Stereoselectivity in the salt-cocrystal products formed by phenylglycinol or phenylglycine with their respective sodium or hydrochloride salts

Article abstract of DOI:10.1002/chir.22103

The salt and stereoselective cocrystal phenomena associated with 2-phenylglycinol and 2-phenylglycine have been studied using X-ray powder diffraction and differential scanning calorimetry. The chiral identities of the free acids and their sodium salts, or the free bases and their chloride salts, were found to play a determining role as to whether a salt-cocrystal product could or could not be formed. In particular, when cocrystallization of an enantiomerically pure basic or zwitterionic substance with its enantiomerically pure acid addition salt was attempted, a salt-cocrystal was only obtained when the absolute configuration of the two reactants is opposite. On the other hand, it has been found that no stereoselectivity in salt-cocrystal formation existed in the cocrystallization of an enantiomerically pure acidic or zwitterionic substance with its enantiomerically pure base addition salt. Copyright

Full text of DOI:10.1002/chir.22103

CHIRALITY (2012)
Stereoselectivity in the Salt–Cocrystal Products formed by
Phenylglycinol or Phenylglycine with their Respective Sodium or
Hydrochloride Salts
HARRY G. BRITTAIN
Center for Pharmaceutical Physics, Milford, New Jersey
ABSTRACT
The salt and stereoselective cocrystal phenomena associated with 2-phenylglycinol
and 2-phenylglycine have been studied using X-ray powder diffraction and differential scanning
calorimetry. The chiral identities of the free acids and their sodium salts, or the free bases and their
chloride salts, were found to play a determining role as to whether a salt–cocrystal product could or
could not be formed. In particular, when cocrystallization of an enantiomerically pure basic or zwit-
terionic substance with its enantiomerically pure acid addition salt was attempted, a salt–cocrystal
was only obtained when the absolute configuration of the two reactants is opposite. On the other
hand, it has been found that no stereoselectivity in salt–cocrystal formation existed in the cocrystal-
lization of an enantiomerically pure acidic or zwitterionic substance with its enantiomerically pure
base addition salt. Chirality 00:000-000, 2012. © 2012 Wiley Periodicals, Inc.
KEY WORDS: cocrystal; stereoselectivity; X-ray diffraction; thermal analysis,phenylglycinol;
phenylglycine
INTRODUCTION
chiral crystals, and the solid solutions could be crystallized to
obtain a chiral enrichment. More recently, stereoselectivity
in the salt–cocrystal products of a-methylbenzylamine and
a-methyl-benzylammonium chloride was demonstrated, where
formation of a salt–cocrystal product could take place only if the
enantiomeric identities of the salt and free base were of
opposite absolute configuration. 16
The methylbenzylamine salt–cocrystal study just mentioned
is of particular importance, as many compounds of pharmaceu-
tical interest are basic in nature and very frequently are formu-
lated in the form of their chloride salts. 17 With the recognition
One of the impediments encountered in pharmaceutics today
is that many drug substances exhibit less-than-desirable
degrees of aqueous solubility. Various formulation and drug
substance modification approaches have been brought to bear
on this issue, and one of the more intriguing advances has been
through the emergence of pharmaceutical cocrystals
the effects that cocrystallization can have on water solubility.
Aakeröy has proposed that a cocrystal is a homogeneous
crystalline solid that contains stoichiometric amounts of
1–3
and
4–6
discrete neutral molecular species that are solids under
7
that exploitation of anion coordination and anion-templated
ambient conditions,
and Schultheiss and Newman have
18
summarized six definitions of a cocrystal that go further. 8
Since the coformers making up both salt and cocrystal
systems contain acidic donor and basic acceptor functionalities,
it is clear that the ionization state and degree of proton transfer
assembly can yield new and interesting structures,
there
has arisen a driving force to improve the physical properties
of drug substances through cocrystal formation. Thus, salt–
cocrystals of fluoxetine hydrochloride were prepared with
benzoic, succinic, and fumaric acids, and the resulting
perturbation on dissolution and solubility behavior evaluated. 19
Similarly, the physical properties of the salt–cocrystal of
tiotropium fumarate with fumaric acid have been studied, and
the product was reported to be the most stable solid form of
the drug substance under ambient conditions. 20
To investigate these phenomena in more detail, and to
determine the generality of the observed stereoselectivity
reported for the a-methylbenzylamine/a-methylbenzylammo-
nium chloride salt–cocrystal system, 16 the products formed by
the enantiomers of 2-phenylglycinol and 2-phenylglycine with
their corresponding hydrochloride or sodium salts have been
investigated. As evident in Figure 1, the cocrystal formers of
the previous and present study can be viewed as being
systematically substituted derivatives of benzylamine. The
in an isolated product is essential to knowing whether one has
9–11
prepared a cocrystal or a salt.
This situation was amply
demonstrated in the cocrystallization of minoxidil with a variety
of carboxylic acids, where seven salts and only one cocrystal
product were actually isolated. 12
The salt-or-cocrystal distinction is ordinarily made on the
basis of atomic positions in a single-crystal structural determi-
nation, but this approach presents difficulties when a crystal
of suitable quality cannot be obtained. Recognizing that the
vibrational frequencies of bonded atoms involved in salt and
cocrystal interactions would be perturbed to a lesser or
greater degree depending on the nature of the proton
transfer, studies have been carried out on cocrystals formed
by salts of benzenecarboxylic acids with their respective free
acids with the aim of developing selection rules for the
13,14
recognition of cocrystal behavior.
Recently, the first example of cocrystal solid solutions for a
*Correspondence to: Harry G. Brittain, Center for Pharmaceutical
Physics, 10 Charles Road, Milford, New Jersey 08848. E-mail: hbrittain@
centerpharmphysics.com
Received for publication 29 December 2011; Accepted 02 July 2012
DOI: 10.1002/chir.22103
Published online in Wiley Online Library
(wileyonlinelibrary.com).
system containing enantiomers of opposite chirality has been
reported for the 1:2 product formed by 4,40-dipyridyl and
15
ibuprofen.
In this work, it was found that the cocrystal
products exhibited significantly higher melting temperatures
and enthalpies of fusion relative to their parent ibuprofen
© 2012 Wiley Periodicals, Inc.

BRITTAIN
were prepared in a similar manner, namely, by dissolving approximately
250 mg quantities in 10 ml of methanol, added a 1:1 stoichiometric
amount of 1.0N hydrochloric acid, evaporating to dryness, transferring
the solids to a mortar, wetting with methanol, and grinding to dryness.
Salt–cocrystal preparation. Cocrystal products were prepared using
the wet-grinding method,24 where accurately weighed, equimolar
amounts of free amine and the corresponding chloride salt (or free acid
and the corresponding sodium salt) were placed in a mortar, wetted with
methanol, and then ground to dryness.
RESULTS AND DISCUSSION
The 2-Phenylglycinol System
Since the XRPD patterns of (S)- and (R)-2-phenylglycinol
(i.e., S-PG and R-PG, respectively) must necessarily be the same,
as would the XRPD patterns of (S)- and (R)-2-phenylglycinol
hydrochloride( i.e., S-PG-Cl and R-PG-Cl, respectively), only
the patterns of the(S)-enantiomers have been shown in Figure 2.
This figure also contains the XRPD patterns of the product
isolated after the solvent-drop grinding of S-PG with S-PG-Cl, as
well as the product resulting from the solvent-drop grinding of
S-PG with R-PG-Cl. Detailed examination of the diffraction
patterns indicated that the XRPD of the (S-PG) + (S-PG-Cl)
product is simply the merged sum of the XRPD patterns of
S-PG and S-PG-HCl¸ while the XRPD pattern obtained for the
(S-PG)–(R-PG-HCl) product is definitely different from the
XRPD patterns of its coformers. These observations are
interpreted to indicate that the interaction of S-PG with R-PG-Cl
yields an authentic salt–cocrystal, while the attempted interac-
tion of S-PG with S-PG-Cl does not.
To further establish the authenticity of the (S-PG)–(R-PG-Cl)
salt–cocrystal product, and the inability to cocrystallize S-PG
and S-PG-Cl, an indexing analysis of the four XRPD patterns
of Figure 2 was carried out. It was found that the XRPD patterns
of S-PG, S-PG-Cl, and the (S-PG)–(R-PG-Cl) salt–cocrystal could
be indexed so that a single set of unit cell parameters (which
have been collected in Table 1) enabled a calculation of all
major peaks in the respective diffraction patterns. Also included
in Table 1 are the scattering angles and relative intensities of
Fig. 1. Hierarchy in the structures of the chiral cocrystal formers, with
reference to the preceding article.
formation of a particular salt–cocrystal product (or the absence of
its formation) was demonstrated by means of X-ray powder
diffraction, differential scanning calorimetry, and the total
volatile content (TVC).
MATERIALS AND METHODS
General Procedures
X-ray powder diffraction (XRPD). XRPD patterns were obtained
using a Rigaku (Woodlands, Texas, USA) MiniFlex powder diffraction
system (X-ray source being nickel-filtered Ka emission of copper),
equipped with a horizontal goniometer operating in the /2 mode. Samples
were packed into the sample holder using a back-fill procedure and were
scanned over the range of 3.5 to 40 degrees 2 at a scan rate of 0.5 degrees
2/min. Using a data acquisition rate of 1 point per second, these scanning
parameters equate to a step size of 0.0084 degrees 2. XRPD patterns were
indexed using the Rigaku PDXL software package, which enables pattern
analysis by means of DICVOL,21 ITO,22 or N-TREOR.23
Differential scanning calorimetry (DSC). DSC thermograms were
obtained on a TA Instruments (New Castle, Delaware, USA) 2910 thermal
analysis system. Samples of approximately 1–2 mg were accurately weighed
into an aluminum DSC pan, and then covered with an aluminum lid that was
crimped in place. Samples were heated at a rate of 10 ꢀC/min from ambient
temperature to termination temperatures in the range of 150 to 225 ꢀC,
depending on the system under study.
TVC. Measurements of total volatile were made using an Ohaus
(Parsippany, New Jersey, USA) model MB45 system. The samples were
heated isothermally at a temperature of 110 ꢀC for a period of 15 min,
and the percentage of lost weight recorded. In some cases, it was noted
that the samples had melted without significant weight loss.
Synthesis Procedures
Starting materials. (R)- and (S)-2-phenylglycinol and (R)- and (S)-2-
phenylglycine were obtained from Aldrich (Milwaukee, Wisconsin,
USA). Prior to their characterization, samples were wetted with methanol
and ground to dryness in a mortar.
Salt preparation. The sodium salts of the 2-phenylglycine enantiomers
were prepared by dissolving approximately 250 mg quantities in 10 ml of
methanol, to which was added a 1:1 stoichiometric amount of 1.0N
sodium hydroxide solution. The solutions were allowed to evaporate to
dryness, whereupon the recovered solids were transferred to a mortar,
wetted with methanol, and ground until a dry solid was obtained. The
chloride salts of the 2-phenylglycinol and 2-phenylglycine enantiomers
Fig. 2. X-ray powder diffraction patterns of (S)-phenylglycinol, (S)-phenyl-
glycinol hydrochloride, the isolated yield of 1:1 (S)-phenylglycinol and
(S)-phenylglycinol hydrochloride, and the 1:1 salt–cocrystal of (S)-phenylglycinol
and (R)-phenyl-glycinol hydrochloride.
Chirality DOI 10.1002/chir

STEREOSELECTIVITY IN THE SALT-COCRYSTAL PRODUCTS FORMED BY PHENYLGLYCINOL OR PHENYLGLYCINE
Chirality DOI 10.1002/chir

BRITTAIN
the observed XRPD peaks, as well as the Miller indices
The Phenylglycine System
assigned to these peaks and deduced from the indexing results.
It may be noted in passing that the unit cell parameters
obtained for S-PG-Cl agreed very well with those obtained
from a single-crystal study, and which have been published
Because amino acids ordinarily exist in the solid state in their
zwitterionic form, the amine and carboxylic acid groups are
hydrogen bonded to each other and the carboxylate proton is
partially transferred. However, this transfer is broken when a
salt of the amino acid is formed, as has been demonstrated in
the case of phenylalanine and phenylalanine hydrochloride.
25
in the Cambridge Structural Database. The existence of a
different set of unit cell parameters for the (S-PG)–(R-PG-Cl)
product that is different from the unit cell parameters of either
of its coformers is proof that this isolated substance is an
authentic salt–cocrystal.
26,27
Owing to the existence of both acidic and basic groups,
phenylglycine is capable of forming salts with either basic or
acidic reagents as long as the energy released upon proton
transfer is larger than the energy required to break the
hydrogen bonding of the zwitterion. Thus, the possibility of
stereoselective salt–cocrystal formation in the phenylglycine
system was investigated for both acidic and basic coformers.
On the other hand, the XRPD pattern of the (S-PG) + (S-PG-Cl)
product could not be indexed to a single set of unit cell
parameters. In fact, the XRPD pattern obtained by the simple
numerical averaging of the XRPD patterns of S-PG and S-PG-Cl
yielded a theoretical XRPD pattern that was the same as that of
the (S-PG) + (S-PG-Cl) product. Collectively, these findings dem-
onstrate the existence of stereoselectivity in the 2-phenylglycinol
salt–cocrystal system, since no salt–cocrystal product was
formed when S-PG and S-PG-Cl were subjected to the same
solvent-assisted, solid-state grinding process as were S-PG and
R-PG-Cl (which did yield a single crystalline phase).
The DSC thermograms obtained for the solid products are
shown in Figure 3, and except for the S-PG-Cl product, these
consisted of endothermic transitions that took place at rela-
tively low temperatures. However, when the samples were
heated isothermally at 110 ꢀC for 15 min, the samples all lost
less than approximately 0.25% of their total mass, demonstrat-
ing that these materials were obtain in the form of non-solvates.
The DSC melting endothermic transition of S-PG was
observed to exhibit a temperature maximum at 78.2 ꢀC
(enthalpy of fusion equal to 160 J/g), while S-PG-Cl melted
with decomposition at a temperature at 171 ꢀC (enthalpy of
fusion approximately equal to 120 J/g). The (S-PG) + (S-PG-Cl)
product melted at a temperature (75.4 ꢀC) that was nearly equal
to that of S-PG, and the enthalpy associated with this transition
(62.9 J/g) was approximately half that of S-PG. On the other
hand, even though the (S-PG)–(R-PG-Cl) salt–cocrystal melted
at the even lower temperature of 64.0 ꢀC, the melting
endotherm was characterized by a substantial (107 J/g)
enthalpy of fusion.
Simple Phenylglycine Salts
The XRPD patterns of (S)-phenylglycine (i.e., S-PGLY),
(S)-sodium phenylglycinate (i.e., S-Na-PGLY), and (S)-phenyl-
glycine hydrochloride (i.e., S-PGLY-Cl) are shown in Figure 4.
Although the diffraction patterns of S-PGLY and S-Na-PGLY
are qualitatively similar, there are a sufficient number of
differences between the two to demonstrate formation of the
sodium salt. The three diffraction patterns were able to be
indexed, and the derived unit cell parameters are found in
Table 2. Also included in Table 2 are the scattering angles
and relative intensities of the observed XRPD peaks, as well
as the Miller indices assigned to these peaks and deduced from
the indexing results. Only the single-crystal structure of the
25
hydrochloride salt has been reported,
but the unit cell
parameters reported in this article do not agree with those of
the present study. This difference in unit cell parameters
signifies that a different polymorphic form of S-PGLY-Cl has
been obtained, which undoubtedly results from the methanol
processing step used in the present work.
DSC was used to establish the properties of phenylglycine
and its two salt forms, and the various thermograms are
collected in Figure 5. The DSC thermograms of S-PGLY itself
consisted of a single melting endothermic transition, which
was characterized by a temperature maximum at 301 ꢀC and
an enthalpy of fusion equal to 298 J/g. The thermogram of
Fig. 3. Differential scanning calorimetry thermograms of (S)-phenylglycinol,
(S)-phenyl-glycinol hydrochloride, the isolated yield of 1:1 (S)-phenylglycinol
and (S)-phenyl-glycinol hydrochloride, and the 1:1 salt–cocrystal of (S)-phenyl-
glycinol and (R)-phenylglycinol hydrochloride.
Fig. 4. X-ray powder diffraction patterns of (S)-phenylglycine, (S)-sodium
phenylglycinate, and (S)-phenylglycine hydrochloride.
Chirality DOI 10.1002/chir

STEREOSELECTIVITY IN THE SALT-COCRYSTAL PRODUCTS FORMED BY PHENYLGLYCINOL OR PHENYLGLYCINE
Chirality DOI 10.1002/chir

BRITTAIN
Fig. 6. X-ray powder diffraction patterns (XRPD) of the 1:1 salt–cocrystals
Fig. 5. Differential scanning calorimetry thermograms of (S)-phenylglycine,
(S)-sodium phenylglycinate, and (S)-phenylglycine hydrochloride.
of (S)-phenylglycine and (S)-sodium phenylglycinate, (S)-phenylglycine and
(R)-sodium phenylglycinate, and (S)-phenylglycine and (R)-phenylglycine
hydrochloride. Also shown is the XRPD pattern of the yield isolated after the
attempted interaction of (S)-phenylglycine with (S)-phenylglycine hydrochloride.
the S-PGLY-Cl salt did not contain a desolvation endotherm
but did exhibited two melting endotherms, indicating that
the isolated product was the metastable polymorphic form.
The first transition was characterized by a temperature
maximum of 216 ꢀC (enthalpy of fusion equal to 67 J/g), and
the second transition was characterized by a temperature
maximum of 242 ꢀC (enthalpy of fusion equal to 246 J/g).
The DSC thermograms of S-PGLY and S-PGLY-Cl were
completely free of low-temperature endothermic transitions,
demonstrating that these products were non-solvated.
included in Table 3 are the scattering angles and relative
intensities of the observed XRPD peaks, as well as the Miller
indices assigned to these peaks and deduced from the
indexing results. It was found that the unit cell parameters
derived from this analysis could account for all major
peaks in the diffraction patterns of the (S-PGLY)–(S-Na-PGLY),
(S-PGLY)–(R-Na-PGLY), and (S-PGLY)–(R-PGLY-Cl). The
existence of different unit cells for the salt–cocrystal products
relative to those of the coformer substances is definitive proof
that these materials are authentic salt–cocrystal substances.
It is to be noted that the diffraction pattern obtained from the
solid isolated by the attempted cocrystallization of S-PGLY with
S-PGLY-Cl could not be indexed, demonstrating that this
product consisted of more than one crystalline phase.
Thermal analysis was used to further study the stereoselec-
tivity in the phenylglycine/phenylglycine hydrochloride
system and the lack of stereoselectivity in the phenylgly-
cine/sodium phenylglycinate system. As shown in Figure 7,
the DSC thermograms of the (S-PGLY)–(S-Na-PGLY) and
(S-PGLY)–(R-Na-PGLY) salt–cocrystal products both featured
a prominent desolvation endotherms having temperature
maxima in the range of 103–105 ꢀC and which were associated
with enthalpies of desolvation in the range of 95–100 J/g.
The desolvated (S-PGLY)–(S-Na-PGYL) salt–cocrystal was
subsequently observed to melt (with decomposition) at a
temperature of 252 ꢀC, while the desolvated (S-PGLY)–
(R-Na-PGLY) salt–cocrystal melted with decomposition at a
temperature of 258 ꢀC.
The DSC thermogram of S-Na-PGLY contained a desolvation
endotherm having
a
temperature maximum of 105ꢀC,
associated with an enthalpy of desolvation equal to 209 J/g.
The TVC of this product was determined to be 15.85%, which
agrees well with the theoretically calculated weight loss
that would be associated with a mono-methanol solvatomorph
(i.e., 15.62%). The desolvated S-Na-PGLY product was subse-
quently observed to melt with decomposition at a temperature
of 263 ꢀC (because of the exothermic decomposition, the
enthalpy of fusion could not be approximated).
Phenylglycine Salt–Cocrystals
The XRPD patterns of the products obtained by the
interaction of 1:1 stoichiometric amounts of S-PGLY with
S-Na-PGLY, S-PGLY with R-Na-PGLY, S-PGL with S-PGLY-Cl,
and S-PGLY with R-PGLY-Cl are shown in Figure 6. Comparison
of the diffraction patterns with those of their respective coformer
compounds indicated that while new cocrystalline solids were
formed by the interaction of (S)-phenylglycine with either
enantiomer of sodium phenylglycinate, only the interaction of
(S)-phenylglycine with (R)-phenylglycine hydrochloride led to
the formation of a salt–cocrystal product.
The XRPD pattern of the solid isolated upon solvent-
drop grinding of (S)-phenylglycine with (S)-phenylglycine
hydrochloride was found to be simply the sum of the
XRPD patterns of the starting materials, and not that of a
new salt–cocrystal.
The DSC thermogram of the non-solvated (S-PGLY)–
(S-PGLY-Cl) salt–cocrystal consisted of a single melting
endothermic transition, which was characterized by a temper-
ature maximum at 247 ꢀC and an enthalpy of fusion equal
to 344 J/g. The DSC thermogram of the solid isolated
after the attempted cocrystallization of (S)-phenylglycine with
(S)-phenylglycine hydrochloride was found to strongly mimic
the DSC thermogram of S-PGLY-Cl, although the enthalpies
associated with the two thermally induced transitions were
different in magnitude.
The authenticity of the three salt–cocrystal products was
established by indexing their respective XRPD patterns, and
the results of this analysis are collected in Table 3. Also
Chirality DOI 10.1002/chir

STEREOSELECTIVITY IN THE SALT-COCRYSTAL PRODUCTS FORMED BY PHENYLGLYCINOL OR PHENYLGLYCINE
Chirality DOI 10.1002/chir

BRITTAIN
LITERATURE CITED
1. Vishweshwar P, McMahon JA, Bis JA, Zaworotko MJ. Pharmaceutical
cocrystals. J Pharm Sci 2006;95:499–516.
2. Shan N, Zaworotko MJ. The role of cocrystals in pharmaceutical science.
Drug Discov Today 2008;13:440–446.
ꢀꢀ
3. Frisci T, Jones W. Benefits of cocrystallization in pharmaceutical
materials science: an update. J Pharm Pharmacol 2010;62:1547–1559.
4. Good DJ, Rodríguez-Hornedo N. Solubility advantage of pharmaceutical
cocrystals, Cryst Growth Design 2009;9:2252–2264.
5. Bethune SJ, Huang N, Jayasanka A, Rodríguez-Hornedo N. Under-
standing and predicting the effect of cocrystal components and pH
on cocrystal solubility. Cryst Growth Design 2009;9:3976–3988.
6. Aakeröy CB, Forbes S, Desper J. Using cocrystals to systematically
modulate aqueous solubility and melting behavior of an anticancer
drug. J Am Chem Soc 2009;131:17048–17049.
7. Aakeröy C, Salmon DJ. Building co-crystals with molecular sense and
supramolecular sensibility. CrystEngComm 2005;7:439–448.
8. Schultheiss N, Newman A. Pharmaceutical cocrystals and their
physicochemical properties. Cryst Growth Design 2009;9:2950–2967.
9. Aakeröy C, Fasulo ME, Desper J. Cocrystal or salt: does it really matter?
Mol Pharm 2007;4:317–322.
Fig. 7. Differential scanning calorimetry (DSC) thermograms of the
1:1 salt–cocrystals of (S)-phenylglycine and (S)-sodium phenylglycinate,
(S)-phenylglycine and (R)-sodium phenylglycinate, and (S)-phenylglycine
and (R)-phenylglycine hydrochloride. Also shown is the DSC thermogram
pattern of the yield isolated after the attempted interaction of (S)-phenylglycine
with (S)-phenylglycine hydrochloride.
10. Childs SL, Stahly GP, Park A. The salt-cocrystal continuum: the
influence of crystal structure on ionization state. Mol Pharm
2007;4:323–338.
11. Aakeröy CB, Rajbanshi A, Li ZJ, Desper J. Mapping out the synthetic
landscape for recrystallization, cocrystallization, and salt formation.
CrystEngComm 2010;12:4231–4239.
The TVCs measured for both the (S-PGLY)–(S-Na-PGLY)
and (S-PGLY)–(R-Na-PGLY) salt–cocrystal products were
both approximately 4.75%, which agrees reasonably well
with the theoretically calculated weight loss that would be
associated with a hemi-methanol solvatomorph (i.e., 4.57%).
This finding is reasonable, considering that the sodium
phenylglycinate coformer was a mono-methanol solvatomorph,
and the phenylglycine coformer is non-solvated.
12. Schultheiss N, Lorimer K, Wolfe S, Desper J. Attempted construction of
minoxidil: carboxylic acid cocrystals; 7 salts and 1 cocrystal resulted.
CrystEngComm 2010;12:742–749.
13. Brittain HG. Vibrational spectroscopic studies of cocrystals and salts.
2. The benzylamine-benzoic acid system. Cryst Growth Design
2009;9:3497–3503.
14. Brittain HG. Vibrational spectroscopic studies of cocrystals and salts.
3. Cocrystal products formed by benzenecarboxylic acids and their
sodium salts. Cryst Growth Design 2010;10:1990–2003.
15. Chen S, Xi H, Henry RF, Marsden I, Zhang GGZ. Chiral cocrystal solid
solution: structures, melting point phase diagram, and chiral enrichment
of (ibuprofen)2(4,4-dipyridyl) CrystEngComm 2010;12:1485–1493.
CONCLUSIONS
The accumulating body of work has demonstrated that the
phenomenon of salt–cocrystal formation appears to be quite
general, and that such products can be readily formed by
the interaction of appropriate coformer reactants. However,
when the cocrystal formers contain centers of dissymmetry,
it has been learned that formation of the salt–cocrystal
product can be complicated by stereoselective effects.
In the systems studied thus far, it has been found
that cocrystallization of an enantiomerically pure basic or
zwitterionic substance (i.e., a-methylbenzylamine, 2-phenylgly-
cinol, or 2-phenylglycine) with its enantiomerically pure acid
addition salt (i.e., its hydrochloride salt) only takes place when
the absolute configuration of the two reactants is opposite.
No cocrystal formation has been detected for a base and
its hydrochloride salt when both coformers are of the same
absolute configuration. This finding indicates that such
coformers of the same configuration cannot pack efficiently,
and instead crystallize separately.
On the other hand, no stereoselectivity has been detected in
the one system studied that entailed cocrystallization of an
enantiomerically pure acidic or zwitterionic substance (i.e., 2-
phenylglycine) with its enantiomerically pure base addition salt
(i.e., its sodium salt). More work is currently being conducted
with related systems to learn if the carboxylate–cation interac-
tion is sufficiently strong so as to override crystal packing
effects that clearly exist with amino–anion interactions.
16. Brittain HG. Vibrational spectroscopic studies of cocrystals and salts.
4. Cocrystal products formed by benzylamine, a-methylbenzylamine,
and their chloride salts. Cryst Growth Design 2011;11:2500–2509.
17. Gould PL. Salt selection for basic drugs. Int J Pharm 1986;33:201–217.
18. Metrangolo P, Pilati T, Terraneo G, Biella S, Resnati G. Anion coordi-
nation and anion-templated assembly under halogen bonding control.
CrystEngComm 2009;11:1187–1196.
19. Childs SL, Chyall LJ, Dunlap JT, Smolenskaya VN, Stahly BC, Stahly
GP. Crystal engineering approach to forming cocrystals of amine
hydrochlorides with organic acids: molecular complexes of fluoxetine
hydrochloride with benzoic, succinic, and fumaric acids. J Am Chem
Soc 2004;126:13335–13342.
20. Pop M, Sieger P, Cains PW. Tiotropium fumarate: an interesting pharma-
ceutical cocrystal. J Pharm Sci 2009;98:1820–1834.
21. Boultif A, Louër D. Powder pattern indexing with the dichotomy method.
J Appl Cryst 2004;37:724–731.
22. Visser JW. A fully automatic program for finding the unit cell from powder
data. J Appl Cryst 1969;2:89–95.
23. Altomare A, Giacovazzo C, Guagliardi A, Moliterni AGG, Rizzi R,
Werner P-E. New techniques for indexing: N-TREOR in EXPO. J Appl
Cryst 2000;33:1180–1186.
24. Trask AV, Jones W. Crystal engineering of organic cocrystals by the
solid-state grinding approach. Top Curr Chem 2005;254:41–70.
25. Bohme U, Wiesner S. Crystal structure file VECWUW. Cited as a private
communication to the Cambridge Structural Database (2005).
26. Al-Jaraggiyku AR, Koetzle TF. Neutron diffraction study of L-phenylalanine
hydrochloride. Acta Cryst 1975;B31:2461–2465.
27. Ravichandran S, Dattagupta JK, Chakrabarti C. D-phenylglycine
hydrochloride. Acta Cryst 1998;C54:499–501.
Chirality DOI 10.1002/chir