Group Title: proposed radical-chain pathway for photoinitiated reductive-dehalogenation and substitution reactions of hetaryl and aryl halides in methanolic methoxide /
Title: A Proposed radical-chain pathway for photoinitiated reductive-dehalogenation and substitution reactions of hetaryl and aryl halides in methanolic methoxide /
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098294/00001
 Material Information
Title: A Proposed radical-chain pathway for photoinitiated reductive-dehalogenation and substitution reactions of hetaryl and aryl halides in methanolic methoxide /
Physical Description: xvii, 277 leaves : ill. ; 28 cm.
Language: English
Creator: Locko, George Allison, 1947-
Publication Date: 1976
Copyright Date: 1976
 Subjects
Subject: Halides   ( lcsh )
Photochemistry   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 271-276.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by George Allison Locko.
 Record Information
Bibliographic ID: UF00098294
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000178121
oclc - 03108611
notis - AAU4625

Downloads

This item has the following downloads:

PDF ( 9 MBs ) ( PDF )


Full Text

















A PROPOSED RADICAL-CHAIN PATHWAY FOR
PHOTOINITIATED REDUCTIVE-DEHALOGENATION AND
SUBSTITUTION REACTIONS OF HETARYL
AND ARYL HALIDES IN METHANOLIC HETHOXIDE







By


GEORGE ALLISON LOCKO


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1976


























To Hom and Dad













ACKNOWLE DGEMENTS


The author wishes to express sincere gratitude to Dr.

John A. Zoltewicz, Chairman of his Supervisory Committee,

for his guidance during the course of the research which is

reported herein. Without the keen insight of Dr. Zoltewicz

much of this work would not have been possible. Apprecia-

tion is also expressed toward the other members of his

committee: Dr. William R. Dolbier, Dr. Paul Tarrant, Dr.

Kuang-Pang Li, and Dr. Charles Allen, Jr.

The author is very grateful to Ms. Ann Kennedy for the

long hours which she spent typing the manuscript.

Very special thanks are extended to Sara for her love,

patience, and understanding.


iii













TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS iii

LIST OF TABLES vii

LIST OF FIGURES xiii

ABSTRACT xvi

CHAPTER:

1. INTRODUCTION 1

2. tMECHANISTIC INVESTIGATIONS INVOLVING
THE PHOTOINITIATED RADICAL-CHAIN RE-
DUCTIVE-DEHALOGENATION AND PHOTOINITI-
ATED SRij REACTION OF 3-IODOPYRIDINE
WITH THIOPHENOXIDE ION 1C

Results 10

Discussion 58

3. STRUCTURE-REACTIVITY CORRELATIONS AND
RELATED INVESTIGATIONS PERTAINING TO
THE PHOTOINITIATED REACTIONS OF HETARYL
AND ARYL HALIDES WITH THIOPHENOXIDE ION
IN METHANOL 89

Results 89

Discussion 144

4. INVESTIGATION INTO THE MECHANICS BY
WHICH PHENYL SULFIDE AND IODOBENZENE
ARE CONVERTED TO BIPHENYL DURING IRRA-
DIATION IN METHANOLIC HETHOXIDE 182

Results 182

Discussion 200







Page


5. EXPERIMENTAL 217

Instrumentation 217

Chemicals 218

Preparations 221

Photolysis Procedures 226

Preparation of Solutions for Photolysis 227

Preparation of Solutions for AIBiJ and
PAT Initiated Experiments 227

Sample Degassing Procedures 228

Column Preparation and Routine Oper-
ating Procedures for Gas-Liquid
Chromatography 228

Cut and Weigh Method of Integration
for Gas-Liquid Chromatography 230

Molar Response Factors for Gas-Liquid
Chromacography 231

Quantitation of 3-lodopyridine, Pyridine,
and 3-Phenylthiopyridine by Gas-
Liquid Chromatography 239

AIBil Controls for Gas-Liquid Chromato-
graphy 253

Control Experiments Undertaken to Deter-
mine Whether the Internal Glc Standard,
Anisole, is Consumed During Sample
Irradiation 255

Products Derived from Thermal Decomposi-
tion of PAT in Hethanolic Methoxide 256

Product Confirmation by GC-HS Analysis 257

Low Voltage GC-MS Analysis of Labeled
Biphenyls 260

Kinetic Study of the Reaction of 2-Bromo-
thiazole with Sodium Methoxide 260







Page


Photolysis of N-terc-Butyl-a- Phenyl-
nicrone and 2-Hethyl-2-Nitsosopro-
pane in Methanolic Hethoxide Con-
taining Thiophenoxide Ion 267











LIST OF TABLES


Table Page

1 Effect of Methoxide Ion Concentra-
tion of the Rate and Extent of the
Photoinitiated Reductive-Dehalogen-
ation of 0.31 M1 3-Iodopyridine in
Methanol 11

2 Effect of Degassing on the Extent of
the Photoinitiated Reductive-Dehal-
ogenation of 0.31 M;I 3-lodopyridine
in 0.43 H Hethanolic Methoxide 15

3 Effect of Degassing on the Extent of
the Photoinitiated Re uctive-Dehal-
ogention of 1.4 x 10- 1 3-Iodopyri-
dine in 0.31 M Methanolic Hethoxide 17

4 Attempted Inhibition of the Photo-
initiated Reductive-Dehalogenation of
1.4 x 10-2 M 3-lodopyridine in 0.31
[I Methanolic Hethoxide 19

5 Product Distributions Resulting from
the Photoinitiated Reactions of 0.30
1 3-Iodopyridine with 0.47 0.49 M
Sodium Thiophenoxide and 0.44 0.47
Il Sodium Hethoxide in Methanol 22

6 Dependence of Product Ratio from the
Photoinitiated Reaction of 0.30 M
3-Iodopyridine on the Concentrations
of Thiophenoxide and Methoxide Ions 24

7 The Small Effect of Methoxide Ion Con-
centration on the Extent of Reaction
of 3-Iodopyridine with 1.3 .1 Sodium
Thiophenoxide in Methanol 27

8 Product Distributions Resulting from
the Photoinitiated Reaction of 1.4 x
10-2 H 3-Iodopyridine with 0.49 M
Sodium Thiophenoxide and 0.43 0.44
M Sodium Methoxide 29


vii









9 Attempted Inhibition of the Photo-
initiated Reaction of 0.30 M 3-Iodo-
pyridine with 0.47 0.49 1M Sodium
Thiophenoxide and 0.45 0.47 M Sodium
Methoxide 34

0 Attempted Inhibition of the Phoco-
initiated Reaction of 1.4 x 10- M
3-lodopyridine with 0.48 MI Sodium
Thiophenoxide and 0.46 MI Sodium Meth-
oxide at High Light Intensity 35

1 The Effect of Potential Inhibitors on
the Photoinitiated Reaction of 1.4 x
10-2 1M 3-Iodopyridine with 0.47 0.49
HI Sodium Thiophenoxide and 0.44 0.47
tM Sodium Methoxide at Low Light Inten-
sity 36

2 Product Distribution Resulting from
the Phocoiniciated Reactions of 3-
Chloro- and 3-Bromopyridine with
Sodium Thiophenoxide and Sodium Meth-
oxide 39

3 Product Distributions Resulting from
AIBN Initiated Reaction of 3-Iodo-
pyridine with Sodium Thiophenoxide
and Sodium Mechoxide in Methanol at
1000 for 60 Minutes 46

4 Product Distribution from the Photo-
initiated Reaction of 3-Iodopyridine
with 0.13 M Phenyl Disulfide in
Methanolic Methoxide 50

5 Ultraviolet Absorption Data for 3-
lodo-, 3-Bromo-, and 3-Chloropyridine
in Methanol 55

i Ultraviolet Absorbence Data for the
3-Halopyridines in Methanol in the
Region for Transmittance of Radiation
by Pyrex Glass 57

7 Photoinitiated Reaction of 0.30 M
4-Bromoisoquinoline with Sodium Thio-
phenoxide and Sodium Mechoxide in
Methanol 90


viii


Table


Page








18 Photoinitiated Reaction of 0.30 M
3-Bromoquinoline with Sodium Thio-
phenoxide and Sodium Methoxide in
Methanol 93

19 AIBN Initiated Reaction of 5.9 x
10-2 i 4-Bromoisoquinoline with
0.49 M Thiophenoxide Ion in Methanol 96

20 Photoinitiated Reaction of 0.30 H.
2-lodothiophene with 0.46 0.48 ;!
Thiophenoxide Ion in Methanol 97

21 AIBN Initiated Reaction of 5.8 x 10-2
I 2-Iodothiophene with 0.49 N Thio-
phenoxide Ion 99

22 Product Distributions from the Irra-
diation of 0.30 M 2-Bromothiazole
at 71 20 in Methanolic Methoxide
which is 0.47 0.48 M in Thiophen-
oxide 101

23 Product Distributions from the Photo-
initiated Reaction of 0.30 0.32 H
lodobenzene with Sodium Thiophenoxide
and Sodium Methoxide in Methanol 106

24 Products Resulting from the Photo-
initiated Degradation of 7.1 x 10-- I.
Phenyl Sulfide in 0.62 H Hethanolic
Methoxide 110

25 Product Distributions from the Photo-
initiated Degradation of Phenyl Sul-
fide in Hethanolic Methoxide in the
Presence of 3-Halopyridines 112

26 Product Distributions from the Photo-
initiated Reaction of 1.0 x 10-2 1.
1-Bromonaphthalene with Sodium Heth-
oxide and 0.48 M Sodium Thiophenoxide
in Methanol 115

27 Product Distributions from the Photo-
initiated Reaction of 0.30 H 1-Bromo-
naphthalene with Sodium Methoxide and
0.37 0.48 M Sodium Thiophenoxide in
Methanol 119


Table


Page








28 Excent of Phocodegradation of l-Phenyl-
thionaphthalene in the Presence and
Absence of 1-Bromonaphthalene in Meth-
anol Containing Sodium Methoxide and
Sodium Thiophenoxide 122

29 Extent of Phorodegradation of 3-Phenyl-
thiopyridine in Methanol in the Presence
and Absence of Methoxide Ion 125

30 Product Distributions from Photoiniti-
ated Reactions of 0.30 M 3-lodopyridine
with Methanolic Methoxide in the Pres-
ence of Sulfur Nucleophiles 130

31 Effectiveness of the Anion of 4-Thio-
pyridone as a Trap for Phenyl Radical
Generated by the Photolysis of Iodo-
benzene in Methanol 133

32 Product Distributions from Photoiniti-
ated Reactions of 0.30 N 3-lodopyridine
with Methanolic Hethoxide in the Pres-
ence of Phosphorus Nucleophiles 136

33 Results which Demonstrate the Ineffec-
tiveness of Carbon, Nitrogen, and
Oxygen Nucleophiles as Traps for 3-
Pyridyl Radical in Methanol 138

34 Competition Between 0.40 M Thiophen-
oxide Ion and Other Thiolate Ions for
Aryl and Hetaryl Radicals in 0.28 -
0.31 I Methanolic Sodium Hethoxide 141

35 Typical Product Distributions for Com-
peting Substitution and Reductive-De-
halogenation Reactions Involving a
Series of Hetaryl and Aryl Halides 156

36 Calculated Rate Constant Ratios for
Reactions of Hecaryl and Aryl o-
Radicals with Thiophenoxide Ion and
with the Hydrogen Atom Donors Methanol
and Methoxide Ion 163

37 A Comparison to Demonstrate the Effect
of an Annular Nitrogen Atom on Reac-
tivity of a-Radicals Towards Thiophen-
oxide Ion 165


Table


Page









38 A Comparison to Demonstrate the
Effect of Annelation on Reactivity
of a-Radicals Towards Thiophenoxide
Ion 167

39 Determination of Labeled BiphenyLs
Produced by the Irradiation of
Phenyl Sulfide in the Presence of
Various Additives 185

40 Hass Spectral Analysis (70 eV) of
Recovered Phenyl Sulfide from a
-2
475 Hinute Irradiation of 7.1 x 10
M Phenyl Sulfide, 0.99 M. Benzene-d6,
and 0.62 11 Sodium Hethoxide in
Ethanol 187

41 Conditions for Unsuccessful Attempts
to Thermally Degrade Phenyl Sulfide
to Biphenyl 190

42 Isotopic Distribution of Deuterated
Biphenyls from the Generation of
Phenyl Radical in the Presence of
Heavy Benzene and Sodium Hethoxide 193

43 Products from the Generation of
Phenyl Radical in the Presence of
Chlorobenzene 196

44 Photoinitiated Reductive-Dechlorina-
tion of p-Chlorobiphenyl in Hethan-
olic Hethoxide 198

45 List of Chemicals and Purification
Procedures 219

46 Holar Response Factors and Retention
Times for Compounds on Column A 233

47 Holar Response Factors and Retention
Times for Compounds on Column B 237

48 Holar Response Factors and Retention
Times for Compounds on Column C 238

49 Retention Times for Degradation Pro-
ducts of AIBN 254


Table


Page









50 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the First Entry in Table 39 261

51 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the Second Entry in Table 39 262

52 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the Fifth Entry in Table 39 263

53 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the First Entry in Table 42 264

54 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the Second Entry in Table 52 265

55 Complete Low Voltage (15 eV) Mass
Spectral Analysis on the Biphenyl
Fraction, Prepared as Indicated in
the Third Entry in Table 42 266


:.: i i


Table


Page













LIST OF FIGURES


Figure Page

1 Extent of Reduction of 3-Iodopyri-
dine as a Function of the Ratio of
the Initial Concentrations of Sodium
Methoxide to Substrate 14

2 Chromatogram Showing the Relative
Peak Areas for Pyridine, 3-Bromo-
pyridine, and 3-Phenvlthiopyridine
Afcer 768 Minutes Irradiation of
the Sample Corresponding to Entry
12-2 41

3 Chromatogram Showing the Relative
Peak Areas for Pyridine, 3-Chloro-
pyridine, 3-Phenylthiopyridine and
Anisole After 811 minutes Irradia-
tion of the Sample Corresponding to
Entry 12-4 42

4 Chromatogram Showing that Pyridine,
3-Phenylthiopyridine, and Anisole
Remain, and that 3-Chloropyridine Has
Been Consumed After 4183 Minutes
Irradiation of the Sample Correspond-
ing to Entry 12-5 43

5 Relationship Between Product Ratio
and Initial Base Concentrations for
AIBN Initiated and Photoinitiated
Reactions of 0.3 M1 3-Iodopyridine
and for Photoinitiated Reactions of
0.3 M 3-Bromopyridine and 0.3 MI 3-
Chloropyridine 80

6 Relationship Between Product Ratio
and Initial Base Concentrations for
AIBN Initiated and Photoinitiated
Reactions of 10-2 M 3-Iodopyridine 82


: i i i







Figure


7 Relationship Between Product Ratio
and Initial Base Concentration for
Photoinitiated, AIBN Initiated, and
Thermally Induced Reactions of 4-
Bromoisoquinoline in Methanol 146

8 Relationship Between Product Ratio
and Initial Base Concentrations for
Photoinitiated and AIBN Initiated
Reactions of 2-lodothiophene in
Methanol 148

9 Relationship Between Product Ratio
and Initial Base Concentrations for
Photoinitiated Reactions of 3-
Bromoquinoline in Methanol 159

0 Relationship Between Product Ratio
and Initial Base Concentrations for
Photoinitiaced Reactions of l-Bromo-
naphthalene, with Emphasis on Data
Corresponding to 15 Minute Sample
Irradiation 160

1 Relationship Between Product Ratio
and Initial Base Concentrations for
Photoinitiated Reactions of lodo-
benzene in Methanol 161

2 Correlation of the Logarithms of
the Rate Constant Ratios for Reactions
of Phenyl, 2-Thienyl, and 3-Pyridyl
Radicals with Thiophenoxide Ion versus
Reaction with Methanol Against the
Logarithms of the Rate Constants for
Capture of the Hydrated Electron by
Benzene, Thiophene, and Pyridine 170

3 FID Response Curve for Analysis of
Pyridine on Column A Using Anisole
as Glc Standard 242

4 FID Response Curve for Analysis of
3-Phenvlthiopyridine on Column A
Using Anisole as Glc Standard 244

5 FID Response Curve for Analysis of
3-Iodopyridine on Column A Using
Anisole as Glc Standard 247


xiv


Page






16 FID Response Curve Corresponding to
On-Column Concentration of 3-Iodo-
pyridine, Column A 249

17 FID Response Curve Corresponding to
Low On-Column Concentrations of 3-
lodopyridine, Column A 251

18 FID Response Curve Corresponding to
On-Column Concentrations of 3-Iodo-
pyridine, Column A2 252







Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

A PROPOSED RADICAL-CHAIN PATHWAY FOR
PHOTOINITIATED REDUCTIVE-DEHALOGENATION AND
SUBSTITUTION REACTIONS OF HETARYL
AND ARYL HALIDES IN FIETHANOLIC METHOXIDE

By

George Allison Locko

December, 1976

Chairman: J. A. Zoltewicz
Major Department: Chemistry

Evidence is presented for the intermediacy of 3-pyridyl

radical in the methoxide ion reduced, photoinitiated re-

duccive-dehalogenation of 3-iodopyridine in methanol. At

fixed reaction times the extent of reaction is increased by

removal of dissolved oxygen, an effect which may be indica-

tive of radical-chain character.

W-hen 3-halopyridines (iodo, bromo, chloro) are photol-

yzed in methanol in the presence of thiophenoxide ion,

products arising from substitution (3-phenylthiopyridine)

and reductive-dehalogenation pyridinee) are obtained. Pro-

duct ratios for the 3-halopyridines are independent of the

identity of the halogen substituent. The overall reactivity

order is iodo > bromo > chloro.

Thermal decomposition of AIBN, a free radical initiator,

was also employed to induce substitution and reductive-

dehalogenation of 3-iodopyridine. Similar substitution to

reduction product ratios were obtained in AIBM initiated

and photoinitiated reactions. This result suggests that


X v






substitution and reduction products arise via competition

between chiophenoxide ion and the hydrogen atom donors

methoxide ion and methanol for an intermediate 3-pyridyl

radical. A radical-chain pathway (S l) is postulated.

The following reactivity order is established for the

reaction of thiophenoxide ion which a series of hetaryl

and aryl radicals: 3-quinolyl > 4-isoquinolyl 3-pyridyl

> 2-thienyl 1 1-naphthyl > phenyl radical. This order

parallels the predicted electron affinities (excluding 1-

naphthyl radical) of the hetaryl and aryl radicals.


: v. i i













CHAPTER 1
INTRODUCTION



It is well known that aryl and hetaryl halides (except

fluorides) undergo photoinitiated reductive-dehalogenation,

a reaction represented by equation 1, in which solvent (SH)

usually serves as hydrogen atom donor.1-14 More recently

it has been reported that this photoinitiated reaction is


h v
ArX + SH > ArH + S. + X- (1)



facilitated by certain bases. Radical-chain mechanisms

have been postulated for these base accelerated

reactions. 6,8-10,1314 In these chain reactions the base

(:B ) may (a) serve as an electron donor to photoexcited

substrate (equation 2) and/or (b) abstract a proton from

solvent derived radicals (equation 3). In either case, the

base facilitates the chain process by forming radical anion



ArX* + :B > ArX + -B (2)


-SH + :B S + HB (3)



intermediates.

The radical-chain mechanisms which have been proposed







are similar to the original mechanism proposed by Bunnetc

and Wamser for the free radical initiated reductive-deiodin-

ation of m-chloroiodobenzene in methanolic sodium methoxide.15

This mechanism is outlined in Scheme I.



Scheme I

R- + CHO3H RH + -CH2OH (4)

*CH9OH + CH3 O == CH OH + CH20 (5)

CH9O + ArI -- CHO + Arl (6)
ArI ----- Ar + I (7)

Ar- + C13OH ArIl + -CH9OH (3)



Zoltewicz, Oestreich, and Sale reported that the re-

ductive-dehalogenation of 4-bromoisoquinoline could be

affected in methanolic sodium methoxide ac temperatures

ranging from 1430 to 1650.16,17 Presumably, the high tem-

peratures which were employed induced free radical formation.

The same group reported that 3-iodopyridine is similarly

reduced via a radical-chain process.17 In both reactions,

equations 9 and 10, reductive-dehalogenation is complicated

by the formation of 4-methoxyisoquinoline and 3-methoxypyri-

dine, respectively. The latter products arise by direct

substitution.


Br H OCH3

1650 + + _aBr
JaOCH, 0, )

Ch3OH (9)
J(9








S 165aO H OCH I
1 C Q + Q+ Nal (10)
NaOCH 30H
CH3OH
Zoltewicz and Oestreich extended their work to thermally
initiated radical-chain reactions in which they were able to
trap the intermediate 4-isoquinolyl radical (of equation 9)
with thiophenoxide ion.16,18 The products of the reaction
are isoquinoline and 4-phenylthioisoquinoline, equation 11.
Only traces of 4-methoxyisoquinoline were found.


Br H SCH
o5

+ NaS H5 47 + O O ,
0 :IIaCc 3 10 :
T CH3OH '
HCH30H



+ (11)




For the above reaction, a radical-chain mechanism
(Scneme II) was proposed in which the intermediate 4-iso-
quinolyl radical (I) may react with either methoxide ion and
solvent (equation 12) or with thiophenoxide ion (equation 13).
Attack by thiophenoxide ion on the intermediate hetaryl
radical generates the radical anion of 4-phenylthioisoquin-
oline (II). Radical anion II then transfers an electron to
substrate, in equation 14, to continue the chain. The over-
all mechanism is similar to chat proposed in Scheme I, with






SCHEIIE II


,.qJ,>'


00I
I


~6'5


sc i
C6 5

OXO@


SC, H
O O5

III


Br

OO,


SC6 5

0N0


Br

OO!,.


(12)









(13)


(1 )







the addition of chain propagation steps 13 and 14. This

mechanism is classified as an SRNi1 type process (Substitu-
19
tion, Radical-Nucleophile, unimolecular). It should be

noted, however, that a small percentage of the sulfide

formed in the above experiments also arises via "classical"

nucleophilic aromatic substitution, a competing reaction.

In the present work, preliminary investigations in-

dicated that the addition of sodium mechoxide to methanolic

solutions of aryl or hetaryl halides results in a greatly

accelerated reduccive-dehalogenation process when samples

are exposed to pyrex filtered ultraviolet light. Further-

more, phocoinitiation is superior to thermally initiated

reductive-dehalogenation since the only product observed is

that resulting from reductive-dehalogenation, i.e., nucleo-

philic attack by methoxide ion on hetaryl halide does not

compete with reductive-dehalogenation.

The goals of our research were to (a) show that aryl

and hetaryl halides can be easily reduced photochemically

in methanolic sodium methoxide, i.e., that photoinitiated

reductive-dehalogenation of aryl and hetaryl halides in the

presence of methoxide ion is a method of general synthetic

utility, (b) develop a method whereby high yields of hetaryl

and/or aryl sulfides could be obtained via photostimulated

SRN1 reactions between the parent halides and thiophenoxide
ion, (c) investigate the reactivity of various nucleophiles

in the phocostimulated SR1 reaction with aryl and hetaryl

halides, i.e., attempt to trap hetaryl or aryl radicals which







nucleophiles other than thiophenoxide ion, and (d) examine

the effect of substrate structure on reactivity in substiru-

tion reactions with a given nucleophile.

In Chapter 2, mechanisms for the photoinitiated reduc-

tive-dehalogenation of 3-iodopyridine and the photostimulaced

SR ji reaction between the 3-halopyridines (iodo-, bromo-,

and chloro-) and thiophenoxide ion are investigated. It is

shown that 3-iodopyridine is rapidly reduced to pyridine in

methanolic methoxide, and that this reaction requires meth-

oxide ion, equation 15. It is also shown that high yields




Sh H
> (15)
IJaOCH3 CH OH
3 3 3'*



of 3-phenylthiopyridine can be obtained when 3-halopyridines

are photolvzed in methanol containing thiophenoxide ion,

equation 16. Pyridine is also a product in these reactions,

and the ratio of 3-phenylthiopyridine to pyridine is deter-

mined by the initial thiophenoxide ion to methoxide ion con-

centration ratio. This radical-chain substitution reaction

is not complicated by "classical" nucleophilic aromatic sub-

stitution; hence, photoinitiation is superior to thermal

initiation in this respect. Evidence is presented which

supports radical-chain mechanisms in both the reductive-

dehalogenation and substitution reactions.











A H. 11 SC6H 5
Q + L aSCH H) SC6 +NaX (16)
5aOCH, CH3' OH J



S= I, Br, C1



In Chapter 3, effects of nucleophile and substrate

structure on the reactivity of aryl and hetaryl radicals to-

ward nucleophiles in the photoinduced Sp1 reaction are ex-

amined, with the expectation that it might be possible to

trap aryl and hetaryl radicals with a variety of nucleophiles,

giving rise to their respective substitution products (equa-
20-26
tion 17).20 Of the many types of nucleophiles which were


h -
Ar- + :R > ArR + e (17)
IHaOCH3, CH30H


employed as potential radical traps, only the thiol anions

are successful. A few experiments also indicate that phos-

phorus nucleophiles might be used with success.. This finding

is corroborated in the literature.27

In examining the effect of substrate structure on re-

activity toward substitution with thiophenoxide ion, several

aryl and hetaryl halides were employed. 3-Halopyridines,

4-bromoiscquinoline, and 3-bromoquinoline all give high

yields of substitution products. In contrast, reductive-







dehalogenation is apparently the predominant reaction path-

way for iodobenzene, 1-bromonaphthalene, and 2-iodothiophene.

For 2-bromothiazole, thermally induced nucleophilic aromatic

substitution competes effectively with the photoinitiated

substitution reaction. The differences in observed substi-

tution to reduction product ratios for these compounds are

rationalized primarily by the effect of annular nitrogen in

stabilizing the incipient radical anion formed by attack

of thiophenoxide ion on the hetaryl radical and by differ-

ences in the electron affinities of the various aromatic and

heteroaromatic rings.

The relative stabilities of some of the aryl and hetaryl

sulfides, compounds which are derived from attack of thio-

phenoxide ion on the aryl or hetaryl radicals, are also

examined. The results are reported and discussed in Chapter

3. The primary mode of decomposition of the sulfides appears

to be cleavage of the radical anion in equations 18 and 19.



ArS + -C H (18)

ArSC6H
o 5
Ar- + C6H5S (19)



Biphenyl appears as a minor product in the photoiniti-

ated decomposition of phenyl sulfide. Although previous in-

vestigators have postulated an intermolecular reaction for

the formation of biphenyl from phenylsulfide, results in-

dicate that both intermolecular and intramolecular processes
29
may be in operation. Investigations into the mechanism




9


for biphenyl formation are presented in Chapter 4.

During the progress of this research, Bunnett and

Creary reported the synthesis of aryl sulfides by the photo-

initiated reaction of aryl halides which thiophenoxide ion

in liquid ammonia via an SR* 1 process.3













CHAPTER 2
MECHANISTIC INVESTIGATIONS INVOLVING THE PHOTOTIITIATED
RADICAL-CHAIN REDUCTIVE-DEHALOGENATION AND
PHOTOINITTATED SR!1I REACTION OF 3-TODOPYRIDINE
WITH THIOPHENOXIDE ION



Results

Photoinitiated reductive-dehalogenation of 3-iodo-

pyridine in methanolic sodium methoxide. 3-Iodopyridine,

when irradiated with pyre:: filtered ultraviolet light in

methanolic sodium methoxide, is rapidly reduced co pyridine.

Results indicate that (a) methoxide ion is required for

reductive-dehalogenation, but that the addition of methoxide

ion in excess of one equivalent has little effect on re-

action race; (b) removal of dissolved oxygen results in a

significant rate acceleration when reactions are carried

out at low substrate concentration; (c) reaction rate is

proportional to light intensity; and (d) the reaction can

be inhibited by some free radical inhibitors.

Reductive-dehalogenation clearly requires methoxide ion,

as demonstrated in the experiments in Table 1. W-hen methanol

solutions of 3-iodopyridine without added methoxide ion are

irradiated with pyrex filtered ultraviolet light, the rate

of reaction must be extremely slow, since in run 1-1 only a

trace of pyridine is detected after an irradiation of 2207

minutes. However, when a solution of 3-iodopyridine in
















I

o


0 C

au
41-



CO




'- .
0 C







0O
C l











0 C








Ll 0
*r

















0 0


a ")


C rC



tn











0
C-
Lr-4 -4
Oo


m .
C *-
0 0>
*rH











L a>






W r4


01
u
C

L-i
Er,





r-l




rn
E









.,i
Co

0




a)












0
c t
.rrl
,4 -A
O -









Co





C

r,-4 C
0L *,I
M E
*r
*l-
a)









0


U -
O








C I


-3 -3 -,7

CM Co Co CM
C. C. C. 0- I, 0-
S E E ES E



r-4 r-4 C14N N N N








ON CO r r-4 Z 1-0
CF O C O O C) O
-4 r- i -4








c0 0 00 r %D L0











C0~ 0 Lf 0 0
C) o i--q r- 4 co











0' ^ Cm N N li
C

cci
Co
















C)
C
0








c IO'
N- 0 0 0 0
N O I 0
-c















*Hl
MC




01




12


methanol which was initially 2.2 1. in sodium methoxide was

irradiated for only 96 minutes, conversion to pyridine was

complete.

Experiments 1-3 through 1-5 demonstrate that reaction

essentially ceases after metho::ide ion has been consumed.

In each of these experiments methoxide ion is the limiting

reagent. In run 1-3, in which the initial concentration

ratio of methoxide ion to substrate was 0.15, the glc yield

of pyridine is 13 percent. In run 1-4, the 71 percent yield

of pyridine agrees closely with the predicted conversion of

74 percent which would be produced by consumption of all of

the methoxide ion (0.23 11/0.31 1 = 0.74). Since doubling

the irradiation time in entry 1-5 does not change the yield

of pyridine significantly (71 vs 66 percent), irradiation

time must not be a limiting factor in entries 1-3 through

1-5. In the last run, 1-6, irradiation time probably is

the limiting factor factor since more than 7 equivalents of

methoxide ion were initially present, and conversion to

pyridine is incomplete.

The average mass balance for the entries in Table 1

is 98 + 6 percent. The high and low values of 108 and 83

percent, respectively, represent glc quantitation error for

3-iodopyridine which is occasionally encountered with the

use of the carbowax column which was employed for the

analysis of samples in Table 1. The problem of 3-iodopyri-

dine quantitation is discussed in greater detail in Chapter

5.







The results plotted in Figure 1 further demonstrate

that reaction almost ceases when methoxide ion, present as

the limiting reagent, is consumed. As the amount of meth-

oxide ion initially present is increased by increments,

proportionate amounts of substrate are consumed until approx-

imately an equivalent of methoxide ion has been added. In

this region the increase in consumption of substrate with

the added methoxide ion declines sharply.

Experiments were also undertaken to determine whether

the presence of dissolved oxygen has any effect on the re-

action race. The effect of oxygen was first determined at

high substrate concentration. Examination of the results

in Table 2 suggests there is a slight rate acceleration

when samples are degassed prior to irradiation. In the first

two runs, samples were not degassed prior to 3 minutes ir-

radiation; the yield of pyridine in each run is 11 percent.

In the third entry, the yield of pyridine increases slightly

(27 + 2 percent) for samples which were degassed prior to

irradiation.

It is interesting to note that doubling the initial

methoxide ion concentration in run 2-2 does not increase the

yield of pyridine relative to the amount obtained in run 2-1.

This also demonstrates that the addition of sodium methoxide

in excess of one equivalent has only a small effect on the

rate of reduccive-dehalogenacion.

The effect of dissolved oxygen on reaction rate is much

more dramatic for reactions which are carried out at low























































































'0 '0
'0 rjI* *:


> 30 '- 0 0 0 0
-.C L,7 ,.T ,-.4

euTpT.I,,dopo.-t E,


1"



,-i L4 rLj






'- U 3





0 OC
Cr- H r-





C: r-4
Lr rJ
















0 E





C3 C J C,.

- .,
-CC
0 O O E *-









S- .-Ti F -i







C-l -r *-4 LJ

7U C 0C
o0 r C









c v LCT3-
0 o


0 .






0 C 0
U0

J Hu -T
rD3 C-)





r -4 L
TO > Ui >







O C O 0F
*Z U 0 U









Si-. Ci
UCO

*'"J 0 3 C







Table 2.


Effect of Degassing on the Extent of the Photo-
initiaced Reductive-Dehalogenacion of 0.31 iI 3-
lodopyridine in 0.43 1-H ethanolic Methoxidea


Nol o
Pyridine


Mass Bal.


Degassed


1 87 11 98 .No

2b 85 11 96 No

3c 69 + 3 27 + 2 96 + 4 Yes



aSamples were irradiated for 3 minutes with one lamp at a
distance of 42 mm.

Solution was initially 0.36 i. in sodium methoxide.


Average of results from three experiments.


Run #
(2- )


Hol i
lodo-







subscrace concentration. In Table 3 are listed the results

from two series of experiments, 3-1 and 3-2, in which both

degassed and nondegassed solutions of 3-iodopyridine were

irradiated for identical periods. There is clearly a large

race increase due co oxygen removal prior to phocolysis, as

evidenced by the almost quantitacive conversion to pyridine

in encry 3-1. The magnitude of this oxygen effect is made

more evident by the obser'.'ation in entries 3-3 and 3-4 chac

subscrace is only slowly converted to pyridine on extended

irradiation of nondegassed solutions.

Experiments 3-4 and 3-5 were carried ouc to demonstrate

that there is no significant rate acceleration on substrate

dilution. A comparison of the results in entries 3-4 and

3-5 indicates chat the ec:enc of reaction actually decreases

slightly on an approximate 22 fold dilution of substrace.

In the latter run 35 percent substrate remains after a 130

minute irradiation, while in the former (low initial sub-

strace concentration) run 60 percent substrate remains.

From the results of degassing experiments and those of

runs 3-4 and 3-5, the following conclusions are reached.

First, the excent of reaction is similar for degassed samples

at high substrate concentration and nondegassed samples at

both high and low substrate concentration. Second, the

extent of reaction is greater for degassed samples at low

substrate concentration.

The average mass balance in Table 3 is 97 + 7 percent.

The values of 103 percent in entries 3-4 and 3-5 are within














03
C


O


r-c

Ga





LI4
Z-








0
> *H


, o.
3 Li





-u 01


CO









a)
C'Ti












C"O
0 0











-c






O w
C

C O








uO
0


LJ i-i


0
rI I



O 0
C


Un r-4


,G




C -)




c> >


3
0
Gl














C











Zr
*l
0







1-1








O O
Ul
























-o


















:3 E
0CI

















0
r- *r











C >'





-o -
0 0













C I
I c


0 0 0 0
>- z Z z z












Lf

+1











O
r-1l-









+,

CD 0C 10c c co
+1

o O I'. I c-






























O O O c O
r--4-


+1

i 0 0 m

















o 0 0O 0 0
rl v-- ( CO 00










0)


-i N c- -..- in


'In




U)
r-i





u
0
U
C


0 H

Li 0a
0 .C C



-l a 0 *



-- E 0
0 i w r
r-l *-l I -H


C X


-i 0 0


LIN cn
0 *F -F



3 m o

u E E
0a 0 0 0 0
Li k4





3-4 n :3 Li Li1
r.l -- cr -H -1
r- ( a l r- -H

0 0 C C
'---

3 0 0 03 03
3 3

n) ao c
r-4 03 0m r-4 r-4

E o U E 5
a > > 03 0
cri < < m LO
ca ,D u -o oi







the limits of experimental error. The high value of -105

percent for the first entry is probably due to inaccuracy in

quanticacion of 3-iodopyridine at low concentration. A

linear regression analysis was employed for quancitation of

3-iodopyridine, and an intercept correction which is made

may lead to a significant error for quancicacion of low on-

column concentrations of substrate.

The extent of reduccive-dehalogenacion is dependent on

light intensity, an effect which is clearly demonstrated in

a comparison of runs 1-4 and 3-2. In the former run two

lamps were employed at a distance of 42 mm and the glc yield

of pyridine is 71 percent. In the latter experiment the

irradiation time was unchanged, but one lamp was employed at

about a 4-foid greater distance of 165 mm; consequently, 3-

iodopyridine does noc react.

A control experiment was run to determine to what ex-

tent reductive-dehalogenacion occurs thermally. A methanol

solution which was initially 1.4 x 10-2 H in 3-iodopyridine

and 0.31 ;1 in sodium nethoxide was heated at 1650 for 6

minutes. Only a trace of pyridine was detected by glc

analysis. For temperatures at which the photoinitiated re-

actions are carried out (<710) 3-iodopyridine should be un-

reactive.

Reduccive-dehalogenation was successfully inhibited

when H -cert-butyl--:-phenylnicrone and 2-methyl-2-nicroso-

propane were employed as additives, Table 4. iigh mole

ratios of inhibitor to substrate were used in these experi-

ments. Unfortunately, inhibition in 4-1 was probably















O-<







Li
4-1
0





TO



0C




k,
fm cu
-:4
0
-4










r"
OO


.1) -
C u
-I


-1 0
(o





Q-C
W 0










C
H C













<
0






4-1
r-i






-u
C c















1-i .:


CMn
=3
-0
I
*0 ,


>. *





U C
















1o
00









aC'
^j













CO


CO CN
ca 0
i-4


oJ Ln
cn -


C 0 0






r-iri


ICN

OH
0,


UX
4-1 X

c co
*r-



CCC
I

r-i



E
1 .
ca CL


E






-u




C:

*r-1
0




W





TO
ca











r.J
-0












0
CJ
0



C2





OJ
.C


Cu


r-







Q4
O
01








U,
O,



















E
0








En

-4
0






cn
11


C



0
0




U,









0
O











J r-
OE
E

,-4
EW



rl. x
a >
0



Cm ,
-r-


C

.- 0
a r_:










x
JHaC



m4
0O C




4-4-0
C r3
*r-1 40

W-J
*SZ .I







U0




LiJ -0
-1
cn


12)
cn T







associated which the nitrone acting as a photochemical filter;

the nitrone strongly absorbs light (methanol, 292 mm,
max
c 17,600) in the region where pyrex transmits ultraviolet

radiation. However, inhiibtion in 4-2 was more likely due

to a radical-chain breaking process since 2-mechyl-2-nitroso-

propane absorbs light (methanol, max 289 nm, c 160) eakly

in the same region. The data in the fourth entry, in which

no additive was employed, serves as the control for these

inhibition experiments.

Pyridine, the product of reductive-dehalogenation, was

also investigated as a potential inhibitor. When a high

concentration of pyridine .was employed, in run 4-3, some

inhibition w.as noted. However, pyridine probably does not

inhibit reductive-dehalogenation at the low concentrations

in which it is formed during a typical photolysis of 1.4 x
-2
10 11 3-iodopyridine.

Photoiniciated reductive-dehaloqenacion and subscitu-

tion reactions of 3-iodopyridine in methanol containing

methoxide and chiophenoxde ions. In Table 5 are shown

results of a series of experiments in which 0.30 1I 3-iodo-

pyridine was irradiated with pyrex filtered ultraviolet

light in the presence of sodium methoxide and sodium thio-

phenoxide. A constant concentration ratio of thiophenoxide

ion to methoxide ion was used in these experiments to make

possible comparisons of substitution to reduction product

ratios at different times. The average mass balance is

90 + 4 percent for fifteen experiments; this indicates

either some substrate or products were consumed via side








reactions, or there was a consistently low quantitation

error in glc analysis. No substrate derived components

other than pyridine or 3-phenylthiopyridine were detected

by glc or gc-ras analysis of product mixtures. However,

analysis of a few product mixtures by glc indicated the

presence of varying amounts of benzene, a product which

probably arises via degradation of the sulfide. (Analysis

of the product mixtures from runs 2 and 4 in Table 6 in-

dicated the presence of 13 and 2 percent benzene, respec-

tively.) Benzene was not detected in all product mixtures

analyzed; it appears that benzene is formed only in the

experiments in which a relatively high concentration of

thiophenoxide ion is present initially and in which long

irradiation times are employed. The mechanism by which

benzene is formed is discussed in a later section.

For the series of experiments represented in entry

5-6, each of six samples was irradiated for 60 minutes.

Tne average product ratio for substitution to reduction is

1.2 + 0.1. The average glc yield for 3-phenylthiopyridine,

taken over values ranging from 30 to 41 percent, is 37 + 3

percent. For pyridine the average yield is 31 + 4 percent

(27 to 38 percent) and for 3-iodopyridine the value is

20 + 6 percent (12 to 30 percent). Reproducibility for

quantitacion of 3-iodopyridine is poor. Therefore, 3-iodo-

pyridine concentration determined by glc is not a quantita-

tive measure of the extent of reaction. The glc yields of

sulfide and pyridine are more reproducible. Thus, sub-











rJ

Li C'J
wrO

.4 .


0 Ln
3 -
O -r:
0 C)
0
o -




U) C/

C0



O0




a.,-
C 0

0 Q


OC 0




.-4 0
aU)























U) 0 -4
0 0
TO

























4-J Or--
C 0


























SLi C
ooi

l-l4












0 C






-* a)

U *, .




U 0wH
Qi 0










30 0_r
*0 4
C 0










FA~cr2


r.*)

oC-

* 0O
U --
C Q
0 E
U w







Li








r-4
m C





- "







0
c





















-a
CO

0 3





*H




0
-4 -0





0 0







a

* 0
1-



C-
*
r E


I--







C m-* I


i -
0 roo

o0 0a
4-I I
r-i CI (1) C,' r4- cli I-) co

rl r-4, r--1 r- r-4 l-< I










+1 +I
o \D Cn c0> C Co 0 r- C
Or oa a,- 0o 0 co co C







Cz C-,














C C N m l ,--T
+i +

0 cj r- C C') C r- 0. r-4




+1 J
00 co r-4 r- %D r- O -Z

























-4 -4
CM co Ln oo %D co r-




















i-1 C1 C- Nl f %. N 0



- CM m -

U



U)


0"3






03
Cu







03
E







.i





.r-
3



4-1
0










r-i


C,-
a


ao
03
Li








-4
r.1

0

01




-0







4-1
3


-4
0






a







-i
en








0-














C3
L1
0

*
U0
-4




U)



r-


U)
0L


4-1
C.
E


a)
0.











4,





-C
0


















Ui
0a













a1







-4


i-
0)
0-



a)


01
o4
U)



a)
01






v,

1^


Li
U)




0




-4a
*r-




Li






C
'-1


c


U)





0





L1



E E
4-1


x-r-4
Qc)



0 0






.C


nn
00 O





4-4 U








0.J

E E -

01 L 0
> 0O
a 0 O


*r- o
o 01 a
u o

) -4n 01
a) -- U





















0a LO U


U) m O
3 01
a -4 )














U "- 0)
m E *H


L i i




01 *0 r-4

U)0 0








3 T0 a)







stitution to reduction product ratios are reliable measures

for che competition between substitution and reductive-

dehalogenation.

Comparison between the results in entry 5-6 and results

for the other experiments in Table 5, which differ only in

the time of irradiation, demonstrates there is a fair

correlation between conversion of substrate to products with

irradiation time. The average substitution to reduction

product ratio does noc change with irradiation time. Neither

sample degassing nor blanketing a sample with oxygen had

any significant effect on the reaction rate or product

ratio.

The last entry in Table 5 demonstrates chat rate is

proportional to light intensity, an observation which was

also noted for reductive-dehalogenacion. In this experiment

a longer irradiation time is employed, but conversion of

substrate to products is significantly less than in runs

5-1 through 5-7 since the light intensity is reduced by

about a factor of 32. The substitution to reduction product

ratio for this experiment is considered to be unreliable

due to the inaccuracy in quantitating pyridine at low

concentrations.

The data in Table 6 demonstrate the product ratio for

substitution to reduction is dependent on the initial con-

centration ratio of thiophenoxide ion to methoxide ion. At

high ratios of thiophenoxide ion to methoxide ion substitu-

tion is the predominant pathway for the reaction of 3-iodo-

pyridine.











I
0
_0
-u
0






C)
*ctj
O
0
1-'4 -H
0
0)
O

CJ -
O

u O
,J .E
u U




WC
u
0 z
o3

LO C
r4

r- 0
C :


O

-4
0 C






0O
C o
r-i










0
U
4-
0


LJC







C
U



0






IC
0 .
C 0


wu>


-4
C
*-













O C
SI-4










O
L -4









I.-4 -












C)
= I-







































cu
TO E














m



U'
u















en

U


r-- a' r- 0 Dn rn











Ut) C) 0 0 C0 C0



















0 0 -4 0 C) 0
0 CN D0 00 CN co














OI' co r- r-.
1 ONi -n CO

















'D co
CN b"| r-4 r-1
















r- cNJ i '.


,-I

C
o
r =
-0
E 0
E *M




i L,

01
.-:








C
M >




U 0
S0



iu -C

oLi




0 C



3 o














m
S .r-4



E
CO >.













.4
0






















a, ,"
o -o









3 0









O c
Li 4-


















r- C-
00






r-4
O ci







<1 -




3 _
O4 --4


0 *







-4 0

Cu '-


LI









c.
Ol
r-j









0
*r-4





0




Li-4
0






c
Cu


C

0
.1














C u0
.. -4











u




u O
N 04
C

0 3






0



.- .
U 0

i ,Ua
V -Z













U U
Li r _






.0 0 Q


-4


u o0



0
01 01




< UO
01 3







C U
U -0
*r-l O


c 'C
CiV


















0
-I 0
TrI c-


m l
UU


Z 2















LI

C4
CP) C
















co
m













--

Q-4

-0


-N C4 m- Z L ..D


CO r' r ..,

















O
Li --4 o 0







0 c1i r- 0 .D I'D

C14 r 01 o 0














D O r L) Cco
0) r- 0 0 CO 00







The first two entries in Table 6 not only demonstrate

that nigh yields of the sulfide may be obtained, but they

also show that substrate is consumed in the absence of

added methoxide ion. (It was noted previously that reduc-

tive-dehalogenation of 3-iodop,'ridine proceeds extremely

slowly in methanol to which no methoxide ion has been added.)

Hence, for reaction to occur in the presence of thiophen-

oxide ion, added methoxide ion is not a requirement.

However, it does seem that added methoxide ion increases

the extent of conversion of 3-iodopyridine to products.

For experiment 6-1 in which no methoxide ion was initially

present, consumption of substrate is incomplete. In run

6-5, which was initially 1.1 H in sodium methoxide, sub-

strate is totally consumed.

Extended irradiation times do not result in a change

either in the product ratio for substitution to reduction,

or in the concentrations of products. This is evident in

a comparison of entries 6-5 and 6-6. Thus, products appear

to be stable to continued irradiation after substrate has

Deen consumed.

An attempt was made to demonstrate more clearly the

influence of methoxide ion concentration on the rate of

the photoinitiated reaction of 3-iodopyridine with thio-

phenoxide ion. In Table 7 are listed results from three

experiments in which 0.12 M 3-iodopyridine was photolyzed

for 185 minutes in the presence of 1.3 M. sodium thiophen-

oxide. Varying the initial methoxide ion concentration












0


C 0
1 O





i ca
" ----


-4 CN m-


!---
0





Cu


01



Li


I r-4

C
MO

Ui
C -c


0 C








o
0 -



c-i






E-o
(

TO .
r-1 CO

U
C 0





0 -.-1









-0
u
0 0
i1-1
r-l
[--c ,,


Ln -4 r-






.0

























-Ts
4 CN 3 T












r-* -
i ,












o'. co CO









--I N ,-.
























n

i-



0






rl
U

rC
1-







U)

Fu
"O




1i






0
-.l










U)
a)











ZO
C




















E



-4

0

a-



l-i






-,I





-1
Cu






r-i


















0
Cu

iO
*r-l


41







:-4







0
Li
a















*rH
.-I










-,












0







-4







0
l)














u
ia)













0








-o

1--
ri


















rjl


:, 0














03





,o
-ln



















0
|li
*i-
r- V
01







from 0.22 co 0.92 H had only a minor effect on che extent

of reaction. The concencracion of 3-phenylchiopyridine is

invarianc for the three runs. Although the pyridine con-

centracion seems to increase slightly with an increase in

mechoxide ion concentration, the glc determined yields for

chis product in che three runs overlap within the limits

of experimental error. Therefore, mechoxide ion, when

present in excess of one equivalent, has a negligable in-

fluence on che race.

The phocoiniciaced reaction of 3-iodopyridine which sodium

thiophenoxide in mechanolic methoxide was also examined

ac low substrate concentration. Results of chis invesciga-
-2
rion are presented in Table 8. lhen 1.4 x 10 1 3-iodo-

pyridine is phocolyzed in the presence of 0.49 H chiophen-

oxide ion and 0.44 I mechoxide ion, substitution to reduc-

cion product radios are obtained which are approximately

50 percent higher than chose observed for the experiments

in Table 5 in which che initial concentration of 3-iodo-

pyridine was 0.30 H. (The average product ratio of 1.S +

0.3 for Table 8 was calculated using the daca in entries

3-3, 4, 5, and 7. The first two entries were excluded

since the concentration of pyridine is coo low for accurate

quancicacion, and the subscitucion to reduction product

ratios are unreliable. The unusually high value in entry

3-6 was also excluded.) The higher produce ratios in

Table 8 may be due, in part, co difficulties in repro-

ducibly quantitating products by glc analysis for boch















0




y.r





Cs




u -IT
O I


0 0



UO
H -








4-i *H
C
m













4JQ
0 *,











* 0
.-4







4 1 C
c-I



























0C
C .C
*H 'J










-H *H
o ,-o
>. o














-C 0.c

O -C 0
0
rz-i
14-1


Cv -1


m










0 W

O


v'


V0
-E
EE

E














cr'


CO
-4
i








0
N-


(n
rA
E

r-4

cM *<


+1
o-- .4 O0 uI
-i in In


+1
co


T '. 0 0














r- r o 0
C C







-o
-- CNI


- -4

Tm




















In-






0'


E-l



-4-4


0 0' 01'









CO %D -
r-j Ln Ln










-n oN -I


0 r-I


In I r


to


r-4Lr-












CN





30



c 3

0 -
Sr~l 1, ,,-4
S*r-4 0 - .0|
D >. E E x -
0* o E : E 0 0 0,
"0 "- "-
C .t, r--- L f n .. -,-4
O E D 0 0 .C: C



SI 0 0 O
0 .C 0
C ,C .
CL LL E >i


O c.-1 -- O
j C t-Il LI a 0
f i U 0 0' C


03 1 *T








C/-" j 0 ,- >O --4
jr




D C C -4



,
S*r- -4 O
0






r-1 4 -
U 0 o









-m C >. (
C *- 0 -
crl i





--0 -4




0 *.I V0 *r



-, .. (4 E 0

0D 0- -- r--E
L-i Q -









0 4 .1 c
0" 0 0 _-AC *0
0"C 0 r- -



S -O 0 a
r E C 0
I 0 -0 D

o ac r- 4 u 0
C N .C f C -
-U,-r > 0















-4.I CU E 3 > ,-
-00. 0 C C 0 U c a C -
'O C D .rl (j


0 ** 3 O OJ C *1 Q
c- m E fw '-2 C
Li I c j O
C o- r cj )-i --- 'L
0 '0 +1--* .O CL (, LIC
0 0 U 3 O 0
01 3 >. "O (OJ

0 ( *(U (- 0 O1
Hf O ci(3C D u0r


i C 0
0 Cz3 '-. '1 '. D D i [3 C







concentrated and dilute samples. These problems are more

thoroughly discussed in Chapter 5.

Degassing apparently has no effect on the rate of re-

action, as indicated by a comparison of the first two entries

in Table 8. In each entry, the yield of products is about

the same. Degassing also has no effect on the product

ratio. For example, the product ratio for substitution to

reduction in entry 8-3, a sample which was degassed prior

to irradiation, is 1.6. The same product ratio is obtained

in entry 8-4, for a sample which was not degassed.

The experiments in Table 8 also demonstrate that rate

is proportional to light intensity. There is a large

difference in the extent of reaction between the first two

entries and the second set of experiments. In entries 8-1

and 8-2 the majority of substrate has not undergone re-

action after 7 minutes irradiation. In runs 8-3 and 8-4

in which the light intensity is increased by a factor of

-16, substrate is completely consumed at 10 minutes irradi-

ation.

The most striking observation of the study of the

competing reductive-dehalogenation and substitution of

3-iodopyridine in the presence of thiophenoxide ion is the

enormous effect of substrate concentration on reaction

rate. Reaction is completely over in entries 8-3 and 8-4

at an irradiation time of only 10 minutes. In contrast,

for the first seven entire in Table 5, in which the 3-

iodopyridine concentration was initially 0.30 H, reaction







is incomplete at irradiation times of from 20 co 129 min-

uces at the same light intensity. A comparison of the

results from run 5-8 with the data in entries 8-5 through

8-7 also demonstrates that more dilute solutions require

shorter irradiation times for reaction to reach completion

when the same light intensities are employed. In entry

5-8, in which the 3-iodopyridine concentration was iniciall,

0.30 H, 77 percent substrate remains after 171 minutes

irradiation. In run 8-7, in which the initial 3-iodopyridine
-2
concentration was 1.4 x 10 2 H, only 16 percent substrate

remains after 82 minutes irradiation.

Known free radical scavengers were employed as additives

in attempts to innibit the reaction of 3-iodopyridine which

thiopllenoxide and methoxide ions. Inhibition was attempted

for the reaction at both high and low substrate concencra-

tions, Tables 9 through 11. 2-Hechyl-2-nicrosopropane and

;-cerc-bucyl-:i-phenylnicrone, compounds which are often

used as spin labels in esr studies, were employed with

limited success. The use of these compounds is ofcen

severely restricted in chac they react with nucleophiles

to yield aminoxy anions.31 A brief invescigacion by nmr

suggested that these scavengers degrade under the conditions

employed in a typical phocolysis experiment. It has been

reported tnac the nicrone undergoes deoxygenarion when

irradiated with ultraviolet light to yield N-benzylidene-

tert-butyl amine. '" This deoxygenacion product was detected

in significant amounts by both glc and gc-ms analyses of







reaction mixtures (retention time of 16 minutes, column A).

These findings prompted the use of high.mole ratios of the

scavengers relative to the concentration of 3-iodopyridine.

Inhibition was effective only at these relatively high con-

centrations. Inhibition was also attempted with the 3-

carbamoyl-2,2,5,5-tetramechyl-3-pyrrolin-1-yloxy free

radical (CIP), but without success.

Phenyl disulfide, azobenzene, and 1,1-diphenylethylene

were also tested as potential inhibitors. Phenyl disulfide,

an oxidation product of thiophenol, was of interest since

this compound is formed as a side product in the reaction

of 3-iodopyridine with thiophenoxide and methoxide ions.

1,1-Diphenylechylene and azobenzene are often used to in-

hibit radical-chain reactions.34,35

Attempts at inhibition of the reaction at high sub-

strate concentration are presented in Table 9. The use of

2-methyl-2-nitrosopropane was unsuccessful. The CMfP free

radical, N-tert-butyl-a-phenylnitrone, and phenyl disulfide

inhibited the reaction only slightly. Interestingly, when

p-nitrothiophenoxide ion was employed as a potential

nucleophile, it was not only unreactive in that capacity,

but it completely inhibited reductive-dehalogenation. The

latter compound exhibits a broad absorption at 417 nm which

tails into the region in which pyrex transmits ultraviolet

light; lit j (methanol) 415 nm (log E 4.16).36
max
In Table 10, data are presented for three experiments

in which attempts were made to inhibit the reaction of
















0-
C


r-l



O
- X
S0

SC
U



S0










-4
"I -L-J










Li 0
'
OO















CM 0
OCu'







C -0
O O
o




O




-1/ .
o" r








0
0







a; -4r
.. -






















U )


L0
-4 0
-1 0







LJ
0 -r
OLi
UC 3
E







0 0
















C

~0J

on D;




U-
r-4




Cu
tf




O3


o/i


-n





0 0




r-




0 >.



o i



C Ci

0 0
r-4 r-

H












01
>
-4
-Q


-0

<


-4

0
+1














I- C o .0 OC CO
r-4 CN r00 o0 CJ













+1
CD 0 Q0 0Y co
o00 r 00








+1

CoJ r*l CNJ C~) (~










--C
+1
0 r. rC" Co -4
CN CNJ -4 CNJ <







'.0
+1
'0 r--4 0 Co 0
r7 C4 <-1 > J







cDl CNI C)J CNI
I I I I
0 0 0 0
-1 '-, 1-4
X X X X
cN .1 '0 0c'

n1 rC CN Co


,-,--4
- 4 1-1 >
I-4










U
N N-< t O


I I
r-4 C
O
0
3
Li.
I tf
,, ,--I



S --4



.-4
4-1






E E

0 W WD
3 CO


.0 r-



aia



0
0.








i I aa
'4 (T3






C 0 >1 C.>
E 0 rr E0
7 -- 4



r-4 E E

C N U
-l --4l H1
'1- C 3
0 -4
E ,-i r*
J I--I.I









C :3 Li
0E ,-4 fj 0
Ct U "- i L -"-
_0 Li C.-4
r --4 r-
*0 -E
CD > .C,- 4 . ,-

Li 1 L 0 1
4 N1 U












03 I
0 > O r0

C -








-4
C- --j 0 >
03 e 0 4-4
r > 0 0 L C*
rl .* .C
0 -- C 0
3 >1 rl C1 *r-l












0 i- -- 0-






"-0 E i.
ci > > 0 r-

03 *H 0 -
r- LJ 03 01









-4 L -(4
0- r-j 0- Crl
3 a 0 03 O


3 Ia
e s-
EY o-',















I ;-

0 X



O Q




I -C
0i 0








4Li
u 0

O
co r-l




O









C
-i

-I -4







Qa)
C a





00
UrO






O. 0-
C O









0
t--





a-) o
Qo *)
-1A0
O jU)











Q))
O









i 0-4
S*H






C-
, .-4
0 C--








OC
r- i
U -T







a

0


C
L-i
Un C








-r
U)I
3






U)











>- -4










..0

O 4




-4





















.r0
Ol

CO












0 0
') -4



*ll
0



-u-

C0
r -4

>M
*l --*


r-4

0
+1
LrT 0 %Z %.0D

r-j r4 C1N 0-I









Lfn
+1
0 r-. 0









+1
S- 0 '.0 Ln











+1











CM CM r-i 0








0 CM) Cr)

0 0 0
r-4 -4 r-4



cn ol l


- 1-N4 -

1-4
H H
H H H






H N 0 0


N



-O
0
N





--4


O *


cu
-4


0 0

0 C


S-1 1



-O O


























E 2

C) rl
Sr- r
TO >



m c0


0 I






S r-
CT3
o a)
Li I










UL ri
U)


c U
OTO








0 r- H
Er

O,

a1 )




o Lj
in
*0 --







-u -
E

O S




--1 (-

O


0 r
-4 L
O .
-- V


E -




3 n





Ca 0
E E


T J3


O
0
u



--4








r4
E





a)







.-i
Ca
0



0



m




r-4

a)








r.Q
1. -4
0 0
E
CO






SO
O)
L0

S-4













-I
E
-ao
0-



CO











0
4- )













U -

Ca
0











I



CO


0 r










u
O





O X





C13
ci0






U-4







H4 C,
C C

0 0
0 0






OC











rC,
C m





-0 x
0 0
H










PcJ



Sr- Li



C --4 -
ow










-4 --1
o CH U







U >
*C 0 H'






-aj c' .-
Q *O (



0 0)
CL, =
*r-1 ,-
U-i -


0 0


LM CO CN 0r N
(01. O co ar CN -1%










CN C CO CM 0 o
-j m C -4 (1 LCI L11


r- -J
CM i-


a' r-I -O
.--1 H- C.-


CN Co MN (C -.1 % .
CM4 CMl -0 -7 (N -I


- I-


0 0 0)
S>'
ZTZ>


I


Li


4-1
u
.1
Li











-0
C

L,-














































C
F.-
-4
0 )






l-4
G-.





Si
0
LI





I4I
IC













1-4
Sr








CM C
.-I












- 0)







'-4- C
03 >!


C



0
0

LI-








03
U
*rJ










CO
tol
03
















U






I
C





mw
(J




















I .,-1
Sfl

II

















>i)
C T3
EC
I -T
0mE



















Ci i
*O
*r-



CO
u


r- 1 I

-iC

>,m A






r^ j3
C-
fc
03
Lu


Cl
-4

E
*-4



10





-1
03

C
"0







cO
3















'-4
CO




1-4

030


1-
.,-I











GEi


cn



c.
3O


U
C .-
-o




CU C





-0
l.. C


t--4
E U


03





-l
I-,







ECr
0
C -^

EC






CO

U
A)*u
X:
c 3






-2
1.4 x 10 M 3-iodopyridine. The CI[P free radical and 2-

mechyl-2-nitrosopropane are almost ineffective as inhibitors.

Azobenzene inhibits reaction to a slightly greater extent.

However, azobenzene strongly absorbs light in the region

in which pyrex transmits ultraviolet radiation; lit37'
max
(ethanol) 225 nm (log c 4.1), 317 (4.3), 441 (2.7). Hence,

inhibition by azobenzene is probably not due to radical-

chain breaking.

Further attempts at inhibition of the reaction of 1.4
-9
x 10 11 3-iodopyridine are summarized in Table 11. In-

hibition was successful in runs 11-3 and 11-4 in which

relatively high mole ratios of 2-methyl-2-nitrosopropane

(methanol, 292, 17,600) were employed. It also
ma3
appears as though some inhibition may have occurred in run

11-2. Unfortunately, it was not possible to achieve in-

hibition using low concentrations of either the nitroso

compound or the nitrone.

Photoinitiated reactions of 3-bromopyridine and 3-

chloropyridine with sodium thiophenoxide in methanolic

methoxide. Results of a series of experiments in which 3-

bromo- and 3-chloropyridine were irradiated in methanol

containing sodium methoxide and sodium thiophenoxide are

presented in Table 12. As was observed for 3-iodopyridine,

irradiation of the bromo- and chloro- compounds yields a

mixture of pyridine and 3-phenylchiopyridine. When either

compound was allowed to react with a mixture of 0.45 MI

sodium methoxide and 0.49 M sodium thiophenoxide, a sub-




38


sticucion co reduction product ratio was obtained which is

essentially identical, within experimental error, to that

which is obtained when 3-iodopyridine is irradiated under

the same conditions. For example, when a solution which
-2
was initially 1.9 x 10-2 H in 3-bromopyridine (encry 12-1)

was irradiated for 120 minutes a product ratio for substicu-

tion to reduction of 1.6 was obtained. The average product

racio for the experiments involving 3-iodopyridine (ac an
_O
initial concentration of 1.4 x 10 i) in Table 8 is 1.8

+ 0.3. In entry 12-2 when a 0.32 H solution of 3-bromo-

pyridine was irradiated for 768 minutes, the observed pro-

duce ratio was 1.3. The average product ratio for the ex-

perimencs involving 0.30 H 3-iodopyridine in Table 5 is 1.2

+ 0.1. In run 12-4, a subscicution to reduction product

ratio of 1.2 is obtained when 0.32 M 3-chloropyridine is

irradiated for 811 minutes. Only in entry 12-5 does the

product ratio differ, and this value (0.97) was obtained

after -3900 minutes irradiation. The product ratios in

entries 12-3 and 12-6 also seem co be low. It is probable

that 3-phenylthiopyridine degrades on extended sample ir-

radiation, and this degradation would be reflected in lower

product radios. Thus, product ratios ac extended irradia-

cion times may not reflect che reaccivity of che 3-halo-

pyridines.

The low mass balances (70-79 percent) may also be in-

dicacive of sulfide degradation. Unfortunately, mass bal-

ances are not available for runs 12-2 and 12-4 in which








Table 12.



Rurn #
(12- )


1

2b,d

3
4c,d

5

6


Product Distribution Resulting from the Photoinitiatec
Reactions of 3-Chloro- and 3-Bromopyridine with Sodiur
Thiophenoxide and Sodium Hethoxidea

Halopyridine, [NaOCH3 0, [NaSC6H5]0 Irradiation
11 II MI Time, min.


Br-,

Br-,

Br-

Cl-,

Cl-,

Cl-,


-2
1.9x10 2

0.32

0.31

0.32

0.32

0.30


0.45

0.45

1. 5

0.45

0.45

1. 5


0.49

0.49

0.34

0.49

0.49

0.33


120

768

3909

811

4183

3909


aSamples were irradiated with two lamps at a distance of 42 mm.

Small amount of unreacted 3-bromopyridine present.

cLarge amount of unreacted 3-chloropyridine present.

Concentrations of pyridine and sulfide were not determined.

Product ratio is unreliable because of the large error in
quantitation of products at low concentrations.


-









Table 12 extended


11ol ,
Pyridine


.lass Bal.


Sulfide


(-0.073)e


1.2


0 97

(-0.13)e


Substn.
Redn.


1.6

1.3


-74


-70








I (1 )
0 -c


OC
0

I
OT 0


0










ioa
*O-
-'-









() r **


4C
0C %











nl i`y
-^ (LI











*H -4


- C






. -1O







0 0
14-J














< O4




0 CH
Hi CL




CF 0
(T1 5 >-








Lr-4 C
-1



LO a






r^ *H ci

(* o..t(U


U (..





































7




/
/









I
0

0
r-















I 01
r-c

U







>-' L-J
4O









-C
CI .-I



















--I Q
'4-1 r











AC
,2 C 4












0

-4 -- 4
-I '








J- -11 !-4



U C.U
0 0


14 C C
> -I -







.. C t-




140-
CI^ C








, .,E
,- r0
















C4





"3


c)



0

C .-





)U Q)
C C)


4-J


E






.C C -.



TU'.


C) 0 f%



..I


r*H Q*
S. C





U,







mEC
U1 Q) Q







4--1 -H
4J 4J
Mcr










-C,
m c < H
0 -1
0' Lo, .t



u o




4 4



samples were irradiated for much shorter periods of time.

An internal standard was not employed in run 12-2, and

overlap of the glc peaks for 3-chloropyridine and anisole

(internal standard) prevented quantication of products in

run 12-4.

It is significant chat both 3-bromo- and 3-chloropyri-

dine react more slowly than does 3-iodopyridine under sim-

ilar conditions. A comparison of the entries for 0.30 .

3-iodopyridine in Table 5 with entry 12-2 for 3-bromopyri-

dine and entry 12-4 for 3-chloropyridine demonstrates chat

3-iodopyridine is largely consumed after ca. 60 minutes

irradiation, while the other halopyridines are only par-

tially consumed at irradiation times of 768 and 811 min-

utes, respectively. Furthermore, a comparison of Figures

2 and 3, which correspond to entries 12-2 and 12-4, res-

peccively, shows that 3-bromopyridine is consumed more

rapidly than 3-chloropyridine. Thus, the reaccivity order

3-iodo- > 3-bromo- > 3-chloropyridine is established.

Another interesting observation can also be made. In

12-1, in which the initial 3-bromopyridine concentration
-9
was 1.9 x 10 1.1, substrate is no longer present after a

120 minute sample irradiation. After a 768 minute irradia-

tion in 12-2, in which the initial 3-bromopyridine concen-

tration was 0.32 M, there is scill unreacted 3-bromopyri-

dine. Thus, dilution of substrate results in a significant

race acceleration for the photoinitiated reaction of 3-

bromopyridine with thiophenoxide ion.








AIBI initiated reaction of 3-iodopyridine in methanol

containing sodium methoxide and sodium thiophenoxide. It

was of interest to determine whether a phenvlthio-substicu-

tion product could be obtained from 3-iodopyridine using a

nonphotochemical method of initiation. Therefore, AIBM, a

free radical initiator, was employed as an alternative

method for the generation of 3-pyridyL radical in the ab-

sence of light. Results for ATBN initiated experiments are

listed in Table 13.

Three of the five experiments in Table 13 were carried

out at low initial 3-iodopyridine concentration (1.4 x 102

IM) rather than at high concentration (0.30 M) since conver-

sion of substrate to products is higher at the low concen-

cration level. The greater extent of reaction in entries

13-3 through 13-5 may be due to either the higher AIBN to

substrate concentration ratio, or to a dilution effect

similar to that which is observed for the photoinitiated

substitution reactions.

Mass balances in the first two entries are consistent

with those for photoinitiated experiments in which the con-

centration of 3-iodopyridine was initially 0.30 IM. However,

for experiments 13-3 through 13-5, which were carried out

at low substrate concentration (1.4 x 10-2 ), mass balances

are lower than those for corresponding photoinitiated ex-

periments in Table 8. The average mass balance in entries

13-3 through 13-5 is 95 + 3 percent. The lower mass

balances for the three entries in Table 13 are probably due








Table 13. Product Distribucions Resulcing from AIBN Iniciaced
Reaction of 3-lodopyridine which Sodium Thiophenoxide
and Sodium tlethoxide in Mechanol ac 1000 for 60
:linutesa


Run ;,
(13- )


[IaOCH3] 0'
H


1~.


1.4

1.8

0.45

0.60

1.1


aSample was initially
in AIBII.


[NaSC6H5 0,
1.1


0.67

0.33

0.49

0.96

0.48


0.30 N in 3-iodoovridine and 0.19-0.20 M


b -n
Sample was initially 1.4 x 10 -2 in 3-iodopyridine and 3.0 x
10-2 4.0 x 10-2"1 in AIBN. Average data are presented for
duplicate experiments.


Hol "
Iodo-


64

38

39+1

34+4


!Rol "
Redn.


11

33

12+1

11+2

28+2


Mol .
Subsen.


11

11

24+1

32+3

32+1








Table 13 extended


lass Balance


Substn.
Redn.


1.0


0.33


74+1

76+5

63+1


2.1+0.1

3.1+0.7

1.2+0.1


[NaSC6H5 0
[NaOCH3]0


0.48

0.18

1.1

1.6







to glc quantication error which arises from deviation in

FID response over the large concencracion range (0.30 M co

-5 x 10i0 M) for which samples are analyzed. Adjustments

were made to correct for this quantitacion error ac low

concencracions, in the analysis of samples for which phoco-

initiacion was employed; anisole was employed as internal

standard in these experiments. However, biphenyl was

employed as glc standard for the AIBN iniciaced runs, and

the same corrections were noc applicable in chese analyses.

Fortunacely, a deviacion in detector response only affects

the mass balances and not product ratios for subscicution

to reduction; i.e., the yields of pyridine and sulfide are

adjusted proportionately.

Reproducibility of experimental results is excellent

in each sec of duplicate runs except series 13-4. The dis-

agreement in substitution co reduction product ratios for

this series is due to difficulty in accurately determining

pyridine concentration ac low conversion (11 + 2 percent).

Substitution to reduction produce radios are dependent

on che initial concentration ratio of sodium thiophenoxide

to sodium methoxide. In encry 13-3, in which che initial

concentration racio of chiophenoxide ion to mechoxide ion

is 1.1, che average subscicucion to reduction product ratio

for two experiments is 2.1 + 0.1. A decrease in the initial

base concentration ratio to 0.44, in entry 13-5, results

in an average product ratio of 1.2 + 0.1.

Photoinitiated reaction of 3-iodopvridine which phenvl








disulfide in methanolic mechoxide. The experiments in

Table 14 were undertaken to determine whether 3-phenylthio-

pyridine may arise by reaction of the 3-pyridyl radical with

phenyl disulfide. (Phenyl disulfide, an oxidation product

of chiophenol, is formed when 3-iodopyridine is phocolyzed

in mechanol contain sodium methoxide and sodium thiophen-

oxide.) Radical displacement reactions on disulfides have

been reported.38,39,40

In entries 14-1 through 14-4 the initial 3-iodopyridine

concentration was 0.30 M. Except for entry 14-2, mass

balances for this series of experiments are excellent. The

high value in entry 14-2 (108 percent) is probably due to

glc analysis error. In entries 14-5 through 14-8 the
_?
initial concentration of 3-iodopyridine was 1.4 x 10 M.

Hass balances in these experiments are considerably lower

chan in the experiments in which 3-iodopyridine concentra-

tion was initially 0.30 M. The lower mass balances prob-

ably reflect degradation of pyridines during sample irradi-

ation since unidentified products were detected by glc.

Although the major reaction pathway is reductive-de-

halogenation in each of the experiments, substitution

competes effectively. The product ratio for substitution

to reduction is dependent on the initial concentration

ratio of phenyl disulfide to sodium methoxide. This is

demonstrated in a comparison of the results for entries

14-2 and 14-4, in which the initial concentration ratios

of phenyl disulfide to methoxide ion were 0.26 I and 0.14 M,








Table 14.


Product Distribution from the Photoinitiated Reaction
of 3-lodopyridine with 0.13 Phenyl Disulfide in
lethanolic Methoxide


[NaOCH3]0,
1.1


0.49


0.92

0.92

0.49

0.49

0.92


Irradiation
Time, Min.a


226

453

226

450

226

453

226


Samples ere irradiated h one lamp at a
Samples were irradiated with one lamp at a


distance of


Solution wias initially 0.30 11 in 3-iodopyridine.

Solution was initially 1.4 x 10-2 0 in 3-iodopyridine. Residual
phenyl disulfide that did not disolve on mixing, did so during
sample irradiation. Unidentified compounds were detected in
the reaction mixture by glc.

These data were obtained by continued irradiation of the pre-
vious sample which had been irradiated for 226 minutes.


Run i#
(14-


ib,d


4b,d

5c

6c,d

7c


Ilol 7.
lodo-


Mol ",
Redn.


165 mm.







Table 14 extended


Ilass Balance
'o


103

108

103

100

77

82

83


Substn.
Redn.


0.43

0.36

0.32

0.22

0.53

0.46

0.29


[C6H5SSC6, H 5 ,
[NaOCH310


0.27

0.27

0.14

0.14

0.27

0.27

0.14


Mol '
Subs cn.




52



respectively. The respective product ratios in these runs

are 0.36 and 0.22.

Increasing the initial concentration of sodium mech-

oxide clearly increases the yield of pyridine, but not that

of sulfide for a given irradiation time. In entries 14-1

and 14-3 an increase in the initial concentration of sodium

methoxide from 0.49 H to 0.92 H results in an 8 percent in-

crease in the yield of pyridine, while the yield of sulfide

is essentially unchanged. Similarly, in runs 14-2 and

14-4 an increase in methoxide ion concentration increases

only the yield of pyridine. Thus, in these experiments

the extent of reductive-dehalogenation is dependent on meth-

oxide ion concentration, but the extent of substitution is

not.

A significant rate acceleration accompanies dilution

of substrate. In run 14-3, in which substrate concentra-

tion was initially 0.30 H, 74 percent of the substrate

remains after 226 minutes irradiation. In run 14-7, which

represent a 22 fold dilution of substrate, only 11 percent

3-iodopyridine remains after an identical irradiation time.

It is noteworthy that the rate for the photoinitiated

reaction of 3-iodopyridine with phenyl disulfide in mech-

anolic methoxide is significantly less than the rate for

the reaction of 3-iodopyridine with thiophenoxide ion in

methanolic methoxide when both reactions are carried out
-9
at low substrate concentration (-10 M). In entry 14-5,

28 percent 3-iodopyridine remains after 226 minutes irradi-








nation. In entry 8-7, after 82 minutes irradiation at the

same light intensity only 16 percent substrate remains.

At high substrate concentration (0.30 MI) there is a smaller

difference in the extent of conversion of substrate to pro-

ducts for the two reactions. In run 14-1, 83 percent sub-

scrate remains after 226 minutes irradiation. In run 5-8,

which involves reaction of 3-pyridyl radical with thiophen-

oxide ion, 77 percent substrate is present after 171 minutes

irradiation.

Irradiation of 3-iodopyridine in methanol-0-d solution

in the presence of methoxide and thiophenoxide ions. To

demonstrate that 3-iodopyridine is reduced to pyridine via

a radical mechanism and not via an ionic pathway, a control

experiment was carried out using a methanol-0-d solution,

which was initially 1.96 x 10- M1 in 3-iodopyridine, 0.195 '1

in sodium thiophenoxide, and 1.60 M in sodium methoxide.

This mixture was irradiated for 10 minutes (two lamps at a

distance of 42 mm).

Analysis of the product mixture by glc indicated the

presence of pyridine, 3-iodopyridine, and 3-phenylthiopyri-

dine. The pyridine fraction was analyzed by gc-ms to

determine the deuterium content of the molecule.

The mass spectrum (70 eV) of pyridine exhibited a

molecular ion at 79.042 (100 percent) and a P+1 peak with

a relative intensity of 8.5 percent. The calculated P+1

intensity for pure pyridine-d0 is 5.9 percent. Thus, the

observed P+1 intensity indicates that the pyridine is







almost completely deuterium free. The slightly high value

for the observed P+1 contribution may be due to a small

amount of deuterium present at position 4, which was in-

corporated via hydrogen-deuterium exchange prior to sample

analysis (the sample was scored at 00 C for approximately

one week prior to analysis by gc-ms).

Electronic absorption and emission spectra of the 3-

halopyridines. To further understand the mechanism of

photoiniciation in the radical-chain reductive-dehalogen-

acion and substitution reactions of the 3-halopyridines,

ulcraviolec absorption spectra were recorded for 3-iodo-,

3-bromo, and 3-chloropyridine in methanol. An attempt

was also made to record the fluorescence emission spectrum

of 3-iodopyridine in methanol in order to identify the low-

est energy excited state involved in photoinitiation and

to investigate the nature of the effect which molecular

oxygen exercs on the rate of reductive-dehalogenation.

The absorption maxima and molar absorptivities for

3-iodo-, 3-bromo, and 3-chloropyridine in methanol are

listed in Table 15. The absorption spectrum of 3-iodopy-

ridine exhibits the following maxima (E): 272 nm (2755)

and 230 nm (7500). The absorption at 272 nm tails into

the region above 300 nm in which excitacion by pyrex fil-

tered light occurs. Although the absorption spectra have

been reported for 3-bromo- and 3-chloropyridine, and cransi-

tions have been assigned to absorption bands, no assignments
have been made for he spectrum of2,43
have been made for the spectrum of 3-iodopyridine.









Table 15. Ultraviolet Absorption Data for 3-Iodo-, 3-Bromo-,
and 3-Chloropyridine in Methanol


3-Halopyridine


.\max(nm)


Holar Absorptivity
(0)


3-iodopyridine



3-bromopyridine





3-chloropyridine


a
s = shoulder.


272

230

275(s)a

268

262

273(s)

267

260


2755

7500



2802

2421



2856

2486








In the spectra of 3-bromo- and 3-chloropyridine a weak r,'-"

absorption appears as a shoulder on a more intense n-n"' ab-

sorption. These n-r*n' bands occur at 275 nm and 273 nm,

respectively.

In a comparison of the absorption speccra of 3-iodo-,

3-bromo-, and 3-chloropyridine (Table 16), the absorbence

is greatest in the region 280-310 nm for 3-iodopyridine.

This observation can also be made from literature data.

An attempt was also made to observe the fluorescence

emission spectrum of 3-iodopyridine in mechanol. No emis-

sion w.as observed between 200 and 600 nm when 2.24 x 10 M !
-3
and 4.48 x 10- M solutions of 3-iodopyridine were irradi-

ated at 272 and 305 nm. The absence of observable fluo-

rescence strongly suggests that the excited molecule exiscs

in a criplec scace.




57



Table 16. Ultraviolet Absorbence Data for the 3-Halopyridines
in Ilethanol in the Region for Transmittance of
Radiation by Pyrex Class


3-Iodopyridine
(E)


2263

2053

1196

339


3-Bromopyridine
(E)


899

554


3-Chloro-
oyridine
(E)


584

229


116

71

40


295

300








Discussion

A proposed mechanism for the photoinitiated reductive-

denalogenation of 3-iodopyridine in methanolic methoxide The

proposal of a radical-chain mechanism for reductive-dehalo-

genacion of 3-iodopyridine is supported by the following ob-

servations: (a) reductive-dehalogenation is photoiniciaced,

requiring irradiation with ultra.-iolet light; (b) the reaction

requires methoxide ion; (c) formate ion, a methanol oxida-

tion product, is detected in reaction mixtures; (d) a rate

increase is observed when samples are degassed to remove

dissolved oxygen; (e) only a small amount of deuterium is

incorporated into the pyridine ring when 3-iodopyridine is

phocolyzed in mechanol-0-d; and (f) reaction may be inhibited

by some free radical inhibitors. Although some of che

evidence outlined above demonstrates that reductive-dehalo-

genacion must proceed via a radical intermediate, caken

as a whole, the evidence suggests that a radical-chain path-

way is the most reasonable one for reduccive-dehalogenation.

The mechanistic implications of each of the above observa-

tions will be discussed in detail.

It is known chat when aryl and heraryl halides (e:x-

cluding fluorides) are irradiated with ulcraviolet light

of appropriate wavelengths, aryl and hetaryl radicals are
1-14
generated via carbon-halogen bond fragmentation. When

the aryl and hetaryl halides are phocolyzed in the presence

of hydrogen atom donors (usually alcohol solvent) products

resulting from reductive-dehalogenation are obtained.







There are two possible photoinitiation mechanisms by

which 3-pyridyl radical can be generated in the reductive-

dehalogenacion of 3-iodopyridine in methanolic methoxide.

These mechanisms are presented in Scheme II. The first mech-

anism involves photoexcitation followed by homolycic cleavage

of the carbon-halogen bond of the excited molecule, equations

20 and 21. This mode of initiation is generally accepted for

reactions involving aryl and hetaryl iodides, and has even

been postulated as a mechanism for initiation in the reductive-

dehalogenation of aryl bromides and chlorides. 1-5,7,1 The

second mechanism for photoinitiation (equations 22 and 23)

involves electron transfer from an electron donor (D ) to the

photoexcited 3-iodopyridine molecule to generate the radical

anion of 3-iodopyridine. Loss of iodide ion by the radical

anion generates 3-pyridyl radical. Recent esr investigations

have demonstrated that pyridyl and quinolyl radicals generated

by this latter pathway are a-radicals.44



Scheme II

Initiation:

3IPr -h--% 3IPyr (20)

3IPyr Pyr- + I- (21)

3IPyr + D 3IPyr + D. (22)

3IPyr Pyr- + I (23)







lethoxide ion probably serves as electron donor in

equation 22. Several investigators have reported that reduc-

tive-dehalogenation of aryl and hetaryl halides is facilitated

by addition of nucleophiles (e.g., hydroxide ion, cyanide

ion, and aliphatic amines) to reaction media (generally alco-

hol solvents).6,8-10 13,14 It is generally postulated that

these bases serve as electron donors to the photoexcited

aryl or hetaryl halide.

The latter mechanism, involving photoinduced electron

transfer from donor to aryl halide is particularly attractive

as a route to carbon-bromine or carbon-chlorine bond frag-

mentation, since cleavage of these bonds requires greater

energy than cleavage of a carbon-iodine bond. Loss of

chloride or bromide ion by a radical anion is a more feasible

process energetically than photoinduced homolysis of the

carbon-halogen bond.14 However, initiation in the case of

3-iodopyridine probably occurs via both mechanisms presented

in Scheme II.

A species such as III might also serve as electron

donor to photoexcited 3-iodopyridine in equation 22. This

type of n-delocalized anionic o-complex has been suggested as

a possible electron donor in a similar (thermally initiated)

chain reaction. 6,17

The phocoinitiated reductive-dehalogenation of 3-iodo-

pyridine in methanol exhibits a requirement for methoxide ion.

There are two possible explanations for this requirement.

First, methoxide ion may function as the electron donor de-















CH 0' -




III

picted in equation 22. Methoxide ion may also be involved

in a chain propagation step (equation 24). In this reaction,

base (:B ) abstracts a proton from hydroxymethyl radical to

generate the radical anion of formaldehyde. The latter

species is a good reducing agent, and is expected to be an

important chain propagating species.45'46 Thus, an increase

in the concentration of methoxide ion should increase the

concentration of formaldehyde radical anion; this would in-

crease the overall rate of reduccive-dehalogenation.

-CH2OH + :B CH20 + HB (24)

The equilibrium represented in equation 24 might also

explain the leveling effect which is observed with increasing

methoxide ion concentration, Figure 1. Although it was de-

monstrated that rate is dependent on methoxide ion concentra-

tion when methoxide ion is the limiting reagent, addition of

excess mechoxide ion (relative to substrate) has little effect

on the rate. These characteristics might be attributable

to the above described equilibrium. Addition of excess

methoxide ion has little effect on the concentration of







formaldehyde radical anion present relative to the sum of the

concentrations of formaldehyde radical anion and hydroxymeehyl

radical, i.e., this ratio so closely approaches unity with

addition of excess methoxide ion, that further addition of

methoxide ion does not change this value significantly.

It is noteworthy chat reductive-dehalogenation of

3-iodopyridine in methanol is accelerated by bases other than

methoxide ion. For instance, 3-iodopyridine was reduced to

pyridine by irradiation in methanol containing cyanide ion.

The effect of cyanide ion might be explained in terms of

the acid-base equilibrium described above, or cyanide ion

may serve as electron donor to photoexcited substrate.

Parkanyi and Lee have reported chat the photoinduced

reductive-dehalogenacion of 3-bromoquinoline in aqueous

methanol requires the presence of nucleophiles such as hvdrox-
10
ide ion or cyanide ion. They postulated a phoioinduced

electron transfer mechanism as the initiation step, but did

not propose a radical-chain mechanism. In contrast, other

investigators have claimed that reductive-dehalogenation is

accelerated by bases, and that acid-base equilibria, analogous

to equation 24, are involved in formation of chain carrying

radical anion intermediates.6,8,9,11'14 Zoltewicz and co-

workers reported that the thermally induced radical-chain

reductive-dehalogenation of 3-iodopyridine in methanol re-

quires methoxide ion.7

The 3-pyridyl radical, which is formed via equations

21 and 23, may abstract a hydrogen atom from either methanol







or methoxide ion. The products of these reactions, in addi-

tion to pyridine, are hydroxymethyl radical and formalde-

hyde radical anion, respectively. Boyle and Bunnett have

reported thac methoxide ion is 45 10 times more reactive
47
as a hydrogen atom donor than methanol.4 In spite of the

better hydrogen atom donating ability of methoxide ion, this

effect alone cannot explain the rate increases which are

observed with added methoxide ion.

Analysis of reaction mixtures by nmr indicated that

format ion is produced during the reductive-dehalogenation

of 3-iodopyridine. Formate ion could not be quantitated

because of signal overlap with ring protons of 3-iodopyridine.

Formaldehyde, a reaction product, is converted to formate ion

by reaction with hydroxide ion, or by reaction with methoxide

ion to yield methyl format, which is then hydrolyzed by

traces of water. The formation of format ion is consistent

with the finding that format ion is also produced during the

photoinitiated reductive-dehalogenation of 1-halonaphthalenes

in methanolic methoxide. (The latter results are from un-

published work in this laboratory.) It has been reported

that format ion is produced during the photoinitiated radi-

cal-chain reductive-dehalogenation of 4-bromoisoquinoline in

methanolic methoxide.16,17

Proposed propagation steps for reductive-dehalogenation

of 3-iodopyridine are presented in Scheme III. Hydrogen

atom abstraction from methanol and from methoxide ion, as

previously described, yield pyridine, hydroxymethyl radical,







and formaldehyde radical anion. Formaldehyde radical anion

is also formed via reaction of methoxide ion with hvdroxy-

mechyl radical. In equation 28, formaldehyde radical anion

transfers an electron to 3-iodopyridine to continue the chain.

The identities of the termination steps have not been deter-

mined.



Scheme III

Propagation:

Pyr. + C 11O -3 PyrH + CI2OH (25)

Pyr- + CH30 PyrH + CHO2 (26)

*CH2OH + CH3 > CHO2 + CH3 OI (27)

CH20 + 3IPyr > CI120 + 3IPyr (28)



Several experiments (Tables 2 and 3) were undertaken

to determine whether the presence of dissolved molecular
_9
oxygen has any effect on the extent of reaction at low (10

1-I) and high (0.3 1-1) substrate concentrations. One obvious

observation is that removal of dissolved oxygen results in

a significant increase in conversion of substrate to product

at low concentrations of 3-iodopyridine, but not at high

concentrations. A parallel observation is that dilution of

substrate results in an increase in extent of reaction for

degassed samples, but not for samples which have not been

degassed.

Two possible conclusions may be drawn from the above

observations concerning the effects of oxygen and of substrate








concentration on reduccive-dehalogenacion. The first of these

is that oxygen inhibits reductive-dehalogenacion; the magni-

tude of this effect is dependent on substrate concentration,

being much more pronounced at low concentrations of substrate.

An alternative conclusion is that dilution of substrate results

in an increase in extent of reductive-dehalogenacion, but that

this dilution effect is cancelled by the inhibitory effect of

oxygen. The latter conclusion is reasonable since other in-

vestigators have noted that in the photodebromination of het-

aryl bromides quantum yields for product formation increase
9,10
with substrate dilution at given irradiation times.910

The inhibitory effect of oxygen is not easily explained.

Inhibition of reactions by small amounts of dissolved oxygen

is often considered good evidence for radical-chain charac-
6 9
cer. '" However, the use of inhibition by oxygen as a crite-

rion for radical-chain character must be considered cautious-

ly when the reaction is photoinitiated, as in this instance.

Paramagnetic materials such as oxygen can affect singlet-

triplet interconversion in photoexcited molecules via en-
48
hancement of spin-orbital coupling.4 Oxygen can also decrease

tre lifetime of excited triplet species by triplet-triplet
49
annihilation. Thus, inhibition of photoinitiated reactions

by dissolved molecular oxygen may be explained by either a

radical-chain breaking process or via some effect on the

nature or lifetime of the photoexcited molecule.

Inhibition by oxygen has been cited previously as evi-

dence for radical-chain pathways in photoinitiated reductive,

dehalogenations of arvl and hetaryl halides. Other authors,







in investigations of very similar reactions, have cited

inhibition by oxygen as evidence for triplet excited

states.10,14

It is noteworthy chat in previous investigations into

mechoxide ion promoted reductive-dehalogenation of 4-bromo-

isoquinoline, a thermally induced reaction, oxygen had no
16r
effect on reaction rate.' This suggests that the inhibitory

effect by oxygen in the photoinitiated reactions of 3-iodo-

pyridine might be due to an effect on the photochemistry,

rather than a chain breaking process.

Since 3-iodopyridine is a heavy atom containing molecule,

it is likely that it exists in a triplet lowest energy ex-

cited state. This postulate is supported by the observation

that fluorescence was not observed when an attempt was made

to take the fluorescence emission spectrum of 3-iodopyridine

in methanol. Hence, it seems possible that inhibition by

oxygen mighc be due to depopulation of the lowest energy

triplet state via triplet-triplet annihilation.

In a recent report concerning photodebromination of

4-bromoisoquinoline and 3-bromoquinoline Pgrkanyi and Lee

stated that the lowest triple states of pyridine-like hetero-
,10 :.
cycles are usually Ti-n states. A n-1 band was invoked

as the lowest excited state in a recent investigation into

the photodebromination of 5-bromopyrimidines by NJasielski

and Kirsch-Demesmaeker. In both of che above described in-

vescigations photodebromination was inhibited by oxygen.

Parkanyi and Lee proposed that oxygen inhibited reaction via







a photochemical process, while ilasielski and Kirsch-Demes-

maeker postulated a chain breaking process.

Reductive-dehalogenation cannot proceed via an ionic

intermediate since only a small amount of deuterium was in-

corporated into the pyridine ring when 3-iodopyridine was

photolyzed in methanol-O-d containing methoxide ion (thio-

phenoxide ion was also present in this experiment). A hydro-

gen acom should be abstracted from the methyl group of

methanol by 3-pyridyl radical, whereas 3-pyridyl anion should

remove a hydroxylic procon. Thus, a mechanism (equation

29) in which pyridine is formed via electron transfer to

3-pyridyl radical to generate the anion and subsequent proton

transfer, is excluded. Instead, pyridine is formed via

hydrogen acom abstraction from solvent or methoxide ion by

3-pyridyl radical.

Pyvr- e- Pyr CH--30H PyrH (29)

About 2.6 percent of monodeuterated pyridine was formed

in the above experiment. Some of this material was probably

produced via reaction of 3-pyridyl radical with a small

amount (1l percent) of proteo methanol present, which was

formed via reaction of thiophenol and methoxide ion. Deu-

terated pyridine probably was also formed via hydrogen-deu-

terium exchange at the 4-position of the ring and by a small

percentage of deuterium atom abstraction by 3-pyridyl radical

from the hydroxyl group of methanol.41,50

Attempts at using free radical traps to inhibit reduc-

tive-dehalogenation were not entirely successful. Although






the reaction was innibited slightly when N-tert-butyl-o-

phenylnitrone and 2-methvl-2-nicrosopropane were employed as

additives, degradation of these compounds during sample irra-

diation necessitated the use of high mole ratios of the inhi-

bitors relative to substrate. Both compounds are known to

undergo reaction with nucleophiles to yield aminoxy anions.31

An investigation by nmr indicated that 2-mechyl-2-nitroso-

propane undergoes reaction at room temperature in methanol

containing methoxide and thiophenoxide ions to yield un-

determined products. Investigations involving glc and gc-

ms analysis of products indicated that N-benzylidene-cerc-

butyl amine is formed in significant amounts when N-tert-

butyl-,:-phenylnitrone is photolyzed in methanol.

Both 2-methyl-2-nitrosopropane and H-tert-buytl-o-

phenylnitrone are known to be photochemicallv labile. The

nitroso compound dissociates as indicated in equation 30

under the influence of red light; however, reports indicate

it is possible to irradiate solutions which absorb light in

the 280 300 nm region without affecting this reaction.51

It has been reported that the nitrone rearranges to the

2 C 4H9Q lged (CH9)2N-0. + HO. (30)
I- light

oxaziridine IV, when irradiated with ultraviolet light, and
32 33
that oxygen is liberated during irradiation of nicrones.32

Furthermore, it has been noted that N-benzoyl-IN-tert-butvl-

nicroxide V, a stable radical, is produced when either N-tert-

butvl-:i-phenylnitrone or the oxaziridine IV is irradiated

wich ultraviolet light in benzene.33 This nitroxide radical

is not detected when the nicrone is phocolvzed in ethanol.33




















IV












o0

V
v

Obviously, the complex photochemistry of N-tert-butyl-

a-phenylnitrone makes it difficult to reach any conclusions

concerning inhibition of reductive-dehalogenation. This com-

pound absorbs light in the region for transmittance of radia-

tion by pyrex glass, and the nitrone, oxaziridine (IV), or

nitroxide radical (V) may be potential inhibitors.

For several experiments in which the nitrone was employed

as additive, samples were degassed prior to irradiation. De-

gassing may be ineffective when the nitrone is present,

because molecular oxygen is probably eliminated during forma-

tion of N-benzylidene-tert-butyl amine.







In summary, it has been shown conclusively chat 3-iodo-

pyridine is reduced to pyridine in a photoinitiated process

which involves an intermediate 3-pyridyl radical. Although

the proposed radical-chain mechanism outlined in Schemes II

and III represents speculation, there is precedent for such

a mechanism.

A proposed mechanism for the photoiniciaced substitution

reaction of 3-halopy:ridines with chiophenoxide ion in mecha-

nol. The foregoing has demonstrated that the phocoiniciaced

reductive-dehalogenacion of 3-iodopyridine in methanolic

mechoxide proceeds via an intermediate 3-pyridvl radical.

It will now be shown that 3-phenylchiopyridine also arises

via an intermediate 3-pyridyl radical when 3-halopyridines

are photolyzed in methanol in the presence of chiophenoxide

ion. The proposal that reduction and substitution products

arise via competition for a common radical intermediate is

supported by the following evidence: (a) although a signifi-

cant rate increase for consumption of substrate is observed

on dilution of substrate, product ratios for substitution

to reduction do not change appreciably; (b) similar product

ratios are obtained when 3-iodo-, 3-bromo-, and 3-chloro-

pyridine are photolyzed in methanol containing thiophenoxide

ion; i.e., substitution to reduction product ratio is in-

dependent of the identity of the leaving group; (c) the ob-

served reactivity order for the 3-halopyridines is iodo :*

bromo > chloro; and (d) similar product ratios for substitution







to reduction are obtained when two different methods of

initiation, AIBN initiation and photoinitiation, are employed.

It will also be shown that a radical-chain mechanism

involving the intermediate 3-pyridyl radical can account for

the observations which have been made. An SR 1 (Substitution,

Radical, Nucleophile, Unimolecular) mechanism is proposed in

which substitution product is formed via attack of thio-

phenoxide ion on the intermediate 3-pyridyl radical. Alter-

native mechanisms which might account for formation of sub-

stitution product are also considered.

The proposed mechanism for formation of 3-phenylchio-

pyridine and pyridine via a common intermediate, 3-pyridyl

radical, is outlined in Scheme IV. The intermediate radical

may abstract a hydrogen atom from either methanol or methoxide

ion to yield pyridine. Attack by thiophenoxide ion on 3-

pyridyl radical generates the radical anion of 3-phenylthio-

pyridine (VI). An electron transfer step is required to

account for formation of sulfide. The most reasonable elec-

tron transfer step is represented in equation 31, in which

halopyridine serves as electron acceptor. The last two

propagation steps, when combined with Schemes II and III,

constitute a radical-chain mechanism for photoinitiated

substitution and reductive-dehalogenation of 3-iodopyridine

(and the other 3-halopyridines) in methanolic methoxide.

Dilution of substrate results in a significant rate

acceleration leading to the formation of substitution and

reduction products. Furthermore, product ratios for sub-







Scheme IV

Propagat ion:










65

c iS


VI + SCI5 (3



stitucion to reduction do not change significantly upon an

approximate 22-fold dilution of substrate. The latter ob-

servation implies that substitution and reductive-dehalogena-

tion reactions muse have similar kinetic orders and chat

both produces are formed via a common intermediate.

Equation 32 describes the competition between substitu-

tion and reductive-dehalogenation depicted in Scheme IV. The

substitution to reduction product ratio is related to the

rates for attack of thiophenoxide ion on 3-pyridyl radical

and for hydrogen atom abstraction by 3-pyridyl radical from

methanol and methoxide ion, k3, k and k.,, respectively.

Since 3-pyridvl radical is a common intermediate, its con-

centration does not appear in the equation. According to

Substn k63[NaSC6H5 0
Redn k[I CH3OH]o + kH2[aOCH3 ]o








equation 32, the observed product ratio should be independent

of substrate concentration.

It was noted that substitution to reduction product

ratios are about 50 percent higher when reactions are carried

out at low substrate concentrations, than when reactions are

carried out at high substrate concentrations (10-2 I versus

0.3 M). For experiments which were carried out at low sub-

strace concentrations, approximately 10-2 1 in 3-iodopyridine,

pseudo first order conditions were assumed, i.e., insignifi-

cant amounts of thiophenoxide ions and methoxide ions were

consumed. However, for experiments which were run at high

3-iodopyridine concentration (%0.3 M) pseudo-first order

conditions did not prevail, and appreciable amounts of chio-

phenoxide ion and methoxide ion were consumed. Consumption

of mechoxide ion in the latter instance would not appreciab-

ly decrease the rate for pyridine formation because methanol,

a good hydrogen atom donor, is present in large excess;

however, the rate for sulfide formation should decline for

experiments run at high 3-iodopyridine concentrations. Hence,

substitution to reduction product ratios are expected to be

somewhat lower for these latter experiments. However, the

observed differences in product ratios are too large to be

explained using the above argument. The reason for this dis-

parity in product ratios is not known.

It is significant that dilution of substrate results

in increased conversion of substrate to products at given

irradiation times. Dilution of substrate is expected to




7/


increase the extent of reaction if concentrations of other

reactants are held constant. This is so, because in simple

reactions greater than first order, pseudo-first order con-

ditions are more closely approached with dilution of substrate.

Extent of reaction at a fixed time will increase as pseudo

first order conditions are more closely approached. How-

ever, such a change may or may not be sufficient to account

completely for the observed increases in product yields at

given irradiation times, which are encountered with 22-fold

dilution of substrate. A definite conclusion cannot be

reached until the kinetic order is established.

An alternative explanation may be offered to account

for observed increases in extent of reaction with substrate

dilution at fixed irradiation times. These increases may

be due to differences in chain lengths for reactions at

different substrate concentrations. Increases in product

yields with dilution of substrate have been described for

photoinitiated reductive-dehalogenations of aryl and hetaryl

halides in which radical-chain mechanisms have been postu-

lated.6,9

It was noted that sample degassing has no effect on

the increase in extent of reaction which is observed with

substrate dilution, i.e., dilution of substrate results in

an increase in the extent of reaction for degassed and non-

degassed samples. There are two possible reasons for this

observation. An obvious possibility is that molecular

oxygen has no effect on extent of consumption of substrate







for reactions carried out at high and low substrate concen-

trations. The second explanation is that oxygen might retard

reaction if present, but that oxygen is rapidly consumed via

oxidation of thiophenoxide ion to phenyl disulfide, an ob-

served product in these reactions. Thus, product yields at

given irradiation times would increase with substrate dilution

due to "chemical degassing". The latter explanation is more

reasonable, because experiments involving reductive-dehalo-

genation of 3-iodopyridine (Tables 2 and 3) demonstrated

that increases in product yields were observed on substrate

dilution for degassed samples, but not for samples in which

oxygen was present.

Plausible explanations may be offered to account for

the observation that although photoinitiated reductive-de-

halogenation of 3-iodopyridine requires methoxide ion, photo-

initiated substitution with thiophenoxide ion does not require

added methoxide ion. In reactions involving reductive-de-

halogenation methoxide ion may function as either an electron

donor (equation 22) or as a base (equation 24) in an impor-

tant chain propagation step. In contrast, substitution does

not require methoxide ion because thiophenoxide ion may ful-

fill several important functions. First, chiophenoxide ion

may serve as an electron donor (equation 33), or a n-delocal-

ized a-complex such as VII may serve as donor (equation 34).

3IPvr + CgH5S 3IPyr + C H S. (33)

Second, as was previously noted, methoxide ion is not required

in chain propagating reactions involving substitution, because






3IPyr + C6H5S S 31Pyr + C6H5 S (33)




I I
3IPyr + H 1- 3IPyr + H (34)

C6H5S CZ6I5S
VII 5

attack by thiophenoxide ion on 3-pyridyl radical generates

a radical anion which muse transfer an electron, equation 31.

Finally, formaldehyde radical anion is probably formed in

very low concentration via the equilibrium between thiophen-

oxide ion and hydroxymechyl radical, equation 35. This equi-

C6H5S + -CH20H C6H5SH + CH20 (35)

librium lies far to the left, because thiophenol is a sub-

stantially stronger acid chan hydro:.:ymethyl radical. The

pKa of thiophenol in methanol is 8.65, while chat of hvdro:.:y-

methyl radical in water is 10.7.5455 The pKa of hydroxy-

methyl radical in methanol has been estimated to be 15.3.

Two observations were made involving the series of

3-halopyridines which lend strong support to the postulate

that both pyridine and 3-phenylthiopyridine arise via competing

reactions involving 3-pyridyl radical as common intermediate.

The first of these observations is that similar product ratios

are obtained for reactions involving 3-iodo-, 3-bromo-, and

3-chloropyridine. Only at extended irradiation times are

somewhat lower substitution to reduction product ratios ob-

served for the bromo- and chloro- compounds; these lower

product ratios are probably due to sulfide degradation.








(Eegradadon of sulfides is considered in Chapter 3.) The

similar product ratios indicate that both substitution and

reductive-dehalogenation involve a common intermediate. If

substitution involved direct displacement of halide, then

different product ratios would probably be observed. The

second observation is that the reactivity order for overall

consumption of substrate decreases in the order iodo- > bromo-

> chloropyridine. This is not the order which is usually

observed for nucleophilic aromatic substitution (Cl -. Br

> I).56

The reactivity order observed for the 3-halopyridines

is that which would be expected for the radical-chain mechanism

outlined in Schemes II, III, and IV. Several explanations

might be offered to account for the observed reactivities.

Reactivities may reflect (a) different molar absorptivities

for the halopyridines in the region for transmittance of

radiation by pyrex glass, (b) different electron affinities

of the halopyridines, and (c) different carbon-halogen bond

energies.

Comparison of the ultraviolet absorption spectra (Table

16) for the halopyridines demonstrates a decrease in molar

absorptivities in the direction iodo > bromo > chloro in

the region between 278 and 310 nm. If reactivity is related

to molar absorptivity in this region, then the reactivity

order should be iodo > bromo > chloro.

Reactivity should also correlate with differences in

electron affinities of the 3-halopyridines. Relative elec-







cron affinities may be estimated by comparison of redox

potentials or by comparison of rates for capture of the

hydrated electron by these molecules. Although race constants

for capture of the hydrated electron are not available for

halopyridines, values have been determined for iodo-, bromo-,

and chlorobenzene. These rate constants (N sec ) are
10 9 8 57
1.2 x 101 4.3 x 10 and 5.0 x 108, respectively.7 The

reactivities of the halopyridines are expected to follow

the same trend as determined for the halobenzenes, since

this trend is generally observed for aryl halides.5 Aryl

iodides and bromides are known to exhibit very high reactivicy

toward capture of the hydrated electron. In fact, it has

been reported chat iodine bonded co carbon and, to some ex-

cent, even bromine and chlorine interact directly which the

hydrated electron.5

Relative reactivities for the 3-halopyridines might

also be correlated with carbon-halogen bond energies, which

increase in the order iodine < bromine < chlorine. Although

the rate determining step for formation of 3-pyridyl radical

has noc been identified, carbon-halogen bond cleavage in

either equation 21 or 23 may be rate determining. The race

for cleavage of the carbon-halogen bond of either che photo-

excited 3-halopyridine or the 3-halopyridine radical anion

should increase with a decrease in carbon-halogen bond

energy. However, cleavage of the carbon-chlorine or carbon-

bromine bond of the radical anion is more feasible energetical-

ly chan cleavage of this bond in che photoexciced molecule.14







Nucleophilic photosubstitution reactions have been
58
reported for aryl and hetaryl halides.58 It was necessary to

establish that a photosubstitution mechanism was not involved

in formation of 3-phenylthiopyridine. An obvious way to

eliminate photosubstitution as a mechanistic alternative

would be to initiate substitution and reductive-dehalogenation

of 3-iodopyridine in the absence of light, using a free radi-

cal initiator. The thermal decomposition of azoisobutyro-

nitrile (AIBN) was successfully used to initiate both reactions.

In order to compare results for experiments in which

AIBII initiation and photoinitiation were employed, equation

32 was linearized by rearrangement to equation 36. Using

the latter equation, the product ratio can be related to the

methoxide ion and thiophenoxide ion concentrations. Note

that equation 36 requires that the product ratio be given as

reduction to substitution. (The inverse of this ratio has

been used previously in discussions concerning the relative

reactivity of thiophenoxide ion versus methanol and methoxide

ion toward 3-pyridyl radical). Equation 36 describes a

straight line with a slope given by k2/k3 and an intercept

of k1[CH3OH]/k3.

Redn SC ] k2[aOCH3] + kl[CHOH],
d Subs n Ja6 5 o k3 k

(36)

In Figure 5, data for the product ratio and initial

methoxide ion and thiophenoxide ion concentrations are plotted,

according to equation 36, for photoinitiated and AIBN initiated








I
O
Z LJ








0 r 0



0 C,
SU C


\ 1- 0 '"--
\-i -



C 0 CD









"0 *0-..D
cI -- -


o '-0 "O



S CC C


\ \O co4 '




\0\ r01- >
\ \c c _
S- -, 0 0 4 --
o 0 ci
L Cj


--











G 0
\ 4 L*
















I Li
CL 0 -4









-i H C' C-











L u ,- -C
G1 O r- %D L- C- Cr l -.- 0-J-
\ i C















C D
-0 H- X--
U






i. J 1 -4 Ci
*- CO C-
U .' .
-' CoCl .-
AZuQ








reactions involving 0.3 11 3-iodopyridine. In addition, data

for phocoinitiated reactions of 0.3 II 3-bromo- and 0.3 H1

3-chloropyridine are plotted. Two least squares lines were

calculated for the data. One line, which is generated by

the data for experiments involving photoinitiation, has a

slope of 0.515, and an intercept of 0.183. The correlation

coefficient is 0.951. The second line includes data for all

of the experiments; some reactions were initiated by thermal

decomposition of AIBN and others by exposure to ultraviolet

radiation. This line has a slope of 0.408, intercept of

0.232, and a correlation coefficient of 0.954. The points

which represent AIBN initiated reactions fall reasonably close

to the line generated by experiments in which reactions were

photoinitiated. Furthermore, the similarity in correlation

coefficients for both lines suggests that the points for

experiments involving AIBN initiation may be included with

the data for photoinitiated reactions of the 3-halopyridines.

Since AIBN initiation and photoinitiation produce similar

results, photosubstitution cannot be a viable mechanism for

substitution. These results support the postulate that sub-

stitution and reduction product arise via competition for an

intermediate 3-pyridyl radical.

In Figure 6, data are plotted for AIBN initiated and

photoinitiated reactions of 3-iodopyridine at low substrate

concentrations (10-2 H). The straight line with the best

least squares fit has a slope of 0.190, an intercept of 0.190,

and a correlation coefficient of 0.810. Although there is












C C
H *,-4
<0






Ci
r


C -4


0
-4 -4
SLJ -4


ICJ
-i C,





SE
--4




O I O
-44
0 -4 Q).





0O CC;

C *Ha
,- 1-.









S -4U -




L-i C- c
U- 0 *












C) Q LJ
044 a s-
Q3e



.. 0
s-i -








c0 o

0. 0 s-i0


-4.-

S0 HD CW C
QLJ Qj
\upou -
I I I I I l rI)











, C c o c -. n-
UPJ \ w -l















\ 3 0 0 00








\lJ*
\ *us-'




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs