Hydrogen bonding in Schiff bases - NMR, structural and experimental charge density studies

Article abstract of DOI:10.1039/c0dt00298d

A series of sixteen Schiff bases (derivatives of salicylaldehydes and aryl amines) was studied to reveal the influence of substituents and the length of the linker on the properties of the H-bonding formed. In theory, two groups of compounds, derivatives of 2-(2-hydroxybenzylidenoamine)phenol) and 2-hydroxy-N-(2-hydroxybenzylideno)benzylamine, can form different types of H-bonds using one or two hydroxyl groups present in the molecules. Two other groups of compounds, derivatives of 4-(2-hydroxybenzylidenoamine)phenol and N-(2-hydroxybenzyideno)benzylamine, can form only one type of H-bond. It was confirmed by ¹⁵N and ¹³C NMR experiments, that in all cases only traditional, H-bonded six-membered chelate rings were formed. The positions of the hydrogen atom in the rings depend on the substituent and phase. Generally, the OH H-bond form dominates in solution, with exception of the nitro derivatives, where the NH tautomer is present. In the solid state the tautomeric equilibrium is strongly shifted to the NH form. Only for the 5-Br derivative of one compound was the reverse relationship found. According to the results of experimental charge density investigations, two intramolecular H-bonds in the 5-methoxy derivative of 2-hydroxy-N-(2′-hydroxybenzylideno) benzylamine) differ significantly in terms of charge density properties. The intra- and intermolecular H-bonds formed by the deprotonated oxygen atom from 2-OH group are strong, with significant charge density concentration at the bond critical point and a straight, well-defined bond path, whereas the second intramolecular H-bond formed by the oxygen atom from the 2′-OH group is quite weak, with ca. five times smaller charge density concentration than in the previous case and a bent bond path. In terms of energy densities, the latter H-bond appears to be a non-bonding interaction, with total energy density being slightly positive. In terms of source contributions to the density at the H-bond critical point from the atoms involved, the intermolecular, linear H-bond is very strong and charge-assisted in the source function classification, the N(1)-H(1N)...O(1) H-bond is medium-strength, while the third H-bond is extremely weak.

Full text of DOI:10.1039/c0dt00298d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 421
www.rsc.org/dalton
PAPER
Hydrogen bonding in Schiff bases – NMR, structural and experimental
charge density studies†
Anna Makal,a Wojciech Schilf,b Bohdan Kamien´ski,b,c Anna Szady-Chelmieniecka,d Eugeniusz Grechd and
Krzysztof Woz´niak*a
Received 13th April 2010, Accepted 10th October 2010
DOI: 10.1039/c0dt00298d
A series of sixteen Schiff bases (derivatives of salicylaldehydes and aryl amines) was studied to reveal
the influence of substituents and the length of the linker on the properties of the H-bonding formed. In
theory, two groups of compounds, derivatives of 2-(2-hydroxybenzylidenoamine)phenol) and
2-hydroxy-N-(2-hydroxybenzylideno)benzylamine, can form different types of H-bonds using one or
two hydroxyl groups present in the molecules. Two other groups of compounds, derivatives of
4-(2-hydroxybenzylidenoamine)phenol and N-(2-hydroxybenzyideno)benzylamine, can form only one
type of H-bond. It was confirmed by 15N and 13C NMR experiments, that in all cases only traditional,
H-bonded six-membered chelate rings were formed. The positions of the hydrogen atom in the rings
depend on the substituent and phase. Generally, the OH H-bond form dominates in solution, with
exception of the nitro derivatives, where the NH tautomer is present. In the solid state the tautomeric
equilibrium is strongly shifted to the NH form. Only for the 5-Br derivative of one compound was the
reverse relationship found. According to the results of experimental charge density investigations, two
intramolecular H-bonds in the 5-methoxy derivative of 2-hydroxy-N-(2¢-hydroxybenzylideno)-
benzylamine) differ significantly in terms of charge density properties. The intra- and intermolecular
H-bonds formed by the deprotonated oxygen atom from 2-OH group are strong, with significant charge
density concentration at the bond critical point and a straight, well-defined bond path, whereas the
second intramolecular H-bond formed by the oxygen atom from the 2¢-OH group is quite weak, with
ca. five times smaller charge density concentration than in the previous case and a bent bond path. In
terms of energy densities, the latter H-bond appears to be a non-bonding interaction, with total energy
density being slightly positive. In terms of source contributions to the density at the H-bond critical
point from the atoms involved, the intermolecular, linear H-bond is very strong and charge-assisted in
the source function classification, the N(1)–H(1N) ◊ ◊ ◊ O(1) H-bond is medium-strength, while the third
H-bond is extremely weak.
commonly form very stable six-membered chelate rings with
Introduction
strong hydrogen bonds. The structure and properties of such
Schiff bases are usually obtained by condensation of substituted
bonds have been investigated in the solution and solid state
salicylaldehydes with both aliphatic and aromatic amines. They
using many physicochemical methods, including heteronuclear
NMR in both phases, FT-IR spectroscopy, X-ray and neutron
aDepartment of Chemistry, University of Warsaw, Pasteura 1, 02-093,
Warszawa, Poland. E-mail: kwozniak@chem.uw.edu.pl; Fax: +48 22
8222892
diffraction.1–32 In the present study, the structures of four groups of
compounds are discussed, obtained by condensation of different
salicylaldehydes with 2- and 4-hydroxyaniline and benzylamine
and 2-hydroxybenzylamine (Scheme 1).
bInstitute of Organic Chemistry, Polish Academy of Science, Kasprzaka
44/52, 01-224, Warszawa, Poland
cInstitute of Physical Chemistry, Polish Academy of Science, Kasprzaka
44/52, 01-224, Warszawa, Poland
The compounds from groups A and A¢ can theoretically form
not only conventional single six-membered chelate rings but also
more complicated systems (Scheme 2). The two remaining groups
of compounds, B and C, contain only one hydroxyl group in a
favored position in a given molecule. Thus, they can form only
conventional hydrogen bonds. Another three compounds, which
do not form intramolecular hydrogen bonds, were selected as
model reference moieties (Scheme 3). Their spectral parameters
dDepartment of Inorganic and Analytical Chemistry, West Pomeranian
University of Technology, al. Piasto´w 42, 71-065, Szczecin, Poland
† Electronic supplementary information (ESI) available: Technical de-
scription of the Experimental section, including details of spectroscopic
and crystallographic procedures, crystal data, full table of critical point
parameters, residual electron density map, structural and multipole
parameters and description of intermolecular interactions. For ESI see
DOI: 10.1039/c0dt00298d
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 421–430 | 421
©

View Article Online
Scheme 1 Classes of Schiff bases studied – the X derivatives of: (A) N-(5-X-salicylidene)-o-hydroxyaniline, (B) N-(5-X-salicylidene)-p-hydroxyaniline,
(A¢) N-(5-X-salicylidene)-o-hydroxybenzylamine, (C) N-(5-X-salicylidene)benzylamine.
Scheme 2 Possible hydrogen bonding in the studied compounds.
are shown in Scheme 3. The reference compounds from the family
B contain a short –CH N– linker joining the two aromatic
rings, an OH group at the para position (unable to form an
intramolecular H-bond), and a substituent X (H, NO2, OMe
or Cl).
The compounds from family C contain a longer –CH N–CH2–
linker and they do not have the second OH group. The solid-
state structure of compound I has already been investigated by
X-ray diffraction methods.27 Its asymmetric unit consists of two
crystallographically independent, but nearly identical, molecules
linked by an O–H ◊ ◊ ◊ O hydrogen bond. In both molecules
typical six-membered hydrogen-bonded rings are found. The two
molecules make a dihedral angle of 73.07(4)◦ in the crystal
lattice. They are both roughly planar; however, the benzene ring
(C1¢, ◊ ◊ ◊ ,C6¢) in one molecule is slightly more twisted with respect
to the (C1, ◊ ◊ ◊ , C6, N1, C7) moiety than for the other moiety,
as indicated by the values of the dihedral angle between these
planes 12.3(1)◦ and 9.7(3)◦, for the first and second molecule,
respectively.
The aim of the present study is to find out which schemes of
hydrogen bonds are present in the groups A and A¢ both in DMSO
solution and in the solid state, and characterise these interactions
at the level of experimental charge densities. The X-ray structures
(Fig. 1) and – in one case – high-resolution charge density studies
have been carried out to verify the results of NMR measurements
for three compounds VII, XV and XVI.
422 | Dalton Trans., 2011, 40, 421–430
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme
3
Model compounds without hydrogen bonding used as
reference systems. All NMR data taken for chloroform solution:
(A) N-benzvlideno-p-hvdroxyaniline. (B) N-benzylidenobenzvlamine. (C)
N-benzylideno-o-hydroxybenzylamine.
Results and discussion
NMR
The results of NMR measurements (chemical shifts) are collected
in Table 1. From previous experience, we know that the hydrogen
atom position in an intramolecular hydrogen bridge can be
determined by analysis of the nitrogen chemical shifts of the imine
atom (quantitative estimation) or carbon chemical shifts of the
carbon atoms bonded to the formal hydroxyl group (qualitative
estimation). The general rule is that a transfer of the hydrogen atom
from the oxygen atom to the nitrogen atom generates a significant
upfield shift of the imine nitrogen signal and a simultaneous
downfield shift of the signal of the C–OH atom. The chemical
shifts of the remaining carbon atoms in the molecule usually
do not supply reliable structural information. This is due to
an overlap of several effects influencing this parameter. These
are mostly the effects created by different substituents in the
aromatic ring. Usually, these effects are very difficult to estimate
in three-substituted aromatic rings. Such effects can have similar
consequences as those created by different proton positions in the
H-bridge.
The proton chemical shift of the OH group can only inform
us about the presence of hydrogen bonding (a large down-
field effect for the hydrogen-bonded structure) but cannot be
used to decide where the proton is located in the H-bridge.
Generally, signals from protons located in the middle of the
bridge (symmetrical H-bonds) have very large downfield shift
compared to signals from the unsymmetrical structures. The
following compounds can be used as model reference structures
without hydrogen bonding (see Scheme 3): (A) N-benzylideno-
p-hydroxyaniline, (B) N-benzylideno-benzylamine, and (C) N-
benzylideno-o-hydroxybenzylamine.
Fig. 1 ORTEP representations of the following compounds: (a) VII, (b)
XV and (c) XVI. Thermal displacement ellipsoids are presented at the 50%
probability level at 100 K. The most important bond lengths for compound
˚
VII, XV and XVI (distances in A) are listed below: VII. O(1)–C(1)1.310(2);
O(3)–C(10) 1.348(2); O(3)–H(3O) 0.93(2); N(1)–C(8) 1.300(2); N(1)–C(9)
1.411(2); N(1)–H(1N) 0.92(2); C(1)–C(2) 1.407(2); C(1)–C(6) 1.432(2);
C(2)–C(3) 1.377(2); C(3)–C(4) 1.403(2); C(4)–C(5) 1.376(2); C(5)–C(6)
1.410(2); C(6)–C(8) 1.423(2); C(9)–C(14) 1.391(2); C(9)–C(10) 1.404(2);
C(10)–C(11) 1.393(2); C(11)–C(12) 1.387(2); C(12)–C(13) 1.388(2);
C(13)–C(14) 1.384(2); XV. O(1)–C(2)1.2979(3); O(2)–C(5) 1.3649(3);
O(2)–C(7) 1.4213(4); O(3)–C(11) 1.3440(3); O(3)–H(3O) 0.8823(3);
N(1)–C(8) 1.3025(3); N(1)–C(9) 1.4631(3); N(1)–H(1N) 0.8784(2);
C(1)–C(2) 1.4284(3); C(1)–C(6) 1.4204(3); C(1)–C(8) 1.4190(3); C(2)–C(3)
1.4264(3); C(3)–C(4)1.3759(3); C(4)–C(5) 1.4129(3); C(5)–C(6) 1.3751(3);
C(9)–C(10) 1.5057(3); C(10)–C(11) 1.4057(3); C(10)–C(15) 1.3940(3);
C(11)–C(12) 1.3987(3); C(12)–C(13) 1.3911(3); C(13)–C(14) 1.3940(4);
C(14)–C(15) 1.3980(4); XVI. Br(1)–C(4) 1.901(2); O(1)–C(1) 1.289(2);
O(2)–C(10) 1.355(2); O(2)–H(2O) 0.86(3); N(1)–C(7) 1.284(3); N(1)–C(8)
1.470(2); N(1)–H(1 N) 0.81(2); C(1)–C(2) 1.420(3); C(1)–C(6) 1.439(2);
C(2)–C(3) 1.373(3); C(3)–C(4) 1.407(3); C(4)–C(5) 1.365(3); C(5)–C(6)
1.407(3); C(6)–C(7) 1.422(3); C(8)–C(9) 1.502(3); C(9)–C(14) 1.389(3);
C(9)–C(10) 1.401(3); C(10)–C(11) 1.391(3); C(11)–C(12) 1.388(3);
C(12)–C(13) 1.386(3); C(13)–C(14) 1.390(3).
The third model compound can theoretically form intramolec-
ular hydrogen bonds using the OH group at the position 2¢, but
examination of the nitrogen chemical shift of the imine atom
shows that this does not take place. The dN value of -51.25 ppm is
typical for an imine without hydrogen bonding. This means that
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 421–430 | 423
©

View Article Online
Table 1 Proton, carbon and nitrogen chemical shifts of investigated compounds. For the definition of compounds, see Scheme 1.
Compound
DMSO solution
CPMAS
Linker
–CH N–
OH, X
4¢, H
d2OH d2¢OH d4¢OH dN
dC2
dC2¢
dC4¢
dC7
dN
dC2
dC2¢
dC4¢
156.0
dC7
159.3,
156.0
Ib
13.4
9.6 -87.1 160.6
157.4 160.6 -99.1,a -133.8 161.0,
163.1
II
–CH N–
–CH N–
–CH N–
–CH N–
–CH N–
–CH N–
–CH N–
–CH N–CH2– —, H
–CH N–CH2– —, NO2
–CH N–CH2– —, OCH3 12.8
–CH N–CH2– —, Br
–CH N–CH2– 2¢, H
–CH N–CH2– 2¢, NO2
–CH N–CH2– 2¢, OCH3 13.0 9.58
–CH N–CH2– 2¢, Br 13.8 9.6
4¢, NO2
15.0
4¢, OCH 12.7
9.8 -99.9 168.2
9.62 -84.4 154.6
8.64 -84.9 159.7
158.2 158.9
III
157.5 160.3 -188.4
157.3 159.2 -166.3
162.1 -215.5
159.7 -213.9
161.9 -200.4
162.4 -192.3
166.9
167.8
167.4
175.7 150.0
178.9 149.4
167.8 150.8
174.2 148.6
157.8
157.4 ov. 157.4 ov
155.4
IVd
V
4¢, Cl
13.5
13.7 9.7
15.7 10.3
2¢, H
-94.6 161.2 151.6
-145.1 173.0 150.9
-91.3 155.1 151.7
-97.2 159.1 147.7
-84.1 161.0
153.8
156.4
154.4
160.2
VI
2¢, NO2
VIIc
VIIIe,f
IX
2¢, OCH3 13.0 9.6
2¢, Cl
12.2 5.8
13.4
14.5
X
XI
XII
XIII
XIV
XVc
XVIc
-161.3 175.8
167.3 -214.4
166.3 -84.84
165.7 -76.3
166.6 -203.4
167.3 -198.4
165.8 -216.7
165.3 -199.0
179.7
154.7
159.2
174.3 156.8
179.6 156.8
170.7 159.8
173.6 156.6
166.2
165.0
163.7
165.7
164.4
163.0
164.2
-80.3 154.8
13.5
13.7 9.6
15.5 10.0
-82.8 160.5
-87.6 161.4 155.7
-182.5 178.0 156.1
-82.5 154.9 155.7
-89.1 161.2 155.7
a Intensity of signals close to 1:1; ov: overlapped signals; –CH N–: Schiff bases obtained from the aniline derivatives; –CH N–CH2–: Schiff bases
obtained from benzylamines. b X-ray from literature. c X-ray from current study. d NMR in acetone. e NMR in CDCl3. f 5¢-CH3.
the acidity of the OH group at the position 2¢ is not high enough
to form such a bond. Similar conclusion can be obtained when
one analyses the OH group proton chemical shifts. This chemical
shift is typical for phenols without intramolecular hydrogen bond
similarly to the first model compound. Comparing the nitrogen
chemical shifts for all model compounds mentioned above, one
can say that a typical value for this parameter for a free imine –
i.e. without hydrogen bonding – is close to -50 ppm.
The structural analysis of the investigated compounds was
started using derivatives I–IV. The proton chemical shifts of the
OH groups (from 12 to 15 ppm) and the nitrogen chemical shifts
close to -90 ppm show that in all these compounds intramolecular
hydrogen bonding exists. In the CPMAS experiments for I, two
sets of signals in 13C and 15N spectra were found. This means
that the mixture of two different forms of I exists in this phase,
which is in agreement with the X-ray data. On the base of the
dN values in the DMSO solution, one can determine that the
proton transfer process is the most advanced (with the largest
upfield shift of the imine signal) in the nitro derivative. In the
solid state, this process is even more advanced. For the 5-methoxy
derivative, the upfield shift of dN between the solution and solid
state is almost 100 ppm. This is rather an unexpected effect, since
for the aliphatic derivative XI, the proton position in the bridge
(and consequently the nitrogen chemical shifts) are almost the
same in the both phases. Similarly, 5-bromo aliphatic derivative
XII exhibits weaker reaction on transfer from the solution to the
solid state (about 10 ppm upfield effect) vs. an effect of about 80
ppm compound IV.
data it is evident that in the 5-nitro derivative the proton transfer
process is the most advanced (dN = -145.1 ppm vs. about -90 ppm
for the other compounds from this group). The carbon chemical
shift of the C2 atom confirms this conclusion. This parameter is
different for each compound from this group and generally fulfils
the condition that the upfield shift of nitrogen signal of imine is
connected with a downfield shift of the carbon atom connected
with the OH group involved in the H-bond formation. The value
of dC2 = 173.0 ppm for VI is clearly different from the equivalent
values for the other compounds, confirming that the C2 atom
is involved in hydrogen-bond formation. In contrast to this, the
dC2¢ values for V–VII are almost the same. This means that the
OH group at this atom does not form H-bonds. The proton and
carbon chemical shifts of VIII have values that are slightly different
compared with V–VII. This is probably because of the different
solvent used for VII (chloroform instead of DMSO) and because
of the presence of the methyl group at the aniline ring. This can
affect the carbon chemical shifts of the atoms from this ring. In
the solid state, V–VIII exist in the NH form with almost the same
proton position (nitrogen chemical shift very close to -200 ppm in
all cases). The dC2¢ values in the solid state are also very close one
to another and to the values found in solution. This means that
also in the solid state, the H-bonds in the N ◊ ◊ ◊ H form exist only
at the 2-OH group. If the OH group from position 2¢ participates
in H-bond formation with proton transfer, a downfield shift of the
C2¢ signal should be expected.
The third group of compounds, IX–XII, can form only one
kind of H-bond because they do not have a second hydroxyl
group at a suitable 2¢ position. One can state – on the basis
of nitrogen chemical shifts of the imine atoms – that in the
case of the compound IX, XI and XII, some very similar
hydrogen bonds with hydrogen atoms located close to the oxygen
atom do exist. The 5-nitro derivative, as in the previous groups
of compounds, forms a different kind of H-bond in DMSO
solution. In this case the bridged hydrogen atom is shifted to
the nitrogen site. The analysis of C2 chemical shifts leads to the
The second group of compounds, V–VIII, is even more inter-
esting because they can form more complicated hydrogen-bond
networks (Scheme 2). In this case, on the basis of proton chemical
shifts of the OH groups and the nitrogen chemical shifts, one can
assume that some intramolecular hydrogen bonds are present. The
proton chemical shift values suggest that only the OH groups at
the position 2 are involved in this process. The d2¢OH values are in
a typical range for a free hydroxyl group. From nitrogen NMR
424 | Dalton Trans., 2011, 40, 421–430
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
same conclusion. In the solid state only three compounds were
investigated and similar effects were found. The nitro compound
in the solid state exists as the NH form, while the 5-MeO and 5-Br
derivatives retain the OH form, resulting in relatively weak
hydrogen bonds.
The crystal structures of VII, XV and XVI
All analyzed Schiff base derivatives crystallize in centrosymmetric
space groups (Table 1S†) in the monoclinic system. In each
case one molecule occupies a general position in crystallographic
asymmetric unit. The ORTEP representations of the compounds
at 100 K, together with numbering scheme, are presented in Fig. 1.
The symmetry of the crystals does not depend on the temperature,
and the general structural motifs, in particular the geometry of
hydrogen bonds network, are similar at room temperature and
at 100 K. More detailed structural analysis is therefore based
on the low temperature datasets, whereas parameters for both
temperatures are reported in Table 1S and in deposited cif files. In
all compounds investigated by X-ray diffraction an intramolecular
asymmetric hydrogen bond is formed, in which the H atom from
the C2 hydroxyl group is transferred to the N1 imine nitrogen
atom. There are no indications of the alternative proton locations,
either at 100 K or at room temperature.
The most interesting is the fourth group of compounds: XIII
to XVI. In this case two different six-membered chelate rings
with intramolecular H-bonds can exist (Scheme 2, the upper
trace). On the basis of nitrogen chemical shifts analysis (solution
NMR spectra) one can say that the compounds XIII, XV and
XVI exist as the OH form with some asymmetric intramolecular
hydrogen bonding, while the 5-nitro compound exists as the
hydrogen bonded NH form. From the nitrogen chemical shifts it
cannot be decided which OH group participates in this hydrogen-
bond formation. To solve this problem it is necessary to analyze
carbon chemical shifts of C2 and C2¢ atoms. The dC2 values are
in good agreement with those in the previously discussed systems,
meaning: the upfield shift of the nitrogen signal is connected with
the downfield shift of the C2 signal. In contrast to this, dC2¢ values
are practically unchanged for all studied compounds and are very
close to the dC2¢ value measured for the third model compound
(dC2¢ = 155.6 ppm) which does not form any intramolecular
hydrogen bonding. A comparison of those two effects leads to the
conclusion that only the OH groups at the position 2 are involved
in hydrogen bond formation with proton transfer, otherwise some
downfield effect on dC2¢ would be observed. In the solid state all
compounds XIII–XVI exist in the NH form, with only very small
differences at the hydrogen atom positions. The downfield shift of
the C2 signals confirms this statement.
The case of the 5-OMe derivative XV in the solid state is a little
surprising, as the largest upfield shift of nitrogen signal observed
in this study is not accompanied by an equally large downfield shift
of the C2 signal. Nevertheless, the carbon signal is still in a range
indicating proton transfer from the 2-OH group to nitrogen. This
could be explained by intermolecular hydrogen bonding formation
in the solid state. This explanation can be verified by analysis of
the X-ray data. In general, there are mainly isotropic interactions
with the solvent molecules in the solution. This is the reason
why the shielding of protons is larger in the solution than in
the solid state, whereas in the solid state, there are some strong
anisotropic intermolecular interactions with the closest molecules
in the crystal lattice.
The molecule of VII is generally flat. The imine linker of two
phenyl rings has a flat geometry and is coplanar with the C1–C6
ring, while the ortho-hydroxy- substituted C9–C14 ring is rotated
with respect to the Schiff-base plane by about 16◦ (see Table 4S†).
The methoxy substituent is rotated slightly away from the C1–
C6 ring by about 6◦. Bonds to C1 carbon, connected to strong
˚
H-bond donor O1 atom, are elongated by 0.3 A with respect
to the analogous bonds in the rest of the ring, while the OMe
substitution does not introduce a similar bias. The C1–C6 ring –
as a whole – presents significant bond alternation. The C9–C14
ring, in contrast, has nearly equal C–C distances. Illustration of
H-bonded dimers formed in the crystal structures of VII at 100
K, XV and XVI is shown in Fig. 2, the packing and the H-bond
networks are shown in Fig. 2S, whereas the details of H-bonds are
given in Table 2.
Compounds XV and XVI represent the A¢ family of Schiff-base
derivatives. An additional methylene group divides the Schiff base
fragment from the phenyl substituent, and introduces a bend into
the molecules of XV and XVI. A measure of this bend may be
the value of torsion around the N1–C9 bond (XV) or N1–C8
bond (XVI), reported in the ESI. In the case of the bromine-
substituted compound XVI, the rotation around the N1–C8 bond
is 10◦ larger than the analogous rotation in XV. In both cases, the
imine fragment is strictly coplanar with the C1–C6 ring.
Fig. 2 Illustration of H-bonded dimers in the crystal structures of: (a) VII at 100 K, (b) XV and (c) XVI.
The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 421–430 | 425
This journal is
©

View Article Online
Table 2 H-bo◦nd geometries for compounds VII, XV and XVI (distances
˚
in A, angles in
)
D–H ◊ ◊ ◊ A
d(D–H) d(H ◊ ◊ ◊ A) d(D ◊ ◊ ◊ A) ∠(DHA)
VII
N(1)–H(1 N) ◊ ◊ ◊ O(1)
0.92(2)
0.92(2)
1.77(2)
2.30(2)
1.63(2)
2.551(1)
2.665(1)
2.558(1)
140(2)
102(1)
179(2)
N(1)–H(1 N) ◊ ◊ ◊ O(3)
O(3)–H(3O) ◊ ◊ ◊ O(1)#1 0.93(2)
XV at 100 K
N(1)–H(1 N) ◊ ◊ ◊ O(1)
N(1)–H(1 N) ◊ ◊ ◊ O(3)
0.87(2)
0.87(2)
1.90(2)
2.45(2)
1.57(2)
2.599(1)
2.886(1)
2.532(1)
137(1)
112(1)
170(2)
O(3)–H(3O) ◊ ◊ ◊ O(1)#2 0.97(2)
XVI at 100 K
N(1)–H(1 N) ◊ ◊ ◊ O(1)
N(1)–H(1 N) ◊ ◊ ◊ O(2)
0.81(2)
0.81(2)
1.89(2)
2.55(3)
1.74(3)
2.589(2)
2.927(2)
2.582(2)
143(2)
110(2)
168(3)
O(2)–H(2O) ◊ ◊ ◊ O(1)#2 0.86(3)
1
#1: -x + 1, y - /2, -z + 5/2, #2: -x + 1,-y + 2,-z.
Significant bond alternation is present in the C1–C6 ring of
XV. Elongation of the bonds connected to the oxygen-substituted
carbon is also noticeable. The methoxy substituent is out of the
C1–C6 ring plane by about 5◦. In the case of XVI, the C1–C6 ring
does not show strict bond alternation, although elongation of the
C1-connected bonds is visible. Differences in bond lengths are less
pronounced at room temperature.
A common packing motif can be observed for all compounds.
For details of the H-bond networks for VII, XV and XVI classified
according to the Etter terminology33 see the ESI.
Hydrogen bonding in VII, XV and XVI
A similar behaviour of the compounds from A and A¢ group in
terms of chemical shifts for atoms potentially involved in hydrogen
bonding is in agreement with the structural analysis. Compounds
VII, XV and XVI present a common structural motif, which is a
specific set of intramolecular hydrogen bonds. In terms of Etter
classification, this implies a S21(9) motif for VII, and S21(10)
for XV and XVI. Apart from additional methylene group in the
compounds of A¢ series, the motif is the same, consisting of one
proton donor – the N1 nitrogen – and two proton acceptors
– the oxygen atoms O1 and O3 or O1 and O2 for XV and
XVI, respectively – resulting in the formation of at least one six-
membered ring. While the intramolecular N1–H1O ◊ ◊ ◊ O1 presents
in each case features of the typical strong hydrogen bond (Fig. 3),
the existence of the actual N1–H1O ◊ ◊ ◊ O3 (or O2) hydrogen bond
may be disputable, due to long donor–acceptor distance and nearly
straight D–H ◊ ◊ ◊ A angle. However, several structural features of
the compounds presented here suggest that the interaction takes
place. These are: the C–N–C angle values are significantly larger
than 120◦, indicating the bending of the structure to place the
second hydroxyl group closer to the hydrogen donor; the geometry
around the O3 or O2 atom, suggesting that the oxygen is bent
still closer to the NH group; and the values of angles H–N–
C, which are similar, suggesting that the hydrogen is not pulled
strongly towards one of the oxygens, but rather is located in
between them. Comparison of the most important geometrical
features of the intramolecular hydrogen bond motif is presented in
Fig. 3.
Fig. 3 Representation of the intramolecular hydrogen-bonding system in
compounds VII, XV and XVI with selected values of (a) bond lengths and
(b) valence angles.
the solid state. While located on the oxygen atom, the proton
cannot participate in an additional hydrogen bond with the second
hydroxyl group. It is also noteworthy that the hydroxyl group
playing the role of hydrogen H1N acceptor, is at the same time
an effective donor in intermolecular hydrogen bonds, which are
important for the crystal structure stabilization, while they do not
exist in the solution.
Experimental charge density analysis of XV
Topological analysis of the final model of experimental charge
density for XV complex was performed. Bond critical points and
bond paths have been associated with all covalent bonds and
expected hydrogen bonds, both intramolecular and intermolec-
ular. In particular, bond path and bond critical point indicate
an additional intramolecular N(1)–H(1N) ◊ ◊ ◊ O(3) hydrogen bond
between the protonated nitrogen atom and the oxygen from 2¢-
OH group. This hydrogen bond could not be detected by means
of NMR study.
The formation of this double hydrogen bond is the main cause
of the preference for proton transfer from the hydroxyl group
on the nitrogen, taking place irrespective of the substituents in
A number of intermolecular short contacts can also be as-
sociated with BCP-s and bond path existence, reflecting the
weak C–H ◊ ◊ ◊ O interactions of the methoxymethyl group with
426 | Dalton Trans., 2011, 40, 421–430
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 3 Most important parameters of experimental charge density at bond critical points in compound XV. Symmetry operation X3: 1 - x, 2 - y, -z
G(rBCP)/r
H(rBCP)/r
|V(rBCP)|/
-3
2
-5
-3
-3
-3
(rBCP) [H e-1]
(rBCP) [H e-1] G(rBCP
)
˚
˚
Bond
r(rBCP) [e A ] — (rBCP) [e A ] G(rBCP) [H a0
]
V(rBCP) [H a0
]
H(rBCP) [H a0
]
O(1) ◊ ◊ ◊ H(1N)
0.26(1)
3.33(1)
2.33(7)
1.34(2)
0.036
0.05
-0.037
-0.075
-0.007
-1.075
-0.83
-0.639
-0.871
-0.995
-1.205
-1.138
-0.598
-0.001
-0.025
0.003
0.92
0.72
1.13
1.03
0.97
0.87
0.96
0.64
0.96
1.05
0.90
-0.03
-0.36
0.31
1.03
1.50
0.64
2.71
2.61
2.62
2.68
4.04
3.00
2.72
2.49
H(3O) ◊ ◊ ◊ O(1)_X3 0.47(4)
O(3) ◊ ◊ ◊ H(1N)
O(1)–C(2)
O(2)–C(5)
0.066(5)
2.60(2)
2.21(1)
1.89(2)
2.29(1)
2.59(1)
2.82(5)
2.69(2)
1.81(1)
0.011
0.396
0.318
0.244
0.325
0.246
0.402
0.419
0.24
-27.15(8)
-18.71(6)
-14.59(6)
-21.38(7)
-48.55(1)
-38.6(4)
-28.87(7)
-11.35(5)
-0.678
-0.512
-0.395
-0.547
-0.749
-0.802
-0.719
-0.358
-1.76
-1.56
-1.41
-1.62
-1.95
-1.92
-1.81
-1.34
O(2)–C(7)
O(3)–C(11)
O(3)–H(3O)
N(1)–H(1N)
N(1)–C(8)
N(1)–C(9)
O(1) and O(3) atoms from the symmetry-related molecules and
C–H ◊ ◊ ◊ p interactions. The summary of topological properties for
the most interesting interactions is presented in Table 3 (Table
2S† contains full topological data for all bonds). Most of the
topological parameters of electron density is close to the values
obtained for similar compounds.34
The intramolecular hydrogen bonds differ significantly in terms
of charge density properties. The O(1) ◊ ◊ ◊ H(1N)–N(1) bond is
strong, with significant charge density concentration at the bond
critical point and a straight, well-defined bond path. The same
is true in the case of intermolecular O(1) ◊ ◊ ◊ H(3O)–O(3) bond.
In contrast, the second intramolecular hydrogen bond is weak,
showing charge density concentration over five times less than
the other H-bonds and a bent bond path, in agreement with
the strained geometry. Nevertheless, the topology of the charge
density suggests its existence. Moreover, ring critical points of
significant charge density concentrations have been found not
only for the benzene rings, but also for the six-membered rings
associated with O(1) ◊ ◊ ◊ H(1N)–N(1) and O(3) ◊ ◊ ◊ H(1N)–N(1)
hydrogen bond formation, which according to Koch and Popelier
35 criteria makes the topology of the molecule reliable and confirms
hydrogen bonding effects (a list of ring critical points reported in
the ESI†). In terms of energy densities, the weak O(3) ◊ ◊ ◊ H(1N)–
N(1) hydrogen bond appears as a non-bonding interaction, with
the total energy density being slightly positive. It can be qualified
as pure closed-shell interaction. This is in line with recent results
of Wood et al.,36 who concluded that H-bonds with DH ◊ ◊ ◊ A
angles of less than about 130◦ have negligibly small energies. The
NH ◊ ◊ ◊ O angle of the weak H-bond in XV is 111.9◦, so both the
geometry, as well as the charge density, points to the weakness of
this bond.
Fig. 4 Molecular graph of the surroundings of the hydrogen-bonding
network in the dimer of XV. Labels of atoms from the molecule generated
by the X3 symmetry operation are mostly omitted for the sake of clarity.
The BCP-s are represented as cylinders. The length of the cylinder is scaled
by r(rBCP), while its cross-sections illustrate ellipticity at BCP, and colour
2
denotes the value of — r(rBCP) (violet – positive; green to red – negative).
Symmetry operation X3: 1 - x, 2 - y, -z.
The negative charge is concentrated at all oxygen atoms and
also at nitrogen. Carbon atoms do not deviate significantly from
neutrality, apart from the methyl group, in which the carbon atom
is positively charged. This may be an effect of polarization or a
sign of slight rotational disorder of the methyl group. The trends
are in general preserved in the case of charges integrated over
atomic basins. The oxygen atoms are all significantly negative, as
well as the nitrogen atom. The C(7) methyl carbon is significantly
positively charged, and so is the C(8) carbon involved in imine
bond. There is a significant charge alternation along the bonds
enclosing two intermolecular hydrogen bonds: O(1)–C(2)–C(1)–
C(8)–N(1)–C(9)–C(10)–C(11)–O(3), while carbon atoms belong-
ing to ring C(1)–C(6) tend to be slightly positive, and those
belonging to C(10)–C(15) tend to be slightly negative. Hydrogen
atoms involved in H-bonds are both distinguished by significant
positive integrated atomic charges, in the case of H(3O) exceeding
0.5 e.
The strong hydrogen bonds both qualify as shared-shell in-
37
teractions with respect to H(rBCP) and G(rBCP)/r(rBCP) criteria
Characteristically for the hydrogen bound system, the charge
concentrations on the O–H and N–H covalent bonds are very
high. The map of the Laplacian illustrates well-defined charge
concentrations on oxygen atoms, indicative of the lone pair
existence. The Laplacian distribution at hydrogen atoms involved
in hydrogen bonding is very significantly polarized, an effect
observed before38 for the hydrogen-bound system. In the case of
H(3O), it is polarized exactly towards the O(1) acceptor, while
in the case of H(1N) the polarization is in the direction between
the two available acceptors. Atomic charges from the monopole
populations are illustrated in Fig. 4–6.
According to Gatti et al.,39 the source function allows for a
classification of hydrogen bonds in terms of characteristic source
contributions to the density at the H-bond critical point from the H
involved in the H-bond, the H-donor (D), and the H-acceptor (A).
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 421–430 | 427
©

View Article Online
strength, consistent with the parallel increase in the electrostatic
character of the interaction. In terms of such a classification, three
hydrogen bonds present in the structure XV are very distinct.
The intermolecular, linear O(3)–H(3O) ◊ ◊ ◊ O(1) hydrogen bond,
already classified as the strongest due to geometrical parameters,
is a very strong and charge-assisted hydrogen bond in the source
function classification, as the H(3O) hydrogen has a positive source
contribution, and the combined hydrogen donor, acceptor and
H(3O) contributions constitute nearly 90% of all inputs to charge
density in this bond. The N(1)–H(1N) ◊ ◊ ◊ O(1) hydrogen bond
is a medium strength hydrogen bond (only about 50% of the
source contributions come from atoms belonging to that bond).
Interestingly, its characteristics (i.e. significantly negative source
contribution from the H(1N) hydrogen and the donor contribution
being larger than that of acceptor by nearly 10%) suggest that this
hydrogen bond is rather polarization- than resonance-assisted.39,40
This is counterintuitive to the location of hydrogen bond inside the
conjugated system. The third hydrogen bond is extremely weak in
terms of source contributions. Only atoms which do not directly
constitute this bond, have positive contributions to the charge
density at its bond critical point. Source contributions to the
critical points of all three hydrogen bonds have been calculated,
and are presented in Table 4.
Fig. 5 Atomic charge distribution with numerical values below (AIM
atomic charges in bold, and charges resulting from monopole populations).
The direction of resulting dipole moment is shown as a yellow arrow.
Atomic charges: O(1), -0.53(2), -1.096; O(2), -0.34(2), -1.033; O(3),
-0.43(2), -1.094; N(1), -0.01(5), -0.919; C(1), -0.08(3), -0.098; C(2),
-0.06(3), 0.501; C(3), 0.01(3), 0.015; C(4), 0.01(3), 0.004; C(5), -0.01(3),
0.415; C(6), 0.06(3), 0.060; C(7), 0.49(4), 0.710; C(8), -0.06(3), 0.469;
C(9), 0.04(4), 0.316; C(10), -0.09(3), -0.104; C(11), -0.00(3), 0.437; C(12),
-0.10(3), -0.093; C(13), -0.17(4), -0.113; C(14), -0.13(4), -0.066; C(15),
0.14(4), 0.125; H(3), -0.02(2), -0.029; H(4), 0.07(2), 0.061; H(6), 0.09(2),
0.075; H(7A), -0.05(1), -0.021; H(7B), -0.05(1), -0.021; H(7C), -0.05(1),
-0.018; H(8), 0.10(2), 0.069; H(9A), 0.13(2), 0.097; H(9B), 0.12(2), 0.099;
H(12), 0.11(2), 0.091; H(13), 0.23(2), 0.187; H(14), 0.13(2), 0.064; H(15),
0.05(2), 0.044; H(1N), 0.09(3), 0.301; H(3O), 0.27(2), 0.505.
Conclusions
A series of sixteen Schiff bases, which can form different types of
intramolecular hydrogen bonds, has been investigated, applying
several NMR methods both in the solution and solid state. Ad-
ditionally, the X-ray structures have been established for three of
them. This series of Schiff bases, derivatives of salicylaldehydes and
aminophenols or hydroxybenzylanine, consists of four subgroups
defined according to presence and position of additional OH
The source contribution from the H appears as the most distinctive
marker of the H-bond strength, being highly negative for isolated
H-bonds, slightly negative for polarized charge assisted H-bonds,
close to zero for resonance-assisted H-bonds, and largely positive
for charge-assisted H-bonds. The contributions from atoms other
than H, D, and A strongly increase with decreasing H-bond
Fig. 6 Laplacian maps of the Schiff base fragment (a) and hydrogen bond network (b).
428 | Dalton Trans., 2011, 40, 421–430 This journal is The Royal Society of Chemistry 2011
©

View Article Online
Table 4 Source function analysis for hydrogen bonds in XV. Symmetry operation X3: 1 - x, 2 - y, -z
-3
2
-5
˚
˚
˚
˚
˚
D ◊ ◊ ◊ A [A]
D–H [A]
H ◊ ◊ ◊ A [A]
r(rBCP) [e A ]
— (rBCP) [e A ]
SD [%]
S
H [%]
SA [%]
SD+H+A [%]
O(1)–H(1N)
O(1)–X3_H(3O)
O(3)–H(1N)
2.6016(3)
2.5332(3)
2.8889(3)
0.8784(2)
0.8823(3)
0.8784(2)
1.8882(3)
1.6598(2)
2.4492(3)
0.26(1)
0.47(4)
0.066(5)
3.33(1)
2.33(7)
1.34(2)
32.5
35.9
-1.0
-9.9
9.8
-56.2
24.7
40.0
-48.3
47.4
85.6
-105.6
group and the length of the linker. Typical hydrogen bonds,
involving hydroxyl groups from the parent aldehydes, are formed
in all investigated compounds. No 5-membered rings involving the
OH-groups in 2¢ position were found.
The X = H, OCH3 and halogen-substituted compounds from
families A and A¢ exist in solution as the OH H-bond form with
some asymmetric intramolecular hydrogen bonding, while the 5-
nitro compounds exist as the hydrogen-bonded NH. Based on
the nitrogen and carbon chemical shifts, it appears that only
the OH groups at the position 2 are involved in the hydrogen
bond formation (with the O to N proton transfer) in all studied
compounds. In the solid state, compounds XIII–XVI exist in the
NH form with only very small differences at the hydrogen atom
positions.
The third group of compounds, IX–XII (group C) is very similar
to group B, and can form only one type of H-bonds because they
do not have a second hydroxyl group at a suitable position (2¢).
In the case of the IX, XI and XII, some very similar hydrogen
bonds exist in both phases with the hydrogen atom located close
to oxygen, as confirmed by the nitrogen chemical shifts of the
imine atoms. Again in the 5-nitro derivative X, the contribution
of the NH form is much higher in solution, and this structure
dominates in the solid state.
energy densities, the weak O(3) ◊ ◊ ◊ H(1N)–N(1) hydrogen bond is
a non-bonding interaction, with total energy density being slightly
positive. It can be qualified as pure closed-shell interaction. The
strong hydrogen bonds both qualify as shared-shell interactions
with respect to H(rBCP) and G(rBCP)/r(rBCP).
The intermolecular, linear O(3)–H(3O) ◊ ◊ ◊ O(1) hydrogen bond
has a positive source contribution from the H(3O) hydrogen, and
the combined hydrogen donor, acceptor and H(3O) contributions
constitute nearly 90% of all inputs to charge density at the BCP
of this bond. In the case of the N(1)–H(1N) ◊ ◊ ◊ O(1) hydrogen
bond, only about 50% of the source contributions come from
atoms belonging to that bond. A significantly negative source
contribution from H(1N) hydrogen, and the donor contribution
being larger than that of acceptor by nearly 10%, suggest that this
hydrogen bond is polarization- rather than resonance-assisted,
which is counterintuitive to the location of hydrogen bond in the
conjugated system. In the case of the third hydrogen bond, it is
extremely weak in terms of source contributions, and it seems to
be an interaction resulting mainly as an effect of the presence of
two strong hydrogen bonds in its vicinity.
Acknowledgements
In the solid state, the contribution of the NH hydrogen bond
form in family B is generally higher than in family C. In DMSO
solution the tautomeric equilibrium for 5-nitro derivative II is less
shifted to the NH form compared to compound X. The remaining
compounds from this group (I, III and IV) have in DMSO solution
very similar structure to XII, XV and XVI respectively.
X-ray structures of the compounds VII, XV and XVI reveal
intricate networks of both intra- and intermolecular H-bonds for
these compounds. In the case of VII (group A), the network
results in the infinite chain of molecules along the [010] crys-
tallographic direction, whereas compounds XV and XVI from
group A¢ display a common motif, forming dimers bound by
intermolecular H-bonds. All three structures proved the existence
of strong intermolecular H-bonds, as well as very weak additional
intramolecular H-bonds involving the nitrogen and the oxygen
from 2¢-OH group. These interactions were undetectable by means
of carbon and nitrogen chemical shift analysis, because they did
not involve proton transfer from 2¢-OH hydroxyl group onto the
nitrogen atom.
The intramolecular hydrogen bonds in XV differ significantly in
terms of charge density properties. According to the results from
the experimental charge densities, the O(1) ◊ ◊ ◊ H(1N)–N(1) bond
is strong, with significant charge density concentration at the bond
critical point, and a straight, well-defined bond path. The same
comment is true in the case of the intermolecular O(1) ◊ ◊ ◊ H(3O)–
O(3) bond. In contrast, the second intramolecular hydrogen bond
is weak, showing charge density concentration over five times
less than the other H-bonds, and a bent bond path. In terms of
Single-crystal X-ray measurements were carried out at the
Structural Research Laboratory of the Chemistry Depart-
ment, Warsaw University, Poland, established with financial
support from the European Regional Development Found in
the Sectoral Operational Programme “Improvement of the
Competitiveness of Enterprises, years 2004–2006” project no:
WKP_1/1.4.3./1/2004/72/72/165/2005/U. KW thanks finan-
cial support from the Ministry of Science and Higher Education –
grant 1 T09A 116 30. KW and AM thank financial support from
the Foundation for Polish Science “MASTER” professorship.
References
1 E. Hadjoudi, Mol. Eng., 1995, 5, 301 and references cited therein.
2 T. Dziembowska, Pol. J. Chem., 1998, 72, 193 and references cited
herein.
3 A. Filarowski, J. Phys. Org. Chem., 2005, 18, 686 and references cited
therein.
4 W. Schilf and L. Stefaniak, Bull. Pol. Ac.: Chem., 1998, 46, 15.
5 A. Koll, Int. J. Mol. Sci., 2003, 4, 434.
6 W. Schilf, B. Kamienski, T. Dziembowska, Z. Rozwadowski and A.
Szady-Chełmieniecka, J. Mol. Struct., 2000, 33, 552.
7 A. Szady-Chełmieniecka, E. Grech, Z. Rozwadowski, T. Dziembowska,
W. Schilf and B. Kamienski, J. Mol. Struct., 2001, 125, 565–566.
8 W. Schilf, B. Kamienski, A. Szady-Chełmieniecka and E. Grech, Solid
State Nucl. Magn. Reson., 2000, 18, 97.
9 W. Schilf, B. Kamienski and T. Dziembowska, J. Mol. Struct., 2002,
41, 602–603.
10 W. Schilf, B. Kamienski, B. Kołodziej, E. Grech, Z. Rozwadowski and
T. Dziembowska, J. Mol. Struct., 2002, 615, 141.
11 T. Dziembowska, K. Ambroziak, Z. Rozwadowski, W. Schilf and B.
Kamienski, Magn. Reson. Chem., 2003, 41, 135.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 421–430 | 429
©

View Article Online
12 G. Wojciechowski, P. Przybylski, W. Schilf, B. Kamienski and B.
Brzezinski, J. Mol. Struct., 2003, 649, 197.
13 W. Schilf, B. Kamienski, Z. Rozwadowski, K. Ambroziak, B. Bieg and
T. Dziembowska, J. Mol. Struct., 2004, 700, 61.
14 G. Wojciechowski, M. Ratajczak-Sitarz, A. Katrusiak, W. Schilf, P.
Przybylski and B. Brzezinski, J. Mol. Struct., 2003, 650, 191.
15 W. Schilf, J. P. Bloxsidge, J. R. Jones and S.-Y. Lu, Magn. Reson. Chem.,
2004, 42, 556.
27 C. C. Ersanli, C. Albayrak, M. Odabasoglu and A. Erdonmez, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2004, 60, o389.
28 I. Majerz, A. Pawlukojc, L. Sobczyk, T. Dziembowska, E. Grech and
A. Szady-Chełmieniecka, J. Mol. Struct., 2000, 552, 243.
29 T. M. Krygowski, K. Woz´niak, R. Anulewicz, D. Pawlak, W.
Kołodziejski, E. Grech and A. Szady, J. Phys. Chem., 1997, 101, 9399.
30 K. Pyta, P. Przybylski, W. Schilf, B. Kołodziej, E. Grech, A. Szady-
Chełmieniecka and B. Brzezinski, J. Mol. Struct., 2010, 967(1–3), 140–
146.
16 W. Schilf, J. Mol. Struct., 2004, 689, 245.
17 B. Kołodziej, P. M. Dominiak, A. Koscielecka, W. Schilf, E. Grech and
K. Wozniak, J. Mol. Struct., 2004, 691, 133.
31 H. Pizzala, M. Carles, W. E. E. Stone and A. Thevand, J. Chem. Soc.,
Perkin Trans. 2, 2000, 935.
18 W. Schilf, B. Kamienski, A. Szady-Chełmieniecka and E. Grech, J. Mol.
Struct., 2004, 700, 105.
19 Z. Rozwadowski, K. Ambroziak, M. Szypa, E. Jagodzinska, S. Spychaj,
W. Schilf and B. Kamienski, J. Mol. Struct., 2005, 734, 137.
20 W. Schilf, B. Kamienski, A. Szady-Chełmieniecka and E. Grech, J. Mol.
Struct., 2005, 743, 237.
32 N. Galic, Z. Cimerman and V. Tomisic, Anal. Chim. Acta, 1997, 343,
135.
33 M. C. Etter, J. C. MacDonald and J. Bernstein, Acta Crystallogr., Sect.
B: Struct. Sci., 1990, 46, 256–262.
34 P. M. Dominiak, E. Grech, G. Barr, S. Teat, P. R. Mallinson and K.
Woz´niak, Chem.–Eur. J., 2003, 9(4), 963–970.
21 Z. Rozwadowski, W. Schilf and B. Kamienski, Magn. Reson. Chem.,
2005, 43, 573.
22 V. A. Ozeryanskii, A. F. Pozharskii, W. Schilf, B. Kamienski, W. Sawka-
Dobrowolska, L. Sobczyk and E. Grech, Eur. J. Org. Chem., 2006, 782.
23 W. Schilf, B. Kołodziej and E. Grech, J. Mol. Struct., 2006, 791, 93.
24 W. Schilf, B. Kamienski, A. Szady-Chełmieniecka, E. Grech, A. Makal
and K. Wozniak, J. Mol. Struct., 2007, 94, 844–845.
25 B. Kołodziej, E. Grech, W. Schilf, B. Kamienski, M. Makowski, Z.
Rozwadowski and T. Dziembowska, J. Mol. Struct., 2007, 32, 844–845.
26 W. Schilf, P. Cmoch, A. Szady-Chełmieniecka and E. Grech, J. Mol.
Struct., 2009, 921, 34–37.
35 U. Koch and P. L. A. Popelier, J. Phys. Chem., 1995, 99, 9747.
36 P. A. Wood, F. H. Allen and E. Pidcock, CrystEngComm, 2009, 11,
1563.
37 P. M. Dominiak, A. Makal, P. R. Mallinson, K. Trzcinska, J. Eilmes, E.
Grech, M. Chruszcz, W. Minor and K. Woz´niak, Chem.–Eur. J., 2006,
12(7), 1941–1949.
38 J. Overgaard, B. Schiøtt, F. K. Larsen and B. B. Iversen, Chem.–Eur. J.,
2001, 7, 3756.
39 C. Gatti, F. Cargnoni and L. Bertini, J. Comput. Chem., 2003, 24, 422–
436.
40 G. Gilli and P. Gilli, J. Mol. Struct., 2000, 552(1–3), 1–15.
430 | Dalton Trans., 2011, 40, 421–430
This journal is
The Royal Society of Chemistry 2011
©