Desmotropy, polymorphism, and solid-state proton transfer: Four solid forms of an aromatic o-hydroxy schiff base

Article abstract of DOI:10.1002/chem.201103508

The Schiff base derived from salicylaldehyde and 2-amino-3-hydroxypyridine affords a diversity of solid forms, two polymorphic pairs of the enol-imino (D1a and D1b) and keto-amino (D2a and D2b) desmotropes. The isolated phases, identified by IR spectrosco

Full text of DOI:10.1002/chem.201103508

DOI: 10.1002/chem.201103508
Desmotropy, Polymorphism, and Solid-State Proton Transfer: Four Solid
Forms of an Aromatic o-Hydroxy Schiff Base
Mirta Rubcˇic´,*[a] Krunoslav Uzarevic´,*[a] Ivan Halasz,[a, b] Nikola Bregovic´,[a]
ˇ
Momir Malisˇ,[c] Ivica ꢀilovic´,[a] Zoran Kokan,[c] Robin S. Stein,[d]
Robert E. Dinnebier,[b] and Vladislav Tomisˇic´[a]
Abstract: The Schiff base derived from
salicylaldehyde and 2-amino-3-hy-
ent molecular conformations and, sub-
sequently, by divergent supramolecular
architectures. The appearance and in-
terconversion conditions for each of
the four phases have been established
on the basis of a number of solution
and solvent-free experiments, and eval-
uated against the results of computa-
tional studies. Solid phases readily con-
vert into the most stable form (D1a)
upon exposure to methanol vapor,
heating, or by mechanical treatment,
and these transformations are accom-
panied by a change in the color of the
sample. The course of thermally in-
duced transformations has been moni-
tored in detail by means of tempera-
ture-resolved powder X-ray diffraction
and infrared spectroscopy. Upon disso-
lution, all forms equilibrate immediate-
ly, as confirmed by NMR and UV/Vis
spectroscopy in several solvents, with
the equilibrium shifted far towards the
enol tautomer. This study reveals the
significance of peripheral groups in the
stabilization of metastable tautomers in
the solid state.
droxypyridine affords a diversity of
solid forms, two polymorphic pairs of
the enol-imino (D1a and D1b) and
keto-amino (D2a and D2b) desmo-
tropes. The isolated phases, identified
by IR spectroscopy, X-ray crystallogra-
phy, and 13C cross-polarization/magnet-
ic angle spinning (CP/MAS) NMR
spectroscopy, display essentially planar
molecular conformations characterized
by strong intramolecular hydrogen
ꢀ
ꢀ
bonds of the O H···N (D1) or N H···O
(D2) type. A change in the position of
the proton within this O···H···N system
is accompanied by substantially differ-
Keywords: desmotropy
·
poly-
morphism · proton transfer · Schiff
bases · tautomerism
Introduction
and activity of various enzymes.[4] Many of these biological
effects rely not only on the presence of different tautomers
but also on the rate at which they interconvert. If a proton
migrates between two electronegative atoms, equilibrium is
generally attained quickly, but it is much slower if the
proton is bonded to, for example, carbon.[2b,c,5] This has espe-
cially significant implications for drug activity because bio-
logical systems usually recognize only one tautomer and the
consumption rate and availability of a drug will depend on
how fast the equilibrium is achieved.[3g,h,6] Consequently,
there has been much commercial and intellectual-property
interest in the proper identification, characterization, and
registration of solid pharmaceutical materials, which has led
to the recognition of numerous tautomeric solids.[7]
Tautomeric equilibria in solution have been thoroughly
investigated and were found to be dependent upon the
choice of solvent, pH, temperature, concentration, and so
forth.[8] In contrast to solution, in which different tautomers
coexist, in the solid state usually only one tautomer appears.
Rare examples in which more than a single tautomer emerg-
es in the solid state mostly include stoichiometric[9] or non-
stoichiometric[7a] mixtures comprising the same crystal. Even
more scarce are cases in which different tautomers form dis-
tinct crystalline phases, which are defined as desmo-
tropes.[10–12] Desmotropic systems are important because
they provide an exceptional opportunity for studying tauto-
Tautomerism is an equilibrium involving two or more iso-
meric structures capable of facile interconversion by migra-
tion of an atom or a small group within a molecule.[1] The
most frequently encountered type of tautomerism refers to
migration of a proton, and is termed prototropy.[2] Tauto-
merism plays a number of key roles in biological systems,
from DNA base-pairing[3] to the regulation of the function
ˇ ´
´
ˇ
´
´
[a] Dr. M. Rubcic, Dr. K. Uzarevic, Dr. I. Halasz, N. Bregovic,
ˇ ´
Dr. I. ꢀilovic, Prof. V. Tomisic
Department of Chemistry, Faculty of Science, University of Zagreb
Horvatovac 102a, 10002 Zagreb (Croatia)
Fax : (+385)1-4606341
E-mail: mirta@chem.pmf.hr
kruno@chem.pmf.hr
[b] Dr. I. Halasz, Prof. R. E. Dinnebier
Max Planck Institute for Solid State Research
Heisenbergstr. 1, 70569 Stuttgart (Germany)
ˇ
[c] M. Malis, Z. Kokan
ˇ
´
ˇ
Ruder Boskovic Institute, Bijenicka 54, 10002 Zagreb (Croatia)
¯
[d] Dr. R. S. Stein
Bruker UK Ltd, Banner Lane, Coventry CV4 9GH (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103508.
5620
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

FULL PAPER
meric forms separately, which is seldom possible in solution,
even in the case of a slow equilibrium. The most common
type of desmotropy refers to the annular desmotropy of
azoles in which a proton migrates between nitrogen atoms
of different five-membered heterocycles (Scheme 1a).[12d]
Less frequently encountered are functional desmotropes in
which the proton resides on different functionalities altering
thus not only the electronic properties of a molecule, but
also the donor/acceptor characteristics of its functionalities
(Scheme 1b and c).[12e,14]
the keto-amino, occurs separately, thereby presenting the
first case of desmotropy in Schiff bases. Moreover, we have
established that each tautomer crystallizes in two poly-
morphic forms. In this multifaceted study, the occurrence of
the four phases was thoroughly investigated in a variety of
crystallization conditions, including solvent-free systems.
The relative stabilities and interconversions of the four
phases were assessed by various solution and solid-state
methods and evaluated against the results of a computational
study. Structural investigations focusing on the correlation
between the molecular geometries of the tautomers and
their supramolecular architectures revealed an intriguing re-
lationship between the N···H···O proton position and the
molecular conformation of a particular tautomeric form.
Results
Syntheses: The Schiff base investigated in this work, 2-[(2-
hydroxybenzylidene)amino]pyridin-3-ol (H2L), was first syn-
thesized and utilized as a ligand for the preparation of me-
chanosensitive molybdenum(VI) complexes.[26] It was then
observed that the ligand itself appeared in three solid forms:
Orange prisms, yellow plates, and red needles. Subsequent
crystallization from several solvents yielded these three
forms concomitantly. The first indication of the ligandꢂs di-
verse solid-state nature was provided by powder X-ray dif-
fraction (PXRD), whereas UV/Vis and solution NMR meas-
urements showed identical spectra for all three forms. How-
ever, the IR spectra indicated significant differences be-
tween the molecular structures of the red solid on one hand
and the yellow and orange solids on the other. This motivat-
ed us to conduct a comprehensive study that would explain
the causes of the observed differences.
Scheme 1. Different types of prototropy: a) Annular (pyrazoles); b) func-
tional (pyrazolinones);[13] c) ketoamino-enolimino (Schiff bases).
Among the variety of organic compounds that can exhibit
functional tautomerism, probably the best studied are Schiff
bases
derived
from
salicylaldehyde
derivatives
This investigation first required a clear definition of the
occurrence domains of each of the solids. It was recognized
that polar solvents afforded orange prisms (methanol, aceto-
nitrile) and yellow plates (n-propanol) in high yields. In con-
trast, red needles were harvested from tert-butanol. Further
solvent screening yielded a new phase from isopropanol in
the form of large dark-red prisms (see Figure 1a and the Ex-
perimental Section). The X-ray structural studies revealed
unambiguously that the four crystalline forms of the title
compound correspond to two pairs of polymorphs of the
enol-imino (D1) and keto-amino (D2) desmotropes[30]
(Scheme 1b and c, respectively). The crystal and molecular
structures of D1b (yellow plates) and D2b (dark-red
prisms) were determined by single-crystal X-ray diffraction,
whereas the crystal structure of D2a (red needles) was
solved on the basis of a high-resolution PXRD analysis. Due
to the poor quality of the crystals, single-crystal X-ray dif-
fraction revealed only a preliminary structural model of
D1a (orange prisms), which was finally refined against high-
resolution PXRD data.
(Scheme 1c).[15] They crystallize predominantly in the enol-
imino form,[16] although the appearance of a particular tau-
tomer also depends upon the substituents in both the salicyl-
aldehyde and amine moieties.[17,18] Proton transfer (PT) in
such systems proceeds along the channel of a resonance-as-
sisted hydrogen bond (RAHB; in this case an intramolecu-
lar N···H···O hydrogen bond),[19] and in the solid state is
often linked to the photo-[20] or thermochromic[21] properties
of a substance.[22,23] Although such compounds and accom-
panying phenomena have been widely studied, the mecha-
nism of this apparently simple rearrangement still presents
a number of challenges.[20c,d] PT in these systems yields meta-
stable tautomeric forms, which revert to the starting tauto-
mer upon removal of the external stimuli (heat or light).
Schiff bases displaying photo- and thermochromism have
raised exceptional interest due to their potential applications
in data storage, information processing, telecommunications,
and optical switching.[24] However, thus far, desmotropes of
Schiff bases have not been observed.
Herein we present a unique case of a salicylaldimine
Schiff base displaying both desmotropy and polymorph-
ism.[25] Namely, each of its tautomers, the enol-imino and
Crystal structures: The presence of a strong intramolecular
ꢀ
ꢀ
O2 H2···N2 (D1) or N2 H2···O2 (D2) hydrogen bond
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5621

ˇ ´
M. Rubcic et al.
Figure 2. Overlapping Mercury[31] POV-Ray rendered views of a) enol-
imino desmotropes D1a (light grey) and D1b (black), and b) keto-amino
desmotropes D2a (light-grey molecule A, dark-grey molecule B) and
D2b (black). The diagrams were constructed by overlapping the salicyl
rings. The dihedral angles between the planes of the salicyl and pyridine
are 6.21(5)8 for D1a, 6.68(10)8 for D1b, 4.67(10)8 for molecule A and
3.47(10)8 for molecule B in D2a, and 9.93(10)8 for D2b.
The main effect of relocation of the H2 proton is on the
intermolecular donor/acceptor properties of the O2 atom
because the N2 atom, due to steric hindrance, remains avail-
able only for the intramolecular hydrogen bond
O2···H2···N2. Consequently, in addition to differences in
their molecular structures, the crystal architectures of the
D1 and D2 forms differ in their primary supramolecular
synthon (Figures 3 and S4–S8 in the Supporting Informa-
tion).
Figure 1. a) Four solid forms of H2L. ORTEP[27] POV-Ray[28] rendered
views of the molecular structures of b) D1b and c) D2b showing the
atomic numbering. Displacement ellipsoids are drawn at the 30% proba-
bility level. Hydrogen atoms are presented as spheres of arbitrary small
radii. The main geometrical parameters [ꢃ and 8] of the intramolecular
Crystal structures of the enol desmotropes D1a (P21/c) and
D1b (Pbca): The primary supramolecular architectures of
the D1 polymorphs, that is, infinite zig-zag chains, are as-
ꢀ
ꢀ
ꢀ
O2 H2···N2 hydrogen bonds in enol desmotropes: D1a: O2 H2
sembled through an intermolecular O1 H1···N1 hydrogen
ꢀ
0.889(4), H2···N2 1.822(4) O2···N2 2.6153(16), aO2 H2···N2 147.4(3);
bond, thus utilizing the functionalities of the pyridine part
ꢀ
ꢀ
D1b: O2 H2 1.01(2), H2···N2 1.71(2), O2···N2 2.612(2), aO2 H2···N2
147(2). The geometrical parameters [ꢃ and 8] of the intramolecular C6
H6···O2 interaction: D1a: C6 H6 0.929(4), H6···O1 2.292(4), C6···O1
2.9325(18), aC6 H6···O1 125.7(3); D1b: C6 H6 0.93, H6···O1 2.10,
C6···O1 2.770(2), aC6 H6···O1 127. The main geometrical parameters
[ꢃ and 8] of the intramolecular N2 H2···O2 hydrogen bonds in the keto
desmotropes: D2a: N2A H2A 1.000(6), H2A···O2A 1.698(7),
N2A···O2A 2.533(3), aN2A H2A···O2A 138.2(4); N2B H2B 0.997(6),
of the molecule. In D1b, the formation of chains is further
ꢀ
ꢀ
assisted by intermolecular O2 H2···O1 hydrogen bonds. Al-
ꢀ
ꢀ
ꢀ
though the D1a and D1b chains appear rather similar, they
actually display substantial differences (Figures 3a, S4, and
S5, and Table S4 in the Supporting Information).
ꢀ
ꢀ
ꢀ
In D1a, the chains run parallel to the c axis and the mole-
cules within them are rotated through approximately 908
with respect to each other (Figures 3 and S4). The chains
ꢀ
ꢀ
ꢀ
H2B···O2B 1.667(7), N2B···O2B 2.536(3), aN2B H2B···O2B 143.2(5);
D2b: N2 H2 1.02(2), H2···O2 1.69(2), N2···O2 2.552(2), aN2 H2···O2
ꢀ
ꢀ
139.1(17).[29]
ꢀ
are further associated into layers through C H···O interac-
tions along the a axis (Figure S4 and Table S4). Although
chains within the same layer assume the same orientation,
the chains in two adjacent layers are oriented in the oppo-
site direction. Neighboring layers are finally assembled
through van der Waals interactions along the b axis into the
final 3D architecture.
(RAHB) and the conjugated p system leads to nearly planar
molecular structures for all four forms (Figures 1b and 2,
and Table S3 in the Supporting Information). The planarity
of molecules of D1 is further facilitated by favorable intra-
ꢀ
molecular C6 H6···O1 interactions (Figure 1b). However,
In D1b, the hydrogen-bonded zig-zag chains extend along
the a axis and the molecules within them are mutually tilted
by around 608 (Figure 3 and S5). The chains, oriented in the
molecules in polymorphs of D1 assume the anti conforma-
tion with respect to the O1 and O2 atoms whereas mole-
cules in polymorphs of D2 adopt the syn conformation,
which corresponds to a flip of the pyridine ring of around
1808 (Figure 1c). Differences in polymorphic pairs are subtle
and appear mostly in the spatial arrangements of the salicyl
(sal) and pyridine (py) rings (Figure 2).
ꢀ
opposite direction, are then joined through C H···O interac-
tions and two types of p···p interactions, along the c axis, to
form layers (Figure S5 and Table S4).
¯
Crystal structures of the keto desmotropes D2a (P1) and
D2b (P212121): In contrast to the structures of D1, the crys-
5622
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

Desmotropy and Polymorphism
FULL PAPER
Figure 3. Primary intermolecular synthons of the D1 and D2 forms:
a) D1a: light grey; D1b: black; b) D2a: light grey; D2b: black. Hydro-
gen bonds are shown with dashed lines. The main geometrical parameters
Figure 4. a) TR-PXRD and b) TR-IR spectra of D2a during the D2a!
D1 transformation. The sample was heated in the range of 30–1708C. No
significant changes were observed below 1008C.
ꢀ
[ꢃ and 8] of the hydrogen bonds in the enol desmotropes: D1a: O1 H1
i
i
i
ꢀ
0.868(4), H1···N1 1.785(4), O1···N1 2.649(2), aO1 H1···N1 173.5(4), i=
x, 1= ꢀy, 1= +z]; D1b:O1 H1 1.02(2), H1···N1i 1.62(2), O1···N1i
ꢀ
2
2
i
1
3
ꢀ
2.6413(18), aO1 H1···N1 175.6(17), i=ꢀ = +x, y, = ꢀz]. The main geo-
the opposite direction, interact through van der Waals inter-
actions along both the b and c axes to form a 3D architec-
ture (Figure S7).
2
2
metrical parameters [ꢃ and 8] of the hydrogen bonds in the keto forms:
D2a: O1A H1A 0.876(6), H1A···O2B 1.865(6), O1A···O2Bi 2.741(3),
i
ꢀ
i
ꢀ
ꢀ
aO1A H1A···O2B 177.6(5), i=1ꢀx, 1ꢀy, 2ꢀz; O1B H1B 0.871(7),
H1B···O2Aii 1.825(7), O1B···O2Aii 2.695(4), aO1B H1B···O2A
ii
ꢀ
179.3(6), ii=ꢀx, 1ꢀy, 2ꢀz]; D2b: O1 H1 0.87(2), H1···O2i 1.71(2),
ꢀ
Solid-state NMR studies, IR spectroscopy, and thermal anal-
ysis: NMR spectroscopy has been recognized as a very pow-
erful method for establishing the presence of particular tau-
tomers in the solid state as well as for investigating the tau-
tomeric equilibrium in solution. In this context, the keto-
enol tautomerism of Schiff bases has been particularly ap-
pealing.[32,33] Among the signals which might be helpful for
distinguishing the two forms, the most sensitive in 13C solid-
sate (SS) NMR spectra is the shift associated with the
carbon atom bearing the keto or enol functionality (here
C8). For the enol form this signal is expected to appear at
around 160 ppm, whereas for the keto form it should be
shifted to around 180 ppm.[34] Because the structures of des-
motropes D1a and D2a have been solved and/or refined
against PXRD data, we looked for additional confirmation
of their structure and correct tautomeric form. 13C cross-po-
larization/magnetic angle spinning (CP/MAS) NMR experi-
ments revealed similar chemical shifts for polymorphs D1a
and D1b (Figure S9 in the Supporting Information). In con-
trast, D2a displayed a significantly different set of signals,
especially for C6 and C8 (included in the RAHB system).
i
i
1
1
ꢀ
O1···O2 2.574(2), aO1 H1O···O2 174(2), i=ꢀ = +x, = ꢀy, ꢀz.
2
2
tal architectures of the D2 polymorphs are dominated by
ꢀ
O1 H1···O2 hydrogen bonds that involve functionalities of
both the pyridine and salicylidene moieties (Figure 3b, S6,
and S7, and Table S5 in the Supporting Information).
In both D2 polymorphs, hydrogen bonding leads to the
formation of zig-zag chains that differ in topology to those
of the D1 polymorphs. In both D2a and D2b, these chains
are parallel to the a axis (Figures S6 and S7). Although in
D2a the molecules are rotated by about 758 with respect to
each other, in D2b this angle is about 508 (Figure 4b). The
wider angle found in D2a allows insertion of the molecules
ꢀ
of the second chain, through C H···O interactions, in
a tongue and groove fashion, between the molecules of the
first chain (Figure S6 and Table S5). Such insertion provides
a growth of layers along the c axis. The layers are stacked
along the b axis and interact only through van der Waals in-
teractions. In contrast, adjacent chains in D2b, running in
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5623

ˇ ´
M. Rubcic et al.
The chemical shifts were assigned by means of nonquaterna-
ry suppression SS NMR experiments and by comparing the
spectra with solution-state assignments (for enol forms, see
the Experimental Section). Namely, in the spectra of D1a
and D1b, the signals assigned to the C8 and C6 atoms
appear at around 161 and 166 ppm, respectively (Figure S9).
In contrast, in the 13C CP MAS NMR spectra of D2a, sig-
nals belonging to the same atoms appear at 177 ppm for C8
and at 155 ppm for C6 (Figure S9). These results, in combi-
nation with the X-ray diffraction structural studies, clearly
demonstrate the presence of the D1 polymorphs in the enol
form and D2a in the keto form.
Polymorphs of the same desmotrope display similar IR
spectra, whereas the spectra of the different desmotropes
reveal interesting dissimilarities indicative of their different
molecular structures and the nature of the intra- and inter-
molecular hydrogen bonds. The IR spectra of all four forms
are characterized by a broad band at around 3440 cmꢀ1,
which has been assigned to the vibration of the pyridine
of 100–1608C. Thus, all the forms show almost identical
melting points and deterioration patterns as D1a (Fig-
ure S10). Given the similarities between the melting points
of the four compounds, tautomeric/polymorphic conversion
into D1a was immediately suspected. To inspect this ther-
mal event samples of D1b, D2a, and D2b were heated, col-
lected just before their melting points, cooled to room tem-
perature, and subjected to PXRD. In all three cases, the
powder patterns matched that of D1a exactly, which reveals
that heating causes both keto forms to undergo solid-state
PT to yield the enol D1a as the final product. On the other
hand, heating D1b causes it to undergo a polymorphic tran-
sition into the D1a phase. However, the course of these
transitions could not be elucidated by DSC alone.
Solid-state reactivity: To fully reveal the processes leading
to the broad endotherms observed in the DSC traces, time-
and temperature-resolved (TR) PXRD and IR spectroscopy
were employed. The TR-PXRD experiments revealed that
above 1208C, D1b starts to convert directly into its poly-
morphic pair, D1a (Figure S11a in the Supporting Informa-
tion). In contrast, D2a changes its tautomeric form above
1008C first to give a mixture of D1b and D1a below 1408C,
and then to give exclusively D1a above 1608C (Scheme 2
and Figure 4a). A similar transformation pattern was estab-
lished for D2b (Figure S11b). However, for D2b, as well as
for D1b, solid-state transformations were hard to trace in
detail due to partial amorphization of the samples (Fig-
ure S11). It is also worth mentioning that the D2!D1 trans-
formation was accompanied by a change in the color of the
macroscopic sample from red to yellow-orange.
ꢀ
O1 H1 group involved in the intermolecular hydrogen
bond. Similarly, a broad band at around 2500 cmꢀ1, observed
ꢀ
ꢀ
for all four phases, has been assigned to O H/N H RAHB
ꢀ
ꢀ
stretching vibrations. When O H/N H groups are not in-
volved in hydrogen bonding, they usually present sharp IR
bands with significantly higher wavenumbers.[19d] In the pres-
ꢀ
ꢀ
ent case, the O H/N H groups are involved in many and di-
verse intra- and intermolecular hydrogen bonds, which
broadens the corresponding bands and shifts them to lower
wavenumbers. However, the largest differences between the
desmotropes are observed in the region around 1600 cmꢀ1,
in which the enol-imino desmotrope (D1) displays two
bands assigned to C=N and C=N(py) vibrations. The keto-
amino desmotrope (D2), on the other hand, shows three
bands corresponding to the C=O, C=C, and C=N(py) vibra-
tions. The positions of the bands at around 1607 cmꢀ1 (C=N)
for D1 and 1623 cmꢀ1 (C=O)
In many respects IR spectroscopy is complementary to
PXRD. Although the latter reveals changes in the crystal
structures, TR-IR spectroscopy detects changes at the mo-
lecular level. Even though the assignment of bands in the
for D2 are consistent with the
resonant nature of these
groups, as established by struc-
tural studies.
Thermogravimetric analyses
of all four phases revealed that
no loss of weight occurs below
2208C. According to differen-
tial scanning calorimetry (DSC)
measurements, D1a does not
display any significant changes
before the melting point, which
was observed at 1708C (onset;
Figure S10 in the Supporting
Information). This thermal
event is immediately followed
by an exotherm corresponding
to the deterioration of the com-
pound. DSC curves of the other
three forms reveal a broad and
shallow endotherm in the range Scheme 2. Reaction conditions and solid-state reactivity of the four desmotropic polymorphs.
5624
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

Desmotropy and Polymorphism
FULL PAPER
Table 1. UV/Vis spectral data for H2L in various solvents.
IR spectra is not always trivial, data collection is considera-
bly quicker than for PXRD (in our case 10–30 s versus
30 min, respectively). Therefore many transformations
hidden during one PXRD data collection cycle can poten-
tially be detected by TR-IR. In this case, the TR-IR experi-
ments confirmed the thermally induced D2!D1 transfor-
mations, but in much shorter time intervals then in TR-
PXRD (the samples were held for 5 min at every measured
temperature). In combination with DSC data (for which the
samples were heated at 58Cminꢀ1), the results indicate that
the temperature of the sample, and not the time the sample
is held at a specific temperature, is crucial for the D2!D1
transformation to proceed.
Solvent
l [nm] (e [104 dm3 molꢀ1 cmꢀ1])
D1
D2
455 (0.158)
359 (1.351)
269 (0.934)
458 (0.136)
361 (1.373)
269 (0.908)
463 (0.116)
363 (1.451)
269 (0.949)
359 (1.491)
269 (0.985)
365 (1.459)
363 (1.669)
455 (0.163)
359 (1.304)
269 (0.898)
458 (0.136)
361 (1.343)
269 (0.908)
463 (0.117)
363 (1.444)
269 (0.977)
359 (1.496)
269 (0.988)
365 (1.439)
364 (1.669)
methanol[a]
ethanol[a]
tert-butanol
acetonitrile
toluene
dichloromethane
For D2a, changes in the IR spectra first became visible at
1008C and became more evident with every temperature in-
crement of 108C. The most significant changes are the shift
of the band at 1623 (C=O) to 1607 cmꢀ1 (C=N), the weaken-
ing of the band at 1571 cmꢀ1, and the appearance of the
band at around 1554 cmꢀ1. The spectrum at 1608C corre-
sponds to that of the pure D1 forms (Figure 4b). For D2b,
the same transition pattern and heating outcome was ob-
served but, compared with D2a, it started to transform at
an approximately 308C higher temperature, which is consis-
tent with TR-PXRD experiments (Figure S12b in the Sup-
porting Information). As expected, the heating of D1b in-
duced only slight spectral changes because the same tauto-
meric form is retained during the transition (Figure S12c).
[a] The UV/Vis spectra of H2L in these two solvents have previously
been recorded and the published data are in agreement with our re-
sults.[37]
be seen that solutions prepared by dissolving crystals of dif-
ferent tautomeric forms of H2L display almost identical
spectra. In aprotic solvents, namely acetonitrile, toluene,
and dichloromethane, only two bands emerge, centered at
approximately 269 and 360 nm (Figure S14 in the Support-
ing Information), which indicates the almost exclusive pres-
ence of the enol-imino tautomer.[36] Both bands can be as-
signed to p!p* transitions: The maximum at 269 nm corre-
sponds to the excitation of the aromatic rings and the band
at around 360 nm is associated with the excitation of the
whole molecule.[36a–c] A slight blue shift of the latter is ob-
served with increase in polarity of the solvent, which is in
agreement with previous studies of similar compounds.[36a,c]
In protic solvents, that is, alcohols, H2L displays an addition-
al band at 460 nm, which has been ascribed to the appear-
ance of the keto-imino tautomer.[36b,c,f,g] An increase in sol-
vent polarity is accompanied by an increase in the intensity
of this band, which implies a greater abundance of the keto
tautomer (Table 1 and Figures S15 and S16 in the Support-
ing Information). However, the tautomeric equilibrium con-
stants could not be estimated on the basis of the UV spec-
tral data in a way described elsewhere[36b,c] because the ex-
tinction coefficients of the individual tautomeric forms ap-
peared to be solvent dependent.
Mechanochemistry: D1b, D2a, and D2b proved sensitive to
mechanical treatment, yielding in all cases pure D1a. On
the other hand, D1a did not experience any phase transi-
tions upon grinding. Furthermore, no amorphous phase or
any kind of decomposition product was observed.
Although all the samples are stable in air for months,
short exposure (1 h) of the D1b, D2a, and D2b phases to
methanol vapors induced an irreversible transformation to
D1a. Moreover, the D2 tautomers changed color from red
to yellow after 10 min and no further changes were then ob-
served. The exposure of D1 polymorphs to assorted solvent
vapors was also tested but did not result in a reverse trans-
formation to any of the D2 polymorphs.
Competitive slurry experiments: The relative thermodynam-
ic stabilities of all the phases were evaluated by slurry ex-
periments performed in isopropanol at room tempera-
ture.[12h,35] In the presence of seed crystals of the D1 forms,
the D2 forms completely convert into the corresponding D1
forms within 5 h (Figure S13 in the Supporting Information).
Although no interconversion between the D2 forms was ob-
served, complete conversion of D1b into D1a was observed
after stirring for 18 h (Figure S13). The results of slurry ex-
periments also confirmed the stability of D1 forms in com-
parison to D2. It can thus be concluded that D1a is the ther-
modynamically most stable form at room temperature.
The tautomeric equilibrium for H2L was also investigated
1
by H and 13C solution-state NMR spectroscopy. The reso-
nance signals were assigned by 2D NMR spectroscopy in
[D4]methanol and [D3]acetonitrile (see Figures S17 and S18
in the Supporting Information). The solvents used for
1H NMR spectroscopy were chosen in accordance with the
UV/Vis experiments. 13C NMR spectra of H2L were record-
ed in acetonitrile and methanol solutions as a result of the
larger differences in the UV/Vis spectra recorded in these
solvents. The relevant NMR data are presented in Table S6
in the Supporting Information.
When assessing fast tautomeric equilibria in solution,
which is usually the case with prototropy, it is expected that
1
Solution studies: The UV/Vis spectral data for both tauto-
both H and 13C NMR spectra display only one set of signals
meric forms in different solvents are given in Table 1. It can
(Figure S19 and Table S6). However, the position of each
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5625

ˇ ´
M. Rubcic et al.
resonance is the population-weighted average of the chemi-
cal shifts of each tautomer and thus NMR measurements
can be exploited for studies of the tautomeric equilibria.[38]
The signal for C8 in [D4]methanol and [D3]acetonitrile ap-
pears at approximately 160 ppm (Table S6), as expected for
the enol-imino form.[34] Similar was observed for the solid
D1 forms. These findings indicate the predominance of the
enol-imino tautomer in both solvents. Another very sensi-
tive parameter is expected to be the chemical shift of the
azomethine proton (H6), which should differ significantly
between the two forms because of its participation in the ar-
omatic system affected by PT. Taken together with the re-
sults of the UV/Vis analysis, the change to a lower chemical
shift of H6 in methanol indicates a greater fraction of the
keto form present in this solvent than in the others. Note
that in the absence of exchange with deuterated solvent, H6
should appear as a singlet in the case of the enol-imino form
and as a doublet for the keto-amino form due to spin-cou-
pling with the amino hydrogen atom. In the case of the
methanol solution, the appearance of a singlet is readily ex-
plained by an exchange of the amino hydrogen with the
deuterated solvent.
intramolecular hydrogen bond, which defines a six-mem-
ꢀ
bered ring. The corresponding O2 H2···N2 RAHB was
found in the solid state.
Structures related to the three most stable enol conform-
ers (Figure 5a,i–iii) were also found in the conformational
space of the keto tautomer. These structural pairs differ
only in the movement of H2 from the proximity of O2 to
N2.
A favorable orientation of the H1 atom towards an elec-
tron-rich area of a RAHB explains the stability of the enol
conformer in Figure 5a,i. This enol structure is not planar
ꢀ
but slightly twisted around the C1 N2 bond, which enables
ꢀ
a better interaction between the O1 H1 group and the
RAHB. The close contact between the H1 and H2 atoms
decreases the stability of the corresponding keto form and
reduces the energy barrier of the keto-to-enol conversion
(the difference between the ketone energy minimum struc-
ture and the corresponding transition state) to 2.87 kJmolꢀ1.
Such a small energy barrier is easily overcome by a light-
weight particle, such as a proton. In the other direction, the
barrier to enol-to-keto transformation is 34.14 kJmolꢀ1, so
the enol form is the most probable structure when the
system is in the lowest vibrational state.
1
The remaining features of the H NMR spectra reveal, as
expected, aromatic protons appearing as multiplets in the
range 7–8 ppm. The H2 and H1 protons are observed as
broad singlets at 12.0–13.5 and 7.0–9.5 ppm, respectively.
These signals show substantial solvent dependency and, due
to deuterium exchange, they do not appear in the spectrum
recorded in methanol solution.[39,40]
The temperature-dependent NMR measurements re-
vealed that, upon heating, the H2 signal broadens substan-
tially and shifts downfield (Figures S20–S23 in the Support-
ing Information). This suggests that the RAHB is also pres-
ent in solution. In an acetonitrile solution, the signal as-
signed to H1, on the other hand, shifts upfield with increase
in temperature. This is most likely due to the formation of
hydrogen bonds with solvent molecules. The remaining sig-
nals are not significantly affected by a change in tempera-
ture.
The second most stable enol-keto pair (Figure 5a,ii) has
H1 oriented away from the RAHB, which destabilizes both
the enol and keto structures but allows them to be planar.
The pyridine and salicyl rings in the enol energy minimum
ꢀ
structure are rotated by about ꢁ108 around the C1 N2
bond. This structure is only 24 Jmolꢀ1 more stable than the
full planar form. This becomes negligible when taking nucle-
ar motions into account because they average the structure
to a planar form. The corresponding keto conformer has the
optimized geometry of the D2 crystal structures with a keto-
to-enol conversion barrier of 10.43 kJmolꢀ1. Again the enol
form is more stable, but because of the lower barrier to
transition (27.05 kJmolꢀ1), the keto form is somewhat more
accessible.
ꢀ
Rotation of the pyridine ring around the C1 N2 bond
leads to the third keto-enol pair (Figure 5a,iii). In this case
the OH functionalities are very nearly in an anti position,
which prevents their favorable interaction. This results in
decreased stability in comparison with the above described
conformers. The third enol conformer has the optimized ge-
ometry of molecules in the D1 crystal structures. The energy
barriers to conversion are similar to the previous enol-keto
pair, being 29.64 and 9.98 kJmolꢀ1, respectively, again caus-
ing the enol form to be the most probable when the system
is in the lowest vibrational state. In general, all the keto
forms (unlike enol) tend to be planar because of the greater
conjugation of the electrons of the N2 atom with the p-elec-
tron system of the pyridine ring. In addition, all the keto
structures have a greater dipole moment then their enol
pairs (Table S7 and Figure S24 in the Supporting Informa-
tion).
Computational studies: A thorough analysis of the molecu-
lar conformational space of both H2L tautomers was per-
formed in vacuum and in a polarizable continuum. A densi-
ty functional theory approach with the wB97X-D[41] ex-
change-correlation functional, instead of the more common
B3LYP,[42] was applied to better account for intramolecular
dispersion interactions. A relatively large basis set (6-311+
GACHTUNGTRENNUNG
(2d,p))[43] was used with the purpose of minimizing the in-
tramolecular basis set superposition error. Two groups of
low-energy conformers were found, one comprising the six
most stable enol tautomers and the other comprising the
three most stable keto tautomers. Within the enol group, all
ꢀ
conformers differ in their rotation around the N2 C1 bond
ꢀ
ꢀ
ꢀ
(torsion angle: C6 N2 C1 N1) and in the orientation of
the OH groups on the salicyl and pyridine rings. The three
most stable enol conformers are characterized by the conju-
gated p system of the benzene ring and an almost collinear
Breaking the RAHB system by moving the H2 atom to
the other side of the benzene ring gives the last set of the
least stable enol structures (Figure 5a,iv–vi), which do not
5626
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

Desmotropy and Polymorphism
FULL PAPER
Figure 5. a) Relative electronic energies of conformers calculated at the wB97X-D/6-311+GACTHUNRTGNEUNG(2d,p) level of theory for structures in vacuum. b) Each line
represents the energies of the different conformers—shown in a)—in assorted solvents (characterized by their relative permittivities) . Each line is posi-
tioned and colored in accordance with the position and the color code of the conformer presented in a).
have a corresponding keto form formed by a simple translo-
cation operation. If the H1 atom is positioned as in the most
stable conformer (Figure 5a,iv), a five-membered angular-
red and dark-red polymorphs of the keto-amino form, D2a
and D2b (Scheme 2).
In the solid state, molecules of the D1 and D2 desmo-
tropes differ not only in the position of the proton but also
in conformation. In both polymorphs of D1, the O1 and O2
atoms adopt an anti conformation (Figure 1b). In contrast,
in the D2 forms, the pyridine ring is rotated by about 1808
and the molecules assume a syn conformation (Figure 1c).
Such changes also lead to an alteration in the spatial ar-
rangement and donor/acceptor characteristics of their func-
tionalities. As a result, the architectures of the D1 and D2
desmotropes differ in their primary supramolecular synthon.
According to a recent CSD search[45] and previous literature
ꢀ
tense O1 H1···N2 hydrogen bond is formed, which is stabi-
lized to some extent by the negative charges of O2 and N2.
ꢀ
Reorientation of the O1 H1 group leads to two unstable
enol conformers (Figure 5a,v and vi), which do not have sig-
nificant intramolecular interactions.
Recalculating the energies of all the obtained conformers
by a polarizable continuum model (PCM) gives an estima-
tion of the stability of each conformer in a particular solu-
tion.[44] The electric permittivities used in these calculations
were chosen in accordance with solution experiments. As
can be seen in Figure 5b, the keto forms are more stabilized
with increasing solvent permittivity than their corresponding
enol forms. Also, the structures that have free polar groups
(v and vi) are stabilized somewhat more than the systems
with intramolecular hydrogen bonds.
reports,[46] it could be expected that the intermolecular O
H···N(py) hydrogen bond represents the most stable syn-
ꢀ
ꢀ
thon, even in the presence of the competing O H···O
hydrogen bond. However, bond rearrangement, induced by
PT from O2 to N2, increases the partial negative charge of
G
ꢀ
O2 and may explain the formation of the O1 H1···O2 hy-
drogen bond in the crystal structures of the D2 forms in-
ꢀ
Discussion and Conclusion
stead of the expected O1 H1···N1ACTHUGNETRNNU(G pyridine) hydrogen bond
found in D1 architectures. It has been postulated that the
stronger hydrogen bond (shorter N···O distance) is associat-
ed with the less stable tautomer.[19e,f,47,48] Our structural stud-
ies revealed that the N···O distance in the D1 forms is
longer than in the D2 forms, which should, according to this
geometric criterion, render the enol-imino tautomer more
stable. The stability of the enol forms with respect to their
Depending upon the solvent, H2L crystallizes in four solid
phases. These phases were obtained either as concomitant
mixtures (even of different desmotropes) or occasionally as
pure phases. In general, polar solvents favor the crystalliza-
tion of the enol-imino forms D1a (orange) and D1b
(yellow). On the other hand, solvents of lower polarity yield
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5627

ˇ ´
M. Rubcic et al.
keto tautomers in the gas phase was confirmed by computa-
tional studies.
features of Schiff bases, intramolecular N···H···O hydrogen
bonds and solution- and solid-state PT. Our findings empha-
size the importance of intermolecular forces in the stabiliza-
tion of the less stable keto tautomer in the solid state.
Therefore we anticipate that control, not only of the tauto-
meric form, but also of potential solid-state tautomeric inter-
conversions, would be accessible through careful selection of
the donor/acceptor functionalities of the peripheral groups.
In the case of Schiff bases, this is directly related to the ther-
mochromic and photochromic properties of the materials.
Taken together, the explored system could serve as a refer-
ence system in future studies for tautomerism and related
phenomena.
NMR and UV/Vis solution studies revealed that all the
solid forms equilibrate immediately upon dissolution. The
enol-imino tautomer is dominant in all solvents, especially
weakly polar and aprotic solvents. The fraction of the keto
tautomer is higher in alcohols, which is consistent with pre-
vious studies on similar systems in which polar and protic
solvents have also stabilized the more polar keto tauto-
mer.[8c,49] However, PCM calculations showed no significant
differences in the stabilization of different tautomeric forms
with increase in polarity of the dielectric continuum. These
findings indicate that the solvent influence on the keto-enol
equilibrium is mainly caused by hydrogen bonding with the
solvent molecules, the formation of which was detected by
temperature-dependent NMR measurements. The fact that
both polymorphs of the keto-amino desmotrope have been
crystallized from weakly polar solvents can be tentatively
explained by a lower solubility of the keto tautomer.
In the system presented herein we applied various sol-
vent-free methods to investigate the possibility of solid-state
PT in all the phases. These techniques include the manipula-
tion of the solid sample by varying the temperature, vapor
diffusion, or mechanochemical methods.[50] In the case of
desmotropically related compounds, only a few articles have
reported unambiguous solid-state interconversions.[12d,e,h,14]
Thermal and solvent-free treatments (grinding and methanol
vapor diffusion) of the D1b, D2a, and D2b phases induced
phase transformations exclusively to D1a (Scheme 2). In the
cases of the D2 desmotropes, transformations are accompa-
nied by a PT to change the tautomeric form. Conversion in
the opposite direction, from D1!D2, was not observed.
The thermodynamic stability of D1a at room temperature
was further confirmed by competitive slurry experiments.
Such ease with which metastable tautomeric phases convert
to the stable one raises caution in the way in which the solid
form of certain tautomers should be handled, stored, and ex-
amined (for example, active pharmaceutical ingredients in
the pharmaceutical industry). It could also be the main
reason why more desmotropes have not been discovered
and reported.
Experimental Section
Synthesis of 2-[(2-hydroxybenzylidene)amino]pyridin-3-ol (H2L), forms
D1a and D1b: Equimolar amounts (0.82 mmol) of 2-amino-3-hydroxy-
pyridine and salicylaldehyde were added to methanol for D1a or n-prop-
anol for D1b (both 5 mL) and heated at reflux for 1 h. From methanol,
brown-orange crystalline D1a was isolated after standing at room tem-
perature for 12 h. Yield: 61%. The identical product was obtained (con-
firmed by PXRD) from acetonitrile, but in significantly lower yield
(20%). Form D1a: elemental analysis calcd (%) for C12H10N2O2: C
67.28, H 4.71, N 13.08; found: C 67.11, H 4.50, N 12.97; IR (KBr): n˜ =
ꢀ
ꢀ
ꢀ
3466 (br, nO H(py)), 3057 (w), 2887 (w, nC H), 2441 (br, nO H-
AHCTUNGTRENNUNG
uct was obtained by microwave syntheses from methanol, ethanol, and n-
propanol. D1b: Yellow platelike crystalline product deposited after
standing at room temperature overnight. Yield: 72%. The identical prod-
uct was obtained (confirmed by PXRD) from n-hexanol, but in a signifi-
cantly lower yield (8%). Form D1b: elemental analysis calcd (%) for
C12H10N2O2: C 67.28, H 4.71, N 13.08; found: C 67.34, H 4.36, N 13.11;
ꢀ
ꢀ
ꢀ
IR (KBr): n˜ =3445 (br, nO H), 2992 (w), 2851 (w, nC H), 2434 (nO H-
AHCTUNGTRENNUNG
Synthesis of 6-[(3-hydroxypyridin-2-ylamino)methylene]cyclohexa-2,4-di-
enone (H2L), forms D2a and D2b: The syntheses were performed by fol-
lowing the procedure described above, except for the choice of solvents.
Both forms were obtained in low yields. Form D2a was deposited from
tert-butanol as dark-red platelike crystals after the volume of the solvent
had been reduced to a third under vacuum. Yield: 30%. D1a emerged
after standing overnight at room temperature. D2b was obtained from
isopropanol, in a concomitant mixture with D1a, as dark-red prismatic
crystals (maximum yield 9%). The crystals of pure D2b were separated
mechanically. Recrystallization of D1a from tert-butanol randomly yield-
ed D2a or D2b, in some cases as pure products and in some cases as
a concomitant mixture with D1a. Form D2a: elemental analysis calcd
(%) for C12H10N2O2: C 67.28, H 4.71, N 13.08; found: C 67.60, H 4.55, N
In the case of the keto forms, solid-state PT induced by
heating is also accompanied by a change in color from red
to yellow. Although these transformations are endothermic,
we can speculate why the reverse transformation was not
observed upon cooling. This may be due to the required
under cooling for nucleation of the low-temperature
phase[51] leaving the system with insufficient energy to break
ꢀ
ꢀ
13.04); IR (KBr): n˜ =3461 (br, nO H), 3049 (w), 2891 (w, nC H), 2520
(RAHB)), 1623 (s), 1571 (s), 1525 cmꢀ1 (s, nC=O, nC=C, and
ꢀ
(nN H
nC=N). Form D2b: elemental analysis calcd (%) for C12H10N2O2: C
67.28, H 4.71, N 13.08; found: 67.42, H 4.68, N 13.21; IR (KBr): n˜ =3435
ꢀ
ꢀ
ꢀ
(br, nO H), 3055 (w), 2921 (w, nC H), 2485 (nN H
ACHTUNGTREN(NUGN RAHB)), 1617 (s),
1574 (s), 1524 cmꢀ1 (s, nC=O, nC=C, and nC=N).
ꢀ
the intermolecular O1 H1···N1 hydrogen bond and to
SS NMR experiments: Experiments on D1a, D1b, and D2a were per-
formed at Bruker UK on a standard-bore Bruker Avance III spectrome-
ter operating at 500.13 MHz for 1H NMR and at 125.77 MHz for 13C
NMR with a Bruker 4 mm double-resonance probe spinning at 10 kHz.
The CH2 site of powdered a-glycine was used as an external reference at
43.1 ppm with respect to TMS. The 13C NMR spectra were acquired by
using cross-polarization for 4 ms at a rf field of approximately 60 kHz
with SPINAL-64 1H decoupling[52] of 90 kHz during acquisition. The
spectrum of D1b was acquired in 4096 scans by using a 3 s recycle delay.
The spectrum of D1a was acquired in 64 scans by using a 6 s recycle
ꢀ
rotate around the C1 N2 bond to form the new intermolec-
ꢀ
ular O1 H1···O2 synthon. Nonetheless, although the conver-
sion between two desmotropes is not reversible, it can be
considered that the system exhibits thermochromic behav-
ior.
To summarize, the investigated system of two polymorph-
ic pairs of functional keto-enol desmotropes has provided
a platform for a thorough investigation of the two important
5628
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

Desmotropy and Polymorphism
FULL PAPER
delay. The spectrum of D2a was acquired in 48 scans by using a 12 s recy-
cle delay. When nonquaternary suppression was applied, the dephasing
period was 60 ms. As a result of the large quantity of sample required for
the SS NMR experiment such measurements were not performed on
D2b. Nonetheless, single-crystal X-ray diffraction data offer unambigu-
ous evidence that the molecules of D2b are in the keto-amino form in
the solid state.
(Rint =0.0849), R1 =0.0400 (I>2s(I)), wR(F2)=0.0730 (I>2s(I)), final
R1 =0.0838 (all data), wR(F2)=0.0993 (all data), S=0.817.
Crystal data for D2a: C12H10N2O2, Mr =214.22, triclinic, a=8.3501(1), b=
10.3548(3),
c=12.3782(3) ꢃ,
a=106.909(2),
b=89.293(2),
g=
102.028(2)8, V=1000.17(4) ꢃ3, T=293 K, space group P1, Z=4, l-
¯
AHCTNUGTREN(GNNU CuKa1)=1.54056 ꢃ, starting angle (2q)=48, final angle (2q)=708, step
width (2q)=0.008538, Rexp =0.010, Rp =0.022, Rwp =0.030, R(F2)=0.010,
S=2.84. Crystal data for D2b: orthorhombic, a=8.5321(9), b=
9.9649(10), c=12.3045(13) ꢃ, V=1046.15(19) ꢃ3, T=293 K, space group
Computational studies: All calculations were undertaken with the Gaus-
sian 09 quantum chemistry software package.[53] For implicit solvation,
a polarizable continuum model using the integral equation formalism
(IEF-PCM)[44] was used to model the following solvents: toluene (er =
2.37), chloroform (4.71), tetrachloroethane (7.10), dichloromethane
(8.93), tert-butanol (12.0), acetone (20.49), ethanol (24.85), methanol
(32.61), acetonitrile (35.69), and dimethyl sulfoxide (46.83).
P212121, Z=4, mACTHNUTRGNEUNG
(MoKa)=0.095 mmꢀ1, 4558 reflections measured, 1196 in-
dependent reflections (Rint =0.0234), R1 =0.0309 (I>2s(I)), wR(F2)=
0.0698 (I>2s(I)), R1 =0.0474 (all data), wR(F2)=0.0728 (all data), S=
0.938.
PXRD: Patterns used for qualitative analysis of the samples were collect-
ed on a Philips PW 3710 diffractometer with CuKa radiation, a flat plate
sample on a zero background in Bragg–Brentano geometry, tension
40 kV, current 40 mA. The patterns were collected in the 2q range of 4–
408 with a step size of 0.028 and 1.0 s counting per step. High-resolution
X-ray powder diffraction patterns (Bruker D8 ADVANCE in Debye–
Scherrer geometry with Vꢄntag-1 position sensitive detector with a 68
Acknowledgements
´
´
We are thankful to Prof. Marina Cindric, Dr. Dalibor Milic, and Dr Nada
¯
ˇ
´
Doslic for critical discussions and helpful suggestions. We are also grate-
ˇ
´
ful to Prof. Snezana Miljanic for the help with TR-IR measurements.
These results are based on work financed by the Ministry of Science,
Education and Sports of the Republic of Croatia (Grant Nos. 119-
1191342-1082, 119-1193079-1084, 119-1191342-2960, 119-1191342-1334,
and 098-0352851-2921) and the Croatian Science Foundation (Grant No.
HrZZ-02.04/23).
opening angle and CuKa1 radiation from a primary GeACHTUNTRGNEU(GN 111) Johansson-
type monochromator, step size of 0.00858 in 2q) were used for the struc-
ture solution of D2a and the Rietveld refinements of D2a and D1a.
Samples were gently crushed to a fine powder in an agate mortar and
packed in 0.5 mm borosilicate glass capillaries. The capillaries were rotat-
ed during data collection to obtain better particle statistics. Selected crys-
tallographic and refinement data for structures D1a and D2a obtained
by the X-ray powder diffraction experiments are reported in Table S1.
All calculations were performed by using the program Topas.[54]
[1] a) F. A. Carey in Organic Chemistry, 5th Ed. (Ed.: S. R. Ober-
broeckling), McGraw-Hill, New York, 2003; b) IUPAC Compendi-
um of chemical terminology (electronic version). http://goldbook.iu-
pac.org/T06252.html; c) R. A. Sayle, J. Comput.-Aided Mol. Des.
2010, 24, 485–496.
[2] a) A. R. Katritzky, J. M. Lagowski, in Advances in Heterocyclic
Chemistry, Vol. 1, (Ed.: A. R. Katritzky), Academic Press, New
York, 1963, pp. 311–338; b) J. Elguero, A. R. Katritzky, O. V. Denis-
ko, in Advances in Heterocyclic Chemistry, Vol 76, (Ed.: A. R. Ka-
tritzky), Academic Press, San Diego, 2000, pp. 2–64; c) J. Elguero,
C. Marzin, A. R. Katritzky, P. Linda in Advances in Heterocyclic
Chemistry: Supplement 1, (Eds.: A. R. Katritzky, A. J. Boulton),
Temperature-resolved PXRD measurements were performed on the
Bruker D8 laboratory diffractometer described above and equipped with
a capillary furnace (mri Physikalische Gerꢅte GmbH). The sample was
placed in a 0.5 mm borosilicate glass capillary. Data collection of each
pattern was set to start after the desired temperature was reached, which
was then held constant during the data collection of each pattern. A
heating rate of 0.58Csꢀ1 was used and data was collected over 30 min for
each pattern in all in situ heating experiments. Patterns were collected in
the 2q region of 5–358.
Single-crystal X-ray diffraction experiments: X-ray diffraction was per-
formed with an Oxford Diffraction Xcalibur CCD diffractometer with
graphite-monochromated MoKa radiation at 293 K by using w (for D1b
and D2b) and f (for D1b) scans. Because all the spectroscopic measure-
ments and high-resolution PXRD data collections were performed at
room temperature, single-crystal XRD experiments were conducted on
D1b and D2b at the same temperature. Details of data collection and
crystal structure refinement are presented in Table S2. The CrysAlis
CCD and CrysAlis RED[55] programs were employed for data collection,
cell refinement, and data reduction. The structures were solved by direct
methods. The refinements were carried out by full-matrix least-squares
methods based on F2 values against all reflections including anisotropic
displacement parameters for all non-hydrogen atoms. The positions of
hydrogen atoms (except for H1 and H2, which were found in the differ-
ence Fourier map) were positioned geometrically and refined by applying
the riding model. Calculations were performed with SHELXS97[56] and
SHELXL97[57] (both operating within the WinGX[58] program package).
The geometries were calculated by using PLATON[59] and PARST,[60] and
the molecular graphics were created with ORTEP[27] and Mercury.[31] Se-
lected bond distances and valence angles are listed in Table S3.
ˇ
Academic Press, New York 1976; d) B. Stanovnik, M. Tisler, A. R.
Katritzky, O. V. Denisko in Advances in Heterocyclic Chemistry,
Vol. 91, (Ed.: A. R. Katritzky), Academic Press, San Diego, 2006,
pp. 1–108.
[3] a) J. D. Watson, F. H. C. Crick, Nature 1953, 171, 737–738; b) J. D.
Watson, F. H. C. Crick, Nature 1953, 171, 964–967; c) L. Pauling,
The Nature of the Chemical Bond, 3rd Ed., Cornell University Press,
Ithaca, New York, 1960, 502–504; d) P. O. Lçwdin, Rev. Mod. Phys.
1963, 35, 724–732; e) H. T. Miles, Proc. Natl. Acad. Sci. U.S. 1961,
47, 791–802; f) J. F. McConnell, B. D. Sharma, R. E. Marsh, Nature
1964, 203, 399–400; g) I. Alkorta, P. Goya, J. Elguero, S. P. Singh,
Natl. Acad. Sci. Lett. 2007, 30, 139–159; h) A. R. Katritzky, C. D.
Hall, B. E.-D. M. El-Gendy, B. Draghici, J. Comput.-Aided Mol. Des.
2010, 24, 475–484.
´
[4] a) D. Milic, T. V. Demidkina, N. G. Faleev, R. S. Phillips, D. Matkov-
ˇ
´
´
ic-Calogovic, D. Antson, J. Am. Chem. Soc. 2011, 133, 16468–
16476; b) M. D. Toney, Arch. of Biochem. Biophy. 2005, 433, 279–
287; c) Y.-l. Lin, J. Gao, Biochemistry 2010, 49, 84–94.
[5] a) A. R. Katritzky, J. M. Lagowski in Advances in Heterocyclic
Chemistry, Vol. 2, (Ed.: A. R. Katritzky), Academic Press, New
York, 1963, pp. 1–26; b) S. Gronowitz, R. A. Hoffman, Ark. Kemi,
1960, 15, 499–512; c) A. R. Katritzky, Chimia 1970, 24, 134–146.
[6] a) Y. C. Martin, J. Comput.-Aided Mol. Des. 2009, 23, 693–704;
b) H. Brandstetter, F. Grams, D. Glitz, A. Lang, R. Huber, W. Bode,
H.-W. Krell, J. Engh, Biol. Chem. 2001, 276, 17405–17412; c) X. J.
Yan, P. Day, T. Hollis, A. F. Monzingo, E. Schelp, J. D. Robertus,
G. W. A. Milne, S. M. Wang, Proteins Struct. Funct. Genet. 1998, 31,
33–41.
Crystal data for D1a: C12H10N2O2, Mr =214.22, monoclinic, a=
4.88823(6), b=18.5684(4), c=11.5673(2) ꢃ, b=102.183(1)8, V=
1026.28(3) ꢃ3, T=293 K, space group P21/c, Z=4, l
ACHTUNGTRENNUNG
R
mACHTUNGTRENNUNG
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5629

ˇ ´
M. Rubcic et al.
[7] a) P. M. Bhatt, G. R. Desiraju, Chem. Commun. 2007, 2057–2059;
b) Z. Bçcskei, K. Simon, R. Rao, A. Caron, C. A. Rodger, M.
Bauer, Acta Crystallogr. C: Cryst. Struct. Commun. 1998, 54, 808–
810; c) M. Bauer, R. K. Harris, R. C. Rao, D. C. Apperley, C. A.
Rodger, J. Chem. Soc. Perkin Trans. 2 1998, 475–481; d) L. A. Filip,
[21] a) J. H. Day, Chem. Rev. 1963, 63, 65–80; b) J. H. Day, Chem. Rev.
1968, 68, 649–657; c) K. Nassau in The Physics and Chemistry of
Color, 2nd ed., Wiley, New York, 2001.
[22] a) A. Samat, V. Lokshin in Organic Photochromic and Thermochro-
mic Compounds, Vol. 2, (Eds.: J. C. Crano, R. J. Guglielmetti),
Kluwer Academic, New York, 1999, pp. 415–466; b) P. Bamfield in
Chromic Phenomena: Technological Apllications of Colour Chemis-
try, Royal Society of Chemistry, Cambridge, 2001.
[23] a) M. D. Cohen, G. M. Schmidt, J. Phys. Chem. 1962, 66, 2442–2445;
b) M. D. Cohen, G. M. Schmidt, S. Flavian, J. Chem. Soc. 1964,
2041–2051; c) J. Harada, H. Uekusa, Y. Ohashi, J. Am. Chem. Soc.
1999, 121, 5809–5810; d) K. Ogawa, Y. Kasahara, Y. Ohtani, J.
Harada, J. Am. Chem. Soc. 1998, 120, 7107–7108.
ˇ
ˇ
M. R. Caira, S. I. Farcas¸, M. T. Bojit¸a, Acta Crystallogr Sect. C:
Cryst. Struct. Commun. 2001, 57, 435–436; e) A. J. Blake, X. Lin, M.
Schrçder, C. Wilson, R. X. Yuan, Acta Crystallogr Sect. C: Cryst.
Struct. Commun. 2004, 60, o226-o228; f) M. Mirmehrabi, S. Rohani,
K. S. K. Murthy, B. Radatus, J. Cryst. Growth 2004, 260, 517–526.
[8] a) F. H. Kamounah, L. Antonov, V. Petrov, G. J. van der Zwan, J.
Phys. Org. Chem. 2007, 20, 313–320; b) D. L. Nedeltcheva, Anto-
nov, J. Phys. Org. Chem. 2009, 22, 274–281; c) A. Afkhami, F. Kha-
javi, H. Khanmohammadi, Anal. Chim. Acta 2009, 634, 180–185.
[9] a) T. Steiner, G. Koellner, Chem. Commun. 1997, 1207–1208; b) H.
Pizzala, M. Carles, W. E. E. Stone, A. Thevand, J. Mol. Struct. 2000,
526, 261–268; c) B. D. Sharma, J. F. McConnell, Acta Crystallogr.
1965, 19, 797–806; d) F. Betchel, J. Gaultier, C. Hauw, Cryst. Struct.
Commun. 1973, 2, 469–472.
[24] a) T. Inabe, New J. Chem. 1991, 15, 129–136; b) Y. Zhang, C.-Y.
Zhao, W.-H. Fang, X.-Z. You, Theor. Chem. Acc. 1997, 96, 129–134;
c) K. Nakatani, J. A. Delaire, Chem. Mater. 1997, 9, 2682–2684.
[25] a) W. C. McCrone in Physics and Chemistry of the Organic Solid
State, Vol. 2, (Eds.: D. Fox, M. M. Labes, A. Weissberger), Inter-
science Publishers, New York, 1965, 725–767; b) J. K. Haleblian, W.
McCrone, J. Pharm. Sci. 1969, 58, 911–929; c) J. K. Haleblian, J.
Pharm. Sci. 1975, 64, 1269–1288; d) T. L. Threlfall, Analyst 1995,
120, 2435–2460; e) J. Bernstein in Polymorphism in Molecular Crys-
tals, 1st ed., Oxford University Press, New York, 2002; f) A. Gavez-
zotti, J. Pharm. Sci. 2007, 96, 2232–2241; g) J. Bernstein, Cryst.
Growth Des. 2011, 11, 632–650.
[10] a) P. Jacobson, Ber. Dtsch. Chem. Ges. 1887, XX, 1732; b) A.
Hantzsch, F. Herrmann, Ber. Dtsch. Chem. Ges. 1887, XX, 2803;
c) P. Jacobson, Ber. Dtsch. Chem. Ges. 1888, XXI, 2628.
[11] J. Elguero, Cryst. Growth Des. 2011, 11, 4731–4738.
[12] a) C. Foces-Foces, A. L. Llamas-Saiz, R. M. Claramunt, C. Lopez, J.
Elguero, J. Chem. Soc. Chem. Commun. 1994, 1143–1145; b) P.
Csomꢆs, L. Fodor, A. Csampai, P. Sohar, Tetrahedron 2010, 66,
3207–3213; c) T. Holczbauer, L. Fabian, P. Csomos, L. Fodor, A.
Kalman, CrystEngComm 2010, 12, 1712–1717; d) M. A. Garcia, C.
Lopez, R. M. Claramunt, A. Kenz, M. Pierrot, J. Elguero, Helv.
Chim. Acta 2002, 85, 2763–2776; e) M. R. Chierotti, L. Ferrero, N.
Garino, R. Gobetto, L. Pellegrino, D. Braga, F. Grepioni, L. Maini,
Chem. Eur. J. 2010, 16, 4347–4358; f) A. Madinaveitia, E. Olay, An.
Soc. Esp. Fis. Quim. 1933, 31, 134–138; g) F. Reviriego, I. Alkorta,
J. Elguero, J. Mol. Struct. 2008, 891, 325–328; h) M. U. Schmidt, J.
Brꢇning, J. Glinnemann, M. W. Hutzler, P. Mçrschel, S. N. Ivashev-
skaya, J. van de Streek, D. Braga, L. Maini, M. R. Chierotti, R. Go-
betto, Angew. Chem. 2011, 123, 8070–8072; Angew. Chem. Int. Ed.
2011, 50, 7924–7926; i) A. J. Cruz-Cabeza, C. R. Groom, CrystEng-
Comm 2011, 13, 93–98.
[13] a) J. Elguero, R. Jacquier, G. Tarrago, Bull. Soc. Chim. Fr. 1966,
2990–2995; b) J. Elguero, R. Jacquier, G. Tarrago, Bull. Soc. Chim.
Fr. 1967, 3772–3779.
[14] G. R. Desiraju, J. Chem. Soc. Perkin Trans. 2 1983, 1025–1030.
[15] a) S. C. Bhatia, J. M. Bindlish, A. R. Saini, P. C. Jain, J. Chem. Soc.
Dalton Trans. 1981, 1773–1779; b) J. Costamagna, J. Vargas, R. La-
torre, A. Alvarado, G. Mena, Coord. Chem. Rev. 1992, 119, 67–88;
c) E. Hadjoudis, I. M. Mavridis, Chem. Soc. Rev. 2004, 33, 579–588.
ˇ
´
ˇ
´
ˇ ´
´
[26] K. Uzarevic, M. Rubcic, I. ꢀilovic, Z. Kokan, D. Matkovic-Calogov-
´
´
ic, M. Cindric, Cryst. Growth Des. 2009, 9, 5327–5333.
[27] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565–566.
[28] http://www.povray.org/.
[29] In our case, the structures of D1a and D2a were solved and refined
on the basis of the high-resolution PXRD data (see the Supporting
Information). Because the exact positions of protons are difficult to
ascertain from PXRD data and restraints on geometrical parameters
were used during structure refinement, the geometries of hydrogen
bonds could deviate from true values. Therefore we discuss in detail
only the hydrogen-bonding patterns of the other two desmotropic
phases (D1b and D2b), which were solved on the basis of the
single-crystal XRD data. The tautomeric forms of all four phases
were unambiguously established by correlation between the IR
spectra of the polymorphic forms and the 13C CP MAS NMR data.
[30] This system provides us with a good opportunity to discuss terminol-
ogy. Although the term desmotropy is clearly defined and has been
present in the literature for more then a century, modern literature
refers to desmotropy as tautomeric polymorphism. The meaning of
the latter term is somewhat unclear, however, because it has not
been used to refer to only desmotropes, but also to crystals compris-
ing tautomeric mixtures.[7d,9] However, had the desmotropes in our
system been termed tautomeric polymorphs, their polymorphs
would be described with a rather cumbersome term, such as poly-
morphs of tautomeric polymorphs. One must bear in mind that,
unlike for “real” polymorphic phases, desmotropes are structural
isomers with different connectivity between the atoms. Accordingly,
molecules in D1 desmotropes are named 2-[(2-hydroxybenzyl-
idene)amino]pyridin-3-ol whereas in D2 they are named 6-[(3-hy-
droxypyridin-2-ylamino)methylene]cyclohexa-2,4-dienone. An excel-
lent survey of heterocyclic desmotropes has recently been written by
Elguero,[11] who additionally illuminates this “grey” area of solid-
state chemistry.
ˇ ´
ˇˇ ´
´
[16] A. Blagus, D. Cincic, T. Friscic, B. Kaitner, V. Stilinovic, Maced. J.
Chem. Chem. Eng. 2010, 27, 117–138.
[17] J. W. Ledbetter Jr., J. Phys. Chem. 1968, 72, 4111–4115.
[18] a) T. Dziembowska, Pol. J. Chem. 1998, 72, 193–209; b) J. M.
Fernꢈndez-G, F. R. Portilla, B. Q. Garcia, R. A. Toscano, R. Salcedo,
´
J. Mol. Struct. 2001, 561, 197–207; c) B. Kamienski, W. Schilf, T.
Dziembowska, Z. Rozwadowski, A. Szady-Chełmieniecka, Sol. St.
´
Nucl. Magn. Res. 2000, 16, 285–289; d) M. Gavranic, B. Kaitner, E.
ˇ
´
Mestrovic, J. Chem. Crystallogr. 1996, 26, 23–28.
[19] a) M. L. Huggins, J. Org. Chem. 1936, 1, 407–456; b) G. Gilli, F. Bel-
lucci, V. Ferretti, V. Bertolasi, J. Am. Chem. Soc. 1989, 111, 1023–
1028; c) P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J. Am. Chem. Soc.
1994, 116, 909–915; d) P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J.
Am. Chem. Soc. 2000, 122, 10405–10417; e) P. Gilli, V. Bertolasi, L.
[31] C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields,
R. Taylor, M. Towler, J. van de Streek, J. Appl. Crystallogr. 2006, 39,
453–457.
ˇ
Preto, A. Lycka, G. Gilli, J. Am. Chem. Soc. 2002, 124, 13554–
[32] H. Pizzala, M. Carles, W. E. E. Stone, A. Thevand, J. Chem. Soc.,
Perkin Trans. 2 2000, 2, 935–939.
[33] a) T. Dziembowska, M. Szafran, A. Katrusiak, Z. Rozwadowski, J.
Mol. Struct. 2009, 929, 32–42; b) Z. Rozwadowski, T. Dziembowska,
Magn. Reson. Chem. 1999, 37, 274–278; c) R. S. Salman, F. S. Ka-
mounah, Spectroscopy 2003, 17, 747–752.
13567; f) P. Gilli, V. Bertolasi, L. Preto, L. Antonov, G. Gilli, J. Am.
Chem. Soc. 2005, 127, 4943–4953.
[20] a) Photochromism: Molecules and Systems, (Eds.: H. Dꢇrr, H.
Bouas-Laurent), Elsevier, Amsterdam, 1990, and references therein;
b) J. R. Scheffer, R. P. Pokkuluri in Photochemistry in Organised and
Constrained Media, (Ed.: V. Ramamurthy), Wiley, New York, 1990,
p. 185.
[34] a) O. Domꢉnguez, B. Rodrꢉguez-Molina, M. Rodrꢉguez, A. Ariza, N.
Farfꢈnc, R. Santillan, New J. Chem. 2011, 35, 156–164; b) M. Jawor-
5630
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5620 – 5631

Desmotropy and Polymorphism
FULL PAPER
ska, P. B. Hrynczyszyn, M. Wełniak, A. Wojtczak, K. Nowicka, G.
8, 4533–4545; d) S. Mohamed, D. A. Tocher, M. Vickers, P. G. Kara-
mertzanis, S. L. Price, Cryst. Growth Des. 2009, 9, 2881–2889.
[47] a) K. Lammertsma, P. V. Bharatam, J. Org. Chem. 2000, 65, 4662–
4670; b) J. A. Long, N. J. Harris, K. Lammertsma, J. Org. Chem.
2001, 66, 6762–6767.
[48] a) G. Buemi, F. Zuccarello, P. Venuvanalingam, M. Ramalingam,
Theor. Chem. Acc. 2000, 104, 226–234; b) G. Buemi, J. Mol. Struct.:
THEOCHEM 2000, 499, 21–34; c) G. Buemi, F. Zuccarello, J. Mol.
Struct.: THEOCHEM 2002, 581, 71–85.
[49] W. M. F. Fabian, L. Antonov, D. Nedeltcheva, F. S. Kamounah, P. J.
Taylor, J. Phys. Chem. A 2004, 108, 7603–7612.
[50] D. Braga, D. DꢂAddario, S. L. Giaffreda, L. Maini, M. Polito, F. Gre-
pioni, Top. Curr. Chem. 2005, 254, 71–94.
´
Krasinski, H. Kassassir, W. Ciesielski, M. J. Potrzebowski, J. Phys.
Chem. A 2010, 114, 12522–12530; c) A. Makal, W. Schilf, B. Ka-
´
mienski, A. Szady-Chelmieniecka, E. Grech, K. Wꢆzniak, Dalton
Trans. 2011, 40, 421–430; d) J. Sitkowski, L. Stefaniak, I. Wawer, L.
Kaczmarek, G. A. Webb, Solid State Nucl. Magn. Reson. 1996, 7,
83–84; e) S. R. Salman, J. C. Lindon, R. D. Farrant, T. A. Carpenter,
Magn. Reson. Chem. 1993, 31, 991–994; f) H. Saitꢊ, I. Ando, A.
Ramamoorthy, Prog. Nucl. Magn. Reson. Spectrosc. 2010, 57, 181–
228.
[35] a) Isopropanol was chosen because all phases proved stable in the
isopropanolic suspension and no transformations were observed in
the absence of seed crystals (in contrast to the methanolic suspen-
sion, in which all phases transform immediately into D1a); b) G. P.
Stahly, Cryst. Growth Des. 2007, 7, 1007–1026; c) K. Jarring, T. Lars-
son, B. Stensland, I. Ymꢋn, J. Pharm. Sci. 2006, 95, 1144–1161;
d) A. Storey, I. Ymꢋn, in Solid State Characterization of Pharma-
ceuticals, 1st ed., Wiley-Blackwell, Chichester, 2011 pp. 20–30.
[36] a) M. Ziołek, K. Filipczak, A. Maciejewski, Chem. Phys. Lett. 2008,
[51] Y. Mnyukh, Fundamentals of Solid State Phase Transformations, Fer-
romagnetism and Ferroelectricity, Authorhouse, 2001.
[52] B. M. Fung, A. K. Khitrin, K. Ermolaev, J. Magn. Reson. 2000, 142,
97–101.
[53] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢍ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford CT, 2009.
[54] Topas, version 4.1. Bruker-AXS, Karlsruhe, Germany.
[55] Oxford Diffraction, Oxford Diffraction, Xcalibur CCD system,
CRYSALIS Software system, Version 1.170, 2003.
[56] G. M. Sheldrick, SHELXS97, Program for the Solution of Crystal
Structures, University of Gçttingen, Germany, 1997.
[57] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[58] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[59] a) A. L. Spek, Acta Crystallogr. Sect. D 2009, 65, 148–155; b) A. L.
Spek, PLATON, A Multipurpose Crystallographic Tool, Utrecht
University, Utrecht, The Netherlands, 1998.
´
ˇ ´
464, 181–186; b) N. Galic, Z. Cimerman, V. Tomisic, Anal. Chim.
´
ˇ ´
Acta 1997, 343, 135–143; c) N. Galic, Z. Cimerman, V. Tomisic,
Spectrochim. Acta Part A 2008, 71, 1274–1280; d) R. M. Issa, A. M.
Khedr, H. F. Rizk, Spectrochim Acta A 2005, 62, 621–629; e) W.
Bruyneel, T. J. Charette, E. De Hoffman, J. Am. Chem. Soc. 1966,
88, 3808–3813; f) H. ꢌnver, M. Yıldız, N. Ocak, T. N. Durlu, J.
Chem. Crystallogr. 2008, 38, 103–108; g) H. ꢌnver, M. Yıldız, A.
Kiraz, ꢍ. ꢍzgen, J. Chem. Crystallogr. 2009, 39, 17–23.
[37] R. M. Issa, A. M. Khedr, H. Rizk, J. Chin. Chem. Soc. 2008, 55,
875–884.
[38] N. S. Golubev, S. N. Smirnov, P. M. Tolstoy, S. Sharif, M. D. Toney,
G. S. Denisov, H. H. Limbach, J. Mol. Struct. 2007, 844, 319–327.
[39] G. O. Dudek, G. P. Volpp, J. Am. Chem. Soc. 1963, 85, 2697–2702.
[40] P. Przybylski, W. Lewandowska, B. Brzezinski, F. Bartl, J. Mol.
Struct. 2006, 797, 92–98.
[41] J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10,
6615–6620.
[42] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[43] R. Krishnan, J. S. Binkley, R. Seeger, J. Pople, J. Chem. Phys. 1980,
72, 650–654.
[44] G. Scalmani, M. J. Frisch, J. Chem. Phys. 2010, 132, 114110–114115.
ˇ ´
ˇ
´
´
ˇ
´
[45] I. Halasz, M. Rubcic, K. Uzarevic, I. ꢀilovic, E. Mestrovic, New J.
Chem. 2011, 35, 24–27.
[46] a) T. Steiner, Acta Crystallogr. Sect. B: Struct. Sci. 2001, 57, 103–
106; b) F. H. Allen, W. D. S. Motherwell, P. R. Raithby, G. P. Shields,
R. Taylor, New J. Chem. 1999, 23, 25–34; c) T. R. Shattock, K. K.
Arora, P. Vishweshwar, M. J. Zaworotko, Cryst. Growth Des. 2008,
[60] a) M. Nardelli, Comput. Chem. 1983, 7, 95–98; b) M. Nardelli, J.
Appl. Crystallogr. 1995, 28, 659–673.
Received: November 7, 2011
Published online: March 23, 2012
Chem. Eur. J. 2012, 18, 5620 – 5631
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5631