13C and 1H nuclear magnetic resonance of methyl-substituted acetophenones and methyl benzoates: Steric hindrance and inhibited conjugation

Article abstract of DOI:10.1002/mrc.1414

The ¹H and ¹³C NMR spectra of 14 methyl-substituted acetophenones and 14 methyl-substituted methyl benzoates were assigned and interpreted with respect to the conformation of the C_ar - C(O) bond. The substituent effects are proportional in the two series and can be divided into polar and steric: each has different effects on the ¹³C SCS of the individual atoms. In the case of C atoms C(O), C(1) and CH₃(CO), the steric effects were quantitatively separated by comparing SCS in the ortho and para positions. The steric effects are proportional for the individual C atoms and also to steric effects estimated from other physical quantities. However, they do not depend simply on the angle of torsion φ of the functional group as anticipated hitherto. A better description distinguishes two classes of compounds: sterically not hindered or slightly hindered planar molecules and strongly sterically hindered, markedly non-planar. In order to confirm this reasoning without empirical correlations, the J(C,C) coupling constants were measured for three acetophenone derivatives labeled with ¹³C in the acetyl methyl group. The constants confirm unambiguously the conformation of 2-methylacetophenone; their zero values are in accord with the conformation of 2,6-dimethylacetophenone. The zero values in the unsubstituted acetophenone are at variance with previous erroneous report but all J(C,C) values are in accord with calculations at the B3LYP/6-311++G(2d,2p)// B3LYP/6-311+G(d,p) level. Copyright

Full text of DOI:10.1002/mrc.1414

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2004; 42: 844–851
Published online 10 August 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mrc.1414
13C and 1H nuclear magnetic resonance
of methyl-substituted acetophenones and methyl
benzoates: steric hindrance and inhibited conjugation
1
2
3
1
1∗
´
´
ˇ
ˇ ˇ
ˇ
´
¨
Milos Budesınsky´, Jirı Kulhanek, Stanislav Bohm, Petr Cigler and Otto Exner
1
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, CZ-16610 Prague 6, Czech Republic
Department of Organic Chemistry, University of Pardubice, CZ-53210 Pardubice, Czech Republic
Department of Organic Chemistry, Prague Institute of Chemical Technology, CZ-16628 Prague 6, Czech Republic
2
3
Received 18 December 2003; Revised 10 March 2004; Accepted 7 April 2004
1
The H and 13C NMR spectra of 14 methyl-substituted acetophenones and 14 methyl-substituted methyl
benzoates were assigned and interpreted with respect to the conformation of the Car — C(O) bond. The
substituent effects are proportional in the two series and can be divided into polar and steric: each has
different effects on the 13C SCS of the individual atoms. In the case of C atoms C(O), C(1) and CH3(CO),
the steric effects were quantitatively separated by comparing SCS in the ortho and para positions. The
steric effects are proportional for the individual C atoms and also to steric effects estimated from other
physical quantities. However, they do not depend simply on the angle of torsion f of the functional
group as anticipated hitherto. A better description distinguishes two classes of compounds: sterically not
hindered or slightly hindered planar molecules and strongly sterically hindered, markedly non-planar.
In order to confirm this reasoning without empirical correlations, the J(C,C) coupling constants were
measured for three acetophenone derivatives labeled with 13C in the acetyl methyl group. The constants
confirm unambiguously the conformation of 2-methylacetophenone; their zero values are in accord with
the conformation of 2,6-dimethylacetophenone. The zero values in the unsubstituted acetophenone are
at variance with previous erroneous report but all J(C,C) values are in accord with calculations at the
B3LYP/6-311++G(2d,2p)//B3LYP/6–311+G(d,p) level. Copyright  2004 John Wiley & Sons, Ltd.
KEYWORDS: NMR; 13C NMR; 1H NMR; substituent effects; steric effect; hindered conjugation
extreme steric hindrance and a perpendicular conformation
INTRODUCTION
°
ꢁꢀ D 90 ꢂ:
The conformation of aromatic ortho derivatives with one
planar substituent has often been examined by 1H, 13C
or 17O NMR spectroscopy. The results have commonly
been interpreted in terms of inhibited conjugation, which
is often called steric hindrance to resonance1,2 (SIR). This
is correlated with the assumed torsion angle ꢀ between
the functional group and the ring plane; as ꢀ increases
the resonance decreases. Anilines,3 benzaldehydes,4 aryl
ketones,4–6 imines,6c benzoic acids7,8 and their esters8 and
amides7a with ortho substituents have been investigated
in this way. The ortho substituents generating the steric
hindrance were mostly alkyl groups of variable size or
in variable number. The angle ꢀ was estimated from the
chemical shifts υ according to Eqn (1), where υ0 relates to the
cos2 ꢀ D ꢁυ ꢀ υ ꢂ/ꢁυ0 ꢀ υ ꢂ
ꢁ1ꢂ
ꢁ2ꢂ
1
1
cos ꢀ D ꢁυ ꢀ υ ꢂ/ꢁυ0 ꢀ υ ꢂ
1
1
There is no great difference when cos2 ꢀ is replaced by cos ꢀ,
Eqn (2).2,3 In addition to NMR shifts, other spectroscopic2,9
and non-spectroscopic10,11 methods have been treated by
Eqn (1) or (2). The values of ꢀ estimated from various
experimental quantities were often in fair agreement and
have been accepted as a reasonable description of the actual
conformation.
Although the basic principle1 of SIR is obviously correct,
objections may be raised against the quantitative extension2
operating with the angle ꢀ. We have argued12–14 that
some values do not agree with the actual conformation
determined in an unambiguous way. At present, the con-
formation is determined most reliably by quantum chemical
calculations,15–17 sometimes supported by X-ray16 or elec-
tron diffraction methods.18 Recently we examined in detail
a series of methyl-substituted acetophenones17 1–14 (see
Table 1) by density functional theory (DFT) calculations
at the B3LYP/6–311CG(d,p) level. The computed angles
ꢀ disagreed with previous estimates from 13C NMR shifts,5
unsubstituted compound with the planar conformation ꢁꢀ D
°
0 ꢂ and υ to the compound (sometimes hypothetical) with
1
ŁCorrespondence to: Otto Exner, Institute of Organic Chemistry
and Biochemistry, Academy of Sciences of the Czech Republic,
CZ-16610 Prague 6, Czech Republic.
E-mail: otto.exner@uochb.cas.cz
Contract/grant sponsor: Ministry of Education of the Czech
Republic; Contract/grant number: LN00A032.
Copyright  2004 John Wiley & Sons, Ltd.

13C and 1H NMR of methylacetophenones and methyl benzoates
845
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

´
846
M. Budeˇsˇınsky´ et al.
electronic spectra2 and dipole moments.2 Still more impor-
tant, ꢀ was rejected in principle as a quantity measuring the
intensity of the steric hindrance since the energy barrier and
the whole potential-energy curve were very different for the
individual members of the series.17 In some cases ꢀ belongs to
a stable conformation whereas in others the energy minima
are so shallow that ꢀ has no physical meaning.
O
CH3
CH3
O
CH3
O
φ
CH3
C
C
C
CH3
CH3
CH3
2A
2B
8
The conformations of some methyl-substituted acetophe-
nones have not been unambiguously determined in spite of
numerous investigations. The most investigated compound,
2-methylacetophenone, was attributed mostly a non-planar
conformation.3–5,7,19 Only in a few cases was the equilibrium
ꢃ
i.e. the possibility of determining ꢀ from Eqn (1). We
proceeded in two ways. First, we investigated the 13C
and 1H chemical shifts of a series of methyl-substituted
acetophenones 1–14 (see Tables 1 and 3) and similarly
substituted methyl benzoates 15–28 (see Tables 2 and 4). We
believed that it would be possible to obtain more significant
results by extending the series and using a systematic
treatment within the framework of correlation analysis.
This procedure has been used, for example, for methyl-
substituted benzoic acids.12 Second, we wanted to obtain a
2A
2B assumed,20 or only one planar form was dominant;6
ꢄ
in the extreme case the same experimental facts were inter-
preted differently.5,6 In contrast, 2,6-dimethylacetophenone
was always assumed3–7,21 to be in a non-planar form, 8, for
which there were only slight differences in the values of ꢀ.
In this work, we intended not only to determine the
conformation of acetophenones more reliably but also to
examine critically the SIR theory in its quantitative aspect,
3
non-empirical proof from the JꢁC,Cꢂ coupling constants of
isotopically labeled samples of 1,2 and 8 with the 13C isotope
in the acetyl-CH3 group.
Table 2. Carbon-13 chemical shifts of methyl-substituted methyl benzoates 15–28 in CDCl3
No.
Substituent
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
CH3
15
16
17
18
19
20
21
22
23
24
25
26
27
28
H
2-Me
3-Me
4-Me
130.11
129.51
130.07
127.35
130.92
126.57
129.30
133.80
127.70
129.96
128.77
130.85
126.87
134.92
129.51
140.11
130.10
129.52
128.30
131.61
138.12
128.99
132.85
131.88
133.64
143.48
128.30
125.62
128.23
128.99
125.11
126.40
136.98
127.52
129.63
137.96
126.93
128.35
136.56
134.08
129.51
130.50
126.68
129.52
127.62
130.77
130.97
134.91
127.12
127.85
127.04
135.14
128.55
129.80
167.05
168.00
167.27
167.12
169.08
168.00
168.22
170.45
167.36
167.46
169.23
170.59
167.51
171.67
52.02
51.71
52.01
51.85
51.85
51.60
51.70
51.81
51.88
51.95
51.73
51.70
51.83
51.92
—
21.63 (2)
21.23 (3)
21.55 (4)
2,3-Me2
2,4-Me2
2,5-Me2
2,6-Me2
3,4-Me2
3,5-Me2
2,3,4-Me3
2,4,6-Me3
3,4,5-Me3
2,3,5,6-Me4
137.90a 137.47a 133.03
16.57 (2); 20.50 (3)
21.72 (2); 21.34 (4)
21.18 (2); 20.73 (5)
19.66 ꢁ2 C 6ꢂ
140.33
135.18
134.91
130.61
127.85
137.26
135.14
128.55
129.80
132.47
131.56
127.52
136.68
137.96
136.34
128.35
136.56
134.08
142.45
132.68
129.29
142.22
135.40
140.31
139.26
140.82
132.21
19.65 (3); 19.97 (4)
21.08 ꢁ3 C 5ꢂ
17.02 (2); 15.62 (3); 21.14 (4)
19.71 ꢁ2 C 6ꢂ; 21.08 (4)
20.49 ꢁ3 C 5ꢂ; 15.76 (4)
16.34 ꢁ2 C 6ꢂ; 19.53 ꢁ3 C 5ꢂ
a The assignment of these signals may be interchanged.
Table 3. Proton chemical shifts of methyl-substituted acetophenones 1–14 in CDCl3
No.
Substituent
H-2
H-3
H-4
H-5
H-6
COꢁCH3ꢂ
CH3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
H
2-Me
3-Me
4-Me
7.952
—
7.770
7.852
—
—
—
—
7.719
7.563
—
—
7.594
—
7.452
7.243
—
7.249
—
7.051
7.121
7.008
—
—
—
6.833
—
—
7.553
7.379
7.374
—
7.241
—
7.184
7.144
—
7.194
—
7.452
7.273
7.343
7.249
7.132
7.070
—
7.008
7.191
—
7.952
7.697
7.750
7.852
7.383
7.632
7.487
—
7.672
7.563
7.307
—
2.595
2.590
2.589
2.570
2.550
2.556
2.564
2.466
2.552
2.572
2.535
2.453
2.561
2.455
—
2.529 (2)
2.409 (3)
2.404 (4)
2,3-Me2
2,4-Me2
2,5-Me2
2,6-Me2
3,4-Me2
3,5-Me2
2,3,4-Me3
2,4,6-Me3
3,4,5-Me3
2,3,5,6-Me4
2.340 (2); 2.300 (3)
2.517 (2); 2.348 (4)
2.476 (2); 2.360 (5)
2.243 ꢁ2 C 6ꢂ
2.298 ꢁ3 C 4ꢂ
2.367 ꢁ3 C 5ꢂ
7.034
6.833
—
2.369 (2); 2.201 (3); 2.311 (4)
2.215 ꢁ2 C 6ꢂ; 2.222 (4)
2.337 ꢁ3 C 5ꢂ; 2.224 (4)
2.089 ꢁ2 C 6ꢂ; 2.205 ꢁ3 C 5ꢂ
—
—
6.950
7.594
—
—
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

13C and 1H NMR of methylacetophenones and methyl benzoates
847
Table 4. Proton chemical shifts of methyl-substituted methyl benzoates 15–28 in CDCl3
No.
Substituent
H-2
H-3
H-4
H-5
H-6
OCH3
CH3
15
16
17
18
19
20
21
22
23
24
25
26
27
28
H
2-Me
3-Me
4-Me
8.040
—
7.859
7.924
—
—
—
—
7.811
7.734
—
—
7.672
—
7.430
7.237
—
7.226
—
7.049
7.123
7.026
—
—
—
6.845
—
—
7.547
7.385
7.365
—
7.274
—
7.200
7.182
—
7.229
—
7.430
7.230
7.322
7.226
7.121
7.040
—
7.026
7.188
—
8.040
7.904
7.836
7.924
7.615
7.824
7.718
—
7.764
7.734
7.513
—
3.912
3.883
3.908
3.891
3.882
3.868
3.881
3.908
3.892
3.894
3.861
3.887
3.885
3.911
—
2.597 (2)
2.400 (3)
2.396 (4)
2,3-Me2
2,4-Me2
2,5-Me2
2,6-Me2
3,4-Me2
3,5-Me2
2,3,4-Me3
2,4,6-Me3
3,4,5-Me3
2,3,5,6-Me4
2.449 (2); 2.313 (3)
2.570 (2); 2.345 (4)
2.544 (2); 2.338 (5)
2.309 ꢁ2 C 6ꢂ
2.301 (3)a; 2.310 (4)a
2.371 ꢁ3 C 5ꢂ
7.017
6.845
—
2.465 (2); 2.206 (3); 2.308 (4)
2.280 ꢁ2 C 6ꢂ; 2.274 (4)
2.318 ꢁ3 C 5ꢂ; 2.214 (4)
2.134 ꢁ2 C 6ꢂ; 2.208 ꢁ3 C 5ꢂ
—
—
6.974
7.672
—
—
a The assignment of these signals may be interchanged.
effect is small.13,14 The steric effect, SCSst, is then estimated
by subtracting the assumed polar effects of all substituents,
ortho substituents being replaced by para. For instance for
2,3-dimethylacetophenone 5 SCSst is given by Eqn (3) where
all SCS are related to the unsubstituted acetophenone:
RESULTS AND DISCUSSION
13C chemical shifts of 1–14 and 15–28 are given in Tables 1
and 2, respectively, and the corresponding 1H chemical shifts
in Tables 3 and 4.
13C SCS and conformation of acetophenones
SCSstꢁ2, 3ꢂ D SCSꢁ2, 3ꢂ ꢀ SCSꢁ3ꢂ ꢀ SCSꢁ4ꢂ
ꢁ3ꢂ
Our measured 13C SCS of substituted acetophenones 1–14
(Table 1) agree well with previously reported22,23 but mostly
unassigned22 spectra of several compounds. In particular, the
shifts given22a for 1,4,6,7,9 and 12 differ from ours essentially
by a constant shift of 0.14 ppm. Agreement with previous
data5 measured as neat liquids is not so good.
The values of SCSst for the carbon atoms C-7, C-8 and
C-1 are given in Table 1 in the last three columns. For 1,
3 and 4 they are zero by definition. For 9, 10, 13 they
should essentially vanish according to the first of the two
basic approximations. The small, non-zero values for these
compounds can thus serve as a measure of the reliability and
accuracy of the whole approach. The justification of SCSst
and the reliability of their values are seen very clearly in
Fig. 1. There is no simple relation between SCS of the atoms
C-7 and C-1 since they are influenced differently by polar and
The substituent effects of the methyl groups can be
rationalized by distinguishing two different mechanisms.
The first, termed the polar effect, operates primarily on
the directly attached carbon atom (on average C9.5 ppm),
but decreases with the carbon atom distance. The second
effect is observed mainly on the carbon atoms of the acetyl
group, denoted C-7 and C-8, and less on C-1 or C-4. It is
termed a steric effect and may be partly connected with the
torsion angle ꢀ and with SIR. The SCS of C-7 and C-8 are
highly correlated: the regression coefficient b D 0.54ꢁ3ꢂ, the
correlation coefficient R D 0.977 and the standard deviation
from the regression s D 0.52. This implies that the steric effect
is transmitted from the carbonyl carbon C-7 to the terminal
methyl C-8 with reduced intensity. This is the only observed
direct correlation of SCS within the acetophenone series.
In an earlier investigation focused on C-7 shifts, sep-
aration of steric effects was attempted5 by referencing to
methyl-substituted benzenes. However, this was found to
be not very reliable: the angle ꢀ was ultimately estimated
without this separation.5 We achieved this separation by
means of the classical procedure2 recently described in
detail,12–15 primarily for methyl-substituted benzoic acids.
The method makes use of two approximations: there are
no steric effects in meta and para positions and polar effects
are equal in ortho and para positions. The latter approxi-
mation has been questioned and corrected24 but has been
found to be adequate for the methyl group where the polar
2,3,5,6
A
2,6
2,3
140
2,4,6
2,3,4
2
3
3,5
2,5
H
b
R
0.57(9)
0.884
3,4
2,4
135
SD 1.34
4
3,4,5
200
210
SCS C(7) ppm
Figure 1. Rough SCS of the C-1 atom plotted vs SCS of the
carbonyl atom C-7 in methyl-substituted acetophenones 1–14.
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

´
M. Budeˇsˇınsky´ et al.
848
2,3,5,6
2,6
10
B
A
10
2,4,6
2,3,4
2
5
5
2,3
2,5
2,4
b = -10.6(13)
R = 0.920
s = 1.83
b
R
0.91(2)
0.9958
SD 0.39
0
0
0
5
10
0.0
0.2
0.4
cos2 f
0.6
0.8
1.0
SCSst C(7) ppm
Figure 2. Steric component SCSst of the C-1 atom plotted vs
SCSst of the carbonyl atom C-7 in methyl-substituted
acetophenones 1–14. The values of SCSst separated
according to Eqn (3).
Figure 3. SCS (the steric component SCSst) of the carbonyl
carbon atom C-7 in methyl-substituted acetophenones 1–14
plotted vs the theoretical measures of the steric effect (cos2 of
the torsion angle ꢀ).
steric effects (Fig. 1). However, there is a close correlation of
SCSst since in both coordinates the same steric effect is seen
(Fig. 2). Several similar examples can be given comparing
different physical properties of the same set of compounds
or the same properties of two different sets.
O
CH3
O
CH3
C
C
+
+
(CH3)n
(CH3)n
The main goal of our investigation was to test the SIR
theory based on Eqn (1) or (2). These equations were used
in the literature3–8 to calculate the assumed, artificial values
of ꢀ from the experimental SCS; some values for methyl-
substituted acetophenones were estimated from 13C-7 SCS
by Dhami and Stothers.5 Previously, we calculated17 the
values of ꢀ within the framework of the DFT25 at the
B3LYP/6–311CG(d,p) level. Correlation of our steric com-
ponents, SCSst, with these ꢀ according to Eqn (1) is shown
in Fig. 3. According to Eqn (2), it would be even worse
ꢁR D 0.889ꢂ. The statistics given in Fig. 3 are not so bad
at first sight but the correlation suffers from an irregular dis-
tribution of points. The main defect is that it would require
non-negligible values of ꢀ for the 2,6 and 7, which are actu-
ally in planar conformations such as 2A. Instead of a series
in which ꢀ gradually increases, there are just two classes of
compounds: those with two ortho methyl groups (8,12 and
14) in non-planar, almost perpendicular, conformations and
the other, planar or nearly planar, conformations. Moreover,
the angle ꢀ has different meanings for these two classes of
compounds and the potential energy curves for the rota-
tion around the C(1)—C(7) bond have different shapes, as
shown in detail previously.17 For the first class this curve
ꢁ4ꢂ
Its reaction energy, 4E, is a measure of the total
substituent effect exerted on the acetyl group by all methyl
groups together. The quantity 4E has a clear physical
meaning and can be calculated unambiguously. In analogy
with SCS, even for 4E it is possible to separate steric and
polar effect as shown in Eqn (3). The steric component 4Est
is then a measure of the total steric effect. The plot of SCSst
vs 4Est (Fig. 4) reveals good linearity, much better than in
Fig. 3. This shows that the SCSst depend on all kinds of steric
effects. In non-planar derivatives the dependence is mainly
on the angle ꢀ, and in planar derivatives the dependence is
on the in-plane steric compression.
We conclude that the SCS are controlled partly by SIR,
which is critical for the non-planar derivatives 8,12 and
14. However, they cannot serve to determine directly the
torsion angle ꢀ since they are influenced also by the steric
hindrance even in planar derivatives. The angles calculated
from experimental SCS, in acetophenones4–6 and in other
compounds,3,7,8 are doubtful quantities, and are often at
variance with the real conformation.
Similar results to those based on SCS of C-7 could be
obtained from SCS of C-8 since the two sets are highly
correlated (see above), or from C-1 since at least the SCSst
are correlated (Fig. 2). Some supporting arguments can be
formulated by an empirical analysis from SCS of the carbon
atoms in the ortho position, C-2 and C-6; at least the planar and
non-planar molecules can be distinguished: when the atoms
C-2 or C-6 bear a methyl group, their SCS are 8.5–10.8 ppm
greater than in similar molecules without this methyl group,
when the molecules are planar. This effect is reduced almost
°
exhibits a fairly deep minimum near 90 ; for the second class
°
there is one shallow minimum near 0 and a much higher
°
secondary minimum between 140 and 150 . For all these rea-
sons, we searched for an energy-related quantity which better
describes the steric hindrance. We suggested17 using the
energy of the isodesmic12,14,15 (and homodesmotic26) reaction
[Eqn. (4)] in which a substituted acetophenone is synthesized
from acetophenone and a polymethyl-substituted benzene:
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

13C and 1H NMR of methylacetophenones and methyl benzoates
849
(Table 5) and the value measured by Marshall et al.28 was
found to be in error. In the low-frequency spectra, there was
probably a misassignment of the lines of C-2, C-6 and C-3,
C-5, which are separated by only 0.27 ppm (Table 1).
B
10
The data for 2 in Table 5 are most revealing. Since
3JꢁC-8,C-2ꢂ is greater than 3JꢁC-8,C-6ꢂ, the C-8 atom must
be in a position opposite to C-2. Hence the conformation
is planar or nearly planar. Small deviations from planarity
cannot be detected. In any case, the sp conformation 2A
strongly prevails or is practically the only one present. This
was noted already by Hansen et al.6d and is in agreement
with our DFT calculations.17 The zero values of 3J in 8
are also in agreement with the assumed, strongly non-
5
b = 0.322(18)
R = 0.981
s = 0.89
3
planar conformation. Very small values of JꢁC-8,C-2ꢂ and
0
3JꢁC-8,C-6ꢂ are expected, and in any case they should be
equal. The very small value (essentially zero) of 3JꢁC-8,C-2ꢂ in
1 was unexpected. We carried out high-level DFT calculations
of both chemical shifts (Table 1, in parentheses) and coupling
constants (Table 5). All calculated and experimental values
of J agree well. In particular, the calculated 1J and 2J confirm
the reliability of the theoretical model: they are only slightly
shifted but reproduce even the small differences between
the individual compounds. The values of 3JꢁC-8,C-2ꢂ D
C0.7 Hz and 3JꢁC-8,C-6ꢂ D ꢀ0.3 Hz are compatible with
the experimental values, too small to detect.
0
10
20
30
∆Est kJ mol-1
Figure 4. 13C SCS (the steric component SCSst) of the
carbonyl carbon atom C-7 in methyl-substituted
acetophenones 1–14 plotted vs the theoretical measures of
the steric effect [the calculated energy of the isodesmic
reaction Eqn (4)].
to zero in the non-planar molecules 8,12 and 14. It would be
possible to derive further empirical regularities of this kind,
but this reasoning is relatively complex and not as convincing
as the above statistical proofs and graphical representations.
Correlation of calculated and observed SCS is somewhat
worse (Table 1). For the eight carbon atoms of 1 we
obtained a reasonable fit, b D 0.999ꢁ19ꢂ, R D 0.9989,
s D 2.39 ppm, but it may not be sufficient for predicting
small differences. For the difference between the C-2 and C-6
atoms we obtained ꢀ1.06 ppm. Considering the relatively
low rotational barrier17 ꢁ23.6 kJ molꢀ1ꢂ, it is evident that the
two signals are coalesced. (The difference SCS2–6 can also
be estimated empirically from SCS2–6 in 2, accounting for
the shift produced by the 2-methyl substituent. We obtained
ꢀ1.0 ppm but the agreement is certainly fortuitous).
Conformation and the J(C,C) coupling constants
We intended to obtain a more direct proof of conformation
from the 3JꢁC,Cꢂ coupling constants using compounds
labeled with 13C. We were encouraged by the pioneering
work of Hansen and co-workers,6d,27 who found important
2
differences between the coupling constants JꢁC-7,C-2ꢂ and
3JꢁC-7,C-3ꢂ in many aryl ketones using compounds labeled
in position 7 (the carbonyl carbon atom). In contrast, we
investigated the constants 3JꢁC-8,C-2ꢂ on 1*,2* and 8* labeled
with 13C in position 8 (the acetyl methyl group). The axis
of rotation is now in the middle of the three-bond system
and we expected, according to the basic Karplus equation,
a greater value of 3JꢁC-8,C-2ꢂ (in the ap position) than of
3JꢁC-8,C-6ꢂ (in the sp position) in 2*, an intermediate value
in 1* (rapid exchange) and a small value of 3JꢁC-8,C-2ꢂ in 8*
(sc or ac, almost perpendicular position). From the reported28
value of 8.4 Hz for 3JꢁC-8,C-2ꢂ in 1*, we assumed that all
3JꢁC,Cꢂ will be sufficiently large for safe conclusions to be
drawn about the conformation. This was not confirmed.
Three of the 3JꢁC,Cꢂ values were indistinguishable from zero
13C SCS and conformation of methyl benzoates
Concerning the methyl-substituted methyl benzoates 15–28,
we expected a great similarity with acetophenones. Correla-
tion of the SCS in equivalent positions is always very close:
for the carbon atoms C-2, C-3, C-4, C-7 and C-8 R is between
0.982 and 0.995. The regression coefficients b were not sig-
nificantly different from unity with an important exception
of the carbonyl carbon atom C-7 (Fig. 5). The much smaller
sensitivity to substituent steric effects in methyl benzoates
agrees with the smaller steric requirements of the COOCH3
group compared with the COCH3 group. This is reflected
even in the correlation of C-1 ring carbon atoms: b D 0.85ꢁ6ꢂ,
R D 0.967 and s D 0.64 ppm.
Table 5. Selected carbon–carbon coupling constants (Hz) of substituted acetophenonesa
No.
1JꢁC-8,C-7ꢂ
1JꢁC-7,C-1ꢂ
2JꢁC-8,C-1ꢂ
3JꢁC-8,C-2ꢂ
3JꢁC-8,C-6ꢂ
1*
2*
8*
43.0 (41.2)
42.0 (40.7)
40.2 (39.1)
52.5 (50.7)
52.6 (51.3)
52.2 (50.1)
13.7 (15.3)
13.2 (15.1)
12.0 (13.4)
¾0 (0.7)
1.5 (1.3)
¾0 (0.3)
¾0 (ꢀ0.3)
¾0 (0.1)
¾0 (ꢀ0.3)
a In parentheses are the values calculated at the B3LYP/6–311CCG(2d,2p)//B3LYP/6–311CGꢁd,pꢂ level.
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

´
850
M. Budeˇsˇınsky´ et al.
found to be inadequate in all examples re-examined hitherto,
based either on the NMR spectra or previously on the UV
spectra.17 Some molecules do not possess the conformations
attributed by this theory; moreover, the angle ꢀ is not
an appropriate physical quantity. Instead of a continuous
series of ꢀ values, a better description distinguishes only
two groups of compounds: nearly planar ꢁꢀ ³ 0ꢂ and non-
172
170
168
2,3,5,6
2,4,6
2,6
°
planar (ꢀ not far from 90 ). (Some more sensitive observable
2,3,4
quantities, for instance the IR spectral shifts,29 permit a
subclass to be distinguished: slightly non-planar molecules
with a buttressing effect of two substituents in positions 2
and 3.)
In any new application of this theory, it is advisable to
determine the actual conformation of each derivative by
direct methods; quantum chemical calculations are very
effective. The strength of the steric hindrance is better
expressed in energy terms (reaction energy of an isodesmic
reaction) than in any geometric parameter (angle ꢀ). The
numerous values of ꢀ reported in the literature should not
be given credence.
Concerning the conformation of the title compounds,
it is now in our opinion determined safely from several
experimental quantities; the 13C NMR shifts have also been
very significant. However, the quantum chemical calcula-
tions have been essential for determining all conformational
details.
2,3
2,5
2,4
b = 0.34(2)
R = 0.987
s = 0.25
2
200
205
SCS C(7) acet. ppm
210
Figure 5. SCS of the carbonyl atom C-7 in methyl-substituted
acetophenones 1–14 plotted vs SCS of the C-7 atom in
methyl-substituted methyl benzoates 15–28.
It follows that the conformation of methyl benzoates
15–28 can be determined from SCS, particularly from C-7, in
the same way as the conformation of 1–14 and with the same
results: division into two classes, planar and non-planar, is
central. This conformation is in accord with that of the parent
acids;12–15 in the case of methyl 2-methylbenzoate 16 it was
determined by gas-phase electron diffraction.18
EXPERIMENTAL
Synthesis of compounds
Syntheses of the substituted acetophenones 1–14 have been
described.17 Methyl benzoates 15–28 were prepared from
pure samples of benzoic acids15 by esterification with
diazomethane and were not further purified except for
simple vacuum distillation.
The 13C-labeled acetophenones were prepared from
[13C]iodomethane (Aldrich) by standard methods30 reported
for the unlabeled compounds: [8-13C]acetophenone
1* and [8-13C]-2-methylacetophenone 2* from labeled
methyl magnesium iodide and benzonitrile or 2-
methylbenzonitrile, respectively30a (yield 33 or 40%), [8-
13C]-2,6-dimethylacetophenone 8* from labeled dimethyl-
cadmium and 2,6-dimethylbenzoyl chloride as described for
2-ethylacetophenone30b [yield 38% after chromatography on
silica with hexane—diethylether (1 : 1) as eluent].
1H SCS
Of the 1H SCS of 1–14 (Table 3), only the shifts of the CH3CO
group are present in all derivatives and can be related to
structure and conformation. Of their possible correlation
with the 13C SCS, only that with C-7 is significant and has a
reversed slope [b D ꢀ0.0105ꢁ14ꢂ, R D 0.907, s D 0.022 ppm].
However, it arises primarily from inductive effects since
there is no improvement for the steric components calculated
according to equations similar to Eqn (3). The hydrogen
atoms are evidently too far from the axis of rotation and
are influenced differently to the proximate atoms. They are
sensitive to strong deviations from planarity but not to steric
compression in the planar conformation. This observation is
confirmed by the behavior of the SCS for CH3O in 15–28.
These quantities are only slightly correlated with 13C(O) SCS
of the same compounds [b D 0.011ꢁ2ꢂ, R D 0.832]. In contrast,
there is a correlation between the 1H SCS of this methyl group
in 1–14 and in 15–28 (Table 4): b D ꢀ0.56ꢁ6ꢂ, R D 0.943,
s D 0.0067 ppm; it is controlled essentially by the polymethyl
derivatives 8,12,14,22,26 and 28. More precise results cannot
be obtained since the values of 1H SCS are generally too
small. Hence their significance for determining conformation
is negligible.
NMR measurements
The NMR spectra were measured on a Varian UNITY 500
Fourier transform NMR spectrometer (1H at 500 MHz; 13C
°
at 125.7 MHz) in CDCl3 at 20 C. The proton chemical shifts
were referenced to internal tetramethylsilane and carbon-13
chemical shifts to the signal of the solvent using the relation
υꢁCDCl3ꢂ D 77.0. Characteristic chemical shifts, coupling
constants and/or substituent effects were used for structural
assignment of protons. Proton NMR spectra were measured
with a spectral width 5 kHz and an acquisition time of
5 s. Zero filling to 128 K data points and Lorentz–Gauss
function (with parameters lb D ꢀ1.5 and gf D 0.75) were
used for resolution enhancement. Carbon-13 NMR spec-
tra were accumulated with broadband proton decoupling
CONCLUSIONS
The concept of SIR is evidently valid as a qualitative
principle within the framework of the resonance theory.
Its quantitative version operating with variable angle ꢀ was
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851

13C and 1H NMR of methylacetophenones and methyl benzoates
851
9. For instance: (a) Wepster BM. Recl. Trav. Chim. Pays-Bas 1957;
76: 335; (b) Yates K, Scott BF. Can. J. Chem. 1963; 41: 2320;
(c) Krueger PJ. Can. J. Chem. 1973; 51: 1363.
10. (a) Zaitsev BA. Izv. Akad. Nauk SSSR, Ser. Khim. 1974; 780;
(b) Lumbroso H, Liegeois C, Goethals G, Uzan R. Z. Phys. Chem.
(Wiesbaden) 1983; 138: 167.
11. Kukol A, Strehle F, Thielking G, Gru¨tzmacher HF. Org. Mass
Spectrom. 1993; 28: 1107.
12. Decouzon M, Ertl P, Exner O, Gal J-F, Maria P-C. J. Am. Chem.
Soc. 1993; 115: 12 071.
13. Fiedler P, Exner O. J. Phys. Org. Chem. 1998; 11: 141.
14. (a) Decouzon M, Gal J-F, Maria P-C, Bo¨hm S, Jime´nez P,
Roux MV, Exner O. New J. Chem. 1997; 21: 561; (b) Otyepkova´ E,
Neveˇcˇna´ T, Kulha´nek J, Exner O. J. Phys. Org. Chem. 2003; 16:
721.
using spectral width 30 kHz, acquisition time 1 s and zero
filling to 128 K data points before Fourier transformation.
Characteristic chemical shifts, relative intensities and known
substituent effects were used for structural assignment of
carbon signals. In ambiguous cases we confirmed the assign-
ment with the heterocorrelated 2D-1H,13C-HMQC spectra.
The 1JꢁC-7,C-1ꢂ couplings in 1, 2 and 8 were determined
from 1D-INADEQUATE spectra31 using following parame-
ters: spectral width ¾27 kHz, 13C transmitter pulses 7.25 µs
°
°
(90 ) and 14.5 µs (180 ), relaxation delay d1 D 5.0–7.0 s,
delay d2 D 5.0 ms calculated for 1J(C,C) D 50 Hz, broadband
proton decoupling, acquisition time 2–2.5 s and number of
scans 4000–7000. Zero filling to 256 K data points and an
exponential weighing function with line broadening 0.2 Hz
were used for data processing.
15. Bo¨hm S, Exner O. Chem. Eur. J. 2000; 6: 3391.
´
16. Cısarˇova´ I, Podlaha J, Bo¨hm S, Exner O. Collect. Czech. Chem.
Commun. 2000; 65: 216.
17. Kulha´nek J, Bo¨hm S, Pala´t K Jr, Exner O. J. Phys. Org. Chem. 2004;
17: in press.
18. Hnyk D, Borisenko KB, Samdal S, Exner O. Eur. J. Org. Chem.
2000; 2063.
19. (a) Montaudo G, Caccamese S, Librando V, Maravigna P. Gazz.
Chim. Ital. 1975; 105: 85; (b) Jin O, Wildman TA. J. Phys. Chem.
1991; 95: 20.
20. (a) Baliah V, Aparajithan K. Tetrahedron 1963; 19: 2177;
(b) Benassi R, Iarossi D. J. Chem. Soc., Perkin Trans. 2 1981; 228.
21. Abraham RJ, Angiolini S, Edgar M, Sancassan F. J. Chem. Soc.,
Perkin Trans. 2 1995; 1973.
Calculations
The calculations of chemical shifts were carried out within
the framework of the DFT25 using the Gaussian 03 program.32
Geometries of all compounds were optimized at the
B3LYP/6–311CG(d,p) level, the absolute shielding constants
were calculated at the B3LYP/6–311CCG(2d,2p)//B3LYP/
6–311CG(d,p) level using the GIAO method. The shielding
constants of TMS were obtained in the same way.
The coupling constants were calculated according to
recent suggestions33 based on analytical implementation of
the NMR indirect nuclear spin–spin coupling constants at
the DFT level. This implementation involves all four contri-
butions of the non-relativistic Ramsey theory: diamagnetic
and paramagnetic spin–orbit contributions and paramag-
netic Fermi contact and spin–dipole contributions. With
respect to known RHF instability problems, we used the
B3LYP (hybrid BLYP) exchange-correlation functional with
the 6–311CCG(2d,2p) basis set.
22. (a) Hachiya I, Moriwaki M, Kobayashi S. Bull. Chem. Soc. Jpn.
1995; 68: 2053; (b) Brook MA, Henry C. Tetrahedron 1996; 52: 861.
23. Andreou AD, Bulbulian RV, Gore PH. Tetrahedron 1980; 36: 2101.
24. (a) Pytela O, Prusek O. Collect.Czech. Chem. Commun. 1999; 64:
1617; (b) Thompson HW, Steel G. Trans. Faraday Soc. 1956; 52:
1451; (c) Bo¨hm S, Fiedler P, Exner O. New J. Chem. 2004; 28: 67.
25. Becke AD. J. Chem. Phys. 1993; 98: 5648.
26. George P, Trachtman M, Bock CW, Brett AM. J. Chem. Soc., Perkin
Trans. 2 1976; 1222.
27. (a) Hansen PE. Org. Magn. Reson. 1978; 11: 215; (b) Marshall JL,
Faehl LG, Kattner R, Hansen PE. Org. Magn. Reson. 1979; 12: 169.
28. Marshall JL, Faehl LG, Kattner R. Org. Magn. Reson. 1979; 12:
163.
Acknowledgements
29. Fiedler P, Exner O. Collect. Czech. Chem. Commun. 2004; 69: 79.
30. (a) Haworth RD, Jones B, Way YM. J. Chem. Soc. 1943; 10;
(b) Elsner BB, Strauss HE, Forbes EJ. J. Chem. Soc. 1957; 578.
31. Bax A, Freeman R, Kemsell SP. J. Am. Chem. Soc. 1980; 102: 4849.
32. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN,
Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Men-
nucci B, Cossi M, Scalmani G, Rega N, Petersson GA, Nakat-
suji H, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J,
Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Klene M,
Li X, Knox JE, Hratchian HP, Cross JB, Adamo C, Jaramillo J,
Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R,
Pomelli C, Ochterski JW, Ayala PY, Morokuma K, Voth GA,
Salvador P, Dannenberg JJ, Zakrzewski VG, Dapprich S,
Daniels AD, Strain MC, Farkas O, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clif-
ford S, Cioslowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P,
Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA,
Peng CY, Nanayakkara A, Challacombe M, Gill PMW, John-
son B, Chen W, Wong MW, Gonzalez C, Pople JA. Gaussian 03,
Revision B.03. Gaussian: Pittsburgh, PA, 2003.
This work was carried within the framework of the research project
Z4055905 of the Academy of Sciences of the Czech Republic and
was supported by the Ministry of Education of the Czech Republic
(Project LN00A032, Center for Complex Molecular Systems and
Biomolecules).
REFERENCES
1. Birtles RH, Hampson GC. J. Chem. Soc. 1937; 10.
2. Braude EA, Sondheimer F, Forbes WF. Nature (London) 1954; 173:
117.
3. Ahlbrecht H, Du¨ber EO, Epsztajn J, Marcinkowski RMK.
Tetrahedron 1984; 40: 1157.
4 (a) Benassi R, Iarossi D, Folli U, Schenetti L, Taddei F. J. Chem.
Soc., Perkin Trans
2 1981; 228; (b) Grimaud M, Pfister-
Guillouzo G. Org. Magn. Reson. 1975; 7: 386.
5. Dhami KS, Stothers JB. Can. J. Chem. 1965; 43: 479.
6. (a) Goethals G, Uzan R, Nadjo L, Doucet JP. J. Chem. Soc., Perkin
Trans. 2 1982; 885; (b) Oakley MG, Boykin DW. J. Chem. Soc.,
Chem. Commun. 1986; 439; (c) Solladie´-Cavallo A, Solladie´ G.
Org. Magn. Reson. 1977; 10: 235; (d) Hansen PE, Poulsen OK,
Berg A. Org. Magn. Reson. 1977; 9: 649.
33. (a) Helgaker T, Watson M, Handy NC. J. Chem. Phys. 2000; 113:
9402; (b) Sychrovsky V, Grafenstein J, Cremer D. J. Chem. Phys.
2000; 113: 3530; (c) Barone V, Peralta J, Contreras RH, Snyder JP.
J. Chem. Phys., Sect. A 2002; 106: 5607.
7. (a) Leibfritz D. Chem. Ber. 1975; 108: 3014; (b) Guilleme J, Diez E,
Bermejo FJ. Magn. Reson. Chem. 1985; 23: 442.
8. Baumstark AL, Balakrishnan P, Dotrong M, McCloskey CJ,
Oakley MG, Boykin DW. J. Am. Chem. Soc. 1987; 109: 1059.
Copyright  2004 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2004; 42: 844–851