Understanding the p-toluenesulfonamide/triphenylphosphine oxide crystal chemistry: A new 1:1 cocrystal and ternary phase diagram

Article abstract of DOI:10.1021/cg201300e

A novel 1:1 cocrystal between p-toluenesulfonamide and triphenylphosphine oxide has been prepared and structurally characterized. This 1:1 cocrystal was observed to form during solid state grinding experiments, with subsequent formation of a known 3:2 cocrystal in the presence of excess sulfonamide. Both cocrystals are stable in the solid state. The ternary phase diagram for the two coformers was constructed in two different solvents: Acetonitrile and dichloromethane. Examination of these diagrams clarified solution crystallization of both the newly discovered 1:1 cocrystal and the previously reported 3:2 cocrystal, and identified regions of stability for each cocrystal in each solvent. The choice of solvent was found to have a significant effect on the position of the solid state regions within a cocrystal system.

Full text of DOI:10.1021/cg201300e

Article
pubs.acs.org/crystal
Understanding the p-Toluenesulfonamide/Triphenylphosphine
Oxide Crystal Chemistry: A New 1:1 Cocrystal and Ternary Phase
Diagram
Denise M. Croker,*,†,§ Michael E. Foreman,†,‡ Bridget N. Hogan,†,§ Nuala M. Maguire,†,‡
Curtis J. Elcoate,†,‡ Benjamin K. Hodnett,†,§ Anita R. Maguire,†,∥ Åke C. Rasmuson,†,§
and Simon E. Lawrence†,‡
‡
†Solid State Pharmaceutical Cluster, Department of Chemistry, Analytical and Biological Chemistry Research Facility, University
College Cork, Cork, Ireland
§Materials and Surface Science Institute and the Department of Chemical and Environmental Sciences, University of Limerick,
Limerick, Ireland
∥Department of Chemistry and School of Pharmacy, Analytical and Biological Chemistry Research Facility, University College Cork,
Cork, Ireland
S
* Supporting Information
ABSTRACT: A novel 1:1 cocrystal between p-toluenesulfonamide and
triphenylphosphine oxide has been prepared and structurally charac-
terized. This 1:1 cocrystal was observed to form during solid state
grinding experiments, with subsequent formation of a known 3:2
cocrystal in the presence of excess sulfonamide. Both cocrystals are
stable in the solid state. The ternary phase diagram for the two
coformers was constructed in two different solvents: acetonitrile and
dichloromethane. Examination of these diagrams clarified solution
crystallization of both the newly discovered 1:1 cocrystal and the
previously reported 3:2 cocrystal, and identified regions of stability for
each cocrystal in each solvent. The choice of solvent was found to have a
significant effect on the position of the solid state regions within a
cocrystal system.
coformers in methanol yielded the cocrystal, but in water
gave only the acid. A phase diagram was constructed by
allowing the coformers and the cocrystal to equilibrate
individually in methanol and water, with varying coformer
ratios. In methanol, it was observed that the solubility curve of
the cocrystal crossed the 1:1 stoichiometric ratio line, meaning
that it is possible to crystallize the cocrystal from a
stoichiometic solution in methanol. In water, the cocrystal
solubility curve does not cross the 1:1 stoichiometric ratio line,
and thus it is not possible to obtain the cocrystal from a
INTRODUCTION
■
Cocrystals, crystalline structures composed of two or more
neutral components, have the potential to alter the physical
properties of solid state materials, and as such are of great
interest to the pharmaceutical industry.1 Research in cocrystal-
lization has greatly increased in recent years, yet large scale
production of a commercial cocrystal product remains elusive.
This is mainly due to a lack of robustness in cocrystal
preparation methods. Solid state grinding, solvent drop
grinding, and solution methods are routinely used to produce
cocrystals,2,3 but with a certain element of trial and error
involved. The majority of the work to date has focused on
discovery of new cocrystal structures, although there has been
recent work on understanding the mechanism of formation of
cocrystals,4−10 as well as combining prediction and experiment
to design novel cocrystal forms.11
́
stoichiometric aqueous solution. Rodriguez-Hornedo et al.
made use of a ternary phase diagram to describe the stability of
two cocrystals (2:1 and 1:1 ratios) of carbamazepine with 4-
aminobenzoic acid.7 Both cocrystals could be obtained from
slow evaporation of solutions of the coformers in ethanol, with
the cocrystal stochiometry dictated by the acid concentration.
Billot et al. used discontinuous isoperibolic thermal analysis
(DITA) to produce phase diagrams of an active pharmaceut-
Ternary phase diagrams can be used to describe the stability
of a desired cocrystal in terms of the concentration of both of
the coformers in a given solvent.4−10 Chiarella et al. used
ternary phase diagrams to understand the formation of a 1:1
cocrystal of trans-cinnamic acid and nicotinamide.4 Evaporation
of a solution containing stoichiometric amounts of both
Received: September 30, 2011
Revised: December 12, 2011
Published: December 15, 2011
© 2011 American Chemical Society
869
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
metrically. Excess solid was charged to solvent at 20 °C and
equilibrated with constant agitation for 24 h, at which point agitation
was stopped and the solids were allowed to settle for 1 h. Three 1 mL
samples of the clear solution were filtered into preweighed glass vials
(M1) and weighed (M2). The solvent was allowed to evaporate in a
fume hood (24 h) before transferring the glass vial to a vacuum oven
icals ingredient (API) and glutaric acid in a range of different
solvents and used the results to build a predictive model for
cocrystal stability in alternative solvents.5 A number of authors
have reviewed the thermodynamics involved in cocrystal
formation.9,12−14
Triphenylphosphine oxide (Ph3PO) is an excellent hydro-
gen-bond acceptor and has therefore attracted attention as a
cocrystal coformer.15 Sulfonamides have also been used as
coformers due to their hydrogen-bond donor ability, for
example, as recently shown by Nangia et al.16 The commonly
observed motif is the R42 (8) ring, usually with a 1:1
stoichiometry. Glidewell et al. reported a 3:2 cocrystal of p-
toluenesulfonamide (TSA) with Ph3PO,17 as well as a 1:1
cocrystal of TSA with the related m-tritolylphosphine oxide.18
o
at 50 C (overnight). The vial was allowed to return to room
temperature before weighing the final dry weight (M3). The formula
(M3 − M1)/(M2 − M3) revealed the solubility, expressed as g of solid/
g of solvent.
The solubility of pure TSA and the 1:1 cocrystal as a function of
Ph3PO concentration in MeCN and CH2Cl2 was determined at 20 °C
by equilibrating the desired phase in solutions of known Ph3PO
concentration. The solubility of pure Ph3PO and the 3:2 cocrystal as a
function of TSA concentration in MeCN and CH2Cl2 was determined
at 20 °C by equilibrating the desired phase in solutions of known TSA
concentration. Gravimetric analysis as described above was used to
calculate solubility, with mass balance used to account for the mass of
the known component.
The solubility of the 1:1 cocrystal in MeCN at 20 o C was measured
in the same way, but efforts to measure the solubility of the 3:2
cocrystal in MeCN, or the 1:1 or 3:2 cocrystal in CH2Cl2, were
unsuccessful due to transformation of the solid form within the
equilibration time.
Invariant points, also referred to as eutectic points or transition
concentrations, are fixed solution concentrations at which two solid
phases can exist together in equilibrium; in the present work, namely,
TSA and the 3:2 cocrystal (C1), the 3:2 cocrystal and the 1:1 cocrystal
(C2), and the 1:1 cocrystal and Ph3PO (C3). These points were
determined by generating a slurry of the two required solid forms,
Figure 1. Chemical structure of (a) p-toluenesulfonamide (M.P. =
134−137 °C) and (b) triphenylphosphine oxide (M.P. = 154−158
oC).
The cocrystallization of TSA with Ph3PO was investigated,
and herein we report a new 1:1 cocrystal of TSA with Ph3PO.
The formation of the two cocrystals was investigated in the
solid state and in solution, and ternary phase diagrams were
constructed to identify regions of stability for each cocrystal in
two solvents, acetonitrile and dichloromethane.
described as (C1), (C2), (C3) above, using the method described by
14
́
Rodriguez-Hornedo et al. The slurries were equilibrated for 24 h at
20 °C, after which the solvent content of the liquid phase was
determined using gravimetric analysis, and the concentration of TSA
and Ph3PO in the liquid phase was determined using HPLC.
The experimental setup for all solubility measurements consisted of
a thermostatic water bath (Grant GR150 with S38 stainless steel water
bath; 26 L; stability 0.005 °C and uniformity 0.02 °C @ 37 °C)
with a serial magnetic stirrer plate placed on the base. Agitation was
provided by use of 10 mm magnetic stirrer bars in 5 mL glass vials.
Construction of the Ternary Phase Diagram. The solubility of
the pure substances and the invariant points were converted to mass
fraction on a total mass basis (TSA + Ph3PO + solvent) and plotted on
a ternary axis in both MeCN and CH2Cl2 to generate the appropriate
ternary phase diagrams using ProSim Ternary Diagram software. Mass
fraction was chosen in preference to mole fraction as the use of mole
fraction tended to compress the solution phase region making
visualization of the solubility curves difficult.
Differential Scanning Calorimetry (DSC). DSC was performed
on a TA Instruments Q1000 incorporating a refrigerated cooling
system. Samples (3−5 mg) were crimped in nonhermetic aluminum
pans and scanned from 30 to 300 °C at a heating rate of 10 °C/min
under a continuously purged dry nitrogen atmosphere.
Single Crystal Diffraction. X-ray diffraction data were collected
on a Bruker APEX II DUO diffractometer using graphite
monochromatized Mo Kα radiation (λ = 0.7107 Å) and cooled
using an Oxford Cryosystems COBRA fitted with a N2 generator. All
calculations were performed using the APEX2 software suite,19,20 and
the diagrams were prepared using Mercury 2.4.21
EXPERIMENTAL SECTION
Chemicals. p-Toluenesulfonamide and triphenylphosphine oxide
were obtained from Sigma and used as received. Acetonitrile and
dichloromethane were reagent grade.
■
Solid State Grinding Experiments. Grinding experiments were
performed using a Retsch MM400 ball mill fitted with 5 mL grinding
jars containing one 2.5 mm stainless steel grinding ball per jar. The
mill was operated at 30 Hz frequency. The initial experiment involved
grinding a 1:1 ratio of the two coformers for 30 min on a 1 mmol scale.
Subsequent experiments were undertaken on a 1 mmol as detailed in
Tables 1 and 2.
Crystallization Experiments. p-Toluenesulfonamide (0.171 g,
1.00 mmol) and triphenylphosphine oxide (0.278 g, 1.00 mmol) were
dissolved in MeCN (6 mL) and placed in a sample vial. Toluene (4
mL) was layered on top of the solution, and the system was left to
stand for 21 days to give crystals suitable for single crystal diffraction.
p-Toluenesulfonamide (0.171 g, 1.00 mmol) and triphenylphos-
phine oxide (0.278 g, 1.00 mmol) were dissolved in CH2Cl2 (10 mL),
placed in a sample vial, and left to stand for 14 days to give crystals
suitable for single crystal diffraction.
Large amounts of each cocrystal form were produced by cooling
crystallization in a HEL Polyblock  a glass reaction vessel with
automated heating and cooling provided by a Julabo UC012T-H
Unichiller. For the 3:2 cocrystal, p-toluenesulfonamide (2.57 g, 0.015
mmol) and triphenylphosphine oxide (4.17 g, 0.015 mmol) were
dissolved in CH2Cl2 (50 mL), heated to reflux, and maintained under
reflux for 1 h. For the 1:1 cocrystal, p-toluenesulfonamide (5.60 g,
0.033 mmol) and triphenylphosphine oxide (11.23 g, 0.040 mmol)
were dissolved in MeCN (50 mL), heated to 70 °C, and maintained at
this temperature for 1 h. The solutions were cooled at 0.1 ° C min−1 to
5 °C, and aged for 24 h. The crystals were isolated, washed with the
pure crystallization solvent (∼10 mL), and dried in a vacuum oven at
50 °C overnight. Cocrystal form was verified using powder X-ray
diffraction (PXRD).
Powder Diffraction. Powder diffraction data were collected on
either a Philips X’Pert-MPD PRO diffractometer with nickel filtered
copper Cu Kα radiation (λ = 1.5418 Å), run at 40 kV and 35 mA, 2θ =
5−35°, with a step size of 0.02° 2θ and a scan speed of 0.02° s−1, or on
a Stoe Stadi MP diffractometer with Cu Kα radiation (λ = 1.5406 Å)
̈
1
run at 40 kV and 40 mA, 2θ = 3.5−60°, with a step size of 0.5° 2θ and
a step time of 30s.
HPLC Analysis. This was performed on either a Waters Alliance
2690 Separations Module with a Waters 486 Detector using a YMC-
Pack ODS-A column (250 mm × 4.6 mm, 5 μm) eluting with
MeCN:H2O (60:40) at 1 mL/min, UV detection at 254 nm, or on a
Solubility Measurements. The solubility of pure TSA and
Ph3PO respectively in MeCN and CH2Cl2 was determined gravi-
870
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
HPLC System consisting of a Shimadzu LC10AT Pump set at 1.5 mL/
min, Shimadzu SPD-6AV Spectrophotometric detector at 254 nm, a
Waters 717 autosampler with a Beckman Coulter Ultrasphere ODS
Column (250 mm × 4.5 mm, 5 μm), eluting with MeCN:H2O
(60:40).
short time (5 min) or a long time (198 min). This was repeated
using a 3:2 molar ratio of the coformers. The results (Table 1)
suggest that the 1:1 cocrystal is formed first, before trans-
forming into the 3:2 cocrystal if sufficient sulfonamide is
present.
Examination of the crystal structures of the 1:1 and 3:2
cocrystals show significant similarity: they both possess two
terminal PPh3O molecules, with sulfonamide molecules acting
as bridges via hydrogen bonding. This suggests that a third
sulfonamide can “slot into” the structure of the 1:1 cocrystal in
the grinding experiments to generate the 3:2 cocrystal.
To test this hypothesis, the following experiments were
undertaken: (i) the 1:1 cocrystal was ground with TSA in a 1:1
ratio, (ii) the 1:1 cocrystal (1 equiv) was ground with TSA (2
equiv), and (iii) the 3:2 cocrystal was ground with Ph3PO in a
1:1 ratio (Table 2 and Supporting Information). Experiments
(i) and (ii) confirm that the 1:1 cocrystal can transform to the
3:2 cocrystal with excess sulfonamide present. Experiment (iii)
shows that the 3:2 cocrystal does not transform to the 1:1
cocrystal when ground with an excess of Ph3PO. PXRD analysis
indicates no changes to either cocrystal over the period of 18
months. Having examined the stability of the cocrystals in the
solid state, we proceeded to undertake solution-based stability
studies.
RESULTS AND DISCUSSION
■
Solid State Grinding. Solid state grinding of an equimolar
mixture of TSA and Ph3PO for 30 min yielded a product which
showed the presence of two sharp endothermic peaks at 135
and 139 °C in the DSC analysis, Figure 2. The 3:2 cocrystal is
Solubility Measurements. The solubility values for TSA
and Ph3PO in each solvent at 20 °C are listed in Table 3. The
order of solubility between TSA and Ph3PO varies in the two
solvents: TSA is more soluble than Ph3PO in MeCN, but much
less soluble than Ph3PO in CH2Cl2. Ph3PO is extremely soluble
in CH2Cl2, exceeding the solubility of TSA by a factor of 35. It
has been reported that the relative solubilities of the two
cocrystal components in a solvent can be used to prepare a
cocrystal in that solvent,4 and a screening method has been
developed around this concept.23
Production of pure cocrystal forms using cooling crystal-
lization was confirmed using PXRD (Figure 4). The solubility
of the 1:1 cocrystal was successfully determined in MeCN. The
concentration of TSA in a solution in equilibrium with the 1:1
cocrystal is lower than the concentration of TSA in a solution in
equilibrium with pure TSA. The solubility of the 3:2 cocrystal
could not be determined in MeCN. Equilibration of the 3:2
cocrystal in MeCN resulted in formation of the 1:1 cocrystal
within 24 h, a phenomenon known as incongruent dissolution.
Both cocrystals dissolved incongruently in CH2Cl2. Equili-
bration of the 3:2 cocrystal resulted in the formation of pure
TSA, and equilibration of the 1:1 cocrystal resulted in the
production of a mixture of the 3:2 cocrystal and TSA.
Incongruent dissolution is an indication of an unsymmetrical
ternary phase diagram and usually occurs when there is a large
difference in solubility between the two pure conforming
phases in that solvent.4
Experimentally determined invariant points in each solvent
are expressed in terms of mass fraction in Table 4.
The ternary phase diagrams in MeCN and CH2Cl2 are
presented in Figures 5 and 6 respectively.
TSA and Ph3PO have a reasonably similar solubility in
MeCN resulting in an approximately symmetrical ternary phase
diagram. The 3:2 (region 4) and 1:1 (region 6) cocrystal are
independently stable across a wide range of solution
compositions. The 1:1 component stoichiometric line (solid
line) intersects the solubility curve for the 1:1 cocrystal,
indicating congruent dissolution for this cocrystal. Dissolution
of the 1:1 cocrystal results in a solution with 1:1 molar
Figure 2. DSC trace of the product obtained from 30 min grinding of
a 1:1 molar mixture of TSA and Ph3PO.
known to have a melting point of 138 °C.17 The PXRD pattern
of this product showed the presence of the known 3:2 cocrystal
and extra peaks which were not due to an excess of either
starting material.
Single crystal analysis of crystals obtained from MeCN,
melting point 134−136 °C, revealed a novel 1:1 cocrystal
(Figure 3). Each amide hydrogen atom is involved in a discrete
Figure 3. The cocrystals of TSA with PPh3O: (i) the 1:1 cocrystal, left,
showing the R42 (8) ring and (ii) the 3:2 cocrystal,17 right, which also
has the R42 (8) motif.
hydrogen bond to a phosphine oxygen atom, leading to a
discrete four molecule hydrogen-bonded complex containing a
R42 (8) ring at the binary level.22 The PXRD data for the
unknown material observed in the grinding experiments is
consistent with the single crystal data for this new 1:1 cocrystal.
Crystals grown from CH2Cl2, melting point 138−140 °C, were
found to match the reported 3:2 structure.17 The X-ray
diffraction pattern of each form was generated in Mercury 2.4
from the respective crystallographic information files (CIF) file
and is presented in Figure 4.
In order to gain some insight into the stability of the
cocrystals, we undertook further grinding experiments. Thus, a
1:1 molar ratio of the two coformers was ground for either a
871
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
Figure 4. PXRD patterns recorded for the 1:1 (A) and 3:2 (B) cocrystals of TSA and Ph3PO. Experimental patterns are shown as solid lines with the
corresponding theoretical pattern given as a dotted line.
Table 1. Effect of Reagent Composition and Grinding Time
upon Cocrystal Output
Table 4. Invariant Points in MeCN and CH2Cl2 at 20 °C
invariant point (mass fraction,
total mass basis)
grind time
(min)
material input
material output
solvent
MeCN
solid forms in equilibrium
XTSA
XPh PO
XSolvent
3
1:1 ratio of coformers
1:1 ratio of coformers
3:2 ratio of coformers
3:2 ratio of coformers
5
198
5
1:1 cocrystal, plus both coformers
1:1 cocrystal
TSA + 3:2 (C1)
3:2 + 1:1(C2)
0.195
0.087
0.018
0.047
0.035
0.031
0.014
0.041
0.150
0.075
0.205
0.357
0.791
0.871
0.832
0.878
0.760
0.612
1:1 cocrystal, plus both coformers
3:2 cocrystal
1:1 + Ph3PO (C3)
TSA + 3:2 (C1)
3:2 + 1:1 (C2)
198
CH2Cl2
Table 2. Effect of Grinding the Cocrystal with One
Coformer
1:1 + Ph3PO (C3)
In CH2Cl2 the significant difference in solubility between
TSA and Ph3PO results in an unsymmetrical diagram, with all
regions skewed toward the Ph3PO axis of the diagram. Both
cocrystals dissolve incongruently, with dissolution of the 3:2
cocrystal initially resulting in solution of invariant composition
in equilibrium with the 3:2 cocrystal and pure TSA (region 3),
potentially continuing to a solution in equilibrium with pure
TSA (region 2). Dissolution of the 1:1 cocrystal initially results
in generation of a solution saturated with respect to the 1:1 and
the 3:2 cocrystal (region 5), with continued dissolution
resulting in solution composition moving to region 4, the
region of stability for the 3:2 cocrystal. The 1:1 component
stoichiometric line will eventually pass through the C1 invariant
point on this diagram, resulting in a saturated solution of
invariant composition in equilibrium with the 3:2 cocrystal and
pure TSA. In each case, the extent of dissolution will be
controlled by the mass of the solvent present. The diagram is in
good agreement with the phase conversions observed during
efforts to measure cocrystal solubility (Table 3).
grind time
material input
(min)
material output
1:1 cocrystal with TSA (1:1 ratio)
1:1 cocrystal with TSA (2:1 ratio)
20
20
20
3:2 cocrystal plus TSA
3:2 cocrystal
3:2 cocrystal with Ph3PO (1:1
ratio)
3:2 cocrystal plus
Ph3PO
Table 3. Measured Solubility Values for each Solid Form in
MeCN and CH2Cl2 at 20 °C, Expressed in Terms of Mass
Fraction (Total Mass Basis, (MF)), and Molarity [M]
solubility
solid phase
at end of
solid
TSA
solvent MFTSA MFPh PO MFsolvent [M] equilibration
3
MeCN
0.23
0.04
0.77
0.87
0.90
1.38
0.40
0.19
TSA
Ph3PO
1:1
Ph3PO
1:1
0.13
0.06
cocrystal
3:2
cocrystal
N/A
1:1
Evaporation from a solution of 1:1 molar composition in
either solvent may be visualized by following the solid line from
the top apex of the diagram to the 1:1 cocrystal composition
point. In MeCN, this line passes straight through region 6
making it possible to isolate the 1:1 cocrystal from this solution,
regardless of the point at which evaporation is stopped. In
CH2Cl2, evaporation of a solution of 1:1 molar composition
could initially result in crystallization of pure TSA as the solid
line skims the solubility curve for TSA. Continued evaporation
will move the solution composition into region 4, making it
possible to isolate a pure 3:2 cocrystal product, but only if
evaporation is stopped within this region. Further evaporation
will result in concomitant crystallization of the 1:1 cocrystal as
the solution composition moves into region 5. This
demonstrates that the product of an evaporative crystallization
can be dependent on the extent of solvent evaporation in a
TSA
Ph3PO
1:1
cocrystal
3:2
cocrystal
CH2Cl2
0.01
0.99
0.56
0.11
3.80
TSA
0.44
Ph3PO
3:2 + TSA
N/A
N/A
TSA
stoichiometry, making it possible to experimentally measure the
solubility of the 1:1 cocrystal in MeCN. In contrast, the 3:2
component stoichiometric line (dashed line) does not intersect
the solubility curve of the 3:2 cocrystal, explaining the observed
incongruent dissolution of this form. This line instead intersects
the solubility curve for the 1:1 cocrystal. Dissolution of the 3:2
cocrystal results in establishment of an equilibrium between a
saturated solution and the 1:1 cocrystal.
872
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
Figure 5. Ternary phase diagram for the TSA−Ph3PO−MeCN system at 20 °C. Regions in the diagram are as follows: (1) solution phase; all other
regions consist of a saturated solution in contact with (2) TSA only, (3) TSA + 3:2 cocrystal, (4) 3:2 cocrystal only, (5) 3:2 cocrystal + 1:1 cocrystal,
(6) 1:1 cocrystal only, (7) 1:1 cocrystal + Ph3PO, and (8) Ph3PO only. Values are in mass fractions.
cocrystallization experiment, depending on the shape of the
ternary phase diagram. A ternary phase diagram represents a
system at thermodynamic equilibrium, and the above discussion
is based on the assumption that thermodynamic equilibrium is
maintained during the evaporation process. Crystallization is a
dynamic process in which kinetic and thermodynamic factors
compete to determine the solid product form, with kinetic
factors most likely to promote the formation of a metastable
product, but this is not specifically addressed in this work.
The same seven solid form regions (regions 2−8) were
observed in each solvent, but the choice of solvent did affect the
positioning of each solid form region within the ternary phase
diagram. The shape of a ternary phase diagram is dependent on
the relative solubility of both coformers in that solvent.
Selection of a solvent which has a particular affinity for either,
or both, coformers is a means of controlling the shape of the
ternary phase diagram, and the region of stability for the desired
crystal form.
Construction of the ternary phase diagram for p-
toluenesulfonamide and triphenylphosphine oxide in MeCN
and CH2Cl2 clearly identified regions of stability for the 1:1 and
3:2 cocrystal in these solvents, allowing for rationalization of
the experimental isolation of the 1:1 cocrystal from a solution
of 1:1 molar stoichiometry in MeCN, and the 3:2 cocrystal
from a solution of 1:1 molar stoichiometry in CH2Cl2. This
result demonstrates that solution stoichiometry during cocrystal
preparation or screening does not necessarily reflect the
stoichiometry of the resulting cocrystal. The solubility of both
cocrystal coformers must be taken into consideration when
performing a cocrystal search to ensure that cocrystals are
discovered and that those of different stoichiometry are not
missed.
Congruent and incongruent cocrystal dissolution was
observed in MeCN for the 1:1 and the 3:2 cocrystal,
respectively, with both cocrystals dissolving incongruently in
CH2Cl2. Incongruent dissolution prevents the measurement of
pure cocrystal solubility with traditional methods, as it becomes
impossible to equilibrate the pure cocrystal in the pure solvent
to generate the required saturated solution. Incongruent
dissolution is likely when there is a significant difference
between the solubility of the pure coformers in the solvent.
Knowledge of the ternary phase diagram informs effective
experimental design to control the desired cocrystal form. The
choice of solvent in a cocrystal system has a major impact on
the shape of the resulting ternary phase diagram.
CONCLUSIONS
■
A novel 1:1 cocrystal of p-toluenesulfonamide and triphenyl-
phosphine oxide has been obtained by crystallizing the two
coformers from acetonitrile, and single crystal analysis
successfully applied to identify the new crystal structure. The
1:1 cocrystal has some structural similarity to the known 3:2
cocrystal, with the latter related by a simple incorporation of an
extra sulfonamide molecule between two bridging phosphine
oxide molecules.
873
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
Figure 6. Ternary phase diagram for the TSA−Ph3PO−CH2Cl2 system at 20 °C. Regions in the diagram are as follows: (1) solution phase; all other
regions consist of a saturated solution in contact with (2) TSA only, (3) TSA + 3:2 cocrystal, (4) 3:2 cocrystal only, (5) 3:2 cocrystal + 1:1 cocrystal,
(6) 1:1 cocrystal only, (7) 1:1 cocrystal + Ph3PO, and, (8) Ph3PO only. Values are in mass fraction, total mass basis.
(4) Chiarella, R. A.; Davey, R. J.; Peterson, M. L. Cryst. Growth Des.
2007, 7, 1223−1226.
(5) Ainouz, A.; Authelin, J.-R.; Billot, P.; Liebermann, H. Int. J.
ASSOCIATED CONTENT
* Supporting Information
X-ray crystallographic information in CIF format, DSC and
PXRD data, and additional figures. This material is available
free of charge via the Internet at http://pubs.acs.org. The
crystallographic data for the 1:1 cocrystal have been deposited
with the Cambridge Crystallographic Data Centre, CCDC
number 843627. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
■
S
Pharm. 2009, 374, 82−89.
(6) Chadwick, K.; Davey, R. J.; Sadiq, G.; Cross, W.; Pritchard, R.
CrystEngComm 2009, 11, 412−414.
(7) Jayasankar, A.; Reddy, L. S.; Bethune, S. J.; Rodriguez-Hornedo,
N. Cryst. Growth Des. 2009, 9, 889−897.
(8) Yu, Z. Q.; Chow, P. S.; Tan, R. B. H. Cryst. Growth Des. 2010, 10,
2382−2387.
(9) Nehm, S. J.; Rodríguez-Sprong, B.; Rodríguez-Hornedo, N. Cryst.
Growth Des. 2006, 6, 592−600.
(10) Chadwick, K.; Davey, R. J.; Dent, G.; Pritchard, R. G.; Hunter,
C. A.; Musumeci, D. Cryst. Growth Des. 2009, 9, 1990−1999.
(11) Arlin, J.-B.; Price, L. S.; Price, S. L.; Florence, A. J. Chem.
Commun. 2011, 47, 7074−7076.
AUTHOR INFORMATION
Corresponding Author
*E-mail: Denise.Croker@ul.ie. Tel.: +353 61 234617. Fax:
+353 61 213529.
■
(12) Schartmann, R. Int. J. Pharm. 2009, 365, 77−80.
(13) Good, D. J.; Rodríguez-Hornedo, N. Cryst. Growth Des. 2010,
10, 1028−1032.
ACKNOWLEDGMENTS
■
(14) Good, D. J.; Rodríguez-Hornedo, N. Cryst. Growth Des. 2009, 9
(5), 2252−2264.
(15) (a) Etter, M. C.; Baures, P. W. J. Am. Chem. Soc. 1988, 110,
This publication has emanated from research conducted with
the financial support of Science Foundation Ireland under
Grant Numbers 07/SRC/B1158 and 05/PICA/B802/EC07.
639−40. (b) Aakeroy, C. B.; Salmon, D. J. CrystEngComm 2005, 7,
̈
439−448.
REFERENCES
(16) Goud, N. R.; Babu, N. J.; Nangia, A. Cryst. Growth Des. 2011, 11,
1930−1939.
■
(1) Schultheiss, N.; Newman, A. Cryst. Growth Des. 2009, 9, 2950−
2967.
(2) Trask, A. V.; Jones, W. Top. Curr. Chem. 2005, 254, 41−70.
(3) (a) Weyna, D. R.; Shattock, T.; Vishweshwar, P.; Zaworotko, M.
J. Cryst. Growth Des. 2009, 9, 1106−1123. (b) Trask, A. V.;
Motherwell, W. D. S.; Jones, W. Chem. Commun. 2004, 890−891.
(17) Ferguson, G.; Glidewell, C. J. Chem. Soc., Perkin Trans. 1988, II,
2129−2132.
(18) Ferguson, G.; Lough, A. J.; Glidewell, C. J. Chem. Soc., Perkin
Trans. 1989, II, 2065−2070.
(19) APEX2 v2009.3-0; Bruker AXS: Madison, WI, 2009.
874
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875

Crystal Growth & Design
Article
(20) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112−122.
(21) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodríguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466−470.
(22) Etter, M. C. Acc. Chem. Res. 1990, 23, 120−126.
(23) Ter Horst, J. H.; Deji, M. A.; Cains, P. W. Cryst. Growth Des.
2009, 9 (3), 1531−1537.
875
dx.doi.org/10.1021/cg201300e | Cryst. Growth Des. 2012, 12, 869−875