Effects on the reactivity by changing the electrophilic center from C=O to C=S: Contrasting reactivity of hydroxide, p-chlorophenoxide, and butan-2,3-dione monoximate in DMSO/H2O mixtures

Article abstract of DOI:10.1002/chem.200801534

Second-order rate constants have been measured spectrophotometrically for the reactions of O-p-nitrophenyl thionobenzoate (1, PNPTB) with HO^-, butan-2,3-dione monoximate (Ox^-, α-nucleophile), and p-chlorophenoxide (p-ClPhO^-, normal nucleophile) in DMSO/H ₂O of varying mixtures at (25.0±0.1)°C. Reactivity of these nucleophiles significantly increases with increasing DMSO content. HO ^- is less reactive than p-ClPhO^- toward 1 up to 70 mol % DMSO although HO^- is over six pK_a units more basic in these media. Ox^- is more reactive than p-ClPhO^- in all media studied, indicating that the α-effect is in effect. The magnitude of the α-effect (i.e., k_ox /k_p-ClPhO) increases with the DMSO content up to 50 mol% DMSO and decreases beyond that point. However, the dependency of the α-effect profile on the solvent for reactions of 1 contrasts to that reported previously for the corresponding reactions of p-ni-trophenyl benzoate (2, PNPB); reactions of 1 result in much smaller α-effects than those of 2. Breakdown of the α-effect into ground-state (GS) and transition-state (TS) effects shows that the GS effect is not responsible for the α-effect across the solvent mixtures. The role of the solvent has been discussed on the basis of the bell-shaped α-effect profiles found in the current study as well as in our previous studies, that is, a GS effect in the H₂O-rich region through H-bonding interactions and a TS effect in the DMSO-rich media through mutual polarizability interactions.

Full text of DOI:10.1002/chem.200801534

FULL PAPER
DOI: 10.1002/chem.200801534
Effects on the Reactivity by Changing the Electrophilic Center from C=O to
C=S: Contrasting Reactivity of Hydroxide, p-Chlorophenoxide, and Butan-
2,3-dione Monoximate in DMSO/H2O Mixtures
Ik-Hwan Um,*[a] Jeong-Yoon Han,[a] and Erwin Buncel*[b]
Abstract: Second-order rate constants
have been measured spectrophotomet-
rically for the reactions of O-p-nitro-
phenyl thionobenzoate (1, PNPTB)
with HOꢀ, butan-2,3-dione monoxi-
mate (Oxꢀ, a-nucleophile), and p-
chlorophenoxide (p-ClPhOꢀ, normal
nucleophile) in DMSO/H2O of varying
mixtures at (25.0ꢁ0.1)8C. Reactivity
of these nucleophiles significantly in-
creases with increasing DMSO content.
HOꢀ is less reactive than p-ClPhOꢀ
toward 1 up to 70 mol% DMSO al-
though HOꢀ is over six pKa units more
basic in these media. Oxꢀ is more reac-
tive than p-ClPhOꢀ in all media stud-
ied, indicating that the a-effect is in
effect. The magnitude of the a-effect
(i.e., kOxꢀ/kp-ClPhOꢀ) increases with the
DMSO content up to 50 mol% DMSO
and decreases beyond that point. How-
ever, the dependency of the a-effect
profile on the solvent for reactions of 1
contrasts to that reported previously
for the corresponding reactions of p-ni-
trophenyl benzoate (2, PNPB); reac-
tions of 1 result in much smaller a-ef-
fects than those of 2. Breakdown of the
a-effect into ground-state (GS) and
transition-state (TS) effects shows that
the GS effect is not responsible for the
a-effect across the solvent mixtures.
The role of the solvent has been dis-
cussed on the basis of the bell-shaped
a-effect profiles found in the current
study as well as in our previous studies,
that is, a GS effect in the H2O-rich
region through H-bonding interactions
and a TS effect in the DMSO-rich
media through mutual polarizability in-
teractions.
Keywords: electrophilic addition ·
enthalpy of solution · kinetics ·
medium effect · polarizability
Introduction
tion-state (TS) stabilization, thermodynamic stabilization of
products, and solvent effects.[2–11] However, none of these
hypotheses is able to explain the a-effect phenomenon.[2]
Particularly, controversial are the hypotheses concerning sol-
vent effect on the a-effect.[2a,8a,12–22]
Nucleophiles possessing an lone pair of electrons at the
atom a to the nucleophilic center have been termed a-nu-
cleophiles and exhibit enhanced reactivity compared with
normal nucleophiles of similar basicity.[1] Numerous studies
have been performed to investigate the cause of the en-
hanced reactivity (i.e., the a-effect) observed for a-nucleo-
philes.[2–22] The most prevalent theories on the a-effect phe-
nomenon include ground-state (GS) destabilization, transi-
The solvent effect of the a-effect has been suggested to
be unimportant on the basis of the observation that the
magnitude of the a-effect is similar for reactions performed
in H2O and in neat organic solvents (e.g., MeCN and tolu-
AHCTUNGTRENNUNG
ene).[8a,12] Also controversial have been results of gas-phase
studies and theoretical calculations into the origin of the a-
effect.[13] In 1983, DePuy et al. performed gas-phase reac-
tions of methyl formate with HOOꢀ and HOꢀ (as the refer-
ence nucleophile) and found that HOOꢀ does not exhibit
any enhanced reactivity.[13a] Accordingly, the a-effect ob-
served for various reactions with HOOꢀ in aqueous solu-
tions was attributed to a solvent effect.[13] By contrast, a-nu-
cleophiles have been reported to be intrinsically more reac-
tive than normal nucleophiles of similar basicity in gas-
phase reactions (e.g., HOOꢀ vs. EtOꢀ) again with methyl
formate.[14] According to recent gas-phase studies including
high-level theoretical calculations, a-nucleophiles exhibit
[a] Prof. Dr. I.-H. Um, J.-Y. Han
Division of Nano Sciences and Department of Chemistry
Ewha Womans University, Seoul 120-750 (Korea)
Fax : (+82)2-3277-2844
E-mail: ihum@ewha.ac.kr
[b] Prof. Dr. E. Buncel
Department of Chemistry, Queenꢀs University
Kingston, ON K7L 3N6 (Canada)
Fax : (+1)613-533-6669
E-mail: buncele@chem.queensu.ca
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801534.
Chem. Eur. J. 2009, 15, 1011 – 1017
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1011

lower activation energies than normal nucleophiles of simi-
lar basicity in gas-phase SN2 reactions.[14–16] Thus, solvent ef-
fects on the a-effect have been suggested to be unimpor-
tant.[14–16]
We have extended our study to reactions of the thiono
ester O-p-nitrophenyl thionobenzoate (1, PNPTB) with
HOꢀ, Oxꢀ and p-ClPhOꢀ in DMSO/H2O (Scheme 1). The
kinetic data have been compared with those for the corre-
sponding reactions of p-nitrophenyl benzoate (2, PNPB) to
examine the effect of modification of the electrophilic
center from a C=O bond (2) to a polarizable C=S bond (1)
on reactivity and the a-effect.
We have found remarkable solvent effects on the a-effect
in nucleophilic substitution reactions at various electrophilic
centers.[17–22] In 1986, we initiated a systematic study to in-
vestigate solvent effects on the a-effect.[17a] Our first study
showed a significant solvent effect on the a-effect for reac-
tions of p-nitrophenyl acetate (PNPA) with butan-2,3-dione
monoximate (Oxꢀ, a-nucleophile) compared with p-chloro-
phenoxide (p-ClPhOꢀ, normal nucleophile) in DMSO/H2O
of varying compositions: the a-effect (kOxꢀ/kp-ClPhOꢀ) increas-
es with increasing DMSO content in the medium up to
about 50 mol% DMSO and then decreases thereafter, that
is, a bell-shaped profile.[17] Similar bell-shaped a-effect pro-
files have been obtained for the corresponding reactions of
aryl acetates,[18] p-nitrophenyl benzoate,[19] p-nitrophenyl di-
phenylphosphinate,[20a] and p-nitrophenyl benzenesulfon-
ACHTUNGTRENNUNG
ate,[20b] although the magnitude of the a-effect is highly de-
Scheme 1.
pendent on the nature of the electrophilic center of the sub-
strates (C=O, P=O, and SO2).[20]
The basicity of Oxꢀ and p-ClPhOꢀ in DMSO/H2O mix-
tures has been found to increase in almost parallel manner
with increasing mol% DMSO, indicating that the bell-
shaped a-effect profiles are not due to a difference in basici-
ty of the two nucleophiles.[20a] Our calorimetric study for the
two nucleophiles has shown that the difference in enthalpy
of solution between OxꢀNa+ and p-ClPhOꢀNa+ (i.e.,
Results and Discussion
The kinetic study was performed under pseudo-first-order
conditions with the concentration of nucleophile in excess
over the substrate concentration. All reactions obeyed first-
order kinetics with quantitative liberation of p-nitrophenox-
ide ion. Pseudo-first-order rate constants (kobsd) were calcu-
OxꢀNa+
p-ClPhOꢀNa+
DDHsol =DHsol
ꢀDHsol
) increases up to 30–40
lated from the equation ln
N
mol% DMSO and then remains nearly unchanged upon fur-
ther addition of DMSO.[17b] Breakdown of the a-effect into
GS and TS contributions through combination of the kinetic
of kobsd versus nucleophile concentration were linear passing
through the origin. Second-order rate constants have been
determined from the slope of the linear plots and summar-
ized in Table 1. It is estimated from additional runs that the
uncertainty in the rate constants is less than ꢁ3%. Detailed
kinetic conditions and data are summarized in the Support-
ing Information.
data with enthalpy of solution (DHsol) data has enACTHNUGTRNEUNGabled us
to conclude that desolvation of the a-nucleophile Oxꢀ is
mainly responsible for the increasing a-effect up to
50 mol% DMSO (i.e., GS effect) while differential stabiliza-
tion of TS contributes to the decreasing a-effect in the
DMSO-rich solvent mixtures.[17b]
Effect of medium on reactivity: Table 1 and Figure 1 show
that the reactivity of the nucleophiles toward both 1 and 2
increases as the DMSO content increases. It is also seen for
1, Oxꢀ is much more reactive than p-ClPhOꢀ or HOꢀ in all
solvents studied, indicating that a sizable a-effect is opera-
A remarkable GS effect on the a-effect has been found
for reactions of S-p-nitrophenyl thioacetate (PNPTA) with
Oxꢀ and p-ClPhOꢀ in DMSO/H2O.[22] The a-effect for the
PNPTA system increases as the mol% DMSO increases up
to 30–40 mol% and then re-
mains nearly constant beyond
Table 1. Summary of second-order rate constants for reactions of 1 (and 2 in parentheses) with Oxꢀ, p-
ClPhOꢀ and HOꢀ in DMSO/H2O of varying composition at (25.0ꢁ0.1)8C.
that point.[22] Consequently, the
a-effect trend in the PNPTA
system is equal to the difference
in enthalpy of solution between
Oxꢀ and p-ClPhOꢀ (DDHsol), in-
dicating clearly that the differen-
tial GS desolvation between the
two nucleophiles is largely re-
sponsible for the a-effect profile
over the whole medium range
studied.[22]
Mol% DMSO
kOxꢀ [mꢀ1 sꢀ1
(22.5)[a]
(20.0)[a]
(34.2)[a]
(75.7)[a]
(186)[a]
(455)[a]
(1070)[a]
(2360)[a]
(5360)[a]
(11700)[a]
]
kp-ClPhOꢀ [mꢀ1 sꢀ1
]
kHOꢀ [mꢀ1 sꢀ1
]
kOxꢀ/kp-ClPhOꢀ
32
(88)[a]
(139)[a]
(185)[a]
(218)[a]
(229)[a]
(206)[a]
(187)[a]
(152)[a]
(120)[a]
(72)[a]
0
48.9
102
254
N
1.53
2.20
3.91
9.57
25.4
63.8
186
511
1470
–
G
0.470
0.750
1.62
4.7
15.8
51.1
159
502
1790
–
ACHTUNGTRENNUNG
10
20
30
40
50
60
70
80
90
A
46
ACHTUNGTRENNUNG
N
65
ACHTUNGTRENNUNG
786
N
82
ACHTUNGTRENNUNG
2560
7790
20700
52300
127000
–
R
101
122
111
102
86
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
T
ACHTUNGTRENNUNG
T
ACHTUNGTRENNUNG
G
–
[a] Data for the reactions of 2 were taken from ref. [19].
1012
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1011 – 1017

Reactivity Effects
FULL PAPER
tive for the reactions of 1. Interestingly, HOꢀ is less reactive
than p-ClPhOꢀ toward 1 up to about 70 mol% DMSO, al-
though the former is over six pKa units more basic than the
latter in these media. In contrast, as shown in Figure 1 for
the corresponding reactions of 2, HOꢀ exhibits much higher
reactivity than p-ClPhOꢀ, indicating that the unusual reac-
tivity of HOꢀ is limited to the reactions of 1.
60 mol% DMSO for the reactions of 1 although the degree
of GS desolvation on addition of DMSO is more significant
for HOꢀ than for Oxꢀ. A contrasting result is demonstrated
for the corresponding reactions of 2. As shown in Figure 2
(PNPB), HOꢀ exhibits significantly larger relative rate con-
stants than Oxꢀ in all solvent compositions.
The enthalpy of transfer from H2O to DMSO/H2O for a
DMSO/H2O
nucleophile, which can be expressed as DDHtr =DHsol
ꢀ
DHsolH O, represents the degree of desolvation of the nucleo-
phile upon changing the medium from H2O to DMSO/
H2O.[17b] Figure 3 illustrates the relationship between DDHtr
and the relative reactivity for the reactions of 1 with Oxꢀ, p-
ClPhOꢀ and HOꢀ. It is seen that logarithmic relative rate
constants for the three nucleophiles follow a similar trend in
the region where DDHtr <5 kcalmolꢀ1, indicating that the
effect of GS desolvation on reactivity is similar for the three
nucleophiles in that region. However, in the region where
DDHtr >5 kcalmolꢀ1, p-ClPhOꢀ exhibits the largest while
HOꢀ the smallest relative rate constant. This result indicates
that GS desolvation on addition of DMSO cannot be solely
responsible for the increasing relative rate constants. Ac-
cordingly, differential stabilization of TS on addition of
DMSO to the medium is considered to contribute to the in-
creasing rate constants in the reactions of 1. The fact that p-
ClPhOꢀ exhibits the largest relative reactivity where
DDHtr >5 kcalmolꢀ1, while HOꢀ is subject to the smallest
rate enhancement, shows that TS stabilization on addition
of DMSO is most significant for p-ClPhOꢀ but least for
HOꢀ.
2
Figure 1. Plots of the logarithmic second-order rate constants versus
mol% DMSO for the reaction of O-p-nitrophenyl thionobenzoate (1)
and p-nitrophenyl benzoate (2) with Oxꢀ, p-ClPhOꢀ and HOꢀ at (25.0ꢁ
0.1)8C.
The effect of the medium on the reactivity is illustrated in
a different manner in Figure 2 (PNPTB), which shows the
relative rate constants, that is, the ratio of the second-order
rate constant in DMSO/H2O over the corresponding rate
constant in H2O, kDMSO/H O/kH O, as a function of mol%
DMSO. It is apparent that Oxꢀ exhibits larger relative rate
constants than p-ClPhOꢀ, which appears to be mainly a re-
flection of the GS energies of the nucleophiles. We have
shown that Oxꢀ is more destabilized than p-ClPhOꢀ as the
DMSO content in the medium increases.[17b] Interestingly,
HOꢀ shows smaller relative rate constants than Oxꢀ up to
2
2
Figure 3. Plots of the logarithmic relative rate constants, logkDMSO/H O/kH O
versus DDHtr for the reaction of 1 with Oxꢀ, p-ClPhOꢀ and HOꢀ at
(25.0ꢁ0.1)8C. The DDHtr data were taken from ref. [17b].
2
2
Effect of medium on the a-effect: As shown in Figure 1,
Oxꢀ is more reactive than p-ClPhOꢀ toward 1 throughout
the DMSO/H2O solvent region. The a-effect for the current
reactions of 1 is illustrated in Figure 4 as a function of
mol% DMSO together with the one reported previously for
the corresponding reactions of 2 for comparison.[19] Figure 4
clearly demonstrates a solvent composition dependent a-
Figure 2. Plots of logarithmic relative rate constants, logkDMSO/H O/kH O
,
2
2
versus mol% DMSO for reactions of 1 and 2 with Oxꢀ, p-ClPhOꢀ and
HOꢀ at (25.0ꢁ0.1)8C.
Chem. Eur. J. 2009, 15, 1011 – 1017
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1013

I.-H. Um, E. Buncel and J.-Y. Han
*
*
Figure 4. Plots of the a-effect (kOxꢀ/kp-ClPhOꢀ) versus mol% DMSO for re-
Figure 5. Plots for the a-effect behavior of 1 ( ) and 2 ( ), as a function
of the differential enthalpy of solution of Oxꢀ versus p-ClPhOꢀ in
DMSO/H2O mixtures. The break in the plot occurs at 40 mol% DMSO.
Data for DDHsol were taken from ref. [17b].
ꢀ
actions of 1 with Oxꢀ and p-ClPhOꢀ
(
) and 2 with Ox and p-ClPhOꢀ
*
*
(
) at (25.0ꢁ0.1)8C.
effect for the reactions of 1 and 2. The a-effect for both sys-
tems increases with increasing the DMSO content in the
medium up to 40 or 50 mol% DMSO, and then decreases
upon further increase in mol% DMSO. However, the shape
and magnitude of the a-effect profiles are contrasting, that
is, 1 exhibits smaller a-effect than 2 and the increase (or de-
crease) in the a-effect on changing the medium composition
is less significant for 1 than for 2.
effect in the DMSO-rich region. It is also noted that the a-
effect for reactions of 1 exhibits lower sensitivity to DDHsol
than that for 2 up to near DDHsol =7 kcalmolꢀ1. As well 1
exhibits much smaller variation in the magnitude of the a-
effect at DDHsol = ꢂ7 kcalmolꢀ1, than 2. Thus, one can con-
clude that another effect than GS contributes to the con-
trasting a-effect profiles shown in Figure 4. This is consid-
ered below.
Dissection of the a-effect into GS and TS effects can be
accomplished through combination of the kinetic data with
enthalpy of solution (DHsol) data reported previously by our
group.[17b]
Polarizability effect on the a-effect: Different studies have
shown that the contrasting a-effect found for the reactions
of 1 and 2 is unlikely due to a difference in the reaction
mechanism. We have recently shown that reactions of aryl
benzoates including 2 with three representative anionic nu-
It was shown that while DHsol for OxꢀNa+ greatly exceeds
that for p-ClPhOꢀNa+ over the entire range of DMSO/H2O
solvent composition, the difference in enthalpy of solution
OxꢀNa+
ꢀ
for Oxꢀ and p-ClPhOꢀ (i.e., DDHsol =DHsol
DHsol
ꢀ
cleophiles (i.e., HOꢀ, CNꢀ and N3 ) proceed through a step-
p-ClPhOꢀNa+
) increases up to near 40 mol% DMSO and
wise mechanism.[23] Alkaline hydrolysis of 1 has also been
reported to proceed through a stepwise mechanism.[24] Be-
sides, studies on reactions of 1 and 2 with primary amines
have shown that the aminolysis of these esters proceeds
through the same stepwise mechanism.[25]
then remains nearly constant upon further addition of
DMSO.[17b] Thus, if DDHsol (i.e., the difference in GS energy
between Oxꢀ and p-ClPhOꢀ) were responsible for the a-
effect, the a-effect should increase up to near 40 mol%
DMSO and then remain nearly constant beyond that point.
Besides, one might expect that reactions of 1 and 2 should
result in a similar a-effect in magnitude, since DDHsol is con-
stant for Oxꢀ and p-ClPhOꢀ for the two systems. However,
the a-effect shown in Figure 4 increases up to 40 or
50 mol% DMSO and then decreases as the DMSO content
in the medium increases further. Thus, the current results in-
dicate that GS effect (i.e., DDHsol) cannot solely be respon-
sible for the a-effect found for the reactions of 2.
The above argument can be further supported by
Figure 5, which shows that the magnitude of the a-effect in-
creases linearly with increasing DDHsol up to about 40 or
50 mol% DMSO but thereafter the a-effect becomes inde-
pendent of DDHsol. This result indicates that GS effect could
be partially responsible for the increasing a-effect up to 40
or 50 mol% DMSO but clearly not for the decreasing a-
The major factor influencing reactivities in the present
system is polarizability. Thus, enhanced polarizability of the
thiono ester compared with its oxygen analogue is consid-
ered to be responsible for the contrasting a-effect between
the two systems, since replacing the C=O bond in 2 by C=S
is accompanied by a significant increase in polarizability.[26]
The enhanced polarizability of the C=S bond in 1 is well re-
flected in the 13C NMR chemical shift of the thiono carbonyl
carbon. The chemical shifts are d=163.8 ppm for the car-
bonyl carbon in 2 and 209.8 ppm for the thio carbonyl
carbon in 1, that is, a 46 ppm downfield shift in the
13C NMR.[26] This is in good agreement with the literature
value 30–50 ppm downfield shift reported for various sulfur
compounds compared with the corresponding oxygen ana-
logues.[27] In addition to 13C NMR spectroscopy, significant
1
deshielding influences have been observed in the H NMR
1014
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1011 – 1017

Reactivity Effects
FULL PAPER
and 19F NMR chemical shifts for a series of thionyl acid flu-
orides, RC(S)F.[28] Such a deshielding effect is not surprising
since an electron deficiency on the thio carbonyl carbon
would be expected to result from inefficient orbital overlap
and/or the ability of sulfur to stabilize a negative charge in
the substrate by distributing the electron cloud through the
3d orbitals.
in the resonance structures I and II for Oxꢀ and III through
VI for p-ClPhOꢀ. The following argument shows that this
delocalization is more effective for p-ClPhOꢀ than for Oxꢀ.
Figure 6 demonstrates that upon increasing the mol%
DMSO, the kC= /kC ratio increases more significantly for
=
S
O
the reactions with p-ClPhOꢀ than for Oxꢀ. A similar result
is shown in Figure 3, that is, p-ClPhOꢀ exhibits much larger
kDMSO/H O/kH O ratios than Oxꢀ for the reaction of 1 in the
region where DDHtr >5 kcalmolꢀ1. These results suggest
that p-ClPhOꢀ is relatively more polarizable than Oxꢀ.
Thus, one can conclude that the contrasting relative reactivi-
ty and the a-effect profile found for the reactions of 1 and 2
is mainly due to the enhanced polarizability of the electro-
philic center upon replacing the C=O bond in 2 by a polariz-
able C=S bond.
2
2
Consequently, one might expect a greater reactivity of a
nucleophile at the electrophilic center of 1 compared with 2.
In fact, as shown in Figure 6, the relative reactivity of 1 to 2
(i.e., kC= /k O) for the reactions with p-ClPhOꢀ and Oxꢀ is
=
C
S
much larger than unity and it increases with increasing
mol% DMSO. In contrast, the thiono ester is less reactive
than its oxygen analogue toward HOꢀ (kC= /k O =0.11–
=
C
S
0.26) in all solvent compositions, indicating that the reactivi-
ty of the polarizable 1 is strongly dependent on the nature
of the nucleophiles.
Role of solvent—Gas-phase reactions: Our present work is
ideal for commenting on the role of the solvent in nucleo-
philic reactions, noting that the DMSO/H2O solvent system
provides the characteristics of both protic (H2O) and aprotic
(DMSO) media, that is, hydrogen bonding in the H2O-rich
region and polarizability interactions in the DMSO-rich
domain.[30,31] These characteristics, when combined with
present experimental results showing bell-shaped a-effect
profiles (Figure 4), have enabled us to discuss GS and TS
solute–solvent interactions in a more detailed manner (cf.
ref. [15]).
As noted in other publications, hydrogen-bonding interac-
tions are believed to be dominant for anionic solutes in
H2O-rich region.[30,31] However, charge dispersion or mutual
polarizability interactions between solutes and solvent mole-
cules become the more important interactions in the
DMSO-rich region.[30,31] Our calorimetric study has shown
that in H2O, Oxꢀ is less strongly solvated than p-ClPhOꢀ by
4 kcalmolꢀ1, and that on addition of DMSO to the medium
up to 40 to 50 mol% DMSO, both Oxꢀ and p-ClPhOꢀ
become greatly desolvated.[17b] However, as mentioned in
the preceding section, the difference in enthalpy of solution
for Oxꢀ and p-ClPhOꢀ (i.e., DDHsol) increases on addition
of DMSO to H2O up to about 40 mol% DMSO and then re-
mains nearly constant thereafter.[17b] The increasing DDHsol
follows a similar pattern to the increasing a-effect profile;
thus, the GS effect is shown to be mainly responsible for the
a-effect profile up to 40 or 50 mol% DMSO.
Figure 6. Plots of the relative reactivity of 1 to 2, kC= /kC versus mol%
=
S
O
DMSO. Note that kC= /k O =0.11–0.26 for the reaction with HOꢀ.
=
C
S
The above results agree well expectations based on the
HSAB principle.[29] It is noted that while HOꢀ is a hard nu-
cleophile, Oxꢀ and p-ClPhOꢀ, although they are oxygen nu-
cleophiles, are considered to be relatively polarizable. The
negative charge on the oxygen atom in Oxꢀ and p-ClPhOꢀ
can be delocalized through resonance interactions as shown
The above interpretation is supported by the measured
activation parameters for reactions of PNPA with Oxꢀ and
p-ClPhOꢀ in various DMSO/H2O mixtures, where it was
found that the reactions are mainly governed by the enthal-
py of activation, DH°; beyond 20 mol% DMSO DH° de-
creases on addition of DMSO to the medium while DS° re-
mains nearly unchanged for both Oxꢀ and p-ClPhOꢀ sys-
tems.[17b] An interesting finding was that the decrease in
DH° for p-ClPhOꢀ is steeper than that for Oxꢀ in the
medium range 50–90 mol% DMSO, which indicates that the
TS for the p-ClPhOꢀ system becomes relatively more stabi-
lized than the TS for Oxꢀ in the DMSO-rich region.[17b]
Thus, differential TS stabilization has been demonstrated to
Chem. Eur. J. 2009, 15, 1011 – 1017
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1015

I.-H. Um, E. Buncel and J.-Y. Han
be responsible for the decreasing a-effect profile in the
DMSO-rich region (see above).[17b]
1) HOꢀ is more reactive than p-ClPhOꢀ toward 2, but less
reactive toward 1 up to 70 mol% DMSO although the
former is over six pKa units more basic in these media.
2) Oxꢀ is more reactive than p-ClPhOꢀ toward 1 through-
out the whole medium range (i.e., the a-effect), but the
a-effect profile for the reactions of 1 contrasts to that
previously reported for the reactions of 2.
3) Dissection of the a-effect into GS and TS effects shows
that the GS effect is not responsible for the a-effect over
the entire solvent composition for reactions of 1.
Since the difference in polarizability of the TS between
the Oxꢀ and p-ClPhOꢀ systems contributes to their differen-
tial TS stabilization in the DMSO-rich region, it follows that
the TS for the p-ClPhOꢀ system is more polarizable than
the TS for the Oxꢀ system. This is consistent with the above
proposal on the basis of the respective resonance structures
and relative rate constants that p-ClPhOꢀ is more polariza-
ble than Oxꢀ.
The bell-shaped a-effect profile found in the current
study (Figure 4) as well as in our previous studies,[17–20] has
revealed a differential GS/TS solvent effect, that is, a GS
effect in the H2O-rich region through H-bonding interac-
tions and a TS effect in the DMSO-rich media through
mutual polarizability interactions.[17–20]
4) The dominant factor influencing reactivity in the present
system is polarizability: a) p-ClPhOꢀ and Oxꢀ are up to
33 times more reactive toward the C=S compound 1 than
toward its oxygen analogue 2, whereas HOꢀ is 4–9 times
less reactive toward the former than the latter. b) En-
hanced polarizability of the electrophilic center is re-
sponsible for the contrasting reactivity as well as the dif-
ference in the a-effect profile for reactions of 1 and 2.
5) The bell-shaped a-effect profile obtained in this work
has provided unique opportunity for discussion of the
role of solvent with respect to GS and TS stabilization/
destabilization through H-bonding versus polarizability
interactions in the DMSO-H2O solvent system.
Recent gas-phase studies including high-level theoretical
calculations have shown that a-nucleophiles are intrinsically
more reactive, lower enthalpies of activation, than normal
nucleophiles of similar basicity (e.g., HOOꢀ vs. EtOꢀ).[14–16]
Patterson and Fountain have reported on the basis of a the-
oretical study for gas-phase reactions of methyl formate
with HOꢀ, EtOꢀ, and HOOꢀ that the a-nucleophile HOOꢀ
exhibits 3.6 kcalmolꢀ1 lower activation barrier than the gas-
phase-acidity-matched normal nucleophile EtOꢀ, as evi-
dence for a gas-phase a-effect.[14] They found that HOOꢀ
does not exhibit enhanced reactivity toward methyl formate
when compared with HOꢀ, which is much more basic than
HOOꢀ in the gas phase.[14] Yamataka et al. performed theo-
retical studies at the G2(+) level on gas-phase SN2 reactions
of alkyl halides with 11 anionic nucleophiles.[15] They found
that normal nucleophiles exhibit linear plots of the calculat-
ed activation barrier versus proton affinity for the reactions
with EtCl and iPrCl, while a-nucleophiles exhibit negative
deviations.[15] The negative deviations exhibited by a-nucleo-
philes were considered as evidence of a gas-phase a-effect
in gas-phase reactions.[15]
An interesting study by McAnoy et al. of the gas-phase
reactions of dimethyl methylphosphonate with CD3Oꢀ and
HOOꢀ anions in an ion-trap mass spectrometer showed four
parallel reactions (i.e., deprotonation which yields a carban-
ion, SN2 at carbon, nucleophilic substitution at phosphorus,
and a reductive elimination process), nucleophilic substitu-
tion at carbon predominant for the HOOꢀ reaction but
proton transfer dominating for CD3Oꢀ.[16] The difference in
the observed reactivities of the two nucleophiles was sug-
gested as evidence for an interesting a-effect in the gas-
phase.[16]
Experimental Section
Materials: Compound 1 was prepared as reported previously.[24,25] Butan-
2,3-dione monoxime and p-chlorophenol were recrystallized before use.
DMSO was distilled over CaH2 under reduced pressure just before use.
Other chemicals were of the highest quality available. Doubly glass dis-
tilled water was further boiled and cooled under nitrogen just before use.
Kinetic measurements: The kinetic study was performed with a UV/Vis
spectrophotometer for slow reactions (t1/2 ꢃ10 s) or with a stopped-flow
spectrophotometer for fast reactions (t1/2 <10 s) equipped with a constant
temperature circulating bath to maintain the temperature in the reaction
cell at (25.0ꢁ0.1)8C. The reaction was followed by monitoring the ap-
pearance of the leaving p-nitrophenoxide ion. All reactions were carried
out under pseudo-first-order conditions in which nucleophile concentra-
tions were at least 20 times greater than the substrate concentration. The
Oxꢀ and p-ClPhOꢀ stock solutions of ca. 0.2m were prepared by dissolv-
ing two equiv of OxH (or p-ClPhOH) and one equiv of standardized
NaOH solution to keep the pH constant in this self-buffered solution. All
solutions were prepared freshly just before use under nitrogen and trans-
ferred by gas-tight syringes. Typically, the reaction was initiated by
adding 5 mL of a 0.02m solution of the substrate in CH3CN by a 10 mL sy-
ringe to a 10 mm quartz UV cell containing 2.50 mL of the thermostatted
reaction mixture made up of solvent and aliquot of the nucleophile stock
solution.
Product analysis: p-Nitrophenoxide was liberated quantitatively and
identified as one of the products by comparison of the UV/Vis spectrum
at the end of reaction with the authentic sample under the experimental
condition.
Conclusion
Our study of the reactions of O-p-nitrophenyl thionoben-
zoate (1) with HOꢀ, p-ClPhOꢀ, and Oxꢀ has revealed major
reactivity differences on changing the electrophilic center
from C=O (2) to C=S (1):
Acknowledgements
The authors are grateful for the financial support from Korea Research
Foundation (KRF-2005-015-C00256) and from NSERC of Canada.
1016
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1011 – 1017

Reactivity Effects
FULL PAPER
Res. 1987, 20, 301–308; e) C. F. Bernasconi, C. J. Murray, J. Am.
Chem. Soc. 1986, 108, 5251–5257.
[11] a) S. Hoz, E. Buncel, Tetrahedron Lett. 1984, 25, 3411–3414; b) E.
Buncel, S. Hoz, Tetrahedron Lett. 1983, 24, 4777–4780; c) S. Hoz, J.
Org. Chem. 1982, 47, 3545–3547.
[12] R. Curci, F. Di Furia, Int. J. Chem. Kinet. 1975, 7, 341–349.
[13] a) C. H. Depuy, E. W. Della, J. Filley, J. J. Grabowski, V. M. Bier-
baum, J. Am. Chem. Soc. 1983, 105, 2481–2482; b) S. Wolfe, D. J.
Mitchell, H. B. Schlegel, C. Minot, O. Eisenstein, Tetrahedron Lett.
1982, 23, 615–618.
[1] J. O. Edwards, R. G. Pearson, J. Am. Chem. Soc. 1962, 84, 16–24.
[2] Reviews: a) E. Buncel, I. H. Um, Tetrahedron 2004, 60, 7801–7825;
b) S. Hoz, E. Buncel, Isr. J. Chem. 1985, 26, 313–319; c) A. P.
Grekov, V. Ya. Veselov, Russ. Chem. Rev. 1978, 47, 631–648; d) N. J.
Fina, J. O. Edwards, Int. J. Chem. Kinet. 1973, 5, 1–26.
[3] a) A. J. Kirby, M. F. Lima, D. da Silva, C. D. Roussev, F. Nome, J.
Am. Chem. Soc. 2006, 128, 16944–16952; b) A. J. Kirby, N. Dutta-
Roy, D. da Silva, J. M. Goodman, M. F. Lima, C. D. Roussev, F.
Nome, J. Am. Chem. Soc. 2005, 127, 7033–7040.
[4] a) F. Terrier, P. Rodriguez-Dafonte, E. Le Guevel, G. Moutiers, Org.
Biomol. Chem. 2006, 4, 4352–4363; b) F. Terrier, E. Le Guevel,
A. P. Chatrousse, G. Moutiers, E. Buncel, Chem. Commun. 2003,
600–601; c) E. Buncel, C. Cannes, A.-P. Chatrousse, F. Terrier, J.
Am. Chem. Soc. 2002, 124, 8766–8767; d) G. Moutiers, E. Le Gue-
vel, C. Cannes, F. Terrier, E. Buncel, Eur. J. Org. Chem. 2001, 17,
3279–3284.
[5] a) K. R. Fountain, J. Phys. Org. Chem. 2005, 18, 481–485; b) K. R.
Fountain, C. J. Felkerson, J. D. Driskell, B. D. Lamp, J. Org. Chem.
2003, 68, 1810–1814; c) K. R. Fountain, D. B. Tad-y, T. W. Paul,
M. V. Golynskiy, J. Org. Chem. 1999, 64, 6547–6553; d) K. R. Foun-
tain, K. D. Patel, J. Org. Chem. 1997, 62, 4795–4797; e) K. R. Foun-
tain, T. W. Dunkin, K. D. Patel, J. Org. Chem. 1997, 62, 2738–2741;
f) K. R. Fountain, R. D. White, K. D. Patel, D. G. New, Y. B. Xu,
A. J. Cassely, J. Org. Chem. 1996, 61, 9434–9436.
[6] a) K. K. Ghosh, D. Sinha, M. L. Satnami, D. K. Dubey, P. Rodri-
guez-Dafonte, G. L. Mundhara, Langmuir 2005, 21, 8664–8669;
b) A. Shrivastava, K. K. Ghosh, J. Mol. Liq. 2008, 141, 99–101.
[7] a) J. B. Domingos, E. Longhinotti, T. A. S. Brandao, L. S. Santos,
M. N. Eberlin, C. A. Bunton, F. Nome, J. Org. Chem. 2004, 69,
7898–7905; b) C. A. Bunton, F. Nome, F. H. Quina, L. S. Romsted,
Acc. Chem. Res. 1991, 24, 357–364; c) C. A. Bunton, M. M. Mhala,
J. R. Moffatt, J. Phys. Chem. 1989, 93, 854–858; d) C. A. Bunton,
L. H. Gan, J. R. Moffatt, L. S. Romsted, G. Savelli, J. Phys. Chem.
1981, 85, 4118–4125; e) C. A. Bunton, L. Sepulveda, J. Phys. Chem.
1979, 83, 680–683.
[14] E. V. Patterson, K. R. Fountain, J. Org. Chem. 2006, 71, 8121–8125.
[15] a) Y. Ren, H. Yamataka, J. Org. Chem. 2007, 72, 5660–5667; b) Y.
Ren, H. Yamataka, Chem. Eur. J. 2007, 13, 677–682; c) Y. Ren, H.
Yamataka, Org. Lett. 2006, 8, 119–121.
[16] A. M. McAnoy, M. R. Paine, S. J. Blanksby, Org. Biomol. Chem.
2008, 6, 2316–2326.
[17] a) E. Buncel, I. H. Um, Chem. Commun. 1986, 595–596; b) I. H.
Um, E. Buncel, J. Org. Chem. 2000, 65, 577–582.
[18] I. H. Um, S. J. Hwang, E. Buncel, J. Org. Chem. 2006, 71, 915–920.
[19] I. H. Um, Y. H. Shin, J. Y. Han, E. Buncel, Can. J. Chem. 2006, 84,
1550–1556.
[20] a) R. M. Tarkka, E. Buncel, J. Am. Chem. Soc. 1995, 117, 1503–
1507; b) I. H. Um, J. Y. Hong, E. Buncel, Chem. Commun. 2001, 27–
28.
[21] a) I. H. Um, Y. M. Park, E. Buncel, Chem. Commun. 2000, 1917–
1918; b) I. H. Um, E. J. Lee, E. Buncel, J. Org. Chem. 2001, 66,
4859–4864.
[22] I. H. Um, E. Buncel, J. Am. Chem. Soc. 2001, 123, 11111–11112.
[23] I. H. Um, H. J. Han, J. A. Ahn, S. Kang, E. Buncel, J. Org. Chem.
2002, 67, 8475–8480.
[24] I. H. Um, J. Y. Lee, H. T. Kim, S. K. Bae, J. Org. Chem. 2004, 69,
2436–2441.
[25] I. H. Um, S. E. Lee, H. J. Kwon, J. Org. Chem. 2002, 67, 8999–9005.
[26] I. H. Um, J. Y. Lee, S. Y. Bae, E. Buncel, Can. J. Chem. 2005, 83,
1365–1371.
[27] B. S. Pedersen, S. Scheibye, N. H. Nilsson, S. O. Lawesson, Bull. Soc.
Chim. Belg. 1978, 87, 223–228.
[28] a) W. J. Middleton, E. G. Howard, W. H. Sharkey, J. Org. Chem.
1965, 30, 1375–1384; b) J. F. Harris, F. W. Stacey, J. Am. Chem. Soc.
1963, 85, 749–754.
[8] a) M. J. Gregory, T. C. Bruice, J. Am. Chem. Soc. 1967, 89, 4400–
4405; b) J. E. Dixon, T. C. Bruice, J. Am. Chem. Soc. 1972, 94, 2052–
2056; c) J. E. Dixon, T. C. Bruice, J. Am. Chem. Soc. 1971, 93, 6592–
6597.
[9] a) W. P. Jencks, Catalysis in Chemistry and Enzymology, McGraw-
Hill, New York, 1969, pp. 107–111; b) D. Herschlag, W. P. Jencks, J.
Am. Chem. Soc. 1990, 112, 1951–1956; c) W. P. Jencks, Chem. Rev.
1985, 85, 511–526; d) W. P. Jencks, M. Gilchrist, J. Am. Chem. Soc.
1968, 90, 2622–2637; e) W. P. Jencks, J. Am. Chem. Soc. 1958, 80,
4585–4588.
[29] T. L. Ho in Hard and soft acid and bases (Ed.: R. G. Pearson), Aca-
demic Press, New York, 1977.
[30] a) A. J. Parker, Chem. Rev. 1969, 69, 1–32; b) C. D. Ritchie, R. E.
Uschold, J. Am. Chem. Soc. 1968, 90, 2821–2824; c) C. Reichardt,
Solvents and Solvent Effects in Organic Chemistry, 3rd ed., VCH,
New York, 2003.
[10] a) C. F. Bernasconi, A. E. Leyes, I. Eventova, Z. Rappoport, J. Am.
Chem. Soc. 1995, 117, 1703–1711; b) C. F. Bernasconi, Adv. Phys.
Org. Chem. 1992, 27, 119–238; c) C. F. Bernasconi, M. W. Stronach,
J. Org. Chem. 1991, 56, 1993–2001; d) C. F. Bernasconi, Acc. Chem.
[31] E. Buncel, H. Wilson, Adv. Phys. Org. Chem. 1977, 14, 133–202.
Received: July 28, 2008
Published online: December 9, 2008
Chem. Eur. J. 2009, 15, 1011 – 1017
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1017