Examining the robustness of a theophylline cocrystal during grinding with additives

Article abstract of DOI:10.1039/c2ce25495f

The robustness of theophylline-p-hydroxybenzoic acid cocrystal (TP·pHBA) while grinding with additives in the solid state was evaluated through a series of solvent-drop grinding experiments with coformers containing various functional groups. Powder X-ray diffraction was used to qualitatively analyse the powders after grinding and identify the species present. The TP·pHBA cocrystal was found to be robust in the presence of benzoic acid (BA), p-nitrobenzoic acid (pNBA), p-(N,N-dimethylamino)benzoic acid (dMABA), m-hydroxybenzoic acid (mHBA), p-nitrophenol (pNP), hydroquinone (HDQ) and benzamide (BZA), but it disintegrates in the presence of salicylic acid (SA), 3,5-dinitrobenzoic acid (dNBA), acetamide (ACA) and melamine (MLM).

Full text of DOI:10.1039/c2ce25495f

View Online / Journal Homepage
Dynamic Article Links
CrystEngComm
Cite this: DOI: 10.1039/c2ce25495f
www.rsc.org/crystengcomm
PAPER
Examining the robustness of a theophylline cocrystal during grinding with
additives{{
Heba Abourahma,*a Jennifer M. Urban,a Nicole Morozowichb and Benny Chana
Received 4th April 2012, Accepted 23rd April 2012
DOI: 10.1039/c2ce25495f
The robustness of theophylline-p-hydroxybenzoic acid cocrystal (TP?pHBA) while grinding with
additives in the solid state was evaluated through a series of solvent-drop grinding experiments with
coformers containing various functional groups. Powder X-ray diffraction was used to qualitatively
analyse the powders after grinding and identify the species present. The TP?pHBA cocrystal was
found to be robust in the presence of benzoic acid (BA), p-nitrobenzoic acid (pNBA), p-
(N,N-dimethylamino)benzoic acid (dMABA), m-hydroxybenzoic acid (mHBA), p-nitrophenol (pNP),
hydroquinone (HDQ) and benzamide (BZA), but it disintegrates in the presence of salicylic acid (SA),
3,5-dinitrobenzoic acid (dNBA), acetamide (ACA) and melamine (MLM).
The cocrystal selected for this study is theophylline-p-hydro-
Introduction
xybenzoic acid, TP?pHBA. It can be synthesized readily in high
yield and purity and has been characterized by single crystal
Pharmaceutical cocrystals,1,2 cocrystals comprised of at least one
X-ray crystallography,22 which provides insight into the hydro-
active pharmaceutical ingredient (API), have been shown to
improve a number of physical properties of the API, including
poor solubility,3–8 poor hydration stability,9,10 poor compressi-
gen bonding interactions sustaining the cocrystal. The crystal
structure of TP?pHBA, shown in Fig. 1, reveals that each TP
bility11 and poor thermal stability.7,12,13 Since pharmaceutical
molecule interacts with two pHBA molecules via supramolecular
2
synthon23 I (R2 (9) motif)24 and synthon II (D(2) motif).
cocrystals offer great potential to address the limitations of
certain APIs, they have gained a lot of interest in recent years.14–16
The selected competing coformers and abbreviations, shown
However, if a promising pharmaceutical cocrystal were to be
formulated into a tablet, how likely is it that it will stay intact in
the presence of additives, excipients and binding materials that go
into making the drug tablet? Grinding has been shown to effect
chemical transformations in the solid state,17,18–20 as well as cause
cocrystal disintegration.21 So how robust is a pharmaceutical
cocrystal during grinding with additives in the solid state? This
study addresses the question by evaluating the robustness of a
pharmaceutical cocrystal containing theophylline, a muscle
relaxant used in asthma medications. Over 40 cocrystals of
theophylline have been reported in the literature22 making it an
excellent model compound for our study.
in Fig. 2, are: p-(N,N-dimethylamino)benzoic acid (dMABA),
benzoic acid (BA), p-nitrobenzoic acid (pNBA), o-hydroxyben-
zoic acid (salicylic acid, SA), m-hydroxybenzoic acid (mHBA),
3,5-dinitrobenzoic acid (dNBA), p-nitrophenol (pNP), hydro-
quinone (HDQ), benzamide (BZA), acetamide (ACA) and
melamine (MLM). Five of the 11 coformers are known to form
cocrystals with TP, namely BA,25 pNBA,25 dNBA,22 SA22 and
pNP,26 and the latter two were characterized by single crystal
X-ray diffraction (CSD refcodes KIGLES and TOPPNP,
respectively). All selected coformers are capable of interacting
with theophylline through one of the synthons illustrated in
aDepartment of Chemistry, The College of New Jersey, 2000 Pennington
Rd, Ewing, NJ, USA. E-mail: abourahm@tcnj.edu; Fax: 609-637-5157;
Tel: 609-771-3302
bDepartment of Chemistry, Penn State University, 104 Chemistry Building,
University Park, PA 16802, USA
{ Electronic supplementary information (ESI) available:
Characterization data of TP?ACA and TP?MLM?DMSO including
CIF, PXRD, 1H NMR, DSC and IR. In addition, complete PXRD data
for all reactions described herein and some DSC data are included.
CCDC reference number 879008. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c2ce25495f
{ The authors acknowledge The College of New Jersey for financial
support (startup funds), Dr Christopher Cahill and Nicholas Deifel for
crystal data collection and Professor Christer Aakero¨y for his invaluable
feedback and the many insightful and fruitful discussions during the
preparation of this manuscript.
Fig. 1 Crystal structure of TP?pHBA (CSD refcode KIGLOC).
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm

View Online
Results
Table 1 gives an overview of the results obtained in this study.
The TP?pHBA cocrystal is found to be robust in the presence of
7 coformers (Table 1, a–b), but disintegrates in the presence of 4
coformers (Table 1, c–e). Grinding time does not appear to have
an effect on the composition of the ground product, qualitatively.
During the course of this study, 2 new theophylline cocrystals
were discovered, namely TP?ACA and TP?MLM?DMSO, and
fully characterized by PXRD, IR, DSC and 1H NMR.
TP?MLM?DMSO was additionally characterized by single crystal
X-ray crystallography and the complete characterization data for
both cocrystals is included in the electronic supplementary
information (ESI). {Details of the competition and selectivity
experiments evaluating the robustness of TP?pHBA are described
herein.
Competition experiments
Competition experiments were conducted to evaluate the robust-
ness of TP?pHBA in the presence of a competing coformer. Fig. 4
shows a typical result from the solvent-drop grinding (SDG) of
TP?pHBA with a stoichiometric amount of a coformer, in this
case BZA.
Fig. 2 Chemical structure and abbreviation of coformers employed in
evaluating the robustness of TP?pHBA.
The PXRD pattern of the product shows that TP?pHBA stays
intact in the presence of BZA and no displacement of pHBA by
BZA is observed. Indeed, this was observed for 4/11 coformers
used in this study (coformer = BZA, BA, pNP and HDQ), as
illustrated in the following equation:
Fig. 1, in addition to hydrogen bonding to the second carbonyl
functionality in TP. The benzoic acid coformers were selected to
have varying competency of hydrogen bonding.
TP?pHBA + coformer A TP?pHBA + coformer
Two general types of experiments, outlined in Fig. 3, were
conducted in this study: (A) competition experiments involved
grinding TP?pHBA with a stoichiometric amount of a coformer
to see if the cocrystal stays intact; and (B) selectivity experiments
involved grinding TP with a stoichiometric binary mixture of
pHBA and a second coformer to determine the affinity of TP for
pHBA. Solvent-drop grinding (SDG) was utilized in these
experiments since it has been shown to enhance reaction kinetics
significantly and improve yields.20,27 The effect of grinding time
on the reaction was examined by sampling the ground mixture
after 20, 40 and 60 min. Powder X-ray diffraction was used to
qualitatively analyse the powders after grinding and identify the
species present. No quantitative measurements were conducted
in this study.
In the case of mHBA, dMABA and pNBA, the cocrystal stays
intact; however, each of the coformers undergoes a phase
change, as determined by PXRD. The outcome from these
experiments is schematically illustrated in Table 1 (b).
When stoichiometric amount of ACA is ground with
TP?pHBA, the cocrystal partially disintegrates. PXRD data
(Fig. 5) shows that a mixture of TP cocrystals is present, namely
Table 1 A summary of the results obtained from experiments evaluat-
ing the robustness of TP?pHBA in the presence of competing coformers:
(a) The cocrystal is robust and the coformer is recovered intact; (b) the
cocrystal is robust, but the coformer undergoes a polymorphic change;
(c) a mixture of TP cocrystals results in addition to a cocrystal of the two
coformers; (d) the cocrystal disintegrates and the coformer is recovered
along with TP and a cocrystal of pHBA and coformer; (e) a new phase
containing all three components forms
Fig. 3 Two general types of experiments conducted: (A) competition
experiment and (B) selectivity experiment.
CrystEngComm
This journal is ß The Royal Society of Chemistry 2012

View Online
Fig. 4 PXRD data for (a) TP?pHBA, (b) BZA, (c) product of grinding
TP?pHBA and a stoichiometric amount of BZA.
Fig. 6 PXRD data for (a) TP?pHBA, (b) TP?SA, (c) product of the
competition experiment involving grinding TP?pHBA with stoichiometric
amount of SA.
TP?pHBA and TP?ACA, in addition to a cocrystal of the two
coformers, ACA?pHBA. The existence of the latter cocrystal was
confirmed by a separate experiment in which pHBA and ACA
were ground together and the PXRD pattern of the product was
found to match that of the ACA?pHBA cocrystal from the above
experiment. The outcome from the competition experiment with
ACA is illustrated in the following equation:
When
a stoichiometric amount of SA is ground with
TP?pHBA, a new phase is obtained, as determined by PXRD
(Fig. 6). PXRD data of the ground product shows a unique
pattern that is different from the PXRD patterns of either
cocrystal; it is also different from any of the starting materials.
The same phase is obtained from solution crystallization of
1
stoichiometric amounts of TP, pHBA and SA. H NMR of the
TP?pHBA + ACA A TP?pHBA + TP?ACA + ACA?pHBA
crystalline material obtained from solution revealed that the
product contains all three components (TP, pHBA, SA) in
1 : 1 : 1 ratio. Furthermore, the DSC heating curve of the
product shows a sharp endotherm at 183.5 uC, compared to the
DSC heating curve of a physical mixture of stoichiometric
amounts of the three compounds which shows three endotherms
(ESI, Fig. S8c{). Therefore it was concluded that grinding of
When stoichiometric amount of MLM is ground with
TP?pHBA, the cocrystal disintegrates and a mixture of products
is present according to the following equation:
TP?pHBA + MLM A TP?pHBA + MLM?pHBA + TP + MLM
TP?pHBA with SA forms
a ternary cocrystal, namely
The existence of the pHBA?MLM cocrystal was confirmed by
a separate experiment in which pHBA and MLM were ground
together and the PXRD pattern of the product was found to
match that of the MLM?pHBA cocrystal from the above
experiment.
TP?pHBA?SA, as shown schematically in Fig. 7. A similar result
is obtained when TP?pHBA is ground with stoichiometric
amount of dNBA. A new phase, the PXRD pattern of which
does not correspond to any of the cocrystals or the starting
materials, is obtained and is also believed to be a ternary
cocrystal. DSC experiments of the new phase compared to a
physical mixture of stoichiometric amounts of the three
compounds shows a sharp endotherm at 229 uC, compared to
three phase transitions for the physical mixture of the three
compounds (ESI, Fig S11c{). Examples of ternary cocrystals
prepared by SDG have previously been reported.28,29
Fig. 5 PXRD data for (a) TP?pHBA, (b) TP?ACA, (c) pHBA?ACA,
(d) product of the competition experiment involving TP?pHBA and
stoichiometric amount of ACA.
Fig. 7 Schematic representation of the competition experiment invol-
ving stoichiometric amounts of TP?pHBA and SA or dNBA.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm

View Online
Grinding TP?MLM with stoichiometric amount of pHBA
results in disintegration of the TP?MLM cocrystal and the
formation of a mixture of TP cocrystals, in addition to
MLM?pHBA, according to the following equation:
TP?MLM + pHBA A TP?MLM + TP?pHBA + MLM?pHBA
Grinding TP?ACA with a stoichiometric amount of pHBA
resulted in a mixture of TP cocrystals, in addition to ACA?pHBA.
This is the same outcome obtained from all previously described
experiments involving ACA.
Excess coformer experiments
The above-described competition and selectivity studies revealed
that TP?pHBA is robust in the presence of a stoichiometric
amount of 7 coformers and that TP selectively cocrystallizes with
pHBA in a stoichiometric binary mixture of pHBA and the same
7 coformers. However, would the presence of excess coformer
push the disintegration of TP?pHBA to completion? Competition
and selectivity experiments addressing this question were con-
ducted with two selected coformers, namely MLM and ACA.
Grinding TP?pHBA with excess MLM results in disintegration
of TP?pHBA and the cocrystallization of pHBA with MLM to
form MLM?pHBA. The existence of the latter cocrystal was
confirmed by a separate experiment in which MLM and pHBA
were ground in stoichiometric amounts, and the PXRD pattern
of the resulting product was found to be different from either
pure compound. Grinding TP with a stoichiometric amount of
pHBA and excess MLM shows that TP still has higher affinity
for pHBA and selectively forms TP?pHBA. Some of the excess
MLM cocrystallizes with pHBA and forms MLM?pHBA. In
addition, the PXRD pattern shows the presence of some TP and
MLM. Scheme 1 summarizes the findings from competition and
selectivity experiments involving excess MLM.
Fig. 8 PXRD data for (a) TP?pHBA, (b) BZA and (c) product of SDG
of stoichiometric mixture of TP, pHBA and BZA. TP selectively
cocrystallizes with pHBA to form a cocrystal while BZA is recovered.
Selectivity experiments
Selectivity experiments were conducted to determine the affinity
of TP for pHBA in the presence of a competing coformer. A
representative result from SDG experiment of TP with a
stoichiometric mixture of pHBA and BZA is shown in Fig. 8.
PXRD data reveals that TP selectively cocrystallizes with pHBA
to yield TP?pHBA, while BZA is recovered unreacted, according
to the following equation:
TP + pHBA + coformer A TP?pHBA + coformer
TP has higher affinity for pHBA in the presence of 7/11
coformers, namely BA, pNBA, dNBA, dMABA, pNP, HDQ,
BZA, and mHBA. In the case of mHBA, dMABA and pNBA,
TP does selectively cocrystallize with pHBA, but each of the
coformers undergoes a polymorphic change. These results are
consistent with the competition experiments involving the same
coformers.
Competition and selectivity experiments involving excess ACA
resulted in a mixture of TP cocrystals (TP?pHBA and TP?ACA),
A different result is obtained when ACA is used. When TP is
ground with a stoichiometric mixture of ACA and pHBA, it
cocrystallizes with each of the coformers resulting in a mixture of
TP cocrystals, namely TP?pHBA and TP?ACA. PXRD data
confirmed the presence of both cocrystals, in addition to
ACA?pHBA. These results are consistent with the results
obtained from the competition experiment involving ACA.
Grinding TP with a stoichiometric binary mixture of pHBA
and SA or pHBA and dNBA results in the same phases obtained
from the competition experiment, which were identified as
ternary cocrystals TP?pHBA?SA and TP?pHBA?dNBA, respec-
tively.
Scheme 1 Competition and selectivity experiments using equimolar and
excess MLM give the same result: TP?pHBA disintegrates in the presence
of MLM.
Reverse experiments
To further confirm that the robustness of the TP?pHBA
cocrystal is due to the high affinity of TP for pHBA and not
that the cocrystal was pre-formed, we conducted experiments
with selected coformers in which TP?coformer was ground with
stoichiometric amount of pHBA. The results from experiments
involving MLM and ACA are described below.
Scheme 2 Competition experiments with equimolar amount and excess
ACA gave the same results as selectivity experiments involving equimolar
amount and excess ACA, which were also consistent with the reverse
competition experiment involving TP?ACA and pHBA.
CrystEngComm
This journal is ß The Royal Society of Chemistry 2012

View Online
as well as ACA?pHBA. This is the same outcome as that
obtained from experiments conducted with stoichiometric
amount of ACA. Scheme 2 summarizes the outcome from all
reactions conducted with ACA.
Discussion
The objective of this study was to determine the robustness of
TP?pHBA while grinding in the solid state in the presence of
coformers with a variety of functional groups. The results
described demonstrate that TP?pHBA is generally robust and
maintains its integrity in the presence of a variety of functional
groups including carboxylic acids, amides and phenols; however,
it does disintegrate in the presence of acetamide (ACA), salicylic
acid (SA), 3,5-dinitrobenzoic acid (dNBA) and melamine
(MLM). In attempting to explain our results, we examined the
findings in the context of the functional groups of the coformers.
The intermolecular interactions sustaining TP?pHBA, coformers
and potential cocrystals were considered since the interplay
between homomeric and heteromeric interactions plays a role in
cocrystal formation.17,30
Fig. 10 Crystal structure of TP?pNP reveals that pNP interacts with TP
…
via a single hydrogen bond, OH O (CSD refcode TOPPNP).
donors. Etter’s rules24 of hydrogen bonding dictate that the best
hydrogen bond donor will interact with the best hydrogen bond
acceptor. In a mixture of TP?pHBA and either SA or dNBA, the
latter two acids are the strongest hydrogen bond donors (pKa of
pHBA is 4.6) and therefore likely to interact with the strongest
hydrogen bond acceptor, which is the nitrogen of the imidazole
ring in TP, resulting in disruption of the TP?pHBA cocrystal.
As for the phenols, they are unable to disrupt TP?pHBA
because the heteromeric TP-phenol interactions are likely to be
weaker than the heteromeric TP-pHBA interactions. Examining
the crystal structure of TP?pNP26 (Fig. 10) clearly shows that
pNP forms a single hydrogen bonding interaction with TP
Considering the amides, our results show that TP?pHBA is
robust in the presence of benzamide (BZA), but it disintegrates
in the presence of acetamide (ACA). Examining the crystal
structure of BZA31 (Fig. 9) reveals that it is sustained by the
…
˚
(OHpNP TPOLC, 2.711 A), compared to three hydrogen bonding
…
interactions between TP and pHBA (Fig. 1), (OHpHBA TPOLC
…
…
˚
˚
˚
2.676 A, CLOpHBA TPHN 2.729 A and OHpHBA TPNLC 2.712 A).
Finally, TP?pHBA disintegrates in the presence of MLM, but
no TP?MLM cocrystal is detected and trace of MLM?pHBA is
present. MLM is sustained by very strong and extensive
homomeric hydrogen bonding interactions that result in a 3D
network. Such interactions are difficult to break unless favour-
able heteromeric interactions form, which happens with pHBA
and not TP.
2
centrosymmetric R2 (8) amide dimer, which further aggregates
into a ladder structure, characteristic of primary amides.32,33,34
Breaking the strong homomeric interactions in this case is
difficult and unfavourable and therefore no disruption of
TP?pHBA cocrystal is observed. The crystal structure of
ACA35 on the other hand is not sustained by the amide dimer,
rather each ACA molecule interacts with 4 other molecules via 2
…
…
single NH O hydrogen bonds and a bifurcated O HN
hydrogen bond. Clearly, the homomeric ACA-ACA interactions
are much easier to break and therefore the heteromeric TP-ACA
interactions are more likely to form.
Although qualitative, plausible explanations are offered for
the results observed herein, it must be stressed that additional
factors, including crystal packing and lattice energy, do play a
significant role in cocrystal disruption.36
In the case of the carboxylic acid coformers, TP?pHBA is
robust in the presence of 4/6 coformers, but not in the presence
of salicylic acid (SA) and 3,5-dinitrobenzoic acid (dNBA).
Considering the relative acidity of the carboxylic acid coformers
shows that SA and dNBA are the most acidic (pKa 3.02 and 2.8,
respectively)22 and therefore the strongest hydrogen bond
Experimental
All chemicals were purchased from Sigma-Aldrich and used as
received without further purification. Melting points were
determined using a Stanford Research System EZ-melt auto-
mated melting point apparatus with digital image processing
technology at a heating rate of 2 uC min21. NMR spectra (1H
and 13C) were collected on a Varian Gemini 300 MHz. IR
spectra were collected on a Perkin Elmer Spectrum-RX FT-IR
over a range of 4000–400 cm21. DSC data were collected using a
Perkin Elmer Pyris 6 using sealed 30 mL aluminium pans and an
empty reference pan sealed in the same way as the sample under
inert nitrogen conditions (flow rate of 30 ml min21) with a
heating rate of 10 uC min21 from 25 uC to 350 uC.
Powder X-ray diffraction. Data were collected on a Bruker
D8 Focus X-ray powder diffractometer using Cu-Ka radiation
˚
(l = 1.5406 A); 40 kV and 40 mA over an angle range of 5–40u 2h
in locked coupled scan mode using a step size of 0.02 and a scan
rate of 1 step/s. Mercury 2.4 was used to calculate PXRD
patterns for cocrystals whose single crystal X-ray structures were
previously determined. The PXRD pattern for each chemical
Fig. 9 Crystal structure of (a) benzamide and (b) acetamide reveals that
benzamide is sustained by stronger hydrogen bonding interactions.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm

View Online
Synthesis of TP?ACA. A binary mixture of acetamide (32.82 mg,
0.555 mmol) and theophylline (100 mg, 0.555 mmol) was SDG with
EtOH (0.2 mL mg21) for 20 min. Vapour diffusion of toluene
into an ethyl acetate solution of the ground product yielded
microcrystalline material that was identified as TP?ACA. 1H NMR
(300 MHz, DMSO-d6) d (ppm) 13.61 (s, 1H, H–NTP), 8.08 (s, 1H,
H–CL), 7.32 (s, 1H, H–NACA), 6.73 (s, 1H, H–NACA), 3.49 (s, 3H,
CH3), 3.28 (s, 3H, CH3), 1.80 (s, 3H). IR (KBr pellet) n/cm21 3357
(N–H, s), 3298 (N–H, s), 3173 (N–H, s), 1670 (CLO, s), 1652
(CLO, s), 1564 (CLO,s); DSC melt endotherm at 269 uC.
was determined before and after 20 min SDG with EtOH to
determine if a polymorphic transition occurs. Polymorphic
transitions were observed for mHBA, dMABA and dNBA.
Single crystal X-ray diffraction. Data were collected using a
Bruker SMART APEX CCD diffractometer with monochroma-
˚
tized Mo-Ka radiation (l = 0.71073 A) connected to a KRYO-
FLEX low temperature device. Data were collected at 100 K.
Lattice parameters were determined from least-squares analysis
and reflection parameters were integrated using SAINT.
Structure was solved using SHELX-97 package. All non-
hydrogen atoms were refined anisotropically. 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.
Synthesis of TP?MLM?DMSO. Melamine (138.3 mg, 1.10 mmol)
was combined with theophylline (198.6 mg, 1.10 mmol) and
dissolved in 20 ml of a DMSO–water (7 : 3) solution. The vial
was heated to 70u until both components dissolved. The clear
1
solution was left uncapped and crystals formed within 2 days. H
NMR (300 MHz, DMSO-d6) d (ppm) 13.61 (s, 1H, H–N), 8.08
(s, 1H, HLC), 6.03 (s, 6H, H2N–), 3.49 (s, 3H, CH3), 3.28 (s, 3H,
CH3). IR (KBr pellet) n/cm21 3090 (N–H, s), 1702 (CLO, s), 1654
(CLO, s). DSC melt endotherm at 178 uC.
SDG experiments. Each experiment was conducted in a 2.5 mL
stainless steel grinding jar equipped with one 6.25 mm stainless
steel grinding ball using a SPEX SamplePrep 8000 M Mixer/Mill
at a rate of 60 Hz.
Conclusions
Competition experiments. In a typical experiment, TP?pHBA
(175 mg, 0.550 mmol) was combined with equimolar amount of
coformer and EtOH (0.2 mL mg21) and ground in the mixer/mill.
The reaction mixture was sampled at 20, 40 and 60 min.
This study examined the robustness of a pharmaceutical
cocrystal (TP?pHBA) in the solid state during grinding with
additives. We found that TP?pHBA is generally robust and does
withstand the presence of a number of functional groups,
including carboxylic acids, phenols and amides; however, it does
disintegrate in some cases (4/11). The findings are significant in
the context of pharmaceutical cocrystals, since a cocrystal must
maintain its integrity during the formulation process while being
ground with additives and excipients. It is therefore crucial that
attention be paid to the nature of the excipient used as it may
have an impact on the robustness of the cocrystal. Knowledge of
crystal structures, crystal lattice energies and the mechanism of
the reaction would provide additional insight into the factors
that result in cocrystal disintegration. Studies involving quanti-
tative analysis of cocrystal mixtures and crystal structure
determination of ternary products are currently underway.
Selectivity experiments. In a typical experiment, stoichiometric
ratio of TP (100 mg, 0.555), pHBA (76.7 mg, 0.555 mmol) and a
coformer were combined along with EtOH (0.2 mL mg21) and
ground in the mixer/mill. The reaction mixture was sampled after
20, 40 and 60 min.
Experiments involving melamine (MLM). Since the TP?MLM
cocrystal is a DMSO-solvate, two sets of SDG experiments
involving MLM were performed: one with EtOH (0.2 mL mg21
)
and one with DMSO (0.04 mL mg21). There was no noticeable
difference in the outcome of the two experiments, except for the
low crystallinity of the product obtained with DMSO.
SDG of individual coformers. Each coformer (ca. 200 mg) was
independently ground with EtOH (0.2 mL mg21) to determine if a
polymorphic change occurs. Polymorphic changes occurred for
mHBA, dMABA and pNBA, all other coformers exhibited no
change.
References
1 S. G. Fleischman, S. S. Kuduva, J. A. McMahon, B. Moulton,
R. D. B. Walsh, N. Rodr´ıguez-Hornedo and M. J. Zaworotko, Cryst.
Growth Des., 2003, 3, 909–919.
¨
2 O. Almarsson and M. J. Zaworotko, Chem. Commun., 2004,
1889–1896.
3 D. J. Good and N. Rodri´ıguez-Hornedo, Cryst. Growth Des., 2009, 9,
2252–2264.
4 J. F. Remenar, S. L. Morissette, M. L. Peterson, B. Moulton, J. M.
SDG of stoichiometric amount of pHBA and coformer. A
stoichiometric binary mixture of pHBA and each of the coformers
¨
MacPhee, H. R. Guzma´n and O. Almarsson, J. Am. Chem. Soc.,
2003, 125, 8456–8457.
(total mass ca. 200 mg) was ground along with EtOH (0.2 mL mg21
)
for 20 min to determine if a new phase forms.
5 D. P. McNamara, S. L. Childs, J. Giordano, A. Iarriccio, J. Cassidy,
M. S. Shet, R. Mannion, E. O’Donnell and A. Park, Pharm. Res.,
2006, 23, 1888–1897.
6 D. J. Good and N. Rodr´ıguez-Hornedo, Cryst. Growth Des., 2010,
10, 1028–1032.
Solution crystallizations
Synthesis of TP?pHBA?SA. Theophylline (65 mg, 0.361 mmol),
pHBA (50 mg, 0.362 mmol) and SA (50 mg, 0.362 mmol) were
dissolved in 20 mL, 10 mL and 15 mL of water–EtOH (1 : 2),
respectively. Heat was needed for complete dissolution. The clear
solutions were combined and left to slow evaporate. Clear
microcrystals formed within ca. 2 weeks.
7 C. B. Aakero¨y, S. Forbes and J. Desper, J. Am. Chem. Soc., 2009,
131, 17048–17049.
8 H. D. Clarke, K. K. Arora, H. Bass, P. Kavuru, T. T. Ong, T. Pujari,
L. Wojtas and M. J. Zaworotko, Cryst. Growth Des., 2010, 10,
2152–2167.
9 A. V. Trask, W. Motherwell and W. Jones, Int. J. Pharm., 2006, 320,
114–123.
CrystEngComm
This journal is ß The Royal Society of Chemistry 2012

View Online
10 A. V. Trask, W. D. S. Motherwell and W. Jones, Cryst. Growth Des.,
2005, 5, 1013–1021.
23 G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311–2327.
24 M. C. Etter, Acc. Chem. Res., 1990, 23, 120–126.
ˇ ´
11 S. Karki, T. Frisˇcic, L. Fa´bia´n, P. R. Laity, G. M. Day and W. Jones,
Adv. Mater., 2009, 21, 3905–3909.
25 M. Sekiya, T. Yoshino, H. Tanaka and Y. Ishido, Bull. Chem. Soc.
Jpn., 1973, 46, 556–561.
12 R. D. B. Walsh, M. Bradner, S. Fleischman, L. Morales, B. Moulton,
N. Rodriguez-Hornedo and M. Zaworotko, Chem. Commun., 2003,
186–187.
26 K. Aoki, T. Ichikawa, Y. Koinuma and Y. Iitaka, Acta Crystallogr.,
Sect. B: Struct. Crystallogr. Cryst. Chem., 1978, 34, 2333–2336.
27 N. Shan and W. Jones, Green Chem., 2003, 5, 728–730.
ˇ ´
13 D. Braga, E. Dichiarante, G. Palladino, F. Grepioni, M. R. Chierotti,
R. Gobetto and L. Pellegrino, CrystEngComm, 2010, 12, 3534–3536.
14 N. Schultheiss and A. Newman, Cryst. Growth Des., 2009, 9,
2950–2967.
28 T. Frisˇcic, A. V. Trask, W. Jones and W. Motherwell, Angew. Chem.,
2006, 118, 7708–7712.
ˇ ´
29 T. Frisˇcic, A. V. Trask, W. Motherwell and W. Jones, Cryst. Growth
Des., 2008, 8, 1605–1609.
15 A. V. Trask, Mol. Pharm., 2007, 4, 301–309.
16 N. Shan and M. J. Zaworotko, Drug Discovery Today, 2008, 13,
440–446.
17 M. R. Caira, L. R. Nassimbeni and A. F. Wildervanck, J. Chem.
Soc., Perkin Trans. 2, 1995, 2213–2216.
ˇ ´
18 A. Delori, T. Frisˇcic and W. Jones, CrystEngComm, 2012, 14,
2350–2362.
19 D. R. Weyna, T. Shattock, P. Vishweshwar and M. J. Zaworotko,
Cryst. Growth Des., 2009, 9, 1106–1123.
20 N. Shan, F. Toda and W. Jones, Chem. Commun., 2002, 2372–2373.
30 C. B. Aakero¨y, A. Rajbanshi, Z. J. Li and J. Desper, CrystEngComm,
2010, 12, 4231–4239.
31 C. C. F. Blake and R. Small, Acta Crystallogr., Sect. B: Struct.
Crystallogr. Cryst. Chem., 1972, 28, 2201–2206.
32 L. Leiserowitz and A. Hagler, Proc. R. Soc. London, Ser. A, 1983,
388, 133–175.
33 S. S. Kuduva, D. Bla¨ser, R. Boese and G. R. Desiraju, J. Org. Chem.,
2001, 66, 1621–1626.
34 C. B. Aakero¨y, B. M. T. Scott and J. Desper, New J. Chem., 2007, 31,
2044–2051.
ˇ ´
21 T. Frisˇcic, L. Fa´bia´n, J. C. Burley, W. Jones and W. D. S.
35 W. Denne and R. Small, Acta Crystallogr., Sect. B: Struct.
Crystallogr. Cryst. Chem., 1971, 27, 1094–1098.
36 N. Schultheiss, M. Roe and S. X. M. Boerrigter, CrystEngComm,
2011, 13, 611–619.
Motherwell, Chem. Commun., 2006, 5009–5011.
22 S. L. Childs, G. P. Stahly and A. Park, Mol. Pharm., 2007, 4,
323–338.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm