Communications
[6] The minimum-energy structure was just 0.2 kJmolÀ1 lower in
able arrangements, that is, the information regarding the
À
energy than the established crystal structure, and in our opinion
this value is so small that the order of energies might just as
easily have been reversed.
particular sequence of the O H···O layers. See Ref. [12].
[15] Neglecting any diffuse scattering is clearly a simplification.
However, the procedures that we describe have the advantage
that they are standard (that is, they are based only on procedures
used for routine crystallographic analyses), require no specialist
knowledge, and can be applied in any crystallographic labora-
tory.
[7] P. Vishweshwar, J. A. McMahon, M. Oliveira, M. L. Peterson,
M. J. Zaworotko, J. Am. Chem. Soc. 2005, 127, 16802.
[8] The highest form II domain ratio observed from 15 single-crystal
data collections is in crystal 1b. For this crystal, conventional
refinement of the form II structure gives: C9H8O4, Mr = 180.15,
monoclinic, space group P21/c, a = 12.1515(10), b = 6.5064(5),
c = 11.3677(9) , b = 111.574(3)8, V= 835.79(12) 3, T=
180(2) K, m(MoKa) = 0.114 mmÀ1, 2qmax = 508, 9778 reflections
measured, 1474 unique reflections (Rint = 0.029), 1149 observed
reflections, R1(I>2s(I)) = 0.041, wR2(all data) = 0.116, S = 1.02.
CCDC-617840 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
m.ac.uk/data_request/cif.
[16] The term “form II data set” refers to the list of intensities (the
hkl file) that is obtained when the diffraction data are indexed on
the basis of the form II unit cell and integrated using standard
procedures. The additional Bragg reflections that could be
described on the basis of the form I unit cell and any diffuse
features are ignored. The term “form I data set” refers to the
opposite situation.
[17] In particular, there may be problems associated with assessment
of the background intensity in the presence of diffuse streaks in
the vicinity of the Bragg reflections.
[9] Compared to the powder X-ray diffraction pattern of pure
form I, the principal features that reveal the presence of form II
domains are peaks at 2q ꢀ 20.9 and 26.08 (at 295 K, see the
Supporting Information).
[18] The correlation is derived from the experimental form I data set
(crystal 3), with a suitable fractional scale factor applied to the
odd l reflections (see the Supporting Information for full
details). The batch refinement must be based on all F2 data.
The Figure provides an immediate approximate guide, based on
standard refinement procedures. Application of the batch
refinement procedure will provide an improved estimate.
[19] R. J. Davey, K. Allen, N. Blagden, W. I. Cross, H. F. Lieberman,
M. J. Quayle, S. Righini, L. Seton, G. J. T. Tiddy, CrystEngComm
2002, 4, 257.
[20] Since interlayer arrangements A and B are structurally and
energetically so finely balanced, disorder is thermodynamically
inherent in the aspirin system. All aspirin crystals should contain
some of arrangements A and B at real temperatures. They are
detectable by X-ray diffraction only when their number becomes
significant.
[10] With extremely rapid cooling of a hot equimolar solution of
aspirin and acetamide in acetonitrile (that is, by plunging the
flask directly into liquid N2 or dry ice/2-propanol), we obtained a
1:1 molecular complex of aspirin and acetamide, which contains
the expected acid–amide heterosynthon. This would argue in
favor of acetamide acting as a growth inhibitor at the (200) faces
of aspirin, leading to platelike crystals that cannot grow easily
along [100]. It is likely that levetiracetam would behave in a
similar way, and this is the morphology reported by PZ for their
aspirin crystal. Crystal data for the aspirin/acetamide (1:1) co-
crystal: (C9H8O4)(C2H5NO), Mr = 239.22, orthorhombic, space
group Pbca, a = 9.5268(4), b = 8.7495(3), c = 28.4520(11) , V=
2371.61(16) 3, T= 203(2) K, m(MoKa) = 0.107 mmÀ1
,
68487
[21] In general, for any system of hydrogen bonds (O H···O or C
H···O) wherein the dimer and catemer alternatives are well
matched, there is a consensus that the catemer is kinetically
favored since it is a higher dimensional pattern than the dimer.
See: D. Das, G. R. Desiraju, Chem. Asian J. 2006, 1, 231.
[22] PZ have reported that “form II” transforms into form I at 100 K.
Our assessment of the stability of the form II domains is based
on re-collection of single-crystal data for 1b after the crystal had
stood in air for two months.
À
À
reflections measured, 3074 unique reflections (Rint = 0.059),
2586 observed reflections, R1(I>2s(I)) = 0.042, wR2(all
data) = 0.119, S = 1.04. CCDC-618012 contains the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
[11] a) H. Jagodzinski, Acta Crystallogr. 1949, 2, 201; b) H. Jagod-
zinski, Acta Crystallogr. 1949, 2, 208; c) “Disorder Diffuse
Scattering of X-rays and Neutrons”: H. Jagodzinski, F. Frey in
International Tables for Crystallography, Vol. B (Ed.: U.
Shmueli), International Union of Crystallography and Kluwer
Academic Publishers, Dordrecht, 2001.
[23] Co-crystal: a) G. R. Desiraju, CrystEngComm 2003, 5, 466;
b) J. D. Dunitz, CrystEngComm 2003, 5, 468. Pseudopolymorph:
c) K. R. Seddon, Cryst. Growth Des. 2004, 4, 1088; d) G. R.
Desiraju, Cryst. Growth Des. 2004, 4, 1089; e) J. Bernstein, Cryst.
Growth Des. 2005, 5, 1661; f) A. Nangia, Cryst. Growth Des.
2006, 6, 1.
[12] H.-B. Bürgi, M. Hostettler, H. Birkedal, D. Schwarzenbach, Z.
Kristallogr. 2005, 220, 1066.
[13] a) K. Dornberger-Schiff, H. Grell-Niemann, Acta Crystallogr.
1961, 14, 167; b) K. Dornberger-Schiff, J. D. Dunitz, Acta
Crystallogr. 1965, 19, 471; c) K. Dornberger-Schiff, Acta Crys-
tallogr. Sect. A 1982, 38, 483; d) D. Schwarzenbach, K. Kirsch-
baum, A. A. Pinkerton, Acta Crystallogr. Sect. B 2006, 62, 944.
[14] In crystallographic parlance, the Bragg reflections with even l
are referred to as “family reflections” and are common to both
the form I and form II structures. The family reflections alone
define a “superposition structure”, which in the case of aspirin
[24] However, one of us has argued (in Ref. [23d]) that difficulties in
the use of the word “pseudopolymorph” might have arisen
because the definition of the term “polymorph” is too rigid with
respect to identity of chemical composition among substances
that are deemed to qualify for the polymorph epithet.
[25] To some extent, this situation parallels that of isomorphous
crystalline solvates, for which patents have been granted for
ranges of solvent content. However, while variable solvent
content is a common occurrence, it is not clear how common the
aspirin situation will be.
À
comprises two O H···O dimers overlaid in a unit cell with c
dimension half that of the true c dimension. The distribution of
the intensities along the odd l rows contains the information to
decompose the superposition structure into chemically reason-
[26] A. R. Verma, P. Krishna, Polymorphism and Polytypism in
Crystals, Wiley, New York, 1966.
[27] G. Coquerel Chem. Eng. Proc. 2006, 45, 857
622
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 618 –622