. WWW MK: J... 4.1.x” I» .. I. r .... IIIUertIfiJH . z .. LEFM...&%.WI«.%§ MW.» at»... ah”... ...I. ...... . .....I; ....u II... ..hrWfir : ... . .... . ”h. ...... .....- . ..n. - --I. .. 3...». .-.nh ..I-I . uIvInLIfi. DIM... .. ....H1 ...)Luuuvvlbmmvuvwm . . :1 .7 - tmLsLIH..- .... ......VMII ,- ... IL '1‘!“ It. III-3. . ..I . . . .... . BahafiflddI. I . “Ni-I Ln .MFflAIIB $.43... . .I .rIur. JmII ”UHMWUI "..Iwnagh IflfiIMIflIIIIhIM.H.LIII.HIII|I.I4.nmI. I41”... .I I I I . . u I. I .mIIIIILMLwMHWWIWIIIIIIII mflIhIRwIIIHVHLI... ..u..u.I.:I . . III.I.I.L.M.I.IM ...-huh... . . . Lfiwwguyfifip. IIIiuuu.I.l.I.u. - ...5.14....IIII.h.I..IIIr ..FwnhrtfimI . ....II IFS-rum. .Ifi. . I . .- . -.L. “6%.? .Iuv. I. .mflgmiJ. ..I... I? I. ...-LII... ..I... I. I . ...... . .....HINIHNIITIHMWIHIWI . . ..IW WVIHIIIHmnIMIhUrvfiHIP. I . ...... ....rI. .I. . VWWIWANIPIWWIF. . «.muIIDFTIIrIIP. .. II ... II II... .1 II"..- I . I. . Luv...»- ..III. I. .. ...... Hm... I'VE ..I». .-I.I.....I.. I . . g II........I..I..II I-IMuHII. ILNI. ..I I. IJIIIII . . III-III 0.95.“..- IIWhWImHI III-....- I . . . . I... I.I»I.I.IIII... ... ...-.. ..I -..-.. 4.... IIIIIuIIIIWIAIIIII I I ..I. . I ..I-...IIIIYIWRFrI. I ..I. “II-LII ....I.II.I.I.I hm”. ..IIIII. Iritnpbfiimflhh. I .UIIxIIIIIII If?!“ I-.- IaPou..I.IIII»«.III.. ... -.I- :IWK . h” ...,Iunl ...)... aria-9.5%”. I --.I... I... I...- I....flnu.... .-.I. I... [tn-AI. III..PI. III-III. -..“? ..I. .0... I... II...” WWW-IIIQIIFI. .Ih I I. .II . IKIIIIF .I. .. .IIIIIhthIIIIIIIIIIIlIidIIIuIP-I. . IIIIP..I¢II.Ifl.IHwIIIIw.|.IIII.IIIIIVII-I“ILQIII ... I I...II4IIIII. . . ......II I... muTl. 6.. , ..I. . 5......“ Ih.wI.I.nu.IIIn.III.hI.III:.IHnuI.u.III.III..IIIII.IHWI -..... -..»...IIIIIII. . «“5... I .I I. It..- ..I HM... ..-. LhquMn-t. .III.I IIIII “NIH...- a... I" ”“1"“ . ’ ’Ik-II'. IVI‘ILOIDI‘IR. I I HildlIVIIII. 13.4.4]: I- (I. I . I...- muk WHI...4L~I8I.T ..I-.1...- IIn...I I rurlflmenII. I-.. . IIN. III..- .PIIIfinFIuIIIqIIIIIWIIIIIIIII III-35W“... ..ILII ThunIthTIIIBmlkflm I. mm...“ II... ”mam-WI. ... fiflwWthlIKflrn/IIIVHIFIIIII.IHIIJHUIUIHHD1.WIHI.IMII.W“I.II- I I .r I. . I-.. _ . .II. . . IIIIII. - a I... .I IrinumuflAh-III huntflrf mnunVIuIIM».uWI.I ...-I“! In“ “III-IND.“ I... ...-I .IIIWE . - . .. . . III-III. IIIIIIIII. I..I.I.. ..IwfiIVLIIIIIIHIlh. III-II “VI-III- \ H . I: I m1IIwwIHIoIIIJ4H¢JIuMIIIIIIIIIIIIII . IIIIrIILIPIr...II I. . . I I III-..IIU PII- .I «IIUIELIADP NIH-1“.“rPIWWHHUmUIquuIMVHMMWII ...I ”I II... ..I. II II.“ ... . .. ..IIIIIuIIhMIMIumINNmI .. . I- ....hh. III“... I. , I. ......I.....IWFIENWNMHIHNIIIfi-Inmtmmrtn v.1...hmrmwnWIInthmhth. ..I... ....I... III...“ ..I. I .... . ., n .. ”Mug-WMWI? “WWII-“N Milan-4..»IIHNMWLIIPIPJHWHIW “...-"..I. . I. .IEgmqmmmmmwmflmflvu .I ....II.I..IIu. I ... IIIqu. IIIIlevwwIIIMWW.‘ «- ..I.. I h5g5- III? “HIM-huh! “Wmmhlmchfl..- flan-II... .- III-...““fiwEfiIbgdqW-IUjflflflnuiI} I .I. ...I - ...IHH...WM4...II:IN.I..MJUII....I. . IIIIIII... I II... I ......I... ”..I .II. II In IMNII-IIWILW Milli-é.“ -..-El! . rtI....I..I..|.-. . “-.IIIIIII.II.IIPILII I III-III“? Inn-nun -..nIIII I--- I I... .u- I . I II. .. III.-. ..I-....“IIFIIIrUHIILINIIIIIIIInI. . ..I-.... ..I III-.IhIIII u 16-4 LIL-I41“! ..qu I IWIII1 u“. II”! . .1 L... . .hHIIIIIII W... . . .. . . “flaw-r“. Sana-“III“..gr-IIF I fig...) .. . .. III-.... . ...-II. ..I” II I “I“-IAIIII -.I... IIwII‘I .I... . . II -.IIhIII-.. I...II..”II.II.II..WP..“.I.”I IIIIIIIIIIHIIIII I. . ..I-..I- IIhIIIIIuIIIII I II... .. .Iu Ivarbuu. hI.II...IuII.HI.u.I. ......- I. “Hfiwmflmnrfimfiéyhfl I..mII.II.I..IWII “17...... "IN ...».nIHIMIrHIIIMIIILUJI. .MHMJIIIIIIII. I. .3294.“ bevrIIInnNuMIIMWNIIIIHHIIIMIL ”wag“ guf.1“d..m..wwmuug n”- I. III I ...”...n..." I. , ...Iwfl NJ“... «III-I .wlfiflwuhnflfhwfi . lawman .....LFIIII. gym-Id“... I..-....vh.k.II....-..II.IH. . I III; 4 IIWIHIIIIumr ......»u .nme. .I IIIIfi.III.. . - . ... I . ..I... I - IIIIIIIIIIWIIIIHII. I . . I III... ....-. .Ww . - , . “I... . -----.IIIIIIIIIIII..-I I. . . . .. I. . uh I. Iflflm... I .I tum. ..I I... . Ifi“...§n.§I.HIMhM.WIIII“.WTnWMI.VIIBIRWh~§ , . , . . .. IUIIII I... . -...aIITII. Iul I .-IWIIII. “PM“. .INhIIIIII.“.Ifl.I.N..II "A“? M I. I II . . I III‘III’I'IIII Ya . l‘ . II I I | . .. .I I. I II»..- - IMMPJIIWFI. , .. . .I... If I III... Jmfinfl ...I...0..“I...thn.II.IIlII.uI. ..I III-III; ..IIIIIIIJII I...- hI-MrkfluIIIIHMHIIHJIIIInIIIII. I. .. I . I IN. ......Ihfl. ..I H“. . um. ..I, I! III“ I III I I. w'lrI‘I‘n. ”Mm; .’. I I “I, . I t. I. III. "II . 1945”,”? ..II. Infill I I I III-.I . -IWIJIIIII. I... ...-I I ’31. :1 u r. JIM-fill: I. t . UR! 1n". I . .II..... . . III ...IIIIIIIIIIIIIIIIIIIIIIHLILQIIIIN Er?! n “I I. . . . . .. l ..IIIIIPIUIHIIIHIIMDW-flnmhmmrwu n1“...fl..W..n.I..4..IIII.I..I..b.II.”.W.wII ..IHIIlInnIIPMInIIKIML“. ......IIIIII... I IIuIM...H. fig .. . .. . .. ”WWI-......IHIMMHMIIMWMWHIHWIHIW,rI...I.h.....|lII.I.I.I..“IJIIIflIII..II.WMMI. “am... ..I... I . . IIIIImIMQIIIUHYuLHIEIHIJ. I III-.. IuIIIIhIIINIIIIII. IIIIIIII IIIflrIIIIIIIIIII H n ..I .I I .l I. I . .-.-(IIIIIIII .mIIIIIIruHIJIIIIIIIIIflI III. III. wLIIIIIII. .III III. .-I! I. .II... h.“.. I“. II'IIIIIIJH! I --III..0IIH.IIIII.ILII% nw~II4IIII . IIIHIJIHUINIII .IIII . -IIIIhIhIIIIIuwIIILIIhIIIIJIHI . III”... .. .. .... «HI... ..I. I I .MIIIII III-III...“ . “IHJILHIIIIM. . .“Idfli... ..I II... IaIPIIIInhIhfihILI. I I I. EIIHI’IWIm l"x III.“ E. gtmwm i%,bu . .....IIIII ......»II..I.I.II "{gEIdrnrfllvngfl““.I “II-......INHPIIIIIIIHIIQ. .UHquII . . . . . .. . ..- I..- .. ...I. “hm“IIuNWHIIIIIIflg .IIIIII I... . ..I! . . . I . fifl.l(.IlII.I. . I. IN. I .. ...IIIIIrIIII..u.III.I.~II.I.r I... ....IuIIWIIIIIIIflIAIIIIn’IIIIIIIIIIIIII . 15...”... .I I II. ..I--.....IIIIIII .. I. I. I. I IIIIIIIIII , I... «fin..- -. ..I ....I . . . I... ..I... III. pdghfigiIIi “IIIEQMNRwMIMMfiIIIHNIUI ..I I IUIIII “In...“ .-.WLNPMNII I...“ “In...“ I . «III-III}... LVN-IIIwHIWLHIIUHhIIIIHWUIWLIIIHNUIWHnIWIgfikIIII. III.- . . . .hunuflmnu. I. III. III I. “..I... .FIIIILI"... . . . IVIIIIIIIIII.I.I.III.EIIIIIIIIIII’9. III MIMI-.... I. . IIIPII. is . ...I . . . “...”ng . II. ..I....I.I..I.IIIIIII.I\.IIIII.I ......IIIHMUII- III... .IIIIIIIIIIII . I . . . . I ... ...... ..I III-......- . r“...IIIIIIIIII.II.III4Is tflVUHIAIthI . : .. M ..u. I . . .. . . I “gig".IfiIIn-“wi I... .I . I. u . II.I.I..IIII.IIIII.r~II.IwMMI.IdIK. 31%... I . I. .. . IIIUIIII. . . . I .. ..rhjnvv. I) ...... ... I... I land”... n “I I IWRVII. “I... r414... H . ..I I II IIHI-HUHINIIIRINIIILIV- Egghhudhhhfllgwy . IIIIIIIIIIIInIIIIII. . .. I..I.I...II.I IIIIIIIIthWIHI- . IIIIIIIIIH “4hr“... «..I I“-..HI|I.III.II.IIII..I .... IEI I . . ..IflmHHII , IIInIIII... -IIIIIIInufiIIIIIIIIIIIflIIIIflII. I - “mg..- .u... . III II IIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII. . u' ‘z‘.“.?»?\".".'\"-“L mm...‘ . . “I”... IrIII-xIIIIIPILIIIIIIIhIIIhIIIMuuNI n... III-..I ..I. II... I “III-...... I- r ..I . hIIIIIII ..IIIIIIIINIIIIIIII. “In“ hIIIIIIII. ..- II. ... ..IIIIHfl-IWUIIWH g“hwh.l .III I-I.- II- . I!“ ..IIUUNI—mnNIIrIHWII.“.IIUN ... III-fidqi . i“flfi“w%fl “Bum-I. . D. .II. III-...... III... ”WWI“. .r 4.8.- . .. “Gun... “II I . I.“ .I .III .1“ I n“? -(IIINIIVINIIIIWI‘IHUIIHIIIIIIIIII “(III-“WWII”. .u. .IIWJNHIIAIIHHHHI... “u“..«UHwWJHM. III-I . . IUIIIIII. . II.IIIIIIIIIII .. $1."? I. II .IIInyIIWIImII. .....HWIIquIIIIIIiI... I I . . III...I..II.II IIIHW. . . .---..IIIII .-IIIIIIIIVIII. III-I. III. .III.I|. 1. ...MNHIMIhUIII. . III.nu.....l.III..I.III..I.I..IIIIv.IIIIIJnH... “I ... “H“EIIIIIIIIIIII...II...I.-III.III.II.I -.I I. .. ..IIl-IIIEIII. .I.. ..I-III. . . I ...... . .. .. ..I III- .... ...III4. . “In ..I... .-I. ..IIIIIIIIIHHIIIAIIII.II.III III... “I. .-. ..I . . .. I....IIIIII..II I III. I“... ..IIMIIIII .-.IIIII-IIII. I -...I.I.III. III-..I... ...MIINNNAII III-.... .I.. ..I..- (”Ignhh Dir-mg um. E“. I IIQIIIIIIIIIIMIIIIII .NIIIIII. . . . .IINII . “III. - - (““IgIIIRuI IIIIIIIIIIIIIIIIIII .- -II II... I . III-Ii“... JD III-“III-.-“ III-....-. III I I -LIINIIunIIIIIIIIIIIIIII . I: “I IIIHIIIIIIIIHIHIIIIII . I III... . I4... I “Ida-III. ..IIIRIIII ..IIIIJHVIII. IIIIIIIIL. -..... .UIIIIIILMAIHMIUn «HI-(III... IIWIWIVIIHIIIIIIII I... ..IAVIIIIIIIUIIIIIIIIIIIIIIIHIIIIIIHIIIIIIIII .lIIIIII PIIIIIII Pan-III .In..“H...lU.IIII.uII.u4II\.nIIIIIIu-IIILHL.IIIIIU ....n. .- 0...“. I. II.III..II..II.IIHmMHI~7LIILUIIIIIIHI“IIIIIINI II"... .llIIIILI “3‘ I.“ ..IIIIIIIIIIII I. ..I.“ u. I. I I IIIIIII II-.. . . I 1.6“”. ...... .III. III. .I .. III-III:- II.WI ..I. $§Wfij§§fliIfluyf - - II III JKIJAIIIIIJIIIIIII . IIIIIIIIIILIHRIPII [IN-IIIIIIIIIIIIWIIIIIIIIIIIIIIIII. III-IIIIIII .InIIIIII.II “g...“ ..I. (“..IHIIIUV“... .-II TH... III-IIIIIIIII ”II-..ILIIIIIII. “Pill-III?!“ I .I I ..II. IIIIFIIIIIIIhUqunmmNIIINInII..I.m.IIIII.. . .I. III...“ . -IIIIIIIIIIIII. ...-4V... . I... III.-.“ IIIIII.I.-..I. .I..-..I.H......II.-I..I. Inn-II.- IJWIIIIIIIIIII ..III...IIIIII|.I.I ...-“Ill.- -. III. . .III. .IHII... IIIIIIII. II....II.II.II.IIIIIIfiN.IIIIIIII.II“.IHI.I. ..ILIHJ........IIII’....II\. III. III. I“ I “fig““yhu’5pl . . -....II II. ..IIIIIIIIIIJII .II..II...I.. IPIII III. I .I ...-III. .. . . .. . wIwquIIImkIquwuoInunIIIIIIIIIIIIIIIInIritIIIIII THESIS This is to certify that the .. thesis entitled PART I. Extended Huckel Molecular Orbital Calculations of the Relative Stability of Isotrindenetrione PART II. Preparation of and Estimation of the Strain Energy in the Diels-Alder Dimers of Two Simple c-Face Annulated Cyclopentadienones presented by 1 Dennis K. Klipa l has been accepted towards fulfillment of the requirements for Ph.D. Chemistry degree in 4402mm? I Major professor Date January 28, 1981 0-7639 w: 25¢ per day per item RETURNING LIBRARY MATERIALS : _________..___——- Place in book return to remove charge from circulation records PART I EXTENDED HDCKEL MOLECULAR ORBITAL CALCULATIONS OF THE RELATIVE STABILITY OF ISOTRINDENETRIONE PART II PREPARATION OF AND ESTIMATION OF THE STRAIN ENERGY IN THE DIELS-ALDER DIMERS OF TWO SIMPLE C-FACE ANNULATED CYCLOPENTADIENONES By Dennis K. Klipa A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1981 64/6 9’ 7? ABSTRACT PART I EXTENDED HUCKEL MOLECULAR ORBITAL CALCULATIONS OF THE RELATIVE STABILITY OF ISOTRINDENETRIONE PART II PREPARATION OF AND ESTIMATION OF THE STRAIN ENERGY IN THE DIELS-ALDER DIMERS OF TWO SIMPLE C—FACE ANNULATED CYCLOPENTADIENONES By Dennis K. Klipa In this thesis two factors which might influence the stability of the unknown isotrindenetrione (2%) were studied. In the first part of this thesis Extended Hickel Molecu- lar Orbital Theory was used to predict whether 2% would be stabilized electronically as a result of electron delocaliza— tion in the central six—membered ring. A comparison of the calculated pi orbital energies of a series of structures including cyclopentadienone (ll), 2e, benzene (g%) and tri— phenylene (ga) with the pi orbital energies of a second series of acyclic structures possessing the corresponding number of double bonds (all trans) showed that although 3e Dennis K. Klipa is predicted to be antiaromatic there does appear to be some relative electronic stabilization. A second comparison was made in which the pi orbital energy per "unit" was compared for a series of structures consisting of a single unit (e.g., benzene), two units (e.g., biphenylene) and three units (e.g., triphenylene). The comparison was made for the benzene, thiophene, furan and cyclopentadienone systems with the latter three units being coupled across their C—faces. This comparison showed that for the benzene, thiophene and furan systems there is essentially no net increase in stabilization per unit for triphenylene (2a), benzo[l,2-C:3,A-C':5,6—C"]trithiophene (OR) and benzo[1,2-C:3,A—C':5,6—C"]trifuran (2%). This comparison does, however, predict an increase in stabiliza- tion for 2% relative to 11 of 12 Kcal per mole per unit. In the second part of this thesis a new method for retarding cyclopentadienone dimerization was explored. The strain which might develop upon Diels-Alder dimeriza- tion of 2% due to the presence of the C-face bridge was examined by studying model dimers (cndo—pentacyclo— [9.6.1.02’10.05,10.012’17]octadeca-A,12(l7)—diene—3,18- dione (gg) and cndo-pentacyclo[8.5.1.0239.O5=9.011315] hexadeca-A,1l(15)—diene—3,16—dione (%%)). These are dimers of two simple C—face annulated cyclopentadienones (U,5,6,7— tetrahydro-2-indenone (TB) and 5,6-dihydro—2(AH)-penta1enone (lg), reSpectively). Cyclopentadienone IQ was prepared by the bromination Dennis K. Klipa of 1,“,5,6,7,7a—hexahydro-2-indenone (fig) with N-bromo- succinimide (NBS) followed by dehydrobromination with po— tassium tert-butoxide in tetrahydrofuran (THF) at —78°C. Monomeric %§ could not be observed or trapped but was isolated as the dimer was prepared via A,5,6,7a-tetra- 8KO8O % Cyclopentadienone l hydro-2(1H)-pentalenone (1%). Enone 1% was prepared by the following route. Alkylation of 2-carboethoxycyclopentanone with prooargyl bromide gave ethyl 1-(2—propynyl)—2-oxocyclo— pentanecarboxylate (Qé)' Mercuric ion catalyzed hydration of the triple bond gave ethyl l—acetonyl—2-oxocyclopentane- carboxylate (QQ). Intramolecular aldol condensation using sodium hydride in refluxing toluene gave 2,3,A,5-tetrahydro- 5—oxo—3a(lH)—penta1enecarboxylate (l2). Saponification and decarboxylation of the resulting acid gave Kg. Bromination of 1% with NBS followed by dehydrobromination with potassium tert—butoxide in THF at -78°C presumably gave 1% which was not directly observed but isolated as the dimer gl. The kinetics of decarbonylation of £9 and 3% in deuterated benzene were measured at three elevated tem- peratures. Decarbonylation of £9 at 150 to 173°C gave a complex mixture of products which when heated to 195°C was converted nearly quantitatively to l,2,5,6,7,8-hexa- hydrospiroEbenz[f]indene—3,1'—cyclopentane]—1—one (2%)° Decarbonylation of £1 at 82 to 109° gave tetracyclo- [10.3.O.0139.03’7J-pentadeca—2,7,ll-trien—1O—one (28). The first order rate constants for the thermal Dennis K. Klipa decarbonylation of 20 and 2% at 109°C are 1.2 x 10—6 sec—1 (extrapolated) and 2050 x 10_6 sec—l, respectively. The activation parameters for the decarbonylation of 20 and 21 ,5 are AH = 35.5 i 1.2 Kcal/mole (20), 23.5 i 0.5 Kcal/mole (21); ASE = 6.3 i 2.7 e.u. (20), -9.8 : 1.u e.u. (21). ’\;'\I ’L’b NW The AAHC indicates that 21 is more strained than 20 by a ’b’b ’L’L minimum of 12 Kcal/mole. The single crystal X-ray structures of 20 and 21 were determined and the strain in 2% is discussed in terms of bond distortions. ACKNOWLEDGMENTS I wish to express my sincere appreciation to Professor Harold Hart for his encouragement, guidance and confidence in me. Appreciation is extended to the National Science Founda— tion and National Institutes of Health for financial support in the form of research assistantships. Finally and mostly, I thank my wife, Jackie, for her patience, understanding and continued support in spite of the innumerable sacrifices she has made these past few years. ii TABLE OF CONTENTS Chapter LIST OF TABLES. LIST OF FIGURES INTRODUCTION. RESULTS AND DISCUSSION. PART I - EXTENDED HUCKEL MOLECULAR ORBITAL CALCULATIONS OF THE RELATIVE STABILITY OF ISO- TRINDENETRIONE. Conclusions ............................. PART II — PREPARATION OF AND ESTIMATION OF THE STRAIN ENERGY IN THE DIELS-ALDER DIMERS OF TWO SIMPLE C—FACE ANNULATED CYCLOPENTADIENONES Section A. Preparation of endo-pentacyclo- [9.6.1.02’10.o5’10.012’17jocta— deca-A,12(l7)—diene—3,l8-dieone (20) via A,5,6,7-tetrahydro-2— indenone Attempts Using d-Functionalization Use of B—Functionalization Section B. Preparation of endo—pentacyclo- [8.5.1.02,9.05’9.Oll’15]hexa— deca-A,1l(l5)-diene-3,16-dione (21) via 5,6-dihydro-2(AH)—pentalenone (l9) iii Page viii 22 23 23 2A Al A6 Chapter Page Section C. Decarbonylation Activation Enthalpy as a Measure of the Relative Strain in cndo—penta— cyclo [9.6.1.02’10.05’1O.012’17]— octadeca-A,12(17)-diene—3,16- dione (20) and endo—pentacyclo- [8.5.1.02’9.05’9.011’15]hexadeca— U,ll(l5)-diene—3,lb-dione (g1) . . 52 Decarbonylation of endo-pentacyclo— [9.6.1.02’10.05’1O.012’l7]octadeca- A,12(17)—diene—3,16-dione (20) and 2 9 m 9 11 15 cndo—pentacyclo[8.5.l.0 ’ .0 ’ .0 ’ ] hexadeca—A,ll(l5)-diene—3,16—dione (2%) . . . . . . . . . . . . . . . . . . . 52 Discussion of the Kinetic Results. . . . . . . . . . . . . . . . . . 68 Section D. Single Crystal X—Ray Structures of cndo-pentacyclo- [9.6.1.02’10.05’1O.ol2’17]- octadeca—A,l2(l7)—diene-3,l8-dione (20) and cndo-pentacyclo— [8.5.1.02’9.05’9.o”’lljhexadeca- U,12(l5)-diene—3,16—dione (2%) . . 7A Crystal Data . . . . . . . . . . . . . . . 7“ Discussion of X—ray Structures . . . . . . 75 EXPERIMENTAL. . . . . . . . . . . . . . . . . . . . 8“ General. . . . . . . . . . . . . . . . . . . . 8A Ethyl 1—(2—propynyl)-2-oxocyclo— hexanecarboxylate (g5) . . . . . . . . . . . . 85 Ethyl l-acetonyl-2—oxocyclo— hexanecarboxylate (Q5) . . . . . . . . . . . . 86 iv Chapter Page 1,A,5,6,7,7a—hexahydro—2— indenone (A0). . . . . . . . . . . . . . . . . 87 Attempted Synthesis of l—bromo— l,A,5,6,7,7a—hexahydro—2—indenone (as) 87 Attempted Synthesis of l-phenyl- seleno—1,A,5,6,7,7a-hexahydro— 2—indenone (A9). . . . . . . . . . . . . . . . 89 2-Trimethylsilyloxy—A,5,6,7- tetrahydro[3aH]-indene (52) 90 2—Trimethy1silyloxy-A,5,6,7— tetrahydro—[IHJ—indene (5%). . . . . . . . . . 91 2,A,6—Triaza-3,5,13—trioxo— A—phenyltetracyclo— [5.5.2.01’8.02’6]tetra— decane (51). . . . . . . . . . . . . . . . . . 92 3, 3a, A ,5 6 ,7 Hexahydro— 3- (l, l, l- triphenylmethyl)- 2— indenone (62). . . . . . . . . . . . . . 92 3,3a,u,5,6,7—Hexahydro—2- indenone dimethylhydrazone (71) . . . . . . . . . . . . . . . . . . . . . 9“ an, Reaction of 3,3a,U,5,6,7— hexahydro—2—indenone dimethyl hydrazone (71) with dichlorodi— cyanoquinone (DDQ) . . . . . . . . . . . . . . 95 7a- Bromo- 1, A ,5, 6, 7, 7a— hexahydro— 2— indenone (76).. . . . . . . . . 96 endo-Pentacyclo[9.6.1.02’1O.05’1O.012’l7] octadeca-A,12(17)—diene—3,18-dione (20). . . . 97 Reaction of 7a- bromo— l, A 5, 6, 7, 7a— hexahydro- 2- indenone (76) with triethylamine in ether . . . . . . . . . . 99 Reaction of 7a— bromo— 1, A ,5 6, 7, 7a— hexahydro— 2— indenone (76) with neat triethylamine. . . . . . . . . . . . . . . . 100 Chapter Page Reaction of 7a—bromo-1,A,5,6,7,7a- hexahydro—2—indenone ( 6) with potas— sium tert—butoxide at °(C. . . . . . . . . . . 100 Reaction of 7a—bromo-1,A,5,6,7,7a— hexahydro—2—indenone ( 6) with lithium hexamethyldisilazide in ether/hexane (1:1). . . . . . . . . . . . . . . . . . . . . 102 Reaction of 7a—bromo—1,A,5,6,7,7a- hexahydro—2-indenone ( 6) with lithium hexamethyldisi azide in THF. . . . . . . . . . . . . . . . . . . . . . 103 Attempted trapping of 18 with N—phenylmaleimide. . . . . . . . . . . . . . . 103 Attempted trapping of 18 with dimethylacetylene dicarboxylate (DMAD) . . . . . . . . . . . . . . 10A Attempted trapping of 18 with ”\J’b Cyclopentadiene. I O O . C C O O O O O . C O C 105 Ethyl l—(2-pr0pynyl)—2-oxo- cyclopentanecarboxylate (86) . . . . . . . . . 105 Ethyl 1—acetony1—2—oxocyclo- pentanecarboxylate (8%). . . . . . . . . . . . 107 Hg++ catalyst on Dowex—50 H+ Resin. . . . . . . . . . . . . . . . . . . . . 108 A,5,6,6a—Tetrahydro—2— (1H)-pentalenone (78). . . . . . . . . . . . . 108 Attempts to improve the yield of A,5,6,6a-tetrahydro—2— (1H)-penta1enone (Z8). . . . . . . . . . . . . 110 Potassium hydride in refluxing toluene. . . . . . . . . 110 Potassium tert—butoxide in refluxing toluene . . . . . . . 111 Potassium tert-butoxide in refluxing THF . . . . . . . . . 111 vi Chapter Page Potassium tert—butoxide in refluxing THF (inverse addition). . . . . . . . . . . . . 112 Potassium tert—butoxide in refluxing ether. . . . . . . . . . 112 cndo-Pentacyclo[8.5.1.02’9.05’9.011’15] hexadeca—A,11(15)—diene—3,16-dione (21). . . . 113 ’\J’\; Attempted trapping of .19 with cyclopentene . . . . . . . . . . 115 Kinetic measurement of the de— carbonylation of endo-pentacyclo- [9.6.1.02’10.05’10.012’l7]octadeca- A,l2(17)—diene—3,l8—dione (2g) and endo-pentacycloI8.5.1.02’9.0 ’9.Oll’15]— hexadeca-A,1l(15)—diene—3,16—dione (21). . . . 116 Thermolysis of endo—pentacyclo— [9.6.1.02’10.05’10.012’171—octadeca- A,12(l7)-diene-3,l8—dione (20) . . . . . . . . 117 Thermolysis of endo—pentacyclo— [8.5.1.02’9.05’9.011’15]-hexadeca— A,11(15)-diene-3,16-dione (g1) . . . . . . . . 119 APPENDIX — Single Crystal X-ray Bond Lengths Bond Angles and Positional Parameters for 20 and 21 . . . . . . . . 121 REFERENCES. . . . . . . . . . . . . . . . . . . . . 13A vii LIST OF TABLES Table Page I Molecular Geometries of Structures Examined by the EHMO Method. . . . . . 12 II a Values of EHMO Calculations. . . . . 16 III Solvents and Temperatures Used in the Constant Temperature Bath for Decarbonylation Reactions. . . . . 53 IV Extent of Decarbonylation (X) of gg and g1 . . . . . . . . . . . . . 61 V First Order Decarbonylation Rate Constants for 20 and 21 . . . . . 65 VI Decarbonylation Parameters for 7-Norborenone Derivatives. . . . . . . 67 Appendix Fractional Atomic Coordinates for 20 . . . . . . . . . . . . . . . . 121 mm Bond Lengths for 20. . . . . . . . . . 123 Bond Angles for 20 . . . . . . . . . . 12A Fractional Atomic Coordinates for 21 . . . . . . . . . . . . . . . . 130 ’Vb viii Table Page Bond Lengths for 21. . . . . . . . . . . 131 Bond Angles for 21 . . . . . . . . . . . 132 ix LIST OF FIGURES Figure Page 1 Calculated ring carbon—carbon bond lengths (A) for the parent radialenes . . . . . . . . . . . . . . 5 2 Comparison of ZEN between cyclic and acyclic structures (Kcal/mole) . . . . 18 3 Comparison of ZEN (Kcal/mole) per ring for cyclic structures 35, SQ and 2 where x = —CH=CH—, S, O, C=O. . . . . . . . . . . . . . . . . 21 A Two schematic views of the constant temperature bath for decarbonylation reactions. . . . . . . 5A 5 Partial 250 MHz lHMR spectra of 20, 21 and their decarbonyla- mm ’1: tion products. . . . . . . . . . . . . . 56 6 Plots of the decarbonylation of 20 at 150.5, 162, 173°C. . . . . . . . 63 7 Plots of the decarbonylation of 21 at 82, 97, and 109°C. . . . . . . . 6A 8 Bond Lengths for 29 and 21 . . . . . . 76 9 Bond angles for 20 and 21. . . . . . . 77 Figure 10 11 Page ORTEP stereoscopic views of 20 and 21 . . . . . . . . . . . . . . . . . 78 Disordered orientations of C(lA) and (15) in crystals of 20 . . . . . . . 79 xi INTRODUCTION Radialenes are carbocyclic compounds per—substituted with exocyclic carbon-carbon double bonds. Although pentaradialene (3) remains unknown, triradialene1 (I), tetraradialene2 (2) and hexaradialene3 (A) have been pre- pared and found to be quite reactive hydrocarbons. Hexa- radialene, for example, is only stable in dilute solution, and is oxidized rapidly on exposure to air. axxx I 2 Alkyl and aryl substituents stabilize radialenes. Thus hexamethyltriradialeneu (5), heptapheny1tetraradialene5 (6), hexamethylhexaradialene6 (z) and hexaethylhexarad- ialene7 (8) are all stable crystalline compounds. Another way to stabilize hexaradialenes is to introduce a bridging group, X, (9) between the termini of the exo-. cyclic double bonds. This permits electronic interaction of the n orbitals at both ends of the double bonds. The oldest and most stable exemplification of this concept O 9 9a a, x = —CH=CH— b, x = S c, x = Se d, x = 0 e, x = C=O f, x = N—R is triphenylene (9a) in which the exocyclic double bonds have been incorporated into three benzenoid rings. Re— cently the trisbenzothiophene8 (9p), trisbenzoselenophene9a )9b and trisbenzopyrrole90 (9g, (9g), trisbenzofuran (9d R = benzyl) have been prepared and shown to be quite stable crystalline compounds. The sulfur-containing compound (2Q) undergoes normal electrophilic aromatic substitution reactions. Whereas X bridges with unshared electron pairs (such as sulfur, selenium or oxygen) result in the formation of conjugated aromatic rings (9p, 28 and 9%, respectively), the introduction of carbonyl bridges would give the unknown isotrindenetrione (as) which would consist of three linked antiaromatic cyclopentadienone rings. The goal of this research was to investigate whether cyclopentadienones linked in this way would possess greater stability than three separate cyclopentadienone moieties. Two factors might work to enhance the stability of such a system. The first factor involves electronic stabilization. Would the arrangement of the six p—type atomic orbitals of the six central carbon atoms result in any electronic "benzenoid" resonance stabilization? One resonance con— tributor which can be drawn is the tris—oxyallyl structure (18) in which the central ring is benzenoid. m . o I H o C) o O O - 5 - 9e IO This question has been considered previously10 for the parent radialenes l, 2, 3 and A. If such delocalization were to occur in hexaradialene (A), a shortening of the ring carbon-carbon bond lengths relative to the correspond— ing bond lengths in tri—, tetra—, and pentaradialene would be expected. Pariser—Parr-Pople SCF calculations, which determine minimum energy geometries, show (Figure 1) no significant difference in the bond lengths throughout the parent radialene series, suggesting that "benzenoid" reson- ance is not important in hexaradialene.lO But since the introduction of bridging groups alters the chemistry of hexaradialene, perhaps "benzenoid" resonance stabiliza- tion is significant in these bridged radialenes. In the first part of this thesis Extended HUckel M.O. theory is used to determine whether any electronic stabilization A >"< * (1.4641450 I 2 3 Figure 1. Calculated ring carbon-carbon bond lengths (A) for the parent radialenes.10 of this type is predicted for isotrindentrione (9%). The second factor is steric in nature and involves the well known cyclopentadienone dimerization. Cyclopenta— dienones, being antifaromatic, are highly reactive.11 Many cyclopentadienones, in the absence of other reagents, undergo a facile [Ans + 2ns] dimerization due to their high energy and cisoid diene structure. For example, even though the parent cyclopentadienone (11) has been detected by low temperature matrix infrared spectroscopy12 and by microwave spectroscopy13 using flash vacuum pyro— lysis, it has never been isolated as the monomer. . l 0 4+2 )-> 11 Most cyclopentadienones which are monomeric, such as 1A 1Ab At, Aié , 1A18 and 1§,19 owe their stability to bulky I2 13 I4 15 (R: aryLalkyl) l6 substituents which hinder dimerization. Ceface annulated cyclopentadienones (16) may possess additional stabilization if the bridging group (x) is short or is constrained to be coplanar with the cyclopentadienone ring, because ring strain may develop upon dimerization. The strain in 17 results from distortion of optimum bond lengths and bond angles due to geometrical constraints of the x' bridge. o I O 4+2 , 16 I7 This strain may contribute to the stability of 1A and 15 but more importantly it may also contribute to the stab- ility of 9%. Unfortunately, the magnitude of this effect in known systems is masked by additional steric and elec— tronic effects. In the second part of this thesis the magnitude of this strain, in the absence of any special electronic or steric effects, has been approximated by preparing and estimating the strain energy in the dimers (20 and 21) of the cyclopentadienones l8 and 19 (respec- tively). I9 RESULTS AND DISCUSSION PART I EXTENDED HUCKEL MOLECULAR ORBITAL CALCULATIONS OF THE RELATIVE STABILITY OF ISOTRINDENETRIONE Hfickel Molecular Orbital (HMO) theory has been used to explain and predict the stabilities, physical proper— ties and chemical reactivities of organic n—systems,21 but has been limited to homoatomic, planar structures. HMO 22 to include theory was extended (EHMO) in 1963 by Hoffmann non-planar structures, differential overlap and, most im- portantly for this study, heteroatoms. Despite these im— provements the EHMO method still has limitations and short— comings. In order to put the results of these calculations into proper perspective, several aspects of the EHMO method should be clarified. The EHMO method does not calculate absolute molecular energies because it fails to include several factors. Among these factors are non-valence shell electrons, nuclear—nuclear repulsion and electron—electron repulsion. In addition, the EHMO method is known to overemphasize "steric" interactions and to inaccurately estimate ring strain.22 Despite these limitations the EHMO method has been quite successful at determining relative stabilities. Thus while the absolute energy values and even the dif— ferences in energies between structures may not be accurate, the relative stabilities of closely related structures are generally in accord with experimental observations. In other words, the stability of a given structure can be estimated by the EHMO method but only in reference to another structure. This is the approach used here to determine whether EHMO theory predicts any stabilization for 9%. 9e One other aspect must be considered before discussing the calculations. Since we are interested only in the n resonance stabilization of 9%, we will compare only the 10 relative n orbital energies of these systems. Since all structures considered in this study are planar and the n molecular orbitals are orthogonal to all 0 molecular orbitals, this restriction will not introduce significant errors. In fact this condition is necessary in order to make a valid comparison because while the number of n bonds is constant or an integral multiple, the number of o C-C bonds and g C-H bonds is not. This is illustrated in the comparison of trans-1,3,5—hexatriene (22) and ben— zene (23). Triene 22 has eight C—H bonds and five 0 C-C ¢¢¢’\\v¢;?\\V§;; :>. 122 .23 bonds while benzene has only six C—H bonds and six 0 C—C bonds. Thus a comparison of the o orbital energies would be meaningless in the context of this study. The EHMO calculations were carried out using a com- puter program written by Professor J. F. Harrison23 of Michigan State University. The program is fairly simple to use as the only inputs necessary are: 11 1. Accurate atomic coordinates. 2. Energy values of valence shell atomic orbitals. (a-VaIUES) Accurate atomic coordinates for known structures were taken from x—ray or microwave data. The coordinates for unknown structures were estimated based on similar struc- tures in the literature. The dimensions of the structures used in this study are given in Table I. The energy values of valence shell atomic orbitals were kindly provided by Professor Harrison or were taken from the literature29 and are listed in Table II. The molecular orbital energy values are calculated before the electrons are entered into the calculations. Therefore, the orbital energies were multiplied by a factor of 2 to obtain the energies of the doubly occupied orbitals. To determine whether EHMO theory predicts any electronic stabilization for 9%, as a result of possible n electron delocalization in the central ring (10), two separate com- parisons were made. 6— 9!: IO 12 .oomA ppm mphdposmpm 0AAomom GA mmchm ccon AA¢*** .conpmo pmnp 0p popzon macaw 03p map pew Op pmnomppw mam mos» conpwo map mp pchEpmpmp kocm map poomfin wpcon mlo Ha¢** .m wo.A mam anpmatA econ muo AA<* :: ANA 00A NOA AOA om OAN.A 0:3.A 0mm.A 0:2.A AAm.A am AN --- 00A AAA SOA om lllll AAm.A 2mm.A 0:2.A AAm.A mm AN an- ..... wOA AAA SOA ..... ruin: on:.A amm.A AAm.A Nm mN 1.. Na AAA NAA om In--- AAA.A mmm.A 0:3.A :Am.A Am mN In- in--- NAA AAA Nm uuuuuuuuuu mN:.A OAm.A AAA.A om mN In- N.NNA N.mAA m.NNA om mwm.A mN:.A NAm.A mNA.A :Am.A mN u- *** 0:2.A 0mm.A OAN.A 0:2.A omm.A NN u: *** 02A.A 0mm.A OAN.A 0:2.A 0mm.A AN In *** IIIIIIIIII ozm.A 0:3.A omm.a mm I- *** ..... 0AA.A omm.A OAA.A omm.A mN II *** IIIII 02:.H omm.H 0:2.A omm.H :m aN --- --- --- --- ONA .......... ----- ----- AAA.A AN *** ll--- lulu: 0mm.A 0:2.A omm.A NN I- ..I NOA AOA SQA ANA ..... QA=.A omm.A 02:.A ozN.A AA u- ANA mOA NOA AQA ONA oqN.A 02A.A 0mm.A 0AA.A 0:3.A am AN In- ©OA AAA ©OA ONA lllll AAm.A Amm.A 03:.A OAA.A em mN --- Nm AAA NAA QNA lllll AAA.A mom.A 0:2.A 0:2.A am AN 1.. mAA ANA ONA ONA ON:.A 0mm.A ON:.A OA:.A OAA.A am .w w > m a m U o n m .02 .eam **AoV tAmEA ecom *Amv npmcmq acom mmzo .ponpoz 02mm esp an pocAmem moLSpozppm mo mwfihpofiomo ANASOvoz .H oAnt 13 a'/h/d// / / / // 25 28 a 29 1A ‘\° lb / c \ \J II \‘b ' \¢ 31 ' 34 15 16 Table II. a Values of EHMO Calculations. Atom Orbital Value H ls —13.6 C 2s -2l.A C 2p -11.A 0 2s -35.57 0 2p —l8.03 S 38 -21.13 S 3p -13.31 17 The first comparison was made between cyclic struc- tures containing 1, 2 and 3 "units" (35, 36 and 9) and acyclic structures containing the same number of double bonds in the all prans configuration (37, 38 and 39, respectively). The EHMO calculations were performed on ("I 8 x) —-> VXWxQ ‘ ax/WXWxQ 37 38 39 the benzene series (23, 29, 9a) and the cyclopentadienone series (11, 3A, 9%). The results of these comparisons are shown in Figure 2. An energy value of zero has been assigned to the acyclic structures. The energy difference between the acyclic and corresponding cyclic structures is listed below each cyclic structure. 18 23 Q 29” 9a "18.0 '22.2 -30 o W W \\\\\\\ \ 22 24 I:>=0 34 +14.4 +235 26 27 28 Figure 2. Comparison of ZEn between cyclic and acyclic structures (Kcal/mole). 19 If a n system is stabilized in going from an acyclic to a cyclic array, then the total n energy decreases (- sign). If there is a net destabilization, the total n energy increases (+ sign). In the benzene series (23, 29, 9%) a decrease in energy is observed for each case as expected for this well known aromatic series. In the cyclopentadienone series (11, 3A, 9e) an increase in energy is observed in each case. Thus EHMO theory predicts not only that 11 is antiaromatic as expected but also that 3A and 9% are also antiaromatic. Although EHMO theory predicts that 9g will be anti— aromatic, closer examination of the data shows that 9g possesses some intrinsic stabilization. In the benzene series (23, 29, 9%) there is a steady increase in the stabilization energy (-l8.0, —22.2, —30.0). Conversely, if there were no additional resonance stabilization in 9% a steady increase in the destabilization energy for the cyclopentadienone series (11, 3A, 9%) would be expected. There is an increase in destabilization energy when going from 11 to 3A (+1A.Ato-+23.6) but there is no further in— crease in destabilization energy when going from 3% to 9% (+23.6 to +22.8). This result suggests that the three cyc10pentadienone units linked in this way (9%) are stabi— lized by n electron resonance stabilization. The second comparison was made by setting the total n energy for 35 (x = —CH=CH-, S, O, C=O) equal to zero 20 and observing how the total n energy per ring varies when two units are brought together to form 36 and when three units are combined to form 2. The results of these EHMO calculations are shown in Figure 3. I If there were no additional n resonance stabilization in 2 as a result of linking three units of 35 in this way, then the difference in ER“ per ring for 9 relative to 35 should be zero. This is approximately what is seen for the benzene, thiophene and furan systems. The cyclopenta— dienone system, however, shows a significant stabilization for 2% (12.2 kcal/mole/ring) relative to an isolated cyclo— pentadienone unit (11). Although the absolute value of this stabilization cannot be known with high accuracy, there is little doubt that EHMO theory does predict sig— nificant n electron resonance stabilization for 2%. One may ask why as is predicted to have additional resonance stabilization, while 9a, 9b and 9d are not. ’b’b ’b’b ”Vb 23 29 9a 0 +3.4 Ar ‘- E) C13 \ / 0 +3.8 31 E) CI 0 +0.4 O 00 e- 3 Figure 3. Comparison of ZEN (Kcal/mole) per ring for cyclic structures and where x = -CH=CH-, s, o, 0:88’ 88 2 22 There is one simple and perhaps naive explanation. Struc— tures 9%, 9b and 9d are made up of three units which are "UV "\1 already aromatic. In order for them to participate in central ring "benzenoid" type resonance each individual unit would have to be disturbed. This would result in the loss of some of that aromaticity with a resultant increase in total energy. On the other hand 9g is made up of three antiaromatic cyclopentadienone rings and thus would lose no individual ring stabilization by participation in a resonance structure such as 10. ’VD Conclusions Both comparisons made in this study indicate that EHMO theory does predict significant stabilization for 9% relative to three isolated cyclopentadienone moieties. Although confidence in the prediction of stabilization is high, the accuracy of the magnitude of that stabilization is not as high. This is an intrinsic characteristic of EHMO calculations. Finally, 9g, although stabilized some— what over its "monomeric" units, is still predicted to be antiaromatic. PART II PREPARATION OF AND ESTIMATION OF THE STRAIN ENERGY IN THE DIELS-ALDER DIMERS OF TWO SIMPLE C—FACE ANNULATED CYCLOPENTADIENONES Section A. Preparation of endo—pentacyclo— [9.6.1.02’10.05’10.012’17]octadeca—A,l2(l7)-diene— 3,18-dieone (20) via A,5,6,7—tetrahydro—2—indenone 18 ”\I’b The most reasonable approach to 18 was from the known bicyclic enone Ag. Enone AC has the desired carbon frame— work and most of the functionality, and requires only the 3» ---------- » 06A 40 ‘3 introduction of a second carbon—carbon double bond to give the desired 18. Excluding aldol type condensations, most methods of introducing unsaturation q,B to existing carbonyls involve introduction of a functional group X either a or B to the carbonyl and subsequent elimination of HX. If 23 2A Ag were functionalized B to the carbonyl (81) elimination 41 \ 42 could occur either to give the desired dienone 18 or to give the more thermodynamically stable dienone 81. On the other hand if 88 were functionalized d to the carbonyl (8%) elimination of HX should initially lead only to the desired dienone 18. Thus hoping to avoid Ag the initial strategy was to functionalize 88 a to the carbonyl. Attempts Using d—Functionalization The hexahydroinden-2-one (Ag) was prepared according 30 to procedures of Dauben and Raphael.31 Treatment of 25 2-carboethoxycyclohexanone (88) with potassium t—butoxide in Et 1) KOIBu/tBuOH + Z)Bnr~s: 5% «on 73% 4!) I46 refluxing t-butyl alcohol followed by addition of pro— pargyltmwmmdeEgave 88. Hydration of the triple bond was accomplished by treating 88 with a catalytic amount of Hg++ on Dowex—50 resin and a trace of H280“ in aqueous methanol. The method is very mild and gives 88 in nearly quantitative yield. Intramolecular aldol condensation using 5% KOH gave 88 in 56% overall yield. Bromination of ketones a to the carbonyl followed by dehydrobromination has been used successfully to prepare 32 The a bromination of ketones 39 a,B-unsaturated ketones. has been accomplished using two different strategies. 26 One method uses acid catalysis,33 and presumably involves attack of Br2 or Br+ on the enol form of the ketone. The b '—*"+ R ‘— \ J X. + Bf Bf acid—catalysed reaction is slow and it was feared that bromination of the double bond in 88 might be competitive under these conditions. Another method for preparing a-bromoketones involves the bromination of enolate anions.34’39 This reaction is very fast even at —78°C and it was hoped R Base Ir, R Br that bromination of the double bond in 88 would be much slower. Thus treatment of 88 with lithium diisopropylamide (LDA) in THF at —78°C, which is known to generate kinetic enolates,35 presumably gave 81. Treatment of this solution 32b with Br2 was expected to give 88. In addition to un— reacted 88, a dark orange liquid was isolated after column 0 I . IDA/THF 3 Br: 48°C 4O 47 48 chromatography. The HMR spectrum of this liquid showed the expected vinyl proton resonance at 65.80 and a doublet at 63.83 (J = 2.5 Hz) which might correspond to the proton attached to the carbon bearing the bromine atom. The low coupling constant suggests that this proton is probably trans to the bridgehead proton. Unfortunately, even if this were the correct structure, the product was only 60% pure (<30% yield by NMR) and decomposed rapidly on standing. AOb Others have also reported q-bromoketones to be quite unstable. Although this result was disappointing, it should be pointed out that this bromination method was 36 introduced originally for saturated or mono-aryl ketones, not for enones. Another popular method for generating N,B unsaturated ketones involves the thermal elimination of d—phenylselen- oxides.37 While d-phenyl- and d-methylsulfoxides38 28 undergo the same reaction, the phenylselenoxide elimination occurs at lower temperatures, by 100-150°C. In View of the high energy of the expected cyclopentadienones, the phenyl— selenoxide elimination was preferred. After treatment of 88 with LDA in THF at —78°C followed by addition of ¢SeBr, an established procedure for generating a-ketoselenides,37 only a small amount of crude material could be isolated which had the expected vinyl resonance at 65.70 and methine doublet 63.A0 (J = 2.0 Hz) for the methine proton geminal to the phenylselenide. The expected product could not be b_. IDA 48°" i ¢5e3r 4!) isolated and the reaction was not reproducible. There are many possible explanations for the failure of these reactions. One possibility is that the expected kinetic enolate (81) was not being generated. At least two other enolates could be formed (88 and 81). To test this possibility, 88 was treated with LDA and the enolate was quenched with trimethylsilyl chloride (TMSC). After work up, mass spectral analysis of the product showed that 29 a silyldienol ether had been formed (M/e M+ = 208), and in high yield. The 130MB showed only ten unique carbon atoms (the three silyl methyls are magnetically equivalent by rapid rotation) indicating that only one of the three 1 possible isomers (5g, 5; or 5%) was present. The HMR showed two vinyl protons of equal area at 55.58 and H.98, 30 thus eliminating isomer 5% which has only one vinyl proton. 0f the remaining two isomers (52 and 54) only 52 can 'Vb ’Vb ’Vb undergo a Diels—Alder cycloaddition reaction with dieno- philes. Therefore, 52 and 5% should be distinguishable on this basis, keeping in mind that a lack of reaction would be inconclusive. Treatment of a methylene chloride /’ ‘0 OSi: 4- Meanwhile ) N.R. 54 0 OS'/ 0' hile ’ “+2 I: + lenop ADDUCI 52 solution of 52/53 with a methylene chloride solution of U-phenyl—l,2,4—triazoline—3,5-dione (55)“1, a pOtent di- enophile, gave an immediate reaction as indicated by the rapid loss of the red color as 55 was slowly added. This is taken as an indication that the isomer in hand was in- deed the expected isomer 52. Thin layer chromatography (CHCl3/SG) of the crude product gave 5% in 50% yield. The expected Diels-Alder adduct (5g) is presumably hy- drolysed to ketone 5% by the silica gel during chroma- tography. 31 52 55 WORK-UP 57 Other data also suggest that 52 is the correct struc- ture for the silyldienol ether. When attempting to purify the silyldienol ether by preparative V.P.C. (10% FFAP on ChromG/l8OOC) the more thermodynamically stable isomer (5%) was obtained in essentially quantitative yield. The GC—MS spectrum was practically identical with that of 52 but the l HMR showed only one vinyl proton at 65.02 and a greater symmetry in the methylene region. The silyl dienol ether 5% 32 is readily accessible from 52 by a 1,5 prototropic shift. It is conceivable that 5% could have been formed from 5M, ’V‘b / _ - / Q . Si— 1.5 Shift 4 O . Si" \ A \ ('l/ 52 53 by a l,3—H shift. This is not a thermally allowed reaction l3-Shift D ./ 1.5 .— 2 ” '\ 54 53 in a concerted manner, however, and the migrating hydrogen is not especially acidic, making an ionic rearrangement mechanism unlikely under these conditions. Although not conclusive, the spectral data, the Diels— Alder reaction with 55, and the facile conversion to 5% ’VL 33 strongly suggest that the silyl dienol isomer 5% had been formed. Therefore, the desired kinetic enolate (3%) had been prepared with LDA as expected. This being the case there must be some other explanation for the failure of the bromination and phenylselenation reactions of $0. One unexplored possibility is that after product forma- tion is achieved rearrangement or decomposition occurs. It is known that the a-hydrogen geminal to the ¢Se group in d—ketophenylselenides is very acidic. Under the basic conditions of the reaction and work-up, including chroma— tography, a sizeable portion of £2 may be in equilibrium with its enol 5Q. Enol 5%, like 52, may undergo l,5-H shifts to give a variety of products (52, ég, 9%), among which 52 would be the most likely. The bromoenone 5% may have suffered a similar fate before decomposing to intractable tars. 3H . a 1.5 0" «‘g__i 6] Although these attempts to functionalize Q0 d- to the carbonyl had not met with the intended success, the a-carbon in 30 had been functionalized as the silyl enol double bond in 52. Perhaps a way could be found to utilize this functionality. In fact Jung and RathkeuZ have con- verted silylenol ethers to q,B-unsaturated ketones via B—hydride abstraction with trityl cation or DDQ followed O / 1OSJ" \ \ cat-functionality 52 by loss of Me 31+. To test this approach, 52 was treated 3 with triphenylmethyl tetrafluoborate under prescribed Sig SiE HP - ESie conditions. It did not give the expected dienone (%§) 3... O ./ ¢3c+ "4- 05r- ‘\ 52 \ 62 36 or its dimer (20), but rather the tritylated enone (ég) in about A0% yield. Rathke and Jung”2 found that when there were no B—hydrogens, nucleophilic attack of the enol electrons on the trityl cations became the predominant re- action, as in the case of fig reacting to give g5, but they .I 0 SI: \ ¢3c+arf €53 4~JI> 63 64 further state that when B—hydrogens are available "nucleo— philic attack is less favored than hydride abstraction, and oxidation is the dominant reaction". The reason that nucleophilic attack predominates in the case of 5% in spite of the presence of a B-allylic hydrogen may be that hydride 0 v © 37 abstraction would lead to the antiaromatic cyclopentadienyl cation g5. G x 66» If the energetically unfavorable intermediate £5 was an obstacle to the formation of %§’ then perhaps the stability of the cyclopentadienyl anion (éé) could be used in some way to facilitate the formation of lg. This leads to the following proposal. Cyclopentadienone I l z’N\\ -<(' in» li:i’>__flgfiflsz~ 67' 67 dimethyl hydrazone ($1) is a very stable distillable 38 derivative of cyclopentadienone which shows no tendency ‘63 o (,n=u-ocua)cu,so. + .n@ j. 67 a 68 to dimerize or react with dienophiles.“3 This is probably due in part to the resonance contributor QZ', as suggested by the molecule's 3.3D dipole moment. Hydrazone éz is prepared from fig and cyclopentadienyl sodium,Ll3 and can */Ok "250‘ > —' — 67' 11 be hydrolyzed to cyclopentadienone ll with H2SOLI.H3 Un— fortunately the requisite cyclopentadienyl anion ég, which would lead to lg, is not readily available nor could the necessary regiochemistry be predicted a priori. 39 as » «A ------- .... N36) 69 70 However, the thermodynamic stability of 10 might provide I l .3_ /N\ [g] > .3_ /N\ 71 70 a sufficient driving force to permit its formation by the oxidation of El in much the same way as 2b is prepared DDQ 72 9b 40 from the tri-sulfide 12.8 To test this possibility, the + 22.11, awn-«L 4O 71(Z+E) dimethylhydrazone 11 was prepared as a mixture of Z and an E isomers, according to the procedure of Fishel, in about 90% yield. While the conversion of 12 to 2b with cow‘- 7] DDQ required extended reflux at 130°C, 1% underwent an im- mediate reaction with DDQ at 25°C in benzene as indicated 41 by the formation of an intensely dark color upon addition of the first drop of the DDQ solution. The only isolable product from the tarry residue of the reaction was the DDQ-hydroquinone (1%). It is suspected that initial oxi- dation occurs by electron transfer from one of the nitro- gens, leading to polymerization instead of hydride ab- straction from the five membered ring as anticipated. At this point, after exploring the possibilities of a—functionalization without success, it was decided that perhaps the original strategy (a-functionalization) may have been a tactical error and that B-functionalization might be more successful. Use of B-Functionalization DePuyl45 has successfully used this strategy to pre- pare cyclopentadienone (ll) itself, although in this case there was no opportunity for elimination exocyclic to the cyclopentenone ring. Bromination of cyclopentenone with N—bromosuccinimide (NBS) gave A—bromocyclopentenone (1%) in good yield. Treatment of (Q with Et3N in ether gave cyclopentadienone dimer (15) in nearly quantitative yield. H2 NBS/AIBN CCh_reflux 73% r 71$ Et3N Ether rt DHAS-Adder L .. A similar reaction sequence should also lead to 1% from £0. Free radical bromination of £0 should give the requisite bromoenone 15 as a result of attack at the more stable tertiary allylic site as opposed to the secondary radical site which would give the unwanted bromoenone (xx). Although base promoted dehydrohalogenation of IQ could lead either to lg or to the more thermodynamically stable 52, the methylene protons a to the carbonyl (lg) are “3 "' to» or 40 Br. Br. 03-. Br 76 77 kinetically the most acidic and therefore the initially formed product could be the desired dienone 18. Treatment of £0 with NBS in refluxing carbon tetra- chloride gave the unstable bromoenone (zé), which decom- posed rapidly when neat but could be kept for a few hours in dilute solution. The yields were as high as 90%. The site of bromination is clear from the 1HMR spectrum which shows the a-methylene protons as two doublets at 63.25 (H1) and 2.78 (H2) with a geminal coupling constant of 19.3 Hz. The downfield proton is assigned as Hl based on its proximity to the bromine atom. AA .b_. NBS/60h .4» reflux/5mm 4!) The reaction is interesting because it is a self indicating reaction when a slight excess of enone ($0) is used. Initially the reaction mixture is water white. As the reaction mixture is brought to reflux, a light tan color develops and deepens as the refluxing continues. After about 5 minutes (under conditions reported in the experimental section) the solution turns almost instan- taneously from brown to water white. Work up at this point shows complete reaction. If the reaction mixture is refluxed much longer noticeable decomposition occurs. If an excess of NBS is used the brown color persists even after 20 minutes at reflux. Unlike the conversion of 1% to 1%, treatment of 18 with triethylamine gave back approximately 80% unreacted bromoenone. Being confident of the correct structure assignment of ZQ it was felt that a stronger base was needed to effect the elimination. Thus treatment of IQ A5 with potassium tertfbutoxide in THF at 0°C gave 2%, pre- sumably via 18, in 29% yield. It was later found that treating 16 with neat Et3N gave 20 in U9% yield. The best yield (60%)was achieved by simultaneous addition of THF solutions of potassium t—butoxide and 16 to a flask kept at —78°C. Attempts to trap 18 with N-phenylmaleimide, di- methyl acetylenedicarboxylate and cyclopentadiene failed. The spectral data were consistent with 2222 geometry for the dimer 20. Compound 20 gave the correct elemental analysis and showed the correct exact mass to .001 AMU. l The IR (KBr) showed a carbonyl stretch at 1765 cm— which is in the range of norbornenone carbonyls. The IR also showed the a,B-unsaturated ketone chromophore stretching l frequencies at 1690 cm- (C=O) and 1615 cm—1 (C=C). The l3CMR showed the requisite 18 carbons with the carbonyl resonances at 205.53 and 198.76 ppm (relative to TMS). 1 Despite the complexity of the HMR spectrum the one vinyl 2C) U6 and three methine protons are clearly discernible. The vinyl proton (65.99, H3) and one methine proton (62.95, HM) were both singlets. The remaining methine protons (H1, H2) appeared as two doublets (63.09, 2.90) with a coupling A6 constant Q = 9.7 Hz which is consistent with endo-gg. Section B. Preparation of cndo—pentacyclo- [8.5.1.0239.05’9.011’15]hexadeca—A,11(15)—diene-3,16- dione (21) via 5,6—dihydro—2(UH)-penta1enone (19) Having successfully prepared 20 from £0 it seemed reason- able to prepare 21 from 78 by a similar sequence. Unfor- tunately, 78 was unknown despite several attempts at its .-——-——»—» , 78 2] synthesis“7 and despite the successful synthesis of several of its simple derivatives 79,“8 80,u8 §%,U7a 82.u7a Attempts to prepare 78 by the standard aqueous 80 BI 82 U7 base—promoted intramolecular aldol condensation of 83 ’47 lead to a complex mixture of products. e on /u,o COMPLEX a MIXIIIRE 83 A synthesis of 78 was planned via 12 which was first reported by BECKGTM8 in 1978 and more recently by Trost.£49 Hydrolysis of the ester under mild conditions followed by decarboxylation under neutral conditions should lead to 18. Becker prepared 19 from the diketoester 88 which he prepared in 33% yield by alkylation of 85 with chloro- 50 acetone according to Herz's procedure. I 02:: l) ref/cine 33% . 2) mi “2“ 35 84 A8 1n the present work a two step procedure increased the yield of 8A to 81%. Thus treatment of 82 with potas- sium t—butoxide in refluxing tfbutyl alcohol followed by the addition of propargyl bromide gave the alkyne 88 in 81% yield. The structure of 88 was clear from its spectral data (see experimental). Hydration of the triple bond51 with Hg++—Dowex 50 resin in aqueous MeOH and a trace of H280” gave 88 quantitatively. Following Becker's procedure 88 was Et , 1)K0t3u/ (8005) ft 2)Br”“~i§; 86 85 H§+ Dowex Nail/Toluene # Et Reflux HQhINL 84- “9 treated with NaH in refluxing toluene to give 82. Although small quantities of Z8 could be separated by gas-liquid chromatography, 1% could not be readily purified in syn- thetically useful quantities. Therefore, crude 82, after bulb-to-bulb distillation, was used in subsequent reactions. Hydrolysis of 82 was achieved using a slight excess of 1% NaOH at 0-10°C for 1.5 hours. Extraction of the non- hydrolyzable impurities with chloroform, acidification with dilute hydrochloric acid and extraction with chloroform gave crude 88. Heating neat crude 81 (100°C, 15 min) was l) 1% MO!) 79 87 78 accompanied by vigorous gas evolution. After dilution with chloroform, extraction of the remaining acids with dilute sodium hydroxide gave 88 ( 90% pure by GC, 1HMR) in 38% yield from 88 as a mobile liquid with an exceptionally sweet aroma. Following the same procedure used for 88, 88 was brominated with NBS in refluxing CClu to give the unstable 50 bromoenone (88). Simultaneous addition of THF solutions of 88 and potassium tert-butoxide at -78°C gave 8%, pre— sumably via 88, in 50% yield from 88. In addition an isomer of 88 was isolated in up to 10% yield from the re- action mixture and was assigned structure 88 on the basis of its spectral data. :..':°" 189 :2: 78 The spectral data for 8% are consistent with the ex- pected endo geometry. The key datum is the coupling 2 “H1H2 is consistent with the endo geometry. A coupling constant EH H W 0 would be expected for the exo isomer.”6 Both 1 2 H3 and HA appear as singlets as would be anticipated for constant {HlH' The observed J for 8% 18 5.2 Hz which 51 13 either isomer. The CMR of 8% showed the requisite number of carbons with the carbonyl resonances at 207.A2 and 196.08 ppm (relative to TMS). The IR spectrum showed the nor- 1 bornenone carbonyl stretch at 1768 cm— and the enone car— bonyl and carbon—carbon double bond stretches at 1690 and 1622 cm.1 respectively. Dimers 88 and 8% are quite similar in terms of crystal- linity, spectral properties and even solubilities. The most striking difference between 88 and 8% is that whereas 88 melts with decomposition at l90—2°C, 8% melts with de- composition at l35-A0°C and crystal fracturing was ob- served as low as 95°C. This dramatic difference in de- composition points is ascribed to an increase in the strain energy in 81 relative to 88. This effect is quantified in the next section by studying the decarbonylation kinetics for 20 and 21. ’b’b ’Vb 52 Section C. Decarbonylation Activation Enthalpy as a Measure of the Relative Strain in endo-pentacyclo— [9.6.1.02’10.05’10.012’17]octadeca-A,l2(l7)- diene—3,16—dione (20) and endo-pentacyclo- [8.5.1.02’9.05’9.0l ’lSJhexadeca—A,ll(l5)-diene- 3,16—dione (88) Decarbonylation of cndg—pentacyclo[9.6.1.02’10.05’10.Ol2’l7]- octadeca—A,l2(17)—diene—3,16-dione (88) and endo-pentacyclo- 8.5.1.0239;05’9,oll’lsjhexadeca-A,ll(l5)—diene—3,l6-dione (881 It was noticed that the decomposition point for 88 (l35—lA0°C) was much lower than that of 88 (190—2°C) and C).52 88 (197—8° Actually 88 exhibits crystal fracturing 21(135-4o°,m) 20090-214“) 75 (197-8'Jec) at temperatures as low as 95°C. It was felt that perhaps this decrease in the decomposition point was a consequence of the increased strain in 88 due to shortening of the X' bridge (see structure 88, page 7). If this were true then the strain energy should be reflected in the activation enthalpies of decarbonylation (the normal mode of thermal decomposition of non—dissociating cyclopentadienone dimersll) of 88 and 88. More accurately, only that strain energy which is relieved in the transition state would be 53 reflected in the AH#. Thus a kinetic study of the decar- bonylation of gg and 3% was undertaken to determine their activation parameters. The decarbonylations of £8 and 3% were carried out in m0.2 M C6D6 solutions which were sealed in 5 mm NMR tubes. The NMR tubes were heated in a constant temperature bath (Figure H) for a desired length of time, then removed and immediately cooled to room temperature, thus stopping the reaction. The temperature of the bath was determined by the reflux temperature of the solvent placed in the bottom flask. The solvents used and the temperatures achieved are listed in Table TIL Since the sealed tubes were less Table III. Solvents and Temperatures Used in the Constant Temperature Bath for Decarbonylation Reactions. Reflux Solvent Temperature (i0.5°C) l,2-dichloroethane 82°C l-propanol 97°C toluene 109°C anisole 150.5°C 1,3,5-trimethylbenzene 162°C l,2,3-trimethylbenzene (Tech.) 173°C dense than the bath medium (ethylene glycol) a holder was 5M 0 0K ' countusm/ e I I q Ln} . :‘-~1 ( counzusm 5 E nsrunu i g I ' . I : E EIHYLENE : E : -——GLYCOL . . ; : IN BATH ; ' I I i : (.— VAPOR : i : ‘“ IRAP : g ' {_“é:i’/) lEFlllX SOLVENT Figure U. Two schematic views of the constant temperature bath for decarbonylation reactions. 55 fashioned from nichrome wire,which not only held the NMR tube completely submerged in the bath,but also allowed rapid removal and cooling of the tubes. The entire bath was insulated to ensure uniform heating and to limit the exposure of the samples to light. At the temperatures used for the decarbonylation of 2% (82, 97, 109°C) only one product (g0) was formed. This conclusion is based on the observation that no vinyl sz-109‘c 060. 21 9C) or aromatic proton resonances (lHMR) were observed other than those assignable to 21 or 20. Trienone 20 is mm ’b ’b the expected product and exhibits vinyl resonances at 66.01, 5.78 and 5.20 andsamethine signal at 62.82 (Figure 5). The l3CIVIR shows the requisite 15 carbon atoms. The IR spectrum shows the loss of the bridging carbonyl group (no stretch above 1750 cm_l) and retention of the 56 turd / 20 5H . Wet. AL... __ __ Ate/(J ! l 173°C . '5 I 29' 30 MIN. n HM» 9.2. M ;' _~_fiJKIIJL~J\.i“*_IIIIJMw\i - ._ iv, lit,M,»/ U\“fi Llninnlnnin111:4111111111;[‘4in...;411A1nl.1.1I4A4JJ11..I.+4 8.0 7.0 6.0 5.0 4.0 3.0 ppm(5) Figure 5. Partial 250 MHz lHMR spectra of £9, gi and their decarbonylation products. 57 conjugated enone (1690 and 1620 cm-l). Compound 90 also exhibits the correct parent ion (1V1+ = 212) in the mass spectrum. In the case of 20 the decarbonylations were carried out at 150.5 to 173°C. At those temperatures 20 gave a mixture of products which appeared to contain some 91, based on the similarity of some of the peaks of the mixture to those in the spectrum of 20 (see Figure 5). When the mixture was heated at 195°C or alternatively when crystalline 20 was heated just above its melting point (m200°C) the sole product formed was 92. It was noticed that the aromatic ’V’b protons of 92 could also be observed in the mixture of "0% decarbonylation products before heating to 195°C (Figure 5). 0:30 ISOMERS ‘92 58 Although a multi-step ionic mechanism for the conversion 91 + 92 might be written, enolization of 91 gives a system (91') with the proper orbital overlap to undergo a 6 elec- tron thermally allowed (l,5)-sigmatropic shift which aromatizes the central ring (92'). If the other two double bonds were involved a perhaps indistinguishable 10 electron (1,9)-sigmatropic shift could accomplish the same result. Enolization Enofizatiou l,5-Shift OH 19-Sff “ '3 )4 '* 4* '00 91' 92' '92 The decarbonylation of B,y-cyclopentenones is a well 59 11.52—5u known cheletropic reaction. Although the rearrange- 11 this appears to be ment of 91 to 92 has much precedent the first example of the formation of a spiro compound from the decarbonylation of a cyclopentadienone dimer.ll When 90 was heated to 195°C a complex mixture of products was formed which was not analyzed. In the kinetic experiments, the extent of reaction was determined by analyzing the 250 MHz lHMR spectrum. For 20 the extent of reaction was calculated from the areas of protons 1, 3 and A (53.09, 2.95 and 5.99, respectively) 2C) 21 relative to the residual protons in deuterated benzene. This was necessary because of the complex mixture of products formed. However, since 21 gave only one product upon decarbonylation at the temperatures used, the extent of reaction was calculated from the areas of protons l, 3 and U (21) (63.U5, 5.86 and 3.01, respectively), and three 60 of the product (90) protons (56.01, 5.20 and 2.82) which have not been uniquely assigned The decarbonylation of 22 and 21 followed first order kinetics. The extent of reaction (X) as a function of time and temperature is listed in Table IV. The data are plotted in Figures 6 and 7 as L AO/(AO-X) versus time, where A0 is the initial concentration (arbitrary units). A least squares linear regression analysis of the combined data of two runs at each temperature gave the first order rate constants k which are listed in Table V along with the cal— culated correlation coefficients. In order to obtain the most accurate results the activa- tion enthalpies of decarbonylation for 20 and 21 were cal— culated using a multiple data set KINFITLI55 computer program which gave a least squares best fit of the entire data set for each compound (Table IV) to the following equations provided by Professor J. L. Dye.56 The second equation _E(l_l ) k __ T R T T — T e ref ref ref and A 0 Ln _ — k t AO X is simply the first order reaction rate law. The first equation is simply derived by dividing the 61 Table IV. Extent of Decarbonylation (X) of 20 and 21. Extent of Reactiona’C 0 Temp. Time (i0.5°C) (sec) Run #1 Run #2 150.5 600 0.038 0.059 3600 0.205 0.209 7200 0.365 0.359 10800 0.A67 0.452 12600 0.u99 0.u86 luuoo b C 0.53M 0.523 A0 ’ =0.660 0.65M 20 162. 900 0.151 0.152 1800 0.278 0.281 2700 0.378 0.381 3600 0.u38 o.uu3 0600 b C 0.501 0.513 A0 ’ =0.628 0.639 20 173. 120 0.082 0.06M 360 0.190 0.193 720 0.321 0.321 1080 0.019 0.u35 luuo 0.500 0.511 1800 b 0.5u0 0.5u5 A0 ’°=0.697 0.705 51 82. 3600 0.u29 0.u23 4800 0.513 0.515 6000 0.59M 0.592 7200 0.6u3 0.651 8000 0.692 0.699 9600 0.73M 0.73M 10800 b C 0.767 0.768 A0 ’ =0.925 0.927 21 97. 300 0.039 0.036 600 0.080 0.076 900 0.120 0.101 1200 0.152 0.116 1500 0.163 0.1u7 1800 0.185 0.162 2100 b C 0.206 ----- A 3 =0.297 0.262 62 Table IV. Continued. Extent of Reaction a,c Temp. Time Comp. (i0.5°C) (sec) Run #1 Run #2 21 109.0 120 0.205 0.18“ ““ 240 0.369 0.359 360 0.473 0.u96 “80 0.576 0.597 600 0.655 0.669 720 0.709 0.723 _ 840 b C 0.761 0.761 A0 ’ =0.929 0.929 aAmount of starting material which has undergone decarbonyla- tion. bInitial concentration. CArbitrary units. dTube broke after removing from bath. 63 .OOmeH .mma .m.omH as aw no soaucaacocscccc are no areas .8 cssmfim 337...: . e9...... _ h . cow... . . I. . 2%... _ _ L F\Wo.e X \\\\\\\\\\M k‘fi .. \\ I... . ‘ .l o \ Iné \. s . m o .\ .a.. w M\ \ l- u-o‘ B \\\\\\ a .< . .. .... 9.53.}. ..s. \ 9.2K 1 .1¢.~ N ...... x ~ 22.0 I 6A .oomOH 9.8 .nm .mw pm flaw. .Ho COHpmazsophwomp 0.3 ..Ho mpoam .N. madman .33 as: 0000' COO» 0000 ......F..P. ‘ \ 9.3 \ . ... ....c I...— ] «2...... "2.2.0 65 Table V. First Order Decarbonylation Rate Constants for 39 and 31. Temp. -6 —1 Compound (i0.5°C) k (x10 sec ) Corr. Coef. 20 173.0 8AA 0.9969 ’b’b 20 162.0 3A2 0.9972 20 150.5 112 0.9983 g0 109.0a 1.2 21 109.0 2050 0.9983 2% 97.0 592 0.9927 21 82.0 163 0.9992 aExtrapolated value. absolute rate expression for any temperature T k T _AH¢ 135 k _ B e IN? . e R _ h by the expression for a reference temperature Tref with a known rate constant kref’ The entropy term and all constants kBTref e RTref , R ref = h e except R drop out leaving the first equation. The error limits (one standard deviation) for the enthalpies are 66 generated by KINFITA based on estimates of the error in the input data (time, :3 sec; extent of reaction, i2%) and the "degree of fit" of the data. The activation entropy was calculated manually from the expression: H5 ref kref ) + A kBTref A85 = R Ln ( where kB = Boltzmann's constant. The square of the stan— dard deviation of the entropy is given by: 2 2 ref 2 (5kref)2+(Tl ) (6AH#)2-+2 Ef—j%———(0krefAH#) ref ref ref 62(A85)= k g . . ¢ . where okrefAH is the covariance of kref and AH and like akref is also calculated by KINFITU. The activation parameters for 20 and 21 as well as comparative data from the literature are listed in Table VI. The activation 2 parameter AH shows that the tetramethylene bridge in 39 (AH5 = 35.5 Kcal/mole) introduces little if any measur- able strain into the dicyclopentadiene-l,8—dione (75) framework (AH5 = 39.5 Kcal/mole). The trimethylene bridge in 21 (AH5 = 23.5 Kcal/mole), however, introduces a minimum of 12 Kcal/mole of strain energy relative to 20. 67 Table VI. Decarbonylation Parameters for 7-Norbornenone Derivatives. AH5 (Kcal/mole) A85 Compound (e.u.) Ref. fig 35. i 6. i 21 23. i —9. i 15 3“. i 9. i 52 25. i 6. i 55c / 93 mm 31. i -3. i 540 / 911 mm 27. i H. i 53a 25. 57 68 Discussion of the Kinetic Results Three explanations have been offered previously to a rationalize observed decreases in AH for other 7-nor- bornenone decarbonylations. The first is "steric compres- 53a sion" at the norbornenone double bond as in 95. The 4 95 authors claim that the methylene bridge is compressed against the norbornenone double bond and that this tends to flatten the six carbon ring in the norbornenone skele- ton. This results in the increase in the overlap of the orbitals of the carbonyl bridge with the double bond. Thus the decreased AH5 (27.9 Kcal/mole) is directly attributed to this increased orbital overlap in the ground state which is a consequence of "steric compression". Inspection of models of 20 and 21 shows that there is 69 no increased compression at the corresponding double bond when changing from a six- to a five-membered fused ring. Therefore "steric compression" cannot be used to rationalize the AAH5 for 20 and 21. 'b’b mm The second rationalization has been used to explain the large AAH¢ (6 Kcal/mole) between 93 and 9%. The authors Ma c claim5 ’ that in 93 there is an increase in the orbital overlap of the carbonyl bridging orbitals and the saturated ethylene bridge. This is a result of the cyclopropane 93 94 "banana" bonds which overlap favorably in 93 but not in 9%. The observed product in the decarbonyl of 93 and 93 70 93 96 94 is cycloheptatriene 96. 97' ‘98 The second rationalization has also been used to ex- g plain the low AH for 9753C (28.5 Kcal/mole). In this case the edge of the cyclobutene ring participates in the 71 decarbonylation to give cyclooctatetraene 99. This ex- planation has also been used to rationalize the fact that LYNNE REFlIlX C. (1 Cl whereas 99 decarbonylates smoothly in refluxing xylene, 999 remains unchanged.5ue flu“ ,. II. R. / I mm m IOO 72 Since both 99 and 9% have the same endo-cyc10pentenone ring at the saturated ethylene bridge it is unlikely that 9% has significantly more "edge" overlap than 99. There- fore this second rationalization also does not explain the large AAH5 between 20 and 21. ’b’b 111% The third explanation5 b is that the stability of the decarbonylation product provides the driving force for the reaction as in the case of 101. Although there is mmm . ION some evidence57 that %9% may be a fleeting transient, it has never been isolated. Since the initial decarbonyla— tion products from 99 and 9% do not possess any special stability, this third rationalization must also be aban- doned. Actually this third case may be an extreme example of the second case where instead of a cyclopropyl "banana" bond or the cyclobutyl "edge" a second n bond overlaps with the bridging carbonyl orbitals. 73 Since none of the previously proposed rationalizations for decarbonylation rate enhancement can explain the large AAH5 there must be some other explanation. Furthermore, since the results of this study are in accord with the predictions proposed earlier in this thesis, it is reason— 9 able to assume that the large AAH may be due to the relief of ring strain in the decarbonylation transition state of 99. The strain energy in 99 relative to 99 may be more than 12 Kcal/mole since only the ring strain which is relieved in the transition state is manifested in the AH5. Finally these results show that monomeric g—face annulated cyclopentadienones may be stabilized relative to their [U+2] dimers by increasing the dimer strain energy if the Q—face bridging group is short enough. It also sug- gests that this same conclusion may be true if the bridg— ing group is constrained to be coplanar with the cyclo— pentadienone ring, as in 99 and 99. In the last section of this discussion the single crystal X—ray structures of 99 and 99 are examined to determine what structural changes occur in the dicyclo— pentadienone framework as a result of the strain in 21. mm 7“ Section D. Single Crystal X—Ray Structures of 2,10 05,10 012’171— endo—pentacyclo[9.6.l.0 octadeca-u,l2(l7)-diene-3,l8-dione (20) and 2 9 5 9 u 11 ““ endo-pentacyclo[8.5.l.0 ’ .0 ’ .0 ’ Jhexadeca— U,l2(15)-diene—3,l6—dione (99) Crystal Data Single crystals of 99 and 99 were grown from methylene chloride—methanol solutions by slow evaporation. The X— ray structures were determined by Dr. D. L. Ward whose efforts are gratefully acknowledged. Crystals of 99, C18Hl8o2 are orthorhomblc; space group P2 2 2 a = 9 170(3), 0 = 19.622(u), c = 10.273(u)fi; z = 1 l l3 u; M = 266.34; pC = 1.28M g cm'3. Crystals of 99, C H o 16 16 2’ are monoclinic, space group P2l/n3 a = 13.220(H), b = 10.600 (5), c = 8.876fi, o = 99.54<3>°; z A; M = 240.30; pC = 1.301. Lattice dimensions were determined using a Picker F FACS—I diffractometer and MOKQ (1 = 0.709268) radiation. 1 The intensity data were measured using MOKQ radiation (20 = 50° (99); 20 = 550 (9%)) yielding lul7 (99) max max and 2380 (99) total unique data, and based on I > 20(1), 1067 (99) and 12U0 (99) observed data. The data were re- duced;58 the structures were solved by direct methods;59 and the refinement was by full matrix least squares tech- niques.60 The final R values were 0.0U3 (99) and 0.065 (99). The final difference Fourier maps showed densities ranging from +.26 to -.21 (20) and from +.A6 to —.A6 (2 ) mm m 75 with no indication of misplaced or missing atoms. Positional parameters, bond lengths and bond angles for 20 and 21 are listed in Appendix 1. Discussion of X—ray Structures The structures of 20 and 21 are shown in Figure 8 with the bond lengths indicated. The bond angles for 20 and 21 are shown in Figure 9 and ORTEP stereoscopic views of 20 and 21 are shown in Figure 10. Structure 21 was solved and refined without difficulty. However, 20 exhibited two anomalies one of which is easily rationalized. The first anomaly was that the six—membered ring containing C(lu) and C(15) was flat and the C(1M)— C(15) bond length was ml.3u3 indicating that it was a double bond (Figure 11). This was chemically unreasonable. Inspection of a model of 20 showed that there were two possible orientations for this four carbon bridge as shown in 20% and 20% (Figure 11). This difficulty was overcome by assigning half atoms to C(lU) and C(15) (as well as their respective hydrogens) and allowing these u half carbons to occupy the positions indicated in 20% and 20%. Since this is a crystal phenomenonenuiégg and 20% are certainly interconverting rapidly in solution and since this portion of the molecule is not expected to contribute significantly to the relative strain in 20 and 21, the aThe hydrogen atoms are omitted for clarity. bBond lengths associated with C(lu) and C(15) have been omitted due to disorder at these positions. . a b Figure 8. Bond Lengths for 20 and 21. 77 aThe hydrogen atoms are omitted for clarity. bBond angles associated with C(lu) and C(15) have been omitted due to disorder at these positions. Figure 9. Bond anglesa for 20b and 21. 78 Figure 10. ORTEP stereoscOpic views of 20 and 21. mm mm 79 20A 208 Figure 11. Disordered orientations of C(lA) and (15) in crystals of 20. observed disorder should not affect the interpretation of the important structural changes between 20 and 21. The second anomaly in 20 is the unexpectedly short 0(7)—0(8) bond (1.50 X instead of ml.5HA) in the other four- carbon bridge (Figure 8). At present there is no explana- tion for this short bond. It is a curious coincidence that the bonds which are involved in these anomalies are es- sentially the bonds which have been "removed" in 21. Although there are several differences in bond lengths between 20 and 21 the most striking change in bond lengths involves the carbon-carbon single bonds which are formed during the Diels-Alder dimerization. The single bonds in 21 (C(l)-C(2) and C(9)-C(10)) are longer than the 80 corresponding bonds in g9 by 0.019 and 0.0113, respectivelY- Since the standard deviation for these bonds are 0.006 and 0.005A, respectively (for both 20 and 21), the differences in bond lengths,although small, are probably significant. In other strained cyclopentadienones which exist as dis- sociating dimers the corresponding bonds are thought to be elongated also.ll Since 21 decarbonylates much more readily than 20 the carbonyl bridging bonds in 21 might be expected to be longer (i.e., weaker) than in 20. This is not the case. In fact, the C(1)—C(l6) (21) bond is essentially the same as the corresponding bond in 20 and the C(10)-C(16) (21) bond is even shorter than the corresponding bond in 20. Surprisingly, the carbonyl bridge is more symmetrically disposed in 21 than in 20. There are several variations in bond and torsion angles between 20 and 21. Essentially all of the significant bond angle variations are centered at the termini of the X'(11) bridging group (C(5) and C(9) in 21). Of course, the bond angles at the termini of the X bridge change in going from 20 to 21 but are unexceptional. 81 One of the explanations for decarbonylation rate ac- celeration discussed in Section C was that there was an increased overlap of the carbonyl bridging orbitals with the pi orbital of the "norbornenone" double bond by de- creasing the angle of the bridge relative to the pi or- bital. As can be seen in Figure 9 the C(16)-C(l)-C(15) and C(16)—C(10)-C(11) angles (21) are within 10 of the corresponding angles in 20. This confirms the conclusion drawn earlier from inspection of models that this type of increased overlap cannot be used to explain the decrease in AH? for 21 relative to 20. At the sp2 hybridized carbon 0(5) two angular distor- tions occur. The C(A)-C(5)-C(6) angle in 21 is expanded to 136.7 i 0.50 from 128.5 i 0.50 for the corresponding 82 angle in 22 which is already distorted from the ideal 120°. The orientation of the C(5)—C(6) bond not only extends this bond angle but also results in a twisting of the C(A)-C(5) double bond. The ORTEP stereoscopic views (Fig- ure 10) of 22 and 21 have been drawn looking down the C(U)— (5) axis to show the increase in the C(3)-C(A)-C(5)-C(6) torsion angle in going from 22 (3.30) to 21 (1A.M°). This results in decreased pi overlap and an increase in the total energy of the molecule. At the other end of the X' bridge three of the six bond angles change significantly in going from 22 to 21. The angles which are enclosed in the norbornenone and cyclo— pentenone rings as well as the C(8)-C(9)-C(10) angle (21) do not change very much. However, the C(2)—C(9)—C(8) (21) angle increases from 115.9° in 22 to l2l.6° in 21 and the C(5)—C(9)-C(10) angle (21) decreases from the normal tetrahedral angle of 109.110 in 22 to 101.2° in 21. The last bond angle change in going from 22 to 21 is a decrease toward the normal tetrahedral angle of m109°. The C(5)-C(9)—C(10) angle (21) decreases from 11A.6° in 22 to 111.5° in 21. In other words this end of the cyclo- pentenone ring is 3° more gngg in 21 than in 22. This was unexpected since it was initially felt that the shortened bridge would pull the cyclopentenone ring up (gig). How- ever, this observation, when coupled with the elongation of the sigma bonds formed in the dimerization (vide supra), 83 can be interpreted as suggesting that 21 is closer to its transition state for dimerization (in terms of energy or distance along the "reaction coordinate") than 2Q is to its transition state. Any conclusions drawn from the single crystal X-ray structures must be tempered with the fact that the solid state structure and solution state structure may differ significantly due to crystal lattice forces and solvation factors. Nevertheless, the idea that short bridges in 9— face annulated cyclopentadienones results in increased dis- tortions of optimum bond lengths and bond angles in their Diels—Alder dimers seems to be justified based on the X—ray data presented here. EXPERIMENTAL General All melting points were recorded on a Thomas—Hoover melting—point apparatus and are uncorrected. Infrared spectra were recorded either as neat films or as KBr pellets on a Perkin—Elmer 167 Grating Spectrophotometer. lHMR spectra were recorded on a Varian T—60, Bruker 180 or Bruker 250 spectrometer and 13C NMR spectra were re— corded on a Varian OFT—20 or Bruker 250 spectrometer. All Spectra were taken at ambient temperatures and are re— corded in delta (6) values relative to TMS. The coupling constants (i) are recorded in Hertz (Hz). All mass spectra were obtained on either a Hitachi EMU-6 or Finnigan—AOOO mass spectrometer. Ultraviolet spectra were recorded on a Cary-l7 spectrophotometer. Elemental analyses were per— formed by Spang Microanalytical Laboratories, Eagle Harbor, Michigan. Exact mass determinations were made at the re- gional NIH facility, Michigan State University. Toluene, tetrahydrofuran, ether, diisopropylamine and hexamethyl— disilizane were dried and distilled in glass prior to use. All other commercially available reagents were used as re— ceived unless otherwise specified. 8A 85 Ethyl l—(2—propyny1)—2-oxocyclohexanecarboxylate (22) 30 2% was prepared according to Dauben's procedure. TO a 2-L flask fitted with a magnetic stirrer, ZSO-mL addition funnel, N2 purge and a reflux condenser was added Egrt-butanol (1110 mL) and potassium metal (30.5 g, 0.78 mole). The tertebutanol was refluxed until all of the potassium had reacted (mu h). To this solution was slowly added neat 2—carboethoxycyclohexanone (126 g, 0.7“ mole) and refluxed for an additional 30 minutes. To this was slowly added propargyl bromide (97.1 g, 0.99 mole). After refluxing for an additional 30 minutes the mixture was dumped onto ice (m500 g). After separating the organic layer the aqueous layer was washed with ether (A x 250 mL). The organic layers were combined, dried (MgSOu) and filtered. The solvents were removed under reduced pressure and the resulting oil was vacuum distilled giving pure 2% (118 g, .5u mole) in 77% yield (lit.30 8u%). B.p. 108°C at 0.65 torr (lit.3° 13u-1370C at 9 torr). 1HMR (60 MHz, CDC13, TMS) 6A.13(2H,q,J = 7 Hz), 2.8-1.5 (11H,br m), 1.2u(3H,t, 1 3280 (m), 2950(m>, 2860, 119(62>, 107(32), 93(28>, 91(75>, 86(22), 82(68), 79(38), 77(51), 22(89), 21(28), 39(27). Exact Mass. Calcd. for C17H17N3O3: 311.12698. Found: 311.12520. 3,3a,2,5,6,7—Hexahydro—3—(1,1,1-triphenylmethyl)-2—indenone 2,2,2 To a magnetically stirred 50—mL round-bottomed flask purged with N2 was added triphenylmethyl tetrafluoborate 93 (1.59 g, 2.80 mmoles), collidine (0.59 g, 2.80 mmoles) and methylene chloride (10 mL, distilled from CaH2). When everything had dissolved a solution of 2—trimethylsily1oxy— 2,5,6,7—tetrahydro-[3aHJ—indene (52) (0.5 g, 2.20 mmoles) in methylene chloride (10 mL) was slowly added (5 min) at 25°C and allowed to stir for 2 h. The reaction mixture was dumped into H20 (50 mL). The organic layer was sep— arated and washed successively with H20 (50 mL), 10% HCl (50 mL), saturated NaHCO3 (50 mL), H20 (50 mL), dried (MgSOu), filtered and the solvent removed under reduced pressure to give a dark yellow solid (1.622 g). TLC (SG/ CH2C12) showed two spots. Trituration with ether gave a White solid (0.351 g). Preparative TLC (SG/CHCl3) of 75 mg of this solid gave 72 mg (96%) of (62) as a white solid with m.p. 218°C (39% total yield). lHMR (60 MHZ, 00013, TMS) 07.07(15H,m), 5.23(1H,s), 3.80(1H,s) 2.8 to 0.8(9H,m). 130MB (20 MHZ, 00013, TMS) 5205.73, 180.17, 127.81, 125.79, 125.31, 123.89, 55.29, 25.82, 32.20, 28.90, 25.25, 23.53. IR (KBr) cm“1 3110(w), 3090(w), 3060(m), 3020(m), 3020(w), 3000(w), 2965(m), 2935(m), 2885(w), 2860(w), 2823(m), 1690(s), 1630(s), 1595(W), 1295(m), 1250(m), 1220(m), 1358(w), 1322(W), 1330(W), 1315(W), 1305(W), 1292(W), 1270(W), 1228(w), 1195(W), 1185(m), 1160(w), 1080(w), 1077(w), 1020(m), 1000(w), 960(w), 950(m), 930(W), 910(w). 885(w), 868(m), 838(m), 792(m), 770(m), 720(s), 720(m), 710(s), 695(m), 662(W), 92 635(m), 615(m). M.s. m/e (rel. int.) 378(3,M+), 222(21) 223(100), 165(23). Exact Mass. Calcd. for C28H26O: 378.19836. Found: 378.19678. 3,3a,2,5,6,7-Hexahydro-2—indenone dimethylhydrazone (71) To a magnetically stirred 25—mL flask fitted with a reflux condenser and N2 purge was added 3,3a,2,5,6,7- hexahydroinden—2-one (20) (1.00 g, 7.35 mmoles), EtOH (7.5 mL) and unsymeN,N—dimethylhydrazine (7.5 g, 125 mmoles). The solution was refluxed for 22 h, cooled to 25°C, and the solvent and excess N,N-dimethylhydrazine were removed under reduced pressure to give a mixture of the g and E isomers (30/70) of the expected dimethylhydrazone (7%) as a dark red oil (1.2 g, 92% crude yield) containing less than 2% starting material (by V.P.C.). A sample of the pure hydrazones was obtained by preparative scale V.P.C. (5% SE—30 on Chrom W, 1/2 in. x 6 ft.) as a colorless liquid which rapidly turned dark red upon exposure to air. lHMR (60 MHz, 00013) 66.23, 5.80(1H, singlets, Vinyl pro- tons from Z and E isomers); 3.2 - 0.8 (17H, br m, including 3: TMS) 6172.67, 172.91, 168.72, 166.25, 123.69, 118.12, 27.91, a sharp singlet at 2.2, N(CH3)2). l3CMR (20 MHZ, 0001 27.22, 27.13, 23.77, 22-36: 37.19, 35-28, 32.69, 30.33: 29.99, 27.20, 25.59. IR (thin film) om"l 2980(m), 2930(8), 95 2855(8), 2815(m), 2770(m), 1625(5), 1270(m), 1228(m), 1355(w), 1255(w), 1200(w), 1162(w), 1025(m), 976(s), 952(m), 920(w), 860(m), 828(w), 820(w), 767(w), 712(w). M.S. m/e (rel. int.) 179(12), 178(100,M+), 177(16), 163(11), 136(13), 132(12), 132(18), 118(10), 107(23), 106(15), 105(12), 92(17), 93(31), 92(20), 91(38), 79(22), 78(11), 77(22), 67(12), 65(17), 53(12), 53(10), 26(16), 25(25), 22(22), 23(89), 22(26), 21(22), 39(12). Exact Mass. Calcd for C11H18N2: 178.12700. Found: 178.12702. Reaction of 3,3a,2,5,6,7-hexahydro-2—indenone dimethyl hydrazone (71) with dichlorodicyanoquinone (DDQ) To a magnetically stirred solution of the hydrazone (71) (0.320 g, 1.91 mmoles) dissolved in benzene (10 mL) in a 50—mL flask protected with a N2 purge was added drop- wise (10 min) a solution of DDQ (0.278 g, 2.11 mmole) in benzene (10 mL). The initially light tan solution of the hydrazone (71) darkened immediately upon addition of the first drop of the 000 solution and had become opaque at the end of the addition. After stirring the mixture for an additional 50 minutes a ball of black solid material had encased the teflon coated magnetic stirring bar. The dark benzene solution was decanted off and the benzene was re— moved under reduced pressure leaving a small amount of un- identifiable black tar—like material. The black ball of 96 material was dissolved in methanol, the magnet was re— moved and the methanol was evaporated leaving a dark solid. The solid was practically insoluble in carbontetrachloride, ether, and chloroform. When it was dissolved in methylene chloride only a small amount of black residue remained which was removed by decantation. Evaporation of the methylene chloride gave a fluffy tan solid. The spectral data for this solid was consistent with the hydroquinone of DDQ but gave no indication of the presence of 70. The lHMR (0001 TMS) showed only one peak at 82.76. The IR 1. 3) (KBr) showed a broad phenolic absorption at 3200 cm— The mass spectrum showed the parent peaks at m/e 228, 230 and 232 in the expected ratio for two chlorine atoms. 7a—Bromo—l,2,5,6,7,7a—hexahydro—2-indenone_(76) To a 50-mL pear-shaped flask fitted with heating mantle, magnetic stirrer, reflux condenser and N2 purge was added 1,2,5,6,7,7a—hexahydro-2—indenone (20) (1.00 g, 7.35 mmole), N-bromosuccinimide (1.20 g, 7.32 mmole) and carbon tetrachloride (30 mL). As the reaction mixture was refluxed it became light brown until after about 2 minutes when the mixture suddenly became water white. The heating mantle was replaced with an ice bath and the re- action mixture was cooled to 5 to 10°C, filtered and the solvent removed under reduced pressure to give a light tan liquid which would darken rapidly if left undiluted 97 and thus was not stable enough to permit either elemental analysis or exact mass determination. Its NMR (1H and 13C) and IR spectra, however, were consistent with W90% pure 7a-bromo—l,2,5,6,7,7a—hexahydro—2—indenone which was used without further purification. lHMR (180 MHz, 00013) 5.912 (lH,d,g = 1.5 Hz), 3.226(1H,d, g = 19.3 Hz), 2.776 (1H,d, g = 19.3 Hz), 2.77—1 0 (8H,m); l3CMR (20 MHz, 00013, TMS) 5203.00, 180.26, 127.39, 66.17, 55.11, 23.26, 27.29, 26.52, 22.28; IR (neat) om'l. 3080(w), 2920(s), 2880(m), 1708(vs), 1610(s), 1250(m), 1238(m), 1208(m), 1350(w), 1322(m), 1278(w), 1256(w), 1222(w), 1220(m), 1180(w), 1120(w), 1070(m), 958(m), 938(m), 860(m), 835(w), 820(m), 760(w), 680(m). 10.05,10. cndo-Pentacyc1o[9.6.1.02’ 012’171octadeca—2,l2(l7)— diene—3,18—dione (20) To 50—mL of THF cooled to —78°C in a 250—mL flask with magnetic stirring and under a N2 purge was added slowly (15 min) and concomitantly THF solutions of KOtBu (0.71 g, 7.39 mmoles in 50 mL of THF) and crude 7a-bromo—1,2,5,6,7,7a- hexahydro-2—indenone (prepared from NBS and 7.35 mmoles of 1,2,5,6,7,7a-hexahydro—2-indenone in 50 mL of THF) as described above. The dark red reaction mixture was stirred for an additional 30 minutes, and allowed to warm to 0°C, when water (7 mL) was added and most of the THF was removed under reduced pressure. The residue was 98 taken up in chloroform (100 mL) and water (50 mL). The aqueous layer was separated and washed with chloroform (1 x 50 mL). The combined organic layers were dried (MgSOu), filtered and the solvent was removed under re— duced pressure to give a dark red residue which when tri- turated with ether gave light tan crystals of endo—penta— 2’10.05’10.012’17]octadeca—2,12(17)-diene—3,l8- cyclo[9.6.l.0 dione (0.587 g, 60% yield). Alternatively, flash chroma- tography61 (5% EtOAc/CH013) effectively isolates this product from the reaction mixture. Recrystallization from MeOH— CH2C12 gave white crystals, m.p. 190—200 (dec). lHMR (180 MHz, 00013, TMS) 65.99(1H,s), 3.09(1H, d, g = 2.7 Hz), 2.95(1H,s), 2.77(1H,m), 2.20(1H,d, g = 2.7 Hz), 1.39—2.23 (15H,m). 130MB (20 MHz, 00013, TMS) 5205.53, 198.76, 181.52, 135.61, 132.89, 132.16, 56.26, 53.25, 52.11, 51.73, 32.28, 30.00, 27.16, 25.07, 22.72, 22.86, 22.70, 22.65(sh); IR (KBr) cm’1 3000(m), 2922(s), 2962(s), 2820(m), 1765(vs), 1690(vs), 1615(s), 1250(m), 1230(m), 1355(m), 1323(w), 1315(w), 1288(m), 1272(m), 1262(w), 1227(m), 1212(m), 1196(m), 1190(w), 1169(m), 1129(w), 1090(w), 1020(w), 976(w), 922(w), 922(w), 899(m), 869(8), 855(w), 822(W), 830(W), 817(w), 807(w), 777(w), 685(w), 670(w), 652(W), 638(W), 612(w); M.S. m/e (rel. int.) 268(3,M+), 221(21), 220(100), 212(21), 211(52), 198(19), 197(18), 183(21), 121(27), 132(28), 129(17), 128(21), 115(20), 91(21), 77(17), 22(28); UV (0H30N) Amax (s) 231 nm (10,650), 320(22), 332(20), 328(19). 99 Anal. Calcd. for Cl8H2002: C, 80.56; H, 7.51. Found: C, 80.22; H, 7.29. Exact Mass. Calcd. for C18H 0 268.12630. 20 2’ Found: 268.12621. Reaction of 7a—bromo-1,2,5,6,7,7a—hexahydro—2—indenone (76) with triethylamine in ether Fresh crude 76 (36.8 mmole), prepared as described above, was dissolved in ether (60 mL) and transferred to a 100—mL addition funnel which was attached to a 250—mL round—bottomed flask containing triethylamine (2.10 g, 20.6 mmole) and ether (20 mL) cooled to —20°C. The bromo— enone (76) solution was slowly added (m10 min) to the mag- netically stirred amine solution. While the mixture was stirred an additional 0.5 h at —20°C a white precipitate began to form. The mixture was allowed to warm to 25°C (0.5 h) and was dumped into 5% NaHCO3 (50 mL) and ether (100 mL). The ether layer was separated and the aqueous layer was extracted with ether (1 x 100 mL). The combined ether layers were washed with 5% hydrochloric acid (1 x 25 mL), saturated sodium chloride (1 x 25 mL), and water (1 X 25 mL), dried (MESOH) and filtered. The ether was removed under reduced pressure to give a light brown oil (8.3 g) whose IR and lHMR spectra were nearly identical with the crude 76 (>80% recovery based on lHMR). 100 Reaction of 7a—bromo—1,2,5,6,7,7a—hexahydro—2—indenone (76) with neat triethylamine Fresh crude 76 (7.35 mmole), prepared as described above, was treated with triethylamine (5 mL, 36.0 mmole) in the absence of solvent and cooling. After the mixture was stirred magnetically for l h the tan slush was diluted with water (30 mL) and the product was extracted with chloroform (l x 100 mL plus 1 x 50 mL). The combined organic layers were washed with 5% hydrochloric acid (1 x 25 mL - washing was acidic to litmus), saturated sodium bicarbonate (1 x 25 mL) and water (2 x 25 mL), dried (MgSOu) and filtered. The chloroform was removed under reduced pressure to give a nearly completely crystal- line mass which when triturated with ether and filtered . gave 20 (0.283 g, 29%). The melting point and spectral data were identical with those of 20 prepared with potas- sium tert—butoxide in THF at —78°C (vide supra). Reaction of 7a-bromo—l,2,5,6,7,7a—hexahydro—2—indenone (76) with potassium tert—butoxide at 0°C Fresh crude 76 (32.7 mmoles), prepared as described above), was dissolved in dry THF and placed in a 50-mL addition funnel attached to a 250—mL round—bottomed flask containing THF (100 mL) and fitted with a N2 purge, mag- netic stirrer and a second 50—mL addition funnel. To 101 the second addition funnel was added potassium tert— butoxide (3.33 g, 32.7 mmole) dissolved in THF (50 mL). The flask was cooled to 0°C and the two reactants were added simultaneously over a period of about 5 min and stirred for an additional 0.5 h. The mixture was dumped into a mixture of ether (200 mL), pentane (100 mL) and water (200 mL) and shaken in a separatory funnel. A large "rag" layer was formed which was broken by filtra- tion and the layers were separated. The aqueous layer was washed with ether (1 x 200 mL) and the combined or- ganic layers were washed with 5% hydrochloric acid (1 x 25 mL, washing was acidic to litmus), saturated sodium bicarbonate (1 x 25 mL) and water (3 x 25 mL), dried (MgSOu) and filtered. The solvents were removed under reduced pressure to give a dark brown, partially crystal— line residue. Crystallization from methanol/pentane and flash chromatographic separation (SC/5% EtOAc/CHClB) of the mother liquor gave crystalline 20 (1.32 g, 29%). The melting point and spectral data were identical with those of 20 prepared with potassium tert-butoxide in THF at —78°C (vide supra). 102 Reaction of 7a—bromo—1,2,5,6,7,7a-hexahydro—2—indenone (16) with lithium hexamethyldisilazide in ether/hexane (1:1) Fresh crude Z6 (7.35 mmole), prepared as described above, was dissolved in a mixture (1:1) of ether and hexane (100 mL) and put in a 250—mL round-bottomed flask fitted with a N2 purge, magnetic stirrer and addition funnel, and cooled to —78°C. To this solution was slowly added lithium hexamethyldisilazide (prepared in the ad- dition funnel from hexamethyldisilazane (1.18 g, 7.35 mmole)) and 2.1 H n—BuLi (3.5 mL, 7.35 mmole) which were magnetically stirred for 0.5 h and diluted with hexane to 50 mL) over a 10 min period. The contents of the flask were allowed to warm to 25°C and the solvent was removed under reduced pressure. The residue was taken up in chloro- form (200 mL) and water (50 mL). The organic layer was separated, washed with water (2 x 50 mL), 5% hydrochloric acid (1 x 25 mL), saturated sodium bicarbonate (1 x 25 mL) and water (1 x 25 mL), dried (MgSOu), and filtered. The solvent was removed under reduced pressure to give a dark brown oil. Flash chromatography (SG/3% EtOAc/CHC13) gave 88 (0.272 g, 28% yield) the melting point and spectral data of which were identical to those of £6 pre- pared with potassium tert—butoxide in THF at —78°C (vide supra). 103 Reaction of 7a-bromo—l,2,5,6,7,7a-hexahydro-2-indenone(16) with lithium hexamethyldisilazide in THF Fresh crude 16, prepared as described above, was treat- ed with lithium hexamethyldisilazide under the same condi- tions as described in the immediately preceding reaction except that THF was used as the solvent and only 50 mL was placed in the reaction flask. Flash chromatography (SC/2% EtOAc/CHCl3) of the crude product gave 26 (0.217 g, 22% yield), the melting point and spectral data of which were identical to those of 66 prepared with potassium tert- butoxide in THF at —78°C (vide supra). Attempted trapping oflfi with N—phenylmaleimide Fresh crude Z6 (3.0 mmoles), prepared as described above, was dissolved in ether (15 mL) and placed in a 25- mL addition funnel which was fitted to a lOO—mL round—bot- tomed flask. To this flask, also fitted with a magnetic stirrer and a N2 purge, was added ether (75 mL), N—phenyl— maleimide (1.02 g, 6.0 mmoles) and triethylamine (0.61 g, 6.0 mmoles). The stirred contents were cooled to -20°C and the solution of 16 was added over a 5 min period, stirred for an additional 15 min, and dumped into a mixture of sat- urated sodium bicarbonate (50 mL) and ether (100 mL). The mixture was shaken and the ether layer was separated, washed with water (1 x 25 mL), 5% hydrochloric acid 102 (1 x 25 mL) and water (1 x 25 mL), dried (MgSOu) and fil- tered. The ether was removed under reduced pressure. Although flash chromatography (SG/30% EtOAc, petroleum ether 30-60) of the residue gave back 50% of 16, and an unquantified amount of N—phenylmaleimide the lHMR spectra of all fractions showed no indication of an adduct between 66 and N—phenylmaleimide. Attempted trgppipg of %6 with dimethylacetylene dicar- boxylate (DMAD) To a 250—mL round-bottomed flask fitted with a N2 purge, magnetic stirrer and two 50-mL addition funnels was added DMAD (1.05 g, 7.35 mmoles) and THF (50 mL). To one funnel was added potassium pgpp-butoxide (0.706 g, 7.35 mmole) dissolved THF (50 mL) and to the other was added fresh crude Z6 (7.35 mmole), prepared as described above, dissolved in THF (50 mL). The contents of the funnels were added slowly (15 min) and concomitantly to the stirred solu— tion of DMAD cooled to -78°C. The mixture was allowed to warm to 25°C and stirred for 12 h. The mixture was par- titioned between methylene chloride (100 mL) and water (100 mL). The aqueous layer was washed with methylene chloride (1 x 25 mL) and the combined organic layers were washed with water (2 x 25 mL), 5% hydrochloric acid (1 x 15 mL), water (1 x 25 mL), saturated sodium bicarbonate (l x 25 mL) and water (1 x 25 mL), dried (MgSOu) and 105 filtered. Flash chromatography of the products gave g6 in 62% yield but no indication (1HMR) of an adduct of $6 and DMAD. Attempted trapping of $6 with cyclopentadiene To a 50—mL flask fitted with a magnetic stirrer was added triethylamine (10 g, 100 mmole) and cyclopentadiene (7.1 g, 108 mmole). To this solution was slowly added (0.5 h) a solution of fresh crude 16 (7.35 mmole), pre— pared as described above. The resulting mixture was stirred at 25°C for 60 h, washed with water (3 x 25 mL), dried (MgSOu) and filtered. The ether and cyclopentadiene were removed under reduced pressure. Flash chromatography of the residue gave 66 (21%) and a small amount of dicyclo— pentadiene but no indication (lHMR) of any adduct of £6 and cyclopentadiene. Ethyl l-(21propynyl)—2-oxocyclopentanecarboxylate (66) t—Butyl alcohol (850 mL) and potassium metal (26.2 g, 0.670 moles) were heated at reflux in a 2—L, 3—necked, round-bottomed flask fitted with a heating mantle, N2 purge, 250—mL addition funnel and tempered water (m20°C) reflux condenser until all of the potassium metal had dis- appeared. Then neat 2—oxocarboethoxycyclopentane (95 g, 0.609 moles) was added to this solution in a slow stream 106 (10 min) and refluxed for 15 min. To this refluxing solu- tion was slowly added propargyl bromide (90.6 g of an 80% solution in toluene, 0.609 moles) over a 25—min period (Caution: Rapid addition can cause the reaction mixture to boil out the condenser since the alkylation is quite fast and very exothermic). After the addition was complete t-butanol (N600 mL) was removed by distillation. The re- action mixture was cooled to room temperature and poured onto ice ( 500 g). The organic layer was separated and washed with water (3 x 200 mL) and the combined organic layers were evaporated under reduced pressure to remove the t—BuOH, taken up in chloroform (750 mL), dried (MgSOu), filtered, evaporated under reduced pressure and distilled (107°C at 2.00 Torr) to yield pure ethyl l-(2-propynyl)— 2—oxocyclopentanecarboxylate as a colorless liquid (95.6 g, 81% yield) 1H NMR (250 MHz, 00013) 52.158 (2H, q, {=7.3 Hz), 2.687, 2.680 (2H, d, AB, gd=2,7 Hz JAB=17 Hz), 2.60—2.06(m,6H), 2.026(1H,t,g:2.7 Hz), 1.227(3H,t,g=7.3 Hz). l3CMR (20 MHz, 00013, TMS) 6213.11, 170.23, 79.97, 70.99. 61.62, 58.75, 38.20, 32.57, 23.13, 19.79, 12.02; IR(neat) cm'l 3280(m), 2980(m), 1750(vs), 1725(vs), 1270(w), 1250(w), 1225(w), 1205(w), 1330(m), 1015(m), 930(w), 860(w), 810(w); M.S., m/e (rel. int.) 192(5,M+),166(38), 129(26), 138(53), 121(79), 120(28), 111(52), 110(31), 109(37), 93(92), 92(22), 91(95), 79(72), 78(30), 77(95), 67(20), 65(82), 62(23), 53(32), 51(22), 23(27), 21(37), 39(100)- 107 Anal. Calcd. for C C, 68.02; H, 7.27. 11H12O3‘ Found: C, 68.09; H, 7.32. Ethyl 1—acetOpyl—2-oxocyclopentanecarboxylate (86) 'U A mixture of ethyl l-(2-propynyl)-2-oxocyclopentane- carboxylate (86.0 g, 0.223 moles), MeOH (600 mL), H20 (150 mL), Hgo impregnated Dowex 50 H+ resin51 (200-200 mesh) (2.0 g) and cone. H2804 (2 drops) was magnetically stirred in a l-L, round-bottomed flask for 60 h at room temperature, filtered to remove the resin and evaporated under reduced pressure to remove MeOH. This mixture was taken up in chloroform (200 mL) and the aqueous layer was separated and washed with chloroform (l x 100 mL). The combined organic layers were dried (MgSOu), filtered and the solvent was removed under reduced pressure to yield pure ethyl l-acetonyl—2—oxocyclopentanecarboxylate (93.0 g, 0.239 moles, 99% yield) as a colorless liquid. lHMR (250 MHz, 00013) 52.121 (2H,q,g=7.3 Hz), 3.182, 2.099 (AB,2H, g=18.3 Hz), 1.95—2.65(6H,m), 2.127(3H,s), 1.237(3H,t,g=7.3 Hz); 13CMR (20 MHz, 00013, TMS) 6212.26, 205.12, 170 25, 61.22, 57.22, 27.22, 37.55, 33.22, 29.86, 19.79, 13.97; IR(neat) cm‘l 2925(m), 2910(m), 1750(vs), 1720(vs), 1220(m), 1250(m), 1205(s), 1368(s), 1325(m), 1282(m), 1255(m), 1230(8), 1168(8), 1128(5), 1110(m), 1050(w), 1030(m), 970(w), 920(w), 920(w), 858(w); M.S., m/e (rel. int.) 212(0.2,M+),167(12), 166(32), 121(7), 139(15), 138(7). 108 122(21), 123(26), 113(13), 111(32), 110(8), 97(17), 95(28): 71(8), 68(8), 67(20), 55(13), 23(100), 21(13), 20(10), 39(7). Apal. Calcd. for 011H160u: 0, 62.25; H, 7.60 Found: C, 62.12; H, 7.70. Hg++ catalyst on Dowex—50 H+ Resin Using a procedure similar to Newman's51 Dowex—50 R H+ resin (1.0 g) was washed with dilute sulfuric acid and was air dried overnight. The resin was suspended in 200 mL of dilute sulfuric acid to which was added mercuric oxide (0.10 g). The suspension was stirred magnetically for 22 h and filtered. The resin was washed with dilute sulfuric acid, air dried for 1 h and stored in a vial until needed. 2,5L6,6a-Tetrahydro—2(1H)-pentalenone (Z6) To a l-L round—bottomed flask fitted with a reflux condenser, 250-mL addition funnel, nitrogen purge, magnetic stirrer and heating mantle was added NaH (2.0 g, 0.167 moles) and dry (from K/benzophenone) toluene (300 mL) which was then brought to reflux. To the refluxing mixture was slowly added a toluene solution (200 mL) of ethyl l—acetonyl- 2-oxocyclopentanecarboxylate (8.0 g, 0.039 moles) over a 2—h period and the mixture was refluxed for 18 hrs. The reaction 109 mixture was cooled to 10°C and carefully acidified with 10% HCl (90 mL, 0.25 moles). The aqueous layer was sepa- rated and washed with ether (3 x 50 mL). The organic layers were combined, washed with saturated brine (l x 50 mL), dried (MgSOu) and filtered. After removing the solvent under reduced pressure, bulb-to—bulb distillation (100°C at 0.1 Torr) gave a colorless liquid (2.22 g) containing the desired ethyl 2,3,2,5-tetrahydro-5-oxo—3a(lH)-pentalene- carboxylate (12) which was not readily isolated on a preparative scale. Therefore the impure product was treat- ed with l% NaOH (100 mL) for 1.5 h at 0-10°C. The result- ing mixture was washed with chloroform (l x 50 mL) acidi- fied (10% HCl), and extracted with chloroform (3 x 50 mL). The chloroform extract was dried (MgSOu) and the solvent removed under reduced pressure to give a clear oil which was heated on a steam bath for 10 min (gas evolution was observed), diluted with chloroform (100 mL), washed with 5% NaOH (1 x 25 mL), dried (MgSOu), filtered and the solvent removed under reduced pressure to give 1.75 g (38% yield) of Z 95% pure (by NMR and VPC) 2,5,6,6a—tetrahydro-2(1H)— pentalene 76. Preparative scale VPC (5% FFAP on Chrom W AWDMCS, 1/2 in. x 6 ft., 160°C) gave pure 76. IR(neat) cm”1 2968(5), 2875(m), 1705(vs), 1625(8), 1252(m), 1375(w), 1315(m), 1258(w), 1176(m), 1158(m), 1108(w), 1082(W), 1028(W), 932(W), 872(m), 835(W), 819(w); UV A (MeOH) max 228 nm (5:12,200), 293(62); M.S. m/e (rel. int.) 122(100,M+), 110 123(8.5), 121(31), 107(22), 95(5), 92(72), 93(11), 91(13>, 81(5), 80(16), 79(52), 77(25), 66(39), 65(13); lHMR<60 MHz. 00013) 65.86(1H,br s),l.7-3.0(9H,m); l30MR<20 MHz, 00013, TMS) 6210.93, 191.21, 122.79, 26.72, 22.36, 31.16, 26.30, 25.55. Exact Mass Calcd. for C8H100: 122.07260. Found: 122.07317. Attempts to improve thepyield of 2,5,6,6a-tetrahydro-2(1H)- » pentalenone (16) The following five experiments were performed in an attempt to improve the yield of 16. In each case only the conditions of the aldol condensation of the diketoester 66 were altered. The rest of the procedure was the same as that described in the preceding experiment. Potassium hydride in refluxing toluene To a l—L round-bottomed flask fitted with a magnetic stirrer, 250-mL addition funnel, heating mantle, reflux condenser and N2 purge was added toluene (300 mL) and potassium hydride (6.7 g, 167 mmoles; the dispersion oil was removed with pentane). There was slowly added a toluene (200 mL) solution of 66 (8.0 g, 38.7 mmole) at 25°C. The mixture was stirred at 25°C for 16 h. VPC analysis (3% UCW/98 on Chrom W, 1/8 in. x 18 in.) of a small sample indicated that no reaction had 111 occurred. The mixture was refluxed and the reaction was monitored by VPC. After 6 h most of 66 had been consumed but no product peaks had appeared. Standard workup, hydrolysis and decarboxylation procedures failed to give any trace of Z6. Potassium tert—butoxide in refluxing toluene To a l—L round-bottomed flask fitted with a reflux condenser, 250-mL addition funnel, magnetic stirrer, N2 purge and heating mantle was added potassium Eggpfbutoxide (2.0 g, 21.6 mmole) and toluene (300 mL) which was brought to reflux. To this mixture was added dropwise (2 h) a toluene (200 mL) solution of 66 (8.0 g, 38.7 mmole). The mixture was refluxed for an additional 18 h. Standard work-up, hydrolysis and decarboxylation procedures gave 16 (0.990 g, 8.11 mmoles) in 21% yield and greater than 90% purity. Potassium tert-butoxide in refluxing THF To a 1—L round—bottomed flask fitted with a reflux condenser, 250—mL addition funnel, magnetic stirrer, N2 purge and heating mantle was added potassium Eggpfbutoxide (3.82 g, 20.0 mmole) and THF (300 mL) which was brought to reflux. To this was added dropwise (2 h) a THF (200 mL) solution of 66 (8.0 g, 38.7 mmole). The mixture was re— fluxed for an additional 18 h. Standard work-up, hydrolysis 112 and decarboxylation procedures gave 16 (0.528 g, 2.29 mmole) in 12% yield and greater than 90% purity. (Note: After quenching the base, most of the THF was removed under re- duced pressure and the residue was taken up in chloroform (100 mL)before continuing the standard work—up.) Potassium tert-butoxide in refluxing THF (inverse addition To a l—L round-bottomed flask fitted with a reflux condenser, 250-mL addition funnel, magnetic stirrer, N2 purge and heating mantle was added 66 (8.0 g, 38.7 mmole) and THF (300 mL) which was brought to reflux. To this was added dropwise (2 h) a THF (200 mL)solution of potassium Eggpebutoxide (2.0 g, 21.6 mmole). The mixture was refluxed for an additional 18 h. Standard work-up, hydrolysis and decarboxylation procedures gave 16 (0.895 g, 7.32 mmole) in 19% yield and greater than 90% purity. (Note: After quenching the base, most of the THF was removed under re- duced pressure and the residue was taken up in chloroform (100 mL) before continuing the standard work—up.) Potassium tert—butoxide in refluxing ether To a 1-L round—bottomed flask fitted with a reflux condenser, 250-mL addition funnel, magnetic stirrer, N2 purge and heating mantle was added potassium tert-butoxide (2.0 g, 21.6 mmole) and ether (300 mL) which was brought 113 to reflux. To this was added dropwise (2 h) an ether (200 mL) solution of 66 (8.0 g, 38.7 mmole). The mixture was refluxed for an additional 18 h. Standard work-up, hy— drolysis and decarboxylation procedures gave Z6 (0.887 g, 7.27 mmole) in 19% yield and greater than 95% purity. (Note: After quenching the base, most of the ether was removed under reduced pressure and the residue was taken up in chloroform (100 mL) before continuing the standard work-up.) cndo-Pentacyclo[8.5.l.02’9.05’9.Oll’lSJhexadeca—2,11(15)- diene-3,16—dione (6%) To a 50-mL pear—shaped flask, fitted with a magnetic stirrer, reflux condenser, N2 purge and heating mantle was added 2,5,6,6a—tetrahydro-2(lH)—pentalenenone (0.97 g, 7.95 mmoles), N—bromsuccinimide (1.30 g, 7.90 mmole) and carbon tetrachloride (30 mL). As the reaction mixture was refluxed, it became light brown until after about 2 min— utes when the mixture suddenly became water white again. The heating mantle was quickly replaced with an ice bath and the mixture was cooled to 5 to 10°C. The reaction mix- ture was filtered and the solvent removed under reduced pressure (m20 Torr) without heating. As soon as the sol- vent stopped distilling the flask was evacuated to 10"1 Torr for 2 minutes to remove as much solvent as possible. The resultant bromoenone, which rapidly decomposes when 112 concentrated, was neither purified nor characterized but immediately dissolved in dry THF (50 mL) and transferred to an addition funnel. The bromoenone solution was added dropwise and concomitantly with a solution of t-BuOK (0.82 g, 8.75 mmoles, in 50 mL of THF) over a 20—minute period to a 250—mL, 3—necked round-bottomed flask containing THF (50 mL) at -78°C and protected by an N2 purge. The dark red reaction mixture was stirred an additional 20 minutes and allowed to warm to just above 0°C when water (5 mL) was added and most of the THF was removed under reduced pressure. The residue was taken up in chloroform (100 mL) the organic layer separated, dried (MgSOu), fil— tered and the solvent removed under reduced pressure. Flash chromatography (10% acetone/CHC13) of the residue gave endo-pentacyclo[8.5.l.02’9.05’9.011’151hexadeca-2,ll(l5)- diene-3,16—dione (0.26 g, 29% yield) which is readily re— crystallized from CH2Cl2/Me0H, m.p. 135—12000 (dec). 1HMR (250 MHZ, CD013) 65.858(1H,S), 3.255(1H,d,l?5.2 HZ), 3.005 (1H,s), 2.672(1H,d,ge5.2 Hz), 2.65-1.60(12H,m); l30MR (20 MHz, 00013, TMS) 6207.22, 196.08, 187.91, 122.38, 120.87, 129.02, 58.60, 56.19, 52.01, 50.56, 31.61, 31.21, 31.12, 26.02, 23.87, 23.66; IR (KBr) cm'1 3025(w), 2950(m), 2875(w), 2855(w), 1768(vs), l690(vs), 1622(s), 1229(w), 1225(w), ‘ 1331(w), 1328(w), 1315(w), 1250(m), 1200(w), 1190(m), 1175(m), 1150(W), 1132(W), 1058(W), 1018(W), 978(W), 959(w), 938(W), 903(w), 889(m), 870(m), 852(W), 818(W), 788(W), 115 735(w), 710(w); M.S. m/e (rel. int.) 220(1.5,M+), 213(15), 212(22), 182(76), 183(21), 171(17), 170(60), 169(33), 168(25), 156(22), 155(25), 122(19), 121(50), 129(26), 128(26), 127(12), 120(23), 117(29), 116(18), 115(68), 91(56), 78(21), 77(28), 65(27), 62(21), 63(36), 53(22): 52(17), 51(25), 50(16), 21(26), 39(100,B): UV (MeOH) Amax(€) 229 nm (10,500), 317(79), 327(69). Anal. Calcd for 0 H 0 - 0, 79.97; H, 6.71. 16 16 2° Found: C, 80.08; H, 6.72. Also isolated by flash chromatography was 2,5—dihydro- 1mm.w0 2(lH)-pentalenone (66) (92 mg, 10%) m.p. 35°C; MHz, 00013) 66.03(1H,s), 5.77(1H,s), 2.83(6H,s); 13CMR (20 MHz, 00013, TMS) 6208.39, 187.22, 125.79, 131.68, 121.06, 36.00, 25.71; IR cm"1 3078(w), 2958(m), 2925(m), 2875(w), 2835(w), l700(vs), 1595(s), 1228(m), 1218(m), 1395(m), 1357(W), 1301(w), 1266(m), 1252(m), 1210(m), 1178(m), 1128(m), 1058(w), 1003(w), 988(m), 938(m), 866(w), 821(m), 800(m), 780(m); M.S. m/e (rel. int.) 121(8), 120(52,M+), 92(21), 91(100), 65(17), 63(11), 51(16), 50(10), 20(70), 39(35): UV (MeOH) A (6) 282 nm (16,300). max Anal. Calcd. for 08H80: 0, 79.97; H, 6 71. Found: C, 80.08; H, 6.72. Attempted trapping ofkg with cyclopentene To a 250—mL round—bottomed flask fitted with a N2 purge, magnetic stirrer and two 50-mL addition funnels was 116 added cyclopentene (22 g, 0.32 moles). To one funnel was added a THF (50 mL) solution of potassium Eggp-butoxide and to the other was added a THF (50 mL) solution of the bromoenone 66 (8.20 mmoles), prepared as described in the preceding experiment. The contents of the two funnels were added slowly (m0.5 h) and concomitantly to the stirred cyclopentene which was cooled to —78°C. The reaction mix— ture was allowed to warm to 25°C overnight, and was treated with water (10 mL). Most of the THF was removed under reduced pressure and the residue was taken up in chloro- form (100 mL) and washed with water (2 x 50 mL). The chloroform solution was dried (MgSOu) and filtered. The chloroform was removed under reduced pressure. Flash chromatography (8% EtOAc (CHC13) of the residue (0.62 g) gave 6% (15%), pentalenone Z6 (2%) and pentalenone 66 (2%) but no indication (lHMR) of any adduct of 66 and cyclopentene. Kinetic measurement of the decarbonylation of endo—penta— 2,10.05,10. 012’171octadeca—2,l2(l7)-diene- 239.053 9.011315]— cyclo[9.6.l.0 3,18-dione (88) and endo-pentacyclo[8.5.l.0 hexadeca—2,ll(15)-diene-3,l6-dione (66) Sealed 5 mm NMR tubes containing m0.2 M solutions of dimer (66) or (6%) in deuterated benzene were heated at the desired temperature (151, 162, 173°C for g6 and 82, 97, 109°C for 6%) in a constant temp bath (see Figure 2). The 117 tubes were periodically removed from the constant tempera- ture bath, and immediately cooled to 25°C and analyzed by 250 MHz NMR. Six to eight analyses were made between 0 and 85% reaction. The first order rate constant calcula— tions were performed by a least squares analysis of the data from duplicate runs and were based on the amount of starting material plus product formed for dimer 6%. Due to the complexity of the isomeric product mixture from dimer (66), however, its first order decarbonylation rate constant calculations were based on remaining starting material rela— tive to the residual protons in the deuterated benzene. The kinetic parameters for dimers (66) and (6%) can be found in Tables (2) and (5). Thermolysis of cndo-pentacyclof9.6.1.02’1O.05’10.012’17]- octadeca-2il2(l7)-diene-3,l8-dione (66) 2’10.05’10.012’l7]octadeca- A. Endo-pentacyclof9.6.l.0 2,l2(l7)—diene—3,l8-dione (66) (0.212 g, 0.791 mmole) was placed in a 25-mL flask and heated at 205°C where it melted and evolved a gas. After 5 minutes gas evolution had ceased and the product was cooled. IR and NMR analysis of the light yellow oil showed no starting material. Purifica- tion by flash chromatography (CHC13) gave l,2,5,6,7,8—hexa— hydrospiro[benz[f]indene—3,1'-cyclopentane1—l-one (66) . . 1 (0.187 g, 93% yield) as a clear Viscous o11. HMR (180 MHz, 00013, TMS) 67.21(1H,s), 7.l7(1H,s), 2.83(2H,m), 2.53(2H,s), 118 l.8l(l2H,m). l3CMR<20 MHz, 00013, TMS) 6186.A5, 159.28, 195.66, 136.98, 13u.3u, 123.95, 123.26, 52.56, M9.u6, A1.95, 30.58, 29.38, 25.02, 22.99, 22.89; IR(neat) cm'l 3015(m), 3990(8), 2870(m), 1712(VS), 1619(8), 1578(W), 1M85(w), 1955(m), 1938(m), 1U10(w), 1336(w), 1320(m), 1288(w), 1258(m), 1230(w), 1195(w), 1162(w), 1113(m), 1062(w), 978(W), 950(W), 928(m), 875(w), 822(w), 757(m); M's. (0.1.) m/e (rel. int.) 2U1(100,M++1), 2AO(52,M+), 212(43), 211(20), 199(55), 198(87), 183(29), 170(35), 169(20), 155(33): 153(23), 152(22), 142(31), lU1(92), 129(39), 128(61), 115(70), 91(25), 82(39), “1(63), 39(53)5 UV (CH3CN) Amax(€) 25“ nm (11,400), 300(3,380), 330(225). Anal. Calcd. for C H 0: C, 8A.95; H, 8.39 17 20 Found: C, 85.08; H, 8.39. B. A sealed 5-mm NMR tube containing a “0.2 M solution Of 39 in C6D6 was heated at 173°C during a kinetic run. After about 80% conversion there were an additional nine peaks in the region 06.0 to 8.0 of the 250 MHZ lHMR spectrum; 07.70, 7.00, 6.75, 6.10, 5.9M, 5.81 (20, vinyl proton), 5.7M, 5.66, 5.61 and 5.A8. The peaks at 55.98, 5.7“ and 5.61 are similar in shape to the vinyl proton peaks in the 250 MHZ lHMR of 90 (derived from 21) at 56.12, 5.89 and 5.31, suggesting the presence of the expected 91 (see Figure 5). When the same solution was heated to 195°C for 12 h only two peaks remained in that region other than C6D5H (67.70 119 and 7.00). These peaks were also observed in mixture be- fore heating at 195°C. When the deuterated benzene was 1 replaced with deuteriochloroform a HMR spectrum was ob- tained which was identical to that obtained for 92 above. Thermolysis of endo-pentacyclo[8.5.1.02’9.05’9.011’151— hexadeca—U,ll(15)-diene-3516‘dione (€11 The sealed 5 mm NMR tubes containing m0.2 M solution of 2% in C6D6 used for the decarbonylation kinetic studies were heated at the temperatures at which the kinetic measure- ments were made until no trace of starting material could be observed by NMR. The resulting 250 MHz lHMR was con- sistent in all cases with a quantitative conversion to the expected tetracyclo[10.3.0.01’9.03’7J—pentadeca—2,7,ll— trien-lO-one (90). The contents of the tubes were com— bined, the solvent was removed under reduced pressure and the light yellow oil was characterized after passing through a short column (SG/CHC13). 1HMR (250 MHz, 0606) 06.12(1H,m), 5.89(1H,s), 5.29(1H,s), 2.91(1H,m), 2.60—0.9 12H,m). 130MB (20 MHz, 00013, TMS) 212.u6, 192.63, 139.69, 128.69, 118.2A, 11U.62, 56.51, 56.13, 39.81, 31.A5, 31.12, 25.23, 2u.86, 22.u9. IR (thin film) cm"l 29u0(s), 2860 (m,sh), l690(vs), 1620(s), 1A25(m), 1368(w), 1300(m), 1237(m), 1205(w), 11U8(m), 1087(w), 896(w), 8A3(m), 830(sh), 783(m). M.S. m/e (rel. int.) 213(9), 212(U0), 211(13), 18u(90), 183(26), 171(21), 170(66), 169(83), 168(27), 120 155(68), 153(22), 1u2(25), l“l(73), 129(81), 128(63), 128(21), 117(75), 116(32), 115(100), 91(62), 78(22), 77(51): 67(22), 65(20), 53(21), “1(26), 39(38). Exact Mass. Calcd for Cl5H16O: 212.12012. Found: 212.11998. APPENDIX APPENDIX 1 Single Crystal X-ray Bond Lengths, Bond Angles and Positional Parameters for 20 and 21 mm ’b Fractional Atomic Coordinates for 20 Atom X Y Z 0(1) .209095 .74602u .237258 0(2) .08u618 .7u6225 .13698u 0(3) -.058502 .780163 .192610 C(A) —.163796 .705689 .183528 0(5) -.101006 .630u17 .136510 0(6) —.170266 .539880 .105229 0(7) —.136739 .516378 .037831 0(8) .023697 .519890 .065560 0(9) .086781 .615572 -.036203 0(10) .059051 .6u3522 .10u635 0(11) .165376 .59u025 .199793 0(12) .126755 .615228 .339011 0(13) .060778 .550126 .u3577g 0(1uA) .067060 .59672u .57162 0(1uB) -.007979 .601066 .550303 C(15A) .091185 .671918 .599398 C(15B) .0305A6 .698922 .567817 0(16) .118779 .750961 .887339 0(17) .153356 .70u170 .362506 0(18) .295018 .662213 .189955 0(19) - 078578 .855A29 .239193 0(20) .417588 .651893 .15u935 H(1) .25279 .794750 .262383 H(2) .102115 .77uu80 .065696 H(u) -.263396 .712321 .208uuu H(61) —.268u85 .592310 .121122 H(62) -.133668 .995828 .168105 H(7l) —.1698u2 .5u9235 .088353 H(72) -.159895 .u51856 .060296 H(81) .022930 .508982 .156001 H(82) .077u77 .977050 .00uu18 H(9l) .026u55 .658392 .092366 H(92) .203291 .619025 .071503 H(11) .176150 .526755 .183785 H(131) .137339 .50u228 .u550u1 H(132) -.050539 .517332 .392599 121 Fractional Atomic Coordinates for 20 - Continued. Atom X Y Z H(l6l) .028889 .787736 .1180116 H(l62) .200596 .772877 .511231 H(1AA1) .169038 .590605 .6082A7 H(1AA2) —.0017u0 .5640H6 .632751 H(luBl) -.02A632 .557161 .624658 H(1MB2) —.103031 .628321 .524100 H(15Al) .182200 .6A5702 .62280“ H(15A2) .OUO706 .703389 .6735Hl H(15Bl) .028278 .726065 .656579 H(15B2) -.O75914 .705358 .531077 123 Bond Lengths for 20 Atoms Distance S.D. Atoms Distance S.D. 0(2) 0(1) 1.537 .006 0(15B) 0(11B) 1.185 .012 C(17) C(1) 1.511 .006 H(11A1) 0(11B) 1.736 .001 0(18) 0(1) 1.536 .006 H(11A2) 0(11B) 1.007 .000 H(1) 0(1) .857 .036 H(11B1) 0(11B) 1.009 .000 0(3) 0(2) 1.515 .007 H(11B2) 0(11B) .995 .000 0(10) 0(2) 1.556 .005 H(15B2) 0(11B) 1.659 .000 H(2) 0(2) .856 .031 C(15B) C(15A) .755 .012 0(1) 0(3) 1.158 .006 0(16) C(15A) 1.651 .010 0(19) 0(3) 1.211 .001 H(11A1) 0(151) 1.390 .000 0(5) 0(1) 1.333 .005 H(lSAl) 0(151) .919 .000 H(1) 0(1) .953 .010 H(15A2) C(15A) 1.003 .000 0(6) 0(5) 1.503 .005 H(15B1) 0(151) 1.112 .000 C(10) C(5) 1.516 .005 H(15B2) 0(151) 1.755 .000 0(7) 0(6) 1.536 .007 0(16) C(15B) 1.385 .010 H(61) 0(6) .916 .037 H(161) C(lSB) 1.581 .051 H(62) 0(6) .972 .011 H(11B2) C(15B) 1.661 .000 0(8) C(7) 1.500 .007 H(15A1) 0(15B) 1.691 .000 H(71) 0(7) .772 .010 H(15A2) C(15B) 1.092 .000 H(72) 0(7) .995 .037 H(15B1) 0(15B) .995 .000 0(9) 0(8) 1.511 .006 H(15B2) C(15B) 1.051 .000 H(81) 0(8) .911 .013 0(17) 0(16) 1.188 .006 H(82) 0(8) 1.073 .011 H(161) 0(16) .987 .067 0(10) 0(9) 1.525 .005 H(l62) 0(16) .852 .050 H(91) 0(9) 1.015 .038 0(20) 0(18) 1.190 .005 H(92) 0(9) 1.130 .013 H(62) H(61) 1.191 .058 0(11) 0(10) 1.559 .005 H(72) H(71) 1.156 .017 0(12) 0(11) 1.506 .005 H(82) H(8l) 1.782 .051 0(18) 0(11) 1.555 .006 H(92) H(91) 1.731 0.56 H(11) 0(11) 1.002 .032 H(162) H(161) 1.621 .081 0(13) 0(12) 1.503 .005 H(15B2) H(161) 1.627 .000 0(17) 0(12) 1.315 .001 H(11A2) H(11A1) 1.633 .000 0(111) 0(13) 1.551 .011 H(15Al) H(11Al) .828 .000 0(11B) 0(13) 1.529 .010 H(1131) H(11A2) .217 .000 H(131) 0(13) .991 .015 H(11B2) H(11A2) 1.730 .000 H(l32) C(13) 1.212 .016 H(11B2) H(11Bl 1.633 .000 0(11B) 0(111) .725 .013 H(15B2) H(11B2) 1.156 .000 0(151) 0(11A) 1.157 .012 H(15A2) H(15A1 1.633 .000 C(15B) 0(11A) 1.532 .011 H(15Bl) H(15A2) .392 .000 H(11Al) 0(11A) 1.012 .000 H(15B2) H(15B1 1.633 .000 H(11A2) 0(11A) 1.010 .000 H(11Bl) 0(11A) 1.157 .000 H(11B2) 0(111) 1.698 .000 H(15A1) 0(11A) 1.380 .000 C(15A) 0(11B) 1.168 .013 121 Bond Angles for 20 Atom 1 Atom 2 Atom 3 Angle 0(2) C(1) C(17) 108.66 0(2) C(1) C(18) 99.82 0(2) 0(1) H(1) 123.19 0(17) C(1) C(18) 96.86 0(17) 0(1) H(1) 103.76 C(18) C(1) H(1) 121.20 0(1) 0(2) 0(3) 113.01 0(1) 0(2) 0(10) 101.66 C(1) C(2) H(2) 115.79 0(3) 0(2) 0(10) 105.11 C(3) 0(2) H(2) 109.09 0(10) 0(2) H(2) 108.11 C(2) C(3) C(1) 107.75 0(2) 0(3) 0(19) 125.21 0(1) C(3) C(19) 126.99 0(3) 0(1) 0(5) 110.72 . 0(3) 0(1) H(1) 122.76 2. 0(5) 0(1) H(1) 126.52 2. 0(1) 0(5) 0(6) 128.18 0(1) 0(5) 0(10) 113.09 . 0(6) 0(5) 0(10) 118.31 . 0(5) C(6) 0(7) 108 11 . 0(5) 0(6) H(61) 110.07 1. 0(5) 0(6) H(62) 107.20 2. 0(7) 0(6) H(61) 112.05 2. 0(7) 0(6) H(62) 111.62 2. H(61) 0(6) H(62) 101.28 1. 0(6) 0(7) 0(8) 111.86 . C(6) C(7) H(71) 115.36 3. C(6) C(7) H(72) 113.23 2. C(8) 0(7) H(71) 103.51 3. C(8) C(7) H(72) 101.32 2. H(71) 0(7) H(72) 110.21 5. C(7) C(8) 0(9) 111.11 . 0(7) 0(8) H(81) 110.15 2. C(7) C(8) H(82) 107.55 2. C(9) C(8) H(81) 110.81 2. 0(9) 0(8) H(82) 103.06 2. H(81) C(8) H(82) 121.02 1. 0(8) 0(9) 0(10) 111.17 . 0(8) 0(9) H(91) 101.12 1. C(8) C(9) H(92) 109.10 1. 0(10) 0(9) H(91) 106.16 2. 0(10) 0(9) H(92) 116.75 1. 125 Bond Angles for 28 — Continued. Atom 1 Atom 2 Atom 3 Angle S.D. H(91) C(9) H(92) 107.77 3.67 C(2) C(10) C(5) 102.81 .33 C(2) C(10) C(9) 115.87 .32 0(2) 0(10) 0(11) 102.71 .31 0(5) 0(10) 0(9) 109.11 .33 0(5) 0(10) 0(11) 111.29 .31 0(9) 0(10) 0(11) 111.50 .35 0(10) 0(11) 0(12) 110.67 .30 0(10) 0(11) 0(18) 98.03 ..28 0(10) 0(11) H(11) 111.18 1.99 0(12) 0(11) 0(18) 96.29 .30 0(12) 0(11) H(11) 112.11 1.83 0(18) 0(11) H(11) 122.91 2.10 0(11) 0(12) 0(13) 126.32 .31 0(11) 0(12) 0(17) 109.08 .33 0(13) 0(12) 0(17) 121.51 .35 0(12) 0(13) 0(11A) 107.53 .19 0(12) 0(13) 0(11B) 111.19 .12 0(12) 0(13) H(131) 106.00 2.39 0(12) 0(13) H(132) 110.33 1.89 0(111) 0(13) 0(11B) 27.19 .17 0(111) 0(13) H(131) 95.21 2.71 0(111) 0(13) H(l32) 122.23 2.18 0(11B) 0(13) H(131) 117.93 2.67 C(11B) C(13) H(132) 97.30 2.08 H(131) 0(13) H(132) 113.69 1.16 0(13) 0(111) 0(11B) 71.15 .93 0(13) 0(111) 0(151) 130.17 .67 0(13) 0(11A) C(15B) 113.37 .60 0(13) 0(111) H(11Al) 109.23 .01 0(13) 0(111) H(11A2) 109.13 .02 0(13) 0(111) H(11Bl) 100.18 .02 0(13) 0(11A) H(11B2) 80.01 .01 0(13) 0(111) H(15A1) 126.73 .01 0(11B) 0(11A) 0(151) 99.95 1.62 0(11B) 0(111) C(15B) 72.50 1.38 0(11B) 0(11A) H(11Al) 175.76 .00 0(11B) 0(11A) H(11A2) 68.68 .02 0(11B) 0(11A) H(11Bi) 59.75 .02 0(11B) 0(11A) H(11B2) 10.76 .00 0(11B) 0(11A) H(15A1) 112.75 .01 C(15A) 0(11A) C(15B) 28.51 .86 0(151) 0(111) H(11A1) 79.38 01 0(151) 0(11A) H(11A2) 111.55 01 C(15A) 0(11A) H(11Bl) 119.86 01 126 Bond Angles for 20 - Continued. Atom 1 Atom 2 Atom 3 Angle .D. 0(151) 0(11A) H(11B2) 89.31 .01 C(15A) 0(111) H(15Al) 12.81 .01 C(15B) 0(11A) H(11Al 107.32 .00 C(15B) 0(111) H(11A2) 109.92 .01 0(15B) 0(11A) H(11Bl) 109.93 .01 C(15B) 0(111) H(11B2) 61.71 .00 0(15B) 0(11A) H(15A1) 70.77 .01 H(1111) 0(11A) H(11A2) 107.71 .02 H(11Al) 0(11A) H(11Bl) 116.89 .02 H(11A1) 0(111) H(11B2) 168.10 .00 H(11Al) 0(11A) H(15Al) 36.60 .01 H(11A2) 0(11A) H(11Bl 10.56 .00 H(11A2) 0(111) H(11B2) 71.56 .02 H(11Bl) 0(11A) H(11B2) 66.67 .02 H(11Bl) 0(11A) H(15A1) 129.51 .02 H(11B2) 0(11A) H(15A1) 132.15 .02 0(13) 0(11B) 0(11A) 78.37 .03 0(13) 0(113) C(15A) 110.66 .56 0(13) 0(11B) C(15B) 117.69 .51 0(13) 0(11B) H(11Al) 80.52 .02 0(13) 0(11B) H(11A2) 111.23 .02 0(13) 0(11B) H(11B1) 109.57 .02 0(13) 0(11B) H(11B2) 110.37 .02 0(13) 0(11B) H(15B2) 120.71 .01 0(11A) 0(11B) 0(151) 50.95 .32 0(11A) 0(11B) 0(15B) 79.71 .16 0(111) 0(11B) H(11Al) 2.17 .00 0(111) 0(11B) H(11A2) 69.19 .01 0(111) 0(11B) H(11Bl) 81.91 .01 0(11A) 0(11B) H(11B2 161.12 .01 0(111) 0(11B) H(15B2) 118.22 .01 C(15A) 0(11B) C(15B) 29.63 .62 0(151) 0(11B) H(11Al) 50.58 .01 0(151) 0(11B) H(11A2) 93.16 .02 C(15A) 0(11B) H(11Bl) 106.11 .02 0(151) 0(11B) H(11B2) 110.66 .02 0(151) 0(11B) H(15B2) 67.96 .02 C(15B) 0(11B) H(11Al) 79.68 .01 C(15B) 0(11B) H(11A2) 113.71 .02 C(15B) 0(11B) H(11Bl) 123.86 .02 C(15B) 0(11B) H(11B2) 81.68 .01 C(15B) 0(11B) H(15B2) 38.56 .01 H(11Al) 0(11B) H(11A2) 67.11 .01 H(11Al) 0(11B) H(11Bl 79.97 .01 127 Bond Angles for 20 - Continued. Atom 1 Atom 2 Atom 3 Angle .D. H(11A1) 0(11B) H(11B2) 161.21 .01 H(11A1) 0(11B) H(15B2) 118.23 .01 H(11A2) 0(11B) H(11Bl) 11.08 .00 H(11A2) 0(11B) H(11B2) 119.50 .01 H(11A2) 0(11B) H(15B2) 128.00 .01 H(11Bl) 0(11B) H(11B2) 109.07 .01 H(11B1) 0(11B) H(15B2) 128.16 .01 H(11B2) 0(11B) H(15B2) 13.20 .00 C(11A) C(15A) C(11B) 29.11 .81 0(111) 0(151) C(15B) 101.18 .58 0(11A) C(15A) 0(16) 121.50 .96 0(11A) 0(151) H(11Al 15.70 .02 0(111) 0(151) H(15A1) 81.19 .01 0(111) 0(151) H(15A2) 122.33 .01 0(111) 0(151) H(15Bl) 133.51 .01 0(111) C(15A) H(15B2) 89.95 .01 0(11B) C(15A) C(15B) 76.12 .11 0(11B) 0(151) 0(16) 110.16 .71 0(11B) C(15A) H(11A1) 71.75 .02 0(11B) 0(151) H(15A1) 110.28 .02 0(11B) 0(151) H(15A2) 107.39 .02 0(11B) 0(151) H(15B1) 110.67 .02 0(11B) 0(151) H(15B2) 61.20 .02 C(15B) C(15A) 0(16) 56.11 .01 C(15B) 0(151) H(1111) 118.60 .01 C(15B) 0(151) H(15A1) 165.28 .00 C(15B) C(15A) H(15A2) 75.32 .02 C(15B) C(15A) H(15B1) 59.11 .02 C(15B) 0(151) H(15B2) 16.18 .00 0(16) C(15A) H(11Al) 121.16 .01 0(16) 0(151) H(15Al) 108.91 .01 0(16) C(15A) H(15A2) 106.20 .02 0(16) C(15A) H(15Bl) 87.17 .02 0(16) 0(151) H(15B2) 70.11 .01 H(11A1) C(15A) H(lSAl) 35.55 .00 H(11A1) 0(151) H(15A2) 125.11 .01 H(11Al) 0(151) H(15Bl) 111.96 .01 H(11Al) C(15A) H(15B2) 135.18 .01 H(15A1) 0(151) H(15A2) 113.19 .02 H(15A1) 0(151) H(15B1) 126.11 .01 H(15A1) C(15A) H(15B2) 168.93 .00 H(15A2) 0(151) H(15B1) 19.69 .00 H(15A2) C(15A) H(15B2) 76.88 .02 H(15Bl) 0(151) H(15B2) 61.63 .02 128 Bond Angles for 20 - Continued. Atom 1 Atom 2 Atom 3 Angle .D. C(11A) C(15B) C(11B) 27.75 .61 C(11A) C(15B) C(15A) 17.01 .93 C(11A) C(15B) C(16) 115.06 .88 C(11A) C(15B) H(161) 111.90 .55 C(11A) C(15B) H(11B2) 61.06 .02 C(11A) C(15B) H(15Al) 50.11 .00 C(11A) C(15B) H(15A2) 90.82 .00 C(11A) C(15B) H(15B1) 111.71 .01 C(11A) C(15B) H(15B2) 107.13 .00 C(11B) C(15B) C(15A) 73.91 .05 C(11B) C(15B) C(16) 126.60 .78 C(11B) C(15B) H(161) 136.09 .31 C(11B) C(15B) H(11B2) 36.30 .01 C(11B) C(15B) H(15Al) 78.03 .01 C(11B) C(15B) H(15A2) 101.15 .00 C(11B) C(15B) H(15B1) 119.38 .01 C(11B) C(15B) H(15B2) 79.73 .01 C(15A) C(15B) C(16) 96.52 .19 C(15A) C(15B) H(161) 132.96 .68 C(15A) C(15B) H(11B2) 109.19 .02 C(15A) C(15B) H(15Al) 8.20 .00 C(15A) C(15B) H(15A2) 62.69 .01 C(15A) C(15B) H(15B1) 80.22 .02 C(15A) C(15B) H(15B2) 152.26 .01 C(16) C(15B) H(161) 38.16 .65 C(16) C(15B) H(11B2) 127.59 .02 C(16) C(15B) H(15A1) 88.39 .02 C(16) C(15B) H(15A2) 120.71 .01 C(16) C(15B) H(15Bl) 109.90 .02 C(16) C(15B) H(15B2) 106.21 .02 H(161) C(1SB) H(11B2) 110.12 .02 H(161) C(15B) H(15A1) 125.20 .01 H(161) C(15B) H(15A2) 121.21 .01 H(161) C(15B) H(15B1) 101.21 .02 H(161) C(15B) H(15B2) 73.31 .01 H(11B2) C(15B) H(15Al) 111.22 .02 H(11B2) C(15B) H(15A2) 111.62 .00 H(11B2) C(15B) H(15B1) 118.66 .01 H(11B2) C(15B) H(15B2) 13.50 .01 H(15Al) C(15B) H(15A2) 67.98 .01 H(15A1) C(15B) H(15Bl) 82.95 .01 H(15A1) C(15B) H(15B2) 157.71 .01 H(15A2) C(15B) H(15Bl) 20.97 .01 129 Bond Angles for 20 - Continued. Atom 1 Atom 2 Atom 3 Angle S.D. H(15A2) C(15B) H(15B2) 115.51 .01 H(15B1) C(1SB) H(15B2) 105.91 .01 C(15A) C(16) C(15B) 27.01 .19 C(15A) C(16) C(17) 108.17 .59 C(15A) C(16) H(161) 107.81 3.56 C(15A) C(16) H(162) 101.39 3.21 C(15B) C(16) C(17) 112.71 .66 C(1SB) C(16) H(161) 81.71 3.18 C(15B) C(16) H(162) 123.32 3.36 C(17) C(16) H(161) 111.33 3.58 C(17) C(16) H(162) 103.51 3.15 H(161) C(16) H(162) 123.52 6.95 C(1) C(17) C(12) 107.13 .35 C(1) C(17) C(16) 128.23 .39 C(12) C(17) C(16) 121.10 .36 C(1) C(18) C(11) 95.66 .33 0(1) C(18) 0(20) 132.98 .39 C(11) C(18) 0(20) 131.35 .36 Fractional Atomic Coordinates for 21 130 Atom x Y 2 0(1) -.100286 .188385 .359223 0(2) -.051211 .111577 .210117 0(3) -.091771 .111226 .071513 0(1) -.009805 .190892 .002011 0(5) .075381 .196101 .102268 0(6) .187310 .210520 .098011 0(7) .239306 .126309 .230021 0(8) .152887 .066798 .302103 0(9) .061803 .157773 .259373 0(10) .061732 .268179 .378008 0(11) -.011283 .368169 .313898 0(12) -.009616 .198393 .252790 0(13) -.121239 .528217 .203108 0(11) -.186291 .113295 .238515 0(15) —.106281 .323682 .305516 0(16) -.003760 .205500 .180511 0(17) -.183763 .129768 .015659 0(18) .013199 .178773 .612687 H(11) -.158875 .150601 .391817 H(21) —.056250 .021197 .251683 H(11) - 017166 .203997 .099571 H(61) .208158 .298003 .121302 H(62) .201712 .190851 .003107 H(71) .277055 .051127 .186700 H(72) .281610 .177055 .316886 H(81) .131683 .020229 .256300 H(82) .166912 .053306 .113112 H(101) .129656 .295111 .128587 H(121) .027603 .501109 .161866 H(l22) .021819 .556222 .328687 H(131) —.112007 .517766 .098901 H(132) -.119617 .599071 .269182 H(111) -.236817 .127156 .297913 H(112) -.233393 .379898 .153002 131 Bond Lengths for 21 Atoms Distance .D. 0(2) 0(1) 1.556 .006 C(15) C(1) 1.509 .006 0(16) 0(1) 1.538 .005 H(11) 0(1) .958 .031 0(3) 0(2) 1.529 .006 C(9) 0(2) 1.555 .005 H(21) 0(2) .935 .031 0(1) 0(3) 1.170 .005 C(17) C(3) 1.215 .001 0(5) 0(1) 1.316 .005 H(11) 0(1) .902 .037 0(6) 0(5) 1.191 .006 0(9) 0(5) 1.192 .005 0(7) 0(6) 1.511 .008 H(61) C(6) .980 .015 H(62) 0(6) .916 .012 0(8) 0(7) 1.513 .007 H(71) 0(7) .968 .016 H(72) 0(7) 1.027 .017 0(9) 0(8) 1.532 .005 H(81) 0(8) 1.019 .010 H(82) 0(8) .985 .035 0(10) 0(9) 1.570 .005 0(11) 0(10) 1.506 .005 0(16) 0(10) 1.537 .005 H(101) 0(10) .915 .027 0(12) 0(11) 1.185 .006 0(15) 0(11) 1.332 .005 0(13) 0(12) 1.539 .007 H(121) 0(12) 1.013 .011 H(122) 0(12) .967 .038 0(11) 0(13) 1.530 .007 H(131) 0(13) .911 .053 H(132) 0(13) 1 012 .061 0(15) 0(11) 1.172 .006 H(111) 0(11) .929 .010 H(112) 0(11) .966 .017 0(18) 0(16) 1.191 .005 H(62) H(61) 1.537 .061 H(82) H(81) 1.593 .050 H(122) H(121) 1.597 .055 H(112) H(111) 1.389 .058 132 Bond Angles for 21 S Atom 2 Atom 3 Angle Atom 1 0311252330.“. 3“. 59392081“ 9117u022u8u5uuu73u 361“».3 33938933030733u311uuu. 355975.“. 96630.”.28080 33333 1 ll 2 21 22 22225 22225 2121U. 0662235.“.28623776181308650560023 85953538328 837931626576 765u567u9979252u870 188892 35923 ............................................ )))))) ) )) )) )) ))))) ))))) ))))) ) ) 56l6ll))l)ll)77)l l))))12122)l2 l22)l2l22))0)0 1 ll 111392922u115uu6 997666668 77777988888 58181 (((((((((((( (:(((((((((((((((((((((( (((((((( CCHCHHCCHCHHCOOCHHCCCCHHHHHCHHHHHCHHHHHCCCCC )\./) ))\I/)\|/\Il)\n/) ))))))))) ))))\I/\./) ))))))))))) ))))) 1 11 111222222 0333qu4.555666666777777888888 99999 ((( ((((((((( ((((((((( (((((((((((((((((( ((((( CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC ))) ) ) ) )\./\/ 556 )\./\/)\./\/)\./\./))\/\./\.l\./ )\./\/\/\.ll)\/\./\./)l )))))l )))\./\.l 2 22 111 1 1133922.” 335.4.46 55577666688777.7998 222 55 {K(((((/.\(/.\((((((((((((((((((((((((((/\(((((((( CCCCCCCCCCCCCCCCCCCCCCCCCCHCCCCCHCCCCCHCCCCC 133 Bond Angles for gl — Continued. Atom 1 Atom 2 Atom 3 Angle S.D. C(8) 0(9) C(10) 111.80 .38 C(9) C(10) 0(11) 109.29 .31 0(9) C(10) C(16) 97.09 .30 C(9) C(10) H(lOl) 117.62 1.6M C(11) C(10) C(16) 96.10 .30 C(11) C(10) H(lOl) 117.62 1.69 C(16) C(10) H(lOl) 115.30 1.63 C(10) C(11) C(12) 138.01 .36 0(10) 0(11) C(15) 109.63 .37 0(12) 0(11) 0(15) 112.30 .ul 0(11) C(12) C(13) 102.83 .37 C(11) C(12) H(121) 111.03 2.29 0(11) 0(12) H(122) 111.62 2.20 0(13) 0(12) H(121) 110.62 2.29 C(13) C(12) H(122) 113.33 2.22 H(121) 0(12) H(122) 107.uu u.71 0(12) C(13) 0(1u) 108.31 .uo 0(12) 0(13) H(131) 113.52 2.99 0(12) C(13) H(132) 112.03 3.08 0(14) 0(13) H(131) 108.73 3.11 0(1u) 0(13) H(132) 103.05 3.31 H(131) C(13) H(l32) 110.20 6.67 0(13) 0(1u) C(15) 102.87 .39 C(13) 0(1u) H(lul) 116.56 2.52 0(13) 0(1u) H(1u2) 115.u5 2.79 C(15) 0(1u) H(1u1) 11u.26 2.u8 C(15) C(14) H(1u2) 11u.05 2.70 H(1u1) 0(10) H(1u2) 9u.30 5.0M C(1) C(15) C(11) 108.61 .35 C(1) C(15) 0(1u) 137.52 .u0 0(11) 0(15> C(1u) 113.70 .39 C(1) 0(16) C(10) 97.95 .33 C(1) C(16) 0(18) 131.32 .39 0(10) 0(16) 0(18) 130.73 .38 REFERENCES 3a. 3b. 3c. 9a. 9b. 9c. 10. 11. REFERENCES P. A. Waitkus, E. B. Sanders, L. 1. Peterson, and G. W. Griffin, J. Amer. Chem. Soc., 89, 6318 (1967). G. W. Griffin and L. I. Peterson, ibid., 8Q, 3398 (1962). A. J. Barkovich, E. S. Strauss and K. P. C. Vollhardt, ibid., 22, 8321 (1977). P. Schiess and M. Heitzmann, Helv. Chim. Acta., 61, 8AA (1978). L. G. Haruff, M. Brown and V. Boekelheide, J. Amer. Chem. Soc., lgg, 2893 (1978). G. K5brich and E. Heinemann, Angew. Chem. Int. Ed. Engl‘: Q, 59“ (1965). Private communication from Professor Fumio Toda, Ehime University, Matsuyama, Japan. H. Hopf and A. K. Wick, Helv. Chim. Acta., HQ, 380 (1961). “ H. Hopf and A. Gati, ibid., 3%, 1289 (1965). M. Sasaoka and H. Hart, J. Amer. Chem. Soc., 199, 4326 (1978). Unreported result of Dr. A. Sy while a postdoctoral student of H. Hart, Michigan State University. M. B. Stringer and D. Wege, Tetrahedron Lett., 3831 (1980). J. H. Gall, C. J. Gilmore and D. D. MacNicol, J. C. S. Chem. Comm., 927 (1979). M. J. S. Dewar and G. J. Gleicher, J. Amer. Chem. Soc., 81, 692 (1965). For a review of cyclopentadienones see: M. A. Ogliaruso, M. G. Romanelli and E. 1. Becker, Chem. Rev., 6%, 261 (1965). 13A 12. 13. 14a. 19b. 15. l6. 17. 18. 19. 20a. 20b. 20c. 21. 22. 23. 2A. 25. 26. 27. 135 O. L. Chapman and C. L. McIntosh, Chem. Comm., 770 (1971). P. A. Baron and R. D. Brown, Chem. Phys., 1, AAA (1973). W K. Ziegler and B. Schnell, Ann., £55, 266 (1925. G. Maier and S. Pfriem, Angew. Chem. Int. Ed. Engl., 11. 519 (1978). F. R. Japp and C. I. Burton, J. Chem. Soc., A20 (1887). C. F. H. Allen and J. A. VanAllan, J. Amer. Chem. Soc., 1%, 5165 (1950). G. Maier and F. Bosslet, Tetrahedron Lett., 1025 (1972). W. Dilthey, I. terHorst and W. Schommer, J. Prakt. Chem., 139, 189 (1935). "““—" C. F. H.IUJ£n1and J. A. VanAllan, J. Org. Chem., 17, 845 (1952). J. M. Holland and D. W. Jones, Chem. Comm., 587 (1969). J. M. Holland and D. W. Jones, J. Chem. Soc. (C), 608 (1971). A. Hassner and D. J. Anderson, J. Amer. Chem. Soc., 95. 8255 (1972). K. Yates, "Huckel Molecular Orbital Theory", Academic Press, New York, 1978, p. 27. R. Hoffmann,J. Chem. Phys., 39, 1397 (1963). Unpublished computer program. F. R. Ahmed and J. Trotter, Acta. Cryst., 16, 503 (1963). J. K. Fawcett and J. Trotter; ibid, 28, 87 (1966). W. Reusch, "An Introduction to Organic Chemistry", Holden-Day, Inc., San Francisco, 1977, p. 110. B. Bak, L. Hansen-Nygaard and J. Rastrup-Andersen, Disc. Far. Soc., 19, 30 (1955). 28. 29. 30. 31. 32a. 32b. 33a. 33b. 3A. 35. 36. 37. 38. 39. A0a. A0b. A00. 136 B. Bak, D. Christensen, L. Hansen-Nygaard and J. Rasgrup-Andersen, J. Molecular Spectroscopy, 7, 58-63 19 1 . H. A. Skinner and H. O. Pritchard, Trans. Far. Soc., A2, 1295 (1953). W. G. Dauben, J. W. McFarland and J. B. Rogan, J. Org. Chem., 26, 297 (1961). A. M. Islam and R. A. Raphael, J. Chem. Soc., A086 (1952). R. Joly, J. Warnant, G. Nominé, and D. Bertin, Bull. Soc. Chim. Fr., 366 (1958). P. L. Stotter and K. A. Hill, J. Org. Chem., 38, 2576 (1973). E. W. Garbisch, Jr., J. Org. Chem., 30, 2109, 1965. H. MA R. Hoffmann and J. G. Vinter, ibid, 99, 3921 (197 ). '—_" B. Smith, Synthetic Communications, 5, A35 (1975). A H. 0. House, L. J. Czuba, M. Gall and H. D. Olmstead, J. Org; Chem., 3%, 232A (1969). H 2 . 0. House, M. Gall and H. D. Olmstead, ibid, 36, 361 (1971). H. J. Reich J. M. Renga and I. L. Reich, J. Am. Chem. Soc., 91, 5&3u (1975). B. M. Trost, T. N. Salzmann and K. Hiroi, ibid, A887 A third technique for preparing a-bromoketones is a base promoted reaction using a quaternary ammonium tribromide salt as the source of Br:°. This methodology was not investigated in this study. D. V. C. Awang and S. Wolfe, Can. J. Chem., A7, 706 (1969). L. A. Paquette, R. H. Meisinger and R. E. Wingard, Jr., J. Am. Chem. Soc., 95, 2230 (1973). M. Fetlzon, M. Jurion and A. Nguyen Trong, Organic Preparation and Preparations Int., 6, 31 (197A). Al. A2. A3. AA. A5. A6. A7a. A7b. A7c. A8. A9. 50. 51. 52. 53a. 53b. 530- 5Aa. sub. 137 J. C. Stickler and W. H. Pirkle J. Org, Chem. 3uuu (1966). ’ ’ 81’ M. E. Jung, Y. G. Pan, M. W. Rathke, D. F. Sullivan and R. P. Woodbury, ibid., 92. 3961 (1977). K. Hafner and K. Wagner, Angew. Chem. Internat. Edit., G. R. Newkome and D. L. Fishel, J. Org. Chem., 31, 677 (1966) C. H. DePuy, M. Isaks, K. L. Eilers and G. F. Morris, J. Org. Chem., 3503 (196A). P. Laszlo and P. V. R. Schleyer, J. Am. Chem. Soc., fié’ 1171 (196A). M. Miyashita, T. Yanami and A. Yoshikoshi, J. Am. Chem. Soc. ,98, A679 (1976). Private communication from S. Danishefsky. Private communication from J. A. Berson. D. Becker, N. C. Brodsky and J. Kalo, J. Org. Chem., Ag. 2557 (1978). B. M. Trost, D. P. Curran, J. Am. Chem. Soc., 102, 5699 (1980)- W. Herz, J. Am. Chem. Soc., 78, 1A85 (1956). M. S. Newman, J. Am. Chem. Soc., 75, A7A0 (1953). J. E. Baldwin, Can. J. Chem., AA, 2051 (1966). R. McCulloch, A. R. Rye and D. Wege, Tetrahedron Lett., 5231 (1969)- M. Sakai, ibid., 2297 (1973). A. Diaz, R. Rubio, R. G. Hernandez, J. P. Daudy and R. Cetina, J. Org. Chem., A5, 150 (1980). B. Halton, M. A. Battiste, R. Rehberg, C. L. Deyrup and M. E. Brennan, J. Am. Chem. Soc., 89 596A (1967). J. M. Landesberg and J. Sieczkowski, J. Am. Chem. Soc., 91, 2120 (1969). 138 5A0. S. C. Clark and B. L. Johnson, Tetrahedron, 27, 3555 (1971). 5Ad. K. Kurabayashi and T. Mukai, Tetrahedron Lett., 10A9 (1972)- 5Ae. G. Kretschmer, I. W. McCay, M. M. Paddon-Row and R. N. Warrener, ibid., 1339 (1975). 55. An updated version of a program published by J. L. Dye and V. A. Nicely, J. Chem. Ed., A8, AA3 (1971). 56. Professor J. L. Dye is a Professor of Physical Chemistry at Michigan State University and is responsible for the development of KINFITA.55 57. S. Eankelevich and B. Fuchs, Tetrahedron Lett., A9A5 19 7 . 58. K.-T. Wei and D. L. Ward, Acta Crystallographica, 93%. 2768 (1976). 59. P. Main, "MULTAN78 A System of Computer Programs for the Automatic Solution of Crystal Structures from X-ray Diffraction Data", Univ. York, England (1978). 60. A. Zalkin, Private communication (197A). d 61. W. C. Still, M. Kahn and A. Mitra, J. Org, Chem., A3, 2923 (1978). 11qu 144 993 H R“ II||II N“ H. TH|| All] “Ill- "I 3 1293 03