Survival of the Fittest: Competitive Co-crystal Reactions in the Ball Mill

Article abstract of DOI:10.1002/chem.201500925

The driving forces triggering the formation of co-crystals under milling conditions were investigated by using a set of multicomponent competitive milling reactions. In these reactions, different active pharmaceutical ingredients were ground together with a further compound acting as coformer. The study was based on new co-crystals including the coformer anthranilic acid. The results of the competitive milling reactions indicate that the formation of co-crystals driven by intermolecular recognition are influenced and inhibited by kinetic aspects including the formation of intermediates and the stability of the reactants.

Full text of DOI:10.1002/chem.201500925

DOI: 10.1002/chem.201500925
Full Paper
&
Crystal Engineering |Hot Paper|
Survival of the Fittest: Competitive Co-crystal Reactions in the
Ball Mill
Franziska Fischer,[a, b] Maike Joester,[a, b] Klaus Rademann,[b] and Franziska Emmerling*[a]
Abstract: The driving forces triggering the formation of co-
crystals under milling conditions were investigated by using
a set of multicomponent competitive milling reactions. In
these reactions, different active pharmaceutical ingredients
were ground together with a further compound acting as
coformer. The study was based on new co-crystals including
the coformer anthranilic acid. The results of the competitive
milling reactions indicate that the formation of co-crystals
driven by intermolecular recognition are influenced and in-
hibited by kinetic aspects including the formation of inter-
mediates and the stability of the reactants.
cannot be synthesized under common synthesis conditions.[6a,7]
Besides the tremendous interest in mechanochemical synthe-
ses and the success of the method for the formation of co-
crystals,[8] a fundamental understanding of the formation path-
ways and driving forces during the milling syntheses is still
lacking.[9]
Introduction
In recent years, the interest in the synthesis of co-crystals has
increased tremendously because the co-crystallisation of an
active pharmaceutical ingredient (API) with an appropriate co-
former can improve the physicochemical properties of the
API.[1] Typically, co-crystals are defined as homogeneous, crys-
talline phases consisting of stoichiometric amounts of two or
more neutral molecular species, which are solid under ambient
conditions.[2] On the basis of noncovalent interactions between
these molecules,[3] a new crystal lattice is formed leading to
a change of the crucial properties of the respective API - like
water solubility, dissolution behavior, melting point, photoem-
ission, electronic behavior, and stability against physical or
chemical stress.[1d,4]
Although considerable work concentrates on the theoretical
calculation of possible crystal structures,[10] up to now, it is im-
possible to predict the products of co-crystal syntheses relia-
bly.
The first real-time and in situ measurements using synchro-
tron X-ray powder diffraction (PXRD) and Raman spectroscopy
provided direct insights in the formation pathway of milling re-
actions.[11] These experiments can offer an understanding of
formation pathways, but information about the driving forces
of mechanochemical syntheses is scarce. Usually, the hierarchy
of supramolecular synthons derived from CSD database entries
is a first reference point. Co-crystal structures containing co-
former with two or more different moieties in the same mole-
cule are investigated.[12]
Until now, various methods for co-crystal syntheses have
been applied, ranging from spray drying, freeze drying and
slow evaporation to the formation from melt.[5] Compared with
these typically employed co-crystallization methods, mechano-
chemistry represents an effective and environmentally sustain-
able alternative. The term “mechanochemistry” covers solid-
state reactions induced by mechanical energy, for example, by
grinding in a ball mill.[6] Mechanochemical reactions bear many
advantages. Typically, no solvent or only a small amount of sol-
vent is needed and the reactions are quantitative. Furthermore,
several co-crystals are only accessible mechanochemically and
Here, we analyze the initial interactions in the milling pro-
cess based on the molecular recognition of the reactants. To
elucidate the key factors of these interactions, a set of multi-
component milling reactions was chosen based on four new
co-crystals including the coformer anthranilic acid (ana) and
the APIs carbamazepine (cbz), salicylic acid (sa), theobromine
(tb) and theophylline (tp). Our results indicate that the co-crys-
tal-milling product is not only related to preferred intermolecu-
lar interaction but also the kinetic aspects have to be consid-
ered.
[a] F. Fischer, M. Joester, Dr. F. Emmerling
BAM Federal Institute for Materials Research and Testing
R.-Willstätter-Strasse 11, 12489 Berlin (Germany)
E-mail: franziska.emmerling@bam.de
[b] F. Fischer, M. Joester, Prof. Dr. K. Rademann
Department of Chemistry, Humboldt-Universität zu Berlin
B.-Taylor-Strasse 2, 12489 Berlin (Germany)
Results and Discussion
Syntheses and characterization of model compounds
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500925: A detailed description of the co-
crystal syntheses, the XRD and synchrotron measurements, the Raman and
ssNMR spectroscopy and the DTA-TG analysis.
We focus our investigation on co-crystals of the coformer ana,
which is known for its excellent ability to form co-crystals.
Chem. Eur. J. 2015, 21, 14969 – 14974
14969
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Grinding experiments with the APIs cbz, sa, tb, and tp were
conducted. All APIs co-crystalize with ana in stoichiometric
ratios of 1:1, 2:1, or 2:3. Additionally, a mechanochemically syn-
thesized theobromine:salicylic acid (tb:sa) co-crystal could be
identified. All pure co-crystals were synthesized by neat or
liquid-assisted grinding. The corresponding PXRD patterns
shown in Figure 1 indicate the completeness of the reaction.
On the basis of the powder patterns, the solution and refine-
ment of the crystal structure of the co-crystals carbamazepi-
ne:anthranilic acid (2:1) (cbz:ana), salicylic acid:anthranilic acid
(sa:ana), and theophylline:anthranilic acid (2:3) (tp:ana) was
possible (Figures S1–S4 in the Supporting Information). The
structure of the tb:sa was solved based on the single-crystal
data. The crystal structures and intermolecular interactions of
the new co-crystals are presented in Figure 2. As expected, all
crystal structures are dominated by the formation of intermo-
lecular hydrogen bonds leading to the formation of layers
(sa:ana), chains (tb:ana), or isolated entities (cbz:ana, tp:ana,
tb:sa).
formation of salts. With the exception of both sa co-crystals, all
co-crystals consist of neutral molecules and a salt formation
can be excluded. The Raman spectrum of the sa:ana co-crystal
shows that the protons of the molecules are bridged strongly,
since the stretching band of the carboxylate group of ana at
n˜ =1373 cmÀ1 does not vanish. The results of the correspond-
ing ssNMR investigations back this assumption, since the acidic
proton of ana leads to a signal at d=16.3 ppm. The tb:sa co-
crystal shows also a high-shifted proton signal at d=15.6 ppm.
The shift of the corresponding Raman absorption bands of the
reactants is not sufficient to indicate a complete salt formation.
The Raman band of sa at n˜ =1635 cmÀ1 (C=O stretching, car-
bonyl group) and of tb at n˜ =3116 cmÀ1 (NÀH stretching, sec-
ondary amine) shifted by approximately n˜ =15 cmÀ1 in the co-
crystal.[13] Because we did not observe the additional Raman
band of an ammonium ion, we could not assume that a com-
plete salt formation had occurred. The investigations using
DTA-TG reveal the thermal stability of the co-crystals, whereas
the co-crystal of cbz and ana has the lowest melting point. The
results of these investigations and the lattice constants of the
co-crystals are summarized in Table 1.
The products were characterized by using differential ther-
mal analysis and thermogravimetric analysis (DTA-TG), Raman
spectroscopy and solid-state NMR (ssNMR) spectroscopy. On
the basis of the Raman and ssNMR data, we can identify the
Figure 1. Powder XRD patterns of the new co-crystals carbamazepine:anthranilic acid 2:1 (cbz:ana) (top left), salicylic acid:anthranilic acid (sa:ana) (top right),
theophylline:anthranilic acid 2:3 (tp:ana) (bottom left), and theobromine:salicylic acid (tb:sa) (bottom right) and their corresponding reactants.
Chem. Eur. J. 2015, 21, 14969 – 14974
www.chemeurj.org
14970
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. General and crystallographic data of the tb:ana, cbz:ana, sa:ana, tp:ana, and tb:sa co-crystal.
Co-crystal
tb:ana
cbz:ana
sa:ana
tp:ana
tb:sa
formula
(C7H8N4O2):(C7H7NO2)
317.3
(C15H12N2O)2:(C7H7NO2)
609.68
triclinic
(C7H6O3):(C7H7NO2)
(C7H8N4O2)2:(C7H7NO2)3
771.45
triclinic
(C7H8N4O2):(C7H6O3)
318.28
triclinic
Mw [gmolÀ1
]
275.26
monoclinic
P21
4.38114(13)
10.91555(42)
14.47135(59)
90
109.1491(20)
90
657.76
2
crystal system
space group
a [Š]
b [Š]
c [Š]
monoclinic
P21/c
6.56078(27)
21.84702(89)
11.01679(44)
90
112.9011(19)
90
1454.61
4
¯
P1
¯
P1
¯
P1
14.74569(67)
11.08411(52)
10.86061(58)
113.9659(27)
93.7022(31)
101.1102(25)
1571.33
2
20.1369(11)
13.29925(68)
7.16109(30)
96.8582(38)
94.8506(37)
72.0511(39)
1809.07
2
6.7950(8)
7.9364(10)
14.0600(18)
94.047(4)
103.658(4)
103.467(3)
710.17
a [8]
b [8]
g [8]
V [Š3]
Z
2
method
T [K]
PXRD
293
PXRD
293
PXRD
293
PXRD
293
single crystal
100
salt formation
Tm [8C]
no
[a]
no
128
yes
137
no
144
no
[a]
[a] Stepwise thermal decomposition.
Competitive and stability experiments
C) Stability milling tests of other ana co-crystals
To understand the driving forces of mechanochemical synthe-
ses, three types of competitive milling reactions were per-
formed: A) Grinding a stoichiometric physical mixture of ana,
tb and a further API (API2); B) Grinding the tb:ana co-crystal
with a stoichiometric amount of API2; C) Grinding the ana co-
crystal of API2 with tb. All grinding experiments were conduct-
ed in a MM400 mixing mill at a frequency of 30 Hz within
25 min to ensure a thorough, uniform interparticular mixture
during the grinding process. To get more information about
the co-crystal stabilities, slurry experiments with a mixture of
a co-crystal and an API were conducted.
Furthermore, these stability tests were performed vice versa by
milling tb with powders of the ana co-crystals of cbz, sa, or tp
(Figure 3C). In this case, three different reactions were ob-
served: An exchange of the API in the final product and the re-
lease of API2 (Figure 3C, pathway 6). In the other cases, the
co-crystal remains stable and no reaction is observed (Fig-
ure 3C, pathway 8). For some combinations, the formation of
a mixture of co-crystals was observed (Figure 3C, pathway 7).
Results of milling and slurry experiments
To obtain further insights in the stability of the co-crystals,
slurry experiments were performed. The tb:ana co-crystal was
slurried with a competitive API. The co-crystals of ana with cbz,
sa, or tp were stirred with tb.
A) Competitive milling experiments
A physical mixture of tb, ana, and an additive API2 was milled
(Figure 3) either by neat grinding or liquid-assisted grinding
(LAG) using ethanol or acetonitrile. The results are schematical-
ly summarized in Figure 3A. Three different reaction pathways
were observed: 1) Formation of the co-crystal theobromine:an-
thranilic acid (tb:ana), while the added API remains unreacted;
2) All three possible two-component co-crystals are formed;
and 3) A co-crystal of the additive API2 and ana is formed. Tb
is still present in its initial crystal structure.
In reactions with cbz, tb, and ana, the cbz:ana co-crystal is
formed preferably. The milling product is formed regardless of
neat or liquid-assisted grinding conditions (Figure 4). This co-
crystal also remains stable during grinding with tb. The forma-
tion of the cbz:ana co-crystal can be explained by analyzing
the eutectic melting points. Because cbz and ana form a eutec-
tic phase with a melting point at 100.38C, which is lower than
the eutectic melting point of tb with ana at 136.28C (the Sup-
porting Information, Table S1), we can assume that the forma-
tion of the cbz:ana co-crystal is favored.
B) Stability milling tests of tb:ana
Abourahma et al. conducted grinding reactions of co-crystals
to investigate the formation of theophylline:p-hydroxybenzoic
acid co-crystals. The results were discussed based on Etter’s
rules. The rule that the best hydrogen bond donor interacts
with the best hydrogen bond acceptor was considered.[14] We
could show that Etter’s rules are important, but a prediction of
favored co-crystals is not possible based on these rules only.
As described in the following, kinetic aspects influence the
milling product and have to be considered also.
A set of mechanochemical experiments was conducted to eval-
uate the stability of a distinct co-crystal. In these experiments
powders of the tb:ana co-crystal were ground together with
either cbz, sa, or tp (Figure 3B). Two types of reaction product
were observed depending on the added API. Either the tb:ana
co-crystal remains intact (Figure 3B, pathway 4) or all three
possible co-crystals are formed next to each other (Figure 3B,
pathway 5).
Chem. Eur. J. 2015, 21, 14969 – 14974
www.chemeurj.org
14971
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Consequently, the formation of cbz:ana co-crystal would be
expected in the stability milling tests (Figure 3B) if only Etter’s
rules were important for the product formation. However, the
tb:ana co-crystal remains stable in the milling experiment with
pure cbz. Obviously, kinetic factors also influence the mecha-
nochemical reactions. As compared with the pure compounds,
the tb:ana co-crystal seems to be energetically favored. This
leads to a higher activation energy for the decomposition of
the crystalline structure in relation to the corresponding reac-
tants. The experiments using three pure reactants (cbz, ana,
tb) suggest that the activation energy for the formation of the
cbz:ana crystal is lowered because of the decomposition of
the crystalline structure of ana. From the stable tb:ana co-crys-
tal present in the reaction mixture, ana is not released and
hence not present for the formation of cbz:ana.
The second model system contains tp as competitive API2
(Figure 5). In contrast to the other systems, the product of the
competition milling reaction with tp depends on the synthesis
parameters (Figure 3A). During neat grinding the tb:ana co-
crystal is preferred, whereas the tp:ana co-crystal is favored
during liquid-assisted grinding. Therefore, it can be supposed
that the kinetic barrier for the formation of the tp:ana is
higher than for the tb:ana co-crystal. The tp:ana co-crystal is
also not formed under neat grinding conditions. During the
application of liquid-assisted grinding the activation energy is
lowered because of the formation of an intermediate liquid
phase, which is based on the higher solubility of tp compared
with that of tb (the Supporting Information, Table S2). Neither
grinding nor slurrying enables the conversion of the tb:ana
into the tp:ana co-crystal. Neat grinding of the tp:ana co-crys-
tal with tb leads to the formation of the tb:ana and pure tp.
No conversion is observed when using liquid-assisted grinding.
These results show that the intermediates of the formation
pathway are also crucial for the course of the mechanochemi-
cal reactions.
The behavior of sa as an additive in the competitive milling
reactions is different than that with tb and ana (Figure 6). Here,
all possible co-crystals (tb:ana, tb:sa, and sa:ana) are formed
both under neat and liquid-assisted grinding (pathway 2,
Figure 3). According to Etter’s rules, it is reasonable that the tb
co-crystals of ana and sa are formed in the same extent, since
both molecules show a comparable acidity. The remaining sa
co-crystalizes with ana. The grinding experiments with the mix-
tures of the tb:ana co-crystal and sa or the sa:ana and tb
reveal similar milling products. In both cases, again all possible
co-crystals were synthesized. Among tp, cbz, and sa, only sa is
able to break the hydrogen bonds of the tb:ana co-crystal
during the grinding process. The sa molecules are small
enough to interact with ana molecules in the corresponding tb
co-crystal, leading to the same product as in the competition
grinding experiments.
Figure 2. Crystal structure and bonding arrangement of the co-crystals theo-
bromine:anthranilic acid (tb:ana), carbamazepine:anthranilic acid (2:1)
(cbz:ana), salicylic acid:anthranilic acid (sa:ana), theophylline:anthranilic acid
(2:3) (tp:ana), and theobromine:salicylic acid (tb:sa). Dashed lines indicate
hydrogen bonds. All hydrogen atoms not involved in hydrogen bonds were
omitted for clarity.
It was not possible to convert the tb:ana co-crystal into the
cbz:ana co-crystal and tb by milling or slurrying with cbz. Be-
cause ana is a strong acid and cbz a stronger base than tb, the
results of the previous competitive milling reaction can also be
explained by Etter’s rules, which indicate that the strongest
acid interacts with the strongest base.[14b]
Slurry experiments using the tb:ana co-crystal with an equi-
molar amount of sa in heptane again led to a mixture of the
three possible co-crystals. On the other hand, the sa:ana co-
crystal stays intact during slurrying with tb. This result shows
that under mechanochemical conditions the energy input per
Chem. Eur. J. 2015, 21, 14969 – 14974
www.chemeurj.org
14972
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. Schematic representation of the: A) competitive, and B,C) stability milling reactions with theobromine (red), anthranilic acid (blue), and an additional
API2 (green). Different possible and observed reaction pathways are shown. The numbers indicate the different observed pathways. The other possibilities
could not be observed.
Figure 6. Results of the competitive grinding reactions, the stability grinding
reactions and the slurry experiments of salicylic acid (sa), theobromine (tb),
and anthranilic acid (ana). The numbers in brackets indicate the pathways
presented in Figure 3.
Figure 4. Results of the competition grinding reactions, the stability grinding
reactions, and the slurry experiments of carbamazepine (cbz), theobromine
(tb), and anthranilic acid (ana). The numbers in brackets indicate the path-
ways presented in Figure 3.
unit time per unit material is strikingly higher than that ob-
tained in the slurry experiments.
Conclusion
The formation of co-crystals under grinding conditions was in-
vestigated to elucidate the factors that influence the formation
of a distinct product. A set of competitive reactions were
chosen, which allowed addressing different aspects of milling
reactions. The energy input during mechanochemical reaction
is sufficient to decompose the crystalline structure of pure
compounds as well as co-crystals. One would suggest that the
formation of the reaction products under these conditions are
based on preferred intermolecular interactions. Our results
show that the co-crystal-milling product is not only triggered
by preferred intermolecular interactions but it is also signifi-
cantly influenced by the crystalline starting phase and possible
intermediates based on the synthesis condition. These results
are in good accordance to previous experiments.[9a] Therefore,
mechanochemical co-crystal reactions are more complex than
Figure 5. Results of the competitive grinding reactions, the stability grinding
reactions, and the slurry experiments of theophylline (tp), theobromine (tb),
and anthranilic acid (ana). The numbers in brackets indicate the pathways
presented in Figure 3.
Chem. Eur. J. 2015, 21, 14969 – 14974
www.chemeurj.org
14973
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Moulton, A. Mukherjee, G. Mukherjee, A. S. Myerson, V. Puri, A. Raman-
an, T. Rajamannar, C. M. Reddy, N. Rodríguez-Hornedo, R. D. Rogers,
T. N. G. Row, P. Sanphui, N. Shan, G. Shete, A. Singh, C. C. Sun, J. A.
Swift, R. Thaimattam, T. S. Thakur, R. Kumar Thaper, S. P. Thomas, S. To-
thadi, V. R. Vangala, N. Variankaval, P. Vishweshwar, D. R. Weyna, M. J. Za-
worotko, Cryst. Growth Des. 2012, 12, 2147–2152.
generally expected. For a successful synthesis, the kinetics of
the reaction have to be considered.
Experimental Section
ˇ
[4] a) D. Yan, D.-K. Bucar, A. Delori, B. Patel, G. O. Lloyd, W. Jones, X. Duan,
Materials: Anthranilic acid (ana), C7H7NO2 (98+%, Acros Organics,
Belgium), carbamazepine (cbz), C15H12N2O (99%, Acros Organics,
Belgium), salicylic acid (sa), C7H6O3 (AppliCem, pure Pb. Eur., Ger-
many), theobromine (tb), C7H8N4O2 (99%, Acros Organics, Belgium),
and theophylline (tp), C7H8N4O2 (99+%, Acros Organics, Belgium)
were purchased commercially and were used without further pu-
rification.
Chem. Eur. J. 2013, 19, 8213–8219; b) M. Morimoto, M. Irie, Chem. Eur. J.
2006, 12, 4275–4282.
[5] a) A. Alhalaweh, W. Kaialy, G. Buckton, H. Gill, A. Nokhodchi, S. P. Velaga,
Aaps Pharmscitech 2013, 14, 265–276; b) M. D. Eddleston, B. Patel,
G. M. Day, W. Jones, Cryst. Growth Des. 2013, 13, 4599–4606; c) K. See-
feldt, J. Miller, F. Alvarez-Nunez, N. Rodriguez-Hornedo, J. Pharm. Sci.
2007, 96, 1147–1158; d) T. Leyssens, G. Springuel, R. Montis, N. Candoni,
S. Veesler, Cryst. Growth Des. 2012, 12, 1520–1530; e) S. J. Nehm, B. Ro-
driguez-Spong, N. Rodriguez-Hornedo, Cryst. Growth Des. 2006, 6, 592–
600.
Synthesis of the pure co-crystals: The milling syntheses were con-
ducted by neat or liquid-assisted grinding (LAG) in a ball mill
(MM400, Retsch, Germany) at 30 Hz for 25 min. A 10 mL steel
vessel with two steel balls (10 mm diameter, 4 g) was used for
a total load of 1 g.
ˇˇ ´
[6] a) S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Friscic, F. Gre-
pioni, K. D. Harris, G. Hyett, W. Jones, Chem. Soc. Rev. 2012, 41, 413–
ˇ ´
447; b) E. Boldyreva, Chem. Soc. Rev. 2013, 42, 7719–7738; c) D. Cincic,
ˇˇ ´
T. Friscic, W. Jones, Chem. Eur. J. 2008, 14, 747–753.
Competitive milling synthesis: The competitive milling syntheses
were conducted by neat grinding or LAG of stoichiometric
amounts of tb, ana and an additional API (cbz, sa, tp) in a ball mill
(MM400, Retsch, Germany) at 30 Hz for 25 min. A 10 mL steel
vessel with two steel balls (10 mm diameter, 4 g) was used for
a total load of 1 g. For the LAG experiments, 250 mL of ethanol or
acetonitrile were added to the reaction mixture.
[7] a) D. Braga, L. Maini, M. Polito, F. Grepioni, Chem. Commun. 2002,
2302–2303; b) S. Karki, T. Friscic, W. Jones, CrystEngComm 2009, 11,
470–481; c) V. R. Pedireddi, W. Jones, A. P. Chorlton, R. Docherty, Chem.
Commun. 1996, 987–988.
[8] a) A. A. L. Michalchuk, I. A. Tumanov, V. A. Drebushchak, E. V. Boldyreva,
Faraday Discuss. 2014, 170, 311 –335; b) I. A. Tumanov, A. F. Achkasov,
S. A. Myz, E. V. Boldyreva, V. V. Boldyrev, Doklady Chem. 2014, 457, 154–
159; c) A. A. L. Michalchuk, I. A. Tumanov, E. V. Boldyreva, CrystEngComm
2013, 15, 6403–6412; d) I. A. Tumanov, A. F. Achkasov, E. V. Boldyreva,
Stability milling synthesis: The stability milling syntheses were
conducted by neat grinding or LAG of stoichiometric amounts of
an existing co-crystal and a competitive API (API2) in a ball mill
(MM400, Retsch, Germany) at 30 Hz for 25 min. A 10 mL steel
vessel with two steel balls (10 mm diameter, 4 g) was used for
a total load of 1 g. For the LAG experiments, 250 mL of ethanol or
acetonitrile were added to the reaction mixture.
ˇˇ ´
V. V. Boldyrev, Russian J. Phys. Chem. A 2012, 86, 1014–1017; e) T. Friscic,
L. Fabian, J. C. Burley, W. Jones, W. D. S. Motherwell, Chem. Commun.
2006, 5009–5011; f) T. Friscic, L. Fabian, J. C. Burley, D. G. Reid, M. J.
Duer, W. Jones, Chem. Commun. 2008, 1644–1646.
ˇˇ ´
[9] a) M. R. Caira, L. R. Nassimbeni, A. F. Wildervanck, J. Chem. Soc. Perkin
Trans. 2 1995, 2213–2216; b) M. C. Etter, S. M. Reutzel, C. G. Choo, J.
Am. Chem. Soc. 1993, 115, 4411–4412.
[10] a) J. B. Arlin, L. S. Price, S. L. Price, A. J. Florence, Chem. Commun. 2011,
47, 7074–7076; b) S. L. Price, Chem. Soc. Rev. 2014, 43, 2098–2111; c) N.
Issa, P. G. Karamertzanis, G. W. A. Welch, S. L. Price, Cryst. Growth Des.
2009, 9, 442–453; d) A. J. Cruz-Cabeza, G. M. Day, W. Jones, Chem. Eur.
J. 2008, 14, 8830–8836.
Slurry experiments: Slurry experiments were conducted by stirring
a slurry of a stoichiometric mixture of an existing co-crystal and
a competitive API2 in heptane for seven days under ambient con-
ditions. The solid phase was gained by filtration and analyzed by
powder X-ray diffraction (PXRD).
ˇˇ ´
[11] a) T. Friscic, I. Halasz, P. J. Beldon, A. M. Belenguer, F. Adams, S. A. J.
Kimber, V. Honkimäki, R. E. Dinnebier, Nat. Chem. 2013, 5, 66–73; b) I.
Acknowledgements
ˇ
´
Halasz, A. Puskaric, S. A. J. Kimber, P. J. Beldon, A. M. Belenguer, F.
ˇ
ˇ
Adams, V. Honkimäki, R. E. Dinnebier, B. Patel, W. Jones, V. Strukil, T. Fris-
ˇ ´
cic, Angew. Chem. Int. Ed. 2013, 52, 11538–11541; Angew. Chem. 2013,
We are grateful to S. Reinsch for DTA-TG measurements, S.
Benemann and E. Ortel for SEM images, and C. Jäger for NMR
spectroscopic investigations.
125, 11752–11755; c) F. Fischer, G. Scholz, L. Batzdorf, M. Wilke, F. Em-
merling, CrystEngComm 2015, 17, 824–829; d) L. Batzdorf, F. Fischer, M.
Wilke, K. J. Wenzel, F. Emmerling, Angew. Chem. Int. Ed. 2015, 54, 1799–
1802; Angew. Chem. 2015, 127, 1819–1822.
[12] a) T. R. Shattock, K. K. Arora, P. Vishweshwar, M. J. Zaworotko, Cryst.
Growth Des. 2008, 8, 4533–4545; b) P. Kavuru, D. Aboarayes, K. K. Arora,
H. D. Clarke, A. Kennedy, L. Marshall, T. T. Ong, J. Perman, T. Pujari, L.
Wojtas, M. J. Zaworotko, Cryst. Growth Des. 2010, 10, 3568–3584;
c) J. A. Bis, P. Vishweshwar, D. Weyna, M. J. Zaworotko, Mol. Pharm.
2007, 4, 401–416; d) A. M. Moragues-Bartolome, W. Jones, A. J. Cruz-
Cabeza, CrystEngComm 2012, 14, 2552–2559; e) C. B. Aakerçy, K. Epa, S.
Forbes, N. Schultheiss, J. Desper, Chem. Eur. J. 2013, 19, 14998–15003.
[13] a) S. Gunasekaran, G. Sankari, S. Ponnusamy, Spectrochim. Acta Part A
2005, 61, 117–127; b) S. Suresh, S. Gunasekaran, S. Srinivasan, Spectro-
chim. Acta Part B 2014, 132, 130–141.
Keywords: crystal engineering · kinetics · mechanochemistry ·
structure determination · X-ray diffraction
[1] a) R. Thakuria, A. Delori, W. Jones, M. P. Lipert, L. Roy, N. Rodriguez-Hor-
nedo, Int. J. Pharm. 2013, 453, 101–125; b) D. P. Elder, R. Holm, H. L.
de Diego, Int. J. Pharm. 2013, 453, 88–100; c) A. S. Sinha, A. R. Maguire,
S. E. Lawrence, Cryst. Growth Des. 2015, 15, 984–1009; d) N. Schultheiss,
A. Newman, Cryst. Growth Des. 2009, 9, 2950–2967; e) T. Friscic, W.
Jones, Cryst. Growth Des. 2009, 9, 1621–1637.
[2] C. B. Aakerçy, D. J. Salmon, CrystEngComm 2005, 7, 439–448.
[3] a) C. B. Aakerçy, M. E. Fasulo, J. Desper, Mol. Pharm. 2007, 4, 317–322;
[14] a) H. Abourahma, J. M. Urban, N. Morozowich, B. Chan, CrystEngComm
2012, 14, 6163–6169; b) M. C. Etter, Acc. Chem. Res. 1990, 23, 120–126.
ˇ
b) M. C. Etter, G. M. Frankenbach, Chem. Mater. 1989, 1, 10–12; c) T. Fris-
ˇ ´
cic, W. Jones, J. Pharm. Pharmacol. 2010, 62, 1547–1559; d) M. J. Zawor-
otko, Cryst. Growth Des. 2007, 7, 4–9; e) S. Aitipamula, R. Banerjee, A. K.
Bansal, K. Biradha, M. L. Cheney, A. R. Choudhury, G. R. Desiraju, A. G. Di-
kundwar, R. Dubey, N. Duggirala, P. P. Ghogale, S. Ghosh, P. K. Goswami,
N. R. Goud, R. R. K. R. Jetti, P. Karpinski, P. Kaushik, D. Kumar, V. Kumar, B.
Received: March 8, 2015
Revised: July 1, 2015
Published online on September 1, 2015
Chem. Eur. J. 2015, 21, 14969 – 14974
www.chemeurj.org
14974
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim