Photo-induced intermolecular electron transfer from donating amines to accepting diarylethylenes in different solvents

Article abstract of DOI:10.1007/s10953-015-0373-6

The fluorescence quenching of newly synthesized diarylethylenes by some aliphatic and aromatic amines has been carried out at room temperature in solvents of various polarities and viscosities including acetonitrile, methanol, ethanol, and cyclohexane usi

Full text of DOI:10.1007/s10953-015-0373-6

J Solution Chem
DOI 10.1007/s10953-015-0373-6
Photo-Induced Intermolecular Electron Transfer
from Donating Amines to Accepting Diarylethylenes
in Different Solvents
1
1
1
Marwa N. El-Nahass
•
Tarek A. Fayed
•
Mohammed A. El-Morsi
Received: 4 May 2013 / Accepted: 7 April 2015
Ó Springer Science+Business Media New York 2015
Abstract The fluorescence quenching of newly synthesized diarylethylenes by some
aliphatic and aromatic amines has been carried out at room temperature in solvents of
various polarities and viscosities including acetonitrile, methanol, ethanol, and cyclohex-
ane using steady-state measurements. Additionally, the cyclic voltammetric behavior of the
investigated diarylethylenes has been studied, in order to determine their reduction
potentials, which are necessary for calculation the Gibbs energy change for the photoin-
duced, electron transfer (ET) reactions in diarylethylene–amine systems. No change in the
shape of the fluorescence spectra was observed, except for diarylethylenes–N,N-dimethyl
aniline pairs in cyclohexane at which an exciplex was observed. The quenching mechanism
is due to the ET from the ground state amines to the excited diarylethylenes. Both linear
and non-linear Stern–Volmer behaviors have been encountered. The observed k values for
q
different diarylethylenes–amine pairs do not correlate with the basicity (pK ) of the amines
b
but decrease with increasing the oxidation potentials of the amines. The bimolecular
quenching rate constants have been correlated with DG° values for the ET reactions
according to the Marcus outer-sphere ET theory.
Keywords Fluorescence quenching Á Cyclic voltammetric behavior Á Exciplex Á Electron
transfer Á Aliphatic amines Á Aromatic amines
1
Introduction
Diarylethenes with heterocyclic rings have been developed as novel classes of therapeutic
compounds shown to exert a wide range of biological activities. Initially employed as
pigments in leather tanning, most of the compounds that exhibited both high levels of
&
Marwa N. El-Nahass
marwacu@yahoo.com; marwa.elnahas@science.tanta.edu.eg
1
Chemistry Department, Faculty of Science, Tanta University, Tanta 31527, Egypt
123

J Solution Chem
fluorescent properties and a capacity to bind with cellular structures were extensively used
as fluorochromes [1]. They have been investigated for their clinical efficacy as an oph-
thalmic drug, to lower intraocular pressure, and for anti-tumor and anti-HIV activities [2].
Also, they have been used as novel fluorescent substrates for the localization of peroxidase
activity [3], antiproliferative agents [4], nonlinear optical materials [5] and emitter material
in organic electroluminance devices [6].
Electron transfer (ET) reactions are one of the most interesting subjects and have
experienced extensive experimental and theoretical studies in recent decades [7, 8]. Studies
on ET reactions have importance from the view point of both academic and applied
implications, as these reactions are important in chemistry and biology. Understanding of
the various factors that control ET reactions is the main impetus in most of studies on ET
processes. Photoinduced ET reactions where either the acceptor or the donor is photoex-
cited to trigger the reaction have been extensively studied to investigate various factors that
control the mechanisms and dynamics of ET in various ET systems. In some of the earlier
studies on bimolecular ET reactions [9–11], it was observed that the quenching process is
largely dependent on the aliphatic and aromatic nature of the quenchers. The fluorescence
quenching for a series of acceptors by different aliphatic and aromatic amines as electron
donors have been investigated in acetonitrile solutions under diffusion conditions, aiming
to understand how the nature of the amines affects the ET process. In addition to the non-
cyclic aliphatic amines, cyclic amines like 1-azabicyclo-[2]-octane (ABCO) and 1,4-di-
azabicyclo-[2]-octane (DABCO) have been also used to obtain better understanding and
interpretation of the observed results [12]. Therefore, our work aimed at studying the cyclic
voltammetric behavior of the newly synthesized diarylethylenes, in order to determine their
reduction potentials, which are necessary for calculation the Gibbs energy change for the
photoinduced, ET reactions in diarylethylene–amine systems. Also, the fluorescence
quenching of the investigated diaryethylenes with some amines as electron donors has been
studied using steady-state fluorescence spectroscopy, to give insights into the factors
affecting the photoinduced ET reactions. Finally, the rationalization of the experimental
data in terms of Marcus theory and Rehm–Weller model has been studied.
2
Experimental
The investigated 1-Ar-2-(2-benzothiazolyl) ethenes (Ar = phenyl, 1-naphthyl and
-naphthyl) were synthesized by condensation of 2-methylbenzothiazole with the corre-
2
sponding arylaldehyde [13], as follows: equimolar quantities of 2-methylbenzothiazole and
the corresponding arylaldehyde were dissolved in an alkaline solution of dimethylfor-
mamide (containing 4 g KOH per 50 mL) then the solution was stirred for 3 h; after that,
-
3
drops of 0.1 molÁdm HCl were added to the pre-cooled reaction mixture reaction mixture
until complete precipitation. The products were removed by filtration and purified by
recrystalization twice from dry ethanol. The purity of the prepared compounds was
checked by UV–Vis, fluorescence spectroscopy and melting point measurements. The
compounds have a pale yellow color. Amine samples were N,N-diethylethanamine (TEA,
0
Aldrich), N,N -dimethylaniline (DMA, Fluka) and aniline which were purified by distil-
lation under reduced pressure, while 1,4-diazabicyclo[2]octane (DABCO, Aldrich) was
sublimed before use. Chemical structures of the diarylethylenes are shown in Scheme 1,
along with their abbreviations. All solvents used were of spectroscopic grade (Aldrich or
Merck) except ethanol and methanol, which were purified by double distillation before use.
1
23

J Solution Chem
Scheme 1 Structures and
abbreviations of the prepared
S
1-Ar = 2-(2-benzothiazolyl)
ethenes
Ar
N
1
-aryl-2-(2-benzothiazolyl) ethenes
Phenyl
(STBT)
Ar =
1-Naphthyl (BTN1)
2- Naphthyl
2
(BTN )
The reduction potentials were measured using cyclic voltammetry. Voltammetric
measurements were carried out using an electrochemical (Pyrex) cell comprising three
electrodes immersed in a solution containing the investigated compounds. One of the three
electrodes is the working electrode, which was a hanging mercury drop (HDME, geo-
2
metrical area 0.026 cm ), the second electrode was the reference electrode (Ag/AgCl), and
the third electrode was the counter electrode.
The absorption spectra were recorded on a Shimadzu UV–Vis 3101 PC spectropho-
tometer, while the emission spectra were measured using a Perkin-Elmer LS50B lumi-
nescence spectrometer using matched quartz cuvettes.
3
Results and Discussion
3
.1 Cyclic Voltammetric Behavior of 1,2-Diarylethylene Derivatives
in Universal Buffer Solutions
To determine the reduction potentials of the 1-Ar-2-(2-benzothiazolyl) ethenes
Ar = phenyl, STBT, 1-naphthyl, BTN and 2-naphthyl, BTN ), which are necessary for
(
1
2
calculation of the Gibbs energy changes of the photoinduced ET reactions in the
diarylethylenes–amine systems, the electrochemical reduction was studied in universal
-
3
buffer solutions having pH values 4, 7 and 10 using sample concentrations of 1 9 10
-3
molÁdm , at room temperature. The recorded cyclic voltammograms show a single well
defined cathodic peak, which was attributed to the reduction of the diarylethylenes. The
?
reduction wave is in the potential range of -0.85 to -1.8 V versus the Ag /AgCl electrode
throughout the pH range studied. At pH = 4, the peak of the reduction wave of BTN is
1
-
1
located at -0.98 V at a scan rate of 500 mVÁs , as an example. On increasing the pH of
the solution, the peak potential (E ) is shifted to more negative values and the peak current
pc
(ip) varied as well (Fig. 1). The larger shift in the Epc values was found at pH = 10,
indicating participation of hydrogen ions in the electrode process. The relation of Epc to pH
is shown in Fig. 2. Examination of the cyclic voltammograms obtained at different scan
123

J Solution Chem
3
3
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
.5
.0
(a)
pH=10
pH=7
pH=4
-0.5
0.8
1.0
1.2
1.4
1.6
1.8
−
Ε (V)
(b) 3.5
3.0
2.5
2
1
1
0
0
.0
.5
.0
.5
.0
pH=10
pH=4
pH=7
-0.5
0.8
1.0
1.2
1.4
1.6
1.8
−
Ε (V)
4
3
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
.0
.5
.0
(c)
pH=10
pH=4
pH=7
-0.5
1.0
1.2
1.4
Ε (V)
1.6
1.8
−
-
1
Fig. 1 Cyclic voltammograms of a STBT, b BTN
solutions
1
, and c BTN
2
at scan rate 500 mVÁs in different buffer
1
23

J Solution Chem
Fig. 2 The relation of Epc versus
pH for the investigated
compounds at scan rate
1
.7
.6
.5
.4
.3
.2
.1
.0
.9
STBT
BTN2
BTN1
1
-1
500 mVÁs
1
1
1
1
1
1
0
4
5
6
7
8
9
10
11
pH
-
1
rates within the range of 100–500 mVÁs at pH = 4 reveals that no anodic counter part is
seen on the reverse sweep, indicating either the reduction is totally irreversible, which is
unlikely, or the reduction product is consumed rapidly by another processes, e.g., proto-
nation or dimerization. Also, the peak potential was shifted to more negative potentials on
increasing the scan rate. A similar behavior is obtained at pHs 7 and 10. The typical
voltammograms recorded at various potential scan rates for the compounds under inves-
tigation in different buffers are shown in Fig. 3. As shown from these figures, the peak
potential (E ) and the peak current (i ) were strongly affected by changing the scan rate.
pc
pc
The large decrease in the peak current indicates that the product formed is strongly
adsorbed at the electrode surface.
1
/2
Since the peak potential, the peak current and the current function (i /v ) depend on
pc
the scan rate, the variations of the cathodic peak potential and peak current with the scan
1
/2
rate were analyzed by plotting ipc against the square root of scan rate (v ). Linear
dependences were obtained as is shown in Fig. 4. These results confirm the previous
conclusion regarding the irreversibility of the process. This linear dependence indicates
that the reduction process is diffusion-controlled [14]. Also, the dependence of the current
1
/2
1/2
1/2
1/2
function (i /v ) on v is shown in Fig. 5. The dependence of i /v on v is an
pc
pc
important diagnostic criterion for establishing the type of mechanism by cyclic voltam-
metry. Tables 1, 2, and 3, show the electrochemical data for the reduction of these com-
1
/2
pounds at different pH values and the different values of i /v for the investigated
pc
compounds as a function of scan rate. From the data reported in these tables, it was found
1
that the value of i /v decreases as the scan rate increases. From these results it was
/2
pc
concluded that the compounds are reduced electrochemically in a diffusion controlled
irreversible process.
3
.2 Steady-State Fluorescence Quenching Studies
With the aim of collecting information about the excitation and emission maxima, which
are necessary for quenching measurements, ground state absorption spectra and steady
state fluorescence spectra of the investigated diarylethylenes were recorded in different
solvents of various polarities and viscosities including acetonitrile (ACN), methanol
(
MeOH), ethanol (EtOH), and cyclohexane. All measurements were carried out in the dark
A
to avoid any photochemical reactions. Table 4 lists the absorption maxima (kmax) and
123

J Solution Chem
Fig. 3 Cyclic voltammograms
of a STBT, b BTN , and c BTN
in pH = 7 and c pH = 10 at
(
a) 2.5
Scan rate mV/s
1
2
5
4
00
00
different scan rates of 100, 200,
2
1
1
0
0
.0
.5
.0
.5
.0
300
200
-1
300, 400 and 500 mVÁs
1
00
1
.2
1.4
Ε (V)
1.6
−
(
b) 3.5
Scan rate mV/s
00
5
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
.0
400
3
2
1
00
00
00
-0.5
1
.2
1.4
Ε (V)
1.6
−
(
c) 3.5
Scan rate mV/s
00
400
5
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
.0
3
2
1
00
00
00
-
0.5
1.2
1.4
Ε (V)
1.6
−
F
max
fluorescence maxima (k ) of STBT, BTN and BTN in the different solvents, while
1
2
Fig. 6 shows the normalized electronic absorption and fluorescence spectra of STBT,
BTN and BTN in different solvents, as representative examples. The absorption spectra
1
2
1
23

J Solution Chem
(a)
3
2
2
1
1
0
.0
.5
.0
.5
.0
.5
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1/2
1/2
-1/2
v
(mv
s
)
(b)
3
3
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
.5
.0
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1/2
1/2
-1/2
v
(mv
s
)
(
c)
3
3
2
2
1
1
0
.5
.0
.5
.0
.5
.0
.5
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1/2
1/2
-1/2
v
(mv
s
)
Fig. 4 Dependence of the cathodic peak current (ipc) on the square root of the scan rate for the investigated
compounds at different pH values
123

J Solution Chem
Fig. 5 Dependence of the
(
a) 0.14
1
/2
current function (ipc/v ) on the
square root of scan rates for the
investigated compounds at
different pH values
0
0
0
0
.12
.10
.08
.06
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1
/2
1/2 -1/2
v
(mv
s
)
(
b) 0.16
0
0
0
0
.14
.12
.10
.08
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1
/2
1/2 -1/2
v
(mv
s
)
(
c) 0.16
0
0
0
0
.14
.12
.10
.08
pH =4
pH =7
pH =10
1
0
12
14
16
18
20
22
24
1
/2
1/2 -1/2
v
(mv
s
)
show maxima between 330 and 363 nm while the fluorescence spectra show maxima in the
range 393–443 nm, with a shift depending on the solvent and the ring size. Generally, the
absorption spectra show small solvent dependence. The shifts of the absorption band
1
23

J Solution Chem
Table 1 Cyclic voltammetric data for the reduction of STBT, BTN
the potential scan rate
1
and BTN
2
in pH = 4 as a function of
Scan rate
)
STBT
BTN
1
BTN
2
-
1
(
mVÁs
1
/2
1/2
1/2
i
pc
-Epc
(V)
i
pc/
v
i
pc
-Epc
(V)
i
pc/
v
i
pc
-Epc
(V)
pc/
i v
1/2
1/2
1/2
(
lA)
(lAÁmV
Á
(lA)
(lAÁmV
Á
(lA)
(lAÁmV
Á
-
1/2
)
-1/2
-1/2
s
s
)
s
)
1
2
3
4
5
00
00
00
00
00
0.74
1.37
1.94
2.46
3.05
1.02
1.03
1.03
1.04
1.041
0.074
0.1
0.8
0.96
0.97
0.98
0.98
0.98
0.08
0.1
0.82
1.48
2.16
2.83
3.52
0.97
0.98
0.99
0.99
1.0
0.02
0.11
1.35
1.84
2.48
3.02
0.112
0.123
0.14
0.11
0.124
0.14
0.16
0.124
0.14
Table 2 Cyclic voltammetric data for the reduction of STBT, BTN
the potential scan rate
1
and BTN
2
in pH = 7 as a function of
Scan rate STBT
)
BTN
1
BTN
2
-1
(
mVÁs
1
/2
1/2
1/2
i
pc
-Epc
(V)
i
pc/
v
i
pc
-Epc
i
pc/
v
i
pc
-Epc
(V)
pc/
i v
1/2
1/2
1/2
(lA)
(lAÁmV
Á
(lA) (V)
(lAÁmV
Á
(lA)
(lAÁmV Á
-
1/2
)
-1/2
-1/2
s )
s
s
)
1
2
3
4
5
00
00
00
00
00
0.58
1.27
0.058
0.82
1.5
1.198 0.082
1.209 0.11
0.886 1.227 0.09
1.646 1.24 0.12
1.251 0.13
1.065 1.278 0.08
1.469 1.28
1.908 1.28
2.359 1.29
0.08
0.1
2.14
2.79
3.38
1.21
1.22
1.23
0.123
0.14
0.15
2.17
2.65
3.26
1.24
1.25
0.133
0.15
0.11
1 2
Table 3 Cyclic voltammetric data for the reduction of STBT, BTN and BTN in pH = 10 as a function of
the potential scan rate
Scan rate STBT
)
BTN
1
BTN
2
-1
(
mVÁs
1
/2
1/2
1/2
i
pc
-Epc
(V)
i
pc/
v
i
pc
-Epc
i
pc/
v
i
pc
-Epc
(V)
pc/
i v
1/2
1/2
1/2
(
lA)
(lAÁmV
Á
(lA) (V)
(lAÁmV
Á
(lA)
(lAÁmV Á
-
1/2
)
-1/2
-1/2
s )
s
s
)
1
2
3
4
5
00
00
00
00
00
0.99
1.59
0.099
1.0
1.484
1.486
0.099
0.096
0.93
1.49
2.07
1.503 0.093
1.506 0.11
1.516 0.12
1.425 1.602 0.100
1.722 1.607 0.99
2.058 1.604 0.103
1.37
1.75
2.17
2.58
1.4869 0.101
1.489
1.49
0.11
0.12
2.568 1.52
0.13
2.49
1.602 0.111
3.04 1.522 0.14
A
max
maximum k
maxima shifts by about 10, 12 and 15 nm for STBT, BTN and BTN , respectively, with
are 2, 5 and 6 nm for BTN , STBT and BTN , while the fluorescence band
2 1
1
2
changing the solvent from EtOH to cyclohexane.
The fluorescence of STBT, BTN and BTN were effectively quenched by some ali-
1
2
phatic (TEA and DABCO) and aromatic amines (DMA and aniline) in the polar solvents
EtOH, MeOH and ACN and non-polar cyclohexane at room temperature (Fig. 7). In fact,
123

J Solution Chem
Table 4 The absorption and emission maxima of STBT, BTN
1
2
and BTN in different solvents
Compound
Solvent
STBT
BTN
1
BTN
2
a
f
a
f
a
f
k
(nm)
k
(nm)
k
(nm)
k
(nm)
k
(nm)
k
(nm)
max
max
max
max
max
max
ACN
331
333
335
330
399
403
403
393
353
358
357
351
435
437
436
424
350
350
352
350
413
418
417
402
MeOH
EtOH
Cyclohexane
(a) 1.2
(b) 1.2
absorption
absorption
emission
emission
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
3
00
400
500
300
400
wavelength (nm)
500
wavelength (nm)
Fig. 6 Normalized absorption and emission spectra of (dotted lines) STBT, (dashed line) BTN
line) BTN in a ACN and b cyclohexane
2
1
and (solid
the absorption spectra of the fluorophore do not show any changes on addition of amines,
which indicates the absence of ground state complex formation. Also, there was no change
in the shape and position of the fluorescence spectra, even with higher concentrations of
the quencher in the solvents, except for cyclohexane. This confirms the absence of any
diarylethylenes–amine excited state complex (exciplex) formation. However, for STBT,
BTN and BTN –DMA pairs in non polar solvents, a new structureless red shifted emis-
1
2
sion band (corresponding to formation of an exciplex) is observed at sufficiently high
-3
concentrations of DMA (0.5–0.65 molÁdm ), where the fluorescence is almost completely
quenched and the exciplex emission is symmetrically increased around this band. Similar
observations were reported for trans-stilbene, disubstituted anthraquinones and substituted
styrylthiophene [15–17].
Quenching of steady state fluorescence intensity of the investigated compounds by the
used amines was analyzed using the Stern–Volmer (SV) relationship as:
I0
¼
1 þ KSV½QŠ
ð1Þ
ð2Þ
I
KSV ¼ kqsf
where I and I are the fluorescence intensities in the absence and presence of quencher, K
o
SV
is the Stern–Volmer rate parameter constant, k is the bimolecular quenching rate constant,
q
[
Q] is the quencher concentration and s is the fluorescence lifetime of diarylethylenes.
f
Representative Stern–Volmer plots are shown in Fig. 8. It can be observed that the plots
are linear for STBT, BTN1 and BTN /DMA and/TEA pairs in EtOH, MeOH and
cyclohexane, as examples, with different quencher concentrations in the range
2
1
23

J Solution Chem
(
a) 9
(b)
[
0
DMA] mole/L
.0
0.06
.1
.15
.2
00
00
00
00
00
00
00
00
00
0
8
7
6
5
4
3
2
1
100
0
0
0
0
0.3
0
0
.25
50
.4
.5
0
4
00
500
600
wave length (nm)
4
00
500
wave length (nm)
600
c) 1000
(
[
DMA] mole/L
0
0
0
0
0
.0
.07
.1
8
6
4
2
00
00
00
00
0
.15
.2
80
70
60
50
40
30
0.25
0
0
0
.3
.45
.65
2
1
0
0
0
4
00
450
500
550
wavelength (nm)
4
00
500
wavelength (nm)
Fig. 7 The fluorescence spectra of a STBT, b BTN1 and c BTN2 with DMA in cyclohexane
4
DMA
DMA
4
Aniline
DABCO
3
2
1
0
3
2
1
0
TEA
Aniline
DABCO
TEA
(
BTN1)
(
STBT)
0
.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
0.0
0.3
0.6
0.9
1.2
1.5
[Q] mole/L
[Q] mole/L
Fig. 8 Stern–Volmer plots for the fluorescence quenching of STBT and BTN1 using different amines as
quenchers in ACN and cyclohexane, respectively
-
3
0
centrations, as in case of STBT, BTN and BTN /DMA pairs in ACN. Similar experi-
.02–1.6 molÁdm . However, they show an upward curvature at higher quencher con-
1
2
mental results were also observed for 3-carboxy-5,6-benzocoumarin, 1,4-bis[2-(2-
methylphenyl)ethenyl]-benzene and 5,6-benzoquinoline [18–20]. In such cases, the values
of KSV were calculated from the linear parts of the plots, which indicate a dynamic process
123

J Solution Chem
Fig. 9 The absorption spectra of
BTN with DMA in MeOH
0.8
1
BTN1
-
--- BTN +DMA
1
0.6
0.4
0.2
0
.0
2
50
300
350
Wavelength (nm)
400
450
in which quenching is mainly due to collision. The upward curvature suggests that the
quenching mechanism is not purely diffusion controlled and may be associated with either
ground state complex formation or to a transient, where the amine can quench the fluo-
rescence of the diarylethylenes without collision (presence of a quenching sphere). To see
whether the ground state complex formation is partly involved, the extended Stern–Volmer
equation [21, 22] was tested:
ðI0=IÞ À 1
À
Á
¼
KSV þ Kg þ KSVKg½QŠ
ð3Þ
½QŠ
where KSV is the Stern–Volmer constant and K is the ground state association constant.
g
From Eq. 3, the values of KSV and K can be easily determined by least-squares fit of the
g
relation between [(I /I)–1]/[Q] and [Q]; the values of K according to Eq. 3 were found to
o
SV
be imaginary in all the solvents. Therefore, the role of ground state complex formation is
ruled out in the present cases, which is also confirmed by the absorption spectral mea-
surements (Fig. 9). These finding show that Eq. 3 is not applicable to the data responsible
for the observed positive deviation in the Stern–Volmer plots. Thus, analysis of this
deviation was made using the quenching sphere static model. The Stern–Volmer constants
were determined from the slopes of the corresponding plots and summarized in Tables 5,
6
, 7, and 8.
To calculate the bimolecular quenching rate constant (k ), the value of the fluorescence
q
lifetime should be known. Unfortunately this technique was unavailable and its value was
measured for BTN and BTN in a few solvents using a frequency modulation technique.
The values in the remaining solvents as well as for STBT in all solvents were theoretically
1
2
calculated. In this aspect, the fluorescence quantum yield (U ) was firstly determined in all
f
used solvents. The fluorescence quantum yield was determined relative to quinine sulphate
-
3
in 0.05 molÁdm H SO (U = 0.54) [23], using solutions having absorbance less than 0.3
2
4
f
at the excitation wavelength. The U values were calculated according to the following
f
equation:
1
23

J Solution Chem
1 2
Table 5 Stern–Volmer constants for STBT, BTN and BTN with different quenchers in MeOH
-
1
3
9
-1
3
-1
1/2
-3/2
Quencher
K
SV (mol Ádm )
k
q
9 10 (mol Ádm Ás
)
[Q]1/2 (mol Ádm
)
U
f
s (ns)
STBT
0.015
0.108
DMA
2.8
26.01
2.96
12.95
6.1
0.24
–
TEA
0.32
1.4
Amiline
DABCO
0.66
1.42
0.66
BTN
1
0.02
0.093
0.047
DMA
2.3
0.1
1.3
0.19
24.5
1.03
13.4
2.03
0.42
–
TEA
Amiline
DABCO
0.8
0.4
BTN
2
0.047
DMA
3.35
0.34
1.72
0.72
71.2
7.32
36.5
15.3
0.16
–
TEA
Amiline
DABCO
0.54
1.4
Table 6 Stern–Volmer constants for STBT, BTN
1
2
and BTN with different quenchers in EtOH
-
1
3
9
-1
3
-1
1/2
-3/2
Quencher
K
SV (mol Ádm )
k
q
9 10 (mol Ádm Ás
)
[Q]1/2 (mol Ádm
)
U
f
s (ns)
STBT
0.019
0.133
DMA
3.11
0.34
1.7
23.4
2.5
0.2
–
TEA
Amiline
DABCO
12.4
5.7
0.64
1.07
0.76
BTN
1
DMA
2.9
0.14
1.5
0.6
30.1
1.5
0.26
–
0.027
0.049
0.096
0.052
TEA
Amiline
DABCO
15.2
6.3
0.7
1.4
BTN
2
DMA
3.8
0.38
1.8
0.9
72.6
7.4
0.22
–
TEA
Amiline
DABCO
35.4
16.9
0.54
0.9
R
s
2
u
2
s
u
f
s
f
n A I ðkfÞdkf
u
f
s
f
U ¼ U
R
ð4Þ
u
n A I ðkfÞdkf
where n is the refractive index of the solvent, A is the absorbance, U is the quantum yield,
f
and the integrals denote the computed area under the corrected fluorescence bands. The
subscripts s and u refer to the standard and test sample, respectively. Tables 5, 6, 7, and 8,
show the fluorescence quantum yield of the investigated diarylethylenes in different sol-
vents at room temperature.
123

J Solution Chem
1 2
Table 7 Stern–Volmer constants for STBT, BTN and BTN with different quenchers in ACN
-
1
3
9
-1
3
-1
1/2
-3/2
Quencher
K
SV (mol Ádm )
k
q
9 10 (mol Ádm Ás
)
[Q]1/2 (mol Ádm
)
U
f
s (ns)
STBT
0.024
0.045
0.011
0.19
DMA
0.58
2.41
3.2
30.4
12.7
16.6
8.8
0.25
0.46
0.35
0.33
TEA
Amiline
DABCO
1.7
BTN
1
0.114
0.065
DMA
4.1
1.99
2.6
1.1
35.8
17.5
22.5
9.3
0.19
0.62
0.46
0.92
TEA
Amiline
DABCO
BTN
2
DMA
6.1
93.6
38.4
62.4
30.8
0.24
0.70
0.33
0.51
TEA
2.50
4.1
Amiline
DABCO
2.01
1
Table 8 Stern–Volmer constants for STBT, BTN and BTN2 with different quenchers in cyclohexane
-
1
3
9
-1
3
-1
1/2
-3/2
Quencher
K
SV (mol Ádm )
k
q
9 10 (mol Ádm Ás
)
[Q]1/2 (mol Ádm
)
U
f
s (ns)
STBT
0.14
0.34
0.44
0.392
DMA
6.9
2.9
5.95
6.6
17.7
7.3
0.12
0.37
0.16
0.12
TEA
Amiline
DABCO
15.1
16.7
BTN
1
0.38
0.23
DMA
4.99
3.6
13.13
9.5
0.23
0.4
TEA
Amiline
DABCO
4.6
11.97
17.2
0.24
0.10
6.53
BTN
2
DMA
8.1
4.4
6.8
8.74
35.2
19.04
29.5
38.0
0.15
0.3
TEA
Amiline
DABCO
0.23
0.1
The radiative rate constants were determined using the following equation:
2
f
kR ¼ 0:67v f
where v is the frequency of the emission maximum (cm ) and f is the oscillator strength,
ð5Þ
2
f
-1
which was calculated as:
1
23

J Solution Chem
Table 9 Fluorescence lifetime
(sf, ns) of diarylethylenes in var-
ious solvents calculated experi-
Compound
Experimental s
f
Calculated s
f
BTN
1
BTN
2
BTN BTN
1
2
STBT
mentally and theoretically
EtOH
0.9
–
0.05
–
0.096
0.093
0.114
0.38
0.052
0.047
0.065
0.23
0.133
0.108
0.19
MeOH
ACN
0.11
0.34
0.06
0.21
Cyclohexane
0.392
f ¼ 4:32 Â 10À9emaxD vꢀ 1=2
ð6Þ
where D vꢀ 1=2 is the wave number at the half band width of the absorption band.
Finally, the fluorescence lifetime (s ) was calculated by:
f
Uf
sf ¼
kR
ð7Þ
The values of fluorescence lifetime calculated theoretically and the experimental ones
for the investigated compounds in different solvents are summarized in Table 9. To test the
precision of the calculated values, they were compared with the experimental ones.
Therefore, the calculated s values could be used safely for calculation of the bimolecular
f
quenching constants. The calculated bimolecular rate constants are summarized in
Tables 5, 6, 7, and 8. From the data, it was found that the k values for diarylethyles–DMA,
q
–
TEA and –aniline systems in ACN are higher than those in cyclohexane, suggesting the
greater charge transfer character of the excited complex in ACN. Also, polar solvents
like ACN enhance the dissociation of the formed ion pairs into solvent separated ones,
which reduces the possibility of back electron transfer. In contrast, the behavior in the
case of diarylethylenes–DABCO pairs is different, where the values of k in ACN are
q
smaller than those in cyclohexane, which may be due to chemical nature of DABCO. For
example, k values for STBT, BTN and BTN –DMA pairs are 30.4, 35.8 and
q
1
2
9
-1
3 -1
9
3
3.6 9 10 mol Ádm Ás in ACN, while in cyclohexane, the values are 17.7, 13.13 and
9
-1
3 -1
5.2 9 10 mol Ádm Ás , respectively. But for STBT, BTN and BTN –DABCO pairs,
1 2
3 -1
9
-1
the k values in ACN are 8.8, 9.3 and 30.8 9 10 mol Ádm Ás while in cyclohexane, the
q
9
-1
3 -1
values are 16.7, 17.2 and 38.0 9 10 mol Ádm Ás . As can be seen, the values of k for
q
all if the amines in MeOH and EtOH are smaller than those in ACN. This is due to the
hydrogen-bond donating ability of alcohols, which suppress the electron transfer process
from the hydrogen bonded amines to the diarylethylenes [24]. For example, the values of
9
kq of STBT, BTN and BTN –TEA pairs in MeOH are 2.96, 1.03 and 7.23 9 10
1
2
-1
3
-1
9
-1
3 -1
mol Ádm Ás while in ACN, the values are 12.7, 17.5 and 38.4 9 10 mol Ádm Ás
,
respectively. Additionally, no general trend was observed for the effect of solvent polarity
for quenching of the investigated diarylethylenes using amines which may be due to the
antagonistic effect of other factors such as hydrogen-bonding character and viscosity of
solvents, which affect the quenching efficiency. Also, aliphatic amines like TEA and
DABCO are less efficient quenchers than aromatic amines such as DMA and aniline. This
is in agreement with that reported by Inada et al. [25], who found that N-donors are less
efficient than p-ones by 0.5 eV. Also, the calculated half quencher concentrations [Q]1/2
(which is the quencher concentration at which I = 1/2 I ) confirm this finding; thus on
using aliphatic amines, especially TEA, the [Q]1/2 values are very large compared to those
f o
of the aromatic amines. As example, the [Q]1/2 values of BTN –DMA, –aniline, –DABCO
2
123

J Solution Chem
Table 10 Energy of singlet–singlet transition, E00, oxidation potentials, E eD xid, and basicity (pK
D
) of the
b
amines
Parameter
Aromatic amines
Aniline
Aliphatic amines
TEA
DMA
DABCO
D
0
a
E
E
(eV)
0
3.91
0.93
9.37
-1
3.81
0.78
5.1
[4.95
1.05
[4.35
0.70
5.3
D
exid
(V)
pK
b
3.19
a
1
eV corresponds to 96.5 kJÁmol
-
3
and –TEA, pairs are 0.24, 0.33, 0.51 and 0.7 molÁdm , respectively, in ACN and in
-
3
cyclohexane the corresponding values are 0.15, 0.23 and 0.1 and 0.3 molÁdm
,
respectively.
A cursory glance at data reported in Table 10, shows that the k values for different
q
diarylethylenes–amine pairs do not correlate with the basicity (pK ) of the amines [26].
b
Accordingly, the k values for aliphatic amines are always lower than those of the aromatic
q
amines, though it was expected to be opposite on the basis of their pK values. Similarly,
b
among different aliphatic amines, although their pK values are quite similar, they have
b
very different k values that cannot be justified on the basis of the difference in basicity of
q
the aliphatic amines.
Considering the quenching mechanism, although some alternatives can be considered,
based on the results, the ET from ground state amines to the excited fluorescent
diarylethylenes seems to be the most likely mechanism, as suggested in earlier literature
[
27]. For the present systems, ED values of the amines are much higher than those of the
100
investigated diarylethylenes; ED (the energy of the singlet–singlet transition was calcu-
100
lated from the intersection between the normalized absorption and emission spectra
Table 11), which rules out the energy transfer as a quenching mechanism. Also, the
bimolecular quenching constant for the different diarylethylenes–amine systems gradually
decreases with increasing oxidation potentials (E eD xid) of the amines. Figure 10 shows the
correlation between k and ED for diarylethylenes–amine pairs in cyclohexane, as an
q
exid
example. As can be seen, BTN is more sensitive to the change in the oxidation potential of
2
the used amine compared to BTN and STBT.
1
Since, the rate of ET reaction depends on the change in Gibbs energy, therefore, to give
a quantitative description of ET process as the mechanism, it was necessary to estimate the
change in the standard state Gibbs energy (DG°) for each diaryletylene–amine system in
ACN and cyclohexane, as examples. The usual expression for calculation of DG° is Rehm–
Weller equation [28], as follows:
A
0
A
red
Table 11 Energy of singlet–singlet transition, E0 , and reduction potentials, E , of the investigated
-
1
compounds, measured in pH = 7 at scan rate = 500 mVÁs
Compound
STBT
BTN
1
BTN
2
A
a
E
E
a
(eV)
3.4
3.12
1.23
3.25
1.25
0
0
A
red
(V)
1.29
-1
1
eV corresponds to 96.5 kJÁmol
1
23

J Solution Chem
5
4
3
2
1
0
0
0
0
0
DABCO
DMA
aniline
TEA
0
0
.65
0.70 0.75
0.80
0.85 0.90
D)
0.95
1.00 1.05
1.10
(
Eoxd (V)
oxid
q 1
Fig. 10 The correlation of k -(unfilled circle) and (unfilled
versus ED for STBT-(filled square), BTN
triangle) BTN –amine pairs in cyclohexane
2
e2
DGo ¼ Eoxid À Ered À E00 À
ð8Þ
er
2
where e /er is the columbic attraction energy. E is the energy required for the transition
0
0
of diaryethylenes from ground state (S ) to first excited electronic state (S ), which is
0
1
obtained from the intersection points of normalized absorption and emission spectra. ED
exid
and EA denote the oxidation potential of the donor and reduction potential of the acceptor,
red
respectively. In the present case, the value of EA was estimated using cyclic voltammetric
red
measurements as discussed previously. The fourth term is negligible in highly polar sol-
vents like ACN [29]. It is not possible to measure the redox potentials in cyclohexane;
hence the Gibbs energy change (DG°) in cyclohexane cannot be calculated using the
previous equation. To obtain DG° in cyclohexane the corresponding values in ACN were
-1
shifted by a value of 0.698 eV (1 eV corresponds to 96.5 kJÁmol ) following the equation
[
30]:
ꢀ
ꢀ
DG ðcyclohexaneÞ ¼ DG ðACNÞ þ 0:698
ð9Þ
As seen from Table 12, the electron transfer reaction from the ground state amines to
the excited diarylethylenes is highly exothermic and energetically favorable. Also, in all
cases the rate constants are found to be the smallest for TEA, which is consistent with the
low (DG°) values. The most interesting observation is the inversion of the ET rate, Fig. 11,
as predicted by Marcus outer sphere electron transfer theory [31], where the rate decreases
in highly energetic systems. However, in cyclohexane the ET reaction rate falls in the
normal and diffusion limit regions.
As mentioned previously, the fluorescence quenching of diarylethylenes in non-polar
solvent such as cyclohexane is accompanied by the appearance of a new broad fluores-
cence band at longer wavelengths, which was attributed to exciplex formation (inset in
Fig. 7). The exciplex emission spectra are associated with the quenching of STBT, BTN1
123

J Solution Chem
Table 12 Gibbs energy changes and bimolecular rate constants of ET reactions in diarylethylenes–amine
systems in ACN and cyclohexane
Compound Quencher ACN
Cyclohexane
9
9
DG°
k
q
9 10
-1
DG°
(kJÁmol
q
k 9 10
-
1
)
3
-1
-1
-1
3 -1
(
kJÁmol
(mol Ádm Ás
)
)
(mol Ádm Ás
)
STBT
TEA
-102
-114
-128
12.67
16.63
30.36
8.78
-35
-46
-61
-69
-14
-25
-40
-74
-24
-36
-50
-72
7.27
15.1
Aniline
DMA
17.65
16.7
DABCO -136
BTN
1
TEA
-81
-93
17.45
22.5
9.97
Aniline
DMA
11.97
13.13
17.2
-107
35.78
16.0
DABCO -142
BTN
2
TEA
-92
-93
38.4
19.04
29.47
35.17
38.0
Aniline
DMA
62.4
-107
93.6
DABCO -140
30.84
(a)
(b) 24.5
2
2
2
2
2
2
2
5.0
4.5
4.0
3.5
3.0
2.5
2
2
2
2
4.0
3.5
3.0
2.5
2.0
-
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1
1.6
-1.4
-1.2
-1.0
-0.8
-0.6
O
O
−
ΔG (eV)
−
ΔG (eV)
Fig. 11 Variation of ln k
2
BTN –amines pairs in a ACN and b cyclohexane
q
1
verus DG° for STBT-(filled square), BTN -(unfilled circle) and (unfilled triangle)
and BTN –DMA pairs in non-polar solvent, cyclohexane. The observed energies at the
2
exciplex fluorescence maxima (E ) are listed in Table 13 and can be compared with the
1
theoretical energies E according to Rehm and Weller:
2
ðDÞ
ðAÞ
red
E2 ¼ E À E À 0:15 Æ 0:1 eV
ð10Þ
oxd
For most stable exciplexes the correlation between E and E is better. The effect of
1
2
structure on the stability of the exciplexes can be seen from the calculation of the exciplex
dissociation enthalpies (E ) [32]:
3
1
23

J Solution Chem
1
Table 13 Comparison of the observed energy at exciplex emission maxima, E , with the theoretical energy
according to Rehm and Weller, E and dissociation enthalpies, E , of the diarylethylenes–DMA exciplexes
2 3
in cyclohexane
STBT
BTN
1
BTN
2
k
E
1
E
(eV)
2
E
(eV)
3
k
(nm) (eV)
E
1
E
(eV)
2
E
(eV)
3
k
E
(nm) (eV)
1
E
(eV)
2
E
3
(eV)
a
a
a
(nm) (eV)
491
2.52
1.92
3.7
534 2.1
1.86
3.4
507 2.44
1.88
3.59
a
-1
1
eV corresponds to 96.5 kJÁmol
D
oxid
A
red
E3 ¼ E00 À E
À E À 0:13 ev
ð11Þ
where Eoo is the singlet excitation energy of diaryletylenes. The higher exciplex dissoci-
ation enthalpies, Table 13, indicate strongly bound exciplexes.
In the line of the above results, it can be concluded that the fluorescence quenching of
1
,2-diarylethylenes by aliphatic and aromatic amines is due to the electron transfer (ET)
from the ground state amines to the excited diarylethylenes. Aliphatic amines like TEA and
DABCO are less efficient quenchers than aromatic amines such as DMA and aniline. The
hydrogen-bond donating ability of protic solvents like MeOH or EtOH plays an important
role in the quenching process, leading to an observable decrease in quenching efficiency
(
kq) compared to an aprotic solvent like ACN. The observed kq values for different
diarylethylenes–amine pairs do not correlate with the basicity (pK ) of the amines but
b
decrease with increasing the oxidation potentials of the amines. The ET reaction from the
ground state amines to excited state diarylethylenes is highly exothermic and energetically
favorable, and k values have been correlated with DG° values for the ET reactions from
q
Marcus outer-sphere ET theory.
References
1
2
3
4
5
. Pang, S., Hyun, H., Lee, S., Jang, D., Lee, M.J., Kang, S.H., Ahn, K.-H.: Photoswitchable fluorescent
diarylethene in a turn-on mode for live cell imaging. Chem. Commun. 48, 3745–3747 (2012)
. Sabbioni, E., Blanch, N., Baricevic, K., Sera, M.A.: Effects of trace metal compounds on HIV-1 reverse
transcriptase. Biol. Trace Etem. Res. 68, 107–119 (1999)
. Krieg, R., Halbhuber, K.J.: Novel oxidative self-anchoring fluorescent substrates for the histochemical
localization of endogenous and immunobound peroxidase activity. J. Mol. Histol. 35, 471–487 (2004)
. Tanifuji, N., Irie, M., Matsuda, K.: Photoswitching of solvatochromism using diarylethenes with 2,5-
disubstituted 3-thienyl unit. Chem. Lett. 36, 1232–1233 (2007)
. Zhao, M., Samoc, M., Prasad, P.N., Reinhardt, B.A., Unroe, M.R., Prazak, M., Evers, R.C., Kane, J.J.,
Jariwala, C., Sinsky, M.: Studies of third-order nonlinearities of model compounds containing ben-
zothiazole, bensimidazole and benzoxazole units. Chem. Mater. 2, 670–678 (1990)
6
7
8
9
. Jeong, Y.C., Ik Yang, S., Ahn, K.-H., Kim, E.: Highly fluorescent photochromic diarylethene in the
closed-ring form. Chem. Commun. 19, 2503–2505 (2005)
. Marcus, R.A., Sutin, N.: Electron transfers in chemistry and biology. Biochem. Biophys. Acta. 811,
265–322 (1985)
. Yoshihara, K., Tominaga, K., Nagasawa, Y.: Effects of the solvent dynamics and vibrational motions in
electron transfer. Bull. Chem. Soc. Jpn 68, 696–712 (1995)
. Lewis, F.D., Dykstra, R.E.: Exciplex formation and behavior in the 9,10-dicyanoanthracene–stilbene
system. J. Photochem. Photobiol., A 49, 109–119 (1989)
123

J Solution Chem
1
1
1
0. Allonas, X., Jacques, P.: Factors affecting adiabaticity in bimolecular photoinduced electron-transfer
reaction between anthracene-derivatives and organic donors. Chem. Phys. 215, 371–378 (1997)
1. Nad, S., Pal, H.: Photo induced electron transfer from aliphatic amines to coumarin dyes. J. Chem. Phys.
116, 1658–1670 (2002)
2. Satpati, A.K., Nath, S., Kumbhakar, M., Maity, D.K., Senthilkumar, S., Pal, H.: Bimolecular electron
transfer reactions in coumarin–amine systems: donor–acceptor orientational effect on diffusion-con-
trolled reaction rates. J. Mol. Struct. 878, 84–94 (2008)
1
1
1
1
3. Vernigor, E.M., Shalaev, V.K., Novoseltesva, L.P., Lukyanets, E.A., Ustenko, A.A., Zvolinskii, V.P.,
Zakharov, V.F.: Khim. Geterotsikl. Soedin. 5, 604 (1980)
4. Bard, A.J., Faulkner, L.R.: Electrochemical Methods-Fundamentals and Applicaions. Wiley, New York
(1980)
5. Ho, T.I.: Structure reactivity relationship for fluorecence quenching of trans-stilbene by tertiary amines.
J. Photochem. 36, 185–192 (1987)
6. Hirendra, G.N., Haridas, P., Dipak, K.P., Tulsi, M., Jai, P.M.: Interaction of the excited singlet state of
disubstituted anthraquinones with aliphatic and aromatic amines: a fluorescence quenching study.
J. Photochem. Photobiol., A 73, 17–22 (1993)
1
1
1
2
7. Chen, C.-C., Pan, K., Wang, S.-L., Ho, T.-I.: Styrylthio phene-amine exciplexes. J. Lumin. 71, 321–328
(1997)
8. Tablet, C., Hilledrand, M.: Quenching of the fluorescence of 3-carboxy-5,6-benzocoumarin by aromatic
amines. J. Photochem. Photobiol A: Chem. 189, 73–79 (2007)
0
0
9. Thipperudrappa, J., Biradar, D.S., Hanagodimath, S.M.: Fluorescence quenching of 2,2 -dimethyl-p-
terphenyl by carbon tetrachloride in different solvents and temperatures. J. Lumin. 126, 339–346 (2007)
0. Carrigan, S., Doucette, S., Jones, C., Marzzacco, C.J., Halpern, A.M.: The fluorescence quenching of
-
-
-
-
5
,6-benzoquinoline and its conjugate acid by Cl , Br , SCN and I ions. J. Photochem. Photobiol., A
99, 29–35 (1996)
2
2
2
1. Behera, P.K., Mishra, A.K.: Static and dynamic model for 1-naphthol fluorescence quenching by carbon
tetrachloride in dioxane–acetonitrile mixtures. J. Photochem. Photobiol., A 71, 115–118 (1993)
2. Zeng, H., Durocher, G.: Analysis of fluorescence quenching in some antioxidants from nonlinear Stern–
Volmer plots. J. Lumin. 63, 75–84 (1995)
3. Maus, M., Rellig, W., Bonafoux, D., Lapouyade, R.: Photoinduced intramolecular charge transfer in a
series of differently twisted donor–acceptor biphenyls as revealed by fluorescence. J. Phys. Chem. A
103, 3388–3401 (1999)
2
4. Oh, S.-C., Shirota, Y., Mikawa, H., Kusabayashi, S.: The effect of the hydrogen bonding on the
photoinduced electron transfer reaction of a naphthalene–triethylamine system in the presence of
alcohols. Chem. Lett. 12, 2121–2124 (1986)
2
5. Inada, T.N., Kikuchi, K., Takahashi, Y., Ikeda, H., Miyashi, T.: A comparative study on electron
transfer by aliphatic and aromatic amines. J. Photochem. Photbiol. A 137, 93–97 (2000)
6. Aggarwal, V.K., Emme, I., Fulford, S.Y.: Correlation between pKa and reactivity of quinuclidine-based
catalysts in the Baylis–Hillman reaction—discovery of quinuclidine as optimum catalyst leading to
substantial enhancement of scope. J. Org. Chem. 68, 692–700 (2003)
2
2
2
2
3
3
3
7. Castner Jr, E.W., Kennedy, D., Cave, R.J.: Solvent as electron donor: donor/acceptor electronic cou-
pling is a dynamical variable. J. Phys. Chem. A 104, 2869–2885 (2000)
8. Rehm, D., Weller, A.: Bonding state and fluorescence spectra of hetero-excimer. Z. Phys. Chem. NF 69,
183–200 (1970)
9. Weller, A.: Photoinduced electron transfer in solution: exciplex and radical ion pair formation free
enthalpies and their solvent dependence. Z. Phys. Chem. NF 133, 93–98 (1982)
0. Chen, J.M., Ho, T.-I., Mou, C.-Y.: experimental investigation of excited-state electron transfer reaction:
effect of free energy and solvent on rates. J. Phys. Chem. 94, 2889–2896 (1990)
1. Jortner, J., Bixon, M. (eds.): Electron Transfer from Isolated Molecules to Biomolecules (Adv. Chem.
Phys. Part 1 and 2), vol. 106, 107. Wiley, New York (1999)
2. Knibbe, H., Rehm, D., Weller, A.: Thermodynamics of the formation of excited eda (electron donor–
acceptor) complexes. Ber. Bunsen-Ges. Phys. Chem. 73, 839–845 (1969)
1
23