Keto - Enol/enolate equilibria in the isochroman-4-one system. Effect of a β-oxygen substituent

Article abstract of DOI:10.1021/ja0112801

The enol of 1-tetralone was generated flash photolytically, and rates of its ketonization were measured in aqueous HClO₄ and NaOH solutions as well as in CH₃CO₂H, H₂PO₄^-, (CH₂OH)₃CNH₃⁺, and NH₄⁺ buffers. The enol of isochroman-4-one was also generated, by hydrolysis of its potassium salt and trimethylsilyl ether, and rates of its ketonization were measured in aqueous HClO₄ and NaOH. Rates of enolization of the two ketones were measured as well. Combination of the enolization and ketonization data for isochroman-4-one gave the keto - enol equilibrium constant pK_E = 5.26, the acidity constant of the enol ionizing as an oxygen acid pQ_a^E = 10.14, and the acidity constant of the ketone ionizing as a carbon acid pQ_a^K = 15.40. Comparison of these results with those for 1-tetralone shows that the β-oxygen substituent in isochroman-4-one raises all three of these constants: K_E by 2 orders of magnitude, Q_a^E by not quite 1 order of magnitude, and Q_a^K by nearly 3 orders of magnitude. The β-oxygen substituent also retards the rate of hydronium-ion-catalyzed ketonization by more than 3 orders of magnitude. The origins of these substituent effects are discussed.

Full text of DOI:10.1021/ja0112801

11562
J. Am. Chem. Soc. 2001, 123, 11562-11569
Keto-Enol/Enolate Equilibria in the Isochroman-4-one System. Effect
of a â-Oxygen Substituent
Y. Chiang,† A. J. Kresge,*,† Q. Meng,† R. A. More O’Ferrall,‡ and Y. Zhu†
Contribution from the Departments of Chemistry, UniVersity of Toronto,
Toronto, Ontario M5S 3H6, Canada, and UniVersity College, Belfield, Dublin 4, Ireland
ReceiVed May 24, 2001
Abstract: The enol of 1-tetralone was generated flash photolytically, and rates of its ketonization were measured
in aqueous HClO4 and NaOH solutions as well as in CH3CO2H, H2PO4-, (CH2OH)3CNH3+, and NH4+ buffers.
The enol of isochroman-4-one was also generated, by hydrolysis of its potassium salt and trimethylsilyl ether,
and rates of its ketonization were measured in aqueous HClO4 and NaOH. Rates of enolization of the two
ketones were measured as well. Combination of the enolization and ketonization data for isochroman-4-one
gave the keto-enol equilibrium constant pKE ) 5.26, the acidity constant of the enol ionizing as an oxygen
acid pQEa ) 10.14, and the acidity constant of the ketone ionizing as a carbon acid pQKa ) 15.40. Comparison
of these results with those for 1-tetralone shows that the â-oxygen substituent in isochroman-4-one raises all
three of these constants: KE by 2 orders of magnitude, QE by not quite 1 order of magnitude, and QK by
a
a
nearly 3 orders of magnitude. The â-oxygen substituent also retards the rate of hydronium-ion-catalyzed
ketonization by more than 3 orders of magnitude. The origins of these substituent effects are discussed.
Although the enol isomers of simple aldehydes and ketones
aldose-ketose isomerases, a process known to involve
enediol/enediolate intermediates; a well-known example is
the conversion of dihydroxyacetone phosphate into glycer-
aldehyde phosphate, a necessary step in the metabolism of
glucose.3
We generated the enol of isochromanone by hydrolysis of
its potassium salt, itself prepared by treating isochroman-4-one
with potassium hydride in an aprotic solvent, eq 3, and also by
hydrolysis of the trimethylsilyl enol ether, eq 4.
are generally quite unstable, both kinetically and thermodynami-
cally, a variety of methods have recently been developed for
generating these substances in solution in greater than equilib-
rium amounts under conditions where they can be observed
directly and rate and equilibrium constants of their reactions
can be measured accurately. Much new information is conse-
quently now available on the chemistry of enols of simple
aldehydes and ketones where enol or enolate ion is the only
functional group present in the molecule.1 Much less is known
about the effect of hetereoatom substituents on this chemistry,
and we have therefore undertaken a program of research
designed to supply some of the missing information. We have
recently reported on the effect of the â-nitrogen substituent in
the N-acetylamino-p-methylacetophenone keto-enol system, eq
1,2 and we now add to that the effect of the â-oxygen substituent
in the isochroman-4-one system, eq 2. In addition to its bearing
To have an unsubstituted system with which to compare the
chemistry of the isochroman-4-one keto-enol pair, we also
examined the 1-tetralone keto-enol system, eq 5. We generated
this enol by flash photolytic Norrish type II photoelimination
of 2-isobutyl-1-tetralone, eq 6.
(1) See, for example: The Chemistry of Enols; Rappoport, Z., Ed.;
Wiley: New York, 1990.
on enol chemistry, a â-oxygen substituent such as this is of
biological interest in connection with the mechanism of
(2) Chiang, Y.; Griesbeck, 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, 8979-8984.
(3) Fersht, A. Enzyme Structure and Mechanism, 2nd ed.; W. H.
Freeman: New York, 1985; pp 439-443. Cui, Q.; Karplus, M. J. Am. Chem.
Soc. 2001, 123, 2284-2290.
* To whom correspondence should be directed. Tel.: (416) 978-7259.
Fax: (416) 978-7259. E-mail: akresge@chem.utoronto.ca.
† University of Toronto.
‡ University College, Belfield.
10.1021/ja0112801 CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/06/2001

Keto-Enol/Enolate Equilibria in Isochroman-4-one
J. Am. Chem. Soc., Vol. 123, No. 47, 2001 11563
measurements were made using a conventional microsecond flash
system that has already been described,10 whose reaction cell was
thermostated at 25.0 ( 0.05 °C. Decay of the enol was monitored at λ
) 290 nm in acidic and buffer solutions, and decay of the enolate ion
was monitored at λ ) 300 nm in basic solutions. Observed first-order
rate constants were obtained by least-squares fitting of an exponential
function.
(b) Isochromanone Enol, Potassium Salt Method. Rate measure-
ments in perchloric acid solution were made by conventional spec-
troscopy using a Cary 2200 spectrometer whose cell compartment was
thermostated at 25.0 ( 0.05 °C. Stock solutions of the potassium enolate
were prepared by treating 0.02 M solutions of isochroman-4-one in
anhydrous tetrahydrofuran with a slight excess of potassium hydride.
This operation was carried out under an argon atmosphere in a plastic
glovebag, and the stock solutions were stored under argon in Pierce
Reacti vials fitted with Pierce Mininert valves; samples for kinetic
measurement were withdrawn by hypodermic syringe. Reactions were
initiated by adding 7 µL aliquots of these stock solutions to 3.0 mL
portions of aqueous acid solutions contained in spectrometer cuvettes
that had been allowed to come to temperature equilibrium with the
spectrometer cell compartment. Reactions were monitored by following
the increase in isochromanone absorbance at λ ) 255 nm or the decrease
in enol absorbance at λ ) 290 nm.
Rates of ketonization in sodium hydroxide solution were too fast to
be measured by conventional spectroscopy in this way, and determina-
tions were therefore made using a High-Tech Scientific SF-S1 stopped-
flow spectrometer. Aqueous sodium hydroxide solutions were mixed
in a 1:1 ratio with an essentially aqueous solution of enol in a very
dilute buffer formed by adding 30 µL of potassium enolate stock
solution to 6.0 mL of 0.005 M aqueous acetic acid. The lifetime of the
enol in this buffer solution was sufficiently long to carry out several
stopped-flow rate measurements before the absorbance change became
too small to be useful.
Observed first-order rate constants for both acid and base solutions
were obtained by least-squares fitting of an exponential function.
(c) Isochromanone Enol, Silyl Ether Method. Rate measurements
in perchloric acid solution were made by conventional spectroscopy
using the Cary 2200 spectrometer. Reactions were initiated by adding
4 µL aliquots of ca. 0.04 M acetonitrile stock solutions of isochroman-
4-one trimethylsilyl enol ether to 2.8 mL portions of acid solution that
had been allowed to come to temperature equilibrium with the
thermostated (25.0 ( 0.05 °C) spectrometer cell compartment. The silyl
ether was not very soluble in these aqueous acid solutions, and the
mixtures had to be shaken for ca. 30 s in order to effect complete
solution. This gave sufficient time for hydrolysis of the silyl ether to
enol to go to completion, and observed first-order rate constants for
enol decay could be obtained by least-squares fitting of a single
exponential function. In some cases, however, this first-order decay
was followed by a subsequent slow downward absorbance drift, and
the least-squares fitting was consequently done using an exponential
plus linear function.
Rate measurements in sodium hydroxide solutions were again done
by stopped-flow spectroscopy, mixing aqueous silyl ether and sodium
hydroxide solutions in a 1:1 ratio. Measurements were limited to sodium
hydroxide concentrations equal to or greater than 0.01 M, where
hydrolysis of the silyl ether, which is catalyzed by hydroxide ion, is
considerably faster than ketonization of the enol, whose rate in this
region is independent of hydroxide ion concentration. Observed first-
order rate constants were obtained by least-squares fitting of a single-
exponential function.
Experimental Section
Materials. Isochroman-4-one was prepared by Dieckman cycliza-
tion of ethyl o-carbethoxybenzyloxyacetate followed by saponification
and decarboxylation of the â-keto ester so produced,4 and also by Perkin
ring closure of o-carboxybenzyloxyacetic acid.4a,5
Isochroman-4-one silyl enol ether was obtained by treating
isochroman-4-one with the trimethylchlorosilane-sodium iodide-
tertiary amine reagent.6 To a stirred solution of 0.86 g of sodium iodide
in 10 mL of anhydrous acetonitrile were added 0.58 g of triethylamine
(dried before use) and 0.70 g isochroman-4-one. When these substances
had dissolved, 0.63 g of trimethylchlorosilane was added, which caused
a white precipitate to form immediately. The resulting mixture was
stirred at room temperature for 3 h. It was then extracted with n-pentane,
and the extracts were concentrated to give 0.82 g (79%) of the silyl
1
ether as a yellow oil. H NMR (200 MHz, CDCl3): δ/ppm ) 7.29-
7.01 (m, 4H), 6.45 (s, 1H), 4.93 (s, 2H), 0.25 (s, 9H). 13C NMR (50
MHz, CDCl3): δ/ppm ) 136.7, 133.1, 130.9, 129.1, 128.2, 127.4, 123.7,
119.5, 68.5, 0.27. HRMS: m/e ) 220.0920 (calcd), 220.0919 (found).
2-Isobutyl-1-tetralone was synthesized by alkylating 2-ethoxy-
carbonyl-1-tetralone7 using a published procedure8 and then saponifying
and decarboxylating the product obtained. A solution of sodium
ethoxide was prepared by dissolving 0.24 g of sodium in 30 mL of
absolute ethanol, and to this was added a solution of 2.00 g of
2-ethoxycarbonyl-1-tetralone in 10 mL of absolute ethanol. This mixture
was heated under reflux for 0.5 h. It was then cooled to room
temperature, 1.6 mL of isobutyl iodide was added, and the resulting
mixture was heated under reflux for 20 h. The reaction mixture was
then poured into water, and the resulting mixture was extracted with
ether. The ether was removed from this extract, and the residue was
purified by flash chromatography on silica gel using 5% ethyl acetate
in hexane as the eluent.
The 2-isobutyl-2-ethoxycarbonyl-1-tetralone so obtained was then
saponified and decarboxylated by heating under reflux for 3 h 1.00 g
of this substance in 20 mL of 95% aqueous ethanol to which 2.0 g of
potassium hydroxide had been added. The reaction mixture was then
cooled, and water was added. The resulting mixture was extracted with
chloroform, and the extract was washed with water and brine and then
dried over magnesium sulfate. The solvent was removed from the dried
extract, and the residue was purified by flash chromatography on silica
gel using 2% ethyl acetate in hexane as the eluent. This produced 0.50
1
g (68%) of 2-isobutyl-1-tetralone, whose H NMR spectrum agreed
with a published report.9
All other materials were the best available commercial grades.
Ketonization Rate Measurements. (a) 1-Tetralone Enol. The enol
of 1-tetralone was generated flash photolytically by Norrish type II
photoelimination of 2-isobutyl-1-tetralone according to eq 6. Rate
Enolization Rate Measurements. (a) Basic Solution. Rates of
enolization of 1-tetralone and isochroman-4-one were measured in
sodium hydroxide solutions using iodine to scavenge the enols as they
formed. The reactions were carried out under first-order conditions using
an excess of iodine ((1-3) × 10-5 M) over ketone (0.5 × 10-5 M) in
the presence of iodide ion ((3-15) × 10-5 M). The reactions were
monitored by following the absorbance change at λ ) 230 nm using
the Cary 2200 spectrometer with cell compartment thermostated at 25.0
(4) (a) Anzalone, L.; Hirsch, J. A. J. Org. Chem. 1985, 50, 2128-2133.
(b) Normant-Chefnay, C. Bull. Soc. Chim. Fr. 1971, 4, 1351-1362.
(5) Nilsen, B. P.; Undheim, K. Acta Chem. Scand. B 1976, 30, 619-
623. Thibault, J. Ann Chim. 1971, 6, 381-390.
(6) Duboudin, F.; Moulines, F.; Babot, O.; Dunogues, J. Tetrahedron
1987, 43, 2075-2088.
(7) Bowman, W. R.; Westlake, P. J. Tetrahedron 1992, 48, 4027-4038.
(8) Chini, M.; Crotti, P.; Macchia, F. J. Org. Chem. 1989, 4, 3930-
3936.
(9) Henin, F.; M’Boungou-M’Passi, A.; Muzart, J.; Pete, J.-P. Tetra-
hedron 1994, 50, 2849-2864.
(10) Chiang, Y.; Hojatti, M.; Keeffe, J. R.; Kresge, A. J.; Schepp, N. P.;
Wirz, J. J. Am. Chem. Soc. 1987, 109, 4000-4009.

11564 J. Am. Chem. Soc., Vol. 123, No. 47, 2001
Chiang et al.
( 0.05 °C. Observed first-order rate constants were obtained by least-
squares fitting of a single-exponential function.
(b) Acid Solution. Rates of enolization of isochroman-4-one were
also measured in acid solutions, but, because the reaction in these media
was very much slower than that in base, an initial rate method that
consumed only about 0.2% of the ketone substrate (initial concentration
) 1.37 × 10-3 M) was used. The reaction was followed by bromine
scavenging of the enol as it formed, and because the acid used was
hydrogen bromide in the concentration range 0.9-2.7 M, most of the
bromine was complexed as the Br3- ion. Measurements were made by
-
monitoring the decrease in Br3 absorbance at λ ) 320 nm using the
Cary 2200 spectrometer with cell compartment thermostated at 25.0
( 0.05 °C.
In a typical procedure, 3.00 mL of aqueous HBr was allowed to
come to temperature equilibrium with the spectrometer cell compart-
ment, and a small amount (less than 2 µL) of bromine water of
concentration sufficient to give an absorbance reading of 0.1-0.3 at λ
) 320 nm was added. The enolization reaction was then initiated by
adding 20.0 µL of a 0.200 M acetonitrile solution of isochromanone.
This produced an initial small fast drop in absorbance, probably due
to bromination of the enol initially in equilibrium with the keto form
of isochromanone, followed by a slower linear decay; the latter was
monitored for about 60 min.
Figure 1. Titration curve for the ionization of isochroman-4-one as a
carbon acid in aqueous solution at 25 °C.
and the fit of log I to different basicity functions was examined.
Three functions were tried: an H- scale based on neutral indole
indicators,14 an H) scale based on indoles bearing carboxylate
groups,14 and an H) scale based on aromatic amines bearing
carboxylate and sulfonate groups.15 The H- scale based on
indoles proved to be the most appropriate; it gave a plot of log
I vs H- whose slope was unity (1.02 ( 0.05), whereas the other
basicity functions produced slopes significantly different from
unity (0.83 ( 0.04 and 0.86 ( 0.04). The data were therefore
analyzed using the titration curve expression shown in eq 9, in
These zero-order rates of absorbance decay, ∆A/∆t, were converted
into first-order enolization rate constants, kE, with the aid of eq 7, in
kE ) {(∆A/∆t)/[ketone]}/ꢀeff
(7)
which ꢀeff is an effective extinction coefficient that relates the total
stoichiometric bromine concentration to the absorbance of Br3-. This
effective extinction coefficient depends on the true extinction coefficient
-
of Br3 at λ ) 320 nm, ꢀ320, the equilibrium constant for the Br2 +
Br- a Br3- association reaction, Kass, and the bromide ion concentra-
A ) (ABH10(-H ) + ABKa)/(10(-H ) + Ka)
(9)
-
-
tion, [Br-], as shown in eq 8. This relationship works well in dilute
ꢀeff ) ꢀ320Kass[Br-]/(1 + Kass[Br-])
(8)
which the H- scale serves as a measure of the basic strength of
the solutions employed. Figure 1 shows that the data conformed
to this relationship well; least-squares analysis produced the
result Ka ) (2.73 ( 0.19) × 10-16 M, pKa ) 15.564 ( 0.030.
Basicity functions such as the H- scale used here are anchored
to an infinitely dilute standard state, and the acidity constant
determined here is therefore a thermodynamic value applicable
at zero ionic strength. This acidity constant was also determined
from kinetic measurements described below that were carried
out at an ionic strength of 0.10 M. Comparison of the two values
therefore requires application of activity coefficients, and use
of f ) 0.83 for the hydrogen ion16 plus f ) 0.79 for the enolate
ion17 leads to Qa ) (4.17 ( 0.29) × 10-16 M, pQa ) 15.380 (
0.030.19
Ketonization of 1-Tetralone Enol. HPLC analysis of spent
ketonization reaction mixtures using tetralone enol generated
by flash photolysis of 2-isobutyl-1-tetralone (cf. eq 6) showed
the presence of only 1-tetralone and unphotolyzed isobutyl-
tetralone. The flash photolysis and ketonization reactions are
therefore clean processes, giving only the intended products.
Rates of the ketonization reaction were measured in dilute
aqueous (H2O) perchloric acid and sodium hydroxide solutions
and also in aqueous (H2O) CH3CO2H, H2PO4-, (CH2OH)3CNH3+,
bromide ion solutions, but, at the high HBr concentrations used in the
present study ([Br-] ) 0.9-2.7 M), it breaks down because of medium
effects on ꢀ320 and/or Kass. We therefore determined ꢀeff for these
solutions empirically by measuring the absorbance at λ ) 320 nm of
solutions containing the same constant stoichiometric concentration of
bromine both in dilute solution with [Br-] ) 0.100 M, A0.1, and also
in more concentrated solutions with variable [Br-] ()x), Ax. These data
are summarized in Table S1.11 The ratio Ax/A0.1 proved to be a linear
function of x over the concentration range employed for the present
rate measurements: Ax/A0.1 ) 1.519 + 0.1125x. This relationship,
together with the dilute solution values ꢀ320 ) 5155 M-1 cm-1 12 and
Kass ) 17.0,12,13 was then used to calculate ꢀeff at the bromide ion
concentrations needed to evaluate the rate data.
Acidity Constant Determinations. The acidity constant of iso-
chroman-4-one ionizing as a carbon acid was determined by monitoring
the reversible UV spectral change that this substance undergoes upon
ionization. Measurements were made at λ ) 320 nm in potassium
hydroxide solutions at a constant stoichiometric concentration (1.26 ×
10-4 M) of isochromanone. The Cary spectrometer operating at 25.0
( 0.05 °C was used.
Results
Isochroman-4-one Acidity Constant. The extent of ioniza-
tion of isochroman-4-one as a carbon acid was determined by
measuring the absorbance of the enolate ion product in a series
of potassium hydroxide solutions over the concentration range
[KOH] ) 0.1-10 M. The data so obtained, summarized in Table
S2,11 were transformed into values of I ()[enolate]/[ketone]),
(14) Yagil, G. J. Phys. Chem. 1967, 71, 1034-1044.
(15) Halle, J.-C.; Terrier, F.; Schall, R. Bull. Soc. Chim. Fr. 1969, 4569-
4575.
(16) Bates, R. G. Determination of pH Thoery and Practice; Wiley: New
York, 1973; p 49.
(17) This activity coefficient was calculated using a modified form of
the Davies equation that works well for bulky organic anions such as
(11) Supporting Information; see paragraph at the end of this paper
regarding availability.
(12) Keeffe, J. R., unpublished work.
substituted phenols.18
(18) King, E. J. Acid-Base Equilibria; MacMillan: New York, 1965; p
101.
(13) Griffith, R. O.; McKeown, A.; Winn, A. G. Trans. Faraday Soc.
1932, 28, 1901-107.
(19) This is a concentration quotient, applicable at ionic strength ) 0.10
M.

Keto-Enol/Enolate Equilibria in Isochroman-4-one
J. Am. Chem. Soc., Vol. 123, No. 47, 2001 11565
This “uncatalyzed” portion of the rate profile is followed by
a region of apparent hydroxide ion catalysis, which may be
attributed to equilibrium ionization of the enol to enolate ion
followed by carbon protonation of that by a water molecule.
This mechanism produces H+ in the rapid equilibrium step but
does not use it in the rate-determining step, which gives an
overall process whose rate is inversely proportional to [H+] or
directly proportional to [HO-]. At sufficiently low acidities, the
position of the enol ionization equilibrium switches from enol
to enolate ion; the latter then becomes the initial state of the
ketonization reaction, and, with water still as the carbon-
protonating agent, this leads to the final “uncatalyzed” plateau
of the rate profile.
The reaction scheme representing these mechanisms is shown
in eq 11, and the rate law that applies is given in eq 12, in
Figure 2. Rate profile for the ketonization of the enol of 1-tetralone
in aqueous solution at 25 °C.
+
and NH4 buffers. The data so obtained are summarized in
Tables S3-S5.11
The measurements in buffers were made in series of solutions
of constant buffer ratio and constant ionic strength (0.10 M),
and therefore constant [H+], but varying buffer concentration.
Observed first-order rate constants were found to increase
linearly with increasing buffer concentration, in keeping with
the buffer dilution rate expression of eq 10, and the data were
therefore analyzed by linear least-squares fitting of this equation.
which kuc is equal either to kKo or to kH′K+QE. Least-squares
a
kobs ) kHK+[H+] + kuc + ko′KQaE(QEa + [H+])
(12)
fitting of this expression gave kKH+ ) (1.80 ( 0.03) × 102 M-1
s-1, kKo ) (1.10 ( 0.08) × 10-2 s-1, kH′K+ ) (7.25 ( 0.60) × 108
M-1 s-1, k′oK ) (7.43 ( 0.13) × 102 s-1, and QEa ) (1.52 (
0.06) × 10-11 M, pQEa ) 10.82 ( 0.02.19
kobs ) ko + kcat[buffer]
(10)
The zero-buffer-concentration intercepts, ko, obtained in this way
were then combined with the rate constants measured in
perchloric acid and sodium hydroxide solutions to construct the
rate profile shown in Figure 2. Values of [H+] needed for this
purpose were obtained by calculation, using thermodynamic
acidity constants of the buffer acids from the literature and
activity coefficients recommended by Bates.16
Some rates of ketonization of 1-tetralone enol were also
measured in dilute D2O solutions of perchloric acid and sodium
hydroxide. The results, summarized in Tables S3 and S4,11 when
combined with their H2O counterparts, give the following
isotope effects: kKH+/kK+ ) 4.56 ( 0.12 and (k′K)H O/(k′K)D O
)
2
2
D
o
o
7.80 ( 0.35. These isotope effects provide good support for
the interpretation of the rate profile given above. The isotope
effect on kKH+, for example, is large, as expected for rate-
determining hydrogen transfer from the hydronium ion to a
substrate of moderate reactivity. Such isotope effects contain
an inverse (kH/kD < 1) secondary component that offsets the
normal (kH/kD > 1) primary component,22 and the result obtained
here is, in fact, a near maximum value;23 similar isotope effects
on this rate constant have been found for other enol ketonization
reactions.20 The isotope effect on k′oK also contains a secondary
component, but this time in the normal direction,22 and that
serves to augment the primary component, making this isotope
effect much larger than that on kKH+. The striking difference in
magnitude between these two isotope effects is, in fact, a
hallmark of enol ketonization reactions. Similarly large isotope
effects on k′oK have been measured for the carbon protonation
by H2O of the enolate ions of isobutyrophenone (kH/kD ) 7.5),24
mandelic acid (kH/kD ) 6.9),25 methyl mandelate (kH/kD ) 7.7),26
and 4-hydroxyisochroman-3-one (kH/kD ) 7.8).27
This rate profile is characteristic of enol ketonization reac-
tions. It may be understood in terms of the generally accepted
mechanism for this reaction, which consists of rate-determining
protonation of the enol or the enolate ion on their â-carbon
atoms by all available acids.20 Since this profile refers to reaction
through solvent-related species only, the available acids will
be the hydronium ion, written here as H+, and the solvent water
itself.
The acid-catalyzed portion of this rate profile at its high
acidity end then represents â-carbon protonation of the un-
ionized enol by H+. This is followed by an “uncatalyzed” region,
which could be due either to carbon protonation of un-ionized
enol by H2O or to equilibrium ionization of the enol to the very
much more reactive enolate ion21 followed by carbon protona-
tion of that by H+; the latter route produces H+ in a rapid
equilibrium step and then uses it up in the rate-determining step,
to give an overall process whose rate is independent of H+
concentration. The second of these two alternative reaction
routes may sometimes be ruled out because the data require a
rate constant for carbon protonation of the enolate ion that is
greater than the encounter-controlled limit, but that is not so in
the present case (vide infra).
(22) Kresge, A. J.; More O’Ferrall, R. A.; Powell, M. F. In Isotopes in
Organic Chemistry; Buncel, E., Lee, C. C., Eds.; Elsevier: New York, 1987;
Vol. 7, Chapter 4.
(23) Kresge, A. J.; Sagatys, D. S.; Chen, H. L. J. Am. Chem. Soc. 1977,
99, 7228-7233.
(20) Keeffe, J. R.; Kresge, A. J. In The Chemistry of Enols; Rappoport,
Z., Ed.; Wiley: New York, 1990; Chapter 7.
(21) Pruszynski, P.; Chiang, Y.; Kresge, A. J.; Schepp, N. P.; Walsh, P.
A. J. Phys. Chem. 1986, 90, 3760-3766. Chiang Y.; Kresge, A. J.;
Santabella, J. A.; Wirz, J. J. Am. Chem. Soc. 1988, 110, 5506-5510.
(24) Chiang, Y.; Kresge, A J.; Walsh, P. A. Z. Naturforsch. A 1988, 44,
406-412.
(25) Chiang, Y.; Kresge, A. J.; Popik, V V.; Schepp, N. P. J. Am. Chem.
Soc. 1997, 119, 10203-10212.

11566 J. Am. Chem. Soc., Vol. 123, No. 47, 2001
Chiang et al.
Table 1. Summary of Catalytic Coefficients for Ketonization of
the Enol of 1-Tetralone in Aqueous Solution at 25 °Ca
catalyst
H3O+
kK/M-1 s-1 b
k′K/M-1 s-1 c
1.80×102
7.25 × 108
5.64 × 107
1.07 × 106
1.62 × 105
2.10 × 104
(7.43/55.5) ×102
CH3CO2H
9.82 × 10-1
-
H2PO4
+
(CH2OH)3CNH3
+
NH4
H2O
a Ionic strength ) 0.10 M. b Rate constant for carbon protonation
of the un-ionized enol. c Rate constant for carbon protonation of the
enolate ion.
The buffer catalytic coefficients, kcat, obtained by analyzing
the buffer data according to eq 10, were separated into their
general acid, kHA, and general base, kB, components with the
aid of eq 13, in which fA is the fraction of buffer present in the
acid form. Both general acid and general base catalysis were
Figure 3. Brønsted plot for the ketonization of 1-tetralone enolate ion
in aqueous solution at 25 °C.
kcat ) kB + (kHA - kB)fA
(13)
the enolate ion reaction determined from the rate profile, despite
the fact that these catalysts make up a rather mixed group of
different charge types. Linear least-squares analysis of the data
gives the Brønsted coefficient R ) 0.47 ( 0.06. This somewhat
small value is consistent with the moderately rapid nature of
these ketonization reactions and their consequent rather early,
reactant-like transition states.
found for the acetic acid buffers, but the other buffers gave only
general base catalysis.
This pattern of buffer catalysis provides further support for
the interpretation of the rate profile of Figure 2 advanced above.
The acidity of the acetic acid buffers used corresponds to the
end of the acid-catalyzed region and the beginning of the central
“uncatalyzed” plateau of the rate profile, where rate-determining
carbon protonation of both enol and enolate ion by H+ is taking
place. The buffer analogues of these H+ reactions are rate-
determining carbon protonation of the enol by the buffer acid,
eq 14, leading to general acid catalysis, and buffer-base-assisted
Ketonization of Isochroman-4-one Enol. HPLC analysis of
spent ketonization reaction mixtures using isochromanone enol
generated by hydrolysis of the potassium enolate (eq 3) or the
trimethylsilyl ether (eq 4) showed that isochroman-4-one was
the principal product formed when these reactions were carried
out in dilute aqueous perchloric acid or sodium hydroxide
solutions. A minor amount of a second unidentified substance,
+
however, was formed as well. In HCO2H, H2PO4-, and NH4
buffers, on the other hand, apparently comparable amounts of
isochromanone and this second substance were produced, and
only rate data collected in HClO4 and NaOH solutions were
consequently used to characterize the enol ketonization reaction.
The results obtained are summarized in Tables S6 and S7.11
The measurements in aqueous (H2O) HClO4 solution were
carried out over the concentration range [HClO4] ) 0.01-0.1
M. Observed first-order rate constants increased linearly with
increasing acid concentration, and linear least-squares analysis
ionization of the enol to enolate ion plus a molecule of buffer
acid followed by rate-determining carbon protonation of the
enolate by buffer acid, eq 15, leading to general base catalysis.
gave kKH ) (9.55 ( 0.49) × 10-2 M-1 s-1 for enol generated
+
by the potassium salt method, in good agreement with kHK
)
+
(8.80 ( 0.29) ×10-2 M-1 s-1 for enol generated from the silyl
ether. The weighted average of these results gives kKH+ ) (9.00
( 0.25) × 10-2 M-1 s-1 as the best value of this rate constant.
Some rate measurements were also made in D2O solutions of
perchloric acid, again over the concentration range [DClO4] )
0.01-0.1 M, and combination of those results with their H2O
The acidity of the remaining buffers, on the other hand,
corresponds to the rate profile region of apparent hydroxide ion
catalysis, where only the analogue of eq 15 is taking place, and
only general base catalysis is observed in these solutions.
The general base catalytic coefficients for the reaction of eq
15 may be formulated as the equilibrium constant for the initial
step, which is equal to the ratio of the acidity constants of the
counterparts gives the isotope effect kKH+/kK ) 3.92 ( 0.20.
+
D
enol, QEa , and the buffer acid, QaHA, times the rate constant, k′K
,
The rate measurements in aqueous (H2O) sodium hydroxide
solutions gave observed first-order rate constants that remained
invariant over the concentration range used ([NaOH] ) 0.008-
0.1 M), as expected for the ketonization via carbon protonation
of the enolate ion by H2O under conditions where enolate is
the principal enol species, and as represented by the final plateau
of the rate profile for ketonization of tetralone enol shown in
Figure 2. Best values of rate constants were therefore determined
as simple averages of all observed values. Once again, the result
obtained using enol generated by the potassium salt method,
k′oK ) 10.4 ( 0.1 s-1, was consistent with that obtained using
enol generated from the silyl ether, k′oK ) 11.2 ( 0.1 s-1. The
HA
for the rate-determining step: kB ) (QE/QHA)k′K . Since both
a
a
HA
QEa and QHa A are known, k′K may be evaluated. The results
HA
obtained in this way are listed in Table 1.
General acid catalytic coefficients for rate-determining proton-
transfer reactions such as these enolate ion ketonizations are
expected to conform to the Brønsted relation. Figure 3 shows
that this is, indeed, the case for the rate constants of Table 1,
taken together with the H+ and H2O catalytic coefficients of
(26) Chiang, Y.; Kresge, A. J.; Schepp, N. P.; Xie, R.-Q. J. Org. Chem.
2000, 65, 1175-1180.
(27) Chiang, Y.; Eustace, S. E.; Jefferson, E. J.; Kresge, A. J.; Popik V.
V.; Xie, R.-Q. J. Phys. Org. Chem. 2000,13, 461-467.
average of these values is k′oK ) 10.8 ( 0.5 s-1
.

Keto-Enol/Enolate Equilibria in Isochroman-4-one
J. Am. Chem. Soc., Vol. 123, No. 47, 2001 11567
Ketonization rate measurements were also made in D2O
solutions of sodium hydroxide, over the concentration range
[NaOD] ) 0.01-0.07 M. Once again, observed first-order rate
constants did not change with changing sodium hydroxide
concentration, and combination of the average value of this rate
constant with its H2O counterpart gave the isotope effect
(k′K)H O/(k′K)D O ) 8.17 ( 0.42.
These isotope effects determined in acid and base solution
are strikingly similar to those obtained for the corresponding
reactions of tetralone enol in acid and base, which, as pointed
out above, are fully consistent with expectation for enol
ketonization in these solutions and provide strong support for
the formulation of the reactions observed as ketonization. The
isotope effect similarity therefore adds strong support to the
assignment of the isochromanone enol reactions as ketonizations
as well, despite the fact that they give a minor amount of a
second unknown product.
solutions. Because these reactions were slow, concentrated acid
solutions over the range [HBr] ) 0.9-2.7 M were used. The
data so obtained are summarized in Table S10.11
Analysis of data obtained in concentrated acid solutions
requires the use of an acidity function, and such analysis was
done here by the Cox-Yates method28 employing the Xo
function,29 which appears to be the best scale currently available
for this purpose.30 This produced the hydronium ion catalytic
2
2
o
o
coefficient for enolization kEH ) (4.94 ( 0.30) × 10-7 M-1
+
s-1
.
This result can be used, in combination with the hydronium
ion catalytic coefficient for ketonization of isochromanone enol,
to provide the keto-enol equilibrium constant for the iso-
chromanone system. Formulation of the equilibrium as shown
in eq 17 leads to the expression KE ) kE+/kK+, and insertion of
H
H
Enolization of Tetralone. Rates of enolization of 1-tetralone
were measured by iodine scavenging in aqueous (H2O) sodium
hydroxide solutions over the concentration range [NaOH] )
0.02-0.1 M. The data so obtained are summarized in Table
S8.11
the values of the rate constants into this expression gives KE )
(5.49 ( 0.37) × 10-6, pKE ) 5.26 ( 0.03.
Observed first-order enolization rate constants increased
linearly with increasing base concentration, and least-squares
analysis gave concordant results for measurements made at two
different initial iodine concentrations: kEHO- ) (3.46 ( 0.05) ×
Another estimate of KE using rate constants determined in
basic solution according to the isochromanone analogue of eq
16 could not be made because the required value of QEa was
not available. A value of this equilibrium constant for tetralone
enol was supplied by analysis of the rate profile for its
ketonization reaction (Figure 2), but a rate profile for keton-
ization of isochromanone enol could not be constructed because
formation of a second reaction product interfered with rate
measurements in buffer solutions (vide supra).
10-2 M-1 s-1 ([I2]o ) 1.33 × 10-5 M) and kHEO ) (3.69 (
-
010) × 10-2 M-1 s-1 ([I2]o ) 2.67 × 10-5 M). The weighted
average of these values is kEHO ) (3.51 ( 0.04) × 10-2 M-1
-
s-1
.
This result can be combined with the rate constant for
ketonization of tetralone enol in basic solutions to provide the
keto-enol equilibrium constant of the tetralone system. For-
mulation of the keto-enol equilibrium as shown in eq 16 leads
The basic solution rate constants can nevertheless be em-
ployed to provide an estimate of the equilibrium constant for
ionization of isochromanone as a carbon acid, QEa , using the
relationship provided by the first part of eq 16, QKa ) (kHEO
/
-
k′K)Qw. This gives QK ) (4.00 ( 0.20) × 10-16 M, pQaK
)
o
a
15.40 ( 0.02,19 in very good agreement with QKa ) (4.17 (
0.29) × 10-16 M, pQKa ) 15.38 ( 0.03, obtained directly by
measuring the extent of ionization of isochroman-4-one in
concentrated potassium hydroxide solutions (vide supra). The
weighted average of these two values then gives QKa ) (4.06 (
0.17) × 10-16 M, pQKa ) 15.39 ( 0.02,19 as the best value of
this constant.
to the expression KE ) (kE -/k′K)(Qw/QE), in which Qw is the
HO
o
a
acid ionization constant of water, and insertion of the appropriate
values of the rate and equilibrium constants into this expression
gives KE ) (4.93 ( 0.21) × 10-8, pKE ) 7.31 ( 0.02.
The first part of the relationship of eq 16 includes the
ionization of 1-tetralone as a carbon acid, and the equilibrium
constant for this reaction, QEa , may consequently be evaluated
Although a rate profile for the ketonization of isochromanone
enol could not be constructed and a value of QEa be determined
from it, this equilibrium constant can nevertheless be evaluated
using the thermodynamic cycle of eq 18. This cycle leads to
as QEa ) (kEHO-/k′K)Qw ) (7.49 ( 0.16) × 10-19 M, pQaE
)
o
18.13 ( 0.01.19
the relationship QEa ) QK/KE, and insertion of the known
Enolization of Isochromanone. Rates of enolization of
isochroman-4-one also were measured by iodine scavenging in
aqueous (H2O) sodium hydroxide solutions over the concentra-
tion range [NaOH] ) 0.02-0.1 M. The data so obtained are
summarized in Table S9.11
a
values of QaK and KE then gives QaE ) (7.39 ( 0.58) × 10-11
pQKa ) 10.13 ( 0.03.19
,
Discussion
Observed first-order rate constants again increased linearly
with increasing base concentration, and least-squares analysis
once more gave concordant results for measurements made at
Equilibria. The present results, summarized in Table 2, show
that introduction of an oxygen atom into the â-position of
1-tetralone raises the keto-enol equilibrium constant, KE, by 2
orders of magnitude, giving a â-oxygen substituent effect of
δR∆G° ) 2.8 kcal mol-1. This effect can be attributed to
destabilization of the keto isomer plus stabilization of the enol
two different initial iodine concentrations: kEHO ) (2.87 (
-
0.05) × 10-1 M-1 s-1 ([I2]o ) 1.33 × 10-5 M) and kHEO
)
-
(2.47 ( 0.07) × 10-1 M-1 s-1 ([I2]o ) 2.67 × 10-5 M). The
weighted average of these values is kEHO ) (2.73 ( 0.04) ×
-
(28) Cox, R. A.; Yates, K. Can. J. Chem. 1979, 57, 2944-2951.
(29) Cox, R. A.; Yates, K. Can. J. Chem. 1981, 59, 2116-2124.
(30) Kresge, A. J.; Chen, H. J.; Capen, G. L.; Powell, M.F. Can. J. Chem.
1983, 61, 249-256.
10-1 M-1 s-1
.
Rates of enolization of isochroman-4-one were also measured
by bromine scavenging in aqueous (H2O) hydrobromic acid

11568 J. Am. Chem. Soc., Vol. 123, No. 47, 2001
Chiang et al.
Table 2. Summary of Rate and Equilibrium Constantsa
form. The oxygen atoms of keto groups are electron-withdraw-
ing, and that gives these groups dipolar character with partial
positive charges on their keto carbon atoms, as illustrated by
the ionic resonance form shown in eq 19; such a partial positive
charge will interact unfavorably with an electronegative â-
oxygen substituent, and that will destabilize the keto isomer.
The hydroxyl groups of enols, on the other hand, are electron-
supplying, and that puts a partial negative charge on the enol
â-carbon atom, as again illustrated by the second ionic resonance
form of eq 19; this partial negative charge will interact favorably
with an electronegative â-oxygen substituent, and that will
stabilize the enol isomer. The enol will also be stabilized by
delocalization of its â-oxygen lone electron pair into the
aromatic ring. All of these effects operate to decrease the energy
difference between keto and enol isomers, and they thus serve
to increase KE.
a Aqueous solution; 25 °C; ionic strength ) 0.10 M.
) 3.7 kcal mol-1. This is because the keto form is the initial
state of this process, and both destabilization of that form as
well as stabilization of the enolate ion contribute to this
substituent effect.
The â-oxygen substituent also increases the acidity constant
of the enol, QEa , but by a much smaller amount than its effect
on KE: the substituent effect here is less than 1 order of
magnitude and amounts to only δR∆G° ) 0.9 kcal mol-1. This
is because the enol isomer rather than the keto form is the initial
state for the acid ionization process, and destabilization of the
keto form, which made a contribution to the substituent effect
on KE, no longer plays a role. Only stabilization of the partial
negative charge on the â-carbon atom of the enol by the
â-oxygen atom and delocalization of the â-oxygen lone pair
into the aromatic ring, plus corresponding interactions in the
enolate ion acid ionization product, operate here. It is likely
that the delocalization effect will be much the same in the enol
and the enolate ion, but the partial negative charge will, of
course, be greater in the negatively charged enolate ion than in
the neutral enol; the enolate ion is consequently stabilized more
than is the enol, and the result is an increase in QKa .
These â-oxygen substituent effects are in the same direction
as the substituent effects of a â-N-acetylamino group (NHAc)
found before.2 The present effects, however, are uniformly
somewhat weaker and amount to only about two-thirds of the
NHAc group effects. This difference is consistent with the facts
that the Swain-Lupton field effect substituent constant, F, for
the simple oxygen-containing group OMe, which could be taken
as a surrogate for the OCH2 moiety in isochroman-4-one, is
also less than that for NHAc and that the ratio of these two F
values is likewise about 2/3.31 The Charton field effect scale,32
however, as well as various scales cited by Hansch,33 give field
effect substituent constants for OMe and NHAc that are
substantially the same.
The â-oxygen substituent effect on QKa , the acidity constant
of the keto isomer ionizing as a carbon acid to give enolate
ion, on the other hand, is greater than the substituent effect on
KE: it is nearly 3 orders of magnitude and amounts to δR∆G°
(31) Swain, C. G.; Unger, S. H.; Rosenquist, N. R.; Swain, M S. J. Am.
Chem. Soc. 1983, 105, 492-502.
(32) Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119-251.
(33) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165-
195.

Keto-Enol/Enolate Equilibria in Isochroman-4-one
J. Am. Chem. Soc., Vol. 123, No. 47, 2001 11569
Another comparison of the acidifying effects of oxygen and
N-acetylamino substituents is provided by the simple series
water (HOH), hydrogen peroxide (HOOH), and acetohydrox-
amic acid (CH3CONHOH), whose pKa values are 15.7, 11.6,
and 9.4, respectively.34 These acidity constants give oxygen an
acidifying effect two-thirds of that of the N-acetylamino group,
just like the influence of these groups on enol acidity. The effects
in the simple series, however, are considerably stronger than
those on the enols, which is consistent with the fact that in the
enols the substituents must act through an intervening carbon-
carbon double bond and their influence is consequently attenu-
ated.
catalyzed hydrolysis of several vinyl ethers.35 Vinyl ether
hydrolysis is a rate-determining proton-transfer reaction with a
positively charged intermediate, eq 22,36 analogous to that
Kinetics. The data of Table 2 also show strong â-oxygen
substituent effects on rates of ketonization. The rate of keton-
ization of un-ionized enol catalyzed by H+, kHK+, for example,
is slowed by more than 3 orders of magnitude, giving the
substituent effect δR∆Gq ) 4.5 kcal mol-1. A substantial part
of this rate retardation must be due to stabilization of the enol
initial state by the â-oxygen atom, but another contributing factor
is provided by the two-step nature of the acid-catalyzed
ketonization of un-ionized enols, which involves a positively
charged carbonyl-oxygen-protonated intermediate, eq 20.20 The
formed in the ketonization of enols, and interaction of the
positive charge of this intermediate with the â-oxygen atom
must be a factor contributing to the â-oxygen effect on vinyl
ether hydrolysis as well. The vinyl ether hydrolysis effect is, in
fact, slightly stronger than that on enol ketonization, which is
consistent with the ability of a positively charged hydroxyl
group, such as that formed in enol ketonization, to distribute
part of its charge onto the solvent through hydrogen bonding.
Such charge distribution cannot, of course, take place from a
positively charged ether group, such as that formed in vinyl
ether hydrolysis, and that makes the effective positive charge
on the substrate in vinyl ether hydrolysis greater than that in
enol ketonization, leading to a stronger â-oxygen effect in the
former case than in the latter. This ability of positively charged
hydroxyl to distribute charge onto the solvent enhances the
electron-supplying ability of hydroxyl over that of methoxyl, a
difference that is manifest in the magnitude of the resonance
effect substituent constants of these groups, that of hydroxyl
being stronger than that of methoxyl.31-33
unfavorable interaction between this positive charge and the
electronegative â-oxygen atom then raises the energy of the
ketonization reaction’s transition state, and that lowers the rate
of reaction. This explanation is supported by the fact that the
â-oxygen substituent effect on the rate of ketonization of ionized
enol effected by proton transfer from H2O, k′oK, at δR∆Gq ) 2.5
kcal mol-1, is considerably weaker than that on kHK+. The
ionized enol reaction is a single-step process with no positive
charge buildup on the substrate,20 eq 21, and the additional
transition state interaction that contributes to the â-oxygen effect
Acknowledgment. We are grateful to the Natural Sciences
and Engineering Research Council of Canada for financial
support of this work.
Supporting Information Available: Tables S1-S10 of rate
and equilibrium data (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
on kKH is consequently absent.
+
Rate retardations effected by â-oxygen atoms similar to that
found here on kKH have been reported for the hydronium-ion-
JA0112801
+
(34) Jencks, W. P.; Regenstein, J. In Handbook of Biochemistry; Sober,
H. A., Ed.; Chemical Rubber Co.: Cleveland, OH, 1970; p J-189. The pKa’s
of H2O and H3O+ are interchanged in this compilation.
(35) Kresge, A. J.; Yin, Y. J. Phys. Org. Chem. 1989, 2, 43-50. Kresge,
A. J.; Leibovitch, M. J. Org. Chem. 1990, 55, 5234-5236.
(36) Kresge, A. J. Acc. Chem. Res. 1987, 20, 364-370.