Evidence of substituent-induced electronic interplay. Effect of the remote aromatic ring substituent of phenyl benzoates on the sensitivity of the carbonyl unit to electronic effects of phenyl or benzoyl ring substituents

Article abstract of DOI:10.1021/jo035521u

Carbonyl carbon ¹³C NMR chemical shifts δ_C(C=O) measured in this work for a wide set of substituted phenyl benzoates p-Y-C ₆H₄CO₂C₆H₄-p-X (X = NO₂, CN, Cl, Br, H, Me, or MeO; Y = NO₂, Cl, H, Me, MeO, or NMe₂) have been used as a tool to study substituent effects on the carbonyl unit. The goal of the work was to study the cross-interaction between X and Y in that respect. Both the phenyl substituents X and the benzoyl substituents Y have a reverse effect on δ_C(C=O). Electron-withdrawing substituents cause shielding while electron-donating ones have an opposite influence, with both inductive and resonance effects being significant. The presence of cross-interaction between X and Y could be clearly verified. Electronic effects of the remote aromatic ring substituents systematically modify the sensitivity of the C=O group to the electronic effects of the phenyl or benzoyl ring substituents. Electron-withdrawing substituents in one ring decrease the sensitivity of δ_C(C=O) to the substitution of another ring, while electron-donating substituents inversely affect the sensitivity. It is suggested that the results can be explained by substituent-sensitive balance of the contributions of different resonance structures (electron delocalization, Scheme 1).

Full text of DOI:10.1021/jo035521u

Evid en ce of Su bstitu en t-In d u ced Electr on ic In ter p la y. Effect of
th e Rem ote Ar om a tic Rin g Su bstitu en t of P h en yl Ben zoa tes on th e
Sen sitivity of th e Ca r bon yl Un it to Electr on ic Effects of P h en yl or
Ben zoyl Rin g Su bstitu en ts
Helmi Neuvonen,* Kari Neuvonen, and Paavo Pasanen
Department of Chemistry, University of Turku, FIN-20014 Turku, Finland
helmi.neuvonen@utu.fi
Received October 15, 2003
Carbonyl carbon 13C NMR chemical shifts δC(CdO) measured in this work for a wide set of
substituted phenyl benzoates p-Y-C6H4CO2C6H4-p-X (X ) NO2, CN, Cl, Br, H, Me, or MeO; Y )
NO2, Cl, H, Me, MeO, or NMe2 ) have been used as a tool to study substituent effects on the carbonyl
unit. The goal of the work was to study the cross-interaction between X and Y in that respect.
Both the phenyl substituents X and the benzoyl substituents Y have a reverse effect on δC(CdO).
Electron-withdrawing substituents cause shielding while electron-donating ones have an opposite
influence, with both inductive and resonance effects being significant. The presence of cross-
interaction between X and Y could be clearly verified. Electronic effects of the remote aromatic
ring substituents systematically modify the sensitivity of the CdO group to the electronic effects
of the phenyl or benzoyl ring substituents. Electron-withdrawing substituents in one ring decrease
the sensitivity of δC(CdO) to the substitution of another ring, while electron-donating substituents
inversely affect the sensitivity. It is suggested that the results can be explained by substituent-
sensitive balance of the contributions of different resonance structures (electron delocalization,
Scheme 1).
In tr od u ction
and conclusions are drawn on the basis of the magnitude
of the reaction constant. A less studied aspect, however,
is how much substitution in other parts of the molecule
can adjust the sensitivity of the reaction in question to
changes in the aforementioned substitution. In some
reaction systems, very complicated data handling has
been needed.2
The CdO functional group occurs in the most impor-
tant organic molecular structures: in aldehydes, ketones,
carboxylic acids, and their derivatives. The reactivity of
these compounds is known to be significantly affected by
the substitution in the groups joined the carbonyl carbon.
In general, the electron-withdrawing substituents in-
crease the reactivity while the electron-donating ones
decrease it.1 The main reason for this phenomenon is still
under investigation. According to the transition-state
theory, changes in reactivity are attributed to changes
in the free energy of activation ∆Gq ) G(transition state)
- G(ground state). So changes in both G(transition state)
and in G(ground state) can contribute. Acyl-transfer
reactions have an active role both in organic chemistry
and in biological systems too. To get a deeper and more
exact understanding about the substituent effects on the
Although the reactivity of an organic molecule is
mainly determined by the functional group participating
the reaction in question, it is also adjusted by other
structural features of the molecule. Substituents attached
to the molecule framework can enhance or diminish the
reactivity. The mechanistic conclusions based on linear
free energy relationships have been extremely fruitful.
This concept relates changes in reactivity caused by
changes in substitution in one reaction series to changes
in equilibrium or reactivity in another series caused by
the same changes in substitution. The most typical and
the most useful linear free energy relationship is the
Hammett equation which correlates rates and equilibria
of side-chain reactions of para- and meta-substituted
aromatic compounds. In its most conventional form, the
Hammett equation is used as shown in eq 1
log k
log ko
) Fσ
(1)
where k is the rate coefficient for a para- or meta-
substituted aromatic derivative, ko that for the unsub-
stituted one, σ is the substituent constant for the sub-
stituent in question, and F is the reaction constant.1
Usually, substitution in one site of the molecule is varied
(1) Kirby, A. J . In Comprehensive Chemical Kinetics; Bamford, C.
H., Tipper, C. F. H., Eds.; Elsevier Publishing Co.: Amsterdam, 1980;
Vol. 10, Chapter 2,
p 161. Connors, K. A. Structure Reactivity
Relationships: the Study of Reaction Rates in Solution; VCH Publish-
ers: New York, 1990; Chapter 7, p 311.
* To whom correspondence should be addressed. Fax: +358-
2-3336700.
(2) Fujio, M.; Kim, H.-J .; Uddin, M. K.; Yoh, S.-D.; Rappoport, Z.;
Tsuno, Y. J . Phys. Org. Chem. 2002, 15, 330.
10.1021/jo035521u CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/01/2004
3794
J . Org. Chem. 2004, 69, 3794-3800

Evidence of Substituent-Induced Electronic Interplay
CHART 1
CHART 3
CHART 4
CHART 2
NMR chemical shift of the carbonyl oxygen to the benzoyl
substituent X. They suggested that the parameter F+,
obtained by correlating the δO(CdO) chemical shift values
with σ+(X) substituent parameters for every different Y,
can be used as an empirical measure of the effect of Y on
the electrophilicity of the carbonyl group. That was
because it was realized that the parameter F+ was the
higher the higher was the δO(CdO) shift value for the
unsubstituted derivative, i.e., X ) H, for each Y. Dell’erba
et al.16 investigated a similar system by studying the
effect of the electron-withdrawing nitro substituents in
the phenyl ring of phenyl benzoates (mono-, di-, and
trinitro substitutions) on the sensitivity of the carbonyl
carbon 13C NMR chemical shift to benzoyl substituent.
The behavior of benzoyl-substituted phenyl benzoates
was shown to be closely similar to that of benzoyl-
substituted methyl benzoates, while 2,4,6-trinitrophenyl
benzoates possessed characteristics similar to those of
phenyl-substituted acetophenones. Our recent 13C NMR
and computational studies have shown for phenyl-
substituted phenyl acetates that substituents in the
aliphatic acyl group control the effect of the phenyl
substituents on the polarization of the CdO unit.3,4 Also,
we have shown for imines, hydrazones, and related
compounds that the sensitivity of the electronic structure
of the azomethine group (Chart 4) to the phenyl substit-
uents X is strongly affected by the group Y.5
electrophilic nature/reactivity of the carbonyl group, we
have continued our work3,4 with respect to the carboxylic
acid derivatives.
The purpose of the present study was to investigate
systematically the effect of a remote substituent on the
sensitivity of the electronic character of the side-chain
carbonyl group to aromatic substitution. Phenyl-substi-
tuted phenyl benzoates (Chart 1) were chosen as the
model compounds, and 13C NMR resonance of the car-
bonyl carbon was used as a tool to clarify (i) the effect of
the substituent at the benzoyl moiety on the sensitivity
of the carbonyl unit to the phenyl substitution and (ii)
the effect of the substituent at the phenyl moiety on the
sensitivity of the carbonyl unit to the benzoyl substitu-
tion, i.e., the cross-interaction between X and Y. 13C NMR
spectra were recorded for the compounds shown in Chart
1, and the effect of Y on the sensitivity of δC(CdO) to
phenyl substitution X as well as the effect of X on the
sensitivity of δC(CdO) to benzoyl substitution Y were
analyzed.
Although the NMR shielding is not determined only
by the electron density, linear correlations with positive
slopes between the atomic charges and the 13C NMR
chemical shifts for probe nucleus in closely similar
surroundings have been observed in several systems
when the substitution was varied.4-11 A carbon nucleus
resonates at the higher field the higher the electron
density at the carbon. A decrease in electron density of
the carbonyl carbon, on then other hand, can be consid-
ered to facilitate nucleophilic attack on the carbon. NMR
studies of transmission of substituent effects of aromatic
ring substituents X in the organic molecules and espe-
cially the effect on the unsaturated side chain π-units
have been widely performed.3-5,9,12-18 However, there are
only few systematic works concerning the influence of the
component Y in such systems (Chart 2).3-5,16,18
Resu lts
The 13C NMR chemical shifts for the carbonyl car-
bon of the different phenyl-substituted phenyl benzoates
1-6 (cf. Chart 1) are collected in Table 1. In Tables 2
and 3 are shown the substituent-induced changes of the
chemical shifts, SCS [δC(CdO)(substituted compound) -
δC(CdO)(unsubstituted compound)], of the CdO carbon
with respect of the substitutions X and Y, respectively.
Discu ssion
Our primary interest was to study if there is a
systematic effectsas seen by the 13C NMR chemical shift
(11) Alvarez-Ibarra, C.; Quiroga-Feijo´o, M. L.; Toledano, E. J . Chem.
Soc., Perkin Trans. 2 1998, 679.
Dahn et al.18 studied in the system shown in Chart 3,
the effect of Y (Y ) H, CF3, COOEt, Br, Cl, F, SEt,
OCOPh, OH, O-Na+, NH2) on the sensitivity of the 17O
(12) (a) Bromilow, J .; Brownlee, R. T. C.; Craik, D. J .; Fiske, P. R.;
Rowe, J . E.; Sadek, M. J . Chem. Soc., Perkin Trans. 2 1981, 753. (b)
Dell’Erba, C.; Sancassan, F.; Novi, M.; Petrillo, G.; Mugnoli, A.; Spinelli,
D.; Consiglio, G.; Gatti, P. J . Org. Chem. 1988, 53, 3564.
(13) Craik, D. J .; Brownlee, R. T. C. Prog. Phys. Org. Chem. 1983,
14, 1 and references therein.
(3) Neuvonen, H.; Neuvonen, K. J . Chem. Soc., Perkin Trans. 2 1999,
1497.
(4) Neuvonen, H.; Neuvonen, K.; Koch, A.; Kleinpeter, E.; Pasanen,
P. J . Org. Chem. 2002, 67, 6995.
(5) Neuvonen, K.; Fu¨lo¨p, F.; Neuvonen, H.; Koch, A.; Kleinpeter,
E.; Pihlaja, K. J . Org. Chem. 2003, 68, 2151.
(6) Hehre, W. J .; Taft, R. W.; Topsom, R. D. Prog. Phys. Org. Chem.
1976, 12, 159.
(14) (a) Brownlee, R. T. C.; Craik, D. J . Org. Magn. Reson. 1981,
15, 248. (b) Brownlee, R. T. C.; Sadek, M.; Craik, D. J . Org. Magn.
Reson. 1983, 21, 616. (c) Reynolds, W. F. Prog. Phys. Org. Chem. 1983,
14, 165.
(15) Hamer, G. K.; Peat, I. R.; Reynolds, W. F. Can. J . Chem. 1973,
51, 897, 915.
(7) Krawczyk, H.; Szatylowicz, H.; Gryff-Keller, A. Magn. Reson.
Chem. 1995, 33, 349.
(8) Lin, S.-T.; Lee, C.-C.; Liang, D. W. Tetrahedron 2000, 56, 9619.
(9) Neuvonen, K.; Fu¨lo¨p, F.; Neuvonen, H.; Koch, A.; Kleinpeter,
E.; Pihlaja, K. J . Org. Chem. 2001, 66, 4132.
(10) Matsumoto, K.; Katsura, H.; Uschida, T.; Aoyama, K.; Machigu-
chi, T. Heterocycles 1997, 45, 2443.
(16) Dell’Erba, C.; Sancassan, F.; Leandri, G.; Novi, M.; Petrillo, G.;
Mele, A. Gazz. Chim. Ital. 1989, 119, 643.
(17) (a) Neuvonen, K.; Fu¨lo¨p, F.; Neuvonen, H.; Pihlaja,K. J . Org.
Chem. 1994, 59, 5895. (b) Neuvonen, K.; Fu¨lo¨p, F.; Neuvonen, H.;
Simeonov, M.; Pihlaja, K. J . Phys. Org. Chem. 1997, 10, 55.
(18) Dahn, H,; Pe´chy, P.; Van Toan, V. Magn. Reson. Chem. 1997,
35, 589.
J . Org. Chem, Vol. 69, No. 11, 2004 3795

Neuvonen et al.
TABLE 1. Ca r bon yl Ca r bon 13C NMR Ch em ica l Sh ifts
δC(CdO) (in p p m ) in CDCl3 for Su bstitu ted P h en yl
Ben zoa tes p-Y-C6H4COO-C6H4-p-X
or Hammett-type substituent constants. For every series
1-6, a good linear correlation is observed between the
13C NMR chemical shifts of the carbonyl carbons and the
Hammett σ values19 of the phenyl substituents X [cf.
Table 4 and Figure 1; δC(CdO) ) F(X)σ(X) + k]. The
correlation coefficients from 0.9968 to 0.9983 indicate
that the substituent effects are electronic in origin.The
F(X) values are negative in all cases and the absolute
value of F(X) is the higher the more electron-donating
the benzoyl substituent Y is (Table 4). The latter fact is
also seen by the slopes of the cross-correlations given in
Table 5. The negativity of the F(X) values (Table 4) means
that the more electron-withdrawing the phenyl substitu-
ent X is, the more shielded is the carbonyl carbon. This
kind of reverse substituent chemical shift effect is in
accordance with our previous observations for the car-
bonyl carbon of phenyl-substituted phenyl acetates,
dichloroacetates, and trifluoroacetates.3,4 Reverse effects
have also been observed for some other unsaturated
aromatic side-chain carbons.5,9,12-18 The behavior has
often been explained by the so-called π-polarization
mechanism.12a,13-15,17 The substituent dipole is then
thought to polarize each π-unit as a localized system as
shown in 7 for an electron-withdrawing substituent. In
that case the process leads to increase in the electron
density at the carbonyl carbon. This model, however, does
X/Y
NO2
Cl
H
Me
OMe
NMe2
NO2
CN
Cl
Br
H
162.46
162.53
163.10
163.01
163.33
163.51
163.68
163.42
163.50
164.11
164.02
164.34
164.54
164.70
164.24
164.33
164.95
164.86
165.19
165.38
165.55
164.28
164.37
165.01
164.91
165.25
165.44
165.60
163.91
164.01
164.66
164.57
164.92
165.11
165.28
164.44
164.55
165.24
165.16
165.50
165.71
165.87
Me
OMe
TABLE 2. Su bstitu en t-In d u ced Ch a n ges of th e
Ch em ica l Sh ift (SCS) for th e Ca r bon yl Ca r bon (in p p m )
in CDCl3 Wh en Va r yin g X for Su bstitu ted P h en yl
Ben zoa tes p-Y-C6H4COO-C6H4-p-X [SCS ) δC(CdO) for
p-Y-C6H4COO-C6H4-p-X - δC(CdO) for p-Y-C6H4COO-C6H5]
X/Y
NO2
Cl
H
Me
OMe NMe2
σ(X)a/σ(Y)a
0.78
0.66
0.23
0.23
0.78
0.23
0
-0.17 -0.27 -0.83
NO2 -0.87 -0.92 -0.95 -0.97 -1.01 -1.06
CN
Cl
Br
-0.80 -0.84 -0.86 -0.88 -0.91 -0.95
-0.23 -0.23 -0.24 -0.24 -0.26 -0.26
-0.32 -0.32 -0.33 -0.34 -0.35 -0.34
0
H
0
0
0
0
0
0
-0.17
-0.27
Me
OMe
0.18
0.35
0.20
0.36
0.19
0.36
0.19
0.35
0.19
0.36
0.21
0.37
a
From ref 19.
TABLE 3. Su bstitu en t-In d u ced Ch a n ges of th e
Ch em ica l Sh ift (SCS) for th e Ca r bon yl Ca r bon (in p p m )
in CDCl3 Wh en Va r yin g Y for Su bstitu ted P h en yl
Ben zoa tes p-Y-C6H4COO-C6H4-p-X [SCS ) δC(CdO) for
p-Y-C6H4COO-C6H4-p-X - δC(CdO) for C6H5COO-C6H4-p-X]
X/Y
NO2
Cl
H
Me
OMe NMe2
σ(X)a/σ(Y)a
0.78
0.66
0.23
0.23
0.78
0.23
0
0
0
0
0
0
0
0
-0.17 -0.27 -0.83
not explain the fact that electron-withdrawing phenyl
substituents are known to increase reaction rates of
nucleophilic acyl substitutions of phenyl esters.1,3,4 Both
the reactivity and 13C NMR chemical shift behavior
described above can be better understood by considering
the resonance structures of phenyl benzoates, i.e., the
substituent-sensitive balance of electron delocalization.
The electron-withdrawing phenyl substituents X induc-
tively destabilize the resonance form 9 (Scheme 1) and
therefore decrease the significance of its contribution
destabilization of the phenyl benzoate structure as result.
They obviously also diminish the importance of the
resonance form 10. Substituents capable to electron-
withdrawing by resonance (NO2, CN) allow the contribu-
tion of 11, which also could mean increase in shielding
of the carbonyl carbon. However, because the correlation
of δC(CdO) with substituent parameter σ (not σ-) was
excellent, the contribution of 11 cannot be significant.
Decreased resonance stabilization can contribute to the
increase in reaction rates of the compounds while de-
creased contribution of 9 results in shielding of the
carbonyl carbon, i.e., high-field (low-frequency) shift. This
conclusion is in agreement with the 13C NMR results of
Dell’Erba et al.16 for some nitrosubstituted phenyl ben-
zoates. It is also supported by our previous findings
concerning 13C NMR chemical shifts of the carbonyl
carbons, ν(CdO) frequency values, and CdO bond orders
of substituted phenyl acetates. The reaction energies of
the isodesmic reactions evaluating the substituent effects
NO2 -1.78 -0.82
0.04 -0.33
0.04 -0.32
0.06 -0.29
0.05 -0.29
0.06 -0.27
0.06 -0.27
0.05 -0.27
0.20
0.22
0.29
0.30
0.31
0.33
0.32
CN
Cl
Br
H
-1.80 -0.83
-1.85 -0.84
-1.85 -0.84
-1.86 -0.85
-1.87 -0.84
0
-0.17
-0.27
Me
OMe -1.87 -0.85
a
From ref 19.
of the carbonyl carbonsof the benzoyl substituent on the
sensitivity of the electronic properties of the carbonyl unit
to the phenyl substitution and also if there is a corre-
sponding effect of the phenyl substituent on the sensitiv-
ity of the electronic effects of the carbonyl unit to the
benzoyl substitution. Therefore, we will discuss sepa-
rately the effect of phenyl and benzoyl substituents on
δC(CdO).
Effect of P h en yl Su bstitu en t X on th e CdO Ca r -
b on R eson a n ce δC(CdO). With every different ben-
zoyl substituent Y (1-6), when the phenyl substitution
X (a -g) is varied the smallest shift value (the lowest
frequency value) for the carbonyl carbon is observed with
the most electron-withdrawing substitution X ) NO2 (σ
) 0.78) and the largest shift value (the highest frequency
value) with the most electron-donating substitution X )
OMe (σ ) -0.27) (Tables 1 and 2). In other words,
electron-withdrawing substituents cause shielding while
the electron-donating ones cause deshielding. This is
opposite to the idea of the generalized electronic effect
(a normal effect), a tendency of the substituent to release
or withdraw electronssusually quantified by Hammett
(19) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
3796 J . Org. Chem., Vol. 69, No. 11, 2004

Evidence of Substituent-Induced Electronic Interplay
TABLE 4. Cor r ela tion P a r a m eter s G(X) or GF (X) a n d GR(X) Obser ved for Differ en t P h en yl Su bstitu ted P h en yl Ben zoa tes
p-Y-C6H4COO-C6H4-p-X Wh en Va r yin g X a n d Cor r ela tin g δC(CdO) Va lu es w ith σ(X) or σF (X) a n d σR(X), Resp ectivelya ,b
Y
F(X) ( s
r
FF(X) ( s
FR(X) ( s
r
f
FF(Y)/FR(Y)
NO2
Cl
H
Me
OMe
NMe2
-1.17 ( 0.04
-1.23 ( 0.04
-1.25 ( 0.04
-1.27 ( 0.04
-1.31 ( 0.04
-1.38 ( 0.04
0.9968
0.9975
0.9977
0.9975
0.9980
0.9983
-1.05 ( 0.05
-1.10 ( 0.05
-1.13 ( 0.05
-1.16 ( 0.05
-1.20 ( 0.04
-1.25 ( 0.04
-1.39 ( 0.11
-1.47 ( 0.11
-1.49 ( 0.10
-1.49 ( 0.10
-1.54 ( 0.09
-1.63 ( 0.09
0.9944
0.9946
0.9959
0.9960
0.9967
0.9970
0.11
0.11
0.10
0.09
0.09
0.08
0.76
0.75
0.76
0.78
0.78
0.77
a
s, standard deviation; r, correlation coefficient; f ) sd/rms, where sd ) standard deviation of the correlation and rms ) root-mean-
b
square of the data. σ values from ref 19; σF and σR values from ref 20, except σF and σR for Br from ref 19; the values are given in the
Supporting Information.
tional equilibrium between the E and Z conformations
of the ester structure. However, in general, the Z con-
formations of carboxylic acid esters are much more stable
than their E conformations and the contributions of the
latter ones in general are insignifcant. In addition, the
E conformations usually possess higher dipole moments
than the Z conformations.21 Therefore, a noticeable sol-
vent effect on the substituent dependence of δC(CdO)
should be observed if the substitution affects considerably
the conformational equilibrium. We have previously
shown for phenyl acetates that although the CdO carbon
in DMSO-d6 resonates on a slightly higher field than in
CDCl3 the reverse substituent dependence and the
sensitivity to the substitution is quite similar in these
two solvents.4
Some previous works with aromatic side-chain deriva-
tives have shown that a better correlation of the 13C NMR
chemical shifts with substituent parameters is obtained
with a dual substituent parameter approach than with
a single parameter treatment.5,12-17
F IGURE 1. Plots of the 13C NMR chemical shifts of the
carbonyl carbon of phenyl-substituted phenyl benzoates p-Y-
C6H4CO2C6H4-p-X versus σ(X)19 for series 1 (Y ) NO2) (O),
series 3 (Y ) H) (b), and series 6 (Y ) NMe2) (0).
Therefore, we also fitted the δC(CdO) values observed
when varying substituent X with eq 2 where SCS is the
13C NMR shift for a substituted compound relative to that
for the unsubstituted one. The correlations are good in
all cases and both FF(X) and FR(X) turn to more negative
when the electron-donating ability of Y increases but no
improvement of the correlation is achieved when com-
pared with the single parameter treatment (cf. Table 4).
So the ratio of inductive and resonance contributions
described by Hammett σ prevails for all of the series. This
is also shown by the constancy of FF(X)/FR(X) when
varying Y.
TABLE 5. Sta tistica l Da ta for th e Cr oss-Cor r ela tion s of
th e 13C Ch em ica l Sh ifts of th e Ca r bon yl Ca r bon of
P h en yl-Su bstitu ted P h en yl p-Y-Ben zoa tes
p-Y-C6H4COO-C6H4-p-X w ith P h en yl-Su bstitu ted P h en yl
Ben zoa tes C6H5COO-C6H4-p-X (i.e., Y ) H)
Y
slope
r
NO2
Cl
Me
OMe
NMe2
0.932 ( 0.009
0.981 ( 0.004
1.014 ( 0.005
1.048 ( 0.005
1.097 ( 0.014
0.9998
0.99995
0.99993
0.99994
0.9996
SCS ) FFσF + FRσR
(2)
SCHEME 1
Effect of Ben zoyl Su bstitu en t Y on F(X). The
correlation parameter F(X) clearly varies when the ben-
zoyl substitution is changed (Table 4). To quantitatively
assess the effect of benzoyl substitution, the F(X)’s
observed were correlated with the Hammett σ substituent
constants of the benzoyl substituent Y. Again a good
linear correlation with the slope of 0.13 ( 0.01 is obtained
(Figure 2; r ) 0.9877). The good correlation shows that
the electronic effects of the benzoyl substituents system-
atically modify the sensitivity of the carbonyl carbon
on the stabilities of para- or meta-X-substituted phenyl
acetates, dichloroacetates, and trifluoroacetates also
agree with this interpretation. Electron-withdrawing
substituents affect upfield CdO carbon chemical shift,
increase the ν(CdO) frequency value and the CdO bond
order while they systematically and clearly decrease the
stability of the acetates.3,4 In principle, an alternative
possible reason for the reverse behavior of δC(CdO) could
be the substituent sensitive variation in the conforma-
(20) Taft, R. W.; Topsom, R. D. Prog. Phys. Org. Chem. 1987, 16, 1.
(21) J ones, G. I. L.; Owen, N. L. J . Mol. Struct. 1973, 18, 1. Exner,
O. In The Chemistry of Double-Bonded Functional Groups; Patai, S.,
Ed.; Interscience: London, 1977; p 1. Eliel, E. L.; Wilen, S. H.
Stereochemistry of Organic Compounds; Wiley-Interscience: New York,
1994; pp 618-621. Larson, J . R.; Epiotis, N. D.; Bernardi, F. J . Am.
Chem. Soc. 1978, 100, 5713.
J . Org. Chem, Vol. 69, No. 11, 2004 3797

Neuvonen et al.
phenyl, 4-nitrophenyl, 2,4-dinitrophenyl, and 2,4,6-trini-
trophenyl benzoates.12,16 For the latter esters it has been
shown that Hammett parameter σ alone is not adequate
to describe the effects of the benzoyl substituents on
δC(CdO) but a dual substituent parameter approach is
needed to separate the contributions of inductive and
resonance effects. The same was also observed in this
work. Correlations with σ show the trend but the cor-
relations are only moderate (Table 6). Therefore, the data
were analyzed by eq 2. The advantage of the dual
substituent parameter approach is that it allows a
characteristic balance between inductive and conjugative
effects for every series. The correlation parameters are
shown in Table 6. The good correlations indicate that the
substituent effects are electronic in origin. Both FF(Y) and
FR(Y) are negative and of the same general extent as
those determined previously for other benzoic acid esters
in CDCl3.12,16 The negativity of FF(Y) and FR(Y) means that
both inductive and conjugative factors of the benzoyl
substituents operate by a reverse way. Inductive effects
are more significant than the resonance effects, the value
of FF(Y)/FR(Y) being around 3, but varying when X is
varied. Electron-withdrawing substituents cause shield-
ing of the carbonyl carbon while the electron-donating
ones cause an inverse effect. These effects can be ex-
plained by a similar way than the effect of the phenyl
substitution discussed above (cf. Scheme 1). Electron-
donating substituents Y inductively stabilize the reso-
nance form 9 and consequently increase its contribution,
decrease in the electron density, i.e., deshielding, at the
CdO carbon as a result. The negative FR(Y) values can
be explained by the influence of the resonance induced
polar effect.9,12a,14a,17 By that way the conjugative effects
can operate reversely, i.e., for example deshielding as a
result for an electron-donating Y due to the contribu-
tion of the resonance structure 12. This explanation
which is based on the substituent sensitive electron
delocalization also agrees with the results of Dahn et al.18
which have shown the high-field shift of the 17O NMR
resonance (i.e., shielding) of the carbonyl oxygen of
several benzoyl derivatives by electron-donating benzoyl
substituents.
F IGURE 2. Plot for the F(X) values observed for phenyl
benzoates p-Y-C6H4CO2C6H4-p-X when varying X versus σ(Y).
resonance to the phenyl substitution. The positive value
for the slope means that the more electron-withdrawing
the benzoyl substitution is the less sensitive is the
shielding of the carbonyl carbon to the phenyl substitu-
tion. This result is in line with the aliphatic acyl
substitution effect observed recently by us for phenyl
acetates. When the electron-withdrawing ability of acyl
substituents in R in the series RCOOC6H4-p-X is in-
creased the sensitivity of the 13C NMR chemical shift of
the carbonyl carbon to phenyl substitution X decreases.3,4
In that case, a correlation of F with substituent param-
eters was not determined. The systematic decrease in the
value of F(X) by electron-withdrawing benzoyl substitu-
ents can be interpreted by the diminished polarization
of the carbonyl unit. The more electron-withdrawing the
substituentY is the less significant is the contribution of
9 (Scheme 1). This means the decreased carbocation
character of the carbonyl carbon. Consequently, when we
consider that X also operates by adjusting the significance
of 9 (cf. discussion above), the smaller is the total
contribution of 9 for the molecule structure the less can
the changes in its contribution alter the electronic
character of the CdO carbon. In other words, the less
significant is the carbocation character of the carbonyl
carbon the less sensitive is its shielding to phenyl
substitution.
To show the contribution of inductive and resonance
effects of Y we also analyzed the effect of the benzoyl
substitution Y on F(X) by eq 3. The F(X)F,Y and F(X)R,Y
values then obtained are 0.089 ( 0.023 and 0.20 ( 0.02,
respectively (r ) 0.9903). The correlation is slightly better
than that obtained by the single parameter correlation
(slope ) 0.13 ( 0.01; r ) 0.9877). The ratio of inductive
and conjugative effects of Y, F(X)F,Y/F(X)R,Y ) 0.44.
E ffect of P h en yl Su b st it u en t X on FF (Y) a n d
FR(Y). Although the variation of the values of FF(Y) and
FR(Y) along the phenyl substitution X is smaller than the
variation of F(X) along the benzoyl substitution Y, it is
systematic (Table 6; cf. also the statistical data of the
cross-correlations in Table 7). When the FF(Y) and FR(Y)
values were correlated with σ values of the phenyl
substituent X, slopes of 0.092 ( 0.013 (r ) 0.9540) and
0.20 ( 0.06 (r ) 0.9635), respectively, could be observed
for FF(Y)X and FR(Y)X. The electronic effects of the phenyl
substituents affect the sensitivity of the carbonyl carbon
13C NMR chemical shift to the benzoyl substitution. The
more electron-withdrawing the phenyl substituent is the
less sensitive is the shielding of the carbonyl carbon to
the benzoyl substitution. The resonance parameter
FR(Y) is somewhat more sensitive to the phenyl substitu-
tion than the inductive parameter FF(Y). This can also
F(X) ) F(X)H + F(X)F,Yσ(Y)F + F (X)R,Yσ(Y)R (3)
Effect of Ben zoyl Su bstitu en t Y on th e CdO
Ca r bon Reson a n ce δC(CdO). With every different
phenyl substituent X (a -g), electron-withdrawing ben-
zoyl substituents Y cause upfield chemical shifts and the
electron-donating ones have an opposite effect (Tables 1
and 3). In principle, this reverse behavior is in accordance
with the observations for benzoyl-substituted methyl,
3798 J . Org. Chem., Vol. 69, No. 11, 2004

Evidence of Substituent-Induced Electronic Interplay
TABLE 6. Cor r ela tion P a r a m eter s G(Y) or GF (Y) a n d GR(Y) Obser ved for Differ en t P h en yl-Su bstitu ted P h en yl Ben zoa tes
p-Y-C6H4COO-C6H4-p-X Wh en Va r yin g Y a n d Cor r ela tin g δC(CdO) Va lu es w ith σ(Y) or σF (Y) a n d σR(Y), Resp ectivelya ,b
X
F(Y) ( s
r
FF(Y) ( s
FR(Y) ( s
r
f
FF(Y)/FR(Y)
NO2
CN
Cl
Br
H
-1.23 ( 0.32
-1.26 ( 0.32
-1.33 ( 0.32
-1.34 ( 0.31
-1.36 ( 0.31
-1.37 ( 0.31
-1.37 ( 0.31
0.8872
0.8915
0.9030
0.9054
0.9073
0.9100
0.9094
-2.41 ( 0.10
-2.44 ( 0.10
-2.49 ( 0.10
-2.49 ( 0.10
-2.50 ( 0.09
-2.51 ( 0.10
-2.51 ( 0.10
-0.74 ( 0.11
-0.77 ( 0.10
-0.89 ( 0.10
-0.89 ( 0.10
-0.92 ( 0.09
-0.95 ( 0.10
-0.93 ( 0.09
0.9949
0.9953
0.9961
9.9962
0.9968
0.9964
0.9933
0.10
0.10
0.09
0.09
0.08
0.09
0.09
3.26
3.17
2.80
2.80
2.72
2.64
2.70
Me
OMe
a
s, standard deviation; r, correlation coefficient; f ) sd/rms, where sd ) standard deviation of the correlation and rms ) root-mean-
b
square of the data. σ values from ref 19; σF and σR values from ref 20, except σF and σR for Br from ref 19; the values are given in the
Supporting Information.
TABLE 7. Sta tistica l Da ta for th e Cr oss-Cor r ela tion s of
th e 13C Ch em ica l Sh ifts of th e Ca r bon yl Ca r bon of
Ben zoyl-Su bstitu ted p-X-P h en yl Ben zoa tes
p-Y-C6H4COO-C6H4-p-X w ith Ben zoyl-Su bstitu ted P h en yl
Ben zoa tes p-Y-C6H4COO-C6H5 (i.e., X ) H)
(Y)H ) -0.93 ( 0.05, FF(X,Y) ) 0.095 ( 0.14, FR(X,Y) )
0.20 ( 0.13, and k ) 165.18 ( 0.02 (r ) 0.9966).
Comparison of the values of FF(X,Y) ) 0.095 ( 0.14 and
FR(X,Y) ) 0.20 ( 0.13 with the sums F(X)F,Y + FF(Y)X )
0.18 ( 0.04 and F(X)R,Y + FR(Y)X ) 0.40 ( 0.08, respec-
tively (cf. eqs 6 and 7), shows that although the accuracy
X
slope
r
NO2
CN
Cl
Br
Me
OMe
0.929 ( 0.022
0.944 ( 0.018
0.989 ( 0.006
0.989 ( 0.005
1.008 ( 0.005
1.005 ( 0.004
0.99889
0.99931
0.99992
0.99994
0.99995
0.99997
is not excellent the values agree satisfactorily [F(X)F,Y
)
0.089 ( 0.023, F(X)R,Y ) 0.20 ( 0.02, FF(Y)X ) 0.092 (
0.013 and FR(Y)X ) 0.20 ( 0.06, cf. the preceding chap-
ters]. In summary, the results show that (i) δC(CdO)
values can be nicely correlated with the substituent
parameters describing the ability of substituents to with-
draw or release electrons in the molecule framework and
(ii) the effect of both phenyl and benzoyl substituents on
δC(CdO) of phenyl benzoates can be considered simul-
taneously. However, (iii) although the adjusting effect of
the neighboring ring substituent could be clearly verified
when considering the aromatic rings separately (cf.
Tables 4 and 6), the accuracy of the cross-interaction
parameters FF(X,Y) and FR(X,Y) obtained by eq 5 is not
very good (large standard deviations). So eq 5 is not
recommendable in this case but better results are ob-
tained by studying the effect of Y on F(X) and the effect
of X on FF(Y) and FR(Y) separately. The reason for that
obviously is the relatively small contribution of the cross-
interaction terms. The latter explain ca. 10% or less of
the change of δC(CdO). Therefore, testing of eq 5 in the
future in some comparative systems possessing greater
effects is challenging.
been seen as the decrease of the ratio FF(Y)/FR(Y) when
the electron-donating ability of X increases (Table 6). The
adjusting effect of X on the extent of FF(Y) and FR(Y)
obviously is caused by the decrease of the contribution
of the charge separated resonance form 9 when the
electron-withdrawing ability of X increases.
Ap p lica bility of th e Mu ltip a r a m eter Equ a tion s.
To deal with the extensive data obtained we also used
two different multiparameter equations: eq 4, which
takes into account at the same time the variations of X
and Y but assumes the parameters F(X), FF(Y) and FR(Y)
to be constant, and eq 5, which takes into account
∂F(X)/∂Y, ∂FF(Y)/∂X and ∂FR(Y)/∂X.
δC(CdO) ) F(X)σ(X) + FF(Y)σF(Y) + FR(Y)σR(Y) + k
(4)
δC(CdO) ) F(X)Hσ(X) + FF(Y)HσF(Y) +
FR(Y)HσR(Y) + FF(X,Y)σF(Y)σ(X) +
FR(X,Y)σR(Y)σ(X) + k (5)
Com p a r ison of th e Effects of X a n d Y. Inspection
of the effects of phenyl substituents X (Table 4) and
benzoyl substituents Y (Table 6) shows that the inductive
effect from the benzoyl side is ca. twice (2.1-2.3 times)
as strong as that from the phenyl side. This is under-
standable due to the shorter distance between the sub-
stituent and the probe site in the former case (five bonds
versus six bonds). Contrary, as regards the resonance
contribution, the effect from the phenyl side is 1.6-1.9
times as strong as the effect from the benzoyl side. In
principle, one would expect an opposite order, i.e., a
similar order than that observed for the inductive effects.
The situation can, however, be explained as follows. The
benzoyl substituents possess two competitive resonance
effects: due to the direct conjugation between the ben-
zene ring and the carbonyl unit there is a normal
resonance effect (13) FR(Y, normal) which is >0, and in
addition there is a resonance induced polar effect (12)
FR(Y, reverse) which is <0. The observed resonance effect
FR(Y)observed is the sum of these two effects. Therefore it
can be that |FR(Y)observed|<|FR(X)observed|. If we consider that
where
FF(X,Y) ) F(X)F,Y + FF(Y)X
FR(X,Y) ) F(X)R,Y + FR(Y)X
(6)
(7)
Eq 5 is derived from eq 4 by stating that
F(X) ) F(X)H + F(X)F,YσF(Y) + F(X)R,YσR(Y) (3)
FF(Y) ) FF(Y)H + FF(Y)Xσ(X)
FR(Y) ) FR(Y)H + FR(Y)Xσ(X)
(8)
(9)
By eq 4 the following parameters were obtained: F(X)
) -1.27 ( 0.03, FF(Y) ) -2.46 ( 0.05, FR(Y) ) -0.89 (
0.05, and k ) 165.18 ( 0.02 (r ) 0.9962). Equation 5
gives: F(X)H ) -1.25 ( 0.06, FF(Y)H ) -2.47 ( 0.06, FR-
J . Org. Chem, Vol. 69, No. 11, 2004 3799

Neuvonen et al.
from the phenyl side only the resonance induced polar
effect is in operation, i.e., FR(X)observed ) FR(X,reverse), and
that the effect of distance is the same than that observed
for inductive effects, a FR(Y, normal) value ca. 2 can be
approximated for the normal resonance effect of substit-
uents Y.
mean that the F values observed for some nucleophilic
acyl substitutions, for instance when substituent at
phenyl moiety of phenyl benzoates is varied, must be
considered paying attention also to the actual substitu-
tion at the benzoyl moiety. The present results are also
interesting taking into account the goals of modern
chemistry to develop molecular systems where weak
intermolecular interactions, such as the electrostatic
ones, occur. Related results with the above NMR works
have recently been obtained by Liu et al., for instance in
theoretical bond dissociation energy studies concerning
para-substituted aromatic silanes22 and para-substituted
anilines.23 There seems to prevail an interesting interac-
tion over the aromatic and side-chain systems in that
case too.
Con clu sion s
Both the benzoyl substituents Y and the phenyl sub-
stituents X have a reverse effect on the 13C NMR chemical
shift of the CdO carbon, δC(CdO), of phenyl benzoates,
p-Y-C6H4CO2C6H4-p-X, investigated. The electron-with-
drawing substituents cause shielding while the electron-
donating ones cause deshielding. The effects can be
correlated with Hammett or Hammett-type substituent
constants. These reverse effects can be understood by
considering the effects of substituents X and Y on the
contributions of different resonance structures of phenyl
benzoates, i.e., on the electron delocalization. Substitu-
ents capable of electron-withdrawal by inductive effects
destabilize resonance structures 9 and 10 (cf. Scheme 1)
while those capable of electron donation stabilize them,
reversing the substituent effect on δC(CdO) as a conse-
quence (FF < 0). The reverse resonance effect (FR < 0)
can be explained by the resonance induced polar effect
(cf. 12). Furthermore, the study of the effect of Y on the
sensitivity of δC(CdO) to the phenyl substituents X, and
for another the effect of X on the sensitivity of δC(CdO)
to the benzoyl substituents Y, revealed an interesting
interplay over the molecule. Because the substituent in
one aromatic ring adjusts the significance of the contri-
bution of the charged resonance structures it at the same
time adjusts the sensitivity of δC(CdO) to the effect of
the another ring substituents. This means a novel
experimental verification of systematic changes in the
polarization of the carbonyl unit (cf. 14) by gradual
changes in the electron-withdrawing/-donating power of
the substituents in the molecule.
Exp er im en ta l Section
Ma ter ia ls. Phenyl benzoates were prepared by standard
procedures from substituted benzoyl chlorides and substituted
phenols in dichloromethane in the presence of triethylamine,
purified by recrystallization from benzene-petroleum ether
(bp 40-60 °C) mixture, and characterized by the melting point
determinations and 13C NMR measurements. The melting
point ranges and the spectral data are given in the Supporting
Information.
Sp ectr oscop ic Mea su r em en ts. The 13C NMR spectra
(125.78 MHz) were recorded in CDCl3 at 25 °C at concentration
of 0.1 mol L-1. A low and constant sample concentration was
used to avoid any disturbances from intermolecular associa-
tions. The deuterium of the solvent was used as a lock signal.
The spectra were measured with a 1H broad-band decoupling
technique. The chemical shifts are expressed in ppm relative
to TMS (0 ppm) used as an internal reference.
Su p p or tin g In for m a tion Ava ila ble: Melting point and
13C NMR data for the prepared substituted phenyl benzoates;
σ, σF, and σR values used in the correlations. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O035521U
This is an important aspect to be considered in the
substitution versus reactivity considerations. From the
reaction mechanistic point of view the present results
(22) Cheng, Y.-H.; Zhao, X.; Song, K.-S.; Liu, L.; Guo, Q.-X. J . Org.
Chem. 2002, 67, 6638.
(23) Song, K.-S.; Liu, L.; Guo, Q.-X. J . Org. Chem. 2003, 68, 262.
3800 J . Org. Chem., Vol. 69, No. 11, 2004