.‘J ‘ 1" . -
“fit: 74135::k‘5ug '

145‘? 13:”
l I'

-‘uu':

 

i“ ‘
It.
'7.

o ‘

4"

 

. V
I l: k:
.‘3 ,l
y -‘ "
. - .
i ”f
, \II c." ‘. ,‘ i?”
. . w h... _
'. ‘31:” ' “‘31" 1‘ I‘
‘ Wu.
U
f 3‘ .
‘ ' 5:” '1‘
. .' a. ' , {wear}. . '1.» .2
. y , , ‘ . _ I o N l‘bt‘l’. g ~:.“~‘.'a‘r:‘~r.'.c‘.”
P m ,. -, . . A ‘\ “w m 4. . --...\..:;:;_
t ‘ . .- . ,, . - . . 21""! I ".‘v ““J‘ . 53.“...u.-.
.3 »' .‘. "‘fl'ml '5.» 5. .-...
‘ ‘ ~ . . ' . - . ‘ 0' H.“]' i ' ‘ l‘
r . _ ‘ 7. -... -_ .. ,_ .. .3 y ! \ }. "JIV.
-‘ ' H . . ‘v 1c \‘ ' V ' ’I’ 'l'
, "h" «'g ‘. ' "‘ “'.'m|.a' " I'I'
:, '5 ..‘ -'1,"\' 4' 't‘q-..'-“ .-‘
K A ..

I , ' J v
5 .‘ . ‘ . . - V '.‘.A. .‘k'fl‘,
‘ ' 7" v «- ,f‘l (.Jf'fiv‘ n‘ .
. 5‘7\"."'":'~"' ‘z‘ "v ‘3; .' ~ ‘6 4 ' "‘ ...n‘ v: ,:|
s 2. w ‘ 2 ~ . ’ ‘ ' . “r: Wm“
we“ “MIN" ‘ ..1 ’ In ‘ VI . . . ‘ “ ‘u‘R.¢.-..~)“~-
.t ‘l‘ .. I . ~
‘ . . .- r ,. ‘r l

"ti“llt‘fi‘
‘5. ... ‘m
. . P “:‘m l
'31“. . . A
.- u .
...-{if} . -
~ ‘ - Can’t. ‘ :1..-
.I,‘ w‘sun-‘fi >v ‘ u 3“" . ‘ :3. ‘3‘ -v.l‘§-.‘"u
-,d~;\“:‘..".:,¢f;:fi "'17.. y ‘ .L . ‘ ‘ " ..u ’ "‘zS-EHZ‘QQ
‘ .p ~ . 3. - f;““‘“‘

n
h 1
3x35";
. . .
.

.
737' . ' u. “
‘ ‘ h. ,_t 1".

 

"iii...

5 L
nu:

' ., . ‘. .. . .. . ..u.‘ . .w’
I .. ' ~ ' | ‘ . .‘ , , ~ V ‘ .' m V . , . _ v ~ ‘ A\'v‘-‘1..‘m~c _
v " w 7. .‘ “ ‘ .‘J

 

U in‘ .
Jen "

. r'

5‘ “Mm

 

....--nq1,v‘1,“-

 

“5‘43. ‘1H"‘y‘:‘:'
1122-" “fi'r'fi'”?

 

 

1149‘?!"

 

OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

Return to book drop to remove
this checkout from your record.

   

 

 

 

 

SYNTHESIS AND PHOTOPHYSICAL PROPERTIES OF
A FLUORESCENT POLYAMIDE SERIES AND
THEIR MONOMER ANALOGS: NOVEL INTRA-MOLECULAR

EXCIMER AND EXCIPLEX INTERACTIONS

By

John Akolisa Ibemesi

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1978

ABSTRACT

SYNTHESIS AND PHOTOPHYSICAL PROPERTIES OF
A FLUORESCENT POLYAMIDE SERIES AND
THEIR MONOMER ANALOGS: NOVEL INTRAMOLECULAR

EXCIMER AND EXCIPLEX INTERACTIONS

By
John Akolisa Ibemesi

The photophysical properties of a fluorescent aromatic
polyamide-series and their monomer analogs have been uti—

lized to investigate:

i) the configurational and conformational require-
ments for intramolecular excimer formation in hetero-chain
polymeric systems.

ii) The effect of chain-binding on the reactivity
of a chemical moiety.

iii). The mechanism of solvent—induced intramolecular
exciplex fluorescence in aromatic hydrocarbon-aliphatic
amine systems; and

iv) The nature of exciplex interaction with polar
molecules.

The series of polyamides was. synthesized from 2,6-
Bis(hkmethyl, methylamino)naphthalene and the diacid
chlorides, Cl-géCHzfig-Cl (n82, A, 6 and 8) by interfacial

John Akolisa Ibemesi

polycondensation. These polyamides, designated as P-2,
P-4, P-6 and P-8, possess widely-separated, doubly-bonded
naphthyl groups in the back-bone. They were characterized
as having flexible, amorphous chains, and with good solu-
bility properties. Their molecular weight was estimated to
be 20,000.

The monomer analogs: 2,6-Bis(methylamino)naphthalene
(II), 2,6-Bis(N-methyl, methylamino)naphthalene (III) and
2,6-Bis(N—dimethyl, methylamino)naphthalene (IV), were
synthesized by reacting 2,6-Bis(bromomethyl)naphthalene with
ammonia, methylamine and dimethyl amine respectively.

The model compound, 2,6-Bis(N-methyl, N-acetyl, methylamino)—
naphthalene (MC) was obtained by acetylation of III.

The room temperature absorption and emission spectra
of dilute solutions of P-2, P-u, P-6 and P-8 were measured.
Their emission spectra showed the formation of intramolecu-
lar excimers between naphthyl groups separated from one
another by as many as 20, 2N and 10C atoms. The fluores-
cence lifetimes and quantum yields were dependent on the
distance of separation between the chromophores, in a manner
that appeared to involve non-neighboring naphthyl groups
in excimeric interactions. Temperature measurements re-
vealed a small energy requirement for the conformational
transitions involved in the formation of these excimers.
Dimer fluorescence was detected at high polymer concentra-

tions.

John Akolisa Ibemesi

Addition of small amounts of triethylamine (quencher)
to dilute solutions of the polyamides and the model com-
pound, led to the quenching of monomer (naphthalene) fluores-
cence and the formation of intermolecular exciplex. A
kinetic analysis of monomer fluorescence quenching was
done, which considered the formation of fluorescent exciplex
and a transitory non-fluorescent charge-transfer (CT)
complex. A plot of the ratio of intensities of exciplex
and monomer emissions, IE/IM, as a function of quencher
concentration, showed the model compound to be more re-
active than the polyamides; and among the polymers, the
order of reactivity is P-8 > P-6 z P-u > P-2. Thus, chain-
binding and chain rigidity appear to have decreased the
reactivity of a chromophore.

The room temperature absorption and emission spectra
of compounds II, III and IV were measured in solvents of
different polarities. All three compounds exhibited intra-
molecular exciplex fluorescence only in polar solvents,
with intensity in the order: IV > III >> II. In non-
polar solvents, substantial quenching of monomer fluores-
cence occurred; this was attributed to a charge-transfer-
induced intersystem crossing enhancement. Evidence for
this quenching mechanism was obtained from the total
emission properties of II, III and IV at 77°K, relative
to those of 2,6-dimethylnaphthalene. Quenching of exciplex

fluorescence in II, III and IV by water showed 'static'

John Akolisa Ibemesi

quenching to be predominant, and appeared to reflect the
specificity of the interaction of exciplex with protic
solvents.

The case of non-protic polar solvents was investigated
by adding small amounts of methyl cyanide (MeCN), N,N-
dimethylformamide (DMF), dimethylsulfoxide (DMSO) and hexa-
methylphosphoramide (HMPA) to dilute solutions of IV in
ether and methylcyclohexane (MCH). In ether, red-shifts
in exciplex emission maxima were observed, which appeared
to indicate the involvement of both specific interaction
and dielectric effect. The former was considered to
arise from the nucleophilic property (donicity) of the
polar species; while the latter was attributed to aggrega—
tion of polar species in a non-polar medium. The emergence
and growth of exciplex fluorescence of IV in MCH at low
concentrations of added polar species was considered
further evidence of the specificity of the interaction of
an excited dipolar species (e.g., exciplex) with surround-

ing dipolar molecules.

To
My Mother

(for pioneering my early education)

ACKNOWLEDGMENT

The author wishes to thank Professor Jack B. Kinsinger
for his warm guidance and support during the course of this
work. The 'special' guidance role of Professor Ashraf
M. El-Bayoumi is most gratefully acknowledged. His
stimulating discussions on all aspects of this work have
been fruitful. The author also extends his gratitude to
Professors: Eugene LeGoff and Lynn Sousa for their dis—
cussions on some aspects of this work. The technical
assistance of Mr. Dave Carr is highly appreciated. The
author is also grateful to the Department of Chemistry
for provisions towards the execution of this work. The
author is most lovingly indebted to his wife for sustain-
ing this work by her patience and encouragement, and by
her dutifulness at home amidst her own heavy academic
loads. Finally, the author wishes to note with great
admiration the early realization and support by their
little daughter, Chinelo (then aged 2-1/2 years) that

"daddy should go to the laboratory".

iii

...J
O

3.

I :1
L
m
[-1
m
0
"1.1
0
£3

3

1
i
n

 

OF TABLES.
OF FIGURES ........
INTRODUCTION.

Definition.
Historical.
Thesis Objective.

THEORIES OF EXCITED STATE MOLECULAR
COMPLEXES . . . . . . . . . . . . .

EXCIMER STATE .

Introduction.
Molecular Exciton Theory.
Charge Transfer Theory.

Excimer State - a 'Pure' Exciton
Resonance State?. .

Excimer State - a 'Pure' Charge
Resonance State?. . . . . .

Bxcimer State - a 'Mixed' Exciton
and Charge Resonance State.

EXCIPLEX STATE.

Exciplex Energy in Solution

SYNTHESIS OF POLYAMIDES AND THEIR
MONOMER ANALOGS . . .

Introduction. . . . . . . . . . . .
2,6-Bis(bromomethyl)naphthalene (3)

. 2, 6-Bis(methylamino)naphthalene (6)

2, 6- BisCN-meth l, methylamino)-

naphthalene (8 . . . . . . . . .
2,6-Bis(N—dimethyl, methylamino)-
naphthalene (10). . . . . . . . .

p.
4

(hoot-4H

39

39
U1
H2

Chapter Page

2 ,6-Bis(N-methyl, N-acetyl, methyl-

amino)naphthalene (11).. . . . . . . H9
Polyamides by Interfacial Poly-
condensation. . . . . . . . . . . . 50

Polyamides of 2,6-Bis(methyl-
amino)naphthalene (l3). . . . . . . . . . 52

Polyamides of 2,6-Bis(N-methyl,
methylamino)naphthalene (1a). . . . . . . 5A

Structural Analysis of Polyamides
of 2,6-Bis(N-methyl, methylamino)-
naphthalene (lA). . . . . . . . . . . . . 56

Molecular Weight Determination. . . . . . 60
A. INTRAMOLECULAR EXCIMER FORMATION IN
POLYAMIDES HAVING NAPHTHYL GROUPS IN
THE BACKBONE. . . . . . . . . . . .-. . . . 63
Introduction. . . . . . . . . . . . . . . 63

Absorption and Fluorescence Spectra
of Polyamides and Model Compound. . . . . 70

Kinetics of Intramolecular Excimer
Formation . . . . . . . . . . . . . . . . 76

Fluorescence Quantum Yields, Decay
Times and Rate Constants of Polyamides
and Model Compound. . . . . . . . . . . . 79

Activation Energy for Excimer
Formation in Polymer, P-2 . . . . . . . . 82

Discussion. . . . . . . . . . . . . . . . 86
Effect of Viscosity on Excimer
Formation in Polymer, P-2 . . . . . . . . 97

'DIMER' FLUORESCENCE AT HIGH POLYMER
CONCENTRATION C O O O O O O C O O O O O O O O 102

Effect of Concentration on Fluores-
cence SpeCtra Of P-20 0 a o o o o o o o o 103

Fluorescence Spectra of P-2 at
Varying Excitation Wavelengths. . . . . . 105

Excitation Spectra of P-2 . . . . . . . . 108

Effect of Concentration on
Fluorescence Spectra of P-4
and MC. 0 0 O C O O O I O O O O O O O O O 110

Chapter Page

Discussion. . . . . . . . . . . . . . . 113
CONCLUSIONS . . . . . . . . . . . . . . . . 118

5. INTERMOLECULAR EXCIPLEX FORMATION IN
POLYAMIDES HAVING NAPHTHYL GROUPS IN

THE MAIN CHAIN. . . . . . . . . . . 120
Introduction. . . . . . . . . . . . . . 120
Exciplexes of Triethylamine with
P-2, P-A, P-6, P-8 and MC . . . . . . . 129

Kinetics of Intermolecular
Fluorescence Quenching and
Exciplex Formation. . . . . . . . . . . 136

Discussion. . . . . . . . . . . . . . . 141

6. SOLVENT-INDUCED INTRAMOLECULAR EXCIPLEX
FLUORESCENCE IN 2, 6-BIS(METHYL AMINO)-

NAPHTHALENES. . . . . . . . . . . 145
Introduction. . . . . . . . . . . . . . 1A5
Absorption and Emission Properties
of 2,6-Bis(methylamino)naphthalenes . . 1A8
Discussion. . . . . . . . . . . . . . . 161
Viscosity Dependence of Exciplex
Emission in III and IV. . . . . . . . . 171
Quenching of Exciplex Fluorescence
in II-IV by Water . . . . . . . . . . . 175
Conclusion. . . . . . . . . . . . . . . 183

7. INTERACTION OF AMINE-HYDROCARBON

EXCIPLEXES WITH POLAR MOLECULES . . . . . . 185

Introduction. . . . . . . . . . . . . . 185

Exciplex Fluorescence Spectra of
2,6-Bis(N-dimethyl, methylamino)-
naphthalene, IV, in Presence of

Small Amounts of Polar Molecule . . . . 193
Onset and Growth of Fluorescence

of Intramolecular Exciplex of IV. . . . 201
Discussion. . . . . . . . . . . . . . . 206

vi

Chapter Page

8. EXPERIMENTAL. . . . . . . . . . . . . . . . 212

I. SYNTHESIS . . . . . . . . . . . . . . . 212
General Procedure . . . . . . . . . . . 212
2,6-Bis(bromomethyl)naphthalene (3) . . 212
2 ,6-Bis(methylamino)naphthalene (6) . . 21A
2, 6- Bis(N-methyl, methylamino)-
naphthalene (8) . . . . . . . .'. . . . 215
2, 6- Bis(N-dimethy1, methylamino)-
naphthalene (10). . . . . . . . . . 216
2,6-Bis(N-methyl, N-acetyl, methyl-
amino)naphthalene (11).. . . . . . . . 217
Poly(2, 6-dimethylene naphthalene,

N ,N'-dimethyl succinimide),P2-14 . . . 218

Poly(2,6-dimethy1enenaphtha1ene, N,N'-
dimethyladipamide), Pfl-lu . . . . . . . 220

Poly(2, 6-dimethylenenaphthalene, N, N'-
dimethylsuberamide),.P P6-lu. . . . . . . 221

Poly(2,6-dimethylenenaphthalene, N,"-
dimethylsebacamide), P8-1A. . . . . . . 222

Attempted Synthesis of Poly(2,6-
dimethylenenaphthalene, N,N'-di-

methylglutaramide), P3-lu . . . . . . . 223

II. SOLVENTS FOR LUMINESCENCE STUDIES . . . 223
III. SPECTRAL MEASUREMENTS . . . . . . . . . 225
Absorption Spectra. . . . . . . . . . . 225
Emission Spectra. . . . . . . . . . . . 225
Fluorescence Quantum Yields . . . . . . 225

IV. FLUORESCENCE LIFETIME MEASURE-

MENTS . . . . . . . . . . . . . . . 228
APPENDIX. . . . . . . . . . . . . . . . . . . . 236
BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . 237

vii

LIST OF TABLES

Table Page

1.1 Classification of Molecular Complexes

of Aromatic Hydrocarbons . . . . . . . . . 7
A.1 Fluorescence Quantum Yields (0),

Decay Times (T) and Rate Constants

(k) of Polyamides and Model Com-

6

pound - (l x 10- M in CHCl /EtOH,

3
1:3) . . . . . . . . . . . . . . . . . . . 81
5.1 Photophysical Parameters for
Exciplex Formation . . . . . . . . . . . . 13“
6.1 Effect of N-methylation on Emis-
sion Properties of 2,6—Bis-
(methylamino)naphthalenes. . . . . . . . . 153
6.2 Exciplex Fluorescence Quantum

Yields and Emission Maxima in
IV in Different Solvents . . . . . . . . . 158

viii

IU

[U

(A)
0

Figure

2“

(A)

l"
I i
(I)
. J
O
"U
8‘]
I 1
C.)
C":
'u
[11
(I)

Page

Potential Energy diagram for excimer

Potential energy curves correspond-

ing to the charge transfer excited

state and ground state of CT

complexes. . . . . . . . . . . . . . . . . 20
Forster's36 Exciton model of excimer

formation in aromatic hydrocarbons . . . . 23
Molecular orbital scheme for inter-

pretation of complex emission. . . . . . . 30
NMR Spectra: a) 2,6-dimethylnaphtha1ene

(l), (b) 2,6-Bis(methylamino)naphtha-

lene (6), (c) 2,6-Bis(N-methyl, methyl-
amino)naphthalene (8), (d) 2,6-Bis(N-di—
methyl, methylamino)naphthalene (10) . . . A“
NMR Spectra: a) 2,6-Bis(N-methy1,
methylamino)naphthalene, b) (a) -

deuterated, c) 2,6-Bis(N-methyl,

N-acetyl, methylamino)naphthalene. . . . . U7

Figure Page

3.3 NMR Spectra: a),2,6-Bis(N-methy1, N-

acetyl, methylamino)naphthalene, b)

P2-1u, c) Pu—lu, d) P6-1u, e) P8—1u. . . . 58
3.” Determination of intrinsic viscosity

of polymer P2-1A . . . . . . . . . . . . . 62
“.1 Room temperature absorption spectra:

a) Monomer model compound, b) P-2

A

(1 x 10‘ M in CHClB/EtOH, 1:3)

lex 290 nm e o o o o o o o o o o o o o o o 71

u.2 Room temperature fluorescence

 

spectra: Model compound,
---- P-2, -.-.- P-u, ..... .- P-6,
.... P-8 (1 x 10"!4 M/L in CHC13/

EtOH, 1:3), X 290 nm . . . . . . . . . . 72

ex
h.3 Fluorescence decay curves of P—A:

Top-monomer (Aem 330 nm); bottom-

excimer (Aem 425 nm) . . . . . . . . . . . 80
A.“ Effect of atomic separation (n),

between chain chromophores on

intramolecular excimer fluorescence:

X)¢FD,.)ID/IM............o. 83

A.5 Temperature Dependence of intra-

molecular excimer emission in polymer,

P-2: -——- 21.5, ---- -16.5, .... -62.3,
—------ -111°C. Solvent: chloroform-
ethanol (1:3); Aex 290 nm. . . . . . . . . 85

X

Figure Page

“.6 Temperature dependence of ID/IM

ratio for P-2 (1 x 10'6 M/L in

CHCl3/EtOH, 1:3) . . . . . . . . . . . . . 87
“.7 Fluorescence depolarization: Top,
P-2, bottom, P—8 . . . . . . . . . . . . . 90

“.8 Non-neighboring and neighboring

group interactions, a) and b)

respectively. Solvent effect on

polymer conformation: compact coil in

poor solvent (a and b), expanded coil

in good solvent (0). . . . . . . . . . . . 92
“.9 Effect of solvent on excimer fluores-

cence of P-“: -———-CHCl3/EtOH (1:3),

----CH2C12, .... CHClB, A 290 nm . . . . 9“

ex
“.10 Fluorescence spectra of P-2 in CHC13/
EtOH/glycerol mixtures. Percent

glycerol by volume: 0, ----- 10,

 

.... 25, —--- “5. Aex 290 nm. . ... . . . 99
“.11 The ratio of intensities of monomer

and excimer bands, IM/ID of P-2 as

a function of solvent viscosity. . . . . . 100
“.12 Concentration dependence of ratio

or intensities of excimer and

monomer fluorescence, ID/IM, in

dilute solutions of P-2. . . . . . . . . . 10“

xi

Figure] Page

“.13 Fluorescence spectra of P-2 in CHc13/

EtOH (1:1) at varying concentrations

(moles/litre): — 1 x 10‘3, --- 2 x 10'3,

————— 1 x 10'2, ... 1 x 10‘1. 18x 290 nm. 106
“.1“ Room temperature fluorescence spectra of

P-2 at varying excitation wavelengths.

a) 2 x 10'” M/L, b) 2 x 10"2 M/L; 18x

(nm): -—-32“, -- 31“, - - - 30“,

... 29“. Solvent: CHCl3EtOH (1:1) . . . . 107
“.15 Excitation spectra of P-2: a) 2 x 10"2

M/L, b) 2 x 10f“ M/L. 16m (nm): — 362,

--- 380; solvent: CHCl3/EtOH (1:1) . . . . 109
“.16 Comparison of fluorescence spectra of

concentrated solutions (2.5 x 10'"2 M/L)

of polyamides (P-2, P-“) and model com-

pound (MC): -——-MC (Aex 32“ nm), —-- P-“

(Rex 32“ nm) ... P-“ (Aex 290 nm), -----

P-2 (A 290 nm) . . . . . . . . . . . . . 111

ex

“.17 Dependence of ratio of intensities of
excimer and monomer fluorescence, ID/IM
on excitation wavelength: 0 P-2 (1 x
10"2 M/L), o P-“ (1 x 10‘2 M/L) I P-“

(2.5 x 10'2 M/L o P-2 (2 x 10'2

M/L) . . . . . . . . . . . . . . . . . . . 112

xii

Figure

5.3

5.“

5.5

6.1

Room temperature fluorescence

spectrum of P-“, containing various

concentrations of TEA as donor. [TEA]:

-——— 0.0 M, --- 0.01 M, ----- 0.02 M,

-----~— 0.0“ M, ... 0.06 M. A 290

nm. Solvent: CH2C12

Room temperature fluorescence spectra

of P-“, containing various concentra-

tions of TEA as donor [TEA]: ——— 0.09 M,
.. 0.11 M, --- 0.19 M, ----- 0.031 M.

Ratio of exciplex to monomer fluores-

cence intensity as a function of

donor concentration. Acceptor: «3) MC,

(I) P-2, (x) P-“, (Q) P-6, (A) P-8

Ratio of monomer fluorescence in-

tensity in the absence (Io) and in

the presence (I) of the donor at

various concentrations.. Acceptor:

(0) MC, (I) P-2, (x) P-“, (O) P-6,

(A) P-8. . . . . . . . . . . . . . . . .

Ratio of exciplex to monomer fluores-

cence intensity as a function of chain

length. [TEA]: (I) 0.06 M, (.)0.09 M.

Room temperature absorption spectra:

——— I, ... II, --- III, ----- IV

xiii

Page

130

131

133

135

l“3

0"]

6.

6.

3

5

6.6

6

.7

6.

8

' U

(n
09

(D

.11 _ _ ,
(1 x 10 M/L) in 2:02, 1

...:
C?
\O

6
ex 2,0 nm

Room temperature absorption spectra
-99 J

of 17 in different solvents: ——— 3-

methylpentane, ... methylene chloride,

.4
K”
...J

--- acetonitrile, ----- methyl alcohol
Enhancement of intra-molecular
exciplex emission on N-methylation:
——— de I, ... de II, --- de III,
----- de IV. Aex 290 nm, Solvent:
ethyl alcohol. . . . . . . . . . . . . . . 152
Fluorescence spectra of compound IV

in solvents of different polarities.

1) 3 Me-pentane, 2) Iso-propyl

alcohol, 3) methanol, “) methylene

chloride . . . . . . . . . . . . . . . . . 155
Exciplex fluorescence quantum yield,

c5, as a function of solvent di-

electric constant, e . . . . . . . . . . . 156
Determination of dipole moment of

exciplex IV. . . . . . . . . . . . . . . . 157
Total emission spectra at 77°K in

3-methyl pentane: ——— I, --- II,

III, ----- IV, Xe

...:
O'\
C)

x 290 nm. . . . . . .

Room temperature absorption spectra:

1) I, 2) III, 3) III + 90% H90, A) III + 1

Figure

6.9

6.10

6.11

6.12

6.13

6.1“

6.15

Page

drop of CF3COOH. Aex 290 nm . . . . . . . 162
Plot of maximum of exciplex emission

for IV against maximum of exciplex

emission for B—NMA . . . . . . . . . . . . 165
Effect of viscosity on fluorescence

spectra of Compound IV. Solvent: Ethanol/
glycerol mixture. % glycerol: 1) 0,

2) S. 3) “o, u) 80; 1 290 nm . . . . . . 172

ex
Plot of monomer/excimer intensity, IM/IE,

ratio versus viscosity of medium. Com-

pound IV in ethanol/glycerol mixture . . . 17“

Effect of water on fluorescence emission

of compound IV. % H20 (volume): 1) 0, 2)

5, 3) 15, 4) 50, 5) 85- Aex 290 nm,

solvent: ethanol. . . . 1 . . . . . . . . 177
Quenching of exciplex emission by

water; Stern-Volmer plot. (I) de II,

(0) de III, (A) de IV. )‘ex 290 nm,

solvent: ethyl alcohol. . . . . . . . . . 178

Water-induced enhancement of monomer

emission in IV. Plot of IM/Ifi versus

% H2O (volume) in ethanol. . . . . . . . . 180

Water-induced enhancement of monomer

emission in III. Plot of IM/Ifi versus
% H20 (volume) in ethanol. . . . . . . . . 181

XV

Figure

7.1

7.3

7.“

7.5

Page

Fluorescence spectra of IV in ether,

in the presence of varying amounts

of DMSO (moles/litre): -—— 0.0 (a),

0.11 (b), 0.31 (c); --- 0.02 (a),

0.16 (b), 0.56 (c); ----- 0.0“ (a),

0.21 (b), 0.72 (c); ------- 0.07 (a),

0.26 (b), 0.9“ (c); .... 0.09 (a),

0.31 (b), 1.18 (c). 16x 290 nm. . . . . . 195
Fluorescence spectra of IV in ether

in the presence of the same amount

(0.05 M/L) of different polar mole-

cules: ——— HMPA, --- DMF, ... MeCN,

-.-.- DMSO . . . . . . . . . . . . . . . . 196
Exciplex spectral shift as a func-

tion of concentration of polar mole-

cule: (O) DMF, (I) HMPA, (I) DMSO,

(*) MeCN . . . . . . . . . . . . . . . . . 197
Exciplex spectral shift as a func-

tion of donor number. Concentration

of polar molecule (moles/litre):
(0)0.05,(A)0.29............ 199
Exciplex spectral shift as a func-

tion of dielectric constant of

medium: (I) HMPA, (O) DMF, (I)

DMSO, (*) MeCN . . . . . . . . . . . . . . 200

xvi

“\l

Page

Onset and growth of fluorescence
of intramolecular exciplex in IV
-“ .

(l x 10 M/L in Me-cyclonexane)

in the presence of varying amounts

of DMF (moles/litre): -—— (0), --—
(0.07), - --- (0.1u), ... (o.u9)
Aex 290 nm . . . . . . . . . . . . . . . . 203

Ratio of intensities of monomer

and exciplex fluorescence as a

function of concentration of polar

molecules. (0) DMF, (*) MeCN . . . . . . . 20“
Ratio of intensities of monomer

and exciplex fluorescence as a

function of bulk di-electric

constant . . . . . . . . . . . . . . . . . 205
Block diagram for time resolved
spectrophotometer. . . . . . . . . . . . . 229
Operation of the TAC . . . . . . . . . . . 231
Fluorescence life time analysis

for MC (2 x 10‘“

M in CHC13/EtOH,
1:3): x (means experimental point),

0 (means calculated point), = (means

UL)
\J'l

experimental and calculated match) . . . . 21

xvii

Table

1.1

5.1

6.1

6.2

LIST OF TABLES

Classification of Molecular Complexes

of Aromatic Hydrocarbons

Fluorescence Quantum Yields (0),

Decay Times (I) and Rate Constants

(k) of Polyamides and Model Com-
pound - (1 x 10-6
1:3) . . . . . . . . . . . .
Photophysical Parameters for
Exciplex Formation .

Effect of N-methylation on Emis—
sion Properties of 2,6-Bis-
(methylamino)naphthalenes. . . .
Exciplex Fluorescence Quantum

Yields and Emission Maxima in

IV in Different Solvents

viii

M in CHCl3/EtOH,

Page

81

13“

153

158

CHAPTER 1

INTRODUCTION

Definition

The chemical characters in this scientific "drama"
(dissertation) are called excimers and exciplexes. These
are defined as: "transient atomic or molecular complexes,
with associated excited electronic states, dissociative
ground states and structureless emission spectra." Their

formation can be illustrated as follows:

I a a ‘1'
(AA): D A A*+D D D;(AD)

Excimer Exciplex

Thus, an encounter between an excited chemical moiety, A*
and an identical or different ground state species leads
to an excimer or exciplex respectively. In general, a
singlet or triplet excited complex can be formed, depend-
ing on the spin state of A’. The spectral characteristics
of these complexes, implicit in the above definition, are

depicted in Figure 1.1, which is a potential energy diagram

Figure 1.1. Potential Energy Diagram’ for Excimer Formation.

 

 
  

 
     

 

 

 

 

 

 

r-“m'm'- A+A
I A;

1 by;

>

O

‘6

C

I.“ 4'
L ........... ‘ A+A

 

Interplanor distance i
I

   
  

Fluorescence

intensity

  

   

I
{’E

 

y—

1

LA)

for the case of excimer formation.1 The excimer emission,
bug is structureless because its ground state is repulsive
(dissociative) while the monomer emission, hv; is struc-
tured because of electronic transitions to various vibra-

tional levels.

The phenomenon of excimer formation was first discover-
ed by Lord Rayleigh2 (1927) when he observed in the ultra-
violet emission spectra of high-pressure mercury vapor, a
broad continuum, superimposed on the partially quenched
atomic line spectrum. The continuous emission was attributed
to the excited triplet excimer of mercury - 3Hg’2**. There-
after (1930), Hopfield3 observed the first singlet excimer
emission from an excited helium dimer - He2. Other noble
gases behave similarly; in fact, the ultraviolet emission
continuum from a high-pressure Xenon (Xe) lamp (a widely
used light source in optical spectroscopy) originates from
the xenon excimer Xez.

For a while, excimer formation remained an event
observable only in the gaseous phase of atomic systems.

This situation was drastically altered in 195“ when Farster

and Kasper,u’5 working with pyrene solutions, observed that,

as the solution concentration increased, a decrease in the

structured molecular fluorescence of pyrene occurred,

followed by the appearance of a broad-structureless band

at lower energies than the quenched molecular fluorescence.
Since the absorption spectra of concentrated pyrene solu-
tions shcwed no new band, they attributed the above-mentioned
"anomalous" fluorescence to emission from an excited dimeric
pyrene complex produced by the association of one singlet
excited molecule and one singlet ground state molecule.

A few years later (1958), similar observations were reported

6

and for 2,5-diphenyloxazole.7 The

.-1
break-through came in 1962 with benzene and naphthalene8 ‘0

for 9-alkylanthracenes

followed by a host of other aromatic hydrocarbons and their

1 1
derivatives%"‘2

Thus, within a decade, excimer fluorescence
in fluid solutions of aromatic systems was established, from
an exception, to a common observance} Similar observations
have been made in other condensed-phaSes of aromatic mole-
cules such as in pure liquids and molecular crystals.13’l’4
Another milestone in the discovery of excimers occurred
in 1965 when Hirayamals published the first reports on intra-
molecular excimers in diphenyl and triphenyl alkanes; this
was closely followed by the paper on paracyclophanes and
vinyl polymers by Vala, et a1.16 Earlier,in 1962, Basilel7
had published the first fluorescence spectrum of pure
polystyrene; and in 1963, the first suggestion of excimer
formation (intramolecular) in dilute solution of isotactic
polystyrene was made by Yanari et a1.18 By binding the two
interacting aromatic groups in the same molecule by means

of a molecular chain, the probability of their encountering

each other within the life time of the excited state becomes
enhanced; thus, even in very dilute solutions, intra-molecular
excimer fluorescence is observable.

The history of exciplexes started in 1963 in an in-
vestigation by Leonhardt and Weller19 of the fluorescence
quenching of perylene by typical electron donors (D) like
anilines. They observed a broad, structureless emission
band about 5000 cm"1 to the red of the normal structured
fluorescence of the aromatic hydrocarbon (A). The new
fluorescence increased in intensity with increasing donor
concentration, at the expense of perylene fluorescence,
while the absorption spectra remained unchanged. They
therefore attributed the anomalous emission to an excited
molecular complex species (A14§[L which was stable in the
excited state but unstable in the ground state. Such a
complex, sometimes called "heteroexcimer" has been observed
in numerous donor-acceptor systems.2o'23 The intramolecular
case was first reported by Chandross and Thomas (1971)2u
in naphthylalkylamines.

There is another class of excited state molecular
complexes called mixed-excimers which are exciplexes formed
between pairs of molecules of similar but different species,
such as molecules with a common aromatic core. Birks and
Christophorouzs’26 (1962) first observed a mixed-excimer
fluorescence from systems like pyrene/l-methylpyrene. Sev-

eral such systems have been reported to give this type of

(n

fluorescence.27’28 The area of molecular complex formation
involving aromatic systems has grown so much in depth and
scope that no attempt will be made here to give a complete
litany of them. Instead some broad classification will be
given. This is necessary because, in addition to the ex-
cited state molecular complexes mentioned in the foregoing
paragraphs, aromatic hydrocarbons and their derivatives

are known to form stable ground state molecular complexes
(both inter and intramolecularly). These ground state
complexes may exhibit their unique absorption and emission
bands. This situation has tended to create some confusion
in the naming of molecular complexes. We will therefore
make some classification of these complexes, not only with a
view to presenting the acceptable descriptions for them, but
also to identify the types that will be encountered in

this work. Complex types, modes of formation and general
characteristics are given in Table 1.1. In this work, we
will be concerned with singlet state excimers and singlet
state AD exciplexes - henceforth referred to as "exciplexes".

Specifically, we will deal with naphthalene excimer and

naphthalene-amine exciplex.

Thesis Objectives

Experimental and theoretical studies on intermolecular

excimer and exciplex formation have shown that the most

.Aa do msfip

Im>Hmoo ocmmo ..w.ov pondooom copuooao cm mo AocHEm afipmndfiam ..w.ov mocoo somuooao
cm mmnpfio MM 0 mmm ”<1Mo o>Hpm>apoo HzHHm no meam ma .< «conmmoopozn oHmeomm ma <

 

.mcoapsaom
ofisvaa CH coamman 0: Congo momma
coauanomow pmomn mopmum ocsopw oHnmpm

.onQHoxo Q¢ CH mm osmm

.Amopompmco
mommcmmu owmmno ou msov o>HuHmcom
upco>aommmcoaudmomnm wcHocommopmoo
o: ”COAmmHEo mmoaopSuosppm .omopm

.om.mmocoomLSpcme ..w.o moefioxo
cm Eomm memom zufiuco HmoHEono <

.cOHpmppcoocoo spa: mmmomo
Isa .mmoEHo op odompsofimppm mmpooam
coaudmomnm moumpm pcsomm oaompm

.o>oom we mean
.o>HpHmcomcH uco>aom mafia

Imocow com .:0fipdmomnm wcfioCOQmompoo
o: chHmmHEo mmoammsuosmum .omomm

.i+a-¢o + A+a-<v

aAIQ+¢v + Q+a<

* A +Ql¢v AI Q+$¢

me + .Aaao

*A¢<V am A<¢V
*A.¢<v + .<+*<

3A<¢v + ¢+a<

onQEoo

A<omv poudmoom
mocoonconpooam

xoadfioxm <Q

menHoxm me

pmefiuuouozm

pmer ompfioxm

moEHoxm ooxaz

mosaoxm

 

moapmamopommmno

coameAom mo opoz

ooze xoadsoo

 

 

.mCOQEMoomomm oaumsom¢ mo moondEoo amasooaoz mo coameHmHmmmHo .H.H manna

probable conformation of these complexes insolutions of
aromatic compounds is one in which the two components are
in parallel-stacked face-to-face (sandwich) arrangement with
a distance of 2.5-3.5 A between them.3l"33 Indirect evidence
for this conformation comes from the observance of excimer
fluorescence from crystalline aromatic hydrocarbons in
which the rings are oriented in a sandwich fashion. How-
ever, when the two interacting groups are bound in the same
molecule, steric and entropy restrictions are introduced
which strongly control the attainment of the favored sand-
wich conformation within the life time of the singlet ex-
cited state, which is on the order of 10'9 seconds.
thus, the formation of intramolecular excimers and exciplexes
is subject to certain geometrical requirements, the study
of which has, not only thrown more light on the intrinsic
nature of these complexes, but also has served as a reliable
method of understanding configuration and conformation in
chemical species in general, and in polymers in particular.
In vinyl polymers (for example) with pendant fluores-
cent aromatic groups, a molecular configuration, involving
a three-carbon-atom separation between the aromatic moieties,
has been found optimum (n83 rule)15 for intra-molecular
excimer and exciplex formation. For a while, this rule
seemed to have wide—spread validity in carbon-carbon chain
composite and polymeric systems. However, exceptions soon

began to surface, first for the case of intramolecular

exciplex formation. Ide et a13“ and Okada et al.35 (1972),
reported the occurrence, in polar solvents, of intramolecular
cular exciplex emission in aromatic hydrocarbon—aromatic
amine systems, in which the donor and acceptor groups are
separated by one methylene group (n81). This was attributed
to increased complex stabilization via solvation, which

seems to have overriden the geometric restrictions.

In 1973, Hayashi and Tazuke36

searched unsuccessfully,
for intra—molecular excimer emission in polyamides containing
widely separated carbazole units in the backbone (n16).
To our knowledge, this was the only reported work for the
case of heterochain polymers at the inception of this work.
During the course of this work, Tazuke and Banba (l976)37
reported the occurrence of intramolecular excimer formation
in polyesters having pendant l-naphthylmethyl groups sepa-
rated by 16 carbon atoms and 2 oxygen atoms. Consequently,
it seems obvious that the issue of geometric requirements
for the formation of fluorescent intramolecular excimers
and exciplexes remains an unsettled one. This work is
intended to address this issue and other related excimer-
exciplex interactions along the following lines:

1. To synthesise and characterize a series of fluores-
cent polyamides with naphthyl groups in the backbone, with
a View to investigate the possibility of intramolecular
excimmr'formation between widely separated naphthyl groups

(n>6),

10

2. To study the "polymer effect" on the reactivity of
a molecular species (e.g., naphthalene) in intermolecular
exciplex formation, involving the above polyamides, their
monomer model compound and triethylamine.

3. To study the effects of solvent on intramolecular
exciplex, formed in the monomer and related compounds, be-
tween the naphthyl group and the amino group separated by
one methylene group (n81).

“. To re-examine the controversial subject of exciplex
interaction with small polar molecules, in order to shed

more light on the mechanism involved in the interaction.

CHAPTER 2.

THEORIES OF EXCITED STATE MOLECULAR COMPLEXES
EXCIMER STATE

Introduction

Experimentally, excimer formation has become a fre-
quently occurring phenomenon in many aromatic systems in
different environments. Its emission indicates a stable
excited state and an unstable ground state; and its occur-
rence in aromatic crystals in which the rings are oriented
in a sandwich fashion indirectly suggests a similar type
of configuration for excimer in solution. These and other
experimental findings formed the basis for theoretical
investigations on the nature of the excimer state. Some
of the fundamental questions are: what is the preferred
steric configuration of the aromatic entities in the excimer?
What type of forces hold them together? and what is the
nature of the transient electronic structure?

Forster (l962)38’39 was the first to address these
issues theoretically. He proposed that the excimer fluores-
cence occurred from the lowest excited singlet level of a
1:1 self complex; such level corresponds to the lowest-
energy exciton component of the 1La state of the aromatic

hydrocarbon. Ferguson (1958)"0 suggested that the excimer

11

12

fluorescence could occur from a low energy charge resonance
state of the dimer. Azumi and McGlynn (196“)!41 showed

that neither the molecular exciton approach nor the charge

. resonance approach could lead to a satisfactory interpre-
tation of excimer energies. KoniJnenberg (1963)”2 introduced
the configuration interaction model, in which the singlet
excimer state is considered as a "mixed" state, produced

by configuration interaction of charge-resonance and exciton-
resonance states. This approach was subsequently adopted

.“3,““ Thus, the general wave function,

by other workers
V for the electronic states of the excimer entity can be

written as:

z i *
W Zai¢1(A B) + Xbi¢i(AB ) + XCi¢i(A B ) + 2di¢i(A B )

 

 

(2.1)

 

Exciton Resonance States Charge Resonance States

Configuration Interaction States

In the following sections, we will describe in some
detail, the three models for the excimer state: exciton
resonance state, charge resonance state and configuration
interaction state. This will be followed by a discussion
on the extent to which each model or any combination of

them, correctly accounts for the excimer binding energy.

.4
(.10

Molecular Exciton Theory

The concept of the exciton was first introduced by
J. Frenkel (1931)"'5 to account for the dissipation into heat,
of excitation energy in atomic crystals, following photon
absorption. Frenkel realized that electronic excitation initi-
ally localized on a particular atom would, under the influence
of the interatomic interactions, migrate to neighboring
atoms. He called this migrating packet of electronic exci-
tation energy, an "exciton". A. S. Davydov (19“8)”6 used
the molecular exciton concept to study the electronic states
of naphthalene crystals. Subsequently, many problems in the
area of molecular and solid state physics have been treated
theoretically using the exciton theory.

A very simplified form"7 of this theory can be used to
describe a self dimer consisting of an excited molecule and
its ground state counterpart. Two degenerate time dependent

wave functions W, and $2 can be written:
*
"I a ¢1¢2
*
W2 ‘ ¢1¢2

The stationary states of the dimer correspond to a linear
combination of "1 and w2. Such stationary states belong to
different irreducible representation of the dimer point

group. The first order energies can be found from the

1“

determinant

H11 ' E H12

12 22 ‘ E

where

:32
II
:1:
I

22 ' ‘¢1¢2IH'¢1¢2>

s
H12 <¢i¢2'"'¢1¢2’
(* denotes electronic excitation)
The roots of the determinant are
E' 8 H11 + H12
E" ' "11 ‘ H12
.and the eigenfunctions are

_. i
w' = 2 l/2(¢1¢2 + ¢1¢;) (2.2)

.. I i
w" . 2 1/2<¢1¢2 - ¢1¢2) (2.3)

w' and w" are the stationary exciton states where the

15

excitation energy is delocalized over the two molecules

forming the dimer. Evaluation of E' and E" yields
I: * it 9(- if
E E1”E2*“"1""2I‘712‘1’1‘1’2>+<¢’1“’2I"’12I‘1’1‘1’2> (2.14)
n: * )I a: _ it- *
E E1+E2+<¢1¢2IV12I¢1¢2> <¢1¢2|V12l¢1¢2> (2.5)

The last term in Equations (2.“) and (2.5) gives the "exci—
ton splitting",

= i6 "’
AE ‘¢1¢2'V12'¢1¢2>

V12 can be expanded into a series of multipole-multipole
interactions - that is dipole-dipole, dipole-quadrupole,
quadrupole-quadrupole etc. For electric-dipole allowed
transition, the most important term is the dipole-dipole
interaction term. Using the point dipole-dipole approxi-
mation, Vl2 can be written as:

. 3 . .
v M2 - ——2—.(M1 R)M2 R)]

l

where R is the intermolecular distance, and M1 and M2
are the transition moments of molecules 1 and 2. The exciton
splitting term can also be written in terms of trignometric

functions:

16

AE - -'Ml|2/R3(2coseicoseg-cosegcoseg-coseicoseg) (2.6)
eix,y,z) and eéx,y,z) are the angles which the transition
moments, M1 and M2 respectively make with the x, y, z axes
of the coordinate frame. The maximum exciton interaction
will occur when the two transition moments are parallel to
each other and at right angles to R ("perfect sandwich

configuration) and the exciton splitting will be given by

N

AB = - (2.7)

$11.3
to

The validity of this expression in explaining the excimer
binding energy will be examined later in the chapter.

Charge Transfer Theory

The formation of electron-donor-acceptor (EDA) molecular
complexes by stable aromatic moieties has been known for
years. One important phenomenological manifestation of such
complexes is the appearance of intense absorption bands
characteristic of the complex. Mulliken (1952)’48 developed
the "charge transfer theory" to explain the electronic struc-
ture, and hence the absorption and emission characteris-
tics of these complexes. His general theory of molecular
complex formation is as follows: For a simple case of a
1:1 complex (D-A) formed by the weak interaction (dipole-

dipole, dipole-induced dipole dispersion interactions) of

17

an electron donor (D) and an electron acceptor (A) molecules,
.the ground state wave function can be written as:

W = WO(D-A)

0
while the wave function of an ionic state (where an electron

is donated to the acceptor) can be written as:

+ .-

A and D are assumed to be neutral, closed-shell molecules.
The true ground state TN of the molecular complex is a mixed

state of W and V , and the corresponding excited state is

l 0
TE (charge transfer state)

TN = awo + bV (2.8)

1

a*V

TN and WE are orthogonal to each other, and on normalization

yield
<v Iv > = fw2dr = a2 + b2 + 2abs = 1 (2 10)
N N N 01 '

<TN|VE> = 0 (2.11)

 

18

S01 is the overlap integral between To and V1:

801 = IVOVldI

The eigenvalues corresponding to TN and WE are obtained by

solving the secular equation

(2.12)

 

 

where
E0 = <wOiH|vO>; E1 = <vl|H|wl>
H01 3 ("O'HIVI>3 H10 = (Wllle0>

For a weak complex, solution of equation (2.12) yields

 

 

2
a (H - s E )
EN E0 - 01 01 o (2 13)
E1 ’ Eo
E = E + (H01 ‘ 801E1)2
E 1 (2.1“)
E1 ' Eo

If V0 and V1 have the same symmetry and spin multiplicity

in the complex, then the smaller the energy difference

19

(El - E0) and the larger the value (H01 - 501E0)2’ the more
stable the complex becomes. The potential energy curves
corresponding to the ionic state, (D+ — A'), the nonbonding
state, D - A, the ground state and the charge transfer state
of the complex are all shown in Figure 2.1, to illustrate
the relations among E0, E1’ EN and EE' R is the inter-
molecular distance between donor and acceptor; Ip is the
ionization potential of the donor and EA is the electron
affinity of the acceptor.

An electronic transition from the ground state of the

complex, EN to its excited state, EE, would then give rise

to an absorption spectrum, with the transition energy given

 

by:
2
(a3 + a1)
hVCT = BE " EN = E]. - E0 + E (2.15)

where do = H01 - 801E0 and a1 = HOl - 801E15 El and E0

can be approximated by,

where Em is energy at infinite separation of donor and
acceptor molecules; Cl is electrostatic energy and other

weak interaction energies occurring between D+ and A" at

2O

.noonQEoo 90 on» no woman pcsomw can madam oouuoxo
mohmcnmu ownmno can on wcaocoamommoo no>nso mwmoco Hempcouom .H.m oaswam

8am. mil 2
33 m. (...-“HI

 

m

 

5.02

 

 

 

 

21

the equilibrium distance of the complex; and CO is the sum
of several interaction energies between D and A, including
electrostatic, dipole-dipole and van der Waals interaction.

Then

so that equation (2.15) becomes

_ G
hVCT - Ip - EA - C 'I' W (2.16)

where G = (“6 + QI)‘ For a weak complex, I >> G or (EA-C)

p
consequently, its energy can be written:

I - - C (2.17)

hVCT = p EA
Equation (2.17) approximates the energy of a "pure" charge
transfer (ionic) state, and its validity in accounting for

the excimer binding energy will be seen shortly.

Excimer State - A "Pure" Exciton Resonance State?

38

The exciton resonance theory was first used by Forster
to explain excimer binding energy in aromatic hydrocarbons

(e.g., naphthalene). He listed the following conditions

22

for excimer fluorescence to occur: (1) The lowest excited
singlet state of the monomer should be the lLb state, which
has a small transition moment to the ground state, 1A.
The relatively long life time of this state allows excimer
formation to occur at reasonable monomer concentration.
(2) The lLb state should be adjacent to a higher energy
lLa state, which has a large transition moment to the
ground state. The large splitting of the 1La state, in
contrast to the usually small splitting of the lLb state,
results in the lowering of one of the 1La exciton components
below the lowest 1Lb component. It is the transition from
this lowest energy lLa exciton component to the ground
state to which Fbrster attributed the sandwich type excimer
fluorescence, Figure 2.2.

The above conditions are not necessary for excimer
fluorescence. Some molecules like 9-methylanthracene and

pyrylene have 1

La as the lowest singlet excited state, and
still show excimer fluorescence. Also condition (2) is not
necessary as shown by Azumi and McGlynn, using the follow-
ing arguments:

Using the exciton splitting, aE = M2/R3, the calculated
mutual separation R, which would yield the observed frequency
difference between the 1La level and the excimer fluores-
cence in naphthalene is 1.8 A, which is physically unten-
able.

Phenanthrene exhibits excimer fluorescence, however,

ofiumEom< :H cofiumemom Loefioxm no Homo: COuHoxm

 

 

 

.mcoommoomcz:
mmm.poumpmm .m.m omswnm

 

 

 

 

 

EMS-O mmZOZOZ

4. c.
m... # z
a //.r .
II— P ”V If D]—
nu .\\\x 2 v
... p \ \ 0...,

\.

0.. \\

 

2“

according to Forster, the separation between 1La and lLb
levels in phenanthrene is so large (m5000 cm-l) that the
exciton splitting of 1La could not possibly produce a stable
excimer state.

Thus, the excimer state cannot be described as a "pure"

exciton resonance state.

Excimer State - A "Pure" Charge Resonance State?

The suggestion for the involvement of a charge resonance
state in excimer binding energy was first made by Fergu-
son (1958)“0 for the case of pyrene crystals. This was
later reiterated by Hochstrasser (1962)“9. And in 1963,
SlifkinSO extended this idea to excimers formed in solu-
tions of a number of aromatic hydroCarbons including
naphthalene.

The energy of the charge resonance state of an excimer

can be expressed as:
hv = I - E - C i A (2.19)

where A‘is the term resulting from the interaction of the
two possible degenerate charge resonance states. With the
assumption that C t A is constant, and using the values of
(Ip - EA) obtained by Hedges and Matsen,51 Azumi and Mc-
Glynn"l showed that hve varied linearly with I - EA.

D

However, the I - EA - C values were found higher than the

p

excimer fluorescence energies except at extremely low
distance of separation. For example, in naphthalene:
Ip - 5A = 8.5 eV, c m 3 eV at 3 1, giving that Ip — EA -
C = 5.5 eV. But naphthalene excimer fluorescence energy

is 3.13 eV. Consequently, the excimer state cannot be

described as a "pure" charge resonance state.

Excimer State - a "Mixed" Exciton and Charge Resonance

 

We have seen from the preceding sections that neither
the exciton resonance state nor the charge resonance state
adequately accounts for the excimer binding energy at
reasonable distances of separation between the excimer
components. It seems a logical option would be a "mixed"
state produced by configurational interaction between the
exciton resonance state and the charge resonance state;

“2

such option was suggested by Konijnenberg.‘ McGlynn et

al52 used this approach to study excimers formed by D2h
molecules such as naphthalene, anthracene and pyrene.

What follows is an outline of their treatment, using naph-
thalene as an example. The relevant molecular orbitals
(MOS) of naphthalene are sketched as shown on page 26.

In the ground state electron configuration of molecule

A, the next-to-highest and the highest filled MO's are
represented as do and d1, respectively, and the lowest

and next-to-lowest energy vacant MO's as ¢2 and ¢,,

26

 

 

 

 

 

 

"3 63
‘1’2 62
¢ 1 e
1 t t 1 o o

 

Molecule A Molecule B

respectively. The corresponding MO's of molecule B are 80

61, 02, and 63, respectively. The two lowest excited
states of molecule A are lIa'and lLb. The 1La excited state
is produced by the excitation of an electron from the highest

occupied orbital to the lowest vacant orbital, ¢l + ¢2.

1Lb excited state is a resultant of the two zeroth-

order degenerate excitations: ¢1 + ¢3 and ¢0-*¢2. The

The

lowest energy excited states contributing to the excimer
state are: charge resonance state and the two lowest exci-
ton states, due to 1La and lLb excited states of the monomer.
The wave functions of these excited states of the excimer

can be written as:

27
Molecular exciton states of 1La origin:
"A ’ i‘4’17‘51'3’18'21 ‘ !¢131E1921
I4"}3 ‘ Id’17529181! " '314’29181'

where the bar denotes opposite spin.

, - 1
Molecular exciton states of Lb origin:

"(3 = loltleotzl - |¢131F062| + l¢137191§3| ‘ ”1318153!
"UD a leltlootzl - [altltcogl + lelFl¢l$3l - |61F131¢3l
Charge resonance states:

We ’ ”1329131I ‘ I61929181l

1M; = I¢l$l¢2€ll - IchfilIgell

The above wave functions can be transformed into irreducible
representations of the D2h excimer point group by linear
combinations of the corresponding functions: Molecular
exciton (EXC) states

¢_(1La) = w, - wB : a

l .
¢+( La) 3 ”(A + IDB ' 8211

28
¢_(1Lb) = we - wD : B2

g

¢+(1Lb) = to + YD : B3u

Charge resonance (CR) states:

¢+(CR) "E + "F : B2u

These results reveal that for a D2h excimer, the 1La exciton

states transform identically with the charge-transfer states,

while the lLb exciton states transform differently from both

lLa exciton and charge-transfer states. There is non-
vanishing configuration interaction between |82u(exc)

and IBzu(CR) and between |B3g(exc) and |B3g(CR) . Con-
sequently, the electronic states of the excimer are expressed
by the superposition of the 1La exciton (EXC) states and

the charge resonance (CR) states:

W(B ) C

¢_(1La) + 02¢_(CR) (2.20)

32 l

w(52u) = C3¢+(1La) + c,¢+(ca) (2.21)

where (01’ C3) and (C2, C“) are weighting factors for the

molecular exciton states and charge resonance states,

29

respectively. Numeral calculations by McGlynn et a152

and other authors,u3’uu

unanimously show that the excimer
fluorescence state (for many aromatic hydrocarbons) is
lB3g' For naphthalene excimer, the interplanar distance
(R) at which calculated results give the observed excimer
energy is 3.2 A, which is reasonable, and compares well
with the magnitude of intermolecular separation in mole—
cular crystals exhibiting excimer emission. However, the
above R value is dependent on the chosen effective nuclear
charge, Z. In fact agreement with experiment is obtained
only at Z > 3. This situation, thus casts some doubt on

the real quantitative value of these semi-empirical tech-

niques.

Exciplex State

The singlet exciplex state has been described as a

charge transfer state, produced according tozlg’23

1A* + D + l(11‘6“) or A + 0* + 1(1113*)

where A is an electron acceptor and D is an electron
donor. A molecular orbital diagram of the emissive exci-
plex state is given in igure 2.3.53 Indirect evidence
in favor of the charge-transfer character of the exciplex
comes from flash photolysis experiments in highly polar

solvents, where transient absorption spectra due to the

30

.cofimmfiem ondEoo no soaumpomomoch pom oEonom Hopfinno amazooaoz .m.m onswfim

 

 

 

a... + (a STA. <
bu-fi _ ». b.
:14 _ Ni...

+|I II.
n
.

 

 

I
I
'§
I
I
I
I
u
I
I
u
I
u
f
l
I
I
n
I
I
I
I

31

radical anions have been observed.53

However, as the charge transfer complex state in most
cases has an energy close to that of the lowest excited
singlet states of the components, a more general descrip-
tion of the emitting exciplex state, which includes locally

excited states, is given:

U)

Q I?
1 “ad - + 1 .
Mam) ... (A D ) fad *(Aw)

The interactions of these locally excited states with the
charge—transfer state depend upon the overlap integrals Sad
between the highest bonding orbitals and 53b, between the
lowest antibonding orbitals, respectively, leading to a
stabilization of the emitting exciplex state. The wave

function of the exciplex may be approximated as:
~ - +
TB ~ Za1°1(A*D) + Zbi¢i(AD*)+zci¢i(A D ) (2.22)

where ci >> a1 a bi
The energy of the "pure" charge-transfer exciplex state

above the separated ground state components is given by:

BET = 1p - EA - c (2.23)

w are I ‘c the ionization potential of the donor

p --
e electron affinity of the acceptor and C is the coulomb

(T
db‘

a traction energy at equilibrium distance.

32

Exciplex Energy In Solutionsu

The exciplex state is dipolar in character, so that in
solution, its energy is expected to be lowered, due to
solvation. The energy of the solvated exciplex can be
determined as follows:

If the exciplex in solution is regarded as a point
dipole, m, and if dipole-dipole interactions between it
and the surrounding solvent molecules are considered, then
the solvent dipoles will orient themselves relative to the
exciplex; this will result in a decrease of the free energy
of the total system. The orientation of the solvent dipoles
will create an electric field, called the reaction field F,
at the location of the exciplex. According to Onsanger
simple continuum model, the reaction field on the solute
dipole, with a dipole moment u, situated at the center of
a sphere of radius a, and surrounded by a homogeneous di-

electric with dielectric constant, e, is given by

F = ufs (2.2“)
where
_ i;_ e—l
fa ' a3 22 + l (2‘25)

The energy of a dipole m, in an electric field F is given

by

33

E = - u-F (2.26)

-1.
Epol - 2'“ F (2.27)

Thus, the free energy of solvation of the exciplex is given

by

_ _ _ l .
E - Em + E - 2 n F (2.28)

E = - 3— --—£:l—— (2.29)

If we represent the Hamiltonian of the exciplex in the gas
phase by H°, and the eigenfunctions by W°, then in solu-
tion, the reaction field caused by solvation will act as a
perturbation, given by the Hamiltonian, H1

1 -

H - - F (2.30)

Hop

(where “o is dipole moment operator) so that the total

9
Hamiltonian H, is given by

H = H° + H1 (2.31)

3“
The wavefunctions of the perturbed states are given by
= o
I g ciwi (2.32)

n = <YuopY> (2.3“)
The solution of V iszlself-consistent field problem. In
addition to the above electronic perturbation, the equilib-
rium values of )3, of the geometric parameters, AJ of a
non-rigid exciplex, will depend on the polarity of the

solvent (6) and also on the condition

BE (A )
<—%r§-> ° = 0
J 531

for all j; where Ea is the free energy of solvation of the

exciplex.
Ea = E + Epol (2.35)
E = <¢(AJ)|H°|(AJ)+Hl(AJ)I¢|(AJ)> (2.36)
and
Epol = %u(lJ)F(lJ) (2.37)

If we consider an exciplex, in which only one locally

excited state (AD)* interacts with the CT state (A-D+),

the wave function Y can be represented by
I = Clw(A‘D+> + €2w<AD)* (2.38)

The matrix element <¢(A'D+)Iuop|W(AD)* vanishes, if inter-
molecular overlap is neglected. Also, if the dipole

moment in the state (AD)* is zero, the final dipole moment,
0 is given by

2 — + - +
u a Cl<w(A D )luoplw(A D )>= Gino (2.39)

where “o is the dipole moment of (A-D+), and under normali-

zation condition

2 2
C1 + C2 3 l (2.40)

The Hamiltonian matrix elements of the exciplex in the gas

phase are denoted by

<w(A'D+)|H°|w(A'D+)> = EC (2.“1)
<w(AD)*lH°I¢(AD)*> = Re (2.u2
<¢(A‘D )lH°|w(AD)*> = o (2.u3)

Using Equations (2.2“), (2.30) and (2.39), the matrix element

of the exciplex in solution is given by

3e
<w(A'D+)|allw(A‘D+)> = -C§uor€ (2.uu)

This approximates the influence of the solvent in terms of
changing the energy difference between the zero-order
states (A'D+) and (AD*). Thus, applying the variation
principle, and substituting equations (2.“1) - (2.““),

leads to the secular equations

n r 2 2 g = .
Vl(HC - Cl: f8 - a) + C20 0 (4.45)
Clo + C2(EO - E) = 0 (2.“6)
which can be combined to give
0.2 C12 2
(EC - E - -'—)(l ‘I' fl) = “OfE (2.87)
Ee-E (Ee-E,

where E is the energy of the resulting exciplex states.
When E is calculated, C? can be obtained from Equations

(2.“0), (2.“5) and (2.“6)

2 d2
(Ee-E)

From Equations (2.2“), (2.28) and (2.““) the polarization

energy, E stored in the solvent can be obtained,

pol

37

FC

The energy of the Frank-Condon (FC) ground state, ES to

which the exciplex transition occurs can be written
FC f

- .1
Eg - sgu - f )Epol (2.50)

where

2
fl=%a3f = nz-l,f=l‘a3f = (2.51)
" 2n +1

 

 

(n = refractive index of the solvent), and E; is the gas
phase energy of the ground state.
The frequency of the exciplex emission is given by
F0
Thus, using equations (2.2“), (2.35), (2.36) and (2.50-
2.52), and assuming that the excited dipole moment, 2,
of the emitting exciplex is independent of the solvent, then

the exciplex emission frequency becomes

2u2

3 (f - #1) (2.53)

hVE a hVE(O) '-
a
Exciplex energies calculated from Equation (2.53), may
deviate significantly from experimentally observed values,
because some of the approximations involved in its deriva-

tion are rather too simplified, if not far from reality.

or example, polar molecules are not expected to form a
spherical dome around the exciplex. Their structure may
be such that neighboring dipoles can approach the central
dipole (exciplex) to within distances smaller than 2a.

It is hard to estimate the error due to any difference

in the structure of actual molecules and Onsanger's

model. Any such departure, will however, make the true re-

(‘I

action field, F, greater than that calcula ed from Onsanger's

Also, at short distances, it seems untenable to consider
the environment of the central dipole as a continuum, since
the amount of polarizable matter contained in a continuum
will be greater (by say a factor a) than the amount con-
tained in the discontinuous environment. Consequently, the
reaction field calculated on the basis of a continuous
environment of the central dipole will be too large by
approximately the factor 6.

Finally, the neglect of specific exciplex-solvent inter-
action (which can occur in polar media) is surely not a
sound assumption; and it will be the subject of a separate

and detailed investigation in Chapter 7 of this thesis.

CHAPTER 3
SYNTHESIS OF POLYAMIDES AND THEIR MONCMER ANALOGS

Introduction

Most of the studies to date on intramolecular excimer
formation in polymers have involved olefinic type pol-

15,16,57

mers, such as polystyrene, polyvinylnaphthalene,
polyacenaphthalene, etc., which either, are commercially
available or can be easily made from readily available
monomers. On the other hand, fluorescent heterochain
polymers, apart from biopolymers,are not commonly avail-
able; the same is true of their monomers. This seems to
account for the little attention, heterochain polymers
have received in the literature36’37 to date, in studies
of geometric requirements for excimer formation.

This is in spite of the fact that heterochain (poly-
condensation) polymers have a number of structural ad-
vantages over the olefinic polymers in studying molecular
interactions in relation to polymer structure (as in
excimer formation). These are: skeletal flexibility;
ease of variation of distance of separation between chain-
incorporated fluorophores, and a high degree of accuracy
in distribution of groups in the chain. In vinyl-type
polymers, uncertainty in polymer structure, arises mostly
from the existence of a strong probability for head-to-

head and tail-to-tail addition of the monomer in the

39

“0

propagation stage.

It seems therefore rational to take advantage of
these structural features offered by heterochain (poly-
condensation) polymers, notwithstanding any concomittant
synthesis. In this chapter, we will discuss the synthesis
of: diamine monomers - 2,6-bis(methyl amino) naphthalene
and 2,6—bis(N methyl, methyl amino)naphthalene; and the
synthesis of polyamides derived from them by interfacial
pclycondensation, with a series of diacid chlorides.
These polyamides have the following structural features
- a built-in fluorescent moiety (naphthalene) in the
backbone; and with the said fluorescent group separated
from the amide linkage by a methylene group. This type of
polyamide structure offers a number of additional advan-
tages in fluorescence studies: it allows for a direct
monitoring of "micro" and macro-brownian motions of the
polymer chain by fluorescence quenching and depolarization
tion; it also prevents any significant ground-state elec-
tronic perturbation of the fluorescent aromatic group by
the amide group. We will also discuss the synthesis of a
few more compounds: 2,6-bis(N-methy1,N-acetyl, methyl
amino) naphthalene - model compound for the above poly-
amides; 2,6-bis(N4dimethyl, methyl amino)naphthalene,
synthesized to probe solvent-induced intramolecular exci-
plex emission; and finally 2,6-bis(bromomethyl)naphthalene

— the starting material for the preparation of the diamines.

“1

2,6-Bls(bromomethyl)naphthalene (3)

 

‘0
II
&H
so + ~2 ». .... J so
i / Br benzoylperoxnde 3. r
. 3 H2 \C (reflux) +
I “ o
"' 0 u
2 H2C’C\
4 NH
H2‘\t//

teo=

2,6-Bis(bromomethyl)naphthalene (3) was obtained by
refluxing a mixture of 2,6-dimethylnaphthalene (1), N-
bromosuccinimide (2) and benzoyl peroxide, in benzene.58
Under heat, and in the presence of a free radical initiator,
bromination occurs predominantly in the side chain. The
resulting product, on crystallization with a 1:3 mixture
of chloroform and tetrahydrofuran gave rod-like crystals
(slightly colored yellow). Multiple recrystallization
was therefore undertaken to get colorless crystals; yield
31.“%. This low yield could result from the formation
of a number of competitive products; for example, the
di-, and tribromo derivatives, and even some ring bromina-

tion products. The melting point of the product (182-

18“°) agrees with that reported in the literature,

142

(182-18“°)59.
The product was used in the synthesis of 2,6-bis(methy1-
amino)naphthalene (6) and its N-methyl derivatives (8 and

10).

2,6-Bis(methylamino)naphthalene (6)

This compound was prepared according to the reaction

scheme:

 

‘ ErHafiH
2..<e*h:":'> ' Z‘NHI

1......

"2““; as

"2N"2

The first step in the reaction is a straight forward
ammonolysis of the bromide (3). However, this method of
synthesis of amines has a serious disadvantage of yielding
a mixture of primary, secondary, tertiary and quarternary
amines. The formation of the last three products can be
reduced by using a large excess of ammonia, preferably

pure liquid ammonia. However, attempts to use pure liquid

“3

ammonia in sealed tubes at room temperature, led to explo—
sions. So, ethanolic solution of ammonia (large excess)
was used. The reaction mixture was refluxed at 30° with a
cold finger attached to the reaction flask. The resulting
amine salt was treated with sodium hydroxide solution to
give the free amine. Recrystallization with a 1:3 mixture
of methylene chloride and cyclohexane gave a white fluffy
powder; yield, 57.“%; mp l25-l30°.

Structural identification of product was done by nmr
and mass spectra. The mass spectrum showed a base peak
at M/e 186 (parent) and another intense peak at M/e 156
(P-CHZNHZ). Nmr spectra (Figure 3.1b) indicate the pres-
ence of the expected groups: -NH2(51.58),2>CH2(53.95) and
naphthalene (67.27-7.77). The -NH2 protons give a sharp
singlet because they are decoupled from the nitrogen and
from the adjacent::CH2 protons (due to fast exchange).

For the same reason their-CH2 peak is a singlet. This
behavior is typical of most aliphatic amines.

Product (6) was used in interfacial polycondensation
with glutaroyl, suberoyl and sebacoyl chlorides to give
polyamides, which were intended for use in fluorescence
studies. However, because they proved highly insoluble,
their N-methyl derivatives were prepared and used instead.
Further use of product (6) was in the study of intra—

molecular exciplex formation (Chapter 6).

““

 

 

 

 

 

 

 

 

 

 

 

 

l L l 1 1L L 11 I
8.0 20 6.0 5.0 4.0 3.0 2.0 L0

ppm (8) TMS

 

 

-igurc 3.1. NMR spectr : a) 2,6-Dimethyl naphthalene
(b) 2,6-Bis(methylamino)naphthalene (6), (
Bis(N-methyl, methylamino) naphthalene (8)

(1),
c)2,6-
(d) 2,6-Bis(N-dimethy1, methylamino)naphthdlene
(10).

“5

2,6—Bis(N—methyl, methylamino)naphthalene (8)

S3":
BrH N
CH3” - 2359”z 9,3
”23' (ethanol) H2NH23f
+ -
Z
~ iNaOH

.5332
2“ 9‘3
CH2NH

Compound (8) was prepared by treating 2,6-bromomethyl-
naphthalene with a large excess of methylamine, dissolved
in ethanol. The reaction mixture was refluxed at “5°,
and trapped with a cold finger. The resulting amine salt
(white) was dissolved in dilute hydrochloric acid solution;
on treating this solution with sodium hydroxide, it was
discovered that precipitation occurred in two distinct
stages - first a milky precipitate (probably tertiary and/or
quarternary amines), then followed by a silvery-white,
flake-like precipitate (identified as the product). Thus,
recovery of the free amine was accomplished by fractional
precipitation from an aqueous solution of the amine salt.

Yield, 86%; mp 52-60°.

“6

The molecular formula (C1“H18N2) of compound (8) was
confirmed by elemental analysis. Its structure was elu-
cidated from nmr and mass spectra. The latter showed
peaks at: M/e 21“ (parent, 6“%), M/e “2 (base, CH2-N=CH2)
and M/e ““ (CHZNH, intense). Nmr spectra (Figure 3.1c)

CH3

show the presence of the expected groups: :=NH(61.“0),
CH2(63.80),.:NCH3(62.“3), and naphthalene (67.22-7.72).

The absorptions of =NH and =CH2 protons have shifted slightly

upfield (cf. Figure 3.1b), due to electron donating effect

of the methyl group on nitrogen. The labile nature of

:NH proton is shown by effect of deuteration (Figure

3.2b); also its intensity is considerably enhanced by water
(Figure 3.2a - spectrum of a partially dried product),
further confirming the active nature of this hydrogen.

Product (8) was used in interfacial polycondensation with

a series of diacid chlorides to give N-methylated poly—
amides (l“), that were soluble. Also, the product was

used extensively in the study of intramolecular exciplex

formation (Chapter 6).

“7

 

(c)

 

 

(b)

 

 

 

 

(o)

 

 

 

 

 

Figure 3.2. NMR spectra: a) 2,6-Bis(N—methyl, methyl
amino)naphthalene, b) (a) - deuterated, c)
2,6-Bis(N-methy1, N-acetyl, methylamino)-
naphthalene.

“8

2,6-Bis(N-dimethyl, methyl amino)naphthalene (10)

 

 

H

I
I (CH3)2N H 1043);!!‘2 H
”Hf-”0123', (liquid) Br 60 . 30,3)?

Br

fl

2

£9

The synthesis of this compound was effected by reacting
2,6-bis(bromomethyl)naphthalene (3) with dimethylamine
(liquid) under the same reflux conditions used for the mono-
methyl derivative (8);product recovery was similarly
carried out. Yield, 85.3%; mp loo-103°.

The molecular formula (016H22N2) of compound (10) was
confirmed by elemental analysis. The mass spectrum showed
a number of expected peaks: M/e 58 (base, CH2N(CH3)2),

M/e 2“2 (parent, “1%) and M/e 198 (P-N(CH3)2, 50%). Nmr
spectra (Figure 3.1d) confirm the presence of: ::£H2(53.50),
-N(CH3)2(62.25) and naphthalene (57.23-7.72).

Product (10) was used in the study of intra-molecular

exciplex formation (Chapter 6) and in the investigation

of exciplex interaction with small polar molecules (Chapter

7).

2,6- B‘s/f—rethy_,.-ace yl, meth vlamino)nanhthalene (ll)

 

Hume”: CH3COCI "3C: N $HH2C as CH3
CH3 H217“... "HCI IEI'E'CHa
(CH3 <3
g 11

~~

The polyamide model compound (11) was made by acyla-
tion of 2,6-bis(N-methyl, methylamino)naphthalene (8)
with acetyl chloride in chloroform. An excess amount of
the diamine was used to remove hydrochloric acid (condensa-
tion by-product). Recrystallization with a mixture of
ether and chloroform gave colorless plate-like crystals;
yield, “9%, mp 202- 20“°.

The molecular formula (018H22N202) of compound (11)
was confirmed by elemental analysis. The mass spectrum
gave the following expected peaks: M/e 298 (parent, 18%),

M/e 22 5 (base, P- -HNcocr .3), M/e 255 (P-cocn “8%, and M/e

3)

226 (P-NCOCH3, 51%). Nmr spectra (Figure 3. 2c) reveal the
H3

expected groups: :EH2(5“.60, 5“. 67 ),.>NC .3(52.90, 02.95),

-COCH3(02.17) and naphthalene (67.22-7.75). The absorp-

tions of2>CHZ and=NCHq protons appear as two-component
J

50

peaks as a result of non-equivalence caused by hindered
rotation around the amide group.

Product (11) was used as the model compound for poly-
amides (14) in the following studies: intramolecular exci-
mer formation (Chapter 4) and "polymer effect" in molecular

interactions (Chapter 5).

Polyamides by Interfacial Polycondensation

60 reaction for acylat-

The well—known Schnotten—Bauman
ing amines and alcohols is the basis for interfacial poly-
condensation - an extremely versatile method, employed
here for the synthesis of polyamides of 2,6-bis(methyl-
amino)naphthalene (6) and 2,6-bis(N—methyl, methylamino)
naphthalene (8). The reaction is believed to take place
at the interface between a solution of a diacid chloride
in a water-immiscible organic solvent and a water solution
of, for example, a diamine, containing an acid acceptor.

It is a fast, low-temperature reaction, usually complete in
1-5 minutes. The reaction is also insensitive to reactant
equivalence and tolerant to the presence of many impurities
in the reactants that have only dilution effect, such as
water, or salt in the diamine or an inert liquid in the
acid chloride. However, one serious disadvantage of this
process is the hydrolysis of the acid chloride - the
shorter-chain aliphatic diacid chlorides being most affect-

ed. This could lead to low or no yield of high molecular

51

weight polycondensation products.

The procedure for mixing the reactants is to dissolve
both the diamine and an acid acceptor in water, and with
this aqueous solution vigorously stirred, the diacid solu-
tion is quickly added. However, adoption of this procedure
in our case, led to precipitation of the diamine from the
aqueous solution prior to the addition of the acid chloride.
Consequently, a separate aqueous solution of an acid accep-
tor-sodium hydroxide had to be made, and was added simul-
taneously with the acid chloride. Also, in all of the poly-
amide synthesis, excess diacid chloride was used for two
reasons: to compensate for losses of the acid chloride
due to hydrolysis, and to ensure total consumption of the
diamine, which could otherwise contaminate the polymer, and
would absorb and emit at the same wavelength region. Con-
sequently, the terminal groups in these polyamides are most
likely the respective acid chloride groups, or ethoxy groups
resulting from reaction with ethanol in the solvent-chloro-
form (a commercial sample with 0.75% ethanol as a preserva-

tive.

Polyamides of 2,6-Bis(methylamino)naphthalene (13)

$3 $3
”Wm + Cl-C-(CHz-Z-C-‘Cl
CH2NH2
6

 

 

 

.. 12
'-HCl
.2. ‘ .x. IQLWER.
"NH“ IC) ("3 3 P3-13
6 P6-I3
_ HzNH'é-6CH2-L-C _n 3 PB-Ia
13
pd

An aqueous solution of diamine (6) and a chloroform
solution of the diacid chloride: glutaroyl chloride (x=3),
suberoyl chloride (x=6) or sebacoyl chloride (x=8) were
allowed to react as described in the preceding section.
The resulting polyamides (white solids) were recovered by
filtration. Products were dialyzed for 2h hours to remove
small molecule impurities, and then vacuum-dried at 62°;
the final products were creamy white, with yields: P3-13
(30%), P6-13 (70%) and P8-l3 (53%). Their melting tem-
peratures were above 300°.

The polymers could not be dissolved in any solvent;

53

not even in strongly acid solvents (for example sulfuric
acid and trifluoroacetic) nor in phenols (like m-cresol).
These solvents are generally used to dissolve polyamides.
Consequently, some structural modification had to be made
to solubilize the polyamides. Polyamides are generally,
difficultly soluble, owing to their partial crystalline
nature. This crystallinity arises from intermolecular
hydrogen-bonding through the amide linkages. Shashoua

and Eareckson61 found that polyterephthalamides from
primary aliphatic diamines were soluble only in the strongly
acidic solvents; however, on replacing the amide hydrogen
by alkyl groups, several of the polymers became soluble

in chloroform, acetone, aqueous alcohol, 80%, and dimethyl-
fOrmamide. They also found that N-alkylation led to a sub-
stantial lowering of the polymer melt temperature. All
this, is as a consequence of the loss of interchain hydro-
gen bonding, on N-alkylation. The result - non-crystalline
(amorphous) polyamides, with remarkably good solubility
properties. This offered us a good prospect, not only

from the stand point of getting these polymers into solution
in the first place, but also, of having some flexibility in
using solvents that are suitable for fluorescence studies.
Consequently, 2,6-bis(N-methyl, methylamino)naphthalene

(8) was synthesized for making the N-methylated derivatives

of the insoluble polyamide series (13).

5h

Polyamides of 2,6-bis(N—methyl,_methylamino)naphthalene (1H)

CH3

"'N""2C 08$ 2CH3 + c1-‘c-(-CH2-)-<!—CI

J.

 

(EH3
N’ OO 0‘39: $3 M8
_. 3 “$44
“2" C+CH2+C 4 94-14
6 P6.“
14 n 8 PE-M

With the exception of P8-lu synthesis (in which carbon
tetrachloride was used in place of chloroform) the poly-
condensation reactions were performed in a chloroform/
water mixture. Use of carbon tetrachloride as a solvent,
led to precipitation of the polymer during the course of the
reaction. The solid product was colored dirty brown. This
was dissolvedin chloroform, and treatment with methanol,
did not precipitate the polymer. In the cases where the
reaction was performed in chloroform/water mixture, the
polymer went into solution in chloroform layer (generally
colored brown).. Reduction of the volume of chloroform

layer, followed by addition of methyl alcohol caused

precipitation of 92-14 and PH-lu out of solution but not
P3-lfl and P6-lh. Cyclohexane and n-hexane were tried as
precipitants; this led to brown rubbery solids, instead.
Consequently, product purification in P6-lh and P8-lh
synthesis had to be accomplished by repeatedly dispersing
the solid polymer into fine particles in methanol to dis-
solve as much of the impurities as possible - this method
yielded creamy white products. However, this approach
failed to rid P3-lu product of its dirty brown color; it
was therefore not used in the photophysical studies.

The following yields were obtained: 92-1u (aux), Pu-1u
(27%), P6-1u (2.6%) and Pe—iu (30%). These yields are
comparatively lower than those obtained with the primary
diamine (6). This is not surprising, since the secondary
diamine (8) is expected to be less reactive than the primary
diamine because of steric hinderance. The low yields could
also result from the hydrolysis of the diacid chlorides;
no products were obtained with malonyl and oxaloyl chlorides,
probably because they are highly susceptible to hydrolysis.

The melt temperatures of the polymers are: P2-lh (115-
120°), Ph-lh (193-195°), P6-lu (llS-126°) and P8-lh (123-
l25°). There is thus a substantial lowering in melting
temperature as a result of N-methylation. These polyamides
are soluble in: chloroform (extremely), ethanol/chloroform
(3:1) mixture, and methylene chloride (with warming). In
the presence of small amounts of‘a non-solvent (for ex-

ample cyclohexane) the polyamides become plastic, and

56

gradually turn into rubbery materials as the amount of
non-solvent is increased. Thus, while thermodynamically
"good" solvents dissolve the polymers with ease, non-sol-
vents act as plasticizers or flexibilizers. The implica-
tion of all this, is that the forces of cohesion between
the polymer molecules in the N-methylated polyamides are
easily diminished by solvent molecules, following the
destruction of their hydrogen bonding capability. We thus
have made a soluble amorphous polyamide series, with linear
and flexible chains - these characteristics will prove use-
ful in studying inter- and intra-molecular interactions in

solution.

Structural.Analysis of Polyamides of 2,6-bis(N-methyl,
methylaminolNaphthalene (1h)

A somewhat limited but adequate identification of these
polyamides was performed, as low yields were obtained in
most cases. Their chemical compositions and structures
were established by elemental analysis and nuclear magnetic
resonance spectra, respectively; While an estimate of
the magnitude of their molecular weight was obtained by
viscometric measurements carried out with one member of
the series.

Results of elemental analysis on P2-lu, PU-lu and
P8-lu for carbon and hydrogen, show excellent agreement

between the calculated and the found, only for hydrogen;

while sizeable deviations occur in carbon content: 3.6%
in F2-lfl and PU-lu, and 1.5% in PE-lu. The calculated
elemental composition was based on the repeating unit of
each polymer; which does not take into account the end
groups. However, t is unlikely that exclusion of two
small-molecular units per polymer molecule, would introduce
any significant error when we are dealing with molecular
weights on the order of 20,000 (see next section). And,
if the discrepancy in carbon content is due to organic
impurities, one would expect the hydrogen content to be
similarly affected. But this was not the case, which
leaves one to speculate that the experimentally-determined
carbon contents could be in error. A repeat of elemental
analysis was not undertaken because sufficient structural
evidence was obtained from nmr spectra (Figure 3.3) of

dilute solutions of the polymers in CDCl with TMS as

3,
reference.

The NMR peaks are broadened; this is generally the case

in solutions of high molecular weight polymers. The line
broadening arises from direct dipole-dipole interactions,
as a result of restricted molecular motion in polymeric
systems. The presence of the diamine (8)-monomer unit in
the polymer structure is confirmed by the absence ofI=NH
group (51.h) and the presence of the following groups:
:2CH2 (attached to ring) (dmu.65). :=N'CH3 (6N2.90) and naph-

thalene (57.22-7.7h). The respective diacid chloride units

 

58

 

 

 

 

TV

4“ ‘- I Ji
AA A
'rvv V V‘r
e

 

 

 

W a
l l I L L l l L L
8.0 20 6.0 5.0 4.0 3.0 2.0 LO 0
ppm (8) TMS
Figure 3.3. NMR spectra: a) 2,6-Bis(N—methyl, N-acetyl,

methylamino)naphthalene, b) P2-lh, c) Ph-lh,

d)

P6-lh, e) P8-lu.

59

in the polymer chain, give rise to methylene proton absorp-
tions at higher fields, identified as follows: terminal
:£H2 protons (next to carbonyl groups) absorb at lower
fields (6N2.35) than the penultimatej:CH2 protons: 61.78
(PU-14), 61.67 (P6-lu) and 61.70 (PS-l“), while the internal
>£H2 protons in P6-1u and P8-1H occur at much higher field,
6~1.A.

Ratios of peak areas are given below, starting from
low to high field absorptions, with expected values in

brackets:

P2-1u, 1.u:1:1.u:o.8 (1.5 1:1.5:1)

PN-lU, l.2:1:l.5:0.9:l(l.5:l:l.5:l:l)
P6-lu, l.5:l:l.6:0.8:l.8(l.5:l:l.5:l:2)
P8-lh, l.2:l:l:l.3:l.3:3.“(l.5:1:l.5:l:3)

Within the limits of experimental error, there is agree-
ment between the experimentally determined peak ratios and
the expected, thus confirming the chain structures estab-
lished by chemical shifts.

The spectra of P2-lu show a two-component peak at
62.77 and 62.95, assigned tozzNCH3 protons (Figure 3.3b).
A similar peak is seen in the spectra of the polymer model
compound (Figure 3.3a). The two—component nature of
both peaks is believed to arise from the non-equivalence

created by restricted rotation about the amide group.

60

Molecular Weight Determination

A relatively quick and reliable method of determining
the molecular weight of a polymer involves the use of the

Mark-Houwink expression:

[n] = KMa

where [n] is the intrinsic viscosity, M is the molecular
weight, and K and a are constants for the polymer-solvent
system at a particular temperature. Values for K and a
for many systems are available in the Polymer Handbook.62
The intrinsic viscosity is obtained by means of expres-
sions derived by Huggins and Kraemer, which give relative

viscosities as a function of concentration:

Huggins equation

nr-l/C = [n] + Klth2C (3.1)
Kraemer equation
lnnr/C = [n] - K2 [nJZC (3.2)

where nr is the relative viscosity at concentration C,
and k1 and k2 are constants. A plot of nr-l/C and/or
znnr/C against C would yield [n] as the intercept.

Relative viscosity measurements were done with polymer

61

P2-14 using a Cannon-Ubbelohde viscometer. The temperature
of the viscometer and its contents was kept constant to
within m0.02°C by means of a thermostated water bath.
Efflux time at any chosen concentration was taken about
five times to get at least three readings that agreed to
within 0.05 sec. A Kraemer plot (Figure 3.4) gave a value

of 0.55 for [n], from which a molecular weight of ~20,000

was estimated for P2-lu.

62

 

055-

0.525

['3

 

 

 

0.0L 1 l l

0-05 0.15 0.25 0.35
CONCENTRATIOMgm/dn

Figure 3.u. Determination of Intrinsic Viscosity of
Polymer P2-lh.

CHAPTER 4

INTRAMOLECULAR EXCIMER FORMATION IN

POLYAMIDES HAVING NAPHTHYL GROUPS IN THE BACKBONE

Introduction

Hirayama's work on diphenyl and triphenyl alkanes15
(prototypes ofvinyl polymers) showed that only 1,3—di-
phenyl propane and 1,3,5-triphenyl pentane exhibited intra-
molecular excimer fluorescence. This led him to propose
what is today known as Hirayama's "n=3 rule", which states
that intramolecular excimer formation will occur only when
the aromatic chromophores are separated by a chain of three
carbon atoms. This geometric restriction arises in that
the optimum condition for excimeric interaction is one in
which the components are in a symmetrical parallel-sand-
wich configuration. This configuration is attainable for
n=3; however, for n=1 and n=2, it is impossible, without
deforming the tetrahedral bond angle of the alkane chain;
while for n>3, conformational instability due to steric
interaction makes such a sandwich configuration difficult
to reach.

An enormous amount of work was undertaken in the early
days to verify the "n=3 rule", and results from these

studies tended to confirm the rule: Yanari, et al.18

53

(1963) made the first observation of intramolecular exci-
mer fluorescence in isotactic polystyrene. Vala, et al.16
(1965) reported its occurrence in polystyrene (atactic
and isotactic) and in atactic po1y(l-vinylnaphthalene).
Kldpffer63 (1969) observed a weak intramolecular excimer
fluorescence in l,3-biscarbazolylpropane. Chandross and

6A (1970) studied a,w-dinaphthy1 alkanes and found

Dempster
that only l,3-dinaphthy1propane showed intramolecular
excimer fluorescence. Johnson and Offen65 (1971) reported
the case of poly(N-vinylcarbazole); and Fox, et a1.66
(1972) observed intramolecular excimer formation in poly-
(2-vinylnaphthalene).

One must point out, however, that all of the above
composite and polymeric systems have one thing in common:
a three-carbon chain separating the two interacting
chromophores. The question then arises: will the 'n=3
rule' apply with the same strictness in heterochain systems?
The first clue to the answer to this question was perhaps
the work of Longworth67 (1966), who reported the occurrence
of intramolecular excimer formation in poly(L-phenylalanine),
poly(l-benzyl-L-histidine) and poly(B-benzyl-L-aspartic
acid). In the first two polypeptides, the phenyl groups
are separated by 5 carbon atoms and l nitrogen atom; while
in the last compound, 6 carbon atoms and l nitrogen atom
separate the interacting groups. Ten years later Tazuke

37

and Banba observed weak intramolecular excimer fluores-

cence in some polyester series and their dimers, in which

65

pendant naphthyl groups are separated by as many as 16
carbon atoms and 2 oxygen atoms. And quite recently,
DeSchryver, et a1.68 (1977), observed intramolecular
excimer formation in a,w~dianthryl esters,in which the
chromophores are separated by upwards of 7 carbon atoms
and 2 oxygen atoms. The conclusion to all these observa-
tions is that heterochain systems do not seem to comply
strictly with the "n=3 rule" most probably, as a conse-
quence of the flexibility of the chain containing hetero
atoms. And one obvious implication of the behavior of the
above heterochain systems‘ is that it emphasizes the im-
portance of the chemical constitution of the chain in
influencing intramolecular excimer interaction.

Exceptions to the n=3 rule for carbon chain systems
were reported. Zachariasse and Kfihnle69 (1976) showed
that intramolecular excimer fluorescence occurs in a,w-
di-(l-pyrenyl)alkane Py(CH2)nPy for n=2-6, 8-12 and 22.
One wonders why it took so long to discover such serious
exceptions to this rule. One obvious reason is the narrow
choice of fluorescent moieties. The early workers used
virtually only phenyl, naphthyl and carbazolyl-containing
systems which were readily available as vinyl polymers,
and simple bichromophoric alkanes. It turned out that the
above three aromatic groups have the following shortcom-
ings, compared to pyrene moeity:

i) Shorter monomer fluorescence lifetime (for example,

66

toluene-39 nsec and B-methylnaphthalene-59 nsec); while
pyrene has a lifetime of 250 nsec.

ii) Smaller intermolecular excimer stabilization
energy (-AH), which is on the order of 5 kcals mole-l

compared to 10 kcals mole'l for the pyrene excimer.

The above two factors are surely essential in intramolecular
excimer formation:

a long—lived excited singlet state would lead to in-
creased probability of encounter between chromophores,
widely-separated in a chain.

and when such an encounter occurs, the binding energy
of the resulting excimer will depend on the extent of
orbital overlap, which in turn depends on the electronic

and geometric structure of the aromatic moiety.

The crucial role of these two factors in determining the
occurrence of intramolecular excimer fluorescence, is
clearly demonstrated in the case of Py(CH2)nPy.

In the foregoing review, three factors are seen to
exercise significant control on intramolecular excimer

formation, namely:

Length of chain;
Chemical constitution of chain; and
Electronic and geometric structure of the aromatic

moiety.

67

From the results of the work of Zachariasse and Kfihnle,
cited earlier, it would appear that the third factor
above - the nature of the fluorescent group, is the pre-
dominant factor.

The work to be described in this chapter will focus
on the effect of the intervening chain (its length and
chemical constitution) on the probability of forming ex-
cimers and the importance of the "n=3 rule" for a given
series of bichromophoric or polychromophoric systems.

This workudJJ.also add to the increasing evidence, obtained
from studies of intramolecular excimer formation in homo-
polymers and copolymers alike, that chain coiling in poly-
mers can produce geometries favorable to excimer formation
between non-adjacent chromophores. Currently, the picture
is not clear: for example, while it is Justifiable to
attribute intramolecular excimer formation in poly(ace-
naphthalene)70 in dilute solution to interactions between
chromophores on adjacent turns of the polymer coil, because
face-to-face interactions between adjacent rings are un-
likely, such proposition in the case of poly(N-vinyl car-
bazole)65 is doubtful. Since face-to-face interactions

are not sterically hindered, the absence of intramolecular
excimer formation in 1,3-bicarbazoly propane could there-
fore be a reflection of the electronic fluorescence
properties of the carbazolyl moiety, rather than geometric

restrictions.

68

The controversy over the involvement of neighboring
and non-neighboring interactions in excimer formation in
polymers becomes even more crucial in studies of fluores-
cent properties of copolymers. This is because the
phenomenon of intramolecular excimer formation is poten-
tially useful in understanding monomer sequence in co-

polymers.7l’72 66

For example, according to reactivity
ratio, copolymer of l-vinylnaphthalene and styrene would
tend towards an alternating copolymer structure. The ob-
servance of intramolecular excimer fluorescence in the
above copolymers, having as low as 3-A mole percent vinyl-
naphthalene units, suggests that sequences of two or more
adjacent units containing naphthyl groups must be dominant;
otherwise, the excimers are formed between non-adjacent
chromophores on the same chain. With the structure of the
copolymer in question remaining uncertain, it is difficult
to access the validity of either interpretation on the
sequence of l—vinylnaphthalene-styrene copolymer. Our
polyamide series, with their well-defined copolymeric
structures, seem to offer a useful system for distinguishing
between adjacent and non-adjacent chromophoric interactions
in a polymer chain. In addition, our studies may shed
light on the general question of possible conformations of
a 'giant' molecule in solution.

In this chapter, we will study the absorption and emis-

sion characteristics of the polyamide series: P2-1U,

69

93-19, P6-ll and

'U

8-1L, and the model compound 11 (their
synthesis is described in Chapter 3). For brevity, we

will re-designate these compounds as P-2, P-9, P-6 and

F-8 and KC respecti rely.
NNZCM% o --.. .EQLYMER
ll 2 P—Q
HZN— 3C+CH 2+6: 4 P—4
6 P—é
r18 P—8
3(:
"ff“: 9‘3
hACl C)

We will present evidence for intramolecular excimer for-
mation in the above polyamide series. Results of fluores-
cence lifetimes and quantum yield measurement of monomer
and excimer emissions will be given. A kinetic and thermo-
dynamic analysis of the temperature effect on excimer
emission in P-2, will also be made. The influence of
viscosity on intramolecular excimer interaction in P-2,
will be discussed. Finally, evidence for the occurrence

of 'dimer' emission at high polymer concentrations will

be presented.

Absorption and Fluorescence Spectra of Polyamides and

 

Model Compound

The room temperature absorption spectra of the monomer
model compound (MC) and polymer, P-2, are shown in Figure
3.1. A dilute solution (1 x 10'“ M/l in CHCl3/Et0H, 1:3)
of each compound was used. The molarity of the polymer
solution was based on the weight of the 'repeat unit'.

The absorption spectra of the other polyamides: P-B,

P-6 and P-8, were similarly obtained, and are identical
to that of P-2. From the above figure, it is clear that
the monomer model compound has the same absorption char-
acteristics with the polymers; their molar extinction
coefficients are similar ($5,000). Also, the diamine
monomer-2,6-bis(N-methyl, methylamino)naphthalene(8), III
possesses an identical absorption spectrum (Figure 6.1).

The absorption bands shown in Figure A.l are due to
the naphthalene moiety. The weak, long-wave-length peaks:

308, 315 and 322 nm are due to the 1L naphthalene transi-

b
'tion; the intense peak, with an absorption maximum at
N275 nm, is due to the 1La transitions. The identical
absorption properties of the above compounds show that
only weak ground state interactions involving the naph-
thalene chromophore occur in the polymer.

The room temperature fluorescence spectra of the

model compound (MC) and the polyamides: P-2, P-A, P-6

and P-8, are shown in Figure U.2. The model compound

«J
.-J

 

50 _ (0)

pl
0

ABSORBANCE
go
c>

 

 

 

 

L 1 I 1
Q0 260 280 soc 320 340 ”260 280 300 320 340

WAVELENGTH (nm)

 

Figure &.1. Room temperature absorption spectra: a
Monomer model compound, b) P-2 (l x 10‘ M

in CHCl3/EtOH, 1:3). Aex 290 nm.

72

 

Km
.5: omm « .Amua .20em\MHozo

as Q\z 10H x Ho m-m.... .mum ..... .sum ..... .mnsuuua

.pcsoasoo Hop EIIII "mpuomom mocmommmOSHm mmsumLmQEmu Boom

AES Ihwzmgm><>>

one.

 

 

_

 

.m.: mpzmHm

 

(90!JD|9J)AJJSNB.LN| SONSOSBHOO'H

 

73

shows only a single fluorescence emission band which is

attributed to the naphthalene emission with a maximum

max

(xem

) at N355 nm. In contrast, the polyamides show in
addition to the naphthalene (monomer) fluorescence band
at 355 nm, a broad band at longer wave lengths, (mu25 nm).

A number of questions immediately arise: what chemical
species gives rise to the new band? What are the photo-
physical processes that lead to such band? It is relevant
to note that the diamine monomer (8), III, shows a two-
component fluorescence band (see Figure 6.3), under the
same conditions used for obtaining the polymer fluores-
cence spectra. As will be discussed in Chapter 6, and ac-
cording to a recent report73 by the author, this broad-
band in the diamine spectrum is attributed to a solvent-
induced intramolecular excited state complex (exciplex),
formed between the naphthalene (acceptor) group and the
amino (donor) group.

Thus it is natural to consider first the possibility
of a donor-acceptor interaction occurring between the
amide nitrogen and the naphthalene group, in the polymer

backbone. However, a number of observations rule out

such possibility:

i) the absence of a similar band in the fluorescence
spectra of the monomer model compound (MC), which possesses
the amide group similarly linked to the naphthalene group,

as in the polyamides.

79

ii) the slight enhancement of the broad band in the
polyamides upon addition of a few drops of trifluoroacetic
acid; a similar treatment to a solution of the diamine (8),
leads to a complete quenching of the broad emission band,
with a considerable enhancement of the monomer (naphthalene)
emission. This is due to protonation which ties up the
nitrogen lone-pair electrons, thus increasing its ioniza-
tion potential.

The next logical candidate for the broad emission band
is the interaction between two naphthalene groups, either
in the same or different polymer chains. The fact that
dilute polymer solutions exhibit these broad bands suggests
that the interaction is intramolecular. Further confirma-
tion of this, which comes from concentration studies, will
be presented later. The next question is whether the
interaction occurs in the ground or excited state. If
it is in the ground state, then it is a 'dimer' complex,
which should exhibit some characteristic absorption band.
Identical absorption spectra of the model compound and
those of the polyamides rule out any significant ground
state interaction between the naphthalene groups. The
observed coincidence of the excitation spectra of the
monomer emission at A 355 nm and that of the broad band

ems

at Rem U25 nm strongly suggests an excited state inter-

action of the naphthalene moieties, and that the new broad

band arises from an intramolecular naphthalene excimer.

75

Subsequent studies in this chapter will support this.

The slight enhancement of the excimer emission due
to addition of acid can be accounted for in terms of the
elimination of the small perturbation effect of the lone-
pair nitrogen of the amide group. Such perturbation effect
in the amide monomer is small compared to the amine monomer
due to the carbonyl group which reduces the basicity of
the nitrogen. Thus, intramolecular excimer interaction
versus intramolecular exciplex interaction is more favor-
able in the polyamide systems. Such an enhancement of
excimer fluorescence in heterochain systems by protonation
or acylation of the lone pair on a chain-nitrogen atom,
has been reported by Wang and Morawetz7u (1976). They found
that di-benzylamine in its basic form, does not show any
detectable fluorescence, while in the protonated or acylated
form (as in N,N-dibenzylformamide) it exhibits a strong
excimer band. In the basic form, it appears that the
nitrogen lone pair, despite the intervening methylene
group, effectively quenches the excited singlet state of
the phenyl group, perhaps via an intramolecular exciplex
mechanism. In this respect, the use of polyamides in
studying intramolecular excimer formation in nitrogen-

containing heterochain systems seems advantageous.

“\l
(J\

Kinetics of Intramolecular Excimer Formation

For a kinetic treatment of the photophysical processes
involved in intramolecular excimer formation, we draw the
following reaction scheme, which is similar to the inter-
molecular case,39’7S except that the rate constant of

association is of first order:

K
* DNI
1M~|-M <— 910*

’ Kmo
7/ at; “no “so

A“ +11VM “A'th nn'an'Ffufi)

where M represents the monomer (in our case - naphthalene)
and 0* is the intramolecular excimer. Birks,76 nomen-
clature is used for the monomer and excimer rate constants,

which are defined as follows:

kFM : fluorescence of monomer

kIM : internal quenching of monomer
kDM : excimer formation

ka : fluorescence of excimer

kwn : dissociation of excimer

RID : internal quenching of excimer

77

For simplified analysis, we also define the following

parameters,
kM = kFM + kIM
kD = kFD + kID

If the system is subjected to steady excitation with light

-1

intensity IO einsteins A'ls , the rate equations are:

1

Siagil e Io _ (kM + kDM)[lM*] + kMD[1D*] (u.1)
1

9La%11.= kDM[1M*] - (kD + kMD)[1D*] (A.2)

Assuming photostationary conditions

d[lM*] _ d[1D*l =

 

 

dt ‘ dt 0
then
I (k +k )
[1M*]stat = (k +2 g+kM2k k (”’3)
D MD M D DM
I k
[113*] :3 0 DM (“'14)

stat (kD+kMD74kM+kaDM

The intensity of monomer fluorescence (photons per unit

time) is given by

78
IM = k[lM*]kFM (u.5)

where k is a constant, dependent on the experimental condi-

tions. Similarly, excimer fluorescence intensity is given

by.
ID = k[1D*]kFD (4.6)

Substituting for [1M*] and [1D*] in equations (A.3) and
(A.h), using Equations (h.5) and (9.6) respectively, the
ratio of the excimer fluorescence intensity to that of the

monomer, becomes:

ID = kFDkDM
IR. kFM(kD+kMD)'

 

(u.7)

The corresponding experimental decay times are given by:
r = (k +k )"1 (u 8)
M FM IM '

(kFD+kID)‘1 (u.9)

TD

The above relations will be utilized in the following
sections to obtain kinetic and thermodynamic parameters
for excimer formation in the polyamides and model com-

pound.

79

Fluorescence Quantum Yields, Decay Times and Rate Constants

of Polyamides and Model Compound

Decay times for monomer and excimer fluorescence were
determined by single photon counting technique using a
nanosecond time-resolved spectrometer. Solutions of the
polyamides and model compound (2 x 10'!4 M/L in CHClB/
EtOH, 1:3) degassed by a three-cycle freeze-pump-thaw,
were used; all measurements were done at room tempera-
ture. Monomer and excimer decay data were monitored at
330 nm and “25 nm respectively and are shown in Figure
A.3. These decay curves provide77 a graphic demonstra-
tion of the reality of the excimer formation process.
Similar decay patterns were obtained for the other poly-
amides.

Analysis of lifetime data was done by KINFIT to obtain
the experimental decay times for monomer (TM) and excimer
(TD) emissions, and the corresponding rate constants kM
and kD. These results, along with the calculated rate
constants for fluorescence emission of monomer (kFM) and
excimer (kFD) are shown in Table A.l. Calculation of

kFM or kFD was done using the relation:

kFM = TM/‘I’FM

where ¢FM is the monomer fluorescence quantum yield. Rela-

tive fluorescence quantum yields of monomer and excimer

80

 

Figure ”.3. Fluorescence Decay curves of P-h: Top-monomer
(Aem 330 nm): bottom-excimer (Aem #25 nm).

81

 

 

OH.m HN.~ m~.nq nuo.o om.m Ho.m o~.- hefl.o ca.o mum

 

 

05¢ £.~ mmém Zoé om.m .36 3.3 and :6 ...-m
om.m ooé end.» ~86 .26 36 3.3 m36 «To sum
oo.e «mg 36m H86 36 36 2.3 «:6 36 NJ“
-- ...... --- --- on.m Hm.m we.m~ eeH.o -- one flees:
Aa-emmmoav AH-emmsoflv Aemmsv one H-emmeodv AH-ueenoflv Aemesv sue mm. eesosaoo
aux ax 9 aka, xx 9 an
umawuxm umaoaoz

 

 

.Amua .moummfiomo as : e-oH x Ho essoeaoo Hoses was
mmvwammaom mo Axv mucpumsoo mama can Apv omega maven .Aev mpamu» arucaac ouauuuuuoaam .H.c «Home

82

were obtained using 2,6-dimethylnaphthalene as a standard.
Undegassed solutions (1 x 10-4 M/L in CHCl3/EtOH, 1:3)
were used. Results are presented in Table “.1, along with
ratios of fluorescence intensities of monomer and excimer,
(ID/IM). A plot of excimer fluorescence quantum yield,
QFD and ID/IM ratio, versus the number of atoms (n) in
the chain between naphthyl groups, is given in Figure u.u.
A minimum in both ¢FD and ID/IM is observed at n=10 and 12
(representing 20, 2N, 6C and 20, 2N, 8C atoms respec-
tively); while maximum values of ¢FD and ID/IM occur at
n=8 and 1“ (representing 20, 2N, DC and 20, 2N, 10C,
respectively). These trends are also reflected in the

fluorescence spectra (Figure 4.2).

Activation Energy for Excimer Formation in Polymer, P-2

Intramolecular excimer formation, like the intermolecu-
lar case, is a diffusion-controlled process. However,
unlike the intermolecular case, the binding of the two
interacting chromophores by a chain (as in composite or
polymeric systems) removes any effects on excimer inter—
action, due to molecular translational diffusion; thus
the main dynamic influence would appear to be due to
molecular rotation. In polymeric systems, an additional
molecular motion comes into effect, namely microbrownian
(segmental) motion, and it appears to be a major deter-

mining factor in intramolecular excimer formation in

83

.zH\QH_ A. 5&9: "oocmomopozfiu poEHoxo amazooHoEmpucH
so monocoosonno :fimno coozpon Acv .coauwpmoom oHEoum mo vacuum .:.: mpzwfim

820.2 oz_zu>mm:z. s .023

 

 

v. N. .

o. q _ . _ o. . w 4 mo.

N. T l N.

v. I l S

0. T l m.
:H

0.xo' m. l l o.
N H

mcase
ON T 1 ON

 

 

 

8A

polymers.

In order to examine the above dynamic effects, the
fluorescence emission of P—2 was taken at different tem-
peratures (Figure H.5). A maximum in the excimer fluores-
cence intensity is evident at -l6.5°, decreasing above
and below this temperature. On the other hand, monomer
fluorescence intensity increases continuously with de-
crease in temperature.

The activation energy for intramolecular excimer forma-

tion in P-2 was computed using Equation A.7,

H

D _ kFDkDM

IM kFM(kD+kMD)

 

The following assumptions were made:

i) RFD and kFM are independent of temperature.
ii) kMD << kD (in the low temperature region).
iii) kMD>> kD (in the high temperature region).

Assumption (1) generally holds over a wide range of tem-
peratures for many aromatic molecules.78

Using assumption (ii), then at low temperature

Since kDM represents an activated process (excimer forma-

tion) we'can replace it with the Arrhenius relation to

85

 

 

 

A
0
.2
~
.2
o
.3
>- :53
_-_
a) \
z 'I s‘
E ,' xp .
o I \ \

3 I! '\. ‘-

{1 \ \

'I \. \s '.

'I. \ "‘.-~

; \

.\ ...\\‘
\ ".. \
\\. ”:13
I "‘=._,]_
350 450
Figure u.5. Temperature dependence of intramolecular

 

5 “’

Wave Lenoih (nm)

excimer emission in polymer, P-2:
o -6203 -.-.-.
chloroform—ethanol (1:3); Aex

---- -16.5, ...
vent:

21.5,
-lll°C. Sol-
290 nm.

 

86

obtain

FM-

rficF

(n.1o)

where ED is the activation energy for intramolecular excimer
formation. A plot of in (:3) versus T'l for P-2 gave a
straight line, in the low temperature region, with a nega-
tive slope (Figure A.6), thus validating assumption (ii)
above. From the slope, which is equal to (- 1?), an activa-

-1

tion energy of 1.8 kcals mole was obtained.

Under assumption (iii), we get that

 

 

I k
-n(D)=2n<FD)+£n:D(M>
Tfi' RF kMD
7‘ 7‘ 7‘
(AH -AH ) (AS -ASD M)
MD DM MD + constant (4.11)
RT R

where (AH;D - AHSM) is the enthalpy change of the monomer-
excimer equilibrium. A plot of in (;E) versus T l for P-2
gave a straight line in the high temperature region, with
positive slope (Figure u.6). From the slope, a value of 2.2
kcal mole'1 was obtained for (AHfiD - AHEM)'

Discussion

We will now discuss the foregoing results of fluores-
cence lifetime, quantum yield and activation energy mea-

surements. First, we examine the structural basis for the

.Amua .moum\maozo
2a A\2 OH x HV mum Low oapmg 2H\QH mo cocoocoomo onzumngEoB .m.: mpswfim

 

87

 

 

:-
...
.-x. ma.
NN 0.0 N.@ 0.0 v.0 0.0 0.? N6 QM tn n O

_ _ _ _ _ d A _ _ _

d1 a 1 mod
0
11mzu
ItVNAv
M.

l Nnd

l ova

 

 

88

formation of intramolecular excimers in the polyamides:

 

For purposes of illustration, we consider one of the poly-
amides, P-2, represented above in its fully extended form.
In this model (Fisher-Herschfelder-Staurt's), the methylene
chain of the dicarboxylic acid moiety, is planar zigzag
and coplanar with the amide linkage.79 The plane of the
naphthalene ring is then perpendicular to that of the amide
group, in order to minimize steric interaction between
the hydrogen atoms of the ring and those on the methylene
groups attached to the ring.

Rotation about the C-N bond of the amide group is
hindered because of its partial double bond character.
This introduces rigidity in the chain, likewise the pres-
ence of the naphthalene ring impacts additional rigidity.
However, molecular rotation is possible about the follow-
ing bonds: C

C Cal-N, and Cal'cal) (al-aliphatic and

ar‘ al’

ar-aromatic). Thus, the naphthyl ring can rotate about

the axis that link it to the two methylene carbon atoms;

89

this motion seems to occur extremely fast, with a relaxa-
tion time on the order of 10'9 sec. Evidence for this,
comes from fluorescence depolarization measurements (Fig-
ure u.7), in which the parallel and the perpendicular com-
ponents of the polarized monomer fluorescence have equal
decay times. This implies that some rotational motion of
the polymer chain, involving the ring, has a relaxation
time greater than the excited state lifetime of the
naphthalene moiety (i.e., 10'9 seconds). Since it is un-
likely that the entire repeat unit would execute a seg—
mental motion that fast, it seems reasonable to attribute
the observed complete fluorescence depolarization in these
polyamides to the free and fast rotation of the ring about
the axis that links it with the two methylene groups; with
the amide groups serving as 'hinges'. In addition, there
is free rotation of the methylene chain (of the carboxylic
acid moiety). Consequently, the combined effect of the
two motions: ring and methylene chain rotations, would
render the polyamides flexible.

In dilute solution, such a flexible polymer molecule,
exists in the form of a contorted coil, continuously chang—
ing its shape. During these writhings of the polymer chain,
long-range forces (dipole-dipole, dipole-induced dipole
and dispersion forces) can lead to the approach to one
another, of groups that are distantly located on the chain.

Such interactions are expected to be pronounced in polyamides

90

 

Figure 4.7. Fluorescence depolarization: top, P-2,
bottom, P-8.

91

where the amide functional groups bear large dipole moments
(3.7 Debye)80 directed approximately perpendicular to the
chain axis. Thus, intramolecular excimer formation in
polymers like ours, that possess widely separated chromo—
phores, would rest largely on the flexible, coiling nature
of the chain.

The observation that the fluorescence quantum yields
of the intramolecular excimers in the polyamides show a
maximum at n=8 and 1a, and a minimum at n=10 and 12, may
be rationalized as follows: We propose that intramolecular
excimer interaction in these polyamides can occur between
neighboring naphthyl groups and/or non-neighboring ones.
Under this premise, the maximum at n=8 (P-2) is probably
due to the relatively greater frequency of encounter between
neighboring naphthyl groups, because of the shorter distance
of separation. As we go to P-h, the increased distance of
separation reduces the rate of encounter between neighbor-
ing naphthyl groups; so that its excimer fluorescence may
arise largely from non-neighboring interactions (Figure
H.8a). The fraction of such non-neighboring excimer sites
seems to be small, hence the observed decrease in excimer
fluorescence intensity. The same effect appears to be
operating in P-6 also. However, as the chain becomes con-
siderably long as in P-8, the flexibilizing effect of the
methylene chain can lead to loops, in which neighboring

naphthyl groups are once more brought into close proximity

Figure u.8.

\\1\

92

. ‘\\ .'
/

(o)

(b)

(c)

Non-neighboring and neighboring group inter-
actions, a) and b) respectively. Solvent
effect on polymer conformation: compact coil
in poor solvent (a and b), expanded coil in
good solvent (c).

93

(Figure H.8b), resulting in the observed increase in exci-
mer fluorescence intensity.

Evidence in favor of the involvement of non-neighboring
naphthyl groups in intramolecular excimer formation, is
provided by a strong solvent effect on excimer fluores-
cence in P—H (Figure u.9). The intensity of excimer emis-
sion is seen to be least in CHCl3 (thermodynamically good
solvent) and highest in CHCl3/EtOH, 1:3 (thermodynamically
poor solvent). In a good solvent, the polymer chain will
be highly expanded (Figure u.8c), thus substantially re-
ducing the number of excimer-forming sites. But in a poor
solvent, the polymer chain assumes a compact conformation
(Figure 4.8a), thus permitting non-neighboring-group inter-
actions, with an increase in excimer fluorescence.

The difference in the decay times of excimer emis-
sions in the P-2, P—u, P-8 excimers (35, A8 and AS nsec,
respectively) may reflect the predominant neighboring
naphthalene interactions in the P-2 polymer. The smaller
lifetime may be rationalized in terms of a less symmetric
excimer.

The activation energy of 1.8 kcalsmole-l obtained for
P-2, is comparable to those reported for poly(l-vinyl-
naphthalene) and poly(2-vinylnaphtha1ene):66 2.1 and 2.0

kcals mole-l

, respectively. Also, activation energies on
this order of magnitude were reported for a series of

polyesters possessing widely-separated pendant naphthyl

9H

. 0
5: omm x « .maomo .... .maomzo -- .Amuflv :Oam

\maomo Illl ”sum mo mocoomoQOSHm LoEHoxm co uco>aom mo vacuum

AESIHOZMJMZQB

 

 

 

 

 

 

 

.m.: opswam

 

(9A!$°|9J)A.LISN3J.NI EDNBDSEHOH'H

 

95
groups.37 In contrast, 1,3-di(l-naphthy1)propane and
l,3-di(2-naphthyl)propane have the following activation

1 respectively. One

energies: 3.3 and H.5 kcals mole"
wonders why these smaller molecules would possess higher
activation energies for intramolecular excimer formation.

Two factors may be considered responsible for this:

i) availability of more favorable conformations to
adjacent or non-adjacent chromophores due to chain coiling.
ii) possible energy transfer along the chain, with a

favorable conformation for excimer formation acting as a

sink.

The above factors, notwithstanding, measured activation
energies for intramolecular excimer formation in polymeric
systems should also reflect the degree of segmental dif-
fusion. The closeness in activation energies between the
above vinyl-polymers (with pendant naphthyl groups) and
P-2 (with doubly-bound naphthyl groups) seems to imply
that our polyamides are equally as flexible, despite the
presence of the amide group and doubly-bonded, chain-
incorporated aromatic rings.

In this respect, it is pertinent to note that Hayashi
and Tazuke36 (1973) did not detect intramolecular excimer

fluorescence in the following series of polyamides:

96

- 1
——H C' ‘C H-g-(CHgi-g“ x=2,3,4,6,8.

L.

 

 

I -n
In this series, as in ours, the fluorescent aromatic moiety
is doubly-bonded in the polymer chain; also the same series
of aliphatic dicarboxylic acid groups are involved. How-
ever, their polyamides appear to be poor candidates for

intramolecular excimer interaction because of a number of

disabilities:

1) Increased chain rigidity resulting from the con-
Jugation of the carbonyl groups with the carbazolyl groups;
this would tend to lead to more extended linear chains in
dilute solution.

ii) Hydrogen bonding of the amide groups causes further
rigidity, and makes the polymers very insoluble in solvents
suitable for fluorescence studies; and

iii) The carbazolyl group itself is a poor excimer-
forming aromatic species; for example, intermolecular
excimers of N-isopropyl carbazole are not observed at all,82
and it has been suggested that the weak exCimer-type emis-
sion in 1,3-dicarbazoly1 propane (which satisfies "n=3
rule") may be due to an intramolecular dimer.65 Thus, in

view of the above chemical and structural disadvantages, it

is therefore not surprising that the polyamide series

97

having the carbazolyl unit in the backbone, failed to show

intramolecular excimer fluorescence.

Effect of Viscosity on Excimer Formation in Polymer, P-2

Intramolecular excimer formation in polymeric systems,
as noted in the preceding section, proceeds via chain
conformational transitions, the rate of which is related
to the internal viscosity, ¢, of the polymer chain. In-
ternal viscosity of a polymer chain can be defined as the
ratio of a force applied to the ends of the chain to the
rate of their separation, when the chain is suspended in a
medium of negligible viscosity. Originally, ¢ was thought
to be independent of the viscosity of the medium. However,
it is now believed to consist of two components: ¢' and
¢". ¢' is the potential barrier separating the chain con-
formations; this component is independent of the viscosity
of the solvent. ¢" arises from the lateral displacement
through the viscous medium of chain elements next to the
location of conformational change; and so it is dependent
on solvent viscosity (n). The following empirical rela-

tion has been found to hold:

135. u.
fn n + B ( 12)

where f is the friction coefficient of a segment.

From our earlier consideration of the rate processes

98

involved in intramolecular excimer formation, we obtained

Equation (N.7), which can be re-written as,

 

IM = kFM(kD+kMD) (u 13)
IS kFD kDM

In Equation (“.13). the viscosity dependent terms are kMD

I
M l
and kDM' If kD >>kMD, then IE is proportional to ES; .
But kDM is the rate constant for a conformational transition
I

involved in excimer formation, so that IE is expected to
be linearly related to the medium viscosity.

To verify the linearity of the above relation, fluores—
cence spectral measurements were carried out at room tem-
perature, using dilute solutions (1 x 10’“ M/L) of P-2,
in a mixture of chloroform, ethanol and glycerol. The
amount of chloroform was kept constant at 25% by volume;
while the amounts of ethanol and glycerol were varied up
to a maximum of #5% by volume of glycerol. Higher amounts
of glycerol could not be used without precipitating the
polymer from solution. The fluorescence spectra obtained
are given in Figure H.10.

At low viscosities (g 10 cp) the intensity of the exci-
mer band appears constant, while the monomer band increases
with increase in medium viscosity. A plot of ;% versus
viscosity (n) of medium is given in Figure “.11. There
is linearity in the plot at moderate viscosities, and

curvature at low viscosities. Similar results were

99

.ea omm xma .m: --

.mm ... .oa ..... I .o III. "mESHo> an Hopoozaw ucmopom
.mmLSutz Honoozau\:0ummaoxo :a mum mo wpuomom cocoomopozam
AES IHOZMJM><>>

Omm 000 one. 0.? Con

 

 

 

 

 

 

(anlaJMilSNBLNI BONEDSBHOFTH

.oa.: magmas

lOO

.zufimoomfi> uco>Hom ho :oHuocsm o no mum mo
QH\EH .mocmn noEHoxo cam LoEocoE mo moauamcouCH mo oaump one .HH.= mhswfim

33 E883

em om m. m. o e o
_ _ _ _ _ _ m

 

 

 

 

 

101

reported by Avouris et a1.,8u for 1,3-dinaphthylpropane.
On the other hand, Wang and Morawetz7u obtained a linear
plot for di-benzylether, at both low and high viscosities.
The non-linearity in our system at low viscosities may be
due to the assumption: kD >> kMD, which is not quite cor-
rect for P—2 at room temperature, at which the fluores-
cence spectral measurements were made, at different vis-
cosities. This is because, :3 in P-2, decreases with rise
in temperature in the vicinity of room temperature, so that
kMD (rate constant for excimer dissociation) is likely to be
competitive.

However, the behavior of monomer and excimer fluores-

cence in P-2 at low and moderate viscosities indicate a

number of things:

i) insensitivity of chain conformational transitions to
low macroscopic viscosities, as shown by the constant
intensity of the excimer band in this region;.

ii) the linear dependence of ig-on medium viscosity at
moderate viscosities, and its finite value at zero vis-
cosity, tend to support the concept of chain internal
viscosity as a sum of two terms: a viscosity-dependent
and a viscosity-independent term;

iii) the enhancement of monomer fluorescence even at
very low viscosities seems to show substantial slowing down

of molecular collision around the excited chromophores,

which would lead to a decrease in thermally-induced

102

radiationless processes, and hence, enhancement of fluores-

cence .

'Dimer' Fluorescence at High Polymer Concentration

'Dimers' are ground state complexes formed by a rever-
sible combination of two unexcited monomer molecules (see
classification of molecular complexes, Table 1.1). They
occur frequently in concentrated solutions of many dye
stuffs, and lead to a change in the shape of the absorption
spectrum of the monomer compound. Dimer formation, in the
ground state, has not been reported in concentrated solu-
tions of neutral, small aromatic hydrocarbons like benzene,
naphthalene and pyrene, at room temperature. 9

8 However, some workers have indicated the formation

of dimers in concentrated solutions of some polymers.
Nishihara and Kaneko85 (1969) showed the existence of dimers
in concentrated solutions of polystyrene, by the change

in shape of the excitation spectra at different concen-
trations; Tazuke and Banba37 (1976) alluded to dimer
formation in concentrated solutions of a polyester con-
taining pendant naphthylmethyl groups. Both studies
did not show any detectable dimer fluorescence. To date,
the only report, to our knowledge, on dimer fluorescence

in concentrated polymer solution, is that of Kaniijo et a1.86

(1976) on polyvinylnaphthalene, which fluoresces, peaking
at 3A2, 360 and 378 nm.

103

In attempting to establish the concentration indepen-
dence, and the intramolecular nature of excimer formation
in our polyamides, we have observed a third peak, at higher
energy than excimer emission. Solutions of P-2, P-4 and
model compound were prepared in a 1:1 mixture of CHCl3
and ethanol, with concentrations ranging from 1 x 10'6
to l x 10"1 M/L. All solutions were used undegassed.

Fluorescence and excitation spectra were obtained in 1 cm

cell, by right-angle viewing.

Effect of Concentration on Fluorescence Spectra of P-2

Though similar spectral characteristics were obtained
for P-2, P-4 and MC, those of P-2 are by far more dramatic,
followed by P-4. A detailed account of the behavior of P-2
is thus given, and similarities and differences between its
behavior and those of P-4 and MC will be pointed out.
Spectral measurements on P-2 at room temperature and at
different concentrations reveal three types of behavior,
depending on the concentration (Figure 4.12).

1) High dilution region (<1o'5 M/L). The ratio of
intensities of excimer and monomer fluorescence, ii, in-
creases with decrease in concentration.

ii) Moderate dilution region (4 x 10'5 - 1 x 10"3
M/L) ;— ratio is independent of concentration.

iii) High concentration region (N2 x 10"3 - l x 10 1 M/L).

.m-a no neofiosaom ooaflao ea .zH\oH .oocoonchosao noeocos
new Loefioxo mo mmauamcouCH no ofipmp mo oocoocoamo :ofiumhucoocoo .NH.= ohzwam

35:835. co... 3.
o. m o v m o

 

104

 

_ fl _ q _ 9.0

 

 

 

 

105

Fluorescence spectra in this region are shown in Figure

4.13. The following observations are made:

continuous decrease in monomer (naphthalene) fluores-

cence (Aem 362 nm) with increase in concentration; complete

1 M.

max
em

quenching occurs at m1 x 10-
the broad excimer peak (A N428 nm) increases pro-
portionally with increase in concentration, and shifts to
the red by ~22 nm.

appearance of a new peak at 380 nm, which begins at

m2 x 10’3 M/L, and grows with increase in concentration.

Fluorescence Spectra of P-2 at Varying Excitation Wave-

lengths

-4 M/L) and a concentrated

A dilute solution (2 x 10
solution (2 x lo"2 M/L) of P-2 in 1:1 mixture of CHC13
and ethanol were excited at different wavelengths (294-
324 nm) at room temperature. The behavior of the two solu-

tions are markedly different. Dilute solution (Figure
4.14a):

the total fluorescence spectra shows no change in shape.
monomer fluorescence has maximum intensity at Aex

294 nm, and decreases by a factor of w5.6 at Rex 324 nm.
Concentrated solution (Figure 4.14b):

the total fluorescence band shows a change in shape:

106

.5: com me . 10H x H .... . IOH x H .....

.mIOH x N III- . IOH x H .AoauHH\moHoEv coHumnucoocoo
wcHzpw> um HH" v mOum\mHomo :H mum mo anuooam cocoomoposHm .mH.: opstm

 

HES IP02M4M><>>

 

 

 

homo So one ooc 98
/..N/.

. xU/

// ..u

.3::: )1 MW

/ a

S

D

3

883 N

D

3

Ann x. l

N

B

Hal

9.

m.

me

 

 

 

107

.AHHHV mmWM\mHozo uuco>Hom .zmm .... .zom ..... .aHm III-
.amm “Haze a max: m-oa x m An .axz w-oH x m in .meamooao>ez
coHumpHoxo wcHznm> pm mam mo mhuooom coco monosHh onsumpodsou Eoom .aH.: opstm

 

A55 Ihwzmqm><>>

 

 

 

fizxv

on... oom one oov ohm or.» con one cow onn
. .:. - thil _ _ d
.J/HHU/ mu
..../.// n
/. , o
x, ~ 3
. 3 . as.
a./ i q.
..,. , i 3
. C .Q 3
xxx. xi; ; mm
is , é a
.ey _ “N
Anzx. ......xC B
i .1
IA
nfl
E.
m
A
@-

 

 

 

 

 

 

108

the new band showing a higher peak height at Ae 294 nm

x
than at Aex 324 nm; the reverse is the case for the monomer
band.

the ratio of intensities of excimer to monomer fluores-

I

cence, —2, increases with decrease in wavelength (Figure

IM
4.17).

Excitation Spectra of P-2

The excitation spectra of the two solutions of P-2
used in the preceding section, were recorded at room tem-
perature. The emission monochromator was set at Aem
362 and 380 nm, to obtain the excitation spectra of the
monomer and the new—peak 'species', respectively. The

following observations are made (Figure 4.15):

in both dilute and concentrated solutions, the excitation
spectra at Aem 362 nm and Aem 380 nm are coincident.

excitation spectra of concentrated solution are, how-
ever, sharper, and with a short wavelength tail (Figure
4.15a).

also, excitation band of the concentrated solution has
a maximum at Aex 325 nm, and that of the dilute solution

is at Xe 292 nm.

X

109

w
.AHuHV menm\mHo:o loco>Hon momm -- .mom -||.uAsov s H
.uxz e-oH x m Ho .axz m-OH x N Am nm-a co nonooan aoaoooaoxm

. ASE Ihozw4m><3
one ooe can con 03 one cow can con

.mH.: opszm

OnN

 

 

 

 

2: A3

 

 

 

 

 

(3"!JDI91)AJJSN3.LNI BONBDSBHOO'H

 

 

110

Effect of Concentration on Fluorescence Spectra of P—4

and MC

Spectral measurements, with concentrated solutions of
P-4 and model compound (MC) were also made at room tem-
perature in a 1:1 mixture of CHC13 and ethanol. Their
fluorescent behavior under the same conditions are then
compared with those of P-2, at a particular concentration
(2.5 x 10"2 M/L) and at excitation wavelengths — 290 and
324 nm (Figure 4.16). Significant similarities and dif-

ferences are observed:

the fluorescence of MC is practically quenched when
excited at 290 nm; however, on excitation at 324 nm, it
shows detectable monomer fluorescence, and with the new
component at N380 nm.

the fluorescent behavior of P-4 as a function of con—
centration, is very similar to that of P-2, with two excep-

tions:

the intensity of the new peak at 380 nm is less in P-4,
at equivalent concentrations.
:3 ratio decreases with increase in excitation wave-
length in P-2; the reverse is the case in P-4 (Figure

4.17).

111

 

FLUORESCENCE INT ENSITY(relaflve)

 

 

 

 

 

Figure 4.16.

 

WAVELENGTH (nm)

Comparison of fluorescence spectra of con-
centrated solutions (2.5 x 10'2 M/L) of poly-
amides (P-2, P-4) and model compound (MC):

MC (13x 324 nm), ---- P-4 (A 324 nm)

. P-4 (Aex 290 nm), ----- P-2 ffiex 290 nm).

 

112

. .Aa\z
m-oH x S m-d 03>. m-oH x .08 ..-..Isz.-S n C e-aogoxz

-3 x d m-o O "newcoHogs coaoofioxo co szoH .oocoonoaoaC
m Losocos new hoEHoxo mo moHuHmcmucH no OHHwL no cocoocoooo

HES Ib02M4m><3
vmn own en «a com com com 8m 8m mom can can

.NH.= oaamfig

 

_ . _ H _ _ _ _ _ _ H

 

 

 

N5
1 0.0
l 0Nd
¢N0
0N0
Nncsfi
0nd

00.0

 

 

113

Discussion

The initial goal for carrying out the spectral mea-
surements with solutions of P-2 and P-4 at different con-
centrations was to establish the intramolecular nature of
excimer formation in them and the other members of the
series. A plot of ;% versus concentration, is seen to be
invariant with concentration in the moderately dilute
region; thus, confirming the occurrence of excimer inter—
action between naphthyl groups of the same chain. The
reason for the increase in Ifiratio with decrease in con-
centration in the extremely dilute region ( l x 10'5 M/L),
is not totally clear at this point, although it seems to
reflect some increased ease of chain conformational transi-
tions, made possible by better isolation of each polymer
chain at such extreme dilutions.

We now proceed to explain the fluorescent behavior
of P-2, P-4 and model compound at high concentrations.

The observed increase in ifi-ratio with increase in concen-
tration conforms to the behavior of small aromatic hydro-
carbons. At high concentrations, quenching of excited
singlet aromatic moieties by their ground state counter-
part, occurs; leading in most cases, to the formation of
intermolecular excimers. But unlike the case of small
aromatic hydrocarbons, concentrated solutions of P-2, P-4

and even the model compound, showed a new peak at 380 nm,

which lies to the blue of excimer fluorescence. This

114

implies the existence of a third emitting species, and

more than one absorbing species. The presence of a second
absorbing species is also indicated by the observed change

in shape of fluorescence spectra on varying the excitation
wavelength. This is even further confirmed by the shift

to longer wavelength (325 nm) of excitation maximum at

high concentration (P—2, Figure 4.15), together with a change
in shape of the excitation spectra.

We therefore believe that at high concentrations,
'dimers' of naphthalene moieties are formed, which fluor-
esce at 380 nm. The absorption and emission properties of
paracyclophanes, with two benzene rings bridged by (CH2)m
and (CH2)n groups, illustrate this type of ground-state,
proximity-induced interaction. For example,16 2,2'-para-

) at N282 nm, and

cyclophane shows absorption maximum (123x

max
em

max max
phane shows lab at 261 nm and Aem

emission maximum (1 at 356 nm; while 4,4'-paracyclo-
at 335 nm. The emis-
sion bands in both compounds are due to intramolecular
excimers. The absorption and emission spectra of 4,4'-
paracyclophane are similar to those of a dilute solution
of polystyrene, which in turn has an absorption spectrum
similar to that of ethylbenzene. Thus, it is seen that
in 2,2'-paracyclophane, there has been a shift to the red
in both the absorption maximum and in excimer emission

maximum. The above differences are attributed to the

intermolecular electronic interactions of the phenyl

115

groups in their ground states because of the proximity
of the two chromophores in 2,2'-paracyclophane. We invoke
the same reasoning to explain our observations in con-
centrated solutions of the polyamides, namely the shifts to
the red of both the excitation maximum and the excimer
emission maximum. Concentrated solutions of polystyrene,
as noted earlier, exhibit similar red shifts in its excita-
tion and excimer bands;this behavior was similarly ration—
alized.85

The contrast in the fluorescence behavior of P-2,
P-4 and MC may be explained in terms of the following

qualitative scheme:

9
l0

 

 

 

 

 

 

I11! I11!
3 4
v; V!
* 5 II 7 *
A+ A —-—>‘6 (AA) <——§—’ (AA)
MONOMER EXCITED DIMER EX IMER
II I2 l3

A+A+th (AA)+hvED (AA)+hI/E

116

Dimer formation between the naphthalene moieties in
concentrated solutions, is considered to arise from en-
hanced dipole-dipole, dipole-induced dipole and dispersion
interactions due to the close proximity of the naphthalene
groups. These dimers on excitation (process 4) can under-

go a number of processes:

1) fluoresce at xem N380 nm (process 12);

ii) form an excimer, a much more stable excited com-
plex (process 7);

iii) dissociate into an excited monomer and a ground

state monomer (process 5).

Generally, excited dimers do not fluoresce because of
their small enthalpy {-AH) of formation, which when less
than RT, would lead to rapid dissociation of the excited
dimer. This might be the case in polystyrene dimers, for
which no detectable fluorescence was reported.85 However,
in our polyamides, the presence of the amide dipoles in
the chain, would give rise to a relatively stronger intra
and interchain secondary (dimeric) interactions, leading
to the formation of stable excited dimers with characteris-
tic emission at 380 nm. Dimers of the model compound,
however, appear less stable, because of the absence of
the 'locking effect' of the polymer chain, which would tend
to hold the interacting naphthyl groups in place, and in

a more or less stable conformation that will permit

117

effective exciton resonance between them. On the other
hand, the fluorescence intensity of the dimer of P-4 is
less intense than that of P-2 because of 'dilution effect',
caused by larger average distance of separation of naph-
thyl groups in the former polymer, in which there will be
smaller density of chromophores for the same moles of
repeating units.

Because of the above 'locking effect' in polymeric
systems, it seems conceivable that excited dimers may also
be formed via process 6, in which there is strong exciton
coupling between an excited monomer and a ground state
monomer. Two experimental observations appear to support
the above suggestion: the coincidence of the excitation
spectra of monomer and dimer fluorescence monitored at
362 and 380 nm respectively; thus suggesting the involve-
ment of excited singlet state of naphthalene in dimer
formation. Also the observed higher intensity of dimer
fluorescence upon excitation at N290 nm (monomer absorption
maximum) suggests that process 6 in the scheme may be
efficient. However, the monomer and over-all fluorescence
intensity is higher on excitation at longer wavelength
(324 nm), which tends to support the existence of another
absorbing species, which may dissociate to give excited
monomers - in other words, processes 1, 4 and 5 may be
predominant at long wavelength excitation at high concen-

trations.

118

Conclusions

 

We have demonstrated experimentally, the formation of
fluorescent intramolecular excimers in a series of N-
methylated polyamides, with doubly bonded naphthyl groups
in the backbone, separated from one another by as many
as 20, 2N and 10C atoms. The fluorescence lifetimes and
quantum yields of the excimers depend on the distance of
separation between the chromophores, in a manner that seems
to imply the involvement of non-neighboring groups in
excimer interactions. Temperature measurements reveal a
sma11_energy requirement for the conformational transitions
involved in the formation of these excimers; and limited
viscosity studies indicate the extent the above transitions
are dependent on the viscosity of the medium. Finally,
we presented experimental evidence for the formation of
fluorescent dimers in concentrated solutions of the poly-
amides and their model compound.

From the above findings, we can reach the following

conclusions:

1) For a given aromatic moiety, intramolecular excimer
formation can occur over long distances of separation
between the chromophores, if the intervening chain is
flexible.

ii) polyamides present an example of flexible hetero-
chain systems that exhibit intramolecular excimer formation.

Our work is however the first known case.

119

iii) Chromophores, doubly-bonded in a chain appear
capable of engaging in intramolecular excimeric interac-
tions; this is the first known case in polymeric systems.

iv) The 'n=3 rule, in its original statement, is no
longer valid; rewording it as follows, seems necessary:
"intramolecular excimer formation is optimized when the
two interacting aromatic moieties are separated by a three-
atom chain". In such a way the statement finds wide spread
validity in homo and heterochain systems, the nature of the
fluorescent moiety, notwithstanding.

v) Finally, the detection of dimer fluorescence in
concentrated solutions of the polyamides and model compound
suggests strongly the need to consider dimer formation in

these systems.

CHAPTER 5

INTERMOLECULAR EXCIPLEX FORMATION IN
POLYAMIDES HAVING NAPHTHYL GROUPS IN THE BACKBONE

Introduction

The binding of a chemical moiety in a polymer chain
can lead to changes in its chemical or physical reactivity,
Such changes, which may be termed, "polymer effect", con-
trast the behavior of the moiety in a monomeric and a
polymeric situation. Rates of small molecule reactions
have been accurately described in terms of collision theory
and molecular dynamics. However, when polymers are involved,
the concept of "billiard ball" collisions between molecules
of nearly the same size, is no longer applicable. In
solution, swollen or molten state of a polymer, the random
motion of the billiard balls is replaced by a more re-
stricted and coordinated motion. This drastic change in
molecular notion, brought about by the steric restric-
tions of the chain, is likely to affect the reactivity of
a moiety. For excimer or exciplex formation, the polymer
effect will depend on the degree of movement of moieties
needed to attain a given geometry during the excited state
lifetime of a specific chromophore.

The phenomenon of intramolecular excimer formation,

which we studied in the preceding chapter might be

120

121

considered a classic example of 'polymer effect', in which
chain binding of chromophores leads to the observance of
excimer fluorescence in dilute solution. Intermolecular
excimer formation between similar chromophores occurs

at higher concentration and is concentration dependent.
Also, the same chain - binding determines as we saw
earlier, to what extent intramolecular excimeric inter-
actions can take place, between bound chromophores.

On the other hand, some reports have shown that, in
certain intramolecular processes, the reactivity of a
functional group remains essentially the same both in the
monomeric and polymeric states. For example, Y. Miron87
(1969), showed that the free energy of activation, AG:
for the rotation around the amide bond in piperazine, at
the covalescence temperature, was almost identical for
the piperazine polyamides and their low molecular
weight analogs. Also, Tobak and Morawetz88 (1970) re-
ported that the isomerization rate of the azobenzene
moiety is not significantly impeded by incorporation into
the polymer backbone.

In the above intramolecular processes, however,
rotational (segmental) motion is probably all that is
required to bring about the necessary interactions; so
that under these circumstances, any effect due to chain

translatory motion, will not be evident. And, as a matter

of fact, the binding of a molecular species in a polymer

122

chain literally removes its ability to do any sustained
translatory motion, unless by a whole-molecule motion
of the polymer, which hardly occurs.

Therefore, in order to get a total picture of poly-
mer effect on the reactivity of a molecular moiety, it
becomes useful to carry out intermolecular interactions
involving polymers and their monomer model compounds.

By so doing, the chain effect on both the translatory

and rotational molecular motions, may become fully ex-
posed. In choosing a reaction for this type of comparative
study, an important criterion is that the reaction does

not lead to any permanent changes in the electronic and
physical states of the polymer or its monomer model com-
pound. In this respect, photophysical processes have
proved to be potentially useful. Three main approaches,
involving the use of spectrophotometry have been reported
in the literature:

In one approach, the absorption characteristics of
ground state charge transfer (CT) complexes of polymers
are compared with those of monomer model compounds. Equi-
librium constants and molecular extinction coefficients
of the CT complexes are determined using the Benesi and
Hildebrand89 relation which, for the case of excess donor

concentration, is given by

2EAl/0D = l/eCT + l/ECTKED] (5.1)

123

where [A] and [D] are concentrations of acceptor and donor
in moles/litre; OD is the optical density; ECT is the molar
extinction coefficient of the complex, and K is its equi-
librium constant in litre mole‘l. Results of some works
based on this approach are not completely in agreement:
Sugiyama and Kamogawa90 (1966) studied CT complexes of
aza-polymers (for example, poly-2-vinylpyridine and poly-
4-vinyl pyridine) with quinone (acceptor) and the cor-
responding complexes of some monomer model compounds (for
example, pyridine and methyl-substituted pyridines).

They reported higher equilibrium constants for polymer
complexes than for monomer complexes. They reasoned that,
even though a longer period is required for quinone to
approach the donor group of the polymer, the complex, once
formed, is not so easily dissociated. Contrasting results
were obtained by Rembaum et a191 (1968), when they studied
CT complexes of poly(N—viny1 carbazole) and N-ethyl car-
bazole, with acceptors such as iodine and tetracyano-
ethylene. They obtained higher equilibrium constants for
monomer complexes than for polymer complexes. They ration-
alized their results in terms of steric hinderance caused
by chain configuration. The situation became even more
confused with the results of Hayashi and Tazuke36 (1973),
who studied CT complexes of a series of polyamides having
carbazole units in the backbone, and that of N-ethylcar-

bazole, using trinitrobenzene (TNB) as acceptor. They

124

reported higher equilibrium constant (K) for ethyl car-
bazole complex than for corresponding polymer complexes
(in agreement with the results of Rembaum et a1). On the
other hand,higher K values were obtained for polymer com-
plexes than for that of the monomer model compound,
(EtCONH)2Cz. They attributed the latter result to a
possible aggregation of carbazole groups in the polymers,
which could lead to a difference in the donor strength
of the carbazole moiety. They did not explain why the
monomer model compound, (EtCONH)2Cz, exhibits less exciplex
interaction with TNB compared with N-ethylcarbazole, even
though the latter has less donor strength. The work of
David et a195 (1976) appears to increase the evidence in
favor of greater reactivity of small model compounds than
the corresponding large molecules towards CT complex
formation. They reported higher K values for naphthalene
and ethylnaphthalene CT complexes with chloranil than for
the corresponding complex with polyvinylnaphthalene.
Although the experimental approach involved in the
preceding works has the advantage of specificity in inter-
action (involving only the combination of a donor and an
acceptor), its sensitivity is limited by the fact that the
weak CT complexes formed, generally do not exhibit any

absorption and emission spectra of their own, and only
slight shifts of the parent absorption peaks are used to

monitor CT interaction.

125

A second spectrophotometric approach is by fluores-
cence quenching. The quenching of fluorescence by a foreign
substance results from a collision, with formation of a

92 between the excited

transistory charge transfer complex
molecule and the quencher. As a result of this bimole-
cular process, the intensity of fluorescence (I) depends
on the concentration (C) of quencher, according to the

Stern-Volmer relation,

Io/I = l + KC (5.2)
where I0 is the intensity in the absence of quencher, and
K is the Stern-Volmer (quenching) constant. Thus, this
approach involves the determination of Stern—Volmer
constants, which serve as a measure of the extent of
fluorescence quenching by molecular collision. The works
of Duportail et a193 (1974) and Tazuke at 3194 (1975)
appear to be the only reported cases of this approach.

Duportail et al, carried out quenching of fluorescence
of diphenylanthracene and perylene with the following
quenchers: dimethylparatoluidine (DMPT), polymers and
copolymers of dimethylaminostyrene. They reported higher
quenching constants for DMPT than for the polymers and
copolymers, irrespective of the fluorescent molecule.

They explained these results in terms of the steric in-

fluence of the chain, which limits the possible ways of

126

approach towards the reacting group. Tazuke et a197,
studied the quenching of fluorescence of the anthryl

group of an ester by the anilino group of a series of
polyesters and their monomer model compound. Again,

higher quenching constant was found for the monomer model
compound than for the polymers. They advanced two ex-
planations for this: smaller diffusion constants of polymer
quenchers and possible overlap of quenching diameters in

the polymers due to congestion in the chain.

However, this approach is not without serious limita-
tions. Deviation from a linear Stern-Volmer plot can
arise mostly at high concentrations of either the fluores-
cent or quenching molecules. In the first case, the
deviation is largely due to resonance energy transfer in
the polymer. In the second case, it arises from a change
of the rate determining step as a function of the concen-
tration of the quencher. Besides, the observance of a
linear plot may not always prove that the fluorescence
quenching is occurring by a bimolecular collisional
mechanism.

The third spectrophotometric approach, utilizes the
formation of an exciplex, which we defined in chapter
one as an excited state molecular complex, formed between
an excited fluorescent moiety and a dissimilar molecular
moiety in the ground state. Its formation is accompanied

by the quenching of monomer fluorescence. This approach

127

appears to offer two ways of determining relative reac-
tivity: by exciplex formation and by fluorescence quen-
ching. To our knowledge, the only reported use of this
approach, appears to be the work of David et a195 (1976).
They studied intermolecular exciplexes, formed by poly-
vinylnaphthalene, polyacenaphthalene and acenaphthalene,
with N,N-diethylaniline (donor) and dicyanoanthracene
(acceptor). Plots of the ratio of intensities of exciplex
and monomer emissions, IE/IM, versus concentration of
quencher, showed greater slopes for the small molecule
model compounds than for the corresponding polymers.

This indicates greater reactivity of the former compounds
towards exciplex formation. However, the use by the
authors, of fairly high amounts of the polymers and model
compounds as quenchers is likely to have caused substan-
tial differences in medium viscosities between polymer

and model compound solutions; and since exciplex formation
is a diffusion-controlled process, its rate constant will
be proportional to the viscosity of the medium. Thus, the
observed lower rate constants for exciplex formation in
the polymers might have resulted, partly from the higher
viscosities of polymer solutions, and partly from lower
frequency factor arising from inaccessibility of reactive
sites of the polymer sidechain. The problem of viscosity
differences, can however be eliminated, by keeping polymer

concentration low and constant, and using varying amounts

128

of a small molecule as quencher.

In any event, the specificity of an exciplex inter-
action, which leads to the formation of a transiently
stable and fluorescent complex, makes it a potentially
reliable technique for studying polymer effect on re-
activity. Besides, the interaction is purely physical in
nature with no permanent alteration in the electronic and
geometric structure of the interacting components.

From the foregoing accounts on some of the early and
recent works on polymer effect on reactivity, it can be
seen that results are not totally in agreement, though
there is greater evidence in favor of small molecules as
being more reactive than the corresponding large molecules.
Furthermore, only the works of Hayashi and Tazuke36,
and Tazuke et al94 dealt with the influence of the distance
of separation between chromophores in the chain, on their
reactivity. While Hayashi and Tazuke found no systematic
trend in the equilibrium constants of polymer CT complexes
as a function of the distance of separation between the
reactive chromophores, Tazuke et al reported smaller
Stern-Volmer quenching constants for polymers with con-
gested reactive chromophores.

In this chapter therefore, we will seek further experi-
mental evidence on any differences in the reactivity of
a chromophore due to its binding in a polymer chain and

separation in a polymer chain with the general objective

129

of throwing some light on:

the flexibility of randomly coiled polymer chains,
and the rate with which such flexible polymer molecules
change their shape.

To do this, we will study fluorescence quenching and
intermolecular exciplex formation, involving the poly-
amides: P-2, P-4, P-6, P-8, and the monomer model com-
pound MC, as acceptors, and triethylamine (TEA) as donor
(quencher). (See Chapter 4 for structures of the above

compounds.)

Exciplexes of P-2, P-4, P-6 and_MC with Triethylamine

To dilute solutions (1 x 10‘” M/L in CH2C12 of the
polyamides: P-2, P-4, P-6, P-8 and MC, were added increas-
ing amounts of triethylamine (TEA). The fluorescence
spectra of these solutions were taken at room temperature
in 1 cm square cell. The fluorescence spectra of P-4/TEA
solutions are given in Figures 5.1 and 5.2. Similar fluores-
cence spectra were obtained for solutions of TEA with the
other polymers and the monomer model compound.

Figure 5.1 indicates a gradual quenching of the
monomer (naphthalene) fluorescence, accompanied by the
emergence and growth of exciplex fluorescence (broad

band), as the concentration of TEA (donor) is increased.

130

20 .NHommo "u:o>Hom .E: omm

« .2 oo.o ... .2 =o.o -..-..- .2 no.0 -.-.- .2 Ho.o --
.2 o.o.lll nw<meu .pocoo no <me no mcoHuwppcoocoo msoHpo>
wchHmucoo aim no Ezpuooom cocoomoposHm chapmhodsou Eoom

HES IPOZMJ w><>>

00¢

 

(”Nola-J) AUSNlLNI EONEOSBHOO‘H

 

 

 

 

.H.m enemas

131

 

 

.2 Hm.o -.-.- .2 mH.o --

 

.2 HH.o .... .2 mo.o "Heme. oosoo no <29 co neoaossocoocoo
mSOHLm> wcHCHmucoo .znm no mppooam oocoomoposHm opspmpoosou Eoom .m.m opstm
ASS 1.52mi m><>>
com com com one 00? 0mm .
.-s .mu;/ _ _ H _ _ n»
//. . \\
IOI. .\HL1:I: .\
/ /. .\\ I /. \
/,,/. x\.\a ....... ../ ,x. \.\
v, I, \.\\z. ...x /, . Hm
....// ./ \ \ ..... . .. / ./ ..
.... / ./. .\. \ .... ... / ./ \. \ ....
a ,, :u-x..\\. . a / \.
....// II‘ \ ...... .... / \ .....
x .xx m //(\ m

 

 

 

(GARDIGJ) AJJSNEINI EDNBDSBHOO'H

132

Monomer and exciplex fluorescence maxima are at 362 and
N500 nm respectively. Figure 5.2 shows that, maximum
exciplex fluorescence intensity is attained at N.09 M/L
of TEA, beyond which, quenching of exciplex emission
occurs, as well as that of monomer fluorescence; also,
past this optimal concentration, the isoemissive point
is lost.

Plots of intensity ratios of exciplex and monomer bands,
IE/IM, as a function of donor concentration, [TEA], are
given in Figure 5.3; the slopes of these lines, 5, are
given in Table 5.1. If IE/IM ratio is taken as a measure
of the extent of exciplex formation, then our results show
that the monomer model compound is more reactive than the
polymers; and among the polymers, the order of reactivity
is as follows: P—8 > P-6 z P-4 > P-2. Stern-Volmer plots
(Figure 5.4) show deviations from linearity for all the
compounds, at high donor concentrations. However, from
the initial slopes at low donor concentrations, higher
quenching constants, K are obtained for the polymers than
for the monomer model compound (Table 5.1). Plots for
P-2, P-4 and P-6 show approximately the same initial
slopes, with that of P-8 being higher.

Fluorescence decay times, TM and TE for monomer (in
absence of quencher) and exciplex, respectively, were
measured by single photon counting technique, and are

reported in Table 5.1. Our results show that the decay

133

 

...-a 23 .o-..
HOV ...-m CC .mlm HIV .0: HOV “pounce: .coHumnucoocoo .8200 .Ho
:oHuocsm a mo muHmcouCH mocoomoposHm LoEoCoE op onoHoxo mo oHumm .m.m opstm

o....\co.oE EM:
000 00.0 ¢0.0 n00 N00 .00

. . . _ . \

J‘IH

 

 

134

Table 5.1. Photophysical Parameters for Exciplex Forma-

 

 

 

tion.

TM kg TE kE
Compound (nsec) (10 3'1) (nsec) (1073-1) K 5
MC 19.60 5.10 11.90 8.40 25 14
P-2 19.20 5.20 14.20 7.00 N43 6.60
P-4 19.70 5.10 12.70 7.90 N43 W8.30
P-6 20.20 5.00 14.20 7.10 N43 N8.30
P-8 19.80 5.10 13.40 7.50 50 9.60

 

 

135

...-a .4. .o-.. .3 ...-a 3 .m-.. i. .2. .o. 3338..
.mcoHumpucoocoo msoHnm> um Locoo on» no aHv oocomopa on» :H

 

 

6C0 A Hv 0020000 05» CH 39H0309CH mofimomthSHh LOEOCOE MO Ofiuwm
92.3.9: 2m:
90. Nm. cN. 0.. 00.
1 q d _ — H — H
‘\\.
. \
O . \\\
O \
0
0
0
0
I
0

.:.m opstm

 

 

136

times for polymer exciplexes with TEA are on the same
order of magnitude; however, that of the monomer model
compound is slightly lower. Also, decay time (TM) of the
naphthalene moiety in absence of quencher remains essen-
tially the same in the polymers and the model compound.
We will attempt, in the following sections, to explain

these results.

Kinetics of Intermolecular Fluorescence Quenching and

Exciplex Formation

In donor-acceptor systems, in which exciplex formation
can be effected by selective excitation of one of the
components, the resulting quenching of monomer fluores-
cence may not be due to exciplex-formation process alone.
Evidence for this, comes from the fluorescent behavior of
some composite systems, with donor-acceptor moieties
separated by a chain. For example, Davidson and Trethe-
way96 (1976), have shown that monomer fluorescence quench-
ing is more efficient in naphthylbutylamine than in
naphthylpropylamine, despite the fact that the latter com-
pound forms a much more stable, and intensely fluorescent
exciplex, than the former. Also, in our work, IE/IM ratio
continues to increase as a function of quencher concentra-
tion, even when the exciplex intensity, IE has started
to decrease. Thus, it appears that in both of the above

two cases, monomer fluorescence quenching may involve

137

the formation of a non-fluorescent excited state complex,
in addition to exciplex formation.

Therefore, for a kinetic analysis of the fluorescence
quenching and exciplex formation occurring in our systems,

we propose the following mechanism:

 

 

 

(AT-«0) ‘ I:
—2'1£\4’[) '<r* «(IXHT37

“° /\ M

(4+0 13+th A+D A'l'D-H'wE

where A and D represent the acceptor and donor respectively.
(A-D+)* is the intermolecular exciplex, which we can
designate as E*; and (A*....D) is a transitory non-fluores-
cent CT complex, which we can designate as an encounter
complex (BC). In our case, A and D will represent naph-
thalene and triethylamine moieties respectively. We will
therefore refer to the fluorescence of A* as monomer fluor—

escence. The following rate constants can be defined:

kFM : fluorescence of monomer
kIM : internal quenching of monomer

kEM : exciplex formation

138

FE : fluorescence of exciplex
: dissociation of exciplex

kIE : internal quenching of exciplex

kl : encounter complex formation
k2 : dissociation of encounter complex
k : thermal de-activation of encounter complex.

For simplification we also define the following parameters:

= kFE + kIE

Under steady excitation with light intensity Ia einsteins

£‘ls-l, the rate equations are:

£551 = Ia + kZEEC] + kME[E*]
- (k1[D1 + kEMEDJ + kM)[A*l (5.3)
9%11 = kEM[A*JEDJ - (kn/1E + kEmm (5.14)
d 5? = k1[A*][D] - (k2 + kc)[EC] (5.5)

Assuming photostationary conditions

 

139

then

I

a
[A*]stat = x-(Y+z7fbj* (5‘6)

 

and

. ksmlatvl
[E Jstat 3 (kME+kE)tx-(y+z)ffiTT (5'7)

 

where Y = klk2/k2+kc, Z = kEMkME/kME+kE and X = [(k1+
kEM)[D]+kM]. The intensity of monomer and exciplex

fluorescence is given respectively by

1M = krMEA*J (5.8)
and

Putting Equations (5.8) and (5.9) into (5.6) and (5.7)

respectively, we get that

I

k
_ FM a
IM ’ x— Y+z D (5'10)

I a kFEkEMIaED] (5.11)
E (kME+kE)IX-(Y+Z)IDII

 

Then

I k k [D]
$2.. k F9kEM+k ) = gtD] (5.12)
M FM ME E

 

140

From Equation (5.10), we have that

 

 

1. . X-<Y+Z>EDJ (5.13)
IM kFMIa
so that
X-(Y+Z)[D]
(I /I) = (5.14)
o M kFM

where I0 is the monomer fluorescence intensity in absence
of donor (quencher)

Equation (5.14) can be simplified to

[klkc(kME+kE)+kEMkE(k2+kc)JED]

O +

 

(IO/1M)

l + KID] (5.15)

If we assume that kE >> kME and k0 >> k2 then Equation

(5.15) reduces to

klkEMED]

RM (5.16)

(Io/I)M = l +

 

Equation (5.16) is a Stern-Volmer-type relation, but with
an additional rate constant, k1, to account for quenching
of monomer fluorescence due to formation of non-fluores-

cent CT complex.

141

Discussion

The relative slopes of the lines in Figure 5.3 may be
explained in terms of Equation (5.12). The slope of each
line is then equal to kFEkEM/kFM(kME+kE). Since the plots
are within the optimum donor concentration for exciplex
formation, we can assume that kE >> kME; and since kFM
does not vary significantly among the polymers and the model
compound (Table 5.1) it is seen that kEM, (the rate constant
for exciplex formation), is larger for the model compound
than for the polymers. This greater reactivity of the
monomer model compound towards intermolecular exciplex
formation may be attributed to two main factors:

First, since exciplex formation is a diffusion-con-
trolled process, we may approximate kEM, the rate constant

for exciplex formation, bygu

kEM N 41rND(YRO)°103

where 7R0 is the effective action diameter, D is the sum
of diffusion constants for acceptor and donor molecules,
and N is avogadros number.

The slow motion of polymer molecules would lead to
comparatively much lower D values. Second, the steric
strain of the polymer chain would restrict the ability
of the naphthalene moiety to assume the optimum configura-

tion for effective exciplex interaction. The model compound,

142

on the other hand, is free to rotate and translate, and
thus can readily attain with donor molecules, the optimum
configuration necessary for exciplex formation.

Among the polymers, we noted that the order of re-
activity is P-8 > P-6 z P-4 > P-2. It appears that re-
activity increases with increase in the length of the methyl-
ene-chain in the polymer backbone. Increase in methylene-
chain length means increase in chain flexibility; conse-
quently, steric restrictions on the motion of chain
naphthalene moieties will decrease in the order: P-2 >
P-4 > P-6 > P-8; leading to the order of reactivity stated
above, and depicted in Figure 5.5.

The indication from the initial slopes of the Stern-
Volmer plots (Figure 5.4) that the quenching constant (K)
for the model compound is smaller than those for the polym-
ers, appears anomalous, on the basis of our earlier argu-
ment that the model compound possesses a much larger dif-
fusion constant than the polymers. However, from our
kinetic analysis of some of the physical processes involved
in monomer fluorescence quenching and exciplex formation,
one can see from Equation (5.15), that the quenching con-
stant, K, is a complex term, involving many rate constants.
And without knowing the values of kME’ kl and k2, it is
hard to assess the contributions of the diffusion-dependent
rate constants, such as kEM'

The observed deviations from linear Stern-Volmer plots

.2 mod HOV .2 mod iv "mimeu .nuwcoH .593 no :oHuocsm
a no zuHmcmucH mocoomoposHm 9050205 on onoHoxo no oHuwm .m.m opszm

2020.3. OZ_Zw>m.-..._._.Z_ no .022:
e. m. o. m o e. m

 

143

 

..0

_ _ _ _ _

—
-

 

 

 

144

at high donor concentrations seems to indicate a change
in the rate determining step for monomer fluorescence
quenching. The rate of collision of reactants may have
ceased to be the rate determining step; instead a state
of equilibrium may exist between donor and acceptor mole-
cules, leading to the formation of non-fluorescent CT
complexes; so that K may in fact be the equilibrium constant
for the CT complex. If that is the case, the greater
stability of the CT complexes with the polymer versus that
with the model compound may be due to a lesser degree of
dissociation in the case of the polymer.

Thus, in conclusion, it seems that use of monomer
fluorescence quenching as a measure g£_reactivity is not
dependable, when more than one process is involved $2.222

quenching. On the other hand, intermolecular exciplex

 

ing the extent to which the physical reactivity g£_a chemi-

 

cal moiety can be affected a§_a_result g£_binding in_a
polymer chain. Also, intermolecular exciplex interaction
involving a chain-bound reactive group appears sensitive

to the flexibility of the intervening chain.

CHAPTER 6

SOLVENT-INDUCED INTRAMOLECULAR EXCIPLEX FLUORESCENCE
IN 2,6-BIS(METHYLAMINO)NAPHTHALENES

Introduction

Studies on intramolecular exciplex formation have
focused on two main issues: the geometric requirements
and the role of the solvent.

In our consideration of intramolecular excimer formation
(Chapter 4) we saw that the observance of excimer fluores-
cence is dependent upon the length and chemical constitution
of the intervening chain, and upon the chemical nature of
the chromophore. Excimer interactiOn, because it arises
predominantly from exciton resonance, is maximized when
the interacting chromophores are in a sandwich configuration;
the attainment of this configuration as we saw, is opti-
mized when the chain between the interacting groups con-
sists of three atoms. Thus, the spectral position of
excimer emission depends only on the free energy of for-
mation, and the ground state repulsion, being generally
insensitive to external parameters, such as the polarity
of the medium. This is largely due to the fact that the
dipole moment of the excimer state is essentially zero.

However, the charge transfer (CT) character of the

145

146

exciplex state (with a net dipole moment) makes it highly
susceptible to solvation. Thus, the energies, fluores-
cence intensities and decay times of exciplexes are known
to decrease with increase in solvent polarity.i2lm23
The additional stabilization of the exciplex state via
solvation has led to the observance of intramolecular
exciplex fluorescence in polar media, under stringent
geometric conditions: Chandross and Thomas (1971)24,
studied naphthyléCHZAnamine~ system, in hydrocarbon solvents,

——.

and observed exciplex emission only in n=2-4. However,
recent reports by Davidson, et a196 and Ibemesi, et a173
show that naphthylmethyl amines (n=l), exhibit solvent-
induced intramolecular exciplex emission. A number of

studies,97'loo

have shown the same solvent effects on intra-
molecular exciplex emission in the system p-(CH3)2NC6Hu-
(CH2)n (9-anthryl or l-pyrenyl), in which only fluores-
cence emissions typical of 9-ethylanthracene (n=1 and 2)
and l-ethyl pyrene (n=l) are observed. However, in ether
solutions, exciplex fluorescence is observed, in addition
to the fluorescence of the excited aromatic chromophore.
In the aromatic hydrocarbon-aromatic amine systems, the
non observance of exciplex emission in hydrocarbon solvents
is not accompanied by quenching of monomer fluorescence.
However, in aromatic hydrocarbon-aliphatic amine

systems, there is considerable quenching of aromatic

(monomer) fluorescence in hydrocarbon solvents, even though

147

exciplex emission is not detectable. Sufficient experi-
mental evidence is lacking on the mechanism of this quench-
ing. The author and coworkers73 have recently proposed

the involvement of a charge transfer state, which may mix
with the locally excited singlet states, resulting in spin-
orbit coupling. This would make intersystem crossing com-
petitive with monomer fluorescence. Such proposition was
based on the observed decrease in the phosphorescence life-
time and enhancement of phosphorescence yield in B-naph-
thylmethylamine (B-NMA) relative to B-methylnaphthalene

120 had demonstrated

(B-MN). Earlier, McGlynn and co-workers
that charge transfer (CT) states resulting from inter-
molecular interaction in donor-acceptor systems produce

a decrease in the phosphorescence lifetime of the donor
molecule and an increase in the phosphorescence-fluorescence
quantum yield ratio , ¢p/¢F. Also, Toumon and El-Bayoumi121
(1972) made similar observations in benzyl derivatives
(such as benzylamine and benzyl alcohol), which showed an
enhancement of phosphorescence yields, a quenching of
fluorescence intensities and a shortening in their natural
phosphorescence lifetime, TS, relative to toluene. Gold-

schmidt et al122

have shown by flash photolysis, the
formation of triplet states in the charge-transfer quench-
ing of fluorescence of aromatic molecules.

The study of the fluorescence properties of 2,6-bis

(methylamino) naphthalenes, II-IV, which are disubstituted

148

symmetric analogs of B-NMA, is thus undertaken to provide
further experimental evidence for fluorescence quenching

via a CT-induced intersystem crossing. Also the effect

of N—methylation on the fluorescence properties of these
amines will help to expose more clearly the involvement of
the lone pair on the nitrogen in both the quenching of
monomer fluorescence and the formation of fluorescent intra-

molecular exciplex, under stringent geometric conditions.

”3 09 a N.

l
(CH3

' < )

2,6-Bis(methylamino)naphthalenes (II-IV)

Absorption and Emission Properties of 2,6-Bis(methylamino)

Naphthalenes, II-IV.

Room temperature absorption spectra of dilute solutions

4

(l x 10- M/L) of compounds I-IV in ethanol, are given in

Figure 6.1. The absorption bands appear very similar,
except there is substantial blurring of the long wavelength
(lLb) bands (N322-300 nm) in compounds II-IV, being most

1

pronounced in compound IV. Also, the intense La transition

149

 

.2: cam xoa
«HHH Ill-I «HH 0 o o

 

 

.2oom ca 2x2 e-oH x av >H -.-.-
.H II.- "mpuooam :oHquomnm ensue-89:6» .2002 .H.@ chasm-Hm
25.2
0v» ONn con ooN CON 0*» ON... 00» 00N 00N
H A). _ _ _ _ 2 2 2 _
/.2
II.) a
’2...
.2
....
2...
..
3 INC V
.2 .u
., mw
.. n
. 2 \
... x. m
2 . I .
a/)\ \ LVO
. \
. a
2 \ ...
’a\.\ 8 .... [0.0
10.0

 

 

150

max
ab

II-IV. A molar extinction coefficient of N5000 (at 290 nm)

band (A N272 nm) appears less resolved in compounds

is obtained for all four compounds in ethanol. The absorp-
tion spectrum of IV in a number of solvents of different
polarities are identical, with a select few displayed in
Figure 6.2. However, there is slightly more blurring of

1

the 1L bands in methanol, while the La band is slightly

b
better resolved in the same solvent.

Room temperature fluorescence spectra of dilute solu-
tions (1 x 10'“ M/L) of compounds I-IV in ethanol, are
shown in Figure 6.3. Compound I, as expected, gives
only the monomer (naphthalene) fluorescence. However,
significant differences are evident in the fluorescence
spectra of compounds II-IV:

i) Monomer fluorescence quenching occurs in all three
compounds to different degrees, in the order, IV > III >
II.

ii) A broad structureless band, due to an intramole-
cular exciplex, is present, the intensity of which, increases
in the order, IV > III > II. Consequently, the ratio of
intensities of exciplex to monomer emissions, IE/IM, is
highest in IV, followed by III, and least in II (Table 6.1).

iii) The exciplex emission maxima in the three com-
pounds, occur at different energies, in the order II > III >
IV, (Table 6.1).

iv) The emission spectra of II, III and IV, taken in

151

.Hocoon 2 we I.I.I
..oHHLpHcouoom III .ooHAOHno oonznuoE ... .ocmucod M2MWoEIm.III
.mu:o>Hom econommHo :H >H no meuomam coHuanomnm onsumpooEop Boom

 

 

 

26:24
2% can com com 8a 3» can can com com
HIIJ/2 _ _ 2 H 2 2 2
xi;
I.l/? (2 I do
1.),
2.22
2
.2
2
2
2
I no
2 s
2”2 x. l to
2 \
3.2
22 l ad
2
/ .\\/
IQ Joo

 

 

 

EDWGHOSSV

.m.w ohsmHm

152

 

 

 

 

 

 

/”‘.
e x \
1: ' \
2 ./ .
a I \.
). \
I'- '\
2 '\
1‘3 \
\
.2. '\
a-Q‘ \
\x \
\ ‘
\x \.
\\ \
\‘ \.
................ ‘\ ‘~ \.\
............ I ~~~=I "I .§.= '
550 650

 

WAVE LENGTFK nm)

Figure 6.3. Enhancement of intra-molecular exciplex emis-
sion on N-methylation: de I, ... de II,
-—-- de III, ----- de IV. Xex 290 nm, sol-
vent: ethyl alcohol.

 

153

 

 

 

02.m 20.0 20.2 mm.0 222 Hocssom 2oonvoe
0m.0 00.0 00.0 mm.0 can ceaeecacz-m 20cmvae
22.0 02.0 02.0 20.0 222 oeooeooo2-m mH\oH
0H.0 .00.0m 00.00 00.0m 00mm osoonoao2-m 2oonsva
200.0 m02.0 0m2.0 00m.0 00mm osmosoao2-m fie
00.m mm.0 m0.0 ----- oomm Hosanna 2meH
200 000 000 ----- 00mm Hosanna 2ssvxmw2
>H HHH HH H onsoonoasoe oaosHom mnaosmso

 

 

.mosonspnamc

IAocHEmHmsuo2vamIm.m mo moHuhoooem conmHEm co COHpmHhsuoEIz no 900002

.H.w mHnme

154

other solvents of high polarities (e > N15) show the same
trends observed in ethanol.

v) In moderately polar solvents (g < 10), only exciplex
fluorescence is observed in IV; while in III, it appears
as a shoulder, and virtually non-existent in II. Since
compound IV, gives the most intense exciplex emission, its
fluorescence spectra in solvents of different polarities
are given in Figure 6.4. Exciplex fluorescence maxima
(lgax) and quantum yields (0E) in these solvents, are given
in Table 6.2. The following trends are observed as solvent
polarity (dielectric constant) increases:

1) Exciplex emission maxima shifts to longer wave-
length.

ii) EXciplex fluorescence quantum yield is relatively
higher in moderately polar solvents ( e <10) than in highly
polar solvents (g > N15). However, there is a reverse
trend in 9E from methanol (8 = 32.70) to acetonitrile (e =
38.8) (Figure 6.5).

A plot of exciplex emission maximum versus solvent
polarity parameter is made (Figure 6.6) using Equation (2.53),

given below.

112 2
_ i 2(6-1) _ n -1

 

A dipole moment of 14.4D is calculated for the exciplex,

assuming a cavity radius of 4.8 A.

155

 

INTENSITY (relative)

 

Figure 6.4.

Ian) I

 

o\\\

I 1 . \‘Eaé - . . I-
450 550
WAVE LENGTH (um)

 

 

650

Fluorescence spectra of compound IV in solvents
of different polarities. l) 3 Me-Pentane,

2) Iso-propyl alcohol, 3) methanol, 4) methylene
chloride.

156

Table 6.2. Exciplex Fluorescence Quantum Yields and Emis-
sion Maxima in IV in Different Solvents.

 

 

max Amax

 

Dielectric M E ¢

Solvent Constant (nm) (nm) E
3-Mepentane 2.020 346 --- -----
Dioxane 2.209 ---- 458 0.020
Diethyl ether 4.235 ---- 429 0.015
Methylene chloride 8.93 —--- 455 0.017
l-Butanol 17.51 339 480 0.011
Ethyl alcohol 24.55 343 491 0.007
Methyl alcohol 32.70 339 498 0.004

Acetonitrile 38.80 --—- N500 0.010

 

 

157

m .w .pcmumcoo oHAuooHoHo
uco>Hom mo :oHuocsm o no . e .nHon Espcmzd cocoomoposHm onoHoxm .m.o onstm

w 2.23.0200 0_m_._.0w-_m=o

ow mm to. o. 0 o
_ _ . . _ _ . _ . _

 

[000.

 

 

 

 

158

 

hr: 22 ate-II al-

? (ac-I) 202-!

22.3 . ”6"”

22.0 -

 

 

20.0

 

 

0.6
Iflan)

Figure 6.6. eetermination of dipole moment of exciplex
IV .

159

From Figure 6.4, it can be seen that compound IV does
not give exciplex emission in 3-methyl pentane (a hydro-
carbon solvent), compound II and III behave similarly.
Monomer fluorescence quantum yield (d?) and decay time
(1%) of II-IV in 3-methylpentane are given in Table 6.1.
Values of o% and 1% increase in the order II > III > IV.

Total emission measurements were carried out at 77 K,
in 3-methylpentane; the observed fluorescence and phos-
phorescence spectra are shown in Figure 6.7. Compound I
shows extremely weak phosphorescence component; while the
total emission spectra of compounds II and III seem to
indicate the presence of a third peak, lying between the
monomer fluorescence and phosphorescence. Compound IV,
on the other hand, shows only monomer fluorescence and
phosphorescence bands, which are well-resolved. Ratios of
intensities of monomer phosphorescence to monomer fluores-
cence, Ip/IF, for all four compounds are given in Table 6.1
with magnitudes in the order IV > III > II > I. Phosphor-
escence life times determined in 3-methylpentane are in the
order: IV > III a II - I (Table 6.1).

Similar total emission measurements were done in ethanol.
The following observations are made:

i) Monomer fluorescence is absent in IV, weak in III,
relatively intense in II and very intense in I.

ii) Monomer phosphorescence increases in intensity,

in the order, IV > III > II > I. Accurate intensity ratios

160

as: cam ”04 Q>H ooooo OHHH coo. QHH nlnlll

"ocmucoathuoEIm :H gap» as caveman :onmHEo kuoe .>.m opsmHm

 

.H

22:22 Ikozw2m><3

0mm

oloHolL m 0 do n c \
alc/ I . e o n V
of.). . o .

 

 

 

 

 

(aAllDlaJ)kLISN3.LNI NOISSIWB W101

 

161

of monomer fluorescence and monomer phosphorescence could
not be taken because of the weak intensity of either one
or the other of the components.

iii) Results of monomer phosphorescence lifetime
measurements in ethanol (Table 6.1) show the following
increasing order: IV > III > II > I.

Exciplex fluorescence life times of compounds II-IV
in polar solvents, were too short (<2 nsec) to be deter-
mined by our single photon counting instrument, which has

a pulse width of N2 nsec.

Discussion

The blurring Of the lLb absorption bands, and the less
resolved nature of the 1La band in compounds II-IV, com-
pared to compound I, appears to indicate some ground state
charge transfer (CT) interaction between the aromatic group
and the amino group, despite the intervening methylene
group. These bands become resolved on addition of acid
or water (Figure 6.8), which tends to support the above
suggestion. Protonation and/or strong hydrogen-bonding
with water would destroy the donating ability of the nitrogen
lone pair.

However, N-methylation does not lead to any detectable
differences in the absorption spectra of compounds II, III

and IV, either in polar or non-polar media. Thus, the

162

 

 

 

 

 

 

Figure 6.8. Room temperature absorption spectra: 1) I,
2) III, 3) III + 90% H20, I) III + 1 drop of

CFBCOOH. Aex 290 nm.

163

observed differences in their emission spectra reflect dif-
ferences in their excited state properties and not ground
state interactions.

The enhanced exciplex intensity on N-methylation seems
to arise from the decrease in ionization potential of the
amines. If we approximate the exciplex energy, E(A'D+),

by Equation 2.23, given below:

E(A D ) = IPD — EAA - c

then compound IV with the least ionization potential, will
form the most stable exciplex, since the other energy terms
(BAA and C) are essentially the same for the complexes of
the three compounds. Thus the increase in exciplex fluores-
cence intensity on N-methylation clearly demonstrates the
involvement of the nitrogen lone pair in exciplex formation.
A further confirmation of this is given by the observed
total quenching of exciplex emission, on protonation of the
amino group.

The observance of exciplex emission only in moderate to
highly polar solvents could be explained as follows: Be—
cause of the asymmetric location of the donating orbital
of the amino group with respect to the accepting orbital
of the aromatic group, the stabilization energy of the CT
state due to coulombic forces may be small. However, since

the dipole moment in the CT state is large (14.4D for IV),

164

solvation in a polar solvent, may be very effective for
the stabilization of the CT state. Increase in solvent
polarity will increase the amount of stabilization by
solvation; so that exciplex energy will be correspondingly
lowered, hence the observed spectralshifts.

It appears that the dipolar character of the exciplex
of IV (u = 14.4D) is very similar to that of B-naphthyl-
methylamine, B-NMA (P = 14.6). The use of Equation (2.53)
in determining absolute dipole moments for these complexes,
is subject to error due to the following assumptions or
approximations:lOl

i) The radius of the solvent cage (a) surrounding the
complex is only an estimate, which could be in error by as
much as 20%.

ii) Equation 2.53 is derived on the assumption that
the electronic structure of the exciplex does not change on
solvation; however, solvent-exciplex interaction can occur
and will induce changes in the geometry of the complex.
Such specific interactions will cause scatter in plots of
the above equation, as may be evident in Figure 6.6.

ax values for two different complexes

However, if 52
are plotted against each other, the solvent-solute interac-
tions largely cancel, and yield relative values of u2/hca3,
which may be compared. This is shown in Figure 6.9 for
compounds IV and B-NMA. A slope of one is obtained, showing

that uz/hca3 (IV) = 112/boa3 (B—NMA); which seems to validate

165

 

24»

 

 

l l l l l 1 I
IS 20 22 24

IO“3 72 (cm")

 

Figure 6.9. Plot of maximum of exciplex emission for IV
against maximum of exciplex emission for
B-NMA.

166

the absolute values obtained for their dipole moments.
The decrease in exciplex fluorescence quantum yield at
high solvent polarities (Figure 6.5) may be due to formation

of non-fluorescent solvated ion pairs from the encounter

complex (with rate constant kq) and from the exciplex (with

rate constant kI)’

At“... D +1——

% \

(A‘o‘f "I a 1:3-«o; ——"fi—> A; + o;
excipiex solvm‘ed ion pair saivaieaian rodiools

encounter complex

 

According to this mechanism,102 illustrated above, kq

and RI are assumed to increase with increase in solvent
polarity. The exciplex quantum yield will be proportional
to kEM/kEM + kq. In intermolecular exciplexes, ion radical
formation has been reported in many Laser photolysis

studies,2]"22’lo3'107 in

which absorptions corresponding
to the ionic dissociation products A; and D; have been
detected. It has also been shown that the dissociation

into ions occurs from the non-relaxed CT system (encounter

167

complex) before the relaxed exciplex state is formed.108
However, such ion radical formation is unlikely in intra-
molecular exciplexes; this probably accounts for the fact
that the latter generally exhibit higher fluorescence
quantum yields in polar media than their intermolecular
counterpart. For example, the intramolecular exciplex of
compound IV has a quantum yield of 0.004 in methanol, but

the intermolecular exciplex of naphthalene and triethyl-
amine is not observed in the same solvent.109

The initial increase in exciplex quantum yield (Figure
6.5) in moderately polar media seems to reflect the solvent-
induced nature of exciplex fluorescence in these amines;
thus indicating that a certain solvent polarity is necessary
for optimum exciplex stability.

The relatively lower quantum yields observed in ethyl
alcohol and methyl alcohol, compared to acetonitrile, seem
to indicate an effect arising from some solvent property
other than polarity. For example, specific interaction be-
tween the hydroxylic solvent molecules and the amino group
could reduce the reactivity of the lone pair in CT inter-
action, with the excited singlet aromatic group. There is
also the possibility of quenching of the exciplex fluores-
cence by specific interaction of the exciplex with hydroxylic
solvent molecules.

The high quantum yield of exciplex fluorescence in

dioxane, despite its low dielectric constant, again suggests

168

some form of specific interaction between solvent molecules
and the exciplex; the high electron density on dioxane
oxygens could exert some stabilizing effect on the exci-
plex, due to the CT character of the latter. A detailed
study of the specificity of exciplex interaction with polar
molecules is undertaken in the next chapter.

On the whole, exciplex fluorescence quantum yields in
IV are quite low, and with very short life time (which
could not be measured by our single photon counting set-
up); all this indicates that intramolecular exciplex forma-
tion in IV as well as in II and III, is very inefficient,
due to geometric restrictions imposed by tetrahedral bond
angles. The overlap of the nitrogen lone pair orbital
with the w-orbital of the ring is very poor.

The observation that monomer fluorescence quenching in
a hydrocarbon solvent increases in the order IV > III > II
(which is the same order of increase in exciplex intensity
in polar solvents) appears to suggest the involvement of
the same molecular intermediate in both monomer fluorescence
quenching and exciplex formation. This intermediate is
probably an encounter complex, with a partial charge trans-
fer character, and an energy that lies close to that of the
locally excited single state. In a hydrocarbon medium,
the encounter complex is relatively high in energy due to
lack of stabilization by solvation; but it may mix with

the locally excited singlet state, leading to enhanced

169

spin-orbit coupling. The latter, in turn, leads to an
increase in intersystem crossing to the triplet state -

process 2 in the scheme below.73

At room temperature, the resulting excited triplet state

will however undergo thermal dissociation. But at 77 K

Ammo

 

 

 

1(Axum? < 1 (1A?‘°°°D) 2 > (flung)

3 . 4 5
v v

Ammo +in limo-rmM Amo

 

 

(liquid N2 temperature), if intersystem crossing (process
2) is competitive with monomer fluorescence (process a),
then from Equations (6.1 to 6.3),110 one would expect to

observe an increase in the phosphorescence/fluorescence

ratio, ¢p/¢F.

ap/aF . kp‘rp(-5];- - 1) (6.1)

170

(6.2)

(6.3)

where

T = observed phosphorescence lifetime

'6

kp = rate constant for phosphorescence

km = rate constant for thermal deactivation
of excited triplet state
kg = rate constant for intersystem crossing

kF a rate constant for fluorescence
k = rate constant for internal conversion of first

excited singlet state.

Total emission measurements at 77 K, carried out with
compounds I-IV in 3-methylpentane, show that ¢p/¢F, has
increased in II, III and IV by a factor of lfl, l6 and 17
respectively, relative to I. From Equation (6.3), it is
clear that the observed decrease in monomer fluorescence
quantum yield at 77 K, must be due to an increase in kg,
since kn will in fact be negligibly low. Similar phos-
phorescence enhancement is observed in ethanol, with com-
pound IV exhibiting only phosphorescence emission; in-

dicating that intersystem crossing is predominant in de—

populating the excited singlet state. Some degree of

171

stabilization of the transient CT state by the frozen polar
solvent molecules may lead to greater mixing with locally
excited state.

The observed increase in phosphorescence life times,
for example, in IV compared to I, appears contrary to
expectation; because enhancement of intersystem crossing
will lead to an increase in kp, and hence a decrease in

I However, from Equation (6.2), it is possible that the

p.
increase in kp may have been offset by a decrease in kM.

Viscosity Dependence of Exciplex Emission in III and IV

The fluorescence emission bands of compounds III and
IV are similarly affected by the viscosity of the medium.
The case of IV is shown in Figure 6.10. The following

observations are made for both compounds:

1) There is a gradual quenching of exciplex fluores-
cence, with corresponding increase in monomer fluores-
cence, as viscosity of the medium increases. However, the
exciplex emission is not completely quenched even in 100%
glycerol.

ii) As the glycerol content is increased, there is a
slight shift to the red in exciplex emission maxima by

6.5 nm in III and mh.S nm in IV.
iii) A plot of the ratio of intensities of monomer

and exciplex fluorescence, IM/IE against viscosity of the

172

 

 

 

 

 

4&3) '

'5
a
.2
5
E I.
m D
p...
25

‘0 ....

\ z
\.°'
\, __
350

 

WAVE LENGTl-Knm)

Figure 6.10. Effect of viscosity on fluorescence spectra
of compound IV. Solvent: Ethanol/glycerol
mixture. % Glycerol: l) O, 2) 5, 3) no,

u) 80; Aex 290 nm.

173

medium (Figure 6.11) is linear in the high viscosity region
(n g 30 cp), but curved towards the abscissa at low vis-
cosities, making a finite intercept.

The above results can be rationalized, using the same
kinetic expression (Equation 4.13) obtained for the case

of intramolecular excimer formation.

kFM(kE+kME)

I /I .
M E kFE'kEM

 

where

kE ' kFE + 1CIE

IM I monomer fluorescence intensity

IE I exciplex fluorescence intensity

FM I rate constant for monomer fluorescence
kEM I rate constant for exciplex formation
kFE I rate constant for exciplex fluorescence
kME I rate constant for exciplex dissociation

kIE I internal quenching of exciplex.

The viscosity dependent terms are kME’ and REM’ and if
k5 >> kME’ then IM/IE should be proportional to k-1 .
But kEM is the rate constant for the conformational change
involved in exciplex formation, so that IM/IE should be
linearly related to the viscosity of the medium.

The observance of a linear plot at high viscosities

seems to conform to the above kinetic analysis. Although,

17’4

.opzuxfia
Hopoozam\aocw no :« >H ccsodsoo .EzacoE ho anamooma>
manpo> oauap . H\2H .aaamzouca posaoxo\noso:os ho uoam .Ha.c opsmfim

as E835.

9. .8 oh .5 on o. iu~
_ d T _

 

 

 

 

 

175

considerable conformational transitions may not be required
to attain the exciplex configuration in these naphthyl-
methyl amines, some rotation about the Cal'Car bond may
be necessary to bring the lone pair orbital into closer
proximity with the ring orbital; such a rotational transi-
tion will be viscosity dependent. The non-linearity in
the plot at moderate to low viscosities seems to indicate
a more complex dependence, in which the assumption -
kE >> kME may not be valid. The observance of a finite
intercept on IM/IE axis reflects some degree of insensivity
of intramolecular exciplex formation in these amines to
low macroscopic viscosities.

The slight shift in exciplex emission maximum may be
due to some change in dielectric constant of the medium
at high glycerol content. It is also possible that some
of the quenching of the exciplex emission may arise from

both polarity effect and specific interaction with the hy-

droxyl groups of glycerol.

Quenchingiof Exciplex Fluorescence in II-IV by Water

Quenching of intramolecular exciplex fluorescence in
compounds II, III and IV is carried out with the following
objectives:

1) To dramatize the effect of hydroxylic solvents on
exciplex emission, with a view to revealing the specificity

of the interaction of such solvent molecules with exciplexes.

176

2) To examine the relative basicity of N-methylated
amines, and hence the electronic and geometric effects of

N-methylation, by a bimolecular fluorescence quenching.

The absorption spectrum of dilute solution (1 x 10"}4

M/L) of III plus 90% water by volume, is given in Figure 6.8.
As noted earlier, addition of water leads to the resolu-

tion of both the lLb and 1

La bands; which suggests that
some specific ground state interaction occurs between the
water molecules and the amine molecules. Compounds II
and IV behave similarly.

Exciplex fluorescence in IV, in the presence of vary-
ing amounts of water, is shown in Figure 6.12. In IV,
as well as in III and II, exciplex emission is gradually
quenched with increasing amounts of water; on the other
hand, monomer (naphthalene) fluorescence is enhanced.
Quenching of exciplex fluorescence is complete on addition
of us% H20 in III and 85% H20 in IV; in II, there is residual
exciplex band above 85% H20.

Plots of the ratio of exciplex fluorescence intensity
in absence (lg) and presence (IE) of water, as a function
of concentration of water, are given in Figure 6.13.
The following Stern-Volmer quenching constants (mole-l
litre) are obtained: 0.068 (II), 0.330 (III) and 0.209

(IV). Also, plots of ratio of monomer fluorescence inten-

sity in presence (IM) and absence (Ifi) of water, as a function

177

 

 

 

 

 

 

i
0
§
:3
“‘ l
h
55 z
-2.
a ‘\ '°
/’ “s‘
.~o-o-C-.4-O-.- “\
’1' 'L- "a ' 650

WAVE LENGTl-Knm)

Figure 6.12. Effect of water on fluorescence emission
of compound IV. 2 320 (volume): 1) 0, 2) 5,
3) 15, A) 50, 5) 85. lex 290 nm, solvent:
ethanol.

178

am .aanoafla Hanna
35:8 :5 cam .x .5 can 3 .HHH can 53 .2 can A!
.uoaq hoEHo>icpoum «nous: an cofimmasm xoanfioxm go wcanocmzo .mH.m madman

. usph< =8: . O «I

'1‘

2 o e
.. _ _ _

 

 

 

179

of concentration of water are given in Figure 6.14 (IV)
and Figure 6.15 (III).

Quenching of exciplex fluorescence in compounds II-
IV, by water, may proceed by two mechanistic path ways,

represented in the scheme below:

 

 

 

(Edie-L130 E'rg—e—(Eoi’

 

-A+Q th A+ Q

1) Static Quenching.

This will involve the formation of a complex (AQ)
between a ground state molecule (A) of the amine, and a
water molecule (Q). This complex, on excitation, can no
longer form an intramolecular exciplex, because the lone
pair on the nitrogen is bound to water molecule; this will
lead to a decrease in exciplex fluorescence intensity, and
the observance of a Stern-Volmer relationship. However,

the quenching constant will represent the equilibrium

180

 

45i-

til?

35-

25-

 

   

 

 

! l I l l
O 20 4O 60 80 IOO

% H20 (by volume)

 

Figure 6.1“. Water-induced enhancement of monomer emis-
sion in IV. Plot of IM/Ifi versus % H20
(volume) in ethanol. ~

181

 

20.—

 

 

 

o n
o 20 4o
7.1-120 (by volume)

 

is aa—

Figure 6.15. water-induced enhancement of monomer emission

in III. Plot of IM/Ifi versus 1 H20 (volume)
in ethanol.

182

constant for the complex formation between A and Q. The
excited complex (A*Q) will emit monomer fluorescence; so
that a decrease in exciplex fluorescence intensity will
be accompanied by an increase in monomer fluorescence

intensity. Also, exciplex fluorescence decay time will

be unaffected.

2) Dynamic Quenching

This will proceed by a bimolecular collisional mechanism,
involving an exciplex moiety (E) and a water molecule
(Q), to form a transient CT complex (EQ)*, which undergoes
thermal dissociation to the ground state. The quenching
will follow a Stern-Volmer relationship; exciplex fluores-
cence life time will decrease, and monomer fluorescence
will not be enhanced.

The two mechanisms may contribute to the observed quench-
ing of exciplex fluorescence by water. However, static
quenching (1), appears predominant at high concentrations
of water (> N20%). This is supported by the observed rise
in monomer fluorescence. Below 20% of water, the quenching
appears to proceed more by dynamic quenching (particularly
in III), as evidenced by the constant value of monomer
fluorescence intensity.

The basicity of aliphatic amines follows the order:
secondary > tertiary > primary. According to this, com-

pound III is the most basic, followed by IV and then II.

183

The Stern-Volmer quenching constants observed for these
compounds follow the order of their basicity. This seems
to support the predominance of static quenching of exciplex
fluorescence by water. Further evidence for this comes
from the effect of water on the absorption spectra of these
compounds. Addition of equal amounts of water (50%)

to their solutions in ethanol intensified the 1

La absorption
band in the order: III > IV > II. Also the observed
resolution of the lLb bands correspond to the above order.
The lack of significant spectral shift of the exciplex
band on addition of large amounts of water(with increase .
in dielectric constant of the medium) lends support to the
smaller role of dynamic quenching. Exciplexes undergo
spectral shifts to the red, with increase in dielectric
constant of the medium.

Consequently, the observed quenching constants may
represent the equilibrium constants for proton exchange
between water molecules and ground state amine molecules;
in which case the relative basicity of these amines appears

to have been established by exciplex fluorescence quench-

ing.

Conclusion

The fluorescent behavior of naphthyléCHzéNRR' system
(II, III and IV) reported in this chapter have provided

further experimental support to the recent findings that

l8u

intramolecular fluorescent exciplexes can be formed under
stringent geometric conditions (n=l), in polar solvents,
due to increased stabilization by solvation. Effects of
N-methylation in these amines have helped to demonstrate
the involvement of the nitrogen lone pair in both intra-
molecular exciplex formation in polar solvents, and quench-
ing of monomer fluorescence in hydrocarbon solvents. The
latter is shown to proceed via a charge-transfer-induced
intersystem crossing enhancement. Quenching of exciplex
fluorescence by water appears to typify the specificity

of interaction of exciplexes with hydroxylic solvents, and
probably other polar solvents. The latter will be reported

in the next chapter.

CHAPTER 7

INTERACTION OF AMINE-HYDROCARBON EXCIPLEXES
WITH POLAR MOLECULES

Introduction

The fluorescence energies, intensities and decay times
of exciplexes are known to decrease with increase in solvent
polarities.2l'23 The decrease in intensity and life time
has been attributed to the formation of solvated ion pairs,
and/or dissociated radical ions. The red shift in the
exciplex emission band with increasing solvent polarity
is generally rationalized on the basis of Born solvation
of the exciplex. Thus the exciplex energy in solution
(Equation 2.53) is believed to depend on solvent polarity
parameters - bulk dielectric constant and refractive index.
The derivation of the above energy expression is based on
Onsanger's continuum model, which envisages the exciplex
at the center of a cavity surrounded by a dielectric con-
tinuum.

This model excludes the possibility of any specific
interaction between the exciplex and polar solvent mole-
cules. However, since exciplexes are highly dipolar in
character, with dipole moments in the range of lO-30D,

it is therefore, highly probable, that in some polar

185

186

media, solvent molecules closest to the exciplex moiety,

can interact with it in a specific manner, involving di-
pole-dipole and dipole-induced dipole forces and/or hydrogen-
bonding (for protic polar molecules). Generally, the oc-
currence of such specific interactions are said to lead to
scatter in the plot of exciplex emission maximum as a
function of polarity parameters (Equation 2.53).

However, E. A. Chandross, and others tend to attribute
the red-shift of exciplex emission band with increasing
solvent polarity, mainly to specific interaction of the
exciplex with individual solvent molecules, rather than to
a bulk_dielectric continuum. Their contention is based on
the experimental observation that addition of small amounts
of polar solutes to a hydrocarbon solution of an exciplex-
forming composite system, causes: (a decrease in exciplex
fluorescence intensity and a red-shift in exciplex spectral
position and a change in exciplex emission band shape.

For example, Chandross and Thomas111 (1971), studied
the interaction of intramolecular exciplex of naphthyl-
propyl amine (in benzene) with polar molecules such as
propionitrile (e I 27), hexamethylphosphoramide, HMPA
(e I 30) and dimethyl formamide, DMF (e = 37). Analysis
of the observed quenching of the exciplex fluorescence gave
a linear Stern-Volmer relationship at low concentrations of
added polar solute. This was believed to be indicative

of a 1:1 interaction of the exciplex with a polar molecule.

187

Also HMPA was found to possess a much higher quenching
constant than propionitrile despite the closeness of their
dielectric constant. The latter result was attributed to
the greater solvating power of HMPA towards a cation.
Furthermore, the observance of a change in band shape of
the exciplex emission, though at fairly high concentrations
of the added polar species (e.g., 5.67 M for HMPA) led
Chandross and Thomas to suggest the presence of a new
emitting species resulting from a 1:1 interaction between
the exciplex and a polar molecule.

However, Selinger and McDonald112 (1971), argued that
Chandross's observations are merely due to a change in the
dielectric constant of the non-polar solvent as a result
of addition of polar molecules. In other words, that the
quenching and shift in exciplex emission maximum will not
occur if the dielectric constant of the medium remains
constant on addition of the polar species. To prove this
point, they (Selinger and McDonald) examined the effect
of excess diethylaniline on the intermolecular exciplex
of anthracene and diethylaniline in cyclohexane and ethyl-
acetate as solvents. Exciplex fluorescence quenching was
reported in the former solvent, but not in the latter, in
which the dielectric constant of the medium was kept
constant. However, the validity of the observation in

113

ethylacetate was challenged by Chandross, who pointed

out that ethylacetate, being a moderately lewis-basic

188

solvent, may have overwhelmed the added polar species by
virtue of its much higher concentration; in which case,
"levelling effect" appears to account for the non-obser-
vance of exciplex fluorescence quenching in ethylacetate.
To buttress his earlier suggestion that dipolar ex-
cited states (e.g., exciplexes) do interact in a specific
manner with surrounding dipolar species, Chandross113
studied the fluorescence properties of 9-(A-dimethy1amino-
phenyl)anthracene (for brevity, anilinoanthracene) in methyl-
cyclohexane and toluene. This compound, in the above sol-
vents, shows a fluorescence band typical of 9-phenylanthra-
cene. However, upon addition of small amounts of N,N-
dimethyl formamide (DMF) or O-dichlorobenzene (DCB), the

fluorescence spectrum was shifted to the red. The follow-

ing differences were observed:

1) in methylcyclohexane, adding different amounts of
DMF (e I 37) and DCB (e I 9.9) to give the same bulk di-
electric constant, showed DMF to cause a much larger red-
shift, and also a more pronounced change in the fluores-
cence band shape.

ii) adding the same amount of DMF to the solutions of
anilinoanthracene in methylcyclohexane and toluene, to
give different bulk dielectric constant, a larger red-
shift of the fluorescence spectrum was found in methyl-
cyclohexane than in toluene, despite the lower bulk di-

electric constant of the solution in the former solvent.

189

The first observation is believed to be due to a weaker
interaction of the exciplex with DCB. In the second ob-
servation, toluene appears to exert a levelling effect,
competing more effectively with DMF in solvating the excited
molecule than does methylcyclohexane; however, the observed
larger red-shift in a solution with a lower bulk dielectric
constant appears to reflect some specific interaction be-
tween the excited molecule and DMF.

Although Chandross's results seem convincing in showing
that the observed red-shift in exciplex band in the
presence of small amounts of polar molecules may be due
to some specific interaction, the accompanying decrease in
exciplex fluorescence intensity is not explicable on the
basis of formation of a new emitting species, believed to
arise from a 1:1 interaction of the exciplex with a polar
molecule. Dissociation to ions is unlikely to be caused
by one polar molecule when the surrounding medium is ali-
phatic hydrocarbon. Besides, the reported change in ex-
ciplex band shape occurs at fairly high concentrations of
added polar molecule (>1 M); in this region, effects due to
bulk dielectric constant may not be completely ignored.
Also, the failure to detect isoemissive points in the emis-
sion spectra of most of the studied exciplex-polar molecule.
interactions, appears to weaken the claim for a new complex
formation.

Recently, Beddard et alllu reported the quenching of

monomer and intramolecular exciplex fluorescence in

190

1-(2-naphthy1)-l-hydroxy-3-(N-piperidinyl)-propane (I)

on addition of small amounts of ethanol and acetonitrile

to a solution of (I) in cyclohexane. A red-shift was
observed in both the steady state and time-resolved emission
bands of the exciplex. Values for the various excited

state rate constants (including that for a 1:1 exciplex-
polar molecule complex formation) were obtained. Thus,

the suggestion was made for a one polar molecule fluoro-
phore interaction. However, the exciplex band was not
resolvable into two components of solvated and unsolvated
species; nor was there any isoemissive point in the exci-
plex spectra as a result of addition of polar molecules.

In any event, the use of system (I) may have introduced

some unknown quantities in the interactions, because:

there is evidence that it interacts in the ground state with
the added polar molecules; also, in a hydrocarbon medium,

it shows a two-component emission - a monomer and an ex-
ciplex band, both of which are quenched by the added polar
molecules.

The foregoing studies by Chandross and Thomas, and
Beddard et a1 represent, to our knowledge, the only pub-
lished work on the subject of "stoichiometric" complex
formation between an exciplex and a polar molecule. Though
their suggestion for a 1:1 interaction has not been clearly
demonstrated experimentally, their results seem to have

provided a strong basis for more investigation of the

191

subject. Besides, in their studies, little or no attempt
was made to correlate the observed specific interaction with
other properties of the polar molecule other than its bulk
dielectric constant. For example, its donicity, which is

a measure of the nucleophilic properties of the polar
species, is expected to play a major role in solvating the
positive end of the exciplex. Donicity is defined as the
negative AH—value in kca1°mol"l for the 1:1 adduct formation
between the reference acid SbClS and the donor in the quasi-
inert solvent 1,2-dichloroethane.115 It has been shown that
the dissociation constants of some ionic compounds in polar
solvents of similar dielectric constants are dependent on
the donor number (DN) of the polar solvent. For example,116
the dissociation constants of quinuclidinium chloride
(QHCl) are much lower in acetonitrile (e I 37.5, DN = 14.1)
and nitromethane (e I 36.67, DN I 2.7) than in N,N-Dimethyl-
acetamide, DMA (e I 37.78, DN 3 27.8) and N,N-Dimethyl-
formamide, DMF (e I 37, DN I 26.6) even though these solvents
have similar dielectric constants. 0n the other hand, QHCl
is equally strongly dissociated in DMF or DMA as in propan-
ediol-1,2-carbonate, PDC (e I 65, DN I 15.1) even though

the latter solvent has a much higher dielectric constant

and dipole moment. It is true that the dissociation

constant of QHCl will also depend on the electrophilic
property of the polar solvent; however, the observation

that the dissociation constant of QHCl in acetonitrile

192

is considerably lower than in DMF (the former being a stronger
electrophile, but a much weaker donor) appears to indicate
that the influence of the nucleophilic effect is dominating.

However, the dissociative tendency of QHCl was found
also to depend on the dielectric constant. It exhibits a
higher dissociation constant in PDC than in acetonitrile
although both solvents have essentially the same nucleo-
philic and electrophilic properties. This was ascribed
to the higher Born solvation energy contribution in PDC
which has a considerably higher dielectric constant.

In the case of an exciplex in a low dielectric medium,
to which is added small amounts of a dipolar species,
'local' dielectric effect may contribute to the red-shift
in exciplex emission band. This is because in solutions
of polar substances in non—polar solvents, the attractive
forces existing between dipoles are known to lead to the

117,118 Such

formation of aggregates of polar molecules.
aggregates, if formed around the exciplex may create "local"
dielectric effects.

Thus, in this chapter we will study the interaction
of the intramolecular exciplex of 2,6-Bis(N dimethyl,
methylamino)naphthalene, IV, with a number of polar
species in order to examine the involvement of specific
interaction and/or "local" dielectric effect in red-shifts

generally observed in exciplex emission maximum as polarity

of the medium is increased. An attempt will be made to

correlate exciplex spectral shifts with the donicity

 

f the polar species in order to distinguish effects due to

a 4 43 - . A o I
spec--ic inoeraction ENC polarity parameters. The solvent—

 

induced -iuoresceut nature of the exciplex of IV, may be
useful in making the above distinctions. Also, the close-
ness between the ground state and excited state geometries

of IV may hei. to simplify the interaction of the exciplex

with the polar molecules

Exciplex Fluorescence Spectra of 2,6-Bis(N-dimethy1,

methylamino)naphthalene, IV, In Presence of Small Amounts

Small amounts of polar species were added succes-
sively to a dilute solution (1 x 10’“ M/L) of IV in
ether; the room temperature absorption and emission spectra
were obtained after each addition. The following polar
species were used: ethyl alcohol, methyl alcohol, dioxane,
methylcyanide (MeCN), N,N-dimethyl formamide (DMF), di-
methyl sulfoxide (DMSO) and hexamethylphosphoramide (HMPA).

The addition of any of the polar molecules caused no
detectabl- change in the absorption spectra of IV in
ether as solvent.

However, with the exception of dioxane, the addition
of small amounts of ethyl alcohol, methyl alcohol, MeCN,

II‘ In

DMF, DMSO and nnrA, caused significant red-shifts in the

«If?

exciplex emission maxim m in it. This is

.4.
Q.

llustrate

19A

in Figure 7.1 for the case of DMSO. The isoemissive point,
observed in all cases, does not seem real since it shifts
to longer wavelength with increase in concentration of the
polar molecule.

No change in exciplex band shape is observed; instead
there is a more or less symmetrical broadening of the peak.

Also, exciplex fluorescence quantum yield (0.011)
stays constant on addition of small amounts of polar mole-
cules, for example ~0.05 M (DMSO), mo.2u M (DMF) and
~0.10 M (MeCN); then it decreases only slightly at higher
concentrations.

The extent of the red-shift of exciplex emission band
is dependent on the nature of the polar molecule. We will
compare the interactions, involving only nonprotic polar
species, in order to eliminate effects due to hydrogen
bonding. Thus a comparison of exciplex emission spectra
(Figure 7.2) obtained by adding the same amount (0.05 M)
of polar species, shows a red shift which increases in
the order: HMPA > DMSO > DMF > MeCN. However, as the
concentration of the added polar molecule increases, the
amount of red-shift caused by DMSO becomes slightly larger
than that caused by HMPA (Figure 7.3). It may be noted
that acetonitrile causes no shift in exciplex band for
concentrations less than m0.02 M. The effect of dioxane
is very slight - causing a red shift of ~3.6 nm at a

concentration of 0.A M.

195

 

 

 

 

 

 

 

 

 

FLUORESCENCE INTENSITY (RELATIVE)

 

 

 

 

3E0 450 . 580 sec
WAVELENGTH (nm)

Figure 7.1. Fluorescence spectra of IV in ether, in the
presence of varying amounts of DMSO (moles/
litre): -—-0.0 (a), 0.11 (b), 0.31 (c);
-- 0.02 (a), 0.16 (b), 0.56 (c); ----- o.ou
(a), 0.21 (b), 0.72 (c); ------- 0.07 (a),
0.26 (b), 0.9a (c); ... 0.09 (a), 0.31 (b),
1.18 (c). Aex 290 nm.

196

 

. . .omzo iiiii .zoa: .
as: in- <m=:.iii ”ooflaooflos oofloo scapegoao no Aux: ma.ev ooaoso

came on» no cocomono can :« .sozao :« >H mo mhuoodm cocoomoposam

A65 Ihwzmqm><>>

 

000 One. 0mm
.-.- irt:5:kmu. _ d _ .\\ a
f... \
,/../... \..
,/.10u m
x.../... .\..
/. \x...
.//,fw \Quu
//JJ/ \\.m
/ ...../ \\.\ ...»
./....//I.\ .\ ..
./...... \

.m.> opzmfim

(BALLV‘IEH) MlSNEiNI BONSOSEHOO'H

 

 

197

.282 f; .923 is .22: 3 .mzn 29 "3.839..
Laaoa no coaumhacoocoo ho coauocsu a an sauna Hmpuoodn xoadfioxm .m.> opzwqm

.mmtémmjoé 38.3.0.2 mica do 20.252828

mm. mm. mm. In. om. 9. ~.. 8 woo
_ _ q _ _ _ _ . o

 

VN .
o o :25

 

 

 

198

A plot of exciplex spectral shift as a function of donor
number of the polar molecule is given in Figure 7.4. At
relatively low concentrations of added polar molecule
(30.05 M), the red shift in exciplex emission maximum
appears linearly dependent on the donor number; thus the
extent of shift increases in the order HMPA > DMSO > DMF >
MeCN. However, at relatively higher concentrations (for
example 0.29 M), some deviation to the above linearity
occurs for polar molecules that differ widely in di-
electric constant; hence the observed red shift by DMSO
becomes greater than that of HMPA.

A plot of exciplex spectral shift as a function of
bulk dielectric constant is shown in Figure 7.5. The bulk
dielectric constant, e, of the medium (mixture of the solvent

and added polar molecule) is estimated from the expression:113

iei (7.1)

where Vi is the volume fraction of solvent (ether) or the
polar molecule, and Si is the dielectric constant of the
pure solvent or pure polar molecule. Figure 7.5 shows that
at a given bulk dielectric constant, the magnitude of the
red-shift increases in the order HMPA > DMSO z DMF > MeCN.
It is to be noted also that at bulk dielectric constant

of A.A0 for the mixture (8 = 4.335 for pure solvent) HMPA

causes a finite red shift of N2.“ nm, while DMSO, DMF and

199

.mm.o A£V .mo.o AOV “Aonu:\moaosv oasooaoe pmaod .3 20:93:06
isoo .mnonssc pocoo no soapocsm a no pagan Hmauoodm xoaqfioxm .:.> onsmfim

A29 mum—>52 mozoo

oo an em o. m
rues.

 

_ . _

ES

 

 

 

 

200

.23: E .85 iv .EE 6V Jazz 28 "sauces . . .
mo ucmumcoo capaooaoac mo coauoczm m an awash Hmpuoodm xoadaoxm m u shaman

u ...—24.5200 UEHUMJME

V.@ 0.0 0 0 ND 0.? 9.4V
_ _ _ _ _ AUG

 

 

  

«N

585
26
co

 

 

 

201

MeCN cause no detectable shift. Above 6 = A.AO, all but
MeCN bring about spectral shift, the latter showing de-

tectable red shift from a m U.5.

Onset and Growth of Fluorescence of Intramolecular Exci—

plex of IV

 

We saw in Chapter 6 that compound IV gives intramole-
cular exciplex fluorescence in moderate to highly polar
solvents, such as diethylether, methylene chloride,ethyl
alcohol and acetonitrile; while in hydrocarbon solvents
only monomer emission is observed, though with substan—
tial quenching. However, compound IV does give an intense
intramolecular exciplex fluorescence in dioxane (g I 2.209).
Thus, the non-observance of exciplex fluorescence in
hydrocarbon solvents, such as methylcyclohexane (e I 2.02)
seems not to be dependent on solvent polarity parameters
alone. Since compound IV acquires a dipolar character in
the excited state due to charge transfer (CT) interaction,
it could therefore interact in a specific manner with sur-
rounding polar solvent molecules. Such an interaction
may be revealed by studying the emission spectra of a
hydrocarbon solution of IV in the presence of small am-
ounts of a polar molecule.

The above experiment was carried out, using methyl-
cyclohexane as the solvent, and N,N-Dimethylformamide (DMF),

methyl cyanide (MeCN) and Dimethylsulfoxide (DMSO) as the

202

polar molecules. Addition of small amounts of DMF and
MeCN led to the appearance and growth of intramolecular
exciplex fluorescence in IV, and quenching of monomer
fluorescence. The polar-molecule-induced exciplex emis-
sion undergoes a red shift as concentration of polar mole-
cule is increased. The case of DMF is shown in Figure 7.6.
Addition of one drop (mo.08 M) of DMSO to a dilute solution
of IV in methyclohexane gave a two component emission band,
consisting of a monomer and an exciplex emission band, and
similar to that obtained with one drop of DMF (0.07 M)
(Figure 7.6). However, additional drops of DMSO remained
immiscible with methylcyclohexane; so that the IE/IM

ratio remained constant at a value of mo.86.

A plot of the ratio of exciplex and monomer fluores-
cence intensities, IE/IM against concentration of polar
molecule is shown in Figure 7.7 for MeCN and DMF. At any
given concentration, a greater IE/IM ratio is obtained
for DMF than for MeCN.

Also, a plot of IE/IM ratio against bulk dielectric
constant, e (estimated from Equation 7.1) is shown in
Figure 7.8. For a < 2.525, DMF shows higher IE/IM ratios
than MeCN; this trend is reversed for e > 2.525. Also,
while MeCN gives zero IE/IM ratio at e m 2.035, DMF gives
a finite IE/IM ratio of 0.16 at e = 2.0, which is slightly
below the bulk dielectric constant of the methylcyclo—

hexane (e I 2.02).

203

Ame.ev ... .Ass.ev -.-.- .xge.ev ---
macroEm wcfizhm> no monomopd on» ma AccoxonoaoaOIoz :H A\:
>H :H ondHoxo amasooHoEmLuCH ho commemopozam no npzoaw

.55 Ihozm4m><>>

 

 

 

 

Km

.5: omm «
.2: I $35335 to t

lea x Hy
so pomco

 

 

(3N1V‘138) wsuslm sowsosasoms

.m.» ooamfim

20H

.28: A: sea A:

AcczzmgoEvmqnquO—z Elan. “.0 ZO_._.<m._.ZMUZOU
0t.

mm.

.moasooaoe pmaoo mo coaummucoocoo mo coauoczm
a mo oocoomoposam xoaoaoxo one poaocos ho mofiuamcoucfi no oaumm .m.» madman

 

_ 1

 

 

Qd

mfiv

Q.

.QN

Nfi

.21.}

AH?

06

 

 

.282 E ..mzo 20v scopaooo osoooofiafio fiao no .8325
o no commemonozam xoanfioxo new Losocos mo moauamcmpCH no oaumm .m.> opswam

w ....ZdeZOU U_m._.om.dm_o

 

205

 

on ad 0N ed Nu QN
_ a _ . _ _ _ . _ _ oo

1 0.0

I Q.

1 IN

.1Nn
an.
to

10¢

l we

 

 

206

Discussion

The observed red-shifts in exciplex emission maximum
on addition of small amounts of MeCN, DMF, DMSO and HMPA
may be explained by considering the contributions of the
following molecular properties:

bulk dielectric constant;

donicity; and

"local" dielectric constant.

According to the usual solvation theory,5u’119’120
the red shift in the exciplex emission band with increasing
solvent polarity is considered to arise from solvation of
the highly polar exciplex by a bulk dielectric continuum.
This solvation theory neglects specific interaction be-
tween the exciplex and the surrounding polar molecules.
Thus, the exciplex fluorescence energy is treated as a
direct function of polarity parameters. On this premise,
one would expect that, by adding equal amounts of the

polar molecules: MeCN (e I 37.5), DMF (e I 37.0), DMSO

(5 I h6.68) and HMPA (e I 30.0), the extent of red-shift
will be: DMSO >> MeCN I DMF > HMPA. However the observed
order: HMPA “ DMSO > DMF >> MeCN contradicts the above

I

order expected on the basis of bulk dielectric effect.
Furthermore, if exciplex energy in solution is purely
dependent on bulk dielectric constant of the medium, then

its energy will be the same in different media of the same

207

bulk dielectric constant. However, our results (Figure 7.5)
indicate that addition of different amounts of the above
polar molecules to give the same bulk dielectric constant,
leads to different amounts of red-shifts in the exciplex
emission maximum. Again the observed order: HMPA > DMSO :
DMF > MeCN does not reflect any polarity effect. In fact
the least polar HMPA is seen to produce the highest spec-
tral shift under both conditions of the same bulk dielectric
constant and the same amount of added polar molecule.
For DMF and MeCN which have similar bulk dielectric con-
stant, the former causes relatively higher red shifts.

The observed red shifts in exciplex emission maximum
appear rationalizable on the basis of the donor number

(DN) of the polar molecule: HMPA (DN

38.8), DMSO (DN I
29.8), DMF (DN I 26.6) and MeCN (DN I lu.l). Thus, at low
concentrations of added polar molecule (<0.05 M) the ob-
served order of red shift is HMPA > DMSO > DMF >> MeCN
(Figures 7.3 and 7.N). At relatively high concentrations
(>0.05 M) of polar molecules of similar dielectric constants,
the observed red shift is still largely dependent on don-
icity, hence the order: HMPA > DMF > MeCN. Under the
condition of the same bulk dielectric constant of the
medium, the observed order of red shift: HMPA > DMSO 2
DMF > MeCN, again lends support to the influence of
donicity. The equality of the red-shifts by DMSO and DMF

can be understood from the similarity of their donor

208

numbers.

Also the higher values of IE/IM ratio obtained with
DMF compared to those of MeCN, seem indicative of the ef-
fect of donicity. On the basis of bulk dielectric effect,
both polar molecules are expected to have similar intensity
ratios. From the plot of IE/IM ratio against bulk dielec-
tric constant (Figure 7.8) DMF has a finite IE/IM ratio
at a bulk dielectric constant that is slightly less than
that of the pure hydrocarbon solvent, in which exciplex
fluorescence is not observed. One implication of this is
that the non-occurrence of intramolecular exciplex fluores-
cence of IV in hydrocarbon solvents is probably not due to
their low dielectric constant alone, but also may result
from their non-nucleophilic character. This seems to be
true, because compound IV exhibits intense intramolecular
exciplex fluorescence in dioxane which is essentially non-
polar (e I 2.209), but has a moderate nucleophilic charac-
ter (DN I 14.8).

The overall implication of the foregoing correlation
of the red shift in exciplex emission maximum with the
donicity of the polar molecule is that the electrostatic
interaction between an exciplex and a polar molecule
could be specific in nature, and may lead to a 1:1 or lzn
complex formation. Obviously, the positive charge on the
exciplex which is localized on the nitrogen, is susceptible

to solvation by nucleophilic polar molecules. The

209

continuous decrease in exciplex energy with increase in
concentration of the polar molecule appears to favor a

loose multi-component complex, involving one exciplex

moiety (or an excited state molecule of IV) and many polar
molecules. This seems to be the case since there is no

real isoemissive point, and no detectable change in exciplex.
band shape.

In spite of the overwhelming evidence in favor of the
specificity of the interaction of an exciplex with a polar
molecule, particularly at low concentrations of the latter,
dielectric effects seem to be evident at concentrations above
m0.05 M (Figure 7.3). Thus the red shift caused by DMSO
is relatively higher than that of HMPA in the above region
of concentration. Also, dielectric effect seems to account
for the deviation observed at relatively high concentration
(0.29 M) in the plot of exciplex spectral shift as a func-
tion of donor number (Figure 7.4). Furthermore the fact
that small amounts of methylcyanide (DN I 14.1) do bring
about red shifts in exciplex emission maximum in ether
(DN I 19.2) as solvent, appears to show that dielectric
effects do not seem altogether negligible, even at the so
called low concentrations of the polar molecule. If the
red shift occurs only by specific interaction which in turn
appears dependent on the donicity of the polar molecule,
then a 'levelling effect' would have been observed in

the case of MeCN.

210

It is possible for the added polar molecules to form
aggregates in the comparatively non-polar solvent.lll’112
Such aggregates can be formed around the exciplex because
of its highly dipolar character. As the concentration of
the polar molecule increases, the surface area of aggrega-
tion increases, and this may lead to a "local" dielectric
effect on the central dipole (exciplex). The 'local'
dielectric effect will gradually lead to a bulk dielectric
effect at very high concentrations of polar molecule.

This seems to explain the continuous lowering of exciplex
energy (with increase in concentration of polar molecule)
to a final value that will be observed in the particular
pure polar medium.

We may compare an exciplex with its surrounding polar
molecules in a non-polar medium with a hydrocarbon monomer
trapped inside a micelle of emulsifier molecules in an
aqueous (polar) medium. The surface area of this micelle
is known to increase with increase in concentration of the
emulsifier, and this leads to increased solubilization of
the non-polar solute (monomer) in water. In the same manner
increased stabilization of the polar solute (exciplex) in
a non-polar medium will be expected with increase in sur-
face area of aggregates of polar molecules, which may be
considered 'emulsifier' agents. This will correspond to

a loose l:n complex between the exciplex and the added

polar molecules.

211

In conclusion, it appears that the red shift in ex-
ciplex emission maximum observed in polar media occurs by

a combination of specific interaction and orientation

 

 

polarization interaction (dielectric effect). The contribu-
tion of the former will probably depend on the nucleo-
philic properties (donicity) of the solvent, and will be
effective at short range, preceeding the dielectric effect.
Also the observed red shift in exciplex emission maximum

in the presence of small amounts of a polar molecule in a
non-polar medium appears explicable on the basis of both
specific interaction and "local" dielectric effect. The
latter is believed to result from aggregation of polar

molecules in a non-polar medium.

CHAPTER 8
EXPERIMENTAL
I. Synthesis

General Procedure

A11 melting points were determined with a Thomas Hoover
melting point apparatus in open capillaries.

A Varian T-60 spectrometer was used to record nmr
spectra, which were recorded as 6 values in ppm downfield
from an internal standard of tetramethylsilane.

Mass spectra were obtained by Mr. Mark M. Weidner on
a Hitachi Perkin Elmer RMU-6 mass spectrometer.

Microanalyses were performed by Span Microanalytical
Laboratory, Eagle Harbor, Michigan.

Viscosity measurements were performed using a Cannon

(75, A32) viscometer.

2,6-Bis(bromomethyl)naphthalene (3)

96

3 28'
~

212

213

58 a mixture of 120 g

Following Golden's procedure,
(0.77 moles) of 2,6-dimethy1naphthalene and 0.70 g of ben-
zoyl peroxide in 1.4 liters of benzene (dried over CaH)
was refluxed and mechanically stirred in a 2 litre round
bottom flask. Then a mixture of 300 g (1.7 moles) of N-
bromosuccinimide (NBS) and 3.5 g of benzoyl peroxide was
added in four equal portions, over a period of 55 minutes.
Refluxing was continued for another one hour.

The hot solution was filtered. The benzene filterate,
on standing, gave a solid, product-l, which was treated
with 0.25 M sodium hydroxide to destroy any succinimide.
Product 1, on recrystallization with a 1:3 mixture of chloro-
form and tetrahydrofuran, gave 35.86 g of colorless rod-
like crystals. mp 182-185° (lit.59’121 182-184°).

The residue from the filtered hot reaction mixture
weighed 226.07 g (expected yield of succinimide was 152.3 g).
This residue was then digested with two 750 ml portions of
benzene to yield product 2; recrystallization with CHClB/THF
mixture gave colorless rod-like crystals which weighed
39.89 g. mp 182-185°. Final weight of succinimide was
141.45 g.

Over-all yield of product was 75.59 g (31.4%).

214

2,6-Bisgmethylamino)naphthalene,(6)

HiNFEF1llllllllk3H2hfltz

~

6.0 g (0.02 moles) of 2,6-bis(bromomethyl)naphthalene
was put in a l-litre round bottom flask equipped with a
cold finger; 300 m1 of ethyl alcohol (absolute) was added,
and the mixture stirred with a magnetic stirrer. To this
mixture was added 76 m1 CM3.13 moles) of ammonia dissolved
in 100 m1 of ethyl alcohol (absolute). The reaction flask
with its contents was then heated to a temperature of 30°
and the reaction mixture refluxed by means of the cold
finger maintained with a dry ice/isopropanol mixture. The
inhomogeneous reaction mixture turned milky after 15 min-
utes; heating was stopped after 25 minutes to minimize the
escape of N33 gas from the system. The reaction was then
left to run at room temperature for 21 hours.

The reaction mixture remained milky. Filteration
yielded a pale white solid (0.321 g) and a clear filtrate.
The filtrate was left to stir for 24 hours in an open beaker
to expel excess NH3 gas; a white residue (the hydrobromide
salt of the amine) was obtained (4.85 g).

The above residue was converted to the free amine by

treating its solution in 0.1 NHCl solution with 1 N NaOH;

215

the-resulting solid amine was recrystallized with a 1:3
mixture of methylene chloride and cyclohexane to give a
white fluffy powder (2.04 s. 57.4%), mp 125-130°. nmr
(CDClB/TMS) 67.27-7.72 (s,aromatic), 63.95 (Ar-032),
61.58 (-NH2) mass spectrum (70 eV): base peak at M/e
186 (parent), M/e 185 (P-l, 9oz, M/e 170 (P-NHZ, 63%),
M/e 156 (P-CHaNflz, 51:).

2,6-Bis(N—methy1,methylamino)naphthalene,(8)

HNHZC 0890‘,

§

C3i£<H3

10 g (0.032 moles) of 2,6-bis(bromomethyl)naphtha1ene
was put in a 2—litre round bottom flask, equipped with a
cold finger. uoo m1 C»lO.37 moles) of methylamine dissolved
in 400 m1 of absolute ethyl alcohol, was added into the
flask, and the mixture refluxed by heating to a temperature
of 45°, and cooling with a dry ice/isopropanol mixture in
the cold finger. Stirring was done by means of a magnetic
stirrer.

A clear solution resulted after about 30 minutes;
refluxing was continued for two hours. On evaporation
of ethanol and excess methylamine, a white needle-like
crystalline solid was recovered. This was dissolved in

0.5 N HCl solution, and treated with 1 N NaOH in a

 

216

selective two-stage precipitation process that yielded
the product (secondary amine) as the final precipitate-
white silvery flakes. The product was then recovered by
suction filtration, and vacuum dried at 42°C to constant
weight (5.905 g, 86%). mp 52-60°. nmr (CDCl3/TMS)
67.22-7.72 (aromatic), 53.8 (Ar-QE2-), 52.43 C>NCH3),
61.4 (ENE).

Anal. Calcd. for 014H18N2‘ C, 78.46; H, 8.47. Found:
C, 77.85; H, 8.02. mass spectrum (70 eV): base peak at
M/e 42 (CHZ-NICHZ), M/e 214 (parent, 64%), M/e 184 (P-
NHCH3, 37%), M/e 154 (P-ZNHCHB, 33%) M/e 44 (CHZNHCH3,
82%).

 

2,6-Bis(NIdimeth 1 meth lamino na hthalene ‘1

(cwaizwnzc as
152

- M043),

5 g (0.016 moles) of 2,6-bis(bromomethy1)naphthalene was
put in a 1-litre round bottom flask equipped with a cold
finger. 100 g (2.218 moles) of anhydrous dimethylamine was
added into the flask. The mixture was heated to 45° and
refluxed for two hours.

The clear reaction mixture was transferred to an open

vessel, and stirred at room temperature to evaporate

 

217

excess dimethylamine. A white finely crystalline solid was
obtained (the hydrobromide salt of the tertiary amine).
This was dissolved in 0.05 N HCl, and on treatment with l N
NaOH solution some slight precipitation occurred (this was
filtered off); further addition of NaOH, gave heavy white
flakes (product). The solid free amine was recovered by
suction filtration, washed with de-ionized water and dried
in a vacuum oven at 40° to constant weight (3.29 g, 85.3%).
mp loo-103°. nmr (CDCl3/TMS) 67.23-7.72 (aromatic), 63.5
(Ar°9§2-), 52.25 (—N(CH3)2). Anal. Cald. for C15H22N2:

C, 79.29; H, 9.15. Found: C, 79.26; H, 9.13. mass spec-
trum (70 eV): base peak at M/e 58 (CH2N(CHB)2), M/e 242
(parent, 41%), M/e 198 (P-N(CHB)2, 50%), M/e 154 (P-2N(CH3)2
9%).

6-Bis N-meth l N-acet 1 meth lamino naphthalene ‘)

3
I'm
(3

”3C": 2
‘05:?“
CD

 

IT

~~

In a 250 m1 3-necked round bottom flask was placed
0.734 g (0.0094 moles) of acetyl chloride dissolved in 85
ml of chloroform. 1 g (0.0047 moles) of 2,6-bis(N-methy1,
mmthylamdno)naphthalene was dissolved in 60 m1 of chloro-

form.and added dropwise to the acetylchloride solution from

 

218

a separatory funnel, with stirring. There was some milky
precipitation (probably hydrochloric acid salt of the amine-
excess of which was used as acid acceptor). Stirring was
continued at room temperature for 20 minutes; the milky
reaction mixture was then suction filtered. The filtrate
was washed with two 20 ml portions of 0.5 N HCl solution (to
remove any unreacted amine), and then with two 20 ml por-
tions of water. After drying with anhydrous sodium car-
bonate, the filtrate was evaporated to dryness to give a
solid residue. Recrystallization with a mixture of ether
and chloroform gave colorless plate-like crystals (0.343
g, 49.3%) mp 202-204°: nmr (CDClB/TMS) 67.22-7.75 (aromatic);
64.60 and 4.67 (Ar-Q§2-), 62.90 and 2.95 C3NCH3); 62.17
C-ECHB). Anal. Calcd. for ClBHZZNZOZ: C, 72.45; H, 7.43.
Found: C, 72.06; H, 7.22 mass spectrum (70 eV): base peak
at 225 (P-HNCOCH3); M/e 298 (parent, 18%), M/e 255 (P-
éoca3, 48%), M/e 226 (P-icocx3, 51%).

H3

Poly(2,6-dimethy1enenaphthalene, N,N'-dimethyl succinimide),

P2-14

P2-44 d

 

 

219

An aqueous solution of the diamine-2,6-bis(N-methyl,
methylamino)naphthalene (8) was prepared as follows: 3 g
(0.014 moles) of (8) was dissolved in 80 m1 of 0.5 N hydro-
chloric acid solution, and then neutralized with 1 N sodium
hydroxide solution. The resulting diamine aqueous solution
was diluted to 410 ml with distilled water and then trans-
ferred to the reaction vessel (a 1—1itre 3-necked round
bottom flask equipped with a mechanical stirrer, and a 250
m1 separatory funnel). 2.25 g (0.056 moles) of sodium
hydroxide was dissolved in 50 m1 of distilled water and
put in the separatory funnel. 4.35 g (0.028 moles) of
succinyl chloride was dissolved in 250 ml of chloroform in
a beaker, and then quickly added to the stirred diamine
solution in the reaction vessel. The separatory funnel
was simultaneously opened to let in sodium hydroxide solu-
tion (acid acceptor).

The reaction mixture turned milky instantaneously;
stirring (controlled with a Variac set at 90V) was continued
for fifteen minutes. The reaction mixture was then trans-
ferred to a 500 ml separatory funnel, to separate the
aqueous layer (clear) from the chloroform layer (slightly
colored brown).

Recovery and purification of the polymer was effected
as follows: the chloroform layer (containing the polymer)
was reduced to about 50 m1 of volume, and the polymer

precipitated by addition of methyl alcohol (a non-solvent)

220

with stirring. Product (recovered by suction filtration)
was slightly colored brown; it was then re-dissolved in

a small amount of chloroform, and the precipitation with
methyl alcohol was repeated two more times; a white powdery
product was obtained (1.836 g, 44%). mp 115-125°. nmr
(CDCl3/TMS) 67.27-7.73 (aromatic), 64.68 (Ar-Q§£-), 62.77
and 2.95 (:NCH3), 61.83 (- CH20). *Anal. Calcd. for
018H20N202: C, 72.95; H, 6.80. Found: C, 70.42; H,

6.48.

Poly(2,6-dimethylenenaphthalene,N,N'-dimethy1adipamide),
94-14

 

 

A 150 ml aqueous solution of 0.5 g (0.002 moles) of
2,6-bis(N-methyl, methylamino) naphthalene was made as in
P2-14. 0.584 g (0.0032 moles) of adipyl chloride, dissolved
in 100 m1 of chloroform was then interfacially polymerized
with the aqueous diamine solution in the presence of 0.265g
(0.007 moles) of sodium hydroxide, as the acid acceptor.

The reaction procedure, and the product recovery and puri-

fication were done as in PZ-IQ. The pure product was a

221

white powder. (0.206 g, 27.0%) mp 193-1958. nmr (CD013/TMS)
67.27-7.72 (aromatic), 64.65 (Ar'QEQ-), 62.87 C>NCH3),

62.38 (or3c32-), 61.78 (b,-g~CH249§2)ZCH2-g-). Anal.

Calcd. for ConzuNzoz: c, 74.04; H, 7.46. Found: c, 71.26;
H, 7.37. '

Pol 2 6-Dimeth lenena hthalene,N,N'-dimethylsuberamide),

P6-ig.

 

7‘8

iii i“

P6-44

 

'7:

Using the experimental procedure of 92-14, 0.6 3 (0.0028
moles) of 2,6-bis(N-methyl, methylamino)naphthalene, 8
(in 90 m1 of distilled water) was interfacially polymer-
ized with 0.652 g (0.003 moles) of suberoyl chloride (in
60 m1 of chloroform) in the presence of 0.247 g (0.0062
moles) of sodium hydroxide (acid acceptor).

Repeated attempts to precipitate the polymer from its
chloroform solution with methanol, failed. Use of hydro-
carbon non-solvents led to a rubbery material - slightly
colored brown; this was treated with methyl alcohol to
disperse the bulky material into fine particles; on
filtering and drying a slightly off white powder
was obtained (0.02578 g, 2.6%) mp 115-126° nmr (CDCl3/

TMS) o7.3o-7.7o (aromatic), 64.63 (Ar'§§2-) 62.87 c=wcx3)

222
62.33 (urgcaz-), 61.47 and 1.67 (b,-g-cnzég§é4ucwz-ge).

Pol 2 6-dimeth lenena hthalene,N,N'-dimethylsebacamide)L

P8-14

 

Eamon—31%;:

P8-44 J

 

Using the experimental procedure of P2-14, 0.5 g
(0.0023 moles) of 2,6-bis(N-methy1, methylamino)naphthalene
(in 110 m1 of distilled water) was interfacially polymI
erized.with 1.12 g (0.0047 moles) of sebacoyl chloride
(dissolved in 100 m1 of carbon tetrachloride). 0.378 g
(0.009mo1es) of sodium hydroxide was used as acid acceptor.

A solid rubbery material (slightly brownish) was re-
covered from the reaction mixture.- Purification was ef-
fected by twice dispersing into fine particles with methyl
alcohol; after drying (in a vacuum oven) a creamy white
powder was obtained (0.265 g, 30%), mp 123-125°. nmr
(CDCl3/TMS) 67.30-7.70 (aromatic), 64.64 (Ar'g§2-), 62.9
(:wcsB), 62.37 (b, ~8-CH2-), 61.33 and 1.70
(b.-S°cnzég§296032-g-). Anal. Calcd. for CzuH32N202:

C, 75.75; H, 8.48. Found: C, 74.61; H, 8.30.

223

Attem ted S nthesis of Pol 2 6-dimeth lenena hthalene

N,N'-dimethyl Glutaramide P - A

 

 

”$43

- 3-1.0.35")

.1

 

fi...

 

Using the experimental procedure of P2-14, 1 g (0.00467
moles) of 2,6-bis(N-methy1, methylamino)naphthalene (in
160 m1 of water) was interfacially polymerized with 0.878 g
(0.00514 moles) of glutaryl chloride (dissolved in 80 ml of
chloroform, 0.412 g (0.0109 moles) of sodium hydroxide was
used as acid acceptor.

Attempts to recover the product by precipitation from
chloroform using non-solvents (methanol and cyclohexane)
failed. Evaporation of solvent gave a reddish brown rubbery
product. No further attempt was made to purify the product;

it was therefore not used in the photophysical studies.

II. Solvents for Luminescence Studies

Methylene chloride, chloroformchethyl Alcohol, and 1-
Butanol were spectroquality solvents by Mallinckrodt
Chemical Works.

224
Acetonitrile, Cyclohexane,ApIDioxane, Methyl Sulfoxide and
N,N-Dimethyl formamide were spectroquality solvents from

Matheson Coleman and Bell (MC/B).

Isopropanol was a "photrex" reagent by J. B. Baker Chemi-

 

cal Company.

Hexamethylphosphoramide was purchased from Aldrich Chemical

 

Company, Inc., and was used without further purification.

Diethyl ether (Ether). This was dried over lithium alumin-

 

um hydride (LiAlH), and distilled (when needed) through a

1 meter vacuum jacket column.

Ethanol. 200 proof ethanol was distilled through a 1 meter
vacuum jacket column, at a slow rate (about 5 drops per
minute). Portions of about 50 ml were collected, and the
absorption spectrum taken in a 10 cm cell to check for
benzene. Distillation was continued till the charac-
teristic benzene UV absorption was no longer apparent.

Ethanol was then distilled and used as needed.

3-Methylpentane (3MP). A modified version of the purifica-

122 was used. 3-methylpentane (Phil-

tion method of Potts
lips Pure Grade) was shaken for 30 minutes with a 50:50

mixture of concentrated sulfuric acid and concentrated

225

nitric acid. It was then shaken 3 times for 30 minutes each
with concentrated sulfuric acid. This was then treated
with sodium carbonate solution until the CO2 production
ceased. The 3MP was then washed several times with water
until the water remained clear compared to the yellow

color of the first wash. The 3MP was stored over sodium
ribbon in a flask over night. It was refluxed through

a vacuum jacketed 1 meter column and distilled for use as

needed.
III. Spectral Measurements

Absorption Spectra. Cary 15 and Cary 17 were used to

obtain absorption spectra.

Emission Spectra. Aminco-keirs and Aminco-Bowman spectro-
phosphorimeters were used to obtain fluorescence spectra.
Also, the phosphorescence spectra were obtained with the
above instruments equipped with a rotating can phosphoro-

scope.

Fluorescence Quantum Yields

Fluorescence quantum yields were determined by com-
parison with two standards of known quantum yields:
1) Quinine sulfate (1 x 10'” M/L in 0.1 NH2SOu.

This was used to determine the relative quantum yields

226

of intramolecular exciplex fluorescence of 2,6-bis(N-
dimethyl, methylamino)naphthalene, IVgin organic solvents
of differentppolarities (Table 6.2). A quantum yield of
0.55 was assumed for Quinine sulfate.123
ii) 2,6-Dimethy1naphthalene (l x 10'“ M/L in the same
solvent as the unknown). This compound was used to deter-
mine the relative quantum yields of monomer fluorescence
of 2p6-bis(methylamino)naphthalenes, II-IV in 3-methyl
pentane (Table 6.1); and monomer and excimer fluorescence
of the polyamides (P-2, P-4, P-6 and P-8) and model compound
in a 1:3 mixture of chloroform and ethanol (Table 4.1).
A quantum yield of 0.3 was assumed for 2,6-dimethylnaph-
thalene.12u
This comparative method for quantum yield determina-
tion is based on Beer's law. For two optically dilute
solutions, with fluorescence intensities F1 and F2 (meas-

ured under identical conditions) the following ratio holds:123

 

Fl 8151012 a (3;)(0ptical density of 1)
F_= $26 202$ ¢ Optical density of 2
_ Area 1
_ Area 2 (8'1)

where o is quantum yield, 2 is molar extinction coefficient,
C is concentration in moles per litre, A is optical path
length of sample cell, 1 and 2 refer to the standard and

the unknown compound respectively.

227

All spectral measurements were performed with dilute
solutions of the unknown (1 x 10'“ M/L) in matched 1 cm
square cells. Solutions were excited at 290 nm, at room
temperature, and the fluorescence was viewed at right
angle. The spectrum of the unknown was first recorded;
thentfluestandard was quickly inserted and its spectrum
taken. The area under the spectra of both the standard
and the unknown were carefully cut out and weighed.

Uncorrected spectra were used because of the close
overlap of the spectra of the standards with those of
the respective unknowns. However, refractive index cor-
rections were made in the relative quantum yields determined
with quinine sulfate (as standard) because the unknowns
were dissolved in different solvents. These corrections

were effected with the following expression:125

.. n1 2
6 - <I>2 (a?) (8.2)

where nl (1.3341)25° is the refractive index of 0.1 N
H280“ and n2 is the refractive index oftflneorganic solvent.
¢2 is the uncorrected quantum yield obtained from equation
(8.1), and $3 is the quantum yield corrected for effects

due to refraction of light.

228

IV. Fluorescence Life Time Measurements

Fluorescence decay times were measured by means of a
Sipgle Photon Time Correlation Spectrometer. This instru-
ment consists essentially of a nanosecond flash lamp, a
special photomultiplier for detecting individual photons,

a time-to-amplitude converter (TAC) for timing the fluores-
cence photons relative to the flash lamp, and a multi-
channel analyzer (MCA) for data storage. A block diagram
of the instrument is shown in Figure 8.1.

The nanosecond lamp is a relaxation oscillator where
the voltage across two electrodes is increasing through
charging a capacitor till the breakdown voltage of the
gas filling the lamp (D2 was used) is reached whereupon
the lamp discharges and the circle is repeated. The width
at half maximum (whm) intensity of the lamp was %2 ns.

The lamp flash is detected with the IP28 phototube, and the
resulting analog pulse, after discrimination against low-
1evel noise (Ortec Model 436) is used to start the TAC.

Fluorescence photons are detected by a fast photomulti-
plier 56 DUVP (Time Resolution 800 picoseconds). This photo-
tube can amplify a single photon into an electrical pulse
of several volts. This pulse is used to stop the TAC.

The time resolution of the apparatus depends upon the
uncertainty of timing the detection of single-photon pulses

relative to the flash-lamp pulse. The principal time

229

 

 

 

 

SGDWP [BASE

 

 

 

 

 

E3 LIFETIME ES:

APPARATUS

 

AMP

 

 

DISC DISC

 

 

 

 

 

 

 

 

 

 

Figure 8.1. Block diagram for time resolved spectro-
photometer.

230

jitter arises from the fact that the amplitude of single-
photon pulses is not constant but covers a broad range in
pulse heights. The jitter associated with timing by a
single-level crossing was eliminated by using a "constant
fraction timing" discriminator (Ortec Model 463).

The heart of the instrument is the TAC (Ortec Model 457).
Its operation is illustrated in Figure 8.2. Each time the
lamp flashes, a pulse is sent to the TAC and initiates its
time sweep (start). If during the time sweep a stop pulse
is received from the photomultiplier, a TAC output pulse
is generated with amplitude proportional to the time
(tstop - tstart)'

The TAC is calibrated with the Ortec 462 Time Calibrator.
The time base is a precision lOO-MHz crystal-controlled
oscillator that is calibrated against WWV, the National
Bureau of Standards frequency, and it is temperature-
compensated for accuracy within the normal operating
range of 0 to 50°C.

The output of the TAC is fed to the MCA. The MCA
(Nuclear Data Model 1100) takes this voltage V and converts
it to an appropriate channel number C. The final result is
a count stored in Channel number 0 which is a record of the
time at which the photon was observed by the photomultiplier.
The analyzer channels now represent increments in time, and
the counts in each channel are proportional to the prob-
ability for fluorescence emission from the sample between

t and t+At, where t is measured from an arbitrary but

231

.oee one do coaoosodo

Al m2:

.m.m ohsmdm

 

 

 

 

nxyhm"

<—-— BOflll'TdWV ES'IFId .deIIO

 

 

232

fixed point. Since the TAC seeks one photon after it is
started, then each pulse of the lamp can produce only one
count. By collecting many of these single photon events
we eventually build in the analyzer the decay curve of the
emitting species.

The contents of the memory of MCA can be displayed on
an oscilloscope or an XY plotter or can be fed to a tele-
type equipped with a tape punch. Through the use of an
accoustic coupler, the teletype is connected with the
CDC 6500 computer where the data are processed. The
processed data are plotted through a digital plotter
(HEWLETT-PACKARD 7200A).

The lamp pulsing rate and the photon counting rate are
monitored through two digital counters (Monsanto Model 150A).
Monitoring the photon counting rate is important since at
too high a count rate there will exist a significant prob-
ability that two photons will be counted while the TAC is
involved in a sweep. Typically the lamp repetition rate
was 50 KHz and the count rate <100 counts/sec.

126 the instrument

According to the Convolution theorem,
output I(t) (when a fluorescence decay curve is measured)

is given by
I(t) I L?G<A)F(t-A)di (8.3)

F(t) is the time dependence of the lamp distorted by the

detection system, and is determined experimentally with a

233

scatterer at the wavelength of the exciting light. G(t)
is the time response of the fluorescence system when ex-
cited with the delta pulse of light. To obtain the de-
sired function G(t), Equation (8.3) is solved by a process
called deconvolution. The various deconvolution tech-
niques have been reviewed by Ware.127
However, if the fluorescence decay is a simple exponen-

tial and is long-lived (compared to the lamp decay) the

decay data can be processed by KINFIT, using the expression:
COUNTS = U(l)*exp(-U(2)*XX(1)*nsec/ch) (8.4)
RESID = CALC - XX(2) (8.5)

where U(2) is the rate constant for fluorescence decay;
XX(1) is the number of channels since tI0; U(l) is the
pre-exponential constant (initial count), nsec/ch is
nanosecond per channel (0.388 nsec) and XX(2) is the
experimental count.

The experimental counts are stored in 128 channels;
since the lamp pulse is of short duration (with a pulse
width of «2 nsec) photon counts from its decay are stored
in the earlier channels (~30). For the KINFIT analysis,
we used counts from channel 68, in order to eliminate the
lamp contribution in the fluorescence decay curve. Good

fits were obtained between the experimental and the

234

calculated points. A typical fit is shown in Figure 8.3.
The KINFIT program used in these analyses is given in

Appendix 8.1.

235

.Aaouds condusoaso one amassedpodxo msdosv n
.aucmod coudasoamo hooray o .Aucaod adpcosanoaxo msdosv x "Amid
.zoum\ Home :« z :iOH 2 my 0: non mumaadcs asap mafia oesoomoaosam .m.m enemas

2.5 h 0.36—- m-M ZO:<—-Uxu awn-GOEO uzuatfiux 2— .Nbdiocluz

_m----m'---v---'m|'--m--.- ----ml--- I---r'--'r----m.--"m----m---nlm----m----m----m----m-'-'m-"'w'--.

 

 

....u x .
OI u
I II
II x
OI ax
.— O I .
I ax
O Ox x
_ xO O
_ I I X
.. NO OI A
x u«..
x.- O
... xO _
x In
O O.
I
x
m. x“ I” 1
I
O x
x Ox
_ O O
x
2 .
1 Ox
a .
ax
O .
ex n
O
: ex ...
.1
I
I
i. u i
u
I u
1 u e
O
L ..
1 a.
m----m----m-"-m-'--m----m'-0'mnl--'m--.l-m---'m""m'-'-m'-'-m-'--m'-"m"--m'---m-"-m--'-m---'m--"

 

Appendix 8.1

gggmuzimmzx. assume” smmm cum—u “um-n

fiL°§:m “LII00L7I'1760

338ml! ‘INIIW
§;Q;;*Eztur76.‘

 

gggfieeeeeeeeeegeeeoeee

 

171:5?

 

 

 

 

 

     

 

 

 

 

 

 

fi-uac no W ode-x00 usuhhfl'”: nae-n m

‘3? 2;? 5:: 2: :§§§: 33:

5 iii: 55; I55 i5

I; gig is lgf E§§3

- 5;; . sis; :§=

is all; 1:: :5 :33; ,s:

1‘: si‘ 5‘ “5 iii if

5- ::§ : 22 5:; g:

2 :30 E . E EEO .

.smsi .. a Is .555 ”55‘ 4 II?
. I 1 I i’ii: o I 3 t e
,g 4% 35.: 5:5: $825 323551;)
3 ¥§§5 i, 5 ' 7“ g ‘ .5 IIIIE III: (.5

 

%.=‘ 0". 0:83;.”

'5. u. at: so?' ”En-£57973: fl. xtW'iEX'Ii. «6'3“ ""352 n:
93% 0:37. 79 070 Le”. :V‘Ct: “C37 OCUN39 If 40‘?: ”0". 58'}. §::
w’ l’iD}/!1¥N a“:
O Ngl’o = (I? ; f'f"‘0 3’. I I. 0' "J..'O' C .000 '30., ON.
’C .9 5 30V CT 3:. 5Y:§ (130,372: 0' 0:12V‘#?§ 0,0133‘303’078{§,9 ONO
J :oevI' l 30357303.vvri;'5 §';. II I"ng§i7 :5}; "°"°““’°"‘ "’°’ 33:
' fig“; 0 ob SQIOIOIOQ I 1% If?! 3::
‘78. IO ONO
45?: .747 no °"‘
6 ”'13:? ”in an: mun new
In. ...
1 c "out one
ec ”but 8::
z ' "31:“ ”a:
g "35: I33. WWII)???»
v Ifl1t7101 “ONOS‘CONOS 8" EH: {N'Ut
5133? 13'- 1:4-0; nu no .
; iguana-u to to as 0::
J! c u "#3“ we
a c u 3..“ °"‘
ONO
utxuut
s 73:!" I}:
a: Infiunu no to n 3
0,02“ no
N.
0::
8: C 833‘NU‘ N.
T" N.
u czmsuu‘ 3::
I
a $2NUI=U‘ 3::
ON.

236

BIBLIOGRAPHY

 

l.

10.
ll.

12.

13.

l“.

15.
16.

17.
18.

BIBLIOGRAPHY

B. Stevens and M. I. Ban, Trans. Faraday Soc., 69,
1515 (1964).

L. Rayleigh, Proc. R. Soc. A, 116, 702 (1927).
J. J. Hopfield, Phys. Rev., 32, 1133 (1930).

Th. Farster and K. Kasper, Z. Phys. Chem. (NF), 1,
275 (195“)-

Ibid; 2. Elektrochem; i9, 976 (1955).

G. A. Tischenko, B. Ya, Sveshnikov and A. S. Cherkasov,
Optika i Spektroskopija, 3, 631 (1958).

I. B. Berlman, J. Chem. Phys., 33, 1083 (1961).

T. V. Ivanova, G. A. Mokeeva and B. Ya. Sveshnikov,
Optica i Spectroskopija, 1a, 586 (1962).

E. Daller and Th. F6rster, Z. Phys. Chem., (NF), 31,
27A (1962).

I. B. Berlman and A. Weinreb, M01. Phys., 2, 313 (1962).

J. B. Birks and L. G. Christophorou, Proc. Roy. Soc.
(London), A, 27“, 552 (1963).

T. V. Ivanova, G. A. Mokeeva and B. Ya. Sveshnikov,
Optics and Spectrosc., 1g, 325 (1962).

R. C. Sangster and J. W. Irvine, J. Chem. Phys., g5,
670 (1956).

H. H. Perkampus and L. Pohl, Z. Phys. Chem. (NF),
39) 162 (196U).

F. Hirayama, J. Chem. Phys., 5;, 3163 (1965).

M. T. Vala, Jr., J. Haebig and S. A. Rice, J. Chem.
Phys., 33, 886 (1965).

6, 220A (1962).

L. J. Basile, J. Chem. Phys.,

S. S. Yanari, F. A. Bovey and R. Lumry, Nature, 00,
282 (1963).

237

 

19.

20.

21.

22.

23.

2M.

25.

26.
27.

28.

29.

30.

31.
32.

33.

3h.

35.

36.

238

H. Leonhardt and A. Weller, Ber. Bunsenges. Phys. Chem.

N. Mataga, T. Okada and K. Ezumi, M01. Phys., 19,
203 (1966).

N. Mataga, T. Okada and N. Yamamoto, Chem. Phys. Lett.
_1_, 119 (1967).

H. Knibbe, D. Rehm and A. Weller, Z. Phys. Chem. (NF),

H. Beens and H. Knibbe and A. Weller, J. Chem. Phys.,
51, 1183 (1967).

E. A. Chandross and H. T. Thomas, Chem. Phys. Lett.,
.2. 393 (1971).

J. B. Birks and L. G. Christophorou, Nature, Lond.,
19A, uu2 (1962).

Ibid, Nature, Lond., igg, 33 (1962).

N. F. Neznaiko, I. E. Obyknovennaya and A. S. Cher-
kasov, Opt. Spectrosc., 21/23, 285 (1966).

M. Amalia, F. Tavares, Trans. Faraday Soc., 66, 2u31
(1970).

R. Calas and R. Lalande, Bull. Soc. Chim. France, 763
(1959).

J. B. Birks, J. H. Appleyard and R. Pope, Photochem.
Photobiol., 2, 493 (1963).

J. N. Murrell and J. Tanaka, M01. Phys. 3, 363 (1964).

T. Azumi, A. T. Armstrong, and S. P. McGlynn, J. Chem.
Phys., fl: 3839 (196“)-

T. Azumi and S. P. McGlynn, J. Chem. Phys., 52, 1675
(1965).

R. Ide, Y. Sakata, S. Misumi, T. Okada, and N. Mataga,
J. S. C. Chem. C0mm., 1009 (1972).

T. Okada, T. Fujita, M. Kubota, S. Masaki and N. Mataga
Chem. Phys. Lett., 14, 563 (1972).

Y. Hayashi, S. Tazuke, Die Makromolekulare Chemie,
171, 57 (1973).

(A) LA.) LA)
\0 (I) -\l

‘—.
C)

r:
i“

: .ll‘
\0

(n
O

51.

56.

\n
\1

\,II
(‘1)

p
(n -q

l\)
LU
\0

Banba, Macromolecules, 9, “51 (1976).
ure Appl. Chem., 5, 121 (1962).

Ibid, Pure Appl. Chem., 1, 73 (1963).

 

J. Ferguson, J. Chem. Phys., 28, 765 (1958).

T. Azumi and S. P. McGlynn, J. Chem. Phys., ll: 3131
(196A).

E. Konijnenberg, Thesis, University of Amsterdam (1963).

T. Azumi, A. T. Armstrong and S. P. McGlynn, J. Chem.
Phys., 31, 3839 (196u).

C4
0

N. Murrel and J. Tanaka, M01. Phys., E, 363 (196“).
J. Frenkel, Phys. Rev., 31, 17 (1931).

A. s. Davydov, J. Exptl. Theor. Phys. (U.S.S.R.), gg,
210 (19u8).

P. Avouris, Thesis, Michigan State University (1973).

:11

S. Mulliken, J. Am. Chem. Soc., 13, 811 (1952).

:11

M. Hochstrasser, J. Chem. Phys. 36, 1099 (1962).

g

3

A. 811? kin, Nature, 200, 766 (1963).

M. Hedges and? A. Matsen, J. Chem. Phys., 28,
950 (1958) '

S. P. McGlynn, A. T. Armstrong and T. Azumi, Modern
Quant. Chem. , edit Oktay Si nanoglu, 203 (1965).

A. Weller, Pure Appl. Chem., 16, 115 (1968)

3. Beens and A. Weller, Org. M01. Photophysics, edit.
J. Birks, 2g, p. 201.

D. Rehm and A. Weller, Z. Phys. Chem. (NF), 69, 183
(1970).

D. Rehm, Z. Naturforch. 25a, lAfl2 (1970).
R. B. Fox, Pure Appl. Chem., 3’, 235 (1973).

J. H. Golden, J. Chem. Soc., 56, 3Tfll (1961).

 

59.

60.
61.

62.

63.
6A.

65.

66.

67.

68.

69.

70.

71.

72.

73.

7b.
75.

76.

2H0

P. W. Storms and P. R. Taussig, J. Chem. and Eng. Data,
;_, 272 (1966).

C. Schotten and J. Baum, Ber., 11, 2598 (188A).

V. E. Shashoua and W. M. Eareckson, J. Polym. Sci.,

J. Brandrup and E. H. Immergut, Polymer Handbook,
Interscience Publishers, New York IV-2 (1966).

w. Klopffer, Chem. Phys. Lett., 5, 193 (1969).

E. A. Chandross and C. J. Dempster, J. Am. Chem. Soc.,
22; 3686 (1970).

P. C. Johnson and H. W. Offen, J. Chem. Phys., ii,
29A5 (1971).

R. B. Fox, T. R. Price, R. F. Cozzens and J. R. Mc-
Donald, J. Chem. Phys., 21, 53A (1972).

J. W. Longworth, Biopolymers, fl, 1131 (1966).

F. C. DeSchryver, N. Boens, J. Huybrechts, J. Daemen
and M. DeBrackeleire, Pure and Appl. Chem., 32, 237
1977).

K. Zachariasse and W. Kfihnle, Z. Phys. Chem. Neue
Folge, Bd 101, 267 (1976).

C. David, M. Piens and G. Geuskens, Eur. Polym. J.,
8, 1019 (1972).

C. David, M. Lempereur and G. Geuskens, Eur. Polym.

J-: 2, 1315 (1973).

L. Alezandru and A. C. Sommersall, J. Polym. Sci.,
Polym. Chem. Ed., 1:, 2013 (1977).

J. A. Ibemesi, M. A. El-Bayoumi and J. B. Kinsinger,
Chem. Phys. Lett., :3, 270 (1978).

Y. C. Wang and H. Morawetz, J.A.C.S., 98, 3611 (1976).

E. Doller and Th. Farster, z. Physik. Chem. (NF),
31. 132 (1962).

J. B. Birks and L. G. Christophorou, Proc. R. Soc.,
A277, 571 (1964).

77.

78.

79.

80.

81.
82.

83.

8A.

85.

86.

87.

88.
89.

90.

91.

92.

93.

291

C. A. Parker, Photoluminescence of Solutions, Elsevier
Publishing Co., New York, 350 (1968).

J. B. Birks, "Photophysics of Aromatic Molecules",
Wiley-Interscience, New York, NY, 310 (1970).

C. W. Bunn, in Fibres from Synthetic Polymers, R. Hill,
edi., Elsevier.

R. M. Meighan and R. H. Cole, J. Phys. Chem., §§,
503 (1964).

J. E. Guillet, Pure and Appl. Chem., 33, 249 (1977)-

W. K10pffer, in Organic Molecular Photophysics, J. B.
Birks, ed-. E1, 357 (1973).

A. Peterlin, J. Polym. Sci. Polym. Lett. Ed., 39,
101 (1972).

P. Avouris, J. Kordas and M. A. El-Bayoumi, Chem. Phys.
Lett., g9. 373 (197u).

T. Nishihara and M. Kaneko, Die Makromolekulare Chemie,
124, 84 (1969).

T. Kamijo, M. Irie, K. Hayashi, M. Aikawa, T. Takemura
and H. Baba, Paper presented to the 29th Polymer
Symposia, Osaka (1975).

Y. Miron, B. R. McGarvey and H. Morawetz, Macromolecu-
les, g, 15A (1969).

D. Tabak and H. Morawetz, Macromolecules, 3, 903 (1970).

H. A. Benesi and J. H. Hildebrand, J. Am. Chem. Soc.,
11; 2703 (1999).

H. Sugiyama and H. Kamogawa, J. Polym. Sci. A-l, 3,
2281 (1966).

A. Rembaum, A. M. Hermann and R. Haack, J. Polym. Sci.,
9. 1955 (1968).

H. Leonhardt and A. Weller, Luminescence of Organic and
Inorganic Materials, Wiley, New York, p. 74 (1962).

G. Duportail, D. Froelich and G. Weill, Eur. Polym.
Jo. l. 977 (197A).

9U.

95.

96.

97.

98.

99.

100.

101.

102.

103.

10A.

105.

106.

107.

108.

109.

110.

242

S. Tazuke, K. Sato and F. Banba, Chem. Lett. p. 132
(1975).

C. David, N. Putman-DeLavarelle and G. Geuskens, Eur.
Polym. J., 33, 365 (1976).

R. S. Davidson and K. R. Trethewey, J. Chem. Soc. Chem.
Comm., 33, 827 (1976).

T. Okada M. Kubota, S. Masaki and N. Mataga, Chem. Phys
Lett. , 15, 563 (1972).

N. Mataga, T. Okada, H. Masuhara, N. Nakashima, Y.
Sakata and S. Misumi, J. Lumin., 33/33, 159 (1976).

T. Okada, T. Saito, N. Mataga, Y. Sakata and S. Misumi,
Bull. Chem. Soc. Jpn., 39, 331 (1977).

S. Masaki, T. Okada, N. Mataga, Y. Sakata and S. Misumi,
Bull. Chem. Soc., Jpn. 33, 1277 (1976).

P. J. Harman, P. R. Nott and B. K. Selinger, Aust. J.
Chem., 3Q, 1875 (1977).

. Beens and A. Weller, Org. M01. Photophysics, edit.
B. Birks, 1;, p. 170.

940 an:

. R. Goldschmidt, R. Potashnik and M. Ottolenghi,
Phys. Chem.,13 1025 (1971).

Tournon and M. Ashraf El- -Bayoumi, J. Chem. Phys.
56, 5128 (1972).

Y. Taniguchi and N. Mataga, Chem. Phys. Lett., 13,
596 (1972).

N. Orbach and M. Ottolenghi, Exciplex, Proc. Meet.
1979, 13, Edith M. Gordon and W. R. Ware (1975).

N. Orbach and M. Ottolenghi, Chem. Phys. Lett., _3,
175 (1975).

Y. Taniguchi, Y. Nishina and N. Mataga, Bull. Chem. Soc.
Jpn-. 32, 76A (1972).

M. G. Kuzmin and L. N. Guseva, Chem. Phys. Lett., 3,
71 (1969).

C. A. Parker, Photoluminescence of Solutions, Elsevier
Publishing Co., New York, p. 88 (1968).

111.

112.

115.
116.
117.
118.

119.
120.

121.
122.
123.
129.
125.

126.

127.

293

E. A. Chandross and H. T. Thomas, Chem. Phys. Lett.,
9. 397 (1971).

B. K. Selinger and R. J. McDonald, Aust. J. Chem.,
2:. 897 (1971).

E. A. Chandross, The Exciplex, edits., M. Gordon and
w. R. Ware, p. 187 (1975).

G. S. Beddard, S. E. Carlin and C. Lewis, J. Chem. Soc.
Far. Trans. 2,13, 189“ (1975).

V. Gutman, Chem. Tech., p. 255 (1977).
U. Mayer, Co-ordination Chem. Reviews, 33, 159 (1972).
R. M. Fuoss, J. Am. Chem. Soc., 66, 1031 (193M).

C. Treiner, J. F. Skinner and R. M. Fuoss, J. Phys.
Chem. , 66, 3U06 (196A).

E. Lippert, Angew. Chem., 13, 695 (1961).

N. Mataga and T. Kubota, "Molecular Interactions and
Electronic Spectra", Chapter 8, Marcel Dekker, New
York (1970).

W. Ried and H. Boden, Chem. Ber., 33, 1981 (1958).

N S. Potts, J. Chem. Phys., 36, 809 (1952).

C A. Parker and W. T. Rees, Analyst, 66, 587 (1960).
B. K. Selinger, Aust. J. Chem., 33, 825 (1966).

J. N.

Demas and G. A. Crosby, J. Phys. Chem., 7

991 (1971)-

W. R. Ware, Technical Report No. 3, Office of Naval
Research, Contract No. 001A-67-A-0113-0006, March 1

(1969).

W. R. Ware, in "Creation and Detection of the Excited
States" V01. 1A, A. Lamola, edit. Marcel Dekker,
New York (1971).