Supramolecular architectures of meloxicam carboxylic acid cocrystals, a crystal engineering case study

Article abstract of DOI:10.1021/cg100514g

Meloxicam is a nonsteroidal anti-inflammatory drug with low aqueous solubility and high permeability prescribed for indications of arthritis, primary dysmenorrhea, fever, and pain. In this contribution, we apply crystal engineering and the supramolecular synthon approach to prepare novel meloxicam cocrystal forms with various pharmaceutically acceptable or toxicologically qualified carboxylic acids. As a result, 19 pharmaceutical cocrystals including one cocrystal of a salt are synthesized by solid-state and solution methods. All resulting cocrystals are characterized by X-ray diffraction, infrared, and thermal analyses. In particular, crystal structures of six meloxicam cocrystals are determined and reported, namely, meloxicam?1-hydroxy-2-naphthoic acid cocrystal (1), meloxicam?glutaric acid cocrystal (2), meloxicam?l- malic acid cocrystal of a salt (3), meloxicam?salicylic acid cocrystal form III (4), meloxicam?fumaric acid cocrystal (5), and meloxicam?succinic acid cocrystal (6). The supramolecular assembly of each cocrystal is analyzed and discussed. It is observed that the meloxicam dimer is robust since this motif is observed in five out of six meloxicam cocrystal structures that have been determined. As part of the continuous development, the resulting meloxicam cocrystal forms will be further investigated to explore improved physicochemical and pharmacological properties.

Full text of DOI:10.1021/cg100514g

DOI: 10.1021/cg100514g
Supramolecular Architectures of Meloxicam Carboxylic Acid Cocrystals,
a Crystal Engineering Case Study
2010, Vol. 10
4401–4413
Miranda L. Cheney,†,‡ David R. Weyna,†,‡ Ning Shan,*,† Mazen Hanna,† Lukasz Wojtas,‡
and Michael J. Zaworotko*,‡
†Thar Pharmaceuticals Inc., 3802 Spectrum Boulevard, Suite 120, Tampa, Florida 33612, and
‡Department of Chemistry, University of South Florida, 4202 East Fowler Avenue, CHE205,
Tampa, Florida 33620
Received April 17, 2010; Revised Manuscript Received August 2, 2010
ABSTRACT: Meloxicam is a nonsteroidal anti-inflammatory drug with low aqueous solubility and high permeability
prescribed for indications of arthritis, primary dysmenorrhea, fever, and pain. In this contribution, we apply crystal engineering
and the supramolecular synthon approach to prepare novel meloxicam cocrystal forms with various pharmaceutically
acceptable or toxicologically qualified carboxylic acids. As a result, 19 pharmaceutical cocrystals including one cocrystal of a
salt are synthesized by solid-state and solution methods. All resulting cocrystals are characterized by X-ray diffraction, infrared,
and thermal analyses. In particular, crystal structures of six meloxicam cocrystals are determined and reported, namely,
meloxicam 1-hydroxy-2-naphthoic acid cocrystal (1), meloxicam glutaric acid cocrystal (2), meloxicam L-malic acid cocrystal
3
3
3
of a salt (3), meloxicam salicylic acid cocrystal form III (4), meloxicam fumaric acid cocrystal (5), and meloxicam succinic acid
3
3
3
cocrystal (6). The supramolecular assembly of each cocrystal is analyzed and discussed. It is observed that the meloxicam dimer
is robust since this motif is observed in five out of six meloxicam cocrystal structures that have been determined. As part of the
continuous development, the resulting meloxicam cocrystal forms will be further investigated to explore improved physico-
chemical and pharmacological properties.
1. Introduction
exhibit by utilizing the Cambridge Structural Database
(CSD)12 and effectively prioritizes all possible guest molecules
for crystal form screening of drugs. Such an approach can be
generally effective11b,d-i but has found particular success in
generating pharmaceutical cocrystals.9,13
Crystal form selection for oral delivery of active pharma-
ceutical ingredients (APIs) remains a high priority for the
pharmaceutical industry.1 In general, pharmaceutical crystal
forms are preferred by developers and regulatory authorities,
because crystallization tends to afford highly pure products
that are superior with respect to reproducibility and scalability.1
Moreover, a unique crystalline form may exhibit distinctive
physicochemical properties and could in turn affect the dis-
solution, manufacturing, physical stability, permeability, and
oral bioavailability of an API.2,3 Thus, it is apparent that
selection of an appropriate crystal form for downstream deve-
lopment and processing is of primary importance in pharma-
ceutical development.4
A typical crystal form selection process comprises two stages
of development after a target API molecule has been selected:
discover as many pharmaceutical crystal forms as possible;
then examine the physicochemical properties of the newly
discovered crystal forms. At the stage of crystal form discov-
ery, two primary approaches are used. The more straightfor-
ward approach is largely based on trial-and-error (e.g., high-
throughput crystal form screening) and has been implemented
to discover crystal forms including, but not limited to, salts,3
hydrates,5 solvates,6 and, more recently, cocrystals.7,8 The
alternative approach for crystal form discovery is the supra-
molecular synthon approach,9 which recognizes supramole-
cular synthons10 as a design tool and can be more selective,
time-efficient, and cost-effective. The supramolecular synthon
approach uses crystal engineering11 to carefully analyze the
relevant supramolecular arrangements that an API might
Pharmaceutical cocrystals are defined as multiple compo-
nent crystals in which at least one component is molecular and a
solid at room temperature (the coformer) and forms a supra-
molecular synthon with a molecular or ionic API.14 A given
API may form cocrystals with numerous pharmaceutically
acceptable or toxicologically qualified materials, and these
cocrystals could exhibit enhanced solubility,9,15,16 stability to
hydration, or compressibility.17 Therefore, pharmaceutical
cocrystals represent an opportunity to diversify the number
of crystal forms of a given API and in turn fine-tune or even
customize its physicochemical properties without the need for
chemical (covalent) modification. In addition, the supramole-
cular synthon approach can be implemented by various co-
crystal preparative methods such as solution, slurry, grinding,
or melt.18,19 Numerous APIs that exhibit undesirable solubility
or stability and possess multiple hydrogen-bonding sites have
been (or potentially can be) studied in the context of cocrystal-
lization.15,20 As one of these potential APIs, meloxicam is identi-
fied and is of interest for cocrystallization.
Meloxicam, 4-hydroxy-2-methyl-N-(5-methyl-2-thiazolyl)-
2H-1,2-benzothiazine-3-carboxamide-1,1-dioxide, is a non-
steroidal anti-inflammatory drug (NSAID) and an antipyretic
drug used for indications of rheumatoid and osteoarthritis,21
postoperative pain,22 and fever.23 Meloxicam is effective in
relieving various types of pain and causes fewer side effects
thanotherNSAIDs.24 Comparedwithother drugs in the same
class such as piroxicam, meloxicam is preferred due to its
ability to selectively inhibit cyclooxygenase-2 (COX-2).25
Originally developed by Boehringer Ingleheim, meloxicam is
*To whom correspondence should be addressed. E-mail addresses:
nshan@tharpharma.com (N. Shan); xtal@usf.edu (M. J. Zaworotko).
r
2010 American Chemical Society
Published on Web 08/30/2010
pubs.acs.org/crystal

4402 Crystal Growth & Design, Vol. 10, No. 10, 2010
Scheme 1. Enolic, Anionic, Cationic, and Zwitterionic Forms of Meloxicam
Cheney et al.
marketed in Europe under the brand names Melox or Recoxa
and in the U.S. as Mobic. Meloxicam is available as an oral
tablet (7.5 or 15 mg dose) and as an oral suspension (7.5 mg/
5 mL dose).
acid. Interestingly, pharmaceutical cocrystals of piroxicam,
an API with a similar molecular structure to meloxicam
(i.e., piroxicam possesses a pyridyl group rather than a thiazole
group), have been reported and are used as a reference for this
study.37 Clearly, pharmaceutical cocrystallization of meloxicam
represents a promising approach to diversify the novel crystal
forms, which can then be used to improve the relevant aqueous
solubility and accelerate the onset of action for mild or medium
level pain relief. In this contribution, we use the supramolecular
synthon approach to effectively prioritize all available cofor-
mers and systematically prepare and characterize a variety of
meloxicam cocrystals with carboxylic acids.
Meloxicam is a yellow solid that is practically insoluble in
water (0.012mg/mL, 25°C)26 and is considereda ClassIIdrug
(i.e., a low-solubility and high-permeability drug) by the
Biopharmaceutics Classification System (BCS).27 Meloxicam
is very slightly soluble in various organic solvents26 and has
variable aqueous solubility related to its pH-dependent ioniza-
tion states. Under acidic conditions, meloxicam is present in
solution in its cationic form, while in basic solutions it is present
in its anionic form. When the molecule is neutral in charge,
meloxicam will either be in its zwitterionic or enolic form,
depending on the polarity of the solvent.28,29 The different
ionization states of meloxicam are shown in Scheme 1.
Due to its low solubility under acidic and neutral conditions,
the Tmax (time to reach maximum concentration) of meloxicam
in the human body is typically 4-6 hours, while it can take more
than 2 hours for the drug to reach its therapeutic concentration
in humans.30 The slow onset of meloxicam prevents meloxicam
from its potential application for the relief of mild or medium
level acute pain. To accelerate its onset of action, various
complexes of meloxicam have been prepared and evaluated
with respect to aqueous solubility, resulting in cyclodextrin
inclusion complexes,31 various solvates,26 ethanolamine,32 am-
monium, and sulfate salts,28 or metal complexes with potassium
and calcium.33 Preparation of polymorphic crystal forms of
meloxicam29 has also been attempted to improve its dissolution
profile.34 Five polymorphic forms of meloxicam have been
reported in the literature.29 The stability and potential for
solid-phase interconversion between various meloxicam poly-
morphic forms has not been investigated.35 Despite all the
efforts that have been taken, a faster onset dosage form for oral
administration of meloxicam remains unknown to date. Only
very recently did Myz et al.36 report the preparation of two
pharmaceutical cocrystals employing maleic acid and succinic
2. Materials and Methods
2.1. Materials. Meloxicam was purchased from Jai Radhe Sales,
India, with a purity of 99.64% and was used without further
purification. All other chemicals were supplied by Sigma-Aldrich
and used without further purification.
2.2. Synthesis of Cocrystals 1-19. Meloxicam was reacted with 18
selected coformers: 1-hydroxy-2-naphthoic acid, glutaric acid,
L-malic acid, salicylic acid, fumaric acid, succinic acid, maleic acid,
malonic acid, gentisic acid, 4-hydroxybenzoic acid, adipic acid,
(þ)-camphoric acid, glycolic acid, benzoic acid, DL-malic acid, hydro-
cinnamic acid, ascorbic acid, and L-tartaric acid. All coformers
(Tables 1 and 2) except ascorbic acid and L-tartaric acid produced
at least one meloxicam cocrystal form. The cocrystallization attempts
resulted in 19 crystal forms, many of which wereprepared via multiple
synthetic techniques including solvent-drop grinding18 and slurrying.
Single crystals suitable for X-ray diffraction were successfully pre-
pared for six of the 19 cocrystals (1-6).
Synthesis of Meloxicam 1-Hydroxy-2-naphthoic Acid (1:1)
3
Cocrystal (1). (a) Solvent-drop grinding: 0.176 g (0.501 mmol) of
meloxicam was ground with 0.0957 g (0.508 mmol) of 1-hydroxy-2-
naphthoic acid and 40 μL of tetrahydrofuran (THF) for 30 min in a
ball-mill. Cocrystal 1 was generated in ca. 100% yield. (b) Slurry:
0.700 g (1.99 mmol) of meloxicam and 0.391 g (2.07 mmol) of
1-hydroxy-2-naphthoic acid were slurried at ca. 250 rpm in 3 mL of
THF overnight sealed under ambient conditions. The resulting solid
was filtered and washed with THF. Cocrystal 1 was obtained in ca.
90% yield. (c) Solution Crystallization: 0.0176 g (0.0501 mmol) of

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4403
Table 1. Molecular Structure, pKa,49,50 ΔpKa, and Melting Point Values for Meloxicam and Coformers That Form Cocrystals 1-6
meloxicam and 0.0095 g (0.0505 mmol) of 1-hydroxy-2-naphthoic
acid were dissolved in 10 mL of ethyl acetate and left to slowly
evaporate. Cocrystal 1 was obtained in ca. 93% yield.
THF overnight sealed under ambient conditions. The resulting solid
was filtered and washed with THF. Cocrystal 4 was isolated in ca.
84% yield. (c) Solution crystallization: 0.0226 g (0.0643 mmol) of
meloxicam and 0.0787 g (0.570 mmol) of salicylic acid were dis-
solved in 8 mL of a 6:2 mixture of ethyl acetate and 1,4-dioxane and
left to slowly evaporate. Single crystals of 4 (ca. 33% yield) grew
concomitantly with meloxicam form I and salicylic acid.
Synthesis of Meloxicam Glutaric Acid (1:1) Cocrystal (2). (a)
3
Solvent-drop grinding: 0.179 g (0.511 mmol) of meloxicam was ball-
milled with 0.0699 g (0.529 mmol) of glutaric acid and 40 μL of
chloroform for 30 min, producing 2 in ca. 100% yield. (b) Slurry:
0.892 g (2.54 mmol) of meloxicam and 0.351 g (2.66 mmol) of
glutaric acid were slurried at ca. 250 rpm in 3 mL of ethyl acetate
overnight sealed under ambient conditions. The resulting solid was
filtered and washed with ethyl acetate. Cocrystal 2 was made in
ca. 96% yield. (c) Solution crystallization: 0.0194 g (0.0552 mmol) of
meloxicam and 0.159 g (1.20 mmol) of glutaric acid were dissolved
in 2 mL of ethyl acetate and left to slowly evaporate. Cocrystal 2 was
isolated in ca. 89% yield.
Synthesis of Meloxicam Fumaric Acid (2:1) Cocrystal (5). (a)
3
Solvent-drop grinding: 0.088 g (0.250 mmol) of meloxicam was ball-
milled with 0.015 g (0.129 mmol) of fumaric acid and 50 μL of THF
for 30 min, generating 5 in ca. 100% yield. (b) Slurry: 0.880 g (2.50
mmol) of meloxicam and 0.150 g (1.29 mmol) of fumaric acid were
slurried at ca. 250 rpm in 3 mL of THF overnight sealed under
ambient conditions. The resulting solid was filtered and washed
with THF. Cocrystal 5 was isolated in ca. 81% yield. (c) Solution
crystallization: 0.100 g (0.284 mmol) of meloxicam and 0.330 g
(0.284 mmol) of fumaric acid were dissolved in 9 mL of a THF and
left to slowly evaporate resulting in single crystals of 5 (ca. 31%
yield).
Synthesis of Meloxicam
3 L-Malic Acid (1:1:1) Cocrystal of a Salt
(3). (a) Solvent-drop grinding: 0.176 g (0.501 mmol) of meloxicam
was ground together with 0.0361 g (0.269 mmol) of L-malic acid and
40 μL of THF for 30 min in a mechanical ball-mill. Cocrystal 3 was
generated in ca. 100% yield. (b) Slurry: 0.897 g (2.55 mmol) of
meloxicam and 0.182 g (1.36 mmol) of L-malic acid were slurried at
ca. 250 rpm in 3 mL of THF overnight sealed under ambient
conditions. The resulting solid was filtered and washed with THF
and afforded a ca. 92% yield of 3. (c) Solution crystallization:
0.0214 g (0.0609 mmol) of meloxicam and 0.0416 g (0.301 mmol) of
L-malic acid were dissolved in 2 mL of a 1:1 mix of 1,4-dioxane and
ethyl acetate and left to slowly evaporate. The resulting single
crystals of 3 were isolated in ca. 82% yield.
Synthesis of Meloxicam Succinic Acid (2:1) Cocrystal (6). (a)
3
Solvent-drop grinding: 0.088 g (0.250 mmol) of meloxicam was ball-
milled with 0.015 g (0.127 mmol) of succinic acid and 50 μL of THF
for 30 min, generating 6in ca. 100% yield. (b) Slurry: 0.880 g (2.50 mmol)
of meloxicam and 0.150 g (1.27 mmol) of succinic acid were slurried at ca.
250 rpm in 3 mL of THF overnight sealed under ambient conditions.
The resultingsolid was filteredandwashed with THF. Cocrystal6 was
isolated in ca. 78% yield. (c) Solution crystallization: 0.100 g (0.284
mmol) of meloxicam and 0.017 g (0.142 mmol) of succinic acid was
dissolved in 10 mL of 1:1 THF and left to slowly evaporate. Single
crystals of 6 (ca. 55% yield) grew concomitantly with meloxicam form
I and succinic acid.
Synthesis of Meloxicam Salicylic Acid (1:1) Cocrystal Form III
3
(4). (a) Solvent-drop grinding: 0.174 g (0.495 mmol) of meloxicam
was ball-milled with 0.0724 g (0.524 mmol) of salicylic acid and
40 μL of ethyl acetate for 30 min. Cocrystal 4 was synthesized in
ca. 100% yield. (b) Slurry: 0.876 g (2.49 mmol) of meloxicam and 0.350 g
(2.53 mmol) of salicylic acid were slurried at ca. 250 rpm in 2 mL of
Synthesis of Meloxicam Salicylic Acid (1:1) Cocrystal Form I (7).
(a) Solvent-drop grinding: 0.177 g (0.504 mmol) of meloxicam was
3
ball-milled with 0.0675 g (0.489 mmol) of salicylic acid and 40 μL of

4404 Crystal Growth & Design, Vol. 10, No. 10, 2010
Table 2. Molecular Structures, pKa,49,51 ΔpKa, and Melting Point Values for Coformers That Form Cocrystals 7-19
Cheney et al.
THF for 30 min. Cocrystal 7 was isolated in ca. 100% yield. (b) Slurry:
0.871 g (2.47 mmol) of meloxicam and 0.352 g (2.55 mmol) of salicylic
acid were slurried at ca. 250 rpm in 2 mL of methanol overnight sealed
under ambient conditions. The resulting solid was filtered and washed
with methanol. Cocrystal 7 was obtained in ca. 91% yield.
Synthesis of Meloxicam Maleic Acid (1:1) Cocrystal (9). (a)
3
Solvent-drop grinding: 0.175 g (0.498 mmol) meloxicam was ball-
milled with 0.058 g (0.498 mmol) of maleic acid and 40 μL of THF
for 30 min, generating 9 in ca. 100% yield. (b) Slurry: 0.750 g (2.13
mmol) of meloxicam and 0.248 g (2.13 mmol) of maleic acid were
slurried at ca. 250 rpm in 2 mL of THF overnight sealed under
ambient conditions. The resulting solid was filtered and washed
with the same solvent employed for the slurry. Cocrystal 9 was
isolated in ca. 92% yield. Cocrystal 9 can also be synthesized via
ethyl acetate slurry.
Synthesis of Meloxicam Salicylic Acid (1:1) Cocrystal Form II
3
(8). (a) Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam
was ball-milled with 0.0705 g (0.510 mmol) of salicylic acid and
40 μL of chloroform for 30 min, generating 8 in ca. 100% yield. (b)
Slurry: 0.869 g (2.47 mmol) of meloxicam and 0.356 g (2.58 mmol) of
salicylic acid were slurried at ca. 250 rpm in 2 mL of chloroform
overnight sealed under ambient conditions. The resulting solid was
filtered and washed with chloroform. Cocrystal 8 was isolated in
ca. 89% yield.
Synthesis of Meloxicam Malonic Acid (1:1) Cocrystal (10). (a)
3
Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam was ball-
milled with 0.052 g (0.498 mmol) of malonic acid and 40 μL of THF
for 30 min, generating 10 in ca. 100% yield. (b) Slurry: 0.900 g

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4405
Table 3. Crystal Structure Parameters for Cocrystals 1-6
1
2
3
4
5
6
chemical formula
C14H13N3O4S2
C
C14H13N3O4S2
C5H8O4
C14H14N3O4S2
C14H13N3O4-
S2 C4H5O5
836.88
triclinic
P1
7.278(2)
8.550(2)
15.150(4)
84.115(3)
81.617(4)
70.489(3)
877.6(4)
1.583
C14H13N3O4S2
C7H6O3
C14H13N3O4S2
C14H13N3O4-
S2 C4H4O4
818.86
triclinic
P1
7.145(5)
8.475(5)
15.088(9)
82.250(9)
81.368(9)
70.519(9)
848.1(7)
1.603
(C14H13N3O4S2)2
C4H6O4
3
3
3
3
3
3
11H8O3
3
3
formula weight
cryst syst
space group
539.57
triclinic
P1
483.51
triclinic
P1
489.51
monoclinic
P21/c
11.0844(5)
11.8433(5)
16.4532(7)
90
820.88
triclinic
P1
˚
a (A)
6.938(3)
11.997(5)
15.107(6)
112.838(5)
92.889(6)
95.816(6)
1147.4(8)
1.562
7.181(2)
8.439(3)
18.364(6)
80.737(4)
85.464(4)
70.865(4)
1037.2(6)
1.548
7.2315(4)
8.4994(5)
14.9383(8)
82.741(4)
80.061(3)
70.313(4)
849.21(8)
1.605
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
97.386(3)
90
3
˚
vol (A )
2141.99(16)
1.518
Dcalcd (g cm-3
)
Z
2
5872
4079
3249
100
2
5390
3688
3298
100
1
4334
3037
1620
293
4
11932
3532
2659
293
1
4399
3031
2298
100
1
5052
2719
1871
100
reflns collected
independent reflns
obsd reflns
T (K)
R1
wR2
GOF
0.0511
0.1200
1.058
0.0338
0.0894
1.049
0.0663
0.1138
0.986
0.0433
0.1103
1.010
0.0451
0.1067
0.995
0.0463
0.0845
0.995
(2.56 mmol) of meloxicam and 0.266 g (2.56 mmol) of malonic acid
were slurried at ca. 250 rpm in 2 mL of THF overnight sealed under
ambient conditions. The resulting solid was filtered and washed
with THF. Cocrystal 10 was isolated in ca. 88% yield.
of meloxicam and 0.061 g (0.500 mmol) of benzoic acid were slurried at
ca. 250 rpm in 3 mL of ethyl acetate overnight sealed under ambient
conditions. The resulting solid was filtered and washed with ethyl
acetate. Cocrystal 17 was isolated in ca. 81% yield.
Synthesis of Meloxicam Gentisic Acid (1:1) Cocrystal (11). (a)
Synthesis of Meloxicam
3
3 DL-Malic Acid (2:1) Cocrystal (18). (a)
Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam was ball-
milled with 0.077 g (0.498 mmol) of gentisic acid and 40 μL of
chloroform or THF for 30 min, generating 11 in ca. 100% yield. (b)
Slurry: 0.850 g (2.41 mmol) of meloxicam and 0.373 g (2.41 mmol) of
gentisic acid were slurried at ca. 250 rpm in 2 mL of chloroform
overnight sealed under ambient conditions. The resulting solid was
filtered and washed with the same solvent employed for the slurry.
Cocrystal 11 was isolated in ca. 85% yield. Cocrystal 11 can also be
synthesized via slurry in ethyl acetate.
Solvent-drop grinding: 0.088 g (0.250 mmol) of meloxicam was ball-
milled with 0.017 g (0.127 mmol) of DL-malic acid and 50 μL of THF
for 30 min, generating 18 in ca. 100% yield. (b) Slurry: 0.880 g (2.50
mmol) of meloxicam and 0.170 g (1.27 mmol) of DL-malic acid were
slurried at ca. 250 rpm in 3 mL of THF overnight sealed under
ambient conditions. The resulting solid was filtered and washed
with THF. Cocrystal 18 was isolated in ca. 79% yield.
Synthesis of Meloxicam Hydrocinnamic Acid (1:1) Cocrystal
3
(19). (a) Solvent-drop grinding: 0.176 g (0.501 mmol) of meloxicam
was ball-milled with 0.075 g (0.499 mmol) of hydrocinnamic acid
and 50 μL of THF for 30 min, generating 19 in ca. 100% yield. (b)
Slurry: 0.702 g (2.00 mmol) of meloxicam and 0.300 g (1.99 mmol) of
hydrocinnamic acid were slurried at ca. 250 rpm in 3 mL of ethyl
acetate overnight sealed under ambient conditions. The resulting
solid was filtered and washed with ethyl acetate. Cocrystal 19 was
isolated in ca. 78% yield.
2.3. Crystal Form Characterization. Single-Crystal X-ray Dif-
fraction. Quality single crystals for X-ray diffraction were obtained
for six compounds. Attempts to crystallize 7-19 did not afford
crystals suitable for single-crystal X-ray crystallographic analysis.
Single crystal analysis for 1-3 was performed on a Bruker-AXS
SMART APEX CCD diffractometer with monochromatized Mo
Synthesis of Meloxicam 4-Hydroxybenzoic Acid (1:1) Cocrystal
3
(12). Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam was
ball-milled with 0.069 g (0.498 mmol) of 4-hydroxybenzoic acid and
40 μL of THF for 30 min, generating 12 in ca. 100% yield.
Synthesis of Meloxicam Adipic Acid (2:1) Cocrystal (13). (a)
3
Solvent-drop grinding: 0.088 g (0.250 mmol) of meloxicam was
ball-milled with 0.018 g (0.123 mmol) of adipic acid and 50 μL of
THF for 30 min, generating 13 in ca. 100% yield. (b) Slurry: 0.880 g
(2.50 mmol) of meloxicam and 0.180 g (1.23 mmol) of adipic acid
were slurried at ca. 250 rpm in 3 mL of THF overnight sealed under
ambient conditions. The resulting solid was filtered and washed
with THF. Cocrystal 13 was isolated in ca. 80% yield.
Synthesis of Meloxicam (þ)-Camphoric Acid (1:1) Cocrystal
3
˚
KR radiation (λ = 0.710 73 A) connected to a KRYO-FLEX low-
(14). Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam
was ball-milled with 0.100 g (0.498 mmol) of (þ)-camphoric acid
and 40 μL of THF for 30 min, generating 14 in ca. 100% yield.
temperature device, while data for 4-6 were collected using Cu KR
˚
radiation (λ = 1.54178 A). Data for 1, 2, 5, and 6 were collected at
100 K. Data for 3 and 4 were collected at 293 K (Table 3). Lattice
parameters were determined from least-squares analysis, and reflec-
tion data were integrated using SAINT.38 Structures were solved by
direct methods and refined by full matrix least-squares based on F2
using the SHELXTL package.39 All non-hydrogen atoms were
refined with anisotropic displacement parameters. All hydrogen
atoms bonded to carbon, nitrogen, and oxygen atoms were placed
geometrically and refined with an isotropic displacement parameter
fixed at 1.2Uq of the atoms to which they were attached. Hydrogen
atoms bonded to methyl groups were placed geometrically and
refined with an isotropic displacement parameter fixed at 1.5Uq of
the carbon atoms.
Synthesis of Meloxicam (þ)-Camphoric Acid (3:2) Cocrystal
3
(15). Slurry: 0.700 g (1.99 mmol) of meloxicam and 0.266 g (1.33
mmol) of (þ)-camphoric acid were slurried at ca. 250 rpm in 4 mL of
chloroform overnight sealed under ambient conditions. The result-
ing solid was filtered and washed with chloroform. Cocrystal 15 was
isolated in ca. 91% yield.
Synthesis of Meloxicam Glycolic Acid (1:1) Cocrystal (16). (a)
3
Solvent-drop grinding: 0.175 g (0.498 mmol) of meloxicam was ball-
milled with 0.038 g (0.498 mmol) of glycolic acid and 40 μL of
chloroform for 30 min, generating 16 in ca. 100% yield. (b) Slurry:
0.950 g (2.70 mmol) of meloxicam and 0.206 g (2.70 mmol) of
glycolic acid were slurried at ca. 250 rpm in 2 mL of ethyl acetate
overnight sealed under ambient conditions. The resulting solid was
filtered and washed with ethyl acetate. Cocrystal 16 was isolated in
ca. 92% yield.
Powder X-ray Diffraction (PXRD). Cocrystals 1-19 were char-
acterized using a D-8 Bruker X-ray powder diffractometer using Cu
˚
KR radiation (λ = 1.54178 A), 40 kV, 40 mA. Data were collected
Synthesis of Meloxicam Benzoic Acid (1:1) Cocrystal (17). (a)
over an angular range of 3°-40° 2θ value in continuous scan mode
using a step size of 0.05° 2θ value and a scan rate of 5°/min.
Calculated PXRD. Calculated PXRD diffractograms were gen-
erated from the single-crystal structures of 1-6 using Mercury
3
Solvent-drop grinding: 0.176 g (0.501 mmol) of meloxicam was ball-
milled with 0.061 g (0.500 mmol) of benzoic acid and 50 μL of THF for
30 min, generating 17 in ca. 100% yield. (b) Slurry: 0.176 g (0.501 mmol)

4406 Crystal Growth & Design, Vol. 10, No. 10, 2010
Cheney et al.
Figure 1. Meloxicam supramolecular chains sustained by sulfonyl-amide dimers and sulfathiazole-alcohol supramolecular synthons.
2.2 (Cambridge Crystallographic Data Centre, UK) for the follow-
ing complexes and compared with the pattern obtained for the
corresponding bulk sample.
Differential Scanning Calorimetry (DSC). Differential scanning
calorimetry was performed on a Perkin-Elmer Diamond DSC with
a typical scan range of 25-280 °C, scan rate of 10 °C/min, and
nitrogen purge of ca. 30 psi.
Fourier Transform Infrared Spectroscopy (FT-IR). FT-IR analy-
sis was performed on a Perkin-Elmer Spectrum 100 FT-IR spectro-
meter equipped with a solid-state ATR accessory.
between an azole and a carboxylic acid, primary amide,
and alcohol were examined using the CSD. Due to the
inability of the azole to form a supramolecular synthon with
itself, only the homosynthon formation of the carboxylic
acid, primary amide, and alcohol moieties in the presence of
an azole was examined. Meanwhile, it was noted that when
the pKa difference (ΔpKa, ΔpKa = pKa(base) - pKa(acid))
between meloxicam and a carboxylic acid coformer is sig-
nificant, the coformer could protonate meloxicam and form
a carboxylate salt. With this in mind, carboxylate-azole
interactions were also searched in the CSD. The analysis
showed that the azole-carboxylic acid and azole-alcohol
supramolecular heterosynthons are persistent (see Supporting
Information).
3.1.2. Selection of Meloxicam Coformers. Based on the
statistical success rate of supramolecular heterosynthon for-
mation between azoles and carboxylic acids, a meloxicam
cocrystal screen with acidic coformers that are pharmaceu-
tically acceptable or toxicologically qualified was conducted.
The focus of the study was rather narrow in scope and did not
include coformers that only possessed alcohol moieties
despite the potential for interaction based upon the CSD
statistics. Meloxicam was thereby reacted with 1-hydroxy-2-
naphthoic acid, glutaric acid, L-malic acid, salicylic acid,
fumaric acid, succinic acid, maleic acid, malonic acid, gen-
tisic acid, 4-hydroxybenzoic acid, adipic acid, (þ)-camphoric
acid, glycolic acid, benzoic acid, DL-malic acid, hydrocin-
namic acid, ascorbic acid, and L-tartaric acid. All coformers
except ascorbic acid and L-tartaric acid produced at least one
cocrystal. The cocrystallization attempts resulted in 19 crystal
3. Results and Discussion
3.1. The Supramolecular Synthon Approach. As previously
mentioned, a key step in generating pharmaceutical cocrys-
tals is to analyze the target API from a crystal engineering
perspective, to evaluate how the target molecule would form
supramolecular synthons. This methodology partitions the
target molecule by its functional groups and statistically40
examines the percentage of occurrence of supramolecular
homo- and heterosynthons for these functional groups. The
targeted supramolecular synthons are typically sustained via
hydrogen bonds because they are strong and directional in
nature. This method is particularly beneficial because most
APIs tend to be rich in functional groups that are capable of
forming strong hydrogen bonds.
Polymorphic form I41 of meloxicam (refcode SEDZOQ)
indicates that meloxicam molecules form supramolecular
chains that are sustained by sulfonyl-amide and sulfathiazole-
alcohol supramolecular heterosynthons, as shown in Figure 1.
The chains are held together by various weak interactions,
stacking along the a-axis in a slipped fashion. Thus for melox-
icam cocrystallization, one or all of these supramolecular syn-
thon motifs must be interrupted.37
forms; namely, meloxicam 1-hydroxy-2-naphthoic acid cocrys-
3
tal, meloxicam glutaric acid cocrystal, meloxicam L-malic acid
3
3
3.1.1. CSD Analysis. A CSD analysis42 was conducted to
examine the occurrence of supramolecular synthon forma-
tion of the amino-azole functionality (five-membered ring
containing a nitrogen and primary amine) with carboxylic
acid, primary amide, and alcohol moieties. It was noted that
many of the searches for the amino-azole moiety coupled
with an additional functional group consisted of a relatively
small number of entries. Any conclusions from the data
might therefore be statistically insignificant. Therefore
further CSD analysis was conducted employing a simple
azole (five-membered ring containing one nitrogen atom and
at least one additional heteroatom). The reliability of supra-
molecular heterosynthon versus homosynthon formation
cocrystal of a salt, meloxicam salicylic acid cocrystal forms I, II,
3
and III, meloxicam fumaric acid cocrystal, meloxicam succinic
3
3
acid cocrystal, meloxicam maleic acid cocrystal, meloxicam
3
3
malonic acid cocrystal, meloxicam gentisic acid cocrystal,
3
meloxicam 4-hydroxybenzoic acid cocrystal, meloxicam adipic
3
3
acid cocrystal, meloxicam (þ)-camphoric acid cocrystal (1:1),
3
meloxicam (þ)-camphoric acid cocrystal (3:2), meloxicam
3
3
glycolic acid cocrystal, meloxicam benzoic acid cocrystal,
meloxicam DL-malic acid cocrystal, and meloxicam hydrocin-
3
3
3
namic acid cocrystal.
3.2. Crystal Structure Descriptions. 3.2.1. Meloxicam
1-Hydroxy-2-naphthoic Acid (1:1) Cocrystal, 1. Meloxicam
3
3
1-hydroxy-2-naphthoic acid cocrystal (1) crystallizes in the

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4407
Figure 2. A supramolecular unit formed by meloxicam and 1-hydroxy-2-naphthoic acid in 1.
Figure 3. Packing of four supramolecular layers in 1.
space group P1. The asymmetric unit contains one meloxicam
and one 1-hydroxy-2-naphthoic acid molecule. The primary
supramolecular unit comprises two meloxicam molecules and
two 1-hydroxy-2-naphthoic acid molecules with the inversion
center located in the center of the meloxicam dimer (Figure 2).
The meloxicam dimer is a common motif found in the structure
of various meloxicam cocrystals and is sustained by two planar
meloxicam molecules interacting via electrostatic interactions
between the alcohol, ketone, and sulfathiazole moieties. It is
observed that meloxicam sustains intramolecular alcohol-
ketone interactions and binds to the neighboring meloxicam
2, is sustained by an S OH interaction between two planar
3 3 3
˚
meloxicam molecules with a distance of 3.213 A. In addition,
an intramolecular OH O hydrogen bond is also observed
3 3 3
on meloxicam. Interestingly, glutaric acid utilizes one of its
carboxylic acid moieties to form a centrosymmetric carboxylic
˚
˚
acid dimer [O4-H4 O3: O O 2.641(2) A, H O 1.801 A,
3 3 3
3 3 3
3 3 3
O-H O 177.76°] with a neighboring glutaric acid. The other
3 3 3
carboxylic acid moiety of glutaric acid hydrogen bonds to the
adjacent meloxicam molecule, resulting in the unexpected 1:1
stoichiometry. The carboxylic acid-azole supramolecular het-
erosynthon dimer is sustained by hydrogen bond interactions
that afford the common two-point recognition motif. The
OH N and NH O interactions are shown in Figure 4 [for
˚
via an S OH interaction at a distance of 3.111 A to form
3 3 3
the meloxicam dimer. The exterior of the meloxicam dimer
hydrogen bonds to two 1-hydroxy-2-naphthoic acid molecules
via two-point recognition carboxylic acid-thiazole/NH inter-
3 3 3
3 3 3
˚
˚
˚
O2-H2O N1, O N 2.6675(19) A, H N 1.831 A,
3 3 3 3 3 3 3 3 3
O-H N 173.96°; for N2-H2N O1, N O 2.838(2) A,
3 3 3 3 3 3 3 3 3
˚
˚
˚
˚
actions[forO2-H2 N1, O N 2.563(3) A,H N1.728A,
3 3 3 3 3 3 3 3 3
H
O 1.979 A, N-H O 164.95°]. The supramolecular
3 3 3 3 3 3
O-H N 172.19°; for N2-H2N O1, N O 2.902(3) A,
homosynthon and heterosynthon dimers ultimately result in
the formation of a zigzag supramolecular chain that cuts through
the ac-plane. The chains, which stack along the a-axis, are
3 3 3 3 3 3 3 3 3
˚
H
O 2.026 A, N-H O 166.48°]. This two-point supra-
3 3 3
3 3 3
molecular heterosynthon motif is quite robust since it is found
in all crystallographically characterized cocrystal forms re-
ported herein. The hydroxyl group of 1-hydroxy-2-naphthoic
acid adjacent to the carboxylic acid moiety is involved in
intramolecular hydrogen bonding, which is expected based
upon Etter’s rules.43 The overall packing, illustrated in Figure 3,
is sustained by the primary supramolecular units of melox-
icam and 1-hydroxy-2-naphthoic acid that stack upon each
other along the a-axis in a slipped fashion with an inter-
˚
separated by a distance of 3.465 A. Figure 5 highlights the close
packing of multiple chains with meloxicam dimers and glutaric
acid dimers disposed in a columnar arrangement. The
C1-C2-C3-C4 torsion angle of glutaric acid in 2 is atypical.
It is noted that the C1-C2-C3-C4 torsion angle in both
polymorphs of pure glutaric acid is ca. 170°, whereas in 2 this
torsion angle is 52.97°. A CSD search for additional glutaric acid
cocrystals afforded multiple hits. Among those, glutaric acid
with smaller torsion angles occurs in only two other cocrystals,
˚
planar spacing of 3.630 A.
3.2.2. Meloxicam Glutaric Acid (1:1) Cocrystal, 2. The
asymmetric unit of the meloxicam glutaric acid cocrystal (2)
that is, caffeine glutaric acid cocrystal (refcode EXUQUJ) and
3
3
theophylline glutaric acid cocrystal (refcode XEJXIU), which
3
3
comprises one meloxicam and one glutaric acid molecule
(Figure 4). Cocrystal 2 crystallizes in the space group P1. The
meloxicam centrosymmetric dimer, which is again present in
exhibit torsion angles of 79.34° and 65.18°, respectively.
3 L-Malic Acid (1:1:1) Cocrystal of a Salt,
3.2.3. Meloxicam
3. The asymmetric unit of the meloxicam L-malic acid
3

4408 Crystal Growth & Design, Vol. 10, No. 10, 2010
Cheney et al.
Figure 4. Supramolecular arrangements of meloxicam and glutaric acid dimers in 2.
Figure 5. Packing of supramolecular chains in 2.
Figure 6. Supramolecular synthons in 3 showing a neutral and cationic meloxicam and an anionic L-malate.
cocrystal of a salt (3) contains one meloxicam cation, one
neutral meloxicam, and one L-malate anion. Cocrystal 3
could therefore be considered a cocrystal of a salt and
crystallizes in the space group P1. Based upon the value of
ΔpKa (Table 1), it was unexpected that L-malic acid
could protonate meloxicam. In 3, the meloxicam dimer is

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4409
Figure 7. Stacked layers of 3. L-Malate anions are highlighted in green.
comprised of one cation and one neutral meloxicam and is
sustained by two S OH interactions with S O distances
3 3 3
3 3 3
˚
of 3.231 and 3.294 A (Figure 6). Cocrystal 3 is also sustained
by the L-malate anion acting as a bridge to connect the melox-
icam dimers, ultimately forming a supramolecular chain. The
L-malate bridge protonates the sulfathiazole ring via O- HNþ
3 3 3
interactions [for O1 H1N-N1, O- N 2.686(17) A,
˚
3 3 3
3 3 3
˚
H
O
O 1.831 A, O H-N 173.33°; for O2 H22N-N2,
3 3 3
3 3 3
3 3 3 3 3 3
˚
˚
N 2.948(16) A, H O 2.111 A, O H-N 164.78°] and
3 3 3 3 3 3
hydrogen bonds to the sulfathiazole of the neutral meloxicam via
OH N and NH O interactions [for O4-H4O N4,
3 3 3 3 3 3 3 3 3
˚
˚
O
N 2.732(17) A, H N 1.915 A, O-H N 173.45°; for
3 3 3
3 3 3
3 3 3 3 3 3
˚
˚
O5 H5A-N5, O N 2.831(17) A, H O 1.989 A,
3 3 3
3 3 3
O-H N 166.48°]. The supramolecular chains exhibited by 3
3 3 3
are reminiscent of those seen in 2.L-Malate and meloxicam adopt
a three-dimensional (3D) arrangement that comprises chains
stacking parallel along the a-axis with a separation distance of
˚
Figure 8. Two-point recognition supramolecular synthon in 4.
3.562 A (centroid to plane), as shown in Figure 7.
3.2.4. Meloxicam Salicylic Acid (1:1) Cocrystal Form III, 4.
The supramolecular unit of meloxicam salicylic acid cocrystal
3
meloxicam molecules and one fumaric acid molecule. In 5,
the meloxicam dimer persists and links to adjacent dimers via
fumaric acid molecules, creating an infinite supramolecular
chain (Figure 10). The supramolecular heterosynthon compris-
ing carboxylic acid and sulfathiazole/NH moieties is observed
3
form III (4) comprises one molecular meloxicam and one
molecular salicylic acid, shown in Figure 8. Cocrystal 4 crystal-
lizes in the space group P21/c. Salicylic acid and meloxicam are
associated by a supramolecular heterosynthon between the
carboxylic acid and the sulfathiazole/NH moieties. The inter-
actions of carboxylic acid with sulfathiazole/NH moiety in-
between fumaric acid and meloxicam. Both O-H N and
3 3 3
NH O hydrogen bonds are observed [O22-H22O N3,
3 3 3 3 3 3
˚
˚
O
N 2.68(3) A, H N 1.821 A, O-H N 174.39°;
3 3 3
3 3 3 3 3 3
clude the OH N hydrogen bond [O2-H2O N1, O
3 3 3 3 3 3
N
˚
˚
3 3 3
N2-H2N O21, N O 2.857(4) A, H O 1.976 A,
3 3 3
3 3 3
3 3 3
˚
˚
2.619(3) A, H N 1.807 A, O-H N 170.36°] and the
3 3 3 3 3 3
N-H O 160.49°]. The supramolecular chain of meloxicam
3 3 3
NH O hydrogen bond [N2-H2N O1, N O 2.968(3)
3 3 3 3 3 3 3 3 3
and fumaric acid exhibited in 5 is again similar to those
observed in 2 and 3. Clearly, the use of dicarboxylic acids with
similar molecular structure orients meloxicam and coformers
similarly in the crystal lattice (Figure 11).
˚
˚
A, H O 2.133 A, N-H O 163.78°]. This hydrogen bond-
3 3 3 3 3 3
ing motif generates a discrete unit containing one meloxicam
and one salicylic acid molecule. In addition, intramolecular
hydrogen bonding is observed on salicylic acid between the OH
moiety and the carboxylic acid. Supramolecular units compris-
ing meloxicam and salicylic acid in 4 translate along the 21
screw axis at a dihedral angle of 88.27° (Figure 9). Interestingly,
the meloxicam dimer does not exist in 4 because the meloxicam
molecules are perpendicular to adjacent meloxicam molecules.
Cocrystal 4 is the only cocrystal reported herein that does not
3.2.6. Meloxicam Succinic Acid (2:1) Cocrystal, 6. Pre-
3
paration of the meloxicam succinic acid cocrystal (6) by
3
solvent-drop grinding has been recently reported but without
determination of its crystal structure.36 The calculated
PXRD of 6 based upon our single-crystal structure data
matches that of the previously reported PXRD. Cocrystal 6
crystallizes in the space group P1 with the asymmetric unit
containing one meloxicam molecule and half a succinic acid
molecule. The crystal structure of 6 reveals that the melox-
icam dimers are associated with adjacent dimers by succinic
acid molecules forming infinite supramolecular chains
(Figure 12). Similar to previous meloxicam cocrystal struc-
tures, the primary intermolecular interactions of 6 are hydro-
gen bonds between meloxicam and succinic acid via the
exhibit the meloxicam dimer. Two other meloxicam salicylic
3
acid cocrystal polymorphs (form I and form II) were also
isolated, but single crystals suitable for single-crystal XRD
analysis were not obtained.
3.2.5. Meloxicam Fumaric Acid (2:1) Cocrystal, 5. The
3
asymmetric unit of the meloxicam fumaric acid cocrystal
3
(5), which crystallizes in the space group P1, contains two

4410 Crystal Growth & Design, Vol. 10, No. 10, 2010
Cheney et al.
Figure 9. Supramolecular units comprising meloxicam and salicylic acid translating along the 21 screw axis.
Figure 10. Supramolecular synthons observed in 5.
carboxylic acid to sulfathiazole/NH supramolecular hetero-
synthon. The OH N and NH OdC hydrogen bonds
characterized by PXRD, FT-IR, and thermal techniques.
Even without the single-crystal XRD data, the stoichiome-
tries of 7-19 were determined as exemplified by 9.
3 3 3
3 3 3
˚
are involved [O7-H7 N3, O N 2.863(4) A, H
N
O
3 3 3
3 3 3
3 3 3
3 3 3
˚
The PXRD and FT-IR profiles of 9 were provided in the
literature, but the stoichiometry of meloxicam and maleic
acid in 9 was not disclosed. Based on the potential supra-
molecular interactions of meloxicam, the most likely stoi-
chiometry of meloxicam and coformer in 9 is either 1:1 or 1:2.
In order to determine the stoichiometry of 9, THF slurries
of physical mixtures of meloxicam and maleic acid in molar
ratios of 2:1, 1:1, and 1:2 were performed at room tempera-
ture overnight. From each slurry experiment, the solid
crystalline powder was separated, washed with THF and
dried for characterization. The absence of pure maleic acid,
water, or THF was confirmed based on the DSC analysis
upon all three solid powders from slurries. PXRD characteri-
zation indicated that the solids generated from the 1:1 and
1.847 A, O-H
N
173.63°; N2-H2 O6,
N
3 3 3
3 3 3
˚
˚
2.849(4) A, H O 1.993 A, N-H O 164.32°]. The supra-
3 3 3
3 3 3
molecular chains of meloxicam and succinic acid in 6 are
reminiscent of the supramolecular chains found in 2, 3, and 5
(Figure 12). As shown in Figure 13, supramolecular chains of
succinic acid and meloxicam stack with an interplanar spa-
˚
cing of 3.386 A. As a result of an overall packing comparison
between meloxicam cocrystal structures, it is observed that 3,
5, and 6 are isostructural.
3.2.7. Meloxicam Cocrystals 7-19. Cocrystals 7-19 were
prepared from solution or solid-state grinding methods.
However, the absence of structural data for 7-19 is the
result of unsuccessful solution-based growth of single crys-
tals. Nevertheless, polycrystalline powders of 7-19 were

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4411
Figure 11. Supramolecular layers stacking in 5.
Figure 12. Supramolecular synthons observed in 6.
Figure 13. Supramolecular layers stacking in 6.
1:2 slurries were identical to the cocrystal form reported by
Myz et al.36 In contrast, the 2:1 slurry produced a physical
mixture of meloxicam form I and the cocrystal. Further 1H
NMR analysis (400 MHz, d6-DMSO) on solids from 1:1 and
1:2 slurries confirmed that the stoichiometry of meloxicam
and maleic acid in 9 is 1:1 (see Supporting Information).
3.2.8. Meloxicam Crystal Forms: Cocrystals or Salts? Phar-
maceutical cocrystals and salts are well-defined and are typi-
cally classified as distinct subsets of crystal forms. Whether
novel meloxicam crystal forms generated by the 17 coformers in
this study are cocrystals or salts has also been investigated. For
crystal forms with single-crystal XRD data (i.e., 1-6), the
conclusion that 1-6 are cocrystals was drawn in a relatively
simple and reliable manner. However, it was less straightfor-
ward to identify whether 7-19 are cocrystals or salts.
The pKa values for meloxicam are 1.09 and 4.18.44 The
value of 1.09 is associated with the enolic OH group, while
the value of 4.18 is linked to the nitrogen atom on the
sulfathiazole ring. The enolic OH is much less accessible
from a crystal engineering perspective because it is involved
in intramolecular hydrogen bonding to the neighboring
ketone or NH moieties. In contrast, the nitrogen atom on
the sulfathiazole ring is the primary target for cocrystal or
salt formation, because it could potentially sustain a supra-
molecular synthon with various hydrogen bond donors.
ΔpKa is widely accepted as the key to predicting whether a
salt or cocrystal form will form.45,46 It is generally considered
that, if ΔpKa < 0, the resulting compound will be a cocrystal,
whereas the result is typically a salt if ΔpKa >3. For
the region of ΔpKa between 0 < ΔpKa < 3, our ability to

4412 Crystal Growth & Design, Vol. 10, No. 10, 2010
Cheney et al.
predict whether the resulting complex will be neutral or
charged is limited.46 Indeed, half of the ΔpKa values re-
ported herein fall into the range of 0 < ΔpKa < 3. How
can one identify whether 7-19 are cocrystals or salts, especially
where the FT-IR spectra may not provide adequate
information?
References
(1) Allen, L. V.; Popovich, N. G.; Ansel, H. C., Ansel’s pharmaceutical
dosage forms and drug delivery systems; Lippincott Williams and
Wilkins: Baltimore, MD, 2005.
(2) Byrn, S. R.; Pfeiffer, R. R.; Stephenson, G.; Grant, D. J. W.;
Gleason, W. B. Chem. Mater. 1994, 6, 1148. Haleblian, J. K.
J. Pharm. Sci. 1975, 64, 1269.
The ΔpKa values of cocrystals 1-6 were used as a reference
to determine whether 7-19 are cocrystals or salts. As shown
in Table 1, 4 possesses the largest value of ΔpKa at 1.21, and
proton transfer was not observed between meloxicam and
salicylic acid. Although the molecular arrangement of various
coformers may have an influence to the electron distribution
and protonation of meloxicam, with the single-crystal XRD
data of 1-6, it is reasonable to assert that all coformers
involved in this study with ΔpKa close to or less than 1.21
would potentially produce a cocrystal rather than a salt with
meloxicam.47 Based on this, with the exception of 9, all crystal
forms prepared in this study can be identified as meloxicam
cocrystals (Table 2). However, 9 remained questionable, be-
cause maleic acid exhibits a ΔpKa of 2.25. Further investigation
into the FT-IR spectrum of 9 was performed to identify the
position of the proton. The carbonyl group of maleic acid in 9
exhibits a distinct peak at 1716 cm-1, indicating that the
carboxylic acid group is neutral rather than negatively
charged.48 Therefore, proton transfer does not occur between
meloxicam and maleic acid, and 9 is a cocrystal of meloxicam
and maleic acid.
(3) Berge, S. M.; Bighley, L. D.; Monkhouse, D. C. J. Pharm. Sci.
1977, 66, 1.
(4) Dublin, C. H. Drug Delivery Technol. 2006, 6, 34.
(5) Khankari, R. K.; Grant, D. J. W. Thermochim. Acta 1995, 248, 61.
(6) Vippagunta, S. R.; Brittain, H. G.; Grant, D. J. W. Adv. Drug
Delivery Rev. 2001, 48, 3.
(7) Zaworotko, M.; Arora, K. Pharmaceutical co-crystals: A new
opportunity in pharmaceutical science for a long-known but little
studied class of compounds. In Polymorphism in Pharmaceutical
Solids, 2nd ed.; Brittain, H. G., Ed.; Informa Healthcare: London, 2009.
(8) (a) Aaltonen, J.; Alleso, M.; Mirza, S.; Koradia, V.; Gordon, K. C.;
Rantanen, J. Eur. J. Pharm. Biopharm. 2009, 71, 23. (b) Yin, S. X.;
Grosso, J. A. Curr. Opin. Drug Discovery Dev. 2008, 11, 771.
(9) Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.;
Zaworotko, M. J.; Sava, V.; Song, S.; Sanchez-Ramos, J. R. Cryst.
Growth Des. 2010, 10, 394.
(10) Desiraju, G. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2311.
(11) (a) Pepinsky, R. Phys. Rev. 1955, 100, 971. (b) Desiraju, G. R.,
Crystal Engineering: The Design of Organic Solids; Elsevier:
Amsterdam, 1989. (i) Schmidt, G. M. J. Pure Appl. Chem. 1971, 27,
647. (c) Shan, N.; Batchelor, E.; Jones, W. Tetrahedron Lett. 2002, 43,
8721. (d) Shan, N.; Bond, A. D.; Jones, W. Cryst. Eng. 2002, 5, 9.
(e) Shan, N.; Bond, A. D.; Jones, W. Tetrahedron Lett. 2002, 43, 3101.
(f ) Shan, N.; Jones, W. Tetrahedron Lett. 2003, 44, 3687. (g) Bis, J. A.;
Vishweshwar, P.; Weyna, D.; Zaworotko, M. J. Mol. Pharmaceutics
2007, 4, 401. (h) Shattock, T. R.; Arora, K. K.; Vishweshwar, P.;
Zaworotko, M. J. Cryst. Growth Des. 2008, 8, 4533.
4. Conclusions
(12) Allen, F. H. Acta Crystallogr. 2002, B58, 380.
€
(13) Almarsson, O.; Zaworotko, M. J. Chem. Commun. 2004, 1889.
Considering the enormous impact of pharmaceutical co-
crystals upon crystal form diversity and the resulting oppor-
tunity to customize the physicochemical properties of APIs, it
is unlikely that the relevance of crystal forms to drug delivery,
intellectual property, and regulatory control will diminish in
the near future. The science of crystal form selection will
continue to evolve. In this context, meloxicam was selected as a
case study to demonstrate how coformers can generate phar-
maceutical cocrystals with the potential to improve the phar-
macokinetic properties and the onset-of-action of the API.
The supramolecular synthon approach afforded 19 cocrystals
with pharmaceutically acceptable or toxicologically qualified
carboxylic acids.
The use of multiple preparative techniques for almost all
meloxicam cocrystals in this study would indicate that large-
scale manufacturing of those cocrystals would be feasible
shouldone ofthese cocrystals exhibitdesiredpharmacological
properties. Among the 19 cocrystals reported herein, crystal
structures of six (1-6) were determined. A salient feature in
1-6 is the two-point recognition carboxylic acid-azole/NH
supramolecular heterosynthon. A prominent aspect of1-3, 5,
and 6 is the presence of the meloxicam dimer. The meloxicam
molecules in 4, however, are juxtaposed such that the melox-
icam dimer is absent. As part of the continuous crystal form
selection process, all meloxicam cocrystals obtained in this
study will be tested further via dissolution studies and in vivo
pharmacokinetic studies.
(14) Shan, N.; Zaworotko, M. J. Drug Discovery Today 2008, 13,
440.
(15) (a) Childs, S. L.; Chyall, L. J.; Dunlap, J. T.; Smolenskaya, V. N.;
Stahly, B. C.; Stahly, G. P. J. Am. Chem. Soc. 2004, 126, 13335.
(b) Banerjee, R.; Bhatt, P. M.; Ravindra, N. V.; Desiraju, G. R. Cryst.
Growth Des. 2005, 5, 2299.
(16) Good, D. J.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2009, 9,
2252.
(17) Karki, S.; Friscic, T.; Fabian, L.; Laity, P. R.; Day, G. M.; Jones,
W. Adv. Mater. 2009, 21, 3905.
(18) (a) Shan, N.; Toda, F.; Jones, W. Chem. Commun. 2002, 2372. (b)
Weyna, D. R.; Shattock, T.; Vishweshwar, P.; Zaworotko, M. J. Cryst.
Growth Des. 2009, 9, 1106.
ꢀꢀ ꢁ
(19) (a) Friscic, T.; Childs, S. L.; Rizvi, S. A. A.; Jones, W. CrystEng-
Comm 2009, 11, 418. (b) Lu, E.; Rodriguez-Hornedo, N.; Suryanarayanan,
R. CrystEngComm 2008, 10, 665.
(20) (a) Remenar, J. F.; Morissette, S. L.; Peterson, M. L.; Moulton, B.;
MacPhee, J. M.; Guzman, H. R.; Almarsson, O. J. Am. Chem. Soc.
2003, 125, 8456. (b) Basavoju, S.; Bostrom, D.; Velaga, S. P. Cryst.
Growth Des. 2006, 6, 2699. (c) Chen, A. M.; Ellison, M. E.; Peresypkin,
A.; Wenslow, R. M.; Variankaval, N.; Savarin, C. G.; Natishan, T. K.;
Mathre, D. J.; Dormer, P. G.; Euler, D. H.; Ball, R. G.; Ye, Z. X.; Wang,
Y. L.; Santos, I. Chem. Commun. 2007, 419. (d) Remenar, J. F.;
Peterson, M. L.; Stephens, P. W.; Zhang, Z.; Zimenkov, Y.; Hickey,
M. B. Mol. Pharmaceutics 2007, 4, 386. (e) Zegarac, M.; Mestrovic,
E.; Dumbovic, A.; Devcic, M.; Tudja, P. Pharmaceutically Acceptable
Co-Crystalline Forms of Sildenafil, WO 2007/080362 A1, 2007. (f)
Childs, S. L.; Rodríguez-Hornedo, N.; Reddy, L. S.; Jayasankar, A.;
Maheshwari, C.; McCausland, L.; Shipplett, R.; Stahly, B. C. Crys-
tEngComm 2008, 10, 856. (g) Shiraki, K.; Takata, N.; Takano, R.;
Hayashi, Y.; Terada, K. Pharm. Res. 2008, 25, 2581. (h) Stanton, M. K.;
Bak, A. Cryst. Growth Des. 2008, 8, 3856. (i) Stanton, M. K.; Tufekcic,
S.; Morgan, C.; Bak, A. Cryst. Growth Des. 2009, 9, 1344. (j) Trask,
A. V.; Motherwell, W. D. S.; Jones, W. Cryst. Growth Des. 2005,
5, 1013.
Acknowledgment. We thank Mr. Raymond K. Houck for
stimulating discussions and valuable input to the project.
(21) Ahmed, M.; Khanna, D.; Furst, D. E. Exp. Opin. 2005, 1, 739.
(22) (a) Thompson, J. P.; Sharpe, P.; Kiani, S.; Owen-Smith, O. Br. J.
Anaesth. 2000, 84, 151. (b) Roughan, J. V.; Flecknell, P. A. Eur. J. Pain
2003, 7, 397.
(23) Engelhardt, G.; Homma, D.; Schlegel, K.; Utzmann, R.; Schnitzler,
C. Inflammation Res. 1995, 44, 423.
Supporting Information Available: Additional characterization of
the meloxicam cocrystals reported herein and crystal structure data
in CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.

Article
Crystal Growth & Design, Vol. 10, No. 10, 2010 4413
(24) (a) Stei, P.; Kruss, B.; Wiegleb, J.; Trach, V. Br. J. Rheumatol. 1996,
35 (Suppl. 1), 44. (b) Lehmann, H. A.; Baumeister, M.; Luetzen, L.;
Wiegleb, J. Inflammopharmacology 1996, 4, 105.
(25) (a) Burdan, F. Toxicology 2005, 211, 12. (b) Harirforoosh, S.;
Aghazadeh-Habashi, A.; Jamali, F. Clin. Exp. Pharmacol. Physiol.
2006, 33, 917.
(26) Seedher, N.; Bhatia, S. AAPS PharmSciTech 2003, 4, No. 33.
(27) Takagi, T.; Ramachandran, C.; Bermejo, M.; Yamashita, S.; Yu,
L. X.; Amidon, G. L. Mol. Pharmaceutics 2006, 3, 631.
(28) Luger, P.; Daneck, K.; Engel, W.; Trummlitz, G.; Wagner, K. Eur.
J. Pharm. Sci. 1996, 4, 175.
(29) Coppi, L.; Sanmarti, M. B.; Clavo, M. C. Crystalline Forms of
Meloxicam and Processes for Thier Preparation and Interconver-
sion, US Patent 6,967,248 B2, 2005.
(30) Davies, N. M.; Skjodt, N. M. Clin. Pharmacokinet. 1999, 36, 115.
(31) Ghorab, M. M.; Abdel-Salam, H. M.; El-Sayad, M. A.; Mekhel,
M. M. AAPS PharmSciTech 2004, 5, No. 59.
(32) Han, H. K.; Choi, H. K. Eur. J. Pharm. Biopharm. 2007, 65, 99.
(33) Defazio, S.; Cini, R. J. Chem. Soc., Dalton Trans. 2002, 1888.
(34) (a) Bock, T.; Saegmueller, P.; Sieger, P.; Tuerck, D. Meloxicam for
Oral Administration. US Patent 6,869,948 B1, 2005. (b) Struengmann,
A.; Freudensprung, B.; Klokkers, K. New Pharmaceutical Composi-
tions of Meloxicam with Improved Solubility and Bioavailability,
International Patent WO 99/09988 A1, 1999.
approach has been successfully used in the various pharmaceutical
cocrystal studies, the CSD statistical result here could be biased.
(41) Fabiola, G. F.; Pattabhi, V.; Manjunatha, S. G.; Rao, G. V.;
Nagarajan, K. Acta Crystallogr. 1998, C54, 2001.
(42) CSD version 5.31, November 2009 release including May 2010
update. Search parameters: organics only, 3D coordinates deter-
mined, R < 7.5%.
(43) Etter, M. C. Acc. Chem. Res. 1990, 23, 120.
(44) Ki, H. M.; Choi, H. K. Arch. Pharm. Res. 2007, 30, 215.
(45) (a) Stahl, P. H.; Wermuth, C. G., Handbook of Pharmaceutical
Salts: Properties, Selection, and Use; WILEY-VCH: Zurich, 2002. (b)
Bhogala, B. R.; Basavoju, S.; Nangia, A. CrystEngComm 2005, 7, 551.
(c) Delori, A.; Suresh, E.; Pedireddi, V. R. Chem.—Eur. J. 2008, 14, 6967.
(46) Childs, S. L.; Stahly, G. P.; Park, A. Mol. Pharmaceutics 2007,
4, 323.
(47) Identification of cocrystals based on ΔpKa values is straightfor-
ward but also could be less rigorous. However, confirmation of
relevant proton locations in 8, 9, and 11-20 remains inconclusive
using available characterization techniques. As part of the contin-
uous study, further attempts of single-crystal XRD characteriza-
tion for 8, 9, and 11-20 are being undertaken.
(48) Silverstein, R. M.; Webster, F. X.; Kiemle, D. J. Spectroscopic
Identification of Organic Compounds; John Wiley & Sons: New York,
2005.
(35) Meloxicam for this study was purchased as form I, which was stable
at ambient conditions throughout the study.
(36) Myz, S. A.; Shakhtshneider, T. P.; Fucke, K.; Fedotov, A. P.;
Boldyreva, E. V.; Boldyrev, V. V.; Kuleshova, N. I. Mendeleev
Commun. 2009, 19, 272.
(37) Childs, S. L.; Hardcastle, K. I. Cryst. Growth Des. 2007, 7, 1291.
(38) Bruker SMART, SAINT-Plus, SADABS, XP and SHELXTL;
Bruker: Madison, WI, 1997.
(49) (a) pKa Data, http://research.chem.psu.edu/brpgroup/pKa_com-
pilation.pdf. (b) Minnick, L. J.; Kilpatrick, M. J. Phys. Chem. 1939,
43, 259.
(50) (a) Chemical Dictionary Online, 1-Hydroxy-2-naphthoic acid.
http://www.chemicaldictionary.org/dic/1/1-Hydroxy-2-naphthoic-
acid_919.html (August 2009); (b) Brittain, H. G.; Prankerd, R. J.
Profiles of drug substances, excipients, and related methodolgy;
Academic Press: New York, 2007; p 369; (c) Cabelli, D. E.; Bielski,
B. H. J. Z. Naturforsch., B 1985, 40, 1731.
€
€
(39) Sheldrick, G. M. SHELXTL, University of Gottingen: Gottingen,
Germany, 1997.
€
(51) (a) Erdemgil, F. Z.; Sanli, S.; Sanli, N.; Ozkan, G.; Barbosa, J.;
ꢁ
Guiteras, J.; Beltran, J. L. Talanta 2007, 72, 489. (b) Castro, J. L.;
(40) It must be noted that the CSD structures have not been system-
atically populated to provide a balanced representation of the
various functional groups. Although the supramolecular synthon
ꢁ
Ramírez, M. R. L.; Tocon, I. L.; Otero, J. C. J. Raman Spectrosc. 2002,
33, 455.