Experimental and computational determination of Bronsted coefficients for equilibrium transfer of the O,O-dimethyl phosphorothioyl group between oxyanion nucleophiles

Article abstract of DOI:10.1002/poc.1903

An experimental determination of the β^Eq value for equilibrium transfer of the O,O-dimethyl phosphorothioyl group between oxyanion nucleophiles in water and methanol at 25 °C is presented. The respective β^Eq values in the two solvents are experimentally the same at -1.45 ± 0.08 and -1.39 ± 0.12. Based on the observation that the Bronsted correlation for the nucleophilic reaction of phenoxides in water with substrate 1d (dimethyl 4-nitrophenyl phosphorothioate, pK_a ^HOAr of 7.14) is linear over the entire range of phenoxides employed (5.53 ≤ pK_a^Nu ≤12.38), the reaction for phenoxide nucleophiles displacing phenoxide leaving groups is probably concerted. The obtained data allow one to calculate, for a symmetrical transition state involving 2,4,5-trichlorophenoxide as a nucleophile and leaving group, an approximately 60% P-OAr cleavage and about 40% P-Nuc bond formation. A computational method is presented for the rapid prediction of the β^Eq values for such processes in water and methanol, and the results are compared with known values from the literature. Copyright

Full text of DOI:10.1002/poc.1903

Research Article
Received: 21 April 2011,
Revised: 24 May 2011,
Accepted: 27 May 2011,
Published online in Wiley Online Library: 31 July 2011
(wileyonlinelibrary.com) DOI 10.1002/poc.1903
Experimental and computational
determination of Brønsted coefficients for
equilibrium transfer of the O,O-dimethyl
phosphorothioyl group between
oxyanion nucleophiles
David R. Edwardsa, Christopher I. Maxwella, Robert W. Harknessa,
Alexei A. Neverova, Nicholas J. Moseya* and R. Stan Browna*
An experimental determination of the βEq value for equilibrium transfer of the O,O-dimethyl phosphorothioyl group
between oxyanion nucleophiles in water and methanol at 25ꢀC is presented. The respective βEq values in the two
solvents are experimentally the same at ꢁ1.45ꢂ0.08 and ꢁ1.39ꢂ0.12. Based on the observation that the Brønsted
correlation for the nucleophilic reaction of phenoxides in water with substrate 1d (dimethyl 4-nitrophenyl phosphor-
othioate, pKHa OAr of 7.14) is linear over the entire range of phenoxides employed (5.53≤pKNa u≤12.38), the reaction for
phenoxide nucleophiles displacing phenoxide leaving groups is probably concerted. The obtained data allow one to
calculate, for a symmetrical transition state involving 2,4,5-trichlorophenoxide as a nucleophile and leaving group,
an approximately 60% P–OAr cleavage and about 40% P–Nuc bond formation. A computational method is presented
for the rapid prediction of the βEq values for such processes in water and methanol, and the results are compared
with known values from the literature. Copyright © 2011 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: Brønsted; hydrolysis; methanolysis; phosphorthioate esters; solvolysis
reaction versus the respective pKHa OAr values of the conjugate
acids of the leaving group (Lg) or the attacking nucleophiles
INTRODUCTION
(Nu), defining the gradients of such plots as the Brønsted coeffi-
Many organophosphorothioate triesters containing a phospho-
rus sulfur double bond (P=S) are commercially important insecti-
cides and acaricides that inhibit acetyl cholinesterase.[1,2] These
materials are less prone to hydrolytic degradation than their
P=O counterparts, and at the same time have less toxicity toward
mammals. Methods have been developed to degrade these en-
vironmentally persistant materials, including conventional hy-
drolysis and reaction with activated alpha-nucleophiles[3,4] and
organopalladium/organoplatinum metal complexes in water.[5–9]
We previously reported that certain Cu(II) and Zn(II) complexes
as well as a palladacycle, [Pd(II)(2-(N,N-dimethylaminobenzyla-
mine)-C1,N)(pyridine)] triflate, promote the decomposition of
neutral O,O-dimethyl O-aryl phosphorothioates in methanol
where catalytic efficiencies of up to about 109 relative to the
backgound reaction are realized.[10–13]
cients, βLg or βNu [14–16] The Leffler parameter, defined as βLg/βEq
.
or βNu/βEq, normalizes the reaction constant to measure the ex-
tent of progression of bond cleavage or formation at the transi-
tion state for the reaction, where the coefficient βEq defines the
total charge change felt by the reporter group in going from re-
actant to product.
S
X = a) 2,4-dinitro
MeO
P
O
b) 2-chloro-4-nitro
c) 2-nitro-4-chloro
d) 4-nitro
X
OMe
1a-e
e) 3-nitro
The neutral phosphorothioates 1a–e respresent a closely re-
lated, contiguous series of model compounds for the general
class of P=S pesticides that we have used[13] for kinetic and
mechanistic studies because of their easily visualized chromo-
phores and the graded leaving group propensity of the O-aryl
groups, as reflected in their acid dissociation constants KHa OAr. De-
tailed study of the kinetics of the O-aryl cleavage of this series
allows the construction of linear free energy relationships that
probe substituent effects on reaction rate. In such studies, one
typically plots the log of the second-order rate constant for the
* Correspondence to: R. Stan Brown and N. J. Mosey, Department of Chemistry,
Queen’s University Kingston, Ontario K7L 3N6, Canada.
E-mail: rsbrown@chem.queensu.ca; nicholas.mosey@chem.queensu.ca
a D. R. Edwards, C. I. Maxwell, R. W. Harkness, A. A. Neverov, N. J. Mosey,
R. Stan Brown
Department of Chemistry, Queen’s University, Kingston, Ontario K7L 3N6,
Canada
J. Phys. Org. Chem. 2012, 25 258–266
Copyright © 2011 John Wiley & Sons, Ltd.

EXPERIMENTAL AND COMPUTATIONAL DETERMINATION OF BRØNSTED COEFFICIENTS
In the simple case of the ionization of phenols via acid disso-
ciation, the total charge change on the reporter oxygen is de-
fined as ꢁ1 (Scheme 1), and for the cleavage of phosphate
triesters, which are closely related to the phosphorothioates 1
of interest here, the effective charge on the aryloxygen in the
substrate is +0.87 and ꢁ1 in the aryloxide, so that βEq is
ꢁ1.87.[15,16] In the direction written in Scheme 1, the change in
the effective charge at the transition state ({) is βF so that the
is that it reduces the effort required for the determination of
βEq values and should present an attractive way of obtaining
their initial estimates for other acyl and phosphoryl transfer
reactions.
EXPERIMENTAL
Materials
Anhydrous tetrahydrofuran (THF) and methanol were obtained
from EMD Chemicals. Water was passed through a Millipore Sim-
plicity (Millipore Scientific, Billerica, MA, USA) water purification sys-
tem. Phenol (99%+), 4-chlorophenol (99%+), 2,4,5-trichlorophenol
(98%), 2,3,5,6-tetrafluorophenol, 2,2,2-trifluoroethanol (99%+), tetra-
butylammonium chloride (>97%), sodium chloride (98%) and so-
dium hydroxide (50% solution in water) were obtained from Sigma
Aldrich (Sigma-Aldrich Canada, Oakville, ON, Canada) and used as
supplied. The O,O-dimethyl O-aryl phosphorothioates used were
available from a previous study with the exception of O,
O-dimethyl O-2,4-dinitrophenyl phosphorothioate, which was
prepared and characterized as described below.
+
K
a
H
+ -OAr
-1.0
OAr
0
H
K
O
P
-
O
P
+ -Nu
OAr
0.87
RO
RO
+ -OAr
-1.0
OR
OR
Nu
F
β
Eq
β
Synthesis of O,O-dimethyl O-(2,4-dinitrophenyl)
phosphorothioate (1a)
Scheme 1. Description of charge changes occuring for the ionization of
phenols, and the equilibium transfer of the dialkylphosphoryl group
between oxyanion nucleophiles where log (K)=βEq•pKa+C
Into a 50-mL round-bottomed flask equipped with a magnetic
stirring bar was added 25mL of anhydrous THF (EMD Drisolv)
(EMD Chemicals, Gibbstown, NJ, USA), 2,4-dinitrophenol (990
mg, 5.4mM, Aldrich, 97%) (Sigma-Aldrich Canada, Oakville, ON,
Canada) and sodium hydride (560mg, 14mM, Aldrich, 60%
dispersion in mineral oil). The reaction mixture was stirred at
ambient temperature for five minutes, after which was added
dimethyl chlorothiophosphate (1.58g, 10mM, Aldrich, 97%) and
the suspension stirred at ambient temperature for 16h. The sol-
vent was removed under reduced pressure and the resulting ma-
terial taken up in diethyl ether, which was washed three times
with saturated aqueous potassium carbonate and once with
brine. The organic fraction was evaporated to dryness and sub-
jected to flash column chromatography, eluting with hexanes:
ethyl acetate (50:50) to yield the title compound.
effective charge on the aryloxygen at that state is 0.87 – βF.
Experimental determination of βEq for cases such as those in
Scheme 1 is tedious because it requires many experimental
kinetic data to obtain the Brønsted coefficients for nucleophilic
attack (βNu) and for leaving group departure (βLg). Provided
these are determined for a reaction where the rate limiting step
for the forward and reverse process is the same, the βEq is then
calculated as (βLg – βNu). This represents the total charge change
felt by the aryloxy or alkoxy reporter group (Nuc+(RO)2P(=O)–
OAr ⇆ Nuc–P(=O)(OR)2+–OAr) during a hypothetical symmet-
rical reaction where the attacking nucleophile and departing
leaving group have identical pKa values for their conjugate acids.
In our past studies of the metal ion-catalyzed alcoholysis reac-
tions of phosphates, phosphonates, phosphorothioates, phos-
phonothioates and carboxylate esters in methanol[10–12,17–22]
we have used the literature βEq values determined in water un-
der the critical assumption that these are the same in methanol
(or in fact other solvents) as long as one accounts for the pKa
differences for the nucleophile or leaving group in water versus
the nonaqueous solvent. This is a critical assumption because it
might be argued that the solvation differences between alcohol
and water might seriously change the various β values for nucle-
ophilic attack or leaving group departure thereby rendering
conclusions made about charges in transition states unreliable.
Herein we report the experimental determination of the βEq
values for the transfer of the O,O-dimethyl phosphorothioyl
group between oxyanion nucleophiles in water and methanol.
These values, now determined to be ꢁ1.45 and ꢁ1.39 respec-
tively, are considered to be the same within experimental uncer-
tainty, validating the assumption, at least in the case of
substrates 1. In addition, we present a straightforward density
functional theory computational model devised for the calculation
of the βEq that we compare to the present experimental βEq
results. This is also applied to compare with other βEq values that
appear in the literature for some acyl and phosphoryl transfer
processes. The main virtue of this computational methodology
1H NMR (600MHz, CDCl3) d 8.81–8.85 (1H, m, ArH), d 8.43–8.48
(1H, m, ArH), d 7.63–7.68 (1H, m, ArH), d 3.93 (6H, d, P-OCH3, J=
13.97Hz). 31P NMR (243.06MHz, CDCl3) d 65.32 referenced to
70% phosphoric acid. Calculated mass for C8H10O7N2PS [M+H]+
is 308.9946m/z; found HRMS(ESI+): 308.9940. Melting point
92–94ꢀC. UV–Vis (MeOH, 25ꢀC) e250 nm=12,053M–1 cm–1.
Methods
The CH3OH+2 concentrations were determined potentiometrically
using a combination glass electrode (Radiometer model XC100-
111-120-161) Fisher Scientific, Pittsburgh, PA, USA calibrated
with certified standard aqueous buffers (pH=4.00 and 10.00) as
previously described.[23] The sspK values in methanol were
obtained by subtracting a correction constant of ꢁ2.24[24] from
the electrode readings. The autoprotolysis constant for methanol
was taken to be 10–16.77 (M)2. The sspKHa OAr values of different sub-
stituted phenols in methanol can be found in a previous
report.[25]
Pseudo-first-order observed rate constants (kobs) were deter-
mined by UV–Vis spectrophotometry observing the rate of phe-
nol product formation. For slow reactions, the method of initial
rates was used and reactions were monitored for <5% in each
J. Phys. Org. Chem. 2012, 25 258–266
Copyright © 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

D. R. EDWARDS ET AL.
case. The rate of reaction, defined in units of absorbance units
per second, was determined by a zero order fit of the data. This
was then used to calculate the first order rate constant by divid-
ing the rate of reaction by the expected total ΔAbs. Extinction
coefficients were determined by plots of ΔAbs versus [HOAr
leaving group] under the experimental conditions. For faster
reactions, full time courses were determined and the kobs values
calculated from a fit of the absorbance versus time data to a
standard first-order exponential equation. Full reaction profiles
of absorbance versus time were obtained for the reaction of
2,2,2-trifluoroethoxide with 1a, 1c and 1d and the attacks of
phenoxide and 3,4-dichlorophenoxide on O,O-dimethyl O-2,
4-dinitrophenyl phosphorothioate. The second-order rate con-
stants for the cleavage of O,O-dimethyl O-aryl phosphorothio-
ates were determined as the gradients of the plots of kobs
versus [ArO–] or [CF3CH2O–] and appear in the Supporting Infor-
mation as Tables S1–S3.
-0.5
-1.5
-2.5
-3.5
-4.5
-5.5
-6.5
-7.5
5
6
7
8
9
10
11
12
13
pKNa u
Computational details
Figure 1. Brønsted plot of log(k2) at 25ꢀC versus pKaNu for the attack in
water of substituted phenoxides and 2,2,2-trifluoroethoxide on (■) O,
O-dimethyl O-4-nitrophenyl phosphorothioate where log(k2)=(0.61ꢂ
0.04) pKNa u – (10.06ꢂ0.40), r2=0.9843 (5 data); (●) O,O-dimethyl O-(2-
nitro-4-chlorophenyl) phosphorothioate where log(k2)=(0.54ꢂ0.03) pKaNu
Geometry optimizations, energy calculations and frequency
calculations of structures were performed at the Becke, three-
parameter, Lee–Yang–Parr (B3LYP)[26,27] level of theory using
GAUSSIAN 09 (Gaussian, Inc., Wallingford CT/ USA).[28] with the ap-
plication of the integral equation formalism variant of the polar-
– (8.94ꢂ0.26), r
2
=0.9951 (4 data); and (▼) O,O-dimethyl O-(2,4-dinitro-
phenyl) phosphorothioate where log(k2)=(0.38ꢂ0.05) pKNa u – (5.68ꢂ
0.45), r2=0.9346 (6 data)
izable continuum model (IEFPCM)[29,30] methanol or
water
solvent model. In all calculations, the 6-31G(d,p) basis set was
used for C and H atoms, while heteroatoms were assigned
the 6-31++G(d,p) basis set. Frequency calculations were con-
ducted to use as a basis for determining free energy values
at 298K. Calculations included explicit solvation where
indicated. Solvent methanol or water was oriented and fully
optimized such that it could donate or accept hydrogen bonds
where required.
S
+
S
NucO-
MeO
+
P
-OLg
+
-OLg
+
P
NucO
OMe
MeO OMe
RESULTS
Cleavage reactions of 1 in water
The kobs values were determined using UV–Vis spectrophotome-
try by monitoring the rate of product formation from the cleav-
age of O,O-dimethyl O-arylphosphorothioates 1 in water at
25ꢀC, I=0.5M (NaCl). The pH was kept constant using aryl or alkyl
alcohol buffers (10–300mM) and a set amount of NaOH. The
second-order rate constants (k2) for the aryloxide/alkoxide
promoted cleavage of 1a-e were determined as the gradients
of the linear plots of kobs versus [ArO–] or [CF3CH2O–] and are
presented in Table S1 of the Supporting Information.
S-
S
NucO -
+
NucO---P dist.
NucO
MeO
P
OLg
OMe
P
OLg
MeO
OMe
Figure 2. A More O’Ferrall–Jencks diagram showing the effect on the
βNuc value by making a change to a better leaving group on the sub-
strates. The net effect is to decrease the extent of NucO––P bond forma-
tion with little change in the degree of P–Olg departure leading to a
decrease in the βNuc value when the substrate has better leaving groups
Shown in Fig. 1 are Brønsted plots of log(k2) versus pKaNu
,
which are fit to standard linear regressions providing gradients
of βNu=(0.38ꢂ0.05) for the cleavage of 1a (pKLag=4.11)[31]
,
(0.54ꢂ0.03) for the cleavage of 1c (pKLag=6.46)[31] and (0.61ꢂ
0.04) for the cleavage of 1d (pKLag=7.14)[31]. The Brønsted plots
indicate that passing from the most activated leaving group of
1a to the least activated leaving group of 1d increases the gradi-
ent (or βNu coefficient) by a factor of almost two, which is consis-
tent with expectations based on the More O’Ferrall–Jencks
theory.[14–16,32,33] As shown by the diagram in Fig. 2, substitution
of the poorer 4-nitrophenoxy leaving group in 1d by the better
2,4-dinitrophenoxy leaving group in 1a moves the transition
state along the reaction coordinate towards the starting materi-
als (earlier, Hammond effect) coupled with an anti-Hammond ef-
fect perpendicular to the reaction coordinate towards a more
dissociative pathway in which there is a decrease in the bond
formation between the (CH3O)2P=S unit and the nucleophile at
the transition state for reaction of 1a relative to 1d.
The βNu values determined for the cleavage of 1a,c,d are line-
arly correlated to the pKHOAr of their repective leaving groups
s
s
a
using Eqn (1). This equation can then be used to calculate βNu
values for reactions of other O,O-dimethyl O-aryl phosphorothio-
ates with oxyanion nucleophiles.
βNu ¼ ð0:074 ꢂ 0:006Þ spKHa OAr þ ð0:07 ꢂ 0:04Þ
(1)
s
wileyonlinelibrary.com/journal/poc
Copyright © 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 258–266

EXPERIMENTAL AND COMPUTATIONAL DETERMINATION OF BRØNSTED COEFFICIENTS
In Fig. 3 are shown the Brønsted plots of log(k2) versus sspKaHOAr
for the attack of 2,4,5-trichlorophenoxide (pKNa u=6.72) and phen-
oxide (pKNa u=9.99) on substrate 1. Linear regressions of the data
provide βLg values of ꢁ0.88ꢂ0.08 (2,4,5-trichlorophenoxide) and
ꢁ0.64ꢂ0.06 (phenoxide) defined as the slopes of the lines. The
attack of the less activated nucleophile, 2,4,5-trichlorophenoxide,
on 1 results in a more negative βLg as compared to the reactions
with phenoxide consistent with a lesser degree of bond cleavage
of the leaving group at the transition state for the latter series of
reactions.
gradients of plots of log(k2) versus sspKaHOAr / sspKaNu appear in
Table S2 of the Supporting Information.)
The sspKHa OAr values of 4-nitrophenol (11.18) and 2,4,5-trichloro-
phenol (10.73) in methanol are sufficiently close to allow
computation of the βEq for transfer of the O,O-dimethyl phos-
phorothioyl group between oxyanion nucleophiles in methanol
as βEq=βLg – βNu=(ꢁ0.76) – 0.63=ꢁ1.39ꢂ0.12. The Brønsted
plots determined for the reaction of series 1 in methanol com-
prise fewer data than those in water and furthermore the βEq is
estimated from a quasisymmetrical reaction where the Δ sspKaHOAr
for the attacking and departing aryloxides is 0.42.[34] Nonethe-
less, the estimated βEq is sufficiently well-defined to allow us to
verify that the βEq values determined in water using aqueous
pKa data are experimentaly indistiguishable from those deter-
The βNu for attack of oxyanion nucleophiles on O,O-dimethyl
O-(2,4,5-trichlorophenyl) phosphorothioate can be computed
from Eqn (1) as +0.57 using sspKaHOAr = 6.72 (the value for the acid
disscociation constant of 2,4,5-trichlorophenol). Thus the βEq
for the transfer of the O,O-dimethyl phosphorothioyl group
between oxyanion nucleophiles with the same pKa values (the
mined in alcohol solvent using pKHOArdata.
s
s
a
quasisymmetrical reaction) is calculated as βEq=βLg – βNu
=
Computational βEq determination
(ꢁ0.88) – 0.57=ꢁ1.45ꢂ0.08. This represents the charge change
experienced by the reporter aryloxy group in passing from reac-
tant to product and suggests that a formal charge of +0.45
resides on the bridging aryloxy atom in 1. A similar analysis of
the data using phenoxide attack on O,O-dimethyl O-phenyl
phosphorothioate as the symmetrical reaction also leads to a
βEq=ꢁ1.45.
For the alkoxide or hydroxide promoted cleavage of alkyl and
aryl dimethyl phosphorothioates presented in Eqn (2), βEq is re-
lated to the equilibrium constant, K*, and pKHa OAr of the leaving
group as given in Eqn (3).
S
P
S
P
K*
-
-
H3CO
+
H3CO
OR
+
OAr
(2)
OAr
OR
H3CO
H3CO
Cleavage reactions of 1 in methanol
Second-order rate constants were determined for the attack of
2,4,5-trichlorophenoxide on 1a–d in methanol at 25ꢀC, I=0.5M
Eq
ꢃ
HOAr
s
logK ¼ β ꢄ pKa
þ constant
((33))
(NBu4Cl). A plot of log(k2) versus pKHOAr fits the equation log
s
a
(k )=(ꢁ0.76ꢂ0.11) pKHOAr + (2.6ꢂ1.1). Second-order rate con-
s
s
2
a
stants were also determined for the attack of substituted phen-
Equation (3) can be expressed in terms of the free energy of
oxides on O,O-dimethyl O-4-nitrophenyl phosphorothioate
reaction, ΔG*, as
where a plot of the log(k ) versus pKNu data fits the equation
s
s
2
a
log(k2)=(0.63ꢂ0.04) sspKaNu – (12.8ꢂ0.5). (The kinetic data and
ðꢁΔGꢃÞ
¼ βEq ꢄ pKaHOLG þ constant
(4)
2:303RT
and defining ΔG*=(GP+G-OAr – GSM – GNu), leads to
-1
-2
-3
-4
-5
-6
-7
À
Á
Gp ꢁ GNu
2:303RT
ðGSM ꢁ GꢁOArÞ
2:303RT
¼ βEq ꢃ pKaHOLg
þ
þ constant (5)
Reactions of all substrates 1 lead to a common product when
a common nucleophile is used, namely, (CH3O)3P=S, if a methox-
ide is the nucleophile. As such the values of GNu and GP are con-
stant across a given series of substrates. This allows the last two
terms in Eqn (5) to be treated as constants to yield
ΔG
2:303RT
βEq ꢃ pKaHOLg þ A
(6)
where ΔG=GSM – G–OAr is the difference in the free energies of
the initial phosphorothioate (1) and the free aryloxide anion
product. Note that ΔG does not correspond to a free energy of
reaction because atom balance is not maintained. However, in-
corporating the two free energies needed to achieve atom bal-
ance (GNu and GP) into the constant A reduces the number of
calculations that must be performed, and thus has computa-
tional benefits. Meanwhile, the value of βEq obtained as the slope
of a plot of ΔG/2.303RT versus sspKHa OAr is unaffected by this sep-
aration of free energy contributions because it only affects the
4
5
6
7
8
9
sspKHa OAr
Figure 3. Brønsted plot of log(k2) versus sspsKaHOAr at 25ꢀC for the attack
of 2,4,5-trichlorophenoxide (■, solid line) on O,O-dimethyl O-aryl phos-
phorothioates where log(k2)=(ꢁ0.88ꢂ0.08) sspKaHOAr + (0.23ꢂ0.49), r2=
0.9823 (4 data); and for the attack of phenoxide (▼, dashed line) on O,
O-dimethyl O-aryl phosphorothioates where log(k2)=(ꢁ0.64ꢂ0.06)
sspKHa OAr + (0.61ꢂ0.37), r2=0.9766 (5 data)
intercept of such a plot. A plot of ΔG/2.303RT versus pKHOAr
s
s
a
J. Phys. Org. Chem. 2012, 25 258–266
Copyright © 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

D. R. EDWARDS ET AL.
βEq
for the hydrolysis of 1 is shown in Fig. 4. The values of ΔG were
obtained at the B3LYP/6-31G(d,p) level using a continuum sol-
vent model to yield βEq=ꢁ3.9ꢂ0.4, which differs significantly
from the experimental value of ꢁ1.45. Analogous calculations in-
cluding one explicit H-bonding interaction (described as vide
infra) yielded βEq=ꢁ3.3ꢂ0.2. The poor agreement between the
calculated and experimental values is likely due to an exaggera-
tion of the destabilization of the negative charge with poor leav-
ing groups.
S
P
S
P
*
K
+ -OAr + H+
+ HOCH3
MeO
Ka
MeO
MeO
OCH3
OAr
MeO
HOAr
'
K
S
P
'
β
β = -1
MeO
MeO
+
OCH3 HOAr
To address the difficulty in modeling charged substrates, an
approach that employed the free energies of the neutral sub-
strates and their corresponding neutral phenols was developed.
The overall reaction is represented by the thermochemical steps
shown in Scheme 2. The first step involves the exchange of a
methanol for a neutral phenol. The associated value of β′ can
be determined according to Eqn (6) from a plot of ΔG/2.303RT
versus sspKaHOAr, where ΔG=GSM – GHOAr. The second step involves
the deprotonation of the phenol, which by definition has a value
of β=ꢁ1.0. Based on this thermochemical cycle, the value of βEq
is given by β′ – 1.0.
Figures 5 and 6 show the Brønsted plots using the experimen-
tal pKaHOAr values that give rise to the computed βEq values for
the hydrolysis and methanolysis of phosphorothioate esters with
no explicit hydrogen bonding included. For these βEq calcula-
tions, aryl leaving groups with pKHa OAr values greater than 10.02
in water (or sspKHa OAr greater than 14.3 in methanol) were ex-
cluded because they are outside the limits of the leaving groups
used for the experimental βEq values, and their use in the com-
putations consistently showed a tendency to deviate from linear-
ity, possibly due to the differences in the H-bonding
environment of phenols in this pKa range. The resulting values
of βEq=ꢁ1.75 (hydrolysis) and ꢁ1.69 (methanolysis) are in much
better agreement with the experimental values of ꢁ1.45 and
ꢁ1.39, respectively, than the values reported above using the
aryloxide anions.
Scheme 2. Cycle linking starting material and corresponding leaving
group phenol, where βΕq = β′–1
-4
-5
-6
-7
-8
-9
4
6
8
10
HOAr
pKa
Figure 5. Computed Brønsted plot for determination of βEq for the hy-
drolysis of dimethyl aryl phosphorothioates where the calculated
βEq=β′ –1.0=ꢁ1.75ꢂ0.08
-4
-5
-6
-7
-8
-9
Inclusion of explicit leaving group solvation
Modelling of the H-bonding environment around the aryloxygen
was attempted in three ways: (1) solvent donating a hydrogen
bond to the Lg aryl oxygen; (2) solvent accepting a hydrogen
bond from the HOAr hydroxyl group; and (3) both 1 and 2
(Table 1). The addition of one H-bond from solvent proton to
10
12
14
HOAr
spKa
s
-5
-10
-15
-20
-25
Figure 6. Computed Brønsted plot for determination of βEq for the
methanolysis of dimethyl aryl phosphorothioates in methanol where cal-
culated βEq=β′ – 1.0=ꢁ1.69ꢂ0.05
leaving group oxygen uniformly decreases the absolute value
of the calculated βEq, making it closer to the experimental
values for phosphorothioates and, less convincingly so, acetates.
The inclusion of H-bonds involving the phenolic proton
increases the absolute βEq values. In view of the somewhat arbi-
trary decisions about which H-bonds to include, we have opted
for the simplest computational protocol where the solvent is
treated by a continuum model without explicit H-bonding inter-
actions to the phenol. These values are shown in column 3 of
Table 1.
6
8
10
HOAr
pKa
Figure 4. A pseudo-Brønsted plot for the determination of βEq for the
hydrolysis of 1 obtained via calculating energy differences between start-
ing materials and aryl oxides as in Eqn (6). Calculated βEq=ꢁ3.9ꢂ0.4
wileyonlinelibrary.com/journal/poc
Copyright © 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 258–266

EXPERIMENTAL AND COMPUTATIONAL DETERMINATION OF BRØNSTED COEFFICIENTS
βEq (Experim.)ref, Calc.
Computational βEq values for acyl and phosphoryl transfer
reactions with other neutral substrates
O
P
(-1.87 +/- 0.13)16, -1.68 +/- 0.07
EtO
EtO
O
O
X
X
The versatility of the computational method was tested by deter-
mining whether it could satisfactorily replicate other experimen-
tally determined βEq values for the transfer of phosphoryl or acyl
groups from the substrates shown in Scheme 3 (data from
Refs. [35–38]). The simplest explicit solvation model (only contin-
uum solvation, no added solvent molecules) results in a good
agreement with the experimental βEq values.
O
(-1.83 +/- 0.07)17, -1.70 +/- 0.08
(-1.70 +/- 0.07)18, -1.8 +/- 0.2
(-1.80 +/- 0.09)19, -1.84 +/- 0.08
P
HO
HO
O
C
H3C
H2N
O
O
X
O
C
X
DISCUSSION
Scheme 3. Computed βEq values and experimental βEq ones for the hy-
drolysis of (a) diethyl aryl phosphates, (b) PO3H2OAr esters, (c) acetate
esters, (d) O-aryl carbamates. Errors in calculated values are standard
deviations of the computed Brønsted plots. Errors in experimental values
calculated from the original data. Calculated value of ꢁ1.68 was for di-
methyl aryl phosphates while experimental value is for the diethyl
analogues
Experimental determination of βΕq for transfer of O,
O-dimethyl phosphorothioyl group between oxyanion
nucleophiles
The two main objectives of this study have been met. These
were: (1) to determine the βEq for transfer of the (CH3O)2P=S
group between oxyanion nucleophiles in both water and meth-
anol and (2) to provide experimental support for our earlier
assumption that the βEq value determined for acyl and phos-
phoryl group transfer in water holds in methanol solution pro-
vided that due attention is paid to providing accurate pKa
values for the nucleophiles and leaving groups in that solvent.
This is apparently true, at least for the phosphorothioates con-
sidered here, since the experimental values for βEq of ꢁ1.45ꢂ
0.08 in water and ꢁ1.39ꢂ0.12 in methanol are experimentally
indistiguishable.
Concerted versus stepwise nucleophilic displacement
reactions on phosphorothioates 1
There is experimental evidence pertaining to whether oxyanion
nucleophilic displacements on neutral P=S containing triesters
such as 1 proceed via a concerted phosphorane-like transition
state, or via a stepwise process involving the formation of a sta-
ble 5-coordinate phosphorane as in Scheme 4. Omakor et al.[3,4]
provided good evidence from linear free energy data that there
was no break in the Brønsted plot of log (knuc) versus pKNa u over
the nucleophile pKa range of 5.53–9.38 for the attack of a series
The βEq of ꢁ1.45 in water determined here is significantly
larger than the value of ꢁ0.88 reported previously.[3,4] In that
study, a βNu of +0.49 was determined for the attack of oxyanion
nucleophiles on O,O-dimethyl O-(3-methyl-4-nitrophenyl) phos-
phorothioate (fenitrothion, pKHa OAr 7.20) and a βLg of ꢁ0.39 was
reported for the attack of phenoxide (pKNa u=9.99) on a series of
substrates 1. The βEq was calculated as (ꢁ0.39 – 0.49)=ꢁ0.88.
There are two major points of difference between the reported
βEq of ꢁ0.88 and the value determined here. The first stems from
the large differences in the βLg values obtained here and in the
previous study[4] and the second stems from not computing
the previously reported βEq for the symmetrical reaction. The nu-
cleophile (phenoxide) used for the previous[3] determination
of the βLg did not have the same pKHa OAr as the leaving group
(3-methyl-4-nitrophenoxide) used for the determination of the
βNu. The use of a nonsymmetrical reaction for the calculation
of the βEq results in an underestimation of the true value.
X-
R'O P OAr
k2
R'O
k1
k-1
X
P
RO
X
P
OR
X
+ -OAr
R'O-
+
OR
OR
-
OAr
RO
RO
R'O P
RO
OAr
OR
Scheme 4. Step-wise (top) versus concerted (bottom) pathways for the
displacement by oxyanion nucleophiles of the –OAr group from nutral
phosphates and phosphorothioates
Table 1. Calculated βEq values for the hydrolysis and methanolysis of substrates with different phenol/solvent H-bonding
environments
Substrate
Solvent
LgOH
LgO
HOR
LgOH
OR
H
LgO
HOR
OR
H
H
1
1
H2O
MeOH
H2O
MeOH
H2O
H2O
ꢁ1.75ꢂ0.08
ꢁ1.69ꢂ0.08
ꢁ1.68ꢂ0.07
ꢁ1.54ꢂ0.07
ꢁ1.70ꢂ0.11
ꢁ1.8ꢂ0.2
ꢁ1.66ꢂ0.09
ꢁ1.54ꢂ0.05
ꢁ1.58ꢂ0.07
ꢁ1.39ꢂ0.04
ꢁ1.61ꢂ0.09
ꢁ1.7ꢂ0.2
ꢁ2.14ꢂ0.05
ꢁ2.00ꢂ0.06
ꢁ2.1ꢂ0.1
ꢁ2.14ꢂ0.03
ꢁ1.71ꢂ0.06
ꢁ2.06ꢂ0.08
ꢁ1.6ꢂ0.1
Dimethyl aryl phosphates
Dimethyl aryl phosphates
PO3H2OAr esters
Acetate esters
O-aryl carbamates
ꢁ1.88ꢂ0.05
ꢁ2.0ꢂ0.1
ꢁ2.0ꢂ0.1
ꢁ2.2ꢂ0.2
ꢁ2.3ꢂ0.2
H2O
ꢁ1.84ꢂ0.08
ꢁ1.8ꢂ0.1
ꢁ2.32ꢂ0.05
ꢁ2.32ꢂ0.05
J. Phys. Org. Chem. 2012, 25 258–266
Copyright © 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

D. R. EDWARDS ET AL.
–
–
of phenoxy anions on fenitrothion (O,O-dimethyl O-(3-methyl-4-
nitrophenyl) phosphorothioate, pKHa OAr=7.20). It is also signifi-
cant that in the present study, the Brønsted correlation for
nucleophilic reaction of phenoxides in water with 1d (pKHa OAr of
7.14) is linear over the entire range of phenoxides employed
(5.53≤pKNa u≤12.38). Theory[15,16] predicts a change in the rate
limiting step for a stepwise (AN+DN) mechanism that would be
evidenced by a downward break in the Brønsted plot at the
symmetrical point where pKNa u=pKaHOAr. That the respective plots
with fenitrothion[3] and 1d in water remain linear for 1.2 and 1.6
pKNa u units below the pKHa OAr of the 3-methyl-4-nitrophenoxy and
4-nitrophenoxy leaving groups is consistent with a concerted
ANDN mechansim for the transfer of the O,O-dimethyl phosphor-
othioyl group between these aryloxide nucleophiles. We address
this question below in the section entitled Lyoxide-Promoted
cleavage of O,O-dimethyl O-aryl phosphorothioates.
nucleophiles such as OH in water or OCH3 in methanol, and
how such processes might depend on the nature of the leaving
group. Few data are available for cleavages of neutral phos-
phorthioates, but the literature gives pertinent information
about such reactions with neutral phosphate triesters, which ap-
parently react by similar nucleophilic mechanisms. Ba-Saif et al.
provided convincing evidence for a concerted transition state
for the transfer of the diphenylphosphoryl group between phen-
oxide anions in water.[35] In the case of –OH reacting with diethyl
aryloxy phosphates, Ba-Saif et al. judged, by analogy with the
above reactions, that the displacement of aryloxy leaving groups
was probably concerted, although with little cleavage of the
ArO–P bond. This conclusion is supported by a 18O–C6H4NO2
leaving group kinetic isotope effect of 1.006 for hydroxide-pro-
moted cleavage of paraoxon (O,O-diethyl O-p-nitrophenyl phos-
phate) which was interpreted[36] as being consistent with a
significant bond order of 0.75 for the P–OAr bond in the ‘SN2-like
transition state of an associative mechanism with concerted,
asynchronous departure of the leaving group.’
Charge map
The experimental data at hand allow construction of the charge
map shown in Scheme 5 for the symmetrical reaction of the
2,4,5-trichlorophenoxide attack on O,O-dimethyl O-2,4,5-trichlor-
ophenyl phosphorothioate in water where the βNu=0.57 and the
βLg=ꢁ0.88. The activated complex is described as slightly loose
because bond cleavage has progressed to the extent of 61%
(a=βLg/βEq=0.61) whereas bond formation has only progressed
to 39% (a=βNu/|βEq|=0.39) at the symmetrical transition state.
A charge imbalance of ꢁ0.14 results from the difference be-
tween 2(ꢁ0.43) units on the aryloxygens and the net ꢁ1 charge
at the transition state: presumably the charge imbalance is dis-
tributed over the extremities of the phosphorothioyl moiety.
In the case of the symmetrical process involving a phenoxide
nucleophile reacting with O,O-dimethyl O-phenyl phosphor-
othioate the reaction proceeds through a tighter transition state
where bond formation has progressed 56% and bond cleavage
has progressed 44%. The partial charge on each phenoxide resi-
due in the transition state is only ꢁ0.19 and the residual charge
imbalance of ꢁ0.62 is distributed over the phosphorothioyl moi-
ety. The charge distribution is different from those discussed for
the 2,4,5-trichlorophenoxy system in Scheme 5 where the elec-
tron withdrawing substituents leads to a better leaving group
and poorer nucleophile, each accommodating more negative
charges leading to a looser transition state.
Recent computational data indicate that there is a gradual
change in mechanism for the –OH attack on phosphate triesters
from concerted (leaving groups where the pKLag is <8) to step-
wise (where the pKLag is greater).[37] In preliminary computational
studies,[38] we have found a similar situation with the –OCH3
attack on phosphorothioates 1 in methanol, where the mech-
anism is concerted when the sspKaLg <11.3 (the sspKHa OAr for
p-nitrophenol in methanol) and stepwise for pKLag ≥ 12.4 (the
s
s
sspKHa OAr for 3-nitrophenol). In the latter case, corresponding to
substrate 1e, the rate limiting step is clearly a nucleophilic attack
followed by the rapid breakdown of the phosphorane. Thus, for
the compounds investigated here, one anticipates that the nu-
cleophilic attack will be in the rate limiting step, whether or
not the reaction proceeds by
mechanism.
a concerted or two-step
Experimentally, the plot of the second-order rate constants for
–OH promoted cleavage of 1a–e in water at 25ꢀC fits the equa-
tion log(k2)=(ꢁ0.32ꢂ0.04) pKHa OAr – (0.2ꢂ0.3). (A tabulation of
the kinetic data, and corresponding Brønsted plot appear in
the Supporting Information as Table S3 and Fig. S3). The Leffler
parameter in water is a=βLg/βEq=ꢁ0.32/–1.45=0.22 meaning
the change in net charge on the leaving OAr group has pro-
gressed 22% of the way from the starting material to the arylox-
ide. The βLg for –OCH3 promoted cleavage of O,O-dimethyl O-aryl
phosphorothioates in methanol at 25ꢀC is ꢁ0.47[13] giving a Lef-
fler parameter of a=βLg/ βEq=ꢁ0.47/–1.39=0.34. Accordingly,
there is a small but significant extent of bond cleavage to the
leaving group at the transition state for the lyoxide-promoted
cleavage of O,O-dimethyl O-aryl phosphorothioates in water
Lyoxide-promoted cleavage of O,O -dimethyl O-aryl
phosphorothioates
There are interesting questions as to whether the concerted na-
ture of the phosphoryl and phosphorothioate group transfers
with the aryloxides discussed above also pertain to better
+0.45
S
S +0.45
S
P
-0.43
-0.43
OArLg
MeO P
O
MeO P
O
NuArO
2,4,5-
Cl
2,4,5-Cl
OMe
OMe
+
+
MeO
-
OMe
-
O
O
2,4,5-Cl
-1.0
2,4,5-Cl
-1.0
βNu = +0.57, βLg = -0.88
βEq = -1.45
Scheme 5. A charge map constructed for the symmetrical reaction of 2,4,5-trichlorophenoxide attack on O,O-dimethyl O-2,4,5-trichlorophenyl phos-
phorothioate in water where βEq is ꢁ1.45
wileyonlinelibrary.com/journal/poc
Copyright © 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 258–266

EXPERIMENTAL AND COMPUTATIONAL DETERMINATION OF BRØNSTED COEFFICIENTS
and slightly more in methanol, although the difference is small
when the experimental errors are considered.
acetate esters, O-aryl carbamates)[35–38] indicates that these are
indistinguishable, and are the same as their calculated values
obtained here within error limits. This is a realtively small sampling
of the available systems where βEq values are known[15,16] and in-
tentionally excludes systems that are charged and therefore sub-
ject to strong specific solvation effects that are difficult to model.
Nevertheless, the straightforward computational methodology
may prove useful as a rapid way to determine βEq values for neu-
tral systems such as those presented in Scheme 3.
The computational results in this paper explicitly deal with
aryloxide leaving groups having a strong electron withdrawing
group that are assumed to be similar enough that differences
in solvation can be ignored. The introduction of one H-bonding
interaction from the solvent proton to the leaving group oxygen
decreases each βEq value by approximately ꢁ0.1 unit, bringing
them somewhat closer to the experimentally determined βEq
values for the phosphorothioates and acetates. However, the in-
clusion of more H-bonding does not improve the agreement,
and in most cases, lessens it.
The data for the lyoxide reactions of the phosphorothioates
can be compared with the experimentally determined hydrox-
ide- and methoxide-promoted cleavages of O,O-diethyl O-aryl
phosphates where βLg values of ꢁ0.43[39,40] (in water) and
ꢁ0.70[19] (in methanol) were reported. These are larger than de-
termined for the phosphorothioates above, but the reported βEq
for diethyl aryl phosphates in water is ꢁ1.87,[15,16] leading to an
associated Leffler parameter for cleavage of the phosphoryl ary-
loxy group of a=0.23. If it is accepted that the βEq values for
equilibrium transfer of the diethoxy phosphoryl group between
oxyanion nucleophiles is the same in water and methanol (after
due accounting from the solvent driven differences in pKHa OAr of
the leaving group), the Leffler parameter for phosphoryl cleav-
age by methoxide in methanol is a=0.37. The closeness of the
a values for lyoxide-promoted cleavage of phosphorothioates
and phosphates in the same solvent, water or methanol, is nota-
ble, suggesting that there is no substantial difference in the ex-
tent of leaving group cleavage attributed to the structural
change from phosphate to phosphorothioate.
CONCLUSIONS
The aim of this study was to determine the βEq value for the trans-
fer of a dimethyl phosphorothioyl group between oxyanion
nucleophiles in both water and in methanol, and to ascertain
whether the solvent changed these values in any significant
way. The results indicate that the βEq values are experimentally
the same (ꢁ1.45 in water and ꢁ1.39 in methanol). The reactions
with phenoxide nucleophiles displacing phenoxide leaving
groups appear to be concerted based on the demonstrated lack
of a break in the Brønsted correlation for reaction of 1d with var-
ious phenoxide nucleophiles, which span a pKHa OAr range that
encompasses that of the leaving group. A highly simplified compu-
tational density functional theory method is presented for deter-
mining the βEq values. This method gives values that seem higher
by up to 0.3 unit than the experimental values for some substrates
given here, but considering the uncertainties in both the computa-
tional and experimental values the agreement is probably much
better.
Computed βEq values
Because numerous kinetic determinations are required to define
βEq in solution, we investigated whether a computational
method might provide a reliable preliminary estimate for βEq
values for various equilibirium transfers of acyl and phosphoryl
groups between oxyanion nucleophiles. The method adopted
determines the trend in computed free energy differences for
the hypothetical process in Scheme 2 using known experimental
pKHa OAr values for the leaving groups in water or methanol. In cal-
culating the free energies, the nucleophile (methanol or water)
and phosphorus containing product are omitted because both
are common to the entire series of reactions, so only the starting
phosphorothioate ester and product aryloxide need to be explic-
itly considered. A second simplification avoids dealing with the
computation of an anionic aryloxide product relative to a neutral
starting material where specific solvation effects must be present,
yet are difficult to model. Thus, the hypothetical process involves
calculation of the neutral starting material and neutral phenol to
give an equilibrium β′ to which is added the value of ꢁ1 to
account for the conversion of the phenol to its phenoxide.
For the computations performed in water or methanol media,
only leaving groups with pKHa OAr values spanning the range of
those used in the present experimental study were used. This
is because these are all closely related phenols having one
strongly electron withdrawing group where the solvation differ-
ences are likely to be small, and perhaps negligible in the series.
The calculations completed without explicit hydrogen bond-
ing are barebone representations of the relative free energy dif-
ferences between the phenoxide and the corresponding sub-
strate without any direct solvent interaction. For the hydrolysis
of 1 the βEq value is ꢁ1.75ꢂ0.08 and the computed value for
methanolysis is the same at ꢁ1.69ꢂ0.05. Although both values
are consistently greater than the experimental values of
ꢁ1.45ꢂ0.08 in water and ꢁ1.39ꢂ0.12 in methanol, considering
the errors in each, at the 95% confidence level, the computed
and experimental numbers are the same.
Acknowledgements
The authors gratefully acknowledge the financial contributions of
the Natural Sciences and Engineering Council of Canada, and the
United States Defense Threat Reduction Agency-Joint Science
and Technology Office, Basic and Supporting Sciences Division
through the award of grant # HDTRA-08-1-0046. In addition, RH
thanks Queen’s University for the award of a Summer Work Experi-
ence Program grant for the period of May to August, 2010.
REFERENCES
[1] A. D. F. Toy, E. N. Walsh, in: Phosphorus Chemistry in Everyday Living,
2nd ed., American Chemical Society, Washington, DC, 1987, 18–20.
[2] L. D. Quin, in: A Guide to Organophosphorus Chemistry, Wiley, New
York, 2000, 367–371.
[3] J. E. Omakor, I. Onyido, G. W. van Loon, E. Buncel, J. Chem. Soc. Perkin
Trans. 2001, 2, 324.
[4] The βLG of ꢁ0.39 determined in ref. 3 was for the reaction of PhO-
with a series of nitroaryl containing phosphorothioates 1, where
log kLG=(ꢁ0.39ꢂ0.04) pKHa OAr - 1.16ꢂ0.08. In the present study,
the βLG determined for the attack of PhO- on phosphorothioates 1 is
ꢁ0.64.
An examination of the βEq values of some neutral acyl and
phosphoryl-containing substrates reported prior to this work
(hydrolyses of diethyl P=O phosphate esters, PO3H2OAr esters,
J. Phys. Org. Chem. 2012, 25 258–266
Copyright © 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

D. R. EDWARDS ET AL.
[5] A. D. Ryabov, in: Palladacycles (Eds: J. Dupont, M. Pfeffer), Wiley-VCH,
Weinheim, 2008, 307–339 and references therein.
[6] G. M. Kazankov, V. S. Sergeeva, E. N. Efremenko, L. Alexandrova, S.
D. Varfolomeev, A. D. Ryabov, Angew. Chem. Int. Ed. 2000, 39, 3117.
[7] M. Kim, Q. Liu, F. P. Gabbaï, Organometallics 2004, 23, 5560.
[8] M. Kim, F. P. Gabbai, Dalton. Trans. 2004, 3403.
[9] M. Kim, A. Picot, P. Gabbai, Inorg Chem 2006, 45, 5600.
[10] M. F. Mohamed, A. A. Neverov, R. S. Brown, Inorg. Chem. 2009, 48,
1183.
[11] Z. L. Lu, A. A. Neverov, R. S. Brown, Org. Biomol. Chem. 2005, 3, 3379.
[12] A. A. Neverov, R. S. Brown, Org. Biomol. Chem. 2004, 2, 2245.
[13] C. T. Liu, C. I. Maxwell, D. R. Edwards, A. A. Neverov, N. J. Mosey, R. S.
Brown, J. Am. Chem. Soc. 2010, 132, 16599.
[14] J. E. Leffler, E. Grunwald, in: Rates and Equilibria of Organic Reactions,
Wiley: New York, 1963.
[15] A. Williams, Acc. Chem. Res. 1984, 17, 425.
[16] A. Williams, in: Concerted Organic and Bioorganic Reaction Mecha-
nisms, CRC Press, Boca Raton, Florida, USA, 2000.
[17] R. S. Brown, Z.-L. Lu, C. T. Liu, W. Y. Tsang, D. R. Edwards, A. A.
Neverov, J Phys Org Chem 2009, 23, 1.
methanol swpH-(ꢁ2.24) = sspH, and since the autoprotolysis constant
of methanol is 10-16.77, neutral sspH is 8.4. Correspondingly, spKa
s
refers to the acid dissociation constant of the phenol or alcohol in
question determined in, and referenced to, methanol.
[25] A. A. Neverov, C. T. Liu, S. E. Bunn, D. R. Edwards, C. J. White, S. A.
Melnychuk, R. S. Brown, J. Am. Chem. Soc. 2008, 130, 6639.
[26] A. D. Becke, Phys. Rev. A. 1988, 38, 3098.
[27] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B. 1988, 37, 785.
[28] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J.
R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J.
Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S.
S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E.
Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.
E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox. Gaussian 09,
Revision A.021. Gaussian, Inc., Wallingford CT/ USA 2009.
[29] J. Tomasi, B. Mennuccia, E. Cancés, E. THEOCHEM. 1999, 464, 211.
[30] J. Tomasi, B. Mennuccia, R. Cammi, Chem. Rev. 2005, 105, 2999.
[31] S. A. Ba-Saif, M. A. Waring, A. Williams, J. Chem. Soc. Perkin 2, 1991,
1653.
[18] R. E. Lewis, A. A. Neverov, R. S. Brown, Org. Biomol. Chem. 2005, 3,
4082.
[19] T. Liu, A. A. Neverov, J. S. W. Tsang, R. S. Brown, Org. Biomol. Chem.
2005, 3, 1525.
[20] D. R. Edwards, C. T. Liu, G. E. Garrett, A. A. Neverov, R. S. Brown, J. Am.
Chem. Soc. 2009, 131, 13738.
[21] D. R. Edwards, A. A. Neverov, R. S. Brown, J. Am. Chem. Soc. 2009,
131, 368.
[32] R. A. More O’Ferrall, J. Chem. Soc. B 1970, 274.
[22] W. Yin Tsang, D. R. Edwards, S. A. Melnychuk, C. T. Liu, C. Liu, A. A.
Neverov, N. H. Williams, R. S. Brown, J. Am. Chem. Soc. 2009, 131,
4158.
[33] W. P. Jencks, Chem. Rev. 1972, 72, 705.
[34] Assuming that the relationship between βNu and pKaLG is the same
as in water (see eq. (1)), a difference of 0.42 pKa units in leaving
group would correspond to a 0.03 units reduction in βNu which in
turn changes the βEq from ꢁ1.39 to ꢁ1.36.
[23] G. Gibson, A. A. Neverov, R. S. Brown, Can. J. Chem. 2003, 81, 495.
[24] For the designation of pH in non-aqueous solvents we use the
forms recommended by the IUPAC, Compendium of Analytical
Nomenclature. Definitive Rules 1997 3rd ed., Blackwell, Oxford,
U. K. 1998. The pH meter reading for an aqueous solution deter-
mined with an electrode calibrated with aqueous buffers is des-
ignated as wwpH; if the electrode is calibrated in water and the ‘pH’
of the neat buffered methanol solution then measured, the term
swpH is used; and if the electrode is calibrated in the same solvent
and the ‘pH’ reading is made, then the term sspH is used. In
[35] S. A. Ba-Saif, M. A. Waring, A. Williams, J. Am. Chem. Soc. 1990, 112,
8115.
[36] S. R. Caldwell, F. M. Raushel, P. M. Weiss, W. W. Cleland, Biochemistry
1991, 30, 7444.
[37] N. Tarrat, J. Mol. Struct: THEOCHEM 2010, 942, 56.
[38] C. I. Maxwell, A. A. Neverov, N. J. Mosey, R. S. Brown, to be published.
[39] S.-B. Hong, F. M. Rauchel, Biochemistry 1996, 35, 10904.
[40] S. A. Ba-Saif, A. Williams, J. Org. Chem. 1988, 53, 2204.
wileyonlinelibrary.com/journal/poc
Copyright © 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 258–266