Tautomerization of 2-acetylcyclohexanone. 1. Characterization of keto-enol/enolate equilibria and reaction rates in water

Article abstract of DOI:10.1021/jo026629x

The keto-enol tautomerism of 2-acetylcyclohexanone (ACHE) was studied in water under different experimental conditions. By contrast with other previously studied β-diketones, the keto-enol interconversion in the ACHE system is a slow process. Under equilibrium conditions, the analysis of the absorbance readings of ACHE aqueous solutions yielded more than 40% of enol content at 25 °C; nevertheless, in aprotic solvents such as dioxane, ACHE is almost completely enolized. In alkaline medium, the enolate ion is the only existing species; the study of the effect of pH on the UV-absorption spectrum of ACHE yielded a value of 9.85 for the overall pKa of ACHE. Under nonequilibrium conditions, the keto-enol tautomerization was studied in water. Several factors affecting the reaction have been investigated, which include H⁺-catalysis, ionic strength effect, buffer catalysis, deuterium isotope effects, temperature effect, or solvent effects.

Full text of DOI:10.1021/jo026629x

Ta u tom er iza tion of 2-Acetylcycloh exa n on e. 1. Ch a r a cter iza tion of
Keto-En ol/En ola te Equ ilibr ia a n d Rea ction Ra tes in Wa ter
Emilia Iglesias
Dpto de Quı´mica Fı´sica e E. Q. I. Facultad de Ciencias, Universidad de La Corun˜a,
15071-La Corun˜a, Spain
qfemilia@udc.es
Received October 30, 2002
The keto-enol tautomerism of 2-acetylcyclohexanone (ACHE) was studied in water under different
experimental conditions. By contrast with other previously studied â-diketones, the keto-enol
interconversion in the ACHE system is a slow process. Under equilibrium conditions, the analysis
of the absorbance readings of ACHE aqueous solutions yielded more than 40% of enol content at
25 °C; nevertheless, in aprotic solvents such as dioxane, ACHE is almost completely enolized. In
alkaline medium, the enolate ion is the only existing species; the study of the effect of pH on the
UV-absorption spectrum of ACHE yielded a value of 9.85 for the overall pKa of ACHE. Under
nonequilibrium conditions, the keto-enol tautomerization was studied in water. Several factors
affecting the reaction have been investigated, which include H+-catalysis, ionic strength effect,
buffer catalysis, deuterium isotope effects, temperature effect, or solvent effects.
In tr od u ction
The generally very low enol content of simple ketones
had, for a long time, restricted studies of keto-enol
interconversion to examination of only the enolization
reaction. In the past few years, Kresge et al.10-15 have
developed new methods to carry out an extensive and
very precise work on keto-enol interconversion of “simple”
enols of â-ketoacids, R-ketoesters, or â-ketoamides.
In contrast to the low enol content of monocarbonyl
compounds, the enol form of the tautomeric species of
1,3-diketones sometimes predominates over the corre-
sponding keto form. Solvent effects on the equilibrium
position of these compounds are pronounced, due to the
enol form is stabilized by intramolecular O-H‚‚‚O hy-
drogen bonds. This fact facilitates the study of their
keto-enol interconversion, and there have been extensive
studies of the reaction in water as well as in organic
solvents.16-19
Enols and enolate ions are critical intermediates in
many important reactions. The work of Lapworth on the
acid-catalyzed halogenation of acetone suggested, for the
first time, the formation of acetone-enol in the rate-
determining step which precedes involvement of halo-
gen.1 Since that time, several thousand papers have made
enolization a well-documented process.2-6
Early estimations of the enol content of simple carbonyl
compounds used the heat of formation of the related
methyl ether or the entropy terms and the Gibbs free
energy differences between enols and ethers to evaluate
enol formation equilibrium constants.7 Other methods
estimated the ketonization rate constants as being equal
to the rate constants for hydrolysis of the corresponding
methyl enol ethers. The equilibrium constants were then
calculated as the ratios between H+-catalyzed enolization
and H+-catalyzed hydrolysis rate constants.8 Usually
these independent methods have led to a much lower enol
content than that obtained by the halogen titration
method, which becomes very difficult to apply when the
enol content is very low, leading to ambiguous results;
in addition, the halogenation rate constant is assumed
as the value of a diffusion-controlled process.9
During the past few years, the keto-enol interconver-
(9) Gero, A. J . Org. Chem. (a) 1954, 19, 469; (b) 1954, 19, 1960; (c)
1961, 26, 3156.
(10) Kresge, A. J . Chem. Soc. Rev. 1996, 275.
(11) Kresge, A. J . Acc. Chem. Res. 1990, 23, 43.
(12) Kresge, A. J . In The Chemistry of the Enols; Rappoport, Z., Ed.;
Wiley: Chichester, England, 1990; pp 399-480.
(13) Chiang, Y.; Gue, H.-X.; Kresge, A. J .; Tee, O. S. J . Am. Chem.
Soc. 1996, 118, 3386. Chiang, Y.; Kresge, A. J .; Nikolaev, V. A.; Popik,
V. V. J . Am. Chem. Soc. 1997, 119, 11183. Chiang, Y.; Kresge, A. J .;
Meng, Q.; Morita, Y.; Yamamoto, Y. J . Am. Chem. Soc. 1999, 121, 8345.
(14) Chiang, Y.; Kresge, A. J .; Schepp, N. P.; Xie, R.-Q. J . Org. Chem.
2000, 65, 1175.
(15) Chiang, Y.; Criesbeck, A. G.; Heckroth, H.; Hellrung, B.; Kresge,
A. J .; Meng, Q.; O’Donoghue, A. C.; Richard, J . P.; Wirz, J . J . Am.
Chem. Soc. 2001, 123, 8979.
(16) Morton, R. A.; Hassan, T.; Calloway, T. C. J . Chem. Soc. 1934,
883.
(1) Lapworth, A. J . Chem. Soc. 1904, 30.
(2) Forse´n, S.; Nilson, M. In The Chemistry of the Carbonyl Group;
Zabicky, J ., Ed.; Interscience: London, 1970; pp 157-240.
(3) Hart, H. Chem. Rev. 1979, 79, 515.
(4) Toullec, J . Adv. Phys. Org. Chem. 1982, 18, 1.
(5) Rappoport, Z.; Biali, S. E. Acc. Chem. Res. 1988, 21, 442.
(6) The Chemistry of the Enols; Rappoport, Z., Ed.; Wiley: Chich-
ester, England, 1990.
(7) Guthrie, J . P. In The Chemistry of the Enols; Rappoport, Z., Ed.;
Wiley: Chichester, England, 1990; pp 75-93.
(8) Guthrie, J . P. Can. J . Chem. 1979, 57, 236; 1979, 57, 797; 1979,
57, 1177.
(17) Murthy, A. S. N.; Balasubramanian, A.; Rao, C. N. R.; Kasturi,
T. R. Can. J . Chem. 1962, 40, 2267.
(18) Mills, S. G.; Beak, P. J . Org. Chem. 1985, 50, 1216.
(19) Toullec, J . The Chemistry of the Enols; Rappoport, Z., Ed.;
Wiley: Chichester, England, 1990; pp 323-398.
10.1021/jo026629x CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/21/2003
2680
J . Org. Chem. 2003, 68, 2680-2688

Tautomerization of 2-Acetylcyclohexanone
yield. The enol so recovered tautomerizes slowly to the keto
form until to achieve the equilibrium proportions.The inte-
grated method was followed throughout, fitting the experi-
mental data (absorbance-time, A-t) to eq 1, where A0, A∞,
and A mean absorbance readings at times 0, infinity, and t,
respectively, and ko represents the pseudo-first-order rate
constant. In every case, satisfactory correlation coefficients (r
> 0.9999) and residuals were obtained. Under these experi-
mental conditions, a decrease in absorbance (A0 - A∞ in eq 1
is positive) at 291 nm from approximately 1 to 0.4 absorbance
units was observed.
SCHEME 1. Str u ctu r es of 1,3-Dica r bon yl
Com p ou n d s Th a t Sh ow Ra p id Keto-En ol
Ta u tom er iza tion
k
e
A ) A∞ + (A0 - A∞)e-k t with ko ) ko + ko
(1)
o
sion of 1,3-diketones or 1,3-ketoesters, as well as their
reactivity, was investigated in aqueous solutions of
surfactant-forming micelles.20-27 The research reported
in the current work upon the isomerization of 2-acetyl-
cyclohexanone (ACHE) shows that, conversely to the
compounds previously studied by us (Scheme 1), or even
to its homologous compound, 2-acetylcyclopentanone,28
the keto-enol interconversion of ACHE is a slow reaction.
Therefore, we were able to perform a detailed kinetic
study of the keto-enol tautomerization of this compound
by analyzing several parameters, including salts effects,
acid-base catalysis, isotope effects, or the influence of
temperature.
As the observed process is an approach to the equilibrium,
ko is the sum of the rate constants corresponding to ketoniza-
k
e
tion (ko ) and enolization (ko ). For each kinetic run, the pseudo-
first-order rate constant for ketonization was then determined
k
29
e
k
as ko ) ko(A0 - A∞)/A0 and ko ) ko - kok. (Of course, ko
)
ko/(1 + KE), but we used the former expression in order to
determine also KE).
Resu lts
Keto-En ol/En ola te Equ ilibr ia . In aprotic solvents,
such as dioxane, the enolic form of ACHE is the majority
species, whereas in water, a mixture of both the keto and
enol tautomers exists. However, contrary to the case of
other previously studied 1,3-dicarbonyl compounds, the
keto-enol interconversion in the ACHE system is slow
enough to follow it by conventional methods. In this
sense, the recorded spectrum just after the addition of
an aliquot of a dioxane solution of ACHE into a sample
of water (dilution factor higher than 250) shows an
absorption band due to the enol (λmax ) 291 nm) which
decreases slowly with time because of the enol conversion
into the keto form. Conversely, when an aqueous equili-
brated ACHE solution is diluted (∼1:29) in dioxane the
absorption band centered at 291 nm increases slowly with
time, as a consequence of the conversion of the keto
tautomer into the enol. The presence of a well-defined
isosbestic point at λ ∼ 228 nm states for the keto-enol
equilibrium. Figure 1 shows these experimental observa-
tions.
Exp er im en ta l Section
Ma ter ia ls. 2-Acetylcyclohexanone, of the maximum purity,
was used as supplied. D2O was >99.9% isotope enrichment
and d ) 1.11. All other reagents were also used as received.
For the buffer plots for the determination of rate constants,
the buffers sodium acetate/acetic acid and its chloro- and
dichloro-derivatives were used. For determination of the pKa
in the basic pH range, use was made of hydrogen carbonate/
carbonate and phosphoric acid(1)/hydrogen phosphate buffers.
Aqueous solutions were prepared with doubly distilled water
obtained from a permanganate solution. Freshly prepared
solutions were used in all experiments.
Meth od s. UV-vis absorption spectra and kinetic measure-
ments were recorded with a double-beam spectrophotometer
provided with a thermostated cell holder. A matched pair of
quartz cells with l ) 1 cm light path was used, especially in
obtaining the spectra. The pH was measured with a pH meter
equipped with a GK2401B combined glass electrode. The glass
electrode was standardized by using commercial standard pH
4.01, 7.01, and 9.26 buffers.
Kinetic measurements were carried out under pseudo-first-
order conditions, with the concentration of ACHE, the limiting
reagent, being more than 20 times lower than that of the
others reactants. The rate of the transformation of the enol
form of ACHE to the equilibrium mixture has been studied in
water by monitoring the decreasing absorbance at λ ) 291 nm
due to the enol form. For that, ACHE was dissolved in dioxane
to give a stock solution 0.018 M. The reaction was initiated
by injecting 10 µL of this solution into 3.0 mL of water
containing the rest of the reagents. Alternatively, ACHE can
be dissolved in alkaline medium to yield the enolate quanti-
tatively, which, after acidification, gives the enol with 100%
The keto-enol equilibrium constant, KE, was measured
in water from the Beer-Lambert law under nonequilib-
rium conditions. The substrate ACHE was added to the
reaction mixture from a stock dioxane solution, and the
rate of approach to equilibrium was observed as a
function of ACHE concentration. As ACHE exists in
dioxane, mainly as the enol tautomer, the extrapolated
absorbance readings at t f 0 (determined from the fit of
A-t to eq 1) correspond to the absorption of the enol, then
eq 2 applies. The plot of A0 against [ACHE]t yields a
291
good straight line passing through the origin. The slope
of this line equals the molar absorption coefficient of the
enol at 291 nm, i.e., ꢀEH ) 15 270 ( 200 mol-1 dm3 cm-1
(r ) 0.9998). Morton et al.16 concluded that a given
tautomer of 1,3-diketone is substantially independent of
the solvent in regard to the spectral location and the
molar absorption coefficient of the maxima, which in
aprotic solvents reaches values close to 15 000 mol-1 dm3
cm-1, in good agreement with the value determined here.
On the other hand, the absorbance readings at t f ∞
(20) Iglesias, E. J . Phys. Chem. 1996, 100, 12592.
(21) Iglesias, E. J . Chem. Soc., Perkin Trans. 2 1997, 431.
(22) Iglesias, E. Langmuir 2000, 16, 8438.
(23) Iglesias, E. Langmuir 1998, 14, 5764.
(24) Iglesias, E.; Dom´ınguez, A. New J . Chem. 1999, 23, 851.
(25) Iglesias, E. Langmuir 2001, 17, 6871.
(26) Iglesias, E. J . Org. Chem. 2000, 65, 6583.
(27) Iglesias, E. J . Phys. Chem. B 2001a , 105, 10287; 2001b, 105,
10295.
(29) See for example Laidler, K. J . Chemical Kinetics, 3rd ed.;
(28) Iglesias, E. New J . Chem. 2002, 26, 1352.
Harper Collins Publishers: New York, 1987; Chapter 2.
J . Org. Chem, Vol. 68, No. 7, 2003 2681

Iglesias
F IGURE 2. (A) Absorption spectrum of ACHE (5.0 × 10-5
M) obtained in water as a function of pH: 1, 7.55; 2, 8.56; 3,
9.37; 4, 9.65; 5, 9.82; 6, 10.05; 7, 10.28; 8, 10.55; 9, [NaOH] )
0.16 M. (B) Spectrophotometric titration curve for the acid
ionization of ACHE (5.0 × 10-5 M) in aqueous solutions of the
buffers HPO42-/H2PO4- and HCO3-/CO32- at 25 °C. The inset
shows the fit of the data A309-[H+] according to eq 6.
F IGURE 1. (A) Repeat scans (1-8) every 2 min (9 at 3 min,
10 at infinite time) showing the decreasing absorbance due to
ketonization of ACHE-enol, [ACHE] ) 6.0 × 10-5 M; [H+] )
0.050 M. (B) Variation of the absorbance at 291 nm as a
function of time for (4) enol-ketonization in water, [ACHE]
) 6.5 × 10-5 M, [H+] ) 0.015 M, and for (O) keto-enolization
in 70% v/v dioxane/water, [ACHE] ) 6.0 × 10-5 M, [H+] )
0.013 M. The temperature was 25 °C.
SCHEME 2. Keto-En ol/En ola te Equ ilibr ia of
ACHE in Aqu eou s Ba sic Med iu m
(determined also from the nonlinear regression analysis
of A-t data to eq 1) correspond to the absorption of the
enol in an equilibrated solution of ACHE, in which case
eq 3 applies. The plot of A∞291 versus [ACHE]t results also
in a good straight line with a slope value of 6472 ( 100
mol-1 dm3 cm-1 (r ) 0.9999). The combination of both
slope values gives KE ) 0.72, which means more than
40% of enol content of ACHE in water.
A0 ) ꢀEHl[ACHE]t
(2)
(3)
291
KE
A∞ ) ꢀEH
l
[ACHE]t
291
1 + KE
mixed acid ionization constant, Ka, whose value corre-
sponds to the pH of the inflection point. This acid
ionization constant measured in water can be defined as
in eq 4.
The acid ionization equilibrium constant, Ka, of ACHE
was determined from the analysis of its absorption
spectrum as a function of pH, which is shown in Figure
2A. This graph is typical of the spectral observations for
enolate formation and implies a clean equilibration of the
acidic species (the keto and enol tautomers) with their
common conjugate base, the enolate (Scheme 2).
[EH] + [KH]
1
1
1
)
)
+
(4)
[E-]t[H+]
Ka
Ka
E
K
Ka
Enolate anions are ambident species and the equilib-
rium enolization constant, KE, linking the two tautomeric
protonated forms, is a function of the acidity constants
of these two species. Thus, titration of an equilibrium
mixture of keto (KH) and enol (EH) tautomers yields a
The absorbance readings at the maximum wavelength
(λ ) 309 nm) describe a sigmoid curve on varying the
pH, Figure 2B. This absorption is due to both the enol
and enolate species; then A309 ) ꢀEHl[EH] + ꢀEl[E-]. By
considering the mass balance equation, [ACHE]t ) [KH]
2682 J . Org. Chem., Vol. 68, No. 7, 2003

Tautomerization of 2-Acetylcyclohexanone
TABLE 1. Obser ved Ch a r a cter istics for th e Th r ee
F or m s (Keto, En ol, a n d En ola te) of
2-Acetylcycloh exa n on e Existin g in Wa ter
keto (KH)
enol (EH)
291
enolate (E)
λ
max/nm
<200
-
309
34200
ꢀ/mol-1 dm3 cm-1
15270
i
pKa
9.62
9.47 (9.47)a
KE
pKa
0.72-0.75 (0.41)a
9.85 (9.85)a
a
Values obtained by Riley and Long30 in the bromination of
this ketone.
+ [EH] + [E-] and the expression of Ka, the eq 5 is
derived, where Aa and Ab refer to the absorbance readings
at very low and very high pH, respectively.
Aa10pK + Ab10pH
a
A309
)
(5)
10pK + 10pH
a
The solid sigmoid curve in Figure 2B was constructed
with the experimental values of Aa ) 0.191 and Ab )
1.565 and the adjusted value of pKa ) 9.85 ( 0.01 (r )
0.9995). From the latter value and by making used of KE
determined previously, the relationships of Ka ) KaK/(1
K
E
+ KE) or Ka ) KaEKE/(1 + KE), give pKa ) 9.62 and pKa
) 9.47, respectively to the pKa values of the keto and enol
tautomers. These results show that the stronger acid
(enol) of a pair of compounds in equilibrium with a
common anion (enolate) is the minor component of the
mixture. Alternatively, the experimental data of A309
-
∞
∞
[H+] may be related through eq 6, where AEH and AE
correspond to the absorbance readings when all the
[ACHE] is in the enol or enolate forms, respectively. The
solid line in the inset of Figure 2B fits eq 6 to the
F IGURE 3. Variation of the observed rate constant for
ACHE-enol ketonization, kok, as a function of [H+] (A) at 25
°C and ionic strength, controlled with NaCl, (b) 0.20 M, (2)
0.40 M, and (1) 0.75 M, and (B) at I ) 0.20 M and different
temperatures.
E
experimental points when Ka ) (3.2 ( 0.1)10-10 mol
dm-3; (1 + KE)/KE ) 2.55 ( 0.08; AEH ) 0.45 ( 0.03,
∞
∞
E
ACHE in alkaline medium ([NaOH] ) 0.017 M); subse-
quently the reaction mixture was acidified with the
required amount of HCl to reach the desired [H+]. The
ketonization rates of the enol generated in situ were
studied as a function of [H+]. Figure 3A shows the
variation of the observed rate constants as a function of
[H+], where eq 7 applies. Least-squares fitting of this
equation to the experimental data yields the rate con-
stants listed in Table 2.
and AE ) 1.608 ( 0.020. These results imply pKa
)
9.49; KE ) 0.65, and ꢀE ) 32 160 mol-1 dm3 cm-1, in good
agreement with the expected values. All these results are
summarized in Table 1.
E
AEH∞[H+] + AE∞Ka
A309
)
(6)
1 + KE
E
Ka
+
[H+]
KE
k0 ) kw + kHk[H+]
(7)
k
k
Ra tes of Keto-En ol Ta u tom er iza tion . The keton-
ization reaction of the enol tautomer of ACHE was
studied by monitoring the decreasing absorbance at λ )
291 nm.
Isotop ic Effects. Rates of ketonization were also
measured in dilute D2O solutions of hydrochloric acid at
I ) 0.20 M (NaCl). The results are reported in Table 2.
One can note the lower rate constant observed in D2O
than in H2O. The combination of the data obtained for
ketonization process in these two solvents at the same
ionic strength gives large kinetic isotope effects for the
In flu en ce of Ion ic Str en gth . The ionic strength
dependence of k0, the first-order rate constant for reaction
of the enol to form the equilibrium mixtures, was
analyzed at [H+] ) 0.05 M (HCl) by varying NaCl
concentration. The results reveal no appreciable effect
of this parameter in the range of ionic strength investi-
gated (0.05 to 1 M); for instance, k0 ) 1.68 × 10-3 s-1 (or
k
k
e
uncatalyzed reaction: kw O)/kw
) 8.2 (or kw
/
(H2
(D2O)
(H2O)
e
kw
) 9.1), whereas lower isotope effects for H+-
(D2O)
k0 ) 1.0 × 10-3 s-1) at I ) 0.08 M; k0 ) 1.65 × 10-3 s-1
catalyzed reaction was observed: kHk/kD ) 3.2 (or kH
/
k
k
e
(or k0 ) 0.98 × 10-3 s-1) at I ) 0.45 M, and k0 ) 1.61 ×
kD ) 3.6).
k
e
k
10-3 s-1 (or k0 ) 0.98 × 10-3 s-1) at I ) 0.95 M.
Activa tion P a r a m eter s. The influence of [H+] at ionic
strength 0.20 M on rates of ketonization of ACHE-enol
in water was studied at several temperatures ranging
In flu en ce of [H+]. Rate constants for ketonization
were measured as a function of [H+] (HCl) at 25 °C at
0.20, 0.40, and 0.75 M ionic strength (NaCl). The set of
experiments at I ) 0.20 M was performed by dissolving
k
e
from 283 to 308 K. The obtained k0 (or k0 ) values as a
function of both [H+] and temperature are displayed in
J . Org. Chem, Vol. 68, No. 7, 2003 2683

Iglesias
TABLE 2. Exp er im en ta l Con d ition s a n d Ra te Con sta n ts Obta in ed in th e Keton iza tion of ACHE P er for m ed in H 2O a n d
D2O a t Sever a l Ion ic Str en gth s a n d Tem p er a tu r es
b
I/M
T/°C
kw/10-3 s-1a
kH/mol-1 dm3 s-1a
kwk/s-1
kHk/mol-1 dm3 s-1
KE
0.20c
0.20
0.40
0.75
0.20d
0.10d
0.20f
0.20
0.20
0.20
0.20
0.20
0.20
25
25
25
25
25
25
25
11.3
15.4
20.3
25.1
30.0
35.0
1.393 ( 0.003
1.395 ( 0.004
1.306 ( 0.005
1.215 ( 0.004
1.35 ( 0.01e
(5.84 ( 0.05) × 10-3
(5.80 ( 0.02) × 10-3
(6.67 ( 0.03) × 10-3
(8.14 ( 0.03) × 10-3
8.52 × 10-4
8.11 × 10-4
7.68 × 10-4
7.15 × 10-4
8.10 × 10-4
8.31 × 10-4
1.06 × 10-4
2.25 × 10-4
3.23 × 10-4
5.25 × 10-4
8.19 × 10-4
11.9 × 10-4
18.1 × 10-4
3.87 × 10-3
3.37 × 10-3
3.92 × 10-3
4.79 × 10-3
0.68
0.72
0.69
0.71
1.377 ( 0.002
0.163 ( 0.001
0.382 ( 0.0016
0.549 ( 0.008
0.892 ( 0.003
1.393 ( 0.005
2.03 ( 0.02
(1.72 ( 0.02) × 10-3
(1.466 ( 0.015) × 10-3
(2.57 ( 0.08) × 10-3
(3.79 ( 0.03) × 10-3
(5.84 ( 0.05) × 10-3
(10.2 ( 0.2) × 10-3
(15.65 ( 0.1) × 10-3
1.12 × 10-3
0.86 × 10-3
1.51 × 10-3
2.23 × 10-3
3.44 × 10-3
6.02 × 10-3
9.21 × 10-3
0.65
0.71
0.71
0.69
0.72
0.66
0.66
3.09 ( 0.01
isotope effects: kwk(H2O)/kwk(D2O) ) 8.2, kHk/kD ) 3.2; kwe(H2O)/kwe(D2O) ) 9.1, kHe/kD ) 3.6
k
e
b
d
k
ako ) kw + kH[H+]. Average value of at least nine determinations. c ACHE-enol generated in situ. Reaction in pure water, kw ) kw
+ kwe; obtained from the fit of eq 1 to the experimental data A-t. e Mean value of two determinations. f Reaction in D2O.
TABLE 3. Activa ted P a r a m eter s Cor r esp on d in g to th e Wa ter -Ca ta lyzed Rea ction (kH2O) a n d H+-Ca ta lyzed Rea ction
(kH) for Both th e Keton iza tion a n d En oliza tion Rea ction s of ACHE
k
e
ketonization reaction in water (ko
)
enolization reaction in water (ko )
reaction
∆Hq (kJ /mol)
∆Sq (J /mol‚K)
∆Hq (kJ /mol)
∆Sq (J /mol‚K)
kw/[H2O]/mol-1 dm3 s-1
kH/mol-1 dm3 s-1
63.8
69.4
-140
-75.5
59.6
68.8
-107.4
-63.8
Figure 3B. There is a strong temperature effect on both
buffer and the measured pH, the acetate concentration,
[AcO-] (or the chloroacetate concentration, [ClAcO-]) is
determined at each buffer concentration. In that case,
k
kw and kHk, whose values are depicted in Table 2. The
Arrhenius plot gives good straight lines in both cases.
Using the standard Eyring equation, the activated pa-
rameters were determined from the gradient () -∆Hq/
R) and the intercept () ln(k/h) + ∆Sq/R, with k being the
Botlzmann’ constant and h the Planck’ constant) of the
graph ln(ki/T) against T-1 (with ki being kw or kH,
respectively to the processes water- or H+-catalyzed,
either for ketonization or enolization). The resulting
values are gathered in Table 3.
Gen er a l Ba se-Ca ta lysis. The ketonization reaction
was also studied in aqueous buffered solutions of acetic
acid-acetate and of its chloro-derivatives at 25 °C and I
) 0.25 M, controlled with NaCl. Parts A and B of Figure
4 show the observed rate constants as a function of both
the buffer concentration and the pH of the solution for
the case of acetic acid-acetate and chloroacetic acid-
chloroacetate buffers, respectively. It can be seen that
k0 increases with both the [buffer] and the pH of the
reaction medium. These features are typical character-
istics of a general base-catalyzed reaction. At constant
pH, k0 increases linearly with total buffer concentration
according to eq 8, where Ka represents the acidity
constant of the acid form of the buffer.
the plot of k0, or k0 , against [AcO-] (or [ClAcO-]) should
k
result in a straight line showing the same slope () kB or
kBk, respectively) independent of the pH of the experi-
ment. The corresponding graphs are shown in parts C
k
and D of Figure 4 for the case of k0 . The catalytic
coefficients, kBk, obtained for each buffer are reported in
Table 4. The Bro¨nsted plot of log(kBk) against pKa of the
three acids studied led to a linear relationship, with a
slope value equals to â ) 0.464 ( 0.003 (r ) 0.9999).
Discu ssion
Keto-En ol/En ola te Equ ilibr ia . The equilibria in-
volved as in acid as in basic media in 2-acetylcyclohex-
anone was studied in the current work by UV-absorption
spectroscopy, in a spectral region where the ketone
exhibits negliglible absorption, since only the enolic
species is available, the most straightforward measure-
ment is of the rate of disappearance of the enol to form
the equilibrium mixture. The obtained results indicate
that ACHE is nearly 43% enolized in aqueous acid
medium, KE ) 0.72. In alkaline medium, ACHE exists
as the enolate ion: the overall pKa was measured as 9.85,
Ka
E
k0 ) ks + kbuf[AcO-] ) ks + kB
[buffer]t (8)
which together with the KE value infers pKa ) 9.47 and
K
Ka + [H+]
pKa ) 9.65. These results agree, in part, with those of
Riley and Long.30 These authors studied the bromination
of 2-acetylcyclohexanone in water at 0.10 M (NaCl) ionic
strength by observing the decrease in absorbance at 390
nm due to bromine. By working with 4 × 10-3 M of
ACHE, they determined the keto-enol equilibrium con-
stant at 25 °C as KE ) 0.41, while the measured pH of
an aqueous ACHE solution partially neutralized gives a
The values of the intercept at the origin (ks) and the
gradient, kbuf () kBKa/(Ka + [H+])) of ko versus [buffer]
plots obtained at constant pH, are collected in Table 4,
k
k
along with values of ks {) ks/(1 + KE)} and kbuf {)kbuf
/
(1 + KE)}. One may observe that ks () kw + kH[H+]) is
practically pH-independent, but kbuf increases with pH,
indicating that the basic form of the buffer catalyses the
reaction, as it is assumed in eq 8. From the pKa of the
(30) Riley, T.; Long, F. A. J . Am. Chem. Soc. 1962, 84, 522.
2684 J . Org. Chem., Vol. 68, No. 7, 2003

Tautomerization of 2-Acetylcyclohexanone
F IGURE 4. Variation of ko for enol-ketonization as a function of (A) acetic acid-acetate buffer concentration and pH: (1) 4.28,
(9) 4.43, (b) 4.60, (2) 4.85, and ([) 5.08; and of (B) chloroacetic acid-chloroacetate buffer concentration and pH: ([) 2.05, (O)
2.25, (1) 2.43, (9) 2.62, (2) 2.80, and (b) 2.95, at 25 °C and I ) 0.25 M (NaCl). Values of ko against (C) acetate ion concentration,
pKa ) 4.64, and (D) chloroacetate ion concentration, pKa ) 2.73.
TABLE 4. F ir st-Or d er , ks, ksk , a n d Secon d -Or d er , kbu f, kbu fk , Ra te Con sta n ts Obta in ed in th e Stu d y of th e In flu en ce of
Bu ffer Con cen tr a tion in th e Keton iza tion Rea ction of ACHE a t I ) 0.25 M (Na Cl) a n d 25 °C
k
a
k a
pH
ks/10-3 s-1
k
buf/10-2a
ksk/10-4 s-1
kbufk/10-2a
KE
kB (kB
)
kHA
Buffer of Acetic Acid-Acetate (pKa ) 4.64)b
5.08
4.81
4.60
4.43
4.28
1.33 ( 0.07
1.36 ( 0.06
1.38 ( 0.06
1.39 ( 0.04
1.33 ( 0.05
13.61 ( 0.05
10.71 ( 0.02
8.46 ( 0.08
6.79 ( 0.04
5.38 ( 0.03
7.6
7.7
7.9
8.1
7.5
7.78
6.08
4.89
3.66
3.02
0.75
0.76
0.73
0.73
0.77
0.179c
(0.104)
7.03 × 103
Buffer Chloroacetic Acid-Chloroacetate (pKa ) 2.73)b
2.95
2.81
2.65
2.43
2.25
2.04
1.41 ( 0.02
1.44 ( 0.02
1.39 ( 0.01
1.43 ( 0.07
1.45 ( 0.02
1.48 ( 0.01
1.475 ( 0.01
1.222 ( 0.006
1.030 ( 0.008
0.784 (0.007
0.574 ( 0.010
0.392 ( 0.010
8.3
8.4
8.3
8.4
8.6
8.8
0.87
0.70
0.71
0.68
0.70
0.69
0.68
0.0234c
(0.0134)
7.03 × 104
0.715
0.613
0.461
0.340
0.233
Buffer Dichloroacetic Acid-Dichloroacetate (pKa ) 1.35)b
1.30
1.37
1.67 ( 0.02
1.64 ( 0.01
0.248 ( 0.006
0.270 ( 0.05
9.65
9.6
0.143
0.158
0.73
0.71
5.26 × 10-3c
4.03 × 105
(3.06 × 10-3
)
In mol-1 dm3 s-1
.
From ref 36. c Determined from the slope of ko (kok) versus [base] plot; see Figure 4C,D.
a
b
pKa value quite similar to that determined by us. We
determined the keto-enol equilibrium constant of 2-ace-
tylcyclopentanone as KE ) 0.40,28 but cyclic compounds
of six-membered rings have higher enol contents than
the corresponding five-membered; therefore, the KE
0.72 determined in the present work appears to be more
)
J . Org. Chem, Vol. 68, No. 7, 2003 2685

Iglesias
SCHEME 3. P ostu la ted Tr a n sition Sta te for
Keto-En ol Con ver sion in ACHE System
reliable. The agreement between both pKa values is
expected, because both results are direct determinations;
the different values obtained for KE could be due to
uncertainties in the end point (or to side reactions) in
bromination titration as a consequence of the high enol
content in ACHE; in fact, enolization is a H+-catalyzed
reaction contrarily to the reported results of Riley and
Long30 which are more proper for the situation where the
bromination of the enol is the slow step.24
Mech a n ism of Keto-En ol Con ver sion . The conver-
sion of the enol tautomer of a â-dicarbonyl compound to
its keto isomer requires the removal of hydrogen from
the carbonyl oxygen and placement of hydrogen on
carbon. The reverse applies for the conversion of the keto
into its enol isomer. The two processes are catalyzed by
both acids and bases, which fact indicates that the
hydrogens move as protons in the rate-determining step.
Acid -Ca ta lyzed Keto-En ol Con ver sion . The reac-
tion mechanism of enol ketonization involves rate-
determining H+-transfer from any available acid to the
â-carbon atom of the enol or its enolate ion. Rate profiles
shown in Figure 3 suggest â-carbon protonation of only
entropy values associated with the uncatalyzed process
are in agreement with the solvent isotope effects deter-
mined for this reaction step.
For an understanding of the previous figures, it is
necessary to remember that when enol-ketonization goes
on in D2O solvent, the enol-H would be a deuteron (D)
and its transfer would contribute an additional compo-
nent to the isotope effect. Enol-ketonization involves
proton/deuteron removal from oxygen by H2O/D2O, which,
from chemical intuition one should to believe that this
transfersbetween two electronegative atoms-sshould be
fast, and the H/D transfer to carbon from H2O/D2O,
whose process would expect to be rate-determining. Then
as a O-H (D) bond is broken in the rate-determining
step, a simple kinetic treatment of kinetic deuterium
isotope effect in which the zero-point-energy difference
between the stretching vibration of a O-H and O-D
bond is allowed to disappear in the TS predicts k(H2O)/
k(D2O) )10.6 at 25 °C.32 However, this theoretical
maximum value, experimentally is found to be consider-
able lower. Variation of bending frequencies in TS affects
the magnitude of isotope effects, and bending motions
in general show much greater variation in frequency than
do stretching motion. Since in enol-ketonization the C(2)
changes from sp2 to sp3, then the hybridization of the
activated complex is intermediate between them. But the
bending vibration frequencies are lower in sp2 than in
sp3 hybridization; thus, in enol-ketonization the bending
vibration frequencies are higher for the TS than for the
reactant. This fact decreases k(H2O)/k(D2O) ratio (the
maximum effect is 1.45).33 In addition, the intramolecular
H-bonds (O‚‚‚H‚‚‚O) in the enol may stabilize this isomer
by >10 kcal/mol,34 in other words, this effect increases
the magnitude of k(H2O)/k(D2O) ratio. By joining all these
factors, the observed value of the isotope effect in enol-
ketonization seems to be quite reasonable.
k
the un-ionized enol by H+. Therefore, in eq 7, kw refers
to the water assisted enol-ketonization reaction, whereas
kH is the rate constant H+-catalyzed. Data in Table 2
k
k
k
show that kw or kH depend on the ionic strength. The
former rate constant refers to a reaction between two
polar molecules (enol + water), whereas the latter
conforms to a reaction between an ion (H+) and a polar
molecule. At ionic strength higher than 0.1 M, the activity
coefficients of nonelectrolytes are given by log γo ) boI,
where bo is an empirical constant.31 For reactions between
two nonelectrolytes (the case of kwk), or between an ion
and a polar molecule (the case of kHk), the variation of γo
with I implies that log k is a linear function of I. This is
just the experimental observed behavior, with the fol-
lowing results: log(kwk) ) (-3.073 ( 0.004) - (0.098 (
0.008)I and log(kHk) ) (-2.52 ( 0.01) + (0.27 ( 0.02)I.
The opposite effects of I on the two processes is a
consequence of the different nature of the TS, vide infra,
depending if the reaction is water- or H+-catalyzed; thus,
the slope of the linear plots log(k) versus I depends on
A
B
q
the difference (bo + bo - bo ), where A and B refer to
both reactants. The same effect is observed for the
enolization process in agreement with the principle of
microscopic reversibility.
Both processes are markedly affected by temperature.
The activation parameters reported in Table 3 indicate
that the keto-enol equilibrium of ACHE is not temper-
ature dependent. This fact is easily understood if the
standard enthalpies of formation of both isomers are
considered, which can be estimated from bond-energies.
The large and negative activation entropy values associ-
ated to the uncatalyzed process would be the expected
ones for keto-enol interconversion involving a transition
state that restricts the degree of freedom of more than
one water molecule as postulated in Scheme 3.
With regard to the H+-catalyzed keto-enol transfor-
mation and proceeding now with keto-enolization, we
assumed the reaction mechanism stated in Scheme 4.
The preequilibrium protonation of the substrate by H3O+
leads to an increase in isotopically sensitive zero-point
energy and an inverse solvent isotope effect on this step
(32) Bell, R. P. The Proton in Chemistry; Chapman and Hall:
London, 1973; Chapter 11.
(33) (a) Lorry, T. H.; Richardson, K. S. Mechanism and Theory in
Organic Chemistry; Harper & Row: New York, 1987; Chapter 2. (b)
Melander, L.; Saunders, W. H. Reaction Rates of Isotopic Molecules;
J ohn Wiley & Sons: New York, 1987; Chapter 6.
Isotop e Effects. At 25 °C, the values of the solvent
isotope effects observed for the uncatalyzed reaction are
e
kwk(H2O)/kwk(D2O) ) 8.2, for enol-ketonization, and kw
-
(H2O)/kwe(D2O) ) 9.1, for keto-enolization. Activation
(34) (a) J effrey, G. A. An Introduction to Hydrogen Bonding; Oxford
University Presss: New York, 1997. (b) Dziembowska, T.; Rozwad-
owski, Z. Curr. Org. Chem. 2001, 5, 289.
(31) Brezonik, P. L. Chemical Kinetics and Process Dynamics in
Aquatic Systems; Lewis Publishers: Florida, 1994; Chapter 3.
2686 J . Org. Chem., Vol. 68, No. 7, 2003

Tautomerization of 2-Acetylcyclohexanone
SCHEME 4. Rea ction Mech a n ism of
Acid -Ca ta lyzed En oliza tion in ACHE System
SCHEME 5. Rea ction Mech a n ism of
Ba se-Ca ta lyzed En ol-Keton iza tion in ACHE
System
of K(H2O)/K(D2O) ) 0.6 to 0.3.35 This effect, however, is
overwhelmed by the primary isotope effect produced in
the rate-determining step, in which a C-H (D) bond is
broken, and the net result is an overall isotope effect in
the normal direction, but smaller: the experimental
value is kH/kD ) 3.5 (see Table 2).
Ba se-Ca ta lyzed Keto-En ol In ter con ver sion . In
both ketonization and enolization processes, the general
base catalysis is more pronounced than the proton-
k
catalysis. The buffer catalytic coefficients, kbuf (or kbuf
,
kbufe) given in Table 4 can be separated into their general
acid, kHA, and general base, kB, components (the latter
increase with the strength of the base). Then eq 9 applies.
[HA]
kbuf ) kB + (kHA - kB)
(9)
[buffer]
k
The plot of kbuf (or kbuf or kbufe) against the fraction of
the buffer present in the acid form, i.e., [HA]/[buffer],
gives a good straight line with negative slope () kHA
-
kB); the absolute value of the slope equals the intercept
() kB), which means that only general base catalysis is
occurring, i.e., kHA is negligible. This general base ca-
talysis observed in enol-ketonization indicates that the
reaction is proceeding through base ionization of the enol
in a rapid preequilibrium step followed by rate-determin-
ing carbon protonation of the enolate ion by the conjugate
acid of the general base (Scheme 5). The rate law for this
process can be rearranged to give an expression for the
general acid catalytic coefficients of the enolate ion
reaction, kHAk, in terms of Ka (the acidity constant of the
F IGURE 5. More O’Ferral-J encks free-energy diagram for
the (a) base-catalyzed and (b) acid-catalyzed keto-enol inter-
conversion of 2-acetylcyclohexanone.
donation to carbon of enolate ion (this is equivalent to â
) 1 - R ) 0.464 for proton removal from carbon in
enolization).
E
conjugate acid) and Ka (the acidity constant determined
This Bro¨nsted exponent points to the formation of
slightly unsymmetrical transition states. The nature of
the transition state (TS) can be deduced as follow. The
Bro¨nsted exponents can be identified with extents of
charge transfer at the TS. Then, by considering the
enolization process, â ) 0.46 means that a charge of
+0.46 has been given to the base removing the hydrogen
on carbon (e.g., a water molecule), which generates a
negative charge of -0.46 on the ACHE molecule, and that
a positive charge of +0.54 has been moved from the acid
donating a hydrogen to the carbonyl oxygen, which
generates +0.54 units of charge on the ACHE molecule
and leaves a charge of -0.54 on the water molecule from
for the enol) according to eq 10.
Ka
k
k
kHA
)
EkB
(10)
Ka
k
The calculated kHA values increase as expected with
the strength of the acid as shown in Table 4. The
Bro¨nsted exponent obtained by least-squares analysis of
the data gives R ) 0.536 ( 0.001 in ketonization proton
(35) Kresge, A. J . Pure Appl. Chem. 1964, 8, 243.
(36) Perrin, D. D. pKa Prediction for Organic Acids and Bases;
Chapman and Hall: New York, 1981.
J . Org. Chem, Vol. 68, No. 7, 2003 2687

Iglesias
which this H+ is being transferred (or +0.46 if the acid
is H3O+). The net charge on the ACHE molecule is then
-0.46 + 0.54 ) +0.08. The Scheme 3 shows the overall
transition state charge distribution. This result suggests
that the keto-enol conversion is an overall base-
catalyzed reaction with proton donation running ahead
of proton removal. A transition state positively charged
is reinforced by the general observation of that 1,3-
dicarbonyl compounds form chelate complexes with metal
cations, and, recently we postulated the existence of
nitrosyl complex intermediate in the nitrosation of 1,1,1-
trifluoro-3-(2-thenoyl)acetone,25 of 2-acetylcyclopentanone,28
and of the enol of 2-acetylcyclohexanone (see the following
paper). On the other hand, a positively charged TS for
keto-enol conversion will respond to structural changes
in the same way as ester hydrolysis, as we have found
in the study of the ester hydrolysis of ethyl-cyclohex-
anone-2-carboxylate.26
free-energies of the reactant and product corners have
not changed, but those of the other two corners, enolate
and protonated-keto at oxygen, have higher and lower
energy values, respectively. These changes will shift the
transition state in the direction shown by the arrow, away
from the enolate ion and toward the protonated keto. This
means that, the reaction mechanism of H+-catalyzed
ketonization resembles more to a consecutive mechanism
with preequilibrium protonation of the substrate, in total
accordance with the observed isotope effects (vide supra).
Con clu sion s
In alkaline medium, 2-acetylcyclohexanone undergoes
rapid ionization to give the enolate. The overall pKa was
measured as 9.85. When the alkaline solution is made
acidic, the enol tautomer is rapidly recovered with a yield
of 100%. But, subsequently, the 2-acetylcyclohexanone
enol ketonizes slowly in aqueous acid medium until to
reach a 43% enol content at equilibrium. Both ketoniza-
tion and enolization are acid and general-base catalyzed
reactions. The base catalysis is stronger than the acid
catalysis and increases with the strength of the base. The
isotopic effect, along with the Bro¨nsted exponents, de-
termined independently for each reaction, points to a
reaction mechanism in which a proton-transfer occurs in
the rate-determining step. The More O’Ferrall-J encks
free-energy diagrams constructed from Bro¨nsted expo-
nents suggest a slightly unsymmetrical transition states
bearing a net positive charge.
The More O’Ferrall-J encks free-energy diagram sug-
gested by the Bro¨nsted coefficients in keto-enolization
is given in Figure 5A, where the position of the transition
state is indicated by the symbol q. The horizontal
coordinate represents proton removal from carbon, with
a scale equal to the Bro¨nsted exponent â, and the vertical
coordinate, with its scale equal to R, represents proton
transfer to oxygen. The energies of various corners of this
diagram can be calculated from the keto-enol equilib-
rium constant, KE ) 0.72, the acidity constant of the keto,
K
KH
pKa ) 9.62, the pKa of the catalytic acid, and the pKa
of the keto isomer protonated on the oxygen. This latter
species undoubtedly is a strong acid with pKa < -5;
e.g., the numbers shown on the diagram are for a
KH
Ack n ow led gm en t. Financial support from the Di-
reccio´n General de Investigacio´n (Ministerio de Ciencia
y Tecnolog´ıa) of Spain (Project No. BQU2000-0239-C02)
is gratefully acknowledged.
catalytic acid of pKa ) 1.75 (the midpoint of our Bro¨nsted
KH
correlation) and a pKa of the protonated enol equal to
-6. When the catalyst system is replaced by H3O+
(pKa ) -1.74) and H2O, the corresponding energies are
those indicated in Figure 5B. It may be seen that the
J O026629X
2688 J . Org. Chem., Vol. 68, No. 7, 2003