), 51.41, 49.37, 48.55 (2C), 48.26, 32.58,

27.82, 25.78, 25.49, and 24.52 (one signal not observed); m/e calcd (M4-)

326.1882, obsd 326.1888.

Anal. Calcd for CaxHaeOs: C, 77.27; H, 8.03. Found. C, 77.22;

H, 7.98. 57

Octadecahydro-3b-hydroxy-l,6 -methanocyclopenta[3,4]pentaleno!2,1,6 -cde]- 'VVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/ pentalenof 2,1,6 -gha]pentaleno-7-methanol (57).

butyl alcohol In benzene (5 mL). Three

drops of triethylamlne were added and

the mixture was irradiated for 16 h through pyrex with a 450W Hanovia lamp. The crystalline material which had formed was filtered and dried to yield 37 mg (76%) of pure 57, mp

298.1933, obsd 298.1940.

Anal. Calcd for C 2oH260 2 : C, 80.50; H, 8.78. Found: C, 80.21;

H, 8.80.

1,la,lb,2,3,3a,4,5,5a,6,6a,6b,6c,6d,6e,6f-Hexadecahydro-1,6-methanocyclo- OAAAAAAAAA/XAAAA/VXAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/WVAAAAAAAAA; penta[3,4]pentaleno[2,1,6-cde]pentaleno[2,1,6-gha]pentalene-7-methanol (58), 'V\AAAAAAAAAAAAAAAAAAAAAAAAAA/WWWWWWWVWW\AAAy\AAAAA/\AAAAAAAAAAAAAV\AAA> To a suspension of diol 57 (80 mg,

a single crystal of £-toluenesulfonic

acid. This mixture was heated to 60 °C

for about 1 h until complete dissolution occurred, cooled, concentrated in vacuo, and applied directly to a pre­ parative TLC plate (silica gel). Elution with 50% ether in hexane gave the desired product: 73 mg (95%) at * 0.6. Recrystallization from ethyl acetate-hexane gave pure 58, mp 300 °C dec; IR (KBr, cm” 1) 3370, rV\j 2940, and 1005; *H NMR (6, CDC19) 4.1 (d, - 7.5 Hz, 2 H ) , 3.7 (m, 58

2 H), and 3.6-1.3 (series of m, 20 H); 19C NMR (ppm, CDC13) 139.72, 137.34,

74.96, 72.24, 70.35, 64.42 (2C), 61.56, 54.23, 53.70, 53.45, 51.70, 50.25,

49.08, 48.84, 46.75, 30.25, 28.74, 27.53, and 24.03;-m/^ calcd (M*)

280.1827, obsd 280.1835.

Anal. Calcd for C3oH2<.0: H, 85.67; H, 8.63. Found: C, 85.44)

H, 8.79.

Octadecahydro-1,6-methanocyclopenta[3,4]pentaleno[2,1,6-cde]pentaleno- 'UWWWUWWWWVWWWWWWWAAAAAAAAAAAAAAAAAAAAAAAAAAAA/VWWWIAAA/ [2,1,6-gha]pentalene-7-methanol (59). WUVlAA/VWWWWWVA/mA/WWlWWm Hydroxy olefin 58 (57 mg, 0.20 mmol) VO was dissolved in 4 mL of an ethanol-

tetrahydrofuran solution (5:1). The solu­

tion was cooled to 0 °C and anhydrous

hydrazine (352 y£, 10.4 mmol) was added,

followed by the dropwise addition of cold 30% aqueous hydrogen per­

oxide (1.1 mL of 30%, 10 mmol) over a period of 1.5 h. The tempera­

ture was increased gradually to 25 °C and stirring was continued for 8 h.

The mixture was added to water (10 mL) and extracted with dichloromethane

(3 x 10 mL). The combined organic extracts were washed with water (2 x

10 mL), dried, and evaporated to give 57 mg (100%) of crude jfy. Prepa­

rative TLC on silica gel (50% ether in hexane elution) gave the pure

alcohol at Rf «* 0.3; mp > 250 °C; XH NMR (6, CDC13) 4.28 (d, *= 7.5

Hz, 2 H) and 3.7-1.4 (series of m, 24 H); 13C NMR (ppm, CDC13) 80.69,

70.25 l 67.09, 65.49, 61.51, 58.45, 55.39, 51.70, 51.56, 51.46, 31.56,

and 31.02; m/e calcd (M+) 282.1984, obsd 282.1988. 59

Anal. Calcd for CaoHaeO: C, 85.06; H, 9.28. Foundi C, 84.66

H, 9.54

Oxidation of ^ with Pyridinium Chlorochrornate. 0ctadecahydro-7- 'VAAAnAAAAAAAAAAAAAAAAAAAAAAAA/WVAAAAA/WWWWVWV/WUWWWWVW^ methylene-1,6-methanocyclopenta[3,4]pentaleno[2,1,6-cdel pentaleno- 'WWWVAAAAAAAAAAAAAAAAAAAAAAAA/WVWVWWVWWWWWWWWWWVAAA^ 12,1,6-gha]pentalene (63) and a,p-Unsaturated Aldehyde (£2)- 'WVWWWWWWWWVAA/^bwWWVWWVWWVAAAAAAAAAAAAA/VAAAAA/

CHO

To a suspension of pyridinium chlorochromate (14 mg, 0.064 mmol) in dry dichloromethane (1 mL) was added a solution of alcohol 59 (12 'VU mg, 0.042 mmol) in dichloromethane (2 mL). After 30 min, ether (5 mL) was added and the organic solution was decanted from the salts, concen­ trated, applied directly to a preparative TLC plate (silica gel), and eluted with 50% ether i n hexane. The band at Rf = 0.8 consisted of the olefin (6 mg, 55%); *H NMR (6, CDC13) 4.90 (br s, 2 H) and 3.6-1.5

(series of m, 22 H) ; m/e^ calcd (M+ ) 264.1878, obsd 264.1885.

A second band (Rf = 0.6) consisted of cr,p-unsaturated aldehyde

as a white crystalline solid; IR (KBr, cm” 1) 2940, 1665, 1615, and 1385; XH N M R (6, C D C 1 S) 10.2 (s, 1 H) and 4.0-1.4 (series of m,

21 H); X9C NMR (ppm, CDC1S) 186.97, 172.96, 137.80, 69.97, 68.70, 67.42,

63.11, 57.96, 57.41, 54.92, 53.95, 51.83, 51.10, 49.64, 48.97 (2C),

33.68, 30.95, 29.98, and 28.28; m/e calcd (M+ ) 278.1671, obsd 278.1680. 60

Octad ecahydro-l,6-methanocyclopenta[3,A]pentaleno[2,1,6-cde]pentaleno- •wwvwwvvn/v/wwwwwwvwwvwwwwwwwvA/wwwwwwwwwww^ [2,1,6-gha]pentalen-7-one ($i). 'WWWWWWWVWWWWWWWAA To a cold (0 °C) solution of

(90 mg, 0.32 mmol) in 6 mL of acetone and

2 mL of tetrahydrofuran was added 15

drops of stock Jones reagent solution

(from 200 g of sodium dlchromate di­

hydrate, 272 g of concentrated sulfuric acid, and 600 mL of water) during

a 10 min period. After 30 min, excess oxidant was quenched by addition

of isopropyl alcohol. The resulting mixture was added to water and ex­

tracted with dichloromethane (3 x 20 mL). The combined organic ex­

tracts were washed with saturated sodium bicarbonate solution, water,

and brine prior to drying and solvent evaporation. There was ob­

tained 43 mg (51%) of crystalline Preparative TLC purification on

silica gel (elution with dichloromethane-ether 9*1) gave colorless

crystals, mp 218-222 °C (from ethyl acetate); IR (KBr, cm” 1) 2920, 1710,

and 1445; *H NMR (6, CDC13) 4.0-0.8 (series of m, 22 H); 13C NMR (ppm,

CDCls) 227.45, 68.93, 65.60, 60.50, 58.86, 57.59, 53.52, 51.52, 49.88,

30.03, and 29.25; m/e calcd (M+) 266.1671, obsd 266.1679.

Manganese Dioxide Oxidation of ^5. VXAA/WWWWVVWWWUVWVWVWVWV

CHO CHO 61

A solution of (48 mg, 0.17 mmol) In degassed dichloromethane

(12 mL) was treated while stirred with 600 mg of commercial manganese

dioxide (Alfa Ventron, 94% purity). Stirring was continued at room

temperature for 18 h, at which time the reaction mixture was filtered

through a Celite pad. The retained solids were washed with copious

amounts of ethyl acetate-dichloromethane (1:1). Concentration of the

combined filtrates afforded an oily solid which was directly subjected

to preparative TLC on silica gel (elution with ether-petroleum ether,

1:1). Two major bands were observed.

(a) Rf = 0.55; 20.2 mg (42%) of the aldehyde dienes ^ as a

crystalline mixture of isomers, mp > 250 °C; IR (CDC13, cm-1) 2940,

1640, and 1600; lH NMR (6, CDC13) 9.94 (s, 1 H) and 4.0-1.5 (series

of m, 19 H); 13C NMR (ppm, CDC19) 191.71, 187.95, 185.70, 180.24, 175.02,

142.92, 142.43, 141.28, 139.52, 71.67, 71.49, 71.12, 69.61, 69.18, 63.78,

61.41, 57.35, 56.07, 54.19, 53.77, 53.16, 52.49, 48.00, 46.97, 46.49,

46.00, 31.74, 31.19, 30.71, 28.70, 28.34, 25.67, 22.64, and 20.57; m/e calcd (M+) 276.1514, obsd 276.1524.

(b) Rf = 0.60; 21.5 mg (45%) of 66 as a colorless crystalline 1 'Vu solid, mp 181-183 °C; IR (CDC13, cm-*) 2940 and 1710; XH NMR (6, CDC13)

9.42 (s, 1 H) and 4.0-1.5 (series of m, 21 H ) ; l3C NMR (ppm, CDC13)

204.69, 141.64, 140.25, 72.70, 69.43, 64.08, 62.81, 62.57, 59.29, 55.41,

54.98, 52.07, 51.40, 51.04, 48.00, 46.18, 29.74, 29.43, 25.85, and

23.24, m/e calcd (M+) 278.1671, obsd 278.1680.

Dissolving Metal Reduction of WWWWAAAAAAAAAAA/WWWWWVWb To 5 mL of liquid ammonia (freshly distilled from sodium) was added mnpqapi aqa J33JV 'a-ifA ®npqapi 3° (3 a-3 m £'g) 3m pappa sba 0 o gi~ 0 3

papooa (mnppos moap papppaspp ^pqsaap) apuorama PTn^TT J° 1® 09 °I CH3

-£H3 0 OeH3 0 H3

VV^/XAAA/VWVWVVVWVWWWWV/VV/WI/WWVVWWVWWVWWVi • (0L) aaepXxoqaao-(HI)^-auapaauad[qS--,£*,3*,1: P^-E*1*1] VWWWWVWWWWWWWWWVWVWVWWWVWWVWWWVn/WWVWVWWVVi -ouaxa3uadxpouaq33m-8* V‘l-T^ll^9®“8~^X0-IP^ll_6~0:lP^llB;:,aPBX9H T^q^aR pua OAAAAAAaAAAAAAAAAATVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/WWWVi (xfV) aaaTKxoqaeo-/-auapeauad(po-£‘Z* IJouaxaauad[aq§-9 *T*Z]ouapaauad WA/WVVWI/WVWWIA/WVWVVVVVWWVWWWWVWWWWVWVVVWWWWW; -[*l*Z la^uadoxaXo-HX-oxo-C-T^q^3®-^-0-1? ^ 509?9^0© I^q39R J° spsaqauXg VVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/VVVVVXAAAAAAAAAAAAAAAAAAAAAAAi •Bpuoramy ppnbpx up nmpqapi q^TA ^ 3° uopaeiiCqaaw-uopaezpio^o aApaortpan

•(H £3 ‘® 3 ° sapaas) 1 ’1 -9 ‘E

puH (h X ‘s) ST*9 (eI0a0 *9) HRN H t {0*8 P«® ‘0*31 *0£6Z *009Z (r-m9

*6 I0a0) HI ipxxos aapqA aApapsuas apa ua sa x^ 3 0 (%00I *3® 0E) PTT°s

aapq» a 3 ab3 onoaA up auaApos aqa 3 0 uopaaaodaAa *paaaaXP3 PUB P9 T'1P

‘aaaeA qap» paqsa» * (pm OS) auBqaaraoaopqopp °3 pappa sba anppsaa aqx

•paaaaodBAa sba epuomme aqa pua aopoD anxq aqa a3 aaqospp oa pappa saw

appaopqo mnpuorame pppos *3 upaapas 3 0 upm o£ peuopapppB U9 J" J V ’upm

01 aaAO aspadoap pappB sba (pm 3) uaanpoap/Cqeaaap up (Xomm 11*0 ‘8ra 3 *g)

loqoopa i/Canq-ajaa pua (xomm IT'O ‘3m

0E) Yty 3° uopanpos y ‘paAposspp paq

mnpqapi 9M3 ipaun uo3aa aapun aanaaaad

-raaa xnppaa aqaaa paaapas s b a aanaxpm

aqi *aapA mnpqapx30 (ae-3m 33*0) 3® 5

39 63

had dissolved, a solution of 37 (390 mg, 0.92 nmol) in anhydrous tetra- OAi hydrofuran (5 mL) was added dropwise until approximately 95% of the di­

ester had been added. After 0.5 h at -78 °C, the remainder of ^ was

added dropwise (the final addition should be discontinued at the point where the deep blue color begins to dissipate). A solution of methyl

iodide (131 mg, 0.92 mmol) in 2 mL of dry tetrahydrofuran was im­ mediately added in one portion. After 15 min, solid ammonium chloride

(ca 500 mg) was introduced and the ammonia was evaporated under a

stream of argon. The residue was added to dichloromethane (250 mL)

and washed with water and brine. Concentration of the dried organic phase left a clear oil which was subjected to preparative TLC on silica

gel (10% dichloromethane-15% ether-75% hexane elution). Three bands were observed:

(a) Rf = 0.6; there was isolated 75 mg (24%) of 7^), mp 108-109 °C

(from ethyl acetate); IR (KBr, cm-1) 3480, 2950, 1700, 1190, and 1115;

XH NMR (6 , CDCls) 5.25 (s, 1 H), 3.60 (s, 3 H), 3.2-0. 8 (series of m,

21 H), and 1.15 (s, 3 H); l9C NMR (ppm, CDC19) 178.71, 82.29, 60.88,

60.44, 59.52, 58.94, 58.36, 57.68, 57.04, 56.32, 51.46, 50.88, 49.47,

48.31 (2C), 47.77, 32.77, 26.85, 26.02, 25.54, 24.22, and 19.47; m/e calcd (M+) 340.2038, obsd 340.2046.

Anal. Calcd for Ca2H 2 e03: C, 77.61; H, 8.29. Found* C, 77.51}

H, 8.30.

(b) Rf “ 0.3} there was obtained 144 mg (46%) of 71, mp 141-142 °C

(from-ethyl acetate)} IR (KBr, cm-1) 2938, 1725, 1265, 1138, and 1099}

XH NMR (6, CDCls) 4.0-0.8 (series of m, 22 H), 3.60 (s, 3 H), and 1.38

(s, 3 H)-, X9C NMR (ppm, CDCls) 227.26, 176.83, 65.17, 59.29 (2C), 59.16, 64

57.34, 56.37, 54.67, 54.01, 53.10, 52.73, 50.91, 50.67 (2C), 38.89,

35.19, 30.94 (2C), 30.34, and 27.73 (22nd signal not observed and may

overlap); m/e calcd (M+) 340.2038, obsd 340.2047.

Anal. Calcd for C2 2 H 2 b0 9 * C, 77.61; H, 8.29. Foundt 77.40;

H, 8.30.

(c) The material having an R{ = 0.1 (4.5 mg) was not identified.

Methyl 0ctadecahydro-3b-hydroxy-7-methyl-l,6 -methanocyclopenta[3,4]- 'WVWVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/VWVWWVWWVAA; pentaleno[2,1,6 -cde]pentaleno[2,1,6 -gha]pentalene-7-carboxylate (72). 'WWVWWWWVWWVVAn/WWWVWWWWWWWWWVWVWVWWVWWWVWU A solution of 7^. (100 mg, 0.29

mmol) in 10 mL of a dry deoxygenated

benzene-tert-butyl alcohol solvent sys­

tem (4tl) was irradiated with a 450 W

Hanovia lamp through pyrex for 16 h

under a nitrogen atmosphere. The resulting pale yellow solution was

concentrated in vacuo to leave 1 0 0 mg of which although not further

purified proved to be quite free of contaminants (TLC analysis); IR

(KBr, cm-1) 3512, 2940, 1715, 1263, and 1132; m/e calcd (M+) 340.2038, obsd 340.2046.

Methyl l,la,lb,2,3,3a,4,5,5a,6,6a,6b,6c,6d,6e,6f-Hexadecahydro-7- 'WWVWUWWVWUWWWVAA/WAAAA/WWWWVWWAAA/WWVWWWVWVWV/ methyl-1 ,6 -methanocyclopenta[3,4]pentaleno[2 ,1 ,6 -cde]pentaleno[2 ,1 ,6 - VWWWlAAAA/WWWWWVWWWVWVWVWWWWWUWWWWWWWWVUVm gha]pentalene-7-carboxylate (^)- 'VAAAAAAAAAAAAAAAAA/UWAAAAAAAAAAAAy A solution of (100 mg, 0.29 c ^ o y j mmol) and jg-toluenesulfonic acid ( 1 0

mg) in 15 mL of dry benzene was heated

at the reflux temperature with continuous HHN D6t *(H E *9) V£*I Pu® ‘(H £ *9 ) g9 ‘£ ‘ (h ZZ *® J° saqjias) 8*0-3*V

(6 X0G3 *9) HHN H t *9011 P“ 8 ‘X93T ‘OEIT ‘5363 (T- ^ **HH) HI * (aJeq

- OAj -ao® x^qqa mojj) o„ 5*9£I-££I dm *PTT° 8 ssajaoxoo b sb pauqBqqo sba \ji

ajna • (uoqqr.xa auBxaq %0I--*3 9Ja %0Z-9UBqaamoaoxqoTP %0I) 1 ^ 8 BOfifS

uo oil ^9 P9TJT^nd 9 BA qoqqA PTIos auTHB^sXaoTraas b j o 3m OOT pap^oj

-jb uojoBjodBAa quaAXOS puB assqd otubSjo aqq jo 3uq£.ia ‘auqaq pus

(X£) J33BA qqpi paqsBA PUB auBqrjaraoJOxqoiP oquq paanod sba uoqqnxos aqq

*0o 53 9 3T Pue Do 0 ^ q 3 *aJJV *8 uTj;iT:qs qqqA ujm qt? jo pojxad b

x o a o ajnqxqm uojjoeax pajooo aqq oq asjrtdojp paonpo-iquq sba (qm VI *3

*%0 E) apTxo~ad u?3 ojdjCh • (“xri 009

*%L6 ) aujzBjp/Cq snoapXquB pappB sba

(qsg) ajnqxqm uBjnjoap/CqBjqaq-xouBqqa

ub jo qm OX UT (lomm I£*0 ‘3m OOT)

XL jo uoqjnxos (0 o o) PI0 3 8 oq '\AAAAAAA/WWVWWWWW\AAAAAAAAAAAAA/VVVUVWWVWWVV< • (V^) ajBX^xoqjBO-^-auax®quad[eqS-g‘X ‘3 ]ouaqaquad[apo O/VVAAAAAAAAAAAOAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi -9*T‘3]°uaI8J8ad[v‘£]BJuadoxo^oouBqqam-9‘x-I^llJ3 ®-Z.-o;iP^H8oapBjoo x^9J9W

T V 6 I T 3 E P9 9 0 ‘EE6T*33E (+H) P3!9^ a/m *2Z*V3

PUB ‘01*83 *65*83 ‘63*0£ ‘I0*TV *68*9V *05*8V ‘£6 *8 V ‘18*05 ‘35*35 ‘8 E T S

‘VI*85 ‘I V 6 S ‘31*39 ‘IT*£9 ‘SV*E9 ‘19*69 ‘8£*3I ‘56*9£T ‘0£*0VT ‘OETIT

(eX0 a0 *®dd) XIWN 0 st * (H £ *9 ) O V T P“ 8 * (H E *9) 89*E * (H 03 *« jo sajxaa) 8*0-I*V (eTDCD *9) HHN Hx *8311 PUB ‘V53T ‘8311 ‘5183 ‘0V63 (t- ® 3 r\f\j **9M) HI *(auHxaq moaj) o„ SCT-EEI dm '£i j o (%Z6) 8 ra 9 3 papjojjs qai^a*

-am sqqq jo (auaxaq uq jaqqa %gq qqqA uoqqnqa) xa3 Boqqqs uo uoqqsoqj

-qjnd 3TX aAqqBJBdaaj ‘SuqpuBqs uo pazqxxajSiCiO qoqqA xi° Aoqqa^ B

aABax oq onoBA uq pajBJodBAa sba quaAqos aqi *q 9 x o j qaqaA jo x^Aoraaa

59 (ppm, CDCls) 177.39, 70.39, 66.51, 64.27, 59.66, 58.25, 57.19, 52.91,

50.92, 50.73, 49.66, 39.17, 31.07, and 30.44; m/e calcd (tff) 324.2089,

obad 324.2097.

Anal. Calcd for C2 2 H 2 8 02: C, 81.44; H, 8.70. Foundi C, 81.50;

H, 8.67.

Octadecahydro-7-methyl-l,6 -methanocyclopenta[3,4]pentaleno[2,1,6 -cde]- VWWVWVVWWWXAAAAAAAAAAA^WXnAAAAAAAAAAAnnAnnAA/WWWWVAAAAAAAAAAA* pentaleno[2,1,6 -gha]pentalene-7-methanol (75). 'WAAAAAAAAAAAAAAAAAAA/WXAAAAAAAAAAAAAAAAAAAAAi A solution of 74 (180 mg, 0.56 mmol) f\A j in anhydrous benzene (10 mL) was stirred

at room temperature while diisobutyl-

aluminum hydride (2 mL of 1 M in hexane,

2.0 mmol) was introduced. The reaction mixture was stirred at 25 °C for 3 h under argon, treated dropwise

with methanol to destroy excess hydride and poured into dichloromethane

(50 mL). The organic layer was washed with dilute hydrochloric acid,

saturated sodium bicarbonate solution, water, and brine prior to drying.

Solvent evaporation left 156 mg (94%) of 75, obtained as a white powder,

1000, and 778; lH NMR (6 , CDC13) 4.14 (s, 2 H), 4.3-0.8 (series of m,

23 H), and 1.11 (s, 3 H); ISC NMR (ppm, CDC1S) 70.54, 67.19, 65.44 (2C),

58.65, 57.43, 57.24, 52.00, 51.41, 50.73, 37.92, 31.31, and 30.88; m/e calcd (M+) 296.2140, obsd 296.2149.

Octadecahydro-7-methyl-l,6 -methanocyclopenta[3,4]pentaleno[2,1,6 -cde]- WA/VAA/WWWWWWWV/VAA/WWWWWWWWWWWWWWWWWAnAA/WWWW pentaleno[2,1,6 -gha]pentalene-7-carboxaldehyde (76). 'XAAAAAAAAAAAAAAAAAAAAAAAAAA/VWVWWWWWWWWWWb A solution of (160 mg, 0.54 mmol) in 1 mL of dichloromethane was 67

added to a suspension of pyridinlum I chlorochromate (200 mg, 0.93 mmol) in 5 mL of the same solvent vith stirring

under argon. After 1.5 h, ether was

added and the organic layer was decanted.

The residual salts were washed with ether (2x) and the combined organic solutions were washed with dilute hydrochloric acid, saturated sodium bicarbonate solution, water, and brine. Following drying and solvent evaporation, there was obtained 160 mg (1 0 0 %) of as a clear oil which crystallized on standing. Recrystallization from ethyl acetate at -10 °C gave pure ^ as white crystals, mp 173-176 °C; IR (KBr, cm” 1) 2920, 1710, and 1445; XH NMR (6 , CDC19) 9.92 (s, 1 H), 4.0-0.6 (series of m, 20 H), and 1.16 (s, 3 H); 19C NMR (ppm, CDC1S) 204.33, 70.21, 66.94, 64.75,

59.17, 57.59, 57.17, 52.61, 50.31, 49.88, 34.77, 31.50, and 30.52; m/e calcd (M+) 294.1984, obsd 294.1991.

0ctadecahydro-7f-methyl-l.6 ,7-metheno-lH-cyclopenta[3,4]pentaleno[ 2,1,6 - /VWWWWWWWVWWVWWWUWWUWWWWUWV/WWWVAnAA/WWWWWWVnA/ gha]pentaleno[1 ,2 ,3-cdlpentalen-2 -ol (^) . 'WVVA/VXAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAj A cold (-78 °C) solution of ^

(150 mg, 0.51 mmol) in 10 mL of a de­

oxygenated 9;1 toluene-ethanol solvent

system was irradiated under nitrogen

with a 450W Hanovia lamp through pyrex.

After a 4-h reaction period, the solution was allowed to warm to 25 °C and the solvent was removed under reduced pressure. Preparative TLC on silica gel (elution with 10% dichloromethane-15% ether-75% hexane) gave 68

three bands. The major band (R^ « 0.8) was a mixture of decarbonylated materials (60 mg) which were not characterized. The Rf • 0.4 band com­ prised alcohol 77 (38 mg, 25%), a colorless crystalline solid; IR (KBr, f\J\j cm"1) 3425 and 2920; XH NMR (6 , CDC1S) 4.11 (d, J - 6.2 Hz, 1 H), 4.0-

0.8 (series of m, 22 H), and 1.21 (s, 3 H); m / e calcd (M*) 294.1984, obsd 294.1991. The Rf = 0.3 band proved to be alcohol^ (30 mg, 22%).

0ctadecahydro-7f-methyl-l,6 ,7-metheno-2H-cyclopenta[3,4]pentaleno!2,1,6- 'V/WV\AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/V\AAA/V gha]pentaleno[1,2,3-cd]pentalen-2-one (78). OAAAAAAAA/WWWVWVVWWWXAA/XA/WWWVW^ A solution of 77 (38 mg, 0.13 mmol)

in dichloromethane (1 mL) was added under

argon to a stirred suspension of pyri-

dinium chlorochromate (35 mg, 0.16 mmol)

in 5 mL of the same solvent. After 2.5 h, ether (15 mL) was added and the organic phase was decanted. The residual brown salts were washed with ether (2 x) and the combined solutions were washed with dilute hydrochloric aciu, saturated sodium bicarbonate solu­ tion, water, and brine prior to drying and solvent evaportation. There was obtained 37 mg (98%) of 78 as a pale yellow oil. The pure sample was w obtained by preparative TLC on silica gel (elution with 10% ether in hexane) and sublimation at 125 °C and 0.1 mm. The colorless crystals gradually decomposed when heated above 180 °C with softening occurring at 255 °C, IR (KBr, cm"1) 2920, 1710, and 1445; *H NMR (6 , CDC13) 4.1-

1.0 (series of m, 21 H) and 1.21 (s, 3 H); l9C NMR (ppm, CDC1*) 229^76,

69.36, 68.15, 67.97, 67.72, 67.12, 65.66, 65.12, 62.93, 61.60, 59.35,

58.74, 53.34, 52.01, 51.83, 51.10, 49.94, 36.72, 34.10, 30.65, and 27.92; m/e calcd (*T*") 292.1827, obsd 292.1809. aanaxqm sqqx *PX3B oxuoxqnsauBqaauioaonxjx*} 3° sdoap 5 papps s b a (qm g)

OAj auBqaaraoaoqqoxp ax (qomm £0*0 * 8 0 02) 08 3° uoxanqos paaaxas b ox OAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/WWWVVU • auaxsiuadlBipr-g'x^louaxBauadtapo-g'x'Zlouaxs^uadl^^lB^uad

OiOAAAAAAAA/V\AAAAAAAAA/VXA/VUVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA>VVWVAj -oqaxCaatuCpxT^tZ ] uapxqXuBqaa[X] -g * qg * z * 9 * I-HI-T^M^am-pg-oapKqBaapBxaH

0 C 2 T*V2 2 psqo ‘T Z Z T * ^ (+H) poquo a/ra ifiZ’SX PBB ‘95*92 ‘*£*05 ‘9I*C€

‘00*2»7 ‘20*15 ‘*0*25 ‘22*25 ‘€8*65 ‘<75*09 ‘58*19 ‘62*£9 *66'C9 ‘69*59

‘€8*59 ‘19*99 ‘56*99 ‘€**29 ‘91*82 ‘20*0*1 ‘*6*5*T (sT0d0 ‘

0 et ! (HE ‘ s ) <72*1 P«e (H 61 ‘« 3° saqass) 6*0-6*£ (eT0CI0 *9) HHN HT

iIV82 PUB ‘0262 ‘TIOC (t-nra ‘Jffil) HI 5(uoxaBraxqqns araos qaqA pax«adtnoa

- o p ) jv Q55-022 uaaAaaq pasodmooap ^qqanpsaS 3 nq ‘Suxaqanx 3° aouapXAa

3ax3 aou pxp sqBas^ao ssaqaoqoo aqi *auBxaq moaj pazjXXBasiCaoaa aq

pxnoo qoxqA ^ jo (%t7S) Sox 05 pBasqosq s b a aaaqx * (auaxaq ux ausqa

-amoaoqqoqp %5 qax* uopanqa) qa3 b o j ITs uo uoxaaoxjxand oqx °3 paaoaf

-q n s s b a anpxsaa aqa p u b paasaodBAa s b a uoxanxos paxooo aqx *q5 a o j

aanaBaadtnaa xnqjaa aqa as paasaq s b a uoqanqos aqa pue ‘ papps s b a p ja s

oXU03Xnsauanxoa-3’ 3 0 qBasxtao b ‘ (qm g) auazuaq snoapKqua ux dn uaqsa

s b a xaiaaasm sqqx ‘SuqpuBas uo pazxxqsas/Cao qoqq« (Sm £g) qqo Aoqqa,

aqBd b sb 3a x 3 oa o u o b a up paAomaa s b a auaAXos aqx *aaaqdsomaB

uaSoaaqu B aapun xaaKd q3noaqa q 9X i°3

dmb x b j a o u b h rt 05* B paasppBaax

s b a (x1 *7) maasKs auaAxos qoqooqa lto nq

-aaaa-auazuaq paasuaSxtxoap b j o qm g ux

(qomrn gx*o ‘ 8 m £g) ^ 5 0 uoxanqoa y -VVXA/WWWVVVVVWWVVVVVVV *(3h) auaqBauad[sqS-g *q *Z ] 'WXZWWVXAA/VyWWWWWXA/VXAAZVWXA/WX/WWXAZXAAZXZVWWVWVXAAAAAZWXA/W -ouaqaauad [apo-g ‘ x * z louaqsauad [*7 ‘ g ]BguadoqoiCoauapxxXxpBUBqaa-g * z ‘ 9 * T 'WWWWWVWVWWWWWWWWIA/UWWWWVWWWWWWWVAA/WWWX/ -qXqaam-£-oapXqBoa pexan- 3 9 ‘ a9 ‘ P9 * 39 * q9 * B9 * 9 ‘ sg * g * ^«Bg * g ‘ jr * qq * b q*- T was stirred at 25 °C under nitrogen for

30 min, treated with water (1 mL), and

extracted with dichloromethane. The or­

ganic phase was washed with water, sa­

turated sodium bicarbonate solution, and brine prior to drying and concentration. Sublimation of the semi­ crystalline residue at 200 °C (1 mm) followed by preparative VPC (5%

SE-30, 250 °C) afforded two fractions. The more volatile fraction proved to be 81 (57% of the volatile components), which was recrystallized from rV \ j hexane. The colorless crystals did not give evidence of melting (capil­ lary sealed under vacuum), but gradually decomposed between 290-360 °C accompanied with sublimation; IR (KBr, cm-1) 2970, 2930, and 2850; NMR

(6 , CDC13) 3.23 (m, 2 H ) , 2.78 (m, 5 H ) , 2.42 (m, 4 H), 2.13 (m, 4 H),

1.78 (t, J <= 1.75 Hz, 1 H), 1.73 (t, J = 1.75 Hz, 1 H), 1.25 (t, J =

1.75 Hz, 1 H), 1.19 (t, J = 1.75 Hz, 1 H), and 1.07 (s, 3 H); 19C NMR

(ppm, CDCls) 67.87, 65.50, 63.27, 62.15, 58.56, 51.04, 48.76, 45.59,

34.48, and 25.06 (3 signals not observed); m / e calcd (M+ ) 274.1721, obsd 274.1730.

0ctadecahydro-7-methyl-l,6,2,5 -ethanediylidenecyclopenta[3,4]pentaleno- f\AAAAAAAAAAAAAAAAAAAAAAAAAAAA/VVAAAA/WWWW\AAAAAAAA/WVW\AAAAAAAAAAAAA/ [2 ,1 ,6 -cde] pentaleno[2 ,1 ,6 -gha] pentalene ({^) . 'UWVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/WVAAAAAAAAA/ To a cold (0 °C) solution of 80

(10 mg, 0.04 mmol) in 5 mL of an ethanol-

tetrahydrofuran mixture (5:1) was added

anhydrous hydrazine (95%, 195 pS.). Hy­

drogen peroxide (30%, 660 p£) was 71

Introduced dropwise to the cooled reaction mixture over a period of 30 min with stirring. Following the addition, the reaction mixture was allowed to warm to 25 cC and stirring was maintained overnight. The re­ action mixture was diluted with dichloromethane and washed with water

(2x) and brine. Drying of the organic phase and solvent evaporation followed by filtration through silica gel (hexane elution) afforded 1 0 mg (100%) of W as a crystalline solid. Recrystallization from hexane gave pure product, mp (sealed, evacuated capillary) no melting - slight decomposition 241-243.5 °C accompanied by sublimation; IR (CC1*, cm-1)

3020, 2930, and 2850; *H NMR (6 , CDC13) 3.47-2.92 (series of m, 16 H ) ,

2.70-2.48 (m, 3 H), 1.58-1.46 (m, 2 H), and 1.16 (s, 3 H), 19C NMR (ppm,

CDCls) 69.96, 69.23, 68.21, 66.12, 65.98, 62.14, 58.94, 52.87, 52.38,

50.05, 33.74, and 32.33 (1 signal not observed); m/e calcd (M+) 276.1878, obsd 276.1886.

Methyldodecahedrane IVWWUVWWVWVWWWWW An intimate mixture of 83 (2 mg. r\ f \ j 0.007 mmol) and 10% palladium on carbon

(ca 1 0 0 mg) was placed in a stainless

steel reactor and flushed with argon.

The catalyst bed was then saturated with hydrogen gas (50 psi) and the reactor sealed. The reaction vessel was lowered into a molten salt bath maintained at 250 °C. After 7 h, the reactor was removed and allowed to cool. The reaction mixture was re­ moved and placed on a short silica gel column. Elution with hexane gave the product mixture (2 mg) which contained 35-40% of 84 (NMR r\f\> analysis). A total of 4.5 mg of 83 was treated in this way. From the eofXT9 uo Dll aA-paBaedajd 03 paqoafqns sba qoqqA XT° -iBaxo 8 33BT asBqd

oquBSao paqap aqq jo uoqqBaquaouoo "auqaq puF (xj) aaqBA qqqA paqsBA

pus (qm 0 0 S ) auBqqamoaoxqoqp 03 pappB sba anpqsaa aqx *uoSjb j o ureaaqs

F Japun pF3FJOdFAF SBA BqUOUmiB Bq3 PUB pappB SBA (3 O ’T B0) apf^OIH0

mnquommB pqqos ‘“Tm SX *3 33V ’aqaa asqAdoap pqdaa b 38 pappB iCqaqBqpam

-mq sba UBanjoapXqBaqaq £ip 3 0 qm g up (qorara 9 £"Z ‘3m ggg) aaqqa x^ugqd

X/Cqqamoaoqqo j o uoqqnqos v * (aqsdqssqp 0 3 suq3aq aoqoo anxq daap aq3

aaaqA quqod aq3 3B panuqquoasqp aq pxnoqs uoqqqppB XBUT3 aq3) asqAdoap

pappB sba 7^ j o aapuqBmaa aqq *0 o 8 £- 3B q g*o J33JV "pappB uaaq psq

aaqsaqp aqq jo %g6 £x3 3 BraTX0J

i\j\j . -oap.'CqB.iaaq srtoapiCuuB uq (qoram 9 £ 7 ‘3 O'l) LZ 3° uoqqnqos b ‘paAqossqp psq ranqqqqx aq3 *3 33V ’oxjn. ranqqqqx 3 ° (qB-3m ^ x ) 3® 001 pappB sba 0 o 81- 03

paxooo (mnqpos moaj paxxi^sqp Xxqsaaj) BjuoramB pqnbqx 3° 031 °X

2H30Md

HOOMd 0 0CH3

WVVWVVWVVVWWWUWVUVVWU\/VWWWVWWW\A<

: p o - g * z * X ] o u a x B q u a d q p o u a q 3 a m -8 * * * X- ( X^q33 tn^XOU3 qd) - 8 “ ^ x o jp /C q -6 - 0 jp ^ q s o a p V\A/\AAAAAATl/WVWlAa/yWWVWVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA< -Bxaq x^q^aw puB (58) aqBx£xoqaBO-£-auaxBquad[po-g‘z * X ] o u a x s q u a d [B q S - 9 ‘ x ‘ 3 ] OAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi -ouaxBquad [ t‘ j g] Bquadoxo^Co-HX-oxo-g- (x^qqaraXxouaqd) -^-ojp^qBoapBqoo X^q^aH

*0£3T**I3 sqo ‘T3ZT'VZ3 (+W) p3IBB •(H £ ‘s)

*X’T P“ B * ( H £ w o jiB U ) 26*3 * ( H 91 *s * q ) 8 £ ' £ (6T3dD *9) ^WN H t ‘.Z Z L

Pu® ‘S63T ‘0*63 (fjrc> H I *(uoqqBmqxqns amos qqqA 0 o 013 aAoqB sanooo

SuquaqaBp iKaBxiqdBO paqBtiOBAa ‘paxsas) 0 o 0£* < dm ‘auazuaq moaq uoqqBzqq

-XBqsXao Kq pauqaqqo s b a (%8 Z ) ?§f 3 0 ' 3m z*x ‘saanqxqm uoqqoBaa pauqqraoo

3 L 73

gel (10% dichloromethane-20% ether-70% hexane elution). Two major bands

were observed:

(a) Rf = 0.31 there was isolated 489 mg (48%) of mp 153-154.5 °C

(from ethyl acetate); IR (KBr, cm-1) 2920, 1735, 1720, 1595, 1238, 745,

and 680; *H NMR (6 , CDCla) 7.4-6.7 (m, 5 H), 4.2-1.1 (series of m, 22 H),

3.81 <%ABq, J - 9 Hz, 1 H), 3.79 ftABq, J = 9 Hz, 1 H), and 3.68 (s, 3 H);

13C NMR (ppm, CDCla) 226.97, 173.44, 158.82, 129.44, 121.25, 115.06, 79.07,

65.24, 61.48, 59.29 (2C), 57.41, 54.74, 54.31, 54.07, 52.98, 52.80, 51.83,

51.16, 50.73 (2C), 35.20, 31.19, 31.01, 30.34, and 27.73; m/e (M^") 432.

Anal. Calcd for C 28H 220„« C, 77.75; H, 7.45. Foundi C, 77.83;

H, 7.51.

(b) Rf = 0.5: there was obtained 224 mg (22%) of 8 6 , mp 132- ■L

1496, 1246, 742, and 682; XH NMR (6 , CDCla) 7.4-6.6 (m, 5 H), 5.29 (s,

1 H), 4.23 &ABq, J = 11 Hz, 1 H), 4.10 &ABq, J ■= 1 1 Hz, 1 H) , 3.67

(s, 3 H ) , and 3=4-0.9 (series of m, 21 H); l9C NMR (ppm, CDC13) 178.56,

159.73, 129.33, 120.30, 114.82, 82.48, 67.53, 62.82, 62.93, 60.01, 57.09,

56.90 (2C), 53.79, 53.11, 51.56, 50.54, 49.42, 48.74, 48.35, 48.21, 32.72,

27.19, 26.07, 25.54, and 24.18; m/e calcd (M+) 432.2300, obsd 432.2310.

Anal. Calcd for C28H 320 4 : C, 77.75; H, 7.45. Found: C, 77.78,

H, 7.55.

Methyl 0ctadecahydro-3b-hydroxy-7-(phenoxymethyl)-l,6-methanocyclopenta- 'WW/WWWWWWVWWWWWV/WWWWV/WWWV/WWVWVUWVAAA/WWWWVWW. [3,4]pentaleno[2,1,6 -cde]pentaleno[2,1,6-gha]pentalene-7-carboxylate (87). VWVWVA/WVnAA/WWWWWVVVAnAA/WVA/VWWWWWWWVWVWVVWVWWWWWlAn, A solution of 85 (260 mg, 0.6 mmol) in 20 mL of a dry deoxygenated ' V b benzene-tert-butyl alcohol solvent system (4:1) containing 2 drops of 74

triethylamine was irradiated with a 450W

Hanovia lamp through pyrex for 18 h under

a nitrogen atmosphere. The resulting

solution was concentrated in vacuo to

yield the crude product as a crystalline

solid. Recrystallization from benzene-ether provided 234 mg (90%) of

87, mp 122-124 °C; IR (KBr, cm"*) 3420, 2930, 1725, 1595, 1491, and f\f\j

1235} NMR (6 , CDC19) 7.5-6.6 (m, 5 H), 4.0-0.9 (series of m, 22 H) ,

3.78 (s, 2 H), and 3.67 (s, 3 H); 13C NMR (ppm, CDC1S) 173.68, 158.82,

129.38, 121.13, 115.06, 97.95, 79.92, 79.13, 64.81, 64.14 (2C), 63.24,

55.34, 53.89, 51.10, 50.79, 31.92, and 29.98; m/e calcd (M+) 432.2300,

obsd 432.2310.

Methyl 1,la,lb,2,3,3a,4,5,5a,6 ,6a ,6b ,6 c,6 d ,6 e ,6 f-Hexadecahydro-7- 'VWWWWWVWWAAAAAA/WWWWWWVWVWWVAAAAA/VWWWWVAAAAAAA/ (phenoxymethyl)-l,6-methanocyclopenta[3,4]pentaleno!2,1,6 -cde]pentaleno- OAAAAAA/VVVAAA/VXAAAAAAAAAAAAAA/VAAAAAAAAAAAAAAAAAAAAAAAA/VAAAAAAAAAAA/VAAAAA. [2,1,6-gha] pentalene-7-carboxylate (f$ ) .

under argon for 8 h. The solvent was

evaporated Iji vacuo to give the crude

product as a crystalline solid. Recrystallization from ethyl acetate

provided pure 8 8 (223 mg, 93%), mp 155.5-157 °C; IR (KBr, cm“0 2940, f\/\j 1728, 1601, 1241, 1221, 1111, 745, and 683; *H NMR (6 , CDC1S) 7.4-6.6

(m, 5 H ) , 4.1-1.1 (series of m, 20 H ) , 3.82 &ABq, • 7.5 Hz, 1 H), 3.69

(JsABq, J - 7.5 Hz, 1 H), and 3.60 (s, 3 H); l3C NMR (ppm, CDCla) 173.71, 75

158.90, 139.92, 137.10, 129.38, 121.13, 115.16, 80.25, 72.39, 69.62,

63.50, 63.16, 62.58 (2C), 55.39, 53.40, 52.67 (2C), 51.02, 48.89, 48.55,

47.04, 30.24, 29.18, 28.30, and 24.32; m/e calcd (M+) 414.2195, obsd

414.2210.

Methyl 0ctadecahydro-7-(phenoxymethyl)-l,6-methanocyclopenta[ 3,4]- f\A/V\AAAAAAAA/\AAAA/WWWV\AAAAAAAAAAAAAAAAA/\AAAAAAAAAAAAAAAAAnAAAAA( pentaleno[ 2,1,6 -cde] pental ■;co[ 2,1,6 -gha] peltalene-7-carboxylate (§g). /\/V/V/V>AAAAAAAAAAAAAAAAAAAAA/v/WinAAAAAAAAAAAAAAAAAAAAAAAAAAAA/V\AAAAAAAAi To a cold (0 °C) solution of

(240 mg, 0.58 mmol) in 35 mL of an etha-

nol-tetrahydrofuran mixture (2.5:1)

was added anhydrous hydrazine (97%, 1.6

mL). Hydrogen peroxide (30%, 5.7 mL) was introduced dropwise to the cooled reaction mixture over a period

of 1 h with stirring. After the addition was complete, the cooling bath was removed and the reaction mixture was stirred overnight at 25 °C, di­ luted with dichloromethane (200 mL), and washed with water (2x) and brine. The organic phase was dried, filtered, and concentrated in vacuo to afford the product as a crystalline solid. Recrystallization from ethyl acetate gave pure 89 (220 mg, 91%), mp 162-164 °C; IR (KBr, cm-1) iy\i 2941, 1731, 1496, 1241, and 748; XH NMR (6 , CDC1S) 7.4-6.6 (m, 5 H ) ,

4.0-0.8 (series of m, 22 H), 3.78 (s, 2 H), and 3.64 (s, 3 H); X3C NMR

(ppm, CDCla) 173.85, 158.95, 129.38, 121.03, 115.16, 79.09, 70.40, 66.46,

64.23, 63.40, 57.38, 54.96, 52.92, 51.07, 50.64, 49.71, 31.36, and 30.39; m/e calcd (M+) 416.2351, obsd 416.2362.

Anal. Calcd for CaaHaaOa: C, 80.73; H, 7.74. Foundi C, 80.57;

H, 7.65. 76

Qctadecahydr0 -7- (phenoxymethyl)-l,6-methanocyclopenta[3,4]pentaleno- 'WWWVAAAA/WWWWWWVWVWWWWWWWWWWWWVWWVWWWWWV/ [2,1,6 -cde]pentaleno[2,1,6-gha]pentalene-7-methanol (90). 'WWWVAAAA/WIA/WVWWWWWWWVVWVAA/WWWWV/WWWWV A solution of 89 (220 mg, 0.53

mmol) in anhydrous benzene (10 mL) was PhOCHg stirred at room temperature while diiso-

butylaluminum hydride (2.65 mL of 1 M

in hexane, 2.65 mmol) was introduced.

The reaction mixture was stirred at 25 °C for 12 h under argon, treated

dropwise with methanol to destroy excess hydride, and poured into di-

chloromethane (200 mL). The organic layer was washed with dilute

hydrochloric acid (2x), water, and brine prior to drying. Solvent

evaporation left 200 mg (97%) of ^ as a crystalline solid, mp 92-

9A °C; IR (KBr, cm"*) 3400, 2930, 1599, 1497, 1240, 746, and 685;

*H NMR (6 , CDCls) 7.4-6.7 (m, 5 H), 4.33 (s, 2 H), 4.0-1.1 (series

of m, 23 H), and 3.60 (s, 2 H ) , 13C NMR (ppm, CDC1,) 159.05, 129.48,

120,98. 114.92, 78.21, 70.54, 67.00, 64.86, 61.61, 61.22, 57.58, 53.55,

51.85, 51.56, 50.98, 31.60, and 30.88; m/e calcd (M+) 388.2402, obsd

388.2411.

Octadecahydro-7-(phenoxymethyl)-1,6-methanocyclopenta[3,4]pentaleno- VWW\AMAM/VWVW\AnAArtA/WlA/WWWrt/WVWWW\M/WWWVWWWWlMAA; [2,1,6 -cde]pentaleno[2,1,6 -gha]pentalene-7-carboxaldehyde (J^). 'WWWVWVWWWWVWWVWVWWWWVWWWVnAAA/WVWmA/VAAAA/WV A solution of ^ (200 mg, 0.52 mmol)

H 0 in 5 mL of dichloromethane was added to PhOCH2 a suspension of pyridinium chlorochromate

(200 mg, 0.93 mmol) in 10 mL of the same

solvent with stirring under argon. After 77

1.5 h, ether was added and the organic layer was decanted. The residual

salts were washed with ether (2x) and the combined organic solutions were washed with dilute hydrochloric acid, saturated sodium bicarbonate

solution (2x), water and brine. Following drying and solvent evapora­

tion, the crude product was filtered through a short silica gel column

(10% ether in hexane elution) to give 191 mg (95%) of mp 159-161 °C

(from ethyl acetate); IR (KBr, cm-1) 2930, 1710, 1492, 1263, 738, and

679; lH NMR (6 , CDC13) 10.17 (s, 1 H), 7.4-6.6 (m, 5 H ) , 3.9-1.1 (series of m, 22 H), and 3.68 (s, 2 H ) ; 13C NMR (ppm, CDC1S) 206.15, 158.82,

129.50, 114.88, 78.65, 6 6 .8 8 , 64.69, 57.73, 54.80, 52.61, 50.25, 49.88,

31.98, and 30.58 (3 signals not observed); m/e calcd (M*) 386.2246, obsd 386.2237.

0ctadecahydro-7f-(phenoxymethyl)-1,6,7-metheno-lH-cyclopenta[3,4]penta- '\AAAAAAAAAAAAAAAAAAAA/VV\AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAnAA/VVAAAAAAAAAAAi leno[2 ,1 ,6 -gha]pentaleno[1 ,2 ,3-cd]pentalen-2 -ol (££). AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/VAAAAAAAAAAAi A cold (-78 °C) solution of 91 A/

^VCHgOPh (195 mg, 0.51 m m o l) in 20 mL of a de­ oxygenated 9:1 toluene-ethanol solvent

system was irradiated under nitrogen

with a 450W Hanovia lamp through pyrex.

After a 5-h reaction period, the solution was allowed to warm to 25 °C and the solvent was removed under reduced pressure. Preparative TLC on silica gel (elution with 10% dichloromethane-15% ether-75% hexane) gave three bands. At Rf * 0.8, a mixture of decarbonylated materials

(45 mg) was obtained. The Rf m 0.6 band comprised alcohol 92^ (70 mg,

36%) as an oil; IR (neat, cm-1) 3580, 2930, 1600, 1495, 1233, 743, 78

and 681; *H NMR (6 , CDC1S) 7.4-6.6 (m, 5 H), 4.41 (d, J « 6 Hz, 1 H ) ,

4.1-1.0 (series of m, 22 H ) , 3.78 (JgABq, J - 9 Hz, 1 H), and 3.62

OsABq, J •= 9 Hz, 1 H ) ; 19C NMR (ppm, CDC1S) 159.14, 129.58, 121.23,

114.92, 85.49, 84.43, 68.74, 67.87, 67.00, 66.41, 63.11, 62.63, 60.73,

59.23, 57.00, 55.30, 53.55, 53.21, 52.63, 50.34, 32.58, 32.33, and

31.75 (2 signals not observed)j m/e calcd (M+) 386.2246, obsd 386.2253.

The R£ = 0.5 band proved to be alcohol ^ (42 mg, 21%).

Octadecahydro-1-(phenoxymethyl)-1,6 ,7-metheno-lH-cyclopenta[3,4]pentaleno- 'V/WVAA/WWWVWWWWWWWWWV/WVA/WWVWVWWVWWWWWWVWWWWVWV [2 ,1 ,6 -gha]pentaleno[1 ,2 ,3-cd]pentalen-2-ol ( ^ ) . (VVVVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA; A solution of 92 (325 mg, 0.84 mmol)

to 50 mL of liquid ammonia (freshly dis­

tilled from sodium). Additional tetra­

hydrofuran was added as required to en­ sure complete dissolution of the alcohol. The solution was allowed to reflux gently under argon as lithium wire (118 mg, 16.8 mg-at) was added in small pieces over 5 min. After an additional 15 min, absolute ethanol (1 mL) in 1 mL of dry tetrahydrofuran was added dropwise over a 30 min period. (Additional ethanol should be added as required to completely discharge the blue color). The ammonia was evaporated under a stream of argon and the residue was diluted with water and extracted with dichloromethane (3x). The combined organic portions were washed with water and brine. Concentration of the dried organic phase left the product as a white foam (328 mg, 100%) which was used directly in the next step; IR (CDC19, cnr*) 3550, 2900, 1685, 1180, 1165, 900, and 720; 79

*H NMR (6 , CDCla) 5.6 (s, 2 H), 4.7-4.5 (br s, 1 H), 4.3-4.2 (m, 1 H)

and 3.6-2.1 (series of m, 28 H).

Octadecahydro-2-hydroxy-l,6 ,7-metheno~lH-cyclopenta[3,4]pentaleno- 'lA/WVWWWWWWWWWWXAAAA/WVWVWWWWWWW/WVWWWVWWVb f2,1,6-gha]pentaleno[1,2,3-cd]pentalene-l-methanol (94). 'WWVWUWWWWWWWVWVAA/VWWWVAAnAAAAnA/WWWWW/ A solution of 93 (135 mg, 0.35

mmol) in 20 mL of tetrahydrofuran was

treated with 3 M hydrochloric acid (1

mL). The resulting solution was stirred

at 25 °C for 3 h and poured into di- chloromethane (30 mL). The solution was washed with water and saturated sodium bicarbonate solution (2x) prior to drying. Solvent evaporation left 107 mg (99%) of ^4 as a crystalline solid, mp softens at 172 °C then first melts at 256.5-257 °C and finally at 264-265.5 °C dec; IR

(KBr, cm-1) 3380, 2920, and 1040; XH NMR (6 , CDCla) 4.35 (d, =

6 Hz, 1 H) and 3.8-1.7 (series of m, 25 H); X3C NMR (ppm, CDC19)

86.12, 80.01, 68.74, 67.77, 66.99, 66.37, 63.02, 62.48, 60.39, 59.18.

56.99, 55.25, 53.60, 53.11, 52.82, 50.34, 32.62, 32.28, and 31.70

(2 signals not observed); m/e calcd (M+) 310.1933, obsd 310.1941.

Octadecahydro-2-oxo-l,6 ,7-metheno-l^-cyclopenta[3,4]pentaleno[2,1,6 -gha]- /VWWWWWVWWWWVWWWWWWWWWWWVWWVVVWVVVWVWVWVWVAAAn/V pentaleno[1 ,2 ,3-cd]pentalene-l-carboxaldehyde ( ^ ). 'VWWVWWWWWWWWWVWWWWWVWWWWWWWt A solution of 94 (107 mg, 0.34 mmol 0 <\f\j .0 j! in 2 mL of dichloromethane was added

to a suspension of pyridinium chloro­

chromate (298 mg, 1.38 mmol) in 10 mL

of the same solvent with stirring under 80 argon. After 4 h, ether was added and the organic layer was decanted.

The residual salts were washed with ether (3x) and the combined organic solutions were washed with dilute hydrochloric acid (2x), saturated sodium bicarbonate solution (2x), water and brine. Following drying and solvent evaporation, the crude product was filtered through a short silica gel column (25% ether in hexane elution) to give 81 mg (77%) of

95 as a white solid; IR (CDC13, cm"1) 2940, 1730, 1700, and 1450; lE VO NMR (

Octadecahydro-1,6 ,7-metheno-2H-cyclopenta[3,4]pentaleno[2,1,6 -gha]- 'WWVWVWVWWWWWWWWWWAAAAAAA/WWWWWWVWVWWWWWWV pentaleno[1 ,2 ,3-cd]pentalen-2 -one (^6 ). 'VVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi A suspension of 9^ (70 mg, 0.23

mmol) in 6 mL of absolute ethanol was

treated with tetrahydrofuran until

dissolution occurred. To the resulting

solution was added a solution of po­ tassium hydroxide (15 mg, 0.26 mmol) in 6 mL of absolute ethanol.

After stirring under argon at 25 °C for 5 h, water was added and the solution was extracted with dichloromethane (3x). The combined organic extracts were washed with water (2x) and brine prior to drying and sol­ vent evaporation. Purification by preparative TLC (10% dichloromethane-

40% ether-50% hexane elution) yielded 31 mg (48%) of 9 ^, mp decomposes above 226 °C without melting (from ether); IR (KBr, cm-1) 2920, 1710,

1445,. and 690; lH NMR (6 , CDC1,) 3.9-1.4 (series of m, 22 H); l9C NMR

(ppm, CDCla) 228.57, 69.66, 68.65, 67.87, 67.35, 65.30, 61.22, 59.31, 59.23, 58.36, 57.90, 55.88, 54.43, 52.41, 52.15, 50.90, 49.69, 37.18,

31.43, and 27.44; m/e .calcd (M+) 278.1671, obsd 278.1685. 81

Hexadecahydro-1,6 ,2,5-ethanediylidenecyclopenta[3,4]pentalenol2,1 ,6-cde]- *WV\AAAAAAAA/WVWWVAAAAAAAAAA/WVAAAAAAA/WWWVAAAAA/WVAAAAAAAAAAAAAAAAAA< pentaleno!2,1,6-gha]pentalen-3b(lH)-ol ( 9 7 ). WVWWWWWVWIAAAA/WV/WV/VWWWVWWWWU OH A s o lu tio n o f 96 (28 mg, 0 .1 mmol) r\y\i In 10 mL of a deoxygenated benzene-te rt-

butyl alcohol solvent system (4 il) con­

taining 2 drops of triethylamine was

Irradiated with a 450W Hanovla lamp

through pyrex for 10 h under a nitrogen atmosphere. The solvent was

removed in vacuo to give 28 mg (100%) of 97 as a crystalline solid, 'Vb mp > 350 °C; IR (KBr, cm-1) 3320, 2920, 1000, and 990; XH NMR (6,

CDCla) 3.6-2.8 (m, 18 H) and 1.8-1.3 (m, 4 H); 1SC NMR (ppm, CDCla)

98.02, 79.47, 69.38, 68.41, 65.98, 64.18, 64.08, 60.73, 60.06, 53.11,

48.98, and 31.99; m/e calcd (M+) 278.1671, obsd 278.1680.

1,la ,lb ,2,3,3a,4,5,5a,6,6a,6b,6c,6d,6e,6f-Hexadecahydro-1,6,2,5-ethane- nAn/v/wwvvwvwvwwwwwvwvwvAn/wwvwwvwwwwxMnnnAAAnAnAnAnnA/ diylidenecyclopenta[3,4]pentaleno[2 ,1,6-cde]pentaleno[2 ,1,6-gha]pentalene nAnAAnAA/WWWWXAAA/WVAAAA/V/WVWVVWWWWWVA/WWWVVWV/WWWVVAAAnAAAj < w - 'WW'-

A solution of 97 (31 mg, 0.11 mmol) f\/\> in 15 mL o f d ry benzens was tre a te d w ith

a few crystals of j>-toluenesulfonic acid

and heated to reflux for 1.25 h. The

cooled s o lu tio n was d ilu te d w ith d i - chloromethane and washed with saturated sodium bicarbonate solution

(2x) and water. Drying and solvent removal yielded 29 mg (100%) of

^8 w h ic h could be recrystallized from hexane, mp > 350 °C; IR (KBr, cm”1) 3020, 2920, 1485, 1440, 750, and 690; *H NMR (6, CDCls) 4.0-2.7

(series of m, 16 H) and 2.1-1.2 (series of m, 4 H); 19C NMR (ppm, 82 CDCla) 142.44, 141.18, 70.74, 67.72, 67.19, 66.71, 65.78, 64.13, 62.92,

61.03, 60.35, 59.08, 58.74, 56.75, 52.53, 51.95, 50.54, 48.21, 33.11, and 30.29; UV (isooctane) max 210 nm (e 7670); m/£ calcd (M+) 260.1565, obsd 260.1571.

Octadecahydro-1 ,6,2,5-ethaned iylidenecyclopenta[3,4]pentaleno[2,1,6 -cde]- ^AA/VWVV\AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/\AAAAAAAAAAA« pentaleno[2 ,1 ,6 -gha]pentalene (#9 ). VWWWWUWWWWVMWWWWWWWl To a cold (0 °C) solution of ^

/ (29 mg, 0.11 mmol) in 14 mL of an etha-

nol-tetrahydrofuran mixture (5:1) was

added anhydrous hydrazine (97%, 450 yf).

Hydrogen peroxide (30%, 1.52 mL) was introduced dropwise to the cooled reaction mixture over a period of

45 min with stirring. Following the addition, the reaction mixture was allowed to warm to 25 °C and stirring was maintained overnight.

The reaction mixture was diluted with dichloromethane and washed with water (2x) and brine prior to drying and solvent evaporation. Prepara­ tive TLC (5% dichloromethane in hexane elution) afforded 19 mg (6 6 %) of 99 as a crystalline solid, mp > 360 °C; IR (KBr, cm-1) 3030, 2940, rV \ j

2860, 1490, 1450, and 700; XH NMR (6 , CDCls) 3.5-2.9 (series of m, 20 H) and 1.6-1.4 (m, 2 H); X9C NMR (ppm, CDC19) 69.56, 68.25, 65.91, 61.90,

52.86, 49.89, and 31.91; m/e calcd (M+) 262.1721, obsd 262.1729.

2,9 3 • 7 /. 2 o 3 is 6.16 6 • 1 5 1 0 * 1 12. 19 13.17 Undecyclo[9.9.0.0 .0 .0 .0 * .0 ’ .0 .0 .0 .0 ]- /VAAAAAAAAAAAA/VVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/VlAAAAAAAAA/VVAAAAAAAAAAAAi eicosane (dodecahedrane) 'WWWWWWWWVWWWWW; An intimate mixture of (1 mg, 0.004 mmol) and 10% palladium on carbon (ca 1 0 0 mg) was placed in a stainless steel reactor and flushed with argon. The catalyst bed was then

saturated with hydrogen gas (50 psi)

and the reactor sealed. The reac­

tion vessel was lowered into a mol-

ton salt bath maintained at 250 °C.

After 4.5 h, the reactor was removed and allowed to cool. The con­ tents were placed on a short silica gel column and washed with hexane to provide 1 mg of the product mixture which contained 40-50% of ^

(NMR analysis). A total of 4.5 mg of 99 was treated in this way. f\J\j From the combined reaction mixtures, 1.5 mg of 4 (34%) was obtained by crystallization from benzene, mp > 450 °C (sealed, evacuated capillary; darkening occurs above 350 °C); IR (KBr, cm-1) 2945, 1298, and 728; Raman (cnT1) 2954, 2938, 1324, 1164, 1092, 840, 676, and

480; XH NMR (6 , CDC13) 3.38 (s); 19C NMR (ppm, CDC1S) 66.93, (d, J.CH =

134.9 Hz); m/e calcd (M*) 260.1565, obsd 260.1571. APPENDIX

NMR SPECTRA IA j u L_L

-I 1---- 1---- 1---- 1---- 1____ I____ I____ ■ I I__ _ __ I---- 1____ I I I____ I____ I I I I____ I____ I---- I

Figure 2. 200 MHz lH NMR Spectrum of 81. 00 Ui Figure 3. 200 MHz NMR Spectrum of 83. 00 O' V4 >Nf» *r**r *m» *< ft * v i I .L 1 -j i i X X » » i i l

Figure 4. 200 MHz 1H NMR Spectrum of 84.

•vj00 Figure 5. 200 MHz XH NMR Spectrum of 00 00 DODECRHEDRRNE

JJM

Figure 6 . 200 MHz *H NMR Spectrum of Dodecahedrane 4. % t&j— rto— ten— ten— ten— ten— ten— rln— ten— *— *— 1a— *— *— *— *— * to tr PPM

Figure 7. Broadband Decoupled C NMR Spectrum of Dodecahedrane 4. VOo Figure 8. Off-Resonance Decoupled l3C NMR Spectrum of Dodecahedrane 4. REFERENCES AND NOTES

(1) Jowett, B. "The Dialogues of Plato" Vol. II, Charles Scribner's Sons, New York, 1911.

(2) Eaton. P. E.; Cole, T. W., Jr. J. Am. Chem. Soc. 1964, 8 6 , 3157. 'WAAi (3 ) (a) Maier, G.; Pfriem, S.; Schaffer, U., Matusch, R. Angew. Chem. Int. Ed. Engl. 1578, 17, 520. (b) Maier, G.; Pfriem, S. Ibid. 1978, 17, 5 1 9 ^ < w w — (4 ) Rauscher, G.; Clark, T.; Poppinger, D.; Schleyer, P. v. R. Angew. Chem. Int. Ed. Engl. 1978, 17, 276. a------— -- --- rV \ A A j --- For a review, see: Eaton, P. E. Tetrahedron 1979, 35, 2189. (5 ) ’ OAAA/--- (6) (a) Paquette, L. A.; Balogh, D. W. J. Am. Chem. Soc. 1982, 104, 774. (b) Paquette, L. A.; Balogh, D. W.; Usha, R.; K^Jftftz, D.; Christoph, G. G. Science (Washington, D.C.) ,1981, 211, 575.

(7 ) (a) Engler, E. M . ; Andose, J. D.; Schleyer, P. v. R. J. Am. Chem. Soc. 95., 8005. (b) Clark, T.; Knox, T. McO.; Mackle. H.: McKervey, M. A. J. Chem. Soc. Chem. Commun. 1975, * ------OAAA. OOD.

(8) Schulman, J. M . ; Venanzi, T.; Disch, R. L. J. Am. Chem, Soc, 1 G7C. Q"7 m g * W W — (9 ) Woodward, R. B.; Fukunaga, T.; Kelly, R. C. J. Am. Chem. Soc. 1964, £16, 3162. For related work in the triquinacene area see; (a) Muller, D. M.; Chem. Weekblad 1963, 59, 334. (b) Jacobson, I. T. Ph.D. Dissertation, University of Lund, 1973. (c) Jacob­ son, I. T. Acta Chem. Scand. 1967, 21, 2235. (d) Jacob­ son, I. T. Chem. Scripta 174. (e) Paquette, L. A. Top. Curr. Chem. 1^7^. 7j£jLI4.

(10) (a) Paquette, L. A.; Itoh, I.; Farnham, W. B. J. Am. Chem. Soc. i , 97, 7280. (b) Paquette, L. A.; Farnham, W. B.; Ley, S. Fbid. 97, 7273. (c) Paquette, L. A.; Itoh, I.; Lip- kowitz, K. B. J. Org. Chem. ^lL» 3524.

(11) (a) Repic, 0. Ph.D. Dissertation, Harvard University, (b) Roberts, W. P.; Shoham, G. Tetrahedron Lett. ^ 8 1 . 4895.

92 93 (12) McKervey, M. A.; Vibuljan, P.; Ferguson, G.; Slew, P. Y. J. Chem. Soc. Chem. Commun. 1981, 912. ------OAAAi (13) Vibuljan, P. Ph.D. Dissertation, National University of Ireland, 1981.

(14) (a) Schleyer, P. v. R.; Nicholas, R. D. Tetrahedron Lett. 305. (b) Fry, J. L.; Schleyer, P. v. R.; unpublished observa­ tions.

(15) Jones, N. J.; Deadman, W. D.; LeGoff, E. Tetrahedron Lett. 1973, 2087. W A '

(16) Eaton, P. E.; Mueller, R. H.; Carlson, G. R.; Cullison, G. F.; Chow, T.-C.; Krebs, E.-P. J. Am. Chem. Soc. 1977, 99, 2751. ------'Wi/V — (17) Paquette, L. A.; Snow, R. A.; Muthard, J. L.; Cynkowski, T. J. Am. Chem. Soc. 100. 1601.

(18) Osborn, M. E.; Kuroda, S.; Muthard, J. L.; Kramer, J. D.; Engel, P.; Paquette, L. A. J. Org. Chem. ,1981, 46, 3379.

(19) Christoph, G. G., Muthard, J. L.; Paquette, L. A.; Bohm, M. C.; Gleiter, R. J. Am. Chem. Soc. 1978, 100, 7782. ------vwii --- (20) Paquette, L. A.; Wyvratt, M. J.; Berk, H. C.; Moerck, R. E. J. Am. Chem. Soc. 1978, 100, 5845. ------r\ A / V \ j --- (21) (a) Schulman, J. M . ; Disch, R. L. J. Am. Chem. Soc. 1978, 100, 5677. (b) Allinger, N. L.; Tribble, M. T.; Miller; Wertz, D. W. Ibid. 1971, 93, 1637. (c) Allinger, N. L. Ibid. 1977, 99, 8127. W V A ' W U V -- (22) Gasteiger, J.; Dammer, 0. Tetrahedron 1978, 34, 2939.

(23) (a) Ermer, 0. Angew. Chem. Int. Ed. Engl. 16, 411. (b) Newton.> M. D.; » Schulman, » J. M. J. Am. Chem.- - Soc.v w i 1974, — 96, 17. (24) (a) Wishnok, J. S. J. Chem. Ed. 1972, 50, 780. (b) Gerzon, K.; Kau, D. J. Med. Chem. 1967, 10, 189. » W \ Aj — (25) Vogel. E. Chem. Ber. 1952, 85, 25.

(28) Helwig, G.; Kramer, J. D. unpublished observations.

(29) The experimental details for the preparation of 24 were developed by F. R. Kearney: Paquette, L. A.; Kearney, F. R.; Drake, A. F.; Mason, S. F. J. Am. Chem. Soc. 103, 5064.

(3 0 ) Wasserman, H. H.; Lipshutz, B. H. Tetrahedron Lett. 1975, 4611. * ’ r * ------

(3 2 ) (a) Gersmann, H. R.; Nieuwenhuis, H. J. W.; Bickel, A. F. Tetra­ hedron Lett. 1963, 1383. (b) Russell, G. A.; Bemis, A. G. J^_ Am. Chem. S o c y ^ f 6 6 , 8 8 , 5491. ------'\AAAj — (33) Doering, W. v. E.; Haines, R. M. J. Am. Chem. Soc. 76, 482.

(34) (a) Trost, B. M . ; Tamaru, Y. J. Am. Chem. Soc. X 2 1 5 * 97, 3528. (b) Trost, B. M . ; Tamaru, Y. Ibid. 19T7, 99, 3 1 0 1 ^

(35) Nokami, J.; Matsuura, N.; Sueoka, T.; Kusumoto, Y.; Kawada, M. Chem. Lett. 1978, 1283. ------'WW, (36) Corey, E. J.; Carney, R. L. J. Am. Chem. Soc. .1971, 93_, 7318.

(37) Corey, E. J.; Danheiser, R. L.; Chandrasekaran, S. J. Org. Chem. 1976, 41, 260. '\AAA< — (38) (a) Paquette, L. A.; Itch, I.; Lipkowitz. K. B. J. Org. Chem. 1976, 41, 3524. (b) Burgess, E. M . ; Penton, H. R. Jr.; Taylor, J. Am. Chem. Soc. 1970, 92. 5224; J. Org. Chem. 1973, 38, 26 ------/w\Ai — ------v m —

(3 9 ) Paquette, L. A.; Wyvratt, M. J.; Schallner, 0.; Muthard, J. L.; Begley, W. J.; Blankenship, R. M . ; Balogh, D. J. Org. Chem. 1979, 44, 3616. 'VWV/

(4 0 ) Wyvratt, M. J. Ph.D. Dissertation, The Ohio State University, 1976.

(41) (a) Allen, M. J. "Organic Electrode Processes," Reinhold: New York, 1958; p 58. (b) Perrin, C. L. "Progress in Physical Or­ ganic Chemistry, Vol 3," Wiley: New York, 1965; p 195. (c) Popp, F. D.; Schultz, H. P. Chem. Revs, 62, 29.

(42) (a) Paquette, L. A.; Wyvratt, M. J.; Schallner, 0.; Schneider, D. F.; Begley, W. J.; Blankenship, R. M. J. Am. Chem. Soc. 98, 6744. (b) Paquette, L. A.; Wyvratt, M. J.; Berk, H. C.; 95 Moerck, R. E. Ibid. 1^78, 100. 5845. (c) Balogh, D.; Begley, W. J.; Bremner, D.; wyvratt, M. J.; Paquette, L. A. Ibid. 101. 749. (d) Paquette, L. A.; Snow, R. A.; Muthard, J. L.; Cynkowski, T. Ibid. 1979, 101. 6991. (e) Paquette, L. A.; Wyvratt, M. J.; SchaYlner, 0.; Muthard, J. L.; Begley, W. J.; Blankenship, R. M . ; Balogh, D. J. Org. Chem. 1979, 44, 3616. 11,1 VWli ”" (43) Grinshaw, J.; Haslett. R. J. J. Chem. Soc., Chem. Commun. 1974, ’ ’ *—------V W b 174.

(44) The likelihood that the new framework bond is formed by radical coupling following reduction of the proximal C-Cl bond in !>0 cannot, of course, be dismissed and remains a distinct possi­ bility.

(45) Balogh, D. W. Ph.D. Dissertation, The Ohio State University, 1980.

(46) (a) Begley, W. J. (b) Banwell, M. G. unpublished results.

(47) Corey, E. J.; Suggs, J. W. Tetrahedron Lett. 1975, 2647.

(48) Corey, E. J.; Schmidt, G. Tetrahedron Lett. 1979, 399.

(49) Reviews: Fatiadi, Synthesis 1976, 65, 133.

(50) Carp.ino, L. A. J. Org. Chem. ,1970, JJj, 3971.

(51) Tomioka, H.; Takai, K.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1981, 1605. m (52) (a) Chatt, J.; Shaw, B. L. Chem. Ind. (London) I960, 931. (b) Strauss, S. H. private communication.

(53) Pfitzner, K. E.; Moffatt, J. C. J. Am. Chem. Soc. 1965, 82., 5661, 5670.

(54) Doering, W. E.; Parikh, J. R. J. Am. Chem. Soc. ^ 7 , 89, 5505.

(55) Corey, E. J.; Kim, C. U. J. Am. Chem. Soc. 1 9 2 % , 94, 7856; Org. Chem. 1973, 38, 1233. — ------'WVU — (56) Cookson, R. C.; Stevens, I. D. R.; Watts, C. T. J. Chem. Soc. Chem. Commun. 1966, 744. ------O A A A T (57) (a) Mancuso, A. J.; Huang, S.-L. Swern, D. J. Org. Chem. 1%?^. 43, 2480. (b) Mancuso, A. J.; Swern, D. Synthesis 1981, 165.

(58) Barton, D. H. R.; Garner, B. J.; Wightman, R. H. J. Chem. Soc. 1964, 1855. For a recent application of this procedure, see Greenlee, W. J.; Woodward, R. B. J. Am. Chem. Soc. 98, 96

(59) Binkley, R. W. Synth. Commun. 1976, 6 , 281; J. Org. Chem. 1976, 41, 3030. ^

(6 0 ) Further examples of this photooxidation have recently appeared: Gibson, T. J. Org. Chem. 1981, 46, 1073; Towsend, C. A.; Neese, S. A.; Thels, A. B. J. Chem. Soc. Chem. Commun. 1982, 116. 'VAAAi (6 1 ) We are aware of a lone example where this Issue has been raised. In their attempts to effect chemospeciflc sulfenylation of methyl 9-oxodecanoate with dimethyl disulfide, Trost, Salzmann, and Hiroi [J. Am. Chem. Soc. 98, 4887] were never able to attain > 30% of the desired a-carbomethoxy sulfenylated product. The extent of reaction which occurred a to the ketone carbonyl was not stated.-

(6 2 ) Jeger,w 0.;' Schaffner, 9 K. Helv. - Chim.' —i Acta 'VWU1970, "" 53, 247.

(6 3 ) Christoph, G. G. private communication.

(6 4 ) Kirchen, R. P.; Ranganayakulu, K.; Rauk, A.; Singh, B. P.; Sorensen, T. S. J. Am. Chem. Soc. 1981, 103, 588 and references cited therein. 'vwu

(6 5 ) Rylander, P. N. "Organic Synthesis with Noble Metal Catalysis" Academic Press, Photochemistry 1979, 4, 1. — ------AAAA — (6 9 ) Mislow, K. private communication. For a discussion of the method used, see Baum, M. W . ; Guenzi, A.; Johnson, C. A.; Mislow K. Tetrahedron Lett. 23, 31.

(7 0 ) The Raman spectrum was obtained on a Spectra Physics Model 164 instrument equipped with an argon laser. We are Indebted to Mark Wisnowsky (Owens-Coming Co.) for the determination of these spectra.

(7 1 ) Ermer, 0. Angew., Int. Ed. Engl. ^977, £, 411.

(72) Discoloration of the crystals Is observed above 350 °C.