Solvolytic reactivity of 2,4-dinitrophenolates

Article abstract of DOI:10.1002/ejoc.201000784

A series of X,Y-substituted benzhydryl 2,4-dinitrophenolates (DNP, 1-5) were subjected to solvolysis in various methanol/water, ethanol/water, and acetone/water mixtures at 25 °C. The linear free energy relationship (LFER) equation, logk = s_f(E_f + N_f), was used to derive the nucleofuge-specific parameters (N_f and s_f) for an S_N1-type reaction. The magnitudes of nucleofugalities (N _f) are around zero, indicating that DNP falls in the middle of the established nucleofugality scale. The slope parameters (s_f) and the Gruenwald-Winstein m_OTs parameters obtained demonstrate that benzhydryl DNPs solvolyze through a late transition state (TS) in which the negative charge delocalization causes considerably diminished solvation. Because of the late TS, the nucleofuge-specific slope parameters, s_f, are relatively high, i.e., the logk vs. E_f plots are steeper than for most of the previously investigated leaving groups. This may lead to intersection of the logk vs. E_f plots that correspond to DNP and to some other leaving groups of similar reactivity, i.e., inversion of the relative reactivities may occur. Such inversion is shown here for DNPs and phenyl carbonates. The nucleofuge-specific parameters determined for 2,4-dinitrophenolate (DNP) according to LFER equation logk = s_f (N_f + E_f) indicate that DNPs solvolyze via late TS in which intense charge delocalization occurs. DNPs with weaker electrofuges solvolyze slower than the same substrates with leaving groups of similar reactivity but lower s_f, whereas strong electrofuges solvolyze faster.

Full text of DOI:10.1002/ejoc.201000784

FULL PAPER
DOI: 10.1002/ejoc.201000784
Solvolytic Reactivity of 2,4-Dinitrophenolates
[a]
[a]
[a]
Mirela Mati c´ , Bernard Denegri,* and Olga Kronja*
Keywords: Reaction mechanisms / Solvent effects / Solvolysis / Nucleofugality scale
A series of X,Y-substituted benzhydryl 2,4-dinitrophenolates
DNP, 1–5) were subjected to solvolysis in various methanol/
water, ethanol/water, and acetone/water mixtures at 25 °C.
volyze through a late transition state (TS) in which the nega-
tive charge delocalization causes considerably diminished
solvation. Because of the late TS, the nucleofuge-specific
(
The linear free energy relationship (LFER) equation, logk =
f f
slope parameters, s , are relatively high, i.e., the logk vs. E
s
f
(E
eters (N
nucleofugalities (N
falls in the middle of the established nucleofugality scale.
The slope parameters (s ) and the Grünwald–Winstein mOTs
f
+ N
f
), was used to derive the nucleofuge-specific param-
and s ) for an S 1-type reaction. The magnitudes of
) are around zero, indicating that DNP
plots are steeper than for most of the previously investigated
leaving groups. This may lead to intersection of the logk vs.
E plots that correspond to DNP and to some other leaving
f
groups of similar reactivity, i.e., inversion of the relative reac-
tivities may occur. Such inversion is shown here for DNPs
and phenyl carbonates.
f
f
N
f
f
parameters obtained demonstrate that benzhydryl DNPs sol-
ates[3] have already been determined. The latter leaving
groups are produced by cleavage of the C–OC(O) bond in
the heterolysis step in solvolysis. Phenolates, in which cleav-
age of the ethereal C–OR bond occurs in heterolysis, have
also been used as leaving groups in solvolytic reactions. For
example, adamantyl picrate was studied to determine the
Introduction
In our previous work we compared the leaving group
(LG) abilities of various groups by using the recently devel-
oped electrofugality/nucleofugality scale based on solvolysis
of benzhydryl derivatives.[ According to that approach, the
1]
heterolysis rate constant for any S 1 solvolysis reaction can
be expressed with the three-parameter LFER equation (1)
N
[4]
solvent effects on leaving groups, and 2,4-dinitrophenol-
ate was used to compare the behavior of tertiary and sec-
[
5]
logk = s
f
(E
f
+ N
f
)
(1) ondary cation fragments. Phenoxide was chosen to be the
next leaving group to be placed on the nucleofugality scale,
–
1
in which k is the first-order rate constant (s ) at 25 °C, s
is the nucleofuge-specific slope parameter, N is the nucleo-
fugality parameter, and E is the electrofugality parameter.
The latter parameter is set up as an independent variable
and refers to ability of the substrate moiety to leave as a
carbocation in the heterolysis reaction (S 1). Since the rela-
tive nucleofugalities depend not only on the substrate but
also on the solvent, the nucleofuge-specific parameters (sf
f
so that its reactivity could be compared with other leaving
f
groups, particularly with those in which cleavage of the C–
O bond also takes place. We intended to determine the nu-
cleofuge-specific parameters (N and s ) for 2,4-dinitro-
f
f
f
phenolate (DNP) using its benzhydryl derivatives as sub-
strates. DNP was the leaving group of choice here because
of the convenient reactivity and solubility of its benzhydryl
derivatives in the series of aqueous solvents.
N
and N ) characterize the leaving group in a given solvent.
f
Such an approach separates the contributions of electrofuge
and nucleofuge in overall solvolytic reactivity. Predefined Results and Discussion
parameters are: E = 0.00 for dianisylcarbenium electrofuge
f
A series of benzhydryl DNPs (1–5), which were prepared
from the corresponding benzhydrols, were subjected to sol-
volysis in aqueous solvents.
(
X = Y = 4-OCH ) and s = 1.00 for chloride nucleofuge in
3 f
[
1a]
pure ethanol. According to equation (1), the nucleofugal-
ity (N ) of the given leaving group is defined as the negative
f
intercept on the abscissa of logk vs. E plot.
f
The nucleofugality parameters in various solvents of
some frequently used halides, carboxylates,[1,2] and carbon-
[
a] Faculty of Pharmacy and Biochemistry University of Zagreb,
Ante Kova cˇ i c´ a 1, 10000 Zagreb, Croatia
Fax: +385-1-48-56-201
E-mail: kronja@pharma.hr
bdenegri@pharma.hr
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000784.
View this journal online at
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 6019–6024
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6019

M. Mati c´ , B. Denegri, O. Kronja
FULL PAPER
The solvolysis rates were measured conductometrically at
To calculate the nucleofugality parameters (N ) and the
f
2
5 °C. In a few cases the rates were measured at three dif- slope parameters (s ) for DNP in various solvents, loga-
f
ferent temperatures and extrapolated to 25 °C. Details are rithms of the first-order rate constants in the given solvents
given in Kinetic methods in the Supporting Information. were plotted against E . The values of E for all the electro-
f
f
[
1a]
The first-order rate constants at 25 °C (measured and ex- fuges used have previously been determined.
Plots of
trapolated) are presented in Table 1.
logk against E obtained for acetone/water binary solvents
f
are presented in Figure 1 (for correlation lines obtained in
aqueous ethanol and aqueous methanol see the Supporting
Information). The nucleofuge-specific parameters were cal-
culated according to Equation (1) and are presented in
Table 2.
Table 1. Solvolysis rate constants of X,Y-substituted benzhydryl
2,4-dinitrophenolates in various aqueous solvents at 25 °C.
Solvent[a] Substrate (X, Y)
[b]
–1 [c]
E
f
k [s ]
1
00M
4 (4-Me, H)
–4.68
–3.47
–2.06
–1.29
–4.68
–3.47
–2.06
–4.68
–3.47
–2.06
–4.68
–3.47
–2.06
–1.29
–4.68
–3.47
–2.06
–1.29
–5.78
–4.68
–3.47
–2.06
–5.78
–4.68
–3.47
–2.06
–3.47
–2.06
–1.29
–3.47
–2.06
–1.29
–4.68
–3.47
–2.06
–1.29
–4.68
–3.47
–2.06
–1.29
–4.68
–3.47
–2.06
8.20ϫ10–6[d,e]
–
–
–
–
–
–
–
–
–
4
3
2
5
4
2
5
3
2
3
2
1
(4-Me, 4Ј-Me)
(4-OMe, H)
(4-OMe, 4Ј-Me)
(1.81Ϯ0.04)ϫ10
(4.15Ϯ0.03)ϫ10
(2.86Ϯ0.04)ϫ10
(3.04Ϯ0.07)ϫ10
(5.70Ϯ0.10)ϫ10
(1.04Ϯ0.02)ϫ10
(7.94Ϯ0.13)ϫ10
(1.44Ϯ0.02)ϫ10
(2.35Ϯ0.03)ϫ10
1.78ϫ10–
9
8
1
0M10W 4 (4-Me, H)
3
2
(4-Me, 4Ј-Me)
(4-OMe, H)
0M20W 4 (4-Me, H)
3
2
(4-Me, 4Ј-Me)
(4-OMe, H)
4 (4-Me, H)
6[d,f]
00E
–5
–4
–3
3
2
1
(4-Me, 4Ј-Me)
(4-OMe, H)
(4-OMe, 4Ј-Me)
(3.23Ϯ0.03)ϫ10
(9.76Ϯ0.21)ϫ10
(7.26Ϯ0.08)ϫ10
1.22ϫ10–
5[d,g]
9
8
7
0E10W 4 (4-Me, H)
–4
–3
–2
3
2
1
(4-Me, 4Ј-Me)
(4-OMe, H)
(4-OMe, 4Ј-Me)
(1.91Ϯ0.05)ϫ10
(5.05Ϯ0.12)ϫ10
(3.45Ϯ0.06)ϫ10
1.75ϫ10–
6[d,h]
0E20W 5 (4-F, H)
–5
–4
–2
Figure 1. Plots of logk vs. E for solvolysis of X,Y-substituted benz-
f
4
3
2
(4-Me, H)
(4-Me, 4Ј-Me)
(4-OMe, H)
(2.89Ϯ0.05)ϫ10
(5.23Ϯ0.06)ϫ10
(1.17Ϯ0.21)ϫ10
3.90ϫ10–
hydryl 2,4-dinitrophenolates in aqueous acetone. Solvent mixtures
are given as v/v ratios; A = acetone and W = water.
6[d,i]
0E30W 5 (4-F, H)
–5
–3
–2
4
3
2
(4-Me, H)
(4-Me, 4Ј-Me)
(4-OMe, H)
(6.08Ϯ0.07)ϫ10
(1.02Ϯ0.02)ϫ10
(1.82Ϯ0.04)ϫ10
9.65ϫ10–
6[d,j]
9
8
7
0A10W 3 (4-Me, 4Ј-Me)
Table 2. Nucleofugality parameters N
f
f
and s for 2,4-dinitrophenol-
–4
–3
–5
–3
–2
2
1
(4-OMe, H)
(4-OMe, 4Ј-Me)
(4.52Ϯ0.12)ϫ10
(3.16Ϯ0.03)ϫ10
(4.13Ϯ0.04)ϫ10
(1.50Ϯ0.03)ϫ10
(1.00Ϯ0.02)ϫ10
6.78ϫ10–
ate in various solvents.
Solvent[a]
Nf[b]
sf[b]
[c] (n)[d]
R
0A20W 3 (4-Me, 4Ј-Me)
2
1
(4-OMe, H)
(4-OMe, 4Ј-Me)
1
9
8
1
9
8
7
9
8
7
6
5
00M
–0.22Ϯ0.07
0.04Ϯ0.17
0.37Ϯ0.18
–0.75Ϯ0.04
–0.18Ϯ0.04
0.22Ϯ0.09
0.36Ϯ0.13
–0.85Ϯ0.06
–0.53Ϯ0.03
–0.37Ϯ0.03
–0.14Ϯ0.05
–0.03Ϯ0.12
1.03Ϯ0.02
0.97Ϯ0.05
0.94Ϯ0.05
1.06Ϯ0.01
1.02Ϯ0.01
1.03Ϯ0.02
0.99Ϯ0.03
1.16Ϯ0.02
1.10Ϯ0.01
1.02Ϯ0.01
0.98Ϯ0.01
0.91Ϯ0.03
0.9995 (4)
0.9989 (3)
0.9985 (3)
1.0000 (4)
0.9999 (4)
0.9995 (4)
0.9989 (4)
0.9998 (3)
1.0000 (3)
0.9999 (4)
0.9998 (4)
0.9994 (3)
0M10W
0M20W
00E
6[d,k]
0A30W 4 (4-Me, H)
–4
–3
–2
–5
–4
–3
–2
–5
–4
–2
3
2
1
(4-Me, 4Ј-Me)
(4-OMe, H)
(4-OMe, 4Ј-Me)
(1.23Ϯ0.02)ϫ10
(3.46Ϯ0.06)ϫ10
(1.93Ϯ0.04)ϫ10
(1.77Ϯ0.05)ϫ10
(3.15Ϯ0.05)ϫ10
(7.29Ϯ0.07)ϫ10
(3.76Ϯ0.09)ϫ10
(4.97Ϯ0.09)ϫ10
(7.34Ϯ0.23)ϫ10
(1.21Ϯ0.04)ϫ10
0E10W
0E20W
0E30W
0A10W
0A20W
0A30W
0A40W
0A50W
6
5
0A40W 4 (4-Me, H)
3
2
1
(4-Me, 4Ј-Me)
(4-OMe, H)
(4-OMe, 4Ј-Me)
0A50W 4 (4-Me, H)
3
2
(4-Me, 4Ј-Me)
(4-OMe, H)
[
a] Binary solvents (v/v) at 25 °C. A = acetone, E = ethanol, M =
[
a] Binary solvents are given v/v at 25 °C. A = acetone, E = eth-
methanol, W = water. [b] Errors shown are standard errors. [c]
Correlation coefficient. [d] Number of data points.
anol, M = methanol, W = water. [b] Electrofugality parameters are
taken from Denegri et al.[
1a]
[c] Average rate constants from
at least three runs performed at 25 °C. Errors are standard de-
viations. [d] Extrapolated from data at higher temperatures by
‡
–1
The nucleofugality parameters presented in Table 2 are
using the Eyring equation. [e] ∆H
=
=
107.1Ϯ2.7 kJmol ,
‡
‡
16.8Ϯ8.4 JK mol . [f] ∆H‡
–1
–1
–1
∆S
S
=
107.9Ϯ0.0 kJmol , around zero in all solvents so, accordingly, DNP falls in the
–1
–1
‡
–1
‡
∆
= 6.9Ϯ1.2 JK mol . [g] ∆H = 97.6Ϯ1.0 kJmol , ∆S
=
=
=
=
=
middle of the established nucleofugality scale, above phenyl
carbonate and below fluorinated carboxylates, heptafluoro-
butyrate, and trifluoroacetate. The nucleofugalities of vari-
ous leaving groups in 80% aq. ethanol and 80% aq. meth-
anol are presented in Figure 2.
–1
–1
‡
–1
‡
–
–
–
0
1
11.6Ϯ3.0 JK1 mol . [h] ∆H = 104.7Ϯ0.8 kJmol , ∆S
3.9Ϯ2.4 JK– mol . [i] ∆H‡
–1
‡
=
=
=
103.4Ϯ0.6 kJmol , ∆S
–1
‡
–
1
–1
–1
‡
1.7Ϯ1.9 JK mol . [j] ∆H
101.7Ϯ1.3 kJmol , ∆S
.1Ϯ4.2 JK mol . [k] ∆H‡
–
1
–1
–1
‡
106.6Ϯ2.7 kJmol , ∆S
3.8Ϯ8.4 JK–1 mol .
–1
6020
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6019–6024

Solvolytic Reactivity of 2,4-Dinitrophenolates
Accordingly, for more reactive substrates DNPs are more
reactive than phenyl carbonates, but for less reactive sub-
strates, i.e., those with weaker electrofuges, such as ada-
[
1c]
mantyl, tert-butyl, or 1-phenylethyl, the carbonates may
be slightly more reactive than the corresponding DNP. It
should be emphasized that the intersection of the plots is
reliable for two reasons: (1) it is not obtained after far ex-
trapolation, but it is in the range of experimental data, and
(2) the correlation of experimental data for both plots in
Figure 3 are excellent (R Ͼ 0.999). Thus, for example, it can
be predicted that 3,3Ј-dichlorobenzhydryl DNP (E = –9.63)
f
[
1c]
solvolyzes slower than the corresponding phenyl carbon-
ate in 80% ethanol, while, for example, dianisyl-
methyl DNP (E = 0.00) is more reactive than the corre-
f
sponding phenyl carbonate.
Inversion of the relative rates of the reactions of phenyl
carbonates and DNPs, as is presented in Figure 3, demon-
strates another advantage of the above three-parameter
LFER approach [Equation (1)] over the use of a single
structure with different LGs. Thus, besides the possibility
for covering a much wider range of reactivities, the cases
where inversion of the reactivity between the substrates with
different leaving groups occurs can also be predicted, which
is not possible if the investigations are carried out with sub-
Figure 2. Nucleofugalities and the nucleofuge-specific slope param-
eters (in parentheses) for some leaving groups in 80% aqueous eth-
anol and 80% aqueous methanol.
Discussion on the Nucleofuge-Specific
Parameters of DNP
The values for the nucleofugality parameters N usually strates in which only the leaving groups differ.
f
give correct information on the relative reactivities of the
The values of the slope parameters (s ) obtained for DNP
f
leaving groups (LGs). However, if the nucleofugalities of in various solvents are generally close to unity, as are the sf
two LGs are close in magnitude, while the s parameters parameters obtained with chlorides and bromides. Whereas
f
differ substantially, a simple comparison of nucleofugalities in aqueous acetone the slopes significantly decrease as the
may be misleading because of the possible intersection of polarity of the solvent increases, in aqueous alcohols the
the logk vs. E plots. In such cases, for correct prediction of trend is so mild that the decrease in the s values is close to
f
f
the relative reactivities, both parameters should be taken in the limits of the experimental error (Table 2). A decrease in
account, the nucleofugality and the slope parameter, i.e., the slope parameters with increasing polarity of the solvent
Equation (1) should be applied. This is particularly impor- has previously been attributed to diminished solvation ef-
tant if the LG with a higher N value produces a lower slope fects in the TS caused by additional charge delocalization
f
[
2a]
because, in such cases, the intersection of the logk vs. Ef in the leaving group moiety.
slopes might correspond to electrofuges that are, with an
To extract the Grünwald–Winstein m values and to com-
appropriate LG, stable structures that solvolyze in the range pare the sensitivity of DNP toward solvent polarity with
of experimental reactivities or in their close vicinity. Fig- other leaving groups, we plotted logarithms of rate con-
ure 3 shows the logk vs. E plots for phenyl carbonate stants of benzhydryl DNPs and of the corresponding chlo-
f
[
6]
(
N = –0.84, s = 0.87) and DNP (N = 0.22, s = 1.03) ride in aqueous ethanol, aqueous methanol, and aqueous
f f f f
[
7]
obtained in 80% aqueous ethanol.
acetone, against the ionizing power (YOTs). The previously
determined mOTs values for phenyl carbonates were also
taken for comparison.
[
2a]
The chloride here represents the
reference compound in which electron delocalization in the
LG does not exist, whereas the phenyl carbonate represents
the LG in which the negative charge delocalization onto
three oxygen atoms occurs in the TS, due to resonance and
[
3]
inverse hyperconjugation. Some mOTs parameters are pre-
sented in Table 3.
Two major factors that determine the values of
Grünwald–Winstein m parameters for limiting solvolysis
presented here are solvation effects and the position of the
TS; thus, an earlier TS and diminished solvation caused by
[
4,8,9]
charge delocalization reduce the magnitude of m.
The
sf parameters are, however, mostly an indication of the posi-
tive charge generated in the TS, similar to the Hammett–
Figure 3. Comparison of logk vs. E
and phenyl carbonate in 80% aq. ethanol at 25 °C.
f
plots for 2,4-dinitrophenolate
+
[10,2a]
Brown ρ parameter.
Relatively high s values for
f
Eur. J. Org. Chem. 2010, 6019–6024
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6021

M. Mati c´ , B. Denegri, O. Kronja
FULL PAPER
Table 3. Values of mOTs parameters from the Grünwald–Winstein logk vs. E plots (Figure 1 and Table 2), i.e., a reduction in
f
correlations for solvolysis of X,Y-substituted benzhydryl chlorides
the s parameters with an increase in the fraction of water
f
(Cl), dinitrophenolates (DNP), and phenyl carbonates (PhCarb).
[2a]
in acetone.
Low m values have usually been coupled with relatively
Solvent[a]
LG
X,Y-Substituents on the benzhydryl ring
MeO, Me MeO, H Me, Me Me, H
[12]
‡
positive entropy of activation. Generally, the ∆S param-
E–W
Cl[b]
DNP
0.78
0.62
eters obtained for DNPs are higher than those for chlo-
0.50
0.50
0.62
0.64
[13]
[2a]
rides and phenyl carbonates. For example, in pure eth-
[
c]
PhCarb
0.45
0.54
‡
–1
–1
anol, ∆S = +6.9Ϯ1.2 JK mol for 4, whereas for the
[b]
M–W
A–W
Cl
0.86
0.71
corresponding chloride ∆S‡ = –15.2Ϯ9.1 JK mol .
The results are consistent with intense charge delocalization
of the negative charge generating in the TS of DNPs, which
leads to diminished solvation and therefore to less loss of
freedom of the solvent molecules in comparison with
phenyl carbonates, and particularly to chlorides.
–1
–1 [13]
DNP
0.55
0.56
0.64
0.67
[d]
PhCarb
[b]
Cl
1.13
0.73
DNP
PhCarb
0.39
0.70
0.44
0.74
0.57
0.92
[c]
[
a] A = acetone, E = ethanol, M = methanol, W = water. [b] The
m
OTs values were calculated from rate constants published by Liu
[
6]
et al. For details see the Supporting Information. [c] The mOTs
values were taken from reference 2a. [d] The mOTs values were cal-
[3]
culated from rate constants published in ref. . For details see the Consideration of the Energy Profile in Solvolysis
Supporting Information.
of DNPs
Nucleofugality has been linked to the basicity of the leav-
DNPs (and for chlorides) are an indication of a late TS.
ing groups in that a weaker base constitutes a better nucleo-
fuge. Even though a reasonably good correlation between
Lower mOTs values for DNP than those for the correspond-
ing chlorides are associated with the ability of DNP to bet-
ter disperse the developing negative charge and hence di-
the reaction rates and pK values of the conjugated acids
a
of the LGs for numerous substrates have been demon-
minish the solvation effect in the late TS. Table 3 shows that
[
14]
strated,
some authors have stated that reactivity shows
the mOTs obtained for phenyl carbonates and DNP in aque-
ous alcohols are virtually the same. On the other hand, the
no correlation when wide ranges of diverse structures are
examined, but exists only if the variations in the LGs are
s values for DNPs are higher than those for phenyl carbon-
f
[
15]
ates in all solvents examined (Figure 2).[ This discrepancy
indicates that, for phenyl carbonates, both the earlier TS
and the delocalized partial negative charge diminish the
mOTs parameter in comparison to chlorides, whereas re-
duction of the mOTs values for DNP mainly arise from the
intense charge delocalization in the LG moiety of the late
TS. Consequently, the solvation effect is less important for
DNP than for phenyl carbonate, which is not surprising
considering that two nitro groups on the phenyl ring take
part in accommodating the developing negative charge in
the TS by resonance.
3]
small.
Figure 4a shows the correlation between the nucleofugal-
ities (N ) of some leaving groups determined in 80% aq.
f
ethanol, and the pK values for their corresponding conju-
a
The solvolytic behavior of DNPs in aqueous acetone is
somewhat different to that in aqueous alcohols. Firstly, the
decrease in the slope parameter s with increasing polarity
f
of the solvent is much more pronounced (Figure 1 and
Table 2), and secondly, the mOTs parameters are consider-
ably lower than those for phenyl carbonate in aqueous ace-
tones, whereas the s parameters are still high. This behavior
f
can be attributed to the lipophilicity of the developing 2,4-
dinitrophenolate anion and hence to a more important role
[
4,11,12a]
of the lipophilic acetone in the solvation of the TS.
This additional solvation, which does not occur in aqueous
alcohols, reduces the mOTs further for DNP. The same phe-
nomenon may be responsible for the decrease in the slope
parameter s in more polar solvents. Increasing electrofugal-
f
ity in the series of benzhydryl DNPs leads to more extensive
positive charge delocalization and thus less stabilization of
the activated complex by more polar solvents. Thus, sol-
vents containing a higher fraction of water enhance the rate
more for substrates with a weak electrofuge than those with
Figure 4. Plots of (a) N
0% aq. ethanol for some leaving groups (DNP and MeCarb are
not included in the correlation) (the pK values and the corre-
f a a
vs. pK and (b) logk (25 °C) vs. pK in
8
a
a strong electrofuge. The net result is convergence of the sponding references are given in the Supporting Information).
6
022 Eur. J. Org. Chem. 2010, 6019–6024
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Solvolytic Reactivity of 2,4-Dinitrophenolates
gated acids obtained in water; Figure 4b shows the corre- solvolyze through a later TS than carboxylates, producing
lation between logk at 25 °C (measured or calculated) for logk vs. E plots with relatively high slope parameters.
f
unsubstituted benzhydryl derivatives with the same leaving
groups obtained in 80% aq. ethanol and the pK values.
a
The plots show that the correlation between the reactivity Experimental Section
and acidity exists only within the same family of com-
Substrate Preparation: The substrates from the corresponding
pounds. Thus, whereas the N values of unsubstituted benz-
f
benzhydroles were prepared according to the substantially modified
hydryl carboxylates correlate well without exception with
[17]
procedure presented previously.
the pK values of the corresponding protonated LG (R =
a
4
-Methoxybenzhydrol, 4,4Ј-Dimethylbenzhydrol, 4-Methylbenz-
0
.997), DNP turned out to be considerably more reactive
than 3,5-dinitrobenzoate (DNB) and p-nitrobenzoate
PNB) even though both of the latter anions are weaker
hydrol, 4-Fluorobenzhydrol: Prepared by reduction of the commer-
cially available substituted benzophenones with sodium boro-
hydride in methanol.
(
bases than the DNP anion. It should be emphasized that
nucleofugality is a kinetic term that is derived from the rate
constant, whereas acidity is a thermodynamic term derived
from the equilibrium constant. Within the same group of
compounds, for example carboxylates here, stabilizing ef-
fects in the TS and in the ground states are similar, so the
4
-Methoxy-4Ј-methylbenzhydrol: Prepared according to the pro-
[3]
cedure given in ref.
-Fluorobenzhydryl 2,4-Dinitrophenyl Ether: Freshly cut potassium
1.0 g, 24.7 mmol) was added to a previously prepared stirring solu-
tion of 4-fluorobenzhydrol (5.0 g, 24.7 mmol) in anhydrous benzene
30 mL), and the solution was stirred for 2 h under an atmosphere
4
(
(
rates increase with decreasing pK values. However, if the
a
of argon in an ice-cold bath. A solution of 1-fluoro-2,4-dinitroben-
zene (9.2 g, 49.4 mmol) in benzene (10 mL) was then added drop-
structure of the leaving group varies considerably, inversion
between rate and basicity may occur. The experimental data wise with vigorously stirring and reaction mixture was stirred fur-
show that DNP is an example of such a leaving group so ther for 1 h. The brown precipitate was filtered off and benzene
its anion is less stabilized than DNB and PNB anions in was evaporated in vacuo to give a pale-yellow oil. The crude prod-
uct was dissolved in diethyl ether (ca. 30 mL) and then about
water, and solvolyzes with a lower barrier than DNB.
50 mL of concd. aq. NaOH was added. The mixture was stirred
Based on the approximation that the two thermodynamic
terms, the endothermicity of formation of the carbocation/
leaving group pair in the initial step of solvolysis, and that
of deprotonation of the acid and formation the proton/leav-
for 12 h, then the organic layer was separated and washed with
water. After drying over anhydrous sodium sulfate, the solvent was
removed in vacuo to give a pale-yellow oil. Recrystallization from
diethyl ether/light petroleum (4:1) afforded pale-yellow crystals
ing group (base) pair are proportional, the pK values re-
1
a
3
(0.30 g, 30%). H NMR (300 MHz, CDCl ): δ = 6.48 (s, 1 H,
flect the relative stability of the LGs. Thus, the plot in Fig-
2 2 2
Ar CH), 7.07–7.48 [m, 10 H, ArH + (O N) ArH], 8.29 [dd, J =
ure 4b essentially presents the correlation between the rela- 2.8, 9.3 Hz, 1 H, (O N) ArH], 8.75 [d, J = 2.8 Hz, 1 H, (O N)
2
2
2
1
3
tive energies of the transition states and the carbocation/
2
ArH] ppm. C NMR (75 MHz, CDCl
3
): δ = 83.9 (Ar
2
CH), 116.0,
leaving group pairs. The fact that DNP solvolyzes faster 116.3, 116.4, 121.7, 126.5, 128.4, 128.6, 128.9, 129.3, 135.1, 138.8,
1
6
55.5, 161.4, 164.6 (Ar) ppm. C19
1.96, H 3.56, N 7.60; found C 62.66, H 3.70, N 7.56.
2 5
H13FN O (368.29): calcd. C
than it would be predicted from the stability of the LG,
indicates that DNPs solvolyze over a lower intrinsic barrier
than carboxylates. That can be, according to principle of 4-Methylbenzhydryl 2,4-Dinitrophenyl Ether, 4,4Ј-Dimethylbenz-
nonperfect synchronization, attributed to a less pronounced hydryl 2,4-Dinitrophenyl Ether, and 4-Methoxybenzhydryl 2,4-Dini-
trophenyl Ether: Prepared as pale-yellow crystals according to the
lag of charge delocalization behind the C–O bond cleavage
in the TS of phenolates.[16] Reactions that proceed through
a lower barrier, producing intermediates with higher energy,
as is the case with DNPs here, have later transition states. 4-Methylbenzhydryl 2,4-Dinitrophenyl Ether: H NMR (300 MHz,
The same conclusion regarding the later TS, in which the
charge separation is already advanced, was drawn above on
procedure described for 4-fluorobenzhydryl 2,4-dinitrophenyl ether,
yielding 30–40% of the desired products.
1
CDCl
3
): δ = 2.30 (s, 3 H, Ar-CH
N) ArH], 8.22 [dd, J = 2.8, 9.3 Hz, 1 H,
ArH], 8.70 [d, J = 2.8 Hz, 1 H, (O
75 MHz, CDCl ): δ = 20.8 (ArCH ), 83.7 (Ar
25.8, 125.9, 128.0, 128.1, 128.6, 129.3, 135.4, 138.1, 138.6, 155.0
Ar) ppm. C20 (364.33): calcd. C 65.93, H 4.43, N 7.69;
found C 65.89, H 4.57, N 7.58.
3 2
), 6.44 (s, 1 H, Ar CH), 7.15–7.46
[m, 10 H, ArH + (O
2
2
3
(
O
2
N)
2
2 2
N)
ArH] ppm. 1 C NMR
the basis of the relatively high s parameters.
f
(
1
(
3
3
2
CH), 115.8, 121.3,
According to Figure 4, a similar inversion of the reactiv-
ities can be observed between methyl carbonates and acet-
ates in 80% aq. ethanol, indicating that benzhydryl methyl
carbonate solvolyzes over a lower intrinsic barrier than
acetate, which is the least reactive leaving group so far
16 2 5
H N O
4
,4Ј-Dimethylbenzhydryl 2,4-Dinitrophenyl Ether: 1H NMR
300 MHz, CDCl ): δ = 2.31 (s, 6 H, Ar-CH ), 6.42 (s, 1 H,
CH), 7.15–7.19 [m, 5 H, ArH + (O N) ArH], 7.34 (d, J =
N) ArH],
8.72 [d, J = 2.8 Hz, 1 H, (O N) ArH] ppm. C NMR (75 MHz,
(
3
3
placed on the nucleofugality scale. The s values obtained
f
Ar
2
2
2
for methyl carbonates are also relatively high and are sim-
8
.0 Hz, 4 H, ArH), 8.25 [dd, J = 2.8, 9.3 Hz, 1 H, (O
2
2
ilar to those obtained for DNP (s = 0.99 in 80% aq. ethan-
13
f
2
2
ol), indicating that methyl carbonates also proceed via a late CDCl ): δ = 21.1 (ArCH ), 84.1 (Ar CH), 116.4, 121.7, 126.3,
3
3
2
TS in the rate-determining heterolysis.
18 2 5
128.6, 129.7, 136.2, 138.5, 155.7 (Ar) ppm. C21H N O (378.35):
In conclusion, solvolytic behavior of DNPs is somewhat calcd. C 66.66, H 4.80, N 7.40; found C 66.91, H 5.08, N 7.24.
different than those of carboxylates, although C–O cleavage
1H
4
-Methoxybenzhydryl 2,4-Dinitrophenyl Ether:
NMR
occurs in both cases during the heterolysis process. X,Y- (300 MHz, CDCl ): δ = 3.76 (s, 3 H, Ar-OCH ), 6.44 (s, 1 H,
3
3
Substituted derivatives of benzhydryl 2,4-dinitrophenolates Ar
2
CH), 6.88 (d, J = 8.7 Hz, 2 H, ArH), 7.18–7.46 [m, 8 H, ArH
6023
Eur. J. Org. Chem. 2010, 6019–6024
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org

M. Mati c´ , B. Denegri, O. Kronja
FULL PAPER
+
(O
J = 2.8 Hz, 1 H, (O
54.8 (ArOCH ), 83.5 (Ar
28.0, 128.1, 128.7, 130.4, 138.7, 155.0, 159.5 (Ar) ppm.
(380.33): calcd. C 63.16, H 4.24, N 7.36; found C
2
N)
2
ArH], 8.24 [dd, J = 2.7, 9.3 Hz, 1 H, (O
2
N)
2
ArH], 8.70 [d, presented in Table 4. Calibration showed a linear response of con-
N)
2
ArH] ppm. 13C NMR (75 MHz, CDCl
3
): δ
ductivity in the presented range of concentration of proton-sponge
base and 2,4-dinitrophenol.
2
=
1
3
2
CH), 114.2, 115.9, 121.3, 125.8, 127.5,
Supporting Information (see also the footnote on the first page of
20 16 2 6
C H N O
f
this article): Correlations of logk vs. E in the series of aqueous
63.25, H 4.31, N 7.22.
1
13
solvents, NMR spectra ( H and C) and rate constants for X,Y-
substituted benzhydryl 2,4-dinitrophenolates at various tempera-
tures.
4-Methoxy-4Ј-methylbenzhydryl 2,4-Dinitrophenyl Ether: Synthe-
sized according to the procedure for 4-fluorobenzhydryl 2,4-dini-
trophenyl ether, except that anhydrous diethyl ether was used in-
stead of benzene as the solvent. From 4-methoxy-4Ј-methylbenz-
Acknowledgments
hydrol (5.0 g, 21.9 mmol), the product was obtained as pale-yellow
1
crystals (1.7 g, ca. 19%). H NMR (300 MHz, CDCl
3
): δ = 2.31 (s,
), 6.42 (s, 1 H, Ar CH), 6.88
d, J = 8.7 Hz, 2 H, ArH), 7.15–7.38 [m, 7 H, ArH + (O N) ArH],
.23 [dd, J = 2.7, 9.3 Hz, 1 H, (O N) ArH], 8.69 [d, J = 2.7 Hz, 1
H, (O N) ): δ = 21.1
ArH] ppm. 13C NMR (75 MHz, CDCl
), 55.3 (ArOCH ), 84.0 (Ar CH), 114.7, 116.4, 121.7, 126.3,
27.9, 128.6, 129.7, 131.1, 136.4, 138.4, 155.6, 159.8 (Ar) ppm.
(394.35): calcd. C 63.96, H 4.60, N 7.10; found C
4.03, H 4.69, N 6.88.
We gratefully acknowledge financial support for this research by
the Ministry of Science Education and Sport of the Republic of
Croatia (Grant No. 006–0982933-2963).
3
(
8
H, Ar-CH
3
), 3.76 (s, 1 H, Ar-OCH
3
2
2
2
2
2
2
2
3
[
1] a) B. Denegri, A. Streiter, S. Juri c´ , A. R. Ofial, O. Kronja, H.
Mayr, Chem. Eur. J. 2006, 12, 1648–1656; b) Correction: B.
Denegri, A. Streiter, S. Juri c´ , A. R. Ofial, O. Kronja, H. Mayr,
Chem. Eur. J. 2006, 12, 5415–5415; c) B. Denegri, A. R. Ofial,
S. Juri c´ , A. Streiter, O. Kronja, H. Mayr, Chem. Eur. J. 2006,
(ArCH
3
3
2
1
21 18 2 6
C H N O
6
12, 1657–1666; d) B. Denegri, S. Minegishi, O. Kronja, H.
Kinetic Methods: Solvents were purified and dried according to
standard procedures. Solvolysis rate constants were measured con-
ductometrically. Freshly prepared solvents (30 mL) were held in a
thermostatically controlled bath (Ϯ0.1 °C) at the given temperature
for several minutes prior to addition of substrate. Typically, 20–
Mayr, Angew. Chem. Int. Ed. 2004, 43, 2302–2305.
[
[
2] a) B. Denegri, O. Kronja, J. Phys. Org. Chem. 2009, 22, 495–
503; b) B. Denegri, O. Kronja, Croat. Chem. Acta, in press.
3] a) B. Denegri, O. Kronja, J. Org. Chem. 2007, 72, 8427–8433;
b) N. Streidl, R. Branzan, H. Mayr, Eur. J. Org. Chem. 2010,
4205–4210.
40 mg of substrate was dissolved in 0.10–0.15 mL of dichlorometh-
ane and injected into the solvent. The increase in conductivity dur-
ing solvolysis was monitored automatically by means of a WTW
LF 530 conductometer, using a Radiometer 2-pole Conductivity
Cell (CDC641T). Individual rate constants were obtained by least-
squares fitting of the conductivity data to the first-order kinetic
equation for 3–4 half-lives. The rate constants were averaged from
at least three measurements.
[4] T. W. Bentley, K. Roberts, J. Org. Chem. 1985, 50, 4821–4828.
[5] I. D. Page, J. R. Pritt, M. C. Whitinh, J. Chem. Soc. Perkin
Trans. 2 1972, 906–911.
[
[
[
6] K.-T. Liu, Y.-S. Lin, M.-L. Tsao, J. Phys. Org. Chem. 1998, 11,
23–229.
7] T. W. Bentley, G. Llewellyn, Prog. Phys. Org. Chem. 1990, 17,
21–159.
8] a) D. J. McLennan, P. L. Martin, Aust. J. Chem. 1979, 32,
361–2370; b) D. J. McLennan, P. L. Martin, J. Chem. Soc.
2
1
2
To achieve complete ionization of the liberated weak acid, the pro-
ton-sponge base [1,8-bis(dimethylamino)naphthalene] was added in
a range of concentrations for each given aqueous binary mixture
Perkin Trans. 2 1982, 1991–1998; c) R. L. Buckson, S. G.
Smith, J. Org. Chem. 1967, 32, 634–639.
[
[
[
9] J. S. Lomas, J.-E. Dubois, J. Org. Chem. 1975, 40, 3303–3304.
10] T. W. Bentley, Chem. Eur. J. 2006, 12, 6514–6520.
11] M. H. Abraham, C. M. Du, J. A. Platts, J. Org. Chem. 2000,
o
Table 4. Experimental concentrations (c ) of base and substrates in
the kinetic measurements of the solvolysis rates of the substituted
benzhydryl 2,4-dinitrophenolates.
65, 7114–7118.
[
12] a) D. N. Kevill, M. S. Bahari, S. W. Anderson, J. Am. Chem.
Soc. 1984, 106, 2895–2901; b) P. R. Luton, M. C. Whiting, J.
Chem. Soc. Perkin Trans. 2 1979, 1507–1511.
Solvent[a]
c
0
(PSB)[b] [m]
c
0
(PSB)/c
(S)[b]
0
[
[
[
13] C. Schade, H. Mayr, Tetrahedron 1988, 44, 5761–5770.
1
9
8
1
9
8
7
00M
12.4–15.6
7.8–26.3
2.6–3.7
84.5–137.2
13.1–52.9
11.0–52.6
8.7–22.9
4.0–7.0
4.0–10.0
1.5–2.0
20.0–30.0
5.0–20.0
5.0–15.0
5.0–10.0
14] D. B. Boyd, J. Org. Chem. 1985, 50, 885–886.
15] a) P. Jaramillo, L. R. Domingo, P. Perez, Chem. Phys. Lett.
0M10W
0M20W
00E
0E10W
0E20W
0E30W
2006, 420, 95–99; b) D. R. Marshall, P. J. Thomas, C. J. M.
Stirling, J. Chem. Soc., Chem. Commun. 1975, 940–941.
16] C. F. Bernasconi, Acc. Chem. Res. 1992, 25, 9–16, and refer-
ences cited therein.
[
[
17] I. D. Page, J. R. Pritt, M. C. Whiting, J. Chem. Soc. Perkin
Trans. 2 1972, 906–911.
[
=
a] Binary solvents (v/v) at 25 °C; E = ethanol, M = methanol, W
water. [b] PSB = proton-sponge base: [1,8-bis(dimethylamino)-
naphthalene]; S = substrate.
Received: May 31, 2010
Published Online: September 17, 2010
6024
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6019–6024