The photodimerisation of the α- and β-forms of trans-cinnamic acid: A study of single crystals by vibrational microspectroscopy

Article abstract of DOI:10.1016/S1386-1425(02)00208-1

Infrared and Raman microspectroscopy have been used to follow the photodimerisation reactions of single crystals, the α- and β-forms of trans-cinnamic acid. This approach allows the starting materials and products - α-truxillic acid that has C_i symmetry and β-truxinic acid, which has C_s symmetry - to be identified. It also allows the topotactic nature of the reaction to be confirmed. Attempts to produce the poorly-defined unreactive γ-form of trans-cinnamic acid resulted only in a mixture of the α- and β-forms. The findings suggest a wide role for these spectroscopic methods in monitoring solid-state organic reactions.

Full text of DOI:10.1016/S1386-1425(02)00208-1

Spectrochimica Acta Part A 59 (2003) 629ꢁ635
/
www.elsevier.com/locate/saa
The photodimerisation of the a- and b-forms of trans-cinnamic
acid: a study of single crystals by vibrational
microspectroscopy
Samantha D.M. Atkinson, Matthew J. Almond ꢀ, Peter Hollins,
Samantha L. Jenkins
Department of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, UK
Received 4 June 2002; accepted 7 June 2002
Abstract
Infrared and Raman microspectroscopy have been used to follow the photodimerisation reactions of single crystals,
a-truxillic acid
the a- and b-forms of trans-cinnamic acid. This approach allows the starting materials and products*
/
that has Ci symmetry and b-truxinic acid, which has Cs symmetry*to be identified. It also allows the topotactic nature
/
of the reaction to be confirmed. Attempts to produce the poorly-defined unreactive g-form of trans-cinnamic acid
resulted only in a mixture of the a- and b-forms. The findings suggest a wide role for these spectroscopic methods in
monitoring solid-state organic reactions.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Trans-cinnamic acid; Truxillic acid; Vibrational microspectroscopy; Photodimerisation; Single crystals
1. Introduction
tions [7] and of reactions between components in
co-crystals [8ꢁ12]. It is true, however, that the
/
Solid state organic reactions are attracting
increased attention at the current time. In part,
this interest arises from the possibility that the
crystal lattice in a solid reactant may direct the
course of a reaction and determine the stereoche-
mical nature of products. The term ‘crystal en-
gineering’ has been applied to such processes [1,2].
This has led to the possibility of asymmetric
mechanisms of such reactions are often poorly
understood. In part, this lack of knowledge results
from the lack of suitable methods for monitoring
reactions within single crystals. Single crystal
diffraction is often inappropriate because the
crystals degrade during the course of the reactions
[13]. Atomic force microscopy has also been
employed and while this allows the progress of a
reaction to be monitored, it gives no information
regarding the structures of intermediates or pro-
synthesis in solid crystals [3ꢁ6], of cascade reac-
/
ducts of the reactions [14ꢁ17].
/
Our aim in the current work was to utilise the
techniques of infrared and Raman microscopy to
ꢀ Corresponding author. Fax: ꢂ
E-mail address: m.j.almond@reading.ac.uk (M.J. Almond).
/
44-1-1893-16332
1386-1425/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 1 3 8 6 - 1 4 2 5 ( 0 2 ) 0 0 2 0 8 - 1

630
S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ635
/
follow photochemical reactions within single crys-
tals. We selected the photochemical dimerisation
of trans-cinnamic acid as a paradigm of such
reactions. This is a process that has been well-
known since the early years of the 20th century
could be distinguished by this approach, to show
that different products could be identified from the
two starting materials and to find evidence for the
topotactic nature of the reactions. There has been
very little use of vibrational spectroscopy to
monitor solid-state organic reactions. Mirsa et al.
[24] used laser Raman phonon spectroscopy and
after extensive photolysis, found evidence for
photodimerisation of p-formyl-trans-cinnamic
acid. In a preliminary study, we have shown that
infrared and Raman microscopy may be used to
follow the photodimerisation of molecules of
cinnamic acid derivatives within single crystals
[26]. The work reported in this paper is, however,
the first account of the use of these methods to
monitor single crystals of specific forms of any
organic crystals. It is, moreover, the first occasion
that such an approach has been used to demon-
strate the topotactic nature of a solid state single-
crystal organic reaction.
[18ꢁ22]. The reactions have been shown to be
/
topotactic in that the a (head-to-tail) form of the
monomer gives the dimeric product a-truxillic acid
(I), while the b (head-to-head) form of the mono-
mer gives b-truxinic acid (II) (Scheme 1) [23]. Even
for this straightforward reaction, however, it has
proved difficult to monitor the reaction in one
single crystal. Some success has been achieved by
single crystal diffraction when the crystal is
irradiated in the very tail of its absorption band
[24]. Atomic force microscopy has shown that low
steps and hills, growing to three times their height
after 3 days of exposure to ultraviolet light, are
seen [14]. It was thought that these might arise
from a hydrate of cinnamic acid, although it was
not possible to obtain direct spectroscopic evi-
dence from atomic force microscopy to back up
this assertion.
2. Experimental section
Our aims in the current work were to apply the
techniques of infrared and Raman microscopy to
monitor the photodimerisation reactions of single
crystals of the a- and b-forms of trans-cinnamic
acid. We wished to show that the a- and b-forms
2.1. Apparatus
Infrared spectra were recorded using a Bruker
Equinox 55 FTIR and IRscope II microscope. A
medium-band liquid nitrogen-cooled detector was
used to monitor the spectral range 600ꢁ
4000 cmꢃ1
/
with a typical resolution of 4 cmꢃ1. Raman
spectra were obtained using a Renishaw (model
1000) laser Raman spectrometer with a Leica
microscope attachment. It was equipped with
frequency-doubled Nd-YAG (532 nm) and diode
(785 nm) laser sources. The former source was
used for all spectra reported in this work. Spectra
were recorded in the range 200ꢁ
3200 cmꢃ1 with a
/
typical resolution of 4 cmꢃ1
.
2.2. Samples
A sample of trans-cinnamic acid was obtained
from Aldrich with a stated purity of ꢀ99%. A
/
sample of a-trans-cinnamic acid was prepared by
recrystallisation from benzene and a sample of b-
trans-cinnamic acid from acetone [27ꢁ
/29]. At-
Scheme 1.
tempts were also made to prepare a sample of g-

S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ
/
635
631
trans-cinnamic acid by recrystallisation from eth-
anol [25]. All samples were photolysed using a
medium-pressure water-cooled mercury lamp
(Photochemical Reactors).
matic), 1497, 1454, 1427, 1290 (broad), 1236
(broad), 1081 (w), 1035 (w), 1005 (w) (d(CÃH)
and n(CÃO)), 935 (broad) (ring breathing), 772,
697, 640 (w) (ring deformation). Raman: 3069 (s),
2994, 2974, 2925 (n(CÃH)), 1610, 1593 (w) (n(Cꢄ
C) aromatic), 1295, 1193, 1039 (d(CÃH)), 1009 (s)
(ring breathing), 996 (w), 935 (w) (d(CÃH)), 814
(n(CÃC)), 775 (w), 625 (ring deformation).
/
/
/
/
2.3. Vibrational spectra
/
/
N.B.: All numbers are wavenumbers measured
in cmꢃ1. Where no intensities are given, the band
may be assumed to have medium intensity. The
/
assignments in the region
B
/
1500 cmꢃ1 are
3. Results and discussion
tentative because of vibrational coupling and
overlap of spectral bands.
The aim of the first part of this study was to
record infrared and Raman spectra of single
crystals of the a- and b-forms of trans-cinnamic
acid. Samples of the a- and b-forms were prepared
following the recrystallisation procedure outlined
by Schmidt [23,27,28]. Confirmation that the a-
and b-forms had actually been produced was
achieved by diffraction measurement of the unit
a-Trans-cinnamic acid (before photolysis). In-
frared: 3089, 3066, 3029 (n(CÃ
/
H)), 1680 (vs)
C) aliphatic), 1578
C), aromatic), 1496, 1450, 1419, 1318,
1289, 1266, 1223, 1169, 1074 (w), 980 (w) (d(CH)
and n(CÃO)), 771 (w) and 774 (w) (ring breathing
modes). Raman: 3088, 3067, 3053, 3037 (s), 2982
(n(CÃH)), 1637 (vs) (n(CꢄC) aliphatic), 1599 (vs)
(n(CꢄC) aromatic), 1494 (w), 1450 (w), 1285,
1258, 1210, 1179, 1161, 1026 (d(CÃH)), 1000 (vs)
(ring breathing), 873, 846 (n(CÃC)), 679, 621 (ring
deformation).
a-Truxillic acid (after photolysis). Infrared:
1705 (vs) (n(CꢄO)), 1578 (w) (n(CꢄC) aro-
matic), 1496, 1422, 1332, 1275, 1219, 1186, 1083
C)), 944 (broad) (ring
(n(Cꢄ
/
O)), 1627 (s) (n(Cꢄ
/
(w) (n(Cꢄ
/
/
˚
cell parameters (A) of the crystals, which are
/
/
/
reported as*
[23]; aꢄ7.79(2), bꢄ
aꢄ7.735(2), bꢄ17.671(4), cꢄ
form: aꢄ31.3, bꢄ4.04, cꢄ6.05 [23]. Moreover,
/
a form: aꢄ
18.07(5), cꢄ
5.582(2) [30]; b-
/
7.79, bꢄ
/
18.07, cꢄ
/
5.67
/
/
/
/
5.67(2) [29] and
/
/
/
/
/
/
/
the two forms of the crystals have quite different
morphologies, which makes them quite easy to
distinguish. The a-form crystallises as flat plates
and the b-form as long laths [27]. We were
therefore in no doubt as to which of our samples
was in the a-form and which was in the b-form.
The infrared and Raman spectra of single
crystals of the a-form of trans-cinnamic acid are
shown in Figs. 1 and 2. Corresponding spectra of
the b-form are shown in Figs. 3 and 4. The spectra
show significant differences. In particular, the
infrared spectrum of the a-form shows prominent
bands at 3089, 3066 and 3029 cmꢃ1 which arise
/
/
(w) (d(CÃ
/
H) and n(CÃ
/
breathing), 773 (w), 743 (w), 707 (w) (ring defor-
mation). Raman: 3058, 2965, 2942 (w), 2918 (w)
(n(CÃ
(w), 1289 (w), 1215, 1207, 1187, 1169, 1156, 1032
(d(CÃH) and n(CÃO)), 1000 (vs) (ring breathing),
808 (n(CÃC)), 787, 768, 618 (ring deformation).
b-Trans-cinnamic acid (before photolysis). In-
frared: 3085 (w), 3056 (w), 3026 (n(CÃH)), 1682 (s)
(n(CꢄO)), 1633 (s) (n(CꢄC) aliphatic), 1579 (w)
(n(CꢄC) aromatic), 1494, 1450, 1422, 1350, 1316,
1206 (w), 1177 (w), 985, 935, 873, 766 (d(CÃH) and
n(CÃO)), 705, 677 (ring deformation). Raman:
3097 (w), 3075, 3045 (n(CÃH)), 1643 (vs) (n(CꢄC)
aliphatic), 1605 (vs) (n(CꢄC) aromatic), 1501,
1449, 1298, 1272, 1218, 1184, 1032 (d(CÃH)), 1007
(s) (ring breathing), 882 (w), 853 (w) (n(CÃC), 626
(w) (ring deformation).
b-Truxinic acid (after photolysis). Infrared:
1712 (s) (n(CꢄO)), 1607, 1595 (n(CꢄC) aro-
/
H)), 1599, 1586 (n(Cꢄ
/
C) aromatic), 1360
/
/
/
/
/
/
/
from n(CÃ
/
H) vibrations of CÃ/H moieties attached
/
to unsaturated carbon atoms. The b-form shows a
different pattern of less intense absorptions at
3085, 3056 and 3026 cmꢃ1 in this region. For the
a-form, there is also an intense band around 1224
cmꢃ1. Although bands in this region are difficult
to assign with certainty because of coupling of
/
/
/
/
/
/
vibrations, this band probably arises from a CÃ
/
H
deformation mode. These features are not promi-
/
/
nent in the infrared spectrum of the b-form,

632
S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ635
/
Fig. 1. Infrared spectra of a single crystal of the a-form of trans-cinnamic acid before (top spectrum) and after (bottom spectrum) 10
min of photolysis using a medium-pressure water-cooled mercury lamp. Approximately 100% conversion to a-truxillic acid is observed.
Fig. 3. Infrared spectra of a single crystal of the b-form of
trans-cinnamic acid before (top spectrum) and after (bottom
spectrum) 10 min of photolysis using a medium-pressure water-
cooled mercury lamp. Approximately 100% conversion to b-
truxinic acid is observed.
Fig. 2. Raman spectra of a single crystal of the a-form of trans-
cinnamic acid before (top spectrum) and after (bottom spec-
trum) 10 min of photolysis using a medium-pressure water-
cooled mercury lamp. Approximately 100% conversion to a-
truxillic acid is observed.
although here a sharp, moderately intense band is
seen at 871 cmꢃ1, which is not seen for the a-form.
In the Raman spectrum, differences are more
subtle although inspection of the spectra repro-
duced in Figs. 2 and 4 shows that there are
significant differences in band position and inten-
the different crystallographic forms (a and b) of
trans-cinnamic acid*has therefore been achieved.
/
The next stage of the project was to monitor the
photodimerisation reactions of these two different
crystal forms. Infrared and Raman spectra of the
two different forms before and after 10 min of
photolysis using a medium pressure mercury lamp
sity in the region 1000ꢁ
samples. The first objective of our work*
vibrational microscopy to distinguish crystals in
/
1600 cmꢃ1 for the two
to use
/
are shown in Figs. 1ꢁ/4. Several striking differences
can be seen between the spectra recorded before

S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ
/
635
633
arising from n(CÃ
is unsaturated (typically above 3000 cmꢃ1
H),
/H) where the carbon atom
)
and the growth of bands arising from n(CÃ
/
where the carbon atom is saturated (typically
below 3000 cmꢃ1) is apparent. The a-form of
the monomer shows prominent n(CÃ/H) bands
at 3089, 3066 and 3029 in the infrared and at
3088, 3067, 3053 and 3037 cmꢃ1 in the
Raman. The b-form shows infrared bands at
3085, 3056 and 3026 cmꢃ1 and Raman bands
at 3097, 3075 and 3045 cmꢃ1. After photo-
lysis, the a-form shows no distinct infrared
bands, but shows Raman bands at 3058, 2965,
2942 and 2918 cmꢃ1. For the b-form, there
Fig. 4. Raman spectra of a single crystal of the a-form of trans-
cinnamic acid before (top spectrum) and after (bottom spec-
trum) 10 min of photolysis using a medium-pressure water-
cooled mercury lamp. Approximately 80% conversion to b-
truxinic acid is observed.
and after photolysis. These changes may be
summarised as follows.
are no clear n(CÃ
/H) infrared bands seen after
photolysis, but the Raman spectrum shows
prominent bands at 3069, 2994, 2974 and 2925
cmꢃ1. In each case, the growth of bands
below 3000 cmꢃ1 is apparent.
1) The Raman band at 1637 cmꢃ1 for the a-form
which corresponds to the aliphatic n(Cꢄ
/
C)
decreases markedly in intensity upon photo-
lysis. The corresponding infrared band at 1627
cmꢃ1 has been photolysed to extinction. For
the b-form, the corresponding bands are seen
at infrared 1633 cmꢃ1 and Raman 1643
cmꢃ1. The discrepancy in position between
the infrared and Raman bands occurs because
of the difficulty in measuring the exact posi-
tion of the infrared band which is broad and
overlaps somewhat with the intense band
4) In the region below 1500 cmꢃ1 it is more
difficult to assign bands with certainty. The
most prominent Raman band for both forms
(especially prominent for the b-form) is the
ring breathing mode of the aromatic ring. This
band is common to both monomer and dimer
and is seen for both a- and b-forms.
These spectra demonstrate that :100% conver-
/
sion to the dimer has occurred after 10 min of
photolysis. It is therefore possible to compare the
infrared and Raman spectra of the dimeric mole-
cules formed from the two different forms of the
monomer.
arising from n(Cꢄ
This overlap means that it is also easier to
monitor the change in intensity of n(CꢄC)
(aliphatic) by Raman spectroscopy.
2) The infrared spectra show intense bands
arising from n(CꢄO) of the carboxyl group,
a vibration which gives only weak Raman
O) is seen to
/
O) of the carboxyl group.
/
If we assume that a topotactic reaction has
occurred, then the products of dimerisation are a-
truxillic acid (I) from the a-form of the monomer
and b-truxinic acid (II) from the b-form of the
monomer. b-truxinic acid has low symmetry: it
belongs to the Cs point group. Crystals of b-
truxinic acid have Cs or m symmetry [27]. The
symmetry of a-truxillic acid depends critically on
whether the C4 ring is planar, thereby introducing
a centre of symmetry. X-ray determinations of the
structures of both a-truxinic acid itself [13] and
/
scattering. The position of n(Cꢄ
/
move to higher frequency as the photolysis
proceeds. This shift reflects the conjugation
between Cꢄ
monomer leading to a weakening of the Cꢄ
O bond. Such conjugation is, of course, lost in
the dimer. For the a-form, n(CꢄO) shifts
from 1680 to 1705 cmꢃ1 and for the b-form,
the shift is from 1682 to 1712 cmꢃ1
3) The n(CÃH) region also provides a convenient
/
O and CꢄC bonds in the
/
/
/
.
several of its derivatives [31ꢁ33] all indicate that
/
/
the C4 ring is indeed planar, so that the molecule
belongs to the Ci symmetry group. It may there-
fore be expected that the mutual exclusion princi-
way to monitor the reaction, especially by
Raman spectroscopy. The decay of bands

634
S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ635
/
ple relating infrared and Raman spectra will apply.
As such, for a-truxillic acid we may expect no
coincidences between infrared and Raman bands,
whereas for b-truxinic acid, we would expect to see
coincidences. For a-truxillic acid, the most promi-
nent infrared bands occur at 1705, 1422, 1275,
2) The techniques allow the photodimerisation of
these monomers to be followed and demon-
strate that the reactions follow a topotactic
pathway.
3) The techniques may also be used to investigate
the structure of the poorly-characterised g-
form of the monomer. It is shown that, under
the conditions of our experiments, the g-form
is not produced. Rather a mixture of the a-
and b-forms is seen.
1219, 1186, 1083, 944, 773, 743 and 707 cmꢃ1
.
There is no strong correlation between these and
the prominent Raman bands. The most intense
Raman bands are seen at 3058, 2965, 2942, 2918,
1586, 1187, 1169, 1156, 1032, 1000, 808, 787 and
618 cmꢃ1. For b-truxinic acid, infrared bands are
seen at 1712, 1607, 1595, 1497, 1454, 1427, 1290,
The findings suggest a wide potential role for
infrared and Raman microspectroscopy in mon-
itoring solid-state organic reactions.
1236, 1081, 1035, 1005, 935, 772 and 697 cmꢃ1
Some of these are matched by Raman bands at
1615, 1593, 1295, 1039, 1009, 935 and 775 cmꢃ1
.
.
These observations provide a body of evidence
that the expected topotactic products have been
formed, i.e. a-truxillic acid from the a-form of the
monomer and b-truxinic acid from the b-form of
the monomer.
Acknowledgements
We thank the Engineering and Physical Sciences
Research Council for the award of the grant
through which the microscopes were purchased
(Grant GR/R06908) and for a studentship to
SDMA. We are grateful to University of Reading
RETF for the award of a studentship to SLJ.
In the last part of the work, we have attempted
to generate crystals of the g-form of the monomer
by recrystallisation from ethanol following the
procedure of Mirsa [24]. Infrared and Raman
spectroscopic analysis showed that what had, in
fact, been obtained was a mixture of crystals of the
a- and b-forms. All attempts to generate the g-
form met with the same result, i.e. a mixture of
crystals of the a- and b-forms. Photolysis of such
References
[1] M.A. Garcia-Garibay, S. Shin, C.N. Sanrame, Tetrahe-
dron 56 (2000) 6729.
crystals resulted in dimerisation*giving a-trux-
/
[2] G. Kaupp, H. Frey, G. Behmann, Chem. Ber. 121 (1988)
2135.
illic and b-truxinic acids. We could find no
spectroscopic evidence for the existence of a third
(g) form of trans-cinnamic acid in any of our
experiments.
[3] M. Sakamoto, N. Sekine, H. Miyoshi, T. Mino, T. Fujita,
J. Am. Chem. Soc. 122 (2000) 10210.
[4] M. Sakamoto, M. Takahashi, W. Arai, T. Mino, K.
Yamaguchi, S. Watanabe, T. Fujita, Tetrahedron 56
(2000) 6795.
[5] G. Kaupp, M. Haak, Angew Chem. Int. Ed. 35 (1996)
2774.
4. Conclusions
[6] K. Tanaka, E. Mochizuki, N. Yasui, Y. Kai, I. Miyahara,
K. Hirotsu, F. Toda, Tetrahedron 56 (2000) 6853.
[7] G. Kaupp, J. Schmeyers, A. Kuse, A. Atfeh, Angew Chem.
Int. Ed. 38 (1999) 2896.
Our experiments have allowed us to form the
following conclusions.
[8] H. Koshima, D. Matsushige, M. Miyauchi, J. Fujita,
Tetrahedron 56 (2000) 6845.
1) Infrared and Raman microspectroscopy pro-
vide a convenient and rapid method to de-
monstrate the existence of different crystal
forms of trans-cinnamic acid. The techniques
allow the a- and b-forms to be readily
distinguished.
[9] Y. Ito, H. Hosomi, S. Ohba, Tetrahedron 56 (2000) 6833.
[10] C.B. Aaakeroy, A.M. Beatty, M. Nieuwenhuyzen, M. Zou,
Tetrahedron 56 (2000) 6693.
[11] H. Ihmels, D. Leusser, M. Pfeiffer, D. Stalke, Tetrahedron
56 (2000) 6867.
[12] H. Irngartinger, M. Skipinski, Tetrahedron 56 (2000) 6781.

S.D.M. Atkinson et al. / Spectrochimica Acta Part A 59 (2003) 629ꢁ
/
635
635
[13] V. Enklemann, G. Wegner, K. Novak, K.B. Wagener, J.
Am. Chem. Soc. 115 (1993) 10390.
[26] S.D.M. Allen, M.J. Almond, J.-L. Bruneel, A. Gilbert, P.
Hollins, J. Mascetti, Spectrochim. Acta Part A 56 (2000)
2423.
[14] G. Kaupp, Angew Chem. Int. Ed. 31 (1992) 592.
[15] G. Kaupp, Angew Chem. Int. Ed. 31 (1992) 595.
[16] G. Kaupp, M. Haak, J. Phys. Org. Chem. 8 (1995) 545.
[17] G. Kaupp, J. Schmeyers, F. Toda, H. Takumi, H.
Koshima, J. Phys. Org. Chem. 9 (1996) 795.
[27] M.D. Cohen, G.M.J. Schmidt, J. Chem. Soc. (1964) 1996.
[28] M.D. Cohen, G.M.J. Schmidt, F.I. Sonntag, J. Chem. Soc.
(1964) 2000.
[29] J. Ladell, T.R.R. McDonald, G.M.J. Schmidt, Acta Cryst.
9 (1965) 195.
[18] A.W.K. de Jong, Receuil 31 (1912) 258.
[19] H. Stobbe, A. Lehfeldt, Chem. Ber. 58 (1925) 2415.
[20] A.W.K. de Jong, Chem. Ber. 55 (1922) 463.
[21] A.W.K. de Jong, Chem. Ber. 56 (1923) 818.
[22] H. Stobbe, F.K. Steinberger, Chem. Ber. 55 (1922) 2225.
[23] G.M.J. Schmidt, J. Chem. Soc. (1964) 2014.
[24] T.N. Mirsa, Vib. Spec. Struct. 17A (1989) 137.
[25] S.D.M. Atkinson, PhD Thesis, University of Reading,
2001.
[30] D.A. Wierda, T.L. Feng, A.R. Barron, Acta Cryst. C45
(1989) 338.
[31] S. Kanao, S. Kashino, M. Haisa, Acta Cryst. C46 (1990)
2439.
[32] R.S. Gopalan, G.U. Kulkarni, Proc. Indian Acad. Sci.
(Chem. Sci.) 113 (2001) 307.
[33] M.A. Fernandes, D.C. Levendis, C.B. de Koning, Cryst.
Eng. 4 (2001) 215.