Transition state imbalance in proton transfer from phenyl ring-substituted 2-tetralones to acetate ion

Article abstract of DOI:10.1021/ja990070+

Rate constants for the acetate ion-catalyzed ketonization of phenyl-substituted 2-tetralone enols have been determined by stopped-flow UV spectroscopy. From these rate constants and the keto - enol equilibrium constants, the rate constants (k_-2) for enolization were calculated. A Bronsted plot of these rate constants (log k_-2) vs the acidity of the appropriate 2-tetralone (pK_a^K) is linear, with a slope ( - α^E) of - 0.78 ± 0.03, except for the point corresponding to 6-nitro-2-tetralone (4b). Rate constants for the ionization of 2-tetralone by substituted acetates were determined directly by NMR, giving a corresponding Bronsted β^E of 0.54 ± 0.03. Both the negative deviation of the point for 4b from the correlation line for α^E and the inequality between α^E and β^E indicate an imbalanced transition state for the proton abstraction of 2-tetralone by acetate ion. This reaction is impeded by a thermodynamic barrier of 11 kcal/mol, along with an intrinsic kinetic barrier of 14 kcal/mol. A comparison of the transition states for proton abstraction of 2-tetralone by hydroxide ion and by acetate ion shows similar transition state imbalance and intrinsic kinetic barriers for both reactions. The relevance of these results to the mechanism of enzymatic acceleration of enolization is discussed.

Full text of DOI:10.1021/ja990070+

6220
J. Am. Chem. Soc. 1999, 121, 6220-6225
Transition State Imbalance in Proton Transfer from Phenyl
Ring-Substituted 2-Tetralones to Acetate Ion
Xudong Yao, Mark A. Gold, and Ralph M. Pollack*
Contribution from the Laboratory for Chemical Dynamics, Department of Chemistry and Biochemistry,
UniVersity of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250, and
Center for AdVanced Research in Biotechnology, 9600 Gudelsky DriVe, RockVille, Maryland 20850
ReceiVed January 5, 1999. ReVised Manuscript ReceiVed April 16, 1999
Abstract: Rate constants for the acetate ion-catalyzed ketonization of phenyl-substituted 2-tetralone enols have
been determined by stopped-flow UV spectroscopy. From these rate constants and the keto-enol equilibrium
constants, the rate constants (k-2) for enolization were calculated. A Brønsted plot of these rate constants (log
K
k-2) vs the acidity of the appropriate 2-tetralone (pKa ) is linear, with a slope (-RE) of -0.78 ( 0.03, except
for the point corresponding to 6-nitro-2-tetralone (4b). Rate constants for the ionization of 2-tetralone by
substituted acetates were determined directly by NMR, giving a corresponding Brønsted âE of 0.54 ( 0.03.
Both the negative deviation of the point for 4b from the correlation line for RE and the inequality between RE
and âE indicate an imbalanced transition state for the proton abstraction of 2-tetralone by acetate ion. This
reaction is impeded by a thermodynamic barrier of 11 kcal/mol, along with an intrinsic kinetic barrier of 14
kcal/mol. A comparison of the transition states for proton abstraction of 2-tetralone by hydroxide ion and by
acetate ion shows similar transition state imbalance and intrinsic kinetic barriers for both reactions. The relevance
of these results to the mechanism of enzymatic acceleration of enolization is discussed.
Introduction
dienolate ion (2) and the flanking transition states by about 11
kcal/mol.6 However, although much is known about the mech-
anisms of both the enzymatic reaction and the acetate ion-
catalyzed reaction,5,6 detailed structures of the transition states
for both of these reactions have remained elusive.
Structure-reactivity correlations can be extremely valuable
in the determination of transition state structures, specifically
the charge distribution. Since this method is not applicable to
1, we have chosen the 2-tetralone system (4) as a model in which
the acidity of the R hydrogens (H-1) can be systematically varied
by phenyl ring substitution (Scheme 2). 2-Tetralone (4a)
provides a good model for the A and B rings of 5-androstene-
3,17-dione (1), and importantly, H-1 of 4a has a pKa (12.8)7
similar to that of H-4 of 1 (12.7).5
Although proton transfer from carbon atoms adjacent to
carbonyl groups (enolization) is an inherently slow reaction,1,2
this reaction is catalyzed by a variety of enzymes with
extraordinary efficiency.3 We have been interested for some time
in the isomerization of 5-androstene-3,17-dione (1) to 4-an-
drostene-3,17-dione (3), catalyzed by 3-oxo-∆5-steroid isomerase
from Pseudomonas testosteroni (3-ketosteroid isomerase, KSI,
EC 5.3.3.1),4 and the mechanism by which this enzyme enhances
the rate over the nonenzymatic reaction. A variety of experi-
ments are consistent with a mechanism of catalysis that involves
the carboxyl group of Asp-38 acting as a proton shuttle, with
the hydroxyl group of Tyr-14 and the carboxyl group of Asp-
99 functioning as electrophilic catalysts (Scheme 1).4c
In our previous studies on the deprotonation of substituted
2-tetralones by hydroxide ion, the transition state was found to
be imbalanced, with charge reorganization through resonance
lagging behind proton transfer.7 In this work, we describe
investigations into the nature of the transition state for the proton
transfer to substituted acetates, a more realistic model for the
KSI reaction. Thus, the pKa of the base can be varied, giving a
value for the Brønsted â, as well as an R value from variation
of the tetralone substituent. We compare the transition states
for enolization of 2-tetralones by hydroxide ion and by acetate
ion, and we discuss the possible relevance of these results to
enzymatic mechanisms of deprotonation of ketones.
Since KSI uses the carboxylate group of Asp-38 as the base
in the initial deprotonation of 1, we have used the reaction of
1 with acetate ion as a model to aid in the evaluation of the
efficiency of KSI.5 A comparison of the free energy profiles
for the acetate ion-catalyzed isomerization and the enzymatic
isomerization revealed that KSI stabilizes both the intermediate
(1) Keeffe, J. R.; Kresge, A. J. In The Chemistry of Enols; Rappoport,
Z., Ed.; Wiley: Chichester, England, 1990; Chapter 7.
(2) (a) Bernasconi, C. F. Acc. Chem. Res. 1992, 20, 301. (b) Bernasconi,
C. F.; Wenzel, P. J. J. Am. Chem. Soc. 1996, 118, 11446. (c) Bernasconi,
C. F. AdV. Phys. Org. Chem. 1992, 27, 119. (d) Bernasconi, C. F.; Panda,
M.; Stronach, M. W. J. Am. Chem. Soc. 1995, 117, 9206.
(3) (a) Sargent, A. L.; Rollog, M. E.; Almlo¨f, J. E.; Gassman, P. G.;
Gerlt, J. A.J. Mol. Struct. 1996, 388, 145. (b) Gerlt, J. A.; Gassman, P. G.
J. Am. Chem. Soc. 1993, 115, 11552. (c) Gerlt, J. A.; Kozarich, J. W.;
Kenyon, G. L.; Gassman, P. G. J. Am. Chem. Soc. 1991, 113, 9667.
(4) (a) Pollack, R. M.; Bounds, P. L.; Bevins, C. L. In The Chemistry of
Enones; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, England, 1989;
p 559. (b) Schwab, J. M.; Henherson, B. S. Chem. ReV. 1990, 90, 1203. (c)
Wu, Z. R.; Ebrahimian, S.; Zawrotny, M. E.; Thornburg, L. D.; Perez-
Alvarado, G. C.; Brothers, P.; Pollack, R. M.; Summers, M. F. Science
1997, 276, 415.
Experimental Section
Unless otherwise mentioned, all chemicals were reagent grade or
better and were purchased commercially and used without further
(6) Hawkinson, D. C.; Eames, T. C. M.; Pollack, R. M. Biochemistry
1991, 30, 10849.
(7) Nevy, J. B.; Hawkinson, D. C.; Blotny G.; Yao, X.; Pollack, R. M.
J. Am. Chem. Soc. 1997, 119, 12722.
(5) Zeng, B.; Pollack, R. M. J. Am. Chem. Soc. 1991, 113, 3838.
10.1021/ja990070+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 06/19/1999

Enolization of Phenyl-Substituted 2-Tetralones
J. Am. Chem. Soc., Vol. 121, No. 26, 1999 6221
Scheme 1
100 s. For reactions with half-lives >60 min, the NMR sample tubes
were incubated at 25.0 ( 0.1 °C and removed periodically to record
spectra. The exchange reactions were generally followed for greater
than four half-lives. Values for pD were obtained by adding 0.4 to the
observed pH of these D2O solutions.13 The apparent pKa values of the
buffers in D2O were determined from the pD values of the solutions
containing the acid and base forms in a 1:1 ratio.
Scheme 2
Results
General Base Catalysis of Ketonization of Substituted
2-Tetralone Enols by Acetate Ion. The enolate ions of
substituted 2-tetralones were generated as before9 by mixing
sodium hydroxide solutions (0.01-0.1 M) with an aqueous
solution of the appropriate tetralone. Since these ketones all
ionize in the pH range,9 significant quantities of the enolate
ions are formed. When these solutions are rapidly quenched in
acetate buffers, rapid protonation occurs on the enolate oxygen
producing the enol. The conversion of these enols to the
corresponding ketones was monitored by observing the loss of
absorbance due to the enol as a function of time. Analysis of
the absorbance decrease gives pseudo-first-order rate constants,
which are linearly related to the concentration of acetate ion in
the buffer. These rate constants are independent of both the
HOAc:OAc- ratio (from 3:7 to 7:3) and the pH (Figure S1,
Supporting Information), demonstrating that these reactions are
catalyzed by acetate ion with no contribution from acetic acid.
The calculated second-order rate constants for ketonization (kK)
are given in Table 1.
purification. Water was purified as previously described.8 All 2-tetra-
lones, except 6-methoxy-2-tetralone, were available from previous
work.7,9 6-Methoxy-2-tetralone was prepared according to the literature
(mp 34.0-35.5 °C, lit. mp 33.5-35 °C).10 Deuterated stock solutions
of the carboxylic acids were prepared by partial neutralization of the
acids with NaOD, followed by dilution in D2O to concentrations that
gave less than 5 atom % exchangeable protium. The pH was measured
at room temperature with a Radiometer PHM85 pH meter and a
Radiometer PHC2406 combination electrode standardized at pH 7.00
and either 4.00 or 10.00. Melting points are uncorrected. 1H NMR
spectra were recorded at 300 MHz with sodium 3-(trimethylsilyl)-1-
propanesulfonate (DSS) or tetramethylsilane (TMS) as the reference.
The rates of ketonization of substituted 2-tetralone enols were
obtained using the sequential mixing capabilities of the Hi-Tech QP/
SF 53 stopped-flow spectrophotometer. Basic solutions of the enolates
were formed by mixing an aqueous solution of the tetralone (ca. 0.5
mM), containing 6.6% (v/v) MeOH, with 0.1 or 0.01 N NaOH in a 1:1
ratio. After a delay of 0.5-2 s, these solutions were rapidly quenched
with excess buffer (acetate containing 3.3% (v/v) MeOH) in a ratio of
1:5. The first-order decay of the absorbance of the enols was monitored
for greater than 10 half-lives at 265 nm (306 nm for 6-chloro-7-nitro-
2-tetralone and 340 nm for 6-nitro-2-tetralone enol). The temperature
was maintained at 25.0 ( 0.1 °C, and the ionic strength was controlled
by NaCl (µ ) 0.16). The final pH of the solution was determined by
mixing equivalent solutions outside the stopped flow. Buffer composi-
tions were calculated from the initial concentrations of buffer and
sodium hydroxide in the enolate solution.
Rate constants for H to D exchange of 2-tetralone in the presence
of acetate buffers and 3.3% (v/v) acetonitrile-d3 (D2O, µ ) 1.0 with
KCl) were determined by NMR spectrometry at 300 or 500 MHz. All
reactions were carried out at 25.0 ( 0.1 °C regulated by a gas flow at
the probe, which was calibrated according to the method of Raiford et
al.11 Buffer was incubated for ca. 15 min at 25.0 ( 0.1 °C, and the
exchange reaction was initiated by addition of substrate in acetonitrile-
d3.12 The reaction mixture was vortexed and transferred to an argon-
flushed NMR tube, and spectra were recorded so that at least 10 time
points were collected during the first three half-lives. The spectrometer
magnet was previously shimmed using a test sample, which enabled
exchange rates to be accurately determined with half-lives as low as
Ketonization (kK) may be divided into two elementary steps:
deprotonation of the enol (k1) and protonation of the resulting
enolate (k2), according to Scheme 3. With the assumption that
protonation on the carbonyl oxygen of the enolate to regenerate
enol (k-1) is much faster than protonation on the carbon (k2),
kK is given by eq 1. Rate constants for protonation of 2-tetralone
enolates at the carbon by acetic acid (k2) and conversion of
ketones to enolate ions by acetate ion (k-2) were calculated from
eqs 2 and 3 (Table 1).
kK ) k1k2/k-1
k2 ) kKKaHOAc/KaE
(1)
(2)
(3)
k-2 ) k2KaK/KaHOAc ) kkKak/KaE
General Base Catalysis of Enolization of 2-Tetralone by
Substituted Acetates. Rate constants for the enolization of
2-tetralone were determined by NMR from the rates of
deuterium exchange of the C-1 protons in solutions of substituted
acetates in deuterium oxide. Pseudo-first-order rate constants
(kobs) for exchange were obtained from nonlinear curve fits of
(12) Stock solutions in acetonitrile-d3 were prepared daily and stored at
-20 °C, since these solutions of 2-tetralone begin to change from clear to
yellow/brown soon after mixing. The effect of this decomposition on kobs
is assumed to be unimportant, since good first-order kinetics were obtained
for all reactions and no new peaks were detectable by NMR during the
course of the reaction (up to 15 h).
(8) Hawkinson, D. C.; Pollack, R. M.; Ambulos, N. P. Biochemistry 1994,
33, 12172.
(9) Yao, X.; Pollack, R. M. Can. J. Chem. 1999, 77, 634-638.
(10) Sims, J. J.; Selman, L. H.; Cadogan, M. Org. Synth. 1971, 51, 109.
(11) Raiford, D. S.; Fisk, C. L.; Becker, E. D. Anal. Chem. 1979, 51,
2050.
(13) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188.

6222 J. Am. Chem. Soc., Vol. 121, No. 26, 1999
Yao et al.
Table 1. Rate Constants for Enolization of 2-Tetralones by Acetate Ion
b
c
substituent
σ- a
pKaK b
pKaE b
pKE
kK (M-1 s-1
)
k2 × 10-6 (M-1 s-1
)
k-2 × 103 (M-1 s-1
)
6-MeO
7-Me
H
6-Cl
6-I
-0.27
-0.17
0
0.19
0.27
0.37
0.71
0.90
1.42
1.23
13.27((0.05)
12.94((0.01)
12.83((0.01)
12.51((0.03)
12.36((0.02)
12.30((0.07)
11.76((0.02)
11.34((0.03)
10.34((0.06)
10.09((0.02)
9.64((0.05)
9.52((0.01)
9.44((0.04)
9.31((0.02)
9.23((0.02)
9.09((0.06)
8.81((0.01)
8.56((0.01)
8.14((0.03)
7.69((0.02)
3.63((0.10)
3.42((0.02)
3.39((0.05)
3.20((0.05)
3.13((0.04)
3.21((0.13)
2.95((0.03)
2.78((0.04)
2.20((0.09)
2.40((0.04)
19.7((0.1)
33.5((0.4)
33.6((0.1)
42.8((0.4)
49.9((1.4)
58.5((0.5)
84.9((0.4)
102((1)
2.01((0.31)
2.60((0.16)
2.17((0.27)
2.05((0.16)
1.99((0.20)
1.69((0.26)
1.29((0.06)
0.87((0.05)
0.60((0.01)
0.20((0.02)
4.6((1.2)
12.8((0.8)
13.7((1.7)
27.0((3.6)
37.0((4.7)
36((13)
7-Cl
7-NO2
6-Cl-7-NO2
5,7-(NO2)2
6-NO2
95((7)
170((20)
1170((290)
680((73)
186((2)
171((2)
a The σ- for 5,7-(NO2)2 is calculated as 2 times the σm of a nitro group; the σ- for 6-Cl-7-NO2 as the sum of the σm of a nitro group and the σp
of a chloro group. The σ values are from the following: Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165. b Reference 9. Either concentration
pKa’s based upon the concentration of all species or mixed pKa’s based upon concentrations of solute and activity of hydronium ion have been used
in the same correlations. c The pKa of HOAc used for calculation is measured as 4.63 ( 0.01 at µ ) 0.16 M, 21.6 °C, and 3.3% MeOH.
Scheme 3
the integrated peak area for the exchangeable C-1 protons
normalized to the peak area for the nonexchanging C-3 protons
vs time. Semilogarithmic plots of peak area versus time for
proton exchange were linear for greater than three half-lives.
Figure 1. Plot of log k-2 (M-1 s-1) vs pKaK for proton abstraction by
acetate ion from substituted 2-tetralones (3.3% (v/v) methanol, µ )
0.16, 25.0 ( 0.1 °C).
Where duplicate measurements of kobs were performed, these
values agree within (5%. Slopes of linear plots of kobs vs the
total concentration of buffer give the rate constants for the
buffer-catalyzed reaction. Plots of these slopes vs the mole
fraction of the base give rate constants for proton abstraction
by each base (kB) (Table S1, Supporting Information). Since
these rate constants are determined from measurements of
exchange of both R hydrogens with deuterium, they need to be
statistically corrected (multiplied by a factor of 2) to be
compared to the k-2 values in Table 1. For the deprotonation
of 2-tetralone by acetate ion, for which the rate constants were
determined by both methods, the corrected exchange rate
constant (kexcorr) is 1.05 × 10-2 M-1 s-1, in good agreement
with the rate constant from stopped-flow UV spectral measure-
is a general phenomenon for reactions that involve delocalization
of charge into a π-acceptor.2a,c,16
We have previously analyzed the charge distribution of the
transition state for abstraction of the R proton (H-1) of
2-tetralone by hydroxide ion.7 This transition state is imbalanced,
with charge delocalization into the phenyl ring and into the
carbonyl oxygen lagging behind C-H bond cleavage and charge
transfer from the hydroxide ion. An estimate of the charge
distribution in the transition state was based upon the slope of
the Brønsted plot for this reaction (R ) 0.60), the negative
deviation of the point for 6-nitro-2-tetralone (4b) from this plot,
and the charge distribution of the fully formed anion.7 These
results indicate a decreased resonance effect in the transition
state relative to the inductive effect. We now have extended
this work to the reaction with substituted acetate ions, which
allows variation of the pKa of the base, giving a value for the
Brønsted â, in addition to an R value.
ments (k-2) of 1.37 × 10-2 M-1 s-1
.
Discussion
Nature of the Transition State for Proton Transfer from
2-Tetralones to Acetate Ion. It has been known for a long time
that ionization of carbon acids generally occurs much more
slowly than ionization of oxygen or nitrogen acids of the same
pKa.14 Since the acidity of carbon acids is typically due to the
presence of a functional group that can stabilize the anion by
resonance, a delay in the delocalization of this charge at the
transition state results in a reaction that is slower than expected
based upon the acidity.2a,c Evidence for this type of imbalance
was first obtained for the ionization of arylnitromethanes,15 and
subsequent studies have shown that transition state imbalance
K
A plot of log k-2 vs pKa for proton abstraction from
substituted 2-tetralones by acetate ion is linear, with a slope
(-RE) of -0.78 ( 0.03 (Figure 1), giving a somewhat greater
dependence on substituent than the reaction with hydroxide ion
(R ) 0.60 ( 0.01).7 As with proton abstraction by hydroxide
ion,7 the point corresponding to 4b shows a negative deviation.
Transition state imbalance due to a delay in charge delocalization
(16) (a) Kim, C.; Traylor, T.; Perrin, C. L. J. Am. Chem. Soc. 1997,
120, 9513. (b) Garcia-Rio, L.; Leis, J. R.; Moreira, J. A.; Norberto, F. J.
Chem. Soc., Perkin Trans. 2 1998, 7, 1613. (c) Lee, I.; Kim, C. K.; Lee,
B.-S.; Kim, C. K.; Lee, H. W.; Han, I. S. J. Phys. Org. Chem. 1997, 10,
908. (d) Moutiers, G.; Le Guevel, E.; Villien, L.; Terrier, F. J. Chem. Soc.,
Perkin Trans. 2 1997, 1, 7.
(14) Bell, R. P. The Proton in Chemistry, 2nd ed.; Cornell University
Press: Ithaca, NY, 1973.
(15) (a) Bordwell, F. G.; Boyle, W. J. J. Am. Chem. Soc. 1971, 93, 511.
(b) Bordwell, F. G.; Boyle, W. J. J. Am. Chem. Soc. 1972, 94, 3907.

Enolization of Phenyl-Substituted 2-Tetralones
J. Am. Chem. Soc., Vol. 121, No. 26, 1999 6223
Scheme 4
Figure 2. Plot of the logarithm of the rate constant for ionization of
substituted 2-tetralones (log kB in M-1 s-1) vs pKa of substituted acetic
acid (3.3% (v/v) acetonitrile, µ ) 1.0, 25 ( 0.1 °C).
transferred of +0.20 at the transition state.19 The charge
distributions20 for the transition states for enolization of 2-te-
tralone by acetate and by hydroxide ion, as well as the fully
formed anion of 2-tetralone,7 are summarized in Scheme 4.7,21,22
Several authors have used inequality of the Brønsted R and
â for a reaction as evidence for an imbalanced transition
state.2c,23,24 In a balanced reaction, the extent of charge transfer
as determined from each of these plots should be identical, after
a correction for solvation.25 On the other hand, an imbalanced
transition state for enolization of 2-tetralone would have a
greater fraction of negative charge on C-1 than a balanced
transition state. Since RE is detected by ring substitution, it will
be greater for an imbalanced transition state than for a balanced
one. Thus, RE should be greater than âcorr in the enolization of
2-tetralone by acetate ion. The large difference in the magnitude
of the RE value observed here (0.78) compared to âE (0.54) or
âcorr (0.45) is consistent with an imbalanced transition state.26
Position of the Transition States for Enolization of
2-Tetralone by Acetate Ion and by Hydroxide Ion. It is
may be probed by a comparison of the observed rate constant
for 4b to that of a hypothetical 6-nitro substitution (pKaK 11.5)7
with no resonance effect.17 From the correlation of Figure 1, a
rate constant for proton abstraction from this hypothetical acid
may be obtained (log kind ) -0.826). The predicted rate constant
for reaction of 4b in which the resonance stabilization from
the 6-nitro group in the transition state is as large as in the fully
formed anion (or a balanced transition state) is obtained from
Figure 1 (log kpred ) 0.269). A comparison of these rate
constants with the observed rate constant for 4b (log k-2
)
-0.167) shows that about 60% of the expected resonance
stabilization of the 6-nitro group operates at the transition state
for the proton transfer to acetate. This result is similar to that
obtained previously for the transition state for deprotonation
by hydroxide, in which 55% of the full resonance stabilization
of the 6-nitro group is manifest.7 Using this result and the
assumption that the fraction of charge transferred to the tetralone
is equal to the fraction of p-orbital formed, we estimated that
55% of a full negative charge is located on the tetralone system
at the transition state for deprotonation by hydroxide.7 For the
acetate reaction, the comparable figure would be 60%.
(18) To estimate the charge transferred from a carboxyl group in the
transition state from this â value, it is necessary to correct for solvation of
the carboxyl acid by hydrogen bonding to water.2c,27 The effect of solvation
(Rsol) of the carboxyl acid is about 0.2, which enlarges the â scale for the
protonation of acetate ion from 1.0 to 1.2. Thus, the corrected value (âcorr
)
can be calculated from the observed â (âobs) by âcorr ) âobs/(1 + Rsol) )
âobs/1.2. For deprotonation of 2-tetralone by acetate, âcorr is then equal to
These calculations depend on the assumption that the fraction
of charge transferred to the 2-tetralone is equal to the fraction
of the p-orbital formed in the transition state.7 This assumption
can be tested by using the effect of a change in the pKa of the
attacking base as an experimental probe for charge transfer (or
C-H bond breakage). From the Brønsted plot of the logarithm
of the rate constants for the deprotonation of 2-tetralones by
substituted acetates (Figure 2), a â value of 0.54 ( 0.02 is
obtained. Consideration of the effect of solvation of the acid
gives a corrected value of âcorr ) 0.45.18 Thus, the charge left
on the acetate ion at the transition state may be calculated to
be -0.55. Since the charge on the tetralone at the transition
state is about -0.60, there must be a charge of ca. +0.15 on
the hydrogen being transferred to balance charge. This conclu-
sion is in qualitative agreement with computational results on
the deprotonation of acetaldehyde by acetaldehyde enolate in
the gas phase, which predict a charge on the proton being
0.54/1.2 ) 0.45.
(19) (a) Saunders, W. H., Jr. J. Am. Chem. Soc. 1994, 116, 540. (b)
Bernasconi, C. F.; Wenzel, P. L. J. Am. Chem. Soc. 1994, 116, 5405.
(20) Charges specified are relatiVe (differences between the neutral
molecule and the anion or the transition state), not absolute. The charge
distribution in the anion of 2-tetralone is calculated from the difference
between the 13C chemical shifts for the anion and for 2-tetralone.
(21) The charge on the transition state for the ionization of 2-tetralone
by hydroxide ion was recalculated by putting a partial positive charge
(+0.15) on the proton being transferred as for the reaction with acetate,
thus increasing the partial negative charge on the hydroxide group (-0.60
instead of -0.45).
(22) The relative distribution of charge at the transition state for various
reactions has also been discussed in terms of effective charge, which is
calculated from Brønsted coefficients. For a review on effective charge,
see: Williams, A. AdV. Phys. Org. Chem. 1992, 27, 1.
(23) (a) Stefanidis, D. J.; Bunting, J. W. J. Am. Chem. Soc. 1990, 112,
3163. (b) Wodzinski, S.; Bunting, J. W. J. Am. Chem. Soc. 1994, 116, 6910.
(24) Lee, I.; Kim, C. K.; Lee, B.-S.; Kim, C. K.; Lee, H. W.; Han, I. S.
J. Phys. Org. Chem. 1997, 10, 908.
(25) For proton-transfer reactions, this probe for transition state imbalance
assumes that there is no partial charge in the transition state on the proton
being transferred. The presence of a partial charge would predict a difference
between the Brønsted R and â, even for a balanced transition state. Thus,
a small difference between R and â should be interpreted with caution.
(26) The use of the 6-nitro substituent as a probe only partially detects
a delay in solvent reorganization, since the majority of the incipient negative
charge on C-1 is delocalized into the carbonyl group. However, the
inequality of R and â could be due to a lag in either solvent reorganization
or rehybridization.
(17) The amount of transition state imbalance is somewhat underestimated
by these methods of detection, since the enolate of 2-tetralone is significantly
stabilized by conjugation between the phenyl ring and the double bond that
does not involve charge transfer.9 This conjugation is also limited by
hybridization changes at the transition state, and it, too, could contribute to
the transition state imbalance. However, the lack of charge transfer means
that this contribution cannot be probed by the methods that we use.

6224 J. Am. Chem. Soc., Vol. 121, No. 26, 1999
Yao et al.
difficult to compare the position of the transition states for
enolization along the reaction coordinates, because bond break-
age (charge transfer) is not synchronous with electron delocal-
ization, and thus there is no single reaction parameter that can
describe the position of an imbalanced transition state. However,
if we define “position” as the degree of proton transfer, âcorr
may be used as an estimate of the position of the transition
state in the acetate reaction. In the deprotonation of 2-tetralones
by acetate ion, the corrected âcorr of 0.45 implies that the
transition state is approximately halfway along the reaction
coordinate. However, the lack of a â value for the hydroxide
ion reaction makes it impossible to directly compare the
positions of the two transition states.
The calculated structures of the transition states for enolization
by acetate and by hydroxide ion are similar, but these calcula-
tions involve some approximations and are thus imprecise. The
use of R values to estimate the extent of proton transfer is
complicated by the existence of transition state imbalance.
However, the similar magnitude of the imbalance for the two
transition states should enable a qualitative comparison of the
two R values. Thus, the more negative R (0.78) for proton
abstraction by acetate than by hydroxide (0.60) suggests a later
transition state for the acetate reaction. Similar phenomena have
been observed in the other proton-transfer reactions.27
Transition State Imbalance and Enzymatic Enolization.
It is of interest to estimate the contribution of transition state
imbalance to the intrinsic barrier for the proton transfer of
2-tetralones to acetate ion, and therefore whether minimization
of this transition state imbalance could be significant in an
enzymatic acceleration of enolization. Bernasconi2c has proposed
a mathematical relationship between a change of the intrinsic
rate constant for a reaction and the imbalance in the reaction.
For a reaction that leads to a resonance-stabilized carbanion,
res
the intrinsic rate constant is given by eq 5,31 where δ log ko
is the difference between the intrinsic rate constant for the
reaction in question and a hypothetical reaction of the same
pKa without resonance stabilization of the anion, λres is the
fraction of the charge that is delocalized in the transition state
relative to the product, â is the total amount of charge transferred
in the transition state (equivalent to the Brønsted â), and δ log
Kares is the difference in pKa between the acid and a hypothetical
acid with no resonance stabilization.
δ log kores ) (λres - â) δ log Kares
(5)
For enolization of 2-tetralone by acetate ion, the corrected
value of â is 0.45, and λres is 60% âcorr, or 0.27. The contribution
of resonance stabilization from the phenyl and carbonyl groups
to the increased acidity of 2-tetralone (δ log Kares) relative to
methane (pKa ∼50)32 can be estimated from the relative
contributions of resonance and inductive effects to the electron-
withdrawing nature of these substituents. To do this, we use
σ- as a measure of the sum of the resonance and inductive
effects and σ as a measure of the inductive effect. Thus, the
fraction of resonance stabilization from both groups is calculated
Intrinsic Barriers for Proton Transfer of 2-Tetralone by
Acetate Ion. The slow rate of abstraction of protons from carbon
atoms adjacent to carbonyl groups has been attributed to
relatively high barriers to reaction that exist even when the
reaction is thermodynamically favorable.1,2 Marcus28 has dis-
sected the kinetic barrier (∆Gq) for a reaction into an intrinsic
barrier (∆Gintq) and a thermodynamic barrier (∆G°) (eq 4).
res
through ∑i(σ- - σi)/∑iσ- ) 43%, and δ log ko ) (0.27-
i
i
∆Gq ) ∆Gintq(1 + ∆G°/4∆Gintq)2 ) ∆Gintq + ∆G°/2 +
0.45) × (43% × 37) ) -2.9. This value corresponds to an
increase in the intrinsic barrier of 4 kcal/mol over that for a
hypothetical balanced transition state.
(∆G°)2/16∆Gint ≈ ∆Gintq + ∆G°/2 (4)
q
We6 have suggested that the rate enhancement for the
isomerization of 5-androstene-3,17-dione by steroid isomerase
(KSI) is due to contributions from lowering of both the
thermodynamic barrier (by ca. 11 kcal/mol, of which ca. 5 kcal/
mol shows up in the transition state) and the intrinsic barrier
(by ca. 3 kcal/mol). Thus, although the predominant mode of
catalysis by KSI involves a decrease in the thermodynamic
barrier, there is a substantial contribution from a decreased
intrinsic barrier, which could be accomplished by minimization
of transition state imbalance. The source of the high intrinsic
barriers in these reactions may be attributed to a combination
of two causes.33 One is the incomplete rehybridization of the
incipient p-orbital at C-1 in the transition state,7,34 which limits
delocalization of the negative charge transferred at the transition
state. The other is a delay in solvent reorganization around the
Because the intrinsic barrier for a proton-transfer reaction, in
principle, describes the rate of the reaction at ∆pKa ) 0
(difference between the acid and the base), it is simplest to
determine it from the Brønsted correlation by extrapolation.
Application of this method to the substituted acetate data (Figure
2) gives the intrinsic rate constant (ko) for 2-tetralone (260 M-1
s-1),29 which can be converted to an intrinsic barrier (∆Go ) of
q
about 14 kcal/mol. Thus, the barrier for deprotonation of
2-tetralone by acetate (20 kcal/mol, k-2 ) 13.7 × 10-3 M-1
s-1) is composed of this intrinsic barrier (14 kcal/mol) and about
half of the thermodynamic barrier (11/2 ) 6 kcal/mol).30 The
total barrier for proton transfer to hydroxide is about 14 kcal/
mol (k ) 376 M-1 s-1).7 The intrinsic barrier for this reaction
(16 kcal/mol) can be estimated from eq 4 and the ∆pK of ca. 3
for water and 2-tetralone. The similar intrinsic barriers for
deprotonation by hydroxide ion and acetate ion are consistent
with a similar transition state imbalance for both reactions.2
(31) Bernasconi has further generalized this equation to quantitate the
increase of the intrinsic barrier of a reaction caused by the transition state
imbalance from any factor.2c
(27) (a) Jencks, W. P.; Haber, M. T.; Herschlag, D.; Nazaretian, K. L.
J. Am. Chem. Soc. 1986, 108, 479. (b) Murray, C. J.; Jencks, W. P. J. Am.
Chem. Soc. 1990, 112, 1880.
(28) (a) Marcus, R. A. J. Chem. Phys. 1956, 24, 966. (b) Marcus, R. A.
J. Chem. Phys. 1957, 26, 867. (c) Marcus, R. A. Annu. ReV. Phys. Chem.
1964, 15, 155.
(29) The intrinsic rate constant (ko) was corrected for a statistical factor
of 2. The relative pKaK of 2-tetralone and acetate was assumed to be invariant
in H2O and D2O.
(32) Pross, A. Theoretical and Physical Principles of Organic ReactiVity;
John Wiley & Sons: New York, 1995; p 200.
(33) The additional delay in conjugation with the phenyl ring with C-1
at the transition state could lead to a further increase in the intrinsic barrier.
There is an increase (2-3 kcal/mol) of the intrinsic barrier for deprotonation
of 2-tetralone by hydroxide ion over those for enolization of acetone (13
kcal/mol: Albery, W. J. J. Chem. Soc., Faraday Trans. 1 1982, 78, 1579)
and for enolization of simple aldehydes and ketones (12.1 kcal/mol: Guthrie,
J. P. Can. J. Chem. 1979, 57, 1177). It is likely that the higher intrinsic
barrier for enolization of 2-tetralone is caused by the phenyl ring fused to
the R carbon of the carbonyl group. Thus, enolization involves conjugation
of a p-orbital at C-1 with the carbonyl group and the phenyl group, both of
which are important in stabilizing the enolate of 2-tetralone.9
(34) Kresge, A. J. Can. J. Chem. 1974, 52, 1897.
(30) According to eq 4, the effect of the thermodynamic barrier on the
energy of the transition state can be estimated as half of the thermodynamic
barrier (∆G°/2) by ignoring the small interaction between the thermody-
namic barrier and the intrinsic barrier for this specific enolization of
2-tetralone [(∆G°)2/16∆Gint ) 0.5 kcal/mol].
q

Enolization of Phenyl-Substituted 2-Tetralones
J. Am. Chem. Soc., Vol. 121, No. 26, 1999 6225
oxygen at the transition state,2a,c,7,35 inhibiting charge dispersal
onto this atom. Although it is unlikely that an enzyme can do
much about the first of these difficulties, preorganization of the
atoms at the enzyme active site that are involved in solvation
of the negative charge at the incipient oxyanion could result in
a substantial lowering of the intrinsic barrier. Thus, imbalance
caused by a lag in solvent reorganization could be eliminated
(or at least reduced).2c,7,36,37
Acknowledgment is made to the National Institutes of Health
(Grant GM 38155) and the donors of the Petroleum Research
Fund, administered by the American Chemical Society, for
support of this research.
Supporting Information Available: Table of rate constants
for proton exchange of 2-tetralone catalyzed by substituted
acetate ions, and a plot for the observed rate constant for acetate-
catalyzed ketonization of 2-tetralone enol vs the concentration
of acetate ion in the buffer (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
(35) Bernasconi, C. F.; Wenzel, P. J.; Keeffe, J. R.; Gronert, S. J. Am.
Chem. Soc. 1997, 119, 4008.
(36) Ydav, A.; Jackson, R. M.; Holbrook, J. J.; Warshel, A. J. Am. Chem.
Soc. 1991, 113, 4800.
(37) Pera¨kyla¨, M. J. Chem. Soc., Perkin Trans. 2 1997, 2185.
JA990070+