Content uploaded by Michał Henryk Jamróz
Author content
All content in this area was uploaded by Michał Henryk Jamróz on Aug 25, 2014
Content may be subject to copyright.
On two alizarin polymorphs†
Micha1 K. Cyra
nski,
a
Micha1 H. Jamr
oz,
b
Anna Rygula,
c
Jan Cz. Dobrowolski,
bd
qukasz Dobrzycki
*
a
and Malgorzata Baranska
*
ce
Received 17th August 2011, Accepted 23rd February 2012
DOI: 10.1039/c2ce06063a
For centuries alizarin has been used as natural pigment, yet, now it is also used in histochemistry and
dermatology, and as a dye for semiconductor solar cells. Here, the crystal structure of the previously
known form of alizarin is determined accurately and its new polymorph is discovered. The two
polymorphs crystallize into the same monoclinic crystal system. The previously known form is
centrosymmetric (P2
1
/c space group) whereas the new one is not (space group Pc). Both crystals are
twinned, disordered, and have the molecules packed into the crystal lattices in a comparable way.
However the powder patterns of both forms are visibly different. In addition the analysis of the
fingerprints of H/O intermolecular interactions based on Hirshfeld surfaces also indicates differences
in the packing. Yet, these similarities between two forms of alizarin make their vibrational spectra
hardly distinguishable.
Introduction
Alizarin, 1,2-dihydroxyantraquinone (Scheme 1), the natural red
textile dye known throughout the ancient world, isolated from
the roots of plants of the Madder family and leaves of the
Polynesian noni tree,
1–3
had been already used in red fabrics
found in the tomb of King Tutankhamun in Egypt.
4
From the
late 17th to the early 20th century, it was used to dye the British
‘‘redcoats’’.
5
In 1869, Caro, Lieberman, along with Graebe and
Perkin patented alizarin syntheses day after day, and since then,
the production of artificial alizarin by Perkin & Sons and by
BASF had driven the fabrication of alizarin from natural
sources.
6
Madder has long held medicinal uses. In Dioscorides’s P3rὶ
glh2 ἰasrikή2 (De Materia Medica), a ‘‘pharmacopeia’’ in use
from the 1st to 17th century, madder is called eruthrodanon and is
reported to be useful as a diuretic and when taken with melicrate
(honey water) in treatment of jaundice, sciatica, and paralysis.
Scheme 1 Structure, atom numbering and ring notation in a single
alizarin molecule.
a
Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw,
Poland. E-mail: dobrzyc@chem.uw.edu.pl; Fax: +48 22 822 28 92; Tel:
+48 22 822 02 11 ext. 360
b
Spectroscopy and Molecular Modeling Group, Industrial Chemistry
Research Institute, 8 Rydygiera Street, 01-793 Warsaw, Poland
c
Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060
Cracow, Poland. E-mail: baranska@chemia.uj.edu.pl; Fax: +48 12
6340515; Tel: +48 12 663 22 53
d
Laboratory for Theoretical Methods and Calculations, National
Medicines Institute, 30/34 Chełmska Street, 00-725 Warsaw, Poland
e
Jagiellonian Center for Experimental Therapeutics (JCET), Jagiellonian
University, 14 Bobrzynskiego Str., 30-348 Krakow, Poland
† Electronic supplementary information (ESI) available: Comparison of
PXRD patterns of polymorph I and of polymorph II. PXRD pattern and
theoretically calculated pattern of polymorph I. PXRD pattern and
theoretically calculated pattern of polymorph II. The Raman spectra of
the alizarin form I and form II crystals in the range 1700–1000 cm
1
.
The Raman spectra of the alizarin form I and form II crystals in the
range 1000–200 cm
1
. The IR spectra of the alizarin form I and form II
crystals in the range 1700–400 cm
1
. of the Raman theoretical
B3LYP/6-31++G** spectra of alizarin monomer and dimer.
Comparison of the IR theoretical B3LYP/6-31++G** spectra of
alizarin monomer and dimer. The n(OH) and n(CH) region of the
experimental IR and Raman spectra. Comparison of the Raman
experimental digitized and theoretical spectra of alizarin. Comparison
of the IR experimental digitized and theoretical spectra of alizarin.
Comparison of the IR theoretical harmonic and anharmonic
B3LYP/aug –cc-pVDZ spectra of the alizarin molecule. Observe that in
the anharmonic spectrum different bands are shifted by different
factors. The B3LYP/aug-cc-pVDZ calculated fundamental frequencies
of the alizarin molecule. The parameters of experimental FT-IR and
FT-Raman spectra of alizarin in polycrystalline powder. Definitions of
the local coordinates for PED analysis of alizarin calculated at the
B3LYP/aug-cc-pVDZ level. CCDC reference numbers 800713 and
800714. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2ce06063a
This journal is ª The Royal Society of Chemistry 2012 CrystEngComm, 2012, 14, 3667–3676 | 3667
Dynamic Article Links
C
<
CrystEngComm
Cite this: CrystEngComm, 2012, 14, 3667
www.rsc.org/crystengcomm
PAPER
For centuries, it has been known that the bones of animals that
ate madder turned pink or red.
7
Around 1900, alizarin was used
to detect calcium deposits in renal tissue and now alizarin and
alizarin red S (ARS, sodium salt of alizarine sulfonic acid) still
are used in histochemistry to highlight calcium and its deposits in
soft tissues,
8
for example in dermal calcium disorders (e.g.,
pseudoxanthoma elasticum and calcinosis cutis).
9
ARS and other
alizarin derivatives have potential uses in tumor, mesenchymal
connective tissue, and osteogenic compound characterization,
evaluation, and screening.
10
Alizarin has recently received much attention as a photo-
sensitizing dye for semiconductor solar cells, photographic
processes, and photodegradation of water polutants.
11–16
Both
experiments with different excitation wavelengths and theoretical
nonadiabatic ab initio molecular dynamic studies
11
have indi-
cated the position of dye excited state in the alizarin/TiO
2
system
to be close to the TiO
2
conduction band edge, while the already
known dye-sensitized semiconductors exhibited a resonance of
molecular donor state with a high density acceptor state(s) of
semiconductors.
12
In contrast to the increasing importance of alizarin as a dye,
medicine, and electronic material, relatively little is known about
its crystalline form. The alizarin crystal and molecular structure
was studied in 1960s by Guilhem,
17–19
who found that it crys-
tallizes in the monoclinic Pa space group with three molecules in
the asymmetric unit. However the quality of the structure has
been doubtful with the final R-factor yielding 26%. The recent
careful study on the morphology of alizarin crystals concluded
the crystals to be solvent influenced rather than exhibit
a polymorphism.
20
Recently, we demonstrated that either SERS or SERRS
spectra are observed for alizarin deposited at the surface of Ag
electrode, depending on the excitation line (488, 514.5 and 647.1
nm), and we interpreted the marker Raman bands by means of
B3LYP/PCM/6-31+G(d,2p)/LANL2DZ(Ag) calculations.
21
In this study we show that there are two polymorphs of aliz-
arin, both twinned and disordered. The Raman spectra of the
two polymorphs show no significant differences, whereas
although subtle differences can be found in n(OH), b(CH) and
s(OH) bands of the IR spectra, they rather cannot be used for
polymorph distinction. Indeed, the calculated anharmonic
vibrational spectra of monomer alizarin molecule showed that all
bands in crystals, but a few significantly disturbed by H-bonds,
can be very well elucidated. However, the subtle differences of
the experimental spectra are hardly reproducible by the spectra
of selected dimers.
Experimental
Crystallization and X-ray diffraction: The sample of alizarin was
purchased from Sigma-Aldrich. The suitable crystals were
obtained by a vapor growth (as described by Algra et al.
20
)orby
slow evaporation from ethanol, methanol, or toluene solutions.
Except from toluene, the crystals of polymorph I were obtained.
Their morphologies were the same as described earlier by Algra
et al.
20
i.e. long needles in the case of vapor evaporation or
triangle shapes obtained from alcohol solutions. In turn, the
crystallization from toluene led to the polymorph II, the form
which has never been extracted and structurally characterized.
The powder diffraction measurement indicated that the starting
material (purchased from Sigma-Aldrich) was composed solely
of this polymorph. Interestingly, the attempts of obtaining form
II by dissolving the crystals of the polymorph I in toluene and by
subsequent slow evaporation of the solvent were not successful.
Form I: The measurements of the crystal were performed on
a KM4CCD k-axis diffractometer with graphite-mono-
chromated Mo-Ka radiation (l ¼ 0.71073
A) at T ¼ 100(2) K.
Empirical correction for absorption was applied.
22
Data reduc-
tion and analysis were carried out with the Oxford Diffraction
programs.
23
The structure was solved by direct methods
24
and
refined using SHELXL
25
and WinGX.
26
The refinement was
based on F2. Scattering factors were taken from Tables 6.1.1.4
and 4.2.4.2 in ref. 27. The crystal is twinned by reticular pseu-
domerohedry. Data were integrated according to two orientation
matrices resulting in non-merged reflection file calculation.
During the structure refinement the HKLF 5 option was used.
The twin components ratio was refined at 0.527(3) : 0.473(3)
level. The structure is disordered and the one of the molecules is
split into two fragments with refined occupancy ratio equal to
0.611(5) : 0.389(5). Another molecule located at the center of
symmetry also indicates some disorder, however due to insuffi-
cient resolution it has not been split into two fragments. To
provide reasonable geometry of the less occupied fragment its
geometry was taken from a more occupied section and refined
iteratively as a rigid body only to acquire comparable geomet-
rical parameters in these two fragments. All hydrogen atoms
were located geometrically and their positions were refined using
a riding model. Hydrogen atoms of the hydroxyl group were
fixed to be coplanar with the alizarin molecules. All non-
hydrogen atoms of the more occupied fragment were refined
anisotropically with restraint providing their more spherical
shape (ISOR) whereas the atoms of less occupied fragment were
refined with the similarity constraint (SIMU). All isotropic
ADPs of hydrogen atoms were set to be 1.5 times bigger than
Ueq of corresponding heavy atoms.
Form II: The measurements of the crystal were performed on
a Oxford Diffraction, Xcalibur, Atlas CCD, Gemini, Ultra
diffractometer with mirror-monochromated CuKa radiation
(l ¼ 1.5418
A) at T ¼ 100(2) K. Empirical correction for
absorption was applied.
22
Data reduction and analysis were
carried out with the Oxford Diffraction programs.
23
The struc-
ture was solved by direct methods
24
and refined using SHELXL
25
and WinGX.
26
The refinement was based on F2. Scattering
factors were taken from Tables 6.1.1.4 and 4.2.4.2 in ref. 27.
The crystal is twinned by reticular pseudomerohedry. Data
were integrated according to two orientation matrices resulting
in non-merged reflection file calculation. During the structure
refinement the HKLF 5 option was used. The twin com ponents
ratio was refined at 0.5643(15) : 0.4357(15) level. The structur e
is disordered and the m olecule is split into two fragments with
refined occupancy ratio equal to 0.771(4) : 0.229(4). To
pro vide reasonable geometry of the less occupied fragment its
geometry was taken from a more occupied section and refined
ite rati vely as a rigid body only t o acquire comparable
geo metrical parameters in thes e two fragments. All hydrogen
atoms were located geometrically and their pos itio ns were
refined using a riding model except two hydrogen atoms of the
hydroxy l group in the more occupied fragment. In these cases
3668 | CrystEngComm, 2012, 14, 3667–3676 This journal is ª The Royal Society of Chemistry 2012
the torsion angles C–C–O–H were free to refine leading to
reasonable geometry with intramolecular hydrogen bond
formation. All non-hydrogen atoms of the les s occupied frag-
ment were refined isotropically. In the less occupied fragment,
10 non-hydrogen atoms were constrained to have comparable
values of ADPs. All isotropic ADPs of hydrogen atoms were
set to be either 1.2 or 1.5 times bigger than Ueq of corre-
spo nding heavy atoms. The structure is non-centrosymmetric,
however due to insufficient anomalous dispersion effect origi-
nated from crystal size, twinning and disorder the Friedel pairs
wer e merged.
Powder diffraction: The powder X-ray diffraction patterns of
both forms were recorded at 293 K on a Seifert HZG-4 auto-
mated diffractometer using Cu K
a1,2
radiation (1.5418
A). The
data were collected in the Brag–Brentano (q/2q) horizontal
geometry (flat reflection mode) between 4 and 40
(2q) in 0.04
steps, at 5 s step
1
. The optic of the HZG-4 diffractometer was
a system of primary Soller slits between the X-ray tube and the
fixed aperture slit of 2.0 mm. One scattered radiation slit of 2 mm
was placed after the sample, followed by the detector slit of
0.2 mm. The X-ray tube operated at 40 kV and 40 mA. Powder
XRD patterns were simulated from single-crystal data using the
program MERCURY.
28
IR and Raman measurements: FT-Raman Spectrometer
(Nicolet, Model NXR 9650) equipped with a Nd:YAG laser
(continuum-wave excitation at 1064 nm) and a liquid nitrogen
cooled germanium detector was used in the experiments. FT-IR
Spectrometer (Nicolet 5700) equipped with ATR diamond
crystal was used in order to collect IR spectra. The vibrational
spectra were registered at room temperature for established
polymorph forms of alizarin crystals. Raman measurements
(2000 scans) were performed with laser power of 500 mW and
spectral resolution of 4 cm
1
, from 100 to 4000 cm
1
, whereas IR
spectra were obtained in 400–4000 cm
1
range using 32 scans per
sample and 4 cm
1
resolution.
Computational
DFT calculations were performed by using the hybrid Becke
three-parameter Lee–Yang–Parr DFT B3LYP functional
29,30
and Dunning’s aug-cc-pVDZ basis set.
31,32
Reliability of B3LYP
functional in calculations of the ground state geometries has been
widely assessed
33
and the Dunning’s basis sets are known to be
adequate to describe organic molecules and their hydrogen-
bonded systems.
34
The alizarin dimers were calculated by using
6-31++G** basis set.
35–37
The stationary monomer and dimer
structures are found by ascertaining that all harmonic frequen-
cies are real. All calculations were carried out with the Gaussian
03 program.
38
The anharmonic vibrations code implemented in
Gaussian 03 is based on the second-order perturbative treatment
of anharmonic effects by the use of effective, finite-difference
evaluations of third and semidiagonal fourth derivatives.
39–41
The
PED calculations were carried out with the aid of the VEDA
program
42
derived from Balga codes.
43
For details of the PED
analysis see the ESI;† the related methodology details have been
published elsewhere.
44
The Raman activities A
Ram
obtained from Gaussian03 calcu-
lations were converted to relative Raman intensities I
Ram
REL,anh
using the following relationship derived from the intensity theory
of Raman scattering:
45
I
REL;anh
Ram
¼
f ðn
0
n
i
anh
Þ
4
A
Ram
n
i
1 exp
hcn
i
kT
where n
0
is the exciting frequency (9394.7 cm
1
), n
i
anh
is the
calculated anharmonic vibrational wavenumber of the ith
normal mode; h, c and k are fundamental constants, and f is
a suitably chosen common normalization factor for all peak
intensities.
Results and discussion
Polymorphism and the crystal structures of alizarin
The comparison of unit cell parameters and the refinements
quality of the already known structure and a new one—pub-
lished here for the first time—is presented in Table 1.
We present here the structures of the two polymorphs of
alizarin. The crystal data and refinement parameters are collected
in Table 2. The already known structure, hereafter denoted as
form I, crystallizes in the centrosymmetric space group whereas
the new polymorph (form II) is non-centrosymmetic. The
structure of form I has been previously incorrectly established as
Pa space group with missing the centre of symmetry. Both
structures crystallize with no solvent moieties and are severely
disordered and twinned. As can be easily seen, the form I has
a unit cell volume ca. 3 times bigger than the form II. Moreover,
the crystal lattices corresponding to both polymorphs can be
transformed from one into the other by a simple geometric
operation. This is illustrated in Fig. 1.
Table 1 Unit cell parameters and the structure quality for the alizarin taken from literature compared with our results
[Ref. 19] [Ref. 20]
Our results
Form I Form II
T RT RT 100 K 100 K
Space group Pa Pa P2
1
/cPc
a 21.04(4)
A 21.02(9)
A 20.059(3)
A 8.1319(4)
A
b 3.75(1)
A 3.72(1)
A 3.6850(4)
A 3.7102(2)
A
c 20.12(4)
A 20.09(10)
A 21.004(3)
A 16.8711(9)
A
b 104.5(1)
104.3
104.424(15)
100.455(5)
V 1536.9
A
3
1522.2
A
3
1503.7(3)
A
3
500.57(5)
A
3
R factor 26% 20.3% 6.96% 4.21%
This journal is ª The Royal Society of Chemistry 2012 CrystEngComm, 2012, 14, 3667–3676 | 3669
Disorder
Thermal ellipsoid plots visualizing the disorder in both poly-
morphs of alizarin are presented in Fig. 2. The centrosymmetric
structure as an asymmetric part of the unit cell contains one
molecule in general position and a half one located at the centre
of symmetry. The structure is disordered with one moiety split
into two fragments with the occupancy ratio equal to ca.
0.6 : 0.4. The other one—located at the special position—is also
disordered but in a moderate way, which can be indicated by the
shape of the ADP of non-H atoms. The new polymorph crys-
tallizing in the polar Pc space group contains one disordered
molecule as an asymmetric unit. The molecule is split into two
fragments with the occupancy ratio equal to ca. 0.8 : 0.2.
Table 2 Crystal data and refinement parameters for alizarin
Empirical formula
III
C
14
H
8
O
4
C
14
H
8
O
4
M
r
240.20 g mol
1
240.20 g mol
1
T 100(2) K 100(2) K
l 0.71073
A 1.54178
A
Space group P2
1
/cPc
Unit cell dimensions a ¼ 20.059(3)
A, b ¼ 3.6850(4)
A, c ¼ 21.004(3)
A,
b ¼ 104.424(15)
a ¼ 8.1319(4)
A, b ¼ 3.7102(2)
A, c ¼ 16.8711(9)
A, b ¼ 100.455(5)
V 1503.7(3)
A
3
500.57(5)
A
3
Z, D
x
6, 1.592 g cm
3
2, 1.594 g cm
3
m 0.118 mm
1
0.991 mm
1
F(000) 744 248
Crystal size 0.35 0.09 0.09 mm 0.16 0.14 0.01 mm
q
min
, q
max
3.15
, 25.05
5.33
, 67.10
Reflections collected, unique
a
3570, – 1781, –
R
int
a
——
Completeness 99.7% 100.0%
Reflections [I >2s(I)] 1582 1329
Absorption correction multi-scan multi-scan
T
max,
T
min
0.989, 0.960 0.990, 0.858
Data, restraints, parameters 3570, 112, 279 1781, 10, 191
GoF on F
2
0.851 0.966
Final R indices [F
2
>2s(F
2
)] R
1
¼ 0.0696, wR(F
2
) ¼ 0.1763 R
1
¼ 0.0421, wR(F
2
) ¼ 0.1026
R indices (all data) R
1
¼ 0.1336, wR(F
2
) ¼ 0.2013 R
1
¼ 0.0600, wR(F
2
) ¼ 0.1082
r
max
, r
min
0.481, 0.250 e
A
3
0.188, 0.173 e
A
3
a
Refinement based on non-merged data due to twinning by reticular pseudomerohedry.
Fig. 1 Smaller blue unit cell (form II) can be transformed to bigger red
one (form I) with help of the matrix written on the right.
Fig. 2 Thermal ellipsoid plot at the 50% probability level and the numbering scheme for the alizarin form I (a) and form II (b).
3670 | CrystEngComm, 2012, 14, 3667–3676 This journal is ª The Royal Society of Chemistry 2012
The disorder occurs in a similar way in the two polymorphs.
The overlaid fragments of disordered and split moieties are
related to one another by the pseudo-two-fold rotation along the
longer axis of the molecule resulting in an averaged geometry of
the molecule placed in general position of the unit cell. Another
description of the disorder is the reflection of the molecule by
a pseudo-mirror plane. Due to such type of disorder the geom-
etry of the alizarin single-molecule from the crystal structures
should be considered carefully and ought to be supported by the
theoretical calculations.
Another common feature for the alizarin structures is twin-
ning. Both crystals are twinned by reticular pseudo-merohedry
with n ¼ 2 and the twin obliquity equal to 0.75(2)
and 2.135(4)
in form I and form II, respectively. In the polymorph I the twin
elements are (100) mirror plane and/or [401] two-fold rotation
axis, whereas in the form II these elements are (100) mirror plane
and/or [201] two-fold rotation axis. Relations between twin
components of the crystals are shown in Fig. 3. In spite of the
metric properties of the crystal lattice the twinning could be also
associated with the disorder – the mirror plane transforming one
fragment of the disordered moiety to the other is simultaneously
the composition plane of the twin. The same stands for the twin
2-fold rotation axis.
In both polymorphs the packing of molecules is very similar.
This is illustrated in Fig. 4. Analogous issues when the
arrangement of molecules in polymorphs is very similar are
examples of aspirin
46
and D-ribose.
47
It is worth noting that in
those literature examples the unit cell of the corresponding
polymorphs could be transformed one to the other with use of
rational elements containing matrices.
Fig. 3 Possible mechanism of twinning of the polymorphs of alizarin
form I (a) and form II (b). Reflecting the red-colored fragment of the
lattice by either (100) – form I or (10
2) – form II leads to the green-
colored one with (100) and (10
2) acting as a composition plane (gray
dotted line) of the twin where the disordered molecules are almost ideally
overlaid in crystals of form I and II, respectively.
Fig. 4 Packing diagrams of the alizarin form I and form II crystals, view
along (100), (010) and (001) starting from the top row.
Table 3 Possible hydrogen bonds in the alizarin polymorphs (
A and
).
Dotted lines stand for intramolecular HB
a
D–H/A d
(D–H)
d
(H/A)
d
(D/A)
<
(DHA)
Form I
O2–H2O/O1 0.84 1.86 2.605(8) 146
O3–H3O/O2 0.84 2.20 2.671(8) 116
O3–H3O/O21#1 0.84 2.04 2.724(7) 138
O2B–H2OB/O1B 0.84 1.86 2.605 146
O3B–H3OB/O2B 0.84 2.20 2.672 116
O3B–H3OB/O1B#2 0.84 2.38 3.117(8) 146
O22–H22/O21 0.84 1.70 2.480(7) 154
O23–H23/O22 0.84 2.30 2.710(9) 111
O23–H23/O1#3 0.84 2.57 3.107(9) 123
O23–H23/O4B 0.84 2.31 2.834(10) 120
O3B–H3OB/O4#2 0.84 1.99 2.710(10) 143
Form II
O2–H2O/O1 0.84 1.81 2.546(4) 146
O3–H3O/O2 0.84 2.23 2.691(4) 115
O3–H3O/O4#4 0.84 1.99 2.712(6) 143
O2B–H2OB/O1B 0.84 1.81 2.546 146
O3B–H3OB/O2B 0.84 2.23 2.691 115
O3B–H3OB/O4B#5 0.84 2.42 2.907 118
O3B–H3OB/O4B#6 0.84 2.57 3.054 118
O3–H3O/O1B#4 0.84 2.29 3.023(7) 146
a
Symmetry transformations used to generate equivalent atoms: #1 x,
y + 3/2, z + 1/2; #2 x +1,y + 1/2, z + 1/2; #3 x, y +1,z;#4x,
y, z + 1/2; #5 x +1,y 1, z;#6x +1,y, z.
This journal is ª The Royal Society of Chemistry 2012 CrystEngComm, 2012, 14, 3667–3676 | 3671
Contrary to the famous case of aspirin, however, both poly-
morphs of alizarin have no distinct domain structure. The mole-
cules are stacked in (010) direction with an average intermolecular
distance equal to 3.42
A and 3.38
A in non-split and split moieties
in polymorph I. In the case of polymorph II this distance is close
to the average of these values in form I yielding 3.39
A.
There are numerous intermolecular hydrogen bonds (HB)
present between the stacks. The view of this type of interaction is
complicated by disorder and many possible relative orientations of
moieties. The hydrogen bonds are listed in Table 3, however
because of the disorder these values are approximated only. In both
polymorphs the strongest HB are intramolecular ones. This type of
interaction involves hydroxyl groups and carbonyl O atoms.
Because both crystalline forms of alizarin are very similar it is
desirable to compare the packing and intermolecular interactions
based on the analysis of Hirshfeld surfaces
48
and fingerprints of
O/H contacts. The analysis was carried out using Crysta-
lExplorer program
49
based on the crystallographic information
files containing information about the occupancies of disordered
molecular fragments. Due to disorder of both structures mani-
festing in averaging of alternative orientation of alizarin moieties
we decided to focus on the fingerprints concerning O/H and
C/C contacts. The comparison of 2-D fingerprints for O/H
interactions (including reciprocal contacts) and C/C interac-
tions is presented below in the Fig. 5.
It can be seen that indeed both forms indicate slightly different
2-D plots visualizing intermolecular interactions. In the case of
the form I there are different plots for different alizarin moieties
forming the independent part of the unit cell. Ordered molecules
in the polymorph I form slightly weaker interactions compared
to disordered molecules in polymorph I and II. Moreover the
plot is in addition non-symmetrical. But not only differences in
strong HB can be seen; the 2-D plots generated for C–C contacts
are also different. The distribution of C–C distances is wider in
the case of polymorph II. In this form shorter C/C contacts can
also be observed.
The PXRD pattern of alizarin polymorphs
Fig. 6 presents experimental and theoretically calculated powder
diffraction (PXRD) patterns of polymorph I and II of alizarin
(more details are available in the ESI† - see Figs. 1SI-3SI). The
patterns are clearly distinct, which confirms that alizarin exists in
two polymorphic forms.
Vibrational spectra of the two polymorphs
The two alizarin polymorphic crystals are different, but only
slightly. Really, the geometrical parameters of the molecules in
the two crystals are indistinguishable, and the intra and
Fig. 5 2-D fingerprints visualizing O/H and C/C interactions based on Hirshfeld surface generated for separate molecules is the crystals of two forms
of alizarin.
3672 | CrystEngComm, 2012, 14, 3667–3676 This journal is ª The Royal Society of Chemistry 2012
intermolecular hydrogen bond distances are very comparable
(Table 3). Thus, it is clear why the Raman spectra of the two
polymorphs show practically no differences (Fig. 4SI and 5SI†),
whereas the IR spectra demonstrate subtle dissimilitude (Fig. 7
and 6SI†). Indeed, the Raman spectra correspond mainly to the
(identical) alizarin single molecules, because differences in similar
hydrogen bonding nets are indistinguishable by Raman: scat-
tering is only weakly sensitive to such intermolecular interac-
tions. On the other hand, IR spectroscopy, although sensitive to
hydrogen bonding, for the twin similar nets formed by the OH/
O]C moieties, for which the n(OH) are broad and the n(C]O)
are also substantially broadened, yield twin similar IR patterns,
as well. Thus, three distinct n(OH) bands of the hydrogen bonded
hydroxyl groups are seen for the form I and three for the form II:
for the former the highest maximum is positioned for the most
red-shifted band, whereas for the latter for the band in the middle
of the contour (Fig. 7a). The other difference can be detected for
the triplet of the b(CH) bands in the 1200–1100 cm
1
region. The
b(CH_R2) band at ca. 1200 cm
1
, is connected with the C–H
bending of the R2 ring, and is stronger for the form I (Fig. 7b).
This may be a consequence of closer intermolecular contact of
the R2 ring C–H groups in the form I crystals. The third
difference is located at 767 cm
1
, where the s(OH) band is well-
shaped for the polymorph I whereas for the form II it is broad-
ened and without a maximum (Fig. 7b). Obviously, the above
differences are not enough for a practical differentiation between
the two polymorphs. Finally, we simulated the spectra of dimers
(Fig. 8) at the B3LYP/6-31++G** level and showed that
comparison with the Raman and IR spectra of monomer and
dimers demonstrates only non-diagnostic changes in the spectra
(Figs. 7SI and 8SI, respectively†). This shows the influence of the
neighboring molecules on the vibrational properties of single
alizarin molecule is negligible.
IR and Raman spectra of polycrystalline alizarin
Studies on alizarin polymorphism prompted us to interpret the
entire experimental Raman and IR spectra of polycrystalline
alizarin (Fig. 9 and 9SI†). The whole spectra, but the 3700–3000
cm
1
region (Fig. 7a, Fig. 9SI†), are excellently reproduced by the
anharmonic, B3LYP/aug-cc-pVDZ vibrational spectra of single
molecule (Table 1SI, Fig. 10SI and 11SI, respectively†).
Moreover, the position of the intramolecular H-bond n(C]
O3/H–O1) band at 1633 cm
1
(Fig. 10) is perfectly predicted by
the anharmonic B3LYP/aug-cc-pVDZ spectrum, because this
very moiety remains practically unaffected in the crystal phase.
The two-times stronger n(C] O4) IR band predicted to be
located at 1683 cm
1
is positioned at 1663 cm
1
. Band position
and the red-shift of 20 cm
1
confirm that the intermolecular H-
bonds engaging the C]O4 are weaker than the C]O3/H–O1
ones in full agreement with the X-ray data (Table 3).
The band fitting routine shows that the experimental IR n(C]
C) bands, with maxima at ca. 1590 and 1460 cm
1
, are composed
of signals with maxima at 1590, 1571, 1477, 1465, and 1452 cm
1
(Fig. 10), as well predicted by the theoretical IR spectra and
confirmed by similar procedure for the Raman n(C]C) contour.
Again, the strongest IR band, assigned to the n(C–O2)
stretching vibrations, is both located and predicted at 1295 cm
1
(Fig. 11). The theoretical n (C–O1) stretching vibrations band is
predicted to be positioned at 1319 cm
1
and to be ten-times
weaker and thus covered by the n(C–O2) one of ca. 30 cm
1
width. Assignment of the b(O1–H) and b(O2–H) bands as well as
of several CH bending and deformation modes is presented
in Fig. 11.
Fig. 6 Experimental and theoretically calculated PXRD pattern of
polymorph I and polymorph II.
Fig. 7 The IR spectra of the alizarin form I (dark blue) and form II (red) crystals in the range of 3700–3000 cm
1
(a) and 1700–700 cm
1
(b).
This journal is ª The Royal Society of Chemistry 2012 CrystEngComm, 2012, 14, 3667–3676 | 3673
The IR bands at 765 cm
1
and 748 cm
1
are assigned to the
modes originating from s
1
(O1–H) deformation vibrations
(Fig. 12). The strongest IR band in the 700–500 cm
1
range is
associated with the s
6
(CH_R2) vibrations at 713 cm
1
, while in
the Raman spectrum it corresponds to b(CCC) vibrations at 662
cm
1
. The middle intensity s(O2–H) mode is predicted in the IR
spectrum at 456 cm
1
, however, it is not observed probably due
to band broadening.
Conclusion
The X-ray study revealed new polymorph of alizarin and threw
a new light onto the structure of the previously known crystalline
form of this dye important in medicine, which serves also as
a component of electronic materials. In spite of difference in
space group—the new form is non-centrosymmetric whereas the
known structure is centrosymmetric—these two polymorphs are
very similar. Both crystals are twinned in a comparable way. The
Fig. 8 Visualization of hydrogen bonds present in the alizarin form II (a) and form I (b) crystals. Only molecules with higher occupancies are shown.
The B3LYP/6-31++G** calculated dimer II (c) and dimer I (d).
Fig. 9 IR and Raman spectra of alizarin powder.
3674 | CrystEngComm, 2012, 14, 3667–3676 This journal is ª The Royal Society of Chemistry 2012
same stands for the disorder and packing of the molecules in the
crystal lattice.
The Raman spectra of the two polymorphs show no significant
differences, whereas in the IR spectra subtle dissimilarities can be
found in the n(OH), b(CH) and s(OH) bands. Nevertheless,
discrimination between the two polymorphs is hardly possible by
the vibrational spectroscopy techniques. The entire experimental
Raman and IR spectra of polycrystalline alizarin, but the n(OH)
Fig. 10 The n(C]O) and n(CC) region of the experimental IR (red) and Raman (blue) spectra.
Fig. 11 The n(C–O), n(CC) and b(O–H) region of the experimental IR (red) and Raman (blue) spectra.
Fig. 12 The region of deformation bands of the experimental IR (red) and Raman (blue) spectra.
This journal is ª The Royal Society of Chemistry 2012 CrystEngComm, 2012, 14, 3667–3676 | 3675
stretching vibration region, are excellently reproduced by the
anharmonic, B3LYP/aug-cc-pVDZ vibrational spectra of single
molecule.
Acknowledgements
Dedicated to Professor Roland Boese in recognition of his
outstanding contribution to crystallography.
AR and MB acknowledge the Polish Ministry of Science and
Higher Education for the financial support by Grant No. N
N204121638. LD acknowledges the Foundation for Polish Science
for the Kolumb stipend. MKC kindly acknowledges Dr I. Madura
and Dr Andrzej Ostrowski (Technical University of Warsaw) for
their valuable help. Single crystal diffraction measurement of the
polymorph I was performed in the Crystallographic Unit of the
Physical Chemistry Laboratory at the Faculty of Chemistry of
the University of Warsaw. This research was supported by the
grant for the statutory activity of ICRI institute. The computa-
tional Grants G18-4 and G19-4 from the Interdisciplinary Center
of Mathematical and Computer Modeling (ICM) at University of
Warsaw are gratefully acknowledged.
References
1 D. De Santis and M. Moresi, Ind. Crops Prod., 2007, 26, 151.
2 S. U. Park, Y. K. Kim and S. Y. Lee, Sci. Res. Essay, 2009, 14, 94.
3 A. Shotipruk, J. Kiatsongserm, P. Pavasant, M. Goto and M. Sasaki,
Biotechnol. Prog., 2004, 20, 1872.
4 G. Goffer, Archaeological Chemistry, Wiley-Interscience, John Wiley
& Sons, Inc., Hoboken, New Jersey, USA, 2007.
5 J. Mollo, Military fashion, a comparative history of the uniforms of the
great armies from the 17th century to the First World War, Putnam,
New York, 1972.
6 I. Holme, Soc. Dyers Colour., Color. Technol., 2006, 122, 235.
7 G. C. H. Derksena and T. A. Van Beek, Stud. Nat. Prod. Chem., 2002,
G26, 629.
8 H. Puchtler, S. N. Meloan and M. S. Terry, J. Histochem. Cytochem.,
1969, 17, 110.
9 S. A. Norton, J. Am. Acad. Dermatol., 1998, 39, 484.
10 C. A. Gregory, W. G. Gunn, A. Peister and D. J. Prockop, Anal.
Biochem., 2004, 329, 77.
11 W. R. Duncan and O. V. Prezhdo, J. Phys. Chem. B, 2005, 109, 365.
12 R. Huber, S. Sporlein, J. E. Moser, M. Gr
€
atzel and J. Wachtveitl,
J. Phys. Chem. B, 2000, 104, 8995.
13 S. Kaniyankandy, S. Verma, J. A. Mondal, D. K. Palit and
H. N. Ghosh, J. Phys. Chem. C, 2009, 113, 3593.
14 L. Dworak, V. V. Matylitsky and J. Wachtveitl, ChemPhysChem,
2009, 10, 384.
15 H. Lachheb, E. Puzenat, A. Houas, M. Ksibi, E. Elaloui, C. Guillard
and J.-M. Herrmann, Appl. Catal., B, 2002, 39, 75.
16 G. Liu, T. Wu, J. Zhao, H. Hidaka and N. P. Serpone, Environ. Sci.
Technol., 1999, 33, 2081.
17 J. Guilhem, Acta Crystallogr., 1961, 14, 88.
18 J. Guilhem, D
etermination par la diffraction des rayons X, de la
structure cristalline des deux dihydroxyanthraquinones,
l’anthrarufine et l’alizarine, PhD thesis, Facult
e des sciences de
Paris, l’Universit
e de Paris, 1967.
19 J. Guilhem, Bull. Soc. Chim. Fr., 1967, 1666.
20 R. E. Algra, W. S. Graswinckel, W. J. P. van Enckevort and E. Vlieg,
J. Cryst. Growth, 2005, 285, 168.
21 A. Baran, B. Wrzosek, J. Bukowska, L. M. Proniewicz and
M. Baranska, J. Raman Spectrosc., 2009, 40, 436.
22 CrysAlis RED, Oxford Diffraction Ltd., Version 171.33.66 Empirical
absorption correction using spherical harmonics, implemented in
SCALE3 ABSPACK scaling algorithm.
23 CrysAlis CCD, Oxford Diffraction Ltd., Version 171.33.66; CrysAlis
RED, Oxford Diffraction Ltd., Version 171.33.66.
24 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990,
A46, 467.
25 G. M. Sheldrick, SHELXL93. Program for the Refinement of Crystal
Structures, University of G
€
ottingen, Germany.
26 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
27 International Tables for Crystallography, ed. A. J. C. Wilson, Kluwer,
Dordrecht, 1992, Vol. C.
28 I. J. Bruno, J. C. Cole, P. R. Edgington, M. K. Kessler, C. F. Macrae,
P. McCabe, J. Pearson and R. Taylor, Acta Crystallogr., Sect. B:
Struct. Sci., 2002, B58, 389.
29 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
30 K. Burke, J. P. Perdew and M. Ernzerhof, Mixing exact exchange
with GGA: When to say when, in Electronic Density Functional
Theory: Recent Progress and New Directions, ed. J. F. Dobson, G.
Vignale and M. P. Das, Plenum, 1998, pp. 57–68.
31 T. H. Dunning Jr, J. Chem. Phys., 1989, 90, 1007.
32 R. A. Kendall, T. H. Dunning and R. J. Harrison, J. Chem. Phys.,
1992, 96, 6796.
33 R. Janoschek, Pure Appl. Chem., 2001, 73, 1521.
34 T. Helgaker, P. Jorgensen and J. Olsen, Molecular Electronic-
Structure Theory, John Wiley & Sons, 2000.
35 W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys., 1972, 56,
2257.
36 J. D. Dill and J. A. Pople, J. Chem. Phys., 1975, 62, 2921.
37 M. M. Francl, W. J. Petro, W. J. Hehre, J. S. Binkley,
M. S. Gordon, D. J. DeFrees and J. A. Pople, J. Chem. Phys.,
1982, 77, 3654.
38 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheesema n, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa , M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox,
H. P. Hratchian, J. B. Cro ss, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Ya zyev, A. J. Austin,
R. Cammi, C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. V. Ortiz, Q. Cui, A. G. Baboul, S. Cliffor d, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi,
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
GAUSSIAN 03 (Revision E.01), Gaussian, Inc. , Wallingford,
CT, 2004.
39 V. Barone and C. Minichino, J. Mol. Struc.–Theochem, 1995, 330,
365.
40 V. Barone, J. Chem. Phys., 2004, 120, 3059.
41 V. Barone, J. Chem. Phys., 2005, 122, 014108–1.
42 M. H. Jamr
oz, Vibrational Energy Distribution Analysis, VEDA 4.0
program, Warsaw 2004.
43 L. Lapinski and M. J. Nowak, BALGA—computerprogram for PED
calculation, Institute of Physics Polish Academy of Sciences, Warsaw.
44 M. H. Jamr
oz, J. Cz. Dobrowolski and R. Brzozowski, J. Mol.
Struct., 2006, 787, 172.
45 V. Krishnakumar, G. Keresztury, T. Sundius and R. Ramasamy, J.
Mol. Struct., 2004, 702,9.
46 A. D. Bond, R. Boese and G. R. Desiraju, Angew. Chem., Int. Ed.,
2007, 46, 618.
47 D. Sisak, L. B. McCusker, G. Zandomeneghi, B. H. Meier, D. Blaser,
R. Boese, W. B. Schweizer, R. Gilmour and J. D. Dunitz, Angew.
Chem., Int. Ed., 2010, 49, 4503.
48 J. J. McKinnon, M. A. Spackman and A. S. Mitchell, Acta
Crystallogr., Sect. B: Struct. Sci., 2004, B60, 627.
49 S. K. Wolff, D. J. Grimwood, J. J. McKinnon, D. Jayatilaka,
M. A. Spackman, Crystal Explorer 2.0, 2005–2007, University of
Western Australia.
3676 | CrystEngComm, 2012, 14, 3667–3676 This journal is ª The Royal Society of Chemistry 2012