molecules
Article
Structure, Solubility and Stability of Orbifloxacin
Crystal Forms: Hemihydrate versus Anhydrate
Olimpia Maria Martins Santos 1,2 , Jennifer Tavares Jacon Freitas 2 ,
Edith Cristina Laignier Cazedey 2,3 , Magali Benjamim de Araújo 2 and
Antonio Carlos Doriguetto 1, *
1
2
3
*
Institute of Chemistry, Federal University of Alfenas, Rua Gabriel Monteiro da Silva, 700,
Alfenas-MG 37130-000, Brazil; [email protected]
Faculty of Pharmaceutical Sciences, Federal University of Alfenas; Rua Gabriel Monteiro da Silva, 700,
Alfenas-MG 37130-000, Brazil; [email protected] (J.T.J.F); [email protected] (E.C.L.C.);
[email protected] (M.B.d.A.)
Faculty of Pharmacy, Federal University of Bahia, Campus Ondina, Rua Barão de Jeremoabo, 147, Ondina,
Salvador-BA 40170-115, Brazil
Correspondence: [email protected] or [email protected]; Tel.: +55-35-3299-1261
Academic Editors: Sohrab Rohani and Derek J. McPhee
Received: 18 January 2016 ; Accepted: 23 February 2016 ; Published: 9 March 2016
Abstract: Orbifloxacin (ORBI) is a widely used antimicrobial drug of the fluoroquinolone class. In the
official pharmaceutical compendia the existence of polymorphism in this active pharmaceutical
ingredient (API) is reported. No crystal structure has been reported for this API and as described
in the literature, its solubility is very controversial. Considering that different solid forms of the
same API may have different physicochemical properties, these different solubilities may have
resulted from analyses inadvertently carried out on different polymorphs. The solubility is the
most critical property because it can affect the bioavailability and may compromise the quality of
a drug product. The crystalline structure of ORBI determined by SCXRD is reported here for the
first time. The structural analysis reveals that the ORBI molecule is zwitterionic and hemihydrated.
ORBI hemihydrated form was characterized by the following techniques: TG/DTA, FTIR-ATR, and
PXRD. A second crystalline ORBI form is also reported: the ORBI anhydrous form was obtained by
heating the hemihydrate. These ORBI solid forms were isomorphous, since no significant change in
unit cell and space group symmetry were observed. The solid-state phase transformation between
these forms is discussed and the equilibrium solubility data were examined in order to check the
impact of the differences observed in their crystalline structures.
Keywords: orbifloxacin; crystalline structure; solubility; stability
1. Introduction
Fluoroquinolones are a class of antimicrobials that are widely used in human and veterinary
medicine [1–3]. This group of antimicrobial agents has significant bactericidal activity against
a wide range of veterinary pathogens and Gram-negative bacteria (Escherichia coli, certain species
of Enterobacter, Klebsiella, Pasteurella, Proteus, Salmonella, and Pseudomonas) and some Gram-positive
bacteria, such as Staphylococcus spp., Chlamydia, Mycobacteria, Mycoplasma, and Ureaplasma spp. [2,4].
Orbifloxacin (ORBI, 1-cyclopropyl-7-[(3S,5R)-3,5-dimethylpiperazin-1-yl]-5,6,8-trifluoro-4-oxo-1,
4-dihydroquinoline-3-carboxylic acid, Figure 1), is a third-generation synthetic fluoroquinolone that
has been developed exclusively for the treatment of gastrointestinal and respiratory infections in
animals [1,5,6].
Molecules 2016, 21, 328; doi:10.3390/molecules21030328
www.mdpi.com/journal/molecules
Molecules 2016, 21, 328
Molecules 2016, 21, 328
2 of 19
2 of 19
F
O
O
F
OH
H3C
N
N
F
HN
CH3
Figure
Figure 1.
1. Chemical
Chemical structure
structure of
of the
the antimicrobial
antimicrobial drug
drug orbifloxacin.
orbifloxacin.
According
to the
According to
the Merck
Merck Index,
Index, ORBI
ORBI is
is aa slightly
slightly yellow
yellow crystalline
crystalline powder
powder that
that is
is freely
freely soluble
soluble
in
water,
soluble
in
methanol,
and
slightly
soluble
in
ethanol,
ether,
ethyl
acetate,
and
chloroform
in water, soluble in methanol, and slightly soluble in ethanol, ether, ethyl acetate, and chloroform [7].
[7].
However,
in
the
British
and
US
Pharmacopoeias,
ORBI
appears
as
white
crystals
or
pale
yellow
However, in the British and US Pharmacopoeias, ORBI appears as white crystals or pale yellow
crystalline
crystalline powder
powder that
that is
is very
very slightly
slightly soluble
soluble in
in water,
water, methanol,
methanol, and
and chloroform;
chloroform; soluble
soluble in
in glacial
glacial
acetic
acid;
and
practically
insoluble
in
anhydrous
ethanol
[8,9].
acetic acid; and practically insoluble in anhydrous ethanol [8,9].
This
temperature between
between 15
15 ˝°C
This solubility
solubility was
was recorded
recorded at
at aa temperature
C and
and 25
25 ˝°C
C according
according to
to the
the
pharmacopeias.
“Freely
soluble
in
water”
[7]
means
that
1
g
of
ORBI
is
soluble
in
1–10
mL
of
solvent;
pharmacopeias. “Freely soluble in water” [7] means that 1 g of ORBI is soluble in 1–10 mL of solvent;
thus,
thus, between
between 0.1
0.1 and
and 11 g
g of
of ORBI
ORBI is
is soluble
soluble in
in 11 mL
mL of
of water.
water. In
In contrast,
contrast, “very
“very slightly
slightly soluble”
soluble” [8,9]
[8,9]
means
of ORBI
ORBI is
is soluble
soluble in
in 1000–10,000
1000–10,000 mL
mL of
of solvent;
solvent; thus,
thus, between
means that
that 11 g
g of
between 0.0001
0.0001 and
and 0.001
0.001 g
g of
of
ORBI
is
soluble
in
1
mL
of
water
[8,9].
This
discrepancy
can
be
attributed
to
crystalline
solid
state
ORBI is soluble in 1 mL of water [8,9]. This discrepancy can be attributed to crystalline solid state
changes,
changes, since
since the
the British
Britishand
andU.S.
U.S.Pharmacopoeias
Pharmacopoeiasreport
reportthe
theexistence
existenceofofpolymorphism
polymorphismininthis
thisAPI.
API.
Polymorphs
least
Polymorphs are
are crystals
crystals that
that have
have the
the same
same chemical
chemical composition
composition but
but are
are arranged
arranged in
in at
at least
two
the solid
solid state
state [10–12].
[10–12]. However,
in pharmaceutical
pharmaceutical sciences
used
two different
different forms
forms in
in the
However, in
sciences the
the term
term is
is used
with aa wider
widerrange
rangeofofspecies
species
that
also
includes
other
solid
monomulticomponent
of
with
that
also
includes
other
solid
monoand and
multicomponent
formsforms
of APIs
APIs
and
excipients
such
as
hydrates,
solvates,
amorphous,
salts
and
co-crystals
[13].
and excipients such as hydrates, solvates, amorphous, salts and co-crystals [13].
Drug
phenomenon known
1960s, when
when Aguiar
Aguiar and
and colleagues
colleagues
Drug polymorphism
polymorphism is
is aa phenomenon
known since
since the
the 1960s,
analysed
analysed two
two solid
solid forms
forms of
of the
the antibiotic
antibiotic chloramphenicol
chloramphenicol palmitate.
palmitate. It
It was
was concluded
concluded that
that forms
forms A
A
and
B
of
chloramphenicol
palmitate
showed
different
dissolution
rates
and
different
blood
levels,
and B of chloramphenicol palmitate showed different dissolution rates and different blood levels, and
and only form B showed adequate bioavailability for the expected therapeutic effect [14].
only form B showed adequate bioavailability for the expected therapeutic effect [14].
Thus, the different solubilities reported for ORBI may have resulted from an inadvertent analysis
Thus, the different solubilities reported for ORBI may have resulted from an inadvertent analysis
of different polymorphs in the powdered material. Solid forms of the same API may have different
of different polymorphs in the powdered material. Solid forms of the same API may have different
physicochemical properties [10,14,15]. The different intermolecular interactions among the various
physicochemical properties [10,14,15]. The different intermolecular interactions among the various
existing solid forms of an API can generate these differences. The solubility is the most critical property
existing solid forms of an API can generate these differences. The solubility is the most critical property
because it can affect the bioavailability and may compromise the development of a bioequivalent
because it can affect the bioavailability and may compromise the development of a bioequivalent
product [10,15–17]. The Food and Drug Administration (FDA) [18] and the International Conference
product [10,15–17]. The Food and Drug Administration (FDA) [18] and the International Conference
on Harmonization (ICH) [19] require preliminary and exhaustive screening studies to identify and
on Harmonization (ICH) [19] require preliminary and exhaustive screening studies to identify and
characterize all the polymorphic forms for each drug to avoid this problem [20].
characterize all the polymorphic forms for each drug to avoid this problem [20].
Therefore, because the polymorphism phenomenon is probable to occur for ORBI, it is very
Therefore, because the polymorphism phenomenon is probable to occur for ORBI, it is very
important to have access to the crystal structures of their different polymorphs and to study whether
important to have access to the crystal structures of their different polymorphs and to study whether
there are differences in their solubilities and other properties. These types of studies can avoid
there are differences in their solubilities and other properties. These types of studies can avoid problems
problems such as toxicity or therapeutic failure [10,12,21].
such as toxicity or therapeutic failure [10,12,21].
Single-crystal and powder X-ray diffraction techniques are the most suitable and commonly
Single-crystal and powder X-ray diffraction techniques are the most suitable and commonly
utilized tools to study and characterize polymorphs in pharmaceutical solids because they provide
utilized tools to study and characterize polymorphs in pharmaceutical solids because they provide
unequivocal proof of polymorphism existence [18]. Following our ongoing research to determine the
unequivocal proof of polymorphism existence [18]. Following our ongoing research to determine
crystal structures of APIs [21–30], we report here for the first time two crystalline structures of ORBI.
the crystal structures of APIs [21–30], we report here for the first time two crystalline structures of
The first ORBI crystal structure, which was determined by single-crystal X-ray diffraction, shows the
ORBI. The first ORBI crystal structure, which was determined by single-crystal X-ray diffraction,
ORBI molecule in a zwitterionic form. Because the crystal is a hemihydrate, this structure is termed
shows the ORBI molecule in a zwitterionic form. Because the crystal is a hemihydrate, this structure
the ORBI hemihydrate form. ORBI hemihydrate was characterized by thermal analysis (TG/DTA),
is termed the ORBI hemihydrate form. ORBI hemihydrate was characterized by thermal analysis
infrared spectroscopy, and powder X-ray diffraction. A second crystalline form, the ORBI anhydrous
form obtained from the hemihydrate form by heating it, is also reported. The solid-state phase
Molecules 2016, 21, 328
3 of 19
(TG/DTA), infrared spectroscopy, and powder X-ray diffraction. A second crystalline form, the ORBI
anhydrous form obtained from the hemihydrate form by heating it, is also reported. The solid-state
phase transformation between the ORBI anhydrous and hemihydrate forms during wetting and drying
is presented. The equilibrium solubility of the hemihydrate and anhydrous forms were compared after
48 h (at 150 rpm and 25 ˝ C) in a physiological pH range (~1 to 7.5) to assess the impact of the crystal
structure differences on solubility.
2. Results and Discussions
2.1. Structure Determination by Single-Crystal X-ray Diffraction
Crystal data, data collection procedures, structure determination methods, and refinement results
for ORBI hemihydrate form are summarized in Table 1.
Table 1. Crystal data and the structures refinement for ORBI hemihydrate form.
Parameter
Value
Empirical Formula
Formula Weight (g¨mol´1 )
Temperature (K)
Wavelength (Å)
Crystal System
Space group
C19 H20 F3 N3 O3 . 1/2H2 O
404.4
150 (2)
0.71073
Orthorhombic
C2221
a = 8.4713(4); b = 20.0456(7);
c = 21.4550(7); α = β = γ = 90
3643.3(2)
8; 1.474
0.123
1688
0.25 ˆ 0.15 ˆ 0.10
1.90 a 29.49
´11ď h ď 11; ´27 ď k ď 25; ´29 ď l ď28
22,842/4627 [R(int) = 0.0281]
100
Full-matrix least-squares on F2
4627/0/267
1.041
R1 = 0.0357, wR2 = 0.0883
R1 = 0.0428, wR2 = 0.0938
0.287 and ´0.202
Cell parameters (Å, ˝ )
Cell Volume (Å3 )
Z; Calc. density (g¨cm´3 )
Absorption coefficient (mm´1 )
F(000)
Crystal size (mm3 )
θ-range (˝ )
Index ranges
Reflexions collected/unique
Completeness to θmax = 29.49 (%)
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
R1/wR2 for I > 2σ(I)
R1/wR2 for all data
Largest diff. peak/hole (e¨Å³)
Figure 2 is an ORTEP-3 view showing the ORBI hemihydrate form asymmetric unit. Three important
structural features should be highlighted before analysing its intra- and intermolecular geometries:
(a)
(b)
The structure is a hemihydrate (one half water molecule for each ORBI molecule), and the water
oxygen atom lies in a special position (lies in the 2-fold axis parallel to the unit cell a axis)
of the C2221 space group. Therefore, the two water hydrogen atoms shown in Figure 2 are
crystallographic dependent by symmetry.
The ORBI molecule is in its zwitterion form with the carboxylic hydrogen transferred to
pyperazine nitrogen. The proton transfer was confirmed by Fourier difference maps, which
show the absence of residual electron density close to the oxygen atoms of the carboxylic group
and the presence of two large peaks of electronic density close to the pyperazine nitrogen, which
clearly reveals the presence of two hydrogens linked to it (Figure 3). Moreover, identical bond
lengths are observed for C1-O1 (1.250(3) Å) and C1-O2 (1.251(2) Å), indicating the formation
of a carboxylate ion (negative charge resulting from deprotonation is delocalized between the
two oxygen atoms).
Molecules 2016, 21, 328
Molecules 2016, 21, 328
Molecules 2016, 21, 328
4 of 19
4 of 19
4 of 19
Figure
ORTEPview
viewofofORBI
ORBIhemihydrate
hemihydrate form
form showing
ellipsoids
at 50%
Figure
2. 2.
ORTEP
showingthe
theatoms
atomsdisplacement
displacement
ellipsoids
at 50%
Figure 2. ORTEP
view
of ORBI hemihydrate
form showing
the atoms
displacement
ellipsoids
at 50%
probability
level.
Non-hydrogen
atoms
are
arbitrarily
labelled
and
hydrogen
atoms
are
shown
as as
probability level. Non-hydrogen atoms are arbitrarily labelled and hydrogen atoms are shown
probability
level. Non-hydrogen
atoms are
arbitrarily
labelled
and are
hydrogen
atoms
are shown
as
spheres
of
arbitrary
radii.
The
protonated
and
deprotonated
moieties
marked
by
dotted
circles
in
spheres of arbitrary radii. The protonated and deprotonated moieties are marked by dotted circles
in
spheres
of
arbitrary
radii.
The
protonated
and
deprotonated
moieties
are
marked
by
dotted
circles
in
blue
and
red,
respectively.The
Themolecules
moleculeswith
with C15(R),C16(S)
C15(R),C16(S) configuration
blue
and
red,
respectively.
configurationwas
wasarbitrarily
arbitrarilychosen.
chosen.
blue and red, respectively. The molecules with C15(R),C16(S) configuration was arbitrarily chosen.
(a)
(a)
(b)
(b)
Figure 3. Residual density maps through the deprotonated (carboxylate) (a) and protonated (piperazine)
Figure
Residual
density
mapsthrough
through
the obtained
deprotonated
(carboxylate)
(a)
Figure
3. 3.
Residual
maps
the
deprotonated
(carboxylate)
(a)and
andprotonated
protonated
(piperazine)
(b)
moieties
fromdensity
ORBI
hemihydrate
form,
by difference
Fourier
synthesis.
Red,(piperazine)
green
and
(b)
moieties
from
ORBI
hemihydrate
form,
obtained
by
difference
Fourier
synthesis.
Red,correspond
green and
lines show
negative, and
zeroobtained
contours by
levels,
respectively.
Contour
levels
(b)blue
moieties
from positive,
ORBI hemihydrate
form,
difference
Fourier
synthesis.
Red,
green and
blue
lines
positive, negative, and zero contours levels, respectively. Contour levels correspond
−3show
to
0.1
e∙Å
intervals.
blue
lines
show
positive, negative, and zero contours levels, respectively. Contour levels correspond to
to 0.1´3
e∙Å−3 intervals.
0.1 e¨ Å intervals.
(c) Since ORBI molecule has two stereocenters (at C15 and C16) and it is crystallized in a
(c) non-centrosymmetric
Since ORBI moleculespace
has two
stereocenters
(at C15 and
and it isincrystallized
in a
group
(Table 1) containing
just C16)
one molecule
the asymmetric
(c) Since
ORBI molecule has two
stereocenters
(at C15
and
C16) andjust
it is one
crystallized
in ain
non-centrosymmetric
non-centrosymmetric
space
group
(Table
1)
containing
molecule
the
asymmetric
unit, its crystal structure is expected to contain a pure enantiomer [31]. However, this is not the
space
(Table
1) containing
just one
molecule
in the
asymmetric
unit,
its crystal is
structure
unit,group
its crystal
is expected
to contain
pure
enantiomer
[31].
However,
the is
case
because
thestructure
piperazine
ring containing
theatwo
stereocenters
has as
internal this
mirrornot
plane
case because
the
ring containing
the
two that
stereocenters
has
as internal
plane
expected
toN2,
contain
a pure
[31]. However,
thisbisects
is not this
the
case
because
the piperazine
(through
N3,piperazine
and
theenantiomer
hydrogens
linked
to N3)
moiety.
The mirror
symmetrical
(through
N2,
N3,
and
the
hydrogens
linked
to
N3)
that
bisects
this
moiety.
The
symmetrical
ring
containing
the
two
stereocenters
has
as
internal
mirror
plane
(through
N2,
N3,
and
nature arising from the presence of two stereogenic centres with identical substitution but oppositethe
nature arising
fromtothe
presence
of two stereogenic
centres
substitution
but opposite
hydrogens
linked
N3)
that
bisects
this moiety.
The with
symmetrical
nature
arising
from
configuration
makes
ORBI
a meso-compound.
As highlighted
byidentical
Hoffmann,
meso-compounds
canthe
configuration
makes
ORBI
a
meso-compound.
As
highlighted
by
Hoffmann,
meso-compounds
can
presence
of two
centres
identical
substitution
opposite
configuration
reach the
samestereogenic
conformation
withwith
a plane
of symmetry
or abut
centre
of inversion
among makes
the
reach
the
same
conformation
with
a
plane
of
symmetry
or
a
centre
of
inversion
among
ORBI
a meso-compound.
As highlighted
by Hoffmann,
meso-compounds
can C15(S),C16(R)
reach the the
same
continuum
of freely accessible
conformations
[32]. The ORBI
C15(R),C16(S) and
continuum
of
freely
accessible
conformations
[32].
The
ORBI
C15(R),C16(S)
and
C15(S),C16(R)
conformation
with a plane
of symmetryon
orits
a centre
ofmirror
inversion
among
the continuum
of freely
isomers are equivalent
(superimposable
internal
image)
considering
the pyperazine
isomers Additionally,
are
equivalentbecause
(superimposable
on
itsand
internal
mirror
image)
considering
theare
pyperazine
accessible
conformations
[32]. The
C15(R),C16(S)
and C15(S),C16(R)
isomers
equivalent
moiety.
the ORBI
quinolone
pyperazine
rings can
freely
rotate
around
the
moiety.bond,
Additionally,
because
the quinolone
and
pyperazine
ringsare
canalso
freely
rotate around
C7-N2
theonwhole
molecules
containing
the
two isomers
equivalent.
ORBIthe
is
(superimposable
its internal
mirror
image) considering
the pyperazine
moiety. Additionally,
C7-N2 bond,
the whole
molecules
containing
the two
twostereogenic
isomers arecentres
also equivalent.
ORBI
is
therefore
optically
inactive;
the
contributions
of
the
to
the
Cotton
effect
because the quinolone and pyperazine rings can freely rotate around the C7-N2 bond, the whole
therefore optically
inactive;
the contributions
of the
two stereogenic
centres
to the
Cotton2effect
compensate
for each
other
Therefore,
C15(R),C16(S)
shown
in Figure
wasthe
molecules
containing
the
two[32].
isomers
are alsothe
equivalent.
ORBI isomer
is therefore
optically
inactive;
compensate
for
each
other
[32].
Therefore,
the
C15(R),C16(S)
isomer
shown
in
Figure
2 was
arbitrarily chosen.
contributions
of the two stereogenic centres to the Cotton effect compensate for each other [32].
arbitrarily chosen.
Therefore, the C15(R),C16(S) isomer shown in Figure 2 was arbitrarily chosen.
Molecules 2016, 21, 328
5 of 19
For the ORBI hemihydrate form, it is observed that its quinolone ring is, as expected, very planar.
The root mean square deviation (r.m.s.d) of the least-square plane thought the two fused-rings atoms
is 0.0623.
The largest
deviations from the calculated least-square plane considering the first
neighbour
Molecules
2016, 21, 328
5 of
19
atoms linked to the quinolone ring occur for O3 (´0.289(3) Å) and C1 (0.295(3) Å). The carboxylate
Fordeviate
the ORBIsignificantly
hemihydrate form,
is observed
that its
quinolone
ring and
is, as 0.638(3)
expected, very
planar.
oxygens also
fromit the
quinolone
ring
(0.167(3)
Å for
O1 and O2,
The root mean square deviation (r.m.s.d) of the least-square plane thought the two fused-rings atoms
respectively). In spite of the overall planar character, there is a slight bend in the quinolone ring along
is 0.0623. The largest deviations from the calculated least-square plane considering the first neighbour
an imaginary
line passing
throughring
C3 and
The(−0.289(3)
least-square
through
containing
atoms linked
to the quinolone
occurN1.
for O3
Å) andplane
C1 (0.295(3)
Å).the
Themoiety
carboxylate
˝
C2, C3, oxygens
C10, and
(r.m.s.d
= 0.0062)
makes
an angle
of (0.167(3)
10.30(7)andrelative
plane
calculated
alsoN1
deviate
significantly
from
the quinolone
ring
0.638(3)to
Å the
for O1
and O2,
throughrespectively).
the remaining
quinolone
two-fused
ring including
and
N1in(r.m.s.d
= 0.0090).
In spite
of the overall
planar character,
there is aC3
slight
bend
the quinolone
ring along
imaginary line
passing
C3 and N1. The least-square
through by
thethe
moiety
containing
Theanpiperazine
ring
has athrough
chair conformation,
which wasplane
confirmed
Evans
and Boeyens
C2,
C3,
C10,
and
N1
(r.m.s.d
=
0.0062)
makes
an
angle
of
10.30(7)°
relative
to
the
plane
calculated
conformational analysis [33]. The N2-C7, C15-C18, C16-C19, C14-H14B, N3-H3b, and C17-H17a bonds
through the remaining quinolone two-fused ring including C3 and N1 (r.m.s.d = 0.0090).
are in the equatorial positions, whereas the remaining bonds are axial.
The piperazine ring has a chair conformation, which was confirmed by the Evans and Boeyens
Theconformational
molecular conformation
wasN2-C7,
analysed
using
MOGUL
[34], a N3-H3b,
knowledge
base derived
analysis [33]. The
C15-C18,
C16-C19,
C14-H14B,
and C17-H17a
bondsfrom the
Cambridge
Database,
CSD
[35], the
which
provides
quick
access to information on the preferred
are inStructural
the equatorial
positions,
whereas
remaining
bonds
are axial.
Theand
molecular
conformation
waschemical
analysed using
MOGUL
a knowledge
derived fromstudy of
length, angle,
torsion
values of the
bonds
in the[34],
structures.
Thisbase
comparative
the
Cambridge
Structural
Database,
CSD
[35],
which
provides
quick
access
to
information
on the
the CSD is
the intramolecular geometry of unpublished crystal structures similar to ones reported in
preferred length, angle, and torsion values of the chemical bonds in the structures. This comparative
useful to check for errors in the refinement or to probe unusual geometries [34]. In the case of ORBI,
study of the intramolecular geometry of unpublished crystal structures similar to ones reported in the
the study
showed that all the bond lengths, bond angles, and torsion angles are in agreement with the
CSD is useful to check for errors in the refinement or to probe unusual geometries [34]. In the case of
expected
values.
ORBI,
the study showed that all the bond lengths, bond angles, and torsion angles are in agreement
Taking
intoexpected
accountvalues.
the intermolecular interactions, a tetramer assembled through intermolecular
with the
Taking
into
account
the intermolecular
interactions,
a tetramer
assembled
through intermolecular
H bonds from molecules related
to 21 - and 2-fold
symmetry
axis
is the building
block for the ORBI
H
bonds
from
molecules
related
to
2
1- and 2-fold symmetry axis is the building block for the ORBI
hemihydrate form (Figure 4). This supramolecular synthon involves four ORBI molecules diametrically
hemihydrate form (Figure 4). This supramolecular synthon involves four ORBI molecules diametrically
oriented (tetramer) with one pair of molecules having the piperazine extremity facing the ring centre
oriented (tetramer) with one pair of molecules having the piperazine extremity facing the ring centre
and another
pair facing
the carboxylate
(Figure
and another
pair facing
the carboxylate
(Figure4).
4).
Figure 4. ORTEP-3 representation of the ORBI tetramer stabilized by hydrogen bonds (dashed line)
Figure 4. ORTEP-3 representation of the ORBI tetramer stabilized by hydrogen bonds (dashed line)
viewed normal to the unit cell a axis. The hydrogen atoms not involved in the depicted hydrogen
viewed bonds
normal
toomitted
the unit
a axis.
The hydrogen
not–zinvolved
the–ydepicted
hydrogen
1/2, y-1/2,
+ 3/2; ii –x +in1/2,
+ 1/2, z + 1/2;
were
forcell
clarity.
Symmetry
codes: i –x +atoms
i
ii
iii x, –y,
bonds were
omitted
–z + 2. for clarity. Symmetry codes: –x + 1/2, y-1/2, –z + 3/2; –x + 1/2, –y + 1/2,
z + 1/2; iii x, –y, –z + 2.
The piperazine nitrogen (N3) acts through H3a as a hydrogen bond donor to O1, and through
H3b as a bifurcated hydrogen bond donor to O2 and O3. Both water hydrogen atoms are hydrogen
Thebonded
piperazine
(N3)
acts
through
H3a as
a hydrogen
bondoxygens.
donor However,
to O1, and
to two nitrogen
diametrical
ORBI
molecules
through
their
O2 carboxylate
thethrough
H3b as water
a bifurcated
hydrogen
bond
donorbond
to O2
and O3.
Both
water
hydrogen
are hydrogen
oxygen does
not act as
hydrogen
acceptor
for any
atom
in this
structure. atoms
The hydrogen
geometry
involved
in the
packing isthrough
given in Table
bonded bond
to two
diametrical
ORBI
molecules
their 2.
O2 carboxylate oxygens. However, the water
oxygen does not act as hydrogen bond acceptor for any atom in this structure. The hydrogen bond
geometry involved in the packing is given in Table 2.
Molecules 2016, 21, 328
6 of 19
Table 2. Hydrogen bonds for ORBI hemihydrate form (Å and ˝ ), where D = donor and A = acceptor.
Molecules 2016, 21, 328
6 of 19
D-H¨ ¨ ¨ A
d(D-H)
d(H¨ ¨ ¨ A)
d(D¨ ¨ ¨ A)
<(D-H-A)
O5ii -H5ii ¨ ¨ ¨ O2ii
0.92(8)
1.98(8)
2.854(3)
159(7)
Table 2. Hydrogen bonds for ORBI hemihydrate form (Å and °), where D = donor and A = acceptor.
D-H···A
d(D-H)
d(H···A)
d(D···A)
<(D-H-A)
0.88(3)
1.86(3)
2.720(2)
164(2)
0.92(8)
1.98(8)
2.854(3)
159(7)
0.86(3)
1.94(3)
2.714(2)
150(2)
0.88(3)
1.86(3)
2.720(2)
164(2)
0.86(3)
2.35(3)
2.955(2)
128(2)
0.86(3)
1.94(3)
2.714(2)
150(2)
0.98
2.42
3.382(3)
168.6
N3-H3b∙∙∙O3
0.86(3)
2.35(3)
2.955(2)
128(2)
d = distance;C11-H11∙∙∙O2
<(D-H-A) = angle.
Symmetry
codes as in
Figures 3 and3.382(3)
6: i –x + 1/2, y-1/2,168.6
–z + 3/2; ii –x + 1/2,
iv
0.98
2.42
iv
N3-H3a¨
¨ ¨ O1i
O5ii-H5ii∙∙∙O2ii ii
N3-H3b¨ ¨ ¨ O2
N3-H3a∙∙∙O1i ii
N3-H3b¨ ¨ ¨ O3
N3-H3b∙∙∙O2ii iv
C11-H11¨ ¨ ¨ O2
ii
-y + 1/2, z + 1/2; ´x, y, ´z + 3/2.
d = distance; <(D-H-A) = angle. Symmetry codes as in Figures 3 and 6: i –x + 1/2, y-1/2, –z + 3/2; ii –x +
1/2, -y + 1/2, z + 1/2; iv −x, y, −z + 3/2.
Each tetramer is surrounded by four others by sharing its ORBI molecules, forming a 2D infinite
Each tetramer is surrounded by four others by sharing its ORBI molecules, forming a 2D infinite
network (layer)
parallel to the (100) plane (Figure 5). The red spheres representing the water molecules
network (layer) parallel to the (100) plane (Figure 5). The red spheres representing the water molecules
in Figurein5Figure
coincide
with the tetramer linkage points. All linkage points contain the hydrogen bonds
5 coincide with the tetramer linkage points. All linkage points contain the hydrogen bonds
discussed
above above
for the
shown
inin
Figure
discussed
fortetramer
the tetramer
shown
Figure4.
4.
5. Partial packing of ORBI hemihydrate form showing a layer formed normal to the unit cell a
Figure 5.Figure
Partial
packing of ORBI hemihydrate form showing a layer formed normal to the unit cell a
axis. The layer was generated from the asymmetric unit applying the symmetry operations included
axis. The layer was generated from the asymmetric unit applying the symmetry operations included in
in Figure 4 captions. The colour scheme refers to four of the eight equivalents by symmetry positions
Figure 4ofcaptions.
colour
scheme
refers to
fourare
of the
eight equivalents
by symmetry
of
the C2221The
space
group.
Water oxygen
atoms
represented
as red spheres.
Hydrogens positions
were
the C222omitted
group.
clarity.Water oxygen atoms are represented as red spheres. Hydrogens were omitted
1 spacefor
for clarity.
The water molecules are connected to the layer by the hydrogen bond involving the O2
carboxylate oxygen (see Figure 4). The layer stacking is governed by the 21-fold screw axis parallel to
Thethe
water
molecules
are connected
to the
layer
by the
bond
involving
the O2 of
carboxylate
unit cell
c axis. Therefore,
the layers
stacked
along
the hydrogen
[100] direction
alternate
a translation
c/2
oxygen (see
Figure
4). The
layer Thus,
stacking
is governed
the 21 -fold
axis
parallel
the
(or b/2)
one over
the other.
the tetramer
centreby
localized
at (x,screw
0, 0), (x,
1, 0),
(x, 0, 1),to(x,
1, unit
1), cell c
and (x, ½,the
½) positions
in the layer
shown
Figuredirection
5 will be displaced
to a
thetranslation
(x, 0, ½) (x, ½,
(x,(or
1, ½),
axis. Therefore,
layers stacked
along
thein[100]
alternate
of0),
c/2
b/2) one
(x, ½,Thus,
1) positions
for the layer
stacked
below or
over theand
other.
the tetramer
centre
localized
atabove
(x, 0,it.0), (x, 1, 0), (x, 0, 1), (x, 1, 1), and (x, ½, ½)
It is clearly observed that the crystallization waters occupy channels parallel to the unit cell a axis
positions in the
layer shown in Figure 5 will be displaced to the (x, 0, ½) (x, ½, 0), (x, 1, ½), and (x, ½, 1)
and coincident with the positions of the tetramers’ centres (also tetramer linkage points). Because the
positionswater
for the
layer stacked below or above it.
oxygen is in the special crystallographic position of the C2221 space group (fraction coordinate
It isx,clearly
observed
crystallization
waters
occupytochannels
parallel
to theitunit
½, ½), the
channelsthat
lies,the
of course,
in the 2-fold
axis parallel
the unit cell
a axis. Later,
will cell
be a axis
and coincident
the strongly
positions
of the
tetramers’ centres
tetramereven
linkage
points).
Because the
discussedwith
that this
layered
supramolecular
structure(also
is maintained
without
the presence
of water
in the channel, and
therefore,
ORBI
hemihydrate/anhydrous
structures
water oxygen
is(ORBI
in theanhydrous)
special crystallographic
position
of the
C222
coordinate
1 space group (fraction
present
zeolite
behaviour.
x, ½, ½), the channels lies, of course, in the 2-fold axis parallel to the unit cell a axis. Later, it will be
The layer stacking is stabilized by non-classical hydrogen bonds involving inter-layer ORBI
discussed that this strongly layered supramolecular structure is maintained even without the presence
molecules. The propyl cycle methine hydrogen works as a hydrogen bond donor to O2 (Figure 6).
of waterTherefore,
(ORBI anhydrous)
in thehydrogen
channel,bond
andacceptor
therefore,
ORBI
hemihydrate/anhydrous
structures
O2 is a trifurcated
because
it is
also an acceptor from the water
present zeolite
behaviour.
oxygen and
piperazine nitrogen (Figure 4). π-aryl-π-aryl interactions with centroids (Ct1…Ct1iv,
= −x, y,stacking
−z + 3/2) separated
by 3.580
also help in thehydrogen
inter-layer stabilization.
Theivlayer
is stabilized
byÅnon-classical
bonds involving inter-layer ORBI
molecules. The propyl cycle methine hydrogen works as a hydrogen bond donor to O2 (Figure 6).
Therefore, O2 is a trifurcated hydrogen bond acceptor because it is also an acceptor from the water
oxygen and piperazine nitrogen (Figure 4). π-aryl-π-aryl interactions with centroids (Ct1 . . . Ct1iv , iv
= ´x, y, ´z + 3/2) separated by 3.580 Å also help in the inter-layer stabilization.
Molecules 2016, 21, 328
Molecules 2016,
2016, 21,
21, 328
328
Molecules
7 of 19
of 19
19
77 of
Figure 6.
6. ORTEP-3
ORTEP-3 view
view of
of the
the inter-layer
inter-layer intermolecular
intermolecular contacts
contacts stabilized
stabilized by
by non-classical
non-classical hydrogen
hydrogen
ORTEP-3
view
Figure
bonds
and
π-aryl-π-aryl
interactions
(dashed
line).
Ct
is
the
aromatic
ring
calculated
centroid.
The
and π-aryl-π-aryl
π-aryl-π-aryl interactions
interactions (dashed
(dashedline).
line).Ct Ct
is the
aromatic
calculated
centroid.
bonds and
is the
aromatic
ring ring
calculated
centroid.
The
hydrogen
atoms
not
involved
in
the
depicted
hydrogen
bonds
were
omitted
for
clarity.
Symmetry
The hydrogen
atoms
involvedininthe
thedepicted
depicted hydrogen
were
omitted
for clarity.
Symmetry
code:
hydrogen
atoms
notnot
involved
hydrogenbonds
bonds
were
omitted
for clarity.
Symmetry
iv −x, y, −z + 3/2.
iv
code:
´x, ivy,−x,
´zy,+ −z
3/2.
code:
+ 3/2.
2.2. ORBI
ORBI Characterization
Characterization
2.2.
2.2.1. Powder
Powder X-ray
X-ray Diffraction
Diffraction
2.2.1.
Figure 77 shows
shows that
that the
the experimental
experimental Powder
Powder X-ray
X-ray Diffraction
Diffraction (PXRD)
(PXRD) patterns
patterns of
of the
the ORBI
ORBI
Figure
standard
(Sigma)
and
the
ORBI
raw
material
are
matched,
indicating
that
the
two
ground
materials
standard (Sigma) and the ORBI raw material are matched, indicating that the two ground materials
crystallize in
inthe
thesame
samecrystal
crystal
phase.
Moreover,
both
patterns
match
the calculated
calculated
PXRD
of
the
crystallize
phase.
Moreover,
both
patterns
match
the calculated
PXRDPXRD
of theof
ORBI
in
the
same
crystal
phase.
Moreover,
both
patterns
match
the
the
ORBI
hemihydrate
form
for
which
crystal
structure
was
determined
herein.
hemihydrate
form for
which
crystalcrystal
structure
was determined
herein.herein.
ORBI
hemihydrate
form
for which
structure
was determined
Normalized
Normalizedintensity
intensity(a.u)
(a.u)
OR
RBBII (S
(Sigm
igmaa standard)
standard)
O
OR
RBBII (raw
(raw m
material)
aterial)
O
OR
RBBII (calc.)
(calc.)
O
4
4
6
6
8
8
10
10
12
12
14
14
16
16
18
18
20
20
22
22
24
24
26
26
28
28
30
30
32
32
34
34
degree (C
(Cuk
ukαα))
22θθ degree
Figure 7. Experimental
Experimental and calculated
calculated PXRD patterns
patterns of ORBI.
ORBI. The calculated
calculated pattern is
is in blue,
blue, and the
the
Figure
Figure 7.
7. Experimentaland
and calculatedPXRD
PXRD patternsof
of ORBI. The
The calculated pattern
pattern is in
in blue, and
and the
experimental
patterns
for
the
raw
material
and
the
Sigma
standard
are
in
red
and
black,
respectively.
experimental
respectively.
experimental patterns
patternsfor
forthe
theraw
rawmaterial
materialand
andthe
theSigma
Sigmastandard
standardare
areininred
redand
andblack,
black,
respectively.
Because there
there were
were no
no pronounced
pronounced broad
broad humps
humps from
from amorphous
amorphous solids
solids or
or extra
extra Bragg
Bragg
Because
Because
therecrystalline
were no pronounced
broad
humps
from
amorphous
solids or form,
extra Bragg
reflections
reflections
from
phases
other
than
that
of
the
ORBI
hemihydrate
upon
observing
reflections from crystalline phases other than that of the ORBI hemihydrate form, upon observing
from crystalline phases other than that of the ORBI hemihydrate form, upon observing the experimental
the experimental
experimental patterns
patterns itit was
was possible
possible to
to say
say that
that the
the Sigma
Sigma ORBI
ORBI standard
standard and
and the
the ORBI
ORBI raw
raw
the
patterns it was possible to say that the Sigma ORBI standard and the ORBI raw material are formed in
material
are
formed
in
bulk
by
the
hemihydrate
phase
of
ORBI
determined
here.
material
arehemihydrate
formed in bulk
by the
hemihydrate
phase
of ORBI determined here.
bulk by the
phase
of ORBI
determined
here.
Molecules 2016, 21, 328
8 of 19
Molecules 2016, 21, 328
8 of 19
2.2.2. Thermal Analysis
2.2.2. Thermal Analysis
The TG/DTA curves recorded for the Sigma ORBI standard show, as expected, an initial mass loss
The TG/DTA curves recorded for the Sigma ORBI standard show, as expected, an initial mass
corresponding to the dehydration (0.5 mol of water per mol of ORBI) of the sample that was shown to
loss corresponding to the dehydration (0.5 mol of water
per mol of ORBI) of the sample that was
be a hemihydrate by the diffraction techniques (Ti = 75 ˝ C; Tf = 104 ˝ C; ∆mTG = 2.1%/∆mcalc . = 2.23%)
shown to be a hemihydrate by the diffraction techniques (Ti = 75 °C; Tf = 104 °C; ΔmTG = 2.1%/Δmcalc.
(Figure
8). Therefore,
the molarthe
ratio
of ratio
1:2 for
water:ORBI
waswas
confirmed
material by
= 2.23%)
(Figure 8). Therefore,
molar
of 1:2
for water:ORBI
confirmedininthe
the raw
raw material
˝ C, at which point
TG/DTA
analysis.
The
dehydrated
product
(ORBI
anhydrous)
is
stable
up
to
~260
by TG/DTA analysis. The dehydrated product (ORBI anhydrous) is stable up to ~260 °C, at which
˝ C).
its thermal
decomposition
begins (DTA
269 at
point its
thermal decomposition
beginspeak
(DTAatpeak
269 °C).
100
3.0
104°C = 97.9% ~ 0.5 H2O
95
2.5
TG (mass %)
-1
DTA (μV.mg )
90
2.0
85
80
1.5
75
1.0
269°C
70
50
100
150
200
250
Temperature (°C)
300
350
400
Figure 8. TG (in black) and DTA (in blue) curves measured from 40 to 400 °C starting with ORBI
Figure 8. TG (in black) and DTA (in blue) curves measured from 40 to 400 ˝ C starting with ORBI
hemihydrate. The arrow shows the temperature of the complete water loss from the structure.
hemihydrate. The arrow shows the temperature of the complete water loss from the structure.
2.2.3. Infrared Spectroscopy with Attenuated Total Reflectance by Fourier Transform (FTIR-ATR)
2.2.3. Infrared Spectroscopy with Attenuated Total Reflectance by Fourier Transform (FTIR-ATR)
Considering that TG/DTA results obtained here indicated two different crystalline phases of
ORBI (hemihydrate
and anhydrous),
measured
the Sigma
ORBI standard
Considering
that TG/DTA
resultsFTIR-ATR
obtained data
herewere
indicated
twofor
different
crystalline
phases of
samples
without
(as
received)
and
with
ex-situ
thermal
treatment
(obtained
by
heating
10
mg
of
Sigma
ORBI (hemihydrate and anhydrous), FTIR-ATR data were measured for the Sigma ORBI standard
ORBI
standard
for 2and
h atwith
105 °C)
(Figure
S1—Supplementary
Materials).
samples
without
(assample
received)
ex-situ
thermal
treatment (obtained
by heating 10 mg of Sigma
The
characteristic
bands
of
ORBI
were
assigned
based
on
that
one’s
attributed to similar
˝
ORBI standard sample for 2 h at 105 C) (Figure S1—Supplementary Materials).
−1
quinolones [36]. The band in the region of 3446 cm refers to the -NH vibrations of the piperazinic
The characteristic bands of ORBI were assigned
based on that one’s attributed to similar
−1 can be assigned to axial deformations of the C-N
ring, whereas the bands between 3018 and 2763 cm
´1 refers
quinolones
[36].
The
band
in
the
region
of
3446
cm
to the -NH vibrations of the piperazinic ring,
bond of piperazine. The intense band in the region of 1700 cm−1 corresponds to the C=O group of the
´
1
whereas
the bands
3018
and
2763
be assigned
to axial
deformations
of the C-N
bond of
−1 shows
carboxylic
acid.between
The region
from
1050
to cm
1000 cmcan
vibrations
related
to axial deformations
of C–F
´1 corresponds to the C=O group of the carboxylic
−1
piperazine.
The
intense
band
in
the
region
of
1700
cm
groups. Vibrations below 1000 cm refer to the angular deformations out of plane of aromatic –CH.
−1,
acid. The The
region
fromsignificant
1050 to 1000
cm´1 shows
vibrations
relatedoccurs
to axial
deformations of
C–F
unique
difference
between
the two spectra
at approximately
3340
cmgroups.
´
1
whichbelow
is the region
of water
axial
deformation.
As expected, an
attenuation
of aromatic
this broad–CH.
band for
Vibrations
1000 cm
refer
to the
angular deformations
out
of plane of
the ORBI
anhydrous
formdifference
was observed.
The
unique
significant
between the two spectra occurs at approximately 3340 cm´1 ,
which is the region of water axial deformation. As expected, an attenuation of this broad band for the
2.3. Interconversion Study between ORBI Anhydrous and Hemihydrate Forms
ORBI anhydrous form was observed.
The TG curves of the Sigma ORBI standard as received (ORBI hemihydrate form), ORBI standard
2.3. Interconversion
Studyanhydrous
between ORBI
Hemihydrate
Formsfollowed by treatment in a
after heating (ORBI
form),Anhydrous
and ORBI and
Standard
after heating
climatic chamber (rehydrated form) were compared (Figure S2—Supplementary Materials). The TG
The TG curves of the Sigma ORBI standard as received (ORBI hemihydrate form), ORBI standard
curve of the ex-situ heated sample at 200 °C for 10h does not show mass loss in the range observed
after for
heating
(ORBI
anhydrous
andthe
ORBI
followed
by treatment
the ORBI
hemihydrate
form.form),
Therefore,
ORBIStandard
anhydrousafter
formheating
can be obtained,
as expected,
in a climatic
chamber
(rehydrated
form)
were
compared
(Figure
S2—Supplementary
Materials).
by heating the hemihydrate form in the 104–269 °C range. Interestingly, the TG curve of a previously
˝
The TG
curve ofORBI
the ex-situ
sample
200 Cchamber
for 10 hatdoes
mass loss
in theforrange
dehydrated
sampleheated
introduced
in theatclimatic
40 °Cnot
andshow
75% relative
humidity
1 h match
that
of the
Sigma ORBIform.
Standard
(hemihydrate
form).
Therefore,form
the ORBI
anhydrous
observed
for the
ORBI
hemihydrate
Therefore,
the ORBI
anhydrous
can be
obtained, as
˝
expected, by heating the hemihydrate form in the 104–269 C range. Interestingly, the TG curve of
a previously dehydrated ORBI sample introduced in the climatic chamber at 40 ˝ C and 75% relative
Molecules 2016, 21, 328
9 of 19
humidity for 1 h match that of the Sigma ORBI Standard (hemihydrate form). Therefore, the ORBI
Molecules 2016, 21, 328
9 of 19
anhydrous form can be easily returned to the hemihydrate one by exposing the former to high
form can be easily
returned to the hemihydrate one by exposing the former to high temperature/
temperature/humidity
conditions.
conditions.
The humidity
experimental
PXRD pattern of the sample heated ex-situ at 200 ˝ C retains the Bragg peaks of
The experimental PXRD pattern of the sample heated ex-situ at 200 °C retains the Bragg peaks of
the ORBI hemihydrate
form (Figure 9c). This result confirms that no crystallographic phase transition
the ORBI hemihydrate form (Figure 9c). This result confirms that no crystallographic phase transition
occurs when
ORBI
hemihydrate
form
water
released
to generate
the ORBI
occursthe
when
the ORBI
hemihydrate
formhas
has its
its structural
structural water
released
to generate
the ORBI
anhydrous
form.
Thus,
the
hemihydrate
and
anhydrous
forms
are
isomorphous
crystals
[37].
anhydrous form. Thus, the hemihydrate and anhydrous forms are isomorphous crystals [37].
Normalized Intensity (a.u.)
(d)
(c)
(b)
021
002
111
022
(a)
020
113
023
10
12
131 041
130
004
8
202
025
200
133
112
110
040
14
16
18
2θ degree (Cuk α)
042
132
024
114
20
201
043
22
203
222
044
220
221
115
24
Figure 9. PXRD patterns of ORBI: (a) Calculated from the ORBI hemihydrate form (the Miller indices
Figure 9. PXRD patterns of ORBI: (a) Calculated from the ORBI hemihydrate form (the Miller indices
identifying the Bragg peaks are included); (b) experimental PXRD patterns of Sigma ORBI Standard
identifying
the Bragg
peaks
are included);
experimental
patterns
of heating
Sigma at
ORBI
Standard
as received
(ORBI
hemihydrate
form); (c)(b)
experimental
SigmaPXRD
ORBI Standard
after
200 °C
as received
(ORBI
hemihydrate
form);
(c)
experimental
Sigma
ORBI
Standard
after
heating
at 200 ˝ C
for 10 h (ORBI anhydrous form); (d) Sigma ORBI Standard after heating at 200 °C for 10 h followed
˝
by
treatment
in
a
climatic
chamber
at
40
°C
with
75%
relative
humidity
during
1
h
(rehydrated
form
for 10 h (ORBI anhydrous form); (d) Sigma ORBI Standard after heating at 200 C for 10 h followed
˝ C withwere
= ORBIinhemihydrate
All thermal
by treatment
a climaticform).
chamber
at 40treatments
75%ex-situ.
relative humidity during 1 h (rehydrated
form = ORBI hemihydrate form). All thermal treatments were ex-situ.
Although it does not elicit a crystallographic phase transition, the water release causes structural
distortions. An evident peak broadening is observed for some hkl Bragg peaks. Specifically, the hkl
Bragg peaks
with
h ≠elicit
0 (e.g.,
110, 111, 112, 113) are
broader
than thosethe
with
h = 0 release
(e.g., 002,causes
021, 020,
Although
it does
not
a crystallographic
phase
transition,
water
structural
023),An
suggesting
ORBI
hemihydrate
form dehydration
is followed
by an
anisotropic
loss of the hkl
distortions.
evidentthat
peak
broadening
is observed
for some
hkl Bragg
peaks.
Specifically,
crystallinity along the unit cell a axis.
Bragg peaks with h ­“ 0 (e.g., 110, 111, 112, 113) are broader than those with h = 0 (e.g., 002, 021,
The dehydration process keeping the crystal structure can be explained considering the packing
020, 023),ofsuggesting
that ORBI form.
hemihydrate
form
dehydration
is followed
by an of
anisotropic
the ORBI hemihydrate
It should be
remembered
that the
water molecules
the ORBI loss of
crystallinity
along the
unit
a axis.
hemihydrate
form
lie cell
in the
channels and are linked to the ORBI molecules only as hydrogen bond
These structural
features in
the very can
stable
structures
formedthe
by packing
The donors.
dehydration
process keeping
theaddition
crystaltostructure
beintralayer
explained
considering
supramolecular
tetramers
of ORBI
molecules
probably responsible
for the
ORBImolecules
water desorption
of the ORBI
hemihydrate
form.
It should
beare
remembered
that the
water
of the ORBI
not disrupting the crystal structure. It is important to emphasize that the void volume occupied by the
hemihydrate form lie in the channels and are linked to the ORBI molecules only as hydrogen bond
water molecule in the ORBI hemihydrate form is 1.7% (calculated by MERCURY (see Experimental
donors. Section
These3.3)
structural
features
in aaddition
to the
very
stable
intralayer
structures
formed by
using a grid
spacing and
probing-sphere
radius
of 0.7
and 1.1
Å, respectively).
It is normal
supramolecular
tetramers
of
ORBI
molecules
are
probably
responsible
for
the
ORBI
water
that channel solvates with void volumes less than 10% do not collapse on solvent removal [38]. desorption
The
intralayer
explains
small widening
of the Bragg
peaks
indices observed
not disrupting
the
crystalstability
structure.
It is the
important
to emphasize
that
the with
void0kl
volume
occupied by the
in
the
experimental
PXRD
pattern
of
the
ORBI
anhydrous
form
(Figure
9c).
The
water
presence
water molecule in the ORBI hemihydrate form is 1.7% (calculated by MERCURY (seeappears
Experimental
to be more important to the interlayer stability, working as an intercalation guest in the layered
Section 3.3) using a grid spacing and a probing-sphere radius of 0.7 and 1.1 Å, respectively). It is
ORBI host structure. This explains the preferential broadening of the Bragg peaks depending on the
normal that
with
void
volumes less than 10% do not collapse on solvent removal [38].
unitchannel
cell a axissolvates
(hkl indices
with
h ≠ 0).
The intralayer stability explains the small widening of the Bragg peaks with 0kl indices observed
in the experimental PXRD pattern of the ORBI anhydrous form (Figure 9c). The water presence appears
to be more important to the interlayer stability, working as an intercalation guest in the layered ORBI
host structure. This explains the preferential broadening of the Bragg peaks depending on the unit cell
a axis (hkl indices with h ­“ 0).
Molecules 2016, 21, 328
10 of 19
Molecules
2016, 21,analysis
328
10 of 19
The PXRD
corroborates the hydration of the ORBI anhydrous form in the climatic
chamber. Moreover, they show that rehydrated sample returns to the ORBI hemihydrate form.
Thepeaks
PXRDobserved
analysis in
corroborates
the hydration
of the ORBI
anhydrous
form
in the
climatic
The broad
the PXRD pattern
of the anhydrous
sample
become
narrow
again
in the
chamber.
Moreover,
they
show
that
rehydrated
sample
returns
to
the
ORBI
hemihydrate
form.
The
PXRD pattern of the rehydrated sample. Thus, the ORBI porous structure also favours the inverse
broad peaks observed in the PXRD pattern of the anhydrous sample become narrow again in the
process of water adsorption not disrupting the crystal structure. Therefore, the ORBI crystal structure
PXRD pattern of the rehydrated sample. Thus, the ORBI porous structure also favours the inverse
can be considered a nanoporous API with adsorption/desorption features.
process of water adsorption not disrupting the crystal structure. Therefore, the ORBI crystal structure
It is important to emphasize that water adsorption/desorption of crystalline materials may
can be considered a nanoporous API with adsorption/desorption features.
be of three types; the crystal structure can undergo no or little change, a reversible change, or
It is important to emphasize that water adsorption/desorption of crystalline materials may be of
an irreversible change. Water adsorption/desorption without crystal structure changes is well reported
three types; the crystal structure can undergo no or little change, a reversible change, or an irreversible
for change.
zeolite materials
[39]. Similarly, desolvated
solvated
APIschanges
may have
the same
crystal
Water adsorption/desorption
withoutand
crystal
structure
is well
reported
for structure
zeolite
with
relatively
small
changes
in
cell
parameters
and
atomic
coordinates
[40].
When
the
structure
materials [39]. Similarly, desolvated and solvated APIs may have the same crystal structure with is
maintained
waterin
adsorption/desorption
it is usually
more
difficult
distinguish
relativelyduring
small the
changes
cell parameters and processes,
atomic coordinates
[40].
When
the to
structure
is
between
hydrated
and
anhydrous
forms
by
PXRD
[40].
Fortunately,
the
hemihydrated
and
anhydrous
maintained during the water adsorption/desorption processes, it is usually more difficult to distinguish
forms
of ORBI
can beand
distinguished
here by
comparing
their
peak widths
(hkl indices
between
hydrated
anhydrous forms
byPXRD
PXRD by
[40].
Fortunately,
theBragg
hemihydrated
and anhydrous
with
h ­“ 0).
forms
of ORBI can be distinguished here by PXRD by comparing their Bragg peak widths (hkl
indices with h ≠ 0).
2.4. Characterization of ORBI Tablets
2.4.
Characterization
ORBIground
Tablets and analysed by PXRD (Figure S3—Supplementary Materials).
ORBAX™
tabletsofwere
Although
the excipient
(mass)and
is higher
thanbythe
API (Figure
mass inS3—Supplementary
the studied tablets Materials).
(the average
ORBAX™
tabletsamount
were ground
analysed
PXRD
Although
excipientwas
amount
is higher
thanofthe
API mass inand
the studied
(the average
weight
(n = the
20 tablets)
296.7(mass)
mg, i.e.,
22.7 mg
orbifloxacin
274 mgtablets
of excipient)
and it
weightcrystalline
(n = 20 tablets)
was 296.7
mg,
22.7 mg
of orbifloxacin
andpossible
274 mg to
of identify
excipient)
and
it
contains
compounds
such
asi.e.,
lactose
monohydrate,
it was
the
Bragg
contains
compounds
such ashere.
lactose
monohydrate,
it was
possible
Bragg
peaks
of thecrystalline
ORBI structure
determined
Although
the Bragg
peaks
with to
hklidentify
indicesthe
with
h ­“ 0
peakstoofhave
the ORBI
structure
determined
Although
the Bragg peaks
hkl indices
h≠0
appear
a width
compatible
withhere.
the ORBI
hemihydrate
form, with
the presence
ofwith
the phases
appear
to
have
a
width
compatible
with
the
ORBI
hemihydrate
form,
the
presence
of
the
phases
corresponding to excipient preclude the definitive assignment of the API as hemihydrate or anhydrous
corresponding
to of
excipient
the definitive assignment of the API as hemihydrate or anhydrous
forms
or a mixture
the twopreclude
in the tablets.
forms
or
a
mixture
of
the
two
in
the
tablets.
Unfortunately, the TG analysis and
FTIR spectroscopy, which were useful to distinguish between
Unfortunately, the TG analysis and FTIR spectroscopy, which were useful to distinguish between
the two ORBI forms in the ORBI raw material, could not be applied in the ORBAX™ case due to
the two ORBI forms in the ORBI raw material, could not be applied in the ORBAX™ case due to the
the relative amount of ORBI compared to the excipient and the presence of hydrated crystalline
relative amount of ORBI compared to the excipient and the presence of hydrated crystalline components
components in excipients (e.g., lactose monohydrate). In TG curves for example, the calculated
in excipients (e.g., lactose monohydrate). In TG curves for example, the calculated percentage of
percentage of water loss from the ORBI in the tablets (excipient + API) is only 0.16%, which is very
water loss from the ORBI in the tablets (excipient + API) is only 0.16%, which is very difficult to
difficult
to detect confidently by this technique.
detect confidently by this technique.
2.5. Solubility Studies of Hemihydrated and Anhydrous Forms of ORBI
2.5. Solubility Studies of Hemihydrated and Anhydrous Forms of ORBI
The
ORBI molecule has two acid dissociation constants: pKa1 = 5.6 for the 3-carboxylic
The ORBI molecule has two acid dissociation constants: pKa1 = 5.6 for the 3-carboxylic group,
group,
and
8.9the
for3,5-dimethylpiperazinyl
the 3,5-dimethylpiperazinyl
[7], as shown
in Scheme
1. As
such,
and pKa2pKa2
= 8.9 =for
group group
[7], as shown
in Scheme
1. As such,
three
three
pH-dependent
acid-base
species
are
possible
in
aqueous
media,
and
depending
on
the
pH,
ORBI
pH-dependent acid-base species are possible in aqueous media, and depending on the pH, ORBI can
canbe
becrystallized
crystallized
a neutral
zwitterionic
form
oranionic
as anionic
or cationic
asas
a neutral
zwitterionic
form
or as
or cationic
salts.salts.
Scheme 1. ORBI acid dissociation constants.
Scheme 1. ORBI acid dissociation constants.
It is important to emphasize that the ORBI molecule is zwitterionic in the two solid-state forms
described
in this work
and anhydrous
Additionally,
considering
the pH-dependent
It is important
to (hemihydrate
emphasize that
the ORBI forms).
molecule
is zwitterionic
in the
two solid-state
, /
acid-basic
ORBI
species
distribution,
the
zwitterionic
(
)
species
is
expected
to be the
forms described in this work (hemihydrate and anhydrous (forms). )Additionally, considering
the
predominant acid-basic
one in the pH
range
between
pKa1 and the
pKa2
(Figure 10). (H1 ORBI 0,`{´
pH-dependent
ORBI
species
distribution,
zwitterionic
) species is
pzwitterionq
expected to be the predominant one in the pH range between pKa1 and pKa2 (Figure 10).
Molecules 2016, 21, 328
11 of 19
Molecules 2016, 21, 328
11 of 19
Molecules 2016, 21, 328
11 of 19
Figure 10. pH-dependent acid-basic ORBI species distribution. The molar fraction (α) of each species
Figure 10. pH-dependent acid-basic
ORBI species
The molar fraction (α) of each species
ı )distribution.
was calculated as follows:”
/ ` (1 (
/ “ )‰˘
/
(
“)) ` ‰; 2 ´ 1
` (
,
/
was calculated as follows: α2 “
H2 ORBI
{C “ 1 ` pKa/1 { H ` ( ` pKa1/ˆ Ka);
{ where
H
qq ;
2
/
(
/
);
pcationq
)
(
”
ı )
”(
ı
´
“ `‰
“ ` ‰2 ¯
, /
0,`{´ 1 and
´
.
Curves
show
pH
α1 “ H1 ORBIpzwitterionq {C “ α2 pKa
q; α0 “( ORBI)panionq {C
“) α2 Ka1 ˆ Ka1 { H
;
( 1{ H
)
(
”
ı ”
ı ”
ı
in steps of 0.1.
0,`{´
`
´
whereFigure
α0 `10.
α1pH-dependent
` α2 “ 1acid-basic
and C ORBI
“ H
`The
H1molar
ORBIfraction
ORBI
.
2 ORBI
species
distribution.
(α) of`
each
species
pcationq
pzwitterionq
panionq
Curves
show
pH
in
steps
of
0.1.
was
calculatedthe
as ORBI
follows:
(1 forms’
(
/ solubility
/
)) range
;
) ( in a neutral
Therefore,
hemihydrate and
anhydrous
pH
(
) / , /
( to/ be lower
/ acid
( or alkaline
/ pH);(cationic
where or
);
(zwitterion prevalence)
than that
( at )either
(
)is/ expected
, /
anionic prevalence)
extremes.
In accord
with
these expectations,
the
lowest
values
of
equilibrium
1 and
.
Curves
show
pH pH range
(
)
(
)
Therefore,
the ORBI
hemihydrate
and
anhydrous
forms’
solubility
in
a
neutral
(
)
solubility
were
observed in the neutral media, including water (pH = 7) and pH 5.8, 6.8, and 7.5
in
steps
of
0.1.
(zwitterion prevalence) is expected to be lower than that at either acid or alkaline pH (cationic or
buffers, for both the hemihydrate and anhydrous ORBI forms (Table 3 and Figure 11).
anionic prevalence) extremes. In accord with these expectations, the lowest values of equilibrium
Therefore, the ORBI hemihydrate and anhydrous forms’ solubility in a neutral pH range
solubility
were
inclassification
the
neutral to
media,
including
water
(pH = after
7) and
5.8,
6.8,(cationic
and
buffers,
Tableobserved
3. Solubility,
final
pH
of solutions
equilibrium
hpH
at 150
rpm
25 °C.7.5or
(zwitterion
prevalence)
is expectedand
be lower
than thatat at
either acid
or48alkaline
pH at
for both
the
hemihydrate
and
anhydrous
ORBI
forms
(Table
3
and
Figure
11).
Anhydrous
SD
Anhydrous/
Hemihydrate
Final
anionic prevalence)
extremes.SD
In accord with these
expectations,
the lowest values
of equilibrium
Media
Solubility *
Classification *
(g·L neutral
)
(g·L )
(g·L
) (pH = 7) andHemihydrate
(g·L ) in the
(%)and
pH 7.5
solubility were observed
media, including
water
pH 5.8, 6.8,
Water
1.32
0.0678
SS
1.38
0.06634
SS
4.38 rpm at 7.0
Table
3.
Solubility,
classification
and
final
pH
of
solutions
at
equilibrium
after
48
h
at
150
25 ˝ C.
buffers,
for both the 1.58
hemihydrate
and anhydrous
ORBI
forms
(Table 3 and
Figure 11).
pH 7.5
0.0644
SS
1.64
0.05327
SS
3.80
7.5
pH 6.8
Media
−1
−1
1.52
0.039
−1
SS
−1
1.59
0.07758
SS
†
4.60
7.0
Hemihydrate
SD
Anhydrous1.98 SD
Anhydrous/
Table
3. Solubility,1.73
classification
at equilibrium
after
rpm at 25 °C.
pH 5.8
0.0601and final
0.07915
SS 48
6.0
Solubility
* SSpH of solutions
Classification
* h at 15014.45
Final pH †
(g¨ L´1 )
(g¨ L´1 )
(g¨ L´1 )
(g¨ L´1 )
-1
Solubility (g.L )
Hemihydrate
(%) 6.0
6.30
0.1739
SS
10.33
0.29854
SP
63.97
HCl 0.01 mol∙L−1
SD
SD
Anhydrous/
Hemihydrate
Final
pH 1.32
4.5
10.41
0.5007
SP *1.38Anhydrous
14.240.06634
0.5539
36.79
4.5 7.0
Water Media
0.0678
SS Solubility
SS SP
4.38
Classification
*
−1)
−1)
−1)
−1)
†
(g·L
(g·L
Hemihydrate
22.24
0.7506
SP 1.64 (g·L
26.23
1.0623
17.94
2.0 7.5
mol∙L−1 (g·L0.0644
pH 7.5 HCl 0.1 1.58
SS
0.05327
SS SP
3.80 (%) pH
1.32
0.0678 SS
SS
1.38 0.07758
0.06634
SS
4.38
7.0 7.0
pH 6.8 Water
1.52
0.039
1.59
SS
4.60
* Legend: SS = slightly soluble; SP = sparingly soluble. Source: The United States Pharmacopeia, 38 ed.,
1.58
SS
0.05327
3.80
7.5 6.0
pH 5.8 pH 7.51.73†
0.0601 0.0644 SS
1.98 1.64 0.07915
SS SS
14.45
2015; The final pH was invariant considering the two solids forms in a same media.
1.52
0.039
SS
1.59
0.07758
SS
4.60
7.0
HCl 0.01 pH 6.8
6.30
0.1739
SS
10.33
0.29854
SP
63.97
6.0
mol¨ L´1 pH 5.8
1.73
0.0601
SS
1.98
0.07915
SS
14.45
6.0
30
−1
pH 4.5
10.41
0.5007
SP
14.24
0.5539
SP
36.79
4.5
6.30
0.1739
SS
10.33
0.29854
SP
63.97
6.0
HCl 0.01 mol∙L
25
HCl 0.1 pH 4.5
10.41
0.5007 SP
SP
0.5539
36.79
4.5 2.0
20 1.0623
22.24
0.7506 25
26.23 14.24
SPSP
17.94
´1
mol¨ LHCl 0.1 mol∙L−1
22.24
0.7506
SP
26.23
1.0623
SP
17.94
2.0
15
-1
Solubility (g.L )
20 SP
* Legend:
SS =SS
slightly
soluble;
sparingly
soluble.
Source:
TheThe
United
States
Pharmacopeia,
3838
ed.,
2015; †
2.0=SP
* Legend:
= slightly
soluble;
= sparingly
soluble.
Source:
United
States
Pharmacopeia,
ed.,
10
The final
pH
was
invariant
considering
the
two
solids
forms
in
a
same
media.
invariant considering the two5 solids forms in a same media.
2015; † The final pH was 15
0
2
-1
Solubility (g.L )
15
1M
l 0.
HC
10
6.0
ORBI Hemihydrate
ORBI anhydrous
4.5
pH
01
l 0.
HC
4.5
5
M
-1
25 5
20 02.0
3
4
5
Final pH
25
4.5
Solubility (g.L )
3010
6
7
8
20
15
10
6.0
7.0
5
5.8
6.8
pH
pH 0
2
Dissolution Media
3
7.5
pH
ter
Wa
7.5
4
5
Final pH
7.0
6
7
8
6.0
Figure 11. Equilibrium solubility
values for ORBI anhydrous and hemihydrate forms at 25 °C, 150 rpm,
ORBI Hemihydrate
0
6.0 0.039 7.0
= 3). The inset shows
48 h. Error bars represent
standard
deviations (among
and 0.7517.5g∙L−1, n 7.0
ORBI anhydrous
the correlation including the linear regression (blue lines) between the equilibrium solubility values
M
.8
ter
7.5 and anhydrous
6.8
4.5
1M
H 5 ORBIpHhemihydrate
with the final pH ofCthe
forpthe
forms.
.01
Wa
pH
pH usedl 0media
l 0. different
H
HC
Dissolution Media
˝
FigureFigure
11. Equilibrium
solubility
values
anhydrousand
and
hemihydrate
forms
150 rpm,
11. Equilibrium
solubility
valuesfor
for ORBI
ORBI anhydrous
hemihydrate
forms
at 25 at
°C,25
150C,
rpm,
n =, 3).
48 h. Error
represent
standard
deviations(among
(among 0.039
0.039 and
48 h. Error
bars bars
represent
standard
deviations
and0.751
0.751g∙L
g¨−1L, ´1
n =The
3). inset
The shows
inset shows
the correlation
including
linear
regression (blue
(blue lines)
thethe
equilibrium
solubility
valuesvalues
the correlation
including
thethe
linear
regression
lines)between
between
equilibrium
solubility
with
the
final
pH
of
the
different
used
media
for
the
ORBI
hemihydrate
and
anhydrous
forms.
with the final pH of the different used media for the ORBI hemihydrate and anhydrous forms.
Molecules 2016, 21, 328
12 of 19
Also as expected, considering the pH-dependent acid-basic ORBI species distribution in solution,
the equilibrium solubility values of both ORBI crystal forms increase significantly in aqueous
hydrochloric acid and in pH 4.5 buffered media. For instance, the equilibrium solubility of the
anhydrous ORBI form is 19-fold greater in HCl 0.1 mol¨ L´1 than in water (Table 3 and Figure 11).
Moreover, Figure 11 shows that there is a linear correlation of the equilibrium solubility for both
forms with the final pH of the used media. The pH-solubility profile for the two ORBI forms studied here
0,`{´
increasing as the pH decreases is due to the protonation (H1ORBIpzwitterionq + H+ Ñ H2ORBIp`cationq ) of
ORBI molecules in solution. Thus, the lower the acid pH, the higher the molar fraction of H2ORBIp`cationq
and consequently the higher the ORBI solubility, whatever the solid form in equilibrium.
In all media for which the equilibrium final pH is greater than pKa1 (except to 0.01 mol¨ L´1 HCl
medium), both ORBI anhydrous and hemihydrate forms are classified as slightly soluble (Table 3)
by USP criteria [9]. This means that 1 g of API is soluble in 100 to 1000 mL of solvent. As discussed
above, the low solubility in this neutral media is due to the prevalence of the zwitterionic species in
the equilibrium solution. In the 0.01 mol¨ L´1 HCl medium, for which the final pH was increased to
~6 during the solubility experiment, the hemihydrate form was classified as slightly soluble (1 g of
API is soluble in 30 to 100 mL of test solvent), whereas the anhydrous form was sparingly soluble.
The increased final pH of the medium using 0.01 mol¨ L´1 HCl can be attributed to the ORBI buffering
0,`{´
behaviour involving the H2ORBIp`cationq and H1ORBIpzwitterionq species released from the solid forms,
which was strong enough to neutralize the introduced HCl solution. In the two media for which the
equilibrium final pH is lower than pKa1 (buffer pH = 4.5 (final pH = 4.5) and 0.1 mol¨ L´1 HCl (final
pH = 2)), both the ORBI anhydrous and hemihydrate forms are classified as sparingly soluble (Table 3).
A remarkable result is that the anhydrous form is significantly more soluble than the hemihydrate
in all acidic media (final pH ď 6). The largest percentage difference (64%) occurs in 0.01 mol¨ L´1 HCl
medium, which makes the solubility classification slightly soluble and sparingly soluble for the ORBI
hemihydrate and anhydrous forms, respectively.
The highest solubility values observed for the anhydrous form compared to the hemihydrate in
the acid media tested here could be attributed to the structural features of the two compared phases.
Considering that hemihydrate and anhydrous forms have the same structure, the latter is expected
to present a higher solubility since it is more porous and has a lower lattice energy. However, this
comparison is valid only when there is no conversion of the solid form in equilibrium with the solution
to a more thermodynamically stable one during the experiment. Because it was shown in Section 2.3
that the anhydrous form can be easily converted to the hemihydrate under wet conditions, the solid
material in equilibrium with the solutions used to determine the ORBI equilibrium solubility was
subsequently dried in a desiccator containing silica for 5 days and analysed by PXRD (Figures 12
and 13 and the Figure S4 in the Supplementary Materials).
All experimental PXRD patterns of the residual solid materials in equilibrium with the solution
prepared from the ORBI hemihydrate and anhydrous forms in water and in the pH 6.8 and 7.5
buffer media match the calculated PXRD pattern of the crystal structure determined here for ORBI
hemihydrate (Figure S4—Supplementary Materials). Moreover, there is no apparent difference in terms
of peak width of hkl indices with h “
­ 0 when the final solid materials starting from the hemihydrate
and anhydrous forms are compared. Therefore, it can be concluded that the anhydrous form converts
into the hemihydrate one in these three dissolution media, and consequently the equilibrium solubility
values obtained for the ORBI anhydrous form could not be considered actual values.
The Bragg peaks of the ORBI hemihydrate phase (Figure 12a) do not match the experimental
ones present in the PXRD pattern of the solid material separated from the ORBI anhydrous form
in the 0.1 mol¨ L´1 HCl medium (Figure 12c). In other words, the PXRD analysis shows that the
ORBI anhydrous form is converted to another form during the solubility determination experiment in
0.1 mol¨ L´1 HCl. The new phase is quite likely an ORBI chloride salt. This assumption is based on
the fact that H2ORBIp`cationq is the unique aqueous ORBI species expected to be present in solution at
pH < 3 (Figure 10), thus enabling its crystallization with the chloride from the HCl medium. However,
Molecules 2016, 21, 328
13 of 19
the experimental PXRD pattern of the residual material from the hemihydrate in 0.1 mol¨ L´1 HCl
Molecules 2016, 21, 328
13 of 19
medium (Figure 12b) shows Bragg peaks from the ORBI hemihydrate phase (Figure 12a) in addition to
those attributed
to the
chloride
(Figure 12c).
Therefore,
thethe
solid
in equilibrium is
pH < 3 (Figure
10),ORBI
thus enabling
its salt
crystallization
with the
chloride from
HClmaterial
medium. However,
−1 HCl
the experimental
PXRD pattern
of the hemihydrate
residual material
fromand
the the
hemihydrate
0.1 mol∙L
a solid-state
mixture containing
the ORBI
form
probableinORBI
chloride
salt form
medium
(Figure
shows BraggThis
peaks
from the
ORBI hemihydrate
phase (Figure 12a)
in is
addition
crystallized
during
the12b)
experiment.
result
suggests
that the hemihydrate
form
more resistant
to those attributed to the ORBI chloride salt (Figure 12c). Therefore, the solid material in equilibrium
than the anhydrous one to conversion to the ORBI salt chloride form. This trend is confirmed when
is a solid-state mixture containing the ORBI hemihydrate form and the probable ORBI chloride salt
1 HCl media
the experimental
PXRDduring
patterns
the residual
separated
from
the 0.01 form
mol¨ is
L´more
form crystallized
the of
experiment.
This material
result suggests
that the
hemihydrate
containing
the ORBI
hemihydrate
andtoanhydrous
forms
are compared.
PXRD
resistant
than the
anhydrous one
conversion to
the ORBI
salt chloride The
form.former
This trend
is pattern
confirmed
whenonly
the experimental
PXRD patterns
of the residual material
separatedthe
from
the 0.01
(Figure 12d)
contains
the Bragg peaks
of the hemihydrate
form, whereas
latter
(Figure 12e)
−1 HCl media containing the ORBI hemihydrate and anhydrous forms are compared. The
indicatesmol∙L
the presence
of spurious Bragg peaks of very low intensity from the ORBI chloride salt
former PXRD pattern (Figure 12d) contains only the Bragg peaks of the hemihydrate form, whereas
form. Therefore, the percentage of conversion of the starting ORBI solid material to the ORBI chloride
the latter (Figure 12e) indicates the presence of spurious Bragg peaks of very low intensity from the
salt form
at the
end salt
of the
solubility
depends
on atsolid
least
two variables:
ORBI
chloride
form.
Therefore,determination
the percentage of experiments
conversion of the
starting ORBI
material
(i) the pH
of
the
aqueous
media;
and
(ii)
the
crystal
form
used
as
starting
material.
Conversion
is
to the ORBI chloride salt form at the end of the solubility determination experiments depends on at
least
two
variables:
(i)
the
pH
of
the
aqueous
media;
and
(ii)
the
crystal
form
used
as
starting
material.
favoured by the use of the anhydrous form as starting material at low pH. Therefore, in spite of the
is favoured
by the use
thesolubility
anhydrousdetermination
form as starting material
at lowinpH.
Therefore,
observedConversion
phase transition
observed
inofthe
experiment
aqueous
HCl media,
in spite of the observed phase transition observed in the solubility determination experiment in
which precludes the measurement of actual solubility values either for the hemihydrate or anhydrous
aqueous HCl media, which precludes the measurement of actual solubility values either for the
form, the
observed or
difference
be indirectly
stability
of the two
hemihydrate
anhydrouscan
form,
the observedcorrelated
difference with
can bethe
indirectly
correlated
withORBI
the crystal
forms. In
otherofwords,
form
is more
and consequently
moreand
susceptible
stability
the two the
ORBIanhydrous
crystal forms.
In other
words,unstable
the anhydrous
form is more unstable
consequently
more
susceptible
to
dissolution
or
concomitant
dissolution/solid-state
transformation
to dissolution or concomitant dissolution/solid-state transformation than the hemihydrate form.
than the hemihydrate form. The higher reactivity of the anhydrous form can be explained by the loss
The higher
reactivity of the anhydrous form can be explained by the loss of crystallinity along the a unit
of crystallinity along the a unit cell axis (as discussed in Section 3.3) and/or by its supersaturation with
cell axisrespect
(as discussed
in Section 3.3) and/or by its supersaturation with respect to the hemihydrate
to the hemihydrate form due to nucleation and/or crystal growth during inter-conversion in an
form due
to nucleation
and/or crystal growth during inter-conversion in an aqueous medium [21,41].
aqueous
medium [21,41].
Normalized Intensity (a.u.)
(e)
(d)
(c)
(b)
(a)
4
6
8
10
12
14
16
18
20
22
24
2θ degree (CuKα)
Figure
12. Calculated
experimentalPXRD
PXRD patterns
ORBI
forms
after after
the equilibrium
solubilitysolubility
Figure 12.
Calculated
and and
experimental
patternsofof
ORBI
forms
the equilibrium
test: (a) ORBI hemihydrate form (calculated); (b) ORBI hemihydrate in 0.1 mol∙L−1 HCl;´1
(c) ORBI
test: (a) ORBI hemihydrate −1form (calculated); (b) ORBI hemihydrate
in 0.1 mol¨ L HCl; (c) ORBI
anhydrous in 0.1 mol∙L HCL; (d) ORBI hemihydrate in 0.01 mol∙L−1 HCl (e) ORBI anhydrous in
´1 HCL; (d) ORBI hemihydrate in 0.01 mol¨ L´1 HCl (e) ORBI anhydrous in
anhydrous
0.1−1mol¨
0.01 in
mol∙L
HCl. L
The extra
Bragg peaks are shown by arrows.
´1
0.01 mol¨ L HCl. The extra Bragg peaks are shown by arrows.
The calculated Bragg peaks of the ORBI hemihydrate form (Figure 13a) agree with the experimental
PXRD patterns of the residual solid phase in equilibrium with the pH = 4.5 acetate buffer and pH = 5.8
Thephosphate
calculated
Bragg peaks of the ORBI hemihydrate form (Figure 13a) agree with the
buffer. However, all the spectra show spurious peaks indicating the partial conversion of
experimental PXRD patterns of the residual solid phase in equilibrium with the pH = 4.5 acetate
buffer and pH = 5.8 phosphate buffer. However, all the spectra show spurious peaks indicating the
partial conversion of the initial solid phase to another one. The new phases are probably the ORBI
acetate (Figure 1b,c and phosphate (Figure 13d,e) salt (see discussion above proposing the formation
of the ORBI chloride salt).
Molecules 2016, 21, 328
14 of 19
Molecules 2016,
21, 328
the initial
solid phase to another one. The new phases are probably the ORBI acetate (Figure 1b,c and
14 of 19
phosphate (Figure 13d,e) salt (see discussion above proposing the formation of the ORBI chloride salt).
Normalized Intensity (a.u.)
(e)
(d)
(c)
(b)
(a)
4
6
8
10
12
14
16
2θ degree (CuKα)
18
20
22
24
Figure 13. Calculated and experimental PXRD patterns of ORBI forms after the equilibrium solubility
Figure 13. Calculated and experimental PXRD patterns of ORBI forms after the equilibrium solubility
test: (a) ORBI hemihydrate form (calculated); (b) ORBI hemihydrate in pH 4.5 acetate buffer; (c) ORBI
test: (a) ORBI
hemihydrate
form (calculated);
(b)hemihydrate
ORBI hemihydrate
pH 4.5 acetate
buffer;
anhydrous
in pH 4.5 acetate
buffer (d) ORBI
in pH 5.8in
phosphate
buffer (e)
ORBI (c) ORBI
anhydrous
in pHin4.5
buffer
(d) The
ORBI
pH by
5.8arrows.
phosphate buffer (e) ORBI
anhydrous
pH acetate
5.8 phosphate
buffer.
extrahemihydrate
Bragg peaks arein
shown
anhydrous in pH 5.8 phosphate buffer. The extra Bragg peaks are shown by arrows.
3. Materials and Methods
3. Materials
and Methods
3.1. Chemicals
Orbifloxacin standard (purity 99.9%) was purchased from Sigma-Aldrich (Seelze, Germany),
3.1. Chemicals
and Orbifloxacin raw material (purity 99.8%) was from LKT Laboratories, Inc. (St. Paul, MN, USA).
® (Rio de
Analytical grade
sodium
hydroxide
and was
hydrochloric
acid were
from Vetec
Orbifloxacin
standard
(purity
99.9%)
purchased
frompurchased
Sigma-Aldrich
(Seelze,
Germany),
Janeiro,
Brazil),
and
potassium
phosphate
monobasic,
glacial
acetic
acid,
sodium
phosphate
and
and Orbifloxacin raw material (purity 99.8%) was from LKT Laboratories, Inc. (St. Paul, MN, USA).
sodium acetate were from Proquímios® (São Paulo, Brazil). Spectroscopy/HPLC grade methanol
was
Analytical grade sodium hydroxide
and hydrochloric acid were purchased from Vetec® (Rio de Janeiro,
purchased from JT Baker® (Xalostoc, Mexico). Ultrapure water was obtained by reverse osmosis and
Brazil), and
potassium
monobasic,
acetic
acid, MA,
sodium
phosphate and sodium acetate
filtration
throughphosphate
a water-direct
Q™ systemglacial
(Millipore,
Bedford,
USA).
® (São Paulo, Brazil). Spectroscopy/HPLC grade methanol was purchased from
were from Proquímios
Tablets of ORBI,
ORBAX™ (Schering-Plough Animal Health Ind. Com. Ltda., São Paulo, Brazil),
® (Xalostoc,
a declaredMexico).
dosage of 22.7
mg API, water
were purchased
at a local
excipients
of through
JT Bakerwith
Ultrapure
was obtained
bypharmacy.
reverse The
osmosis
and(mixture
filtration
inactive
pharmaceutical
ingredients)
used
in
the
PXRD
analysis
to
facilitate
the
ORBI
identification
in
the
a water-direct Q™ system (Millipore, Bedford, MA, USA).
ORBAX™ tablets was prepared with the following ingredients: lactose monohydrate (91%) povidone
Tablets of ORBI, ORBAX™ (Schering-Plough Animal Health Ind. Com. Ltda., São Paulo, Brazil),
(1.5%), starch glycolate sodium (6.0%), magnesium stearate (1.0%), and silicon dioxide (0.5%), which
with a declared
dosagefrom
of 22.7
API, were purchased at a local pharmacy. The excipients (mixture of
were purchased
localmg
pharmacies.
inactive pharmaceutical
usedonly
in the
facilitate
theand
ORBI
The excipients ingredients)
composition lacked
the PXRD
Opadryanalysis
green dye,tocarnauba
wax,
the identification
ORBI
compared with
the ORBAX™
reference
product.
The total mass
of the excipients
in themonohydrate
formulation
in the ORBAX™
tablets
was prepared
with
the following
ingredients:
lactose
(91%)
was
obtained
by
determining
the
weight
of
the
ORBAX™
22.7
mg
tablets
(n
=
20),
and
the
amount
of
povidone (1.5%), starch glycolate sodium (6.0%), magnesium stearate (1.0%), and silicon dioxide (0.5%),
each excipient (except the lactose) was determined according to the average proportion of each
which were purchased from local pharmacies.
component used in the production of tablets in general [42]. The amount of lactose monohydrate,
Thewhich
excipients
composition
lacked
the Opadry
green
carnauba
and the ORBI
functions
as a diluent/filler,
wasonly
calculated
by subtracting
thedye,
amounts
of otherwax,
components
compared
with
the ORBAX™ reference product. The total mass of the excipients in the formulation was
from
100%.
obtained by determining the weight of the ORBAX™ 22.7 mg tablets (n = 20), and the amount of each
excipient (except the lactose) was determined according to the average proportion of each component
used in the production of tablets in general [42]. The amount of lactose monohydrate, which functions
as a diluent/filler, was calculated by subtracting the amounts of other components from 100%.
3.2. Crystal Preparation
ORBI crystals were obtained by a slow solvent evaporation technique; approximately 10 mg of
ORBI raw material was dissolved in approximately 50 mL of a solution of ethanol–chloroform (1:1 v/v).
After five days at room temperature (25 ˝ C) and protected from light, well-shaped white single crystals
were formed. Crystals were separated from the solution for the single-crystal X-ray diffraction study.
Molecules 2016, 21, 328
15 of 19
3.3. Structural Determination by X-ray Diffraction for Single Crystals
The single-crystal X-ray diffraction (SCXRD) measurement was performed at low temperature
(150 K) on a Gemini A-Ultra diffractometer (Oxford Diffraction, Blacksburg, VA, USA) equipped
with an Atlas CCD detector using graphite-monochromatized Mo or Cu Kα beams. The programs
CrysAlis CCD and CrysAlis RED [43] were used for data collection, cell refinement, data reduction,
and multi-scan method absorption correction. The structure was solved using the direct method of
structure factors phase retrieval using the software SHELXS-2013 [44] and refined by a full-matrix least
squares on F2 using the software SHELXL-2013 [44]. All non-hydrogen atoms were found from the
electronic density map constructed by Fourier synthesis and refined with anisotropic thermal parameters.
The hydrogen atoms bonded to carbons were stereochemically positioned following a riding model
with fixed C—H bond lengths of 0.93, 0.96, 0.97, and 0.98 Å for the aromatic, methyl, methylene, and
methine groups, respectively. The hydrogen atoms of the water molecule and the secondary amine were
refined freely after assignment using the difference Fourier map. The isotropic thermal parameters of
all hydrogens depended on the equivalent isotropic thermal displacements of the atoms bonded to
them [Uiso(H) = 1.2Ueq(N-amine, C-aromatic, C-methylene, C-methine) or 1.5Ueq(O-water, C-methyl)].
The zwitterionic feature of ORBI (carboxyl deprotonated and secondary amine protonated) was
unambiguously determined from the electronic density map constructed by Fourier synthesis.
The software WinGX [45] was used to treat the crystallographic data and generate tables.
The MERCURY [46] and ORTEP-3 [47] software programs were used for the crystallographic analysis
and artwork representations. CCDC 1447435 contains the supplementary crystallographic data for the
ORBI structural analysis. These data have been deposited at the Cambridge Crystallographic Data
Centre [35] and can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html
(or from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44 1223 336033; E-mail:
[email protected]).
3.4. Powder X-ray Diffraction Analysis
Powder X-ray diffraction (PXRD) data were recorded at room temperature (293 K) using a Ultima
IV diffractometer (Rigaku, Tokyo, Japan) with θ–2θ geometry. CuKα radiation (λ = 1.5418 Å) was
generated using a sealed tube at 40 kV and 30 mA. The data were collected with step size of 0.02˝ .
The speed of scan was 1˝ 2θ per minute from 3 to 35˝ in 2θ. The samples were finely ground and
mounted on a grooved glass slide employed as a sample holder. The experimental PXRD patterns
were compared with the one calculated by importing the crystal structure determined here for ORBI
using single-crystal X-ray diffraction (SCXRD) data into MERCURY. The PXRD pattern calculated from
low-temperature crystal structures (150 K, see Table 1) is, as expected, slightly shifted (right direction)
from the experimental patterns performed at room temperature (298 K).
3.5. Infrared Spectroscopy with Attenuated Total Reflectance by Fourier Transform (FTIR-ATR)
FTIR-ATR spectra were obtained using an Affinity-1 Fourier transform infrared spectrophotometer
(Shimadzu™, Tokyo, Japan) coupled to a Pike Miracle™ attenuated total reflectance sampling accessory
with ZnSe waveguides (Pike Technologies™, Madison, WI, USA). Spectra were recorded at room
temperature in the 4000–600 cm´1 range. After recording a background spectrum, the sample was
placed on the ZnSe crystal, and 32 scans were recorded with a resolution of 2 cm´1 .
3.6. Thermal Analysis
TG and DTA curves were obtained by simultaneous thermal analysis module SDT-Q-600,
manufactured by TA Instruments (New Castle, DE, USA). The system was calibrated against the
weight and temperature using high-purity indium, and the DTA baseline was performed and verified
before analysis. The analysis were permormed using open aluminium crucibles, a heating rate of
10 ˝ C¨ min´ 1 , nitrogen flow at 100 mL¨ min´ 1 , and samples of approximately 8 mg.
Molecules 2016, 21, 328
16 of 19
3.7. Interconversion Study between ORBI Anhydrous and Hemihydrate Forms
Approximately 50 mg of ORBI hemihydrate form (Sigma ORBI standard) was heated at 200 ˝ C
for 10 h to convert them to the ORBI anhydrous form. After allowing the samples to reach room
temperature in a desiccator containing silica, complete water release from the sample was confirmed
by TG measurements, which did not show mass loss in the range observed for the ORBI hemihydrate
form. The dehydrated sample was studied by PXRD.
To verify the stability of the anhydrous form to humidity, 50 mg of Sigma ORBI standard
previously heated at 200 ˝ C for 10 h was placed in a climatic chamber at 40 ˝ C with 75% of relative
humidity for 1 h. Thereafter, the sample was submitted to thermogravimetric and PXRD analysis.
3.8. ORBI Quantification
ORBI quantification in the solubility test was performed by a stability-indicating method with high
performance liquid chromatography as described by Casedey and collaborators [48]. The chromatographic
separation was achieved using a Prominence model Shimadzu™ liquid chromatograph (Kyoto, Japan)
with a DGU-20A 3R degasser, SIL-20AC HT autosampler, CTO-20A oven, LC-20AD pump, and
SPD-M20A diode array detector. The following chromatographic parameters were used. A Shim-pack
CLC-ODS Shimadzu column (Kyoto, Japan) (dimensions 250 mm ˆ 4.6 mm diameter and 5.0 µm
particle size) at 25 ˝ C was the stationary phase. The mobile phase used was a solution of 5% (v/v)
acetic acid:methanol (80:20 v/v) at a flow rate of 0.7 mL¨ min´ 1 . UV detection was performed at
a wavelength of 290 nm, and the injection volume of the sample in the chromatographic system was 20 µL.
A calibration curve was prepared using a stock solution in the mobile phase of ORBI Sigma
standard (previously dried for 4 h at 105 ˝ C in an oven) at 5 concentrations in triplicate (5, 10, 15, 20,
and 30 µg¨ mL´ 1 ). The results were analysed by linear regression of the peak area versus concentration.
The regression equation for the calibration curve was y = 134.377x ´ 3751 with a correlation coefficient
of (r) = 0.9999.
3.9. Solubility Studies
The solubility at equilibrium was determined by the classical saturation shake-flask method [49].
The ORBI hemihydrate and anhydrous (obtained by heating the hemihydrate form at 200 ˝ C)
forms were added in excess to 2-mL Eppendorf tubes containing 1 mL of the following media:
ultrapure water, hydrochloric acid 0.01 mol¨ L´ 1 , hydrochloric acid 0.1 mol¨ L´ 1 , pH 4.5 acetate buffer
(0.050 mol¨ L´ 1 ), pH 5.8 potassium phosphate buffer (0.054 mol¨ L´ 1 ), pH 6.8 potassium phosphate
buffer (0.072 mol¨ L´ 1 ), and pH 7.5 potassium phosphate buffer (0.088 mol¨ L´ 1 ).
The samples in each medium were prepared in triplicate. Subsequently, the tubes were shaken
at 150 rpm on a SOLABTM SL DT 180 (São Paulo, Brazil) shaker table at room temperature (25 ˝ C)
and protected from light. After 48 h of stirring, the samples were filtered on modified PTFE filter
(13-mm diameter, porosity of 0.50 µm, Advantec™ MSF, Dublin, CA, USA) and diluted in the mobile
phase with the necessary dilutions. The concentration of ORBI in each solvent was determined by the
chromatographic method (described in Section 3.8). The remaining solid was dried in a desiccator
containing silica for 5 days and subsequently analysed by PXRD to verify that the resulting material
was the same as the starting material.
4. Conclusions
A crystal structure of the antimicrobial API orbifloxacin was determined for the first time. This first
form is zwitterionic and hemihydrated. A second form (anhydrous form) was obtained by heating
the hemihydrate form. The phase transition occurs at 104 ˝ C and the anhydrous form is stable up to
269 ˝ C. The anhydrous form can be returned to the hemihydrate one under wet conditions. The crystal
structure determined for the hemihydrate form does not collapse on water removal. Therefore, the
hemihydrate and anhydrous forms of ORBI are isomorphous crystals. In spite of their isostructurally,
Molecules 2016, 21, 328
17 of 19
the anhydrous form is more porous and less stable (lower lattice energy) than the hemihydrate one
and consequently the former is expected to be more soluble than the later. Confirming our expectation,
the lower solubility of the hemihydrate form in comparison with the anhydrous one, leads us to infer
that the hemihydrate form is the one described in British Pharmacopoeia (2011) and in the USP 38
(2015), although it is presented as a dry form in these official compendia.
Supplementary Materials: Supplementary materials can be accessed at: http://www.mdpi.com/1420-3049/21/3/328/s1.
Acknowledgments: The authors thank FINEP (0134/2008), CNPq (552387/2011-8; 308354/2012-5; 448723/2014-0),
CAPES (AUX-PE—PNPD—2347/2011), and FAPEMIG (APQ-00273-14, and APQ-02486-14) for financial support.
We also thank CNPq and CAPES for research fellowships (EC and AD). This work is a collaboration research project
of members of the Rede Mineira de Quimíca (RQ-MG) supported by FAPEMIG (Project: CEX-RED-00010-14).
The authors express sincere thanks to Lab-Cri-UFMG for the X-ray facilities and Iara Maria Landre Rosa for
single-crystal X-ray diffraction measurements.
Author Contributions: O.M.M.S., A.C.D., M.B.d.A. and E.C.L.C. conceived and designed the experiments;
O.M.M.S. performed the SCXRD experiments; O.M.M.S. and J.T.J.F. performed the solubility and stability
experiments; A.C.D., J.T.J.F. and O.M.M.S. analyzed the data; Contributed reagents/materials/analysis tools:
A.C.D., E.C.L.C. and M.B.d.A. A.C.D. and O.M.M.S. wrote the paper. All authors approved the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations
The following abbreviations are used in this manuscript:
ORBI
API
SCXRD
PXRD
TG/DTA
FTIR-ATR
orbifloxacin
Active Pharmaceutical Ingredient
Single-Crystal X-ray Diffraction
Powder X-ray Diffraction
Thermogravimetric/Differential Thermal Analysis
Fourier transform infrared spectroscopy-Attenuated Total Rectance
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Ihrke, P.J.; Papich, M.G.; Demanuelle, T.C. The use of fluoroquinolones in veterinary dermatology. Vet. Dermatol.
1999, 10, 193–204. [CrossRef]
Dudhatra, G.B.; Kumar, A.; Awale, M.M.; Kamani, D.R.; Chauhan, B.N.; Patel, H.B.; Mody, S.K. Determination of
orbifloxacin in sheep plasma by high performance liquid chromatography with ultraviolet detection after
intravenous and intramuscular administration. J. Pharmacol. Toxicol. Methods 2013, 67, 50–53. [CrossRef] [PubMed]
Cheng, G.; Dong, X.; Wang, Y.; Peng, D.; Wang, X.; Hao, H.; Xie, S.; Qu, W.; Liu, Z.; Yuan, Z. Development of
a novel genetically modified bioluminescent-bacteria-based assay for detection of fluoroquinolones in
animal-derived foods. Anal. Bioanal. Chem. 2014, 406, 7899–7910. [CrossRef] [PubMed]
Albarellos, G.A.; Landoni, M.F. Current concepts on the use of antimicrobials in cats. Vet. J. 2009, 180,
304–316. [CrossRef] [PubMed]
Reynolds, J.E.F. Martindale the Extra Pharmacopoeia, 35th ed.; Pharmaceutical Press: London, UK, 2007; p. 278.
Goudah, A.; Abo-El-Sooud, K. Pharmacokinetics and milk penetration of orbifloxacin after intravenous and
intramuscular injections to dromedary lactating camels (Camelus dromedaries). J. Vet. Pharmacol. Ther. 2008,
31, 276–280. [CrossRef] [PubMed]
O’Neil, M.J. The Merck Index: An Encyclopedia of Chemicals, Drugs and Biologicals, 14th ed.; Merck & Co.:
Whitehouse Station, NJ, USA, 2006; p. 1182.
British Pharmacopoeia; Her Majesty’s Stationery Office: London, UK, 2011; Volume 1, pp. 88–90.
United States Pharmacopeia National Formulary—USP38 NF33; U.S. Pharmacopeial Convention: Rockville,
MD, USA, 2015; p. 3000.
Brog, J.P.; Chanez, C.L.; Crochet, A.; Fromm, K.M. Polymorphism, what it is and how to identify it:
A systematic review. RSC Adv. 2013, 3, 16905–16931. [CrossRef]
Molecules 2016, 21, 328
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
18 of 19
Láng, P.; Kiss, V.; Ambrus, R.; Farkas, G.; Szabó-Révész, P.; Aigner, Z.; Várkonyi, E. Polymorph screening of
an active material. J. Pharm. Biomed. Anal. 2013, 84, 177–183. [CrossRef] [PubMed]
Bonfilio, R.; Leal, J.S.; Santos, O.M.; Pereira, G.R.; Doriguetto, A.C.; Araújo, M.B. Analysis of chlorthalidone
polymorphs in raw materials and tablets and the effect of forms I and II on the dissolution properties of
drug products. J. Pharm. Biomed. Anal. 2014, 88, 562–570. [CrossRef] [PubMed]
Aaltonen, J.; Allesø, M.; Mirza, S.; Koradia, V.; Gordon, K.C.; Rantanen, J. Solid form screening—A review.
Eur. J. Pharm. Biopharm. 2009, 71, 23–37. [CrossRef] [PubMed]
Aguiar, J.A.; Krc, J., Jr.; Kinkel, A.W.; Samyn, J.C. Effect of polymorphism on the absorption of chloramphenicol
from chloramphenicol palmitate. J. Pharm. Sci. 1967, 56, 847–853. [CrossRef] [PubMed]
Lee, A.Y.; Erdemir, D.; Myerson, A.S. Crystal polymorphism in chemical process development. Annu. Rev.
Chem. Biomol. Eng. 2011, 2, 259–280. [CrossRef] [PubMed]
Byrn, S.R.; Pfeiffer, R.R.; Stowell, J.G. Solid-State Chemistry of Drugs, 2nd ed.; SSCI: West Lafayette, IN, USA, 1999.
Purohit, R.; Venugopalan, P. Polymorphism: An overview. Resonance 2009, 14, 882–893. [CrossRef]
Food and Drug Administration (FDA). Guidance for Industry—ANDAs: Pharmaceutical Solid Polymorphism
Chemistry, Manufacturing, and Controls Information, 2007. Available online: http://www.fda.gov/downloads/
Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm072866.pdf (accessed on 18 December 2015).
International Conference on Harmonisation (ICH). Q6A: Test Procedures and Acceptance Criteria for
New Drug Substances and New Drug Products: Chemical Substances; ICH: Genebra, Switzerland, 1999.
Available online:
http://www.ich.org/products/guidelines/quality/article/qualityguidelines.html
(accessed on 18 December 2015).
Censi, R.; Di Martino, P. Polymorph impact on the bioavailability and stability of poorly soluble drugs.
Molecules 2015, 20, 18759–18776. [CrossRef] [PubMed]
Santos, O.M.M.; Silva, D.M.; Martins, F.T.; Legendre, A.O.; Azarias, L.C.; Rosa, I.M.L.; Neves, P.P.;
Araujo, M.B.; Doriguetto, A.C. Protonation pattern, tautomerism, conformerism, and physicochemical
analysis in new crystal forms of the antibiotic doxycycline. Cryst. Growth Des. 2014, 14, 3711–3726. [CrossRef]
Martins, F.T.; Bocelli, M.D.; Bonfilio, R.; Araújo, M.B.; Lima, P.V.; Neves, P.P.; Veloso, M.P.; Ellena, J.;
Doriguetto, A.C. Conformational polymorphism in racemic crystals of the diuretic drug chlortalidone.
Cryst. Growth Des. 2009, 9, 3235–3244. [CrossRef]
Martins, F.T.; Paparidis, N.; Doriguetto, A.C.; Ellena, J. Crystal engineering of an anti-HIV drug based on the
recognition of assembling molecular frameworks. Cryst. Growth Des. 2009, 9, 5283–5292. [CrossRef]
Martins, F.T.; Doriguetto, A.C.; Ellena, J. From rational design of drug crystals to understanding of nucleic
acid structures: Lamivudine duplex. Cryst. Growth Des. 2010, 10, 676–684. [CrossRef]
Martins, F.T.; Legendre, A.O.; Honorato, S.B.; Ayala, A.P.; Doriguetto, A.C.; Ellena, J. Solvothermal preparation
of drug crystals: Didanosine. Cryst. Growth Des. 2010, 10, 1885–1891.
Martins, F.T.; Lima, P.V.; Azarias, L.C.; Abreu, P.J.; Neves, P.P.; Legendre, A.O.; Andrade, F.M.; Oliveira, G.R.;
Ellena, J.; Doriguetto, A.C. Increasing the symmetry of drug crystals: A monoclinic conformational polymorph
of the platelet antiaggregating agent ticlopidine hydrochloride. CrystEngComm 2011, 13, 5737–5743. [CrossRef]
Legendre, A.O.; Silva, L.R.R.; Silva, D.M.; Rosa, I.M.L.; Azarias, L.C.; Abreu, P.J.; Araújo, M.B.; Neves, P.P.;
Torres, C.; Martins, F.T.; et al. Solid state chemistry of the antibiotic doxycycline: Structure of the neutral
monohydrate and insights into its poor water solubility. CrystEngComm 2012, 14, 2532–2540. [CrossRef]
Martins, F.T.; Abreu, P.J.; Azarias, L.C.; Villis, P.C.M.; Campos Melo, A.C.; Ellena, J.; Doriguetto, A.C.
Form III-like conformation and Form I-like packing in a chloroform channel solvate of the diuretic
drug chlortalidone. CrystEngComm 2012, 14, 6173–6177. [CrossRef]
Ellena, J.; Bocelli, M.D.; Honorato, S.B.; Ayala, A.P.; Doriguetto, A.C.; Martins, F.T. Base-paired and base-stacked
structures of the anti-HIV drug lamivudine: A nucleoside DNA-mimicry with unprecedented topology.
Cryst. Growth Des. 2012, 12, 5138–5147. [CrossRef]
Martins, F.T; Bonfilio, R.; Rosa, I.M.L.; Santos, L.M.; Santos, O.M.M.; Araújo, M.B.; Doriguetto, A.C. The form
II of the antihypertensive drug chlorthalidone. CrystEngComm 2013, 15, 3767–3771. [CrossRef]
Flack, H.D. Chiral and achiral crystal structures. Helv. Chim. Acta 2003, 86, 905–921. [CrossRef]
Hoffmann, R.W. Meso compounds: Stepchildren or favored children of stereoselective synthesis?
Angew. Chem. Int. Ed. 2003, 42, 1096–1109. [CrossRef] [PubMed]
Boeyens, J.C.A.; Evans, D.G. Conformational analysis of ring pucker. Acta Crystallogr. B 1989, B45, 581–590.
Molecules 2016, 21, 328
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
19 of 19
Bruno, I.J.; Cole, J.C.; Kessler, M.; Luo, J.; Motherwell, W.D.; Purkis, L.H.; Smith, B.R.; Taylor, R.; Cooper, R.I.;
Harris, S.E.; et al. Retrieval of crystallographically-derived molecular geometry information. J. Chem. Inf.
Comput. Sci. 2004, 44, 2133–2144. [CrossRef] [PubMed]
Allen, F.H. The Cambridge structural database: A quarter of a million crystal structures and rising.
Acta Crystallogr. B 2002, 58, 380–388. [CrossRef] [PubMed]
Sahoo, S.; Chakraborti, C.K.; Mishra, S.C.; Nanda, U.N.; Naik, S. FTIR and XRD investigations of
some fluoroquinolones. Int. J. Pharm. Pharm. Sci. 2011, 3, 165–170.
Herbstein, F.H. IUCr Monographs on Crystallography—18: Crystalline Molecular Complexes and Compounds:
Structures and Principles; Oxford University Press Inc.: New York, NY, USA, 2005.
Nesterenko, V.F.; Bondar, M.P.; Ershov, V. Instability of plastic flow at dynamic pore collapse. In Proceedings
of the AIP Conference Proceedings, Colorado Springs, CO, USA, 28´2 July 1994; Volume 309, pp. 1173–1176.
Simonot-Grange, M.H. Thermodynamic and structural features of water sorption in zeolites. Clays Clay Miner.
1979, 27, 423–428. [CrossRef]
Byrn, S.; Pfeiffer, R.; Ganey, M.; Hoiberg, C.; Poochikian, G. Pharmaceutical solids: A strategic approach to
regulatory considerations. Pharm. Res. 1995, 12, 945–954. [CrossRef] [PubMed]
Borgadus, J.B.; Blackwood, R.B., Jr. Solubility of doxycycline in aqueous solution. J. Pharm. Sci. 1979, 68, 188–194.
Aulton, M.E. Delineamento de Formas Farmacêuticas, 2nd ed.; Artmed: Rio de Janeiro, Brazil, 2005; p. 675.
CrysAlis Pro 1.7.; Oxford Diffraction Ltd.: Abingdon, UK, 2006.
Sheldrick, G.M. A short history of SHELX. Acta Crystallogr. A 2008, A64, 112–122. [CrossRef] [PubMed]
Farrugia, L.J. WinGX suite for small-molecule single-crystal crystallography. J. Appl. Crystallogr. 1999, 32,
837–838. [CrossRef]
Macrae, C.F.; Bruno, I.J.; Chisholm, J.A.; Edgington, P.R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.;
Taylor, R.; van de Streek, J.; Wood, P.A. Mercury CSD 2.0—New features for the visualization and
investigation of crystal structures. J. Appl. Crystallogr. 2008, 41, 466–470. [CrossRef]
Farrugia, L.J. ORTEP-3 for Windows—A version of ORTEP-III with a Graphical User Interface (GUI).
J. Appl. Crystallogr. 1997, 30, 568. [CrossRef]
Cazedey, E.C.L.; Othman, A.; Garg, S.; Salgado, H.R.N. A validated stability-indicating lc method for
orbifloxacin in the presence of degradation products. Curr. Pharm. Anal. 2011, 7, 176–181. [CrossRef]
Baka, E.; Comer, J.E.; Takács-Novák, K. Study of equilibrium solubility measurement by saturation shake-flask
method using hydrochlorothiazide as model compound. J. Pharm. Biomed. Anal. 2008, 46, 335–341. [CrossRef]
[PubMed]
Sample Availability: Not available.
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).