ABSTRACT SYNTHETIC APPROACHES TO THE TRICYCLO[7.2.l.01’5]DODECANE PING SYSTEM By John N. Patterson, Jr. The ascending importance of the tricyclo[6.2.l.01’5]undecane and the spiro[4.5]decane ring systems in sesquiterpene chemistry motivated this study in which tricyclo[7.2.l.01’5]dodec-6,7-ene-2,8- dione (l) is proposed as a potential intermediate for the synthesis of several naturally occurring derivatives of these two ring systems. Three synthetic routes to the tricyclo[7.2.l.01a5]dodecanes were investigated. The first involved the preparation of decalone g, which could undergo an intramolecular alkylation to give the desired skeleton. The second approach was similar in that an internal alkylation of decalone g was proposed for the formation of the tricyclo- dodecane. Attempts to synthesize decalones g and g were however unsuccessful. R0 John w. Patterson, Jr. In the third route to the tricyclododecane skeleton, a bis- alkylation of cyclohexane-l,3-dione was effected with bromo-3-methylene- 4-oxopentane (3) giving 2-acetylspiro[4.5]deca—6,lO-dione (g). The The relatively straightforward preparation of bromo enone g and the conjecture that only an aldol condensation and dehydration is required to convert g into 1, makes this route to the tricyclo[7.2.l.01:5]— dodecane l the most promising of those investigated. 8...: SYNTHETIC APPROACHES TO THE TRICYCL0[7.2.l.O]’6]DODECANE PING SYSTEM By John N. Patterson, Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1970 6'92 ‘337 7- [‘70 To Carol ii ACKNOWLEDGMENTS The author expresses his gratitude to Professor William H. Reusch far his guidance and assistance throughout the course of this investigation and for arranging financial support from April 1967 to September 1968. Appreciation is extended to the National Institutes of Health fOr a ore-doctoral fellowship from September 1968 thru December 1969. iii TABLE OF CONTENTS PAGE INTRODUCTION 2 RESULTS AND DISCUSSION l0 EXPERIMENTAL 32 General 32 Methoxy- -2- (3'-oxobutyl)cyclohexen- 3- -one (4 ) and 3- methoxy- 3- methylbicyclo[4. 4. D]- 2- oxadec- l ,C- en- -7- -one (g4) 32 2-acetyl diethyl succinate diethyl ketal (4Q) 33 2-acetylbutane-l,4-diol diethyl ketal (31, R=H) 33 2-acetylbutane-l,4-diol diacetate (4g, P=OCDCH3) 34 3-methylene-4-oxopentanol acetate (42, R=OCDCH3) 35 Methoxy- 2- (2'-ethylacetoxy- 3'-oxobutyl)cyclohexen- 3- -one (S3) 36 3-methyltricyclo[7.4.O.D3’7]-2,4-dioxatridec-l,9- en~lD-one (gfi) 36 3-methylene-4-oxopentanol pivalate (49, R=0COC(CH3)3) 37 2-acetylbutane-l,4-diol dibenzyl ether (fig, R=OCH2C6H5) 38 3-methylene—4-ox0pentanol benzvl ether (42, R=OCH2C5H5) 39 2-(2'-cyclohexane-l:3'-dione)acetic acid ethyl ester (g2) 40 2-(2'-cyclohexane-l:3'-dione-2'-(3"-oxobutyl))acetic acid ethyl ester (Qg) 41 2-(B-carboxyethyl)-6—carbethoxymethyl-3-methylcyclo- hex-Z-enone (g3, R=H) 4l 4a- -carbethoxymethyl- -4 ,4a, 7, 8- -tetrahydronaphthalene- 2 ,5(3l' ,6H)- dione (Q1) 42 iv TABLE OF CONTENTS - Continued 4aB-carboxymethyl-4,4a,5,6,7,8-hexahydronaphth-58- ol-2(3H)-one lactone (pg) 4aB-carboxymethyl-2-(l:3'-dioxolane)-l,4,4a,5,6,7- hexahydronaphth-SB-ol lactone (pg) 2-(l;3'-dioxolane)-l,4,4a,5,6,7-hexahydro-4a8- (2"-hydroxyethyl)naphth-SB-ol (g2, R=H) 2-(l;3'-dioxolane)-l,4,4a,5,6,7-hexahydro-4a8- (2"-hydroxyethyl)naphth-SB-ol mono p-nitro- benzoate (£2, P=CDC6HHN02) 4aB-carboxymethyl-3,4,4a,5,6,7,8,8a-octahydro- naphth-SB-ol-2(lH)-one lactone (Z3) 3-hydroxymethylpent-3-enol (Bl) Chloro-3-chloromethylpent-3-ene(g3) Bromo-3-methylene-4-oxopentane (49, R=Br) 2-acetylspiro[4.5]deca-6,lD-dione (ZS) FIGURES REFERENCES 43 44 44 4E 47 48 48 so 89 LIST OF FIGURES FIGURE l. 10. ll. 12. l3. l4. Infrared spectrum of methoxy-Z-(3'-oxobutyl)cyclo- hexen-3-one (4%) Infrared Spectrum of 3-methoxy-3-methylbicyclo- [4.4.0]-2-oxadec-l,6-en-7-one (4Q) Infrared Spectrum of 2-acetylbutane-l,4-dioI diacetate (QQ, R=DC0CH3) Infrared spectrum of 3-methylene-4~oxopentanol acetate (49, R-DCDCH3) Infrared spectrum of methoxy-Z-(Z'-ethylacetoxy~ 3'-oxobutyl)cyclohexen-3-one (fig) Infrared spectrum of 3-methyltric clo[7.4.0.03’7]- 2,4-dioxatridec-l,9-en-lO-one (SQ Infrared Spectrum of 2-acetylbutane-l,4-diol dipivalate (fig, R-DCOC(CH3)3) Infrared spectrum of 3-methylene-4-oxopentanol pivalate (42, R=OCOC(CH3)3) Infrared spectrum of 2-acetylbutane-l,4-diol dibenzyl ether (QQ, R=DCH2CEH5) Infrared spectrum of 3-methylene-4-ox0pentanol benzyl ether (42, R=OCH2C6H5) Infrared spectrum of 2-[2'-cyclohexane-l'3Ldione- 2-(3"-oxobutyl)]acetic acid ethyl ester (QQ) Infrared spectrum of 2-(3-carboxy ethyl)-6- carbethoxymethyl-3-methyl cyclohex-Z-enone (g3, R=H) Infrared spectrum of 4a-Carbethoxymethyl-4,4a,7,8- tetrahydronaphthalene-Z,5(3H,6H)-dione (Q1) Infrared spectrum of 4aB—carboxymethyl-4,4a,5,6,7,8- hexahydronaphth-SB-ol-2(3H)-one lactone (§§) vi Page 51 U"! 1".) 52 53 53 54 54 55 55 56 56 57 57 LIST OF FIGURES - Continued FIGURE l5. 16. I7. 22. 23. 24. 25. 26. 27. 28. 29. 30. Infrared spectrum of 4aB-carboxymethy-2-(l',3'- dioxolane)-l,4,4a,5,6,7-hexahydronaphth-56-ol lactone (pg) Infrared spectrum of 2-(l',3'-dioxolane)-l,4,4a,5,6,7- hexahydro-4a-(2"-hydroxyethyl)naphth-SB-ol (Q2. R'H) Infrared Spectrum of 2-(l',3'-dioxolane)-l,4,4a,5,6,7- hexahydro-4a8-(2"-hydroxyethyl)naphth-SB-ol mono p-nitrobenzoate (Q2, R=COC6H4N02) Infrared spectrum of 4a8-carboxymethyl-3,4,4a,5,6,7,8,8a- octahydronaphth-SB-ol-2(lH)-one lactone (13) Infrared Spectrum of 3-hydroxymethylpent-3-enol (§l) Infrared spectrum of chloro-3-chloromethylpent-3-ene (g2) Infrared Spectrum of bromo-3-methylene-4-ox0pentane ($9, R=Br Infrared spectrum of 2-acetylspiro[4.5]deca-6,lD- dione (Z§) Nmr spectrum of methoxy-Z-(3'-oxobutyl)cyclohexen- 3-one (42) (CDCl3) Nmr spectrum of methoxy-Z-(3'-oxobutyl)cyclohexen- 3-one (42) (pyridine) Nmr spectrum of 3-methoxy-3-methylbicyclo[4.4.0]- 2-oxadec-l,6-en-7-one (4g) Nmr spectrum of 2-acetylbutane-l,4-diol diacetate (4g, R=DCOCH3) Nmr spectrum of 3-methylene-4-oxopentanol acetate (42, R=OCDCH3) Nmr Spectrum of methoxy-Z-(Z'-ethylacetoxy-3'- oxobutyl)-cyclohexen-3-one (fig) Nmr spectrum of 3-methyltricyclo[7.4.0.03’7]-2,4- dioxatridec-l,9-en-TD-one (§§) . Nmr spectrum of 2-acetylbutane-l,4-diol dipivalate (gg, R=0C0C(CH3)3) vii Page 58 58 U1 ‘0 60 60 61 El 62 63 64 65 66 67 69 LIST OF FIGURES - Continued FIGURE 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. Nmr spectrum of 3-methylen-4-oxopentanol pivalate (42, R=DCOC(CH3)3) Nmr spectrum of 3-methylene-4-oxopentanol benzyl ether (49, R=OCH2C6H5) Nmr Spectrum of 2-[2'-cyclohexane-l:31dione-2-(3"- oxobutyl)]acetic acid ethyl ester (6Q) Nmr spectrum of 2-(B-carboxy ethyl)-6-carbethoxymethyl- 3-methyl cyclohex-Z-enone (£3, R=H) Nmr spectrum of 4a-carbethoxymethyl-4,4a,7,8—tetra- hydronaphthalene-2,5(3H,6H)-dione (g1) Nmr spectrum of 4aB-cartoxymethyl-4,4a,5,6,7,8-hexa- hydronaphth-Sp-ol-2(3H)-one lactone (Qg) Nmr spectrum of 4a8carboxymethyl-2-(l',3'-dioxolane)- l,4,4a,5,6,7-hexahydronaphth-58-ol lactone (QC) Nmr spectrum of 2-(l',3'-dioxolane)-l,4,4a,5,6,7- hexahydro-4a6-(2"-hydroxyethyl)naphth-SB-ol (£2. R=H) Nmr spectrum of 2-(l',3'—dioxolane)-l,4,4a,5,C,7- hexahydro-4a8-(2"-hydroxyethyl)naphth-SB-ol mono p-nitrobenzoate (Q2, R=COC6H4H02) Nmr spectrum of 4a8-carbox thyl-3,4,4a,5,6,7,8,8a- octahydronaphth-SB-ol-2(lH)T:ne lactone (73) Nmr spectrum of 3-hydroxymethylpent-3-enol (Bl) Nmr spectrum of chloro-3-chloromethylpent-B-ene (gg) Nmr spectrum of bromo-B-methylene-4-oxopentane (49, R=Br) Nmr spectrum of 2-acetylspiro[4.5]deca-6,lD-dione (lg) Mass spectrum of methoxy-Z-(3'-oxobutyl)cyclohexen- 3-one (42) Mass spectrum of 3-methox -3-methylbicyclo[4.4.0]- 2-oxadec-l,6-en-7-one (4g Mass spectrum of 3-methyltricyclo[7.4.D.D3’7]- 2,4-dioxatridec-l,9-en-lO-one (SQ) viii Page 70 7l 72 73 74 76 77 LIST OF FIGURES - Continued Page FIGURE 48. Mass spectrum of bromo-3-methylene-4-oxopentane ($2. R=Br) 87 49. Mass spectrum of 2-acetylspiro[4.5]deca-6,lD-dione (lg) 88 ix SYNTHETIC APPROACHES TO THE TRICYCLO[7.2.l.O]’GJDDDECANE RING SYSTEH INTRODUCTION This investigation concerns the preparation of a suitably function- alized derivative of tricyclo[7.2.l.01’6]dodecane (l) as a potential inter- mediate in the synthesis of two classes of sesquiterpenes: the spiro[4.5] decanes (g) and the tricyclo[6.2.l.01’SJundecanes (2). However, before discussing the synthetic methods, it is useful to survey some recent developments in the sesquiterpene field. These remarks will illustrate the type of substitution and stereochemistry present in the naturally occurring derivatives of ring systems 2 and 3. 0' 96 During the past several years 2‘the spiro[4.5]decane2 ring system has assumed a prominent place in sesquiterpene chemistry. The first example of this ring system to be reported was acorone (4)! H... CH3 84> CH3)\ CH3 In 1965 5.0. Bhattacharyya reported2 the isolation of agarospirol and assigned structure S to this compound. Classical degradation and spectrosc0pic analysis were used to establish the gross structure of agarospirol. Independent synthesis of the spiro ketone g and comparison of this with the same compound derived from agarOSpirol revealed only 2 3 minor differences attributable to the stereochemical nonhomogenity of the synthetic material. However the stereochemistry shown in structure S was deduced on tenuous arguments concerning the nmr spectra of agarospirol CH, 0' ° CH, 8 and the reactivity of its hydroxyl group. In addition to these two new sesquiterpenes, Marshall and co-workers have demonstrated that B—vetivone (1) possesses the spiro[4.5]decane skeleton. This was accomplished by first synthesizing3 the three epimeric 6,l0-dimethyl-§_i_s__-decahydroazulene-8-ones (Q) and observing each to be different from desisopropylidenedihydro-B-vetivone. Revision of the azu- lenic structure 9" previously assigned to B-vetivone is therefore required. The original degradative studies on g-vetivone were also consistent with structure 1, proposed by Marshall, and this structure was confirmed by correlation of g-vetivone with a synthetic spiro[4.5]decane of known stereo- Chemistry4. A total synthesis of gyvetivone has been reporteds, beginning with the known spiro[4.5]decadienone 13. Several rather obvious steps trans- formed this into compound ll (mixture of C2 epimers). One of the epimers is hinesol and the other may have been agarOSpirol. Marshall and his coaworkers were, however, unable to determine the stereochemistry at C2. This mixture was converted to B-vetivone by a known procedure, involving oxidation of the allylic methylene group and dehydration of the alcohol. N‘ CH 3 12 A significant feature of this work on B-vetivone is that it requires revision of the structures of all the sesquiterpenes correlated to isovetivane (lg). Among these are hinesol (L2)6, bicyclovetivenol7, the isovetivenenes (lg)8, and acorenone (l§)9. As mentioned above, during the course of the B-vetivone synthesis a stereoisomeric mixture of alcohols ll was obtained. One of these must 5 have been hinesol. Very recently Marshall and Brady have synthesized hinesol (13) by an unambiguous routeio. Reaction of the known tricyclic dienone (lg) with dimethyl copper lithium introduced the C2 methyl group, but the desired epimer (ll) constituted only 20 percent of the mixture. In several steps the carbonyl was transposed to C8 and the enone was con- verted to diol 12° The secondary hydroxyl was converted to a methane sulfonyl ester and fragmentation gave the unsaturated ketone 28 which was converted to hinesol. X-ray crystallographic analysis was used to estab- lish the stereochemistry of L1, and that in conjunction with the synthetic route requires hinesol to have configuration 13. Although Marshall‘s work on hinesol was successful, it has several imperfections: introduction of the C2 methyl in L1 gave predominantly the wrong isomer, and the ketone carbonyl was in the wrong position necessitating a tedious transposition to C8. Scheme 1 outlines a possible route to agaroSpirol (and hinesol) employing a suitably functionalized tricyclo[7.2.l.0]’6]dodecane intermediate. Scheme 1: The essential features of this approach are: introduction of a thioketal a to the carbonyl of’gg, Haller-Bauer type cleavage yielding 24, desulfurization of the resulting thioacetal and finally introduction of the two remaining methyl groups. 'The second class of sesquiterpenes for which the tricyclo[7.2.l.01’6]- dodecanes serve as potential precursors possess the tricyclo[6.2.l.01’5]- undecane ring system. The first three members of this group have recently been reported: zizanoic acid (g§)1]zizaene (22)]2 and khusimol (QQ)11 Scheme 2 outlines a short reaction sequence by which g§ could be transformed into zizanoic acid. Bromination a to the carbonyl followed by Favorskii rearrangemnt should give the carboxylic acid 2Q. Transposition of the carbonyl and a-dimethylation generates £1 and introduction of the methylene group via a Hittig-type reaction fOrms zizanoic acid. Scheme 2: 0 H0 —-|1 -- O —> H OR ‘\.OR -——4- gfi R=C02H g1 g2 R=CH20H 3Q R=CH3 Having established the potential synthetic usefulness of 2,8-difunc- tionalized tricyclo[7.2.l.01’6 ]dodecanes , let us now consider some synthetic approaches to these intermediates. A useful technique in planning the synthesis of a polycyclic compound is to consider the structures formed by breaking bonds in such a way as to reduce the number of rings. Hapefully these structures with fewer rings can be synthesized in such a manner as to permit subsequent closure of the ruptured bond. This technique leads to many different approaches to the tricyclo[7.2.l.01’6 ]dodecane ring system. For example, cleavage of the C1-C1], Cg-C10 or C6-C7 bonds generates the bicyclic systems 31, 32 and 33. In order to reform the broken bonds, it will be necessary to introduce appropriate sites of reactivity (i.e. functional groups) into these bicyclic ring systems. Two general and widely used methods of forming carbon-carbon bonds are the alkylation of ambident enolate ions and aldol type condensations. If C1" of structure 31 is bonded to a 8 X HZCI CH, x Y “Y 00 Y GOT 3,1 4?} €53 nucleophilic leaving group and the group X is a carbonyl group then an intramolecular alkylation could be used to close the broken bond as shown in equation l. Synthesis of intermediate 34 having the C6 hydrogen gj§_ to the C9 side chain insures that the stereochemistry of the alkylation product will be that shown in structure 3g. Similarly,the ruptured bond of 32 can conceivably be closed via alkylation. The necessary intermediate in this case is compound gg,which could cyclize to either CZ or 38. A considerable body of literature suggests that glgfdecalones such as 3Q preferably enolize to C9 rather than C7, thus 31 is a reasonable product from internal alkylation. In the case of the Spirodecane 33 an aldol condensation could provide a convenient means of closing the final bond. Thus, an inter- mediate of type 32 chould give the desired tricyclic product. 3% 49 Synthetic routes to the intermediates 34, 3g and 32 have been investigated in detail and are the subject of this dissertation. RESULTS AND DISCUSSION 1,6 The first approach to the tricyclo[7.2.l.0 ]dodecane ring system ‘3 depicted in Scheme 3. According is based on the work of I.N. Nazarov to the Russian workers, condensation of cyclohexane-l,3-dione with methyl vinyl ketone in the presence of potassium Carbonate gave the triketone 41. Since the B-diketone moiety existed primarily in the enol form, reaction with diazomethane produced the enol ether 42. On treatment with potassium t-butoxide this enol ether underwent an intramolecular aJdol condensation, followed.by dehydration generating the methoxy dienone ~§° Scheme 3: 0 0 0CH3 0 CH3 CH3 -—§ 0 0 $1 4?. .CH3 ——* ‘9 O 43 "4% A repetition of Nazarov's work gave in one experiment an isomeric llroduct, hemiketal 44. This hemiketal was acidic, i.e. soluble in aqueous Sodium bicarbonate, and reacted with diazomethane to give the bicyclic ketal 42. The structure of the bicyclic ketal followed from its Spectral lO 11 properties. The nmr spectrum exhibits two Singlet methyl groups at r 6.77 and 8.56; the ultraviolet spectrum has a maximum at 257 mu (e=15,400); 1 and the infrared Spectrum Shows strong absorption at 1650 and 1625 cm- , O ' O I'llII llllll 0 OH 0 OCH, éfi CH3 fig cH3 indicative of an ae-unsaturated carbonyl group, but lacks any absorption attributable to a methyl ketone. In addition, the mass spectrum of 46 has ions at m/e 196, 124 and 72. These are consistent with the fragment- ation of the parent ion as Shown in Scheme 4. SCheme 4: CH3 OCH3 ° X +0 CH2 2 OCH, . CH, O 5. D 0 H2 OCH, H2 ’,/f:i3 + . + k ' 0 CH3 CH2 0 CH3 CH7 m/e 124 m/e 72 In contrast to these observations, the enol ether 42 shows a saturated methyl ketone in the infrared spectrum and has mass spectral 12 ions at m/e 196, 153 and 43, which are consistent with simple a-cleavage of the methyl ketone as shown in Scheme 5. In this case m/e 43 is the base peak, a common feature in many methyl ketones. This fragmentation Scheme 5: OCH, CH2+ , 4‘ CH3CEO ' 0CH3 01' l’,,/””)' 0 m/e 153 CH3 0 \ 0CH3 Hz' + CH3CEO+ m/e 43 0 is supported by a metastable peak at m/e 119.5, which corresponds to a m/e 196 + m/e 153 ion decomposition. The nmr spectrum of the mono- cyclic enol ether 42 showed an interesting solvent effect. In CCl4 or d6-dimethy1 sulfoxide the methylene groups in the 3-oxobuty1 side chain appeared as a four-proton singlet at r 7.67; however, in pyridine these methylene groups were Split into a complex multiplet. These two isomeric enol ethers possess chemical behavior which is in agreement with the assigned structures. Compound 42 undergoes an aldol condensation and dehydration to yield 43 on treatment with potassium t-butoxide as reported by Nazarov, whereas 46 is inert to this reagent. In general, glpc analysis of the enol ether mixture obtained by NazarOv's procedure indicated 5-10 percent of the bicyclic ether 46, l3 and the high yield of 43 observed in the first experiment could not be reproduced. The application of Nazarov's work to the synthesis of the desired tricyclododecane requires a substituted methyl vinyl ketone such as 49, R=DCOCH This has been prepared by the route shown in Scheme 6. 30 Scheme 6: H,C,O OC,Hs CH, CH3 COZCZHS C0,C,H5 0 O R R ““it CH, --%> CH, CH R 7 fig 49 ’Vb ‘4 from acetyl diethyl succinate, Reduc- Ketal diester 46 was prepared tion with lithium aluminum hydride gave the ketal diol 41, R=OH. On treatment with acetic anhydride in pyridine the diol was converted to the diacetate 41, R=DCOCH3, which could be hydrolysed to the keto diacetate 48, R=0CDCH3, by shaking a chloroform solution of the ketal with 10 percent hydrochloric acid. Distillation of this keto diacetate from triethanolamine (at 2 mm pressure) occurred with elimination of the acetoxy group a to the carbonyl function, giving the acetoxy enone m9, l4 R=OCOCH3. Without purification of the intermediates, acetyl diethyl succinate can be converted to the unsaturated ketone 42, R=OC0CH3, in 65 percent overall yield. It should be pointed out that a more obvious route to enones of type 49, namely synthesis of ketones of type 5Q followed by introduction of the £3 methylene, was also investigated. A classical method for 0 O M R R CH 2 $2 42 introduction of a methylene a to a carbonyl is the aldol condensation with formaldehyde followed by dehydration. Many examples of this type of reaction have been reported and suggest that condensation usually occurs on the methylene side of a methyl ketone. However, examination of the experimental procedures reveals the majority of them to involve commercially available ketones which reacted with only 0.1 - 0.3 equiv- alents of formaldehyde. This is necessary to avoid multiple condensations ‘5 in a detailed study of the con- but severely limits the yield. Landon densation of 2-butanone and formaldehyde feund use of 0.2 equivalents of formaldehyde resulted in the maximum yield of 2-methy1-1-butene-3-one. Another common means of introducing a methylene group is the 16, this condensation Mannich condensation. As reported by Hagermeyer proceeded at C3 of 2-butanone in good yield. Unfortunately, pyrolySis of the hydrochloride salt of the B—amino ketone under a variety of conditions gave considerable 2-hutanone in addition to the desired 15 2-methyl-l-buten—2-one. Since these two products were difficult to separate by distillation, it was concluded that the Mannich condensation would be of little use in the synthesis of 42. Several promising condensations with levulinic acid have been described. For example, the Claisen condensation between ethyl formate 17 and ethyl levulinate is reported to occur at the methylene group a to the ketone carbonyl; however, several attempts to repeat this reaction have failed. The Mannich reaction with levulinic acid gave condensation 0 0 /U\/\C02C2H5 ___§ CH3 k CO C H CH3 2 2 s OH 18 at both sides of the ketone carbonyl , the two products being separable by crystallization. In the present study several repetitions of this experiment gave primarily condensation at the methyl group. 19 by a circuitous route beginning with acetyl diethyl succinate and proceeding thru 3-acetyl butyrolactonezo. 0 o “bigoa CH3 COZCZHS -—-> CHS/U\n/\COzC2HS CH2 Br 51 ’V‘b The enone ester 51 has been prepared In view of the fact that the enones sought here are to be used in a Michael condensation requiring base catalysis, enone 1 is of questionable utility because of the possibility of isomerization to the fully conju- gated isomer. For this reason the Michael condensation of 51 with cyclohexane-l,3-dione was not investigated. The acetoxy enone 42, R=OCOCH3, reacted with cyclohexane-l,3-dione 16 and a catalytic amount of potassium carbonate to give the mono Michael addition product 52 in higher yield than the corresponding product from methyl vinyl ketone. This can be attributed to the steric hindrance of the B-acetoxyethyl side chain of 52 which inhibits the formation of the bis-adduct. As in the model system, treatment of the Michael adduct 52 with diazomethane gave the enol ether 53. However aldol condensation dehydration of 53 with potassium t-butoxide led to approximately equal amounts of two products. Scheme 7: O O CH, -——4> OCOCH, 0 CE CH, O O I 1.» SS c: OCOCH3 OCH, 0 0N ‘T Q o 0 0 9’5 £32 5:3 One of the products (54) is analogous to that obtained in the model system. The other product was assigned structure 55 on the basis of its spectral properties and elemental analysis. The infrared Spectrum showed that no hydroxyl was present and absorptions at 1665 and 1635 cm"1 were characteristic of an enone system. This chromophore was further identi- fied by its absorption maximum at 259 mu (e=16,800) in the ultraviolet spectrum. A singlet methyl signal at r 1.88 and a two-proton multiplet 17 at T 5.9-6.3 in the nmr Spectrum (the remainder of the protons being an unresolved multiplet at r 7.5-8.2) agree with structure 55. A strong parent ion appears in the mass Spectrum at m/e 208 and two rearrangement ions are observed at m/e 124 and 84 (base peak). The latter two ions could arise via the same type of fragmentation observed for enol ether 45, shown in Scheme 4. A reasonable mechanism for the formation of 55 is illustrated in Scheme 8. This mechanism involves solvolysis of the acetate and cycliz- Scheme 8: OCH3 O O CH, ___) a 0 09 ‘ 0 O OCH3 OH, -—-—> a ‘— <-— 3% ation of an intermediate hemiketal. Indeed, tricyclic ether 55 could be prepared in good yield by basic hydrolysis of the acetate of 52 to give an aqueous solution of anion 56, which on acidification gave 55. Scheme 9 illustrates a a reasonable mechanism for the formation of 55 under acid catalysis. 18 Scheme 9: G o “3 ~—» w 2 O 0” OH O —> ‘_. -——> 4 0H CH, If this mechanism for the formation of 55 is valid, then this 0 I O undesireable side reaction can be blocked by replacement of the acetate with a protecting group which is inert to nucleophilic bases. To this end the pivalic ester 42, R=DC0C(CH3)3, was prepared by a procedure completely analogous to that for preparation of acetate 42, R=0CDCH3. This enone was used to synthesize the pivalic ester anologue of,§3. Unfortunately, the aldol condensation again gave a mixture of 54 and 55. Ethers are known to be resistent to basic reaction conditions and the benzyl ether protecting group was next investigated. Ketal diol 41, R=OH, reacted with benzyl chloride in the presence of sodium hydride in dimethyl sulfoxide to give the ketal dibenzyl ether NZ, R=0CH CHHS' The keto dibenzyl ether 48, R=OCH C H5, was then obtained 2 2 6 by mild acidic hydrolysis of the diethyl ketal. Unfortunately compound 48,R=0CH2C6H5, failed to eliminate the benzyloxy group n to the carbonyl grouo to yield 42, R=0CH2C6H5, on distillation from triethanolamine. An attempt to generate the benzyloxy enone $3, R=OCH206H5, jg_situ 19 and to trap it by Michael addition to cyclohexane-1,3—dione was then madezi In this experiment a mixture of the dibenzyl ether 48, R=DCH2CCH5, and cyclohexane-1,3-dione were treated with potassium t-butoxide, however the dibenzyloxy ketone merely decomposed giving an intractable black oil as the neutral product along with a nearly complete recovery of the cyclohexane-l,3-dione as the sole acidic product. In one experiment, distillation of the ketal dibenzyl ether NZ, R=OCH2C6H5, resulted in elimination of ethanol, producing the enol ether 56. Mild acidic hydrolysis of the enol ether function resulted in con- comitant elimination of benzyl alcohol giving the enone fig, P=0CH2CCH5 OC,HS CH 1’, OCH,C,HS 0 u, R=CCH2C6H5 —-) 2 -_’ CHZCGHS OCH,C,HS C“3 CH, pg 32, R=OCH,C,H5 This later success was obtained only recently and other approaches to the tricyclododecanes have been investigated. The second approach to the tricyclo[7.2.l.01’CJdodecane ring system 22 of 4a-methyl-4,4a,7,8- to be described here is based on the synthesis tetrahydronaphthalene-2,5(3H§H)-dione (57) and its selective reduction to 4aB-methy1-4,4a,5,6,7,8-hexahydronaphth-58-ol-2(3H)-one (58)23. 0 0 CH, CH, O O 20 0 H0 CH3 H3 _§ ._§ 0 0 £1 48 This reaction sequence was applied to the synthesis of decalin 36 by effecting a base catalysed Michael addition of the readily avail- able 2-(2'-cyclohexane-l',3'-dinne)-acetic acid ethyl ester (52)24 to methyl vinyl ketone. 0 CO,C,H5 0 CO,C,H5 ———a- 0 o 0 £2 89 £1 CO,C,Hs The desired intramolecular aldol condensation-dehydration of the adduct 60 proved to be elusive, Since 60 gave a variety of products depending on the catalyst employed. Thus, a catalytic amount of pyrrolidine in refluxing benzene (the conditions used in the prepar- ation of 51) gave only recovered starting material, while p-toluene- sulfonic acid in refluxing benzene gave enone 63, R=H. The bicyclo- [3.3.1]nonane ketol 62 is thought to be a reasonable intermediate in this latter reaction, as acid catalysed aldol condensations of 2-(3'-oxobutyl)cyclohexanones frequently yield 2-hydroxy-9—oxo-2- methylbicyclo[3.3.l]nonanes as the isolated productzs. 2'1 The desired aldol condensation and dehydration to the enedione 61 was ultimately achieved in 60 percent yield by using 1.1 equivalents of a 1:1 mixture of pyrrolidine and acetic acid at room temperature. Due to the instablity of 61 toward oxygen, the structure assignment was based on its spectroscopic properies and the characteristic reactions discussed below. Reduction with a 10 percent excess of sodium borohydride converted enedione 61 to the enone lactone 65. The gi§_relationship of the hydroxyl and methylcarbethoxy groups in structure Qg_was inferred from the spontaneous lactonization of this intermediate hydroxy ester 64. 0 .1 22 The enone lactone 66 was transformed in high yield to an ethylene ketal derivative 66. The migration of the double bond to the 8,8a-position is normal and is supported by spectroscopic evidence. In the nmr spectrum of the enone 66 the vinyl proton is a singlet with a half-height width of 3 cps, whereas in the ethylene ketal the vinyl proton is a very broad signal with a width of 10 cps. This large increase in the spin-Spin coupling suggests that the double bond moves to the A“.“3 position. In addition the infrared stretching frequency of the lactone carbonyl in the ketal is 25 cm"1 higher than in the enone. Examination of molecular models reveals that shifting the double bond increases the strain in the five-membered lactone ring and therefbre increases the carbonyl stretching frequency. 65 —> «K. 8?. In order fur lactone'66 to be a useful intermediate in this synthesis, the lactone ring must be opened in such a way that the terminal carbon atoms can be chemically distinguished. An ideal solution would leave the terminal carbons in different oxidation states, and to this end two methods were investigated. The first was founded on the report26 that disiamylborane reduces y-lactones to hydroxy aldehydes. However, when ketal lactone 66 was reduced with this reagent, the intermediate cyclic hemiacetal could not be induced to open to the hydroxy aldehyde 61. 23 OH H.. H‘.‘ CHO w 09666 .2 .. .52 In the second approach, oxidation of the ketal lactone with Sarett's reagent did not generate the keto acid 66, but gave instead recovered starting material at room temperature and intractable tars at higher temperatures. 0 CO,H O o) 8% Although these attempts at selective lactone opening were unsuc- 88 '€><‘l/\CO,C,H5 —-—-4> CH, 0” CO,C,H5 OH 12 RR 27 Because the ketal function in diols 6Z and 66 proved to be so troublesome, the synthesis of dihalides not having this functiOn was considered. To this end, compound 66 was prepared from ethylidine diethyl succinate28 (66) by lithium aluminum hydride reduction to diol OH 1 /__<—CO,C,H5 _ _ — CH, C COzcsz CH, OH CH, 1 81 AZ 82 66 followed by reaction with thionyl chloride under very mild conditions. The unstable dichloride 66 was obtained in 45 percent yield after distillation at reduced pressure and could be stored at -10° for several days. Another mild reagent for the conversion of primary alcohols to chlorides, tri-n-octyl phOSphine in carbontetrachloridezg, gave a slightly lower yield of 66 from diol 66 than was obtained with thionyl chloride. The dichloride (66) reacted with ethyl acetoacetate in the presence of two equivalents of sodium ethoxide in ethanol to give the desired bis-alkylation product 66. Although 66 was a mixture of 0 0 CH, . CH CH 3 T 6% -——> 3 fit double bond isomers and consequently failed to give a solid deriv- ative suitable fer elemental analysis, its Spectral properties were in 28 good agreement with the assigned structure. When this bis-alkylation procedure was applied to cyclohexane- l,3-dione, the keto ester 66 was obtained as a mixture of isomers. This difficulty was not unexpected in view of the known lability of . (— 'l 0 C 3 8% -—-—> -—-—~> H C O C ‘\~ CH3 522 _ ° _ O 88 ER 2,2-disubstituted cyclohexane-l,3-diones toward nucleOphilic cleavage. Unfortunately, the dichloride proved unreactive toward the conjugate base of cyclohexane-l,3-dione in non-nucleophilic solvents. A successful bis-alkylation of cyclohexane-1,3-dione was finally accomplished with the bromo enone 66, R=Br, which was prepared from diol 61, R=OH. Thus, treatment of 6Z, R=OH, with p-toluenesulfonyl H,C,O _ OC,HS OH 050 C H CH CH, ___) CH3 2 6 u 3 0H OSO,C,H,CH, 41 4% 0 R —-’ CH3 CH, 42 chloride in pyridine fellowed by an acidic workup gave the unstable 29 keto ditosylate 66. R=0502C6H4CH3, Triethylamine at room temperature catalysed the elimination of the tosyloxy group which is 8 to the carbonyl function, yielding the enone tosylate 66, R=OSOZC6H4CH3, and the tosylate group in 66 was replaced with a bromine atom by reaction with anhydrous lithium bromide in acetone30. By this sequence the bromo enone 66, R=Br, was obtained in 29 percent overall yield from diol 6Z. Although the bromo enone was unstable , it could be distilled at reduced pressure and its spectral properties were in good agree- ment with structure 66, R=8r. In particular the infrared spectrum revealed an aB-unsaturated ketone, and the nmr Spectrum yielded to simple first order analysis. The enone 66, R=Br, reacted with one equivalent of cyclohexane- l,3-dione and 1.1 equivalents of sodium hydride in 1,2-dimethoxyethane giving the spiro triketone Z6. The nmr Spectrum of Z6 consists of a singlet at r 7.30 for the methyl ketone and a four-proton triplet at r 7.36 due to the two nearly equivalent methylene groups a to the carbonyl functions on the six-membered ring. The mass spectrum shows a molecular ion at m/e 208 and fragment ions at m/e 193 and 165 corresponding to loss of CH3 and COCH3 via a-cleavage of the methyl ketone. The infrared spectrum has a broad carbonyl absorption between 1715 and 1700 cm'l. CH3 + Br -—-> CH3 0 0 30 The more direct route to lg, namely using one of the three inter- mediate tosylates in the preparation of the brome enone 33 as a bis- alkylation reagent, was not attempted because sulfonic acid esters are known to give exclusive O-alkylation of cyclohexane-l,3-dione31. A minor product of this reaction between $2, R=Br, and the conju- gate base of cyclohexane-l,3-dione in dimethoxyethane, and the major product when the reaction was conducted in dioxane was the tricyclic ketal gg. A plausible mechanism for the formation of QQ is shown in Scheme 10. In view of the previous study of the Michael addition Scheme 10: 0 Br -—-). 0H cu, 32 it of the conjugate base of cyclohexane-l,3-dione to methyl vinyl ketone, which showed that the hemiketal 5% was soluble in aqueous sodium bicarbonate, anion Q1 appears to be a reasonable intermediate in this transformation. The relatively straightforward preparation of bromo enone $2, R=Br, and the conjucture that only an aldol condensation and dehydration is required to convert ZQ into the desired tricyclododecane 31 1Q, makes this synthetic route the most promising of those investi- gated. EXPEPIMENTAL General. Melting points were taken in capillary tubes on the Hoover Thomas apparatus. Infrared spectra were recorded on a Perkin-Elmer 327 B spectrophotometer in carbon tetrachloride solution, with the exception of Figure 16. Nuclear magnetic reasonance spectra were taken on a Varian P~60 spectrometer. Tetramethylsilane was used as an internal standard in all cases. The ultraviolet Spectra were recorded on a Unicam SP 800 spectrophotometer. Pn Hitachi RM-EO Spectrometer was used to obtain the mass spectra. Absolute ethanol refers to commercial absolute ethanol further purified by distillation from magnesium ethoxide. 3—methoxy-2—(3'-oxobutyl)cyclohex~2—enone (£2) and 3-methoxy-3-methyl— bicyclo[4.4.0]s2-oxadec-l,S-en-7-one (£5), repetition of hazarov's procedure13 for the preparation of 4; gave a mixture containing 8“- 90 percent of fig and 10-15 percent of $5. A pure sample of each was obtained by preparative glpc using a 6'xl/4”, 4% QF-l column (175°). The infrared spectra are in Figures 1 and 2 (p 51), the nmr spectra in Figure 23-25 (pp 62-4) and the mass spectra in Figures 45 and 46 (pp 84, 85). As previously discussed, these spectra are in good agreement with the assigned structures. 33 Acetyl diethyl succinate diethyl ketal (46). This compound was prepared from the readily available acetyl diethyl succinate by the method of E. C. Kornfield14 using the modified workup described below. A solution of 90 g of acetyl diethyl succinate, 85 g of triethyl- orthoformate, 25 g of absolute ethanol and 10 drops of concentrated sulfuric acid was allowed to stand at room temperature for three days in a closed flask. The reaction mixture was cooled in an icebath and 5 ml of triethanolamine and 300 m1 of methylene chloride were added. The organic phase was separated and washed twice with 100 m1 of water and dried over sodium sulfate. Removal of the solvent under reduced pressure gave ketal Ag of sufficient purity for use in the next reaction, a glpc analysis (10'x1/8", 20% SE-30 column at 185°) indicated this crude product to be at least 90 percent ketal fig. 2-acety1butane-1,4-diol diethyl ketal (47, R=0H). To a 2 1, three- necked flask equiped with a Hirshberg stirrer, calcium sulfate drying tube, condenser, addition funnel and heating mantle was added 300 m1 of dry ether, 1200 ml of dry tetrahydrofuran and 22 g of lithium aluminum hydride. The crude ketal diester prepared from 90 g of acetyl diethyl succinate was dissolved in 75 m1 of tetrahydrofuran and added drapwise over 90 minutes and the reaction mixture was then refluxed fer two days. After cooling to room temperature, 40 ml of water and 40 ml of 5 percent aqueous sodium hydroxide were slowly added with caution to the rapidly stirred mixture. The insoluble salts were filtered, the filtrate was dried over sodium sulfate and the solvent was evaporated to give 64.2 g of diol 31, R=OH. The aluminum salts were agitated with 300 m1 of methanol for one hour and filtered again. The methanol was evaporated and the residue treated with 200 m1 of 34 ether and 5 g of sodium sulfate. Filtration and evaporation of the solvent gave another 13.0 g of diol (89 percent total yield). The infrared spectrum (neat) of 41, R=OH, showed strong bands at 3600-3200, 3000-2875, 1448, 1380, 1230, 1130 and 1045 cm-]; the nmr Spectrum consisted of a two-proton singlet at r 5.10 (hydroxyl protons), an eight-proton multiplet at r 6.2-6.9 (protons a to oxygen atoms), a three-proton multiplet at 1 7.9-8.5 (aliphatic protons) and a nine- proton multiplet at r 8.7-9.1 (methyl groups). This diol could not be distilled without decomposition, but proved to be satisfactory for use in the various transformations described below. 2-acetylbutane-lJ4-diol diacetate (48, R=0C0CH3). To a cold solution of 77.2 g of diol 41, R=OH, in 400 ml of pyridine (ice bath) were added with stirring three 40 m1 portions of acetic anhydride at 10 minute intervals. The ice bath was removed and, after standing for two hours at room temperature, the pyridine was removed by evaporation at reduced pressure. The residue was dissolved in 500 m1 of chloroform and this solution was washed with 200 m1 of water, 125 m1 of cold 10 percent hydrochloric acid, Shaken for five minutes with 125 m1 of 10 percent hydrochloric acid, and finally washed with 200 m1 of saturated aqueous sodium bicarbonate. After drying with sodium sulfate, the solvent was evaporated leaving 68.8 g of keto diacetate 48, R=0COCH3 (86 percent). A pure sample of 48, R=0COCH3, was obtained for analytical purposes by glpc. The infrared spectrum (Figure 3, p 52) showed carbonyl 1 absorptions at 1745 and 1720 cm" . The nmr Spectrum (Figure 26, p 65) 35 displays a six-proton singlet at r 8.02 (acetate methyls), a three- proton singlet at r 7.85 (acetyl methyl), a one-proton pentet at r 7.15 (methine proton, J=6.5 cps), a two-proton doublet at r 5.91 (a acetoxy protons, J=6.0 cps), a two-proton triplet at r 6.04 (a acetoxy protons, J=6.0 cps) and a multiplet at r 8.2 which is par- tially obscured by the methyl groups. 3-methy1ene-4-ox0pentanol acetate (49, R=0C0CH3). A mixture of 68.8 g of keto diacetate 48, R=0C0CH3, and 9 ml of triethanolamine was Slowly distilled at 2 mm pressure thru a six inch wire gauze heated column, and the distillate (54 g) was dissolved in 300 m1 of ether and washed with excess aqueous sodium carbonate to remove any acetic acid. After evaporating the ether solvent the residue was distilled and gave 34.6 g of enone 42, R=0C0CH3, bp 84-6°/4.5 mm. This represents an overall yield of 55 percent for the five steps beginning with acetyl diethyl succinate. 32 a 2,4-dinitr0phenylhydrazone It was possible to prepare of 42, mp ll3.7-114.7°, without concomitant isomerization of the double bond to the trisubstituted position. Agal;_ Calc fer cl4Hl606N4: C, 50.00; H, 4.80; N, 16.66. Found: C, 50.08; H, 4.81; N, 16.64. The infrared Spectrum (Figure 4, p 52) Shows absorption at 1735, 1670 and 1625 cm'], as anticipated for the ester and unsaturated ketone carbonyls and the double bond. The nmr Spectrum (Figure 27, p 66) shows two three-proton singlets at r 8.10 (acetate methyl) and r 7.70 (acetyl methyl), two-proton triplets at r 7.50 (protons a to the acetoxy group, J=6.0 cps) and r 5.95 (allylic protons, J=6.0 cps) and two one-proton singlets at r 4.15 and 3.85 (vinyl protons). 36 1ethoxy-2-(2'-ethyl acetoxy-3'-oxobutyl)-cyc1ohexen-3-one (53). A solution of 8.7 g of cyclohexane-l,3-dione, 12.0 g of enone 4g, R=0C0CH3, 50 ml of methanol, 35 ml of water and 1.3 g of potassium carbonate was heated at 70-80° for 100 minutes. The reaction mixture was cooled to room temperature, poured into a slurry of 200 ml of chloroform and 200 g of ice, and treated with concentrated hydrochloric acid (added dropwise with stirring) until pH 4 was reached. The organic layer was separated and the aqueous phase was washed with 50 ml of chloro- form. The combined organic layers were dried with magnesium sulfate and the solvent was removed leaving the crude Michael adduct 42. A solution of the crude triketo ester 44 in 75 ml of ether was treated overnight with excess ethereal diazomethane at room temper- ature. Evaporation of the ether and distillation of the remaining oil thru a Short-path apparatus gave 2.74 q of recovered 44 and 1-methoxy- cyclohexen-3-one, bp 60-70°/0.07 mm and 15.12 g (67 percent yield based on enone 4g) of enol ether 44, bp 187-92°/0.07 mm. The infrared spectrum of 44 (Figure 5, p 53) shows strong bands at 1740, 1715, 1655 and 1620 cm'1 characteristic of the ester, methyl ketone and methoxycyclohexen-3-one carbonyl groups present in this molecule. The nmr Spectrum (Figure 28, p 67) consists of a partially obscured two-proton triplet at r 6.00 (protons a to the acetoxy group) and three-proton singlets at r 6.10 (methoxyl group), 7.80 (acetate methyl) and 7.95 (acetyl methyl). 3-methyltrigyclo[7.4.0.03’71-2,4-dioxatridec-l,9-en-lO-one (55). A solution of 10.0 g of acetoxy enone 4g, R=OCOCH3, 7.2 g of cyclohexane- l,3-dione, 1.2 g of potassium carbonate, 30 ml of water and 45 m1 of 37 methanol was heated for 1 hour at 80°. The reaction mixture was cooled to 50°, a solution of 7.2 g of potassium hydroxide in 20 m1 of water was added, and this mixture was stirred for 40 minutes at 50°. After cooling to room temperature, the solution was diluted with 150 ml of water, and extracted twice with 70 m1 of chloroform. The aqueous solution was then cooled in an ice bath and acidified with hydrochloric acid. Extraction of this aqueous mixture with three 75 ml portions of chloroform gave a mixture of tricyclic ketal 44 and unreacted cyclohexane-l,3-dione. This mixture was then treated with excess ethereal diazomethane in order to convert the cyclohexane-1,3-dione to its methyl enol ether and distillation of this product mixture thru a Short-path apparatus gave 8.0 g of 44 (60 percent yield based on enone 44), bp 120-25°/0.l mm as a colorless oil which solidified on standing. Recrystallization from ethanol fellowed by sublimation gave an analytical sample, mp 42-3°. 5921: Ca1c for C12H1603: C, 69.21; H, 7.74. Found: C, 69.62; H, 7.73. The infrared spectrum of 44 (Figure 6, p 53) shows strong absorption at 1665 and 1635 cm']. The nmr spectrum (Figure 29, p 68) exhibits a two-proton multiplet at r 6.00-6.30 (C5 methylene protons) and a three-proton singlet at r 8.62 (methyl protons). 3-methylene-4-oxopentanol pivalate (49,§R=OCOC(C53)3), To a cold solution of 40 g of diol 41, R=OH, in 200 ml of pyridine (ice bath) was slowly added 56 g of pivaloyl chloride. This mixture was stirred for two hours at room temperature and then worked up in the same manner as the preparation of the corresponding diacetate. Distillation thru 38 a Short vigreaux column gave 32.5 g of keto diester 42, R-OCOC(CH bp 130-32°/0.2 mm. 3)3’ d32 A 2,4-dinitrophenylhydrazone was prepare , mp 101-2° after three recrystallizations from 90 percent ethanol. Anal. Calc fer C22H3208N4: C, 54.99; H, 6.71; N, 11.66. Found: C, 55.09; H, 6.76; N, 11.76. The infrared Spectrum of 42, R=OC0C(CH3)3, (Figure 7, p 54) 1 shows a broad carbonyl absorption at 1735-15 cm' . The nmr spectrum (Figure 30, p 69) has a peak assignment similar to that discussed fer the keto diacetate 42, R=0COCH3. Distillation of 42,,R=0C0C(CH3)3, from triethanolamine at 2 mm pressure gave the enone 42, R=OC0C(CH3)3, bp 98-100°/5 mm. A 2,4-dinitro- 32, mp 119-20°. Anal. Calc for C17H2206N4: C, 53.96; H, 5.86; N, 14.81. Found: C, 53.82; H, 5.98; N, 14.79. phenylhydrazone was prepared The infrared Spectrum of 42, R=0C0C(CH3)3, (Figure 8, p 54) 1 shows the ester and enone carbonyls at 1735 and 1670 cm" and the double bond at 1625 cm']. The nmr Spectrum (Figure 31, p 70) Shows the two vinyl protons as singlets at r 3.85 and 4.17, the two methylene triplets at r 5.85 and 7.42 (J=6 cps), the acetyl methyl at r 7.70 and the t-butyl group at r 8.90. 2-acety1butane-1347diol dibenzyl ether(48, R=0CH2C5H5);_A solution of 16 g of ketal diol 42, R=OH, in 50 ml of dimethyl sulfoxide was added to 10.5 g of sodium hydride (52%, washed with pentane to remove the mineral oil) in 250 m1 of dimethyl sulfoxide under a nitrogen atmos— phere and then 35 m1 of benzyl chloride was added over a 45 minute 39 period with stirring and cooling (20° water bath). After stirring for 10 hours at room temperature the reaction mixture was poured onto 1 kg of ice and the dibenzyl ether was extracted with ether. Removal of the solvent and distillation of the residue gave 18.0 g of ketal dibenzyl ether 41, R=OCH206H5, bp 130-40°/0.02 mm (61 percent yield). A solution of 17 g of 42, R=0CH2C6H5, in 70 ml of acetone was treated with 20 m1 of 3 percent hydrochloric acid at room temperature for 20 minutes. The ketone was then extracted with ether and evapor- ation of the solvent gave 13.5 g of 42, R=OCH206H5 (95 percent). The infrared spectrum of 42 (Figure 9, p 55) displays a saturated ketone carbonyl absorption at 1710 cm']. ‘2" 3-methy1ene-4-oxopentanol benzyl ether (49, R=0CH9C£H5);_ In one experiment, distillation of the crude 2-acetylbutane-l,4-diol dibenzyl ether diethyl ketal resulted in elimination of ethanol, giving an impure product which appeared fronlthe nmr Spectrum to be the enol ether 24. A solution of 9.0 g of this crude enol ether in 100 ml of methylene chloride was shaken with 50 ml of 5 percent hydrochloric acid for 5 minutes. The organic phase was separated, washed with water, dried and after evaporation of the solvent yielded 6.13 g of fairly pure enone 42, R=0CH2C6H5. A 2,4-dinitrophenylhydrazone, mp 98-99°, was prepared32. 5231, Calc for C19H2005N4: C, 59.37; H, 5.24; N, 14.58. Found: C, 59.38; H, 5.26; N, 14.67. A sample of 42 was purified by distillation, bp 11070.04 mm, for Spectral analysis. The infrared Spectrum (Figure 10, p 55) shows absorption characteristics of an unsaturated carbonyl (1680 cm"). 40 a double bond (1625 cm-l), aromatic substitution bands (1945, 1875 and 1800 cm") and aromatic hydrogen absorptions (3100-3000 cm"). The nmr spectrum (Figure 32, p 21) is similar to that of related enones (structure 42) discussed above. 2-(2'-cyclohexane-l:3'-dione)acetic acid ethyl ester (59). The syn- thesis described by H. Stetter24 was modified to permit facile prepar- ation of large quantities of 22. Under a nitrogen atmosohere, cyclohexane-l,3-dione (110 g) was added in one portion to a solution of 23 g of sodium in 500 ml of absolute ethanol. A solution of 120 m1 of ethyl bromoacetate in 100 ml of ethanol was then added dropwise over a period of 45 minutes, and this reaction mixture was stirred at room temperature for 12 hours. The ethanol was removed by distillation (beginning at atmospheric pressure and finishing ar reduced pressure) and the resulting semi- solid was dissolved in 75 m1 of water and 500 m1 of methylene chloride. The organic phase was separated and extracted with ice cold 5N sodium hydroxide solution until the pH of the aqueous layer reached 10. The combined aqueous layers were cooled in an ice bath and neutralized by dr0pwise addition of concentrated hydrochloric acid to the vigor- ously stirred mixture. Although the alkylated diketone (22) began to precipitate during the addition of acid, a characteristic color change from orange to white occurred when neutralization was complete. After stirring the aqueous slurry an additional hour a 0° to insure complete precipitation, the product was filtered and washed with ice water. Drying the crude product overnight gave 86-110 9 (44-55 percent) of 42 sufficiently pure for use in the next reaction. 41 2-(2'-cyclohexane-l',3'-dione-2'-(3"-oxobutyl))acetic acid ethyl ester(60). 40 g of diketo ester 22 and 20 ml (15 percent excess) of freshly distilled methyl vinyl ketone were added to a solution of 0.2 g of sodium in 150 ml of absolute ethanol, and this mixture was refluxed two hours under nitrogen. Following evaporation of the ethanol and excess methyl vinyl ketone under reduced pressure, the residue was dissolved in 250 m1 of methylene chloride and washed three times with 100 ml of water. The methylene chloride solution was dried with sodium sulfate and evaporated. Crystallization of the crude solid residue from 1:1 ethyl acetate-pentane gave two crops of Michael adduct 22 totalling 39 g (73 percent). An analytical sample, mp 95.5-96.5°, was obtained by further recrystallization from ethyl acetate-pentane. 5331, Calc for C14H1805: C, 63.15; H, 6.81. Found: C, 62.87: H, 6.87. The infrared spectrum of 22 (Figure 11, p 56) shows strong absorptions at 1725 and 1695 cm“. The nmr spectrum (Figure 33, p 72) includes Signals a r 5.96 (two-proton quartet, J=7 cps, methylene protons of the ethyl group), 1 7.03 (two-proton singlet, protons a to the ester carbonyl), 1 7.90 (three-proton singlet, acetyl methyl) and 1 8.76 (three-proton triplet, J=7 cps, methyl group of the ethyl ester). 2-(B-carboxymethyl)-6-carbethoxymethy1-3-methy1cyclohex-Z-enone (63, R=H). A mixture of 3.1 g of Michael adduct 22, 0.28 g of p-toluenesulfonic acid mono hydrate and 50 ml of dry benzene was refluxed for three hours. After cooling this solution was siluted with 100 m1 of methylene chloride and washed with 20 m1 of water to remove the sulfonic acid. The acidic product, carboxylic acid 22, R=H, was then extracted by 42 Shaking with 100 m1 of 2 percent aqueous sodium carbonate. Acidifi- cation of the basic water layer with hydrochloric acid followed by extraction with 25 ml of methylene chloride, gave after evaporation of the solvent 1.6 g (55 percent)yield) of 22. The nmr Spectrum of 22 (Figure 34, p 73) shows the presence of an ethyl ester (triplet at T 8.77 and a quartet at T 5.93, J=7 cps), a vinyl methyl (singlet at T 8.05) and the absence of any vinyl protons. The infrared spectrum (Figure 12, p 56) displays absorptions at 1735, 1710, 1665 and 1625 cm" charactersitic of the ester, acid and enone carbonyl groups and the double bond. 0n treatment with excess ethereal diazomethane at room tempera- ture for 20 minutes, 22, R=H, was converted to its methyl ester, which proved to be homogeneous by glpc analysis on 20 percent SE-30 (6'x1/8" column at 240°). 4a-carbethoxymethyl-4,4a,ZJB-tetrahydronaphthalene-2g5(3H,6H)-dione (61). To an ice cold solution of 1.60 ml of glacial acetic acid and 2.2 m1 of pyrrolidine in 75 ml of benzene was added under nitrogen 6.0 g of triketo ester 22, and after stirring the reaction mixture for 30 minutes the cooling bath was removed. The reaction mixture was stirred for 10 hours at room temperature and then poured into 100 m1 of water and 100 m1 of chloroform. The organic layer was separated, washed succes- sively with 50 m1 of 5 percent hydrochloric acid and 50 ml of saturated sodium bicarbonate and dried over sodium sulfate. Removal of the solvent by evaporation at reduced pressure gave 3.90 g of 24 as a tan oil. . This enedione could be purified by preparative glpc of 4 percent 43 QF-l at 200° or by distillation a l75°/0.002 mm. The infrared spectrum (Figure 13, p 57) shows absorptions at 1735, 1715, 1675 and 1520 cm" for the ester, ketone and enone carbonyls and the double bond. The nmr Spectrum (Figure 35, p 74 ) exhibits a vinyl proton (singlet at T 4.22) and an ethyl group (quartet at T 6.96 and triplet at T 8.92, J=7 cps). The ultraviolet spectrum (methanol) contains a maximum at 243 mu (e=11,900). 4aB-carboxymethyl-4,4a,5)6,7-hexahydronaphth-58-ol-2(3H)-one lactone (65). A solution of 3.48 g of the crude enedione 24 in 20 m1 of 95 percent ethanol was cooled in an ice bath under a nitrogen atmOSphere while 0.140 g (1.1 equivalents) of sodium borohydride in 80 m1 of ethanol was added dr0pwise and with stirring over a two hour period. The reaction mixture was stirred for an additional half hour and them 1 m1 of acetic acid was added to destroy any excess hydride. The precipi- tated lactone was filtered and the filtrate concentrated for a second crop.. A total of 1.68 g (36 percent yield for the two steps from 22) of lactone 22 was obtained. Two recrystallizations from ethanol gave an analytical sample, mp 149-50°. Anal. Calc for C C, 69.89; H, 6.84. Found: C, 69.82; 12H1403‘ H, 6.90. The infrared spectrum (Figure 14, p 57 ) Shows bands at 1785, 1675 and 1625 cm']. The ultraviolet spectrum contains a maximum at 236 mu (e=12,700). The nmr Spectrum (Figure 36, p 75 ) displays a one-proton singlet at T 4.13 (vinyl proton), a one-proton multiplet at T 5.71 (C5 hydrogen) and a two-proton singlet at T 7.30 (protons a to the lactone carbonyl). 44 4a8-carboxymethyl-2-(l',3'-dioxolang)-1J4,4a,5,6,7-hexahydronaphth- 58-01 lactone (66), A solution of 1.34 g of the enone lactone (22) in 40 ml of benzene containing 1.5 ml of ethylene glycol and a few crystals of p-toluenesulfonic acid was refluxed in a Dean-Stark apparatus for 90 minutes. The reaction mixture was cooled, dissolved in 100 ml of chloroform, washed twice with 25 m1 of water, dried over sodium sulfate and concentrated at reduced pressure. An unsaturated carbonyl absorption was barely perceptible in the infrared spectrum of the crude product, and recrystallization from ethyl acetate gave 1.28 g (79 percent yield) of ketal 22. An analytical sample, mp 151- 52°, was obtained by further recrystallization from ethyl acetate. figgl, Calc for C14H1804: C, 67.18; H, 6.84. Found: C, 66.81; H, 6.90. The infrared spectrum of 22 (Figure 15, p 58 ) shows a strong absorption at 1760 cm:l for the lactone carbonyl. The nmr Spectrum (Figure 37, p 76 ) includes one-proton multiplets at T 4.4 and 5.6 (vinyl and 05 protons), a four-proton Singlet at T 6.05 (dioxolane protons) and two-proton singlets at T 7.40 and 7.68 (protons o to the carbonyl and the C1 protons). 2-(1',3'-dioxolane)-l,4,4a,5,6,7-hexahydro-4ae-(2"-hydroxyethyl)gaphth- 58-01 (69, R=H). A solution of 1.27 g of ketal lactone 22, R=H, in 25 ml of tetrahydrofuran was added dr0pwise over a 15 minute period to 0.160 g of lithium aluminum hydride and 25 m1 of ether in a three- necked flask fitted with a condenser and drying tube. The mixture was then refluxed for three hours, cooled and 1 ml of water and 1 ml of 5 percent aqueous sodium hydroxide were carefully added. The salts 45 were filtered and thoroughly washed with ether. Evaporation of the solvent gave a white solid product, the infrared spectrum of which was devoid of carbonyl absorption. Recrystallization from ethyl acetate gave 1.04 g (81 percent yield) of the diol. An analytical sample, mp 127-28°, was obtained by further recrystallization from ethyl acetate. 5231, Calc for CI4H2204: C, 66.12; H, 8.72. Found: C, 66.13; H, 8.65. The infrared spectrum (Figure 16, p 58 ) shows a strong D-H stretching absorption between 3600 and 3100 cm']. The nmr spectrum (Figure 38, p 77 ) diSplayS a broad one-proton singlet at T 4.65 (vinyl proton), a two-proton singlet at T 5.05 (hydroxyl protons), a four=proton singlet at T 6.10 (dioxolane protons), and a three-proton multiplet at t6.35 (protons a to the hydroxyls). 2-(1',3'-dioxolane)-1J4,4a,5,6,7-hexahydro-4a8-(2"-hydroxyethyl)gaphth~ 55-01 mono p-nitrpbenzoate (22,AR=CDC5H4N02). To an ice cold solution of 0.400 g of ketal diol 22, R=H, in 10 ml of pyridine was slowly added a solution of 0.340 g of p-nitrobenzoylchloride in 5 ml of chloroform. The reaction mixture was stirred in an ice bath for 1 hour followed by 3 hours at room temperature, and the poured onto ice. The product was extracted with methylene chloride and the organic phase was washed with cold 10 percent hydrochloric acid. Removal of the solvent gave an oily product which, after two recrystallizations from ethyl acetate pentane, yielded 0.301 g (47 percent) of a mono ester, mp ll9-22°. Further recrystallization gave a sample, mp 130.5-131.5°, suitable for elemental analysis. 46 Anal. Calc for 021H2507N: C, 62.52; H, 6.25; N, 3.47. Found: C, 62.77; H, 6.22; N, 3.55. The infrared spectrum of this substance (Figure 17, p 59) 1, and appropriate bands for the p-nitrobenzoate function at 1725 and 1525 cm'1. The most Shows hydroxyl absorption at 3610 and 3500 cm' interesting feature of the nmr spectrum (Figure 39, p 78 ) is the down- field shift (from T 6.27 to 5.40) of the triplet signal due to the protons a to the primary hydroxyl, clearly indicating that esterification of this group has occurred. 4a8-carboxymethy1-3,4,4a,5,6,7,8,8a-octahydronaphth-58-ol-2(1H):one lactone (73). A slurry of 2.85 g of enone lactone 22 in 50 ml of tetrahydrofUran was added to a solution of 0.445 g of lithium in a liquid ammonia (200 ml) - tetrahydrofuran (70 ml) mixture. The addition required 5 minutes, following which the reaction was allowed to reflux (dry ice condenser) with mechanical stirring for 80 minutes. After cooling the flask in a dry ice - acetone bath for 10 minutes, ammonium chloride was added to destroy the excess lithium and the ammonia was evaporated. Water was added to the residue and the aqueous solution was extracted eight times with 50 m1 portions of methylene chloride. The combined organic washes were dried with sodium sulfate and the solvent was removed under reduced pressure to give 2.95 g of crude keto lactone 22. This crude product was chromatographed on 40 g of silica gel. The first fractions were eluted with methylene chloride (100 ml) and contained hydrocarbons (from the lithium coating); the fourth thru tenth fractions (eluted with 1:9 ether - methylene chloride) contained a total of 1.75 g (61 percent yield) of keto lactone Z2. 47 Recrystallization from ethyl acetate, followed by sublimation at 0.005 mm gave an analytical sample, mp 100-101.5°. Anal. Calc for 612H1603: C, 69.21; H, 7.74. Found: C, 69.36; H, 7.83. The infrared spectrum (Figure 18, p 59) of the keto lactone Shows a lactone carbonyl (1890 cm") and a saturated ketone carbonyl (1723 cm'l). The nmr Spectrum (Figure 40, p 79) shows the C5 proton as a multiplet at T 5.7-6.0. 3-hydrpxymethy]pent-3-enol (81). 80 g of ethylidine diethyl succinate28 in 100 ml of tetrahydrofuran was added slowly (two hours) to 20 g of lithium aluminum hydride dispersed in 1100 m1 of tetrahydrofuran and 300 m1 of ether in a 2 l three-necked flask fitted with a mechanical stirrer, condenser and calcium sulfate drying tube. On completion of the addition, the reaction mixture was refluxed 16 hours and, when worked up in the same manner as the reduction of acetyl diethyl succinate diethyl ketal, yielded 39 g of crude diol. Distillation thru a six inch vigreaux column gave 29.2 g (64 percent yield) of diol 24, bp 88-94°/0.5 mm. A sample, bp 87-89°/0.5 mm, obtained by distillation thru a apinning band column, was used for spectral analysis. This was judged to be a mixture of double bond isomers from the fact that the bis- p-nitrobenzoate melted over a wide range even after repeated recrys- tallization. The infrared spectrum (Figure 19, p 60 ) shows 0-H absorption at 3600-3100 cm']. The nmr spectrum (Figure 41, p 80 ) yields to simple first order analysis. Thus, the C1 and C2 methylene hydrogens appear as triplets (T 6.40 and 7.68, J=6 cps), the allylic protons a to the other hydroxyl group as a Singlet (T 6.08), the 48 hydroxyl protons as a singlet (T 4.93) and the vinyl hydrogen as a quartet (r 4.45) coupled to the methyl group ( doublet, T 8.38, J=6 cps). Chloro-3-chloromethyl-31pentene (82). A solution of 25.9 g of diol 24 in 115 ml of chloroform and 62 ml of triethyl amine in a three- necked flask, equiped with a mechanical stirrer, drying tube, thermo- meter and addition funnel, was cooled to -30° and 37 ml of thionyl 33 in 50 ml of chloroform was added at a rate which allowed chloride the reaction mixture temperature to be maintained below -15°. After being kept at 5° for 12 hours, the dark reaction mixture was poured onto ice and the organic layer was separated and washed twice with 100 ml of saturated aqueous sodium carbonate. Evaporation of the solvent gave 39.2 g of crude dichloride, which on distillation thru a six inch vigreaux column gave 16.6 g (43 percent) of materail bp 53-70°/5mm. glpc analysis (4% QF-l column, 6'xl/4", 102°) indicated this to be a three component mixture with the major product (85 percent) having nmr and infrared spectra consistent with structure 22. The infrared Spectrum (Figure 20, p 60) shows the absence of 0-H stretching and the presence of a double bond (1660 cm'1). The nmr spectrum (Figure 42, p 81) can be interpreted in the same manner as that for diol 24. Bromo-3-methylene-4-oxopentane_(49, R=Br). A solution of 20.0 g of ketal diol 44, R=OH, in 100 ml of pyridine was cooled in an ice bath and 50 g of p-toluenesulfonylchloride dissolved in 100 ml of pyridine was added dropwise over a 1 hour period. After stirring the reaction mixture an additional 2 hours at 0° followed by 45 minutes at room temperature, it was poured into 500 ml of ice water and extracted with methylene chloride (150-100 ml). The combined organic extracts were washed twice with 400 ml of cold 10 percent hydrochloric acid, once with 100 ml of water and then dried over sodium sulfate. The resulting solution of keto ditosylate 42, R=0502C6H4CH3, was cooled in an ice bath and 150 ml of triethylamine added. After 1 hour the ice bath was removed and the solution stirred 3 hours at room temperature. The triethylamine was removed by washing twice with 300 ml of cold 10 percent hydrochloric acid. Removal of the solvent under reduced pressure at 35° gave the enone tosylate 42, R=0502C6H4CH3. The crude enone tosylate in 50 ml of dry acetone was added over a 30 minute period to an ice cooled mixture of 30 g of anhydrous lithium bromide (dried at 110°/0.1 mm for 10 hours) and 250 m1 of acetone. The ice bath was removed after 1 hour and the reaction mixture was stirred for 11 hours at room temperature. Two thirds of the acetone was removed under reduced pressure at 35°, the residue poured onto 500 g of ice and the product was extracted with methylene chloride. Removal of the solvent at 35° gave 10 g of crude bromo enone 42, R=Br. Distillation thru a short path apparatus gave 5.0 g (29 percent yield for the four steps from 44, R=0H) of 42, R=Br, bp 62-67°/4.5 mm, which glpc analysis on 4% QF-l (6'x1/4" at 110°) indicated to be about 90 percent pure. Samples of this purity could be stored several days at -10° without decomposition. The infrared Spectrum (Figure 21, p 61 ) shows an unsaturated ‘ and a double bond at 1625 cm“. carbonyl at 1677 cm“ The nmr spectrum (Figure 43, p 82 ) is similar to that of the other enones of structure 42. The mass spectrum (Figure 48, p 87 ) displays parent ions at 50 m/e 178 and 176, the two possible a cleavages of the methyl ketone at m/e 163, 161 and 135, 133 and loss of bromine at m/e 97. 2-acetylspiro[4.5]deca-6,10-dione (75). To a slurry of 0.600 0 of sodium hydride (52 percent, washed with pentane to remove the mineral oil) and 150 m1 of dimethoxyethane was added a solution of 1.51 g of cyclohexane-1,3-dione in 50 ml of dimethoxyethane. The reaction mixture was stirred at room temperature for 45 minutes under nitrogen and then a solution of 1.77 o of enone 42, R=Br, in 20 11 of dimethoxy- ethane was added and the reaction mixture refluxed for 190 minutes. Filtration of the sodium bromide and evaporation of the solvent then gave 2.2 g of crude 42, which on distillation thru a short path apparatus gave 1.5 g of a fraction bp 133-38°/0.35 mm. glpc analysis of this fraction indicated it to be a 4:1 mixture of 42 and 22. Chromatography of this material on 55 g of silica gel (eluting 25 ml fractions with 9:1 methylene chloride ether) gave 0.80 g of pure 42 in fractions 6 thru 8. A sample of 42 was collected by preparative glpc on a 4% 0F-1 column (210°) for spectral analysis. The infrared Spectrum (Figure 22, p 61 ) shows carbonyl absorption at 1730, 1715 and 1705 cm’1. The nmr spectrum (Figure 44, p 83 ) contains a four-proton triplet at T 7.35 (J=7 cps) for the protons o to the carbonyls in the six-membered ring and a three-proton singlet at T 7.90 for the acetyl methyl. The mass spectrum (Figure 49, p 88 ) has a parent ion at m/e 208 and fragment ions at m/e 193 and 165 for the two possible a cleavages of the methyl ketone. FIGURES 51 .Awwv mcoimicwxwzoru»ufifizuznoxo-.mvimizxo;ume mo Eacuumam cmcmcwcm .N oczmwd . _ . - 111111114111. 1. .1 1 ‘ . . . 1 1 :1 11:11 1111111 1 .2 O V O o "‘41 FDNVLIIWSNVUI om oow 08. so? 83 1,1,, 3 _. . _ ooa ' O ., o o 1%) SDNVUIWSNVUI o___2._ _. oo , 2 . 1- 1 a. 11 . 1, . . . . . . *oe .Awwv ocoim1cwim.—-uonmxo-m-flc.¢.egoFuzuwa—zgpmsim-zxozuoe-m mo Ezcuumom umceemcH .F weaved 52 .Amzuoucum .wwv mumumum pocmucmgoxoiwimcmstameim we Eacuumam umcmcwcm .w mcamwd O V "/1 aoNvuIWSNvaI O 0 .0m zoouo m5 ‘3 {"61 JDNVIIIWSNVHJ. .Aquououu awwv mumgmummn Fowniq.fiimcmp=apzumuoim mo Escuummm nmcmcmcH .m weaned 53 .Awwv mcoiopicmim.Flumcwcpmxomuie.mim ..o.o.¢.~Ho—u>urcupxnumsim mo Ezcuumam umcmgwcu .m weaned 111i 1 .11 1110 1 1 -.M.M W _ . mxu o 1 a V ovN S M w m .......... 001 om com coo. 00m, 004. ooo. com. 005. .. ..- nn 1 1 .u- 11s, more .T 1 1 o _ , mxuooo ,1 2 ~18 w 1- a itiN. e w 11m ”on , «mmv1wcoim1cmxmsopuzuAquanoxoi.mixxogmum—zcumi.Nvimiaxospme mo Eaguumam umgmcpcfi .m we: mu 54 .Amfimxuvuououm .wwv mpp~m>wa Focpucmaoxoicicmpzcumeim mo Escuumgm nmcmcmcfi .w oczowu A...“ mfimzovuouo «:8 Dnlam, I}, 11.1.‘,. \J‘Qd‘_ \JDJ I van \'.,\lom oon m 000m MAmxuvuouo namxuvuou -.__:: (11 .Amflmxuvucucna .mwv mumpe>wawe Foweiv.Fiwcep:npmpoumim mo Echumam ememcecu .5 acumen 171141]! IIWHNVHI .Amxpuwxucum .wwv cmsum Fchma Pocmucmaox01Q1mceFzsuweim mo Eacuumem umgmgmcm .o— weaned 111.111 1 a. 1. 1 N :u 9.. 1 1 m. , mxouwzuo 18 m 1 1 V on W. T- o M b m 1 , w r) 00 .1... 1 - _om . v 4 u . 1 .. i 1 UN 00m C 9 ago 1 00mm 009 on.” m 009. m elegazu -—_- -——————.. mIeUNIue {it rFIK/illlilile .Amzeoaxuoum .mwv seeps Paneoeee _owe-e._1aeee=e_speoe-m ea asepwpam-eoeeeeeH .o aesmwe 56 .38 .Ax1m .wwv wcozw1N1xmso~uauFacume1m1Fxgum5>xonumncmu w1A~A2um>xoach1mviN mo Eacauwam umcmcwcH .Np mczmwu O '3 -'./-1 DNvuIWSNvaI daemon: .73.. . 0 O a) o 52 000m 00.». m 000». (J 1'7.’-1 iDNVlllWSNVlfl w 1-1. 1. .--- 11.00” coumm _>;um upoe opu mommA”zusuoxo1.mv1N1mcon1_m. F1mcmchoru>U1.Ng1N mo Ezcuumam umcmcwcH .FF mgzmwu 57 .Awwv mcopump mco1AImVN1Fo1mm1cu5amcoLuXcmxmz1w.N.u.m.m<.¢1pxcumszxongm01mmc mo Eaguomam umgmg$=m .vp mgamwu 1* . L l)NVJ l IW‘1NVHI 1onn OOmN 000m damn mzauNou 2: o on Q!_-.: .Awwv aco_u1axo.zmvm.N1mcmFmgucamcoguzsoguuu1m.N.m¢.¢1p>;pmsxxocuwngmU1me No Eagaumam umgogmcH .mp mgamwm DNVI H"; .5 1: 58 Ce; .2& avg .--o— , .. Fo1mm1cu5am=Achumzxogvzc1=Nv1um¢1oguzcoxmc1m.u.m.me.¢.F1amca—oxomu1.mnpv1N we Ezguuoam nmgmgmcm .o_ wgamvm 1 4 M o « . ‘ Q 1O :1. I.|i| 1 - |u1l||I1II1l1$III 1 1O. '1 ' 'L C W . . 1 . 1 . . , . on 6 V W SDNVJJIWSNVM O '0 O on oon N 000m 002” 000». .Awwv acouuop fl ON. C) C v BDNVIIIWSNVUL O '0 10m _: ~o1mm1nu2amcoguxcmxms1u.w.m.me.¢.P1chmpoxowU1.mflPv1N1pzcume>xongmu1mn¢ mo Esguumam umgmgycH .mp mgsmwm .AWNV mcouump mco1AIPVN1Po1mm1cpgamcognzzmuuo1mw.w.N.w.m.m¢.¢.m1PzzpmexxoanU1mm¢ mo Eagpuwam cmgmgmcm .w_ mgzmwm , 3 C) (‘1 c. x ‘J r) (1 "1." T)NVI HWGNVHI com ha, . , ..,\ O u ’1) -o {‘74} JDNVI IIWS‘NVHI oz:xmuooo om 1 1 1.5. . & .Auoz::muouua .mwv mumo~cmaoguwc1n one: po1mm1guccmc A~asumxxoguxg1:Nv1mm¢1oguzgwxmg1n.w.m.m¢.¢.P1Amcm~oxowv1.mflpv1N mo Ezgpuwam vmgmgmcu .NF mgzuwu f: 60 ga—N. com 1 man? .11 .. “-1 ..1..... . .. .91 comm 000» .0031 000v 1|v|vlll1PIl|olIlll1l 11 111.|. jC :c O (‘J .....g. .--—A--.. ..._. ..-..- :c (I)- v 1 1-_-..—---.-.--- 1. c —-C).-- ..- (D L .Awmv pocm1M1ucoachuwemxoguxg1m mo anguumam umgmgmcH .mp mgsmvm v: O O ‘VIBDNVJUWSNle O O 3DNV11IWS‘NV81 uw 61 .NN mgamwu .Amwv mcowu1op.m1muwMWm.¢HogwamPauwum1N No Ezgpumam wwgmgmcfi o nzu ON o o m w w \M um I Will/Ill», 9°. 00m goo. oo~1 can” on”. man» 90m. 1 1 11m 1Jn1 m 1 c m . Hm a , w .M 1.. m . M .1m .Agmum awv mcmucmaoxo11¢1wcmFxgums1M1oEOLQ yo Ezgpumcm umgmg¥c~ .PN mgzcwm 62 .AMpuouv “wwv ocO1m1cmxmcopuzuAquanoxo1.mv1N1zxocume we Ezgpumam LE: .mN mgamwm Qm 3w 3 Em 5 ed _ . p p b p . h — p p . p b p p .— ml . p P p b - mxu mzuo 63 .Amcwvnguv away mco1m1cwxmgopu>uflquagoxo1.mv1N1>xo;uoE No Ezguumam LE: .¢N mgamrm 151...? .o._«. 1 mxu ‘ J! o.» 1 N 1 mxuo .Azpuuv Awwv mco1~1cm1w.F1umvwx01N1mo.v.cgopuzUVQPkgpmz1m1zxonums1m mo Ezgpumnm L52 .mN mgzawu 153.1 ON o...” 06 111..111q1111fl1111q1111_11 o o.— . 1 4 i ‘ _ ‘ ‘ ‘1 _ Q 64 .A:Fuuv Amzuoooum .WWV muuumumwn Fomu1¢.P1mcmp=apzumum1N mo Ezgpuwam LE: .oN mgamwu m 3 2.: N o m . .1; . _ _ d a _ _ d _ _ _ _ a _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ Iucuc zucuc - .. .“.o- -F. . ‘- ‘w W-, “-1... -—--.-~ -w. -r . .lJ /_ m.n ce.n cm.~ cm.n op.n oo.n ~14 q _ _ _ 41_ _ _ _ _ _ _ d _ _ _ _ _ _ _ _ _ _ _Ifi1_ .LJ I I“ 66 .Aummcv Amzuoucum .wwv mumumum FocmpcmgoxC1¢1mcmnguma1m yo Esgpuomm LE: .NN wgaopu 0 m n 2.: o #14 fl _ # _.___ . _ _ _ _ 4 d .141fi _ _ _ _ _ N A _ _ _ _ __ _ _ _ _ .0- - -—-—-.—c-.- m '22 L) coo C7 .AJPuuv Awwv m:o-m1coxw;o~u>U1APangox01.M1zxouoqumme1.Nv1N1>xo;qu mo Eagpomam LEN .mN ogzuwu $5.3. o m N o .11._%_~_____d~__aa~___._fi.fl_qq_____a-______—a_d__~4l g..—.— o- -- mzuouo m5 _ - .-1..? .- —~.... g—.— -~.. .“ A 3:- .AMFUQQV Awmv «c01oF1cm1o.p1umcwgumeWU1q.N1fi~.mc.o.¢.nNo_u>u?gp_>;pwe1m we Eagpomgm Lax .oN mgzuwu Q5 cum 3 ”E E c; :1 1— % 4 4 d d I d d u —1 I u d u q 1 + — d d d d - I 1 d 4 — 1 1 1 mzu WWW mpmpm>wng powc1¢.P1mcmp:nP>umum1N Lo EaLpumgm LE: .om mLaowL .LJLUQLLmLmzuvucuguu . .LH c.c p.u c.N “LL Lag c.L L.m a.; DP-L-un-b-—-P-~_ppbrb-ub-F-hnrbbPDb—nbbnnnhbn_h--pP-P— C1“ mLm;uvuouc .AJFUUV Amfimzuvucucua . V mum m>wa —ocmucmgox01¢1mcmP>zume1m Lo EsLuumnm LE: .Fm mLzqu 0.0 05 gm 0K 3 LL“. c u o.m c5 7n dd‘dI—d«4141:661—‘1411—11qddfldd+4fidqu u u _ m H L 4 _ N: _m _ :5 w .M mLmzuvuooc LL _ L. .__ _ L _._. . L. ,———~ .. _- n--~-— _-‘ --- 71 .LMLLQUL LmzmoN1901m .mev .m— o.m Qw IPDDPbe—bb-htPPD—h ngum Fauna: Focmgcmgoxo1¢1mcmngum51m Lo EaLpuoom LEz .Nm mLzmwu E 3 2% 3 9m 0.: OJ DbbhlbD—hDD-DPLPD—bbb bP-bb—PbbPDDDLb—bbbbbb+bI—I ~20 mzmuwzuc Lu 72 .AMPUQQVAme memw Panpm uvum ovumummfiPapanox01=mv1N1mcoLc1.mup1mcmxmnopuxU1.Ng1N Lo asLuumam Lax .mm mLame 3 .o.« .L 1% 53.... .o.» .3 ‘1 1 [q — 1 G ‘ 1 “ 6 G ‘ fl ‘ q d _ J ‘ G 1 - ‘ ‘ J — l 1 ‘ ‘ A. 73 .A3FUUVAzua.mwv mcocm1N1xmcoFuzungumsum1FagumemxogumuLmU1o1apmzpmzxoaLmu1mv1N Lo EzLuumLm L52 .qm mL:m_u Au.— 1 1 1 — 1 1 -. --’_.—.— u’L... .--. 'm.‘v- a. ~— I...“ ‘ -—o—-.n-. moo... . com. . 1 1- '1 1 1 1Av u..-o~ .« 251.... . 1— 14 1 d1 ‘ Avgm 1 1 _ 1. . :«ou 13 1 1-1 1 ‘1 1 _ T11 1 1 ~20 QNoNQmI 74 S .AMPUQUVA—uv mcowu1AIw.zmvm.N1mcmPmchQMLOLuxgmLump1w.m.m¢.v1pxcumemxogpmaLmo1mq mo EzLuumom Lsz .mm mLaowu oL.. new Lam . “xv Lmvccmm nan Awe 11LJ1 111111411J111111H1111+T111_1111_T111_1111. 11.11 o WWW LL mzuuuou o AMPQQQVAwwV mcouumP mco1A1mVN1F01mm1cpcgmcoanzmxmzwm.m.m.m.m¢.¢1FzgpmsxxonmU1umq Lo EaLpumam LL 0 m u :JEJ o fi____%__L__::L_fiLAL_: .Lm mL=GLL n ”1 _:__111L_L_L1_L_L_LL1«L 1 31113 75 ,llli'l“ .UI' . I ’11 I 1 1.. . . . .LMLUQUL wav 1COLU1L 1 po1mm1cugamcovacmxog1u L m cc w P1Amcm~oxowu1.mflFv1N1FzsumsxxonLQU1uo¢ Lo aaLpumam Lax .Nm mLaowL o m L 32%: n .1 fia_qLL4LfiL_L:__:_L:__________:_:L__LLLL:_L__ 76 77 (gm .LMLUDUV Axum .wwv Lo 1 m15pcumcaszpmmxoLu>L1=Nv1me1ochgmxmc1n.u1m.m¢.c.P1Amcmpoxowu1.mflpv1N Lo EzLuumom Le: .mm stme _ o w va an; L _ o n ______4___L4_JL_:LLLALL____4__:1___41_:_J_:___4__L o o (I: 78 .LmLuauv L~021zmuouug .va mLmONcmLOLLLC1a ocoa Lo 1 1mm1gu5achchamxxOvaz1=NV1mw¢10LcanmxmL1m.w.m.m¢.<.F1Amcmpoxowv1.mupv1N Lo EzLuumam LE: .mm mLzovu 0:16. 0K o6 Rm 3:“: Ca GM 0.“ EprP-ppb—P-ppp-pbn—ub-PbP--——bnnph-pup—Pb-b-bbhn—bbppbb-P—bbP-pbP#-F- \ II: o: Nozvzmuou. .383 Amwv mcouum_ mLO1ALFVN1F011m1zuggm:0Lv>;mpu01mw.w.m.m.m.mq.q.m1ngumezxoanU1mm¢ Lo EzLuumom LE: .ow mLauLL Qm 9w 3 FE 0.x o6 HLLLLL_._._.Lb._LL—____PLLLPbPrrL_PL 80 .2110— ésev L o S 3 Focm1M1uconnguweszLu>g1m Lo EsLuumLm LE: .Fv mLzawL w \1 o .J47L1 JL « L4 1 . L L L LL '1 1 11111111311111 1- 11 ii: 4 _ m :c :u :0 l1] .AJ—uuv Aewv mcw1m1ucmaFxnmeOLoPLU1M10LoFLU Lo EzLuumnm LEz .Nv mLzme (fig 0 m n o n 1 «71+fl4. 11“?.41fififiqjidddqd—L«11114—44—qqaaflqqfidfiqj4 1 m _.u 5 m \V”\ 8 n11. nu .A3—uuv ALNum .wwv mcmpcmpoxo1V1mccFzzpws1m1oaoLn Lo asLuuoam LE: .me cLaqu 0.3 9m 1c; 0K 3 Em ad Qm 0.: F — b b r h bl P n b p p F n b p p P b p — p P n P — p P L amp moo CLL om. omm mag own cum M b _ L _ L Ll P . h L _ N 0 LL mzo .Am—uuv ANN mcop 1 . L V .u c_ o1mumumm.¢uoLwcm—mpmum1N Lo Eszumam LE _ : .qq wLaaLL o._ E E . ON 1 1 1 o.” w #1 I 1 1 fl [- 7 _ 1 1 1 1- 1 1 1 1 1 1— 1 CON owH .Awwv mco-m-cmxm;o_u>uAFXuacoxou.mv-N->xo;umE we as; {a qua omH ooa om _ , . _ . » OJ’J'NVU mmmwpk l—u- .mq mgauwm 03 ,~ 1 : mxu o mzuo 4____ r3 I. OOH J0 aSaunOJaa (o a need 339 .Awwv mco-n-cm-w.F-uocmxo-m-ho.q.ugopu>uwg_x:pms-m->xocuma-m mo Eaguumnm mmmz .mv mgamwm a} CON owH 00H OQH ONH OOH ow . _ P ._ _ P P4 h — P a— m— — aunt .— o u . . . . a \Q s s s ax ~ MW :3 mzuo o o 0N OH on 8 R 8 8 om ow OOH go aflequaoaag h Ca. x305 ase- S .Ammv m:o-o_-cm-o.F-umn?gumxcwu-q.m-fl 36 OOH b . mzu OOH .OVH N.no.0.TDo—uzutfizfigé .B 2:535 32.. ONH — OOH b __. . "ll'Ph O¢ .nv mgsmvu ON C O 9% oz .4 5} C>~ (3 C) t- \O ‘U\ x995 9333 30 aSequeozad O 130 O O\ OOH .Agmua .wwv mcmucmaoxo-c-mcm—x;um5-muoeogn $0 Ezguumam mmmz .mq mgzmwm m\E om— omr ovp omp cop ow om ov om P . _ h _ . — . _ . _ I A. __ __ _ o N5 2... r 00" need 3523 J0 aoequanad 88 .Awwv acowc-o_.m-mumcmm.cgogwnmrxumum-m vo Ezggumam mmm: .oc «gamma o\EV . . OOH OOH ,OvH ONH OOH ow 00 ON 3 , ~_ _ ___m_€_ _ o O H 8 O M I l C) c: In wt xeea aseg 30 azaqueoaag I O \D "5 l O l‘ O l C no I O O" I OOH REFERENCES 10. 11. 12. 13. 14. Comm., 29, 539 (1964). 89 REFERENCES J. Vrkot, J. Jonas, V. Herout and F. Sorm, C011.Czech. Chem. K. R. Varma, M. L. Maheshwari and S. C. Bhattacharyya, Tetrahedron, 21, 115 (1965). J. A. Marsha11, N. H. Anderson and P. C. Johnson, J. Am. Chem. Soc. , 82, 2748 (1967). J. A. .)Marsha11 and P. C. Johnson, J. Am. Chem. Soc. , 82, 2750 1967 J. A. Marsha11 and P. C. Johnson, Chem. Comm., 391 (1968). w. 2. Chow, 0. Moth and F. Sorm, C011. Czech. Chem. Comm., 21, 1914 (1962). B.Chiurdoglu and J. Decot, Tetrahedron, 4, 1 (1958). ?. Romanuk and V. Herout, C011. Czech. Chem. Comm., 22, 2540 1960 . J. Vrkoc, V. Herout and F. Sorm, ibid., 3183 (1961). R. J. No C1ure et a1, Chem. Comm. W35 968) give an X-ray structure deteranation. J. A. Harsha11 and S. F. Brady, Tetrahedron Letters, 1387 (1967). F. Kido, H. Uda and A. Yoshikoshi, Tetrahedron Letters, 2815 (1967) and 1247 (1968). R. Sakuma and A. Yoshikoshi, Chem. Comm., 41 (1968). An X-ray crysta1 ana1ysis has recent1y confirmed these structures, see R. M. Coates, F. Farney, S. Johnson and I. C. Pau1, Chem. Comm., 999 (1969). I. N. Nazarov and S. I. Zavya1ov, Izvest. Akad. Nauk. SSSR, 0tde1. Khim. Nauk., 207 (1957). E. C. Kornf1e1d, 9, 9:9, Chem., 12, 1671 (1954). 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 131. £32. 90 E. F. Landon and E. P. Irany, 9, 999, 9999,, lg, 422 (1946). H. J. Hagermeyer, Jr., _J_. 9m_. 91929. 999., a, 1117 (1949). D. R. Gupta and H. D. Pathak, 999,_9, 9999,, g, 373 (1964). H. Kataoka, ChemicaT Abstracts, 92, P451a. V. B. Piskov, Z9, Obschch. Khim., 99, 1390 (1960). The a1do1 condensation between 1evu1inic acid and forma1dehyde provides a more direct route to 3-acety1 butyr01actone, but the yie1d is 0n1y 8 percent, cf. J. Rothe and H. Zimmer, 9, 999, Chem., 99, 586 (1959). For a recent examp1e of j9_situ generation of an unsaturated ketone from a B-keto ether see E. Wenkert and D. A. Berges, 9, 99, Chem. 999,, 92, 2507 (1967). P. HieTand and K. Miescher, He1v. Chim. Acta., 99, 2215 (1950). S. Ramachandram and M. S. Newman, 999, 999,, 91, 38 (1961). C. B. C. Boyce and J. S. Nhitehurst, 9, Chem. 999,, 2680 (1960). H. Stetter and H. Dierichs, Chem. 999,, 99, 61 (1952). For severa1 examp1es of this see: E. J. Corey and S. Nozoe, 9, 99, Chem. 999,, 91, 5728 (1965) and the references cited erein. H. C. Brown and D. B. Big1ey, 9, 99, Chem. 999,, 99, 486 (1961). For examp1e E. 0ka, Chemica1 Abstracts, Qg,P15568i suggests heating the di01 wtih excess thiony1 0h10ride in benzene and pyridine at 50° for three hours. M. c. K1oetze1, 9. 99. Chem. 599., 19, 3571 (1948). J. H002 and S. S. H. Gi1ani, 999, 9, Chem., 99, 86 (1968). This 1ithium bromide acetone of converting tosy1ates to bromides is due to S. F. Brad , M. A. I1ton and w. 5. Johnson, 9, 99, Chem. 999,, 99, 2883 (1968). H. Stetter, "Newer Preparative Methods of Organic Chemistry", V01. 2, Academic Press, New York, N. Y., 1963 , pp 51-100. N. Cheronis and J. Entrikin, "Semimicro QuaTitative Organic Ana1ysis“, 2nd Ed., Interscience Pub1ishers, New York, N. Y., 1957, p 393. 33. 91 Purificarion of the thiony1 ch10ride is essentia1 to the success of the reaction, for a simp1e procedure see L. F. Fieser, "Experiments in Organic Chemistry", D. C. Heath and Company, Boston, 1957, p 345. ’\ v L's 1111111” 072 ”11111111111WW