University Microfilms, Inc., Ann Arbor, Michigan TERPENOID NMR STUDIES: NMR PARAMETERS FOR

TERPENOID NMR STUDIES: NMR PARAMETERS FOR BICYCLO(3.1.1)HEPTANES AND REVISED STRUCTURES FOR ARCHANGELIN AND PEREZONE

Item Type text; Dissertation-Reproduction (electronic)

Authors Thalacker, Victor Paul, 1941-

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 29/09/2021 07:42:31

Link to Item http://hdl.handle.net/10150/290208 This dissertation has been microfilmed exactly as received THALACKER, Victor Paul, 1941- TERPENOID NMR STUDIES: NMR PARAMETERS FOR BICYCLO[3.1.1]HEPTANES AND REVISED STRUCTURES FOR ARCHANGELIN AND PEREZONE.

University of Arizona, Ph.D., 1968 Chemistry, organic

University Microfilms, Inc., Ann Arbor, Michigan TERPENOID NMR STUDIES: NMR PARAMETERS FOR

BICYCLo[3.1.1]HEPTANES AND REVISED STRUCTURES

FOR ARCHANGELIN AND PEREZONE

Victor Paul Thalacker

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 68 THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my

direction by Victor Paul Thalacker

entitled Terpenoid MR Studies': NMR Parameters for 6107010" [jS. 1. i] heptanes and Revised Structures for Archangelin and Perezone be accepted as fulfilling the dissertation requirement of the

degree of Doctor of Philosophy

7C Dissertation Director Date

After inspection of the dissertation, the following members

of the Final Examination Committee concur in its approval and

recommend its acceptance:*

Aut.. 1.IW7

2 7. JU7 l?C7

*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. 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 bor rowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permlssiont provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or re production 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 in terests of scholarship. In all other instances, however, permission must be obtained from the author.

SlGNEDt ACKNOWLEDGMENTS

The author wishes 4to express his gratitude to Dr. Robert B.

Bates for his counsel and aid throughout the course of this research and to Dr. S. K. Paknikar for his contributions to the synthesis prob lem. Thanks are also given to Mr. H. S. Craig for writing the plotter program.

Support for this research was provided by Public Health Service

Grant No. GM-11721.

iii TABLE OF CONTENTS

} Page

LIST OF ILLUSTRATIONS vi

LIST OF TABLES viil

ABSTRACT ix

INTRODUCTION 1

Nuclear Magnetic Resonance Spectra of Terpenoids I

Archangel in 4

PsoraLens 6

Peiezone ...... 7

DISCUSSION 9

NMR Spectral Parameters in Bicyclo^3.1.llheptanes: Ot-Pinene, Myrtenal, and Verbenone •••••• 9

Chemical Shifts 10

Coupling Constants 12

Terpenoid NMR Spectral Compilation ..... 17

Archangelin 17

Psoralens 25

Perezone 27

EXPERIMENTAL 28

General Methods ..•• 28

NMR Spectral Parameters in Bicyclo^3.1.lJheptanes 28

Terpenoid NMR Spectral Compilation 29

iv V

TABLE OF CONTENTS--Continued

Page

Synthesis of Archangelln ...... 29

2,6-Dimethyl-5-heptene-2-ol (XVII) 29

Ot-Cyclogeraniolene (XVIII) ••••«•••• 30

Acetylcyclogeranioiene (XX) , . .. 31

Regeneration of XX From Its Semicarbazone 32

^-Cyclolavandulic Acid (XXI) . .. . 32

Methyl 0-Cyclolavandulate •••••••«••• ... . 33

^Cyclolavandulol (XV) 34

0-Cyclolavandulyl Acetate .. 34

P-Cyclolavandulyl Bromide . 35

Sodium Salt of Umbelliferone (XXIV) 35

{9-Cyclolavandulyl Umbelliferonyl Ether (XXV) 36

Acetic Acid Cleavage 37

Cleavage of Archangel in ••.••••••••••••• 37

Reduction of Psoralen Mixture 38

Reduction of Linalool-ff)f-Dimethylallyl Alcohol ••••»• 39

APPENDIX A . 40

APPENDIX B 45

APPENDIX Cl NMR AND IR SPECTRA 53

LIST OF REFERENCES 68 LIST OF ILLUSTRATIONS

Page Figure

I. Preparation of j£»cyclolavandulyl umbelliferonyl ether .. 21

2a. Verbenone (IX) experimental spectrum (100 Mc) 54

2b. Verbenone (IX) simulated spectrum (100 Mc) 54

3a. Myrtenal (X) experimental spectrum (100 Mc) 55

3b. Myrtenal (X) simulated spectrum (100 Mc) 55

4a. 01-Pinene (XI) experimental spectrum (100 Mc) 56

4b. 0(-Pinene (XI) simulated spectrum (100 Mc), system 1 (Insert) 56

4c. 01-Pinene (XI) simulated spectrum (100 Mc), system 2.. 56

5. Myrtenal (X) experimental and simulated spectra (60 Mc; upper negative, middle J^ positive) ..... 57

6. Commercial methylheptenone (XVI) 57

7. 2,6-Dimethyl-5-heptene-2-ol (XVII) . ••• 58

8. 0t+ ^-Cyclogeraniolene (XVIII + XIX) 58

9. Commercial methylheptenone (XVI} neat) 59

10. 2,6-Dimethyl-5-heptene-2-ol (XVII; neat) . .. 59

11.

12. Acetylcyclogeraniolene (XX; neat) 60

13* Acetylcyclogeraniolene (XX) ...... « .. 61

14. ^-Cyclolavandulic acid (XXI) .. . . . 61

15. ^-Cyclolavandulic acid (XXI; KBr) .. . . . 62

16. f-Cyclolavandulol (XV; CC14) 62

vi vil

LIST OF ILLUSTRATIONS—Continued

Page Figure

17. ^Cyclolavandulol (XV) . .. 63

18. f?-Cyclolavandulyl bromide (XXIII) ...... 63

19. Natural archangelln (XXVI; 60 Mc) ••...••••... 64

20. Natural archangelln (XXVI; 100 Mc) ...... 64

21. ^-Cyclolavandulyl umbelliferonyl ether (XXV; 60 Mc) ... 65

22. ^-Cyclolavandulyl umbelliferonyl ether (XXV; 100 Mc) .. 65

23. Natural archangelln (XXVI; CCl^) .. ... 66

24. ^-Cyclolavandulyl umbelliferonyl ether (XXV; CCl^) ... 66

25. Compound B 67

26. Perezone (VIII) 67 LIST OF TABLES

Page Table

1. NMR ABSORPTIONS REPORTED FOR ARCHANGELIN 6

2. CHEMICAL SHIFTS (T) AT 100 Mc 11

3. COUPLING CONSTANTS IN CPS 14

4A. RESULTS OF ITERATION ON VERBENONE (IX, 60 Mc) 40

5A. ERROR VECTORS AND PROBABLE ERRORS 42

6A. TABLE OF ORDERED LINES 43

7B. LIST OF TERPENOIDS Cj-C^ 45

8B. LIST OF TERPENOIDS C15»C19 47

9B. LIST OF TERPENOIDS C. -C. 50 200 40n 10B. TERPENOIDS OVER C^ AND STEROIDS 51

vlii ABSTRACT

NMR spectral parameters were derived, by computer analysis, for three bicyclo^3.1.ljheptane derivatives. As expected, the coupling constants vary little among the three compounds. A 4-bond coupling con stant of +5.9 to +6.5 cps was observed between bridgehead protons, and two 5-bond couplings of 1.8 cps were found in cf-pinene.

The structure of archangelin, a natural furocoumarin, was re vised by NMR analysis and synthesis of the alkyl portion, y?-cyclolavandulol. A new ether, ^-cyclolavandulyl umbelliferonyl ether, was synthesized and its NMR spectrum compared with that of archangelin.

This is the first reported occurrence of the ^-cyclolavandulyl carbon skeleton in nature.

A furocoumarin recently isolated from Hercleum candicans was shown to be a mixture of the known compounds, imperatorin and 8-geranoxypsoralen, by stereospecific Birch reduction and NMR analysis*

The structure of perezone, a natural terpenoid, was revised by inspection of its NMR spectrum.

ix INTRODUCTION

Nuclear Magnetic Resonance Spectra of Terpenoids

In addition to charge and mass, which all nuclei have, some iso

topes possess spin or angular momentum. Since a spinning charge gener

ates a magnetic field, there is associated with this angular momentum

a magnetic moment (fJI). Nuclei which possess spin when placed in a mag

netic field will precess about the direction of the field. An increase

in the field strength causes the nuclei to precess faster.

By applying a second, much weaker, magnetic field at right

angles to the first magnetic field and causing this second field to ro

tate at exactly the precession frequency, the nuclei can be caused to

align themselves with the strong magnetic.field. The frequency at which

the nuclei flip can be observed by a receiver coil.

Interactions of the electrons and nuclei in a molecule modify

.the magnetic environment in which various nuclei are found and this

leads to the observation of chemical shifts and spin coupling. The

chemical shift is a result of the induced orbital motion of the elec

trons when the molecule is placed in an external field and is propor

1 tional to the applied field, Hq. Ramsey' first developed a theory for

the chemical shift.

Nuclear spin-spin coupling, which causes the fine structure, is

independent of the external field and arises from magnetic fields within 2 the molecule itself. Ramsey and Purcell developed the first successful

theory to explain these interactions.

1 2

Resolution sufficient to distinguish chemical shifts of non- equivalent nuclei in the same molecule is referred to as "high resolu tion" nuclear magnetic resonance (NMR) spectra. By observing such spectra in liquids, where the direct magnetic dipole Interaction is averaged to zero by the rapid motion of the molecules, sufficient reso lution may be obtained to observe the fine structure due to nuclear spin-spin interaction.

The chemical shift and spin-spin coupling parameters can com pletely describe high resolution NMR spectra, but in most cases these are related in such a complex manner that a lengthy analysis must be carried out in order to obtain these parameters from the spectrum.

Pople, Schneider, and Bernstein 3 have described procedures for such analyses for certain systems. During the past several years much effort has been directed toward extracting information regarding spin-spin coupling from the fine structure of high resolution NMR spectra, espec ially with regard to coupling between protons.

Theoretical calculations explaining and predicting spin coupling A 5 constants have been developed using molecular orbital ' and valence 6*8 bond methods. Karplus' calculations show especially good agreement with regard to vicinal coupling constants in substituted ethanes and ethylenes.

Nuclear spin coupling decreases rapidly through saturated bonds and coupling through as many as four saturated bonds has been reported 9-13 14 in only a few cases. Barfield has calculated proton spin coupling across four bonds in both saturated and unsaturated systems, but his calculations are based on exchange integrals for unstrained and 3

unsubstituted hydrocarbons and extension to strained or substituted sys

tems is risky.

Complete spectral analysis is a laborious and sometimes impos

sible undertaking, especially when many spins are involved* Computers

have reduced the time and effort necessary to analyze spectra and have

extended the scope of analysis to 7-10 spins. A number of computer pro- 15-19 grams have been written which are helpful in determining chemical

shifts and coupling constants accurately using iterative techniques. 3 20*23 A number of books ' dealing with nuclear magnetic resonance

spectroscopy, which include theory, fundamentals of instrumentation, and

some applications to organic chemistry, have appeared. Other predomin- 24 25 ately theoretical treatments ' have also been published. A monograph

explaining and illustrating the basic knowledge necessary to obtain

structural information from an NMR spectrum has been written with refer- 26 ence mainly to steroid systems.

The chemistry of the terpenoids has been discussed in a number 27-29 30 31 of books. In addition, physical constants ' and infrared spec- 30 tra have been compiled for a large number of mono- and sesquiterpen

oids. The latter (31) contains literature references to sources,

structures, and IR, UV, and NMR spectra of sesquiterpenoids.

Nuclear magnetic resonance spectroscopy has been used to great advantage in determining the structures of natural products. Correla

tions between structure and NMR spectra have been reported for the 32 33 34 lower terpenoids and for triterpenoids. *

Varian Associates and Sadtler Research have compiled NMR

spectra of a wide range of commercially available compounds. Of use to 4

natural product chemists would be a similar undertaking involving as

many terpenoids as possible. Part of the purpose of this work was to

collect and interpret the NMR spectra of as many terpenoids as could be

obtained commercially and from chemists working in this field.

Archangelin

Furocoumarins of the type I have been isolated from a number of 37-43 different plants. Thus, 5-methoxypsoralen (I, Xj»Ht X^OMe),

X 1

5-hydroxypsoralen {I, X^«Hf X^OH), and 5-geranoxypsoralen (I, X^*H, 39 *2™OC10H17^ have been isolated from bergamot oil { 5,8-dimethoxypsora-

len (I, Xj-Xj^QMe) from lime oil ; 5-geranoxypsoralen (1, Xj»H, X^a

OC H 10 17>' 8-geranoxypsoralen (1, XJ-OCJQH^, X2>H) and byakangelicin

(I, X^»OCH2^H^(CH^)2» X2»0Me) from lemon oil^; imperatorin (I, Xj-

OC^H^, X^-H) and isoimperatorin (I, Xj-H, X2»OC^H9t where is 37 38 y,Jf-dimethylallyl) from Imperatoria ostruthium. ' The nature of the 44 aromatic portion of these terpenoids is easily established, but the X 45 part is more difficult to elucidate. A study was made on the stereo

chemistry and synthesis of some of these terpenoids. From the root of Angelica a rchange Ilea JL., an Indian medicinal 46 plant, five crystalline compounds were isolated. Three of these were furocourmarinsjangelicin, prangolarlne, and archan gelin, C21H2204.

Archangel in, mp 132°, crystallized from methanol in thick rods and was shown to have no methoxyl, methylenedioxy, or active hydrogens,

but to contain at least one group. Its furocoumarin nature was shown by its behavior towards 57. aqueous and alcoholic alkali. The ultraviolet 222<1o8C *.38), 251(log£ 4.19), and 310 m|i( loge 4.08)],

infrared ^5.8p(conjugated lactone), 8.9|i(ether), and 9.3|i(benzofuran)J, and NMR ^doublets (J«2.5cps) at 2.38 and 2.75Tand doublets (J»10cps) at

1.85 and 3.7*|] further confirmed its furocoumarin structure.

Hydrogenolysis, pyrolysls, or acid hydrolysis yielded a known phenol, isobergaptol (II), plus a volatile terpenoid compound having a

OH II

-C^q unit. This suggested that archangelin was an isobergaptol ether of a monoterpene alcohol.

The possibility of the fragment being geranyl or its stereo isomer was ruled out from comparison of their NMR spectra. The mass number of archangelin also excluded the possibility of the terpenoid fragment as dehydro or tetrahydrogeranyl but rather suggested a cyclic structure. 46 The authors suggested from the NMR data (Table 1) that the structure of the side chain was III.

0 —

III

TABLE 1

NMR ABSORPTIONS REPORTED FOR ARCHANGELIN

Group Resonance (r) No. of Protons gem-dimethyl 9.08(s) 6 olefinic methyl 8.32(B) 3 coumarin 1.85(d), 3.7(d) 2 furan 2.38(d), 2.75(d) 2 benzene 2.99(a) 1 olefinic 4.35(a) 1 methylenes 7.58-8.9(m) 7 s • singlet; d • doublet; tn • multiplets

Psoralens

Two other furocoumarins were recently Isolated from Heracleum candlcan roots.The furocoumarin nature of the two compounds (A and B) was indicated by three strong absorption maxima in their ultra* violet spectra. Infrared measurements on both compounds showed the 7

absence of free hydroxy 1 groups and the presence of a 6-membered i unsaturated lactone carbonyl. Acid catalyzed degradation of both A and

B resulted in only xanthotoxol (IV) and geraniol (V) OH

Ozonolysis of both compounds gave levulinlc aldehyde and acetone, which were identified as 2,4-dinitrophenylhydrazone derivatives. This sug gested that they were 8-geranoxypsoralen (I, Xj»0-geranyl, X^-H)and

S-neroxypsoralen (I, Xj«0-nerylf X^H). Comparison with an authentic sample showed that compound A was identical with the known 8-geranoxy psoralen, but it remained to be established whether or not B was

8-neroxypsoralen.

Perezone

Isolated over thirty years ago from the roots of Perezia cuer- 47 navucana was a compound, ci5H20°3* c®He<* perezone* It was assigned structure VI on the basis of some poorly understood reactions of itself and its derivatives in warm concentrated sulfuric acid. Pyrolysls of perezone gave two products identified as 4and ^-pipitzol. Romo et al.*® recently reported the structures Vila and Vllb for o( and^-pipitzol,

I 8

VI respectively. A mechanism involving a cyclobutane intermediate and two alkyl shifts was proposed for the perezone-pipitzol conversion.

Vllb

It seemed that a much more straightforward mechanism (see arrows in structure VIII) could be written for this unusual reaction if the

VIII structure of perezone were VIII, Examination of molecular models shows that there should be no serious nonbonded interactions or angle deforma tions in the transition state for the concerted process, although a stepwise mechanism cannot be ruled out* DISCUSSION

NMR Spectral Parameters in Bicvclo^3.1. ljheptanesi

QC-Pinene. Myrtenal, and Verbenone

Among the bicyclo^3.1.lj heptanes are members of the pinene fam

ily of naturally occurring terpenoids. In this thesis the Laocoon II

computer program^ was used to analyze completely the spectra of three

of these compounds, verbenone (IX), myrtenal (X), and o(-pinene (XI),

the latter of which is the most abundant natural cyclobutane.

CH CHO H

CH CH CH CH

IX XI

The NMR spectra of related compounds containing cyclobutane

rings, including bicyclo^2al.ljhexanes,^'^ bicyclo^l.l.ljpentanes,^

and tricyclo^l.l.l.O^'^Jpentanes,have previously been analyzed.

Bicyclo|^2,2. lj heptanes, isomeric with the ^3.l.lj heptanes, have been

studied extensively.^ In some of these compounds, long range couplings £ (through more than 3 bonds) have been observed. Some progress has been 14 made in developing the theory of long-range coupling, but a thorough

understanding awaits the measurement of more such couplings in compounds

of known molecular geometry such as the ones discussed in this thesis.

9 10

The experimental spectra (100 Mc) of IX, X, and XI as shown in

Appendix C, Figures 2at 3a, and 4a were analyzed by computer techniques.

A set of chemical shifts and coupling constants which fits the NMR spec trum of each of these compounds has been found and is given in Tables

2 and 3. The spectra which result from the calculated parameters are shown in Appendix C, Figures 2b, 3b, and 4b. The excellent agreement between the experimental and calculated spectra is evident. These re sults are discussed below. All the chemical shift values in this thesis are given in T units.

Chemical Shifts

The simplest case, that of IX, will be considered first. In the spectrum (Fig. 2a), methyl groups a and b appear as sharp singlets at

9.00 and 8.52, respectively. The assignment of the higher field reson ance to a is based on its position above the IT cloud of the olefinic double bond; the contribution from If electrons to shielding of groups 20 situated above the plane of the double bond is well known. The ole finic methyl absorbs at 8.00 and the olefinic proton at 4.27. The doublet centered at 7.94 (one line obscured by the olefinic methyl ab sorption) is assigned to proton because of the relationship of this proton to the olefinic double bond. The 6-line pattern at 7.22 is due to H^, since, of the remaining three protons, only should be split by three protons with couplings of 5.5 cps or larger. A decision as to which of and gives rise to which of the "triplets" at 7.60 and 7.38 is difficult, and a tenuous assignment of the lower field peaks to was made because it was thought that this proton wocld be 11 deshielded more by the carbonyl than would be by the carbon-carbon double bond*

The assignments of chemical shifts (Table 2) for X (Pig. 3a) and

XI (Fig. 4a) are based primarily on the ones discussed above for IX.

TABLE 2

CHEMICAL SHIFTS (T) AT 100 Mc

IX X XI

CH3-a 9.00 9.26 9.16

CH3-b 8.52 8.67 8.74

CH3-C 8.00 8.35

7.94 8.96 8.84 H1 7.38 7.80 7.92 H2 7.22* 7.51 7.66 H3 7.60 7.12 8.06 H4 7.43 7.79,7.81 VH6 4.27 3.29 4.88 H7 CHO 0.48

The most striking difference is the large upfleld shift (1 T unit) ob served for in going from IX to X and XI, This difference should re- 2 suit from the sp hybridization of the carbonyl carbon in the former 3 compound as compared to the sp hybridization of the corresponding carbon in the other compounds, but it is not clear how much of the extra shielding in X and XI is due simply to the presence of the H^-C bond, how much to the difference in the field around the IT system, and how much to a slight change in conformation which could put closer to the

ITsystem in X and XI, Whatever the reason for this difference is, the methyl group (CH^-a) corresponding to undergoes parallel but smaller shifts in going from IX to X and XI,

Part of the 6-line pattern due to is easily discernable in the spectra of X and XI at 7,51 and 7,66, respectively. The "triplet" pattern at 7.60 in IX due to should change little among the three compounds, except for the chemical shift, and is observed at 7,12 in X and 8,06 in XI, These large differences in the chemical shift of can be rationalized in terms of the anisotropic magnetic fields around the unsaturated systems. H^, unlike H^, should show greater multiplicity in

X and XI than in IX due to the introduction of and H^. clearly absorbs at 7.80 in X and, since it should have about the same chemical shift and splitting pattern in X and XI, apparently gives rise to the partially obscured absorption centered at 7.92 in the spectrum of XI,

The remaining absorption centered at 7,43 in X and 7.80 in XI must be due to and H^« In each compound and should have very nearly the same chemical shift because of their very similar environ ments, and this is observed to be the case.

Coupling Constants

Coupling constants (Table 3) were estimated for IX and then im proved using a least squares refinement program (see Appendix A for computer results). Using the values thus obtained as starting points, the coupling constants for X and XI were determined by trial and error. 13

Hj appears in all cases as a doublet with a separation of 8*5 to

9,1 cps. This separation also occurs in the pattern and therefore represents It is somewhat larger than the corresponding coupling in bicyclo^2.1.ljhexanes.^' Using a plot of gem vs. H-C-H angle for other systems,our finding of • 8.5 to 9.1 cps predicts a bond angle of about 113°• This geminal coupling constant is assigned as 52 53 negative by analogy with related systems. '

The signs of the coupling constants given in Table 3 are based 13 on recent work which has related C-H couplings, which are believed to be positive, to vicinal H-C-C-H couplings. Other experiments in- 57 58 volving high resolution analysis, double resonance, and double quan- 59 turn transition spectra have shown that the sign of the geminal proton-proton NMR coupling constant is opposite from the sign of the vicinal proton-proton coupling constant.

The very small coupling between and the remaining protons on the four membered ring is no doubt due to the unfavorable dihedral 8 angle. The Karplus equation for vicinal cases gives poor agreement when applied to small strained ring systems,and so it would be dangerous to deduce dihedral angles in this case.

From the appearance of the absorptions for H^, H^, and in IX,

and are a11 it is apparent that JJJ, ^4* ^34 About 6 cps. Starting with each of these couplings as 6 cps and iterating gave the values shown in Table 3 for IX. The signs of the vicinal couplings 3^$ atM*

are undoubtedly positive,but the sign of the large 4-bond coupling, ^24' rema*ned t0 found. Barfield^*^ summarized the available data on 4-bond couplings and came to the conclusion that

1 14

TABLE 3

COUPLING CONSTANTS IN CPS

IX X XI

0.32a 0.3 0.3 12 -9.08a -8.9 -8.5 13 0.04a 14 5.53a 5.9 5.7 23 6.50a,d 5.9C,d 5.9d 24 3.0 2.8 25 3.0 2.8 26 1.41a,d 1.4d 1.4d 27 5.904 5.7 5.7 34 o.ua 37 1.32a,d 1.4C»d 1.4d 47 b b 56 3aO 2.8 57 3.0 2.8 67 1.8d c5 d c6 1.8 -1.5 -1.5 c7

a. Results from least squares refinement using Laocoon II, part II. b. Spectrum insensitive to this constant. c. Sign verified by spin tickling. d. Sign based on calculations by M. Barfield and J. Reed.67 large couplings of this type are positive. That is positive in the present case was verified in two ways. First, by calculating and then plotting the 60 Mc spectrum for X , changing only the sign o£ J^» a noticeable change in the spectrum is observed (Fig. 5). This change in the spectrum is observed only at 60 Mc, presumably due to the greater peak distortion at this field strength. For positive, the calcu lated spectrum shows the absorption as a broad band of peaks, virtu ally identical with that found in the observed spectrum. When J^ is negative, however, the absorption of becomes less well resolved. In addition, the appearance of the two peaks at about 7.55, due to H^, fits much better with positive. A second test of the sign of Jwas 58 accomplished by spin-tickling experiments. Irradiation at the low field line of the triplet causes distortion of the low field side of the absorption, indicating and to have the same sign.

Long range coupling has been reported by some to require a near planar arrangement of the protons for detectable couplings, while a re cent article^ has observed that planarity of the system may not be dom inant in determining the size of coupling over four bonds. In this series, the coupling through four saturated bonds (J24) large even though models show that the system is not planar.

Our finding of 5.9 to 6.5 cps for the bridgehead-bridgehead coupling constant (^24) *n these bicyclo^3.l.lj heptanes compares with values of 18 cps for £l.l.lj^ and 1.5 cps for systems.

Surprisingly, a value for the ^2.1.lj system, probably the only other bicyclic system which has a long-range bridgehead-bridgehead coupling constant over 1.5 cps, has not been reported. The 2.1.1 constant is 16 probably between the ^l.l.lj and ^3.1.lj value, and should be much closer to the latter value since the number of 4-bond paths seems to be dominant in determining the magnitude of these couplings.

Jissmaller for Xand XI by 0.6 cps while increases by

0.2 to 0.4 cps. The removal of the carbonyl from the ring system and 2 3 the resulting hybridization change from sp to sp is probably responsi- ble for these coupling changes.

J2j and are observable 4-bond couplings ranging in magnitude from 1.3 to 1*5 cps and were calculated^ to be positive. By spin tick ling, and were shown to have the same sign as J^. In all three 2 compounds J^ and are through at least one sp hybridised carbon, and the atoms are rigidly arranged in the "W" shape, which seems to pro vide maximum coupling.^®

The values of 2.8 to 3.0 cps for the vicinal couplings

J^, and in X and XI conform with expectation for dihedral angles of

60°.8

The aldehydic proton of X is a sharp singlet at 0.48. This is 9 consistent with results reported by others for aldehydes in which the carbonyl group is able to become coplanar with an CKv/?-double bond. The olefinic methyl appears as a doublet (J * -1.5 cps) in IX and a 1x3s3s1 quartet (J " +1.5 cps) in XI. accounts for the splitting in IX, but two additional couplings to the vinyl methyl in XI are required. These additional couplings are almost undoubtedly 5-bond couplings with Hj and

This coupling is confirmed by the close agreement between the H^,

H^, and vinyl methyl absorbances in the observed (Fig. 4a) and 17 calculated spectra (Fig. 4c; for this comparison, the seven spin system, vinyl methyl, H^, H^, H^, and was used).

Some of the coupling constants found above by calculations were confirmed by decoupling experiments. Thus, decoupling of X gave the fol

lowing results: Irradiation at causes to become a triplet (J ™

5.8 cps), H2 to become 7 peaks with average distance 3.0 cps, and a doublet (J - 5*8 cps). Irradiation at in XI collapses the vinyl methyl absorption to a doublet (J * 1.5 cps). 69 Earlier investigators have assigned gem coupling constants on 3 sp carbon from -12.0 to -20.0 cps. Our results show that in the case of X, J56 can be varied over a wide range (0 to -16 cps) with no appre ciable effect on the calculated spectrum.

Terpenoid NMR Spectral Compilation

The NMR spectra of terpenoids which were obtained will not be reproduced here nor will a detailed analysis of individual spectra be attempted. Rather, the complete spectra, together with the assignments and indexing done during this research, are to be published by Varian

Associates, Palo Alto, California. An alphabetical list of the com pounds which will appear in the compilation is given in Appendix B,

Tables 7B-10B. It is hoped that these spectra will be of use, not only to natural product chemists, but also to organic chemists in general, because of the diverse number of structural types included.

Archangelin 46 Natural archangelin as obtained from Chatterjee was reinvesti gated by NMR with careful integration. The NMR spectra (60 and 100 Mc, Figs. 19 and 20; see Appendix C for NMR and IR spectra) show a 6 proton singlet at 9.08, a 2 proton triplet (J • 6 cps) centered at 8.56, a broadened singlet for 3 protons at 8.30, a broadened singlet for 2 pro tons at 8.16, a broad multiplet for 2 protons at 7.75, a sharp singlet for 2 protons at 5.01, doublets (J " 10 cps) at 3.65 and 1.92 (each 1 proton), doublets (J - 2.5 cps) at 2.94 and 2.32 (each 1 proton), and a sharp singlet for 1 proton at 2.80. The absorptions in the region 2.2*

3.7 are in agreement with the isobergaptol (II) portion of archangelin which was previously found.

Analysis of the 5-10T portion of this spectrum leads to a dif ferent conclusion than that reported. The absorption at 5.01 Integrates 46 for 2 protons and is a singlet. The structure reported, III, for the terpenoid side chain would demand at least a broad peak and more prob ably a triplet for this absorption. The resonance at 7.75 is broad, but a triplet and a separation of 6 cps is suggested by careful inspection.

The same separation is observed in the triplet at 8.56 and these two facts suggest the partial structure -CH^CH^- . The absorption at 8.30 is in the well known region for methyl groups attached to double bonds.

The absorption at 8.16 is an isolated methylene while the sharp singlet at 9.08 is a gem dimethyl group. Putting together these observations leads to -the partial structures Xlla and Xllb which are both consistent with the NMR spectrum of this natural compound.

A tentative choice between Xlla and Xllb was made on biogenetic grounds. It is not possible to derive the carbon skeleton which these structures contain by cyclization of geraniol without carbonium ion 19

0—

Xlla Xllb

rearrangements, but this skeleton can be put together from two y,y- dimethylallyl units via lavandulol XIV as shown below.

XIII XIV XV

This biogenetic route clearly favors structure Xlla over Xllb for the

terpenoid side chain in archangelin, and we proceeded under the tenta

tive assumption that Xlla was correct. Although this carbon skeleton

has apparently not been found previously in nature, the alcohol XVt

termed ^-cyclolavandulol, had been prepared in the laboratory several 70-72 73 times, once by the acid-catalyzed cyclization of lavandulol.

In order to test this proposal, the synthesis of ^-cyclolavan*

dulol (XV) was undertaken. Since the methyl ether of isobergaptol was 20 commercially available in limited quantities, a complete synthesis of archangelin could also be attempted, using a procedure analogous to that used to prepare umbelliprenin, the farnesyl ether of umbelliferone.^

^-Cyclolavandulol (XV) was prepared according to Fig, 1. Methyl heptenone (XVI) (Fritzsche Brothers) was reacted with methylmagnesium iodide to give 2,6-dimethyl-5-hepten-2-ol (XVII). The alcohol showed NMR

(Fig, 7) singlets at 8,76(6 protons), 8,35(3 protons), 8,29(3 protons), and 5.90(1 proton). A broad multiplet centered at 7.95(4 protons) and a triplet at 4.80(1 proton) were all consistent for the expected struc ture. The IR (Fig. 10) showed the free -OH at 2.9||,

Dehydrative cyclization of XVII was accomplished by heating with oxalic acid. The optimum yield (57%) of a mixture of cyclogeraniolene isomers (0L+XVIII and XIX) was obtained when the molar ratio of al cohol to acid was 111. The remaining portion of the reaction was an acyclic dehydration product, 2,6-dimethyl-2,5-heptadiene. The cyclogeraniolenes were separated from the acyclic product by spinning band distillation. The NMR spectra of the geranlolene mixture (Fig. 8) shows absorptions for the gem dimethyl at 9.03(9.00 for ft isomer), methylene absorptions at 7.8-8.8, olefinic methyl at 8.32 and the olefinic proton at 4.58(4.75 for isomer). The approximate ratio of Of to ft isomers is

3:2 based on the integration of the olefinic protons in the NMR. The IR

(Fig, 11) spectrum (identical with that reported) shows a weak OC stretch at 6,0p and a strong bending mode at 12,45||, confirming the cyclohexene nature of the hydrocarbon. 21

CH3MgI W H2C2O4

•> n!' XVI XVII

COCH.

Br2,NaOH ^ CH3C0C1 ^ SnCl.

V XX XVIII XIX

COOH CH2OH

2) LiAlH,

PBr XXI XV

CH2Br

DMF

XXIII XXV

Na.EtOH

"*Na "0

XXII XXIV

Figure I. Preparation of |9-cyclolavandulyl wnbelllferonyl ether* 22

Ho attempt was made to separate the isomers because the OCisomer

should react preferentially in the succeeding reactions.

Acylation of the cyclogeraniolene mixture was accomplished under

Friedel-Crafts conditions* Treatment of XVIII with acetyl chloride in

the presence of stannic chloride yielded acetylcyclogeraniolene (XX), in moderate yield (447.). The NMR (Fig. 13) shows a gem dimethyl singlet at

9.10, a methylene triplet at 8.60, a broadened singlet for the olefinic methyl at 8.2, the keto-methyl singlet at 7,90, and a broadened methyl* ene at 7.73. The remaining methylene is obscured by the other absorp

tions. The IR (Fig, 12) shows conjugated 00 stretch at 5.95^ and OC stretch at 6.12|1. After recrystallization, the semicarbazone of XX, 71 prepared in the usual way, had mp 196-8° (lit. 201-3°).

p-Cyclolavandulic acid (XXI) was prepared from the ketone by us ing the bromine modification of the iodoform reaction. Treatment of XX with bromine in sodium hydroxide yielded XXI, mp 111.5-113° (lit#^

110-111°) in 38% yield as white needle-like crystals. The NMR (Fig. 14) has a singlet for gem dimethyl at 9.05, a methylene triplet (J » 6 cps) centered at 8.58, one allylic methylene at 8.00, the olefinic methyl at

7.85, a broad allylic methylene absorption at 7.58, and the acid proton at -1.33. The olefinic methyl is shifted downfield by 0.5 ppm from that usually observed., This is presumably because of its close proximity to the carboxyl group. The IR (Fig. 15) shows the broad -OH at 3-3.6)1, the 00 at 5.95p, C-C at 6.2)1, and C-0 at 7.8ft.

Reduction of the acid to the alcohol with lithium aluminum hy dride was found to proceed in poor yield unless the methyl ether was first prepared. Methyl p-cyclolavandulate (from the acid plus 23 diazomethane) was reduced with lithium aluminum hydride in ether to

^•cyclolavandulol (XV) in 94% yield. The alcohol, bp^ 105-106*, gave an NMR spectrum (Fig* 17) showing gem dimethyl at 9.16, a triplet (J -

6 cps) at 8.72 for 2 protons, a broadened absorption at 8.35 with a shoulder at 8*3 (total number of protons, 5), a broad multiplet at 7.98 for 2 protons, a -OH peak at 6.10 (shift changes on change in concentra tion), and a 2 proton singlet at 6.05. The IR (Fig. 16), identical with that reported, shows -OH stretch at 2.95)1, a very weak OC at 6.0|i, and the C-0 at 8.25p.

Comparison of the NMR spectrum of /?-cyclolavandulol (XV) with the portion of the NMR spectrum of archangelln above 5.Or (see Figs. 17 and 19) shows the very close similarity both in chemical shifts and gen eral appearance. The methylene attached to oxygen is at 5.01 in arch angel in and 6.05 in the alcohol. This difference would be expected because of the deshielding effect of the aromatic system in archangelln.

The deshielding power of the aromatic ring decreases as the alkyl por tion becomes farther removed so the other alkyl resonances are not as deshielded. The broad multiplet at 7.75 in archangelln is reproduced at 7.88 in the alcohol. The olefinic methyl at 8.30 in archangelin is at 8.35 in ^-cyclolavandulol. The methylene absorption at 8.16 in arch angelin corresponds to the shoulder seen in the alcohol at 8.3. The triplet (J • 6 cps) at 8.56 in archangelin is found at 8.72 in the alco hol. The gem dimethyl appears as a sharp singlet at 9.08 in archangelin and 9.16 in the alcohol.

The aromatic portion of archangelin, isobergaptol (II), was available commercially as its methyl ether. Methyl ethers have been 24

cleaved by a number of reagents, but because of the variety of function

al groups present in isobergapten (II, -OH - (Me), it was necessary to

use a mild, selective reagent. The cleavage of methyl ethers In steroid 74 systems was accomplished by Johnson et al. using pyridine hydrochlor

ide. The procedure failed, however, with isobergapten) the IR showed

that the product was not isobergaptol.

Because of the limited supply of isobergapten it was decided to

prepare an ether between cyclolavandulol and a more readily available

hydroxycoumarln and compare the NMR spectrum of the ether with that of

archangelin. ttnbelliferone (XXII) was available and its sodium salt was

prepared by treatment with sodium metal in ethanol. Reaction of the

sodium salt with cyclolavandulyl bromide (XXIII) (from the alcohol and

phosphorous tribromide) yielded a white crystalline solid of mp 92-4°.

The microanalysis was correct for C^H^O^ an<* comparison of its NMR

spectrum, at 100 Mc (Fig. 22), with that of archangelin (Fig, 20) re

veals the close similarity. The ether (XXV) shows a gem dimethyl at

9.06 (9.08 in archangelin), an olefinic methyl at 8.25 (8.30 in arch

angelin), a 2 proton triplet (J • 6 cps) at 8.60 (impurity at 8.68)

(8.56 in archangelin), a broadened singlet for 2 protons centered at

8.18 (8.16 in archangelin), a broad multiplet at 7.88 for 2 protons

(7.75 in archangelin), and a singlet for 2 protons at 5.45 (5.01 in archangelin). The aromatic portion of XXV shows a pair of doublets

(J « 10 cps) at 3.75 and 2.35 (coumarin protons), a singlet at 3.18, a doublet (J - 10 cps) at 3.13 (6ne peak partially obscured by the sing

let at 3.18), and a doublet (J «* 10 cps) centered at 2.65. The latter 25

absorptions correspond to the three aromatic protons. The IR (Fig. 24)

shows the Q(fp-unsaturated lactone carbonyl at 5.75y. This data all fits

the proposed structure XXV for the ^-cyclolavandulyl umbel1iferonyl

ether.

Acid catalyzed degradation of archangelin yielded a terpene ace-

tate in very small yield, but it corresponded to a known sample of p-cyclolavandulyl acetate prepared from the alcohol. The known acetate

had an of 0.55 and the acetate from archangelin had an R^ of 0.56, on

thin layer chromatography with silica gel. The infrared spectra were

nearly identical.

Even though archangelin itself was not synthesized, the evidence

presented points to this furocoumarin as the^cyclolavandulyl ether of

isobergaptol (II + Xlla • XXVI), This being the case, it is the first

reported natural occurrence of this cyclic monoterpene skeleton*

XXVI

Psoralens

A method developed for investigating the side chain in mycelia- 45 amide was applied to help resolve the structure of compound B (see In

troduction, page 6). Birch reduction of compound B and VPC analysis of 26 the product showed peaks attributed to 2-methyl-2-butene (XXVII) and trans-2.6-dlmethvl-2.6-octadiene (XXVIII; methylgeraniolene). Birch

XXVII XXVIII XXIX reduction of a linalool-¥,)f-dimethylallyl alcohol mixture, under the same conditions as those used with the unknown, gave peaks with reten tion times identical with known XXVII (2 minutes), cis-2,6-dimethyl-2.6- octadiene (methylnerolene, XXIX, 38 minutes), and XXVIII (40 minutes).

It is noteworthy that no XXIX was observed in the reaction product from B.

The presence of both and C^Q side chain fragments in the psoralen mixture and their relative amounts was confirmed by analysis of the NMk spectrum. The spectrum (Fig. 25) shows a doublet at 4.82 due

to -OCHjOC- which occurs in both and C^Q compounds while the absorp tion at 7.91 comes from -OCCH^Cl^OC-, which occurs only in the compound. The integrated intensities show a 2 to 1 ratio in favor of the -OCH^OC- protons. Thus the mixture must consist of 3 times as much imperatorin (I, X^ • OC^Hg, X^ • H) as 8-geranoxypsoralen (I, X^ -

OC10H17- *2' H) - 27

Perezone

A high resolution NMR spectrum of perezone (Fig, 26) clearly shows coupling (J « + 1.7 cps) between the quinone methyl (doublet at

7.90) and quinone hydrogen (quartet at 3,38). Previous NMR studies of 75 76 quinones have shown ' that a coupling constant of this magnitude is

observed only when a methyl and proton are attached to the same quinone double bond, thus excluding structure VI for perezone, which must be

VIII. This structure fits much better the pyrolysis to pipitzols, which can proceed in a concerted fashion as envisioned previously (see Intro duction, page 8). The validity of structure VIII for perezone has been confirmed by others on the basis of further degradative and synthetic work.77 EXPERIMENTAL

General Methods

The experimental NMR spectra were run on 10% solutions in DCCl^ with Varian A-60 and HA-100 instruments using tetramethylsilane (TMS) as an internal standard. Chemical shifts are reported in T units*

IR spectra were obtained with a Perkin-Elmer Model 137 spectrom eter, either neat, on carbon tetrachloride solutions, or with KBr discs.

Melting points were taken using a Thomas-Kofler micro hot stage instrument, model 6886-A, and are uncorrected.

VPC analyses were obtained with F and M instruments, models 609 and 770, The columns were constructed as follows and will be referred to by code number: VPT-002, 0.25" o.d. x 5', packed with 20% Carbowax

20M on 30/60 mesh firebrick} VPT-004, 0.5" o.d. x 8*, packed with 20%

Carbowax 20M on 60/80 mesh acid washed Chromosorb W.

Spinning band distillations were carried out with a Nester-

Faust, 6ram x 18", vacuum jacketed column.

NMR Spectral Parameters in Blcvclo|^3.1 *lj heptanes

The theoretical spectra were calculated using the Laocoon II program of Castellano and Bothner-By, modified for use on an IBM 7072 computer equipped with an XY plotter. The calculated parameters for compound IX were obtained by least squares refinement using part II of

Laocoon II. The results of the iteration are listed in Appendix A,

Tables 4A, 5Ay and 6A. Table 4A gives the input parameters, the

28 29

parameter sets which were varied independently (the computer could handle only 10 sets at a time so various combinations of 10 were used to cover all 15 of the parameters; the set given in this table is a repre sentative one), the root mean square error after each iteration, and the

best values obtained for each of the parameters* Table 5A gives the error vectors and the probable errors (see Castellano and Bothner-By^ for a full explanation of how these are calculated and their meaning).

Table 6A tabulates the observed spectral lines, the calculated lines, and the error in fitting,

oc-Pinene (XI) was used as obtained from the Aldrich Chemical

Co., myrtenal (X) was obtained from the Glidden Co., and verbenone (IX) 78 was prepared according to the procedure of Dupont et al.

Terpenoid NMR Spectral Compilation

The experimental spectra were determined as 10% solutions in

DCC13, primarily with a Varian A-60 spectrophotometer. Those compounds which'were not available in sufficient quantity for an A-60 spectrum were run on a Varian HA-100 instrument as solutions of less than 10X concentration. Dimethyl sulfoxide-dg was used as solvent for those com

pounds insoluble in DCCl^. The spectra were run at room temperature with sweep times of 250 sec. (60 Mc) and 500 sec. (100 Mc).

Synthesis of Archangelin

2.6-Dimethvl-5-heptene-2-ol (XVII) 79 The procedure of Callen, Dorafeld, and Coleman was followed for the preparation of methylmagneslum iodide. Into a 1 liter round- bottom flask fitted with a nitrogen inlet, stirrer, water condenser, and 30 dropping funnel, magnesium turnings (36 g., 1.5 mole) were added and covered with 100 ml. anhydrous ether. While stirring the mixture, methyl iodide (198.8 g., 1.4 mole) in 100 ml. anhydrous ether was added dropwise (addition time 2 hours). Anhydrous ether (400 ml.) was added, .

in 100 ml. portions, to the reaction flask during the 2-hour addition

period.

* Methylheptenone (XVI; Fritzsche Brothers; 63 g«, 0.5 mole) in

100 ml. ether was added dropwise to the Grignard reagent over a 2-hour .

period so that the ether was continually refluxing. After the addition

was complete, the mixture was heated at reflux over a steam bath for 1

hour. Hydrolysis of the reaction was accomplished by pouring it slowly

into 2 liters of ice water. The hydrolyzed mixture was transferred to a

separatory funnel and allowed to stand. The water (the bottom layer)

was periodically withdrawn and then the white foamy ether layer was dis

tilled at atmospheric pressure to remove the ether and finally, with re

duced pressure (30 mm), the desired alcohol was steam distilled. The

water drawn off in the separatory funnel was used to replenish the water

supply during steam distillation. The alcohol separated from the water

in the receiving flask and was easily collected. Distillation gave 50

80 g. (711)of alcohol, bp8 71-3°. Infrared (Fig. 10) and NMR (Fig. 7)

spectra were consistent with the desired product (XVII).

Ot-Cyclogeranlolene (XVIII)8*

2,6-Dimethyl-5-heptene-2-ol (XVII; 10 g., 0.07 mole) and oxalic acid (6.5 g., 0.07 mole) were introduced into a 100 ml« round-bottom flask equipped with a nitrogen inlet and water condenser. The mixture was heated at 130-140° for 5 hours, cooled, and extracted three times with ether (100 ml.). The combined extracts were distilled first at atmospheric pressure to remove the solvent and then under reduced pres sure (37 nni) with a spinning band column* Using a reflux ratio of 30tl, three fractions were collected! Fraction 1 (5.0 g., bp 51-3*), fraction

2 (1.0 g., bp 54-8°), and fraction 3 (1.0 g".', bp 58-62#). VPC (130#, column VPT-004), NMR, and IR analysis showed that fraction 1 was cyclogeraniolene (mixture of isomers), fraction 2 contained cyclogeraniolene and 2,6-dimethyl-2,5-heptadiene (1:1), and fraction 3 consisted of cyclogeraniolene and 2,6-dimethyl-2,5-heptadiene (1:6).

The cyclogeraniolene isomers were identified as XVIII and XIX by NMR analysis (Fig. 8) and comparison of IR spectra with those re- 81 ported. No attempt was made to separate the isomers. The yield of cyclogeraniolenes was 57%.

Acetylcyclogeraniolene (XX)^*

Into a 50 ml. round-bottom flask, equipped with a nitrogen in let, dropping funnel, water condenser, and magnetic stirrer, acetyl chloride (3.9 g., 0.05 mole) and stannic chloride (0.3 g., 0.001 mole) were introduced. After cooling to 0°, 5.0 g. (0.04 mole) of the cyclo geraniolene mixture described above was added slowly. The solution darkened during the addition. Stirring was continued for 1 hour at 0°.

Workup of the reaction was accomplished by pouring it into 50 ml. of 10% hydrochloric acid and extracting three times with ether. The combined extracts were evaporated and the organic residue treated with saturated sodium bicarbonate and extracted with ether. The ether layer was then washed with water, dried over sodium sulfate, and evaporated. Distilla

tion of the oil through the spinning band column gave a fraction (2.8 g.

44%) bpj 53°) which was acetylcyclogeraniolene. NMR (Fig. 13) and 1R

(Fig. 12) spectra were consistent with the expected structure. The semicarbazone, prepared in the usual way and crystallized once from methanol-water and once from methanol-benzene, had mp 196-8° (lit.^

201-3°).

82 Regeneration of XX From Its Semicarbazone

The semicarbazone (571 mg., 2.56 mnole) was dissolved in 100 ml. of acetone containing 3 ml. of concentrated hydrochloric acid and refluxed for 1 hour. The yellow solution was cooled and treated with stannous chloride (2.5 g», 0.013 mole) dissolved in 10 ml. of concen

trated hydrochloric acid. Water (15 ml.) was added and the mixture refluxed in a nitrogen atmosphere for 1 hour. The acetone was stripped off with a rotary evaporator at room temperature, the residue extracted

three times with 20 ml. of benzene, and the organic layer washed with

IN hydrochloric acid until it was clear. The benzene solution was washed with a saturated sodium bicarbonate solution and water and was

then evaporated with a rotary evaporator. The crude material was dis solved in ether, dried over sodium sulfate, and the ether evaporated to give 207 mg. (521) of the ketone (XX).

|f?-Cyclolavandulic Acid (XXI

In a round-bottom flask equipped with dropping funnel, cone- driven stirrer, and water condenser was placed a solution of sodium hydroxide (7.0 g., 0.18 mole) in water (25 ml.). After cooling to 0°, bromine (3 ml., 0,06 mole) was added. Acetylcyclogeraniolene (XX; 1.7

g., 0.01 mole) in dioxane (16 ml.) was added dropwise. The ice bath was

removed and stirring continued for 4 more hours. The reaction mixture was then heated at 60° for % hour, cooled, and the excess sodium hypo-

bromite decomposed with saturated sodium bicarbonate. After acidifica

tion with IN hydrochloric acid and extraction three times with ether,

the combined extracts were shaken three times with 0.1N sodium hydrox

ide. The alkaline solution was then acidified and extracted with ether.

Evaporation of the ether solution gave f?-cyclolavandulic acid (XXI), which, after one recrystallization from methanol-water, yielded 1 g.

(58%) of white needle-like crystals, mp 111.5-113° (lit.^ 110-111°).

Methyl ft-Cyclolavandulate 83 Following the Organic Synthesis procedure, to 50 ml. of ether

in a 250 ml. Erlenmeyer flask, 15 ml. of 40% potassium hydroxide was added. While magnetically stirring the solution, it was cooled in an

ice bath and 2.5 g. of nitrosomethylurea was added in small portions.

The ether layer became yellow indicating the presence of diazomethane.

The ethereal solution of diazomethane was added in small quanti

ties to ^-cyclolavandulic acid (XXI{ 500 mg., 2.8 mmole) dissolved in

25 ml* of ether. When the yellow color persisted for several minutes

the addition was stopped. The excess diazomethane was decomposed with acetic acid, the ether layer was washed with a saturated sodium bicar

bonate solution and water, and was dried over sodium sulfate and then

evapbrated to give 490 mg. (97%) of sweet smelling oil identified as the

ester from its IR and NMR spectra. 34 fl-Cvclolavandulol (XV)

Anhydrous ether (20 ml.) and lithium aluminum hydride (140 mg.,

3,7 mmole) were added to a 100 ml. flask fitted with a paddle stirrer, condenser, and dropping funnel. The ester described above (490 mg.t 2.7 mmole), dissolved in ether (20 ml.), was added dropwise over a % hour period. Stirring was continued for 2 hours, then 2 ml. water, 2 ml. of

15% sodium hydroxide, and 10 ml. of water were added in that order. The ether was decanted, the precipitate washed twice with fresh ether, and the combined extracts dried over sodium sulfate. Careful evaporation of the ether yielded the alcohol (XV; 390 mg«, 94%), identified by IR and

NMR.

5-Cyclolavandulyl Acetate^

Into a 50 ml. round-bottom flask equipped with a dry-ice conden ser, nitrogen inlet, dropping funnel, and magnetic stirrer, ^-cyclolavandulol (XV; 154 mg., 1.0 mmole) and pyridine (110 mg., 1.4 mmole) in 10 ml. of ether were introduced. While cooling with a dry-ice ace tone bath, acetyl chloride (118 mg., 1.5 naaole) in 10 ml. of ether was added dropwise (addition time hour). After the addition was complete, the dry-ice bath was removed and the reaction mixture allowed to come to room temperature. Stirring was continued for 4 hours, after which 10 ml. of water was added and all of the solid matter dissolved. The ether and water layers were separated and the latter washed once with 10 ml. of ether. The combined extracts were washed twice with 10% sulfuric acid, twice with saturated sodium bicarbonate, dried over sodium sulfate, and evaporated to give 140 rag. (71%) of acetate* The IR of the acetate was 72 identical with that reported.

0-Cyclolavandulvl Bromide

Following the general procedure of Schauble,^ a 50 ml. round- bottom flask fitted with a dropping funnel, nitrogen inlet, magnetic stirrer, and dry-ice condenser was placed in a dry-ice acetone bath.

Cyclolavandulol (XV; 154 mg.v 1 mnole) in 5 ml. ether was added to the flask. Phosphorous tribromide (2.71 g., 10 mmole, 1 ml.) in 10 ml. of ether was added dropwise over a ^ hour period. After the addition was complete, the dry-ice bath was removed and the reaction flask stirred at room temperature for 5 hours. The reaction was diluted with ether and the excess phosphorous tribromide and phosphoric acid decomposed by slowly adding 50 ml. of 2% potassium hydroxide. The ether layer was separated, washed twice with 50 ml. portions of water, and dried over sodium sulfate. Evaporation under nitrogen gave 168.mg. (781) of the bromide (identified by the loss of -OH absorption in the IR).

Sodium Salt of Umbelllferone (XXIV)^

Crude umbelliferone (XXII; K and K Laboratories, Inc., mp 210°, lit.^ 225-8°) was purified by sublimation at 190° (1 mm). The light yellow product recovered after a single sublimation melted at 225-233*.

Sodium metal (49 mg., 2.1 tnmole) was dissolved in 10 ml. of anhydrous ethanol in a 50 ml. round-bottom flask fitted with a water condenser. Purified umbelliferone (XXII; 162 mg., 1.0 mnole) was added to the reaction flask and the mixture refluxed until all of the 36 umbelliferone dissolved (less than 1 hour). Ether (20 ml,) was added and a yellow precipitate immediately formed. The precipitate was col lected by suction filtration through a fritted glass funnel using a rubber dam to prevent undue exposure to the air. The sodium salt (XXIV}

150 Dig., 82%) was yellow and when dissolved in deuterium oxide gave an 45 NMR spectrum identical with that reported.

A second preparation of the sodium salt (XXIV) was made and im mediately dissolved in 10 ml. of DMF for use in the following reaction

(a bluish-yellow solution was observed when the salt was dissolved in

DMF). ft-Cvclolavandulyl Umbelliferonyl Ether (XXV)

^-Cyclolavandulyl bromide (XXIII; 100 nig., 0.45 mmole) in 10 ml. of DMF was introduced into a 50 ml. round-bottom flask fitted with a dropping funnel, water condenser and nitrogen inlet, and the solution cooled to -78®. The sodium salt (XXIV) in DMF was added and the solu tion allowed to warm to room temperature. While still under the nitro gen atmosphere, the mixture was refluxed at 90° for 12 hours. After cooling, ether was added, followed by ice cold 2% potassium hydroxide, and the ether layer immediately collected. The aqueous layer was washed four times with fresh ether and the combined extracts washed once with ice cold 7X potassium hydroxide. The ether was dried over anhydrous sodium sulfate and evaporated under nitrogen and then vacuum to give 69 eng. (51%) of an oily residue. Recrystallization from methanol-bensene was unsuccessful. Sublimation of the oily residue at 100* (0.4 mu) yielded 40 mg. of white crystals, mp 92-4°• 37

C H I C 76,48 H 7 43 Anal. Calcd. for 19 22°3 » f » * » Found! C,

76.88; H( 7.64.

The NMR (Fig. 21) and 1R (Fig. 24) spectra revealed that the de

sired ether (XXV) was formed.

84 Acetic Acid Cleavage " -

The ether (XXV) described above (40 mg«, 0.13 nmole) was heated with glacial acetic acid (1 ml.) at 115° for l*j hours. After standing

overnight, the mixture was diluted with water, extracted three times with hexane, and the extracts washed with saturated sodium bicarbonate and water. The hexane was dried and evaporated to give a product (trace amount) which had an R^ value (0.56) identical with a known sample of

cyclolavandulyl acetate. |jhin layer chromatography was carried out on silica gel plates (0.25 mm), the elutant was 60% light petroleum ether-

407. chloroform (v/v), and the developer was 1% potassium permanganate .J

The phenol (umbelliferone, XXII) was recovered by acidification of the aqueous layer, extraction with ether, and evaporation. The product (5 mg.) had mp 215°.

Cleavage of Archangelin

Archangelin (25 mg., 0.07 nmole) and glacial acetic acid (1 ml.) were heated in a 10 ml. pear-shaped flask fitted with a water condenser and calcium chloride drying tube for 6 hours at 110°. The solution was

then allowed to cool and stand overnight. After diluting with water (5 ml.), the solution was extracted three times with hexane. The combined

extracts were washed with a sodium bicarbonate solution and water and

then dried over sodium sulfate. Slow evaporation of the solvent yielded 38

6.7 mg. (46Z) of a residue whose IR spectrum resembled that of authentic cyclolavandulyl acetate and whose R^ value (0.55) on TLC (same condi tions as previously) was nearly the same as the authentic acetate.

The aqueous solutions were combined and extracted with ether.

The ether was dried and evaporated to give 5 mg. of recovered archange

lin, mp 115°. The resulting aqueous solution was acidified, extracted

twice with ether, the ether dried over sodium sulfate, and evaporated to give a residue (5.2 mg.) whose IR was not identifical with that pub- 85 lished for the desired phenol (isobergaptol, 11), but was different from that of archangelin.

On the assumption that this was isobergaptol, a regeneration of archangelin was attempted. The bromide (XXIII) was prepared from

0-cyclolavandulol (XV) as described above and the sodium salt of the

residue (5.2 mg.) was likewise prepared as previously described. These were combined under the conditions used for preparing ^-cyclolavandulyl umbelliferonyl ether and worked up in the same way. An oily residue

(6 mg.) was obtained which failed to crystallize before or after evapor ative distillation. The IR spectrum of the oil did not closely resemble that of archangelin.

Reduction of Psoralen Mixture 86 After the procedure of Greenlee and Wiley, ammonia (75 ml.,

3.5 mole) was condensed in a 500 ml. round-bottom flask equipped with a dry-ice condenser, cone-driven stirrer, dropping funnel, and nitrogen atmosphere. Sodium metal (3 g., 0.13 mole) was added in small pieces,

producing a blue solution. To the sodium-ammonia solution, with vigorous stirring, 500 mg. of the psoralen mixture (compound B), in 20 ml. methanol (large excess of methanol due to the low solubility of the mixture) was added dropwise over hour. The blue color of sodium- anmonia disappeared during the addition so more sodium was added in small pieces to maintain the blue color. Stirring was continued for 1 hour, then heptane (25 ml.), granulated ammonium chloride (10 g.)v and water (50 ml.) were added to the refluxing reaction mixture. The hep tane layer was collected. The aqueous layer was washed three times with

25 ml. portions of heptane and the combined extracts were washed with water until the washings were neutral. The heptane was dried over an hydrous magnesium sulfate and subjected to VPC analysis.

Using column VPT-002 on an F and M 609 flame ionization chroma- tograph at 80°, the heptane extract showed peaks attributed to 2-methyl-

2-butene (XXVII) and trans-2.6-dimethyl-2.6-octadiene (XXVIII) by comparing retention times with known standards. The former had a re tention time of 2 minutes while the latter had a 40-minute retention time under the same conditions.

Reduction of Llnalool-y.y-Dimethvlallyl Alcohol

Linalool (36 mg., 0.23 mmole) and ]T,^-dimethylallyl alcohol (72 mg.f 0.84 mmole) in 5 ml. methanol were added dropwise to a solution of sodium (46 mg., 2 nmole) in 25 ml. liquid anmonia according to the above procedure. After stirring for 1 hour at -78°, a red color appeared at the top of the solution. Workup as described above and VPC analysis under Identical conditions gave peaks for XXVII (retention time 2 min utes), cls-2.6-dlmethyl-2.6-octadiene (XXIX, retention time 38 minutes), and XXVIII (retention time 40 minutes). APPENDIX A

TABLE 4A

RESULTS OF ITERATION ON VERBENONE (IX, 60 Mc).

NN » 5 Preq. Range 0.00 500.0 Min. Intensity 0.050

Input Parameters

W(I) - 124.5 A(l,2) - 0.0 A( 2,4) - 5.9

W(2) - 159.0 A(l,3) - -8.7 A(2,5) - 1.5

W(3) - 169.0 A(l,4) - 0.0 A(3,4) - 5.9

W(4) - 145.5 A(l,5) - 0.0 A(3#5) - 0.0

W(5) - 347.0 A(2,3) - 5.9 A(4,5) " i'5

Parameter Sets

1 A(l,2) 5 A(2,3) 8 A(3f4)

2 A(Ir3) 6 A(2,4) 9 A(3,5)

3 A(I,4) 7 A(2,5) 10 A(4,5)

4 A(I,5)

Iteration 0 RMS Error m 1.861

Iteration 1 RMS Error - 0.584

Iteration 2 RMS Error - 0.278

Iteration 3 RMS Error - 0.273

Iteration 4 RMS Error • 0.273

40 TABLE 4A—Continued

Best Values

W(l) - 124.500 A(l,2) - 0.484 A(2,4) - 6.487

W(2) - 159.000 A(l,3) - -9.281 A(2,5) - 1.570

W(3) - 169.000 A(l,4) - 0.024 A(3,4) - 5.572

W<4) - 145.500 A(l,5) - 0.029 A(3,5) - -0.081

W(5) - 347.000 A(2,3) - 5.899 A(4,5) - 1.267 TABLE 5A

ERROR VECTORS AND PROBABLE

0.7557 0.3443 0.3049 0.1121 -0.1527 0.3571 -0.1547 -0.1568 -0.0523 0.0527 Probable Error • 0.136

-0.5387 0.4385 0.2020 0.0025 -0.3136 0.5436 0.1536 0.2095 -0.1084 0.0602 Probable Error • 0.111

-0.1928 -0.5215 0.7037 0.1950 -0.2044 -0.0057 0.0837 -0.3130 -0.0477 -0.0927 Probable Error - 0.098

0.0448 -0.1474 -0.3538 0.8299 -0.1894 0.0781 0.2921 -0.0434 -0.1074 0.1472 Probable Error • 0.095

-0.1591 0.4928 0.3224 0.2567 0.4493 -0.3598 0.1981 -0.2878 0.1680 0.2834 Probable Error » 0.103

-0.1190 -0.1934 -0.1998 -0.0229 0.5133 0.6389 -0.0870 -0.4281 0.2129 0.0147 Probable Error • 0.107

0.1432 -0.0320 0.0169 -0.2008 -0.2437 0.0234 0.6969 -0.1571 0.5490 -0.2574 Probable Error • 0.124

0.1519 -0.2186 0.3128 0.1730 0.4295 0.1684 0.1633 0.7264 0.1851 -0.0005 Probable Error «• 0.129

-0.1264 0.2202 -0.0532 0.3463 -0.0877 -0.0651 -0.4823 0.0763 0.5175 -0.5428 Probable Error • 0.103

-0.0231 -0.1340 -0.0329 -0.0639 -0.2895 -0.0003 -0.2617 0.0861 0.S429 0.7224 Probable Error • 0.113 TABLE 6A

TABLE OF ORDERED LINES

• Line Exp. Freq. Calc. Freq. Inten. Error

130 119.500 119.625 0.768 -0.125 52 119.500 119.670 0.800 -0.170 1 119.500 119.698 0.800 -0.198 153 119.500 119.740 0.811 -0.240 39 119.500 119.787 0.810 -0.287 16 119.500 119.833 0.751 -0.333 198 119.500 119.934 0.778 -0.434 97 119.500 120.201 0.765 -0.701 108 128.300 127.799 1.065 0.501 204 128.300 127.999 1.103 0.301 27 128.300 128.126 1.119 0.174 141 128.300 128.295 1.172 0.005 210 128.300 128.308 1.133 -0.008 171 128.300 128.336 1.140 -0.036 187 128.300 128.365 1.206 -0.065 76 128.300 128.383 1.203 -0.083 17 134.100 134.238 0.118 -0.138 131 135.200 135.570 0.076 -0.370 98 136.100 136.189 0.117 -0.089 199 137.100 137.410 0.078 -0.310 207 138.100 137.925 0.523 0.175 190 138.100 137.983 0.412 0.117 156 139.200 139.199 0.524 0.001 79 139.200 139.246 0.412 -0.046 134 143.000 142.735 0.635 0.265 178 143.000 143.043 0.593 -0.043 20 144.000 143.919 0.638 0.081 57 144.000 144.287 0.604 -0.287 184 145.000 144.575 1.265 0.425 145 145.000 144.871 1.007 0.129 68 146.000 145.870 1.240 0.130 31 146.000 146.196 0.998 -0.196 55 149.800 149.636 1.899 0.164 123 149.800 149.706 1.685 0.094 4 151.000 150.883 1.918 0.117 9 151.000 150.972 1.687 0.028 205 151.700 151.549 0.334 0.151 209 151.700 151.857 1.021 -0.157 109 152.800 152.806 0.214 -0.006 166 153.800 153.343 0.964 0.457 154 157.800 157.704 2.609 0.096 193 157.800 157.897 1.922 -0.097 142 157.800 158.437 0.152 -0.637 44

TABLE 6A--Continued

Line Exp. Freq. Calc. Freq. Inten. Error

182 158.900 158.507 0.431 0.393 40 158.900 158.977 2.761 -0.077 87 158.900 159.391 1.956 -0.491 28 158.900 159.757 0.089 -0.857 66 160.300 160.014 0.419 0.286 200 160.300 160.960 2.385 -0.660 99 161.100 161.196 2.514 -0.096 120 164.200 164.560 0.685 -0.360 53 164.200 164.605 0.947 -0.405 132 166.000 165.712 1.464 0.288 18 166.000 165.869 1.524 0.131 2 166.000 165.942 1.033 0.058 6 166.000 166.076 0.709 -0.076 155 167.500 167.115 0.455 0.385 41 167.500 167.367 0.324 0.133 161 169.200 169.331 1.371 -0.131 208 169.200 169.334 1.323 -0.134 54 172.200 171.880 0.354 0.320 3 172.200 172.054 0.249 0.146 56 174.000 174.419 0.896 -0.419 177 174.000 174.452 0.881 -0.452 194 175.000 175.374 0.574 -0.374 88 175.000 175.379 0.542 -0.379 7 180.300 180.482 0.403 -0.182 121 180.300 180.505 0.430 -0.205 163 345.600 345.605 1.000 -0.005 206 345.600 345.608 1.000 -0.008 175 345.600 345.636 1.000 -0.036 104 345.600 345.872 0.995 -0.272 60 347.000 346.848 1.000 0.152 157 3<»7.000 346.881 1.000 0.119 115 347.000 346.893 0.995 0.107 82 347.000 346.899 1.000 0.101 23 347.000 347.055 0.993 -0.055 169 347.000 347.093 1.000 -0.093 93 347.000 347.098 1.000 -0.098 45 347.000 347.145 1.000 ' -0.145 34 348.500 348.219 0.993 0.281 12 348.500 348.364 1.000 0.136 71 348.500 348.388 1.000 0.112 5 348.500 348.392 1.000 0.108 APPENDIX B

TABLE 7B

LIST OF TERPENOIDS C.-C,, 5 14

Agglomerone Dimethyl-p~ C13HU°4 °10M14° tolylcarbinol Ascaridole °10H16°2 Elaholtzia Ketone C10H14°2 Borneol C10MU° Eucarvone C10H14° Calythrone C12HU°3 Fenchone C10HU°

Camphene C H 10 16 9(-Fenchyl Alcohol C10H18°

Camphor C H 10 16° Geranial °10H16°

^-Carene C H 10 16 Geraniol C10HW° Carquejol °IOhK° ctf-Ionone C13H20° Carvomenthone C10H18° P-Ionone C13H20° Carvone °10H14° neo-Of-Irone CHH22°

1,8-Cineole C H 10 1S° neo-Iso-of-irone C14H22° Citronellal C10H18° Isomenthone CI0H18°

j?-Citronellol C H 10 20° Isophorone C9 H14° p-Cymene C10HU Isopulegol °10H18° Dehyd roangus 11one CUH14°3 Lavandulol C10H18° Linalool trans-Dihydro-a(- C10H18° C10H20° terpineol Linalyl Acetate C12H20°2 Dimethyl C5K8°2 Acrylic Acid trana-p-Menthan- C10H20° 8-ol y,y-Dime thy la 1 ly 1 C5H10° Menthol Alcohol C10H20°

45 46

TABLE 7B—Continued

Menthone Sabinene C10H18° CL0H16 Methyl-trans-chrysan Tasmanone _CllH18°2 C14H20°4 themum Monocarboxyl< ate Q(-Terpinene C10HU Myrtenal $-Terpinene cIOhR° C10H16 Neral of-Terpineol C10H16° C10H18° Nerol C10H18° Terpinen-4-ol C10HU° trans-P-Ocimene ^-Thujaplicin C10H16 ClOH12°2 o(-Phe 11 and rene o(-ThuJene C10H16 C10H16 o(-Plnene Thujone C10H16 C10H16° ^-Plnene Thymol C10HU w Thymoquinone Pinocampheol C10H18° C10H12°2

Pinocamphone Verbenol C10H16° Verbenone Piperitenone cxo"u° C10HU°

Pulegone C10H16°

Rose Oxide C10H18° TABLE SB LIST OF TERPENOIDS C-.-C,,

Acoric Acid oC-Cyperone C15H24°4 C15H22° Agarofuran Dams in cisV C15H20°3 Ambrosiol Dime thylisoshelloate °15H22°4 C17H24°6 Arborescin Dimethylshelloate C15H20°3 C17H24°6 AriatoLone Drimenoi C1JH24° C15H26° Balchanolide Dunnione C15H22°3 C15H14°3

Balduilin C17H20°5 Elemol C15H26° P-Bergamotene Eremophilone C15H24 C15H22° P-Bisabolene °15K24 /?-Eudesmol °15H2t° ^-Bourbonene Flexuosine A C15H24 C17H24°6 Bulnesol Germacrone C15H26° C15H22° S-Cadinene Gibberellic Acid °15H24 CWH20#6 f-Cadinene Globicin C15H24 C17H22°5 oi-Gadinol C15H26° Globulol C15H26° Carissone C15H24°2 Guaiol C15H26° Carotol ^•Himach&lene C15H26° C15H24 Caryophyllene o(«Humulene C15H24 C15H24 Caryophyllene Oxide Humulene Diepoxide C15H24° C15H24°2 oC-Cedrene Hyd roxyd ihyd ro- C15H24 C15H22°2 eremophilenolide P'Cedrene C15H24 6-Hydroxyeremo- C15H22°3 Cedcol philone C15H26° Confertifolin Hyd roxyperezone C15H22°2 C15H20°4

Cryptoacorone C15H24°2 Hydroxyvaleranone C15H26°2 Cyperene Imperatorin C15H24 C16HL4°4 48

TABLE 8B—Continued

IresIn C15H22°» Patchouli Alcohol C15H26° Isoacorone Perezone C15H24°2 C15H20°3 Isoalantolactone Petashione C15H20°2 C15H18°2 Ivalin C15H20°3 Picrotin C15HU°7 Jatamanshic Acid Q(-Picrotoxinic C15H22°2 Vie0* Acid Juniper Camphor C15H2«° &>Ficrotoxinic C15H18°6

Tricyc1ocyperene C15H24

Tutin °15H18°6

Valerenal C15H22° Valeranone C15H26°0 Vlridiflorol C15H26° o(-Ylangene C15H2* Zierone C15H22° 50

TABLE 9B

LIST OF TERPENOIDS C2Q-C40

Abietic Acid Mongynol A C20H30°2 C20H32°0 9-Acetoxyroyie- Paeniflorin anone C22H30°5 C33H38°16 Pentaacetate Ambrein C30H52° Paeniflorin C31H36°15 ^-Amryin Tetraacetate C30H50° Araucarenolone Palustric Acid C20H28°4 C20H28° Araucarolone Pentaacetyl Monotropeln C20H30°3 C26H32°16 Methyl Ester Archangelin C21H22°4 Pimaric Acid C20H30°2 ^-Carotene C40H56 Quassin C22H30°6 Columbia C20H22°5 Rimuene C20H32 Dehydroabietic Acid C20H28°2 Royleanone C20H28°3 Eremolactone C20H26°2 Salaninn C34H44°9 Euphol C30H50° Sand racop imar ic C20H30°2 H«xaacetyl Aucubia Acid C27H34°15 Humulone Sclareol C21H30°5 C21H30° Isopimaric Acid Swietenine C20H30°2 C31H40°9 Levopimaric Acid Swietenolide C20H30°2 C27H34°8 Lycopene Tetraacetyl Aaper- C40H56 C26H32°16 uloside Manool C20H34° Tinophyllone C21H24°6 Marrubiin C20H28°4 Uabellipenin C24H30°3 Mexicanolide C27H32°7 TABLE 10B

TERPENOIDS OVER C, AND STEROIDS 40n

Androstadledlone C1?H24^2

5o(-Androstan-3°(-oi-17-one ^19^30^2

5o(.-Androstan-3^-ol- 17-one C19**30^2

5^-Androstan-3o(-ol-17-one ^19H30^2

Androstenedlone **19^26® 2

4-Androstene-3,11,17-trione C19H24°3

/^'^'^-Androstatriene-3,17-dione ^19H22®2

C Calciferol (Vitamin D2> 28^44

Cevadine C32H49N09

Cholestanol C^H^O

Cholesterol C--H.,0 27 46 Conessine C,.H.-N, 24 40 2 Cortexolone ^21^30^4

Corticosterone C21H30^4

Cortisone Acetate C23**30^6

Cucurbitacin I (Elatericin B) C^qH^Oj

Desoxycortlcosterone Acetate C23H32^4

Digoxigenin C23H34°5

Diosgenin C27H«°3

Dolichol C100H164°

Estriol C18H2*°3

Fucosterol C^H^O TABLE 10B—Continued

Gitoxigenin G23H32°5

Hecogenin Acetate ^29^44^5

Hydroxycortisone Acetate

IsofucosteroL C^H^gO

24-Ketocholesterol co-»H/, °-i 27 44 2 Lanosterol C30H50°

Lumisterol-3 C-.H..0 28 44 Lupeol Acetate ^32^52^

Prednisolone ^21^28^5

Predisone C21H26°5

5^-Pregnane-I7o(l21-dihydroxy-3,ll, 20- C..H..0. trione-2l*acetate

5^-Pregnane-3o(, 17o(-diol-ll, 20-dione C21H32°4

Pregnenolone Acetate C23H34°3

Progesterone C21H30°2

Shottenol C^H^gO

Smilagenin Acetate C^gH^gO^

Solanesol C^H^O

Stigmasterol C29H48®

Testosterone C19H28^2

(^•Tocopherol C29H50°2

Ubiquinone (Q1(J)

Vitamin D3 C27H44 APPENDIX C

NMR AND IR SPECTRA

53 - w aJuil L A A-X II M ff 5.0 70 mn(T)mtir) "

Figure 2b. Verbenone (IX) simulated spectrum (100 Mc),

290I I MOI SO

A ' • • • i.... y . i 4.0 5.0 «.0 KT) 7.0 1.0 9.0

Figure 2a. Verbenone (IX) experimental spectrum (100 Mc). Figure 3b. Myrtenal (X) simulated spectrum (100 Mc).

9.0 4.0 7.0 (T) «.o 9.0

Figure 3a. Myrtenal (X) experimental spectrum (100 Mc). li h A4 i J V > 4.0 i 5.0 7.0 FPM(T) (.0 9.0 Figure 4c. flf-Pitiene (XI) simulated spectrum (100 Mc), system 2.

+ +

1 CPS uo i JL

5.0 6.0 ffM(r)

Figure 4b. 0(-Pinene (XI) simulated spectrum (100 Mc), system 1 (Insert). Ut Figure 4a. 0(-Pinene (XI) experimental spectrum (100 Mc). Jlui

lit' 1 'Hdffl'-TI

Figure 5* Myrtenal (X) experimental and simulated spectra (60 Mc} upper 3^ negative, middle J^ positive).

Figure 6. Commercial methylheptenone (XVI)* Figure 7. 2,6-Dimethyl-5-heptene-2-ol (XVII).

7.0 •0

Figure 8. Ct+f -Cyclogeraniolene (XVIII + XIX). 59 WAVELENGTH (MICRONS] 0.0 0.0

S.20

£•30 §.40CO .40 <.50 .50 .60 .60 .70 .70 1.0 1.0 1.5 4000 3000 2000 1500 1200 1000 900 800 700 CM

Figure 9. Commercial methylheptenone (XVI; neat).

WAVELENGTH MICRONS 3 4 5 6 7 8 9 10 11 12 13 14 15 J : 1 1 » I I I ' . I I • . A . ! . I ). . 1 I I . I. . I I ( . 11

-1

4000 3000 1000 900

Figure 10. 2,6-Dimethyl-5-heptene-2-ol (XVII; neat). 60

WAVELENGTH (MICRONS) -I3 H 4i ! 51 1 .6 1 . 7 8 9 10 1 11' . I . 12 . , 13I 1 I 14.1. I 15L-

4000 3000 1000 900

Figure 11. Ct+fi -Cyclogeraniolene (XVIII + XIX; neat).

! WAVELENGTH (MICRONS) 1 J3 _l 4 I 5I—,—i, 6I ,1 71,1 8I rJ 9L I 10 ' 11111 12 13I I 14I 15

4000 3000 1000 900

Figure 12. Acetylcyclogeraniolene (XX; neat). 61

Figure 13. Acetylcyclogeraniolene (XX).

7T 1.0

Figure 14. ^-Cyclolavandulic acid (XXI). WAVELENGTH (MICRONS)

0.0 0.0

.40

.60 .70 1.0 1.0 1.5 4000 3000 2000 1500 1200 1000 900 800 700 CM

Figure 15. ^-Cyclolavandulic acid (XXI; KBr).

WAVELENGTH (MICRONS)

0.0 0.0

U.20

.40CO 40 <.50 .60 .70 .70 1.0 1.0

4000 3000 2000 1500 1200 1000 900 800 700 CM

Figure 16. ^•Cyclolavandulol (XV; CCl^). 1 m » mI m e 9* i

ft - . X

. i j i - j» 1 >!> ' A —i" 1 'J» 1 U 16 ii • 1 1

Figure 17. p-Cyclolavandulol (XV).

3.0 4.0 1.0 to

Figure 18. £-Cyclolavandulyl bromide (XX11I). Figure 19* Natural archangelln (XXVI; 60 Mc).

Figure 20• Natural archangelln (XXVI; 100 Mc). 65

.JiUJVJjL

1.4' ' 'MClV '4»

Figure 21. ^-Cyclolavandulyl umbelllferonyl ether (XXV; 60 Mc).

T ' | 1 , T 1 | ' T , .TJirt.,y,. , J , I"; • ;:T:7-| >*• 11 m no M NO • at i i

f *• M

rWA-jA ic. "" T . 1 ' - i • A ii •»" »

Jl , - vsJ^Lj^Cj

( 1.... 1... jg 1 1 .1... 11 A " ' 11 'w\»v '—«!i" ' ' —i!r • ' • i 1

Figure 22, ^-Cyclolavandulyl umbelllferonyl ether (XXVj 100 Mc). 66 WAVELENGTH (MICRONS) 6 7 8 9 10 11

I I

r1. 4000 3000 1000 900

Figure 23. Natural archangelin (XXVIj CCl^).

WAVELENGTH 100 100

80 1U Z 60 60

4000 3000 2000 1500 1200 1000 900 800 700 CM'

Figure 24. fl-Cyclolavandulyl umbelliferonyl ether (XXV; CCl^).

I rir—

Figure 25. Compound B.

«jo mtlfl Ji- JE£_

u A,,i* i '••WiV»r I

-i!r •A11 'mini'—ir -j!r tr~

Figure 26, Perezone (VIII). LIST OP REFERENCES

1. N. F. Ramsey, Phya. Rev., 78, 699(1950).

2. N. F. Ramsey and E. M. Purcell, ibid.. 85. 143(1952).

3. J. A. Pople, W. 6. Schneider, and H. J. Bernstein, "High-Resolution Nuclear Magnetic Resonance," McGraw-Hill Book., Inc., New York, 1959.

4. H. M. McConnell, J. Chem. Phys., 24, 460(1956).

5. J. A. Pople and A. A. Bothner-By, ibid.. 42, 1339(1965).

.6. M. Karplus, D. H. Anderson, T. C. Farrar, and H. S. Gutowsky, ibid.. 27, 597(1957).

7. M. Karplus and D. H. Anderson, ibid.. 30. 6(1959).

8. M. Karplus. ibid.. 30, 11(1959).

9. S. Sterahell, Rev. Pure and Appl. Chem., 14, 15(1964).

10. S. Musamune, J. Am. Chem. Soc.( 86. 735(1964).

11. J. Meinwald and A. Lewis, ibid., 83, 2769(1961).

12. K. B. Uiberg, B. R. Lowery, and B. J. Nist, ibid., 84, 1594(1962)

13. J. Meinwald and Y. C, Meinwald, ibid.. 85. 2514(1963).

14. M. Barfield, J. Chem. Phys.t 41,' 3825(1964); ibid., 46, 811(1967)

15. S. Castellano and A. A. Bothner-By, ibid., 41, 3863(1964).

16. J. D. Swalen and C. A. Reilly, ibid.. 37, 21(1962).

17. W. D. Keller, T. R. Lusebrink, and C. H. Sederholm, ibid., 44, 782(1965). 18. R. A. Hoffman, ibid., 33, 1256(1960).

19. Y. Arata, H. Shimisu, and S. Fujiwara, ibid., 36, 1951(1962).

20. L. M. Jackman, "Applications of Nuclear Magnetic Resonance Spec- troscopy in Organic Chemistry," Pergamon Press, London, 1959.

68 69

21. J. D. Roberts, "Nuclear Magnetic Resonance," McGraw-Hill, New York, 1959.

22. The NMR-EPR Staff of Varian Associates, nNMR and EPR Spectroscopy," Pergamon Press, New York, I960.

23. D. Chapman and P. D, Magnus, "Introduction to Practical High Reso lution Nuclear Magnetic Resonance Spectroscopy," Academic Press, New York, 1966.

24. K. B. Wiberg and B. J. Nist, "The Interpretation of NMR Spectra," Benjamin, New York, 1962.

25. J. D, Roberta, "An Introduction to the Analysis of Spin-Spin Splitting in High Resolution Nuclear Magnetic Resonance Spectra," Benjamin, New York, 1961.

26. N. S. Bhacca, D. H. Williams, "Applications of NMR Spectroscopy in Organic Chemistry*" HoIden-Day, Inc., San Francisco, 1964.

27. J. L. Simonson, "The Terpenes," Cambridge University Press, Cam bridge (Eng.), Vol. I-V, 1937-47.

28. P. de Mayo, "Mono- and Sesquiterpenolds," Interscience Publishers, Inc., New York, 1959| ibid.. "The Higher Terpenoids."

29. A. R. Plnder, "The Chemistry of Terpenes," John Wiley and Sons, New York, 1961.

30. L. Pliva, V. Horak, V. Herout, and F. Sorm, "Die Terpene," Akademic Verlag, Berlin, 1960.

31. G. Ourisson, S. Munavalli, and C* Ehret, "Selected Constants of Sesquiterpenolds," Pergamon Press, New York, 1966.

32. C. W. Sigel, B.S. Thesis, University of Illinois, 1963.

33. G. Ourisson and J. M. Lehn, Bull. Soc. China. France, 1137(1962); ibid., p. 1832.

34. A. M. Shamma, R. E. Glick, and R. 0. Mumma, J. Org. Chem., 27, 4512(1962).

35. N. S. Bhacca, D. P. Hollis, L. F. Johnson, and E. A. Pier, "NMR Spectra Catalog," Varian Associates, Palo Alto, California, 1963.

36. Sadtier Research Laboratories, "The Sadtler Standard Spectra," Sadtler Research Laboratories, Inc., Philadelphia, Pa., 1966.

37. E. Spath and H. Holzen, Ber., 66B. 1137(1933). 70

38. E. Spath and L. Kahovic, ibid.. 1146(1933).

39. E. Spath and P. Kainrath, ibid., 70B, 2272(1937).

40. W. L, Stanley and S. H. Vannier, J. Am. Chem. Soc., 79. 3488(1957).

41. Y. N. Sharma, A. Zaman, and A. R. Kidwai, Tetrahedron, 20. 87(1964).

42. Y. N. Sharma, R. C. Sharma, A. Zaman, and A. R. Kidwai, Naturwis* senschaften, 5JL, 537(1964).

43. A. G. Caldwell and E. R. H. Jones, J. Chem. Soc., 540(1945).

44. D. P, Chakraborty and S. K. Chaudhary, Trans. Bose Res. Inst., 24, 15(1961).

45. J. H. Schauble, Ph.D. Thesis, University of Illinois, 1963.

46. A. Chatterjee and S. Sen Gupta, Tetrahedron Letters, 1961(1964).

47. F. Kogl and A. G. Boer, Rec. Trav. Chim., Pays-bas, 54, 779(1935).

48. F. Walls, J. Padilia, P. Joseph-Nathan, F. Giral, and J. Romo, Tetrahedron Letters, 1577(1965).

49. K. Wiberg, G. M. Lampman, R. P. Ciula, D. S. Connor, P. Schertler, and J. La vanish, Tetrahedron, 2,1, 2749(1965).

50. K. C. Ramey, D. C. Lini, R. H. Moriarity, H. Bopal, and H. G. Welsh, J. Am. Chem. Soc., 89, 2401(1967), and references therein.

51. H. S. Gutowsky, M. Karplus, and D. M. Grant, J. Chem. Phys., 31. 1278(1959).

52. E. Lustig, ibid.. 37, 2725(1962).

53. K. L. Servis and J. D. Roberts, J. Phys. Chem., 67, 2885(1963).

54. N« Karplus, J. Am. Chem. Soc., 84. 2458(1962).

55. P. C. Lauterbur and R. J. Kurland, ibid.. 3405(1962).

56. R. Freeman and W. A. Anderson, J. Chem. Phys., 39, 806(1963).

57. R. R. Fraser, R. U. Lemieux, and J. D. Stevens, J. Am. Chem. Soc., 83, 3901(1961).

58. R. Freeman and W. A. Anderson, J. Chem, Phys. 37. 2053(1963). 71

59. K. A. Lauchlan and D. H. Wiffen, Proc. Chem. Soc., 144(1962).

60. K. B. Wlberg and B. J. Nist, J. Am. Cham. Soc.( 83, 1226(1961).

61. 0. L. Chapman, Ibid.. 85. 2014(1963).

62. G. V. Smith and H. Kriloff, ibid.. 2016(1963).

63. P. Laszlo and P. von R. Schleyer, Ibid.. 2017(1963).

64. F. Kaplan and J. D. Roberts, ibid., 83, 4666(1961).

65. R. Freeman, K. A. McLauchlan, J. I. Musher, and K. G. R. Pachler, Mol. Phys., 5, 321(1962).

66. R. Freeman and N. S. Bhacca, J. Chem. Phys., 38, 1088(1963).

67. M. Barfield and J. Reed, private communication, 1967.

68. A. Raasat, C. Jefford, J. Lehn, and B. Wage 11, Tetrahedron Letters, 233(1964).

69. M. Barfield and D. M. Grant, J. Am. Chem. Soc., 85, 1899(1963) and references therein.

70. von A. Brenner, U. Steiner, and H. Schinz, Helv. Chim. Acta., 35, 1341(1952).

71. von A. Brenner and H. Schinz, ibid.. 1621(1952).

72. A. Lauchenauer and H. Schinz, Helv. Chim. Acta., 31_f 1271(1949).

73. C. Ferrero and H. Schinz, ibid., 39, 2109(1956).

74. W. S. Johnson, 1. A. David, H. C. Dehm, R. J. Highet, E. W. Warn- hoff, W. D. Wood, and E. T. Jones, J. Am. Chem. Soc., 80, 661(1958).

75. E. J. Zavarin, J. Org. Chem., 23. 1198(1958).

76. N. R. Trenner, B. H. Arison, C. H. Shunk, D. E. Wolf, and K. Folkers, J. Am. Chem. Soc., 81, 2026(1959).

77. E. Cortes, M. Salmon, and F. Walls, Bol. Inst. Quim. Univ. Nacl. Auton. Mex., JL7, 19(1965).

78. G. DuPont, R. Dulou, and 0. Mondou, Bull. Soc. Chim. France, 60(1953). 72

79. J* E. Callen, C. A. Dornfeld, and G, H, Coleman, "Organic Synthe sis," Coll. Vol. 3, John Wiley and Sons, Inc., New York, 1955, p. 27.

80. von A. Brenner, Helv. Chim. Acta., 3£, 1340(1952).

81. L. Bateman, J. I. Cunneen, and E. S. Waight, J. Chem. Soc., 1714(1952).

82. J. Demaecker and R. H. Martin, Nature, 173. 266(1954).

83. F. Arndt, "Organic Synthesis," Coll. Vol. 2, John Wiley and Sons, Inc., New York, 1943, p. 165.

84. Y. N. Sharma, A. Zaman, A. R. Kidwai, R. B. Bates, and V. P. Thalacker, Tetrahedron, 22. 3221(1966).

85. G. Rodigherio and C. Antonello, IL Farmaco, Ed. Sci., 10. 1899 (1955).

86. K. W. Greenlee and V. G. Wiley, J. Org. Chem., 27, 2304(1962).