Mechanisms of reactions conducted on α-amido-α-aminonitrones, determined based on the structures of their crystalline products and DFT calculations

Article abstract of DOI:10.1039/c0nj00193g

The crystal structures of several compounds obtained from recently synthesised α-amido-α-aminonitrones were determined by X-ray diffraction. The compounds belonged to three different classes: 1,2,5-oxadiazin-4-ones, amidines and dibenzo[d,f][1,3]diazepine

Full text of DOI:10.1039/c0nj00193g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Mechanisms of reactions conducted on a-amido-a-aminonitrones,
determined based on the structures of their crystalline products and DFT
calculationsw
Bartosz Trzewik,*a Tomasz Seidler,a Ewa Broc$awikb and Katarzyna Stadnickaa
Received (in Montpellier, France) 12th March 2010, Accepted 4th May 2010
DOI: 10.1039/c0nj00193g
The crystal structures of several compounds obtained from recently synthesised a-amido-a-
aminonitrones were determined by X-ray diffraction. The compounds belonged to three different
classes: 1,2,5-oxadiazin-4-ones, amidines and dibenzo[d,f][1,3]diazepines. In spite of the fact that
they were yielded in reactions of various kinds, all the products contained the same amido-
amidine moiety. We discovered that some geometrical parameters within the moiety had
non-typical values; for example the Csp2–Csp2 single bonds were outstandingly long. A detailed
analysis of the products’ geometry, aided by DFT calculations for selected structures, allowed
us to propose mechanisms of their formation.
hydrogen transfer and prove its intramolecular nature after
Introduction
isotope labelling.
Although the determining of crystal structures is a widely used
As can be seen in the above examples, X-ray or neutron
method in the elucidation of enzymatic reaction paths, there
diffraction can serve as a powerful tool for confirming
have only been a few examples confirming the mechanisms
proposed reaction mechanisms, even in cases where the reactions
of typical organic reactions, either in the solid state or in
are performed in solution. It should also be kept in mind that
solution, by this means. Usually, the structures of crystalline
spectroscopic methods routinely used for investigating organic
products are determined in order to cast light on reaction
reaction products and reaction mechanisms do not always
mechanisms.
provide full and unequivocal information about the three-
In some cases, however, crystal structures have been found
dimensional molecular structures of the products.
for both the substrates and products of the reactions
In this paper, the results of X-ray diffraction measurements,
conveyed, respectively: only in solution,1 in the solid state
performed for crystals of novel a-amido-a-aminonitrones and
various products of their transformations, allowed us to
and in solution,2 and exclusively in the solid state.3 In the case
of cyclotriphosphazene hexachloro derivatives,1 the structural
propose the mechanisms of these reactions. The crystallo-
data for the substrate and products of subsequent nucleophilic
graphic data were combined with appropriate DFT calcula-
substitutions by t-BuNH groups gave rise to conclusions
tions concerning the molecular and electronic structures of
about the fine detail of the mechanism and revealed the
the substrate, the products of reactions under dialkylation
reasons for the suppression of hexasubstitution. Extensive
conditions and the postulated deprotonated intermediate in
studies by Fu et al.2 upon the versatile reactivity of 9,10-
order to support the selected steps of the proposed mechanisms.
dihydro-9,10-ethenoanthracene derivatives allowed a distinction
to be made between different photolytic reaction paths in the
solid state and in solution. Furukawa and Matsumoto3a
succeeded in observing a structural change in a single crystal
during the photopolymerisation of muconates diastereomers.
Last but not least, Hosoya et al.3b observed the topochemical
photoisomerisation of N,N-dibenzyl-1-cyclohexenecarbothio-
amide to an optically active b-thiolactam and were able
to determine—using neutron diffraction—the mechanism of
Results and discussion
Crystal structures of the products
During the research upon using novel a-amido-a-amino-
nitrones, such as 5a,4 for constructing heterocyclic systems,
several new compounds, shown in Scheme 1, were synthesised.5
The compounds discussed in this paper belong to three
different classes: 1,2,5-oxadiazin-4-ones (6a,b), amidines
(7a,b) and dibenzo[d,f][1,3]diazepines (11a,b). Crystal structures
of four of them—6b, 7a, 7b and 11a—are described here for
the first time. Crystal data, experimental conditions and details
of refinement procedures are given in Table 1.
a Faculty of Chemistry, Jagiellonian University, ul. Romana Ingardena 3,
´
30-060 Krakow, Poland. E-mail: trzewik@chemia.uj.edu.pl;
Fax: +48 12 634 05 15; Tel: +48 12 663 22 97
b Institute of Catalysis and Surface Chemistry PAN, ul. Niezapominajek
8, 30-239 Krako´w, Poland. E-mail: broclawi@chemia.uj.edu.pl;
Tel: +48 12 663 20 23
The results of the X-ray crystal structure analysis performed
for the products of the reactions provided unequivocal
information about the molecular structures of 6, 7 and 11.
In reactions with aromatic biphenyl-2,20-diamine, seven-
membered rings of dibenzo[d,f][1,3]diazepines 11 were formed
as the result of disubstitution on the carbon atom adjacent to
w CCDC reference numbers: 750328–750331. Electronic supplementary
information (ESI) available: (1) geometric parameters of weak inter-
actions in the crystalline state; (2) packings of molecules 6b, 7a, 7b and
11a. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c0nj00193g
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
2220 | New J. Chem., 2010, 34, 2220–2228

View Article Online
Scheme 1 Chemical structural formulae of the products obtained from nitrones 5a,b.5 Reagents and conditions: (a) (CH2)nI2, NaH (2 equiv.),
anhyd DMF, ꢂ10 1C to r.t., 1.5 h; (b) biphenyl-2,20-diamine, TsOH (cat.), AcCN, reflux, 2 h. Compound 7a is reported here for the first time.
Table 1 Crystal data, experimental conditions and details of refinement procedures
6b
7a
7b
11a
Chemical formula
M/g molꢂ1
Crystal system, space group
T/K
a/A
b/A
c/A
a (1)
C21H18N4O2
358.39
Triclinic, P-1
293(2)
10.3512(3)
10.5145(3)
10.8469(4)
116.607(2)
115.147(2)
94.549(2)
900.91(6)
2
1.321
0.09
Yellow
0.60 ꢃ 0.38 ꢃ 0.25
Multi-scan
0.949, 0.978
5519, 4077, 3031
0.026
C19H16N4O
316.36
Monoclinic, C2/c
293(2)
26.1851(8)
9.1055(3)
13.7491(6)
90.000
98.881(1)
90.000
3238.9(2)
8
1.298
0.08
Yellow
0.57 ꢃ 0.32 ꢃ 0.05
Multi-scan
0.954, 0.996
5189, 3630, 2814
0.022
C20H18N4O
330.38
Monoclinic, P21/c
293(2)
13.5564(3)
10.1570(3)
13.3227(4)
90.000
109.473(1)
90.000
1729.50(8)
4
1.269
0.08
Yellow
0.60 ꢃ 0.57 ꢃ 0.03
Multi-scan
0.953, 0.998
6633, 3945, 2532
0.030
C19H14N4O
314.34
Triclinic, P-1
293(2)
10.3641(2)
11.7961(3)
14.4961(4)
89.766(1)
69.834(1)
70.166(1)
1552.07(7)
4 (2)
1.345
0.09
Orange
0.30 ꢃ 0.12 ꢃ 0.07
Multi-scan
0.973, 0.994
9212, 6016, 3947
0.033
b (1)
g (1)
V/A3
Z (Z0)
Dx/Mg mꢂ3
m/mmꢂ1 (Mo-Ka)
Crystal colour
Crystal size/mm
Absorption correctiona
Tmin, Tmax
Reflections collectedb, unique, observed [F2 4 2s(F2)]
Rint
ymax (1)
27.5
27.5
27.5
26.0
R1 [F2 4 2s(F2)], wR2 (all F2), S
Reflections/parameters
Weighting schemec a, b
(D/s)max
0.048, 0.135, 1.00
4077/244
0.0609, 0.2162
o0.001
0.045, 0.115, 1.04
3630/223
0.0407, 1.7843
0.001
0.054, 0.125, 0.99
3945/233
0.0442, 0.5708
o0.001
0.061, 0.129, 1.05
6016/445
0.0402, 0.4713
o0.001
Drmax, Drmin/e Aꢂ3
0.18, ꢂ0.18
0.17, ꢂ0.17
0.14, ꢂ0.15
0.19, ꢂ0.20
a
HKL DENZO and Scalepack.6a b COLLECT.6b c Weighting scheme: w = 1/[s2(Fo2) + (aP)2 + bP] where P = [max(Fo2, 0) + 2Fc2]/3.
the nitrone moiety (Scheme 1). The results of the reactions of 5
with dialkylation agents turned to be more stimulating:
compounds 6 and 7, though chemically different, were formed
under similar conditions. The only difference was in the
number of methylene groups between the iodine atoms in
the dialkylation agent molecules (Scheme 1). On the other
hand, there was a pronounced similarity of the chemical
structures of amidines 7 and compounds 11, in spite of the
fact that they were formed in totally different processes.
Molecules of 11 differed from those of 7 in the presence of a
single bond between the phenyl rings (for the overall shapes of
the molecules under discussion, see Fig. 1). This made 7 and 11
interesting for a comparison from a crystallographic point of
view. The most salient details of the geometrical parameters of
the molecular core, which is the same in all the products, are
given in Table 2.
of the C26–C36 single bonds in the molecules of 11. However,
significant differences in the values of the C21–N2–C3–N3
torsion angles were observed.
The C3sp2–C4sp2 single bonds in the molecules of all the
studied compounds were particularly long (in the range
between 1.517 A in 11a and 1.536 A in 6a). We attribute
this phenomenon to the electron withdrawing effect of the
substituents at both ends of the bonds (N2 and N3 atoms at
C3, and O4 and N5 atoms at C4). Concerning the oxadiazine
rings of 6a,b, the effect was enhanced by strain in the six-
membered rings, caused by the relatively short C4–N5, C3–N2
and N2–O1 single bonds (see Table 2). In the molecules of 7a,b
and 11a,b, the elongation of the C3sp2–C4sp2 single bonds
could additionally be attributed to the effect of short intra-
molecular separations of the N2–H2ꢁ ꢁ ꢁO4 and N5–H5ꢁ ꢁ ꢁN3
type. These ‘‘short intramolecular separations’’ can be
considered ‘‘hydrogen-bond-like interactions’’9a (see Table 3),
closing five-membered quasi-rings. This conclusion is based
Surprisingly, the dihedral angles between the phenyl rings in
the molecules of 7 and 11 were not influenced by the presence
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2220–2228 | 2221

View Article Online
Fig. 1 The overall shapes of the molecules under discussion, in comparison to substrate molecules 5a_a and 5a_b, presented as ORTEP37
projections onto the core N-atoms (N2, N3 and N5). Displacement ellipsoids of non-hydrogen atoms are drawn at the 50% probability level. The
numbering schemes are given only for significant atoms. Molecules 11a_a and 11a_b both have two chirality axes, and in crystal structure they
constitute pairs of conformational diastereomers.
Table 2 Comparison of selected geometrical parameters (distances A and angles 1) for the molecular corea of the compounds from this work: 6b,
7a, 7b and 11a, and of those previously published: 5a,4 6a5 and 11b8
5a_a
1.347(3)
5a_b
1.342(3)
6a
6b
7a
7b
11a_a
1.370(3)
11a_b
1.375(3)
11b
1.364(1)
N2–C3
C3–N3
C3–C4
C4–N5
N2–O1
N2–C3–C4
1.387(2)
1.263(2)
1.536(2)
1.369(2)
1.416(2)
1.386(2)
1.264(2)
1.535(2)
1.368(2)
1.420(2)
1.356(2)
1.286(2)
1.526(2)
1.349(2)
1.351(2)
1.285(2)
1.525(2)
1.349(2)
1.315(3)
1.511(3)
1.343(3)
1.336(2)
1.311(3)
1.516(4)
1.343(3)
1.344(2)
1.275(3)
1.517(3)
1.347(3)
1.279(3)
1.517(3)
1.352(3)
1.278(1)
1.524(1)
1.335(1)
122.5(2)
123.0(2)
114.2(1)
114.0(1)
110.5(1)
110.6(2)
113.1(2)
114.9(2)
112.7(1)
N2–C3–N3
O4–C4–N5
C3–C4–N5
117.8(2)
126.0(2)
114.3(2)
13.9(3)
118.2(2)
126.9(2)
113.8(2)
ꢂ10.3(4)
ꢂ57.1(3)
121.4(3)
123.6(2)
ꢂ57.8(3)
171.1(2)
159.7(2)
120.9(1)
124.6(2)
116.1(1)
179.9(2)
ꢂ14.9(2)
170.0(2)
ꢂ13.0(2)
162.1(2)
4.6(3)
120.4(1)
124.2(1)
116.2(1)
ꢂ178.1(1)
ꢂ17.0(2)
168.9(2)
ꢂ13.5(2)
160.6(2)
7.5(2)
133.2(1)
125.2(1)
114.0(1)
ꢂ171.2(1)
177.9(1)
ꢂ1.6(2)
179.7(1)
ꢂ0.8(2)
9.4(3)
132.9(2)
125.2(2)
113.9(2)
ꢂ174.1(2)
177.0(2)
ꢂ2.7(3)
130.2(2)
125.3(2)
113.5(2)
143.1(2)
ꢂ175.0(2)
8.3(3)
128.7(2)
125.0(2)
113.9(2)
130.0(2)
179.1(2)
ꢂ0.7(3)
ꢂ179.1(2)
0.6(3)
ꢂ49.7(3)
ꢂ6.8(4)
149.0(2)
ꢂ29.1(3)
ꢂ35.2(3)
146.7(2)
45.76(8)
31.62(8)
130.1(1)
126.0(1)
113.7(1)
139.6(1)
178.5(1)
ꢂ4.9(2)
173.7(1)
ꢂ2.8(2)
ꢂ44.4(2)
4.9(2)
148.6(1)
ꢂ29.8(2)
ꢂ34.2(2)
147.5(1)
37.71(5)
26.51(5)
C21–N2–C3–C4
N3–C3–C4–O4
N2–C3–C4–O4
N2–C3–C4–N5
N3–C3–C4–N5
C21–N2–C3–N3
N2–C3–N3–C31
C31–C36–C26–C25
C35–C36–C26–C25
C31–C36–C26–C21
C35–C36–C26–C21
+corea, C21^C26
+corea, C31^C36
+C21^C26, C31^C36
61.7(3)
ꢂ115.0(3)
ꢂ118.6(2)
64.7(3)
178,5(2)
ꢂ171.1(2)
ꢂ1.7(2)
5.7(3)
40.7(3)
ꢂ169.2(2)
6.1(4)
6.8(3)
ꢂ166.1(2)
168.9(2)
171.6(1)
8.2(3)
9.8(3)
ꢂ148.5(2)
30.2(3)
33.9(3)
ꢂ147.4(2)
34.23(8)
25.62(12)
32.17(11)
37.21(6)
76.69(6)
82.31(7)
46.79(8)
78.46(5)
75.78(7)
46.99(5)
55.46(6)
33.39(6)
56.40(6)
58.21(7)
23.39(7)
32.56(12) 32.01(6)
a
The best plane defined by atoms of the amido-amidine moiety, identical for all the compounds: N2, C3, N3, C4, O4, N5.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
2222 | New J. Chem., 2010, 34, 2220–2228

View Article Online
Table 3 Geometry of intermolecular hydrogen bonds and hydrogen-
bond-like intramolecular interactions (marked with asterisks)
conformations were preserved due to intramolecular hydrogen-
bond-like interactions of the N–Hꢁ ꢁ ꢁO and N–Hꢁ ꢁ ꢁN type, the
effect of which was clearly pronounced.
D–Hꢁ ꢁ ꢁA
7a
D–H/A Hꢁ ꢁ ꢁA/A Dꢁ ꢁ ꢁA/A D–Hꢁ ꢁ ꢁA/1
The presence of such interactions, even in solution (CDCl3),
could be also related to downfield shifts of the H-atom signals
1
for CONH and NH groups visible in H NMR spectra; the
N2–H2ꢁ ꢁ ꢁO4*
N2–H2ꢁ ꢁ ꢁN51i
N5–H5ꢁ ꢁ ꢁN3*
7b
0.89(1)
0.89(1)
0.87(1)
2.16(2)
2.27(1)
2.19(2)
2.633(2)
3.089(2)
2.645(2)
112(1)
154(1)
112(1)
signals were recorded, respectively, at 10.54 and 8.42 ppm for
7b, and 10.25–10.21 and 7.62–7.61 ppm for 11a,b.5
N2–H2ꢁ ꢁ ꢁO4*
N2–H2ꢁ ꢁ ꢁN51i
N5–H5ꢁ ꢁ ꢁN3*
11a
0.88(1)
0.88(1)
0.88(1)
2.18(2)
2.21(2)
2.17(2)
2.634(2)
3.026(2)
2.646(2)
111(2)
155(2)
113(1)
A further premise for the existence of hydrogen-bond-like
intramolecular interactions of the N–Hꢁ ꢁ ꢁOQC type in
molecules 7b, 11a and 11b were the nCQO values in the IR
spectra acquired in KBr (i.e. in the solid state). They were
equal to 1702, 1687 and 1694 cmꢂ1 for 7b, 11a and 11b,
respectively. These values are remarkably high for amide
CQO bonds.
Compound 11a crystallised with two symmetry-independent
conformational diastereomers in the asymmetric unit, 11a_a
and 11a_b. This conformational diastereoisomerism arose
from the presence of two chirality axes: one of them is defined
by the C26–C36 biphenyl single bond, the other embraces the
C3–C4 single bond (the single C3–C4 bond links the two
distinct parts of the molecules: the amide part, including the
pyridine ring, and the benzodiazepine moiety).
N2a–H2aꢁ ꢁ ꢁO4a*
N5a–H5aꢁ ꢁ ꢁN3a*
N2a–H2aꢁ ꢁ ꢁN51b
N5a–H5aꢁ ꢁ ꢁO4bi
0.89(1)
0.89(1)
0.89(1)
0.89(1)
2.24(2)
2.20(2)
2.40(1)
2.45(2)
2.17(1)
2.30(2)
2.18(2)
2.25(2)
2.699(2)
2.631(3)
3.248(3)
3.151(2)
3.023(3)
2.734(2)
2.634(3)
3.000(2)
112(2)
110(2)
160(2)
137(2)
159(2)
110(2)
111(2)
143(2)
N2b–H2bꢁ ꢁ ꢁN51aii 0.89(1)
N2b–H2bꢁ ꢁ ꢁO4b*
N5b–H5bꢁ ꢁ ꢁN3b*
N5b–H5bꢁ ꢁ ꢁO4a
11b
0.89(1)
0.89(1)
0.89(1)
N2–H2ꢁ ꢁ ꢁO4*
N5–H5ꢁ ꢁ ꢁN3*
N2–H2ꢁ ꢁ ꢁO4i
0.88(1)
0.87(1)
0.88(1)
2.22(1)
2.19(1)
2.31(1)
2.677(1)
2.644(1)
3.108(1)
112(1)
112(1)
151(1)
Symmetry codes for 7a: (i) x, ꢂy, z + 12; for 7b: (i) x, ꢂy + 21, z ꢂ ; for
1
2
11a: (i) x + 1, y, z; (ii) x ꢂ 1, y, z; for 11b: (i) ꢂx, ꢂy + 1, ꢂz + 1.
Considering a molecule with a chirality axis, one has to
determine the configuration around the axis according to
Cahn–Ingold–Prelog rules (see Fig. 2). It should be stressed
that the configuration is irrespective of the view direction. In
the cases of molecules 11a_a, 11a_b and 11b, the configurations
around the chirality axes, embracing the C3–C4 bonds, can be
on solid state and solution state evidence. Other electronic
factors supposedly influencing the length of the C3–C4 bonds
will be discussed in detail in the next section.
While discussing the X-ray diffraction data, it is worth
noting that there are no examples of single Csp2–Csp2 bonds
with the chemical neighbourhood –(OQC)–(CQN)– in the
ITC.9b
related to the signs of the
–C3–C4–
(Fig. 2a) or
–C4–C3– (Fig. 2b) torsion angles. Their signs are positive
in the cases of R configurations, otherwise they are negative
(see the signs of angles N2–C3–C4–O4 and N3–C3–C4–N5 in
Table 2 for comparison). These relations depend on the
relative priorities of the substituents in each studied case.
In respect to the C3–C4 axis, 11a_a has with no doubt an R
configuration [N2–C3–C4–O4 = 8.3(3)1 or N3–C3–C4–N5 =
5.7(3)1]. In molecule 11a_b, the picture is more complex as the
respective angles are very close to 01 and have opposite signs
[N2–C3–C4–O4 = ꢂ0.7(3)1 or N3–C3–C4–N5 = 0.6(3)1]
(Fig. 2c and d). This made us unable to differentiate between
the S and R configuration. Looking from the C4 end of the
axis towards C3, the determined configuration would be S,
while looking from the C3 end of the axis towards C4 one
would determine an R configuration (Fig. 2). However, as
was mentioned earlier, the absolute configuration should be
independent from the direction of observation.
Although there are many examples of structures containing
molecules with the –(OQC)–(CQN)– core in the CSD
(2008),9c only six of them have the extended
–N(–H)–C(QO)–C(QN–)–N(–H)– core that is characteristic
of the molecules under discussion. Two of them, with refcodes
LIVZEW and PILPAC, are 5a and 11b. From the remaining
four examples (DADHAR, FOWPOW, ICOLAO and
PAPKEW), only in one of them (FOWPOW)9d is the discussed
Csp2–Csp2 distance greater than 1.500 A.
The six-membered rings of 6 adopted boat conformations
with pseudo-mirror planes (Cs) through the O1, C4 and O4
atoms. The seven-membered rings of 11 adopted twist-boat
conformations with pseudo-mirror planes (Cs) through the N3
atoms and the centres of the C21–C26 bonds in each molecule.
The values of the C21–N2–C3 angles were particularly high
[129.0(6)1 on average] in the cases of the oxadiazine rings of 6,
due to the previously mentioned ring strain. In the cases of 7, it
was probably due to the resonance effect within the
N3QC3–N2 fragment. Whereas in the cases of 11, both of
these effects are lacking and the values of the angles are typical
[122(1)1 on average]. Other high values of valence angles were
those of N2–C3–N3 [131(2)1 on average]: in 7—due to intra-
molecular hydrogen-bond-like interactions (Table 3); in
11—interactions of this type additionally modified the strain
in the non-planar seven-membered rings.
Regarding the C26–C36 single bond, the absolute
configuration of 11a_a could be designated as
S
[C31–C36–C26–C21 = 33.9(3)1 or C35–C36–C26–C25 =
30.2(3)1] and that of 11a_b as R [C31–C36–C26–C21 =
ꢂ35.2(3)1 or C35–C36–C26–C25 = ꢂ29.1(3)1].
Discussing the molecular geometry of molecules 11a_a and
11a_b, one should keep in mind that the crystal structure is
racemic (space group P-1). The structure of 11b is simpler:
in the asymmetric unit, it had one molecule with an R
configuration on the biphenyl C26–C36 single bond
[C31–C36–C26–C21 = ꢂ34.2(2)1 or C35–C36–C26–C25 =
ꢂ29.8(2)1] and an S configuration on the C3–C4 single bond
The N3–C3–C4–O4 angles were nearly s-cis [ꢂ16(1)1 on
average] in the molecules of 6, and almost ideally s-trans
[180(3)1 on average] in the molecules of 7 and 11. The s-trans
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2220–2228 | 2223

View Article Online
Proposed reaction mechanisms and DFT calculations
The versatile reactivity of nitrones 5 against various dialkylation
reagents encouraged us to propose mechanisms for these
reactions. The mechanisms, based on the results of the X-ray
structure analyses and DFT calculations, are depicted in
Scheme 2 and will be discussed in detail, along with the results
of the calculations.
We did not attempt to model full reaction pathways since
the crucial steps would have required an explicit inclusion of
alkyl diiodides, which would have made the calculation of
critical structures very cumbersome (because of the heavy
iodine atoms and the almost free rotation of CH2 groups).
The geometry of starting nitrones 5 (for example 5a) were
determined both experimentally in the crystalline state and
in silico for single molecules in a vacuum and in a reaction
medium (DMF). The obtained results are compared in
Table 4.
The core part of molecule of 5a (Fig. 1) is stabilised by an
intramolecular N3–H3ꢁ ꢁ ꢁO1 hydrogen bond.
Analysis of the bond lengths, and calculations of NBO10a
and Merz–Singh–Kolman (MK)10b,c atomic populations
(Table 5), prompted us to conclude that only mesomeric
structures with a meaningful negative charge on the O1 atom
gave a significant contribution to the structure of nitrones 5.
This means that, in contrast to (Z)-N-[(phenylamino)methylene]-
aniline oxide,11 only one tautomeric (nitrone) form prevails in
the cases of 5. As can be seen from Fig. 1 on example
compound 5a, nitrones of type 5 have a Z configuration about
the C3QN2 double bond. The values of the torsion angles
O4QC4–C3QN2 in the core moiety of 5a_a and 5a_b were
very close to ꢄ601. This allowed us to describe the mutual
arrangement of the double bonds as gauche.
Fig. 2 The amido-amidine cores of molecules 11a_a and 11a_b in
ORTEP37 projections approximately along their C3–C4 chirality axes.
The cores on the left are seen from the C3 ends of the axes; the cores on
the right are seen from the C4 ends. In each projection, numbers in
circles denote the priority of substituents around the axis according to
Cahn–Ingold–Prelog rules. The substituents closer to the observer are
of higher priority, the other rules remain the same as in the case of a
As we mentioned in the previous section, we noticed key
differences between the conformations of starting nitrones 5
and the resulting 1,2,5-oxadiazin-4-ones 6, amidines 7 and
dibenzo[1,3][d,f]diazepines 11 (Table 2).
We will now discuss the influence of steric and electronic
factors that allowed us obtaining from 5—under similar
conditions—quite different systems, 6 and 7 (Scheme 1).
Let us recall that both 6 and 7 were isolated after reactions
under dialkylation conditions with the use of methylene
diiodide (one carbon atom) and ethylene diiodide (a two
carbon chain), respectively. The reactions required the addition
of a two-fold molar amount of NaH to complete. The first step
was a deprotonation of the amine N3 atom by the first mole of
NaH—the addition of which seemed to be crucial (without
NaH, no reaction was observed)—leading to anion 50.
Concerning substrate molecules 5 (Fig. 1), the existence of
the intramolecular N3–H3ꢁ ꢁ ꢁO1 hydrogen bond apparently
diminished the reactivity of the nitrones in dialkylation reactions,
leading to products 6 and 7. The hydrogen-bond needed to be
weakened (if not broken) by an appropriate solvent of high
polarity to make the reactions possible. Deprotonation was
then facilitated in a highly polar solvent (DMF, e = 38.2). It
should be noted that in less polar solvents, such as THF
(e = 7.5) or toluene (e = 2.4), the reactions failed. The above
observations were strongly supported by the results of DFT
calculations in solution: both the lengthening of the H3ꢁ ꢁ ꢁO1
C
abcd chirality centre. The absolute configurations around the chirality
axes were determined in each case and are given under each projection.
[N2–C3–C4–O4 = ꢂ4.9(2)1 or N3–C3–C4–N5 = ꢂ2.8(2)1],
and, obviously, its centrosymmetric image was present in the
structure (space group P-1).8
Geometrical details of the intermolecular hydrogen bonds
and hydrogen-bond-like intramolecular interactions of
N–Hꢁ ꢁ ꢁO and N–Hꢁ ꢁ ꢁN type in molecules 7a, 7b, 11a and
11b8 are given in Table 3.
Moving to the products of the reactions under dialkylation
conditions—7 and 11—the N2–H2ꢁ ꢁ ꢁO4 and N5–H5ꢁ ꢁ ꢁN3
intramolecular hydrogen-bond-like interactions stabilised
the molecular conformations by forming five-membered
quasi-rings so that the double bonds C4QO4 and C3QN3
were in s-trans positions [the mean value of N3–C3–C4–O4 =
180(3)1].
All the above values of geometrical parameters given for 6a
were confirmed by DFT calculations, performed for single
molecules both in a vacuum and in a DMF environment
(Table 4).
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
2224 | New J. Chem., 2010, 34, 2220–2228

View Article Online
Table 4 Selected geometrical parameters (distances A and angles 1) of molecules of 5a and 6a calculated in vacuo and in a DMF reaction medium,
and of a molecule of deprotonated tentative intermediate form 50a in DMF, compared with those obtained by X-ray diffraction for 5a and 6a
5a_a
(X-ray)
5a_b
(X-ray)
5a
5a
50a
(DFT in vacuo) (DFT in DMF) (DFT in DMF) (X-ray)
6a
6a 6a
(DFT in vacuo) (DFT in DMF)
N2–C3
C3–N3
C3–C4
C4–N5
N2–O1
1.347(3)
1.342(3)
1.374
1.327
1.500
1.374
1.302
1.324
1.361
1.509
1.364
1.320
1.349
1.323
1.526
1.369
1.346
1.387(2)
1.263(2)
1.536(2)
1.369(2)
1.416(2)
1.398
1.273
1.541
1.387
1.418
1.387
1.278
1.541
1.376
1.414
1.315(3)
1.511(3)
1.343(3)
1.336(2)
1.311(3)
1.516(4)
1.343(3)
1.344(2)
N2–C3–C4
N2–C3–N3
O4–C4–N5
C3–C4–N5
122.5(2)
123.0(2)
122.2
120.9
116.6
113.1(2)
114.7
115.0
117.8(2)
126.0(2)
114.3(2)
118.2(2)
126.9(2)
113.8(2)
171.1(2)
159.7(2)
152.2(2)
ꢂ32.0(3)
ꢂ10.3(4)
ꢂ57.1(3)
115.4
124.9
112.6
166.7
140.4
166.2
ꢂ14.4
ꢂ8.7
116.7
125.5
112.8
167.0
150.8
163.9
ꢂ17.3
ꢂ9.0
120.5
124.3
112.8
161.7
157.9
146.4
ꢂ38.6
ꢂ18.5
ꢂ53.7
130.2(2)
124.6(2)
116.1(1)
4.6(3)
168.9(2)
ꢂ77.8(2)
107.9(2)
179.9(2)
170.0(2)
120.3
124.3
116.5
11.9
166.6
124.6
ꢂ63.0
ꢂ10.7
ꢂ17.2
120.3
124.7
116.7
8.4
165.4
126.0
ꢂ61.4
ꢂ13.6
ꢂ17.3
C21–N2–C3–N3 ꢂ169.2(2)
N2–C3–N3–C31 ꢂ166.1(2)
C3–N3–C31–C36 ꢂ147.9(2)
C3–N3–C31–C32
C21–N2–C3–C4
N2–C3–C4–O4
35.0(4)
13.9(3)
61.7(3)
ꢂ43.1
ꢂ51.6
Scheme 2 Proposed mechanisms of forming 1,2,5-oxadiazin-4-ones 6 and amidines 7.
Table 5 Computed values of selected atomic charges in molecule of 5a in vacuo and in solvents of various polarity
Atomic charge/e
In vacuo
In toluene (e = 2.4)
In THF (e = 7.5)
In DMF (e = 38.2)
Atom
NBO
MK
N2
O1
N2
O1
0.0170
ꢂ0.5656
0.1686
ꢂ0.5080
0.0065
ꢂ0.5985
0.1391
ꢂ0.5520
ꢂ0.0032
ꢂ0.6267
0.1208
ꢂ0.5921
ꢂ0.0089
ꢂ0.6418
0.1101
ꢂ0.6149
distance and the lowering of the H3 proton abstraction energy
were positively correlated with the increasing solvent polarity
(Table 6).
Within the ꢂO1–+N2QC3–N3 fragment of the deprotonated
50 molecule, the lone electron pair from the N3 atom formed a
new N3QC3 double bond, and the electron pair from the
previous C3QN2 double bond counterbalanced the positive
charge on the nitrogen N2 atom. The former C3QN2 bond
then became a C3–N2 single bond. Such an electron density
flow is possible only within a planar moiety providing
significant p orbital overlap.
This observation was confirmed by quantum-chemical
calculations. The value of the C3QN3 bond length, equal to
1.324 A in 5a, increased to 1.349 A in molecule 50a, whereas
Table 6 Relative energies of 5a before and after the deprotonation and the H3ꢁ ꢁ ꢁO1 distance in a molecule of 5a in various environments
Environment DE (= E5a ꢂ E5 a)/kcal molꢂ1 Deviation from DE in vacuo/kcal molꢂ1 dH3ꢁ ꢁ ꢁO1/A H3ꢁ ꢁ ꢁO1 elongation from d in vacuo/A
0
In vacuo
Toluene
THF
347
326
310
301
2.032
2.050
2.088
2.104
ꢂ21
ꢂ37
ꢂ46
0.018
0.056
0.072
DMF
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2220–2228 | 2225

View Article Online
Table 7 Bond populations for selected bond natural orbitals in the
5a, 50a and 6 molecules (in vacuo)
Table 8 The differences in energies of molecule 6a and the molecule
of the putative seven-membered compound 60a, compared with the
energy of a –CH2–group in vacuo and in DMF
Bond population
Environment DE (= E6 aꢂ E6a)/kcal molꢂ1 DE ꢂ ECH /kcal molꢂ1
0
2
5a
50a
6
NBO description
In vacuo
DMF
ꢂ39.317
0.000
0.002
p C3–N2
p C3–N3
LP N3
LP O4
s* C3–C4
1.924
—
1.716
1.858
0.074
—
—
ꢂ39.319
1.834
1.812
1.864
0.091
1.842
1.822
1.850
0.104
equal to ECH (Table 8). Apparently, the reason accounting for
2
the reaction running along path B, leading to products 7,
seems to be not thermodynamic, but rather kinetic in nature.
Path B consists in a reduction of monoalkylated nitrone
moiety 500 by the second mole of NaH. It is known that the
N–O nitrone bond is rather weak and is prone to breaking.12
In the reaction of 5 with methyl diiodide, paths A and B
compete to some extent: along with the main products 6, small
amounts of products 7 are also formed.
the length of the C3–N2 bond, equal to 1.361 A in molecule
5a, decreased to 1.323 A in 50a (Table 4). Simple geometrical
considerations were also fully paralleled by NBO analysis, as
summarized in Table 7.
The double bond between C3 and N2, populated by nearly 2
electrons in 5a, looses its p component in 50a and 6. On the
contrary, the double bond between C3 and N3 showed up only
in 50a and 6. An antibonding C3–C4 s* orbital gained some
population that goes in line with bond elongation in the series
5, 50a and 6 (both measured and calculated).
The double bonds in the molecules of products 7 adopt an
s-trans conformation, instead of s-cis as in 6. This can be
reasoned by the forming of several stabilising interactions;
between them, the two of intramolecular hydrogen-bond-like
type play the most important role (Table 3). The formation of
11 from 5 and biphenyl-2,20-diamine is an example of quite
a different reactivity of a-amido-a-aminonitrones 5. The
mechanism seems to be simple and unequivocal here: the
reaction can be treated as an acid-catalysed bi-nucleophilic
attack of the two amine groups of biphenyl-2,20-diamine on
the C3 atom in molecules of nitrones 5. Formally, this reaction
is both a transamination and an imine formation.
An analysis of the corresponding second order energy
contributions indicated, however, that the s* orbital population
was mainly due to the interaction with the O4 lone pair in 5a
(E2 = 20.48 kcal molꢂ1), while it was with the N3 lone pair in
50a and 6 (E2 = 12.68 and 13.93 kcal molꢂ1, respectively).
A somewhat surprisingly low population of the N3 lone pair
in 5a came from its strong interaction with antibonding p*
orbitals of the C3QN2 bond and the phenyl ring—this inter-
action is not critically relevant for geometrical parameters.
In products 7, the mixing with the N3 lone pair should
be ineffective (in analogy to substrates 5). Therefore, the
lengthening of C3–C4 bond in 7a in comparison to 5a (1.526
vs. 1.511–1.516 A) should be attributed to the intramolecular
hydrogen-bond-like interactions of the N–Hꢁ ꢁ ꢁN and
N–Hꢁ ꢁ ꢁO type. This could also be concluded from the down-
field shifts of the signals of the respective H atoms in their
1H NMR spectra, as mentioned earlier.
In this case, the nitrone moiety acts as an active equivalent
of a carbonyl group.13 The s-trans arrangements of the CQO
and CQN bonds in the cores of products 11 is an effect of the
two hydrogen-bonding-like stabilisations (analogous to that in 7),
as well as an attempt to minimize steric hindrance between the
rigid and bulky benzodiazepine moiety and the amide moiety
in molecules 11.
Anions 50 remained nearly planar within the discussed
moiety, as could be seen from the calculated value of ꢂ9.91
for the O1–N2–C3–N3 torsion angle. We postulate that
the nitrone oxygen atom O1 is alkylated first, leading to
monoalkylated intermediates 50 0 (Scheme 2).
Conclusions
The results of the X-ray structure analysis of both a-amido-
a-aminonitrones and products of their transformations were
utilised in three ways.
The deprotonation of the amide N5 nitrogen atom by the
second mole of NaH made the second alkylation possible. In
the case of using methyl diiodide (Scheme 2—path A), the ring
closure indeed occurred, leading to 1,2,5-oxadiazin-4-ones 6.
A question arose as to why, in the case of using ethylene
diiodide, no alkylation took place and the whole reaction was
instead a deoxygenation, affording 7 (Scheme 2—path B).
DFT calculations suggest that the dramatic change in the
last step of the reaction was not the result of the difference in
size of the –CH2– and –CH2–CH2– groups, as one might
assume. In the case of using ethylene diiodide, the second
alkylation could have formed seven-membered rings of 60 (see
Scheme 2). The calculations revealed that the introduction of
an additional –CH2– group into these putative products 60
would have not introduced any additional strain. As can be
seen for examples 6a and 60a, the computed DE is practically
From a synthetic point of view, the unequivocal determination
of the constitution of the reaction products allowed us to show
the flexibility and utility of the starting nitrones in organic
synthesis.
Regarding the geometrical parameters of the compounds in
question, we discovered an outstandingly long, up to 1.536(2) A,
Csp2–Csp2 single bond. The bond joins the amide and the
amidine parts of the molecular core, common to all the
compounds: the substrates and the products. It seems that
this feature, together with the high value of the Nsp3–Csp2–Nsp2
valence angle, 131(2)1 on average in 7 and 11, are connected
with the specific constitution of the core. It is worth noting
that the intramolecular hydrogen-bond-like interactions, of
N–Hꢁ ꢁ ꢁO and N–Hꢁ ꢁ ꢁN type, significantly influenced both the
geometrical parameters of the cores of product molecules and
the mechanisms of their formation.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
2226 | New J. Chem., 2010, 34, 2220–2228

View Article Online
Most importantly, the detailed structural data analysis cast
light on the nature of the studied reactions, especially those
leading, under similar conditions, to the different products
6 and 7.
It should be mentioned here that some of the compounds
reported earlier crystallised with solvent molecules: 0.425
molecules of methanol per one molecule of 5a4 and 0.5
molecules of toluene per one molecule of 11b.8
The DFT calculations performed for isolated molecules of
5a and 6a reproduced correctly, and rationalised the unusual
length of the Csp2–Csp2 single bond. On the other hand, the
modelling performed for molecules 5a, 50a, 6a and 60a in the
reaction medium (DMF) strongly supported the importance of
a polar medium in the proposed mechanism for the formation
of the oxadiazines 6, and rationalised the reasons for yielding
amidines 7 under similar conditions.
Crystal structure determination
Single-crystal diffraction data of the studied compounds—6b,
7a, 7b, and 11a—were collected on a Nonius KappaCCD
diffractometer using Mo-Ka radiation (l = 0.71073 A)
selected by a graphite monochromator. The scan angle and
the distance of the crystal to the detector were fixed at 11 per
frame and 40 mm, respectively. The diffracted intensity was
collected at a temperature of 293 K up to 27.51 of the y angle,
except of 11a, for which there were no significant Bragg
reflections above 26.01. Data reduction, including absorption
corrections by the multi-scan method, was performed using
the HKL DENZO and SCALEPACK programs.6a All X-ray
structures were solved by direct methods using the SIR92
program.14a Anisotropic displacement parameters were refined
against F2 for all non-hydrogen atoms using SHELX97.14b
The H-atoms of the aromatic rings, methyl and methylene
groups were found in the Fourier difference maps; they were
included in the refinement following the appropriate geometrical
constraints (aromatic C–H = 0.93 A, methylene C–H = 0.97 A
and methyl C–H = 0.96 A) and using the riding model with
the isotropic displacement factors Uizo = 1.2Ueq of the parent
atoms. The methyl groups were disordered and described by
two sites rotated by 601 with respect to one from another.
H-atoms attached to N-atoms were located in the Fourier
difference maps and were refined using DFIX restraints with a
target value of the N–H bond length (0.88 A), and with an
assumed estimated standard deviation of 0.01 A. Crystal data,
experimental conditions and details of the refinement procedure
are given in Table 1.
In our opinion, the results presented above point to the
possibility of applying combined structural research (X-ray
structure determination and DFT calculations) to the elucidation
of organic reaction mechanisms, even those carried out in
solution.
Experimental
Melting points were determined on an Electrothermal IA9000
digital melting point apparatus (ꢄ0.5 1C) and were not
corrected. Yields are given for the pure products.
Synthesis5 and crystal preparation
(Z)-2-Anilino-2-[oxido(phenyl)imino]-N-pyridin-2-yl-acetamide
(5a). (2.70 g, 81%): white powder; mp 152.0–153.0 1C.
(E)-2-Phenyl-3-(phenylimino)-5-(pyridin-2-yl)-1,2,5-oxa-diazinan-
4-one (6a). (0.36 g, 70%): yellow prisms; mp 122.5–123.5 1C
(from MeOH).
(E)-5-(4-Methylpyridin-2-yl)-2-phenyl-3-(phenylimino)-1,2,5-
oxadiazinan-4-one (6b). (0.28 g, 52%): yellow prisms; mp
141.0–143.0 1C (from MeOH).
DFT calculations
N-(4-Methylpyridin-2-yl)-2-(phenylamino)-2-(phenylimino)-
ethanamide (7b). (0.17 g, 35%): light yellow needles; mp
163.5–164.5 1C (from MeOH).
The calculations were performed by the use of the Turbomole
quantum chemical package15a for geometry optimisations and
the Gaussian03 suite of programs15b for NBO and MK
analyses. Density functional theory (DFT) with the
B3LYP (Gaussian) functional15c–e and aug-cc-pVDZ15f,g or
6-311++(d,p) basis sets, respectively, were employed. The
COSMO solvent model with default parameters for atomic
radii implemented in Turbomole and dielectric constants (e) of
2.4, 7.5 and 38.2 were used to simulate the solvent effects in
toluene, THF and DMF, respectively. Atomic charges were
assessed from NBO or Merz–Singh–Kolman populations.
The models for 5a and 6a were taken from their crystal
structures.4,5 Geometries were fully optimised in vacuo and
then re-optimised within the COSMO model in solvents
of various polarity: toluene, THF and DMF. For the
deprotonated species 50a, derived from 5a, the geometry
optimisations were performed after proton abstraction.
In order to assess the relative stability of the putative
analogue, 60a, of 6a with a seven-membered ring (a conceivable
reaction product for the dialkylation of 5a with ethylene
diiodide), the geometry of 6a was re-optimized after introducing
an extra –CH2– group into the ring. The energy of the CH2
N-Pyridin-2-yl-5H-dibenzo[d,f][1,3]diazepine-6-carboxamide
(11a). (0.21 g, 65%): orange needles; mp 180.0–182.0 1C.
N-(4-Methylpyridin-2-yl)-5H-dibenzo[d,f][1,3]diazepine-6-
carboxamide (11b). (0.23 g, 69%): orange needles; mp
203.0–204.0 1C.
Details of the experimental procedures, elemental analyses
and spectral data were reported earlier.5
Single crystals of 7a formed simultaneously, in a small
amount, together with the crystallization of 6a from a
methanolic solution. Apart from X-ray experiments, they were
not investigated by other analytical or spectroscopic methods.
Single crystals of 6b suitable for X-ray diffraction were
obtained from an acetonitrile solution, 7a and 7b from
methanol, and 11a from toluene: a hot solution of the
appropriate compound was filtered, left to cool to ambient
temperature and filtered again. The crystals were grown by the
slow evaporation of the resulting saturated solution at room
temperature.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2220–2228 | 2227

View Article Online
moiety was calculated as the average of the energy differences
between consecutive alkanes for the number of carbon atoms
from n = 2 up to n = 6. The relative stability of the putative
seven-membered ring product 60a was calculated as:
10 (a) A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988,
88, 899–926; (b) B. H. Besler, K. M. Merz and P. A. Kollman,
J. Comput. Chem., 1990, 11, 431–439; (c) U. C. Singh and
P. A. Kollman, J. Comput. Chem., 1984, 5, 129–145.
11 A. G. Giumanini, N. Toniutti, G. Verardo and M. Merli, Eur. J.
Org. Chem., 1999, 141–144.
12 (a) D. Geffken, H. Zydowitz and A. Ploetz, Z. Naturforsch., B:
Chem. Sci., 2005, 60, 967–972; (b) Y. Kamitori, Heterocycles, 2000,
53, 107–113; (c) V. Gouverneur and L. Ghosez, Tetrahedron,
DE = Eseven-membered
ꢂ Esix-membered
and compared
ring
ring
with Emethylene group
.
1996,
52,
7585–7598
and
references
cited
therein;
Acknowledgements
(d) M. B. Gravestock, D. W. Knight, K. M. A. Malik and
S. R. Thornton, J. Chem. Soc., Perkin Trans. 1, 2000, 3292–3305
and references cited therein; (e) M. B. Gravestock, D. W. Knight
and S. R. Thornton, J. Chem. Soc., Chem. Commun., 1993,
169–171; (f) S. Nakanishi, J. Nantaku and Y. Otsuji, Chem. Lett.,
1983, 341–342.
The authors thank the X-Ray Laboratory, Faculty of
Chemistry, Jagiellonian University, for making the Nonius
KappaCCD diffractometer available.
13 (a) M. Boruah, D. Konwar and D. Dutta, Indian J. Chem., Sect. B:
Org. Chem. Incl. Med. Chem., 2003, 42, 2112–2114;
(b) A. Freisleben, P. Schieberle and M. Rychlik, J. Agric. Food
Chem., 2002, 50, 4760–4768; (c) P. S. Branco, S. Prabhakar,
A. M. Lobo and D. J. Williams, Tetrahedron, 1992, 48,
6335–6360; (d) K. G. B. Torsell, in Nitrile Oxides, Nitrones, and
Nitronates in Organic Synthesis: Novel Strategies in Synthesis, ed.
H. Feuer, VCH Publishers, Weinheim, 1988, ch. 3, pp. 75–93.
14 (a) A. Altomare, M. C. Burla, M. Camalli, G. Cascarano,
C. Giacovazzo, A. Guagliardi and G. Polidori, J. Appl. Crystallogr.,
1994, 27, 435; (b) G. M. Sheldrick, Acta Crystallogr., Sect. A: Found.
Crystallogr., 2008, 64, 112–122.
References
1 S. W. Bartlett, S. J. Coles, D. B. Davies, M. B. Hursthouse,
˙
˙
¨
H. Ibioglu, A. Kilic¸ , R. A. Shaw and I. Un, Acta Crystallogr.,
Sect. B: Struct. Sci., 2006, 62, 321–329.
2 (a) T. Y. Fu, Z. Liu, G. Olovsson, J. R. Scheffer and J. Trotter,
Acta Crystallogr., Sect. B: Struct. Sci., 1997, 53, 293–299;
(b) T. Y. Fu, J. R. Scheffer and J. Trotter, Acta Crystallogr., Sect.
B: Struct. Sci., 1997, 53, 300–305; (c) T. Y. Fu, G. Olovsson,
J. R. Scheffer and J. Trotter, Acta Crystallogr., Sect. C: Cryst.
Struct. Commun., 1996, 52, 2347–2349; (d) T. Y. Fu, G. Olovsson,
J. R. Scheffer and J. Trotter, Tetrahedron Lett., 1995, 36,
4353–4356.
15 (a) Turbomole v. 6.0, Turbomole GmbH, Karlsruhe Germany,
from
2009
(available
http://www.turbomole.com);
3 (a) D. Furukawa and A. Matsumoto, Macromolecules, 2007, 40,
6048–6056; (b) T. Hosoya, H. Uekusa, Y. Ohashi, T. Ohhara,
I. Tanaka and N. Niimura, Acta Crystallogr., Sect. B: Struct. Sci.,
2006, 62, 153–160.
4 M. Hodorowicz, K. Stadnicka, B. Trzewik and B. Zaleska, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, o599–o600.
(b) M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador,
J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels,
M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision B.02), Gaussian, Inc., Wallingford, CT, 2004;
(c) A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38,
3098–3100; (d) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652;
(e) J. P. Perdew, Phys. Rev. B: Condens. Matter, 1986, 33,
8822–8824; (f) D. Feller, J. Comput. Chem., 1996, 17, 1571–1586;
(g) K. L. Schuchardt, B. T. Didier, T. Elsethagen, L. Sun,
V. Gurumoorthi, J. Chase, J. Li and T. L. Windus, J. Chem. Inf.
Model., 2007, 47, 1045–1052.
5 B. Trzewik, D. Ciez, M. Hodorowicz and K. Stadnicka, Synthesis,
2008, 2977–2985.
˙
6 (a) Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,
307–326; (b) COLLECT: Data Collection Software, Nonius BV,
Delft, The Netherlands, 1998.
7 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
8 M. Hodorowicz, K. Stadnicka, B. Trzewik and B. Zaleska, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, o4115.
9 (a) The term ‘‘hydrogen-bond-like’’ was proposed in:
H. Rostkowska, M. J. Nowak, L. Lapinski and L. Adamowicz,
Phys. Chem. Chem. Phys., 2001, 3, 3012–3017; (b) F. H. Allen,
O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen and
R. Taylor, in International Tables for Crystallography, vol. C:
Mathematical, Physical and Chemical Tables, ed. A. J. C. Wilson,
Kluwer Academic Publishers, Dordrecht, 1995, pp. 685–706;
(c) Cambridge Structural Database, v. 5.30, 2008; F. H. Allen, Acta
Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380–388;
(d) V. L. Rusinov, T. V. Dragunova, V. A. Zyryanov,
G. G. Aleksandrov and O. N. Chupakhin, Khim. Geterotsikl.
Soedin., 1986, 1668.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
2228 | New J. Chem., 2010, 34, 2220–2228