Sertraline racemate and enantiomer: Solid-state characterization, binary phase diagram, and crystal structures

Article abstract of DOI:10.1021/cg901197b

The racemate and enantiomer of sertraline free base were prepared and characterized. The crystalline sertraline enantiomer is relatively less polymorphic compared with the sertraline-HCl salt. The solid-state nature of sertraline racemate was identified to be a racemic compound through a binary melting point phase diagram and spectroscopy analysis. The crystal structures of the racemate and enantiomer were determined to be monoclinic P12₁/n1 and P2₁, respectively.

Full text of DOI:10.1021/cg901197b

DOI: 10.1021/cg901197b
Sertraline Racemate and Enantiomer: Solid-State Characterization,
Binary Phase Diagram, and Crystal Structures
2010, Vol. 10
1633–1645
Quan He, Sohrab Rohani,* Jesse Zhu, and Hassan Gomaa
Department of Chemical and Biochemical Engineering, The University of Western Ontario London,
Ontario, N6G 4R3, Canada
Received September 29, 2009; Revised Manuscript Received January 22, 2010
ABSTRACT: The racemate and enantiomer of sertraline free base were prepared and characterized. The crystalline sertraline
enantiomer is relatively less polymorphic compared with the sertraline-HCl salt. The solid-state nature of sertraline racemate
was identified to be a racemic compound through a binary melting point phase diagram and spectroscopy analysis. The crystal
structures of the racemate and enantiomer were determined to be monoclinic P121/n1 and P21, respectively.
1. Introduction
and enhanced manufacturability. However, in some cases, the
free base/acid is a better choice especially when API has a low
pKa value and the resulting salts are less stable or more
hygroscopic or when they exhibit complex polymorphism/
pseudopolymorphism.8,9 Exploring other possible solid forms
than the currently used sertraline HCl salt for drug develop-
ment is therefore a worthwhile attempt.
Sertraline is one of the selective serotonin reuptake inhibi-
tors (SSRI) that has been shown to be effective in the treat-
ment of major depression. It may also be used for panic
disorder, social phobia, obesity, or obsessive-compulsive
disorder (OCD). Its chemical name is (1S,4S)-4-(3,4-
dichlorophenyl)-1,2,3,4-tetrahydro-N-methyl-1-naphthalene-
amine.1 The structures of sertraline racemate and enantiomers
are shown in Scheme 1.
Additionally, the present manufacturing process of enan-
tiomerically pure sertraline involves a classical resolution via
diastereomeric salt crystallization, in which (R)-mandelic acid
is introduced as a resolving agent to form the less soluble
(1S,4S)-sertraline mandelate. Subsequently, the single enan-
tiomer sertraline free base (1S,4S)-sertraline is isolated from
mandelate and converted into its hydrochloride.10 The classi-
cal resolution by diastereomeric salt formation has many
advantages such as the simplicity of crystallization operation,
the ease of process scale-up, and the availability of both
enantiomers.11 However, the process involves many steps.
Very often, several recrystallization steps are needed to
achieve high optical purity. Moreover, the intermediate dia-
stereomeric salt has to be decomposed resulting in the desired
enantiomer and recovery of the resolving agent. Thus, devel-
opment of other more efficient chiral separation techni-
ques has been of great interest to both the pharmaceutical
industry and academic researchers. Other possible separation
methods without introducing foreign chiral agents such as
directcrystallization, enrichmentcrystallization, and coupling
crystallization with other separation processes are very
attractive.
The specific method that can be applied depends on the
solid nature of a given racemate which is basically classified
into conglomerate, racemic compound, or pseudoracemate
(solid solution).12 The determination of the solid nature of a
certainracemateisthe prerequisite for rational design of chiral
resolution and purification process of pharmaceuticals. It can
provide a valuable thermodynamic basis for chiral separation,
chiral enrichment, and chiral purification.13 For example, if
racemate belongs to the conglomerate, direct crystallization
will be applied.14 If it is a racemic compound, the chiral
purification by crystallization can be employed as long as
the enantiomeric excess (ee) of the starting mixture enriched
with the desired enantiomer provided by asymmetric synth-
esis, simulated moving bed (SMB) chromatography or other
Sertraline is marketed in the form of its hydrochloride salt,
brand name Zoloft, which is one of the most highly poly-
morphic salts known today. Up to now 28 polymorphic
crystalline forms of sertraline HCl have been disclosed in
patents,2 including pure salt forms, hydrates, and solvates.
Therefore, much effort has been made to obtain single poly-
morph since it is desirable to control the specific form and
avoid a mixture of several forms for formulations of the active
pharmaceutical ingredient (API).3 A critical polymorphic
analysis of other salt forms demonstrated that seemingly
minor differences in salt former can have profound unpre-
dictable effects on the number of polymorphs and solvates.4-6
For example, the sertralinebromidewas foundtobe muchless
polymorphic than sertraline chloride.5 Since there is currently
no accurate method to predict the polymorphic behavior of a
given solid form (e.g., a salt or cocrystal vs free base/acid),
extensive experimentation is needed for screening polymorphs
in the early stage. Thus, it may be possible to find a proper
solid form with less polymorphic propensity and make for-
mulation, manufacture, and storage more reliable, robust,
and cost-effective.7,8
There are several forms of solid-state API including the
drug free acid/base, neutral molecules, solvates, salts, or
cocrystals. The choice and design of the preferred solid form
is based on the comparison of each form’s physicochemical
properties (solubility, bioavailability, hygroscopicity, melting
point, dissolution rate, and stability) as well as the required
properties for dosage and drug delivery.9 Salt forms are
frequently used in formulations due to their advantages
includingmodifiedsolubility, improved physicalandchemical
stabilities, masked unpleasant taste, increased bioavailability,
*To whom correspondence should be addressed. E-mail: rohani@
eng.uwo.ca.
r
2010 American Chemical Society
Published on Web 02/18/2010
pubs.acs.org/crystal

1634 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
Figure 1. DSC and TGA of sertraline enantiomer.
Scheme 1. Structures of Sertraline Racemate and Enantiomer
Figure 2. Hot stage micrographs of sertraline enantiomer. (a) T =
25 °C; (b) T = 67 °C; (c) T = 70 °C; (d) T = 73 °C.
method exceeds the ee of eutectic point of racemate.15 In order
to identify the solid-state nature of racemate, establishment of
a binary phase diagram by measuring the melting behavior of
the mixtures with different enantiomeric compositions is
commonly used.16-19
In the present work, we prepared the sertraline enantiomer
base as a crystalline form, characterized its physical proper-
ties, and performed a preliminary polymorphs screening.
The sertraline enantiomer base is a potential alternative for
pharmaceutical formulations instead of sertraline salts. The
advantages of using the free base for formulation lies in that
the free base is less polymorphic and can be further purified by
crystallization during the preparation process. Second, we
also obtained crystalline sertraline racemate base, namely,
xcis-4-(3,4-dichlorophenyl)-1,2,3,4-tetrahydro-N-methyl-
1-naphthaleneamine (a mixture of equal (1R,4R-) and
(1S,4S)-enantiomer). Thus, a binary melting point phase
diagram of the mixture of sertraline enantiomers on the basis
of thermoanalytical data was established. The solid-state
nature of sertraline racemate was identified to be a racemic
compound. The eutectic point was at Xe = 0.84 (the molar
fraction of (1S,4S)-sertraline enantiomer) and Te (melting
point of eutectic composition) was 58 °C. Finally, crystal
structures of sertraline racemate and enantiomer were ob-
tained and compared to each other, which demonstrated that
sertraline racemate could exist as a stable racemic compound.
The solid-state nature investigation of sertraline provides
valuable thermodynamic basis for exploring other possible
separation methods rather than classical resolution and the
possibility of chiral purification by direct crystallization.
Figure 3. DSC traces of sertraline enantiomer under heat-cool-
heat cycle.
Figure 4. XRPD patterns of solid crystallized from the melt of
sertraline enantiomer.
1H NMR experiments were performed on a Varian INOVA 600
spectrometer at 25 °C and DMSO-d6 as solvent. The X-ray powder
diffraction (XRPD) spectra were collected on a Rigaku-Miniflex
powder diffractometer (Carlsbad, CA) using CuKa (λ for KR =
2. Experimental Section
˚
1.54059 A) radiation obtained at 30 kV and 15 mA. The scans were
2.1. General. Sertraline enantiomer and sertraline racemate
hydrochloride salts were donated by Apotex PharmaChem Inc.
(Canada). All solvents were HPLC grade and were used without
further purification.
run from 5.0° to 40.0° 2θ, increasing at a step size of 0.02° 2θ with a
counting time of 5 s for each step. Specific rotations of salts were
measured by Autopol IV Digital Polarimeter Rudolph America
(Hackettstown, NJ) at 589 nm, equipped with a quartz cell of

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1635
100 mm path length. Solid-state FTIR spectra were recorded with a
Fourier transformation infrared spectrometer (Bruker Vector 22,
Billerica, MA) equipped with OPUS v3.1. The samples were ana-
lyzed by FTIR in transmittance mode through a diamond window.
The number of scans was 32 over the 600 to 4000 cm-1 spectral
region with a resolution of 4 cm-1. The background was collected in
the same range for air. Hygroscopicity was measured on dynamic
vapor sorption (DVS) Advantage-1 system (Surface Measurement
Systems Ltd., Alperton, Middlesex, UK). Approximate 100 mg
samples were weighed into the sample pan and exposed to water
partial pressure (p/p0) cycle of 0-90%. Equilibrium at each step was
determined by a dm/dt of 0.0005%/min. Melting behavior and other
thermal transition of solid samples were investigated by Lincam
LTS350 hot stage equipped by Lincam TNS 94 programmable
heater and a ZEISS microscope.
methyl-1-naphthalenamine hydrochloride (6.8 g, 0.02 mol) was
suspended in NaOH aqueous solution (50 mL, 5 N) and ethyl
acetate (50 mL) was gradually added. The mixture was stirred until
two phases became clear. The organic phase was separated from the
water phase and dried over Na2SO4. The ethyl acetate was removed
in vacuo and oily free base was obtained. Hexane (10 mL) was added
and heated to 50 °C. When cooled to room temperature, the solution
became cloudy. The solution was further cooled to 4 °C and white
crystals formed. Crystals were filtered off and dried at room
temperature overnight in vacuo to give sertraline free base 4.8 g
(yield: 77%).
Melting point: 65.58-67.25 °C (onset of DSC). Specific rotation
[R]15D = þ59.8° (c = 1, methanol). 1H NMR (DMSO-d6/TMS),
δ(ppm): 1.67-1.72 (m, H), 1.87-1.94 (m, 2H), 1.96 (s, 1H),
2.00-2.06 (m, H), 2.35 (s, 3H), 3.61 (d, 1H), 4.08 (t, 1H), 6.69 (d,
1H), 7.08 (t, 1H),7.14-7.17 (m, 2H), 7.375 (d, 1H),7.405 (d, 1H),
7.545 (d, 1H). IR ν(cm-1): 2937(w), 2846(w), 2778(w), 1735(s),
1469(s), 1363(s), 1214(s), 1126(s), 1027(s).
2.2. Preparation and Characterization of Sertraline Base. Enan-
tiomer. (1S,4S)-(þ)-4-(3,4-Dichlorophenyl)-1,2,3,4-tetrahydro-N-
Racemate. cis-4-(3,4-Dichlorophenyl)-1,2,3,4-tetrahydro-N-methyl-
1-naphthalenamine hydrochloride (6.8 g, 0.02 mol) was suspended in
NaOH aqueous solution (50 mL, 5 N) and ethyl acetate (50 mL) was
gradually added. The mixture was stirred until two phases became clear.
The organic phase was separated from the water phase and dried over
Na2SO4. The ethyl acetate was removed in vacuo and oily free base was
obtained. Heptane (7 mL) was added and heated to 70 °C. When cooled
to room temperature, white crystals formed and were filtered off and
dried at room temperature overnight in vacuo to give sertraline free base
4.9 g (yield: 80%).
Melting point: 69.25-69.87 °C (onset of DSC). 1H NMR
(DMSO-d6/TMS), δ(ppm): 1.67-1.72 (m, H), 1.87-1.94 (m, 2H),
1.96 (s, 1H), 2.00-2.06 (m, H), 2.35 (s, 3H), 3.61 (d, 1H), 4.08
Figure 5. HSM of sertraline enantiomer crystallized from melt. (a)
T = 25 °C; (b) T = 65 °C; (c) T = 69 °C; (d) T = 71 °C.
Table 1. Solubilities of Sertraline Enantiomer Free Base and Its Salts
solubility mg/mL
melting point °C
free basea
0.13
<0.1
3.8
0.6
0.1
1.9
4.2
0.3
65.87
free baseb
hydrochlorideb
hydrobromidec
p-toluenesulfonatec
lactatec
249
266
265
150
201
150
Figure 7. Representative DSC curves of mixtures with different
enantiomeric compositions. 1, racemate; 2, enantiomer; 3, 4, 5, 6, 7
are mixtures with X1S,4S = 0.85, 0.55, 0.65, 0.98, respectively.
methanesulfonatec
benzenesulfonatec
Table 2. Experimental Thermal Properties of Sertraline Racemate and
Enantiomer
a Solubility was measured at room temperature in deionized water;
the pH value of saturated solution at dissolution equilibrium is 7.66, the
pKa value of sertraline free base is 9.5 from ref 6. b Data are from ref 6.
c Data are from ref 5.
sertraline racemate
sertraline enantiomer
melting point (°C) (onset)
enthalpy of fusion (J/g)
69.5 ( 0.5
70.0 ( 1.0
66.5 ( 0.5
80.0 ( 1.0
Figure 6. DVS profiles of sertraline enantiomer (left) and sertraline HCl (right).

1636 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
(t, 1H), 6.69 (d, 1H), 7.08 (t, 1H),7.14-7.17 (m, 2H), 7.375 (d, 1H),
7.405 (d, 1H), 7.545 (d, 1H). IR ν(cm-1): 2937(w), 2846(w), 2778(w),
1735(s), 1469(s), 1363(s), 1214(s), 1126(s), 1027(s).
covered aluminum crucible having pierced lids to allow the escape of
volatiles. The sensors and samples were under nitrogen purge during
the experiments. The temperature calibration was carried out using
the melting point of highly pure indium in a medium temperature
range. Heating rates of 3, 5, 10 °C/min were used. The heating
rate did not affect the results. The heating rate of 5 °C/min was
employed.
2.3. Establishment of Binary Melting Point Phase Diagram. The
mixtures with different enantiomeric compositions were prepared
by mixing a weighed sertraline racemate (cis-sertraline) and enantio-
mer ((1S,4S)-sertraline). The resolution of balance was 0.00001 g.
Differential scanning calorimetry (DSC) and spectroscopic analyses
were performed on different parts of the sample and gave consistent
results which confirmed sufficient mixing. The binary melting point
phase diagram was established using Mettler Toledo DSC 822e
differential scanning calorimeter (Greifensee, Switzerland) by mea-
suring the temperature at the beginning and the end of fusion of
enantiomeric mixture. The samples (3-12 mg) were prepared in a
2.4. Single Crystal Preparation and Structure Determination.
Sertraline racemate single crystals were grown from a concentrated
ethanol solution by slow evaporation at room temperature. The
translucent light colorless chip-like crystal was mounted on a glass
fiber and data were collected at low temperature -150 K on a
Nonius kappa-CCD area detector diffractometer with COLLECT
˚
using monochromatic Mo-KR radiation (λ=0.71073 A). Sertraline
enantiomer single crystals were grown from a concentrated ethanol
solution by slow evaporation at room temperature. The data were
collected at temperature of 295 K on a 2D detector (Bruker Smart
6000 CCD) and 3-circle diffractometer (Bruker D8) using Cu-KR
˚
radiation (λ=1.54184 A). The structures were solved with direct
method using the SHELXS-97 program and refined on F2’s by full-
matrix least-squares with the SHELXS-97 program. The calculated
XRPD patterns based on single crystal data are in agreement with
measured XRPD patterns. The comparisons are presented in the
Supporting Information.
3. Results and Discussion
3.1. Solid Form of Sertraline Enantiomer. As described in
Experimental Section, the solid sertraline enantiomer free
base (1S,4S)-sertraline was isolated from its hydrochloride
salt. Free base exists preferably in supercooled liquid form.
Its solid form can crystallize from nonpolar solvent such as
hexane, heptane, and toluene. In the thermograms of sertra-
line enantiomer, a single melting endothermic peak at 66.5 (
0.5 °C (onset) was observed. Above 70 °C, there was no other
obvious thermal event in the DSC trace until in the range of
250-360 °C, at which, another broad endothermic peak
indicated the decomposition accompanying the almost com-
plete weight loss in the thermogravimetric analysis (TGA)
trace as shown in Figure 1. The heat of fusion, calculated by
integration of the melting endotherm in the total heat flow,
Figure 8. Binary melting point phase diagram of sertraline. The
solid curves represent calculated liquidus lines; circle and triangle
represent the temperatures of the beginning and the end of fusion
respectively.
Table 4. Crystal Structure Data of Sertraline Racemate and Enantiomer
cis-sertraline
(1S,4S)-sertraline
empirical formula
formula weight (g/mol)
temperature (K)
C17H17Cl2N
306.22
150
0.71073
monoclinic
P121/n1
11.0604(10)
8.6714(8)
16.2287(15)
90
103.937(6)
90
C17H17Cl2N
306.22
295
1.54178
monoclinic
P21
8.7569(5)
8.7731(5)
10.7921(6)
90
110.814(4)
90
˚
wavelength(A)
crystal system
space group
˚
a (A)
˚
b (A)
˚
c (A)
R (°)
β (°)
γ (°)
3
˚
V (A )
1510.7(2)
1.346
775.00(8)
1.312
Dcalc (g/cm3)
Z
4
2
crystal size (mm)
reflections collected
goodness-of-fit on F2
final R indices (I > 2σ(I))
0.06 ꢀ 0.08 ꢀ 0.2
0.4 ꢀ 0.4 ꢀ 0.4
10167
1.017
4996
1.053
R1 = 0.0332
wR2 = 0.0762
R1 = 0.0469
wR2 = 0.0827
R1 = 0.0347
wR2 = 0.0981
R1 = 0.0354
wR2 = 0.0993
Figure 9. XRPD patterns (top) and FTIR spectra (bottom) of the
racemate (lower traces) and enantiomer (upper traces).
R indices (all data)
Table 3. XRPD Peaks of Sertraline Racemate and Enantiomer
2θ/°
enantiomer
racemate
15.20 17.65 19.15 20.31
13.12 15.23 17.42 19.25
23.25 23.81 24.85 25.90
21.28 21.89 23.30 23.84 24.49
28.41 28.90 31.31 32.45
26.48 28.13 29.00
32.99

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1637
Figure 10. Atomic-numbering molecule scheme (left) and four molecules in the unit cell (right) of sertraline racemate.
was 70.0 ( 1.0 J/g. Further hot stage microscope (HSM)
investigation confirmed the interpretation of thermal events
happened when the crystalline free base experienced the
heating process in the DSC crucible. The photomicrographs
are displayed in Figure 2. Figure 2a shows the powder
samples of sertraline at 25 °C. When heated to 65 °C, a small
move of crystals was observed. At 67 °C shown in Figure 2b,
the crystals started to melt and the edge became round, then
quickly the crystals melted completely to form bigger liquid
droplets at 71 °C shown in Figure 2c,d.
The polymorphic outcome of crystallization is usually
system-specific, and it could be influenced by a number of
factors such as temperature, solvent, supersaturation, cool-
ing rate, agitation rate, impurity, and additives.20 The
mechanism of these effects is not well-known, and the
Figure 11. H-bond patterns of (1R,4R) (left) and (1S,4S) (right)
columns in the racemate.
quantitative relationship between the operational factors
and the polymorphic outcome is not clearly understood.21
Exploring new crystal forms relies basically on traditional
experimental methods. Since most APIs are purified and
isolated by crystallization from an appropriate solvent, the
most prevalent method of polymorph screening starts with
crystallization of APIs from a number of solvents.9,22 The
solvents used in this study included ethanol, methanol, 2-
propanol, and mixtures of hexane and alcohol. All the
resulting solids showed the same melting point and XRPD
patterns as those of the original sertraline enantiomer free
base crystals. The sertraline free base was further suspended
in the above-mentioned organic solvents and water under
vortex for 48 h. There was no evidence indicating the
existence of solvate or hydrate by analyzing the solid phases
at equilibrium. This result is further supported by the latter
crystal structure investigation work in which all concerned
single crystals grown from different solvents showed the
same unit cell and the theoretical calculated X-ray patterns
of single crystal were completely coincident with the powder
X-ray patterns of bulk samples.
Heat-cool-heat cycles and crystallization from melt are
efficient ways to detect new polymorphs. So the crystalline
sertraline enantiomer was subjected to a heat-cool-heat
cycle in DSC measurements shown in Figure 3a, namely,
heating from 20 to 120 °C at 5 °C/min, then cooling from 120
to 20 °C at 10 °C /min, and reheating from 20 to 120 °C at
5 °C/min. The DSC curve from the first heating segment 1
showed one sharp endotherm peak and matched the melting
behavior of sertraline enantiomer. There were no obvious
sharp exothermal peaks in the DSC curve when samples in
the DSC crucible underwent the cooling process (segment 2)
and then immediately the reheating process (segment 4);
there were no obvious sharp endotherm peaks in the DSC
curve from the reheating run. Small thermal events happened
during segments 2 and 4 marked in red circles. The DSC
curve indicated that the melt did not crystallize immediately
and completely during the cooling process. Since the super-
sooling is a necessary condition for crystallization and the
degree of supercooling depends on the nature of substance
and is influenced by the heat transfer as well as the cooling
rate,23 the sertraline enantiomer melts after the first heating
process were subjected to different cooling rates, namely, 0.1,
1, 5, and 10 °C /min, respectively, in expectation to provide
enough time to reach thermal equilibrium and let the mole-
cule relax, nucleate, and grow. The DSC traces under these
cooling rates did not change much and were still similar to
that in Figure 3a.
However, the sample in the crucible after the first melting
process was taken out from the DSC and left at room
temperature overnight, then was run on DSC and XRPD
again. The DSC curve of the second heating run displayed
the same endotherm peak as first heating run; just the onset
of melting point and heat of fusion decreased a little bit
probably due to the presence of a small amount of liquid
remaining or degradation of the sample. The XRPD pattern
was consistent with that of the original sertraline enantio-
mer. Both XRPD and DSC results implied that the solid
crystallized from melt at room temperature was same as the
original form of sertraline enantiomer. The assumption was
verified by extending the annealing time of sample in the
DSC crucibles at room temperature. The samples staying at
room temperature more than one week after the first heating
and melting process exhibited almost the same melting point
and heat of fusion as the original sertraline enantiomer.

1638 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
During the cooling process, a small endotherm peak
and then an exotherm peak (in red circles) were observed,
which might result from phase transition or crys-
tallization. During the annealing process at 20 °C, several
flat exotherm peaks (in red circles) were seen in the first
140 min as shown in Figure 3b of segment 3. They might be
crystallization, crystal growth, or phase transformation. The
very flat line appeared in the last 460 min shown in segment 3
implied no other thermal events happened. Upon the second
heating, only one sharp endotherm peak (segment 4) was
observed that corresponded to the same melting behavior of
starting sertraline enantiomer. These results are in agreement
with the crystal annealed at room temperature under air
atmosphere.
The crystallization from melt was also tracked in situ by
recording XRPD patterns profile as a function of time at
room temeprature. The sertraline free base was heated and
melted on the XRPD sampler directly, and then the XRPD
patterns were recorded continuously. As shown in Figure 4,
the starting melt did not exhibit any peaks, and after 30 min
there emerged some peaks showing the occurrence of crystal-
lization. More and more sertraline enantiomer-related peaks
appeared and their intensity increased. After 2-3 h, all the
characteristic peaks appeared. These peaks could be ascribed
to the sertraline enantiomer. However, even after 96 h, the
diffraction peak intensities were still lower than those in the
original sertraline enantiomer, which may be caused by a
small amount of melt, disordered component, or amorphous
phase in the resulting solid. Besides the characteristic peaks
of original sertraline enantiomer solid, there were no other
detectable peaks. Therefore, no obvious phase transforma-
tion was observed. Although we cannot completely exclude
the possibility of existence of other metastable phases as well
as phase transition during the melt crystallization process,
the XRPD patterns shown in Figure 4 confirmed that the
final solid crystallized from melt was the same as the original
form of sertraline enantiomer.
Observations from HSM provided further support to our
conclusion. When the melting liquid under the HSM was
cooled and annealed at room temperature, the sample did
not crystallize immediately. Only part of liquid transformed
into crystals as shown in Figure 5a. Part of the melt samples
on the slide under HSM still existed as small liquid droplets
even after one week. The newly crystallized solid from melt
was reheated in hot stage, and the snapshots of melting
behavior were recorded. At 65 °C, the crystals started to
blur and turned into transparent liquid as shown in
Figure 5b-d upon melting.
In light of DSC, XRPD, and HSM investigations, the melt
sertraline enantiomer cannot crystallize immediately and
requires enough annealing time at room temperature to
crystallize as the original form. The crystallization is a
very complicated process especially in polymorphic
systems, which may involve nucleation, crystal growth, the
competition between polymorphs as well as the effects of
supercooling degree and impurity, etc.23,24 According to
Ostwald’s rule,25 the metastable polymorph preferably crys-
tallizes first and then converts to the more stable form. So the
above observed crystallization phenomenon might result
from the combination of nucleation kinetics and growth
kinetics. These endo/exo thermal events in segments 2, 3,
and 4 in Figure 3a,b are most likely the results of metastable
phase nucleation and growth, metastable phase transforma-
tion into stable phase as well as stable phase crystallization,
Figure 12. Crystal packing diagram of the racemate. Purple and
orange represent (1R,4R) and (1S,4S) molecules, respectively;
dotted blue and red lines represent hydrogen bonds.
In order to further track the crystallization during the
cooling and annealing process, the sample in the DSC
crucible after the first heating run segment 1 was cooled
down to 20 °C (segment 2), kept at that temperature for 10 h
under inert N2 atmosphere (segment 3), and then experi-
enced the second heating run in segment 4, in which the
heating and cooling rates were exactly same as that in
Figure 3a. The recorded DSC trace is depicted in Figure 3b.

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1639
Figure 13. Atomic-numbering enantiomer molecule scheme (left) and two molecules in the unit cell (right) of sertraline enantiomer.
provided in Supporting Information. The pH value of
saturated solution at equilibrium was measured using a pH
meter. The solubilities of sertraline free base and its salts are
summarized in Table 1.
Hygroscopicity is another important physical property of
the API for formulation. The hygroscopicity of sertraline
enantiomer free base and sertraline hydrochloride was mea-
sured by dynamic vapor sorption (DVS). The relative humi-
dity (RH) ranged from 0% to 90%, with steps of 10%. Both
sertraline free base and its hydrochloride salt have very low
hygroscopicity as is shown in Figure 6. The equal mass
increase and decrease in a DVS measurement cycle also
allude to the nonexistence of sertraline hydrate, which is
consistent with the observation from solvate screening in
Section 3.1.
Onthe basisof the aboveinvestigation, crystalline sertraline
Figure 14. H-bond pattern of (1S,4S) column in the enantiomer.
enantiomer has hygroscopicity similar to sertraline hydro-
chloride but exhibits much less polymorphic diversity, which
provides an advantage for a potential alternative API for-
mulation. Although the solubility of sertraline base is very
low, the drawback can be overcome with proper formulation
strategies suchas nanosizing, adding a solubilityenhancer and
using novel drug delivery systems including micelle, micro-
emulsion or emulsion, solid lipid nanoparticles, etc.28-30
3.3. Binary Phase Diagram of Sertraline. The solid sertra-
line racemate was also obtained as described in Experimental
Section and its solid nature was investigated. The thermal
analysis of mixtures of sertraline racemate (cis-sertraline)
and sertraline enantiomer ((1S,4S)-sertraline) by DSC was
studied in the range of 0.5-1.0 molar fraction of sertraline
enantiomer.
The preparation of mixtures with different enantiomeric
compositions was conducted in three different ways to
eliminate the influence of sample preparation method. First,
a weighed amount of sertraline enantiomer was physically
mixed with sertraline racemate and crushed in a mortar and
pestle to obtain a uniform mixture. Second, a weighed
mixture was dissolved in methanol to ensure uniform mixing
and then the solvent was evaporated. Third, a weighted
mixture was melted and then solidified. Namely, the samples
obtained by mortar and pestle mixing were run on DSC,
annealed one day in a crucible, and then run on DSC
for a second time. All three methods gave consistent DSC
curves and demonstrated that the physical mixing method to
prepare samples by mortar and pestle was feasible. In this
work, mixtures with different enantiomeric compositions
were prepared by a physical mixing method. The interesting
point was that even the mixtures obtained by melting did
not give a different DSC curve, which suggested that no
etc. Although we cannot verify crystallization and transfor-
mation kinetics clearly, the final obtained form from the melt
is undoubtedly the same polymorphic form as the original
sertraline enantiomer.
The above polymorph screening investigation provides no
evidence of other stable polymorphs and solvates of sertra-
line enantiomer. Since the number of polymorphs of a given
compound is proportional to the time and money spent on
the compound,26 we cannot prove definitively that the sertra-
line enantiomer has no polymorphic distribution. However,
under our current investigation, at least we believe that it was
not easy to generate other forms, especially solvate forms.
The obtained crystalline form was quite stable at room
temperature. Interestingly, the preliminary inference coin-
cided with the phenomenon observed in another antidepres-
sant drug, venlafaxine, which demonstrated no polymorphism
of the free base compared to the venlafaxine hydro-
chloride.27
3.2. Properties of Solid-State Sertraline Enantiomer. Solu-
bility of API in water and/or organic solvents is a key
physical parameter for compatibility with various dosage
forms. The solubility of sertraline enantiomer free base in
water was measured by gravimetric method. The starting
solid was suspended in water for 24, 48, 72, and 96 h,
respectively. Experimental results indicate that 72 h is suffi-
cient to reach dissolution equilibrium. The undissolved solid
phase at equilibrium was analyzed by XRPD, Raman, and
DSC. All the characterization methods confirmed that the
final form was identical to the starting solid phase. The
comparison of XPRD patterns belonging to the starting
solid and the undissolved solid in solubility experiments is

1640 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
and 2, respectively, as shown in Figure 7. The thermal data
are listed in Table 2.
Other mixtures with different enantiomeric compositions
had two distinct peaks as curves 3, 4, 5, 6, and 7 shown in
Figure 7. The first endothermic peak represented the fusion
of the eutectic. The solidus temperature (Te) was determined
by the onset temperature of the first peak. The second peak,
however, represented the melting effect of the excess (1S,4S)-
enantiomer in the melting process. The liquidus temperature
(Tf) was determined by peak temperature of the second peak.
Curve 3 denotes the mixture with molar fraction of 0.85 as an
exception. It exhibited only one melting peak with the onset
temperature of 56.85 °C. The first peaks’ melting tempera-
ture of all other mixtures coincided with 56.85 ( 1 °C, which
implies that the eutectic point is located at X1S,4S=0.85 with
a melting point of 56.85 °C.
The DSC curves present a typical racemic compound
feature. However, it is well-known that the partial solid
solution may exist in certain enantiomeric composition
regions. Partial solid solution happens most often in the
vicinity of racemate and/or enantiomer12 and less often in the
other regions.31 The mixture with X1S,4S=0.98 still displays a
notable melting event at the eutectic melting temperature
and rules out the possibility of solid solution. As for X1S,4S
from 0.98 to 1.0, we cannot provide a reasonable explanation
if there is a partial solid solution in the narrow vicinity of
enantiomer. Usually, a Tammann-plot using the heat of
fusion of eutectic vs enantiomeric composition can rule
out the possibility of solid solution in the vicinity of pure
enantiomer.12 However, in this case, the Tammann-
plot was not feasible in that the melting temperature of
56.85 °C and enantiomeric composition of 0.84 of eutectic
were fairly near to those of enantiomer and the merging of
two melting peaks in DSC curves made it difficult to
determine the heat of fusion by integrating areas of peaks
accurately.
The resulting binary diagram is depicted in Figure 8 for the
mole fraction of (1S,4S)-sertraline X1S,4S ranging from 0.5 to
1. The experimental values agree well with the two theo-
retical liquidus curves (the solid curves in Figure 8) assuming
the system is a racemic compound. The two theoretical
liquidus curves were calculated respectively on the basis of
the melting point and fusion enthalpy of sertraline enantio-
mer using Schroder-van Laar equation,
!
ΔHR,f
1
1
ln XR
¼
-
,
R
TR,f Tf
and the melting point and fusion enthalpy of cis-sertraline
racemate using Prigogine-Defay equation,
!
2ΔHrac,f
1
1
ln 4XRð1 -XRÞ ¼
-
,
R
Trac,f Tf
where XR is the molar fraction of enantiomer, ΔHR,f and TR,f
are the enthalpy of fusion in J/mol and the melting point of
pure enantiomer in K, ΔHrac,f, and Trac,f are the enthalpy of
fusion of racemate in J/mol and the melting point of race-
mate in K, and R is a gas constant.12
Figure 15. Crystal packing diagram of the enantiomer. Dotted blue
lines represent hydrogen bonds and the disorder of the methylamine
group is not shown for clarity.
other phases formed from the melt crystallization. This
implies less polymorphic potency of sertraline enantiomer
and racemate.
The DSC curves of pure racemate and enantiomer exhi-
bited one sharp endothermic peak represented by curves 1
The two theoretically calculated curves intersect at X1S,4S
=
0.84 and Te = 58 °C, which is in good agreement with the
experimental values of X1S,4S = 0.85 and Te = 56.85 °C.
Therefore, the sertraline system is determined as a racemic
compound.

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1641
Table 5. Bond Lengths, Bond Angles for Sertraline Racemate
˚
Bond Lengths (A)
C1-N2
1.454
0.98
0.98
0.98
1.47
0.85
1.513
1.53
1.0
1.525
0.99
0.99
1.535
0.99
0.99
1.534
1.0
C8-C9
1.406
1.386
0.95
1.381
0.95
1.380
0.95
0.95
1.388
1.397
1.395
0.95
1.394
1.738
1.385
1.740
1.379
0.95
C1-H1A
C1-H1B
C1-H1C
N2-C3
C9-C10
C9-H9A
C10-C11
C10-H10A
C11-C12
C11-H11A
C12-H12A
C13-C14
C13-C18
C14-C15
C14-H14A
C15-C16
C15-Cl19
C16-C17
C16-Cl20
C17-C18
C17-H17A
C18-H18A
N2-H2A
C3-C8
C3-C4
C3-H3A
C4-C5
C4-H4A
C4-H4B
C5-C6
C5-H5A
C5-H5B
C6-C7
C6-H6A
C6-C13
C7-C12
C7-C8
1.52
1.398
1.400
0.95
Bond Angles (°)
N2-C1-H1A
N2-C1-H1B
N2-C1-H1C
H1A-C1-H1B
H1A-C1-H1C
H1B-C1-H1C
C1-N2-C3
C1-N2-H2A
C3-N2-H2A
N2-C3-C8
109.5
109.5
109.5
109.5
109.5
109.5
C12-C7-C6
119.21(17)
121.90(17)
118.62(18)
122.31(17)
119.06(18)
121.5(2)
119.2
C8-C7-C6
C7-C8-C9
C7-C8-C3
C9-C8-C3
C10-C9-C8
113.82(17)
106.6(16)
109.3(16)
108.22(16)
110.74(17)
108.8
111.34(16)
108.8
108.8
110.53(16)
109.5
109.5
109.5
109.5
108.1
110.38(17)
109.6
109.6
109.6
109.6
108.1
111.31(15)
111.31(15)
111.57(15)
107.1
112.24(16)
107.1
107.1
C10-C9-H9A
C8-C9-H9A
119.2
C11-C10-C9
C11-C10-H10A
C9-C10-H10A
C12-C11-C10
C12-C11-H11A
C10-C11-H11A
C11-C12-C7
C11-C12-H12A
C7-C12-H12A
C14-C13-C18
C14-C13-C6
C18-C13-C6
C13-C14-C15
C13-C14-H14A
C15-C14-H14A
C16-C15-C14
C16-C15-Cl19
C14-C15-Cl19
C15-C16-C17
C15-C16-Cl20
C17-C16-Cl20
C18-C17-C16
C18-C17-H17A
C16-C17-H17A
C17-C18-C13
C17-C18-H18A
C13-C18-H18A
119.5(2)
120.3
120.3
119.8(2)
120.1
120.1
121.7(2)
119.1
119.1
118.16(17)
120.93(16)
120.90(17)
120.64(17)
119.7
N2-C3-C4
N2-C3-H3A
C8-C3-C4
C8-C3-H3A
C4-C3-H3A
C5-C4-C3
C5-C4-H4A
C5-C4-H4B
C3-C4-H4A
C3-C4-H4B
H4A-C4-H4B
C4-C5-C6
C4-C5-H5A
C4-C5-H5B
C6-C5-H5A
C6-C5-H5B
H5A-C5-H5B
C13-C6-C7
C13-C6-C7
C13-C6-C5
C13-C6-H6A
C7-C6-C5
119.7
120.34(17)
120.46(15)
119.20(14)
119.39(18)
121.43(15)
119.19(15)
120.17(18)
119.9
119.9
C7-C6-H6A
C5-C6-H6A
C12-C7-C8
121.29(18)
119.4
119.4
118.88(18)
Torsion Angles (°)
C1-N2-C3-C8
C1-N2-C3-C4
N2-C3-C4-C5
C8-C3-C4-C5
C3-C4-C5-C6
C4-C5-C6-C13
C4-C5-C6-C7
C13-C6-C7-C12
C5-C6-C7-C12
C13-C6-C7-C8
C5-C6-C7-C8
C12-C7-C8-C9
163.01(18)
-74.7(2)
-70.0(2)
50.5(2)
-64.9(2)
170.76(16)
45.0(2)
C9-C10-C11-C12
C10-C11-C12-C7
C8-C7-C12-C11
C6-C7-C12-C11
C7-C6-C13-C14
C5-C6-C13-C14
C7-C6-C13-C18
C5-C6-C13-C18
C18-C13-C14-C15
C6-C13-C14-C15
C13-C14-C15-C16
C13-C14-C15-Cl19
-0.4(3)
-0.4(3)
0.9(3)
179.63(18)
-115.5(2)
118.3(2)
64.2(2)
-62.0(2)
0.6(3)
-179.64(16)
0.6(3)
41.0(2)
166.85(17)
-140.31(18)
-14.4(2)
-0.5(3)
-179.45(14)
C6-C7-C8-C9
-179.17(17)
C14-C15-C16-C17
-0.8(3)

1642 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
Table 5. Continued
Torsion Angles (°)
C12-C7-C8-C3
C6-C7-C8-C3
N2-C3-C8-C7
C4-C3-C8-C7
N2-C3-C8-C9
C4-C3-C8-C9
C7-C8-C9-C10
C3-C8-C9-C10
C8-C9-C10-C11
-179.65(17)
1.6(3)
Cl19-C15-C16-C17
C14-C15-C16-Cl20
Cl19-C15-C16-Cl20
C15-C16-C17-C18
Cl20-C16-C17-C18
C16-C17-C18-C13
C14-C13-C18-C17
C6-C13-C18-C17
179.20(14)
179.20(14)
-0.8(2)
102.3(2)
-19.6(2)
-76.9(2)
161.20(17)
-0.4(3)
-0.2(3)
179.83(15)
1.4(3)
-1.6(3)
178.66(18)
178.82(17)
0.9(3)
In addition, the comparison of the structural difference
between enantiomers and racemates by spectra analysis such
as FTIR, Raman, or SSNMR and XRPD patterns of
racemate and enantiomer further confirms the racemate
solid nature. It is generally accepted that identical spectra
and XRPD patterns suggest a conglomerate, similar spectra
and XRPD patterns imply a pseudoracemate and different
spectra and XRPD patterns represent a racemic compound.
The XRPD pattern of sertraline racemate was diffe-
rent from that of the corresponding enantiomer shown in
Figure 9 (top). The racemate compound had peaks at the
following 2θ angles: 8.79°, 15.23°, 17.42°, 19.25°, 21.28°,
21.89°, 23.30°, 23.84°, 24.49°, 26.48°, 28.13°, 29.09°, 32.99,
whereas the enantiomer had peaks at 8.89°, 13.46°, 15.20°,
17.65°, 19.15°, 20.31, 23.25°, 23.81°, 24.85°, 25.90°, and
28.41°, 28.90°, 31.31°, 32.45°. They have some similar peaks;
however, the enantiomer has unique peaks at 13.46°, 20.31°,
25.90°, 31.31°, 32.45° and the racemate has unique peaks at
13.12°, 21.28°, 21.89°, 26.48°, and 32.99°. These values are
tabulated in Table 3 for clarity.
angle of 84.19°. The hydrogen-bonding plays a key role in
determining the crystal packing. Surprisingly, there are no
hydrogen-bonding interactions between the heterochiral
(1R,4R) molecule and (1S,4S) molecule, whereas there are
hydrogen-bonding interactions between homochiral mole-
cules. As shown in Figure 11 (left), (1R,4R) molecules are
linked to each other head to tail by hydrogen bonds, namely,
a head H of N to a tail Cl of another neighboring molecule,
which form a catemer hydrogen-bonding motif. The hydro-
˚
gen bond is relatively strong with N-Cl = 3.229 A. The
homochiral molecules are aligned and extended along the
b-axis to form an infinite hollow column viewed from the
b-axis. (1S,4S) molecules are orientated head to tail by
hydrogen bond in the same manner displayed in Figure 11
(right), and thus another homochiral column along the b-axis
is also obtained. These (1R,4R) and (1S,4S) columns are
parallel one to another and are assembled alternatively to
form a compact crystal structure favorable to the stability
of the racemate. The packing diagrams are illustrated in
Figure 12 viewed from the a, b, and c-axis, respectively.
In the crystal of enantiomer, two (1S,4S) molecules exist in
the unit cell in Figure 13 (right). The atomic-numbering
molecule is shown in Figure 13 (left). The molecule skeleton
has almost the same shape as the racemate molecule includ-
ing the methylamine group as a head part, the phenyl and
hexane ring as a central part and a tail part of chlorine-
substituted phenyl ring. The phenyl rings and cyclohexane
ring are in a well-defined arrangement. The two aromatic
planes in one molecule make an angle of 83.33°, which is
slightly different from that in the racemate. However, there
are some disorders in the methylamine head part in the
molecule. This part can adopt two orientations relative to
the central phenyl ring and cyclohexane ring, differing from
each other in the N1 position. The atomic coordinates
of N1 in conformation 1 and N1A in conformation 2
Figure 9 (bottom) exemplifies the FTIR spectra of sertra-
line racemate and enantiomer. FTIR of the racemate is
almost superimposable on that of enantiomer and the
most notable difference is at wavenumber 1365-1394 cm-1
and 3450 cm-1
.
These FTIR spectropic analyses and XPRD patterns
provide further support of the racemic compound nature
of sertraline racemate free base.
3.4. Molecular and Crystal Structure of Sertraline. The
crystal structures of both sertraline racemate and enantiomer
were obtained through X-ray crystallographic diffraction
analysis. It was confirmed that sertraline racemate existed as
a racemic compound. The structure was solved and refined
successfully in P121/n1 space group with Z=4, whereas the
structure of the sertraline enantiomer belonged to P21 space
group with Z=2. Detailed crystal structure data are sum-
marized in Table 4.
˚
are (-0.1365, -0.1738, 0.8492) A and (-0.1058, 0.2536,
0.8708) A, respectively. The head part C11, N1, and C1 in
˚
In the crystal of the racemate, two (1R,4R) and two
(1S,4S) molecules are paired and packed in crystal unit cell
to form the racemate as shown in Figure 10 (right). All
molecules have the same conformation. The atomic-
numbering molecule is shown in Figure 10 (left). The mole-
cule skeleton basically includes a head formed by the methyl-
amine group, a central part composed by a phenyl and
cyclohexane ring from C6-C12, and a tail part of a chlorine-
substituted phenyl ring from C13-Cl20. Except for a small
deviation of C4 and C5, the central phenyl ring and cyclo-
hexane ring are in one plane. The C1, N2, and C3 are in
another plane, making an angle of 82.63 with the central
plane. The tail part aromatic ring define another plane which
is almost vertical to the central plane with an inter planar
conformation 1 forms a plane making angle of 59.28° with
the central part aromatic ring plane, which are significantly
different from that of 82.63° in the racemate, whereas the
C11A, N1A, and C1A plane in conformation 2 makes an
angle of 80.42° with the central part aromatic ring plane,
which are slightly different from that of 82.63° in the
racemate. (1S,4S) molecules are also oriented head to tail
and linked by hydrogen bonds, between a head H of N to a
tail Cl of another neighboring molecule as illustrated in
˚
Figure 14. The distance of N-Cl is 3.274 A. (1S,4S) mole-
cules are aligned along the b-axis to form an infinite hollow
column in the same way as the racemate. Thus, two columns
are parallel to one another and stack to form homochiral
crystals. The packing diagram is clearly shown in Figure 15.

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1643
Table 6. Bond Lengths, Bond Angles for Sertraline Enantiomer
˚
Bond Lengths /A
C(1)-C(9)
1.489(4)
1.515(5)
1.535(10)
0.9800
1.449(14)
0.8600
0.9600
0.9600
0.9600
1.440(5)
0.8600
0.9600
0.9600
0.9600
1.530(4)
0.9700
0.9700
1.530(4)
0.9700
0.9700
1.521(3)
1.523(3)
C(4)-H(4A)
C(5)-C(6)
C(5)-C(10)
C(5)-H(5A)
C(6)-C(7)
C(6)-H(6A)
C(7)-C(8)
C(7)-H(7A)
C(8)-C(9)
C(8)-H(8A)
C(9)-C(10)
C(10)-C(20)
C(10)-C(60)
C(20)-C(30)
C(20)-H(20A)
C(30)-C(40)
C(30)-Cl(30)
C(40)-C(50)
C(40)-Cl(40)
C(50)-C(60)
C(50)-H(50A)
C(60)-H(60A)
0.9800
C(1)-C(2)
1.372(4)
1.392(3)
0.9300
1.341(5)
0.9300
1.366(5)
0.9300
1.430(4)
0.9300
1.394(3)
1.379(3)
1.389(4)
1.390(3)
0.9300
1.384(3)
1.727(2)
1.378(4)
1.730(2)
1.369(4)
0.9300
C(1)-N(1)
C(1)-H(1B)
N(1)-C(11)
N(1)-H(1A)
C(11)-H(11A)
C(11)-H(11B)
C(11)-H(11C)
N(1A)-C(11A)
N(1A)-H(1C)
C(11A)-H(11D)
C(11A)-H(11E)
C(11A)-H(11F)
C(2)-C(3)
C(2)-H(2A)
C(2)-H(2B)
C(3)-C(4)
C(3)-H(3A)
C(3)-H(3B)
C(4)-C(10)
C(4)-C(10)
0.9300
Bond Angles /°
C(9)-C(1)-C(2)
111.7(2)
123.2(4)
89.1(5)
110.3
110.3
110.3
103.9(10)
128.1
128.1
109.5
109.5
109.5
109.5
109.5
109.5
123.8
109.5
109.5
109.5
109.5
109.5
109.5
110.5(3)
109.5
109.5
109.5
109.5
108.1
109.9(3)
109.7
109.7
109.7
109.7
108.2
110.5(3)
109.5
109.5
109.5
C(10)-C(4)-H(4A)
C(3)-C(4)-H(4A)
C(6)-C(5)-C(10)
C(6)-C(5)-H(5A)
C(10)-C(5)-H(5A)
C(7)-C(6)-C(5)
C(7)-C(6)-H(6A)
C(5)-C(6)-H(6A)
C(6)-C(7)-C(8)
C(6)-C(7)-H(7A)
C(8)-C(7)-H(7A)
C(7)-C(8)-C(9)
C(7)-C(8)-H(8A)
C(9)-C(8)-H(8A)
C(10)-C(9)-C(8)
C(10)-C(9)-C(1)
C(8)-C(9)-C(1)
C(5)-C(10)-C(9)
C(5)-C(10)-C(4)
C(9)-C(10)-C(4)
C(20)-C(10)-C(60)
C(20)-C(10)-C(4)
C(60)-C(10)-C(4)
C(10)-C(20)-C(30)
C(10)-C(20)-H(20A)
C(30)-C(20)-H(20A)
C(40)-C(30)-C(20)
C(40)-C(30)-Cl(30)
C(20)-C(30)-Cl(30)
C(50)-C(40)-C(30)
C(50)-C(40)-Cl(40)
C(30)-C(40)-Cl(40)
C(60)-C(50)-C(40)
C(60)-C(50)-H(50A)
C(40)-C(50)-H(50A)
C(50)-C(60)-C(10)
C(50)-C(60)-H(60A)
C(10)-C(60)-H(60A)
107.0
107.0
122.1(3)
118.9
118.9
119.5(3)
120.2
120.2
121.0(3)
119.5
119.5
C(9)-C(1)-N(1)
C(2)-C(1)-N(1)
C(9)-C(1)-H(1B)
C(2)-C(1)-H(1B)
N(1)-C(1)-H(1B)
C(11)-N(1)-C(1)
C(11)-N(1)-H(1A)
C(1)-N(1)-H(1A)
N(1)-C(11)-H(11A)
N(1)-C(11)-H(11B)
H(11A)-C(11)-H(11B)
N(1)-C(11)-H(11C)
H(11A)-C(11)-H(11C)
H(11B)-C(11)-H(11C)
C(11A)-N(1A)-H(1C)
N(1A)-C(11A)-H(11D)
N(1A)-C(11A)-H(11E)
H(11D)-C(11A)-H(11E)
N(1A)-C(11A)-H(11F)
H(11D)-C(11A)-H(11F)
H(11E)-C(11A)-H(11F)
C(1)-C(2)-C(3)
C(1)-C(2)-H(2A)
C(3)-C(2)-H(2A)
C(1)-C(2)-H(2B)
C(3)-C(2)-H(2B)
H(2A)-C(2)-H(2B)
C(2)-C(3)-C(4)
C(2)-C(3)-H(3A)
C(4)-C(3)-H(3A)
C(2)-C(3)-H(3B)
C(4)-C(3)-H(3B)
H(3A)-C(3)-H(3B)
C(1)-C(2)-C(3)
121.3(3)
119.4
119.4
117.0(3)
122.4(2)
120.6(3)
119.1(2)
119.0(2)
121.8(2)
118.1(2)
120.9(2)
121.0(2)
120.9(2)
119.6
119.6
120.0(2)
120.63(17)
119.40(16)
119.2(2)
119.45(19)
121.32(18)
120.5(2)
119.8
119.8
C(1)-C(2)-H(2A)
C(3)-C(2)-H(2A)
C(1)-C(2)-H(2B)
121.3(2)
119.4
119.4
Torsion Angles/°
C(9)-C(1)-N(1)-C(11)
C(2)-C(1)-N(1)-C(11)
C(9)-C(1)-C(2)-C(3)
N(1)-C(1)-C(2)-C(3)
C(1)-C(2)-C(3)-C(4)
C(2)-C(3)-C(4)-C(10)
C(2)-C(3)-C(4)-C(10)
-86.6(10)
157.7(9)
-51.0(4)
74.7(5)
C(10)-C(4)-C(10)-C(5)
C(3)-C(4)-C(10)-C(5)
C(10)-C(4)-C(10)-C(9)
C(3)-C(4)-C(10)-C(9)
C(10)-C(4)-C(10)-C(20)
C(3)-C(4)-C(10)-C(20)
C(10)-C(4)-C(10)-C(60)
-40.1(3)
-166.0(2)
143.2(2)
17.3(3)
120.2(2)
-113.3(3)
-61.7(3)
64.2(4)
-45.6(4)
-171.9(3)
0.9(5)
C(10)-C(5)-C(6)-C(7)
C(3)-C(4)-C(10)-C(60)
64.8(3)

1644 Crystal Growth & Design, Vol. 10, No. 4, 2010
He et al.
Table 6. Continued
Torsion Angles/°
C(5)-C(6)-C(7)-C(8)
C(6)-C(7)-C(8)-C(9)
C(7)-C(8)-C(9)-C(10)
C(7)-C(8)-C(9)-C(1)
C(2)-C(1)-C(9)-C(10)
N(1)-C(1)-C(9)-C(10)
C(2)-C(1)-C(9)-C(8)
N(1)-C(1)-C(9)-C(8)
C(6)-C(5)-C(10)-C(9)
C(6)-C(5)-C(10)-C(4)
C(8)-C(9)-C(10)-C(5)
C(1)-C(9)-C(10)-C(5)
C(8)-C(9)-C(10)-C(4)
C(1)-C(9)-C(10)-C(4)
-1.2(5)
0.5
0.5(4)
-178.4(3)
22.1(3)
-81.8(6)
-159.0(3)
97.0(6)
C(60)-C(10)-C(20)-C(30)
C(4)-C(10)-C(20)-C(30)
C(10)-C(20)-C(30)-C(40)
C(10)-C(20)-C(30)-Cl(30)
C(20)-C(30)-C(40)-C(50)
Cl(30)-C(30)-C(40)-C(50)
C(20)-C(30)-C(40)-Cl(40)
Cl(30)-C(30)-C(40)-Cl(40)
C(30)-C(40)-C(50)-C(60)
Cl(40)-C(40)-C(50)-C(60)
C(40)-C(50)-C(60)-C(10)
C(20)-C(10)-C(60)-C(50)
C(4)-C(10)-C(60)-C(50)
0.1(3)
178.21(19)
-0.1(3)
179.90(17)
0.0(3)
180.0(2)
179.20(17)
-0.8(3)
0.2(4)
0.1(4)
-176.5(3)
-0.9(3)
178.0(2)
175.8(2)
-5.3(3)
-179.1(2)
-0.1(4)
0.0(4)
-178.1(2)
On the basis of the above crystal structure descriptions, the
molecules in either the racemate or enantiomer have a similar
backbone consisting of head, central, and tail three parts and
are assembled in the same way. The significant differences
between the racemate and enantiomer lie in conformation of
the molecular tail structure, namely, the methylamine group.
A thorough examination of detailed bond parameters sum-
marized in Tables 5 and 6, respectively, reveals that the
torsion angles of C1-N2-C3-C8 and C1-N2-C3-C4 in
the racemate are 163.01° and -74.7°, while these of
C9-C1-N1-C11 and C2-C1-N1-C11 in the enantiomer
are -86.6° and 57.7°. The significant different torsion angles
demonstrate the molecular tail structure differences. In both
the racemate and enantiomer, identical chain-like hydrogen-
bonding patterns are observed. One molecule acts as an
acceptor for the other molecule which donates its proton.
The molecules with identical chirality are related by hydro-
gen-bonding along the b-axis to form helical columns and
paralleled columns are linked by van der Waals interactions.
It is generally believed that density of the crystal is the
most important criterion to evaluate the relative stability
between the racemate and enantiomer. The denser the crystal
is, the more stable the crystal is.12,32 Density of sertraline
racemate is higher than the enantiomer, and the hydrogen-
bonding interactions connected the molecules in the race-
mate are stronger than those in the enantiomer based on the
corresponding hydrogen bond distance. In addition, the
disorder in the enantiomer may weaken a compact crystal
packing in racemate. This evidence reveals that the racemate
crystals are packed efficiently and are more stable than
enantiomers, so the sertraline racemic mixture preferably
crystallizes as a racemic compound rather than a conglo-
merate from the standpoint of molecular structure. These
results are consistent with the conclusion drawn from the
binary melting point phase diagram as well as DSC, PXRD,
and FTIR analysis in the present work (Section 3.3).
formulation strategy and adopting novel drug delivery
systems.
The crystalline sertraline racemate was obtained, and its
solid-state nature was identified to be a racemic compound.
The crystal structures of sertraline racemate and enantiomer
are provided for the first time in the work. The racemate
crystallizes in monoclinic P121/n1 space group, whereas the
enantiomer crystallizes in monoclinic P21 space group. The
comparison of crystal structures between the racemate and
enantiomer confirm that the sertraline racemate is a racemic
compound rather than conglomerate or pseudoracemate.
Acknowledgment. The authors gratefully acknowledge
financial support of the Natural Sciences and Engineering
Council of Canada through Individual Discovery and Strate-
gic Grants. ApotexPharmaChem Inc. kindly supplied the API
used in this study.
Supporting Information Available: The calculated and experi-
mental XRPD patterns of sertraline racemate and sertraline
enantiomer; XRPD patterns of starting solid and undissolved solid
phase of sertraline enantiomers. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Zhou, M. X.; Foley, J. P. J. Chromatogr. A 2004, 1052, 13.
(2) (a) U.S. patent US005248699; (b) U.S. patent US2003055112;
(c) U.S. patent US2001041815; (d) World patent WO0032551;
(e) European patent EP0928784, (f) World patent WO0132601;
(g) World patent WO0145692.
(3) (a) CA2594198;(b) U.S. patent US2008161607;(c) U.S. patent
US2006229472;(d) World patent WO2006027658.
(4) Almarsson, O.; Hickey, M. B.; Peterson, M. L.; Morissette, S. L.;
Soukasene, S.; McNulty, C.; Tawa, M.; MacPhee, J. M.; Remenar,
J. F. Cryst. Growth Des. 2003, 3, 927.
(5) Remenar, J. F.; MacPhee, J. M.; Larson, B. K.; Tyagi, V. A.;
Ho, J. H.; McIlroy, D. A.; Hickey, M. B.; Shaw, P. B.; Almarsson,
O. Org. Process Res. Dev. 2003, 7, 990.
(6) Banerjee, P.; Bhatt, P. M.; Ravindra, N. V.; Desiraju, G. R. Cryst.
Growth Des. 2005, 5, 2299.
(7) Chow, K.; Tong, H. H. Y.; Lum, S.; Chow, A. J. Pharm. Sci. 2008,
97, 2855.
(8) Bastin, R. J.; Bowker, M. J.; Slater, B. J. Org. Process Res. Dev.
4. Conclusions
The crystalline sertraline enantiomer was prepared, char-
acterized, andcompared withitshydrochloride salt in terms of
solubility, hygroscopicity, and polymorphism propensity.
Sertraline enantiomer is found to be less polymorphic
than sertraline HCl salt, which provides an alternative for-
mulation scheme. Although the solubility of free base is much
lower than that of sertraline HCl salt, which is a disadvantage
in formulation, it can be overcome by using the proper
2000, 4, 427.
(9) Lu, J.; Rohani, S. Curr. Med. Chem. 2009, 16, 884.
(10) Taber, G. P.; Pfisterer, D. M.; Colberg, J. C. Org. Process Res. Dev.
2004, 8, 385.
(11) Kozma, D. CRC Handbook of Optical Resolutions via Diastereo-
meric Salt Formation; CRC Press: Florida, 2002.
(12) Jacques, J.; Collet, A.; Wilen, S. Enantiomers, Racemates and
Resolutions; John Wiley and Sons: New York, 1981.
(13) Wang, Y.; Chen, A. Org. Process Res. Dev. 2008, 12, 282.

Article
Crystal Growth & Design, Vol. 10, No. 4, 2010 1645
(14) Coquerel, G. Preferential Crystallization. Top. Curr. Chem. 2007,269,1.
(15) Lorenz, H.; Sheehan, P.; Seidel-Morgenstern, A. J. Chromatogr. A.
2001, 908, 201.
(16) Neau, S. H.; Shinwari, M. K.; Hellmuth, E. W. Int. J. Pharm. 1993,
999, 303.
(17) He, Q.; Zhu, J.; Gomaa, H.; Jennings, M.; Rohani, S. J .Pharm. Sci.
(24) Chen, S.; Xi, H.; Yu, L. J. Am. Chem. Soc. 2005, 127, 17439.
(25) Ostwald, W. Z. Phys. Chem. 1897, 22, 289.
(26) McCrone, W. C. Polymorphism, In Physics and Chemistry of the
Organic Solid State; Fox, D., Labes, M. M., Weissberger, A., Eds.;
Wiley-Interscience: New York, 1965; Vol. 2, p 725.
(27) Roy, S.; Aitipamula, S.; Nangia, A. Cryst. Growth Des. 2009, 5,
2261.
2009, 98, 1835.
(18) Zhang, G. Z.; Paspal, S.; Suryanarayanan, R.; Grant, D. J. W.
J. Pharm. Sci. 2003, 92, 1356.
(28) Kesisoglou, F.; Panmai, S.; Wu, Y. H. Adv. Drug Delivery Rev.
2007, 59, 631.
(19) Lorenz, H.; Seidel-Morgenstern, A. Thermochimica 2004, 415, 55.
(20) Bernstein, J. Polymorphism in Molecular Crystals; Clarendon Press:
Oxford, 2002.
(29) Jannin, V.; Musakhanian, J.; Marchaud, D . Adv. Drug Delivery
Rev. 2008, 60, 6,734.
(30) Constantinides, P. P.; Tustian, A.; Kessler, D. R. Adv. Drug
(21) Kitamura, M. Cryst. Growth Des. 2004, 4, 1153.
(22) Aaltonen, J.; Alles, M.; Mirza, S.; Koradia, V.; Gordon, K.;
Rantanen, J. Eur. J. Pharm. Biopharm. 2009, 7, 23.
(23) Mullin, J. W. Crystallization; Butterworth Heinemann Ltd: Oxford,
1993.
Delivery Rev. 2004, 56, 1243.
(31) Kaemmerer, H.; Lorenz, H.; Black, S. N.; Seidel-Morgenstern, A.
Cryst. Growth Des. 2009, 9, 1851.
(32) Brock, C. P.; Schweizer, W. B.; Dunitz, J. D. J. Am. Chem. Soc.
1991, 113, 9811.