Infrared and Nuclear Magnetic Resonance Properties of Benzoyl Derivatives of Five-membered Monoheterocycles and Determination of Aromaticity Indices

Article abstract of DOI:10.1002/jhet.5570400504

Benzophenones, 2-benzoylthiophenes, 2-benzoylpyrroles, and 2-benzoylfurans, which have substituents at m- and p-positions of the benzoyl ring were prepared and their ir and nmr spectra were obtained in 0.1 M chloroform-d solution. The chemical shift values of each series were plotted against the Hammett substituent parameters to give good correlation, with the exception of the ortho-Hs and -Cs. The slopes as well as the differences in chemical shift gave sets of meaningful values for the indices of aromaticy.

Full text of DOI:10.1002/jhet.5570400504

Sep-Oct 2003
Infrared and Nuclear Magnetic Resonance Properties of Benzoyl
Derivatives of Five-membered Monoheterocycles and Determination
of Aromaticity Indices
763
Kyu Ok Jeon, Jung Ho Jun, Ji Sook Yu, and Chang Kiu Lee*
Department of Chemistry, Kangwon National University, Chuncheon 200-701, S. Korea
Received April 23, 2003
Benzophenones, 2-benzoylthiophenes, 2-benzoylpyrroles, and 2-benzoylfurans, which have substituents
at m- and p-positions of the benzoyl ring were prepared and their ir and nmr spectra were obtained in 0.1 M
chloroform-d solution. The chemical shift values of each series were plotted against the Hammett sub-
stituent parameters to give good correlation, with the exception of the ortho-Hs and -Cs. The slopes as well
as the differences in chemical shift gave sets of meaningful values for the indices of aromaticy.
J. Heterocyclic Chem., 40, 763(2003).
There are a variety of methods for estimating the aro-
In order to complete the estimation of the aromaticity
indices of the five-membered monoheterocyclic com-
pounds under the same conditions we needed to have sets
of substituted phenyl derivatives of the heterocycles for
thiophene, pyrrole and furan. Therefore, we prepared the
benzoyl derivatives of the heterocycles and obtained their
NMR spectra. This paper reports the results of our
attempts to obtain the indices of aromaticity.
maticity of five-membered monoheterocyclic compounds,
namely thiophene, furan, and pyrrole. The degree of aro-
maticity is typically compared to that of benzene whose
aromaticity index is set as 1 [1]. A set of aromaticity
indices of 1.00, 0.75, 0.59, and 0.46 were reported for ben-
zene, thiophene, pyrrole, and furan, respectively, by com-
parison of the ring currents measured by the chemical
shifts of the protons [2]. Consideration of bond length and
the number ofp electrons in the ring gives a different set of
the indices: benzene 1, thiophene 0.93, pyrrole 0.91, and
furan 0.87 [3]. There are a number of sets of the indices
reported in the literature [1-3].
We have reported a method of estimating the aromaticity
index by comparing the chemical shifts of the carbonyl
carbon atoms of m- and p-substituted phenyl esters of the
five-membered heterocyclic carboxylic acids to those of
benzoic acids [4]. The results were: benzene 1.00, thio-
Results and Discussions.
The m- and p-substituted benzoyl compounds 1-4 were
prepared by Friedel-Crafts benzoylation of benzene, thio-
phene, pyrrole, and furan in 30-85% yields. The ketones
were purified by either recrystallization or column chro-
matography. Their ir, nmr, and mass spectra were consis-
tent with the desired structures. Positions 3-, 4-, and 5- of
the 2-substituted heterocycles can be considered as ortho,
meta, and para, respectively, and such notation has been
used throughout the present report.
phene 0.85, and furan 0.75 in dimethyl sulfoxide-d ; and
6
benzene 1.00, thiophene 0.90, and furan 0.78 in chloro-
form-d. A similar approach with m- and p-substituted
anilides gave a slightly different set of the indices: benzene
1.00, thiophene 0.79, and furan 0.52 in dimethyl sulfoxide-
In order to have a good correlation of the chemical shifts
and the Hammett s values, it is essential to observe the
nmr spectra at the same concentration (0.1 M). Therefore,
the purity of the compound was found to be critical [4,5].
1
13
d [5]. The poor solubilities of the anilides in chloroform-
Signals of H and C shift linearly upfield as the concen-
6
d did not allow determination of a set of indices in that sol-
vent. Our effort to estimate the index of aromaticity of
pyrrole by similar method was not successful because of
the difficulty of preparing the ester, m- and p-substituted
phenyl 2-pyrrolecarboxylates and the 2-pyrrolylanilides.
These esters and amides were very difficult to purify
because they hydrolyze readily.
tration increases. The plots (not shown) of d against con-
H
centration show slopes of –9.3, -15.0, and –15.8 Hz/M for
o-, m-, and p-H, respectively, of 1k. The correlation coef-
ficients are better than 0.999 for all cases. Similar plots of
d
show slopes: CO –24.5; i-C 16.9; o-C –16.9; m-C
C
1
13
–13.7; p-C –13.9 Hz/M. The H and C nmr chemical
shift values of 1-4 are listed in Tables 1 and 2, respectively.

764
K. O. Jeon, J. H. Jun, J. S. Yu and C. K. Lee
Vol. 40
Table 1 (continued)
Table 1
1
H Chemical Shift Values of Substituted Phenyl Ketones 1-4 in
Chloroform-d (0.1 M)
o-H
m-H
p-H 2’-H 3’-H 4’-H 5’-H 6’-H CH
3
2k 7.65
3a 6.92
3b 6.89
3c 6.89
3d 6.92
3e 6.89
3f 6.87
3g 6.86
3h 6.87
3i 6.89
3j 6.89
3k 6.90
4a 7.38
4b 7.27
4c 7.27
4d 7.24
4e 7.21
4f 7.34
4g 7.26
4h 7.25
4i 7.23
4j 7.21
4k 7.24
7.17
6.41
6.36
6.36
6.34
6.34
6.39
6.35
6.35
6.34
6.33
6.34
6.68
6.62
6.61
6.60
6.58
6.67
6.61
6.60
6.59
6.57
6.60
7.73 7.87 7.50 7.60
o-H
m-H
p-H 2’-H 3’-H 4’-H 5’-H 6’-H CH
3
7.25 8.75
7.18 8.03
7.18 7.87
8.43 7.70 8.23
7.69 7.36 7.82
7.54 7.42 7.78
1a 7.81
1b 7.79
1c 7.79
1d 7.81
1e 7.80
1f 7.80
1g 7.77
1h 7.77
1i 7.75
1j 7.78
1k 7.81
2a 7.66
2b 7.64
2c 7.64
2d 7.68
2e 7.68
2f 7.63
2g 7.62
2h 7.63
2i 7.64
2j 7.65
7.54
7.50
7.50
7.48
7.48
7.53
7.49
7.49
7.47
7.47
7.48
7.23
7.19
7.18
7.16
7.12
7.22
7.18
7.18
7.16
7.16
7.66 8.62
7.62 7.94
7.61 7.78
8.45 7.71 8.14
7.72 7.37 7.71
7.56 7.43 7.67
7.15 7.42
7.15 7.70
7.11 7.39 7.48 3.87
7.39 7.37 7.70 2.43
7.59 7.36
7.14 7.38 7.34 3.86
7.24 8.03 8.35
7.17 7.78 7.63
7.17 7.86 7.46
7.13 7.94 6.99
7.14 7.83 7.29
7.58 7.63
7.40 7.36 7.58 2.42
7.66 7.94 8.34
7.60 7.68 7.63
7.60 7.76 7.46
7.56 7.83 6.96
7.57 7.74 7.28
7.59
3.89
2.44
3.89
2.44
7.16 7.91 7.49 7.57
7.77 8.88
8.48 7.72 8.35
7.73 8.11
7.72 7.38 7.91
7.82 8.71
7.76 7.99
7.76 7.83
7.72 7.38
8.46 7.73 8.20
7.72 7.38 7.78
7.57 7.44 7.74
7.13 7.41 7.45
7.34 7.36 7.63
7.72
795
7.55 7.43 7.86
7.71 7.48
7.13 7.40 7.56 3.87
7.70 7.75
7.40 7.37 7.70 2.43
7.76 8.31 8.36
7.71 7.87 7.64
7.70 7.94 7.46
7.69 8.03 6.99
7.68 7.88 7.28
7.61 7.66
7.83 8.02 8.36
7.74 7.74 7.65
7.74 7.82 7.48
7.69 7.91 6.99
7.70 7.79 7.30
7.71 7.97 7.50 7.60
Table 2
C Chemical Shift Values of Substituted Phenyl Ketones1-4 in Chloroform-d (0.1 M)
13
C=O
i-C
o-C
m-C
p-C
1’-C
2’-C
3’-C
4’-C
5’-C
6’-C
CH
3
1a
1b
1c
1d
1e
1f
1g
1h
1i
194.15
195.15
195.24
196.53
196.96
194.77
195.61
195.46
195.54
196.49
196.77
185.59
186.58
186.70
187.95
188.03
186.28
187.05
186.90
186.86
187.93
188.24
181.94
183.06
183.19
184.54
185.05
182.53
183.56
183.44
183.62
184.58
184.85
179.60
180.76
180.84
182.29
182.77
136.23
136.88
136.93
137.58
137.73
136.26
137.25
137.23
138.26
137.93
137.56
142.36
142.96
142.99
143.53
143.34
142.60
143.03
143.17
143.81
143.76
143.60
130.30
130.66
130.70
131.08
131.20
130.56
130.77
130.80
131.15
131.20
131.11
151.84
151.90
151.90
152.17
152.25
129.99
130.00
130.00
130.02
130.02
130.67
129.92
128.90
129.70
130.29
130.04
135.54
135.12
135.07
134.86
133.90
135.71
134.79
134.75
133.99
134.47
134.84
120.28
119.84
119.82
119.48
119.41
120.37
119.52
119.46
118.55
119.04
119.52
121.24
120.92
120.90
120.69
120.49
128.72
128.45
128.44
128.24
128.21
128.66
128.39
128.38
128.16
128.18
128.25
128.43
128.14
128.13
127.94
127.73
128.40
128.06
128.04
127.74
127.84
127.94
111.74
111.31
111.29
111.03
110.93
111.70
111.22
111.20
110.80
110.88
111.02
112.80
112.42
112.40
112.18
112.10
133.35
132.83
132.82
132.41
132.31
133.45
132.66
132.62
131.86
132.13
132.40
135.29
134.82
134.80
134.24
134.59
135.48
134.57
134.52
133.40
133.82
134.20
126.69
125.89
125.89
125.29
125.16
126.68
125.68
125.62
124.65
124.95
125.30
147.77
147.43
147.41
147.16
147.02
139.04
139.45
139.24
138.86
138.13
142.86
136.28
135.82
130.14
134.86
137.56
139.41
139.88
139.67
139.37
138.62
143.30
136.80
136.37
130.65
135.38
138.11
139.65
140.12
139.92
139.59
138.31
143.61
137.04
136.61
130.91
135.61
138.32
138.27
138.88
138.64
138.48
138.25
124.70
132.77
129.88
114.28
130.43
130.07
131.60
131.44
132.54
129.91
130.04
123.95
131.98
129.11
113.71
129.93
129.91
131.71
130.56
131.59
129.37
129.15
123.84
131.82
129.65
113.51
129.43
129.76
131.59
130.33
131.15
129.09
128.94
124.36
132.18
129.73
113.79
129.64
148.07
122.55
134.55
159.54
137.60
123.52
131.54
128.62
113.52
128.95
128.25
148.04
122.60
134.61
159.56
137.75
123.68
130.68
128.74
113.66
129.07
128.39
148.08
122.49
134.48
159.51
138.13
123.57
130.47
128.63
113.59
128.99
128.30
148.13
122.58
134.57
159.59
137.25
126.70
135.25
132.33
118.85
133.17
149.81
127.49
138.87
163.19
143.22
132.40
126.61
135.09
132.21
118.59
132.97
149.77
127.24
138.68
163.07
143.03
132.25
126.23
134.69
131.78
118.17
132.59
149.59
126.69
138.15
162.72
142.46
131.82
126.93
135.42
132.49
118.93
133.31
129.62
129.86
129.61
129.19
128.06
135.42
128.53
128.08
122.86
127.35
55.45
21.34
55.47
21.64
1j
1k
2a
2b
2c
2d
2e
2f
2g
2h
2i
129.80
130.00
129.04
129.37
127.94
134.69
127.64
127.21
121.72
126.06
55.44
21.66
55.47
25.61
2j
2k
3a
3b
3c
3d
3e
3f
3g
3h
3i
129.59
129.90
128.94
129.30
128.13
134.56
127.46
127.02
121.50
126.16
55.40
21.38
55.42
21.57
3j
3k
4a
4b
4c
4d
4e
129.71
129.99
129.27
129.39
128.19
135.02
127.82
127.37
121.82
126.43
55.45
21.31

Sep-Oct 2003
Infrared and Nuclear Magnetic Resonance Properties of Benzoyl Derivatives
765
Table 2(continued)
13
C Chemical Shift Values of Substituted Phenyl Ketones1-4 in Chloroform-d (0.1 M)
C=O
i-C
o-C
m-C
p-C
1’-C
2’-C
3’-C
4’-C
5’-C
6’-C
CH
3
4f
180.35
181.24
180.99
181.14
182.18
182.54
151.86
152.12
152.09
152.60
152.34
152.24
121.43
120.59
120.51
119.64
120.09
120.55
112.80
112.38
112.32
112.03
112.04
112.17
147.89
147.22
147.15
146.51
146.78
147.08
142.10
135.85
135.36
129.80
134.49
137.22
130.28
131.73
130.70
131.70
129.37
129.24
123.60
130.87
128.68
113.68
129.04
128.38
149.96
127.68
138.96
163.26
143.32
132.54
4g
4h
4i
4j
4k
55.46
21.55
The assignments of each peak were made by analysis of
H- H COSY and H- C HETCOR spectra. In addition,
replacing the hydrogen of benzaldehyde with phenyl,
2-thienyl, 2-pyrrolyl, and 2-furyl rings the C=O signals
1
1
1
13
13
the correlations of the values of the heterocycles against
those of the corresponding benzene derivatives make the
assignment unambiguous. Such plots for the protons and
carbons in the substituted phenyl group (like Figure 4)
show slopes of near unity with correlation coefficients of
0.999-1.000. This could also be used to make accurate
assignments.
The average values of H and C chemical shifts for
each series of 11 derivatives (a-k) are listed in Table 3.
For the purpose of comparison the chemical shift values of
benzene, thiophene, pyrrole, and furan in chloroform-d at
0.1 M concentration are also listed in the Table.
shift by +3.65, -5.05, -8.48, and -10.81 ppm, respectively,
with the minus sign indicating upfield shift. Unlike the
benzene ring, which causes a downfield shift, the hetero-
cyclic rings induce shielding of the carbonyl carbon atom
with oxygen, the most electronegative oxygen atom, caus-
ing the most upfield shift. This also seems consistent with
the observation of the upfield shift of the C=O in vari-
ous benzoyl derivatives having electron-withdrawing sub-
stituents at m- and p-positions [7].
13
1
13
Recently, Neuvonen, et al. reported inverse correlation
13
of the C=O shift of m- and p-substituted phenyl esters of
acetic acid (5), dichloroacetic acid (6), and trifluoroacetic
acid (7) with the Hammett s parameters, and proposed a
novel concept to explain the reversed effect of the elec-
tron-withdrawing substituents [8].
Table 3
Averaged Chemical Shift Values of the Benzoylketones 1-4 in
Chloroform-d (0.1 M) and Their Differences
1
2
3
4
2,5( )-H[a] 7.36
3,4( )-H[a] 7.36
7.35
7.13
0.22
7.65 ± 0.03
7.18 ± 0.05
7.74 ± 0.09
0.52
6.85
6.26
0.59
6.89 ± 0.03
6.36 ± 0.05
7.17 ± 0.08
0.63
7.45
6.39
1.06
7.26 ± 0.12
6.61 ± 0.07
7.72 ± 0.05
0.87
( - )
Ortho-H
Meta-H
Para-H
(o- )
0.00
7.79 ± 0.02
7.49 ± 0.05
7.60 ± 0.06
0.43
They concluded that electron-withdrawing substituents
destabilize the carbonyl derivatives in ground-state by
decreased resonance stabilization. Although the correla-
2,5( )-C[a] 128.33
3,4( )-C[a] 128.33
125.11
126.86
-1.75
117.16
108.16
9.00
142.52
109.44
33.08
( - )
C=O
0
tion was considered fair by the authors the correlation of
195.70 ± 1.55 187.10 ± 1.51 183.67 ± 1.73 181.34 ± 1.74
137.25 ± 1.01 143.19 ± 0.83 130.87 ± 0.33 152.12 ± 0.48
129.96 ± 0.26 134.82 ± 0.92 119.57 ± 1.02 120.64 ± 1.00
128.37 ± 0.35 128.04 ± 0.39 119.19 ± 0.55 112.33 ± 0.47
132.62 ± 0.83 134.52 ± 0.96 125.62 ± 1.07 147.22 ±0.67
13
C=O with n
was reported to have a negative slope
Ipso-C
Ortho-C
Meta-C
Para-C
(i-o)
C=O
2
(–0.07) with r = 0.8142 for m- and p-substituted phenyl
acetate (5) [8]. The values of the carbonyl stretching fre-
quencies seem to be consistent with the conventional
thinking that the force constant of C=O bond increases by
the presence of an electron-withdrawing group whereas
the bond distances decrease [9].
7.29
8.37
11.30
31.48
[a] Parent aromatic compounds.
However, reexamination of the values reported in the
Supporting Information [8] and the values we obtained for
the same series of 5 with 11 substituents [5] can hardly be
The difference in chemical shifts of the ortho- and
meta-Hs of the five-membered heterocycles are consid-
ered to be a measure of relative aromaticity [6]. The
effect of the heterocycles on the chemical shift of protons
and carbons in the unsubstituted benzoyl groups of 1k,
2k, 3k, and 4k is quite remarkable. For example, by
considered a fair correlation of n
with the Hammett s .
C=O
-
A plot of n
1
vs. s for the ester shows a slope of 4.73 cm
C=O
with r = 0.428, while a plot of d
vs. s shows a slope
C=O
of 1.35 ppm with r = 0.992. The observation strongly

766
K. O. Jeon, J. H. Jun, J. S. Yu and C. K. Lee
Vol. 40
suggests that the substituent should affect differently the
chemical shift and the carbonyl stretching frequency.
13
The C=O signals of unsubstituted (Z = H) 5, 6, and 7
appear at 169.49, 163.02, and 155.85 ppm, respectively,
-1
while their n
signals are at 1765, 1777, and 1800 cm ,
C=O
respectively [8]. Plot of the values shows the reverse cor-
relation in 5-7 with a slope of –0.379 (r = 0.989) as in
Figure 1. Apparently, a strongly electron-withdrawing
group bonded to the carbonyl carbon causes shielding of
the carbon nucleus, but it makes the force constant of the
C=O bond greater than an electron-donating group does.
13
Figure 2. Correlation between the C=O frequency and the C chemical
shift of the carbonyl carbon in 1k-4k . The slope of the straight line is
0.346 and r = 0.998.
Because of the striking contrast between Figures 1 and
2, the n
values of 1-5 were examined. The results are
C=O
listed in Table 4.
As shown in Table 4 the average frequency of meta-sub-
-1
stituted benzophenones 1a-e is 1657 ± 4 cm whereas that
-1
of para-substituted compounds 1f-j is 1650 ± 2 cm .
Since the unsubstituted benzophenone 1k shows a fre-
-1
quency of 1652 cm , it can be concluded that the presence
of a substituent at the para position reduces the force con-
stant of the C=O bond. A substituent at the meta position
strengthens it regardless of the electron-withdrawing or -
donating nature of the substituent. Similar trends are
apparent with the series 2, 3, 4, and 5 as shown in Table 4.
The difference in the average values of C=O frequencies
13
Figure 1. Correlation between the C=O frequency and the C chemical
shift of the carbonyl carbon in 5-7. The slope of the straight line is -0.379
and r = 0.989.
In contrast, the aryl and heteroaryl rings show a mixed
signals of 1-4. As men-
tioned earlier, the C=O signals shift to upfield going
values shift to lower wave num-
bers: 1652 cm for 1k, 1629 cm for 2k, and 1614 cm
for 3k. However, 4k is an exception, showing a value of
1647 cm , which is the largest among the heterocyclic
ketones. A plot of the chemical shifts against the stretch-
ing frequencies for the carbonyl groups of 1k, 2k, and 3k
shows an excellent correlation (r = 0.998) with a positive
slope of 0.346, but the furan compound 4k shows a clear
exception as shown in Figure 2.
13
effect on the C=O and n
C=O
between meta- and para-substituted compounds are 5-7
13
-1
cm for 1-5. Therefore, the differences in the n
values
C=O
from 1 to 4, but the n
C=O
of 1k, 2k, 3k, and 4k should originate from the presence of
the phenyl and the heteroaryl rings.
Aryl rings are generally known to have electron-donat-
ing ability through conjugation forms like I, and the fre-
quency of carbonyl stretching decreases from acetophe-
-1
-1
-1
-1
-1
-1
none (1685 cm ) to benzophenone (1652 cm ). 2-
Benzoylpyrrole (3k) shows the smallest n value (1614
C=O
-1
cm ) among 1k-4k, indicating that conjugation is most
effective with the 2-pyrrolyl ketone like II-IV. This is not
surprising if we consider the fact that the electronegativity
of nitrogen atom is smaller than that of oxygen atom, and
the sizes of carbon and nitrogen atoms are most compati-
ble to form a C-N double bond to achieve IV. Also, the
hydrogen bonding such as for V makes the C=O bond have
single bond character and consequently, the frequency
becomes the smallest.
Table 4
-1
Averaged Carbonyl Stretching Frequencies (in cm ) of the Benzoyl
Compounds 1-5
1
2
3
4
5
All-(11)
H
1653 ± 8
1652
1630 ± 5
1629
1614 ± 10
1614
1644 ± 9 1765 ± 6
1647 1764
1647 ± 6 1768 ± 3
1640 ± 7 1762 ± 2
As mentioned earlier the order of d
values is 1 > 2 >
Meta-(5) 1657 ± 4
Para-(5) 1650 ± 2
1633 ± 2
1628 ± 3
1614 ± 14
1611 ± 13
C=O
3 > 4, but the order of n
values is 1 > 4 > 2 > 3. The
C=O

Sep-Oct 2003
Infrared and Nuclear Magnetic Resonance Properties of Benzoyl Derivatives
767
exceptional behavior of the furyl ketones 4 can be
explained by the interaction between the oxygen atoms in
the ring and the carbonyl group. It is well known that both
oxygen atoms lie close together because the most favor-
able conformation of the carbonyl derivative of 5-mem-
bered heteroaromatic compounds is syn such as for VI
[10]. The lone pair orbitals in both oxygen atoms lie in
close proximity like VIII so that one of the lone pair elec-
trons may move into C=O bond making its force constant
greater. In this way the C=O stretching frequency
becomes larger than those of 2 and 3. It may be a similar
phenomenon to that of a -halogenated cyclic ketones in
which the equatorial isomer absorbs at higher wave num-
and dual substituent parameter (DSP) approaches, which
are represented in Equations 1 and 2, respectively, have
been extensively studied for such practice [12].
d = rs + d
(1)
(2)
o
d = r s + r s + d
I
I
R
R
o
The effects of m- and p-substituents in the benzoyl group
1
13
on the chemical shift of H and C in benzene (1) and
heterocyclic rings (2-4) were analyzed by the Hammett
equation (Eq. 1). The results are listed in Table 5.
1
13
The plots of H and C chemical shifts against various
substituent parameters show best correlation with the
-1
-1
+
13
ber (1745 cm ) than the axial isomer (1725 cm ) [11].
Anti conformation such as for VII should have no through-
space interaction between the ring hetero atom and oxygen
atom in the carbonyl group.
Hammett s values [13]. Other s values such as s or
s
[14] do not show a reasonable correlation. Although DSP
analysis shows good correlation the results are not listed
because the major objective of the present report is to
There are several factors that influence the chemical shift
of a nucleus. Electronic shielding (or deshielding) originat-
ing from the inductive effect of neighboring groups is the
most influential among them. In this regard the effect of a
substituent on the chemical shift of benzene derivative can
be correlated with a substituent parameter such as the
Hammett s values. The single substituent parameter (SSP)
determine the aromaticity indices of five-membered
heterocyclic compounds.
The correlations show several interesting phenomena.
First of all, o-Hs show no correlation while m- and p-Hs
show fair correlations with s values. The slopes of p-Hs
are larger than those of m-Hs by 11.5, 26.2, 19.7 Hz for 1,
2, and 3, respectively. But the opposite is the case with 4.
Table 5
1
13
Best Fit of the Single Substituent Parameter Equation for the H and C Chemical Shifts of 1-4 in Chloroform-d in Hz.
1
2
3
4
r
r
r
r
r
r
r
r
Ortho-H
9.9
0.478
-2.8
0.150
-2.8
0.132
56.0
0.906
Meta-H
Para-H
C=O
Ipso-C
Ortho-C
Meta-C
Para-C
Ipso-C’
25.2
36.7
-207.4
-190.5
17.2
54.2
141.5
757.4
0.965
0.983
0.787
0.992
0.564
0.976
0.993
0.785
26.3
52.5
0.939
0.984
0.739
0.969
0.957
0.990
0.990
0.833
25.9
45.6
0.934
0.968
0.819
0.937
0.951
0.988
0.992
0.789
36.4
31.7
0.931
0.935
0.774
0.933
0.941
0.982
0.981
0.779
-178.7
-138.9
138.2
64.1
183.5
825.6
-241.0
-83.7
145.1
90.8
191.3
761.8
-228.0
-65.6
139.5
78.7
116.2
724.5

768
K. O. Jeon, J. H. Jun, J. S. Yu and C. K. Lee
Vol. 40
The difference in the magnitude is only 4.7 Hz. Again, the
electronic effect of oxygen atom is so significant that the
effect of the substituents do not seem to make much differ-
ence. Even the correlation coefficient (r) for the para
series (0.967-0.984) is better than those of the meta series
(0.934-0.965) in 1-3. Contrastingly, they are virtually the
behavior of the p-OH group may be due to favorable
hydrogen bonding between the carbonyl oxygen atom and
p-OH group [17].
It should be noted that the absolute values of the slopes
are very close (i.e., 270.9 vs. 287.9 Hz). This indicates
that the magnitude of the upfield shift is about the same
regardless of the sign of s s as long as their absolute values
are same. This observation together with the very close
same in 4 (r = 0.931 vs. r = 0.935).
m
p
13
The C=O shows a trend of a negative slope with poor,
n
values for p-substituted 1 would be better explained
but similar, correlation coefficients (r = 0.739-0.819) for 1-
C=O
13
by the p polarization with opposite partial charges such as
IX and X.
4. The inverse correlation of C=O vs s has been
reported for various benzoyl series, Z-C H -C(=O)X (X =
6
4
NH , F, OEt, OH, Me, H) [7], m- and p-substituted phenyl
2
esters R-C(=O)-O-C H -Z [4], m- and p-substituted
6
4
anilides R-C(=O)-NH-C H -Z [5,15], and ethyl m- and p-
6
4
substituted cinnamates, Z-C H -CH=CH-C(=O)-O-C H
6
4
2 5
[16]. The correlations were much better for the series of
esters (r = 0.989) [4]. Therefore, the rather poor correla-
tion seems to warrant a further investigation. In order to
examine the effect of the electron-donating and –with-
drawing groups additional m- and p-substituted benzophe-
13
In contrast to C=O, the i-Cs of 1 show an excellent cor-
nones (Z = p-C H , 1l; p-tert-C H , 1m; p-NH , 1n; m-
2
5
4
9
2
relation with s (r = 0.992) with a negative slope (r = -190.5
Hz). The i-Cs of the heterocycles also show good correla-
tions (r = 0.933-0.969), but the magnitudes of the slopes
decrease in the order of 1 > 2 > 3 > 4. The magnitude of the
slope is the measure of the sensitivity of the chemical shift
to the electronic effect of the substitutent. i-C is the very first
carbon bonded to the substituted benzoyl group. Therefore,
the substituent effect should be most profound if the trans-
mission of the effect takes place through bonds. The r val-
ues listed in Table 5, however, are not consistent with such
an expectation. If only the absolute values are considered,
NH , 1o; p-NMe , 1p; p-OH, 1q; p-OCOCH , 1r) were
prepared.
2
2
3
With total eighteen benzophenones (1), the slope is
–102.1 Hz with no apparent correlation pattern (r = 0.481).
However, there is a clear trend that shows a change in the
signs of the slopes as shown in Figure 3.
With ten substituents that have negative s values includ-
ing H (s = 0), the slope is positive (270.9 Hz) with r =
0.843. On the other hand, a negative slope (-287.9 Hz)
with r = 0.914 is observed with eight electron-withdrawing
substituents (i.e., +s values). The correlation would have
been much better (r = 300.3 Hz, r = 0.931) if the point cor-
responding to p-OH were excluded. The exceptional
r of 1 is the largest among all r values of 1-4. But r s are
i
p
the largest in 2 and 3 while r is the largest in 4. In 4, r is
o
i
the smallest. The inductive effect of the oxygen atom is so
great that the 2- and 5-Cs of the furan ring in 4 are not
affected as much as in 1-3 by the electronic effect of the sub-
stituent. All r values are smallest among each series of 1-
m
3, except 1. The latter case cannot be compared because
there is no apparent correlation in the chemical shift of o-C
with s (r = 0.564). This is the only case which does not
show a meaningful correlation of d and s . It is known that
both rings of 1 do not lie in the same plane [18]. Instead,
they are skewed a little due to the interaction between o- and
o’-Hs. Therefore, the o-C and o-H of 1 is under the influ-
ence of the field effect of the carbonyl group quite differ-
ently, and such effect should be greater than the effect of
substituent transmitted through bonds or through space.
It is notable that r and r of 3 are the largest among r s
o
p
o
and r s of 1-4. As mentioned earlier, conjugation such as
p
II-IV is very feasible in 3. Therefore, the o- and p-Cs of
the pyrrole ring are likely to have a partial positive charge.
The charge density is most likely to be influenced by the
electronic effect of the substituent.
Figure 3. Correlation between s and d
data line is 2.71 ppm (r = 0.843) and the slope of the rectangle data line is
-2.88 ppm (r = 0.914).
in 1. The slope of the circle
C=O

Sep-Oct 2003
Infrared and Nuclear Magnetic Resonance Properties of Benzoyl Derivatives
769
1
13
As in the case with H r of 4 is larger than r in
C
o
p
correlation. The inductive effect of oxygen atom directly
bonded to p-C should have profound effect. The excellent
correlation of d with s may be explained better by a modi-
fied p polarization mechanism for the transmission of the
substituent effect [7,19]. The electronic effect of the sub-
stituent seems to travel as if it is a vector representing a
dipole like XI and XII. Those atoms along the vector
should be more sensitive to the effect. If the p polarization
determines the sign of the r , the vector should determine
the magnitude of the r . The vector seems to pass through
the i- and p-C atoms of 1-4. The significant sensitivity of
r of 2-4, though smaller than r , should be due to the s-
o
p
trans conformation of the carbonyl and C2-C3 double
bond which makes C3 positive like II.
Figure 4.Correlations between Dd
of 1 and Dd
of 2-4.
ipso-C
ipso-C
Pyrrole and furan rings are about 11% and 12% more sen-
sitive than benzene ring. The i-Cs are apparently less sen-
sitive to the effect of the substituents. The values of the
slopes 0.72, 0.44, and 0.35 may be a reflection of the aro-
maticity indices for thiophene, pyrrole, and furan, respec-
tively, if the index of benzene is set to 1.00.
Figure 4 is an illustration of the correlation of d s of 2-4
C
against d s of 1. Table 6 lists the slopes and correlation
C
coefficients of such plots.
Except for o-H and o-C the correlations are excellent.
As expected the slope of near unity with r = 0.999 for i-C’s
indicates that the assignments are accurate for all ipso car-
bons of the substituted benzoyl group. No correlation with
the o-H is due to the fact that there is no correlation of
these shifts with the Hammett s . Because the chemical
shifts of the o-Cs of 1 do not correlate with s the correla-
tion coefficients reflect merely a trend in the case of o-Cs
of 2-4 (r = 0.780-0.827).
As listed in Table 3, a -Hs of the heterocycles appear
downfield and the difference between a - and b-Hs are
0.22, 0.59, and 1.06 ppm for thiophene, pyrrole, and furan,
respectively. If an assumption is made that the difference
of a - and b-Hs of benzene is zero because it is fully aro-
matic and it corresponds to the index of aromaticity of
1.00, then the relative indices of aromaticity may be
related to the difference between d
and d . A set of
a -H
b-H
indices may be proposed for thiophene 0.89 (= 1.00 –
0.22/2), pyrrole 0.71 (= 1.00 – 0.59/2), and furan 0.47 (=
1.00 – 1.06/2). This seems quite reasonable not only
because the difference in chemical shift of a - and b-Hs is
likely to originate from the presence of the heteroatom but
One of the striking features of such plots is that the slope
13
of C=O is 1.00 (r = 0.998) for 2. This is a clear indica-
tion that the thiophene ring as a whole responds most sim-
ilarly as benzene ring to the effect of the substituents.
Table 6
1
13
Slopes and Correlation Coefficients of the Plots of H and C Chemical Shift Values of the 2-Benzoylthiophenes (2),
2-Benzoylpyrroles (3), and 2-Benzoylfurans (4) vs. Those of the Benzophenones (1) in Chloroform-d
2
3
4
r
r
r
r
r
r
Ortho-H
Meta-H
Para-H
C=O
Ipso-C
Ortho-C
Meta-C
Para-C
Ipso-C’
0.46
1.04
1.41
1.00
0.72
3.84
1.16
1.27
0.99
0.517
0.995
0.999
0.998
0.979
0.827
0.991
0.994
0.999
0.49
1.06
1.26
1.11
0.44
3.91
1.65
1.35
0.99
0.457
0.993
0.997
0.998
0.946
0.780
0.996
0.998
0.999
1.22
1.47
0.88
1.12
0.35
3.92
1.44
0.82
0.96
0.410
0.981
0.967
0.999
0.947
0.806
0.994
0.989
0.999

770
K. O. Jeon, J. H. Jun, J. S. Yu and C. K. Lee
Vol. 40
Table 7
Yields, Mp, Elemental Analysis, and n
Data of Compounds 1-4
C=O
Compd
Yield
%
Mp
C
Calcd
X, %
Observed
X, %
n
C=O
o
-1
C, %
H, %
S, %
C, %
H, %
S, %
cm
1a
1b
1c
1d
1e
1f
1g
1h
1i
89
84
92-94
liquid
85-87
liquid
liquid
68.72
59.80
72.07
79.23
85.68
3.99
3.47
4.19
5.70
6.16
3.99
3.47
4.19
5.70
6.16
5.53
6.71
7.61
5.62
5.62
6.71
5.09
5.03
3.03
2.64
3.17
4.62
4.98
3.03
2.64
3.17
4.62
4.98
4.28
3.73
3.22
3.92
5.51
5.99
3.73
3.22
3.92
5.51
5.99
5.30
3.25
2.81
3.41
4.98
5.41
3.25
2.81
3.41
4.98
5.41
4.68
6.16[a]
30.60[b]
16.36[c]
68.55
60.05
72.33
78.99
85.82
68.66
59.96
72.15
79.50
85.85
85.72
85.88
85.89
78.99
79.02
79.78
5.31
75.08
56.84
49.25
59.08
66.32
71.33
56.88
49.56
59.47
66.21
71.44
70.35
61.36
52.65
64.40
71.55
77.95
61.25
52.74
64.08
71.65
77.85
77.38
60.66
52.88
63.69
71.54
77.45
60.98
52.61
64.08
71.35
77.59
76.80
4.03
3.55
4.48
5.57
6.01
4.05
3.43
4.33
5.55
6.41
5.50
6.82
7.58
5.68
6.75
6.52
6.41[a]
30.45[b]
16.12[c]
1655
1661
1653
1659
1656
1652
1649
1651
1650
1649
1652
1649
1649
1588
1648
1599
68
79
136-138 68.72
6.16[a]
30.60[b]
16.36[c]
6.33[a]
30.57[b]
16.31[c]
80-82
75-77
60-62
56-57
48-49
liquid
liquid
121-124 79.17
81-84
88-90
132-135 78.77
69-70 74.99
107-110 56.65
60-62
56-58
liquid
liquid
59.80
72.07
79.23
85.68
85.69
85.68
85.67
1j
1k
1l
15
45
94
88
11
1m
1n
1o
1p
1q
1r
2a
2b
2c
2d
2e
2f
7.10[a]
7.10[a]
6.22[a]
7.35[a]
7.39[a]
6.45[a]
79.17
79.97
78.88
1562
95
51
53
11
39
35
8
77
87
64
66
77
28
29
57
33
44
32
43
32
40
25
36
6
13
15
10
35
2
10
18
10
36
22
5.22
3.20
2.50
3.25
4.66
5.01
3.36
2.61
3.33
4.38
5.12
4.40
3.81
3.44
4.03
5.56
6.12
3.82
3.32
3.98
5.59
6.15
5.45
3.38
2.90
3.44
5.25
5.56
3.44
2.91
3.55
5.00
5.54
5.01
1651
1631
1631
1632
1635
1634
1628
1626
1631
1629
1627
1629
1615
1628
1610
1605
1612
1615
1624
1614
1605
1602
1614
1645
1633
1648
1646
1645
1639
1654
1651
1632
1644
1647
6.01[a]
29.91[b]
15.92[c]
13.75
12.00
14.40
14.69
15.85
13.75
12.00
14.40
14.69
15.85
17.03
5.92[a] 13.80
29.98[b] 11.78
15.85[c] 14.58
14.52
49.46
59.33
66.03
71.26
15.68
173-175 56.65
6.01[a]
29.91[b]
15.92[c]
5.88[a] 13.48
29.84[b] 11.81
15.64[c] 14.25
14.39
2g
2h
2i
97-100
95-97
69-72
69-71
48-50
82-83
76
86
liquid
66
49.46
59.33
66.03
71.26
70.19
61.11
52.83
64.25
71.63
77.81
61.11
2j
15.58
16.86
2k
3a
3b
3c
3d
3e
3f
3g
3h
3i
12.96[a]
5.60[a]
6.81[a]
6.96[a]
7.56[a]
12.96[a]
5.60[a]31.95[b]
6.81[a]17.24[c]
6.96[a]
7.56[a]
8.18[a]
6.45[a]
31.83[b]
12.80[a]
5.85[a]
6.59
6.74
7.48
12.77
5.40
6.95
6.68[a]
7.35[a]
8.01[a]
6.51[a]
31.56[b]
17.23[c]
31.95[b]
17.24[c]
31.77[b]
17.01[c]
162
123-124 52.83
114 64.25
109-110 71.63
115
71
31.80[b]
17.12[c]
3j
77.81
77.17
3k
4a
4b
4c
4d
4e
4f
4g
4h
4i
107-110 60.83
52-54
liquid
liquid
liquid
52.62
63.94
71.28
77.40
17.16[c]
165-171 60.83
6.45[a]
31.83[b]
17.16[c]
6.52[a]
31.59[b]
17.03[c]
63-65
59-61
59-60
liquid
liquid
52.62
63.94
71.28
77.40
76.73
4j
4k
[a] Nitrogen; [b] Bromine; [c] Chlorine.
because the values are in the range of reported sets of aro-
maticity indices [1].
The o-H signals shift to downfield by the electronic
effect of the carbonyl group. The magnitudes of the
averaged shifts are 0.43, 0.52, 0.63, and 0.87 ppm for
1, 2, 3, and 4, respectively. The difference may be
related to the aromaticity and the set of values 1.00,
0.83, 0.68, and 0.49 can be calculated by dividing the
difference of 1 with the differences of 2, 3, and 4,
respectively.

Sep-Oct 2003
Infrared and Nuclear Magnetic Resonance Properties of Benzoyl Derivatives
771
pale yellow liquid which was purified by chromatography on a
column of silica gel eluting with hexane-ethyl acetate. Solid
recrystallization was repeated until analytical purity was
achieved. Yield, mp, elemental analysis data, and carbonyl
stretching frequencies are listed in Table 7.
1
13
Unlike H, the C chemical shifts are difficult to corre-
late with an index of aromaticity. Introduction of the ben-
zoyl group causes downfield shift of the i-C signal. Such
a shift is the most significant with the thiophene series
showing an average shift of 18.08 ppm. On the other
hand, benzene shows the least effect by the presence of
the benzoyl group and Dd 8.92 ppm is observed. The
inductive effect of oxygen atom in furan causes the most
deshielding of a -C and the difference between a - and b-
Cs is 33.08 ppm for furan itself. But introduction of a
benzoyl group at a -position causes a similar magnitude of
deshielding of both the a - and b-Cs and the difference
between i- and o-Cs is 31.48 ppm for 4.
Acknowledgments.
We thank Professor Ned Martin of the University of North
Carolina at Wilmington and Dr. Gary Kwong of the 3M Co. for
help in preparing the manuscript. This research was supported by
Korea Research Foundation (KRF-2001-005-D20012).
REFERENCES AND NOTES
[1a] M. J. Cook and A. R. Katritzky, Adv. Heterocyclic Chem.,
17, 255 (1974); [b] B. Y. Simkin, V. I. Minkin and M. N. Glukhovtsev,
Adv. Heterocyclic Chem., 56, 303 (1993); [c] A. R. Katritzky, M.
Karelson and N. Malhotra, Heterocycles, 32, 127 (1991); [d] F. R.
Cordell and J. E. Boggs, J. Mol. Struct., 85, 163 (1981); [e] C. W.
Bird, Tetrahedron, 41, 1409 (1985).
The difference between the shifts of i- and o-Cs of 1-4
may be used as a base for calculation of the aromaticity
index. If we divide Dd of 1 by those of 2, 3, and 4, 0.87,
0.65, and 0.23 are obtained, respectively, which may be
considered as a set of aromaticity indices. The exception-
ally low value of 0.23 for 4 may be due to the profound
electronic effect of the oxygen atom in the furan ring.
In conclusion, a set of aromaticity indices is proposed
based on the correlation of the chemical shifts of a series
of substituted benzophenones to an identical series of sub-
stituted benzoyl derivatives of thiophene, pyrrole, and
furan. The indices thus obtained are 0.72, 0.44, and 0.35,
respectively.
[2] Elvidge, J. A. Chem. Commun., 160 (1965).
[3] A. Julg and P. Francois, Theor. Chim. Acta, 8, 249 (1967).
[4a] C. K. Lee, J. S. Yu and H.-J. Lee, J. Hererocyclic Chem.,
39, 1207 (2002); [b] C. K. Lee, J. S. Yu and J. S. Park, Bull. Korean
Chem. Soc., 21, 49 (2000).
[5] C. K. Lee, J. S. Yu and Y. R. Ji, J. Hererocyclic Chem., 39,
1219 (2002).
[6] F. Fringuelli, G. Marino, A. Taticchi and G. Grandolini, J.
Chem. Soc. Perkin Trans. II, 332 (1974).
[7] J. Bromilow, R. T. C. Brownlee, D. J. Craik, P. R. Fiske, J.
E. Rowe and M. Sadek, J. Chem. Soc. Perkin Trans. II, 753 (1981).
[8] H. Neuvonen, K. Neuvonen, A. Koch, E. Kleinpeter and P.
Pasanen, J. Org. Chem., 67, 6995 (2002).
EXPERIMENTAL
[9] N. B. Colthup, L. H. Daly and S. E. Wiberley, Introduction
to Infrared and Raman Spectroscopy, Academic Press, New York,
1964, p 241.
[10] M. Fiorenza, A. Ricci, G. Sbrena, G. Pirazzini, C. Eaborn
and J. G. Stamper, J. Chem. Soc. Perkin Trans. II, 1232 (1978).
[11] Reference 9, p 243.
[12a] D. J. Craik and R. T. C. Brownlee,Prog. Phys. Org. Chem.,
14, 1 (1983); [b] S. Ehrenson, R. T. C. Brownlee and R. W. Taft, Prog.
Phys. Org. Chem., 10, 1 (1973); [c] C. K. Lee, J. H. Jun and J. S. Yu, J.
Hererocyclic Chem., 37, 15 (2000); [d] K. O. Jeon, J. S. Yu and C. K.
Lee, Bull. Korean Chem. Soc., 23, 1241 (2002).
[13] F. Ruff and I. G. Csizmadia, Organic Reactions: Equilibria,
Kinetics and Mechanism, Elsevier, Amsterdam, 1994, p 164.
[14] C. D. Slater, C. N. Robinson, R. Bies, D. W. Bryan, K.
Chang, A. W. Hill, W. H. Moore Jr., T. G. Oray, M. L. Poppelreiter, J.
R. Reisser, G. G. Stablein, V. P.; Wilkinson, W. D. Waddy III and W.
A. Wray, J. Org. Chem., 50, 4125 (1985).
Melting points were determined on a Fischer MEL-TEMP
apparatus and are uncorrected. Nuclear magnetic resonance
(nmr) spectra were recorded on a Bruker DPX-400 FT NMR
spectrometer in the Central Lab of Kangwon National University
1
at 400 MHz for H and 100 MHz for 13C and were referenced to
tetramethylsilane. The concentration of the solution was 0.10 M
in chloroform-d. Infrared spectra were recorded on a JASCO
FT/IR-460 Plus spectrophotometer. Elemental analyses were per-
formed by the Central Lab of Kangwon National University.
Meta- and p-substituted benzoic acids are all commercially
available from which the corresponding benzoyl chlorides were
prepared by a standard procedure using thionyl chloride.
Benzoyl chloride, 2-furoyl chloride, and 2-thiophenecarbonyl
chloride are also commercially available. Compounds 1c, 1e, 1g,
1h, 1i, 1j, and lk are also commercial products.
[15a] H. Suezawa, T. Yuzuri, M. Hirota, Y. Ito and Y. Hamada, Bull.
Chem. Soc. Jpn., 63, 328 (1990); [b] T. Yuzuri, H. Suezawa and M.
Hirota, Bull. Chem. Soc. Jpn., 67, 1664 (1994).
Preparation of the Ketones.
An Illustrative Procedure.
[16] C. N. Robinson, G. E. Stablein and C. D. Slater,
Tetrahedron, 46, 335 (1990).
Benzoyl chloride (7 mmoles) was added dropwise over 10
minutes to a mixture of benzene (15 ml) and aluminum chloride
(1.44 g, 10 mmoles) with rapid stirring. The mixture was heated
at reflux for 36 hours. After neutralization with saturated sodium
bicarbonate solution the aqueous mixture was extracted with
dichloromethane (6 x 25 ml). The organic layer was washed with
water (25 ml) and then dried with magnesium sulfate overnight.
After suction filtration the solution was concentrated to give a
[17] E. Breitmaier and W. Voelter, Carbon-13 NMR
rd
Spectroscopy, 3 Ed., VCH: Weinheim, Germany, 1987; p 117.
[18] O. Exner, in The Chemistry of Double-bonded Functional
Groups, Supplement A, Part 1; Patai, S. Ed., Wiley, New York, 1977, p
81.
[19] R. T. C. Brownlee and D. J. Craik, J. Chem. Soc. Perkin
Trans. II, 760 (1981).