Effects of β-Cyclodextrin on the Keto-Enol Equilibrium of Benzoylacetone and on Enol Reactivity

Article abstract of DOI:10.1021/jo982407g

Both β-cyclodextrin and sodium dodecyl sulfate micelles shift the benzoylacetone keto-enol equilibrium to the enol tautomer by preferentially binding the enol form. The UV-vis spectroscopic method was used to quantify the temperature and solvent effects on the keto-enol equilibrium of benzoylacetone in aqueous acid medium. The comparison between the thermodynamic parameters resulting from the binding of the benzoylacetone enol to sodium dodecyl sulfate micelles and from the inclusion of both keto and enol tautomers into the β-cyclodextrin cavity allows us to draw a picture of the possible complex formed in each case. ¹H NMR results suggest that benzoylacetone-enol protrudes deeper inside the β-cyclodextrin cavity, whereas the keto tautomer could only have the phenyl ring enclosed in the β-cyclodextrin cavity interior. Nitrosation in acid medium of benzoylacetone in the presence of β-cyclodextrin is reduced below that of free benzoylacetone, indicating that the cyclodextrin complex protects the benzoylacetone enol tautomer, which is in perfect accordance with our picture of the enol·β-cyclodextrin complex.

Full text of DOI:10.1021/jo982407g

3954
J . Org. Chem. 1999, 64, 3954-3963
Effects of â-Cyclod extr in on th e Keto-En ol Equ ilibr iu m of
Ben zoyla ceton e a n d on En ol Rea ctivity
Emilia Iglesias* and Vicente Ojea-Cao
Departamento de Quı´mica Fundamental e Industrial, Facultad de Ciencias, Universidad de La Corun˜a,
15071 La Corun˜a, Spain
Luis Garc´ıa-Rio and J . Ramo´n Leis
Departamento de Quı´mica Fı´sica, Facultad de Quı´mica, Universidad de Santiago de Compostela,
15706 Santiago de Compostela, Spain
Received December 9, 1998
Both â-cyclodextrin and sodium dodecyl sulfate micelles shift the benzoylacetone keto-enol
equilibrium to the enol tautomer by preferentially binding the enol form. The UV-vis spectroscopic
method was used to quantify the temperature and solvent effects on the keto-enol equilibrium of
benzoylacetone in aqueous acid medium. The comparison between the thermodynamic parameters
resulting from the binding of the benzoylacetone enol to sodium dodecyl sulfate micelles and from
the inclusion of both keto and enol tautomers into the â-cyclodextrin cavity allows us to draw a
picture of the possible complex formed in each case. 1H NMR results suggest that benzoylacetone-
enol protrudes deeper inside the â-cyclodextrin cavity, whereas the keto tautomer could only have
the phenyl ring enclosed in the â-cyclodextrin cavity interior. Nitrosation in acid medium of
benzoylacetone in the presence of â-cyclodextrin is reduced below that of free benzoylacetone,
indicating that the cyclodextrin complex protects the benzoylacetone enol tautomer, which is in
perfect accordance with our picture of the enol‚â-cyclodextrin complex.
In tr od u ction
water.11,12 As a consequence of the perturbation of the
keto-enol equilibrium of benzoylacetone (BZA) brought
on by the presence of micelles, the nitrosation reaction
of BZA in aqueous acid micellar solutions is greatly
modified.13 In the present study, we investigate the
effects of â-cyclodextrin (â-CD) on the keto-enol equi-
librium of BZA and dibenzoylmethane (DBM) and on the
nitrosation reaction of the enol of BZA, DBM, and
acetylacetone (AcAc) in aqueous acid medium. To arrive
at a quantitative explanation of the results, we first
studied the influence of temperature and solvents on the
keto-enol equilibrium of BZA. For that, the absorption
spectra of BZA were recorded as a function of the
surfactant sodium dodecyl sulfate (SDS) concentration
at different temperatures and at two fixed percentages
of dioxane in the aqueous surfactant solutions. Thermo-
dynamic parameters corresponding to the association
process of the enol of BZA to the SDS micelles were
obtained and compared with those found in studying the
influence of temperature on the complexation processes
between both keto and enol tautomers of BZA and â-CD.
Cyclodextrins, doughnut-shaped molecules, are typical
host compounds made up of six to eight glucose units
linked together covalently by oxygen atoms and held in
shape by means of hydrogen bonding between the sec-
ondary hydroxy groups on adjacent units at the wider
rim of the cavity.1-5 An array of properties seems to make
cyclodextrin a reasonable choice for an enzyme model,6,7
including its water solubility, the fact that the guest is
bounded reversibly in the cavity, and the fact that a
number of equilibrium8 or chemical9,10 reactions may be
modified by the addition of cyclodextrins.
In previous works, we have shown that the study of
the influence of micellar solutions on the absorption
spectrum of â-dicarbonyl compounds, like benzoylacetone,
can provide a new method for determining keto-enol
equilibrium constants of â-dicarbonyl compounds in
(1) Saenger, W. Angew. Chem., Int. Ed. Engl. 1980, 19, 344.
(2) Szejtli, J . Cylodextrin Technology; Kluwer: Dordrecht, Nether-
lands, 1988.
(3) Tee, O. S. Avd. Phys. Org. Chem. 1994, 29, 1.
(4) (i) Liu, Y.; Zhang, Y.-M.; Chen, R.-T.; Yamamoto, K.; Wada, T.;
Inoue, Y. J . Org. Chem. 1997, 62, 1826. (ii) Matsuhita, A.; Kuwabara,
T.; Nakamura, A.; Ikeda, H.; Ueno, A. J . Chem. Soc., Perkin Trans. 2
1997, 1705.
(5) Haskard, C. A.; May, B. L.; Kurucsev, T.; Lincoln, S. F.; Easton,
C. J . J . Chem. Soc., Faraday Trans. 1997, 93, 279.
(6) Tabushi, I. Acc. Chem. Res. 1982, 15, 66.
(7) D’Souza, V. T.; Bender, M. L. Acc. Chem. Res. 1987, 20, 146.
(8) (i) Park, J . W.; Choi, N. H.; Kim, J . H. J . Phys. Chem. 1996,
100, 769. (ii) de Rossi, R. H.; Sa´nchez, A. M. J . Org. Chem. 1996, 31,
3446. (iii) Davies, D. M.; Savage, J . R. J . Chem. Soc., Perkin Trans. 2
1994, 1525.
(9) (i) Tee, O. S.; Fedortchenko, A. A.; Soo, P. L. J . Chem. Soc., Perkin
Trans. 2 1998, 123. (ii) Tee, O. S.; Gadosy, T. A. Can. J . Chem. 1996,
74, 745. (iii) Tee, O. S.; Du, X. J . Am. Chem. Soc. 1992, 114, 620.
(10) (i) Granados, A.; de Rossi, R. H. J . Am. Chem. Soc. 1995, 117,
3690. (ii) Evans, C. H.; Feyter, S. D.; Viaene, L.; van Stam, J .; Schryver,
F. C. D. J . Phys. Chem. 1996, 100, 2129.
Exp er im en ta l Section
Ma ter ia ls. Ketones and â-cyclodextrin, Aldrich products of
maximum purity, sodium dodecyl sulfate, a Sigma product,
and D2O (Solvents, Documentation, Synthe`ses Laboratories)
were used without further purification. All other reagents were
supplied by Merck and were used as received. Solutions were
prepared with doubly distilled water obtained from a perman-
ganate solution.
Meth od s. BZA, DBM, and AcAc were dissolved in dioxane
(spectrophotometric grade). From this stock solution, the
(11) Iglesias, E. J . Phys. Chem. 1996, 100, 12592.
(12) Iglesias, E. J . Chem. Soc., Perkin Trans. 2 1997, 431.
(13) Iglesias, E. Langmuir 1998, 14, 5764.
10.1021/jo982407g CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/29/1999

Keto-Enol Equilibrium of Benzoylacetone
J . Org. Chem., Vol. 64, No. 11, 1999 3955
working aqueous solution was prepared daily by diluting 0.2-
0.4 mL to 25 mL final volume. All other solutions were
prepared in water by directly dissolving the appropriate weight
of the desired product in the required water volume.
UV-vis absorption spectra and kinetic measurements were
recorded with a Kontron-Uvikon (model 942) double-beam
spectrophotometer provided with a thermostated multiple cell
holder. The procedure used for obtaining the UV-vis spectra
of BZA in the presence of SDS, to determine the keto-enol
equilibrium constant in water KE and the association constant
E
Ks of the BZA-enol to SDS micelles, has previously been
reported.11,12 The absorption spectra in the presence of â-CD
were recorded following the same procedure, and equal â-CD
concentrations were placed in both the cell sample and
reference.
1H NMR spectra were obtained with a 200 MHz spectrom-
eter (Bruker AC200F). Chemical shifts were measured with
respect to the residual water signal at δ ) 4.60 ppm. For these
experiments, solutions containing 0.8 mM of BZA were pre-
pared in D2O. A small amount of the BZA methanolic stock
solution was placed in a volumetric flask, and methanol was
evaporated with a gentle stream of Ar followed by the addition
of D2O and/or the desired volume of a â-CD solution in D2O.
All solutions were stirred overnight.
Kinetic studies were performed under pseudo-first-order
conditions, with [nitrite] at least 20 times greater than
[ketone]. All reactions were observed by recording the decrease
in absorbance due to enol consumption at λ ) 312 nm for BZA,
at 345 nm for DBM, and at 245 nm for AcAc.12 Every kinetic
experiment was started by adding, to the rest of the reaction
mixture, 0.10-0.20 mL of NaNO2 aqueous solution, previously
thermostated at the required temperature. In every case the
integrated method was followed, fitting the experimental
absorbance-time data to the first-order integrated equation.
Satisfactory correlation coefficients (>0.999) and residuals
were attained in every case. All kinetic experiments were
carried out at 25 °C.
F igu r e 1. Absorption spectra of BZA obtained (a) in water-
dioxane mixtures of (1) 0.2%, (2) 10%, (3) 20%, (4) 30%, (5)
40%, (6) 50%, (7) 60%, (8) 70% v/v of dioxane ([BZA] ) 7.0 ×
10-5 mol dm-3) and (b) absorption spectra of BZA (8.0 × 10-5
mol dm-3) obtained in the presence of [HCl] ) 0.033 mol dm-3
Resu lts a n d Discu ssion
as a function of SDS concentration at (1) 0.0, (2) 8.8 × 10-3
,
,
(3) 1.1 × 10-2, (4) 1.32 × 10-2, (5) 1.76 × 10-2, (6) 2.8 × 10-2
(i) Tem p er a tu r e a n d Solven t Effects on Keto-
En ol Equ ilibr iu m . The BZA keto-enol equilibrium was
studied in water-dioxane mixtures (an homogeneous
solvent whose physical properties change with the mix-
ture composition), in aqueous micellar solutions of SDS
(i.e., in a microheterogeneous medium, in which the
existence of two main regions of quite different proper-
ties, such as the bulk water pseudophase and the micellar
pseudophase, may be remarked), and in aqueous â-cy-
clodextrin solutions.
Figure 1 displays the effect of water-dioxane (ꢀ ) 2.2)14
mixtures and SDS aqueous micellar solutions on the UV
spectrum of benzoylacetone. The amount of the enol
tautomer increases as either the dioxane percentage
(Figure 1a) or the surfactant SDS concentration (Figure
1b) is increased. In dioxane, poorly defined isosbestic
points are obtained and the maximum wavelength ab-
sorption due to the enol tautomer (λ ≈ 312 nm) shifts to
lower wavelength as the solvent polarity decreases (a
common observation in π f π* transitions).15 By contrast,
in the presence of SDS micellar solutions, if we assume
that the keto form resides in the bulk water phase and
the enol form is solubilized in the Stern layer of the SDS
micelles (a very hydrated region of dielectric constants
or relative permittivity16sestimated between 46 and 56),
(7) 4.4 × 10-2, (8) 6.6 × 10-2, (9) 0.132, (10) 0.28 mol dm-3
.
no change in the microenvironment of both absorbing
species will occur on increasing the surfactant concentra-
tion. Consequently, two well-defined isosbestic points are
drawn on varying the [SDS], which also means that the
enolization constant, KE ) [enol]/[keto], remains inde-
pendent of the parameter being varied.17
The influence of [SDS] on the BZA spectrum was
studied at several temperatures and in the presence of
two constant percentages of dioxane in the aqueous SDS
micellar solutions. We followed the recently proposed
method11,12 to analyze the experimental results. Hence,
the variation of the absorbance of the solution at 312 nm
or at 250 nm, the two maximum absorption bands, is
related to the micellized surfactant concentration ([SDS]m
) [SDS]t - cmc, with cmc being the critical micelle
concentration calculated from the same experiment18) to
E
obtain the unknown parameters KE and Ks ; the corre-
sponding equilibrium processes are included in Scheme
1 for comparison purposes with the case of â-cyclodextrin.
E
Both KE and Ks can be determined by nonlinear (or
linear) regression analysis of the experimental data Aλ
- [SDS]m (or to the linearized equation resulting from
the dependence of A312 with [SDS]m). The obtained
(14) Anderson, J . E. J . Phys. Chem. 1991, 95, 7062.
(15) Reichardt, C. Solvents and Solvent Effect in Organic Chemistry,
2nd ed.; VCH Verlagsgesellschaft: Weinheim, 1988; Chapter 6.
(16) Drummond, C. J .; Grieser, F.; Healy, T. W. J . Phys. Chem. 1988,
92, 2604 and references therein.
(17) Drago, R. S. Physical Methods for Chemists, 2nd ed.; Saunders
Coll. Publ.: Orlando, FL, 1992; Chapter 4.
(18) Dom´ınguez, A.; Ferna´ndez, A.; Iglesias, E.; Montenegro, L. J .
Chem. Educ. 1997, 74, 1227.

3956 J . Org. Chem., Vol. 64, No. 11, 1999
Iglesias et al.
Sch em e 1
Ta ble 1. Exp er im en ta l Con d ition s a n d Th er m od yn a m ic P a r a m eter s Obta in ed by Stu d yin g th e In flu en ce of [SDS] on
th e Absor p tion Sp ectr a of BZA
data of A312 - [SDS]m fitted to A312 ) [A°312(1 + KsE[SDS]m)]/[1 + ((KEKsE)/(1 + KE))[SDS]m]
mediuma
t/°C
cmc/mol dm-3
A°312
KsE/mol-1 dm3
KE/(1 + KE)
KE
cc
SDS
SDS
SDS
16.9
25.0
32.0
41.2
25.0
25.0
2.5 × 10-3
2.5 × 10-3
2.0 × 10-3
2.1 × 10-3
4.0 × 10-3
5.0 × 10-3
0.4001
0.362
0.3568
0.3339
0.4762
0.5670
284 ( 8
235 ( 7
191 ( 6
161 ( 5
85.5 ( 2.5
40 ( 2
0.422 ( 0.002
0.390 ( 0.001
0.389 ( 0.002
0.383 ( 0.002
0.481 ( 0.002
0.562 ( 0.004
0.73
0.64
0.63
0.62
0.93
1.28
0.9992
0.9997
0.9994
0.9994
0.9994
0.9994
SDS
SDS/10%b
SDS/20%b
SDS/2%c
25.0
1.0 × 10-3
0.100d
1100
0.39
0.68
0.99
Data of A250 - [SDS]m fitted to A250 ) [A°250 + (A∞250KEKsE/(1 + KE))[SDS]m]/[1 + ((KEKsE)/(1 + KE))[SDS]m]
mediuma
t/°C
cmc
A°250
A∞
KsEKE/(1 + KE)
KE
cc
250
SDS
SDS
SDS
16.9
25.0
32.0
41.2
25.0
25.0
2.5 × 10-3
2.5 × 10-3
2.0 × 10-3
3.0 × 10-3
3.5 × 10-3
5.0 × 10-3
0.7443
0.7340
0.7709
0.7638
0.7085
0.6638
0.440 ( 0.002
0.451 ( 0.002
0.473 ( 0.002
0.485 ( 0.002
0.448 ( 0.003
0.440 ( 0.005
110 ( 3
99 ( 3
74.5 ( 3
62 (2
0.63
0.73
0.64
0.63
1.01
1.33
0.9993
0.9995
0.9991
0.9994
0.999
SDS
SDS/10%b
SDS/20%b
43 (2
23 (1
0.999
SDS/2%c
25.0
1.0 × 10-3
0.238e
0.077e
475
0.75
0.99
a
b
Aqueous solutions of SDS in the presence of [HCl] ) 0.033 mol‚dm-3
.
Refers to the percentage of dioxane in the aqueous SDS
solutions. c Dibenzoylmethane at 1.2 × 10-5 mol‚dm-3
.
At 345 nm. e At 254 nm.
d
parameters are independent of the analysis procedure
used and are listed in Table 1.
fies the solubility of hydrophobic molecules, like BZA-
E
enol, and consequently Ks decreases strongly (from 235
mol-1 dm3 in water, to 40 mol-1 dm3 in 20% v/v dioxane-
water). In the same sense, as the structure of water is
more ordered at low temperatures (reaching maximum
The KE values do not appear to be temperature
dependent (at least, in the range studied), which is
indicative of the small value of the enthalpy (∆H°)
corresponding to the enolization equilibrium in water.19
Also, one can see that a decrease in the dielectric constant
makes the enol content greater; e.g., values of KE are 0.63
in water, 0.95 in 10% v/v dioxane-water, and 1.3 in 20%
v/v dioxane-water. Most studies20 concerning the deter-
mination of ∆H° and ∆S° for â-dicarbonyl compound
enolization refer to acetylacetone, ethylacetoacetate, and
dimethylacetoamide. Almost all studies have been carried
out in apolar solvents and have employed the NMR
method. Inspection of the data reveal important conclu-
sions: the enol formation is always favored enthalpically
and disfavored entropically and ∆G° differences between
apolar solvents and water are mainly due to the more
negative ∆S° values in the latter.
E
at 4 °C), then Ks values diminish on increasing temper-
ature (e.g., from 284 mol-1 dm3 at 17 °C to 161 mol-1 dm3
at 42 °C).
E
The van’t Hoff plot corresponding to Ks draws a good
straight line, yielding the values of ∆S° ) -16.3 J ‚mol-1
K-1 and ∆H° ) -18.3 kJ ‚mol-1. van der Waals interac-
tions between the enol and the hydrocarbon chains of the
surfactant monomers in the micelle account for the
favorable enthalpic contribution, which is also due, in a
great part, to the gain in the number of intramolecular
hydrogen bonds with the enol tautomer in the micellar
phase. Because of resonance with the phenyl ring, these
intramolecular interactions can be of types O‚‚‚H-O or
-O‚‚‚H-O, whose energy ranges between (50-150) kJ
mol-1 21
Despite the increase in entropy accompanying
.
E
In regard to the values of Ks , the addition of dioxane
the enol departure of the water phase, a greater loss is
associated with enol solubilization in the micelle; how-
ever, the unfavorable entropic factor is less significant
than the enthalpic one, which causes the association
process to be energetically favorable.
to water reduces the solvent organization, which simpli-
(19) Toullec, J . In The Chemistry of Enols; Rappoport, Z., Ed.; J ohn
Wiley & Sons: Chichester, England, 1990; p 323.
(20) For example, in the case of enolization of acetylacetone in water
∆H° ) -11.29 kJ /mol and ∆S° ) -51.8 J /mol‚K, and then KE varies
from 0.21 to 0.16 in the temperature range of 15-40 °C. (Spencer, J .
N.; Holmboe, E. S.; Kirshenbaum, M. R.; Firth, D. W.; Pinto, P. Can.
J . Chem. 1982, 60, 1178.
(21) J effrey, G. A. An Introduction to Hydrogen Bonding; Oxford
University Press: 1997.

Keto-Enol Equilibrium of Benzoylacetone
J . Org. Chem., Vol. 64, No. 11, 1999 3957
E
E
A°312(1 + Kc [â-CD])
KcK + Kc KE
A312
)
with Z )
1 + KE
1 + Z[â-CD]
(1)
A°250 + A∞250Z[â-CD]
1 + Z[â-CD]
A250
)
with
E
KcK + Kc KE
Z )
(2)
1 + KE
Fitting the experimental data of Aλ - [â-CD] to eqs 1
or 2, the parameters reported in Table 2 were obtained.
The results obtained in the study of the influence of â-CD
on the absorption spectrum of dibenzoylmethane are also
included. The UV-absorption spectrum of acetylacetone
does not show appreciable changes by the presence of
â-CD.
If one assumes that the inclusion complexation of the
keto form with â-CD is negligible (in which case Z )
KcEKE /(1 + KE), as in the case of SDS micellar solutions),
the model also fits quite well to the experimental data,
but the resulting KE values (≈ 0.9) are much higher than
those expected (KE ) 0.62) and reported in Table 1 or
determined with other methods.22 As one cannot accept
that the presence of â-CD could modify the value of KE,
because the properties of the bulk water solvent remain
unchanged by the presence of â-CD, this apparent
discrepancy can be solved by considering as well as the
formation of an inclusion complex between the keto
K
tautomer and CD: KH‚CD, with Kc being the stability
constant (Scheme 1b). Such an alternative could also
explain the appearance of two poorly defined isosbestic
points in the BZA spectra as a function of [â-CD]; thus,
the microenvironments of both absorbing species are
changing.
Values of KcK were determined from Z {)(KcK + KcEKE)/
(1 + KE)} by using known values of KcE and KE. The latter
was obtained from data in Table 1 (and results already
published11) as the average value of KE ) 0.62, not
temperature dependent.
F igu r e 2. (a) Absorption spectra of BZA (7.0 × 10-5 mol dm-3
)
in the presence of [HCl] ) 0.033 mol dm-3 as a function of
â-CD concentration at (1) 0.0, (2) 0.36, (3) 0.72, (4) 1.08, (5)
1.44, (6) 2.16, (7) 2.89, (8) 4.31, (9) 5.39, (10) 6.47 mmol dm-3
(b) influence of â-CD concentration on the absorbance readings
at (b) 10 °C and at (2) 36 °C; lines fit eqs 1 (ascending curve)
and 2 (descending curve).
;
Figure 2 exhibits the effect of â-CD concentration on
the absorption spectra of BZA. As in the case of SDS
aqueous micellar solutions, addition of cyclodextrin shifts
the benzoylacetone keto-enol equilibrium to the enol
tautomer. Nevertheless, unlike with SDS, the isosbestic
points are not very well-defined, as occurring in the case
of water-dioxane mixtures. Remembering that the enol
tautomer is stabilized by intramolecular hydrogen bond-
ing, the increases in absorbance at 312 nm with [â-CD]
may be the consequence of the formation of an inclusion
compound between the enol and â-CD of 1:1 stoichiom-
etry. Since, in aqueous solutions, intramolecular hydro-
gen bonding is widely replaced by intermolecular inter-
actions with the solvent, the BZA-enol is more extensively
stabilized in the hydrophobic cavity interior of CD. But
on first sight the keto tautomer could also form inclusion
complexes with â-CD (vide infra), as stated in Scheme
1b, where KH, EH, KH‚CD, and EH‚CD stand for the
keto molecules, enol, keto‚â-CD complex, and the enol‚
â-CD complex, respectively.
The results reported in Table 2 indicate that BZA-
enol forms an inclusion complex more stable than that
of the keto tautomer. Taking into account that ∆G )
-RT‚ln K ) ∆H - T∆S, the study of the influence of
E
K
temperature on Kc and Kc allows us to obtain the
thermodynamic parameters of the corresponding equi-
E
K
librium processes. The plots of ln Kc (or ln Kc ) against
1/T are good straight lines, and from the intercept and
slope values the following thermodynamic parameters
can be obtained: (i) for enol inclusion ∆SEH ) -1.4 J /mol‚
K and ∆HEH ) -14.7 kJ /mol; (ii) for keto inclusion ∆SKH
) -82.5 J /mol‚K; ∆HKH ) -33.7 kJ /mol.
The geometries of both tautomers of BZA were opti-
mized,23 and while the enol is a planar molecule, the keto
form is not. The vertical distance between H14 and H16 is
4.35 Å, whereas the horizontal distances between H15 and
O11, in the enol, and between H15 and C10, in the keto,
are 8.30 and 8.65 Å, respectively. Considering the shape
and dimensions of â-CD1,24,25 (Scheme 2), the only way
Following the same procedure as for that of micellar
solutions,11 starting from Scheme 1b and taking into
account that BZA concentration () 7.0 × 10-5 mol‚dm-3
)
(22) Bunting, J . W.; Kanter, J . P.; Nelander, R.; Wu, Z. Can. J .
Chem. 1995, 73, 1305.
(23) PCMODEL, ver. 6.0, Serena Software, 1996.
(24) Ramamurthy, V. In Photochemistry in Organized & Constrained
Media; Ramamurthy, V., Ed.; VCH Publishers: New York, 1991; p 319.
(25) Li, S.; Purdy, W. C. Chem. Rev. 1992, 92, 1457.
is much smaller than [â-CD] (i.e., [â-CD]t ) [KH‚CD] +
[EH‚CD] + [â-CD] = [â-CD]) one can easily arrive to eqs
1 and 2 to relate the absorbance variation at two
maximum absorption bands with [â-CD].

3958 J . Org. Chem., Vol. 64, No. 11, 1999
Iglesias et al.
Ta ble 2. Th er m od yn a m ic P a r a m eter s Obta in ed in Wa ter by Stu d yin g th e In flu en ce of â-Cyclod extr in Con cen tr a tion on
th e Absor p tion Sp ectr a of BZA a t 6.7 × 10-5 m ol‚d m -3 a n d in th e P r esen ce of [HCl] ) 0.033 m ol‚d m -3
data of A312 - [â-CD] fitted to eq 1
b
t/°C
A°312
KcE/mol-1 dm3
Z/mol-1 dm3
KE
KE
KcK/mol-1 dm3
10.0
15.3
20.3
25.0a
31.1
36.0
0.5252
0.4915
0.5111
0.3865
0.4658
0.4550
431 ( 17
387 ( 16
348 ( 10
314 ( 9
283 ( 16
254 ( 8
215 ( 11
186 ( 10
162 ( 6
145 ( 6
126 ( 6
112 ( 5
0.995
0.889
0.889
0.896
0.802
0.776
0.62
0.62
0.62
81.1
61.4
46.7
40.2
28.7
24.0
0.62-0.7222
0.62
0.62
25.0c
0.273d
445 ( 26
274 ( 19
1.60
0.65
163
ln(KcE) ) A + B(1/T); A ) (-0.185 ( 0.096); B ) (1.77 ( 0.03) × 103; r ) 0.9995
data of A250 - [â-CD] fitted to eq 2
b
t/°C
A°250
A∞
Z/mol-1 dm3
KE
KE
KcK/mol-1 dm3
250
10.0
15.3
20.3
25.0a
31.1
36.0
0.9498
0.9343
1.0176
0.7593
1.0134
1.0260
0.538 ( 0.012
0.536 ( 0.003
0.588 ( 0.005
0.421 ( 0.003
0.579 ( 0.006
0.533 ( 0.008
214 ( 13
186 ( 12
164 ( 10
144 ( 8
126 ( 14
113 ( 13
0.986
0.935
0.851
0.847
0.802
0.801
0.62
0.62
0.62
0.62
0.62
0.62
79.5
53.0
48.3
38.6
28.7
25.5
ln(KcK) ) A + B(1/T); A ) (-8.4 ( 0.3); B ) (4.05 ( 0.10) × 103; r ) 0.999
a
b
At [BZA] ) 5.0 × 10-5 mol dm-3
.
Calculated values by assuming Z ) KcEKE/(1 + KE). c Dibenzoylmethane 2.3 × 10-5 mol dm-3 in
d
2% dioxane. At 345 nm.
Sch em e 2
both keto and enol tautomers can enter the â-CD cavity
is lengthwise, and in no circumstances could both BZA
tautomers be encapsulated completely in the â-CD cavity.
In light of thermodynamic results and molecular
modeling calculations, one could say that the enol form
is a molecule capable of deeply protruding into the CD
cavity. The enol adopts a cyclic form stabilized by
intramolecular hydrogen bonds.Then, the interactions
between the enol and CD are van der Waals forces. The
practically insignificant entropic factor in the enol inclu-
sion can be accounted for in the following way.
The keto tautomer is not a planar molecule: the phenyl
ring is in a different plane from that of the carbonyl
groups. Consequently, the degree of penetration of the
keto tautomer in the CD cavity is expected to be less than
that of the enol tautomer. In addition, the carbonyl
groups need to be stabilized by hydrogen-bond interac-
tions by means of either secondary hydroxyl groups of
â-CD or water molecules. At this point, one reasonable
possibility is the formation of a complex in which only
the phenyl ring was inside the â-CD cavity with the
carbonyl group close to the phenyl ring forming hydrogen
bonds with the OH groups located on the wider rim and
with the other carbonyl group outside the â-CD cavity
solvated by water molecules. This should mean that the
keto form could penetrate the â-CD interior as for as
about half the total deepness of the cavity. Such a picture
should explain, first, that the important decrease in the
entropic factor (a lower degree of penetration means that
the included water molecules may not be entirely expelled
in the complexation process) and, second, that the higher
value of ∆HKH, which is more than two times the value
obtained in the case of the enol tautomer, may be
attributed to the hydrogen-bonding formation between
the â-CD host and keto guest. Molecular docking between
the keto tautomer and â-CD was performed by molecular
modeling techniques.23 Docking was performed starting
from the minimum â-CD geometries (energy ) 26.29 kcal/
mol) and keto tautomer (energy ) 13.28 kcal/mol). The
energy for the calculated minimum conformation, where
The â-CD cavity is filled with water molecules,2,26 and
then their release from the â-CD interior, together with
the desolvation of the guest in the inclusion process, leads
to an increase in entropy. This gain in entropy seems
almost entirely compensated for by the entropy reduction
when the two species, enol and CD, combine to form one
species: this complexation process results in the loss of
a translational degree of freedom. On the other hand, the
enthalpic factor, which approximates the ∆H obtained
in the solubilization process of the enol in the SDS
micellar interphase, is due to the host-guest van der
Waals interactions and to interactions between the
expelled water molecules and the others in the bulk water
phase.
(26) Saenger, W.; J acob, J .; Gessler, K.; Steiner, T.; Hoffmann, D.;
Sanbe, H.; Koizumi, K.; Smith, S. M.; Takaha, T. Chem. Rev. 1998,
98, 1787. D´ıaz, D.; Vargas-Baca, I.; Garc´ıa-Mora, J . J . Chem. Educ.
1994, 71, 708. Fendler, J . H. Membrane Mimetic Chemistry; J . Wiley
& Sons: New York, 1982; Chapter 7.

Keto-Enol Equilibrium of Benzoylacetone
J . Org. Chem., Vol. 64, No. 11, 1999 3959
estimate the equilibrium association constant of BZA-
enol to â-CD; the resulting Kc value is 352 mol-1 dm3,
E
which is in perfect agreement with the value reported in
Table 2 obtained at the same temperature (∼20 °C) from
UV-vis spectrophotometry. We can also note from Figure
3 that the relative intensity of the peaks (keto-enol)
decreases as [â-CD] increases, again in accordance with
the spectrophotometric measurements, which evidence
a displacement of benzoylacetone keto-enol equilibrium
toward the enol tautomer on increasing [â-CD]. The
relative areas of both peaks (calculated only in an
approximate way) yield the percentages of both tau-
tomers reported in Table 3. Finally, analysis of the shifts
corresponding to the protons of â-CD1,27,28 can be related
to the location of the guest in the cyclodextrin cavity. The
glucose H3 and H5 protons are located within the cavity,
whereas protons H6 are located at the narrower entrances
and protons H2 and H4 at the wider entrance. In the
presence of benzoylacetone, the H3, H5, and H6 â-CD
signals shift by 0.02, 0.08, and 0.02 ppm, respectively,
whereas a much smaller shift (<0.006 ppm) is observed
for H4 and H2. This observation suggests that since enol
and keto tautomers of BZA penetrate into the â-CD cavity
lengthwise and the enol tautomer protudes deeper inside
than the keto tautomer does the methyl group of the
latter should be outside the cavity.
Even though DBM is a greater molecule than BZA, the
results agree with the formation of a 1:1 complex. The
existence of isosbestic points near λ ) 380 and 270 nm,
even if not well-defined, indicates this. A plausible
explanation may be that DBM yields only one indistin-
guishable enolic form, since it is symmetrically substi-
tuted. Methyl orange anion of the structure (two phenyl
rings attached at both ends of -NdN-) close to DBM
also yields an inclusion complex with â-CD of 1:1 stoi-
chiometry.5
Further evidence for the formation of this complex-type
can be obtained from the kinetic results of the nitrosation
reaction of the enol form. The reactive species is the enol;
i.e., the nitrosation occurs by an electrophilic attack of
the nitrosating agent (X-NO, X ) Cl, Br, SCN, ...) on
the double bond of the enol (-(OH)CdCH-). Therefore,
we might expect a reduction in the reactivity of the
complex EH‚CD because of the lower polarity of the CD
interior and of the restricted access of the nitrosating
agent to the reaction center.
(ii) Kin etic Resu lts. In aqueous acidic solutions of
sodium nitrite at low concentrations, the equilibria put
forward in Scheme 3 can be stablished to account for the
formation of the possible nitrosating agents: NO+, and
also nitrosyl halides (XNO) when X- () Cl-, Br-, SCN-,
...) is present in the medium.29
F igu r e 3. 1H NMR for the methyl signal of keto and enol
tautomers of benzoylacetone (8.0 × 10-4 mol‚dm-3) in D2O in
the absence (A) and in the presence of 0.89 × 10-3 (B) and
1.20 × 10-3 mol‚dm-3 (C) of â-cyclodextrin.
the keto tautomer is included in the â-CD cavity with
the carbonyl group close to the phenyl ring forming
H-bonds with secondary OH groups of â-CD, was 20.67
kcal/mol. Even though it is a rough simulation of what
can occur in the water solution, the energy of the complex
decreases considerably.
Further arguments in favor of the proposed complex
structure are obtained from NMR measurements. Figure
1
3 shows the effect of â-CD on the H NMR for the methyl
signals corresponding to both keto and enol tautomers.
The position for the chemical shift is determined by the
average environment sensed by the probe taking into
account the fraction bound and free. It can be seen that
no shift is observed for the methyl signal corresponding
to the keto tautomer; this observation is in complete
agreement with our picture of the â-CD‚keto complex.
By contrast, the methyl signal of the enol tautomer shifts
when the complex is formed; shift values are collected in
Table 3 as a function of â-CD concentration. Following
the work of Bohne et al.,27 we used these shift values to
Sch em e 3
KNO
HNO2 + H+ y z NO+ (or NO2H2+: hydrated NO+)
KXNO
HNO2 + H+ + X- y
z
XNO (X- ) Cl-, Br-, SCN-, ...)
(28) Schneider, H.-J .; Hacket, F.; Ru¨diger, V. Chem. Rev. 1998, 98,
1755. Mun˜oz de la Pen˜a, A.; Ndou, T. T.; Zung, J . B.; Greene, K.L.;
Live, D. H.; Warner, I. M. J . Am. Chem. Soc. 1991, 113, 1572.
(29) Williams, D. H. L. Nitrosation; Cambridge University Press:
New York, 1988; Chapter 1.
(27) Barros, T. C.; Stefaniak, K.; Holzwarth, J . F.; Bohne, C. J . Phys.
Chem. A 1998, 102, 5639.

3960 J . Org. Chem., Vol. 64, No. 11, 1999
Iglesias et al.
Ta ble 3. 1H NMR Ch em ica l Sh ifts, δ (p p m ), of -CH3 P r oton s of Both Keto a n d En ol Ta u tom er s a n d of CH P r oton s of
â-Cyclod extr in in D2O Mea su r ed w ith Resp ect to th e Resid u a l Wa ter Sign a l a t δ ) 4.60 p p m
K
E
[â-CD]/M
[BZA]/M
δMe (area)
δMe (area)
∆δ
% enol
% keto
0.00 × 10-3
0.89 × 10-3
1.18 × 10-3
2.36 × 10-3
0.80 × 10-3
0.80 × 10-3
0.80 × 10-3
0.80 × 10-3
2.154 (1.00)
2.148 (1.00)
2.151 (1.00)
2.148 (-)
2.029 (0.465)
2.047 (0.640)
2.053 (0.695)
2.063 (-)
32
39
41
68
61
69
0.018
0.024
0.034
CH protons of â-CDa
δ (H3)
δ (H5)
δ (H6)
δ (H2)
δ (H4)
9.0 × 10-3
0.89 × 10-3
1.20 × 10-3
0.0
3.753
3.734
3.734
3.628
3.545
3.545
3.670
3.648
3.651
3.469
3.465
3.465
3.418
3.414
3.414
0.80 × 10-3
0.80 × 10-3
Assignment of the 1H NMR signals of â-CD as in refs 27 and 28.
a
on the nitrosation of BZA at [H+] ) 0.030 mol‚dm-3 and
[nitrite] ) 1.67 × 10-3 mol‚dm-3 in the absence and
presence of 4.6 ×10-3 mol‚dm-3 of â-CD. In both situa-
tions, the observed rate constant varies linearly with
[Cl-] (see Figure 4b), whose slope and intercept values
are statistically different from zero in both cases. The
intercept values were (3.6 ( 0.1) × 10-4 and (2.42 ( 0.03)
× 10-4 s-1, in the absence and presence of â-CD, respec-
tively; and the slope values were (11.2 ( 0.2) × 10-4 and
(6.4 ( 0.1) × 10-4 mol-1 dm3 s-1, in the absence and
presence of â-CD, respectively. These results indicate that
â-CD inhibits the nitrosation reaction of BZA-enol either
in the nitrosation by NO+ (intercept term) or by ClNO
(slope term).
We then studied the influence of â-CD concentration
under different experimental conditions: when only the
NO+ was the nitrosating agent, i.e., by using HClO4 as
the source of H+ (see Figure 5); or in the presence of X-
() Cl-, Br-, or SCN-), i.e., when nitrosation by nitrosyl
halides, XNO, also takes place (see Figure 6). In both
situations ko decreases as [â-CD] increases.
The results obtained in the previous section point out
that the enol of BZA froms a 1:1 inclusion complex with
â-CD. On the other hand, in electrophilic substitutions,
as appropiate to nitrosation reactions, the enol is the
reactive form. Then the mechanistic hypothesis shown
in Scheme 4 for the case of benzoylacetone, with the rate-
controlling steps being the nitrosation of the enol, is
assumed. The values reported for the equilibrium con-
stants have been obtained in the previous section and
refer to 25 °C, except for the equilibrium constant
between the two enol forms,30 which are kinetically
indistinguishable because the reaction center (dCH-) is
the same in both isomers.
F igu r e 4. (a) Influence of (b) [nitrite] at [H+] ) 0.040 mol
dm-3 and (2) [H+] at [nitrite] ) 2.5 × 10-3 mol dm-3 in the
nitrosation of dibenzoylmethane in water at [DBM] ) 2.5 ×
10-5 mol dm-3; (b) influence of Cl- concentration in the
nitrosation of BZA at [nitrite] ) 1.67 × 10-3 and [H+] ) 0.031
mol dm-3 (b) in the absence of â-CD and (2) in the presece of
The reaction rate will be the sum of the rates of the
reaction via free and complexed enol, and by NO+ or XNO
as stated in eq 4.
rate ) (kNO[NO+] + kXNO[XNO])[EH] +
(kNOc[NO+] + kXNOc[XNO])[EH‚CD] (4)
[â-CD] ) 4.6 × 10-3 mol dm-3
.
The nitrosation reaction of BZA, and other â-dicarbonyl
compounds, has been studied in water.12 The reaction is
first order in both [ketone] and [H+] and is also first order
with respect to total nitrite concentration (see Figure 4a
for the case of DBM). Therefore, eq 3 can be formulated
in which k1 is the rate constant for the reaction via NO+
and k2 refers to the rate constant for the reaction via
XNO.
The anions ClO4-, SCN-, Br-, and Cl- form inclusion
complexes with â-CD. The corresponding stability con-
stants, in mol-1 dm3, are reported in the literature31 as
K(ClO4-) ) 13.7, K(Br-) ) 1.66, or K(ClO4-) ) 26.7,
K(SCN-) ) 9.9, K(Br-) ) 6.5, and K(Cl-) ) 2.56.32 The
(30) Gorodetsky, M.; Luz, Z.; Mazur, Y. J . Am. Chem. Soc. 1967,
89, 1183.
(31) Tee, O. S.; Bozzi, M.; Hoeven, J . J .; Gadosy, T. A. J . Am. Chem.
Soc. 1993, 115, 8990.
(32) Rohrbach, R. P.; Rodr´ıguez, L. J .; Eyring, E. M.; Wojcik, J . F.
J . Phys. Chem. 1977, 81, 944.
rate ) (k1 + k2[X-])[ketone]t[H+][nitrite] (3)
To test possible mechanistic changes induced by the
presence of â-CD, we first examined the influence of [Cl-]

Keto-Enol Equilibrium of Benzoylacetone
J . Org. Chem., Vol. 64, No. 11, 1999 3961
F igu r e 5. (a) Variation of ko as a function of free [â-CD] in
the nitrosation of (2) DBM at [nitrite] ) 2.5 × 10-3 mol‚dm-3
and at [H+] ) 0.040 mol‚dm-3 (HClO4) and of (b) BZA at
[nitrite] ) 1.67 × 10-3 mol dm-3 and at [H+] ) 0.032 mol dm-3
(HClO4); solid lines fit to eq 5; for parameters, see Table 4; (b)
reciprocal plot of ko against free [â-CD].
F igu r e 6. (a) Variation of ko as a function of total [â-CD] in
the nitrosation of BZA at (b) [nitrite] ) 3.1 × 10-3 mol dm-3
,
[H+] ) 0.031 mol dm-3, and [Br-] ) 0.031 mol dm-3 and at
(2) [nitrite] ) 1.67 × 10-3 mol dm-3, [H+] ) 0.031 mol dm-3
and [Cl-] ) 0.031 mol dm-3; solid lines fit to eq 5; for
parameters, see Table 4; (b) reciprocal plot of ko against total
[â-CD].
,
small values for the stability constants of the complexes
formed between â-CD and Cl- or Br-, together with the
small concentration used of these ions, means that no
corrections are necessary to the free [â-CD], which is
assumed to be the total [â-CD]. The same applies to the
case of SCN-, which has been used at 8.0 × 10-4
mol‚dm-3, but since we have used perchloric acid, it was
to the nitrosation through the complexed enol, and finally
Z ) (Kc + Kc KE)/(1 + KE).
K
E
ko )
[nitrite][H+][(k1 + k2[X-]) + (k1c + k2 [X-])Kc [â-CD]]
c
E
-
necessary to do the appropriate corrections due to ClO4
1 + Z[â-CD]
ions.
(5)
We worked with the values reported in ref 31 in order
to get better agreement in the Z values determined in
this section and those reported in Table 2. Then, taking
into account that [BZA]t ,[â-CD]t, we can assume that,
when HCl or HBr was used to acidify the reaction
medium, the total amount of cyclodextrin concentration
A multiple regression analysis of eq 5 using experi-
mental ko and [â-CD] values, along with KcE and Z values
determined in the previous section, affords the results
listed in Table 4, where experimental conditions are also
indicated; values of k1 and k2, determined here or taken
from published data,12 are also reported. Curves in
Figures 5 and 6 are the theoretical fits of the experimen-
tal data to eq 5. Figures 5b and 6b show the reciprocal
plot of ko against [â-CD], and we can note two different
behaviors depending on if the reaction is performed in
the absence (Figure 5b) or in the presence of X- (Figure
6b). According to eq 5, a linear plot of 1/ko vs [â-CD]
indicates that the complexed enol is unreactive; i.e. the
situation found when the nitrosation agent is the NO+,
either with the enol of BZA or DBM. Similar behavior
has been observed for reactions between a complexed
-
is free; but, when ClO4 ions are present, we must
consider the quantity of â-CD-forming complexes with
perchlorate ions. Free cyclodextrin concentration is de-
termined by solving the equation: [CD]f2 + [CD]f{Ki-1
+
[ClO4-] - [CD]t} - [CD]t/Ki ) 0 at each â-CD concentra-
tion, with Ki being the stability contant formed between
-
ClO4 ion and â-CD.
Hence, starting from eq 4 and Scheme 4, one easily
arrives at eq 5, which relates the variation of ko with free
[â-CD]. In this equation, k1 ) kNOKNOKE/(1 + KE); k2 )
kXNOKXNOKE/(1 + KE), i.e., the nitrosation through the free
c
c
enol, whereas the same expressions for k1 and k2 refer

3962 J . Org. Chem., Vol. 64, No. 11, 1999
Iglesias et al.
Sch em e 4
Ta ble 4. Exp er im en ta l Con d ition s a n d Resu lts Obta in ed in th e Nitr osa tion of BZA a n d DBM in Aqu eou s Acid Med iu m
in th e P r esen ce of â-CD
[H+]a
[nitrite]a
[X-]a
k1
k2
k2cKc
Zd
k2
b
c
E
c
medium
system benzoylacetone (BZA)-â-cyclodextrin (â-CD)
H2O/HClO4
H2O/HClO4
H2O/HCl
H2O/HBr
H2O/HClO4
0.023
0.030
0.030
0.030
0.030
1.67 × 10-3
1.67 × 10-3
1.67 × 10-3
3.1 × 10-3
1.67 × 10-3
7.24
7.24
156 ( 4 - 152
133 ( 4 - 129
155 ( 3
0.030 (Cl-)
7.91e
10.6e
14.65e
22
100
10220
204 ( 8
427 ( 13
1400 ( 20
0.65
1.36
4.46
0.030 (Br-)
145 ( 6
8.0 × 10-4 (SCN-)
system dibenzoylmethane (DBM)-water
160 ( 9
H2O/HClO4
H2O/HClO4
H2O/HClO4
variable
0.040
0.030
2.5 × 10-3
variable
5.8
5.3
5.6
2.5 × 10-3
variable/Cl-
10.1
system dibenzoylmethane (DBM)-â-CD
H2O/HClO4
0.040
2.5 × 10-3
.
5.8
293 ( 9 - 285
K
In mol‚dm-3
.
mol-2 dm6 s-1
c mol-3 dm9 s-1
.
Values of Z ) (Kc + KcEKE)/(1 + KE) in mol-1 dm3 obtained by fitting the data of ko
a
b
d
- [â-CD] to eq 5, and obtained from the linear plot of 1/ko against [â-CD]. e Corresponding values of k1 + k2[X-].
substrate and H3O+, i.e., specific acid hydrolysis,33 and
have been explained as due to the noninclusion of H3O+
cation into the â-CD cavity. In fact, whereas the binding
of anions of many types by CDs has been observed, and
it can be quite strong,31,32,34 the binding of cations has
rarely been observed, and only for large organic dyes,5
long-chain surfactants,35 and metal ions with organic
ligands.36 In contrast, the binding of simple cations
appears to be relatively unfavorable, e.g., piperidinium
ions bind only weakly and much less than neutral
piperidine.37 The foregoing considerations along with the
small NO+ concentrations (KNO ≈ 3.5 × 10-7 mol-1 dm3)38
explain the absence of reaction of the complexed enol
toward NO+.
4 indicate that the nitrosation of the complexed enol also
occurs. Nevertheless, values of k2 , i.e., the term of the
c
reaction of the complexed enol with nitrosyl halides (or
pseudohalides), are smaller than those obtained in water
in every case, e.g., more than 30 times when ClNO is the
nitrosating agent. This effect can be attributed to the
lower polarity of the â-CD cavity. The trend of reactivity
found in water k2(ClNO)/k2(BrNO)/k2(SCNNO) as 1:4.5:
460 is associated with the nucleophilicity order of the
anions and is often found in nitrosation studies.29,39 If
the nucleophilicity is still the driving force for catalysis
in the nitrosation of the complexed enol, the change in
c
c
c
the efficiency detected: k2 (ClNO)/k2 (BrNO)/k2 (SCNNO)
as 1:2:7 is probably explained by the inversion of the
nucleophilicity order in the cyclodextrin interior. It is
hown that the order Br- > Cl- > SCN- is only applicable
when the nucleophile is deactivated by hydration, whereas
the “natural” order of nucleophilicity SCN- > Br- > Cl-
is observed in solvents such as acetonitrile, acetone, etc.15
A similar reduction in the catalysis efficiency is often
found in nitrosation reactions carried out in dioxane-
On the contrary, when the nitrosation reaction is
promoted by XNO, results in Figure 6b and data in Table
(33) Iglesias, E. J . Am. Chem. Soc. 1998, 120, 13057.
(34) Park, H.-R.; Mayer, B.; Wolscham, P.; Ko¨hler, G. J . Phys. Chem.
1994, 98, 6158.
(35) Gonza´lez-Gaitano, G.; Crespo, A.; Compostizo, A.; Tardajos, G.
J . Phys. Chem. 1994, 98, 6158.
(36) God´ınez, L. A.; Patel, S.; Criss, G. M.; Kaifer, A. E. J . Phys.
Chem. 1995, 99, 17449. J ohnson, M. D.; Reinsborough, V. C. J . Solution
Chem. 1994, 23, 185.
(37) Barra, M.; de Rossi, R. H.; de Vargas, E. B. J . Org. Chem. 1987,
52, 5004.
(38) Ridd, J . H. Adv. Phys. Org. Chem. 1978, 16, 1.
(39) Williams, D. L. H. In Organic Reactivity: Physical and Biologi-
cal Aspects; Golding, B. T., Griffin, R. J ., Maskill, H., Eds.; The Royal
Society of Chemistry: London, 1995; p 320 and references therein.

Keto-Enol Equilibrium of Benzoylacetone
J . Org. Chem., Vol. 64, No. 11, 1999 3963
water40 or acetonitrile-water41 mixtures and also for the
reactions.43 Therefore, the results attained in the nitro-
sation of AcAc rule out this possibility, at least for the
experimental conditions of this work.
nitrosation in the micellar phase.13
The kinetic results obtained for the nitrosation of DBM
corroborates the BZA outcome, i.e., the formation of a
1:1 inclusion complex between DBM-enol and â-CD that
is unreactive toward NO+. The enol form of DBM is also
a planar molecule and is larger than the enol molecule
of BZA, but since it is symmetrically substituted, there
is only one indistinguishable DBM enol, which explains
the formation of only 1:1 complex.
Con clu sion s
This investigation demonstrates that both keto and
enol tautomers of BZA form inclusion compounds with
â-cyclodextrin of 1:1 stoichiometry. A detailed description
of the complexation was achieved by employing comple-
mentary techniques that provide information on different
aspects of the complexation process. Values of the
thermodynamic functions obtained for the inclusion
processes suggest that the enol form protudes deep inside
the CD cavity, with “hydrophobic interactions” being the
main driving force of the inclusion process; meanwhile,
the keto tautomer is about half included in the CD cavity,
forming hydrogen bonding with the OH groups located
on the wider rim of the CD host. 1H NMR measurements
corroborate these complex structures. The particular
structure of the complex enol‚CD causes this species to
be less reactive in nitrosation reactions than the uncom-
plexed enol, due mainly to the lower polarity of the CD
interior and of the restricted access of the nitrosating
agent to the reaction center.
We also examined the influence of [â-CD] on the
nitrosation of acetylacetone (AcAc) at [AcAc] ) 1.5 × 10-4
mol‚dm-3, [nitrite] ) 1.67 × 10-3 mol‚dm-3, [H+] ) 0.015
mol‚dm-3, and [Br-] ) 0.017 mol‚dm-3, by observing the
decreased absorbance at 245 nm due to enol consumption.
The results of the observed rate constant showed to be
independent of [â-CD]; in fact, ko ) 3.23 × 10-4 s-1 in
the absence of â-CD and ko ) 3.09 × 10-4 s-1 at [â-CD] )
9.0 × 10-3 mol‚dm-3. Because AcAc is quite soluble in
water, no inclusion complex appears to form under these
experimental conditions. Moreover, the results of AcAc
evidence that the presence of â-CD has no effect on
nitrosation reactions in acid media; in other words, since
â-CD has many alcoholic groups, an inhibition of the
nitrosation rate by the presence of â-CD could also be
attributed to a reduction of the nitrosating agent through
the equilibrium process: â-CD‚OH + HNO2 h â-CD‚ONO
+ H2O, which is very common for alcohols42 and is
responsible for the inhibition of alcohols in nitrosation
Ack n ow led gm en t. E.I. wishes to thank the Direc-
cio´n General de Investigacio´n Cient´ıfica y Te´cnica of
Spain for financial support (Project PB96-1085). The
authors thank J orge Otero for his assistance with the
NMR experiments.
J O982407G
(40) Bravo, C.; Herve´s, P.; Iglesias, E.; Leis, J . R.; Pen˜a, M. E. J .
Chem. Soc., Perkin Trans. 2 1990, 1969.
(41) Bravo, C.; Herve´s, P.; Leis, J . R.; Pen˜a, M. E. J . Chem. Soc.,
Perkin Trans. 2 1991, 2091.
(42) Williams, D. L. H. Nitrosation; Cambridge University Press:
New York, 1988; p 150 and references therein.
(43) (i) Casado, J .; Lorenzo, F. M.; Mosquera, M.; Rodr´ıguez-Prieto,
M. F. Can. J . Chem. 1984, 62, 136. (ii) Ridd, J . H. Quart. Rev. 1961,
15, 418.