Liquid forms of pharmaceutical co-crystals: Exploring the boundaries of salt formation

Article abstract of DOI:10.1039/c0cc04485g

We present evidence of hydrogen bond formation, not salt formation, as the driving force in the liquefaction of a solid pharmaceutical in the form of a neutral acid-base complex, as exemplified by the liquid formed from a mixture of the local anesthetic l

Full text of DOI:10.1039/c0cc04485g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Liquid forms of pharmaceutical co-crystals: exploring the boundaries
of salt formationw
Katharina Bica,*ab Julia Shamshina,c Whitney L. Hough,c Douglas R. MacFarlaned and
Robin D. Rogers*bc
Received 18th October 2010, Accepted 17th November 2010
DOI: 10.1039/c0cc04485g
We present evidence of hydrogen bond formation, not salt
formation, as the driving force in the liquefaction of a solid
pharmaceutical in the form of a neutral acid–base complex, as
exemplified by the liquid formed from a mixture of the local
anesthetic lidocaine with fatty acids; these complexes exist at the
boundary between simple eutectics and partially ionised ionic
liquids.
The constant search for new drug forms and formulations to
feed the pharmaceutical pipeline has raised tremendous interest
in crystal engineering of pharmaceutical co-crystals formed
between a molecular or ionic active pharmaceutical ingredient
(API) and a co-crystal former that is a solid under ambient
conditions.1 The attractive features of a co-crystal composi-
tion typically arise as a result of specific supramolecular inter-
actions (e.g., hydrogen bonding) between the active ingredient
Fig. 1 Local anesthetic lidocaine 1 and fatty acids 2–6.7
and the co-former which crystallize together in a new solid
state form without the need to make or break covalent bonds.2
As the name ‘co-crystal’ implies, the goal of this field is to
produce new crystalline pharmaceuticals, in keeping with the
long-standing reliance of the pharmaceutical industry and
regulators on crystalline drug forms. However, co-crystals
suffer from some of the same problems as any solid drug
form, including polymorphism.3 These same issues motivated
us to explore whether liquid salt forms of pure drugs could be
used as an additional strategy to improve the API performance.
Such liquid salts form part of the large family of ionic liquids
(ILs), being salts that melt below 100 1C; here specifically
below room or body temperatures.4,5
a phenomenon whereby liquefaction of solid acids and bases
occurred due to hydrogen bonding and not proton transfer.
The close analogy between this phenomenon and the field of
co-crystals and the overlapping goals of that field with ours led
us to examine the potential utility of ‘liquid co-crystals’ as
another strategy to solve the numerous problems with modern
pharmaceuticals.
Unlike an IL form of an API, which might be considered a
new pharmaceutical entity, the approach we propose here
combines molecules whose toxicity profiles are already well
known and whose use in pharmaceutical applications is
well accepted. In addition, these new liquid compositions
can be prepared with variable stoichiometry in contrast to
pharmaceutical co-crystals which are crystalline and of defined
stoichiometry.
Our search for new IL formulations of the local anesthetic
drug lidocaine (Lid, 1) incorporating lipophilic counterions
such as fatty acids led us to study short (hexanoic acid, 2),
medium (decanoic acid, 3), and long-chain (stearic acid, 4)
fatty acids, as well as mono (oleic acid, 5), and double
(linoleic acid, 6) (Z-)unsaturated acids (Fig. 1).6 We observed
In order to develop straightforward syntheses for these new
formulations without the risk of solvent-, halide-, or metal
contamination, Lid as the free base was melted with a stoichio-
metric amount of the corresponding fatty acid until a clear
liquid was formed. For stearic acid (4), a colorless solid
crystallized after cooling with a melting point (Tm) of 42.8 1C
(Fig. 2; Fig. S8, Table S1, ESIw).
For the remaining mixtures, crystallisation was never observed
and low viscosity liquid samples with glass transition tempera-
tures (Tg) of BÀ50 1C or below were obtained, even when
solid Lid was mixed with the solid decanoic acid (3). The
introduction of double bonds has a dramatic influence, for
example comparing stearic (4) to oleic (5) acid, producing no
observable Tm and further reduction of Tg after introduction
of a second double bond. This behaviour is not totally
a Institute of Applied Synthetic Chemistry,
Vienna University of Technology, 1060 Vienna, Austria.
E-mail: kbica@ioc.tuwien.ac.at
b The Queen’s University of Belfast, QUILL,
School of Chemistry and Chemical Engineering,
Belfast BT9 5AG, Northern Ireland
c The University of Alabama, Department of Chemistry
and Center for Green Manufacturing, Tuscaloosa, AL 35487, USA.
E-mail: rdrogers@as.ua.edu
d School of Chemistry, Monash University, Wellington Rd., Clayton,
Victoria 3800, Australia
w Electronic supplementary information (ESI) available: Experimental
procedures and copies of spectra and graphs. See DOI: 10.1039/
c0cc04485g
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 2267–2269 2267

View Article Online
involved; only a glass transition was observed (Fig. 3; Fig. S4
and S5, ESIw). This is remarkably similar to the ‘‘Deep
Eutectic Solvents’’ described by Abbot for aprotic quaternary
ammonium salts in combination with strong hydrogen bond
donors such as amines, amides, or carboxylic acids.12 It
was further noted that once lidocaine was present in excess
(woleic o 0.5), the excess base can crystallize from the mixture,
which was confirmed by single crystal and powder X-ray
diffraction of the precipitate.
Our results suggest that the Lid : decanoic acid and
Lid : oleic acid systems might be best described as ‘‘deep eutectics’’,
with strong interactions of acid and base present in a certain
range resulting in a sudden drop in Tm. An NMR study of the
neat mixtures revealed that as woleic increased, a gradual shift
of the a-CH–N protons of Lid towards a lower field was
observed, indicating that the amine becomes involved in
a COO–HÁ Á ÁN hydrogen bond. This is in agreement with
15N HMBC spectroscopy, where the 15N chemical shift increases
as woleic increases, with a maximum shift difference for woleic
0.33–0.6, indicating the strongest hydrogen bonding in the
deep eutectic region. Additionally, a sharp drop in viscosity
was observed in this region of composition. Finally, IR spectra
also revealed the changes in the compositional range woleic
0.3–0.5 where a broad new peak at 1690 cmÀ1 was observed
only for these compositions (Fig. 4).
Fig. 2 Reduction or elimination of the melting point for 1 : 1
lidocaine : fatty acid mixtures.
unexpected, given that for the fatty acids themselves, the
presence of a double bond results in a lower Tm.8
Interestingly, despite a DpKa E 3 between fatty acid and
Lid, IR spectroscopy indicates no salt formation, although a
DpKa of B3 would normally be sufficient to generate substan-
tial proton transfer in aqueous solution.9 In all cases, the
CQO stretch of the fatty acid is in the range of B1690 cmÀ1
rather than B1560 cmÀ1, as would be expected for the COOÀ
anion.10 The drastic reduction or elimination of Tm thus has
to be explained by the presence of a non-ionic interaction
between Lid and the various fatty acids. When we performed
DSC analyses of the Lid : stearic acid formulations at different
compositions, we observed the presence of 2 endothermic
transitions, corresponding to the solidus (eutectic) and liquidus
transitions of the mixture. A eutectic point was found to
appear at a composition of Lid : stearic acid 2 : 1 (wstea 0.33)
at 42.9 1C (Fig. 3).11
The eutectic transition is clearly observed between wstea
0.1 and wstea 0.8. The decrease in the transition temperature
on either side of this range suggests solid solubility of the
minor phase in the major phase and hence solidus behaviour.
However, the phase diagrams of the systems lidocaine :
decanoic acid and lidocaine : oleic acid differ significantly.
For both formulations, we observed the expected reduction
of the melting point starting from the pure lidocaine or the
pure acid, until a certain composition at which the sample
does not crystallize at all, at the cooling and heating rates
Considering ionicity and the research focus on protic ILs,
the systems observed here seem to mark an unrecognized state
of proton interaction between acid and base in pure, solvent-
free mixtures. Previously a low degree of proton transfer had
been identified as producing ‘‘poor ILs’’ as a result of their
position in the Walden plot, meaning that, in fact they were
mixtures of the parent acid and base plus rather small amounts
of the salt.13 However, not all acid/base combinations having a
low degree of proton transfer exhibit the behavior observed
here; in fact at the other extreme, some acid/base combina-
tions are not even miscible when the hydrogen bond is not
sufficiently strong. Hence we envisage proton interaction/
transfer increasing in the order: immiscible mixtures o
eutectics o deep eutectics o partially ionized ILs o fully
ionized ILs. Thus, stronger proton interactions produce the
deeper eutectic that consequently merges into partial ioniza-
tion creating a 4 component mixture deepening the eutectic
Fig. 3 Transition temperatures for lidocaine : stearic acid (liquidus
line: red; solidus line: blue), lidocaine : decanoic acid (black), and
lidocaine : oleic acid (green).
Fig. 4 IR spectra of lidocaine : oleic acid mixtures.
c
2268 Chem. Commun., 2011, 47, 2267–2269
This journal is The Royal Society of Chemistry 2011

View Article Online
still further. Liquid samples were only obtained for the inter-
mediates between the eutectic and fully ionized salt, indicating
that the hydrogen bond interaction rather than proton transfer
reduces the Tm, an effect that we have previously observed in
oligomeric ILs.14
5 K. Bica, C. Rijksen, M. Nieuwenheuzen and R. D. Rogers, Phys.
Chem. Chem. Phys., 2010, 12, 2011–2017.
6 For other examples of fatty acid-derived, non-protic IL, see:
S. M. Murray, R. A. O’Brien, K. M. Mattson, C. Ceccarelli,
R. E. Sykora, K. N. West and J. H. Davis Jr., Angew. Chem.,
Int. Ed., 2010, 49, 2755.
7 The transition for Lid : Oleic acid in the DSC differs in shape from
a normal Tg, we therefore refer to this transition temperature as Tt.
8 D. M. Small, The Physical Chemistry of Lipids, Plenum Press,
New York, 1986, pp. 523–554.
9 J. Stoimenovski, E. I. Izgorodina and D. R. MacFarlane, Phys.
Chem. Chem. Phys., 2010, 12, 10341–10347.
10 S. Zhu, M. Heppenstall-Butler, M. F. Butler, P. D. A. Pudney,
D. Ferdinando and K. J. Mutch, J. Phys. Chem. B, 2005, 109,
11753.
11 For an example of eutectic formulations of Lid on the market, see:
S. Fiala, S. A. Jones and M. B. Brown, Int. J. Pharm., 2010, 393,
68.
12 (a) A. P. Abbott, D. Boothby, G. Capper, D. L. Davies and
R. K. Rasheed, J. Am. Chem. Soc., 2004, 126, 9142;
(b) A. P. Abbott, J. C. Barron, K. S. Ryder and D. Wilson,
Chem.–Eur. J., 2007, 13, 6495; (c) A. P. Abbott, G. Capper,
D. L. Davies, K. J. McKenzie and S. U. Obi, J. Chem. Eng. Data,
2006, 51, 1280; (d) A. P. Abbott, G. Capper and S. Gray,
ChemPhysChem, 2006, 7, 803; (e) A. P. Abbott, G. Capper,
D. L. Davies, R. Rasheed and V. Tambyrajah, Chem. Commun.,
2003, 70.
In summary, we have presented new liquid formulations of
the local anesthetic lidocaine with various fatty acids that,
based on their hydrogen-bonded nature, can be considered
as liquid equivalents to pharmaceutical co-crystals. It is
important to note that, in contrast to our previous work with
ionic liquids composed of pharmaceutically active ingredients,
these liquids are prepared from known compounds of known
toxicity and in current clinical use. Furthermore, the low ionicity
observed here in comparison to fully ionized salts provides the
possibility for these liquids to penetrate membranes more
efficiently and for a possible future use in transdermal drug
delivery and local anaesthesia.
Notes and references
1 N. Shan and M. J. Zaworotko, Drug Discovery Today, 2008, 9,
440.
2 M. C. Etter, J. Phys. Chem., 1991, 95, 4601.
3 P. H. Karpinski, Chem. Eng. Technol., 2006, 29, 233.
4 J. Stoimenovski, D. R. MacFarlane, K. Bica and R. D. Rogers,
Pharm. Res., 2010, 27, 521.
13 M. Yoshizawa, W. Xu and C. A. Angell, J. Am. Chem. Soc., 2003,
125, 15411.
14 K. Bica and R. D. Rogers, Chem. Commun., 2010, 46,
1215.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 2267–2269 2269