1-Benzoylazoles: An experimental (NMR and crystallography) and theoretical study

Article abstract of DOI:10.1016/S0022-2860(01)00759-1

Five N-benzoylazoles (imidazole, pyrazole, indole, benzimidazole and carbazole) have been prepared following modified literature procedures. Their NMR spectra in solution (¹H, ¹³C and ¹⁵N) have been measured. The crystal structures of 1-benzoylindole and 9-benzoylcarbazole have been determined by X-ray crystallography and the corresponding ¹³C NMR spectra in the solid state have been measured by the CPMAS technique. Whereas 1-benzoylindole presents a standard behaviour, 9-benzoylcarbazole shows an unexpected ¹³C CPMAS spectrum with additional splittings. In order to understand this fact, the ¹H and ¹³C NMR spectra in dimethylether at -143°C (130 K) have been recorded and ab initio calculations (RHF/6-311G**) carried out. The corresponding absolute shieldings (GIAO/RHF/6-311G**) together with the X-ray structure and the ¹³C chemical shifts at low temperature have been used to discuss the CPMAS spectrum. We propose that the supplementary splittings of this spectrum are due to its conglomerate structure.

Full text of DOI:10.1016/S0022-2860(01)00759-1

Journal of Molecular Structure 605 ꢀ2002) 199±212
www.elsevier.com/locate/molstruc
1-Benzoylazoles: an experimental ꢀNMR and crystallography)
and theoretical study
a
R.M. Claramunt , P. Cornago , D. Sanz , M.D. Santa-Marõa , C. Foces-Foces ,
a
a
a
b
Â
I. Alkortac, J. Elgueroc,
*
a
  Â
Departamento de Quõmica Organica y Biologõa, Facultad de Ciencias, UNED, Senda del Rey 9, 28040 Madrid, Spain
b
  Â
Departamento de Cristalografõa, Instituto de Quõmica-Fõsica Rocasolano, CSIC, Serrano 119, 28006 Madrid, Spain
Â
c
  Â
Instituto de Quõmica Medica, Centro de Quõmica Organica `Manuel Lora Tamayo', CSIC, Juan de la Cierva 3, 28006 Madrid, Spain
Received 30 April 2001; revised 12 June 2001; accepted 12 June 2001
Abstract
Five N-benzoylazoles ꢀimidazole, pyrazole, indole, benzimidazole and carbazole) have been prepared following modi®ed
literature procedures. Their NMR spectra in solution ꢀ1H, 13C and 15N) have been measured. The crystal structures of 1-
benzoylindole and 9-benzoylcarbazole have been determined by X-ray crystallography and the corresponding 13C NMR spectra
in the solid state have been measured by the CPMAS technique. Whereas 1-benzoylindole presents a standard behaviour, 9-
benzoylcarbazole shows an unexpected 13C CPMAS spectrum with additional splittings. In order to understand this fact, the 1H
and 13C NMR spectra in dimethylether at 21438C ꢀ130 K) have been recorded and ab initio calculations ꢀRHF/6-311G**)
carried out. The corresponding absolute shieldings ꢀGIAO/ RHF/6-311G**) together with the X-ray structure and the 13C
chemical shifts at low temperature have been used to discuss the CPMAS spectrum. We propose that the supplementary
splittings of this spectrum are due to its conglomerate structure. q 2002 Elsevier Science B.V. All rights reserved.
Keywords: Azolides; NMR; CPMAS NMR; Rotational barriers; X-ray crystallography; ab initio calculations
1. Introduction
obtained for ®ve N-benzoylazoles: namely, 1-benzoy-
limidazole 1, 1-benzoylpyrazole 2, 1-benzoylindole 3,
1-benzoylbenzimidazole 4 and 9-benzoylcarbazole 5.
We have been studying the physico±chemical prop-
erties of azolides for a long time. These studies concern
mainly their NMR properties, generally in solution [1±
5], but also in the solid state [6], and their behaviour in
mass spectrometry [7,8]. Recently, we have written a
review [9] and a subsequent paper on ab initio calcula-
tions of their NZCHO, NZCOCH3, NZCOCF3 and
NZCO2CH3 derivatives [10].
The present work reports the results we have
* Corresponding author. Tel.: 134-91411-0874; fax: 134-91564-
4853.
E-mail address: iqmbe17@iqm.csic.es ꢀJ. Elguero).
0022-2860/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved.
PII: S0022-2860ꢀ01)00759-1

200
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
Table 1
Selected geometrical parameters ꢀA,8). CentꢀA, B, C or D) stands for the centroid of the corresponding A, B, C or D rings as shown in Fig. 1
Ê
Compound 3
Compound 5
N1ZC10
C10ZO17
1.394ꢀ2)
1.211ꢀ2)
N1ZC14
C14ZO15
1.400ꢀ7)
1.212ꢀ7)
N1ZC10ZO17
C9ZN1ZC10
C2ZN1ZC14
C2ZN1ZC9
C9ZN1ZC10ZO17
N1ZC10ZC11ZC12
A ^ B
120.8ꢀ1)
N1ZC14ZO15
C2ZN1ZC14
C13ZN1ZC14
C2ZN1ZC13
C2ZN1ZC14ZO15
N1ZC14ZC16ZC17
A ^ B
119.3ꢀ5)
125.9ꢀ1)
126.0ꢀ1)
107.1ꢀ1)
26.1ꢀ2)
245.0ꢀ2)
1.3ꢀ1)
124.6ꢀ5)
127.6ꢀ4)
107.7ꢀ4)
223.5ꢀ9)
245.7ꢀ8)
1.3ꢀ2)
B ^ C
5.2ꢀ2)
Hydrogen interactions:
DZH
H´´´A
D´´´A
DZH´´´A
Compound: 3
C8ZH8´´´O17
0.97ꢀ2)
1.02ꢀ2)
0.98ꢀ3)
1.01ꢀ3)
2.42ꢀ2)
2.72ꢀ2)
2.79ꢀ2)
2.85ꢀ3)
2.938ꢀ2)
113ꢀ2)
135ꢀ2)
140ꢀ2)
138ꢀ2)
C16ZH16´´´CentꢀB)ꢀ1 2 x, 2 y,1 2 z)
C13ZH13´´´CentꢀA)ꢀ1 2 x,1 2 y,2 2 z)
C3ZH3´´´CentꢀD)ꢀ1 2 x, 2 y,2 2 z)
Compound: 5
3.520ꢀ2)
3.600ꢀ2)
3.665ꢀ2)
C3ZH3´´´O15
0.96ꢀ7)
0.99ꢀ6)
1.01ꢀ7)
2.26ꢀ7)
2.66ꢀ6)
2.87ꢀ7)
2.890ꢀ9)
3.393ꢀ7)
3.867ꢀ8)
122ꢀ5)
131ꢀ5)
167ꢀ6)
C4ZH4´´´O15ꢀ1/2 2 x,1 2 y,1/2 1 z)
C19ZH19´´´CentꢀC)ꢀ1/2 2 x, 2 y, 2 1/2 1 z)
being liquid at room temperature: 1, m.p. 20±218C
[11], the other solids, in which cases the melting
points coincide with those of the literature: 2, m.p.
468C [12], 3, m.p. 698C [13], 4, m.p. 91±928C
[14,15], 5, m.p. 988C [16].
2.2. X-ray crystallography
The main geometrical intra and intermolecular
parameters for compounds 3 and 5 are reported in
Table 1. The molecular structure of 3 presents similar
features regarding the `amide' fragment to those
described for other 1-COR-indole derivatives [9,17],
that is, NZC, CyO distances, NZCyO angle and, as
in almost all structures, the CyO group pointing
towards the benzene ring [Fig. 1ꢀa)]. Greater discre-
pancies ꢀCyO distance and NZCyO angle) between 5
[Fig. 1ꢀc)] and the only 9-COR-carbazole derivative
previously reported ꢀ14-acetyl-di-indolo[2,3-a:20,30-
c]carbazole; CSD refcode: INDCBZ10) [18], are
observed. They are probably due to the tricyclic
fused ring system and to the low precision achieved
in INDCBZ10 ꢀR ˆ 10.4%). In 3 and 5, the indole
and the carbazole moieties ꢀN-substituted with benzoyl
groups) show small but signi®cant distortions from
planarity. The angles between the least-squares planes
2. Results and discussion
2.1. Chemistry
The ®ve compounds are known compounds, one

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
201
Fig. 1. Two perpendicular views of the molecular structure of compound 3 [ꢀa), ꢀb)] and compound 5 [ꢀc), ꢀd)] showing the numbering scheme
and the conformation of the molecules. Thermal ellipsoids are drawn at 30% probability level.
of the six-membered ꢀA or C carbocycles) and the
®ve-membered ꢀB heterocycle) rings, seem to be
in¯uenced by the intramolecular CZH´´´OyC
contact as re¯ected by the CZNZCyO torsion
angle, in both compounds, and by the narrowing
of the corresponding external angle at N1 mainly
in 5 ꢀC2ZN1ZC14 , C13ZN1ZC14, Table 1)
where the angle between B and C is signi®cantly
greater than the A, B one [5.2ꢀ2) vs 1.3ꢀ2)8]. In
addition, the O and C atoms of the carbonyl group
deviate from the pyrrole plane by 0.535ꢀ2),
The crystal packing of 3 can be described as strands
in a head-to-tail arrangement along the b axis
stabilised by CZH´´´p contacts [Fig. 2ꢀa)] without
other contacts between chains than those of the van
der Waals ones. In 5, CZH´´´O interactions link mole-
cules into chains along the c axis that are then
connected by CZH´´´p contacts [Fig. 2ꢀb)]. The rele-
vance of these hydrogen interactions in the crystal
structure of organic molecules as well as in biological
molecules is well known as stated by Desiraju and
Steiner [19].
Ê
0.213ꢀ2) in 3 and 0.222ꢀ5), 20.127ꢀ6) A in 5,
respectively [see Fig.1ꢀb) and ꢀd)]. The phenyl
ring of the substituent has similar dihedral angles
in both compounds ꢀTable 1).
2.3. Solution NMR results
We have collected in Tables 2 ꢀ1H), 3 ꢀ13C) and 4

202
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
ꢀ15N) the results obtained for compounds 1±5.
be determined at 175 K ꢀno broadening observed
neither in proton, Table 2, nor in carbon NMR,
Table 3).
The barrier of 5 ꢀ29.7 kJ mol21) has been determined
by Lunazzi et al. by 13C NMR spectroscopy at
25.16 MHz, but the chemical shifts at 136 K were not
reported [20]. Recently, at our request, Professor
1
Lunazzi recorded again the H ꢀ300.08 MHz) and 13C
ꢀ75.45 MHz) NMR spectra of 5 in dimethyl ether to
which a little acetone-d6 was added for providing a
deuterium lock. The experiments were run at 2888C
ꢀ185 K, where the rotation is still fast) and at 21438C
1
ꢀ130 K) where the rotation process is slow. The H
NMR spectrum at 21438C is shown in Fig. 3.
Before discussing these data, we will summarise
the previous knowledge on the dynamic aspects of
compounds 2, 3 and 4. This information, gathered
in a recent review [9], come from low temperature
NMR experiments. At low temperatures, nothing
is observed for pyrazole 2, indole 3 and benzimi-
dazole 4 derivatives, not necessarily because the
barrier should be very low, but because only one
conformer is present in solution. These confor-
mers, 2a, 3a ꢀsee X-ray structure) and 4a are
those predicted from all the previous information
gathered on azolides [9].
Therefore, the only compounds where some
splitting or, at least broadening, could be expected
are 1 ꢀsimilar populations) and 5 ꢀidentical popu-
lations of two degenerate rotamers). From data on
the barriers of N-formyl and N-acetylimidazole
[9], as well as those on formamides, acetamides
and benzamides [22], the estimated barrier for 1
should lie between 30 and 33 kJ mol21, too low to
The 13C spectrum at 130 K shows a splitting of only
0.4 ppm for carbons C1 and C8 while for 9-acetylcarba-
zole ꢀ6), the observed splitting is much larger ꢀ3.0 ppm)
[20]. It is interesting to compare the four compounds
of Scheme 1 ꢀNMR data of 6 and 7 from Ref. [9]).
All the chemical shifts are quite consistent. The
fact that on going from 7 to 3, Hꢀ7) moves from
8.37 to 8.59 while from 6 to 5 the shift is opposite
in sign ꢀ8.60±8.52 ppm) may be related to the
twisted structure of 5 and the probable non exis-
tence of the weak hydrogen bond between Hꢀ1)
and the CyO, present in 3 and probably in 6
and 7. The main difference between the acetyl
and the benzoyl series concerns the region oppo-
site to the oxygen where the effect of the phenyl

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
203
Table 2
1H Chemical shifts ꢀppm from TMS) and some H±1H coupling constants of N-benzoylazoles 1±5 at different temperatures
1
Comp.
Solv. ꢀT, K)
Hꢀ2)
Hꢀ3)
Hꢀ4)
Hꢀ5)
Hꢀ6)
Hꢀ7)
Hꢀ20)
Hꢀ30)
Hꢀ40)
1
1
1
1
1
2
2
CDCl3 ꢀ303)
C6D6 ꢀ303)
8.05
7.80
8.18
8.05
8.34
±
±
7.13
7.01
7.16
7.07
7.24
6.52
6.56
J45 ˆ 2.8
6.76
7.62
7.60
7.73
7.84
7.77
7.89
Hꢀ3)
Hꢀ6)
7.36
7.31
7.36
7.482
7.319
7.400
0.08
7.51
7.22
7.66
7.58
7.91
8.43
8.50
J35 ˆ 0.7
8.25
7.33
7.27
7.39
7.42
7.38
7.50
Hꢀ4)
Hꢀ5)
8.01
8.06
8.13
8.236
8.236
8.236
0.00
±
±
±
±
±
±
±
±
±
±
±
±
±
±
7.76
7.28
7.82
7.83
7.96
8.12
8.20
7.53
6.92
7.60
7.57
7.69
7.50
7.48
7.65
7.06
7.73
7.68
7.80
7.60
7.60
±
DMSO-d6 ꢀ303)
THF-d8 ꢀ303)
THF-d8 ꢀ175)
CDCl3 ꢀ300)
THF-d8 ꢀ300)
±
±
±
7.80
7.79
J34 ˆ 1.5
7.99
6.62
6.64
6.83
±
±
2
3
3
3
4
4
4
THF-d8 ꢀ175)
CDCl3 ꢀ303)
THF-d8 ꢀ303)
THF-d8 ꢀ173)
CDCl3 ꢀ303)
THF-d8 ꢀ303)
THF-d8 ꢀ170)
±
±
±
8.13
7.75
7.75
7.88
7.80
7.88
8.02
Hꢀ20)
Hꢀ60)
7.73
7.70
7.72
7.767
7.59
7.53
7.53
7.66
7.59
7.60
7.72
Hꢀ30)
Hꢀ50)
7.53
7.53
7.55
7.599
7.71
7.61
7.61
7.74
7.69
7.69
7.81
Hꢀ40)
7.31
7.37
7.62
8.22
8.31
8.69
Hꢀ1)
Hꢀ8)
7.52
7.48
7.48 ꢀvb)
8.524
6.464
7.494
2.06
7.40
7.34
7.46
7.46
7.41
7.54
±
8.42
8.43
8.59
8.19
8.21
8.34
±
±
±
Hꢀ2)
Hꢀ7)
7.32
7.27
7.30
7.60b
7.156
7.378
0.44
5
5
5
5
CDCl3 ꢀ303)
±
±
±
±
7.65
7.65
7.69
7.740
THF-d8 ꢀ303)
CH3OCH3 ꢀ185)a
CH3OCH3 ꢀ130)a
Center
Splittings
a
Professor Lunazzi measurements ꢀvb: very broad).
Under the 7.599 signal.
b
group on the adjacent protons is deshielding in 3
and shielding in 5.
The assignments of the signals ꢀ1H, Table 2 and
13C, Table 3) of compound 5 at 303 K in CDCl3 has
been done using the following techniques: ꢀi) directly
linked protons and carbons are detected by direct C,H
correlation ꢀHMQC); ꢀii) in CDCl3 long distance
correlation shows that Cꢀ1) ꢀ115.8 ppm) and Hꢀ20,60)
Scheme 1.

204
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
2.4. Solid state NMR 913C CPMAS)
The 13C CPMAS NMR spectrum of 1-benzoy-
lindole ꢀ3) is virtually identical to that obtained in
solution [see Table 3 and Fig. 4ꢀa)] proving that the
structure should be very similar in both phases. On the
other hand, the spectrum of 9-benzoylcarbazole ꢀ5) is
much more complex in the solid state than in solution
[see Fig. 4ꢀb)]. Part of the complexity arises from the
absence of0 free rotation about the Nꢀ9)ZCyO and
OyCZCꢀ1 ) bonds which splits all the signals of the
carbazole ring and most of those of the phenyl ring,
but the complexity is such that it cannot be due only to
this fact.
We have assumed that in the CPMAS spectrum
[Fig. 4ꢀb) and Table 3] there is a double splitting:
the ®rst is similar to the one observed in solution at
130 K for the carbazole signals Cꢀ1) to Cꢀ9a); the
second one is another expected splitting affecting
carbon atoms Cꢀ20)/Cꢀ60) and Cꢀ30)/Cꢀ50) because
the barrier about OyCZCꢀ10) is too low to be slowed
in solution while in the solid state these carbons are
structurally different. Due to a lack of resolution, in a
previous publication, reporting results at 50 MHz, we
were unable to observe splitted signals in the case of
9-acetylcarbazole ꢀ6) [6], although they have been
observed in solution at 25 MHz [20].
However, the most extraordinary feature of the 13C
CPMAS spectrum of compound 5 is the presence of a
second splitting [see Fig. 4ꢀb): CPMAS spectrum]
which affects the CyO group and several other carbon
atoms. The only possible explanation is the presence
in the crystal of two different molecules of 5. This is
apparently in contradiction with the crystallographic
result that there is only one independent molecule.
Note, however, that compound 5 exists in the studied
crystal as a single enantiomer. The other enantiomer
[by rotatio0n about the two bonds of the Nꢀ9)
ZCOZCꢀ1 ) moiety] is certainly present in the crys-
tallisation crop. This spontaneous resolution indicates
that the crude product is a conglomerate and not a
racemate [21]. Therefore, we make the following
suggestion: the second splitting is due to the con-
glomerate structure. We propose that before the two
enantiomers were separated by crystallisation, the
conglomerate contained them in different conforma-
tions, which are responsible for the second splitting.
The values, in ppm, of the second one are also
Fig. 2. Crystal packing along the c axis: ꢀa) compound 3 and
ꢀb) compound 5. Dotted and dashed lines represent CZH´´´O
hydrogen interactions and CZH´´´p contacts, respectively.
ꢀ7.73 ppm) are close and a NOE2D experiment proves
the proximity of Hꢀ1) ꢀ7.52 ppm) and Hꢀ20,60)
ꢀ7.73 ppm). These experiments are still clearer in
THF-d8 where the proton spectrum is better resolved.
1
Since the data on 15N chemical shifts and H±15N
coupling constants of azolides are scarce, we have
collected in Table 4 those of compounds 1±5.

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
205
Table 3
13C Chemical shifts ꢀppm from TMS) and some H±13C coupling constants of N-benzoylazoles 1±4 at different temperatures
1
Comp. Solv. ꢀT, K)
Cꢀ2)
Cꢀ3)
±
Cꢀ4)
Cꢀ5)
Cꢀ6)
±
Cꢀ7)
±
Cꢀ3a)
±
Cꢀ7a)
±
CO
C6H5
1
1
1
1
1
2
2
3
3
3
3
4
4
CDCl3 ꢀ303)
C6D6 ꢀ303)
138.1
138.3
138.4
139.0
130.8
131.3
130.4
131.5
117.9
117.9
118.4
118.7
166.8 131.8 Cꢀ10)
129.6 Cꢀ20,60)
128.8 Cꢀ30,50)
133.5 Cꢀ40)
165.9 132.4 Cꢀ10)
129.7 Cꢀ20,60)
128.7 Cꢀ30,50)
132.9 Cꢀ40)
166.1 131.7 Cꢀ10)
129.8 Cꢀ20,60)
128.9 Cꢀ30,50)
133.5 Cꢀ40)
166.7 133.5 Cꢀ10)
130.7 Cꢀ20,60)
129.6 Cꢀ30,50)
134.0 Cꢀ40)
167.0 133.1 Cꢀ10)
131.3 Cꢀ20,60)
129.8 Cꢀ30,50)
134.4 Cꢀ40)
166.4 132.7 Cꢀ10)
132.7 Cꢀ20,60)
128.6 Cꢀ30,50)
133.6 Cꢀ40)
166.7 132.3 Cꢀ10)
132.8 Cꢀ20,60)
128.8 Cꢀ30,50)
134.0 Cꢀ40)
168.5 134.5 Cꢀ10)
129.0 Cꢀ20,60)
128.4 Cꢀ30,50)
131.7 Cꢀ40)
168.9 135.9 Cꢀ10)
130.0 Cꢀ20,60)
129.3 Cꢀ30,50)
132.5 Cꢀ40)
169.2 135.3 Cꢀ10)
130.6 Cꢀ20,60)
129.5 Cꢀ30,50)
133.0 Cꢀ40)
168.6 134.8 Cꢀ10)
128.8 Cꢀ20,60)
128.8 Cꢀ30,50)
132.0 Cꢀ40)
167.1 132.9 Cꢀ10)
129.5 Cꢀ20,60)
129.0 Cꢀ30,50)
133.2 Cꢀ40)
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
DMSO-d6 ꢀ303)
THF-d8 ꢀ303)
THF-d8 ꢀ175)
THF-d8 ꢀ300)
THF-d8 ꢀ175)
CDCl3 ꢀ303)
THF-d8 ꢀ303)
THF-d8 ꢀ173)
CPMAS
±
±
±
±
1J ˆ 215.2
1J ˆ 190.5 1J ˆ 195.7
140.0
131.4
109.9
110.8
120.8
118.8
131.0
131.4
123.8
±
±
±
144.9
145.7
108.7
±
±
±
±
±
127.5
124.8
116.7 130.3
135.9
137.1
136.7
135.7
132.1
133.5
1J ˆ 189.7 1J ˆ 175.1 1J ˆ 160.1 1J ˆ 159.4 1J ˆ 160.1 1J ˆ 168.0
128.6
129.5
126.4
143.0
108.7
108.7
107.4
121.5
121.8
119.5
120.6
121.2
124.4
124.7
122.7
125.3
125.5
125.6
125.5
122.7
125.8
125.9
117.1 131.8
117.0 131.7
116.7 130.4
115.4 144.0
116.1 145.7
CDCl3 ꢀ303)
THF-d8 ꢀ303)
±
±
144.4
1J ˆ 215.0
167.1 134.5 Cꢀ10)
130.5 Cꢀ20,60)
129.6 Cꢀ30,50)
133.5 Cꢀ40)
1J ˆ 162.9 1J ˆ 159.9 1J ˆ 160.5 1J ˆ 167.4

206
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
Table 3 ꢀcontinued)
Comp. Solv. ꢀT, K)
Cꢀ2)
Cꢀ3)
±
Cꢀ4)
Cꢀ5)
Cꢀ6)
Cꢀ7)
Cꢀ3a)
Cꢀ7a)
133.1
CO
C6H5
4
THF-d8 ꢀ170)
145.8
121.0
125.8
126.1
116.2 145.2
168.2 134.1 Cꢀ10)
131.2 Cꢀ20,60)
129.8 Cꢀ30,50)
133.9 Cꢀ40)
Carbazole
Cꢀ4a)
Cꢀ4b)
Cꢀ1)
Cꢀ8)
Cꢀ2)
Cꢀ7)
Cꢀ3)
Cꢀ6)
Cꢀ4)
Cꢀ5)
Cꢀ8a)
Cꢀ9a)
CyO
Cꢀ10)
Cꢀ20,60) Cꢀ30,50)
Cꢀ40)
5
5
5
CDCl3 ꢀ303)a
MeOMe ꢀ185)b
MeOMe ꢀ130)b
Splittings
115.8
126.7
123.4
119.8
126.0
139.2 169.6
139.2 129.0
135.8
128.9
140.12 170.03 137.06
115.8
126.7
123.4
119.8
126.0
132.3
116.69
116.69
116.56
116.94
127.64
127.64
127.06
128.46
124.29
124.29
123.99
124.84
120.99
120.99
121.39
121.08
126.82
126.82
126.72
126.66
140.12 130.19 129.93 133.44
140.36 170.11 136.93
139.45 130.31 130.00 133.58
Cꢀ1,8)
Cꢀ2,7)
Cꢀ3,6)
Cꢀ4,5)
Cꢀ4a,4b) Cꢀ8a,9a)
0.38
1.40
125.6
0.85
124.0
0.31
120.4
0.06
124.8
0.91
CPMASa
114.8
138.8 167.3
135.8
113.8
118.7
122.1
119.4
118.2
120.7
1.0
138.3 166.2
139.3 168.4
137.9 130.7
137.5 129.7
138.3
135.4
136.2
128.8
128.0
115.7
115.7
114.8
116.6
126.9
1.3
124.8
0.8
124.8
0.0
131.8
131.2
132.4
Splittings
0.9
0.9
1.9
1.8
3.4
2.5
1.0
0.8
2.2
0.8
0.8
1.2
a
This work.
b
Professor Lunazzi measurements.
Ê
reported in Table 3 ꢀprobably there are other small
splittings). According to these values, the largest
perturbation due to the conglomerate arises in the
proximity of the Cꢀ4)/Cꢀ5) part of the molecule.
0.005 A and the second one a shortening of
Ê
0.002 A; ꢀii) about the NZCyO and NZCZC
Ê
angles, the ®rst one increases ꢀ0.27 A) and the second
Ê
one decreases ꢀ1.16 A). These features are probably
related to differences in the resonance forms ꢀsee
formulae 5 and 6).
The most signi®cant difference between both
azolides concerns the dihedral angle u[Cꢀ9a)ZNꢀ9)-
ZCyO]. While 6 is planar ꢀu ˆ 0.008), 5 is twisted
ꢀu ˆ 237.28). This fact has probably more in¯uence,
on the lowering of the barrier to the rotation, than the
lengthening of the NZC bond. In compound 6, Cꢀ1) is
rather close to the oxygen [a Cꢀ19)ZH´´´OyC weak
hydrogen bond is reasonable] and Cꢀ8) to the methyl
group; to this situation corresponds a difference of
3. Theoretical calculations
3.1. Geometries
The comparison between the geometries of 9-acetyl
[10] and 9-benzoylcarbazole 5 ꢀthis work)
6
concerning the ªamideº fragment show the following
features ꢀfrom 6 to 5): ꢀi) concerning the NZC and
CyO, distances, the ®rst one shows a lengthening of

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
207
Table 4
Chemical shifts ꢀppm from nitromethane for 15N) and some 1H±15
3.0 ppm. In compound ꢀ5) the non-planar conforma-
tion lowers this difference to 0.4 ppm.
Finally, there are some differences between
the experimental and calculated geometries for
N
coupling constants ꢀHz) ꢀthe coupling constants have been
measured on the 1H±15N 2D spectra) of N-benzoylazoles at room
temperature ꢀ,300 K)
Ê
compound 5: NZC 1.400/1.394 A, CyO 1.212/
Comp. Solvent
Nꢀ1)
Nꢀ2)
±
Nꢀ3)
Ê
1.187 A, NZCyO 119.3/121.08 and
u
223.5/
237.28, but they are small enough to consider that
the calculated molecule is a good approximation to
the solid state one.
1
THF-d8
2 176.3
2J ˆ 7, Hꢀ2)
2 110.3
2J ˆ 11, Hꢀ2)
2J ˆ 7, Hꢀ4)
2 109.6
2J ˆ 12, Hꢀ2)
2J ˆ 7, Hꢀ4)
±
1
DMSO-d6 2 174.6
±
2J ˆ 7, Hꢀ2)
3.2. Absolute shieldings
2
3
THF-d8
CDCl3
2 143.0
2 71.2
For compound 5 we have used the same basis
set ꢀRHF/6-311G**) as in our previous work on
N-acetylazoles for the sake of consistency [10]. The
absolute shieldings, in ppm, were calculated within
the GIAO approximation ꢀsee experimental part)
and are: Cꢀ1) 76.20, Cꢀ2) 63.21, Cꢀ3) 69.06, Cꢀ4)
2 204.4
±
±
2J ˆ 8, Hꢀ2)
3J ˆ 7, Hꢀ3)
2 192.9
4
5
THF-d8
CDCl3
±
±
2 123.3
2J ˆ 12, Hꢀ2)
±
2J ˆ 9, Hꢀ2)
2 218.6 [Nꢀ9)]
1
Fig. 3. Experimental H NMR spectrum of 5 at 21388C and 300.08 MHz ꢀin bold, the values of the model and the assignment).

208
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
Fig. 4. ꢀa) Experimental 13C CPMAS spectrum of 3 at 125 MHz. ꢀb) Experimental 13C CPMAS spectrum of 5 at 125 MHz.
68.13, Cꢀ4a) 65.11, Cꢀ4b) 66.39, Cꢀ5) 69.20, Cꢀ6)
68.23, Cꢀ7) 61.33, Cꢀ8) 74.38, Cꢀ8a) 48.28, Nꢀ9)
110.80, Cꢀ9a) 49.72, CꢀO) 22.66, ꢀO)0-153.55, Cꢀ100)
54.83, Cꢀ200) 56.73, Cꢀ30) 62.13, Cꢀ4 ) 56.04, Cꢀ5 )
63.68, Cꢀ6 ) 57.96, Hꢀ1) 23.72, Hꢀ2) 24.59, Hꢀ3)
24.80, Hꢀ4) 23.97, Hꢀ5) 23.98, Hꢀ6) 24.96, Hꢀ7)
25.08, Hꢀ8) 25.80, Hꢀ20) 23.81, Hꢀ30) 24.53, Hꢀ40)
24.47, Hꢀ50) 24.89 and Hꢀ60) 24.73.
For 13C NMR, we have compared the experi-
mental chemical shifts at 130 K ꢀTable 3) as
well as the CPMAS data ꢀcentre of the multiplet)
with the absolute shieldings corrected by 204.3 ppm
to account for the reference. The CyO signal
was excluded:
d13C solution
ˆ ꢀ0:905 ^ 0:002†‰204:3 2 s13C calculatedŠ;
n ˆ 22; r2 ˆ 1:000
ꢀ1†
ꢀ2†
d13C CPMAS
ˆ ꢀ0:896 ^ 0:002†‰204:3 2 s13C calculatedŠ;
n ˆ 18; r2 ˆ 1:000

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
209
These equations predict for the CyO, 164.4 instead of
170.1 ꢀsolution) and 162.7 instead of 167.3 ppm
ꢀsolid). At this level of calculations, aromatic and
CyO carbons are mutually inconsistent.
has been done on N-acetyl derivatives [9], the
present work extends these studies to the almost
unexplored ®eld of N-benzoylazoles.
1
The H NMR spectrum is in good agreement with
the calculated shieldings, in this case all H atoms are
linked to aromatic carbons ꢀ32.47 is the absolute
shielding of TMS):
5. Experimental
5.1. Materials
d1H solution
Melting points were determined with a hot-stage
microscope and are uncorrected. IR spectra were
recorded in a Philips PU9714. Column chroma-
tography was performed on silicagel Merck 60 ꢀ70±
230 mesh) using the appropriate eluent.
ˆ ꢀ0:969 ^ 0:002†‰32:47 2 s1H calculatedŠ;
n ˆ 15; r2 ˆ 1:000
ꢀ3†
Compounds 1 to 5 have been prepared according
to one of the two following procedures, depending on
the acidity of the NH. For azolides 1, 2 and 4 proce-
dure 1 has been used. The azole ꢀ25.5 mmol) in solu-
tion in a dry solvent ꢀ50 mL) was placed in a three-
necked, round bottomed ¯ask provided with a
condenser under argon atmosphere and with
magnetic stirring. To this solution 1.5 mL
The0 la0rgest difference ꢀsee Fig. 3) is found for the
Hꢀ2 ,6 ) protons. This is probably related to the con-
formation of the benzoyl group about the Cꢀ10)ZCO
bond. The calculations correspond to a ®xed confor-
mation and the experimental value to free rotation
ꢀalthough with different populations of the rotamers).
According to the calculations, the proton Hꢀ20) close
to the CyO and the proton Cꢀ60) close to Hꢀ8) should
appear at 8.39 and 7.50 ppm, respectively, in the
twisted conformation.
In our previous paper about azolides, we have
described an empirical relationship between the
absolute shieldings ꢀcalculated at the same level)
and the chemical shifts for the 15N nucleus
ꢀd15N ˆ 2130 2 0.75s15N) [10]. In the case of 5,
s 15Nꢀ9) ˆ 110.80 ppm which corresponds to
d15N ˆ 2212.5 ppm, close to the experimental
value of 2218.6 ꢀTable 4).
ꢀ12.75 mmol)
of
benzoyl
chloride
ꢀ99%,
d ˆ 1.21 g cm23) were added and the mixture was
maintained at r.t. for 24 h. The solution was ®ltered
off ꢀazole hydrochloride) under argon and the solvent
evaporated; an oil remains. 1-Benzoylimidazole 1,
solvent benzene, oil, yield 94%. 1-Benzoylpyrazole
2, solvent diethyl ether, m.p. 47±488C, yield 75%. 1-
Benzoyl-benzimidazole 4, solvent tetrahydrofuran,
m.p. 93.0±93.58C, yield 86%, IR ꢀKBr cm21
)
nCyO ˆ 1665. For azolides 3 and 5 procedure 2 has
been used. A solution of azole ꢀ30 mmol) in 150 mL
of a mixture of dry benzene/dimethylformamide ꢀ1:1
ratio) was placed in an ice/salt bath. To this solution
60% NaH ꢀ1.2 g, 30 mmol) was added and the reac-
tion mixture magnetically stirred for 30 min. After
cooling, 4.21 g ꢀ30 mmol) of benzoyl chloride
ꢀ99%, d ˆ 1.21 g cm23) in 10 mL of benzene were
added controlling that the temperature remains
below 08C, then the mixture was left at r.t. for 2 h.
Sodium chloride was ®ltered off and the solution
evaporated under vacuum. After column chromato-
graphy using hexane:EtOAc 9:1, azolides 3 and 5
were obtained. 1-Benzoylindole 3, m.p. 68.0±
68.58C, 60% yield, IR ꢀKBr cm21) nCyO ˆ 1665. 9-
Benzoylcarbazole 5, m.p. 98±1008C, 63% yield, IR
ꢀcm21) nCyO ˆ 1665.
4. Conclusions
The combined use of crystallography, CPMAS
NMR, low temperature solution NMR and ab
initio calculations ꢀincluding absolute shieldings)
is an extraordinarily powerful tool to understand
the solid state and its relation with solution.
Thanks to this approach, the assignment of all
the signals of compound 5 in solution at low
temperature has been possible. This compound
presents a case of spontaneous resolution, although
in solution both enantiomers equilibrate very
quickly due to its low activation barrier ꢀabout
30 kJ mol21). Most structural work on azolides

210
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
Table 5
Crystal analysis parameters at room temperature
3
5
Crystal data
Chemical formula
Crystal colour
Crystal description
Crystal size ꢀmm)
Crystal system
Mr
C15H11NO
C19H13NO
Colourless
Prism
Colourless
Prism
0.67 £ 0.50 £ 0.50
Triclinic
221.258
P1
0.40 £ 0.13 £ 0.13
Orthorhombic
271.318
Space group
Ê
a/A
P
12121
2
12.1874ꢀ7)
7.6925ꢀ4)
6.2297ꢀ3)
97.226ꢀ5)
101.756ꢀ4)
90.047ꢀ7)
2
17.2138ꢀ20)
14.4042ꢀ16)
5.5668ꢀ7)
90
Ê
b/A
Ê
c/A)
a/8
b/8
90
g/8
Z
90
4
Dx/gr cm23
1.296
567.04ꢀ5)
67
1.306
1380.3ꢀ3)
43
Ê 3
V/A
No. re¯ections for lattice parameters:
u range for lattice parameters:
Absorption coef®cient ꢀcm-1)
2±45ꢀ8)
6.46
2±45ꢀ8)
6.36
Data collection
Diffractometer type
Philips PW1100
Seifert XRD3000-S
Graphite oriented monocromator.
Measurement time: 1 min./re¯ection.
Detector apertures: 1 £ 1 ꢀ8)
Collection method: w/2u scans,
Scan width ˆ 1.6 ꢀ8)
Two standard re¯ections every 100 min:
No variation
umaxꢀ8)
No. of independent re¯ections
No. of observed re¯ections
67.0
67.5
2022
1292
1908ꢀI . 2sꢀI) criterion)
909ꢀI . 1sꢀI) criterion)
Re®nement
Treatment of hydrogen atoms
Re®nement
Difference synthesis
Least-squares on Fo, Full matrix
R, wR
0.040, 0.061
0.051, 0.059
No. of parameters re®ned
Extinction parameters ꢀ £ 1024
199
0.44ꢀ6)
243
0.07ꢀ1)
)
Degrees of freedom
1709
666
Ratio of freedom
Ê 3
9.6
3.7
ꢀDr)max ꢀe/A )
kShift/errorl
Max. thermal value ꢀA )
0.21
0.001
0.23
0.011
Ê 2
U22[O17] ˆ 0.131ꢀ2)
U33ꢀO15) ˆ 0.168ꢀ6)
5.2. NMR spectroscopy
spectra were recorded at 300 K using a Bruker DSX
500 instrument ꢀ125.77 MHz) with a 4 mm probe, a
10 kHz rotation frequency and standard CPpulse
sequences were employed. Chemical shifts ꢀd) in ppm
The 1Hand 13C and 15NNMR spectra insolutionwere
recorded on a Bruker DRX-400 instrument working at
400.13 ꢀ1H), 100.62 ꢀ13C) and 40.56 MHz ꢀ15N) using
standard conditions. Solid state 13C CP MAS NMR
1
are referred to external Me4Si for H and 13C and to
external NO2CH3 for 15N NMR spectra.

R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
211
5.3. X-ray crystallography
References
Crystal data and experimental details are given
in Table 5. The structures were solved by direct
methods ꢀSIR97) [23]. The lack of suitable anomalous
scatters in compound 5 did not allow us to determine
the absolute con®guration of the conformational
enantiomer studied as illustrated by the value of
20.3ꢀ17) obtained for the Flack parameter [24] and
thus the Friedel pairs were merged. The weighting
schemes were established [25] in an empirical way
as to give no trends in kwD2Fl vs kFol or ksinu/ll:
w ˆ K/[ꢀa 1 b z Fo)2][c 1 d z sinu/l]; the a, b, c and d
parameters were adjusted to ¯atten the initial trends.
Secondary extinction correction was applied ꢀTable 5)
according to the Zachariasen formalism [26]. The
scattering factors were taken from the International
Tables for X-Ray Crystallography [27]. Most of the
calculations were performed using the Xtal3.6 system
of programs [28].
[1] J. Elguero, A. Fruchier, R. Jacquier, Bull. Soc. Chim. Fr.
ꢀ1966) 3041.
[2] L. Pappalardo, J. Elguero, C. Marzin, C.R. Acad. Sci. Ser. C
277 ꢀ1973) 1163.
[3] J. Elguero, C. Marzin, L. Pappalardo, Bull. Soc. Chim. Fr.
ꢀ1974) 1137.
[4] J. Elguero, C. Marzin, M.E. Peek, Org. Magn. Reson. 7 ꢀ1975)
445.
[5] M. Begtrup, R.M. Claramunt, J. Elguero, J. Chem. Soc. Perkin
Trans. 2 ꢀ1978) 99.
[6] R. Faure, E.J. Vincent, A. Rousseau, J. Elguero, Heterocycles
26 ꢀ1987) 333.
[7] A. Maquestiau, Y. van Haverbeke, R. Flammang, R.M.
Claramunt, J. Elguero, Bull. Soc. Chim. Fr. ꢀ1975) 2693.
[8] A. Maquestiau, A. Tommasetti, C. Pedregal, J. Elguero,
R. Flammang, Bull. Soc. Chim. Belg. 93 ꢀ1984) 1057.
[9] J. Elguero, C. Foces-Foces, D. Sanz, R.M. Claramunt,
Azolides: Structural aspects, in: C.J. Moody ꢀEd.), Advances
in Nitrogen Heterocycles, vol. IV, Jay Press Inc, Stamford,
Conneticut, 2000, pp. 295±367.
[10] R.M. Claramunt, D. Sanz, I. Alkorta, J. Elguero, C. Foces-
Foces, A.L. Llamas-Saiz, J. Heterocycl. Chem. 38 ꢀ2001) 443.
[11] H.A. Staab, Angew. Chem. Int. Ed. 1 ꢀ1962) 351.
[12] W. HuÈckel, H. Bretschneider, Ber. Dtsch. Chem. Ges. 70
ꢀ1937) 2024.
The CIF ®les were deposited with the Cambridge
Crystallographic Data Center ꢀCCDC 149366 and
149367).
[13] P. Grammaticakis, C. R. Hebd. Seances Acad. Sci. 250 ꢀ1960)
4373.
[14] T.H. Fife, R. Natarajan, M.H. Werner, J. Org. Chem. 52
ꢀ1987) 740.
5.4. Computational details
[15] Y.H.R. Jois, H.W. Gibson, J. Org. Chem. 56 ꢀ1991) 865.
[16] S. Ghosh, D.B. Datta, I. Data, T.K. Das, Tetrahedron 45
ꢀ1989) 3775.
Geometry optimisation on compound 5 was carried
out at the RHF/6-311G** level of the theory [29,30]
and absolute shielding calculated over that geometry
within the GIAO approximation [31,32]. All these
calculations were carried out using the Gaussian 98
facilities [33].
[17] F.H. Allen, J.E. Davies, J.J. Galloy, O. Johnson, O. Kennard,
C.F. Macrae, E.M. Mitchell, J.F. Mitchell, J.M. Smith, D.G.
Watson, J. Chem. Info. Comput. Sci. 31 ꢀ1991) 187.
[18] T. Kaneko, M. Matsuo, Y. Iitaka, Chem. Pharm. Bull. 29
ꢀ1981) 3499.
[19] G.R. Desiraju, T. Steiner, The Weak Hydrogen Bond in Struc-
tural Chemistry and Biology, Oxford University Press,
Oxford, 1999.
Acknowledgements
[20] A. Cipiciani, P. Linda, D. Macciantelli, L. Lunazzi, J. Chem.
Soc. Perkin Trans. 2 ꢀ1979) 1045.
Financial support was provided by the Spanish
DGICYT ꢀProject nos. BQU-2000-0252 and BQU-
2000-0868) and by the Comunidad de Madrid ꢀ07N/
0001/1999). We thank Dr. Stefan Steuernagel ꢀBruker
Spectrospin GmbH) for recording the 13C CP MAS
spectra of compounds 3 and 5 with a DSX 500 MHz
instrument. This work would have been completely
different if Professor Ludovico Lunazzi ꢀUniversity
of Bologna) would have not studied the NMR spectro-
scopy of compound 5 at very low temperatures: our
most warmest thanks.
[21] J. Jacques, A. Collet, S.H. Wilen, Enantiomers, Racemates,
and Resolutions, Krieger Publishing Co, Malabar, FL, 1994.
[22] W.E. Stewart, T.H. Siddall, Chem. Rev. 70 ꢀ1970) 517.
[23] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi,
A.G.G. Moliterni, M.C. Burla, G. Polidori, M. Camalli, R.
Spagna, SIR97. A Package for Crystal Structure Solution by
Direct methods and Re®nement, University of Bari, Italy, 1997.
[24] H.D. Flack, Acta Crystallogr, Sect. A 39 ꢀ1983) 876.
[25] M. Martinez-Ripoll, F.H. Cano, `PESOS' unpublished
program, Instituto Rocasolano, CSIC, Spain, 1975.
[26] W.H. Zachariasen, Acta Crystallogr. 23 ꢀ1967) 558.
[27] International Tables for X-Ray Crystallography, 1974,
Birmingham, Kynoch Press, England.

212
R.M. Claramunt et al. / Journal of Molecular Structure 605 92002) 199±212
[28] S.R. Hall, D.J. du Boulay, R. Olthof-Hazekamp ꢀEds.),
The Xtal System of Crystallographic Software, `Xtal3.6'
User's Manual, The University of Western Australia,
Australia, 1999.
S. Dapprich, J.M. Millam, A.D. Daniels, K.N. Kudin, M.C.
Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi,
B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K.
Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman,
J. Cioslowski, J.V. Ortiz, B.B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R.L. Martin, D.J. Fox,
T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P.M.W. Gill, B.G. Johnson,
W. Chen, M.W. Wong, J.L. Andres, M. Head-Gordon, E.S.
Replogle, J.A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
[29] A.D. McLean, G.S. Chandler, J. Chem. Phys. 72 ꢀ1980) 5639.
[30] R. Krishnan, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem.
Phys. 72 ꢀ1980) 650.
[31] R. Ditch®eld, Mol. Phys. 27 ꢀ1974) 789.
[32] F. London, J. Phys. Radium 8 ꢀ1937) 397.
[33] Gaussian 98 ꢀRevision A.1), M.J. Frisch, G.W. Trucks, H.B.
Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, V.G.
Zakrzewski, J.A. Montgomery, R.E. Stratmann, J.C. Burant,