Secondary deuterium isotope effects for enolization reactions

Article abstract of DOI:10.1021/ja942053k

Secondary α- and β-deuterium isotope effects for enolization reactions and equilibria have been determined by ab initio calculations, ¹H NMR spectroscopy, and triton exchange kinetics. Kinetic and equilibrium α-deuterium isotope effects for hydroxide ion-catalyzed enolization of acetaldehyde calculated by ab initio methods are normal and depend on the orientation of the secondary hydrogen with respect to the carbonyl group. The computed transition state structure indicates a small degree of bond rehybridization at the transition state. Experimentally measured secondary isotope effects on the deuteroxide ion-catalyzed proton exchange of acetophenone are k(H)/k(D) = 1.08 ± 0.07 for α-CH₃ exchange and k(H)/k(D) = 0.96 ± 0.08 for α-CH₂D exchange. For α-CH₂T exchange in water, the corresponding secondary isotope effect is k(H)/k(D) = 1.06 ± 0.02, assuming the rule of the geometric mean is valid. These effects are smaller than the calculated equilibrium isotope effect for formation of the enolate ion-water complex: K(H)/K(D) = 1.11-1.22 at the MP2 level. The normal kinetic isotope effects are smaller than might be expected due to a loss in hyperconjugation of the out-of-plane C-H bond and a lag in structural reorganization that contributes to the intrinsic barrier for proton transfer from carbon. Ionization of protonated acetone gives rise to an inverse secondary isotope effect of 0.97/D for the C-L bond adjacent to the carbonyl group and is consistent with a loss in hyperconjugation upon formation of the neutral ketone.

Full text of DOI:10.1021/ja942053k

+
+
6562
J. Am. Chem. Soc. 1996, 118, 6562-6569
Secondary Deuterium Isotope Effects for Enolization Reactions
William C. Alston II,‡ Kari Haley, Ryszard Kanski,§ Christopher J. Murray,*,† and
Julianto Pranata*
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Arkansas,
FayetteVille, Arkansas 72701
ReceiVed June 27, 1994. ReVised Manuscript ReceiVed NoVember 28, 1995X
Abstract: Secondary R- and â-deuterium isotope effects for enolization reactions and equilibria have been determined
by ab initio calculations, 1H NMR spectroscopy, and triton exchange kinetics. Kinetic and equilibrium R-deuterium
isotope effects for hydroxide ion-catalyzed enolization of acetaldehyde calculated by ab initio methods are normal
and depend on the orientation of the secondary hydrogen with respect to the carbonyl group. The computed transition
state structure indicates a small degree of bond rehybridization at the transition state. Experimentally measured
secondary isotope effects on the deuteroxide ion-catalyzed proton exchange of acetophenone are kH/kD ) 1.08 (
0.07 for R-CH3 exchange and kH/kD ) 0.96 ( 0.08 for R-CH2D exchange. For R-CH2T exchange in water, the
corresponding secondary isotope effect is kH/kD ) 1.06 ( 0.02, assuming the rule of the geometric mean is valid.
These effects are smaller than the calculated equilibrium isotope effect for formation of the enolate ion-water
complex: KH/KD ) 1.11-1.22 at the MP2 level. The normal kinetic isotope effects are smaller than might be
expected due to a loss in hyperconjugation of the out-of-plane C-H bond and a lag in structural reorganization that
contributes to the intrinsic barrier for proton transfer from carbon. Ionization of protonated acetone gives rise to an
inverse secondary isotope effect of 0.97/D for the C-L bond adjacent to the carbonyl group and is consistent with
a loss in hyperconjugation upon formation of the neutral ketone.
Introduction
Such kinetic5 and equilibrium6 effects have been measured,
but in some cases interpretation of the data is difficult due to
uncertainties associated with the experimental technique.7 1H
NMR spectroscopy provides a useful method for measuring
structure-reactivity effects in proton transfer reactions of carbon
acids,8-10 and if separate signals of the isotopic species (CHH
vs CHD) can be measured simultaneously in the same sample,
secondary deuterium isotope effects can be measured.11,12
Computations of kinetic isotope effects using vibrational
frequencies obtained from ab initio calculations can also provide
useful insights into the interpretation of secondary KIEs.13
We report here secondary R- and â-deuterium isotope effects
for enolization reactions and equilibria as shown in Scheme 1,
Proton transfers from carbon acids activated by the carbonyl
group generate enolate or enol intermediates in enzyme reac-
tions.1 Estimates of the lifetimes of enzyme-bound enolate
intermediates suggest that enzymes stabilize these intermediates
or avoid their formation through concerted reactions.2 Second-
ary isotope effects can provide useful probes of the intermediacy
of enol(ate)s in enzyme reactions,3,4 but in order to interpret
these effects accurate measures of the magnitude of secondary
kinetic (KIE) and equilibrium (EIE) isotope effects for simple
enolization reactions in solution are required.
‡ Present address: Department of Biochemistry and Biophysics, Texas
A&M University, College Station, TX 77843.
§ Present address: Department of Chemistry, Radiochemistry and Radia-
tion Chemistry Laboratory, Warsaw University, Al. Zwirki i Wigury 101,
02-089 Warsaw, Poland.
(5) (a) Emmons, W. D.; Hawthrone, M. F. J. Am. Chem. Soc. 1956, 78,
5593-5596. (b) Jones, J. R. Trans. Faraday Soc. 1965, 61, 95-99. (c)
Hine, J.; Kaufmann, J. C.; Cholod, M. S. J. Am. Chem. Soc. 1972, 94,
4590-4595. (d) Kresge, A. J.; Weeks, D. P. J. Am. Chem. Soc. 1984,
106, 7140-7143. (e) Andraos, J.; Kresge, A. J.; Obraztsov, P. A. J. Phys.
Org. Chem. 1992, 5, 322-326.
† Address correspondence to this author at Genencor International, Inc.
925 Page Mill Rd., Palo Alto, CA 94304.
X Abstract published in AdVance ACS Abstracts, July 1, 1996.
(1) (a) Rose, I. A. Meth. Enzymol. 1982, 87, 84-97. (b) Richard, J. P.
In The Chemistry of Enols; Rappoport, Z., Ed.; John Wiley and Sons:
Chichester, 1990; Chapter 11. (c) Keeffe, J. R.; Kresge, A. J. In The
Chemistry of Enols; Rappoport, Z., Ed.; John Wiley and Sons: Chichester,
1990; Chapter 7.
(6) Arnett, E. M.; Cohen, T.; Bothner-By, A. A.; Bushick, R. D.;
Sowinski, G. Chem. Ind. 1961, 473-474.
(7) (a) Abad, G. A.; Jindal, S. P.; Tidwell, T. T. J. Am. Chem. Soc. 1973,
95, 6326-6331. (b) Chiang, Y.; Kresge, A. J.; Morimoto, H.; Williams,
P. G. J. Am. Chem. Soc. 1992, 114, 3981-3982.
(2) (a) Thibblin, A.; Jencks, W. P. J. Am. Chem. Soc. 1979, 101, 4963-
4973. (b) Gerlt, J. A.; Gassman, P. G. J. Am. Chem. Soc. 1992, 114, 5928-
5934. (c) Gerlt, J. A.; Gassman, P. G. Biochemistry 1993, 32, 11943-
11952. (d) Cleland, W. W.; Kreevoy, M. M. Science 1994, 264, 1887-
1890. (e) Alagona, G.; Chio, C.; Kollman, P. A. J. Am. Chem. Soc. 1995,
117, 9855-9862.
(3) (a) Rose, I. A. In Isotope Effects on Enzyme Catalyzed Reactions;
Cleland, W. W., O’Leary, M. H., Northrop, D. B., Eds.; University Park
Press: Baltimore, 1977; pp 209-232. (b) Cleland, W. W. In Isotopes in
Organic Chemistry; Buncel, E., Lee, C. C., Eds.; Elsevier: New York, 1987;
Vol. 7, Chapter 2. (c) Anderson, V. E. In Enzyme Mechanisms from Isotope
Effects; Cook, P. F., Ed.; CRC Press: Boca Raton, FL, 1991; Chapter 16.
(4) (a) Pascal, R. A.; Walsh, C. T. Biochemistry 1984, 23, 2745-2752.
(b) Bahnson, B. J.; Anderson, V. E. Biochemistry 1989, 28, 4173-4181.
(c) Xue, L.; Talalay, P.; Mildvan, A. S. Biochemistry 1990, 29, 7491-
7500. (d) Denu, J. M.; Fitzpatrick, P. F. Biochemistry 1994, 33, 4001-
4007.
(8) Washabaugh, M. W.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111,
683-692.
(9) Murray, C. J.; Jencks, W. P. J. Am. Chem. Soc. 1990, 112, 1880-
1889.
(10) Amyes, T.; Richard, J. P. J. Am. Chem. Soc. 1992, 114, 10297-
10302.
(11) Jarret, R. M.; Saunders, M. J. Am. Chem. Soc. 1985, 107, 2648-
2654.
(12) Perrin, C. L.; Thoburn, J. D.; Kresge, A. J. J. Am. Chem. Soc. 1992,
114, 8800-8807.
(13) (a) Saunders, M.; Laidig, K. E.; Wolfsberg, M. J. Am. Chem. Soc.
1989, 111, 8989-8994. (b) Duan, X.; Scheiner, S. J. Am. Chem. Soc. 1992,
114, 5849-5856. (c) Houk, K. N.; Gustafson, S. M.; Black, K. A. J. Am.
Chem. Soc. 1992, 114, 8565. (d) Storer, J. W.; Raimondi, L.; Houk, K. N.
J. Am. Chem. Soc. 1994, 116, 9675-9683. (e) Wiest, O.; Black, K. A.;
Houk, K. N. J. Am. Chem. Soc. 1994, 116, 10336-10337. (f) Yoo, H. Y.;
Houk, K. N. J. Am. Chem. Soc. 1994, 116, 12047-12048.
S0002-7863(94)02053-6 CCC: $12 00 © 1996 American Chemical Society

+
+
Deuterium Isotope Effects for Enolization Reactions
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6563
ammonium ion as an internal standard and recording the spectrum
immediately after mixing with suppression of the water peak. The
apparent spin-lattice relaxation times for the R-CH3 protons (T1 )
4.5 s), the R-CH2D protons (T1 ) 6 s), and the R-CHD2 protons (T1 )
12 s) of acetophenone were measured by the inversion recovery method
(180°-τ-90°)18 at 1.0 M ionic strength (KCl) in D2O.
Scheme 1
Experimental data were fit to eqs 1-3 using SigmaPlot (Jandel
Scientific) or GraFit (Erithacus Software Ltd.).
Acetophenone Triton Exchange in H2O. First-order rate constants
for R-C-3H exchange were determined in H2O, ionic strength 1 M
maintained with KCl by monitoring the decrease in the ratio of H to
where L ) H or D, using 1H NMR spectroscopy, triton exchange
kinetics, and ab initio calculations.14 The inverse â-secondary
isotope effect of 0.97/D for the C-L bond adjacent to the
carbonyl group in the ionization of protonated acetone is
consistent with a decrease in hyperconjugation upon ioniza-
tion.15,16 Hyperconjugation is unchanged for the in-plane CH
bond of the ketone in Scheme 1, but it is further lost in the
out-of-plane CH in the enolization reaction. This gives rise to
a normal secondary R-deuterium KIE ) 1.06 ( 0.02. Ab initio
calculations suggest that the magnitude of the isotope effect is
dependent on the position of the secondary CH bond relative
to the carbonyl group, consistent with opposing effects of sp3-
sp2 rehybridization of the OPL CH bonds and loss of hyper-
conjugation of the CH bonds with the carbonyl in the transition
state.
3
14C in acetophenone in basic solution. Experiments with [1-3H]-
acetophenone and [1-3H,1-2H2]acetophenone were generally carried out
on the same day. Radiolabeled substrate (15 µM; 3H/14C ratio ≈ 1.15)
was added in a volume of 10 mL containing 1.0 M KCl and varying
amounts of KOH. Samples were quenched with HCl and frozen until
HPLC separation. TOH and radiolabeled acetophenone were separated
by reverse phase chromatography using a Shimadzu LC-610 system
on a Shimadzu 4 × 250 mm C-18 column. Samples were eluted at a
flow rate of 1 mL/min with a 55:45 methanol-water mobile
phase. One-milliliter fractions were collected directly into 7-mL
scintillation vials, and after adding 5 mL of Opti Phase HiSafe 3
scintillation cocktail (Wallac Scintillation Products), radioactivity was
determined by liquid scintillation counting on a Wallac 1410 liquid
scintillation counter.
Acetophenone Proton Exchange in D2O. Rate constants for R-CH-
exchange were determined in D2O at 25 °C and ionic strength 1 M
maintained with KCl. Acetophenone (1 µL) was added to 358 µL
of deuterium oxide in a 5-mm NMR tube and sonicated followed
by the addition of 375 µL of 2.0 M KCl. The final concentration
of acetophenone (ca. 10 mM) was below the limit of solubility
(ca. 20 mM) in 1.0 M KCl. The magnet was carefully shimmed prior
to the addition of 3-5 µL of 0.1 M KOD to the NMR tube and
reinsertion into the probe. Spectra were recorded with a digital
resolution of 0.05 Hz/point and acquisition parameters were set so that
the steady state z magnetization between pulses was g95% of the
equilibrium magnetization of the R-CHD2 signal corresponding to the
longest T1.18
Experimental Section
Materials. Acetophenone-d3 (99.9 atom % D), acetone-d5 (99.9
atom % D), and deuterium oxide (D2O, 99.9 atom % D) were from
Isotec, Inc. [1-14C]Acetophenone was from Amersham, Inc. Acetophe-
none and Acetophenone-d3 were distilled under vacuum and stored at
-20 °C in sealed vials. Tritiated water (TOH; 5 Ci/mL) was from
ICN Biomedicals, Inc. (Costa Mesa, CA). TOH was diluted to a
specific activity of 0.1 Ci/mL in H2O or D2O. [1-3H]Acetophenone
was synthesized by tritium exchange in TOH, giving a final specific
activity of 0.273 µCi/mmol. [1-3H,1-2H2]Acetophenone was synthe-
sized from acetophenone-d3 in TOD (containing less than 1% HOD)
to give a final specific activity of 0.123 mCi/mmol. Trimethylammo-
nium sulfate was prepared by titrating trimethylamine with sulfuric acid,
removing the water under vacuum, and drying in a vacuum oven at 40
°C overnight. Sulfuric acid-water mixtures were prepared by dilution
of commercial 98% acid and titrated with standard NaOH. All other
chemicals were reagent grade and used without further purification.
Methods. Reactions were carried out at 25 °C. Solution pH and
pD were measured with an Orion Model 701A pH meter and a
Radiometer GK2321C combination electrode. The value of pD was
obtained by adding 0.4 to the observed pH of solutions in D2O.17 The
optimal pD for the acetophenone R-CH exchange experiments was
dictated within fairly narrow limits by the buffering capacity at low
deuteroxide concentrations and signal-to-noise considerations at higher
base concentrations that lead to faster rates of exchange. Only solutions
where the pD did not change by more than 0.02 unit were used in
analysis of the data.
Data Analysis. For the equilibrium data, the chemical shift
differences of the methyl singlet of acetone and the methyl quintet of
acetone-d5, relative to trimethylammonium ion, were fit to the equations
derived by Bagno et al.19 that describe the equilibria in Scheme 2. The
method requires estimates of limiting values of the chemical shifts ∆νB,
+
+
∆νBCH , and ∆νBH corresponding to the free base, hydrogen-bonded
complex, and protonated base, respectively, as well as values of m*,
c
+
+
m*, pKBCH , and pKBH that define the protonation equilibria in Scheme
2. The chemical shifts are nonlinearly related so they were optimized
first by minimizing the sum of the squares of the errors, then the m*
and pK values, which are linearly related, were optimized. The fitting
procedure used the acidity function summarized in Bagno et al.19d
For the kinetic data, the experimental data were fit to Lorentizian
line shapes using the spectrum deconvolution subroutines of VNMR.20
The statistically corrected integrated areas of the R-CH3 singlet (P),
the R-CH2D triplet (Q), and the R-CHD2 quintet (R) were simulta-
neously fit to eqs 1-3 that describe three consecutive first-order
processes.21
Proton NMR spectra were recorded on a Varian VXR-500S
spectrometer in sulfuric acid/H2O mixtures for equilibrium studies and
in D2O, 1 M ionic strength (KCl) for kinetic studies. The chemical
shift of acetone and acetone-d5 in sulfuric acid-water mixtures was
determined by adding 1-2 µL of acetone and 50 µL of acetone-d5 in
600 µL of sulfuric acid/water mixtures containing 0.01 M trimethyl-
(18) Derome, W. Modern NMR Techniques for Chemistry Research;
Pergamon Press: Oxford, 1987; pp 168-170.
(14) The term “hydron” refers to the hydrogen cation (L+) without regard
to nuclear mass. The specific names “proton” (1H, H), “deuteron”, (2H,
D), and “triton” (3H, T) refer to specific isotopes (Bunnett, J. F.; Jones, R.
A. Y. Pure Appl. Chem. 1988, 60, 1115-1116).
(19) (a) Bagno, A.; Lucchini, V.; Scorrano, G. Bull. Soc. Chim. Fr. 1987,
563-572. (b) Bagno, A.; Lucchini, V.; Scorrano, G. J. Phys. Chem. 1991,
95, 345-352. Bagno et al. have shown that the pKa of protonated acetone
can be obtained with a precision of (0.03 pK units. The excellent
agreement between values of the fitting parameters obtained independently
provides strong support for the approach of Bagno et al. We note that the
observation that ∆ν for acetone-d5 is less than for acetone at all values of
H0 examined requires that pKd5 > pKh6. (c) Bagno, A.; Lovato, G.;
Scorrano, G. J. Chem. Soc., Perkin Trans. 2 1993, 1091-1098. (d) Bagno,
A.; Scorrano, G.; More O’Ferrall, R. A. ReV. Chem. Intermed. 1987, 7,
313-352.
(15) Hogg, J. L. In Transition States of Biochemical Processes; Gandour,
R. D., Schowen, R. L., Eds.; Plenum Press: New York; 1978; Chapter 5.
(16) Several workers, most notably Schowen, and his colleagues, have
used the magnitude of â-deuterium isotope effects as a probe of transition
state structure in nucleophilic acyl substitution reactions. (a) Kovach, I.
M.; Hogg, J. L.; Raben, T.; Halbert, K.; Rodgers, J.; Schowen, R. L. J.
Am. Chem. Soc. 1980, 102, 1991-1999. (b) Harrison, R. K.; Stein, R. L.
Biochemistry 1990, 29, 1684-1689.
(20) Weiss, G. H.; Ferretti, J. A. J. Magn. Reson. 1983, 55, 397-407.
(21) Bateman, H. Proc. Cambridge Phil. Soc. 1910, 15, 423-427.
(17) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188-191.

+
+
6564 J. Am. Chem. Soc., Vol. 118, No. 28, 1996
Alston et al.
P ) P0e-λ t
(1)
(2)
1
Scheme 2
λ1
λ1P0
1
2
Q )
P0e-λ t
+
+ Q0 e-λ t
(
)
λ2 - λ1
λ1 - λ2
λ1λ2P0
λ1λ2P0
e-λ t
+
+
1
R )
[
(λ2 - λ1)(λ3 - λ1)
(λ1 - λ2)(λ3 - λ2)
λ2Q0
λ1λ2P0
λ2Q0
e-λ t
+
+
+ R0 e-λ t (3)
2
3
]
[
]
λ3 - λ2
λ2 - λ3
(λ1 - λ3)(λ2 - λ3)
In eqs 1-3, P0, Q0, and R0 correspond to the initial peak areas, and λ1,
λ2, and λ3 are the pseudo-first-order rate constants for deuteroxide ion
catalyzed exchange of the R-CH3, R-CH2D, and R-CHD2 protons,
respectively. The triplet and quintet were not completely resolved at
this field strength as shown in Figure 2, especially for early percent
exchange. The integral for the quintet was adjusted for this effect by
multiplying the area of the peak at -20.2 Hz by 2.
Ab Initio Calculations. In order to provide a context for interpreta-
tion of the experimentally measured kinetic isotope effects, ab initio
calculations were carried out on a simple model reaction of acetaldehyde
with hydroxide ion. Structures relevant to the aldehyde-to-hydroxide
ion hydron transfer reaction that lead to enolate formation include
acetaldehyde, hydroxide ion, the aldehyde-hydroxide ion complex, the
transition state, the enolate-water complex, the enolate ion, and the
enol isomer of acetaldehyde. Numerous attempts to explore the reverse
proton transfer from water to the enolate to form the enol-hydroxide
ion complex were unsuccessful, suggesting that the enol-to-hydroxide
proton transfer is a very favorable process that proceeds without a
barrier. Additional calculations were also performed relating to the
nucleophilic addition reaction of hydroxide to acetaldehyde.
Initial calculations were performed at the RHF/6-31+G(d) level.22
Complete geometry optimizations were performed on all the structures.
Transition states were located, and the intrinsic reaction coordinate
method23 was used to ensure that they properly connect the reactants
and products; in fact, the reactant and product complexes were actually
located as end points of the IRC. The stationary points were then
reoptimized at the MP2/6-31+G(d) level.22 Single point calculations
at higher levels, up to MP4(SDTQ)/6-311++G(d,p), were then
performed at the MP2-optimized geometries. The calculations were
performed using the Gaussian-92 program24 on a Silicon Graphics
Indigo R-4000 workstation at the University of Arkansas.
contributions from vibrational excited states (EXC), and a tunneling
correction (Q). νrH, νrD, νqH, and νq are the (real) vibrational
D
frequencies for the hydrogen isotopomer in the reactant, the deuterium
isotopomer in the reactant, the hydrogen isotopomer in the transition
state, and the deuterium isotopomer in the transition state, respectively.
νiH and νiD are the imaginary frequencies for the hydrogen and
deuterium isotopomers in the transition state.
Equilibrium isotope effects were calculated in a similar manner,25
with the vibrational frequencies of the product taking the place of the
transition state; in addition, of course, the tunneling correction (eq 8)
is not applicable.
Frequencies from both RHF/6-31+G(d) and MP2/6-31+G(d) cal-
culations were used. For the KIE calculations the RHF frequencies
were scaled by 0.9135, the MP2 frequencies by 0.9646.26 The effects
of substitution of deuterium for hydrogen at both positions of the methyl
group of acetaldehyde were calculated separately (the third position is
occupied by the proton being transferred, so it affects the primary, not
secondary, kinetic isotope effect). The effects of a second deuterium-
for-hydrogen substitution were also calculated.
Equilibrium isotope effects were also calculated for the protonation
of acetone, using frequencies obtained from RHF/6-31+G(d) and MP2/
6-31+G(d) calculations. The effect of a single deuterium substitution
and the substitution of five deuteriums were calculated.
Results
â-Deuterium Equilibrium Isotope Effects. Bagno et al.19
have shown that a complete description of the protonation
equilibria of a weak base such as acetone (B, in Scheme 2) in
strong acid solutions must include the formation of a hydrogen-
bonded complex, BCH+, as well as the fully protonated base,
BH+. Because the equilibria in Scheme 2 are established rapidly
on the NMR time scale, changes in the chemical shift of the
Secondary KIEs for the reaction of hydroxide ion with acetaldehyde
were computed from isotopic frequencies and calculated using the well-
known formulas:13b,25
KIE ) MMI × ZPE × EXC × Q
(4)
where
+
+
R-CH resonance can be used to determine pKBH and pKBCH
,
MMI )
(νrD/νrH) (νqH/νq
)
(5)
D
∏
∏
as well as the associated m* parameters that measure the
sensitivity of the protonated species to hydrogen-bonding
solvation effects.
ZPE ) exp{(h/kT)[ (νrH - νrD) -
(νqH - νqD)]} (6)
∑
∑
The dependence of the chemical shift of acetone-h6 and
acetone-d5 on the H0 acidity function19b is shown in Figure 1.
The raw data are summarized in Table S127 and the fitting
parameters are summarized in Table 1 along with the corre-
sponding values determined by Bagno et al.19a for acetone.
+
EXC )
{[1 - exp(-hνrH/kT)]/[1 - exp(-hνrD/kT)] ×
∏
{[1 - exp(-hνqD/kT)]/[1 - exp(-hνqH/kT)] (7)
∏
Q ) [1 + (hνiH/kT)2/24]/[1 + (hνiD/kT)2/24]
(8)
Taken together, these data yield a best estimate for pKBH
)
+
-3.09 ( 0.03 for acetone and pKBH ) -3.15 ( 0.005 for
Thus, the kinetic isotope effect can be decomposed into effects due to
the mass and moments of inertia (MMI), the zero point energies (ZPE),
acetone-d5, corresponding to an isotope effect of Kh6 +/Kd5
)
+
BH
BH
0.87 ( 0.04.
(22) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A.; Ab Initio
Molecular Orbital Theory; Wiley Intersciences: New York, 1986.
(23) (a) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154-
2161. (b) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523-
5527.
(24) Frisch, M. J.; Trucks, G. W.; Head-Gordon, M.; Gill, P. M. W.;
Wong, M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M.
A.; Replogle, E. S.; Gomperts, R.; Andres, J. L.; Raghavachari, K.; Binkley,
J. S.; Gonzalez, C.; Martin, R. L.; Fox, D. J.; DeFrees, D. J.; Baker, J.;
Stewart, J. J. P.; Pople, J. A. Gaussian 92, ReV. B; Gaussian, Inc.: Pittsburgh,
PA, 1992.
r-Deuterium Kinetic Isotope Effects. Second-order rate
constants for R-C-L exchange from acetophenone catalyzed
by lyoxide ion in aqueous solution were determined by
detritiation8,9 or H NMR10 at 25 °C and ionic strength 1.0 M
1
(25) (a) Bigeleisen, J.; Mayer, M. G. J. Chem. Phys. 1947, 15, 261-
267. (b) Wigner, E. Z. Phys. Chem. 1932, 19, 203-216.
(26) Pople, J. A.; Scott, A. P.; Wong, M. W.; Radom, L. Isr. J. Chem.
1993, 33, 345-350.
(27) See note concerning supporting information at the end of this paper.

+
+
Deuterium Isotope Effects for Enolization Reactions
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6565
a
Table 1. Summary of Fitting Parameters for Equilibrium Protonation of Acetone and Acetone-d5
chemical shift difference, Hzb
hydrogen bonded complex, BCH+
protonated base, BH+
+
+
+
+
substrate
∆νB
∆νBCH
∆νBH
m*
pKBCH
m*
pKBH
c
c
acetone-h6
acetone-h6
acetone-h6
339
342
341
284
283
283
-29.4
-21.7
-25.3
0.40
0.42
0.42
-1.65
-1.59
-1.59
0.35
0.35
0.35
-3.06
-3.11
-3.09
-3.09 ( 0.03d
acetone-d5
acetone-d5
358
358
286
286
-20.0
-17.8
0.40
0.40
-1.63
-1.68
0.34
0.34
-3.15
-3.14
-3.15 ( 0.005d
1
a Determined at 25 °C in concentratred sulfuric acid/water mixtures by H NMR spectroscopy. b Positive values represent upfield shifts (in Hz)
fro the trimethylammonium ion, at 500 MHz. c Reference 19a. Determined at 60 MHz. Chemical shift values are corrected for the difference in
field strengths. d Final values are the average (( standard deviation) of replicate measurements.
1
Figure 1. Dependence of the H chemical shift difference between
trimethylammonium ion and acetone-h6 (O) or acetone-d5 (b) on H0.
The theoretical curves are the best fit of the data to the model of Scheme
+
2 with values of pKBH ) -3.09, m* ) 0.35, pKBCH ) -1.59, and
+
m*) 0.42 for acetone-h6 and pKBH ) -3.14, m* ) 0.34, pKBCH
)
-c1.68, and m*) 0.40 for acetone-d5.
c
Figure 3. Statistically corrected areas of the R-CH3 (O), R-CH2D (b),
and R-CHD2 (2) signals as a function of time for the deuteroxide ion
catalyzed exchange of acetophenone in D2O at pD ) 11.70 and 25 °C,
ionic strength 1 M (KCl). The data were fit to the integrated rate
equations describing consecutive first-order processes (Scheme 3) with
rate constants 3kHHH[OD-] ) 4.20 × 10-4 s-1, 2kHHD[OD-] ) 2.52 ×
10-4 s-1, and kDHD[OD-] ) 1.25 × 10-4 s-1 with the corresponding
secondary isotope effects kHHH/kHHD ) 1.11 and kHHD/kDHD ) 1.01.
Scheme 3
Figure 2. Partial 500-MHz 1H NMR spectrum of the partially
exchanged methyl protons of acetophenone in D2O. The top spectrum
corresponds to the experimental data, the center spectrum is the full
fit, and the bottom spectrum shows the individual component signals.
The data were fit assuming a Lorentizian line shape.
fit of the integrated areas as a function of time according to the
mechanism of Scheme 3. The second-order rate constants for
hydron exchange are defined by kiLj where L refers to the
primary hydron that is cleaved and i and j refer to the remaining
secondary hydron sites. The raw data are summarized in Tables
S2 and S3 and the second-order rate constants and secondary
KIEs calculated from the fit of the exchange data are sum-
marized in Table 2.
Because the isotope effects are small, it is important to assess
possible sources of error. The errors quoted represent replicate
errors and may be subject to systematic errors. The following
points suggest that the observed isotope effects are reliable. (1)
The second-order rate constant for triton exchange catalyzed
by hydroxide ion kHTH ) 4.74 ( 0.08 × 10-3 M-1 s-1 is in
(KCl) under pseudo-first-order conditions. Figure 2 shows
1
representative data for R-C-1H exchange monitored by H
NMR at 500 MHz indicating incomplete resolution of the
multiplets for the R-CH2D (2JHD ) 2.2 Hz) and R-CHD2 (2JHD
) 2.2 Hz) signals. The coupling constants and isotopic shifts
are similar to values reported for acetone and ethylthioacetate.10
In order to account for incomplete resolution, the peaks were
deconvoluted assuming a Lorentizian line shape as described
in the Experimental Section. Figure 3 shows a representative

+
+
6566 J. Am. Chem. Soc., Vol. 118, No. 28, 1996
Alston et al.
Table 2. Second-Order Rate Constants and Secondary Deuterium
Isotope Effects for Lyoxide Ion Catalyzed R-C-L Exchange from
Acetophenonea
R-C-L/OL- 103kL, M-1 s-1
CH3/OD-
208 ( 10
kHHH/kHHD
kHHD/kDHD
kHTH/kDTD
1.08 ( 0.07b
CH2D/OD- 193 ( 9
CHD2/OD- 201 ( 15
0.96 ( 0.08b
CH2T/OH-
CD2T/OH-
4.74 ( 0.08
1.12 ( 0.04
Figure 4. MP2-optimized structures of acetaldehyde-OH- complex
(1), transition state for proton transfer (2), and enolate-H2O complex
(3). Numbers shown are the O-H or C-H nonbonded distances (in
angstroms) and, for 2, the bond angles of the methyl group (in degrees).
4.22 ( 0.12
a A 25 °C in L2O, ionic strength 1 M (KCl). b Values represent the
average (( propogated error) of the ratio of second-order rate constants.
The corresponding averages and standard errors based on the individual
isotope effect measurements are kHHH/kHHD ) 1.08 ( 0.03 and kHHD/kDHD
) 0.96 ( 0.06. The smaller errors represent cancellation of systematic
errors when data are determined at the same time in the same
instrument.
for the proton transfer reaction: the acetaldehyde-hydroxide
ion complex, the transition state, and the enolate-water
complex. Secondary KIEs for the enolate forming reaction are
summarized in Table 4. We emphasize that these calculations
do not take into account solvent effects, which undoubtedly have
an important role in experimental situations. Nevertheless,
trends in the experimental isotope effects are well reproduced
by the calculations.
For ground state complexes qualitatively similar results are
given by computations at the various levels. For example, the
complexation energy is about 20 kcal/mol between the aldehyde
and hydroxide ion and is slightly smaller between the enolate
and water (note, however, that these values are not corrected
for basis set superposition errors33). The enol tautomer of
acetaldehyde is less stable than the keto tautomer by about 12-
15 kcal/mol, as found previously.34
The calculated activation energy for the proton transfer
reaction depends significantly on the computational level. The
reaction has a significant energy barrier at the RHF level, but
the barrier height is drastically reduced when electron correlation
is taken into account. In fact, at the highest computational level
the relative energy of the transition state is negative, i.e., the
reaction does not have a barrier. It should be noted, however,
that calculations at the MP3 and MP4(SDQ) level (results not
shown) predict small but positive energy barriers. Overall we
conclude that the energy barrier for proton transfer from
acetaldehyde to hydroxide ion is quite small in the gas phase.
For the competing nucleophilic addition reaction the calcula-
tions show that the tetrahedral adduct formed is less stable than
the enolate-water complex by about 5 kcal/mol and the reaction
has a small energy barrier (Table 3). In this case the dependence
on the computational level is not as great. This is consistent
with many other examples of computations dealing with
nucleophilic addition to carbonyl compounds.35 It is notable
that the IRC calculation from this transition state resulted in an
aldehyde-hydroxide ion complex of identical structure to the
one obtained in the proton transfer reaction. Also, both proton
transfer and nucleophilic addition reactions have small barriers.
Thus, at least for this particular system, there does not appear
to be a large intrinsic preference for one reaction over the other.
For the protonation of acetone, the computed equilibrium
isotope effects are summarized in Table 5. The optimized
good agreement with kHTH ) 4.7 - 5.4 × 10-3 M-1 s-1
reported previously.28 (2) For R-C-1H exchange, uncertainties
in the pseudo-first-order rate constants arise largely from
variations in resolution set by shimming the magnet. For
individual kinetic runs, the line widths did not change by more
than a few hundredths of a hertz during the course of the
experiments. This is consistent with the fact that the errors
estimated statistically from the goodness of fit are too small to
contribute to the overall observed error in the isotope effects.
(3) The simultaneous fitting of all three data sets places severe
restrictions on the quality of the data required for the fit to
converge.29 No restrictions on the amplitude or rate terms in
eqs 1-3 were required for the fits to converge rapidly. (4)
Systematic errors due to saturation effects caused by incomplete
relaxation between pulses were ruled out by showing there was
no change in the isotope effect when the relaxation delay times
were varied from 6 to 10 T1 for the R-CH2D peak and 3 to 6 T1
for the R-CHD2 peak. (5) We examined several methods of
accounting for the lack of complete resolution of the multiplets
including adding 0.1 Hz of line broadening, Gaussian weighting
of the FIDs, and using peak heights instead of integrals.12 These
had no effect on the isotope effects, within the overall
experimental error. (6) We examined the effect of dissolving
the acetophenone in D2O, followed by the addition of KCl and
doubly distilling the deuterium oxide under nitrogen to reduce
dissolved carbon dioxide.30 There was no change in the isotope
effect, within experimental error. Finally, we note that the
second-order rate constant for deuteroxide catalyzed exchange
of the first R-proton, kHHH ) kOD ) 0.208 M-1 s-1, when
combined with a value kOH ) 0.249 × 0.6 ) 0.149 M-1 s-1 in
H2O,31 gives a solvent isotope kOD/kOH ) 1.4, in good agreement
with solvent isotope effects for enolization reactions catalyzed
by lyoxide ion.32
Calculations. Computed energies of the stationary points
for the reactions between acetaldehyde and hydroxide ion are
summarized in Table 3, and Figure 4 shows relevant structures
(28) (a) Jones, J. R.; Marks, R. E.; Subba Rao, S. C. Trans. Faraday
Soc. 1967, 63, 111-119. (b) Jones, J. R. Trans. Faraday Soc. 1968, 64,
440-444.
(29) Ashwell, M.; Guo, X.; Sinnott, M. L. J. Am. Chem. Soc. 1992, 114,
10158-10166.
(30) Jencks, W. P.; Altura, R. A. J. Chem. Ed. 1988, 65, 770-771.
(31) (a) Chiang, Y.; Kresge, A. J.; Wirz, J. J. Am. Chem. Soc. 1984,
106, 6392-6395. (b) Calculated by correcting for the difference in the
activity coefficient of hydroxide ion with ionic strength using the Debye-
Hu¨ckel equation: -log f( ) 0.516I1/2/(1 + 1.15I1/2) at I ) 1.0 M. Lange’s
Handbook of Chemistry, 14th ed.; McGraw-Hill: New York, 1992, pp 8.4
and 8.5.
(33) Scheiner, S. In ReViews in Computational Chemistry; Lipkowitz,
K. B., Boyd, D. B., Eds.; VCH: New York, 1991; Vol. 2, pp 172-179.
(34) Wiberg, K. B.; Breneman, C. M.; LePage, T. J. J. Am. Chem. Soc.
1990, 112, 61.
(35) (a) Weiner, S. J.; Singh, U. C.; Kollman, P. A. J. Am. Chem. Soc.
1985, 107, 2219-2229. (b) Madura, J. D.; Jorgensen, W. L. J. Am. Chem.
Soc. 1986, 108, 2517-2527. (c) Blake, J. F.; Jorgensen, W. L. J. Am.
Chem. Soc. 1987, 109, 3856-3861. (d) Krug, J. P.; Popelier, P. L. A.;
Bader, R. W. F. J. Phys. Chem. 1992, 96, 7604-7616. (e) Francisco, J.
S.; Williams, I. H. J. Am. Chem. Soc. 1993, 115, 3746-3751. (f) Pranata,
J. J. Phys. Chem. 1994, 98, 1180-1184.
(32) Casamissina, T. E.; Huskey, W. P. J. Am. Chem. Soc. 1993, 115,
14-20.

+
+
Deuterium Isotope Effects for Enolization Reactions
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6567
Table 3. Relative Energies of Stationary Points along the Reaction Coordiantes for Reaction of Hydroxide Ion with Acetaldehydea
aldehyde aldehyde-OH- transition enolate-H2O enolate
enol
nucleophilic
nucleophilic
+ OH-
complex
state
complex
+ H2O + OH- addition TS addition product
RHF/6-31+G(d)
18.7
19.9
19.6
20.0
0
0
0
0
7.3
1.2
-0.6
-0.3
-18.1
-17.7
-22.0
-20.7
-1.3
2.2
-2.9
-1.6
34.7
35.5
31.6
33.0
2.7
1.3
1.2
0.6
-13.4
-15.0
-15.0
-14.7
MP2/6-31+G(d)
MP2/6-311++G(d,p)
MP4(SDTQ)/6-311++G(d,p)
a Energies are in kcal/mol relative to the aldehyde-OH complex, whose total energy is -228.35632, -228.97806, -229.12074, or -229.17690
au at the various computational levels listed in the table.
Table 4. Calculated Secondary Deuterium KIEs for Enolate Formation
MMI
ZPE
EXC
Q
overall KIE
EIEa
EIEb
RHF Frequencies
kHH/kHD
kHH/kDH
kHD/kDD
kDH/kDD
1.006
0.987
0.987
1.006
1.053
0.981
0.992
1.065
1.042
1.001
1.005
1.005
1.001
1.105
1.057
1.060
1.107
1.279
1.179
1.187
1.288
1.200
1.099
1.106
1.208
1.088
1.077
1.032
MP2 Frequencies
kHH/kHD
kHH/kDH
kHD/kDD
kDH/kDD
0.986
0.970
0.970
0.986
1.024
0.960
0.971
1.036
1.036
1.004
1.004
1.004
1.003
1.050
1.002
1.005
1.053
1.324
1.224
1.231
1.332
1.214
1.103
1.111
1.222
1.071
1.062
1.027
a Calculated from frequencies of isolated acetaldehyde and enolate. b Calculated from frequencies of the aldehyde-hydroxide ion and enolate-
water complexes.
of CHH and CHD are not significantly different from one
another. In the case of triton exchange, the effect of a single
deuterium substitution may be given by the rule of the geometric
mean: kHTH/kHD ) (kTHH/kDTD 1/2 ) 1.06 ( 0.02. Combined with
)
the corresponding values for proton exchange this gives a
weighted average isotope effect of a single R-deuterium sub-
stitution on the rate of ionization of a CH bond: kH/kD ) 1.06
( 0.01.37 We must however be cautious in applying the rule
of the geometric mean since it implies that the effect of isotopic
substitution is completely localized to the bond being substituted.
Indeed, the calculated isotope effects (Table 4) indicate a much
larger effect when the deuterium is syn to the carbonyl group.
In any event the calculated equilibrium isotope effects in Table
4 ranging from 1.10 to 1.32 are larger than this value.
There is extensive evidence for a substantial degree of proton
transfer to the base catalyst (Scheme 4) in the transition state
for enolization reactions as measured by Brønsted â coefficients
of ≈0.5.38 In addition, values of kOD/kOH that are intermediate
between 1 and the equilibrium solvent isotope effect for the
ionization of water are consistent with a proton that is half-
transferred.32 If rehybridization of the R-carbon is synchronous
with the degree of proton transfer to the base catalyst, the
R-secondary KIE should be intermediate between 1.0 and the
average equilibrium isotope effect of 1.10-1.32 (Table 4) as is
Figure 5. MP2-optimized structure of protonated acetone. Numbers
displayed are the C-H bond lengths (in angstroms) and the H-C-
C-O dihedral angles (in degrees).
structure of protonated acetone is shown in Figure 5. Proto-
nation causes an approximately 15-20° rotation of both methyl
groups in opposite directions, thus each of the six methyl
hydrogens is unique. Table 5 displays the isotope effects for
substitution at each of these six positions. Acetone itself is C2V-
symmetric at the RHF level; however, optimization at the MP2
level resulted in a structure with C2 symmetry, with the methyl
groups rotated by about 6°.
Discussion
The Nature of the Transition State for Base-Catalyzed
Enolization. The rate constants summarized in Table 2 lead
to three separate determinations of the effect of a single
R-deuterium substitution on the rate of ionization of a CH bond.
The isotope effects determined by 1H NMR for proton exchange
(37) Bevington, P. R.; Robinson, D. K. In Data Reduction and Error
Analysis for the Physical Sciences, 2nd ed.; McGraw-Hill: New York, 1992;
pp 58-60.
(38) (a) Bell, R. P. In The Proton in Chemistry, 2nd ed.; Cornell
University Press; Ithaca, New York, 1973; Chapter 10. (b) Bell, R. P.;
Grainger, S. J. Chem. Soc., Perkin Trans. 2 1976, 1367-1370. (c) Chiang,
Y.; Kresge, A. J.; Santabella, Wirz, J. J. Am. Chem. Soc. 1988, 110, 5506-
5510. (d) Stefanidis, D.; Bunting, J. W. J. Am. Chem. Soc. 1991, 113,
991-995.
(36) (a) Defrees, D. J.; Taagepera, M.; Levie, B. A.; Pollack, S. K.;
Summerhas, K. D.; Taft, R. W.; Wolfsberg, M.; Hehre, W. J. J. Am. Chem.
Soc. 1979, 101, 5532-5536. (b) Pross, A.; Radom, L.; Riggs, N. V. J.
Am. Chem. Soc. 1980, 102, 2253-2259. (c) Saunders, M.; Cline, G. W. J.
Am. Chem. Soc. 1990, 112, 3955-3963.

+
+
6568 J. Am. Chem. Soc., Vol. 118, No. 28, 1996
Table 5. Calculated Equilibrium Isotope Effects for the Protonation of Acetonea
Alston et al.
MMI
ZPE
EXC
overal EIE
RHF Frequencies
MMI
ZPE
EXC
overall EIE
Kh6/Kh15
Kh6/Kh25
Kh6/Kh35
Kh6/Kh45
Kh6/K55
Kh6/K56
Kh6/Kd15
Kh6/Kd25
Kh6/Kd35
Kh6/Kd45
Kh6/Kd65
Kh6/Kd45
1.026
0.999
0.999
0.994
0.999
0.999
1.026
1.089
0.936
1.013
1.061
0.955
1.000
0.979
1.006
0.998
0.983
1.001
1.020
1.066
0.940
1.005
1.042
0.955
0.993
0.987
0.988
0.992
0.987
0.987
1.034
0.980
1.132
1.049
1.004
1.110
0.973
0.991
0.969
0.974
0.987
0.973
1.000
0.958
1.084
1.014
0.979
1.067
MP2 Frequencies
Kh6/Kh15
Kh6/Kh25
Kh6/Kh35
Kh6/Kh45
Kh6/K55
Kh6/K56
Kh6/Kd15
Kh6/Kd25
Kh6/Kd35
Kh6/Kd45
Kh6/Kd65
Kh6/Kd55
0.999
0.999
0.998
1.001
0.999
0.998
1.033
1.059
0.932
1.011
1.039
0.951
0.990
0.983
1.004
0.990
0.985
1.000
1.022
1.040
0.933
1.001
1.022
0.949
0.994
0.994
0.995
0.993
0.995
0.995
0.974
0.952
1.077
0.996
0.970
1.055
0.969
0.975
0.959
0.969
0.973
0.962
0.939
0.923
1.027
0.958
0.938
1.010
a Kdn5 is the equilibrium constant for the reaction shown for Khn5 in the figures with replacement of D with H and vice versa.
Scheme 4
magnitude of these isotope effects in enolization reactions. The
computed KIEs are normal, qualitatively consistent with ex-
pectations for a reaction where a CH out-of-plane bending mode
changes from high frequency to low as the sp3-hybridized carbon
is converted to sp2. As shown in Figure 4, in the transition
state, the three remaining substituents on the R-carbon have bond
angles of about 112° with each other, consistent with a largely
tetrahedral sp3-like structure. In addition, the MP2-optimized
structure suggests an early transition state: the distance from
the hydrogen atom being transferred is 1.22 Å to the carbon
atom of the aldehyde and 1.52 Å to the oxygen of the hydroxide
(the RHF optimized structure suggests a somewhat later
transition state, the distances being 1.30 and 1.35 Å, respectively,
but this would still be considered early). The computed
structures provide support for the conclusion that a lag in
structural reorganization is partially responsible for the large
intrinsic barriers for proton transfers to and from carbon.39
observed. However, there is also extensive evidence for a lag
in structural and solvational reorganization associated with these
reactions, such that rehybridization at the R-carbon is expected
to lag behind proton transfer.39
The computed transition state structures and KIEs summarized
in Table 4 provide some insight into the origins of the small
The data in Table 4 show that deuterium substitution in the
position syn to the carbonyl oxygen (corresponding to the rate
constant kHD) leads to larger overall KIEs than for deuterium
substitution in the out-of-plane position (corresponding to kDH).
In the optimized structure of acetaldehyde, the C-H bond
lengths are 1.0916 Å for the in-plane (syn) position and 1.0962
Å for the out-of-plane position (these are MP2-optimized values;
the RHF values are 1.0818 and 1.0868 Å). The syn position is
(39) (a) Bernasconi, C. F. AdV. Phys. Org. Chem. 1992, 27, 116-238.
(b) Bernasconi, C. F.; Wentzel, P. J. J. Am. Chem. Soc. 1994, 116, 5405-
5413.
(40) Streitwieser, A., Jr.; Jagow, R. H.; Fahey, R. C.; Suzuki, S. J. Am.
Chem. Soc. 1958, 80, 2326-2332.

+
+
Deuterium Isotope Effects for Enolization Reactions
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6569
not affected by hyperconjugation with the carbonyl group and
has a correspondingly shorter CH bond. Acetaldehyde is a
textbook case of hyperconjugation: electron donation from the
CH orbital of the methyl group to the π* orbital of the carbonyl
strengthens the C-C bond and weakens the C-H bonds.
However, due to symmetry the in-plane C-H bond is not
affected.
The in-plane and out-of-plane CH bond lengths become more
nearly equal in the transition state with MP2 values of 1.0936
and 1.0956 Å and RHF values of 1.0863 and 1.0885 Å,
respectively. This is consistent with the notion that hypercon-
jugation decreases in the transition state compared to the
reactants. Decreased hyperconjugation means that the out-of-
plane CH bond is relatively stronger in the transition state, which
causes a tendency toward an inverse KIE. The net effect is a
smaller normal KIE for deuterium at the out-of-plane positions
(calculated to be kHH/kDH ) 1.057 (RHF) or 1.002 (MP2)).
Deuterium substitution at the syn position leads to a slightly
larger KIE: kHH/kHD ) 1.105 (RHF) or 1.050 (MP2). An
analogous argument regarding the compensating effects of
hyperconjugation and sp2-sp3 rehybridization has been made
to explain the small magnitude of secondary kinetic isotope
effects observed in the ketonization of enols.41
are observed when deuterium is substituted for hydrogen at the
position syn to the carbonyl oxygen. Protonation causes rotation
of both methyl groups by about 15-20°, so that one hydrogen
on each methyl group is in a nearly perpendicular position.
These hydrogens are most subject to hyperconjugation effects,
and are observed to have the longest bonds to carbon (Figure
5). In unprotonated acetone, these hydrogens are out of plane,
still affected by hyperconjugation, but not as much. Thus,
deuterium substitution at these position leads to an increase
isotope effect (Kh6/K3h5 and Kh6/Kh65). The in-plane hydrogen in
acetone ends up only slightly out of plane upon protonation,
resulting in no significant change in the hyperconjugation
effects, and the expected isotope effect is minimal (Kh6/K1h5
and Kh6/Kh45). The remaining hydrogen is subject to greater
hyperconjugation effects in the unprotonated species, resulting
in a normal isotope effect (Kh6/K2h5 and Kh6/Kh55). Thus, the
changes in hyperconjugation that lead to the computed isotope
effects can be attributed primarly to geometric changes, rather
than any inherent electronic effects. The results shown in Table
V are consistent with this analysis. The actual observed isotope
effect would presumably be some weighted average of these.
Since substitution at different sites results in different effects,
predicting the effects of multiple deuterium substitution is not
straightforward. For the d5 isotopomer, MP2 calculations appear
to predict an overall small, inverse isotope effect, qualitatively
in agreement with the experimental results.
Conclusions. The effect of secondary R-deuterium substitu-
tion on the rate of ionization of a CH bond is small for
enolization reactions, consistent with a loss in hyperconjugation
as well as a lag in structural reorganization at the R-carbon,
which gives rise to a large intrinsic barrier for proton transfer
from carbon. The equilibrium isotope effect of 0.97/D on the
acidity constant of protonated acetone is consistent with a small
increase in hyperconjugation between the R-CH bond and the
positively charged carbonyl group upon protonation. These
isotope effects serve as limiting models for interpretation of
corresponding secondary isotope effects in enzyme reactions
which will be reported in further communications.
Unfortunately, there are still ambiguities in the calculations:
in the aldehyde-hydroxide ion complex, the corresponding MP2
bond lengths are 1.0950 and 1.0976 Å, and their difference is
intermediate between that for isolated acetaldehyde and the
transition state; however, the RHF bond lengths are 1.0864 and
1.0873 Å, the difference being smaller than in the transition
state.
â-Deuterium Isotope Effects. Protonation of Acetone.
The â-deuterium equilibrium isotope effect Kh6 +/Kd5 ) 0.87
+
BH
BH
( 0.04 for ionization of protonated acetone is consistent with
a loss of hyperconjugation upon going from protonated acetone
to acetone in concentrated acid solutions. The observed isotope
effect corresponds to an equilibrium isotope effect of (0.87)1/5
) 0.97 per deuterium, if the effects of deuterium substitution
are additive. This value is similar to the calculated â-deuterium
isotope effects in Table 5 (average Kh6/Kd5 ) 0.99 at the MP2
level) and values of 0.93-0.96 per deuterium for the equilibrium
(CL3)2C+ + X- a (CL3)2C-X.42 These values are less inverse
Acknowledgment. We gratefully acknowledge the National
Institutes of Health (GM 43251) and the Petroleum Research
Fund administered by the American Chemical Society for
support of this work. Kari Haley was supported by a summer
undergraduate research fellowship from the National Science
Foundation (CHE-9200036). We thank Professors A. Bagno
and James Keeffe for helpful discussions and Marvin Leister
for help in obtaining the NMR spectra.
than the EIE for the acidity constant of protonated acetophenone,
h3
BH
K
+/Kd3 ) (0.78)1/3 ) 0.92/D.6 This latter value was
+
BH
determined using the Hammett indicator method and is subject
to the experimental uncertainties associated with that tech-
nique.19 Comparison of the inverse isotope effect with EIE for
ketone to alcohol equilibria of 0.95/D that corresponds to
complete loss of hyperconjugation43 suggests that there is a small
increase in hyperconjugation in protonated acetone relative to
acetone.
Supporting Information Available: Table S1 summarizing
chemical shift differences of acetone and acetone-d5 as a
function of weight percent sulfuric acid, Tables S2 and S3
summarizing pseudo-first-order rate constants for acetophenone
proton exchange in D2O and triton exchange in H2O, and Tables
S3 and S4 summarizing RHF and MP2 optimized structures
(12 pages). Ordering information is given on any current
masthead page.
The calculated equilibrium isotope effects for the protonation
of acetone summarized in Table 5 show again that larger effects
(41) Andraos, J.; Kresge, A. J.; Obraztsov, P. A. J. Phys. Org. Chem.
1992, 5, 322-326.
(42) Williams, I. H. J. Chem. Soc., Chem. Commun. 1985, 510-511.
(43) Cook, P. F.; Blanchard, J. S.; Cleland, W. W. Biochemistry 1980,
19, 4853-4858.
JA942053K