The Chemistry of Non-Stabilized Sulfenes

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1976 The heC mistry of Non-Stabilized Sulfenes. Huei-nan Lin Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Lin, Huei-nan, "The heC mistry of Non-Stabilized Sulfenes." (1976). LSU Historical Dissertations and Theses. 2928. https://digitalcommons.lsu.edu/gradschool_disstheses/2928

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

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

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

1.The sign or "target" for pages apparently lacking from the document p h o to g rap h ed 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 m ay have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

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

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

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

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

Xerox University Microfilms 300 North Zaeb Road Ann Arbof, Michigan 48106 76-25,271

LIN, Huei-Nan, 1944- THE CHEMISTRY OF NON-STABILIZED SULFENES.

The Louisiana State University and Agricultural and Mechanical College Ph.D., 1976 Chemistry, organic

Xerox University Microfilms, Ann Arbor, Michigan 4B106 THE CHEMISTRY OF NON-STABILIZED SULFENES

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the Degree of

Doctor of Philosophy in The Department o f Chemistry

by Huei-Nan Lin B.S., Fu-Jen University, 1968 May, 19T6 For My Parents acknowledgment

The author wishes to express his gratitude and appreciation to Dr. N. H. Fischer for his guidance, direction, and inspiration.

Without Dr. Fischer's encouragement and patience, the accomplish ment of this work would have been impossible.

Grateful acknowledgment is also recognized for the financial assistance from the Dr. Charles E. Coates Memorial Fund of the

Louisiana State University Foundation for expenses incurred in the publication of this Dissertation. Special thanks are given to the author's cousin, Dr. Nai-Teng

Yu, Associate Professor at the Georgia Institute of Technology, for his assistance and understanding during the author's study in the United States of America.

The author is also grateful to his wife, Laura Hui-Shue, for her sacrifice of many weekends.

iii TABLE OF CONTENTS

PAGE ACKNOWLEDGMENT...... i i i

ABSTRACT ...... v

I . INTRODUCTION ...... 1

I I . RESULTS AND DISCUSSION ...... 19

A. On the Mechanism of Sulfene Formation ...... 19

B. Attempted Generations of Non-stabilized Sulfenes 33

via Thermolysis of Bicyclic Sulfones ......

C. Solvent E ffect and Neighboring Group 55

Participation in Sulfene Formation ......

I I I . EXPERIMENTAL...... 71

i. Reaction of Sulfonyl Chloride with

Metal Hydrides ...... 73

ii. Synthesis of Thietane-s,s-dioxides ...... 77

iii. Synthesis of a-Thiopyran-1,1-dioxide and

Thermolysis of Bicyclic Sulfones ...... 79

iv. Reactions of D-Camphor-10-sulfonyl Chloride

and Camphane-10-sulfonyl Chloride with

Triethylamine ...... 95

REFERENCES ...... 106

APPENDIX A: Nmr S p e c tr a ...... 113

APPENDIX B: I r S p e c tr a ...... 123

VITA...... 133

iv ABSTRACT

Sulfenes are comnonly generated from aliphatic sulfonyl chlorides with tertiary amines. The accepted E2 mechanism for the HCl-elimination from the sulfonyl chloride was re-examined using bases other than tertiary amines. Bases of equal or higher strength than tertiary amines but 1cm nucleophilicity were chosen for the experiments.

Aliphatic sulfonyl chlorides failed to react with sodium hydride, lithium hydride or the Corey base, bases which are generally con sidered to be stronger than tertiary amines. In a kinetic study, using phenylmethanesulfonyl chloride and triethylamine, a second order rate dependence for triethylamine was observed. Based on these experimental data, the E2 pathway for sulfene formation from aliphatic sulfonyl chlorides and tertiary amines is not likely.

Instead, a mechanism involving initial attack of the tertiary amine at the sulfonyl group via a sulfone ammonium intermediate followed by proton abstraction (substitution-elimination pathway) is in agreement with the experimental results.

A new method for the generation of non-stabilized sulfene by the thermolysis of allylic, bicyclic sulfones was attempted.

Dimethyl 2,2-dioxido-2-thiabicyclo[2,2,2]octa-5,7*diene-5,6-dicarboxylate and dimethyl 2 , 2-dioxido- 2-thiabicyclo[ 2 , 2 ,2 ]o c ta - 7-ene-

5 , 6 -dicarboxylate were thermolyzed with and without sulfene trapping agents. The thermolysis reactions proceeded via a novel SOg- extrusion-rearrangement reaction under formation of cycloheptatriene derivatives. Trapping experiments involving electron rich alkenes

(enamines) did not lead to detectable amounts of sulfene cycloaddition products. v Attempted cycloaddition reactions between in situ generated sulfene (methanesulfonyl chloride/triethylamine) and an electron deficient diene (thiopyrane- 1, 1-dioxide) were without success.

The chemistry of D-camphor-10-sulfonyl chloride with tr i ethylamine, a reaction known for many decades, was explored in greater detail. (Z)- and (E)-D-camphor-lO-chlorosulfoxide were successfully separated and obtained in pure form. In contrast to the reaction of phenylmethanesulfonyl chloride and triethylamine, the reaction of D-camphor- 10-sulfonyl chloride with triethylamine did not show any solvent dependence. The lack of solvent dependence is attributed to the involvement of the camphor-carbonyl oxygen which stabilizes the positive carbon center of the sulfene inter mediate by neighboring group participation. Experimental evidence for the necessity of the carbonyl group in the formation of the

D-camphor-10-chlorosulfoxides was provided by the reaction of camphane-10-sulfonyl chloride with triethylamine. The latter reaction failed to give camphan«-10-chlorosulfoxides. Instead, the aldehyde 7,T-dimethyl-l-formylbicyclo[2,2,l']heptane was formed in low y ie ld .

vi I. INTRODUCTION

Several reviews related to the generation and chemistry of sul

fenes have appeared in recent years . 1 ,2 ,a ’4 Therefore, this intro duction is not intended to represent a comprehensive review on sul

fenes but to briefly summarize methods of generation and the most

typical reactions of sulfenes.

Formally, sulfenes are the S,S'-dioxides of thioaldehydes (_1,

R ■ a lk y l, a ry l R' 3 H) or thioketones (l.> R * R' » alkyl, aryl).

R 0

They can also be classified as the inner anhydride of sulfonic acids, just as ketenes represent the inner anhydrides of carboxylic acids.

Furthermore, they can be looked at as derivatives of sulfur trioxide, in which one oxygen atom is replaced by an alkylidene group .2 In sulfenes, all the electrons of the sulfur atoms are involved in bonding.

From the reactions of sulfene with various reagents, it can be concluded that several resonance forms (la-W) may be formulated for this molecule.

0 0 0 . 0® V X- s I CHa CHa®

la lb lc Id

1 2

The accum ulative s tru c tu re Uj , which is neutral locally, as well as overall, indicates the ability of sulfene to react in two directions, on the one hand, to form with water derivatives of methane- sulfonic acid, and on the other, to give derivatives of hydroxy- methanesulfinic acid. Both types of processes do occur, but as will be seen later, most sulfene reactions correspond to the ones which proceed via the ylide structure la_ .2

In 1911, Wedekind and Schenk 5 coined the name sulfene to express the relationship to the ketenes. They made the first attempt to synthesize a sulfene from an aliphatic sulfonyl chloride and tri ethylamine in analogy to Wedekind's ketene synthesis , 6 In a reaction of phenylmethanesulfonyl chloride 2_ with triethylamine they obtained triethyiaranonium chloride and trans-stilbene ^ , which they assumed was derived from phenylsulfene ^ , possibly via^the carbene intermediate Ij- . In 1916, Staudinger and Pfenninger 7 attem pted

NEt3xHCl

2 3

-S02

H Ph ^ dim eriz.

Ph H 3 the synthesis of a sulfene by the reaction of diphenyldiazomethane

_6 with sulfur dioxide. Their investigations did not provide the expected diphenylsulfene £ but resulted in the formation of the episulfone 8 as o u tlined in Scheme 1.

Scheme 1:

0 -Na 2 CgHgCN2 + SQ2 [Ph 2C~N s N] > [Pt^C-SCfe] I 0 SO2 6 7

P h2 CN2

Similar reactions involving diazoalkanes and sulfur dioxide were later studied by Kloosterziel8, Hesse 9 ' 16 and Vargha10.

Further references 11’ 12113 related to sulfenes were made at about the same tim e. However, no research was c a rrie d out in th is area be tween 1S90 and 1950- The only report on sulfene during these two decades appeared in Suter's book14, which pointed out that "a sulfene of structure R2C=SC£ could not be prepared", the author referring to the above mentioned sulfene literature.

In 1952, Backer, Kloosterziel and co-workers recalled the name sulfene to the chemical literature . 15 In a series of papers, they extended and clarified the earlier experiments of Staudinger and

Pfenninger.15’17, In i 960 , sulfene chemistry experienced further rejuvenation when DeMayo and co-worker reported that ultraviolet irradiation of six-member ring diene sultones *-n methanol gave the sulfonic ester _11 which was regarded as arising from the sulfene intermediate 10 . 18

JbL CH3 OH SO3 CH3

2 12. -U

Similar results were obtained by King et al . 19 and by Weintraub et a l .*1

Today, most comnonly sulfenes are generated in situ by the

Wedekind method, that is, from sulfonyl chlorides bearing a-hydrogens and tertiary amines. Triethylamine is most frequently used as a base since it can be easily purified, added to and removed from the reaction mixture. Furthermore, its hydrochloride is practically insoluble in the common organic solvents and can therefore be removed by filtration.

The existence of sulfene intermediates in the Wedekind reaction was demonstrated independently by King and Durst 21’22 and Truce .23

They reasoned that, if the reaction of a base with an alkanesulfonyl chloride 12 were to proceed vfa. a sulfene 1^. , then, in the pre sence of D^O or deuterated alcohols (DOR) the products should contain only one atom of deuterium at the carbon a to the sulfonyl group Ij; .

H R-CH2 -SQ2 -Cl + R^N ------> [R-CH=S0e] + RiN Cl" © 12 a: Z - OD R-CH=S02 + DZ ------> R-CHD-SO2Z b . z * OR

lU By contrast, a direct substitution at sulfur by DZ or R3'N would lead to the nondeuterated compound 15 . Any random exchange of hydrogen for deuterium would be expected to give a mixture of unsubstituted, mono- and dideuterated products.

+Et3N R-Chs-Stfe-Cl + DZ . Et3ND c l - > R-Clfe-SQsZ 0 11

Experimentally, in the presence of triethylamine as a base, with an array of sulfonyl chloride and DsO and DOR as substrates, the products were almost exclusively monodeuterated. In the absence of base, the esterification reaction proceeded much slower without deu terium exchange.

In the absence of trapping agents, in situ generated sulfenes give a variety of products depending on substrates, reaction conditions and solvents. Opitz and co-workers 24 ’2 5 ’28 obtained the relatively stable trimethylammoniun adduct of mesyl sulfene 16 under conditions given in Scheme 2. Dimeric sulfene, the four-membered ring 17 represents a minor product in reactions of methanesulfonyl chloride with triethylamine in a variety of solvents.

Scheme 2: $ -2HC1 SQ2 -NEt3 2 CH3-S0.-C1 + 3Et3N ln qgaVJ^rt. > 4 , Scheme 3

s q - c i RfH >CRtC = S Q ] RtCHSQp R -CHRt

\ SQ a Cl f — CRt

s q - c i Cl S Q -C fi, I *CRZ ^ C R t qf Ti*,cv -CWR, •'CHRt 4 7

R. Fusco e t a l .27 observed that the reaction of 2-propane sulfonyl chloride with triethylamine in non-polar solvents yielded oligomeric sulfones. They postulated that the oligomers would arise from the corresponding sulfenes as outlined in Scheme 3- Similar results were later reported by Grossert and Bharadwaj .28 In acetonitrile at -h0°, they obtained the oligomerization products derived from mesyl sulfene.

Dimerization of a sulfene in a Diels-Alder fashion was first observed by Fusco, et_ al.29: Benzoylmethanesulfonyl chloride (l 8) and triethylamine gave the sultone 20 via the sulfene 19 intermediates as shewn in Scheme 14-.

Scheme U:

A broad interest in sulfene chemistry was generated by the pub lication of Stork and Borcwitz30, and by Opitz and Adolph31. In their studies related to the sulfonation of eaamines, the formation of sulfene-enamine cycloaddition products, four-membered ring aminosulfone of the type 21 , were obtained. These findings followed further studies on cycloaddition reactions of in situ generated sul fenes with electron-rich olefins such as other enamines, ketene 8 acetals and amlnals .32,33

CM—so—Cl + •so.

Double cycloaddition between disulfene and enamines were reported by T. Nagai . 34 Paquette et al . 3 5 *36 observed highly stereoselective reactions of bicyclic enamines with sulfenes as shown for the forma tion of spiro ( 5 -norborene- 2 , 2 '-thietane)-l,l-dioxide ( 22).

A Qon-concerted mechanism for sulfene cyclo-addition reaction with enamines was suggested by Paquette et al.3S,3S for the reaction of l , 3 -bis(dimethylamino)-l-alkene with sulfenes derived from methanesulfonyl chloride and phenylmethanesulfonyl chloride. Opitz and co-workers 37 described the reaction of sulfenes with ketene-N,N-acetals

25 which led directly to a mixture of thietene-S,S-dioxides 2k and

2*5 by elimination of one amino group R—CH so2—ci I! O f i 0

23 26

from che initial cycloaddition product 26 . The cycloaddition of

ketene 0,N-acetals with sulfenes is also followed by spontaneous

elimination of alcohols from the intermediate 28 , leading to the 3 ”

dialkylaminothietane- 1, 1-d io x id e ( 2 £ ).38,39 R A

v . •SOt ( p t Q p — G CCH^N- O C f t \ ZL 2£

f t

CCH%\N ''R

29 10

From the sulfene J2, generated either from the sulfonyl chlorides

JO or 2JL an<* ketene acetal, Truce and co-workers 32*40’41 obtained

mixtures of the thietane- 1, 1-dioxide derivative Jij- and the six- membered rin g sulfone 53 •

CHa = CH-CHa-SOa-Cl

[CHa = CH-CH - SOa] 22

CH3-CH = CH-SQa-Cl CHa = C (0 E t)2

H 22. 2k Vinyl ethers JJ do not react in ether but undergo cycloaddition with in situ generated mesyl sulfenes in acetonitrile to give the

four-membered ring sulfones J 6 _ together with small quantities of

the enol ester 5 7 .45

Ynamines j 8 react similarly with sulfenes to give isomeric 3-(di~ ethylamino)-thiet-l, 1-dioxide (J 2. and 4o ) . 43,44,45 i l

CH;—C ^ —wr.

fc-C| / * I 1 I p,H,-CHl—sqr a fa q, d

1 2 UO

Most recently, Hiraoko and Kobayashi 46 described the first example of a successful reaction of sulfene and a Schiff's base kl, to form the isomeric 1, 2 -thiazetidine- 1,1 -dioxides ("42^ and (lt -3 )

XSH>

Diels-Alder type reactions of conjugated sulfenes, in which the sul fene represents the diene, are less common. Some examples are the earlier mentioned dimerization of sulfene (Fusco et al.29), and the reaction of allyl sulfene with ketene diethylacetals (Truce et al.40’41).

Opitz and Temple 47 demonstrated that sulfene can act however as a dienophile toward 0-aminovinyl ketones (Wf) providing enol sultones of the type 'k5 .

i i 12

OplCz and Schweinsberg 40 reported that butadlenylamines (U 6 ) re a c t with sulfene to give 1, 14-- and 1,2 -cycloaddition products (h 8 and h7, respectively), reactions that probably proceed in a stepwise manner.

SOi

Sulfenes also undergo 1,3-dipolar cycloadditions but only with a limited number of substrates. Rossi et al .48 and Truce and co-workers50, reported on the l,3"

(h9) with sulfene to form a seven-membered ring azasultone (50).

r? w SO.

ii2

In situ generated sulfenes, aerived from primary sulfonyl chlorides by the dehydrohalogenation method, and diazoalkanes form episulfones

51 which can be converted thermally to the corresponding olefins 52 were reported by Opitz and Fischer51’52; Rossi and Maiorana 49 and

Neureiter53. FL-CHt— SO,— CI »rR — C H = S O j I CHjAi R—GHV-----

R —CH—CHt + SO,

The reactions of sulfene with stable phosphorus ylides was des cribed by Ito et al.54. Treatment of a stable phosphorus ylide bearing an ot-hydrogen with methanesulfonyl chloride in the presence of triethylamine resulted in a mixture of an alkene 52 an<* tl*e stable ylide 5 ^ • R

+ Ccn =sq j

c h ^ c t '*-

Ph^P-CH—sq-cH R R 14

King e t a l . 55 observed that sulfene generated from phenylmethanesulfonyl

chloride reacted with triphenylphosphine to give the benzyItriphenyl- phosphonium salt.

A highly stereospecific synthesis of butadiene derivatives by reaction of stable sulfoxenium ylides with sulfene was described by

Idess and is outlined in the following 3cheme.

• m

A number of interesting approaches toward the generation of sul

fenes by thermal and photochemical rearrangement reactions have been described in the literature. As a part of the classic studies of

thermal rearrangement of 1,5-dienes, Cope et al . 37 suggested that heating of the allyl vinyl sulfone (£ 5 .) could possible produce the su lfen e .

55 £ However, pyrolysis of at various temperatures only led to decom

position and polymerization products. In 19T1, King and Harding 53

demonstrated that Cope's original idea had been correct. It was

evident from the more recent work that Cope's failure to obtain de

finite rearrangement products was due to the unfamlllarity with the

earlier sulfene literature that sulfenes were unable to be isolated.

Mulder59 reported an example of sulfene generation by a reaction ana logous to the Wolff rearrangement. Irradiation of a-diazosulfone %£ in methanol gave low yields of methyl sulfonate 60. The formation of

60 was postulated to proceed via the sulfene 3 9, which represents the rearrangement product of the carbene £ 8 .

Recently, R. Langendrides 30 described the irradiation of 3 *

thietanone- 1, 1-d io x id e ( 6 l) in acetonitrile or in a 9:1 mixture of

acetonitrlle/methanol in tne presence of diphenyl methanol. The

initially formed sulfene, reacted with diphenyl methanol to first give

its mesylate 62 from which the ether was derived

by further substitution.

CPhfHOSq-CtiJ

C^SOJi + Ft)%CH-~0~CHPhz

22 King and deMayo 31*32 described the generation of sulfene by flash

therm olysis of c h lo ro s u lfo n y la c e tic a c id a t 650°C to give SQg-CHs, which with methanol on a cold finger was converted to the stable methyl sulfonate. King and ccworkers reported the IR spectrum of free

sulfene which had been obtained at -196°C by flash thermolysis of

CISO2 -CHa-COOH and CH3 -SO2 -OSO2 "CH3 . The ir spectrum showed bands

a t 3170, 30^ 0 , 1330* 1230, and 950 cm"1; the first two bands were assigned to the sulfene CH 2 group and th e second two ab so rp tio n s

appeared to be due to the S0& group in sulfene. On warming, these

bands disappeared. When methanol was deposited along with the ther molysate, the Infrared spectrum showed the characteristic sulfonate bands a t 1330 and 1230 cm"1, the remainder being obscured by the methanol absorptions.

It had been postulated that sulfene has a full C-S double bond

and partial S -0 double bonds33. In the dlmerization of sulfenes, a

two-step pathway Initiated by the attack of carbon at sulfur is

suggested. Pertubation molecular orbital (MO) theory postulates that

the TT-frontier orbitals are directing the early stages of combination.

Accordingly, the highest occupied molecular orbital (HCMO) of one

addend w ill engage the lowest unoccupied molecular orbital (LUMO) of

the other. The bond orders and total charges densities were cal

c u la te d by Boyd 64 and are shewn in Figure 1.

- 0.62 Figure 1: s +1.2 17

The HCMO's and LUMO's of sulfene, obtained by ab inito calculations

by Houk et. a l.35, are summerized in Figure 2. The calculations do not include d orbitals at sulfur. -.233) Figure 2

■ 7/73 l u m o

+ . 57/4. -.1344 HOMO 3♦ .5T14

CHARGES

Houk66, using an ab inito technique to optimize the geometry of sul

fene obtained the following data (Figure 3).

Figure 3:

Contrary to the generally accepted hypothesis that electron withdrawing substituents at a sulfene carbon stabilize the sulfene,

semiempirical MO calculations, obtained by Snyder63, suggested that

sulfenes should be stabilized by electron-donating substituents. As

electron-withdrawing substituents are replaced successively by more

electron-donating ones, the positive charge on sulfur is diminished 18 while that at carbon undergoes a reversal from negative to positive.

It was predicted that electron-rich substituents reduce the electro-

philicity of sulfur and eliminate the nucleophilicity of the sulfene

carbon. Indeed the highly stable diaminothiourea-S,S'-dioxide ( 6 ^) can easily be prepared by oxidation of thiourea67. However, Opitz's definition of a sulfene requires the presence of hydrogen, alkyl c - aryl groups on the carbon attached to sulfur. Therefore, the thiourea derivative does not represent a sulfene, but the data still demon strate which factors contribute to sulfene stabilization.

6 h Using CNDO/2 calculations, Houk et a l . 65 stated that in cyclo-

addltion reactions, sulfenes resemble Isocyanates more than they

resemble ketenes. For instance, they should be less prone to undergo

concerted reactions. The lowest vacant orbital of sulfene is a TT

orbital heavily localized at the sulfur atom. From the data,

the authors concluded that sulfene should not undergo concerted cycloaddition reactions with alkenes. However, sufficiently nucleo-

philic alkenes may add to sulfene in a stepwise reaction, whereas,

electron-rich dienes could react in a concerted £2+ + 2] fashion. The

calculations also predicted that electron deficient dienes could

possibly add concertedly, if the cumulene HCMO-LUMO interaction

became sufficiently great. A. ON THE MECHANISM OF SULFENE FORMATION

Although In the reaction of aliphatic sulfonyl chlorides and tertiary amines the intermediacy of sulfenes is now well established, the detailed mechanism for the formation of these reactive inter mediates is still in question. Mechanisms ranging from multistep processes with initial formation of sulfonyl chloride-amine charge-transfer complexes, to the concerted, E2-type sulfene forma tion have been suggested by different authors .2 The first step in the generation of sulfene from methanesulfonyl chloride and tertiary amines could possibly proceed via the inter m ediates 6 £ , 66 o r 6 l

+ + CHg—SQg —Cl CH3 —SQg—NRg Cl

65 66

CH3 -SO2 - NR3 Cl

Fusco eit a l .29 and King and Durst 19 postulated the initial abstraction of a proton from the oe~carbon of the sulfonyl chloride to give inter m ediate 65 . 1 from which by the subsequent loss of the chloride ion the sulfene would be formed. Opita and Fischer 68' 09 observed that methanesulfonyl chloride and trimethylaming in ether formed a pre- c ip a ta te which shewed no NH-anmonium bands in the 1R spectrum . Upon treatment of the aduct with dry DC1, the nondeuterated sulfonyl chloride was regenerated. This indicated that in the aduct no proton had been abstracted from the sulfonyl chloride, which was used as an

19 20 argument to exclude Fusco's structure 6£ • With aniline, the Opitz-

Fischer-aduct gave the sulfonamide, and with enamines the appropriate cycloaddition products were formed. In contrast, when methanesul fonyl chloride and triethylamine were reacted under analogous con ditions, a precipitate was obtained which contained ammonium ions and had lost the ability to react with amines and enamines . 68*69 The formation and reactions of the methanesulfonyl chloride- trimethylamine aduct suggest that the HC1-elimination with trialkylamines from other sulfonyl chlorides also proceeds via labile inter mediates of type 66 and/or . The generation of sulfenes was therefore formulated by Opitz 2 and Fischer 68 as outlined in the following scheme:

0 SQg-CI R t N „ ft,A/—»SQ,—C! R,N—S0, Cl~ i/4 — i/4 ’ U

67 66

*//j-ft, N-Ha

RJV—SO, SOt «------| CA 69 68

The formation of 66 and/or 6£ seems to be necessary to increase the acidity of the methyl hydrogens. For Instance, when methanesul fonyl chloride was reacted with methylmagnesium iodide in ether, practically no methane evolved, whereas the trimethylamine aduct pro duced methane when treated with the Grignard reagent . 68 The re a c tio n 21 of phenylmethanesulfonyl halides with the stronger base phenyl lithium resulted in mono-, di- and trisulfones the formation of which does not necessarily require sulfene intermediates but could more likely in volve carbanionic sulfones .70

In a kinetic study of the reaction of methanesulfonyl chloride and trlethylamine in the presence of aniline as a sulfene trap, King and Lee 71 provided evidence that the reaction is first order in both the sulfonyl chloride and trlethylamine. From these data it was con cluded that the generation of sulfene must involve an E2-type mech anism.

Later, King and Lee 72 again applied kinetic methods, using axial and equatorial rate ratios, in the study of sulfene formation. The axial sulfonyl chloride 72, was found to react 71 times faster than the equatorial epimer 71 with trlethylamine in the presence of aniline at -25°C, a reaction which was known to proceed via the sul fene intermediate J2 ,71

s a - c i

3 7 T *

SQ—N-Ph H H H

JU (minor) (major)

In the presence of water at 50°C without added base, the sulfonyl chloride 71 reacted lk times faster than the Isomer 72 > to pro duce the axial and equatorial sulfonic acids 75 . 76 , resp ec- 22

liv e ly .

sq-a H

7 0 71 J HP i SQH H

11 76

From the rate data, King and coworkers again concluded that c'ne form ation of sulfene J2 from JO and Jl_ by the action of triethylamine must proceed via the E2 mechanism on the basis of the following argum ents:

Elimination studies with 4-cia- and U-trans-t-butylcyclohcxane derivatives had shewn that the axial epimers react distinctly faster than the equatorial epimer .73 For instance, phenyl cis-U-t-butylcyclohexanemethyl sulfoxide (T7) thermolyres about 6 times faster than its trans (equatorial) epimer (T 8) to give 1-t-butyl-U-methylene- cyclohexane (T 9).

A lso, c i s -k - t-butylcyclohexanemethyl bromide (80) undergoes elimina tlon with potassium t-butoxide in t-butyl alcohol at 1C0°C about 9 25 times faster than its trans (equatorial) epimer ( 8l ).

The rate differences here shoved the same trend as the ones ob served for the reactions of the sulfonyl chlorides in vhich the axial sulfonyl chloride JO underwent bimolecular elimination with a distinctly faster rate than its equatorial epimer J1 . In con trast, the hydrolysis of the axial sulfonyl chloride JO and the equatorial epimer J1 to the corresponding sulfonic acid J^ and

j 6 by direct attack of HgO at the sulfonyl chloride, in the ab sence of base, the rates were reversed. Here, the equatorial sul fonyl chloride reacted faster than its axial epimer JO . On the basis of these data, King et al. concluded that, initial attack of the tertiary amine at the sulfonyl chloride group to give inter mediates of type 66 could be excluded since the observed rate data for the conversion of JO and Jl^ to form J2 were not in agree ment with the hydrolysis reactions of the sulfonyl chlorides JO and J1 but showed similar trends in the rate ratios as the model elimination processes involving the sulfoxides JJ_ and j 8 . On the basis of the rate data, King and coworkers concluded that dehydrohalogenation of sulfonyl chlorides with tertiary amines were to 2 k proceed via an E2 mechanism.

In 1972, King and coworkers 74 reported the reaction of methanesulfonyl chloride in 1, 2 -dimethoxyethane containing DaO ( 15 $) w ith

1,^-diazabicyclo [2,2,2]octane (DABCO). The reaction provided the

CD3SO3 salt as the major product, with diminishing amounts of CHTaSO^",

CH2DSQ3 and CH3SQ3 . In a series of experiments, King and coworkers observed that monoexchange occurs with relatively bulky amines (NEt3) and multiexchange increases as the non-bonding repulsion in the vicinity of the nitrogen decreases (NMe3; l,4-diazabicyclo[2,2,2]octane).

The above data clearly demonstrate the influence of tertiary amines on the stabilization of the sulfene intermediates.

It became obvious that in the generation of sulfenes with bases other than trialkylamines should be considered in order to study the behavior of nonstabilized sulfenes. if the formation of sulfene followed the

E2 path without previous nucleophilic attack of the trialkylamine at the sulfonyl group to form the more acidic sulfon- ammonium intermediate , then any other base of equal or higher strength could replace the trialkylamine. Bases should be chosen which fulfill the following requirements: (a) similar or greater base- strength than trlethylamine, (b) low nucleophilicity and (c) as little as possible stabilizing contribution to the sulfene intermediate via labile intermediates of type 66 and/or 6 £. For the planned experi ments, metal hydrides such as sodium or lithium hydride were therefore considered as replacements for the trialkylamines in the sulfene gene ration. This approach appeared particularly attractive since other authors had successfully replaced amines by sodium hydride in similar elimination reactions.

For instance, Atkins and Burgess 75 reported that the inner salt of ethyl N-sulfonylcarbamate (84) is formed in the reaction of car-

bethoxysulfamoyl chloride ( 82.) w ith 2 equivalent of trlethylamine in

benzene at 30°Ct which is trapped by aniline to give the N-carbethoxy-

N-phenylsulfamide (85).

The reaction of carbmethoxysulfamoyl chloride ( 86 ) with sodium hydride

at -78°C in tetrahydrofuran affords the salt 87 which rapidly

decomposes under formation of sodium chloride and the solvent complex

of methyl N-sulfonylurethane ( 88)7e. This species 88 demonstrates

a high degree of electrophilic reactivity in cycloadditlons with alkenes.

86 81

88 The formation of N-sulfonylcarbamate from reaction of carbethoxy-

sulfamoyl chloride and trlethylamine was postulated to proceed via

the E2 mechanism. This was supported by the fact that the reaction of the carbomethoxysulfamoyl chloride with sodium hydride in the same solvent can also generate the N-sulfonylcarbamate. If the sulfene 26 formation from sulfonyl chloride and triethylaraine were an E2 process it should also be accomplished by reaction of sulfonyl chloride with any base of higher basicity than triethylamine.

Metal hydrides represent strong bases, but unlike trialkylamines show lew nucleophilicity. With aliphatic sulfonyl chlorides, the hydride ion can act as a base by abstracting a proton a to the sulfonyl group but can not stabilize the sulfonyl group by the formation of an intermediate similar to complex 66 nor to a charge transfer complex 6j_. Therefore, experiments as outlined in

Figure ^ were undertaken

Figure U:

R F(C — S O — Cl H + 27

A number of sulfonyl chlorides were reacted with sodium hydride in various solvents under reflux conditions. Methanesulfonyl chloride and sodium hydride in refluxing THF or p-dioxane showed no evolution of hydrogen indicating that no sulfene was generated under those conditions. In both cases, the sulfonyl chloride was rearranged.

Experiments were then attempted with more acidiic Qf-hydrogens on sulfonyl chloride; such as, phenylmethanesulfonyl chloride and propensulfonyl chloride were chosen as substrates. Again, under the above conditions the two sulfonyl chlorides failed to react with sodium hydride and the starting components were recovered.

Finally, a sulfonyl chloride containing the strongly electron withdrawing trifluoromethyl group was chosen. In trifluoroethanesulfonyl chloride, the a-hydrogens were expected to be considerably more acidic than in the previously used sulfonyl chlorides. When the trifluoro compound was subjected to reaction with sodium hydride in refluxing p-dioxane, again no evolution of hydrogen was noticed. In order to increase the hydride concentration in the organic solvents, the more soluble lithium hydride was chosen as a base. Again, re actions of the various sulfonyl chlorides with LiH were without success. The attempted reactions of sulfonyl chlorides with metal hydride are summarized in Table 1 (Expts 1-7)*

Another strong base with low nucleophilicity is the anion derived from dimethylsulfoxide (Corey base)77’78. This base, which is con veniently generated from dimethylsulfoxide (DMSO) end sodium hydride, shows considerably higher basicity than trialkylamines. However, methanesulfonyl chloride resisted reaction with Corey base in boiling

THF (Table 1, Expt. 8). These experiments clearly indicate that 28

TABLE 1

Sulfonyl chloride Base Solvent Conditions

CH3 -SO2 -Cl NaH THF Reflux

CsHg -CHa -SO2 -Cl NaH THF Reflux

CH3 -S02-C1 NaH p-dloxane Reflux

CHa-CH-CIfe-SQs-Cl NaH p-dioxane Reflux

CF3 -CH2 -SQa-Cl NaH p-dioxane Reflux

C6 H5 -CHa -SO2 -Cl LiH E ther Reflux

CF3-CHg -SQ2 -CI LiH E ther Reflux

CH3 -SQ2 -CI Na:CHe-S-CH3 THF Reflux tl the bases stronger than trialkylamines fail to abstract a hydrogen

from the g-carbon of a sulfonyl chloride. If sulfene formation were to proceed via the E2 mechanism outlined in Figure k, a t

least the carbanion 85) should have been formed in the reaction of

the sulfonyl chloride with sodium hydride, lithium hydride or the

Corey base. It therefore appears that, in spite of King's kinetic evidence, the E 2 mechanism is less likely and the trialkylamines play a more complex role in the sulfene formation. It could be pos sible that, the previously discussed pseudo-first order data obtained by King and coworkers were in error. The simple fact that in King's rate studies t the rate of formation of chloride ion depends on

the concentration of the trapping agent (aniline) should be enough

reason to evaluate these kinetic data with greater caution. In the general for the reaction:

H R3N + R'R"C - SQ2CI ------> [R'R"C = SQ2 ] + R3N Cl" ■ ® the reaction rate can be expressed as outlined in Eq. (1)

Rate = k[R ’R"CH - Stfe - C1]X [KgN]7 eq. ( l )

In a simple kinetic study to determine the order in trialkyl

amine, the concentration of sulfonyl chloride and the trapping agent

should be kept constant and the rate dependence on the trialkylamine

concentration should be measured. Experimentally, it is expected

that a reaction first order in trialkylamine would show a doubling of

the reaction rate if the concentration of trialkylamine were

doubled. A fourfold rate increase would be observed if the reaction were second order in trialkylamine. We attempted to obtain kinetic data for the reaction of phenylmethanesulfonyl chloride with triethylamine in the presence of p-toluenesulfonyl chloride as trapping agent ,79 by measuring the disappearance of the methyl signal of the sulfonyl chloride by nmr spectroseopy as a function of time. The experi mental results are shewn in Tables 2a and 2b.

The data in Table 2a and 2b clearly Indicate that by measuring the rate of sulfonyl chloride consumption, the reaction rate is not first order in trialkylamine. In fact, the data are in reasonable agreement with a reaction which is second order in trialkylamine. More experimental facts and more accurate measurements would certainly be necessary to obtain reliable results. Also, a wider range of trialkylamine concentrations should be chosen. Experiments of this type are presently in progress.

Considering all the previously discussed experimental facts [(a) isolation of a complex of type , (b) difference in acidity be tween methanesulfonyl chloride and its trimethylamine complex toward

CHaMgBr and (c) failure to abstract arhydrogens from sulfonyl chlorides with metal hydrides and Corey basej the steps outlined in Scheme

5 appear to be a reasonable alternative route for the formation of s u lfe n e .

Scheme 5 : 31

TABLE 2a (sulfonyl chloride/triethylamine/trapping agent = 1:2:5) original concentration of sulfonyl chloride is 2 .0 x 10-% Time after add. Cone, of sulfonyl of base fmins.) chloride consumed ( 10-%)

0 0 .0

30 1.0

60 1.09

90 1.06

120 1.07

150 1.09 180 1.10

TABLE 2b (sulfonyl chloride/triethylamine/trapping agent = 1:1:5) original concentration of sulfonyl chloride is 2 .0 x 10"% Time after add. Cone, of sulfonyl of base (mins.) chloride consumed (x 0~%)

0 0.0

25 0.11

30 0.13

50 0.19 60 0.21

90 0.27

120 0.29

150 0.31

180 ' 0.32 32

The initial attack of the trialkylamine at the sulfonyl chloride pos sibly represents a fast step leading via the charge-transfer complex

90 to the ammonium sulfone 2i • This is followed by a slow step involving the abstraction of a proton from the ^-carbon, either in a direct E2 fashion providing the sulfene 22. or via the zwitterion 22, which is in equilibrium with 22.* Sulfene stabilizing substituents (R and R') should favor the direct formation of sulfene (path i) whereas the presence of destabilizing, electron withdrawing substituents would preferentially lead to the formation of the zwitterion 2£: ■ B. GENERATION OF NON^STABILIZED SULFENES VIA BICYCLIC SULFONES — rsrsfw — ™— — — ■■■■

In recent years, a number of approaches toward the generation of

non-stabilized sulfenes have been attempted. One possible route was

reported by King et al . 31,62 applying flash thermolysis of chloro- sulfonyl acetic acid at 650°C. Other pathways toward non-stabilized

Cl-S02-CHa-C0sH ~ft59°c > [CHa » SOfe] sulfenes, involving retro Diels reactions, were also considered. The simplest reaction leading to a sulfene by a retro Diels-Alder reaction would be the thermal conversion of 3 ,4-dihydrothiopyran- 1, 1-dioxide

(94) to provide butadiene and sulfene .80

+ CHf=-S02

94 95 69

An approximate thermochemical estimate indicated that the reaction should be endothermic by 40-50 kcal/mol. if a molecule were so con stituted that the diene is part of an incipient aromatic ring system, it could be expected to react more readily. Using essentially the q i method of Benson et al, one would expect the fragmentation of bicyclo[2,2,23octadiene (96) into benzene and ethylene to be ^H° a +1 kcal/mol or less, whereas a value of -t^ 9*T Kcal/mol was estimated

t g y — „ ( d ) C H ^ — C t i 2

26 for the conversion of cyclohexene into butadiene and ethylene. In addition, as has been pointed out by Kwart and King82, such a frag-

53 mentation process would be expected to be accompanied by an increase in entropy and hence be favored by an increase in temperature.

The simplest bicyclic sulfone that would give sulfene and an aromatic co-product is 2-thiabicyclo[2,2,2]octa-5,7- diene (100). Our attem pt to o b tain compound 100 from the cycloaddition of &-thiopyran-l, 1-dioxide ( 2 T) 83 with vinyltrimethylanmonium bromide (^ 8) followed by elimination of trimethylammonium hydrobromide proved unsuccessful. We therefore explored the formation

98 99 100

of bicyclic sulfones of type 102 using other dienophiles. Indeed heating of a-thiopyran- 1, 1-d io x id e ( 97) with diraethylacetylene dicarboxylate ( 101) provided dimethyl 2 , 2 -dio x id o - 2 -thiabicyclo£ 2 ,2 ,2 ] octa-5,7-diene-5, 6 -d ica rb o x y la te ( 102) in good y ie ld . This compound was expected to generate sulfene and dimethylphthalate ( 105 ) as the only by-product when subjected to thermolysis.

£ s C Q f H , COzCHi

Cp + C o p * — * 35

While our work was in progress, King and Lewars 84 reported on the thermolysis of 10k, the k-phenyl derivative of 102.

The experiment provided dimethyl biphenyl-2,3-dicarboxylate (105) and sulfene, which was trapped with 1-N-morpholinecyclohexene (107) to give the aminosulfone 106 . 0

10k 106 105

Thermolysis of 10k alone also gave high yields of dimethyl biphenyl-

2,3 -dicarboxylate (105) and sulfur dioxide; the fate of the missing methylene group being unclear. The authors discussed as a possible route initial formation of sulfene and subsequent reactions leading to sulfur dioxide and unidentified methylene fragments. However, our r e s u lts 85 on the thermolysis of the bicyclic sulfone 102 in d ic a te th a t compound 10k possibly underwent a sulfur dioxide extrusion- rearrangem ent rea ctio n as shown in Scheme 6.

As outlined before, heating of arthiopyxan-1,1-dioxide (97) with dimethyl acetylenedicarboxylate (101) at 100°C for 60 hrs. gave good yields of sulfone 102 . Thermolysis of 102 at 230°C for 10 minutes in the presence of 1-N-morpholinocyclohexene (107) provided less than

5$ of the amino sulfone 106 e5’2. When 102 was thermolysed with and without a sulfene trap, thin-layer chromatography (tic) runs in dicated that a mixture of several compounds was obtained. Samples 4

Scheme 6

C02Me I COtMe C 100°

60 hrs- C02Me C“?C02Me 3 ? iOl |02 103 225 225 /h -SO. -SO. -:CH.

m Hi

C02Me CO. Me [1-5] [1.5] hydrogen hydrogen C 0 2Me shift CO. Me shift ON 112 37

taken from the gas phase of the thermolysis tube showed by gas

chromatographic - mass spectral (GC-M3) analysis ( 8 ' porapak-S column) only one component with major peaks at m/e 60 , 64, 48, and 3 8. The mass spectrum was identical with that of sulfur dioxide .36 The G.C.-

M.S. analysis of the crude oil obtained from the thermolysis of 102 gave on the GC-trace one well-resolved peak followed by a minor and a major broadened peak. The mass spectrum of the first eluted com ponent was identical with that of dimethyl phthalate (1C3). The other two G.C. peaks exhibited highly similar mass spectral patterns, showing parent peaks at m/e 208 (CnHi 204+) and intense peaks at m/e 149

(base peak) and 91. The mass spectral fragments at m/e 149 and 91 were indicative of methyl tropylium carboxylate ions (CgHgOa"*") and tropylium ions (C 7'H7+), respectively.

Column chromatography on silica gel of the crude mixture resulted in a partial separation and an enrichment of the different components.

Purified samples which were essentially free of 103 showed UV maxima a t 211 and 275 and i r bands a t 2945 (m) C-H s tre tc h in g ; 1725

C=0 ( s ); 1620 (m) C=C; 1440 (s) conjugate C-C; 1260 (s) C-O-R; 1125

(s) C-0 and loOO (s) C»C cm-1. These UV and ir spectral para meters were in good agreement with data reported in the literature 97’88 for the cycloheptatriene carboxylates, strongly suggesting dimethylcycloheptatriene dicarboxylates in the thermolysis mixture obtained from 102 . Structural assignments and compositional analysis of the components were made by inspection of increasing and decreasing nmr signals in spectra obtained from different column chromatography fractions. Initial fractions contained a mixture of dimethyl phthalate

(103) and dimethyl cycloheptatriene-l,2-dicarboxylate (109). Sub- 38 sequent fractions provided increasing amounts of dimethyl cycloheptatriene-l,7-dicarboxylate (ill) and dimethyl cycloheptatriene-

2,3-dicarboxylate (112). The nmr assignments for compounds 109 ,

111 and 112 are summarized in Table 3 . The fourth isomer, dimethylcycloheptatriene-3,Ij--dicarboxylate was not detected. In tegration of the carbomethoxy region in the nmr spectrum of the crude thermolysis mixture of 102 showed the following composition:

103 (2U*), 102 (18£), 111 (27#) and 112 (31*). Thermolysis of a mixture of 102, (28* ), 102 (33* ), 111 (29*) and U 2 ( 10*) a t 220-

225°C for 1 hour resulted in an increase of 112 (33*) at the expense of compound 109 (7*) whereas the amount of 111 (30*) and 103 (30*) remained practically constant. Heating of a mixture of 103 (11*),

109 (19*)» H I (29*) and 112 (W*) gave a similar relative ratio of cycloheptatriene derivatives as in the previous reaction, indicating that 109, 111 and 112 readily interconvert at 220°C. In all cases, equilibrium already establishes after 20 minutes at 220°C.

The form ation o f compounds 103, 109, 111 and 112 from 102 is of considerable mechanistic interest. A possible formation of the cycloheptatriene derivatives 109, 111 and 112 from 102 could proceed under extrusion of sulfur dioxide to give first diradical

113 followed by a collapse to the norcaradiene derivatives 108 and

110. The reaction could also involve initial formation of dimethyl- norborna-2,5-diene-2,3-dicarboxylate (11^). When the latter com pound was thermolyzed under the same conditions as described for

102, the starting material was recovered exclusively, thus excluding dimethyl norborna-2, 5 -d ie n e -2 , 3 -dicarboxylate (11^) as an inter m ediate in the form ation o f compounds 109, 111 and 112. Another 39

Table 3• NMR Data of Dimethyl Cycloheptatrlenedicarboxvlatea

Compounds ______nmr signals ______

109 2.52 (d,2H, J -7 .0 Hz, C-T p ro to n s); 3-79 (a, 3H, -COaMe)

3.81 (s, 3H, -COaMe); 5.^8 (dtr, m, J=8 .8 and 7 .0 Hz, C-6 protons)

8-33 (dtr, 1H, J“ 8.8 and 3*0 Hz, C-5 proton)

6.82 (brd, 2H, J#*3*0 Hz, C-3 and C-i+- protons)

111 3-61 (s, 3H, -COaMe), 3.81 (s, 3H, -COaMe), l* .l 8 (d, 1H, J*8.0 Hz, C-7 proton)

5.91 (brdd, 1H, J=8.0 and 10 Hz, C -6 proton)

6.b2 (brdd, 1H, J=10 and 6 Hz, C-5 proton)

6.6b (brdd, 1H, J=10 and 6 .0 Hz, C-3 proton) 6.81* (dd, 1H, J*6.0 and 10 Hz, C-lf p ro to n ), 7*38 (brd, 1H, J*6.0 Hz, C-2 proton)

112 2.1*5 (dd, 2H, J=6.5 and 7 .5 Hz, C-7 proton)

3 . 7!* (s, 3H, -COaMe), 3.82 (s , 3H, -COaMe)

5,86 (d tr, 1H, J*9*5 and 7*0 Hz, C -6 proton)

6.35 (dd, 1H, J«9.5 and 5*5 Hz, C-5 proton)

6.52 (tr, 1H, J*7.5 Hz, C-l proton), 7.65 (d, 1H, J-5.5 Hz, C-l* proton)

COz M* 40

attractive mechanism would be a linear f 2 + 2 + 2 1 or a non- LTT a a a a s J lin e a r r 2 + 2 + 2 1 cheletropic sulfur dioxide extrusion- “TT S rr cl rr “ rearrangement reaction (shown for conversion 102 to 110) a lso

leading to the norcaradiene in term ed iates 108 and 110. From the

latter, cycloheptatriene derivatives 109 and 112 would be derived by skeletal rearrangements followed by [ 1. 5 ]-hydrogen shifts80** to give 111 . Sulfur dioxide extrusions in the thermolysis of non- cyclic allylie sulfones 11^ to form rearranged alkenes 116 were first reported by Combe and Stewart 393 and later by Hendrickson £t al.s9b

HC=CH CH— CHR— S 0 2 — CR$

R: allyl or substituted allyl R ': unsubstituted or substituted alkyl or aryl

The concertedness in the S0^ extrusion from bicyclic sulfones of

type 102 is supported by Mock's observation that thermolysis of

the cis and trans stereoisomers of 2 , 5 -di® ethyl- 2 , 5 -dihydrothiophene-

1, 1-dioxide ( 117) occurs with high stereospecificity under forma

tion of the hexa 2,4-dienes ( 118) . 90

, R '= H

b- R~H , R'=CHi The formation of 103 may be formulated either as a sulfene extrusion reaction from 102 or possibly a carbene extrusion reaction of intermediate 108 and 110. Examples for the loss of carbene from cycloheptatriene derivatives via norcaradienes have been de scribed in the literature92. Thermolysis of 102 in the presence of cyclododecene as a carbene acceptor as well as prolonged heating of a mixture of 103, 109. I l l and 112 did not result in an increase of 103. Thus, formation of 103 from intermediates

108 and 110 by a carbene tra n s fe r re a c tio n seems to be an unlikely route. The experimental data indicated that in the thermolysis of 102, two independent processes seem to be com peting: Sulfur dioxide extrusion leading to cycloheptatrienes

109. Ill and 112 represents the predominant route, whereas retro

Diels-Alder reaction under liberation of sulfene from 102 is energetically less favored at 220°C. The presence of a phenyl group at C-h of the bicyclic sulfone 102 surpressess norcaradiene formation and direct the process via the sulfene route exclusively. In the latter case, the fate of the methylene fragments, possibly derived from sulfene, is not clear. In the thermolysis of 102, minor nmr signals typical for cyclopropane ring protons at about 5 0,5 were observed in a short-time thermolysis of 102 at 220°C, but were not detected in the 10 minute reactions. These signals could be due to labile, sulfene-derived, methylene transfer product.

Recently, King et al . 93 reported the thermolysis of 2,2- dioxido-^-phenyl- 2 -thiabicyclo( 2 , 2 ,2 )-o c ta - 5 , 7*diene- 5 , 6 - dicarboxylate anhydride ( 119) to give equal amounts of the biphenyl derivative ( 120) and cyclopeptatriene derivative 121.

King's results confirmed our observations on the thermolysis of

102 and dimethyl 2 , 2 -dioxido- 2 -thiabicyclo[ 2 , 2 , 21o c ta - 7-ene- 5 ,6 -

dicarboxylate (122). Whereas thermolysis of 102 gave mostly SOg-

extrusion products, and the thermolysis of lf-phenyl derivatives of 102.

119 120 121

proceeded predominantly via the retro Diels reaction,

thermolysis of the bicyclic anhydride 119 gave about

equal amounts of SCfe extrusion and retro Diels-Alder products. The reason for this is not completely clear, but the structural influences upon the reaction path can be explained in

terms of the most likely transition state of the reactions.

12k 123 From the previous discussion and the references85,89>90*91*93) it seems reasonable to discuss the SO^ extrusion-rearrangement reactions in terms of a cyclic, concerted mechanism, the tran sition state for which may be presented by 123 ; the c o rre s ponding transition state for the reverse Diels-Alder process is given by 12k . In the reaction of k-phenyl derivative of

102 , the phenyl group is capable of conjugative stabilization of 12k , but not in the case of 123 . This may be a main factor leading to the occurrence of the reverse Diels-Alder re action as the principal thermal reaction for the k-phenyl derivative o f 102 , but not for the u n su b stitu te d compound 102 ; in other words, the resonance of the phenyl group will stabilize the transition state fl2k). The partial SOa extrusion-rearrangement process of 119 could also be due, at least in part, to the same effect. The presence of the cyclic anhydride system leads to a nonbonding repulsion between the phenyl group and the adjacent carbonyl oxygen pre- c venting coplanarity of the phenyl group and the forming inter mediate. In the k-phenyl derivative of 102, the same interaction may be avoided by rotation of the carbomethoxy group, whereas in the anhydride 119. only the phenyl group can rotate. The conforma tion of the phenyl group which minimizes the interaction with the carboxyl group is less favorable for stabilization of 12k and there fore leads to an increase in the energy of transition state 12k in the reaction of 119 relative to the k-phenyl derivative of 102.

In order to get more insight into the competition be- kh tween the retro Diels-Alder reaction and the sulfur dioxide ex trusion of bicyclic sulfones, the synthesis of the two dihydro derivatives, dimethyl 2 , 2 -d io x id e - 2 -thiabicyclo( 2 , 2 , 2)o c ta -

5 -en e- 5 , 6 -dicarboxylate ( 125) and dimethyl 2,2-dioxido-2- thiabicyclo(2,2,2)octa-T-ene-5,6-dicarboxylate (122) were at tem pted.

When subjected to thermolysis, compounds 125 and 122 would be expected to provide cycloheptadiene derivatives, which due to a lack of one double bond for delocalization should be less stable than the cycloheptatriene derivatives.

122

The synthesis of the bicyclic sulfone 125 was attempted by catalytic hydrogenation of 102. It was hoped that for sterlc reasons the tetrasubstituted double bond in 102 would be considerably less reactive toward hydrogenation than the disubstituted double bond. Hydrogenations attempts using different catalysts (Pd/C and Ft^O) were without success; in each case, the starting material was recovered. The failure of the hydrogenation of 102 could be due to the poisoning of the catalyst by the sulfur atom in the molecule. Compound

122 was obtained in about 50 $ yield by heating crthiopyran- 45

1,1-dioxide (2L) with dimethyl maleate at 100-120°C for 75 hours. The sulfone 122 gave i.r. bands at 1750 cm -1 ty p ic a l for a carbonyl group. Bands at 1240 and 1120 cm "1 were assigned to th e SQ2 absorption. The nmr and MS data were also in agree ment with structure 122 . Short-time thermolysis experiments of 122 at 225°C, 250°C and 275°C provided the unchanged starting material indicating that the activation energy of thermolysis for this bicyclic sulfone is higher than that of the previously dis cussed diene 102. Compound 122 was then subjected to ther molysis at 300°c fo r 3 minutes which provided a black solid residue and a small amount of green oil. The evolved gas mix ture was subjected to gas chromatographic-mass spectral (GC-MS) analysis at 70 ev., using an 8 foot porpak S column. Several fractions were detected in the GC trace. The data GC-MS data are summarized in Table 4,

From the data summarized in Table 4, it is clear the ther molysis of 122 at 300°C occurs with considerable fragmentation.

Formation of COg, methyl acetate and methyl ether could pos sibly proceed via a path as shown in the following diagram:

9 i

-f 2C ' i - o c H ^— l ►cH3c q ^ 3 4 cO j U6

Table 4: Hass Spectral Data of the Gas Phase Components from the Thermolysis of (122) at 300°C for 3 Minutes

S tru c tu ra l Fraction Parent mass Base peak m/e (intensity) assignments

1 74 43 74 (74.3) 0 15 (44.8) ^ c h 3 ch 3 0 J 59 (24.8) 44 (14.3)

2 44 44 28 (9.5) CQ2 16 (6 . 0 )

30 28 (28.6) 3 27 CH3 -CH3 30 (24.5)

4 46 45 k6 (52.6) 29 (2 5 . 6 ) CH3 -0-CH3 15 (14.7)

5 64 64 48 (4.8) 65 (5-5) S0s 66 (3 .5 )

6 58 43 56 (52) 57 (25) CH3 -CH2 **CHg-CHq 41 (19) C4H10 4o (8)

62 62 7 47 (44.6) ch 3 ch 3 45 (22.0) V 61 (2 2 . 0 )

8 60 60 32 (24.3) SCO 62 (4)

*-*With the exception of GC band 8, the MS data of all the other bands are identical with the mass spectra of the known compounds.3^ ^7

The detection of sulfur dioxide in the gas phase indicates a process typical for allylic sulfones which evolved SCfe on thermolysis.

Ethane and butane could be formed by a combination of methylene

groups derived from sulfene. Hiraoka 91 reported the formation

carbonyl sulfide from sulfene in the absence of trapping agent and

sometime, ethane and ethylene were formed. In our pyrolysis mix

tu re the compound w ith a mass number 60 could be due to carbonylsulfide. The major i.r. signals for the greenish liquid obtained

from the thermolysis of 122 gave typical C-H stretching at 3000 and 1*4-30; carbonyl absorption at 1800 and 1750 * and ester ab

sorptions at 1270 and 1220; alkene absorption at 1125 and 635 ; aromatic and conjugative double bonds at 1080 and 900 cm-1 .

There was not sufficient material available to run an nmr spectrum. The black solid residue shewed typical alkane absorption at 2950 and 1^50 cm_1; carbonyl absorption

a t 1750 and alkene absorption and conjugative double bonds at

1550, 730 and 720 cm-1. The i.r. data indicated that the

thermolysis residue from 122 contained compounds with con

jugated double bond and methyl carboxylate group. The G.C.-M.S.

analysis of the thermolysis residue gave in the GC-trace a well resolved peak followed by four broad peaks. The mass

spectrum at 70 eV of the component first eluted was in good

agreement with that of dimethyl phthalate (103). At higher

temperature, the initially formed dimethyl 1, 2 -dihydrophthalate

(127) could possibly give dimethyl phthalate (103) by the

loss of hydrogen, formation of the aromatic system 103 being the driving force. The formation of 105 is indicative of a retro

Diels-Alder reaction initially providing sulfene and 127T the

l a t t e r being the precursor fo r compound 105 as outlined in

Scheme 7.

The second component in the G.C.-M.S. showed a base peak at m/e 91 and a parent peak at m/e 210 (1,6). Further intense signals were observed at m/e: 15 ( 15 ), 59 (24.2), 77 (2k.k),

79 (1 3 .M , 92 (2 0 ), 95 (55), 105 (27.3), 118 ( 1 8 ), 150 (29),

I65 (50), 178 (l 6 ). The intense mass spectral fragments at m/e 93 and 210 indicated the presence of a dihydrotropylium ion

(CyHg*) and C 11H14O4 , respectively, which would be in agree ment w ith compounds 128, 129 and 150. The th ird and fourth components in the G.C.-M.S. gave base peaks at m/e 177, parent peaks a t 208 and intense signals at m/e 59, 9 1 , 118 , a p a tte rn typical for the cycloheptatriene dicarboxylates (C 11H12O4 ).

The fifth component was quite similar in the mass spectral pat tern to the third and fourth compounds with a parent peak m/e

208 and a base peak at m/e 177*

The parent peaks at m/e 208 in these compounds indicated the presence of cycloheptatrienes 109 . I l l and 112 which a t high temperature ( 5 00°) could have been formed from 129 and 150 by the loss of hydrogen. Thick layer chromatography of the thermolysis residue of 122 , developed in benzene, resulted in a partial separation of compounds but no definite products

«jould be isolated; one band gave a syrup, the nmr spectrum of which showed signals typical for dimethyl phthalate, and b9

Scheme J: o CH, C O zC H s 3 97 c m > 12 6 0

o y ^ y ^ - O C H3 ? 0 V ^ r O C H 3 ' ^ v - o c h , 128 0 127 A J.

O ^sT o^S © is & 129 130 c> 105 o i a

/" n A o CH, O j r i Art ^ V / ^ ) Y CW3 \ 0 0 / ^ V ^ O C H , 112

w \ ^ ? c h ;

h i possibly 3_niethyl dimethylphthalate. The nmr spectrum of a more polar band indicated the signals typical of the cycloheptatriene derivatives. A mixture of compounds present in the more polar region of the thick layer chromatogram when analyzed by G.C.-M.S. indicated that compounds 103 and 109 , 111 and 112 were indeed produced in the thermolysis of 122 .

Since the gas phase from the thermolysis of 122 showed strong fragmentation of 122 , further purification and detailed analysis of the thermolysis products were discontinued. It is however important to note that in the gas phase thermolysis products of 122 the formation of both sulfur dioxide and sulfene seems to occur. It is therefore reasonable to conclude that in the thermolysis of 122 , analogous to the thermolysis of

102, two routes are competing, sulfur dioxide extrusion and the form ation o f su lfen e as shown in Scheme 2, although in the case of compound 122 a large number of side products are formed.

In the case of sulfone 122 , the activation energy for thermal degradation is much higher than in 102. This could be due to the lack of a double bond necessary for the de localization in the transition state of the reaction. Con sequently, transition states similar to 123 and 12i<- are le ss favorable in the therm olysis o f compound 122 . In the diene system 102 , both double bonds must be involved in the initial step via 123 and 12k but in the case of 122 , only one double bond is available.

Finally, the synthesis of the parent bicyclic sulfone 100 was attem pted by an approach o u tlin e d in Scheme 8. Heating of a-thiopyran- 1, 1-dioxide ( 2£) with maieic anhydride ( 151 ) a t

110cC for 2 hours gave high yields of 152 . Spectroscopic studies and elemental analysis supported the structure of 152 .

The i.r. bands were typical anhydride absorptions at I 85 O, 1175 can*1 and SQg absorption at 1510 and 1250 cm*1. The nmr and mass spec tral data were also in agreement with structure 152 . (See experimental.)

2,2-Dioxido-2-thiabicyclo[2,2,21-octa-7-ene-5>6-dicarboxylic acid (155 ) was obtained by refluxing and stirring the anhydride

152 in 10$ HC1 solution. The mixture, which was a suspension at the beginning, became homogeneous when the reaction was com plete. After removal of the water, 155 was used for the following reactions without further purification.

Oxidative conversions of dicarboxylic acid into olefin with lead tetraacetate are well documented in the literature94. For in sta n c e , Chapman, e t a l . 95 successfully decarboxylated 1- carboethoxy bicyclo[ 2 , 2 , 2 ]-o cta n e- 2 ,3 -dicarboxylic acid ( 15 M to obtain the corresponding olefin 155 in fair yield.

When 135 was reacted with lead tetraacetate, the work-up pro cedure which was analogous to that used in other successful dicarboxylation 52

Scheme 8 :

0 Q +

152 97 151

•SOL

^ electrolytic decarboxylation

100

100 reactions did not provide the desired sulfone 100 . Another attempt involving removal of water from 155 at elevated tem perature under high vacuum, followed by lead tetraacetate oxidation, also failed to provide the desired product. In this case it was possible that 133 had been dehydrated at high temperature by recyclization to give anhydride 152 . An hydrides are known to resist oxidative decarboxylations. The decarboxylation of 155 was then attempted by electrolytic decarboxylation, a process which had been previously reported by Corey and Casanova96, Bacha and Kochi?7, and Elakovich90. The experiments by Corey and Casanova96, and by Bacha and Kochi 97 had been successful in providing the desired olefins. Hcwever, in our experiment the electrolysis mixture turned Into a deep black solution from which no decarboxylation product could be isolated,

Snyder 63 in his semiempirical M.O. calculation suggested that sulfene may be stabilized by electron-donating substituents.

Houk65 using frontier MO energies and total charges cal culated by CNDO/2 , suggested that sufficiently nucleophilic alkenes may add in a stepwise fashion or electron-rich dienes react in a concerted (V*"2) fashion. Electron-deficient dienes may cycloadd in a concerted fashion if the cumulene HO-diene LU interaction becomes sufficiently great. With these idea in mind, cycloaddition between sulfene and an electron deficient diene was attempted by the reaction of a-thiopyran- 1, 1-dioxide ( 97) with sulfene, generated in situ from methanesulfonyl chloride. 5b

Sulfene generated from mesyl chloride and trlethylamine was re acted w ith compound 21. • A fter s tir r in g a t room tem perature fo r

16 hours, the mixture was worked-up to provide only the starting m ateria l 9T_ but no cycloaddition products. The reason might be that the diene is not sufficiently electron deficient.

Reaction of in situ generated sulfene with an electron- rich diene was also attempted by reacting sulfene with 1-methoxy-

1,5 -butadiene ( 156). Although a sim ila r £hn+2) cy clo ad d itio n reaction had been previously reported by Truce and L in",

OR OR

reaction of 156 and sulfene did not give the expected product.

This could be due to the decomposition or polymerization of 156

before undergoing cycloaddition with sulfene; or because of in s u f f ic ie n t n u c le o p h ilic ity o f compound 156 . SOLVENT EFFECT AND NEIGHBORING GROUP PARTICIPATION IN SULFENE FORMATION

In 1911> Wedekind and Schenk 5 reported that the reaction of phenyimethanesulfonyl chloride with triethylamine yielded trans- stilbene and suggested that phenylsulfene (PhHC = SQ 2 ) was an intermediate in the reaction. Similarly, in 1923, Wedekind and co w orkers 100 described the transformation of D-camphor-10-sulfonyl chloride (137) with either triethylamine or pyridine to give D- camphor- 10-chlorosulfoxide ( 138).

o 137 13KJ

The s tru c tu re o f 138 was confirmed by King and co-workers 19 using spectroscopic data. Later, the reactions of D-Camphor-10-sulfonyl c h lo rid e ( 137) and phenyimethanesulfonyl chloride with triethylamine or pyridine were reinvestigated by Strating 79 and King and Durst101.

In 1966, King and Durst 101 confirmed Wedekind's results on the reaction of phenyimethanesulfonyl chloride with triethylamine. In addition, they reported the isolation of cis and trans-oxythiobenzoyl chloride (l^ 6 a) and (l^ b ). This isomeric mixture represented the first example of geometrical isomerism about a carbon-sulfur double bond; as a matter of fact, about any double bond in which one of the atoms is of a higher atomic number than nitrogen.

It was observed by King and Durst 101 that the reaction of

55 56 phenyimethanesulfonyl chloride with triethylamine showed a strong dependence on the solvent used in the reaction. The solvent de pendence data are summarized in Table 5 .

Table 5 :

Solvent Products

Stilbene Chlorosulfoxide Diphenylethylene sulfone Ether + trace +

Cyclohexane trace +

Benzene + + +

The proposed mechanism for the formation of the different products, as formulated for the reaction of phenyimethanesulfonyl chloride with trie th y la m in e , is summarized in Scheme 9 .2 ’79

It is of particular interest that the formation of the chlorosulfoxides 13 8 , 1^6 a and l46b represents the only examples in the literature in which sulfenes react out of the 1,3 -dipolar re sonance form as represented by 139b . Although the solvent- dependence of the reaction of phenyimethanesulfonyl chloride with triethylamine indicates a strong influence of the solvent upon the phenylsulfene intermediate, no reasons were given by King and D urst 101 nor by other authors. From the experimental data it appears that, in non-polar solvents, the 1,3 -dipolar resonance form 139b of phenylsulfene is more favored due to greater internal stabilization of the positive benzylic carbon center by the phenyl substituent, whereas the ylid-form 139a experiences greater external stabilization by more polar solvents, e.g., ether. 57

Scheme 9:

^,CH=sa H 0

^ \ + J 129b (

%H j S j O j ^ s ^ - c i \pr - -J ©-%-$& £*er if 1 S®2 © S H , • ***" F °- s o z u»it (£ )

140 { (2) 1

s I a

l4 l / c II

S O , I lU6fa f > - + ^H ( s r ° ( g r 0*- * ? 142 58

The exclusive formation of D-camphor-10-chlorosulfoxide ( 138)

from D-camphor-10-sulfonyl chloride (13T) and tertiary bases could

possibly be due to steric reasons, that is, the relatively bulky

norbornyl skeleton could prevent the formation of intermediates analogous

to 1^0 and ll+l therefore not permitting alkene formation.

However, we considered the carbonyl function in the D-camphor-10- sulfonyl chloride to be wholly responsible for the exclusive direc

tion of the reaction toward the chlorosulfoxide 13 8 . If intra molecular stabilization of the sulfene intermediate by the carbonyl group were to represent a major contribution, minimal influence of the solvent polarity upon the direction (alkene or chlorosulfoxide) of the reaction would be expected.

From phenyimethanesulfonyl chloride and triethylamine, King

and co-workers had obtained cis and trans-oxythiobenzoyl chloride. However, the (E)- and (z)-D-camphor-lO-chlorosulfoxides ( 158a) and

( 138b) had never been separated. Therefore, we first attempted

to separate the two isomers and obtain them in pure form. Reac

tion of D-camphor-10-sulfonyl chloride with triethylamine in ether

gave a mixture of ether-insoluble triethylanmonium chloride and

triethylammonium camphor- 10-sulfonate, and an ether-soluble brown

syrup. The nmr spectrum of the syrup did not indicate signals be

low 3 ppro excluding alkene formation in the above reaction. The

thin layer chromatogram of the syrup, using silica gel/benzene,

showed two distinct spots with * 0.M5 and R^ ~ 0.37* Column

chromatographic separation on silica gel using benzene/ethyl acetate

( 12:l) as an eluant provided pure samples of the two reaction 59 products. The less polar ccxnpound was tentatively assigned the structure (E)-D-camphor-lO-chlorosulfoxide (138b) on the basis of polarity comparison with compounds l46a and 1^6 b ; then, the more polar constituent had to be (z)-D-camphor-lO-chlorosulfoxide

(I3§a).

Further arguments for the structural assignments of the new com pounds were provided by spectroscopic data. Compound 138b , a waxy solid at room temperature, easily crystallized in the deep freezer, m.p. = 5°C. The i.r. spectrum of 138b showed typical carbonyl absorption at 1750 cm "1 and C ■ S absorptions at 1290,

1130 cm*1. Furthermore, S * 0 absorptions at 1080, 1050 and 1020 cm"1 were observed. The nmr spectrum gave bands at 5 - 1.68 and

1.71 ppm due to the two geminal methyl groups and an envelop be tween 3.0 ~ 1.5 ppm due to protons of the camphor skeleton. The mass spectrum shewed a parent and base peak at 232 indicating a molecular weight in agreement with structure 138b . The more polar isomer 138a was crystalline compound, m.p.

8l - 82°C. The i.r. spectrum showed a typical carbonyl absorption at 1750» a C = S absorption at 1280 and S - 0 absorptions at 1080, 60

1050 and 1020 cm-1. The nmr spectrum gave signals at 6 = 1.61 and

1.65 ppm which were assigned to the two geminal methyl groups. Further more, an envelop between 2.7 - 1.5 ppm was observed. The mass spectrum of 158a gave a parent and base peak at 232 which is in accord with the molecular formula of 138a . The E:Z ratio of the crude mixture is approximately 5 :1). and the combined yield for 138b and

158a is nearly 72$.

In cyclohexane, the reaction of D-camphor-10-sulfonyl chloride with triethylamine gave about 66 $ of a mixture of the (e)- (z)-

D-camphor-10-chlorosulfoxides ( 158a) and ( 138b) showing the same

E:Z ratio as found in ether. These results indicated that, unlike

the reaction of phenyimethanesulfonyl chloride with triethylamine,

the reaction of D-camphor-10-sulfonyl chloride (137) w ith t r i ethylamine shewed no solvent dependence and (E)- and (Z)-D-camphor-

10-chlorosulfoxide ( 138a) and ( 138b) were the main products.

From the above results, it is reasonable to propose that the absence of a solvent effect in the reaction of D-camphor-10-

sulfonyl chloride ( 137) with triethylamine is due to a strong

neighboring group participation of the carbonyl oxygen of the

camphor moiety. The lone pair electrons of the carbonyl oxygen of

the sulfene intermediate 147 could stabilize the positively polarized

carbon center of the sulfene. This intramolecular stabilization

should be reasonably strong and independent of the solvent polarity.

147 147a 61

Effects of this type have been well documented in the litera ture. For instance, Pasto and Serve 102 provided examples for neighboring group participation by a carbonyl group. The data in dicated that the p-TT bonding pair of electrons of a keto carbonyl group were involved in the formation of a carbocation in a silver ion assisted solvolysis of chloroketones of the type lh 8 via the stabilized intermediate lh9 . In the solvolysis of a series of chloroketones,

lk-9 the rate of solvolysis decreases as n increases from 2 to If. This is expected if the ionization process nvolves the formation of a cyclic intermediate, the ring sizes being 5 ~» 6 -, and 7-membered, re s p e c tiv e ly , as shown in l^t -9 •

Paquette and Scott 103 studied the kinetics of the acetolysis of oxocan-^-yl brosylate ( 150 ) and found a sevenfold rate retardation relative to cyclooctyl brosylate (151). In striking contrast,

/ 0 5 s w - N O ? O C r " 150 151 152 62 oxocan- 5 -y l 3 »5 _di-nitrobenzoate ( 152 ) hydrolyzed at a i+8, 000-

fold rate enhancement relative to 151 . Paquette and Scott in

terpreted the rate increase to be due to the intramolecular participation of oxygen stabilizing the carbocation intermediate derived from 152 , whereas there was no such stabilization in

lg l and 150 since in those compounds the oxygen only acts as an electron withdrawing group which destabilizes the carbocation, consequently, decreasing the rate of solvolysis. In a similar

study, Oae investigated the solvolysis of v-bromobutyl phenyl ketone ( l^ ) and observed that 155 solvolysed considerably faster than 1+-phenyl-1-broraobutane . Oae derived from the

rate enhancement that intramolecular stabilization of the carbo- ro4 cation by the carbonyl oxygen must occur as shown in 155 •

0 II (o>—C —CH-—nH— n u —R-

1 5 it

The above experimental evidence strongly favor the concept

that the carbonyl oxygen can participate in the stabilization of

carbocationic centers. On the basis of these data it is rea sonable to explain the absence of a solvent effect in the

reaction of D-camphor- 10-sulfonyl chloride with triethylamine 63

by the participation of the carbonyl oxygen in the stabilization of the positively charged carbon center of the sulfene intermediate, as shown in lVf .

In order to study the effect of the carbonyl oxygen in greater detail, the reaction of camphane-10-sulfonyl chloride (157) with trie th y la m in e was considered. Clemmenson red u ctio n o f D-camphor-

10-sulfonic acid gave camphane-10-sulfonic acid (1 56 ). Reaction of 156 with phosphorus oxychloride provided camphane-10-sulfonyl chloride in low yield.

156 157

The i.r. spectrum of 157 indicated the disappearance of carbonyl absorption at 1750 cm*1 and showed typical SO 2 bands a t

1250 and lloO cm"1. The nmr spectrum gave a sharp methyl sig n a l a t 0 .S 5 ppm for the two methyl groups, a singlet at 3*75 ppm

for methylene protons and an envelop at 1.8l ~ 1.1 ppm. In con 6k

trast to the chiral D-camphor-10-sulfonic acid, camphane-10-

sulfonyl chloride ( 157 ) is achiral, thus explaining why the nmr

absorption for the methylene protons adjacent to the sulfonyl group represents a singlet in l*jj , whereas the two methylene protons adjacent to the sulfonyl group in D-camphor-10-sulfonyl c h lo rid e ( 157 ) are diastereotopic and therefore appear as two

one-proton doublets (J**ll Hz). The mass spectrum of the sulfonyl

c h lo rid e 157 showed no parent peak but strong peaks at 138 and

137 which are due to the camphane fragments.

The reaction of camphane-10-sulfonyl chloride (157) with triethylamine in ether formed a white precipitate which gave an nmr spectrum identical with that of synthetic triethylamnonium chloride. The nmr spectrum of the crude syrup, obtained after evaporation of the solvent, showed a more complicated spectrum.

An absorption at 9*75 ppm* appearing as a sharp singlet, indicated

the presence of an aldehyde. Purification of the reaction mixture by column chromatography (silica gel/benzene) was attempted. How ever, decomposition occurred forming a blue-green band on the column. The reaction was repeated a number of times with the same observation, that is, decomposition of the material on the column. The earlier fractions of the chroma

tographic runs provided small amounts of a crystalline material which was further purified by sublimation to give pure crystalline compound, m.p. 152-15^°.

In the i.r. spectrum, a typical carbonyl absorption was ob served a t 1730 can"1 and the nmr spectrum gave a sharp singlet at 65

9*75 PP®> indicated an aldehyde group. Further ranr signals, a

singlet at 1.28 ppm (methyl groups) and an envelop between 1.3

and 2.38 ppm, were in agreement with the presence of a camphane skeleton. The integration of all proton signals gave a ratio

1:6:9* From the above spectral data the new compound was tenta

tively assigned structure 160 .

- s o .

159

The mass spectrum gave a strong parent peak at m/e 152 which re presents the molecular weight of aldehyde 160. The base peak at ra/e 151 (M-l) provided further evidence for the presence of an aldehyde. An Intense signal at m/e 137 must be due to the loss of a methyl group from the parent molecule, the aldehyde 160.

Another intense signal at m/e 123 is indicative of the camphane skeleton after loss of CHO (29 m.u.). The elemental analysis data were not completely in agreement with the theoretical values; the difference could be due to air-oxidation of the aldehyde to the corresponding carboxylic acid.

The formation of aldehyde 160 can be visualized as an intra molecular reaction of sulfene to give oj-sultine 159 which was formed from the sulfonyl chloride 157. Compound 162 is probably

formed from the zwitterion l6l on the silica gel during the chromotographic procedure. As in Scheme 10, it is quite reason- a b le fo r compound 158 to form 160 since no Intramolecular stabili zation of the sulfene 158 is possible.

S c h e m e 1 0 : R-C H —$C£-CI

151

\ - r ° v0 t v 158 1 no

\- - s q - M * 3 -SO

l&l 159 \ w ' I

RCH,-SC% flEt3 H* > o t CSOJ

162 l£2

Cl ^ j O - S 0 R - f i - H - < 0 ) 67

In the case of the generation of sulfene from D-camphor-

10-sulfonyl chloride, sulfene stabilization by the carbonyl oxygen on positively polarized carbon center of sulfene seems to occur.

Since intramolecular stabilization of the sulfene 158 is not pos sible, it stabilizes by aldehyde formation via sultine 159. Due to stabilization, intermediate lVfa has a longer lifetime to allow the chloride ion to attack the positive carbon center of the sulfene with formation of D-camphor- 10-chlorosulfoxide via the mixed anhydride. In the case of camphane-10-sulfonyl chloride (157) the absence of a carbonyl group forces the sulfene 158 to acquire stabilization through formation of the zwitterion l 6 l or the sultine

159 and, subsequently, the aldehyde l60. The lack of the carbonyl group depresses the caraphane- 10-chlorosulfoxide ( 165 ) pathway.

Reactions of sulfonyl chlorides with triethylamine without trapping agent have been carried out before. Fischer 68 was not able to isolate a definite product in the reaction of methanesulfonyl chloride with triethylamine in ether. Paquette 105 and co-workers observed that 9-fluoroene-sulfonyl chloride (164) with triethylamine gave a dark purple solution

loU 182

which was believed to be a solution of sulfene 165 which poly- 6 8 tnerized daring Che work-up procedure. Again, a pure product from the reaction of 164 with triethylamine could not be isolated.

When 164 was stirred with triethylamine for 2-4 hours, followed by the addition of N,N-dimethyl~l-isobutenylamine ( 166 ) no cycloaddition products were obtained, indicating that sulfene 165

166 had undergone further reactions.

When camphane-10-sulfonyl chloride (157) was reacted with triethylamine until the methylene signal of the sulfonyl chloride had completely disappeared in an NMR run, reaction with aniline did not give the sulfonamide 167 , clearly indicating that the

167 sulfene, after its formation had also undergone further reactions.

A compound was formed, which gave in the i . r . spectrum a

C = N absorption at 1600, 1720 cm "1 and bands due to the phenyl group a t 1175» 1100 and 780 cm"1. The nmr spectrum showed signals typical of a phenyl group ( 6 * 7 . 27)> »n envelop at 2 .0 ~ 1.4 ppm,

(camphane ring) two methyl group signals at 1.0 ppm and a proton sig n a l a t 5*5 ppm. The mass spectrum o f the new compound gave a parent peak at m/e 227 end a base peak at 109. It also showed 69 intense signals at m/e 152 which is indicative of the loss of a phenyl group from the molecular ion. The peak at m/e 157 was assigned the camphane skeleton (152 m.u.) minus a methyl group. A signal a t m/e 123 indicated a dimethyl norbomyl ion and peaks at m/e 95 and 9? were assigned the norbornane and aniline fragments, res pectively. The above spectroscopic data are in good agreement with th e S c h iff's base 168 . P u rific a tio n attem pts o f compound 168 by column chromatography over silica gel led to the hydrolysis of

168 , one of the products being the aldehyde 160 , The Schiff's base 168 is most likely derived from the aldehyde 160 o r pos sibly its precursor 159 •

160 168

On the basis of the above results together with the failure to obtain camphane- 10-chlorosulfoxide ( 165 ), it is reasonable to conclude that the carbonyl group in the D-camphor-10-sulfonyl c h lo rid e ( 157 ) does stabilize the sulfene via intramolecular par ticipation of the lone pair electrons of the camphor oxygen. TO EXPERIMENTAL

General Information

The reagents and solvents used in all reactions were reagent grade conmercial chemicals. The melting points were measured on a

Fisher-Johns Melting Point Apparatus and were uncorrected. Infra red spectra (i.r.) were recorded on a Perkin-Elmer Infracord Model

1J7- Polystyrene was used as the calibration standard. Ultra violet spectra (UV) were run on a Cary 1^ Spectrometer. Proton nuclear magnetic resonance spectra (nmr) were recorded on a Perkin-

Elmer Model R12B (60 M Hz) spectrometer or Varient HA-60 (60 M

Hz) and V arient HA-100 (100 M Hz). Mass sp e c tra (M.S.) were determined on a Hitachi Perkin-Elmer Model RMS-4 Mass Spectro meter using ionization energy of 70 eV. Gas chromatography trace-mass spectra (G.C.-M.S.), G. C. were done on Perkin-Elmer

990 and the mass spectra were detected on the above RMS -14-.

Microanalyses were carried out either by Mr. R. Seab, L. S. U.,

Baton Rouge, or by Galbraith Laboratory, Knoxville, Tennessee.

Anhydrous ether was dried over metallic sodium. Tetrahydro- furan (THF) and p-dioxane were refluxed with calcium hydride, distilled and stored over calcium hydride. Dimethyl formamide (DMF) was dried by stirring with molecular sieves and distilled. Di- methylsulfoxide (DMSO) was refluxed with calcium hydride and dis tilled in vacuo. Triethylamine was purchased from Backer Chemical

Company and distilled with a small amount of ethyl isocyanate.

Sodium hydride was purchased from Alfa Inorganic Ventron as a 57% suspension in paraffin oil* the sodium hydride was washed with

71 72 ether and then used immediately. 1-N-Morpholinocyclohexene was prepared by the method reported by Domaschke106. Methanesulfonyl chloride, reagent grade, was purchased from Fisher Chemical

Company, and was p u rifie d by d i s t i l l a t i o n . 2 ,2 ,2 -T riflu o ro e th a n e - sulfonyl chloride was purchased from Willow Brook Labs, Inc. and was used without further purification. Phenylmethanesulfonyl chloride was prepared by the method reported by Johnson and

Ambler 107 in If 1 $ yield; nmr, 6 ^ ^ ^ : ^.82 {S. 2H, -CHa-SOa-);

7.^8 (S. 5H aromatic protons). 2-Propene-l-sulfonyl chloride was synthesized by the method of Johary and Oweer 108 in 89$ fPUO yield; nmr S^™t : 5-73 (m, 1H, -CH-); 5 .6 (m, 2H, CHa«); k.J>6 (d, J=U.5Hz, 2H, -Ctfe-S02-). 73

I. Reactions of Sulfonyl Chlorides with Metal Hydrides

I-A. Reaction of methanesulfonyl chloride with sodium hydride in THF.

In a 250 ml, three-necked round-bottomed flask, fitted with a reflux condenser with a drying tube and a dropping funnel,

3.12 g. (0.13 mol) of sodium hydride, 9»732 g. (0.11 mol) of cyclohexene and 80 ml of THF were mixed. Under stirring with a magnetic stirrer, 11. 1+5 g. (0.1 mol) of freshly distilled methane sulfonyl chloride was added dropwise at room temperature. Pro visions were made to trap the gas which might be formed in the reaction; the gas was measured by displacement of water in a cylinder. After completion of the addition of the methanesulfonyl chloride, the mixture was refluxed for 4 hours. There was no evolution of hydrogen within experimental error due to tempera ture change. The reaction mixture was cooled to room tempera ture and the excess of sodium hydride filtered off. The filtrate was treated with 95 $ ethanol to destroy dissolved excess sodium hydride. Evaporation of the solvent from the filtrate left 10 g. of black residue. The nmr spectrum of the black residue was identical with that of the starting methanesulfonyl chloride.

I-B. Reaction of methanesulfonyl chloride with sodium hydride in

p-dioxane.

A mixture of 2.88 g. (0.12 mol) of sodium hydride,

8 A g. (0.1 mol) of dihydropyran in 100 ml of p-dioxane were reacted with 11. 1+5 g* ( 0.1 mol) of methanesulfonyl chloride as described in (IA). Only the starting materials were recovered. 7^

I-C. Reaction of phenylmethanesulfonyl chloride with sodium hydride

in THF.

A solution of 20 g. (0.105 mol) of phenylmethanesulfonyl chloride in 50 ml of THF were added into a mixture of 3*12 g.

(0.13 mol) of sodium hydride and 9*2 g. (0.11 mol) of cyclohexene in I 5 O ml of THF as described in (IA). After work-up, only the starting materials were recovered,

I-D. Reaction of phenylmethanesu1fony1 chloride with sodium hydride

in p-dioxane.

The reaction described in I-C was carried out in boiling

P-dioxane. Work-up procedures provided the unchanged starting m a te ria ls .

I-E. Reaction of 2-propene-1-sulfonyl chloride with sodium hydride in p-dioxane.

A mixture of I .67 g. (0.07 mol) of sodium hydride and

5.0 g. (0.05 mol) of dihydropyran in 100 ml o f p-dioxane were reacted with 7 g. (0.05 mol) o f 2-propene-l-sulfonyl chloride under conditions described in (lA). No hydrogen evolved and the starting materials were recovered.

I-F. Reaction of 2.2.2-trifluoroethanesulfonyl chloride with

sodium hydride in p-dioxane.

A mixture of 1.3 g. (0*015 mol) of dihydropyran and

1.0 g. (0 . 0^ mol) of sodium hydride in 100 ml o f p-dioxane were reacted with 1.83 g* ( 0.01 mol) of trifluoroethanesulfonyl chloride as described in (lA). No cycloaddition product was formed. The labile starting sulfonyl chloride decomposed during the work-up 75

procedure. The reaction was also attempted in DMSO as a solvent.

The addition of sulfonyl chloride to the DMSO solution resulted

in a violent explosive reaction and a blow-up. Violent reactions

between DMSO and acyl chlorides had previously been reported in

the literature . l09t110*111

I-G. Reaction of phenylmethanesulfony1 chloride with lithium

hydride in anhydrous ether.

A solution of 5*7 g» (0.03 mol) of phenylmethanesulfonyl chloride in 25 ml of anhydrous ether was added dropwise into a s lu rry o f 0.3 g. (0.033 mol) o f 90$ lithium hydride in 50 ml o f anhydrous ether. The mixture was stirred at room temperature

for 18 hours, without evolution of hydrogen. Excess lithium hydride was filtered off, the filtrate was washed with water;

then the aqueous solution was extracted twice with 50 ml anhyd rous ether. The combined ether washings were dried over anhydrous magnesium sulfate and the magnesium sulfate was filtered off.

Removal of the solvent left a white solid, the nmr spectrum of which indicated the recovery of the starting materials, phenyl methanesulfonyl chloride.

I-H. Reaction of 2.2.2-tri fluorethanes ulfonvl chloride with

lithium hydride in anhydrous ether.

A slurry of 0.107 g. (0.012 mol) of 90$ lithium hydride in 50 ml of anhydrous ether was stirred with 0 . 92^ g.

(0.011 mol) of dihydropyran. To this solution, 1.82 g. (0.01 mol) of 2 ,2 , 2 -trifluoroethanesulfonyl chloride in 25 ml o f anhydrous ether wae added dropwise, and the mixture was stirred for an 76

additional 18 hours without detecting any evolution of hydrogen.

Work-up of the reaction mixture was analogous to experiment (iG).

The residue turned brown with decomposition of the trifluoroethanesulfonyl chloride.

I-I. Reaction of Corey base with methanesulfonyl chloride.

A mixture of 2.88 g (0.12 mol) of sodium hydride in

1+0 ml of anhydrous THF was placed in a 250 ml three-necked round-bottomed flask, fitted with a condenser and a drying tube on the condenser, a magnetic stirrer and a dropping funnel. Di- methylsulfoixde, 7*8 g. (0.1 mol) was added to the mixture and the temperature was raised to 70-8O°C to generate the Corey b ase . 77’78 After completion of the reaction, the solution was cooled to room tem perature and 8. 1+ g. (0.1 mol) of dihydropyran was added. The mixture was then stirred at 0° under dropwise addition of 11.5 6 « (0*1 mol) of methanesulfonyl chloride in

1+0 ml of THF. Upon addition of the sulfonyl chloride, the tem perature of the reaction mixture rose to about 20°C then dropped to 9°C and maintained that temperature. After the addition of the sulfonyl chloride, the reaction mixture was stirred at room temperature overnight and then was hydrolyzed with water. The aqueous solution was extracted with ether and the combined ether extracts were evaporated. The residue was extracted with chloroform, and the aqueous solution was extracted twice with chloroform. The combined chloroform extracts were dried over anhydrous magnesium sulfate and filtered off. Evaporation of the solvent provided about 2 g. of crude syrup, the nmr spectrum TT of which did not indicate the presence of sulfene cycloaddition product.

I-J. Kinetic study for reaction of phenylmethanesulfonyl chloride

with triethylamine in deuterated chloroform.

A solution of O.O 58 g. (0.2 mmol) of phenylmethane sulfonyl chloride and 0.190 g. (l.O mmol) of p -to lu en esu lfo n y l chloride in 1 ml of deuterated chloroform was transfered to an nmr tube, and 0.020 g. (0.2 mmol) of triethylamine in 0 A ml of deuterated chloroform was added to the tube; then, the tube was sealed immediately. After thorough mixing, the tube was placed into the nmr spectrometer and an nmr spectrum was taken every 30 minutes. Each 30 minutes the decrease of the methylene group in the phenylmethanesulfonyl chloride was determined by the decrease of the integrated area. The same operations were repeated in an experiment with the double amount of the triethylamine and leaving all other concentrations constant.

II. Synthesis of Thietane-S, S-Dioxides

II-A. Reaction of methanesulfonyl chloride with 1-N-mopholino-

cyclohexene and triethylamine . 34

In a 500 ml round-bottomed flask with dropping funnel and reflux condenser, a stirred solution of 16.7 g. (0.1 mol) o f

1-N-morpholinocyclohexene and 10.0 g. (0.10 mol) of triethylamine in 200 ml anhydrous dioxane was reacted with 11.5 g* ( 0.1 mol) of methanesulfonyl chloride by a dropwise addition. After the addition of sulfonyl chloride, the reaction mixture was allowed 78 to stir at room temperature for 15 hours. It was then poured into a 200 ml ice-water mixture and the aqueous mixture was extracted with ether. A white solid which was insoluble in both water and ether, was filtered off and dried to give 8.5 g. (34. 5 $) o f sulfo ne 169 m.p. IJZ-ljk0 ( L it.34 133°-13J+°C), NMR, sER?13 ; XMb k. 3 , ( t , lKa ), 3 .9 ~ Jj-.l (s , 2H^), 3 . 65 , ( t , 1+Hd), 3.55 (m, ^H .),

2.65 (m, ^He ), 1.7 (m, lfHf ).

II-B. Reaction of 2-propene-1-sulfonyl chloride with 1-N-

morpholinocyclohexene. 112

A solution of 3-85 g. (0.035 mol) of 1-N-morpholinocyclohexene and 3-5^ g. ( 0*035 mol) of triethylamine in 150 ml anhydrous tetrahydrofuran was cooled to 0°C. Under stirring, ^.95 g. (0.035 mol) o f 2 -propene-l-sulfonyl chloride in 20 ml o f anhy drous THF were added dropwise. The mixture was then stirred at room temperature for one more hour. The white precipitate was filtered off and washed with ether then with water to solubilize triethyl ammonium chloride. The ether washings were concentrated to give predominantly 3 *(l"morpholinocyclohexene)-propene- sulfonate and some cyclosulfone. The residual solid, after washing several times with water, was dried to provide 2 A 5 g- (25 . 6 $) of sulfone 1J0 m.p. 132-134°C (L it.112: m.p. 136°C) nmr,

6CDC13 : V 5,96

III-C. Reaction of 2.2,2-trifluoroethanesulfonyl chloride with

1-N-morpho 1inocyclohexene and triethylamine.

A solution of I .67 g. (0.01 mol) of 1-N-morpholino- 79 cyclohexene and 1.01 g. (0.01 mol) of triethylamine were stirred in Uo ml of anhydrous ether and 1.82 g. (0.01 mol) of 2,2,2- trifluoroethanesulfonyl chloride in 15 ml o f e th e r were added dropwise under formation of a white precipitate. The resulting mixture was stirred for 2 hours. Filtration of the solid residue gave 1.308 g. (95$) of the triethylammonium chloride. Evapora tion of the solvent provided 2.32 g. of a yellow solid which was recrystallized from cyclohexane to give 1.68 g. (55 * 8$) of 0 KBr pale yellow needles, which were sulfone 171 m.p. 83-85 C. I.R . v IS 13X 1320, 1250 (CF3), 1335, 1120, (Spa), 1090 (C-0), 920-1000 cm "1

(C-N); NMR, the nmr spectrum of the yellow solid showed cis and

rtiw C trans cycloaddition products. : U.72 (q, 1H, J=9.0 Hz)

Hb, 5-^9 (t, 1H, J-U.O Hz) H , 3.71 (t, UH, J-5.0 Hz) 3.98

(bt, UH, J=5.0 Hz ) H , 2.68 (m, 2H, Hf ), 2.85 (m, 2H, H ) 2.20 § J- c (bm, 2H, H^), 1.75 (b*n, 2H, He). The mass spectrum gave a parent mass at 313 (25*7) and base peak at 167, intense signals at Ul

(15.6), 55(10.3), 68 ( 10. 6 ), 78( 10. 0 ), 80(15.1), 82(11.8), 166(19.1),

168(22.1). Anal, calcd. for Ci^HigNFaQsS: C, U6.01; H, 5-75; N, U.U 7

F, 18.21; S, 10.22; 0, 15.33. Found: C, U 5 .U5 ; H, 5.78; N, U. 63 .

III. Synthesis of ry-Thiopyran-1.1 -dioxide (97)63

Although the procedure for the synthesis of crthiopyran-

1, 1-dioxide was published in a communication83, a more detailed experimental description is outlined as follows.

Ill-A. Preparation of glutaraldehyde.

A 25$ aqueous solution of conmercially available glutaraldehyde (MCB Chemical Company) was saturated with sodium

chloride and filtered. This saturated solution was extracted

twice with methylene chloride. After further saturation of the aqueous layer, it was again extracted with methylene chloride.

The methylene chloride extracts were evaporated to give a deep yellow, thick syrup which solidified after standing at room temperature overnight. This polymeric glutaraldehyde was de polymerized by heating and distilled at ^5-55°C/0.ImmHg just before its use for the subsequent reactions.

III-B. Preparation of v-thiopyran.

A solution of 100 g. (l.O mol) of glutaraldehyde in

500 ml of dichloromethane was placed into all. three-necked round-bottomed flask, fitted with a gas inlet, gas outlet, and a thermometer. The solution was cooled to -^0°C; then a mixture o f H2S/HCI in 1:2 ratio was passed over the surface of the solu tion for about 5 hours. During this time the temperature of the solution was maintained at -20°C to -30°C. Then, the solution was cooled to - 65 °C and the ice formed in the reaction was quickly filtered off. The filtrate was left at room temperature over anhydrous magnesium sulfate. After filtration, the methylene chloride was removed by passing a nitrogen stream over the mix ture, providing a thick yellow residue. N,N-diethylaniline

(300 g ., 2 ,0 mol) was quickly added and the mixture was heated to 135°C for 30 minutes, then cooled to 80°C and the y'thiopyran was distilled off at if0-80°C/13 mmHg. The highly unstable V" thiopyran was converted directly to the tetrabromide without 81 further purification.

III-C. Preparation of 2,5.5, 6 -tetrabromotetrahydrothiopyran (172).

The v-thiopyran obtained from the above re a c tio n was dissolved in 75 ml of chloroform and cooled in a dry ice-acetone bath to about -40°C. A solution of 10 ml of bromine in 75 ml of chloroform was added until the reaction mixture remained red- brown, The precipitated tetrabromide was immediately filtered off. Evaporation of the filtrate gave a brown residue which was washed with chloroform and then filtered. The crude residue was recrystallized from carbon tetrachloride to obtain white needles, 30*27 g* ( 7*3$, based on glutaraldehyde), m.p. 132-

134°C (L it.63, l43-l44°C). NMR, 5™ci3 : 5*2 J=s5Hz, 2H&);

4.7 (dd, J=5,4.5Hz 21^); 3*0 (t, J-4.5, 2H.).

III-D. Preparation of 2.5.5. 6 -tetrabromotetrahvdrothiopvran-1.1-

dioxide ( 173).

A suspension of 12.54 g. (0.05 mol) of 2,3,5,6-tetrabromotetrahydrothiopyran in a mixture of glacial acetic acid

(200 ml) and 40 ml of acetic anhydride was oxidized to the sul fone 173 using 30# hydrogen peroxide. The suspension was heated to 70° and under stirring 30# hydrogen peroxide was added slowly at such a rate that the temperature remained at 70 + 2°C without external heating. When all the solid was dissolved and the temperature of the reaction mixture dropped upon further addition of 30# HqOz , the temperature of the reaction mixture was maintained at 70°C by external heating and several drops of the peroxide were occasionally added. The progress of the 82

reaction was monitored by thin layer chromatography (tic). The

temperature should not be raised until the starting material is

completely oxidized; otherwise uncontrolled decompositions occur.

It usually required two days for the complete oxidation of starting material to the sulfoxide intermediate. Then, the temperature o f the m ixture was ra is e d to 100°C and 30$ HaO^ was o ccasio n ally added until the sulfoxide was not detected by tic. [The R^ values (silica gel/benzene, 10 cm plate) sulfide: 0.7; sulfone:

0.3 and sulfoxide: 0.1] Then, the mixture was cooled to room

temperature and the solution was tested for excess peracid by shaking an aqueous KI solution with carbon tetrachloride. Re moval of the solvent in vacuo provided a yellow solid residue which was washed with a small amount of methanol and then further purified by column chromatography (silica gel/benzene) to give

7.07 g. (60$) o f 172 m-P- 180-182°C (Lit.03, I8l-l8l.5°c). TMS The nmr spectrum conformed the structure. 5 CD3COCD3 : 5 M (d, J«UHz, 2Hc ). J-5 .5 Hz, 2H3 ); UA,

III-E. Preparation of /v-thiopyran-1.1-dloxide (97).

A stirred suspension of 2.0 g. of zinc powder and

36.0 g. (0.08 mol) of 2,3,5,6-tetrabromotetrahydrothiopyran-

1 ,1-dioxide 173 in 200 ml o f 90$ ethanol (in a nitrogen atmos phere) was gradually heated to boiling. The heating source was removed and 22 g. of zinc dust were introduced at such a rate

that the reaction mixture kept boiling. After the addition of zinc dust was complete, the reaction mixture was refluxed for one more hour during which its appearance turned from brown to green. The mixture was cooled to 0°C and filtered. The light yellow filtrate was concentrated in vacuo to give a brown oil which was intensively stirred with 100 ml of water for one hour.

The insoluble yellow solid was filtered and washed with a small amount of water.

The filtrate was acidified with concentrated hydrochloric acid and extracted four times with chloroform, 25 ml each.

The aqueous layer was then stirred with benzene for 12 hours.

The combined benzene and chloroform extracts were dried over anhydrous magnesium sulfate, filtered and concentrated to give a brcwn oil which crystallized in the refrigerator. Purifica tion by column chromatography (silica gel/benzene) gave 4.0 g.

(40#) of pure (100), m.p. 55-56°C (Lit83: 56 ~ 57°C). NMR,

5c s c i3 : 3 'k2> (bd’ 2V ' 6,28 (m’ 2H)’ 6,6 fm’ 2H)* III-F. Synthesis of dimethyl 2,2-dioxido-2-thiabicyclo[ 2 .2.21

octa-5.T-diene-5. 6 -dicarboxylate (102).

A mixture of 1.2 g. (1.09 mmols) of a-thiopyran-1,1 dioxide (97) and 2.5 g. (1.J6 mmol) of dimethyl acetylenedicarboxylate was kept at 100°C under nitrogen iri vacuo for 60 hours.

The progress of the reaction was followed by nmr. After 3 hours, 10# of 102 was formed; after 24 hours, 60# and after

60 hours, the nmr signals for the starting thiopyran-1,1- dioxide (97) had disappeared. Excess dimethyl acetylene di carboxylate was removed in vacuo. The dark brown residue crystallized from 5 of ethyl acetate at -20°C, providing

1.62 g. (60#) of the bicyclic sulfone 102 . Recrystallization 34

from ethyl acetate gave colorless crystals, m.p. 108- 110°; i r ,

max 1T55 (-COOCH3 ), 1652, 1656 (double bonds), I 305 and I I 55 cm' 1

(SQs-stretching vibration); nmr, 6^ ^ ^ : 2.75 (d, J , 2 . 7Hz),

2.77 (d, J , 3.0Hz, -Clfe-SOa-); 3.85 (s, -COOCH3 ); 4.32 (brdq, J ,

5.5; 3-0 Hz, H-4); 5.06 (dd, J , 5 . 5 ; 2.0Hz, H -l); 6.67 (m , H-7 and H- 8).

Anal. Calcd. for C^H^OsS: C, 48.53; «, 4.44; 0, 35-27;

S, 11.75; (MW, 272). Found: C, 48.59; H, 4.23; S, 12.00 (MW- MS. 272).

III-G. Thermolysis of dimethyl 2,2-dioxido-2-thiabicyclo f2,2.21-

octa-5.7-diene-5. 6 -dicarboxylate ( 102).

In a pyrex tube, sealed with a septum, 200 mg of 102 were heated for 10 minutes in vacuo at 220-225°. With a syringe, a 3 ®1 gas sample was taken from the tube and subjected to GC-MS analysis using an 8 * porapak-S column with temperature programing between 60° and 200° with 6 ° per minute. Sulfur dioxide was the only gas detected by GC, being identified by MS spectroscopy.

The nmr spectrum of the brown, oily residue (156 mg) in d icated a mixture of at least 3 components. The GC-MS analysis using a 12* SE-30 column at 200° three well-resolved bands were found in the GC trace. The first eluted component shewed a mass- spectral pattern identical with that of dimethyl phthalate 102 a s .

Lew resolution MS [70 eV, ra/e (intensity)] peaks > 2$ of the base peak are recorded.

2nd band: 208(3.6); 207(4.5); 195(14.l); 178(3 . 6 ); 177(34.0);

176 (31 . 8); 164(3.2); 163 (6 . 8); 161(9.1); 150 ( 11. 8); 149(100); 11*8(10.0); 133(11*. 1 ); 121(6 . 8); 119(27-3); 118(24.0);

106 (6 . 3 ); 105 (28. 6 ); 92(7-3); 91(72.7); 90(43. 0 ); 89(29.6 );

79(1*.l); 78(9.5); 77(23.2); 65(13.2); 64(6.4); 63 (21. 0 );

62 (6 . 8 ); 59(25.5); 51(11.**); 50(4.6); 45(4.1); 39(13-7); 15 (11. 8).

3rd band: 208(2.4 ; 207(2.4); 195(2.6); 177(1**. 8); 176 (12. 0 );

164(2.6); 163 (2 . 2 ); 161 (2 . 8); 150(10.5); 149(100); 148(4.1);

147(4.8); 133(7.2); 121(6 . 1); 119(9*5); 118(10. 0 ); 106 (3 .3 );

105(11.1); 92(3.0); 91(33.0); 90(17.6); 89(15.7); 79(2.6);

7 9 (3 -5 ); 77( 12. 0 ); 65(4.6); 64(3.0); 63(9-3); 62 (3 .3 );

59 ( 10. 2 ); 51(5.0); 39(6.5); 15(9.3).

The thermolysis product (150 mg) were chromatographed on a silicagel column (length, 20 cm; diameter, 1 cm) using benzene as eluting solvent and taking 2 ml fractions, using a fraction collector. Fractions 1 to 40 were free of organic material.

The amount and composition of material recovered from fractions

4 l to 170 were:

fractions 4l - 50 (11 mg)

51 - 70 (33 mg)

7 1 - 9 0 (21 mg)

91 - 120 (20 mg)

121 - 170 (6 mg)

The nmr spectrum o f compounds 109 > 111 and 112 are sum marized in the following:

TMS dimethyl cycloheptatriene-l, 2 -dicarboxylate ( 109): : 8 6

2.52 (d, 2H, J=T.O Hz, C-7 p ro to n s), 3 .7 9 (s, 3H, CCbMe), 3.81

(s, 3H, CPgMe), 5*1*8 (d t, 1H, J« 8. 8, 7 .0 Hz, C-6 proton), 6.33

(dt, 1H, J=8. 8, 3-0 Hz, C-5 p ro to n ), 6.82 (brd, 2H, J*3*0 Hz, C-3

and C-4 protons). Subsequent fractions provided increasing

amount of dimethyl cycloheptatriene-l, 7-dicarboxylate fill)

6 CDC13 : 3 *61 ( s ’ 3H’ CQ2Me5> 5*81 (s, 3H, COOMe), i*.l 8 (d, 1H, J=8.0 Hz, C-7 p ro to n ), 5.91 (brdd, 1H, J-8.0, 10 Hz, C -6

proton), 6.42 (brdd, 1H, 7=10, 6 Hz, C-5 p ro to n ), 6.64 (brdd,

1H, J=10, 6 .0 Hz, C-3 p ro to n ), 6.84 (dd, 1H, J=10, 6 Hz, C-4

proton), 7.38 (brd, 1H, J=6,0 Hz, C-2 proton)] and dimethyl TMS cycloheptatriene- 2 ,3 _dicarboxylate ( 112) 6^ ^ : 2 .45 (dd,

2H, J=6.5» 7-5 Hz, C-7 p ro to n s), 3*74 (s, 3H, C 02Me), 3*82 (s,

3H, COfeMe), 5.86 (d t, 1H, J-9.5, 7 Hz, C-6 p ro to n ), 6.35 (dd,

1H, j=9.5, 5.5 HZ, c-5 p ro to n ), 6.52 ( t , 1H, J^r.5 Hz, C -l

p ro to n ), 7.65 (d, 1H, J”5.5 Hz, C-4 p ro to n )].

III-H . Thermolysis o f a m ixture o f compounds 103 . 109 . I l l

and 112 .

Heating of a mixture of 103 (11$), 109 (19$)» HI

(29$) and 112 (1*4$) in a non-sealed glass tube at 220-225 ° for

1 hour gave the following mixture of compounds:

103 - 10$

111 - 6$ 7$

113 - 45$ 100$ 50 $

114 - 39$ 43$

The nmr analysis was performed by electronic integration of the

carboraethoxy region using a 60 MHz Perkin-Elmer R-12 instrument. 87

The d ata are le s s r e lia b le than th e one performed on the HR-100

instrument.

III-I. Thermolysis of a mixture of compounds 103 . 109 . Ill

and 112 .

A mixture of 10^ (28$), 10£ (33#), 111 (2996) and

112 (10$) was heated in a sealed glass tube at 220° for 20 and

60 minutes, respectively. NMR analyses (100 MHz, Varian HR-100)

by electronic integration of the carbomethoxy region gave the

following composition of compounds:

I n i t i a l Compound Composition 20 min. at 220° 60 min. at 220° ______W ______(*) ______f f l

105 28 30 30

109 53 6 7 111 29 31 30

112 10 33 33

I I I - J . Therm olysis of compound 102 w ith cyclododecene.

A mixture of 30 nig of compounds 102 and 85 mg of cyclododecene were heated in a nmr tube for 10 minutes at 220°.

Electronic integration of the 100 MHz nmr spectrum of the re sulting mixture showed the same ratio of compounds 103 » 109 »

111 and 112 as was observed for the thermolysis of 102 alone.

III-K . Therm olysis o f compound 102 w ith 11^ .

A mixture of 102. (33 mg) and 11^ (lj-0 mg) was th e r- molysed at 220° for 10 minutes. Electronic integration of the carbomethoxy region in the nmr spectrum (100 MHz) gave essentially 88 the same com position o f compounds 1Q3 , 109 , 111 and 112 as found for the thermolysis of 102 alone.

III-L. Thermolysis of dimethyl-2,5-norbornadiene-2.5-dicarboxylate (ll4).

About 100 mg of 114 were heated to 230° for 10 minutes. Nmr analysis indicated no change during the pyrolysis of

(114).

III-M . C a ta ly tic hydrogenation of 102 .

A suspension of 25 mg. of 5$ Pd/C catalyst in 5 ml of reagent grade methanol was placed into a reduction flask, which was flushed first with N 2 then filled with H 2 . Under stirring,

109 mg. (0.4 mmol) of 102 in 25 ml of methanol were added by injection into the reduction flask with a syringe. The mixture was stirred for 3 hours, filtered and the solution concentrated to provide a yellow syrup. Crystallization from ethyl acetate gave pale yellow crystals, which were identical with 102 by nmr.

The same operation was repeated with Pt20 as the catalyst.

Again, only the starting material was recovered.

III-N. Dimethyl 2.2-dioxido-2-thiabicyclo-f2.2.21 -octa-7-ene-5.6-

dicarboxylate (122).

A mixture of 2.0 g. (15 mmol) of (*-thiopyran-l, 1- dioxide (9T) and 2.88 g. (24 mmol) of dimethyl maleate in a pyrex tube was flushed with nitrogen, evacuated and then heated to

110-120°C for 75 hours. The progress of the reaction was monitored by tic. Excess maleate was extracted with chloroform and the

CHC13- insoluble yellow solid was recrystallized from ethyl acetate to give 1.55 g. of 122 . Evaporation of the mother liquid under high vacuum and recrystallization from ethyl acetate gave another 0.5 g. o f -122 , a to ta l o f 2.05 g. (49$), m.p. I 5 I-I 53 0; i r , v ^ 1725 (-COOCH 3 ) ; 1280 and 1120 cm-1 (SQa-stretching).

Nmr’ ^DtE0-d: ^*17 (ddd> dis> 2.0; Ji6, 2.0; Jiy, 6.0 Hz, H-l);

2.86 (dd, J34, 2.5,* J3 a,3 b» 15.0 Hz, H-3; 5.22 (dd, J3,4, 2.5, d3a,3b» 13.° Hz, H-5; 3.55* (H-4 and H- 5 ); 5-92 (dd, Js,i, 1.8;

Js,s, 10.5 Hz, H-6); 6.4* (h -T and H-8); 3-57 (C5-C00CH3); 3.56

(C6-C00CH3). *Complex m ultiplet with the center of AB part given.

The mass spectrum 70 eV, m/e (in te n s ity ) 274(0.7), 2*1-3(9-2), 210

( 18. 0 ), 151 (20. 0 ), 150 (51 . 7 ), 118(8.3), 105(4.67), 91(100), 6 *1-(3 .7 ),

59 (30. 0 ), 44(3.6), 31(1.3), 15(7.3).

Anal. Calcd. for CnHi406S: C, 48.17; H, 5.11; 0, 35.04;

S, 11.68. Found: C, 48.29; H, 5.15; S, 11.80.

III-0. Thermolysis of Dimethyl-2.2-dioxido-2-thiabicyclo-f2.2,21 -

octa-7-ene-5,6-dicarboxylate (122).

A pyrex tube sealed with septum containing 300 mg.

( 1.09 mmol) o f 122 was flushed w ith n itro g e n then evacuated and placed into an oil bath preheated to 225°C for 5 minutes. The nmr spectrum of the thermolysis mixture showed only signals for the s ta r tin g m a te ria l 122 .

Repetition of the procedure at 250°C and 275°C gave similar results as described above. The pyrex tube was then placed into a sand bath, preheated to 300°C, for 3 minutes. A black- residue and a green yellow oil formed on the wall of the tube.

From the gas phase, a 3 ml sample was subjected to GC-MS, (column; 90

3', porapak-S, 70°C/200°C, 6°C/min.). MS(70ev.). The GC-trace indicated several components which on the basis of their mass spectra were identified86 as: methyl acetate, carbon dioxide, ethane: methyl ether, sulfur dioxide, butane and dimethyl e th e r.

Fraction Parent Mass Base Peak Intense Signals Structure

74 43 74(74.3) q 15(44.8) » 59(21,. 8) 3 44(14.3) 3

44 44 28(9.5) CO, 15 (6 . 0 )

3 30 28 27(28. 6 ) 30(24.5) 4 k6 45 46(52.6) L-0-CH, 29(25.6) CHs 15(14.7) 64 64 48(4.8) 65(5.5) SCL 66(3.5)

6 58 43 56(52) 57(25) 41(19) 4 10 40(8)

62 62 47(44.6) c jj 45(22.0) 6 1 (22. 0 ) $

60 60 32(21,-3) 6 2 (1,.0 ) SCO 91

The green-yellow oil shewed ir bands at v *iquid 3000. m a x ^ ’ 1850, 1800, 1750, i k 30, 1270, 1220, 1125, 1080, 900, 635 CM-1.

However, not enough material was available to run an nmr spectrum.

The ir bands for the black residue were: v 2950, 1750 ,

1^50, 1375) 75C> 720 CM”1, The combined solid and liquid residue were purified by preparative layer chromatography with benzene as the developing solvent giving four major regions of material.

Extraction of the four bands with methanol, filtration of the silica gel and evaporation of the solvent gave the following results: The first two bands contained only very small amounts of organic material. The top band of lowest polarity gave 15 mg, the third band provided 25 mg, and the fourth band 30 nig of syrup. The individual bands, dissolved in CHC13, were sub jected to GC. It was shown that they represented mixtures of several compounds. The top band contained two components, and the other bands 5 components each. The nmr spectra of the

f f T L lO organic residues of each band follow: Top band: : CDCI3 1.25 ( s ) , 1.70 ( s ) , 2.15 ( s ), 2 .3 2 ( t ) , 3.16 (d), 3.38 (s ),

3.90 (t), 6.22 (s), 7.26 (s). Third band: 6 ™ ^ : 1.26 (2H, t ) , 1.56 (1H, b . s . ), 2.15 (1H, s)„ 2.38 (2H, s ) , 2.62 (1H, s ) ,

3 .7 (2H, b s), 3 .9 (2H, s ) , 7.25 (2H, s ) , 7-38 (2H, s ) , J.b (1H, dd), 7 .8 5 (1H, s ) . The fo u rth band: : 1.25 (1H, b s),

2 .0 (1H, s), 2.3 (2H, s), 2.68 (1H, s), 3.65 (*•«, m), 3-86 (3 H, t ) ,

6 .0 (1H, bm), 7 .4 (2H, s ), 7.65 (2H, dd). 92

III-P. Reaction of /v-thiopyran-1,1 -dioxide (97) with vinyl

trimethylamnonium bromide.

In a pyrex tube, 130 mg. ( l mnol) o f compound 97 were mixed with 166 mg. (l nmol) of vinyl trimethylanraonium bromide, flushed with N 2 , evacuated and heated to 110-120°C for

17 hours. The reaction mixture was worked up by dissolving un reacted ammonium bromide in H^O and filtration of the H 2O- insoluble solid. The nmr spectrum of the organic residue in dicated that no cycloaddition products were formed. Repetition o f the same re a c tio n by h eatin g compound %]_ and the vinyl trimethyl ammonium bromide in p-dioxane under reflux for

3 hours, gave similar results.

III-Q. Preparation of 2.2-dioxido-2-thiabicyclo T2.2.21octa-7-

en e- 3 .6 -dicarboxylic anhydride ( 132).

In a 10 ml round-bottomed fla s k , I .30 g. (10 mmol) of the a-thiopyran- 1, 1-dioxide ( 97) and I .30 g. (13 mmol) of maleic anhydride were mixed, the flask was flushed with nitrogen and then evacuated. The flask was heated to 110°C for 2 hours.

First, the starting materials melted, then a pale yellow solid was formed. After cooling, excess maleic anhydride was re moved by e x tra c tio n w ith chloroform and th e re s id u a l so lid was recrystallized from ethyl acetate to give 2.049 g. (89-9$) of compound 132 , m.p. 277-278°C; i r , v max 3500, i860, l8?0, 1780, 11*0 0 , 1300, 1230, 1220, 1120, 1090, 1000 , 940, 78O, 750 ,

685 . CM"1; nmr, : H-l: 4.07 (br dd Ji,Si 2.0; Ji,y,

5.2 Hz, H-l): 2.82^(b dd J 3a ,4 » 4 .0 ; J3a ,3 b> 13*0 Hz, H-3a ) : 93

5.17 (br dd J3b,4, 2 .0 ; J3a>3 b, 13.0 Hz, H-3b ): 3-42* (h-4 and

H-3): 3.76 (dd J6>1, 2.0; J5,6, 10.5 Hz, H-o): 6.4* (H-7 and

H-8).

# Multiplet appearing as a br. doublet (J:2.0; 5 . 2 )

* Complex multiplet with the center of the AB part given. Mass apectrum (70 ev) showed parent mass at m/e (intensity) 168 (0.8) and base peak at 55 , in ten se peaks a t 18 (6 .9 ), 45 (3*7), 52 (5 -6 ),

64 (4.0), 66 ( 9.9), 91 ( 2 .6 ), 92 (80), 9b (4 .6 ), 166 (3 . 3 ), 168

(0 .8 ).

Anal, calcd, for C 9H805 S: C, k j.31; H, 3-51; S, 14,04; 0, 35.09;

Found: C, 1*7.25; H, 3.52; S, 14.13.

III-R . H ydrolysis o f compound 152 .

In a round-bottomed flask, 350 ml of 10$ hydrochloric acid were added to 1.530 g. ( 6 .7 nmol) of anhydride 152 . The suspension was stirred at 90°C for 24 hours, slowly forming a c le a r so lu tio n . Removal o f most o f the w ater in the high vacuum gave the diacid l^ which was used for the subsequent electrolysis- decarboxylation reactions without further purification. Complete removal of water at elevated temperature after hydrolysis resulted in the reformation of the anhydride 152 .

I I I - S . E le c tro ly s is decarboxylation of compound 155 .

In an electrolysis flask, 1.3 g. of the diacid 155 in 5 ml of water, 5 ml of triethylamine and 48 ml of pyridine were mixed96’97’98. The flask was equipped with two platinum elec trodes, a nitrogen inlet, a reflux condenser, and a magnetic s t i r r e r . The fla sk was flushed w ith N 2 for 5 minutes and then the electrolysis was started at 13h volts. During the electrolysis,

the system was continuously stirred in a nitrogen atmosphere

under reflux. The current dropped from 0.14 ampere to 0.01

ampere at 13^- volts. At the end of 10 hours, the brown solution

turned black. Evaporation of the solvent gave a black residue,

the nmr spectrum of which did not indicate the formation of

the desired decarboxylation product 100 . An experiment using

the tetraacetate decarboxylation method113 was also carried out

without success.

III-T. Reaction of n>-thiopyran-l. 1-dioxide (97) with methane-

sulfonyl chloride/ triethylamine.

In a round-bottomed flask, fitted with a magnetic

stirrer, addition funnel and reflux condenser, 260 mg (2 mmol)

of crthiopyran-1,1-dioxide (97) in 10 ml of freshly distilled

p-dioxane, and 202 rig. (2 nmol) of triethylamine were mixed.

Under stirring 230 mg. (2 nmol) of methanesulfonyl chloride was

slowly added and then allowed to stir at room temperature for

another 16 hours. The reaction mixture was then poured into a

beaker with ice-water. The aqueous phase was stirred for a

few minutes. Evaporation of the solvent provided a black residue which was stirred with 30 ml of water and then extracted twice with 25 ml of ether and 25 ml of chloroform. The combined ether

and chloroform extracts were evaporated giving a brown residue,

the nmr spectrum of which was identical with the starting material.

97 . In acetonitrite at -40°C the reaction gave analogous results. 95

III-U. Reaction of 1-methoxv-1.5-butadiene with methanesulfonvl

chloride/triethvlamine.

To a solution of 0.42 g. (0*5 nmol) of 1-methoxy-

1,3-butadiene and 0.606 g. (0.6 tnmol) o f trieth y lam in e in 50 ml

of anhydrous ether, was added dropwise under stirring, a solu

tion of 0.684 g. (0.6 nmol) of methanesulfonyl chloride in 30 ml

of anhydrous ether. The resulting mixture was then stirred for

12 hours. The brown precipitate was filtered off and the sol

vent removed from the filtrate, providing a dark brown syrup,

the nmr spectrum of which gave no indication of cycloaddition

p ro d u cts.

IV. Reaction of D-Camphor-10-sulfonyl Chloride and Camphane-10-sulfony1

chloride with triethylamine

IV-A. Preparation of D-camphor-10-sulfonyl chloride (157).

In a round-bottomed flask, fitted with a conderser and a drying tube, 25 g. (0.108 mol) of D-camphor-10-sulfonic acid was mixed with 20 ml of thionyl chloride. The mixture was heated until homogenous and kept refluxing for 2 hours.

The re a c tio n m ixture was cooled to room tem perature and poured into a beaker with ice to hydrolyze the excess thionyl chloride. The pale yellow crude product was filtered by suction and recrystallized from cyclohexane to get 21 g. (77$) of pale yellow crystals, m.p. 51~57°C. Lit. 5°"57°C114. The nmr spectrum confirmed the structure of the D-camphor-10-sulfonyl chloride (137). 5 ^ ^ : 4.05 dd (j-l4 Hz, 2H)*; 2.5 ~ 1.6 (e^ lop) 96

1.16 (a, 3H); 0.9 (s, 3H).

* dd with center given.

IV-B. Reaction of D-camphor-10-sulfonyl chloride with triethyl

amine in ether.

In a 3-necked round bottomed flask, fitted with drying tube, addition funnel and a magnetic stirrer, 5.02 g.

(20 mmol) of the D-camphor-10-sulfonyl chloride were dis solved in 175 ml of anhydrous ether under nitrogen. Triethyl amine (b.2k g ., k-2 mmol) in 30 ml of anhydrous ether was added dropwise. After completion of the addition of triethylamine, the nitrogen was turned off and the mixture was stirred for another

16 hours. During the addition of triethylamine, a white solid was formed.

At the end of the 16 hrs., the solid was filtered off to get k.019 g* of the solid. The nmr and ir spectrum of it showed that there were triethylanmonium chloride and triethylanmonium camphor-10-sulfonate. The pale yellow filtrate, upon evapora tion of all the solvent, k.O g. of brown thick syrup was ob tained. The thin layer chromatography (tic) of the syrup, developing in benzene, shewed two distinct spots, first with

R^ o f 0.1 j6 and second of R^ 0.37 and a spot at original point which was ammonium salt in 10 cm silica gel plate. The syrup was purified by column chromatography in silica gel with benzene/ethyl acetate (12/1). Different fractions were collected. From 3.8 g. of the syrup, 0.70^ g. of the pure first spot was obtained, which was E-D-camphor-10-chloro- 97 sulfoxide ( 138b) there was O .557 g. of pure second spot collected, which was (Z)-D-camphor-10-chlorosulfoxide (138a) and 0.30 g. of mixed first and second spots. The total y ie ld is 35 * 1$*

Work up the pure fra c tio n s and the f i r s t spot was a waxy oil at room temperature, but crystallized well in deep freezer and the second spot was a fine crystals at room temperature, in.p. 8l=82°C. The E-D-camphor-10-chlorosulfoxide ( 138b) showed i . r . v “ I 1750 ( J L ) ; 1290 and I I 30 ( /I n ) ; 1080, 1050 0 and 1020 (/Sn), and 830 cm"1 (c-cl). The nmr spectrum : CDCI3 1.68 (s, 3H), 1.71 (s, 3H), 3.0 1.5 (envelop, 7H). The mass spectrum (70 ev) gave a parent and base peak at 232 (100), intense signals at 197 (38. k ), 183 (12. 6 ), 137 (71-9), 107

(75-3) and 109 (7 5 .3 ). Anal. Calcd. for CxoHiaGlOfeS: C, 51.61; H, 5.59; Cl, 15.27;

S, 13.76 ; 0, 13.76 Found: C, ; H, 5-71 ; Cl, 15.09 , s , 13.87 .

The (z)-D-camphor-lO-chlorosulfoxide (138a) showed the

v 1750 (X ), 1280 ( C=S); 1080, 1050 and 1020 IDoa ,PMC ( S=0) and 780 cm" (C-Cl). The nmr spectrum gave 6 _ „ , : CDCI3 1.61 (s, 3H), 1.63 (s, 3H), 2.7 r- 1.5 (envelop, 7H). The mass spectrum gave a parent and base peak at 292 ( 100) and intense peaks at 215 (3 . 0 ), 197 ( 6 . 0 ), 183 (3«l), 173 (19* 8),

137 (9 .1 ), 107 (7 A ) and 105 (7.*0.

Anal. Calcd. for CloHi 3C102S: C, 51.61; H, 5.59; Cl, 15.27;

S, 13.76 . Found: C, 51.90; H, 5*72; Cl, 15 . 08; S, 13. 90. IV-C. Reaction of D-camphor-10-sulfonvl chloride with triethvl-

amine in anhydrous cyclohexane.

A solution of 5.02 g, (0.02 mol) of D-camphor-10- sulfonyl chloride ( 137) in 175 ml of cyclohexane, predried over

Na, was s tir r e d under N3 p assin g over i t . k.2l+ g. (O.CA-2 mol) of triethylamine in 30 ml of cyclohexane was added slowly and the resulting mixture was stirred for 16 hrs. During that time, a white solid was formed which was filtered at the end of

16 hrs. and gave k.2.0 g. of white solid. The nmr and ir spectrum of the solid shewed a mixture of triethyl ammonium chloride and

D-camphor-10-sulfonate salt.

The filtrate, after evaporation gave 1,95 g. of deep yellow syrup. The nmr spectrum of the syrup showed no signals lower than 3*3 ppm, and the tic indicated only two distinct spots,

I .85 g. of the syrup was purified over a column chromatography on silica gel and developed with 1 to 12 ratio of ethyl acetate and benzene. The different fractions were collected by fraction collector and checked by tic. Among the fractions, it gave O. 65 O of the first spot and Q. 5 IQ g. of the second spot and 0.30 g. of mixture of the first and second spots. The ir and nmr of these two pure fractions showed exactly the same signals as those of the (e)- and (z)- of D-camphor-10-chlorosulfoxide

( 158b) and ( 158a) as that obtained in the previous reaction.

IV-D. Preparation of camphane-10-sulfonic acid ( 156 ).

A. Reduction of D-camphor-10-sulfonic acid.

k2 g. (0.18 mol) of the D-camphor-10-sulfonic 99 acid were dissolved in 270 ml of hot glacial acetic acid in a

3-necked round bottomed flask, fitted with a mechanical stirrer, additional funnel, and a reflux condensor. To this solution,

♦zinc-amalgam was added followed by 270 ml o f concentrated hydrochloric acid. The mixture was stirred for 7 hrs. and another 70 nil of the concentrated hydrochloric acid was added and another 60 ml of concentrated HC1 was added after another

7 hrs. The mixture was then stirred for 2h h rs . The m etals were filtered off through glass wool, and most of the solvent was removed to give 22 g. ( 56 -5$) of the reduction product which was converted to sulfonyl chloride without further purification.

♦Preparation of zinc-amalgam115.

To a hot solution, containing 15*5 g* of mercuric chloride and 8 ml of concentrated hydrochloric acid and 120 ml of w ater, 137 g. of zinc dust was added slowly with vigorous stirring. The zinc-amalgam was separated by decantation of the water and was used immediately.

B. Preparation of camphane-10-sulfonyl chloride (157).

To 21.8 g. (0.1 mol) of the reduced acid in a round bottomed flask, 17 g. of the phosphorus oxychloride were added and refluxed for 3 hrs. then cooled to room temperature.

The reaction mixture was poured into a beaker containing ice- water to hydrolyze any excess phosphorus oxychloride. The black solid was filtered and was dissolved in chloroform and refluxed with charcoal for decolorization. The charcoal was filtered off and chloroform was evaporated. The deep 100

brown solid was purified by column chromatography. Recrystalliza tion from petroleum ether gave 6.0 g. ( 25 . 37$) of pale yellow

c ry s ta ls , m.p. 80-82°C. The i . r . fo r compound 157 —— max 1250 and ll60 ( SQ2 ), and 780 cm "1 fo r S-Cl ab sorptio n. The TMS nmr spectrum gave &CDCl3: 0*8? (s , 6 h ); 1.8 ~ 1.1 (envelop, 9H);

3.75 (s , 2H). The mass spectrum (JO ev) gave a parent peak at

172 (0 . 7 ), base peak 137 ( 100) and intense signals at 27 (20),

29 (1*0, 39 ( 26 ), kl (50 ), k3 (21. 5 ), 53 (1 8 .5 ), 55 ( 1^ . 0 ), 57

( 10. 0 ), 6k (3 . 6 ), 67 (29. 5 ), 69 (22. 5 ), 77 ( 16 .2 ), 79 ( 21. 0 ),

81 (88. 0 ), 82 ( 12. 5 ), 91 ( 13. 0 ), 93( 22. 5 ), 95 ( 75 . 0 ), 105 (^ A ),

109 (5 .0 ), 119 ( 6 . 0 ), 138 ( 8. 5 ).

Anal. Calcd. for CloHiTC10sS: C, 50.71*; H, 7.19; S, 13.53; Cl, 15.01;

0, 13. 53 . Found: C, 50.77; H, 7.13; S, 13.60; Cl, 15 . 20.

IV-E. Preparation of camphane-10-phenyIsulfonamide (167).

r A solution of O .606 g. (6 mmol) o f trieth y lam in e and O.558 g. (6 nmol) of aniline in 60 ml of anhydrous ether was stirred under N2 . To this solution, 1.18 g. (5 nmol) o f camphane-10-sulfonyl chloride in 20 ml of anhydrous ether was added slcwly, and the resulting mixture was stirred for 17 hrs.

A white solid was formed during that time and was filtered off to get 0.32 g. of triethylanmonium chloride. Evaporation of all the solvent from the filtrate to get 0.95 6* (61$) of yellow crystals, which after recrystallization from cyclohexene gave 0.60 g. (i;0$) of sugar-like, colorless, transparent crystals, m.p. 121-123°C. The i.r. v max 3250 (N-H), 3080 (c -h ), 1600 , 720 ( ), 790, 770 (A /-® ), 1370, 1165 cm*1, 101

R-SCt-NH-. The nmr gave 6 ^ ^ : 0.78 (s, 6 h ), 1.20-1.92 (envelop,

9H), 3.14 (s , 2h), 7.05-7.15 (bm, 1H), 7 .2 -7 .4 (m, ?H). The mass spectrum gave a parent mass at 295 (6.0), and base peak at 55 (100), and intense signals at 27 (10), 59 (1 6 .7 ), 55 (1 8 .0 ), 56 (1 4 .7 ),

64 (9 .7 ), 65 (13-7), 67 (17.3), 69 (11.3), 77 (9.7), 79 (13.7),

81 (45), 84 ( 36 . 7 ), 91 ( 8. 7 ), 92 ( 10. 7), 94 (13.7), 95 ( 25 . 0 ),

107 (6 . 9), 121 (7 .3 ), 136 ( 18.3 ), 137 (2 3 .3 ), 157 (1 1 .3 ), 292 (30. 0 ). The crystals were dried over refluxing ethanol under vacuum.

Anal. Calcd. fo r CisHasNOaS: C, 65-53,* H, 7 . 85 ; N, 4.77. Found:

C, 65 .51; H, 8.06; N, 4 .61 .

IV-F. Reaction of camphane-10-sulfonvl chloride (157) with

triethylamine in ether.

A solution of 2.36 g. (0.01 mol) of camphane-10- sulfonyl chloride ( 157 ) in 80 ml of anhydrous ether was transfered to a three-necked, round bottomed flask, fitted with a drying tube, additional funnel, and N2 inlet. This solution was stirring with N2 passing over it, and 2.121 g. (0.021 mol) of tri ethylamine in 30 ml of anhydrous ether was added dropwise and the resulting solution was stirred for 16 hrs. After the addition of triethylamine was complete, the N2 was turned off.

Work-up of the reaction mixture by filtration gave

1.21 g. of solid which was formed during the reaction. The nmr spectrum of the solid was exactly the same as that of the TMS synthetic triethylanmonium chloride 5 : 1.42 (t, J=6.0 Hz, CDCI3 9H), 3.22 (q, 6H, J=6.0 Hz ). The filtrate was evaporated to give a brown syrup (2.4 g .). The nmr spectrum 102

of the syrup showed trace amounts of aldehyde in a complicated spectrum.

The crude reaction product was purified by column chromatography

using benzene and petroleum ether eluant and then with ethyl acetate mixed with benzene. Upon the addition of the

syrup onto the silica gel, the syrup turned blue

green. Obviously, decomposition occured on the silica gel. Different fractions were collected. 2-15 (0.04-5 g.),

16-20 (0.140 g.), 21-to (0.137 S.), tl -50 (0.0 g.), 51-65 (0.337 g .) , last (1.420 g.). The i.r. and nmr spectrum of the first 15

fractions indicated the recovery of the starting materials.

Fractions 16-20 showed mostly the aldehyde which was further

purified by sublimation at 110-120°C under high vacuum to get KBr 0.1 g. of pure aldehyde. The i.r. v : 1750 cm-1 (C*0), tQflX US nmr spectrum gave a typical aldehyde signal at 9.75 ppm. . CDCI3 9.75 (3 , 1H), 2 . 38- 1.30 (m, 9H), 1.28 (s, 6h). The mass spectrum

gave a parent mass at 152 (75*0)* base peak at 151 (100), and

intense peaks at 137 (15-5), 135 (18.2), 123 (72.7), 109 ( 63 . 6 ),

95 (24.5), 99 (19.1), 83 (33.6), 81 (54.5). Fractions 21-to were the green materials which crystallized after evaporation of

all the solvents. The nmr spectrum gave 6^^^: 4.2 (q, J^.O Hz,

2H), 3.0 (s, 2H), 1.32 (t, j»7.0 Hz, 3H), 1.8-0.9 (envelop, 9H),

0.8 (s, bH). The mass spectrum gave a parent mass at 180 (4.7)

and base peak at 42 (100), intense peaks at 27 ( 58 . 9), to ( 10. 5 ),

53 (24.2), 64 (15.5), 67 (31. 6 ), 69 ( 17. 4 ), 79 ( 28. 9), 81 (50 . 0 ),

91 ( 16 . 8), S3 (47.4), 95 (26.3), 99 (17.4), 107 ( 10. 5 ), 135 (18. 9),

136 (25 . 8), 137 (1 6 .8 ), 151 (6 .3 ), 162 (3 .2 ). F ractio n s 51-65

showed a very complicated spectrum. Fractions after 65 did not 103

come out of the column, but were washed out by methanol. The nmr spectrum of this methanol wash was similar to the nmr spectrum of the synthetic triethylanmonium camphane- 10-sulfonate. 5 " “, : CDCI3 3 .2 (q, J-7 .0 Hz, 6 H), 2.81 (s, 2H), I .38 (t, J^T.O Hz, 9H),

1 .8 ^ -1 .(envelop, 9H), 0.82 (s, 6 h ).

IV-G. Reaction of camphane-10-sulfonyl chloride (157) with

triethylamine followed by aniline addition.

A solution of 2.36 g. (0.01 mol) of caraphane-10- sulfonyl chloride ( 157 ) in 100 ml o f anhydrous e th e r was tra n s - fered into a three-necked, round bottomed flask, fitted with a drying tube, additional funnel, and N 2 inlet. This solution was stirred with N2 passing over it, and 2.020 g. (0.02 mol) of triethylamine in 20 ml of ether was added slowly. After the addition of triethylamine was complete, the N 2 was turned off.

The resulting solution was stirred for 20 hrs. and the nmr spectrum indicated that there was still a small amount of starting material 1^7 left, at the end of lj-0 hrs., it indicated that all the starting materials were gone, then 1.0 g. (0.011 mol) of aniline was added with Ng passing over the reaction mixture.

The solution was stirred for another 20 hrs. There was solid formed before the addition of aniline. Work-up by filtration off the solid to get 1.27 g. (97.0$) of solid whose nmr spectrum was the same as that of synthetic triethylanmonium chloride. The filtrate was evaporated off and the syrup (3-0 g.) was dissolved in chloroform. The chloroform solution was washed twice (50 ml each) with 5$ HC1 then with water to get 104 rid of any excess triethylamine and aniline. The chloroform solution was then dried over anhydrous magnesium sulfate. After evaporation of all the chloroform, a brown syr..p (l.O g, ) remained.

The i.r. spectrum of the camphane-10-phenylimine (168) gave

v ^ Uld: 2990 (C-H), 1720, 1600 (ON), 1175, H00, 780 ( ^ 2 > ) . TMS The nmr spectrum gave : 7.27 (s, 5H), 2.0-1.4 (envelop,

9H), 1.0 (s, 6 h), 5*5 (s, 1H) and impurities at 3*02 (1H), O. 85 -

0.95 (m» 2H). The mass spectrum gave a parent mass at 227 (3*2), base peak at 109 ( 100), intense peaks at 170 (5 . 1 ), 158 (4 .6 ),

152 (30. 2 ), 137 ( 18. 1), 136 (30 . 2 ), 135 (8. 8), 123 (21.9), 121

( 15 . 8), 110 (15.3), 99 ( 21,9)» 95 (39.5), 93 (46. 5 ), 91 ( 18.6 ),

83 (4 8 .5 ), 31 ( 60 . 5 ), 79 (34.9), 77 (25.l). The syrup was purified by column chromatography on silica gel/benzene and different fractions were collected.

Fractions Weight (mg)

1-7 0 .0

8-16 0.168

17-18 0.078

19-20 0.220

21-22 0.088

23-26 0.045

27-40 0.037

46-75 0.038 The nmr spectrum of fraction 8-16 showed the same spectrum as camphane- 10-aldehyde ( 160 ). 17 and 18 showed small amount of aldehyde and mostly a complicated spectrum. Fractions 19 and

20 showed a complicated nmr spectrum, a phenyl signal between 7.l8-7.33» a quartet at 3*8* a singlet at 3*1 ppm, several singlets at 2.1-3.0 ppm, camphane envelop at 1,83-1.5 PP® , complex signals a t 0 . 9- 1.3 ppm. and two methyl group from camphane at 0 .8 ppm.

Obviously, the Schiff's base decomposed on the silica gel. The latter fractions also showed complex nmr spectrum, totally different from that of the crude reaction syrup before chromatograph. REFERENCES

1. T. J. Wallace, Quart. Rev. (Chem. Soc. London). 20, 67 (1966).

2. G. O pitz, Angew. Chem. I n t e r . E d it. 6 , 107 ( 1967).

3. W. E. Truce and L. K. Liu, Chem. Ind. 4, 145 ( 1969).

4. J. F. King, Accounts of Chemical Research. 8, 10 (1975).

5. E. Wedekind and D. Schenk, Ber. dtsch. Chem. Ges. 44, 198 (1911).

6. E. Wedekind and W. Weisswange, B er. d tsc h . Chem. Ges. 59.

1631 ( 1906 ).

7. H. Staudinger and F. Pfenninger, Ber. dtsch. Chem. Ges. 49.

1941 (1916).

8. H. Kloosterziel, M. Deinema, and H. Backer, Reel. Trav. Chim.

Pavs-Bas. J lt 1228 (1952).

9. G. Hesse, E. Reichold and S. Majmudar, Chem. Ber. . 90. 2106

(1957).

10. L. Von Vargha and E. Kovacs, Ber. dtsch. Chem. Ges. 75. 794

(1942).

11. T. Zincke and R. Brune, Ber. dtsch. Chem. Ges. 4l, 902 (1907).

12. G. Schroeter, Justus Liebigs Ann. Chem. 4l8. 161 (1919)-

13. A. Locher and H. E. Fierz, Helv. Chim. Acta. 10. 642 ( 1927),

14. C. M. Suiter, The Organic Chemistry of Sulfur; Wiley, New York.

N. Y. 1944 p. 659.

15. H. Kloosterziel and H. J. Backer, Reel. Trav. Chim. Pays-Bas,

XI, 1235 ( 1952 ).

16. G. Hesse and S. Majmudar, Chem. Ber. 95, 1129 (i 960 ).

17. H. Kloosterziel, J. Boerema and H. Backer, Reel. Trav. Chim.

Pays-Bas, 72, 612 (1953).

106 107

18. E. Hennno, P. de Mayo, A.B.M.A. S a tta r and A. S to e ssl, Chem.

Soc. London. 238 (1961).

19- J. F. King and T. Durst, Tetrahedron Letters. 56 . 585 (1963).

20. S. T. Weintraub and B. F. Plumner, J. Org. Chem. 36. 361 (1971).

21. J. F. King and T. Durst, J, Amer. Chem. Soc. 86. 287 (1964).

2 2 . J. F. King and T. Durst, J. Amer. Chem. Soc. 87. 5684 (19 65 ).

23. W. E. Truce and R. W. Campbell, J. Amer. Chem. Soc. 88. 3599 (1966 ).

24. G. Opitz, M. Kleemann and D. Bucher, G. tfalz and K. Rieth,

Angew. Chem. I n t e r . E d it. £, 59^ (I 966 ).

25 . G. Opitz and D. Bifchler, Tetrahedron Letters. 5263 ( 1966 ).

26. G. Opitz and H. R. Mohl, Angew. Chem. I n t e r . E d it. 8, 73 (1969).

27. R. Fusco, S. Rossi, S. Maiorana and G. Pagani, Gazz. Chim. Ital..

22, 77^ (1965). 28. J. S. Groosert and M. M. Bharadwaj, J. Chem. Soc. , Chem. Comm.

4, lkk (197*0.

29. R. Fusco, S. Rossi and S. Maiorana, Chim. e. Ind. (Milano). 4f>.,

564 (1963). 30. G. Stock and I. J, Borowitz, J. Amer. Chem. Soc. 84. 313 ( 1962 ).

31. G. Opitz and H. Adolph, Angew. Chem. Inter. Edit. ,1, 113 (1962).

32. W. E. Truce, J. J. Breiter, D. J. Abraham and J. R. Norell,

J. Amer. Chem. Soc. 84. 30J0 ( 1962 ),

33* Fusco, S. R ossi and M aiorana, Chim. e. In d . (M ilan). 44,

873 ( 1962 ).

3**. T. Nagai, H. Namikoshi and N. Tokura, Tetrahedron Letters. 40,

^329 ( 1968). 108

35* L. A. Paquette, J . Org. Chem. 2 9, 2851 (1964).

36 . L. A. Paquette, J . Org. Chem. 29, 2854 (1-964).

37* G. O pitz, H. Schetnpp and H. Adolph, Liebigs Ann. Chem. 684,

92 (1965).

38. R. H. Hasek, P. G. Gott, R. H. Meen and J. C. Martin, J. Org.

Chem. 28, 2496 ( 1963 ).

39. R. H. Hasek, R. H. Meen and J . C. M artin, J . Org. Chem. 3 0 ,

1495 (1965). 40. W. E. Truce and J . R. N o rell, J . Amer. Chem. Soc. 85 , 3231 ( 1963 ).

41. W. E. Truce, R. W. Campbell and J . R. N o rell, J . Amer. Chem.

Soc. 86 , 288 (1964).

42. G, Opitz and D. Buchler, Tetrahedron Letters. 5269 (1966).

43. A. M. Hamid and S. Trippet, J. Chem. Soc. (c), 1612 ( 1968).

44. W. E. Truce, R. H. Barry and P. S. Bailey, Tetrahedron Letters.

5651 ( 1968).

45 . M. H. Rosen, Tetrahedron Letters. 647 ( 1969).

46. T. Hlraoka and T. Kobayashi, Bull, of Chem. Soc. Japan. 48.

480 (1S75).

47. G. Opitz and E. Temple, Angew. Chem. Inter. Edit. 754 (1964).

48. G. Opitz and F. Schweinsberg, Angew. Chem. Inter. Edit. 4,

786 (1965).

49. S. Rossi and S, Maiorana, Tetrahedron Letters. 263 ( 1966 ).

5 0 . W. E. Truce and A. R. Naik, Can. J. Chem. 44. 297 (1966).

51. G. Opitz and K. Fischer, Angew. Chem. 77. 4l ( 1965 ).

52. N. H. Fischer, Synthesis. 393 (1970).

53. N. P. N e u reiter, J . Amer. Chem. Soc. 88, 558 ( 1966 ). 109

54. Y. Ito, M. Oklano and R. Oda, Tetrahedron. 2137 (1967).

55. J. F. King, E. G. Levars and L. J. Danks, Can. J. Chem. 5 0 .

866, (1572).

56. J. Ide and M. Okano, Tetrahedron Letters. 51. 3491 ( 1968).

57. A. C. Cope, D. E. M orrison and L. F ie l, J . Amer. Chem. Soc. 7 2 ,

59 ( 1950 ).

58. J. F. King and D, R. K. Harding, J. Chem. Soc. . Chem. Comm. 959

(1971). 59. R* Mulder, A. M. Van Leusen and J. Starting, Tetrahedron

Letters. 3057 (1967).

60. R. Langendries, P. DeSchryner, P. de Mayo, R. Marty and J.

Schutyser, J . Amer. Chem. Soc. 96 , 2964 (1974).

61. J. F. King and de Mayo, Can. J. Chem. 47, 4509 ( 1969).

62. J . F. King, R. A. M arty, P. de Mayo and D. L. Verdum, J . Amer.

Chem. Soc. 6304 (1571).

63 . J. P. Snyder, J. Org. Chem. 21, 3965 (1575).

64. R. J. Boyd, Can. 1. Chem. 51. 1151 (1973).

65 . K. N. Houk, R. W. Strozier and J. A. Hall, Tetrahedron Letters.

897 (1974). 66. K. N. Houk, Private Comm. L.S.U. Baton Rouge. La. (1976).

67. E. B. B arn ett, J . Chem. Soc. 97, 63 (1910).

68. N. H. Fischer, Dissertation. University of Tflbingen, Germany

(1965). 69. G. Opitz and K. Fischer, Z. ftfr Naturforschung. 186, 775 (1963).

70. J . F. King and T. S. Lee, J . Amer. Chem. Soc. 91. 6524 ( 1969).

71. J. F. King and T. S. Lee, J . Amer. Chem. Soc. 91, 23 ( 1969). 110

72. J. F. King and T. S. Lee, Can. J. Chem. 49. 3724 ( 1971).

73* J. F. King and M. J. Coppen, Can. J. Chem. 49. 3714 (1971).

74. J. F. King, E. A. Luinstra and D. R. K. Harding, J. Chem.

Soc. , Chem. Comm. 1313 (1972).

75. G. M. Atkins and E. M. Burgess, J. Amer. Chem. Soc. 90 . 471*4 (1968).

76. E. M. Burgess and W. M. William, J. Amer. Chem. Soc. 94. 4386

(1972).

77. M. Chaykovsky and E. J. Corey, J. Amer. Chem. Soc. 84, 866

(1962 ).

78. M. Chaykovsky and E. J. Corey, J. Org. Chem. 28, 254 ( 1963 ).

79. J. Strating, Recutl. 83, 94 (1964).

80. J. F. King and E. G. Lewars, Can. J. Chem. £1, 3044 (1973).

81. S. W. Benson, F. R. Cruickshank, D. M. Golden, G. R. Haugen,

H. E. O'Neal, A. S. Rodges, R. Shaw, and R. Walsh, Chem. Rev.

279 (1969). 82. H. Kvart and J. F. King, Chem. Rev. 68 , 4 l? ( 1968).

83. E. Molenaar and J . S tra tin g , Rec. Trav. Chim. 86 , 1047 (1967).

84. J. F. King and E. G. Lewars, J. Chem. Soc. . Chem. Comn. 700

(1972).

85 . N. H. Fischer and H. N. Lin, J. Org. Chem. 38, 3073 (1973).

86 . E. Stenhagen, S. Abrahamson, F. W. M cLafferty, A tlas o f Mass

Spectral Data. Interscience Publisher. ( 1969).

87. H. Prinzbach and R. Schwesinger, Angew. Chem. Inter. Edit. 11.

940 (1972).

88. a. R. Darm, T. T h re lfa ll, M. Pesaro and A. Eschenmoser, Helv.

Chim. A cta. 46, 2895 ( 1963 ). Ill

b. V. A. Mironov, E. V. Sobolev and A. N. Elizarova,

Tetrahedron 1£, 1939 ( 1963 ).

89. a. E. M. LaComb and B. Stew art, J . Amer. Chem. Soc. 8£, 3**57

(1961).

b. J. B. Hendrickson and R. Bergeron, Tetrahedron Letters.

3609 (1973).

90. a. W, L. Mock, J . Amer. Chem. Soc. 97. 3666 (1575).

b. W. L. Mock, J . Amer. Chem. Soc. 97. 3673 (1975).

91. H. Hiraoka, J. Chem. Soc. . Chem. Coron. 101k, (197k).

92. A. P. Ter Borg, E. Razenberg and H. Kloosterziel, Reel. Trav.

Chim. 8£, 77^ ( 1966 ).

93* J. F- King, R. M. Enanoza and E. G. Lewars, Can. J. Chem. 52,

21*09 (1974).

94. C. A. Grob, M. Ohta and A. W eiss, Angew. Chem. 70. 3^-3 (1958).

95. N. B. Chapmann, S. Sotheeswaran and K. J. Toyne, J. Chem. Soc.,

Chem. Comn. 2lk ( 1965 ).

96. E. J. Corey and J. Casanova, J. Amer. Chem. Soc. 85 . 2lk ( 1965 ).

97. J. D. Bacha and J. K. Kochi, Tetrahedron. 2215 (1968).

98. S. D. Elakovich, Dissertation. L.S.U. . Baton Rouge, (1971)*

99. W. E. Truce and C. I. Lin, J. Amer. Chem. Soc. 95. 1*1*26 (1973).

100. E. Wedekind, D. Schenk and R. Stusser, Chem. Ber. 56. 633 (1923).

101. J. F. King and T. Durst, Can. J. Chem. 1*1*. 819 (1966).

102. D. J. Pasto and M. P. Serve, J. Amer. Chem. Soc. 8 7 . 1515 (1965).

103. L. A. Paquette and M. K. Scott, J. Amer. Chem. Soc. 9I*. 6760

(1972).

10l*. S. Oae, J . Amer. Chem. Soc. 7 8 , 1*030 (1956).

105. L. A. Paquette, J, P. Freeman and R. W. Harsser, .J. Org. Chem.

2k, 290 (1969). 112

106. G. Domschke, J. Prakt. Chem. 52. lkk (1966).

107. I. B. Johnson and J. A. Ambler, J. Amer. Chem. Soc. 56. 372

(191*0. 108. N. J. Johary and L. N. Oweer, J. Chem. Soc. 1307 (1935).

109. F. G. Bordwell and B. M. P itt, J. Amer. Chem. Soc. 77. 572

(1955). 110. S. A. Heininger and J, Dazzi, Chem. and Engr. News. 35. No. 9,

87 (1957). 111. M. F. Lappert and J. K. Smith, J. Chem. Soc. London. 322k, ( 1961 ).

112. S, Bradamante, S. Maiorana and G. Pagani, J. Chem. Soc. Perkin

I, 282, (1972).

113. E. E. Van Tamelen and S. P. Pappas, J. Amer. Chem. Soc. 85 .

3298 (1963).

Ilk. Handbook of Chemistry and Physics, The Chemical Rubber Co..

k9th Edit. 1968- 1969.

115. H. N. Lin, Thesis, Northeast Louisiana State College. Monroe,

Louisiana. 1970. 2 .0 3.0 4 .0 3 .0 PfMlT) *.0 7.0 t 0 ».o ib HZ

MO XXI iOO

'wrJ*J jj j '

t.o 7.0 6.0 5.0 rpM(2) 4.0 3.0 2.0 1.0

Nmr Spectrum o f Compound 171 f> o NKt MU) SOLVENTi CDC1

Nmr Spectrum o f Compound 102 m * ( T )

400 300 300 100 0 Ht m

io n MHt NMR

sor.vtN T i c d c

Ml V\

60 5:0 TTT 4 .0 30 20

Nmr Spectrum o f Compound 122 FH ft(T) ooo I soo 000 200 100 0 Hi I I 100 I so 100 Wt SOLVEHTt -MSO-d

DMS 0

■

. I . . . . . ,,, ~T 6,0 S ^ orfTOl 4.0 3.0 2.0

Nmr Spectrum o f Compound 2.0 3.0 4.0 5.0 PPM |T) 6.0 7.0 80 ' 9.0 1*0 'I I ' 1 t „ ...... 1 < ■ I ■ - - , - . 1000

400 JOO LOO -sq— a 100 »

T « t

8.0 7.0 6.0 5.0 PPM (i) 4.0 3.0 2.0 1.0

"■J Nmr Spectrum o f Compound 1 ’': ’ r " ' --|':....:' I ^ | ■:1 ■1"11 1: "' -4" ■:1" | ■

40 0 31X1 200 100

4.0 3.0 20 to

CD Nmr Spectrum of Compound 138a A

4 0 0 JOO 200 100

100 MHi SOLVENT* CDC1

I ^ ^ . 1 ^ ^ . - L - * . . l . * . . 1 .... i .... t , . . _ l .... I , - . . i . . , . \ . ... l . 1 40 3 0 zo /.o 'O Nmr Spectrum o f Compound 138b ‘SO-CI

Nmr Spectrum o f Compound ljj£ 120 2.0 3 0 4.0 PfM(Tl 6.0 7.0 9.0

300 1*0

t.O 7.0 5.0 4.0 3.0 2.0 1.0

Nmr Spectrum o f Compound 167 X

300 100

ion rHz N*H

SOl.VENTi CDC 1

ilfl** I'

4.0 3.0 2.0 /

Nmr Spectrum of Compound 160 r ^ /W 4000 3(JoO 2000 1500 CM 1000 900 800 700 ]0Q [uulu u h x i 100

Z60

40 oc *-20

WAVELENGTH (MICRONS)

Ir Spectrum of Compound 171 4000 3c)oO 2000 1500 CM* 1000 900 800 700 u tilu n li ) ± 1 1 a,i- L - i Li,Li) j.LiJ -,1 X j_J _i 1—i— 1—i— I— l _ L lu J 11.> il i.m i-i x.i.1 Li-j-uJ-Lxj i L ui, i.

s ' ' A

WAVELENGTH (MICRONS)

Ir Spectrum o f Compound 102^ 4000 3doo 2000 1500 CM 1000 900 800 700 100

LU Z60 fV'V V » vw*v V /

*40

QC *“20 ■ocw,

WAVELENGTH (MICRONS)

Ir Spectrum o f Compound 122 ro I

4000 3c)00 2000 1500 CM 1000 900 800 700 i o o | u u W ^ —Uin..liii 111 m J 100

260

,?40

WAVELENGTH (MICRONS)

Ir Spectrum o f Compound 15^ 2000 1500 CM 800 700 100

Z60

SQ— Ct 40

WAVELENGTH (MICRONS)

Ir Spectrum of Compound 137

ro -3 4000 3(j00 2000 1500 CM 1000 900 800 700 , 1j i.i 11 n u 11.1 lO oh ^ 1111 100

Z60 60

------

WAVELENGTH (MICRONS)

Ir Spectrum o f Compound 138a -I 4000 3doO 2000 1500 CM 1000 900 800 700 Miml.i.Mu-i i-11 l.iij-Jj Li. 1.1 1. j—L.i I i —I 1 N m 111111 ■ 111 11 11 i I1 ■ ■ i I i i i i I i i i H e & 100 'G c i

80 80 OJ V ' - ' i U 260 60 < fro llll \\l 1/ 40 Z i n ;i , < ’ | i fill I oc *-20 “ 1 i n i .. 20 ! Ili j 11 3 4 5 6 7 8 9 10 11 12 13 14 15 WAVELENGTH (MICRONS)

Ir Spectrum of Compound 1^8^ I

-I 4000 3(JoO 2000 1500 1000 900 800 700 null m I i i l i I i ■ 11111 i I i i i 1 i I J Llu.i1 in 111 i i —lI i i l _ l 1 l i u i I i i i I i i i

WAVELENGTH (MICRONS)

Ir Spectrum o f Compound 157

hi o 4000 3doO 2000 1500 CM 1000 900 800 700

[miltn 11 i i i i I i-i 1 I 1 i 1.1 1-L.l-i—L_l_l ..I—L - i. 1 i ,1 . i. , (n II1 M I I I 1 1 I 1 I I I 1 I I I I 1 * I ' 1 1 1 I \ 1 1 100

80 80 .11, f i ; ]f 1 iT l| ., ;

!! I:. Z60 A il l ! r, h ' 40 3z * ° i! < i . DC •“20 20 r I .1 '■ I I iu li ■/ ■jCCL- 7 8 9 10 11 12 13 14 15 WAVELENGTH (MICRONS)

Ir Spectrum of Compound 167 (WW 4000 3(JoO 2000 1500CM 1000 900 000 700 n u l l II I I I i I 1 0 0 100

260

WAVELENGTH (MICRONS)

I r Spectrum of Compound 160

ro VITA

Huei-Nan Lin was born September 29, I 9IA in Pingtung, Taiwan, and is the son of Dr. and Mrs, Ko-Pen and Huan Lin. Ke was educated in Chungcheng Elementary School and graduated from Tainan

First High School in 19^3* After attending Taipei Medical College for a short period of time in 1963» transfered to Fu-Jen

University in Taipei from which he received his Bachelor of Science degree in 1968. He received the Master of Science degree from

Northeast Louisiana State College in January, 1970 and attended the Graduate School of the University of Maryland, Coliege Park, in Spring 1970 and transferred to the Graduate School of the

Louisiana State University in Baton Rouge, Louisiana in September,

1970. He married Laura Hui-Shue Kwo in 1973 and is currently a candidate of Doctor of Philosophy in the Department of Chemistry,

Louisiana State University.

133 EXAMINATION AND THESIS REPORT

Candidate: Huei-Nan Lin

Major Field: Chemis try

Title of Thesis: The Chemistry of Non-Stabilized Sulfenes

Approved:

Major Professor and Chairman

Dean of the Graduate School

EXAMINING COMMITTEE:

4 < - ^ ^ /Mjr u

f '

Date of Examination:

April 22, 1976