AN EWESTEQATEON OF THE SYNTHESIS
AND BASE ENDUCED REARRANGEMENT
OF SOME DlTHBENYL DIKETONES

Thesis for the Degree of Ph. D.
MESHIGAN STATE UHWERSETY
GEORGE PETER NILLES
1970

.rayrRI;

0-169

Date

This is to certify that the

thesis entitled

AN INVESTIGATION OF THE SYNTHESIS
AND BASE INDUCED REARRANGEMENT

OF SOME DITHIENYL DIKETONES
presented by

George Peter NiIIes

has been accepted towards fulfillment
of the requirements for

Ph. D. Chemistry
degree in __

 
 
 

/

Major profssor

   

8/26/70

 

 

'1

1“, at“

 

Y

lAjCl
UUAV \.

.—
~ -rv'r-——-

  

w
amomc av v

 

! n. 5":
_ 9.. cu

.sity r

v! =0 '«

  

HUAG & SDNS'

unluwl SINGERS

  
   

ABSTRACT

AN INVESTIGATION OF THE SYNTHESIS
AND BASE INDUCED REARRANGEMENT
OF SOME DITHIENYL DIKETONES

By

George Peter Nilles

An investigation of the synthesis and reactivity of various dithienyl

diketones (thenils, III) has been carried out.

£3R——£33 —E£3—R3—~ £33 £33R

III
These compounds, listed ”below, were prepared, for the most part, by

Vilsmeier—Haack formylation of various substituted thiophenes, followed by
cyanide catalyzed condensation of the resulting aldehydes (I) to give the

thenoins (II). Oxidation of the thenoins, usually by CuII

, gave the thenils.
Two of the diketones (VII) and (XV) were prepared by reacting the cor-
responding thienyllithiums with dimethyl oxalate. One of them (IX) was formed
by oxidation of the thenoin prepared by reacting 5-methyl-2-thienylmagnesium
iodide with 2-thienylglyoxal. This new reaction appears to have general syn-

thetic utility. The structures of the thenils were confirmed by nmr, infrared

and ultraviolet spectroscopy.

George Peter Nilles

Various Thenils Prepared In This Investigation

 

 

 

2,2'-thenil (IV) 3,3'-thenil (XI)
5,5'-dimethyl-2,2'-thenil (V) 2-thienylphenyl diketone (XII)
5,5'-dichloro-2,2'-thenil (V I) 5,5'-methoxy—2,2'-thenil (XIII)
5,5'-difluoro-2,2'-thenil (VII) 5,5'-isopropoxy-2,2'-thenil (XIV)
5,5'-di-(2"-thienyl)-2,2'-thenil (VIII) 5,5'-di-(l"-adamantyl)-2,2'-thenil(XV)
5-methyl-2,2‘-thenil (IX) 3,3'-benzo[b]thenil (XVI)
2,2'-benzo[b]thenil (X)

 

Upon treatment with potassium hydroxide, the thenils were rearranged
to the thenilic acids (XVII). These acids, which are unstable at room
temperature in the solid state, were converted to their methyl esters with
diazomethane for characterization and further reaction studies.

Four of the esters, prepared from thenils (IV), (VI), (X) and (XI) were
transesterified with N-methyl-3-piperidinol to give dithienyl isosteres of
JB 336 (XVIII). These amino esters may have significant anti-cholenergic

and psychotomimetic properties (l).

R

.$

@313} —» £33 £3 gr. 03‘3““

XVII CH3 XVIII R

Thenils (XIII) and (XVI) gave anomalous products when rearrangement was
attempted. Thenil (XV) proved to be much to insoluble in organic solvents
for any further investigations.

The kinetics of the thenil-thenilic acid rearrangement were studied in

a 2:l dioxanezwater solution at temperatures of lS-80°.

George Peter Nilles

Rate data are given for compounds (IV) through (XII). The rearrangement of
2,2'-thenil, as determined by the loss in base as a function of time, was
second order overall, analogous to the previously established benzilic acid
rearrangement (2). The rate of rearrangement of 2,2'-thenil was found to be
12.2 times faster than benzil under the same conditions. This may be
attributed to the strong -I effect of the 2-thienyl group (3).

Thermochemical parameters, evaluated from the Arrhenius and Eyring equations
were in accord with the Ingold mechanism for the rearrangement (4).

Determinations of the dissociation constants of nine substituted and
unsubstituted thenoic acids were made at 49.5° with a glass electrode. A new
series of o values, 00’ based on these pK's are suggested for use in cor-
relating reactivities and other physical parameters in Hammett type rela-
tionships for thiophene. Improved correlations in the thenil-thenilic acid
rearrangement, r=0.994 using 06, vs r=0.923 using a were noted.

The strong deviations noted in the Hammett plot and the isokinetic plot
for the halothenils (VI) and (VII) indicate the existence of an equili-
brium step involving the thenil and hydroxide in the rearrangement mechanism.
This is in accord with the Ingold mechanism and at variance with the Ott—
Clark mechanism (5).

In addition, many new fluorothiophenes were synthesized and long range
ring fluorine to side chain nmr couplings were determined.

The previously reported colorations of thienylgylcolic acids in sul-
furic acid were shown to be due to the formation of a-carboxy carbonium
ions. In particular, the ion generated from 2,2'-thenilic acid and methyl

2,2'-thenilate was studied by nmr and visible spectroscopy (6).

George Peter Nilles

References

. J. H. Biel, L. G. Abood, H. K. Hoya, P. A. Nuhfer, and E. F. Kluchesky,
J. Org. Chem., 2Q, 4096 (l96l).

 

. F. H. Hestheimer, J. Amer. Chem. SOC., SB) 2209 (1936).

 

. Salo Gronowitz in "Advances in Heterocyclic Chemistry", Vol. I, A. R.
Katritzky, Ed., Academic Press, New York, N.Y., T963, pp 89-9l.

. S. Selman and J. F. Eastham, Quart. Revs., 1&2 22l (l960).

 

. M. T. Clrak, E. G. Hendley, and O. K. Neville, J. Amer. Chem. Soc.,‘ZZ,
3280 (T955); D. G. Ott and G. G. Smith, ibid., 12, 2325 (1955).

 

. G. P. Nilles and R. D. Schuetz, Tetrahedron Lett., 43l3 (l969).

 

AN INVESTIGATION OF THE SYNTHESIS
AND BASE INDUCED REARRANGEMENT
OF SOME DITHIENYL DIKETONES

By

George Peter Nilles

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1970

9\
a

>1.)
0(\

DEDICATION
To: My Mother and Father who

dont always understand but
who never lose their faith.

ii

ACKNOWLEDGMENTS

The author has on occasions refered to himself, and has been refered
to as "The Great Nilles" with tongue-in-cheek agreement. Such a sobriquet
implies a status like unto a self sufficient being. In demonstrable re-
futation, he wishes to list at random those persons who have performed
invaluable service to his development. He humbly extends acknowledgment to:

Professor Robert D. Schuetz, major professor, for his constant en-
couragement, scientific inspiration, and timeless philosophy; and to
Professor Edward Leete for two fascinating and instructive years as an
undergraduate research assistant; other members of the faculty, and in
particular; Professor Andrew Timnick for his friendship and for serving
on both of the author's examining committees; Professor Harry Eick for
numerous intercessions on the author's behalf both financial and adminis-
trative and for serving as a member of the author's guidance committee;
Professor William Reusch, also a guidance committee member, for many services
and recommendations; Professor Donald Farnum for many kind words when the
author was deeply in need of them; and to Professor Matt Zabik, former fel-
low grad student and future employer, for the loan of various instruments.

Many of the friendships formed here at MSU were probably instrumental
in keeping the author from going ”over the edge” as it were. He hesitates
to name names for fear of overlooking anyone.

Turning the other cheek, the author extends a thank you to those indi-
viduals who, by happenstance or whatever, became the sharp shards in the
author's path. A callous or two will serve to make future encounters of a
like nature more bearable.

He also extends deep appreciation to accounts ll-3653 for financing
his research and to ll-365l for the financing of his existence in the form
of teaching assistantships from Sept., l964 to March, l970---and of course

to the people who administered them.

In everyone's existence, there are those "deities" who guide the de~
velopment of what the person is and will be. In the author's case, Henry
Miller, Salvador Dali, Robert Burns Woodward, and Bob Dylan, and many others
have meant much to him in terms of his atman, his own breath of being, and
they are rightfully acknowledged.

I didn't know half of you haZf’as well as I should like;
and I like less than half of you haZf‘as well as you deserve.

J. R. R. Tolkien

The Lord of the Rings

iv

TABLE OF CONTENTS

INTRODUCTION AND HISTORICAL ...................

DISCUSSION, PART I: SYNTHETIC EXPLORATIONS ...........

Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis

of
of
of
of
of
of
of
of
of
of

2,2'-thenil .................
5,5'-dimethyl-2,2'-thenil ..........
5,5'-dichloro-2,2'-thenil ..........
5,5'-di-(2"-thienyl)-2,2'-thenil .......
3,3'-thenil .................
3,3'—benzo[b]thenil .............
2,2'-benzo[b]thenil .............
2-thienylphenyldiketone ...........
5-methyl-2,2'-thenil .............
5,5'-dimethoxy-2,2'-thenil and 5,5'-

diisopropoxy-2,2'-thenil ...............

Synthesis of 5,5'-di-(l"-adamantyl)-2,2'-thenil and
5,5'-difluoro-2,2'-thenil ...............

Three unsucessful attempts to prepared some thenils . . . .

Synthesis of the thenoic acids ..............

Synthesis of the piperidyl thenilates ...........

DISCUSSION PART II: KINETIC INVESTIGATIONS ...........

SUMMARY .............................

EXPERIMENTAL ..........................

Preparation

Preparation

Preparation

Preparation

Preparation

Preparation

of 2-thenaldehyde ...............
of 2,2'-thenoin ................
f0 2,2'-thenil ................
of 2,2'-thenilic acid .............
of diazomethane ................
of methyl 2,2'-thenilate ...........

Page
I

ll

14
T6
l6
l7
T7
18
20
21
24

25

28
34
45
47

50
89
91

92
92
93
93
94
94

Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation

of
of
of
of
of
of
of
of
of
of
of
of
of
of
of

3,3'-thenil
of 3,3'—thenilic acid

TABLE OF CONTENTS - Continued

2-methylthiophene ............
5-methyl-2-thenaldehyde .........
5,5'-dimethyl-2,2'-thenil ........
methyl 5,5'-dimethyl-2,2-thenilate
5-chloro-2-thenaldehyde .........
5,5'-dichloro-2,2'-thenil ........
methyl 5,5'-dichloro-2,2'-thenilate . . .
2,2'-bithienyl . . . ..........
5-(2'-thienyl)-2-thenaldehyde ......
5,5'-di-(2"-thienyl)-2,2'-thenoin . . . .
5,5'-di-(2"-thienyl)-2,2'-thenil .
methyl 5,5'-di-(2"-thienyl)-2,2'-thenilate
3-bromothiophene
3-thenaldehyde
3,3'-thenoin

of methyl 3,3'-thenilate ..........

of
of
of
of

3-bromothianapthene ...........
3-thianapthaldehyde ...........
3,3'-benzo[b]thenoin ..........
3,3'-benzo[b]thenil ...........

Attempted preparation of methyl 3,3'-benzo[b]thenilate .

Attempted preparation of 3,3'-benzo[b]thenil ......

Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation
Preparation

of
of
of
of
of
of
of
of
of

2-thenoylphenylmethanol .........
2-thienylphenyl diketone

2-thienylglyoxal ............
5-iodo-2-methylthiophene
5'-methyl-2,2'-thenoin

5-methyl-2,2'-thenil ..........
methyl 5-methyl-2,2'-thenilate

2-iodothiophene

2-methoxythi0phene

vi

Page

95

95

96

97

97

98

99

99
100
100
101
101
102
102
103
104
104
104
105
106
107
107
107
108
109
109
110
110
111
111
112
113
113

TABLE OF CONTENTS - Continued

Preparation of 5-methoxy-2-thenaldehyde ........
Preparation of 5,5'-dimethoxy-2,2'-thenil .......
Preparation of 2-isopropoxyth10phene .........
Preparation of 5,5'-diisopropoxy-2,2'-thenil .....
Preparation of 2-thianapthaldehyde ..........
Preparation of 2,2'-benzo[b]thenoin ..........
Preparation of 2,2'-benzo[b]thenil ..........
Preparation of methyl 2,2'-benzo[b]thenilate .....
Preparation of 2-(1'-adamantyl)thi0phene .......
Preparation of 5,5'-di-(l"-adamantyl)-2,2'-thenil . . .
Preparation of 2-fluorothiophene ...........
Preparation of 5-fluoro-2-thenaldehyde ........
Preparation of 5,5'-difluoro-2,2'-thenil .......
Preparation of methyl 5,5'-difluoro-2,2'-thenilate

Preparation of 5-acetyl-2-fluorothiophene .......

Prepartions Directed Toward the Synthesis of 2-Trifluoro-
methylthiophene .....................

Preparation of 5,5,5-trifluorolevulinic acid .....
Attempted preparation of 2-trifluoromethylthiophene . .

Preparations Directed Toward the Synthesis of 2-Phenoxy-
thiophene ........................

Via the Ullmann reaction ...............
Preparation of bis(2-thienyl)iodonium salts ......

Reaction of sodium phenolate with bis(2-thienyl)iodon-
ium salts .......................

Preparation of 2-hydroxythi0phene ...........

Attempted preparation of 2-(2',4'-dinitr0phenoxy)thio-
phene .........................

Preparation of 2-chloro-3,5-dinitr0phene .......
Preparation of 2-phenoxy-3,5-dinitrothiophene .....
Attempted reduction of 2-phenoxy-3,5-dinitrothiophene .
Attempted preparation of 5,5'-dinitro-2,2'-thenil . . .
Preparation of l,l-di-(2'- and 3'-thienyl)ethylene glycol
Degradation of 5,5'-dimethoxy-2,2'-thenil .......
General preparation of thenoic acids .........

vii

Page

114
114
115
116
117
118
118
118
119
120
121
122
122
123
124

125
125
125

126
126
127

128
128

129
130
131
131
132
132
132
133

TABLE OF CONTENTS - Continued Page

Preparation of 2-diacetoxymethy1-5-nitrothi0phene . . . 134
Preparation of 5-nitro-2-thenoic acid ......... l34
Purification of 3-benzo[b]thenoic acid ........ 135

Prepartion of methyl 2,2-di-(2'-thienyl)-2-ethoxyacetate 135
Preparation of N-methyl-3-piperidyl-2',2"-thenilate

hydrochloride ..................... l36
Nonsynthetic Experimental Procedures ........... l38
Reagents ....................... 138
Determination of the carbonium ion Spectra ...... 139
A. The ultraviolet spectra .............. 139
B. The nuclear magnetic resonance spectra ....... 140
Determinations of the ionization constants of the then-
oic acids ....................... 140
Procedure for the kinetic determinations ....... 142
Ionization constant calculations ........... 144
Calculations ..................... 145
REFERENCES ......................... 157

viii

LIST OF TABLES

TABLE Page
1. Spectral Properties of Various Thenils ........... 39
2. Some Statistical Parameters For Imoto's Thiophene Studies . 52
3. Ionization Constants and Sigma Values for Nine Thenoic Acids 55
4. Second Order Rate Constants for the Thenil-Thenilic Acid Re-

arrangement at Various Temperatures ............ 6O
5. Thermodynamic Constants for the Thenilic Acid Rearrangement. 67
6. Rho Values, Standard Deviations, and Correlation Coefficients

for the Thenilic Acid Rearrangement in the Temperature Range

of 15-80° .......................... 71
7. Thenoic Acids Prepared by Oxidation of the Corresponding

Aldehydes ......................... 134
8. Extinction Coefficients for the Di-(Z-thienyl)carboxycarbonium

Ion as a Function of Acid Concentration .......... 139

9. Actual Volume of a Mixture of 100 ml of Dioxane and 50 m1 of
Water as a Function of Temperature ............. 144

ix

LIST OF FIGURES

FIGURE

oowcnmnwm

10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.

. Visible spectrum of an ethanolic solution of sodium hydrox-

ide and 2,2'-thenil .....................
Mass spectrum of compound (XXXV) ............. .
Mass spectrum of benzoin ..................
PMR spectrum of 5-f1uoro—2-thena1dehyde ...........
F19 nmr Spectrum of 5-f1uoro-2-thena1dehyde .........
PMR spectrum of 5-acetyl-2-f1uorothiophene .........
F19 nmr spectrum of 5—acetyl-2-fluoroth10phene .......

Nmr spectra of the aromatic region of various dithienylcar-
binols ...........................

. Nmr spectrum of 2-diacetoxymethy1-5-nitrothiophene .....

Plot of 0 VS 00 .......................
Second order kinetic plot by Equation 9_for 2,2'-thenil at 60°
Arrhenius plot of 2,2'-thenil ................
Arrhenius plot of 5,5'-dimethyl-2,2'-thenil .........
Arrhenius plot of 5-methy1-2,2'-theni1 ...........
Arrhenius plot of 5,5'-dichloro-2,2'-theni1 .........
Arrhenius plot of 5,5'—dif1uoro-2,2'-theni1 .........
Arrhenius plot of 5,5'-di-(2"-thieny1)-2,2'-thenil ......
Arrhenius plot of 2,2'-benzo[b]thenil ............
Arrhenius plot of 2-thienylphenyl diketone .........
Arrhenius plot of 3,3'-theni1 ................
Hammett plot for the thenilic acid rearrangement at 15° . . .
Hammett plot for the thenilic acid rearrangement at 30° . . .
Hammett plot for the thenilic acid rearrangement at 40° . . .
Hammett plot for the thenilic acid rearrangement at 50° . . .
Hammett plot for the thenilic acid rearrangement at 60° . . .
Hammett plot for the thenilic acid rearrangement at 70° . . .
Hammett plot for the thenilic acid rearrangement at 80° . . .

Page

15
23
23
30
31
32
33

42
48
56
59
63
63
63
64
64
64
65
65
65
68
68
69
69
7O
7O
71

LIST OF FIGURES - Continued

FIGURE

28.
29.

30.
31.

32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.

Isokinetic plot for the thenilic acid rearrangement . . . .

Second order kinetic plots for 5,5'-dimethoxy-2,2'-theni1

by equations 2_and §Q_at 80°
Arrhenius plot of 5,5'-dimethoxy-2,2'-theni1

Second order kinetic plots for 5,5'-diisopropoxy-2,2'-theni1

by equations 9_and gg_at 50°

Kinetic plots for 3,3'-benzo[b]theni1 by equations 2_and gl_

Visible spectrum of ion (LXII) in strong acid at 25°

Nmr spectrum of ion (LXII) in ClSO3H-CH

Cl at -55° . . . .

2 2

Nmr spectrum of ion (LXIII) in C1503H-CH2C12 at +30°

Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared
Infrared

Ultraviolet spectra of various thenils
Ultraviolet spectra of various thenils
Ultraviolet spectra of various thenils

spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum
spectrum

of
of
of
of
of
of
of
of
of
of
of
of
of

3,3'-thenil
2,2'-thenil
5,5'-diisopropoxy-2,2'-theni1 .....
5,5'-dimethoxy-2,2'-thenil ......
5,5'-dichloro-2,2'—thenil
5,5'-difluoro-2,2'-thenil
5,5'-dimethy1-2,2'-theni1 ......
5-methy1-2,2'-thenil .........
5,5'-di-(l"-adamanty1)-2,2'-theni1 . .
5,5'-di-(2"-thienyl)-2,2'-thenil . . .
2,2'-benzo[b]thenil
3,3'-benzo[b]theni1
2-thienylphenyl diketone .......

xi

Page

72

79
80

80

82

84

86

87
147
147
148
148
149
149
150
150
151
151
152
152
153
154
155
156

INTRODUCTION AND HISTORICAL

One could justifiably describe Atropa Belladonna L. as quite an eye

 

opener. Courtesans of Louis XIV were wont to apply an effusion of the
berries to their conjunctival sacs, thereby relaxing the mydriatic muscles,
dilating the pupils, and making the eyes appear more attractive, albeit

at the expense of blurred vision (1).

This property of mydriasis is shared by Atropa Belladonna L. (Deadly

 

Nightshade), Hyocyamus Niger L. (Black Henbane), Datura Stramonium L.

 

 

(Jimson weed), and many other members of the Solanaceae family. The active
principle atr0pine (I) and atropines levo isomer, hyoscyamine, form about
0.25% of the roots and leaves of the plants.

/CH

N 3

(I) R: phenyl

“ (II) R: 2-thienyl
H

Its striking physiological properties attracted early attention, and as a
result, the active principles have been known for some 140 years (2).
Pharmacologically, atropine is classed as an anticholenergic and is most

useful clinically as an antiSpasmodic, affecting all smooth muscle tissue.

2

Usually it is given in 0.5 to 2 mg doses to counter the discomfort of
peptic ulcer, arterial spasm, postoperative nausea and motion sickness
as well as being used extensively as a preanesthetic.

It is, unfortunately, a rather general antispasmodic and its non-
specificity in regard to its properties and concomitant side effects,
such as mydriasis, bradycardia, urinary retention, and occular hyperbar-
icity,have led to the medicinal chemist to search for analogs which will
retain the desired qualities and reduce or eliminate the side effects.

One of the most obvious variations in structure is isosteric re-
placement of various noncritical functional groups. Steinkopf (3) suc-
ceeded in synthesizing the thienyl atropine (11), although not in suf—
ficient quantity to permit physiological evaluation.

As the study in this area progressed, it was determined that quite
extensive changes in the gross structure of the parent atropine molecule
could be made while retaining considerable spasmolytic activity.

Biel and Abood (4) have reviewed the structural modifications necessary
to elicit anticholenergic and spasmolytic activity from the atropine-like

molecules whose basic structure is shown below.

O‘C
)n 33 OH 0 l \
3

(III) (IV)

[3‘33

Maximum desired response is obtained when n=1 or 2, R1=methy1 or ethyl,

R2=phenyl, and R3 is cycloalkyl, phenyl, or 2-thienyl.

Many of these compounds had reached the clinical evaluation stage when
it was noted that one of them, designated JB 336 (IV), exhibited potent
psychotomimetic activity in humans (5). Hallucinatory manifestations ranged
from aberrant interpretation of all sensory stimuli to the illusion of
space time regression. While this aspect of the compound limited its use-
fulness as a spasmolytic, it was quickly recognized that it had the poten-
tial to serve as a valuable adjunct in psychotherapy (1,6) analogous to
LSD-25.

It was also discovered, by Biel (7) that JB 336 had 0.6 times the activ-
ity of atropine against acetylcholine induced Spasms in the isolated guinea
pig ileum. Isosteric replacement of one of the phenyls by 2-thienyl gave a
compound which registered approxiametly three times the activity of JB 336.

Consequently, this property became of interest in these laboratories,
and work was directed toward the synthesis of compounds in which both phenyls
are replaced by thienyls and substituted thienyls.

The piperidyl esters (VII) are readily prepared (7) by transesteri-
fication of 2-hydroxy-2,2-bis(aryl)-acetates (v) with N-methyl—3-piperidinol
(VI).

0 AR
II I
3;: 13R OH O'C‘C-OH
CH30 (E OH + ———-> AR
RR '1 3
CH3 CH3

(V) (VI) (VII)

Prior to the initiation of this study only 2-hydroxy-2,2-bis(2'-thienyl)—
acetic acid, i.e. 2,2'-thenilic acid and the 3,3'-isomer had been described

(8,9).

The acids were reported to be unstable, resinifying in a few hours at room
temperature. Esters of these acids were unknown. It was anticipated they
should be accessible via the "thenil-thenilic acid" rearrangement, followed
by treatment with diazomethane. Thus the problem resolved itself into the
preparation of the here-to-fore unknown thenils.

There is certainly no dearth of methods for the preparation of 1,2-
diaryl-l,2-diketones. The methods listed below are intended to be illus-
trative and not exhaustive.

One of the most efficient methods is the direct introduction of the 1,2-

diketone function on an appropriately substituted substrate.
THF

1) R-I + Ni(CO)4 -————a- R-fi-%-R ref (10)
O O
A1C1 " ref (11)
2 R - - -
) H + C1 N g C1 3,I
0 0
3) R-Li + CH3OEEOCH3—Ay- " ref (12)

OO
Oesoxybenzoins may be oxidized to benzils by selenium dioxide in nearly

quantitative yield (13). Since desoxybenzoins may be readily prepared by

the following routes

 

 

 

 

0 0
ll 11
4) R-CH + RCHNZ ; R-C-CHZR ref (14)
9
5) R-CHz-C-NHZ + 12ng 3 " ref (15)
H so H
6) R-CEC-R g 4/ ; " ref (16)
9
7) R-CHZ-t-m + RH f A1C13 # 1' ref (17)

they could serve as a useful basis for the synthesis of the desired thenils.

5

Far more general is the simple and efficient oxidation of benzoins
to benzils using nitric acid (18), sulfuryl chloride (19), cupric-sul—

II (21), or a

fate-pyridine (20), ammonium nitrate in the presence of Cu
stream of oxygen in dimethylsulfoxide (22).

Various procedures have been evolved for the synthesis of benzoins (23).

 

 

 

.. 333
8) R-CCl ; 3°C : R- C-C-R ref (24)
° I
2 H
0
H "M I" " ref (25)
9) R-COH ——9——>
OH
10) R-E-ENHZ JNEBL———_e— ref (23)
0
11) R—C-fiH + RH 31913 a: " ref (23)
H
0 0
co g,_
12) RMgX COC12 ’7 " ref (26)
R -
l3) R-CH CN }- r93 (23)

 

Of these, the most generally applicable is method (13) known as the benzoin
condensation. Recently, a somewhat more complex reaction (14) involving

the condensation of anils with cyanide has been reported. Mechanistically,
it is reminiscent of the benzoin condensation, and furnishes diketanils

which may be hydrolyzed to diketones in quite good yield.

+
CN' AR-C-—C-AR H AR'E“fi'AR
-————€" n n

———->-
,N N 0 0
AR AR

H
14) AR-C=N-AR

ref (27)
In terms of the thienyl aldehydes, this was largely an unknown area, indeed
only 2-thienylcarbaldehyde, 3-thienylcarbaldehyde, and the 2 and 3-thia-

napthylcarbaldehydes had been shown to undergo the benzoin condensation.

6

Method (3) for the synthesis of thenils had not been reported at the
inception of this work. Method (2) is limited to reactions involving sub-
strates with strongly electron donating substituents. Method (1) was found
to be inapplicable to the synthesis of thenils. Therefore, the most direct
approach seemed to be to synthesize the thenils via oxidation of the thenoins,
and the thenoins via the "thenoin condensation" of the aldehydes with cyanide.

Considerable work has been previously reported in correlating Hammett
and Taft sigma parameters in semi empirical equations with structure activ-
ity relationships in biological systems (28). An objective of this inves-
tigation was to synthesize as many 5,53disubstituted piperidyl thenilates

(VIII) as possible in order to determine the physiologic response of the

R
substituent.
\
0
II
O-C-—C-OH
\
'l
/
CH3
R
(VIII)

In addition to being motivated by the possibility of develOping some
highly useful pharmaceuticals, we were intrigued by the observation that
rather scant attention had been paid to substituent effects in the benzil—
ic acid rearrangement.

The discovery of the benzilic acid rearrangement postdated that of the
isolation of atropine by five years. Ninety years later, Ingold (29) pro-

posed a mechanism, shown on the next page, which is well supported by more

recent mechanistic investigations.

0 o 0‘ o 0 0‘ 0 OH

II ll _ k1 112 113 k2 I I _II I
C—C-AR + OH :_ HO-C—C-AR —>HO-C—C-AR —: O-C—C-AR

I<__1 . J l '

AR AR

The first step is the rapid and reversible addition of hydroxide ion to
one of the carbonyls. This is supported by Urey's findings (30) that benzil

18 enriched water containing sodium hydroxide much fas-

exchanges 018 from O
ter than it rearranges to benzilate anion. The exchange occurs, although at a
slower rate, even in the absence of hydroxide ion. Water is unnecessary for
the reaction to occur, as shown by Evans and Dehn (31) who obtained potassium
benzilate from potassium hydroxide and benzil in anhydrous ether.

That proton transfer is most likely not involved in the rate determining
step was shown by Hine (32) who failed to obtain an isotope effect using sod-
ium deuteroxide in DZO-dioxane. Indeed, the reaction was 85% faster in 020-
dioxane than in HZO—dioxane.

In accordance with Ingold's mechanism, the rate of the reaction should

be represented by
d[OH] : k1k2[OH][benzil]

dt k_]

 

 

with the observed rate constant, kr=k1k2/k_]. Westheimer (33) has shown that
the reaction is second order overall, first order in both benzil and hydroxide.
Pfeil et al. (34) have reported half lives for the reaction of benzil,
4,4'-dichlorobenzil, and 4,4'-dimethy1benzil with various bases. They found
that dichlorobenzil rearranges faster than benzil, which in turn rearranges
faster than dimethylbenzil. This is in accord with the postulate that elec—
tron withdrawing groups should increase k],by making C-2 more positive, and k2
by stabilizing the developing negative character at C-1, increasing the elec-

trophilicity of the migration terminus, and weakening the C—1 to C-2 bond.

8

If these statements are valid, they may gain support by observing the
rate of rearrangement of bis substituted compounds and determining if a
Hammett correlation holds according to the following equation: log k/k0= 20p
where k is the observed rate constant of the rearrangement of the substi-
tuted compound, k0 is the rate constant for the rearrangement of the un-
substituted compound, 0 is the log of the ratio of some other physical
parameter of the substituted vs the unsubstitited compound of similar struc-
ture, e.g. ionization constants of the corresponding acids, and o is the
slope of the linear relationship indicative of the electron demand of the
transition state. If both substituents affect the rate of the rearrange-
ment, 0 would be doubled to obtain the required relationship. That both
substituents have an effect on the activation energy may be determined from
the observation of the rate of rearrangement of a mono substituted compound.

An alternate hypothesis to the Ingold mechanism has been postulated by
two laboratories (35, 36). These authors propose that hydroxide attack and
rearrangement occur as a concerted process. That is to say, the established
(30) equilibrium between hydroxide and benzil plays no role in the mech-
anism, but occurs only as a side reaction, although at a rate greater than
the rearrangement. They base their view on the fact that in singley labeled

14

benzils such as (IX) in which one carbonyl is C , the relative migratory

O O
u “*
R C--C
(IX)
aptitude of the substituted vs nonsubstituted ring gave, after correction
for a small isotope effect, a linear op plot.
Thus one goal of the present study was to search for evidence that the

equilibrium between the diketone and hydroxide is a discrete step in the

rearrangement mechanism.

One additional factor must be reckoned with, namely the nature of the
ortho effect of the heterocyclic sulfur. One may take three approaches to
the effects of an ortho substituent. Most experimenters have simply ig-
nored any errors that result in using Hanmett o constants in substituted
thienyl (or other heterocyclic) systems. There is a tendency to do this in
view of the fact that thiophene is a rigid aromatic system, thereby elim-
inating steric perturbations and field effects on 0. However, resonance
and inductive effects would not be expected to be linearly correlated in
both thiophene and benzene. This can lead to scatter in some Hammett plots
that use sigma based on the ionization constants of benzoic acid.

A second approach is to make use of a modified Hammett equation (37-39)

of the type

1°9 3’0 = Ci 1 °Z°I°I
1

where ai and Ci may be constants used to best fit the plot to a linear rela-
tionship and may usually be ascribed to steric effects (Taft equation) and
X is the summation of the 0 values of the individual contributing sub-
stituents.

A third alternative ignores all previous computed values and instead re-
turns to the original definition of the Hammett equation: log E— = on
but uses a different base for computing sigma. Thus, an approprIate base
for the thienyl system might be the log of the ratio of the ionization
constants of the thenoic acids rather than the benzoic acids. Automatically,
this would be expected to take in all effects on the substituents arising
from the substrate whether from inductive, resonance, steric or field effects.
The possibility existed that a simple relationship between these 00 values

BHd(3 might be found. This would extend the usefulness of the large compil-

ation of constants already evaluated for benzene and increase the accuracy

10

of predicting physical parameters for the thiophene ring.

It was a purpose of the present work to investigate such a possibility
in regard to the thenil-thenilic acid rearrangement, dispite the complexity
of the reaction and the uncertainty in the mechanism.

In summary, the purpose of the present investigation was: to synthesize
and describe the properties of a number of thienyl isosteres of JB 336 as
gastrointestinal antispasmodics and potential psychopharmaceuticals; to
investigate qualitatively the substituent effect in regard to the applica-
tion of the benzoin condensation to some thienyl aldehydes; to explore the
substituent effect in synthetic, kinetic, and thermodynamic aspects of the
benzilic acid rearrangement as applied to bis(thienyl)-l,2-diketones (thenils);
to determine quantitative differences, if any, in substituent effects in
thiophene vs benzene; and to attempt a quantitative evaluation of the nature
of the 2-thienyl group as a substituent. In addition, several serendipitous
observations were uncovered and will be fully described in the discussion

section of this work.

DISCUSSION, PART I: SYNTHETIC EXPLORATIONS

It was the choice of the most suitable substituents that guided the
overall design of this phase of the research. There were two points in
the total synthesis scheme where consideration of the substituents was
of paramount importance. The first was the thenil-thenilic acid rearrange-
ment, i.e. (XII) to (XIII) in which a major share of the total endeavor
was concerned with the substituent effect. The second point of concern
was the very last product in the synthetic sequence, namely the piperidyl
thenilates (XV). Substituents had to be chosen which could possibly en-

hance the desired physiologic properties of the parent molecule.

CN [01
[3,7 30" 3[,3;*[33X-1-I§[3
(x) [fa/OHM 1011‘
0H I 22 H
,[33[3 ~§-[3c|[3, :[3 [3R
0 (‘00 of
N

(XV) l (XIV) (XIII)

("J -—-('5-O

It was envisioned that all of the desired thenils could be prepared by

the synthetic sequence outlined above.

11

12

The unsyrrmetrical thenoins (XI) should be accessible by cocondensation of
an equimolar mixture of the two appropriate aldehydes, since it has been
shown that this process usually results in the formation of only one of the
four possible products (23, 40).

Since the first objective in the thenoin condensation study was the syn-
thesis of the aldehydes, the substituent problem had to be resolved at this
point. For a monosubstituted thiophene aldehyde, there are six positional

isomers possible.

CHO R/j%%_:§
\ / \
R IJ33;3§*‘CH0 R”3Z:;E: s

(XVI) (XVII) R (XVIII)
R CHO
2‘3 /\ b

s CHO 5 CHO

(XIX) (xx) (XXI)

From a pharmacologic viewpoint, aldehydes (XVIII), (XX), and (XXI) are un-
desireable, since it has been shown that in the piperidyl benzilates, ac-
tivity decreases when the benzene rings are substituted in such a manner

that reduces their freedom to achieve a coplaner conformation (4l), as in

(XXII).

CH3 (XXII)

In addition, the proximity of the ring substituent to the aldehyde func-

tion would later introduce an obfuscating steric effect in the'Hanmett studies.

l3

Furthermore all of the aldehydes except (XVI) are inordinately difficult to
synthesize, indeed most of them are unknown. Thus, only 5-substituted-2-then-
aldehydes were deemed best suited to the present investigation.

The choice of substituents was further restricted to those which would
survive the strongly basic conditions of the thenilic acid rearrangement.

A more meaningful comparison of substituent effects in thiophene vs benzene
would only be arrived at by evaluations on both the positive and negative
side of the sigma scale using groups having accurately established primary
sigma values. For these reasons, the final choices for substituents were
reduced to methyl, methoxy, chloro, fluoro, benzo[b], phenoxy, and nitro

in order to achieve as wide a range of electromeric effects as possible. In
addition, three other functions were selected; the l-adamantyl group since
its substituent constant had not been previously determined and because of
its desirable phannacologic properties; the 2-thienyl group which would
give an experimental measure of the electron density at the 2-position of
the thiophene ring; and the trifluoromethyl group, of interest since its
substituent effect is probably purely inductive.

The most convenient and efficient method for the preparation of most of
the thiophene aldehydes was found to be by Vilsmeier formylation. Fortun-
ately, electrophilic attack on Z-substituted thiophenes normally gives rise
to considerable 2,5-disubstitution regardless of the nature of the directing
group. Indeed, for all but -I-M substituents, the 2,5-product is the only
one obtained (42). Based on this rather general procedure, the synthesis of

each aldehyde, thenoin, and thenil was undertaken.

l4

0 0
Synthesis of 2,2'-thenil / \ "" / \ (XXIII)
s c-c \\

S

 

Thiophene was reacted with phosphorus oxychloride and dimethylform-
amide to obtain a good yield of 2-thenaldehyde. The aldehyde was condensed
with itself in the presence of potassium cyanide to give the known thenoin
(8). Since it is known (23) that traces of impurities inhibit the benzoin
condensation, an excess of cyanide was used rather than the usual catalytic
amount. It was also observed that the reflux time of the reaction is rather
critical, being about 15 minutes. Shorter reflux time results in consid-
erable unreacted starting material, while longer reaction times result
in much tar formation.

The typical intense color formation observed during the benzoin conden-
sation was noted for the thenoin condensation as well. The reaction of 2-
thenaldehyde and potassium cyanide in ethanol-water produces a deep emerald
green color. Such behavior has been ascribed (43) to radical formation of

the type shown in (XXIV).

AR—fi-—§-AR
0 OH
(XXIV)

The green color of 2,2'-thenoin in ethanolic sodium hydroxide is rapidly
discharged upon shaking in air or acidification. The color persists just long
enough for the visible spectrum of the solution to be determined. Absorption
maxima were recorded at 558, 604, and 655 nm as well as a broad band be-
ginning at 720 nm to beyond the limits of the instrument as shown in Figure l,
on page 15. It must also be considered that the color of the solution may
be due to the presence of ion (XXV) or to a hypothetical complex between the

thenoin and the thenil.

absorbance

15

 

._a
cameo

 

thmON

 

 

I
I l l I I

l L I
430 500 550 600 650 700 750 800 850

 

1 l
I I

 

wavelength, millimicrons

Figure l. Visible Spectrum of an ethanolic solution of sodium hydroxide
and 2,2'-thenoin.

022.213

S
(XXV)

These colors were noted for the other thienyl aldehydes upon reaction
with cyanide in nonaqueous media as well. Indeed, formation of color was
used as a convenient sign indicating whether or not the desired condensa-
tion was occuring.

The best reagent for the oxidation of 2,2'-thenoin to 2,2'-thenil was
iodine and sodium methoxide in methanol. The success of the method depends
on the skill of the experimenter since the iodine must be added and the
reaction mixture quenched as rapidly as possible. Delay causes considerable
tar to form, presumably by reaction of the thenil with sodium methoxide (44).

In this manner, 2,2'-thenil was obtained in 20% yield from thiophene.

I6

/\33/\
Synthesis of 5,5'-dimethyl-2,2'-thenll [/1<:—:}¥-C-C-1[:;:>K\\(XXVI)

CH3 5 CH3

The starting material, 2-methylthiophene,was prepared by Huang—Minlon

 

reduction of 2-thenaldehyde. Vilsmeier formylation of Z-methylthiophene gave
the known 5-methyl-2-thenaldehyde (45). All attempts to prepare 5,5'—
dimethyl-2,2'-thenoin by conventional thenoin condensation conditions in
ethanol-water failed.

It should be pointed out that only sodium and potassium cyanide are
readily effective in catalyzing the benzoin condensation (23), therefore
recourse to other cyanides was not attempted.

However, the characteristic deep green coloration of the condensation
developed at room temperature using dimethyl sulfoxide as the reaction medium.
Unfortunately, only intractable tar resulted when this mixture was quenched
with water. The tar apparently contains some amounts of the thenoin, since
oxidation of it with the cupric sulfate-pyridine complex gave 5,5'-dimethyl-
2,2'-thenil in a 29% yield based on the starting aldehyde. This represents

a 20% overall yield from thi0phene.

O 0
Synthesis of 5,5'-dichloro-2,2'-thenil ’1éf—T§L_3__£_ZZi—3§\\ (XXVII)
c1 5 5 c1

Fortunately, 2-chlorothiophene is one of the more reasonably priced

 

commercially available thiophenes. It was formylated smoothly by reaction
with phosphorus oxychloride and dimethylformamide. Like 5-methyl-2-then-
aldehyde, however, 5-chloro-2-thenaldehyde fails to give an isolable thenoin
upon treatment with cyanide. When the condensation was carried out in
tetrahydrofuran containing 2% dimethyl sulfoxide, an intense prussian blue

color was produced.

l7

Neutralization of this solution with acetic acid, followed by oxidation with
cupric sulfate-pyridine gave the desired thenil in 26% yield based on 2-

chlorothiophene.
Synthesis of 5,5'-di-(2"-thienyl)—2,2'-thenil (XXVIII)

This compound has been described previously (46). The starting aldehyde
is readily prepared by the Vilsmeier formylation of 2,2'-bithienyl(47) (XXIX).
The later is prepared by the cupric chloride catalyzed coupling of 2-lithio-

thiophene (48).

[533/ 3&8“ [/SP\§””me

(XXIX) UHlKCNS
EtOH- H2 0

HOAc- -H20

(XVIII)

The aldehyde undergoes the thenoin condensation and may be easily iso-
lated. Weiss (49) discovered the facile oxidation of a-hydroxyketones to
l,2-diketones using CuII in the presence of annmmium nitrate as the second-
ary oxidant. Aqueous acetic acid is the usual solvent and yields range from
80% to quantitative. Employing this method in the synthesis of 5,5'-di-
(2"-thienyl)-2,2'-thenil gave the desired product in a 35% overall yield
based on thiophene. 9 9

C-—-C
Z E E 5 XXX
Synthesis of 3,3'-thenil / \ / \ 3 )

S S

 

)8

The strong preference for electr0philic attack at the 2-position vs the
3-position in thiophene has been long established experimentally and recently
supported by SCF calculations (50).

The cumbersome Sommelet synthesis Via 3-methylthi0phene to 3-thenylbromide
to the hexamethylenetetramine salt to 3-thenaldehyde (Sl) has been super-

seded by the general method shown below.

nBuLi
HOAc - -60° -60°

Normally, 3-bromothiophene is prepared by tetrabrominating thiophene with
elemental bromine followed by reduction with zinc dust in acetic acid (52).
This results in a 3:2 mole ratio mixture of 3-bromothiophene and 3,4-dibromo-
thiophene. In the present work, a good yield of 3-bromothiophene was obtained
from 3,4-dibromothiophene,available from previous thiophene studies in these
laboratories, by prolonged reflux in aqueous acetic acid with intermittent
addition of zinc dust.

It was known that 3—lithiothiophene must be prepared at -60° or lower.

At higher temperatures, rapid transmetalation occurs to give 2-lithiothio-
phene. The bromothiophenes do not metalate directly, but they may be formed
by transmetalation with n-butyllithium. Gronowitz' procedure (53) gave the
aldehyde as shown above. The aldehyde was reacted with potassium cyanide as
before to give the known thenoin (54). It was easily oxidized to the reported

(9) 3,3'-thenil in l8% overall yield based on 3,4-dibromothiophene.

Synthesis of 3,3'-benzo[b]thenil (XXXI)

 

Electrophilic attack on thianapthene (XXXII) occurs at the 3-position.

19

This is in accord with a large body of experimental evidence and recent

SCF calculations which show that the highest n-electron density is at the
3-position (50). While Vilsmeier formylation does give 3—thianapthaldehyde
(55), the low yield, 9%, prompted the use of a new synthesis. Campaigne pre-
pared the aldehyde in 48% yield via the Sormielet reaction sequence (56).

The method illustrated below gave the aldehyde in 55% yield from thianapthene.
While this work was in progress, this synthesis was reported (57).

Despite numerous attempts, the condensation of 3-thianapthaldehyde with
cyanide gave yields that were l5-20% of theory compared to the 73% yield
reported by Campaigne (56). The oxidation of the 3,3'-benzo[b]thenoin to
3,3'-benzo[b]thenil went smoothly, however, using the cupric acetate-ammon-

ium nitrate method. The melting point of the product did not agree with that

reported.
[::::I:T—:]Br . nBuLi
s “3501““ 18"] 2. DMF 40°11! :CH
(XXXII)
KCN EtOH-H O
2
IE—cH
Cu( WOAC

(XXXI)

Infrared evidence showed the carbonyl absorbance at l640 cm-l, typical of
l,2-diaryl-l,2-diketones. The mass spectrum gave a parent peak at m/e 322
(22%), calculated for 3,3'-benzo[b]thenil, 322. As expected, the molecule
underwent cleavage through the two acyl groups to give a peak at m/e l6l

(l00%) which in turn gave the decarbonylated signal at m/e 133 (29%).

20

Thus the structure was confirmed and the thenil was realized in a 7.7%
overall yield from thianapthene (XXXII).

O 0
I l

S C"‘C 5

Synthesis of 2,2'-benzo[b]thenil (XXXIII)

 

The direct metalation of thianapthene with alkyl lithiums gives 2-sub-
stitution. This is expected since the C-2 hydrogen is more acidic than the
C-3 hydrogen. However, this quick rational is somewhat too naive, since it
does not explain the lithiation of thiophene in the 2-position, in which the
C-3 hydrogen is more acidic than the C-2. A deeper look at the metalation
process (58) is shown schematically below.

[3 :i [39 [3——*[3

S L?//\\\//»\\\ S.+ H 3'I s Li

LI - Ll + CH CH CH CH
' \\\//’\\\ 3 2 2 3

The first step is the coordination of the lithium with the thiOphene sulfur.
Secondly, the C-2 hydrogen is removed by attack of the alkyl carbanion, a
process whose equilibrium lies far to the right since the estimated dif-
ference in pK values between butane and thiophene is about l0. Lastly, tau-
tomeric shift of the lithium to the 2-position occurs.

Accordingly, thianapthene was lithiated and treated with dimethylformamide
to give 2-thianapthaldehyde. The difference in ease with which 2-thianapth-
aldehyde undergoes the thenoin condensation is remarkable compared to 3-thia-
napthaldehyde. An 85% yield of the known (59) 2,2'-benzo[b]thenoin was
realized after 20 minutes of reflux in ethanol-water-cyanide. It will be re-
called that 3-thianapthaldehyde gave only about 20% yield of the thenoin in
the same length of time. Yields of about 50% were obtained from 2-thenaldehyde

and 3-thenaldehyde in the thenoin condensation.

21

Once again, the ammonium nitrate-cupric acetate oxidation was used to
give an excellent yield of the previously unknown 2,2'-benzo[b]thenil. The

overall yield of the diketone was 57% based on thianapthene.

O 0
Synthesis of 2-thienylphenyl diketone {/ \5 g_lcl_© (xxx1v)
S

The mixed benzoin condensation between 2-thenaldehyde and benzaldehyde

 

could result in two possible products.

<3 <1: <3 (”<3 <3§“°<:>

(XXXV) XXXVI)

In practice only one product is isolated, although in the earlier report
of this compound (7) the structure was not determined. Campaigne and
Bourgeois have shown (9) that the condensation of 3-thenaldehyde with

benzaldehyde gives the mixed benzoin (XXXVII).H
l

C—
/ \k u 9-<<:::>
S O OH
(XXXVII)
This was shown by the conversion of the benzoin to its oxime, followed by

Beckmann rearrangement to give benzaldehyde and 3-cyanothiophene.
Naively,it might be assumed that 2-thenaldehyde and benzaldehyde should
give the benzoin (XXXV). It is difficult to rationalize this conclusion from

the proposed mechanism for the condensation (23).

- 9:57 _

CN' 9 9H H- c- AR' 9H 9 -HCN 9 9

AR-CHO :32 AR-cH- :1— AR- |c 1/-——e> AR c— c- AR'-—->- AR- c— c- AR'
CN CN CN H H

(XXXVIII)

The cyanide ion should initially attack the most highly unsaturated carbonyl.

22

Both 2-thenaldehyde and 3-thenaldehyde have carbonyl absorptions in the
infrared about 40 cm"1 lower energy than benzaldehyde, 1670 cm.1 for the
former thiophene compounds and l7l0 cm-1 for benzaldehyde. Therefore, the
cyanide ion would be expected to attack the benzaldehyde carbonyl prefer-
entially and benzoin (XXXVI) should be formed.

However, since the benzoin condensation takes place in a slightly basic
milieu, it is reasonable to assume that ion (XXXVIII) can tautomerize to the
structure in which the carbonyl is adjacent to the more electron donating
aromatic center thus increasing the conjugation path of any electron donating
substituent. Since thienyl appears to be a better electron donor than phenyl
(60) (more saturated carbonyl) it is likely that the initially formed ion
tautomerizes to give benzoin (XXXV).

A firmer basis for the structure of the benzoin was sought. Ideally, the
mass spectrum might reveal the difference in the two proposed structures
if cleavage occurred as shown.

m=lll m=l07 m=ll3 m=l05
I OI IOH 3 l j I l
i/ 3 33%;) @939
5 ' S 3 CG
H H

The mass Spectrum of the mixed benzoin failed to Show a parent peak at

 

 

m/e 2l8, but did give rise to a very weak P-2 signal at m/e 2l6 (5%), shown
in Figure 2, page 23. The signals at m/e ll] (55%), m/e 105 (l00%) and the
absence of a parent peak indicate that the molecule probably oxidizes as

shown upon electron impact.

+

+
0 OH 0 -OH 0 OH 0 q
'I I _e" H l -H' II II -H II I '
AR—Cs—(IZ-AR' ——.—, AR—(Ec-AR' ——- AR-C —C-AR' —-———>- AR-C- -AR

H H

% relative intensity

% relative intensity

l00

KO
0

00
O

\J
O

05
O

01
O

8

00
O

100

90
80

70

60

50

30

20

IO

23

 

 

 

 

 

 

r— l05
r—n—n
lll

_.
e- 40 77
- 44
__ 39

5I

l216
l T r

 

I I I I T I I I
20 40 60 80 100 120 I40 I60 I80 200 220 240 m/e

Figure 2. Mass spectrum of compound (XXXV)

 

 

 

 

 

 

 

 

105 I”% intensity X l0
.__.. /,
/
/
"' x’
/
/
3"" /
/
’l
‘" 77 I’ I22
I
.__ I
: l35
*" : 150 2l0
I
__ 40 5l I
l
__ I
I
I
ll I .
I I" I I If I I I l T I I
20 40 60 80 l00 l20 T40 l60 l80 200 220 240

m/e
Figure 3. Mass spectrum of benzoin

24

No precedent exists for this in the literature since the mass spectra of
benzoins have apparently been neglected. A determination of the mass
spectrum of benzoin in these laboratories reveals the same behavior as
the mixed benzoin (XXXV); it showed no parent and only a very weak P-2
peak at m/e 2l0 .

The structure of the mixed benzoin was finally elucidated by its unam-
biguous synthesis. The reaction of 2-thienylglyoxal with phenylmagnesium
iodide gave compound (XXXV). Compound (XXXV) proved to be identical with
the benzoin resulting from the condensation of 2-thenaldehyde with benzald-
ehyde by melting point, mixed melting point, and infrared spectrum. In
terms of the of synthesis of the thenil, it didn't matter, of course, which
benzoin had been obtained.

Oxidation with ammonium nitrate-cupric acetate gave the known (7) 2-thienyl-

phenyl diketone in 5l% overall yield from 2-thenaldehyde.

Synthesis of 5-methyl-2,2'-thenil (XL)

 

Several attempts were made to condense 2-thenaldehyde with 5-methyl-2-
thenaldehyde. The procedure used in the synthesis of 5,5'-dimethyl-2,2'-
thenil and 5,5'-dichloro-2,2'-thenil also failed to give the desired pro-

duct. It was obtained by the application of the new procedure illustrated.

/ \

(XXXIX)
CuSO4-pyridine

 

<33 E<3

25

Selenium dioxide oxidation of 2-acetylthiophene gave 2-thienylglyoxal (61).
Although 5-methyl-2-iodothiophene had been previously prepared by Steinkopf
(62), by treating 2-thienylmercuric chloride with potassium triiodide, it

was far more convenient to prepare it by treating 2—methylthiophene with
mercuric oxide and iodine analogous to the method described for 2-iodothiophene
(63).

A thorough search of the literature failed to reveal the reaction of
arylglyoxals with Grignard reagents, although the reaction of arylglyoxilic
acids with Grignards to give arylglycolic acids has received considerable
attention (9). It appeared reasonable to assume that the more sterically
accessible and more highly unsaturated aldehyde carbonyl might react selec-
tively with a Grignard to give the desired thenoin. Indeed, addition of 5-
methylmagnesium iodide to a solution of 2-thienylglyoxal at -50° gave an
immediate reaction with formation of a brick red adduct. Workup with ammonium
chloride gave a rather poor but practical yield of 37% of the needed thenoin
(XXXIX). The structure proof rested on the subsequent oxidation of this mater-
ial to 5-methyl—2,2'-thenil (XL) in 75% yield by the cupric sulfate-pyridine
method.

- This Grignard reaction with arylglyoxals should be quite general in scope
although no attempt was made to test it further or to find optimum conditions.
It has been previously mentioned that phenylmagnesium iodide reacts with

2-thienylglyoxal to give benzoin (XXXV) in 2l% yield.

Synthesis of 5,5'-dimethoxy-2,2'-thenil and 5,5'—diisopropoxy-2,2'-thenil

/ \ .. / \
133—<33 £3332- <30;2

CH3O

 

(XLI) (XL)II CH3

26

The alkoxythiophenes are readily obtained by diSplacement of a thienyl
halide with alkoxide ion (64, 65). The prefered method for the preparation
of 2-iodothiophene is by oxidative iodination using iodic acid and iodine.
In this manner, a quantitative yield of Z-iodothiophene may be realized (66).

The synthesis of S-methoxy-Z-thenaldehyde has been reported (64). The
heretofore unknown 2-isopropoxythiophene was prepared in poor yield by Sice's
method (64). A solution of sodium isopropoxide in isopropanol was refluxed
30 hours with 2-iodothiophene and cupric oxide to obtain an ll% yield of the
ether. The ether forms a azeotrope with unreacted 2-iodothiophene and the
mixture had to be grignardized to remove the halide.

Previous studies (ll) have shown that oxalyl chloride can be utilized to
give benzils via a double Friedel-Crafts acylation. The method is restricted
to aryl rings bearing strongly electron donating substituents such as alkoxy
or tert-amino. Less reactive aromatic systems require more vigorous condi-
tions under which oxalyl chloride decarbonylates to phosgene, especially in
the presence of lewis acids. This results in the formation of diarylmono-
ketones.

In the present investigation, the alkoxythenils were formed by reaction
of the appropriate alkoxythiophene with oxalyl chloride in carbon disulfide
at 0° using stannic chloride at the catalyst. While the yield is not good,
33% for the 5,5'-dimethoxy compound, it was particularly bad, l7%, for the
5,5'-diisopropoxy thenil.The latter molecule suffers from severe degrada-

tion during the acylation probably of the type shown.

C{+30 QQHf‘jig/[X—Q' flfla» CHZ/lH\CH3

H/'(
CH3 CH

3 resm

27

Indeed, 2-tertbutoxythiophene and a trace of acid give an excellant yield
of the highly unstable 2-hydroxythiophene (67).
It was necessary to exclude the very slight chance that oxalylation had
occurred at the 3-position rather than at position 5 in either or both rings.
Gronowitz has shown (68) that the nmr coupling constants for some 65 di-

substituted thiophenes fall in the range shown below.

H H4 02_3 = 4.90-5.80 Hz

/ \ 02_4 = 1.25-1.70 Hz

“2 5 H5 02_5 = 3.20-3.65 Hz
03_4 = 3.45- 4.35 Hz

The coupling constant in 5,5'-dimethoxy-2,2'-thenil was determined to be
4.4 i.0.2 Hz. This indicates 2,5-substitution but does not firmly exclude
2,3-substitution.

Oxidation of the molecule should give the known 5-methoxy-2-thenoic acid.
The thenil turned out to be surprisingly resistant to all common oxidizing
agents. It was recovered unchanged after treatment with refluxing potassium
permanganate, or aqueous periodic acid at 90° for two hours. Likewise
chromium trioxide in acetic acid at room temperature for 30 minutes had no
effect. It was completely destroyed (no aromatic protons in the nmr ) by
90% hydrogen peroxide for 30 minutes. Finally upon reaction with sodium
cyanide and ammonium chloride (2], 70) followed by reflux with aqueous sod-
ium hydroxide and acidification, 5-methoxy-2-thenoic acid was obtained. The
identity of the cleavage product was confirmed by comparison of the melting

point and infrared spectra to that of an authentic sample of the acid.

Thus the thenil prepared was the desired one. Further characterization of

5,5'-diisopropoxy-2,2'-thenil was not attempted.

28

Synthesis of 5,5'-di-(l"-adamantyl)-2,2'-thenil (XLIII)
and 5,5'-difluoro-2,2'-thenil (L)

The adamantylation of thiophene by l-bromoadamantane and stannic chloride
produces both 2- and 3-(l'-adamantyl)-thiophene in a 2:l mole ratio respec-
tively. Selective chloromercuration serves to separate the two isomers and

the chloromercuri function is readily removed by refluxing hydrochloric acid.

/ \ DMF-POC13 / \

5 s CHO

\ SX—E

(XLIII) (XLIV)H

V

Vilsmeier formylation gave the aldehyde shown (7l) for which no set of
reaction conditions could be found that would give the desired thenoin (XLIV).
A similar situation occurred in the attempted synthesis of 5,5'-difluoro-

2,2'-thenil( (XLIX)

(0I230-'6 --c [\fl" ”MN

n u 1 DMF
\‘E-E: / S\ '————*’ / \ ~EéLa~ S\ II (Miyéi—T§K\F

I
C C

Li F s CHO F I
(XLV) (XLVI) H

DYw—z— <3,r

(XLVII) (XLVIII)

29

The previously unknown 5-fluoro-2-thenaldehyde was prepared as illustrated
on the previous page. An attempted thenoin condensation of the aldehyde gave
only intractable tar under all of the earlier sucessful reaction conditions
for the other thenoins.

A brief digression at this point will be made to consider some new
fluorothiophene chemistry. Despite the ready availability of 2-fluoro-
thiophene (72, 73), its chemical reactivity has not been reported. It was
found that fluorine is more electron withdrawing in thiophene than in ben-
zene, cf p 55.Consequently, it became of interest to determine the position
of electr0philic substitution in this molecule.

The lithiation of 2-fluorothiophene proceeded with facility. The lithium
salt was reacted with dimethylformamide to give an aldehyde which was demon-
strated to be the S-aldehyde (XLVI) on the basis of the nmr Spectrum, Figures
4 and 5, pp 30 and 3l. Remarkably, JF-CHO was 4.2 Hz. The magnitude of this
coupling prompted an attempt to synthesize the S-acetyl compound in the ex-
pectation of being able to observe a side chain ring fluorine coupling.

Treatment of 2-fluorothiophene with acetyl chloride and stannic chloride
in carbon disulfide gave 5-acetyl-2-fluorothiophene. Both the pmr and F19
nmr revealed a coupling constant of 0.45 Hz between the fluorine and the
acetyl protons, Figures 6 and 7, pp 32 and 33. A haloform reaction of the
acetyl compound (XLVII) gave 5-fluoro-2-thenoic acid (XLVIII) which was
identical by melting point and nmr spectrum to the acid prepared by silver
oxide oxidation of 5-fluoro-2-thenaldehyde (XLVI).

Further confirmation of the substituents' positions was based on studies
by Gronowitz (73) who determined the coupling constants in Z-fluorothiophene.

He found JF-H3= 1.62 Hz, J

= 3.07 Hz, JF-H = 3.l0 Hz and JH -H = 3.89 Hz.

F'H4 5 3 4

30

.auxgmu_mcm;3-m-ozo:_<-m

mo Esgpumam mZQ .w mgzmwu

 

 

 

 

 

 

 

P S w m m m a o

_ |\\ _ _ _ _ h

\W NN_.m NWN :._o
.wa, _, t _70
_
WEE
N: e._ I NI w.m fr.
NI com IEE mmszw NI qé #4 N: «é

NI N.< It+.AI

 

 

31

.muzsmupmcmgp-muogoz_wum we Esgpumam LE: m .m mgzmwu
m~
.wkuo__ +

 

NIX muIHon mmmZm _ II .fi.

 

3-N_Uu-m_uu-d

NI oopuzkon mmmzm

NI w.~

NI N.¢

32

P OH

mZP

NI om “IHQHI ammzm

 

NI mu.o:l4y

 

 

 

mcmzcovngLODPWIN-Fxpmumum mo Eggpomam mza .m mgsmwm

N m m a m
_ _ _ _
mo.“ mo.m mN.N

_

NI oom ”IHQHI ammzm

 

NI ¢._. =r=i

NI o.m
.kgfil NI N.¢

 

 

33

m.m.mF~ +

NI om ”IHQHI ammzm

 

 

NI o.m

111+.

 

 

 

 

 

wcmcqovcpogo:_m-mufixpmomum mo Eagpumam LE: m .N mgsmwm

’ID kl! P" b.-’:.
‘1 Il‘ll‘ ‘Xl‘ll {I 1‘

. lINI med

I ¢.~

mP

 

DD {In 'FDEE Elli! Pb! PI 5 erIIDDDI Dill-D IDI'IIIDI t—i
ii 1‘11“1 £111 1‘ {I‘ll 1( ‘1 I‘ll) III «I: #

 

a-N~uu-NFuu-d

34

As may be seen in Figures 4 and 5, the very small "ortho" J is observed

F-H
3
thereby confiming 2,5-substitution. Furthermore, JH -H = 4.2 Hz is in the
3 4
range found for other 2,5—substituted compounds (68) of this type.

Since Wynberg had shown (71) that lithiation of 2—(l'-adamantyl)-thiophene
occurs in the 5-position, the stage was set for the facile application of a
newly reported thenil synthesis. The addition of 2- and 3-thienyllithium
to dimethyl oxalate at -70° gave 2,2'-thenil and 3,3'-thenil respectively,
(l2). Application of this method to 2-fluorothiophene gave the desired 5,5'-
difluoro—2,2'-thenil in 7% yield from thiophene. The adamantylthenil (XLIII)

was prepared from 2-(l'—adamantyl)-thi0phene in the same manner in 9% yield

based on l-bromoadamantane.

Three unsucessful attempts to prepare some thenils

 

Not all of these attempts to prepare the thenils met with sucess. The
nitrothenil could not be prepared due to failure of 5—nitro-2-thenaldehyde
to undergo the thenoin condensation. The other methods, such as the double
Friedel-Crafts and the reaction of thienyl lithiums with dimethyl oxalate
were clearly not applicable. An attempt was made to nitrate 2,2'-thenil.
This should have resulted in a presumably separable mixture of nitro
products. A solution of 2,2'-thenil could be recovered unchanged after
heating for 30 hours at 600 in a mixture of 90% nitric acid and acetic
acid. Further attempts to prepare this compound were abandoned.

Kabbe (74) in these laboratories had made an earlier futile attempt to
synthesize 2-trifluoromethylthiophene (LI) by the reaction of phenylsulfur-
trifluoride with 2-thenoic acid. In the present study, an effort to prepare
this compound by treatment of the sodium salt of 5,5,5-trifluorolevulinic

acid (LII) with phosphorus heptasulfide met with similar success.

0 0 0 0
.. u l. Na H II 1. NaH / \
CF3-C-0Et + (EtO—C-CH2-)2 ___,.CF3-C—CH2CH2-C-0H -_—-;- s CF3
2. H2504 2. P457
(LII) (L1)

The Ullmann reaction is a well established method for the synthesis of
diaryl ethers (75). Since thiophenes readily undergo nucleophilic attack
to give alkyl aryl ethers (64, 65) and dithienylsulfides (76, 77), there
was ample precedent for the synthesis of 2-phenoxythiophene by the Ullmann
reaction.

The reaction failed under all reasonable reaction conditions; either
starting material was recovered in most cases or the reaction mixture was
totally degraded. Reaction times ranged from one hour to one week; temper-
atures up to the boiling point of phenol were employed. The usually ef-
ficacious copper salts failed to have the desirable result.

It is known that the usual mechanism for nucleophilic displacement of
an aryl halide from a nondeactivated aromatic center involves a benzyne
intermediate (78). No really firm evidence for thiophyne (dehydrothiophene)
as an intermediate in any reaction has ever been presented (78, 79).

From this one might infer that nucleophilic displacement of a thienyl halide
occurs exclusively through an addition—elimination mechanism rather than
elimination-addition. A thiophyne intermediate would be expected to result
in the formation of some 3- as well as 2-substituted product (cine substi-
tution). Such an occurrence has not been reported.

Neglecting steric factors, the addition-elimination reaction should be
favored as the polarity of the carbon—halogen bond increases. In the reaction
of 2-halogeno-5-nitrothiophenes with piperidine, which because of the nitro
substituent almost certainly proceed by the addition-elimination mechanism

the relative rates were: Cl, l.00; Br, 0.64; I, 0.076, (80).

36

However, 2-bromo, 2-chloro, and 2-fluorothi0phene gave the same results as
2-iodothiophene when treated with phenolate.

The use of a more labile leaving group might aid the displacement pro-
cess, especially with weak nucleophiles such as the phenolate ion. Beh-
ringer (8l) in some rather elegant work on phenylating agents, has prepared
diaryl ethers by the reaction of diaryliodonium salts with various phenolates,

under quite mild conditions.

+ +- .
AR-I-AR M O‘AR> AR-O—AR' + M+x‘ + AR-I

X

 

This appeared to be quite attractive in the present study since the displace-
ment should be well favored by the adjacent positive charge on the iodine.
Depending on the relative nucleophilicity, there can be competition from
the iodonium salt anion. Only the halothiophenes, resulting from this internal
attack could be isolated when bis(2-thienyl)-iodonium iodide or bromide were
treated with either sodium or potassium phenolate in solvents or neat.

Hurd and Kreuz (82) have prepared 2-phenoxy-3,5-dinitrothiophene (LIII)

in the following manner.

[—INOiONO 3-Acg: / 5\ HNO -HSO _3_2.>.4/ \

Cl 0° NO S NaO N02

2 \O\\lllll
(L111)

It should be fairly straightfoward to remove the nitro groups by reduction

N02 NO

to the amino function, double diazotization, and reduction of the diazon-
ium groups to hydrogen using formaldehyde. While the aminothiophenes are
notoriously unstable compounds, 2-aminothiophene only recently was adequately

characterized (83), they are quite stable as their chlorostannate salts (84).

37

All attempts at reduction of (LIII) with tin and hydrochloric acid gave a
vigorous evolution of hydrogen sulfide and resinification of the entire
reaction mixture.

Next this process was examined from the reverse direction. An attempt
was made to synthesize 2-(2',4'-dinitrophenoxy)thiophene as illustrated,
since it is known that the 2,4-dinitrophenyl moity can survive reduction

to the 2,4-diamino radical (85).

 

1‘ Mg > _H+,_ CH CH
AE:;3&"12. [::::L\ AZT;EX\‘c3\>(: ZZ:§\0H+ 3\\EE;’ 3

C-OO

0 Et3N - 2,4—DNFB

N0
S 0
N02

Excellent yields of 2,4—dinitroaryl ethers have been obtained from the re-
action of 2,4-dinitrofluorobenzene with various phenols in the presence of
triethylamine (86). The reaction of 2-hydroxythiophene, triethylamine, and
2,4-dinitrofluorobenzene was totally anomalous, producing almost instantly
a dark brown tar which was partially soluble in water. Careful examination
of the reaction products failed to reveal any of the expected dinitrophen-
oxythiophene.
Other procedures which were cursorily examined were also in vain. Reference

numbers are given for analogously sucessful reactions with benzene derivatives.

38

These reactions were: The reaction of sodium phenolate with 2-nitrothio-

phene and sodium 2-thienylsulfonate, both neat and in dimethylformamide;

rb/ S\§ ---—%> IP* ref (87)
$03 Na
‘\N
é/ W\> a IP ref (88)
z/’

* intractable products

 

The ring closure of sodium phenylsuccinate with phOSphorus heptasulfide;

P437
l I ———————5- IP ref (89)
O O

[:::]/O ONa
The trapping of aprotically generated benzyne by 2-hydroxythiophene.

NHZ amyl nitrite 0“
I -—-—->- IP ref (79)
COOH

At this point, it was concluded that the synthesis of 2—phenoxythiophene
was definitely not a trivial exercise in well known chemistry.

The H3-H4 coupling constants of the 5,5'-disubtituted-2,2'-thenils were
4.3 :_O.2 Hz. The infrared spectra of the thenils all showed a Characteristic
carbonyl absorption at l630 :.30 cm'] as given in Table l. p 39. The
constant appearance of this absorption, together with the corroborative
nmr evidence and satisfactory elemental analysis, not to mention their
bright yellow to rust orange color, was taken as sufficient justification

for the proposed structures for the thenils.

39

Table l. Spectral PrOperties of Various Thenils

 

 

 

_ 17 +1.:

Compound J (H3-H4) Hz VC=O cm A(109,)

2,2'-theni1 - 1650 310(4.236)
3,3'-theni1 - 1655 273(4.261)
5,5'-dimethoxy-2,2'-thenil 4.5 l620 350(4.335)
5,5'-diisopropoxy-2,2'-thenil 4.5 16l0 354(4.63l)
5,5'-dich1or0-2,2'-theni1 4.2 1633 335(4.278)
5,5'-dif1u0ro-2,2'-theni1 4.4 1625 314(4.482)
5,5'-dimethy1-2,2'-theni1 4.4 1633 324(4.282)
5-methy1-2,2'—thenil - 1635 317(4.236)
5,5'—di-(2"-thienyl)-2,2'-thenil - l620 397(4 775)
5,5'-d1-(1"—adamanty1)-2,2'-theni1 4.5 1630 324(4.557)
2,2'—benzo[b]then11 - 1640 336(4.723)
3,3'-benzo[b]thenil - 1655 3l6(4.459)
2-thienylphenyl diketone - 1620 290(4.0l4)

 

Before any kinetic work or further synthetic investigations could be
carried out, it was necessary to show the nature of the product(s) ob-
tained by the interaction of the thenils with hydroxide.

Three different laboratories had shown earlier (8, 9, 7) that 2,2'-
thenil, 3,3'-thenil, and 2-thienylphenyl diketone respectively rearranged
to give the correSponding thenilic acids when treated with hydroxide (the
thenil-thenilic acid rearrangement).

Of these, 2,2'-thenilic acid and 3,3'-thenilic acid were found to be
singularly unstable in the solid state, resinifying in a desiccator in a
few hours at room temperature. This was found to be the case in the present
studies as well, however, the acids are perfectly stable, for at least two

years at -20° or in solution. Fortunately, their isolation is not necessary.

The ethereal extract obtained from the product isolation of the rearrange-
ment reaction mixture may be treated with diazomethane to give good to ex-

cellent yields of the methyl esters.

4o

R©§_§@R—* RUG HQR RUG HQR
0// \OH o// \ocn

(LIV)

The structures of all of the thenilic esters (LIV) were checked by nmr

o—n-o
0—0-0

3

and showed pr0per field positions and integral ratios for aromatic to methyl

protons. All esters showed a shift in the carbonyl absorptions from 1630 :_30
- . . . ' -l

cm 1 for the thenils to typical aliphatic ester absorptions at l725 :.l0 cm .

All new esters were submitted for elemental analysis and all of them checked

satisfactorily.

One slight doubt remained, however, and that was the possibility that a
l,2 proton shift on the migrating thienyl ring might occur during the thenilic

acid rearrangement.

5:) /'0 0 OR
H043; C//O—a- HO —C/ "_" Ilia—9m) — C/C/
7‘ /
3/ s/ C 3

This would appear to depend on the extent of anionic character of the mi-

 

grating carbon, the relative acidities of the hydrogens at the 2 or 3 posi-

tions, and the relative rate of proton vs thienyl migration. The problem was
amenable to nmr investigation. However, a further complication arose at this
point in that the thiophene rings are not‘magnetically equivalent and one

or both of them suffers from anisotropic perturbation by the ester carbonyl.
To obviate this problem, the methyl thenilates were reduced by lithium alum-

inum hydride to the l,l-bis(thienyl)-ethyleneglycols.

41

OH
04? CH2

OCH

 

3 H0
The aromatic region of their nmr spectra was then compared to the same re-
gion of all three dithienylcarbinols which were unambiguously synthesized

from the 2- and 3-thenaldehydes by reaction with 2- and 3—thienylmagnesium
F OH
I
m ———>- Low
S CHO S g S

/Q—?H
5Q . Q
MgBr < ?H

4 c
(/ S \S [/ S\S 1+ 2/ S\§

The comparison of these spectra are shown in Figure 8, p 42. It is clearly

halides.

Q... ‘

 

evident that neither 3,3'-thenil or 2,2'-thenil undergoes prototropic rear-
rangement during the thienyl migration.

With but three exceptions, 5,5'-diisopropoxy-2,2'-thenil, 5,5'-dimethoxy-
2,2'-thenil, and 3,3'-benzo[b]thenil which gave anomalous products, the re-
arrangement was especially clean and no evidence for the formation of any
other compound or degradation product could be found. This was particularly
important since the kinetic determinations were to be made only by following

the rate of disappearance of the hydroxide.

42

0H OH
I I
oz-c-o2 93-C-93

I 2 '
CHZOH g = ZZ:_E§\\ CHZOH
s

0H
1 I 1
oz-c-e2 93-c—0 9 -c—o
I I '
H H H

 

 

 

l
1

Figure 8. Nmr spectra of the aromatic region of various dithienylcarbinols.

43

Neither kinetic nor synthetic observations concerning the rearrangement
of 5,5'-di-(l"-adamantyl)-2,2'-thenil were possible owing to the low solu-
bility of the compound in virtually all solvents. In the synthetic study
of the rearrangement, the solvent used in all cases was water with the ad-
dition of sufficient dioxane to effect solution at the reaction temperature.
The use of ethanol or other alcohols occasionally lead to cleavage of the
thenil and the formation of methyl thenoates after workup.

The reaction of 3,3'-benzo[b]thenil with potassium hydroxide gave only
3-benzo[b]thenoic acid, isolated as the methyl ester after the usual reac-
tion workup with diazomethane. The ester was identified on the basis of its
mass spectrum. Signals were found at m/e T92 (parent, 35%), at m/e l6l (P-3l,
loss of CH30-, 65%), and m/e l33 (loss of CO from m/e l6l, l7%).

The cleavage of benzils by hydroxide in the absence of alcohols appears
to be without precedent in the literature. It did not occur with any other
thenil studied. The mechanism of this reaction could well be analogous to
Dakin and Harrington's reaction (2l) which was used earlier to degrade 5,5'-
dimethoxy-2,2'-thenil to S-methoxy-Z-thenoic acid. That procedure involved
the cyanide catalyzed cleavage of the diketone in the presence of alcohol

to give an aldehyde and ester derived from the two halves of the benzil.

o ‘0
H II '
AR-c—c-AR —Cl—> AR-CHO + AR-COOR
R-OH

A mechanism is proposed for the hydroxide cleavage reaction which is sim-
ilar to the one proposed by Kwart (70) for the aforementioned cyanide cleavage.
Presumably, the 3-thianapthaldehyde formed is converted to 3-benzo[b]thenoic
acid and 3-hydroxymethylthianapthene by the Cannizzaro reaction. This would

result in a 75% theoretical yield of 3-benzo[b]thenoic acid from the thenil.

44

 

It is uncertain why this one thenil of all the thenils studied should undergo
this anomalous reaction.

When the rearrangement of 5,5'-dimethoxy-2,2'-thenil was attempted, the
thenil was degraded into unidentifiable products. Repeated efforts at iso-
lating the ester after the usual workup from the rearrangement reaction mix-
ture failed. Purely on the basis of olfactory and visual evidence, the fail-
ure to obtain the expected thenilic acid may be rationalized by the forma-
tion of a Z-hydroxythiophene via a transient Meisenheimer type complex (LV).
It is known that such complexes can be formed with hydroxide (90) and from

thiophenes(

W: way—'11.

H0"

”,7/SO\

cn3o,

 

45

Alternatively, one may visualize an SN2 attack on the methoxy carbon. In
either case, the resulting hydroxythenil or hydroxythenilic acid, would
most likely be as unstable as 2-hydroxythiophene itself (92).

It would have been quite tedious to prepare a sufficient quantity of
5,5'-dii50propoxy-2,2'-thenil for synthetic investigations. However, one
kinetic run of this compound indicated the same abnormal type of behavior
as the methoxythenil. The decrease in the apparent second order rate con—
stant for the disappearance of hydroxide compared to the methoxythenil in-
dicates that either of the two decomposition mechanisms may be valid on
the basis of steric arguments.

Two more synthetic endeavors remained to be accomplished; the synthesis
of a series of substituted thenoic acids corresponding for the most part,
to the substituted thenils; and the synthesis of the piperidyl thenilates,
the original prime objective for this research.

Synthesis of the thenoic acids

 

A series of substituted thenoic acids was prepared so that primary sigma
values for substituents on thiophene might be determined. From this data,
a measure of the deviation of 0 values in benzene vs 0 values in thiophene
was evaluated. The results are treated in detail in Part II of the discussion
section of this thesis. The same criteria used in selecting substituents for
the thenils guided the selection of substituted thenoic acids which were
prepared. Accordingly, 5-methyl, 5-methoxy, 5-nitro, 5-chloro, 5-fluoro-2-
thenoic acids as well as 2-thenoic, 3-thenoic, 2-benzo[b]thenoic, and 3-
benzo[b]thenoic acids were prepared. All of these were readily available,
with the exception of 5-nitro-2-thenoic acid by oxidation of the previously

described aldehydes with silver oxide in aqueous sodium hydroxide.

46

Of these,only 5-fluoro-2-thenoic acid had not been previously reported.

As a consequence of the strong -I-M character of the aldehyde func-
tion, the nitration of 2-thenaldehyde gives a mixture of 4— and 5—nitro-
2-thenaldehydes depending on quite critical reaction conditions.

Buu-Hoi (93) nitrated 2-thenaldehyde with fuming nitric acid in acetic
anhydride and obtained the S-nitroaldehyde (LIX) together with a "small
amount" of the 4-isomer. Foye (94) found only 4-nitro-2-thenaldehyde using
nitric acid in sulfuric acid as the nitrating medium. A eutectic mixture
of the 4- and S-aldehydes was reported by Giver (95) who nitrated 2-then—
aldehyde diacetate (LVI) with nitric acid in acetic anhydride. Fournari
awn: Chane (96) repeated both the nitration in acetic anhydride and the hi-
tnwation in sulfuric acid. They found that neither aldehyde was formed, in
aceiric anhydride, but rather a mixture of the 5- and 4-nitro-2-thenaldehyde
diacxetates (LVII) and (LVIII) in an 87:l mole ratio respectively. Nitra-
ticni of 2-thenaldehyde in sulfuric acid gave the 4-nitroaldehyde exclusively.

[From the simplified scheme illustrated, it is evident that the extent of
4-ni tro products depends on the position of the k1/k_1 equilibrium and
wherther or not equilibrium has been reached before introduction of the hi-

trati ng agent.

m 79° [ mp 78°]
k2 N0. [p 1 k4 N02
(/ \5 __.. 247C \ ——=— U
"‘ ' A
5 CH0 N02+ 5 CHO k_4 . s /0 C
1“”
k1“ k_ 1 OAc
+ (LVIII)
+
H ACZO [mp 730] k5 [mp 750]

k
3 g
/S\ 3... 702. / \ /0A. .__k 5 A
3w 2 S |C\H - N02 3 CH0

OAc OAc
(LVI) (LVII) (LIX)

47

It depends on the relative values of k2 and k3 (competitive nitration).
Thus, the products which are isolated depend on the temperature of the
reaction, the order of addition of the reagents, and the position of the
k4/k_4 and k5/k_5 equilibria. The identification of the products is com-
plicated by the fact that all four possible products melt within six de-
grees of each other.

In the present study, 90% nitric acid was added to a 40% solution of
2-thenaldehyde in acetic anhydride at lO°, and the reaction mixture was al-
lowed to stand at least l2 hours at 0°. The only product in evidence was
2-diacetoxymethyl-5-nitrothiophene. The nmr spectrum of the purified mat-
erial revealed an H3-H4 coupling constant of 4.2 Hz, Figure 9, p 48, there-
by confirming 2,5-substitution. No other proton signals were detected in
the aromatic region. A coupling constant of 0.6 Hz of the diacetoxymethyl
proton with the ring H3 proton was noted.

The diacetate was smoothly oxidized to 5-nitro—2-thenoic acid by potas-

sium dichromate in aqueous sulfuric acid.

Synthesis of the piperidyl thenilates

 

The final synthetic effort was directed toward a brief exploration of
the scope of the transesterification of the methyl thenilates with N-methyl-
3-piperidinol. The objective, of course, was the preparation of certain
thiophene isosteres of dB 336 or N-methyl-3-piperidyl benzilate.

The primary interest in these compounds was their potential physiologic
properties, both anticholenergic and psychotomimetic. Thus a sufficient
amount of each amino ester was to be prepared to permit preliminary eval-

uation of these properties. Four of the esters were prepared as shown on p 49.

48

mcmcqowgpocpwcumupxsumexxoumomwnum mo Echumqm LEz .m mczmwd

P S m m N m m 3 m m H o T m-
_ _ _ _ _ _

m2» #

Nzx — “IHQHZ mmmzm

 

 

N: 00— ”IHQHZ awmzm

 

 

NI N.¢

 

 

 

 

 

I;
1

N: N.¢
N: 0.0 II.

 

 

 

49

 

OH N OCH OH
I I
R—C-R a 3 2 R- C -R
1 i I
,C ,C
0’ ‘0CH3 UOH 0, ‘0
N U
(LX) I (LXI) N
CH3 I
CH3
R = 2-thienyl R = Z-thienyl
R = 3-thienyl R = 3-thienyl
R = 5-chloro-2-thienyl R= 5-chloro-2-thienyl
R = 2-benzo[b]thienyl R = 2-benzo[b]thienyl

The sharp decrease in research funding from various agencies of the
federal government and other sources made the initiation of the pharm-
acologic work impossible. For this reason, it was decided to show only
the general nature of the transesterification reaction and not prepare

all possible piperidyl thenilates.

DISCUSSION, PART II, KINETIC INVESTIGATIONS

Certainly, one of the main expedients of the chemist, or any scientist
for that matter, is the reduction of qualitative observations to quanti-
tative data. An extremely useful tool to the chemist is the ability to pre-
dict the extent of change in physical pr0perties for a molecule as a re-
sult of structural modifications of the molecule.

Some 33 years ago, Louis Hammett elucidated the phenomenological cor-

relation in equation 1,

k
109 — = 00 1_
k0

The universal nature and countless experimental and theoretical justifications
for its validity render any background discussion here redundant. Suffice it
to say, the original definition predicated on the reactivities of substi-
tuted benzenes has been extended, with more or less success,to correlate
data involving many other rate and equilibria measurements in all types of
molecules and in addition such physical properties as nmr data, infrared
absorbances, half-wave potentials, and biological activity (97).

The chemist as statistician must be concerned, however, not merely with
whether or not correlations exist between his data and some other body of
factors, but rather with how well_do they correlate.

Considering the Hammett equation in particular, the logs of the ratio of
the rate constants for a reaction of a series of substituted vs nonsub-

stituted compounds are plotted as a function ofcxas defined in equation 2,

0= log-E 2
ref '—

50

51

The slope of the line is defined as 0 ,usually calculated by the least
squares method. The certainty that all the points fall on the line is in-
dicated by r, the correlation coefficient, as defined in equation 3 (98),

in which x and y are the coordinates of a given point.

 

n
12:,(><1-->< )(1-y)
r = 3
n n 1/2 —
[:be 'X)2 (y1-- y 2]
1=1i=l

The values of r will range from +l, perfect positive correlation between
rates and 0, to 0, no correlation whatsoever, to -l, perfect negative cor-
relation between rates and 0. Whether or not a "good" correlation has been
obtained is obviously a matter of rather arbitrary definition. Jaffe (99)
defines correlations for r>O.99 as excellent, r>0.95 as good, and r>O.90
as fair. In terms of equation 1, r<0.90 is unacceptable and may indicate
serious secondary perturbation in the reaction. In addition, nonreliability
of the Hammett equation is indicated when the standard deviation,s, from the
regression line, as given by equation 4 exceeds certain values dependant
on the value of p. Jaffe's maximum allowable limits for s are 0.4 for p>4,

0.3 for o>3, 0.25 for p>2, and 0.2 for p>l.

 

 

These definitions shall be employed in the following discussion.

52

By way of definition of the problem, we will look at three previous
investigators efforts in the light of these criteria.

Imoto and his co-workers have performed several investigations into
the possibility of extension of the Hammett equation to thiophenes (100).
Some of this work,shown in Table 2,indicates that a reasonable fit to the
Hammett equation has been obtained in most cases, providing that the num-

ber of compounds, n, entering into the plot is small or carefully selected.

Table 2. Some Statistical Parameters For Imoto's Thiophene Studies

 

 

 

Reaction 0 s r n
pK of 5-R-2-thenoic acids 1.10 0.10 0.988 5
Base hydrolysis of ethyl 5-R-2-thienyl l.86 0.30 0.975 4
carboxylates
Acid catalyzed methanolysis of 5-R-2- -0.34 0.l0 0.884 5
thenoic acids
E1/2 of 2-nitro-5-R-thiophenes 0.20 0.06 0.718 l2
Base hydrolysis of ethyl 5-R-3-thienyl l.63 0.07 0.998 4
carboxylates

 

His 0 values were the classical one based on the ionization of substi-
tuted benzoic acids (lOl). As the number of entries is increased, e.g. the
plot of E1/2 of 2-nitro-5-R—thiophenes, the correlation coefficient dr0ps
to totally unacceptable levels.

The same order of correlation; r=0.972, p=0.44, s=0.053, was obtained by
Schuetz and Teller (l02) for the thermal decomposition of some bis(thenoyl)
peroxides. It must be noted that this reaction displays a rather low order
of dependence on the substituent and as a consequence would not be expected
to show large standard deviations.

Gronowitz has pointed 0ut that large deviations from linearity should

be expected for strong +M substituents in thiophene (l03).

53

He has also stated that an alternate set of 6 values should be used for
correlations in thiophene systems (42).
To this end, Janssen (l04) in 1965 reported an investigation of the hy-

drolysis of five 5-substituted-2-thienyl ethyl esters. He defined 0 as

th
in equations §_and16

* *

0th:C-UH é-

*
o — log kb/ka - log kOb/koa .6

where kb and ka are the rate constants for the basic and acidic hydrolysis
respectively and kob and koa are the rate constants for the basic and acidic
hydrolysis of ethyl acetate at 25°. His rate constants at 25° were extrapo-
lated from data taken at l00° and 50°. Since only two temperatures entered
into his Arrhenius plot, we have no means of estimating errors for the first
tenn in equation 6. A plot of log k/kO vs 0th gives r=0.608, indicating far
too much scatter of the data to have any real significance.

The concept of a "0th" is an attractive one, however. By basing values
on some intrinsic property of the thiophene ring, perturbations associated
with it, such as steric, resonance, inductive and field effects would pre-
sumably be minimized.

Barlin and Perrin(ll2) have tabulated the pKa's of a number of thenoic

acids. A fair prediction of the values of the acids is given by equation 7

pK = 4.20 - (0.72 + a) Z

a

where 4.20 is the pKa of benzoic acid, 0.72 is the difference in pKa between
benzoic acid and 2-thenoic acid and 0 has its usual meaning. The pKa's, how-
ever, all come from one laboratory (Imoto's). Furthermore nothing was report-

ed of the method of measurement of these values or their accuracy.

54

Therefore work was initiated in these laboratories to determine the
ionization constants of a series of thenoic acids. We planned for reasons
outlined in the introduction to this thesis, to limit the study mainly to
the 5-substituted-2—thenoic acids.

The pKa's of the acids were determined potentiometrically with a glass
electrode in water at 49.5° using the method described by Albert and Ser-
jeant (105). Potassium hydroxide was employed as the titrant to eliminate
the sodium ion error inherent in the glass electrode. The concentrations of
the acids were typically in the range of lO'3 molar in order to minimize
the need for activity corrections*. A complete description of the procedure
is given in the experimental section. The results are collected in Table 3,
p 55.

The difference between the pKa of a substituted acid and the pKa of 2-
thenoic acid is defined as 06 for that substituent. A plot of 06 vs Ham-
mett's 0 gives p=l.l8, s=0.12, and r=0.943 as seen in Figure l0, p 56.

Several points are immediately evident. The value for p is in agreement
with that found by Imoto, Table 3. The correlation coefficient has decreased
relative to Imoto's r, indicating greater scattering of the data and the non-
linear relationship between 0 and 09. The halogens, especially fluorine,
appear to be much more electron withdrawing in thiophene than in benzene as

reflected in the ionization constants of the corresponding acids.

 

* Corrections to true thermodynamic pKa values may be made. For a uni-uni-
valent electrolyte, the extended Debye-Huckel equation reduces to:

-log f+ = 0.5 Vi" where g the ionic strength is approxiametly constant
- l + 2 Vfi' at 0.00l. This gives a value for the mean activity
coefficient of 0.966 and raises the values given for the ionization con-
stants in Table 3 by 0.0l5 pK units (l06). This value cancels, of course

in calculating 06.

55

_o_ mucmcmwmcu

uwom owocmggum to mxu we“ so ummmnu

co.0_c.cae man

.0

.cumm mcovpmcwEmemn m we was; m yo mcowpmw>mv vcmncmgm mzp mo mmmcm>m mg“ men women; . m.m< pom

 

 

No.0.“ Neo.o mmo.o.H _mm.o N_o.o.u 0mm.m v.0. avocacpfln_o~can-m
No.0.“ Neo.o aomo.o.u mmp.o omo.o.u Nmo.e e.g. 0.0:a50mauo~=05-m
---- ---- «Fo.o.u NmF.¢ v.0. o.oc¢;p-m

No.0.“ wmm.o- omo.o.H mNN.o- 420.0.“ Nem.m 8.0a a.o:a;p-m-»xo;pae-m
No.0.“ Noo.o omo.o.H Nmp.o ¢_o.o.u ome.m 8.0a a.o:a;u-m-ozos_e-m
No.0.“ mee.o emo.o.u Nam.o m_o.o.u ON©.N e.g. a.oca:p-~-Oprc-m
No.o.u NNN.O omo.o.u ONN.O ¢_o.o.u _¢m.m emu. awoca;0-~-0topea-m
No.0.“ Ok_.o- mmo.o.H 50F.o- m_o.o.u emk.m n.0m a.oca;g-~-_»;pae-m
---- nooo.o o_o.o.u N_o.m v.0. a.ocm;u-~

Auv o oo mxa ccaoasou

 

 

muwu< uwocmcp mcwz no» mm:_m> msmwm use

mpcmumcou cowpmecoH .m mpnmh

56

 

0.9 "

0.8 F-

O.6 “

0.5 ‘-

’ -0CH
1

3

 

-CH3

1

l

l

 

-O.2

Figure 10. Plot of 0 vs

-O.1

O

0.1

06.

0.2

0.3

0.4

0.5

0.6

0.7

 

57

The same holds true for the 3,4-benzo group and the nitro function. The pKa
of 3-thenoic acid is quite comparable to benzoic acid and the difference in
pK between 3-thenoic acid and 3-benzo[b]thenoic acid gives a a value more in
agreement with Hammett's sigma for the 3,4-benzo group.

The increases noted forcfi1for the electron withdrawing groups are prob-
ably not due to a steric effect but rather to inductive effects via polari-
zation of the o and n framework of the molecule or to through space (field)
effects. For comparison, the same phenomenon is observed in the ionization
of methylthioacetic acid, pKa=3.72 and n-butyric acid, pKa=4.81 (107).

It may also be pointed out that Imoto obtained much better correlations
between Hammett's 0 values and the rate of hydrolysis of ethyl 5-R-3-thienyl
carboxylates compared to ethyl 5-R-2-thienyl carboxylates. Presumably, 3-
thienyl,in which the effects of the sulfur atom are shielded by the extra
intervening carbon atom, is much more like benzene in its reactivity.

It will be demonstrated that better correlations are obtained by the use
of 00 instead of o for a reaction involving considerable negative charge
in the transition state, namely the thenil-thenilic acid rearrangement.

In Part I of this discussion, the rearrangement of thenils to thenilic
acids was described. With the exceptions noted of 3,3'—benzo[b]theni1, 5,5'-
dimethoxy-2,2'-thenil and 5,5'-diisopropoxy-2,2'-thenil, the rearrangement
gave the thenilic acids, isolated as their methyl esters, in yields of 62-
94%. No evidence was found of any reaction that might interfere with a quan-
titative study of this reaction.

The reaction of the thenils with potassium hydroxide was followed by ti-
trimetric determination of the loss in base as a function of time. All de-
terminations of the rate constants were carried out in a 2:1 mixture by vol-

ume of dioxane and water.The temperature of the reaction mixture was held

58

to :_0.04° or better. Reaction times were measured by three different timing
devices initially triggered simultaneously. Further details are described in
the experimental section.

Westheimer had already sh0wn that the benzilic acid rearrangement is
first order in benzil and hydroxide, second order overall (33). For a re-
action of this type in which the stoichiometry is 1:1 and the initial con-
centrations of the reagents are deliberately set such that [0H-]1>[DK]1, the

rate constant is given by equation §_(108).

K H
Inf-1+4]
k_ 0H1 K 8

t[OH - DK]1

 

where t is time, i refers to initial concentrations, and DK and 0H refer to
thenil and hydroxide concentrations respectively. This equation is con-

veniently rearranged to give

H 551
1n [g—K] = kt[OH - 0K]1 - 1n H .

.1

Ito

Thus, a plot of the instantaneous 1n of the ratio of hydroxide to thenil
concentration vs time should give a straight line. A representative ex-
ample of equation g_is shown in Figure 11, p 59 for the reaction of 2,2'-
thenil at 60° with hydroxide. Although rate constants can be calculated
from the slope of this line, it was simpler to obtain them directly from
equation 8,

A good estimate of the accuracy of the rate constants can be made in the
following manner. The temperature of the various kinetic determinations was

known to :0.013%. Time could easily be measured to 1 second in 10,000 or 0.01%.

59

 

 

 

 

 

O 2 4 6 8 10 12 14 16 18

t (sec. x10'3)

Figure 11. Second Order Kinetic Plot by Equation 9_for 2,2'—thenil at 60°

A11 glassware was calibrated to 0.01 ml and all weighings were made to 0.1
mg. Total systematic errors were therefore, estimated at not greater than
0.2%. Thus, the uncertainty in the rate constants must arise principally
from random errors. These random errors are expressed as the standard de-
viation, i.e. the 68.3% confidence level of the individual rate determinations.
The usual practice was to make only one run at a given temperature and to
reject any determination that fell outside the limits of experimental error
in the Arrhenius plot. Occasionally, random runs were redetermined to insure
the reproducibility of the rate determinations. This was especially neces-
sary in the determination of the rate constant for 2,2'-theni1 at 50°, since
much of the data used in the Hammett plots would depend on this rate con-
stant. With the exception of the three compounds mentioned earlier, all of
the thenils gave consistent rate constants in accordance with equation 8
and equation 9. A complete surrmary of the rate data is presented in Table 4,

p 60.

6O

.UcoommIm—OE\Lmuw— mm mucmumcou mum.» :.m Low “.55 ms.—

 

 

 

 

.pOFQ mzwcmgcg< mcp soc; noun—oamgpxm mmocp wcm mcowpmw>mc ugmvcmpm pzogpwz czonm mucwwmcou mummm
c-0mem.o ¢-o_x¢~.emOo.e ¢-o_xFF.8mNo.m __=a;p-_m.m
m-o_xmm.o+mm.m m-o_xuo.o+we.F ¢-o_xmm.o+mk.m 0:00a¥.e _»ea;a_»ca.zo-m
P-o_xmm.~ P-0Fxmo.oHN~.F N-o_xmm.owme.m FacagpmaHONcaa-.N.N
N-oerm.o+_o.~ N-QFXNo.o+e_.. m-o_xmm.o+__.m Facagp-.m.m-fi_»ca.;p-=mv-_u-.m.m
_-o_xmm.m _-oFxmo.m _-opxoo._ _.:a;p-.~.N-OLO:FC.e-.m.m
o~._ _-o_xo~.o F-ormoo.m _.cacp-.m.m-oao_;a.u-_m.m
m-ofixuo.o+o¢.m m-o~xuo.o+mm._ m-o_xmo.o+oo._ Facagp-.m.m-2»;aae-m
m-o_xm_.o+nm._ ¢-o_xmq.pwem.~ 4-0Fx_m.owqm.m _.cacp-.N.N-F»;0aE.a-.m.m
m-o_x_p.o+mm.m m-oFx_F.o+mm.¢ m-o_xmo.o+©q.m __:a;0-.N.N

com .0“ com
¢-o_xm_.qukm._ 3-0FXNo.qu_m.P m-opxmm.u m-opx_m.m Fweaea-_m.m
3-0Fxmr.qwmo.m q-o_xa_.oumm.P m-opxp_.m m-o_xmm._ acopaxwe Feca;a_»:awep-m
N-o_xno.o+om.m m-o_xmm._+NN.a m-oFx¢m.m ¢-o_xmm.m _.:a;aflngo~can-.~.m
m-o_xm_.owkm.m m-opxmr._ e-o:me.q 3-0rmqm._ _.cagp-.N.N-AFecawep-=mv-.u-_m.m
N-o_xmm.o+eo.m N-ormOm.m N-oano.pH_w._ m-oerF.pH__.m _wca;0-.N.N-OLo=~c.u-.m.m
P-o_xmm._ N-ofixep.o+m_.m N-o_xmm.o+om.q N-o_xqo.o+mN.P _.=m;p-.~.N-OLopea.u-.m.m
e-orx_m.QHom.m e-o_xm_.qH¢N.N ¢-o_xmm._ m-o_x¢m.m Frame“-.N.N-FX;pas-m
3-0Pxom.8m®m._ m-o_xm~.8me.m m-o_xom.¢ m-o_XNN.P P.cmep-.~.N-F»;uaE.u-.m.m
m-oFx¢o.o+N~._ 4-0Fx¢_.o+mm.m ¢-o_xmm.m m-o_x~o.m Pagan“-.m.~
com .oe com om, n.a ucaoaeou

 

 

mmczpmcmaemh msowgm> pm acmsmmcmccmmm UHU< uwpwcmshupwcmgh mgp com mpcmumcou mung cmugo ucoumm .e mpnm»

61

In order to obtain accurate rate constants, a few minor corrections
were occasionally necessary. As might be expected, the strongly alka-
line solutions employed in the reaction necessitated something other
than glass for a reaction vessel. Polyethylene containers were tried
but it was noted that values for the rate constants tended to decrease
slightly with time, and correspondingly, the vessels were stained a bright
yellow. This problem was corrected by employing teflon bottles in all
rate constant determinations.

Despite this, determinations of k (run in teflon) now trended upward
slightly on long runs at high temperatures. This effect must be ascribed
to consumption of base by the solvent, since a blank run, that is a
typical kinetic run under the same conditions but omitting the thenil,
still showed a small consumption of base with time. By determining a
blank run simultaneously with each kinetic run, an appropriate cor-
rection could be made for base which was not being consumed by the re-
arrangement. This was found to be necessary only at temperatures of 70°
or 80° and for reaction times longer than 105 seconds. This correction
was never more than 3% of the total base consumed.

A small temperature dependent correction in initial concentration of
the reatants arose from the departure from ideality of the mixture of
dioxane and water. That is to say, the 150 ml of 2:1 dioxane-water em-
ployed in a given kinetic run was slightly less than 150 m1 at various
temperatures. The correction is easily made by comparing the determined
density of the mixture as a function of the temperature of the mixture
(109) with the weighted sums of the densities of the individual com-

ponents as a function of temperature (110).

62

This correction amounts to about 2% in the temperature range of 50-80°.
A table showing the correction employed at each temperature is given in
the experimental section, p 144.

It was difficult to obtain accurate and precise rate constants for
reactions with half lives <150 seconds. At the other end of the time
scale, tedium and the aforementioned reaction of the solvent with hy-
droxide made accurate work difficult for reactions with half lives greater
than 200,000 seconds. While direct measurements of the rate constants
were not made at all temperatures for a given thenil, they were readily

calculated from equation 10, the Arrhenius equation.

-Ea/RT Ea
k = Ae or 1n k = 1n A - ——— 10

m

Such calculated rate constants are indicated in Table 4 by omission of
the standard deviation.

The slope of equation lg_is equal to the activation energy, Ea times
the gas constant R. Arrhenius plots with slopes calculated by least squares
analysis are given in Figures 12-20, pp 63-65. The intercept of the Arr-
henius plot gives the ln of the pre-exponential factor A. For a reaction
in solution, the activation energy can be converted to the thermodynamical-

ly more useful enthalpy of activation AH* by equation 11,

Ea = AHt + RT 11

The Eyring equation, 12_may be used to calculate AS*, the entropy of
activation with K, the transmission coefficient being assumed equal to
unity (111a).

k = Kk.1 -AH*/RT As*/R
_—h-'e e TT’

ln k

1n k

1n k

63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.8 2.9 3.0 3.1 3.2 3.3 1000
T
Figure 12. Arrhenius plot of 2,2'-thenil
' 1 T 1 ' 1 1 1
5' r = 0.999 ‘1
F— —l
—- -H
I I l I I l I L
2.8 2.9 3.0 3.1 3.2 3.3 1000
T—
Figure 13. Arrhenius plot of 5,5'-dimethyl-2,2'-thenil
' l I 1 ' l I l
L
r
2.8 2.9 3.0 3.1 3.2 3.3 1000
'"T__

Figure 14. Arrhenius plot of 5-methy1-2,2'-thenil

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l I I l I 1 T 1 I
-- "l
'21-
-.
-3 __
x I—
.5 -4 __
I l
3.0 3.1 3.2 3.3 3.4 3.5 1000
T—
Figure 15. Arrhenius plot of 5,5'-dichloro-2,2'-theni1
I
-3 _-
-4 __
x
.5 ._
-5 __
l
3.0 3.1 3.2 3.3 3.4 3.5 1000
T
Figure 16. Arrhenius plot of 5,5'-difluoro-2,2'-theni1
] I
x —
,S
I I
2.8 2.9 3.0 3.1 3.2 3.3 1000
T

Figure 17. Arrhenius plot of 5,5'-di-(2"-thienyl)-2,2'-thenil

65

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
-3 _. __
F 1
-4 —-
x .. —-l
C.
l— -5 _ __1
I l L 1 II. J I 1 I
2.8 2.9 3.0 3.1 3.2 3.3 1000
T
Figure 18. Arrhenius plot of 2,2'-benzo[b]thenil
I I I I I I I I I
l
2.8 2.9 3.0 3.1 3.2 3.3 1000
T
Figure 19. Arrhenius plot of 2-thienylphenyl diketone
I I I I I I I I I
— -i
'7 'T r = 0.997 '—
-3 r- .-
x _ _
,S
-g - .-
l I I 1 31 .J I l I
2.8 2.9 3.0 3.1 3.2 3.3 1000
T

Figure 20. Arrhenius plot of 3,3'-theni1

66

Equation 13, readily derived from equations 19, 113 and 12_ gives 45* di-
rectly in terms of A, the pre-exponential factor, when time is expressed
in seconds and the temperature is approximetely 323°K. Within a range of
:_30° , the maximum error arising from the use of equation l§_to calculate

A5* at other temperatures is only 0.4%.

t
AS = 4.576logA-60.689 13

According to Benson (113) the errors in AS1 and AH; are easily calcu-
lated. Use of the Arrhenius equation to determine Ea leads to the frac-

tional error in Ea to be

T 2 T 2 T

 

I_A_EI2=_T_2_2_I_ . __
E T1-T2 T1 T1-T2 T2 ln(k2/k1) k1 k2

We have already seen that errors in the temperature are negligible, there-
fore the first two terms in equation 14_may be dropped, leaving the error
dependant only on the standard deviations in the rate constants. In the
error equation 14, k2 and k1 are the experimentally measured rate constants
at the extremes of the Arrhenius plot.

Wiberg has shown that the error in As* is linerally related to the
error in AH* (114). He derived equation l§_which expresses the fractional
error in AS1 as a function of a, the error in AH; and a, the error in the
rate constants. In the present studycx was taken to equal the average of

the standard deviations in all the rate constants for a given compound.

Ms‘ = 6(1/T) + R1n(l +.) 15

The thermodynamic pr0perties of the thenil-thenilic acid rearrangement

are summarized in Table 5.

67

Table 5. Thermodynamic Constants for the Thenilic Acid Rearrangement

 

 

 

Compound a AH323°K A3323°K AF323°K logA

2,2'-thenil 15.0:0.3 -25.6:1.3 23.3:0.7 7.676
5,5'-dimethyl-2,2'-theni1 l4.5:1.0 -30.8:5.5 24.4:2.8 6.522
5-methyl-2,2'-thenil 13.3:0.4 -32.5:2.4 23.8:1.2 6.162
5,5'-dichloro-2,2'-thenil 13.5:O.6 -20.3:j.6 20.0:1.1 8.820
5,5'-difluoro-2,2'-thenil l4.6:0.4 -20.5:1.5 20.6:0.9 8.773
2,2'-benzo[b]thenil 16.8:1.0 -l4.2:1.8 21.4:1.6 10.155
5,5'-di-(2"-thienyl)-2,2'-thenil 15.2:l.3 -23.6:4.0 22.8:2.6 8.104
2-thieny1phenyl diketone l5.6:0.6 -26.l:3.0 24.0:1.6 7.569
3,3'-thenil 8.40:0.8 -49.5:5.4 24.4:2.5 2.447

 

a AH: and AF3 are in units of kcal/mole 05* is in cal/°K-mole

 

The determination of p, the reaction constant, was of importance from

a mechanistic interpretation and as a test of the validity of using 06 in-
stead of o, in the Hammett plot. Only the first six compounds in Table 5
are suitable for a cap plot. The plot of log k/k0 for these six thenils at
the seven temperatures employed in the kinetic studies are shown in Figures
21-27epp 68-71.

At 50° a plot of log k/k0 vs (ygiven in Table 3 shows a rather poor cor-
relation coefficient of 0.923, Figure 24, in contrast to the“exce11ent" values
given in Table 6, p 71. Table 6 represents a complete summary of p values,

standard deviations, and correlation coefficients as a function of temperature.

Theof1 values used in these plots were those of Table 3, p 55.

log k/kO

68

 

2.0.-

 

 

 

I I I I I I. I l I I I I
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8

 

°e

Figure 21. Hammett plot for the thenilic acid rearrangement at 15°

 

 

I I I I I I I I I I I I
-0.4 -O.2 0.0 0.2 0.4 0.6 0.8

Figure 22. Hammett plot for the thenilic acid rearrangement at 30°

 

 

 

O0

69

 

    

 

 

 

 

 

 

 

l l 1 1 1* I 1 1 1 l 1 1
2.4-— ‘
2.0-— d
1.6-— _
0
ii 1.2__ o benzo _-
x
E 0.8— -
0.4'- "
0.0.. ‘
-0.4 - ‘
-0.8 _
I I 3 3 I. I I I I I I I I
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 00
Figure 23. Hamhett plot for the thenilic acid rearrangement at 40°
I T T l
2.4__ I l l l l l I l _
_ o
2.0-— C] A .-
Ago A F- o
3: 1 6 —- -4
UV
.9 1.2 _. A 0 benzo ._
0.8" plotted bye ----plotted by 06 -—
0.4I- -4
0.0-— . 2,2! _-
-O.4 - CH ‘ -—
- . .. --- CH .-
0 8 I (I 3)! I J I I I I I I I
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 09

Figure 24.Hammett plot for the thenilic acid rearrangement at 50°

log k/kO

log k/ko

7O

 

 

 

 

’0.8- -_(CH ) -1
I 3 I I I I I I I I I I
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 00

Figure 25. Hammett plot for the thenilic acid rearrangement at 60°

 

 

 

—

 

 

I P’ZIIIIIIIIII
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 09

Figure 26. Hammett plot for the thenilic acid rearrangement at 70°

 

 

1|
‘1
11!

[I

71

 

2.4-
2.0..

1.61-
0 1.2..

0.8 —

log k/k

---(CH )
I I, 312 I

 

 

I l I_

 

L I

 

-0.4 -0.2 0.0

0.2

0.4 0.6

0.8 06

Figure 27. Hammett plot for the thenilic acid rearrangement at 80°

In Figures 21-27, the thenils were represented by the symbols (CH3)2

for 5,5'-dimethy1-2,2'—theni1; CH

3

for 5-methyl-2,2'—thenil; 2,2' for

2,2'-theni1; benzo for 2,2'-benzo[b]thenil; F— for 5,5'-dif1uoro-2,2'-thenil;

and CI- for 5,5'-dichloro-2,2'-theni1.

Table 6. Rho Values, Standard Deviations, and Correlation Coefficients for
the Thenilic Acid Rearrangement in the Temperature Range of 15—80°.

 

 

 

15° 30° 40° 50° 60° 70° 80°
0 2.624 2.640 2.636 2.662 2.662 2.667 2.649
5 0.238 0.194 0.173 0.147 0.215 0.138 0.151
r 0.985 0.990 0.992 0.994 0.988 0.995 0.994

 

a calculated by the least squares method

 

72

Leffler has derived a quantity 8, called the isokinetic temperature,(111b)
defined as the slope of a plot of AH: as a function of A51 . Theory pre-
dicts that p will undergo a sign inversion, that the op plot will exhibit
maximum scatter and the effect of substituents on the reaction rate will
be a minimum at the isokinetic temperature. Therefore, it is essential
that rate studies have been made at temperatures remote from 8 before any
mechanistic interpretation of either the magnitude or the sign of p can
be assessed. The isokinetic plot of all nine thenils given in Table 5 is

shown in Figure 28.

 

      
   
    
    

 

 

 

 

 

I 1 1 I I l I f 1 1
17 —— Jr:> .-
2,2'-benzo[b]
16 _— 2-thienylphenyl I. “
11 ’//"dithieny1
15 —- - “
2,2'-thenIl
14 —— dimethy-I “/5 I diquorO —
dichloro
“5% 13 monomethyl1/j' ‘K’/ _-
12 '7
ll “
10 "
9 3,3'-thenil _
J I I I I l I I
-50 -40 -30 -20 -10 0 As"

Figure 28. Isokinetic plot for the thenilic acid rearrangement.

73

The halothenils were omitted in the calculation of the isokinetic temp-
erature. A postulate will be presented later that indicates the non-
concerted nature of the thenilic acid rearrangement, which causes a
strong deviation by the halothenils from the isokinetic line. Indeed,
deviation from the isokinetic line often indicates the occurance of an
additional perturbation on the mechanism.

The correlation coefficient, based on seven points, for the isokinetic
plot was 0.971 and the isokinetic temperature was 242:24°K. It must be
pointed out that the isokinetic temperature is arrived at in an ex post
fagtg_manner, that is after the rate constants have been determined in
some arbitrarily chosen temperature range. In the present case, it can
be seen that the kinetic determinations were fortuitously carried out in
a valid temperature range.

Many observations and a great deal of postulation can now be presented.

The entropies of activation are in agreement with those expected for
a reaction involving considerable restricted geometry in the transition
state. The entropy of activation of -14.2 eu for the rearrangement of 2,2'-
benzo[b]thenil indicates that a lesser degree of ordering may be necessary
in going from the ground state to the transition state for the reaction.
This may be due to steric interactions in the ground state which invoke a
specific conformation of the benzo[b]thieny1 rings. Coincidentally, this
geometry may be similar to that required by the transition state and hence
shows up as a more positive entropy of activation, relative to the other
thenils.

In a similar manner, the high value of AS? of -49.5 eu for the 3,3'-

thenil rearrangement could indicate that the geometry of the transition

state bears little resemblence to the ground state. It must be kept in

74

mind, however, that according to the Ingold mechanism for the rearrangement
the observed thermodynamic data would be a function both of the equili-
brium reaction between the thenil and hydroxide and the rearrangement of
that intermediate to the thenilate anion.

Other factors, such as differences is solvation between the ground
state and the transition state will also affect the entropy of activation,
but such effects should be small relative to steric effects.

The rearrangement of 2,2'-thenil is faster (I<=I.22x10‘3

l/mole-sec)
than benzil (k=l.OOXlO-4 l/mole-sec, ref 32) at 50° in 2:1 dioxane-water.
This is a rather clear manifestation of the electron withdrawing nature of
the 2-thienyl group which parallels the acidity of benzoic acid (pKa=4.229
at 49.5°) vs 2-thenoic acid (pKa=3.6l7 at 49.5°). One is tempted to pos-
tulate stabilization of the incipient negative charge at the migration
origin via d-orbital interaction, although at this point there is nothing
to substantiate this. It can be noted, though, that 3,3'-thenil in which
the sulfur is one more carbon removed from the active site of the diketone
rearranges at a rate almost the same as benzil.

Gronowitz has stated (116) that the -1 effect of the thiophene ring is
greater than benzene. If this is true, substitution of the 5- and 5'-
hydrogens in 2,2'-theni1 by 2-thienyl groups should increase the rate of
rearrangement. This was found to be the case, cf Table 4, although the
effect is not very dramatic. Using the value of 2.66,p,in the Hammett
plot yields a secondary value for 00 of +0.04 for 2-thienyl.

In the interpretation of the rate data, a valid point may be raised
that the rates of two of the diketones, 5,5'-di-(2"-thienyl)-2,2'-theni1

and 2,2'-benzo[b]thenil were not determined at the same concentrations

as the other thenils, and a correspondingly smaller amount of hydroxide

75

was used in the kinetic determinations. As a consequence of this, the rate
constants would not be comparable if the rearrangement were subject to
large salt effects. More than likely, this is a negligible factor as West-
heimer has shown (33) that the salt effect in the benzilic acid rearrange-
ment is very small. He found that an increase in the sodium ion concentra-
tion by a factor of 15 (compared with 3 for potassium ion in the present
study) caused only a 14% increase in the rate constant for the rearrange-
ment of benzil.

It is simple to calculate the rate constants for the rearrangement (k)
of a hybrid compound like 2-thienylphenyl diketone from equation 16, where
ko and k' are the rate constants for the rearrangement of benzil and 2,2'-

thenil.

1/2 16

k=(kk0) _
Thus the rate constant for the rearrangement of 2-thienylphenyl diketone

by equation 1§_is 3.49le4

to be 3.68XlO—4. In a similar manner, the calculated rate constant for the

at 50°, while the experimental value was found

rearrangement of 5-methyl-2,2'-thenil is 4.89X1O"4 based on the rate constants
for 2,2'-thenil and 5,5'-dimethyl-2,2'-thenil. The experimentally mea-

4. These calculated values are within the limits

sured value was 5.20X10'
of experimental error.This simple relationship should allow one to pre-
dict the rate constants for any substituted thenils provided the values of
the other two k's related to it in this manner are known.

The fairly large positive and constant value for ;)at all temperatures
indicates that the reaction is facilitated by electron withdrawing sub-

stituents in accord with both the Ott-Clark (35, 36) and Ingold (117) mech-

anisms. The developing negative character at the migration origin as well

76

as the facilitating unsaturation of the carbonyl migration terminus should
both be favored by electron withdrawing groups. For this reason, the 09
values used in the Hammett plot were taken as the sum of the two substi-
tuents.

Shown below are the two mechanisms proposed for the benzilic acid

 

 

 

rearrangement.
THE INGOLD MECHANISM
k T 5 ‘
9 u 1 9;>R:> ,9 P 1 kg 9 9
AR-C—C-AR' ‘1“ A -c— -AR' -» c’ /; 3 —> HO-C-C-AR
k H0 / \‘ —’,C — AR I '
'1 OH ‘\ AR AR
HOT - -
THE OTT-CLARK MECHANISM
0 0 Q "' " H T - - 0 0 0H
II II %' /’I _ II I
AR-C - C-AR' —>- C C'__ AR. ——> O-C -C-AR
0 ’/ AR'
HO‘ ‘AR’
L .I

 

 

The essential difference between the two is the existence of an initial
rapid equilibrium of the hydroxide and the diketone in the Ingold mechan-
ism. If the Ingold mechanism is the correct one, then the observed rate

constant will be given by equation 17, where K is the equilibrium constant.

k

- I 3
kobs ‘ (1"2 “‘2 17—

For substituents of the same type, alkyl, aryl for example, K should be
nearly constant. For strongly electron withdrawing groups, e.g. halo or
nitro, the equilibrium constant may be quite different. This naturally
will have an effect on kobs which cannot be accounted for purely on the
basis of the substituent effect, unless one knows the value of the equili-

brium constant.

77

In the thenilic acid rearrangement, the two methyl compounds and the
benzo[b] compound did correlate well with 2°0’ at all temperatures. It
must be noted here that 00 gives a much better fit thano for the benzo[b]
function. Thehalo o's even using the larger values of 06 do not cor-

relate well unless o is multiplied by three.

0
Several points now need to be borne in mind. If the Ott-Clark mechanism
is valid, one would expect to observe a deuterium isotope effect on the
experimental rate constant since their mechanism pr0poses that proton
migration occurs as a concerted process with the rearrangement. Hine (32)
failed to find an isotope effect, but instead found rate enhancement when
the rearrangement was conducted using NaOD in 020. Since we do not know
a priori, the enhancement of rate due to the greater basicity of NaOD in
020 relative to NaOH in H20, it is possible that an isotope effect may exist
which does not completely compensate for the isotopic rate enhancement.
Therefore, Hine's study does not rule out the Ott-Clark mechanism.
In addition, Ott and Clark contend that the observed rapid uptake of

18 from 018—enriched water (30) occurs through an equilibration process

0
that is only incidental to the rearrangement.

Ott and Clark have measured the relative migratory aptitude of one aryl
group vs the other in various unsymmetrically substituted carbonyl C14
monolabeled benzils. Since the ratios of the migrating aryls correlated
with a they argued that the reaction had to be concerted. However, it is
difficult to explain why the equilibration step should affect the subse-
quent relative migration of one group with respect to the other.

In the present study, it was found that the halothenils did not cor-

relate with 00 or 0. Indeed, the 0 values for both chloro and fluoro

0

had to be multiplied by 3 to fit the o plot. This coefficient of 3 may

60

78

be purely fortuitous. If we consider a more generalized Hammett equation

_I_§_

10911:- = on: 01. l_g
o

it is seen that we have assumed that n=1, i.e. the proper value of the
substituent constant in the 00° plot is obtained from the simple sum of
the two sigmas. Considering Ingold's mechanism, we must include the equi-
librium step in the value for kObs , but we do not have any estimate of
the proportion of contribution by each step in the mechanism to the mag-
nitude of o.

In other words, n may have any value other than unity making it imposs-
ible to determine p unless the equilibrium constant K is known. However,
the existence of the equilibrium step is verified by the nonconsistent
values for n, or for what is actually observed, deviations from the linear
relationship in the Hammett plot. Phrasing it still another way, the large
rate enhancement due to the halo substituents, above that predicted on the
basis of their cnavalues, strongly suggests that the equilibrium step is
part of the nonconcerted rearrangement in accord with the Ingold mechanism.

It was mentioned earlier that the reaction of hydroxide with 5,5‘-
dimethoxy-2,2'-thenil does not yield identifiable products. It was, how-
ever, somewhat surprising that the disappearance of hydroxide with time
fits the rate law shown in equation 12, that is second order in hydroxide

only.

J—ld 0“ = k[0H]2 19

dt ——-

A plot of ln[OH/DK] vs time gave a sharply rising plot, while the plot of
of [l/OH] vs time was linear to over two half lives of the hydroxide as

may be seen in Figure 29.

79

 

   
 

 

 

”1% . . r . . 3 WI"
6.5" ‘130
60*- - “120
5.5 ' I “110
5.0-— —100
4°5__ ------- by equation 20_-_90
4.0— ~80
3.5-— —70
3.0— ...60
2'5 ----------- by equation 9_ '150
2.0 740
1.5 ._30
1.0 -20
0.5 -10.0

I I I J I I I I

 

 

1 2 3 4 5 6 7 8 9 10 11

time (sec XlO'4)

Figure 29. Second order kinetic plots for 5,5'-dimethoxy-
2,2'-thenil by equations 9_and 20_at 80°.

Equation 19_is easily integrated to give equation 29,

k = 1/t(1/0Ht - 1/0H1) _20

This rate expression holds reasonably well in the temperature range 40-80°
as seen by the linearity of the Arrhenius plot in Figure 30.

It is easily hypothecated that the reaction depends in some manner on
the attack of hydroxide at the alkoxy group. The more sterically hindered

5,5'-diisopropoxy-2,2'-theni1 gave what appears to be mixed order kinetics

for the reaction with hydroxide although a roughly linear

80

was obtained by equation 29, Figure 31.

correlation

 

 

 

 

_.I
3

 

 

 

 

 

 

3.3

%% I I T I T l l l I I T 1 [0H]-]
2.7 - '1 46
2.6 r - 45
2.5 - by equation 9_ ------- - 44
2.4 - 43
2.3 - 42
2.2 - 41
2.1 - 40
2.0 ~ 39
1.9 - 38
1.8 - 37
1.7 - 36
1.6 - 35
1.5 ~ 34
1.4 4 a 33

I 1 l J I I I L i J I L I

l 2 3 4 5 6 7 8 9 10 ll 12 13

time (sec X10'4)

Figure 31. Second order kinetic plots f0r 5,5'-diisoprop0xy-

2,2'-thenil by equations 9_and 2Q_at 50°.

81

Approxiametly 5 times as great a reaction period was required to reduce
the concentration of the isopropoxy compound by one half compared to the
methoxythenil. This slow rate of disappearance strengthens the argument
that the alkoxy group is involved directly with the reaction site. Two
reasonable theories, one involving the intermediacy of a Meisenheimer
type complex to give a labile hydroxythiophene and the other, an 5N2 at-
tack on the alkoxy carbon were discussed on p 44.

On p 43, the anomalous reaction of 3,3'-benzo[b]thenil with hydroxide
was mentioned. The thenil was cleaved and at least one of the products was
identified as 3-benzo[b]thenoic acid. A nearly unvarying apparent rate

constant was given by equation 21,

k = ln[OK/OHJ, + ln[OH/DK] 21
t-ln[OH/DK][OH - DK]1

 

A plot of ln[OH/DK] as a function of time gives the steadily rising curve
shown in Figure 32, p 82, while a plot of ln[OH/DK] vs k from equation 9
shows a linear relationship. This behavior could indicate a second consec-
utive reaction which according to the hypothetical scheme on p 44 would be
the Cannizzaro reaction of the intermediate 3-thianapthaldehyde.

The 2-thienyl group appears to be a -I+M substituent in which either
the inductive effect or the resonance effect can predominate depending on
the reaction requirements. We have seen that the -I effect is dominant in
the thenilic acid rearrangement. However, a serendipitous example of the
+M effect was uncovered during the course of this investigation.

In a report of the synthesis of diarylglycolic acids by Blicke (118)
he noted that they displayed characteristic colors of red to blue when

dissolved in concentrated sulfuric acid. During the present synthetic

1n [OH/OK]

82

 

 

by equation 2l;---

  
  

---by equation 9_

 

   

I I I I I I I I l I I
0 2 4 6 8 'H) 12 14 16 18 20 22

0.4 0.8 1.2 1.6 2.0 2.4

Figure 32. Kinetic plots for 3,3'-b€nZ°[b]the"11 by
equations 9_and 21,

 

I
24 time(sec X10'3)
2.8 k(X100)

ln [OH/OK]

82

 

 

by equation 21;---

   
  

---by equation 9_

 

 

 

I I I I I I I I
0 2 4 6 8 10 12 14 16 18 20 22“724 t1me(sec x10'3)

0.4 0.8 1.2 1.6 2.0 2.4 2.8 k(X100)

Figure 32. Kinetic plots for 3,3'-benzo[b1th9011 by
equations 9_and 21:

83

work, overacidification of the alkaline reaction mixture from the rearrange-
ment of 2,2'-thenil gave a transient pink coloration. Sensing that this might
be due to a "stable" carbonium ion (119) solutions of 2,2'-thenilic acid
and its methyl ester in strong acids were examined by visible and nmr
spectroscopy.

The visible spectrum, Figure 33, p 84 was determined in 50% (w/w)
chlorosulfonic acid-methylene chloride, and shows Amax at 518 nm (log a =

4.52) indicative of an aryl stabilized carbonium ion.

QIQ QIQ

\

O OH . 0//C OCH3

(LXII) (LXIII)

The decomposition of (LXII) could be monitored by observing the absor-
bance at 518 nm. Increasing the concentration of chlorosulfonic acid above
50% or decreasing it below 1% hastened the decomposition. At 10% ClSO3H,
the absorbance at 518 nm fell to half its initial value in 21 hours at 25:2°,

'5 molar. Stable solutions of (LXII),

with a starting concentration of 2X10
i.e. no change in Amax or initial log 6, could be formed in sulfuric acid-
water mixtures down to 80% (w/w) sulfuric acid. The rate of decompsition
increased as the percent of water was increased.

If the ruby red solutions of (LXII) in ClSO3H-CH2C12 were quenched at
-50° with anhydrous methanol or ethanol, and treated with diazomethane after

workup, the methoxy (LXIV) or ethoxy (LXV) dithienyl acetic esters could

be isolated in better than 95% yield. The nmr spectra of these compounds

clearly indicate that no quenching of an acylcarbonium ion occurred.

84

0mm um kum mcospm :- AHHXJV :0- we Esspumam m—nwmw> .mm mesmed

 

E: coo 0mm oom oma ooa 0mm com omN oom
- . - _ _ _ a _ _ _ _ _ .
/
I \\I// 180.2
I \II/
I x I
/ \
, \ .
II. , \INFUNIUIImomFu c? uwoe owgzepzm :wIIIIIII .Ilgooo om
, \
/ \
.II \ II. 3
\
1 I898
\
//\
_ _ _ _ _ - T _ L _ _ 08.9-

 

 

 

0R
U1 fl (LXIV) R=CH3
(LXV) R=C H
,C 2 5

/ \

Olah and Pittman (120) have reported on the effect of an adjacent car-
bonium ion on the chemical shift of thienyl protons. The nmr spectrum
appeared to be considerably complicated by coupling to adjacent protons.
In contrast, the nmr spectrum of (LXII) as an 8% solution in 25% (w/w)
ClSOBH-CHZCl2 at -55° with tetramethylammonium tetrafluoroborate (t=6.90,
ref 121) as a second internal standard gave the simple spectrum shown in
Figure 34, p 86. This spectrum together with large downfield shift of the
thienyl protons may be taken as evidence for considerable and symmetrical

participation by the thiophene rings in stabilizing the carbonium ion as

in (LXVI).

 

(LXVI)

Although ion (LXII) was rapidly decomposed at room temperature, the
nmr spectrum of ion (LXIII) could be determined at +30°. It was unchanged
from the spectrum at -55° and strikingly similar to the spectrum of ion
(LXII) under the same conditions. The spectrum of ion (LXIII) is shown in
Figure 35, p 87. The downfield chemical shift of the methoxy protons of
(LXIII) compared with methyl 2,2'-thenilate (T=6.20) is probably due to
the inductive effect of the adjacent carbonium ion. Of course, there is
the possibility that some of this difference may be due to solvent effects.
Use of external tetramethylsilane as a standard, in general gave chemical

shifts 0.2 ppm lower field than internal tetramethylannmnfitmItetrafluoroborate.

.omm- pa N-UNIQ-Imom_U e- A-Hx-v so. to Est-saga 252 .4m mesa-a

0N m m z m m H

_ _ _ _ _ _ _

(D

971 1 pa -11 43

. :fi
8
maeqz- I//\f\\\IIIIIIIIIIIIIJ//\/\\\I
JHHH N: m. a I m. a NI m. aL
NI m.a

 

 

 

 

 

 

 

 

 

m.
S :u

87

 

mIH<ZH

o

Fm.m

 

 

Io

.oom+ “mm—umxouzmompu cw AHHHxJV cow mo Ezgpumam LE2 .mm mgzmwm

 

 

N_orIu

: m

N: m.¢
NI m.¢

 

 

 

 

 

N
_

Pm.m

 

 

 

NI m.v

Nn.~

 

 

H o
_ XII
mm.o

 

 

 

 

 

 

N: ¢.F
N: m.¢

 

88

While small rapid reversible protonation of the ester function cannot be
ruled out, it may be noted that no quenching of any such protonated form by
ethanol occurred, i.e. no ethyl esters could be isolated from either (LXII)
or (LXIII) when quenched with ethanol.

There is the possibility that the carbonium ion may be stabilized by some
participation of the type shown below. In either the monomeric or the dimeric
form, the methyl group should be more effective in stabilizing the positive

charge than hydrogen and this may explain the observed greater thermodynamic

stability of (LXIII) vs (LXII). 2 3 92
9, x
\f" //0 R \0 /C\+’492
R=H or CH 2”C C "
3 9 \,/ 2,.- \ /6\
‘0 9 A C R
I 2 u
R 9 0

Indeed, ion (LXIII) is so stable that it is readily formed at room temper-
ature simply by adding the solid ester to a mixture of chlorosulfonic acid
and methylene chloride. An attempt was made to perform the same experiment
with methyl benzilate. The initially orange-red solution decomposed before
its nmr spectrum could be examined, that is within 3 or 4 minutes. This is
in strong contrast to the solution of (LXIII) which gave a virtually un-
changed nmr spectrum after 20 hours at room temperature in a sealed nmr tube.

This greater stability of the thiophene carbonium ion is a qualitative
demonstration of the +M effect of the 2-thienyl group.

It would seem that the +M effect will dominate in those reactions in-
volving positive character at the reaction site and likewise, the -1 effect
will dominate for reactions that involve negative character in the transition
state. For this reason, the value of 0 will be reaction dependent and no

6
one value for it can be properly assigned that will hold for all reactions.

SUMMARY

Thus we have come in a quasi-anfractuous manner through a rather mixed
aggregate of thi0phene chemistry. Many supposedly unique observations owed
more to the author's naivete than to originality. However, listed below are
the major and some minor conclusions unearthed for the first time.

The cyanide catalyzed condensation of various thiophene aldehydes, the
thenoin condensation, was studied. It was found to be as unpredictable as
the normal benzoin condensation, in regard to sucess or failure to yield
the desired products.

The synthesis of ten new thenils, dithienyl diketones, was accomplished
by various methods including the oxidation of the corresponding thenoins
and the reaction of thienyl lithiums with dimethyl oxalate.

The thenils were subjected to base induced rearrangement to give, after
treatment with diazomethane, the methyl esters of many heretofore unreported
thenilic acids in good to excellent yields.

Four of these esters were transesterified with N-methyl-3-piperidinol to
give dithienyl isosteres of dB 336. They may have significantly useful pharm-
acologic properties.

A new synthesis of benzoins, which is applicable to both symmetrical and
unsymmetrical benzoins was formulated. It consists of the reaction of aryl
glyoxals with aryl grignard reagents.

The mass spectrum of benzoins was found to be characterized by the absence

of a parent peak as well as the presence of a P-2 peak, indicating that the

89

90

initial fragmentation process is the loss of hydrogen to give a benzil.

The abnormal cleavage 3,3'-benzo[b]thenil by hydroxide was noted, as
well as the anomalous reaction of two alkoxythenils with potassium hydroxide.

The kinetics of the thenil-thenilic acid rearrangement were studied for
the parent compound and for many substituted thenils. The parent 2,2'-thenil
rearranges faster than benzil in accord with the prediction based on the -1
character of the Z-thienyl group relative to the phenyl group.

The ionization constants for a series of nine thenoic acids were deter-
mined potentiometrically with the glass electrode in water at 49.5°. From

these a new series of substituent constants, C , was calculated.

6

A plot of the log of the relative rate constants for rearrangement for
some susbtituted thenils vs 2,2'-thenil was found to give better correla-
tion with 06 than 0.

A demonstration of the existence of the equilibrium step in the Ingold
mechanism of the benzilic acid rearrangement was infered from the abnormal
rate enhancement due to halo substituents.

Long range ring fluorine to side chain couplings in 5-fluoro-2-thenalde-
hyde and 5-fluoro-Z-acetylthiophene were recorded. Several other fluorothienyl
compounds were synthesized for the first time.

An a-carboxyl carbonium ion was generated and examined. Solutions of it
were examined by nmr and visible spectroscopy in order to confirm the struc-

ture. The stability of the ion emphasizes the +M character of the 2-thienyl

group.

EXPERIMENTAL

Melting points were taken on an Electrothermal Melting Point Apparatus
calibrated with furnished standards. Infrared spectra were recorded on a
Perkin-Elmer 237B grating spectrophotometer as KBr pellets for solids or
as films between NaCl plates for liquids. All spectra were calibrated at
1601.4 cm"1 with polystyrene film. Nuclear magnetic resonance spectra were
recorded on a Varian A60, a Varian A56/60, or a Jeolco C60H instrument at
ambient probe temperature using tetramethylsilane as internal standard
for proton spectra, and Freon 112 as internal standard for F19 spectra.
Ultraviolet and visible spectra were determined on a Unicam SP8OO or a
Cary 14 spectrophotometer in 95% ethanol. Mass spectra were determined by
Mrs. Lorraine Guile of this department on a Hitachi-Perkin-Elmer RMU-6
mass spectrometer at 70 eV ionization potential. Vapor phase chromatograms
were run on a Varian A90-P3 gas chromatograph equipped with a 5 ft X l/4 in
20% SE-30 on Chromosorb w column. Refractive indices were detennined with
a Bausch and Lomb Abbe-3L refractometer at the temperature indicated. Ele-
mental analyses were performed by Galbraith Laboratories, Knoxville, Tenn.

Thiophene, technical grade, was purchased from Pennsalt Chemicals Corp.,
Philadelphia, Pa. All thienyl aldehydes were stored at -l0° in a refriger-
ator until needed. Unless otherwise noted, anhydrous sodium sulfate was

used to dry extracts in all product isolation procedures.

91

92
Preparation of 2-thena1dehyde

 

The procedure of Campaigne and Archer (122) was used in this synthesis.
A 500 ml three-neck flask fitted with a reflux condenser and a thermometer
to extend below the level of the flask contents, was charged with 126 g
(1.50 moles) of freshly distilled thiophene and 138 g (1.89 moles) of re-
cently distilled dimethylformamide. The mixture was cooled in an ice bath
and stirred during the rapid but cautious addition of 288 g (1.87 moles) of
phosphorus oxychloride. The ice bath was removed and the reaction mixture
temperature was allowed to rise to 90-100° with occasional cooling to hold
its temperature in this range. After spontaneous heating had ceased, the
reaction mixture was heated on a steam bath for one hour and poured into 1
kg of crushed ice. The solution was neutralized to pH 6 (Hydrion B) with
5 M sodium hydroxide. The oily aqueous suspension was extracted continuously
with ether for 16 hours. The extract was dried,and the ether evaporated on a

steam bath. The residue was distilled under vacuum to obtain 144 g (1.28 moles,

24

86.4%) of 2-thenaldehyde: bp 49-50°(1.4 torr); nD
(123) bp 44-45° (1.1 torr); nD20 1.5920.

1.5870; reported values

Preparation of 2,2'-thenoin

The reported procedure (8) was modified in this synthesis. A solution of
28.0 g (0.250 mole) of 2-thenaldehyde in 250 ml of 95% ethanol was heated to
reflux and 19.5 g (0.300 mole) of potassium cyanide in 100 ml of water was
added in one portion through the condenser. The mixture was refluxed for
15 minutes. Longer or shorter reflux time resulted in the formation of a very
tarry product. The dark green reaction mixture was cooled to 20° and acidified
with 2N hydrochloric acid to pH 4. A total of 250 ml of water was continuously

added during this process with stirring. The pale orange precipitate was

93

collected and air dried to yield 11.2 g (0.050 mole, 40.0%) of product. This
material was used without further purification for the preparation of 2,2'-

thenil.

Preparation of 2,2'-thenil

 

The literature procedure (124) was somewhat modified in this synthesis. The
previously prepared crude 2,2'-thenoin, 7.36 9 (0.0331 mole) was dissolved
in 200 ml of boiling methanol (purified by Fieser's method, ref 125). To this,
a solution of sodium methoxide prepared by dissolving 1.61 g (0.070 mole) of
sodium metal in 60 m1 of purified methanol was added. Heating of the dark
green solution was discontinued and 10.2 g (0.040 mole) of solid iodine was
added as rapidly as possible. Immediately upon completion of the iodine ad-
dition, the entire reaction mixture was rapidly quenched in a mixture of 300 g
of ice and 10 ml of acetic acid. The resulting yellow-brown product was col-
lected to obtain 6.09 9 (0.0274 mole) of crude thenil. The crude material
was extracted continuously with 30-50° petroleum ether in a Soxhlet appar-
atus for 24 hours. The petroleum ether was removed with a stream of air and
the residue was recrystallized from methanol to obtain 4.20 9 (0.0189 mole,
57.1%); mp 82-84°; reported (124) 81-82°. The infrared spectrum is shown on

p 147. The ultraviolet Spectrum is shown on p 154.

Preparation of 2,2'-thenilic acid

 

A suspension of 2.00 9 (0.00900 mole) of 2,2'-thenil in an alkaline sol-
ution prepared from 6.00 g (0.107 mole) of potassium hydroxide in 40 ml of
water was heated with stirring at 80° in an oil bath for 1.5 hours. After
cooling in an ice bath to 20°, the reaction mixture was rapidly acidified
to congo red with concentrated hydrochloric acid and extracted with ether

(3X25 ml).

94

The combined ether extracts were dried, and the ether was removed under re-
duced pressure to yield 1.86 9 (0.00726 mole, 86.0%) of 2,2'-thenilic acid:
mp 98-100° dec; reported (8) 80° dec; neutralization equivalent: calc'd for
C10H80352’ 240.2; found, 241. This material decomposes in a few hours at room

temperature, but is stable for at least 2 years at -20°.

Preparation of diazomethane

 

Diazomethane was prepared from N-methy1,N-nitrosourea (126) as described

(127) in the literature. Typical analyses gave 61-65% yield of diazomethane.

Preparation of methyl 2,2'-thenilate

 

This compound is most conveniently prepared directly from the combined
dried ethereal extracts of the 2,2'-theni1ic acid preparation described pre-
viously. Alternatively, it may be prepared from the isolated acid which is
dissolved in ether and then treated with diazomethane. Thus, the ethereal
thenilic acid solution from the rearrangement of 2.22 g (0.0100 mole) of
2,2'-theni1 was stirred and treated with 50 ml of distilled ethereal diazo-
methane. The diazomethane was prepared from 2.38 9 (0.0231 mole, 56% excess)
of N-methyl,N-nitrosourea.

The stirring was continued for 15 minutes and the ether was then removed
under reduced pressure in a bell jar with stirring but with no external heat.
The dried ester was recrystallized from 60-90° petroleum ether to obtain 1.70 9
(0.00693 mole, 69.3%) of pure product. For analysis, the ester was recrystal—
lized three times from 60-90° petroleum ether: mp 93°.

Analysis, Calc'd for C H 0 ' C, 51.95; H, 3.96; S, 25.22.

ll 10 3 2'
Found: C, 51.86; H, 3.80; S, 25.02.

95

Preparation of 2-methy1thiophene

 

This material was prepared essentially by the method of King and Nord
(45). A three liter, three-neck flask equipped with an overhead stirrer, a
thermometer extending below the level of the reaction mixture, and a reflux
condenser with a distillation takeoff, was charged with 140 g (1.25 mole) of
2-thenaldehyde, 250 ml (4.42 moles) of 85% hydrazine hydrate and 1000 ml of
ethylene glycol. The solution was stirred and heated until 280 ml of water
and hydrazine hydrate had been removed by distillation. The distillate was
extracted with ether (2X50 ml) and the ether extract was added to the cooled
reaction mixture which was then heated to remove the ether by distillation.
When the thermometer indicated 60°, 250 g of potassium hydroxide was added to
the reaction mixture in one portion causing a rapid temperature rise to 90°,
along with a strong evolution of gas. After the gas evolution had subsided,
the mixture was refluxed 15 minutes, and the 2-methylthiophene was removed
azeotropically. It was separated from the water, dried, and heated to 70° to
remove a small amount of low boiling material. The remaining 2-methylthiophene
was colorless and sufficiently pure to be used directly in the preparation of

5-methy1-2-thenaldehyde. The yield was 95.2 g (0.971 mole, 77.7%): nD24 1.5151;

reported (45) "020 1.5203.

Preparation of 5-methyl—2-thena1dehyde

 

A 300 ml three-neck flask equipped with a condenser and magnetic stirrer
was charged with 23.2 g (0.236 mole) of 2-methylthiophene and 21.9 g (0.300
mole) of dimethylformamide. To this mixture, 44.4 g (0.290 mole) of phos-
phorus oxychloride was added in one portion with stirring. After the highly
exothermic reaction had susbsided, the reaction mixture was heated on a steam

bath for two hours, cooled and poured onto 100 g of crushed ice. An ice bath

96

was used to maintain the solution temperature below 40° while the pH was ad-
justed to 8 by the addition of 5M sodium hydroxide. The oily aqueous suspen-
sion was extracted with ether (3X70 ml). The combined, dried extracts were

flash distilled, and the residue was vacuum distilled to obtain 25.6 g (0.203
mole, 86.0%) of the desired aldehyde: bp 62-64° (1.3 torr), nD24 1.5791; re-

20

ported (128) bp 81-82° (6 torr) nD 1.5742.

Preparation of 5,5'-dimethy1-2,2'-then11

 

A mixture of 2.5 9 (0.0510 mole) of sodium cyanide, 12.6 g (0.100 mole) of
5-methyl-2-thena1dehyde, and 50 ml of dimethyl sulfoxide in a 100 ml round
bottom flask under a nitrogen atmosphere was sealed and shaken occasionally
during two days. The reaction mixture gradually changed from green to brown.

At the end of this period, the reaction mixture was poured into 500 ml of
water and the aqueous suSpension was extracted with ether (3X100 ml). The
ether extracts were combined and the ether was removed by evaporation. The
oily residue was treated with 20 g of cupric sulfate pentahydrate in 200 m1
of pyridine. The deep blue solution was then heated on a steam bath for 2 hours
and poured into 1.5 l. of cold water. After being set aside overnight in the
refrigerator, the solution was filtered and the collected precipitate air
dried. This solid material was extracted continuously in a Soxhlet apparatus
with 30-50° petroleum ether for 30 hours. Upon evaporation of the petroleum
ether, the bright yellow-orange thenil was obtained. It weighed 3.35 9 (0.0148
mole, 29.6%). For analysis it was sublimed twice at 70° (0.05 torr): mp 85-86°.
The infrared spectrum is shown on p150. The ultraviolet spectrum is shown on
p 156.
Analysis, Calc'd for C12H100252: C, 57.57; H, 4.03; S, 25.62.

Found: C, 57.44; H, 4.01; S, 25.60.

97

Preparation of methyl 5,5'-dimethyl-2,2'-thenilate

 

A 0.50 g (2.0 mnoles) quantity of 5,5'-dimethyl-2,2'-thenil was added to
1.0 g (18 mmoles) of potassium hydroxide in a mixture of 15 ml of water and
3 m1 of dioxane previously heated to 80° in an oil bath. After solution was
effected, in about 4 hours, the dark violet solution was cooled in an ice
bath to 10°, and acidified to pH 3 with concentrated hydrochloric acid. The
precipitated thenilic acid was immediately extracted with 25 ml of ether, and
the aqueous soltuion was extracted with another 25 m1 portion of ether. The
combined ether extracts were shaken with a mixture of anhydrous sodium sulfate
and Norit-A and filtered. The stirred filtrate was treated with an ethereal
solution of diazomethane, prepared from 0.5 g of N-methyl,N-nitrosourea. The
ether was removed at reduced pressure at room temperature. The yield of ester
was 0.46 g (1.6 mmoles, 80%). The crude product was crystallized twice from
60-90° petroleum ether for analysis: mp 57-58°.

Analysis, Calc'd for C H 0 S ° C, 55.29; H, 5.00; S, 22.71.

13 14 3 2'
Found: C, 55.60; H, 4.95; S, 21.47.

Preparation of 5-chloro-2-thena1dehyde

 

A mixture of 119 g (1.00 mole) of 2-chlorothiophene (Columbia Chemical 60.,
Columbia, 5.6.) and 80.4 g (1.10 moles) of dimethylformamide was stirred while
176 g (1.15 moles) of phosphorus oxychloride was added dropwise to the mixture.
Heat was applied with a steam bath until the strongly exothermic reaction was
initiated. The reaction temperature was held below 130° by occasional cooling
in an ice bath. After the reaction had subsided, heating was continued for
an additional 2 hours. The reaction mixture was cooled in an ice bath and
hydrolyzed by the cautious addition of 500 g of crushed ice. The pH was ad-

justed to 8 with solid sodium hydroxide, keeping the temperature below 50°.

98
The oily product was extracted with ether (4X300 ml). The combined ether
extracts were dried and the ether removed by flash distillation. The residue
was vacuum distilled to yield 91.9 g (0.627 mole, 62.7%) of the desired
aldehyde: bp 50-52° (0.5 torr); reported (122) bp 52-53° (0.5 torr).

Preparation of 5,5'-dichloro-2,2'-thenil

 

A suspension of 7.05 g (0.144 mole) of sodium cyanide in 300 m1 of anhy-
drous tetrahydrofuran containing 2% dimethyl sulfoxide was prepared under a
nitrogen atmOSphere. To this, 41.4 g (0.282 mole) of 5-chloro-2-thena1dehyde
was added in one portion. The flask was sealed and the contents magnetically
stirred for 30 hours. The flask was opened momentarily to add 7 ml of glacial
acetic acid. The flask was reclosed and the contents stirred an additional
30 minutes. Following this neutralization, 300 m1 of anhydrous pyridine, and
70.5 g (0.282 mole) of cupric sulfate pentahydrate was added and the reaction
mixture was stirred vigorously for 3 hours. At the end of this period, the
entire contents of the flask were poured into 2 1. of cold water and refrig-
erated overnight. The ice cold suspension was filtered and the dark brown fil-
ter cake was washed well with water and air dried. The solid was continuously
extracted with 30-50° petroleum ether in a Soxhlet apparatus until the bright
orange color was exhausted. The petroleum ether was removed with a stream of
air and the residue was crystallized from 300 m1 of methanol to obtain 14.8 9
(0.0508 mole, 41.2%) of the thenil. For analysis the thenil was crystallized
three times from methanol: mp 121-122°. The infrared spectrum is shown on p
149. The ultraviolet spectrum is shown on p 156.

Analysis: Calc'd for C H C1 0 S ° C, 41.25; H, 1.38; 5, 22.02.

10 4 2 2 2'
Found: C, 41.36; H, 1.49; S, 22.00.

99

Preparation of methyl 5,5'-dichloro-2,2'-theni1ate

 

A solution of 0.80 g (0.014 mole) of potassium hydroxide in 15 m1 of water
was heated to 90° in an oil bath. To this alkaline solution, 1.00 9 (0.00345
mole) of 5,5'-dichloro-2,2'-thenil was added and the suspension was stirred
under nitrogen until solution was effected in about 45 minutes. The reaction
mixture was cooled to 15° in an ice bath and acidified to pH 3 with concen-
trated hydrochloric acid. The acid solution was quickly extracted with ether
(2X50 ml). The combined, dried ether extracts were stirred with an ethereal
solution of diazomethane, prepared from 1.5 g of N-methyl,N-nitrosourea. Remo-
val of the ether under reduced pressure gave an oil, weighing 1.02 9 (0.00316
mole, 91.6%). For analysis, the oil was subjected to distillation by micro
molecular still at 130° (0.01 torr).

Analysis, Calc'd for C H Cl 0 S ' C, 40.87; H, 2.49; S, 19.84.

11 8 2 3 2'
Found: C, 41.05; H, 2.63; S, 19.47.

Preparation of 2,2-bithienyl

 

This compound was prepared using a minor modification of the procedure
described by Gronowitz (48). A solution of 30.0 g (0.360 mole) of thiophene
in 100 ml of dry ether was added dropwise with overhead stirring to 250 ml of
1.6N n-butyllithium in hexane ( Foote Chemical Co., Exton, Penn.), at such
a rate that the temperature could be maintained below 10° by an ice bath.
After stirring an additional 2 hours in the ice bath, the thienyllithium
solution was refluxed for 15 minutes and then cooled to 5°. By means of a
Gooch tube attached to one neck of the reaction flask, 50.0 g (0.37 mole) of
anhydrous cupric chloride was added cautiously in one portion. The reaction
mixture became quite exothermic deSpite immersion in the ice bath. Following

the addition of the cupric chloride, the SUSpension was refluxed 3 hours,

100

cooled, and 300 m1 of 2N hydrochloric acid was cautiously added. The organic
layer was separated and the aqueous layer was extracted with ether (2X100 ml).
The ether solutions were combined and the ether was removed on a steam bath.
The residue was steam distilled. After one liter of distillate was collected,
it was extracted with ether (3X150 ml) and the combined ether extracts were

dried and evaporated to obtain 20.0 g (0.12 mole, 67%) of 2,2'-bithienyl.
Preparation of 5-(2'-thienyl)-2-thenaldehyde

A vigorously stirred mixture of 16.9 g (0.110 mole) of phosphorus oxy-
chloride and 9.04 g (0.110 mole) of dimethylformamide was cooled in an ice
bath and 16.6 g (0.100 mole) of 2,2'-bithienyl was added in one portion. After
15 minutes of stirring, the reaction temperature rose rapidly to 80-90° and
the mixture solidified. Following heating on a steam bath an additional 30
minutes, the semisolid was cooled and added to 60 g of sodium acetate in
100 ml of water. This suspension was stirred for one hour and basified to pH
10 with solid sodium hydroxide. The basic solution was extracted with ether
(4X100 ml). The ether extracts were combined and washed successively with
200 m1 of 2N hydrochloric acid, 200 m1 of saturated sodium bicarbonate, and
200 ml of water. Evaporation of the solvent gave 16.1 9 (0.0830 mole, 83.0%)
of the aldehyde. It was recrystallized from methanol-water: mp 55-56°; re-

ported (47) mp 56°.

Preparation of 5,5'-di-(2"-thienyl)-2,2'-thenoin

 

A solution of 0.59 g (0.010 mole) of potassium cyanide in 4 m1 of water
was added in one portion to 1.94 9 (0.0100 mole) of 5-(2'-thieny1)-2-then-
aldehyde in 10 m1 of refluxing ethanol. After 15 minutes of refluxing, the

reaction flask was plunged into an ice bath. The pH was adjusted to 3 with

101

concentrated hydrochloric acid. The acid solution was heated to boiling and
water was added to incipient cloudiness. After further cooling in the ice
box, the thenoin crystallized and yielded 1.46 9 (0.00396 mole, 75.2%) of
product: mp 117-121°. It was recrystallized from carbon tetrachloride: mp

126-128°; reported (46) 125-130°.
Preparation of 5,5'-di—(2"-thienyl)-2,2'-theni1

This compound was prepared by oxidizing the thenoin in the manner described
by Weiss (49). A mixture of 1.46 g (3.76 mmoles) of 5,5'-di-(2"-thienyl)-
2,2'-thenoin, 0.40 g (5.0 mmoles) of ammonium nitrate, approxiametly 0.1 g
of cupric acetate and 20 m1 of 80% acetic acid-water was stirred and refluxed
for one hour. The reaction mixture was then cooled in the ice box and the thenil
was collected by filtration. The yield of the dark brown thenil was 1.22 g
(3.16 mmoles, 84.7%). It was recrystallized from dioxane-water: mp 188—189°;
reported (46) 189-191°. The infrared spectrum is shown on p 151. The ultra—

violet spectrum is shown on p 155.

Preparation of methyl 5,5'-di-(2"—thieny1)-2,2'-thenilate

 

A 1.00 g (0.259 mmole) quantity of 5,5'-di-(2"-thieny1)-2,2'-theni1 was
suspended in a mixture of 10 m1 of dioxane and 5 ml of water. To this, 1.5 g
(27 moles) of potassium hydroxide was added and the mixture was refluxed
under nitrogen for 4 hours, during which time most of the thenil dissolved.

The solution was cooled and acidified with 6N hydrochloric acid to congo red.
After extraction of the aqueous layer with ether (3X25 ml), the extracts

were combined and dried. This ethereal solution was inmediately treated with

a solution of diazomethane in ether prepared from 0.51 g of N-methyl,N-nitroso-

urea. Removal of the solvent under reduced pressure gave an oil which was

102

crystallized from 60-90° petroleum ether to obtain 0.98 g (0.23 mmoles,
94%) of the ester: mp 112-113°. For analysis, it was recrystallized from
60-90° petroleum ether.

Analysis, Calc'd for C H 0 S ° C, 54.52; H, 3.37; S, 30.64.

l9 l4 3 4'
Found: C, 54.62; H, 3.37; S, 30.70.

Preparation of 3—br0mothiophene

 

A three liter, three-neck flask fitted with an overhead stirrer and reflux
condenser was charged with 900 m1 of water, 400 g (6.12 moles) of zinc dust_
and 750 m1 of glacial acetic acid. The suspension was heated to reflux with
vigorous stirring and 1290 g (5.33 moles) of 3,4-dibromothiophene was added
in one portion. After an hour of refluxing, 100 g of fresh zinc dust was added.
A second portion of 100 g of zinc dust was added an hour later, and a final
portion of 100 g of zinc dust was added after another two hours. The reaction
mixture was then refluxed an additional 9 hours with vigorous stirring.

After cooling the reaction flask, a Dean-Stark trap was interposed between

the condenser and the reaction vessel. The reaction mixture was then heated

until only one phase distilled over. The distillate was washed with water

(100 m1) and saturated sodium bicarbonate (100 ml). The oily product was

dried and distilled through an 8 inch Vigreaux column under vacuum to obtain

614 g (3.76 mole, 68.0%): bp 83-86° (58 torr), nD24
20

bp 149-151° (760 torr), nD 1.5861. The yield based on recovery of 232 g

of 3,4-dibromothi0phene was 86.1%.

1.5892; reported (129)

Preparation of 3-thenaldehyde

 

This material was prepared as described by Vincent (130). A three liter,

three-neck flask was equipped with a nitrogen inlet tube, a dr0pping funnel,

103

a thermometer extending below the level of the reaction mixture, a condenser,
and an overhead stirrer. Lithium metal, 17.5 g (2.52 moles), previously ham-
mered and scissored (131) was added to 400 m1 of anhydrous ether in the flask.
The ether SUSpension of lithium metal was cooled to -5° by immersion in a dry
ice-isopropanol bath and 157 g (1.15 moles) of n-butylbromide was added with
stirring during two hours. After one hour, the n-butyllithium solution was
cooled to -60° and 163 g (1.00 mole) of 3-bromothiophene in 100 m1 of dry
ether was added dr0pwise during 1.5 hours maintaining the temperature below
-60°. The 3—thienyllithium solution was then treated dropwise during one
hour at —60° with a solution of 100 g (1.26 moles) of freshly distilled di-
methylformamide in 100 ml of dry ether. The reaction temperature was main-
tained below -60° with constant stirring for an additional 6 hours, and then
poured into 200 m1 of saturated aqueous ammonium chloride. The organic layer
was separated and the aqueous layer was extracted with ether (3X200 ml). The
combined ether extracts were dried and the ether was removed by flash dis-

tillation. The residue was vacuum distilled to obtain 71.2 g (0.635 mole,

63.5%): bp 50-53° (1.3 torr) n024

n020 1.5860.

1.5806; reported (132) bp 78° (14 torr)

Preparation of 3,3'-thenoin

This compound was prepared by the method of Campaigne and LeSuer (54). A
refluxing solution of 28.0 g (0.250 mole) of 3-thenaldehyde in 250 ml of
95% ethanol was treated in one portion with 19.5 g (0.300 mole) of potas-
sium cyanide in 100 m1 of water. The bright red-orange solution was refluxed
2 hours, cooled, and acidified to pH 3 with 2N hydrochloric acid. After fur-
ther dilution with 600 m1 of cold water and cooling in the refrigerator, the
precipitate of crude thenoin was collected. It amounted to 12.0 9 (0.0536

mole, 42.8%): mp 111-113°. Upon concentration of the mother liquor to one

104
third its volume, an additional 2.2 g of product were obtained. The total

yield of product was 14.2 9 (0.0633 mole, 50.7%).

Preparation of 3,3'-theni1

 

The procedure described by Campaigne and Bourgeois (9) was employed with
some modifications in this synthesis. A solution of 32.6 g (0.130 mole) of
cupric sulfate pentahydrate in 45 ml of pyridine and 25 ml of water was
heated on a steam bath and 11.8 9 (0.0527 mole) of 3,3'-thenoin was added.
After heating and stirring for two hours, the reaction mixture was poured
into 600 ml of cold water. The dark brown precipitate was collected by fil-
tration after refrigerating the reaction mixture overnight. It was air dried
and continuously extracted with 30-50° petroleum ether for 24 hours. The pet-
roleum ether was evaporated with an air stream and the bright yellow thenil
weighing 9.88 9 (0.0445 mole, 84.4%) was obtained. It was sublimed for further
purification: mp 75-76°; reported (9) mp 75-76°. The infrared spectrum is

shown on p147. The ultraviolet Spectrum is shown on p154 .

Preparation of 3,3'-thenilic acid

 

This compound was prepared from 3,3'-thenil under exactly the same condi-
tions as described for the synthesis of 2,2'-thenilic acid from 2,2'-thenil.
The crude acid is apparently less stable than its 2,2'-isomer, precluding a
meaningful % yield determination. The crude acid, a pale tan powder, chars at
85-90° and melts at 101-103°. It is reported (9) to char below 90°. Neutraliza-
tion equivalent, calc'd for C10H80332: 240; found: 257 and 259.

Preparation of methyl 3,3'-theni1ate

 

A solution of 15.0 g (0.268 mole) of potassium hydroxide in 100 ml of water

105

under a nitrogen atmosphere was heated to 80° in an oil bath with vigorous
stirring, and 5.55 g (0.0250 mole) of 3,3'-thenil was added. Heating and
stirring were continued until solution was effected in about 4 hours. The
reaction mixture was cooled below room temperature in an ice bath and neu-
tralized to congo red with concentrated hydrochloric acid. The acidified
solution was extracted with ether (3X50 m1) and the combined extracts were
dried. This stirred solution was immediately treated with an ethereal sol-
ution of diazomethane prepared from 5.0 g of N-methyl,N-nitrosourea. After
stirring for 30 minutes, the ether was removed under reduced pressure. The
residue was recrystallized from 60-90° petroleum ether to obtain 4.48 9
(0.0176 mole, 70.3%) of the ester: mp 80-81°. For analysis, the ester was
recrystallized twice from 60-90° petroleum ether.

Analysis, Calc'd for C H 0 S ' C, 51.95; H, 3.96; S, 25.18.

11 10 3 2'
Found: C, 52.06; H, 3.97; S, 25.18.

Preparation of 3-bromothianapthene

 

The procedure described by Szmuszkovicz (133) was used without modifi-
cation in this synthesis. A two liter, three-neck flask was charged with
71.9 g (0.535 mole) of thianapthene, 73 g (0.89) of sodium acetate and 380 ml
of chloroform. Nith vigorous overhead stirring, 28 ml (86.8 g, 0.543 mole) of
bromine in 70 m1 of chloroform was added dropwise during 30 minutes to the
thianapthene solution. The reaction mixture was cooled by occasional immer-
sion in an ice bath to keep the temperature near 25°. After stirring an add-
itional hour following the bromine addition, 100 ml of water was added, and
the chlorform solution was separated, washed with 200 ml of water, 100 m1 of
5% sodium hydroxide, 200 m1 of water, and finally, 200 ml of saturated sodium

chloride. The chloroform solution was dried, and the solvent removed by

106

rotary evaporator. The residue was vacuum distilled to yield 93.8 g (0.440
mole, 82.3%) of 3-bromothianapthene: bp 83-84° (0.4 torr) nD25 1.6592; re-
ported (133) bp 90-105° (1.5 torr).

Preparation of 3-thianapthaldehyde

 

This compound was prepared by the reaction of 3-thianapthyllithium and
dimethylformamide, and has since been reported (57). A solution of 75 m1
of 1.6N n-butyl lithium in hexane (Foote Chemical Co., Exton Penn.) was
cooled to -70° in a dry ice-isopropanol bath. A solution of 21.3 g (0.100
mole) of 3-bromothianapthene in 50 ml of dry ether was added during a per-
iod of 20 minutes with constant stirring. The reaction temperature was al-
lowed to warm to -20° and stirred an additional 30 minutes. The thianapthyl-
lithium solution was treated with 8.78 g (0.120 mole) of freshly distilled
dimethylformamide in 50 ml of dry ether during 30 minutes with stirring.
After one hour of further stirring, the reaction mixture was hydrolyzed by
pouring it into 60 ml of 3N hydrochloric acid and 100 g of ice. The ether
layer was separated and the aqueous layer was extracted with ether (3X100 ml).
The combined dried ether solutions were flash distilled to remove the sol—
vent. The crude aldehyde was dissolved in 50 ml of ethanol and treated with
30 g of sodium bisulfite in 100 ml of water. After being set aside for sev-
eral minutes, the bisulfite adduct crystallized and was collected by vacuum
filtration, washed with ether, and air dried. It was dissolved in 100 m1 of
warm water, and 100 ml of saturated sodium carbonate was added. The oily al-
dehyde crystallized on being set aside in the ice box. The yield of product

was 10.8 g (0.0666 mole, 66.6%): mp 53-55°; reported (57) mp 56-57°.

107

Preparation of 3,3'-benzo[b]thenoin

This compound was prepared exactly as described by Campaigne and Neiss
(56). The procedure was repeated at least 20 times, however, no increase
in yield could be realized. The melting points in the original article for
both 3,3'-benzo[b]thenoin and 3,3'-benzo[b]thenil are in error (134). A mix-
ture of 6.0 g (0.037 mole) of 3-thianapthaldehyde, 20 m1 of 95% ethanol and
1.0 g (0.015 mole) of potassium cyanide was refluxed for 3 hours. The reac—
tion mixture was set aside overnight in the ice box, and then neutralized
with glacial acetic acid. After one week, the product precipitated and was
collected by filtration. After recrystallization from ethanol, the thenoin

melted at 128-130° and amounted to 1.00 g (0.00309 mole, 16.7%).

Preparation of 3,3'-benzo[b]thenil

 

This compound was prepared by oxidation of the thenoin using Weiss' method
(49). A mixture of 1.00 9 (0.00309 mole) of 3,3'-benzo[b]thenoin, 0.36 9
(0.0045 mole) of ammonium nitrate, approxiametly 0.1 g of cupric acetate,
and 10 ml of 80% acetic acid-water was refluxed for 90 minutes with stirring.
After cooling the reaction mixture in an ice bath, the thenil crystallized.

It was collected by vacuum filtration, and washed with ethanol, then ether and
air dried to obtain 0.82 9 (0.00255 mole, 82.5%) of product: mp 188-189°,
mol wt (by mass spectrum) calc'd for C H 0 S ° 322; found: 322. The infrared

18 10 2 2'
spectrum is shown on p 152. The ultraviolet spectrum is shown on p 154.

Attempted preparation of methyl 3,3'-benzo[b]thenilate

Various quantities of 3,3'-benzo[b]thenil were subjected to reaction
conditions that had previously been shown to sucessfully result in rearrange-

ment of a thenil to a thenilic acid. With 3,3'-benzo[b]thenil, however, they were

108

uniformly unfruitful. Following exactly the procedure described for the
preparation of 2,2'-thenilic acid, but using 3,3'-benzo[b]thenil, gave,
after treatment of the reaction mixture workup with diazomethane, a residue
which resisted crystallization attempts. An infrared spectrum showed a typ-

ical ester carbonyl absorption at 1720 cm"1

but no -0H absorption. The mass
spectrum showed a strong parent peak at m/e 192 (72%) and no peaks beyond
this ratio. Since methyl 3-thianapthylcarboxylate has a mass of 192, and is

a liquid, the evidence suggests this compound as the product.

Attempted preparation of 3,3'-benzo[b]thenil

 

An alternate synthesis of diaryl diketones described by Bauld (10) was
tried in the synthesis of 3,3'-benzo[b]thenil. A solution of 3.23 ml (4.25 g,
0.025 mole) nickel tetracarbonyl (K&K Laboratories, Plainview, N.Y.) was added
in one portion* to 6.50 g (0.025 mole) of 3-iodothianapthene in 50 ml of
dry tetrahydrofuran. The reaction apparatus had been previously swept clear
of air with a stream of purified nitrogen. The reaction mixture was heated
slowly until a thermometer an inch above the level of the reaction mixture
registered 65°. After cooling, the reaction mixture was set aside for 6 hours,
and 150 ml of water was then added. Another 12 hours were allowed to elapse
before further product isolation. The solid material was removed by fil-
tration, air dried, and extracted continuously with 30-50° petroleum ether.
Evaporation of the petroleum ether gave 4.2 g of pale yellow product in the

form of needles, mp l72-l74°. The infrared spectrum showed no absorptions in

 

* The nickel tetracarbonyl was most conveniently removed from the steel
shipping cylinder by a short piece of tygon tubing attached to a glass
inlet on the top of the addition funnel. Gentle suction applied to the
apparatus then effected removal. Caution is emphasized; antidotes for

 

nickel tetracarbonyl poisoning are only partially effective.

109

the 1620-1660 cm‘1 range, typical of thenils prepared in this study. The
product did show a strong carbonyl absorption at 1685 cm-1. Since 3,3-di-
thianapthylketone melts at 174° (135) this compound may have been formed

instead of the desired product.

Preparation of 2-thenoylphenylmethanol

 

The procedure described by Biel (7) was slightly modified for the syn-
thesis of this material. A solution of 11.2 g (0.100 mole) of freshly dis—
tilled 2-thena1dehyde, 10.6 g (0.100 mole) of benzaldehyde and 10 g of potas—
sium cyanide in 50 ml of 95% ethanol and 20 m1 of water was refluxed for
30 minutes. The hydroxyketone spontaneously crystallized when the reaction
mixture was cooled in an ice bath. The product was collected by vacuum fil-
tration and triturated with 300 ml of 5% aqueous sodium bicarbonate for
30 minutes. The suspension was filtered and the filter cake washed well
with water and dried: mp 123-124°; reported (7) 132-134°. This compound was
further purified by washing it with several portions of 50/50 (v/v) ether-
30-50° petroleum ether to obtain 16.6 9 (0.0732 mole, 73.2%) of product.
This material is pure enough to be used directly in the preparation of 2-

thienylphenyl diketone.

Preparation of 2-thienylphenyl diketone

 

The procedure described by Biel (7) was followed exactly in this prep-
aration. A suspension of 10.9 g (0.050 mole) of 2-thenoylphenylmethanol,
0.5 g of cupric acetate and 5.6 g (0.070 mole) of amnonium nitrate in 100
m1 of 80% acetic acid-water was refluxed and stirred for 90 minutes. After
cooling the reaction mixture to room temperature, 100 ml of water was added

and the product was allowed to crystallize in the refrigerator. The diketone

110

was collected by vacuum filtration and air dried to obtain 7.54 9 (0.0349
mole, 69.8%). It was further purified by crystallization from 50/50 meth-
anol-water: mp 67-68°; reported (7) 59—60°. The infrared spectrum is shown

on p 153. The ultraviolet spectrum is shown on p 155.

Preparation of 2-thienylglyoxal

 

The procedure described by Kipnis and Ornfelt (61) was used to prepare
this compound. A mixture of 160 ml of dioxane, 5.5 ml of water, and 27.8 g
(0.250 mole) of selenium dioxide was stirred and heated at 60° until the
solid dissolved. To this selenous acid solution, 31.5 g (0.250 mole) of
2-acetylthiophene was added, and the mixture was heated at reflux for 4 hours.
Gray selenium metal was deposited as a lump which caused some interference
with the stirrer. The solution was filtered hot through a sintered glass
funnel to remove the selenium metal, and the dioxane was removed under re-
duced pressure with gentle heat. The residue of 2-thienylglyoxa1 and 2-
acetylthiophene was vacuum distilled and the fraction boiling at 62-72° (
0.5 torr) was collected. The yield was 21.6 g (0.154 mole, 61.6%). This mat-
erial was 88% pure by nmr, the only impurity being 12% 2-acetylthiophene. No

further purification was attempted.
Preparation of 5-iodo-2-methylthiophene

The procedure described here has not been previously reported in the
synthesis of this compound. A solution of 23.8 g (0.243 mole) of 2-methyl-
thiophene in 30 m1 of benzene in a 500 m1 Morton flask was cooled in an ice
bath and small alternate portions of 43.3 g (0.200 mole) of yellow mercuric
oxide and 63.0 g (0.248 mole) of iodine were added during a period of 30

minutes, with vigorous stirring. The suspension of mercuric iodide was

111

removed by filtration and the filter cake was washed with ether. The com-
bined organic solvents were washed with 50 m1 of saturated sodium thio-
sulfate solution, dried, and the solvents were removed on a steam bath. The
residue was vacuum distilled to obtain 46.8 g (0.209 mole, 83.6%) of pro-
duct: bp 61-64° (0.5 torr); reported (136) bp 88-89° (14 torr).

Preparation of 5'—methyl-2,2'-thenoin

A Grignard reagent was prepared from 17.9 9 (0.0800 mole) of 5-methyl,2-
iodothiophene and 1.94 9 (0.0800 mole) of magnesium in 200 ml of dry ether,
under nitrogen in a dropping funnel. With constant overhead stirring, the
Grignard reagent was added dr0pwise to 10.4 9 (0.0732 mole) of 2-thienyl-
glyoxal in 200 m1 of dry ether at -50°. The addition required about 60 minutes.
After stirring an additional 60 minutes, during which time the brick red
precipiate of the adduct changed to beige, the reaction temperature was a1-
1owed to rise to 0°, and 50 ml of saturated aqueous ammonium chloride was
added in one portion. The suspension was allowed to stir until solution was
effected. The ether layer was separated, and the aqueous layer was extracted
with ether (3X100 ml). The combined ether solutions were dried, and the ether
was removed in a rotary evaporator. The residue of thenoin weighing 6.43 9
(0.0270 mole, 36.9%) mp 118—120°, was sufficiently pure to be used for the

synthesis of 5-methy1-2,2'—thenil.

Preparation of 5-methyl-2,2'-theni1

 

A 3.75 9 (0.0158 mole) quantity of the crude 5'-methyl-2,2'-thenoin was
dissolved in a mixture of 15 ml of pyridine and 8 ml of water. To this, 10.0 9
(0.0400 mole) of cupric sulfate pentahydrate was added, and the mixture was
stirred and heated in an oil bath at 80° for 90 minutes. The solution was

poured into 300 m1 of cold water and refrigerated for 4 hours. After

112

neutralizing the reaction mixture with concentrated hydrochloric acid, the
cold solution was extracted with methylene chloride (4X50 ml). The combined
extracts were dried and the solvent was removed with a stream of air. The
residue was continuously extracted with 30-50° petroleum ether in a Soxhlet
apparatus until the bright yellow color was exhausted. After evaporation of
the petroleum ether with a stream of air, the yellow thenil was collected
and amounted to 2.81 g (0.0119 mole, 75.3%): mp 42-45°. For analysis it was
sublimed twice at 70° (0.01 torr) mp 44-46°. The infrared spectrum is shown
on p 150. The ultraviolet spectrum is shown on p 155.

Analysis, Calc'd for C H 0 S ° C, 55.91; H, 3.41; S, 27.14.

11 8 2 2'
Found: C, 55.74; H, 3.41; S, 26.97.

Preparation of methyl 5-methyl-2,2'-theni1ate

 

A suspension of 472 mg (2.00 mmoles) of 5-methyl-2,2'-thenil in 15 m1
of water and 1.0 g (18 mmoles) of potassium hydroxide was heated at 80° under
nitrogen until solution was effected, about 3 hours being required. The re-
action mixture was cooled to 10° and acidified to pH 2 with concentrated
hydrochloric acid. The precipitated thenilic acid was inmediately extracted
with ether (2X30 ml). The combined ether solutions were shaken with a mix-
ture sodium sulfate and Norit-A and filtered. The ether solution was immed-
iately treated with an ether solution of diazomethane prepared from 0.5 g
of N-methyl,N-nitrosourea and set aside for one hour at room temperature. The
solvent was removed under reduced pressure at room temperature, and the
residue of ester crystallized from 60-90° petroleum ether. The yield of pro-
duct was 326 mg (1.12 mmole, 56.1%): mp 93-96°. For analysis, it was recry-

stallized twice from 60-90° petroleum ether: mp 95-96°.

113

Analysis, Calc'd for C11H1203SZ: C, 53.91; H, 4.15; 5, 23.99.

Found: C, 53.71; H, 4.22; S, 24.10.

Preparation of 2-iodothiophene

 

This compound was prepared by the method described by Wirth et al. (66).
The process is more convenient than the more usual method using iodine and
mercuric oxide. A 500 m1 Morton flask was charged with the following reagents
in the order listed: 120 ml of water, 5.6 m1 of concentrated sulfuric acid,
320 ml of glacial acetic acid, 80 m1 of carbon tetrachloride, 32.8 g (0.186
mole) of iodic acid, and 84.0 g (1.00 mole) of thi0phene. This mixture was
stirred vigorously with an overhead stirrer while 81.6 g (0.321 mole) of
iodine was added in small portions during a 30 minute period. The reaction
mixture was poured into a seperatory funnel and 100 ml of carbon tetrachloride
was added. The aqueous layer was separated, and the organic layer was washed
with 100 ml of water, followed by 100 m1 of 10% sodium thiosulfate. The or—
ganic phase was dried and the solvent was removed by distillation through
an 8 inch Vigreaux column at atmOSpheric pressure. The residue was vacuum
distilled to obtain 132 g (0.629 mole, 62.9%) of 2-iodothiophene based on
thiophene. The yield was 98.0% based on iodine: bp 62-72° (12 torr); reported
(66) 73-78° (18 torr).

Preparation of 2-methoxythiophene

 

The method reported by Sice (64) was employed in the preparation of this
compound. A solution of 59.1 g (2.57 mole) of sodium metal in 700 ml of an-
hydrous methanol (freshly distilled from magnesium) was prepared. To this
sodium methoxide solution, 177 g (0.843 mole) of 2-iodothiophene and 33.6

g of powdered cupric oxide were added. The mixture was refluxed with moderate

114

stirring for 30 hours. The cooled reaction mixture was poured into 1.5

l. of cold water and extracted with ether (4X400 ml). The ether extracts
were combined and washed with water (2x200 ml). The ether solution was dried
and the solvent removed by flash distillation. The residue was vacuum

distilled to obtain 74.4 g (0.652 mole, 77.3%) of 2—methoxythiophene:

24
D

4

bp 71-74° (50 torr) n 1.5258; reported (64) bp 74-75° (50 torr) nD2 1.5261.

Preparation of 5-methoxy-2-thena1dehyde

 

With constant stirring, 16.9 g (0.110 mole) of phosphorus oxychloride
was added dr0pwise to 8.76 g (0.120 mole) of dimethylformamide during a
period of 15 minutes, maintaining the reaction temperature below 50°. To
this mixture 11.4 g (0.100 mole) of 2-methoxythiophene was added dropwise,
maintaining the temperature below 30°. The reaction mixture formed a thick
paste, which was heated to 50° for a few minutes on a steam bath. After being
set aside for 30 minutes, the semisolid material was hydrolyzed with 50 m1
of ice and water. The pH of the mixture was adjusted to 6 with 5M sodium
hydroxide and the aqueous oily suspension was extracted with ether (3X50 ml).
The combined dried ether extracts were flash distilled to remove the ether.
The residue was vacuum distilled to obtain 5.65 9 (0.0398 mole, 39.8%) of
the aldehyde: bp 71-72° (0.1 torr); reported (64) 79-81° (0.9 torr).

Preparation of 5,5'-dimethoxy-2,2'-thenil

 

A solution of 11.4 g (0.100 mole) of 2-methoxythiophene in 150 ml of
carbon disulfide was cooled to 5° and 26.1 g (0.100 mole) of stannic chloride
in 30 ml of carbon disulfide was added dropwise with stirring. This was
followed by the addition of 6.35 9 (0.0500 mole) of oxalyl chloride in 30

ml of carbon disulfide. Both additions were made a rate to maintain the

115

reaction temperature below 5°. The mixture was stirred an additional 2 hours
at 0-5°, and then poured into 100 ml of ice water. The reaction mixture was
filtered to remove the bulk of the crude product.

The filtrate was poured into a separatory funnel and the carbon disulfide
was separated. The aqueous layer was extracted with ether (4X100 ml). The
combined organic phases were washed with saturated sodium bicarbonate until
the aqueous layer was basic to litmus. The organic phase was dried, and evap-
orated to obtain a tar residue. The residue was dissolved in a minimum amount
of methylene chloride and chromatographed on alumina (Merck Act. II, column
prepared in 30-50° petroleum ether). The column was eluted with acetone,
and all bright yellow fractions were collected and combined. Evaporation of
the acetone with a stream of air gave a bright orange solid which was cry-
stallized from methanol and combined with the material isolated from the
original reaction mixture following hydrolysis. The total yield was 4.64 9
(0.0165 mole, 33.0%): mp l36-138°. For analysis the thenil was recrystallized
three times from methanol. The infrared Spectrum is shown on p 148. The
ultraviolet spectrum is shown on p 155.

Analysis, Calc'd for C12H100452: C, 51.05; H, 3.57; S, 22.72.
Found: C, 51.08; H, 3.56; S, 22.79.

Preparation of 2-isopropoxythiophene

 

To a solution of 23.0 g (1.00 mole) of sodium metal in 500 ml of dry
isopropanol (freshly distilled from sodium) 105 g (0.500 mole) of 2-iodo-
thiophene and 20 g of cupric oxide were added. The mixture was stirred well
and refluxed for 30 hours. The reaction mixture was then cooled until it
became quite viscous, and 300 ml of water was added with constant stirring.

This mixture was poured into 2 l. of water in a separatory funnel and the

116

lower layer was separated. The upper layer was extracted with methylene
chloride (4X100 ml). The lower layer was combined with the methylene
chloride extracts and filtered through a bed of celite, and then through
anhydrous magnesium sulfate. The solvent was removed in a rotary evaporator
at water pump pressure. The residue was distilled to obtain 8.02 9 (0.0564
mole, 11.3%): bp 50-65° (13 torr), of material slightly contaminated with
2-iodothiophene. The 2-iodothiophene can be removed only by reacting the
mixture with magnesium under grignard conditions, and hydrolyzing with

water. A final distillation serves to remove the thiophene.

Preparation of 5,5'-diisopropoxy-2,2-theni1

 

A mixture of 1.42 g (10.0 mmoles) of 2-isopropoxythiophene and 2.61 g
(10.0 mmoles) of stannic chloride in 50 m1 of carbon disulfide was cooled
to 0° and 0.64 g (5.0 mmoles) of oxalyl chloride in 10 ml of carbon disul-
fide was added with constant stirring during 30 minutes. After an addi-
tional 15 minutes of stirring, the reaction mixture was poured into 50 g of
ice and the reaction flask was washed with 50 ml of ether. The carbon disul-
fide-ether-water solutions were combined and the aqueous layer was separated
and extracted with ether (2X50 ml). The combined organic phases were washed
with water and saturated sodium bicarbonate until the washings were basic to
litmus. The solvents were removed with a stream of air, and the tarry res-
idue was dissolved in 50 ml of methanol, treated with Norit, filtered, and
set aside. The solid material which separated from the solution was sublimed
at 140° (0.1 torr) to yield 0.28 g (0.828 mmole, 16.6%) of the bright yellow
thenil: mp 101-103°. For analysis, it was resublimed. The infrared Spectrum

is shown on p 148. The ultraviolet spectrum is shown on p 156.

117
Analysis, Calc d for C16H180452: C, 56.78; H, 5.36; S, 18.95.

Found: C, 56.80; H, 5.49; S, 19.07.

Preparation of 2-thianapthaldehyde

 

The procedure described by Shirley and Danzig (59) was followed with
minor modifications for the synthesis of this compound. A two liter, three-
neck flask, equipped with an overhead stirrer was charged with 400 ml of
l.6N n-butyllithium in hexane (Foote Chemical Co., Exton, Penn.) . With
constant stirring, 67.0 g (0.500 mole) of thianapthene in 200 ml of dry
ether was added to the butyllithium solution dr0pwise over a period of 1.5
hours in an atmOSphere of nitrogen at -10°. After all the thianapthene had
been added, the reaction temperature was raised briefly to 20° to allow the
evolution of butane and then recooled to 0°. The lithiothianapthene was then
treated dropwise with 36.6 g (0.500 mole) of freshly distilled dimethylform-
amide in 200 m1 of dry ether during 30 minutes. The reaction mixture was
then refluxed for 1.5 hours and poured into 400 m1 of 3N hydrochloric acid
and 500 g of crushed ice. After separation of the ether layer, the aqueous
layer was extracted with ether (3x300 ml). The organic phases were combined,
dried, and the ether was removed in a rotary evaporator. The residue of al-
dehyde was dissolved in 300 m1 of 95% ethanol and 200 g of sodium bisulfite
in 700 ml of water was added. The resulting bisulfite adduct was collected
by vacuum filtration, washed copiously with ether until white, and air dried.
The adduct was dissolved in 500 ml of water with gentle heat, and 500 m1 of
saturated sodium carbonate was added to decompose it. The resulting alde-
hyde separated from solution as an oil which solidified on being set aside
in the ice box. The yield was 54.9 g (0.339 mole, 67.8%): mp 33-35°; reported
(59) mp 34-34.5°.

118

Preparation of 2,2'-benzo[b]thenoin

 

This compound was prepared according to the method described by Shirley
and Danzig (59). Thus, 32.4 g (0.200 mole) of 2-thianapthaldehyde in 400 m1
of 50% ethanol-water and 2.0 g of potassium cyanide was refluxed for 20 min-
utes. The thenoin separated from solution spontaneously during the reflux
period. After cooling to room temperature, the thenoin was collected, washed
well with water and air dried. The yield was 27.4 9 (0.0846 mole, 84.6%):
mp 152-154°; reported (59) 154-156°.

Preparation of 2,2'-benzo[b]theni1

 

Weiss' oxidation method was used to prepare this thenil (49). A mixture
of 30.3 9 (0.0942 mole) of 2,2'-benzo[b]thenoin, 12.0 g (0.15 mole) of
ammonium nitrate, l g of cupric acetate, and 200 ml of 80% acetic acid-
water was heated at reflux for 90 minutes. The thenil crystallized spontan-
eously from the violet solution on cooling. It was collected by vacuum fil-
tration, washed with cold methanol and ether, and dried. The yield of pro-
duct was 29.9 9 (0.0928 mole, 98.7%) mp 236-238°. For analysis, it was
recrystallized twice from benzene: mp 239-240°. The infrared spectrum is
shown on p 152. The ultraviolet Spectrum is shown on p 154.

Analysis, Calc'd for C H

18 1002 2‘
Found: c, 67.08; H, 3.08; 5, 19.96.

S C, 67.06; H, 3.12; S, 19.89.

Preparation of methyl 2,2'-benzo[b]thenilate

 

This compound was prepared under the typical thenilic rearrangement con-
ditions described previously. Thus, 1.61 g (5.00 mmoles) of 2,2'-benzo[b]thenil

and 1.12 g (20 mmoles) of potassium hydroxide in 20 m1 of water was heated

119

to 80° in an oil bath with stirring under a nitrogen atmosphere. Complete
solution of the thenil was effected in about one hour. After cooling to
room temperature in an ice bath, the reaction mixture was acidified to
congo red with concentrated hydrochloric acid, and then immediately ex-
tracted with ether (3X50 ml). The dried extracts were inmediately treated
with an ether solution of diazomethane prepared from 1.03 g of N-methyl,
N-nitrosourea. Evaporation of the solvent gave a solid which was recrystall-
ized from 60-90° petroleum ether to obtain 1.10 g (3.10 mmoles, 62.0%) of the
ester: mp 103-104°. For analysis it was recrystallized from 60-90° petro-
leum ether.

Analysis, Calc'd for C H 0 S ° C, 64.38; H, 3.98; S, 18.09.

19 10 3 2'
Found: C, 64.15 H, 3.95; S, 17.98.

Preparation of 2-(1'-adamantyl)thiophene

 

This compound was prepared by following Wynberg's procedure exactly (71).
A vigorously stirred, refluxing solution of 64.5 g (0.300 mole) of l-bromo-
adamantane (Aldrich Chemical Co., Milwaukee, Wisc.) in 500 ml of freshly
distilled thiophene was treated dropwise with 83.0 g (0.320 mole) of stannic
chloride during 30 minutes. The reaction was allowed to continue an addi-
tional 15 minutes after the evolution of hydrogen bromide ceased. A mixture
of 200 ml of concentrated hydrochloric acid and 200 g of ice was added to the
cooled reaction mixture. The organic phase was separated and washed with 100
m1 of 4N sodium hydroxide and 200 ml of water. After drying the solution
with magnesium sulfate, the thi0phene was removed by distillation. The res-
idue was distilled and all material boiling between 120-150 (1 torr) was col-
lected. This distillate, amounting to 50.8 g (0.243 mole) of a mixture of 2-

and 3-(1'-adamanty1)thiophenes was dissolved in 230 m1 of ethanol. To this

120
well stirred refluxing solution, a mixture of 93 ml of saturated aqueous
sodium acetate and 62.5 g (0.226 mole) of mercuric chloride in 230 m1 of
ethanol was added dr0pwise during tWo h0urs. After refluxing an additional
15 minutes, the suspension of 2-chloromercuri-5-(l'-adamantyl)thiophene
was cooled to 50° and filtered. The solid was air dried and then refluxed
in 500 m1 of 6N hydrochloric acid containing a few mls of benzene until the
solid had all dissolved. The acid solution was cooled and extracted with
benzene (3X100 ml). The benzene was removed in a rotary evaporator at
reduced pressure and the residue of 2-(1'-adamantyl)thiophene was crystal-

lized from ethanol. The yield was 26.8 g (0.123 mole, 41.0%).
Preparation of 5,5'-di-(l"-adamantyl)-2,2'-theni1

A 15 m1 quantity of 1.6N n-butyllithium in hexane (Foote Chemical Co.,
Exton, Penn.) was added in one portion with stirring to 4.36 g (0.020 mole)
of 2-(1'-adamantyl)thiophene in 30 ml of dry ether under a nitrogen atmOSphere.
This reaction was carried out in a jacketed addition funnel. After 10 min-
utes of stirring, the jacket of the addition funnel was filled with a mix-
ture of dry ice and isopropanol. The thoroughly chilled solution of ada-
mantylthienyllithium was added dropwise over a period of 20 minutes to
1.18 9 (0.0100 mole) of dimethyl oxalate in 50 ml of dry ether with stirring
at -60° or lower. After one hour of stirring at -60°, the reaction temp-
erature was allowed to rise to -10° and 50 ml of 5% sulfuric acid was added.
The organic phase was separated and the aqueous phase was extracted with 50 m1
ether. The combined dried extracts were evaporated to yield an orange res-
idue. This was triturated with 2:1 benzenezmethanol to give a bright yellow
solid: mp 244-247°. It was recrystallized from cyclohexane to obtain 1.07 9

(0.00218 mole, 21.8%): mp 257-258°, of the thenil. For analysis it was

121

recrystallized twice from cyclohexane. The infrared spectrum is shown on
p 151. The ultraviolet spectrum is shown on p 156.
' I o . o
Analys1s, Calc d for 030H345202' C, 73.42, H, 6.98. S, 13.07.
Found: C, 73.17; H, 6.83; S, 12.91

Preparation of 2-fluorothiophene

 

This compound was prepared using a combination of the procedures devel-
oped by Schuetz (72) and Gronowitz (73). Lithium metal, 7.29 g (1.05 moles)
was hammered and scissored (131) and the pieces added to 150 m1 of dry ether
in a one liter, three-neck flask. After cooling the lithium SUSpension to
-10° by means of a dry ice-isopropanol bath, 54.5 g (0.500 mole) of ethyl
bromide was added with stirring while holding the temperature between 0°
and -10°. The ethyllithium solution was stirred an additional hour after the
lithium metal had dissolved, and then was treated dropwise with 33.6 g (0.400
mole) of freshly distilled thiophene in 100 m1 of dry ether during 20 min—
utes. The ethane was driven off by heating the reaction mixture to 30°
for a few minutes. The thienyllithium solution was recooled to -10° and
a strong stream of perchloryl fluoride (Columbia Chemical Co., Columbia,
S.C.) was added. The addition was accompanied by a strong blue luminescence.
The addition of the perchloryl fluoride was stopped when the luminescence
was no longer noted and the reaction mixture began to turn dark brown. The
solution was purged with nitrogen for 45 minutes and then poured into 200
m1 of 5M sodium carbonate solution. After filtering the reaction mixture,
the aqueous su5pension was extracted with ether (3x200 ml). The ether ex-
tracts were combined, dried and distilled from a water bath heated to 40°
to remove the ether. The residue was examined by nmr and was found to be

50% 2-fluorothiophene, 30% thiophene, and 20% ether. The yield of

122

2-fluorothiophene was, therefore, 8.4 9 (0.0823 mole, 21%). This material
was distilled through a 24 inch spinning band column at atmospheric pressure
to give an azeotrope boiling at 77°. Examination of this material by nmr
showed it to consist of 88% 2-f1uorothiophene and 12% thiophene. This mat-

erial was used in all subsequent experiments calling for 2-fluorothiophene.
Preparation of 5-f1uoro-2-thena1dehyde

A solution of 1.6N n-butyllithium in hexane, 70 ml (Foote Chemical Co.,
Exton, Penn.) was cooled in an ice bath to 3° and 10.2 g (0.100 mole) of 2-
fluorothiophene in 50 ml of dry ether was added dropwise during one hour.
After stirring 15 minutes following the 2-fluorothiophene addition, a sol-
ution of 8.0 g (0.11 mole) of freshly distilled dimethylformamide in 40 ml
of dry ether was added dropwise during 30 minutes while maintaining the re-
action temperature below 10°. The reaction mixture was allowed to stir 15
minutes longer, and then poured onto 100 g of ice. After acidifying with
50 ml of 6N hydrochloric acid, the organic phase was separated, and the
aqueous phase was extracted with ether (4X100 ml). The organic layers were
combined, dried and the ether was removed by flash distillation. The residue
was vacuum distilled to obtain 8.2 g (0.063 mole, 63%): bp 60-6l° (63 torr)
11023 1.5482, of aldehyde. A 2,4-dinitrophenylhydrazone was prepared of the
aldehyde in the usual manner and recrystallized twice from ethanol for
analysis: mp 257-258° dec.

Analysis, Calc'd for C11H6FN 0 S: C, 42.72; H, 1.96; N, 18.12; S, 10.37.

44
Found: c, 42.64; H, 2.01: N, 17.96; 5, 10.44.

Preparation of 5,5'-dif1uoro-2,2'-theni1

This compound was synthesized in a manner analogous to the preparation

123

of 5,5'-di—(1"-adamantyl)-2,2'-thenil. A solution of 3.69 9 (0.0362 mole)
of 2-f1uorothiophene in 30 ml of anhydrous ether was stirred and treated
under a nitrogen atmosphere, with 25 m1 of l.6N n-butyllithium in hexane
(Foote Chemical Co., Exton Penn.). The mixture was stirred 15 minutes at
room temperature, then cooled to -60° and added dropwise to 2.18 9 (0.0181
mole) of dimethyl oxalate in 100 ml of dry ether also cooled to -60°. The
addition was performed at such a rate that the reaction temperature was
maintained below -50°. After one hour of additional stirring, the reaction
mixture was allowed to warm to room temperature and 50 m1 of water was added,
followed by 20 m1 of 10% hydrochloric acid. The layers were separated and
the aqueous layer was extracted with ether (2X50 ml) and combined with the
original ether layer. The ether solution of the thenil was dried and the
solvent was evaporated with an air stream to leave a black iridescent res-
idue. The residue was continuously extracted with 30-50° petroleum ether in
a Soxhlet apparatus. Evaporation of the petroleum ether gave 1.60 9 (0.00620
mole, 34.3%) of the thenil: mp 89-95°. The product was purified by sublimation
at 70° (0.5 torr) to give the analytically pure thenil: mp 126-128°. The
infrared spectrum is shown onla 149 The ultraviolet spectrum is Shown on
(J 155

Analysis, Calc'd for C H F 0 S

104222‘
Found: H, 46.70; H, 1.65; S, 24.62.

C, 46.50; H, 1.56; S, 24.83.

Preparation of methyl 5,5'—dif1uoro-2,2'-thenilate

 

A solution of 0.6 g (11 mmoles) of potassium hydroxide in 10 m1 of
water was heated to 80° in an oil bath with stirring under nitrogen and
0.190 g (0.737 mmole) of 5,5'-difluoro-2,2'-thenil was added in one por-
tion. The suspension was heated and stirred until solution was complete,

about 45 minutes being required. The reaction mixture was cooled in an ice

124

bath below room temperature and acidified to congo red with 6N hydrochloric
acid. The acid solution was quickly extracted with ether (3X25 ml). The
combined ether extracts were dried and treated with an ethereal solution of
diazomethane prepared from 0.2 g of N-methyl,N-nitrosourea. Evaporation of
the ether gave a pale yellow oil, which solidified on being set aside. The
thenilic ester, 0.166 g (0.572 mmole, 77.7%) was sublimed twice for analysis:
mp 55-57°.

Analysis, Calc'd for C H F 0 S ° C, 45.51; H, 2.78; S, 22.09.

11 8 2 3 2‘
Found: C, 45.74, H, 2.63; S, 22.18.

Preparation of 5-acety1-2-fluorothiophene

 

A solution of 1.02 g (0.010 mole) of 2-fluorothiophene and 0.87 g (0.011
mole) of acetyl chloride in 15 ml of carbon disulfide was cooled to 5° in an
ice bath. The cold solution was then treated dropwise with 2.87 g (0.011 mole)
of stannic chloride with stirring during a one hour period. Stirring was con-
tinued for an additional hour, and the purple solution was then hydrolyzed
by cautiously adding 10 ml of water. After separation of the carbon disul-
fide layer, the aqueous layer was extracted with carbon disulfide (2X20 ml).
The combined extracts were washed with 20 m1 of water and dried. The dried
organic solution was filtered and flash distilled to remove the solvent. The
residue, 1.15 9 (0.0072 mole, 72%) was distilled in a micro molecular still
at 80° (15 torr) to obtain the pure product. A 2,4-dinitrophenylhydrazone
of the ketone prepared in the usual manner, was twice crystallized from
ethanol for analysis: mp 246-248°.

Analysis, Calc'd for 012“ FN 0 S: C, 44.58; H, 2.49; N, 17.33; S, 9.92.

8 4 4
Found: C, 44.46; H, 2.54; N, 17.21; S, 9.84.

125

Preparations Directed Toward the Synthesis of 2-Trif1uoromethylthiophene

Preparation of 5,5,5-trifluorolevulinic acid

The procedure described by Brown et. al. (137) was used with some mod-
ification in this synthesis. A well stirred mixture of 39.2 g (0.276 mole)
of ethyl trifluoroacetate, 48.2 g (0.276 mole) of dimethyl succinate, and
6.35 g (0.276 mole) of sodium metal pellets (K&K Chemical Co., Plainview,
N.Y.) was heated cautiously to initiate the reaction. After the exothermic
reaction had cooled to room temperature, 50 m1 of dry ether was added and the
reaction mixture was refluxed for 18 hours. The ether was removed by distil-
lation and 100 ml of 10N sulfuric acid was added cautiously. The acidic sol-
ution was extracted with ether (6X50 ml). The extracts were combined and
dried by filtration through magnesium sulfate. After removal of the ether
by evaporation on a steam bath, the residue was distilled to give 56.8 g

(0.208 mole, 74.8%) of diethyl trifluoroacetosuccinate: bp 100-135° (0.5 torr)

19 20
"0 0

fluxing it in 350 ml of 43% aqueous sulfuric acid for 7 hours. After cooling,

1.4000; reported (137) n 124104. The distillate was hydrolyzed by re-
the acid solution was extracted with ether (6X50 ml). The ether extracts were
combined and dried, and the ether was removed in a rotary evaporator. The
crude acid was distilled to obtain 17.4 g (0.102 mole) of trifluorolevulinic
acid: bp 66-68° (0.1 torr). The acid was recrystallized from benzene to obtain
12.3 9 (0.0723 mole, 26.2%) of pure product based on ethyl trifluoroacetate:
mp 53-56°; reported (137) 54-56°.

Attempted preparation of 2-trifluoromethylthiophene

 

An attempt was made to synthesize the title compound using as a guide

the procedure described by Hartough (138) for the synthesis of 2-methylthiophene.

126

A 12.3 9 (0.0723 mole) quantity of 5,5,5-trif1uorolevu1inic acid was dis-
solved in 50 m1 of dry ether, and 1.73 9 (0.0720 mole) of sodium hydride

was added in small portions with stirring. The ether was then removed in a
rotary evporator, and the residue of sodium 5,5,5-trifluorolevulinate,

13.8 9 (0.0719 mole, 99.4%) was mixed with 35 g of phosphorous pentasulfide
and 50 g of sand. This mixture was cautiously heated to 190° during 90 min-
utes. A continuous stream of nitrogen was passed through the apparatus during
the heating period. Provision was made to trap any volatile material in a
cold trap at 0°. At l30-l35°, a strong evolution of white fumes began but

no material could be trapped. The pot residue consisted of a totally intract-

able tar. None of the desired product was in evidence.

Preparations Directed Toward the Synthesis of 2-Phenoxythiophene

Via the Ullmann reaction

 

Various 2-halothiophenes were treated under typical Ullmann reaction condi-
tions with sodium or potassium phenolate. The method reported by Bacon
(139) and the one described by Jones and Moodie (76) were used as guides
in these attempted syntheses of 2-phenoxythiophene.

A quantity of the halothiophene ranging from 10 to 100 mmoles was dis-
solved in 10 times its volume of solvent. To this solution, an equivalent
of cupric oxide, or cuprous oxide, or finely divided copper metal was added
along with a 5% excess of sodium or potassium phenolate. This mixture was heat-
ed at reflux usually 24 hours, with stirring under nitrogen. In one exper-
iment, using 2—iodothi0phene with dimethylformamide as the solvent and cu-
prous oxide as the catalyst, the reaction time was one week. The solvents were

either removed by evaporation at reduced pressure or extracted with water.

127

The residue was examined by nmr after distillation unless vpc had indicated
that no reaction had occurred. The following reaction systems were examined:
2-iodothiophene in water, dimethylformamide, methanol, ethanol, dimethyl sul-
foxide, dimethylacetamide, molton phenol, and pyridine; 2-bromothiophene was
tried in all these solvents except water; 2-chlorothiophene was tried in di-
methylformamide and dimethyl sulfoxide; 2-fluorothiophene was tried in di-
methylformamide, dimethylacetamide and dimethyl sulfoxide. In no case could
even a trace of material be isolated as to indicate that the desired reaction
had occurred. In almost all cases, the halothiophene was recovered in better

than 95% yield.

Preparation of bis(2-thienyl)iodonium salts

 

The procedure described by Behringer (140) was utilized for this synthesis.
A well stirred mixture of 130 ml of acetic acid, 40 ml of acetic anhydride,
20 g (0.094 mole) of potassium iodate and 58 g (0.69 mole) of thiophene was
cooled below 10° by immersion in an ice bath. A cold mixture of 30 m1 of
acetic acid and 24 m1 of sulfuric acid was added to the above mixture, while
taking extreme care not to allow the reaction temperature to rise above 10°.
After stirring overnight at room temperature, the reaction mixture was di-
luted with 200 ml of water and extracted with ether (3X100 ml). The ether
extracts were discarded and the aqueous layer was treated with Norit-A and
filtered. The colorless filtrate was treated with a solution of 15 g of po-
tassium iodide in 100 m1 of water. The precipitated bis(2-thieny1)iodonium
iodide was collected by filtration and air dried to obtain 8.50 9 (0.0202
mole, 21.6%) mp 127-128°; reported (140) 128-129°. The substitution of 15 g
of sodium bromide in 100 ml of water in place of the potassium iodide gave
8.15 9 (0.0218 mole, 23.4%) of bis(2-thieny1)iodonium bromide: mp 185-186°.

The crude iodonium bromide was recrystallized from dimethyl sulfoxide: mp

128

210-212°; reported (140) 215-219°. These compounds were not purified fur-

ther.
Reaction of sodium phenolate with bis(2-thienyl)iodonium salts

Behringer's method (81) for the synthesis of diaryl ethers was used as a
guide in this reaction attempt, as well as a modification described by
Crowder et al. (141). A stirred suspension of 4.20 9 (0.0100 mole) of
bis(2-thienyl)iodonium iodide and 5.80 9 (0.0500 mole) of sodium phenolate
(City Chemical Corp., New York, N.Y.) was heated in 50 m1 anhydrous methanol
at reflux for 24 hours. The methanol was then removed in a rotary evapor-
ator and 100 m1 of water and 100 m1 of ether was added to the residue. After
Shaking the mixture, the ether was removed and dried. Evaporation of the
ether gave only phenol and 2-iodothiophene, identified by vpc. The same pro-
cedure was tried using water as the reaction solvent. Again only phenol and
2-iodothiophene could be isolated. The use of bis(2-thienyl)iodonium bromide
in the procedure described above yielded only phenol and presumably a mix-

ture of 2-bromothiophene and 2-iodothiophene.

Preparation of 2-hydroxythiophene

 

The method reported in Organic Synthesis was modified for use in this
preparation. A Grignard reagent prepared from 84.0 g (0.400 mole) of 2-iodo-
thiophene, and 10.0 g (0.412 mole) of magnesium turnings in 400 ml of dry
ether was cooled to 0-5° by immersion in an ice bath. A solution of 62 g
(0.32 mole) of tert-butyl perbenzoate in 100 ml of dry ether was added during
45 minutes to the cold, well stirred Grignard solution. After stirring an ad-
ditional 3 hours, the solution was hydrolyzed by pouring it into 500 g of

ice and 5 ml of concentrated hydrochloric acid. The ether layer was quickly

129

separated and the aqueous layer extracted with ether (2X100 ml). The com-
bined ether solutions were extracted with 2N sodium hydroxide (3X60 m1),
washed with water until neutral, and dried. After establishing the absence
of peroxides, the ether was removed by distillation in a water bath held at
50°. The residue was distilled at water pump pressure to obtain 37.6 g (0.240
mole, 75.0%), of 2-tert-butoxythiophene.

A 7.81 9 (0.0500 mole) quantity of 2-tert-butoxythiophene was trans-
fered to a round bottom flask fitted with a distillation head. The appar-
atus was purged with nitrogen, and 0.1 g of p-toluenesulfonic acid was added.
The distillation apparatus was again purged with nitrogen by several flush-
evacuate cycles. The distillation flask was immersed in an oil bath pre-
viously heated to 125° and kept there until a brisk evolution of isobutane
ensued, which was drawn off by a water aspirator. The water aspirator was
replaced by an oil pump when the gas evolution had ceased and the 2-hydroxy-
thi0phene was distilled. The yellow product, 4.82 9 (0.00482 mole, 96.4%)
bp 51° (1 torr); reported (67) 91-93° (13 torr) was stored in the freezer

at -20° until needed.

Attempted preparation of 2-(2',4'-dinitrophenoxy)thi0phene

 

A stirred solution of 2.00 9 (0.0200 mole) of 2—hydroxythiophene in 20 m1
of anhydrous ether was treated in small portions with 0.48 9 (0.0200 mole) of
sodium hydride. After the evolution of hydrogen had ceased, a solution of
3.72 9 (0.0200 mole) of 2,4-dinitrof1uorobenzene in 10 ml of ether was added.
An immediate dark blue color formed. After stirring for 30 minutes, a brown-
violet precipitate formed and was collected. This materia1,weighing 4.2 g,
was water soluble and was discarded without further characterization. Fur-

ther workup failed to reveal any of the desired phenoxythiophene.

130

The reaction was repeated under conditions described by Reinheimer et al.
(86) for the preparation of 2,4-dinitropheny1 ethers of various phenols.

The main difference in this procedure is that triethylamine is substituted
for sodium hydride, and the reaction is carried out in acetone. In the pre-
sent study, the same brown-violet solid was formed when Reinheimer's method
was used as was noted previously. The product was apparently a condensation
product with the 2,4-dinitrof1uorobenzene, Since without this material being
present, no reaction occurred. Further characterization of the product was

not attempted.

Preparation of 2-chloro-3,5-dinitrothiophene

 

The procedure described by Hurd and Kreuz (82) was followed to prepare
this compound. A well stirred solution of 37 g (0.31 mole) of 2-chlorothio-
phene, and 100 ml of acetic anhydride was cooled to -10° in an isopropanol-
dry ice bath. To this solution, a mixture of 50 g of 90% nitric acid (0.72
mole) and 100 ml of acetic anhydride was added dropwise at such a rate that
the reaction temperature remained below 0°. After 8 hours at this temperature
the reaction mixture was poured into 400 g of crushed ice. The cold solution
was extracted with 30-50° petroleum ether (3x250 ml). The organic phase was
filteredauuithe ether was removed by evaporation to give a yellow oil. The
oil was vacuum distilled and yielded 27 g (0.16 mole, 52%) of product: bp
57° (0.3 torr).

The oily 5-nitro-2-chlorothiophene was added with stirring during a 30
minute period to a mixture of 150 g of 90% nitric acid and 150 g of concen-
trated sulfuric acid cooled by immersion in an ice bath. The temperature was
maintained at 0-5° during the addition of the thiophene compound. The di-

nitro compound spontaneously crystallized from the nitrating mixture as it

131
formed. The reaction mixture was quenched by pouring it into 300 g of
crushed ice. The dinitro compound was collected by filtration and air dried

to obtain 22.5 g (0.11 mole, 69%). It was used without further purification.

Preparation of 2-phenoxy-3,5-dinitrothiophene

 

The method used by Hurd and Kreuz (82) was employed on a larger scale in
this synthesis. A 4.17 9 (0.0200 mole) quantity of 2-chloro-3,5-dinitro-
thiophene was placed in a large porcelain mortar and just covered with
30 ml of 30-50° petroleum ether. In small portions, and with constant
grinding, 2.32 9 (0.0200 mole) of sodium phenolate was added. The grinding
was continued until all of the ether evaporated. The resulting salmon colored
solid was recrystallized from 2:1 methanol-benzene to obtain 2.98 9 (0.0114

mole, 57.0%) of the ether: mp 153-154°; reported (82) 151.5-152.5°.
Attempted reduction of 2-phenoxy-3,5-dinitrothiophene

An attempt was made to reduce the nitro functions of 2-phenoxy-3,5-
dinitrothiophene, using a procedure developed for the reduction of 2-nitro-
thiophene to 2—aminothiophene (84). A mixture of 5.32 9 (0.0200 mole) of
2-phenoxy—3,5-dinitrothiophene and 20 m1 of concentrated hydrochloric acid
was warmed to 40°. In small portions, 7.08 9 (0.0600 mole) of 30 mesh tin
was added during a 15 minute period. The reaction temperature was held to
40° or lower during the addition of the metal by means of an ice bath. After
about one fourth of the tin had been added, a strong evolution of hydrogen
sulfide ensued. The tin addition was continued until all had been added,
during which time an intractable tar separated from the reaction mixture.
All attempts to separate the expected stannic chloride salt of the diamino-

thiophene failed.

132

Attempted preparation of 5,5'-dinitro-2,2'-thenil

 

A 0.50 9 (0.0023 mole) quantity of 2,2'-thenil was dissolved in 20 m1 of
acetic acid and 2 m1 of 90% nitric acid was added in one portion. The temp-
erature was held at 60° in an oil bath for 30 hours. The yellow solution was
poured into 50 ml of ice water and stirred until the oil crystallized. This
solid proved to be starting material, identified by mixed melting point, and

recovered in 97% yield.

Preparation of l,l-di-(2'-and 31thienyl)ethylene glycol

 

A 0.254 g (1.00 mmole) quantity of methyl 2,2'-thenilate or methyl 3,3'-
thenilate in 15 m1 of dry ether was allowed to stir for 18 hours with 50 mg
(1.32 mmoles) of lithium aluminum hydride. The reaction mixture was then re-
fluxed 30 minutes, and cooled. The solution was cautiously hydrolyzed with
1% hydrochloric acid. The ethereal layer was separated and the aqueous layer
extracted with 15 m1 of ether. The combined ether solutions were dried and
the ether was removed in a rotary evaporator. The residue was recrystallized
from 60-90° petroleum ether in the case of the 2,2'-compound. The 3,3'-com-
pound was an oil and was not further purified. The yield of l,l-di-(2'-thienyl)
ethylene glycol was 0.190 g (0.852 mmole, 85.2%): mp 62-63°. The yield of
l,l-di-(3'-thienyl)ethylene glycol was 0.170 g (0.752 mmole, 75.2%). Examin-
ation of the aromatic portion of the nmr Spectrum of each compound was made.
These spectra are shown on p 42. They clearly confirm that prototropic rear-

rangement does not occur during the thenilic acid rearrangement.
Degradation of 5,5'-dimethoxy-2,2'-theni1

The method described by Dakin and Harington (21) was used to degrade

133
5,5'-dimethoxy-2,2'-theni1 to 5-methoxy-2-thenoic acid. A mixture of 0.20 g
(0.71 mmole) of 5,5'-dimethoxy-2,2'-theni1, 15 ml of water, 10 m1 of meth-
anol, 0.3 g of sodium cyanide, and 0.3 g of ammonium chloride was refluxed on
the steam bath for 12 hours. The mixture was taken to dryness on a rotary
evaporator, and 20 m1 of 50/50 (v/v) ethanol-water and 2.0 g of sodium hy-
droxide was added to the residue. This mixture was refluxed for 3 hours,
treated while hot with Norit-A and filtered. The filtrate was cooled in the
ice box and then acidified with concentrated hydrochloric acid. The crude
acid was collected and air dried. It was sublimed at 110° (0.05 torr) and
weighed 0.11 g (0.69 mmole,97%). The pure material was shown to be iden-
tical to 5-methoxy-2-thenoic acid by mp and mixed mp: mp 162-163°; reported
(64) 162-163°, and by infrared spectrum.

General preparation of thenoic acids

 

The thenoic acids used in the pKa determinations were usually prepared
by silver oxide oxidation of the corresponding aldehydes. A general descrip-
tion of the procedure used in the preparation of 2-thenoic, 3-thenoic, 5-
chloro-2-thenoic, 5-methy1-2-thenoic, 5-methoxy-2-thenoic, 5-f1uoro-2-then-
oic, and 2-benzo[b]thenoic acid follows.

An alkaline silver oxide suSpension was prepared from 3.0 g (18 mmoles)
of silver nitrate and 2.0 g (50 mmoles) of sodium hydroxide in 30 ml of water.
A 10 mmole quantity of the aldehyde was added in one portion to the silver
oxide suspension, and the mixture was allowed to stir for one hour or at
least 30 minutes after a silver mirror had formed on the surface of the re-
action vessel. The solution was filtered, shaken with a small amount of Norit-A
and refiltered. The solution of the acid salt was acidified to pH 3 with 2N

hydrochloric acid. The precipitated acid was collected and dried. The acid

134
was recrystallized from water using a small amount of Norit-A for decolor-
ation. For the pKa determinations, the acids were recrystallized three times
from water, or until their melting points were constant. They were then dried
at 56° in a drying pistol at 0.1 torr, and stored in a desiccator over phos-
phorous pentoxide for one week. The acids prepared by this method are shown

in Table 7.

Table 7. Thenoic Acids Prepared by Oxidation of the Corresponding Aldehydes

 

 

 

Acid % yield melting point lit. melting point
2-thenoic acid 84 129-130° 129-130° (142)
3-then0ic acid 91 135-137° 138.4° (142)
5-chloro—2-thenoic acid 72 151-151.5° l46-147° (142)
5-methy1-2—thenoic acid 88 l38-139° 138-138.5° (142)
5-methoxy-2-thenoic acid 75 l62-l63° 162-163° (64)
5-f1uoro-2-thenoic acid 48 l45-146° ---
2-benzo[b]thenoic acid 47 243-244° 240-241.5° (143)

 

Preparation of 2-diacetoxymethyl-5-nitrothiophene

 

A solution of 22.4 g (0.200 mole) of 2-thena1dehyde in 50 m1 of acetic
anhydride was cooled to 10° and maintained there or below during the addition
of 12.1 ml (18.2 g, 0.260 mole) of 90% nitric acid. After the addition of the
nitric acid was completed, the reaction mixture was set aside in the ice
box overnight, and then poured into 400 ml of water. The solid product was
immediately collected, air dried, and recrystallized from 90% methanol. The

yield was 29.5 g (0.114 mole, 57.0%): 71-73°; reported (96) mp 73°.

Preparation of 5-nitro-2-thenoic acid

 

A suSpension of 5.18 9 (0.0200 mole) of 2-diacetoxymethyl-5-nitrothiophene
and 8.82 9 (0.0300 mole) of potassium dichromate in 30 ml of water was

treated dropwise with 20 m1 of concentrated sulfuric acid at such a rate that

135

the reaction temperature did not exceed 60°. The solution was refluxed 30
minutes, cooled and 20 ml of water was added. The precipitated acid was col-
lected, washed copiously with water and air dried. The yield of 5-nitro-2-
thenoic acid was 1.73 9 (0.0100 mole, 50.0%): mp 159-161°. For the pKa
determination, it was dried in the same manner as described previously for

the other 2-thenoic acids. The melting point of the purified acid was 159-160°;
reported (142) 158°.

Purification of 3-benzo[b]thenoic acid

 

A sample of 3-benzo[b]thenoic acid previously prepared in these labor—
atories by Dunigan (144) was purified as described previously for the thenoic

acids. The melting point was l62-l64°; reported (143) 174-175°.

Preparation of methyl 2,2-di-(2'-thieny1),2-ethoxyacetate
Method A

A solution of 2,2'-thenilic acid, prepared from 1.00 g (4.52 moles) of
2,2'-thenil,in 15 ml of methylene chloride was cooled to —70° in a dry ice-
isopropanol bath. To this solution, 6 ml (10.6 g, 91.5 mmoles) of chloro-
sulfonic acid was added in one portion with vigorous stirring. After 5 min-
utes, the dark red solution was quenched by pouring it into 25 m1 of anhydrous
ethanol, freshly distilled from sodium metal and previously cooled to -70°.
After adding 20 m1 of water, the acid solution was neutralized to congo red
by adding, in small portions, solid sodium carbonate while maintaining the
temperature below 0°. The solution was filtered, and the filtrate was ex-
tracted with methylene chloride (2X20 ml). The combined methylene chloride
solutions were dried and immediately treated with an ether solution of

diazomethane, prepared from 3.0 g of N-methyl,N-nitrosourea. After stirring

136

the solution for one hour at room temperature, the solution was treated with
Norit-A and filtered and the ether was removed at water pump pressure and
finally with an oil pump. The yield of crude ethoxyester was 1.19 g (4.21
mmoles, 93.3%). For analysis, the ester was distilled twice in a micro mol-
ecular still at 110° (0.3 torr). The nmr Spectrum showed T 2.7-3.3 (6H'S, m);
T 6.30 (3H'S, S); I 6.65 (2H's, q); T 8.80 (3H'S, t).

Analysis, Calc'd for C13H1403SZ: C, 55.29; H, 5.00; S, 22.71.

Found: C, 55.13; H, 4.94; S, 22.95.

Method B

Following the same procedure as outlined in Method A, 1.00 g (3.94 mmoles)
of methyl 2,2'-thenilate was treated with the same amount of chlorosulfonic
acid as used before. The treatment with diazomethane was omitted. The com-
bined dried methylene chloride extracts of the ethoxyester were taken to
dryness in a rotary evaporator at reduced pressure. The residue of product,
1.07 g (3.79 mmoles, 96.3%) was examined by nmr and found to be identical
with the material from Method A. In particular, no transfer of the methyl
group from the alcohol p0rtion of the ester to the acid portion had occured

in the carbonium ion.

Preparation of N-methyl-3-piperidyl-2',2"-thenilate hydrochloride

 

The synthesis of this compound was based on a procedure reported by Biel
et al. (7). To 30 m1 of n-heptane (dried over sodium) was added 1.00 g(3.94
mmoles) of methyl 2,2'-theni1ate and 0.459 g (3.98 mmoles) of N-methy1-3-
piperidinol (Aldrich Chemical Co., Milwaukee, Wisc.). The mixture was stirred
and heated to reflux. When all the reactants had dissolved, approxiametely
0.02 g of sodium methoxide was added in one portion, and the reaction was

allowed to continue for an additional 6 hours. At the end of this reaction

137

time, 15 ml of solvent was removed by distillation, the reaction mixture

was cooled, and 15 m1 of water was added. The ether-heptane solution was
extracted with water (2X20 ml) and the organic phase was dried. The solvents
were removed in a rotary evaporator at reduced pressure to obtain an orange
oil. The oil was dissolved in 20 ml of isopropanol and 30 m1 of ether, pre-
viously saturated with hydrogen chloride, was added. After being set aside
for a few minutes, the product crystallized, and was collected to obtain

0.35 g (0.946 mmole, 24%): mp 201-204° of the ester. For analysis, the amino-
ester hydrochloride was recrystallized three times from 90% ethanol; mp 201-
204°.

Analysis, Calc'd for C16H20C1N03SZ: c, 51.39; H, 5.39: N, 3.75: 5, 17.15.

Found: C, 51.22; H, 5.43; N, 3.86; S, 17.18.

In the same manner as described for the synthesis of N-methyl-B-piperidyl-
2',2"-theni1ate hydrochloride, the N-methyl—3-piperidy1 ester hydrochlorides
of three other thenilic acids were prepared.

From methyl 3,3'-thenilate, the yield of N-methyl-3-piperidyl-3',3"-
thenilate hydrochloride was 0.71 g (1.90 mmoles, 47.7%): mp 228-231°.

Analysis, Calc'd for C16H20C1N03SZ: C, 51.39; H, 5.39; N, 3.75; S, 17.15

Found: C, 51.44; H, 5.45; N, 3.66; S, 17.10.

From methyl 5,5'-dichloro-2,2'-thenilate, the yield of N-methyl-3-pip-
eridyl-5',5"-dichloro-2',2"-theni1ate hydrochloride was 0.65 g (1.47 mmoles,
37.4%): mp 203-205° dec.

Analysis, Calc'd for C16H18C13N0332: C, 43.40; H, 4.10; N, 3.16; S, 14.48.

Found: C, 43.54; H, 4.07; N, 3.02; S, 14.20.

From methyl 2,2'-benzo[b]thenilate, the yield of N-methyl-B-piperidyl-
2',2"-benzo[b]thenilate hydrochloride was 0.80 g (1.66 mmoles, 42%) mp 233-
234°.

138

Analysis, Calc'd for C24H24C1N03S2:

Found: C, 61.05; H, 5.17; N, 2.84; 5, 13.32.

C, 60.81; H, 5.10; N, 2.96; 5, 15.53.

Nonsynthetic Experimental Procedures

Reagents

Reagent grade acetone, “Spectroguality” methylene chloride, and "Iron
Free" chlorosulfonic acid were purchased from Matheson, Coleman and Bell,
East Rutherford, N.J., and were used as received. Tetramethylammonium tetra-
fluoroborate (TMATFB) was purchased from Aldrich Chemical Co., Milwaukee,
Wisc., and was dried in a vacuum desiccator over phosphorus pentoxide be-
fore use. Reagent grade dioxane was further purified by stirring it with li-
thium aluminum hydride for 24 hours, then refluxing it for 6 hours, and
finally distillation. It was stored over 5A molecular seive. After three days,
any unused dioxane was repurified. Potassium hydroxide solution was prepared
from 0.1 N "Acculutes" standard solution ampules (Anachemia Chemical Co.,
Champlain, N. Y.) and stored in a polyethylene container. It was standardized
against potassium hydrogen phthalate (145). Addition of a sample of the potas-
sium hydroxide solution to 2N barium hydroxide solution produced no observable
turbidity. Hydrochloric acid was prepared as a 0.1 N stock solution and stan-
dardized in the manner described by Laitinen (146). The acid used in the kin-
etic studies was diluted to 0.01 N. Sulfuric acid was standardized against
standard potassium hydroxide. Distilled water, used for all dilutions, was
boiled 15 minutes to remove carbon dioxide and stored in a container pro-
tected from the atmosphere by an Ascarite tube. For the kinetic studies, the
thenils were sublimed (3X) at 120° (0.1 torr), except for 5,5'-dimethoxy-
2,2'-thenil and 5,5'-di-(2"-thienyl)-2,2'-theni1 which were recrystallized

(3X) from methanol and dioxane, respectively.

139

Determination of the carbonium ion spectra

A. The ultraviolet spectra

Aliquots (100 ul) of a 0.01013 M solution of 2,2'-theni1ic acid in gla-
cial acetic acid were added to 50.0 ml samples of various weight percentage
solutions of aqueous sulfuric acid as shown in Table 8 below. The spectra
of the resulting red carbonium ion solutions were Scanned in the region of
650 nm to 340 nm on a Cary 14 spectrophotometer at 25:2°. Similarly, 100 u1
aliquots of the thenilic acid solution were added to 50.0 ml samples of var-
ious weight percentage solutions of chlorosulfonic acid in methylene chloride
as shown in Table 8. Since some of the solutions were somewhat unstable, the
time of mixing was noted and refered to to. The spectra were scanned three

t1mes 1n the reg1on of Ama and e for Amax was extrapolated back to to. The

x
loss in absorbance was linear as a function of time.

Table 8. Extinction Coefficients for the Di-(2-thieny1),carboxycarbonium Ion
as a Function of Acid Concentration.

 

 

wgt % H2504 6t (A=516 nm) wgt % ClSO3H at (x=518 nm)

 

0 0

98.86 21,200 100 30,000
89.87 22,000 50.0 33,050
85.02 24,300 10.0 33,050
80.29 20,300 5.00 33,100
69.91 * 1.00 33,800

0.50 31,200

0.10 24,900

0.0024 *

 

*
unmeasurable

 

140

B. The nuclear magnetic resonance Spectra

The nmr spectra of the di-(2-thienyl)carboxycarbonium ion were deter-
mined on a Jeolco C60H high resolution spectrometer at -55° (measured by an
internal thermocouple) and at +30° (measured by a thermometer inserted in the
probe). In practice, a magnetically stirred solution of 0.30 g of 2,2'-
thenilic acid and 0.1 g of TMATFB in 2.0 ml of methylene chloride was cooled
to -70° in a dry ice-isopropanol bath. To this solution, 0.5 ml of chloro-
sulfonic acid was added in one portion and allowed to stir for 15 seconds. A
portion of this solution was transfered to an nmr tube precooled to -70°, and
after the sample had equilibrated in the Spectrometer, the spectrum was scanned
from -7 to +10 T. The carbonium ion solution derived from 2,2'-theni1ic acid
was not sufficiently stable to allow its spectrum to be determined at +30°.
However, the ion derived from methyl 2,2'-theni1ate could be observed at this

temperature. The same procedure was used as in the acid derived spectra.
Determinations of the ionization constants of the thenoic acids

The ionization constants of the thenoic acids were determined following
the potentiometric method described by Albert and Serjeant (105). A titra-
tion cell was constructed from a 125 m1 glass jacketed resin kettle as il-
lustrated on p 141. In practice, a continuous flow of water at 49.5° sup-
plied by the same constant temperature bath as was used in the kinetic studies,
was maintained through the jacket during all pH measurements. A Beckmann
Model 1019 research pH meter was standardized with Matheson pH 4 buffer sol-
ution (pH=4.06 at 50°) and the meter was checked with Beckmann pH 7 buffer
solution (pH=6.975 at 50°). The buffer solutions and the electrodes (Corning
# 476002 SCE reference electrode, Beckmann # 40498 glass electrode) were pre-

viously equilibrated in the titration cell and continually checked until no

141

h
Anschutz thermometer ‘——“-*“~
v

u ? s—nitrogen inlet
1
---b- water out

~*—-—-—-10 m1 buret

 

 

 

 

 

 

 

Acrylic top :=

 

 

 

 

 

 

 

 

 

t r- . —+ . . .
W6 e 1" magnetic st1rr1ng bar

no drift was noted on the pH meter.

After standardization, the cell was emptied, dried, and 100.0 m1 of
COZ-free water was pipetted into the cell. A quantity of the purified then-
oic acid, never less than 0.10 mmole or more than 0.12 mmole,was weighed to
0.1 mg, and dissolved in the cell by stirring. The cell was flushed with
nitrogen, and a slow stream of nitrogen, which had been previously passed
through a wash bottle containing C02-free water, was maintained in the cell
during the entire pKa determination.

After the entire apparatus, including the electrodes had equilibrated,
l.000:0.002 ml of 0.1002:0.0001 N potassium hydroxide was added to the sol-
ution of the thenoic acid by means of a certified class A 10 m1 buret pro-
tected from the atmosphere by an Ascarite tube. The pH meter reading was re-
corded after the meter had stabilized at the new value. Nine sucessive 1.000
ml amounts of potassium hydroxide were added to the thenoic acid solution,

taking the pH meter reading after each addition. These readings were used to

142

compute the pKa of the thenoic acids as described on p 144. The pKa of

each acid was determined three times in succession.At the end of a series of
measurements on one acid, the cell was recharged with fresh pH 4 buffer and
the meter was rechecked. The pH 4 reading consistently checked within 0.004

pH units of the correct value.

Procedure for the kinetic determinations

 

A constant temperature bath of eight gallons capacity (the large volume
helped to minimize temperature variations during the heating cycle) was re-
gulated to :0.04°. Temperatures were measured by a Will Scientific Inc.

(New York, N.Y.) #26846, -5° to +101° thermometer graduated to 0.1° and
stated to comply with the National Bureau of Standards Circular #8. The
temperature was further checked by a calibrated Hewlett-Packard (Palo Alto,
Cal.) #2801A quartz digital thermometer. It was not possible to detect any
variation in temperature at different points in the bath. Temperatures below
room temperature were achieved by using a pump driven heat exchanger im-
mersed in a dry ice-isopropanol bath at -70°. The pump was activated by the
thermostat.

Kinetic measurements for all of the thenils, with the exception of 3,3'-
benzo[b]thenil, 2,2'-benzo[b]thenil, and 5,5'-di-(2"-thieny1)-2,2'-thenil,
were carried out in the same manner. A sample of the thenil, in the range of
2.5 to 3.0 mmoles, was weighed to 0.1 mg and placed in a 250 m1 teflon screw
cap bottle, along with a magnetic stirring bar. The thenil was dissolved
in 100.0 ml of equilibrated dioxane with stirring. The solution was contin-
uously stirred during the kinetic run by means of a G. Frederick Smith sub-
mersible magnetic stirrer mounted directly beneath the reaction vessel. A
50.00 ml quantity of 0.1 N temperature equilibrated potassium hydroxide 501-

ution was pipetted into the thenil solution with vigorous stirring. The vessel

143

was sealed with a screw cap arrangement that permitted the insertion and
withdrawal of a pipette. A 10.00 ml aliquot of the thenil solution was
immediately withdrawn. The aliquot was quenched in a 10 ml sample of ice
cold acetone and the clock was started. Time was measured by three dif-
ferent timers which could be triggered simultaneously. They consisted of

a common wall clock and two rotary timers, one measuring in tenths of a
minute and the other in hundredths of a minute. No differences could be de-
tected in time duration with any of the timers in periods up to 300,000 se-
conds. The sample was titrated to either a phenolphthalein (pH=8.3) or thy-
mol blue (pH=8.0) end point with 0.01 N hydrochloric acid. The initial
aliquot determination was designated as [OHJi in equation 9, p 58. At var-
ious intervals, depending on the rate of potassium hydroxide disappearance,
10 ml aliquots were withdrawn and titrated as described.

Simultaneously with the kinetic runs, a blank sample was determined under
the same conditions described above, except that the thenil was omitted. In
this manner it was possible to correct for any acidic impurity present in
the thenil and also for any reaction of the base with the solvent. During
the 80° runs and to a lesser extent at 70°, a small correction (by never more
than 3%) was found to be necessary. At other temperatures, the correction
was too small to be measured.

Samples of 3,3'-benzo[b]thenil, 2,2'-benzo[b]thenil, and 5,5'-di-(2"-thieny1)-
2,2'-theni1 were too insoluble to be determined at the concentrations used
above for the other thenils. Accordingly, samples in the range 0.3 to 0.8
mmole, but never less than 0.1000 g were used in the kinetic determinations.
The amount of dioxane used was 100 m1, but the 50 m1 of potassium hydroxide
was replaced by 15 m1 of potassium hydroxide and 35 ml of water. All other
conditions were the same. The solubility of 5,5'-di-(l"-adamantyl)-2,2'-thenil

did not permit kinetic determinations to be made at any concentration.

144

Since a mixture of dioxane and water does not constitute an ideal
solution in the thermodynamic sense, an additional correction was employed
in calculating the initial concentrations of the reactants. The following
equation was used to calculate the actual volume of the reaction mixture

at the start of a given kinetic run

V . V](p]) + V2(92)

 

pmix

where v1 and v2 are the volumes of dioxane and water measured out, p1 and 92

are the densities of dioxane and water at a given temperature (110) and

pmix is the density of the mixture at the same temperature (109). In these

calculations, the thenils and potassium hydroxide are assumed to behave

ideally. Thus, by equation 22, 150 m1 of the mixture of dioxane and water

actually has the volumes indicated in Table 9 as a function of temperature.

Table 9. Actual Volume of a Mixture of 100 ml of Dioxane and 50 ml of Water
as a Function of Temperature.

 

 

Temperature 15° 30° 40° 50° 60° 70° 80°
Volume (ml) 147.8 147.3 147.1 147.0 147.0 147.0 147.0

 

Ionization constant calculations

By definition:

then

pKa = pH + -1og —fl—- = pH + log H5
HA A

where "pH" is read from the pH meter.

145

However, [HA]

[HA]st - [“Aldis = [HA1st - [n+1

the stoichiometric amount of the acid less that which

that is, [HA]

has dissociated,
- - - +
and[A]=[A]S,+[HA1dis =[Aht+m1

that is, [A-] = the stoichiometric amount of the anion plus that which
arises from the dissociation of the acid. This gives the pKa as in equation

23.

+
pKa . pH , ,Og [H41 - [8*] pg,
[A 1 + [H 1

 

Naturally, each stoichiometric concentration must be corrected with the

appropriate dilution factor as the titrant is added.

Calculations

 

All calculations in this study were performed on a Wang 320 electronic
calculator. This device was used in preference to a larger computing sys-
tem, which is actually more time consuming since it requires the transpo-
sition of data to punch cards, running the data and program decks, and wait-
ing for the print out. Even this assumes that an error free program is
available and there is no waiting to get on line.

The introduction of the directly addressable computing system such as
the Wang 700 series or the Hewlett-Packard 9100 series obviates the need
for a computor language such as FORTRAN, ALGOL or even the much simpler

BASIC. Furthermore, the capacity of these systems, 8K bit core memory in

146

the Wang 700B and up to twenty 960-Step programs, provides any organic
chemist certainly, and most scientists in general, with the capacity to
perform virtually any desired mathamatical manipualtion, but with the
highly desired advantage of instant output, as visual or hard copy.

The author believes that the next generation or two of such “desk
top computors" with greater capacity and reduced access time will render
obsolete present computing systems except, perhaps, for the most complex

calculations, or as large information storage systems.

Fvcmnpu.m.m we Ezspuwam umcagwcH .mm msamwm
Eu com ooofl ooNH OOIH oomfl OOIH ooom comm ooom comm coo:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aoueqatmsuedq %

147

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m m w: 3 “a4 4 [A m #1 _
ON
o:
08
r\>\\¥ om
r _ _ _ _ _ _ _ _ _ I
chmgu:.m.m mo Esguumam umcmcycH .mm «Lame;
« «1 3 fl m w m 1+1 “ +11 m *1
3
in. ON
Ii. o:
. om
11 om
_ _ _ _ p F _ _ _ [I

 

 

 

_.1EU com ooofi oowfi 00.2 003 coma ooom comm ooom oomm coo:

aoueathsued1 %

148

PoooIo-.N.N-oxooooeoo-.o.o to

szgpomam umsogwcH .mm mcomwm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F150 ooo oooo ooNH oooo oooo oooo ooom oomm ooom oomm ooo:
— ~ d F H — F E H M W
4 7: .3 4 .D A m _ q d

1
J
D.
w
13w
1..
1
D.
U
11 oo m

[I oo

_ _ _ _ _ _ _ _ _ _ _
Foams“-.N.N:>xoaosoomwwu:.m.m we Eacgumom vmcocwcH .mm meamwm
r In I M « [I I I I .
14 II om 2p.

1
J
D.
w
1: [L o: w.
AI?
1
w
1. 1L oo m

11 C 1. oo

_. h _ L _ _ _ _
Eu ooo oo H oomo ooIo oooo oooo ooo oomo ooom oomm ooo:

149

EU

oom

oooH

coma

COJH

_oouoo-.~.o-oeoo_coo-.m.m co Eoeoooom ooeaecoo .Po oeoooo

oomH

coma

ooom

comm

ooom

comm

ooo:

 

n

 

 

 

 

 

 

 

I

 

 

 

 

:j

 

 

 

 

 

 

n

j

 

 

 

 

 

H
a

H
u

_

H
.

A
d

H
4

 

.II om

o:

om

 

 

_wooed-.m.~-oeo_oooo-.m.m co Eoeoooom ooooecoo .oa oeoooo

 

Ii

H

 

 

—~

 

 

~-

 

 

 

 

 

 

 

 

 

 

 

_-

 

 

“

_

fl

_

H
q

_

0N

o:

.11 om

11 em

 

 

EU

oom

ooofi

ooNH

OOJH

coma

coma

ooom

comm

ooom

comm

ooo:

aouenatwsuedn %

aoueaatmsueda %

150

Foouoo-.~.~-PoIooe-m to soeoooom ooeoeeoo .mo oeoo_a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_-Eo ooo oooo ooNH ooIH oooo oomo ooou oomm ooom oomm ooo:
H1 _ I
. _ q; 1 “ a
TI ll ON
1 g 1:
11 .1 oo
1 om
_ _ o, _ _ o1 _ _ _ _
Pwcmgui.m.~-_ozumswnu.m.m mo Easpumam umcoswcH .Nv msomoa
1 I 1 a
T Q 1 o:
1\II:\||
E II ow
_ _ o I) _ I _ _ _ _ _
Eu ooo oooo oooo ooso oooH oooo oooN oomm ooom oomm ooo:

aoueantwsuedn %

aouezatwsuedq %

151

Propoo-.N.N-A_oooooo-.mo-oo-.o.o eo Eoeooaom ooeoecoo .mo oeoooo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Eu ooo oooH 002 005 com: com: ooom comm ooom oomm ooo:
I n 4 n u u H H H 1
3 q o - u - _ A q - m
I. 11 on
I. L o:
1. I. ow
I ow
h _ _ _ _ _ r — _ _ _
Fooooo-.N.N-AFooanooo-=_v-oo-.m.o to soeoooom oososcoo .oa osoooo
— n m n n H i 1 L n
_ _ _ 4 _ a q 1 _
TI 11 om
1 L o:
J om
l om
_ _ _ L _ _ _ _ _ _
Eu oom ooo“ coma 00+: 003 oomfl Doom comm ooom comm ooo:

aoueaatwsueda %

aoueqatwsuedq %

152

so ooo

coca

chH

cc:H

coca

chmgpmncoNcmau.m.m co Eocuumam umgacmcH .nv mcamou
oooo ooom

comm

cccm
—

comm
H

 

F

H

n

n

1—

 

—

4

 

 

 

 

 

 

n

_

d

 

 

 

 

 

 

_

 

 

 

 

 

_

 

 

_

_

_ _

_

_

_

r1

ccc:

 

 

Foamsumnco~cmni.u.~ oo Escuumam umcocwcH .o¢ mgoowm

 

-

n

H

 

44

 

 

 

 

 

 

 

 

 

 

I—Ibl

 

 

 

_

 

 

 

 

_

 

 

or

 

 

 

 

_

A

_

#

d

111

_

_

cm

0:

co

co

cm

o:

co

co

 

 

EU com

cccH

chH

cOJH

coca

ccoH

cch

comm

cccm

comm

coo:

aoueqthsu941 %

aoueaatwsuedz %

153

EU

mcoomxou Focmgapxcmwnpim co Eocuumam uwgocmco .oe ms:m_m

 

 

 

cco cccH chH chH coca coca cch comm cccm comm ccc:
I) m i 3 o i. m o I. c
. cm
I c:
C 8
cm
_ _ _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

33U911LNSU941 %

extinction coefficient

50,000

40,000

30,000

20,000

10,000 .

154

 

 

—1I I l 1 1 I 1 I 1

_ 1 2,2'-thenil _
x~\ . 2,2'-benzo[b]thenil—-—-

__ ' /\ 3,3'-benzo[b]theni1—-—:_
I" l, l' \, 3,3'-thenil ---------
l \

_ ' 1 1 ' - ..
.' 11 .’ \.

L. | \I 1 \‘ __
i \1 'I \1

_ , I. I \ _
! 11 I, ‘

- I ‘1 ' 1 .1
. 1 ’l 1

- 1 1 / 1 —.
I / ”x '\

"7% i \ ' // \\ \ -——

\ti . 1‘
I /

 

 

 

 

200 250 300 350 400 450 .4

Figure 49. Ultraviolet Spectra of various thenils

extinction coefficient

155

 

 

 

 

 

T I TTT I I 1 l
- 5,5'-di-(2"-thieny1)-2,2'-thenil-—————-
5,5'-difluoro-2,2'-thenil ——————
50,000 — 5-methy1-2,2'-thenil —-——-—- —-
2-thienylphenyl diketone ------------------
40,000 —‘
30,000 — ,\
/
/ \
' I \
I \
20,000 — / \
.A. \
K / / \‘ \
\. - ~ 4/ /' \\ \
1-—\ ' s‘ :_
10,000 \/ ’ \/ ‘% 5‘ \ \
o.‘-~ ’0‘ ’-/’ \\\ \‘ \
17“." \ \\
\\~~‘ \\ \
1 1 1 IL 1, ‘i-. | ‘35:1
200 250 300 350 400

Figure 50. Ultraviolet spectra of various thenils

450 ,

extinction coefficient

156

 

I 1 l I l 1 I

F 5,5'-diisopropoxy-2,2'-theni1

5,5'-di-(1"-adamantyl)-2,2'-theni1-—--— —-
5,5'-dimethoxy-2,2'-theni1 -

_ 5,5'-dimethyl-2,2'-theni1———--—- —- ——

5,5'-dichloro-2,2'-theni1-----------------

 

 

50,000

40,000 "’

30,000 — / \

20,000 I 1’

10,000 ;\ / / '\ “i ‘

 

l L 1 l l J 1 “\&\‘£ | 7S

 

200 250 300 350

Figure 51. Ultraviolet spectra of various thenils

 

REFERENCES

1.

10.
ll.
12.
13.
14.
15.
16.

17.

REFERENCES

J. S. Pierce in "Medicinal Chemistry," 2nd ed, A. Burger, Ed., Inter-
science Publishers, New York, N. Y., 1960, p 471. E. P. Claus,
"Pharmacognosy," Lea and Febiger, Philadelphia, Penn., 1961, pp 302-315.

. A. Mein, Justus Liebigs Ann. Chem., 6, 67 (1833).
’b

 

. w. Steinkopf and A. Wolfram, Justus Liebigs Ann. Chem., 421, 22 (1924).

 

J. H. Biel, L. G. Abood, w. K. Hoya, H. A. Leiser, P. A. Nuhfer, and
E. F. Kluchesky, J. 0rg_. Chem. , 26, 4096 (1961); see also reference 22
and references therein.

 

. L. G. Abood, A. M. Ostfeld, and J. H. Biel, Proc. Soc. Exp. Biol. Med

 

97, 483 (1958).

. An excellant review on psychotomimetic substances may be found in: D. F.

Downing, Quart. Revs. (London), 16’, 133 (1962).

 

. H. Biel, E. P. Sprengler, H. A. Leiser, J. Horner, A. Drukker, and
. L. Friedman, J. Amer. Chem. Soc., 71, 2250 (1955).

IC...

 

 

. 5. Z. Cardon and H. P. Lankelma, J. Amer. Chem. Soc., ZR) 4248 (1948).
. E. Campaigne and R. C. Bourgeois, J. Amer. Chem. Soc., 25, 2702 (1953).

 

N. L. Bauld, Tetrahedron Lett., 1841 (1963).

 

. Tuzun, M. Ogliaruso, and E. J. Becker, Org. an., 41, 1 (1960).

x0

 

. Nyberg, Acta Chem. Scand., 2%; 1087 (1966).

m
o

J. Corey and J. P. Schaefer, J. Amer. Chem. Soc., 82, 918 (1960).
’VD

 

. P. Gutsche, Org. React., 8, 364 (1954).

 

S. Jenkins, J. Amer. Chem. Soc., 56, 1137 (1934).

 

_.¢, U1 (1
0

March "Advanced Organic Chemistry," McGraw Hill Inc. , New York, N. Y.
968, p 582, and references therein.

 

R. C. .)Fuson, J. Corse, and C. H. McKeever, J. Amer. Chem. Soc. , 62, 3250
1940

157

18.

19.
20.

21.
22.

23.

24.

25.
26.

28.

29.

30.
31.
32.
33.
34.
35.

36.
37.

38.

158

L. F. Fieser, "Organic Experiments," 2nd ed, Raytheon Education Co.,
Lexington, Mass., 1968, p 219.

 

L. F. Fieser and Y. 0kumura, J. Org. Chem., 27, 2247 (1962).
'\/\1

L. F. Fieser and M. Fieser, "Reagents for Organic Synthesis," John Wiley
and Sons, Inc., New York, N. Y., 1967, p 164.

H. D. Dakin and C. R. Harington, J. Biol. Chem., 55, 487 (1923).

 

"Dimethyl Sulfoxide as a Reaction Solvent," Crown Zellerbach Corp.,
Camus, Wash., 1968, p 20.

An excellant although out of date review has been written by: W. S. Ide
and J. S. Buck, Org. React., 4, 269 (1948).

 

I. I. Lapkin, G. Y. Anvarova, and T. N. Povarnitsyna, J. Gen. Chem. USSR
36, 1945 (1966).

 

M. Gomberg and W. E. Bechmann, J. Amer. Chem. Soc., 50, 2762 (1928).

 

M. Ryang and S. Tsutsumi, Bull. Chem. Soc. Jap., 34, 1341 (1961).

 

. J. S. Walia, J. Singh, M. S. Chatta, and M. Satyanarayana, Tetrahedron

Lett., 1521 (1969).

W. P. Purcell and J. M. Clayton in "Annual Reports in Medicinal Chemistry,"
C. Cain, Ed., Academic Press, New York, N. Y., 1968, pp 314-321 and re-
ferences therein.

C. K. Ingold, "Annual Reports on the Progress of Chemistry," The Chemical
Society, London, 1928, p 24.

I. Roberts and H. C. Urey, J. Amer. Chem. Soc., 60, 880 (1938).

 

T. W. Evans and W. H. Dehn, J. Amer. Chem. Soc., 52) 252 (1930).

 

J. Hine and H. W. Hawaorth, J. Amer. Chem. Soc., 803 2274 (1958).

 

F. H. Westheimer, J. Amer. Chem. Soc., 583 2209 (1936).

 

E. Pfeil, G. Geissler, W. Jacquemin,and F. Lomker, Chem Ber.,89, 1210 (1956).

M. T. Clark, E. G. Hendley, and 0. K. Neville, J. Amer. Chem. Soc., 77,
3280 (1955). V“

 

D. G. Ott and G. G. Smith, J. Amer. Chem. Soc., 77, 2325 (1955).

 

K. B. Wiberg, "Physical Organic Chemistry," John Wiley and Sons, Inc.,
New York, N. Y., 1964, p 413.

G. B. Barlin and D. D. Perrin, Quart. Revs. (London), 29, 75 (1966).

 

39.

40.
41.

42.

43.
44.
45.
46.

47.

48.
49.
50.
51.

52.

53.
54.
55.
56.
57.
58.
59.

159

C. D. Ritchie and W. F. Sager in "Progress in Physical Organic Chemistry,"
Vol. 2, S. Cohen, A. Streitwieser, and R. Taft, Eds.,John Wiley and Sons,
Inc., New York, N. Y., 1964, p 323.

see reference 16, p 714.

J. G. Cannon, J. 0r . ChenL, 25, 959 (1960). S. B. Kadin and J. G. Cannon,
J. Org. Chem., 27, 348 (I962).

 

'V'b

S. Gronowitz in "Advances in Heterocyclic Chemistry,“ V01. 1, A. R.
Katritzky, Ed., Academic Press, New York, N. Y., 1963, p 47.

L. Michaelis and E. S. Fetcher, J. Amer. Chem. Soc., 52, 1246 (1937).

 

S. Selman and J. F. Eastham, Quart Revs. (London), 14, 221 (1960).

 

 

W. J. King and F. F. Nord, J. Org. Chem., 14, 638 (1949).

A. R. Dodson, Ph. D. Thesis, Michigan State University, East Lansing,
Mich. 1961, p 18.

J. H. Uhlenbroek and J. D. Bijloo, Rec. Trav. Chem. Pays—Bas, 79,
1181 (1960). 'W

 

S. Gronowitz and H. 0. Karlsson, Ark. Kemi, ll, 89 (1961).
M. Weiss and M. Appel, J. Amer. Chem. Soc., 72, 3666 (1948).

 

M. J. S. Dewer, and N. Trinajstic, J. Amer. Chem. Soc., 92, 1453 (1970).

 

E. Campaigne, R. C. Bourgeois, and W. C. McCarthy, Org. Syn. Coll. Vol. 4,
p 918. “

 

R. L. Burns, Ph. D. Thesis, Michigan State University, East Lansing,
Mich., 1966.

S. Gronowitz, Ark. Kemi, 8, 441 (1965).

 

E. Campaigne and W. M. LeSuer, J. Amer. Chem. Soc., 70, 1555 (1948).
’\N\:
A. W. Weston and R. J. Michaels, J. Amer. Chem. 5°C°’,333 1422 (1950).

 

E. Campaigne and E. S. Neiss, J. Heterocycl. Chem., 3, 46 (1966).

UVV

R. P. Dickinson and B. Iddon, J. Chem. Soc.,C, 2733 (1968).

 

 

see reference 42, p 75.

D. A. Shirley and M. J. Danzig, J. Amer. Chem. Soc., 336 2935 (1952).

 

60. see reference 42, p 90.

61.

F. Kipnis and J. Ornfelt, J. Amer. Chem. SOC-’.§§g 2734 (1946).

 

62.
63.
64.
65.
66.

67.
68.
69.
70.
71.

72.

73.

74.

75.
76.
77.
78.
79.

80.

81.

82.
83.
84.

160

W. Steinkopf and W. Hanske, Justus Liebigs Ann. Chem., 527, 264 (1936).

WL

 

W. Minnis, Org. Syn. Coll., Vol. 2, p 357.

 

J. Sice, J. Amer. Chem. Soc., 75, 3697 (1953).

 

E. Profft, Justus Liebigs Ann. Chem., 622, 196 (1959).

 

H. 0. Wirth, 0. Konigstein, and W. Korn, Justus Liebigs Ann. Chem., 634,
84 (1960) ”W”

 

C. Frisell and S.-0. Lawesson, Org. Syn. 43, 55 (1963).
see reference 42, p 8.

N. P. Buu-Hoi, Justus Liebigs Ann. Chem., 256, 1 (1949).

 

H. Kwart and M. M. Baeysky, J. Amer. Chem. Soc., 89, 580 (1958).

 

 

W. Hoek, J. Strating, and H. Wynberg, Rec. Trav. Chem. Pays-Bas, 85,
1045, (1966). m“

R. D. Schuetz, D. D. Taft, J. P. O'Brian, J. L. Shea, and H. M. Mork,
J. Org. Chem., 28, 1420 (1963).

 

S. Rodmar, B. Rodmar, M. K. Sharma, S. Gronowitz, H. Christiansen, and
U. Rosen, Acta Chem. Scand., 22, 907 (1968).

 

F. Kabbe, Ph. D. Thesis, Michigan State University, East Lansing,
Mich. 1967.

H. E. Ungnade, Chem. Rev., 38} 407 (1946).
E. Jones and I. M. Moodie, J. Chem. Soc., 7018 (1965).

 

E. Jones and I. M. Moodie, Tetrahedron, 21, 2413 (1965).

 

T. Kauffmann, Angew. Chem. Int. Ed., 4, 543 (1965).

 

R. W. Hoffman in "Dehydrobenzenes and Cycloalkynes," Academic Press,
New York, N. Y., 1967, p 292.

0. Spinelli, C. Dell'Erba, and A. Salvemini, Ann. Chim. (Rome), 52,
1156 (1962).

F. M. Behringer, A. Brierley, M. Drexler, E. M. Grindler, and C. C.
Lumpkin, J. Amer. Chem. Soc.,zg, 2708 (1953).

 

 

C. D. Hurd and K. L. Kreuz, J. Amer. Chem. Soc., 74, 2965 (1952).
Mb
G. W. Stacy and D. L. Eck, Tetrahedron Lett., 5201 (1967).

 

see reference 128, p 513.

 

85. M. T. Bogert and R. L. Evans, Ind. Eng. Chem., 18, 299 (1926).

86.

87.
88.
89.
90.
91.
92.
93.
94.

95.
96.
97.

98.

99.
100.
101.
102.
103.
104.
105.

106.

107.
108.

161

J. D. Reinheimer, J. P. Douglas, H. Leister, and M. B. Voelkel,
J. Org. Chem.,“ZAZA3 1743 (1957).

 

E. H. Nollau and L. C. Daniels, J. Amer. Chem. Soc., 396 1885 (1914).

 

see reference 16, p 498.

P. E. Wolf and K. Folkers, Org. React., Q, 410 (1951).

 

C. A. Fyfe, M. I. Foreman and R. Foster, Tetrahedron Lett., 1521 (1969).

 

G. Illuminati, G. Dodd, and F. Stegel, Chem Commun., 953 (1969).

 

 

C. D. Hurd and K. L. Kreuz, J. Amer. Chem. Soc., 72, 5543 (1950).
N. P. Buu-Hoi and D. Lavit, J. Chem. Soc., 1721 (1958).

 

W. 0. Foye, J. J. Hefferren, and E. G. Feldman, J. Amer. Chem. Soc., 76,
1378 (1954). mm

 

 

G. Giver, J. Amer. Chem. Soc., 77, 577 (1955).
P. Fournari and J. P. Chane, Bull. Soc. Chim. Fr., 479 (1963).

 

H. H. Jaffe and H. L. Jones in "Advances in Heterocyclic Chemistry,"
Vol. 3, A. R. Katritzky, Ed., Academic Press, New York, N. Y., 1964,
p 209, and references therein.

C. Mack, "Essentials of Statistics," Plenum Press, New York, N. Y.,
1967, p 118.

H. H. Jaffe, Chem Rev., 53, 191 (1953).
mm
see reference 97, p 240.

D. H. McDaniel and H. C. Brown, J. Org. Chem., 23, 420 (1958).
’\/\a

 

R. D. Schuetz and D. M. Teller, J. Org, Chem., 37, 410 (1962).

 

S. Gronowitz and R. A. Hoffman, Ark. Kemi, 16, 539 (1960).
P. A. TenThije and M. J. Janssen, Rec. Trav. Chem. Pays-Bas, 84, 1169 (1965).

 

A. Albert and E. P. Serjeant, "Ionization Constants of Acids and Bases,"
John Wiley and Sons, Inc., New York, N. Y., 1962, Chapter 2.

H. A. Laitinen, "Chemical Analysis," McGraw-Hill, Inc., New York, N. Y.,
1960, p 86.

see reference 105, p 124.

S. W. Benson, "The Foundations of Chemical Kinetics," McGraw-Hill, Inc.,
New York, N. Y., 1960, Chapter 2.

109.
110.

111.

112.
113.
114.
115.
116.
117.

118.

119.

120.

121.

122.
123.

124.
125.

126.
127.
128.
129.

162

H. Hartmann, Z. Phys. Chem. (Leipzig),,6191, 197 (1942).

 

F. Hovorka, R. A. Schaefer and D. Dreisbach, J. Amer. Chem. Soc., 58,
2264 (1936).

 

a, J. E. Leffler and E. Grunwald, "Rates and Equilibria of Organic
Reactions," John Wiley and Sons, Inc., New York, N. Y., 1963, p 71;
b, Chapter 9.

G. B. Barlin and D. D. Perrin, Quart. Rev. (London), 20, 75 (1966).

 

see reference 108, p 91.
see reference 37, p 379.
see reference 111, p 324.
see reference 42, pp 89-91.

C. K. Ingold, "Annual Reports on the Progress of Chemistry," The Chemical
Society, London, 1928, p 24.

 

F. F. Blicke and M. M. Tsao, J. Amer. Chem. SOC-.,gée 1645 (1944).

. A. Olah, C. U. Pittman, and M. C. R. Symons in "Carbonium Ions,"
. A. Olah and P. R. Schleyer, Eds., John Wiley and Sons, Inc., New York,
. Y., 1968, p 156.

2639

C. U. Pittman and G. A. Olah, J. Amer. Chem. Soc., , 5123 (1965). The
chemical shift for the thienyl protons Shou1d be 2. -2.571 . G. A. Olah,
Case Western Reserve University, personal communication, 1969.

 

 

N. C. Deno, J. S. Lin, J. 0. Turner, D. N. Lincoln, and R. E. Fruit,
J. Amer. Chem. SOC-9:§Za 3000 (1965).
E. Campaigne and W. L. Archer, J. Amer. Chem. SOC".$§¢ 989 (1953).

 

H. D. Hartough, "Thiophene and Its Derivatives," Interscience Publishers,
Inc., New York, N. Y., 1952, p 317.

I. Deschamps, W. J. King, and F. F. Nord, J. Org. Chem.,,Ie, 184 (1949).

 

L. F. Fieser, "Experiments in Organic Chemistry," 3rd ed, 0. C. Heath
and Co., Boston, Mass., 1957, p 289.

F. Arndt, Org. Syn. Coll., Vol. 2, p 461.
see reference 126, p 165.
see reference 123, p 318.

see reference 123, p 208.

130.

131.
132.
133.
134.
135.
136.
137.

138.
139.
140.

141.

142.
143.

144.

145.

146.

163

G. Vincent, M. S. Thesis, Michigan State University, East Lansing,
Michigan, 1966, p 12.

 

R. B. Woodward and E. C. Kornfeld, Org. Syn. Coll., Vol. 3, p 413.
’1;
see reference 123, p 317.

J. Szmuszkovicz and E. J. Modest, J. Amer. Chem. Soc., 72, 571 (1950).

 

E. Campaigne, Indiana University, personal communication, 1968.

G. Komppa and S. Weckman, J. Prakt. Chem., 138, 109 (1933).
WI;

 

see reference 123, p 213.

 

P. Brown, J. Burdon, T. J. Smith, and J. C. Tatlow, Tetrahedron, 10,
164, (1960). mm

see reference 123, p 495.

R. G. R. Bacon and 0. J. Stewart, J. Chem. Soc., 4953 (1965).

 

F. M. Behringer, H. E. Bechofner, R. A. Falk, and M. Leff, J. Amer. Chem.

 

_S_g_c_., 9p, 4279 (1958).

J. R. Crowder, E. E. Glover, M. F. Grundon, and H. X. Kaempfen, J. Chem.
S99., 4758 (1963).

see reference 123, p 375.

H. D. Hartough, "Compounds with Condensed Thiophene Rinds," Interscience
Publishers, Inc., New York, N. Y., 1954, p 145.

E. P. Dunigan, M. S. Thesis, Michigan State University, East Lansing,
Michigan, 1961, p 53.

I. M. Kolthoff and E. B. Sandell, "Textbook of Quantitative Inorganic
Analysis," 3rd ed, The McMi11an Co., New York, N. Y., 1952, p 528.

see reference 106, p 86.

 

 

 

”1111111111111111111111111111111“