Intramolecular general base catalysis in the hydrolysis of a phosphate diester. calculational guidance to a choice of mechanism

Article abstract of DOI:10.1021/jo302498g

Notwithstanding its half-life of 70 years at 25 C, the spontaneous hydrolysis of the anion of di-2-pyridyl phosphate (DPP) is thousands of times faster (ca. 3000 at 100 C, over 10000-fold at 25 C) than expected for a diester with leaving groups of pK_a 9.09. The kinetic parameters do not permit a conclusive choice between five possible mechanisms considered, but the combination of kinetics and calculational evidence supports a single-step, concerted, S_N2(P) mechanism involving the attack of solvent water on phosphorus assisted by intramolecular catalysis by a (weakly basic) pyridine nitrogen acting as a general base. Catalysis is relatively efficient for this mechanism, with an estimated effective molarity (EM) of the general base of >15 M, consistent with the absence of catalysis by typical buffers. Further new results confirm that varying the nonleaving group has minimal effect on the rate of spontaneous diester hydrolysis, in striking contrast to the major effect on the corresponding reaction of triesters: though protonation of one nitrogen of DPP^- increases the rate of hydrolysis by 6 orders of magnitude, in line with expectation.

Full text of DOI:10.1021/jo302498g

Featured Article
pubs.acs.org/joc
Intramolecular General Base Catalysis in the Hydrolysis of a
Phosphate Diester. Calculational Guidance to a Choice of Mechanism
Anthony J. Kirby,*,† Michelle Medeiros,‡ Jose R. Mora,‡ Pedro S. M. Oliveira,‡ Almahdi Amer,§
́
Nicholas H. Williams,*,§ and Faruk Nome*,‡
†University Chemical Laboratory, University of Cambridge, Cambridge CB2 1EW, U.K.
‡Departamento de Química, Universidade Federal de Santa Catarina, 88040-900 Florianop
́
olis, SC, Brazil
§Centre for Chemical Biology, Department of Chemistry, University of Sheffield, Sheffield S3 7HF, U.K.
S
* Supporting Information
ABSTRACT: Notwithstanding its half-life of 70 years at 25
°C, the spontaneous hydrolysis of the anion of di-2-pyridyl
phosphate (DPP) is thousands of times faster (ca. 3000 at 100
°C, over 10000-fold at 25 °C) than expected for a diester with
leaving groups of pKa 9.09. The kinetic parameters do not
permit a conclusive choice between five possible mechanisms
considered, but the combination of kinetics and calculational
evidence supports a single-step, concerted, SN2(P) mechanism
involving the attack of solvent water on phosphorus assisted by
intramolecular catalysis by a (weakly basic) pyridine nitrogen
acting as a general base. Catalysis is relatively efficient for this
mechanism, with an estimated effective molarity (EM) of the
general base of >15 M, consistent with the absence of catalysis
by typical buffers. Further new results confirm that varying the nonleaving group has minimal effect on the rate of spontaneous
diester hydrolysis, in striking contrast to the major effect on the corresponding reaction of triesters: though protonation of one
nitrogen of DPP− increases the rate of hydrolysis by 6 orders of magnitude, in line with expectation.
Scheme 2), and we initiated an investigation of the catalytic
properties of the 2-pyridyl group in phosphate esters.
INTRODUCTION
■
We recently reported1 “first results” of an investigation
prompted by the intriguing report of Liu and Wulff2 that
their attempt to prepare di-2-pyridyl phosphate (DPP) as a new
template “failed owing to its instability”. 2-Hydroxypyridine,
which exists in water almost exclusively as 2-pyridone (Scheme
1), has a nominal pKa of 9.09,3 and its anion is not obviously a
Scheme 2. Suggested Special Mechanism for “Efficient”
Intramolecular GBC of the Hydrolysis of TPP1
Scheme 1. Equilibria Involving 2-Hydroxypyridine3
Our first results were for the triester TPP, which we found to
be hydrolyzed near pH 7 some 107 times faster than expected
on the basis of a linear free energy relationship derived for
dialkyl aryl phosphate triesters. However, we soon established
that this enhanced reactivity is fully accounted for by the
activating electronic effects of the aromatic nonleaving groups4
so that there is no need to invoke intramolecular catalysis.
We now report our results with the diester DPP, the original
target of the investigation. We confirm that the hydrolysis of
good leaving group. So this apparent instability suggested that
the neighboring pyridine nitrogens could be involved in
intramolecular catalysis. The geometry is not favorable for
intramolecular nucleophilic catalysis, which would involve the
formation of a four-membered ring, but the system appears to
be well set up for intramolecular general base catalysis (IGBC,
Received: November 21, 2012
Published: February 1, 2013
© 2013 American Chemical Society
1343
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
the DPP anion is very slow, as shown previously by Brown and
Zamkanei,5 who found that DPP− was “not hydrolysed” under
their mild conditions (pH 6−8, 37 °C, in the presence or
absence of Co(2+) or selected complexes). But we find that
DPP does exhibit significantly enhanced reactivity, consistent
with intramolecular catalysis of hydrolysis. We discuss the
mechanism of its hydrolysis, via the more reactive monoester
MPP, to inorganic phosphate (Scheme 3) and present new
information on the more general question of the effect of the
nonleaving group on diester reactivity.
Scheme 3. Hydrolysis Reaction of Di-2-pyridyl Phosphate
(DPP)
Figure 1. pH−(log) rate profile for the hydrolysis of DPP at 25 °C
and ionic strength 1.0 M (KCl). The points for kobs are experimental;
the curves are calculated using the rate constants and pKa’s shown in
Table 1 but setting k0 to zero for kcalc. For full data see the Supporting
Information.
RESULTS AND DISCUSSION
■
Nonleaving Group Effects on Diester Hydrolysis.
Identifying the remarkably large effects of varying the
nonleaving (so-called “spectator”) groups on the rates of
hydrolysis of phosphate triesters4 prompted us to look again at
their apparently minimal effects on the reactions of diesters.
The original two-point comparison6 was based on the
observation that bis-2,4-dinitrophenyl phosphate was hydro-
lyzed at 39 °C, only 2.88 times faster than the methyl 2,4-
dinitrophenyl ester, little more than the statistical factor of 2.
Scheme 4. Three Ionic Forms of DPP Significant in the pH
Region
−
We have extended the data set to four diesters RO(PO2 )−O-
2,4-DNP, with R = Me, Ph, 4-nitrophenyl, and 2,4-
dinitrophenyl, with 2,4-dinitrophenol being the common
leaving group. The data and the Brønsted (nonleaving group)
plot are presented in the Supporting Information (Table S.1.5
and Figure S.1). The minimal dependence of the rate of
spontaneous hydrolysis on the nonleaving group RO is
confirmed by the new measurements of k0 at 100 °C, which
phosphate was found to be too slow to measure at 100 °C.7 We
made extensive efforts to measure the rate of the reaction for
DPP− directly at 25 °C, using the method of initial rates,
finding levels of random error over numerous experiments to
be frequently unacceptable. The rate constant quoted in Table
1 (corresponding to a half-life of 70 years) is the result of a
are correlated by a βNLG of −0.031
0.001 for the four
compounds. The contrast with triesters, with βNLG of −0.35
Table 1. Kinetic Data (Rate Constants and pKa’s) Derived
from the pH−rate Profile of Figure 1 for the Hydrolysis of
Di-2-pyridyl Phosphate DPP at 25 °C and Ionic Strength 1.0
0.02 per nonleaving group at 25 °C,4 is striking.
Hydrolysis of DPP. The pH−rate profile for the hydrolysis
of DPP (Figure 1) shows acid and base-catalyzed limbs
separated by a hint of a pH-independent region: the observed
rate at the pH 9.4 minimum was 4−5 times faster than
calculated for the sum of the acid- and base-catalyzed reactions
of the anion (curve labeled kcalc in Figure 1). Furthermore, this
rate is over 10000 times faster than predicted for the
spontaneous hydrolysis of a simple diaryl phosphate diester
with leaving groups of pKa = 9.09.7 (For details, see the
Supporting Information). Other things being equal, this degree
of acceleration would have been expected to generate a
significant pH-independent region in the rate profile, but, as a
result of the high reactivity of the N-protonated form DPP
(Scheme 4), the rate minimum is shifted to higher pH (9−10)
compared with rate profiles for other diaryl phosphates, which
typically show minima in the region of pH 4−5.7
a
M (KCl)
ionic
form
b
rate constants
pKa
pKLG
DPP+
kH = 6.81 0.89 x10−3 s−1
0.33 0.2
0.75
0.75
9.09
c
DPP
k
= 3.56 0.44 x10−4 s−1
2.73
DPP− k0 = 3.12 0.85 x10−10 s−1
;
kOH = 1.47 0.26 × 10−6 M−1s−1
a
A complete data set is in Table S.1.1 of the Supporting Information.
b
c
Literature value.3 Measured value.
cognate series of best measurements, made in parallel at
different buffer (mainly CHES) concentrations at the pH
minimum of 9.4−9.7, using the maximum practical substrate
concentrations (for details see the Experimental Section). The
measurements cited gave acceptably reproducible rate constants
at 25 °C and support the following conclusions: (1) The rate at
the pH minimum is (marginally) too high to be explained by
Phosphate diesters are well-known to be the least reactive
esters of phosphoric acid in the absence of acid or base: in a
comparable system, the spontaneous hydrolysis of diphenyl
1344
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
the sum of the hydroxide and acid-catalyzed reactions (Figure
1). (2) There is no catalysis by standard buffers of pKa < 10.
The presence or absence of measurable buffer catalysis of
hydrolysis is a simple, but generally reliable test of intra-
molecular catalytic efficiency.4 (3) The initial product, the
monoester MPP (Scheme 3), is hydrolyzed faster than DPP
between pH 1.2 and 12 (Figure 2), so initial rate calculations
Table 2. Kinetic Data for the Hydrolysis of Di-2-pyridyl
Phosphate anion DPP−, at the pH Minimum (9.4−9.7 at
25°C): In CHES Buffer, Ionic Strength 1.0 M (KCl) at a
Range of Temperatures
pH
T (°C)
kobs (s−1
)
pH
T (°C)
kobs (s−1
)
9.4
9.4
9.7
25
45
60
3.12 × 10−10
5.44 × 10−9
3.70 × 10−8
9.7
9.7
9.7
75
90
1.35 × 10−7
8.50 × 10−7
2.06 × 10−6
a
100
a
Measured by HPLC. ΔH⧧ = 25.1 0.5 kcal/mol, ΔS⧧ = −17.8 1.6
cal/K/mol, ΔG⧧ = 30.4 kcal/mol, at 25 °C.
Table 3. Solvent Isotope Effect Measurements: Parallel Runs
in Triplicate in CHES Buffer, pH 9.4, Ionic Strength 1.0 M
(KCl)
mean kobs (s−1
)
kH2O/kD2O
DPP− in H2O, 25 °C
3.37 0.28 × 10−10 kH/kD(25 °C) = 1.28 0.22
2.64 0.53 × 10−10
DPP− in D2O
b
DPP− in H2O, 45 °C
1.51 × 10−8
1.26 × 10−8
kH/kD (45 °C) = 1.20
DPP− in D2O
b
Single measurements at 45 °C. For full data see the Supporting
Information.
kobs = kH[DPP+] + k [DPP ] + k0[DPP−]
+ kOH[HO−]·[DPP−]
(1)
Below pH 5 the profile (Figure 1) is not a straight line of
slope −1 but shows inflections near pH 3 and pH zero
indicative of two ionisations. The reaction at acid pH (<2)
represents the hydrolysis of DPP+; while between pH 2.5 − 5
we see the reaction of the zwitterionic species DPP (the
independently measured pK2 of the pyridyl group is 2.73). This
is the ionization responsible for the marked shift of the rate
minimum to higher pH evident for the hydrolysis of DPP. The
reactions above pH 9 are the spontaneous and alkaline
hydrolysis of DPP−.
Two rate constants in the pH region are unusually high for a
phosphodiester with leaving groups of pKa 9.09. That for the
hydrolysis of DPP is simply explained: protonation of a
pyridyl N converts the leaving group to 2-pyridone, a species
over eight powers of ten less basic than its anion. In terms of a
βLG (leaving group) above unity the observed rate difference
(only 106-fold faster than DPP−) is consistent with a significant
rate enhancement for the reaction of the anion. As described
above, k0 for the spontaneous hydrolysis of the anion DPP− is
indeed significantly faster than expected at 25 °C: the observed
value of k0 at the pH minimum is estimated to be some 10000
times faster than estimated for the diester of a phenol of pKa =
9.09. (See the Supporting Information.)
Activation Parameters. Table 4 compares enthalpies and
entropies of activation for the spontaneous hydrolysis of DPP−
with those for the same reaction of representative dialkyl and
diaryl phosphodiesters taken from the literature. BMIPP
(bis[2-(1-methyl-1H-imidazolyl)phenyl] phosphate) is in-
cluded because it is known to be hydrolyzed with a
combination of intramolecular nucleophilic and general species
catalysis involving two (imidazole) nitrogens.8
The sequence illustrates very clearly the powerful effect of
efficient intramolecular nucleophilic catalysis by imidazole N,
expressed in both enthalpic and entropic terms for BMIPP, and
worth 8.6 kcal mol−1 in ΔG⧧. The relatively modest
Figure 2. pH profiles compared for the hydrolysis of DPP (curve from
Figure 1) and the product MPP. The points are experimental; the
curves are calculated from the data in Table 1.
are based in most cases on the release of 2 equivalents of 2-
pyridone. Where the rates of the sequential reactions are closely
similar, rate constants are derived from the appropriate
equation for consecutive first-order reactions, as described in
the Supporting Information. The corrections turn out to be
small.
The hydrolysis of MPP is itself of interest as a model for the
general mechanism of hydrolysis of phosphate monoester
monoanions but presents no obvious problems of measurement
or interpretation and will discussed in more detail elsewhere.
The spontaneous hydrolysis is (by far) the slowest reaction
of DPP, so should be better defined, and more easily measured,
at higher temperatures. We have measured the rate of
hydrolysis, specifically at the pH minimum, under our standard
conditions, of pH 9.4−9.7 and buffer at 25 °C (see the
Experimental Section), at a range of temperatures up to and
including 100 °C. The data for kobs (Table 2) give an excellent
linear Eyring plot, consistent with a constant mechanism over
the range 25−100 °C, and the thermodynamic parameters are
shown in Table 2. For the reaction at 100 °C the release of 2-
pyridone and the disappearance of DPP were followed to
completion, using HPLC. The rate measured at 100 °C is also
faster than predicted, on the basis of a linear free energy
relationship measured specifically for diaryl phosphate esters at
100 °C,7 by a factor of ca. 3000.
We interpret the log kobs/pH rate-profile of Figure 1 in terms
of reactions of the three ionic forms shown in Scheme 4
(above), according to eq 1.
1345
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
conditions by this, the most basic buffer used, is not
unexpected.
Table 4. Activation Parameters for the Spontaneous
Hydrolysis of Selected Phosphate Diester Anions
Mechanisms ii and iv: Intramolecular Catalysis. Why
should intramolecular catalysis be observed for the spontaneous
hydrolysis of DPP when we see little or none for the reaction of
TPP? An important difference is that the pyridine nitrogens of
DPP− are significantly more basic, with pKa = 2.73 compared
with −0.22 for TPP.1 The rate enhancements (of the order of
10000 and 3000 at 25 and 100 °C) are substantial, and might
suggest nucleophilic rather than general base catalysis by the
neighboring pyridine N. Effective molarities (EM)12 are not
available for general base catalysis (GBC) of phosphate diester
hydrolysis by pyridines (because the observed intermolecular
reactions are primarily nucleophilic6), but we can make an
informed estimate, of ca. 14 M, for the EM required to account
for the observed rate of hydrolysis of DPP− by a pyridine of pKa
2.73 by intramolecular nucleophilic catalysis (for details of the
calculation, see the Supporting Information).
spontaneous hydrolysis (k0)
ΔH⧧
ΔS⧧
ester
reaction
(C−O
(kcal mol−1
)
(cal mol−1 K−1
)
ref
9
dimethyl
phosphate
(25.0)
(−16.8)
cleavage)
(P−O
cleavage)
(P−O
cleavage)
(P−O
cleavage)
(P−O
cleavage)
dineopentyl
phosphate
29.5
−28.9
10
Table 2
11
DPP
25.1 0.5
19.0 1.2
19.0
−17.8 1.6
−9.05 0.5
−25.5
BDMIPP
bis-2,4-DNPP
7
acceleration of the hydrolysis of DPP− is fully accounted for by
an increase in the entropy of activation.
Intramolecular Catalysis. The significant rate acceler-
ations of the spontaneous hydrolysis of DPP− at 25 and 100 °C
is prima facie evidence for participation in the reaction by the
pyridine nitrogen(s). Several possible mechanisms suggest
themselves. (i) A reaction involving the attack of hydroxide on
the conjugate acid DPP (Scheme 5, Nu = HO−), which is
In the case of DPP−, intramolecular nucleophilic catalysis
would involve a 4-membered ring intermediate (Scheme 6),
Scheme 6. Mechanism (ii): Intramolecular Nucleophilic
a
Catalysis of the Hydrolysis of DPP by Pyridine Nitrogen
Scheme 5. Strong Nucleophiles Such As Hydroxide
(Mechanism (i)) Could React Faster, Even at High pH, with
Very Small Amounts of DPP present, than with the
Predominant DPP Anion
a
The reactive intermediate formed would be converted rapidly to
product by the attack of water on the central phosphorus atom.
and a low EM would be expected: so the estimate of 14 M is
inconclusive. However, a similar EM would be expected for the
reaction of DPP , with identical geometry for intramolecular
nucleophilic attack. An estimated EM for this system, with a
much better leaving group but more weakly basic nucleophile,
is a barely significant 0.05 M (see the Supporting Information),
which makes this mechanism a good deal less likely also for the
DPP− reaction. Kinetic evidence consistent, though not
uniquely consistent, with intramolecular nucleophilic catalysis
is the low solvent deuterium isotope effect kH2O/kD2O = 1.28
0.22 at 25 °C (Table 3). However, this figure is not
conclusively smaller than the solvent deuterium isotope effect
of 1.55 observed at 39 °C for the presumed SN2(P) hydrolysis
of bis-2,4-dinitrophenyl phosphate.7 (For further discussion of
the solvent kinetic isotope effect, see below.) The entropy of
activation, of ΔS⧧ = −17.8 cal mol−1 K−1, is also inconsistent
with a unimolecular transition state, though higher than
expected for a spontaneous SN2(P) reaction involving solvated
water as the nucleophile.
SN2(P) Mechanisms iii. The “default” mechanism for the
spontaneous hydrolysis of a diaryl phosphate involves the
concerted displacement of the leaving group by a water
molecule, assisted by its hydrogen-bonded solvation partners
acting as general bases to disseminate the developing positive
charge into bulk solvent. There is general agreement that the
hydrolysis reactions of phosphate diesters with good leaving
groups do not involve pentacovalent phosphorane dianion
intermediates.13
kinetically equivalent to the direct attack of the water on DPP−.
(ii) Intramolecular nucleophilic catalysis by the pyridine
nitrogen of the nonleaving group. (iii) Direct attack of water
on the DPP− anion, presumably (iv) with assistance from a
pyridine nitrogen, acting as a general base.
In the context of mechanism i, it is important to notice that
the relevant pre-equilibrium (DPP− + H2O ⇌ DPP + OH−),
is controlled by a large difference in acidity between water and
DPP , consistent with a value of pK2 of about 13 and,
therefore, lies very much on the side of DPP− + H2O as a global
minimum. In order to check the viability of the reaction, we
discuss first the observable reactions of DPP with external
buffers, acting as nucleophiles (Scheme 5).
Reactions with Nucleophiles. The first indication that the
spontaneous hydrolysis of TPP could not be subject to efficient
intramolecular catalysis was our observation of buffer catalysis
in the pH-independent region.4 Buffer acids and bases are
typically stronger acids, bases and nucleophiles than neutral
water, but are not expected to compete with efficient
intramolecular catalysis.12 We looked very carefully for
indications of catalysis of the hydrolysis of DPP− by typical
buffer species (TRIS, CHES, etc.): we find none at 25, 45 or
even 75 °C (see the Experimental Section). Typically rates are
slightly reduced at higher buffer concentrations, suggesting that
these are medium effects. Medium effects could also account
for the weak rate increase observed near pH 10 at 70 °C with
increasing [carbonate], but measurable catalysis under these
1346
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
The evidence from kinetics tells us that the spontaneous
hydrolysis of DPP− is subject to moderately efficient catalysis,
presumably involving neighboring pyridine nitrogen: and
suggests strongly that the reaction involves water as the
immediate nucleophile involved in nucleophilic attack on the
DPP anion. Since the observed intermolecular reactions of
pyridines with reactive diesters are primarily nucleophilic, the
rate constants define only upper limits for general base
catalysis: so that the EM of the order of 14 M estimated
above for nucleophilic catalysis would represent a lower limit
(and at the same time a relatively high value12) for IGBC.
However, the kinetic evidence does not offer unambiguous
support for one particular mechanism. So we have performed
calculations for all the candidate mechanisms for the
spontaneous hydrolysis of DPP−: which do allow an informed
choice. As a basis for comparison we performed a parallel
calculation for the reaction of the simple diaryl ester bis-4-
chlorophenyl phosphate (4ClPP−), derived from a phenol with
pKa (9.38) close to that of 2-hydroxypyridine.
Table 5. Activation Free Energies and Rate Constants for the
Hydrolysis of DPP− at 25 °C Calculated at the B3LYP/6-31+
+G(d,p) Level of Theory
mechanism
1010kcalc (s−1
)
ΔG⧧ (kcal/mol)
mechanism i (1H2O)
mechanism i (3H2O)
mechanism ii (0H2O)
mechanism ii (1H2O)
mechanism ii (2H2O)
mechanism iiia (1H2O)
mechanism iiia (2H2O)
mechanism iiia (3H2O)
mechanism iiia (4H2O)
mechanism iiia (5H2O)
mechanism iiib (1H2O)
mechanism iv (1H2O)
mechanism iv (2H2O)
mechanism iv (3H2O)
observed
4.16 × 10−7
2.39 × 10−5
1.98 × 10−3
2.54 × 10−4
7.20 × 10−5
4.84 × 10−7
5.24 × 10−4
2.82 × 10−3
2.42 × 10−3
4.48 × 10−3
3.62 × 10−7
1.72 × 10−2
39.8
37.4
34.8
36.0
36.8
39.7
35.6
34.1
33.3
34.3
39.9
33.5
a
a
0.84 [5630]
0.52
31.2 [32.7]
31.5
a
a
3.12 [16.000]
30.4 [31.7]
Computational Methods and Models. Density func-
tional theory (DFT) calculations for the hydrolysis reactions of
DPP− and 4ClPP− were performed at the B3LYP/6-31+
+G(d,p) level of theory using the GAUSSIAN 09 package
implemented in Linux operating systems.14 The default
parameters for convergence, viz. the Berny analytical gradient
optimization routine, were used; convergence on the density
matrix was 10−9 atomic units, the threshold value for maximum
displacement was 0.0018 Å, and the maximum force 0.00045
hartree/bohr. The energy minimum and the transition state
were identified in each case, and their structures characterized
by frequency calculations at 1 atm and both 298.15 and 373.15
K.15 Since solvent effects are important, and indeed crucial
when the solvent is a reactant, we optimized the substrate
structures using the SCRF keyword and the polarizable
continuum model (PCM) and the solvation model density
(SMD).16 The effects of adding a series of up to 5 explicit water
molecules were examined, as we have reported recently in
related systems.17,18 Transition states (TS) were obtained by
the quadratic synchronous transit (QST) protocol, and
identified by the single imaginary frequency. Intrinsic reaction
coordinates (IRC) were computed to confirm the reaction
paths.
a
Values in square brackets correspond to kcalc and ΔG⧧ calculated at
100 °C.
Scheme 7. Mechanism iiia: SN2(P) Mechanism for
Hydrolysis Using a Single Water Molecule
Scheme 8. Mechanism iiib: Two-Step Mechanism via a
Pentaoxyphosphorane Intermediate Using a Single Water
Molecule
Thermodynamic Parameters. We calculated potential
energy surfaces (PES) for the hydrolysis mechanisms of
interest. Stationary points are located to characterize reactants
(R), transition states (TS), intermediates, and products and
thus to obtain activation parameters (Table 5). Five possible
mechanisms were considered. The calculation for mechanism i,
the kinetically equivalent reaction of hydroxide with DPP
(Scheme 5) gave one of the highest free energies of activation,
confirming the tentative conclusion reached above from the
kinetic results, that this mechanism could be excluded. As
Scheme 9. Mechanism iv: SN2(P) Mechanism for Hydrolysis
Using a Single Water Molecule with General Base Catalysis
by a Nitrogen Atom of the Non-Leaving Group
previously discussed, the relevant pre-equilibrium (DPP−
+
H2O ⇌ DPP + OH−), is consistent with pK2 ∼13 and with
DPP− + H2O as a global minimum. In this sense, the energy
barrier was determined with respect to this global minimum,
resulting in a free energy of activation of 37.4 kcal/mol, which
is consistent with a very slow reaction.
Mechanisms ii, iiia, iiib, and iv (Schemes 7−9) were
considered in more detail. Mechanism ii is the two-step
process involving intramolecular nucleophilic attack by a
pyridine nitrogen via a 4-membered cyclic transition state, as
described in Scheme 6. The reactive intermediate formed
would be converted rapidly to product by the attack of water on
the central phosphorus atom, so this step not involved in the
1347
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
calculation). In the absence of an explicit water molecule,
mechanism ii naturally shows the lowest activation energy
(Table 5); but the calculated free energy of activation increases
as water molecules are added to the calculation, and it
immediately becomes the least favored of the remaining
pathways. Since the reaction is carried out in water this
mechanism can also be discarded with some confidence.
Mechanism iii involves attack by water oxygen on
phosphorus following either (a) a classical concerted SN2(P)
mechanism or (b) a two-step displacement of the leaving
group, proceeding with formation of a pentacovalent addition
intermediate, similar to that described for the hydrolysis
reactions of phosphate triesters.17,18 Finally, mechanism iv
involves attack by a water oxygen atom on phosphorus assisted
by general base catalysis by the nitrogen atom of the nonleaving
group. Table 5 compares initial calculated and experimentally
observed parameters for all the mechanisms considered.
For the SN2(P)-based mechanisms iii, the addition of discrete
waters to the calculation leads to a decrease in the free energy
barrier, which for the concerted mechanism iiia levels out at 3−
5 H2O to give a value of ΔG⧧ 3−4 kcal·mol−1 above that
observed experimentally. (A parallel calculation, described
below, for the comparable diaryl ester bis-4-chlorophenyl
phosphate (4ClPP‑), gave a calculated rate close to that
expected for the 4-chlorophenoxide leaving group.)
The calculation for the two-step mechanism (iiib), involving
a single H2O and the formation of a pentaoxyphosphorane
addition intermediate, found a transition state with a free
energy of activation similar to that found for the concerted
SN2(P) mechanism. However, all attempts to optimize the
structure of the pentacoordinate intermediate led to cleavage of
the P−OLG bond and reversion to the bimolecular TS. The P−
OLG bond was longer in the hypothetical intermediate (1.824
Å) than in the transition state (1.690 Å), indicating that the
phosphorane is not an energy minimum. Adding two and three
water molecules to the calculation did not affect this result.
Thus, the two-step mechanism (iiib) is not supported by the
calculations. Similar results for phosphate diester hydrolysis
have been widely discussed in the literature.19a,20 Notice that it
is possible to find pentacoordinate intermediates for phosphate
diester hydrolysis under strongly acidic conditions,19a where
these compounds are present in a neutral form, and so can
behave like triesters.17,18 Similarly, in the acid-catalyzed and
spontaneous hydrolysis of dineopentyl phosphate, a compound
with a rather poor leaving group, a stepwise addition−
elimination mechanism was consistent with the theoretical
calculations.19b
Figure 3. IRC profile for the hydrolysis reaction of DPP.
hydrolysis of bis-4-chlorophenyl phosphate 4ClPP in the
presence of two to four discrete water molecules at 100 °C.
The calculated free energy of activation is some 6 kcal·mol−1
higher than observed for the hydrolysis of DPP−, giving a
calculated rate close to that expected for 4ClPP− at 100 °C
(Table 6).
Table 6. Activation Free Energies and Rate Constants for the
Hydrolysis of 4ClPP by Mechanism iiia with Two, Three,
and Four Discrete Water Molecules Present: Calculated at
100°C at the B3LYP/6-31++G(d,p) Level of Theory
no. of water molecules
1010 kcalc (s−1
)
ΔG⧧ (kcal/mol)
mechanism iiia (2 H2O)
mechanism iiia (3 H2O)
mechanism iiia (4 H2O)
experimental (extrapolation)7
1.84 × 10−7
15.5
40.3
37.1
37.6
37.9
7.71
4.93
To allow a more detailed comparison of the reactions of
DPP− and 4ClPP we also constructed the IRC (Figure 4) for
For mechanism iv, involving intramolecular general base
catalysis by the nitrogen center of the nonleaving group, the
calculated free energy of activation is in good agreement with
the experimental value when two or three discrete water
molecules are present. This mechanism accounts satisfactorily
for the rate enhancement observed for the hydrolysis of DPP−
when compared to symmetric phosphodiesters with compara-
ble leaving groups7 and so is examined in more detail.
Figure 4. IRC profile for the hydrolysis reaction of 4ClPP.
The transition state (TS) for the hydrolysis reaction by
mechanism iv was verified by IRC calculations. The IRC profile
(Figure 3) is consistent with a single step, concerted reaction,
with two water molecules playing well-defined roles in TS1.
(Cartesian coordinates for reactant, TS and product are given
in the Supporting Information.)
the hydrolysis reaction of the latter, simple diaryl phosphate in
the presence of three waters. The apparently high energy of the
product with respect to reactant in Figure 4 is due to the fact
that the proton transfer from hydronium (H3O+) to 4-
chlorophenolate anion is not included in this step. The IRC
confirms the concerted single-step pathway, with the three
water molecules playing well-defined roles. In particular, the
To confirm the rate enhancement for DPP−, we performed
the same calculation (for the SN2(P) mechanism iiia) for the
1348
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
Figure 5. Optimized structures for the transition states TS for the hydrolysis reactions of DPP and 4ClPP. For a key to the atom numbering, see
Scheme 10. The roles of the water molecules are discussed in detail in the following section.
water molecule closest (and hydrogen-bonded) to the
nucleophilic H2O is acting as the immediate general base
(Figure 5 and Scheme 10).
state TS with a close to linear arrangement with the oxygen
atom of the leaving group, as expected for a SN2(P)
mechanism. The calculated angles O2···P1···O7 are 171.6° and
172.57° for DPP and 4ClPP, respectively (Table 7), with
imaginary frequencies of 666.24i and 190.95i corresponding to
the rocking vibration leading to product formation. The
displacement vector (DV) associated with the imaginary
frequency is consistent in each case with a transition state in
which P1−O2 formation and P1−O7 cleavage are concerted.
The vectors are illustrated in Figure 5 to show the physical
nature of the effect. The imaginary frequencies, the temperature
of 100 °C, and the barriers including the zero-point energies
(ZPE) for reactants, transition state, and product allow an
estimate of the tunneling contribution for this reaction using
the Truhlar derivation21,22 and the Bell treatment.23 The values
obtained, 1.33 for DPP and 1.02 for 4ClPP, indicate that any
tunneling contribution to this mechanism is negligible.
Scheme 10. Transition States for the Hydrolysis of DPP and
4ClPP Detailing the Atom-Numbering Scheme
Bond Order Analysis. To compare bond order evolution
in the transition state with respect to reactant, we performed a
natural bond orbital (NBO) calculation as implemented in
In each case, the hydrolysis reaction is initiated by the
approach of an appropriately positioned solvent water molecule
to the phosphorus center, to form a pentacoordinate transition
Table 7. Structural Parameters for Reactant (R), Transition State (TS), and Product (P) at the B3LYP/6-31++G(d,p) Level of
a
Theory
Interatomic Distances (Å)
DPP
P1−O2
O2−H3
H3−N4
N4−C5
C5−O8
P1−O7
P1−O8
P1−O6, P1−O9
N10−H11
R
3.55
1.98
1.64
0.98
1.31
1.91
1.95
1.18
1.03
1.33
1.34
1.35
1.38
1.34
1.33
1.66
1.85
2.98
1.67
1.74
1.73
1.51, 1.51
1.52, 1.52
1.50, 1.50
1.85
1.80
1.82
TS1
P
4-ClPP
P1−O2
O2−H3
H3−O4
O4−H5
H5−O6
P1−O7
P1−O8
P1−O6, P1−O9
O6−H11
R
4.51
1.93
1.68
0.98
1.02
1.44
1.80
1.56
1.06
0.99
0.97
0.98
1.76
3.08
3.23
1.65
2.41
5.39
1.65
1.63
1.65
1.52, 1.50
1.50, 1.50
1.51, 1.50
4-ClPP
1.82
1.82
1.80
TS1
P
DPP
angle O2−P1−O7 of TS (deg)
imaginary frequency (cm−1
171.6
172.6
)
666.24i
190.95i
a
The atom numbering is defined in Scheme 10.
1349
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
Table 8. Wiberg Bond Indices for Reactant (R), Activated Complex (TS), and Product (P) at the B3LYP/6-31++G(d,p) Level
of Theory
P1−O2
O2−H3
H3−N4
N4−C5
C5−O8
P1−O7
P1− O8
Sy
DPP
BiR
BTi S
BjP
0.0032
0.3613
0.6651
54.1
0.6726
0.2684
0.0379
63.7
0.0553
0.4498
0.6766
63.5
1.3974
1.2893
1.2392
68.3
0.9537
1.0603
1.1003
72.7
0.5988
0.4256
0.0169
29.8
0.5930
0.5140
0.5030
87.8
0.889
%Ev
BiR
BTi S
BjP
4ClPP
0.0004
0.3433
0.5906
58.1
0.6590
0.5341
0.1710
25.6
0.0646
0.1367
0.4825
17.3
0.6177
0.1420
0.0002
77.0
0.654
%Ev
Gaussian 09.24−27 Wiberg bond indices were estimated from
population analysis. Bond-breaking and -making processes
involved in the rate limiting step were examined by means of
the synchronicity (Sy) concept proposed by Moyano et al.28
The synchronicity parameter offers a measure of the concerted-
ness of a reaction, varying from zero, in the case of an
asynchronous process, to unity, for a perfectly synchronous
process. The results are shown in Tables 8 and S2 (Supporting
Information). Wiberg bond indices Bi were calculated for all the
bonds involved in the hydrolysis reactions of DPP and 4ClPP,
as indicated in Scheme 10. Bonds undergoing negligible
changes were not considered.
Transition State and Mechanism. The similar intera-
tomic distances P1−O2 of 1.98 Å and 1.93 Å in the transition
states for the reactions of DPP and 4ClPP, respectively, indicate
comparable amounts of bond formation to the incoming
nucleophile: but the P1−O7 distances to the leaving group, at
1.85 and 2.41 Å (Table 7), differ substantially. Evidently the
transition state for the phosphate transfer process is reached
earlier for the more reactive DPP, consistent with the
involvement of the pyridine nitrogen, which is a more effective
because it is stronger general base compared with a second
water molecule. Thus, proton transfer from the nucleophilic
water to nitrogen is further advanced (O2−H3 increases from
0.98 to 1.31 Å in the transition state for the DPP reaction,
compared with 1.02 Å for 4ClPP), as the developing positive
charge is stabilized more effectively on nitrogen than by solvent
water. The more effective general base catalysis is reflected also
in the Wiberg bond indices (Table 8), which indicate that
proton transfer to nitrogen in the TS for the DPP reaction is ca.
65% complete, while the proton transfers involved in the
dispersal of the developing positive charge on the nucleophilic
water clearly lag behind for the 4ClPP reaction.
The estimated bond orders for bond-making and -breaking in
the transition state for the reaction of DPP involving IGBC
indicate similar degrees of change (“evolution”) for all but one
of the bonds directly involved: giving a synchronicity parameter
close to unity (Sy = 0.889). Most advanced are the changes
more closely involved in general base catalysis, with proton
transfer from O2 to N4 well advanced and the formation of the
new P1−O2 bond over 50% complete. The bond P1−O8,
nominally to the nonleaving group, actually shows the highest
degree of evolution in the TS (87.8%, Table 8): because it is
affected both directly by the decreasing electrophilicity of
phosphorus and by the development of positive charge at N.
Only the cleavage of the P1−O7 bond to the leaving group is
not well developed (29.8%) at the point where the transition
state is reached for this very slow reaction.
(%Ev = 58.1) (Table 8). Here the (even slower) reaction is
driven primarily by the (poor) leaving group. P−O bond
making and breaking are more closely synchronous, but the
proton transfers involving solvent water molecules necessary for
general base catalysis lag behind. These differences are the
likely basis for the low solvent deuterium isotope effect, kH2O
/
kD2O = 1.28 0.22 at 25 °C, observed for the hydrolysis of
DPP. It is recognized that the magnitudes of such effects can be
reduced when there is a large difference in pKa between the
proton donor and acceptor, resulting in a very early or late
transition state.29 In the present case both requirements are
satisfied: the (water) donor and pyridine acceptor pKa s are
15.7 and 2.73, respectively; and the proton transfer of H3 from
O2 to N4 from is well over half complete in the transition state,
as indicated by both bond length and bond order calculations
(Tables 7 and 8, above).
CONCLUSIONS
■
The combined kinetic and calculational evidence leads to the
firm conclusion that the spontaneous hydrolysis of DPP−
involves intramolecular general base catalysis by pyridine
nitrogen. The substantial rate increase over expectation for a
symmetrical diester with leaving groups of pKa 9.09, ca. 3000-
fold at 100 °C and at least 10000-fold at 25 °C, corresponds to
an EM of at least 15 M at 25 °C, of the order expected for
relatively efficient IGBC.12
The rate acceleration is fully accounted for by a favorable
entropy of activation, consistent with the involvement of an
intramolecular nitrogen as the general base. The surprisingly
low solvent deuterium isotope effect, kH2O/kD2O = 1.28 0.22
at 25°, finds close precedent in the hydrolysis of the ester 4-
nitrophenyl 2-dimethylaminopropionate (Scheme 11), another
system demonstrating IGBC because the structure rules out
intramolecular nucleophilic catalysis; for which kH2O/kD2O = 1.4
at 39 °C.30
Scheme 11. IGBC by Tertiary Nitrogen Is Observed When
the Nucleophilic Mechanism Is Ruled out by Strain in a
Potential 4-Membered Ring Intermediate30
For 4ClPP, in contrast, P1−O7 cleavage is the most advanced
coordinate (%Ev = 77.0), followed by P1−O2 bond formation
1350
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
DPP was determined by hydrolyzing an aliquot of the reaction
solution to completion in HCl and measuring the absorbance at the
same pH in the same buffer solution as the original reaction.
EXPERIMENTAL SECTION
■
General Methods. Organic solvents were carefully dried using
molecular sieves (type 3A or 4A) and synthetic reactions carried out in
strictly anhydrous conditions under argon. Chemicals and inorganic
salts were of the highest purity available and were used as purchased.
1H and 31P NMR spectra were recorded on a spectrometer (400 and
200 MHz) with sodium 3-(trimethylsilyl) propionate (TSP) as
At 25 °C, the DPP anion has a half-life of over 70 years, and more
concentrated solutions cannot be used for initial rate measurements
because the absorbance of the substrate (λmax = 270 nm) overlaps with
that of the released 2-pyridone. Thus, the appearance of 2-pyridone at
294 nm produces only very small changes over very long periods
(typically 0.01 absorbance units over 2 days), placing this reaction on
the borderline of what is measurable at 25 °C. Numerous
measurements over 2 years, in our two different laboratories, in
different hemispheres, produced constants in the k0 (pH 9−10) region
which we considered to be accurate to no more than an order of
magnitude: these have been discarded. We are confident that the data
cited in Table 1 are the best that can be obtained under these
circumstances: and their validity is supported by the excellent
agreement with data collected at higher temperatures (Table 2), as
defined by an Eyring plot with r > 0.999.
1
internal reference for H NMR and 85% phosphoric acid as external
reference for 31P NMR spectra. Chemical shifts are reported as δ
(ppm). The time-of-flight (TOF) mass analyzer was used for HRMS
measurements.
General Procedure for the Synthesis of Phosphate Diesters
2,4-DNPO-(PO2−)-OAr. A solution of 2,4-dinitrophenol (10 mmol)
and triethylamine (10 mmol) in 20 mL of dry CHCl3 was added
dropwise to a cold solution of the appropriate phosphorochloridate
(10 mmol, 4-nitrophenyl or phenyl−OP(O)Cl2) in 30 mL of dry
CHCl3, and the reaction mixture was stirred overnight at room
temperature. Unreacted chloride was hydrolyzed by adding triethyl-
amine and water. This solution was extracted twice with CHCl3 and
the organic layer dried and evaporated. The diester was precipitated as
the cyclohexylamine salt and recrystallized to give a colorless solid.
2,4-Dinitrophenyl phenyl phosphate (2.28g, 52%): 1H NMR
(CD3CN, 200 MHz) δ (ppm) 8.68 (d, 1H), 8.45 (dd, 1H), 7.85 (d,
1H), 7.31 (m, 2H), 7.14 (m, 3H); 31P (CD3CN, 200 MHz) δ (ppm)
−13.6; ESI-MS negative-ion mode m/z calcd for C12H8N2O8P− 339.0,
found 339.1 (100) [M]−. 2,4-Dinitrophenyl 4-nitrophenyl phos-
phate (2.13 g, 44%): 31P (CD3CN, 200 MHz) δ (ppm) −14.9 ppm;
ESI-MS negative-ion mode m/z calcd for C12H7N3O10P 383.9, found
384.0 (100) [M]−.
Synthesis of Salts of Di-2-pyridyl Phosphate (DPP). A solution
of 2-pyridone (2.48 g, 26.08 mmol) and triethylamine (2.63 g, 26.08
mmol) in 50 mL of dry THF was added dropwise under nitrogen to an
ice-cooled solution of POCl3 (2 g, 13.04 mmol) in 100 mL of dry
THF. The reaction mixture was stirred for a further 15 min at 0 °C
and then for 18 h at room temperature. Triethylamine (3 mL) in 20
mL of water was added and the reaction mixture stirred for 15 min.
The THF was evaporated in vacuo and the remaining aqueous
solution diluted with water then washed with CH2Cl2. After
evaporationof the water, the crude product was dissolved in acetone
and the triethylammonium chloride removed by filtration and
evaporated to dryness. The product triethylammonium salt was
dissolved in water and passed through an ion-exchange column (IR-
120 resin, Na form) to give 2.43 g of DPP (Na salt, 68%): δH (250
MHz, D2O) 8.10 (2H, dd, J = 5.33, 1.9 Hz), 7.74−7.81 (2H, m),
7.14−7.17 (2H, m), 7.10 (2H, d, J = 8.35 Hz); δP (250 MHz, D2O)
−11.90; ESI-MS positive-ion mode m/z calcd for C10H8N2O4Na2P
297.0017, found 297.0018 (100) [M + Na]+.
The k0 data at higher temperatures complement and support the
results at 25 °C. pHs of buffer solutions were generally measured at 25
°C and the same conditions used at higher temperatures. Detailed
measurements for the runs at 75 °C (Table 9) showed that the pH
Table 9. DPP Hydrolysis in CHES Buffer at 25 and 75 °C, I
a
= 1.0 M
[CHES] (mM)
pH
T (°C)
kobs (s−1
)
50
200
50
9.4
9.4
7.8
8.2
8.6
7.8
8.2
8.6
25
25
75
75
75
75
75
75
2.0 0.2 × 10−10
1.4 0.2 × 10−10
3.2 0.03 × 10−7
1.6 0.02 × 10−7
1.1 0.01 × 10−7
2.6 0.02 × 10−7
1.5 0.01 × 10−7
1.1 0.01 × 10−7
50
50
200
200
200
a
Note: [DPP] = 15 mM. The total change in the absorbance at 294
nm is calculated for the formation of 2 equiv of 2-pyridone.
(and thus the apparent pKa of the buffer) had shifted downward by
some 1.2 units, but so too had the minimum in the pH profile for the
reaction (Figure 6). Much of this shift can be attributed to the
Kinetic Methods. Reactions were initiated by adding 10 or 30 μL
of a stock solution of the substrate DPP (5 × 10−3 M in water) to 1 or
3 mL of buffered aqueous solution at the appropriate pH. Buffers used
were HCl (pH < 2), chloroacetate (pH 2−3.5), acetate (pH 4−5.5),
Bis-TRIS (pH 6−7), TRIS (pH 8−9), CHES (pH 8.7 − 10.4),
carbonate (pH ≈ 10), and KOH (pH >10). The reactions were
followed, at constant temperature and ionic strength 1.0 (kept
constant with KCl or NaCl, without significant difference in rate
constants), by monitoring the appearance of pyridone at 294 nm on a
spectrophotometer equipped with a thermostatted cell holder.
Reactions at low pH (<3.5) and at pH > 12 were monitored to
completion. At intermediate pH’s, rate constants for the hydrolysis of
DPP were determined by the initial rate method, using more
concentrated solutions (5 to 15 mM) of substrate. In a typical run,
100 μL of 150 mM aqueous solution were added to 900 μL of a buffer
solution equilibrated to the desired temperature. The absorbance of
the released product was monitored by UV−vis spectroscopy at 294
nm. The observed rate constants were obtained by dividing the initial
rates of change of absorbance by the final change in absorbance under
the reaction conditions (hydrolysis of DPP produces 2 equiv of 2-
pyridone in the acid-catalyzed and pH-independent regions, but 1
equiv at pH >10). The total change in the absorbance of hydrolysis of
Figure 6. pH−rate profile for the hydrolysis of DPP at 75 °C
compared with the same measured at 25 °C (from Figure 1).
1351
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
decrease in pKw (which falls from 14 at 25 °C to 12.24 at 100 °C).31
The calculated curves are naturally shallow close to the pH minimum,
where the rate is identical or very close to the pH-independent k0; e.g.
the rate difference at 75 °C between the observed minimum at about
pH 8.87 and pH 9.4 (the minimum at 25 °C) is only some 20%. Thus,
full pH−rate profiles were generally not measured at the highest
temperatures.
The data shown in Table 9, at pH’s 7.8, 8.2, and 8.6 (30, 50, and
70% free base buffer) measured at 75 °C demonstrate very clearly the
absence of buffer catalysis. Rates are actually reduced at higher buffer
concentrations, as they are also at higher pH, corresponding to the
approach to the pH minimum.
Sciences ResearchCouncil; EP/E01917X; UK); and to The
Libyan Ministry of Higher Education (to A.A.) for support of
this work.
REFERENCES
■
(1) Kirby, A. J.; Medeiros, M.; Oliveira, P. S. M.; Brandao, T. A. S.;
̈
Nome, F. Chem.Eur. J. 2009, 15, 8475.
(2) Liu, J. Q.; Wulff, G. J. Am. Chem. Soc. 2008, 130, 8044.
(3) Albert, A.; Phillips, J. N. J. Chem. Soc. 1956, 1294.
(4) Kirby, A. J.; Medeiros, M.; Oliveira, P. S. M.; Orth, E. S.; Brandao,
̃
T. A. S.; Wanderlind, E. H.; Amer, A.; Williams, N. H.; Nome, F.
Chem.Eur. J. 2011, 17, 14996.
The rate constant at 100 °C is based on an independent
measurement in which the reaction was followed to completion by
HPLC, in 70% free base CHES buffer (Figure 7).
(5) Brown, R. S.; Zamkanei, M. Inorg. Chim. Acta 1985, 108, 201.
(6) Kirby, A. J.; Younas, M. J. Chem. Soc., Sect. B 1970, 1165.
(7) Kirby, A. J.; Younas, M. J Chem Soc., Sect. B 1970, 510.
(8) Orth, E. S.; Brandao, T. A. S.; Souza, B. S.; Pliego, P. R.; Vaz, B.
G.; Eberlin, M. N.; Kirby, A. J.; Nome, F. J. Am. Chem. Soc. 2010, 132,
8513.
(9) Wolfenden, R.; Ridgway, C.; Young, G. J. Am. Chem. Soc. 1998,
120, 833.
(10) Schroeder, G. K.; Lad, C.; Wyman, P.; Williams, N. H.;
Wolfenden, R. Proc. Nat. Acad. Sci. U.S.A. 2006, 103, 4052.
(11) Orth, E. S.; Brandao, T. A. S.; Milagre, H. M. S.; Eberlin, M. N.;
Nome, F. J. Am. Chem. Soc. 2008, 130, 2436.
(12) Kirby, A. J. Adv. Phys. Org. Chem. 1980, 17, 183.
(13) Lassila, J. K.; Zalatan, J. G.; Herschlag, D. Annu. Rev. Biochem.
2011, 80, 669−702.
(14) Frisch, M. J.; G. W. Trucks, H. B. S., Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(15) McQuarrie, D. Statistical Mechanics, New York, 1986.
(16) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378.
Figure 7. Spontaneous hydrolysis of DPP in CHES buffer (70% free
●
base, I = 1.0 M KCl) at 100 °C, quantified by HPLC: = pyridone
○
appearance at 224 nm, = substrate disappearance at 275 nm.
Activation Parameters for DPP Hydrolysis. Activation
parameters for the spontaneous hydrolysis of DPP were calculated
from rate constants measured in CHES buffer (0.05 M, 70% free base,
I = 1.0 M (KCl)) at 25, 45, 60, 75, and 90 °C (Table 2). At each
temperature the initial rate of the reaction was monitored using UV−
vis spectroscopy at 294 nm. At 100 °C, the reaction was followed to
completion using HPLC. The observed first order rate constants for
water attack and second order rate constants for hydroxide attack were
analyzed by Eyring plots for the two reactions (see the Supporting
Information).
(17) Kirby, A. J.; Mora, J. R.; Nome, F. Biochim. Biophys. Acta 2013,
1834, 454−463.
ASSOCIATED CONTENT
* Supporting Information
Tables and figures giving kinetic and theoretical study data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(18) Mora, J. R.; Kirby, A. J.; Nome, F. J. Org. Chem. 2012, 77, 7061.
(19) (a) Iche-Tarrat, N.; Barthelat, J. C.; Rinaldi, D.; Vigroux, A. J.
Phys. Chem. B 2005, 109, 22570. (b) Kamerlin, S. C. L.; Williams, N.
H.; Warshel, A. J. Org. Chem. 2008, 73, 6960.
■
S
(20) Zhou, D. M.; Taira, K. Chem. Rev. 1998, 98, 991.
(21) Skodje, R. T.; Truhlar, D. G. J. Phys. Chem. 1981, 85, 624.
(22) Skodje, R. T.; Truhlar, D. G.; Garrett, B. C. J. Phys. Chem. 1981,
85, 3019.
(23) Caldin, E. F. Chem. Rev. 1969, 69, 135.
(24) Lendvay, G. J. Phys. Chem. 1989, 93, 4422.
(25) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899.
(26) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985,
83, 735.
(27) Wiberg, K. B. Tetrahedron 1968, 24, 1083.
AUTHOR INFORMATION
Corresponding Author
*Tel: +55-48-3721-6849. Fax: +55-48-3721-6850. E-mail:
faruk.nome@ufsc.br, ajk1@cam.ac.uk, N.H.Williams@sheffield.
ac.uk.
■
Notes
The authors declare no competing financial interest.
(28) Moyano, A.; Pericas, M. A.; Valenti, E. J. Org. Chem. 1989, 54,
573.
(29) Grzyska, P. K.; Czyryca, P. G.; Purcell, J.; Hengge, A. C. J. Am.
Chem. Soc. 2003, 125, 13106.
ACKNOWLEDGMENTS
We are grateful to INCT-Catal
and CAPES in Brazil; to EPSRC (Engineering and Physical
■
́
ise, PRONEX, FAPESC, CNPq,
1352
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353

The Journal of Organic Chemistry
Featured Article
(30) Kirby, A. J.; Lloyd, G. J. J. Chem. Soc., Perkin Trans. 2 1976,
1748.
(31) Whitfield, M. J. Chem. Eng. Data 1972, 17, 124.
1353
dx.doi.org/10.1021/jo302498g | J. Org. Chem. 2013, 78, 1343−1353