Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 73-30,250

CHEMICAL CONSTITUENTS OF ARTEMISIA FILIFOLIA (TORR.)

Item Type text; Dissertation-Reproduction (electronic)

Authors Torrance, Sterling Jay, 1943-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 05/10/2021 04:38:05

Link to Item http://hdl.handle.net/10150/290336 INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

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

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

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

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

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 73-30,250

TORRANCE, Sterling Jay, 1943- CHEMICAL CONSTITUENTS OF ARTEMISIA FILIFOLIA TORR.

The University of Arizona, Ph.D., 1973 Chemistry, organic

l University Microfilms, A XEROX Company , Ann Arbor, Michigan

THTS riTSSRRTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. CHEMICAL CONSTITUENTS OF ARTEMISIA FILIFOLIA TORR.

Sterling Jay Torrance

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF CHEMISTRY

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

19 7 3 THE UNIVERSITY OF ARIZONA.

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my direction by Sterling Jay Torrance entitled CHEMICAL CONSTITUENTS OF ARTEMISIA FILIFOLIA

TORR. be accepted as fulfilling the dissertation requirement of the degree of DOCTOR OF PHILOSOPHY

7A/73 Dissertation Director Date

After inspection of the final copy of the dissertation, the following members of the Final Examination Committee concur in its approval and recommend its acceptance:""

^/•73

tHfvvC/ Urt '

This approval and acceptance is contingent on the candidate's adequate performance and defense of this dissertation at the final oral examination. The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination. t

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: ACKNOWLEDGMENTS

The author wishes to express his gratitude to Drs.

C. Steelink, the director of this research, T. A. Geissman,

J. Wolinsky, and R. G. Carlson for assistance during the course of this work.

The author is especially grateful for financial assistance in the forms of a National Science Foundation

Summer Traineeship (Summer, 1971), Atlantic Richfield

Company Fellowship (I Semester, 1971-72), and a University of Arizona Predoctoral Fellowship (II Semester, 1972-73).

iii TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS viii

LIST OF TABLES . . . X

ABSTRACT xi

INTRODUCTION 1 { RESULTS AND DISCUSSION 4

A. filifolia Monoterpene Lactones 4 Minor Monoterpene Constituents of A. filifolia 32 Other Constituents of A.' filifolia 36

BIOSYNTHETIC IMPLICATIONS 43

EXPERIMENTAL 59

Collection of the Plant Material 60 Steam Distillation of the Plant Material 60 Fractional Distillation of the Steam Distillate 61 Hydrolysis of the A. filifolia Lactones 12 and 13; Isolation of Hydroxy-acids W and M . . 61 FractionaT^Crystallization of 17 and 18 64 Regeneration of Lactone 12 from^the Hydroxy-acid 17 . . 66 Regeneration oflactone 13 from the Hydroxy^HCid 13 •••••••••••••••• 66 Preparation of i^enchone Oxime (Figure 4) 67 Acid Ifydrolysis of Fenchone Oxime; oc- and P-Fencholenic Acid Nitriles (Figure 4) 68 Selective Saponification of the Mixture of a- and P-Fencholenic Acid Nitriles; 3- Fencholenic Acid and a-Fencholenic Acid Amide (Figure 4) 69 Hydrolysis of a-Fencholenic Acid Amide; a-Fencholenic Acid (Figure 4) 71 Preparation of Geranic Acid (Figure 4) 72 Preparation of (jO-Filifolone (Figure 4) 74

iv V TABLE OF CONTENTS—Continued

Page

Hydrolysis of Filifolone; a-Fencholenic Acid (Figure 4) 75 Iodolactone of a-Fencholenic Acid (Figure 4) . . . 75 Dehydroiodination of a-Fencholenic Acid Iodolactone; Preparation of Lactone 2 (Figure 4) m 76 Hydrogenolysis of A. filifolia Lactones 12 and 13; Preparation of Carboxylic Acids' 9 ancPlO 78 Preparation of a-Cyclogeranic Acid (5) 79 Hydrogenation of the Carboxylic Acid Mixture 9+10 from A. filifolia; Preparation of 1f;he Carboxylic Acid 11 81 Preparation of 2,4-Dime^hyl-l,3-pentadiene (Figure 5) 83 Preparation of the Carboxylic Acid 9 (Figure 5) ? 84 Hydrolysis of Chrysanthenone; Preparation of Carboxylic Acid 9 (Figure 5) 85 Hydrogenation of Synthetic 9; Preparation of Carboxylic Acid 11 (Figure 5) 86 Methanolysis of the^A. filifolia Lactone (12 + 13) Mixture; Preparation of 19 88 Oxi$ationwof 19; Preparation of the Keto-acid 89 Photolysis of^Verbenone; Preparation of Chrysanthenone (Figure 8) 90 Oxidation of Chrysanthenone; Chrysanthenone Epoxide (Figure 8) 93 Solvolysis of Chrysanthenone Epoxide; Preparation of the Acid 17 (Figure 8) 95 Thermal Rearrangement of Chrysanthenone Epoxide; Preparation of Lactone 12 (Figures 8 and 9) T 96 Bromination of the Acid 9; Preparation of Lactone 12 (Figure 10 )w 96 Photolysis^of Chrysanthenone; Preparation of the Ketone 22 (Figure 11) 98 Oxidation of Preparation of Epoxide 23 (Figure ll)MMtt ^ ... . 101 Thermal Rearrangement of Epoxide 23; Preparation of Lactone 13 (Figure 11) 104 Solvolysis of Epoxide 23 ^Preparation of Hydroxy-acid 18 (Figure 11) 106 Preparation of Methyl—2-butenoic Acid (Senecioic Acid, p,0-Dimethylacrylic Acid) (Figure 12) 107 vi TABLE OF CONTENTS—Continued

Page

Attempted Lewis Acid-catalyzed Diels-Alder Reaction of Senecioic Acid and Isoprene (Figure 12) 108 Preparation of Ethyl 3-Methyl-2-butenoic Acid (Ethyl Senecioate) 109 Attempted Lewis Acid-catalyzed Diels-Alder Reaction of Ethyl Senecioate and Isoprene (Figure 12) Ill Performic Acid Oxidation of 9; Preparation of Lactone 28 (Figure 13) . .m 112 Iodolactonization of the Carboxylic Acid 9; Preparation of Iodolactone 30 (Figure l^f) ... 113 WAAj4 Preparative GLC of the A. filifolia Steam Distillate; Isolation of the Mixture of Lactones 12+13 114 Preparative^^LC o¥ Lactone 12; Isolation of the Mixture of Lactones l^w+ 13 115 TTHeating of_ TLactone . 12T „ mm mm 115 Steam Distillation ol Lactone 12 116 Steam Distillation of 17; Isolation of the Mixture of Lactones 1'$ + 13 116 Photolysis of ( + )-Verbenone'f^Obtention of (+)-Chrysanthenone (Figure 8) 117 Oxidation of (+)-Chrysanthenone; Preparation of (-)-Chrysanthenone Epoxide (Figure 8) . . . . 118 Thermolysis of (-)-Chrysanthenone Epoxide; Preparation of (-)-Lactone 12 (Figure 8) . . . . 119 Hydrolysis of (-)-Chrysanthenone Epoxide; Preparation of (+)-17 (Figure 8) 119 Preparative GLC of the A. filifolia Steam Distillate; Isolation of (-)-Verbenone, Borneol Piperitenone, and the Enedione 34 (Figure 15) 120 Preparation of (+)-Verbenone 2,4- Dinitrophenylhydrazone 122 Preparation of Piperitenone 2,4- Dinitrophenylhydrazone 123 Preparation of [3-Isophorone (Figure 17) 123 Preparation of 4-Hydroxy-3,5,5-trimethylcyclohex- 2-en-l-one (Figure 17)..... 125 Oxidation of 4-Hydroxy-3,5,5-trimethylcyclohex- 2-en-l-one; Preparation of the Ene-dione 34 (Figure 17) T. . . 126 Preparation of Phenylsemicarbazide 127 Preparation of the Mono-phenylsemicarbazone of 34 . 128 mm vii

TABLE OF CONTENTS—Continued

Page

Chloroform Extraction of the A. filifolia Plant Material 129 Chromatography of the Acidic Extract of Artemisia filifolia: Isolation of Acacetin (36) and the Keto-acid 35 130 Preparation of the Keto-acid 35 132 Preparation of Anisic Anhydricfe 133 Preparation of Sodium Anisate 133 Preparation of Acacetin (36) 134 Preparation of Acacetin Diacetate 135 Chromatography of the Neutral Chloroform Extract of Artemisia filifolia: Isolation of the Sesquiterpene Lactone Colartin 136

APPENDIX A. NUCLEAR MAGNETIC RESONANCE SPECTRA . . . 139

APPENDIX B. ULTRAVIOLET SPECTRA 199

REFERENCES 205

/ LIST OF ILLUSTRATIONS

Figure Page

1. Major Constituents of Artemisia filifolia ... 2

2. Infrared Spectrum of the cioHi4°2 Lactone Mixture . . 5

3. 100 MHz NMR Spectrum of Clf.Hn .0_ Lactone Mixture (in CCl^) 6

4. The Synthesis of Lactone Structure 2 9

5. The Preparation of Carboxylic Acid 11 12

6. Infrared Spectra of Pure 12 and 13 14 c mm mm 7. 60 MHz NMR Spectra of Pure Lactones 12 and 13 (in CC1.) T 15 mm 4 8. A Synthesis of Lactone 12 16 J mm 9. The Thermal Rearrangement: Possible Mechanisms 18

10. A Second Synthesis of Lactone AAA12 MA 20 11. The Synthesis of Lactone 13 21 J mm 12. The First Unsuccessful Synthesis of Lactone 13 24 mm 13. The Second Unsuccessful Synthesis of Lactone 13 25 mm 14. The Third Unsuccessful Synthesis of Lactone 13 27 mm 15. Typical GLC Chromatogram of the A. filifolia Steam Distillate 33

16. NMR Spectrum of Ketoisophorone (34) (in. CCl^)_ \ MAMA 35r-

17. The Preparation of Ketoisophorone (34)MAMA 37 viii IX

LIST OF ILLUSTRATIONS—Continued

Figure Page

18. NMR Spectrum of Acacetin (36) (in pyridine-d5) T . . 40

19. NMR Spectrum of Colartin (37) (in CDC13) T 42

20. Classical (Head-to-Tail) Biogenesis of Borneol, Camphor, Piperitenone, Verbenone, and 1,8-Cineole 44

21. Classical (Head-to-Tail) Biogenesis of Filifolone 45

22. Classical (Head-to-Tail) Biogenesis of 12, 13, Isophorone, 34, 35, and Filifolone**. ... 47 mm' c ' mm' hi ' 23. Proposed Biogenesis of Lavandulol (44) 50

24. Proposed Biogenesis of Artemisia Ketone (46), Santolina Hydrocarbon (47), ciirysanthemic Acid Esters (48T^ and Lavandulol (44) 52 mm 25. Lavandulol-like Biogenesis of 12, 13, _ . _ mm' mm' Isophorone, 34, and 35c 54C/1 c ' mm' mm 26. Lavandulol-like Biogenesis of Piperitenone, Verbenone, Camphor, Borneol, and 1,8- Cineole 55

27. Proposed Biogenesis of the Flavone Acacetin (36) 56 mm 28. Proposed Biogenesis of the Sesquiterpene Colartin (37) 57 mm 29. GLC Chromatogram of Chrysanthenone Photolysis Products 100

30. GLC Chromatogram of Epoxides 23 and 24 105 /VA/M ftwrn 31. Mechanism of Formation of 3-Mpthyl-3- phenylbutanoic Acid 110 LIST OF TABLES

Table Page

1. Steam Distillations of Artemisia filifolia ... 62

2. Fractional Distillation of the A. filifolia Steam Distillate 63

3. The Photolysis of Verbenone 91

4. The Photolysis of Chrysanthenone 99

x ABSTRACT

The western sand sage Artemisia filifolia Torr. has

been examined in order to determine the nature of its

secondary metabolites. The steam volatile oil of A. fili-

folia was analyzed by fractional distillation and preparative

GLC and was found to contain ten monoterpenes. Of these,

four had been previously reported to be constituents of A.

filifolia. This group consists of isophorone, 1,8-cineole,

(-)-camphor, and (-)-filifolone. The remaining six mono

terpenes of A. filifolia have not been reported as constitu

ents of this plant. Four of the latter are of known structure

and were found to be piperitenone, borneol, (-)-verbenone,

and 3,5,5-trimethylcyclohex-2-en-l,4-dione. The last two

unreported monoterpenes of A. filifolia are new compounds,

shown by independent synthesis to be 1(S), 5(S) — (—)— 5—

hydroxy-2,2,4-trimethylcyclohex-3-ene-l-carboxylic acid

Y^lactone, and 1(R), 3(R)-(+)-3-hydroxy-2,2,4-trimethyl-

cyclohex'-4-ene'-l-carboxylic acid' y-lactone. Some synthetic

preparations toward the structure proof of these 2 lactones

led to three other new monoterpenes. These are 2-hydroxy-

a,a,3-trimethylcyclopent-3-ene-l-acetic acid y-lactone, cis, v cis-3,4-dihydroxy-2,2,4-trimethylcyclohexane-l-carboxylic

acid 5-lactone, and cis-4-hydroxy-trans-3-iodo-2,2,4-tri-

methylcyclohexane-l-carboxylic acid 6-lactone.

xi xii Artemisia filifolia was also extracted with chloro form. The sodium hydroxide extract of the chloroform extract of the plant provided two acidic compounds. These were found to be the known flavone acacetin (5,7-dihydroxy-

4 *-methoxy-flavone) and the new monoterpene-1(S)-(+)-5-keto-

2,2,4-trimethylcyclohex-3-ene-l-carboxylic acid. The latter structure was proved by independent preparation.

Column chromatography of the neutral chloroform extract of A. filifolia resulted in the isolation of one sesquiterpene lactone, shown to be the known colartin.

Biosynthetic pathways are proposed for all compounds isolated from Artemisia filifolia. INTRODUCTION

The genus Artemisia is a member of the Anthemideae

(mayweed) tribe of the family Compositae of flowering plants.

Perhaps the most well-known species is Artemisia tridentata, or common sage brush, the state flower of Nevada. Artemisia filifolia grows in sandy soil at elevations of from 4,000 to

6,000 feet (Hall and Clements, 1923, pp. 130-131). It is found in the central, north-central, and northeastern por tions of the state of Arizona.

Artemisia filifolia, along with A. tridentata and A. frigida, was used medicinally by Indians and early white settlers. The latter was also used by Hopi Indians to flavor roasted sweet corn (Berney and Peebles, 1951, p. 938). This historical interest, together with its relatively large per centage (ca. 1% by weight) of odiferous steam-volatile oil, led to an interest in A. filifolia. This oil was initially noted to be largely terpenoid in nature (McCaughey and

Buehrer, 1961) and has been related to the natural resistance of this plant to herbivore browsing (Martin and Steelink,

1961). Preliminary analysis of the A. filifolia steam- volatile oil indicated that its major constituents were

(Figure 1) (-)-camphor, 1,8-cineole, isophorone (Onore,

1 (-)-camphor 1,8-cineole (eucalyptol)

isophorone (-)-filifolone

Figure 1. Major Constituents of Artemisia filifolia 1967), (-)-filifolone (2,6,6-trimethylbicyclo[3.2.0]pent-

2-en-7-one) (Bates et al., 1967), and an apparently insepa rable mixture of two lactones of composition cioH14°2 (Tammami, 1970).

The continued investigation of A. filifolia presented in this dissertation was prompted by several questions: (l)

What are the structures of the lactones of composition

C10Ii14°2?' can t^lese lactones be prepared in the laboratory?, and (3) What are the possible modes of bio genesis of the unusual terpenoid structures (especially isophorone and the two lactones) present in A. filifolia? The latter included a search for possible biogenetic inter mediates in the plant. RESULTS AND DISCUSSION

A. filifolia Monoterpene Lactones

Preliminary work (Tammami, 1970) on Artemisia fili folia had indicated that a major component was lactonic in nature. This material was collected by preparative GLC of the steam distillate of A. filifolia. It was found by the same worker to be an inseparable 1:1 mixture of two lactones, both of the molecular formula cioH14°2" mixture showed a single molecular ion in its mass spectrum with m/e 166.

The elemental analysis was consistent with the latter molecular formula. That both lactones are strained was indicated by the single infrared absorption at 1770 cm-"*" (see

Figure 2). The NMR spectrum of this mixture, obtained by

Tammami, is presented in Figure 3, which also includes the original peak assignments. Based upon the NMR and IR spectra the lactones were initially proposed (Tammami, 1970) to be represented by structures ^ and 2:

filifolone 5 Wavelength (micro ns)

•4000 3000 20OO ISOO IOOO 9oo 3oo 7oo

Frequency (cm""'')

Frequency Structural Assignment

810 cm""'" trisubstituted C=C -1 1385 cm' gem-dimethyl -1 1770 cm' Y-lactone C=0

Figure 2. Infrared Spectrum of the cioH14°2 Lactone Mixture 6

_y\A__/v_A- 4- i t I 1 • i ppr (6) 5 4 3 2

Assignments:

Chemical No. Chemical No. shift (6) Protons Ass' t shift Protons Ass' t 3.85 1 H-l 4.4 1 • H-l 5.4 1 H-3 5.1 1 H-3 2.3 3 H-4,4',5 2.3 1 H-4,4',5 1.8 3 C—2 Me 1.8 3 C-2 Me 1.1 6 C-6 Me's 1.1 6 C-6 Me's

Figure 3. 100 MHz NMR Spectrum of C.. _H.. .0o Lactone Mixture (in CC14) 10 14 2 7

These are biogenetically likely structures because they are simple Baeyer-Villiger oxidation products of filifolone, another A. filifolia constituent. These structures are also

consistent with Tammami's observation that hydrogenation of the lactone mixture yielded a neutral and an acidic product.

The latter presumably had arisen by hydrogenolysis of the allylic lactone 2, whereas 1 is non-allylic and was assumed 1 mf m to have undergone simple hydrogenation. Strained allylic lactones such as (W2, e.g.,' 3 and 4frfi (Meinwald,' Seidel,* and Cadoff, 1958) are known to undergo ready and nearly

jut>° (xy°

3 4 m m exclusive hydrogenolysis in preference to simple addition of hydrogen.

The present work on Artemisia filifolia began with the demonstration that the proposed structures 1 and 2 for r c mm the two lactones were incorrect. A literature search revealed that structure 1 is, in reality, carvenolide, originally prepared (Wallach, 1899) long ago. A comparison of the NMR spectrum of carvenolide (Wolinsky, 1971) with that of the A. filifolia lactone mixture demonstrated that neither lactone is carvenolide. The NMR spectrum of carvenolide and those of all other compounds mentioned in 8 this dissertation may be found in Appendix A. Furthermore, structure 2 (as yet unreported) was prepared as shown in

Figure 4. Again, an NMR spectral comparison revealed that neither A. filifolia lactone is 2.

When an attempt was made to reproduce the results of Tammami (above) by hydrogenation of the A. filifolia lactones, surprisingly both lactones underwent hydrogenolysis to yield a mixture of two unsaturated carboxylic acids of formula cioH16®2* 0f tlie ac^^s with this molecular formula, a-cyclogeranic acid (5) seemed likely because it could explain the unusual occurrence of isophorone in the plant.

That is, a-cyclogeranic acid would arise by hydrogenolysis of lactone structure 6:

COOH

The biogenetic route to isophorone in the plant could be as represented below:

2 isophorone 9

KOH CN (+)-fenchone

OOH

•OMH,

COOH OH HCO- OOH CONH.

oc-fencholenic acid P-fencholenic acid

COOH NlaOAc Ac,P filifolone

geranic acid

*DBN = 1,5-diazabicyclo[4.3.0]non-5-ene =

Figure 4. The Synthesis of Lactone Structure 2 10

The elimination of carbon dioxide from the keto-acid 7 M\ (above) is very feasible because this is the Michael analog of the well-known facile elimination of carbon dioxide from

f3-keto-acids. To test this hypothesis, a-cyclogeranic acid

(5) was prepared from geranic acid (8) (Mondon and Teege,

1958; Stork and Burgstahler, 1955):

COOH

COOH BF3*Et20

MeOH

8 5 m

The a-cyclogeranic acid (5) prepared as above was compared by NMR to the two acids of molecular formation ciqH16^2

which arose by hydrogenolysis of the A. filifolia lactones and found to be different from both.

Finally, the structures of these two carboxylic acids were deduced by identity of their NMR spectra with those reported (Erman, Wenkert, and Jeffs, 1969) for 9 m and

COOH COOH

9 10 11 m N/M mm 11

The mixture of 9AM and 10nMAM derived from the A.^ filifolia lactones was re-subjected to hydrogenation to produce only cis-2.2.4-tri-methylcvclohexane-l-carboxvlic acid (11).

This showed that the A. filifolia acids were simply double bond isomers. Compound 11AM AM was prepared as in Figure 5 and shown to be identical to the saturated acid prepared from the A. filifolia lactones by spectral comparison and an undepressed mixed melting point. These data indicate that the A. filifolia lactones must be represented by structures

4J2tj2t £ 13 14 15 16 mm mm mm mm

Lactones 12 and 13 would produce acids 9 and 10. respec- mm mm c m\ mm' c tively, upon hydrogenolysis. Lactones 14, 15, and 16 would J ' c •* 3 J mm' mm' mm also produce 9 and/or 10 upon hydrogenolysis, but can be AM M\AM ruled out because they are, respectively, 6-, 6-, and P- lactones, none of which fits the infrared data (1770 cm"''") of the A. filifolia lactones.

The two lactones from Artemisia filifolia were finally separated from each other by saponification and fractional crystallization of their hydrolysis products, 17 12

COOH

COOH 11 chrysanthenone AMAM

Figure 5. The Preparation of Carboxylic Acid 11 MAMA 13 and 18. The latter could then be heated to regenerate the

OH /H2 +

COOK

12 13 17 18 mm mm mm mm

lactones. The IR and NMR spectra of the pure lactones are

presented in Figures 6 and 7, respectively„

Once separation had been achieved, structure 12 was AM AM shown to be identical to one of the A. filifolia lactones.

For example, methanolysis of one lactone produced 19. The

allylic alcohol 19 could easily be oxidized to the ot,P-

unsaturated ketone mm20 with activated manganese3 dioxide:

NaOMe MeOH :00M«. OOM«. 12 19 20 mm mm mm

The actual synthesis of 12 was accomplished in two ways. AMMN The first is shown in Figure 8. Hydrolysis of chrysanthenone

epoxide has been shown (Chretien-Bessiere and Retamar, 1963)

to give 17. Structure 17 was shown to be identical to the ^ mm mm hydrolysis product of one of the A. filifolia lactones by spectral comparison and an undepressed mixed melting point. 14

Wavelength (microns)

1000 9oo 800 Frequency (cm""'')

Wavelength (microns)

3 s 1 9 M <3

4000 3000 looo 1600 000 7oo Frequency (cm~^)

Figure 6. Infrared Spectra of Pure 12 and 13 /MWA MM 15

J\ f\s_ AJ I I I I I i 1 I pprv*(5} 8 7 6 5 4 3 2 O

_/v_ A_ v 4^ I i * i i i pr (S) a 7 4-3 1 I o

Assignments:

Chem. No. Multiplicity Ass 11 Shift (6) Protons (J.Hz) Ass 11 1.08 6 s* C-2 Me1s 1.80 3 d (1) C-4 Me 2.3 3 m H-1,6,6' 4.40 1 m H-5 5.10 1 br s H-3

1.10 3 s C-2 Me 1.20 3 s C-2 Me 1.80 3 s C-4 Me 2.3 3 m H-1,6,6' 3.87 1 s H-3 5.43 1 m H-5

*s = singlet, d = doublet, m = multiplet, br = broad.

Figure 7. 60 MHz NMR Spectra of Pure Lactones 12 and 13 (in CC14) 16

hv m-Cl—C^H.-CO-H -O verbenone chrysanthenone chrysanthenone epoxide

OH/EtOH

cooh

Figure 8. A Synthesis of Lactone 12 17 The acid 17 from chrysanthenone epoxide could be heated to

give lactone 12, identical to one of the A. filifolia AM/M —— •—- "-HI I •• lactones.

Interestingly, chrysanthenone epoxide, when heated

to 160-170°, was converted directly and quantitatively into

lactone 12. This unprecedented thermal rearrangement, if

concerted, would be a [^g + a2g] cycloaddition type reac

tion and, according to the Woodward-Hoffmann rules

chrysanthenone 12 mm epoxide

(Woodward and Hoffmann, 1970, p. 70), symmetry-forbidden.

Hence, the reaction must proceed by a non-concerted

mechanism, some of which are presented in Figure 9. This

type of cycloaddition has recently been utilized by Carlson

(1973) in the synthesis of macrolides. However, Carlson's

work employs photolysis to bring about the reaction, e.g.:

+ other products 18

Figure 9. The Thermal Rearrangement: Possible Mechanisms 19 This is entirely plausible, because the [ 2 + 2 ] cyclo- ' o s o s~ J addition is predicted by Woodward-Hoffmann rules to be

allowed in the excited state (i.e., photochemically). The

photochemical rearrangement (Carlson, 1973) is probably

either concerted or involves a stepwise version of

mechanism c (Figure 9):

>=< ft • -V- -°

The second synthesis of lactone 12 consists of one J mm step, utilizing the carboxylic acid 9 (prepared as in Figure

5) as starting material. This preparation is presented in

Figure 10. This reaction presumably proceeds via allylic

bromination of 9 to give the bromo-acid 21 (not isolated) m a mm

which immediately undergoes an internal Sn2 displacement

yielding directly ^2. Lactone 12 prepared by this route

was also identical to one of the A. filifolia lactones.

The independent synthesis of lactone MA13 AM proved to be considerably more difficult than that of 12. The route ffnrm which was finally successful was analogous to chrysanthenone

epoxide —> 17 and chrysanthenone epoxide —12 (Figure 8).

This route is shown in Figure 11. The epoxide 23 was 3 • mm prepared from 2,4,4-trimethyl-bicyclo[3.1.l3hept-2-en-6-one 20

cooh

HBr

12 mm

Figure 10. A Second Synthesis of Lactone 12 21

hy y / Iciu

o chrysanthenone verbenone

cojh

OH --O o

COOH

V- AWVA24

13 mm

Figure 11. The Synthesis of Lactone 13 NAM 22 (2^) by peracid oxidation. The ketone 22 was prepared by

photolysis of chrysanthenone (Erman, 1967). Epoxide 23

could be hydrolyzed to the acid 18. identical to the

hydrolysis product of one of the two A. filifolia lactones.

The epoxide 23 was also found to undergo the same thermal

rearrangement (not concerted) that was seen for

chrysanthenone epoxide. This resulted directly in lactone

identical to one of the A. filifolia lactones.

This preparation was complicated by several diffi

culties. Firstly, the ketone 22 had to be separated from

unreacted chrysanthenone, from which it was prepared.

Chrysanthenone and 22 were found to be separable only by

careful and laborious preparative GLC. Moreover, this had

to be carried out at temperatures below 120° due to the

thermal lability of the two ketones. The second difficulty

encountered in the synthesis of lactone 13 is as follows. J mm Although chrysanthenone reacted smoothly with peracid to

give only one epoxide product, the ketone AM22 M gave both possible stereoisomers, 23 and 24. Due to their extreme 1 ' Mm mm thermal lability, 23 and 24 could be separated only by very

careful preparative GLC. Other means failed to accomplish

separation. The above difficulties in purification of

intermediates made the synthesis presented in Figure 11 of

lactone 1^ tenable only on a very small scale. For this

reason, although the above preparation was successful, alternate means of preparation of lactone 13 were sought,

all of which, in the end, failed.

The first alternate route selected for the synthesis of /IMAM13 is presented in Figure 12. This is analogous to the

preparation of lactone MA12 AM represented in Figure 10. A Lewis acid-catalyzed Diels-Alder reaction was selected for the preparation of acid 10 because from steric considerations a AM NV\ thermal Diels-Alder reaction would probably give the product of incorrect orientation:

However, the Lewis acid-catalyzed reaction was totally unsuccessful using both senecioic acid (25) and its ethyl ester. It has been noted (Fieser and Fieser, 1967, p. 32) that Lewis acid-catalyzed Diels-Alder reactions are successful with only a small group of dienophiles, notably acrolein and acrylate esters.

A second unsuccessful preparation of lactone 13 can AM MA be seen in Figure 13. This performic acid oxidation reaction has been shown (March, 1968, p. 617) to proceed via an epoxide (26 in this particular case) which is then opened by MAMA

S,.T2 attack by a molecule of water to give the diol (27 in JM MAAM 24

naoci

OOH

Figure 12. The First Unsuccessful Synthesis of Lactone 13 •* AAA AAA 25

HCOOH

COOH cooh

Figure 13. The Second Unsuccessful Synthesis of Lactone 13 26

this case). Unfortunately, the diol 27 reacted further to AWM\ give 28 instead of the desired 13. ^ /maw mm Still a third unsuccessful preparation of 13 was

designed after the preparation of lactone 2 (Figure 4),

utilizing iodolactonization. This sequence is presented in

Figure 14. Unfortunately, the iodolactone formed from the

acid 9 was not the desired 29, but instead 30„ Reaction of m mm' mm 30 with 1 5-diazabicyclo[4.3.0]-non-5-ene (DBN) in refluxing

toluene afforded only 9.AW Finally, Baeyer-Villiger oxidation of chrysanthenone,

under a variety of conditions (see Experimental Section)

failed to provide jL3, the only product being chrysanthonone

epoxide.

The A. filifolia lactones, now shown to be 12 and — ' mm 13, were found to be easily interconvertible. Interestingly,

when the lactone mixture was isolated by preparative GLC

(above 170°) of the steam distillate of Artemisia filifolia. the two lactones were always obtained in a 1:1 ratio. How

ever, when the lactone mixture was isolated in other ways

(fractional distillation or column chromatography of the steam distillate), the ratio of 12:13 in the mixture was ' mm mm approximately 5:1. These ratios were obtained by integra tion of NMR spectra. These data make one suspect that 12 and 13 equilibrate upon gas chromatography: 27

ooh

1>BN

Figure 14. The Third Unsuccessful Synthesis of Lactone 13 MMi 28

GLC

> 160

12 13 m/YA mm

Indeed, when pure 12 was subjected to preparative GLC a 1:1 AW AW mixture of 12 and 13 was collected. However, simple heating MAM AM AM

of lactone /M"A12 to 170 does not produce this result. In addition, the hydroxy-acid derivatives of

lactones 12 and 13 (17 and 18. respectively) are readily AW AW AW AW AW AW AW AW' ^ J J

steam

x / \ 12 13 mm mm

interconvertible» Pure 17, upon steam distillation, with or mm ' without added mineral acid, resulted in the isolation of a

1:1 mixture of lactones 12 and 13 in the distillate. mm mm Several equilibria may be operating here. The hydroxy-acids

17WW and 18AWWA could equilibrate under these conditions, as could 29 lactones 12 and 13. To examine this, pure 12 was steam rrArtn rfANA * WiAW distilled and was recovered unchanged in the distillate.

Therefore, 12 and 13 do not directly equilibrate upon steam trArm rmrfn distillation; rather, there is a rapid equilibrium operating

between their hydroxy-acid precursors, 17 and 18. This u c ' mm mm behavior is to be expected of such allylic alcohols,

especially those which are also carboxylic acids:

± HjP.

COOH

± Hz0

COOH OOH

The above data lead to the conclusion that the

lactones 12 and 13 are not artifacts of the steam distilla- mm mm tion used in their isolation. If they were artifacts

arising from 17 and/or 18, they should be obtained in a 1:1 3 mm mm' J ratio. They are, however, as mentioned previously (unless subjected to preparative GLC), obtained in a 5:1 ratio.

Moreover, chloroform extraction of Artemisia filifolia and

subsequent chromatography failed to provide either 17MAlWV or 18AAAMA 30

The A. filifolia lactones 12 and 13 are optically 1 AMAM AM AM active: [<*3^ -33.2° and +43.8°, respectively. The

determination of their absolute stereochemistry was under

taken. The absolute stereochemistry of (-)-chrysanthenone,

obtained by photolysis of (-)-verbenone, is known (Erman,

1967). Since lactone 12 can be prepared from chrysanthenone AM AM (Figure 8) and the absolute stereochemistry of the latter

is known, then the absolute stereochemistry of the former

can be determined by utilizing optically active

chrysanthenone in the preparation of optically active

12. That is, (+)-chrysanthenone should give, according to

the reaction scheme presented in Figure 8, lactone 12 of absolute stereochemistry shown below:

(+)-verbenone (+)-chrysanthenone (-)-chrysanthenone 12 epoxide

The optically active 12, prepared in this way, if of the same absolute stereochemistry as the lactone 12 from A. mm — filifolia. will have a rotation in the same direction as the latter. If synthetic 12 is of opposite absolute stereo chemistry, the rotations of natural and synthetic samples will be in opposite directions. 31

Most photolysis experiments described in this

dissertation were carried out at a temperature of ca. 50°.

This was due to the heat generated by the mercury arc lamp

employed. Unfortunately, these high temperatures resulted

in racemization of the chrysanthenone prepared from

verbenone (see Experimental Section) by photolysis. A special experiment was therefore designed in order to obtain optically active chrysanthenone. A Hanovia 450 broad spectrum mercury arc lamp was cooled with a large fan so

that the temperature of the photolysis vessels dropped to

34°. Utilizing these conditions, (+)-verbenone (partially active, [a]^ +93.7°) gave (+)-chrysanthenone (partially active, +31°). The latter was converted, according to the scheme of Figure 8. to optically active 12, [

( + »-U.[< +8-2° =

toon (+)-17 (+)-chrysanthenone (-)-chrysanthenone AMM epoxide

Since Artemisia filifolia contains (-)-12, which can be 1 - /IM/m hydro ly zed to ( + )-jjL^, the absolute stereochemistries of these two compounds ,are as shown above. 32

Because lactones 12 and ^ are.interconvertible (by

GLC) then the absolute stereochemistry of 13 from A. mmAAA AAA mm filifolia must be as shown below:

> 170 QLC

(-)-12 (+)-13 mm mm

That the thermal isomerization product (13) of (-)-12 is r mm mm identical to (+)-13 from A. filifolia can be seen by the mm — •* positive specific rotation (+6°) of the 1:1 (approximately)

mixture of 12MAM and 13.Wrt.W isolated by preparative GLC of pure 12 ([a]^ -33.2°) from Artemisia filifolia. This is because

(+)-13 has a larger specific rotation (+43.8°) than (-)-12 mm 13 mm (-33.2°).

Minor Monoterpene Constituents of A. filifolia

The steam distillate of Artemisia filifolia was re

examined by preparative GLC for minor monoterpene constitu

ents, especially possible intermediates in the biogenesis of

isophorone and the lactones 12 and 13. It has been mentioned

previously that this method of analysis had already demon

strated the presence of (-)-camphor, 1,8-cineole, (-)-

filifolone, isophorone (Onore, 1967). and lactones 12 and 13 ' r ' ' mm mm in A. filifolia. Four other monoterpenes were isolated in

very small amounts by preparative GLC. Figure 15 shows a 33 5 Solvent

4 6 8 XO 12. Retention Time (R^.), minutes

Relative Intensity Rt(min) Compound (by peak area)

0.9 1,8-cineole 35 2.0 filifolone 10 2.6 camphor 30 3.2 isophorone 40 4.0 borneol+34 , mm 2 4.7 verbenone 1 8.6 piperitenone 2 10.2 12 + 13 mm mm 10

Figure 15. Typical GLC Chromatogram of the A. filifolia Steam Distillate — 20% Carbowax on Chromosorb P (acid washed)—h" x 5'. Flow rate = 100 ml/min; column = 178°, detector = 220°, injector = 210°. 34

typical chromatogram of the A. filifolia steam distillate.

Two of these monoterpenes were preliminarily identified by

NMR and IR spectroscopy (see Appendices A and B) as (-)-

verbenone (^1) and piperitenone (32). Confirmation of

identity was made by mixed melting points of derivatives

with derivatives of authentic samples.

A third minor GLC peak was collected preparatively

It was deduced by NMR spectroscopy to consist of borneol

(^) and apparently one other compound. The borneol was

allowed to sublime off of the sample mixture. The remaining

liquid was found by mass spectroscopy to have a molecular

weight of 152. Elemental analysis provided a molecular

formula of C9Hi2°2* strong infrared absorption at 1680 and

1625 cm""1' and an ultraviolet absorption at 238 nm indicated

an a,P-unsaturated ketone. The NMR spectrum (Figure 16)

revealed the presence of the following atomic groupings

(assignments may be found in Figure 16):

cW O 0 3 ii n ' JL CH3 — C-CHX- 35

A -J h- 4 i I I i I I PP" (5) 7 fc 4- 3 2. O

.<2

34mm (ketoisophorone)(3,5,5-trimethylcyclohex-2-en-l,4-dione)

Chemical Shift (6) No. Protons Assignment

1.18 6 C—5 Me's 1.98 3 C-3 Me 2.63 2 H-6, 6' 6.46 1 H-2

Figure 16. NMR Spectrum of Ketoisophorone (34) (in CCl^) 36 These features may be accommodated bv structure 34 J mm ("ketoisophorone") (Isler et al., 1956; Wada, 1964):

34 mm

This compound was prepared according to the procedure of Isler (1956) as in Figure 17. Identity of natural and synthetic samples of 34 was shown by superimposable IR and

NMR spectra and an undepressed mixed melting point of a simple derivative.

Both isophorone and ketoisophorone (34) have been found but once previously in nature; they were isolated from saffron (Zarghami and Heinz, 1971). Saffron is a spice composed of the dried stigmas of Crocus sativus L.

Other Constituents of A. filifolia

In order to determine whether Artemisia filifolia contains carboxylic acid (non-steam volatile) precursors to the lactones 12 and 13, and also to verify that the latter mm mm' J are not artifacts of the steam distillation used in their isolation, the plant was dried, ground, and extracted with chloroform. The chloroform extract was then separated into neutral and acidic components by extraction thereof with

5% sodium hydroxide. 37

MeMgl, FeClg ©, then H7H20 isophorone P-isophorone

HCOOH H2°2

MnO,

Figure 17. The Preparation of Ketoisophorone (34) 38

The acidic portion was subjected to column

chromatography. The hydroxy-acid derivatives of the A.

filifolia lactones 12 and 13 (17 and 18. respectively) were mm mm mm mm' ^ *

X>H

&COOH cooh 13 17 18 mm12 mm mm mm not found to be present. However, one carboxylic acid, biogenetically related to these lactones, was isolated.

This was 5-keto-2,2,4-trimethyl-cyclohex-3-ene-l-carboxylic acid (35). The keto-acid 35 was prepared from 17 by mm ^ ^ mm, J HO.6 "6 COOH coom

(+)-17 (+)-35 mm mm manganese dioxide oxidation. Superimposable IR and NMR spectra and an undepressed mixed melting point indicated that the natural and synthetic materials were identical.

The carboxvlic acid 35 isolated from A. filifolia was J mm — optically active, +128°. That the absolute stereo chemistry of (+)-35 is as shown above can be seen by the observation that manganese dioxide oxidation of (+)-17, a 24 o [a]n +33.8 , from A. filifolia. with absolute 39 stereochemistry shown above (see Chapter 1), gave (+)-35,

[a]£4 +128°.

Also isolated from the acidic extract of A. filifolia by column chromatography was the flavone acacetin (36). Preliminary identification was made by an examination of the NMR (Figure 18) and ultraviolet

(Appendix B) spectra and the structure was confirmed by independent synthesis (according to Robinson and

Venkataraman, 1926):

OH Mei O

H -tKeh KOH/Eton/A OH

Undepressed melting points were obtained for mixtures of natural and synthetic samples of both acacetin and its diacetate.

The neutral chloroform extract of A. filifolia was also analyzed by column chromatography. Both lactones 12 and 13 were found therein. Also present was a crystalline compound of molecular weight 252 (mass spectroscopy). The elemental analysis was consistent with a molecular formula of ci5H24°3* Infrared data (3509, 1778 cm~"L) suggested the 40

Jl A- I I l i i ppnn(S) s 4 3 Z o

HO (f V^-OtAc. T T lT\=/ 1' 3'

Chemical Multiplicity Shift (6) (J. Hz) No. Protons Assignment

3.73 s* 3 -OMe 6, 56 d (1.5) 1 H-6 6.69 d (1.5) 1 H-8 6.78 1 H-3 7.00 d (9) 2 H-31,5' 7.87 d (9) 2 H-21.6'

*s = singlet, d = doublet

Figure 18. NMR Spectrum of Acacetin (36) (in pyridine-d^) 41

presence of a hydroxyl group and a Y-lactone. An examina

tion of the NMR spectrum (Figure 19) led to the speculation that this compound was the sesquiterpene lactone colartin

(Irwin and Geissman. 1969) (37): ' AAA AAA

H0-'

An authentic sample (Geissman, 1973) of colartin was obtained, and identity with the A. filifolia material was shown by superimposable infrared spectra and an undepressed

mixture melting point. The specific rotation of colartin from A. filifolia (+ 19.4°) was close to that (+ 11.4°) of

Geissman's material. Hence, the absolute stereochemistries of the two samples are the same. 42

i i I I . I o PP («) 8 6 3 Z

Her

Chemical Multiplicity Shift (6) (J. Hz) No. Protons Assignment

0.99 sa 3 C-10 Me 1.20 d (7) 3 C-ll Me 1«33>> s 3 C—4 Me 2.93 m 1 -OH 4.06 m 1 H-6

s = singlet, d = doublet, m = multiplet.

'Disappears when D2O added.

Figure 19. NMR Spectrum of Colartin (37) (in CDC1,) BIOSYNTHETIC IMPLICATIONS

The biogenesis of monoterpenes has been discussed

many times (cf., Geissman and Crout, 1969, pp. 233-266;

Newman, 1972, pp. 350-361). There is no need here to repeat

the well-known pathway leading to the formation of

isopentenyl pyrophosphate (38) and dimethylallyl pyrophos

phate (39). In the biosynthesis of nearly all monoterpenes

38 and 39 are presumed to combine in a head-to-tail fashion tfnrrn frnrrn to form geranyl pyrophosphate (40) (Figure 20). Geranyl MA AM pyrophosphate is then assumed to undergo whatever cycliza-

tion and/or oxidation necessary for the formation of the

monoterpene in question. Indeed, such a conventional

scheme is quite adequate to explain the biogenesis of six

of the A. filifolia monoterpenes: borneol, camphor, 1,8-

cineole, piperitenone, verbenone, and filifolone. Such

routes are presented in Figures 20 and 21.

One comment on the possible biogenesis of filifolone

(Figure 21) should be made here. The proposed ketene

intermediate 41 is quite logical for the following reasons.

Firstly, this is the same intermediate that has been pro

posed (Beereboom, 1965) for the chemical synthesis of

filifolone by base-catalyzed cyclization of geranic acid.

Secondly, distillation of filifolone results in some thermal

decomposition, the product being isogeranic acid. The

43 44

oPP

borneol :OH piperitenone CO|

camphor 1,8-cineole

Figure 20. Classical (Head-to-Tail) Biogenesis of Borneol, Camphor, Piperitenone, Verbenone, and 1,8- Cineole 45

H40 PPi

OPP OH

40 NMNi geraniol

^ ^ ISOMGRIZE ^ f-03 .ooh -< coon "*^~ HO

isogeranic geranic geranial acid acid

HzO

ryUo

41 filifolone mm

Figure 21. Classical (Head-to-Tail) Biogenesis of Filifolone 46

latter is the compound which would arise by addition of

water to ketene intermediate 41:

NlttOAc

geranic acid isogeranic acid filifolone

A conventional scheme (involving geranyl pyrophos-

phate as the key intermediate) must be stretched a bit to allow for the formation of the five remaining A. filifolia monoterpenes.c These are the lactones 12mm and 13,mmi isophorone,r i the ene-dione 34, and the carboxylic acid a hypothetical conventional biogenesis of them is depicted in Figure 22.

The photochemical transformations shown in Figure 22 are credible because there is ample sunlight in the desert regions where the plant grows, and these same reactions can be reproduced in the laboratory. Filifolone is included in

Figure 22 because it, too, is a photolysis product of verbenone (see Experimental Section). The verbenone iso lated from A. filifolia is (-)-verbenone. The proposed biogenesis of lactones 12 and 13, 35 and filifolone pre- ^ mm mm' mm ^ sented in Figure 22 is therefore unlikely for the following reasons. Firstly, (-)-verbenone has been shown (Erman, 47

verbenone

40 mm

13 filifolone mm

OOH

WjO

HO" OOH COOH OOH 35 mm isophorone CO-

COOH

Figure 22. Classical (Head-to-Tail) Biogenesis of 12, 13, Isophorone.~ ' 34,MAMA' 35,AM AM' and Filifolone 48

1967) to give (+)-filifolone on photolysis. Since Artemisia

filifolia contains (-)-filifolone the latter could not have

arisen by photolysis of (-)-verbenone. In addition, inde

pendent preparation from (+)-verbenone gives a product

((-)-12) with absolute stereochemistry the same as that of M\MI\

/IMAM12 isolated from A. filifolia.1 Since this *preparation (Figure 8) parallels the biogenetic scheme of Figure 22,

(-)-verbenone is not the biogenetic precursor of (-)-12;

nor is it the precursor of (+)-13 or (+)-35 because both of c mm mm these compounds can be prepared from (-)-12.

The last-mentioned five monoterpenes can all be seen

to have or be derivable from a precursor (e.g., 43) with the AM AM carbon skeleton 42: mm

COOH COOH

12 13 isophorone 34 35 42 43 mm mm mm mm mm mm

The carbon skeleton 42 has been found previously in nature exclusively in the plant family Umbellifereae. in eight compounds characterized by Bohlmann (Bohlmann and

Zdero, 1969; Bohlmann and Grenz, 1970): 49

H R. % ?, i •T& CHO

R = -H R = -OH R ~ ~H CH_OAc | ^yU R = -OH R = -OAc R = -COC=C^H R = -OAc 3

Lavandulol (44) (Naves, 1960) is a monoterpene with AM AM ' * an unusual carbon skeleton whose biogenesis cannot easily be rationalized as proceeding from geranyl pyrophosphate (40):

It has been proposed (Richards and Hendrickson, 1964, p. 212) that lavandulol arises biogenetically by a non-head-to-tail coupling of isoprene units. This is presented in Figure 23, which also includes the proposal that thiamine pyrophosphate

(TPP) plays a catalytic role in the coupling process. The latter has been suggested (Geissman and Crout, 1969, p. 255) because TPP has been demonstrated to catalyze the dimeriza- tion of farnesyl pyrophosphate units in a tail-to-tail manner in squalene biosynthesis. 50

tj PPi HiO IPP ISOMCRIZE

Figure 23. Proposed Biogenesis of Lavandulol (j^) Recently, numerous reports (Popjak et al., 1969;

Rilling et al., 1971; Altman, Kowerski, and Rilling, 1971;

van Tamelen and Schwartz, 1971; Coates and Robinson, 1972;

Poulter et al., 1972) have been made of the role of a

cyclopropylcarbinyl cation as the key intermediate in the

tail-to-tail coupling of two farnesyl pyrophosphates in

squalene biosynthesis. The same type of intermediate (45)

has been proposed (Bates and Paknikar, 1965; Bates and

Feld, 1965) as the key intermediate in the biogenesis of

certain monoterpenes which are apparently not geranyl

pyrophosphate derived. Three such monoterpenes are

artemisia ketone (46) (Zalkow, Brannon, and Uecke, 1964;

Takemoto and Nakajima, 1957), the Santolina chamaecyparris-

sus hydrocarbon 47 (Thomas and Willhalm, 1964), and esters of chrysanthemic acid (48) (Bohlmann, Bornowski, and Arndt,

1963). Figure 24 shows the proposed biogenesis of the

cyclopropylcarbinyl cation 45 and the manner in which the

latter is presumed to proceed to 46, 47, and 48. This nW,W\ AMMA MAAN\ scheme incorporates all features analogous to those of squalene biosynthesis. Figures 23 and 24 show, however,

that jfavandulol (44) is apparently not derived from the mm J cyclopropylcarbinyl cation 45; but rather from its pre

cursor (49). mm In any case, the same non-head-to-tail coupling of

isoprene units which explains the biogenesis of lavandulol

(i.e., via intermediate 49) can be quite attractively OPOjHL

(lavandulol) COOR.

Figure 24. Proposed Biogenesis of Artemisia Ketone (46), Santolina Hydrocarbon (47), Chrysanthemic^cid EstersC48) and Lavandulol (44) 53

applied to the biosynthesis of the five A. filifolia

monoterpenes of unusual structure. This proposed

biogenetic pathway is presented in Figure 25, and is largely

self-explanatory.

It should be noted here that isophorone and 34 ^ mm isolated from saffron probably do not arise biogenetically

in the same way as do isophorone and 34 from A. filifolia.

That is, in Crocus sativa these two compounds are apparently

derived from the major constituent safranal (or its pre

cursor):

HO OOH

safranal isophorone 34 c mm

Although it is not necessary that a single biogenetic

pathway be operating in a given plant, Figure 26 demon

strates that verbenone, piperitenone, borneol, 1,8-cineole,

and camphor can also be derived via the lavandulol route

(i.e., from intermediate 49). ' mm The biosyntheses of flavones (Richards and

Hendrickson, 1964, pp. 155-164) and sesquiterpenes (Clayton,

1965) are well documented, and only brief summaries of each are presented in Figures 27 and 28. These two charts represent proposed biogeneses of the flavone acacetin (36), 54

ppo' PP' 49 nm

HO'

.'ooh :00h :ooh

isophorone :ooh oom

34 13 12 mm AMM mm

Figure 25. Lavandulol-like Biogenesis of 12, 13, Isophorone, 34, and 35 55

PP( 49 OPP mam

PPO

t o.s in Figure 20

•OH piperitenone 1,8-cineole verbenone borneol

camphor

Figure 26. Lavandulol—like Biogenesis of Piperitenone, Verbenone, Camphor, Borneol, and 1,8-Cineole 56

o o N«i ;QOH it. c-hichcooh WOOCv chjxcooh ch2ccooh

6h oh phenyl- phenylalanine shikimic acid prephenic acid Pyruvic acid Nrt,

3 Acetates or 1 Acetate + 2 Malonates

cinnamic acid jD-coumaric acid oh .OH

OOH

oh OH

OH OMe S-adenosyl methionine

m o Acacetin

Figure 27. Proposed Biogenesis of the Flavone Acacetin (36) MM 57

Figure 28. Proposed Biogenesis of the Sesquiterpene Colartin (37) and the sesquiterpene lactone colartin (37), respectively.

These pathways are self-explanatory. EXPERIMENTAL

Microanalyses were performed by Huffman

Laboratories, Inc., Wheatridge, Colorado and Scandanavian

Microanalytical Laboratories, DK 2730 Herlev, Denmark.

Melting points were determined on a Fischer Mel-Temp

apparatus and are uncorrected. Ultraviolet spectra were

measured on a Cary recording spectrophotometer, Model 14.

All ultraviolet (UV) spectra may be found in Appendix B.

Infrared (IR) spectra were recorded on Perkin-Elmer Models

137 and 337. The listings of infrared bands include those

which are relevant to the structural argument. NMR (nuclear

magnetic resonance) spectra were measured on Varian

Associates A-60-A, T-60, and HA-100 instruments and peak

positions are given in 6-values, using tetramethylsilane as an internal standard. The NMR spectra of all compounds

mentioned in this Experimental Section are included in

Appendix A; the pertinent data may be found therein. Mass spectra were recorded on a Hitachi/Perkin-Elmer Model RMU-

6E Mass Spectrometer. Optical rotations were measured on a Cary Model 60 Recording Spectropolarimeter by scans from

.5800 to 6000 %, selecting the value of 5893 A as the sodium

D line. Gas chromatography (GLC) was carried out on a

Varian Aerograph Model 90-P, using a Varian Aerograph Model

20 strip chart recorder. The columns used for GLC were 15%

59 60 SE 30 on Chromosorb P, 3/8" x 20' aluminum tubing (here

after referred to as Column I); 20% Carbowax 20 M on

Chromosorb W (acid-washed), 1/4" x 5' copper tubing (here

after referred to as Column II); and 18% SE 30 on Chromosorb

P, 1/4" x 6' copper tubing (hereafter referred to as Column

III). The sequence of experiments presented here generally

follows the sequence found in the Results and Discussion

Section of this dissertation.

Collection of the Plant Material

All plant material utilized in this study was

collected about 2 miles northeast of Willcox, Arizona on the

following dates: 10/31/69, 8/6/70, and 9/30/71. The whole

plants were collected except for woody trunks and roots.

The mechanics of collection consisted of shearing off whole

branches and placing them in gunny sacks for transportation.

Steam Distillation of the Plant Material

Whole, wet plant material gathered on 10/31/69 and

8/6/70 was steam distilled as follows. A 2- to 8-kg portion

of wet plant was placed in a ten-gallon milk can and the can

was half filled with water. A special top with a vertical

exit tube was then clamped on the milk can. To this vertical

metal tube were then attached three condensers in series.

The drum was heated with a Meeker burner until about four

liters of distillate (cooled in ice) had been collected.

This process takes about four hours. A visible yellow layer 61 of oil was present on top of the water in the distillate.

The steam distillate was separated into two 2-liter portions and each was extracted with ether (2 x 500 ml). All ether extracts were combined, dried over anhydrous sodium or magnesium sulfate, and the solvent was carefully removed in vacuo to give the steam-volatile oil as the residue. In some experiments chloroform was used in place of ether as the solvent for the extractions. The remainder of the procedure was the same. A total of seven such steam dis tillations were carried out. The pertinent data for each may be found in Table 1.

Fractional Distillation of the Steam Distillate

The steam distillate (steam-volatile oil) of

Artemisia filifolia was distilled in vacuo through vacuum- jacketed Vigreux columns. This was done twice; once with a

250-mm column, and again with a 600-mm column. All perti nent data are included in Table 2.

Hvdrolvsis of the A. filifolia Lactones 12 and 13: J -* — MM MM ' Isolation of Hydroxy-acidsJ 17 and 18 MM Wl M Fraction #6 from the second distillation (see Table

2) of the A. filifolia steam distillate (9.0 g, 80% ^ + 13) was dissolved in ether (100 ml) and the ethereal solution was washed with 5% aqueous sodium hydroxide (2 x 100 ml) to remove isogeranic acid„ After washing with water (100 ml), 62 Table 1. Steam Distillations of Artemisia filifolia

Date of Plant Weight (wet) Weight of Steam- Per Cent Collection of Plant Volatile Oil (by Weight)

10/31/69 3.5 kg 33.8 g 0.97 8/6/70 2.4 kg 16.1 g 0.67a 8/6/70 3.6 kg 53.1 g 1.48 8/6/70b 5.7 kg 51.7 g 0.91 8/6/70b 7.5 kg 56.8 g 0.76 8/6/70b 6.7 kg 54.4 g 0.82 8/6/70b 8.2 kg 57.1 g 0.70 Total 37.6 kg 293.1 g Average 0.77

aApparatus observed to be leaking slightly.

Grateful acknowledgment is made to Mr. A. H. Wilkinson for his assistance in the steam distillation of these batches. 63 Table 2. Fractional Distillation of the A. filifolia Steam Distillate

Fraction # b.p. ( C) Weight (g) Composition0 First Distillation: Pressure = 2mm Hq. Amount Distilled =45 q:

1 24-55 15 Ci:C:F:I 8:4:1:2

2 56-63 5 Ci:C:F:I:L 1:20:2:21:3

3 68-86 4 Ci: C:F:I :L 1:1:1:1:50

4 87-89 4 ca. 98% L pot residue 15 mostly IGA*"

Second Distillation: Pressure = 2mm Hq. Amount Distilled = 166 q:

1 30-46 17.0 Ci:C:F:I 11:1:1:1

2 48-62 44.3 C:F:I 8:1:1

3 64-71 6.3 C:I:L:0 20:20:1:7

4 72-81 9.1 I:L :0 3:3:4

5 84-90 12.2 L:0 3:2

6 91-109 9.0 > 80% L + some IGA

7 110-118 20.5 > 90% IGAb

Determined by GLC and NMR; Ci s 1 8-cineole, C = camphor, F = filifolone, I = isophorone, L = lactones 12 + 13, O = others.

IGA = isogeranic acid, a thermal decomposition product of filifolone. 64 drying over anhydrous magnesium sulfate and evaporation of the ether ill vacuo. there remained 7.5 g of the mixture of lactones 12rrnfrn + 13. NMR integration revealed this mixture to consist of 85% 12 and 15% 13. The latter mixture was then rmtm MAW placed with potassium hydroxide (4.5 g) in water (150 ml) and refluxed for 140 minutes. The reaction mixture was cooled, diluted with water (200 ml), and extracted with ether (2 x 100 ml). The ether extracts were combined, dried over anhydrous magnesium sulfate and concentrated in vacuo to yield 2.9 g of unreacted lactones as a yellow oil.

The still-basic aqueous phase was then acidified with hydrochloric acid and extracted with ether (3 x 100 ml).

The ether extracts were combined, dried over anhydrous magnesium sulfate, and the solvent removed in vacuo. This resulted in 4.5 g of a yellow oil which crystallized on standing. The latter was found by NMR to consist of a 10:1 mixture of 17:18. AW.YA MAMA

Fractional Crystallization of 17. and jj,§

The crude crystalline mixture of 17 and 18 (4.5 g) J MAMA MAMA ^ from above was recrystallized from carbon tetrachloride.

The first crop of crystals (2.3 g) collected was pure 1/7, m.p. 134-6°. Recrystallization from cyclohexane gave colorless plates, m.p. 140-1°. The reported (Chretien-

Bessiere and Retamar, 1963) melting point of racemic MAMA17 is 140°. This sample of 17 from A. filifolia was optically 65 active, [ct]^ +33.8° (c 0.5, CHCl^). The infrared (KBr:

3350; 3000, very broad; and 1685 cm""''), NMR (Appendix A) and mass (m/e 184, M+; 166, - 1^0; 122, - CC^; and 107, base, - (f^O + CO2 + *CH^)) spectra were consistent with structure 17. mm

Anal. Calculated for cioH16°3: C' ^.19 ; H, 8.75. Found: C, 65.20 ; H, 8.68.

The mother liquor from which pure 17 was crystal lized yielded a second crop of crystals (0.8 g). This was shown by NMR to consist of a 3:1 mixture of 18:17. This J mm mm material was recrystallized from carbon tetrachloride, providing 0.4 g of a mixture now (by NMR) 5:1, 18:17. * ^ ^ ' mm mm Another recrystallization from carbon tetrachloride gave colorless prisms (0.26 g) with m.p. 122-34°, which was by

NMR analysis ca. 95% 18. Still another recrystallization, 1 AW*"* this time from benzene, gave pure (> 98%) 18 (0.16 g) as colorless prisms, m.p. 134-6°. The hydroxy-acid 18 prepared in this way was optically active, [a3^ -58.0°

(c 0.9, CHC13). The NMR (Appendix A), IR (KBr: 3230; 2970, very broad; and 1680 cm-"'') and mass (m/e 184, M+; 166,

- HgO; 123; 107, - (*^0 + CC>2 + 'CH^); and 84, base) spectra were consistent with structure 18. AMAflA

Anal. Calculated for C^gH^O^: C, 65.19 ; H( 8.75. Found: C, 64.95 ; H, 8.63. 66

Regeneration of Lactone 12 from the Hydroxy—acid 17 _ turn z i nm. The pure carboxylic acid 17 (1.84 g, 0.0100 mole) M\/M from A. filifolia. m.p. 140-1°, [ct]^ +33.8, was heated

(neat) in a glass tube in an oil bath at a temperature of

160-175° for 60 minutes. After this time, the reaction

mixture was washed out of the tube with ether (50 ml). The

ethereal solution was dried over anhydrous magnesium sulfate

and concentrated in vacuo to provide 1.64 g (0.00988 mole;

98.8%) of the lactone 12 as a pale yellow oil. This material mm r J was homogeneous by GLC and its NMR spectrum showed no

extraneous peaks. Lactone MAMA12 prepared in this way was 4 optically active, [cc]J -33.2° (c 0.4, CHC13). The NMR

(Appendix A), IR (Neat; Figure 6: 1770; and 1670 cm"1) and

mass (m/e 166, M+; 107, base, - (CC^ + 'CH^); 69; and 41) spectra were consistent with structure 12. mm Anal. Calculated for cioH14°2: C' ^2.26 ; H, 8.49. Found: C, 72.12 ; H, 8.41.

Regeneration of Lactone 13 from the Hydroxy-acid 18 a mm 2 mm_ This experiment was performed exactly as was the preceeding experiment. The carboxylic acid 18 (0.100 g, MAMA 0.00543 mole) from A. filifolia. m.p. 134-6°, -58°, when heated at 160-175° for one hour provided, after workup,

0.0901 g (0.00543 mole; 100%) of lactone 13MAMA as a pale yellow oil. This material was homogeneous by GLC and its NMR spectrum showed no extraneous peaks. Lactone 1^ prepared in this way was optically active, [a]^ +43.8° (c 1.2,

CHClg). The NMR (Appendix A), IR (Neat; Figure 6: 1770;

and 1650 cm-'1') and mass (m/e 166, M+; 123, 107, - (CO 2 +

•CH^); 95; 83; and 43, base) spectra were consistent with

structure ^3.

Anal. Calculated for cioH14°2: C» ^2.26 ; H, 8.49. Foun/3: C, 72.19 ; H, 8.51.

Preparation of Fenchone Oxime (Figure 4)

The general procedure of Shriner, Fuson, and Curtin

(1965, p. 289) was followed. (_+)-Fenchone (Pfaltz and Bauer)

(100 g, 0.68 mole) was added to a mixture of hydroxylamine

hydrochloride (100 g, 1.4 mole), potassium hydroxide (400 g,

7.1 mole), 95% ethanol (1 1) and water (100 ml). The

reaction mixture was then refluxed for 2 hours, after which

time thin layer chromatography indicated the reaction to be

largely incomplete. Refluxing was therefore continued for

another 15.5 hours. Still the odor of unreacted fenchone

was present. The solution was cooled, filtered, poured into

water (1.5 1), and extracted with chloroform (2 x 500 ml).

The chloroform extracts were combined, dried over anhydrous sodium sulfate and concentrated under vacuum to yield 68 g of recovered fenchone. The still-basic aqueous layer was then acidified with 6 N hydrochloric acid. When no

precipitate of fenchone oxime appeared, the solution was

extracted with chloroform (2 x 500 ml). The chloroform extracts were combined, dried over anhydrous sodium sulfate

and concentrated in vacuo. The resulting yellow oil (38 g)

crystallized on standing; the crystals of fenchone oxime

were collected by trituration with 95% ethanol and filtra

tion, and were recrystallized from the same solvent. Four

crops of crystals yielded a total of 34.4 g (0.21 mole;

30%) of fenchone oxime.

The unreacted fenchone (68 g, 0.45 mole) recovered above was resubjected to essentially the same procedure.

It was dissolved in 95% ethanol (1 1) and water (500 ml) to which was added potassium hydroxide (128 g, 2.3 mole) and hydroxylamine hydrochloride (112 g, 1.6 mole). The result ing mixture was then refluxed for 45.5 hours. Workup as above afforded an additional 18.0 g (0.11 mole) of product.

These two preparations therefore provided a total of 52.4 g

(0.31 mole; 46%) of (+)-fenchone oxime, m.p. 110-113°.

Recrystallization from 95% ethanol gave fenchone oxime of m.p. 121-3°. The reported (Rappoport, 1967, p. 166) melting point of (jO-fenchone oxime is 123°. The NMR spectrum

(Appendix A) of the material obtained here was entirely consistent with that predicted for fenchone oxime.

Acid Hydrolysis of Fenchone Oxime: cx~ and P-Fencholenic Acid Nitriles (Figure 4)

The procedure of Cockburn (1899), adapted after that of Wallach (1892), was generally followed. Fenchone oxime (50.0 g, 0.30 mole) was added to a solution of 69

concentrated sulfuric acid (10 ml) in water (250 ml). The

resulting mixture was refluxed for 1.5 hours, cooled and

extracted with chloroform (3 x 200 ml). The chloroform

extracts were combined, dried over anhydrous sodium sulfate,

and concentrated in vacuo to yield the crude product as a

yellow oil (56.6 g). The latter was partially distilled

under water aspiration (22 mm). The low-boiling material

(b.p. 36-58°) was mostly chloroform plus some fenchone.

The material remaining in the distillation flask was found

to be high-boiling (b.p. > 178°) and had a tendency to foam

violently. It was therefore not further distilled, and used

as is for the next experiment. This material (41 g, 0.28

mole; 93%) showed infrared absorption at 2250 cm-1, as

expected. Cockburn (1899) reports that the mixture of a-

and (3-fencholenic acid nitriles boils at 219° at atmospheric

pressure.

Selective Saponification of the Mixture of ct- and P-Fencholenic Acid Nitriles: P-Fencholenic Acid and a-Fencholenic Acid Amide (Figure 4)

The procedure of Cockburn (1899) was generally

followed. The crude mixture (purified by partial dis

tillation) of a- and P-fencholenic acid nitriles (41 g,

0.28 mole) was added to a solution of potassium hydroxide

(17 g, 0.30 mole) in 95% ethanol (200 ml). This mixture

was refluxed for 45.5 hours, cooled, poured into water (2 1)

and extracted with chloroform (3 x 500 ml). The chloroform 70

extracts were combined, dried over anhydrous sodium sulfate

and concentrated in vacuo to yield 35.8 g (0.22 mole; 80%)

of an orange oil which later crystallized. Recrystalliza-

tion from 95% ethanol gave colorless needles, m.p. 110-2°.

The reported (Wallach, 1892) melting point of a-fencholenic

acid amide is 114°. This material had infrared (CHCl^:

3530; 3500; 3420; 3180; 1680; and 1600 cm"1) and MNR

(Appendix A) spectra consistent with a-fencholenic acid

amide.

The still-basic aqueous phase, from which a-

fencholenic acid amide was extracted, was acidified with

dilute sulfuric acid and extracted with chloroform (3 x

200 ml). The chloroform extracts were combined, dried over

anhydrous sodium sulfate, and evaporated jLn vacuo to yield

11.3 g (0.07 mole; 25%) of an orange oil. The latter

crystallized on standing, giving needles, m.p. 45-55°.

Recrystallization from 95% ethanol provided 9.6 g (0.06

mole; 20%) of P-fencholenic acid as colorless needles,

m.p. 66-8°. The reported (Cockburn, 1899) value is 72-3°.

This material had NMR (Appendix A), infrared (CHCl^: 3100,

very broad; and 1710 cm-1) and mass (m/e 168, M+; 123,

base, - (H* + CC^); 81; and 67) spectra in accord with

(3-fencholenic acid. 71

Hydrolysis of a-Fencholertic Acid Amide: a-Fencholenic Acid (Figure 4)

a-Fencholenic acid amide was found to be extremely resistant to saponification. The procedures of both Wallach

(1892) and Cockburn (1899) were largely unsuccessful. This experiment was carried out no less than five times, each time only a small amount of a-fencholenic acid being obtained. The five hydrolyses were substantially the same; the procedure for only the most successful is included here.

The reaction times and yields of the other hydrolyses are also presented.

To a-fencholenic acid amide (31 g, 0.18 mole) was added a solution of potassium hydroxide (150 g, 2.7 mole) in methanol (450 ml) and water (150 ml). The mixture was then refluxed for 189 hours, cooled, poured into water (2 1) and extracted with chloroform (2 x 500 ml). The chloroform extracts were combined, dried over anhydrous sodium sulfate, and evaporated jLn vacuo to yield 20 g (0.12 mole; dl%) of recovered crystalline a-fencholenic acid amide.

The still-basic aqueous phase, from which the amide had been extracted, was acidified with 18 N sulfuric acid and extracted with chloroform (2 x 500 ml). The chloroform extracts were combined, dried over anhydrous sodium sulfate, and evaporated in vacuo to leave a residue of crude a- fencholenic acid (10 g, 0.06 mole; 33%) as an orange oil. The results from the other similar experiments are:

moles KOH mole amide reaction time (hr) yield (q) yield (%)

1.5 125 1 3 1.5 189 2 6 15 96 2 6 15 166 2.5 8

The crude a-fencholenic acid from all five experi

ments was combined (total = 17.5 g) and distilled under vacuum (1.7-2.0 mm) through a 200-mm Vigreux column. Pure a-fencholenic acid (10.0 g, 0.06 mole; 9%) was obtained as a slightly yellow oil, b.p. 113-5°. Reported (Cockburn,

1899) are b.p.12 140.5-144.5°, and b.p.760 254-6° (slight decomposition). Beereboom (1965) reports that a-fencholenic acid is a crystalline solid, m.p. 45-6°. None of the material prepared and used in this work was crystalline.

The infrared (Neat: 3100, very broad; and 1705 cm-1), NMR

(Appendix A) and mass (m/e 168, M+; 107; and 81, base, -

("C^H^ + CC^)) spectra of the material prepared here were consistent with those reported (Beereboom, 1965) for a- fencholenic acid.

Anal. Calculated for cioH16°2: C' ^1.39 ; H, 9.59. Found: C, 71.08 ; H, 9.80.

Preparation of Geranic Acid (Figure 4)

The procedure of Howard and Stevens (1960) was followed. Silver oxide was prepared by adding a solution of silver nitrate (236 g, 1.38 moles) in water (500 ml) to a

solution of sodium hydroxide (140 g, 3.76 moles) in water

(500 ml). The mixture was continuously shaken during the

addition to insure complete reaction, and this resulted in a

brown semi-solid suspension of silver oxide. To this

mixture, cooled in an ice-water bath, was added geranial

(cis- + trans-) (100 g, 0.662 mole) in small portions with

stirring. The reaction mixture was stirred for 30 minutes

after addition was complete. The still-basic solution was

filtered to remove silver and then extracted with chloroform

(2 x 250 ml) to remove unreacted geranial. Combination of

the chloroform extracts, drying over anhydrous sodium

sulfate, and evaporation of solvent afforded 14.3 g (0.094

mole; 14.3%) of recovered geranial.

The aqueous phase was then acidified with hydro

chloric acid and extracted with chloroform (4 x 250 ml).

The chloroform extracts were combined, dried over anhydrous

sodium sulfate, and concentrated in vacuo to give crude

geranic acid as an orange oil (134.6 g). The latter was

distilled under vacuum (1.0 mm) through a 300-mm Vigreux

column. This yielded geranic acid (73.5 g, 0.437 mole;

77.5%, based upon recovered geranial) as a pale brownish oil, b.p. 137-147°. GLC analysis and NMR integration

revealed this to be a mixture of approximately 30% cis- and

70% trans-geranic acids. 74 Preparation of (±)-Filifolone (Figure 4)

The procedure of Beereboom (1965) was employed. A mixture of geranic acid (cis + trans) (70 g, 0.44 mole), anhydrous sodium acetate (27.6 g, 0.34 mole) and acetic anhydride (72.5 g, 0.71 mole) was stirred under reflux for

17 hours. The resulting dark brown solution was poured into ice-water (1 1), stirred, and then extracted with chloroform

(3 x 500 ml). The chloroform extracts were combined, washed with 5% sodium hydroxide (2 x 500 ml), washed with water (500 ml), dried over anhydrous sodium sulfate, and evaporated to yield an orange oil (78.0 g). The latter was distilled under vacuum (0.7 mm) through a 250-mm vacuum- jacketed Vigreux column. Four fractions were collected:

Fraction # b.p. (°C) Weight (g) Composition (by GLC. NMR)

1 36-45 1.6 F :jd—Me—a-Me-styrene;10:1 2 48-50 13.0 100% filifolone (F) 3 86-94 14.5 100% piperitenone 4 113-126 18.0 geranic, isogeranic acids pot residue ca. 30 mostly isogeranic acid

The isogeranic acid produced is apparently a thermal decomposition product of filifolone. The (+)-filifolone from distillation fraction #2 (above) (13.0 g, 0.087 mole;

22%) had NMR (Appendix A), IR (Neat: 1770 cm"""'") and mass

(m/e 150, M+; 122, - CO; 107, - (CO + 'CH^; and 80,. base) spectra consistent with those reported (Beereboom, 1965).

Anal. Calculated for <-'io^140: ^9.96 '> 9.39. Found C, 79.86 ; H, 9.52. 75

Hydrolysis of Filifolone: oc-Fencholenic Acid (Figure 4T

The procedure of Beereboom (1965) was followed.

(-j-)-Filifolone (3.0 g, 0.020 mole, prepared as above) was

added to a 15% ethanolic solution (100 ml) of potassium

hydroxide. The reaction mixture was then refluxed for 13

hours, cooled, diluted with water (300 ml), and extracted

with chloroform (2 x 150 ml). The chloroform extracts were

combined, dried over anhydrous sodium sulfate, and con

centrated in vacuo to provide 0.62 g (0.0041 mole; 21%) of recovered filifolone.

The still-basic aqueous solution, from which

filifolone was extracted, was acidified with hydrochloric

acid and extracted with chloroform (2 x 150 ml). Workup as above afforded 2.3 g (0.014 mole; 70%) of crude a- fencholenic acid as a light brown oil. This oil was identical (IR, NMR) to the a-fencholenic acid prepared

(described previously) from fenchone oxime.

Iodolactone of a-Fencholenic Acid (Figure 4)

The procedure of Beereboom (1965) as adapted from van Tamelen and Shamma (1954) was followed. A suspension of freshly distilled a-fencholenic acid (1.0 g, 0.0060 mole) in water (40 ml) was neutralized with 5% sodium hydroxide. To this solution was added sodium bicarbonate (0.30 g, 0.0055 mole) and then iodine (2.0 g, 0.0079 mole) and the resulting suspension was stirred at room temperature for 18.5 hours. 76 The reaction mixture was then extracted with ether (3 x 30 ml). The ether extracts were combined, washed with saturated sodium thiosulfate (4 x 50 ml) until colorless, washed with water (100 ml), dried over anhydrous magnesium sulfate, and concentrated jLn vacuo. This resulted in 1.8 g

(0.0060 mole; 100%) of the crude iodolactone as a pale yellow oil which crystallized after standing for 3 hours, providing cubelets, m.p. 45-55°. Two recrystallizations from pentane gave 1.3 g (0.0044 mole; 13%) of the pure iodolactone as colorless cubelets, m.p. 72-4°. The reported

(Beereboom, 1965) melting point is 74-5°. This material decomposed readily at room .temperature, turning brown in the process. The NMR (Appendix A), IR (CHCl^: 1770 cm~^), and mass (m/e 294, M+; 167, - I*; 139; 107; and 69, base) spectra of this sample were in accord with a-fencholenic acid iodolactone.

In some experiments, the iodolactone of a- fencholenic acid failed to crystallize. The crude oil, however, gave substantially the same results for the dehydroiodination reaction (immediately following) as did crystalline iodolactone.

Dehydroiodination of a-Fencholenic Acid Iodolactone: Preparation of Lactone 2 (Figure 4)

The iodolactone (1.8 g, 0.0060 mole) of a- fencholenic acid was dissolved in toluene (15 ml). To this solution was added 1,5-diazabicyclo[4.3.0]non-5-ene (DBN, 1.5 g, 0.012 mole) and the reaction mixture was stirred under reflux for 50 minutes. The mixture was cooled, diluted with water (100 ml), and extracted with ether

(2 x 100 ml). The ether extracts were combined, washed with 5% hydrochloric acid (2 x 100 ml), water (100 ml), saturated sodium thiosulfate (4 x 100 ml) until colorless, and then water (100 ml). The ether solution was dried over anhydrous magnesium sulfate and evaporated to yield 1.2 g of a yellow oil. NMR analysis revealed this oil to be ca. 80% lactone 2; it was subjected to preparative GLC (Column III, column temperature 170°). The major peak was collected

(there appeared to be some thermal decomposition of 2) resulting in pure lactone 2 as a colorless oil, m.p. about

10° (0.70 g, 0.0042 mole; 70%). The NMR (Appendix A), IR

(Neat: 1760 cm-'1'), and mass (m/e 166, M+; 122, - CC^; 107, base, - (CO2 + an<^ 91) spectra were in accord with structure 2. Also interesting was the similarity of the

NMR spectra of 2, carvenolide (1), and filifolone (see

Appendix A).

2 carvenolide filifolone

Anal. Calculated for cioH14°2: C' ^2.26 ; H, 8.49. Found: C, 72.11 ; H, 8.63. 78

Hvdroqenolvsis of A. filifolia Lactones 12 and 13: — — WW • • WW' Preparation of Carboxylic Acids 9 and 10 c MA AM MA Several similar experiments were performed; there

fore, only one shall be described in detail here. The

products from all such experiments differed only in the

relative ratios of 9:10. That is, lactone 13 underwent w ww ' ww hydrogenolysis more readily than lactone 12. Therefore,

shorter reaction times gave a 9 + 10 mixture richer in acid ^ W WW 10 than did longer reaction times. ww 3

The —A. filifolia lactone mixture (12WW + 13),WW ' cpurified by fractional distillation (Table 2; first distillation,

fraction #3)(1.21 g, 0.00730 mole) was dissolved in absolute

ethanol (10 ml). To this was added 10% palladium on carbon

(0.050 g) and the resulting suspension was hydrogenated at

atmospheric pressure and room temperature for 17 hours. The

amount of hydrogen absorbed (215 ml) calculates for slightly

more than that required (194 ml) for 1 mole H„/mole 12AWM + 13.AM MA The ethanolic mixture was then filtered through Celite, diluted with 5% aqueous sodium hydroxide (100 ml), and

extracted with ether (2 x 100 ml). Sodium bicarbonate solution was found to be an ineffective solvent for these

p,p-di-substituted (sterically hindered) carboxylic acids

(9 and 10). The ether extracts were combined, washed with W WW ' 5% sodium hydroxide (2 x 100 ml), water (100 ml), dried over

anhydrous sodium sulfate, and evaporated to yield 0.224 g of

neutral hydrogenation products. GLC analysis showed this 79

material to consist of cineole, isophorone, camphor, dihydrofilifolone, and unreacted lactones 12 and 13. J ' mm mm The above sodium hydroxide extracts were combined

with that used to dilute the original solution, acidified

with hydrochloric acid, and extracted with ether (3 x 100

ml). The ether extracts were combined, washed with water

(100 ml), dried over anhydrous sodium sulfate, and

evaporated in vacuo to yield 0.987 g of acidic product as a yellow oil. The NMR spectrum of this oil showed it to con sist of a 3:1 mixture of 9:10. The yield of the 9+10 m mm J m mm mixture was calculated to be 0.00528 mole (72.2%). The infrared (Neat: 29 50, very broad; and 1710 cm""'') and mass

(m/e 168, M+; 153, - "CHg? 123; and 107, base) spectra were consistent with structures 9 and 10 and the NMR spectrum m mm c agrees with those reported (Erman et al., 1969) for 9 and 10. mm

Preparation of a-Cyclogeranic Acid (5)

Two preparations of 5 were undertaken. The first, according to Mondon and Teege (1958) was as follows.

Geranic acid (13.0 g, 0.078 mole, prepared as described previously) was cooled in an ice-water bath and then added dropwise to a stirred solution of 98% formic acid (220 ml) and concentrated sulfuric acid (19.5 g). The reaction mixture was stirred at room temperature for 24 hours. The resulting solution was then poured into ice-water (1 1), 80 and extracted with chloroform (200 ml). The chloroform was

extracted with aqueous 5% sodium hydroxide (200 ml) and the latter was acidified and re-extracted with chloroform (2 x

100 ml). The last-mentioned chloroform extracts were combined, washed with water (100 ml), dried over anhydrous sodium sulfate, and concentrated iri vacuo to yield 2.1 g of a dark yellow oil. NMR analysis showed that this oil was approximately 70% a-cyclogeranic acid (5; 10.8%). The contaminating material was not characterized.

The second and much more successful preparation of a-cyclogeranic acid was accomplished according to the procedure of Stork and Burgstahler (1955). Freshly distilled geranic acid (5.5 g, 0.033 mole, b.p. 105-112°/

1-2 mm) was dissolved in dry benzene (10 ml). The resulting solution was then added dropwise at room temperature, with stirring, to a solution of boron trifluoride-etherate

(5.5 g, 0.044 mole) in dry benzene (50 ml), over a period of

10 minutes. After addition was complete, the reaction mixture was heated at 45° for 20 minutes. The resulting mixture was then poured into a separatory funnel containing ice-water (100 ml), shaken, and the aqueous layer drawn off.

The organic layer was then washed with water (100 ml), dried over anhydrous sodium sulfate, and the solvent removed in vacuo. The residue (6.8 g) crystallized after a few hours. Recrystallization from petroleum ether (b.p. 30-60°) gave a total of 5.0 g (0.030 mole; 91%), in two crops, of ct-cyclogeranic acid (5) as colorless needles, m.p. 101-2°.

The reported (Stork and Burgstahler, 1955) melting point is

105-6°. The NMR (Appendix A), IR (CHCl^: 3000, very broad;

and 1710 cm-"''), and mass (m/e 168, base, M+; 153, - "CHg;

123; 111; 107; 81; and 69) spectra indicated that the

product was a-cyclogeranic acid, and also that the latter

was not a product of hydrogenolysis (described earlier) of

the —A. filifolia lactones JM12 + 13.mm

Hydrogenation of the Carboxylic Acid Mixture 9 + 10 from —A.1~ ^^"filifolia: 1 Preparation of the Carboxylic Acid |VAAV\11 The mixture of carboxylic acids of formula cioH16°2' obtained by hydrogenolysis of the A. filifolia lactones

(0.30 g, 0.0018 mole, described previously) was dissolved

in 95% ethanol (10 ml). To this was added platinum oxide

(0.050 g) and the resulting suspension was subjected to

hydrogenation at atmospheric pressure and room temperature

for 48 hours. The amount of hydrogen absorbed (49 ml) after this time was approximately equal to the amount calcu

lated (47.5 ml) for 1 mole H^/mole cioH16°2* A£ter removal of the catalyst by filtration through Celite and evaporation of the solvent there remained 0.29 g of a pale yellow oil

which could not be made to crystallize. An explanation for

the contamination (and therefore failure to crystallize) of

11 is described later in the discussion of its preparation mm c from synthetic 9. This material was purified by accident

in the following manner. 82 The crude hydrogenation product was to be converted

to a solid derivative, its anilide. The procedure of

Shriner et al. (1965. p. 236) was followed. Crude 11 ' c AM/Wt (0.080 g, 0.00047 mole) was added to thionyl chloride (5 ml)

and the mixture was refluxed for one hour. There followed

the addition of aniline (5 ml) and heating for 5 minutes.

The resulting mixture was cooled, diluted with ether (25 ml)

and then water (dropwise, 20 ml) and extracted with 5%

sodium hydroxide (50 ml), then 5% hydrochloric acid (50 ml),

and finally water (50 ml). The ethereal solution was dried

over anhydrous sodium sulfate and evaporated to yield 0.032

g of a brown oil. GLC and IR analysis revealed that very

little, if any, anilide was present in this material. Its

composition was not determined.

The sodium hydroxide extract of the reaction mixture

was acidified and extracted with ether (2 x 50 ml). These

ether extracts were combined, washed with water (50 ml),

dried over anhydrous sodium sulfate and concentrated in

vacuo to yield recovered 11 in the form of 0.055 g (0.00032

mole; 63%) of crystals, m.p. 59-69°. Recrystallization from

acetic acid/water gave pure MAW11 as colorless tiny flakes, m.p. 83-4°. ' The reported (Kotake and Nonaka, 1957) melting

point is 81-2°. The acid prepared in this way had NMR

(Appendix A), IR (CHCl^: 2960, very broad; and 1710 cm"'1'),

and mass (m/e 170, M+; and 83, base) spectra in accord with 83

structure 11.. In addition, 11 was independently prepared

(described later) and shown to be identical to the material

described here by identical spectra (IR, NMR) and an

undepressed (82-4°) mixture melting point.

Preparation of 2 4-Dimethyl-l.3-pentadiene (Figure 5)

A published procedure (Jitkow and Bogert, 1941) was

employed. Methyl magnesium iodide was prepared from methyl

iodide (160 g, 1.1 mole), anhydrous ether (240 ml), and

magnesium turnings (28 g, 1.2 mole) in the usual manner.

When nearly all of the magnesium had gone into solution, the

mixture was cooled to 0°, and a solution of mesityl oxide

(112 g, 1.3 mole) in dry ether (100 ml) was added dropwise,

maintaining the temperature below 15°. The reaction mixture was refluxed for 30 minutes and then allowed to stand at room temperature overnight. It was then poured into a saturated aqueous solution of ammonium chloride (150 ml), maintained at 0°. The ether layer was separated, washed with water (100 ml), dried over anhydrous sodium sulfate, concentrated in vacuo. and the residue distilled with a few crystals of iodine. The hydrocarbon product was separated from the water in the receiver, dried over anhydrous sodium sulfate, and distilled over sodium. The final product

(42.2 g, 0.44 mole; 38%) boiled at 94-5°. The NMR spectrum

(Appendix A) was consistent with 2,4-dimethyl-l,3- pentadiene. 84 Preparation of the Carboxylic Acid 9 (Figure 5)

Again the procedure of Jitkow and Bogert (1941) was generally followed. A mixture of 2,4-dimethyl-l,3- pentadiene (40.0 g, 0.42 mole) and acrylic acid (40.0 g,

0.55 mole) was heated in a stainless steel autoclave under a back pressure of 1500 psi of nitrogen at 150° for 20 hours."'" The product was washed out of the autoclave with ether and then vacuum (1 mm) distilled through a 250-mm vacuum-jacketed Vigreux column. Collection of material boiling between 104 and 106° resulted in 43.3 g (0.26 mole;

62%) of crude 9MA as colorless oily needles,I m.p.It 55-60°. This material was found by NMR analysis to be contaminated with approximately 45% of its double bond isomer 10./MAW The 9 ^ 10 isomerization by acid catalysis has been reported m mm J J r (Erman et al., 1969); the catalyst in this case being acrylic acid, from which /M9 was distilled. Repeated recrystallizations from acetic acid/water and 95% ethanol removed some of the acid 10 from 9, but not all of it. mm m' Sublimation of this material gave the same result, pro viding colorless cubelets, m.p. 61-3°. Four recrystallizations of crude 9 from 95% ethanol gave purified 9, m.p.

63-6°, still containing 25% (by NMR integration) 10. The reported (Erman et al., 1969) melting point is 84-5°. The

1. Gratitude is expressed to the University of Arizona High Pressure Laboratory for carrying out this process. 85

NMR spectrum, however, of the acid prepared here was

identical to that reported (Erman et al., 1969) for 9,MA excepting peaks due to the impurity 10. The IR (CHCl^:

3000, very broad; and 1710 cm-"'") and mass (m/e 168, M+;

153, - "CHg; 123; 107, base; 96; 81; and 67) spectra were in

accord with structure 9. MA

Anal. Calculated for C^H^C^: C, 71.39 ; Hf 9.59.

Found: C, 70.98 ; H, 9.47.

Hydrolysis of Chrysanthenone: Preparation of Carboxylic Acid 2 (Figure 5)

Chrysanthenone (2.74 g, 0.0183 mole, described

later) prepared by the photolysis of verbenone was added

to potassium hydroxide (10.0 g) in 95% ethanol (70 ml) and

the mixture was refluxed for 17.5 hours. The reaction

mixture was then poured into cold water (500 ml) and

extracted with ether (200 ml). The ether extract was

washed with water (100 ml), dried over anhydrous sodium sulfate, and evaporated jLn vacuo to provide 0.201 g of a

yellow oil. This material was shown by NMR spectroscopy to

be verbenone, a contaminant of the starting material, which

was prepared from it.

The still-basic aqueous phase, from which the

verbenone was extracted, was acidified with hydrochloric

acid and extracted with ether (200 ml). The ether was

washed with water (100 ml), dried over anhydrous sodium sulfate, and concentrated in vacuo to give 2.95 g (0.0176 86

mole; 96.3%) of crude 9 as a yellow oil which later

crystallized, m.p. 59-67°. Recrystallization from acetic

acid/water gave colorless needles, m.p. 81-2°. NMR

analysis showed that 9 prepared in this way was slightly

(< 10%) contaminated with 10, its double bond isomer. The

explanation for this has been described previously. The NMR

spectrum of this material was identical to the NMR spectrum

of 9 prepared by the Diels-Alder route.

Hydrogenation of Synthetic 9; Preparation of Carboxylic Acid 11 (Figure 5)

The carboxylic acid 9, prepared from chrysanthenone

(0.50 g, 0.0030 mole, described above) was dissolved in 95%

ethanol (20 ml). To this was added 10% palladium on carbon

(0.050 g) and the resulting suspension was hydrogenated at

atmospheric pressure and room temperature for 21 hours. The amount of hydrogen consumed (74 ml) was close to the amount calculated (78 ml) for one mole H /mole 9. The mixture was 20 Ml then filtered through Celite, washing with chloroform, and concentrated in vacuo to yield 11 as a pale yellow oil oil (0.50 g, 0.0029 mole; 97%) which crystallized on standing. The crude crystals had m.p. 50-60°. The anilide purification procedure, described previously for

11 from the A. filifolia lactones, was followed. From 1.0 g WAIVA — ' ^ of crude 11 there was obtained 0.6 g of purified 11, m.p.

65-70°. Recrystallization from acetic acid/water gave pure 87 11 as tiny colorless flakes, m.p. 83-4°. The reported

(Kotake and Nonaka, 1957) melting point is 81-2°. The NMR

(Appendix A), IR (CHCl^: 2960, very broad; and 1710 cm-"*"), and mass (m/e 170, M+; and 83, base) spectra were in accord with structure 11 and identical to those of the acid pre- mm c pared from 12 and 13. c mm mm Anal. Calculated for C, 70.55 ; H, 10.66.

Found: C, 70.47 ; H, 10.59.

The NMR spectra of the crude hydrogenation products

(before purification by attempted anilide preparation) of 9 and 10 from both A. filifolia and chrysanthenone are mm — J identical. Such a spectrum may be found in Appendix A.

This NMR spectrum shows that both cis- and trans-2.2,4- trimethyl-cyclohexane-l-carboxylic acids were obtained.

COOH COOH 11 mm

That is, this spectrum contains three gem-dimethyl signals, more than that expected for just one stereoisomer. After purification by attempted anilide preparation and recrystallization, the product obtained (presumably stereoisomer 11) * Wttfn had only two gem-dimethyl signals (they are not equivalent).

Whether this obtention of only one stereoisomer is due to 88

preferential unreactivity of one isomer or due to isomeriza-

tion to the most thermodynamically stable isomer is diffi

cult to decide. Probably the latter process is unlikely

under the conditions of anilide preparation. The acid 11 c r m Mi described in this dissertation is assigned the cis-

stereochemistry because it is predicted to be slightly less

reactive toward thionyl chloride/aniline than the trans-

stereoisomer. This assignment, however, is only tentative.

Methanolysis of the A. filifolia Lactone (12 + 13) z — to ma— Mixture: Preparation of 19 £1 tML The A. filifolia lactone mixture, obtained by frac

tional distillation of the steam distillate (Table 2; first

distillation, fraction #4, 0.67 g, 0.0041 mole) was

dissolved in absolute methanol (20 ml). This was added dropwise, over a period of 20 minutes, to a solution of sodium (0.12 g, 0.0052 mole) in absolute methanol (75 ml)

and the mixture was stirred at room temperature for 2 hours

after addition was complete. At this time, GLC analysis showed that approximately 50% of the lactone mixture had been consumed. The reaction mixture was poured into water

(200 ml) and extracted with ether (2 x 100 ml). The ether

extracts were combined, washed with water (100 ml), dried over anhydrous magnesium sulfate and concentrated iri vacuo.

The residue was then chromatographed over silica gel

(Baker, 30 x 500 mm). Elution with 1:9 chloroform/benzene 89 afforded recovered lactone mixture (0.32 g, 0.0019 mole;

46%), which by NMR analysis was markedly decreased in

content of lactone 12. Elution with 4:1 chloroform/benzene yielded the methyl ester 19 as a colorless oil (0.33 g, MAM ^ ' 0.0017 mole; 42%). The NMR (Appendix A), IR (Neat: 3410;

1730; and 1600 cm "*"), and mass (m/e 198, M+; and 107, base) spectra were consistent with structure 19. c AM AM

Oxidation of 19; Preparation of the Keto-acid 20 uml wl The hydroxy-acid 19 (0.050 g, 0„00025 mole) was AW/M dissolved in chloroform (50 ml). To this was added active manganese dioxide (10.0 g, 0.12 mole, prepared according to

Ball, Goodwin, and Morton, 1948) and the resulting suspen sion was stirred, under anhydrous calcium chloride, at room temperature for 25 hours. The manganese dioxide was then removed by filtration through Celite and the chloroform evaporated in vacuo to yield 0.045 g (0.00023 mole; 93%) of a dark yellow oil, homogeneous by GLC. The NMR

(Appendix A), IR (Neat: 1735; 1680; and 1650 cm"'*'), and mass (m/e 196, M+; and 137, base, - COOMe) spectra indicated that this material was an a,^-unsaturated ketone carboxylic acid of structure MAM20. This conclusion was confirmed byJ the ultraviolet spectrum (Appendix B). 90

Photolysis of Verbenone: Preparation of Chrysanthenone (Figure 8)

Verbenone has been shown (Hurst and Whitham, 1960;

Erman, 1967) to yield chrysanthenone upon photolysis. This was therefore chosen as the means to be used for the prepara tion of chrysanthenone for this work."'" Several such photolyses were carried out and the results thereof are tabulated in Table 3. These photolyses were all conducted with a Hanovia 450 broad-spectrum mercury arc lamp at temperatures of approximately 50°. The workups of all reactions were identical, except for the photolyses employing glacial acetic acid as the solvent. In the latter case, the photolysis vessels were cooled and the contents poured into a separatory funnel, rinsing with ether. More ether (200 ml) was then added and the acetic acid was largely removed by partitioning it between ether and water. That is, the ether solution was washed with water (3 x 1000 ml). The last of the acetic acid was removed by washing with 5% sodium bicarbonate (2 x 300 ml). The ether solution was then washed again with water (100 ml), dried over anhydrous magnesium sulfate, concentrated in vacuo. and the residue fractionally distilled. When the solvent employed for the photolysis was cyclohexane, the photolysis vessels were cooled, the contents poured into a round-bottom flask,

- 1. Gratitude is expressed to the Organic Chemicals Group, Glidden-Durkee Company, Division of SCM, Jacksonville, Florida, for a generous sample of verbenone. Table 3. The Photolysis of Verbenone

Yield,a %

Run Time, hr Vessel Amt. Verbenone, g Solvent (Amt., ml) V F P I C

1 6.5 Pyrex 13.8 acetic acid (330) 45 9 1 0 0

2 5.5 Pyrex 19.0 cyclohexane (275) 16 0 5 5 46

3 7.5 Pyrex 21.7 cyclohexane (180) 17 0 5 6 46

4 5.5 Vycor 9.4 cyclohexane (200) 28 0 6 9 31

5 22.5 Vycor 10.1 cyclohexane (200) 9 0 8 8 67

6 13.3 Vycor 13.3 cyclohexane (180) 9 0 7 8 52

Composition and yield determined by NMR and GLC analysis. V = verbenone, F = filifolone, P = piperitenone, I = isopiperitenone C = chrysanthenone. 92 rinsed with ether, and the cyclohexane carefully evaporated

under vacuum. The residue was then fractionally distilled.

All fractional distillations were carried out through a

250-mm vacuum-jacketed Vigreux column at reduced pressures

(20, 16, 13, 13, 4.3, and 1.5 mm). The elevated tempera

ture (50°) used for these photolyses resulted in the

racemization of the chrysanthenone product, even though the

verbenone employed as the starting material was optically

active ([a]^ +93.7°). The reason for employing an

apparatus operating at such a temperature was simply that it

was the only one available. Although not included in Table

3, some isogeranic acid was obtained as a thermal decomposi tion product of chrysanthenone and filifolone in some of the distillations. The observed boiling points for chrysanthe none were: 96-8° (16 mm), 88-92° (13 mm), 69-72° (4.3 mm), and 52.5-53° (1.5 mm). The chrysanthenone obtained in this way had an infrared spectrum (Neat: 1775 cm"1) characteris tic of cyclobutanones, the total of which was identical to that published (Teresa, Bellido, and Bellido, 1962) for natural chrysanthenone from Chrysanthemum indium. The NMR

(Appendix A) and mass (m/e 150, M+; 135, - 107, base,

- (CC>2 + 'CHg); 91; and 82) spectra were also in accord with those expected for chrysanthenone.

Anal. Calculated for C^QH^O: C, 79.96 ; H, 9.39.

Found: C, 80.39 ; H, 9.24. 93

The byproducts of the photolysis were identified by

NMR spectroscopy.

Oxidation of Chrysanthenone: Chrysanthenone Epoxide (Figure 8)

This reaction was carried out in two ways. The first was adapted after the method of Chretien-Bessiere and

Retamar (1963). Chrysanthenone (2.98 g, 0.0198 mole) was added dropwise to a solution of meta-chloroperbenzoic acid

(Aldrich lot #081307, technical, 85%, 3.77 g, 0.0318 mole) in dry ether (75 ml). The reaction mixture was then stirred at room temperature under anhydrous calcium chloride for 24 hours. The ethereal solution was washed with 10% potassium hydroxide (2 x 50 ml), then water (50 ml), dried over anhydrous magnesium sulfate and evaporated to yield 2.41 g

(0.0145 mole; 74.0%) of a yellow oil, homogeneous by GLC.

This was shown by NMR (Appendix A), XR (Neat: 1770 cm "*"), and mass"*" (m/e 166, M+; 122, - C02l anc^ 107, base, - (CC^ +

•CH^)) spectral analysis to be chrysanthenone epoxide.

Anal. Calculated for cioH14°2: ^' 72.26 ; H, 8.49. Found: C, 72.51 ; H, 8.57.

The second method of oxidation was designed after that of Corey et al. (1969, 1971). This method was shown by Corey et al. to provide Baeyer-Villiger oxidation of

1. Chrysanthenone epoxide, under the mass spectroscopy conditions employed (inlet temperature 200 ) isomerizes largely to lactone 12. 94 olefinic ketones in preference to epoxidation. Hence, it

was hoped that this technique would oxidize chrysanthenone directly to lactone 1^. Unfortunately, only chrysanthenone

epoxide was obtained. The procedure follows.

Chrysanthenone (0.767 g, 0.0051 mole) was dissolved

in dichloromethane (25 ml). To this was added, after

cooling to 0°, meta-chloroperbenzoic acid (Aldrich lot

#081307, technical, 85%, 1.10 g, 0.0064 mole) and sodium bicarbonate (0.548 g, 0.0065 mole). The reaction mixture was stirred under anhydrous calcium chloride, maintaining the temperature below 4°, for 2 hours. After this time

GLC analysis showed the reaction to be complete. The

mixture was then poured into water (200 ml) and the organic layer separated. The latter was washed with 5% potassium hydroxide (2 x 50 ml), then water (100 ml), dried over anhydrous sodium sulfate, and evaporated to provide 0.846 g

(0.0051 mole; 100%) of chrysanthenone epoxide as a pale yellow oil. The NMR and IR spectra of this material were identical to those of chrysnnthenone epoxide prepared as described previously. It is to be noted that the inclusion of sodium bicarbonate in the oxidation mixture (according to Corey et al.) does not give Baeyer-Villiger oxidation products in this case. However, both the rate of epoxida tion and the yield of epoxide are greatly increased. 95

Solvolysis of Chrysanthenone Epoxide* Preparation of the Acid 17 (Figure 8)

This was performed according to Chretien-Bessiere and Retamar (1963). Chrysanthenone epoxide (0.558 g,

0.00336 mole) was placed with potassium hydroxide (1.5 g) in 95% ethanol (5 ml) and water (10 ml). The resulting mixture was then refluxed for 3 hours. GLC showed no chrysanthenone epoxide remaining after this time. The solution was diluted with water (100 ml), acidified, and extracted with chloroform (3 x 50 ml). The chloroform extracts were combined, washed with water (100 ml), dried over anhydrous sodium sulfate, and evaporated in vacuo to yield 0.324 g of a yellow oil which crystallized on stand ing. These crystals (0.00176 mole; 52.5%) had m.p. 132-6°.

Recrystallization from benzene gave 0.289 g of 17 as ma ma colorless cubelets, m.p. 139-40°. The reported melting point (Chretien-Bessiere and Retamar, 1963) of 17 is 140°.

This material was identical to 17 from the hydrolysis of fmrm lactone 12 isolated from A. filifolia, except that it is racemic. This identity was demonstrated by an undepressed mixed melting point (m.m. p. 138-40°) and identical IR and

NMR spectra.

When 17 (0.0587 g), prepared from chrysanthenone epoxide, was heated to 160-177° for one hour, MAAM12 and water were formed in quantitative yield (total weight = 0.0583 g).

The NMR spectrum of the latter mixture had no extraneous peaks and showed the molar ratio of 12:Ho0 to be 1:1. C MAMA 2 Excepting the water present, this NMR spectrum was identical to that of 12 isolated from Artemisia filifolia. AMMA • '1

Thermal Rearrangement of Chrysanthenone Epoxide: Preparation of Lactone 12 (Figures 8 and 9) . um _ Attempted preparative GLC (> 160°) of chrysanthenone epoxide resulted in the collection of 12 as the only c A/AMA J product. It was therefore suspected that the former was undergoing thermal rearrangement to the latter. To verify this, the following experiment was performed.

Chrysanthenone epoxide (0.0838 g, 0.000506 mole) was placed in an NMR tube and heated (neat) in an oil bath at 160-180° for 15 minutes. The product (0.0829 g, 0.000496 mole; 98%) was shown by NMR and IR spectroscopy to be exclusively lactone 12. Lactone 12 prepared in this way could be saponified to the hydroxy-acid 17, identical to that prepared as previously described (m.p. 17 = 138-40°, Irnnfi m.m.p. = 138-40°).

Bromination of the Acid 9; Preparation of lactone ^2 (Figure 10)

Several similar experiments were performed. One representative is presented here.

The carboxylic acid 9 (2.1 g, 0.012 mole) was dissolved in carbon tetrachloride (40 ml) and to this was added N-bromosuccinimide (3.1 g, 0.017 mole) and benzoyl peroxide (0.20 g, 0.00083 mole). The mixture was stirred 97 and refluxed under a nitrogen atmosphere for 22.5 hours and then stirred at room temperature for an additional 2 hours.

The reaction mixture was cooled in ice and the precipitate of succinimide was removed by filtration, washing with cold carbon tetrachloride. Evaporation of the filtrate provided as a residue 4.2 g of a yellow oil. This oil was taken up in chloroform (150 ml) and extracted with 5% sodium hydroxide (2 x 150 ml). The sodium hydroxide extracts were combined, acidified, and extracted with chloroform (2 x 150 ml). These chloroform extracts were combined, washed with water (100 ml), dried over anhydrous sodium sulfate, and evaporated to yield 0.28 g of acidic product as a yellow oil, which later crystallized. NMR analysis revealed this to be primarily benzoic acid plus some succinimide. The bromo-acid 21 was not observed as an isolable product. MAMA The chloroform solution, from which the acidic products were extracted, was washed with water (100 ml), and dried over anhydrous sodium sulfate. Evaporation of the solvent djn vacuo afforded 1.5 g of a yellow oil. An NMR examination of this oil revealed that it contained 27% lactone 12, along with at least two unidentified products. mm' 3 c The oil was combined with the oils obtained in the same way from two other brominations of the acid 9 to give a total of 4.0 g of crude lactone 12 (these other experiments started with 2.1 and 3.8 g of 9). This was subjected to short-path distillation at reduced (0.3 mm) pressure. The 98

fraction boiling at 79-83° was collected, providing 1.1 g

(0.0066 mole; 20%) of pure 12 as a colorless oil. The NMR

and IR spectra of 12AWM\ prepared in this way were identical to

those of 12AM/VW from —A. filifolia.

Photolysis of Chrysanthenone; Preparation of the Ketone 22 (Figure 11) MAAM J Chrysanthenone, obtained by photolysis of verbenone

(Table 3), was photolyzed in substantially the same way.

Therefore, the pertinent data are presented in Table 4

without much additional comment. The tricyclic ketone

("TCK") and the bicyclic hydrocarbon ("HC") were struc

turally deduced by a comparison of their NMR spectra to

those reported by Erman (1967) in his photolysis work. The

NMR spectra of these two compounds are in Appendix A. The

workup and isolation of products was as follows. The solvent (cyclohexane) was carefully evaporated jLn vacuo

and the residue was subjected to preparative GLC on Column

I or Column II (column temperatures used were from 95 to

115°). A typical chromatogram is shown in Figure 29. The recovery of products by preparative GLC was calculated to be only approximately 55-60%. When Column I was used for

preparative GLC (Figure 29) collection of the peak with a retention time (R. ) of 21.6 minutes afforded the ketone 22. t rmtrn This.material had NMR (Appendix A), IR (Neat: 1780; and

1650 cm-"'"), and mass (m/e 150, M+; 122, - CO; 108; 107, Table 4. The Photolysis of Chrysanthenone

Yield ,a %

Run Time, hr Vessel Amt. Chrysanthenone, g Solvent (Amt., ml) C 22 TCK HC

1 39.5 Vycor 7.30 cyclohexane (200) 44 50 3 3

2 20 Vycor 1.0 cyclohexane (100) 20 62 ND ND

3 24 Vycor 2.20 cyclohexane (100) 45 45 ND ND

4 46 Vycor 6.80 cyclohexane (70) 16 23 3 58

aYields and composition determined by GLC and NMR. C = chrysanthenone, 22 = ketone 22, TCK = tricyclic ketone (below), HC = hydrocarbon products (mainly the one shown^below), ND = not determined.

22 tricyclic ketone (TCK) major hydrocarbon product (HC) m\.W 100

UL/ C—i 1 ' i 6 15 /8 21 24 27 Retention time (R.), minutes

Relative Ratio ft Product (Teak-Area) 5.3 ib>* 4.3 6.4 21.6

21.6 MM22 9.4 23.3 chrysanthenone 5.7

26.2 dk 1 o *This compound was not isolated, but is the minor hydrocarbon product described by Erman (1967) in his photolysis of chrysanthenone.

Figure 29. GLC Chromatogram of Chrysanthenone Photolysis Products — Column I; flow rate = 250 ml/min.; temperatures = column, 115°; detector, 191°; injector, 187°. 101 base; 93; 91; and 80) spectra consistent with those reported

(Erman, 1967) for 22.

Anal. Calculated for C^QH^O: C, 79.96 ; H, 9.39.

Found: C, 80.10 ; H, 9.21.

Oxidation of 22; Preparation of Epoxide 23 WiM*' f_ NtM (Figure 11)

When the method of Corey et al. (1969, 1971) was applied to the ketone 22, some (10-20%) Baeyer-Villiger oxidation did occur, in contrast to the peracid oxidation of chrysanthenone (described previously). The ketone 22

(0.7011 g, 0.00474 mole) was added dropwise to a mixture of meta-chloroperbenzoic acid (Aldrich, technical, 85%, 0.95 g,

0.0048 mole) and sodium bicarbonate (0.43 g, 0.0050 mole) in dichloromethane (12 ml). The resulting suspension was then stirred under anhydrous calcium chloride at room temperature for 2 hours. The same results were obtained when the reaction was carried out at 0°. The stirring was stopped and the reaction mixture was allowed to stand at room temperature overnight. The mixture was diluted with chloroform (100 ml), washed with 5% potassium hydroxide

(100 ml), then water (100 ml), dried over anhydrous sodium sulfate, and evaporated. This resulted in 0.0828 g of a yellow oil, shown by NMR analysis to consist of 10% 12

(Baeyer-Villiger oxidation product), 10% stereoisomeric epoxides (2 of them) of 12, and 70% 23 + 24. All products 102

were collected by preparative GLC on Column II at a column

temperature of 146°. Epoxide 24 was obtained in extremely

small amount and epoxide 23 was converted largely to the

1:1 mixture of lactones 12 and 13 (R. 24.5 minutes), MVM ,Wi.Wi t ' although some thermally unrearranged material (R^ 19.2

minutes) was obtained. Collection of the peaks with

retention times of 28.7 and 30.5 minutes resulted in the obtention of the a,a-epoxy-lactone (0.043 g, 0.00024 mole;

5.1%) and the p,a-epoxy-lactone (0.033 g, 0.00018 mole;

4.0%), respectively. Both compounds were purified by

a,a-epoxy-lactone (3 , a-epoj

-1 81-2°, had NMR (Appendix A), IR (CHC13: both 1770 cm ), and

mass (m/e 182, M+; 126; 96, base; 83; 55; 43 and m/e 111, base; 83; 55; 43, respectively) spectra in accord with the structures proposed. The structural assignments were made by NMR spectroscopy, although it was not definitely shown which isomer was which. However, the compound with Rt 28.7 minutes was obtained in amounts slightly greater than that with Rfc 30.5 minutes, and the former was assigned the 103 a,a- stereochemistry because the ketone 22, structurally

related, gives mostly a-epoxide product (23).

Because of these difficulties, a method of epoxide

preparation was sought which would give a cleaner product.

The added sodium bicarbonate, even though it greatly

(> 10 x) accelerates the reaction rate, leads to 10-20%

unwanted Baeyer-Villiger oxidation of 22. Better means of

preparation of lactone 12 have been described previously.

Therefore, the bicarbonate was omitted as a reactant. A

typical experiment, one of four, is presented below.

To the ketone 22 (0.157 g, 0.00105 mole) in

dichloromethane (3 ml) was added meta-chloroperbenzoic acid

(Aldrich, technical, 85%, 0.21 g, 0.00122 mole) in dichloro

methane (15 ml). The reaction mixture was then stirred

under anhydrous calcium chloride at room temperature for 20

hours. After this time the solution was diluted with

chloroform (50 ml), washed with 5% sodium hydroxide (3 x

150 ml), washed with water (100 ml), and dried over anhydrous

magnesium sulfate. Evaporation of the solvent in vacuo

afforded 0.112 g (0.000675 mole; 64.3%) of crude epoxide

product as a pale yellow oil. NMR analysis (Appendix A) of

this oil revealed the presence of the two stereoisomeric

epoxides 23 and 24, and the absence of Baeyer-Villiger r AM MA Hfm '

products. By GLC analysis (Column I) MAAM23 and MAMA24 were present in the ratio 3:1, respectively. Column chromatography over

silica gel did not separate 23 and 24, and chromatography 104- over alumina (Brockmann activity grade III) resulted in

their decomposition. Careful preparative GLC on Column I

finally provided good separation. Figure 30 shows a typical

chromatogram. Preparative GLC of the 23 + 24 mixture pre- MAMA Whim pared here resulted in 0.0420 g (0.000253 mole; 25.1%) of

pure 23. The NMR (Appendix A) IR (Neat: 1780 cm"1), and frAftn mass1 (m/e 166, M+; 123; and 107, base) spectra of this

material were in accord with structure 23.

c H : C 72 26 H Anal. Calculated for 1o i4°2 ' - # 8.49. Found: C, 71.99 ; H, 8.57.

Pure epoxide NAM24 was obtained in amounts too small for adequate structural analysis.

Thermal Rearrangement of Epoxide 23; Preparation of Lactone 13 (Figure 11)

Pure epoxide 23 (0.0420 g, 0.000253 mole), collected r MM ^ ' ' by preparative GLC, was heated (neat) in an NMR tube in an oil bath at 160-175° for 12 minutes. At 160° the material started to turn brown, and it was dark brown in color when the temperature of the oil bath had reached 175°. The tube was cooled and weighed (0.0419 g, 0.000252 mole; 99.7%) and the NMR spectrum recorded. The latter was found to be representative of pure 13, with no extraneous peaks. The infrared spectrum of this sample was identical to that of

1. Epoxide 23, under the conditions of the mass spectroscopy (inlet temperature 200°), isomerizes to lactone 13. MM 105

R Retention time ( t), minutes

ft Compound Relative Ratio 15.0 24 1 mm

24.8 u»AM23 3 27.6 13 1 (from thermolysis of 23 during M \ MAMA mm GLC process)

Figure 30. GLC Chromatogram of Epoxides 23 and 24 — Column I: flow rate = 300 ml/min; temperatures: column, 148°; detector, 187°; inuector, 191°. 106

13 isolated from A. filifolia. This synthetic sample of lactone 13 could be saponified to the hydroxy-acid 18. »*m c mm' identical to the material from A. filifolia as shown by an undepressed (133-6°) mixture melting point. The sample of

18 prepared from synthetic 13 was recrystallized from cyclohexane before the mixed melting point was attempted.

Solvolysis of Epoxide 23; Preparation of Hydroxy-acid 18 (Figure 11) i MAW Pure epoxide 23 (0.050 g, 0.00030 mole) was refluxed c mm ^' with potassium hydroxide (0.16 g, 0.029 mole) in 95% ethanol

(5 ml) and water (2 ml) for 3 hours. The reaction mixture was then cooled, diluted with water (20 ml), and extracted with ether (2 x 25 ml). The ether extracts were combined, washed with water (25 ml), dried over anhydrous magnesium sulfate, and evaporated to yield less than 1 mg of neutral material of undetermined composition. The still-basic aqueous phase was acidified with hydrochloric acid and extracted with ether (2 x 25 ml). The ether extracts were combined, washed with water (20 ml), dried over anhydrous magnesium sulfate, and evaporated in vacuo to provide 0.042 g of a brown oil. The latter was chromatographed over silica gel (Baker, 20 x 350 mm). Elution with ether gave crystalline 18 (0.029 g, 0.00017 mole; 58%), m.p. 120-6°.

Recrystallization from cyclohexane gave pure 18 (0.022 g) as colorless cubelets, m.p. 134-6°. That this was identical 107

to the 18 from A. filifolia was shown by an undepressed

(133-6°) mixture melting point and identical IR and NMR

spectra.

Preparation of 3-Methyl-2-butenoic Acid (Senecioic Acid, g.p-Dimethylacrylic Acid) (Figure 12)

An alteration of the procedure of Smith, Prichard,

and Spillane (1955) was employed. To commercial chlorine

bleach (5.25% sodium hypochlorite, 3 1, 2.1 moles

hypochlorite) was added dioxane (200 ml) and mesityl oxide

(50 g, 0.51 mole). The reaction mixture was stirred under

a condenser for 4 hours, after which the temperature of the

solution had dropped to that of the room. Excess sodium

hypochlorite was destroyed by the addition of 5% sodium

bisulfite solution (100 ml). Sufficient 50% sulfuric acid

was then added to make the solution acidic to Congo Red

paper. The solution was cooled in ice and extracted with

chloroform (8 x 400 ml). The chloroform extracts were

combined, dried over anhydrous sodium sulfate, and

evaporated in vacuo to yield an oily residue containing much

elemental sulfur. The residue was taken up in chloroform

(1 1) and washed with 5% sodium hydroxide (2-x 600 ml). The

sodium hydroxide extracts were combined, washed with

chloroform (500 ml), acidified, and extracted with chloroform

(4 x 200 ml). The latter chloroform extracts were combined,

washed with water (300 ml), dried over anhydrous sodium sulfate, and evaporated to yield 25.6 g (0.26 mole; 51%) of 108 senecioic acid as a yellow oil which later crystallized.

Recrystallization from water gave pure senecioic acid

(20 g, 0.20 mole; 39%) as colorless needles, m.p. 65-6°.

The reported (Smith et al., 1955) melting point is 66-67.5°.

The NMR spectrum of this material was consistent in all respects with that expected for senecioic acid (see

Appendix A).

Attempted Lewis Acid-catalyzed Diels-Alder Reaction of Senecioic Acid and Isoprene (Figure 12)

This reaction was carried out according to the

Chisso Company Patent (The Chisso Company, 1967) under a variety of conditions, all of which failed to provide the desired product, 10. A typical experiment is presented here.

A solution of senecioic acid (1.0 g, 0.010 mole) in benzene (5 ml) was added dropwise to a stirred suspension of anhydrous aluminum chloride (2.7 g, 0.02 mole) in benzene (20 ml) which had been warmed to 40°. The tempera ture was maintained between 39 and 45° at all times and the reaction was conducted under a nitrogen atmosphere. When all of the senecioic acid had been added, the solution was yellow in color and most of the aluminum chloride, but not all of it (as reported by the Chisso Company), had gone into solution. To the reaction mixture was then added, dropwise, a solution of isoprene (0.68 g, 0.010 mole) in benzene

(5 ml) over a period of 10 minutes. The resulting mixture 109

was stirred for an additional hour, cooled, diluted with 5 N

hydrochloric acid (50 ml), and benzene (50 ml). The mixture

was shaken and the benzene layer was separated, washed with

water (50 ml), and extracted with 5% potassium hydroxide (3

x 50 ml). The potassium hydroxide extracts were combined,

acidified, and extracted with ether (3 x 50 ml). The ether

extracts were combined, washed with water (50 ml), dried over anhydrous magnesium sulfate, and evaporated to yield

1.0 g of a pale yellow oil which crystallized on standing.

An NMR (Appendix A) analysis of this oil revealed the absence of ^0, and instead the presence of a 1:1 mixture of unreacted senecioic acid and 3-methyl-3-phenylbutanoic acid.

The latter probably arose by a Friedel-Crafts type reaction involving the solvent (benzene). This mechanism is presented in Figure 31.

Preparation of Ethyl 3-Methyl-2-butenoic Acid (Ethyl Senecioate)

A slight alteration of the procedure of Micovic

(1943) was employed. A mixture of senecioic acid (8.0 g,

0.080 mole), absolute ethanol (13.8 g, 0.30 mole), toluene

(6.9 g), and concentrated sulfuric acid (4 drops) was refluxed for 5 hours. The water-ethanol-toluene ternary azeotrope (b.p. 75-8°) was then distilled off, dried over anhydrous potassium carbonate, poured back into the dis tilling flask, and the mixture re-distilled until a temperature of 79° was observed for the distillate. The 110

AlOl, 5-

COOH

AlCl AlOl

H o h2o

OOH 0^C)H

A1C1- AlOl © -

Figure 31. Mechanism of Formation of 3-Methyl-3-phenylbutanoic Acid Ill

contents of the distilling flask were then transferred to a

smaller flask and distilled under vacuum (20 mm). Ethyl

senecioate was collected as a colorless liquid, b.p. 76-7°

(7.0 g, 0.055 mole; 69%). The NMR spectrum of this material

(Appendix A) was consistent in all respects with that pre

dicted for ethyl senecioate.

Attempted Lewis Acid-catalyzed Diels-Alder Reaction of Ethyl Senecioate and Isoprene (Figure 12)

This reaction was carried out exactly as was that of senecioic acid, employing 2.6 g (0.020 mole) of ethyl senecioate, 2.6 g (0.020 mole) of aluminum chloride, and

1.5 g (0.021 mole) of isoprene. After quenching by addition of dilute hydrochloric acid, the reaction mixture was diluted with ether (50 ml) and the organic layer was separated, washed with water (20 ml), dried over anhydrous

magnesium sulfate, and evaporated. This resulted in 2.8 g of a yellow oil, shown by NMR to contain none of the desired product, the ethyl ester of 10. Instead, the NMR spectrum

(Appendix A) revealed that this oil was a 3:1 mixture of unreacted ethyl senecioate and ethyl 3-methyl-3-phenylbutanoate, respectively. The latter apparently resulted by the same reaction with solvent (benzene) as did its parent acid, described previously.

Utilization of cyclohexane or dioxane as solvent, in place of benzene, for this reaction, with both senecioic acid and its ethyl ester, resulted only in recovery of 112 unreacted starting material (except isoprene, which polymerized).

Performic Acid Oxidation of 9; Preparation of Lactone 28 (Figure 13) sm _ The carboxylic acid 9 (1.00 g, 0.00595 mole) was placed with 98% formic acid (3 ml). To this was added dropwise a solution of 30% hydrogen peroxide (7.0 g, 0.0062 mole) in 98% formic acid (8 ml) and then enough (3 ml) 95% ethanol was added to bring the mixture to homogeneity.

The reaction mixture was stirred in a water bath (24-5°) for 18 hours. After this time, the mixture was diluted with water (200 ml), excess peracid was destroyed by the addition of saturated sodium thiosulfate (200 ml), and the solution was made alkaline to pH 8-9 with aqueous sodium hydroxide. The total aqueous solution was extracted with chloroform (2 x 100 ml) and the chloroform extracts were combined, washed with water (100 ml), dried over anhydrous magnesium sulfate, and concentrated in vacuo. This resulted in a residue consisting of 1.02 g (0.00554 mole; 91.3%) of a pale yellow oil, homogeneous by GLC. The NMR (Appendix

A), IR (Neat: 3400; and 1735 cm"''') , and mass (m/e 184, M+;

113; 85; 72; and 43, base) spectra were consistent with structure 28. The stereochemistry of 28 was deduced from mm J mm the mechanism (Figure 13) of its formation.

Anal. Calculated for CioH16°3: C/ ®^.19 » 8*75. Found: C, 65.37 ; H, 9.00. 113

Iodolactonization of the Carboxylic Acid 9; Preparation of Iodolactone 30 (Figure 14) _ um It has recently been reported (Barnett and Son,

1972) that the ring size of iodolactones is dependent upon the conditions employed in their preparation. That is, long reaction times and added potassium iodide result in the larger of the two possible iodolactones, while short reac tion times and no added potassium iodide provide the smaller iodolactone. The iodolactonization of 9MA was attempted both with and without added potassium iodide, and the product was the same in each case: the iodolactone AWMA30, not MAMA29. Possibly the results observed by Barnett and Son apply only to P,Y-unsaturated acids (to which their investigation was confined) and not to Y,6-unsaturated' acids such as 9.MA

Since both iodolactone preparations from 9MA were substantially identical except for added potassium iodide

(one equivalent) and reaction time, only a single repre sentative is included here.

A suspension of the carboxylic acid 9 (5.0 g, 0.030 mole) in water (200 ml) was carefully neutralized with 5% sodium hydroxide. To this solution was then added sodium bicarbonate (2.5 g, 0.030 mole) and then iodine (10 g,

0.039 mole). The resulting suspension was stirred at room temperature for 22 hours (after 2 hours, a considerable amount of white precipitate began forming). The reaction mixture was then extracted with ether (3 x 150 ml) and the 114 bright yellow ether extracts were combined. The ethereal solution was washed with saturated sodium thiosulfate (2 x

200 ml) until colorless, washed with water (100 ml), dried over anhydrous magnesium sulfate, and evaporated in vacuo.

This gave crude iodolactone 30 as a pale yellow oil (8.7 g,

0.22 mole; 74%) which crystallized on standing. Recrystal- lization from pentane gave 30 as colorless cubelets, m.p.

78-80°. A second recrystallization from the same solvent provided pure material, m.p. 79-81°. The IR (CHCl^: 1750 cm-"*"), NMR (Appendix A), and mass (m/e 294, M+; 167, - I*;

123, - (I* + CO2)', and 84, base) spectra were in accord with structure 30. mm

Anal. Calculated for C^qEL^C^I: C,40.79 ; H,5.10 ; 1,43.23.

Found: C,40.68 ; H,5.32 ; 1,43.47.

Preparative GLC of the A. filifolia Steam Distillate; Isolation of the Mixture of Lactones 12 + 13

The total steam distillate of Artemisia filifolia

(gathered on 10/31/69) was subjected to preparative GLC on

Column II at a column temperature of 178°. A typical chromatogram can be seen in Figure 15. From a total of

1.30 g of steam distillate, collection of the peak with R^

10.2 minutes gave 0.10 g (7.7%) of the mixture of lactones

12 + 13. An NMR Spectrum (Figure 3) of this material mm mm c 3 revealed, by an integration of peak areas, that it was a

1:1 mixture of 12 and 13. mm mm 115

Preparative GLC of Lactone 12; Isolation of the Mixture of Lactones 12 + 13 mm mm Two experiments were performed. In the first, pure lactone 12 ([a]^ -33.2°), from A. filifolia (0.088 g), was mm L D ' 3 ' subjected to preparative GLC on Column III at a column temperature of 168°. Collection of the only peak (R^. 9.0

minutes) resulted in 0.075 g (85%) of a mixture of lactones

12 and 13. NMR integration revealed the composition of this

mixture to be 42% 13 and 58% 12. This mixture was optically mm mm c J active, [cc]^ +6° (c 1.51, CHC13). The probable reason that the mixture of 12 and 1^ was not exactly 1:1 in composition is that the temperature of the chromatography (168°) was not high enough nor the time of heat exposure (9 minutes, flow rate = 150 ml/minute He) in the instrument long enough to effect full equilibration of 12 £ 13. ^ mm mm In the second experiment, pure lactone 12 from A. filifolia (0.053 g) was subjected to preparative GLC on

Column II under the conditions found in Figure 15. Collec tion of the only peak (Rt 10.2 minutes) resulted in 0.041 g

(77%) of a 1:1 (by NMR integration) mixture of 12 and 13. WuVA /MW An optical rotation was not recorded for this material.

Heating of Lactone 12 ^ mu*a Pure lactone 12 (synthetic) (0.056 g) was heated mm J 3 (neat) in an NMR tube in an oil bath at 160-170° for 15 minutes. After this time the sample (which had turned light 116 brown in color) was cooled, weighed (0.056 g; 100% recovery)

and its NMR spectrum was recorded. The latter showed that

12 was completely unchanged.

Steam Distillation of Lactone 12AM.VA Pure lactone 12 (synthetic, 0.0623 g) was subjected

to steam distillation until 200 ml of distillate had been

collected (after 2.75 hours). The distillate was then

extracted with dichloromethane (2 x 100 ml). The extracts

were combined, dried over anhydrous sodium sulfate, and

concentrated in vacuo to yield 0.0413 g (66% recovery) of a

pale yellow oil. This material was homogeneous by GLC and its NMR spectrum revealed it to be pure 12 (i.e., no 13 was c c WUWl ' Wl*"

produced from 12MAMA during the steam distillation process).

Steam Distillation of 17; Isolation of the ami Mixture of Lactones 12 + 13 MM mm Pure 17 (synthetic) (0.121 g, 0.000658 mole) was NMt/i J ^' steam distilled for 5.5 hours, after which time 475 ml of distillate had been collected. The distillate was poured into a separatory funnel, salt was added, and the mixture was extracted with ether (2 x 100 ml). The ether extracts were combined, dried over anhydrous sodium sulfate, and evaporated in vacuo to provide 0.0585 g (0.000353 mole;

53.6%) of a pale yellow oil. The latter was shown by NMR to be a 1:1 mixture of lactones 12 and 13. 117

Photolysis of (+)-Verbenone: Obtention of T+)-Chrysanthenone (Figure 8)

It has been mentioned previously that photolysis of

(+)-verbenone at a temperature of 50° gave largely racemic

products. Therefore, a large fan was used to cool the

Hanovia 450 broad spectrum mercury arc lamp employed for the

photolysis. This brought the temperature down to 34°.

Under these conditions, (+)-verbenone (partially active,

[a]^4 +93.7°, 37% optically pure, 10.28 g, 0.0685 mole) in

cyclohexane (250 ml), in two Vycor vessels, was photolyzed

for 21.0 hours. After this time, a GLC analysis (Column

II) showed the composition of the reaction mixture to be

40.7% chrysanthenone, 58.2% verbenone, and only traces

(< 1%) of other products. The cyclohexane was then removed

in vacuo to provide 11.2 g of a yellow oil. The latter was

distilled under vacuum (7 mm) through a 250-mm vacuum-

jacketed Vigreux column. Four fractions were collected:

fraction # b.p. (°C) weight (q) composition1

1 74-6 0.04 88% C; 12% V 2 76.5-77 2.57 87% C; 13% V 3 77.5-79.5 1.30 80% C; 20% V 4 80-2 0.85 60% C; 40% V pot residue —— 4.8 80% V + some IGA, I, P

1. By GLC and NMR analysis. C = chrysanthenone; V = verbenone; I = isopiperitenone; P = piperitenone; IGA = isogeranic acid, a thermal decomposition product of chrysanthenone. 118

Chrysanthenone from distillation fraction #2 was

optically active, [°0^ +31° (c 3.7, CHCl^) (corrected for

13% verbenone). This corresponds to an optical purity of

29%, based upon a specific rotation of +108° (Erman, 1967)

for pure chrysanthenone. This sample of chrysanthenone had

an NMR spectrum identical, excepting peaks due to the 13%

verbenone, to that of chrysanthenone prepared as described

previously.

Oxidation of (+)-Chrysanthenone; Preparation of )-Chrysanthenone Epoxide (Figure 8)

To (+)-chrysanthenone (1.08 g, 0.00720 mole, from

distillation fraction #2, above, containing 13% verbenone)

in chloroform (30 ml) was added 85% meta-chloroperbenzoic

acid (1.46 g, 0.00733 mole, based upon 85% purity) and

sodium bicarbonate (0.61 g, 0.0072 mole). The resulting

mixture was stirred at room temperature under anhydrous

calcium chloride for 2 hours. The mixture was then diluted

with water (50 ml) and chloroform (50 ml), shaken, and the

organic layer separated. The chloroform solution was

washed with 5% sodium hydroxide (2 x 50 ml), then water

(50 ml), dried over anhydrous magnesium sulfate, and

concentrated in vacuo to provide 1.18 g (0.00710 mole; 99%) of a yellow oil. This was shown by NMR analysis to be

chrysanthenone epoxide, contaminated only with verbenone

(13 mole %). This material was optically active, -26°

(c 4.8, CHClg) (corrected for 13 mole %, 12% by weight, 119 verbenone). If the chrysanthenone epoxide prepared here is

29% optically pure, as was its precursor, (+)-chrysanthenone, then the specific rotation (unreported) of pure (-)-

chrysanthenone epoxide is -90°.

Thermolysis of (-)-Chrysanthenone Epoxide: Preparation of (-)-Lactone 12 (Figure 8) mm ~ (-)-Chrysanthenone epoxide, [a]^4 -26° (0.1683 g,

0.001013 mole, still containing 13 mole % verbenone) was heated (neat) in a glass tube in an oil bath at 170-180° for

20 minutes. An NMR spectrum of the product showed it to be lactone 12, contaminated only with verbenone (13 mole %. 12% mm' •* ' by weight). This material was a pale brown oil (0.1648 g,

0.000993 mole; 98%) and was optically active, [cc]^4 -8.1°

(c 1.8, CHCl^) (corrected for 13 mole % verbenone). This calculates for an optical purity of 24%.

Hydrolysis of (-)-Chrysanthenone Epoxide; Preparation of (+)-17 (Figure 8) r mm _ (-)-Chrysanthenone epoxide, [oO^4 -26° (0.44 g,

0.00265 mole) was refluxed with potassium hydroxide (2.0 g) in 95% ethanol (20 ml) and water (2 ml) for 2 hours. The mixture was then cooled, diluted with water (75 ml), washed with chloroform (2 x 50 ml), acidified, and re-extracted with chloroform (2 x 50 ml). The latter chloroform extracts were combined, washed with water (50 ml), dried over anhydrous magnesium sulfate, and evaporated to give 0.0824 g of a pale yellow oil. This oil later crystallized after 120 trituration with ether. The crystals (0.0532 g, 0.000289 mole; 10.9%) were recrystallized from benzene to colorless platelets, m.p. 136-8°. An NMR spectrum of this material was identical to that of 17MAM prepared as previously described. This sample of the hydroxy-acid 17 was optically active, [°0j^ +8.2° (c 0.56, CHCl^). This calculates for an optical purity of 24%.

Preparative GLC of the A. filifolia Steam Distillate; Isolation of (-)-Verbenone. Borneol. Piperitenone. and the Enedione 34 (Figure 15) mi The total steam distillate of Artemisia filifolia

(6.5 g, collected 8/6/70) was subjected to preparative GLC on Column II under the conditions specified in Figure 15.

Collection of peaks with retention times of 0.9, 2.0, 2.6,

3.2, and 10.2 minutes provided 1,8-cineole (1.7 g; 26%),

(-)-filifolone (0.48 g; 7.4%), (-)-camphor (1.5 g; 23%), isophorone (1.9 g; 29%), and the mixture of lactones 12 and

13 (0.50 g; 7.7%), respectively. The latter has been mm ^' •* described previously in this Experimental Section and the identities of the other four compounds, shown already

(Onore, 1967) to be constituents of A. filifolia. were deduced by spectral (IR, NMR) and GLC comparisons with authentic specimens.

Collection of the peak with a retention time of 4.7 minutes provided 0.041 g (0.63%) of a yellow oil. The NMR

(Appendix A), IR, and UV spectra of this material were 121 identical to those of verbenone. The rotation of this

sample from A. filifolia. -184° (c 0.2, CHCl^),

indicated that this was (-)-verbenone. The specific rota

tion of pure (-)-verbenone is -256°. Identity of the

material of 4.7 minutes with (-)-verbenone was con

firmed by an undepressed mixed melting point of 2,4-

dinitrophenylhydrazones of natural and authentic specimens.

The derivative preparation is described later.

Collection of the peak with a retention time of 8.6

minutes gave 0.066 g (1.0%) of an orange oil. The NMR

(Appendix A), IR, and UV spectra of this material were

identical to those of piperitenone. The identity of the

compound from A. filifolia of Rfc 8.6 minutes with

piperitenone was confirmed by an undepressed melting

point of a mixture of 2,4-dinitrophenylhydrazones of natural and authentic specimens. The preparation of this derivative is described later.

Collection of the GLC peak with a retention time of 4.0 minutes provided 0.070 g (1.1%) of a pale yellow ail. The NMR spectrum of this oil (Appendix A) revealed that it was an approximately 1:1 mixture of borneol and apparently one other compound. The sample was allowed to stand in an open container at room temperature for 3 days.

After this time nearly all of the borneol had sublimed off, as seen by the NMR spectrum of the residue. This NMR spectrum suggested that the compound remaining was 122 3,5,5-trimethylcyclohex-2-en-l,4-dione (34). The NMR spectral assignments can be found in Figure 16 and

Appendix A. Independent synthesis (described later) of 34 gave a compound with spectra (IR, NMR, UV, MS) identical to those of the material from A. filifolia. This identity was confirmed by an undepressed melting point of a mixture of the mono-phenyl-semicarbazones of natural and synthetic specimens. The derivative preparation is described later.

Preparation of (-f)-Verbenone 2.4- Dinitrophenylhydrazone

Essentially the procedure of Shriner et al. (1965, p. 253) was employed. The 2,4-dinitrophenylhydrazine solution consisted of 2,4-dinitrophenylhydrazine (0.4 g), concentrated sulfuric acid (2 ml), water (3 ml), and 95% ethanol (10 ml). To (+)-verbenone (authentic specimen,

1.0 g, 0.0067 mole) in 95% ethanol (20 ml) was added excess freshly prepared 2,4-dinitrophenylhydrazine solution (30 ml). Orange-red crystals of the 2,4-dinitrophenylhydrazone of verbenone soon formed and these were collected by filtration, washing with cold ethanol. One recrystallization from ethanol gave orange-red needles, m.p. 103-110°

(0.80 g, 0.0024 mole; 36%). Two more recrystallizations from the same solvent provided the pure 2,4-dinitrophenylhydrazone of (+)-verbenone (0.55 g, 0.0017 mole; 25%) as orange-red needles, m.p. 114-6°. The reported (Kergomard and Sandris, 1954) melting point is 116°. 123

The same procedure provided the 2,4-dinitrophenyl-

hydrazone (0.022 g, 0.000067 mole; 33%) of (-)-verbenone

from A. filifolia (0.030 g, 0.00020 mole) as orange-red

needles, m.p. 112-5°. The melting point of a mixture of

dinitrophenylhydrazones of natural and authentic verbenones

was undepressed (111-5°).

Preparation of Piperitenone 2.4- Dinitro phenylhydra zone

The procedure utilized in the preparation of (+ )-

verbenone dinitrophenylhydrazone was employed. Authentic

piperitenone (1.0 g, 0.0067 mole), prepared from geranic

acid (described earlier), gave, after two recrystallizations

from 95% ethanol, 0.72 g (0.0022 mole; 33%) of the 2,4- dinitrophenylhydrazone derivative as deep orange needles,

m.p. 183-4°. The reported (Naves, 1942) melting point is

184-4.5°. Similarly, piperitenone (0.035 g, 0.00070 mole) from A. filifolia gave the dinitrophenylhydrazone (0.028 g,

0.000085 mole; 12%) as deep orange needles, m.p. 181-3°.

The melting point of a mixture of 2,4-dinitrophenylhydrazones of natural and authentic piperitenone samples was un depressed (181-4°).

Preparation of g-Isophorone (Figure 17)

The procedure of Kharasch and Field (1941) was followed. Isophorone (Aldrich) was distilled under vacuum

(19 mm) through a 400-mm Vigreux column immediately before 124 use. The purified isophorone was collected over the range of 115-6°.

To a stirred suspension of magnesium turnings (19.8 g, 0.814 mole) in dry ether (150 ml) was added dropwise, at a rate which kept the ether refluxing, a solution of methyl iodide (117 g, 0.814 mole) in dry ether (150 ml).

Refluxing and stirring were continued for 15 minutes after addition of methyl iodide was complete and after this time most of the magnesium had dissolved. The Grignard reagent

(in ether) was then cooled to 5° and to it was added freshly fused ferric chloride (1.32 g, 1 mole %). The solution turned from dark gray to black when the ferric chloride was , added. A solution of freshly distilled isophorone (94.5 g,

0.685 mole) in dry ether (135 ml) was then added dropwise to the cooled mixture, at a rate which maintained the temperature between 10 and 12°. After addition was complete, the reaction mixture was refluxed for 2 hours, allowed to stand at room temperature overnight and then cooled to 5°. The Grignard adduct was hydrolyzed by the careful addition of chopped ice (~ 250 g) and then glacial acetic acid (50 ml). The ether layer was separated and the aqueous phase extracted with ether (200 ml). All ether layers were combined, washed with water (400 ml), 5% sodium hydroxide (400 ml), and then water (400 ml). The ethereal solution was dried over anhydrous magnesium sulfate and concentrated _in vacuo to provide 90 g of a yellow oil. The 125 latter was then distilled under vacuum (1.5-1.8 mm) through a 250-mm vacuum-jacketed Vigreux column, collecting four fractions:

Fraction # b.p. (°C) Weight (q) Composition'*" (by NMR)

1 35-38 5.0 100% HC 2 40-50 5.3 0-1:HC ; 5:1 3 53-58 24 p-I:I ; 6:1 4 65-68 26 p-I:I ; 1:10 pot residue — 29 mostly I

Fractions #2 and 3 were combined giving p-isophorone (29.3 g) of 85.3% purity. The yield of pure p-isophorone is then

28 g (0.20 mole; 41%). The NMR spectrum of the major component in this sample was consistent with that of 0- isophorone (see Appendix A).

Preparation of 4-Hydroxy-3.5.5-trimethylcyclohex- 2-en-l-one (Figure 17)

The procedure of Isler et al. (1956) wad employed.

To (3-isophorone (85.3% purity, 20.0 g; 0.123 mole pure (3- isophorone), cooled in ice, was added dropwise 88% formic acid (8 g). The mixture was stirred, maintaining the temperature at ~ 5°. To this was added dropwise,

1. I = isophorone; 0-1 = 0-isophorone; HC = hydrocarbon products, probably the following two: 126

maintaining the temperature between 10 and 12°, a mixture

of 88% formic acid (20 g) and 30% hydrogen peroxide (17 g,

0.15 mole pure peroxide). After addition was complete the

mixture was stirred at 5° for 2 hours and then at 25° (in a

water bath) overnight. Excess peracid was then destroyed

by the addition of sodium sulfite, sodium hydroxide was

added to pH 8-9, the mixture was stirred for one hour, and

then extracted with ether (2 x 200 ml). The ether extracts

were combined, washed with water (200 ml), dried over

anhydrous magnesium sulfate, and concentrated jLn vacuo to

provide 19.2 g of a yellow oil, which partially crystal

lized. The crystals (of undetermined composition) were

removed by filtration, washing with ether. Evaporation of

the ether from the filtrate gave a yellow oil (12 g) which,

by NMR (Appendix A), consisted of approximately 60%

isophorone and 40% 4-hydroxy-3,5,5-trimethylcyclohex-2-en-l- one. This was used without further purification for the next experiment.

Oxidation of 4-Hydroxy-3.5.5-trimethylcyclohex- 2-en-l-one; Preparation of the Ene-dione 34 (Figure 17) AAA AAA *' The crude (40%) 4-hydroxy-3,5,5-trimethylcyclohex-

2-en-l-one (11.5 g, 0.030 mole pure hydroxy-ketone), pre pared above, was dissolved in chloroform (500 ml). To this was added active manganese dioxide (80 g, 0.92 mole), prepared according to Ball et al. (1948). The resulting 127 suspension was stirred under anhydrous calcium chloride at

room temperature overnight, filtered through Celite and the

filtrate was concentrated in vacuo. This resulted in 10.0 g

of a pale yellow oil, shown by NMR to consist of approxi

mately 40% 34 and 60% isophorone. This corresponds to a

yield for 34 of 4.0 g (0.026 mole; 87%), based upon the

amount of 4-hydroxy-3,5,5-trimethylcyclohex-2-en-l-one in

the starting material. The ene-dione 34 and isophorone were MAMA not readily separated by fractional distillation; separation

was accomplished by preparative GLC on Column II at a column

temperature of 150°. Pure 34, obtained in this way, had NMR MAMA (Appendix A, Figure 16) IR (Neat: 1680; and 1625 cm""'") , UV

(Appendix B), and mass (m/e 152, M+; 96, base; 82; and 68)

spectra in accord with 3,5,5-trimethylcyclohex-2-3n-l,4-

dione. The IR and UV spectra were identical to those

reported (Isler et al., 1956).

Anal. Calculated for c^i2°2: C' ^1.03 ; H, 7.95.

Found: C, 71.27 ; H, 8.12.

Preparation of Phenylsemicarbazide

The procedure of Wheeler (1932) was followed.

Phenylurea (34 g, 0.25 mole), 99-100% hydrazine hydrate

(25 g, 0.50 mole), and absolute ethanol (25 ml) were heated

together on a steam bath for 24 hours. The hot solution was decolorized with Norite, concentrated to a volume of 40 ml, and cooled in ice. This resulted in the crystallization of 128

phenylsemicarbazide (29 g, 0.19 mole; 76%) in 4 crops,

as colorless needles, m.p. 121-2°. The reported (Wheeler,

1932) melting point is 121-3°.

Preparation of the Mono-phenylsemicarbazone of 34 ^c-1 mm This derivative was chosen for the ene-dione 34 mm because it is the only one reported in the literature. The

procedure employed was that of semicarbazone preparation

(Shriner et al. , 1965, p. 253).

The synthetic diketone 34 (1.0 g, 0.0066 mole),

prepared as previously described, was dissolved in 95%

ethanol (10 ml). Water was added until the solution was

fairly turbid and the turbidity was removed by the addition of a few drops of ethanol. To this solution was then added

phenylsemicarbazide (1.0 g, 0.0066 mole) and the mixture was shaken vigorously, boiled briefly, and allowed to cool to room temperature. Cooling in ice resulted in the

crystallization of the phenylsemicarbazone derivative, which was collected by filtration. Recrystallization from

ethanol/water gave the pure mono-phenylsemicarbazone of 34

(1.4 g, 0.0048 mole; 72%) as colorless needles, m.p. 190-2°, softens, and 228-9°. The reported (Isler et al., 1956) values are 190-2° and 229-30°.

Preparation of this derivative of 34 (0.025 g,

0.00016 mole) from A. filifolia in the same way gave color less needles, m.p. 189-90°, softens, and 227-8°. A mixture 129 of the mono-phenylsemicarbazones of both natural and synthetic samples of 34 had an undepressed (189-91° and MAMA 227-9°) melting point.

Chloroform Extraction of the A. filifolia Plant Material

Artemisia filifolia. collected on 9/30/71, was ground in a Wiley mill and allowed to air dry (the heat generated during the milling process was sufficient to dry most of the plant material). The dry, ground plant (3.0 kg) was then extracted (6 x 400 g of plant by Soxhlet and 600 g by percolation) with chloroform. All chloroform extracts were combined and concentrated iii vacuo to provide 310 g of a brownish-green sludge. The latter was partitioned between

methanol (1.5 1) and hexane (1.5 1); the hexane layer was set aside. The methanol layer was washed with hexane (600 ml) and concentrated under vacuum to give a residue (190 g) consisting of a brown oil. This oil was taken up in chloroform (1 1) and extracted with 5% sodium hydroxide

(1 1). This resulted in a terrible emulsion requiring 2 days to separate. The chloroform layer was then extracted again with 5% sodium hydroxide (2 x 1 1), washed with saturated aqueous sodium chloride, dried over anhydrous sodium sulfate, and evaporated to give the neutral chloroform extract (105 g) of A. filifolia as a dark brown oil. This was set aside (its analysis is described later). 130

Chromatography of the Acidic Extract of Artemisia filifolia:Isolation of Acacetin (36) mm and the Keto-acid 35 mL All sodium hydroxide extracts of the chloroform

extract of A. filifolia were combined, acidified with

concentrated hydrochloric acid, cooled, and extracted with

chloroform (2 x 1 1). These chloroform extracts were

combined, washed with water (500 ml), dried over anhydrous sodium sulfate, and concentrated in vacuo to yield 85 g of

a dark brown oil. This acidic extract of A. filifolia was then chromatographed over silicic acid (Mallinckrodt, 100 mesh, 50 x 850 mm). Elution with 1:10 benzene/chloroform afforded the only crystalline material obtained in the chromatography. An NMR spectrum, of the latter showed the

presence of a flavone and the keto-acid 35. Recrystalliza- tion of this material from chloroform gave the flavone acacetin (36) (0.030 g) as a yellow powder, m.p. 246-7°, with decomposition. Another recrystallization from chloroform yielded pure acacetin as a pale yellow powder,

m.p. 259-261°, with decomposition. This sample was con firmed to be acacetin by means of an undepressed melting point of a mixture of natural and synthetic specimens.

The independent preparation of acacetin is described later.

Removal of the solvent from the mother liquor from which acacetin was crystallized afforded 0.183 g of a crystalline residue. The latter was shown by NMR to 131

consist of M\.YA35, still contaminated with acacetin. It was therefore acetylated by addition of acetic anhydride (10 ml)

and pyridine (10 ml) and stirring of the resulting mixture

at room temperature overnight. The mixture was then

diluted with water (50 ml), heated for 5 minutes and the

water, pyridine, and acetic acid were removed _in vacuo.

This resulted in a brown oily residue (0.185 g) which was

taken up in chloroform (200 ml) and extracted with 5%

sodium hydroxide (4 x 100 ml). The sodium hydroxide extracts

were combined, acidified, and extracted with ether (2 x 100

ml). These ether extracts were combined, washed with

water (100 ml), dried over anhydrous magnesium sulfate, and

evaporated to give 0.032 g of the keto-acid M\VA35 as yellow cubelets, m.p. 99-104°. Recrystallization from benzene pro

vided pure 35 as colorless cubelets, m.p. 106-8°. This c mm ' material was optically active, [oc]^ +128° (c 1.0, CHCl^), and, from this rotation, has the same absolute stereochemis try as the keto-acid 35 prepared by oxidation of optically active 1^ (described later) from A. filifolia. That the natural and synthetic specimens were identical was demon strated by an undepressed mixture melting point (described later) and identical spectra (IR and NMR).

The chloroform layer from which 35 had been J mm extracted was washed with water (100 ml), dried over anhydrous sodium sulfate, and evaporated to give acacetin diacetate (0.133 g) as yellow needles, m.p. 192-6°. 132

Recrystallization from ethanol gave pure material as colorless needles, m.p. 199-201°. This specimen and synthetic acacetin diacetate (described later) were identical as seen by an undepressed mixture melting point and superimposable NMR spectra.

Preparation of the Keto-acid 35 ^ um. The hydroxy-acid 17 ([cOH^ +33.8°, from A. J J mm u D ' — filifolia) (0.70 g, 0.0038 mole) was dissolved in dichloromethane (100 ml) to which was then added active manganese dioxide (5.0 g, 0.063 mole), prepared according to Ball et al. (1948). The resulting suspension was stirred under anhydrous calcium chloride for 24 hours, after which the reaction mixture was filtered through Celite and the filtrate evaporated to provide 35 (0.55 g, 0.0030 mole;

79%) as a yellow oil, which soon crystallized, m.p. 101-4°.

Recrystallization from benzene afforded pure 35 (0.42 g,

0.0023 mole; 60%) as colorless cubelets, m.p. 107-9°. The reported (Chretien-Bessiere and Retamar, 1963) melting point for racemic 35 is 107°. The material prepared in this way had NMR (Appendix A), IR (KBr: 3000, very broad; 1720 and

1660 cm-"1"), UV (Appendix B), and mass (m/e 182, M+; and

137, base, - (*H + CC^)) spectra in accord with structure r T 24 O / 35. This material was optically active, [ccj^ +128 (c www ' D

1.05, CHC13). 133

Anal. Calculated for <-ioH14°3: C' ®^.92 r H» 7.74. Found: C, 65.64 ; H, 8.06.

A mixture of this synthetic sample of 35 and the keto-acid isolated from A. filifolia had an undepressed

(107-9°) melting point.

Preparation of Anisic Anhydride

The procedure of Robinson and Venkataraman (1926)

was employed. Anisic acid (Eastman, 100 g, 0.658 mole) was stirred and refluxed for 18 hours with acetic anhydride

(300 g, 2.94 moles). After cooling of the mixture, no

crystals of anisic anhydride appeared; acetic acid and

excess acetic anhydride were therefore evaporated _in vacuo.

This gave an oily crystalline residue. Trituration with

ether, discarding the triturate, furnished anisic anhydride

(66.6 g, 0.262 mole; 81%) as colorless needles, m.p. 65-70°.

Recrystallization from ether gave pure material as colorless needles, m.p. 98-9°. The reported (Robinson and

Venkataraman, 1926) value is 99°.

Preparation of Sodium Anisate

Anisic acid (Eastman, 10.0 g, 0.0658.mole) was dissolved in water (200 ml) containing sodium bicarbonate

(5.5 g, 0.067 mole). The mixture was filtered and the

filtrate was concentrated to a volume of 70 ml and diluted with 95% ethanol (100 ml). Cooling in ice resulted in the 134

precipitation of sodium anisate (10.6 g, 0.0609 mole; 92.5%)

as tiny colorless needles.

Preparation of Acacetin (36)

The procedure of Robinson and Venkataraman (1926) was followed. Phloracetophenone (5.0 g, 0.030 mole), anisic anhydride (42.4 g, 0.17 mole), and sodium anisate (6.0 g,

0.034 mole) were heated (neat), with stirring, at 175-185° for 6 hours. After this the mixture was boiled with potassium hydroxide (32 g), water (20 ml), and 95% ethanol

(300 ml) for 25 minutes. The alcoholic alkaline solution was decanted (about 20 g of brown solid material remained stuck fast to the flask) into ice-water (1.5 1). This solution was then concentrated to a volume of approximately

1600 ml (to remove ethanol), acidified to pH 3, and extracted with ether (3 x 400 ml). The ether extracts were combined, washed with water (200 ml), and cooled. From this ethereal solution precipitated a tan powder (0.26 g). This powder, surprisingly, was pure acacetin (0.00092 mole; 3.1%), m.p.

260-2°, with decomposition. The reported (Robinson and

Venkataraman, 1926) melting point of acacetin is 262°. The material prepared in this way had NMR (Appendix A), IR

(KBr: 3160; 1670; and 1610 cm-1), UV (Scott, 1964;

Appendix B), and mass (m/e 284, base, M+; 152; and 132) spectra consistent with those of acacetin. 135

Anal. Calculated for ci6Hi2°5: c» 65.60 ; H, 4.25. Found: C, 65.31 ; H, 4.49.

That the flavone isolated from A. filifolia and

acacetin, prepared here, were identical was shown by an

undepressed (259-262°, dec.) mixture melting point and

identical spectra (UV and NMR).

Preparation of Acacetin Diacetate

The ether extract, from which the tan powder

(acacetin) separated, was evaporated in vacuo. and the

residue (7.1 g of a brown oil) was taken up in 3:1 95%

ethanol/glacial acetic acid and decolorized with Norite.

The clarified solution was then diluted with water (300 ml)

and concentrated to a volume of 350 ml. From the resulting

orange solution crystallized only recovered anisic acid

(5.78 g) as yellow feathery needles, m.p. 171-2°. The

mother liquor from which anisic acid was crystallized was

extracted with ether (3 x 200 ml). The ether extracts were

combined, washed with saturated sodium bicarbonate (6 x 100

ml), then water (100 ml), dried over anhydrous magnesium sulfate, and concentrated jLn vacuo to give a brown oily residue (0.63 g) which would not crystallize. The latter

was chromatographed over silica gel (Baker, 30 x 450 mm) and

elution with chloroform gave crude acacetin, which still

would not crystallize. This residue (0.52 g) was therefore acetylated with acetic anhydride (5 g) and pyridine (3 g), 136 by stirring at room temperature for 19 hours and then

refluxing for 2 hours. The resulting solution was poured

into ice-water (100 ml) and extracted with ether (2 x 100

ml). The ether extracts were combined, washed with 5%

hydrochloric acid (2 x 100 ml), 5% sodium hydroxide (2'x

100 ml), and then water (100 ml), dried over anhydrous

magnesium sulfate and evaporated to yield 0.60 g of a brown

oil. The latter was chromatographed over silica gel (30 x

200 mm) and elution with 1:3 chloroform/benzene finally

afforded the diacetate of acacetin (0.33 g), m.p. 185-193°.

Recrystallization from ethanol gave pure acacetin diacetate

as pale yellow needles, m.p. 200-2°. The reported (Pillon,

1954) melting point is 202-3°. The material prepared in

this way had NMR (Appendix A), IR (KBr: 1760; 1640; and 1610

cm-"*"), UV (Appendix B), and mass (m/e 368, M+; 326; and 284, base) spectra consistent with those of acacetin diacetate.

This synthetic specimen was identical to the material

prepared from acacetin isolated from A. filifolia. as shown by an undepressed (199-201°) mixture melting point.

Chromatography of the Neutral Chloroform Extract of Artemisia filifolia: Isolation of the Sesquiterpene Lactone Colartin

The residue (105 g) of the chloroform extract of A. filifolia. from which acidic compounds had been extracted, was chromatographed over silicic acid (Mallinckrodt 100 mesh, 50 x 860 mm). Elution with benzene gave monoterpene 137

lactones 12mm and 13,mm' obtained in the ratio 5:1,# respectively.r j The only crystalline, or otherwise homogeneous material obtained in this chromatography was eluted from the column

with chloroform. After evaporation of the eluent, this

material consisted of 8.0 g of a yellow, partially

crystalline oil. The latter was rechromatographed over silica gel (30 x 450 mm) and elution with 1:1 benzene/

chloroform afforded purified material (6.2 g), still yellow oily crystals. This material was further purified by repeated trituration with hot pentane (2 x 50 ml) and then hot hexane (2 x 50 ml). The triturates were carefully decanted off of the yellow oil remaining in the flask, combined, and concentrated in vacuo to provide 4.2 g of

colorless pasty crystals, m.p. 79-89°. Recrystallization from hexane gave colorless oily crystals, no better than before the recrystallization. These crystals were triturated with hot hexane (2 x 75 ml), discarding the oily residue. The triturate was boiled briefly with Norite, filtered, and concentrated to a volume of 50 ml. Crystals of purified material (2.2 g) formed at room temperature in the form of colorless (still slightly oily) tiny flakes,

m.p. 85-92°. Another recrystallization from hexane brought the melting point up to 103-5° and one more such process finally provided pure material (1.5 g) as colorless tiny

4 flakes, m.p. 107-8°, [a]^ +19.4° (c 0.88, CHC13). The reported (Irwin and Geissman, 1969) melting point and 138 specific rotation of the sesquiterpene lactone colartin are

107-8° and + 11.4°, respectively. The NMR (Figure 19 and

Appendix A), IR (CHCl^: 3509; and 1778 cm-"'"), and mass

+ (m/e 252, M ; 237, - *CH3; 234, - E^O; 219, - CCH3 + H20);

206; 191; 167; 121; 93; 55; 43, base; and 41) spectra of

this specimen from A. filifolia were consistent with those

expected and reported (Irwin and Geissman, 1969) for

colartin (37). urn i Anal. Calculated for ci5H24°3: c» 71.39 ; H, 9.59.

Found: C, 71.06 ; H, 9.67.

A sample of crude cqlartin (4 mg), m.p. 81-94°, was obtained (Geissman, 1973). It had an infrared spectrum identical to the material from A. filifolia. Recrystalliza- tion of this authentic specimen from hexane gave colorless tiny flakes, m.p. 106-7°. That the material isolated from

A. filifolia was in fact colartin was confirmed by an undepressed (106-8°) melting point of a mixture of natural and authentic specimens. APPENDIX A

NUCLEAR MAGNETIC RESONANCE SPECTRA

139 60 MHz NMR Spectrum of carvenolide (1) (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment f A 5.37 br.s 1 H-3 B 3.46 br.d 9 1 H—1 C 2.95 hextet 6,7.5,9 1 H-5 D 2.48 m 2 H-4,4' carvenolide ( E 1.82 br.s — 3 C-2 Me F 1.37 s — 3 C-6 Me G 1.29 s — 3 C-6 Me

B C

A /ajimaA—i • -

—i— —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of fenchone (neat)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 1.3-2.2 complex 7 H-2,2'3,3',4,5,5' B 1.07 s 3 C-l Me C 0.98 s 6 C—6 Me's fenchone

—I— ~t 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of fenchone oxime (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 9.34 s 1 =NOH B 1.6-2.2 complex 7 H-2,2',3,3',4,5,5' C 1.57 s 3 C-l Me D 1.15 s 6 C-6 Me's fenchone oxime D

-it

f— —i— -fh f 11.0 10.0 9.0 8.0 4.0 3.0 2.0 1.0 ppm (6) H N> 60 MHz NMR Spectrum of 0-fencholenic acid (in CCl^) COOH Chemical Shift Multiplicity J, Hz Protons Assignment

A 11.70 1 -COOH B 1.7-3.0 complex 6 H-2,2'.3,3',5,5' C 1.60 s 6 C-6 Me's D 1.23 s 3 C-l Me S-fencholenic acid C

VJ I

Jhr —I— —I— i I 13.0 12.0 11.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of a-fencholenic acid amide (in CCl^)

Chemical Shift Multiplicity J, Hz Protons .CONH, 5.30 H-2 2.33 -NH. 1.6-2.4 complex 1.74 C-3 Me a-fencholenic acid 1.30 C-6 Me's amide

, , 1 1 1 1 i j j 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 100 MHz NMR Spectrum of a-•fencholenic acid (in CC14)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 11.69 s 1 -COOH J ^ COOH B 5.20 br.s 1 H-2 C 1.6-3.0 complex 5 H-l,4,41,5,5' D 1.73 br.s 3 C-3 Me a-fencholenic acid E 1.08 s 6 C-6 Me's B

u ~fh A

. . / h-r- —|— 1— 12.0 11.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of cis- + trans-geranic acid (in CCl^) COOH Chemical Shift Multiplicity J, Hz Protons ' Assignment

A 12.0 s 1 -COOH B 5.63 s 1 H-2 C 5.07 br.s 1 H-6 D 2.0-2.5 complex 4 H-4,4',5,5' geranic acid E 2.17 s 3 trans-C-3 Me (cis + trans) F 1.92 s 3 cis- C-3 Me G 1. 63 s 3 C-7 Me H H 1.60 s 3 C-7 Me E

—ff- Lyv JU 4-

i— 1 -ff- —I— ( J- -I— 13.0 12.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) it^ as 60 MHz NMR Spectrum of piperitenone (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.73 cr,s 1 H-2 B 2.2-2.9 m 4 H-4,41,5,5' C 2.00 s 3 C-3 Me D 1.90 s 3 C-7 Me piperitenone E 1.79 s 3 C-7 Me

B A

j j , , 5 | , , , 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (5) 60 MHz NMR Spectrum of jo-methyl-a-methylstyrene (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 7.32 d 8.5 2 H-2,6 B 7.08 d 8.5 2 H-3,5 C 5.31 br.s 1 H-8 D 5.00 br.s 1 H-81 ]D-methyl-a-methyl- E 2.18 s 3 C-4 Me styrene F 2.05 s 3 C-7 Me

A B U I

j 1 1 i i i 1 i 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) H oo 60 MHz NMR Spectrum of filifolone (in CC14 )

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.36 br.s 1 H-3 B 3.97 br.s 1 H-l C 2.51 m - 3 H-4,4',5 D 1.75 br.s - 3 C-2 Me filifolone E 1.16 s - 3 C-6 Me F 1.08 s - 3 C-6 Me

_A. Jj^~

—i— —i— —,— —I— I !~ 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of isogeranic acid (in CCl^)

Chemical COOH Shift Multiplicity J, Hz Protons Assignment

A 9.67 br.s 1 -COOH B 5.2 m 2 H-4,6 C 3.00 s 2 H- 2, 21 D 2.67 br.t 8 2 H-5,5' E 1.77 d 1 3 C-3 Me F 1.63 s 3 C-7 Me G 1.60 s 3 C-7 Me

B a/1AV JV •ff-

i— --T—/A -1 —I— i 10.0 9.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) H Ul O 60 MHz NMR Spectrum of ct-fencholenic acid iodolactone (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 4.97 d 5 1 H-l B 1.6-3.1 complex 5 H-3,3',4,4',5 C 2.20 s 3 C-2 Me D 1.27 s 3 C-6 Me ct-fencholenic E 1.20 s 3 C-6 Me acid iodolactone

C D

JL -4

i —i 1 1 1 1 1 1 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 100 MHz NMR Spectrum of lactone 2 (in CCl.) AM 4 Chemical Shift Multiplicity J, Hz Protons Assignment :0 A 5.60 br.s 1 H-3 B 5.04 br.d 4 1 H-l C 2.77 q 4,7 1 H-5 D 2.38 m 2 H-4,4' lactone 2 E 1.82 d 1 3 C-2 Me F 1.24 s — 3 C-6 Me G 1.12 s — 3 C-6 Me

C D B J\ av AfVj

r— —1— -71— —J— 1 ~l— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 100 MHz NMR Spectrum of dihydrofilifolone (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 3.47 t 8 1 H-l B 2.41 t 7 1 H-2 C 2.4 m — 1 H-5 1 D 1.8 m — 4 H-3,3 ,4,4' dihydrofilifolone E 1.14 s _ 3 C-6 Me F 1.11 d 7 3 C-2 Me G 0.89 s — 3 C-6 Me

B.C A __/V_ u 4

i— —I —I— —i "1 —r~ 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 100 MHz NMR Spectrum of 9 + 10, from hydrogenolysis of 12 + 13 (CC1.) mm mm 4 Chemical Protons Shift Multiplicity J, Hz 9 10 Assignment ' m mm

A 5.30 br.s 1 H-5 COOH COOH B 5.07 br.s 1 ' H-5 C 2.3 m 2 2 H-5,51 H-3,3' D 1.8 m 3 3 H-1,6,61 H-1,6,6' 9 10 C-4 Me (from 12 + iTt E 1.61 s 3 3 C-4 Me mm mm F 1.15 s 3 C-2 Me G 1.05 s — 3 C-2 Me H 1.00 s 3 C-2 Me H I 0.97 s — 3 C-2 Me

A B M.

I -J —1— —I— —I— —I— —I— —I— -| 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of 2,4-dimethyl-l,3-pentadiene (neat)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.59 br.s 1 H-3 B 4.84 br.s 1 H-l C 4.68 br.s 1 H-l' 2,4-dimethyl-l,5- D 1.76 s 9 C-4 Me's, C—2 Me pentadiene

B C

A Ml 4

I i I —I— —I— —i— -I 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of synthetic 9 (+10) (in CCl^)

Chemical Protons Assignment Shift Multiplicity J, Hz 9 10 9 10 ' Nti /MAM M/l mm A 5.30 br.s 1 H-5 COOH COOH B 5.05 br.s 1 H-3 C H-5,5' H-3,3' 10 2.3 m 2 2 m mm D 1.8 m 3 3 H-l,6,6" H-l,6,6' (synthetic) E 1.63 s 3 3 C-4 Me C-4 Me F 1.15 s 3 C-2 Me G 1.04 s _ 3 C-2 Me H 0.97 s 3 3 C-2 Me C-2 Me

< ~ < • i i t I | 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz M Spectrum of a-cyclogeranic acid (in CC14)

Chemical .COOH Shift Multiplicity J, Hz Protons Assignment

A 11.90 s - 1 -COOH B 5.53 br.s - 1 H-3 & C 2.50 s - 1 H-l D 1.6-2.3 m - 4 U-4,4',5,5' a-cyclogeranic acid E 1.67 s - 3 C-2 Me F 0.96 s - 3 C-6 Me G 0.89 s - 3 C-6 Me

B \Mf -ff- TV.

i i // 1 1 T 1 1 » 1 12.0 11.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of ^ (from chrysanthenone) (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 9.46 br.s 1 -COOH COOH B 5.05 br.s 1 H-3 C 2.3 m 2 H-5,5' H-1,6,6' 9 D 1.8 m 3 M E 1.63 s 3 C-4 Me F 1.15 s 3 C-2 Me G 0.97 s 3 C-2 Me

B VJ _A_ -ff- Jl

-r-4S- i —I— —I— —I— —I— 10.0 9.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 100 MHz NMR Spectrum of carboxylic acid 10 (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.30 br.s 1 H-5 B 2.3 m 2 H-3,3' COOH C 1.8 m 3 H-1,6,6' D 1.61 s 3 C-4 Me 10miM E 1.05 s 3 C-2 Me F 0.97 s 3 C-2 Me

B A

i "T— —I— —I— I— —I— -I 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of cis + trans-11 (in CCl,) c 1 1 M 4 Chemical Shift Multiplicity J, Hz Protons Assignment

A 11.93 1 -COOH B 1.2-2.3 complex 8 H-l,3,31,4,5,5',6,6' COOH COOH C 1.03 s 3 cis-C-2 Me D 1.00 s 6 trans-C-2 Me cis- + trans-11 — ——— mm E 0.98 s 3 cis-C-2 Me F 0.89 d 3 C-4 Me D

« 1—/ f—i 1 1 1 1 , r 12.0 11.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of cis-11 (in CCl.) _ mm 4 Chemical Shift Multiplicity J, Hz Protons Assignment

A 11.66 br.s 1 -COOH B 1.3-2.2 complex 8 H-l,3,3',4,5,5',6,6- COOH C 1.03 s 3 C-2 Me D 0.98 s 3 C-2 Me cis-11 E 0.89 d 6 3 C-4 Me

A yv. -ih

, , // r- —r~ i —r~ —I— ~1— I 12.0 11.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of 17 (in CDC13 + 3 dps DMS0-dg)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.70 br.s 2 -OH + -COOH B 5.07 br.s 1 H-3 COOH C 4.00 br.t 7 1 H-5 D 2.43 H—1 dd 3,12 1 17wm E 1.9-2.2 m — 2 H-6,6' F 1.67 s — 3 C-4 Me G 1.11 s — 3 C-2 Me H 0.97 s — 3 C-2 Me H

I —1— ~l— v t —I— i 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) H N) 60 MHz NMR Spectrum of 18 (in CCl. + 3 dps DMSO-d ) c wi/w 4 c 6c ..OH Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.30 br.s H-5 1 COOH B 3.47 br.s 1 H-3 C 2.24 m 3 H-l,6,6' D 1.70 s 3 C-4 Me 18 E 1.00 s 3 C-2 Me mm\ F 0.93 s 3 C-2 Me

A A 4-

< 1 1 1 1 1 1 1 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of lactone 12 (in CC1.) ^ mm 4 Chemical Shift Multiplicity J, Hz Protons Assignment JyO A 5.10 br.s 1 H-3 B 4.40 m 1 H-5 C 2.3 m 3 H—1,6,6' D 1.80 d 1 3 C-4 Me lactone 12 E 1.08 s 6 C-2 Me's

A B

A fv i

I " i i I I T" I I > 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of lactone 13 (in CC1.) AM AM *± Chemical Shift Multiplicity J, Hz Protons Assignment O'1" A 5.43 m 1 H-5 B 3.87 br.s 1 H-3 C 2.3 m 3 H-1,6,6' D 1.80 3 C-4 Me lactone 13 s Utnti E 1.20 s 3 C-2 Me F 1.10 s 3 C-2 Me

B JV 1 X

—I— —I —i— —I— —:— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 50 MHz NMR Spectrum of the hydroxy-ester 19 (in CCl^) HQ.. Jt Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.02 br.s .. 1 H-3 B 3.91 br.t 7 1 H-5 COOMe C 3.62 s — 3 COOMe D 3.45 br.s — 1 -OH 19 E 2.39 dd 4,10 1 H-1 WAM\ F 1.8-2.3 complex — 2 H-6,6' G 1.70 s _ 3 C-4 Me H 1.06 . s — 3 C-2 Me I 0.91 s - 3 C-2 Me

A /vr\

—i— —I— —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of the keto-ester 20 (in CC1 ) F ATM 4A Chemical Shift Multiplicity J, Hz Protons Assignment

A 6.23 br.s - 1 H—3 B 3.63 s - 3 -COOMe COOMe C 2.0-3.0 complex (ABX) - 3 H—1,6,6' D 1. 68 s - 3 C-4 Me 20 E 1.23 s - 3 C-2 Me AM AAA F •1.07 s - 3 C-2 Me

JL. J

i— —i— 1 —I— I— —I— —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectarum of verbenone (in CC14)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.61 q 1 1 H-2 B 2.3-3.0 complex - 4 H-4,6,7,7" C 1.99 d 1 3 C-3 Me D 1.47 s _ 3 . C-5 Me E 0.93 s — 3 C-5 Me

B Jj A_ 4-

• ••ill i 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.33 br.s 1 H-4 B 2.53 br.s 4 H-2,5,5',6 C 1.69 br.s 3 C—3 Me D 1.18 s 6 C-7 Me1s chrysanthenone

i— —i -1 I —I —I— I —I— T 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of chrysanthenone epoxide (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 3.17 br.s - 1 H-2 B 2.75 m - 1 H-6 C 1.9-2.4 complex - 3 H-4,5,5' D 1.37 s - 3 C-3 Me chrysanthenone E 1.30 s - 3 C-7 Me epoxide F 1.10 s - 3 C-7 Me

\J I

i i I I -i 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of the hydrocarbon "HC" (in CCl^)

Chemical Shift Multiplicity J, Hz Protons

A 5.01 br.s 1 H-4 B 2.0-2.4 m 2 H- 5, 51 C 1.74 d 1 3 ' C-3 Me D 1.2-1.6 complex 2 . H-2,6 "HC" E 1.02 s 3 C-l Me F 0.79 s 3 C-l Me

B A

i— —i— i —I— —T -i 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) H -J H 60 MHz NMR Spectrum of tricyclic ketone "TCK" (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 2.50 d 3 1 H-6 B 2.23 br.s — 1 H-2 C 2.20 dd 3,8.5 1 H-7(3 1 o D 1.2-2.2 complex O - 4 H-3,3 ,4,4' "TCK t C E 1.67 d 1 H-7CC • F 1.11 s — 3 C-8 Me G 0.95 s — 3 C-5 Me

A B

i r- 1 1 i 1 r 1 1 r- 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of isopiperitenone (in cci4)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.75 br.s 1 H-2 B 4.84 br.s 1 H-8 C 4.68 br.s 1 H-81 D 2.80 t 4 1 H-6 isopiperitenone E 2.30 m 2 H-4,4' F 1.9-2.2 m 2 H-5,5* G 1.94 s 3 C-3 Me H 1.74 s 3 C-7 Me

B C D E F A _AA_ A—/V-'VJ \w~ I

~l— -j r~ —i— —I -i 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum^ of ketone 22MMWl (in CC1-)4 Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.00 br.s 1 H-4 B 2.8 m 2 H-2,6 C 1.9-2.8 complex 2 H-7,71 D 1.76 d 1.5 3 C-3 Me E 1.09 s 6 C—5 Me1s

A -i

i— —I —I— —1 —r~ -I I 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of a-epoxy-a-lactone 12 (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 3.90 dd 2, 5.7 1 H-5 B 1.6-2.7 complex 4 H-l,3,6,6' C 1.43 s 3 C-4 Me D 1.07 s 6 C—2 Me 1s a,, a-epo xy -1 ac to ne

—i— —I— I —r~ 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 H ppm (6) U1 60 MHz NMR Spectrum of j3-epoxy-a-lactone 12 (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 4.53 dd 1.4 1 H-5 B 1.7-2.7 complex 4 H-l,3,6,6' C 1.43 s 3 C-4 Me D 1.16 s 6 C— 2 Me1 s p,a-epoxy-lactone

JA. —J KJ In—

j— —»— i i i i 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of epoxides 23MMM + 24WA.VA (in CC1.)rt Chemical Shift Multiplicity J, Hz Protons Assignment

A 1.6-3.3 complex 5 H-2,4,6,7,7' (23+24) B 1.43 s _ 3 C-5 Me (Ml) 6 C 1.17 s _ 3 C-3 Me _ C-3 Me m) 23 24 D 1.09 s 3 AW.W /MMA E 1.07 s 3 C-3 Me m) (^) F 1.02 s 3 C-3 Me D C

f 1 V \ 1 1 1 1 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of pure epoxide 23 (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignnv

A 3.10 br. 1 H-6 B 2.52 s 1 H-4 C 2.5 m 1 H-2 1 D 1.8 ra 2 .H-7,7 epoxidee 23 E 1.43 s 3 C-5 Me mm F 1.17 s 3 C-3 Me G 1.07 s 3 C-3 Me

i 1 i ; t » r 1 1 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of senecioic acid (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment )=\ A 12.20 br.s - 1 -COOH ' COOH B 5.64 br.s - 1 H-2 C 2.19 s - 3 C-l Me D 1.93 s - 3 C-l Me senecioic acid

W j. -A m A —

1 J / f- J 1 1 1- 13.0 12.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) H -O VO 60 MHz NMR Spectrum of 3-methyl-3-phenylbutanoic acid (CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment ,COOH A 11.60 br.s 1 -C00H B 7.2 m 5 Ar-II's C 2.59 s 2 H-2,2' 3-methyl-3-phenyl- D 1.48 s 6 C-3 Me's butanoic acid

k -f

B 13.0 12.0 11 I-

i— —I— —I— —I— —I— —I— I —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of ethyl senecioate (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment V3 it ' COOCH CH A 5.59 br.s 1 H-2 2 3 B 4.05 q 2 H-4,41 C 2.07 s 3 C-1 Me ethyl senecioate D 1.86 s 3 C-1 Me E 1.21 t 3 C-4 Me

A u

i 1 1 » 1 1 1 1 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of ethyl 3-methyl-3-phenylbutanoate (CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment * OOCH2CH3 A 6.9 m 5 Ar-H's B 3.87 q 2 H-4,4' C 2.40 s 2 H-2,2' ethyl 3-methyl-3- D 1.37 s 6 C-3 Me1s phenyl-butanoate E 1.16 t 3 C—4 Me

ilP JJu r i

i— —I— —I— —i— I —I— 1- 1 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of 28 (in COL.)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 4.2 br.s 1 -OH B 3.30 br.s 1 H_3 C 1.5-2.3 complex 5 H_1,5,5',6,6' D 1.36 s 3 C-4 Me 28 E 1.00 s 6 C-2 Me1s

sample shaken with D2O

A /V

I 1 1 ' I t I I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of iodolactone 30 (in CC1.) c NAM 4 Chemical Shift Multiplicity J, Hz Protons Assignment

A 4.07 1 H-3 B 1.8-2.3 complex 5 H—1,5,5* 6.6' C 1.51 s 3 C-4 Me D 1.27 s 3 . C-2 Me E 1.05 s 3 C-2 Me

—! I— —r~ 8.0 7.0 6.0 5.0 4.0 3.0 ppm (6) 60 MHz NMR Spectrum of 1,8-cineole (neat)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 1.3-2.3 complex - 9 B 1.13 s 6 C 0.93 s - 3 1,8-cineole no assignments made

I —5 -I— —I— —I— l —r~ 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of isophoron^ (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment 5L A 5.73 br.s 1 H-2 B 2.17 br.s 2 H-4,4' C 2.09 s 2 H-6,6' D 1.92 s 3 C-3 Me isophorone E 1.02 s 6 C-5 Me's

i 1 1 1 1 f 1—; 1 > 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of camphor (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 1.2-2.4 complex - 7 B 0.94 s 3 C 0.S3 s 6 camphor no assignments made

I 1 T • 1 1 1 1 1 | 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 100 MHz NMR Spectrum of 1:1 lactone mixture 12 +13 (in CC1.) ftnrcn AMANV Chemical Shift Multiplicity J, Hz Protons Assignment

see data for pure 12 and 13

12 13 mm mm

ATLjvJI

—I— —I— —r~ ... I —r— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR SDectrumf of borneol + 34 mixture (from —A. filifolia) (CC1A)

Chemical Shift Multiplicity J, Hz Protons Assignment

for assignments, see NMR spectra of borneol and 34 borneol + . /Wl34 YA mixture

A 4-

—J— —J— —I 1— I 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of borneol (in CCl^)

Chemical OH Shift Multiplicity J, Hz Protons Assignment

no assignments made

borneol

_/V\_

r- 1— —,— i— —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) O 60 MHz NMR Spectrum of 2 tetramethylcyclohexadienes (in CCl^)

Chemical Shift Multiplicity J, Hz Protons

A 5.83 br.s 1 B 5.43 br.s 1 C 4.98 br.s 1 D 4.62 br.s 1 E 1.6-2.2 m 16 F 0.93 s 6 G 0.89 s 6

no assignments made

A B CD ^vA_AA I

—I— —I— —I— —r~ 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of (3-isophorone (in CCl^)

Chemical Shift Multiplicity J, Hz Protons Assignment

A 5.40 br.s 1 H-4 B 2.62 s 2 H-2,2' C 2.22 s 2 H-6,61 D 1.70 s 3 •C-3 Me 3-isophorone E 1.01 s 6 C-5 Me's

A A u i-

i— —I— i —I— i —I— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of 4-hydroxy-isophorone (in CC14>

Chemical Shift Multiplicity J, Hz Protons Assignmer

A 5.75 br.s —• 1 H-2 B 3.95 br.s 1 H-4 C 2.23 s — 1 ' H-6 D 2.18 s — 1 . H-6' (contaminated with E 2.03 d 1 3 C-3 Me isophorone) F 1.25 br.s — 1 -OH G 1.08 s — 3 C-5 Me H 1.01 s - 3 C-5 Me

other peaks due to isophorone (which see) G

A B yv. J-

i— —i— i —i— -i— -I 8.0 7.0 6.0 5.0 4.0 3.0 .0 1.0 0 H ppm (6) VD W 60 MHz NMR Spectrum of ketoisophorone (34) (in CC1.) ft Chemical Shift Multiplicity J, Hz Protons Assignment

A 6.46 br.s 1 H-2 B 2.63 s 2 H-6,61 C 1.98 s 3 ' C-3 Me 6 .C-5 Me's 34 D 1.18 s mm,

A. L / I

— —i— i r —i— 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm (6) 60 MHz NMR Spectrum of keto-acid 35 (in CDC1,) e mm 3 Chemical Shift Multiplicity J, Hz Protons Assignment

A 9.6 br. — 1 -COOH B 6.38 br.s — 1 H-3 C 2.4-3.0 complex — 3 ' H-1,6,6 COOH — . C-4 Me 35 D 1.77 s 3 WAflM E 1.33 s — 3 C-2 Me F 1.13 s _ 3 C-2 Me

-f

i —r~f h —I— —I— —I— I 10.0 9.0 6.0 5.0 4.0 3.0 ppm (6) 60 MHz NMR Spectrum of acacetin (36) (in pyridine-dj-) _____ MAMA 3 Chemical Shift Multiplicity J, Hz Protons Assignment

A 7.87 d 9 2 H-21,6' B 7.00 d 9 2 H-31,51 C 6.78 s _ 1 H-3 D 6.69 d 1.5 1 H-8 E 6.56 d 1.5 1 H-6 F 3.73 s — 3 -OMe

—r— —r~ 8.0 7.0 6.0 5.0 4.0 3.0 ppm (6) H va CTl 60 MHz NMR Spectrum of acacetin diacetate (in CDCl^) Me Chemical Shift Multiplicity J, Hz Protons Assignment l :0 s! '0> ^5 A 7.75 d 9 2 H-2',6' B 7.26 d 1.5 1 H-6 e^5s C 6.93 d 9 2 H-31,5' 5T D 6.80 d 1.5 1 H-8 OAc E 6.61 s 1 H-3 F 3.80 s 3 -OMe acacetin diacetate G 2.40 s 3 -OAc H 2.30 s 3 -OAc H

t 1 1 i 1 r~ —I- -» 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 ppm (6) 60 MHz NMR Spectrum of colartin (37) (in CDC1_) f_ mm 3 Chemical Shift Multiplicity J, Hz Protons

A 4.06 m 1 H-6 B 2.93 br.s 1 -OH C 1.33 s 3 C-4 Me D 1.20 d 7 3 C-ll Me E 0.99 s 3 C-10 Me

r— —I t —I— —r— 8.0 7.0 6.0 5.0 4.0 3.0 ppm (6) H VD 00 APPENDIX B

ULTRAVIOLET SPECTRA

199 4.0 -

3.0 COOMe 20 in 100% EtOH 235 nm (log s 4.00) max ^ tn 0 2.0

1.0

0.0- i i t 200 220 240 260 280 300 320 wavelength (\), nm

to o o 4.0-

34 3.0" mm, in 95% EtOH X „ 238 nm (log e 4.16) max 3

2.0 0> o

1.0-

0.0- 200 220 240 260 280 300 320 wavelength (X), nm to o H 4.C-

3.0. COOH

mm in 100% EtOH X 242 nm (log e 4.10) max ^

2.0"

1.0-

0.0— 200 220 240 260 280 300 wavelength (X), nm to o to in 100% EtOH OMe 5.01- 100% EtOH + added NaOEt

36 (acacetin) NANA

4.0

CP o basic 3.0 X 330 (4.32) X 376 (4.15) max max X 298 (4.21) \ 295 (4.32) max max X 269 (4.31) X 278 (4.52) max maxv 2.0

\ 1.0 200 220 240 260 280 300 320 340 360 380 wavelength (X), nm N> O OJ 5.0 "

4.0 .OMe

AcO tn o 3.0 OAc 0 acacetin diacetate in abs. EtOH 341 (log 4.32) max 320 (log 4.30) 2.0 293 (log 4.28) 264 (log 4.24) 246 (log 4.22)

1.0. 200 220 240 260 280 300 320 340 360 380 wavelength (X), nm to o Ji". REFERENCES

Altman, L. J., Kowerski, R. C., and Rilling, H. C., J. Amer. Chem. Soc.. 93. 1780 (1971). ' MAAW' Ball, S., Goodwin, T. W., and Morton, R. A., Biochem. J.. 42, 516 (1948). MM ' Barnett. W. E., and Son. W. H.. Tet. Letters. 1972. 1777. ' ' ' ' ' MAMAWMM' Bates. R. B., and Feld. D.. Tet. Letters, 1965. 4875. ' ' ' ' ' mwwww' Bates, R. B., Onore, M. J., Paknikar, S. K., Steelink, C., and Blanchard. E. P., Chem. Comm.. 1967, 1037. ' ' ' MA MAMAMA ' Bates, R. B., and Paknikar, S. K., Tet. Letters, 1965, 1453. ' ' ' ' ' M\,VWM«A' Beereboom, J. J.. J. Orcr. Chem., 30. 4230 (1965). ' ' ' MAMA ' Berney, T. H., and Peebles, R. H., "Arizona Flora," University of California Press, Berkeley, 1951.

Bohlmann, F.. Bornowski. H., and Arndt, C.. Ann., 668, 51 (1963).

Bohlmann, F., and Grenz, M., Tet. Letters. 1970. 1453. ' ' ' ' ' MAMAMAMA Bohlmann, F., and Zdero, C., Ber., 102. 2211 (1969). ' ' ' ' ' MANAMA' Carlson, R. G., University of Kansas, personal communica tion, 1973.

The Chisso Company, British Patent 1076304 (1967); Chem. Abstr.. 68. P48177f (1968). ' MAMA ' Chretien-Bessiere, Y., and Retamar, M. J.-A., Bull. Soc. Chim. France. 19 63. 884. ——————' WA.WNA"*' Clayton, R. B., Quart. Revs.. 19, 168 (1965).

Coates, R. M. , and Robinson, W. H. , J. Amer. Chem. Soc. ." 94, 5920 (1972). mm

Cockburn, G. B.. J. Chem. Soc.. 75. 501 (1899). ' ' ' MAMA'

205 206 Corey, E. J., Shirahama, H., Yamamoto, H., Tereshima, S., Venkatesawarlu, A., and Schaaf, T. K. , J. Amer. Chem. Soc.. 93. 1491 (1971). ' MWWi Corey, E. J., Weinhenker, N. M., Schaaf, T. K., and Huber, W.. J. Amer. Chem. Soc.. 91. 5675 (1969). ' ' WWWl' Erman. W. F., J. Amer. Chem. Soc., 89, 3829 (1967). ' ' ' M,VA'

Erman, W. F., Wenkert, E., and Jeffs, P. W.f J. Org. Chem.. 34, 2196 (1969). M' Fieser, L. F., and Fieser, M., "Reagents for Organic Synthesis," John Wiley and Sons, Inc., New York, 1967.

Geissman, T. A., University of California, Los Angeles, personal communication, 1973.

Geissman, T. A., and Crout, D. H. G., "The Organic Chemistry of Secondary Plant Metabolism," Freeman and Cooper & Co., San Francisco, 1969.

Hall, H. H., and Clements, F. E., "The Phylogenetic Method in Taxonomy," Judd and Detweiller, Inc., Washington, D. C., 1923.

Howard.' G. A.. and Stevens. R.. J. Chem.———————— Soc.. I960.mamaaawa 161

Hurst. J. J.. and Whitham. G. H.# 1J. Chem. Soc., 1960,MAMAMAMA 2864.

Irwin. M.^ A., and Geissman. T. A., Phytochemistry.1 1969.mamamana 2411.

Isler, O. , Lindlar, H. , Montavon, M. , Riiegg, R. , Saucy, G. , and Zeller. P.. 'Helv. 1 Chim. Acta. M\M39. 2041 (1956). Jitkow. O. N. . and Bocrert. M. J. , J. Amer. Chem. Soc. , 63. 1981 (1941).

Kergomard, A., and Sandris, C., Bull. Soc. Cbim. France. 1954, 1260. WA/W\WA«\' Kharasch. M. S., and Field, E. K. J, Amer, Chem, Soc,. 63, 2308 (1941).

Kotake, M. and Nonaka, H. Ann.. 607 153 (1957). 207

March, J., "Advanced Organic Chemistry: Reactions, Mechanisms and Structure," McGraw-Hill Co., New York, 1968.

Martin, P. S., and Steelink, C., Bull. Ecol. Soc. America. 41, 129 (1961). mm' McCaughey, W. F., and Buehrer, T. F., J. Pharm. Sci.. 50, 658 (1961).

Meinwald, J., Seidel, M. C., and Cadoff, B. C., J. Amer. Chem. Soc.# 80, 6303 (1958). ' mm' Micovic, V. M., Org. Syn., Coll. Vol. II, 264 (1943). f » a '— ' MmMmmmmmmmmmm'

Mondon,It A. and Teege,t G. I Ber..__ I 91,I 1014 (1958).

Naves.» Y. R..» Helv.1 •" 1 " 11 Chim. Acta, 25,W" 732 (1942). Naves. Y. R.. Bull. Soc. Chim. France. 1960. 1741. ' ' ' mmmm' Newman, A. A., ed., "The Chemistry of Terpenes and Terpenoids," Academic Press, New York, 1972.

Onore, M. J., Ph.D. Dissertation, The University of Arizona, Tucson, Arizona, 1967.

Pillon. D.. Bull. Soc. Chim. France, 1954, 9. ' ' ' mmmm Popjak, G., Edmond, J., Clifford, K., and Williams, V., J. Biol. Chem.. 244. 1897 (1969). ' mmm' Poulter, C. D., Muscio, O. J., Spillner, C. J., and Goodfellow, R. G., J. Amer. Chem. Soc.. 94, 5921 ( > MAMA 1972).

Rappoport, Z., ed., "Handbook of Tables for Organic Compound Identification," Third Edition, Chemical Rubber Co., Cleveland, 1967. Richards, J. H., and Hendrickson, J. B., "The Biosynthesis of Steroids, Terpenes and Acetogenins," W. A. Benjamin, Inc., New York, 1964.

Rilling, H. C., Poulter, C. D., Epstein, W. W., and Larsen, B., J. Amer. Chem. Soc., 93, 1783 (1971). ' ' mm'

Robinson,* R.,. * and Venkataraman, K. J. Chem.• Soc.. mYVApVYmn1926, b J 4 • 208

Scott, A. I., "Interpretation of the Ultraviolet Spectra of Natural Products," Pergamon Press, London, 1964.

Shriner, R. L., Fuson, R. C., and Curtin, D. Y., "The Systematic Identification of Organic Compounds," Fifth Edition, John Wiley & Sons, Inc., New York, 1965.

Smith, L. I., Prichard, W. W., and Spillane, C. J., Org. Syn.. Coll. Vol. Ill, 302 (1955).

Stork, G., and Burgstahler, A. W., J. Amer. Chem. Soc., 77, 5068 (1955).

Takemoto. T.. and Nakaiima. T.. Yakuqaku Zasshi. 77. 1339 (1957). ^

Tammami, B., M. S. Thesis, The University of Arizona, Tucson, Arizona, 1970.

Teresa, J. de P., Bellido, H. S., and Bellido, I. S., Anales Real Soc. Espan. Fis. Quim. (Madrid). Ser. B. 58, 339 (1962). '

Thomas. A. F.. and Willhalm. B.. Tet. Letters. 1964. 3775. ' ' ' ' ' mmmm' van Tamelen, E. E., and Schwartz, M. A., J. Amer. Chem. Soc., 93, 1780 (1971). —— ' mm van Tamelen. E. E. . and Shamma, M. . J. Amer. Chem. Soc. . 76, 2315 (1954).

Wada. T., Chem. Pharm. Bull. (Tokyo), 13, 43 (1964). ' ' ' ' mm' Wallach, 0., Ann., 269, 332 (1892). ' ' ' mmm' Wallach, O., Ann., 305, 249 (1899). ' ' — ' /MWAMA' Wheeler. A. S.. Orcr. Svn.. Coll. Vol. I. 450 (1932). ' ' 3 '— ' mmmmmmmmmmmm' Wolinsky, J., Purdue University, personal communication, 1971.

Woodward, R. B., and Hoffmann, R., "The Conservation of Orbital Symmetry," Verlag Chemie , Weinheim, 1970.

Zalkow, L. '"H. , Brannon, D. R. , and Uecke, J. W. , J. Org. Chem.. 29, 2786 (1964). and references cited therein. ' mm' ' Zarahami. N. S.. and Heinz. D. E.. Phvtochemistry, 1971, o "71? t? ' ' MMMMiYA ^ /OO*