The Chemistry of Azasulfonium Salts

INFORMATION TO USERS

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

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

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

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

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

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

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

Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 74-17,777 GRUETZMACHER, Gordon Dwight, 1946- THE CHEMISTRY OF AZASULFONIUM SALTS. The Ohio State University, Ph.D., 1974 Chemistry, organic

University Microfilms, A XEROX Company , Ann Arbor, Michigan BEE CHEMISTRY OF AZA.SULFONHJM SALTS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Gordon Gruetzmacher5 B.S.

-jf- * #

The Ohio S ta te U n iv e rsity

197]l-

Reading Committee: Approved Ry

Dr. John S. Swenton

Dr. Jack Hine

Dr. William G. layers

Dr. Paul Go Gassman A dviser Department of Chemistry To Patricia ACKNOWLEDGEMENTS

The author -wishes to thank Dr. Paul G. Gassman for suggesting this problem., for his many ideas throughout the course of this work, and for many stimulating discussions about chemistry and sundry other topics.

To his collegues in chemistry, the author is indebted for their assistance in solving the day to day problems that arise while doing chemical research. To Clinton Harrington, the author leaves without any remorse the Sfaces. Special mention must be made of Rogie Drewes, who was a good friend of, beer drinker with, and screw-off with the a u th o r. *

Finally, the author is also deeply grateful to his wife, Pat, for her patience, understanding, and encouragement during the course of his graduate career. VITA

Gordon Gruetzmacher, son of Carl H. and Viola Mueller Gruetzmacher,

■was born on November 7, 19^6 in Columbus, Texas. He attended primary and secondary schools in various small towns in South Texas. For two years,

September, 1965 thru May, 1967> he attended Del Mar College in Corpus

Christi, Texas. In September, 1967 he entered The University of Texas at Austin where he recieved a B. S. in Chemistry in May, 1969- He married Patricia H ill of Sinton, Texas, on August 22, 1969. Ctoe month later, he entered the Graduate School of The Ohio State University.

While at Ohio State, he held the positions of teaching assistant and research associate. He recieved his Ph. D. in organic chemistry from

The Ohio State University in March of 197^-- TABLE OF CONTEST IS

Page

ADKN0WLEDGMEN1S...... i i

Y I T A o...... i i i

TABLESrx

ILLUS TRATIONS...... x

INTRODUCTION...... 1

PART I. Historical ...... 1

PART I I . The Erohlem ...... 26

RESULTS AM) DISCUSSION...... I ...... 35

PART I. The Synthesis, Thermolysis, and Basic Rearrangement of Azasulfonium Salts ...... 35

PART II. The ortho-ALkylation of Anilines ...... 6 k

PART III. Die Reaction of Anilines with Halodimethylsulfonium Halides ...... 8l

EXPERIMENTAL...... 90

IT-te rt-Butylaniline (80) ...... 9°

N -ter t -Butyl -p -toluid ine ( 8 3) ...... 90

IT -tert -Butyl -chlor oaniline (102) ...... 90

N-tert -Butyl-4-fluoroaniline (97) ...... 90

p-(N-te rt-Butylamino)nitrohenzene (82) ...... „ 9°

N-te rt-Butylanilinodimethylsulfonium Chloride ( 8 3) ...... 91

IT-te rt-Butyl-p-toluidinodixaethylsulfonium Chloride ( 83.)...... 91

iv Page

II-te rt-Butyl-4 -chloroarxilinodimet hv 1sulfonium Chloride ( 8 8) . . . 92

H-te rt-Butyl-4-f luor oanilinodlme thylsulf onium Chloride ( 8 6 ) . . . 99

IT-te rt -Butyl -p-t oluidinotetr amethylene sulf onium Chloride (Sg.)...... 94

N -te rt -Butylanilinodime thylsulf onium Trifluorpacetate (§4).«.. 95

IT-tert -Butyl -p -t oluidinodimethyls ulf onium IT if luor o - a c e ta te (§5.). T ...... 96

IT -te rt -Butyl -4 -chlor oan ilinodiine thylsulf onium Trifluoroaeetate (25.) ...... 96

Thermolysis of N-te rt -Butylanilinod ime thylsulf onium Chloride (&£-)...... 96

Thermolysis of N -tert-Butyl-p-toluidinodimethylsulfonium C hloride ( 8 5)...... 97

Thermolysis of IT -te rt -Butyl -4 -chlor oanilinodime thylsulf onium C h lorid e (88 ) ...... '...... 98

Ihermoly s i s of IT -te r t -Butyl -4 -f luor oanilin odimethyls ulf oni um C hloride ( 8 6 ) ...... 99

Kinetic Procedure ...... 100

Thermolysis of IT-tert -Butyl-p-toluidinotetramethylene - _. Sulf onium Chloride (§2.)...... 103

Pyrolysis of 4-Chloro-n-butyl-9 -(4-te rt-butylaminotolyl)- sulfide Hydrochloride...... 10J

Bis-3-(4-aminotolyl) Disulfide (10£) ...... 104

6 -Amino-m-toluenethiol Hydrochloride (108) ...... 104

4-Chloro-n-hutyl-3-(4-aminotolyl) Sulfide (105) ...... 105

N-te rt -Butyl-(N-methy 1thio) -p-toluidine (122.) ...... 106

Sodium Methoxide Rearrangement of N-te rt -Butylanilino- dimethylsulf onium Chloride ( 8j ) ...... 107

v Page

Sodium Methoxide Rearrangement of II -te rt - Butyl -u- toluidinodimethylsulf onium Chloride ( 8^ )...... 108

Sodium Methoxide Rearrangement of IT-te rt-Butyl-4-chloro- an ilin o d im eth y lsu lfo n iu m C hloride IB5JT7" ...... 108

Sodium Methoxide Rearrangement of N-te rt-Butyl-7~fluoro- anilinodimethylsuif onium Chloride ^ 8 6 )...... 109

Sodium Methoxide Rearrangement of IT -te rt -Butylanilino - dimethylsulfonium Trifluoroaeetate 110

Sodium Methoxide Rearrangement of h-tert-Butyl-p- toluidinotetramethylene sulf onium Chloride(82.) ...... 110

IT-t e r t -B u ty l-o -to lu id in e ( 1 ^ 6 ) ...... I l l

IT-te r t-Butyl -2, ^ -xylidine (l^X) ...... 111

^•-Methoxy-2-(thiomethoxymethyl)aniline (iB j) ...... 112

U-Methyl-2-(thiomethoxymethyl)aniline (16^) ...... 115 6 2-(Thiomathoxymetbyl)aniline (160) ...... H ^

k -Carhoethoxy-2-( thiomethoxymethyl)aniline (178) ...... 115

4-Chloro-2-(thiomethoxymethyl)aniline (171) ...... - 116

U-ITitro-2-(thiomethoxymethyl)aniline (l§2) ...... 117

2 -Amino~ 5 -(thiomethoxymethyl)hiphenyl (l 8o ) ...... 118

^ 6 -Methyl-2-(thiomethoxymethyl) aniline (l 6 £ )...... 119

5-Methoxy-2-( thiomethoxymethyl)aniline (l84) and 5-Methoxy-2-(thiomethoxymethyl)aniline (iS^.)...... 120

5-Methyl-2-(thiomethoxymethyl)aniline (167) and 5 -Methyl-2 -(thiomethoxymethyl)aniline (168) ...... 122

5-0hloro-2-(thiomethoxymethyl)aniline (172.) anc^ 5-Chloro-2-(thiomethoxymethyl)aniline (17^.) ...... 125

preparation of 2-Chloro-6-(thiomethoxymethyl)aniline ( 176 ) . . . . 12k

IT-Methyl-2-(thiomethoxymethyl)aniline (201) ...... 12k

v i Page

N-M ethyl- 2 -(l-thioethoxyethyl)aniline ( 202) ...... 125

N-Methyl-2-(1 -thiopropoxyprcpyl)aniline (P&i) ...... 126

2-(2-Tetrahydrothienyl)aniline (196) ...... 127

2 -( 2 -Tetrahydrothiopyranyl)aniline (l^j) ...... 128

N -A liy la n ilin e (19,5.) ...... «...... 129

Benzenesulfenanilide (l§2) from n-Butyl Phenyl Sulfide ...... 130

Benzene sulf enanilide (1^2) from Phenyl Isopropyl Sulfide 131

W-2 Raney Nickel ...... 131 o-Toluidine (l 6 l ) ...... 132

4-Chloro-o-toluidine (20^) ...... 132

4-Carhoetho:xy-2-methylanil:me (20j) ...... 133

2 ,6 -X ylidine (204)...... 134

2-Methyl-p-anisidine (206)...... 134

2,N-Dimethylaniline (j210) ...... 134

2-Ethyl-N-methylanilime (211) ...... 135

2-n-Butylaniline (208) ...... ’...... 135

2-n-Pentylaniline {20^}...... 135

4 -Metho:xy-2 -(thiomethoxymethyl)aniline ( 187 ) from 227 ...... 136

Preparation of iBj from 128...... 137

Preparation of 16^ from 128 ...... 138

Preparation of 160 from 128 ...... 138

Preparation of 171 from 128 ...... 139

Preparation of 182 from 128- ...... 139

v ia Page

Preparation of IjS from _128 ...... 1^1

Preparation o f 160 from. 2 ^ 1 ...... l ^ i

P re p a ra tio n of 1§6 from ...... 1^2

Preparation of lgS from 222...... 1^5

REPEREI'TCES...... lM -

v i i i TABLES

Table Page

1. Rates and Products of N-Chloroaziridine Solvolysis ...... 18

2. Yields of Azasulfonium Chlorides ...... 38

3. Rates of Thermolysis of ^--Substituted-W-te rt-butylanilinodimethylsulfonium Chlorides in the Presence of Lithium Chloride in Dimethyformamide ...... 48

4. Base Rearrangement of Azasulfonium Salts to 2-Thio- m ethoxym ethylaniline s ...... 62

5. Alkylation of Anilines via Aluminum Anilides ...... 66

6 . Conversions of Anilines into 2-Thiomethoxymethylanilines ...... 71

7- Ratios of 3~ and 5“Substituted-2-(thiomethoxymethyl)- anilines from meta-Substituted Anilines ...... 74

8. Reduction of 2-Thioalkoxyalkylanilines to Yield ortho-Alkylaniline s ...... 77

9. 4-Substituted-2-Thiomethoxymethylanilines from Chlorodimethylsulfonium Chloride ...... 87

10. . Gas Phase Chromatography Peak Areas of 2-Thiomethoxy- methyl-p-toluidine and Phenanthrene ...... 101

11. Data Used to Obtain Second Order Rate Constant ...... 102 ILLUSTRATIONS

Figure Page

1. Plot of

2. Plot of a vs. lo g k ...... 51

X INTRODUCTION

Part I. Historical

In modern organic chemistry, a great deal of work has been devoted

to the chemistry of carbonium ions, trivalent electron deficient carbon 1 species (l)» It was only recently that Gassman and his collaborators

(l) For a review of carbonium ions and their chemistry see G. A. Olah and P. v.R. Schleyer, eds., “Carbonium Ions,” Interscience, New York, N. Y., 1968c

i £ 2 it began a systematic study of nitrenium ions, divalent positively charged 2 nitrogen species (2) analogous to carbonium ions. A nitrenium ion

(2) P.. G. Gassman, Accounts Chem. Res., 5_, 26 (197°)•

was expected to exist in either of two electronic? states, a singlet

(£) in which the unshared pair of electrons were in the same orbital

or a triplet (4) in which the unshared pair of electrons were in different orbitals with the same spin state. Hms, a dichotomy could

arise in the chemistry of nitrenium ions, The singlet could give rise

to chemistry analogous to carbonium ions while the trip let could result

in chemistry analogous to that observed for nitrogen radicals. It

was further expected that a nitrenium ion should be much more reactive

than a carbonium ion since nitrogen is more electronegative than carbon.

Subsequent to the first reports of nitrenium ions, there have been

o several theoretical studies on nitrenium ions. Lee and Morokuma

(3) S. T. Lee and K. Morokuma, J . Amer. Chem. S o c ., 93> 6863 (19T1)«

carried out ab initio SCF-CI calculations on the lowest singlet and

triplet states of M 2' These calculations shoved that the ground state was a linear or almost linear triplet ( 3Bi). Die lowest singlet state

( 1A1) was found to be about k-5 kcal/mole higher in energy than the

lowest trip let. The singlet was found to have an H-N-H bond angle of o + c a. 115 . Die population analysis suggested that the singlet M 2 H- w ill react more like a carbonium ion and that the triplet M 2 w ill + + react like a trip let methylene. Comparison of NH2, CH2? and CH 3 showed

that all three species have a 2p orbital that is completely or almost

completely unoccupied and as a result, these intermediates are electro-

philes. It was postulated that these three reagents attack electro-

philically with the 2p orbital pointing toward the substrate. The

weakest electrophile would be CH2« The net charge on the 2p o r b ita l

on singlet M 2 a t 120° and ChJ are nearly the same, thus singlet M 2 3 would be a stronger electrophile because of its greater electronega t i v i t y .

As the H-N-H angle decreased the energy separation between the triplet and singlet states decreased. Hence, if a nitrenium ion is generated in a cyclic structure where the angle Rx-N-R^ cannot increase because of molecular restrictions and remains at its singlet equi librium angle or smaller, the singlet state gains extra stability and the singlet reaction could predominate in such cases. 4 A subsequent calculation indicated that the triplet was the ground state, being ca. 36 kcal/mole lower in energy than the singlet.

The triplet was found to have a minimum energy configuration at a

(k ) S. Y. Chu, A. K. Q. Siu, and E„ F„ Hayes, J. Amer. Chem. Soc., 2969 (1972).

. o bond angle of ca. 140 and the singlet was found to be most stable at a bond angle of ca. 120°. Interestingly, in the two previously discussed theoretical studies of nitrenium ions, ?3 4 that although energy convergence m s predicted for the lowest lying singlet and triplet states, no crossover ms observed as the H-N-H bond angle was varied between 90° and 180 . If calculations are performed on methylnitrenium ion ( 5 ) and dimethyl - nitrenium ion (6), an energy crossover can be observed in both cases.

(5) G. F. Koser, Chem. Comm., 461 (1973)* h

1 6

SFC-MO calculations of the DIDO type were performed on nitrenium ions j? and 6 . It was found that the triplets of _5 and 6 are more stable than the corresponding singlets at their minimum energy configurations

"by 27 and 26 kcal/mole. As the interatomic bond angle, 9, was decreased from l 8o°, the singlet-triplet energies of j? and <5 converge and finally cross at IO 30 fo r 5, and 109° fo r 6 . Ihus, if 0 was con strained to 90°, the singlet states of J? and _6 would be th e gpound states by 10 and 37 kcal/mole respectively. Also, for dimethylni- trenium ion ( 6 ) a t 0 values near 109°, the singlet vras found to be

3 kcal/mole lower in energy than the triplet, therefore, singlet- triplet interconversion should occur readily and singlet chemical reactivity might be expected.

A fourth theoretical study of nitrenium ions involved a comparison "*■ + + 6 of NH2, HHF , and NF 2 using ab initio calculations. The difference

( 6 ) J. F. Harrison and C. W. Eakers, J. Amer. Chem. Soc., 95j 3^7 (1973). 5 in energy, 45 kcal/mole, and bond angles 150° and 120° of the triplet

and singlet NH 2 are in agreement with other calculations . 354 I n t e r - + estingly, NF2 was found to have a ground state singlet with F-N-F bond

an g le o f 105° and the singlet was 33 kcal/mole lower in energy than

the trip let, which was found to have an optimal calculated bond angle

of 122°. This ordering of the m ultiplicities was in agreement with

that observed with the isoelectronic carbenes, CH 2 and CF2. An inver

sion of the car bene order was found with NHF+, in which the trip let with a bond angle of 135° was only V kcal/mole more stable than the

singlet with a bond angle of IO 7 0. This increased stability of the

singlet nitrenium ion was attributed to overlap of the unshared elec

trons of the fluorine with the empty p orbital on the nitrogen to stabi- 0 lize the positive charge on the nitrogen.

These theoretical studies are in agreement with and lend inter

pretative support to the observations on the chemistry of nitrenium

ions. Experimentally, nitrenium ions were first observed by Stieglitz

and co-workers in a series of papers on the rearrangement of substituted 7 hydr oxylamine s and W-chlor amine s. Ihis group found that tritylhy-

droxylamines and trityl-N -chlor amines rearranged readily to give, after

( 7 ) J* Stieglitz and P. N. Leech, Chem. Ber., 45, 2147 (1913) | J. Stieglitz and P. N. Leech, J. Amer. Chem. Soc., 3 6 , 272 (1914); J. Stieglitz and B. A. Stanger, ibid., 35, 2046 (191£>).

hydrolysis, benzophenone and aniline; however, disubstituted hydroxyl-

amines and disubstituted N-chloramines either did not rearrange or they rearranged only under extremely forcing conditions. Stieglitz concluded that the rearrangement m s occurring via the intermediacy of a monovalent nitrogen species, a nitrene. More recent data , 8 showed that if suitable conditions are used the disubstituted compounds can

(8 ) R. T. Conley and H. Brandman, unpublished results.

also be rearranged. These results lead to the conclusion that the rearrangements observed actually involved a divalent positively charged nitrogen intermediate, nitrenium ion, rather than a univalent nitrogen intermediate, a nitrene.

Newman and Hay 9 provided evidence# for a nitrenium ion inter - mediate in a study of the migratory aptitudes of aryl groups in the

(9) M. S. Neman and P. M. Hay, J. Amer. Chem. Soc., j^5, 2^22 (1953)*

rearrangement of tritylhydroxylamines. As the electron-donating ability of the para-substituent increased, the migratory aptitude of the aryl group increased. The thermal decomposition of trity l azides 7 was insensitive to the nature of the aryl group . 10 This rearrange ment was almost certainly via a nitrene. These two observations lend

(10) R. A. Abramovitch and B. A. Davis, Chem. Rev., 64, 149 (1964).

credence to the intermediacy of a nitrenium ion in the Stieglitz rearrangement.

A nitrenium ion may be postulated as an intermediate in some cases of the Beckmann rearrangement . 11 When oximes are treated with

( l l ) For a review of th e Beckmann rearrangem ent, see P .. Smith in ‘‘Molecular Rearrangem ents,” P. Mayo, e d ., V ol. 1 , I n te r sc ie n c e , New York, N. Y., 1963? pp 483-507.

strong acids they are converted to substituted amides. The group that usually migrates is the one trans to the hydroxyl. In some cases, however, the cis group migrates and in cases where both substituents are alkyl a mixture of the two possible amides is formed. Thus, it appears that a concerted, backside attack by the migrating group does not always occur and that a discrete unsaturated nitrenium ion, iminium c a tio n , may be involved in th e p ro ce ss.

0 II R2 -C-NH-Ri

0 £ 12 Lansbury and co-workers attacked the problem of unsaturated nitrenium ions in the Beckmann rearrangement of indanone oximes.

(12) P. Lansbury and N. Mancuso, Tetrahedron Letters, 2445 ( 1965 ).

These researchers reported that when steric considerations precluded the possibility of prior isomerization of the oxime and made aryl migration unlikely owing to steric compression in the transition state, the alkyl group cis to the hydroxyl group migrated preferentially.

Iminium cations were considered as possible intermediates in this reaction. A novel intramolecular insertion reaction supported this proposal. t

In this reaction 4 -bromo-7-t-butyl-1 -indanone oxime ( 7 ) was con verted to imine ( 8 ) in 73 p yield by treatment with polyphosphoric 13 n acid. In the insertion reaction leading to _o, it was possible that

(13) P* Lansbury, J. Colson, and N. Mancuso, J. Amer. Chem. Soc., 86 , 5225 (1964).

the species attacking the proximal C-H band could be either the cationic

CH.

1 £ nitrogen intermediate _9, or the vinyl nitrene 10. In order to distinguish "between these two possible intermediates, rearrangement of oxime was carried out in deuterated polyphosphoric acido If the reaction was proceeding through the iminium cation 9, no deuterium would "be found in the product, while if the nitrene 10 were the inter mediate, the product should have one deuterium atom at the carbon

CHq

!+ N-H

8 -H+

Br 10 adjacent to the imine carbon. Since no deuterium was found in the product, the authors concluded that the unsaturated nitrenium ion was the intermediate.

Dialkyl nitrenium ions were postulated by Emmons 14 in 1957 in his study of the chemistry of oxaziranes. He found that 2-t-butyl-3- phenyloxazirane (ll) when treated with aqueous acid gave benzaldehyde

(1*0 Wo D. Emmons, J. Amer. Chem. Soc., 79 , 5739 (1957).

and t-butylhydroxylamine presumably via protonation on the oxygen, ring opening to give the indicated carbonium ion, and hydrolysis. However, when the carbon atom of the oxazirane ring did not possess a carbonium 10

OH ' / • ° \ H+ + | CeHsC N-£(CH3 )3 -> ! c sh 5c h -n -c(ch 3 )3

11 J HI CeHsCHO + (CH3 )3C-N-OH

ion stabilizing group such as phenyl, the ring was cleaved in a different manner. The ring opening of 2-n-butyloxazirane (12) pro duced formaldehyde, ammonia, and butyraldehyde. In this case, pro tonation occurred again on the oxygen j however, in contrast to 11, th e N -0 bond was cleaved to form a nitrenium ion followed by hydrogen

8 migration to the electron deficient nitrenium ion. Methyl migration to an electron deficient nitrogen was observed in the opening of

2 -t-butyloxazirane ( 15 ) which afforded formaldehyde, methylamine, and acetone.

HO /\ H+ I + ch 2 n -(c h 2 )3ch 3 -> ch 2 -n -(c h 2)3ch 3

12 H-migration

HO I + CHsO + HH3 + CH3 (CH2 )2CH0 f- CH2 -N-CH(CH2 ) 2CIt3

II 11

HO H+ CH2 ~ —~N -C ( (513)3 ■) CH2-N-C(CH3)3

CH3 -migration

0 HO If + CH20 + CH3HH2 + CH3CCH3 {■ ch2 -n -c ( gh3 )2

ch3

In 1961 Gassman and co-workers began a systematic study of the chemistry of nitrenium ions. Priority was given to establishing that alkyl migration could occur -to a divalent electron-deficient nitrogen species where both substituents on nitrogen were alkyl groups. Uiis approach was dictated by the tendency of alkyl groups to migrate to cationic centers and not to radical centers, in contrast to the behavior of aryl groups and hydrogen which are known to migrate to both cationic and radical centers. Thus, the migration of an alkyl group from carbon to divalent nitrogen would indicate that the nitrogen was significantly electron deficient.

Refluxing N-chloroisoquinuclidine (l4) in a methanolic solution of silver nitrate gave a 60p y ie ld of 2 -methoxy-l-azabicyclo[ 3 . 2 . l ] - octane (IT ) . 15 Such a skeletal rearrangement plainly requires the migration of an alkyl group with its bonding pair of electrons from 12

(15) P° G. Gassman and B. L. FoxP Chem., Cammun ., 153 (1966); P- G. Gassman and B. L. Fox, J. Amer., Chemo S o c ., 89, 33$ (1967).

<+ L 16

+ vJT CH3OH

H 17 carbon to nitrogen. The reaction could involve loss of chloride ion to give the nitrenium ion (l£) as an intermediate which could undergo an alkyl migration to form 16 followed by solvent capture to produce

17• An alternative route would involve concerted loss of chlorine and migration of the allyl group with its pair of bonding electrons 13 to y ie ld 16 directly. Regardless of which path was operative in this rearrangement the alkyl group must have migrated with its electron pair, and thus, an electron-deficient nitrogen species must have teen involved. A similar rearrangement was observed in the methanolysis of ^,7,7-trimethyl-2-chloro-2-azabieyelo[2.2.1]heptane (l8) which gave

19, 20, and 21 in 59 5 20, and ’fp yields respectively.16

(16) P. G. Gassman and R. L. Cryberg, J. Amer. Chem. S o c ., 90, 1355 (1968); P. G. Gassman and R. L. Cryberg, i b i d . , 91 5 20"^7 (1969).

CH; CH- CH;

+ CH< Cl CH3 O H

18 23 20 21

Unfortunately, the observations discussed thus far did not permit a distinction to be made between a divalent nitrogen species with a unit positive charge and a slightly electron-deficient nitrogen species which represents only a transitory point on the reaction pathway. As previously discussed, the solvolytically generated singlet nitrenium ion (22) could undergo spin inversion to the triplet (2£) which could 14 abstract two hydrogen atoms from solvent to give the protonated form of the secondary amine (_2£). The first evidence for a discrete nitrenium ion intermediate 'with a fu ll positive charge was provided by C-ass- man and Cryberg.1'" They found that when 18 was solvolyzed in solvents

(17) P. G. Gassman and R. L. Cryberg, J . Amer. Chem. Soc., 91> 5176 (1969).

R-R -R R-R -R . C_g3.0H ) r VE I H

22 23 2k containing heavy atoms, such as chloroform or bromoform, the percen tage of the secondary starting amine in the product mixture increased.

Die authors proposed that, since heavy atom solvents are known to

18 catalyze spin inversion, the occurrence of the secondary starting amine in the product mixture was due to hydrogen atom abstraction by

(l8) A. G. Anastassiou, J. Amer. Chem, Soc., 88, 2322, (1966); C. D. Dijkgraaf and G. J. Hoijtink, Tetrahedron Suppl., _2, 179 (1963).

the triplet nitrenium ion followed by neutralization of the protonated amine in work-up. In order for spin inversion to occur, the nitrenium ion must have existed as a discrete entity with unit positive charge on nitrogen.

Ring expansions and contractions via nitrenium ions have also been observe do 19 When 26 /•was solvolyzed in methanol containing silver trifluoroacetate a mixture of 0 ( 85$) and 29 (7b) was obtained.

(19) P. G. Gassman and A. C a rra sq u illo , Chem. Commun., 495 (19&9)} P. G. Gassman and A. Carrasquillo, Tetrahedron Letters, 109 (197l)«

iisH s

CeHs / N a / m \ CHS 'sCHc N-Cl I CHS 27 b

CHS

0 N 0 H II II II - I CsRsCCHaCHaOCHs <------CsHsCCHaCHsOCHs CeHsCCHsCI^CHa

52 28 Methanolysis of N-chloro-2-phenylazetidine (j5l) in the presence of silver ion gave benzaldehyde and a mixture of and ^b. These reactions can be rationalized in terms of the generation of a singlet nitrenium ion followed by alkyl migration thus producing a reasonance stabilized species which is hydrolyzed.

Kinetic evidence for the heterolytic cleavage of the N-Cl bond tinder solvolytic conditions was provided by a study of the kinetics 1 6 J? 6H5 Ag+ CeHs

% \ Cl

32a ^2b

CeHsCHO + H2N(CH2 ) aOCH3 + HaN(CH2)2OH

S i

of N-chloroaziridine ionization in polar solvents. The solvolytic ring opening of cyclopropyl tosylates and chlorides, ? 20 which21 obey

th e o r b i ta l symmetry c o n sid e ra tio n s of Woodward and Hoffman, were22

(20) C. H. DePuy, L. G. Schnack, J . W. Hauser, and W. Weidemann, J . Amer. Chem. S o c ., 8 7 , 4006 (1965 ).

(21) P. v .R . Schleyer, G. W. VanDine, U. Schollkopf, and J . P au st, ibid., 88, 2868 ( 1966 ); U. Schollkopf, F. Fellenberger, M. Patsch, P. v.R. Schleyer, T. Su, and G. W. VanDine, Tetrahedron Letters, 3639 (1967); P. v.R. Schleyer, T. Su, M. Saunders, and J . C. Rosenfield, J 0 Amer. Chem. S o c ., 91, 517^ (1969)*

(22) R. B. Woodward and R. Hoffman, i b i d . , 87 , 395 (1965)* Also see H. Co Longuet-Higgins and E. W. Abrahamson, ib id ., 87 , 20b5 (1965 ).

the basis of this study. Studies of the concerted electrocyclic ring

opening of cyclopropyl cations to allylic cations have shown that a

consideration of electronic and steric interaction coupled with molecu

lar orbital symmetry arguments permits predictions of reaction rates

and products that are in complete agreement with experimental

fin d in g s . 2 0 , 2 1 17

As shown by Brois, N-chloroaziridines (5 5 ) are stereochemically stable since inversion at nitrogen is a relatively slow process.23 24 Gassman and co-workers reported that solvolysis of various N-chloro-

(2 3 ) So Jo B rois, J . Amer. Chem. S o c ., 90, 506 , 508 (1968 ).

(24) P. g . Gassman and D. K. Dygos, ib id ., _91, 1543 (1969); P. G. Gassman, D. K. Dygos, and J. E. Trent, ibid., J92, 2084 (1970).

aziridines ) in methanol proceeded with widely divergent rate constants to yield in all cases a product mixture that upon hydrolysis gave two moles of carbonyl compound and one mole of ammonium chloride.

As shown in Table 1, when no carbonium ion stabilizing groups are availble, as in ^5a, the reaction was slow. However, when methyl groups are available to stabilize the incipient carbonium ion, the effect must be balanced with steric strain generated during the disrotatory ring opening. The solvolysis of N-chloroaziridines exhibited kinetic behavior identical to that of the cyclopropyl chlorides and tosylates.

This indicates that heterolytic cleavage of the N-Cl bond was involved.

0 0 H II RiCR2 + R3 CR4 + NH4 CI TABLE 1

Rates and Products of N-Chloroaziridine Solvolysis

N-Chloroaziridine ^rel Products

R1=R2=R3=Ra =H 1 2 CH20 + NH4 C I 27b: R1=R2=R3 ; R4=ch3 15 CH3 CH0 + CH20 + NHaCI ^ J c : R!=R2=Ra ; R3=CH3 210 CH3 CH0 + ch2o + nhIci ^ T d : Ri =R3 =H ; R ^ R ^ C H s 1^90 2CH3 CH0 + MHaCI 3Te: Ri =R2=H; R3=rI=GH 3 i860 CH3 COCH3 + CHpO + NHaCI 27£: Ri"R4=H; R2=R3=CH3 155,000 2 CH3 CHO + M D C l

Once a firm basis for nitrenium ion theory had been established,

several examples of their synthetic utility began to appear in the

literature. Certain azabicyclics can be generated via intramolecular addition of a nitrenium ion intermediate to a double bond in a “it route” ,, . 25 synthesis. it was found that solvolysis of the R-chloramine J 56 in the presence of silver ion gave jy? in hyjo yield, while cyclization of

(25) P. G. Gassman, F. Hoyda, and J. Dygos, J. Amer. Chem. Soc., 90, 2716 (1968); P. G. Gassman and J. Dygos, Tetrahedron Letters, W 5 (1970).

^8 gave 39 and as the major bicyclic products. In both cases the nucleophile added trans to the attacking nitrenium ion. 19 ch 3

ch2 - n - c i

38 39 40

Hie solvolysis of N-chloroaziridiaes was utilized by Horwell and

Rees26 in a synthesis of isoquinoline. dhe aziridine (4l) was converted

(26) D. C. H orw ell and C. W. Rees, Chem. Commun., 14-28 (1969).

to its N-chloro analogue which solvolyzed to give isoq.uinol.ine (44) via presumably 4^» ihe mechanism is analogous to that described for H-chloroaziridines, heterolytic cleavage of the N-Cl bond to give iminium ion (4^), which then loses a proton to give isoquinoline.

H Cl Cl-

4 l 42

-H+

44 20 2T Edwards and co-workers discovered a transannular insertion reaction of a nitrenium ion into a C-H bond. N-chloroazacyclononane

(27 ) 0. E. Edwards, D. Vocelle, J. W. ApSimon, and F. Hague, J„ Amer. Chem. S o c., 8 7 , 678 (1965).

(J+5.), upon treatment with silver ion, gave a 6&p yield of indolizidine

(46). The reaction was proposed to go thru a mechanism involving

Ag+ I

45 46

Ri + Ri ^1\T H-C; -4 N-H + C — Rv+ / \ H z" *^ \ * E/ H - v

hydride abstraction followed by ring closure to give the protonated tertiary amine. 5

Kovacic and co-workers have studied the aluminum chloride cata lyzed rearrangements of W,H-dichloroamines. Treatment of l-N,N-di- chloroaminoadamantane (47) at low temperature with aluminum chloride in methylene chloride gave 48 which could be trapped by methoxide to , 28 yield 49 in ca. 5 Qp yield. More recently it was found that treatment of 1-N,N-dichloroaminoapocamphane ( 50) with 2 equivalents of aluminum •

(28) P. Kovacic, J.-H. Liu, E. M. Levi, and P. D/Roskos, J. Amer. Chem. s o c ., 23, 5801 (1971 ). 21 Cl Cl Cl ^ / \ N N+ Cl

l a

CH.

J— ~7

l a 48 chloride in methylene chloride at ca. 75 produced l-chloro-3,3-dimethyl- 23 2-azabicyclo[2.2.2. ]octane ( 51) in 1C$ yield. The major nonbasic

(2 9) R. D. F is h e r, T. D. Bogard, and P. Kovacic, J . Amer. Chem. S o c., <£, 3646 (1973).

products _52, 52., and 54 from rearrangement of £ 0 , result from 6 scission of the C-C bond present in the shortest bridge. Little or

NC12 Cl 0 0

N-H + + +

5it 22 no inhibition was apparent when radical inhibitors were added. In both of these examples ring expansion occurred to a nitrenium ion which was not constrained in a ring.

When N, N-dichlorotri -n-butylcarbinamine (j?5) was treated with aluminum chloride at 0° in methylene chloride followed by acid hydrol ysis, di-n-butyl ketone ( 5 6 ) and n-butylamine (_5T) were produced in , 30 95 and 9 respectively. The mechanistic interpretation involved

(30) T. A. Kling, R. E. White, and P. Kovacic, Jo Amer. Chem. Soc., 9^, 7^16 (1972).

AICI3 (n-Bu)3C-N-Cl2 ------— r —) (n-Bu)3C-N-Cl -C l 55.

1 C l +C1 : + Cl (n-Bu)2C-N-n-Bu (n-Bu)2C-N-n-Bu Cl

H20 , H+

H (n-Bu)2C-N-n-Bu (n-Bu)2C0 + n-BuNH2 OH 56 SL abstraction of chloride ion by aluminum chloride and migration of an alkyl group from carbon to a nitrenium ion. This was the first example of 1,2 shift to a positive nitrogen derived from a simple open-chain

N-haloalkylamine. 23

fhese three rearrangements are probably examples of the stabil

ization of a nitrenium ion by the unshared pair of electrons on a

heteroatom as predicted by the ab initio calculations of Harrison and Q E akers.

Nitrenium ions can also be generated frctn precursors other than

H-chloramines. When the hydrochloride of 58 was treated with 1.1

equivalents of isoamyl nitrite at 50° in methanol,31 21 and 20 were

produced in 67 and 21$ yield, respectively, a finding analogous to

(31) P* G. Gassman and K. Shudo, J. Amer,, Chem. ooc., £ 3 , 5899 (1971)*

+ CK H 20

16 results previously discussed. A similar rearrangement was observed

when N-bromamine (59) was treated with a methanolic silver perchlorate

solution.n 4. • • 32 Reaction at 25 O for thirty minutes gave 20 as the only

(3 2 ) P. G. Gassman, K. Shudo, R. L. Cryberg, and A. B attisti, Tetra- hedrcn Letters, 875 (1972). 2 h isolable product in JOfo yield, while solvolysis at 0° for thirty minutes led to the isolation of a 3% y ie ld of 00 and a 2C$ yield of

2 0 . Ihese results again are analogous to previously discussed fin d in g s . 16

CH3 ' w - ' ch3 JL c iT s

A / \ Br H 52. 60

Gassman and Hartman found that hydroxylamine derivatives could he 33 used as nitrenium precursors. They found' that the methanolysis of a series of piperidin-l-yl benzoates gave a p of + 0 .6 8 while methanolysis

(33) P* G. Gassman and G. D» Hartman, J. Amer. Chem. Soc., 95s W-9 (1973).

of a series of 1 -phenylcyclohexyl benzoates gave a p of ca. + 1 .

This indicated that the transition state for heterolytic cleavage of an N -0 bond to generate a nitrenium ion occurred earlier in the bond breaking process than for heterolytic cleavage of a C-0 bond to produce a tertiary carbonium ion. Further evidence for a nitrenium ion inter mediate in the solvolysis of hydroxylamine derivatives was the 25 methanolysis of 61 that produced 20 , 21 , and p-nitrobenzoate anion.

A competing reaction of transesterification also afforded equivalent amounts of 62 and methyl-p-nitrobenzoate. Pant II. The Problem

Gassman and Campbell have done a systematic study of the aryl 3 4 substituted nitrenium ion (£ 9?), anilenium ion. been

(34) P. G. Gassman, G. A. Campbell, and R. C. F re d e ric k , J . Amer. Chem. Soc., 94s 3884 (19T2); P. G. Gassman and G. A. Campbell, J . Amer. Chem. S o c ., 94, 3391 (1972).

previously demonstrated that various chlorinating agents produced o- and p-chloroanilines and that IT-chlorination preceded chlorination of the aromatic nucleus 35 No mechanistic studies had been done and two reaction pathways seemed plausible. Hie in itially formed N-chloroaniline

(35) R* S. Neale, R. G. Schepers, and M. R. Walsh, J. Qrg. Chem., 29, 3390 (1964); P. Haberfield and D. Paul, J. Amer. Chem. Soc., Bj, 5502 (1965).

R R N. \ R R N N \ N N

+ *

2 6 could either ionize to form amide anion and positive chlorine followed hy electrophilic attack on the aromatic ring hy the chlorine (path a), or cleavage to form a nitrenium ion and chloride anion followed by nucleophilic attack of chloride on the aromatic ring (path b).

In order to distinguish between these two mechanisms, a series of H-tert -butyl-N-chloroanilines were synthesized and the rates of 34 solvolysis measured. ' ph.e substituents gave an excellent correlation with Hammett cr+ values, giving a p of -6.35» This large negative p provided evidence that the solvolysis of N-chlor oaniline s proceeded via an anilenium ion, phenyl nitrenium ion, in which the charge was extensively delocalized into the aromatic nucleus. For comparison, the solvolysis of 1 -chloro-l-arylethanes produced a p of -4.50 . 36

(3 6 ) C. Mechehynck-David and P. J. C. Fierens, Tetrahedron, 6, 232 (1959).

Thus, it would appear that the transition state for the solvolysis of a 1-chloro-l-aryle thane to yield a trivalent carbon cation involved

less charge delocalization into the aromatic ring than did the transition

state for the solvolysis of N-tert-butyl-N-chloroanilines to produce a nitrenium ion.

Synthetically it was found that the products obtained from the

silver trifluoroacetate assisted methanolysis of N-te rt-butyl-N-chloroanilines differ greatly, depending upon the nature of the substituents 34. on the aromatic ring. ‘ For electron-rich N-chloroanilines that are not substituted in the para position, the major process involved form ation of anisidines. For electron-rich N-chloroanilines substituted in the para position with carbonium ion stabilizing groups the principle products were derivatives of 2, 5-cyclohexadienone. When strongly electron-withdrawing substituents are situated on the aromatic ring of the N-chlor oaniline, the major reaction pathway became ring chlori nation. fhese observations can be rationalized in terms of a silver catalyzed reaction to form a “tight ion pair” which became

“tighter” as the ring substituents became more electron-withdrawing, i.e. anilenium ion stability decreased. 29 Aryl nitrenium ions have also teen implicated in the mechanistic process for the carcinogenic behavior of some aromatic amines and

37 5 38 amides. It was believed that organic compounds are carcinogenic

(37) For a review of this subject see: J. A. Miller, Cancer Research, $ 0 , 559 (1970).

(3 8) J. Do Schribner and H. K. Naimy, Cancer Research, Jjg., 1159 (1973)°

because they possess a latent electrophilic site that is converted enzymatically to an “ultimate carcinogen” that possesses an electro philic site. This electrophilic site then attacks the nucleophilic centers of nucleic acids in the cell. Such a carcinogenic amide is ft 2-acetylaminofluorene ( 6 5 ). It is converted in vivo to the N-hydroxy compound 66. Synthetic 66 proved to be much more carcinogenic than 6 5 .

o N-CCH3

However, since 6 6 did not react with proteins and nucleic acids, some further activation to the ultimate carcinogen must be occurring. This was subsequently shown t o be conversion o f the IT-hydroxy compound6 6 50 to its sulfuric acid ester &7, which provided the good leaving group needed for nucleophilic aromatic substitution. Biis was accomplished in the lab by incubation of soluble rat liver proteins with 66,

5’ -phosphoadenosine-5’ -phosphosulfate, and Mn++ or Mg-H-. Specifically, when 67 reacted in vitro with methionine (68), 1-thiomethoxy-2-acetylaminofluorene (69) was obtained in 65$ yield, ihe livers of rats have a high level of sulfotransferase activity and 69 was isolated from the livers of rats fed 66. As indicated in Chart I, Miller proposed a reaction involving a nucleophilic attack by sulfur on the nitrenium ion precursor in either an Sp2 or Spl manner.

§SL Chart I 51

R-S-CH;

0 0 !i •CCH3 R-S-CH 3 cch 3 .|-CH3 ^ “ F

cch 3 r - s - c h 3 o N t o H/ \S-CH 3 n R +

nh 2 ✓H o O Ac >S-CII3

2 —- 6=0 32

Ihe amides 70 , 71 , 72 , and 72 were also found to "be carcinogenic to varying degrees . 38 Simple molecular orbital calculations were done in an attempt to correlate the stability of the incipient nitrenium ion

12. li.

and carcinogenic activity. It was postulated that the more stable the nitrenium ion, i.e. less likely to undergo singlet to triplet inversion, the better carcinogen the compound.

In 1922 Pope and Mann 39 found that diethyl sulfide reacted with chloramine-T (7^) to afford the sulfilimine J'6 as a crystalline solid.

(39) F. G. Mann and W. J. Pope, J. Chem. Soc., 121, IO 52 (1922).

This compound was formed presumably via the azasulfonium salt 72 which lost hydrochloric acid under the work-up conditions. 53

S-N (C2H5)2S + ch 3

l i t

-HC1

T6 Appel and co-workers 40 found that azasulfonium salts 77 could he isolated and characterized before treatment with a strong base to form

(40) R. Appel, W. Buchner,-and E. 'Guth, Justus Liebigs Ann. Chem., 6 l 8, 53 (1958); R. Appel and W. Buchner, Angew. Chem., 71, 701 T3558), Chem. Ber., 95, 8U9, 855, 2220 (1952 J.

R2S + CIMHs 2^ R2S-MH2 r 2s c i 2 + m 3

nh 2"

R2SesMI

a sulfilimine. Ihese salts could either be prepared from the dialkyl sulfide and chloroamine or from chlorodialkylsulfonium chloride and ammonia. In most cases the yields of azasulfonium salts were low. 3h

Thus, several reasons were evident for a study of the chemistry of anilinodialkylsulfonium salts, azasulfonium salts ( 78 ). They could be useful as intermediates in nucleophilic aromatic substitution by exchange of X" to various nucliophiles. A detailed kinetic study of this nucleophilic substitution could lead to insights on the mechanism of carcinogenesis by aromatic amines and amides. These azasulfonium salts appeared to be accessible by reaction of the corresponding

N-chloroaniline and the appropriate sulfide.

R / "II X

78 RESULTS AND DISCUSSION

Part I. The Synthesis, Thermolysis, and Basic Rearrangement of

Azasulfonium Salts.

Since the chemistry of N-anyl-N-te rt-butyl nitrenium ions was well documented, 34: a study of the synthesis of such nitrenium ions

from azasulfonium salts was undertaken. N-tert -hutylanillnes are readily available from the reaction of aniline hydrochlorides and tert-

butyl alcohol34541 or hy nucleophilic displacement on the aryl fluoride hy te rt-butylamine.42 Thus, heating aniline hydrochloride (79) with

(4l) A. Bell and M. B. Knowles, U. S. Patent 2 , 692,287 (1956); Chem. A hstr., j?0, 2666e (1956).

(h-2) H. Suhr, Justus Liehigs Ann. Chem., 687 , 175 (1961).

150' NH2 + ( CH3)3coh HC1

80 22. te rt-hutyl alcohol at 150° in a steel homh gave after work-up 55$ of

N-tert-butylaniline ( 8 0). Reaction of jg-fluoronitrobenzene (8l) with te rt-hutylamine in dimethylsulfoxide produced p-(N-te rt-hutylamino)nitro-

henzene (82).

55 36

DMSO 'N-C(CH3 )3 + ( ch 3 )3c- nh 2 > H . N02 no2

81 82

These N-te rt-butylanilines, once in hand, could easily he chlori nated essentially quantitatively with solid calcium hypochlorite.34

These chlorinations were effected hy dissolving the aniline in a non polar solvent such as pentane or carhon tetrachloride and adding the powdered calcium hypochlorite in portions to these solutions. For example, N-te rt-butyl-p-toluidine (8q) was dissolved in pentane, cooled to -10°, and stirred with calcium hypochlorite for 1 hour. The salts were then removed by filtration and the pentane removed in vacuo to yield the N-chloro compound (84) as a dark oil.

N-C(CH3 )3 n - c( ch 3 )3 Ca(0Cl)2 H \ > Cl ch3 ch 3 84

The N-chloroanilines were used without further purification in the next step. The neat N-chloroanilines were maintained under a static nitrogen atmosphere and excess dimethyl sulfide was added with vig orous stirring. Depending upon the stability of the N-chlor oaniline 37 these reactions were sometimes initially cooled. Alter a few minute

induction period a white precipitate was formed;.this precipitate was found to he extremely hygroscopic and had to he maintained under an anhydrous atmosphere. Specifically, N-tert-hutyl-N-chloro-p-toluidine

(84) was cooled to -10° and dimethyl sulfide was added slowly with s tir ring, the cooling hath was removed after 10 minutes, and the reaction

stirred at room temperature for 4 hours; this procedure produced 85_ in

7&p yield. The filtrate consisted of mainly N-tert-hutyl-p-toluidine and a small amount of N-tert -hutyl-2-chloro-p-toluidine. In order to dry and prepare analytical samples of the salts synthesized, the azasulfonium salts were dissolved in chloroform, stirred over 4a molec ular sieves, filtered, concentrated and recrystallized from chloroform-ether.

N-C(CH3 )3 Cl

§5.

Tahle 2 lists the N-te rt-hutylanilinodimethylsulfcnium salts synthesized in this study.43 The formation of azasulfonium salts was

(43) For a preliminary report of this work see: P. G. Gassman, G. Gruetzmacher, and R. H. Smith, Tetrahedron Letters, 497 (1972). 38

TA.BDE 2

Yields of Azasulfonium Chlorides

Compound Y ields

N-te rt-butyl-p-toluidinodimethylsulfonium c h lo rid e ( 8 5) 7 &[o

N-tert -hutyl-4 -f luoroanilinodimethyl- sulfonium chloride ( 8 6 ) 83$

N-tert-Butyl-anilinodimethylsulfonium chloride ( 8 7 ) 8C$

R-tert-Butyl-k -chloroanilinodi - methylsulfonium chloride (88) 5 %

not limited to dimethyl sulfide; 89 could he synthesized from 8U and tetrahydrothiophene in 6 &fo y i e l d .44

(W-) Dr. R ichard H. Sm ith, Jr., performed some of the in itial experi ments on the chemistry of azasulfonium salts. This author thanks Dr. Smith for his assistance in this work.

n - c( ch 3 )3 S Cl / CH3 CH; Cl

8k 39 Attempts were made to synthesize 90 s and 91. In the case of

9£, under conditions analogous to those used in the synthesis of 88, the precipitate formed was t-(N-te rt-butylamlno)nitrobenzene hydro c h lo rid e as shown hy in fra re d (ir) and nuclear magnetic resonance (nmr) spectroscopy. The filtrate consisted of a solution of mainly ^-(R-tert- butylamino)-3-chloronitrobenzene and 4-(N-te rt-hutylamino)nitrobenzene.

S ~ CH3

Si.

The synthesis of 91 was attempted via the reaction N-te rt-butyl-p- anisidine (§2) and chlorodimethylsulfonium chloride.45 The only solid

(45) The basis for attempting the synthesis in this way w ill be discussed in Part III of Results and Discussion.

m aterial, however, that was isolated was triethylamine hydrochloride as indentified by ir and nmr. The filtrate when concentrated in vacuo produced only an intractable tar. The synthesis of 91 was also attempted via the chlorination of 92 with te rt-butyl hypochlorite in pentane at -10° followed by addition of dimethyl sulfide. Again no azasulfonium salt m s isolated. This failure may be due to the extremely rapid rate at which N-te rt-butyl-N-chloro-p-anisidine solvolyzes.34

It was also observed that the preparation of azasulfonium salts in solution gave lower yields of azasulfonium salt and higher yields of starting material and tar.

The ionic nature of these salts was borne out by the fact that the chloride anion could be exchanged for trifluoroacetate anion.

Thus, 9^3 and £5 were prepared in 76, 82, and 79$ yield, respectively, from the corresponding azasulfonium salts by dissolving the azasulfonium salt in dry methanol and adding one equivalent of silver trifluoro acetate in dry methanol.

n - c( ch 3)3 S-j-CHs / ch 3 CF3C02

§2.: X = CH3

<&: X = H

25.: x = c i

When an azasulfonium salt was thermolyzed in an aprotic, non- nucleophilic solvent such as dimethylformamide (DMF), we obtained thioanisidines from the reaction mixture. For example when 8_5 was heated in DMF at 100° for 2 hours, we obtained of 8^ and 58$ of 9 6 .

The reaction also produced methyl chloride which was identified by high resolution mass spectroscopy. in a similar manner 86 produced n -c (g h 3)3 n - c ( ch 3 )3 H

ch 3 sch 3 Cl

6% of 97, 20 % of 93, and %■ of 99* likewise 87 yielded of 8 0,

J>Ofo of 100, of 101, and % of 102, and 88 gave 5*$ of 102 and l8-/a of lOJo Thermolysis of N-te rt-bubyl-p-toluidinotetrame thy lenesulfonium

c ( ch 3 ); 7 c (ch 3 )3 c( ch 3 ); / n - c( ch 3 )3 n - h w-H

s - ch 3 sch3+ 1

22.

n - c( ch 3 )3

s —ch 3

ch 3 c i 80 100

Cl sch 3 101 102

^ n - c ( c h 3 )3 N-C(CH3 )3 42

c h lo rid e ( 8 9) produced bjfo of N-te rt-butyl-p- 1oluid ine ( 8 3) and 3Jp

of 4-chloro-n -butyl -3 -(4-tert -butylaminotolyl) sulfide (104).43

N-C(CH3 )3

The pyrolysis of the hydrochloride of 104 produced after a basic work-up 4-chloro-n-butyl- 3 -(^4--aminotolyl) sulfide (lOj?). An authentic

sample of 10£ was synthesized by the hydrolysis of 106 to y ie ld 107?465 47

(46) L. Horwitz and C. A. Clark, U. S. latent 3?102,l42 ( 1963 ); Chem. Abstr,, 60, lT60e (i960).

(47) M. T. Bogert and L. Smidth, J. Amer. Chem. Soc., 50? 428 ( 1928).

followed by reductive cleavage of 10J to 6-amino-m-toluenethiol hydro- 4 8 c h lo rid e 108, and coupling of 108 and l-bromo-4-chlorobutane by A. A. sodium methoxide to produce 105.

(48) M.. Schubert, Justus Liebigs Ann. Chem., 558, 10 (1947). k 3

H HC1 \ * n - c( ch 3 )3 m 2

ch 3 S-(CH2)4-C1 CHs s -( ch 2 )4- c i 10^

m 2 • h c i

CIi3' J O r : - C H 3 ^ s ^ A sh

106 io x 108

At first glance a very pleasing nitrenium ion mechanism may he written for the thermolysis of azasulfonium salts. Heterolytic cleavage of the N-S hond could produce a nitrenium i'on lCg_ and dimethyl sulfide.

,0 (0 1 1 3 ) c ( ch 3 )3 ,C(CH3 ); n - c( ch 3 )3 \ — > s— ch3

122.

+ CH3SCII3 + Cl"

H H C(CH3 )3 /

- > ✓ -S-OHs 122. X SCH ch 3 110 111

+ CH3 C1 As discussed, In the solvolysis of the N-chloroanilines, the positive 34 charge delocalizes into the ring. Nucleophilic attack on 109 by the unshared pair of electrons on the sulfur to produce 110, after rearomatization, could then occur. Nucleophilic displacement by chloride anion on one of the sulf onium methyl groups could yield the o-thioanisidine (ill) and methyl chloride. Ihe results are very reminiscent of the types of products obtained in the solvolysis of N-chloroanilines.

As previously noted, as the stability of the nitrenium ion decreased, the ease of singlet to triplet inversion increased; thus, the more electron withdrawing substituents promote the regeneration of starting aniline. Thus, solvolysis of Ii-tert-butyl-N-chioro-p-toluidine (84) in buffered ethanol produced 1 $ of 8^,^while Ii-te rt-butyl-IT,p-dichloro- aniline (112) yielded 1 fp of IT -te rt -butyl -p -chlor oaniline (102). As the stability of the nitrenium ion increased, yields of ring chlorinated

/ C(CH3 ) s /CCCII3)3

,c( ch 3 )3 ,c( ch 3 )3 / /

1 1 2 1 02 b 5

products increased. In the case of the thermolysis of azasulfonium

salts, since dimethyl sulfide is approximately 10 times more nucleo philic than chloride anion in methanol,4'8 it was not surprising that thioanisidines were obtained instead of chloroanilines.

(49) R. G. Pearson, H. S oh el, and J . Songstad, J . Amer. Chem. S o c ., 90, 319 (1968).

When an N-chloroaniline 84 substituted in the para-position with a strongly electron'donating substituent was treated with methanolic

silver trifluoroacetate, a 70 fo yield of 115 was obtained. Bnus, if the thermolysis of _85 were proceeding via a nitrenium ion, and if the 8 reaction was carried out in methanol, 1T5 should be produced. However, when 85_ was sealed in a tube with methanol and heated for 2 hours at

100° no thioanisidine or chloroanilines were detected by gas phase chromatography. If the reaction was allowed to go for 24 hours only very small amounts of thioanisidine were detected (ca. 1$), however the yield of 8^3 was 40^>. Ihe thermolysis of 85-88 in methanol gave

g(ch3 )3 c(ch3 )3 AgOgCCF 3 Cl ------c h 3o h ch3o

115. k 6 no evidence for the formation of 2,5-cyclohexadienone derivatives.

Thus, the first hint was provided to indicate that the thermolysis of azasulfonium salts did not proceed via the formation of a nitrenium ion.

The mechanism of the decomposition of trim ethyl- and tribenzyl -

sulfonium salts in 9C^ acetone-10$ water was studied "by Swain and 5 0 Kaiser. Earlier workers had reported the decomposition to he S-^l;

(5°) C. G. Swain and L. E. Kaiser, J. Amer. Chem. Soc., 80, h089 (1958) and references therein.

however, Swain and Kaiser found that the rate determining step involved a reaction of the sulfonium cation with the anion present. It was found that trimethylsulfonium perchlorate does not react at 100°, but when lithium chloride was added rapid consumption of chloride was observed with good second-order kinetics. If trimethylsulfonium chloride was decomposed pseudo-first-order kinetics were observed. Hie authors attributed this observation to exactly compensating salt effects. These observations were substantiated by similar observations by Hughes, 51 Ingold, and Pocker.

(5l) E, D. Hughes, C. K. IngaLd, and Y. Pocker, Chem. Ind. (London), 1282 (1959).

Ihe reactions of trimethylsulf onium cation with anions X“ (X = OEt,

CN, I, SCH, N3, Cl, and F) to form dimethyl sulfide and CH^X in ethanol and methanol at 100° were shown to be S^2 by Pocker and Parker . 52

(52) Y. Pocker and A. J. Parker, J. Qrg. Chem., j51, 1526 (1966).

Hydrolysis and methanolysis of the sulfonium cation can also proceed by an Spj2 mechanism, and methanol and ethanol have been shown to be more nucleophilic toward the sulfonium cation than water. The kinetic rate law derived for the formation of dimethyl sulfide and methyl bromide m s studied in great detail, but cnly to 2C$> completion because of complications arising from the reverse reaction. Nevertheless, it was ascertained that mechanisms involving separated ion pairs are indis tinguishable in terms of kinetic rate law from the ones involving intimate ion pairs.

In order to lend some insight into the mechanism of the decompo sition of azasulfonium salts, the rates of decomposition of 89-88 were measured in dry DMF at 80° in the presence of lithium chloride. Table 5 lists the second order rate constants for the reaction of azasulfonium cation with chloride anion, the square root of the initial ionic strength of the solution, and the logarithm of the rate constant at the square root of the ionic strength equal to O.V555* 53 4 *54: The activity-rate theory, proposed by Bronsted and Bjerrum, predicts that in reactions between oppositely charged ions, the rate of reaction should be slower, the greater the ionic strength of the

solution, and that a plot of log k vs the square root of ionic strength, TA.BLE 3

Rates of Thermolysis of 4-Substituted-II-te rt-butylanilinodimethylsulf onium Chlorides in the Presence of Lithium Chloride in Dimethylformamide

Compound 4-Substituent k(^ m ole "1 sec 1) log k at /l= 0.^355 a 1 1 6 .54x10'~3 0.3083 5. 36 x10"3 0.3246 85 CHS 3.37x10_s 0.4428 2 . 08x10"3 0.5127 -2.49

8.67x10“3 0.3000 86 F 4.56x10“3 0.3931 3 .37x10 "3 0.4762 -2.40 e a

6 . 87 x10“3 0.3496 87 H 4.73x10 ~3 0.4132 3 =44x10“3 0.4709 -2.38

2.49x10"2 0„ 4070 88 Cl 1.36x10“2 0.4672 1.04x10 "2 0.5346 -1.72

Values obtained by interpolation of a least squares fit.

CO (53) J* H. Bronsted, Z. Physik. Chem., 102, 169 (1922); N. Bierrum, ib id . , 108, 82 (1924").

(54) For general discussions of ionic reactions see: W. J. Moore, ^Physical Chemistry,” jvd. ed., Prentice-Hall, Inc., Englewood Cliffs, H. J ., 1962, Chapter 9; K. J. Laidler, “Chemical K inetics,” 2nd ed., McGraw-Hill, Hew York, H. Y., 1965? Chapter 5.

/I, should give a straight line. Brief perusal of Table 3 shows that indeed the rate of thermolysis of 85-88 decreased as the ionic strength increased. The compounds listed gave fair second order kinetic plots

(average correlation coefficient ca. O. 9 9 5) for Ig- to 2 half-lives.

After this time deviations from linearity were observed. Figure 1 1 + 55 shows a plot of log k at /I = 0.4355 vs. o . The points for compounds

(55) H. C» Brown and Y. Ckamoto, J. Amer. Chem. Soc., 80, 4-979 (1958).

8^ , 86 , and 87 corresponding to the g- Me, F, and H compounds, fa ll on a straight line giving a p of + 0 .3 7 with a correlation coefficient of

0.998; however 88, the point for the p- Cl compound, lies much above this line. It was interesting to note that this compound was very hygroscopic and a satisfactory elemental analysis could not be obtained for it. A plot of log k at /i = 0.4355 vs. a was also made^correlation coefficient of 0.910. Figure 2 shows this plot. In this plot the correlation between log k and the substituent constant, cr, was not very good for compounds &?, 8 6 , and 8 7 ; however all four points corre late much better than in the a+ plot in Figure 1, correlation coefficient log k iue 1. Figure ot f r vs lg k. log s. v cr+ of t lo P - 0.3 - 0.2 - CT 0.1 +

0.0

0.1 0.2 50 log k - 1.7 -1 - 1.9 -1 - - - 2.3 -2 2.4 -2 - 2.5 -2 iue . ot f vs lg k. log s. v a of t lo P 2. Figure 1.6 1.8 2.1 2.0 2.2 2.6 - 0.2 - 0.1 CT 0.0 0.1 0.2 for all four points of 0. 76 O. I t may be e n tir e ly fo rtu ito u s th a t the points corresponding to compounds 8£, 86, and 8 j correlate so veil in Figure 1.

Thus, it appeared that the thermolysis of azasulfonium salts in

DMF was a second order process, first order in azasulfonium cation and first order in chloride anion, dhis was the first step of the reaction, and it could not be ascertained from the data if this was the rate determining step of the reaction. It was found that the azasulfonium salts substituted cn the aromatic ring with electron-donating groups reacted faster than those substituted with electron-withdrawing groups.

This observation could be rationalized by an inductive effect of the electron-donating groups which decreased the positive charge on the sulfur making it less electrophilic.

A possible rationale is the intermediacy of a sulfurane (llA ),50 a tetracoordinate sulfur compound; this is attractive in that

(56) Dialkoxydialkylsulfuranes have recently been synthesized and studied, R. J. Arhart and J. C. Martin, J. Amer. Chem. Soc., 9k, *+997j 5003 (1972) and references therein.

C(CH3 )3 55 decomposition of lit can give an amide anion 11£ and chlorosulf onium. cation (li£>), an electrophilic reagent. The negative charge could delocalize onto the ring and electrophilic attack “by 116 on the ring

c( ch 3)3 9 9 1 N -

l i t « * X

/ k ch3 ch3 116

CH3C I +

X SCH, X

111

would produce 1U. Credence is lent to this mechanism by the observa tions of Wiegand and McEwen 57 who pyrolyzed phenyl-p-tolyl-2, 5 -dim ethyl-

(57) G. H. Wiegand and W. E. McEwen, J. Qrg. Chem., 26J1 ( 1968 ).

phenyl sulfonium chloride 117 a t 250°, and obtained 119, 120, and 121 in CH. Br 54 CH. Br

CH;

i i l 118

s k

+ D J * £ >

120 121 the ratio 1.6j: 1.00: 5*00. They rationalized their results "by proposing a tetravalent sulfur compound 118 which could decompose preferentially

in a manner to relieve steric strain. Oae and Kim58 found that tritolyl-

sulfonium bromide in ether, when treated with phenyl lithium underwent

(58) Y. H. Kim and S. Oae, Bull. Chem. Soc. Jap., 42, 1968 (1969).

nucleophilic attack on the sulfur about JOjfo of the time; an additional

20/j of the products arose from a benzyne type of reaction. Trost and co-workers59 found that the reaction of triphenylsulfonium fluoroborate

(59) B. M. Trost, R. LaRochelle, and R. C. Atkins, J. Amer. Chem. Soc., 91, 2175 (1969). 55 (122) with vinyl lithium generated diphenyl sulfide (12,3) and styrene

(12U) quantitatively * They suggested that a sulfurane was an inter mediate in this reaction; however, this could also he an aromatic nucleophilic displaeement.

L i / CeHs

(c6H5)3s + bf4~ ------(C6H5 )2S +

122 122, 124

Spectroscopic evidence for the intermediacy of a sulfurane was obtain ed by Helmkamp and co -w o rk ers.60 When e q u iv a le n t amounts of

(60) Do C. Owsley, G. K. Helmkamp, and M. F. Rettig, J. Airier. Chem. Soc., 91, 5239 (1969).

cyclooctene-S-methylepisulfonium 2,^,6 -trinitrobenzenesulfonate (12^) and tetraphenylarsonium chloride were mixed at roam temperature in an nmr tube, the nmr, when recorded immediately, showed singlets at 122 and

128 Hz downfield from QMS. After 15 minutes the peak at 122 Hz had disappeared. This peak at 122 Hz was attributed to the methyl group of the sulfurane 126. This sulfurane decomposed to 127, which had a

TUBS 56 singlet in its nmr spectrum at 128 Hz. At no time during the reaction period could a singnal at 158 Hz be detected for the methyl group of

125 or at 171 Hz for me thane sulfenyl chloride. Furthermore, no signal could be detected in the olefinic region of the spectrum, a fact which indicates the absence of cyclooctene. Simple molecular orbital calcu lations suggested that the most stable form of the sulfurane was a square pyramidal structure. The mechanism by which the trans stereo chemistry of 127 arises is not known.

Thus, there are some grounds for postulating an azasulfurane as an intermediate in the thermolysis of azasulfcnium salts. Ihe azasul furane llJ7 can form by nucleophilic attack on the positive sulfur; to relieve steric strain it could decompose to form and 116. The lack of formation of thioanisidines in methanol can be rationalized by abstraction of a proton by 115 from the solvent to form the parent a n ilin e .

An attempt to prove this mechanism was made by studying the reaction of the sodium salt of N-te rt-butyl-p-toluidine with chlorodimethylsulfonium chloride (128 ).~5 Since this reaction should have 115 a-nd 116

1 2 8 57 present in the reaction mixture, if they are present in the decompo s itio n of 85, then 96 should he formed. Chlorodimethylsulfonium chloride is stable only at lew temperatures (- 78 °). The reaction was carried out by adding a solution of 128 in methylene chloride at - 78 ° to a solution of the amide anion in DMF at -40°, warming the reaction to room temperature, and heating the reaction mixture at 100° for 2 hours. Analysis of the reaction mixture by gas phase chromatography showed that only yjo of N-te rt-butyl-4 -methyl-o-thioanisidine (95) was formed. A control experiment was done in which 8j was reacted with

128, in this instance 2$> of J95 was formed. Ihese results are incon clusive and neither prove or disprove the sulfurane-amide anion mechanism.

c( ch 3 )3 / N-C(CH3 )3

Another possible mechanism involves the intermediary of a sulfenamide (129) which could arise through nucleophilic attach upon one

^ ( 013)3

- c( ch 3 )3 N \ sch 3 + ch 3c i CH s— ch3 CHa^ / + ch 3 C l

8^ 129 of the methyl groups of the azasulfonium salt "by chloride anion to produce methyl chloride; then 129 can rearrange to give j§ 6 . I t had teen reported ty Moore and Johnson 61 that 2-nitrobenzenesulfenanilide

(6 l ) M. L. Moore and T. B. Johnson, J . Amer. Chem. Soc., 57 j 1517 (1935); i t i d . , 57 , 223b (1 9 3 5 ) ;~TbTd., 1091 (1936); i t i d . , .£85 1960H936).

1^0 when heated in aniline at 160 gave, in 70 $ yield, 7' -amino-2-nitro- diphenylsulfide (jh3l)» More recently this reaction has teen reinvesti gated ty David, et. al. ,g2 who found only 12$ of 1^1 formed from 130,

(62) F. A. Davis, JR. ,B. Wetzel, T. J. Devon, and J. F. Stackhouse, J . Org. Chem., 356 , 799 (1971).

S-N

131 however, they found 5$ of 132, % of 133. and 373 of 1^7. With the nitro group in the 3 position, only 136 and 137 were produced from 1^5 in 22 and 60$ yieldj respectively. Ihe mechanism of these rearrange ments of sulfenanilides to produce diphenyl sulfides is not known. In view of the results ty David et. a l., it was believed that perhaps a H I o KHc

sulfenamide was an intermediate in the thermolysis of azasulfonium salts.

®ius, 1^8 was prepared by the reaction of 85 with methane sulfenyl chloride63 and triethylamine. However, when 1^8 was heated for 2 hours

(6 3 ) I. B. Douglas, J. Qrg. Chem., 2h} 200h (1959)•

CCCHa); c ( ch 3^33) / / N \ sch 3 d — > CH3 CH; SCH3

9k 60 in DMF, no trace of 96 was detectable by gas phase chromatography.

When azasulfonium salt 8^ was heated for 2 hours In DMSO a kzfi y ie ld of 96 was obtained. Heating of lg8 far 2 hours in DMS0-ds, adding a drop of MS, and taking an nmr spectrum showed an aromatic region identical with that obtained for N-tert -butyl-p -toluidine (&5). Hius, it appears that 96 is not arising via a sulfenamide.

Although the experiments described are hardly definitive, several tentative conclusions may be drawn. No products were obtained from the thermolysis of 85_-<38 in methanol which would lead one to believe that a nitrenium ion was present, i.e. no anisidines or 2,5-cyclohexadienone derivatives were observed. In these thermolyses the reaction in DMF was faster for the salts substituted on the aromatic ring with electron- withdrawing groups; in contrast, the N-chloroanilines substituted with electron-withdrawing groups on the aromatic ring solvolyzed in methanol much more slowly than did N-chloroanilines substituted on the aromatic ring with electron-donating substituents. The intermediacy of a sul fenamide is ruled out because its thermal decomposition in DMF gave no o-thioanisidine. Hence, at this time, the most reasonable mechanism is via a sulfurane, which decomposes to an amide anion and chlorosulfonium c a tio n .

As an offshoot of this study, a very useful reaction of azasul fonium salts was discovered. Treatment of with aqueous sodium hydroxide resulted in the formation of N-tert -butyl-2-(thiomethoxymethyl)aniline (l^l)» In this case, the base removed one of the protons 61

on a methyl group "bonded to the sulfur to give the ylid, 1/39, which

rearranged to give ibO. Proton shift and accompanying rearomatization

transformed lt-0 into itl. This rearrangement was analogous to the

C(CH3 )3

a-C H s S—ch 3

ch 3 cf3co 2 m

c( ch3):

c ( ch 3 )3 / / N f r n ' h L — ch 2sch 3

ch 2sch 3 H

I k l i h o

Sommelet-Hauser rearrangement of ammonium64 and sulfonium65 ylids.

(6^4-) . M. Sommelet, C. R. Acad. S ci., 205, 56 (193?); G. C. Jones and C. R. Hauser, J. Qrg. Chem., 27, 3572 (1962); G. C. Jones, W. Q,. Beard, and C. R. Hauser, ihid., 28, 119 (1963)- For a review of the Sommelet-Hauser rearrangement see: H. J. Shine, “Aromatic Rearrangements,” Elsevier Put. Co., New York, N. Y., 19675 pp 3 1 6 - 3 2 6 .

(6 5 ) C. R. H auser, W. W. K antor, and W. R. Brasen, J . Amer. Chem. S o c ., .7£, 2660 (1953).

Sommelet found that heating benzhydryltrimethylammonium hydroxide produced o-benzyl-benzyldimethylamine. This product arose via a similar 62 type of rearrangement to that which produced l4l. Benzyldimethyl- sulfonium bromide rearranges to 2-methylbenzyl methyl sulfide with sodamide in liquid ammonia by an analogous mechanism.

Thus, it was discovered that azasulfonium salts S£-_8S and 94 can be rearranged in methanol by sodium methoxide. Table 4 lists the yields of 2-thiomethoxymethylanilines obtained in these rearrangements. This rearrangement was not confined to anilinodimethylsulfonium salts.

TABLE 4

Base Rearrangement of Azasulfonium Salts to 2-Baiomethoxymethylanilines

c( gh 3 )3

X Compound Anion $ Yield of 2-Thiomethoxymethylaniline

ch 3 C l" 92 Q £ 2 ) F 86 C l" 88 Q&) H 87 C l" 95

C l 88 C l" 90 ( i ^ t ) H 94 CF3C02“ 85 ( l4 l) N-tert -butyl -p-toluidinotetramethylenesulf onium chloride (8£) was

rearranged to give ^6fo of 3^ and of lh^.

,C(CH3); c( ch 3 )3 c ( ch 3 )3 / / / Cl‘ N \ + CH* H o CHS ■ ® s §2,

/ C(CH3 ); /,c( ch 3 )' H

1^7

Compounds 1^1 and lk2 were reduced to o-toluidines lh6 and 1^7 in

69 and 76p yield respectively by W-2 Raney nickel.66 Thus, this

(66) R. Mozingo, Org. Synthesis, Coll. Vol. l8l (1955)*

procedure of forming the N-chloroaniline, reaction of the N-cbloro-

aniline with a sulfide to form an azasulfonium salt, treatment with base, followed by Raney nickel reduction could be a useful method for

alkylating anilines in the ortho-position. Part II. The ortho-Alkylation of Anilines

Of the various methods available for the formation of carbon- carbon bonds to aromatic rings, the classical Friedel-Crafts reaction is undoubtedly the best known and the most widely used.67 In the

(67) C. Friedel and J. M. Crafts, C. R. Acad. Sci., 84, 14-50 ( 1877 ). For a detailed account of the Friedel-Crafts reaction see G. A. Olah, ‘‘Friedel-Crafts and Related Reactions,” Interscience, New York, N. Y., 1963 -1965 .

Friedel-Crafts alkylation an alkyl halide, olefin, or alcohol is treated with an aromatic compound in the presence of a Lewis acid catalyst.

Thus, in general 148 reacts with an alkyl chloride in the presence of aluminum chloride to produce an alkyl benzene (1^9). Friedel-Crafts

+ RC1 + AICI3 ------;------>

]J+8 14-9

alkylation introduces a group that activates the ring; thus, di- and polyalkylations are frequently observed. An additional lim itation of the Friedel-Crafts alkylation procedure is that phenols, anisoles, and anilines do not facilitate the reaction; the Lewis acid catalyst coordinates with these basic groups.

64- 65

Aromatic amines can undergo alkylation if olefins are used as reagents and aluminum anilides as catalysts. 6sajtl in this process,

(68) (a) For a review see: R. Stroh, J. Ebersbergei’, H. Haberland, and W. Hahn in “Newer Methods of Preparative Organic Chemistry, ” Vol. 2, W. Foerst, Ed., Academic Press Inc., New York, N. Y., 1963j pp 227-252. This article also appeared in Angew. Chem., 69, 669 (1957); (b) G. G. Ecke, J. P. Napolitano, A. H. Filbey, and A. J. Kolka, J. Org. Chem., 22, 639 (1957)*

aluminum in the form of dust, powder, or shavings was dissolved in aniline (150) on warming with the evolution of hydrogen. Ethylene was taken up rapidly when forced into this solution at 200 atmospheres, and the temperature increased to 350°• fhe reaction was complete after two moles of ethylene were taken up, and 2,6-diethylaniline (l£l) was produced in 9^ yield. Several observations can be made on this reaction.

Die olefin entered the aromatic ring as an alkyl group, principally in the position ortho to the amino group. The reactivity of the olefins decreased in the order ethylene > propylene > butylene, isobutylene.

Die olefin became attached to the aromatic ring at the olefinic carbon 66 possessing the larger number of alhyl groups. N~Alkylanilines are more reactive than the corresponding primary anilines, although only one group can he introduced into the ring. Tahle 5 shows the starting amine, product and yield obtained in some of these alkylations. 00<3«

TABLE 5

Alkylation of Anilines via Aluminum Anilides

S ta rtin g Amine Product $ Y ield

Aniline 2,6-diethylaniline 96 Aniline 2-ethylaniline ^9 o-chloroaniline 2-chloro-6-ethylaniline 85 Aniline 2-isopropylaniline 15 p-toluidine 2-isopropyl-g-toluidine 52 Aniline 2 -ter t -butylaniline 65 N-methylaniline 2 -ethyl -Ii -me thylaniline 85

The mechanism of this reaction was believed to involve a-cyclic intermediate in which the jt electron pair of the olefin, in the case

shown ethylene (i^2)5 interacted with the electron deficient aluminum of the aluminum trianilide (152.) which had been formed. Reorganization of the electrons resulted in the alkylation at the ortho position. The aluminum trianilide was regenerated by reaction with another mole of aniline and rearomatization produced 15^* This method was the only my

in which aromatic amines could be alkylated directly, albeit under

somewhat strenuous conditions. O m CH2CH2-A l(m 0)2 N N v a / k ^ CH2 h H— N ) ^ !]_ Cl I; /O'

0M 2 i s .

m 2

+ iS L ch 2ch 3

A 2t

Ihe usual procedure used to produce ortho-alkylanilines is to nitrate the appropriate compound, separate isomers and reduce the nitro functionality to the amine. Particularly, ethylbenzene was nitrated, the isomers separated, reduced, and nitrated again to produce'

2 -ethyl-5 -nitr oaniline. 6 9

(6 9 ) H. S a n ie le v ic i, L. F lo ru , and M» Cibu, Rev. Chim. (B u ch arest), 673 (1963)> Chem. Abstr., 6l, 16195b (19&4).

In great profundity the patent literature lists derivatives of ortho-alkylanilines as dyes,69 photoconductive resins,70 photosensi tive components of color film ,'’1 cyan couplers in photographic film ,72 accelerators for polyepoxides,73 gasoline additives to decrease carbu- rator icing,7^ analgesics,75 and inhibitors of steel corrosion.76 68

(70) A.-G. K ane, British Patent 977,399 (196*0; Chem. Ahstr., 62, 8570d (1965).

(71) E. Wainer, U. S. P atents 3,0*4-2,515 and 3,0*42,516; Chem. A b str ., _58, I2100e (1963).

(72) M. Iwama, K. Sakamoto, I. Inoue, Y. Takei, and T. Endo, German Patent 1,959,59*4 (1971); Chem. A bstr., 7£, 12*4989x (1971). P. Ramello, German Patent 2,028,601 (1970;; Chem. Abstr., 75, 22505 (1971).

(73) M. T. Harvey, U. S. Patent 3,207,813 ( 1967 ); Chem. A b str ., 6 7 , 12219z (1967).

(7*0 E. L. Fareri. and J.. P. Pellegrini, Jr., U. S. Patent 3 , 085,867 (1963); Chem. Abstr., 59, 2570f ( 1963 ).

(T5) G. A. Starmer, S. McLean, and J. Thomas, Toxicol. Appl. Pharmacol., 20 ( 1971 ).

(76 ) R. L. Every and R. L. R iggs, J r . , M ater. P r o te c t., J5, *46 (196 *4).

A particularly interesting compound is 21,6' -diethyl-N-methoxymethyl-2- chloroacetanilide (l££')• It was claimed to be effective as a fungicide,

weed killer, insecticide, nematocide, algicide, and bactericide. TT

(77) Monsanto Co., Netherland Appl 6,602,56*4 (1967 ); Chem. Abstr., 61, 99832r (1967). . 6 9

Undoubtedly, the use of is related to the ease with which 1£1 can be prepared.

Clearly, it appeared that a simple way to make ortho-alkylanilines was needed. A route to such ortho-alkylated anilines appeared to be available in the reaction of aniline (l£ 0 ) with a chlorinating agent to produce N-chloroaniline 0 l 5 6 )s addition of a sulfide, such as dimethyl sulfide, to form the azasulfonium salt (l^T); followed by base rearrangement to give 2-(thiomethoxymethyl) aniline (160). Raney nickel

Cl H / N. / •2 \ CH3SCH3 N H CH3

ch 3 Cl

WaOCH3 CHr

- \ H CH3 H / f /

CH2 SCH3 M A -> T ch 3 / N =S CHs

W-2 Raney Ni RH2

160 161 70

reduction of 160 would produce o-toluidine 6 (ll). Two major

difficulties were anticipated in this work. Would a N-chloroaniline

"be stable enough to react with the sulfide and would the sulfilim ine

138 be formed in this reaction? It was also known that 162 did not

undergo a Somme let-Hauser rearrangement presumably because the cyano

group stabilizes the ylid 1 &5 . 7 8

( 7 8 ) W. A. Beard, Jr. , and C. R. Hauser, J. Org. Chem., _2£, 33k (i9 6 0 ).

+ ch - n ( ch 3 )3

m e ■'

1 6 2 165,

In fact, aniline (l£0) could be converted into 160 in 90?o y ie ld .

Specifically, a solution of aniline was cooled to -78°, one equivalent

of tert-butyl hypochlorite was added at 7 - 8 ° from a jacketed constant

pressure addition funnel. Ihis produced a light yellow solution of

N-chloroaniline. About three equivalents of dimethyl sulfide were added

to the solution. Hie reaction mixture was stirred for. one hour after

which sodium methoxide in methanol was added to effect the base rear

rangement. The reaction was carried out with various substituents on

the aromatic ring; Table 6 summarizes these results . 7 9 Ihe reaction ft 71

TABLE 6

Conversions of Anilines into 2-Ihiomethoxymethyianilines

e 5

4 .(7)T X X \ ch 2sch 3 / Q X

Starting Material Products $ Y ield3, $> Conver

Aniline (1^0) x==H 90 82 p-T oluidine (l6jX) 165., x=4-CH3 9b lb m-Toluidine (l66) i&x, x=3-CH3 , + 1683 x==5-CFfl3 86 58 o-Toluidine (l6l) 162 , x==6 -CH3 7^ 85 p-Chloroaniline ( 170 ) m , x=4 -Cl 83 75 m-Chloroaniline ( 172 ) x=3-Cl, + l £ b x==5 ~ci 87 b9 o-Chloroaniline (1X5.) 1x6 , x==6 -Cl 70 37 Benzocaine (iJX) i l 8, x=lf-C0sC2H2 65 c o-Aminobiphenyl (.179) 180^ x==6 -CeH5 b2 86 p-Nitroaniline (iSlJ 182 5 x=4 - no 2 71 75 m-Anisidine (X§5.) lS ^ , x =3-0CH3 , + x=:5-och 3 55 87 p-Anisidine (186) 1&7 x==4-0CH3 3 97 a Based on unrecovered starting material. Percent of unrecovered aniline. cNo attempt was made to recover starting material in this case.

(79) For a preliminary report of this work see P. G. Gassman and G. Gruetzmacher, J. Amer. Chem. Soc., 588 (1975).

could he performed with little difficulty on a 0.5 mole scale. In this case, l60 was produced from 1^0_ in 85/a yield.

In cases where the ring contained electron withdrawing sub stituents, sulfilimines such as 158 were formed and were stable inter mediates. In those cases, the sulfilimines were refluxed with triethylamine in either toluene or acetonitrile. Claus and co-workers 80 found

(80) (a) P. Claus and W. Vycudilik, Tetrahedron Letters, 360J (1968); (b) P. Claus and W. Vycudilik, Monatsh. Chem., 101, 396, 2+05 (1970; (c) P. Claus, W. Vycudilik, and W. Rieder, ibid., 102, 1571 (1971) (d) P. Claus and W. R ied er, ib id . , 10^., I I 63 (1972JT'”

sulfilimines to be intermediates in the reaction of anilines with dimethylsulfoxide in the presence of phosphorous pentoxide and triethylamine. These workers also found that upon heating, these compounds rearranged via a Sommelet -Hauser type reaction to 2-thiamethoxymethyl- anilines. In this case, the aniline attacks the sulfur of the

DMSO-P4O3.0 complex to form an intermediate such as 188. This was in contrast to the reaction of N-chloroaniline with dimethyl sulfide, in which the sulfide was probably the nucleophilic reagent.

188

-P40s (0H)2 73 In all cases studied, varying amounts of starting aniline were

obtained. This was probably because of simple hydrolysis of unre- sod rearranged sulfilimine upon work-up. Claus and Reider found that

rearrangements of stable sulfilimines such as lS9 in ethanol at 5°°

produced p-nitroaniline (lSl) and ^-nitro-2-(thiomethoxymethyl)aniline

( 187 ) in the ratio 7 * 1: 1*

c h 3

N- ■2 C2H sOH CH, + NO; 181 i§2.

When meta-substituted anilines were used, a mixture of 3 - and

5-substituted-2-(thiomethoxymethyl)anilines was obtained. It appeared

that as the electron-donating ability of the substituent increased,

th e more 5 -substituted compound was obtained. There could also be a

steric effect due to the methoxy and methyl groups blocking the

ortho-position. The ratios of products were determined by nmr.

Table 7 summarizes these results.

Perusal of Table 6 shows that p-anisidine gave only a 3$ yield of

4-methoxy-2-thiomethoxymethylaniline (l3j)• This is attributed to the

fact that electron donating groups facilitate the solvolysis of

N-chloroanilines. Thus, it appeared that once N-chloro-p-anisidine

formed, it immediately solvolyzed. Thus, strongly electron-donating l h

TABLE T

Ratios of 3- and 5-Substituted-2-(Thiomethoxymethyl)anilines from Meta-Substituted Anilines

3 -Substituted -2 - (Ttiio - 5 -Substituted -2 - (Thio ■ Meta-Substituent methoxymethyl) aniline methoxymethyl)aniline

OCH3 (^) 1 (l§ii) 3 QJ£)

CH3 (166) 1 (167) ‘ 1 (1§8)

c i (.172 ) 3 ( m ) 2 ( l£ i)

groups cannot be present in the ortho or para ^position on the aromatic ring for this alkylation procedure to work in high yield.

The reaction with N-chloroaniline (JL56) with simple dialkyl sulfides was attempted. Reaction of JL 36 with diethyl, di-n-propyl, di-isopropyl, or di-n-butyl sulfide resulted in the recovery of aniline

1§0 and tar. These results can be rationalized by formation of the sulfilixnine 1§0 ; the anion on the nitrogen could abstract a p proton from one of the alkyl groups in a five membered transition state to give sulfenamide Igl and olefin. The sulfenamide then decomposed to

H / H— C— R‘ H n * vi n + CH2~ CHR s S-R / + R I 9O aniline. This sulfenamide could be isolated when the R group was phenyl, The reaction of N-chloroaniline with n-butyl phenyl sulfide in the usual manner produced henzensulfenanilide (192) in 70 $ y ie ld .

In a similar manner 192 was produced in 76$ yield using phenyl isopropyl sulfide.

When allyl methyl sulfide was used, again no ring alkylation occurred; a 7^$ yield of N-ally lan iline (125.) was obtained. In this case, the sulfilimine formed, 125.3 attacked the double band in a

Michael fashion to produce Igh which decomposed to produce 525-

H H I C ch 2 CH / N I! CH CH2

CH3 195 19^

If the p-carbon atoms are “tied back,” as in a cyclic sulfide, so that abstraction of a p proton is precluded, then alkylation of the ring can occur. Hence, 12.6 was prepared in 31$, 6kf> yield based on unrecovered starting material, from N-chloroaniline and tetrahydrothiophene. Similarly, 121 was prepared in yield based on unrecovered starting m aterial, from tetrahydrothiopyran and N-chloro

aniline. Attempts to use trimethylene sulfide and thiirane to produce

lg8 and 1§£5 yielded only aniline and tar, perhaps because of the

instability of the azasulfonium salt.

The alkylation of N-alkylanilines should be feasible using simple

dialkyl sulfides, since the azasulfcaaium salt formed would not have a

0 proton on the nitrogen. Thus, reaction of N-chloro-N-methylaniline 200 with dimethyl, diethyl, and di-n-propyl sulfide produced 201, 202 and

202. in 59? ^5j anh 2 k°[o yield respectively. CH;

201

CH3

N. H Cl CHSCH2CH3 202 I ch3 / CHa ■n : H

CHSCH2CH2CII3 205 CH2CH3 Reduction of the sulfides produced by this procedure using

¥-2 Raney nickel66 yielded ortho-alkylanilines in good to excellent yields. Table 8 summarizes the yields obtained in these reductions.

TABLE 8

Reduction of 2-Thioalkoxyalkylanilihes to Yield ortho-Alkylanilines

Product % yield Based on Sulfide

o-toluidine (l 6 l ) 63 2 , 6 -xylidine (204) 66 4-chloro-o-toluidine ( 205.) 72 a 2 -methyl-p-anisidine (25o) 50 4-carboethoxy-2-methylaniline ( 207 ) 88 2 -n-butylaniline ( 208) 62 2 -n-pentylaniline ( 209) 68 2 ?n-dimethylaniline~C 21;0 ) 72 2-ethyl-N-methylaniline (211) 69

St 1C$> o-toluidine (l 6 l) was also obtained by concurrent reduction of the chlorine.

The process described for this alkylation procedure occurs via a cyclic mechanism in which little , if any charge develops in the

aromatic ring. Thus, the reaction works equally well in the presence

of electron-donating or electron-withdrawing substituents. This

provides a distinct advantage over the classical Priedel-Crafts reaction.

The process was limited in that primary anilines cannot be alkylated with dialkyl sulfides except dimethyl sulfide. However, cyclic sulfides

such as tetrahydrothiophene and tetrahydropyran can be used to alkylate 78 primary anilines. N-alhylan ilines can "be alkylated with dialkyl sulfides since the azasulfcmium salt formed does, not have a proton on the nitrogen which could be abstracted to form a sulfilimine.

However, the process described is a simple method for aromatic sub stitution which w ill prove useful in the synthesis of a veritable plethora of aromatic compounds.

Several examples of logical extensions of this procedure have 81 appeared in the literature. Gassman and Dr ewes have prepared diphenylmethanes (212) via 213 and 2l4 using benzyl phenyl sulfide.

(8l) P. G. Gassman and H. R. Drewes, Chem. Comm., 488 (1975).

•m2 NH; -» X chsc 6 h5 X ch2 cs h 5

215 214

212

The alkylation of aminopyridines m s accomplished in a manner analogous 8 2 to that previously described for the alkylation of anilines. Thus,

(82) P. G. Gassman and C. T. Huang, J . Amer. Chem. S o c ., 95» 4453 (1975). 79

2 -aminopyridine (2 1 2 ) was converted to 2 -amino - 3 -thiomethoxymethyl - pyridine (216) in 7C$ yield; Raney nickel reduction gave 2T£ in

73$ y ie ld .

Ihe use of a substituted sulfide, such as methylthio-2 ~propanone instead of a dialkyl sulfide, in the reaction with N-chloroaniline

(i32.) resulted in the formation of 2 -methylindole (218) . 3 3 This procedure could he modified to produce indoles2 1 £ from methylthio- acetaldehyde and l^i* * Oxindoles 220 could he prepared analogously from 13p' and ethyl methylthioacetate. 8 5 All of these indoles and

P. G. Gassman and T. J. van Bergen, J. Amer. Chem. S o c ., 95j 590 (1975). m P. G. Gassman and T. J. van Bergen, i b i d . , 95., 591 (1973).

(85) P. G. Gassman and T.' J. van Bergen, ibid., 23, 2718 (1973 ).

and oxindoles could he reduced with Raney nickel. The synthesis of indoles and oxindoles hy this method provides a simple route to a hroad category of substituted indoles and oxindoles. 80

H / N 0 SCH' \ I' Cl + ch 3 sch 2 cch 3

H 156 218

^SCH3 SCH,

QO Part III. The Reaction of Anilines with Halodimethylsulfonium Halides

In 1965 Moffatt and Burdon 8 6 reported the acid-catalyzed reaction of phenols with dimethyl sulfoxide (DMSO) and dicyclohexylcarhodiimide

( 8 6 ) M. G. Burdon and J. G. Moffatt, J. Amer. Chem. Soc.,8 £ , 4656 ( 1 9 6 5 ); M. G. Burdon and J. G. Moffatt, ibid/, 8 8 , 5855 (1966) J M. G. Burdon and J. G. M o ffa tt, i b i d . , 8 £ , 4J25 1 ( 9 6 7 )°

(DGC). In this reaction, the acid catalyzed the condensation of DMSO and DGC to form 221, a sulfonium isourea, which was irreversibly attacked by phenol giving the phenoxysulf cnium salt 222. Loss of a proton from2 2 2 gave the ylid 2 2 2 . which rearranged in the previously

222

221

CH3 82 described Sominelet-Hauser type rearrangement to give 2-thiomethoxy- methylphenol (22*+) in 27 $ yield. This process also leads to dial- kylation and 2,6-di(thiomethoxymethy 1)phenol (22^) was obtained in

17$ y ie ld .

Moffatt and Lerch87 found that p-nitroaniline (l8l) reacted with

DCC, DMSO, and anhydrous polyphosphoric acid to yield S , S -dime thy 1-N- | p-nitrophenylsulfilimine (226) in 82$ yield. However, when p-anisidine

(8 7 ) U. Lerch and J . G. M o ffa tt, J0 Or go Chem. , 3 6 , 3861 (1971)*

(1 86 ) was treated under similar conditions, no analogous sulfilimine m s obtainedo

DCC ch3 DMBO N“S HH2 P 2 O 5 CHj

181 226

As previously discussed, Claus and co-workers8 0 were able to form sulfilim ines by the reaction of aniline with DMSO, phosphorous pentoxide, and triethylamine by nucleophilic attack upon the positive sulfur of a sulfonium type complex. Claus also found that DMSO, pyridine, and sulfur trioxide effected the thiomethoxymethylation of phenol to produce

22h in 3 % yield and 22J in 2% y i e l d .88

(88) P. Claus, Monatsh. Chem., 102, 913 (1971).

• • 39 More recently, a communication by Vilsmaier and Sprugel appeared in which the complex 22£ formed in the reaction of N-chlorosuccinimide

(8 9) E. Vilmaier and W. Sprugel, Tetrahedron letters, 625 (1972).

H CHp /

CH3 — > y r v

Cl ^C H s Cl" and dimethyl sulfide was reacted with anilines to form azasulfonium salts. The yields reported for this reaction were quite high ranging from 7 for naphthylamine to 91$ for p-anisidine. Reactions of this complex with phenols resulted in formation of dimethyl-(p-hydroxy- phenyl)sulfonium salts ( 228).

As previously noted, Appel and co-workers reported that ammonia reacted with complexes of chlorine and sulfides to form azasulfonium s a lt s . " However., ammonia and amines are more n u c le o p h ilic and more tasic than phenols and anilines . 4 9 i^e iow- yields in these reactions could perhaps he ascribed to a base rearrangement of the halosulfonium halide to form an a-chlorosulfide.

Halogen-sulfide complexes have been long known, formed from the reaction of halogen and sulfide. Such a complex was first proposed by Lawson and Dawson in the destruction of mustard gas, p, p '-dichlorodi- ethyl sulfide. 9 0 Bromination and chlorination of sulfides bearing an

(90) W. E. Lawson and T. P. Dawson, J. Amer. Chem. Soc., b$> 3119 (1927).

a-proton at low temperatures in nonpolar aprotic solvents often led to the formation of metastable adducts that decompose upon warming to

9 X give O'-halo sulfides. The major component of the bromine-tetrahydrothiophene adduct would seem to be ionic on the basis of X-ray and nmr data . 0 2 in the solid state the tricoordinate sulfur atom was (91) P. Haas, Biochem. J ., 2%, 1297 (1935)5 F. Runge, E. Rrofft, and R. Brux, J . P ra k t. Cb.em., 2, 279 (1955) 5 H. Bolune and E. Boll, Z. Anorg. A llg. Chem., 2£0, 17 (1957) 5 H. Fenselau and J. G. Moffatt, J. Amer. Chem. Son., 8 8 ,.1 7 6 2 (1966 ); G. E. Wilson and R. Albert, J. Orgo Chem., 3lL 2156, 2160 (1973)5 F* G. Bordwell and B. M. P itt, J. Aiaer. Chem., Soc., JX> 572 (1955)} H. Bohme and H. J. Gran, Justus Liebigs Ann. Chem. , 133 (1953)5 F. Boberg, G. Winter, and G. R. Schultze, Chem. Ber., 82 j 1160 (1956).

(92) G. Allegra, G. E. Wilson, Jr., E. Benedetti, C. Pedone, and R. Albert, J. Amer. Chem.' Soc., 92, ^002 (1970)*

pyramidal and the S-Br-Br arrangement was linear as in 22£. The nmr data indicated a highly charged sulfur atom, while self-consistent charge extended Hiickel calculations indicated a + 0. 3^- charge on the sulfur and a -0.8 charge on the terminal bromine with very little Br-Br interaction. An X-ray structure determination of the chlorine complex of bis(p-chlorophenyl) sulfide (£750) showed a trigonal bipyramid about the sulfur atom with the p-chlorophenyl groups and an unshared pair of electrons occupying the equatorial plane and the chlorine atoms 9 3 occupying axial positions. Thus, such complexes appeared to hear

(95) N. C. Baenzinger, r, e. Buckles, R. J. Manner, and T. D. Simpson, J. Amer. Chem. Soc., 9JL? 57^9 (1969)*

a positive sulfur which would he susceptible to nucleophilic attack hy an aniline.

Reaction of a halogen with a sulfide, addition of an aniline to produce an azasulfonium salt, and subsequent base rearrangement should provide a new method of preparing 2-thiomethoxymethylanilines. This

could also provide a, method of preparing 2-thiomethoxymethylanilines with strongly electron-donating groups on the aromatic ring. This reaction sequence proved possible. Condensation of chlorine in a

graduated test tube, transferal to methylene chloride at - 78 ° , and addition of dimethyl sulfide produced a clear solution of chlorodi-

nfithylsulfonium chloride (128) at -78°. Addition to this complex of a

1:1 solution of aniline and triethylamine produced the azasulfonium

salt 1£7 and triethylamine hydrochloride. The triethylamine was

Cl CH3SCH3 + Cl2

NaOCH 3 was present to scavenge the hydrochloric acid generated. Addition of 9k sodium methoxide in the usual manner gave 160. . Table 9 lists the

(9*0 For a preliminary report of this work see: P. G. Gassman, G. Gruetzmacher, and T. Jo van Bergen, J. Amer. Chem. Soc., 95. 5608 (1973).

yields of 2-thiomethoxymethylanilines obtained by this method. Exam ination of the table shows that 7-methoxy-2-thi omethoxymethylan iline was obtained in 62$ yield versus an isolated yield of 2$ in the reaction of N-chloro-p-anisidine and dimethyl sulfide. Thus this was a very superior procedure to use in the alkylation of anilines substituted with strongly electron-donating groups. It was not as good as the

TABLE 9

4-Substituted-2-Thiomethoxymethylanilines from Chlorodimethylsulfonium Chloride

-Substituent $ Y ield

0CHs (3SX) 62 CH3 (1£3) 5*1- H (160) 67 c i ( m ) *f5 COsCsHs (1 7 8 ) 35 K02 (182) 31 previously described method for electron-withdrawing groups, since they were less nucleophilic and did not displace the chloride from the sulfonium salt as readily as the more electron-rich anilines.

For comparison, p-anisidine was treated with the N-chlorosuccinimide: dimethyl sulfide complex, followed by treatment with base. In this case a 59$ yield of h-methoxy-2-(thiomethoxymethyl) - aniline (l8j) was obtained. Thus, the yields of product obtained in the two processes were similar.

This reaction was not limited to complexes of chlorine and di methyl sulfide. The complex formed between bromine and dimethyl sulfide

(2^1)95 when treated with aniline and triethylamine produced, after

(95) Bromodimethylsulfonium bromide (orange crystals, dec. 8 1-82° ) can be stored for several weeks at room temperature in the absence of light and moisture, N. Furukawa, T. Inoue, T. Aida, and S. Oae, Chem. Commun., 212 (19T3)«

base rearrangement, a 69$ yield of 160. Aniline could react with either

Br 1) ( c2h 5 )3f

Br" 2) KaOCH3 S i . 160 . X 1 ) 0® 2, (C2H5)3N s+ X nh2 the chlorine or hromine complex of tetrahy drothiophene to yield, after

base rearrangement, 1§6 in 2C Vp y ie ld .

The use of P-keto sulfides and G!-carboalkoxysulfides in this

reaction provided a similar route to indoles and oxindoles. In this manner 2$h was prepared in y ie ld and 225. in 53$ y ie ld .

H 23^

In summary, the use of halosulfonium. halides with substituted anilines provides a simple process-for the preparation of ortho-alky-

lated anilines, indoles, and oxindoles. This process is particularly

important in that methoxylated anilines can be used to prepare ortho -

alkyl anisidines and methoxylated indoles, which constitute a portion

of numerous indole alkaloids, and in the synthesis of methoxylated

oxindoles. Variation of the substitution patterns of the p-keto sulfides

and a-carboalkoxy sulfides used, should provide a ready access to a wide variety of methoxylated indoles of value as key intermediate in the

synthesis of certain natural products. Coupled with the reaction of

N-chloroanilines and various sulfides, followed by base rearrangement,

a wide variety of ortho-substituted aromatic amines, indoles, and

oxindoles can be prepared. EXPERIMENTAL

Melting points and boiling points are uncorrected. Infrared

spectra were recorded on a Perkin-Elmer Model 137 Infracord Spectro

photometer as neat liquids, solutions in carbon tetrachloride, or

powdered solids in potassium bromide disks. Nuclear magnetic reasonance

spectra were obtained.on a Varian Associates Model A_6o or A-60-A

Spectrometer and are reported in tau (t) units relative to tetra-

methylsilane (t = 10.00) as the internal standard. Elemental analyses were performed by the Scandinavian M icroanalytical Laboratory, Herlev,

Denmark.

N-tert -Butylaniline (80.). This compound was prepared by the procedure

of Bell and Knowles.

N-te rt-Butyl-p-toluidine ( 83.)= This compound was prepared by., the

34 method of Gassman, Campbell, and Frederick.

N-te rt-Butyl-4-chloroaniline (102). The method of Gassman, Campbell,

and Frederick was used to prepare this compound.

N-te rt-B utyl-fluoroaniline (5LT)* The procedure of Gassman, Campbell,

and Frederick was used to prepare this compound . 34 p-(N-te rt-Butylamino)nitrobenzene (82). This compound was prepared according to the method of Suhr.

90 91

N -te rt -Bufcylnnilinodimetbylsulfonium Chloride ( 8 7 ) • I 11 a th r e e - necked round -"bottomed flask containing 150 ml of pentane under a nitrogen atmosphere was placed 7°96 g (0 . 053b mol) of N-te rt-butylaniline. The solution was cooled to -10°, 63 . h g (ca. 10 eg..) of calcium hypochlorite was added; the mixture was stirred for 1 hr. The excess calcium hypochlorite was removed by filtration, and the solution was concentrated by rotary evaporation. The N-chloramine which was formed was used without further purification. The N-chloro-N-te rt- butylaniline was placed under nitrogen in a three-necked flask and the flask cooled to - 10°, and 40 ml of dimethyl sulfide was slowly added

(about 5 min) to the N-chloramine. After 5 min, the reaction flask was removed from the bath, allowed to ccane to room temperature, and the reaction mixture ms stirred for 3y hr* The white precipitate which ms formed, ms collected by filtration to yield 10.39 g

(0 . 0^23 m ol, 80fo yield) of N-te rt-butylanilinodimethylsulfonium chloride

(8 7 ). An analytical sample was prepared by low temperature recrystalli zation (three times from methylene chloride-ether): mp 1^-7 -1^-8° ; i r

(KBr) 3*39, 6 . 56 , 8.33, 10.05, 12.66, 13.79 M-J nmr (CDC1S) t 2 .3 0 -2 .85

(5H, m), 6.72 (6 h , s), 8 .5 6 (9H, s ) .

Anal. Calcd for Cr 2H20CMS: C, 58. 6 3 ; H, 8 .2 0 ; N, 5* 7° I S, ±3 . Oh.

Found: C, 58. 6 3 ; H, 8 .2 5 ; N, 5*675 S, 12.98.

N-tert-Butyl-p-toluidinodimethylsulfonium Chloride(85 .). In 25O ml o f pentane under a nitrogen atmosphere in a three-necked, round-bottomed flask was placed 10.00 g (0.0578 mol) of N-tert-butyl-p-toluidine. 92

Tills solution was cooled to -10°, and 82 g (ca. 10 eq.) of calcium hypochlorite was added. After the reaction mixture was stirred for

1 hr at -10°, the excess calcium hypochlorite m s filtered off, and the solution m s concentrated by rotary evaporation. The K-chloramine that m s prepared was used in the next step without further purification.

The chloramine under a nitrogen atmosphere was placed in a three-necked, round-bottomed flask and cooled to -10°, 100 ml of dimethyl sulfide was added slowly (about 5 min) with stirring. The flask ms kept in the cooling bath for 5 min after completion of the addition of the sulfide.

The mixture was stirred at room temperature for 4 hr. Hie salt formed m s collected by filtration in a dry box to give 11.96 g (0.0460 mol, 76 $ yield) of 8^. (This salt is extremely hygroscopic and must be maintained under an anhydrous atmosphere.) An analytical sample m s prepared by low temperature recrystallization (three times from chloro form -ether). Between recrystallizations the salt ms dissolved in chloroform and stirred over 3A molecular sieves. The purified sample had mp 153-154° I ir (KBr) 3-40, 6 . 67 , 7.27, 8.47, 10. 15, and 13.42 n; nmr (CDC1S ) t 8 .5 8 (9H, s), 7-53 (3H, s), 6 .7 2 (6h, s ) , 2 .45 -2 .8 8 ( 4 h , m).

Anal. Calcd for C 13H22CTNS: C, 60.09; H, 8. 5 3 ; N, 5*39; S, 12.34;

C l, 13.64.

Found: C, 59*80; H, 8. 3 7 ; N, 5-43; S, 12 .0 4 ; c i , 13.53.

N -te rt -Butyl -4-chlor oanilinodimethylsulf cnium Chloride (88). In a three- necked, round bottomed flask with 250 ml of pentane under nitrogen m s placed 10.00 g ( 0 .0 5 4 mol) of N-tert-butyl-4-chloroaniline. To this solution was added 78 g (ca. 10 eq.) of calcium hypochlorite. The

resultant solution was stirred for 3 hr at room temperature. The

excess calcium hypochlorite was filtered off, the pentane was removed

hy rotary evaporation, and the N-chloramine that was found was used in

the next step without further purification. The N-chloramine was

placed in a flask under nitrogen, 100 ml of dimethyl sulfide was added

over a period of about 15 min. The solution was stirred for 6 hr a t

room temperature. The salt that was formed was collected "by filtration

in the dry box and dried. The yield of 88_ was 9*07 g (0.0322 mol, 59$):

mp Ib2-lh5°, i r (KBr) 3 .^ 5, 6 . 76 , 8.Vf, 9*22, 15°79, 1^.18 n; nmr (CDC13)

t 8.62 (9H, s), 6.75 ( 6 H, s ) , 2.02-2.83 (^-H, m). A satisfactory ele

mental analysis could not be obtained for 7-chloro-N-te rt-butylanilino-

dimethylsulfonium chloride presumably because it m s extremely hygroscopic.

N-te rt-Butyl-fluoroanilinodimethylsulfonium Chloride ( 8 6 ). In a 250 m l,

round-bottomed flask wa s placed 2.00 g (0.0113 mol) of N -tert-butyl-4-

fluoroaniline and 100 ml of dry pentane-j to this solution was added, in

small portions, 20 g (ca. 10 eq.) of calcium hypochlorite. This mixture

was stirred for 2 hr at room temperature, at which time the excess

calcium hypochlorite was removed by filtration. The solvents were

removed, in vacuo to leave the N-chloramine as a dark yellow oil which was used without further purification. To this neat chlor amine under

nitrogen at room temperature was added with vigorous stirring 10 ml

of dimethyl sulfide. A voluminous white precipitate formed within ca.

30 sec; after stirring for 5 m in, 100 ml of dry ether were added and 9k

the mixture was stirred for 3 hr. The white precipitate was collected

by filtration in a dry box to yield 2.52 g (0.0092 mol, 80$) of IT-te rt-

butyl-4-fluoroanilinodimethylsulfonium chloride ( 8 6 ): ir (KBr) 3 . 36 ,

6 .2 9 , 6 .7 1 , 80IT, 10.18, 12.20 p; nmr (CDC13) t 8. 5^ (9H, s), 6 .6 8

(6 h , s ) , and 2.51-2.95 (^-H, m). An analtyical sample was prepared by

drying over kA molecular sieves in chloroform, concentration of the

solution, and recrystallization from chloroform-ether, mp 160 -162 °.

Anal. Calcd. for C12H3.9CIFNS: C, 5k,6k) H, 7-26; N, 5.31*. S, 12.15.

Found: C, 53*92; H, 7*38; N, 5*91; S, 12.27.

E -tert -Butyl -p-t oluidinotetr amethylenesulf onium Chloride ( 89.) <>

E-te rt-Butyl-IT-chloro-p-toluidine was prepared as described in the 8 8 O preparation of 85. To the dark red chloramine solution, cooled to -12

under nitrogen, was added dropwise 25 ml of tetrah y d ro th io p h en e . Hie

resulting deep red solution was allowed to warm slowly to room temper

ature, during which time a large amount of precipitate formed. The

reaction mixture was stirred at room temperature for 1 hr and filtered

in a dry box to give 5-97 g (0 .0 2 1 m ol, 6 &?o) of a light purple solid.

Purification was accomplished by stirring a chloroform solution of salt

over kA molecular sieves for 12-16 hr, concentration of the solution in

vacuo, and low temperature recrystallization from chloroform-ether. Hais

process was repeated twice to produce an analytically pure sample of 85?;

mp 128. 5 - 129. 5° ; i r (CHCI3 ) 3-35, 6 . 6 3 , 7*28, 8.O-8. 3 , 8.53, 9*00,

10.3 (j,; nmr (CDC13 ) t 8.62 (9H, s), 8.20-8.75 (2H, m), 7*61 (3H, s),

7 A 9 -8.O6 (2h, m), 6 . 3 1 -6 .9 3 (2H, m), 5*57-5*99 (2H, m), 2 . 56 -2 .9 8

(4h, m). Anal. Calcd. for Ci 5H24C1NS: C, 63.02 ; H, 8 .4 6 ; C l, 12.40;

N, 4 .9 0 ; S, 1 1 .22 0

Found: C, 63.24; H, 8 .35; C l, 12.19;

N, 4.93 5 S, IO . 98.

N-te rt-Butylanilinodimethylsulfonium Trifluoroacetate (g4). In 50 ml of dry methanol was placed 5*40 g ( 0.0220 mol) of N-te rt-buty lan ilino- dimethylsulfonium chloride. To this solution was added 4.86 g (0.0220 mol) of silver trifluoroacetate in 50 ml of dry methanol over a period of about 3 min. Upon commencement of the addition of the trifluoro acetate solution, a white precipitate formed. After 2 hr at room temperature, the silver chloride was filtered off, and the solvent was removed by rotary evaporation. This procedure yielded a reddish- brown oil which was triturated with dry ether to yield a white salt.

This salt was dried to yield 5*43 g (0.0168 mol, JQfo) of N-tert- butylanilinodimethylsulfonium trifluoroacetate. Three low temperature recrystallization from methylene chloride-ether gave an analytical sample: rap 149-150°; ir (KBr) 3*33, 5°92., 8.40, 8.97, 10.26, 12.20,

13.99, l4.l8 p.; nmr (CDCI 3 ) t 8.61 (9H, s), 6 .91 ( 6 h , s ) , 2 . 3 6 - 2.89

(5H, m).

Anal. Calcd. for C 14H19C1F3N02S: C, 52.00; H, 6 .2 3 ; N, 4 .3 3 ;

S, 9 .9 2 . „

Found: C, 51.82; H, 6.24; W, 4.26; s , 1 0 .1 9 . 9 6

N-tert -Butyl-p-toluidinodimethylsulf cmium Trifluoroacetate (22).

By a procedure analogous to that used in the preparation of IT-te rt-

butylanilinodimethylsulfonium trifluoroacetate, 52. was prepared in

82 % yield from. 8$.: mp 150-152°; ir (KBr) 3*313 5*92, 8. 34, 9*96,

10.27, 12.16, 12.52, 13.97 Pj nmr (CDC13 ) t 8.60 (9H, s ) , 7 .6 1 (3H, s ) ,

7.01 (6H, s ), 2 . 58-3 .1 4 (4h, m).

Anal. Calcd. for Ci5H22F3N02S: C, 53.40; H, 6.57; N, 4.15; S, 9*50.

Found: C, 53*04; H, 6.59; N, 4.03; S, 9*81.

N-te rt -Butyl-4 -chloroanilinodimethylsulfonium Trifluoroacetate (92.) •

By a procedure analogous to that used in the preparation of N-te rt- butylanilinodimethylsulfonium trifluoroacetate, 92. was prepared in 792 yield from 88: mp 142-143°; ir (KBr) 3.27, 5. 89, 6 . 70 , 8. 31, 8. 92,

12.13, 12.51, and 13*94 p| nmr (CDCI 3 ) t 8 .6 l (9H, s ) , 6.95 (6H, s ),

2.36-2.94 (4h, m).

A nal. C alcd. fo r Ci4Hi9 C1F3N02S: C, 46-99; H, 5*35; N, 3 .91.

Found: C, 47.01; H, 5*40; N, 3*90.

thermolysis of N-te rt-Butylanilinodimethylsulfonium Chloride ( 8 7 ).

A suspension of 4.0 g (O.OI 63 mol) of 87 in 100 ml of dry dimethyl-

formamide was heated at 100° under nitrogen for 2 hr. The resultant

deep purple solution was cooled to 0° and poured onto 200 ml of cracked

ice. The pH of the solution was adjusted to 11 hy the addition of

1G/3 aqueous sodium hydroxide, this solution was extracted with three

150 ml-portions of ether, the combined organic extracts were washed with saturated sodium chloride solution, dried over anhydrous potassium 9 7 carbonate, filtered, and concentrated in vacuo to give a viscous oil»

Column chromatography on a silica gel (Skelly B-benzene eluant) yielded 0.26 g of a yellow oil which on purification by preparative gas phase chromatography ( 3$ SE-30 Chrom G, 5' x l/h" column at l80°) gave 0. l4 g (O .7 2 mmol, hfi) of U-te rt-butyl-o-thioanisidine

^ (n e a t) 2 . 95? 3 * 3 2 ? 6 . 29, 6 . 62 , 6 . 89, 8 . 21, 13.50 |ij nmr

(CC14) t 8.60 (9H, s), 7-75 (3H, s), 5.O5 (3H, broad s), and 2.58-3„67

m) ; ng 3 1 . 56 O7 .

Anal. Calcd. for CnHiyNS: C, 67 . 6 ^ ; H, 8077i N? 7.17; S, 160^2.

Found: C, 67-75; H, 8. 7 7 ; N, 7-5^; S, 1 6 . 3 0.

Ihe second compound eluted amounted to 1 .5 0 g which on purification hy preparative gas phase chromatography ( 3$ SE-30 on Chrom G 5 ' x l./k" column a t 190° ) gave 0.69 g (3«5 mmole, 21$) of N-tert-butyl-p-thioanisidine ( 100_): ir (neat) 2 . 90, 3 * 3 1 ? 6 . 22, 6 . 65 , 8. 21, 12.22 n; nmr (CC14 ) t 8.71 (9H, s), 7 .6 6 (3H, s ) , 6 .7 0 (1H, broad s ) , 3 . 2 5 -

3.65 (2H, m) and 2.70-3.20 (2H, m); mp ^5-^6°.

Anal. Calcd. for Ci 3.H17 NS: C, 67 . 6 ^ ; H, 8 .7 7 ; N, 7.17-

Found: C, 6 7 . 6 7 ; H, 8. 8l ; N, 7.01.

The reaction was repeated and analyzed on a 10$ k:l Carbowax

20M: KOH. on Chrom W column using biphenyl as an internal standard which showed 36 $ of 80, % of 101, 36 $ of 100, and 3$ of 102.

Thermolysis of N-te r t -Butyl -p -1 oluid in o d ime thy 1 s ulf on ium Chloride ( 8 ^ ).44

N-te rt-butyl-p-toluidinodimethylsulfcnium chloride was thermolyzed on a preparative scale in a procedure identical to that used in the ther molysis of 87 . Preparative gas phase chromatography ( 3$ SE-30 on 9 8

Chrom G 5' x l/4" column at 150 ) of the resultant oil yielded 24$ of

IT-tert -tutyl-p-toluidine ( 8 3) as identified “by ir and nmr spectroscopy and 42$ of N-te rt -butyl-4 -methyl -o-thioanisidine ( 9 6 ): np 2*5 1. 5522; ir (neat) 2.90, 3*30, 6 .60, 8.20, 12.40 p; nmr (CDC13) t 8 .6 3 (9H, s ) ,

7.77 (3H, s ) , 7.71 (3H, s ) , 5.35 (1H, hroad s ) , 2. 76 -3 .1 0 (3H, m).

Anal. Calcd. for Ci 2Hl 9ITS: C, 68 . 8 5; H, 9.15 5 N, 6 .6 9 ; S, 15*32.

Found: C, 6 8 . 8 3; H, 9-21; N, 6 . 8 3 ; S, 15.II.

The reaction was repeated and analyzed on a 10$ 4:1 Carbowax

20M-K0H on Chrom W column using naphthalene as an internal standard which showed 32$ of 83 and 58$ of 9 6 .

Ihe reaction was repeated in a Carius tube fitted with a stopcock.

The tube with the stopcock closed was heated at 100° for 2 hr, cooled to - 196°5 and placed on a vacuum line. Hie stopcock was opened and the tube was evacuated at diffusion pump pressure. The tube was then gradually warmed to room temperature and all readily volatile materials were transferred to a similar tube at -196°. The peak observed at m/e

50 was due to CH3CI. 35 Anal. Exact theoretical mass molecular weight, calculated for

CH3CI35: 49.9923.

Found: 49-9924,

Thermolysis of N-te rt-Butyl-4-chloroanilinodimethylsulfonium Chloride

(8 8). F-tert -Butyl-4 -chlor oanilinodimethylsulf onium chloride was thermolyzed on a preparative scale in a procedure identical to that used in the thermolysis of N-te rt-butylanilinodimethylsulfonium chloride.

Preparative gas phase chromatography of’the resultant oil on a 3C$ 9 9 SE-30 on Chrom W column a t 175° y ie ld e d ^0$ of N-t e r t -b u ty l-4- chlor oaniline (102) and lkc/> of IT-t e r t -butyl -4 -chlor o -o-th ioani s i d ine

(103.): np 3 "2 1.570^; i r (n e a t) 3-3^-? 6.33? 6 . 67 , 8. 26 , 12.42 |ij nmr

(CC14 ) t 8.62 (9H, s ) , 7.63 (3H, s ) , 5-20 ( in , broad s ) , 2.6 3 -3 .4 6

(3H, m).

Anal. Calcd. for Ci ^Hi BC1KS: C, 57-50; H, 7-02; N, 6.10; S, 13-95-

Found: C, 57-53; H, 6 .9 3 ; N, 6 . 3 3 ; s , 13-99-

Ihe reaction was repeated and analyzed on a 10$ Carbowax

20M: KCH (4:1) on Chromsorb W column using naphthalene as an internal standard which showed 54$ of 102 and 18$ of 103.

Thermolysis of IT-te rt -Butyl-4 -fluor oanilinodimethylsulf onium Chloride

(8 6 ). In a preconstricted test tube was placed 931 (3»53 nmol) of

8 6 , 5 ml of dry dimethylformamide was added, the tube was sealed, and placed in an oil bath at 100° fo r 2 hr. The tube was opened, and the dimethylformamide was distilled off at reduced pressure. To the resultant residue was added 10 ml of 10$ aqueous sodium hydroxide; this solution was extracted three times with 10 ml-portions of methy lene chloride, the combined organic extracts were dried with anhydrous magnesium sulfate, filtered, and concentrated in vacuo to yield a dark oil. Column chromatography on silica gel (% ether-hexane eluant) gave 112.8 mg (0 . 53 mmol) of N-t e r t -b u ty l-4 - f lu o ro -o -th io - anisidine (£7): bp 75° ( 0 .2 0 mm); n f 3' 9 1.5^13*, ir (neat) 2.97? 3 -3 0 ,

6 . 58, 8.16, 11.13, and 12.35 n; nmr (CC14 ) t 8.80 (9h, s ), 7 .8 6 1 0 0

(3H, s ) , 6 .0 9 (OH, "broad s ) , 3 .OO-3. 6 I (3H, m).

Anal. Calcd. for C 1XH1SMS: C, 61 .92; H, 7*56; N, 6 . 5 9; S, 15. 03.

Found: C, 61 .0 2 ; H, 7*56; N, 6 .^ 3 ; S, 15-98.

Exact theoretical mass molecular weight, calculated for

Ch H16FNS: 213. 0987.

Found: 213. 0989.

Kinetic Procedure. Into a clean, dry, tared volumetric flask were weighed the internal standard and lithium chloride. Ihe flask was placed in a dry box and the azasulfonium salt was placed in the flask.

Hie flask was removed from the dry box, weighed and filled to the mark

Q Q with dry dimethylformamide. The flask was placed in a' constant

(96) The dimethylformamide had been stirred overnight with anhydrous magnesium sulfate, filtered, and distilled at reduced pressure from barium oxide into a 1 I round-bottomed flask with a side arm and septum inlet containing hk molecular sieves.

temperature bath and 0 .2 ml aliquots were withdrawn at intervals and placed in stoppered test tubes precooled to - 78 ° (dry ice-acetone).

The test tubes were cooled for an additional 10 min before ca. 0.25 ml of diethylamine was added to effect base rearrangement of any unreacted azasulfonium salt. The aliquots were analyzed via gas phase chroma tography to determine the ratio of internal standard to 2 -thiomethoxymethylaniline.

Ihese reactions were followed for lj to 2 half-lives. After this time the rate of thermolysis became faster and deviations from 1 0 1

linearity were observed. The concentrations of azasulfonium salt varied from JxlO "2 mole Jo~x to 7 x l0 “2 mole X"1 and the total initial i > ionic strength of the solutions varied from 0 .0 9 mole 2 kg”2 to 0 .2 8 I _I mole 2 kg 2. At least seven points were taken per run. Each k in

in Table 3 represents one run. A control was done in which a solution was divided into two equal parts and each part was treated as described

above. The agreement between the two runs was within %. Table 10

shows the data obtained from one run. A dedector response factor, K,

TABLE 10

Gas Phase Chromatography Peak Areas of 2-Thiomethoxymethyl~p-toluidine (142) and Phenanthrenea

Time (sec) Area p-Toluidine Area Phenanthrene

0 3649 1749 500 3245 1976 1000 2794 2235 1500 1751 1818 2000 llj-32 1893 2500 1145 1886 3000 729 I 6 O3 3500 539 1485

Q# Gas phase chromatography peak areas were obtained using a Vartan Associates Gas Chromatograph Model 1200, with a Hewlett-Packard Integrator Model 337° A.

could be obtained from the values at time =0 sec and the relationship;

(wt. 142) (area phenanthrene) (wt. phenanthrene) (area 142) 1 0 2

This relationship can be algebraically manipulated to give the number of moles of 2-thiornethoxymethy 1 -p-toluidine ;

K(wfc. phenanthrene) (area 1^2) moles of 1*1-2 = (MW of 1*1-2 ) (area phenanthrene)

Since all of the quantities on the right hand side of the equation are known at various time, the number of moles of 2-thiomethoxymethyl-p- toluidine at various times are known. This quantity corresponds to the number of moles of azasulfonium salt remaining at any one time.

Thus, a plot of In vs. t should give a straight line with a slope of k, the second order rate constant; a is the initial concentration of azasulfonium salt; b is the in itial total concentration of chloride anion j and x is the amount of azasulfonium salt that has reacted after time, t. Table 11 summarizes these results. This gives a k=3»37x10“3 I mole-1 sec-1 where 1=0.1961 mole2 kg 2.

TABLE 11

Data Used to Obtain Second Order Rate Constant

a-x (mole/f) b-x (mole/l) (^/m ole)

0 0. 05*1-5 0.1866 -0.0001 500 0c 0*51 0.1752 1.2992 1000 o . 0302 0.1623 3.4127 1500 0. 02*1-1 0.1562 *(-.8308 2000 0.0193 0.1514 6 . 275 S 23OO 0.0150 0. 1*1-71 7.9653 3000 0.0112 0 . 1*1-33 9*9791 3500 0.0086 0 . 1*1-07 11. 8*1-01 103 Thermolysis of N-te rt-Butyl-p-toluidinotetramethylenesulfonium Chloride

(82.) • 4 N-te rt-butyl-p-toluidinotetramethylenesulfonium chloride was thermolyzed on a preparative scale in a procedure identical to that used in the thermolysis of N-te rt-butylanilinodimethylsulfonium chloride.

Column chromatography on silica gel (ether-benzene eluant) of the resultant dark oil produced 32 $ of N-te rt-butyl-p-toluidine ( 8 3) as identified by ir and nmr and 2Cfjo of 4 -chloro-n-butyl -3 - (4-te rt -butylaminotolyl)sulfide (104): ap 2*5 1 . 5^+9 7 ) ir (neat) 2 . 9 0, 3*55 5 6 . 60 ,

8 .2 0 , 12.35 nj nmr (CDC13 ) t 8.64 (9H, s), 7.91-8.44 (4h, m), 7*77

(3H, s), 7.28 (2H, t), 6.47 (2H, t), 2.73-3.11 (3H, m).

Anal. Calcd. for C 15H24C1NS: C, 6 3 .02; H, 8.46; Cl, 12.40; N, 4.90;

S, 11.22.

Found: C, 6 3 .10; H, 8 .4 5 ; C l, 12.45; N, 4 .9 1 ; s , 11.23.

Ihe re a c tio n was rep e ate d and analyzed on a ^ SE-30 on Chrom W column using diphenylmethane as an internal standard which shewed kjfo o f 83 and 35fo of 104. pyrolysis of 4-Chloro-n-butyl-3-(4-te rt-butylaminotolyl)sulfide Hydro c h lo rid e . 44 A sealed tube containing 0.100 g ( 0 . 3I mmol) of the amine hydorchloride under nitrogen was completely submerged in an oil bath at 215° for 1 hr. Ihe tube was cooled and opened, and 10 ml of lOfo aqueous sodium hydroxide was added. Ihe resultant basic mixture was extracted three times with 15 ml portions of ether, the combined organic layers were washed with saturated sodium chloride solution, dried over 10l|- anhydrous potassium carbonate, filtered, and concentrated to give a

dark oil. Column chromatography of this oil on silica gel (Skelly

B-benzene, and ether-benzene solutions as eluant) gave as the first

compound to be eluted 23 mg (0.122 mmol, 39$) of 4-chloro-n-butyl-

3 -(4 -a m in o to ly l)su lfid e (1Q5.). This sample had i r and nmr sp e c tra

identical to an authentic sample prepared in these laboratories.

Bis-3-(4-aminotolyl) Disulfide (10J.).44 Following the procedure of

L. Horwitz and C. A. Clark46 for the preparation of 6 -amino -m-toluene -

thiol from the reaction of 2-amino-6-methylbenzothiazole (106) and

fused sodium hydroxide-potassium hydroxide-sodium monosulfide hydrate, the only product isolated was was 10J, 82$ yield, m/e 276 parent; mp 88°• This product is presumably formed by an oxidation of the

aminothiol during work-up as has been previously reported by Bogert

and Smidth.47

6-A m ino-m -toluenethiol H ydrochloride (.108 ) . 44 Following the procedure of Bogert and Smidth, bis-3-(4-aminotolyl) disulfide (10 j) was

reductively cleaved to its zinc salt by the action of zinc and boiling

acetic acid. The crude zinc salt, a white powder, mixed with metallic

zinc was dissolved completely, with the evolution of hydrogen in a

minimum amount of hot concentrated hydrochloric acid. Cooling gave an

88$ yield, based on disulfide, of 108, as a tan precipitate: mp

225-226°; ir (KBr) 3 .4 -3 .6 , 3 . 8 3, 6 .7 1 , 12.19 M3 nmr (DMS0-d6) t 7 .7 3

(3H, s), 2.37-2.85 (3H, m), 1.37 (4h, broad s). This last reaction

has been previously reported but the physical properties of the product

were not described.48 1 0 5

4-Ch2oro-n-butyl-3~(4-aminotolyl) Sulfide (105 ) . 44 To a suspension of 9.0 g (O.O 51 mol) of 6 -amino-m-toluenethiol hydrochloride (lOS) in 180 ml of dry tetrahydrofuran at 0° under nitrogen was added, in one portion, 5-54 g (0.103 mol) of sodium methoxide. The reaction mixture was stirred at 0° fo r 20 min and then stirred an additional 40 min at room temperature. The reaction mixture was again cooled to

0° and a solution of 8 .7 8 g ( 0 .0 5 1 mol) l-bromo-4-chloro-n-butane in 100 ml dry tetrahydrofuran was added. The reaction mixture was stirred at 0° for 1 hr and for 18 h r a t 3 The re a c tio n m ixture was concentrated in vacuo, dissolved in 200 ml of w ater and made a c id ic w ith 10f, aqueous hydrochloric acid. This_produced a large amount of white precipitate which was only sparingly soluble in water or aqueous hydrochloric acid. Collection of this precipitate by filtering the solution yielded, after washing with ether and drying 24 hr in a vacuum desiccator, 1 3 .0 g (0.049 mol, 95/0 of the desired amino-sulfide hydrochloride. Recrystallization five times from absolute ethanol gave an analytical sample of 4-chloro-n-butyl-3-(4-aminotolyl)sulfide hydro chloride: mp 169.5-170.5°; ir (KBr) 3 . 4 5 , 3 . 78 , 6 .2 0 , 6 .7 5 , 12.28 (i; nmr (DMS0-ds ) t 7.88-8.54 (4h, m), J.63 (3H, s ) , 6 .9 6 (2H, t ) , 6 .3 2

(2H, t), 2.34-2.94 (3H, m), 0.46 (3H, broad s).

Anal. Calcd. for CnHiyClaNS: C, 49-63; H, 6.44; Cl, 26 . 6 3 ;

N, 5*26; S, 12.04.

Round: C, 49.70; II, 6 . 3 5 ; Cl, 26.55;

N, 5 . 1 2 ; s , 1 2 .1 4 .

The free amine 10£ was obtained by vigorously stirring the hydro chloride salt with a two phase system of 10$ aqueous sodium hydroxide 1 0 6 and ether□ Ihe ether layer m s separated, mshed with saturated sodium chloride solution, dried over anhydrous potassium carbonate, and con- contrated to give a brownish oil: ir (neat) 2 . 85, 2. 95, 5 .4 0 , 6 . 20,

6 . 70 , 7-71, 12.29 nj nmr (CC14) t 7-92-8.43 (4h, m), 7-79 (3H, s),

7.26 (2H, t), 6.53 (2H, t), 5.98 (2h, broad s), 3-49 (1H, d), 3.14

(OH, d), 2. 86 (1H, m).

N-tert -Butyl-(N-methylthio) -p-toluidine (129) • I*1 a graduated test tube, 1.0 ml (ca. 1.5 eq..) of chlorine m s condensed at - 78 ° ; 15 nil of dry ether cooled to -j8° in a jacketed addition funnel m s added to the chlorine. This solution was transferred to a jacketed addition funnel, cooled to - 78 °, and added dropwise to a solution of 1.44 g (ca. 1.5 el-) of dimethyl sulfide in 100 ml of dry ether at -78°. This

go procedure produced a pale yellow sqlution of methane sulfenyl chloride which was stirred for 10 min. To this solution was added a solution of

3 .2 6 g (20.0 mmol) of N -tert -butyl -p -toluidine and 3 i'll of triethylamine in 10 ml of dry ether, ihe resultant solution was stirred for 30 min a t - 78 ° at which time the cooling bath was removed and the solution m s allowed to warm to room temperature. Ihe reaction m s quenched by the addition of 100 ml of 10p aqueous ammonium hydroxide. The layers were separated. The aqueous phase was extracted two times with 50 ml portions of ether. The combined organic layers were washed with saturated sodium chloride solution. Ihe sodium chloride solution was back extracted with 75 ml of ether. The combined organic layers were dried over anhy drous magnesium sulfate, filtered, and concentrated to yield a yellow oil. This oil was distilled using a molecular still with the receiver cooled to - 7 8 ° to g i v e 5 .1 2 g ( l 4.9 mmol, o f 1 2 2 ,: lap 55-57°

(O.35 mm); n ^3*4 1.5284; ir (neat) 3*3^ 6 . 18, 6 . 63 , 7 . 21, 7 . 35, 8. 22,

12.21 |x; nmr (DMS0-d6 ) t 8 .7 8 (9H, unsummetrical doublet in ratio 2:1),

7*79 (3H, m), 7*74 (3H, s), 2.87-3*39 (^H, m). All attempts to obtain pure sample of this material for combustion analysis failed.

Anal. Exact theoretical mass molecular weight, calculated for

C1 2 H1 9 NS: 209. 1238.

Found: 209.1240.

Sodium Methoxide Rearrangement of IT-tert-Butylanilinodimethylsulfonium

C hloride (8 7 ). In 5 ml of dry methanol, 1 .0 0 g (4.06 mmol) o f 87 was d isso lv e d and co o led to 0 ° w h ile 0 . 5 g ( 5 * 3 5 eq.) of sodium was placed in 15 ml of dry methanol and cooled to0 °. Ihe methoxide solution was added to the sulfanium salt solution over a period of about 1 0 min a t 0°. After the addition of the sodium methoxide m s completed, the ice bath was removed and the solution m s stirred at room temperature fo r 2 h r. Ihen 1 0 ml of water was added to the reaction mixture and the mixture was extracted three times with 20-ml portions of ether. Ihe organic layer was washed with saturated sodium chloride solution,

stirred over anhydrous magnesium sulfate, and filtered. Ihe ether m s removed by rotary evaporation to yield a yellow oil. Molecular distillation gave 0 . 8l g (3*87 mmol, 95^ ) °f N-tert-butyl-2-(thiomethoxymethyl) aniline ( l4 l) : np6 * 4 1 * 5532; ir (neat) 3 * 2 8, 5* 92, 6 .7 8

8.06, 12.66 n; nmr (CC14 ) t 8.64 (9H, s ) , 8 .1 5 (3H, s ) , 6.46 (2H, s ) ,

5. 56 - 5.85 (lh, broad s), 2.84-3.64 (4h, m). 1 0 8

Anal. Calcd. for C 12Hl 9KS: C, 68 . 8 5; H, 9-15; N, 6 .6 9 ; S, 15. 31.

round: 0, 68 . 8 6 ; H, 9 .06 ; N, 6 . 5 8; S, 15o07.

Sodium Methoxide Rearrangement of N-tert -Butyl-p-toluidinodimethyl-

sulfonium Chloride 8 ( 5 ). In. 5 ml of dry methanol, 1.00 g (3 .6 1 mmol)

of 85, was dissolved, while 0 .5 g (6 . eq..) of sodium was allowed to

react with 15 ml of dry methanol in a separate vessel. Both solutions o were cooled to 0 and the sodium methoxide solution was added over a

period of about 10 min to the azasulfonium salt solution. After the

addition of the sodium methoxide solution, the ice hath was removed,

and the solution was stirred for 4 hr at room temperature. At this

tim e, 1 5 ml of water was added to the reaction mixture, and this a solution was extracted three times with 2 5-ml portions of ether. The

organic layer was washed with saturated sodium chloride solution,

stirred over anhydrous magnesium sulfate, and filtered. Rotary

evaporation removed the ether to yield a light yellow oil. Molecular

distillation of the oil yielded O .75 g (3 .3 6 mmol, 92/a) of II-t e r t - b u ty l-

2 -(thiomethoxymethyl)p_-toluidine (l42): n ^6*8 1.54731 ir (neat) 3*33 5

6.71, 7.41, 7*63, 12.42 n; nmr (CC14) t 8 .6 9 (9H, s ) , 8.20 (3H, s ) ,

7.87 (3H, s), 6.52 (2H, s), 5.79-5.98 (3H, broad s), 3 .03-3.42 (3H, m).

Anal. Calcd. for C 13H21NS: C, 69-70; H, 9-48; N, 6 . 2 7 ; S, 14.35.

Found: C, 70.00; H, 9-67; N, 6 .2 3 ; S, 14.11.

Sodium Methoxide Rearrangement of N-tert -Butyl-4 -chlor oanilinodimethyl -

su lf onium Chloride (88). In an anhydrous atmosphere (or in a dry box)

' 1*^7 g (5»25 mmol) of 88 was placed in a tared three-necked flask. The 1 0 9 flask was removed from the dry box, nitrogen passed thru the flask, and 25 ml of dry methanol was added. The flask was cooled to 0° ;

0o40 g (1.4 eq.) of sodium methoxide was dissolved in 25 ml of methanol o and cooled to 0 . Ihe sodium methoxide was added over a period of about 15 min to the azasulfonium salt solution, after which time the reaction mixture was allowed to stir at roam temperature for 2 h r ;

25 ml of water was added. This solution was extracted three times with

25-ml portions of ether. The combined organic layers were washed with saturated sodium chloride solution, stirred over anhydrous magnesium sulfate, and filtered. Removal of the ether by rotary evaporation produced a light yellow oil. Molecular distillation of the oil gave

1.16 g (4.76 mmol, 9056) of IT-te rt-butyl-4-chloro-2- (thiomethoxymethyl) - 24 * 2 aniline (l44): np 1 .5 6 4 9 1 in (neat) 3-33, 6 . 62 , 6 .8 5 , 8. 2 6 , 12.42 p,; nmr (CDC13 ) t 8 . 61 (9H, s ) , 8 .0 3 (3H, s ) , 6.40 (2H, s ) , 5-85-5.97

(OH, broad s ) , 2.69-3-43 (3H, s).

Anal. Calcd. for Ci 2H18CMS: C, 59-12; H, 7-44; IT, 5-75; S, 13-15-

Round: C, 59-19; H, 7 -5 1 I N, 5-50; S, 12.99.

Sodium Methoxide Rearrangement of IT-te rt -Butyl -4 -fluor oanilinodime thyl - s u lf onium' C hloride ( 8 6 ). In a procedure identical to the one used in the sodium methoxide rearrangement of IT-te rt-butyl-4-chloroanilinodimethylsulfonium chloride, 86 was rearranged in 88$ yield to IT-te rt-butyl-

4-fluoro-2-(thiomethoxymethyl)aniline (142.): bp 85° (0 .1 5 mm); a ^3 *7

1.5363; ir (neat) 2. 9 6 , 3-33, 6.75, 8.14, 12. 36 , and 13-64 p,; nmr

(CCI4 ) t 8.61 (9H, s), 8.08 (3H, s), 6.47 (2H, s ) , 5-99 (IK, Iroad s) 1 1 0 and 3.12-3.47 (3H, m).

Anal. Calcd. for CiaHiSFNS: C, 63 .4 0 ; H, 7*98; R, 6 . 1 6 ; S, 14.10.

Found: C, 6 3 . 2 3 ; H, 8. 0 3 ; R, 6.10; S, 14.17.

Sodium Methoxide Rearrangement of R-te rt-Butylanilinodimethylsulfonium

Trifluoroacetate (j§4). In 10 ml of dry methanol, 1.00 g ( 3 .09 mmol) of 9k -was cooled to 0° , w hile 0 .5 g sodium (7 eq.) was allowed to react o w ith 15 ml of dry methanol in a separate flask and also cooled to 0 .

Ihe sodium methoxide was added over a period of about 10 min, after which time the ice bath, was removed. Ihe reaction mixture was stirred for l|- hr at roam temperature. The methanolic solution was taken up in 20 ml of water, and the solution was extracted three times with

50-ml portions of ether. The combined? organic layers were washed with saturated sodium chloride, stirred over anhydrous magnesium sulfate, and filtered. Removal of the ether by rotary evaporation yielded a light yellow oil. Molecular distillation of the oil produced 0 . 9k g

(2 .5 8 mmol, 83ifo) of R-tert-butyl- 2 -(thiomethoxymethyl)aniline (l4l) identical in all respects to previously prepared samples.

Sodium Methoxide Rearrangement of R-te rt .Butyl-p-toluidinotetramethylene - sulfonium C hloride (8)9 ) . 44 In a procedure analogous to the one used in the sodium methoxide rearrangement of R-tert-butyl-p-toluidinodimethylsulfonium chloride, 89 gave, after column chromatography on silica gel (ether-hexane eluant), 293 of R-tert-butyl- 2 -( 2 -tetrahydrothienyl)- p-toluidine (l4£): mp 38.5-40.5°; ir (KBr) 2.91, 3-32, 6.60, 8.22,

12.35 n; nmr (CCI 4 ) t 8.62 (9H, s), 7.52-8.00 (7H, m), 6.73-7.12 (2h, m), I l l 6.06 (IK, -broad s), 5.^0 (TH, t), 3.03-3.27 (3H, m).

■Anal. Calcd. for CisH^NS: C, 72 . 2 3 ; H, 9-29; N, 5.62; S, 12.86.

Found: C, 72-32; H, 9-33; N, 5 .5 9 ; S, 1 2 . 76 .

The second compound eluted was 33$ of N-te rt-butyl-p-toluidine

(82 )- The reaction was repeated and analyzed via gas phase chromatography

on a jjo SE-30 on Chrom W column using diphenylmethane as an internal

standard which showed 56 $ of 83. and 70$ of 1V5.

N-te rt-Butyl-o-toluidine (176). To a vigorously stirred solution of

0.507 g (2.7l mmol) of N-te rt-hutyl-2(thiomethoxymethyl)aniline in

20 ml of absolute ethanol was added 1 3A- teaspoonfuls (ca. 5 g) of

W-2 Raney nickel66 and about 70 ml of absolute ethanol used to wash the spoon and powder funnel.. This mixture was refluxed for 2 hr,

200 ml of water was added and the N -tert-butyl-o-t oluidine was azeotropically distilled with the water-ethanol. Tie distillate was extracted with three 100 ml portions of"ether, the combined organic layers were washed with saturated sodium chloride, dried with anhydrous potassium carbonate, and filtered. Tie organic solvents were removed hy distillation through a Vigereux column. Molecular distillation affo rd ed O.2 7 1 g (1.66 ramol, 6 9$) of N-tert -butyl-o-toluidine (176): n j 5*9 1.5178 (lit 34 ng 5 ,0 1 . 5178 ).

N-tert -Butyl-2,7 -zylidine (l7l). In a procedure identical to that

used in the preparation of N -tert-butyl-o-t oluidine, iTj "was prepared 1 1 2 in 7 &p yield from N-tert -butyl -2 -(thiome thoxymethyl) -p-toluidine: n 2 5 ,2 1. 5136 ; ir (neat) 2 . 90, 3 .3 4 , 6 . 58, 6 . 85, 8. 20, 12.42 p,; nmr

(CCI4) t 8.69 (9H, s ) , 7.94 (3H, s ) , 7 .8 2 (3H, s ) , 3 . 2 2 -3 .5 7 (3H, m).

No N-H resonance was seen.

Anal. Calcd. for C 12HlSN: C, 81. 3O; H, 1 0.80 ; N, 7*90.

Pound: C, 81 .08; H, 10.84; N, 8. 03.

4-Methoxy-2-(thiomethoxyme thyl)aniline (lSx). In 150 ml of methylene chloride was placed 4.65 ml (3 eq..) of dimethyl sulfide which was cooled to - 78 ° under nitrogen. In a jacketed constant pressure addition funnel 2 .1 7 g (0 .0 2 mol) of te rt-hutyl hypochlorite in 10 ml of methylene chloride was cooled to - 78 ° and added dropwise to the dimethyl sulfide solution. Ihe reaction mixture was stirred for 10 min. In a non-jacketed addition funnel was placed 2.46 g (0.02 mol) of p-anisidine in 20 ml of methylene chloride which was added dropwise to the reaction mixture. The reaction ms stirred for 2 hr o a t -78 . In 25 ml of ahsoltue methanol was dissolved 2.28 g (2 e q .) of sodium methoxide; this solution was added dropwise to the reaction mixture which was stirred for 15 min at -7 8°. At this time the dry ice-acetone bath was removed, and the reaction was allowed to come to room temperature. Ihe reaction m s quenched with 5° ml of water; the layers were separated; the aqueous phase was extracted two times with 50-ml portions of methylene chloride. Ihe combined organic extracts were dried over anhydrous magnesium sulfate, filtered, and concentrated to yield a dark oil. Ihis oil was distilled to give a 11 3 yellow oil [bp 80-120° (o*l ram)] which solidified. Preparative gas chromatography on a 1C$ Carbowax 20M-KOH on 60-80 Chromsorb W column at 190° with a flow rate of 200 ml/min was used to separate product from starting material. The first compound eluted was p -a n is id in e (O .5 6 g , k.6 mmol); the second compound eluted was b- methoxy-2-(thiomethnxymethyl)aniiine ( 187 ) (82 rag, 0 . mmol, 2 .2 $ ,

2 .9$ based on recovered starting material): n^4*4 1.5965; ir (neat)

2 .9 0 , 3 . 58, 6 .6 1 , 8. 03 , 9. 6 O, and 12 . 35 nj nmr (CCI4) t 8.06 ( 3H, s ) ,

6.16 (2H, s), 6.32 (3H, s), 6 .3 0 (2H, broad s ) , 3 .18 (3H, s ) .

Anal. Exact theoretical mass molecular weight, calculated for

C9H13N0S: 183. 0717 .

Found: 183. 0720 .

1-Methyl-2-(thicmethoxymethyl)aniline ( 165 ). A solution of 11.50 g

(0.1075 mol) of p-toluidine in 100 ml of methylene chloride was vigorously stirred and cooled to - 65 ° under nitrogen. To this solution was added dropwise 11.68 g ( 0.1075 mol) of te rt-butyl hypochlorite in

20 ml of methylene chloride also at - 65 °. After the addition was complete, the solution was stirred for 15 min. ihe addition funnel was rinsed with ^-0 ml of methylene chloride and 25 ml (ca. 3 ©1*) of dimethyl sulfide were placed in it and cooled to - 65 °. Bie dimethyl sulfide was added dropwise to the N-chloro-p-toluidine solution and stirred for 3f hr at - 6 5 °• In 50 ml of methanol, 7*0 g (0.1290 mol,

1.2 eq..) of sodium methoxide was cooled to -65°? and added dropwise to the reaction mixture. This solution was allowed to stir for 1 hr at 1 1 4

-65° after which time the dry ice-acetcne hath was removed and the

reaction mixture was allowed to come to room temperature. In order

to remove the inorganic salts, 150 ml of water was added to the

reaction, the layers were separated, and aqueous phase was washed

twice with 100-ml portions of methylene chloride. The combined

organic layers were dried over anhydrous magnesium sulfate and filtered.

The solvents were removed in vacuo to yield a red oil. This oil was fractionally distilled to yield 3*04 g (0.028 mol, 2Bfo) of

p-toluidine, bp 85-90° (15 mm), and T »50 g of 16 >5j> bp 9 0-92. 5°

(0 .0 3 mm), which was recrystallized from pentane to yield 7.28 g

(0.044 mol, kOp) mp 46.5-47° (litsoc 42-45°). Bie yield of 16 £

based on unrecovered p-toluidine was ^ 5>>•

2-(Thiomethoxymetbyl)aniline (l6o). A vigorously stirred solution of

10.0 g (O.IO75 mol) of aniline in 250 ml of methylene chloride was

cooled to -78° under nitrogen. In a jacketed constant pressure

addition funnel 11.68 g (O.IO75 mol) of te rt-butyl hypochlorite was

also cooled to - 78 ° and added dropwise to the aniline solution over a

5 min period. The stirring was continued for 25 additional min, at which time the reaction mixture had turned dark green. The addition

funnel was rinsed with 10 ml of methylene chloride and 20 ml (ca. 3 e(l* )

of dimethyl sulfide was placed in the funnel and cooled to -78 .

Upon addition of the dimethyl sulfide an exotherm was observed; this solution was stirred for 40 min. In 5° ml of methanol, 7*0 g

(0 .1 3 m ol, 1 .2 eq.) of sodium methoxide was placed in the funnel, 1 1 5 cooled to -78°, and added quickly to the reaction mixture. This solution ms allowed to stir for 1 hr at - 78 ° at which time 100 ml of water was added and the heterogenous mixture allowed to come to room temperature. The layers were separated, the aqueous layer was extracted with two 100-ml portions of methylene chloride, the combined organic extracts were washed with saturated sodium chloride, dried over anhydrous sodium sulfate, filtered, and the solvent was removed in vacuo to leave a dark red oil. This oil was factionally distilled to yield 1.84 g (0.02 mol) of aniline and 11.82 g ( 0.0 7 7 mol) of 1^0, bp 135-159 (10 mm), fo r a 72 $ yield based on starting material and

90$ yield based on unrecovered starting material: ftp4’0 1.6083} (lit80C nD 1.6042).

4-Garboethoxy-2-(thiomethoxymethyl)aniline -(!!§.)• A solution of

16.52 g (0.10 mol) of benzocaine in 400 ml of methylene chloride was vigorously stirred with a Herschberg stirrer and cooled to -78 under nitrogen. To this solution was added dropwise 10 .8 5 g (0.10 mol) of te rt-butyl hypochlorite in 10 ml of methylene chloride. After the reaction mixture had stirred for 2 h r , 25 ml (3 eq.) of dimethyl sulfide was added to the solution; this solution was stirred for 18 h r.

A solution of 6.48 g (1.2 eq.) of sodium methoxide in 5° ml methanol was added to the reaction mixture} the resultant solution was stirred o for 1 hr at -To at which time the dry ice-acetone bath was removed and the reaction mixture was allowed to come to room temperature. 1 1 6

The Inorganic salts formed were dissolved by the addition of 100 ml of water, the layers were separated, and the aqueous phase was washed with two 150-ml portions of methylene chloride. Die combined organic layers were dried over anhydrous magnesium sulfate, filtered, and the solvents evaporated to produce a dark red oil which solidified upon standing. This red solid was refluxed in 400 ml of toluene with

25 ml of triethylamine for 24 hr. The solvent was removed by rotary evaporation to produce again a red oil which solidified upon standing.

The solid was recrystallized in two crops from absolute ethyl alcohol to y ie ld 14.70 g (O.O65 mol, 65 $) of 178, nip 83-84. 5°* An a n a ly tic a l sample was prepared by recrystallizing a small portion twice from pentane-ether: 84.3-85*5°.; (KBr) %.95s 3*52, 5*93> 6 . 13 , 6 . 25,

7 .8 1 , 8 . 37 , 12.96 ijl; nrar (CDCI3 ) t 8.66 (3H, t), 8.04 (3H, s), 6 .3 3

(2H, s ) , 5.68 (2H, q), 5*60 (2H, broad s), 2 .7 8 (3H, m).

Anal. Calcd. for CnHisNOsS: C, 58.64; H, 6 .71} N, 6 . 23.

Found: C, 58. 6 6 ; H, 6 .6 6 ; N, 6 . 15.

4-Chloro-2-(thiomethoxymethyl)aniline (l£l). In 150 ml of methylene c h lo rid e , 5.50 g (0.043 mol) of p-chloroaniline m s dissolved and cooled to -78° under nitrogen. In a jacketed constant pressure addition funnel, 4.67 g (0.043 mol) of te rt-butyl hypochlorite o in 10 ml of methylene chloride was cooled to -78 and added dropwise to the aniline solution. The reaction mixture was stirred for 3° min, during which time the solution turned dark yellow. Die funnel was rinsed with 15 ml of methylene chloride, 25 ml (ca. 5 ©1*) of dim ethyl o sulfide was placed in the funnel and cooled to -78 • Ihe sulfide was 1 1 7 added dropwise and the reaction was stirred for 5§ To 'the reaction mixture was added 7»00 g (0 .0 5 2 mol, 1.2 eq.) of sodium methoxide in 5° ml of methanol. The resultant solution was stirred for 1 hr a t -78 , the hath was removed, and the reaction mixture was allowed to warm to room temperature over night. The reaction was quenched by the addition of 100 ml of water, the layers were separated, and the aqueous phase was extracted two times with 150-ml portions of methylene chloride. The combined organic extracts were washed with saturated sodium chloride solution, dried over anhydrous sodium sulfate, filtered, and the solvent evaporated to yield a red solid. Chromatography on silica gel {2% ether-pentane solvent) separated the product from starting material. The first compound eluted was 1X1 (5*^ gj

0.029 mol): mp 79-80° (litsoC mp 78 - 79 ° ) ; 67 ^ (83$ based on recovered starting material). The remaining I .85 g of material eluted was a mixture of product and starting material. h-Nitro-2-(thiomethoxymethyl)aniline (l82_). In a 1 1, three-necked flask, equipped with a mechanical stirrer, was placed 2 .7 6 g (0.02 mol) of h-nitroaniline, 300 ml of acetonitrile, and 100 ml of methylene

, o chloride. This solution was cooled to -hO to -50 by means of a dry i c e SCffo' aqueous methanol bath. From an addition funnel, 2.17 g

(0.02 mol) of tert-butyl hypochlorite in 5 ml of methylene chloride was added dropwise. This solution was stirred for 3° min c&» o -45 ; 5 ml (ca. 3 eq.) of dimethyl sulfide was added dropwise ; and the reaction mixture was stirred for 18 hr while maintaining the o temperature at about -40 . At this time, 5 ml of triethylamine was 1 1 8 added dropwise., and the solution was stirred for 1 hr. The cooling hath was removed, and the reaction mixture was refluxed for 4-8 hr.

The solvents were removed by rotary evaporation. To the resultant red oil was added 100 ml of 1C$ aqueous sodium hydroxide and 100 ml of methylene chloride. The layers were separated, the aqueous phase was extracted with two 100-ml portions of methylene chloride, the organic phases were combined, dried over anhydrous magnesium sulfate, filtered, and the solvents removed in vacuo. The resultant red-yellow solid was chromotographed on silica gel with hexane-ether to y ie ld 2 .1 0 g (10.6 mmol) of 182: mp 76 - 77 ° (litsoc mp 75 -77 ° ) , 53$ (71^ based on unrecovered starting material). The second compound eluted was 4-nitroaniline (O .7 0 g, 5 mmol, 2%).

2-Amino-3-(thiomethoxymethyl)biphenyl (180). To a vigorously stirred solution of 400 ml of methylene chloride and 8.46 g ( 0 .0 5 mol) of

2 -aminobiphenyl, cooled with a dry ice-acetone bath, was added dropwise 5.4-2 g (0.05 mol) of te rt-butyl hypochlorite in 10 ml of' methylene chloride. The resultant solution was stirred 1 hr; 12. 5 ml of dimethyl sulfide (ca. 3 eq. ) was added, while maintaining the exotherm to less o than 10 . This solution was stirred for 4 hr. In 30 Ml of a b so lu te methanol was dissolved 3.24 g ( 1 .2 eq.) of sodium methoxide; this solution was added dropwise to the reaction mixture and stirred for

1 hr. The cooling bath was removed, the reaction mixture was allowed to come to room tem p eratu re, 200 ml of water was added, the layers were separated, the aqueous phase ms washed two times with 2 00-ml 1 1 9 portions of methylene chloride, and the combined organic phases were

dried with anhydrous magnesium sulfate, The solution was filtered,

and the solvents were removed in vacuo to yield a dark oil. The oil

was vacuum transferred to give 6 .5 5 g of a yellow oil, which ms

chromatographed on silica gel with ether-pentane as eluant. This

procedure yielded 1 .2 5 g (0.0074 mol) of 2 -aminobipbenyl and 4.02 g

( 0.0175 mol) of h3<0 ( 55/^ yield, 4i$ based on unrecovered starting material): np5'° 1.6455; ir (neat) 2.94, 3.48, 6 . 25, 6 .9 1 , 13. 20,

13.45, and 14.31 \i\ nmr (CClJ t 8.15 (3H, s), 4.39 (2H, s), 3 .9 7

(2H, broad s), 2. 55-3 .60 (8h , m).

Anal. Calcd. for C 14H i5NS: C, 73.32; H, 6.59) N, 6.11.

Found: C, 7,3.34; H, 6 .6 0 ; N, 6 . 23.

6 -Methyl-2-(thiomethoxymethyl)aniline (l6j9). To a stirred solution of

11.5 g (0.1075 mol) of o-toluidine in 400 ml of methylene chloride, o cooled to -78 under nitrogen, was added dropwise from a jacketed

constant pressure addition funnel, 11.7 g ( 0.1075 mol) of tert- o butyl hypochlorite at -78 . This solution ms stirred for 40 min

during which time the solution turned dark green. The funnel m s rinsed with methylene chloride^ 25 ml (ca. 3 eq..) of dimethyl sulfide was placed in the funnel and cooled. Upon dropwise addition of the

sulfide, the solution turned dark red. After 1^- hr a voluminous white precipitate had formed; the reaction mixture was stirred for an additional 30 min. In the funnel was placed 7.0 g (0.13 mol, 1.2 eq.)

of sodium methoxide dissolved in 50 ml of methanol. This m s cooled 1 2 0 and quickly added to the reaction mixture. The "bath was allowed to come to room temperature overnight. The inorganic salts that had formed were dissolved in 100 ml of water. The heterogeneous mixture was separated; the aqueous layer was washed two times with 1 ^0 -ml portions of methylene chloride; the organic layers were combined, dried over anhydrous sodium sulfate, filtered, and concentrated on the rotary evaporator to yield a red oil. Rractionation of the resultant o i l gave 1 .8 0 g (0 .0 1 7 mol) of o-toluidine and 11.38 g (0 .0 6 8 mol) of

182. yield based on starting material and 7 *$ y ie ld based on unrecovered starting material): bp 82. 5-84. 5° (0*05 mm); n^4’ °

1.5963 (lit 800 nD I . 5998).

3-Methoxy-2-(thiomethoxymethyl)aniline (184) and 5-Methoxy-2-(thiomethoxymethyl)aniline ( 183). Under nitrogen, in a 1 1, three-necked flask equipped with a mechanical stirrer, m s placed 12 .3 2 g (0 .1 0 mol) of m-anisidine and 400 ml of methylene chloride. This solution was cooled to - 78 °. In a jacketed constant pressure addition funnel was p laced IO .8 5 g (0 .1 0 mol) of tert-butyl hypochlorite in 15 ml of methylene chloride. This solution m s cooled to ca. - 65 °, added dropwise to the reaction mixture, and stirred for 5 min. The addition funnel was rinsed wifc h to ml of methylene chloride, and 25 ml (ca.

3 eq.) of dimethyl sulfide was placed in the addition funnel, cooled to ca. - 6 5 ° and added dropwise to the reaction mixture. The solution was stirred for 36 min, during which time a white precipitate formed.

Sodium methoxide (6.48 g, 0.12 mol) in 50 ml of anhydrous methanol 1 2 1 was cooled and quickly added to the reaction mixture. The resultant solution was stirred for 15 min, the cooling hath was removed, and the solution was allowed to come to room temperature, 150 ml of water was added, the layers were separated, and the aqueous phase was washed with two 100-ml portions of methylene chloride. The combined organic phases were dried over anhydrous magnesium sulfate, filtered, and the solvents removed to leave a red oil. This oil was taken up in U 00 ml of tert -butyl alcohol and refluxed for 12 hr with 11.20 g (l eq.) of potassium te rt-butoxide. The te rt-butyl alcohol was removed in vacuo. The residue was taken up in 100 ml of.water and 100 ml of methylene chloride; the layers were separated^ and the aqueous phase was washed with two 100-ml portions of methylene chloride. The combined organic phases were dried over anhydrous magnesium sulfate, filtered, and the solvents were removed by rotary evaporation to yield a dark oil. D istillation of this oil gave 1.62 g

(0.013 mol, 13fO of m -an isid in e and 8 .7 7 g [ 0. 0U8 mol, bdfo { 5 % based on unrecovered starting m aterial)] of a mixture ( 3 :1 by nmr) of l 8j? and JJ34, bp 141-1^5° (°*^5 mm). By preparative gas phase chroma tography, using a 1C fp Carbowax 20M:K0H on 60/80 chram W column 10’ x

±/k" at 220°, a pure sample of 185. could be isolated: n^2’ 8 1*8023 J i r (n ea t) 2 . 96 , 3 .^ 2 , 6 . 17 , 6 . 8 2, 7 . 9 3, 9 ;3 5 , 1 2 .8 0 p; nm (CC14 ) t

8.10 (3H, s), 6 .3 2 (2H, s ) , 6 .2 8 (3H, s ) , 6 .0 3 (2H, broad s), 3 .8 1

(2H, m), 3.18 (TH, m).

Anal. Calcd. f o r CgHi3N0S: C, 58. 9 8} H, 7*15) N, 7*6^.

Found: C, 59*27} H, 7*09} N, 7*75. 1 2 2

Hie sample of 184 was always contaminated with small amounts

of the 5-methoxy isomer: nmr (CCI 4 ) t 8.13 (3H, s), 6.48 (2H, s),

6 .3 6 (3H, s), 6.15 (2H, broad s), 3-88 (2H, m) and 3 .2 5 (IE , m).

3-Methyl-2-(thiomethoxymethyl)aniline (l 6 £) and 5 -Methyl-2-(thiomethoxymethyl) aniline (168). To a vigorously stirred solution of

11.5 g (O.IO75 mol) of m-toluidine in 400 ml of methylene chloride o cooled to -78 under nitrogen, was added dropwise 1 1 .7 g (0.1075 mol) of tert-butyl hypochlorite in 10 ml of methylene chloride (also cooled to -78° in a jacketed constant pressure addition funnel). The

solution turned red and was stirred for 30 min. The funnel was rinsed w ith 15 ml of methylene chloride and 25 ml (ca. 3 eq.) of dimethyl

sulfide was cooled to - 78° , and added "dropwise to the reaction mixture which was then stirred for 3 hr. A solution of 7*0 g (O .13 mol, 1.2 eq. ) of sodium methoxide in 50 ml of methanol was cooled in the funnel and added quickly to the reaction mixture. The cooling bath was allowed to warm to room temperature over night. The reaction mixture was diluted with 100 ml of water, the heterogeneous mixture was separated, and the aqueous layer was extracted twice with 150-ml portions of methylene chloride. The combined organic layers were washed with saturated sodium chloride solution, dried over anhydrous

sodium sulfate, filtered, and concentrated on the rotary evaporator to yield a dark purple oil. Fractionation of the resultant oil gave

4 .8 5 g (0.045 mol) of m-toluidine and 9*40 g (O.O 56 mol) of an approx

imately 1:1 mixture of the two named products for a 5 Ofo y ie ld (90$ 123 yield "based upon unrecovered starting m aterial), "bp 8 7 -90° (0.06 mm).

All attempts to separate these two isomers failed.

3-Chloro-2-(thiomethoxymethyl)aniline (172.) and 5-Chloro-2-(thiomethoxymethyl)aniline ^ solution of 13«7 g (0.1075 mol) of m-chloroaniline in ^00 ml of methylene chloride was vigorously stirred o and cooled to -Jo under nitrogen. To this solution, from a jacketed constant pressure addition funnel was added 11.7 g (0.1075 mol) of o tert -"butyl hypochlorite at -Jo . The dropwise addition required

5 min and the reaction mixture was stirred an additional 35 min. The funnel was washed with 15 ml of methylene chloride and 25 ml (ca.

3 eq.) of dimethyl sulfide was placed in it. The sulfide was added dropwise, and the solution was stirred for 6 hr. A solution of Jo0 g

(0 .1 3 mol, 1.2 eq.) of sodium methoxide in .50 ml of methanol was cooled in the funnel and quickly added to the reaction mixture. The dry ice-acetone bath was allowed to warm to room temperature over night.

The inorganic salts formed were hydrolyzed by the addition of 100 ml of water, the layers were separated, and the aqueous phase was washed with two 150-ml portions of methylene chloride. The combined organic layers were dried over anhydrous sodium sulfate, filtered, and the solvents evaporated to produce a dark purple oil. The oil was fractionally distilled to yield 6 .9 8 g (0.055 mol) of m-chloroaniline and 9.99 g (°*°53 mol) of an approximately 60 :kO m ixture o f 173. and

17 ji fo r a 50 fo y ie ld (87 ^ yield based on unrecovered starting material), bp 128-1300 (0.08 mm). These two isomers could be separated by 124 preparative gas phase chromatography on a ldfJ Carbowax 20M:K0H on

6 0 /80 chromosorb \

1J4: n^5,° 1.61571 ir (neat) 3 . 00 , 3 . 49, 6 .2 4 , 6 . 89, 10. 25, 12.94 p; ram* (CC14 ) t 8.04 (3H, s), 6 .1 5 (2H, s ) , 5.92 (2H, broad s), 2.95-3.85

(3H, m).

Anal. Calcd for C 8H10C1NS: C, 51.19; H, 5*57) N,

Found: C, 51.10; H, 5-39; N, J.bO.

Die second compound eluted was IJis nD5'° 1.8153) ir (neat) 2 . 98,

3,45, 6.29, 6 . 75 , 11.01 p,; nmr (CC14) 8.10 (3H, s), 6.45 (2H, s),

5*95 (?H, broad s), 3 * 08-3 .5 3 (3H, m).

A nal. C alcd. f o r CsHioClDS: C, 51.19; H, 5»37; N, 7.^6.

Found: C, 51.13; H, 5 M ’, N, 7.38.

Preparation of 2-Chloro- 6 -(thicmethoxyme thyl)aniline ( 176 ). Die procedure was identical with that used in the preparation of 3 - and

5-c h lo ro - 2 -(methylthiomethyl)aniline. _In this manner 176 was syn thesized from o-chloroaniline in 26 ?o y ie ld (5*25 g 5 0.029 m ol): n^3*7 1.6160 ( l i t soC n-Q I . 6730 ). Also recovered was 8.68 g (0.068 mol) of o-chloroaniline. Die yield based on unrecovered starting material was 7Offo.

W-Methyl-2-(thiomethoxymethyl)aniline (201). In a 1 1, three-necked flask, equipped with a mechanical stirrer, m s placed 1 0 .7 0 g (0.10 mol) of N-methylaniline and 400 ml methylene chloride. Diis mixture o ms cooled to ca. -70 with a dry ice-acetone bath. To this solution was added dropwise 1 0 .8 5 g (0.10 mol) of tert-butyl hypochlorite in 10 ml of methylene chloride. The resultant dark green solution was s t i r r e d fo r 5 min* 25 ml (ca. 3 ecl*) of dimethyl sulfide was added dropwise while maintaining the exotherm to less than 10° , and the mixture was stirred for 3 hr. At this time 6,h-S g (0.12 mol) of sodium methoxide in 5° elL of anhydrous methanol was added and the solution was stirred for 1 hr. Hie cooling hath was removed, the solution was allowed to come to room temperature, and 200 ml of 10$ aqueous sodium hydroxide was added to the reaction mixture. Hie layers were separated; the aqueous phase m s washed twice with 200-ml portions of methylene chloride. The organic phases were combined, dried over anhydrous magnesium sulfate, and concentrated in vacuo to yield a dark oil. Ihis oil was distilled to yield 1 .7 6 g (0.016*1- mol, 10$) of

N-metliylaniline and 9° 94 g ( 0 . 059*1- mol, 59$5 72 $ based on unrecovered starting material) of 201: bp 8 6 -89° (0 .3 5 mm); n ^3 ' 4 1* 599*1-; i r

(neat) 2.92, 3.41, 6 . 23, 7 . 65 , and 13.41, |x; nmr (CC14 ) t 8.16 (3H, s ) ,

7*20 (3H, s), 6.42 (2H, s), 5*62 (XH, broad s), 2 . 7 O-3 . 6 O (4h, broad m).

Anal. Calcd. for C9H13NS: C, 64 .6 2 ; H, 7 , 8 3 ; N, 8. 37 .

Found: C, 64.64; H, 7*89; N, 8. 36 .

W~Methyl-2-(l-thioethoxyethyl)aniline (202). In a procedure analogous to that used to prepare N-methyl- 2 -(thiomethoxymethyl)aniline, 202 was prepared in 4 jf o y ie ld : bp 130-132° (6 mm); n 22*8 1. 573° 5 (n ea t)

2.9 0 , 3 . 27 , 6.16, 6.54, 7 . 6 O, and 13.40 y,; nmr (CC14) t 8.90 (3H, t ) ,

8.41 (3H, d), 7*73 (2H, q.), 7*17 (3H, s), 6.02 (Iff, q.), 5*21 (iff, broad s), and 2.77-3*64 (4h, m). 126

A nal. Calcd. f o r C1;lH17NS: C, 6 7 . 6 k ; H, 8 .7 7 ; N, 7.17=

Found: C, 67 . 3 2 ; H, 8 .7 0 ; N, 7-05.

N-Methyl-2-(1 -thiopropoxypropyl)aniline (20^). A solution of 5*35 g

(0 .0 5 mol) of N-methylaniline was placed in 200 ml of dry methylene chloride under a static nitrogen atmosphere and cooled to -70°. A solution of 5-^3 g ( 0-05 mol) of tert -butyl hypochlorite in 15 ml of methylene chloride cooled to - 70 ° was added dropwise from a jacketed addition funnel while maintaining the exotherm to less than 10°.

The resultant green solution was stirred for 5 min. A solution of

11.0 ml (O.O73 mol) of di-n-propyl sulfide in 15 ml of dry methylene chloride was cooled to - 70 ° and added dropwise while maintaining the exotherm to less than 10°. The resultant solution was stirred for

3 hr at ca. -70°. The basic rearrangement was effected by addition of

5.^0 g (0 .1 0 mol) of sodium methoxide in 25 ml of absolute methanol.

This solution was stirred for 1 hr at -70°; the bath was removed and the reaction mixture was allowed to warm to room temperature. The reaction was worked up by addition of 200 ml of water, separation of the layers, and washing of the aqueous phases two times with 200-ml portions of methylene chloride. The combined organic phases were dried over anhydrous magnesium sulfate, filtered, and concentrated in vacuo to give a clear oil. Column chromatography on silica gel (ether- hexane eluant) gave 2 .7 7 g (0.012 mol, 2kp) of 20£ as a clear oil: 2 3 • 8 1- 55^9 ; i* (neat) 2 . 90, 3 - 28, 6 . 18, 6 .5 5 , 8. 52, 13.39 p,; nmr

(CC14) t 7.60-9.30 (02H, m), 7-lk (3H, s), 6.21 (lH, t), 5-10 (1H, 127 broad s), and 2 . 7 O-3 .5 8 (hH, m). The second compound eluted was

N-methylaniline (0.012 mol, 2 The y ie ld of 2CQ based on unrecovered starting material was 32$.

Anal. Calcd. for C 13H2iWS: C, 69 .9 0 ; H, 9 -^8 ; N, 6 . 27 .

Found: C, 69.575 H, 9.^3 5 N, 6 . 17 .

2-(2-Tetrahydrothienyl)aniline (196). To a vigorously stirred

solution of 10.0 g (0. IO 75 mol) of aniline in 200 ml of methylene o chloride cooled to -78 under nitrogen was added by means of a jacketed constant pressure addition funnel, 11.7 g ( 0.1075 mol) of te rt-butyl hypochlorite (also cooled to - 78 °). Hi is' solution was stirred for

10 min. The funnel was rinsed with a few ml of cold methylene chloride and hh ml (ca. 5 eq..) of tetrahydrothiophene was cooled to -j8 and added dropwise to the reaction mixture which was stirred for 3 hr. A solution of 50 ml of methanol and 7*0 g ( 0*13 m ol, 1.2 eq.. ) of sodium methoxide m s placed in the funnel, cooled, and added quickly to the reaction mixture. The dry ice-acetone -cooling bath was removed and the solution was stirred for 2 hr. The reaction was diluted by the addition of 100 ml of water, the heterogeneous mixture was separated, and the aqueous layer was extracted two times with 100-ml portions of methylene chloride. The combined organic layers were washed with saturated sodium chloride solution, dried over anhydrous sodium sulfate, and filtered. The solvent was removed in vacuo to produce a dark red oil. Fractional distillation of the oil gave 5*03 g

(0 . 05^ mol) of aniline and 6.11 g ( 0 . 03^ mol) of 196 for a 31$ y ie ld 12 8 based on starting material (64$ yield based on unrecovered starting material). The properties of 196 were: bp 130-134° (0.19 mm); n 26.8 1. 6 2 5 8 ; ir (neat) 2 . 92, 3 * 5 6 , 6 . 12, 6 . 70 , and 13*40 pi; nmr

(CC14) t 7.82 (4h, m), 6.99 (2H, m), 6.06 (2H, s), 5*52 (2H, t),

3 .2 0 (4h, m).

Anal. Calcd. for C 10H13NS: C, 66.99* H, 7*31; N, 7°8l.

Found: C, 67 .0 4 ; H, 7*29| N, 7*74.

2-(2-Tetrahydrothlopyranyl)aniline (l^X). In a 1 j I , three-necked flask equipped with a mechanical stirrer was placed 9*30 g (0.10 mol) of aniline in 400 ml of methylene chloride. This solution was cooled to -70° by means of dry ice-acetone bath. A solution of 10 .85 g

(0.10 mol) of tert -butyl hypochlorite in 25 ml of methylene chloride was added dropwise to the aniline solution. Hie resultant N-chloroaniline solution was stirred for 5 rain and 25 ml (ca. 3 eq.) of tetrahydrothiopyran was added at such a rate as to maintain the exotherm at less than 10°. This dark solution was stirred for 4 hr, 25 ml of triethylamine was added dropwise, and the solution was stirred for o r o 1 hr at -70 to -60 . Ihe cooling bath was removed and the solvents were removed in vacuo. This yielded a dark residue which was refluxed overnight in 200 ml of acetcnitrile containing 25 ml of triethylamine.

Qhe solvents were removed by rotary evaporation, 200 ml of 1C$ aqueous sodium hydroxide and 200 ml of methylene chloride were added, the layers were separated, and the aqueous phase was washed with two

200-ml portions of methylene chloride. . Ihe combined organic phases 129 were dried over anhydrous magnesium sulfate, filtered, and the solvents removed try rotary evaporation to produce a dark oil. D istillation of

this oil yielded 0 .5 6 g (0.0060 mol) of aniline and 7*83 g (0.0^06 mol,

k b ° [o "based on unrecovered starting material) of 157 , t>p 139-1^0°

(0.10 mm). An analytical sample was prepared by column chromatography

on silica gel using hexane as an eluant: n ^4 *3 1. 6 1 0 7 ; ir (neat) 2 . 9 2,

3.36, 6.12, 60 76, 7 .9 b , 13.38; nmr (CC14) 7 . 7 O-8 .7 5 (6h, m), 7.03-

7.52 (2H, m), 6.20 (1H, m), 6.04 (2H, s), 2 . 8O-3 .5 9 (4h, m). This

sample solidified in the refrigerator: mp 65 . 5-67 °;. "the original

distillate had also solidified.

Anal. Calcd. for CnHisKS: C, 6 8 . 3 5 ; H, 7 .8 2 ; N, 7.25.

Found: C, 68.1f6; H, 7-93 J N, 7-3^. a h-AUylaniline (19|5). In a 5°0 ml, three-necked, round-bottomed flask

equipped with a low temperature thermometer, jacketed addition funnel,

and a gas inlet tube was placed 1.86 g (0.02 mol) of aniline. Ihe

apparatus was cooled to - 78 °. In the jacketed addition funnel was p laced 2 .1 7 g (0 .0 2 mol) of te rt-butyl hypochlorite and 20 ml of methylene chloride. This solution was cooled to - 78 ° and added dropwise to the aniline solution. The resultant yellow solution was

stirred for 10 min. Ihe addition flannel was washed with 20 ml of methylene • chloride, and charged with 2 . 6 ^- ml ( 1 .2 eq.) of allyl methyl

sulfide and 20 ml of methylene chloride. This solution was cooled

to -"{8° and added dropwise to the N-chloramine solution. The resultant

solution was stirred for ^ hr at -78°. A solution of 1.30 g (1.2 eq.) 130

of sodium methoxide in 20 ml of dry methanol was added to the reaction mixture. This solution was stirred 1 hr at -78 , the cooling bath was removed, and the reaction was allowed to warm to room temperature.

To the reaction mixture was added 100 ml of 1C$ aqueous sodium hydroxide, the layers were separated, and the aqueous phase was extracted two times with 100-ml portions of methylene chloride, ihe combined organic phases were dried over anhydrous magnesium sulfate, filtered, and concentrated in vacuo to produce a yellow oil. Dis tillation of the yellow oil gave 1.97 g (0.0148 mol) of 1§£, bp

122-126 (2 .8 5 mm), fo r a 7 hfo y ie ld ; n ^5*0 1.5612 ( l i t 97 np 1. 56 l 4 ).

(97) J . E. Hyre and A. R. Bader, J. Arner. Chem. S o c ., 80, 437 (1958).

The ir and nmr spectra were consistent with the assigned structure.

Benzenesulfenanilide (192.) from n-Butyl Ehenyl Sulfide. In a vigorously stirred solution of 300 ml of acetonitrile and 100 ml of methylene chloride was placed 1 .8 6 g (0 .0 2 mol) of aniline and 4.98 g

(0.03 mol) of n-butyl phenyl sulfide. This solution was cooled to

-45 to -40 . A solution of 3*26 g ( 0 .0 3 mol) of tert -butyl hypo chlorite in 10 ml of methylene chloride was added dropwise to the solution and stirred for 3 hr. A solution of 2.16 g (0.04 mol) of sodium methoxide in 20 ml of absolute methanol was added slowly to the solution and stirred for 3° min. The cooling bath was removed, and the reaction mixture was allowed to warm to room temperature. The solvents were removed at reduced pressure, and 100 ml of water and 151

100 ml of methylene chloride we re added to the residue. Die layers were separated, the aqueous phase was washed two times with 100-ml portions of methylene chloride, and the combined organic extracts were dried over anhydrous magnesium sulfate. The solution was filtered, the solvents were removed on the rotary evaporator, and the resultant red oil was refluxed in 100 ml of anhydrous toluene with 5 ml of triethylamine for 2 hr. Upon removal of the solvents, the dark oil was chromatographed on silica gel with pentane-ether as eluant to yield 2.8l g (0.01*1- mol, JC$) of 192, mp 57-59° (lit98 53-55°). Ihe

(98) H. Lercher, F. Holschneider, K. Koberle, W. Speer, and P. Stocklin, Chem. Ber., _58, 409 (1935 )•

8 ir and nmr spectra were consistent with the proposed structure.

Benzenesulfenanilide (192) from Phenyl Isopropyl Sulfide. Benzene- sulfenanilide (192) was prepared in 76 $ yield using phenyl isopropyl sulfide in a manner similar to the one described using n-butyl phenyl sulfide except that the period of stirring at -*i-0o was for hr and the reflux period was for 4 hr. The ir and nmr spectra were identical to those of 192 prepared from n-butyl phenyl sulfide.

W-2 Raney Nickel. The W-2 Raney nickel used in these experiments was obtained from W. R. Grace & Co., Raney Catalyst Div., So. Pittsburg,

Tenn. as No. 26 Raney Active Nickel Catalyst in Water. A portion of this was placed in a beaker and washed with distilled water until neutral to pH paper and then several more times with distilled water, three times with 95$ ethanol, and three times with absolute ethanol. 1 5 2

The catalyst under absolute ethanol was stored in brown bottles until use. This material was identical in its reduction of sulfides to

66 W-2 Raney nickel produced by the method of Mozingo. o-Toluidine (l 6 l). To a vigorously stirred solution of 2.50 g

(0.0163 mol) of 2 -(thimethoxymethyl) aniline in 100 ml absolute 1 ethanol was added ca. 30 g (10 level tsp) of W-2 Raney nickel. This solution was stirred for 30 min at room temperature. The Raney nickel was removed by filtration and washed five times with 100-ml portions of absolute ethanol. The ethanol was removed at reduced pressure.

The resultant residue was taken up in 200 ml of methylene chloride, and dried over magnesium sulfate. The drying agent was removed by filtration, and the solution was concentrated by rotary evaporation to yield a clear oil. Distillation of this oil gave 1.10 g (0.0103 mol,

63 $ yield) of 161, n^2' 8 1.5680 (lit 99 n^° 1 . 5688). The o-toluidine

(99) H. G. Tanner and P. A. Lasselle, J. Amer. Chem. Soc., kS, 2165 (1926 ).

produced was identical in all respects to an authentic sample.

4-Chloro-o-toluidine (20£). To a vigorously stirred solution of

1.10 g (O.OO59 mol) of 4-chloro-2-(thiomethoxymethyl)aniline in

30 ml of absolute ethanol cooled to 0° was added ca. 9 g (3 tsp) of

W-2 Raney nickel. After vigorously stirring this mixture for 10 min, the Raney nickel was removed by filtration, and washed twice with 133 30-ml portions of methylene chloride. Ihe solvents were removed by rotary evaporation. Ihe resultant residue was taken up in 25 ml methylene chloride and dried over anhydrous magnesium sulfate. The drying agent was removed by filtratio n and washed with two 50-ml portions of methylene chloride. Ihe solvents were removed at reduced pressure to yield 0 .6 6 g of a mixture of 20£ and h3lo Ihis mixture was analyzed by vapor phase chromatography using a 10p Carbowax

20M:KOH on 60/80 chromosorb W column at 150 , which showed the mixture to be 9°$ 20£ and 1C$ l6 l . The y ie ld of 20£ (0. 59^ g, 0.00if2 mol) was 72 $ and the yield of l 6 l (0.066, 0.00062 mol) was 10p.

^-Carboethoxy-2-methylaniline (207). To a vigorously stirred solution of 3 .6 7 g (O.OI63 mol) of h carboethoxy-2-(thiomethoxymethyl)aniline in 100 ml of absolute ethanol was added ca. 30 g (10 tsp) of ¥-2 Raney nickel. Ihis mixture was stirred for 30 min at room temperature.

The Raney nickel was removed by filtration, washed twice with 100-ml portions of absolute ethanol and twice with 100-ml portions of methy lene chloride. The combined washings were concentrated at reduced pressure to leave a clear oil. Ihe oil was taken up in 100 ml of methylene chloride and dried over anhydrous magnesium sulfate. The magnesium sulfate was removed by filtratio n , and the solution was concentrated at reduced pressure to give a clear oil which solidified upon standing. This procedure gave 2.66 g (0.01^3 mol, 8850) of 20J: mp 76-77°; 5r (KBr) 2 . 92, 5-90, 6 . 21, 7-72, 7 . 88, 8. 36 , 13.00 nmr

(CDCI3 ) t 8.67 (3H, t), 7.85 (3H, s), 5.96 (OH, broad s), 5-69 (2H, q .) 9 1 3 4

2.12-3.50 (3H, nO* Ihis compound is in the literature . 100 Anal. Calcd. for CloHi 3N02: C, 67.02; H, 7.31; N, 7.82.

Pound: C, 66 . 8 9; H, 7 .2 4 ; N, 7 . 87 .

(100) F. J. Viliani, U. S. Patent 2,764,519 (1958), Chem. Abstr., 51, p4443c (1957).

2,6-Xylidine (£04). 2,6-Xylidine, n 24' 4 1.5598 (lit 101 n ^4 *75 1-5615), was produced from 6 -m ethyl- 2 -(thicmethoxymethyl)aniline in 66$ y ie ld

(101) K. von. Auwers, Justus Liebigs Ann. Chem., 422, 160 (1921).

in a fashion analogous to the procedure used to prepare 4-carboeth oxy -2 -methylaniline.

2 -Methyl-p-anisidine (£06). 3n a procedure identical to that used to prepare 4-carboethoxy- 2 -methylaniline, 206 , n ^ '’ 1.5692 ( l i t np° 1.5647), was prepared in 56jo yield from 2 -(thiomethoxymethyl) - p-anisidine.

(102) R. C. Weast, Ed., “Handbook of Chemistry and Fnysics,” 50th ed., Chemical Rubber Pub. Co., Cleveland, Oh., 1969, p C-514.

2,H-Dimethylani 1 ine (210). By a method similar to the cne used to prepare 4-carboethoxy- 2 -methylaniline, £ 10, n 24*4 1.5622 ( l i t 103 135

n^° 1.5649), was prepared in 72$ yield from 201.

(103) H. Ley and G. Pfeiffer, Chem. Eer., 363 (1921).

2-Ethyl-N-methylaniline (211). 2-Ethyl-N-methylaniline, n ^5*1

1.5538 (lit6sa n^° 1 . 5553)5 was prepared from 202 in 69$ yield by

a procedure analogous to that used to prepare tocarboethoxy- 2 -

methylaniline.

2-n-Butylaniline (20S). 2-n-Butylaniline w s produced from 196 in 62 $

yield in a manner analogous to the procedure used to prepare o-tolu

id in e ; 208 gave spectral data, nmr and ir, consistent with the 0 structure: nfj3,4 1.5362. Ihe hypochloride was prepared, mp 139-

lkO° ( l i t 104 137 ° ) .

(lO^f) R. R. Read and D. B. M ullin, J . Amer. Chem. S o c ., 1763 (1928).

2 -n-Pentylaniline (20^). By a procedure identical to that used to

p r e p a r e k-carboethoxy -2 -m ethylaniline, 20£ was prepared from lgX in

68 $ y i e l d , bp 80° (9 mm)J n ^ 2 *4 1.5292; i r ( n e a t ) 2.86, 3«355 6.13,

• 6.614-, 1 3 .3 7 M-J (CCI4 ) t 8. 2 1 -9 .2 5 (9H, m), 7^0-7.75 ( 2H, m),

6 . 6 ^ (2H, s ) , 2.89-3.62 (to, m).

Anal. Calcd. for cu Hi 7 N: C, 8 0 .9 2 ; H, 1 0 . 5 0 ; N, 8. 58.

Found: C, 80.86; H, 10.35; N? 8.62. 1 3 6

4-Me thoxy-2-(thiomethoxymethyl) aniline (l8y) from 22J. A solution

2»76 g (0.02 mol) of N-chlorosuccinimide in 75 ml of dry methylene chloride was cooled to 5°j 2.22 ml (O.OJ mol) of dimethyl sulfide

in 10 ml of dry methylene chloride was added dropwise while keeping the temperature below 10°. A voluminous white precipitate formed.

The reaction mixture was stirred for 30 nun a t ca. 5°« A so lu tio n of 2.46 g (0.02 mol) of p-anisidine in 10 ml of dry methylene chloride was added dropwise while maintaining the temperature at less than 10°.

Ihe clear solution was stirred for 10 min at 50j warmed to room temperature, and stirred for 1 hr. A solution of 1.62 g (0,0j5 mol) of sodium methoxide in 15 ml of absoltue methanol was added dropwise

(2° endotherm). Ihis solution was stirred for 1 hr. The reaction was quenched by the addition of 100 ml of water, the phases were separated, the aqueous phase was extracted two times with 100-ml portions of methylene .chloride. The combined organic phases were dried over anhydrous magnesium sulfate, filtered, and concentrated in vacuo to produce a dark oil. Column chromatography on silica gel

(ether-hexane eluant) followed by distillation of. the resultant oil gave 1.97 g (0.011 mol, 55?) of 18j: np3*4 1*6011 (a previously prepared sample had n^4'4 1*5965); hp 91-95° (0.l4 mm). The ir and nmr spectra were identical to those of previously prepared samples by the K-chloroaniline method.

Anal. Calcd. for C 9H13NOS: C, 58.93; H, 7*15} N, 7*64.

Found: C, 58. 8 8; H, 7*15; N, 7*62. 157

Preparation of l 8j from 126. In a graduated test tube was condensed

2.0 ml (O.Okk mol) of chlorine at -78°; 10 ml of dry methylene chloride

was added to the chlorine; the solution was allowed to warm slightly,

stirred with a spatula, and poured into 100 ml of dry methylene o chloride at -78 which was maintained under a static nitrogen atmos

phere. Ihe graduated test tube was washed with an additional 10 ml

of methylene chloride, and this was added to the chlorine-methylene

chloride mixture. Ihe temperature of the pale yellow solution was o allowed to go back to ca. -70 , and 3-90 ml (0*05 mol) of dimethyl

sulfide in 10 ml of dry methylene chloride was added. Ihe exotherm o was kept to less than 5 5 ^nd the yellow color had dissipated upon

completion of the addition of the dimethyl sulfide solution. Ihe

solution was stirred for 5 min and a solution of 2 .7 6 g (0 .0 2 mol) of

p-anisidine and 2 .8 0 ml (0 .0 2 mol) of triethylamine in 10 ml of dry

methylene chloride was added dropwise. Ihe resultant purple solution o was stirred for k hr at -70 , and 3 »2k g (0.05 mol) of sodium methoxide

in 15 ml of absolute methanol was added dropwise to the reaction

mixture. Ihe solution as allowed to warm to room temperature over

night (ca. 16 hr). The reaction was diluted with 150 ml of 10$

aqueous sodium hydroxide; the layers were separated; the aqueous phase

was extracted two times with with 1 00-ml portions of methylene chloride;

the organic layers were combined, dried with anhydrous magnesium

sulfate, filtered, and concentrated in vacuo to yield a dark oil.

Column chromatography on silica gel (ether-hexane) gave 2.2k g

• (0.012k mol) of 132: bp 93-100° (0.16 mm), 6 2$, ng3 *6 1.5998 138

(n ^3 *4 1.6011 for an analytical sample). The ir and nmr spectra were identical to those of previously prepared samples.

Preparation of 165 from 128. 4-Methyl-2-(thiomethoxymethyl)aniline

(163 ) was prepared in 54$ yield after column chromatography on silica gel (ether-hexane elucant) and distillation [bp 9 0-95° (0 .0 3 ram)] in a procedure identical to that used in the preparation of 160 frcm 128. The distilate solidified upon standing (mp 45-47 ° 5 lit8°C mp 42-45°). Hie ir and nmr spectra were identical to those of previously prepared samples.

Preparation of 160 from 128. Chlorodimethylsulfonium chloride was prepared as described in the preparation of l 8j. A solution of 1.86 g

(0 .0 2 mol) of aniline and 2 .8 0 ml (0 .0 2 mol) of triethylamine in 10 ml of dry methylene chloride was added dropwise to the clear solution of chlorodimethylsulfonium chloride. A white precipitate formed after _ o ca. 1 hr. The reaction mixture was stirred for 6 hr at -70 . Hie reaction mixture was allowed to warm to room temperature overnight

(ca. 16 hr). A work-up procedure identical to that used in the preparation of 187 gave a dark oil. Hiis oil was refluxed for 5 hr in 100 ml of acetonitrile containing 5 nil of triethylamine. Hie solvent was removed in vacuo; the resultant oil was chromatographed on silica gel (ether-hexane) to produce 2.04 g (O.OI 33 mol, 67 ^) of

160, bp 80-82° (0.08 ram), n ^1 *8 1.6094 (n^4*° I. 6 O83 for a previously prepared sample. Hie ir and nmr spectra were identical to those of previously prepared samples. 1 5 9 Preparation of 1J1 from 128. Chlorodimethylsulfonium chloride was prepared as described in the preparation of l8j. A solution of 2.55 g

(0.02 mol) of 4-chloroaniline and 2.80 ml of triethylamine in 10 ml of dry methylene chloride was added dropwise to the clear solution of

128. The resultant solution was stirred for 8 hr at ca. -J0°. A solution of 5*2^ g (0.06 mol) of sodium methoxide in 15 ml of absolute methanol was added, the reaction mixture was allowed to warm to room temperature, and the reaction mixture stirred overnight. The reaction was diluted with 100 ml of 10$ aqueous sodium hydroxide; the layers were separated, and the aqueous phase was washed with two 100-ml portions of methylene chloride. The combined organic phases were dried over anhydrous magnesium sulfate, filtered, and concentrated in vacuo to give a red oil. This oil was taken up in 100 ml of acetonitrile containing 5 ml of triethylamine, the solution was refluxed for 18 hr. The solvents were removed in vacuo to yield a red oil which solidified upon standing. Column chromatography on silica gel (ether-hexane eluant) gave 1.68 g (0.009 mol) of 1£L, ^5$9 mp 76 - 78 ° (the analytical sample previously prepared had a mp 78 -80° ) .

The ir and nmr spectra were identical to those of previously prepared sam ples.

Preparation of 182 from 128. In a graduated test tube was condensed

2 .0 ml (0.0li4 mol) of chlorine at - 7 8 ° (dry ice-acetone); 10 ml of dry methylene chloride was added to the chlorine; the solution was allowed to warm slightly, stirred wtih a spatula, and poured into a solution of 75 nil of acetonitrile and 25 ml methylene chloride at o -50 to -bO which was maintained under a static nitrogen pressure.

The graduated test tube was washed with an additional 10 ml of

methylene chloride, and this was added to the solution. A solution of

3 .9 0 ml (0 .0 5 mol) of dimethyl sulfide in 10 ml of dry methylene o chloride was added, ihe exotherm was kept to less than 5 "by drop-

wise addition. Ihe yellow color had dissipated upon completion of

the addition of the dimethyl sulfide solution. Hie solution was then

stirred for 5 min, and a solution of 2 .7 6 g (0.02 mol) of p-nitro-

aniline and 2.80 ml (0.02 mol) of triethylamine in 20 ml of acetoni

trile and 20 ml of methylene chloride was added dropwise while main-

. o taining the temperature to -40 or less. A voluminous precipitate

formed and the reaction mixture was stirred for 9 hr while maintaining o the temperature between -50 and -M3 . A solution of 3 . 2b g (0.06 mol)

of sodium methoxide in 15 ml of methanol m s added, the cooling bath was removed, and the solution was stirred overnight. The reaction was diluted by the addition of 100 ml of 1C$ aqueous sodium methoxide,

the layers were separated, and the aqueous phase was extracted two

times with 100-ml portions of methylene chloride. The combined organic

phases were dried over anhydrous magnesium sulfate, filtered, and

concentrated in vacuo to produce a yellow solid. This solid was taken

up in 150 ml of dry toluene containing 5 ml of triethylamine and the

solution was refluxed for ^8 hr. Hie solvents were again removed

in vacuo to yield a yellow solid. Column chromatography on silica gel 141

(ether-hexane eluant) gave 1.24 g (O.OO 63 mol, 31$) of 182,, mp 75 -75 °

( l i t 800 mp 75 -77 °)* The ir and nmr spectra were identical to

previously prepared samples. Also, I .3 0 g (0.0094 mol, 42$) of

4-nitroaniline was recovered.

Preparation of 1J8 from 128. Chlorodimethylsulfonium chloride was

prepared as described in the preparation of 182. A solution of 5 . 3O g

(0 .0 2 mol) of benzocaine and 2 .8 0 ml (0 .0 2 mol) of triethylamine in

in 10 ml of dry methylene chloride was added dropwise to the clear

solution of 128o A voluminous white precipitate formed, and this mixture was stirred for 6 g- hr. At this time, 8.40 ml (0.06 mol) of

triethylamine was added to the reaction mixture, and the cooling bath was removed. The solution was stirred overnight. A work-up procedure

identical to that used in the preparation of 1§2 gave a dark red oil which solidified upon standing. Hiis oil was taken up in 100 ml of

acetonitrile containing 5 ml of triethylamine and refluxed for 48 hr.

The solvents were removed in vacuo to leave a red solid. Recrystalli

zation from absolute ethanol gave 1.60 g ( 0.0071 mol, 39$) °T 12&? nip 83-84° (an analytical sample previously prepared had a mp 84.5-

85°5°)* The ir and nmr spectra were identical to those of previously prepared samples.

Preparation of 160 from 231. A solution of 3*20 g (0.02 mol) of bromine in 100 ml dry methylene chloride was cooled to - 7 $ (dry ice-acetone) under a static nitrogen pressure. To this dark red solution was added 1.55 ml (0 .0 2 mol) of dimethyl sulfide in 10 ml of methylene chloride. A yellow precipitate formed immediately (Note; the yellow color persists even if another equivalent of dimethyl sulfide is added); The resultant solution was stirred for 3° min.

A solution of 1.86 g (0.02 mol) of aniline and 2.80 ml (0.02 mol) of triethylamine in 10 ml of dry methylene chloride was added dropwise o while maintaining the exotherm at less than 5 • At the end of the addition, the reaction was clear with no red color, The solution was stirred for 6 hr, and no precipitate was observed. A solution of 3*24 g (0.06 mol) of sodium methoxide and 15 ml of absolute methanol was added, the solution was allowed to warm to room temperature, and the reaction mixture was stirred overnight (ca. 16 hr). The reaction mixture was diluted with 100 ml of lOfo aqueous sodium hydroxide; the layers separated, and the aqueous phase was washed with two 100-ml portions of methylene chloride. The organic phases were combined, dried over anhydrous magnesium sulfate, filtered, and concentrated in vacuo to yield a yellow oil. This oil was taken up in 100 ml of acetonitrile containing 5 ml of triethylamine and refluxed for 5

The solvent was removed in vacuo, and the resultant oil was chromato graphed on silica gel (ether-hexane eluent) to produce 2.10 g (0.0137 mol, 6 <~Pp) of 160, bp 8 8-90° (0.12 mm), n^4'8 1.6093 (lit800 n^

1.6042 vs. a previously prepared sample n^4*8 I . 6085 ). The ir and nmr spectra were identical to those of previously prepared samples.

Preparation of 1§6 from 233.* Bromotetramethylenesulfonium bromide

(233) 'was prepared in a procedure identical to that used in the preparation of 231. 2-(2-Tetrahydrothienyl)aniline was prepared in 1^3

19$ yield, in a procedure identical to that used in the preparation

of 160 from 2^1 except that the acetonitrile-triethylamine reflux time was overnight (ca. 16 hr).

Preparation of 1§6 from 2^2. Chlorotetramethylenesulfonium chloride was prepared in a procedure identical to that used in the preparation

of 128. 2-(2-Tetrahydrothienyl)aniline was prepared in 2C$ yield,

in a procedure identical to that used in the preparation of 160 from

chlorodimethylsulfonium chloride except that the acetonitrile- triethylamine reflux time was overnight (ca. 16 hr). After column

chromatography on silica gel (ether-hexane eluant) and distillation,

lg6 was obtained, bp 1 02-105° (0.08 mm), n^2-2 1.6251 (an analytical

sample previously prepared had n^6"8 I 06258 ). The i r and nmr sp e c tra were identical to those of previously prepared samples0 REFERENCES

1. For a review of carbonium ions and their chemistry see G. A. Olah and P. v.R. Schleyef, eds. , “Carbonium Ions,” Interscience, New York, N. Y., 1968.

2. p. G. Gassman, Accounts Chem. Res., 26 (1970).

5. S. T. Lee and K. Morokuma, J. Amer. Chem. Soc., 9^, 6869 (1971)*

4. S. Y. Chu, A. K. Q. Siu, and E. F. Hayes, ibid., £4, 2969 (1972).

5. G. F. Koser, Chem. Comm., 461 (1973)*

6. J. F. Harrison and C. W. Eakers, J. Amer. Chem. Soc., 3467 (1973). 7o J. Stieglitz and P. N. Leech, Chem. Ber., 46, 2147 (1913)5 J. Stieglitz and P. N. Leech, J. Amer. Chem. Soc., ;?6, 272 (1914); J. Stieglitz and B. A. Stanger, ibid., 3 8, 2046 (191§j.

8. R. T. Conley and H. Brandman, unpublished results.

9. M. S. Newman and P. M. Hay, J. Amer. Chem. Soc., 2J22 (1933)*

10. R. A. Abramovitch and B. A. Davis, Chem. Rev., 64, 149 (1964).

11. For a review of the Beckmann rearrangement see P. Smith in “Molecular Rearrangements,” P. Mayo, ed., Vol. 1, Interscience, New York, N. Y ., 19&3j PP 484-507.

12. P. Lansbury and N. Mancuso, Tetrahedron Letters, 2445 (1965).

1 3 . P. Lansbury, J. Colson, and N. Mancuso, J. Amer. Chem. Soc., 86, 5225 (1964).

14. W. D. Emmons, i b i d . , 79> 5739 (1957)<■

15* P. G. Gassman and B. L. Fox, Chem. Commun., 153 (1966); p. G» Gassman and B. L. Fox, J. Amer. Chem. Soc., 89., 338 ( 1967 ).

16. P. G. Gassman and R. L. Cryberg, ibid., 50, 1355 (1968); P. G. Gassman and R. L. Cryberg, ibid., ^1, 2047 (1969).

144 Iks

IT* P. G. Gassman and R. L. C ryberg, J . 'Amer. Chem. S o c., 91, 5176 (1969). ------

18. A. G. Anastassiou, ibid., 88, 2322, ( 1966 ); C. D. Dijkgraaf and G. J. Hoijtink, Tetrahedron Suppl., 2, 179 (1963 ).

19- P. G. Gassman and A. Carrasquillo, Chem. Commun., Jj-95 (1969); P. G. Gassman and A. Carrasquillo, Tetrahedron Letters, 109 (1971).

20. C. H. DePuy, L. G. Schnack, J. W. Hauser, and W. Weidemann, J . Amer. Chem. S o c ., 8 7 , 4006 (1965 ).

21. P. v.R. Schleyer, G. W. VanDine, U. Schollkopf, and J. Paust, ibid., 88, 2868 ( 1966 ); U. Schollkopf, F. Fellenberger, M. Patsch, P. v.R. Schleyer, T. Su, and G. W. YanDine, Tetrahedron Letters, 3639 (1967); P« v.R. Schleyer, T. Su, M» Saunders, and J. C. Rosenfield, J. Amer. Chem. Soc., jgl, 517^ (1969).

22. R. B. Woodward and R. Hoffman, i b i d . , 87 , 395 (1965)- Also see H. C. Longuet-Higgins and E. W. Ahrahamson, ib id ., 87 , 20^5 (1965 ).

23. s. J. Erois, ibid., $0, 506 , 508 (1958).

2ko P. G. Gassman and D. K. Dygos, ibid., £1, 15^+3 (1969); P. G. Gassman, D. K. Dygos, and J. E. Trent, ibid., 203k (197°).

25. p„ G. Gassman, F. Hoyda, and J. Dygos, ibid., ^0, 2716 (1968 ); P. G. Gassman and J. Dygos, Tetrahedron Letters, ^7^5 (1970).

26. D. C. H orw ell and C. W. R ees, Chem. Commun., 1^28 ( 1969 ).

27 . 0. E. Edwards, D. Vocelle, J. W. ApSimon, and F. Hague, J. Amer. Chem. S o c ., 87 , 678 (1965).

280 P. Kovacic, J. -H. Liu, E. M. Levi, and P. D. Roskos, ibid., 2 2 , 5801 (1971 ).

29. R. D. Fisher, T. D. Bogard, and P. Kovacic, ibid., 36^6 (1973).

30. T, A. Kling, R. E. White, and P. Kovacic, ibid., 2lb> 7kl6 (1972).

31. P. G. Gassman and K. Shudo, ibid., 22.9 5899 (1971).

32. p. G. Gassman, K. Shudo, R. L. Cryberg, and A. B attisti, Tetra hedron Letters, 875 (1972).

33- P. G. Gassman and G. D. Hartman, J. Amer. Chem. Soc., j97, ^ 9 (1973). l4 6

34. P» G. Gassman. Go A. Campbell, and R. Co Frederick, J. Amer. Chem. Soc., 94, 3884 (1972) ; P. G. Gassman and G. A. Campbell, Ibid.', 9jf, 5§9i (1972).

35« R« S. Neale, R. G. Schepers, and M. R. Walsh, J. Qrg. Chem. , 2£, 3390 (196*0; P. Haberfield and D. Paul, J. Amer. Chem. Soc., 8x , 5502 (1965).

3 6 . C. Me chebynck-David and P. J. C. Fierens, Tetrahedron, 6 , 232 (1959).

37* For a review of this subject see J. A. M iller, Cancer Research, 559 (I 970 )o

3 8. J. D. Schribner and N. K. Naimy, ibid., 22, 1159 (1973)*

39» Fo G. Mann and W. J . Pope, J . Chem. S o c., 121, IO 52 (1922).

40. R. Appel, Wo Buchner, and E. Guth, Justus Liebigs Ann. Chem., 613, 33 (1958); R. Appel and W. Buchner, Angew. Chem. , J l , 701 TI9 5 8), Chem. Ber., 95, 849, 855, 2220 (l9o2j.

4l<, A. Bell and M. B. Knowles, U. S» ,Patent 2 , 692,287 (1956); Chem. A bstr., ^O, 2666e (1956).

42. H. Suhr, Justus Liebigs Ann. Chem., 68X, 175 ( l9 6 l) .

43* For a preliminary report of this work see P. G. Gassman, G. Gruetzmacher, and R. H. Smith, Tetrahedron Letters, 7-97 (1972).

44. Dr. Richard H. Smith, Jr., performed some of the initial experi ments on the chemistry of azasulfonium salts. This author thanks Dr. Smith for his assistance in this work.

7-5. The basis for attempting the synthesis in this way w ill be discussed in Part III of Results and Discussiono

46„ L. Horwitz and C. A. Clark, U. S. Patent 3?102,l42 ( 1963 ); Chemo A b s tr ., 60, 1760e (i 9 6 0 ).

47 o M. T, Bogert and L. Smidth, J. Amer. Chem. Soc., j?0, ^28 (192 8 ).

48. M. Schubert, Justus Liebigs Ann. Chem., -1-0 (1947)°

49. R. G. Pearson, II. Sobel, and J. Songstad, J. Amero Chem. Soc., 2 0 , 319 (1968). 1 4 7 50. C. G. Swain and L. E. Kaiser, J. Amer. Chem. Soc., 80, 4089 (1958) and references therein.

51. E. D. Hughes, C. K. Ihgold, and Y. Pocker, Chem. Ind. (London), 1282 (1959).

52. Y. Pocker and A. J. Parker, J. Org. Chem., 3 1 , 1526 (1966).

53* J* Bronsted, Z. Physik. Chem., 102, 169 (1922); W. Bjerrum, ibid., 108, 82 (1924).

54. For general discussions of ionic reactions see W. J. Moore, “Physical Chemistry , ” 3rd ed., Prentice-Hall, Inc., Englewood Cliffs, N. J ., 1962, Chapter 9; K. J. Laidler, “Chemical K inetics,” 2nd ed., McGraw-Hill, Hew York, H. Y., 1965, Chapter 5*

55* H. C. Brown and Y. Okamoto, J . Amer. Chem. S o c ., 80, 4979 (1958).

5 6 . Dialkoxydialkylsulfuranes have recently been synthesized and stu d ie d , R. J. Arhart and J. C. M artin, J. Amer. Chem. S o c ., 4997, 5003 (1 9 7 2 ) and references therein.

57* G. H. Wiegand and W. E. McEwen, J. Org. Chem., 23 j 2671 (1968).

58. Y. H. Kim and S. Oae, Bull. Chem. Soc. Jap., 42, 1968 (1969).

59* B. Mo frost, R. LaRochelle, and R. C. Atkins, J. Amer. Chem. Soc., 3 1 , 2 1 7 5 (1969)*

60. D. C. Owsley, G. K. Helmkamp, and M. F. Rettig, ihid., 91, 5239 (1969)*

61. M. L. Moore and T. B. Johnson, ibid., 57, 1517 (1935); ibid., 2J, 2234 (1935); ibid*, 109171936); ibid*, . 58, 196o“Tl936).

62. F. A. David, R. B. Wetzel, T. J. Devon, and J. F. Stackhouse, J. Org. Chem., ^6, 799 (1971)*

6 3 . I. B. Douglas, ibid., 24, 2004 (1959)°

64. M. Sommelet, C. R. Acad. Sci., 202, 56 (1937); G. C. Jones and C. R. Hauser, J. Org. Chem., 2J, 3572 (1962); G. C. Jones, W. Q. Beard, and C. R. Hauser, ibid., 23, 119 (1963)* For a review of the Sommelet-Hauser rearrangement see JI. J. Shine, “Aromatic Rearrangements,” Elsevier Pub. Co., Hew York, N. Y., 1967, PP 3 1 6 -326 . 148

65 . Co R. Hauser, W. W. Kantor, and W. R. Brasen, J. Amer. Chem. Soc., 7£, 2660 (1953.).

66. R. Mozingo, Org. Synthesis, Coll. Vol. l8l ( 1 9 5 5 ).

67 . C. F rie d e l and J . M. C ra fts , C. R., Acad. S c l ., 84, 1450 ( 1877 ). For a detailed account of the Friedel-Crafts reaction see G. A. Olah, “Friedel-Crafts and Related Reactions,” Interssience, New York, N. Y., 1963 -1965 .

68. (a) For a review see R. Stroh, J. Ehersberger, H. Haberland, and W. Hahn in Viewer Methods of Preparative Organic Chemistry,” 1963s PP 227-252. This article also appeared in Angew. Chem., &9, 669 (1957); (b) G.- G. Ecke, J . P. Napolitano, A. H. F ilb e y , and A. J. Kolka, J. Org. Chem., 22, 639 (1957).

69 . H. Sanielevici, L. Floru, and M. Cibu, Rev. China. (Bucharest), 14, 673 (1963); Chem. Abstr., 61, 16195b (1964). "

70. A.-G. Kalle, British Patent 977,399 (1964); chem. Abstr., 62- 8570d (1965). 71. E. Wainer, U. S. Patents 3,042,515 and 3,042,516; chem. Abstr., j)8, 12100e (1963). 72. M. Iwama, K. Sakamoto,. I. Inoue, Y. Takei, and T. Endo, German Patent 1,959,594 (1971); Chem. A bstr., 7 5 , 124989X (1971). P. Ramello, German Patent 2,028,601 ( 1970 ) ; Chem. A b s tr ., JJ5, 22505 (l97l)o

73» M. T. Harvey, U. S. Patent 3,207,813 ( 1967 ); Chem. Abstr., 67 12219z (1967).

74. E. L. Fareri and J. P. Pellegrini, Jr., U. S. Patent 3 , 085,867 . (1963); Chem. Abstr. , 2570f ( 1963 ).

75* G. A. Starmer, S. McLean, and J. Thomas, Toxicol. Appl. Pharmacol., 20 (1971).

76. R.- L. Every and R. L Riggs, Jr., Mater. Protect., ^3, 46 (1964).

77- Monsanto Co., Netherland Appl 6,602,564 ( 1967 ); Chem. A b s tr., 6X, 99332r ( 1967 ).

78 . W. A. Beard, Jr., and C. R. Hauser, J. Org. Chem., 2£, 334 (i 960 ).

79. For a preliminary report of this work see P. G. Gassman and G. G ruetzm acher, J . Amer. Chem. S o c ., 588 (1973)» 149

80 o (a) P. Claus and W. Vycudilik, 'tetrahedron Letters, 3607 (1968 ); f t ) P. Claus and W. Vycudilik, Monatsh. Chem., 101, 398, 405 (1970)} (c) P. Claus, W. Vycudilik, and W. Rieder, ibid., 102, 1571 (1971) ; (d) P. Claus and W. Rieder, ibid., 1Q3., II 63 (1972).

81. P. G. Gassman and H. R. Brewes, Chem. Comm., 488 (1973)*

82 . P. G. Gassman and C. T. Huang, J . Amer. Chem. S o c ., 934 4453 (1973).

83. P. G. Gassman and T. J. van Bergen, ibid., 590 (1973)*

84. P. G. Gassman and T. J. van Bergen, ibid., 9 3, 591 (1973)*

85. P. G. Gassman and T. J. van Bergen, ib id .,

8 6 . M. G. Burden and J. G. Moffatt, ibid., 874 4656 (1965 ) } M. G. Burdon and J. G. Moffatt, ibid., 8b, 5855 (1986); M. C-. Bur don and J. Gc Moffatt, ibid., Sgj 4725 ( 1967 )* 87. U. Lerch and J. G. Moffatt, J. Org. Chem., 3 8.* 3861 (1971)* 88 . P. Claus, Monatsh. Chem., 102, 948 (1971)*

89. E. Vilmaier and W. Sprugel, Tetrahedron Letters, 625 (1972).

90. W. E; Lawson and T. P. Dawson, J. Amer. Chem. Soc., 4$, 3H9 (1927).

91. P. Haas, Biochem. Jo, 29, 1297 (1935)5 E. Runge, E. Prof ft, and R. Brux, J. Prakt. Chem., 2, 279 (1955); H° Bohme and E. Boll, Z. Anorg. Allg. Chem., 2£0, 17 (1957)5 H. Eenselau and J. G. M o ffa tt, J . Amer. Chem. S o c., 83, 1762 (1966 ); G. E. Wilson and R. Albert, ~j7~~Crg. Chem., 3.8? 2156, 2160 (1973) 5 E» G. Bordwell and B. M. P i t t , J. Amer. Chem. S o c ., 33.5 572 (1955); H. Bohme and H. J. Gran, Justus Liebigs Ann. Chem., 3 8 1, 133 (1953); F. Boberg. G. Winter, and G.~R. Schultze, Chem. Ber., 89., 1160 (1956).

92. G. Allegra, G. E . Wilson, Jr. , E . B e n e d e tti, C. Pedone, and R. Albert, J. Amer. Chem. Soc., J92, 4002 (1970).

• 95. N. C« Baensinger, R» E. Buckles, R. J. Manner, and- T. D. Simpson, ibid., 91, 5749 (1969). 150

9k. For a preliminary report of this work see P. Go Gassman, G. Gruetzmacher, and T. Jo van Bergen, J . Amer. Chem. S o c., 95* 5608 (1975).

95° Branodimethylsulfonium ‘bromide (orange crystals, dec. 81-82°) can be stored for several weeks at. room temperature in the absence of light, and moisture, H. Furukawa, To Inoue, To Aida, and S» Oae, Chem. Commun., 212 (1973)•

96. The dimethylformamide had been stirred overnight with anhydrous magnesium sulfate, filtered, and distilled at reduced pressure from barium oxide into a 1 Z round-bottomed flask with a side arm and septum inlet containing kA molecular sieves.

97. J. E. Hyre and A. R. Bader, J. Amer. Chem. Soc., 80, kj>j (1958).

98. Ho Lercher, F. Holschneider, K. Koberle, W. Speer, and P. Stocklin, Chem. Ber., j?8, kOfy (1935).

99* H. G. Tanner and P. A. Lasselle, J. Amer. Chem. Soc., 2165 (1926 )o

100. F. J. V ilia n i, U. S. P aten t 2,76t,519 (1956), Chem. A bstr. , £1, ptttjc (1957).

1 0 1 . Ko von. Auwers, Justus .Liebigs Ann. Chem., t22, 160 (1921).

102. R. C. Weast, Ed., “Handbook of Chemistry and Physics,” 50th ed., Chemical Rubber Pub. Co., Cleveland, Oh., 1969? p C-5lt.

105. II. Ley and G. P f e if f e r , Chem. B e r., J§t, 363 (1921). lO t. R. R. Read and D. B. M u llin , J . Amer. Chem. S o c., jjO, 1763 (1928)o