Enolization of aldehydes and ketones: Structural effects on concerted acid-base catalysis

Article abstract of DOI:10.1021/ja9729544

The third-order term (k(AB)) for the concerted acid-base catalyzed enolization of a selection of simple aldehydes and ketones has been measured in a series of substituted acetic acids at 25°C at constant ionic strength 2.0 (NaNO₃). While there is no direct correlation of the magnitude of the third-order term with either the rate constants for acid (k(A)) or base (k(B)) catalysis, a simple log-log relationship exists between the product of the consecutive rate constants (k(A)·k(B)) and the concerted (third order) rate constants (k(AB)). This implies that the concerted pathway is important only when both the general acid and the general base terms are significant; this will be useful in designing other systems which might show such concerted catalysis. In the case of aldehydes, a slope of 0.97 was found for this plot, which compares to the result for 4-substituted cyclohexanones (0.51) and other ketones (0.59), as measured in acetic acid buffers. The resultant Bronsted β(AB) value of 0.20 found for propanal (2) is consistent with the overall observation that concerted catalysis is largely independent of the buffering species, and that process is overall base catalyzed. The solvent isotope effect on the concerted acid-base catalyzed enolization rate term, k(AB)(H₂O)/k(AB)(D₂O) = 1.33, indicates that the transition state for proton transfer to the carbonyl is more advanced than in the case of ketones. In general we have found that carbonyl compounds with large measured (or estimated) enol contents show significant third-order terms.

Full text of DOI:10.1021/ja9729544

2290
J. Am. Chem. Soc. 1998, 120, 2290-2296
Enolization of Aldehydes and Ketones: Structural Effects on
Concerted Acid-Base Catalysis
Anthony F. Hegarty,* Joseph P. Dowling, Stephen J. Eustace, and Michelle McGarraghy
Contribution from the Chemistry Department, UniVersity College Dublin, Belfield, Dublin 4, Ireland
ReceiVed August 22, 1997. ReVised Manuscript ReceiVed December 23, 1997
Abstract: The third-order term (kAB) for the concerted acid-base catalyzed enolization of a selection of simple
aldehydes and ketones has been measured in a series of substituted acetic acids at 25 °C at constant ionic
strength 2.0 (NaNO3). While there is no direct correlation of the magnitude of the third-order term with either
the rate constants for acid (kA) or base (kB) catalysis, a simple log-log relationship exists between the product
of the consecutive rate constants (kA‚kB) and the concerted (third order) rate constants (kAB). This implies that
the concerted pathway is important only when both the general acid and the general base terms are significant;
this will be useful in designing other systems which might show such concerted catalysis. In the case of
aldehydes, a slope of 0.97 was found for this plot, which compares to the result for 4-substituted cyclohexanones
(0.51) and other ketones (0.59), as measured in acetic acid buffers. The resultant Brønsted âAB value of 0.20
found for propanal (2) is consistent with the overall observation that concerted catalysis is largely independent
of the buffering species, and that process is overall base catalyzed. The solvent isotope effect on the concerted
acid-base catalyzed enolization rate term, kAB(H2O)/kAB(D2O) ) 1.33, indicates that the transition state for
proton transfer to the carbonyl is more advanced than in the case of ketones. In general we have found that
carbonyl compounds with large measured (or estimated) enol contents show significant third-order terms.
Introduction
Scheme 1
The most reliable method for evaluation of tautomeric
equlibrium constants, KE ([enol]/[ketone]), is by direct measure-
ment of forward (kE) and reverse (kK) rate constants, measured
under the same conditions. Rate measurements for the eno-
lization of ketones (kE) are easily determined, for example, by
halogen scavenging.1 However, it is only within the past decade
that reliable data for the ketonization rate constants (kK) have
been obtained. Photochemical generation of enols has allowed
the accurate determination of equilibrium constants for many
monocarbonyl compounds.2
K ) k /k
K
(1)
E
E
Tautomerization can occur by both acid and base catalyzed
mechanisms, which have been well established.3,4 As an alterna-
tive to these well-accepted consecutive acid-base catalyzed
pathways, the possibility of a concerted mechanism involving
both catalytic species is plausable. This is represented in
5
(1) (a) Lapworth, A. J. Chem. Soc. 1904, 30. (b) Dubois, J. E.; Toullec,
J. Tetrahdron 1973, 29, 2859. (c) Dubois, J. E.; El-Aloui, M.; Toullec, J.
J. Am. Chem. Soc. 1981, 103, 5393.
Scheme 1, where the partial charges on the transition state were
calculated from Brønsted values given by Hegarty and Jencks.5
To date, data are available only for a few aldehydes and ketones
for this possible pathway so that there is little evidence on how
f
(
2) (a) Haspra, P.; Sutter, A.; Wirz, J. Angew. Chem., Int. Ed. Engl. 1979,
1
8, 617. (b) Chiang, Y.; Kresge, A. J.; Tang, Y. S.; Wirz, J. J. Am. Chem.
Soc. 1984, 106, 460. (c) Chiang, Y.; Kresge, A. J.; Wirz, J. J. Am. Chem.
Soc. 1984, 106, 6392. (d) Chiang, Y.; Hojatti, M.; Keefe, J. R.; Kresge, A.
J.; Schepp, N. P.; Wirz, J. J. Am. Chem. Soc. 1987, 109, 4000. (e) Keefe,
J. R.; Kresge, A. J.; Schepp, N. P. J. Am. Chem. Soc. 1988, 110, 1993. (f)
Chiang, Y.; Kresge, A. J.; Santaballa, J. A.; Wirz, J. J. Am. Chem. Soc.
6
the magnitude of kAB varies with structure. There are limita-
tions, however, on the number of ketones which are sufficiently
water soluble to permit direct measurement of enolization rates
in the absence of cosolvents. Moreover, unsymmetrically
substitued ketones can in most cases enolize on either side of
1
988, 110, 5506. (g) Hochstrasser, R.; Kresge, A. J.; Schepp, N. P.; Wirz,
J. J. Am. Chem. Soc. 1988, 110, 7875. (h) Chiang, Y.; Kresge, A. J.; Schepp,
N. P. J. Am. Chem. Soc. 1989, 111, 3977. (i) Chiang, Y.; Kresge, A. J.;
Walsh, P. A. J. Am. Chem. Soc. 1986, 108, 6314. (j) Chiang, Y.; Kresge,
A. J.; Walsh, P. A.; Yin, Y. J. Chem. Soc., Chem. Commun. 1989, 869.
(5) (a) Dawson, H. M.; Spivey, E. J. Chem. Soc. 1930, 2180. (b) Bell,
R. P.; Jones, P. J. Chem. Soc. 1953, 88. (c) Lowry, T. M.; Faulkner, I. J.
J. Chem. Soc. 1925, 127, 2883. (d) Swain, C. G. J. Am. Chem. Soc. 1950,
72, 4578. (e) Hand, E.; Jencks, W. P. J. Am. Chem. Soc. 1975, 97, 6221.
(f) Hegarty, A. F.; Jencks, W. P. J. Am. Chem. Soc. 1975, 97, 7188.
(6) Hegarty, A. F.; Dowling, J. J. Chem. Soc., Chem. Commun. 1991,
996.
(
3) (a) Bell, R. P. Acid-Base Catalysis; Oxford University Press: Oxford,
941. (b) Jones, J. R. The Ionisation of Carbon Acids; Academic Press:
London, 1973.
1
(4) (a) Pederson, K. J. J. Phys. Chem. 1934, 38, 581. (b) Lienhard, G.
E.; Wang, T. C. J. Am. Chem. Soc. 1969, 91, 1146. (c) Toullec, J. AdV.
Phys. Org. Chem. 1982, 18, 1.
S0002-7863(97)02954-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/25/1998

Enolization of Aldehydes and Ketones
J. Am. Chem. Soc., Vol. 120, No. 10, 1998 2291
the carbonyl group, which could make the interpretation more
difficult. We have therefore studied both aldehydes and
4-substituted cyclohexanones, despite the experimental difficul-
ties involved. Taken together, the chosen acyclic and cyclic
ketones, together with the aldehydes, give a reasonable structural
variation. Moreover, it has recently been predicted on theoreti-
7
cal grounds that aldehydes would show marked concerted
catalysis, and the results which we now report have confirmed
this.
In general, contributions from each of the catalyzed pathways
may be described by eq 2. This may also be expressed more
simply, in eq 3, in terms of the observed rate constant, kobs, as
measured in the presence of a catalytic buffer, AH, which
exhibits general acid, general base, and concerted catalysis,
denoted by kA, kB, and kAB, respectively. The buffer independent
rate constant, which of course includes terms for hydronium,
hydroxide, and water catalysis, is denoted by ko.
+
d[KH]/dt ) [KH]{∑i kBHi+
[BH ] + k [B ] +
∑i
i
B
i
i
+
∑i kB ,BH
+
[B ][BH ]}(2) (2)
i i
i
i
kobs ) k + k [AH] + k [A ] + k [AH][A-]
-
(3)
o
A
B
AB
Results
Figure 1. First-order rate constants for the enolization of cyclohex-
anone (17) plotted against the acetate anion concentrations.
We have extended our study to include a series of substituted
cyclohexanones. Available keto-enol equilibrium constants are
listed in Table 1, along with second- and third-order buffer-
catalyzed rate constants, measured in acetic acid in aqueous
solution. Of particular interest is the observation that the same
kAB term is obtained at 25%, 50%, and 80% of the buffer
ionized, which would rule out artifacts such as salt or solvent
effects giving rise to the curvature of the observed rate vs
concentration plots (see Figure 1). Specific salt effects can cause
upward or downward deviations in plots of rate against buffer
concentration if the buffer salt and the salt used to maintain
ionic strength constant have different effects on the ratios of
activity coefficient for the reactants and transition state. Thus
increase in the pKaK of the keto tautomer and hence the base
catalyzed rate constant, kB, of enolization decreases. This is
probably due to the electron donating inductive effect of the
methyl group making the forming carbanion less stable.4c
Similarly other electron donating substituents are expected to
have the same effect. The reduction in the size of the base-
catalyzed rate constant could also be explained in terms of
increasing steric hindrance to the approaching base.
A preliminary examination was carried out on the ketone
3-hydroxy-3-methyl-2-butanone (11) as a model for the aromatic
-
-
carbonyls. This compound was chosen on steric grounds as it
is expected to interfere with concerted catalysis much as an
aromatic ring would. The compound however fits the loga-
rithmic relationship between the consecutive and concerted rate
constants, with a slope of 0.59, see Figure 2. The rate constant
for the base-catalyzed reaction is quite substantial even com-
pared to that for acetone (8). The value for the acid-catalyzed
rate constant is smaller than that for the unsubstitued butanone
NO 3 seems to be a good model for CH3CO2 in this instance.
At the highest buffer concentrations up to 10% acetic acid was
present, which might have induced a medium effect (“solvent
sorting”), but the observation that the same kAB term is obtained
at high and low pH indicates that this is not important.
There were limitations on the number of ketones that could
be studied under the present conditions; the substrates were
required to be sufficiently water soluble to permit direct
measurement of the enolization rates in the absence of cosolvent.
For example, 4-chlorocyclohexanone was synthesized but was
not included in the study for this reason. Moreover unsym-
metrically substituted ketones can in most cases enolize on either
side of the carbonyl group, making interpretation more difficult.
This study also encompasses a number of aromatic ketones and
aldehydes, despite the experimental difficulties involved in the
measurement of the latter compounds. For all the compounds
examined the second-order rate constant for the base-catalyzed
reaction was greater than that catalyzed by acid, see Table 1.
Manipulation of the structure can have different effects on
the magnitude of the pKa of the carbonyl, enol, or the protonated
carbonyl species and subsequently on the value of the different
rate constants. Increasing the alkyl chain length causes an
(9); this is probably a consequence of the electron-withdrawing
hydroxy substituent, which would be expected to cause a
K
decrease in pKa . Although the rate constant for the base-
-
8
-2
catalyzed mechanism is dominant, a value of 7.5 × 10
M
-1
s
was calculated for the concerted pathway. Thus, it appeared
likely that even though the substituents were bulky they did
not interfere in this case significantly with the concerted route.
Discussion
It is useful to explain the observation of concerted catalysis
in terms of a three-dimensional free energy diagram, as seen in
7
a
the More O'Ferrall-Jencks diagram in Figure 3. The struc-
tures of the product (top right-hand corner) and the intermediates
in the consecutive acid- and base-catalyzed reactions are given
with the associated energies relative to the reactant in (kcal/
mol). The concerted reaction is independent of either of the
intermediates shown, and is represented by the diagonal line
(
7) (a) Albery, W. J.; Gelles, J. S. J. Chem. Soc., Faraday Trans. 1982,
8, 1569. (b) Kresge, A. J.; Keefe, J. R. The Chemistry of Enols; John
Wiley & Sons: Chichester, 1990; pp 399-480.
7

2292 J. Am. Chem. Soc., Vol. 120, No. 10, 1998
Hegarty et al.
Table 1. Second- and Third-Order Rate Constants for Acetic Acid Buffer Catalysis in the Enolization of Aldehydes and Ketones in Aqueous
Solution at 25 °C and Ionic Strength 2.0 M
a
Units M-1
-1. b Units M-2 s-1. c Reference 2d. d Reference 2i. e Reference 2j. Total (E and Z) enol content. g Reference 2e. h Reference 9a.
f
s
i
j
k
Reference 5f. Reference 9c. Reference 2c.
for a hypothetical situation with a Brønsted âAB ) 0. Experi-
mentally, the reaction shows a net base catalysis with a Brønsted
âAB ) +0.15.5 This is also represented as the transition state
to the right of the diaganol, indicating that proton donation to
the base is running slightly ahead of protonation of the carbonyl
group. The concerted reaction, while entropically unfavorable,
f
5f

Enolization of Aldehydes and Ketones
J. Am. Chem. Soc., Vol. 120, No. 10, 1998 2293
or by the introduction of an electron withdrawing group, an
increase in the third-order term is observed. It is apparent that
when the acid catalyzed and base catalyzed terms are both large
that is when the kAB term is most significant.
Although accurate thermodynamic data are now generally
available for many ketones, especially the pKE values, it is only
in recent years that reasonable methods have been developed
for the evaluation of pKa(
OH+)
values of protonated ketones,
although several experimental problems still remain regarding
9,10
aldehydes.
We have looked for various relationships between
kAB and structural or other thermodynamic or reactivity data
and have found that there is a simple relationship between log
kAB and log kA + log kB. This implies that kAB is largest only
when kA and kB are both significant; while a larger variation is
found in kB than in kA (43-fold and 5-fold, respectively), the
correlation between kAB and either kB or kA alone shows scatter.
In this respect the availability of data for the cyclic ketones is
particularly useful since the acyclic ketones have kA values
which are all quite similar. Consistent with analysis, cyclobu-
tanone (12) and 3-pentanone (10) do not show a measurable
third-order term within the limits of experimental error, while
the kinetic data for cyclobutanone (12) could be correlated by
using kB alone.
Figure 2. Correlation showing the log-log relationship between k
and kAB, for a selection of nonaromatic ketones.
‚k
A B
The aromatic ketones acetophenone (15) and 4-acetylbenze-
nesulfonic (16) acid (sodium salt) were also studied. Substantial
rate constants were observed for both of the consecutively
catalyzed mechanisms for each of the aromatic ketones, but no
concerted catalysis was observed.
Aldehydes. Compared to the study of keto-enol tautomer-
ism of ketones, the tautomerization of aldehydes has been much
less extensively studied, probably because the enolization
processes are often complicated by oxidation and hydration.
Arguments based on free energy relationships (Figure 3) predict
that aldehydes should show a marked concerted catalysis. In
these cases the product corners of the free energy diagrams will
be much lower and there will be a greater energetic advantage
to avoiding the intermediates along the consecutive routes.
Consistent with this, the kAB term for acetaldehyde (1) was found
to be an order of magnitude greater than that for acetone (8).
Replacing the electron donating methyl group next to the
carbonyl by a hydrogen, as is the case in going from acetone
(8) to acetaldehyde (1), causes a decrease in pKa of the keto
tautomer which in turn shows as an increase in the overall base-
catalyzed rate constant. As in the case of the ketones, the simple
logarithmic relationship between kA‚kB and kAB was obeyed
(Figure 4), but with an increased slope (0.97).
Aldehydes are expected to be less basic than their ketone
9
c
Figure 3. More O'Ferrall-Jencks diagram for the enolization of acetone
8) catalyzed by acetic acid, energies in kcal/mol.
counterparts, for example, benzaldehyde (pKa ) -4.48) is less
9
c
(
basic than acetophenone (15) (pKa ) -3.87). Conversely,
the pKa values for aldehydes for C-deprotonation are generally
lower than their ketone counterparts, and since they are more
acidic, the values for the base-catalyzed rate constants are quite
substantial for the aldehydes. Although aldehydes are generally
less basic than ketones, the acid-catalyzed rate constant for the
enolization reaction, kA, does not vary significantly. This
avoids the high-energy intermediates associated with the
consecutive reactions. This is only true for weak acid-base
pairs, and agrees with the observation of concerted catalysis
only for acids of pKa ) 2.8. Finally, the solvent isotope effect
kAB(H2O)/kAB(D2O) ) 2.0 is also quite different from that
expected for a stepwise reaction.
By this analysis it is anticipated that stable enols are also
more likely to use the concerted route because in these cases
the product corners will be lower, increasing the energetic
advantage by avoiding the consecutive pathway intermediates.
This agrees with the findings of this study. In general those
enols which have, or are expected to have, a large KE value
also show a large concerted rate constant. A trend in the
magnitude of the kAB term has emerged: on reducing the basicity
of the carbonyl group, i.e., in going from a ketone to an aldehyde
(
8) (a) Arnett, E. M.; Joris, L.; Mitchell, E.; Murty, T. S. S. R.; Gorrie,
T. M.; Scheleyer, P. v. R. J. Am. Chem. Soc. 1970, 92, 2365. (b) Arnett, E.
M.; Mitchell, E. J.; Murty, T. S. S. J. Am. Chem. Soc. 1974, 96, 3875.
(9) (a) Levi, A.; Modena, G.; Scorrano, G. J. Am. Chem. Soc. 1974, 96,
6
585. (b) Bagno, A.; Lucchini, V.; Scorrano, G. Bull. Soc. Chim. Fr. 1987,
563. (c) Bagno, A.; Lucchini, V.; Scorrano, G. J. Phys. Chem. 1991, 95,
345.
(
Lucchini, V.; Modena, G.; Scorrano, G.; Cox, R. A.; Yates, K. J. Am. Chem.
Soc. 1982, 104, 1958. (c) Bagno, A.; Scorrano, G.; More O'Ferrall, R. A.
ReV. Chem. Intermed. 1987, 7, 313.
10) (a) Cox, R. A.; Yates, K. J. Am. Chem. Soc. 1978, 100, 3861. (b)

2294 J. Am. Chem. Soc., Vol. 120, No. 10, 1998
Hegarty et al.
Figure 4. Correlation showing the log-log relationship between k
and kAB for a selection of aldehydes (compound 5 was excluded).
‚k
A B
Figure 6. Correlation showing the log-log relationship between k
and kAB for a selection of 4-substituted cyclohexanones.
‚k
A B
analogous aliphatic aldehyde, the acid-base catalyzed rate
constant for the aromatic aldehyde was somewhat larger than
that for propanal (2).
Cyclohexanones. In the series of cyclohexanones studied it
is expected that the substituents, even though in the γ-position,
will destabilize the keto tautomer, resulting in larger KE values,
as a result of polar effects. The substituted cyclohexanones,
with the exception of 4-carboxycyclohexanone (22), all exhibit
a greater value for the third-order term than those previously
reported for other aliphatic ketones, the enol content of which
is on average one or two orders of magnitude less than that
expected for the substituted cyclohexanones. These compounds
also conform to the log-log relationship (Figure 6) that was
found in the case of aldehydes and ketones, with a slope of
0
.51. The kA value for the 4-substituted cyclohexanones is close
Figure 5. Brønsted plot of the third-order rate constants for the
enolization of propanal 2 catalyzed by a selection of substituted acetic
acids.
to that for cyclohexanone (17), which indicates that there is a
partial cancellation of effects in the acid-catalyzed reaction in
contrast to that for the base-catalyzed reaction. The magnitude
of the rate constant for the base-catalyzed enolization reaction
is, however, much greater than that reported for any of the
unsubstituted cycloalkanones.
suggests that the increased rate of C-deprotonation of the
protonated species compensates for the drop in carbonyl basicity.
Brønsted Exponent. The third-order rate constant, kAB, for
the enolization of propanal (2) was determined for a variety of
Conclusion
-
7
-2 -1
substituted acetic acids: cloroacetic acid, 5.63 × 10
M
s ;
-
6
-2 -1
The third-order term for the concerted acid-base catalyzed
enolization is readily observable in the case of simple aldehydes
and ketones. Third-order terms are observed for weak acid base
pairs (pK G 2.8), where both the general acid and general base
terms are significant. This is reflected quantitatively in a simple
log-log relationship that exists between the product of the
consecutive rate constants (k ‚k ), and the concerted rate
constant (kAB). This is more significant in the case of aliphatic
aldehydes, with a dependence of approximately 1.0, which
compares to the value obtained for cyclic and acyclic ketones,
which have a dependence of approximately 0.5, and finally in
the case of aromatic ketones, no third-order term is observed.
These results would tend to imply that steric effects may be
important, although at this point the exact structure of the
transition state is not clear. There also appears to be a
relationship between the third-order term and the enol content
of the carbonyl compound when comparing structurally similar
compounds.
methoxyacetic acid, 1.22 × 10 s ; ethoxyacetic acid,
M
-
6
-2 -1
-7
-2 -1
1
.70 × 10
M
s
; glycolic acid, 6.12 × 10
s ; acetic acid, 9.97 ×
s . The
M
s ;
-
6
-2 -1
chloroacetic acid, 1.03 × 10
M
-
7
-2 -1
-6
-2 -1
1
0
M
s ; and propanoic acid, 2.93 × 10
M
a
resultant Brønsted plot (Figure 5) gives a âAB value of 0.2. This
compares with the value obtained for the enolization of the
5f
ketone acetone (8), 0.15, reported by Hegarty and Jencks. This
is consistent with the overall observation that concerted catalysis
is largely independent of the buffering species, and that process
is overall, net base catalyzed. The solvent isotope effect on
the concerted acid-base catalyzed enolization rate term,
kAB(H2O)/kAB(D2O) ) 1.33, is smaller than that observed for
A
B
5f
acetone (8), which indicates that at the transition state, proton
transfer to the carbonyl is more advanced in this case.
Aromatic Aldehydes. There was no measurable acid-base
catalyzed term in the case of the aldehyde, 2-phenylethanal (6).
However, a significant contribution to the overall rate was made
by the concerted rate constant in the case of the longer chain
3
-phenylpropanal (7). It may be inferred from this result that
Experimental Section
even though larger concerted catalyzed terms are expected for
aldehydes than ketones, the concerted route may not always be
viable (probably for steric reasons). When the values obtained
for 3-phenylpropanal (7) were compared to those for the
Kinetics. All rate constant were measured in aqueous
solution at 25 °C. Acetic acid buffers were prepared by partial
neutralization of the acid with aqueous potassium hydroxide,

Enolization of Aldehydes and Ketones
J. Am. Chem. Soc., Vol. 120, No. 10, 1998 2295
and ionic strengths were maintained at 2.0 M by the addition
of the appropriate quantity of potassium nitrate. The rates of
enolization of the aldehydes and ketones were measured using
the iodine trap method which depends on a rapid and irreversible
[ketone or aldehyde]) is related to the Taft σ* values (eq 6),
where ∆ is the number of aldehydic hydrogens. This predicts
that the percentage of hydration of 3-chloropropanal (3) should
be 75%. Although the percentage hydration of acetone (8)
-
14g
7b
trapping of the enol form by I2/I3 . The concentration of I2/
(0.2%) and cyclopentanone (13) (0.2%) reported in water
at 25 °C is small, a value of 7.4% has been reported for
-
11
I3 was measured at an isosbestic point (351 nm) where ꢀ )
4
7b
(
2.6 ( 0.1) × 10 by using the equilibrium constant for triiodide
cyclohexanone. No correction was made to take account of
-
1 5f
formation appropriate to µ ) 2.0 (630 M ). The carbonyl
the small amount of hydrate in the case of ketones. This
approximation does not appreciably change the correlations
observed, since concentration appears in both coordinates.
2
3
compounds were used in a 10 -10 )-fold excess. Under these
conditions reversal of iodination is negligible12 and iodine,
unlike bromine or chlorine at low pH, does not induce oxidation
h
2
c,13
of the aldehyde.
The zero-order rate of iodine uptake was
pK ) 1.70Σσ* + 2.03∆ - 2.81
(6)
measured in deoxygenated solutions and was reproducible under
these conditions. The individual rate constants were obtained
from buffer dilution curves, typically covering 12 buffer
dilutions over the range 0.01 to 2.0 M total buffer concentration.
However, due to the dominance of the base-catalyzed and the
third-order terms in the ionization of aldehydes, the kA values
were not obtained with the same precision and are accurate only
to (20%. Plots of absorbance against time were linear in each
case from 10 s after mixing to at least 90% iodine consumption.
Independent experiments showed that the observed rate was
independent of the initial iodine concentration and was propor-
tional to the ketone or aldehyde concentrations. The observed
rate constants (see eq 4) were obtained with eq 5, where dA/dt
is the rate of change of absorbance at 351 nm, where iodide
was maintained in a constant large excess.
All the aldehydes were distilled under nitrogen and used fresh.
As has been noted, acetaldehyde (1) in particular causes
experimental difficulties if not pure, due to the formation of
2
c
oligomers, such as a cyclic trimer. It was found that when
the rate of iodine uptake was measured in air there was an initial
rapid generation of iodine when the aldehyde was added. This
was most marked with acetaldehyde (1), but decreased with
increasing chain length. This initial increase in absorption could
be eliminated by following the rates of reaction in sealed
cuvettes using solutions which had been flushed with oxygen
free nitrogen. There was a residual initial small increase in the
case of acetaldehyde, but the experimental pseudo-zero-order
rate constant was not affected by this.
In acetic acid buffer solutions the observed first-order rate
constants can be written as in eq 3. Since Ka is the acidity
constant for the ionizaton of the acid, this can be rewritten as
d[I ]/dt ) kobs[ketone]
(4)
(5)
2
-
+
eq 7, or eq 8 where r is the buffer ratio [AH]/[A ] ) [H ]/
Ka. This is a quadratic curve and kAB was obtained by using a
least-squares computer program, using the values of kA and kB
from the initial linear portion of the buffer dilution plots, where
the contribution of the kAB term is negligible.
-
kobs ) 1/[ketone]ꢀI3
-
{1 + (1/KI3 [I ])} dA/dt
-
The concentration of free aldehyde in solution was obtained
by correcting for the hydrate, measuring aldehyde/hydrate ratios
in 2.0 M aqueous KNO3 as follows: acetaldehyde (1), 45%
hydrate; propanal (2), 41.5% (H2O), 44.7% (D2O); 3-chloro-
propanal (3), 80%; 2-methylpropanal (4), 33.3%; and butanal
+
-
-
+
- 2
kobs ) k + k [H ][A ]/K + k [A ] + k [H ][A ] /K
a
o
A
a
B
AB
(7)
(
5), 31.5%. These values were obtained by integration of the
-
- 2
kobs ) k + {kA
o
•
r + k }[A ] + k
B
•
r[A ]
AB
(8)
R and â protons of the aldehyde and the hydrate at 270 MHz,
when present. Although it is known that the hydrate concentra-
tion decreases as the ionic strength increases,14a the hydrate
Buffer ratios of 25% (r ) 3), 50% (r ) 1), and 80% base
r ) 0.25) were used in each case. Similar results were obtained
(
concentration was shown not to vary once ionic strength was
kept constant. The percentage hydrate obtained where they
could be compared with literature values at 25 °C in water (m
)
2
not obtain a reproducible value for 3-chloropropanal (3) so the
values for the rate constants in Table 1 were calculated by
assuming 0% hydrated. Electron-withdrawing substituents are
known to increase the percentage of hydrate at equilibrium, and
-
-
by using plots of (kobs - ko)/[AcO ] against [AcO ], which
gave rkAB as slope and (kAr + kB) as intercept; such plots were
useful in determining the relative significance of each term. In
the case of the aldehydes, the kA term is in all cases a relatively
small contributor (typically 10%) to the reaction flux and the
error in this term is correspondingly larger ((20%).
1
4b,c
0) compare quite closely: acetaldehyde (1) 51.5%
and
-methylpropanal (4), 30%.1
4d,e
It is surprising that we could
Commercially available substrates were freshly distilled
immediately prior to use. All other substrates were prepared
by suitable modifications of existing literature procedures.15
Melting points were determined on a Gallenkamp melting point
block or on a B u¨ chi 530 melting point apparatus and are
uncorrected. Elemental analyses of compounds were performed
by the Microanalytical Laboratory, University College Dublin.
1
4f
h
Luz and Samuel have shown that the pK () -log[hydrate]/
(
11) Harper, E. T.; Bender, M. L. J. Am. Chem. Soc. 1965, 87, 5625.
(12) Carey, A. R. E.; Fukata G.; More O'Ferrall, R. A.; Murphy, M. G.
J. Chem. Soc., Perkin Trans. 2 1985, 1711.
13) (a) Dawson, H. M.; Burton, D.; Ark, H. J. Chem. Soc. 1914, 105,
275. (b) Goswami, G.; Banerji, K. K. Z. Naturforsch. B 1971, 26, 383. (c)
(
1
1
H NMR spectra were recorded at 60 MHz on a JEOL JNM-
McTigue, P. T.; Gruen, L. C. Aust. J. Chem. 1963, 16, 177. (d) McTigue,
P. T.; Sime, J. M. Aust. J. Chem. 1967, 20, 905. (e) Talvik, A. J.; Hiidmaa,
S. O. Org. React (USSR) 1968, 5, 121, 297 (f) Baigrie, L. M.; Cox, R. A.;
Slebocka-Tilk, H.; Tencer, M.; Tidwell, T. T. J. Am. Chem. Soc. 1985,
(15) (a) MacLeod, A. J.; Rossiter, J. T. J. Chem. Soc., Perkin Trans 1.
1983, 717. (b) Aldersley, J. B.; Burkhardt, G. N.; Gillam, A. E.; Hindley,
N. C. J. Chem. Soc. 1940, 10. (c) Jones, E. R. H.; Sondheimer, F. J. Chem.
Soc. 1949, 614. (d) Bowers, A.; Jones, T. G.; Jones, E. R. H.; Lemin, A. J.
J. Chem. Soc. 1953, 2548. (e) Hunziker, F.; M u¨ llner, F. X.; Schattenger,
H. HelV. Chim. Acta 1955, 38, 1943. (f) Dominguez, X. A.; Lopez, I. C.;
Franco, R. J. Org. Chem. 1961, 26, 1625. (g) House, H. O.; Babad, H.;
Toothill, R. B.; Noltes, A. W. J. Org. Chem. 1962, 27, 4141. (h) Weinberg,
N. L. Can. J. Chem. 1965, 43, 24. (i) Campbell, N. R.; Hunt, J. H. J. Chem.
Soc. 1950, 1379. (j) Overberger, C. G.; Kozlowski, J. H.; Radlmann, E. J.
Polym. Sci. A-1 1972, 10, 2265.
1
07, 3640.
14) (a) Cox, B. G.; McTigue, P. T. Aust. J. Chem. 1964, 17, 1210. (b)
(
Kurz, J. L. J. Am. Chem. Soc. 1967, 89, 3524. (c) Bell, R. P.; Onwood, D.
P. Trans. Faraday Soc. 1962, 58, 1557. (d) Hine, J.; Houston, J. G.; Jensen,
J. H. J. Org. Chem. 1965, 30, 1184. (e) Gruen, L. C.; Mc Tigue, P. T. J.
Chem. Soc. 1963, 5217. (f) Greenzaid, P.; Luz, Z.; Samuel, D. J. Am. Chem.
Soc. 1967, 89, 749. (g) Hine, J.; Redding, R. W. J. Org. Chem. 1970, 35,
2
769.

2296 J. Am. Chem. Soc., Vol. 120, No. 10, 1998
Hegarty et al.
PMX spectrometerstetramethylsilane was the internal reference
used. Deuterated chloroform, deuterated methanol, deuterated
DMSO, and deuterated water were the solvents used. The IR
spectra were recorded on a Perkin-Elmer 1710 Fourier Trans-
form spectrometer or on a Mattson Instruments Galaxy Series
FTIR 3000 spectrometer. Spectroscopic measurements in the
ultraviolet and visible regions were obtained using a Cary 210
spectrometer, equipped with moveable thermostated cell com-
partment, the temperature of which was controlled by using a
Techne TE8D Temp-ette water bath and pump and a Techne
RB12 Refrigerated Bath. pH measurements were made with a
Radiometer pH meter 26 and a Metrohm combined pH glass
electrode, which was standardized using E. I. L. buffers to (0.02
pH units.
standard techniques. Inorganic materials used in kinetic mea-
surements were AnalaR grade, where possible. Potassium
nitrate was recrystallized from water and its crystals finely
ground and dried for 12 h under vacuum at 100 °C. The
composition of the potassium hydroxide pellets was determined
by titration with a standard solution of hydrochloric acid, which
had been prepared by dilution of Rh oˆ ne Poulenc Volucon
ampules. Water used for kinetic measurements was initially
purified by doubly distilling it by using an Exelo Water Still
and then degassed with a Millipore Filtration Apparatus.
Organic solvents used during kinetic work were HPLC grade
where possible.
Reagents and solvents were purified, where stated, using
JA9729544