‘r-y' . .135! ..'. F “M... h ‘ r I ‘ " “ ‘ ' r ‘ v ' V Ir a , ‘ . .‘ - w. I r» ‘ vl u y . ‘ o , q W')":L:H?);Q}n4xwflt _ .'}>“u;fl') J. {9'}. ,- - .' u" .‘njjfl. ‘Uu. . .9 ”.3. .n .I Q.- .«A ,| 4.: u .. ”a. .‘ ‘ n -.I ~-. . " ,. ‘Jn‘ .\ r. .....I.. .- w . .1 . V .. A .u‘ H , ":1“ .- ,. u. ,. v . .~ ... r', _~Vb‘l.l.I|Iv-. ""vK--avu-up....q... ‘I- A: . ‘... ... . .v‘wur 1-” .~.-.. . .-. .. ‘ ”wt-h -Im..u-_ r...“ ‘ ‘ - "~n ... ‘ ‘ ‘n .3“ ‘u‘ ‘9'“ . .. . . . . :‘;1,~.E;‘.‘ .— ‘ THE svm-nasss AND CHEMISTRY or vic-CYCLOPROPANEDIQLSf Thesis for the Degree of Ph. D. MICHiGAN STATE UNIVERSITY DUANE B. PRIDDY 1971 LI BRA R Y Michigan State University This is to certify that the thesis entitled The Synthesis and Chemistry of m-Cyclopropanediol 5 presented by Duane B. Priddy has been accepted towards fulfillment of the requirements for Ph.D. degreein chemistry 7mm W Major professor Date—MEL] 9 . 1971 0-7639 ABSTRACT THE SYNTHESIS AND CHEMISTRY OF gingYCLOPROPANEDIOLS gig¢Cyclopr0panediols IV, V and VI were prepared by lithium in ammonia reduction of 2,2-dimethy1cyclohex- ane-1,3-dione derivatives I, II and III respectively. V was also prepared by Clemmensen reduction of II. All of the gig-cyclopropanediols were readily converted to their respective diacetate derivatives and V was also converted to the dimethyl derivative IX. 0 o no H 3’ OR’ R 'lll' R. E 'R R 5 11 R R R R R R R Y Y I-III Iv-VI VII-x I R=H; Y=(CH3)2 . VI R=CH3; Y=0H,H II R=CH3; Y=O VII R=H, R'=AC; Y=(CH3)2 III R=CH3; Y=OH,H VIII R-CHB; R'-Ac; Y-O Iv R=H; Y=(CH3)2 Ix R=R'=CH3; Y=O v R=CH3; Y=O x R=CH3; R'aac; Y=0AC,H Acid and base catalyzed rearrangements of V and VI and the reaction of VII with base were studied and found to react as shown. sto4 V 7 0 0H HCl V \ H20 ’ 1) NaH, d6DMSO \ 2) HC) aqueous acid VI > or base 0 OH methanolic V I > I KOH Possible mechanisms for these reactions and the reaction of molecular oxygen with vic-cyclopropanediols are discussed. THE SYNTHESIS AND CHEMISTRY OF ViC-CYCLOPROPANEDIOLS BY AQXI Duane BEWDriddy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1971 ‘ f. 'I / . ,’ .I. I \ (v ' 1‘ DEDICATION To Donna Mae ACKNOWLEDGMENTS The author wishes to express his appreciation to Professor William Reusch for his assistance and encourage- ment during the course of this investigation. This work was supported by National Institutes of Health grant AM 10849. TABLE OF CONTENTS Page INTRODUCTIONOOOOOOCOOOOO0.000000COOOOOOOOOOOCCOO 1 RESULTS AND DISCUSSION. 0 O O O O O O O O O O O O C O C O O C O O C C O . 6 EXPERIMENTALOOOOOOOOOO.OOOOOOOOOOOOOOOOOOOOOOOOO 25 Genera-1......0.0.0.0.0.0....0.00.00.00.00... 25 Preparation of 2,2,4,4,6,6-hexamethy1-1,3 S trihydroxybicyclo['3.1.0] hexane (185.. 25 2,2,4,4,6,6-hexamethyl-5-hydroxycyclohexane- 1'3-dione (19)....OOOOOOOOOOOOOOOOOOOO. 26 2,2,4,4,6,6-hexamethyl-5-acetoxycyclohexane- 1’3-dione (20)....O.OOOOOOOOOOOOOOOOOOO 27 2,2,4,4,6,6-hexamethyl-l,3,5-triacetoxybicy- clo[ 3.1.0] hexane (21)................ 27 2,2,4,4,6,6-hexamethyl-l,3-dihydroxy-5-acet- oxybicyclo[ 3.1.0] hexane (22)......... 28 2,2,4,4-tetramethy1-cis-3,5-dihydroxy-S-iso- propylcyclopentanone (23).............. 28 l-isopropyl-Z,2,4,4-tetramethyl-5-oxo-cis-l, 3-cyclopentylene cyclic sulfite (24)... 3O 2,2,5,5-tetramethyl-l,3—acetoxybicyclo [ 3.1.0] hexane (27)....O...OOOOOOOOOOOOOOCCOCCO 30 2,2,5,5—tetramethyl-3-hydroxycyclohexanone (28)....OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 31 2,2,4 4,6,6-hexamethyl-l,3-dihydroxybicyclo (3.1.0] hexane-S-one (29).............. 31 2,2,4,4,6,6-hexamethyl-l,3-diacetoxybicyclo 301.0] hexane-5-0118 (30)oooooooooooooo 32 TABLE or CONTENTS (Cont.) Page 2,2,4,4-tetramethy1-S-isopropy1-5-hydroxy- cyclopentane-l,3-dione (32)........... 33 2,2,4,4,6,6-hexamethyl-5-methoxycyclohexane- 1'3-dione (34)...COOOOOOOO00.0.0000... 33 2,2,4,4,6,6-hexamethyl-l,3-dimethoxybicyclo [3.1001hexane-5-one(35)ooooooooooooo 34 2,2,4,4-tetramethyl-5-isopropylidenecyclo- pentane-l,3-dione (38,000.000000000000 35 2,2,4,4-tetramethylcyclopentane-l,3—dione (39 OOCOOOOOOOOOOOOOOOO0......00...... 3S Attempt to prepare 2,2,4,4,6,6-hexamethyl— 1-hydroxy-3-methoxybicyclo[ 3.1.0] hexane- S-One (41)...0..0......OOOOOOOOOOOOOOOOOOCO 35 BIBLIOGRAPHYOOO....0...OOOOOOOOOOOOOOOOOOOCCOOO 37 APPENDIXOOOOOOOOOOOOO0..OOOOOOOOOCOOOOOOOOOOOOO 39 LIST OF FIGURES Figure Page 1. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl-1,3,5-trihydroxybicyclo[ 3.1.0] hexane...‘0.00.000...OOOOOOOOOOOOOOOOOOOOO 39 2. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl-S-hydroxycyclohexane-l,3-dione..... 4O 3. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl—S-acetoxycyclohexane—l,3-dione..... 41 4. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl—1,3,5-tracetoxybicyclo[ 3.1.0] hexaneOOOOOOOOOOOOOOOOOOOOOOOOOOOO00...... 42 5. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl-l,3-dihydroxy-5-acetoxybicyclo [3.1.0]hexaneOOOOOOOOOOOOOOOOOOOOOOOOO... 43 6. Infrared spectrum of 2,2,4,4-tetramethyl- cis-3,5-dihydroxy-S-isopr0py1cyclopentanone 44 7. Infrared spectrum of l-isopropyl-2,2,4,4- tetramethyl-S-oxo-cis-l,3-cyclopenty1ene eyelic sulfiteOOOO...OOOOOOOOOOOOOOOOCCOOO 45 8. Infrared spectrum of 2,2,5,5-tetramethyl- 1,3-acetoxybicyclo[ 3.1.0] hexane......... 46 9. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl-l,3-dihydroxybicyclo [3.1.0] hexane- 5-oneOOOOOOOOOOOOOOOOOCOOOOOOOOOCOCOOCOOOO 47 10. Infrared spectrum of 2,2,4,4,6,6-hexamethyl- 1,3-diacetoxybicyclo[ 3.1.0] hexane-S-one. 48 ll. Infrared spectrum of 2,2,4,4-tetramethyl- S-isopropyl-S-hydroxycyclopentane-l,3-dione 49 LIST or FIGURES (Cont.) Figure Page 12. Infrared spectrum of 2,2,4,4,6,6-hexa- methyl-S-methoxycyclohexane-l,3-dione..... 50 13. Infrared spectrum of 2 ,2 ,4 ,4 6 ,6-hexa- methyl-1,3-dimethoxybicyclo[ 3.1.0] hexane- S-one...:.OOOOOOOOCOOOOOOOOOOOOOOOOOOCOCCC 51 14. Infrared spectrum of 2,2,4,4-tetramethy1- 5-isopropylidenecyclopentane-l,3-dione.... 52 15. Infrared spectrum of 2,2,4,4-tetramethyl— CYClOpentane-l, 3-di0ne.................oo. 53 16. Nmr spectrum of 2, 2 ,4 ,4 6, 6- -hexamethy1- 1, 3, 5- -trihydroxybicyclo[3 .1. 0] hexane.... 54 17. Nmr spectrum of 2,2,4,4,6,6-hexamethy1- 5-hydroxycyclohexane-l,3-dione............ 55 18. Nmr spectrum of 2,2,4,4,6,6-hexamethyl- 5-acetoxycyclohexane-1,3-dione............ 56 19. Nmr spectrum of 2,2 A ,4 6 ,6-hexamethy1- 1, 3 ,5 triacetoxybicyclo[ 3.1.0] hexane.... 57‘ 20. Nmr spectrum of 2,2,4,4-tetramethy1-cis- 3,S-dihydroxy-S-isopr0py1cyclopentanone... 58 21. Nmr spectrum of l-isopropyl-Z,2,4,4-tetra- methyl-S-oxo-cis-l,3-cyclopenty1ene cyclic SUlfiteooooooooooooooo00000000000000.0000. 59 22. Nmr spectrum of 2, 2 ,5 ,5- -tetramethy1- -1, 3- acetoxybicyclo[ 3.1.HO] hexane............. 60 23. Nmr spectrum of 2, 2 ,4 4, 6 ,6-hexamethy1- 1 ,3-dihydroxybicyclo[ 3.1.0] hexane-S-one. 61 LIST OF FIGURES (Cont.) Figure Page 24. Nmr spectrum of 2,2,4,4,6,6-hexamethyl- 1,3-diacetoxybicyclo[ 3.1.0] hexane-S-one. 62 25. Nmr spectrum of 2,2,4 4,6,6-hexamethy1- 1,3-dimethoxybicyclo [' 3.1.0 ] hexane-S-one. 63 26. Nmr spectrum of 2,2,4,4-tetramethy1-5- isopropylidenecyclopentane-l,3-dione...... 64 27. Nmr spectrum of 2,2,4,4-tetramethy1cyclopen- tane-l'B-dioneoooooooooooooooooooooooooooo 65 28. Nmr spectrum of product mixture from the attempt to prepare 2,2,4,4,6,6-hexamethyl- l-hydroxy-B-methoxybicyclo[ 3.1.0] hexane- S-Oneoooooooooooooooo.oooooooooooooooooooo 66 29. Nmr spectrum of 2,2,4,4,6,6-hexamethy1-l,3- dihydroxybicyclo[ 3.1.0] hexane in fluoro- sulfonic acid at various temperatures..... 67 INTRODUCTION This thesis reports the synthesis of zigycyclopro- panediols via intramolecular pinacol reductions of/g—di- ketones, and describes some of the characteristic chemical reactions of these compounds. Pinacols have been prepared by many methods, the most common being the treatment of ketones with active metals in aprotic or weakly protonating solvents.1 CH3 CH3 CH3 1) Mg/Hg O > H OH < 2) H® CH3 CH3 CH3 Pinacols have also been formed from ketones by elec- trochemical2 and photochemical3 reductions and may occur as by-products in Clemmensen4 and metal-ammonia5 reduc- tions. Intramolecular reductions of diketones to cyclic pinacols can occur, but have not been widely observed. The work of Wenkert and Yoder6, involving Clemmensen reduction of the 1,4-diketones 2,2,6-trimethy1cyclohep- tane-l,4-dione £1) and tetracyclo [6.3.01'8.05'9.O4'11] undecan-3,6-dione (2) to the yiggcyclobutanediols 3 and ~ ‘4 respectively, illustrate these transformations. 0 OH ———————> x O 0 OH ,1 3 ———9 N OH 2 “0 4 H ~ In 1933, Khuda7 subaected 5,5-dimethy1-cyclohexane- 1,3-dione (dimedone) (E) to Clemmensen reduction and Obtained a saturated monoketone which proved not to be the expected 3,3-dimethylcyclohexanone. On repeating this work, Dey and Linstead8 were able to identify the anomolous product as the ring contracted ketone 2,4,4- trimethylcyclohexanone (6). N; E. 2. Staschewski9 has proposed a mechanism for this re- arrangement (equation l), involving the formation of a zigfcyclopropanediol intermediate 1 which suffers acid catalyzed ring opening to 8 followed by subsequent reduc- tion to 6, ~ 3 (1) 69 H OH 2H M\]:;;;]/ ’E' -——————9 -——————> Z. l 6) further H3 m [0H2 2 reduction 0' E Having observed that brief Clemmensen reduction of 2-methy1-2-acetylcyclohexanone (10) gave a mixture of w ketol 11 and ketone 12 (the former diminishing on more vigorous reduction), Karin and Wenkert10 proposed an alternate mechanism (equation 2) in which the vic-cyclo- CH H 0 CH30 3 H O 3 CH3 CH3 CH3 0 propanediol intermediate is opened to anCXEketol which is then further reduced to a ketone. R’ R” , (2) R’R’ RI RI] 0 0 2H@ H R R 2 -; n ’ H e H R R R R n69 R’R” R’R”l H H o in? o 9 HO Chuang and Scott11 have described the Clemmensen reduction of/g-diketone 13, which can only proceed via the Wenkert mechanism, since formation of an(}%fllunsat- urated ketone is prohibited by the structural constraints of the[:2.2.1] bicycloheptane ring system (i.e. Bredt's Rule). As expected, ketol 14 proved to be the major product. , . ._ ! th—Ms‘ .0 OH l 42 . .5 o o 43 Recently, Curphey and McCartney12 provided further I t. I” support for the Wenkert mechanism by isolating the ketol 16 from a Clemmensen reduction of 2,2,4,4,6,6-hexamethy1- 1,3,5-cyclohexanetrione (15). OH 0 o o o -———-> + 0 o o 13 16 ~ After the investigations described in this thesis were essentially finished, two other research groups reported the isolation of vic-cycloPropanediol derivatives from reduction of 1,3-diketones. Curphey and co-workers13 found that electrochemical reduction of 10 and 15 in K.) tetrohydrofuran, using acetic anhydride as a trapping agent, gave the diacetates l6 and 17. 0 COACH3 2e- fifcna 10 > a» A020 \\v/ 16 Ac OAc 15 2e 'fi% “~ A020 0 17 14 " More recently, Le Goaller and his co-workers have investigated the reduction of 3,3-dimethyl-1,3-pentane- dione (18) by a sodium dispersion in the presence of tri- methylsilyl chloride. A mixture of products including trimethylsilyl ethers of cis and trans-cycloprOpanediols, was obtained. x0 H [H \\n/><\n// ____) xéipgx + ki;ix .+ ’;r;x:1i\ O C) ' X0 ‘OX 1 8 20% 10% 1 5% . .H H HA + W + W xo ox x0 ox XO ox 5% 20% 30% X a (CH3)3Si RESULTS AND DISCUSSION In this investigation, the preparation of gig: cyclopropanediols was first attempted by reduction of 2,2,4,4,6,6-hexamethyl—l,3,5—cyclohexanetrione with an excess of lithium metal dissolved in an ammonia-tetra- hydrofuran solvent mixture. The relatively insoluble powder thus obtained in 65-95% yield was essentially transparent in the carbonyl stretching protion of the infrared, except for a slight impurity absorbing at 1710 cm-1. Crystallization of the powder from acetic acid and pyridine followed by sublimation neither improved the melting point (melts with decomposition above 1600 ) nor removed the impurity. This material, however, gave a clearly defined nmr spectrum (p.54) which was consis- tent with its assignment as 2,2,4,4,6,6-hexamethy1-1,3, S-trihydroxybicyclo. {3.1.0} hexane (18). Treatment of 18 with refluxing acetyl chloride in acetic acid gave the triacetate derivative 21, charac- terized by a strong infrared absorption at 1750 cm-1. 'The nmr spectrum showed nine acetoxy protons at€52.12, one proton geminal to an acetoxy group at 4.95 and eighteen methyl protons arranged as singlets at 1.65, 1.15, 1.07 and 1.0 in a ratio of 3:6:6:3 respectively. 7 O O H H O O . ——-> :— OH 0 L5. / L8. AcO OAc H OH O O <——-— 0 OAc OAc OAc 2.3 23 29 Crude 18 was readily oxidized to the hydroxydiketone 19 by mild oxidizing agents such as aqueous-methanolic ferric chloride, chromic oxide in pyridine (Sarett's reagent) or molecular oxygen. Compound 19 exhibited 1 I strong infrared absorption at 1690 and 1720 cm- (typical of a non-enolizable/g-diketone) and the nmr spectrum displayed methyl singlets at(51.27, 1.21 and 1.19 in a ratio of 3:9:6, and two one proton doublets typical of a secondary hydroxy group at 3.69 and 3.51 (J==5 cps). Treatment of 19 with a refluxing solution of acetyl chloride in acetic acid gave a monoacetate derivative 29, which was characterized by infrared absorption at 1690, 1720 and 1730 cm-1. The nmr spectrum of 29 showed 3 acetoxy protons at(52.04, one proton geminal to an acetoxy group at 5.0 and 18 methyl protons appearing as three singlets at 1.10, 1.23 and 1.28 in a ratio of 6:9:3 respectively. Reduction of 19 with lithium in an ammonia—tetrahydrofuran solution produced a sample of 18 free from the carbonyl containing impurity mentioned previously. Furthermore, reduction of 29 by the slow addition of two equivalents of lithium to a refluxing ammonia-ether solution of the compound gave the monoacetate derivative 22. This result indicates that addition of two electrons to the/g-diketone moiety occurs much faster than reduction of the ester. The stereochemistry of compounds 18, 2} and 22 could not be established solely by spectrosc0pic analysis; however, chemical evidence points to a Sis-orientation for all the hydroxyl groups. Treatment of 18 with re- fluxing hydrochloric acid, refluxing methanolic-potassium hydroxide or thermolysis in an evacuated sealed tube at 2500 gave the gig-dihydroxy cyclopentanone 23, charac- l terized by a strong absorption at 1735 cm- in the infrared. The nmr spectrum of 23 showedtwo one proton doublets at (52.48 and 3.80 characteristic of a secondary hydroxyl group, a tertiary hydroxyl proton singlet at 2.88, a tertiary isopropyl group with restricted rotation dis- played as a one proton multiplet at 1.90 and two doublets (3H each) at 0.72 and 1.07 (J296.S cps), and four methyl singlets at 0.79, 1.07, 1.15 and 1.18. The gig relation- ship of the hydroxyl groups in 23 was established by its conversion to the cyclic sulfite ester 24, characterized by strong infrared absorption at 960, 1210 and 1760 cm-1. The nmr spectrum of 23 shows a one proton singlet at (54.20, a one proton multiplet at 2.18, two three proton doublets at 1.10 and 1.20 (J=:6.5 cps) and twelve methyl protons appearing as three singlets at 1.51, 1.28 and 1.12 in a ratio of 6:3:3 respectively. it} “0 Since the configuration at carbon atoms 2 and 4 in 23 cannot have changed during the reaction with thionyl chloride, it is clear that the stereochemistry of the three hydroxyl groups in 18 must also be cis. Reduction of 2,2,5,5-tetramethy1 cyclohexane-1,3- dione (25) by lithium in an ammonia-tetrahydrofuran sol- vent mixture gave an unstable white crystalline solid. This material,which was assumed to be the corresponding yigrcycloprOpanediol 26, was essentially transparent in the carbonyl stretching region of the infrared; but upon 10 exposure to air it reverted to the diketone 25 in a few hours. Fortunately, this diol could be trapped as the stable diacetate derivative 23 by treatment with a refluxing solution of acetyl chloride in acetic acid. The structure of 23 was confirmed by a strong acetate absorption at 1735 cm—1 in the infrared and a nmr spectrum displaying a six proton acetoxy singlet at(52.10, a four proton AB quartet for the methylene protons at 1.65 and 2.00 (J‘tl4 cps) and 12 methyl protons displayed as three singlets at 0.92, 0.96 and 1.16 in a ratio of 3:3:6 respectively. Saponification of this diacetate with refluxing methanolic-potassium hydroxide under nitro- gen gave 2,2,5,S-tetramethyl-3-hydroxycyclohexanone (28), which proved to be identical to an authentic sample prepared by partial reduction of 25.17 o 0 HO OH Ac GM: 0 OH = __._) —-——-) as a6 .23 as The slow addition of two equivalents of lithium to a solution of 25 in a refluxing ammonia and ether solution produced 2,2,4,4,6,6-hexamethyl-l,S-dihydroxybicyclo- [3.1.0 ] hexane-B-one (29). Crystallization of the 11 crude reduction product from ether gave 29 in the form of its monohydrate as indicated by a four proton hydroxy signal in the nmr. Sublimation of this hydrate gave pure 2?, which was characterized by strong infrared absorption at 1735 cm-1 and a nmr spectra displaying a two proton tertiary hydroxyl signal at(54.68 and 18 methyl protons appearing as four singlets at 1.18, 1.12, 1.02 and 0.93 in a ratio of 6:3:6:3 respectively. Treat- ment with acetyl chloride in acetic acid gave a diacetate derivative 29, displaying infrared absorptions at 1750 and 1735 cm-land nmr singlets at51.17, 1.27, 1.43 and 2.14 in a ratio of 3:9:6:6. H OH AcO 0A0 15 —--+> —————> O O 29 30 Some of these cyclopropanediol derivatives were examined by mass spectrometry, but the results were not very informative. Since the molecular ions could not be detected and since complex fragmentation patterns were observed, the molecular weights of 21 and 30 were confirmed by vapor pressure osmometry (due to the rapid air oxida- tion of dilute solutions of 29, its molecular weight could 12 not be determined by this method). This was done to eliminate any possibility of these compounds being dimeric pinacols such as 21 which might conceivably possess simi- lar chemical and spectrosc0pic properties. The observed molecular weights agreed well with the expected values: + 21, 347 - 10 (calc 340); 30, 297 i 6 (calc 296). 1 H? (in o{xr}o 4H2. 3. In Investigations of the chemistry of 2? yielded many interesting results. Treatment of 29 with a refluxing solution of methanolic—potassium hydroxide in the absence of oxygen gave the hydroxycyclohexadione 19. However, refluxing concentrated hydrochloric acid transformed 29 12 , which was identical into the hydroxycyclopentadione 22 with the product obtained from Jones oxidation of 23. Two modes of cyclopropyl ring opening have been ob— served (i.e. path a and path b). Cyclopropanediol 28 opens exclusively by path a with either acid or base OH 0 HO ‘l/~\\ OH ’//,//? lIIII o \ l3 catslysis, cyclopr0panediol diacetate a] is opened by path b on treatment with base and cyclopropanediol 22 opens by path a with acid and path b with base. Compound l§ only reacts via path b since reaction by path a would produce a sterically unfavored six membered ring a? due to crowding of the substituents on the ring. The planarity of the five membered ring in Z? is able to accomadate the bulk of all the substi- tuents much better. 0 OH 9322/22? OH 18 ~ path b {33 ‘\“~fi> 23 Since both modes of cyclopropyl ring opening were observed with 29 while only path a was observed for lg, the base catalyzed reaction of E? is apparently modified by participation of the carbonyl group (possibly as shown in the following equation). 9 H HO gsufisfi 06 i469 0 e0 06 6b o o 09 <———> <———> 0 De 06 14 Attempts to prepare the dimethyl derivative (§§) of cycloprOpanediol 22 revealed an unexpected reaction of the big conjugate base. When 29 was treated with two equivalents of sodium hydride in a 1:1 dimethylformamide- benzene solution followed by the addition of excess methyl iodide the major product was the methoxycyclohexanedione 35 characterized by infrared absorption at 1720 and 1690 cm.1 and a parent ion at m/e 226 in the mass spectrum. In contrast, similar treatment of a hexamethylphosphoramide (HMPA) solution of 2? gave the diether §§, characterized by infrared absorption at 1735 cm-1. The nmr spectrum of §§ shows the six methoxy protons as a singlet at 63.49, and 18 methyl protons as signlets at 1.34, 1.28, 1.23 and 1.11 in a ratio of 3:6:6:3. CH3. 0 CH3 CH3 0 0 ~12! 3,2 Apparently the conjugate base of 29 is sufficiently reactive to abstract a proton from dimethylformamide. Indeed, a similar proton transfer from dimethyl sulfoxide was disclosed by the isolation of 29 from the reaction of 2? with an excess of sodium hydride in perdeuterated dimethylsulfoxide followed by an aqueous work-up. 15 The surprisingly slow conversion of 2? to 33 in acid (8 hours in refluxing concentrated hydrochloric acid) prompted a study of the Clemmensen reduction of 15 under mild conditions. This led to the remarkable ~ discovery that reduction of I? with amalgamated zinc dust in a refluxing ethanol-concentrated hydrochloric acid mixture for two and a half hours gave 2? in nearly quantitative yield. Although cycloprOpanediol deriva- tives were recently isolated from Clemmensen type reduc- tions of/g-diketones, this work represents the first case in which a cyclopropanediol itself has actually been isolated from a Clemmensen reduction. Thus the intermediacy of cyclopropanediols in the abnormal Clemmensen reduction of 1,3-diketones has been established beyond question. The unusual stability of 29 in acidic media suggested that its conjugate acid might have the trishomocyclopro- penyl cation structure 22‘ Evidence supporting less highly substituted cations of this kind as intermediates in solvolsis reactions of 3-substituted bicyclo [3.1.0] hexane derivates has been reviewed by Winstein and co- workers.15 The importance of steric hinderance by the ggg. dimethyl groupings in 23 is difficult to estimate; however, the chair-like conformation required by 23 must suffer serious non-bonded methyl compressions. Further— more, Olah and co-workers16 have noted that the positive charge in protonated ketones resides mainly on the oxygen atom; consequently the effect of homoaromatic stabili- zation may be obscured by Charge localization by the hydroxyl substituents. Additional information concerning the question of homoaromatic stabilization of the conjugate acid of 2? was obtained from its nmr spectrum in strong acids. If 23 were formed,one would expect the methyl resonance signals in this ion to appear either as a single peak or a pair of equally intense peaks depending upon the rate of ring inversion. In any event, a significant simplification of the original nmr spectrum was antici- pated. The four methyl singlets displayed in the nmr spectrum of 22 in perdeuterated dimethyl sulfoxide solution were 17 essentially unchanged in mixed fluorosulfonic acid and sulfur dioxide solution at -400 . However, when the temperature of this solution was increased to -ld3 , several new peaks appeared, and on recooling this solu- tion to -4d3 the complex spectrum suffered only a slight signal broadening. The absence of any structural rear- rangement was demonstrated by recovery of unchanged 2? upon quenching the sample in ice water. Thus homoaromatic stabilization does not seem to be significant in this system. When 23 was dissolved in fluorosulfonic acid at room temperature and allowed to stand for forty-five minutes (or in concentrated sulfuric acid for twenty-four hours), it was transformed to the isopropylidenecyclopentanedione 38. This volatile crystalline solid was characterized 1 I by infrared absorptions at 1750, 1690 and 1606 cm- and an nmr spectrum displaying four methyl singlets at 62.30, 2.16, 1.38 and 1.09 in a ratio of 3:3:6:6. Further confirmation of this structure (39) was obtained by retro- aldol cleavage in refluxing methanolic-potassium hydroxide, which gave acetone (dinitrophenylhydrazone derivative mp 125-1260 ) and tetramethylcyclopentanedione 32 (a volatile oil having infrared absorption at 1760 and 1725 cm'l, nmr signals at\§1.16 (6H), 1.26 (6H) and 2.68 (2H), and a parent ion at m/e 154 in the mass spectrum. 18 O O ‘19“? —> +Y O O O 1.8 as The nature of the species obtained from solutions of 2? in strong acids and the mechanism for the rearrangement of 3g are not immediately obvious. Three possible re— arrangement pathways can in fact be conceived: HO R=H0r803F 19 Path 1 involves the formation of 13 as an intermed- iate, and the subsequent 1,2-acy1 shift is analogous to the acid catalyzed transformation of 2,2,5,5-tetramethy1- 3-hydroxycyclohexanone to 2—isopr0pylidene-4,4-dimethyl— cyclopentanone reported by Eschenmoser et a1.17 Although a solution of 13 in fluorosulfonic acid does in fact rearrange to §§ at a slightly faster rate than 2?, I? has never been detected in any acid catalyzed reactions of 2?. Path 2 is considered primarily because the intermediate was actually isolated from the reaction of 22 in aqueous acid, as described previously. Indeed, a solution of 32 in fluorosulfonic acid rearranged to 3g within a few seconds at room temperature. This unexpectedly facile dehydration of anCXEketol may proceed as a concerted elimination or via an epoxide. Path 3, in which a conjugate acid or a fluorosulfonate ester (£9) serves as a leaving group in the ring opening step, does not proceed gig an isolable intermediate. Nevertheless, this mechanism deserves serious consideration, since formation of the unsymmetrical intermediate 39 provides a rationale for the nmr observations described previously and would easily give 23 back again upon aqueous treatment. In this respect it should be noted that several attempts to obtain the unsymmetrical methyl ether 41 failed 20 (a mixture of 35 and 38 were always obtained from treatment of 29 with sodium hydride and one equivalent of methyl N iodide in HMPT solution); presumably due to a similar elimination-rearrangement reaction. c1130S fl Won > 35 + -—-——> 38 + CHaoe A: ~ 0 It was mentioned previously that yigfcyclopropanediols are readily oxidized to diketones by the action of oxygen. yig-Cyclopropanediols appear to be much more reactive toward oxygen than are cyclopropanols. Thus, an ethyl acetate solution of 2? absorbed one molar equivalent of oxygen in 30-60 minutes, giving 16 as the sole product (ca. 95% yield): and gé was readily oxidized to 2,2,5,5- tetramethylcyclohexane-l,3-dione even in the solid state. This should be compared with 24-48 hr. time reported for the oxidation of methyl substituted cycloprOpanols in hexanela. The absorption of oxygen by solutions of 23 was easily followed with the aid of a small gas buret, and the for- mation of a peroxide containing product was determined by iodometry. The results from several dozen experiments showed considerable variation: oxygen uptake was 70-95% of theoretical (assuming a 1:1 stoichiometry) and the 21 peroxide concentration always ran a bit lower. Since the gasometric and iodometric measurements corresponded more closely during the early stages of the oxidation, it appears that the peroxide suffered a slow decomposition. Other facts germane to this reaction are: l) A short induction period (ca. 5 min.) was normally observed. 2) The oxidation of 2? to 15 could also be effected by nitric oxide; however, the stoichimetry was complex (ca. 1:3.5 respectively). A transient blue color was noted in these reactions, but no intermediates could be isolated. 3) The dimethyl ether 36 and diacetate 39 derivatives of 23 were not affected by prolonged exposure to oxygen or nitric oxide. These facts suggest that the reaction of yigrcyclo— prOpanediols with oxygen proceeds by a radical mechanism (equation 3) similar to that prOposed by Gibson and DePule. The opening of the three membered ring is shown to be concerted with hydrogen abstraction in order to ration- alize the reactivity order: yig-cyclopropanediol>cyclo- propanol >other cyclic>3O -a1cohols. 22 HO H OH 0 + '02H ——-—) . '1' H202 O O OH O . O .O + 02 ———> + .0211 OH O O H 1+) 123an O 13 Numerous attempts to trap the intermediate carbinol radical by conducting the oxidation of 2? in solvent mix- tures having large mole excesses of tri-n-butylstannane or triphenylstannane were all unsuccessful. The conver- sion of 2? to 1§ proceeded at the customary rate and gave no detectable I? (as little as 0.5% VI could have been detected by our infrared and nmr analysis). These results suggest that carbinol radicals are oxidized by molecular oxygen with extraordinary ease; indeed, a similar oxygen effect was reported by Pitts et. al.19 in their study of the photoreduction of benzophenone. When three molar equivalents of pschlorothiophenol was added to a solution of 23 in ethyl acetate, the oxida- tion to 15 was completely stopped (ie. no change in the 23 concentration of 23 and no uptake of oxygen was detected over a 24 hr. period). The reasons for this behavior are not clear. It may be that the thiophenol scavanges hydroperoxy radicals so effectively that the chain reaction cannot function (this explanation implies that molecular oxygen does not itself significantly attack cyclopropanols). Alternatively, thiOphenol inhibition may be due to the hydrogen bonding effect discussed in the following para- graph. In the course of these studies a curious solvent effect on the rate of oxidation of 2? was noted: CH3C02C2H5>CHBCOCH3>C2HSOH>>Pyridine, DMSO. The time required for a test solution to absorb greater than 90% of the stoichiometric amount of oxygen ranged from 30-60 minutes for ethyl acetate and acetone to 24—48 hr. for pyridine and DMSO. A similar reactivity order was observed for nitric oxide oxidations. In addition, small amounts of DMSO were observed to inhibit oxidations in ethyl acetate solution. These facts can be explained by hydrogen bonding of the cyclOpropanol hydrogen atoms. Since intramolecular hydrogen bonding is precluded by the molecular geometry of the yig-cyc10propanediols, inter- molecular bonding, as in 22 can occur when strong hydrogen bond acceptors (eg Y = DMSO, pyridine) are present in the reaction mixture. These interactions apparently render radical attack at O-H more difficult. 24 Compound 22 also forms a hydrate, which may have a cyclic hydrogen bonded structure similar to 23. It was found that solutions of this hydrate in ethyl acetate are oxidized much more slowly than pure 23. EXPERIMENTAL EXPERIMENTAL General Melting points were taken in capillary tubes on a. Hoover-Thomas apparatus. Infrared spectra were recorded on a Perkin-Elmer 327B spectrophotometer in potassium bromide pellets or as neat liquid smears. Nuclear mag- netic resonance spectra were obtained with a Varian A-60 spectrometer. Tetramethylsilane was used as an internal standard with the exception of figure 23 in which tetra- methylammonium tetrafluoroborate was used. Elemental anal- yses were obtained from Spang Microanalytical Laboratory. 2,2J4,4J6,6-hexamethy1-1l3,S-trihydroxybicyclo[‘3.1.0 J hexane (18). To a 250 ml, three necked flask equipped with a magnetic stirrer, dry ice-acetone condenser, soda lime drying tube and addition funnel was added 100 ml of anhydrous ammonia and 270 mg of lithium metal. Two grams of 2,2,4,4,6,6-hexamethy1cyclohexane-l,3,5-trione20 (13) was dissolved in 25 ml of tetrahydrofuran and added drop- wise over 10 minutes to the refluxing ammonia solution. After a five minute reflux period, 5 g of ammonium chloride was added to the reaction mixture. The ammonia was allowed to evaporate under a stream of air and the residue was slurried with water and filtered to yield 1.9 g (94.5%) 0 of a white powder, melting with decomposition above 150 . 25 26 Anal. Calc for C12H2203: C, 67.29; H, 10.28. Found: C, 67.31; H, 10.25. The infrared spectrum of 18 is shown in figure 1 on page 39. The nmr spectrum is shown in figure 16 on page 54. 2,2,4444646-hexamethy1-5-hydroxycyclohexane-l,3-dione (19). a) A 100 mg sample of 18 was dissolved in a mixture of 100 m1 of methanol and 1 ml of concentrated hydrochloric acid. This mixture was kept at 40-500 for 3 days (exposed to the air) and the solution was allowed to evaporate under a stream of air until a thick slurry was obtained. Filtration gave 60 mg (60%) of long needles, mp 66-670 . b) To a mixture of 100 mg of chromic anhydride in 2 ml of pyridine was added a solution of 10 mg of 18 in 2 ml of pyridine. After stirring at room temperature for 20 minutes, 20 m1 of ether was added and the resulting solution was filtered. The filtrate was washed with water, 5% hydrochloric acid and allowed to evaporate under a stream of air, yielding 5 mg (50%) of white solid, mp 65-670 . c) A slurry of 4.28 g of 18 in 150 ml of water and 100 ml of methanol was treated with a solution of 16 g of ferric chloride hexahydrate in 150 ml of water. After stirring at room temperature for one hour, the solution was extracted twice with 100 ml portions of ether. The 27 combined ether extracts were dried over anhydrous sodium sulfate and evaporated on a rotary evaporator to yield 4 g of a pink oil. This oil was crystallized from hexane to give 3.5 g (82%) of long white needles, mp 66-670 . 5221. Calc for C12H2003: C, 67.92; H, 9.43. Found: C, 67.92; H, 9.39. The infrared spectrum of 19 is shown in figure 2 on page 40. The nmr spectrum is shown in figure 17 on page 55. 242,4,4,6,6-hexamethyl-5-acetoxygyclohexane-lyS-dione (20). A solution of 500 mg of 19 in a mixture of 10 m1 of acetic acid and 10 ml of acetyl chloride was heated at gentle reflux in an open beaker until only an oil remained. This oil was crystallized from hexane to give 350 mg (58%) of large white crystals, mp 76-79C . final. Calc for C14H2204: C, 66.14; H, 8.66. Found: C, 66.10; H, 8.70. The infrared spectrum of 29 is shown in figure 3 on page 41. The nmr spectrum is shown in figure 18‘on page 56 . 2424444.616-hexamethy1-1J3,S-triacetoxybicyclo[ 3.1.0] hexane (21). A 1 g sample of 18 was dissolved in a mixture of 50 ml of acetic acid and 10 m1 of acetyl chloride and heated at a gently boil in an open beaker until only an oil remained. The oil was crystallized twice from 50% 28 methanol giving 0.9 g (56%) of white needles, mp 158-1600 . Anal. Calc for C18H2806: C, 63.52; H, 8.24. Found: C, 63.52; H, 8.22. The infrared spectrum of 21 is shown in figure 4 on page 42 . 2,244,4,616-hexamethy1-lj3—dihydroxy55-acetoxybicyclo [3.1.0] hexane (22). To a 250 ml, three necked flask equipped with a magnetic stirrer and soda lime drying tube, was added 500 mg of 22, 100 m1 of anhydrous ether and 50 ml of ammonia. The flask was placed in a dry ice-acetone bath for 5 minutes, 30 mg of lithium metal was then added and the cooling bath was removed. After 15 minutes, the reaction mixture was decomposed with excess ammonium chloride and the ammonia was evaporated under a stream of air. The ether which remained was washed with water, dried over anhydrous sodium sulfate and evaporated to give 100 mg (20%) of large prisms. éflél- Calc for C14H24O4: C, 65.62; H, 9.40. Found: C, 65.54; H, 9.56. The infrared spectrum of 23 is shown in figure 5 on page 43 . 2,2,4,4-tetramethy1-cis-315-dihydroxy-5-isopropyl-cyclo- pentanone (23). a) To a 250 m1 three necked flask equipped with a magnetic stirrer and a nitrogen flushing valve was added 1.08 g of 18, 75 m1 of deoxygenated ethanol, and 50 m1 of 6N hydrochloric acid. The reaction mixture was 29 refluxed under nitrogen for 15 hours, concentrated to about half volume under a stream of nitrogen and the resulting solution extracted with ether. The ether ex- tracts were dried over anhydrous sodium sulfate and evaporated to an oil which was crystallized from hexane to give 600 mg (55.5%) of large white crystals, mp 89-91? . b) To a 250 ml, three necked flask equipped with a magnetic stirrer and nitrogen flushing valve was added 1.08 g of 18, 2 g of potassium hydroxide and 100 ml of deoxygenated methanol. This solution was refluxed for 4 hours under nitrogen, concentrated to about half its volume under a stream of nitrogen and the resulting solution extracted with ether. The ether extracts were dried over anhydrous sodium sulfate and evaporated to an oil, which was crystallized 15 times from hexane to give 150 mg (13.9%) of large prisms, mp 89-90.5o . c) A 1.08 g sample of sublimed 18 was placed in an evacuated pyrex tube (11 mm x 15 cm) and heated to 256) for 1 hour. The oil thus obtained was recrystallized 15 times from hexane to give 200 mg (18.5%) of large prisms, mp 89-90.5o . 5231. Calc for C12H2203: C, 67.30; H, 10.28. Found: C, 67.29; H, 10.58. The infrared spectrum of 13 is shown in figure 6 on page 44-. The nmr spectrum is shown in figure 20 on page 58 . 30 l-isopropyl-ZJ214,4-tetramethy1-5-oxo-cis-ll3—cyclopent- ylene cyclic sulfite (24). A solution of 100 mg of 23 in 15 ml of pyridine was cooled to -50 and 2 ml of thionyl chloride was added dropwise over 10 minutes. The reaction mixture was maintained at -50 for seven hours and then poured onto ice and extracted with ether. The ether solution was dried over anhydrous sodium sulfate and evaporated. The brown solid which remained was recry- stallized from hexane to give 70 mg (59.5%) of white needles, mp 68-700 . 5331. Calc for C12H2004S: C, 55.38; H, 7.68. Found: C, 55.51; H, 7.78. The infrared spectrum of 24 is shown in figure 7 on page 45. 242J5,5-tetramethyl-l,3-acetoxy bicyclof 3.1.0] hexane (27). To a 250 m1, three necked flask equipped with a magnetic stirrer and a nitrogen flushing valve was added 150 ml of ammonia, 180 mg of lithium metal and 2 g of 2,2,5,5- tetramethylcyclohexane-l,3-dionel7. After being stirred at reflux for one hour, the reaction mixture was decom- posed with an excess of ammonium chloride and the ammonia was completely evaporated under a stream of nitrogen. The residue was dissolved in a mixture of 50 ml of acetic acid and 50 ml of acetyl chloride; and this solution was heated to reflux under nitrogen for 30 minutes, and then poured into a beaker and evaporated at a gentle boil until 31 only a thick slurry remained. The slurry was added to cold water and extracted with hexane. The hexane extracts were washed with 10% sodium bicarbonate solution and con- concentrated to ca. 50 ml. This solution was cooled and the solid that appeared was removed by filtration to give 1.2 g (40%) white needles, mp 87-890 . 5&31. Calc for C12H2003: C, 67.92; H, 9.43. Found: C, 67.83; H, 9.53. The infrared spectrum of 22 is shown in figure 8 on page 46. The nmr spectrum is shown in figure 22 on page 60 . WW 2!2,545-tetramethyl-3—hydroxycyglohexanone (28). To a 100 ml, three necked flask equipped with a magnetic stirrer, a reflux condenser and a nitrogen flushing valve was added 100 mg of 23, 1 g of potassium hydroxide and 35 m1 of methanol. This mixture was heated at reflux under nitrogen for 1 hour and then concentrated to about half volume under a stream of nitrogen. 30 m1 of water was then added and the resulting solution was extracted with ether. The ether extracts were dried over anhydrous sodium sulfate and evaporated to give a white solid which was recrystallized from methanol to give 20 mg of white 17 0 crystals, mp 55—56.6o (lit mp 54-55 ). 2,2,4,4,6,6-hexamethyl-1,3-dihydroxybicycloI 3.1.0] hexane- S-one (29). 32 a) A 2.1 g sample of 15 was dissolved in 100 ml of anhydrous ether placed in a 250 ml, three necked flask equipped with a magnetic stirrer and soda lime drying tube. Anhydrous ammonia (75 ml) was added followed by 140 mg of lithium metal (in seven portions). After the blue color dissipated, the reaction mixture was decomposed with excess ammonium chloride and the reac- tion mixture was evaporated under a clean air stream. The residue was slurried in ether and water and filtered to give a white solid, which on sublimation gave 800 mg (38%) of white crystalline solid melting with decompos- ition above 1600. b) A solution of 6.3 g of 15 dissolved in 60 m1 of ethanol was added to a 500 ml, three necked flask containing 40 g of amalgamated zinc dust in 100 ml of water. 50 m1 of concentrated hydrochloric acid was added and the reaction mixture was refluxed two and one-half hours. The reaction mixture was cooled, fil- tered and allowed to stand. The crystals which formed were separated by filtration; resulting in 5.5 g (87%) of a white crystalline solid. £331. Calc for C12H2003: C, 67.92; H, 9.43: Found: C, 67.87; H, 9.45. The infrared spectrum of 29 is shown in figure 9 on page 47 . The nmr spectrum is shown in figure 23 on page 61. 2,2, 4,4,6,6-hexamethy1-l,5-diacetoxybicyclo L 3.1. o 1 hexane- 3-one (30). A 500 mg sample of 29 was added to a mixture 33 of 10 ml of acetic acid and 10 ml of acetyl chloride in a beaker and refluxed gently until only an oil remained. This syrup was crystallized from hexane to give 500 mg (71.7%) of white crystals, mp 128-1300 . 5331. Calc for C H 0 ° C, 64.87; H, 8.11: 16 24 5' Found: C 64.71; H, 8.07. I The infrared spectrum of 39 is shown in figure 10 on page 48 . The nmr spectrum is shown in figure 24 on page 62 . 21214,4-tetramethyl-5-isopropyl—S-hydroxycyclopentane- 1,3-dione (32). To a 100 ml, three necked flask equipped with a magnetic stirrer a reflux condenser and a nitrogen flushing valve was added 50 ml of concentrated hydro- chloric acid and l g of 3g. This mixture was heated at reflux under nitrogen for three hours, and then extracted with ether. The ether extracts were dried over anhydrous sodium sulfate and evaporated to give 1 g (100%) of a solid. Sublimation of this material gave a white cry- 12 37-3s°). stalline solid, mp 37-38.5O (Lit The infrared spectrum of 33 is shown in figure 11 on page 49 . 21245454616-hexamethy1-5-methoxycyclohexane-l,3-dione 1331, To a 100 m1, three necked flask equipped with a magnetic stirrer, a nitrogen flushing valve and an add- ition funnel was added 107 mg of 33, 25 ml of dry benzene, 25 m1 of dry dimethylformamide and 50 mg of sodium hydride. This mixture was stirred under nitrogen until hydrogen evolution ceased, and 1 m1 of methyl iodide was then added 34 followed by another 2 hours of stirring. Water and ether were added, the immiscable layers were separated, and the organic layer was dried over anhydrous sodium sulfate and evaporated to give 87 mg (77%) of an oil. Glpc analy- sis of this oil showed that it was ca. 98% pure. The mass spectrum of 35 showed a parent ion at m/e 226 (calc P + l and P + 2 for C 13.44 and 1.43 13H22°37 respectively; Found: 13.1 and 1.2. The infrared spectrum of 35 is shown in figure 12 on page 50 . 21244J4J646-hexamethy1-113-dimethoxybicyclo[ 3.1.0] hexane- 5—one (35). To a 100 ml, three necked flask equipped with a magnetic stirrer, nitrogen flushing valve and addition funnel was added 214 mg of 33, 10 ml of dry hexamethylphosphoramide and 100 mg of sodium hydride. After nitrogen evolution ceased, 1 ml of methyl iodide was added and the stirring was continued for one hour. The reaction mixture was quenched by 20 ml of water and the resulting solution was extracted with pentane. Evap- oration of the pentane extracts gave a solid which was sublimed to give 100 mg (45.5%) of a white crystalline solid, mp 113-1210 . C Anal. Calc for C 70.00; H, 10.00 14H2403‘ ' Found: C, 70.11; H, 10.04. The infrared spectrum of 35 is shown in figure 13 on page 51 . The nmr spectrum is shown in figure 25 on page 63. 35 2,244,4-tetramethy1-5-isgpropy1idenecyclopentane-lLB-dione 13311 A 400 mg sample of 33 was dissolved in 1.8 m1 of concentrated sulfuric acid, and after standing at room temperature for 24 hours, the mixture was poured into ice water and extracted with ether. The ether solution was dried over anhydrous sodium sulfate and evaporated to leave an oil, which on sublimation gave 270 mg (74.5%) of white crystals, mp 34.5—36C . The mass spectrum of 3g showed a parent ion at m/e 194 (calc P + 1 and P + 2 for C12H1802; 13.33 and 1.22 respectively; Found: 13.8 and 1.4). The infrared spec- trum is shown in figure 14 on page 52 . The nmr spectrum is shown in figure 26 on;page 64. 21214,4-tetramethylcyclopentane-1J3-dione (39). A mix- ture of 582 mg of 39, 174 mg of potassium hydroxide, 40 ml of 95% ethanol and 10 ml of water was heated at reflux for four hours, cooled and then diluted with water and extracted with ether. The ether extracts were washed with water, dried over anhydrous sodium sulfate and evap- orated to give 300 mg (65%) of a volatile brown oil. The mass spectrum of 33 showed a parent ion at m/e 154 (calc P + 1 and P + 2 for C9H1402; 10.03 and 0.85 respectively; Found: 10.9 and 1.1. The infrared spec- trum is shown in figure 15 on page 53 . Attempt to prepare 212,414,6,6-hexamethy1—l-hydroxy-3- 36 methoxybicyclo[ 3.1.0]hexane-5—one (41). A 200 mg sample of 29 was placed in a 100 m1, three necked flask equipped with a nitrogen flushing valve and magnetic stirrer. 10 m1 of dry hexamethylphosphoramide was added followed by 100 mg of sodium hydride. The reaction mixture was then stirred at room temperature for 10 minutes followed by the addition of 145 mg of methyl iodide. After stirring for another hour, 20 m1 of water was added and the re- sulting solution then extracted with pentane. Evaporation of the pentane gave an oil. Glpc analysis of the oil showed that it contained about 45% each of 35 and 38. The nmr spectrum of the oil is shown in figure 28 on page 66. Iodometric titration of mixture obtained during the oxida- tion of 2,g4414,6L6-hexamethyl-l,B-dihygroxybicyclo [_3.1.0jghexane-5-one. After a mixture of 22 in the desired solvent had been subjected to molecular oxygen, it was added to 10 times its volume of glacial acetic acid. An excess of saturated potassium iodide solution was added and the resulting mixture allowed to stand for 5 min. The liberated iodine was then determined by titration with 0.1 normal sodium thiosulfate solution. __ _ __.1 fl a: BIBLIOGRAPHY 11. 12. 13. 14. 15. BIBLIOGRAPHY R. Adams and E. W. Adams, Org. Syn. Coll. Vol. 1, 459 (1938). N. A. Izgaryshev and I. I. Aryamova, Doklady Akad. ‘ Nauk SSSR. 84, 313 (1952). firi N. C. Yang and Ding-Djung H. Yang, J. Am. Chem. Soc. 80, 2913 (1958). G. S. Risinger and J. A. Thompson, J. Appl. Chem. 13, 346 (1963). B. 0. de Montellano, B. Loving. T. Shields and P. , P. Gardner, J. Am. Chem. Soc. 89, 3365 (1967). E. Wenkert and J. E. Yoder, J. Org, Chem., 35, 2986 (1970). ' M. Qudrati-Khuda, Nature 132, 210 (1933). A. N. Dey and R. P. Linstead, J. Chem. Soc. 1063 (1935). . D. Staschewski, Angew. Chem. 71, 726 (1959). E. Karin and E. Wenkert, Israel J. Chem. 5, 68 (1967). V. T. C. Chuang and R. B. Scott, Chem. Comm. 758 (1969). T. J. Curphey and R. L. McCartney, J. Org. Chem. 34, 1964 (1969). T. J. Curphey, C. W. Amelotti, T. P. Layloff, R. L. McCartney and J. H. Williams, J. Am. Chem. Soc. 91, 2817 (1969). R. Le Goaller, M. Rougier, and C. Z. P. Arnaud, sub- mitted for publication. S. Winstein, E. C. Fredrich, R. Baker and Yang-i Lin, Tetrahedron, Suppl. 8, Part II, 621 (1966). 37 l6. 17. 18. 19. 20. 38 G. A. Olah, M. Calin and D. H. O'Brian, J. Am. Chem. Soc. 89, 3589 (1967). A. Eschenmoser, H. Sching, R. Fischer and J. Colonge, Helv. Chim. Acta 287, 2329 (1951). D. H. Gibson and C. H. DePuy, Tet. Letters, 2203 (1969). J. N. Pitts, R. L. Letsinger, R. P. Taylor, J. M. Patterson, G. Recktenwald and R. B. Martin, J. Am. Chem. Soc. 81, 1068 (1959). J. L. E. Erickson and G. C. Kitchens, J. Org. Chem. 27, 460 (1962). APPENDIX mcmxmn Ho.a.m HoHomoflnmxoup>£flhplm.m.Hlawnumemxmnnm.o.v.v.m.N mo Esuuommm UmHmHMCH .H musmfim com 000— comp 00: 002 009 a «.44 u «4 4 Il‘ 3 m _ . mgr 9%4\\£Jj\¥\/\Jr{q n3... 40 mCOHUIm.lecmxm£OHO>O>XOHU>£|mlaasuofimxoSIm.o.v.v.~.~ Ho sshuoomm UonHmcH com 000— oom_ oov— oop— .N whamwm oom— 41 ocoflolm.HchmxmcoaU>o>xoumomlmuaxnaoamxmnlm.0.v.v.m.N mo Enhuommm UmHMHMCH .m madman 000” 009 003 002 009 II. II! I...) : .4. 94'14) n1 1.“: q... .1 1 .Wru‘qo _ .. 8 ON O0fin>xouoomunhulm£.thnhnumeaxoalo.o.v.v.m.m mo annuowmm vohmumcH :v warm 000 000— 00m_ 003 002 009 42 mcmxmc a 0.4.m HoHomoflawxoumomlmlhxoupwgflvnm.Hlawnqumxmnlm.0.v.v.m.N mo esuuommm pwumuwcH .m madman 000w , 00E 003. 002 003 .lrlll ) w. 3 o<0 : 5%: 43 m m k..— xfixa wcocmucomoHUxoahmoumomwlmlhxonbwnwulm.mumfiolaanumsmuuwul¢.w.N.N mo Esuuoomm noumumcH .0 whamwh 000 000— 00a 2 003 002 000—. 44 oufimaam UHH0>U ocmahuchOHUMUIm.Hunwouoxonmla>nuw5muumulv.v.N.Nnawmoum0mflla mo Esuuoomm woumHMCH 000 000.. 005. 003. 002 45 .5 0.33m 46 mcmxmn ~0.H.mw OHU>OHQ>xouwomlm.Hla>cwwemnumunm.m.m.m mo Esnuummm UmHMHMCH .m ousmflm 000— 00.9. 003 002 002 « )Hu..~l|« . 4 . 1 .4.)«llfi . fi 4. .u- .. . q a ,lqulla“ _ a J).:— A la a - -1; («lonqt MAG: ~ : u . u a )4 a . i11|lJIIIHIU Mm <21... C £8 47 occlm -mcmxmn 3o.a.m0 oHoaanmxouusrauum.Huassumsmxonuo.m.v.v.~.~ mo gunpowmm umumumcH 000 0003 00$ 003 002 .m unseat 00.2 occlm lwcmxmsmo.a.mg oHo>oflQ>x0pmomfionm.Haa>£umemxmanm.m.v.v.m.N mo Esuuommm UmumnmcH .0H whoafim 0003 009 003 002 002 -»fil....H.l.fl -a n a a . a .(IFL)~.:H-; aid 3 . a H a a a JIJLL.L)H1|F.L.I-FIL.---F .1 48 5.. r u . . u in 5.321....» 49 mcoaunm.a locmucmmoaohuwxonvhnlmlammoumomwumlahsuwemuuwplw.v.N.N mo Eshpowmm 60HMHMCH 000 0003 00m_ oovp 000_ .HH whamfim oom_ 50 wcofivimaAlocmstoauhohxosuofilmlahsuoemxmnlw.0.¢.¢.~.N mo ennuoomm UmumumcH 000 000— 00N_ 00v— 0003 .NH ohduwh 0003 maelm -mnman fio.a.m _oaosoansxonumsau-m.Hnamnumemstnm.m.¢.¢.~.~ mo asuuommm cmumumcH 000 000— 00m_ 00v— 000— 51 .maonaxm 000. mCOAUIm.HamcmucmmoaUhomcmpfia>moumomflumlawzumEMHuwule.w.N.N mo Esuuommm UmHMHMCH .vH wusoflh ooo_ com; 0041 cool oomi q . q « a .W! Ll).~l|u — q a _ d 4 9%.! a. .u J 4. d o\\ //I 52 yfiefifxfiu 3.1!]. 53 000 ocowunm.HlmcmuchOHO>UH>n90EMHuouI¢.v.N.N mo anhuoomw uoumumcH 000_ 00m_ 00v. 0003 I an. O .ma manuam 0003 1 wcmxmc 20.H.m 54 g oaoxofinaxouc>nfiuunm.m.Hlamcummemslm.m.v.¢.m.m mo Esauommm HE: 038 o . 10;- .ma musofim . o.« O.” d1)414%14‘4d4‘ 4.11m- 1 41* 1 3111 I 4 A! I .- H‘. 114 “I“ I 3H — o u 0 d d o 1‘ q 1 .1 ‘ _ T 4 a . . q .q _ . 1 q q — 4 . u . — . q . d . wH =0 55 mcofiplm.Hsmcmchoaomuhkouo>nlmla>£poemxo£nm.0.¢.v.~.N mo Esuuummm HE: och .ha gunman .o .. . . 0.. I o.« . o.” 3. 1)1| 1 —\ 1 1 1 d 1 11 1 1H 1 1 1 1 ‘11 1‘1 1iflul1l " ._,. -- --——- —.—-—~———-——.—“ mo 56 mCOwoum.HawcmxonoHo>o>xoumUMIm|H>nummewnlm.0.¢.¢.N.N mo Esnuowmm Mac one .ma musoflm o o; - .----_-- i: o¢o mamxm: 20.H.mw oHo>0flQ>xouwomfilem.m.Hlfihnuoemxmnlm.m.v.v.m.N mo Esupommw HE: was 57 .ma masons -- --. -- -.o;-‘-- .-.,-ioa.- -- .. .Jod-- 9v .eezs. as _- 4 fl 4 a 1-.--41H--4.a.-.«41. _ _ ....... .1 . _. .1 _ .7 m . ; _ \ fi. \_ q I L 040 . BBQ/00¢ I . . . comm HE: 036 .ON musmwh wcocmucwmoaoxoammonmomfllm1>x0HU>£HCIm.mtmfiblamnumemupmu v v m N mo EDHD . It a o.— o." 0.0 9v 0.0 0.0 1111 1 1 1 1i)“ 1 11111 ‘11 11H 11 11 «l1 1 11H I 1 11 d 11 141 1 11 < 1 1 1 ‘H1 1 1 1 d 11 11H1 -.—-—. 58 ~_—4“—~--“ —.-- .‘—'.. .- 2U o 8. 8n g ouflwflnm Uflauxo mcmamucmmoauxolm.AlmaUIOXOImlawnqumuumuuv.¢.N.muaxmonmomwla mo Esuuommm HE: one .HN mhswflm - o 3 .o.u o... .. a I __L._ZJT.;1._.1_I2.... W .319. _ i. I I _ l 9 U 5 . I I . .. ol . /VN U /m )xml 60 >fl 200 100 OCPS> Ac 0A0 27 . i s : _L..-.l....l....l... is..1.........1.-.,.....1...‘...ij- .‘ 3.0 2.0 1.0 0” Figure 22. The nmr spectrum of 2,2,5,5-tetramethy1-l,3-acetoxy- bicyclo[ 3.1.0] hexane 61 mconm Imcmxmc w0.H.mH odowoflnwxouvwnflonm.Haaxnpomemcnm.0.¢.v.m.N mo Bohbowmm use one .mm ousmfim .o..- ...+o.f-. od 3.. 0...... _ _ _ .-..)3...-_:.)_.....123 — — ~_ —_.flqd—-.~.~.Jmod’u u.—o-.-—d...—.oou‘—..] .......- --.:T.z_--’-_ _-_--—...,,.—__-—_._._ g - ”a... -~- §-----. - 62 ozonm Imcmxmn H0.H.mw oaomufln>xouoowflolm.Hna>numemxmclw.m.v.¢.m.m mo Esuuoomm use one .VN ouswfim . 4 - . OH; O.“ O.” 0.? +. . . . J _ - ..44.fi- - . a -. -..L.. a- . _. . w- J _ 1i . _ .nd ... . .3. ..\1.... _ . .. ._ _ . . .1 .fi & . . 1 -h ‘- ——.——_—-. -~_—-.—-—- m...” o 35 \Jfi/ooa 63 ocoun Iocmxozmo.fi.m HoHozoubmxosuochnm.HufimcuomeonI0.0.v.v.m.N mo Enhpooam he: o:9.mm onsmfim o .3 gm 0.0 0.? 1 1 141' I 1 1 1 . d 1 1 1 1 H 1 1 1 1 q 1 1 I + F 1 1 1 1 d 1 1 111 # 1 1 1 1 d 1 I 1 + T — 1 4 J . < . — q < . q < d‘ . 1 d m < 1 . i— < . d . d 4 1 — . d a < . q q fl 1) .1 1 F o—o--- o—....---_ , -.. --——. -——.-— _. -. --——‘—— -- - -—-- .- n :11 O >§ m U 64 mCOHUIm.HamCMpcmmoHowomcmvflaxmoumomwlmlahnumEmHumulv.¢.N.m Mo Esupommm HE: was 0 .mm musofim ‘3 1 12} 1 o.» 3 . _ a _ 2431.137 — .4 --1— d u a . u L 1 11- 11 L. ...m. 1 1 11 1 1 11 a e t . m a J o . . m M w m . . h m‘ z u . M m w m _ u _ _ mm “ _ o -— -——————.—---- —..- ————-—-_ u g.. _—. ..——-._ ”at": \\ 64 mCOfivlm.HumcmucmmoHU>omcmvflaxmonmomfilmIH>£qumHumulv.v.m.N mo Esuuommm HE: was .mm wusoflm o.« od . 4L lilL; J ILonwJ . _ fl .q. - -~.- _._~___ . ..—... -—.———-—-~-- - ——.7- _—._-—_..— -— —..-. u ‘m;97_‘. '_' . -- ‘ua-‘o- J..- JL... 1 .-- —-o———*_ 2 ox\ XI mcoflwlm.HnmcmucmmoaU>oaxnww5muuwuiw.v.m.N mo Esuuummm HES mze .wm mudoflm '4-0 .a . - .3 . - . - - M. I. .-!-'h --'n -.—‘_ -~- --Q.—~ 65 - - _ *fi—‘k . . . . ‘ maoumumcmxmn Ho.H.mH 0Huxownaxon#melma>XOHU%£|aIa>npmemxmn um w v v m N mummmum ou unamppm mnu Eoum muquwE unavoum may no Esauommm was $39 .mm musmam 3+0 ‘ . 4 - 4 0.. o.u o.m 0:» q ‘ ‘ 1 ‘— 1 1 ‘ ‘ d I ‘1 ‘l‘H ‘ ‘ ‘ ‘ d 1 1 44+ —‘ 1 ‘0 1! 1 - 1 ‘ ‘ I —‘ a . _ q . J- _ . . 1 1 . W . d _ d . . . . «1 . u d . q . u 1 . 1 I“ _ — _ — _ u . 1 . — . A . q . ‘ . dll. j —- - H.371: ‘- EL 0 .‘I . . I “W --O—--—-—-—o --ou .. \ 66 _. o... co...- w 67 ff 4 m -204 f m 4 m; 104 m Figure 29. The nmr spectrum of 2, 2 H4 4, 6, 6-hexamethy1-—1,3-dihydroxy- bicyclo[;3. 1. O] hexane in fluorosulfonic acid at various temperatures MICHIGAN STAIE UNIVERSITY LIBRARIES IIIIIIIIIIIHIIIIIIIIHIIIIIIIIIIIIIIIIIH 31293‘03115 6178