N-(Hydroxybenzyl)benzamide Derivatives: Aqueous pH-Dependent Kinetics and Mechanistic Implications for the Aqueous Reactivity of Carbinolamides

Article abstract of DOI:10.1021/acs.joc.9b02812

The rate constants for the aqueous reaction, between pH 0 and 14, have been determined for a series of amide substituted N-(hydroxybenzyl)benzamide derivatives, in H₂O, at 25 °C, I = 1.0 M (KCl). The N-(hydroxybenzyl)benzamide derivatives were found to react via three distinct mechanisms with the kinetically dominant mechanism being dependent on the pH of the reaction solution. It has been shown that the carbinolamides react via a specific-base-catalyzed mechanism (E1cB-like) under basic and pH neutral conditions. At lower pH values, an acid-catalyzed mechanism was kinetically dominant and, last, a water reaction was postulated at pH values where neither the hydroxide-dependent nor the general-acid-catalyzed mechanism was dominant. Contrary to earlier studies with N-(hydroxymethyl)benzamide compounds, no evidence for mechanistic variation based upon the nature of the amidic substituent was observed for any of the N-(hydroxybenzyl)benzamide derivatives studied between pH values of 0-14. The rate for the acid-catalyzed reaction (k_H, ρ = -1.17), the apparent second-order hydroxide rate constant (k1′, ρ = 0.87), the hydroxide-independent rate (k₁, ρ = 0.65), and the pK_a's of the hydroxyl group of the carbinolamide (ρ = 0.23) are reported.

Full text of DOI:10.1021/acs.joc.9b02812

Article
pubs.acs.org/joc
Cite This: J. Org. Chem. XXXX, XXX, XXX−XXX
N‑(Hydroxybenzyl)benzamide Derivatives: Aqueous pH-Dependent
Kinetics and Mechanistic Implications for the Aqueous Reactivity of
Carbinolamides
Takaoki Koyanagi, Paul M. Siena, David E. Przybyla, Mohammad I. Rafie, and Richard W. Nagorski*
Department of Chemistry, Illinois State University, Normal, Illinois 61790-4160, United States
S
* Supporting Information
ABSTRACT: The rate constants for the aqueous reaction,
between pH 0 and 14, have been determined for a series of
amide substituted N-(hydroxybenzyl)benzamide derivatives,
in H2O, at 25 °C, I = 1.0 M (KCl). The N-(hydroxybenzyl)-
benzamide derivatives were found to react via three distinct
mechanisms with the kinetically dominant mechanism being
dependent on the pH of the reaction solution. It has been
shown that the carbinolamides react via a specific-base-
catalyzed mechanism (E1cB-like) under basic and pH neutral
conditions. At lower pH values, an acid-catalyzed mechanism
was kinetically dominant and, last, a water reaction was
postulated at pH values where neither the hydroxide-
dependent nor the general-acid-catalyzed mechanism was
dominant. Contrary to earlier studies with N-(hydroxymethyl)benzamide compounds, no evidence for mechanistic variation
based upon the nature of the amidic substituent was observed for any of the N-(hydroxybenzyl)benzamide derivatives studied
between pH values of 0−14. The rate for the acid-catalyzed reaction (kH, ρ = −1.17), the apparent second-order hydroxide rate
constant (k′1, ρ = 0.87), the hydroxide-independent rate (k1, ρ = 0.65), and the pKa’s of the hydroxyl group of the carbinolamide
(ρ = 0.23) are reported.
interesting observations having been reported.7,32−35 A valid
question that has resulted from these studies is how
characteristic of “normal” carbinolamide reactivity patterns
INTRODUCTION
■
At first appearance, carbinolamides are structurally similar to a
variety of carbonyl derivatives that possess a broad range of
reactivity.1 Coupled to the problem of a broad potential
are the N-(hydroxymethyl)benzamide derivatives given the
well-known electrophilicity of formaldehyde36−38 vs other
reactivity range within this family of functionalities is a lack of
aromatic aldehydes?39−43 To broaden the understanding of the
information concerning reactions having an amide group
acting as a leaving group.2−7 Additionally, there are aspects of
effects of structure on the aqueous reactivity of carbinolamides,
the aqueous kinetics for a series of N-(hydroxybenzyl)-
benzamide derivatives (2) were investigated in H2O, as a
function of pH, at 25 °C, I = 1.0 M (KCl).
the mechanism of reaction of carbinolamides that are still not
well understood, and this is surprising given that the first report
of compounds possessing this functionality appeared in the
literature as early as the 1870s.8 One could rationalize this
statement as being the result of the rarity of the functionality
within our world. If one accepts this argument, it would be
surprising to learn of the diverse roles that carbinolamide
containing compounds have in a broad variety of venues. For
example, an increasing number of compounds possessing this
functionality have been discovered and are being developed
that have interesting pharmaceutical applications.9−23 Carbi-
nolamides have also been found to be intermediates in an array
of biological venues with both positive24−28 and negative29−31
outcomes for the associated organism. Our studies, to date,
have focused on the mechanisms and methods of catalyzing
the reaction of these compounds and the role that structure
plays in their reactivity patterns.7,32−35 Much of the work that
has been performed, in our group, on the N-(hydroxymethyl)-
benzamide derivatives (1) has been published, with several
The studies involving N-(hydroxymethyl)benzamide deriv-
atives (1) led to a number of somewhat unexpected outcomes,
one of which was the reexamination of the generally accepted
reaction pathways at different pH values.7,32−35 The previously
accepted mechanism for the acid-catalyzed reaction was
thought to involve a specific-acid protonation of the
carbinolamide, followed by rate-limiting decomposition into
Received: November 7, 2019
Published: December 12, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
the tautomer of the amide and protonated aldehyde (Scheme
1).32,44,45 This proposed mechanism was invalidated by the
Scheme 4. Specific-Base-Catalyzed Reaction
Scheme 1. Previously Accepted Specific-Acid-Catalyzed
Breakdown of Carbinolamides
upon the trends observed for those compounds having
electron-withdrawing substituents or having no substituent
(apparent break in the Hammett plot).7,32,35 Product analysis
of the reaction products of the derivatives of 1, bearing
strongly electron-donating groups, showed the presence of
large amounts of carboxylate product present.35 That is, there
was more hydrolyzed amide present than could be explained
by carbinolamide breakdown followed by amide hydrolysis,
based upon the conditions of the reaction (H2O, 25 °C, I = 1.0
M (KCl)).35 These observations led to the conclusion that the
decrease in the nucleofugality of the benzamidate, upon the
addition of the electron-donating groups, increased the
stability of the conjugate base of the carbinolamide (3),
where R is an H (see Scheme 4), to the degree that alternative
reaction pathways became energetically competitive.35 Would
this observation hold true for the carbinolamides not generated
from formaldehyde?
observation of buffer catalysis in the acid-catalyzed region of
the pH−rate profile for the derivatives of 1.34,35 Buffer catalysis
provided evidence for an acid-dependent reaction which
involved rate-determining proton transfer.33,34 This provided
a new complication, as two kinetically equivalent mechanisms
could be proposed for the acid-catalyzed reaction that involve
rate-determining proton transfer. These mechanisms are
general-acid catalysis (Scheme 2) and a specific-acid followed
by a general-base mechanism (Scheme 3).33,34
Scheme 2. General-Acid-Catalyzed Mechanism
These observations, which form the foundation of what is
currently known about the reactivity of carbinolamides, were
determined as a result of studies of the N-(hydroxymethyl)-
benzamide derivatives (1). It should be noted that the
derivatives of 1 are a unique class of carbinolamides, as they
are the product of the reaction of a benzamide derivative with
formaldehyde. The electrophilicity of formaldehyde results in
the conjugated base of the carbinolamide (3) not having strong
electronic assistance to the departure of the benzamidate
leaving group in the breakdown of the carbinolamide.36−38 The
results for the N-(hydroxybenzyl)benzamide derivatives
represent a series of compounds where the nucleofugality of
the benzamidate leaving groups was the same as the studies
with 1, but the electronic assistance to the departure of the
benzamidate, by the less electrophilic aromatic aldehydes,
should be greater.39,40 What effect will this have on the
reactivity of the derivatives of 2 vs 1, and will the mechanisms
of reaction remain the same? Reported here are the kinetics of
the aqueous reaction of N-(hydroxybenzyl)benzamide and the
amide substituted derivatives as a function of pH at 25 °C, I =
1.0 M (KCl), and the implications of these results on the
accepted reaction mechanisms of carbinolamides under
aqueous conditions.
Scheme 3. Specific Acid Followed by General Base
A second outcome, stemming from the work focused on the
reactivity of derivatives of 1, occurred in the high [HO−]
region of the pH−rate profile for those compounds bearing
electron-donating groups. The generally accepted mechanism
for the breakdown of carbinolamides under basic conditions
involves a specific-base-catalyzed deprotonation of the
hydroxyl group of the carbinolamide, followed by rate-limiting
breakdown of the conjugate base into the aldehyde and an
amidate (Scheme 4).7,32,35,44−48 Experimental evidence for the
mechanism, shown in Scheme 4, involved a lack of buffer
catalysis and the existence of a pH-independent region at high
[HO−] where the carbinolamide would be fully deprotonated.
However, it was found that derivatives of 1, bearing strongly
electron-donating groups on the amidic portion of the
carbinolamide structure, reacted faster than anticipated based
RESULTS
■
The carbinolamides used in the studies described here were
synthesized from the amide and aldehyde starting materials
(see Scheme 5). The morpholine aminal of benzaldehyde (4)
was reacted with the benzamide derivatives, using a solvent-
free method, to generate the monoacylaminal synthetic
intermediate (5).49 The monoacylaminals were then subjected
to solvolysis conditions, maintained at “pH” ∼4.2 in an
B
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
the pH−rate profile, and the equation that fits this data is also
shown: the acid-dependent reaction, the hydroxide-dependent
reaction, and, at higher pH, a hydroxide-independent reaction.
Previous studies also reported a region, between the acid- and
hydroxide-dependent reactions, where there was no change in
rate as the pH of the solution was modified.32,44,45,53 The
reaction, in this region of the pH−rate profile, was said to be
reacting via a water-catalyzed reaction. The derivatives of 2
discussed here show only a very small region that might be
called the water reaction as compared to previous studies
involving derivatives of 1.32
Scheme 5. Synthesis of Carbinolamide Compounds
Figures 1 and 2 show the effects of changes in the
hydronium ion concentration on the observed rate of
acetone/water solution (2:1, respectively), which was a
modification of a method used by Bundgaard et al. and led
to the synthesis and eventual isolation of the target
carbinolamides.44,49
All kinetic experiments were performed in H2O at 25 °C and
I = 1.0 M (KCl) and were initiated by the injection of
concentrated solution of the carbinolamide, in DMSO, into the
reaction solution to yield a final [substrate] of ∼1 × 10−4 M.
Rate constants for experiments between pH values of 0−4 and
6−9 were obtained by following the change in absorbance as a
function of time, utilizing a UV−vis spectrophotometer.
Between pH values of 10 and 14, a stopped-flow apparatus
was used to perform the kinetic studies, as the reactions were
too fast to follow by other means. The stopped-flow
experiments were performed similarly to the studies described
above; however, the stock solution of substrate was injected
into H2O or 1 M KCl solution and these solutions were mixed
with the hydroxide solutions, in the stopped-flow, to obtain
kobsd at a specific [HO−]. All of the carbinolamides studied
exhibited good first-order kinetics with stable infinity
absorbancies. Reactions quenched, prior to the completion of
the reaction, showed only starting material, amide, and
benzaldehyde by HPLC analysis. At all pH values, the reaction
of the carbinolamides was much more rapid than subsequent
hydrolysis of the benzamide based upon available data.50,51
For experiments performed in the pH range from 4 to 6,
HPLC was used to monitor the reaction progress as a function
of time. The chromatogram showed separate peaks for the
carbinolamide, amide, and benzaldehyde so that the dis-
appearance and appearance of each component could be
followed independently. The rate of reaction was obtained
using eq 1, where Acar was the area of the peak for the
carbinolamide at some time t and Aald was the area of the
benzaldehyde peak at the same time. The observed area of the
benzaldehyde peak was corrected by Rx to account for
differences in the molar extinction coefficients of benzaldehyde
and the carbinolamides.52 (The Rx values used are reported in
Table S1 of the Supporting Information.)
Figure 1. Plot of the effect of [hydronium ion] (M) on the observed
rates of reaction (kobsd, s−1) for N-(hydroxyphenylmethyl)-4-nitro-
◆
▲
benzamide ( ), N-(hydroxyphenylmethyl)-4-chlorobenzamide ( ),
■
N-(hydroxyphenylmethyl)benzamide ( ), N-(hydroxyphenylmeth-
●
yl)-4-methylbenzamide ( ), and N-(hydroxyphenylmethyl)-4-me-
▼
thoxybenzamide ( ) in H2O, at 25 °C, I = 1.0 M (KCl).
Figure 2. Plot of the effect of acetate buffer (M) of the observed rates
of reaction (kobsd, s−1) at pH 4.0 for N-(hydroxyphenylmethyl)-4-
▲
chlorobenzamide ( ), N-(hydroxyphenylmethyl)-4-methylbenza-
●
▼
mide ( ), and N-(hydroxyphenylmethyl)-4-methoxybenzamide (
)
in H2O, at 25 °C, I = 1.0 M (KCl).
carbinolamide breakdown and the effect of buffer concen-
tration on kobsd, respectively, in H2O at 25 °C and I = 1.0 M
(KCl). The second-order rate constants for the acid-dependent
reaction (kH) from the plots in Figure 1 are reported in Table
1. For studies between pH values of 2 and 6, the acid-
dependent reaction was buffer-catalyzed with kobsd, at a specific
pH value, being determined by extrapolation to zero [buffer]
(see Figure 2). Extensive investigations of the second-order
i
y
Acar
j
z
j
z
lnj
j
z = kobsdt
z
z
j
(Acar + (Rx)(Aamide or ald))
(1)
k
{
Shown in the graphical abstract is the pH−rate profile for 2d
(see the pH−rate profiles for the other compounds in the
Supporting Information). Three distinct regions are present in
C
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Constants for the Aqueous Reaction of N-(Hydroxybenzyl)benzamide and Derivatives and Their pKa’s in Water at 25
°C, I = 1.0 (KCl)
a
bc
,
b
d
e
compound
kH (10−2 M−1 s−1
)
k1′ (M−1 s−1
)
k1 (s−1
)
Ka (10−13 M)
pKa
krel
2a
2b
2c
2d
2e
2f
0.55 0.02
−
2.0 0.1
3.2 0.1
5.6 0.2
12.0 0.1
1.35 × 103
7.4 × 102
4.0 × 102
2.5 × 102
1.9 × 102
1.6 × 102
71.4 0.8
36.8 0.6
28.9 0.6
18.2 0.4
15.6 0.4
13.7 0.4
1.93 0.05
1.97 0.05
1.31 0.05
1.39 0.05
1.14 0.05
1.19 0.05
12.71
12.71
12.88
12.86
12.94
12.92
3.92
2.02
1.59
1.0
0.86
0.75
a
Apparent second-order rate constant for the hydroxide-dependent breakdown of the carbinolamides in water calculated from k1 and Ka using k1′ =
b
c
(k1Ka)/Kw and Kw = 1 × 10−14 M2. Errors obtained from nonlinear least-squares fits of kobsd vs [HO−] according to eq 2. First-order rate constant
d
for the pH-independent breakdown of the deprotonated carbinolamide. Ionization constant for the hydroxyl group of the carbinolamide.
Calculated by dividing k1 for 2d into the k1 value for each amide.
e
rates of buffer catalysis were not pursued, and at most pH
values, the buffer concentrations were kept quite low (0.05−
0.1 M) and this, ultimately, limited the utility of the second-
order buffer catalysis constants obtained under these
experimental conditions.
this functionality, were focused on the N-(hydroxymethyl)-
benzamide derivatives (1), as these compounds represented
the most straightforward access to the carbinolamide
functionality.7,32−35 The synthesis of carbinolamides using
less electrophilic aldehydes remained problematic, although
biological systems seem to be quite able to make such
compounds.9−17
Shown in Figure 3 is the effect of [HO−] on kobsd for
carbinolamide breakdown into amide and benzaldehyde in
It was shown, by Bundgaard and co-workers,44 that N-
(hydroxybenzyl)benzamide could be synthesized by an
adaptation of a procedure from Macovski and Backmeyer
wherein the morpholine monoacylaminal was synthesized from
a 1:1:1 mixture of benzamide, benzaldehyde, and morpho-
line.56 The monoacylaminal formed was subsequently hydro-
lyzed to yield 2d. Unfortunately, for the purposes of the work
described here, the monoacylaminal formation reaction was
most successful with benzamide and other benzamide
derivatives yielded poor to nonexistent amounts of mono-
acylaminals for reasons that we did not explore. The
development of a solvent-free method to synthesize the
monoacylaminals from the morpholine aminal of benzaldehyde
and the amide, followed by hydrolysis, provided a route to the
compounds discussed herein (see Scheme 5).49,57
Initial kinetic studies, by Bundgaard and co-workers, were
focused on the aqueous reaction of carbinolamides having a
broad variety of structures, with their primary interest directed
toward the utility of carbinolamides as pro-drugs.44−48 As such,
the Bundgaard studies focused on pH ranges and at a
temperature that were relevant for biological systems.44−48
Bundgaard et al. did perform studies with N-(hydroxybenzyl)-
benzamide (2d) in the pH range from 2 to 10.44 The second-
order rate constant for the acid-catalyzed reaction (kH = 0.13
M−1 s−1) and the apparent second-order rate constant for the
hydroxide-dependent reaction (k1′ = 3.0 × 102 M−1 s−1 in H2O,
I = 0.5 M and 37 °C) were reported.44 No substituted
derivatives of 2d were investigated, and kinetic data points
leading to the kH for 2d were limited.44
Acid-Catalyzed Studies. Shown in Figure 1 are the plots
of kobsd vs [HCl] for five of the derivatives of 2. All five
compounds show a linear dependence on the acid concen-
tration, and the second-order rate constants for the acid-
catalyzed reaction are reported in Table 1. From Figure 1 and
Table 1, the methoxy derivative (2f) reacts most quickly and
the compound bearing the nitro group (2a) reacts more slowly
than the others studied. This substituent effect suggests that
the rate-limiting step of the reaction involves a species with a
positive charge or a developing positive charge that would be
destabilized by electron-withdrawing groups. The effect of the
substituents is better illustrated in Figure 4, which shows the
Figure 3. Plot of the effect of [hydroxide] (M) of the observed rates
of reaction (kobsd, s−1) for N-(hydroxyphenylmethyl)-4-nitrobenza-
◆
mide ( ), N-(hydroxyphenylmethyl)-4-(trifluoromethyl)benzamide
▲
(⬢), N-(hydroxyphenylmethyl)-4-chlorobenzamide ( ), N-(hydrox-
■
yphenylmethyl)-benzamide ( ), and N-(hydroxyphenylmethyl)-4-
▼
methoxybenzamide ( ) in H2O, at 25 °C, I = 1.0 M (KCl).
H2O, at 25 °C and I = 1.0 M (KCl). No evidence for buffer
catalysis was observed between pH values of 5 and 12, but in
most cases, the buffer concentrations were kept low (0.01−
0.05 M total buffer). From Figure 3, low [HO−] yielded a first-
order dependence between [HO−] and kobsd, while at higher
[HO−], the rate of the reaction was independent of the
[HO−]. The dependence of kobsd on [HO−] can be fit by eq 2
(see Scheme 4), to obtain the limiting rate of reaction (k1) and
the Ka of the hydroxyl group (see Table 1).
Ka[HO−]
HO
(kobsd
)
= k
1 Kw + Ka[HO−]
(2)
DISCUSSION
■
While carbinolamides have been known for quite some
time,8,54 the structural diversity of the known compounds is
not as broad as might be expected, as the majority of the
isolated synthetic compounds result from the reaction of
amides with highly electrophilic aldehydes, such as form-
aldehyde and chloral.55 The first studies, from our group, with
D
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
eq 3, where (kobsd)H is the observed rate for the acid-dependent
reaction.
H
(kobsd
)
= kH[HA]
(3)
The second proposed mechanism for the acid-catalyzed
reaction is shown in Scheme 3. The mechanism involves an
initial protonation of the carbonyl oxygen (specific-acid
catalysis).32 The protonated species then undergoes rate-
determining general-base-catalyzed decomposition with a base
removing the hydroxyl proton during breakdown. The rate
expression for the mechanism in Scheme 3 is shown in eq 4
and has the same dependence on [HA], as seen in eq 3.
(Equations 3 and 4, where KaHA is the Ka of the buffer catalyst
and kd[A−] can be dropped from the denominator of the rate
expression, as it would be slow compared to k−p.) The
mechanism shown in Scheme 3 resembles the accepted
specific-acid-catalyzed mechanism for the breakdown of
acetals.58,59 It is acknowledged that, given the available data,
no conclusion concerning which mechanism dominates can be
made based upon the results presented here, but we have
argued in a previous publication that it is the mechanism
shown in Scheme 3 that is the acid-dependent reaction.33
From Table 1, the kH values are similar to that reported by
Bundgaard et al. for 2d with Bundgaard’s value being ∼4-fold
faster. The difference between the values in Table 1 for kH of
2d and that reported by Bundgaard can be rationalized due to
differences in the experiment temperatures (37 °C vs 25 °C)
and ionic strength (0.5 vs 1.0) at which the studies were
performed vs the studies reported herein.
The similarity of the rate expressions for the reactions shown
in Schemes 2 and 3 does not allow for a definitive conclusion
with respect to the mechanism of the acid-catalyzed reaction.
What is clear is that all of the derivatives of 2 react via the same
mechanism (see Figure 4), and at a much faster rate than that
observed for the derivatives of 1. Very little work on the acid-
catalyzed reaction of carbinolamides has been published;
however, the rate constants for the acid-catalyzed reaction of 1
and the 4-chloro substituted derivative of 1 have been reported
to be kH = 7.3 × 10−6 M−1 s−1 and kH = 5.7 × 10−6 M−1 s−1,
respectively.32,33 These rate constants, for 1 and its derivative,
are ∼4000-fold slower than that for the similarly substituted
derivatives of 2 (see Table 1). The relative reactivity of 1 vs 2
Figure 4. Hammett plot of log(kHX/kHH) for the acid-catalyzed
reaction of a series of N-(hydroxybenzyl)benzamide derivatives in
H2O, at 25 °C, I = 1.0 M (KCl) vs σ.
Hammett plot of the kH values. The linear correlation, shown
in Figure 4, supports the conclusion that all five of the
compounds react via a similar mechanism. The plot, shown in
Figure 4, has a ρ value of −1.17.
The negative ρ value for the acid-dependent rates is an
indication that there is buildup of positive charge on the
amidic portion of the carbinolamide in an intermediate or
during the rate-determining step of the reaction. (It is unlikely
that the positive charge is on the hydroxyl portion of the
carbinolamide, as it has been shown and will be shown later,
during the discussion of the hydroxide-dependent reaction,
that the substituents on the amide portion of the structure have
only a minor effect on the pKa of the hydroxyl group of the
carbinolamide.)7,32 The effect of the substituents on the
reactivity of the derivatives of 2 is similar to that observed for
the derivatives of 1.32 The observed substituent effects support
a mechanism of the type shown in Scheme 1; however, the
observation of buffer catalysis in the acidic portion of the pH−
rate profile indicates that proton transfer must be occurring
during the rate-limiting step of the acid-catalyzed reaction (see
Figure 2).
Shown in Figure 2 is the effect of [buffer] on the rate of the
acid-catalyzed reaction with acetate buffer at a single pH value.
The same observation was made with both acetate and
chloroacetic acid which were the only two buffers used in the
acid region of these studies. The kobsd value for the buffer-
independent rate at each pH value was obtained from the y-
intercept values of the plots shown in Figure 2. For most of the
studies performed, total [buffer] ([Btot]) was kept quite low
(<0.05 M = [Btot]), resulting in limited data concerning the
second-order rate for buffer catalysis for the compounds
discussed here. The important conclusion connected to the
observation of buffer catalysis was the reaction must be
occurring via a mechanism involving proton transfer during the
rate-determining step.
mirrors the relative electrophilicity of formaldehyde (KHyd
=
2420)36−38 vs benzaldehyde (KHyd = ∼0.01),40,43 as measured
by the hydration constant.
kpkd[H+][A ]
kpkdKaHA[HA]
−
H
(kobsd
)
=
=
−
k−p + kd[A ]
k−p
(4)
Hydroxide-Dependent Reaction. From the graphical
abstract and Figure 3, at lower [hydroxide], there is a linear
dependence between kobsd and the [HO−], but at higher
[HO−], kobsd becomes independent of hydroxide. Such
observations are consistent with a specific-base-catalyzed
reaction where, at lower [HO−], the first-order dependence
on hydroxide is a result of the shift in equilibrium between 2
and 3 (in Scheme 4). As the Ka value of the hydroxyl group of
2 is approached, with increasing pH, the kobsd dependence on
[HO−] transitions from first-order to zero-order as the
equilibrium is fully shifted to 3 and further increases in
hydroxide have no effect on the concentration of the reactive
species in solution. This observation supports the conclusion
From above, two mechanisms have been proposed based
upon the observation of buffer catalysis in the aqueous acid-
dependent reaction of carbinolamides (see Schemes 2 and
3).33 Shown in Scheme 2 is the general-acid catalysis reaction
where the rate-determining step involves proton transfer from
an acid source as the carbinolamide decomposes into the
amide and protonated aldehyde. The atom being protonated
during the decomposition step is arbitrary, as it could be either
the nitrogen or the oxygen of the amidic species. The rate
expression for the reaction shown in Scheme 2 is as shown in
E
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
that the rate-determining step is the decomposition of 3 into
benzaldehyde and the benzamidate leaving group.
apparent second-order rates for the derivatives of 1 vs 2, the
effect of the substituents on the leaving group ability of
amidate remains the same.7,35
Further evidence for a mechanism of the type shown in
Scheme 3 was provided by the lack of buffer catalysis for the
kinetic studies for pH values between 5 and 12. Each of the
reactions for pH values between 5 and 12 were performed at,
at least, three [buffer], at the same pH, and no evidence for
buffer catalysis was observed for any of the buffers that were
utilized. This observation indicates that proton transfer does
not occur during the rate-determining step of carbinolamide
breakdown under conditions when the hydroxide-dependent
mechanism dominates. The rate expression for the hydroxide-
dependent reaction is shown in eq 2, where Kw is the
dissociation constant of water. The observed rate data for the
hydroxide-dependent reaction was fit to eq 2, yielding the
limiting rate constant (k1) and the Ka of the hydroxyl group of
the carbinolamide (see Table 1).60
From Figure 3 and Table 1, the effect of the substituents on
the hydroxide-dependent reaction are the opposite of that
observed for the acid-catalyzed reaction. The derivatives of 2
bearing electron-withdrawing substituents react more quickly
than those having electron-donating substituents. Such an
observation would be in accord with the type of mechanism
shown in Scheme 4, as electron-withdrawing groups would
more effectively stabilize the developing negative charge on the
benzamidate leaving group during the rate-determining step.
(Benzamide pKa data is limited, but available data suggests a
pKa range of ∼16−21 for the benzamide derivatives studied
herein.36)
The linearity of the Hammett plot of k′1, for the derivatives
of 2, was not unexpected, as all of the derivatives of 1 also
reacted by a single mechanism at lower [HO−]. It was only at
higher [HO−] that the derivatives of 1 bearing electron-
donating groups began to exhibit mechanistic variation.35
Substituent Effect on Ka. The Ka values for the derivatives
of 2 were determined by the nonlinear least-squares fit of eq 2
to the dependence of kobsd upon the [hydroxide]; see Figure 3.
The Ka values determined in this manner are listed in Table 1,
and these values show a greater dependence upon the nature of
the substituent attached to the ring than was observed for the
derivatives of 1, which had a ρ value of 0.07, albeit over a
smaller substituent range.7 However, like the derivatives of 1,
no significant changes in pKa of the derivatives of 2 were
observed. The pKa of the hydroxyl group of 2d (pKa = 12.9)
was essentially the same as the pKa reported for the hydrate of
benzaldehyde (pKa = 12.8).40 A similar observation between
the pKa of the hydroxyl group of 1 and the pKa of the hydrate
of formaldehyde was noted previously.7 These two observa-
tions would suggest that the inductive effect of the benzamide
group in the carbinolamide is similar to that of a hydroxyl
group in hydrates of aldehydes.
The Hammett correlation of the Ka values of the derivatives
of 2 was linear (see the Supporting Information) and had a ρ
value of 0.23. The plot was linear over all of the compounds
that were studied, which contrasted with studies of the
derivatives of 1 which underwent a substituent-dependent
mechanism change for the hydroxide-dependent reaction
which yielded odd variations in the Ka when kobsd was
incorrectly used to fit the [HO−] using eq 2.7,35 For the
derivatives of 2, the substituent-dependent change on the pKa
of the hydroxyl group of the carbinolamide was larger than that
found for 1 (ρ = 0.07). The change in Ka seen for the
derivatives of 2 was more in line with the expected effect of the
addition of methylene units between the carbon bearing the
ionizable group and the aromatic ring.37,61
While the changes in the Ka of the hydroxyl group are more
substantive for derivatives of 2 as compared to those observed
for 1, there is no clear rationale for this observation, as the only
difference is the presence of an aromatic ring in the aldehyde
portion of the carbinolamide structure. It was noted in the
previous studies of 1 that substituents in the ortho position of
the aromatic ring also did not have any apparent effect on the
Ka of the hydroxyl group, which was in contrast to the effect of
ortho substituents on the Ka of benzoic acids which tend to be
quite significant.7 This result further illustrated the oddity of
the invariance of the Ka values for the derivatives of 1.7 The Ka
values shown in Table 1 for the derivatives of 2 show an effect
of the magnitude expected based upon other studies studying
the effect of remote substituents on Ka in other molecular
structures.37,61
Limiting Rate (k1). It has been proposed that the rate-
limiting step for the breakdown of carbinolamides under the
hydroxide-dependent mechanism is the departure of an
amidate leaving group and the formation of the aldehyde
product (see Scheme 4). This proposal was based upon a lack
of buffer catalysis in the region of the pH−rate profile where
the hydroxide-dependent reaction dominated and a change
from a first-order to zero-order dependence on the [hydroxide]
in this same region. The reaction can be thought of as an E1cb-
k1Ka[HO−]
HO
(kobsd
)
=
Kw
(5)
Previous studies of the reaction of 2d, performed by
Bundgaard and co-workers, under aqueous conditions were
completed at 37 °C and I = 0.5 M.44 Due to the limited range
of pH over which this study was performed, only the linear
portion of the effect of hydroxide on the rate of reaction was
investigated. As a result, the authors only reported an apparent
second-order rate constant (k′1, see eq 5 where k1Ka/Kw = k1′)
for the effect of hydroxide on the rate of reaction.44 Reported
in Table 1 are the apparent second-order rate for the hydroxide
reaction for compounds 2a−f. From Table 1, 2d has an
apparent second-order rate constant of k1′ = 250 M−1 s−1 at 25
°C, I = 1.0 M (KCl), as compared to k′1 = 300 M−1 s−1
reported by Bundgaard et al. at 37 °C and I = 0.5 M.44 Given
the difference in the experimental conditions, the two results
are similar to one another, showing that the temperature and
ionic strength have a smaller effect on the hydroxide-
dependent reaction than that observed for the acid-catalyzed
reaction.
From eq 5, it is apparent that k′1 incorporates the effects of
the substituents on both the Ka of the hydroxyl group and the
limiting rate of the breakdown of 3. There is a rate
enhancement of ∼10-fold on moving from the 4-methoxy
(2f) to the 4-nitro compound (2a). The Hammett correlation
of the apparent second-order rate constants for the derivatives
of 2 is linear (see the Supporting Information) with a ρ value
of 0.87. This value is the same as that found for the Hammett
plot of the apparent second-order rate constants for the
derivative N-(hydroxymethyl)benzamide (1) (ρ = 0.87)
bearing electron-withdrawing groups.7,35 While there is an
approximate 3 orders of magnitude difference between the
F
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
like reaction and represents a unique form of reactivity wherein
an amide (amidate) acts as a leaving group. The properties of
amides acting as leaving groups is not something that has
received a great deal of attention.7,26,29
The k1 values or limiting rates, shown in Table 1, follow a
similar trend to that observed for the apparent second order
rates (k1′). That is, the amidic functionality bearing electron-
withdrawing groups undergoes elimination more quickly than
those bearing electron donating groups. This result is in accord
with the previous studies involving derivatives of 1 and was
attributed to electron-withdrawing groups providing better
stabilization of the developing negative charge of the amidate
as it departs.7,32,35 Figure 5 shows the Hammett plot of limiting
thus, benzaldehyde provides better electron assistance to the
departure of the benzamidate leaving group. Implicit in this
statement is the idea that the energetics of the amidic
hydrolysis pathway for carbinolamide breakdown would not
change too dramatically between 1 and 2.
In addition, the similarity of the two ρ values found for the
derivatives of 1 (ρ = 0.67) and 2 (ρ = 0.65) indicates that the
substituents have the same relative effect on the rate-limiting
breakdown of the conjugate base of the carbinolamide. This
result provides further evidence that the substituent effect is
solely on the departure of the leaving group and that the extent
of charge stabilization on the benzamidate leaving group, in the
transition state, must be similar for 1 and 2. This result may
seem to be curious given the difference in the relative limiting
rates of reaction for the derivatives of 1 vs 2. When the limiting
rates for 2a, 2b, and 2d are compared to their related N-
(hydroxymethyl)benzamide derivative, the derivatives of 2
react, on average, 430 times faster than their related derivatives
of 1. The effect of the aromatic ring on the carbinol portion of
2 must provide significant stabilization/assistance during the
departure of the benzamidate, leading to a rate enhancement vs
derivatives of 1.
CONCLUSIONS
■
Presented here are the results of studies investigating the
mechanism of the aqueous reaction of amide substituted
derivatives of N-(hydroxybenzyl)benzamide as a function of
the pH of the solution. The N-(hydroxybenzyl)benzamide
derivatives (2) show much greater reactivity than their N-
(hydroxymethyl)benzamide derivative (1) counterparts. The
degree of the reactivity difference was dependent on the pH-
dependent mechanism, but the derivatives of 2 were 400−4000
times more reactive than the derivatives of 1. These studies
were undertaken to broaden the understanding of the aqueous
reactivity of carbinolamides as a function of the structure of the
carbinolamide itself. Very little information is available in the
literature concerning the reactivity of carbinolamides derived
from less electrophilic aldehyde sources,38−42 and the data
presented herein represents the first complete pH−rate profile
study of compounds of this type. The studies were performed
between pH values of 0 and 14 with buffers used to maintain
pH between 2 and 12. The pH−rate profile for 2 can be seen
in the graphical abstract, and four distinct regions are evident.
There was an acid-catalyzed reaction, hydroxide-dependent
reaction, hydroxide-independent reaction, and water reaction
(see the equation in the graphical abstract). As with the
derivatives of 1, the hydroxide-dependent reaction dominated
the largest portion of the pH−rate profile.
The acid-catalyzed reaction was observed between pH
values of 0 and ∼4 with electron-donating groups increasing
the rate relative to electron-withdrawing groups (ρ value =
−1.17). The reaction itself was buffer-catalyzed, and two
kinetically equivalent mechanisms can be proposed. Both a
general-acid-catalyzed reaction (Scheme 2) and a specific-acid
followed by general-base-catalyzed reaction (Scheme 3) can be
postulated. The results presented here cannot distinguish
which mechanism was energetically more favorable; however,
all of the compounds studied react via a single mechanism (see
Figure 4) and the derivatives of 2 mirror the observations
reported for 1, while being much more reactive than similar
derivatives of 1. For the acid-catalyzed reaction, the data
supports a common mechanism for both the derivatives of 1
and 2.
X
Figure 5. Hammett plot of log(k1 /k1H) for the hydroxide-dependent
reaction of a series of N-(hydroxybenzyl)benzamide derivatives in
H2O, at 25 °C, I = 1.0 M (KCl) vs σ.
rate constants. The plot is linear, indicating a single mechanism
over the range of compounds reported here, and the ρ value is
0.65. The ρ value found for the derivatives of 1 had a ρ value of
0.67.7
The linearity of the plot in Figure 5 provides evidence that
the mechanism is the same over the six compounds studied
which is in contrast to studies involving derivatives of 1.7,26,29
It has been shown that all derivatives of 1 react via a single
mechanism at moderate [HO−] (Hammett plot of the
apparent second-order rate constants (k′1, see above)), but at
higher concentrations of [HO−], those derivatives of 1 bearing
electron-donating groups transitioned to a different mecha-
nism.7,26,29 Product analysis of the derivatives of 1 bearing
electron-donating groups showed large amounts of amide
hydrolyzed product both at 50% reaction and after reaction
completion, relative to those derivatives of 1 bearing electron-
withdrawing groups which only showed the presence of amide
after reaction completion.35 A second observation, for
derivatives of 1 bearing electron-donating groups, was that
the limiting rate (k1) became invariant with addition of strong
electron-donating groups.35 These observations were rational-
ized as the hydrolytic mechanism becoming competitive with
the normal carbinolamide reaction due to the poor
nucleofugality of the amidate bearing electron-donating
groups.35 The linearity of the plot shown in Figure 5 supports
the conclusion that all of the derivatives of 2 studied react by
the same mechanism and the hydrolysis route, observed for
those derivatives of 1 bearing electron-donating groups, does
not become energetically competitive as compared to the
accepted specific-base-catalyzed reaction for the derivatives of
2 studied. Presumably, this observation is the result of the
greater electrophilicity of formaldehyde vs benzaldehyde;36−43
G
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Benzamides. Benzamide, 4-(trifluoromethyl)benzamide, and 4-
chlorobenzamide were obtained from commercial sources and used
without further purification. In general, the other benzamide
derivatives were synthesized from the reaction of their benzoic acid
precursors with PCl5 to yield the acyl halide intermediate which was
subsequently reacted with concentrated ammonium hydroxide, to
yield the crude amide product.62 Crude amide products were then
recrystallized and melting points determined: 4-methoxybenzamide
(methanol, mp = 166−167 °C, lit. mp = 164−167 °C),63 4-
methylbenzamide (water, mp = 160−162 °C, lit. mp = 162−163
°C),64 4-nitrobenzamide (water, mp = 207−209 °C, lit. mp = 206
°C).65
For the hydroxide-dependent reaction, the data reported
supports a specific-base-catalyzed process for all of the
derivatives of 2 studied ((E1cb)R-like reaction). Once again,
the derivatives of 2 react much faster than their related
derivatives of 1 but show similar substituent effects to the
derivatives of 1 bearing electron-withdrawing groups. The
change in reactivity can, in part, be explained by a minor
difference in the pKa of the hydroxyl group of 2 vs 1. The
derivatives of 1 had pKa values of ∼13.1, whereas those
compounds discussed herein had pKa values of ∼12.8. This
difference in pKa would lead to a larger concentration of the
conjugate base of 2 vs 1 in solution at a specific pH and
resulting in a relative rate increase. The remainder of the
difference in rates between 1 vs 2 was due to differences in the
limiting rate (k1) between the two types of carbinolamides.
The limiting rates (k1) for derivatives of 2 were much larger
than similarly substituted derivatives of 1. Presumably, this
pattern of reactivity was as a result of the greater electro-
philicity of formaldehyde vs benzaldehyde, as the leaving
groups in both studies are the same. Further, the similarity of
the ρ values between 1 and 2 indicates that the transition
state/charge development must be similar between the two
sets of compounds, with the benzamidate leaving group having
similar charge development in the transition state for both 1
and 2. The most significant divergence between the derivatives
of 1 and 2 was that all derivatives of 2 react via the same
mechanism under all conditions where the hydroxide-depend-
ent reaction dominates.
It becomes more apparent with each group of carbinola-
mides studied that the acid-catalyzed reaction occurs with rate-
limiting proton transfer. The problem is which of the two
mechanisms (general-acid vs specific-acid followed by general-
base) is the actual mechanism of the acid-dependent reaction?
Additionally, was it the stability of 3 for the derivatives of 1
that led to the competitive amide hydrolysis reaction? At what
point does the electrophilicity of the aldehyde involved in the
formation of the carbinolamide lead to this amide hydrolysis
mechanism becoming energetically competitive? Lastly, by
which mechanism does the PAL enzyme catalyze the
breakdown of the carbinolamide? The acid-catalyzed reaction
involves rate-limiting proton transfer but also requires
relatively acidic conditions with only moderate returns in
reactivity. The hydroxide-dependent reaction yields much
larger catalysis potential (see graphical abstract), but how
would the enzyme affect the reactivity of the conjugate base of
the carbinolamide (3 in Scheme 4) to achieve catalysis?
Aminal Synthesis. Dimorpholinophenylmethane (4) was synthe-
sized utilizing a standard method66 with the resulting product
requiring no further purification and with mp = 100−102 °C (lit. mp
= 105−106 °C).57
Monoacylaminal Synthesis. The monoacylaminal derivatives
were synthesized using a solvent-free method described in the
literature.49 The general method involved mixing equimolar amounts
of the benzamide derivative (∼0.01 mol) and 4 in a 50 mL
Erlenmeyer flask. The flask was heated using a Bunsen burner until
the two solids melted, yielding a viscous liquid. Heating was
continued for 5−7 min, with constant swirling of the flask. After
heating was discontinued, the flask was allowed to cool until warm to
the touch, whereupon diethyl ether was added to promote
crystallization (∼25 mL). The monoacylaminal was isolated by
vacuum filtration with the crystals being washed with ice-cold 3:1
diethyl ether and 95% ethanol to remove residual amide.
N-(Morpholinobenzyl)-4-nitrobenzamide (5a). 5a with a melting
point of 161−164 °C, lit. mp = 161−164 °C.49
N-(Morpholinobenzyl)-4-(trifluoromethyl)benzamide (5b). 5b
was obtained by reacting 1.00 g (5.29 × 10−3 mol) of 4-
(trifluoromethyl)benzamide with 1.39 g (5.29 × 10−3 mol) of 4 as
described above. The crude product was isolated from ether by
vacuum filtration and recrystallized from 1:1 H2O:95% ethanol. The
product was a white powder weighing 0.2967 g (7.88 × 10−4 mol,
1
15% yield) with a melting point of 187−189 °C. H NMR 500 MHz
(DMSO-d6) δ 9.13 (d, 2H); 8.14 (d, 2H); 7.88 (d, 2H); 7.54 (d,
2H); 7.40 (t, 2H); 7.32 (t, 1H); 5.93 (1H, d); 3.67−3.56 (m, 4H);
2.55−2.49 (m, 4H). 13C{1H} NMR 125 MHz (DMSO-d6): δ 166.6,
2
139.3, 138.6, 131.7 (q, JCF = 32 Hz, C−CF3), 129.1, 128.7, 128.1,
1
127.9, 125.7, 124.4 (q, JCF = 273 Hz, CF3), 71.9, 66.7, 49.3. HRMS
(ESI-TOF) m/z: Calcd for C19H19F3N2O2H [M + H]+ 365.1471;
Found 365.1501.
N-(Morpholinobenzyl)-4-chlorobenzamide (5c). 5c with a melt-
ing point of 140−143 °C, lit. mp = 140−143 °C.49
N-(Morpholinobenzyl)benzamide (5d). 5d with a melting point of
160−162 °C, lit. mp = 166−167 °C.57
N-(Morpholinobenzyl)-4-methylbenzamide (5e). 5e with a
melting point of 134−136 °C, lit. mp = 134−136 °C.49
N-(Morpholinobenzyl)-4-methoxybenzamide (5f). 5f was ob-
tained by reacting 1.012 g (6.70 × 10−3 mol) of 4-methoxybenzamide
with 1.74 g (6.62 × 10−3 mol) of 4 as described above. The crude
product was isolated from ether by vacuum filtration and recrystal-
lized from 1:1 H2O:95% ethanol. The product was a white powder
weighing 1.273 g (4.0 × 10−3 mol, 60% yield) with a melting point of
EXPERIMENTAL SECTION
■
All chemicals were reagent grade from commercial sources and were
used without further purification unless otherwise noted. HCl,
potassium acetate, potassium hydrogen phosphate, potassium
carbonate, potassium chloride, potassium pyrophosphate, chloroace-
tate acid, and potassium hydroxide were obtained from commercial
sources. The water used in the kinetic and HPLC studies was house
DI-water that was then passed through a water purification system.
Standard potassium hydroxide and HCl solutions were obtained by
dilution of concentrated solutions with the final hydroxide or HCl
concentration obtained by titration to the phenolphthalein end point
using standard HCl and hydroxide solutions, respectively.
1
158−161 °C. H NMR 400 MHz (CDCl3) δ 7.8 (d, 2H); 7.55 (d,
2H); 7.25−7.45 (m, 3H); 6.95 (d, 2H); 6.4−6.5 (bd, 1H); 5.9 (1H,
d); 3.9 (s, 3H); 3.7−3.8 (m, 4H); 3.6−3.7 (m, 2H); 3.5−3.6 (m,
2H). 13C{1H} NMR 125 MHz (DMSO-d6): δ 166.9, 162.2, 139.8,
130.1, 129.8, 128.6, 127.9, 126.8, 113.9, 71.6, 66.8, 55.9, 49.4. HRMS
(ESI-TOF) m/z: Calcd for C19H22N2O3H [M + H]+ 327.1703;
Found 327.1700.
Carbinolamide Synthesis. All N-(hydroxybenzyl)benzamide
derivatives were synthesized using a similar general method based
upon a previously published method.44 Approximately 1 g of the
monoacylaminal was dissolved in 30 mL of a 1:2 mixture of DI-water
and reagent grade acetone. The reaction solution was stirred and
gently warmed (∼30 °C) for 0.5−1.5 h. The “pH” of the solution,
monitored using a pH meter, was maintained at a pH reading between
All melting points are uncorrected. The NMR spectra were
obtained in deuterated DMSO or CDCl3 utilizing a 400 or 500 MHz
instrument.
Synthesis. All of the derivatives of N-(hydroxybenzyl)benzamide
were synthesized using the general pathway shown in Scheme 5.
H
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
4 and 4.5 via the addition of small amounts of 0.10 M HCl or 1.00 M
HCl, depending upon the observed rate of change in the “pH” of the
reaction solution. As the reaction was nearing completion, changes in
the observed “pH” would slow and precipitate would begin to collect
in the reaction vessel. When the reaction was complete, the crystals
were collected by vacuum filtration and further purified by washing
with ice cold 95% ethanol.
N-(Hydroxybenzyl)-4-methoxybenzamide (2f). 2f was obtained
by reacting 0.704 g (2.17 × 10−3 mol) of 5f in 30 mL of a 1:2 mix of
DI-water and reagent grade acetone. The pH was monitored, and the
pH was maintained between 4 and 4.5 by the addition of 0.1 M HCl
or 1.0 M HCl. The product precipitated out of solution as the
reaction proceeded. After no further change in pH was observed, the
reaction solution was cooled, and the product was collected by
vacuum filtration. The product was washed with ice-cold 95% ethanol
and dried in a vacuum desiccator. The product was a white powder
that weighed 0.189 g (7.35 × 10−4 mol, 33% yield) with a melting
N-(Hydroxybenzyl)-4-nitrobenzamide (2a). 2a was obtained by
reacting 0.830 g (2.43 × 10−3 mol) of 5a in ∼30 mL of a 1:2 mix of
DI-water and reagent grade acetone. The pH was monitored, and the
pH was maintained between 4 and 4.5 by the addition of 0.1 M HCl
or 1.0 M HCl. The product precipitated out of solution as the
reaction proceeded. After no further change in pH was observed, the
reaction solution was cooled, and the product was collected by
vacuum filtration. The product was washed with ice-cold 95% ethanol
and dried in a vacuum desiccator. The product was a pale-yellow
powder that weighed 0.278 g (1.02 × 10−3 mol, 42% yield) with a
1
point of 183−186 °C. H NMR 500 MHz (DMSO-d6): δ 8.87 (d,
1H), 7.85 (d, 2H), 7.40 (d, 2H), 7.27 (t, 2H), 7.20 (t, 1H), 6.89 (d,
2H), 6.45 (dd, 1H), 6.27 (d, 1H), 3.72 (s, 3H). 13C{1H} NMR 125
MHz (DMSO-d6): δ 165.7, 162.2, 143.0, 130.0, 128.4, 127.8, 126.9,
126.6, 113.9, 74.2, 55.8. HRMS (ESI-TOF) m/z: Calcd for
C15H15NO3H [M + H]+ 258.1124; Found 258.1120.
Kinetics. The rate constants for the aqueous reaction were
determined as a function of pH. Since the rates of the aqueous
reaction varied by several orders of magnitude over the entire pH−
rate profile under examination (see graphical abstract), several
techniques were required to follow the breakdown of the
carbinolamides studied.
UV−vis Spectroscopy. The kinetic studies observed between pH
∼6−9 and 0−4 were followed using a UV−vis spectrophotometer
with a temperature-controlled sample transport tray maintained at 25
°C. Experiments were initiated using ∼2−3 μL of a 0.1 M stock
solution of the carbinolamides dissolved in DMSO, into 3 mL of a
thermally equilibrated reaction solution in a quartz cuvette, to yield a
final substrate concentration of ∼6.67 × 10−5 to 1.00 × 10−4 M. The
compounds studied were monitored through the change in
absorbance at 250 nm as a function of time with kobsd determined
using the software supplied with the instrument. For those
experiments where buffers were used to maintain the pH of reaction
solution, the observed rate constant was determined at, at least, three
different buffer concentrations, at the same pH. The observed rate,
under buffered conditions at a particular pH value, was determined by
the linear extrapolation back to zero [buffer]. The pH values of the
kinetic solutions were measured after the reaction was completed via
standardized combination glass electrode attached to a pH meter. The
reported kobsd at each pH was the average of, at least, three trials at the
specific pH value.
1
melting point of 106−109 °C. H NMR 500 MHz (DMSO-d6): δ
9.44 (d, 1H), 8.23 (d, 2H), 8.07 (d, 2H), 7.44 (d, 2H), 7.31 (t, 2H),
7.23 (t, 1H), 6.49−6.46 (m, 2H). 13C{1H} NMR 100 MHz (DMSO-
d6): δ 164.7, 149.6, 142.4, 140.4, 129.6, 128.5, 128.1, 126.6, 123.9,
74.5. HRMS (ESI-TOF) m/z: Calcd for C14H12N2O4H [M + H]+
273.0869; Found 273.0865.
N-(Hydroxybenzyl)-4-(trifluoromethyl)benzamide (2b). 2b was
obtained by reacting 0.1835 g (5.04 × 10−4 mol) of 5b in ∼50 mL of
a 1:1 mix of DI-water and reagent grade acetone. The pH was
monitored, and the pH was maintained between 4 and 4.5 by the
addition of 0.1 M HCl. The product precipitated out of solution as
the reaction proceeded. After no further change in pH was observed,
the reaction solution was cooled, and the product was collected by
vacuum filtration. The product was washed with ice-cold 95% ethanol
and dried in a vacuum desiccator. The product was a light-yellow
powder that weighed 0.1103 g (3.74 × 10−4 mol, 74% yield) with a
melting point of 127−128 °C. 1H NMR 500 MHz (DMSO-d6) δ 9.40
(d, 1H), 8.12 (2H, d), 7.85 (2H, d), 7.51 (2H. d), 7.38 (2H, t), 7.31
(1H, t), 6.56 (1H, dd), 6.52 (1H, d). 13C{1H} NMR 125 MHz
2
(DMSO-d6): δ 165.2, 142.6, 138.5, 131.7 (q, JCF = 32 Hz, C−CF3),
129.0, 128.5, 128.0, 126.6, 125.7, 124.4 (q, 1JCF = 273 Hz, CF3), 74.4.
HRMS (ESI-TOF) m/z: Calcd for C15H12F3NO2H [M + H]+
296.0893; Found 296.0885.
N-(Hydroxybenzyl)-4-chlorobenzamide (2c). 2c was obtained by
reacting 0.800 g (2.42 × 10−3 mol) of 5c in ∼30 mL of a 1:2 mix of
DI-water and reagent grade acetone. The pH was monitored, and the
pH was maintained between 4 and 4.5 by the addition of 0.1 M HCl
or 1.0 M HCl. The product precipitated out of solution as the
reaction proceeded. After no further change in pH was observed, the
reaction solution was cooled, and the product was collected by
vacuum filtration. The product was washed with ice-cold 95% ethanol
and dried in a vacuum desiccator. The product was a white powder
that weighed 0.360 g (1.37 × 10−3 mol, 57% yield) with a melting
The hydroxide concentration was calculated using eq 6 with an
apparent activity coefficient for hydroxide of γHO− = 0.79,68
determined from the measured pH of known [HO−] in water at I
= 1.0 M (KCl) at 25 °C.
)
10(pH−K
γHO−
w
[HO−] =
(6)
HPLC Kinetic Studies. The kinetic experiments for pH values
between 4 and 6 (pH 2−5 for 2a) were too slow to be followed
conveniently with the UV−vis spectrophotometer. For those
experiments, the reaction was followed by means of an HPLC
instrument. A general HPLC kinetic experiment commenced with the
∼3−4 μL injection of a stock solution of the N-(hydroxybenzyl)-
benzamide derivative in DMSO (0.1 M) into 4 mL of the reaction
solution (with varying buffer concentration) to yield a substrate
concentration of ∼6.7 × 10−5 M. Prior to substrate addition and
subsequent to addition, samples were thermally equilibrated
throughout the reaction period in an autosampler, maintained at 25
°C. The HPLC utilized had dual pumps and sphotodiode array
detector, all remotely controlled by computer. The column was a C-
18 reverse phase column with a guard column containing the same
material. Reaction progress was monitored by injection of 100 μL
volumes of the reaction solution into the instrument at timed
intervals. Reaction progress was followed by the observation of the
disappearance of the carbinolamide (Acar) and the concurrent
appearance of the amide (Aamide) or aldehyde (Aald), depending on
the substrate, as a function of time (s) (see eq 1). The pH of the
reaction solutions was determined prior to the initiation of the
1
point of 110−112 °C. H NMR 500 MHz (DMSO-d6): δ 9.23 (d,
1H), 7.96 (d, 2H), 7.54 (d, 2H), 7.50 (d, 2H), 7.38 (t, 2H), 7.30 (t,
1H), 6.54 (dd, 1H), 6.46 (d, 1H). 13C{1H} NMR 125 MHz (DMSO-
d6): δ 165.2, 142.7, 136.7, 133.5, 130.0, 128.7, 128.4, 128.0, 126.6,
74.3. HRMS (ESI-TOF) m/z: Calcd for C14H12ClNO2H [M + H]+
262.0629; Found 262.0625.
N-(Hydroxybenzyl)benzamide (2d). 2d was obtained in a 75.7%
yield with a melting point of 112−114 °C (lit. mp = 114−115 °C).44
1H NMR 400 MHz (DMSO-d6): δ 9.14 (d, 1H), 7.92 (d, 2H), 7.41−
7.58 (m, 5H), 7.37 (t, 2H), 7.28 (t, 1H), 6.57 (m, 1H), 6.42 (d, 1H).
13C{1H} NMR 100 MHz (DMSO-d6): δ 166.5, 143.05, 134.9, 132.0,
128.9, 128.7, 128.3, 128.1, 126.8, 74.5.
N-(Hydroxybenzyl)-4-methylbenzamide (2e). 2e was obtained in
51.1% yield with a melting point of 102−105 °C (lit. mp = 105 °C).67
1H NMR 400 MHz (DMSO-d6): δ 9.05 (d, 1H), 7.84 (d, 2H), 7.49
(d, 2H), 7.40 (t, 1H), 7.38 (t, 2H), 7.25 (d, 2H), 6.51−6.58 (m, 1H),
6.42 (d, 1H), 2.35 (s, 3H). 13C{1H} NMR 100 MHz (DMSO-d6): δ
165.6, 142.4, 141.2, 131.5, 128.8, 128.7, 127.9, 127.6, 126.1, 73.7,
20.9.
I
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
reaction and upon completion of the reaction. No significant changes
in pH (>0.05) were ever observed.
(6) Valente, E.; Gomes, J. R. B.; Moreira, R.; Iley, J. Kinetics and
mechanism of hydrolysis of N-acyloxymethyl derivatives of azetidin-2-
one. J. Org. Chem. 2004, 69, 3359.
(7) Tenn, W. J., III; Murphy, J. L.; Bim-Merle, J. K.; Brown, J. A.;
Junia, A. J.; Price, M. A.; Nagorski, R. W. Amidates as Leaving
Groups: Structure/Reactivity Correlation of the Hydroxide-Depend-
ent E1cB-like Breakdown of Carbinolamides in Aqueous Solution. J.
Org. Chem. 2007, 72, 6075.
(8) Pinner, A. Ueber die Producte der Einwirkung von Chlor und
Brom auf den Aldehyd der Aethylreihe. Justus Liebigs Annalen der
Chemie. 1875, 179, 21.
(9) Vela, M.; Kohn, H. The Synthesis and Reactivity of [N(8)-
C(3′)]-Cyclized Bicyclomycin. Evidence of the Role of the C(1)-
Triol Group in Bicyclomycin-Mediated Processes. J. Org. Chem. 1992,
57, 5223.
(10) Miyoshi, T.; Miyari, N.; Aoki, H.; Kohsaka, M.; Sakai, H.;
Imanaka, H. Bicyclomycin, A New Antibiotic. I. Taxonomy, Isolation,
and Characterization. J. Antibiot. 1972, 25, 569.
(11) Miyamura, S.; Ogasawara, N.; Otsuka, H.; Niwayama, S.;
Takana, H.; Take, T.; Uchiyama, T.; Ochiai, H.; Abe, K.; Koizumi, K.;
Asao, K.; Matuski, K.; Hoshino, T. Antibiotic No. 5879, a New Water-
Soluble Antibiotic Against Gram-Negative Bacteria. J. Antibiot. 1972,
25, 610.
(12) Asami, Y.; Kakeya, H.; Onose, R.; Yoshida, A.; Matsuzaki, H.;
Osada, H. Azaspirene: A novel angiogenesis inhibitor containing a 1-
oxa-7-azaspiro[4.4]non-2-ene-4,6-dione skeleton produced by the
fungus Neosartotya sp. Org. Lett. 2002, 4, 2845.
(13) Nakai, R.; Ogawa, H.; Asai, A.; Ando, K.; Agatsuma, T.;
Matsumiya, S.; Akinaga, S.; Yamashita, Y.; Mizukami, T. UCS1025A,
a novel antibiotic produced by Acremonium sp. J. Antibiot. 2000, 53,
294.
The observed rates of the reaction utilizing the HPLC were
calculated using eq 1. The areas observed for the amide or aldehyde
were statistically corrected for differences in the molar absorptivities
between the carbinolamide and the amide or aldehyde product. The
correction factor (Rx) was determined by following the reaction by
HPLC for ∼2−5 half-lives and determining the differences in the total
area changes between the carbinolamide and the amide or aldehyde
(Rx values shown in Table S1 of the Supporting Information). The
correction factor was determined at, at least, two other pH values, and
good agreement between the values was found in all cases.
Stopped-Flow Kinetics. Studies were performed using a stopped-
flow spectrophotometer for reactions that were complete in less than
1 min. For all kinetic experiments, the solutions were equilibrated at
25 °C for at least 10 min prior to the initiation of any experiment.
Each experiment consisted of injection of a 0.1 M solution of the
carbinolamide under study in DMSO into an aqueous solution. The
stock solution was injected into the reaction solution (water or 1 M
KCl solution) to yield a final concentration of approximately 1 × 10−4
M carbinolamide. All compounds studied were monitored at an
absorbance of 250 nm. The observed rate constants were calculated
using instrument supplied software, and the kobsd values reported are
the average of at least five kinetic experiments.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.9b02812.
pH−rate profiles and hydroxide plots for all compounds
studied and the observed rate constants that were used
to determine kobsd at specific pH values (PDF)
(14) Suzuki, S.; Hosoe, T.; Nozawa, K.; Kawai, K.; Yaguchi, T.;
Udagawa, S. Antifungal substances against pathogenic fungi,
talaroconvolutins, from Talaromyces convolutus. J. Nat. Prod. 2000,
63, 768.
(15) Singh, S. B.; Goetz, M. A.; Jones, E. T.; Bills, G. F.; Giacobbe,
R. A.; Herranz, L.; Stevensmiles, S.; Williams, D. L. Oteromycin - a
Novel Antagonist of Endothelin Receptor. J. Org. Chem. 1995, 60,
7040.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 309 438 8978. Fax: 309 438 5538. E-mail: rnagor@
ilstu.edu.
(16) Kakeya, H.; Takahashi, I.; Okada, G.; Isono, K.; Osada, H.
Epolactaene, a Novel Neuritogenic Compound Human Neuro-
blastoma Cells, Produced by a Marine Fungus. J. Antibiot. 1995, 48,
733.
ORCID
Richard W. Nagorski: 0000-0002-0769-9972
Notes
(17) Hayashi, Y.; Shoji, M.; Yamaguchi, J.; Sato, K.; Yamaguchi, S.;
Mukaiyama, T.; Sakai, K.; Asami, Y.; Kakeya, H.; Osada, H.
Asymmetric Total Synthesis of (−)-Azaspirene, a Novel Angiogenesis
Inhibitor. J. Am. Chem. Soc. 2002, 124, 12078.
(18) Lawson, M. R.; Dyer, K.; Berger, J. M. Ligand-induced and
small-molecule control of substrate loading in a hexameric helicase.
Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 13714.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Department of Chemistry and
Illinois State University.
(19) Williams, R. M.; Durham, C. A. Bicyclomycin: Synthetic,
Mechanistic, and Biological Studies. Chem. Rev. 1988, 88, 511.
(20) Williams, R. M.; Tomizawa, K.; Armstrong, R. W.; Dung, J.-S.
Mechanism, Biological Relevance, and Structural Requirements for
Thiolate Additions to Bicyclomycin and Analogues: A Unique Latent
Michael Acceptor System. J. Am. Chem. Soc. 1987, 109, 4028.
(21) Agatsuma, T.; Akama, T.; Nara, S.; Matsumiya, S.; Nakai, R.;
Ogawa, H.; Otaki, S.; Ikeda, S.-i.; Saitoh, Y.; Kanda, Y. UCS1025A
and B, new antitumor antibiotics from the fungus Acremonium
species. Org. Lett. 2002, 4, 4387.
(22) Emoto, M.; Yano, K.; Choijamts, B.; Sakai, S.; Hirasawa, S.;
Wakamori, S.; Aizawa, M.; Nabeshima, K.; Tachibana, K.; Kanomata,
N. Azaspirene analogs inhibit the growth of human uterine
carcinosarcoma in vitro and in vivo. Anticancer Res. 2015, 35, 2739.
(23) Li, L.; Tang, M.-C.; Gao, S.; Soliman, S.; Hang, L.; Xu, W.;
Tang, Y.; Tang, S.; Ye, T.; Watanabe, K. Genome Mining and
Assembly-Line Biosynthesis of the UCS1025A Pyrrolizidinone Family
of Fungal Alkaloids. J. Am. Chem. Soc. 2018, 140, 2067.
REFERENCES
■
(1) Jencks, W. P. Catalysis in Chemistry an Enzymology; Dover
Publications Inc.: Mineola, NY, 1975.
(2) Iley, J.; Lopes, F.; Moreira, R. Kinetics and mechanism of
hydrolysis of N-amidomethylsulfonamides. J. Chem. Soc.-Perkin Trans.
2001, 2, 749.
(3) Iley, J.; Moreira, R.; Rosa, E. Acyloxymethyl as a Drug Protecting
Group. Kinetics and Mechanism of the Hydrolysis of N-
Acyloxymethylbenzamides. J. Chem. Soc., Perkin Trans. 2 1991, 2, 563.
(4) Lopes, F.; Moreira, R.; Iley, J. Acyloxymethyl as a drug
protecting group. Part 5. Kinetics and mechanism of the hydrolysis of
tertiary N-acyloxymethylsulfonamides. J. Chem. Soc., Perkin Trans. 2
1999, 2, 431.
(5) Moreira, R.; Santana, A. B.; Iley, J.; Neres, J.; Douglas, K. T.;
Horton, P. N.; Hursthouse, M. B. Design, synthesis, and enzymatic
evaluation of N-1-acyloxyalkyl- and N-1-oxazolidin-2,4-dion-5-yl-
substituted beta-lactams as novel inhibitors of human leukocyte
elastase. J. Med. Chem. 2005, 48, 4861.
J
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(24) Young, S. D.; Tamburini, P. P. Enzymatic Peptidyl α-Amidation
Proceeds Through Formation of an α-Hydroxyglycine Intermediate. J.
Am. Chem. Soc. 1989, 111, 1933.
(25) Eipper, B. A.; Mains, R. E. Peptide a-Amidation. Annu. Rev.
Physiol. 1988, 50, 333.
(26) Bradbury, A. F.; Finnie, M. D. A.; Smyth, D. G. Mechanism of
C-terminal amide formation by pituitary enzymes. Nature 1982, 298,
686.
(47) Bundgaard, H.; Johansen, M. Pro-drugs as Drug Delivery
Systems VIII. Bioreversible Derivatization of Hydantoins by N-
hydroxymethylation. Int. J. Pharm. 1980, 5, 67.
(48) Johansen, M.; Bundgaard, H. Prodrugs as Drug Delivery
Systems. VI. Kinetics and Mechanism of the Decomposition of N-
hydroxymethylated Amides and Imides in Aqueous Solution and
Assessment of Their Suitability as Possible Prodrugs. Arch. Pharm.
Chemi, Sci. Ed. 1979, 7, 175.
(49) Sansone, M. F.; Koyanagi, T.; Przybyla, D. E.; Nagorski, R. W.
Solvent-free synthesis of monoacylaminals from the reaction of
amides and aminals as precursors in carbinolamide synthesis.
Tetrahedron Lett. 2010, 51, 6031.
(50) Bunton, C. A.; Nayak, B.; O’Connor, C. Alkaline hydrolysis of
benzamide and N-methyl- and N,N-dimethylbenzamide. J. Org. Chem.
1968, 33, 572.
(51) Slebocka-Tilk, H.; Bennet, A. J.; Keillor, J. W.; Brown, R. S.;
Guthrie, J. P.; Jodhan, A. Oxygen-18 exchange accompanying the
basic hydrolysis of primary, secondary, and tertiary toluamides. 2. The
importance of amine leaving abilities from the anionic tetrahedral
intermediate. J. Am. Chem. Soc. 1990, 112, 8507.
(52) In general, the course of the reactions, followed by HPLC, was
performed by following the disappearance of the carbinolamide and
the appearance of the aldehyde (see the Rx values in the Supporting
Information). Those experiments where the appearance of amide was
compared to the disappearance of the carbinolamide yielded rate
constants that were within error of one another.
(27) McIninch, J. K.; McIninch, J. D.; May, S. W. Catalysis,
Stereochemistry, and Inhibition of Ureidoglycolate Lyase. J. Biol.
Chem. 2003, 278, 50091.
(28) Takada, Y.; Noguchi, T. Ureidoglycolate Lyase, a New
Metalloenzyme of Peroxisomal Urate Degradation in Marine Fish
Liver. Biochem. J. 1986, 235, 391.
(29) Chung, F. L.; Nath, R. G.; Nagao, M.; Nishikawa, A.; Zhou, G.
D.; Randerath, K. Endogenous Formation and Significance of 1,N2-
propanodeoxyguanosine Adducts. Mutat. Res., Fundam. Mol. Mech.
Mutagen. 1999, 424, 71.
(30) Marnett, L. J. Lipid Peroxidation-DNA Damage by
Malondialdehyde. Mutat. Res., Fundam. Mol. Mech. Mutagen. 1999,
424, 83.
(31) Vilanova, B.; Fernandez, D.; Casasnovas, R.; Pomar, A. M.;
Alvarez-Idaboy, J. R.; Hernandez-Haro, N.; Grand, A.; Adrover, M.;
Donoso, J.; Frau, J.; Munoz, F.; Ortega-Castro, J. Formation
mechanism of glyoxal-DNA adduct, a DNA cross-link precursor. Int.
J. Biol. Macromol. 2017, 98, 664.
(32) Tenn, W. J.; French, N. L.; Nagorski, R. W. Kinetic dependence
of the aqueous reaction of N-(hydroxymethyl)benzamide derivatives
upon addition of electron-withdrawing groups. Org. Lett. 2001, 3, 75.
(33) Mennenga, A. G.; Johnson, A. L.; Nagorski, R. W. General-
Buffer Catalysis of the Reaction of N-(Hydroxymethyl)benzamide: A
New Pathway for the Aqueous Reaction of Carbinolamides.
Tetrahedron Lett. 2005, 46, 3079.
(53) Bundgaard, H.; Kahns, A. H. Chemical stability and plasma-
catalyzed dealkylation of peptidyl-α-hydroxyglycine derivatives -
intermediates in peptide α-amidation. Peptides (N. Y., NY, U. S.)
1991, 12, 745.
(54) Jacobson, P. Ann. Chim. Phys. 1871, 157, 245.
(55) Zaugg, H. E.; Martin, W. B. α-Amidoalkylations at Carbon. Org.
React. 1965, 14, 52.
(56) Macovski, E.; Bachmeyer, C. N-(α-1-Piperidylbenzyl)-
benzamides. Ber. Dtsch. Chem. Ges. B 1944, 77B, 495.
(57) Sakai, H.; Sekiya, M. Reaction of amide homologs. XXIII.
Syntheses of α-amidobenzylamines. Chem. Pharm. Bull. 1969, 17, 32.
(58) Cordes, E. H. Mechanism and catalysis for the hydrolysis of
acetals, ketals, and ortho esters. Prog. Phys. Org. Chem. 2007, 4, 1.
(59) Cordes, E. H.; Bull, H. G. Mechanism and catalysis for
hydrolysis of acetals, ketals, and ortho esters. Chem. Rev. 1974, 74,
581.
(60) The rate and equilibrium constants were obtained using a
nonlinear least-squares fit of the data to eq 2.
(61) Jaffe, H. H. A Reexamination of the Hammett Equation. Chem.
Rev. 1953, 53, 191.
(62) Vogel’s Textbook of Practical Organic Chemistry, 5th ed.; Furniss,
B. S., Hannaford, A. J., Smith, P. W. G., Tatchell, A. R., Eds.; John
Wiley & Sons: New York, 1989; p 1073.
(63) Grammaticakis, P. Medium ultraviolet and visible absorption of
N-aroylaryl-amines. III. o-, m-, and p-methoxybenzoylarylamines. Bull.
Soc. Chim. Fr. 1964, 924.
(64) Zhang, L.; Wang, S.; Zhou, S.; Yang, G.; Sheng, E. Cannizzaro-
Type Disproportionation of Aromatic Aldehydes to Amides and
Alcohols by Using Either a Stoichiometric Amount or a Catalytic
Amount of Lanthanide Compounds. J. Org. Chem. 2006, 71, 3149.
(65) Dictionary of Organic Compounds, 6th ed.; Buckingham, J. B.,
Ed.; Chapman & Hill: New York, 1996.
(66) Sakoda, R.; Matsumoto, H.; Seto, K. Novel synthesis of α-
acetylstyrylphosphonates. Synthesis 1993, 1993, 705.
(67) Katritzky, A. R.; Rao, M. S. C. A novel synthesis of acyclic N-
(α-hydroxybenzyl)benzamides. Synthesis 1990, 1990, 663.
(68) Amyes, T. L.; Richard, J. P. Determination of the PK(a), of
Ethyl Acetate - Bronsted Correlation For Deprotonation of a Simple
Oxygen Ester in Aqueous Solution. J. Am. Chem. Soc. 1996, 118, 3129.
(34) Ankem, R. V.; Murphy, J. L.; Nagorski, R. W. Rate of formation
of N-(hydroxymethyl)benzamide derivatives in water as a function of
pH and their equilibrium constants. Tetrahedron Lett. 2008, 49, 6547.
(35) Murphy, J. L.; Tenn, W. J., III; Labuda, J. J.; Nagorski, R. W.
Rapid amidic hydrolysis: A competitive reaction pathway under basic
conditions for N-(hydroxymethyl)benzamide derivatives bearing
electron-donating groups. Tetrahedron Lett. 2009, 50, 7358.
(36) Hine, J.; Hine, M. The relative acidity of water, methanol and
other weak acids in isopropyl alcohol solution. J. Am. Chem. Soc. 1952,
74, 5266.
(37) Bell, R. P.; McTigue, P. T. Kinetics of the aldol condensation of
acetaldehyde. J. Chem. Soc. 1960, 2983.
(38) Bell, R. P. The reversible hydration of carbonyl compounds.
Adv. Phys. Org. Chem. 1966, 4, 1.
(39) McClelland, R. A.; Coe, M. Structure-reactivity effects in the
hydration of benzaldehydes. J. Am. Chem. Soc. 1983, 105, 2718.
(40) Greenzaid, P. Reversible addition of hydroxide to substituted
benzaldehydes. J. Org. Chem. 1973, 38, 3164.
(41) Sayer, J. M. Hydration of p-nitrobenzaldehyde. J. Org. Chem.
1975, 40, 2545.
(42) Bell, R. P.; Soerensen, P. E. Kinetics of the addition of
hydroxide ions to substituted benzaldehydes. J. Chem. Soc., Perkin
Trans. 2 1976, 2, 1594.
(43) Guthrie, J. P. Equilibrium constants for a series of simple aldol
condensations and linear free energy relations with other carbonyl
addition reactions. Can. J. Chem. 1978, 56, 962.
(44) Bundgaard, H.; Johansen, M. Hydrolysis of N-(α-
hydroxybenzyl)benzamide and other N-(α-hydroxyalkyl)amide De-
rivatives: Implications for the Design of N-acyloxyalkyl-type Prodrugs.
Int. J. Pharm. 1984, 22, 45.
(45) Bundgaard, H.; Buur, A. Prodrugs as Drug Delivery Systems.
65. Hydrolysis of α-hydroxy- and α-acyloxy-N-benzoylglycine
Derivatives and Implications for the Design of Prodrugs of NH-
acidic Compounds. Int. J. Pharm. 1987, 37, 185.
(46) Bundgaard, H. Design of Prodrugs; Elsevier: Amsterdam, The
Netherlands, 1985; p 1.
K
DOI: 10.1021/acs.joc.9b02812
J. Org. Chem. XXXX, XXX, XXX−XXX