Crystal Growth & Design
ARTICLE
and similarly to the 1:1 cocrystals, a columnar assembly of the
molecules is obtained.
Although all four cocrystals utilize a predictable supramole-
cular synthon and similar hydrogen bonds are present in related
structures, the extended hydrogen bond motifs and the different
compositions of the cocrystals would be difficult to predict. Some
physicochemical properties of the fenamic acids, related to the
stability of the acid crystals, could be correlated with the stability
of the cocrystals. Yet, other properties, also associated with the
stability of the acid crystals, showed no such correlation. Similar
studies on groups of related cocrystals will be required to confirm
whether the correlations discussed in this work are of general
utility.
This similarity in the overall construction of the 1:1 and 1:2
cocrystals is in contrast with their different hydrogen bond
motifs. Presumably, the sizable hydrophobic region of the
fenamic acid molecules limits the extension of hydrogen bonds
into two- or three-dimensional networks, while favorable stack-
ing interactions between the fenamic acid molecules help the
formation of molecular columns. With regard to hydrogen
bonding, the 1:2 cocrystals show more structural resemblance
to nicotinamide than to the 1:1 cocrystals (Figure 1). In effect,
the only shared motif between all four cocrystals is the acidꢀaromatic
N synthon.
’ ASSOCIATED CONTENT
The cocrystal structures give little indication why 1:1 fenamic
acid:nicotinamide stoichiometry ispreferred over 1:2. The packing
coefficient of each cocrystal is 0.70ꢀ0.71, so improvements in
space filling cannot be used as an adequate explanation. These
cocrystals are formed because the acidꢀaromatic N synthon is
energetically more favorable than any of the interactions possible
in the one-component crystals. The relative weight of this
interaction is higher in the 1:1 cocrystals. The similarity of the
1:2 cocrystals to nicotinamide may further reduce their kinetic
stability, by facilitating the formation of nicotinamide crystals
during cocrystal decomposition.
S
Supporting Information. Solubility data, hydrogen bond
b
tables, PXRD and DSC data for the cocrystals, Rietveld refine-
ment plots, and CIF files. This material is available free of charge
’ AUTHOR INFORMATION
Corresponding Author
*Tel: +353 21 490 3143. Fax: +353 21 427 4097. E-mail: s.lawrence@
ucc.ie.
There is no obvious reason why flufenamic acid and niflumic
acid could not form 1:2 cocrystals similar to those of tolfenamic
and mefenamic acid. It is, however, clear from the relative
difficulty of their preparation that the 1:2 cocrystals are less
stable than the 1:1 cocrystals. Therefore, it is reasonable to
assume that when starting from a 1:2 mixture of flufenamic or
niflumic acid and nicotinamide, the mixture of a 1:1 cocrystal and
nicotinamide is a more stable product of the cocrystallization
reaction than a pure 1:2 cocrystal.
’ ACKNOWLEDGMENT
This publication has emanated from research conducted with
the financial support of Science Foundation Ireland, under Grant
Nos. 05/PICA/B802 TIDA 09, 05/PICA/B802/EC07, and 08/
RFP/MTR1664. We gratefully acknowledge Jana Cairns,
Lorraine Donaghy, and Tracy Walker for the preparation and
analysis of cocrystals. We are grateful to Fabia Gozzo (Paul
Scherrer Institute) for collecting synchrotron data.
The failure to form meclofenamic acid/nicotinamide cocrys-
tals is also hard to explain. For example, comparison with the
3 6 structure shows that the additional 60-Cl atom would be
3
positioned on the outside of the hydrogen-bonded helices. While
such a substitution may render close packing less optimal, it is not
in obvious conflict with the hydrogen bond network. Presum-
ably, the failure of cocrystallization is determined by interactions
between the hydrophobic fragments of the molecules, which are
much less predictable than hydrogen bonds.
’ REFERENCES
(1) Khansari, P. S.; Halliwell, R. F. Neurochem. Int. 2009, 55,
683–688.
(2) Lee, S.-H.; Bahn, J. H.; Whitlock, N. C.; Baek, S. J. Oncogene 2010,
29, 5182–5192.
(3) Pedersen, S. B. Pharmacol. Toxicol. 1994, 75 (Suppl. 2), 22–32.
(4) Glazko, A. J.; Chang, T.; Borondy, P. E.; Dill, W. A.; Young, R.;
Croskey, L. Curr. Ther. Res. 1978, 23, 22–41.
(5) Cafaggi, S.; Russo, E.; Caviglioli, G.; Parodi, B.; Stefani, R.; Sillo,
G.; Leardi, R.; Bignardi, G. Eur. J. Pharm. Sci. 2008, 35, 19–29.
(6) Vavia, P. R.; Adhage, N. A. Pharm. Dev. Technol. 2000, 5,
571–574.
’ CONCLUSION
Of the five fenamic acids investigated, four formed cocrystals
with nicotinamide, demonstrating the robustness of the car-
boxylic acidꢀaromatic N heterosynthon. The acid:amide stoi-
chiometry of the cocrystals was either 1:1 or 1:2, depending
on the acid involved. The 1:1 cocrystals were more stable and
were formed with the lowest melting, most soluble fenamic acids.
The structure of the 1:1 cocrystals suggests that steric conflict
involving the 20-substituents in the other three acids prevents
them from forming 1:1 cocrystals.
Obtaining the two 1:2 cocrystals in sufficient purity for
structure determination was only possible using grinding and
sonication, implying that scale-up of these cocrystals using
solution based methods would require a detailed knowledge
of the ternary phase diagram. Analysis of the structures of the
1:2 cocrystals does not provide a clear explanation for why
only two of the five fenamic acids formed cocrystals with this
stoichiometry.
(7) Del Arco, M.; Fernandez, A.; Martin, C.; Sayalero, M. L.; Rives,
V. Clay Miner. 2008, 43, 255–265.
(8) Pandey, M.; Saraf, S. A. J. Pharm. Res. 2010, 3, 146–150.
(9) (a) Fang, L.; Numajiri, S.; Kobayashi, D.; Ueda, H.; Nakayama,
K.; Miyamae, H.; Morimoto, Y. J. Pharm. Sci. 2004, 93, 144–154. (b)
Fonari, M. S.; Ganin, E. V.; Vologhzanina, A. V.; Antipin, M. Y.;
Kravtsov, V. C. Cryst. Growth Des. 2010, 10, 3647–3656.
(10) Thybo, P.; Kristensen, J.; Hovgaard, L. Pharm. Dev. Technol.
2007, 12, 43–53.
(11) McNamara, D. P.; Childs, S. L.; Giordano, J.; Iarriccio, A.;
Cassidy, J.; Shet, M. S.; Mannion, R.; O'Donnell, E.; Park, A. Pharm. Res.
2006, 23, 1888–1897.
(12) (a) Bꢀathori, N. B; Lemmerer, A.; Venter, G. A.; Bourne, S. A.;
Caira, M. R. Cryst. Growth Des. 2011, 75–87. (b) Eccles, K. S.; Deasy,
R. E.; Fꢀabiꢀan, L.; Maguire, A. R.; Lawrence, S. E. J. Org. Chem. 2011,
76, 1159–1162. (c) Lemmerer, A.; Bꢀathori, N. B; Bourne, S. A. Acta
3527
dx.doi.org/10.1021/cg200429j |Cryst. Growth Des. 2011, 11, 3522–3528