Experimental and theoretical molecular and electronic structures of the N-oxides of pyridazine, pyrimidine and pyrazine

Article abstract of DOI:10.1016/j.tet.2012.05.004

The structures of pyridazine N-oxide, pyrimidine N-oxide and pyrazine N-oxide have been determined by X-ray diffraction for the first time. Comparison with theoretical predictions of the equilibrium structures using the B3LYP method together with a cc-pVTZ basis set, show close agreement with the structural parameters observed, and experimental dipole moments, which suggests that the charge distribution is realistic. An 'atoms in molecules' (AIM) analysis of the computed wave-functions shows total electron densities rather different from the classical picture of a dative bond, whereas the same wave-functions subjected to Mulliken analysis show a more conventional view of the electron distribution. This latter procedure allows a bond dipole analysis of the N-oxide charge distribution.

Full text of DOI:10.1016/j.tet.2012.05.004

Tetrahedron 68 (2012) 5845e5851
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Experimental and theoretical molecular and electronic structures of the N-oxides
of pyridazine, pyrimidine and pyrazine
R. Alan Aitken , Bernd Fodi , Michael H. Palmer ,*, Alexandra M.Z. Slawin a,*, Jing Yang a
a
,
b
b
*
a
EaStCHEM School of Chemistry, University of St Andrews, North Haugh, St Andrews, Fife, KY16 9ST, UK
School of Chemistry, The University of Edinburgh, West Mains Road, Edinburgh, EH9 3JJ, UK
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The structures of pyridazine N-oxide, pyrimidine N-oxide and pyrazine N-oxide have been determined by
X-ray diffraction for the first time. Comparison with theoretical predictions of the equilibrium structures
using the B3LYP method together with a cc-pVTZ basis set, show close agreement with the structural
parameters observed, and experimental dipole moments, which suggests that the charge distribution is
realistic. An ‘atoms in molecules’ (AIM) analysis of the computed wave-functions shows total electron
densities rather different from the classical picture of a dative bond, whereas the same wave-functions
subjected to Mulliken analysis show a more conventional view of the electron distribution. This latter
procedure allows a bond dipole analysis of the N-oxide charge distribution.
Received 7 February 2012
Received in revised form 15 April 2012
Accepted 1 May 2012
Available online 8 May 2012
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
derivative with difficulty. This is generally true of the azine N-ox-
ides when a para position is available. Less well-known is the ease
1
The importance of the heterocyclic amine oxides, such as pyri-
dine N-oxide (1, Fig. 1), lies in the ring susceptibility to electrophilic
substitution, where it is enhanced yielding a 4-nitro-derivative
with ease.1 This contrasts with pyridine, which yields a 3-nitro-
of cross-coupling reactions with aryl bromides at positions ortho to
the N-oxide. In the present paper we present structural and
3
electron distribution results for the series of compounds 1e5
(Fig. 1). A further factor of importance for these amine oxides, is
that some 4-nitro-derivatives of pyridine and quinoline N-oxides
,2
4
,5
are active against a number of tumours. The dative bond shown
1
in 6 (Fig. 1) for this series of compounds is widely used to explain
4
4
4
the reactivity of the 4-position in 1, with similar explanations for
the 2-position. This brings into focus both permanent and reaction
induced properties of the ring 6, which are discussed further below.
5
6
3
5
6
3
5
6
3
N
2
2
N
2
6
N1
O
1
N1
O
N
O
1
Although their synthesis was reported over 50 years ago, the
crystalline structures of the isomeric diazine (mono) N-oxides 2e4
have not so far been determined. We now report the results of
single-crystal X-ray diffraction studies on these three compounds.
There have been no previous X-ray diffraction (XRD) studies of
2
3
4
7,8
O
N
simple diazine N-oxides, but there are X-ray structures for 1 and
N
3
4
9
10e12
13
5
6
5
, as well as microwave (MW)
and electron diffraction (ED)
5
3
structures for 1; the MW studies are by far the most detailed, since
the X-ray and ED studies made a number of assumptions. These
data are shown in Table 1 and are compared with the corre-
sponding theoretical data presented here.
2
N1
O
2
N
6
N 1
O
6
O
5
14
Structures for a few substituted pyrazine N-oxide, and pyr-
4
0
15
azine-N,N -dioxides have been determined; a complex 1,2,5-
16
Fig. 1. Compounds studied with numbering scheme used.
oxadiazolo[3,4-d]pyridazine-5,6-dioxide appears to be the only
example of a crystal structure with the pyridazine-N,N -dioxide
0
function present. Although originally thought unlikely owing to the
0
adjacent positive charges, pyridazine N,N -dioxides have been
*
Corresponding authors. E-mail address: raa@st-and.ac.uk (R.A. Aitken).
0
040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.05.004

5846
R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
H
H
H
1
.0787
1
.0794
1.0813
Table 1
122.5 121.3
121.9
1
21.2
121.1
117.1
ꢀ
ꢀ
H
.0809
1.3903
117.6
H
H
1.0810
1.3888
121.9
1.3850
21.3
H
H
1.3870
120.6
1.3371
Comparison of experimental and theoretical bond lengths (A) and angles ( ) for
1
1
116.2
1.0826 1.0805
120.9
118.9
pyridine N-oxide 1 and pyridine
118.3
113.5
1.3292
N
118.8
1
20.6
118.9 119.2 125.3
1.3714
119.6
1.3775
1
.3771
1.3075
Pyridine N-oxide 1
Pyridine
121.4
118.4
1.3689
124.9
113.8
20.8
125.5 120.6 117.5
125.5 120.0 124.3 122.0
1
.0776
H
1.0768
121.2
N
1.0781
117.0
N
1.0803
XRD8,a ED13,b
MW11,44 DFT
XRD23,c MW24
DFT
N
H
H
113.9
1.3750
20.3
N
1.3440
118.5
1.2440
H
114.5
113.7
H
1.3647
1.3798
1
1
122.0 121.0
1.2675
1
.2711
Bond
N
N
1
eO
eC
1
2
3
4
5
1.330
1.348
1.013
1.371
1.078
1.375
0.867
1.375
1.078
1.371
1.013
1.348
1.290
1.384
1.2806
1.3644
1.271
d
d
O
O
O
N
O
1
2
1.369 1.336
1.078 0.979
1.377 1.381
1.081 0.949
1.390 1.377
1.080 0.964
1.390 1.377
1.081 0.949
1.377 1.381
1.078 0.979
1.369 1.336
1.3376 1.3330
1.0857 1.0846
1.3938 1.3908
1.0818 1.0815
1.3916 1.3882
1.0811 1.0822
1.3916 1.3882
1.0818 1.0815
1.3938 1.3908
1.0857 1.0846
1.3376 1.3330
C
C
C
C
C
C
C
C
C
C
2
eH
(1.070) 1.0775
1.381 1.3814
(1.070) 1.0820
1.393 1.3953
(1.070) 1.0787
1.393 1.3953
(1.070) 1.0820
1.381 1.3814
(1.070) 1.0775
1.3348
1
H
H
H
N
115.1
24.2
H
1.0833
1
17.2
2
3
3
4
4
5
5
6
6
eC
3
1
21.8
1
18.7
1
eH
1.3660
.3774
eC
4
125.2 119.9
16.9
123.1
1.0771
1
16.4
N
1
1.0773
H
eH
114.9
115.1
H
N
1
1.3692
H
1.3675
H
121.8
eC
5
121.6
.2597
1.2705
eH
O
O
eC
eH
eN
6
6
1
H
H
1.0759
1.384
1.3664
1
.0782
1.3507
N
Angle
H
.0792
1.3822
O
1.2295
H
1.3762
1.3698
O
C
C
2
eN
eN
1
1
2
2
2
eO
eC
eC
eH
eC
1
119.5
121.0
119.7
111.1
129.2
120.9
113.2
125.7
117.9
121.1
121.1
120.9
125.7
113.2
119.7
129.2
111.1
120.4
119.2
121.0
120.8
d
d
d
1
1
.0796
N
2
6
120.9
118.0
(120.5) 113.8
(121.5) 125.2
124.6
(108.5) 118.4
(126.9) 121.1
114.1
(123.0) 121.2
(123.0) 121.2
124.6
(126.9) 121.1
(108.5) 118.4
118.4 116.6
121.4 123.6
113.8 116.1
124.9 120.3
120.6 118.6
118.3 119.6
121.1 121.8
117.7 118.8
121.2 120.6
121.2 120.6
120.6 118.6
121.1 121.8
118.3 119.6
121.4 123.6
124.9 120.3
113.8 116.1
116.9
123.8
116.0
120.2
118.5
120.1
121.4
118.4
120.8
120.8
118.5
121.4
120.1
123.8
120.2
116.0
117.4
123.6
116.2
120.3
118.5
120.3
121.2
118.6
120.7
120.7
118.5
121.2
120.3
123.6
120.3
116.2
1
19.3
121.3
1
.3820
21.8
18.7
1.4736
115.6
1.3775
1.3584
N
N
H
1
eC
eC
eC
3
1
19.5 118.0
122.5
126.2 119.0 122.8 118.6
1
1
2
2
N
118.9
N
1.0740
1
126.4
114.8
3
H
O
H
H
121.4 119.7
1.2656
1
24.4 113.1
C
C
2
eC
eC
3
eC
eH
eC
4
121.0
2
3
3
H
O
H
3
eC
3
4
C
C
3
eC
eC
4
eC
5
117.6
Fig. 2. Theoretical structures for the present series.
3
4
eH
4
5
4 4
H eC eC
C
C
4
eC
eC
5
eC
6
121.0
5
eH
5
6
1
6
0
0
4
pyridazine N,N -dioxide and pyrimidine N,N -dioxide are shown for
completeness.
Before going on to describe the new experimental structural
results for compounds 2e4, a comparison of available experimental
(XRD, ED, MW) data with the results of the present theoretical
equilibrium structures for the N-oxides 1 and 5 are shown in
Tables 1 and 2, where the theoretical results are rounded to a sim-
ilar number of significant figures to the corresponding experi-
mental data.
H
5
eC
5
eC
eN
eH
C
C
5
eC
eC
6
118.0
(121.5) 125.2
(120.5) 113.8
121.0
5
6
H
6 6 1
eC eN
a
CCDC Ref Code PYRDNO11.
Values in parentheses are either assumed (lengths), or based on non-bonded
7
,8
9
b
distance estimates (for angles).
c
CCDC Ref Code PYRDNA01dMean of four molecules.
0
synthesised by direct oxidation, initially for cinnoline N,N -di-
17
3. Experimental study
oxide, the benzo-compound, and subsequently for pyridazine
0
18
N,N -oxide.
In the present study, we make comparisons between the azines
and their N-oxides, because the structure and electron density in
Structural data have
previously been determined by a range of techniques for the parent
Samples of the diazine N-oxides 2e4 were prepared by oxida-
tion of the corresponding diazines with either m-chloroperbenzoic
3
6,45
19,20
acid (mCPBA) or hydrogen peroxide in acetic acid.
Compounds
each pair of compounds are so different.
3
and 4 readily formed good quality prisms suitable for X-ray dif-
21e24
25e29
30e33
fraction upon simple recrystallisation; however, the low melting
point of 2 meant that suitable crystals were only formed with great
difficulty and the R-factor of the resulting structure is rather poorer.
The X-ray structure of pyridazine N-oxide 2 showed eight
molecules per unit cell with two distinct molecular geometries A
and B (Table 3, Fig. 3). As displayed in Fig. 3 these are centrosym-
azines: pyridine,
pyrazine.3
pyridazine,
pyrimidine
and
0,34e36
We also compare with the equilibrium structures determined
The wave-functions
can generate molecular properties, which are close to experiment
37,38
using density functional (DFT) methods.
37,38
for a number of molecules;
hence the wave-functions produced
A B A B
can be compared with experimental properties such as dipole
metrically arranged:
and there are no significant
moments and 14N quadrupole coupling, giving additional checks on
B A B A
intermolecular interactions.
reliability of the calculations. All of these aspects contribute to
knowledge of the nature of these polar molecules.
The X-ray structure of pyrimidine N-oxide 3 showed only a sin-
gle molecular geometry and four molecules per unit cell (Table 4,
Fig. 4).
2
. Theoretical methods
The X-ray structure of pyrazine N-oxide 4 again showed eight
molecules per unit cell with two distinct molecular geometries A
and B (Table 5, Fig. 5).
The equilibrium structures for the azines and their N-oxides
3
9
were determined using the Gaussian-03 suite of programmes. We
As displayed in Fig. 5 these are arranged:
use a cc-pVTZ basis set40 with the B3LYP hybrid version of the DFT
B
suite. The total electron densities at the atomic centres were de-
termined by the AIM method using the AIMPAC package.4
1,42
The
A
A
A
B
B
B
same wave-functions were also subjected to a classical Mulliken
Analysis,43 giving the results discussed later. The theoretical
structures for all the compounds are presented in Fig. 2, where
A

R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
5847
Table 2
ꢀ
ꢀ
Comparison of theoretical and experimental bond lengths (A) and angles ( ) for
pyrazine-di-N-oxide 5 and pyrazine
Pyrazine-di-N-oxide 5
Pyrazine
XRD35,b
XRD9,a
DFT
ED30
DFT
Bond
N
N
1
eO
1
1.296
1.353
0.930
1.362
0.930
1.358
1.296
1.353
0.930
1.362
0.930
1.358
1.2705
1.3675
1.0771
1.3660
1.0771
1.3675
1.2705
1.3675
1.0771
1.3660
1.0771
1.3675
d
d
d
1
eC
2
1.333
0.939
1.388
0.939
1.333
d
1.339
1.115
1.403
1.115
1.339
1.3320
1.0839
1.3906
1.0839
1.3320
d
C
C
C
C
2
eH
2
2
3
3
eC
eH
eN
3
3
4
Fig. 3. Unit cell of pyridazine N-oxide 2 (viewed along a-axis).
N
N
4
eO
4
4
eC
5
1.333
0.939
1.388
0.939
1.333
1.339
1.115
1.403
1.115
1.339
1.3320
1.0839
1.3906
1.0839
1.3320
C
C
C
C
5
eH
5
Table 4
ꢀ
ꢀ
eC
eH
eN
6
Comparison of theoretical and experimental bond lengths (A) and angles ( ) for
5
6
6
pyrimidine N-oxide 3 and pyrimidine
6
1
Pyrimidine N-oxide 3
Pyrimidine
XRD31,a
Angle
2
eN
6
eN
2
eN
1
1
1
2
2
2
eO
eO
1
1
121.1
120.6
118.3
120.9
119.6
119.5
120.8
119.6
119.6
118.3
120.6
121.1
120.9
119.6
119.5
120.8
119.6
119.6
121.8
121.8
116.4
121.8
115.1
123.1
121.8
123.1
115.1
116.4
121.8
121.8
121.8
115.1
123.1
121.8
123.1
115.1
d
d
d
XRD
DFT
MW32/ED30
DFT
C
C
C
d
d
d
Bond
eC
eC
eH
6
116.2
121.9
115.9
122.1
121.9
122.1
115.9
116.2
d
115.6
122.2
113.9
123.9
122.2
123.9
113.9
115.6
d
116.1
122.0
117.2
120.9
122.0
120.9
117.2
116.1
d
N
N
1
eO
eC
1
1.308
1.359
0.950
1.315
1.347
0.951
1.375
0.950
1.369
0.950
1.359
1.2674
1.3796
1.0802
1.3071
1.3370
1.0813
1.3870
1.0805
1.3774
1.0781
1.3649
d
d
d
N
N
H
1
eC
eC
eC
3
1
2
1.331
1.010
1.337
1.339
0.959
1.387
0.921
1.388
0.966
1.337
1.317
1.097
1.329
1.351
1.096
1.394
1.098
1.393
1.087
1.361
1.3322
1.0842
1.3322
1.3322
1.0807
1.3871
1.0844
1.3871
1.0844
1.3322
1
2
2
C
C
2
eH
eN
2
3
4
4
eC
3
2
C
C
2
eC
eC
3
eN
eH
4
N
3
eC
2
3
3
C
C
C
C
C
C
4
eH
H
3
eC
3
4
4
eN
4
4
5
5
6
6
eC
5
C
C
3
eN
eN
eC
eO
5
eH
5
3
4
5
eC
eH
eN
6
O
N
N
H
4
eN
4
eC
d
d
d
6
1
4
eC
4
eC
5
eC
5
eC
6
121.9
115.9
122.1
121.9
122.1
115.9
122.2
113.9
123.9
122.2
123.9
113.9
122.0
117.2
120.9
122.0
120.9
117.2
5
5
eH
5
Angle
eC
6
C
C
C
2
eN
6
eN
2
eN
1
1
1
2
2
2
3
4
4
4
eO
eO
eC
eN
eH
eN
eC
eC
eH
eC
eC
eH
1
1
119.9
121.4
118.7
124.0
118.0
118.0
117.3
122.0
119.0
119.1
118.9
120.6
120.6
119.1
120.5
120.4
121.0
122.0
117.0
124.3
113.7
122.0
118.8
121.0
117.1
121.9
118.9
121.5
119.6
120.0
125.5
114.5
d
d
d
C
C
5
eC
eC
6
eN
eH
1
d
d
d
5
6
6
6
115.9
126.8
115.3
117.9
116.3
121.9
113.9
124.0
116.5
118.9
124.6
122.6
124.7
112.5
116.0
127.8
116.1
116.1
116.0
121.2
117.9
120.9
117.9
125.0
125.0
121.2
120.9
117.9
116.0
127.0
116.5
116.5
116.0
122.2
116.6
121.2
116.6
121.7
121.7
122.2
121.2
117.9
H
6 6 1
eC eN
N
N
H
1
eC
eC
eC
3
a
CCDC Ref Code AHEMAB.
CCDC Ref Code PYRAZI01.
1
2
2
b
3
C
2
eN
4
N
N
H
3
eC
eC
eC
5
3
4
4
Table 3
ꢀ
ꢀ
5
Comparison of theoretical and experimental bond lengths (A) and angles ( ) for
pyridazine N-oxide 2 and pyridazine
C
C
4
eC
eC
5
5
6
4
5
Pyridazine N-oxide 2
Pyridazine
XRD25,a
H
5
eC
5
eC
eN
eH
6
1
6
C
C
5
eC
eC
6
XRD A
XRD B
DFT
MW/ED28,29
DFT
5
6
Bond
H
6 6 1
eC eN
N
N
N
1
1
2
eO
eN
eC
1
1.242
1.357
1.330
0.951
1.384
0.950
1.351
0.951
1.378
0.950
1.377
1.278
1.359
1.304
0.950
1.373
0.951
1.373
0.951
1.367
0.950
1.350
1.244
1.344
1.329
1.083
1.385
1.079
1.389
1.081
1.372
1.077
1.375
d
d
d
a
CCDC Ref Code PRMDIN01.
2
1.346
1.325
0.979
1.395
0.934
1.371
0.945
1.390
0.988
1.330
1.3370
1.3379
1.0787
1.4000
1.0707
1.3846
1.0707
1.4000
1.0787
1.3379
1.3295
1.3296
1.0831
1.3926
1.0813
1.3779
1.0813
1.3779
1.0831
1.3296
3
C
C
C
C
C
C
C
C
3 3
eH
4
4
. Discussion of the theoretical study
.1. Correlation of theory and experimental structures
The calculated azine N-oxide data is shown in Fig. 2. In order to
gain some insight into the general reliability of the theoretical
method for this type of compound, and hence its predictive value,
we show a comparison with other spectral data in the Tables 1e5.
3
4
4
5
5
6
6
eC
4
eH
4
eC
5
eH
5
eC
eH
eN
6
6
1
Angle
eN
N
2
1
eO
1
118.5
120.7
120.8
117.7
124.4
118.0
117.7
117.1
121.5
121.4
120.5
119.7
119.8
119.6
120.2
120.2
116.7
121.6
121.6
116.5
126.2
116.8
117.0
116.0
122.0
122.0
119.5
120.2
120.3
120.2
119.9
119.9
118.5
120.3
121.2
117.5
125.3
113.4
121.3
116.2
121.3
122.5
119.2
121.9
118.9
120.6
125.5
113.9
d
d
d
C
C
6
eN
eN
1
eO
eN
1
d
d
d
6
1
2
119.3
118.9
123.9
115.5
120.6
117.1
120.1
122.8
116.9
123.8
119.3
123.9
120.0
116.1
119.4
119.4
123.7
114.9
121.3
116.9
120.7
122.4
116.9
122.4
120.7
123.8
121.3
114.9
119.5
119.5
123.7
115.1
121.3
116.9
120.9
122.2
116.9
122.2
120.9
123.7
121.3
115.1
N
N
N
H
1
2
2
3
eN
2
eC
3
eC
eC
eC
3
eC
eH
eC
eC
eH
eC
eC
6
eH
4
3
3
3
4
C
C
3
eC
eC
4
5
3
4
4
H
4
eC
4
5
C
C
4
eC
5
eC
5
4
5
6
1
6
H
5
eC
5
eC
eN
eH
C
C
5
eC
eC
6
5
6
H
6 6 1
eC eN
a
CCDC Ref Code VOBJEB.
Fig. 4. Unit cell of pyrimidine N-oxide 3 (viewed along b axis).

5848
R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
Table 5
then the fits are somewhat poorer, with A 0.079(60), B 1.135(48), R
ꢀ
ꢀ
Comparison of theoretical and experimental bond lengths (A) and angles ( ) for
pyrazine N-oxide 4 and pyrazine
0.967 and SD 0.042 for the bond lengths.
Clearly, the theoretical study relates to isolated molecules, and
Pyrazine N-oxide 4
Pyrazine
XRD35,a
the MW and ED studies are more relevant than X-ray studies of the
solid state. If we correlate the X-ray bond length data with the
theoretical results (Tables 1e5) for the azines, then the coefficients
are A ꢂ0.041(51), B 1.028(38) with R 0.982, SD 0.006 over 30 points;
this is still clearly acceptable. However, a correlation over all X-ray
data for the four N-oxides gives A ꢂ0.412(128), B 1.309(95), R 0.940,
SD 0.014 over 27 discrete points. This is little changed (R 0.959, SD
XRD A
XRD B
DFT
DFT
Bond
N
N
1
eO
1
1.293
1.364
0.951
1.362
0.949
1.340
1.339
0.951
1.372
0.950
1.355
1.272
1.359
0.950
1.359
0.949
1.348
1.348
0.950
1.359
0.950
1.363
1.2597
1.3692
1.0773
1.3774
1.0833
1.3348
1.3348
1.0833
1.3774
1.0773
1.3692
d
d
1
eC
2
1.333
0.939
1.388
0.939
1.333
1.333
0.939
1.388
0.939
1.333
1.332
1.084
1.391
1.084
1.332
1.332
1.084
1.391
1.084
1.332
C
C
C
C
2
eH
2
2
eC
3
eH
3
eN
3
3
4
5
5
7
0.013) by removal of the older X-ray data for 1. The fit of the azine
N
4
eC
N-oxides angles is: A 19.4(78), B 0.84(7), R 0.899, SD 0.97 over 41
C
C
C
C
5
eH
angles. Almost all differences between theory and experiment are
5
eC
6
eH
6
eN
6
ꢀ
less than 2 , and this must be seen as a satisfactory result. The
6
1
wider correlation over all 58 angles in azines and N-oxides is:
Angle
y
Calcd¼AþBxObsd, with A ꢂ7.8(50), and B 1.07(4), with correlation
C
C
C
2
eN
6
eN
6
eN
1
1
1
2
2
2
eO
eO
1
1
120.4
121.3
118.3
119.0
120.5
120.6
124.9
117.6
117.5
114.3
124.2
117.9
117.9
119.3
120.3
120.4
121.0
121.3
117.8
119.3
120.4
120.3
125.0
117.5
117.6
113.8
124.3
117.9
117.8
119.8
120.1
120.1
121.6
121.6
116.9
119.8
114.9
125.2
124.2
118.7
117.2
115.0
124.2
117.2
118.7
119.8
125.2
114.9
d
d
coefficient R 0.959 and overall SD 0.856. Easily the worst discrep-
ancy is with the C(4)eC(5)eH(5) angle in pyrimidine, where the
present data suggests the presence of an error in the ED
publication.
Hence, as expected, we conclude (a) that the difference between
the calculated and X-ray structural results arise from solid state
effects. (b) Clearly the MW correlations are excellent, and we an-
ticipate that future MW studies of the N-oxides will give very
similar values to the present calculated data.
d
d
eC
eC
eH
2
116.2
121.9
115.9
122.1
121.9
122.1
115.9
116.2
121.9
115.9
122.1
121.9
122.1
115.9
116.1
122.0
117.2
120.9
122.0
120.9
117.2
116.1
122.0
117.2
120.9
122.0
120.9
117.2
N
N
H
1
eC
eC
eC
3
1
2
2
eC
3
C
C
2
eC
3
eC
3
eN
eH
4
2
3
H
3
eC
eN
3
4
5
5
5
eN
4
C
3
eC
eC
eH
5
N
N
H
4
eC
eC
eC
6
4
5
5
eC
6
4
.2. Structural variations
C
C
5
eC
eC
6
eN
eH
1
5
6
6
Both theory (Fig. 2) and experiment (Tables 1e5) show that the
H
6 6 1
eC eN
ꢀ
bond lengths (A) have comparatively narrow ranges in the mono N-
oxides, with NeO 1.25 to 1.27, CeN 1.33 to 1.38 and CeC 1.37 to
a
CCDC Ref Code PYRAZI01.
ꢀ
1.39 A. The longest CeC bonds lie furthest from the N-oxide group,
as in the 3,4-bond (2) and 4,5-bond (3), respectively. The theoretical
CeH bond lengths, show systematic variation, with the longest
ꢀ
bonds (>1.08 A) being meta to the N-oxide group. Overall the
median difference between theory (T) and experiment (E) (‘the
error’) for the bond lengths is (EeT): ꢂ0.011 (CeC), þ0.006 (CeN)
ꢀ
and þ0.029 A (NeO). Clearly all the calculated NeO bond lengths
are slightly too long by this theoretical method.
The theoretical NeO bond length is more variable in the di-N-
oxides, being shortest in the case of pyridazine di-N-oxide. The
NeN bond lengths of the mono- and di-N-oxides of pyridazine
ꢀ
show a major difference, with lengths of 1.344 and 1.473 A, re-
spectively. Clearly the latter shows a lengthening towards ring
4
6
opening,
which would ultimately lead to the di-nitroso-
compound. However, this structure (Fig. 2) is rather different
47
from that for the cis- or trans-dimers of nitrosomethane, where
ꢀ
the NN, CN and NO bonds are 1.31, 1.31 and 1.47 A, respectively. In
the N-oxides studied, by theory and/or experiment, the ring angle
at the N-oxide group is significantly less than 120 , with pyridazine
Fig. 5. Unit cell of pyrazine N-oxide 4 (viewed along b axis).
ꢀ
In this Section, we will correlate theory and experiment using the
equation yCalcd¼AþBxObsd, giving the correlation coefficient (R) and
overall standard deviation (SD). In these data, SD for the parameters
N-oxide as the sole exception. Similarly, the ring angle para to the
ꢀ
N-oxide is also smaller than 120 , with the result that the other
ꢀ
internal ring angles are larger than 120 . In all cases of the theo-
(
A/B) is shown in parentheses.
Easily the best fit for the theoretical and experimental results is
retical structures, the cis-group OeNeCeH has an NCH angle lower
ꢀ
ꢀ
than 120 by 10e12 when compared with the angle on the op-
posite side of the HC bond. This is not found with all the present
X-ray structures, but this may be a result of the poorly defined
position of the H-atoms.
with the microwave (MW) structural data (Tables 1e5). For 43 MW
bond lengths (A), we have the linear correlation Equation above
ꢀ
ꢂ3
ꢀ
with A 0.020(11), B 0.981(8), R 0.9985, SD 7.8ꢁ10 . The bond an-
ꢀ
gles for the azines have a range from 115 to 128 . Using the same
Equation, a similar fit between theoretical bond angles and ex-
perimental MW values, yields, A 1.938 (2644), B 0.984 (22), R
4.3. Electron density considerations
0
.9876, SD 0.449 over 52 points. Since the SD for the intercept in
The overall electron distributions as measured by the molecular
dipole moments (Table 6) are very close between experimental (X)
and present theoretical (Y) values. A linear correlation has
this angle correlation is larger than its intercept, the line passes
through the origin, while that for the bond lengths lies close to the
origin. However, if we include the larger group of azine data (41
bond measurements) where most is by electron diffraction (ED),
YCalcd¼ꢂ0.077(49)þ0.987(15)XObsd, with overall correlation co-
efficient (R) 0.9993 and standard deviation (SD) 0.065 over the 8

R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
5849
Table 6
Comparison of experimental and theoretical dipole moments (Debye)
0,0313
0.0422
H
0.0471
H
H
0
.0508
H
0.0564
H
0.0504
H
0
.0389
H
0.0131
0.5427
0
.0129
Azine/Azine N-oxide
Experiment
Ref.
Theory
H
H
0.5360
–0.0214
N –1.1835
Pyridine
Pyridazine
Pyrimidine
Pyrazine
2.35
4.22
2.334
0.0
4.13
5.21
3.65
1.66
d
21
26
32
d
2.177
4.105
2.280
0.0
4.014
5.087
3.514
1.454
5.745
4.039
0.0231
0.0260
0.3836
H
0.4040
H
1
.0220
0.3934
–0.6311
N
H
N
O
N
O
N
O
H
–
0.5661
–0.0743
0.0948
–0.5722
–0.5405
0.0844
0.0924
0.0990
Pyridine N-oxide 1
Pyridazine N-oxide 2
Pyrimidine N-oxide 3
Pyrazine N-oxide 4
11,44
49
50
49
d
–0.4971
–
0.5528
8
9
7
00255
H
H
0
Pyridazine N,N -dioxide
0.0514
H
–1.1371
N
0
0395
0
H
H
Pyrimidine N,N -dioxide
d
d
0.0233
H
H
H
–
0.0118
H
H
0
.5811
–00102
–
0.0149
.5452
0
.3678
0.5306
points, and SD for the parameters in parentheses. This leads to
confidence that the theoretical study is close to reality. For com-
pleteness, the values for the di-N-oxides are included.
N
0
H
N
H
N
–0.5410
–0.5222
0
.0927
H
N
–0.6034
0.0429
0.0261
–1.1709
O
The molecular dipole moments show that the substitution of
12
1
0
11
4 4
C H in 1 by N in 4 leads to a reduction in dipole moment but the
H
effect is relatively small. This similarity extends to pyrazine di-N-
oxide 5. Similarly, the electron distribution in pyridazine di-N-oxide
0
.0354
H
0.0384
H
–1.1488
N
H
H
–
1.1676
is relatively similar to that of its mono-N-oxide 2; an exception is N
2
–0.0214
N
0.5360
of the latter, where the positive charge arises from small contri-
butions from all ring atoms. The dipole moment (in-plane) di-
rections for the pyridazine- and pyrimidine N-oxides lie nearly
0.5473
1.0955
H
N
H
H
N
0.0400
0.0489
14
perpendicular to the N
1
eN
2
and N
1
eC
2
bonds, respectively.
13
The present theoretical value for the DM of pyridine N-oxide
Fig. 6. AIM charges at the atomic centres based on electron density summations.
compares favourably with a recent 6-311G(2d,2p) basis set using
4
8
the BPW91 methodology (3.80 D).
A direct measure of electron densities is the ‘Atoms in Mole-
There is clearly no indication that an N-oxide group enhances the
electron density at the para centre in the ground state molecules.
The enhanced para reactivity must therefore be a polarisation effect
at the time of reaction with an electrophile.
The Mulliken charge densities are shown in Fig. 8. As discussed
above, the distributions, being based on quite different criteria, do
show a number of similarities to the AIM total electron densities,
and both have a use in rationalisation of spectral and reactivity data
however, and hence the Mulliken populations are discussed in this
Section. The electron density in pyridine- and pyrimidine N-oxides
are fairly similar, as expected on molecular orbital grounds, where
41,51
cules’ (AIM) procedure.
In that procedure, the total electronic
wave-function, determined as densities at a grid of points, is in-
ꢀ
tegrated out to a defined distance (9 A in the present cases) yielding
the total electron density at the atomic centres.
The AIM electron densities for the azines and their N-oxides are
shown as net atomic charges in Fig. 6; these were obtained by direct
integration of the electron density at each nuclear site in turn, and
hence are directly related to experimental electron densities. These
electron densities are rather different from the Mulliken pop-
ulations shown later (Fig. 8); Mulliken populations attribute the
total population of each basis function (BF) to the centre carrying
the BF, irrespective of the spatial distribution of the electron den-
sity; however, it is important to note that the same wave-functions
are used as in the AIM procedure.
the back-donation from oxygen is through the
p-system, and the
meta-orientation of the N-atoms leads to a weak interaction.
0.0058
H
0.0027
H
The most obvious features of the AIM densities, for both types of
compound, azines 11e14 and the N-oxides 7e10 (Fig. 6), is that the
H-atoms are relatively close to neutral. The CeH bonds have very
0.0113
H
0.0135
H
0
.0156
H
0
.0247
0.0233
H
H
0.0380
0.0362
0.0054
5
2
low C
direction shown. However, an N-atom attached to C causes a large
bond dipole C (ꢂ), in the azines 11e14 where is
(þ)eN
.53ee0.56e. This is repeated in the N-oxides, where a high charge
density occurs at both N and N in pyrazine N-oxide 10, for ex-
ample. This is clearly not a result of the N-oxide group, since in
pyridine N-oxide, no similar effect occurs. Indeed, the C atoms in
d(ꢂ)eH
d(þ), dipoles, where
d <0.1e with the dipole bond
–0.1515
–0.1470
–0.0280
N
d
d
d
H
N
O
H
0.0495
N
O
0
.6048
0.5291
0.0583
0
1
4
–0.4971
–0.5528
8
–12
7
–11
2
the N-oxides 7e10 have much lower positive charge (w0.4e) than
other C-atoms; this suggests that the N-oxide is actually a donor to
0.0071
H
0
.0130
H
0.0117
N
C
2
/C
6
rather than C
4
in the ground state molecule. If we compare
0.0150
H
–
0.0046
H
H
the N-oxide directly with the corresponding azine, e.g., 7 with 11
etc. (Fig. 7), we find that the O-atom of the N-oxides 7, 9 and 10
carry a negative charge of ꢂ0.50e to 0.55e, with the attached N-
0.0451
N
–0.0159
0.0000
–0.1682
–
0.0735
–0.1433
atom þ0.60e. In the case of the pyridazines (8), the O
d
(ꢂ)eN (þ)
d
H
N
dipole is smaller, and part of the electron density is transferred to
the adjacent N atom. Using the same oxide/azine comparisons, the
H
N
H
0.6078
0.5222
0
.0548
0.6104
0.0543
0
.0501
2
–
ortho C-atoms show a surplus density of w0.15e in each pair of
compounds (7,11; 8,12; 9,13; 10,14). However, this completes the
main charge re-distribution in the N-oxides; the meta C-atoms
carry a small electron deficiency, with the same at the para sites.
O
O –0.5405
10–14
9
–13
Fig. 7. Electron density differences between azine N-oxide and corresponding azine.

5850
R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
0.1298
H
0
.1332
H
0
.1156
H
close relationship between the dipole moments, as determined
theoretically and by experiment, extends this conclusion to the
electron distributions. The variation in bond lengths and angles
with structure show systematic effects in the theoretical study,
and some of these are reflected in the experimental crystal
structures. The limitations of these comparisons probably arise
from the effects of neighbour N-oxide molecules in the solid
state, and in particular to the dipoleedipole interactions, which
occur.
The charge densities in the N-oxides have been a matter for
discussion, and we have endeavoured to both rationalise the sub-
stituent effects across the series of compounds, and draw attention
to difference between the total electron density (AIM) and the
Mulliken populations.
0
.1362
H
0
.1278
H
–0.1312
0.1353
H
–0.1384 0.1435
–0.0533
H
H
H
–
0
0.1635
.0172
–
0.1016
N
–
0.0775
–0.0719
–0.0656
–0.0616
–
0.0887
–0.0592
H
N
–0.1331
N
H
0.1359
H
N
O
H
0.1477
N
0.2187
0
.2405
0.2699
0.1486
0.1401
μ 4.104 D
μ 5.087 D
μ 3,514 D
–
0.4120 O
–
0.4284
–0.3547 O
0
.1327
H
0.1361
–0.1860
N
O
N
H
H
H
–
0.1133
0
.1510
H
–
0.0322
H
H
H
H
H
–
0.0540
–
0.0856
0
.2039
O
–0.0565
H
N
0
N
.2403
0
.1514
H
N
O
N
O
4
The ready reactivity of the pyridine N-oxide at C to electrophiles,
0
.2129
μ 1.454 D
O
0.1583
–0.3938
is most likely to be due to a polarisation effect in the presence of the
reagent, rather than to a high electron density at that centre.
All of the above study relates to ground state properties, and in
particular to structural features, such as bond lengths and angles,
and their relationship to theoretical results by the two main
methods available to computational chemists, namely AIM and
Mulliken analyses.
The differences between these methods are significant, and
this must be understood to avoid the conclusion that there is only
one measure of electron density in a molecule. The density in-
tegration procedure in AIM is clearly not practicable for large
molecules, since it would require summations at many centres. In
those circumstances, the Mulliken method of summing the den-
sity in individual molecular orbitals, about the centre where the
basis function is centred, is clearly much more easily scalable to
large molecules. Provided that there are only valence type func-
tions, the AIM and Mulliken methods may well give similar re-
sults; however, when diffuse functions are employed, as is
necessary for charged species and very polar molecules, the two
methods can give very different results, and we see that occurring
here.
–0.4166
–0.3203
μ 5.746 D
0
.0143
H
H
0.1442
1.1380
N
0.0142
0.0123
H
1
.0098
N
H
O
H
H
N
0.9793
0.9423
–0.1000
0.2232
–
0.0115
.1591
H
1.0417
H
–0.0722
1.0258
H
0
H
N
N
O
H
N
H
1
.3098
1.2879
0.1511
0.0136
0.0141
μ 4.036 D
–
0.3980 O
1.6001
O
1.5532
π-electrons only
π-electrons only
Fig. 8. Calculated Mulliken charges and dipole moments.
The ease of electrophilic substitution at C
has long been attributed to back-donation as in 6. The present data
in Fig. 8 shows that although the 8 -electron system undergoes
considerable reorganisation from the classical N eO
4
in pyridine N-oxide
p
(
þ)
(ꢂ)
repre-
sentation, only about 0.4e is lost from the O-atom, and most of that
is localised on the N-atom, thereby reducing the positive charge.
This situation is very similar to the AIM results. For pyrazine N-
4
oxide, the additional N-atom draws a further charge to the N po-
Now the question of reactivity raises further issues, and in the
case of the N-oxides, it is facile electrophilic substitution at posi-
tions para to the N-oxide, which is the group characteristic. Both in
sition, but otherwise the effects are small. Thus if we take the dif-
ference in Mulliken atomic populations at each centre between
pyrazine N-oxide and pyrazine, we obtain the following (approxi-
acid catalysed proton exchange and nitration we are probably
mate) values: O
þ0.02, H /H þ0.02, H
1
ꢂ0.39, N
1
þ0.39, N
4
ꢂ0.03, C
2
/C
6
ꢂ0.03, C
3 5
/C
þ
comparing the pyridinium cation (C
5
H
5
NH ) with the N-hydrox-
2
6
3
/H
5
0.00e. After allowance for the electron
þ
ypyridinium cation (C
5
H
5
NOH ). The former must have a de-
distribution in the azines (Fig. 7), these figures are somewhat
similar to the AIM conclusions. The extent of this back-bonding
seems general for these compounds, and is largely localised in
the O/N moiety.
However, a further effect, which will have a significant effect on
the reactivity in the N-oxides is the Madelung potential (Eq.1),
where the presence of internal electric dipoles in the molecule have
a distinct effect on both the electron density and the potential at the
ficiency in electron density available at the 2- and 4-positions for
electron donation to an incoming electrophile. The latter in contrast
is closer to the situation in a phenol, where ready electrophilic
substitution occurs.
There is no doubt that para-localisation energy is facilitated at
C
4
in pyridine N-oxide 1; in the case of proton exchange at that
position, a tetrahedral C centre is generated by out of plane attack
by the incoming electrophilic H in this instance (or NO
4
þ
þ
2
in ni-
4
2
nucleus. The potential energy and electron density are modified
by a local dipole (q ) at an angle w and at a distance r from the
tration).
p
-Electron release from the N-oxide oxygen atom must be
i
i
i
þ
significantly easier than from the protonated ring NH of the pyr-
centre (i), as in Eq. 1 Clearly this term is maximal when the cosine is
unity, i.e., in line with the centre under scrutiny. This situation
applies to the para position of the azine N-oxides in particular,
because of the very high bond dipoles, but is less elsewhere in the
molecules. It provides a partial interpretation of the higher electron
idine ring, owing to the high
NeOH groups.
p-electron density on the NeO or even
þ
6. Experimental section
4
density at N than is given by the Mulliken populations.
X
6.1. General
qicoswi
DE
¼
(1)
ri
i
NMR spectra were run at 300 MHz using a Bruker Avance 500
instrument on solutions in CDCl with internal Me Si as refer-
3
4
5
. Conclusions
ence. Chemical shifts are reported in parts per million to high
frequency of the reference and coupling constants J are in Hertz.
The credibility of the theoretical structures is demonstrated
by comparison with microwave structures for the azines; the
Melting points were determined on
microscope.
a Reichert hot-stage

R.A. Aitken et al. / Tetrahedron 68 (2012) 5845e5851
3. Leclerc, J.; Fagnou, K. Angew. Chem., Int. Ed. 2006, 45, 7781e7786.
5851
6
.2. Synthesis
4. Moore, P. R.; Mannering, G. J.; Teply, L. J.; Kline, B. E. Cancer Res. 1960, 20,
6
28e634.
6
.2.1. Pyridazine N-oxide 2.3 A solution of pyridazine (2.0 g,
5
. Nunoshiba, T.; Demple, B. Cancer Res. 1993, 53, 3250e3252.
2
5 mmol) and mCPBA (8.625 g, 50%, 25 mmol) in CH
P (3.275 g, 12.5 mmol),
stirring for a further 4 h, then evaporation and chromatography of
the residue (SiO , EtOAc to 20% MeOH in EtOAc) gave a major
2
Cl
2
(150 mL)
6. Koelsch, C. F.; Gumprecht, W. H. J. Org. Chem. 1958, 23, 1603e1606.
7
€
. Ulku, D.; Huddle, B. P.; Morrow, J. C.; Marsh, R. E. Acta Crystallogr., Sect. B 1971,
€
was stirred at rt for 48 h. Addition of Ph
3
27, 432e436.
8
. Kapon, M.; Hu, S.; Herbstein, F. H. Acta Crystallogr., Sect. B 2002, 58, 62e77.
9. N a€ ther, C.; Kowallik, P.; Jess, I. Acta Crystallogr., Sect. E 2002, 58, o1253eo1254.
2
10. Snerling, O.; Nielsen, C. J.; Nygaard, L.; Pedersen, E. J.; Sorensen, G. O. J. Mol.
fraction consisting of a mixture of product and starting material.
Vacuum distillation of this gave unreacted pyridazine followed by
the title product (1.62 g, 68%) as a colourless liquid, bp 175 C at
Struct. 1975, 27, 205e211.
11. Sorensen, G. O.; Tang-Pedersen, A.; Pedersen, E. J. J. Mol. Struct. 1983, 101,
ꢀ
263e268.
12. Heineking, N.; Dreizler, H.; Endo, K.; Kamura, Y. Z. Naturforsch., A 1989, 44,
2
6
H
0 Torr; d 7.19 (1H, ddd, J 7.5, 5.1, 1.0), 7.76 (1H, m), 8.24 (1H, dt, J
6
1196e1200.
.3, 1.0) and 8.52 (1H, m). The product solidified upon cooling (lit.
1
3. Chiang, J. F. J. Chem. Phys. 1974, 61, 1280e1283.
14. Sato, N.; Matsumoto, K.; Takishima, M.; Mochizuki, K. J. Chem. Soc., Perkin Trans.
1 1997, 3167e3172.
ꢀ
mp 38e39 C) and crystals suitable for X-ray diffraction were
obtained from Et
ꢀ
2
O by cooling at ꢂ78 C.
1
5. Kucharski, T. J.; Oxsher, J. R.; Blackstock, S. C. Tetrahedron Lett. 2006, 47,
569e4572; Cong, Q.; Lin, S.; Wang, H. Jiegou Huaxue 1989, 8, 31e35.
6. Fruttero, R.; Ferrarotti, B.; Gasco, A.; Calestani, G.; Rizzoli, C. Liebigs Ann. Chem.
988, 1017e1023.
17. Palmer, M. H.; Russell, E. R. R. Chem. Ind. (London) 1965, 157e158.
8. Suzuki, I.; Nakadate, M.; Sueyoshi, S. Tetrahedron Lett. 1968, 9, 1855e1868.
9. Katritzky, A. R.; Randall, W.; Sutton, L. E. J. Chem. Soc. 1957, 1769e1775.
4
6.2.2. Pyrimidine N-oxide 3. Pyrimidine was prepared by a litera-
1
53
ture method. A solution of pyrimidine (1.25 g, 15.6 mmol) and
mCPBA (5.38 g, 50%, 15.6 mmol) in EtOAc (150 mL) was heated
under reflux for 4 h. Evaporation followed by chromatography of
1
1
1
the residue (SiO
4%) as colourless crystals, mp 82e84 C (lit. 95e96 C);
1H, ddd, J 6.6, 4.8, 1.0), 8.22 (1H, dd, J 4.8, 1.5), 8.37 (1H, ddd, J 6.6,
2
,15% MeOH in EtOAc) gave the title product (0.51 g,
2
0. Sosc uꢁ n, H.; Castellano, O.; Berm uꢁ dez, Y.; Toro-Mendoza, C.; Marcanob, A.;
Alvarado, Y. J. Mol. Struct.: THEOCHEM 2002, 592, 19e28.
ꢀ
6
ꢀ
3
(
H
d 7.28
2
1. Sorensen, G. O.; Mahler, L.; Rastrup-Andersen, N. J. Mol. Struct. 1974, 20,
19e126.
1
2
.1, 1.5) and 8.96 (1H, dd, J 2.1, 1.0).
22. Pyckhout, W.; Horemans, N.; Van Alsenoy, C.; Geise, H. J.; Rankin, D. W. H.
J. Mol. Struct. 1987, 156, 315e329.
23. Mootz, D.; Wussow, H.-G. J. Chem. Phys. 1981, 75, 1517e1522.
4. Mata, F.; Quintana, M. J.; Soerensen, G. O. J. Mol. Struct. 1977, 42, 1e5.
5. Blake, A. J.; Rankin, D. W. H. Acta Crystallogr., Sect. C 1991, 47, 1933e1936.
26. Werner, W.; Dreizler, H.; Rudolph, H. D. Z. Naturforsch., A 1967, 22, 531e543.
7. Innes, K. K.; Lucas, R. M. J. Mol. Spectrosc. 1967, 24, 247e250.
8. Almenningen, A.; Bjoernsen, G.; Ottersen, T.; Seip, R.; Strand, T. G. Acta Chem.
Scand. 1977, A31, 63e68.
29. Cradock, S.; Purves, C.; Rankin, D. W. H. J. Mol. Struct. 1990, 220, 193e204.
30. Cradock, S.; Liescheski, P. B.; Rankin, D. W. H.; Robertson, H. E. J. Am. Chem. Soc.
.2.3. Pyrazine N-oxide 4.6 A solution of pyrazine (2.5 g, 31 mmol)
in AcOH (30 mL) was stirred at 70e80 C while a mixture of 30%
6
2
2
ꢀ
H
2
O
2
(3.55 g, 32 mmol) and AcOH (25 mL) was added dropwise
2
2
over 30 min. After the addition, heating was continued for 5 h then
the mixture was evaporated under reduced pressure and the resi-
due was taken up in CH
Na CO and evaporation gave a solid, which was recrystallised from
hexane to give the title product (1.48 g, 50%) as colourless crystals,
2
Cl
2
(125 mL). Drying over anhydrous
2
3
1988, 110, 2758e2763.
3
1. Furberg, S.; Groegaard, J.; Smedsrud, B. Acta Chem. Scand. B 1979, 33, 715e724.
ꢀ
6
ꢀ
mp 113e114 C (lit. 113e114 C);
.3. Crystallography
X-ray diffraction data were recorded on a Rigaku diffractometer
H
d 8.11 (2H, m) and 8.48 (2H, m).
3
3
3
3
3
2. Blackman, G. L.; Brown, R. D.; Burden, F. R. J. Mol. Spectrosc. 1970, 35, 444e454.
3. Khetrapal, C. L.; Patankar, A. V.; Diehl, P. Org. Magn. Reson. 1970, 2, 405e407.
4. Wheatley, P. J. Acta Crystallogr. 1957, 10, 182e187.
5. de With, D.; Harkema, S.; Feil, D. Acta Crystallogr., Sect. B 1976, 32, 3178e3184.
6. Bormans, B. J. M.; De With, G.; Mijlhoff, F. C. J. Mol. Struct. 1977, 42, 121e128.
6
37. Palmer, M. H.; Nelson, A. D. J. Mol. Struct. 2006, 825, 93e100.
38. Palmer, M. H.; Nelson, A. D. J. Mol. Struct. 2006, 782, 94e105.
ꢀ
with MoK
a
radiation (
l
¼0.71073 A). The structures were solved by
3
9. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, 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, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, AM64L-G03RevD.01; Gaussian: Wallingford CT, 2004.
direct methods and refined using full-matrix least-squares
methods.
Crystal data for 2: C
4
H
4
N
2
O, M¼96.09, colourless platelet, crystal
dimensions 0.30ꢁ0.05ꢁ0.05 mm, monoclinic, space group P2
1
/n,
ꢀ
ꢀ
a¼3.819(4),
b¼9.993(5),
c¼22.171(8)
A,
b
¼90.72(2) ,
ꢀ
3
ꢂ3
V¼846.0(10) A , Z¼8, D
c
¼1.509 Mg m , T¼93(2) K, R¼0.2412,
R
W
¼0.5926 for 1142 reflections with I>2
s(I) and 129 variables.
Crystal data for 3: C
4
H
4
N
2
O, M¼96.09, colourless prism, crystal
dimensions 0.10ꢁ0.05ꢁ0.05 mm, monoclinic, space group P2
1
/n,
ꢀ
ꢀ
a¼5.437(3), b¼4.0051(19), c¼19.440(10) A,
b
¼95.443(17) ,
ꢀ
3
ꢂ3
V¼421.4(4) A , Z¼4, D
¼0.1567 for 603 reflections with I>2
Crystal data for 4: C
c
¼1.515 Mg m , T¼93(2) K, R¼0.0619,
4
4
0. Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007e1023.
1. Bader, R. F. W. Atoms in Molecules e a Quantum Theory; Clarendon: Oxford,
R
W
s(I) and 66 variables.
4
H
4
N
2
O, M¼96.09, colourless prism, crystal
1990; p 259.
dimensions 0.20ꢁ0.10ꢁ0.06 mm, monoclinic, space group P2
1
/n,
42. van der Avoird, A. Chem. Commun. 1970, 727e728.
43. Mulliken, R. S. J. Chem. Phys. 1955, 23, 1833e1840.
44. Brown, R. D.; Burden, F. R.; Garland, W. Chem. Phys. Lett. 1970, 7, 461e462.
5. Mosher, H. S.; Turner, L.; Carlsmith, A. Org. Synth. 1953, 33, 79e81.
6. Heiberg, A. B.; Talberg, H. J. Adv. Mol. Relax. Processes 1977, 9, 199e210.
47. Germain, G.; Piret, P.; Van Meerssche, M. Acta Crystallogr. 1963, 16, 109e112.
8. Olsson, L.; Cremer, D. J. Phys. Chem. 1996, 100, 16881e16891.
9. Kubota, T.; Watanabe, H. Bull. Chem. Soc. Jpn. 1963, 36, 1093e1096.
ꢀ
ꢀ
a¼11.416(3), b¼3.6545(10), c¼20.889(6) A,
b
¼91.970(7) ,
ꢀ
3
ꢂ3
V¼871.0(4) A , Z¼8, D
¼0.1619 for 1216 reflections with I>2
Crystallographic data (excluding structure factors) for the
c
¼1.466 Mg m , T¼93(2) K, R¼0.0628,
4
4
R
W
s(I) and 129 variables.
4
4
5
structures in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication nos.
0. Ogata, M.; Watanabe, H.; Tori, K.; Kano, H. Tetrahedron Lett. 1964, 5, 19e24.
51. Biegler Konig, F. W.; Bader, R. F. W.; Tang, T. J. Comput. Chem. 1982, 13, 317e328.
2. Palmer, M. H.; Findlay, R. H.; Gaskell, A. J. J. Chem. Soc., Perkin Trans. 2 1974,
5
4
CCDC 865893 (2), 865894 (3) and 865895 (4).
5
420e428.
References and notes
5
3. Bredereck, H.; Gompper, R.; Morlock, G. Chem. Ber. 1957, 90, 942e952.
4. Copies of the data can be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK, (fax: þ44 (0) 1223 336033 or e-mail:
deposit@ccdc.cam.ac.uk.
5
1
. Ochiai, E. J. Org. Chem. 1953, 18, 534e551.
2. Eichhorn, E. L. Acta Crystallogr. 1956, 9, 787e793.