Reactivity of the insecticide chlorpyrifos-methyl toward hydroxy! and perhydroxyl ion. Effect of cyclodextrins

Article abstract of DOI:10.1002/poc.1502

The reactivity of Chlorpyrifos-Methyl (1) toward hydroxyl ion and the α-nucleophile, perhydroxyl ion was investigated in aqueous basic media. The hydrolysis of 1 was studied at 25° C in water containing 10% ACN or 7% 1,4-dioxane at NaOH concentrations between 0.01 and 0.6m; the second-order rate constant is 1.88 × 10^-2m^-1 s^-1 in 10% ACN and 1.70 × 10^-2m^-1 s^-1 in 7% 1,4-dioxane. The reaction with H₂O₂ was studied in a pH range from 9.14 to 12.40 in 7% 1,4-dioxane/H₂O; the second-order rate constant for the reaction of HOO ion is 7.9 m^-1 s^-1 whereas neutral H₂O₂ does not compete as nucleophile. In all cases quantitative formation of 3,5,6-trichloro-2-pyridinol (3) was observed indicating an S_N2(P) pathway. The hydrolysis reaction is inhibited by α-, β-, and γ-cyclodextrin showing saturation kinetics; the greater inhibition is produced by γ-cyclodextrin. The reaction with hydrogen peroxide is weakly inhibited by α- and β-cyclodextrin (β-CD), whereas g-cyclodextrin produces a greater inhibition and saturation kinetics. The kinetic data obtained in the presence of β-or γ-cyclodextrin for the reaction with hydroxyl or perhydroxyl ion indicate that the main reaction pathway for the cyclodextrin-mediated reaction is the reaction of HO^- or HOO^- ion with the substrate complexed with the anion of the cyclodextrin. The inhibition is attributed to the inclusion of the substrate with the reaction center far from the ionized secondary OH groups of the cyclodextrin and protected from external attack of the nucleophile. Sucrose also inhibits the hydrolysis reaction but the effect is independent of its concentration. Copyright

Full text of DOI:10.1002/poc.1502

Research Article
Received: 20 August 2008,
Revised: 7 November 2008,
Accepted: 14 November 2008,
Published online in Wiley InterScience: 29 December 2008
(www.interscience.wiley.com) DOI 10.1002/poc.1502
Reactivity of the insecticide
chlorpyrifos-methyl toward hydroxyl and
perhydroxyl ion. Effect of cyclodextrins
a
a
a
*
´
Raquel V. Vico , Rita H. de Rossi and Elba I. Bujan
The reactivity of Chlorpyrifos-Methyl (1) toward hydroxyl ion and the a-nucleophile, perhydroxyl ion was investigated
in aqueous basic media. The hydrolysis of 1 was studied at 25 8C in water containing 10% ACN or 7% 1,4-dioxane at
NaOH concentrations between 0.01 and 0.6 M; the second-order rate constant is 1.88 ꢀ 10ꢁ2
M
ꢁ1 sꢁ1 in 10% ACN and
1.70 ꢀ 10ꢁ2
M
ꢁ1 sꢁ1 in 7% 1,4-dioxane. The reaction with H2O2 was studied in a pH range from 9.14 to 12.40 in 7%
1,4-dioxane/H2O; the second-order rate constant for the reaction of HOOꢁ ion is 7.9 Mꢁ1 sꢁ1 whereas neutral H2O2 does
not compete as nucleophile. In all cases quantitative formation of 3,5,6-trichloro-2-pyridinol (3) was observed
indicating an SN2(P) pathway. The hydrolysis reaction is inhibited by a-, b-, and g-cyclodextrin showing saturation
kinetics; the greater inhibition is produced by g-cyclodextrin. The reaction with hydrogen peroxide is weakly inhibited
by a- and b-cyclodextrin (b-CD), whereas g-cyclodextrin produces a greater inhibition and saturation kinetics. The
kinetic data obtained in the presence of b- or g-cyclodextrin for the reaction with hydroxyl or perhydroxyl ion indicate
that the main reaction pathway for the cyclodextrin-mediated reaction is the reaction of HOꢁ or HOOꢁ ion with
the substrate complexed with the anion of the cyclodextrin. The inhibition is attributed to the inclusion of the
substrate with the reaction center far from the ionized secondary OH groups of the cyclodextrin and protected from
external attack of the nucleophile. Sucrose also inhibits the hydrolysis reaction but the effect is independent of its
concentration. Copyright ß 2008 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: organophosphorus insecticides; cyclodextrins; host–guest systems; hydrolysis; hydrogen peroxide; reaction
mechanisms; inclusion complexes
registered for use in the control of insect pests on certain stored
INTRODUCTION
grain, including wheat, barley, oats, rice, and sorghum as well as
for empty grain bins.[12] Annual usage of Chlorpyrifos-Methyl is an
estimated 80 000 pounds active ingredient for approximately
267 497 000 bushels of grain.[12] Cyclodextrins are cyclic
oligomers of a-D-glucose which have a well defined cavity
surrounded by the secondary OH in the wider rim and by primary
OH in the smaller rim.[13–15] They are soluble in water and the
cavity is relatively non-polar compared with the solvent, so it
provides a special microenvironment for organic reactions.
Cyclodextrins have attracted interest in several different areas,
e.g., they have been used as models for enzyme-catalysed
reactions[16–19] and also for many industrial applications.[20]
Organophosphorus compounds are an economically important
class of chemical compounds with numerous uses, such as
pesticides, industrial fluids, flame retardants, therapeutics, and
nerve agents. Many of the synthetic organophosphorus
compounds are tailor-made to inhibit acetylcholinesterase
(AChE), an enzyme essential for life in humans and other animal
species.[1,2] Hence the chemistry of organophosphorous com-
pounds is important to a wide range of interests. Available
methods for destroying organophosphorous compounds include
thermic, enzymatic, or photochemical degradation as well as
reactions with strong oxidants or nucleophiles.[1–6] Detailed
knowledge of the kinetics of the alkaline and neutral hydrolysis
pathways is critical to several areas of environmental chemistry;[7]
they are particularly important to understand their persistence
and fate in natural environments and to find ways to destroy
them rapidly and safely. Hydrogen peroxide has been used for
decontamination of chemical warfare agents,[1,2] and in advanced
oxidation processes for environmental remediation, such as
Fenton and photo Fenton.[8–11] This oxidant has the advantage
over others that is environment friendly leading in some cases to
the total mineralization of the organic pollutant.[8]
´
Correspondence to: E. I. Bujan, Instituto de Investigaciones en Fisicoqu´ımica de
*
´
Cordoba (INFIQC), Dpto. de Qu´ımica Organica, Facultad de Ciencias Qu´ımicas,
Universidad Nacional de Cordoba, Medina Allende y Haya de la Torre,
´
´
´
X5000HUA, Cordoba, Argentina.
E-mail: elba@fcq.unc.edu.ar
´
R. V. Vico, R. H. de Rossi, E. I. Bujan
a
´
Dpto. de Qu´ımica Organica, Instituto de Investigaciones en Fisicoqu´ımica de
Cordoba (INFIQC), Facultad de Ciencias Qu´ımicas, Universidad Nacional de
Chlorpyrifos-Methyl (O,O-dimethyl O-(3,5,6-trichloro-2-pyridynil)-
phosphorothioate) (1) is a cholinesterase inhibiting pesticide
´
´
Cordoba, Medina Allende y Haya de la Torre, X5000HUA, Cordoba, Argentina
´
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
Our results were confirmed by 31P NMR. The degradation of a
1 ꢀ 10ꢁ3 M solution of 1 in 1,4-dioxane/D2O/H2O (20:40:40) in
the presence of 0.6 M NaOH was followed by 31P NMR over a
period of 6 h. The initial reading showed a single peak
at 64.42 ppm, corresponding to 1. After 3 h of reaction a
single signal was observed at 58.23 ppm corresponding to
O,O-dimethylphosporothioate (4),[28,29] this signal remain
unchanged over the period of time studied. (Fig. 1).
The kinetics of the hydrolysis of 1 was followed by measuring
the increase in absorbance at 320 nm, the lmax of 3. The observed
rate constants for this reaction are shown in Table S1. The plot of
k
obs versus HOꢁ concentration is linear (Fig. 2) and from the slope
the value of the second-order rate constant kOH ¼ (1.88 ꢂ
We are particularly interested in the effect of cyclodextrins on
the mechanism of hydrolysis of esters[21–23] and amides,[24] and
therefore we undertook a study of the reaction of phosphor-
othioates in basic aqueous solution in the presence of
cyclodextrins.[25] We have previously reported that the basic
hydrolysis of Fenitrothion (O,O-dimethyl O-(3-methyl-4-nitrophenyl)-
phosphorothioate) (2) is inhibited by complexation with
b-cyclodextrin (b-CD) and that the main reaction pathway for
the cyclodextrin-mediated reaction is the reaction of HOꢁ ion
with the substrate complexed with the anion of b-CD.[25] Since 1
is an insecticide of the same family as 2, we considered of interest
to study its hydrolysis, the reaction with hydrogen peroxide and
the effect of cyclodextrins on them. We found that both
compounds behave in a similar way although 1 is more reactive
and HOOꢁ is more than 400 times more reactive than HOꢁ
toward 1. The hydrolysis of 1 is inhibited in different magnitude
by complexation with b- and g-CD with saturation kinetics
whereas the reaction with hydrogen peroxide is only inhibited
with saturation kinetics by g-CD. The difference in the magnitude
of the inhibition observed with the two CDs is due to differences
in the binding abilities of the two hosts with the guest.
0.05) ꢀ 10ꢁ2
M
ꢁ1 sꢁ1 was calculated. The intercept of this plot is
equal to zero, within experimental error, indicating that H2O does
not compete with HOꢁ ion as nucleophile; this is in agreement
with the fact that t1/2 for the reaction of 5 in natural waters ranges
from 19 to 80 days,[7,26,30] and t1/2 reported for the hydrolysis of 1
at neutral pH ranges from 8 to 23 days.[31] The value of kOH
calculated for the hydrolysis of 1 is one order of magnitude
higher than that for 2, i.e., (2.0 ꢂ 0.1) ꢀ 10ꢁ3
M . The
ꢁ1 sꢁ1 [25]
difference in reactivity of the two insecticides, 1 and 2, can be
explained by the difference in pKa of their leaving groups. The pKa
of 3-methyl-4-nitrophenol, the leaving group of 2, is 7.26,[32] and
that of 3, determined by us, is 4.77 (Lit.[26] 4.55). Similar leaving
group effect was observed previously in the reaction of PhOꢁ
with a series of substituted dimethyl posphorothioates.[33]
Besides, it is a well documented fact that in many families of
esters, their reactivity increases with the decrease in the pKa of
the leaving group.[34–37]
Other factors that may contribute to the difference in reactivity
observed between 1 and 2 are the ionic strength and the organic
solvent content in the reaction medium. The hydrolysis of 2[25]
was studied at I ¼ 1 M and 2% ACN/H2O whereas for 1 it was
necessary to work at I ¼ 0.6 M and 10% ACN/H2O due to its lower
solubility in water. It was previously reported that 5 is less soluble
in sea waters than in distilled water and the rate of hydrolysis in
natural waters is highly dependent on the medium salinity.[26]
Williams and coworkers[38] informed that the second-order rate
constant for the alkaline hydrolysis of 4-nitrophenyl diphenyl
phosphate increases by a factor of 1.6 as the ionic strength of the
reaction medium decreases from 0.5 to 0.1 M. On the other hand,
the increase in polar aprotic solvent content produces an increase
in the reactivity of some phosphate and phosphorothioate
esters.[39,40]
RESULTS AND DISCUSSION
Hydrolysis reaction
The hydrolysis of 1 was studied under pseudo-first-order
conditions in 10% ACN/H2O at 25 8C at several NaOH
concentrations and at constant ionic strength (I ¼ 0.6 M). The
UV–Vis spectrum of the product matches that of
3,5,6-trichloro-2-pyridinol (3) at the expected concentration;
therefore, the only reaction taking place is P—O bond fission,
(Eqn (1)). Some authors have previously demonstrated for
Chlorpyrifos (5), that above pH 8, the only reaction that occurs is
elimination of the pyridinol.[7,26] Jans and coworkers[27] had
reported on the hydrolysis of 1 between pH 5.4 and 9.7 indicating
that the observed hydrolysis rate is independent of pH over the
range 5.4–8.6, and that at pH 9.0, formation of 3 accounts for 76%
of the substrate that reacted, desmethyl chlorpyrifos-methyl
being the other product formed by attack of H2O on the aliphatic
a-carbon.
100
50
0
-50
-100
ppm
Figure 1. 31P NMR spectrum in 1,4-dioxane/H2O/D2O (20:40:40) at room
temperature of 1 (0.080 M) (lower) and of 1 (1.1 ꢀ 10ꢁ3 M) with 0.62 M
NaOH after 6 h of reaction (upper)
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702

REACTIVITY OF CHLORPYRIFOS-METHYL
14.0
expression for kobs is given by Eqn (2). The plots of kobs versus
[H2O2]0 according to Eqn (2) at each pH are linear (Fig. 3) and from
the slope of these plots the values of the second-order rate
12.0
10.0
8.0
constants (k
0 ) at each pH, were calculated (Table S3). The
2
½H O ꢃ
2
intercepts of the plots of Fig. 3 are indistinguishable from zero,
within experimental error, indicating that H2O and HOꢁ ion do
not compete with H2O2. The value of k
and it can be seen that it increases with pH.
is given by Eqn (5)
½H O ꢃ
2
2
0
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
kobs ¼ kH O þ kHO ½HO ꢃ þ kH O ½H2O2ꢃ þ kHOO ½HOO ꢃ
2
2
2
ꢁ
ꢁ
ꢁ
¼ kH O þ kHO ½HO ꢃ þ ðkH O XH O þ kHOO XHOO Þ½H2O2ꢃ
2
2
2
2
2
0
6.0
(2)
½H2O2ꢃ
XH O
¼
(3)
(4)
4.0
2
2
½H2O2ꢃ0
½HOOꢁꢃ
½H2O2ꢃ0
ꢁ
XHOO
¼
2.0
ꢁ
ꢁ
k
¼ kH O XH O þ kHOO XHOO
½H2O2ꢃ0
2
2
2
2
0.0
0.00
0.20
0.40
0.60
0.80
ꢁ
ꢁ
¼ kH O þ ðkHOO ꢁ kH O ÞXHOO
(5)
2
2
2
2
[NaOH] (M)
From a plot of k
versus XHOOꢁ according to Eqn (5), the
½H O ꢃ
2
2
0
second-order rate constants for H2O2 and HOOꢁ can be
calculated from the intercept at XHOOꢁ ¼ 0 and 1, respectively
(Fig. 4). The intercept at XHOOꢁ ¼ 0 is zero, within experimental
error; therefore H2O2 does not compete as nucleophile with
HOOꢁ. The intercept at XHOOꢁ ¼ 1 gives kHOOꢁ ¼ 7.9 ꢂ
Figure 2. Plot of kobs versus [NaOH] for the hydrolysis of 1 at 25 8C. Solvent
contains 10% ACN; ionic strength I ¼ 0.6 M (NaCl); [1]0 ¼ 3.04 ꢀ 10ꢁ5
M
The hydrolysis of 1 was also studied in 7% 1,4-dioxane/H2O at
I ¼ 0.6 M (NaCl) at 25 8C with 0.101 M < [NaOH] < 0.600 M. As it was
observed in 10% ACN/H2O, 3 was formed quantitatively. From the
slope of the linear plot of kobs versus [NaOH] (Fig. S1), kOH was
0.4 Mꢁ1 sꢁ1
.
The reaction of 1 with H2O2 was evaluated also from another
set of data. A series of experiments was conducted at constant
[H2O2]0 ¼ 4.71 ꢀ 10ꢁ3 M over a pH range from 9.14 to 12.4 and the
results are collected in Table S4. When kobs/[H2O2]0 was plotted
against XHOOꢁ a straight line was obtained (Fig. S2), and the value
of kHOOꢁ was calculated as 7.8 ꢂ 0.1 Mꢁ1 sꢁ1, in good agreement
with the value obtained before. By plotting together all the values
calculated as (1.70 ꢂ 0.03) ꢀ 10ꢁ2
M
ꢁ1 sꢁ1. This value is similar to
that obtained in ACN/H2O indicating that both organic solvents
have similar influence on the reaction rate.
Reaction with H2O2
The reaction of 1 with H2O2 was studied in a pH range from 9.14
to 12.40 in 7% 1,4-dioxane/H2O, I ¼ 0.6 M (NaCl) at 25 8C; buffers of
H2O2 or 0.05 M Borax were used to control the pH of the solutions.
We have used this organic solvent due to the known reaction of
ACN with H2O2.[41] The reaction was followed by measuring the
appearance of 3 at 320 nm. Quantitative formation of 3 indicates
that the only reaction taking place is SN2(P) as was observed in
the hydrolysis reaction. Considering the fact that 3 was formed
quantitatively, that during the kinetic measurements evolution of
bubbles was observed and taking into account reported
data,[42,43] we suggest that the reaction taking place is that
shown in Scheme 1.
70.0
pH 10.9
60.0
pH 11.3
pH 11.5
pH 11.6
50.0
pH 11.9
40.0
30.0
20.0
10.0
0.0
Values of kobs were determined at different pH values varying
the analytical concentration of H2O2 ([H2O2]0) at each pH; the
results are shown in Table S2. The nucleophiles present in
the reaction solutions are H2O, HOꢁ, H2O2, and HOOꢁ, so the
0.0
4.0
8.0
3
12.0
16.0
20.0
10 [H O ] (M)
2
2 0
Figure 3. Plot of kobs versus [H2O2]0 for the reaction of 1 with H2O2 at
25 8C at different pHs. Solvent contains 7% 1,4-dioxane; ionic strength
I ¼ 0.6 M (NaCl); [1]0 ¼ (3.07–3.10) ꢀ 10ꢁ5
M
Scheme 1.
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
8.0
6.0
4.0
2.0
0.0
A number of studies have demonstrated the a-effect in
nucleophilic attack at the P center.[33,38,42,43,45,46] It was reported
an a-effect of 210 for the HOOꢁ/HOꢁ pair for the reaction with
2,[33] this is nearly half of the value that we calculated. However,
the same authors note that the magnitude of the a-effect is
substrate-dependent.[33]
Effect of cyclodextrins
The effect of b-CD on the hydrolysis of 1 was determined in 10%
ACN/H2O at three HOꢁ ion concentrations and the data are
summarized in Table S5. The addition of b-CD produces a
nonlinear decrease in the observed rate constants and a
saturation effect as shown in Fig. 6.
In order to determine the importance of the size of the cavity of
the cyclodextrin in the observed effect, the rate constants were
determined in the presence of a- and g-CD at 0.50 M NaOH. The
effect of sucrose,
a non-reducing disaccharide, was also
0.0
0.2
0.4
0.6
0.8
1.0
investigated. The data are collected in Table 1. It can be seen
that the greater inhibition is produced by g-CD whereas the
effect of a- and b-CD are similar. On the other hand, it was
previously observed that b-CD produces a larger inhibition of the
hydrolysis of 2 than the others CDs.[25] These contrasting results
indicate a high specificity of cyclodextrins in the recognition of
the nature of the guest.
As the greater inhibition was observed with g-CD, we
investigated the effect of increasing concentrations of g-CD on
the hydrolysis of 1 at 0.50 M HOꢁ ion concentration (Table 2). As
was observed with b-CD, the addition of g-CD produces a
nonlinear decrease in the observed rate constants and saturation
kinetics as shown in Fig. S3.
We have also studied the effect of increasing amounts of
sucrose on the hydrolysis of 1 in weight amount equal to
solutions (0.05–0.2) ꢀ 10ꢁ2 M in b-CD at 0.50 M NaOH. The addition
of sucrose produces an inhibition of the hydrolysis of 1 which is
independent of the amount of sugar; the observed rate constants
under these conditions have a mean value of (6.1 ꢂ 0.6) ꢀ
10ꢁ3 sꢁ1 (Table S6).
The effect of b- and g-CD on the reaction of 1 with H2O2 was
determined at pH 11.30 at constant [H2O2]0 ¼ 4.71 ꢀ 10ꢁ3 M in 7%
1,4-dioxane/H2O, I ¼ 0.6 M (NaCl) at 25 8C and the data are
summarized in Table S7. The presence of increasing amounts of
b-CD produces a slight decrease in kobs independent of b-CD
concentration with a mean value of (11.4 ꢂ 0.3) ꢀ 10ꢁ3 sꢁ1. On
the other hand, g-CD produces a nonlinear decrease in the
observed rate constants and a saturation effect as shown in Fig. 7.
We have also evaluated the effect of a-CD and sucrose on the
reaction of 1 with H2O2 under the same conditions. In Table 3 are
summarized the results for all the sugars. It can be seen that
again, g-CD produces the greater inhibition whereas a- and b-CD
show only a little inhibitory effect and sucrose has no effect,
within experimental error.
XHOO-
Figure 4. Plot of k[H202]0 versus XHOOꢁ for the reaction of 1 with H2O2 at
25 8C. Solvent contains 7% 1,4-dioxane; ionic strength I ¼ 0.6 M (NaCl);
[1]0 ¼ (3.07–3.10) ꢀ 10ꢁ5
M
of kobs obtained in the two series of experiments against [HOOꢁ]
(Fig. 5), kHOOꢁ was calculated as 7.9 ꢂ 0.1 Mꢁ1 sꢁ1
.
As the pKa of H2O2 is 11.6,[33] in the pH range used for this
study, both H2O2 and HOOꢁ are present in the reaction solution.
The results indicate that only the HOOꢁ ion acts as nucleophile
toward 1. Besides, this anion is 465 times more reactive than HOꢁ
in spite of the fact that it is less basic than the latter
(pKHa 2O ¼ 15.74);[44] this could be attributed to the fact that
HOOꢁ is an a-nucleophile.
80.0
60.0
40.0
20.0
0.0
The observation of inhibition of the reaction with saturation
kinetics on the hydrolysis upon addition of b- and g-CD and on
the reaction with H2O2 upon addition of g-CD, indicates that
these hosts form inclusion complexes with 1. The fact that the
greater inhibition is observed with g-CD may indicate a better fit
of the insecticide in the cavity of this CD.
The addition of cyclodextrins or sucrose does not affect the
reaction products; therefore the only reaction taking place, with
or without CDs, is P—O bond fission (Eqn (1)).
0.0
2.0
4.0
6.0
8.0
10.0
103[HOO-] (M)
Figure 5. Plot of kobs versus [HOOꢁ] for the reaction of Chlorpyrifos-
Methyl (1) with H2O2 at a pH range from 9.14 to 12.40 at 25 8C. Solvent
contains 7% 1,4-dioxane; ionic strength I ¼ 0.6 M (NaCl); [1]0 ¼
(3.07–3.10) ꢀ 10ꢁ5
M
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702

REACTIVITY OF CHLORPYRIFOS-METHYL
(a) 10.0
Table 1. Effect of a-, b-, and g-CD and sucrose on the
hydrolysis of Chlorpyrifos-Methyl (1) at 25 8C at 0.5 M NaOHa
9.0
8.0
7.0
6.0
5.0
3
10 k
(sꢁ1
)
koCbDs =k
CD
obs
OH
CD
[CD] (M)
obs
0
9.40
6.52
6.49
4.68
6.48
a
0.020
0.020
0.020
0.69
0.69
0.50
0.69
b
g
Sucrose
b
a Solvent contains 10% ACN; ionic strength I ¼ 0.6 M (NaCl);
[1]0 ¼ (2.90–3.10) ꢀ 10ꢁ5 M; [NaOH]eff ¼ 0.500 M.
b Weight amount equal to a solution 0.02 M in b-CD.
0.0
5.0
1.0
2.0
3.0
It was reported before that a-, b-, and g-CD have promotive
inclusion-catalytic effects on the degradation of some organo-
phosphorous pesticides solubilized in neutral aqueous media;
pesticide degradations were especially accelerated in the
systems a-CD plus Diazinon and b-CD plus 5.[47] Kamiya
et al.[48] have also informed that in buffer solution (pH ¼ 8.5)
containing humic acids, b-CD inhibits the hydrolysis of Parathion
and Methyl-Parathion and catalyses the hydrolysis of Paraoxon
while g-CD inhibits the hydrolysis of the three pesticides. To
interpret the medium dependence of CD effects they proposed
that as in alkaline media the hydrogen-bond linkages of the
peripheral hydroxyl groups in the rim of CD cavity are destructed
by their ionizations, it is highly probable that electrostatic
repulsions of the ionized hydroxyl groups lead to widening of the
CD torus which would allow deep inclusions of the pesticides in
alkaline media. This may cause the degradation inhibition of the
included pesticides.[47]
(b)
4.5
4.0
3.5
3.0
2.5
2.0
As 1,4-dioxane was used instead of ACN in the reaction of 1
with H2O2, the absence of saturation kinetics in this reaction in
the presence of increasing amounts of b-CD and the lower
inhibition observed in the presence of g-CD, may be a
consequence of the fact that 1,4-dioxane has a greater
0.0
1.0
2.0
3.0
2.0
1.8
1.6
1.4
1.2
1.0
0.8
(c)
Table 2. Dependence of observed rate constants on [g-CD]
for the hydrolysis of Chlorpyrifos-Methyl (1) at 25 8Ca
102 [g-CD] (M)
103 kobs (sꢁ1 b
)
0.0512
0.200
0.501
0.701
1.00
1.20
1.50
1.70
2.00
9.0 ꢂ 0.1
6.3 ꢂ 0.6
5.874 ꢂ 0.008
4.6 ꢂ 0.1
4.57 ꢂ 0.01
5.303 ꢂ 0.007
4.7 ꢂ 0.2
0.0
1.0
2.0
3.0
102 [β-CD] (M)
3.9 ꢂ 0.2
3.8 ꢂ 0.1
CD
obs
Figure 6. Plots of k versus [b-CD] for the hydrolysis of 1 at 25 8C at
2.00
4.679 ꢂ 0.003
constant [NaOH]. Solvent contains 10% ACN; ionic strength I ¼ 0.6 M
(NaCl); [1]0 ¼ (2.9–3.1) ꢀ 10ꢁ5 M; [NaOH]eff ¼ (a) 0.500, (b) 0.250, and
(c) 0.100 M
a Solvent contains 10% ACN; ionic strength I ¼ 0.6 M (NaCl);
[1]0 ¼ 3.10 ꢀ 10ꢁ5 M; [NaOH]eff ¼ 0.500 M.
b Errors are the standard deviation of the fit of the absorbance
versus time data to a single exponential equation.
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
13.0
12.0
11.0
10.0
9.0
Although the magnitude of the inhibition of the hydrolysis of 1
produced by sucrose is similar to that observed with a- and b-CD,
the origin of the inhibition is different as no saturation kinetics
was observed. The inhibition of the reaction by sucrose can be
attributed to a medium effect or to some type of unspecific
association of the substrate with the sugar. We do not think that
the effect is due to changes in pH. At the high OHꢁ ion
concentration used, the consumption of hydroxide ion due to
reaction with the HO of the sugar is very small since the pKa of
sucrose is not expected to be lower than that of b- or g-CD. It is
known that starch and some linear oligosaccharides can
self-associate and form complexes with various types of
compounds.[52] There are examples in the literature of changes
in the reaction rates, accelerations, or inhibitions of different
substrates produced by the presence of linear sacchar-
ides.[34,53,54]
8.0
7.0
Spectroscopic studies
6.0
0.0
0.4
0.8
1.2
1.6
Inclusion complex formation between CD and different guest
molecules can be detected by a variety of spectroscopic
methods.[13] UV–Vis spectroscopy is one of the most frequently
used technique to determine the association constants between
an organic compound and CDs.[55]
102 [γ-CD] (M)
Figure 7. Plot of kobs versus [g-CD] for the reaction of 1 with H2O2 at
25 8C. Solvent contains 7% 1,4-dioxane; ionic strength I ¼ 0.6 M (NaCl);
[1]0 ¼ 3.07 ꢀ 10ꢁ5 M; [H2O2]0 ¼ 4.71 ꢀ 10ꢁ5 M; pH ¼ 11.30
We intended to determine the binding constant of 1 with
b- and g-CD by UV–Vis spectroscopy. Thus, we carried out the
spectra of 1 in the presence of increasing amounts of b- or g-CD.
The UV–Vis spectrum of 1 in the presence of b-CD is identical to
that in the presence of g-CD. Although the spectrum of 1 in the
presence of any of the CDs shows a small hypsochromic shift,
the changes in absorbance observed are not great enough to
enable the determination of the association constants (Fig. S4).
We have also studied the interaction of 1 with b- and g-CD by
circular dichroism. The addition of b-CD to a solution of 1 gives an
induced circular dichroism spectrum (ICD) with a negative peak
at 230 nm and a positive one at 289 nm; these peaks coincide
with the maxima in the UV–Vis spectrum (Fig. 8). On the other
hand, g-CD produces a more intense ICD spectrum with two
negative peaks at 240 and 269 nm (Fig. 8). As is observed in other
cases where CD is the host,[56–59] these peaks are slightly
displaced from the maxima in the UV–Vis spectrum. The fact that
the ICD signal at the higher wavelength changes from negative to
positive and the ellipticity of that signal decreases when the host
changes from g- to b-CD indicates an extrusion of the guest from
association constant with b- or g-CD than ACN. Reported values
for these constants are Kass ¼ (20.0 ꢂ 2.5) Mꢁ1 for a 1:1 association
of 1,4-dioxane with b-CD or its anion, Kass ¼ (0.60 ꢂ 0.05) Mꢁ2 and
(0.50 ꢂ 0.05) Mꢁ2 for a 2:1 association of ACN with b-CD and its
anion, respectively;[49] Kass ¼ 0.54 Mꢁ1 for a 1:1 association of ACN
with b-CD.[50] The organic co-solvent competes with 1 toward the
cavity of the CD; as the affinity of the co-solvent with the host
increases, a greater amount of the substrate will be displaced
from the cavity of the host, increasing the amount of free
substrate and decreasing the observed inhibition effect.[51] The
kinetically determined association constant for 1 with b-CD in
ACN/H2O (see bellow) 105 Mꢁ1 is small, so as 1,4-dioxane has a
greater affinity with b-CD than ACN, it is reasonable to think that
in the presence of dioxane, a lower concentration of 1:b-CD
would be responsible for the absence of saturation kinetics in the
reaction with H2O2. Assuming a greater Kass for 1,4-dioxane with
g-CD than for ACN, the smaller inhibition produced by g-CD on
the reaction of 1 with H2O2 than in its hydrolysis can be explain by
the same arguments.
the cavity as was suggested before by others.[60] If we compare
[61]
˚
the distance between the OCH3 groups on P (5.573 A)
or
of
[61]
˚
between the Cl substituents on C3 and C5 (6133 A)
insecticide 1 (Fig. 9) and the cavity diameters of b-CD (internal
Table 3. Effect of a-, b-, and g-CD and sucrose on the reaction
of Chlorpyrifos-Methyl (1) with H2O2 at 25 8C and pH ¼ 11.30a
[14]
diameter ¼ 5.6 A) and g-CD (internal diameter ¼ 6.8 A),[14] it is
evident that 1 can be more deeply included in the cavity of g-CD
than in the cavity of b-CD.
˚
˚
3
10 k
CD
obs
OOH
obs
CD
[CD] (M)
(sꢁ1
)
koCbDs =k
We have also studied the interaction of 3 with b- and g-CD by
circular dichroism. As can be seen in Fig. 8, the spectra obtained
are less defined and intense than those observed with
1 suggesting a weaker interaction with the hosts than that of
the insecticide.
In order to confirm that the formation of inclusion complexes
between 1 and b- or g-CD is responsible of the ICD spectra
observed, we studied the interaction of 1 with sucrose and
raffinose, two non-reducing saccharides. Although the UV–Vis
spectra of 1 in the presence of sucrose or raffinose are identical to
that in the presence of b-CD, these compounds induce weak ICD
0
11.76
11.08
10.96
7.853
11.94
a
0.015
0.015
0.015
0.94
b
0.93
0.67
1.02
g
Sucrose
b
a Solvent contains 7% 1,4-dioxane; ionic strength I ¼ 0.6 M
(NaCl); [1]0 ¼ 3.07 ꢀ 10ꢁ5 M; [H2O2]0 ¼ 4.71 ꢀ 10ꢁ3 M.
b Weight amount equal to a solution 0.015 M in b-CD.
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702

REACTIVITY OF CHLORPYRIFOS-METHYL
2
signals (Fig. S6), different to those observed with the CDs, which
can be attributed to some type of unspecific association of the
substrate with the sugar.
A
B
C
D
NMR is the most widely used method for structural elucidation
of organic compounds in solution. It is also very useful to provide
evidence for the formation of inclusion complexes in solution
since this will affect the nuclei environment in both the host and
the guest and will hence be reflected by chemical shift variation.
It allows the calculation of the binding constant between the host
and the guest as well as the binding modes in the
complexes.[55,62,63] Complex formation between 1 and b-CD
1
0
1
was also investigated by 31P and H NMR.
-1
Proton-decoupled 31P NMR spectra were recorded for 1
(8.4 ꢀ 10ꢁ4 M) in aqueous media in the absence and presence of
8.7 ꢀ 10ꢁ4 M b-CD. In the presence of b-CD two signals were
observed, one at 64.43 ppm corresponding to 1 and the other at
0.38 ppm downfield from that of 1 (Fig. 10) corresponding to 1
included in b-CD. Downfield changes in the 31P NMR chemical
shift for the organophosphorus pesticide diazinon (7) were
reported before in the presence of a-, b-, and g-CD and
interpreted as formation of an inclusion complex.[64]
-2
0.4
0.3
0.2
0.1
0.0
The spectra of b-CD in 1,4-dioxane/D2O (10:90) were
recorded in the absence and the presence of 1. All the signals
corresponding to b-CD in the presence of 1 appear at lower fields
than in its absence (Table S8). The fact that DdH3 < DdH5
indicates that the guest is deeply included inside the cavity of the
200
250
300
350
Wavelength (nm)
1
host.[65] As was observed in 31P NMR, in the H NMR of 1 in the
Figure 8. UV–Vis (lower) and ICD (upper) spectra in 10% ACN/H2O, ionic
strength I ¼ 0.6 M at 25 8C of (A) 1 (3.10 ꢀ 10ꢁ5 M) with 0.010 M b-CD; (B) 3
(3.07 ꢀ 10ꢁ5 M) with 0.010 M b-CD; (C) 1 (3.10 ꢀ 10ꢁ5 M) with 0.008 M g-CD;
(D) 3 (3.07 ꢀ 10ꢁ5 M) with 0.008 M g-CD
presence of b-CD, two signals for each type of protons appears
corresponding to free and complexed substrate. In Table S9 are
listed the chemical shifts of 1 in the presence and absence of
b-CD; it can be observed that all the signals are shifted to higher
fields in the presence of the host and the most affected is the
signal corresponding to the aromatic proton.
Figure 10. 31P NMR spectrum in 10% [D8]1,4-dioxane/D2O at room
temperature of 1 (8.4 ꢀ 10ꢁ4 M) (lower) and of 1 in the presence of
b-CD 8.7 ꢀ 10ꢁ4 M (upper)
Figure 9. Schematic representation of the structure of 1 as calculated by
Hyper Chem 7.5
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
A 2D ROESY was conducted for 1 in the presence of b-CD. Due
to the low solubility of the pesticide in water, ROESY-2D spectra
have low resolution and only interaction between the aromatic
proton of the guest with H-3 of b-CD could be clearly detected.
The signals corresponding to the OCH3 groups of the guest in the
complex appear close to those of b-CD and no definitive
conclusion about their interaction could be established.
The NMR experiments indicate that 1 is deeply included in the
cavity of b-CD, there is an important interaction of the aromatic
proton with H-3 of b-CD and the rate of exchange of 1 between
the cavity and the bulk solution is slower than the time scale of
NMR since two signals for the H or P in the guest were observed in
the presence of the host.
Scheme 2.
parameters given in Eqn (9) and (10).
a þ b½CDꢃ
1 þ c½CDꢃ
Reaction mechanism in the presence of CDs
kobs
¼
(8)
We have found that H2O does not compete with HOꢁ ion in the
hydrolysis of 1 and that H2O, HOꢁ, and H2O2 do not compete with
HOOꢁ ion in the reaction with 1. Considering that the pKa of b-CD
is 12.2 and that of g-CD is 12.08,[66] as demonstrated in previous
work with other substrates,[67] and with 2,[25] the reaction of 1
with HOꢁ or HOOꢁ in the presence of CDs may take place as
indicated in Scheme 2. In Scheme 2 Nuꢁ represents HOꢁ or HOOꢁ
and k0Nu, k1Nu, k2, and kN3u represent the reactions of the free
substrate, the substrate complexed with neutral b- or g-CD
(CDOH), the reaction with ionized b- or g-CD (CDOꢁ) and the
reaction of the substrate complexed with CDOꢁ reacting with
HOꢁ or HOOꢁ ion, respectively.
b ¼ k1NuK1½Nuꢁꢃð1 ꢁ fÞ þ ðk2 þ k3Nu½NuꢁꢃÞK2f
¼ fk1NuK1Kb þ ðk2 þ k3Nu½NuꢁꢃÞK2gf
c ¼ K1ð1 ꢁ fÞ þ K2f
(9)
(10)
The hydrolysis reactions in the presence of b-CD were studied
at 0.10, 0.25, and 0.50 M NaOH while 0.50 M NaOH was used for the
reactions carried out in the presence of g-CD. Under the
condition of a constant HOꢁ ion concentration of 0.50 M, f ꢄ 1;
then Eqn (6) simplifies to Eqn (11).
k0HO½HOꢁꢃ þ ðk2 þ k3HO½HOꢁꢃÞK2½CDꢃ
kobs
¼
(11)
The observed rate constant for the mechanism shown in
Scheme 2 is given by Eqn (6), where f represents the fraction of
ionized b- or g-CD as defined in Eqn (7) with Kb ¼ KW/Ka, and CD is
the stoichiometric concentration of b- or g-CD.
1 þ K2½CDꢃ
The values of a, b, and c obtained using Eqn (8) to fit the data
are given in Table 4, where the experimentally determined value
of k0HO [HOꢁ] is included.
Taking the value K2 as the value of c at 0.5 M HOꢁ, we can
calculate the association constants of 1 with b- and g-CD as 105
and 461 Mꢁ1, respectively. A good fit to Eqn (8) was obtained for
all the [HOꢁ] studied in the presence of b-CD using the same
value for c. So, according to Eqn (10), K2 must not be significantly
different from K1, as was previously reported for other guests.[68]
A value of Kass ¼ 90 ꢂ 28 Mꢁ1 was informed for the complexation
of 5 with b-CD, in good agreement with that calculated here.
A plot of b/f versus HOꢁ ion concentration (Fig. S6) for the data
in the presence of b-CD, is linear with slope 0.9 Mꢁ2 sꢁ1. The
intercept of this line is indistinguishable from zero within
experimental error, indicating that the main reaction pathway of
k0Nu½Nuꢁꢃþk1NuK1½Nuꢁꢃð1ꢁfÞ½CDꢃ þ ðk2 þk3Nu½NuꢁꢃÞK2f½CDꢃ
kobs
¼
1 þ K1ð1 ꢁ fÞ½CDꢃþK2f½CDꢃ
(6)
½HOꢁꢃ
f ¼
(7)
½HOꢁꢃ þ Kb
In all cases where saturation kinetics was observed, the
observed rate constants were fitted to an equation of the form of
Eqn (8) where a ¼ kN0u [Nuꢁ] and b and c are adjustable
Table 4. Parameters of Eqn (8) for the reaction of Chlorpyrifos-Methyl (1) with HOꢁ or HOOꢁ in the presence of b- and g-CD at 25 8Ca
Nuꢁ
CD
[Nuꢁ]eff (M)
103 ab
b
c
f
b/f
HOꢁ
HOꢁ
HOꢁ
HOꢁ
HOOꢁ
b
b
b
g
g
0.50
0.25
0.10
0.50
9.3 ꢂ 0.2 (9.4)
4.6 ꢂ 0.1 (4.7)
0.51 ꢂ 0.03
0.14 ꢂ 0.01
0.067 ꢂ 0.009
1.7 ꢂ 0.1
105 ꢂ 4
105c
1.00
0.94
0.86
1.00
0.11
0.51
0.149
0.078
1.7
1.68 ꢂ 0.06 (1.88)
9.6 ꢂ 0.4 (9.4)
105c
461 ꢂ 22
0.00157
11.8 ꢂ 0.2 (11.76)
2.47 ꢂ 0.08
366 ꢂ 9
0.73
a Solvent contains 10% ACN for the reaction with HOꢁ or 7% 1,4-dioxane for the reaction with HOOꢁ.
b Values in parenthesis are k0Nu [Nuꢁ].
c The fit was made keeping c constant.
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702

REACTIVITY OF CHLORPYRIFOS-METHYL
Scheme 2 is that including k3, as was previously observed for 2.[25]
decrease in rate for the reactions with g-CD is mainly due to the
higher binding constant.
ꢁ1
With K2 ¼ 105 M , k3HO is calculated as 8.57 ꢀ 10ꢁ3
M
ꢁ1 sꢁ1, which
is about two times smaller than the value for the rate constant for
the reaction of the free substrate, kN0u. As we proposed before for
the hydrolysis of 2,[25] this may be attributed in part to
electrostatic repulsion of the negative hydroxide ion and in part
to steric hindrance to nucleophilic attack imposed by the
cyclodextrin rim. The fact that the main reaction pathway for the
cyclodextrin-mediated reaction is the reaction of HOꢁ ion with
the complexed substrate and that there appears to be no
significant nucleophilic reaction of the ionized secondary OH
of the cyclodextrin may indicate that in the structure of the
complex the thiophosphate group is not at an appropriate
distance from the ionized secondary OH of the cyclodextrin rim.
The electronegativity of sulfur (2.44) is smaller than that of oxygen
The kinetic data of the reaction of 1 with H2O2 in the presence
of g-CD were fitted to Eqn (8) and the values of parameters a, b,
and c obtained are shown in Table 4, where the experimentally
determined value of k0HOO [HOOꢁ] is also included. Assuming that
K1 ꢄ K2, and that the main reaction pathway is that involving
k3HOO, as was observed previously in the hydrolysis reaction,[71]
the association constant of 1 with g-CD in 7% 1,4-dioxane/H2O
could be calculated from parameter c and k3HOO from parameter b
ꢁ1
(Table 5). The value 366 ꢂ 9 M obtained for K2 is 20% smaller
than that kinetically determined in the hydrolysis reaction in the
presence of 10% ACN. In the presence of g-CD, the ratio
k3HOO=k3HO ¼ 583 indicates a greater a effect for the reaction of the
substrate included in the cavity with the nucleophile than for the
free substrate with the nucleophile.
˚
(3.50) whereas the atomic radio of sulfur (1.27 A) is greater than
[69]
˚
that of oxygen (0.60 A). Besides, the value of log PO/W for 1 is
There are two possible modes of inclusion of the insecticide 1
in the cavity of the CDs as shown in Scheme 3, with the aromatic
moiety or with the thiophosphate moiety inside the cavity. For
most of the esters having an aromatic group it is proposed that
this region is included in the cavity of the CDs. This was suggested
for 2, Parathion, Pharathion-Methyl, and Paraoxon.[48,72,73] It is
known that the observed acceleration in the release of phenol
from substituted acetates in the presence of CDs is independent
of the electronic nature of the substituents but it strongly
depends on the position that they occupied in the aromatic ring.
In the substrates having a meta substituent in the aromatic ring,
the ring is not deeply included in the cavity and the carbonylic
carbon of the ester is located near the secondary OH of CD, thus
catalysis of the reaction of phenol release is observed.[34] On the
other hand, in esters having para substituted phenols, the
inclusion is deeper placing the carbonylic group far away from
the secondary OH with a weaker catalysis of the reaction.[34] For
instance, it was suggested that the t-butyl group of
m-t-butylphenylacetate is included in the cavity of a-CD whereas
the aromatic region remains in the aqueous solution consistent
with the fact that the UV–Vis spectrum of the complex shows only
a little perturbation compared with that of the substrate.[34] In
contrast, p-t-butylphenylacetate shows an important spectral
change in the presence of a-CD suggesting the inclusion of the
aromatic region.[34] Moreover, the inhibition of the hydrolysis of
the esters CF3(CF2)nCOOPh (n ¼ 1 and 2) produced by b-CD was
suggested to arise from the fact that the reaction take place with
the hydrophobic perfluorinated alkyl chain included in the
cavity.[22] The complexation of diazinon (7), with a-, b-, and g-CD
was investigated by NMR and computational studies.[64] Binding
4.28 whereas that for the phosphate analog is 3.55.[70] Thus, the
thiophosphate group is more hydrophobic than the phosphate
and the first one may be more deeply included in the cavity of
the CDs. This is in agreement with the fact that monothiopho-
sphates form inclusion complexes with CDs that are more stable
than those of the corresponding monophosphates.[69]
In Table 5 are summarized the values of the rate and
equilibrium constants calculated for the mechanism depicted in
Scheme 2 for the hydrolysis of 1 in the presence of b- and g-CD
along with those previously calculated for 2.[25]
Comparing the results of 1 with b-CD with those for 2,[25] we
found that for the latter compound K2 is about four times greater
and kH3O is sixteen times smaller than those for compound 1. On
the other hand, K2 for 1 with g-CD is comparable to that for 2 with
ꢁ1
b-CD.[25] With K2 ¼ 461 M
for g-CD, kH3O is calculated as
7.37 ꢀ 10ꢁ3
M
ꢁ1 sꢁ1, which is about 2.5 times smaller than the
value for the rate constant for the reaction of the free substrate,
k0HO. With the values of k3HO with g- and b-CD shown in Table 5, we
can calculate k3HOðgÞ=k3HOðbÞ as 0.86, indicating that the rate
constants for the reaction of the substrate included in the cavity
of b and g-CD are very similar. Therefore, the observed higher
Table 5. Rate and equilibrium constants calculated for the
hydrolysis of Chlorpyrifos-Methyl (1) and Fenitrothion (2) and
for the reaction of 1 with H2O2
1
2a
constants determined by 1H and 31P NMR follow the order
ꢁ1 [64]
102 k0HO (Mꢁ1 sꢁ1 b
102 k0HO (Mꢁ1 sꢁ1 c
)
1.88 ꢂ 0.05
1.70 ꢂ 0.03
105 ꢂ 4
8.57
0.20 ꢂ 0.01
g-CD > a-CD ¼ b-CD, being that for g-CD, i.e., 390 ꢂ 63 M
,
similar to K2 determined by us for 1. The computational
)
K2bꢁCD (M
103 k3HObꢁCD (Mꢁ1 sꢁ1 b
)
ꢁ1 b
415 ꢂ 72
0.53
)
K2gꢁCD (M
103 k3HOgꢁCD(Mꢁ1 sꢁ1 b
k0HOO (Mꢁ1 sꢁ1 c
)
ꢁ1 b
461 ꢂ 22
7.37
)
7.9 ꢂ 0.1
366 ꢂ 9
4.30
)
Kg2ꢁCD (M
kH3OOgꢁCD (Mꢁ1 sꢁ1 c
)
ꢁ1 c
)
a Taken from Reference [25].
b Solvent contains 10% ACN.
c Solvent contains 7% 1,4-dioxane.
Scheme 3.
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
calculations indicate that while in a- and b-CD the aromatic
moiety of 7 is located inside the cavity and the phosphoryl
residue is largely outside it, in g-CD the heterocyclic residue and
the phosphoryl residue of the guest are both largely encrypted
in the CD cavity with the methyl substituent on C5 emerging from
the wider rim of CD.
In this work, we present clear evidence of the formation of
inclusion complexes between 1 and a-, b-, or g-CD, although we
cannot assure which of the two possible modes of inclusion
depicted in Scheme 3 is taking place with each of the CDs.
pKa determination
The pKa of 3,5,6-trichloro-2-pyridinol (3) was determined by
spectrophotometric titration in 10% ACN/H2O at 25 8C and ionic
strength 0.6 M. Solutions for the various pH ranges were prepared
using HCl (2.00–3.04), acetic acid/sodium acetate (4.00–5.50),
K2HPO4/KH2PO4 (5.40–8.00), borax (8.80), Na2CO3/NaHCO3 (9.70),
and NaOH (12.0).
Kinetic procedures
Reactions were initiated by adding the substrate dissolved in ACN
or 1,4-dioxane to a solution containing all the other constituents.
The reaction temperature was 25.0 ꢂ 0.1 8C, the ionic strength
was 0.6 M and NaCl was used throughout as compensating
electrolyte. The solvent contained 10% ACN (hydrolysis) or 7%
1,4-dioxane (hydrolysis and reaction with H2O2). In the reactions
with H2O2, water and 1,4-dioxane were degassed and the stock
solution of NaCl was filtered by nylon membranes before
preparing the reaction solutions. After the pH was adjusted to the
desired value, the solution was sonicated for 5 min and
thermostated to 25 8C before the substrate was added.
All kinetic runs were carried out under pseudo-first-order
conditions, with substrate concentrations of (2.9–3.1) ꢀ 10ꢁ5 M.
The reactions were followed by measuring the increase in
absorbance of the reaction mixture at 320 nm, the lmax of 3. The
[HOꢁ]eff values reported in Tables 1–4 were calculated from
stoichiometric concentration of NaOH and CD and the pKa of
CDOH (12.2 or 12.08 for b- or g-CD, respectively).[66]
CONCLUSIONS
Chlorpyrifos-Methyl (1) forms inclusion complexes with a-, b-,
and g-CD in aqueous solutions as is evidenced by kinetic and
spectroscopic methods.
Chlorpyrifos-Methyl is decomposed quantitatively to
3,5,6-trichloro-2-pyridinol (3) in basic aqueous media by HOꢁ
or HOOꢁ through a SN2(P) pathway with an important a-effect.
The hydrolysis of 1 is inhibited by a-, b-, and g-CD and sucrose;
the greater inhibition was observed with g-CD and saturation
kinetics was found in the case of b- and g-CD. The reaction of 1
with H2O2 is slightly inhibited by a- and b-CD and inhibited by
g-CD with saturation kinetics. The effect of g-CD is greater for the
hydrolysis reaction than for the reaction with H2O2; this may be a
consequence of differences in solvation/desolvation of the two
nucleophiles, HOꢁ or HOOꢁ as was proposed before to explain
the a-effect in the reactions of other phosphorous com-
pounds.[46] The kinetic data indicate that the main reaction
pathway for the cyclodextrin-mediated reaction is the reaction of
HOꢁ or HOOꢁ ion with the substrate complexed with the anion
of the cyclodextrin, as was previously found for the hydrolysis of
Fenitrothion (2).[25]
Spectroscopic procedures
The circular dichroism measurements were recorded at 25 8C in
10% ACN/H2O at an ionic strength of 0.6 M by the addition of NaCl;
[1]0 ¼ 3.1 ꢀ 10ꢁ5 M; the concentration of b- or g-CD was equal to
0.01 M while sucrose or raffinose were used in a weight amount
equal to a solution 0.01 M in b-CD. To remove background
contributions, circular dichroism measurements were always
made using the free CD, sucrose or raffinose solution as a
reference system. Water and aqueous salt solutions were filtered
by nylon membranes before the measurements were done.
EXPERIMENTAL SECTION
Materials
Chlorpyrifos-Methyl (1) was characterized by 1H and 13C NMR and
GC–MS. The hydrolysis product 3,5,6-trichloro-2-pyridinol (3), was
obtained by hydrolysis of Chlorpyrifos-Ethyl (5) with KOH in 20%
ACN at room temperature and characterized by 1H and 13C NMR
and GC–MS. a-, b-, g-cyclodextrin, sucrose and raffinose were
used as received, but the purity was periodically checked by
UV–Vis spectroscopy.
Aqueous solutions were prepared using water purified with a
Millipore Milli-Q apparatus. Acetonitrile (HPLC grade) was used as
received and 1,4-dioxane was purified as described previously.[74]
All of the inorganic reagents were of analytical-reagent grade and
were used without further purification. Hydrogen peroxide
solutions were titrated with a KMnO4 solution standardized by
sodium oxalate.
UV–Vis spectra were recorded on a Shimadzu UV-2101 or HP
Multispect 1501 spectrophotometer and the change in absor-
bance during a kinetic run was measured on the same
instruments. Circular dichroism spectra were recorded on a
JASCO Model J-810 spectropolarimeter which was calibrated with
D-(þ)-amonium camphorsulfonate. 31P and 1H NMR spectra were
recorded on a Bruker Ultra Shield 400. GC–MS were performed on
a Shimadzu Model CQ5050 instrument.
NMR experiments
The 31P NMR spectra were recorded on a spectrometer operating
at 161.97 MHz. Chemical shifts were measured with respect to the
external standard of 85% H3PO4 in D2O/H2O (75:25). Spectra were
acquired with a longitudinal relaxation time (D1) of 30 s with 128
accumulations and proton decoupling. In the experiments
conducted for products identification (Fig. 1), 1,4-dioxane was
used while in those for investigation of complex formation
[D8]1,4-dioxane was used.
Sample preparation: For complex investigation by 31P NMR,
70 ml of a stock solution of 1 8.06 ꢀ 10ꢁ3 M in [D8]1,4-dioxane was
added to 0.6 ml of D2O or to 0.6 ml of b-CD solution 9.9 ꢀ 10ꢁ4
M
in D2O. The final concentrations were: [1]0 ¼ 8.4 ꢀ 10ꢁ4 M,
[b-CD] ¼ 8.7 ꢀ 10ꢁ4 M, [D8]1,4-dioxane ¼ 10%.
For complex investigation by 1H NMR a stock solution of b-CD
was prepared by dissolving 0.004 g in 3 ml of D2O
([b-CD] ¼ 1.17 ꢀ 10ꢁ3 M). For the spectra of b-CD, 70 ml of
[D8]1,4-dioxane were added to 0.6 ml of the stock solution of
b-CD. Final concentrations were: [b-CD] ¼ 1.04 ꢀ 10ꢁ3 M,
[D8]1,4-dioxane ¼ 10%. For the 1H NMR and ROESY-2D
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702

REACTIVITY OF CHLORPYRIFOS-METHYL
experiments of 1 in the presence of b-CD, to 0.6 ml of the stock
solution of b-CD, 60 ml of [D8]1,4-dioxane and 70 ml of a stock
solution of 1 8.06 ꢀ 10ꢁ3 M in [D8]1,4-dioxane were added. Final
concentrations were: [1]0 ¼ 7.73 ꢀ 10ꢁ4 M, [b-CD] ¼ 9.56 ꢀ 10ꢁ4 M,
[31] U. Jans, M. H. Miah, J. Agric. Food Chem. 2003, 51, 1956–1960.
[32] R. J. Maguire, E. J. Hale, J. Agric. Food Chem. 1980, 28, 372–378.
[33] J. E. Omakor, I. Onyido, G. W. vanLoon, E. Buncel, J. Chem. Soc. Perkin
Trans 2001, 324–330.
[34] R. L. VanEtten, J. F. Sebastian, G. A. Clowes, M. L. Bender, J. Am. Chem.
Soc. 1967, 89, 3242–3253.
1
[D8]1,4-dioxane¼ 18%. All chemical shifts for H are relative to
1
TMS using H (residual) chemical shifts of [D8]1,4-dioxane as a
[35] S. A. Ba-Saif, A. B. Maude, A. J. Williams, J. Chem. Soc. Perkin Trans
1994, 2395–2400.
secondary standard.
´
[36] M. Fenandez, R. H. de Rossi, J. Org. Chem. 1999, 64, 6000–6004.
[37] G. O. Andres, A. B. Pierini, R. H. de Rossi, J. Org. Chem. 2006, 71,
´
7650–7656.
[38] S. A. Ba-Saif, M. A. Waring, A. Williams, J. Am. Chem. Soc. 1990, 112,
8115–8120.
Acknowledgements
This research was supported in part by Consejo Nacional de
[39] K. Sorensen-Stowel, A. C. Hengge, J. Org. Chem. 2006, 71, 7180–7184.
[40] I.-H. Um, E.-J. Lee, S.-E. Jeon, Bull. Kor. Chem. Soc. 2001, 22, 1301–1302.
[41] N. D. Gillitt, J. Domingos, C. A. Bunton, J. Phys. Org. Chem. 2003, 16,
603–607.
[42] J. Epstein, M. M. Demek, D. H. Rosenblatt, J. Org. Chem. 1956, 21, 796–
797.
´
Investigaciones Cient´ıficas y Tecnicas, Argentina (CONICET),
Agencia Nacional de Promocion Cient´ıfica y Tecnologica (FON-
´
´
CYT), and Secretar´ıa de Ciencia y Tecnolog´ıa, Universidad Nacio-
´
nal de Cordoba, Argentina. R.V.V. is a CONICET fellow. R. H. de R.
and E. I. B. are Career Investigators of CONICET. a-, b-, and
g-cyclodextrins were kindly delivered by Cerestar.
[43] J. B. Domingos, E. Longhinotti, T. A. S. Brandao, L. S. Santos, M. N.
Eberlin, C. A. Bunton, F. Nome, J. Org. Chem. 2004, 69, 7898–
7905.
[44] J. March, Advanced Organic Chemistry, 4th Ed. John Wiley & Sons,
New York, U. S. A., 1999 p. 251.
[45] D. Herschlag, W. P. Jenks, J. Am. Chem. Soc. 1990, 112, 1951–1956.
[46] R. M. Tarkka, E. Buncel, J. Am. Chem. Soc. 1995, 117, 1503–1507.
[47] S. Ishiwata, M. Kamiya, Chemosphere 1999, 39, 1595–1600.
[48] M. Kamiya, S. Mitsuhashi, M. Makino, Chemosphere 1992, 25,
1783–1796.
REFERENCES
[1] Y.-C. Yang, Acc. Chem. Res. 1999, 32, 109–115.
[2] Y.-C. Yang, J. A. Baker, J. R. Ward, Chem. Rev. 1992, 92, 1729–1743.
[3] H. J. Cristau, E. Torreilles, J. F. Ginieys, J. Chem. Soc. Perkin Trans 1991,
13–16.
[4] C. A. Bunton, N. D. Gillitt, A. Kumar, J. Phys. Org. Chem. 1997, 10,
221–228.
[5] F. M. Menger, M. J. Rourk, Langmuir, 1999, 15, 309–313.
´
´
[49] L. Garcia-Rio, P. Herves, J. R. Leis, J. C. Mejuto, J. Perez-Juste, P.
Rodriguez-Dafonte, Org. Biomol. Chem. 2006, 4, 1038–1048.
[50] J. H. Park, T. H. Nah, J. Chem. Soc. Perkin Trans 1994, 1359–1362.
[51] Although the association equilibrium constant for the substrate
could in principle be corrected using the known value of the
association constants of the solvents, we have not done that because
the possibility of formation of ternary complexes CD þ ACN þ
Substrate can not be discarded, therefore we prefer to deal with
the apparent association constant experimentally determined. If
there is no ternary complex formation, the observed inhibition is
smaller than the real inhibition because the amount of the free CD
available for complexation with the substrate is smaller than the
analytical concentration. See L. Garcia-Rio, J. C. Mejuto, M. Nieto, J.
Perez-Juste, M. Perez-Lorenzo, P. Rodriguez-Dafonte, Supramol.
Chem. 2006, 17, 649–653.
ˇ
[6] M. Bavcon Kralj, M. Franko, P. Trebse, Chemosphere 2007, 67, 99–107.
[7] D. L. Macalady, N. L. Wolfe, J. Agric. Food Chem. 1983, 31, 1139–1147.
[8] C. Walling, Acc. Chem. Res. 1975, 8, 125–131.
[9] K. H. Chan, W. Chu, Chemosphere 2003, 51, 305–311.
[10] A. S. Derbalah, N. Nakatani, H. Sakugawa, Chemosphere 2004, 57,
635–644.
[11] M. I. Badawy, M. Y. Ghaly, T. A. Gad-Allah, Desalination 2006, 194,
166–175.
[12] Report on FQPA Tolerance Reassessment Progress and Risk Manage-
ment Decision for Chlorpyrifos-methyl. http://www.epa.gov/pesti-
cides/reregistration/REDs/cpm_tred.pdf. Accessed June 26, 2008.
[13] M. L. Bender, M. Komiyama, Cyclodextrin Chemistry, Springer, New
York, 1976.
[14] I. Tabushi, Y. Kuroda, Adv. Catal. 1983, 32, 417–466.
[15] I. Tabushi, Y. Kuroda, Chem. Rev. 1998, 98, (5), thematic issue.
[16] H. Yu, Y. Makino, M. Fukudome, R. G. Xie, D. Q. Yuan, K. Fujita,
Tetrahedron Lett. 2007, 48, 3267–3271.
[17] H. L. Chen, B. Zhao, Z. B. Wang, J. Incl. Phenom. Macrocycl. Chem.
2006, 56, 17–21.
[18] L. G. Marinescu, M. Bols, Angew. Chem. Int. Ed. 2006, 45, 4590–
4593.
´
´
[52] T. Loftsson, A. Magnu´sdottir, M. Masson, J. Sigurjonsdottir, J. Pharm.
´
´
Sci. 2002, 91, 2307–2316.
[53] R. A. Moss, P. K. Gong, Langmuir 2000, 16, 8551–8554.
[54] N. Hennrich, F. Cramer, J. Am. Chem. Soc. 1965, 87, 1121–1126.
[55] K. A. Connors, Binding Constants. The Measurement of Molecular
Complex Stability, John Wiley & Sons, U.S.A, 1987.
[56] M. Kawamura, M. Higashi, J. Incl. Phenom. Macrocycl. Chem. 2005, 51,
11–15.
[57] H. Shimizu, A. Kaito, M. Hatano, Bull. Chem. Soc. Jpn. 1979, 52,
2678–2684.
[58] H. Shimizu, A. Kaito, M. Hatano, Bull. Chem. Soc. Jpn. 1981, 54,
513–519.
[19] C. Rousseau, B. Christensen, M. Bols, Eur. J. Org. Chem. 2005,
2734–2739.
[20] A. R. Hedges, Chem. Rev. 1998, 98, 2035–2044.
[59] M. Kamiya, S. Mitsuhashi, M. Makino, H. Yoshioka, J. Phys. Chem. 1992,
96, 95–99.
´
[21] G. O. Andres, O. F. Silva, R. H. de Rossi, Can. J. Chem. 2005, 83,
1281–1296.
[60] S. Hamai, T. Ikeda, A. Nakamura, H. Ikeda, A. Ueno, F. Toda, J. Am.
Chem. Soc. 1992, 114, 6012–6016.
´
[22] M. A. Fernandez, R. H. de Rossi, J. Org. Chem. 2003, 68, 6887–
[61] The distances were calculated by optimizing the geometry of the
insecticide by Hyper Chem 7.5, Evaluation version 7.5 by Hyperab,
Inc. U.S.A.
[62] H.-J. Schneider, F. Hacket, V. Ru¨diger, Chem. Rev. 1998, 98, 1755–1785.
[63] L. Fielding, Tetrahedron 2000, 56, 6154–6170.
[64] D. Churchill, J. C. F. Cheung, Y. H. Park, V. H. Smith, G. vanLoon, E.
Buncel, Can. J. Chem. 2006, 84, 702–708.
6893.
[23] M. A. Fernandez, R. H. de Rossi, E. Cervello, C. Jaime, J. Org. Chem.
´
2001, 66, 4399–4404.
[24] A. M. Granados, R. H. de Rossi, J. Org. Chem. 2001, 66, 1548–1552.
´
[25] R. V. Vico, E. I. Bujan, R. H. de Rossi, J. Phys. Org. Chem. 2002, 15,
858–862.
[26] B. Liu, L. L. McConnell, A. Torrents, Chemosphere 2001, 44, 1315–1323.
[27] T. Wu, U. Jans, Environ. Sci. Technol. 2006, 40, 784–790.
[28] V. K. Balakrishnan, J. M. Dust, G. W. vanLoon, E. Buncel, Can. J. Chem.
2001, 79, 157–173.
[29] X. Han, V. K. Balakrishnan, G. W. vanLoon, E. Buncel, Langmuir 2006,
22, 9009–9017.
[65] A. Calderini, F. B. T. Pessine, J. Incl. Phenom Macrocycl. Chem. 2008, 60,
369–377.
[66] J. Szejtli, Chem. Rev. 1998, 98, 1743–1753.
[67] L. E. Viola, R. H. de Rossi, Can. J. Chem. 1999, 77, 860–867.
[68] R. E. Galian, A. V. Veglia, J. Phothochem. Photobiol. A 2007, 187,
356–362.
[30] M. T. Moore, R. Schulz, C. M. Cooper, S. Smith, J. H. Rodgers, Chemo-
sphere 2002, 46, 827–835.
[69] K. Mochida, Y. Matsui, Y. Ota, K. Arakawa, Y. Date, Bull. Chem. Soc. Jpn.
1976, 49, 3119–3123.
J. Phys. Org. Chem. 2009, 22 691–702
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
R. V. VICO, R. H. DE ROSSI AND E. I. BUJAN
[70] Calculated with Chem Draw Ultra, version 8.0 (2003). Cambridge Soft
Corporation, U.S.A.
orientation of the substrate in the complex and this does not change
with the nucleophile.
[71] Although we did not determine the rate of the reaction of H2O2 at
different pH in the presence of cyclodextrin, it is very unlikely that
the k2 pathway is significant with this nucleophile while it is not with
water. The rate of this pathway is very much dependent on the
[72] M. Kamiya, K. Nakamura, Pestic. Sci. 1994, 41, 305–309.
[73] M. Kamiya, K. Nakamura, C. Sasaki, Chemosphere 1995, 30, 653–660.
[74] R. H. de Rossi, E. B. de Vargas, J. Am. Chem. Soc. 1981, 103, 1533–
1540.
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 691–702