Some Chemical Constituents of the Brown Alga, Agarum Cribosum and the Effect of Electrostatic Interactions on the Stereochemistry of the S(N)2' Reaction

University of New Hampshire University of New Hampshire Scholars' Repository

Doctoral Dissertations Student Scholarship

Spring 1980

SOME CHEMICAL CONSTITUENTS OF THE BROWN ALGA, AGARUM CRIBOSUM AND THE EFFECT OF ELECTROSTATIC INTERACTIONS ON THE STEREOCHEMISTRY OF THE S(N)2' REACTION

JOAN DEBORAH NEWBURGER

Follow this and additional works at: https://scholars.unh.edu/dissertation

Recommended Citation NEWBURGER, JOAN DEBORAH, "SOME CHEMICAL CONSTITUENTS OF THE BROWN ALGA, AGARUM CRIBOSUM AND THE EFFECT OF ELECTROSTATIC INTERACTIONS ON THE STEREOCHEMISTRY OF THE S(N)2' REACTION" (1980). Doctoral Dissertations. 1255. https://scholars.unh.edu/dissertation/1255

This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. INFORMATION TO USERS

This was produced from a copy of a document sent to us for microfilming. 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 material submitted.

The following explanation of techniques is provided to help you understand markings or notations 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 through an image and duplicating adjacent pages to assure you of complete continuity.

2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacentTrame.

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

4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.

University Microfilms International 300 N. ZEEB ROAD, ANN ARBOR, Ml 48106 18 BEDFORD ROW, LONDON WC1 R 4EJ, ENGLAND 8027795

N e w b u r g e r, Jo a n D e b o r a h

SOME CHEMICAL CONSTITUENTS OF THE BROWN ALGA, AGARUM CR1BOSUM AND THE EFFECT OF ELECTROSTATIC INTERACTIONS ON THE STEREOCHEMISTRY OF THE S(N)2’ REACTION

University o f New Hampshire Ph.D. 1980

University Microfilms International300 N. Zeeb Road, Ann Arbor, MI 48106 18 Bedford Row, London WC1R 4EJ, England PLEASE NOTE: In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark v ' ' .

1. Glossy photographs ______2. Colored illustrations ______3. Photographs with dark background______

4. Illustrations are poor copy

5. °rint shows through as there is text on both sides of page ______

6. Indistinct, broken or small print onseveral pages s /

7. Tightly bound copy with print lost in spine ______

8. Computer printout pages with indistinct print ______

9. Page(s) ____ lacking when material received, and not available from school or author

10. Page(s) ______seem to be missing in numbering only as text follows

11. Poor carbon copy ______12. Not original copy, several pageswith blurred type ____ 13. Appendix pages are poor copy ______14. Original copy with light type ______15. Curling and wrinkled pages ____ _ 16. Other

University M icrdnims international 300 N ZEES RD.. ANN AR30B Ml 48106'31 31 761-4700 SOME CHEMICAL CONSTITUENTS OF THE BROWN ALGA, AGARUM CRIBOSUM

And

THE EFFECT OF ELECTROSTATIC INTERACTIONS ON THE STEREOCHEMISTRY OF THE Sn 2' REACTION

JOAN D. NEWBURGER B. A. , Grinnell College, 1974

A DISSERTATION

Submitted to the University of New Hampshire in Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY in CHEMISTRY

May, 1980 This dissertation has been examined and approved.

John Uebel ;try

Paul R. Jones, gzofessor of Chemistry

r f c w o l $ > , ______Jairtes D. Morrison, Professor o£ Chemistry

Alexander R. Amell, Professor of Chemistry

4 A S Miyoshi/'Ikawa, Professor of Biochemistry

% m o DdEe DEDICATION

This dissertation is dedicated with great love and affection to my parents, Adele and Sylvan, and to my sister, Harriet. ACKNOWLEDGMENT

I wish to express my sincere gratitude and appreciation to my advisor, Dr. J. John Uebel, for his invaluable guidance and encouragement during the course of my graduate education and dissertation research. I would also like to thank Dr.

Paul R. Jones and, in particular, Dr. Kenneth K. Andersen for their help during the sabbatical year of Dr. Uebel.

I wish to thank Dr. Arthur Matheson and his students for their considerable assistance in the phycological aspects of my work.

I wish to thank Dr. Robert Gagosian of the Woods Hole •4 Institution of Oceanography for his collaboration on the sterol research and Dr. Cathy Costello of the Massachusetts

Institute of Technology for her help in the fatty acid data collection and for high resolution mass spectra.

For their technical assistance, I wish to acknowledge

Michael Pazdon and Kathy Gallager. In particular, Mike spent many hours helping me with the HPLC and GC/MS instru ments .

For their help in the preparation of this manuscript,

I wish to thank Diana Schuman and Dee Cardin.

I am grateful to the Chemistry Department for financial support in the form of a Teaching Assistantship and to the

Leslie S. and Iola Hubbard Marine Program Fund for summer financial support and funding for supplies. A very special acknowledgment is due Dr. Gene Wubbels of Grinnell College and Dr. Edward Radford of the University of Pittsburgh (formerly of Johns Hopkins University) for their advice, encouragement, and understanding during my undergraduate career.

V ABSTRACT

SOME CHEMICAL CONSTITUENTS OF THE BROWN ALGA, AGARUM

CRIBOSUM AND THE EFFECT OF ELECTROSTATIC INTERACTIONS ON

THE STEREOCHEMISTRY OF THE SN2 ' REACTION

JOAN D. NEWBURGER

University of New Hampshire, May 1980

Part 1: Some Chemical Constituents of the Brown Alga,

Agarum cribosum.

Agarum cribosum was examined for major chemical con stituents. Halogenated terpenes found in red algae were not present in this seaweed. The sterol content was de termined by GC/MS analysis and six sterols were identi fied: fucosterol, 24-methylenecholesterol, cholesterol,

24-ketocholesterol, saringosterol, and desmosterol. The presence of desmosterol in brown algae had not been pre viously confirmed. A seventh sterol could not be identified from its mass spectrum. The fatty acid content was also determined by GC/MS analysis. The major fatty acids were

^16-0’ ^16*1’ ^ 2 0 - 4 anc* ^ 2 2 -5 ‘ suSar mannitol was isolated. Part 2: The Effect of Electrostatic Interactions on the

Stereochemistry of the SN2' Reaction.

The stereochemistry of the S^2' reaction was studied using (trans-6 -t-butyl-2 -cyclohexen-l-yl)trimethylammonium tetrafluoroborate as a substrate for nucleophilic attack. The substrate was reacted with two nucleophiles, piperidine and sodium propanethiolate. To the extent that transition state electrostatic interactions can influence the direction of nucleophilic attack, it was expected that piperidine would attack anti to the leav ing group while propanethiolate would attack syn. The experiments confirmed these predictions, piperidine attacking 80% anti and propanethiolate attacking 92% syn, It was concluded that transition state electro static interactions play a role in determining the di rection of nucleophilic attack. TABLE OF CONTENTS

ABSTRACT ...... vi

LIST OF TABLES (Part 1) xii

LIST OF FIGURES (Part 1 ) ...... xiii

LIST OF TABLES (Part 2) xvii

LIST OF FIGURES (Part 2 ) ...... xviii

PART 1: SOME CHEMICAL CONSTITUENTS OF THE BROW

ALGA, Agarum Cribosum ...... 1

I. INTRODUCTION ...... 2

II. RESULTS AND DISCUSSION ...... 21

1. Column Chromatography Separation ...... 21

2. Sterols of Agarum cribosum ...... 24

3. Fatty Acids of Agarum cribosum ...... 50

4. Isolation of Mannitol ...... 67

III. EXPERIMENTAL...... 69

G e n e r a l ...... 69

Extraction of Agarum cribosum ...... 70

Open Column Chromatography of Nonpolar Constituents ...... 70

Saponification of Crude Extract ...... 71

Separation of Non-saponifiable Lipids ...... 72

Identification of Sterol Compounds ...... 72

Isolation of Fucosterol from Ascophyllum n o d o s u m ...... 74

24-Ketocholesteryl Acetate ...... 75

viii 24-Ketocholesterol ...... 76

24-Ethynyl-24-hydroxycholesterol ...... 77

24-Hydroxy-24-vinylcholesterol (saringosterol) . 77

Formation of Synthetic Sterol Acetates ...... 78

Isolation of Fatty Acids ...... 79

Formation of Fatty Acid Methyl Esters ...... 79

Identification of Fatty Acid Methyl Esters . . . 80

Isolation of Mannitol ...... 81

IV. BIBLIOGRAPHY...... 82

V . APPENDIX ......

A. Mass Spectra of Sterol Acetate Standards . . 8 8

B. FAME Mass Spectra and Ionization Intensity P l o t s ...... 95

G. l-H-NMR S p e c t r a ...... 105

D. IR Spectra ...... 121

PART 2: THE EFFECT OF ELECTROSTATIC INTERACTIONS

ON THE STEREOCHEMISTRY OF THE S^'

REACTION...... 124

I. INTRODUCTION...... 125

II. RESULTS AND DISCUSSION...... 139

1. Synthesis of Substrate 1^...... 139

2. Reaction of Substrate 1 with Piperidine . . . 142

3. Reaction of Substrate 1 with Propanethiolate

Nucleophile ...... 161

III. CONCLUSION...... 167 IV. EXPERIMENTAL...... 169

General ...... 169

4-t-Butylcyclohexanone (3)...... 170

4-t-Butyl-2-bromocyclohexanone Ethylene

Ketal (4) ...... 171

4-t-Butyl-2-cyclohexen-l-one Ethylene

Ketal (5) ...... 171

4-t-Butyl-2-cyclohexen-l-one (6^)...... 172

4-t-Butyl-2-cyclohexen-l-ol ( 7 ) ...... 173

4-_t-Butyl-2-cyclohexen-l-yl 2,2,2-Trichloro-

ethanimidate (8 ) ...... 173

2,2,2-Trichloro-N-(6-t-Butyl-2-cyclohexen-

l-yl)acetamide (9)...... 174

N-(trans-6-t-Butyl-2-cyclohexen-l-yl)amine

(1 0 ) ...... 175

N-(trans-6-t-Butyl-2-cyclohexen-l-yl)-

N,N-dimethylamine (11)...... 175

(trans-6-t-Butyl-2-cyclohexen-l-yl)trimethyl-

ammonium Tetrafluoroborate (1)...... 176

S^2' Piperidinolysis of 1^...... 177

4-jt-Butyl-2-cyclohexen-l-yl 2,6 Dichloro-

benzoate (2 0 ) ...... 179

N-(cis and trans-4-t-Butyl-2-cyclohexen-

l-yl)piperidine (13^, 12) Alternate

Synthesis 179

x N-(cis and trans-4-t-Butyl-l-cyclohexen-

1-yl)piperidine and Hydrogenation to

give 14 and L5...... 180

3^2' Reaction of Propanethiolate with

Substrate 1...... 181

1-(Propylthio)-trans and cis-4-t-butyl-

2-cyclohexene (24, 25) Alternate

S y n t h e s i s ...... 182

Kinetic Experiment with the Piperidine

Nucleophile ...... 183

V. BIBLIOGRAPHY...... 184

VI. ADDENDUM

High Resolution Mass Spectrum of 2,2,2-

Trichloro-N-(6-t-Butyl-2-cyclohexen-l-

yl)acetamide (9)...... 187

VII. APPENDIX

A. XH-NMR SPECTRA...... 188

B. 1 3 C-NMR SPECTRA ...... 213

C. IR SPECTRA...... 220

xi LIST OF TABLES

Part 1

Table I: Results of Open Column Chromatography

E x p e r i m e n t ...... 22

Table II: Sterol Composition of Agarum cribosum . 27

Table III: Side Chain Fragmentations of Sterol

A c e t a t e s ...... 28

Table IV: Sterols of Brown Algae ...... 42

Table V: Fatty Acid Methyl Esters of Agarum

c r i b o s u m ...... 55

Table VI: Fatty Acids of Brown Algae ...... 61

Table VII: Comparison of Newburger and Ackman FAME

% Weight D a t a ...... 65

xii LIST OF FIGURES

Part 1

Introduction, Results and Discussion

Figure 1: Bromophenols of Red Algae ...... 4

Figure 2: Halogenated Terpenes in Laurencia and

Rhodophytin ...... 6

Figure 3: Gamones of Brown Algae ...... 8

Figure 4: Biogenesis of Gamones in Brown Algae .... 10

Figure 5: C 1 1 Hydrocarbons of Dictyopteris ...... H

Figure 6 : Sulfur Compounds of Dictyopteris ...... 12

Figure 7: Phloroglucinol Derivatives of Brown

A l g a e ...... 14

Figure 8 : Other Phenolic Compounds of Brown Algae . . 17

Figure 9: Terpenes of Brown Algae ...... 18

Figure 10: Gas Chromatograph Trace of Sterol Acetates

(Fraction 4) from GC/MS Analysis ...... 26

Figure 11: Mass Spectrum of GC Peak I (Cholesteryl

A c e t a t e ) ...... 29

Figure 12: Mass Spectrum of GC Peak II (Desmosteryl

A c e t a t e ) ...... 30

Figure 13: Mass Spectrum of GC Peak III (24-Methylene-

cholesteryl Acetate) ...... 31

Figure 14: Mass Spectrum of GC Peak IV (Fucosteryl

A c e t a t e ) ...... 32

xiii Figure 15: Mass Spectrum of GC Peak V (24-Keto-

cholesteryl Acetate) ...... 33

Figure 16: Mass Spectrum of GC Peak VI (Saringosteryl

Acetate) ...... 34

Figure 17: Mass Spectrum of GC Peak VII (unknown) . . 35

Figure 18: Sterols of Agarum, cribosum ...... 36

Figure 19: MS Fragmentation of Saringosteryl Acetate. 45

Figure 20: Proposed Transformation of Cycloartenol

to C2 4 “substituted Sterols 46

Figure 21: Demethylation of Lanosterol at C-^ . . 47

Figure 22: Demethylation of Cycloartenol at C^ 48

Figure 23: Side Chain Alkylation of Cycloartenol 49

Figure 24: Fatty Acid Methyl Ester Chromatogram . 51

Figure 25: Total Ionization Plots #6097 and #6098

from MIT GC/MS Analysis ...... 53

Figure 26: Ionization Plot for FAME ...... 53

Figure 27: Identification of Overlapping FAME's by

Use of Ionization Intensity Plots .... 57

Appendix A

Figure 1: MS Fucosteryl Acetate (Ascophyllum nodosum) 89

Figure 2: MS Synthetic 24-Ketocholesteryl Acetate 90

Figure 3: MS Synthetic Saringosteryl Acetate .... 91

Figure 4: MS Synthetic Sa r i n g o s t e r o l ...... 92

Figure 5: MS Cholesteryl Acetate (Fisher) . . . . . 93

Figure 6 : MS Desmosteryl Acetate (Applied Science) . 94

xiv Appendix C

Figure 1: ^-H-NMR (CDClg) Column Chromatography

Fraction 20-27 ...... 106

Figure 2: ^H-NMR (CDCI3 ) Column Chromatography

Fraction 28-29 (Chlorophyll) ...... 107

Figure 3: •'-H-NMR (CDCI3 ) Column Chromatography

Fraction 30-31 (Sterols) ...... 108

Figure 4: ^"H-NMR (CDCI3 ) Column Chromatography

Fraction 32-35 ...... 109

Figure 5: ^H-NMR (CDCl^) Column Chromatography

Fraction 37-48 ...... 110

Figure 6 : ^H-NMR (CDCI3 ) Non-saponifiable Lipids

(Fraction 1 ) ...... Ill

Figure 7: ^-H-NMR (CDCI3 ) Non-saponif iable Lipids

(Fraction 2) P h y t o l ...... 112

Figure 8 : ^-H-NMR (CDClg) Non-saponif iable Lipids

(Fraction 3) Fucosterol and 24-Methylene-

cholesterol ...... 113

Figure 9: ^H-NMR (CDCl^) Non-saponifiable Lipids

(Fraction 4 ) ...... 114

Figure 10: ^H-NMR (CDCl^) Fucosteryl Acetate ...... 115

Figure 11: -^H-NMR (CDCI3 ) 24-Ketocholesteryl Acetate . . 116

Figure 12: 1 H-NMR (CDCI3 ) 24-Ketocholesterol ...... 117

Figure 13: ^-H-NMR (CDCI3 ) 24-Ethynyl-24-hydroxv-

cholesterol ...... 118

xv Figure 14: ■^H-NMR (CDCI3 ) 24-Hydroxy-24-vinyl-

cholesterol (saringosterol) ......

Figure 15: ■^H-NMR (CDCl^) Fatty Acid Mixture . .

Appendix D

Figure 1: IR (CDCI3 ) 24-Ethynyl-24-hydroxy-

cholesterol ......

Figure 2: IR (KBR) 24-Hydroxy-24-vinylcholesterol

(saringosterol) ...... LIST OF TABLES

Part 2

Table I: Summary of S^2' Reaction Types ...... 133

Table II: GC Data for Piperidinolysis...... 145

Table III: Kinetic Experiment ...... 160

Table IV: GC Data for Propanethiolate

Nucleophile...... 166

xvii LIST OF FIGURES

Part 2

Introduction, Results and Discussion

Figure 1: Concerted Mechanism for the S^2 1

Reaction...... 126

Figure 2: Ion Pair Mechanism for the S^2'

Reaction...... 126

Figure 3: S^2' Reactions...... 128

Figure 4: Possible Transition State Interactions

for the S^2 1 R e a c t i o n ...... 136

Figure 5: Synthesis of Substrate 1...... 140

Figure 6 : Gas Chromatogram of the Piperidinolysis

Reaction Products ...... 144

Figure 7: Gas Chromatogram of the Piperidinolysis

Hydrogenated Reaction Products...... 147

Figure 8 : Gas Chromatogram of the Coinjection of

the Saturated and Unsaturated

Piperidinolysis Reaction Products . . . 148

Figure 9: Gas Chromatogram of the S^2 Products

12 and 1 3 ...... 151

Figure 10: Gas Chromatogram of the Coinjection of

S^2 Products 12^ and 13^ and the S^2'

Products...... 152

xviii Figure 11: Gas Chromatogram of the Alternate

Synthesis 14 and 1J5...... 153

Figure 12: Gas Chromatogram of the Synthetic

12, 13, 14 and 1 5...... 155

Figure 13: Gas Chromatogram of the Coinjection of

the Alternate Synthesis 14- and 15_ and

the S^2' Hydrogenated Products ...... 156

Figure 14: Gas Chromatogram of the S^2 1 Reaction

of Sodium Propanethiolate and

Substrate (1)...... 162

Appendix A

Figure 1: ^H-NMR (CDCip 4-t-Butylcyclohexanone

(3)...... 189

Figure 2: 1 H-NMR (CDC13) 4-t-Butyl-2-bromo-

cyclohexanone Ethylene Ketal (4) . . . . 190

Figure 3: "^H-NMR (CDCl^) 4-_t-Butyl-2-cyclo-

hexen-l-one Ethylene Ketal (5) ...... 191

Figure 4: 1 H-NMR (CDC13) 4-t-Butyl-2-cyclo-

hexen-l-one (6 ) ...... 192

Figure 5: ^H-NMR (CDC13) 4-t-Butyl-2-cyclo-

hexen-l-ol (7)(87% trans)...... 193

Figure 6 : ■'"H-NMR (CDC13> 4-£-Butylcyclohexanol

from (7) ...... 194 Figure 7: ^H-NMR (CDCl^) 4-t:-Butyl-2-cyclo-

hexen-l-yl 2,2,2 Trichloro-

ethanimidate (?5)...... 195

Figure 8 : 1 H-NMR (CDC13) 2,2,2-Trichloro-N-

(6-t>Butyl-2-cyclohexen-l-yl)

acetamide (9)...... 196

Figure 9: ^"H-NMR (CDCl^) N - (trans-6 -t-Butyl-

2-cyclohexen-l-yl)amine (10) ...... 197

Figure 10: ^"H-NMR (benzene-dg) N- (trans-2-t-

Butylcyclohexyl) amine...... 198

Figure 11: XH-NMR (CDC13) N-(trans-t-Butyl-2-

cyclohexen-l-yl)-N, IT—dime thy lamine

( 1 1 ) ...... 199

Figure 12: ^H-NMR (acetone-dg) (trans-6 -t-Butyl-

2 -cyclohexen-l-yl)trimethylammonium

Tetrafluoroborate (1)...... 200

Figure 13: ^H-NMR (acetone-d^) (260 MHz) (trans-

6-t-Butyl-2-cyclohexen-l-yl)trimethyl-

ammonium Tetraf luoroborate ( 1 ) ...... 201

Figure 14: 1 H-NMR (CDC13) N-(cis-4-t-Butyl-2-

cyclohexen-l-yl)piperidine (13)

(Sn 2 * ) ...... 202

Figure 15: ^H-NMR (CDC13) 4-_t-Butyl-2-eyclohexen-

1-yl 2, 6 -o-Dichlorobenzoate (20) .... 203

xx Figure 16: ^"H-NMR (CDCl^) N- (cis and trans-

4-t>Butyl-2-cyclohexen-l-yl)piperidine

(13, 12) (Sn 2 ) ...... 204

Figure 17: ^H-NMR (CDCl^) N-(cis and trans-

4-t-Butylcyclohexyl)piperidine

(14, 1 5 ) ...... 205

Figure 18: 1 H-NMR (CDC13) N- (4-t-Butyl-l-cyclo-

hexen-l-yl)piperidine...... 206

Figure 19: 1 H-NMR (CDC13) N-(cis and trans-4-

_t-Butylcyclohexyl) piperidine (14, 15). . 207

Figure 20: ^H-NMR (CDC13) N-(trans-6 -t-Butyl-2-

cyclohexen-l-yl)-N,N-dimethylamine

(11) from Piperidinolysis Experiment . . 208

Figure 21: 1 H-NMR (CDCip 1-(Propylthio)-4-t-

Butyl-2-cyclohexene (92% trans)

(24, 2 5 ) ...... 209

Figure 22: ^H-NMR (CDC13) 1-Propylthio)-4-t-

Butyl-2-cyclohexene (87% cis)(S^2) . . . 210

Figure 23: ^H-NMR (CDC13) N-(trans-6 -t-Buty1-

2 -cyclohexen-l-yl)-N,N-dimethylamine

(11) from the' Thiolate S^2 1 Experiment. 211

Figure 24: ^H-NMR (CDC13) Piperidinium Tetrafluoro

borate 2 1 2

xx i Appendix B

Figure 1: ^3 C-NMR (acetone dg) (trans-6 -t-Butyl-

2 -cyclohexen-l-yl)trimethyl ammonium

Tetrafluoroborate (1)...... 214

Figure 2: 3 3 C-NMR (CDCl^) N-(cis-4-t-Butyl-

2-cyclohexen-l-yl)piperidine (13)

(Sn 2 ' ) ...... 215

Figure 3: 1 3 C-NMR (CDC13) N-(cis-4-t-Butyl-2-

cyclohexen-l-yl)piperidine (13)

(SN 2 ) ...... 216

Figure 4: ^3 C-NMR (CDCl^) N-(cis and trans-4-

_t-Butylcyclohexyl)piperidine (14, 15). . 217

Figure 5: 1 3 C-NMR (CDC13) 1-(Propylthio)-4-t-

Butyl-2-cyclohexene (24, 25) (92%

trans) (S^2 ')...... 218

Figure 6 : 1 3 C-NMR (CDC13) 1-(Propylthio)-4-t-

Butyl-2-cyclohexene (24, 25) (87%

cis) (Sn 2 ) ...... 219

Appendix C

Figure 1: IR (Neat) 4-t-Butyl-2 cyclohexen-l-yl

2, 2, 2-Trichloroethanimidate (.8 ) ...... 221

Figure 2: IR (KBr) 2,2,2-Trichloro-N-(6 -t-Butyl-

2 -cyclohexen-l-yl)acetamide (.9)...... 2 2 2

xxii Figure 3: IR (Neat) N-(trans-6-t-Butyl-2-cyclo-

hexen-l-yl)amine (10)...... 223

Figure 4: IR (Neat) N-(trans-6-t.-Butyl-2-cyclo-

hexen-l-yl)-N,N-dimethylamine (11) . . • 224

Figure 5: IR (Neat) 1-(Propylthio)-4-1:-Butyl-2-

cyclohexene (24, 25) (9270 trans)

(Sn 2 ' ) ...... 225

Figure 6 : IR (KBr) (trans-6-t-Butyl-2-cyclo-

hexen-l-yl)trimethylammonium Tetra-

fluoroborate (_1 ) ...... 226

Figure 7: IR (Neat) N-(cis and trans-4-t-Butyl-

2 -cyclohexen-l-yl)piperidine (13., 1 2 )

(Sn 2 ) ...... 227

Figure 8 : IR (Neat) N-(cis and trans-4-t-Butyl-

cyclohexyl)piperidine (14, 15) ...... 228

xxiii Part 1: Some Chemical Constituents of the Brown Alga,

Agarum cribosum INTRODUCTION

The field of marine natural products chemistry, with an

emphasis on the search for drugs from the sea, has experienced

tremendous growth in the past fifteen years. The research has proved interesting in both its chemical and pharmaco

logical aspects. From the medicinal viewpoint, plant and

animal marine organisms have yielded numerous compounds dis

playing high degrees of antibacterial and antitumor activity. 1 2 Reviews by Ruggieri and by Faulkner et: a!L. describe the

various pharmacological uses for which they have been employed.

The potential for the isolation of anticancer drugs is under- 3 lined by the survey of Weinheimer and Kams which identified

8.9 percent of greater than 1500 marine animal extracts as

showing significant anticancer activity. For chemists,

these same compounds are noteworthy for their unique struc

tures and structural substituents. Faulkner etal. suggest

that the reason for this diversity may lie in the higher

pressures and lower temperatures of the marine environment.

These factors could cause changes in biochemical rates and

might stabilize normally unstable compounds in marine

organisms. Scheuer's^ book Chemistry of Marine Natural

Products provides a review of the field through 1972, and

the prolific number of articles published since then

emphasizes the productivity of marine natural products

research.

2 3

The macroalgae, belonging to the three phyla-Rhodophyta

(red algae), Phaeophyta (brown algae), and Chlorophyta (green algae) are a group of marine organisms which have shown extensive biological activity. Hard reported that antibiotic active substances are secreted by algae in 1917.5 Since then numerous studies have been undertaken and the antibacterial

fi activity of many species is known. The microorganism screenings: used either specimens of the whole plant or plant extracts. These data are used to target species which might contain specific compounds having pharmacological activity.

The driving force behind this research is the possibility of isolating new drugs of drug prototypes.

When specific substances were isolated from seaweeds and their biological activity confirmed, it was evident that these compounds were not primary metabolites; i.e., they were not involved in the essential metabolism and nutrition of the plants. Seiburth^7 suggested that the function of these secondary metabolites was ecological in nature. The biosynthesis of the compounds evolved as a chemical defense against predators and pathogens in the plants' environment.

The isolation of many new compounds in recent years is con firming this hypothesis.

One area of research where novel compounds have been isolated consistently is the study of red algae. Early studies of seaweeds concentrated upon identifying and quantifying the primary metabolites such as fatty acids, 4

sugars and pigments. This work is presented in Algal Bio chemistry and Physiology.^ Work since 1965 has excited great interest. The discovery of new classes of halogenated compounds has changed the emphasis of research and stimulated a new era in marine natural products chemistry. The review article by F e n i c a l ^ details the history and chemistry of these compounds, and only a broad outline will be presented here.

The presence of bromine in red algae was first reported in 1926, and in the early 1950s the isolation of several bromophenols was reported. Since then, many new bromophenols from a variety of red algae have been reported, of which lanosol is representative (Figure la). A more recent report suggests that many of the reported structures are actually artifacts of the extraction procedures and that, in vivo, the bromophenols occur as sulfate esters (Figure lb).l®

Figure 1

Bromophenols of Red Algae

Br Br

HO Br HO Br OH

a b 5

Often, one species may contain several bromophenols and

different species may contain the same compound.

The bromophenols were only the first of several classes

of compounds found in red algae which incorporate halogen.

The genus Laurencia synthesizes at least nine different

sesquiterpene and diterpene carbon skeletons (Figure 2).

While some of these carbon skeletons are found in terestrial

plants, spirolaurenone, opposital and iriediol are unique.

Of the sesquiterpenes, the bromochamigrene structures are

the most common and are found in a number of species. Two

intermediates, y-bisabolene^--*- and bromomonocyclo famesol.^^

have been postulated as biosynthetic precursors of these natural products. The presence of caespitol is evidence for

the former intermediate while synderol would be synthesized

from the latter. With one exception, all of the halogenated

sesquiterpenes and diterpenes have been found in species of

the genus Laurenciae (Bromoditerpene alcohols have been

isolated from Sphaerococcus coronopifolius ). Halogenated monoterpenes have been found in the genera Desmia and Plocamium.

Nonterpenoid halogenated aliphatic terpenes have been

isolated from red algae. Some Laurencia species synthesize

Cq5 compounds containing a conjugated pentenyne residue, of which rhodophytin (Figure 2) is an example. The volatile

components of Bonnemaisonia hamifera and Asparagopsis taxi-

formis contain a variety of C 3 , and Cy halogenated ketones,

haloforms and dihalomethanes. Recently halogenated acetic

and acrylic acids have been isolated as well.-^ Figure 2. Halogenated Terpenes in Laurencia and Rhodophvtin

,Br Brt Br Br Br OH nidificene caespitol L. nidifica L. caespitosa furocaespitane L. caespitosa

Br Br OH sp irolaurenone snyderol OH ^ L . glandulifera L . snyderae oppositol aplysin L. supooposita L . okamuri Br

OH OH

Br OH Br HO Br iriediol concinndiol Chondria oppositiclada Laurencia sp L. concinna 7

The research on red algae continues to be productive. 1 fi A 1972 review of bromo compounds listed only sixteen. Yet by 1975, Fenical was able to report the structures of seventy- five compounds. In the past twenty-one months, over twenty articles on red algae have appeared in Phytochemistry and

Tetrahedron Letters, the major journals for the chemistry of marine natural products.

By contrast, research on the brown algae has not been as productive. With a few exceptions, the research has dealt with the identification and quantification of primary meta- bolities. The research which goes beyond these studies describes novel work in four areas - sex attractants, C-^. hydrocarbons and sulfur compounds, phenolic compounds and terpenes. 17 In 1854, Thuret postulated that there might be a chemical attraction between spermatozoids and eggs in the

Fucales from his observation of sexual reproduction in sea- 18 weeds. Almost a century later, Cook, Elvidge andHeilbron made the observation that during reproduction in Fucus vesiculosus and F. serratus the motile spermatozoa are attracted to the mature eggs by a chemical substance secreted by the eggs. The substance was characterized as volatile and could be collected in cold traps by passing an inert gas through the solution. Hlubucek et_ al. identified the attractant in F. vesiculosus as n-hexane. Muller and 20 Jaenicke identified the attractant in F. serratus as a 8

1,3,5-octatriene and called the compound fucoserratene. The stereochemistry of fucoserratene was later determined to be the 1„3-trans,5-cis-octatriene.^

Figure 3

Gamones of Brown Algae

Ectocarpus siliculosus Cutleria multifida

ectocarpene aucantene

multifidene

Fucus serratus

\ F. vesiculosus

fucoserratene C H3C H2C H2CH2C HjC h 3 9

The mal'e'isex attractants of two other species' have also been identified. Ectocarpus siliculosus produces ectocarpene

(Figure 3), a hydrocarbon with a seven-merabered ring.^

Three C-q hydrocarbons are excreted by the female gametes of Cutleria multifida, aucantene, multifidene (Figure 3)

2 2 and small amounts of ectocarpene. Multifidene was deter mined to be the male sex attractant while aucantene and ecto carpene were found to be inactive.

The biogenesis of fucoserratene, ectocarpene,multifidene

0 / and aucantene appears to be related. Jaenicke proposed that the C-^g. g fatty acid (18 carbons with three double bonds) linolenic acid, is the precursor (Figure 4). Beta oxidation and oxidative decarboxylation of this acid yield a C^]_ triene-ol intermediate from which the sex attractants can be formed.

A series of C-q hydrocarbons was isolated from the es sential oils of both Dictyopteris australis and D. plagioramma^-*

(Figure 5). It is interesting to note that dictyopterene D' is identical to the sex attractant ectocarpene, but in

Dictyopteris it is not related to reproduction. These two species also contain sulfur compounds related to the hydrocarbons (Figure 6).^,27 Moore^S has proposed biosynthetic pathways for the formation of all these C-j^ derivatives.

Both dictyopterene A and B undergo a thermal Cope re arrangement to yield the enantiomers of dictyopterene C' and Figure 4. Biogenesis of Gamones in Brown Algae. OH

OH OH

OH OPP ©x©y ©

/© i © Figure 5. C-q Hydrocarbons ojf Dictyopteris

dictyopterene D

polyenes 12

Figure 6: Sulfur Compounds of Dictyopteris vw\AM0 0

0 0

0 v w ^ y w w n- a,3.v OAc 0 s— s' / W = v W \

/ v r w / ^ y u w OAc

and D 1. Therefore, in vivo , thermal sigmatropic rearrangements

cannot account for the stereochemistry of dictyopterene C 1 and

D 1. The seaweed is fully exposed to sunlight in its habitat. 29 Pickenhagen has shown that, in vitro, the trans cyclopro- panes can be isomerized to a cis cyclopropane which can then undergo either thermal or photochemical sigmatropic rearrange ment to give the stereochemical product found in nature.

The study of tannins in brown algae is one area which has received considerable attention recently. Many brown algae produce phenolic substances which may be involved in their chemical defense. Crato first suggested the presence of phloroglucinol (1 ,3,5-hydroxybenzene) in brown algae in 30 1892. He based his theory upon observations that there 13 were substances present in intracellular bodies of the plant

(physodes) which reacted with vanillin HC1. In 1938 Kalle reported that seawater contained yellow substances termed 31 "gelbstoff” which absorbed UV light. Craigie and McLachlan showed that the gelbstoff was secreted into the surrounding medium by several species of brown algae (species of red and green algae gave negative results). Their observations led them to conclude that the release of the substances was not closely related to metabolic processes and that the gelbstoff was initially secreted colorless, turning color over a period of time. The material was isolated and gave phloroglucinol upon treatment with base. It was not until

1973 that phloroglucinol was first successfully isolated by

Glombitza^O from an extract of Cystoseira tamariscifolia.

Glombitza also examined extracts from twenty-six other species of brown algae and identified phloroglucinol in seventeen of these species.

In further studies, Glombitza isolated and characterized eleven phloroglucinol derivatives with four basic structures

(Figure 7). Bifuhalol (Bifurcaria bifurcata,^2 Cystoseira tamariscifolia^ ) , trifuhalol (Halvdris siliquosa,3 4 35 B. bifurcata ), tetrafuhalol, pentafuhalol and heptafuhalol O C (B. bifurcata-3-') are polyhydroxyphenylethers consisting of one pyrogallol residue and one to six phloroglucinol deriva tive. Diphlorethol (C. tamariscifolia,^3 g. bifurcata,

Laminaria ochroleuca-^) and triphlorethol (L. ochroleuca^ ) are also polyhydroxyphenylethers composed only of phloroglucinol Figure 7. Phloroglucinol Derivatives of Brown Algae.

HO HO OH -OH HO OH HO -OH HO OH HO OH HO, HO OH

OH HO OH HO tetrafucol b tetrafucol a HO.

OH HO OH HO HO HO HO' OH HO OH HO

HO HO HO HO OH triphlorethol trifucol difucol OH OH

HO OH OH ‘OH diphlorethol trifuhalol bifuhalol 15

O C residues. The four fucoles (B. bifurcata, Fucus vesiculosus^?) consist of phloroglucinol residues coupled through the ring.

In AscophyTlum nodosum, the phenolic precursor to these compounds was identified by Ragan and Jensen^S as 1,2, 3 , 5- tetrahydroxybenzene-2,5-disulfate ester. This compound was detected in a total of eight species of brown aglae and a monosulfated ester was also present in fourteen species.

Phloroglucinol and its derivatives may act as environ mental defense agents in brown algae similar to the role played by halogenated terpenes in red algae. Conover and

Seiburth39 reported a "marked animal toxicity and supression in tidepools" in which Ralfsia verrucosa grows. They found this water toxic; and it affected trochopores, veligers and nereid worms. The toxicity was due to the presence of gelbstoff. Seiburth and Jensen^ postulated that brown algae exude phenolic precursors which in seawater react with carbohydrates and proteinaceous material. Extension of their results to the marine environment showed that in a littoral well-populated with Fucus vesiculosus, one gram of phenol per square meter was produced per day. The polyphenols which formed over time were toxic to plarce larvae. From these studies, it is not unreasonable to suggest that forma tion of phloroglucinol and its derivatives in algae evolved as a defense mechanism for maintaining their ecological niche.

Another class of phenolic compounds has also been dis covered. These compounds (Figure 8 ) are also probably 16

defensive in origin since some are highly toxic to fish.

Stypotriol and stypoldione are two fish toxins isolated from the tropical brown alga, Stypopodium zonale.^ Closely related

/ 0 is taondiol, isolated from Taonia atomara. Possible bio synthetic precursors to taondiol are found in Sargassum tortile^ and Cystophora torulosa.^ In these compounds the ring system is only partially closed.

In only one case have brominated phenols been reported.

Pedersen and Fries^~* reported the isolation from Fucus vesiculosus of 2,3-dibromo-4,5-dihydroxybenzyl alcohol, 2,3- dibromo-4,5— dihydroxybenzylethylether, and 3 ,5 -dibromo-]3 - hydroxybenzyl alcohol. These results are open to question in light of Glombitza's work on phenols in brown algae and

0 7 in particular his work on Fucus.

There are few examples of true terpenes isolated from brown algae in the literature. The terpene producing species, with one exception, belong to the order Dictyotales (Figure 9).

Several novel diterpene alcohols containing a hydroazulene skeleton were isolated from Dictyota dichotoma^ »^ A Q (dictyol A, B and E), Pachydictyon coriaceum (pachydictyol

A) and Glossophora galapagensis.^ The sesquiterpene zonarene and several related aromatic compounds were isolated from Dictyopteris zonaroides . It is interesting to note that the enantiomer of chromazonarol was shown to occur in the sponge, Disidea pallescens. From Cystoseira crinita in the order Fucales, two novel linear terpenoids were 52 isolated, oxocrinitol and crinitol. Figure 8. Other Phenolic Compounds of Brown Algae

HO .iu> HO O taondiol stypoldione

HO HO OH stypotriol

OH OH

Sargassum tortile S . tortile

CH3)

OCH. CH, CorH)

Cystophora torulosa C. torulosa

-vl 18

Figure 9: Terpenes in Brown Algae

OH crinitol

(AcO)HO

Glossophora galapagensis G. galapagensis

HO pachydictyol A dictyol A

HO OH

CH CH. .CH

zonarol chromazonaro1 zonarene 19

The growth in research on brown algae has not kept pace with red algae. Attention has been directed at the red

algae because of the excitement over halogenated compounds.

In contrast, only the bromophenols of Fucus vesiculosus have

been reported in brown algae. Yet the biosynthetic pathway

for the production of terpenes does exist in some species.

The presence of volatile, nonpolar brominated compounds was

demonstrated by Lunde53 in Ascophyllum nodosum and Laminaria

hyperborea. These compounds were extracted into cyclohexane

for analysis, excluding the possibility that they were phenols.

Most of the novel compounds reported in brown algae have

been isolated from tropical species . With the exception of

the phloroglucinol research, there has been very little work

reported on the northern cold water species. Based upon

Lunde's results, we felt that there was a reasonable pos

sibility that comparable halogenated hydrocarbons might: also

be present in species native to New Hampshire. The theory

that secondary metabolites are synthesized by algae as a

means of chemical defense provided the criteria for choosing

a species. Environmental and physical observations led to

the choice of Agarum cribosum (order: Laminariales) for

s tudy. This algae is of special interest to phycologists

because it is often found in areas otherwise closely cropped

by marine animals. The plant itself is usually free of

macroscopic epiphytes except for the hold fasts. Microscopic 20

examination revealed an absence of diatoms on the surface.

These factors led us to believe that the plant might contain defensive secondary metabolites. A further factor in favor of this species was that it belonged to the same taxonomic family as Laminaria hyperborea, a species which Lunde had found to contain non-polar brominated organic compounds.

Experiments were designed to look for novel non-polar halo genated compounds in Agarum cribosum. RESULTS AND DISCUSSION

In October 1976, Agarum cribosum was collected in the drift after a major storm at Rye Beach, New Hampshire. The seaweed was air-dried, milled and 4.5 kg was extracted with ether to yield 33 g of a green-brown oil. On a silica gel open column, 1 1 g of the extract was fractionated into ca.

200 mL portions by continuous elution with hexane, hexane/ benzene, benzene, benzene/ether, ether, ether/methanol and methanol. The forty-eight fractions obtained from elution with the solvents hexane through ether were examined and like fractions were combined. ^H-NMR analysis was used to determine the types of compounds present in each fraction.

The results of the chromatography experiment are summarized in Table 1.

The open column chromatography experiment was designed to discover and separate terpenoid compounds from the crude algal extract and was modelled upon those methods used suc-

54 >55 cessfully by Fenical and by Waraszkiewicz. In their chromatography experiments, halogenated terpenes were eluted in the benzene and benzene/ether fractions. When halogenated terpenes have been isolated from red seaweeds, they have constituted a considerable portion of the crude algal extract.

Some examples are B-snyderol^ which was isolated as 25 per cent of a Laurencia snyderae chloroform extract, chondrocoles

21 22

TABLE I, RESULTS OF OPEN COLUMN CHROMATOGRAPHY EXPERIMENT

Fraction No. Appearance Amount ^-H-NMR Analysis

4-19 insignificant

20-27 orange oil 0 . 0 2 g triglycerides

28-29 green solid 2.34 g chlorophyll triglycerides

30-31 green oil 1.94 g sterol esters* triglycerides

32-35 green oil 1.32 g triglycerides

36 brown crystal triglycerides line solid (highly unsatu rated "drying oils")

37-48 orange oil 0.90 g trigylcerides

*Fraction No. 30-31 (1.74 g) was saponified to yield Q.438 g of free sterols. 2.3

A"^ which was isolated as 25 percent of a Chondrococcus c 5 homemanni ether extract, and laurinterolJ which was isolated as 36 percent of a L. nidifica ether extract. If significant amounts of terpenes were present in Agarum cribosum, then

1 1 g of extract should contain detectable quantities in isolatable amounts.

Individual nonpolar components were eluted from the column in narrow bands. During the fractionation, the two non-fatty acid components, chlorophyll and the sterol esters, appeared in only two consecutive fractions. Free sterols were present as only 4 percent of the total extract. Yet the sterol ^H-NMR angular methyl signals were identifiable in the spectrum of Fraction No. 30-31 even though the sterols were bound to fatty acids. These fatty acids along with pigments represented 75 percent of this fraction. Assuming their presence, even small amounts of terpenes should have been eluted in a discreet band and their isoprene methyl signals would have been discerned by ^H-NMR. From this chromatography experiment, it was concluded that Agarum cribosum does not contain halogenated terpenes or other special nonpolar metabolites in amounts comparable to the quantities reported in the literature. 24

Since all of the fractions contained small amounts of other materials, it was necessary to remove the triglycerides in order to further examine the other lipid components. For this purpose, a base saponification of the crude extract was performed. The saponification also allowed a more complete examination of the sterol fraction.

Crude extract (16 g) was saponified in 10 percent KOH/ methanol at 45° for 4 hours. Upon workup, the non-saponifiable lipids were recovered as ared oil (2.7 g). The oil was sepa rated using preparative High Performance Liquid Chromatography and fractions were cut guided by the refractive index detector.

Fractions with similar ^H-NMR spectra were combined into four fractions.

Fraction 1 (.32 g) consisted of one major and two minor compounds (by TLC) and was not identified. Fraction 2, a clear oil (.56 g), was identified by ^H-NMR comparison as phytol,^ a cleavage by-product of chlorophyll. The third and fourth fractions (1.08 g and .019 g, respectively) were identified as sterols. The ^H-NMR spectrum of Fraction 3 had a quartet at 5.28 ppm( 6 ) which is characteristic of the

24-ethylidene substituent in fucosterol, the major sterol of brown algae. The ■*'H-NMR spectrum of Fraction 4 differed significantly from Fraction 3 in the region of 5-6 ppm(S).

Analysis of the two sterol fractions was obtained by glass capillary GC/MS using the facilities of Dr. Robert

Gagosian at Woods Hole Oceanographic Institution. Each 25

sterol fraction was acetylated and the sterol acetates iden tified by mass spectrometry.

Fraction 3 contained fucosterol and 24-methylenecholesterol. Fraction 4 was a complex mixture whose GC trace is presented in Figure 10. The mass spectra of the seven sterol acetates are given in Figures 11-17. Table XI presents the molecular ions of the sterol acetates and their percentages of the total sterol composition. The chemical structures of the sterols are given in Figure 18. Table III gives the major MS fragmentation ions above m/e 253.

Identification of each sterol acetate except for 24- methylenecholesterol was made by a comparison of its mass spectrum with that of an authentic sample. Cholesterol was purchased from Fisher and desmosterol was purchased from

Applied Science. Fucosterol was isolated from Ascophyllum nodosum and purified as the acetate (mp 118-119.5°, lit. mp

118-119°). 24-Ketocholesterol acetate (mp 128.5-130°, lit. mp 127-128°) was synthesized by the Ozonolysis of fucosteryl acetate. Saringosterol (mp 158-160.5°, lit. mp 160-161.5°) was synthesized by the addition of potassium acetylide to 24-ketocholesterol followed by partial reduc tion of the triple bond with Lindlar's catalyst.

The identify of 24-methylenecholesterol was confirmed by its mass spectrum which was in agreement with the pub lished s p e c t r u m . 59 yhe ^h-NMR of Fraction 3 contained a peak at 4.7 ppm (6 ) corresponding to the terminal methylene m

Cholestane

T7TT

0 100 200 300 400 500

Figure 10: Gas Chromatograph Trace of Sterol Acetates (Fraction 4) from GC/MS Analysis TABLE IL. STEROL COMPOSITION OF AGARUM CRIBOSUM

% of Total Sterols Sterol Identity M+ -60 (Combined Fractions 3 and 4)

I cholest-5-en-33-ol 368 <0.05 (cholesterol)

II cholesta-5,24-dien-3$-ol 366 <0.05 (desmosterol)

III 24-methylenecholest-5-en-3$-ol 380 10.80 (24-methylenecholesterol)

IV stigmasta-5,(E)-24(28)-dien-33-ol 394 88.70 (fucosterol)

V 38-hydroxy-cholest-5-en-24-one 382 <0.05 (24-ketocholesterol)

VI 24£-stigmasta-5,28-dien-38,24-diol 410 0.40 (saringosterol)

VII unknown 410 <0.05 28

TABLE III. -STDE CHAIN FRAGMENTATIONS OF STEROL ACETATES

Sterol Acetate of m/e cholesterol 368, 353, 255 desmosterol 366, 351, 281, 282, 283, 253

24-methylenecholesterol 380, 365, 296, 281, 253 fucosterol 394, 379, 296, 297, 281, 253

24-ketocholesterol 382, 367, 296, 281, 255 saringosterol 410, 392, 377, 367, 349, 296,

281, 282, 283, 253 unknown 410, 395, 255 29

l H4 ACAKUI1 CRlBOitiri PR.* '2 STEROL EXTRACT ACETYl-ATED FH4 •

0050 B075 Q1G0 0125 0153 0175 0200 0225 0250 0275 030E

i HI M'-V'i'Un Cl :j.EOBUM F R . 9 'Z STEROL EXTRACT ACETYLATED FH4 ■I'M i'K'll HUMBhR 0215 - 0200

0325 0330 0375 0400 0425 045fe

Figure 11: Mass Spectrum of GC Peak I (Cholesteryl Acetate) 30

F'n ALARUM CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED FH4 SPECTRUM HUMBER 0231 - 0240

uj l| i» » »*»| • i i n'r'i' i*i* i ‘T'j’i11 * i * i * j11 'T M 0040 0065 0030 0115 0140 0165 0130 0215 0240 0265 0236

FH4 ALARUM CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED FH4 SPECTRUM NUMBER 0231 - 0240

J Jjjij ft t/ 0200 0225 0250 0275 0o00 0325/-i^ncr 0350rt- ? c m r*~?~7f=0375 (0400 l A c m 0425r»A o c c 0456*a irr

Figure 12: Mass Spectrum of GC Peak II (Desmosteryl Acetate) 31

>114 AGARUM CRIBOSUM FR.9-'2 STEROL EXTRACT ACETYLATED FH4 il-'ECTRUM HUMBER 0255 - 0270

0040 0O65 0099 3115 0140 0165 0190 0215 0240 0265 0296

FH4 AGARUM CRIBOSUM FR.3-'2 STEROL EXTRACT ACETYLATED FH4 T-ECTRUM HUMBER 0255 - 0270

'{'E y 1^1 T |a nr^-l^ ,fa

■ 0200 0225 0250 0275 1300 0325 0350 0375 0400 0425 0456

Figure 13: Mass Spectrum of GC Peak III (24-Methylene- cholesteryl Acetate) 32

FH4 AGARUM CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED FH4 Gl’LCTEUII t UMBER 0306 - 0318

0040 0065 0090 0115 0140 0155 0190 0215 0240 0265 0296

Ri 14 AGARUI1 CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED FH4 •_! hClFUH NUMBER 0306 - 0313

l,1 I J fr t p |‘i j r p 1 1*11 r p T 1 A | ' 1 P T,|,,l"rTTT i I 1 I * 1 lT r | _ _ _ - . . m J < l->A 1 -1 r“ LT. ' 0250 8275 0300 0325 0350 0375 0430 0425 0450 0475 0536

Figure 14: Mass Spectrum of GC Peak IV (Fucosteryl Acetate) 33

fH4 AGAMJI-1 CRIBOSUM FF.9-2 STEROL EXTRACT ACETYLATED FH4 '.r'L.i. IRUM NUMBER 0324 - 0318

y m I ...... M W l 1 0030 0075 O100 S125 01’50 0175 0200 0225 0250 0275 0306

MIT i-tCAkUM CRIBOSUM FR.9 '2 STEROL EXTRACT ACETYLATED FH4 CI LCIKUH NUMBER 0324 - 0318

1 O250 0275 O300 0325 0353 0375 0400 3425 0450 0475 050C Figure 15: Mass Spectrum of GC Peak V (24-Ketocholesteryl Acetate) 34

! i> 1 AG,'.RUM CEICOSUM FR.9 2 STEROL EXTRACT ACETYLATED FH4 ij.un humour: 0431 - 042a

0O4a W&65 0030 61 15 0146 0165 0100 G215 0240 0265 0296

I 111 iTUiRUM CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED I'Ei IRUfl HUMBER 0431 -- 0420

W- 0240 0265 0290 031 0340 0365 0330 0415 0440 G465 0496 Figure 16: Mass Spectrum of GC Peak VI (Saringosteryl Acetate) 35

rn? riGrtRLiri CRTBOSUH FR.9.'2 STEROL EXTRACT ACETYLATED FH4 I.’-- I l -WI I I I'JI r j "• V" T'JW

0275 O300 G325 0350 0375 0400 0425 0450 0475 0506

MI-1 h u h RUII CRIBOSUM FR.9'2 STEROL EXTRACT ACETYLATED FH4 ..rU-TI oil HUMBER 0479 - 9450

0940 9065 9090 0115 0165 9190 0215 82400140

Figure 17: Mass Spectrum of GC Peak VII (unknown) Figure 18. Sterols of Agarum cribosum.

cholesterol desmo sterol 24-methylenecholesterol

fucosterol 24-ketocholesterol saringosterol 37

protons at C 2 4 / This sterol is the second most abundant sterol in the total mixture which is characteristic of 24- methylenecholesterol in brown algae.

The mass spectral fragmentation patterns of sterols are very complex; however, they provide substantial structural information, particularly about side chain substitution.

Certain fragmentations are common in all of the A5 sterol esters. The lower molecular weight fragments (>253) are dominated by a repeating pattern of n- 2 , n, n+ 2 twelve to fourteen units apart from m/e 43-201. This series arises from fragmentations in the ring.

A distinguishing fragment at m/e 213 is the result of the cleavage of the D ring. Loss of the side chain gives a fragment at m/e 255 or m/d 253 and a common fragmentation is seen at M+-60-15 from the cleavage of an angular methyl.

The important side chain cleavages for each sterol acetate of Agarum cribosum are discussed below.

The cholesterol spectrum is very simple. Its spectrum ean be rationalized from the above discussion. Desmosterol is similar but contains fragments at m/e 281, 282, 283 which are attributed to cleavage between C2 Q and C 2 1 , with or with out hydrogen transfer . The spectra of 24-ketocholesterol,

24-methylenecholesterol and fucosterol are distinguished by a strong fragment at m/e 296 due to a McLafferty rearrange ment. A possible mechanism is presented for 24-ketocholesterol. OH + T + m/e 29 6

In contrast to a classical McLafferty rearrangement, the

charge remains with the ring in these sterol acetates. Loss

of an angular methyl from the m/e 296 fragment gives rise to

a peak at m/b 281. In addition, fucosterol has a strong ion

at m/e 297, possibly due to simple allylic cleavage.

The mass spectrum of saringosterol is much more complex,

and an explanation of the fragmentation pattern is presented

in Figure 19. This scheme is rationalized in the following manner. There are three basic fragmentations. Water is

usually eliminated from alcohols via a 1,3 or 1,4 hydrogen

abstraction. In saringosterol this gives rise to the frag ment at m/e 392 and m/e 377 (lT*"-60-18-15) . However, the

triad m/e 281, 282, 283 suggests the presence of a A24 type moiety so a rearrangement is proposed. Cleavage alpha to CH, / 3 -H + CH OH n'c h . OH

A - C a H 7

OH :OH« +

•OH

if H + V ’ ny

the carbinol carbon is a common fragmentation and gives the ion, m/e 367. A rearrangement to initiate the loss of water yields a very stable ion at m/e 349. Another common frag mentation in alcohols is the concerted elimination of water and an olefin giving the fragment at m/e 296.

Sterol VII remains unidentified. The molecular weight of the sterol acetate suggests an isomer of saringosterol; however, the fragmentation pattern was not at all similar to that of saringosterol and there was no evidence for a second hydroxyl group. The concentration of this material was quite small and the mass spectrum contained a high background so that further structural interpretation was not possible.

The sterols of Agarum cribosum were present as 0.05 percent of the dry algal weight. This figure is variable according to the extraction conditions and the length of extraction. The non-saponifiable lipids constituted 17 per cent of the crude extract and the sterols made up 40.7 per cent of the non-saponifiable lipids. These values cor respond well with values for some brown algae reported by

fi 1 Tsuda, Hayatsu, Kishida and Akage, 12.5-24.8 percent and 32.2-56.6 percent, respectively.

The sterols of brown algae were first investigated by fi 9 Heilbron et al. in 1934. From Fucus vesiculosus, they isolated a C2 9 sterol having two double bonds. Melting points of the free sterol and of its acetate did not cor respond to any known sterol and the new sterol was 41

named fucosterol. In 1938 fucosterol was shown to be a

6 3 A5 sterol and in 1950 the second double bond was placed

at A24 (28).64 The side chain double bond was assigned the

E configuration after comparison with model compounds in

a 1968 NMR s t u d y . 65

The Hedlbron study prompted the examination of many

species of brown algae and eventually it was found that the

sterol fraction contained a second minor component, 24-

methylenecholesterol. Investigations in recent years, aided

by better instrumentation, in particular gas chromatography

and mass spectrometry, revealed that the sterol mixture in

brown algae is more complex. The sterols of a number of

brown algae reported in the literature are presented in

Table IV.

The identification of fucosterol and 24-methylenecholes-

terol and their percentage composition of the total sterol

fraction in Agarum cribosum is comparable to many species of

brown algae. The percentage of cholesterol is lower than

the 0.4-8 percent usually reported by Safe, Wong, and

C h a n d l e r . 6 6 it is probable that more cholesterol was present

in HPLC Fraction 3 of the sterol mixture butwa.s not visible

to instrumental detectors because of the large amounts of

fucosterol and 24-methylenecholesterol. The detected

amount of desmosterol may also have been reduced in the same

way. The more polar sterols, 24-ketocholesterol and sarin

gosterol, were more likely to be present in HPLC Fraction 4 Table IV. Sterols of Brown Algae

Percentage of T al Sterols

I II III IV V VI % of 24- 24- Dry Fucos- Methylene- Saring- Choles- Ketochol- Desmos- Species______Weight terol Cholesterol osterol terol esterol terol

LAMINARIALE S

Agarum cribosum 0.05 88.7 10.8 0.40 <0.05 <0.05 <0.05

Laminaria saccharinaa 0 . 2 0 87 11 1.8 0.05 0.05

L . faeroensis^3 74 20 4 1 - - <1

Alaria crassifoliac 0.03 81.5 4 1 1.5

Costaria costatac 0.02 75.5 16.5 3 2.5

FUCALES

Fucus evanescensc »^ 0.16 78.5 13.5 2.5 0.5

(65)d < - ) d (+)d ( - ) d

-P'K5 Table IV (continued) 7> Dry I II III IV V VI Ascophyllum Weight nodosumat 0 . 1 0 90 2 6 0.05 < 1

Sargassum r ingo 1 di anumc 0 . 2 1 91.5 4.5 2 . 1 1.7 -- —

S. confusumc 0.18 90 2 0 . 1 8 -- --

S. thunbergic 0 . 1 0 87 9 trace 4.5 --

S. fluitanse'' 0.14 %50 + + --

Cystophyllum hakodatensec 0.07 8 8 8 0.5 3 — --

Pelvetia wrightiic 0 . 2 1 97 1.5 0 . 1 0.5 --

DICTYOX&LES

Dictyopteris divaricata0 0.24 97 1.5 1

a Reference 6 6 ; b Reference 67; c Reference 6 8 ; d Reference 69; e Reference 70 Other Sterols: t 3,5,(E)-24(28)-Stigmastatrien-7-one , 10.67,; 5 , (E)-24(28)-stigmastadien-36 ,7a-diol, 0.37,. £ Brassicasterol, <170. * trans-22-Dehydrocholesterol; 24-ethylcholesterol, 24-methylcholesterol, 24-methvl-trans- 22-dehydrocholesterol, 24-ethyl-trans-22-dehydrocholesterol: tentative assignment.

(j O 44

and their values probably reflect the actual amounts in the non-saponifiable lipids. The chromatographic methods used in this experiment may preclude the accurate determination of the percentage of some of the sterols. However, preliminary chromatographic separation was essential in order to keep the two major sterols from overwhelming the minor ones.

Saringosterol has been reported in varying amounts.

Knights^ postulated that it is actually an oxidation arti fact formed during the milling and extraction procedures.

24-Ketocholesterol has not been reported frequently and may also be an oxidation product of fucosterol according to

Knights.

Of particular note is the detection of the sterol desmosterol. Although this sterol is common in red algae, until now its presence has not been confirmed in brown algae.

Patterson^? tentatively identified desmosterol in Laminaria digitata and L. faeorensis by GC retention times. Our mass spectral analysis positively confirms this sterol's presence in a brown alga. This confirmation suggests that small amounts of desmosterol are present in other brown algae pre viously examined but that the methods of detection were not sensitive enough. The presence of desmosterol in brown algae might be explained by proposing its intermediacy in the 72 biosynthesis of the C-24 alkylated sterols. However, Goodwin postulated that the and C-j^ demethyl at ions of cycloartenol, the plant sterol precursor, occur after the alkylation at 45

Desmosterol is proposed as the precursor of choles- 7 ^ terol. J A more detailed discussion of sterol biosynthesis

in brown algae is presented below.

The area of plant sterol biosynthesis was reviewed by

Goodwin.^ In the plant kingdom, the cyclization of

squalene-2 ,3-oxide leads to cvcloartenol instead of the

animal sterol precursor lanosterol.

HO' HO

lanosterol cycloartenol

Labeled mevalonate or labeled acetate yields cycloartenol

but not lanosterol, and the cycloartenol is synthesized

directly not via a rearrangement of lanosterol. The path

way leading to the C 2 4 alkylated sterols is presented in

Figure 20. The existence of this pathway is supported

by the isolation of each of these intermediates as natural

products. Mechanisms for the C-j^ and C^a demethylations

are shown in Figures 21 and 22, respectively. The A5

double bond is formed via the sequence A8 -> A7 A 5 , 7 -> Figure 20. Proposed Transformation of Cycloartenol to C2 4 ~Substituted Sterols.

HO HO HO cycloartenol 24-methylenecycloartenol cycloucalenol

HO HO HO obtusifoliol 24-methylenelophenol 24-ethvlidenelophenol Figure 21. Demethylation of Lanosterol at 0 ^4 .

c=o x

Z-^

Figure 22. Demethylation of Cycloartenol at C^.

HO HO

HO COOH

HO 49

A5. The transformation to the A7 occurs with the elimina

tion oj?* the^'3 hydrogen and the A5 bond is formed by the elimination of the 6a hydrogen. The saturation of A7

occurs by trans addition of hydride from NADPH to 7a and a proton to 8$.

A mechanism for the alkylation of the side. Chain

(Figure 23) was postulated by Castle, Blondin and N e s ^

in 1963 and was later confirmed by a series of labeling experiments.^5 It was found that in germinating peas, the methyl group of labeled methionine was incorporated into both the C2 8 and ^29 positions of the side chain of 24- ethylcholesterol and at C 2 8 in 24-methylcholesterol. Labeled ethionine did not incorporate in similar experiments, supporting the two-step alkylation. When Sphacelaria

(Phaeophyta) was cultured in the presence of methyl-^^C- methionine, labeled fucosterol was produced. The mechanism in Figure 23 has gained general acceptance.

H + (C—28)

H (C~28) y Figure 23: Side Chain Alkylation of Cycloartenol 50

Fatty Acids of Agarum cribosum

The fatty acid composition of Agarum cribosum was also determined. First the non-saponifiable lipids were removed from a KOH/methanol saponified extract. The remaining extract was acidified and the organic acidic materials recovered by ether extraction. This crude fatty acid oil was esterified with BC^/methanol. The resulting fatty acid methyl esters (FAME) were subjected to gas chroma tographic analysis for identification and quantitation.

A seven foot x one ^eighth inch column packed with a 12 per cent DECS stationary phase was used. Upon analysis of the

FAME mixture chromatogram (Figure 24), twenty-six peaks were ascertained. Data was collected to determine the peak areas, and the sample was then analyzed byGC/MS at the Massachusetts Institute of Technology National

Institutes of Health mass spectrometer facility to identify the individual components.

The mechanics of data collection at the MIT facility were the following. The sample was gas chromatographed on the same column used in our own laboratories. Once the first GC peak was detected, one mass spectrum was taken every three seconds for the duration of the run. The data were stored in a computer and retrieved on microfilm. For 3E

Figure 24: Fatty Acid Methyl Ester Chromatogram Ln 52

each mass spectral run, the data consisted of the total ionization plot for the GC/MS run, individual mass spectra, and an ionization intensity plot for every mass unit, m/e.

The total ionization plot (Figure 25) is a graph resembling the shape of the gas chromatogram and is a plot of the ionization intensity vs. spectrum numbers. As the amount of material in the ionization chamber increases and decreases when the peaks elute from the gas chromato graph, the overall ionization intensity increases and decreases. The maximum ionization intensity observed for each individual spectrum of the run is then plotted. The graph shows the association of each peak in the gas chromato gram with a particular series of spectra.

A further treatment of the data is the ionization in tensity plots. For each mass unit, m/e, a plot is made of the intensity of that ion in each spectrum vs. the spectrum number. This graph is superimposed over the total ionization plot. For any mass unit, spectra containing that particular molecular fragment can be designated at points where the maxima of the two superimposed plots coincide. In this manner, peak #9 in Figure 24 can be confirmed as FAME

(Figure 26) even though its spectrum (see Appendix B) con tains fragments at both m/e 270 and m/e 268.

A combination of the information provided in the three data sets made it possible to designate most of the peaks in the gas chromatogram as arising from a specific fatty acid 53

LEflEL C- 9 RLN 2 TOTAL IcNI2ATIQN p lo t RLN MD. GO3 7

20 60 100 120 140 160 aeo 200 220 240 260 2 90 300 320 340 360 390 400 SFETTRLW IfOEX fU €E R

LEBEL AC-9. RJ'J 3 TOTAL ICNI2ATICN PUTT RLN W). 6000

SPE l TRUJ I lC a MJJ3ER

Figure 25: Total Ionization Plots #6097 and 7#6098 from GC/MS Analysis

M/E-2GB MRX. TNT. * 1195 RUN NO. * 06097

aoo

Figure 26: Ionization Intensity Plot for C-^.^

i 54

methyl ester. Only seven of the major peaks, less than

7 percent of the total weight,remain unidentified.

The data was collected during two chromatographic injections (see Figure 25). The first injection, series

#6097, gives data through peak #20. In the second injection, series #6098, data was not collected until peak #18 began to elute. This was a larger injection to increase the intensity of the later eluting peaks. In several cases due to background noise, unequivocal molecular ions could not be obtained from these runs. For these fatty acid methyl esters, data was used from a subsequent run, #6106. Although only onq mass spectrum is presented for each FAME, the molecular ion was clearly visible in the spectra preceding and fol lowing the chosen spectrum.

Table V presents the fatty acid methyl esters of Agarum cribosum and their weight percentages of the total mixture.

The percentages are adjusted for the varying FID relative 76 responses. The spectra of these FAME's and the ionization intensity plots of their M+ ions are given in Appendix B.

The GC resolution of the individual FAME's was varied. Some spectra, such as #13-6097 contain only one molecular ion and the fragmentation pattern can be easily distinguished from the background. In other spectra, the components over lapped as they entered the ionization chamber and several molecular ions appear. Spectrum #278-6097 contains the M+ ions for C-^.g (m/e 264), (m/e 262), and C-^g^ (m/e 55

Table V. Fatty Acid Methyl Esters of Agarum cribosum

Peak % Weight Composition .gure 24) FAME M+ of Total FAME Mixture

1-3

4 C1 4 :0 242 4.4 5 -- 0.4

6 C15:0 256 1 . 1 7 -- -- 0.3

8 270 29.5 C 1 6 : 0 9 268 19.0 C 1 6 :1

1 0 C 17 : 0 284 0 . 2

1 1 C16:2 266 1.9

1 2 -- 0.5

13 298 1 . 2 C18 : 0

14 C16 : 3 264 4.1

C 1 8 :1 296 15 C16 : 4 262 5.7

16 C 18:2 294 2 . 1

17 -- 1 . 2

18 C18 : 3 292 1.7 19 326 1.4 C 2 0 : 0

2 0 290 2 . 6 C18:4

2 1 -- 0.4

2 2 C20 : 3 320 0 . 6

23 C20:4 318 9.0 24 -- -- 2.3

25 -- 1.3

26 C22:5 344 9.2 56

294) (Figure 27) . The ionization intensity plots were used to designate which peak in the chromatogram was due to a par ticular FAME.

Overall, interpretation of the data led to unequivocal assignments. The few instances where the peak assignments were tentative was in the case of peaks # 1 0 and # 1 1 , peak

#14, and peak #22. For peaks #10 and #11, the maxima in the ionization intensity plots of m/e 284 and m/e 266 overlap so that the actual peak identities cannot be verified. Peak #10 was assigned to FAME Ci 7 ;Q because its maximum is much less intense and an odd-number fatty acid would not be expected to be present in large amounts. Peak #14 is a combination of FAME Ci6 : 3 anc* ^18-I* i°nizati°n intensity plots overlap completely so no differentiation can be made. Peak

#22 definitely gives an M+ ion at 320; however, the cor responding ionization intensity plots in runs #6097 and

#6098 have too much background for a positive identification.

In beginning the discussion of the fatty acid results, it must be stated that the quantitation of the data is at best an estimation. Not having at our disposal the most efficient GC columns, i.e., capillary or open tubular, complete separation of all the components of the FAME mix ture was not possible. As a result, the peak shape and area of overlapping peaks had to be approximated. These esti mates were based upon the observation that the cleanly separated peaks were Gaussian. In drawing the peak shapes - 270 BIOS MAX * 9955

M/E 264 MAX. INT. * 1245 RUN NO. * 06D97

06097

M/E 294 MAX. INT. * BBO RUN NO. * 06097

Figure 27: Identification of Overlapping FAME's by Use of Ionization Intensity Plots 58

for overlapping peaks, it was assumed that these peaks would also display Guassian behavior. 76 However, according to Ackman, there are other errors in quantitation inherent in the procedure. There can be column losses due to adsorption on the column and also due to oxidation of the higher molecular weight fatty acid methyl esters during elution. Programming can also inter fere with detector response. It is estimated that "for proper operating conditions, there is a probability of an

8 to 20 percent error for a 1 to 3 percent (weight) com ponent , 4 to 8 percent error for the 3 to 10 percent com ponent range, and 2 to 4 percent for the 10 to 30 percent components."

Exact calibration and determination of detector response were not feasible for an experiment of this limited scope.

However, the approximate nature of this data does not in validate it. The absolute values of the individual peaks are questionable, but the relative percentages can be used to comment upon the fatty acid content of this seaweed.

However, the problems and deficiencies of the method must be kept in mind in terms of this data and the literature data.

Literature concerning the fatty acid content of algal species is s p a r s e . 7 7 , 78 ,8 0 , 81 prior to the present work,

GC/MS analysis has not been used for identification.

Instead, identification has been accomplished using GC 59

retention times. Variations in data can stem from the types of columns used (open tubular or packed columns) and the method used to determine peak assignment. In the earlier experiments with packed columns,^ »78 identification of the carbon chain length was determined from the linear relationship between the logQQ) °f the retention time and the number of carbon atoms. The number of double bonds was determined by plotting the log retention time on a polar phase vs. the log retention time on a nonpolar phase. This analysis was adequate for the low resolution separations.

A more precise method of identifying unknown fatty acids was introduced by A c k m a n . 7 9 His graphical identification of individual fatty acids was based upon a relationship between unsaturated fatty acids, the chain length, and the position of the double bonds relative to the terminal methyl group. In a plot of log tr^/tr2 (tr^ = retention time of each FAME; tr^ = retention time of a reference, methyl octadecanoate) vs. carbon number, a critical line is drawn through the points for the monoethylenic methyl esters of

9-octadecenoic acid, 11-eicosenic acid and 13-docosenic acid. For each series of FAME's having the same degree of unsaturation and whose double bond sequence begins the same distance from the terminal methyl group, a plot of log tr^/t^ vs. carbon number can be made and the resulting line will have a slope equal to that of the critical line.

In this manner, unknown fatty acids can be identified by 60

plotting the relative retention times of a few reference standards. This method is superior to the earlier methods because it locates the positions of the double bonds as well as the degree of unsaturation.

Retention time data is accepted as a valid means for

FAME identification. Use of open tubular and capillary columns has increased the accuracy of the data. It must be kept in mind, however, that without comparison to an authentic sample, mistakes can be made.^ Such factors as column age and temperature programming can influence the order and time of elution and columns must be recalibrated often for the data to be valid.

The data on fatty acids from the literature is presented in Tables Via and VIb. Chuecas and Riley,^ conclude that there are no substantial differences in the fatty acid content of brown algae and Tables Via and VIb bear this out.

In most of the sets of data (even allowing for the dis- crepencies in values that Cheucas and Riley and Klenk et: al. report for Fucus serratus and F. vesiculosus), C ^ . q ,

^18-1 anc^ ^18-2 Predominate. This is true for both tropical and northern species.

Interestingly, the one exception is the data on Agarum cribosum. For this species, and predominate while Ci 8 ;l and C-^g.2 are minor components. Also, C2 2 . 5 is a substantial component. This data is different even from species in the same family, Laminariales. 61

TABLE Via. FATTY ACIDS OF BROWN ALGAE

% of Weight of Total Fatty Acids

Fatty AN PC FP FV FV FS FS LS LD Acid la Ila Ha IVa IVb Va Vb Via Vila

C14:Q 8.9 7.9 1 1 . 6 15.3 1 1 . 2 14.5 9 .7 10.9 8 . 1 0.7 0.7 -- 0.7 -- C15: 0

C16:0 9.5 8 . 6 24.5 13. 3 25.6 15.8 26.4 21. 3 24.5

3.9 2.4 2 . 0 2.4 1 . 6 7.1 2 . 2 10.7 1 1 . 8 C16 :1

-- -- 0 . 2 -- 0 . 2 -- 0 . 1 -- -- C16 : 2

C16 : 3 -- -- 0.4 -- 0.3 -- 0.4 -- --

-- -- 0.3 -- 0.4 0 . 1 C16 : 4

C17 : 0 -- 0.3 -- 0.4 0.4

-- 2.3 0 . 8 0 . 8 1 . 0 0 . 8 2.4 2 .3 C18 : 0

44.1 46.6 16.2 48. 8 17.2 36.3 29.6 C18 : 1 18. 7 30.6 9.5 14.5 9.0 C18 : 2 7.6 12.3 7.1 15.5 10.5 8.3

-- 7.1 -- 8 . 2 -- 6 . 0 -- -- C 1 8 : 3

C18 : 4 7.0 -- 6 . 0 -- 7.0 -- -- + -- -- 0.3 -- 0.5 -- -- C 2 0 : 0

3.4 1.5 -- 1 . 0 0 . 6 -- -- c 2 0 :2

C20:3 6 . 2 1 . 2 1 . 1 -- 1 . 1 0.4 --

C20 : 4 7.2 11.2 10.9 4.0 1 0 . 1 6 .0 1 0 . 0 7.0 6 . 1

C20:5 1 . 8 1 . 0 7.8 -- 7.6 8 . 1 2 . 1 3.9 + c22:5 -- + + -- FUCALES: I = Ascophyllum nodosum; II = Pelevtia canaliculata; III = Fucus platycarpus; IV = F. vesiculous; V = F. serratus. LAMINARIALES: VI = Laminaria saccharina; VII = L. digitata. a = Reference 77; b = Reference 78. 62

TABLE VIb. FATTY ACIDS OF BROWN ALGAE

7o Weight of Total Fatty Acids

Fatty Hormorsira* Ralfsia* Dictyota* Agarum# Acid banksii sp dichomota cribosum

0.08 tr 0.09 0.13 C 1 2 : 0

__ CI3 : 0 0.15 3. 43 7.17 10.32 5.36 C14:: 0

c14 : 1 0.80

C14:: 1 7 0 . 2 1 0.48 0.40 0.53 1.73 C15:: 0

__ 2 . 2 0 C15:: 1 m 8

-- 0 . 2 0 C15::1 m 6

24. 65 . 8 8 C16 :: 0 35 23.05 23.68

C16 ::lw9 0.36 0.16

C16 : 1m 7 1.93 1.43 1.57 13. 20

C16 : lay 5 tr 0.13 8.09 1.48

C16 ::la)13 0.17 0.97 0.71

C16 :2 w 6 tr 0.13 -- 0.60 -- C16:: 2oj4 0.16 0.34 3.29 tr 0.05 -- 0.30 C 1 6 ::3m 6 tr 0.09 -- C16 ::3a)4

C16 :;3w3 tr 0.19 -- 1 . 08

__ C16 :: 4m 3 0.09 0.83 6.60

0 . 2 1 tr C17 :0 0.69

C17 ::1 m 8 1.50

1.15 0. 54 2.40 1 . 2 0 C18 : 0 63

Table VIb (continued) ^Reference 79; #Reference 80

% Weight of Total Fatty Acids

Fatty Hormorsira* Ralfsia* Dictyota* Agarum# Acids banksii sp dichomota cribosum

C18:la)9 18.69 14.92 22.67 5.93 1.05 1.57 0.89 C18:1o)7 C1 8 :2w6 6.39 3.80 {2. 89 2.16 c18:3w3 7.98 6 . 70 3.74

C18:3w6 0.34 0.48 {8.81 0.19 c 1 8 : 4to3 3.80 5.87 1 . 2 1

Cl8:4wl -- 0.82

C19 : 0 0 . 2 1 tr 0.53 0.16 0.19 0.35 0.80 0.59 C 2 0 : 0

__ -- 0 . 1 2 C2 0 :lwll

0 . 2 2 0.27 -- C20:lw9 0.16 c20:lw7 0.13

1. 63 0.16 0 . 1 1 C2 0 :2w 6

4.11 0. 76 3 . 1 1 0 . 1 2 C20:3w6 c20:3oo3 0.44 0.17 0.09 12.98 6.83 {11.46 7.19 c2 0 :4w6 6.64 8.69 4.61 c 20:5o>3

1.57 0 . 8 8 tr C20:4o)3

0 . 1 2 tr tr C 2 2 : 0

0 . 1 2 tr {1.60 0.53 c2 2 :4w 6

C2 2 : 5co3 0.07 tr 5.17 c22:5o)6 tr tr -- -- Coo . o 0.13 tr — _ — — 64

The possibility of an error in either isolation pro cedures or chromatography is ruled out upon comparison of the two data sets on Agarum cribosum (Table VII). At the same time that our lab was obtaining FAME data on Agarum cribosum by GC/MS analysis, Ackman was looking at the same research problem by retention time analysis using open tubular columns (SILAR-5CP and Apiezon L).. The two data sets are complementary. Ackman's GC work has the resolu tion lacking in our own chromatogram and subsequent estima tion of peak area. This method of identification also locates the double bond positions. However, the identifica tion of peaks in Ackman's chromatogram based upon graphical relationships must remain tentative. In this sense, our identification, while it does not predict the positions of unsaturation, is unequivocal because for each peak (even in cases of overlap) we have the mass spectral molecular ion. Considering the very different analyses by which the data sets were obtained, a comparison of the two data sets in Table VII shows them to be in reasonable agreement.

The difference between the fatty acid content of Agarum cribosum and other species cannot be explained at present.

Greater numbers of brown algal species need to be examined to see if other species have a fatty acid distribution similar to Agarum cribosum.

In concluding the fatty acid discussion, it should be noted that the ideal method for analyzing these FAME mixtures 65

TABLE VII. COMPARISON OF NEWBURGER AND ACKMAN FAME % WEIGHT DATA

FAME Newburger Ackman

4.4 5.4 C1 4 : 0

C1 4 :1 0.4 1 . 0

C15 : 0 1 . 1 1.7

C15 :1 -- 2.4 29.5 23.7 C16 : 0 19.0 C16:l 14.7 1.9 3.9 C16:2

4.1 8 . 2 C16:3, 18:1 Cl'6:4 5.7 6.7

0 . 2 0.7 C17 : 0 C17 :1 -- 1.7

O 2 . 1 2 . 2 t— 00 N3

C18 : 3 1.7 3.9

2 . 6 C 1 8 : 4 2 . 0

C19 : 0 0 . 2

C19 :1 0 . 2

1.4 0 . 6 C2 0 : 0

0 . 1 C2 0 :l

0 . 1 C2 0 : 2

0 . 6 C20:3 0 . 2 9.0 C 2 n • l 7.2 66

TABLE VII (continued)

FAME Newburger Ackman

C20:5 2 - 3 4 ‘6

C22:4 1 , 3 ° - 5

C22:5 9 , 2 5 , 2 67

would be to have an open tubular or capillary column GC

instrument interfaced with a mass spectrometer. Maximum resolution could then be obtained and mass spectra could be used to corroborate tentative GC identification based on retention.time data.

Isolation of Mannitol

After the extraction with ether, methanol was added to the milled seaweed and the mixture refluxed gently. After twenty-four hours, a substantial amount of white crystals deposited on the sides of the flask. Some of the crystals were removed and recrystallized in methanol. The resulting white needle crystals were identified as the hexose, mannitol (mp 166-169°; lit. mp 166-168°). Upon filtering the methanol extract, large quantities of the crystalline sugar were found mixed in with the ground seaweed particles.

No attempt was made to determine the total amount of mannitol present (authentic sample mixed mp 166-168°).

CH^OH

HO H HO H H OH H OH

CH-jOH mannitol 68

82 Mannitol is ubiquitous in brown algae. Recently

Q O Mehta et: a^. examined fifteen species of brown algae

for their mannitol content. In different genera, they

found variations in the amount of mannitol present. They

also found that the mannitol content of the same species

could vary depending upon the location and the time of year of collection. The mannitol content of some species

of Laminaria was reported to be from 6 . 8 percent to 23.5

O A percent of the dry weight. EXPERIMENTAL

General. Proton magnetic resonance (^H-NMR) spectra

were recorded on a JEOL JNM-MN 100 or a Varian A-60 spectro

meter. Unless otherwise stated, the spectra were run in

CDCl^ and chemicals shift are reported in parts per million

(6 ) downfield from tetramethylsilane as an internal standard

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

337 spectrometer as neat liquids, solutions, or in a KBr

pellet.

High Pressure Liquid Chromatography (HPLC) was per

formed on a Waters Prep LC 500 with a silica column and a

refractive index detector.

Melting points were determined on a Thomas-Hoover Uni-

Melt apparatus and are uncorrected. Melting points are

recorded in degrees Celsius.

Elemental Analyses were performed at the University of

New Hampshire on an F&M 185 instrument.

Materials. Thin layer chromatography was performed

on precoated silica gel F-254 plates purchased from E. Merck

Dry column chromatography was performed in nylon tubing

filled with "Silica Gel Woelm for Dry Column Chromatography" purchased from ICN Pharamceuticals, Inc. Preparative TLC was performed on glass plates prepared with "Silica Gel 60

PF-254" purchased from E. Merck. The silica layer was 1 mm

thick.

69 70

All chemicals and reagents were used as purchased unless

otherwise noted.

All extraction solvents were distilled. Removal of

solvents at reduced pressure was accomplished on a rotary

evaporator unless otherwise noted.

Extraction. Agarum cribosum was collected in the drift

after a major storm at Rye Beach, New Hampshire, in October

1976. The seaweed was free of macroscopic epiphytes except

on the holdfasts which were discarded. The seaweed was air

dried and ground in a Wiley Mill with a hole mesh size #20.

The ground seaweed (4.5 kg) was gently refluxed in ether

(6 L) for several days. The ether was decanted and more

ether added to return the solvent volume to the original

level. This procedure was repeated twice. The ether ex

tracts were filtered, concentrated to one liter at reduced

pressure, dried (MgSO^), and concentrated at reduced pres

sure to yield a green-brown oil. A total of 33 g of crude

extract were obtained.

Open Column Chromatography of Nonpolar Constituents.

A 90 cm x 5 cm column was packed with silica gel (400 g;

40-140 mesh; Baker) using hexane as the solvent. Crude

algal extract ( 1 1 g) was absorbed on silica gel (4 . 1 g) and

layered on top of the column. The extract was eluted suc

cessively with hexane (1 L) , 10 percent benzene/hexane

(1 L), 50 percent benzene/hexane (2 L), benzene (0.5 L),

10 percent ether/benzene (0.5 L), 50 percent ether/benzene

(4 L), ether (0.5 L), 10 percent methanol/ether (0.5 L), 71

50 percent methanol/ether (0.7 L) , and methanol (2.4 L).

Fractions of 200 mL were collected and flasks were changed as different pigment bands were eluted. A total of

64 fractions was obtained. Fractions 1-48 were analyzed by

TLC and the following similar fractions were combined: 4->6,

7-11, 12-13, 14-19; these fractions contained an insignifi cant amount of material.

20-27, 0.02 g of an orange oil

28-29, 2.34 g of a green oil

30-31, 1.94 g of a green oil

32-35, 1.32 g of a green oil

36, - - a brown crystalline solid

37-48, 0.90 g of an orange oil

l-H-NMR analysis of these combined fractions showed all to contain triglycerides, with Fraction 28-29 and

Fraction 30-31 containing significant amounts of chlorophyll and sterols, respectively. A portion of Fraction 30-31

(1.74 g) was saponified (see saponification procedures) to yield 0.44 g of free sterols.

Saponification of Crude Extract.^2 in a typical saponification, crude extract (16 g) was heated in 1 0 percent

KOH in methanol (300 mL) for 4 h at 45°. The mixture was then divided in half. Each portion was poured into 300 mL

H 2 O and extracted with four 250 mL portions of ether until 72

the ether layer was colorless. The combined ether fractions were washed with H 2 O until neutral and dried (MgSO^). The ether was evaporated at reduced pressure to give 2.7 g of a red oil, the non-saponifiable lipids.

Separation of Non-Saponified Lipids. The non-saponifiable lipid oil (2.7 g) was separated by prep-HPLC using a 50 per cent ether/hexane solvent mixture. Detection by refractive index enabled the separation to be followed closely. Frac tions 1 and 2 were not identified. Fractions 3, 4, and 5 contained 0.561 g of phytol /^iH-NMR (100 mHz, DCDI 3 ) 6 5.5

(t, Hr C = C), 4.2 (d, H 2C - 0), 1.6 (s, CH3 - C = C), 0.8-

0.9 (2d, CH 3 - CH)[7. Fractions 6 , 7 and 8 (1.08 g) were determined by -^H-NMR to contain sterols. Fraction 9 (0.019 g) was also a sterol fraction but gave a different ^-H-NMR spectrum in the region of 5-6 pmm (6 ). Fraction 6 , 7, 8

(fucosterol and 24-methylenecholesterol): ^H-NMR (60 mHz,

CDCI 3 ) 6 5.3 (1H, H - C = C), 4.7 (d, H 2C = C ) , ca. 3.5 (b,

1H, H - C - 0), 1.6 (d, C = CHCH 3 ) , 1.0, 0.9, 0.7 (-CH3 ).

Identification of Sterol Components. The procedure described below was performed by the staff of Dr. Robert

Gagosian at the Woods Hole Institution of Oceanography.

The sterol fractions were converted to their respective acetate derivatives with pyridine and acetic anhydride

(Supelco, Inc., Bellefonte, PA) and analyzed by high resolu tion glass capillary gas chromatography on a Carlo Erba

Fractovap 2150 gas chromatograph. The helium carrier gas 73

flow was 1.5 - 2 ml per minute. Injector and detector were operated at 275°C; the derivatized samples (1-2 yl) were injected splitless at 150°C, heated to 250° at 4°/minute and held isothermally for thirty minutes. The glass capillary column (20 m x 0.32 mm i.d.) used was coated with SE-54

(methyl silicone gum, containing 5 percent phenyl and 1 percent vinyl groups)(H & G Jaeggi, Trogen, Switzerland). 90 The film thickness measured according to Grob et_ al. was approximately 0 . 1 1 y.

The concentrations of the sterols were determined using an internal standard (5a-cholestane, 10 ng/yl). The cal culation of peak areas and the corresponding concentrations x^ere accomplished by a programmable integrator (Supergrator 3,

Columbia Scientific Industries) interfaced with the gas chromatograph. The sterol concentration variability for consecutive multiple injections is 4- 15 percent for standard mixtures at the 1 0 ng individual sterol level.

A gas chromatograph-mass spectrometer-computer (GC-MS-C) system was used to identify individual sterol acetates. The gas chromatograph used for GC-MS work was a Varian Aerograph

1400 equipped with a glass capillary column (17 m x 0.32 mm i.d.) coated with SE-52. The column was prepared in the laboratory of Dr. Robert Gagosian. The temperature program was the same as described earlier for the Carlo Erba instru ment. The gas chromatograph was interfaced by glass capillary tubing to a Finnigan 1015C quadrupole mass spectrometer. The 74

mass spectrometer was run in the electron impact mode; the ionization potential was 70 eV. The data were collected and edited by a DEC PDP 8 -E (16 K core) computer equipped with a

System Industries System 150 data system, Diablo Series 30 disc drive, and Tektronic 4010 CRT display unit and 4610 copier.

Isolation of Fucosterol from Ascophyllum nodosum and

Formation of Fucosteryl Acetate. Ascophyllum nodosum was collected at Adams Point, New Hampshire, in May 1978.

It was first air dried and then oven dried at 75°. The dried seaweed was hand-ground to give 1.78 kg (particle

O size: powder to 2 mm ). The seaweed was gently refluxed in*ether (4 L) for 4 days. The ether solution was filtered and concentrated to give 6 8 g of a crude green oil. The extract was saponified according to the procedure for

Agarum cribosum, yielding 10 g of non-saponifiable lipids, a red oil. Upon standing, the oil solidified to give 6.31 g of solids. The solids were recrystallized two times in methanol to give 3 g of beige crystals (m.p. 90-105°; lit. m.p. 124° )^. The crystals were not further purified and were used to make the acetate. Fucosterol (2 g; 0.0049 moles) was dissolved in dry pyridine (8 mL) and acetic anhydride (5 mL) was added. The solution was heated for

1 h on a steam bath. After cooling, 50 mL methanol were added, and the mixture was boiled for 2 min giving a clear solution. Upon cooling, crystals precipitated. Upon 75

recrystallization from methanol, 1.9 g (85 percent) of white crystals of fucosteryl acetate were obtained (m.p. 118-119.5° lit. m.p. 118-119°).62 1 H-NMR (100 mHz, CDC13) 6 5.3 (b,

1H, H - C = C ) , 5.1 (q, 1H, C = C - HCH 3 ) , 4.5 (b, 1H, H -

C - 0), 2.0 (s, CH 3 -0), 1.5 (d, C = CHCH3 ) , 1.0, 0.9,

0.7(-CH3).

24-Ketocholesteryl Acetate .^ »^ 4 Powdered fucosteryl acetate (0.4 g; 0.00088 moles) and glacial acetate acid

(16 mL) were placed in a 50 mL test tube. The reaction mixture was cooled, with stirring, in an ice bath. A slow stream of ozone (4 percent in air) was added until most of the solids had dissolved. After warming to room temperature, zinc dust (0.45 g; 0.0069 moles) was added, and the mixture was stirred for seven min. The zinc dust was filtered and

40 mL H 2 O was added. The aqueous mixture was set aside.

The ozonlysis was repeated. After filtering the zinc dust from the second reaction, 60 mL H 2 O was added. Both aqueous mixtures were combined and extracted with ether

(100, 50, 50 mL portions). The combined ether layers were washed successively with H 2 O, aq. Na 2 C0 3 , and H 2 O, and then dried (MgSO^). Upon concentration at reduced pressure,

0.76 g of a white solid were obtained. Preparative TLC separation (15 percent ethyl acetate/hexane; 2 elutions) yielded 0.2 g recovered fucosteryl acetate and 0.3 g (39 percent) of a white solid, 24-ketocholesteryl acetate

(m.p. 126-130°; lit. m.p. 127.5-128°) . 6 4 1 H-MMR (60 mHz, 76

CDC13) 6 5.3 (b, 1H, H - C = C) , ca. 4.5 (b, 1H, H - C - 0), ca. 2.6 (H2 CC =0), 2.0 (CH3 - 0), 1.1, 1.0, 0.7 (-CH3> .

24-Ketocholesterol.^5 To 24-ketocholesteryl acetate

(0.39 g; 0.00088 moles) dissolved in methanol (16 mL) was added K 2 CO3 (0.14 g) dissolved in H 2 O (1 mL). The reaction was stirred at room temperature overnight. After 15 h, the clear solution was poured into 100 mL H 2 O and extracted with ether (100, 50, 50 mL portions). The combined ether layers were washed with aq. NaCl and dried (MgSO^). Concentration at reduced pressure yielded 0.36 g of a white solid. Re crystallization from hexane gave white crystals 0.31 g ( 8 8 percent) of 24-ketocholesterol (m.p. 136.5-137°; lit. m.p.

137-138°64). 1 H-NMR (60 mHz, CDCI3 ) 6 5.3 (b, 1H, H - C = C ) . ca. 3.5 (b, 1H, H - C - 0), ca. 2.5 (b, H 2 CC = 0), 1.2-1.0,

0.7 (-CH3 ).

24-Ethynyl-24-Hydroxycholesterol.^6 A 50 mL 3-neck round bottom flask was equipped with a dropping funnel having a nitrogen inlet, a condenser with a drying tube and an acetylene inlet. Potassium (0.38 g) was dissolved in t-amyl alcohol (13 mL; freshly distilled from sodium) with refluxing under nitrogen and with stirring. The resulting potassium alkoxide solution was cooled to 0°. 24-Keto- cholesterol (0.29 g; 0.000725 moles) dissolved in dry ether

(13 mL; freshly distilled from sodium and benzophenone) was added to the dropping funnel. A moderate stream of acetylene was bubbled through the alkoxide solution for several minutes and then the sterol solution was added 77

dropwise, keeping the temperature below 5°. After the ad

dition, the reaction mixture was allowed to come to room

temperature. Acetylene was passed through for six more hours. The reaction mixture was poured into 40 mL iced

IN H 2 SO4 and the aqueous mixture extracted with ether. The

ether solution was washed with Na 2 C0 3 solution, NaCl solution,

and then dried (MgSO^.) . Concentration at reduced pressure yielded 0.49 g of an orange glass. The glass, separated by preparative TLC (45 percent ethyl acetate/hexane; 2 elutions), yielded seven components (visualized by spraying the side of the plate with 50 percent aq. H 2 SO4 and heating in the oven at 75°). The sterol band, indicated by a purple spot, was cut and extracted into ether. Concentration of the sterol fraction showed it to be a white solid, 0 . 2 g.

This solid was separated by preparative TLC (25 percent ethyl acetate/hexane; 4 elutions) to give four bands (aq.

H 2 SO4 visualization). The band with the highest Rf value was identified as 24-ketocholesterol. The band with the next highest Rf value yielded 0.026 g (7.6 percent) of

24-ethynyl-24-hydroxycholesterol (m.p. 184-186°; lit. m.p. 184-184.5°87) „ IR (CDCl^ 3600, 3300, 3020, 2940,

1600. 1 H-NMR (60 mHz, CDCI3 ) <5 5.3 (b, 1H, H- C = C) , ca.

3.5 (b, 1 H, H - C - 0), 2.4 (s, H - C = C ) , 1.1-0.9, 0.7

(-CH3 ).

24-Vinyl-24-Hydroxycholesterol (Saringosterol) .8 8

24-Ethynyl-24-hydroxycholesterol (0.065 g; 0.00014 moles), 78

Lindlar's Catalyst (5 percent Pd on CaCC^, Pb(OAc)^ poisoned;

Engelhard; 0.01 g), methanol (8 mL) and quinoline-methanol solution (1.2 mL; 1 drop Fisher synthetic distilled quinoline in 20 mL methanol) were placed in a 25 mL round bottom flask equipped with a stirring bar. The flask was attached to an atmospheric pressure hydrogenator. The system was evacuated and flushed with hydrogen several times and then stirring was commenced. Hydrogen uptake began after 8 min. and ended after 20 min. After a total of 60 min, the reaction vessel was opened and the catalyst filtered. Concentration of the clear solution at reduced pressure yielded 0.067 g of an off-white solid. TLC (20 percent ethyl acetate/hexane; 4 elptions) showed two major spots, with the top one corres-

■ponding to 24-ethynyl-24-hydroxycholesterol. Preparative

TLC on 10 percent AgNC^/silica plates (Analtech; 33 percent ethyl acetate/hexane; 4 elutions) gave 0.04 g of a white solid. Recrystallization from hexane afforded 0.023 g

(35 percent) of saringosterol ( mp 156-160°; lit. mp 160-

161.5°87); ir (KBr) 3380, 3090, 2920, 1700, 1650, 1465,

1350, 1060, 995, 955, 920, 840, 800. 1 H-NMR (60 mHz, CDCI3 )

6 ca. 5.8 (CH = CH2) , ca. 5.3 (H - C = C ) , ca. 5.2 (H2C = CH), ca. 3.5 (H - C - 0), 1.0-0.8 , 0.7 (-CH3 ).

Formation of Synthetic Sterol Acetates. The sterol

(0 . 0 0 2 g) was placed in a round bottom flask which was capped with a septum. Pyridine (0.1 mL) and acetic anhydride

(50 jjL) were injected and the reaction was heated on a steam 79

bath for 1 h. The flask was cooled, 0.5 mL methanol was added and the mixture was boiled for 2 min. Water (10 mL) and ether (10 mL) were added and the organic phase was withdrawn. Evaporation of the ether with nitrogen yielded a yellow solid (0.0035 g for 24-ketocholesteryl acetate).

The crude sterol acetate was chromatographed on silica plates

(15 percent ethyl acetate/hexane; 2 elutions) to give the pure sterol as a white solid, 0.0013 g (58 percent) for

24-ketocholesteryl acetate). Solutions of 0.1 mg per 1 mL hexane were made for GC/MS work.

Isolation of Fatty Acids. ^ Crude Agarum cribosum extract (2g) was saponified in 10 percent KOH in methanol

(80 mL) for 6 h at 40-50°. The mixture was poured into

200 mL H 2 O and extracted with a total of 900 mL ether to remove the non-saponifiable lipids. Hydrochloric acid

(6 W, 25 mL) was added to the remaining aqueous solution to adjust the pH to 1. The acidified mixture was extracted with six 150 mL ether portions. The combined ether extracts were washed with 150 mL (2) ^ 0 portions and dried (MgSO^).

Concentration at reduced pressure yielded 0.8 g of a brown green oil (see ^H-NMR, Appendix C).

Formation of Fatty Acid Methyl Esters (FAME). The fatty acid oil (0.025 g) was added to methanol (10 mL) in a 100 mL round bottom flask. The solution was stirred and BCI3 gas was bubbled through for 5 min. The solution was re fluxed for 1 0 min and then the methanol was removed at 80

reduced pressure. A 1 percent solution of the fatty acid methyl ester mixture in methylene chloride was prepared for

GC/MS analysis.

Identification of Fatty Acid Methyl Esters. Gas chromatography to determine run conditions and peak areas was performed on a Perkin Elmer 3290 gas chromatograph with a flame ionization detector. The column (7 ft. x 1/8 in.) was packed with 12 percent DEGS stationary phase on Anakrom

A. Peak area data was collected under the following condi tions :

Flow rate: 22.5 mL ^ / m i n

Injector: 195°

Detector: 200°

Column: 150°, initial period: 25 or 32 min; program

to 175° at 16°/min

Gas chromatography at MIT was performed with the same column on a Perkin Elmer 900 gas chromatograph (FID) inter faced with a Hitachi-Perkin Elmer RMU- 6 L low resolution mass spectrometer, and an IBM 1800 computer system. Data was collected under the following conditions:

Helium carrier gas

Injector: 195°

Detector: 200°

Column: 150°, initial period: 16 min; program

to 175° at 16°/min 81

The FAME mixture was injected as a 1.0 percent solu tion in methylene chloride.

Isolation of Mannitol. After the Agarum cribosum solids were extracted with ether (see extraction procedure), they were refluxed in methanol (4 L). After 24 h, a thick layer of white crystals deposited about the circumference of the flask. Some of the crystals were removed and re crystallized from methanol. Upon filtering of the methanol extraction mixture, copius amounts of white solids were found mixed in with the seaweed solids. These were not further isolated. Recrystallization from methanol of the crystals gave white needle crystals which were identified as mannitol (m.p. 166-169; lit. m.p. 1 6 6 -1 6 8 ^ ) . BIBLIOGRAPHY

1. Ruggieri, G. D. Science 1976, 194, 491.

2. Faulkner, D. J . ; Andersen, R. J. In "The Sea"; J. W. Wiley: New York, 1974; Vol. V, Chapter 19.

3. Weinheimer, A. J.; Karns, T. K. B. In "Procedings of the Food-Drugs from the Sea Conference 1974"; Webber, H. H. and Ruggieri, G. D., Eds.; Marine Technological Society: Washington, DC, 19.74; 491.

4. Scheuer, P. J. "Chemistry of Marine Natural Products"; Academic Press: New York, 1973.

5. Khaleafa, A. F. Bot. M a r . 1975. 1 8 , 163.

6 . Hornsey, I. S.; Hide, D. Br. Phycol. J . 197Ji, 9_, 353.

7. Sieburth, J. M. Adv. Microbiol. Sea. 1968, 1, 63.

8 - Stewart, W. D. P. "Algal Physiology and Biochemistry"; University of California Press: Berkeley, 1974•

9. Fenical, W. J. Phycol. 1975, 11, 245.

10. Weinstein, B.; Rold, T. L.; Harrell, C. E., Jr.; Burns, M. W., III: Waaland, J. R. Phytochemistry 1975, 14, 2667.

11. Gonzalez, A. G.; Darias, J.; Diaz, A.; Fourneron, J. D. Martin, J. D. ; Perez, C. Tetrahedron L e t t . 197J5, 3051.

12. Wolinsky, L. E . ; Faulkner, D. J. J. Org. Chem. 1976, 41, 597. ^

13. Cafieri, B.; DeNapoli, L.; Fattorusso, E. Tetrahedron Lett. 1979, 963.

14. Burreson, B. J.; Moore, R. E.; Roller, P. P. J. Agric. Food Chem. 19J6, 24, 856.

15. Woolard, F. X.; Moore, R. E.; Roller, P. P. Phyto chemistry 1979, 18^, 617.

82 83

16. Webb, J. S. In "Proceedings of the Food-Drugs from the Sea Conference 1973"; Worthen, L. R . , Ed.; Marine Technological Society: Washington, DC, 1973; 275.

17. Pollock, E. G. Planta 19J70, 92^, 85.

18. Cook, A. H . ; Elvidge, J. A.; Heilbron, I. Proc. R . Soc. London, Ser. B 1948, 135, 293.

19. Hlubucek, J. R.; Hora, J.; Toube, T. P.; Weedon, B. C. L Tetrahedron Lett. 1£70, 5163.

20. Muller, D. G.; Jaenicke, L. FEBS Lett. 1973, 30, 137.

21. Jaenicke, L . ; Sefariadis, K. Chem. Ber. 19J5, 108, 225.

22. Muller, D. G.; Jaenicke, L.; Donike, M.; Akintobi, T. Science, 1971, 171, 815.

23. Jaenicke, L . ; Muller, D. G . ; Moore, R. E. J. Am. Chem. Soc. 197^, 96, 3324.

24. Jaenicke, L. Coll. Gessellshift. Biol. Chemie 1974, 25, 307.

25. Moore, R. E.; Pettus, J. A., Jr.; Mistysyn, J. J. Org. Chem. 1.9.74. 39, 2201.

26. Moore, R. E. Chem. Commun. 1971, 1168.

27. Moore, R. E . ; Mistysyn, J.; Pettus, J. A., Jr. J. Chem. Soc., Chem. Commun. 1 972., 326.

28. Moore, R. E. Lloydia 1976, 34, 181.

29. Pickenhagen, W.; Naf, F.; Ohloff, G.; Muller, P.; Perlberger, J. Helv. Chim. Acta 1 9.7,3, 5 6 , 1868.

30. Glombitza, K.-W.; Rosener, H.-V.; Vilter, H.; Rauwald, W. Planta M e d . 1973, 24, 301.

31. Craigie, J. S.; McLachlan, J. Can. J. Bot. 1964, 42, 23

32. Glombitza, K.-W., Rosener, H. V. Phytochemistry 19, 13, 1245.

33. Glombitza, K.-W.; Rosener, H. V.; Muller, D. Phyto chemistry 1975^., 14, 1115.

34. Glombitza, K.-W.; Sattler, E. Tetrahedron Lett. 1973, 4277. 84

35. Glombitza, K.-W.; Rosener, H.-V.; Koch, M. Phyto chemistry 1926, 15, 1279.

36. Glombitza, K.-W.; Koch, M. ; Eclchardt, G. Phyto chemistry 197b, 1082.

37. Glombitza, K.-W.; Rawauld, H.-W.; Eckhardt, G. Phytochemistry 1975, 14, 1403.

38. Ragan, M. A.; Jensen, A. Phytochemis try 19 79, 18, 261.

39. Conover, J. T . ; Sieburth, J. McN. In "Proceedings of the Fifth International Seaweed Symposium"; Young, E. G., and McLachlan, J. L., Ed.; Pergamon Press: New York, 1966; 207.

40. Sieburth, J. M c N . ; Jensen, A. J. Exn. Mar. Biol. Ecol. 1969, 3, 275. '

41. Gerwick, W. H.; Fenical, W.; Fritsch, N.; Clardy, J. Tetrahedron Lett. 1979, 145.

42. Gonzalez, A. G.; Darias, J.; Martin, J. D. Tetrahedron Lett. 19_71, 2729. :

43.' Kato, T.; Kumanireng, A. S.; Ichinose, I.; Kitahara, Y.; Kakinuma, Y.; Nishihira, M.; Kato. M. Experentia 1915, 31, 433.

44. Gregson, R. P.; Kazlauskas, R.; Murphy, P. T.; Wells, R. J. Aust. J. Chem. 1977, 30, 2527.

45. Pedersen, M.; Fries, L. Z. Pflanzenphysiol. 1975, 74, 2 7 2 ~

46. Fattorusso, E.; Magno, S.; Mayol, L.; Santocroce, C.; Sica, D . ; Amico, V. ; Oriente, G . ; Piattelli, M . ; Ringali, C. J. Chem. Soc., Chem. Commun. 1976, 575.

47. Danise, B.; Minale, L.; Riccio, R.; Amico, V. ; Oriente, G.; Piattelli, M . ; Triengal, C.; Fattorusso, E.; Magno, S. ; Mayol, L. Experentia 1977 > 33, 413.

48. Hirschfeld, D. R . ; Fenical, W . ; Lin, G. H. Y . ; Wing, R. M . ; Radlick, P.; Sims, J. J. J. Am. Chem.Soc. 1973, 95, 4049.

49. Sun, H. H. ; Fenical, W. Phytochemistry 1979_, 18*, 340.

50. Fenical, W. ; Sims, J. J.; Squatrito, D. ; Wing, R. M. ; Radlick, P. J. Org. Chem. 1973, 38, 2383. 85

51. Cimino, G.; de Stephano, S.; Fenical, W.; Minale, L.; Sims, J. J. Experentia 1975., 3 1 , 1250.

52. Fattorusso, E.; Magno, S.; Mayol, L.; Santacroce, C.; Sica, D.; Amico, V.; Oriente, G.; Piattelli, M.; Tringali, C. Tetrahedron Lett. 1976, 937.

53. Lunde, G. J. Am. Oil Chem. Soc. 1973, 50, 24.

54. Fenical, W. , Scripps Institution of Oceanography, personal communication, 1976.

55. Waraszlciewicz, S. M. Ph.D. Dissertation, Clark University, Worcester, Mass., 1975.

56. Howard, B. H . ; Fenical, W. Tetrahedron Lett. 1976, 41. ---

57. Burreson, B. J.; Woolard, F. X.; Moore, R. E. Tetrahedron Lett. 1975, 2155.

58. Bhacca, N. S.; Johnson, L. F.; Shoolery, J. N. "High Resolution NMR Spectra Catalog," Vol. 1; National Press: USA, 1962; no. 346.

59'. Piretti, M. V.; Viviani, R. Comp. Biochem. Physiol. 1976, 55B, 229.

60. Wyllie, S. G . ; Djerassi, C. J. Org. Chem. 1968, 33, 305. ~

61. Tsuda, K.; Hayatsu, R.; Kishida, Y.; Akagi, S. J . Am. Chem. Soc. 1958, 8 0 , 921.

62. Heilbron, I.; Phipers, R. F.; Wright, H. R. J. Chem. Soc. 1934, 1572.

63. Larsen, C. D. J. Am. Chem. Soc. 1963, 60, 2431.

64. Hey, D. H . ; Honeyman, J.; Peal, W. J. J. Chem. Soc. 195.0, 2881.

65. Bates, R. B . ; Brewer, A. D. Tetrahedron Lett. 1968, 6163.

66. Safe, L. M.; Wong, C. J.; Chandler, R. F. J. Pharm. Sci. 1963, 63, 464.

67. Patterson, C. W. Comp. Biochem. Physiol. 1968, 24, 501. ’

68. Ikekawa, N.; Morisaki, N.; Tsuda, K.; Yoshida, T. Steroids 1968, 12, 41. 86

69. Ikekawa, N.; Morisaki, M.; Hirayama, K. Phytochemistry 1972, 11, 2317.

70. Smith, L. L.; Dhar, A. K.; Gilchrist, J. L.; Lin, Y. Y. Phytochemistry 1973., 12, 2727.

71. Knights, B. A. Phytochemistry 1910, 9, 903.

72. Goodwin, T. W. Biochem. J . 1971, 123, 293.

73. Goodwin, T. W. In "Algal Physiology and Biochemistry"' Stewart, W. D. P., Ed.; University of California Press: Berkeley, 1974; Chapter 9.

74. Castle, M . ; Blondin, G. A.; Nes, W. R. J. Am. Chem. Soc. 1963, 85, 3306.

75. Castle, M . ; Blondin, G. A.; Nes, W. R. J. Biol. Chem. 1967, 242, 5796.

76. Ackman, R. G. In "Progress in the Chemistry of Fats and Other Lipids"; Holman, R. T., Ed.; Pergamon Press: New York, 1972j Vol. 12, 165.

77. Chuecas, L . ; Riley, J. P. J. Mar. Biol. Ass. U. K. i 1966, 46, 153.

78. Klenk, E. ; Knipprath, W. ; Eberhagen, D.; Koof, H. P. Zeitschrift f. Physiol. Chemie 1£63, 334, 44.

79. Ackman, R. G. J. Am. Oil Chem. Soc. 196^3, 40, 558.

80. Johns, R. B.; Nichols, P. D.; Perry, G. J. Phyto chemistry 19^79, 18, 799.

81. Ackman, R. G., Fisheries and Environment Canada - Halifax Laboratory, personal communication, 1977.

82. Craigie, J. S. In "Algal Physiology and Biochemistry"; Stewart, W. D. P., Ed.; University of California Press: Berkeley, 1974,; Chapter 7.

83. Mehta, B. R . ; Parekh, R. G. B o t . Mar. 1978, 21, 251; Chem. Abstr. 1978,, 89, 87170q.

84. Bukhryakova, L. K . ; Levanidov, I. P. Rast. Resur. 1969, 5, 183; Chem. Abstr. 1967, 6 6 , 9883j.

85. Adams, W. J.; Patel, D. K.; Petrow, V.; Stewart-Webb, I. A. J. Chem. Soc. 1954, 1825. 87

86. Thomson, A. F . , Jr.; Burr, J. G . ; Shaw, E. N. J. A m . Chem. Soc. 1941, 63^, 184.

87. Ikekawa, N.; Tsuda, K.; Morisaki, N. Chem and Ind. 1966^, 1179.

88. Augustine, R. L. "Catalytic Hydrogenation"; Marcel Dekker: New York, 1965_; Chapter 4.

89. Stecher, Paul G. "The Merck Index," 8th ed.; Merck and C o . : Rahway, 1968; 664.

90. Grob, K., Jr.; Grob, G.; Grob, K. J. Chromatography 1978, 156, 1. APPENDIX A: Mass Spectra of Sterol Acetate Standards 1 1 O0-i

.-■u-

tj-Ji rfrrrA l^ r j t HrrJj li*n i lllt'i >*> HI 6 Q 80 100 120 140 160 ISO 206 100-1

50-

‘|*rt 200 £20 240 260 280 300 320 340 3 66 100-1 75-

50- 25-

Figure 1 MS Fucosteryl Acetate (Ascophyllum nodosum).

0 0 VO I

.TpU-

V r;,.

jW>Tp»AV|i ilU rrl^Wp'i^lfPi'J jill A A j I I'l’A V ri r i*i'l 10 on 100 120 140 160 130 £0g 10i iT

H »I’r1*1*1» I'M tiS mT|'iM rr\ i rl’rp r*ri | i »■» i'ii i p ir n r ilf i*i*iTl ' i n i | ZOO 220 £40 26o 280 300 320 340 -----366 100- |

5 0 -

Figure 2. MS Synthetic 24-Ketocholesteryl Acetate. 150

350

•400 430 500

Figure 3. MS Synthetic Saringosterol Acetate.

VO h-1 Lu

Figure 4. MS Synthetic Saringosterol .jB-

0-Ljt r% i ii *> f ihi myri-iVnVi'iiMrJ^i'n yTi'i'rl^ 'IvrlljHtt^-iMRyyYivrl'l'lWrth^rririVi 40i 60 80 100 120 140 160 180 206 100-1 75- 5Q-

5 0 - 25'

0-* I1»in i nii|iin| iiri'i H'l »| 111 ryi rii| 1111| 11 ii | n ii| 11 n |i I i i | m 1111111 260 380 400 420 440 460 480 500

Figure 5. MS Cholesteryl Acetate (Fisher).

VO 1L- . „ I, IL . 111.! IllL .. illllll .llllll.i i.ll li Jlljlii. - i l.r. i/ 1 1 I * I 1 ! 1 1 ' 1 1 \ ' 1 31.1U

Jtjij -.^u

•I X ll, ,. ( ,.-r-r j- f 1 -1 1 1 1 1 T-'i—i- i —i"*T—y*T~r"« I""*' f 1 f 4 Ti0 !mu

Figure 6 . MS Desmosteryl Acetate (Applied Science). APPENDIX B: FAME Mass Spectra and Ionization Intensity Plots 96

013 6097 MRX « 995B-

T ■*• 100 300

M/E 242 MRX. INT. * 2023 RUN NO. = 06097

GC Peak 4 14:0

035 MRX - 27596097

100 300 400

M/E 256 MRX.' INT. * 643 RUN NO. = 06097

GC Peak 6 97

079 6097 MRX * 9955-

100 2 0 0 300

M/E 270 MRX. INT. * 9713 RUN NO. * 06097

GC Peak 8

251 .6106 MRX * 9956-

100 300

M/E-268 MRX. INT. * 1195 RUN NO. = 06097

GC Peak 9 C 1 6 :1 98

146 6097 MRX * 2615

100 300 400

140 6097 MRX * 2554

100 300 400

M/E ^B6 MRX. INT. 1 438 • RUN NO. = 06097

GC Peak 11 c16:2

L PPP 6097 MRX - 4330

1-vH^ ‘i*r ■>I^,v*t *|‘i ‘t*^ i ‘n-.-i 100 an 300

M/E 264 MRX. INT. 1 1245 RUN NO- * 06097

GC Peak 14

570 6106 MRX 1 9955-

100 300

* M/E 262 MRX. INT. * 346 RUN NO. = 06097 1 0 0

508 6097 MAX * 2403

100 300 400

RUN NO. - 06097

aie 6097 MAX * 4026

100 2D0 300

M/E 296 MAX. INT. - 353 RUN NO. * 06097

I Jil fli idiA iVlMkJu'I'i ado GC Peak 14 '18:1 1 0 1

244 6097 MAX 1 4401

100 aoo 300

M/E 294 MAX. INT. 1 680 RUN NO. * 06097

* * I

GC Peak 16 18:2

6097 MRX - 3102

■») 2 0 0 4 0 0

M/E 292 MRX. INT. 1 337 RUN NO. * 06097

GC Peak 18 18:3 1 0 2

014 BO 98 MRX * 4070

100 £00 300

M/E 326 MRX. INT. : f 404 RUN NO. i= 0605' /•

GC Peak 19 C20 :0

6097 MRX * E898

100 £00 300

M/E 290 MRX. INT. » 275 RUN NO. . 0B097

GC Peak 20 18:4 103

133 S106 MRX - 3665

*^*1 100 an 300 — 400

M/E 330 MRX. INT. * 311 RUN NO. = 06097

M/E 330 MRX. INT. = 344 RUN NO. * 06098

GC Peak 22

} 1 104

140 s i o g MRX * 6010

T»1 1 T i-T ''| ''I '*t | ''I 'T *"l »'»' I ' ton

M/E 310 MRX. INT. * 437 RUN NO. * 06096

0 \

100 6090 MRX * 3000

M/E 344 MRX. INT. * 201 RUN NO. * 06096 0 I APPENDIX C: •^H-NMR Spectra iue . HNR (CDCI ^H-NMR 1. Figure Jra- 3 Clm hoaorpyFato 20-27 Fraction Chromatography Column ) Figure 2. ^H-NMR (CDCI3 ) Column Chromatography Fraction28-29 (Chlorophyll) 107 Figure 3: 1 H-NMR (CDCI3 ) Column Chromatography Fraction 30-31 (Sterols) 108 nSw^-

Figure 4: lH-NMR (CDGI3 ) Colmun Chromatography Fraction 32-35 Figure 5: ^H-NMR (CDClg) Column Chromatography Fraction 37-48

Figure 7: ^-H-NMR (CDCI3 ) Non-saponifiable Lipids (Fraction 2) Phytol BR?

1000 600 400 200

600 400 ■300 200 100

300 200 :iso 100

120 ;jo:

i.. l_J !. .1 _L1 113 Figure 8 ^H-NMR (CDClo) Non-saponifiable Lipids (Fraction 3) Fucosterol and 24-Methylenecholesterol 114 Figure 9: ^H-NMR (CDClg) Non-saponifiable Lipids (Fraction 4) Figure 10: 1-H-NMR (CDCI3 ) Fucosteryl Acetate U O O M i j 1 millSaiiiilill

O.Sppm

OO'T) Ml

Figure 11: ■'•H-NMR (CCDl^) 24-Ketocholesteryl Acetate ™isa

rri I'rrr-r

JO jipm 1 200H i |.;.840ij-l|i li “ ""I1 tO ppm , - l i S O o i !! ■IDO,,,

Sppm I'ljji-JoO'l:-

2pp m

I ppm ! li'liil

O .S p p m

^ I!

ppm(^) I 1 1 1 1 1 1 1 1 I 1 L J i J

1 117 Figure 12: H-NMR (CDCI3 ) 24-Ketocholesterol 100

300 v ' 1

2pp- 120

"20

w Ms

Figure 13: ^H-NMR (CDCI3 ) 24-Ethynyl-24-hydroxycholesterol r

i 7-ri i t r'r i

000 6 0 0 , {hi; 200

600 100

3 0 0 ; ISO

(20 100

I ppm

i.l-J J_ L X 1 l_i. 10 9 8 7 6 5 4 i 2 I 0 119 Figure 14: 1 H-NMR (CDCI3 ) 24-Hydroxy-24-Vinylcholesterol(Saringosterol) i-irv'o afip i--:: 200 H •ItH

Hr-

H—CM.

OPPM i-'MHil L_i/. i i)t i i I i I I :l. il l I ."I I -I.V I I I I I V 1 I l i I 120 Figure 15: ^H-NMR (CDCI3 ) Fatty Acid Mixture APPENDIX D: IR Spectra 1 2 2

4-0 MICRONS S-0 I—I I I 1

TT*V i-t"~ — TT-*" r- —*\

TXJT t; r* tr Txri

■

'nn.i!lliinlltii|! = nii;i IiIM i1111' 11 i i 1111 i i 1111 i 'I H 111 !i I<1h'111, i ll ■ <. 1.11111 ■. Ill 4000 3500 3000 2500 2000 1500

FREQUENCY (CM 1)

9-0 10-0 MICRONS 15.0 25-0

1200 1100 1000 900 800 700 6 0 0 500 FREQUENCY (CM ')

Figure 2: IR (KBr) 24-Hydroxy-24-vinylcholesterol saringosterol iue : R CC^ 24-Ethynyl-24-hydroxycholesterol (CDCl^) IR 1: Figure

ABSO k BANC O-Oh •40 jr: •40 o : 4000 "'■'fc'c. J! * O j a .J"!I m . L i . r B T ± t t r t ± 1300 DcT . 1200 o »^ 7 d _.clu s n o i + . ^ y o s 5030 2500 3000 3500 ::;::!~~: “ m 9-0 1100 00900 1000 rrz: 1 H i-t-H-H-f 1H (CM->) Y C N E U Q E R F CM '| ' M (C Y C N E U Q E R F 1+rr - MCO5 5-0 MICKON5 4-0 MICRONS 800 UtLlz U 700 2000 15.0 1500 500600 r . trf n r c - i S li 25-0 400 123 Part 2: The Effect of Electrostatic Interactions on the

Stereochemistry of the S^2' Reaction. INTRODUCTION

In Sj^jl nucleophilic substitution reactions, special phenomena are often observed with allylic systems that are not seen in their saturated analogs. With allylic systems,

there is a pronounced rate enhancement due to stabilization by it bond overlap. An associative rearrangement of the

leaving group prior to nucleophilic attack (internal return)

can lead to isomeric product mixtures as can a preliminary

ionization forming a dissociated carbonium ion intermediate.

In some cases, the degree of reagent nucleophilicty can

influence product ratios. All of the phenomena are based upon the availability of the tt bond as a reactive center.

Allylic systems can also undergo normal S^2 bimolecular

substitution. However, when the leaving group carbon is

sterically hindered, attack may occur exclusively at the

y-carbon leading to a rearranged product. The first

authentic example of this type of reaction, denoted S^2', O was reported by Kepner, Winstein, and Young in 1949. One possible mechanisn for an S^2' reaction involves a con

certed (although not necessarily synchronous) attack by

the nucleophile and exit by the leaving group (Figure 1).

While examples of the S^2' reaction have been shown to Q / follow bimolecular kinetics ’ , the concertedness of the

C fi reaction has been questioned by Bordwell. » Instead,

125 1 126

Bordwell promotes a mechanism involving the formation of a

discreet allylic intimate ion pair (not freely dissociated)

of the type proposed by Sneen et al. (Figure 2 ).7,8,9

Figure 1

Concerted Mechanism for the Sjj2' Reaction

^ R ^ CH^CH-C-rX* — ^— >- CH2 -CH=C

^ V ^ "R'

Figure 2

Ion Pair Mechanism for the S^2' Reaction

H2 c = c h - c - x xr'

A, A -I

R / A / h 2 c = c h - c +x \ H 2 C - C — C R •h ' x - XR' A:

y R h 2 c -c h = c :

n « * 127

If k_-^ is much greater than k 2 , the reaction would display

second order kinetics. The regioselectivity of the reaction

is explained by assuming that k_^ is also much greater than

kg. Much of the data available on the reactions can be

interpreted by either the concerted or the ion pair mechanism.

Recently, Georgoulis and Ville^ ’ ^ presented evidence for

direct attack by a neutral or a polarized species which is

less ionized than an intimate ion pair.

Because of its synthetic importance, the stereochemistry

of the Sj^2' reaction has received much attention. The direc tion of nucleophilic attack determines product stereochemistry.

The nucleophile can approach the substrate on the same side

as the leaving group (syn attack) or on the opposite side

(anti attack). The original definitive work on this problem O was by Stork and WhiteJ in 1952. At the time, they reported

that for trans-6-alkyl-2-cyclohexene-l- yl 2,6-dichloro-'

benzoate esters, attack by a nucleophile (piperidine) occurs

stereospecifically with the nucleophile approaching ex

clusively syn to the leaving group. 128

Figure 3. S^2' Reactions

1. Chiche, Coste, Christol and Plenat 14 0 0 C-OR C-OR

N olOH HjO/ MeOH

.C-OR C-OR COOH •CO OH

2. ICirmse, Scheidt and V a t e r ^ MeO H, £1 H, .D ,0 ?r»'D / H »" 1 Na.OMe r\ 1 mil Q H‘ MeO 'Cl H D MeO

3. Ikola and Ganem16

OAc AcOLi

OAc 129

Figure 3. S^2' Reactions

4. Welch, Hagan, White, Fleming and T r o t t e r ^

5. Martel, Toromanoff, Mathieu and Nomine 18 0H

////

6. Stork and White^

~ o — « - o T T ' V 0 7. Uebel, Milaszewski and Arlt^

Ph SPh SPh

^ \ ^ o c h 2c f 3 c f3ch2oh > u

3 yh / anti ^ I Figure 3. S^2' Reactions

7. Uebel, Milaszewski and Arlt

PhS'V Ph

IOt 1 J l syn 8. Stork and Kreft-^ QR a N oi H 9 0 % sun PR . 2 % anti

H 100%

I + RS' 'SR h-buOH 9 1 H M P A 6 0 40

+ R S >

n-buOH RSNcc 65 RS‘Li + 50

Stork and Kreft^ Figure 3. SN2' Reactions

10. Stork and Schoofs^

O C A r "ntt CH

11a. Magid and Fruchey^l

R a N H H H \ / Ra_N H NC = ( / .CH, S.C-C. ■CH, D" V C"* H \ H H Cl s y n

H M R aN H R a N H V-KC— C u .c-c^ / \X .»Nn. o D C'^CH: H \ C — C H 3 D Cl H

lib. Oritani and Overton^

H H ch3 h 4- V/ C = C v o*r H 1 X D' C ^ C 2H 5 Ph 'n h 2 O C A r H 0

c h 3 h

/ \ ^ 0/) Ph NH —CHD— C /° / + S N 2 (2 0

\ H

Syn/ a n - t i 6 0 : 4 0 132

1 9 Recent work by Stork and Kreft and by Dobie and

Overton-^ has shown that the course of this reaction is more complex. Analysis of the reaction products using modem tools showed that some anti product was produced along with the syn. Further complicating interpretation of the data, it was found that the starting material was unstable to reaction conditions. Some underwent rearrangement prior to nucleophilic attack. Stork and Kreft repeated the work using the mesitoate ester (stable to reaction conditions) and reported that the reaction did occur with predominant syn attack by the nucleophile (2 percent anti attack).

Other stereochemical studies have been performed and these reactions are listed in Figure 3. A summary of the stereochemistry of the reactions is presented in Table I.

The reactions are classified according to reaction type, substrate, and transition state polarization of the nucleophile and leaving group (see discussion below).

Clearly, there is a definite preference for syn nucleophilic attack. Several theories have been invoked to account for this preference. In Reaction 8, with the piperidine nucleophile, hydrogen bonding between nucleo phile and leaving group is proposed to lead to the syn products.3 More generally, molecular orbital considera- O O 2 A- tions predict a syn preference. ° Anh, in a treatment similar to that of sigmatropic reactions, proposed that a syn attack was preferred if bond breaking of the leaving Table I. Summary of S^2' Reaction Tynes

Reaction tvne T.S. (inter- or Polarization Nucleonhilic Author intramolecular) Substrate Nu: L.G. Attack

1. Chiche et al. intra cyclic 6“ 6” svn

2. Kirmse et al. inter cyclic 6" 6" syn

3. Ikola et al. inter cyclic 5" 6" syn

4. Welch et al. intra cyclic 6" 6" syn

5. Martel et al. intra acyclic 6" 6" syn

6. Stork and White inter cyclic 6" 6~ syn

7. Uebel et a l . inter cyclic 6“ &+ syn - anti intra cyclic 6 6~ syn - anti

8. Stork and Kreft inter a,a' cyclic 6+ 6" syn inter b,b' cyclic 6~ syn, anti

9. Stork and Kreft intra acyclic 6“ 6" anti

10. Stork and Schoofs intra acyclic 6' 6" anti

11a. Magid and Fruchey inter acyclic 6+ 6" syn lib. Oritani and Overton inter acyclic 6+ 6" 60% syn 134

group begins prior to bond formation with the nucleophile

(non-synchronous). A synchronous reaction would lead to

anti attack. Liotta^^ uses an "orbital distortion technique"

to justify syn attack. In cyclohexenyl systems, it has been

proposed that the leaving group must be axial and that the

leaving group and nucleophile must be coparallel with the

it system. A conformational bias to avoid the boat conforma- n fi tion forces syn attack.

Yates et al.27 postulated "that the allylic framework

plus the nucleophile and leaving group constitute a 6 tt

electron system which will be more stable in a cis geometry

(Huckel aromatic system) than in a trans geometry." To sup port their hypothesis, they performed ah initio (ST0-4G)

calculations. They chose two different nucleophiles, OH-

and NH 3 (L.G. = F-) which were representative of transition

states where the nucleophile and leaving group were either

both negatively polarized (OH-) or oppositely polarized

(NH3 ). They found that although syn attack was favored by

tt overlap factors for both nucleophiles, in the case of

the OH- nucleophile, the anti transition state was actually

lower in energy. They concluded that electrostatic inter

actions were dominating the stereochemical course of the

reaction even though tt overlap factors were present.

The stereochemistry of Reactions 6 -lla can also be

interpreted as being controlled by transition state electro

static interactions. Looking at the cyclohexenyl system 135

as a model, four possible transition state situations can

be envisioned (Figure 4). In Case I, the electrostatic

interactions favor syn attack; in Case II, anti attack is

favored; in Case III; syn attack is favored; and in Case IV,

anti attack is favored.

With the exception of the work by Uebel el: al. , all

of the literature examples involve a leaving group which

is negatively polarized in the transition state (Cases I and

II). It was decided to examine the stereochemistry of a

system wherein the leaving group had some degree of oosi-

tive charge in the transition state (Cases III and IV).

For this purpose, (trans-6-t-butyl-2-cyclohexen-l-vl)tri-

methylammonium tetraboroflurate (1) was synthesized.

Piperidine and sodium propanethiolate were chosen as nucleo

philes so that the stereochemical results using this sub

strate could be directly compared with those results of

Stork and Kreft.^

i. 136

Figure 4. Possible Transition State Interactions between

Nucleophilie and Leaving Group

Nu: = neutral

L.G. = negative x V

L I

a Nu: = negative

x'S- L.G. = negative E

Nu: = negative

L.G. = neutral

P— - " N a ’*S*

m Nu: = neutral

L.G. = negative 137

/ hiiK N v

JlIH

C H ?C H 2 CH 2S N q- n (c h 3)3 b f h'

Reaction a would lead toaCase IV transition state whereas Reaction b would be an example of a Case III

- N — s c h 2c h 2c h 3 H

n (c h 3)3

transition state. If electrostatic interactions are important in directing nucleophilic attack, Reaction a 138

would be expected to give predominantly anti products while Reaction b should give predominantly syn products. RESULTS AND DISCUSSION

The tetrafluoroborate 1 was synthesized via a ten

step procedure which is shown in Figure 5. Since the

reaction procedures were taken from the literature, this

discussion will include only those aspects of the synthesis

which deal with the isomeric purity of the products.

In order to perform product studies on the substrate

1, it was necessary to have the pure trans isomer. Prior

to Reaction e (Figure 5), the question of cis and trans

product ratios was not relevant. However, when ketone

was reduced, it was expected that both cis and trans isomers

would be produced. Hydrogenation of 1_ (Pd on C) produced

a mixture of cis(13%) and trans (87%) -4-t-butylcyclo-

hexanol. The ^H-NMR showed two signals for the carbinol proton at 3.96-3.84 and 3.52-3.2 ppm (6). The former signal

was assigned to the cis proton (due to its downfield posi

tion and smaller bandwidth) and the latter to the trans

proton.The spectrum was also checked against a spectrum

of an authentic mixture of cis and trans - 4-t-butylcyclo-

hexanol (Aldrich).

It was decided to continue the reaction sequence with O C the isomer mixture since preliminary experiments bv TollJJ

indicated that formation of 9 might occur stereoselectively.

Compound 9_ was formed via a rearrangement of the imidate 8.

139 Figure 5. Synthesis of

n (c h 3x b f 4

OH a ■> 0

(2) (3) (4)

-K >/w OH-^ \ ___/12% cis (5) (6) (7)

-K >-o-c-cci3 H"" 1 \ / 100% (8) HN HNCCCI3 n h 2 (9) 0 (10)

H11m — Jiiim

n (c h 0 2 n (c h 3)3 bfh~ 00 (I)

a. Na 2 Cr07 , H 2 SO4 2 8 ; b. Br2 , H0 CH 2 CH2 0 H, HBr2 9 ; c. NaOMe, DMSO29; d. H 2 S04 , Dioxane, H 2 0; e. LiAH4 , A1C1 3 3 0 f. N e CCC13 , NaH31; g. Xylene, reflux; h. NaOH, E+OH/H2 0 3 1 i. HCOOH, H 2 CO3 2 ; j_. (CH3 )3 0+ BF4-, CH 2 C1 2 3 3 141

During the reaction, substantial amounts of trichloro-

acetamide were also formed. Purification of 9_ was diffi

cult due to cocrystallization of the trichloroacetamide;

and, therefore, the corresponding crystalline mixture was

hydrolyzed (Reaction h) . Upon work up, pure 10 was isolated

and purified by distillation.

The stereochemistry of 10 was determined by hydro- 36 genation. Booth el: al. found that cis and trans-2-

alkylcyclohexylamines could be differentiated by their

•*-H-NMR signals for the proton on the amino-substituted

carbon. In benzene, the trans proton was not clearly

resolved from the ring protons. The cis proton was found

to be a discreet signal in the region of 2.9-3.2 ppm (6).

The ^H-NMR (benzene) spectrum of 10 (H2 ) showed no evidence

of a signal in the region downfield from 2.9 ppm. There

fore, it appears that 10 (H2 ) is the pure trans isomer and that

in Reaction g only the trans isomer of 9 successfully under went rearrangement.

The possibility for isomerization of the amine under

the acidic conditions of Reaction i was also recognized

and the final product 1^ was examined again for isomeric

purity. ^-H-NMR (260 mHz) and -^C-NMR spectra both indi

cated only one isomer (within detectable limits, l-27o) . 142

Reaction of Substrate (1) with Piperidine

lllll

+ +

13. 15.

-f-

— (u.<^ V^N(CH3)i

n (c h 3)2

In a solvolytic reaction, a mixture of distilled piperidine (2.5 mL) and 1 (0.5 g) was sealed in a glass tube and heated to 131° for 24 h. During this time the liquid turned yellow, and upon cooling, needle crystals precipitated.

The tube was opened and ether added to precipitate any tetrafluoroborate salts. After filtering, the solids were tentatively identified as piperidinium tetrafluoroborate from the ^-H-NMR snectrum. There was no indication of a 143

methyl signal from the t-butyl group of 1 and it was assumed that the reaction had gone to completion.

To analyze the piperidinolysis products, the ether was distilled from the reaction solution at atmospheric pressure

The concentrated solution was gas chromatographed on a 8 ft x 1/8 in column (5% SE-30; initial temperature 140°, 7 min; program to 175° at 6°/min). The gas chromatogram of the piperidinolysis products is presented in Figure 6. Data from the gas chromatogram and subsequent GC/MS analysis

(Table II) was obtained in the following manner. The experi ment was conducted twice to insure reproducibility. For each experiment, peak areas for three seoarate injections were determined by planimeter. The relative peak areas were calculated from the total area (Peaks 1-6) , although it is recognized that the FID relative responses for Peaks

1-3 and Peaks 4-6 would be different. Additionally, Peaks

5 and 6 are given as a percentage of their individual total area. The identification of Peaks 5 and 6 is dis cussed below, and the identification of Peaks 1-3 will be discussed at a later time.

From the M-^ ion at m/e 221, it was assumed that Peaks

5 and 6 were due to 12^ and 1_3 or another isomer (Peak 4 was not concentrated enough for MS analysis). The mixture was separated by preparative TLC (30% Methanol/Cl^C^;

2 elutions) and a small amount of pure Peak 5 was isolated.

The 1-H-NMR spectrum showed signals for the pioeridinyl Figure 6. Gas Chromatogram of the Piperidinolysis Reaction Products 145

Table II. GC Data for Piperidinolysis

Relative Retention Relative Time Assignment Peak Areas 3eak (min)a M+ (compound) (%)b

1 3.75 181 16 15

2 4.60 181 17 5

3 5.30 181 17 3

4 13.80 -- <1

5 14.95 221 12 62 (80)*

6 16.40 221 13 15 (20)*

aRetention times are recorded relative to the beginning of the piperidine peak

^Determined with an FID detector

*7o composition of Peaks 5 and 6 146

protons (1.5 and 2.5 ppm (6)), for the vinyl protons (5.85

ppm, 2H), and for the t-butyl protons (0.9 ppm, 9H). A

•^C-NMR spectrum had eleven signals. These data are con

sistent with structures 12 and 13.

Final assignment of the cis and trans configurations

was to be determined by GC retention times. Retention times

in the literature are reported only for the saturated

analogs 14 and 1 5 ^ so the reaction mixture was hydrogenated.

Prior to hydrogenation, butanol (10 mL) was added to the

reaction solution which was then distilled to a volume of

ca. 2 mL. This procedure was performed to remove the

majority of the piperidine from the solution. The remaining

solution was hydrogenated (Pd. on C; ethanol) and then gas

chromatographed.

The gas chromatogram for the hydrogenation is presented

in Figure 7. Two major peaks are evident which both have

an M+ ion at the expected m/e 223. The peak area ratio

of 60:40 shows that some isomerization has occurred during

■the hydrogenation. An injection of a mixture of the satu

rated and unsaturated amines is shown in Figure 8. In

between the two major peaks of Figure 7 are two peaks which

are too small for MS analysis. It is possible that a small

amount of the unsaturated products is present due to in

complete hydrogenation . However, the possibility of simple

S^2 substitution to give 18 also exists and the NMR data

reported above would also be consistent with 18. n: iifij mm

Figure 3. Gas Chromatogram of Coiniection of 12_, 13>, 14, and 15^ from S ^ 1 Reaction 148 The relative retention times reported by Stork and

i 2 Kreft for the four isomers are trans 19 < cis 19 < cis 15^ < trans 14.

In order to confirm the structures of the two isomers at m/e 221, alternate syntheses of 12% 13, 14, and L5 were undertaken. 4-t-Butyl-2-cvclohexen-l-ol (7_) was available as starting material for both 12_ and 1 3 .

S'n2 >

7 20 r\ | i "xOl « - i - — Cl l3/o CIS 87 % trans 13 % cis

13 % \ 2_ (trans) 8 7 % 13 (cis)

The isomeric composition of this compound was known to be 13% cis. It was esterified with o-dichlorobenzoic acid to give 20_. Simple S^2 substitution on 20 with piperidine 150

(solvolysis, 131^) was expected to give an inverted product mixture of 13%, trans and 87% cis isomer.

Chromatography of this mixture did give two peaks

(Figure 9). Based upon the theoretical course of an S^2 reaction, the larger peak having the shorter retention time was assigned to the cis isomer L3 and the smaller peak was assigned to L2. The oil was distilled and purified by preparative TLC. The trans isomer could not be separated from another impurity; however, the pure cis isomer was cleanly separated. The NMR spectra of 13 were identical with Peak 5 (Figure 6). GC coinjection of a solution of the synthetic 2^ and _3 and the S^2 ’ reaction products gave only one set of peaks (Figure 10). Peak 6 was therefore tentatively assumed to be the trans isomer 2.

As further proof of these assignments, 14 and L5 were synthetized. Reduction of enamine YL leads to satu rated amines 14 and 15^, as determined by -^C-NMR spectro

scopy and gas chromatography (Figure 11).

14 and r5 ■HFOi ■HFOi i i H -j-H -iM

Figure 9. Gas Chromatogram of the SN2 Products 12 and 13.

Ul Figure 10. Gas Chromatogram of the Coinjection of Products 12 and 13 and 152 the S^2' Products.

154

GC coinjection of synthetic T2, L3, 14 and 15^ (Figure 12) led

to a chromatogram similar to that of the saturated and un

saturated S^j21 products (Figure 8). Coinjection of synthetic

14 and 15^ and the hydrogenated Sn 2 ' reaction mixture led toi

total overlap of the major peaks (Figure 13).

Stork and Kreft^ reported that at 190° (5%. SE-30) ,

the GC retention times for the saturated amines were: trans

19, 5.8 min; cis 19_, 6.2 min; cis 15^ 7.2 min; and trans 14,

8 min. Since the cis and trans isomers 14 and 15^ are clearly

separated in our data, it would be expected that any S^2 product, cis 19_, formed along with the S*,j2' products would

be eluted as a separate peak. It is possible that Peak 6

(Figure 6) which appears in both the S^2' and the alternate

synthesis piperidinolyses is the S^2 product 19. However,

it doesn't seem very likely that in the alternate synthesis,

substitution at the hindered Y-carbon could compete with

direct S^2 attack. Therefore, this assignment is not pro posed. To summarize, Peaks 5 and 6_ (Figure 6) were assigned

structures and L2, respectively, after comparison with

authentic saturated and unsaturated samples.

The three initial peaks in Figure 6 had the same molecular weight as N-(trans-6-t-Butyl-2-cyclohexen-l-yl)-

N,N-dimethy1amine. Peak 1 was isolated by preparative TLC

(30% methanol / CH2 CI2 ; 2 elutions) and identified by com parison of GC and 1-H-NMR data from an authentic sample. Figure 12. Gas Chromatogram of Coiniection of 12, 1 3 , 14, and 15 (alternate svnthesis). T T i J i m l

! I I l '

i ! ! i !

H+r

H-i-Hfjjt}I . I j i i.i.i r { ! 1)1 i-M-i-M-j- ■ ;-1—i- -U -i-UU . L U J J -LiJ f_L_Ll h i i ' i nil"i i 11 h if vrii J.l'niltfij i i'. ;'i'!T] u UuJ iLU ... i.i.i.i L j ! I 1 I ■ . 1 • • ■ • i; : ! i ! !I I * . i f 7 I 7 ! l ! ! ; - 1 1 | ;! !_i 1 ! ! I I l.| , j j J j J i . | ( , : - \ • j I ! I ‘ j ; . j I | ; 1 ; , I ■ , ; , ; ; ■ ; l | ; ■ ; | . | ; : • . | Figure 13. Gas Chromatogram of the Coinjection of" the Alternate'Synthesi 14 and 15^ and the Sj^21 Hydrogenated Products. Lni-* 157

Its formation can be explained by piperidine displacement on

a methyl group of 1. Formation of Peaks 2 and 3 can only

be explained if a rearrangement process is occurring.

H-O* H ^ 3

N (CH3)3 + .c •*"N(c h 3)5

If this dissociation can occur under the reaction

conditions, then it is possible that 12^ and 13 are also

forming through a dissociative mechanism.

H iii< :N(CH,V - — |«ii^ n ^ (c h 3)3

22 m ajor n (c h A

(cis) 13 1)5

21 m1n or (trans) J_2_ 158

A tight amine-cation pair in a solvent cage would be expected to recombine to give a high rate of 22/23. Sub sequent S^2 attack on 13 or 14 would give the observed products with the correct relative stereochemical ratios.

Such an internal return mechanism would be expected to exhibit 1st order kinetics because dissociation to the intermediate would be the slow step.

In order to determine whether 12 and 1_3 were being formed by a three-step process having a dissociated inter mediate or via direct S-^2' attack by the nucleophile, a kinetic experiment was performed.

The experiment was designed to run equal concentrations of substrate (1) with two equivalents or four equivalents of nucleophile. If the dissociation mechanism (4) is oc curring, it would be expected to follow 1st order kinetics and the rate should be independent of the nucleophile con centration. The bimolecular mechanism would lead to reac tion of four equivalents of nucleophile at twice the rate of the two equivalents reactions.

The substrate 1_ (25 mg) was weighed into each of four ampoules. Stock solutions of 0.35 M piperidine in xylene

(2 equivalents) and 0.7 M piperidine in xylene (4 equiva lents) were prepared. The former solution (0.5 mL) was added to two of the ampoules as was the latter. One of the two equivalents ampoules and one of the four equivalents 159

ampoules was heated for 6 h at 131° and the other set was heated for 11 h at 131°.

Table III gives the GC relative peak areas for the two experiments. The correlation of the % trans isomer formed in these experiments (~20%) with the % trans isomer formed in the solvolysis experiment (20%) is an indication that the two reactions are proceeding by similar pathways

(Problems with the gas chromatograph may account for the slightly higher values in the 11 h experiment). Since the reaction with four equivalents occurs at almost twice the rate of that with two equivalents, it is concluded that a bimolecular S^2' mechanism is more consistent with the data than the mechanism (4). Overall, the reaction pro ceeded with predominant anti attack by the nucleophile via an Sjq2' process. 160

Table III. Kinetic Experiment

Relative Piperidine Amounts 7> of Concentration______of Products Trans Isomer

2 eq 1 19

4 eq 1.83 20

llh

2 eq 1 21

4 eq 1.86 24 161

Reaction of Substrate 1 with Sodium Propanethiolate

s c h 2c h 2 c h 3

CH3 CH2CHaS~r< ) +

n (c h 3)3

i'tiSCH2CH2CH3 B F 4" H ...

To a mixture of substrate 1 (0.4 g) and n-butanol

(5 mL) was added a mixture of sodium propanethiolate (2 equivalents) in butanol (3 mL). The mixture was refluxed for five hours and then cooled in an ice bath. Pentane

(3 mL) was added to precipitate additional dissolved NaBtfy or sodium propanethiolate. ^H-NMR (acetone) of these solids did not display a t-butyl methyl signal and it was assumed that the reaction went to completion. The solution of reaction products was gas chromatographed on a 5 percent

FFAP 8 ft x 1/8 in column (65°, 23 min; program to 120° at 12°/min).

Gas Chromatography (Figure 14) revealed three major peaks. Their retention times and M+ ions are listed in

Table IV. Values for the \ areas of Peaks 2 and 3 were p j a B f j y

Figure 14. Gas Chromatogram of the Stq2' Reaction of Sodium Propanethiolate and Substrate (1).

ON 163

obtained by the same procedure used in the piperidinolysis experiment. Stork and Kreft^^ report that GC retention times of 24 and 25 follow the order 24 > 25^. Therefore,

Peak 2 was tentatively assigned to cis 25^ and Peak 3 was assigned to trans 2 4 .

The reaction mixture was separated bv preoarative TLC

(pentane; 3 elutions) and gave four bands. The first elu tion component was identified as dipropyl disulfide and corresponded to Peak 1 (Figure 14). The third eluting component was identified as N-(trans-6-t-Butyl-2-cyclohexen-

1-yl)-N,N-dimethylamine (11). Its formation was assumed to be analogous to that in the piperidinolysis, with propane thiolate displacing on a methyl group of 1. The fourth eluting compound was unidentified.

The second eluting component was shown to be a mixture of Peaks 2 and 3 by gas chromatography. The ^H-NMR spectrum was consistent with 24 or 25^ and the -^C-NMR showed the expected eleven major signals. Many of these signals were doubled and these extra signals were assumed to be due to the minor isomer.

Confirmation of the assignments was obtained by an alternate synthesis of 24 and 2b from the dichlorobenzoate ester 2() via simple S^2 substitution with oropanethiolate.

Since starting material was 87 percent trans isomer, it was predicted that the products would be formed in an inverted ratio, . 13 percent trans and 87 percent cis. The 164

\rfscH2CH2CH3 Cl \ = s _24 13%

HO r 9 CH3CHaCH25NoL 20 Cl 87 % trans /-----V 13% c;s -jim/ VSCH2CH2CH5

2 5 87 %

l^C-NMR spectrum gave the same total number of peaks. How

ever, in cases where the doubling of signals had occurred,

the magnitude of the signals was reversed. This was as

expected and each signal could now be assigned as resulting

from the cis or trans isomer.

Using theoretical SN2 considerations to interpret

Reaction (7) and based upon a comparison of -^C-NMR spectra

for Reactions (6) and (7), Peak 2 was assigned to Z5 and

Peak 3 was assigned to 24. The possibility that the ob

served products were the result of the dissociation mech

anism (5) was discounted because such a mechanism would be

expected to give the opposite product ratios. The S^2'

reaction proceeded with predominant syn attack by the

thiolate nucleophile. 165

In repeating the experiment, a fourth Peak with a GC retention time of 33.7 and an MT*- ion at 136 was observed in the GC chromatogram ( 8 percent of total area). This peak was assumed to be due to an elimination reaction instead of substitution. An explanation of why elimination occurred in the second experiment but not the first might involve the reaction procedure. In generating the propanethiolate anion, excess n-propyl thiol was used to insure complete reaction with sodium methoxide. If an error was made, excess sodium methoxide might have induced elimination.

Another possibility might be that the pot temperature in the second experiment was slightly higher and that energy difference may have been enough to have competitive elimina tion with RS- as the base. Regardless of the reason, it is believed that the elimination reaction was independent of the substitution pathway since the product ratios for the two experiments (7 percent for the first experiment,

9 percent for the second) remained essentially the same. 166

Table IV. GC Data for the Propanethiolate Nucleophile

Relative Retention Assignment % of Peak Time (min)a (compound) Total Area

19.7 150 dipropyl disulfide

2 47.1 212 25 8

3 52.5 212 24 92

aRetention times are recorded relative to the beginning of the butanol peak. CONCLUSION

Theoretical considerations, according to the majority of the literature, ^3-27 suggest that syn attack is favored in Sjg21 reactions. These theories, briefly mentioned in the Introduction, could be used to explain the observed stereochemistry of our reaction of propanethiolate with substrate 1. However, the observed stereochemistry of the piperidinolysis is contrary to these hypotheses. In par ticular it brings into question the cyclohexenyl conforma tional bias for syn attack.

The theory proposed by Yates et al. ,27 suggested that electrostatic interactions in the transition state control the direction of nucleophilic attack in S^j21 reactions. This would explain the anti product predominance in the piperidin olysis experiment. It can also be invoked to explain the results with the propanethiolate nucleophile.

Our results are complementary to those of Stork and 12 Kreft. Using a cyclohexenyl substrate, we chose a leaving group which would exhibit polarization in the transition state opposite to that of Stork and Kreft. ye have per formed the experiments using the same nucleophiles under the same conditions. The transition state theory of Yates el: al.27 predicts that we would obtain the reverse of the Stork and Kreft results. This is essentially what we have seen.

167 168

In light of our results, it appears that the importance of transition electrostatic interactions cannot be overlooked in a discussion of S^2' stereochemistry. EXPERIMENTAL

General. Proton magnetic resonance (^H-NMR) spectra were recorded on a JEOL JNH-MH 100 or a Varian A-60 spectro- 1 0 meter. C-NMR were recorded on a JEOL FX90Q Fourier Trans

form NMR spectrometer. Unless otherwise stated, the spectra were run in CDCI3 and chemical shift are reported in parts

per million (6 ) downfield from tetramethylsilane as an

internal standard.

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

337 spectrometer as neat liquids, solutions, or in a KBr

pellet.

Gas chromatography was performed on a Perkin-Elmer 881

gas chromatograph. This instrument was interfaced with

a Hitachi-Perkin-Elmer RMU-6 E mass spectrometer for GC/MS

analyses.

High resolution mass spectra were obtained from the

MIT NIH facility on a CEC 21-110B high resolution mass

spectrometer.

Melting points were determined on a Thomas-Hoover

Uni-Melt apparatus and are uncorrected.

Elemental Analyses were performed at the University

of New Hampshire on an F&M 185 instrument.

Materials. Thin layer chromatography was performed

on precoated silica gel F-254 plates purchased from E. Merck.

Preparative TLC was performed on glass plates prepared with

169 170

"Silica Gel 60 PF-254" purchased from E. Merck.

All chemicals and reagents were used as purchased unless otherwise noted.

Solvent evaporation at reduced pressure was carried out on a rotary evaporator unless otherwise noted.

9 Q 4-t-Butylcyclohexanone (3)t An oxidizing solution of

Na2 Cr0 y •2 H 2 O (60 g; 0.24 moles), 1^0 (300 mL) and cone.

H 2 SO4 (47.5 mL) was prepared and the solution was cooled to 30°. A slurry of 4-t-butylcyclohexanol (75 g; 0.48 moles) and H 2 O (240 mL) was prepared and two-thirds of the dichromate solution was added. The mixture was heated to

50° to initiate the reaction. The remainder of the dichromate solution was added and the mixture was kept at 55-60°, with cooling as necessary. After the tempera ture no longer rose above 55°, the mixture was stirred until all the alcohol dissolved, and then allowed to stand for

2 h. The solids were filtered, washed with 150 mL HgO and dissolved in ether. The aqueous filtrate was extracted

3 times with ether and all ether layers were combined. The ether solution was washed with H 2 O, NaHCOg solution and

NaCl solution and dried (MgSOA.) . The ether was removed at reduced pressure yielding 61.9 g (84%) of colorless crystals (mp 45.5-48°; lit. mp 49-50°37). The iH-NMR spectrum agreed with the spectrum reported by Toll.35 171

4-t-Butyl-2-bromocyclohexanone Ethylene Ketal (4). The procedure used was modeled after that reported by Garbisch.^

To a 1 L 3-neck round bottom flask equipped with a drying

tube, a thermometer, a magnetic stirrer and a dropping

funnel, was added 4-t-butylcyclohexanone (65.8 g; 0.43 moles) and ethylene glycol (230 mL) . HBr (1.5 g) was bubbled into ethylene glycol (300 mL), the solution was added to the flask and the reaction mixture heated to 50-

55°. Bromine (68.5 g; 0.43 moles) was added dropwise at a rate which always maintained a faint bromine color while the reaction temperature was kept at 50-55°. Extra bromine was then added until the color nersisted for several minutes.

The reaction mixture was poured into a stirred mixture of

Na2 C0 3 (107 g) in pentane (400 mL). After stirring several minutes, H 2 O (500 mL) was added. The layers were separated and the aqueous layer was washed with pentane (100 mL).

The combined pentane layers were dried (MgS04) and evaporated at reduced pressure to yield 118.3 g of a pale yellow oil. The procedure was repeated with the ketone

(64.5 g; 0.42 moles) yielding 109.3 g of oil. Vacuum dis tillation yielded 145.6 g (627o) of a colorless liquid 29 (bp 121-125°, 0.3 mm; lit. bp 94-96°, 0.05 ram ) . The

1-H-NMR spectrum agreed with the spectrum reported by T o l l . 35

4-t-Butyl-2-cyclohexen-l-one Ethylene Ketal(5).^ To a solution of 4-t-butyl-2-bromocyohexanone ethylene ketal 172

(77.8 g; 0.28 moles) and dry DMSO (250 mL) was added sodium methoxide (52.5 g; 0.97 moles). The mixture was heated and

stirred on an oil bath for 15 h. It was then cooled and poured into 500 mL NaCL solution. The mixture was extracted with pentane twice, dried (MgSO^.) and evaporated at atmospheric pressure and room temperature. The light brown oil was distilled to yield 27.9 g (51%) of a color less oil (bp 81-89°, 0.3 mm; lit bp 76-78°, 0.5 m m ^ ) .

The 1-H-NMR spectrum agreed with the spectrum reported by

Toll . 3 5 29 4 - 1 -Butyl-2-cyclohexen-l-one(6 ). To a solution of

4-t-butyl-2-cyclohexen-l-one ethylene ketal (53.5 g; 0.27 moles) in dioxane (120 mL) was added 15%, H 2 SO4 (180 m L ) .

The mixture was stirred at room temperature for 30 min and extracted with ether three times. The ether layers were washed with H^O, NaHCC>3 solution and H 2 O and then dried

(MgSO^.) . The ether was evaporated at reduced pressure and the residual oil was distilled to yield 29 g of a colorless oil (bp 102-104°, 3.7 mm; lit. bp 88-89°, 4.6mm2^). Com parison of the ^-H-NMR spectrum with the spectrum reported by Toll2^ showed the product to be contaminated, with a second t-butyl signal appearing at 1.1 ppm. Integration of the two-t-butyl groups showed the product to be 93%, pure

(6 6 % overall yield). Further purification was not attempted at this stage. 173

30 4-t-Butyl-2-cyclohexen-l-ol (7) • To an ice-cooled suspension of LiAlH^ (5.5 g; 0.14 moles) in dry ether

(500 mL) was added AlCl^ (6.09 g; 0.05 moles). The mix ture was stirred under nitrogen for 1 0 min and allowed to come to room temperature. A solution of 4-t-butyl-

2-cyclohexen-l-one (26.3 g; 0.17 moles; 94% pure by

■^H-NMR) in dry ether (100 mL) was added and the mixture was stirred under nitrogen at room temperature for 1 h.

Aqueous 15% KOH (35mL) was added dropwise and stirring continued for 10 min. The solids were filtered, washed with ether, and the combined ether filtrates dried

(MgSO^). The ether was evaporated at atmospheric pres sure to give 19.4 g (74%; 94% pure by ^H-NMR integration of _t-butyl signals) of a waxy white solid (mp 41.8-43.8°).

The ^H-NMR spectrum agreed with the spectrum reported by 35 Toll. A portion of the alcohol (120 mg) was hydro genated with 5% Pd on C (20 mg) and abs. ethanol (20 mL) i at 2 atm for 1% h. H-NMR showed the alcohol to be a mixture of cis and trans isomers, 13% and 87%, respec tively, by integration of the carbinol proton.

4-t-Butyl-2-cyclohexen-l-yl 2,2,2-Trichloroeth-

_ . animidate (8 ). In a 3-neck 100 mL round bottom flask equipped with a magnetic stirrer, thermometer, nitrogen inlet and a septum, a solution of trichloroacetonitrile

(3.2 g; 0.022 moles) in dry ether (32 mL) was cooled to

-5 to 0° under nitrogen. In a separate flask,

4-t-butyl-2-cyclohexenol (7)(4.87 g; 0.032 moles) was 174 dissolved in dry THF (32 mL). A hexane slurry of

NaH (0.229 g; 0.0048 moles; 50% oil dispersion washed with hexane twice) was added, the flask capped with a

septum and the mixture stirred 5 min under nitrogen. The alkoxide solution was added dropwise by needle transfer at a rate to maintain the temperature below 0°. The re action was stirred for 1.5 h and then concentrated at reduced pressure at <25°. Pentane (150 mL containing

6 drops of methanol) was added, the mixture shaken vigorously for 1 min and then gravity filtered. Evapora tion of the pentane at reduced pressure gave 8.3 g of 8

(8 8 %). The crude imidate was not further purified. The

^H-NMR spectrum agreed with the spectrum reported by

Toll . 3 5

2,2,2-Trichloro-N-(6-t-Butyl-2-cyclohexen-l-yl) acetamide (9). The crude trichloroethanimidate 8

(9.5 g; 0.032 moles) was refluxed for 3h in xylene

(110 mL). The xylene was distilled (short path) at

0.3 mm Hg and a white solid mixed with a brown oil re mained. The mixture was recrystallized from hexane to i yield 3.1 g of 9. H-NMR integration of the amide proton signal at 6 . 6 ppm(S) showed that trichloroacetamide (20%)had cocrystallized with product. The product was not further purified (overall yield: 43%). A small amount was purified for analytical purposes (mp 142 -

142.5; lit. mp 140-14235). IR (KBr) 3280, 3025, 2960,

2900, 1680, 1620, 1185, 1060, 875, 835, 755, 715, 695, 175

650. 1 H-NMR (60 mHz, CDC13) 6 ca^ 6 .6 (b, 1H, H-N),

6.2 - 5.3 (b, 2H, H-C = C ) , ca^ 4.6(b, 1H, H-C-N),

2.3 - 1.1(5H, H-C, H 2 C), 1.0 (s, 9H, -CH3>. Anal. Calcd

for C 1 2 H 1 8 N0CC13 : C, 48.28; H, 6.03; N, 4.69.

Found: C, ; H, ; N, . See page 187.

N-(trans-6-t-Butyl-2-cyclohexen-l-yl)amine (10). ^

2,2,2-Trichloro-N-(6-t-butyl-2-cyclohexen-l-yl)

acetamide (8.3 g; 0.28 moles), 95% ethanol (150 mL) and aq, 6 N NaOH (140 mL) were stirred under nitrogen for 30 h.

The reaction mixture was extracted with ether (300, 100,

100 mL portions) and the combined organic phase was concentrated at reduced pressure. The residue was taken up in boiling hexane and filtered. The filtrate was evaporated at reduced pressure to yield 3.3 g of an orange oil. Distillation of the oil yielded 2.2 g (51%) of 10 (bp 53°, 0.3 mm). IR (Neat) 3370, 3300, 3050,

2920, 1650, 1590, 1460, 1350, 1090, 1045, 930, 855, 790,

720, 670. 1 H-NMR (60 mHz, CDC13) 66.0-5.4 (2H, H-C=C), ca. 3.3(b, 1H, H-C-N), 1.3 (s, 2H, H 2 N ) , 1.0 (s, 9H,

-c h 3 ).

To check the isomeric purity of the product, the amine (100 mg) was hydrogenated at 2 atm with Pd on

C (20 mg) in abs. ethanol (15 mL) for 2 h. ^H-NMR

(CgDg) showed the amine to be the pure trans isomer with Ofi no detectable signal at >2.96 for the cis proton. N-(trans-6-t-Butyl-2-cyclohexen-l-yl)-N,N- dimethylamine (_11) . The cyclohexenylamine _10 (1.49 g; 176

0.01 moles) was cooled to 0° in an ice bath. Formic acid

(8 8 %; 1.14 mL) was added to the amine followed by formal

dehyde (37%; 1.74 mL). The flask was equipped with a

stirrer and reflux condenser, and heated for 24 h at 80°.

The reaction mixture was then cooled and 6 N HC1 (3 mL)

was added. The non-basic reaction products were extracted with ether and discarded. The solution was made basic with 50% NaOH and extracted with ether (10, $ 5 mL por

tions). The ether layer was washed with IL^O (5 mL) and

dried (Na2 CC>3). Evaporation of the solvent at reduced pressure gave 1.24 g (71%) of a yellow oil, 11. IR (Neat)

3010, 2850, 1660, 1440, 1260, 1095, 970, 900, 845, 795,

730, 655. XH-NMR (60 mHz, CDC13) 66.1-5.4 (2H, H-C=C),

ca. 3.0 (b, 1H, H-C-N), 2.2(s, 6 H, H 3 C-N), 2.1-1.2 (5H,

H-C, H 2 C), 0.95(s, 9H, -CH3 ).

(trans-6-t-Butyl-2-cyclohexen-l-yl)trimethylammonium _

______Tetrafluoroborate (1)._ T a glove n bag,-u tnmethyloxonium4. • 4--u i • ~ tetrafluoroborate (1 . 2 1 g; 0.0082 moles) was weighed and

CH2 Cl 2 ( 2 0 mL; distilled from 1*2^5’ stored over K 2 CO 3 ) was added. The flask was capped and the N,N-dimethylamine 11

(1.24 g; 0.007 moles) was injected by syringe. The re action solution was stirred for 2 h at room temperature.

After solvent removal at reduced pressure, the solids were purified by several precipitations from acetone with ether to yield 1.4 g (71%) of white crystals, _1 (mp 168-169°).

IR (KBr) 3050, 2960, 1650, 1470, 1400, 1350, 1010-1110,

890, 855, 725, 700, 580, 530. 1 H-NMR (60 mHz, CD 3 C0CD3 ) 177

6 6.7 - 5.9 (2H, H-C =C) , ca^ 4.0 (b, 1H, H-C-N), 3.2

(s, 9H, H3C-N), 1.0 (s, 9H, -CH3>. 13C-NMR (CD3COCD3>

6 141.1, 119.1, 69.7, 52.2, 52.0, 51.8, 39.5, 34.9, 29.0,

28.0, 22.9, 20.4. Anal. Calcd for ^ 3 ^ 2 6 ^ ^ 4 :

C, 55.14; H, 9.25; N, 4.95. Found: C, 54.90;. H, 9.26;

N, 4.73.

S^2' Piperidinolysis of 1_. The reaction was modelled ------TO upon the reaction of Stork and Kreft . The tetrafluoro borate 1. (0.5 g; 0.0018 moles) and distilled piperidine

(2 mL) were sealed in a glass tube. The tube was heated at 131° (refluxing chlorobenzene bath) for 24 h and then allowed to cool, at which time needle crystals precipi tated. Upon opening, the vial contents were poured into an Erlenmeyer and then rinsed with 0.5 mL piperidine.

Ether (40 mL) was added to precipitate any remaining salts.

The mixture was filtered and the solids were examined immediately by ^H-NMR (CD3COCD3). They appeared to be piperidinium tetrafluoroborate and no starting material was present. Ether was distilled from the reaction filtrate at atmospheric pressure and the remaining solu tion was analyzed by GC using an 8 ft. x 1/8 in. column

(5% SE-30 on Chromasorb W, AW). The gas chromatogram

(140°, 7 min; program to 175° at 6°/ min) showed 6 peaks with the following retention times (relative to piperidine) and peak area ratios: 3.75(15%), 4.6(5%),

5.3(3%), 13.8(<1%) , 14.95(62%), 16.4(15%). MS analysis 178

showed the first three peaks to have an M+ ion at m/e 181

and the fifth and sixth peaks at m/e 2 2 1 .

Preparative TLC (30% methanol/CH^C^; 2 elutions):

No bands were visible to UV light so the plate was cut

based upon the values for the analytical TLC. Using

this procedure, the bands traveling 8 cm from the origin

(solvent travels 14 cm) and 12 cm from the origin were

isolated relatively pure (GC analysis). After work up

(ether extraction), the former band was identified as 1 (>

and the latter as 1J3. ‘'"H-NMR 13 (60 mHz, CDCl^)

6 6.15 - 5.6 (2H, H-C =C), ca^ 3.1(b, 1H, H-C-N),

2.7 - 2.35 (H2 C-N), 0.9 (s, 9H, -CH3). 1 3 C-NMR 6 132.3,

129.8, 58.3, 50.6, 45.1, 33.1, 27.6, 26.7, 24.8, 22.2,

21. 8 .

Hydrogenation procedure: Piperidine was removed

from the S^2 1 reaction mixture by codistilling with 10 mL butanol. The mixture was distilled at atmospheric pres

sure and the remaining butanol was removed at reduced

pressure. The mixture of amines (157 mg) was dissolved

in abs. ethanol (10 mL) and hydrogenated at 2 atm with

Pd on C (25 mg) for 7.5 h. An aliquot was withdrawn

and hydrogenation resumed for 3 h more. Gas chromatog

raphy of the hydrogenated aliquot showed two major peaks with retention times (relative to ethanol) of 25.5 min

(60% total peak area) and 31.8 min (40%). Each peak had

a small shoulder which persisted after 10.5 h of hydro

genation. 179

4-t-Butyl-2-cyclohexen-l-yl 2,6 -dichlorobenzoate

^ * 4-t-Butyl-2-cyclohexenol (87% trans) (1.83 g,

0.012 moles) was dissolved in dry pyridine (4 mL) and a solution of o-dichlorobenzoyl chloride (2.49 g, 0.012 moles) in pyridine (3 mL) was added. Under nitrogen, the mixture was warmed on a heating mantle with stirring and then heated on a steam bath for 3 h. After cooling, the mixture was poured into cold 5% HC1 (100 mL) and extracted three times with ether (50 mL). The ether was washed with NaHCO^ solution, NaCl solution and dried

(MgSO^). Upon evaporation of the ether at reduced pres sure, an orange oil (2CO remained. The oil solidified upon standing. Attempts to recrystallize the product were unsuccessful and it was not purified prior to fur ther reactions.

N-(cis and trans-t-Butyl-2-cyclohexen-l-yl) piperidine (13, _12) Alternate Synthesis. The crude orange ester 2j) (1 g; 0.003 moles) and piperidine (3 mL) were sealed in a glass tube and heated at 131° for 26 h. The mixture was poured into 20% NaOH (25 mL) and extracted with ether (50, 25 mL). The ether solution was washed with NaCl solution (25 mL) and dried (Na2 C0 2 ). The sol vent was removed at reduced pressure, and the residual oil distilled at 95-97°, 0.2 mm (lit. bp 131-132°,

10 mm^). GC analysis (5% SE-30; 140°, 7 min; program to

175° at 6 °/min) showed a major and minor peak with the major peak eluting first. The major peak was assigned 180 to 13 and the minor peak to '12. IE. (Neat) 3030, 2930,

1680, 1650, 1480-1450, 1360, 1390, 1220, 1115, 1100, 875,

865, 845, 795, 730. 1 H-NMR - identical to S^' product.

1 3 C-NMR (CDC13 ) <5 132.5, 129.5, 58.4, 50.7, 45.2, 33.1,

27.7, 26.6, 24.8, 22.4, 21.7.

N-(cis and trans-4-t-Butyl-l-cyclohexen-l-yl) —— ------gg— piperidine and Hydrogenation to give 14 and 15. A solu tion of 4-t-butylcyclohexanone (10 g; 0.065 moles), piperidine ( 8 mL; 0.08 moles) and j>-toluenesulphonic acid in toluene (50 mL) was refluxed in a 250 mL round bottom flask with a Dean-Stark trap equipped with a condenser for 4 h. The toluene was removed at atmospheric pressure and the remaining oil distilled at 0 . 2 mm and the fraction boiling 93-108° yielded 7.3 g (51%) of the enamine.

1 H-NMR (60 mHz, CDC13) 6 4.8 - 4.6 (1H, H-C=C), 2.9 - 2.6

(4H, H 2 C-N), 1.75 - 1.4 (6 H, H 2 C-CH2 -N), 0.9(s,9H, -CH3).

Hydrogenation: The enamine (1.5 g; 0.007 moles), was dissolved in abs. ethanol (25 mL) and hydrogenated

(5% Pd/C; 0.2 g) at 3 atm for 16 h with heating (=50°).

The catalyst was filtered and the ethanol evaporated at reduced pressure. The crude oil was distilled (short path) to give 1.1 g (70%) of a mixture (GC analysis) of

14 and 15 (bp 110-116°, 0.2 mm; lit. bp 136-137°,

7 mm3 ). IR (Neat) 2940, 1475-1450, 1390, 1360, 1220,

1110, 1095, 1015, 955, 860, 740. 1 H-NMR (CDC13 ), 60 mHz)

6 2.6 - 1.0 (20 H), 0.85 (s, 9H, -CH3 ). 1 3 C-NMR (CDC13)

6 64.5, 50.2, 48.2, 32.7, 29.3, 28.8, 27.6, 26.9, 26.6,

24.9, 21.5. 181

S^j21 Reaction of Propanethiolate With Substrate _1.

The reaction was modelled on the reaction by Stork and

Kreft.^ Sodium methoxide (0.15 g, 0.0028 moles) was weighed into a round bottom flask in a glovebag and capped with a septum. Nitrogen purged n-butanol (3 mL) was injected by syringe, followed by freshly distilled

1-propanethiol (0.23 g; 0.003 moles; 0.27 m L ) . The mixture was stirred with heating for 1 0 min and then added to a mixture of tetrafluoroborate 1 (0.4 g; 0.0014 moles) in 5 mL butanol. The reaction mixture was re fluxed under nitrogen for 5 h, cooled in an ice bath, and filtered. The flask was washed with pentane (3 mL) and butanol (0.5 mL). The solids were examined by

■^H-NMR (CD^COCD^) and did not show a t-butyl methyl signal so it was assumed that the reaction had gone to completion.

The filtrate was gas chromatographed on an 8 ft. x 1/8 in. column (5% FFAP on Chromasorb G: 65°, 23 min; program to

1 2 0 ° at 1 2 ° /min) and showed three peaks at retention times (relative to n-butanol) of 19.7 min, 47.1 min and

52.2 min. The last two peaks had areas of 8 % and 92%, respectively, relative to their total area. The M+ ion of Peak 1 was at m/e 150 and for Peaks 2 and 3 at m/e 212.

Preparative TLC (pentane; 3 elutions) gave 4 bands. The band 13 cm from the origin (solvent travels 14 cm) was identified as n-propyl disulfide. The band at 2.5 cm from the origin was N-(trans-6-t-butyl-2-cyclohexen-l-yl)-N,N- dimethylamine (_11) . The band at the origin was unidenti 182

fied. The band at 10.5 cm from the origin was identi

fied as cis and trans- 1 -(propylthio)-4 -t-butyl-2 -

cyclohexene (24, 25). IR (Neat) 3000, 2945, 2855, 1475,

1375, 1285, 1200, 1080, 1005, 930, 835, 800, 725.

1 H-NMR (60 mHz, CDC13> 6 5.75 (s, 2H, H-C=C), cau 3.3

(b, 1H, H-C-S), 2.5(t, 2H, H 2 C-S), 1.0 (t, 3H, H 2 C-CH3 ) ,

0.9 (s, 9H, -CH3 ). 1 3 C-NMR 6 131.2, 130.0, 45.6, 41.1,

32.7, 31.8, 31.1, 27.2, 24.6, 23.4, 13.7. Exact mass,

Calcd for C 1 3 H 2 4 S: 212.15987. Found: 212.15886.

1-(Propylthio)-trans and cis-4-t-butyl-2-cyclohexene

(24, 25) Alternate Synthesis. Sodium methoxide (0.37 g;

0.007 moles), nitrogen purged n-butanol (4 mL) and propanethiol (0.84 mL; 0.009 moles) were placed in a round bottom flask. The flask was flushed with nitroger^ capped with a rubber septum and heated briefly. A solution of the crude o-dichlorobenzoate ester 2J3 (1 g; 0.003 moles) in n-butanol (4 mL) was added and the mixture refluxed under nitrogen for 4 h. The reaction mixture was poured into 10% NaOH (20 mL), extracted with pentane (50 mL, twice) and dried (Na2 C03 ). After evaporation of the pentane at reduced pressure, a small amount of the products was chromatographed (pentane; 3 elutions). The band next to the highest band was cut and extracted with pentane.

Evaporation of the pentane led to an oil that was a mixture

(by 1 3 C-NMR) of 24 and 25. ^H-NMR (60 mHz, CDC13) 6 5.75

(d, 2H, H-C =C), ca. 3.3 (b, 1H, H-C-S), 2.5 (t, 2H,

H 2 C-S), 1.0 (t, 3H, H 2 C-CH3 ), 0.9 (s, 9H, -CH3 ). 1 3 C-NMR 183

(CDC13 ) 6 131.6, 127.9, 46.2, 40.7, 34.2, 32.7, 28.8,

27.3, 23.4, 19.6, 13.7.

Kinetic Experiment with the Piperidine Nucleophile .

Into each of four ampoules was weighed 25 mg (0.00008 moles) of substrate _1. A 0.35 M solution of distilled piperidine ( 2 equivalents) in distilled xylene was pre pared by adding piperidine (0.174 mL) to a 5 mL volumetric flask and then adding xylene. A 0.7 M solution of dis tilled piperidine (4 equivalents) was prepared in the same manner. The 0.35 M solution (0.5 mL each) was added to two of the ampoules and the 0.7 M solution (0.5 mL each)was added to the other two; the ampoules were sealed. A 2 equivalents ampoule and a 4 equivalents ampoule were heated together at 131° for 6 h. The other set was heated together for 11 h. The ampoules were cooled, opened and gas chromatographed (5% SE-30). Peak areas were determined by planimeter. The relative ratio of product formation for the 6 h experiment was 1:1.83,

2eq: 4eq, and for the 11 h experiment, 1:1.86, 2eq:4eq.

The amount of cis isomer formed relative to the trans isomer in each ampoule was: 6 h, 2eq - 19%; 6 h,

4 eq - 20%; 11 h, 3eq - 21%; 1 1 h, 4eq - 24%. BIBLIOGRAPHY

1. Streitwieser, A. , Jr. "Solvolytic Displacement Reactions"; McGraw-Hill: New York, 1Q&2.

2. Kepner, R. E.; Winstein, S.; Young, W. G. J. Am. Chem Soc. 1949, 71, 115.

3. Stork, G . ; White, W. N. J. Am. Chem. Soc.1956, 78, 4609. ~

4. Uebel, J. J.; Milaszewski, R. F.; Arlt, R. E. J. Org. Chem. 19JJ, 42, 585.

5. Bordwell, F. G. Acc. Chem. Res. 19^0, _3, 287.

6. Bordwell, F. G . ; Mecca, T. B. J. Am. Chem. Soc. 1969, 91, 362. "v

7. Sneen, R. A. Acc. Chem. Res. 1973, 6, 46.

8. Sneen, R. A. 9 Carter, J. V. J. Am. Chem. Soc 94, 6690.

9. Sneen, R. A. 9 Larsen, J. W. J. Am. Chem. Soc 91, 362.

10. Georgoulis, C,.; Ville, G. J. Chem. Res. (S) 248.

11. Georgoulis, C,.; Ville, G. J. Chem. Res. (M) 3344.

12. Stork, K. G. 9 Kreft, A. K. J. Am. Chem. Soc. 99, 3850.

13. Dobie, A. A.; Overton, K. H. Chem. Commun. L9J77 ? 722

14. Chiche, L.; Coste, J.; Christol, H.; Plenat, F. Tetrahedron Lett. 1978, 3251.

15. Kirmse, W . ; Scheidt, F . ; Vater, H.-J. J. Am. Chem. Soc. 1978, 100, 3945.

16. Ikola, N . ; Ganem, B. J. Am. Chem. Soc. 1978, 100, 351

184 185

17. Welch, S. G.; Hagan, C. P.; White, D. H. ; Fleming, W. P.; Trotter, J. W. J. Am. Chem. Soc. 1977, 99_, 549.

18. Martel, J.; Toromanoff, E.; Mathieu, J.; Nomine, G. Tetrahedron Lett. 19J72, 1491.

19. Stork, G . ; Kreft, A. F. J. Am. Chem. Soc. 1^7], 99, 3851.

20. Stork, G . ; Schoofs, A. R. J. Am. Chem. Soc. 1979, 101, 5081.

21. Magid, R. M . ; Fruchey, 0. S. J. Am. Chem. Soc. 1979, 101, 2107.

22. Oritani, T . ; Overton, K. H. Chem. Commun. 1978, 454.

23. Drendth, W. Reel. Trave. Chim. Pays-Bays 86, 318, 1967.

24. Anh, N. T. Chem. Commun. 1968, 1089.

25. Liotta, C. L. Tetrahedron Lett. 1975, 523.

26. Toromanoff, E. Tetrahedron 1978, 34, 1665.

27. Yates, R. L. ; Epiotis, N. D. ; B e m a r d i , F. J. Am. Chem. Soc. 1975, 97, 6615.

28. Adams, R.; Johnson, J. R. ; Wilcox, C. F. "Laboratory Experiments in Organic Chemistry"; Macmillan: Toronto, 1969.

29. Carbisch, E. W. J. Org. Chem. 1965, 30, 2109.

30. Dauben, W. G.; Williams, R. G.; McKelvey, R. D. J. Am. Chem. Soc. 1973_, 95, 3932.

31. Overman, L. J. Am. Chem. Soc. 1976, 98^, 2901

32. Pine, S. H . ; Sanchez, B. L. J. Org. Chem. 1971, 36, 829.

33. Caret, R. Ph.D. Dissertation, University of New Hampshire, Durham, New Hampshire,1974-

34. Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. "Spectrometric Identification of Organic Compounds"; John Wiley and Sons: New York, 1963. 186

35. Toll, J. Masters Thesis, University of New Hampshire, Durham, New Hampshire, 1977.

36. Booth, H.; Gidley, G. C.; Franklin, N. C. Tetrahedron 1965, 21, 1070.

37. Schmerling, L. J. Am. Chem. Soc. 1947, 69, 1121.

38. Baumgarten, H. "Organic Syntheses"; John Wiley and Sons: New York, 197,3; Vol. 5, 808. ADDENDUM

High Resolution Mass Spectrum of 2,2,2 -Trichloro-N- (.6 -

Butyl-2-cyclohexen-l-yl)acetamide (9).

Obs. m/e Calc, m/e El, Comp.

262.17582 262.07655 C 12H 18N0C12 264.07360 264.07325 C 12H 18N0C1C1,

212.95236 212.95151 C 6H 6NOC13

214.94774 214.94855 C 6 H 6 N0C1 2 C1'

216.94493 216.94560 C 6 H 6 N0C1C12 '

136.12789 136.12520 C10H 16

187 APPENDIX A: 1 H-NMR Spectra omi

Figure 1. Hl-NMR (CDCI3 ) 4-t-Butylcyclohexanone (3)

00 16 ^

Figure i-r~Tj! ,1 2 Q U I U '

O .S ppm

L l L i l.J .

Figure 3. ^H-NMR (CDCI3 ) 4-t-Butyl-2-cyclohexenone Ethylene Ketal (5) Figure 4. ^H-NMR (CDCI3 ) 4-t-Butyl-2-cyclohexenone M i i } m i i | i i i i i i i i i | i i i i | r r n -7 t r n | i i i i | t i i i | r -n | r n i \ i i ) i j I i’l i | I M » | i n v j r n i |- i i n - ) t i 1 y i i | i i i i j . » n - j -

h i [ 1700Hz .... 2J0P1.. . H li 1-lhi^‘i 600 !' 1 0 0 ...... H i.'u J :i Slliiiil i i i l E U II ijfetfiiP l i

O .Sppm 06 i i l ' l p p i i

ppm (5) I I . 1 L I . L !

Figure 5. H-NMR (CDCI3 ) 4-t-Butyl-2-cvclohexenol (7) (87% trans) 193 1200Hz

600 ‘H i l i f r i f j M i 1 I M i ! : , 300 I00:;iili1

|ij;;^ O Ji|i;!{ r i i ' l f f i i v.\mi

O .Sppm

lilililiiiili

L-l L 1.1 1- t_l I I--I-I 1 I P P -n (5 )

Figure 6. ^H-NMR (CDCl^) 4-t-Butvlcyclohexanol from (7) 194 1200H z

0 . 5pprn

P P ^ ( 5 )

Figure 7. ^H-NMR 4-t-Butyl-2-cyclohexen-l-yl 2,2,2 Trichloroetanimidate (8) .l- L X - L X 1 L _ J I I .1.1 1.1

Figure 8 iH-NMR (CDCI3 ) 2,2,2-Trichloro-N-(6-t-Butvl-2-cyclohexen-l-yl) acetamide (9) i ji ji j jl j !' J I200HI !; 2 0 0 !;: -.1 1 55i360;ii | ijjlilao p i!! l|0!ii! III J ' r i ! i l50 :i it s l l i f p i i i i i p j f | T ii V'l '■ IS ' i b ’ii i iliii lit! IlijHOiil !:• n ;i! I jS ii i ii j m ; ; < i b ^ •U'jiili i f i l l ! III f j iffii-s m 0 .5 ppm [ititiililii; •1 ' J :• :t: :r:t-: : : 0 ‘ iliiil ilrifflii l III! Ipiil p : l i i : ; | III i! i-ii'il 1 | 1 1 1 i i l - i i ’ i l l i : il: I; jliii; | | lilt jl! !! } !! ! ill!'! iiliiii m 1 1 III ' ■! I ll ij i jl j i l i i ! ililiill! i l l - i i i l i

i | | • j H j - . i t 1 !!:|lili! m ___ i i i _ . lil't:!!! d L - i i i i ifeliiri'

• i l il i :: ‘ U i x ' 1 .1:: U - i — I t I W i l l IPvil

b i s : J i l k p M l j l l t i l S iiiiji rijiiiji! 1 1 ■iibiibipii: - b j ; . i i k l i l f i l l i r l l l i b ' v - : »■ I...i l I L i J ! i l ■! m . 1 1 1 1 1 ! 1 t t 1 1 1 1 1 1 1 1 1 i 1 1 1 1 1 ...... ill * 11111111 lilt l_ l_ t ± l 1 1 1 1 1 1 1 1 1 t ! 1 I 1 1 1 1 1 1 i t 1 1 1 1 1 1 1 1 ■ : 1 1 1 1 • ' 1 1 I ■ I ! 1 . i I. : lll.il. ppm (5; ' £* 5

Figure 9. -^-H-NMR (CDCI3 ) N- (trans-6-t-Butvl-2-cvclohexen-l-vl) amine (10) 197 rp rrr rrrr rri r i i r i i i rr i i m | i i i

1200Hz liiiiriifiiB 1 M jiijiilj

ItlljsliijlliiM f ;i:d ' ■ .t.111 1 : 1 ',r Shj Hi ;;i! !!i : iri llliiilii jjiiiilih:!1 motm !"''1 I | l i i p i i ! i W I!''!' | f f j l l j tr tt j [ r I l l l i l " W l I ' M ' i r /I i li j li i it m m l i i i l i i i i i i n IlK lli .llilfiuIII O .Sppm (

l ,| J w m i iiiil ^ l i i tiiiiiii!• it fi i it i iliiiHsi ^ 1 l'i i l l l i l !! m m

ijijiiii ' ~ r ~

i l l

U i U i l i n ! .iili’ ii: ; s tjiii: t i l

hijjiiii iiiitii!!li l.-.j I::;;J 1 .=' ‘iliitiliiiliiSI ’-Oil! I1' ''11 I1!1'.' IlliilillllilillIllNl'l I p m ] \ r- i.l : i.i ■ l 1,1. i.l : 1 1 ll l_l 1.1 M 1 1 1 [ 1 1 I t 1 I t 1 1 1 1 1 1 1 1 I 1 I 11 1 . 11 1 1 1 1 1 1 1 1 1 I.'-I 1 I 1 1 1 11 P P n ( o )

Figure 10. lH-NMR (benzene-dg) N-(trans-2-t-Butylcvclohexyl)amine 1200Hz

I I I - !l ■ P Llh-'

O .Sppm

iiii-iiii

Figure 11. iH-NMR (CDClg) N-(trans-6-t-Butyl-2-cvclohexen-l-vl)-N^T-dimethylamine (11) 199 i 1 i i i i ~rr i i | rr-ri

: i l l 1:1200Hz 1000 hwjiii! m

t:!

■ ■ '■ I t it i

Figure 12. ^-H-NMR (acetone-dg) (t r ans - 6 - 1 - Butyl - 2 - cy clohexen-1 -y 1) trimethylammonium Tetrafluar

Figure 13 1 H-I,JMR (acetone-d6) (260 mHz) (trans-6-t-Butyl-2-cyclohexen-l-yl)-

trimethylammonium Tetrafluoroborate 201 n o o H r i|;

PPm (6)

Figure 14. ^H-NMR (CDClo) N-(cis-4-t-Butvl-2-cyclohexen-l-yl)t)iDeridine (13.) (SN2 ' ) “ 202 ^

^ ^ ^

Figure 15 1 H-NMR (CDCI3 ) 4-t-Butyl-2-cyclohexen-l-yl 2,6 -o-Dichloro- benzoate (2 0 ) JPffTT"

; | i i 1 i

}O i'pm liOOHi

I SO • li

I.II ! i!i !.(! -

lllii iNTpP

1.L.C.J .1

Figure 16. iH-NMR (CDC1-0 N-(cis and trans-4-t-Butvl-2-cyclohexen-l-yl)r)ineridine (13, 12) (SN2) 204 :l?OOHx! ^

iSiliislkilihil 'iilliliti!: tl ii I;-1:1111 llllll rp>n(ft)

Figure 17. ^-H-NMR (CDClq) N-(cis and trans-4-t-Butylcyclohexvl)pineridine ID " 205 I i t * | i I m | I 1.1 i | m M | i m I | M I i | i m r r nr] r i in I~ » ■ ■ | r r I r | i HI m j! jl2 0 0 H ii Iti 4 IPS i’soo.J"!; m iK# i'ijipK ? ? I I,',1??;!'

]4m 1.11 wm

).5 p p m i!(|f:io i!ii

Hi!! Sill 76

ii!-':!;; Si !l!l!ii|l

^ 48

ppm (6) ■ i ■« i i - 1 i 1.1 i I ii

Figure 18. ^-H-NMR (CDCI3 ) N-(4-t-Butyl-l-cyclohexen-l-yl)piperidine 206 i i I i I i i n ■i'fe [}•! i!cnfi: i!i

5ppm i ^

Moldmiprr-fjijln yiiliuiiiijl: m O .Sppm

ililllii .

Ilf! I

i—1—1-1 L L t I I I I J ppm (a)

Figure 19. Ir -NMR (CDCl-^) N-(cis and trans-4-t-Butylcyclohexyl)niperidine (14, 15) “ ~ 207 Figure 20. ^-H-NMR (CDCl^) N-(trans-6-t-Butyl-2-cyclohexen-l-yl)-N,N-dimethyl amine (11) from Pineridinolysis Experiment i i i r 1 i i m I I I ] I I M I I II | ! I I I I f I T T f l 1 -it:: 20ppm 1200Hz ! ii riooo i: • 600 200 ! •• ! ' ;i '■liilii- :• H ! " 1 • 1 i;:: ■ lOppm : 600 ii!!i!.500:;i!i: loo . : i -1:

Sppm :■ .300 :: !: •! 00 1 50:-i •;! m m l:::i lliiliiliii! 2ppm 120 -i !:& !:i:ilOO. ■ :!!: i •»:!:: 20 v:: i I- : ii;--': :• 1 : Ippti •• 60 ; I i Mv: s o : I ; 10 • ' ' L j: "H"; i- ~r-;- • O.Sppm 20mh: UJ: j . — \ JJ. : Mr*m l CH.CH lilliiliiiil

.!■ '

J ; ' :J;; - !: WM- lljll !!:::;]■

•liiii:: -4--— - grl i W - IB IHPt? :• J M u r p . v j j ' ■I rHifn*: lUju;:: ll;

.:i. ..

» I ■■ i i ! I ; i ■ i ! i i , . n 1 1 1,1 1 I ) I I l I t I I 1 > i i i J rill • t r 1 1 . 1 I ppm(i) 10 ,3 • 2 •

Figure 21. -^H-NMR (CDClo) 1-(Propylthio)-4-t-Butyl-2-cyclohexene (92% trans) "(24, 25) “ 209 1 T - r i - j 1 1 r - r r r p r it't ; i i i i i i r 11 : i | i! 20ppfn ij! :il2 00Hzj; Kim ‘lii! -Ii!!.: :‘!:i1 0. ■ iiu..;:: i:;:UL ;iil 1fji i, i i i lO ppm r 600:h; 1 m ;! ill 1 • 'rii II ill B l ? m 5ppm III! i n wijlr ...... lii! iiiiiL 2ppm ::i 20:r;:: ! -IliliO .iiiiE.'t iiiilii! _ _ iM; Ip p m :: 60 40 9 ;lil| •".ii' O .Sppm ■0!

IhS C H ^ H

! ii; I i

III 1 !' '

!ii|; [liiiii!! E l M i £ 11 > i i .i»i i ppm (i)

Figure 22. ^H-NMR (CDCI3 ) 1-(Propylthio)-4-t-Butyl-2-cvclohexene (87% cis) Sn2 210 lO ppm 600 SOO 200:

Sppm 3 0 0 : ;ioo

O .Sppm

Figure 23. ^H-NMR (CDCI3 ) N-(trans-6-t-Butyl-2-cyclohexen-l-yl)-N,N dimethylamine (11) from S^j21 Reaction 211 Figure 24. ^H-NMR (CDCI3 ) Piperidinium Tetrafluoroborate

ilOOOjji eooiji 200

rSO.Qi j'200

300 120 ;i.oo;

Ip p m

0 . 5ppm APPENDIX B: 13C-NMR Spectra 13 Figure 1. C-NMR (acetone-dg) (trans-6-t-Butyl-2-cyclohexen-l-yl)trimethvl- ammonium Tetrafluoroborate (1)

+•50001S!||j!! f*40(X > t Ii! 1: 250Oj ,,2000 ’ilMi i i ^2000 m & m ^ 1000 m IfiL

i i i i i 214 M Hi 1 ^ Figure 2. C-NMR (CDCI3 ) N-(cis-4-t-Butyl-2-cyclohexen-l-vl)piperidine (13) (SN2')

fi5000p

inn 92 P i i 700tV

j o o ”

! mi SHf

iiii liiiit [illiiill Figure 3. ^C-NMR (CDCI3 ) N-(cis^-4-t-Butyl-2-cyclohexen-l-yl)piperidineXl3) (S^2)

M Ml ii1H!i 200

ii!m!ililillii m s m 216 w

13 Figure C-NMR (CDCI3 ) N-(cis and trans-4-t-Butylcyclohexyl)piperidine (14, 15)

THrf

411!+ilOOO

ililliillMill 217 Figure 5. 13C-NMR (CDClo) 1-(Propylthio)-4 trans) (SN 2')

^ Figure 6 . ^C-NMR (CDCI3 ) 1-(Propylthio)-4-t-Butyl-2-cyclohexene (87% cis)(SN2) APPENDIX C: IR Spectra iue . R Na) --uv--ylhxnly 2,2,2- 4-t-Butvl-2-cyclohexen-l-yl (Neat) IR 1. Figure

o c APP-ORRANCF 1-0 1300 rclrehnmdt ( Trichloroethanimidate 6-0 1200 9-0 1100 1000 10-0 QUti CI I ') I I |C Y tliC U tQ H ■ ( 0 000 900 4-0 5-0 MICRONS MICRONS 8 ) GGO 20-0 250 400700 1-0 0-0 221 iue . R Kr 2,2,2-Trichloro-N-(6-t-Butyl-2-cyclo- (KBr) IR 2. Figure ’/' I L BSOKBANC -*■! t ■■ 30 20 10 00 0 OG7uo OOG 900 1000 1100 1200 1300 ee--laeaie ( hexen-l-yl)acetamide - MCOS 5-0 MICRONS 4-0 IRM "15*0 MIcROMS ; i : i 9 1 I .1 ) i i ii i r1 223

MICRONS , ■ 1 5 .0 o-o 9-0 10-0 70-0 25-0 0-0 i ii

r + .

: • i rrr

VO 1-0

o c 1300 1200 1100 1000 900 800 700 SCO •100

3-5 4-0 MICRONS 5-0 6-0

4000 3500 3000 2500 2000 1500

Figure 3. IR (Neat) N-(trans-6-t-Butyl-2-cyclohexen-l-yl) &mine (1 0 ) iue . R Na) N-(trans-6-t-Butyl-2-cyclohexen-l-vl)- (Neat) IR 4. Figure

ARSORI’A U 1— - O C - 8-0 7-5 0-0 4000 - 3-0 2-5 ,-iehlmn ( N,N-dimethylamine 3500 9-0 9-0 1100 10-0 1000 3000 .5 3

2500 1 1 -0 4 MICRONS MICRONS 5-0 5-0 MICRONS ) 2000 6-0 1500 '5-0 c c 0-0 o o 1-0 40 224 225

2-5 ; 3-0 3 b 4-0 i il' KOMS 5-0 6 0 0-0 0-0 0-0

u j - 2 0

i i 1-0 1-0

c o L — - i 4000 3000 2500 1500

MICRONS

Figure 5. IR (Neat) 1-(Propylthio)-4-t-Butyl-2-cyclohexene (24, 25) (92% trans). (SM 2 ') 226

4-0 MICRONS 5-0

%#($)%&&*)%#((*#'&%)$%&%++*%*&()%$&$)#)%(

SiUillliMIQ i.iiiLiniiiiiini ililiillliiliiiirilllliiiiiiillliiiillihh 4000 3S00 3000 2500 2000 1500

FREQUENCY |CM J|

HH-KUNi

1300 1200 1100 1000 900 800 700 600 500 400

FREOUEHCY (CM -1 Figure 6. IR (KBr) (trans-6-t-Butyl-2-cyclohexen-l-yl)- trimethylammonium Tetrafluoroborate (1)■ 227

4-0 MICRONS 5-0

h.'n.'iilnHr1

M I

4000 3500 3000 2500 2000 1500 FREQUENCY (CM-*)

MICRONS

FREQUENCY (CM-1)

Figure 7. IR (Neat) N-(cis and trans-4-t-Butyl-2-cyclohexen-l-viypiperidine (13, 12J (S^2) iue . R Na) -cs n trans-4-t-Butylcyclo- and N-(cis (Neat) IR 8. Figure

ABSORBANCE - ' 8-0 ' 7-5 25 30 - 40 IRN 50 - 8-0 6-0 5-0 MICRONS 4-0 3-5 3-0 1 2-5 • 4000 1200 ey)ieiie 1, 15TT (14, hexyl)piperidine i 503000 3500 ® * a LLLLiUJJ 10 00 0 800 900 1000 1100 llllllillilhlillllTTTlJiTiTTi! u 10-0 FREQUENCY(CMJ) FREQUENCY (CH-MFREQUENCY 2500 n i v . r \ v j i N i 15-0 2000 S K B U H S S U U i f t 00 25-0 20-0 1500

•40 •70 1-0 •60 •50 •30 0-0 •20 •10