Inhibition of Acid-induced decomposition of diphenyltriazenes by complexation with cyclodextrins

Article abstract of DOI:10.1002/kin.20503

Acid-promotedN Nbond cleavage in 1,3-diphenyltriazenes (X-Ph-N=N-NH-Ph-X X = H, 4-OCH₃), leading to formation of diazonium ions and anilines, is strongly inhibited in aqueous solutions in the presence of cyclodextrins (CDs). The inhibition is ascribed to the formation of inclusion complexes that render the guest diphenyltriazene significantly less basic as a result of the less polar nature of the CD cavity (amicrosolvent effect). Association equilibrium constants for 1:1 host-guest complexes increase in the order α-CD <β-CD ～permethyl-β-CD < hydroxypropyl-β-CD, with values for X = 4-OCH₃ being larger than those for X = H. In the case of α-CD, formation of 2:1 host-guest complexes is also involved.

Full text of DOI:10.1002/kin.20503

Inhibition of Acid-Induced
Decomposition of
Diphenyltriazenes by
Complexation with
Cyclodextrins
´
TINGTING XU, ATEFEH VAEZ-ZADEH ASADI, MONICA BARRA
Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
Received 23 October 2009; revised 3 February 2010; accepted 3 February 2010
DOI 10.1002/kin.20503
Published online 25 June 2010 in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Acid-promoted N N bond cleavage in 1,3-diphenyltriazenes (X-Ph-N=N-NH-Ph-X;
X = H, 4-OCH3), leading to formation of diazonium ions and anilines, is strongly inhibited in
aqueous solutions in the presence of cyclodextrins (CDs). The inhibition is ascribed to the for-
mation of inclusion complexes that render the guest diphenyltriazene significantly less basic as
a result of the less polar nature of the CD cavity (a microsolvent effect). Association equilibrium
constants for 1:1 host–guest complexes increase in the order α-CD <β-CD ∼ permethyl-β-CD
<
hydroxypropyl-β-CD, with values for X = 4-OCH3 being larger than those for X = H. In the
case of α-CD, formation of 2:1 host–guest complexes is also involved. ꢀC 2010 Wiley Periodicals,
Inc. Int J Chem Kinet 42: 567–574, 2010
INTRODUCTION
ity to modify (alkylate) DNA result from the ability
of triazenes to generate diazonium ions via heterolytic
N2 N3 bond cleavage triggered by protonation of N3.
The likelihood of triazenes to undergo (significant) de-
composition under typical human gastric pH condi-
tions, however, limits the capability of aryltriazenes
known for their antiviral [3] and trypanocidal [4] ac-
tivity, for instance, to be administered orally. Such a
limitation could in principle be overcome by devel-
oping coated formulations based, for example, on the
use of host molecules such as cyclodextrins, as drug
carriers.
Cyclodextrins (CDs) are doughnut-shaped cyclic
oligosaccharides of six to eight α-D-glucose units and
represent one of the most commonly employed (water-
soluble) host systems [5]. Since water is the sol-
vent where most biological processes take place, the
Triazenes (compounds characterized by having a dia-
zoamino group, N1 N2 N3<) are commonly used
in organic synthesis to protect or generate amines [1]
and are known as well for their mutagenic and car-
cinogenic properties [2]. Both the utility of triazenes
as protecting groups/source of amines and their capac-
Correspondence to: M o´ nica Barra; e-mail: mbarra@uwaterloo
.ca.
Contract grant sponsor: Department of Chemistry of the Univer-
sity of Waterloo, Waterloo, Canada.
Contract grant sponsor: Natural Sciences and Engineering Re-
search Council (NSERC) of Canada.
Experimentally observed rate constants are available as Support-
ing Information in the online issue at www.interscience.wiley.com.
ꢀ
c 2010 Wiley Periodicals, Inc.

5
68
XU, ASADI, AND BARRA
Figure 1 Target substrates and hosts.
−
5
characterization of inclusion compounds (complexes)
derived from water-soluble CDs continues to be a sub-
ject of significant interest for applications, among oth-
ers, in drug transport and delivery, catalysis, and the
binding and detection of biomolecules [6].
The present article reports on the effects of cy-
clodextrin complexation on the acid-catalyzed decom-
position of 1,3-diphenyltriazene (1) and 1,3-bis-(4-
methoxyphenyl)triazene (2), two diaryltriazenes of
known biological activity [3,7]. Rate constants for
acid-promoted N2 N3 bond cleavage were measured
in the presence of α-, β-, 2-hydroxypropyl-β-, and
permethyl-β-cyclodextrin (Fig. 1), to assess the in-
fluence of host (cavity) size and rim substitution on
triazene degradation.
tions ranging in the order of (3–4) × 10 M. The
buffer concentration employed was 0.05 M, and the
ionic strength of the solutions was kept constant at
0
.5 M using NaCl as compensating electrolyte. Ki-
◦
netic traces were obtained at (21.0 ± 0.1) C by moni-
toring the disappearance of target diphenyltriazenes at
3
52 nm (1) or 367 nm (2). All reactions were followed
until at least 80–90% conversion of the starting mate-
rial was observed. Values for observed rate constants,
as well as for rate coefficients and association equilib-
rium constants, were obtained by fitting to appropriate
functions using the general curve-fitting procedure of
KaleidaGraph (version 3.6.4) from Synergy Software.
Fittings based on Eqs. (2)–(6) were carried out setting
[
CD] = [CD]0, where [CD]0 refers to the total cy-
clodextrin concentration. Proton concentrations were
calculated from the observed pH by using a value of
0
.732 for the proton activity coefficient [9]. Induced
EXPERIMENTAL
Reagents
circular dichroism spectra were recorded on a JASCO
J-715 spectropolarimeter (0.2-nm resolution, 10 accu-
mulations, 1-cm cell) at (24 ± 1) C.
◦
1,3-Diphenyltriazene and 1,3-bis-(4-methoxyphenyl)-
triazene employed were existing samples from a pre-
vious study [8]. Methanol and acetonitrile (OmniSolv
grade; EM Science, Gibbstown, NJ), 2-hydroxypropyl-
β-cyclodextrin (FW = 2,354), α-cyclodextrin, and
β-cyclodextrin (Aldrich, St. Louis, MO), permethyl-
β-cyclodextrin (TCI, Portland, OR), soluble starch
RESULTS AND DISCUSSION
Acid-promoted decomposition of aryltriazenes dis-
solved in aqueous media typically involves general
acid catalysis (A-SE2) [10] or specific acid catalysis
(Baker, Phillipsburg, NJ), and D-glucose (EM Science)
(A1) [8,11]. Decomposition of 1, under experimen-
were used as received. Buffer solutions were prepared
using analytical grade salts (EM Science) and water
purified in a Millipore apparatus.
tal conditions analogous to those of this study, has
been found to follow an A1 mechanism (Scheme 1),
as rates increase with decreasing pH and are indepen-
dent of the buffer concentration [12]. The expression
Instrumentation and Procedures
H2
N
H
H+
Decomposition of target diphenyltriazenes was moni-
tored photometrically using a Varian Cary 1 Bio spec-
trophotometer with a thermostated cell compartment
connected to a circulating water bath. Reactions were
initiated by adding 41 μL of an appropriate stock tri-
azene solution in methanol (unless noted otherwise)
to 2 mL of an aqueous buffer solution containing all
the other constituents; resulting aqueous solutions thus
contained 2% (v/v) MeOH and substrate concentra-
N
N
N
Ar
N
T
Ar
Ar
N
Ar
K
a
TH+
kN2-N3
Ar N2 + H N Ar
2
Scheme 1
International Journal of Chemical Kinetics DOI 10.1002/kin

INHIBITION OF ACID-INDUCED DECOMPOSITION OF DIPHENYLTRIAZENES
569
6
4
2
0
0.3
0.2
0.1
0
would render the N3-protonated forms of 1 and 2 more
+
BH
acidic than the conjugate forms of aniline (pK
=
a
+
BH
4
.63 [16a]) and p-anisidine (pKa
= 5.34 [16b]),
respectively.
The linear plots shown in Fig. 2 (circles) yield
3
−1 −1
s and
kN2−N3/Ka values of (1.33 ± 0.03) × 10 M
4
−1 −1
s for 1 and 2, respectively.
(
5.3 ± 0.1) × 10 M
k
The second-order rate coefficient kN2−N3/Ka clearly in-
creases with increasing electron-donating (ED) ability
of the phenyl ring, in agreement with previous reports
on substituent effects on acid-catalyzed decomposi-
tion of aryltriazenes [8,11a,17]. The trend is attributed
to an increase in N3 basicity and in stabilization of
the developing aryldiazonium ion with increasingly
stronger ED groups. Figure 2 also displays observed
rate constant values (triangles) for decomposition of 1
in solutions containing 11 mM BCD. The correspond-
0
1.1
×
10
2.2 × 10
+
[
H ] , M
Figure 2 Proton dependence of the observed rate constant
for acid-promoted decomposition of 1 ( , aqueous solu-
tion; ꢀ, 11 mM BCD aqueous solution) and 2 (
◦
•
, aqueous
solution).
−
1
−1
ing slope, namely (196 ± 8) M
s , is significantly
smaller than that determined for 1 in the absence of
CD, indicating a clear retardation effect on the rate of
acid-promoted N2 N3 heterolytic bond cleavage.
To assess the dependence of this inhibitory effect
on CD (cavity) size and rim substitution, for both sub-
strates and for each of the CDs employed in this study,
kinetic traces were recorded using a series of solu-
tions at constant pH but varying host concentration. In
for the observed rate constant (kobs) corresponding to
Scheme 1 is given by Eq. (1), where Ka and kN2−N3 rep-
resent, respectively, the acid dissociation equilibrium
constant (expressed in terms of concentration) for the
N3-protonated intermediate (TH ) and the first-order
rate constant for heterolytic cleavage of the N2 N3
bond.
+
CD
+
all cases, corresponding kobs values (i.e., observed rate
kN2−N3[H ]
Ka + [H ]
kobs =
(1)
constants in the presence of CDs) are found to decrease
nonlinearly with the CD concentration (Figs. 3–5). The
lack of any effect in the presence of glucose (10 mM)
and of only a slight retardation (if at all significant)
upon addition of soluble starch (in an amount equiv-
alent to 2 mM BCD) of the rate of decomposition
of 1 in aqueous phosphate buffer solutions (pH 6.16)
+
For both substrates 1 and 2, decay traces were collected
using a series of solutions at different pH, and in all
cases traces follow first-order kinetics. Resulting kobs
vs. [H ] plots are linear (Fig. 2, circles), indicating
∗
+
that, under the experimental conditions of this study,
+
Ka ꢁ [H ] in the denominator of Eq. (1). Instability
of triazenes in acidic media precludes measurement of
their basicity; however, theoretical calculations have
systematically shown that protonation of N3, in spite
of being thermodynamically less favorable, is able to
compete with that of the (more basic) N1 position [13].
6
5
4
3
2
1
0
0.14
0.12
5
4
3
2
1
0
0
0
0
0
0
0
.1
+
k
That the pKa of TH (for 1 and 2) is much lower than
.08
.06
.04
.02
the lowest pH used in this study (i.e., pKa ꢂ 6) is fully
consistent with the pKa value of 4.95 reported for acid-
promoted decomposition of triazene (HNNNH2) [14],
since resonance effects are expected to decrease the
electron density, and hence basicity, of the triazeno
moiety in diphenyltriazenes relative to that of triazene.
It is also in agreement with the assumption that the
electron-withdrawing character of the N NAr moi-
ety (e.g., σm = 0.32 and σp = 0.39 for N NPh [15])
0
0.005 0.01 0.015
[BCD], M
k
0
0.005
0.01
0.015
[
BCD], M
Figure 3 Influence of β-cyclodextrin concentration on the
observed rate constant for acid-catalyzed decomposition of
∗
1 (
◦
) and 2 ( ) in 2% MeOH aqueous phosphate buffer (pH
Experimentally observed rate constants are available as Sup-
•
porting Information.
6.16). Inset: reciprocal plot of data for 2.
International Journal of Chemical Kinetics DOI 10.1002/kin

5
70
XU, ASADI, AND BARRA
T
5
0.14
.12
K
CD
T + CD
T.CD
2
2
50
00
1
0
8
6
4
2
0
0
4
3
2
1
0
H+
CD
H+
K
K
a
a
150 0.1
TH+
+
1
5
00
0
TH .CD
k
0.08
0.06
0.04
CD
N2-N3
kN2-N3
k
0
Products
0
0.005 0.01 0.015
k
[
PMBCD], M
Scheme 2
0
0
.02
the observed rate constant corresponding to Scheme 2
would be given by Eq. (2), where KT represents
CD
0
0.005
0.01
0.015
the association equilibrium constant (concentrations
quotient) for cyclodextrin–neutral substrate complex
[
HPBCD], M
Figure 4 Influence of 2-hydroxypropyl-β-cyclodextrin and
permethyl-β-cyclodextrin (inset) concentration on the ob-
served rate constant for acid-catalyzed decomposition of
CD
CD
(
T.CD), whereas Ka and k
represent, respec-
N2−N3
tively, the acid dissociation equilibrium constant and
the first-order rate constant for heterolytic cleavage of
1
(
(
◦
) and 2 (
•
) in 2% MeOH aqueous phosphate buffer
+
the complexed N3-protonated intermediate (TH .CD).
pH 6.16).
CD
obs
k
ꢀ
ꢁ
ꢂ
ꢃ
ꢄ
CD
CD
T
CD
+
clearly indicates that the kinetic effects observed in
the presence of CDs are due to the formation of inclu-
sion complexes and not just to unspecific interactions;
thus, and assuming the conventional 1:1 host–guest
stoichiometry, acid-catalyzed decomposition of target
triazenes in the presence of CDs can be represented
as shown in Scheme 2. Bearing in mind that under
the experimental conditions of this study, target sub-
strates are essentially in their neutral form (i.e., mo-
lar fraction of neutral substrate ≈1, as inferred from
the linear plots shown in Fig. 2), the expression for
(kN2−N3/Ka) + k
K
K
[CD] [H ]
N2−N3
a
=
T
1
+ K [CD]
CD
(2)
For BCD as host, nonlinear curve fittings accord-
ing to Eq. (2) render the first-order rate coefficients
and association equilibrium constant values listed in
Table I. For HPBCD and PMBCD as hosts, how-
ever, fittings according to Eq. (2) reveal that the term
CD
N2−N3
CD
T
(
k
/Ka )K [CD] is actually negligible in com-
CD
CD
obs
parison to kN2−N3/Ka, in agreement with linear 1/k
vs. [CD] plots (e.g., Fig. 4 inset). Thus, nonlinear curve
fittings according to Eq. (3) ultimately render asso-
ciation equilibrium constant values for HPBCD and
PMBCD (Table I).
6
5
4
3
2
1
0
0
.14
2
2
1
1
5
0
5
0
5
0
0.12
+
(kN2−N3/Ka)[H ]
CD
0
0
0
0
0
0
.1
k
=
(3)
obs
T
1 + K [CD]
CD
k
.08
.06
.04
.02
For ACD as host, data cannot be explained on the
basis of the mechanism proposed in Scheme 2 and
0
0.008
ACD], M
0.016
k
CD
[
Eq. (2). As illustrated in Fig. 5 (inset), plots of 1/kobs
vs. [CD] curve upward, particularly in the case of data
for decomposition of 2. This nonlinear dependence is
consistent with a mechanism involving formation of
host–guest complexes with 2:1 stoichiometry
0
0.004
0.008
0.012
(Scheme 3), a common observation when using
[
ACD], M
Figure 5 Influence of α-cyclodextrin concentration on the
observed rate constant for acid-catalyzed decomposition of
T
TCD
KCD
K
CD
1
(
◦
) and 2 (
•
) in 2% MeOH aqueous phosphate buffer (pH
T + CD
T.CD
+
CD
T.CD
2
6.16). Inset: reciprocal plot of data for decomposition of 2 in
phosphate buffer (pH 6.16) with different organic cosolvents
, 2% MeOH; ꢀ, 2% MeCN).
(
•
Scheme 3
International Journal of Chemical Kinetics DOI 10.1002/kin

INHIBITION OF ACID-INDUCED DECOMPOSITION OF DIPHENYLTRIAZENES
571
Table I First-Order Rate Coefficients and Association Equilibrium Constants Obtained for Decomposition of 1 and 2
a
in the Presence of CDs
1
2
+
T
TCD
+
T
TCD
kN2−N3[H ]/Ka
K
K
kN2−N3[H ]/Ka
K
K
CD CD
CD CD
−
3
−1
T
CD
(102
−1
5
−2
−3 −1
T
CD
(102
−1
5
−2
M )
CD
(10
s
)
K
M
)
(10
M
)
(10
s
)
K
M
)
(10
ACD
ACD
BCD
1.26 ± 0.01
7.4 ± 0.2
9.1 ± 0.4
0.30 ± 0.05
51.2 ± 0.8
36.3 ± 0.1
50.5 ± 0.5
(1.1 ± 0.2)
49.1 ± 0.4
46.5 ± 0.9
14 ± 2
7 ± 2
b
4.0 ± 0.8
29 ± 1
1.4 ± 0.4
1.26 ± 0.01
0.077 ± 0.009)
1.26 ± 0.01
1.28 ± 0.01
c
c
(
HPBCD
PMBCD
32.1 ± 0.8
10.9 ± 0.4
77 ± 2
22 ± 1
a
◦
Solvent contains 2% (v/v) methanol (unless stated otherwise); T = 21 C; μ = 0.5 M (NaCl); 0.05 M phosphate buffer (pH 6.16). Errors
given correspond to the standard deviations.
b
Solvent contains 2% (v/v) acetonitrile.
Value for k
c
CD
N2−N3
+
CD
[H ]/Ka .
ACD as host. Based on the two-equilibria model
depicted in Scheme 3, the expression corresponding
to the observed rate constant for acid-promoted
the slope value calculated from Fig. 2 (triangles) for
the BCD series.
The fact that, for any given target substrate and re-
gardless of the type of CD employed, decomposition
of complexed triazenes is much slower than that of the
free (uncomplexed) substrates, can be rationalized in
terms of the dependence of the reaction on the polarity
of the reaction medium. Second-order rate coefficients
for acid-catalyzed N2 N3 bond cleavage of 1 and 2
decomposition of target triazenes in the presence of
CD2
ACD would be given by Eq. (4), where K
and
a
CD2
k
represent, respectively, the acid dissociation
equilibrium constant and the first-order rate constant
for heterolytic cleavage of 2:1 cyclodextrin–TH
N2−N3
+
complexes.
ꢀ
ꢁ
ꢂ
ꢃ
ꢁ
ꢂ
ꢃ
[CD]2ꢄ
[H ]
CD CD
CD
N2−N3
CD
T
CD2
CD2
T
TCD
+
(
kN2−N3/Ka) + k
K
K
[CD] + k
K
K
K
CD
obs
a
CD
N2−N3
a
k
=
(4)
T
CD
T
TCD
2
1
+ K [CD] + K
K
[CD]
CD CD
CD
/K aC D)K [CD] +
T
Curve fittings reveal that {(k
in aqueous buffer solutions containing 30% (v/v) THF
N2−N3
CD
CD2
N2−N3
CD2
T
TCD
2
−1 −1
and
(
k
/K )K K
CD CD
[CD] } ꢂ kN2−N3/Ka in the
have been reported as being (21.3 ± 0.7) M
s
a
− −1 †
1
numerator of Eq. (4), in agreement with linear
(976 ± 9) M
s , respectively [8], i.e., significantly
CD
obs
(
1/k − 1/kobs)/[CD] vs. [CD] plots (not shown).
smaller than the values reported here for decomposition
in 2% (v/v) MeOH aqueous buffer solutions. Proto-
nation constants for anilines in dioxane–water solvent
mixtures have been shown to (systematically) decrease
as the concentration of the aprotic nonpolar organic
co-solvent increases [18]. On the other hand, and based
on the Hughes and Ingold theory [19], kN2−N3 values
would be expected to increase (albeit only slightly) as
the concentration of the organic cosolvent increases
since the transition state for heterolytic cleavage of
Thus, Eq. (4) reduces to Eq. (5), from which nonlinear
fittings yield association equilibrium constants for 1:1
and 2:1 ACD complexes (Table I).
+
(
kN2−N3/Ka)[H ]
+ K [CD] + K
CD
obs
k
=
(5)
T
T
TCD
[CD]2
1
K
CD
CD CD
In all cases, calculated first-order rate coefficients cor-
responding to reaction in the bulk solution lead to
kN2−N3/Ka values that agree very well with those de-
termined from the pH-dependence series in the absence
of CD. Furthermore, from the values for the three fit-
ted parameters given in Table I for decomposition of 1
in the presence of BCD, one obtains {(kN2−N3/Ka) +
+
TH involves charge dispersion. Thus, the (noticeable)
decrease in reactivity on going from 2% MeOH to 30%
THF would indicate that the rate of decomposition of
†
−pH
Values were obtained from kobs vs. 10
plots and correspond
CD
CD
a
T
CD
T
−1
(
s
k
/K )K [CD]}/(1 + K [CD]) ≈ 195 M
N2−N3
CD
indeed to kN2−N3/(γH+ ), where γH+ represents the proton activity
K
a
−1
(at [CD] = 11 mM), in excellent agreement with
coefficient.
International Journal of Chemical Kinetics DOI 10.1002/kin

5
72
XU, ASADI, AND BARRA
target triazenes is dominated by their basicity. Inclu-
sion in the cavity of CDs certainly renders a relatively
nonpolar environment, at least in comparison with wa-
ter, which leads to a decrease in triazene basicity and,
ultimately, in the rate of decomposition. The fact that
BCD is the only host, among those used in this study,
ACD
4
2
0
HPBCD
BCD
+
for which reactivity of the cyclodextrin–TH complex
is detected suggests guest triazenes experience lesser
environmental changes upon inclusion in BCD than
upon inclusion in the narrower ACD (ultimately with
two host units) or in any of the BCD derivatives (in
which alkyl substituents elongate the cavity and in-
crease the hydrophobicity of its edges [20,21]).
-
2
PMBCD
0.4
0.2
0
As indicated in Table I, KT values derived from
CD
the dependence of the observed first-order rate con-
stants on CD concentration increase in the order ACD
300
350
Wavelength, nm
400
<
BCD ∼ PMBCD < HPBCD. This trend is consis-
tent with existing reports in the literature that show
an increase in binding constants for aromatic guest
species with increasing cavity size (ACD vs. BCD)
as well as with increasing rim substitution (BCD vs.
HPBCD and PMBCD) [22]. It is important to point
out here that both target substrates 1 and 2 have two
binding sites: N1-phenyl and N3-phenyl end groups;
Figure 6 Induced circular dichroism spectra in the presence
of 7 mM CD (top) and absorption spectrum (bottom) of 1 in
% MeOH aqueous borax buffer (pH 9.94).
2
the wavelength shift (approximately 3–5 nm) and vari-
ations in signal intensity are rather small for native
and substituted BCDs (spectra not shown), spectral
changes are certainly noticeable in the case of ACD.
Upon addition of increasing amounts of ACD, besides
a bathochromic shift, a decrease in signal intensity is
observed when using solutions of 1 (Fig. 7), whereas an
increase in absorption ultimately results in the case of
(Fig. 8). Interestingly, no isosbestic point remains
throughout the ACD concentration range employed
([ACD] ≤ 21 mM) in either case. Moreover, fittings
hence, isomeric 1:1 complexes can form in each case.
T
CD
The K
values kinetically determined would cor-
respond to the complexes that more strongly influ-
ence the rate of decomposition. Interestingly, equilib-
rium constants for complexation of 2 are ca. twice
as large as those for 1, and values corresponding to
BCD complexation are comparable to literature values
for inclusion of azobenzene derivatives (X-Ph-N=N-
2
+
−
−1
Ph-N(CH3) Cl ), namely 860 and 1300 M , respec-
2
tively, for X = H and CH3 [21].
Further evidence of inclusion complex formation
was obtained by means of induced circular dichroism
0
0
.7
.6
(ICD) and UV–visible absorption spectroscopy, using
borax buffer solutions (pH 9.98) to minimize substrate
decomposition while acquiring corresponding spectra.
For instance, ICD spectra corresponding to solutions
of 1 in the presence of ACD, BCD, or HPBCD show
a single peak with positive sign in the 280–430 nm
region (Fig. 6), while that of a solution of 1 in the
presence of PMBCD displays a split-type pattern. In-
terestingly, these spectral characteristics are compara-
ble to those reported for CD complexes of azobenzene
derivatives; recent reports, for example, show a posi-
tive peak in the ICD spectra of 4-hydroxyazobenzene
and 4-aminoazobenzene upon inclusion in BCD [23],
and a splitting pattern in the case of complexation with
PMBCD [24]. On the other hand, UV–visible absorp-
tion spectroscopy reveals that addition of CDs to solu-
tions of 1 or 2 generally leads to a bathochromic (red)
shift of the longest wavelength absorption band. While
a
0.5
b
3
40 360
0
0
.4
.3
,
nm
0.2
0
.1
0
250 300 350 400 450 500
Wavelength, nm
Figure 7 Absorption spectra of 1 in 2% MeOH aqueous
borax buffer (pH 9.94) as a function of ACD concentration
(0, 0.5, 1, 2, 4, 5, 10, 13.5, and 21 mM, from a to b). Inset:
cutoff normalized spectra.
International Journal of Chemical Kinetics DOI 10.1002/kin

INHIBITION OF ACID-INDUCED DECOMPOSITION OF DIPHENYLTRIAZENES
573
0
0
0
0
0
0
.6
.5
.4
.3
.2
.1
icantly affected by the presence of (aprotic) organic
cosolvents; competition for the CD cavity between
guest and cosolvent may lead to a (significant) de-
crease in binding constant [21,26]. Methanol, among
organic solvents, is the closest to water in structure
and properties; thus, a decrease in binding affinity on
going from 2% MeOH to 2% MeCN would not be un-
expected. To verify this reasoning, observed first-order
rate constants for decomposition of 2 were determined
in 2% MeCN aqueous phosphate buffer solutions and
from the dependence on ACD concentration (Fig. 5
inset; Eq. (5)) lower association equilibrium constants
in comparison with values for 2% MeOH in fact result
(Table I, second row). Also, the calculated first-order
rate coefficient corresponding to decomposition of
b
a
0
2
50 300 350 400 450 500
Wavelength , nm
Figure 8 Absorption spectra of 2 in 2% MeCN aqueous
borax buffer (pH 9.94) as a function of ACD concentration
2
in 2% MeCN aqueous solution is lower than that
in 2% MeOH (but higher than in 30% THF), in excel-
lent agreement with the dependence of the reaction on
solvent polarity presented in a preceding paragraph.
In summary, acid-promoted decomposition of 1,3-
diphenyltriazenes in aqueous solutions is strongly in-
hibited upon inclusion by cyclodextrins. For any given
substrate, the inhibitory effect, attributed to the de-
crease in substrate basicity resulting from a decrease
in the polarity of the reaction medium, depends both
on the cavity size and the rim substituents of the host
molecule. Data reported herein should prove valuable
to applications requiring modulation of (aromatic) tri-
azenes stability under acidic conditions.
(
0, 1, 2, 4, 5, 10, 13.5, and 21 mM, from a to b).
according to Eq. (6) of the change in absorbance (ꢀA)
resulting from addition of ACD,
T
ꢀ
εK [CD][T]0
CD
ꢀ
A =
(6)
T
1
+ K [CD]
CD
where ꢀε and [T]0 represent the difference in extinc-
tion coefficients between a 1:1 host–guest complex and
free substrate and the total concentration of substrate,
respectively, render KT values that are wavelength
CD
dependent. This result, together with the lack of isos-
bestic points, is a very clear indication of formation of
higher complexes [25], consistent with the kinetic anal-
ysis presented above based on the involvement of 1:1
and 2:1 (host–guest) ACD complexes. Unfortunately,
the small variations in signal intensity (ꢀA < 0.1)
and limited number of ꢀA values obtained precluded
a statistically significant fitting to a function involv-
ing ꢀε and KCD terms for 1:1 and 2:1 (host–guest)
complexes (i.e., with four adjustable parameters) to
determine equilibrium constants.
Absorption spectra displayed in Fig. 8 correspond
indeed to aqueous solutions of 2 containing 2% MeCN
instead of 2% MeOH, since appreciable decomposi-
tion of 2 stock MeOH solution would typically take
place within the time period needed for completion
of spectral titrations. Interestingly, during the course
of these experiments it was noticed that for any given
ACD concentration, spectral changes recorded for 2%
MeCN aqueous solutions of 2 were less significant
than those corresponding to 2% MeOH aqueous so-
lutions, which suggests lower association equilibrium
constants in the former solvent system. It is well es-
tablished that binding properties of CDs can be signif-
Special thanks are due to Prof. G. Guillemette for the use of
the spectropolarimeter.
BIBLIOGRAPHY
1
. Kimball, D. B.; Haley, M. M. Angew Chem, Int Ed 2002,
1, 3338–3351.
4
2
. For representative articles see (a) Marchesi, F.;
Turriziani, M.; Tortorelli, G.; Avvisati, G.; Torino, F.;
De Vecchis, L. Pharmacol Res 2007, 56, 275–287; (b)
Carvalho, E.; Francisco, A. P.; Iley, J.; Rosa, E. Bioorg
Med Chem 2000, 8, 1719–1725; (c) Rouzer, C. A.;
Sabourin, M.; Skinner, T. L.; Thompson, E. J.; Wood,
T. O.; Chmurny, G. N.; Klose, J. R.; Roman, J. M.; Smith,
R. H., Jr.; Michejda, C. J. Chem Res Toxicol 1996, 9,
172–178; (d) Wilman, D. E. V. Cancer Treat Rev 1988,
15, 69–72; (e) Kolar, G. F. In Chemical Carcinogens,
2nd ed.; Searle, C. E., (Ed.); ACS Monograph 182;
American Chemical Society: Washington, DC, 1984;
Vol. 2, Ch. 14.
3. Tonelli, M.; Vazzana, I.; Tasso, B.; Boido, V.; Sparatore,
F.; Fermeglia, M.; Paneni, M. S.; Posocco, P.; Pricl, S.;
International Journal of Chemical Kinetics DOI 10.1002/kin

5
74
XU, ASADI, AND BARRA
La Colla, P.; Ibba, C.; Secci, B.; Collu, G.; Loddo, R.
Bioorg Med Chem 2009, 17, 4425–4440.
R. H., Jr.; Michejda, C. J. J Org Chem 1988, 53, 3433–
3436; (c) Nguyen, M-T.; Hoesch, L. Helv Chim Acta
1986, 69, 1627–1637.
4
. Campbell, M.; Prankerd, R. J.; Davie, A. S.; Charman,
W. N. J Pharm Pharmacol 2004, 56, 1327–1332.
. Dodziuk, H. (Ed.). Cyclodextrins and Their Complexes:
Chemistry, Analytical Methods, Applications; Wiley-
VCH: Weinheim, Germany, 2006.
. For recent review articles see (a) Stella, V. J.; He, Q.
Toxicol Pathol 2008, 36, 30–42; (b) Vyas, A.; Saraf, S.;
Saraf, S. J Inclusion Phenom Macrocyclic Chem 2008,
14. Sutherland, J. W. J Phys Chem 1979, 83, 789–795.
15. Hansch, C.; Leo, A.; Taft, R. W. Chem Rev 1991, 91,
165–195.
16. Lide, D. R., (Ed.). CRC Handbook of Chemistry and
Physics; CRC Press: Boca Raton, FL, 1996; (a) pp. 8–
49, (b) pp. 8–52.
5
6
17. Svodoba, P.; Pytela, O.; Ve cˇ e rˇ a, M. Collect Czech Chem
Commun 1986, 51, 553–563.
6
2, 23–42; (c) Cal, K.; Centkowska, K. Eur J Pharm Bio-
pharm 2008, 68, 467–478; (d) Brewster, M. E.; Loftsson,
T. Adv Drug Delivery Rev 2007, 59, 645–666; (e)
Oshovsky, G. V.; Reinhoudt, D. N.; Verboom, W. Angew
Chem, Int Ed 2007, 46, 2366–2393; (f) Villalonga, R.;
Cao, R.; Fragoso, A. Chem Rev 2007, 107, 3088–3116.
. (a) Roehrl, M. H. A.; Wang, J. Y.; Wagner, G. Biochem
18. Demirelli, H.; K o¨ seo g˘ lu, F. J Sol Chem 2004, 33, 1501–
1515.
19. Reichardt, C. Solvents and Solvent Effects in Organic
Chemistry; Wiley-VCH: Weinheim, Germany, 2003;
Ch. 5.
7
20. Immel, S.; Lichtenthaler, F. W. Starch/St a¨ rke 1996, 48,
225–232.
2
004, 43, 16067–16075; (b) Nifontov, V. I.; Bel s´ kaya,
N. P.; Krupnova, L. V.; Ershova, Y. A.; Fedorova, A. P.
Pharm Chem J 1994, 28, 611–615.
. Chen, N.; Barra, M.; Lee, I.; Chahal, N. J Org Chem
21. Sueishi, Y.; Kasahara, M.; Inoue, M.; Matsueda, K. J
Inclusion Phenom Macrocyclic Chem 2003, 46, 71–75.
22. (a) Connors, K. A. Chem Rev 1997, 97, 1325–1357; (b)
Szejtli, J. Pure Appl Chem 2004, 76, 1825–1845.
23. Liu, Y.; Zhao, Y-L.; Chen, Y.; Guo, D-S. Org Biomol
Chem 2005, 3, 584–591.
24. Shi, J., Guo, D-S., Ding, F. Liu, Y. Eur J Org Chem 2009,
923–931.
25. Connors, K. A. Binding Constants: The Measurement of
Molecular Complex Stability; Wiley: New York, 1987;
Ch. 4.
26. The detailed coverage of solvent effects on binding con-
stants is beyond the scope of this article; only a few
representative publications are listed here: (a) Lo Meo,
P.; D’Anna, F.; Riela, S.; Gruttadauria, M.; Noto, R.
Tetrahedron 2009, 65, 2037–2042; (b) Garc ´ı a-Rio, L.;
Herv e´ s, P.; Leis, J. R.; Mejuto, J. C.; P e´ rez-Juste, J.;
Rodr ´ı guez-Dafonte, P. Org Biomol Chem 2006, 4, 1038–
1048; (c) Porras, S. P.; Sarmini, K.; Fanali, S.; Kenndler,
E. Anal Chem 2003, 75, 1645–1651; (d) Partyka, M.;
Au, B. H.; Evans, C. H. J Photochem Photobiol A:
Chem 2001, 140, 67–74; (e) Connors, K. A.; Mulaki,
M. J.; Paulson, A. J Org Chem 1992, 57, 1794–1798.
8
2002, 67, 2271–2277.
9
. Harned, H. S.; Owen, B. B. The Physical Chemistry
of Electrolytic Solutions; Reinhold: New York, 1958;
p. 748.
0. (a) Hanusek, J.; B eˇ lohlavov a´ , H.; P ˇr ikryl, J.; Mach a´ cˇ ek,
V. Dyes Pigm 2009, 80, 136–140; (b) Ludwig, M.;
Bedn a´ ˇr ov a´ , I; Pa ˇr ´ı k, P. Collect Czech Chem Com-
mun 1999, 64, 1654–1672; (c) Pytela, O.; Nev e˘ cˇ n a´ , T.;
Kav a´ lek, J. Collect Czech Chem Commun 1990, 55,
1
2
701–2706; (d) Svodoba, P.; Pytela, O.; Ve cˇ e ˇr a, M. Col-
lect Czech Chem Commun 1987, 52, 2217–2224.
1. (a) Bene sˇ , J.; Ber a´ nek, V.; Zimprich, J.; Vete sˇ n ´ı k, P.
Collect Czech Chem Commun 1977, 42, 702–710; (b)
Zv eˇ ˇr ina, V.; Reme sˇ , M.; Divi sˇ , J.; Marhold, J.; Matrka,
M. Collect Czech Chem Commun 1973, 38, 251–256.
2. Chen, N. M. Sc. thesis, University of Waterloo, Water-
loo, Canada, 2000.
1
1
1
3. (a) Rakotondradany, F.; Williams, C. I.; Whitehead,
M. A.; Jean-Claude, B. J. J Molec Struct (Theochem)
2001, 535, 217–234; (b) Schmiedekamp, A.; Smith,
International Journal of Chemical Kinetics DOI 10.1002/kin