Effect of β-cyclodextrin on the thermal Cis-trans isomerization of azobenzenes

Article abstract of DOI:10.1021/jo951028+

The cis-trans thermal isomerization of p-methyl red (1), o-methyl red (2), and methyl orange (3) was inhibited by β-cyclodextrin (β-CD) at constant pH. Their isomerization rate decreased 4, 8, and 1.67 times, respectively, in a solution containing 0.01 M β-CD. This effect can be attributed to the formation of an inclusion complex between the substrate and β-CD which hinders the rotation of the N=N bond. The isomerization rate of methyl yellow (4), 4-(dimethylamino)-4′-methoxy-azobenzene (5), and naphthalene-1-azo[4′-(dimethylamino)benzene] (6) was not affected by β-CD due to the presence of an organic cosolvent in the solution which displaces the azobenzene from the cavity, and the complex formed is probably equatorial. In addition, the transition state for the isomerization of compounds 1-3 involves rotation and that of 4-6, which have only electron-donating groups, inversion. This latter process brings about less volume change than rotation so it is less hindered by the complexation with β-CD.

Full text of DOI:10.1021/jo951028+

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
3446 J . Org. Chem. 1996, 61, 3446-3451
Effect of â-Cyclod extr in on th e Th er m a l Cis-Tr a n s Isom er iza tion
of Azoben zen es
Ana M. Sanchez and Rita H. de Rossi*
Instituto de Investigaciones en Fı´sico-Quı´mica de Co´rdoba (INFIQC), Facultad de Ciencias Quı´micas,
Departamento de Quı´mica Orga´nica, Universidad Nacional de Co´rdoba, Sucursal 16, CC 61,
5016 Co´rdoba, Argentina
Received J une 6, 1995X
The cis-trans thermal isomerization of p-methyl red (1), o-methyl red (2), and methyl orange (3)
was inhibited by â-cyclodextrin (â-CD) at constant pH. Their isomerization rate decreased 4, 8,
and 1.67 times, respectively, in a solution containing 0.01 M â-CD. This effect can be attributed
to the formation of an inclusion complex between the substrate and â-CD which hinders the rotation
of the NdN bond. The isomerization rate of methyl yellow (4), 4-(dimethylamino)-4′-methoxy-
azobenzene (5), and naphthalene-1-azo[4′-(dimethylamino)benzene] (6) was not affected by â-CD
due to the presence of an organic cosolvent in the solution which displaces the azobenzene from
the cavity, and the complex formed is probably equatorial. In addition, the transition state for the
isomerization of compounds 1-3 involves rotation and that of 4-6, which have only electron-
donating groups, inversion. This latter process brings about less volume change than rotation so
it is less hindered by the complexation with â-CD.
In tr od u ction
The isomerization of azobenzenes is a subject of great
current interest due to their possible application in
energy storage systems or in photochemical devices.7 The
thermal Z-E isomerization of p-aminoazobenzenes pro-
ceeds by two routes as shown in Scheme 1: (a) inversion
of one or both of the nitrogen atoms through a linear (sp-
hybridized) transition state in which the double bond is
retained and (b) disruption of the nitrogen-nitrogen
p-bond, with rotation around the remaining σ-bond giving
the E isomer (rotation mechanism).8,9
Since the isomerization reaction takes place with a
considerable volume change, we considered it of interest
to determine the effect of â-cyclodextrin (â-CD) on the
thermal cis-trans isomerization of azo compounds 1-6,
and these results are reported here.
Cyclodextrins are largely used as hosts for organic
molecules. They possess a hydrophobic cavity due to a
cyclic arrangement of six (R-CD), seven (â-CD), and eight
(γ-CD) D-(+)-glucopyranose units, linked by R-(1,4) gly-
cosidic bonds. The macrocycle can be described as a
truncated cone, the narrow rim bearing the primary -OH
groups and the wide rim the secondary -OH groups.1
The binding of a molecule or ion to the cavity of a
cyclodextrin results in the formation of an inclusion
complex that involves relatively weak nonspecific inter-
action. The association of the guest molecule normally
occurs by partial or full fitting of the cavity. The main
driving forces for substrate and host binding are as
follows: (a) Van der Waals forces, (b) hydrophobic
interactions, and (c) hydrogen bonding. The inclusion
phenomenon produces changes in the processes under-
gone by the guest due to steric effects as well as changes
in the microenvironment.
Several kinetic and equilibrium studies have been
reported on the inclusion complexes of cyclodextrins with
azo dyes of variable structural complexities.2,3 Azoben-
zene derivatives exhibit photoinduced reversible cis-
trans isomerism, and the formation of an inclusion
complex of azobenzene with R-CD or â-CD leads to a
decrease in the quantum yield for the trans-cis isomer-
ization of azobenzene.4 The cis-trans photoisomerization
of p-methyl red has been investigated in the host-guest
Langmuir-Blodgett films prepared with amphiphilic
â-CD derivatives.5,6
Resu lts a n d Discu ssion
X Abstract published in Advance ACS Abstracts, April 1, 1996.
(1) Bender, M. L.; Komiyama, M. Cyclodextrins Chemistry; Springer
Verlag: New York, 1978.
(2) Clarke, J . R.; Coates, J . H.; Lincoln, S. F. Carbohydr. Res. 1984,
127, 181-191.
(3) Hersey, A.; Robinson, B. H. J . Chem. Soc., Faraday Trans. 1
1984, 80, 2039-2052.
(4) Bortolus, P.; Monti, S. J . Phys. Chem. 1987, 91, 5046-5050.
(5) Yabe, A.; Kawabata, Y.; Niino, M.; Ouchi, A.; Takahashi, H.;
Tamura, S.; Tagaki, W.; Nakahara, H.; Fukuda, K. Chem. Lett. 1988,
1.
(6) Yabe, A.; Kawabata, Y.; Niino, M.; Ouchi, A.; Takahashi, H.;
Tamura, S.; Tagaki, W.; Nakahara, H.; Fukuda, K. Thin Solid Films
1988, 160, 33
Equ ilibr iu m Stu d ies. The spectrum of compounds
1-3 was determined in water, those of 4 and 5 in
ethanol-water 20/80 v/v, and the one for 6 in ethanol-
water 50/50 v/v. The solutions were prepared with NaOH
at the same concentration used for the kinetic studies
(see below). The fact that compounds 4-6 are too
(7) Sekkat, Z.; Dumont, M. Appl. Phys. B 1992, 54, 486.
(8) Wildes, P.; Pacifici, J . G.; Irick, G.; Whitten, D. G. J . Am. Chem.
Soc. 1971, 93, 2004.
(9) Asano, T.; Okada, T. J . Org. Chem. 1984, 49, 4387.
S0022-3263(95)01028-0 CCC: $12.00 © 1996 American Chemical Society

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Thermal Cis-Trans Isomerization of Azobenzenes
J . Org. Chem., Vol. 61, No. 10, 1996 3447
Sch em e 1
F igu r e 1. Spectra of 3 in the presence of â-CD. Tempera-
ture: 25 °C; solvent: H2O; [methyl orange] ) 6.8 × 10-5 M;
[â-CD] ) 0 (s) to 10 (‚‚‚) mM; NaOH 0.1 M.
insoluble in water precludes the determinations without
an organic cosolvent. In all cases the addition of â-CD
to the diluted solution of the different substrates shifted
the maximum wavelength to the blue and decreased the
extinction coefficient. The spectrum of compound 5 was
also measured in water-acetonitrile 80:20 v/v containing
NaOH 0.1 M where it has λmax 444 nm and a shoulder at
412 nm. The ratio of extinction coefficients is 1.21, and
the addition of 0.01 M of â-CD lowered this ratio to 1.07.
The changes in the spectrum in the presence of â-CD
show that there must be some specific interaction be-
tween compounds 1-6 and â-CD because a similar
addition of soluble starch did not modify the spectrum.
A well-defined isosbestic point was not obtained in any
case, suggesting that there is more than one type of
interaction.10 The spectrum for compound 3 is shown in
Figure 1.
Ta ble 1. Effect of â-Cyclod extr in on Cis-Tr a n s
Isom er iza tion of Azoben zen esa
o-methyl redb
kobs, s-1
p-methyl redc
kobs, s-1
methyl orangec
[â-CD], M
kobs, s-1
0.0000
0.297
0.053d
0.216
0.100
0.0606
0.063
0.179
0.173
0.0001
0.0002
0.00025
0.0003
0.0005
0.0008
0.0010
0.0020
0.0030
0.0040
0.0050
0.0060
0.0075
0.0090
0.0100
0.170
0.165
0.161
0.159
0.190
0.170
0.117
0.090
0.084
0.054
0.039
0.0328
0.0326
0.024
0.147
0.129
0.127
0.122
If the interaction is of 1:1 stoichiometry, the system
can be described as in eq 1 where S, â-CD, and S‚â-CD
0.0247
0.114
0.109
K11
0.034
0.024
S + â-CD y z S‚â-CD
(1)
0.055d
0.239e
0.083e
0.022
0.0231
0.0203
stand for the substrate molecules, â-cyclodextrin, and the
â-cyclodextrin-substrate complex, respectively. The
change in absorbance, ∆A, at constant wavelength is
given by eq 2 which can be rearranged to eq 3 where
0.0120
0.0150
0.0170
0.101
0.098
0.035
a
Temperature, 25 ( 0.1 °C; ionic strength 0.20 M; rate
constants are average of at least three determinations, and the
b
∆ꢀ11K11[â-CD]o[S]o
mean deviation is less than 5%. [NaOH], 0.01 M; solvent, water.
c [NaOH], 0.1 N; solvent, water. Solvent, EtOH/H2O (20:80);
d
∆A )
(2)
(3)
1 + K11[â-CD]o
[NaOH], 0.01 M. e Solvent, EtOH/H2O (50:50); [NaOH], 0.01 M.
e Soluble starch in concentration equivalent to the â-CD.
[â-CD][S]o
∆A
[â-CD]o
1
)
+
constant K11 for the association of the substrate with CD
(eq 1) is obtained. Alternatively, a nonlinear fit of the
data to eq 2 yields K11 and both methods should give the
same results.10 Only the data for compounds 1-3 fit to
eqs 2 and 3 but for compound 1 and 3 the change in the
wavelength of analysis yields significantly different equi-
librium constants (Table 2) indicating that more than one
type of complex is formed. In a previous study on the
interaction of p-methyl red with R- and â-CD at pH 9.5,
the same equilibrium constant was obtained at different
K11∆ꢀ11
∆ꢀ11
[â-CD]o and [S]o represent the initial concentration of
â-cyclodextrin and that of the substrate, respectively; ∆ꢀ11
is the difference in the molar absorptivity between the
free and complexed substrate. From the slope and
intercept ratio of a plot according to eq 3, the equilibrium
(10) Connors, K. A. Binding constants: The measurement of molec-
ular complex stability; J ohn Wiley and Sons: New York, 1987

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
3448 J . Org. Chem., Vol. 61, No. 10, 1996
Sanchez and de Rossi
Ta ble 2. Ca lcu la ted Ra tes a n d Equ ilibr iu m Con sta n t for th e Cis-Tr a n s Isom er iza tion of Azoben zen es 1-6
compd
kct, s-1
kctCD, s-1
Kc, M-1 d
K11, M-1 e
p-methyl reda
0.100
0.025
4400 ( 1400
1300 ( 300f
400 ( 100g
204 ( 35h
893 ( 80i
699 ( 75j
484 ( 78k
o-methyl reda
0.297
0.179
∼0
654 ( 150
105 ( 20
methyl orangea
0.072
methyl yellowb
0.0088
0.0081l
0.0061m
0.0021m
0.0017m
4-(dimethylamino)-4′-methoxyazobenzeneb
0.0018
0.002
naphthalene-1-azo-[4′-(dimethylamino)benzene]c
Temperature, 25 ( 0.1 °C; solvent, water. Solvent, H2O/EtOH (80:20). c Solvent, H2O/EtOH (50:50). Kinetically determined from
a
b
d
the fit to eq 6. e Spectrophotometric value determined from the fit to eq 2 and 3. f Measured at λ ) 500 nm. λ ) 424 nm. λ ) 479 nm.
g
h
i
j
k
l
m
λ ) 510 nm. λ ) 480 nm. λ ) 465 nm. Observed rate constant in a solution containing â-CD ) 0.006 M. Observed rate constant in
a solution containing â-CD ) 0.01 M.
[S]o[â-CD]oK11(∆ꢀ11 + ∆ꢀ12K12[â-CD])
1 + K11[â-CD]o(1 + K12[â-CD]o)
∆A )
(4)
The fact that the data for 1-3 give a reasonable fit to
eqs 2 and 3 indicates that in the range of â-CD concen-
tration used K12[â-CD] < 1. The different values of the
equilibrium constants obtained at different wavelengths
may be due to different ∆ꢀ11/∆ꢀ12 ratios. Attempts to fit
the data to eq 4 failed. The reason may be that eq 4 has
four adjustable parameters, namely ∆ꢀ11, ∆ꢀ12, K11, and
K12, and the number of data points was 10-12. It was
not possible to get more data because the maximum
change in optical density (∆A) which is obtained for
solutions containing â-CD 0.01 M is smaller than 0.1.
NMR Deter m in a tion s. To obtain more direct evi-
dence for the â-CD inclusion complex structure, we
measured the 1H-NMR spectra of 1 with and without
â-CD in D2O/NaOD 0.1 M. The chemical shifts were
determined relative to water which appears at 4.63 ppm
relative to external tetramethylsilane. In Table 3 are
tabulated the changes in chemical shift (∆δ) of 1 and CD
in the mixed solutions relative to their values in sepa-
rated solutions. The NMR spectrum of pure â-CD
consists of peaks due to five kinds of protons H-1 (4.78
ppm, doublet), H-3 (3.75 ppm, triplet), H-5,6 (3.64-5.58
ppm, unresolved), H-2 (3.35 ppm doublet of doublets), and
H-4 (3.26 ppm, triplet).14 The numbering of protons is
indicated in structure 7.
F igu r e 2. Schematic representation of one of the 1:1 com-
plexes of 1 with â-CD (carboxylate inclusion).
F igu r e 3. Schematic representation of one of the 1:1 com-
plexes of 1 with â-CD (dimethylamino inclusion).
F igu r e 4. Schematic representation of the 1:2 complex of 1
with â-CD.
wavelengths11 and the values of K11 reported, i.e., 3810
M-1 10 and 6230 M-1 12
, are significantly higher than those
of this work. The differences may be attributed to the
fact that our values were determined at higher pH (see
the Experimental Section) and under these conditions a
significant amount of â-CD is ionized because its pKa is
12.3.13 In addition, our spectrophotometric values are
apparent equilibrium constants since there is evidence
that more than one type of complex is formed. Complexes
of the type shown in Figures 2 and 3 or complexes of 1:2
stoichiometry such as the one shown in Figure 4 may also
be formed. The lack of isosbestic points in the spectra is
consistent with the formation of complexes of stoichiom-
etry higher than 1:1. If complexes of 1:2 stoichiometry
were formed the change in absorption with the addition
of â-CD would be given by eq 4.10
There is a significant upfield shift of the H-3 and H-5
signal of â-CD (the two hydrogens inside the cavity) and
H-6. This shift can be attributed to diamagnetic anisot-
ropy of the benzenoid moiety of the included azo com-
pounds. The resonances of the H-1, H-2, and H-4 located
outside of the â-CD torus are relatively unaffectd (∆δ <
0.01 ppm) by the addition of the guest suggesting that
the association does not take place at the exterior of the
torus. Since splitting of the â-CD signals is not observed,
(11) Tawarah, K; Khouri, S. Carbohydr. Res. 1993, 245, 165.
(12) Tawara, K. Ber. Bunsen Ges. 1993, 97, 727.
(13) Van Etten, R. L.; Clowes, G. A.; Sebastian, J . F.; Bender, M. L.
J . Am. Chem. Soc. 1967, 89, 3253.
(14) The hydrogen has been assigned on the basis of literature
data: Guo, Q-X.; Li, Z.-Z.; Ren, T.; Zhu, X-Q.; Liu, Y.-C. J . Incl. Phenom.
Mol. Rec. 1994, 17, 149.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Thermal Cis-Trans Isomerization of Azobenzenes
J . Org. Chem., Vol. 61, No. 10, 1996 3449
Ta ble 3. Ch em ica l Sh ift Va r ia tion s (∆δ) in p p m of th e
Cyclod extr in a n d p-Meth yl Red P r oton s In d u ced by
Com p lexa tion a t 300 K
p-methyl red
â-CDa
∆δ
H-5,5′
H-4,4′
H-3,3′
H-2,2′
-CH3
0.05
0.03
0.00
-0.01
0.03
-0.01
-0.01
-0.05
-0.01
-0.10
-0.06
H-1
H-2
H-3
H-4
H-5
H-6
a
The numbers refer to structure 7.
the inclusion process must be fast in the NMR time scale.
The more affected signals for 1 are those ortho to the
carboxylate group. This may indicate that the major type
of interaction occurs as shown in Figure 2 with the
carboxylate group protruding out of the smaller rim of
the â-CD cavity. The protons bonded to the ring bearing
the dimethyl amino group are not greatly affected but
the methyl signals are shifted downfield significantly.
Since under the conditions of the NMR experiment equal
concentrations of â-CD and 1 are used, it is very unlikely
that complexes of 1:2 stoichiometry are formed; therefore,
the modification of the hydrogens bonded to the two
aromatic rings probably means that complexes of the type
shown in Figures 2 and 3 are formed.
F igu r e 5. Dependence of the observed rate constant with the
â-CD concentration for the cis-trans isomerization of methyl
orange.
Sch em e 2
Kin etic Stu d ies. Solutions of compounds 1-6 in the
same solvent as those used for the spectroscopic studies
and at constant pH were irradiated with a medium-
pressure mercury lamp (wavelength emitting maximally
at 365 nm) until a photostationary state was reached. A
bleaching of the solution occurs at the wavelength of
maximum absorption but the spectrum returns to its
original shape in the dark. The thermal isomerization
reaction of methyl yellow is slow enough to allow for the
determination of the UV-vis spectra as a function of time
and good isosbestic points are obtained.15,16 The solutions
can be irradiated several times, always returning to the
original spectrum which is indicative of a wholly revers-
ible reaction and that the only reaction taking place is
the cis-trans isomerization, eq 5. In the presence of
â-CD similar results are obtained.
rate of isomerization of compounds 4-6 within the
experimental error (Table 2).
The mechanism of cis-trans isomerization in the
presence of â-cyclodextrin can be described as shown in
Scheme 2.
The observed rate constant for Scheme 2 is given by
eq 6 which predicts nonlinear dependence on â-CD
concentration. The data for methyl orange and o- and
kct + KckctCD[â-CD]
kobs
)
(6)
1 + Kc[â-CD]
p-methyl red were fitted to eq 6. For 2 the best fitting
was obtained when the isomerization rate for the in-
cluded substrate is considered zero which indicates that
kct > 10KckctCD[â-CD], and consequently the ratio kct/kct-
CD must be higher than 65.
The calculated equilibrium and rate constants are
shown in Table 2. The values for Kc obtained from the
kinetic measurements are not in good agreement with
the value determined spectrophotometrically (Table 2).
It should be noted that the kinetically determined values
pertain to the inclusion of the cis isomer whereas the
spectrophotometric values correspond to the inclusion of
the trans isomer. Besides, if there is more than one type
of complex formed, the value of Kc kinetically determined
pertains to the complex which more strongly influences
the isomerization rate.
(5)
A significant decrease in the isomerization reaction
rate of p- and o-methyl red is observed in presence of
â-cyclodextrin and a smaller decrease in methyl orange.
Several factors may contribute to the observed effect. The
mechanism proposed for cis-trans isomerization of sub-
strates having electron-donor and electron-acceptor sub-
stituents is rotation around the N-N bond through a
dipolar transition state8 which is favored in polar sol-
vents. Inclusion places the substrate in a more hydro-
The observed rate constant for 1-3 decreases when
â-cyclodextrin is added and it reaches a platteau (Table
1, Figure 5 is representative). The kobs decreases 8.7-
fold for o-methyl red, 4.16-fold for p-methyl red, and 1.64-
fold for methyl orange at â-CD 10-2 M. On the other
hand, the presence of 0.01 M â-CD does not change the
(15) Sanchez, A. M.; de Rossi, R. H. J . Org. Chem. 1993, 58, 2094.
(16) Sanchez, A. M.; de Rossi, R. H. J . Org. Chem. 1995, 60, 2974.