Mechanochemical reaction of sulfathiazole with carboxylic acids: Formation of a cocrystal, a salt, and coamorphous solids

Article abstract of DOI:10.1021/cg401673z

The anhydrous solvent-free mechanochemical reaction of sulfathiazole, STZ, polymorphs I, III, and V with 10 carboxylic acids was monitored by powder X-ray diffraction (PXRD), attenuated total reflectance infrared (ATR-IR), and near-infrared (NIR) spectroscopy. A 1:1 cocrystal was observed with glutaric acid and the strongest acid, oxalic acid, gave a 1:1 salt. A principal components analysis of the glutaric acid NIR data showed that forms I and V proceeded to the cocrystal, but that form III transformed to form IV before cocrystal formation. The oxalic acid salt was formed via complete amorphization. The crystal structures of the cocrystal and the salt were determined. Sulfathiazole comilled with l-tartaric and citric acids gave coamorphous systems that were stable at 10% RH for up to 28 days. Comilling sulfathiazole with dl-malic acid gave mixtures of form V and the amorphous form.

Full text of DOI:10.1021/cg401673z

Subscriber access provided by UNIV OF CONNECTICUT
Article
Mechanochemical Reaction of Sulfathiazole with Carboxylic
Acids: Formation of a Cocrystal, a Salt and Coamorphous Solids.
Yun Hu, Katarzyna Gniado, Andrea Erxleben, and Patrick McArdle
Cryst. Growth Des., Just Accepted Manuscript • Publication Date (Web): 20 Dec 2013
Downloaded from http://pubs.acs.org on January 1, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Crystal Growth & Design is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Mechanochemical Reaction of Sulfathiazole with Carboxylic
Acids: Formation of a Cocrystal, a Salt and Coamorphous Solids.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yun Hu, Katarzyna Gniado, Andrea Erxleben* and Patrick McArdle *
School of Chemistry, National University of Ireland, Galway, University Road, Galway,
Ireland
* Corresponding author. Tel.: +353 91 49 2487; Fax: +353 91 525700.
Email address: p.mcardle@nuigalway.ie (P. McArdle).
andrea.erxleben@nuigalway.ie (A. Erxleben)
Abstract
The anhydrous solvent free mechanochemical reaction of sulfathiazole, STZ,
polymorphs I, III and V with ten carboxylic acids was monitored by powder Xꢀray
1
ACS Paragon Plus Environment

Crystal Growth & Design
Page 2 of 40
1
2
3
4
5
6
7
8
9
diffraction (PXRD), attenuated total reflectance infrared (ATRꢀIR) and nearꢀinfrared
(NIR) spectroscopy. A 1:1 cocrystal was observed with glutaric acid and the strongest
acid, oxalic acid gave a 1:1 salt. A PCA analysis of the glutaric acid NIR data showed
that forms I and V proceeded to the cocrystal but that form III transformed to form IV
before cocrystal formation. The oxalic acid salt was formed via complete amorphization.
The crystal structures of the cocrystal and the salt were determined. Sulfathiazole
comilled with Lꢀtartaric and citric acids gave coamorphous systems that were stable at 10%
RH for up to 28 days. Comilling sulfathiazole with DLꢀmalic acid gave mixtures of form
V and the amorphous form.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords:
Sulfathiazole; Cocrystal; Amorphous; Stability; NIR spectroscopy; Powder Xꢀray
diffraction; ATRꢀIR spectroscopy
1. Introduction
Solubility and dissolution rates are important factors in determining the efficacy as
well as activity of pharmaceutical solids.1, 2 Thus, the discovery of methods which will
improve the solubility and dissolution rate of active pharmaceutical ingredients (APIs) is
an important challenge in the pharmaceutical industry. Some of the methods currently
being examined include the discovery of new polymorphs, amorphous forms, solid
dispersions, cocrystals, salt forms and inclusion complexes.1 Cocrystals and salt forms in
particular, have recently attracted significant interest in the pharmaceutical industry
because of their potential to improve the physicochemical properties of APIs. The wide
2
ACS Paragon Plus Environment

Page 3 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
range of GRAS3 coformer properties and the multiple possible interactions in the solid
state provide many strategies for the control of cocrystal and salt properties, including,
enhanced dissolution rates, thermal stability and improved mechanical properties.4ꢀ6
Alternatively, transferring a crystalline drug into an amorphous form can also increase
dissolution rate and apparent solubility. However, the physical stability of amorphous
forms during manufacture or storage is a problem, and it is necessary to find ways to
stabilize the amorphous state. Recently, the concept of coamorphous systems has been
introduced as a potential solution to this problem.7ꢀ9 In coamorphous systems, a
combination of two complimentary drugs or a drug and a suitable GRAS small molecule
is used instead of drugꢀpolymer combinations.
Many methods and technologies have been described for the preparation of
cocrystals. Preparation from solution may be the only suitable method where large single
crystals are required for structure determination. However, this approach is not efficient
for cocrystal screening due to the time consuming solvent selection process, the potential
for multi component crystallization and solvate formation.6 As an alternative solidꢀstate
milling methods, including neat milling and solvent drop milling, have been explored as a
more efficient means of cocrystal synthesis.5, 10ꢀ15 Comilling with additives has also been
employed in an attempt to control solidꢀstate phase transformation, and to both stabilize
and suppress the generation of amorphous forms.8, 9, 16ꢀ19
Sulfathiazole (STZ) (Figure 1) is a model drug which is known to have at least five
polymorphs, forms over 100 solvates and numerous adducts and has more than 6000 STZ
references in SciFinder.20ꢀ22 STZ was also part of the first pharmaceutical application of a
cocrystal when the 1:1 cocrystal of sulfathiazole and proflavine, branded as flavazole,
was used as an antibacterial agent during the Second World War.23 Thus, it has been
extensively investigated as a model system in polymorphism research, and as a model
cocrystal former.20, 21, 24ꢀ26 There are some inconsistencies in the nomenclature of
sulfathiazole forms in the literature and in this study the numbering of the polymorphic
forms FI−FV is in accordance with that used by the Cambridge Crystallographic Data
Centre (CCDC).27
In this study, we extend our previous research on the mechanochemistry of STZ
with an investigation of the reaction of STZ with a series of carboxylic acids, half of
which are on the GRAS list, in an effort to understand the factors which control cocrystal,
salt and coamorphous formation. The molecular structures of STZ and the carboxylic
acids are given in Figure 1.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

Crystal Growth & Design
Page 4 of 40
1
2
3
4
5
6
7
8
9
S
O
O
HO
O
O
S
N
N
H
OH
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
oxalic acid
OH
H2N
sulfathiazole
O
HO
OH
O
OH
O
O
O
O
OH
tartaric acid*
HO
OH
OH
OH
citric acid*
HO
O
O
O
fumaric acid
( ) n
HO
OH
O
OH
OH
n = 1, 2, 3, 4 and 5
HO
malonic, succinic*, glutaric,
adipic* and pimelic acids.
O
malic acid*
Figure 1: Chemical structures of sulfathiazole and the carboxylic acids used. GRAS acids are
denoted by a *.
2. Experimental
2.1. Materials
Sulfathiazole was purchased from SigmaꢀAldrich with a purity of 98%. HPLC grade
dichloromethane and spectrophotometric grade acetone, were also obtained from
SigmaꢀAldrich. Oxalic acid, oxalic acid dihydrate, malonic acid, succinic acid, glutaric
acid, adipic acid, pimelic acid, DLꢀmalic acid, Lꢀtartaric acid, anhydrous citric acid, and
fumaric acid were purchased from SigmaꢀAldrich.
2.2 Methods
2.2.1 Preparation of sulfathiazole polymorphs and physical mixtures
21
Commercial sulfathiazole was determined to be FIII and it was used as received.
Sulfathiazole FI was prepared by heating commercial sulfathiazole samples in an oven
for 30 min at 180 ºC as described in a previous study.21 Crystals of FV were obtained by
evaporating a boiling aqueous solution of sulfathiazole to dryness, followed by drying at
105 ºC.20 A physical mixture of sulfathiazole and each carboxylic acid was prepared by
gentle mixing of both at a molar ratio of 1:1 in an agate mortar with pestle.
4
ACS Paragon Plus Environment

Page 5 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
2.2.2 Milling experiments
Milling experiments were performed using an oscillatory ball mill (Mixer Mill
MM400, Retsch GmbH & Co., Germany) using a 25 mL stainless steel milling jar
containing one 15 mm diameter stainless steel ball at 25 Hz. Each experiment was carried
out with a fresh 0.5 g sample. Sulfathiazole samples were comilled with the carboxylic
acids in a 1:1 molar ratio for 7.5, 15, 17, 20, 25, 30, 45, 60, and 90 min. The resulting
milled powder samples were analyzed immediately by PXRD, IR and NIR spectroscopy.
In order to avoid overꢀheating the samples, pause periods of 5 min were made after every
30 min of milling. The maximum temperature recorded on the outside of the milling jars,
during milling, using a digital thermometer was 37 °C.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2.3 Preparation of sulfathiazole cocrystal and salt
The cocrystal and salt powders were prepared by grinding stoichiometric amounts of
solid STZ with glutaric and oxalic acids for 60 min. Single crystals were grown by slow
evaporation of solutions of these powders from dichloromethane and acetone
respectively.
2.2.4 Storage conditions
The stability of the coamorphous and cocrystal and salt samples was studied at
o
ambient temperature (22 ±2 C) under different relative humidity (RH) conditions
ranging from 10 to 75% RH. RH values of 10, 43, and 75 % were achieved in desiccators
sealed using silica gel and using solutions of the salts K2CO3, NaCl, and KCl.28 Repeated
measurements suggest that RH values were maintained at ±3% of the stated values.
Stored samples were analyzed after 7 and 28 days, and two months by NIR spectroscopy
and PXRD.
2.3 Analytical techniques
2.3.1 Single Crystal X-ray Diffraction
An Oxford Diffraction Xcalibur system was used to collect Xꢀray diffraction data at
150 K. The crystal structures were solved by direct methods (Shelxsꢀ97) and refined by
full matrix leastꢀsquares using Shelxlꢀ 97 within the Oscail package.29, 30
2.3.2 X-ray Powder Diffraction (XRPD)
Xꢀray powder diffraction data were collected on a Siemens D500 powder
diffractometer which was fitted with a diffracted beam monochromator. Diffraction
patterns were recorded between 5 and 40o (2θ) using CuKα radiation with steps of 0.05o
with a 2 second counting time per step. The Oscail software package was used to
generate theoretical PXRD patterns of the STZ forms.31
2.3.3 Thermal Analysis
Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC)
were performed on a STA625 thermal analyzer from Rheometric Scientific. The samples
5
ACS Paragon Plus Environment

Crystal Growth & Design
Page 6 of 40
1
2
3
4
5
6
7
8
were scanned at 10 oC/min from 20 to 300 o C under a nitrogen purge of 50 mL/min in
open aluminum crucibles. The system was calibrated using an indium standard.
9
2.3.4 Attenuated Total Reflectance-Infrared Spectroscopy
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ATRꢀIR spectra were recorded from 4000 to 650 cmꢀ1 using a PerkinElmer
Spectrum 400 (FTꢀIR/FTꢀNIR Spectrometer) with 32 accumulations at a resolution of 4
cmꢀ1. This instrument was equipped with a DATR 1 bounce Diamond/ZnSe Universal
ATR sampling accessory.
2.3.5 Near-Infrared Spectroscopy
NIR spectra were collected in glass vials (15 mm × 45 mm) on a PerkinElmer
Spectrum One fitted with an NIR reflectance attachment. Spectra were collected using
interleaved scans in the 10000 – 4000 cmꢀ1 range with a resolution of 8 cmꢀ1, using 32
coadded scans.
2.3.6 Data Analysis
Data analysis was carried out using the multivariate data analysis software The
Unscrambler v.9.8 (Camo, Norway). Principal components analysis (PCA) was used to
investigate the spectral variation during the comilling process and stability of comilled
samples under various RH conditions. Standard normal variate (SNV) and second
derivative preꢀprocessing methods were sequentially applied to NIR data for PCA
analysis. SavitzkiꢀGolay second derivative calculations were performed with a window
size of 11 points and a second order polynomial
.
2.3.7 Hydrogen bonded adduct modelling
The molecules were docked using the molecular modelling functionality, Moilin,
within the Oscail software package.32 DFT calculations were carried out using the Firefly
quantum chemistry package,33 which is partially based on the GAMESS (US) source
code34.
3. Results and Discussion
3.1 Comilling experiments
The main differences between the STZ polymorphs lie in hydrogen bonding and
packing of the molecules in the crystal lattice.35 In an analysis of the crystal structures of
the five forms using the XPac program Gelbricht et al. 36 have shown that the low density
forms FI and FV have different structures that are unrelated to those of FII, FIII and FIV.
The latter three forms have structures based on the same monoꢀlayer and forms II and IV
have biꢀlayers which differ in a slip of the monoꢀlayers while FIII is more complex in that
it consists of the biꢀlayers of both FII and FIV. Consequently, these three polymorphs are
very similar in their spectral and physicochemical behavior and the properties of FIII are
to a large extent the average of the superposition of the properties of FII and FIV.21
6
ACS Paragon Plus Environment

Page 7 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Therefore, to maximize the effectiveness of monitoring the milling of STZ polymorphs in
the presence of a range of carboxylic acids forms I, III and V were used and the
experimental results are summarized in Table 1.
As is clear from the Table 1 four different results have been observed: cocrystal
formation with glutaric acid (GA); salt formation with oxalic acid (OA); stable
coamorphous generation with citric (CA) and Lꢀtartaric (TA) acids and STZ polymorph
transformations with the six other acids. It was also observed that cocrystal, salt and
coamorphous formation were dependent on the carboxylic acid used but not on the
starting STZ polymorph whereas the polymorph transformations did show some variation
with the starting polymorph.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.1.1 Polymorph transformations
The simplest result was observed in the presence of fumaric acid where all three
starting forms gave mixtures of FI and the amorphous form (FA). This result is identical
to that observed for neat milling of these three forms.37 This may be because fumaric acid
has by far the highest melting point of any of the acids used and is the least likely to act
like a solvent. Succinic acid with the second highest melting point gave results closest to
those of fumaric acid the only difference being that samples that started with form V (FV)
contained FV. Of the other transformations observed DLꢀmalic is perhaps the most
interesting as milling all three forms led to no change when starting from FV or to FV/FA
mixtures in the other two cases. The only reported preparative methods for STZ FV
involve crystallization either from boiling water at 100 or from nꢀpropanol at 95 °C.24
The temperature of the outside of the milling jars reached a maximum of 37 °C during
milling and it is possible that a combination of temperature and the extra hydroxyl group
provide hydroxylic solvent like conditions which promote the generation of FV. Malonic
acid gives results that are closest to those of DLꢀmalic acid but here some FI was
observed in samples which started with FI and FIII. Adipic and pimelic acids have
closely related structures and the results differ only in the amount of FIII which remained
in samples which started with FIII.
3.2 Sulfathiazole glutaric acid cocrystal
When STZ and GA (βꢀpolymorph) were comilled a new diffraction pattern was
observed, Figure 2a. It was possible to grow the tabloid shaped crystals of this STZꢀGA
cocrystal from DCM solution that are shown in Figure 2b. The crystal structure of the
cocrystal was determined and the crystal data are given in Table 2. The hydrogen
bonding in the crystal structure is shown in Figure 3 and hydrogen bonding distances and
angles are in Table 3. In the crystal structure both the STZ and GA molecules form
dimers via ꢀꢁꢁꢂ8ꢃ rings.38 The GA dimers are linked via a bifurcated hydrogen bond to
N3 and O2 of two adjacent STZ molecules and there is also a hydrogen bond between
N3ꢀH3B and the O1 of an adjacent STZ molecule. The experimental PXRD pattern was
an excellent match to the pattern calculated using the atom coordinates, Figure 2a.
7
ACS Paragon Plus Environment

Crystal Growth & Design
Page 8 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 (a) PXRD patterns for Glutaric acid, STZ FI, STZꢀGA cocrystal experimental and
STAꢀGA cocrystal calculated, (b) Morphology of the STZꢀGA cocrystals.
Figure 3 Hydrogen bonding in the STZꢀGA cocrystal structure. H atoms not involved in hydrogen
bonding have been omitted for clarity.
(a)
(b)
(c)
(1)
(2)
X 3
(1)
(2)
(3
)
(1)
(2)
(3)
(3)
0
50
100
150
200
250 3600 3200 2800
1750 1500 1250 1000 750
Wavenumber (cm-1)
7000 6500 6000
5000
4500
4000
Temperature (o)
Wavenumber (cm-1)
Figure 4 (a) DSC graphs, (b) IR spectra and (c) NIR spectra of (1) STZ FI, (2) GA and (3) the
8
ACS Paragon Plus Environment

Page 9 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
STZꢀGA cocrystal.
8
9
Figure 4a shows the DSC curves of STZ FI, βꢀglutaric acid, and the cocrystal. The
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
melting point of FI is around 202 C. There are two endothermc peaks in the curve for
pure glutaric acid, the first and shallower one at 75 oC is the transition peak of β form to
α form, and the second and deeper one at 97 oC is due to melting of αꢀglutaric acid.39 The
o
endothermic peak at 120 C in the cocrystal curve indicates a decomposition of the
cocrystal. The IR spectra of sulfathiazole FI, glutaric acid and the cocrystal are compared
in Figure 4b. Particular differences between the spectra of the cocrystal and of pure
sulfathiazole are found in the range 3500 to 3300 cmꢀ1, corresponding to the N–H
stretching vibrations. Four bands at 3481, 3399, 3385 and 3326 cm−1 are observed in IR
spectra of the cocrystal compared two IR bands for STZ FI at 3464 and 3358 cm−1. In a
similar way, the largest spectral difference in the NIR is in the 6980ꢀ6400 cm−1 range
(Figure 4c).
Principle component analysis (PCA) was used to give a better insight into the
solidꢀstate STZꢀGA transformation and to visualize the different rates of cocrystal
formation. The PCA scores and the corresponding loadings plots for the NIR spectra of
the milled samples in the regions 6980–5800 and 5130–
4000 cmꢀ1 are shown in Figure 5.
In Figure 5a PM is used to indicate a physical mixture before milling started. A three
component PCA model was used to explain 96.6% of the variance observed in the NIR
data. Overall, the differences in the solidꢀstate behavior during comilling of the three STZ
forms with GA were seen in the score plot. PC1 explained 57.4% of the total variance
because the NIR spectra of the cocrystal exhibits unique features that are not present in
the NIR spectra of pure STZ forms or GA and vice versa. This is illustrated by the
positive bands in the PC 1 loadings. PC2 accounted for 25.0% of the total variance and it
describes the variation between the STZ FIIIꢀGA PM and STZ FV–GA PM, the positive
bands in the loadings at 6138 and 5046 cmꢀ1, and the negative bands at 6126 and 4896
cmꢀ1 could be from STZ FVꢀGA PM and STZ FIIIꢀGA PM, respectively. PC3 explained
14.2% of the variation, and differentiated samples containing STZ FIꢀGA PM from the
other two, as indicated by the bands at 6150, 5082, and 4184 cmꢀ1 in the corresponding
loadings plot. Figure 5a illustrates the overall process of cocrystal formation in a single
plot. The rate at which the scores travel toward the STZꢀGA cocrystal region depends on
the sulfathiazole starting material. It can be seen that the STZ FIIIꢀGA PM move toward
the cocrystal region takes the longest time (30 min).15
9
ACS Paragon Plus Environment

Crystal Growth & Design
Page 10 of 40
1
2
3
4
5
6
7
8
9
(a)
(b)
FI
4184
0.2
0.1
FIII
FV
5082
0.08
0.06
0.04
0.02
0
STZ FIII -GA PM
6150
PC3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.0
STZ FV-GA PM
-0.1
-0.2
6126
6138
4172
0.2
0.1
30 min
5046
20 min
PC2
PC1
-0.02
-0.04
-0.06
-0.08
0.0
20 min
STZ-GA cocrystal
-0.1
-0.2
4896
6126
6138
5100
5022
0.2
0.1
6696
6904
4554
-0.1
0.2
6592
STZ FI-GA PM
0.0
0.1
P
C
-0.1
-0.2
0
0.05
2
5042
(
2
0
5
.
0
4176
-0.1
%
-0.05
)
-0.2
-0.1
7000 6500 6000
5000 4750 4500 4250 4000
Wavenumber (cm-1)
Figure 5 PCA scores plots (a) and loading plots (b) of the NIR spectra of STZꢀGA cocrystal
formation prepared by comilling. The arrows represent the direction of cocrystal formation, and
PM denotes physical mixtures.
Also initially STZ FIIIꢀGA PM moves along PC3 before heading in the cocrystal
direction. This is because STZ FIII, unlike FI and FV transforms to STZ FIV before it
forms the cocrystal.
The possibility of an intermediate polymorphic conversion, or amorphous phase
formation, prior to the cocrystal formation was further tested by using a combination of
PXRD, NIR and IR spectra. It is known that the shift or appearance/disappearance of new
IR and NIR bands can suggest the occurrence of a polymorphic transformation or the
formation of an amorphous phase. 40 In the STZ FIꢀGA mixture, the NIR bands shift and
become broader, which indicated the formation of small amounts of amorphous content
prior to cocrystal formation. However, the cocrystal formation seems direct in the STZ
FVꢀGA mixture during milling as detected by IR spectra. The STZ FIIIꢀGA mixture is
more complicated, Figure 6. Firstly, there is an increase in the intensity of the NIR band
at 6126 cmꢀ1 which suggests an STZ FIII to FIV conversion. This result is in agreement
with our previous study.37 The appearance of new NIR bands around 6880 and 5080 cmꢀ1
also suggested that the formation of some STZ FI and FA prior to cocrystal formation.
The physical stability of the STZꢀGA cocrystal samples obtained from the three
starting STZ polymorphs was investigated, and the results showed that they were stable
under RH in the 10ꢀ75% range at room temperature at least for 2 months.
10
ACS Paragon Plus Environment

Page 11 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
5080
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
FIV
6126
6880
7000 6800 6600 6400 6200
5100 4950 4800 4650
Wavenumber (cm-1)
Figure 6 NIR spectra of STZ FIII comilled with GA for various times. The plots from bottom to
top: STZ FIIIꢀGA PM, milled for 7.5, 15, 20 and 25 min, STZ FI, and STZ FA (amorphous
sulfathiazole obtained from cryoꢀmilling)
3.3 Sulfathiaole oxalic acid salt
Milling STZ with oxalic acid also gave a new crystalline solid, as evidenced by a
PXRD pattern that did not contain any reflections belonging to the starting materials
(Figure 7). Crystals of this new solid were grown from acetone solution and the crystal
structure was determined. Crystal data are given in Table 2 and the hydrogen bonding
network is shown in figure 8 and hydrogen bond lengths and angles are given in Table 3.
This compound is a salt. Proton transfer from OA to the STZ NH2 has taken place. The
oxalate anions form hydrogen bonded dimers via ꢀꢁꢁꢂ10ꢃ rings which bridge adjacent
STZ molecules. The experimental PXRD pattern corresponds to the pattern calculated
from the atom coordinates Figure 7. Figure 9a shows the DSC curves of STZ FI, OA, and
o
the salt. It can be seen that the STZꢀOA salt has a melting point around 190 C. The IR
spectra are given in Figure 9b with the spectrum of STZ FIꢀOA PM milled for 15 min for
comparison. Two IR bands at 3464 and 3366 cmꢀ1 were ascribed to amorphous
sulfathiazole, and they disappeared when the salt formed. It is suggested that there is no
molecular interaction between amorphous sulfathiazole and oxalic acid in this
coamophous system. Furthermore, there is a new sharper IR band at 3121 cmꢀ1 which
appears on salt formation.
11
ACS Paragon Plus Environment

Crystal Growth & Design
Page 12 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7 Calculated and experimental PXRD patterns of STZꢀOA salt.
Figure 8 Hydrogen bonding in the STZꢀOA salt. H atoms not involved in Hꢀbonding have
been omitted for clarity.
12
ACS Paragon Plus Environment

Page 13 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
(a)
(b)
8
9
(1)
(2)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
X 3
(1)
(2)
(3
)
(3)
(4)
3600 3200 2800
1750 1500 1250 1000 750
Wavenumber (cm-1)
0
50
100
150
200
250
Temperature (o)
Figure 9 (a) DSC graphs and (b) IR spectra of (1) STZ FI, (2) OA and (3) STZꢀOA salt, (4) IR
spectrum of STZ FIꢀOA PM milled for 15 min.
Figure 10 shows NIR monitoring of the formation of the STZꢀOA salt with milling
time. It is clear that the salt forms from each of the starting polymorphs but unlike the
STZꢀGA cocrystal formation a complete coamorphous STZꢀOA system is first formed, as
evidenced by the PXRD patterns shown in the figure inserts (Figure 10). The salt formed
with extended milling after the system was amorphous. Milling STZ FIIIꢀOA required
the longest time (60 min).
The physical stability of the STZꢀOA salt was also investigated and the results
showed that it was stable at 10% RH and room temperature at least for 28 days. When
RH value is above 43%, NIR spectra clearly indicated that the salt took up water.
The formation of a salt by proton transfer as opposed to a cocrystal of the same
stoichiometry has been studied by several researchers. In general it has been suggested
that a minimum pKa difference of 3 units is required for salt formation.41 However, while
pKa values measured in solution may not be directly relevant to transformations which
take place under anhydrous solvent free conditions the pKa difference for STAꢀOA is
close to 6 units. In general it does appear, probably due to the charge separation involved,
that salt formation is a more difficult process than cocrystal formation. For example it has
been reported that mortar and pestel ground physical mixtures of pimelic acid and a
ferrocene base did not react until they were exposed to solvent vapour and that
nonꢀhydroxylic solvents gave a cocrystal whereas hydroxylic solvents gave a salt.42 There
is a report that milling of the bases trimethoprim and pyrimethamine with carboxylic
acids, without PXRD monitoring, gave salts under anhydrous solvent free conditions in
40% of the cases studied.43 However in every one of the cases that gave a salt, in the
absence of solvent, the melting point of the carboxylic acid was less than 100 °C. It is
possible that actual temperature inside the mill could have given these acids sufficient
mobility to allow them to act in a solvent like manner. Interestingly, it was found that that
13
ACS Paragon Plus Environment

Crystal Growth & Design
Page 14 of 40
1
2
3
4
5
6
7
8
9
when the milling of benzyladenine with maleic acid was monitored by PXRD that that
system went amorphous before the diffraction pattern of a salt appeared just as described
here.44 It has also been reported that milling glycine/anhydrous oxalic acid mixtures gave
no reaction but that milling glycine/oxalic acid dihydrate mixtures gave a salt.45 We have
found that milling STZ with OA dihydrate gives a new PXRD pattern, Figure 11, which
is different from that of either of the starting materials or the STZꢀOA salt. It was not
possible to obtain single crystals of this compound and its composition or structure has
not been established.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In summary it appears that solid state anhydrous salt formation by proton transfer is
a higher energy process than cocrystal formation and that in the absence of a hydroxylic
catalyst in the form of a solvent or a relatively low melting acid salt formation is difficult.
(a)
(b)
(c)
5
10 15 20 25 30 35 40
5
10 15 20 25 30 35 40
5
10 15 20 25 30 35 40
60min
30 min
15 min
30 min
15 min
30 min
15 min
STZ FI-OA PM
STZ FIII-OA PM
STZ FV-OA PM
6750
6000 5000
4500
4000 6750
6000 5000
4500
4000 6750
6000 5000
4500
4000
Wavenumber (cm-1)
Figure 10 NIR spectra of sulfathiazole polymorphs (a) STZ FI, (b) STZ FIII, and (c) STZ FV
comilled with OA, respectively, at various milling times. The insert figure represents the PXRD
patterns of STZ polymorphs comilled with OA for 15 min, PM denotes physical mixtures.
14
ACS Paragon Plus Environment

Page 15 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11 PXRD pattern observed when sulfathiazole was milled with oxalic acid dihydrate.
3.4 Milling sulfathiazole with Lꢀtartaric and citric acids
3.4.1 Stability of comilled amorphous samples
It has been mentioned above that a combination of two drugs or a drug and a small
molecule could form a single phase amorphous system of enhanced stability. In this study,
comilling STZ with Lꢀtartaric acid (TA) and citric acid (CA) in a 1:1 mole ratio for 60
and 90 min was found to give samples which were nonꢀcrystalline as determined by
PXRD. The importance and nature of the effect of moisture on amorphous materials is
especially important when formulating amorphous products. Water can act as a
plasticizer which increases molecular mobility and recrystallization rate.46 The samples
were stored at 10% RH for 28 days and at 43 and 75% RH for 7 days. NIR spectra were
used to monitor the samples using PCA analysis. To reduce the interference due to the
NIR bands of water, a combination of the 6980–5800 and 5130–4000 cmꢀ1 regions was
chosen for PCA analysis. Figure 12 shows the PCA scores scatter plots obtained using the
first and second PCs of the amorphous samples stored under different humidity
conditions. Overall, the PCA model showed clustering of the samples based on relative
humidity.
For STZꢀTA systems (Figure 12a), the first two principal components summarized
83.6% of the variation in the spectral information. The differences arose from FA with
other polymorphic forms (FII, FIII and FIV) along PC1, and the second component, PC2,
accounted for variability of amorphous and FI and FV. At 10% RH, the amorphous
samples remain amorphous up to 28 days irrespective of the starting sulfathiazole
polymorph used (Figure 12c (1)). Similarly, at 75% RH, the FA recrystallized mostly to
FIV with a small amount of FII and FIII within 7 days irrespective of the starting STZ
polymorph used (Figure 12c (6)). However, at 43% RH, the STZ polymorphs which
15
ACS Paragon Plus Environment

Crystal Growth & Design
Page 16 of 40
1
2
3
4
5
6
7
8
9
recrystallized were not only dependent on the starting sulfathiazole polymorph used, but
also on milling time. When the starting polymorphs are FI and FV, the amorphous
samples recrystallized to FI and FV (with small amount of FI), respectively, without
correlation to milling time (Figure 12c (2) and (5)). If FIII was used as starting material,
the amorphous samples crystallized to the mixture of FIV with FII and FIII when the
milling time was 60 min (Figure 12c (3)), and only FI formed when the milling time was
extended to 90 min, as indicated by the NIR band at 6876 cmꢀ1, Figure 12c (4). Apart
from STZ, amorphous tartaric acid also recrystallized detected by the bands at 5146 cmꢀ1
at 43 and 75% RH.
In the STZꢀCA systems, the first two principal components summarized 84.3% of
the variation in the spectral information. In contrast to the STZꢀTA system, PC1 can
differentiate the variability of FA, FI and FV, PC2 shows spectral differences arising
from the FA with other polymorphic forms (FII, FIII and FIV) along PC1. The FA
samples remain amorphous for up to 28 days at 10% RH, but recrystallize to a mixture of
FII, FIII and IV within 7 days at 75% RH irrespective of the starting sulfathiazole
polymorph used (Figure 12d (6)). At 43% RH, when the starting polymorphs are FI and
FV, the amorphous samples recrystallized to FI and FV (with a small amount of FI),
respectively. (Figure 12c (2) and (5)). If FIII was used as starting material, the amorphous
samples crystallized to a mixture of FIV, FII and FIII when the milling time was 60 min
(Figure 12c (3)), and FI was formed when the milling time was extended to 90 min, as
indicated by the NIR band at 6876 cmꢀ1, shown in Figure 12c (4). Amorphous citric acid
easily absorbed water (see NIR bands caused by water), and recrystallized at 43 and 75%
RH as indicated by the band at 6800 cmꢀ1. The above results strongly suggested that the
stability of these two coamorphous systems is very strongly influenced by relative
humidity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
ACS Paragon Plus Environment

Page 17 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
(a)
(b)
STZ-TA
STZ-CA
0.04
0.02
0.04
(1)
10% RH
(2)
(5)
10% RH
(1)
0.02
0.00
75% RH
(3)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(4)
(6)
0.00
43% RH
(3)
(4)
(2)
-0.02
-0.04
-0.06
-0.08
-0.10
-0.02
-0.04
-0.06
-0.08
-0.10
43% RH
(6)
75% RH
FI
FIII
FV
FI
FIII
FV
(5)
-0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06
PC1 (58.0%)
-0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08
PC1 (60.7%)
(c)
(d)
FIII
4897
FIII
4897
CA
6800
FV
6138
FV
6138
H2O
TA
5146
FI
6876
(6)
H2O
FI
6876
(6)
(5)
(4)
(3)
(2)
(1)
(5)
(4)
(3)
(2)
(1)
7000 6500 6000
5000 4750 4500 4250 4000
Wavenumber (cm-1)
7000 6500 6000 5250 5000 4750 4500 4250 4000
Wavenumber (cm-1)
Figure 12 PCA scores plots of the NIR spectra of the 60 and 90 min comilled STZ samples with
(a) TA and (b) CA stored at different humidity; (c) NIR spectra of STZ FIꢀTA milled for 60 min
stored at 10% RH for 28 days and (2) at 43% RH for 7 days; (3) STZ FIIIꢀTA milled for 60 min
and (4) for 90 min at 43% RH for 7days; (5) STZ FVꢀTA milled for 60 min at 43% RH for 6 days;
(6) STZ FVꢀTA milled for 60 min at 75% RH for 7 days; (d) NIR spectra of STZ FIꢀCA milled
for 60 min stored (1) at 10% RH for 28 days and (2) at 43% RH for 7 days; (3) STZ FIIIꢀCA
milled for 60 min and (4) for 90 min at 43% RH for 7days; (5) STZ FVꢀCA milled for 60 min at
43% RH for 7 days; (6) STZ FIꢀCA milled for 60 min at 75% RH for 7 days.
3.4.2 Characterization of coamorphous systems
FTIR spectra were used to monitor amorphization and to detect any intermolecular
interactions. Figure 13 shows the evolution of these two coamorphous systems with time.
As can be seen from the figure, the IR bands due to crystalline STZ, TA and CA
disappeared after 60 min milling. The PXRD patterns of both mixtures milled for 60 min
also confirmed that the products were coamorphous. The IR results also showed that
when the milling time was extended to 90 min the samples remained amorphous.
When the IR spectra of amorphous STZ was compared to that of the coamorphous
17
ACS Paragon Plus Environment

Crystal Growth & Design
Page 18 of 40
1
2
3
4
5
6
7
8
9
systems no band shifts were observed. For example, there was no shift in the band at
3464 cmꢀ1 of amorphous STZ. This suggests that the coamorphous system contains small
particles of the components which are stabilized by particle to particle hydrogen bonding.
The observed FTIR spectra arise from the bulk of the particles and not from the surface
molecules which are involved in the interparticle interactions. DSC results for the
coamorphous samples at two scan rates are given in the supplementary information. At
the scan rates used the TA samples showed a single crystallization exotherm while the
CA samples showed a double exotherm at the slower scan rate. Double exotherms are not
characteristic of coamorphous systems as they have been observed even for amorphous
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
systems generated from a single phase and have been ascribed to surface and bulk
47, 48
crystallization phenomena.
However, in the present case the quality of the DSC
results is not great and it is difficult to draw definite conclusions from them.
(a)
(b)
STZ FIII-TA PM
STZ FIII-CA PM
5
10 15 20 25 30 35 40
5
10 15 20 25 30 35 40
STZ FA
STZ FA
X 3
FA
3463
X 3
FA
3463
3600 3200 2800
1500 1400 1300 1200 1100 1000
Wavenumber (cm-1)
3500 3150 2800
1650 1500 1350 1200 1050
Wavenumber (cm-1)
Figure 13 IR spectra of STZ FIII comilled with (a) tartaric acid (TA) and (b) citric acid (CA) at
various times. Figure inserts represent the PXRD patterns of STZ comilled with TA and CA for
60 min. The IR spectra from top to bottom: physical mixture (PM), mixtures milled for 15, 30, 60
and 90 min, and amorphous sulfathiazole obtained from cryomilling.
3.4.3 Modelling STZ TA/CA interactions.
It is interesting to consider why TA and CA are the only acids which stabilize coamorphous
systems. A likely possibility is that these are the only acids which not just equal but exceed STZ
in hydrogen bonding ability. STZ has 3 hydrogen bond donors and 4 hydrogen bond acceptors
while TA and CA have 4 and 6 and 4 and 7 hydrogen bond donors and acceptors respectively.
Thus small particles of STZ, TA and CA will have surface molecules which will have spare
hydrogen bonding capacity for binding to other particles. In an attempt to demonstrate this we
have modeled gas phase adducts of STZ with TA and CA using DFT calculations with B3LYP
functionals and 6ꢀ31G* basis sets. The molecules were docked so that the acids spanned the SO2
and either the ꢀNH2 or the thiazole ring. The results are shown in Figure 14.
18
ACS Paragon Plus Environment

Page 19 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 14 calculated adduct structures for STZꢀCA (a) and (b) and STZꢀTA (c) and (d).
Citric acid with three carbons between its –COOH groups was better able to span the
STZ molecules than TA. In each case spare hydrogen bonding capacity remains in both
molecules. The adducts in Figure 14 were more stable than their components by 52.79,
70.45, 51.68 and 53.49 kJ/mol respectively.
4. Conclusions
Milling sulfathiazole with carboxylic acids led to the formation of a 1:1 cocrystal
with glutaric acid a 1:1 salt with oxalic acid and stable coamorphous systems with citric
and glutaric acids. While it is not possible to rationalize either the cocrystal or salt
formation nevertheless the formation of a salt by proton transfer under anhydrous and
solvent free conditions is a higher energy process than cocrystal formation. The stable
coamorphous systems were formed by acids that have hydrogen bonding capacities
which exceed those of sulfathiazole. Polymorphic transformation to FV in the presence of
DLꢀmalic acid has been attributed to its hydroxylic solvent like behavior. Relative
humidity up to 75% did not transform either the cocrystal or the salt. However, the
coamorphous systems were more sensitive to relative humidity. At 10% RH, the
coamorphous systems remained amorphous for up to 28 days. At higher RH
coamorphous sulfathiazole crystallized to different polymorphs the ratio of which
depended both on RH and on the starting sulfathiazole polymorph used.
Acknowledgement
This work was supported by Science Foundation Ireland under Grant No.
[07/SRC/B1158] as part of the Solid State Pharmaceutical Cluster (SSPC). Mr. Dermot
McGrath and Dr. Gerard Wall are thanked for DSC and microscope measurements.
19
ACS Paragon Plus Environment

Crystal Growth & Design
Page 20 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
Table 1 Results of milling sulfathiazole with different carboxylic acids for 60 min.
8
9
Starting material
Carboxylic
acids
m.p.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
FIII
FI
FV
Glutaric acid
Oxalic acid
95
cocrystal
salt
cocrystal
salt
cocrystal
salt
189
LꢀTartaric acid 171
coamorphous
coamorphous
FI+FA a
coamorphous
coamorphous
FI+FA
coamorphous
coamorphous
FV+FA
FV+FA
FV
Citric acid
Adipic acid
Pimelic acid
Malonic acid
156
152
103
135
FIII+FI+FA
FV+FI+FAa,b
FV+FA a
FI+FA
FI+FA
FV+FI+FA
FV+FA
DLꢀMalic acid 130
FV
Succinic acid
184
FI+FA
FV+FI+FA
FI+FA
Fumaric acid
287
FI+FA
FI+FA
a: a small amount of FIII can be detected by IR spectra.
b: 90 min milling time
Table 2 Crystallographic data for STZꢀGA cocrystal and STZꢀOA salt
Sample name
STZꢀGA
STZꢀOA
Empirical formula
Formula weight
Temperature, K
C14H17N3O6S2
387.42
C11H11N3O6S2
345.35
150.1(1)
150.1(1) K
0.71073
Wavelength, Å
Crystal system
0.71073
Monoclinic
Monoclinic
P21/c
Space group
Unit cell dimensions
Å and °
P21/n
a = 8.8854(5)
a = 8.0392(6)
b = 19.9339(10)
c = 8.6360(3)
b = 13.3231(5)
c = 14.7403(7)
β
=101.159(5)
β
= 97.715(4)
Volume, Å3
Z
1711.98(14)
1371.41(13)
4
4
21
ACS Paragon Plus Environment

Crystal Growth & Design
Page 22 of 40
1
2
3
4
5
6
7
8
9
3
Density (calculated), Mg/m
Absorption coefficient, mm
F(000)
1.503
1.673
ꢀ1
0.348
0.423
808
712
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Crystal size, mm
0.40 x 0.10 x 0.10
0.22 x 0.10 x 0.08
3.137 to 25.346
4727
Theta range for data collection, ° 2.917 to 25.347
Reflections collected
10819
Independent reflections
3131 [R(int) = 0.0406] 2510 [R(int) = 0.0249]
Completeness to theta = 25.242° 99.8 %
Data / restraints / parameters 3131 / 0 / 289
99.8 %
2510 / 0 / 200
1.211
2
Goodnessꢀofꢀfit on F
1.013
R1 = 0.0356,
wR2 = 0.0848
R1 = 0.0399,
wR2 = 0.1043
0.426 and ꢀ0.537
Final R indices [I>2sigma(I)]
Largest diff. peak and hole, e.Åꢀ3 0.270 and ꢀ0.380
Table 3. Hydrogen bonds for STZꢀGA and STZꢀOA [Å and °].
DꢀH...A
d(DꢀH)
0.86
0.86
0.86
0.82
0.82
0.82
0.86
0.89
0.89
0.89
0.82
d(H...A)
2.05
2.13
2.36
2.09
2.64
1.77
1.85
1.98
2.24
2.14
2.04
d(D...A)
2.905(3)
2.986(3)
3.154(3)
2.743(4)
3.289(4)
2.580(13)
2.697(3)
2.847(3)
2.953(3)
3.025(3)
2.745(3)
<(DHA)
170.9
171.5
153.4
136.8
137.5
170.2
170.5
165.4
137.3
174.1
143.4
STZꢀGA
N(1)ꢀH(1N1)...N(2)#2
N(3)ꢀH(3A)...O(2)#3
N(3)ꢀH(3B)...O(1)#4
O(3)ꢀH(3)...N(3)#5
O(3)ꢀH(3)...O(2)
O(5)ꢀH(5A)...O(6)#6
N(3)ꢀH(3)...O(6)
STZꢀOA
N(1)ꢀH(2N1)...O(4)#2
N(1)ꢀH(3N1)...O(4)#3
N(1)ꢀH(1N1)...N(2)#2
O(3)ꢀH(1O3)...O(5)#4
Symmetry transformations used to generate equivalent atoms STZꢀGA:
#2 ꢀx+1,ꢀy+2,ꢀz
#3 xꢀ1,y,z
#4 ꢀx+1/2,y+1/2,ꢀz+1/2
#5 x+1,y,z
#6 ꢀx+1,ꢀy+3,ꢀz+1
STZꢀOA: #2 xꢀ1,y,z
#3 xꢀ1,ꢀy+3/2,z+1/2
#4 ꢀx+1,ꢀy+1,ꢀzꢀ1
22
ACS Paragon Plus Environment

Page 23 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
Table of Contents
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Sulfathiazole Carboxylic Acids
cocrystal
salt
Milling
coamorphous
10
15
20
25
10
15
20
25
10 15 20 25 30
Synopsis: Comilling sulfathiazole polymorphs I, III and V with carboxylic acids gave a
1:1 cocrystal with glutaric acid, a 1:1 salt with oxalic acid and coamorphous systems with
Lꢀtartaric and citric acids. A PCA analysis of NIR data showed that forms I and V
proceeded to the cocrystal but that form III transformed to form IV before cocrystal
formation.
ASSOCIATED CONTENT
Supporting Information
DSC plots and Crystallographic cif files and tables are available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Eꢀmail: andrea.erxleben@nuigalway.ie (A. Erxleben) p.mcardle@nuigalway.ie (P.
McArdle). Tel.: +353 91492487; Fax: +353 91525700.
Notes
The authors declare no competing financial interest.
23
ACS Paragon Plus Environment

Crystal Growth & Design
Page 24 of 40
1
2
3
4
5
6
7
References
8
9
(1) Masuda, T.; Yoshihashi, Y.; Yonemochi, E.; Fujii, K.; Uekusa, H.; Terada, K., Cocrystallization
and amorphization induced by drug–excipient interaction improves the physical properties of
acyclovir. International Journal of Pharmaceutics 2012, 422, (1–2), 160ꢀ169.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2) Thakuria, R.; Delori, A.; Jones, W.; Lipert, M. P.; Roy, L.; RodríguezꢀHornedo, N.,
Pharmaceutical cocrystals and poorly soluble drugs. International Journal of Pharmaceutics 2013,
453, (1), 101ꢀ125.
(3) See
US
FDA
GRAS
List.
http://www.fda.gov/food/ingredientspackaginglabeling/gras/default.htm
(4) Brittain, H. G., Pharmaceutical cocrystals: The coming wave of new drug substances. Journal of
Pharmaceutical Sciences 2013, 102, (2), 311ꢀ317.
(5) Qiao, N.; Li, M.; Schlindwein, W.; Malek, N.; Davies, A.; Trappitt, G., Pharmaceutical
cocrystals: An overview. International Journal of Pharmaceutics 2011, 419, (1–2), 1ꢀ11.
(6) Lu, E.; RodriguezꢀHornedo, N.; Suryanarayanan, R., A rapid thermal method for cocrystal
screening. CrystEngComm 2008, 10, (6), 665ꢀ668.
(7) Löbmann, K.; Laitinen, R.; Grohganz, H.; Gordon, K. C.; Strachan, C.; Rades, T.,
Coamorphous Drug Systems: Enhanced Physical Stability and Dissolution Rate of Indomethacin and
Naproxen. Molecular Pharmaceutics 2011, 8, (5), 1919ꢀ1928.
(8) Löbmann, K.; Strachan, C.; Grohganz, H.; Rades, T.; Korhonen, O.; Laitinen, R.,
Coꢀamorphous simvastatin and glipizide combinations show improved physical stability without
evidence of intermolecular interactions. European Journal of Pharmaceutics and Biopharmaceutics
2012, 81, (1), 159ꢀ169.
(9) Chieng, N.; Aaltonen, J.; Saville, D.; Rades, T., Physical characterization and stability of
amorphous indomethacin and ranitidine hydrochloride binary systems prepared by mechanical
activation. European Journal of Pharmaceutics and Biopharmaceutics 2009, 71, (1), 47ꢀ54.
(10) Braga, D.; Giaffreda, S. L.; Grepioni, F.; Pettersen, A.; Maini, L.; Curzi, M.; Polito, M.,
Mechanochemical preparation of molecular and supramolecular organometallic materials and
coordination networks. Dalton Transactions 2006, (10), 1249ꢀ1263.
(11) Trask, A.; Jones, W., Crystal Engineering of Organic Cocrystals by the SolidꢀState Grinding
Approach. In Organic Solid State Reactions, Toda, F., Ed. Springer Berlin Heidelberg: 2005; Vol. 254,
pp 41ꢀ70.
(12) Trask, A. V.; van de Streek, J.; Motherwell, W. D. S.; Jones, W., Achieving Polymorphic and
Stoichiometric Diversity in Cocrystal Formation: Importance of SolidꢀState Grinding, Powder
Xꢀray Structure Determination, and Seeding. Crystal Growth & Design 2005, 5, (6), 2233ꢀ2241.
(13) Yamamoto, K.; Tsutsumi, S.; Ikeda, Y., Establishment of cocrystal cocktail grinding method for
rational screening of pharmaceutical cocrystals. International Journal of Pharmaceutics 2012, 437,
(1–2), 162ꢀ171.
(14) Lin, H.ꢀL.; Zhang, G.ꢀC.; Hsu, P.ꢀC.; Lin, S.ꢀY., A portable fiberꢀoptic Raman analyzer for fast
realꢀtime screening and identifying cocrystal formation of drugꢀcoformer via grinding process.
Microchemical Journal 2013, 110, (0), 15ꢀ20.
(15) Chieng, N.; Hubert, M.; Saville, D.; Rades, T.; Aaltonen, J., Formation Kinetics and Stability of
Carbamazepine−Nicotinamide Cocrystals Prepared by Mechanical Activation. Crystal Growth &
Design 2009, 9, (5), 2377ꢀ2386.
(16) Lin, S.ꢀY.; Hsu, C.ꢀH.; Ke, W.ꢀT., Solidꢀstate transformation of different gabapentin
polymorphs upon milling and coꢀmilling. International Journal of Pharmaceutics 2010, 396, (1–2),
83ꢀ90.
(17) Balani, P. N.; Ng, W. K.; Tan, R. B. H.; Chan, S. Y., Influence of excipients in comilling on
mitigating millingꢀinduced amorphization or structural disorder of crystalline pharmaceutical
actives. Journal of Pharmaceutical Sciences 2010, 99, (5), 2462ꢀ2474.
(18) Curtin, V.; Amharar, Y.; Hu, Y.; Erxleben, A.; McArdle, P.; Caron, V.; Tajber, L.; Corrigan, O.
I.; Healy, A. M., Investigation of the Capacity of Low Glass Transition Temperature Excipients to
24
ACS Paragon Plus Environment

Page 25 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Minimize Amorphization of Sulfadimidine on Comilling. Molecular Pharmaceutics 2012, 10, (1),
386ꢀ396.
(19) Brittain, H. G., Polymorphism and solvatomorphism 2010. Journal of Pharmaceutical Sciences
2012, 101, (2), 464ꢀ484.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20) Anwar, J.; Tarling, S. E.; Barnes, P., Polymorphism of sulfathiazole. Journal of Pharmaceutical
Sciences 1989, 78, (4), 337ꢀ342.
(21) Apperley, D. C.; Fletton, R. A.; Harris, R. K.; Lancaster, R. W.; Tavener, S.; Threlfall, T. L.,
Sulfathiazole polymorphism studied by magicꢀangle spinning NMR. Journal of Pharmaceutical
Sciences 1999, 88, (12), 1275ꢀ1280.
(22) Bingham, A. L.; Hughes, D. S.; Hursthouse, M. B.; Lancaster, R. W.; Tavener, S.; Threlfall, T.
L., Over one hundred solvates of sulfathiazole. Chemical Communications 2001, 0, (7), 603ꢀ604.
(23) McIntosh, J.; Robinson, R. H. M.; Selbie, F. R., ACRIDINEꢀSULPHONAMIDE COMPOUNDS
AS WOUND ANTISEPTICS CLINICAL TRIALS OF FLAVAZOLE. The Lancet 1945, 246, (6361),
97ꢀ99.
(24) Abu Bakar, M. R.; Nagy, Z. K.; Rielly, C. D.; Dann, S. E., Investigation of the riddle of
sulfathiazole polymorphism. International Journal of Pharmaceutics 2011, 414, (1–2), 86ꢀ103.
(25) Munroe, Á.; Rasmuson, Å. C.; Hodnett, B. K.; Croker, D. M., Relative Stabilities of the Five
Polymorphs of Sulfathiazole. Crystal Growth & Design 2012, 12, (6), 2825ꢀ2835.
(26) Caira, M. R., Sulfa Drugs as Model Cocrystal Formers. Molecular Pharmaceutics 2007, 4, (3),
310ꢀ316.
(27) Allen, F., The Cambridge Structural Database: a quarter of a million crystal structures and
rising. Acta Crystallographica Section B 2002, 58, (3 Part 1), 380ꢀ388.
(28) Lu, T.; Chen, C., Uncertainty evaluation of humidity sensors calibrated by saturated salt
solutions. Measurement 2007, 40, (6), 591ꢀ599.
(29) Sheldrick, G., A short history of SHELX. Acta Crystallographica Section A 2008, 64, (1), 112ꢀ122.
(30) McArdle, P.; Gilligan, K.; Cunningham, D.; Dark, R.; Mahon, M., A method for the prediction
of the crystal structure of ionic organic compoundsꢀthe crystal structures of oꢀtoluidinium chloride
and bromide and polymorphism of bicifadine hydrochloride. CrystEngComm 2004, 6, (53), 303ꢀ309.
(31) McArdle, P.; Hu, Y.; Lyons, A.; Dark, R., Predicting and understanding crystal morphology:
the morphology of benzoic acid and the polymorphs of sulfathiazole. Crystengcomm 2010, 12, (10),
3119ꢀ3125.
(32) McArdle, P., J. Appl. Crystallogr. 1995, 28.
(33) Granovsky,
A.
A.
Firefly
(previously
PC
GAMESS).
http://classic.chem.msu.su/gran/firefly/index.html
(34) Schmidt, M. W.; Baldridge, K. K.; J.A.Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.;
Koseki, S.; atsunaga, N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A.,
J.Comput.Chem. 1993, 14, 1347ꢀ1363
(35) Blagden, N.; J. Davey, R.; F. Lieberman, H.; Williams, L.; Payne, R.; Roberts, R.; Rowe, R.;
Docherty, R., Crystal chemistry and solvent effects in polymorphic systems Sulfathiazole. Journal of
the Chemical Society, Faraday Transactions 1998, 94, (8), 1035ꢀ1044.
(36) Gelbrich, T.; Hughes, D. S.; Hursthouse, M. B.; Threlfall, T. L., Packing similarity in
polymorphs of sulfathiazole. CrystEngComm 2008, 10, (10), 1328ꢀ1334.
(37) Hu, Y.; Erxleben, A.; Hodnett, B. K.; Li, B.; McArdle, P.; Rasmuson, Å. C.; Ryder, A. G.,
Solidꢀstate Transformations of Sulfathiazole Polymorphs: the Effects of Milling and Humidity.
Crystal Growth & Design 2013.
(38) Etter, M. C., Encoding and decoding hydrogenꢀbond patterns of organic compounds. Accounts
of Chemical Research 1990, 23, (4), 120ꢀ126.
(39) Zhang, S.; Rasmuson, Å. C., Thermodynamics and Crystallization of the Theophylline–Glutaric
Acid Cocrystal. Crystal Growth & Design 2013, 13, (3), 1153ꢀ1161.
(40) Van Eerdenbrugh, B.; Taylor, L. S., Application of midꢀIR spectroscopy for the
characterization of pharmaceutical systems. International Journal of Pharmaceutics 2011, 417, (1–2),
3ꢀ16.
(41) Childs, S. L.; Stahly, G. P.; Park, A., The Salt−Cocrystal Continuum: The Influence of Crystal
25
ACS Paragon Plus Environment

Crystal Growth & Design
Page 26 of 40
1
2
3
4
5
6
7
8
9
Structure on Ionization State. Molecular Pharmaceutics 2007, 4, (3), 323ꢀ338.
(42) Braga, D.; Giaffreda, S. L.; Grepioni, F.; Chierotti, M. R.; Gobetto, R.; Palladino, G.; Polito, M.,
Solvent effect in a "solvent free" reaction. CrystEngComm 2007, 9, (10), 879ꢀ881.
(43) Trask, A. V.; Haynes, D. A.; Motherwell, W. D. S.; Jones, W., Screening for crystalline salts via
mechanochemistry. Chemical Communications 2006, (1), 51ꢀ53.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(44) McHugh, C.; Erxleben, A., Supramolecular Structures and Tautomerism of Carboxylate Salts
of Adenine and Pharmaceutically Relevant N6ꢀSubstituted Adenine. Crystal Growth & Design 2011,
11, (11), 5096ꢀ5104.
(45) Losev, E. A.; Mikhailenko, M. A.; Achkasov, A. F.; Boldyreva, E. V., The effect of carboxylic
acids on glycine polymorphism, salt and coꢀcrystal formation. A comparison of different
crystallisation techniques. New Journal of Chemistry 2013, 37, (7), 1973ꢀ1981.
(46) Grisedale, L. C.; Belton, P. S.; Jamieson, M. J.; Barker, S. A.; Craig, D. Q. M., An investigation
into water interactions with amorphous and milled salbutamol sulphate: The development of
predictive models for uptake and recrystallization. International Journal of Pharmaceutics 2012, 422,
(1–2), 220ꢀ228.
(47) Chattoraj, S.; Bhugra, C.; Telang, C.; Zhong, L.; Wang, Z.; Sun, C., Origin of Two Modes of
Nonꢀisothermal Crystallization of Glasses Produced by Milling. Pharm Res 2012, 29, (4), 1020ꢀ1032.
(48) Trasi, N.; Byrn, S., Mechanically Induced Amorphization of Drugs: A Study of the Thermal
Behavior of Cryomilled Compounds. AAPS PharmSciTech 2012, 13, (3), 772ꢀ784.
26
ACS Paragon Plus Environment

Page 27 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1: Chemical structures of sulfathiazole and the carboxylic acids used. GRAS acids are denoted by a *.
154x158mm (300 x 300 DPI)
ACS Paragon Plus Environment

Crystal Growth & Design
Page 28 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 (a) PXRD patterns for Glutaric acid, STZ FI, STZ-GA cocrystal experimental and STA-GA cocrystal
calculated, (b) Morphology of the STZ-GA cocrystals.
61x25mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 29 of 40
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3 Hydrogen bonding in the STZ-GA cocrystal structure. H atoms not involved in hydrogen bonding
have been omitted for clarity.
80x77mm (300 x 300 DPI)
ACS Paragon Plus Environment