Polymorphism in acesulfame sweetener: Structure-property and stability relationships of bending and brittle crystals

Article abstract of DOI:10.1039/c0cc00028k

Acesulfame is found to exist in two crystalline forms of which Form I (needles) shows bending upon mechanical stress. Crystal structures explain their mechanical response. This is the first case of aliphatic organic compounds featuring a bending phenomenon. Form I is physically more stable than Form II in ambient conditions.

Full text of DOI:10.1039/c0cc00028k

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Polymorphism in acesulfame sweetener: structure–property and stability
relationships of bending and brittle crystalsw
a
a
a
mb
¨
Sitaram P. Velaga,* Venu R. Vangala,z Srinivas Basavoju and Dan Bostro
Received 19th February 2010, Accepted 14th April 2010
First published as an Advance Article on the web 27th April 2010
DOI: 10.1039/c0cc00028k
8
Acesulfame is found to exist in two crystalline forms of which
Form I (needles) shows bending upon mechanical stress. Crystal
structures explain their mechanical response. This is the first
case of aliphatic organic compounds featuring a bending pheno-
menon. Form I is physically more stable than Form II in ambient
conditions.
than that of sucrose. This study deals with the discovery
of polymorphism in acesulfame and unique mechanical beha-
viour of one of the crystal forms. Polymorphs of acesulfame,
Forms I and II, were characterized with respect to their
physical and mechanical properties and thermodynamic
stability relationships between the forms were established.
The mechanical properties of the two forms were tied to their
crystal structures.
Polymorphism in crystalline materials is of paramount impor-
1
tance to the pharmaceutical as well as the food industry.
Different crystalline forms of a solid compound can have
different physical, chemical and mechanical properties. Thus,
polymorphic behaviour of pharmaceutical or food materials
must be thoroughly investigated for the successful develop-
ment of a quality product. Polymorphs with different mecha-
nical properties can have profound implications for the
Acesulfame potassium (Ace K) was neutralized with HCl to
afford a salt free acesulfame (ESIw). Good quality needle
shaped crystals were obtained from ethyl acetate (EtOAc) by
the evaporative crystallization of acesulfame in ambient con-
ditions (Fig. 1a). Motivated by our interest in the mechanical
properties of crystals and for curiosity, we applied stress with
the help of two metallic needles on both ends of the needles,
and to our surprise, they became bent (Fig. 2). However, these
crystals could not be straightened back completely by a
counter stress. In search of other crystal forms, these needles
2
processing, formulation and eventually product performance.
In fact, control of the mechanical properties requires an
3
improved understanding of structure–property relationships.
4
A seminal contribution by Desiraju and co-workers demon-
strates that organic crystals exhibit bending and shearing
5
phenomena just as inorganic materials do when mechanical
stress is applied. They proposed that the sorting of visually
indistinguishable polymorphs of 6-chloro-2,4-dinitroaniline is
6
possible based on the mechanical properties. Incidentally, the
(
B15 mg) were ground in a Retsch mixer mill for 10 s. A few
drops of CH Cl were added to the ground powder placed on a
bending is observed in one of the crystal forms of an anti-
7
convulsant drug, venlafaxine hydrochloride. The bending/
2
2
glass slide while observing under the microscope. Interestingly,
few prismatic crystals were formed instantaneously (Fig. 1b).
As opposed to needles, these crystals showed breakage upon
mechanical stress.
shearing phenomenon in organic crystals is attributed to the
˚
corrugated sheet or layered structure with 4 A short axis and
anisotropic interactions. However, all these compounds are
broadly aromatic and this communication unveils the bending
phenomenon in aliphatic cyclic compounds for the first time.
Acesulfame (6-methyl-1,2,3-oxathiazin-4(3H)-one 2,2-dioxide)
is an aliphatic calorie-free sweetener and its potassium salt is
widely used in food products, beverages, vitamins, table-top
sweeteners, cosmetics, and in pharmaceutical formulations.
The sweetening power of acesulfame is about 180 times greater
The contrast in morphologies as well as mechanical proper-
ties of the crystals indicated the existence of polymorphism in
acesulfame and prompted a detailed characterization of these
phases. However, it was difficult to generate prismatic crystals
2 2 3
from the bench-top solvents, CH Cl , CHCl , EtOAc and
ethanol by slow evaporations, as needles were obtained all
the time. Therefore, seeds were used to control the nucleation
9
and to promote exclusive growth of the prisms. With this
strategy, good quality crystals (prisms) were obtained in
reasonable quantities.
a
Department of Health Sciences, Lulea˚ University of Technology,
Lulea˚ S-971 87, Sweden. E-mail: sitram.velaga@ltu.se;
Fax: +45-920-493850; Tel: +46-920-493920
Energy Technology and Thermal Process Chemistry,
Thermal analysis, DSC and TGA, was carried out for needle
and prismatic crystals. DSC traces of these crystalline phases
were different (Fig. 3a). Needles showed one major endotherm
b
Umea˚ University, Umea˚ S-901 87, Sweden
w Electronic supplementary information (ESI) available: Thermal and
Raman data, conformations, H-bond geometries, simulated PXRDs,
heat–cool–heat DSC, slurry product phase DSC and PXRD details.
CCDC 766997–766998. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0cc00028k
z Present address: Institute of Chemical and Engineering Sciences,
A*STAR (Agency for Science, Technology and Research), 1 Pesek
Road, Jurong Island, Singapore, 627833, Singapore. E-mail: venugopal_
vangala@ices.a-star.edu.sg; Fax: +65 6316 6183; Tel: +65 6796 3862.
Fig. 1 Optical micrographs of (a) needle and (b) prismatic crystals.
This journal is ꢀc The Royal Society of Chemistry 2010
3
562 | Chem. Commun., 2010, 46, 3562–3564

View Online
Fig. 4 (a) Crystal packing of Form I along the c axis. Note the
N–Hꢃ ꢃ ꢃO catemers between the two symmetry independent molecules.
b) Stacks view down the a axis and the bending which occurs in the
001] plane.
(
[
Fig. 2 Bending sequence in acesulfame needle crystals.
exhibiting bending. To our knowledge this is the first case of
an aliphatic organic crystal that featured bending behaviour.
ꢂ1
at B122 1C (enthalpy of transition, 130.8 ꢁ 2.6 J g ) that is
ꢀ
Form II crystallizes in the triclinic, P1 space group with
attributed to a melting event. Two endotherms were observed
ꢂ1
0
Z = 1. Its conformation was similar to one of the two indepen-
for prisms: a minor endotherm at B83 1C (5.4 ꢁ 0.5 J g ) that
dent molecules of Form I.w Form II prisms could not be bent in
any plane upon mechanical stress. Not surprisingly, analysis of
the crystal structure revealed that it was not a layered structure.
could be interpreted as a solid–solid phase transition and a
ꢂ1
major endotherm at B122 1C (132.4 ꢁ 4.9 J g ) corresponding
to a melting event (Fig. 3a). TGA profiles showed the absence
of solvent molecules in the crystal.w
The powder X-ray diffraction patterns of these two phases
˚
However, it was found to have the short axis of B4 A. In fact, it
was claimed that it is a necessary but not a sufficient condition
4
for a crystal to show shearing or bending. The major inter-
were distinctly different, confirming that they are polymorphs
action type was a discrete centrosymmetric N–Hꢃ ꢃ ꢃO imide
(
Fig. 3b). We classified the needle crystals that featured
1
0
˚
dimer synthon (d/A, y/1: 1.77, 169) (Fig. 5a). A close packing
bending as Form I and the prisms that were brittle in nature
as Form II.y Furthermore, Raman analysis corroborates the
DSC and PXRD conclusions and provides molecular insights
into differences between these two solid forms.w
was efficiently achieved in 3D by C–Hꢃ ꢃ ꢃO hydrogen bonds
˚
(
d/A, y/1: 2.49, 157; 2.65, 125; 2.67, 145; 2.68, 151) (Fig. 5b). As
rationalized in the literature, brittleness appears to be because
of the subtle differences in the relative strengths of the inter-
4
actions in three orthogonal directions. The simulated and
The acesulfame dimorphs were structurally characterized by
single crystal X-ray diffraction to reveal the relationship
between the structure and the mechanical property.z Form I
experimental PXRD patterns for Forms I and II were perfectly
matched.w
crystallizes in the monoclinic, P2
1
/c space group with two
0
In this part, we sought to understand the stability relation-
ships of the Forms I and II. Thermodynamic analysis of DSC
data formed an excellent basis for the construction of a
molecules of acesulfame in the asymmetric unit (Z = 2). The
molecules adopt two different conformations.w Interestingly,
the crystal structure analysis reveals that the acesulfame
molecules form a layered structure. The layers were stabilized
1
1
qualitative energy–temperature diagram (Fig. 6). Physical
properties of the acesulfame polymorphs are given in Table 1.
The transitions were endothermic and the melting tempera-
ture differences were small or insignificant, and hence according
to heat of transition and heat of fusion rules, the polymorphs
˚
by the robust imide catemers with N–Hꢃ ꢃ ꢃO synthons (d/A,
y/1: 1.74, 169; 1.77, 156) to form one-dimensional tapes along
the a-axis (Fig. 4a). These catemers were propagated in a
second dimension with the support of strong and specific
bifurcated C–Hꢃ ꢃ ꢃO interactions to form layers. Consequently,
the crystals have grown along the c-axis. The short axis of the
1
2
are enantiotropes (Fig. 6). The PXRD pattern of Form II
heating pass B83 1C in the DSC was matched with that of
Form I, which implies that the Form II had transformed to
Form I. Further, heat–cool–heat studies in the DSC suggested
that these enantiotropic polymorphs were non-reversible.w
In general, enantiotropic transitions are thermodynamically
reversible with temperature and pressure, however, they may in
˚
unit cell is 7.4940 (B2 ꢄ 4 A). The layers were stacked in an
˚
˚
antiparallel fashion with separations of B3.2 A and B3.3 A
with weak and non-specific C–Hꢃ ꢃ ꢃO bonds. All the molecules
in the layers were arranged in parallel to the face (001) and
the bending had occurred along this plane (Fig. 4b). The
face indexing of the crystal was in accordance with the bending
direction and the solid-state packing.w This is not the corrugated
sheet structure that is typical of some of the bending struc-
1
3
turn be discernible as irreversible in practical investigations. In
any case their free energies become equal at the transition
II–I
temperature T
p
.
4
The transition temperatures observed in DSC could be
tures, however, Form I is a classical case of layered crystal
1
4
of kinetic origin and slurry transformation studies are a
practical way to determine the thermodynamic stabilities of
different solid forms. The PXRD and DSC analysis of the
Fig. 3 Acesulfame (a) DSC traces and (b) PXRD patterns for Forms
Fig. 5 Acesulfame Form II. (a) Centrosymmetric N–Hꢃ ꢃ ꢃO dimer
synthon, and (b) the effective close packing via C–Hꢃ ꢃ ꢃO interactions.
I and II.
This journal is ꢀc The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3562–3564 | 3563

View Online
The authors thank the Kempe Foundation for an instrumen-
tation grant and SPV and DB acknowledge the Swedish
Research Council (SRC) for a project grant.
Notes and references
y Baseline physical characterization of neutral acesulfame shown to be
Form I.
z Crystallographic data for the acesulfame dimorphs (C H NO S,
4 5 4
M = 163.15, Kappa CCD, ꢂ123 1C). Form I: monoclinic, a =
˚
7
1
.4940(15), b = 11.784(2), c = 15.238(3) A, b = 101.36(3)1, V =
3
319.3(4) A , D = 1.643 g cm , space group = P2
ꢂ3
˚
1
/c, Z = 8,
Fig. 6 Energy–temperature diagram of an enantiotropic dimorphic
ꢂ1
m(Mo-Ka) = 0.443 mm , size 0.22 ꢄ0.18 ꢄ 0.13 mm. 9845 total
p f
system: G, free energy; H, enthalpy; T , transition temperature; T ,
reflections of which 5031 were independent, 4024 observed [I 4 2s(I)].
melting temperature.
2
Refinement against F with 183 parameters, R1 [I 4 2s(I)] = 0.0432,
wR2 = 0.1433. Form II: triclinic, a = 6.3310(2), b = 7.4720(3),
˚
c = 7.7150(15) A, a = 62.5460(15), b = 73.1710(18), g = 77.874(2)1,
Table 1 Physical properties of acesulfame, Forms I and II
3
V = 308.73(2) A , D = 1.755 g cm , space group = P1, Z = 2,
ꢂ3
˚
ꢀ
ꢂ1
Modification
Form I
Form II
m(Mo-Ka) = 0.473 mm , size 0.28 ꢄ0.26 ꢄ 0.18 mm. 1307 total
reflections of which 1307 were independent, 1256 observed [I 4 2s(I)].
a
2
M.p. (1C) DSC onset temp.
Enthalpy of transition/J g
—1 22.27 ꢁ 0.13
1.643
67.8
—
Refinement against F with 91 parameters, R1 [I 4 2s(I)] = 0.0313,
wR2 = 0.0910.
ꢂ1
ꢂ3
a
82.57 ꢁ 0.57
Calculated density/g cm
Packing co-efficient Ck*
True density/g cm
1.755
72.7
1
J. Bernstein, Polymorphism in Molecular Crystals, Clarendon Press,
Oxford, 2002; R. Hilfiker, in Polymorphism in the Pharmaceutical
Industry, Wiley-VCH, Weinheim, Germany, 2006; D. J. W. Grant, in
Polymorphism in Pharmaceutical Solids, ed. H. G. Brittain, Marcel
Dekker, Inc., New York, 1999, pp. 1–34; J. M. Aceves-Hernandez,
ꢂ3
b
b
1.6154 ꢁ 0.0021
1.7097 ꢁ 0.0094
Metastable
ꢂ29.83
Stability at 22 1C
Lattice energies/kcal mol
Stable
ꢂ28.25
ꢂ1
a
b
n = 3 determinations. n = 10 determinations, true densities of
Forms I and II are significantly different (T-test, p o0.05).
I. Nicola
´
V. M. Castan
s-Va
´
zquez, F. J. Aceves, J. Hinojosa-Torres, M. Paz and
ˇ
o, J. Pharm. Sci., 2009, 98, 2448–2463; Z. Ma and
B. Moulton, J. Chem. Crystallogr., 2009, 39, 913–918.
C. C. Sun and D. J. W. Grant, Pharm. Res., 2001, 18, 274–280;
C. C. Sun, J. Pharm. Sci., 2009, 98, 1671–1687.
2
3
product phase from slurry studies (at room temperature (RT)
of about 22 1C) indicated that the Form II had transformed to
Form I. This result confirmed that the Form I is thermo-
dynamically more stable than the Form II in ambient con-
ditions and the transition temperature could be at or below
RT. Thus, adequate attention must be paid with respect to
evaluating the physical stabilities of enantiotropic polymorphs
at storage conditions.w
J. Bernstein, Nat. Mater., 2005, 4, 427–428; S. Karki, T. Fric
L. Fabian, P. R. Laity, G. M. Day and W. Jones, Adv. Mater.,
009, 21, 3905–3909.
ic,
ˇ ´
´
´
2
4
S. Basavoju, C. M. Reddy and G. R. Desiraju, Acta Crystallogr.,
Sect. E: Struct. Rep. Online, 2005, 61, o822–o823; C. M. Reddy,
K. A. Padmanabhan and G. R. Desiraju, Cryst. Growth Des., 2006,
6, 2720–2731; C. M. Reddy, M. T. Kirchner, R. C. Gundakaram,
K. A. Padmanabhan and G. R. Desiraju, Chem.–Eur. J., 2006, 12,
2
G. E. Dieter, Mechanical Metallurgy, SI Metric edn, McGraw-Hill,
222–2234.
The calculated densities and packing fractions were obtained
from the single crystal structures at ꢂ123 1C and true densities
for these forms were measured using a pychnometer at RT
5
Singapore, 1988, pp. 651–678.
6 C. M. Reddy, R. V. Gundakaram, S. Basavoju, M. T. Kirchner,
K. A. Padmanabhan and G. R. Desiraju, Chem. Commun., 2005,
(
Table 1). The calculated density of Form II was higher than
2
C. M. Reddy, R. V. Gundakaram, S. Basavoju, M. T. Kirchner,
439–2441.
that of Form I, which is in line with the close packing observed
in Form II. Lattice energy calculations further indicated that
7
8
K. A. Padmanabhan and G. R. Desiraju, Chem. Commun., 2005,
3
ꢂ1
945–3947.
the Form II had lower free energy (B1.6 kcal mol ) com-
J. H. Chu, in Handbook of Pharmaceutical Excipients, ed.
R. C. Rowe, P. J. Sheskey and S. C. Owen, American Pharma-
ceutical Association, Washington, D.C., 5th edn, 2006, pp. 4–5;
FAO/WHO, World Health Organ. Tech. Rep. Ser., 1991, No. 806.
pared to Form I at ꢂ123 1C, and therefore, Form II is
thermodynamically stable at ꢂ123 1C.w In fact, the stability
order has changed somewhere between ꢂ123 1C and RT as
indicated by equilibrium slurry studies. However, the density of
Form II was higher than that of Form I at ambient tempera-
ture. As cautioned in the literature, the density rule can deviate
9
A. A. Llina
98–210; W. Beckmann, Org. Process Res. Dev., 2000, 4, 372–383.
0 S. Basavoju, D. Bostrom and S. P. Velaga, Pharm. Res., 2008, 25,
530–541.
`
s and J. M. Goodman, Drug Discovery Today, 2008, 13,
1
1
¨
0
15
11 N. Zencirci, T. Gelbrich, D. C. Apperley, R. K. Harris,
V. Kahlenberg and U. J. Griesser, Cryst. Growth Des., 2010, 10,
for solids with Z a 1 which is the case for Form I.
In conclusion, the two polymorphic forms of acesulfame
sweetener were discovered and unequivocally characterized.
One of these aliphatic polymorphs showed bending (Form I)
and the other was brittle (Form II) in nature. The bending
phenomenon in Form I is very well correlated to the layered
structure. Forms I and II were enantiotropically related. Form I
is thermodynamically more stable than Form II in ambient
conditions. Future prospects are to investigate the influence
of mechanical properties (bending/brittle) on the powder
compaction or tableting behaviour.
3
02–313.
1
1
2 A. Burger and R. Ramberger, Mikrochim. Acta, 1979, II,
259–272; A. Burger and R. Ramberger, Mikrochim. Acta, 1979,
II, 273–316.
3 L. Yu, J. Pharm. Sci., 1995, 84, 966–974; K. Kawakami, J. Pharm.
ndez and
J. A. Ellena, J. Pharm. Sci., 2006, 95, 1075–1083.
Sci., 2007, 96, 982–989; D. Vega, A. Petragalli, D. Ferna
´
14 Y. Hu, H. Wikstro
¨
m, S. R. Byrn and L. S. Taylor, J. Pharm.
Biomed. Anal., 2007, 45, 546–551.
1
5 C. C. Sun, J. Pharm. Sci., 2007, 96, 1043–1052; S. Thirunahari,
S. Aitipamula, P. S. Chow and R. B. H. Tan, J.Pharm. Sci., 2010,
DOI: 10.1002/jps.22061.
3
564 | Chem. Commun., 2010, 46, 3562–3564
This journal is ꢀc The Royal Society of Chemistry 2010