Fluid Phase Equilibria 363 (2014) 97105

Contents lists available at ScienceDirect

Fluid Phase Equilibria

journal homepage: www.elsevier.com/locate/fluid

Solubility of carbamazepine, nicotinamide and

carbamazepinenicotinamide cocrystal in ethanolwater mixtures
Ali Shayanfar a,b, , Sitaram Velaga c , Abolghasem Jouyban a
a
b
c

Drug Applied Research Center and Faculty of Pharmacy, Tabriz University of Medical Sciences, Tabriz, Iran
Students Research Committee, Tabriz University of Medical Sciences, Tabriz, Iran
Pharmaceutical Materials Research, Department of Health Sciences, Lulea University of Technology, Lulea S-971 87, Sweden

a r t i c l e

i n f o

Article history:
Received 12 September 2013
Received in revised form 9 November 2013
Accepted 12 November 2013
Available online 22 November 2013
Keywords:
Solubility
Cocrystal
Solvent mixtures
Carbamazepine
Nicotinamide

a b s t r a c t
Solubility is an important physiochemical property of pharmaceutical compounds, and cocrystallization
is one method used to improve the solubility of drugs. Carbamazepine is a drug from class II, according to
the biopharmaceutical classication system, and it forms a cocrystal with nicotinamide. Carbamazepine
cocrystallized with nicotinamide was synthesized using the solvent evaporation approach, and its characteristics were determined using differential scanning calorimetry and powder X-ray diffractometry.
The solubility of various solid phases in ethanol + water mixtures was investigated at different temperatures using the shake-ask method, and the resulting precipitates were characterized. The solubility
of carbamazepine was increased with the addition of ethanol up to a mass fraction of 0.8. Nevertheless,
maximum solubility of NIC is observed in neat solvent (water). While the solubility of a cocrystal depends
on the concentration of the coformer and its stability in the solution.
2013 Elsevier B.V. All rights reserved.

1. Introduction
Solubility is a crucial physiochemical property in drug discovery and development [1]. Crystal engineering is claimed to be one
of the essential strategies of improving drug solubility. Solvates or
hydrates, salts, polymorphs, and cocrystals are examples of different crystal engineering strategies. A solid-state structure that
contains solvent molecules is known as a solvate, and if the solvent is water, it is called a hydrate [2]. Salt formation is another
approach used to alter the physicochemical properties of ionizable drugs [3]. Cocrystals are dened as multicomponent systems
containing two components in a stoichiometric ratio that are crystalline single-phase materials. The cocrystals are formed by means
of non-covalent interactions (often hydrogen bonding) [47].
Cosolvency (mixing solvent with water) is another method
for solubilizing and crystallizing pharmaceutical compounds [8].
Solvent mixtures can be used in cocrystalization experiments to
thermodynamically suppress solvate formation [9]. Ethanol is a
widely used cosolvent in the pharmaceutical industry and includes
dosage forms such as parenterals and soft gelatin pharmaceutical
formulations for low soluble drugs [10]. It is also one of the main

Corresponding author at: Department of Medicinal Chemistry, Faculty of Pharmacy, Tabriz University of Medical Sciences, Tabriz, Iran. Tel.: +98 4113341315;
fax: +98 4113344798.
E-mail addresses: shayanfara@tbzmed.ac.ir, alishayanf@yahoo.co.uk
(A. Shayanfar).
0378-3812/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.uid.2013.11.024

solvent systems in the crystallization of drugs [11]. The amount of

water in a crystallization solvent can affect solubility and the crystal habit of the drugs [12]. In addition, solution-mediated phase
transformation is a crucial issue in the evaluation of dissolution and
bioavailability of drugs [13]. The inuence of the ethanol + water
composition on the solubility of different drugs such as nevirapine [12], sodium naproxen [14], and carbamazepine (CBZ) [15] was
investigated in the literature. In addition, solvent mediated transformation is very important in cocrystal studies, because it can
dissociate in solution environments [7] and is a tool used to prepare
cocrystal polymorphs [16].
CBZ is an anti-epileptic drug and belongs to the biopharmaceutics classication system (BSC) class II (low solubility and high
permeability) [17]. CBZ exists in different crystalline forms with different solubility, dissolution rate, and bioavailability. CBZ is rapidly
converted to dihydrate form in water which has a lower solubility
than the anhydrate form [18,19]. It is one of the most frequently
investigated drugs in cocrystal studies to change physicochemical
properties such as solubility [20]. CBZ can form a hydrogen bond
with different coformers via its amide group [21]. Nicotinamide
(NIC) (or vitamin B3 ) is a highly soluble coformer that forms a
cocrystal with CBZ [22]. CBZ cocrystal with NIC (CBZNIC) is a common model cocrystal in crystal engineering [23,24], and different
cocrystal forms of CBZ have different levels of stability. High soluble coformers with CBZ are dissociated in water [25]. The aqueous
solubility of CBZNIC has no signicant difference with that of CBZ
according to a recent report [21], whereas signicantly different
solubility in ethanol was reported [24].

98

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

NH2

List of symbols

apparent molar enthalpy changes (kJ mol1 )

apparent molar entropy changes (J mol1 K1 )
apparent molar Gibbs free energy changes
(kJ mol1 )
relative contributions by enthalpy
% H
% TS
relative contributions by entropy
BCS
biopharmaceutics classication system
carbamazepine
CBZ
CBZNIC CBZ cocrystal with NIC
DSC
differential scanning calorimetry
K
equilibrium constant
NIC
nicotinamide
powder X-ray diffractometry
PXRD
R
gas constant
RSD
relative standard deviation
temperature (K)
T
Thm
the mean harmonic temperature
H
S
G

N
NH2

O
(a)

(b)

Fig. 1. Chemical structure of (a) CBZ, and (b) NIC.

The main objectives of this study are to determine and compare

the solubility of CBZ, NIC and CBZNIC in ethanol + water mixtures
at different temperatures and the solid state of CBZ and CBZNIC
after dissolution.
2. Experimental
2.1. Materials

The study of solubility and thermodynamic properties are

important issues in crystal engineering as they relate to controlling the phase transformation in pharmaceutical processing and
storage. The solubility of drugs as well as their crystal habits and
polymorphism can change as a function of solvent composition
and temperature [2]. These variables are also crucial in solventmediated anhydrate to hydrate transformation of CBZ [26].
Determining drug solubility is a time-consuming process, but
tting the solubility data to the mathematical model can conrm
the accuracy of data points and detect possible outlier data points
[8]. In this study, the vant Hoff equation was used to t the solubility data of drugs in a mono-solvent at different temperatures
[27]:
log xTSat = A +

B
T/K

(1)

In this model xTSat is solubility of drugs in mole fraction unit. T is

temperature in Kelvin unit, A and B are constants of the model.
Thermodynamic properties, such as apparent molar enthalpy
(H), entropy (S), and Gibbs energy (G) changes, can be calculated using a modied version of the vant Hoff equation [28,29]:
log xTSat =

H
R

 1

T/K

1
Thm /K

n

(1/(T/K))
i=1

S =

H G
Thm

(3)

(4)
(5)

Relative contributions by enthalpy (% H ) and entropy (% TS ) of

CBZ, NIC and CBZNIC solutions in ethanol + water mixtures can be
calculated by Eqs. (6) and (7), respectively.
%H = 100

|H|
|H| + |TS|

(6)

|TS|
|H| + |TS|

(7)

%TS = 100

The CBZNIC was prepared using the solvent evaporation

approach following a published method with minor modications
[21]. A 1:1 mixture of CBZ (2.363 g, 0.01 mol) and NIC (1.221 g,
0.01 mol) was dissolved in 20 mL absolute ethanol, heated, and
stirred for 30 min; then the mixture was left at 303.2 K for 72 h
for evaporation.
2.3. Characterization of CBZNIC

where n is the number of temperatures studied

G and S are computed by the following equations:
G = RThm intercept

2.2. Preparation of CBZNIC

(2)

where R is the gas constant using (8.314 J K1 mol1 ) and Thm is the
mean harmonic temperature that can be calculated using:
Thm /K =

CBZ (Fig. 1a) (IUPAC name: dibenzo[b,f]azepine-5carboxamide) (0.99 mass fraction) was purchased from the
Arasto Company (Tehran, Iran), and NIC (Fig. 1b) (IUPAC name:
pyridine-3-carboxamide) (0.99 mass fraction) was purchased from
the Zahravi Pharmaceutical Company (Tabriz, Iran). Anhydrous
ethanol (0.99 mass fraction) for preparation of solutions and
ethanol (0.935 mass fraction) for dilution of the samples were
obtained from Scharlau Chemie (Barcelona, Spain) and Jahan
Alcohol Teb (Arak, Iran), respectively. Chemicals were used as
received from the companies without further purication. Distilled
water was used to prepare the solutions. Table 1 lists a summary
of chemicals used in this work and their source and purity.

The X-ray diffraction pattern of the drug powders, the synthesized cocrystals, and the solid state of the saturated solutions were
determined using powder X-ray diffraction (PXRD) (Siemens-850,
Munich, Germany) within a range of 2 (440 ) with steps of 0.05 .
The thermal properties of the solid phase were also studied using
differential scanning calorimetry (DSC) (Shimadzu, Kyoto, Japan).
The 5 mg samples were analyzed using aluminum pans and heated
at a rate of 10 K min1 to 308.2473.2 K.
Table 1
Source and purity of chemical used in this study.
Component

Supplier

Mass
fraction
purity

CBZ
NIC
Anhydrate ethanol
Ethanol
Distilled water

Arasto Pharmaceutical Chemicals Inc.

Zahravi Pharmaceutical Company
Scharlau Chemie
Jahan Alcohol Teb
Lab made

0.99
0.99
0.99
0.94

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

Intensify

Intensity

99

a
4

14

19

24

29

34

39

14

24

Angle 2

Fig. 4. PXRD patterns of anhydrate CBZ (a), and dihydrate CBZ (b).

Fig. 2. PXRD patterns of (a) CBZ, (b) NIC, (c) CBZNIC.

2.4. Solubility determination using saturation shake-ask method

3. Results and discussion

Binary solvent mixtures of ethanol + water were prepared by

mass fraction from 0.00 to 1.00 at 0.10 intervals with the uncertainty of 0.01 in mass fraction. Solubility values for CBZ, CBZNIC,
and NIC in the solvent mixtures were determined using the classical saturation shake-ask method [30]. Excess amounts of powder
were equilibrated at 298.2, 303.2, 308.2 and 313.2 K in a shakerincubator equipped with a temperature controlling system having
an uncertainty of 0.2 K (Heidolph, Schwabach, Germany). Only
a small excess should remain in the solution to avoid difculties in sampling and the impact of an amount of excess solid.
To ensure equilibrium and completeness of possible transformation of excess solid, samples were incubated for 72 h. Attainment
of the constant concentration was veried by solubility measurements at different time intervals. Samples of the saturated
solutions were centrifuged at 10,000 rpm for 10 min (MSE Micro
Center MSB010.CX2.5, SANYO, Muriguchi City, Japan). The saturated solution of CBZ and CBZNIC were diluted with ethanol
(0.935) 204000-fold to maintain linearity according to the calibration curve, and CBZ concentrations calculated by absorbance
of the diluted solutions were measured at 284 nm [24] using a
UVvis spectrophotometer (Beckman DU-650, Fullerton, USA). In
this wavelength NIC has no signicant absorbance. The solubility
of NIC in ethanol + water mixtures was determined gravimetrically
because of its very high solubility. In this method, 0.5 mL of ltered
saturated solution was evaporated to dryness and the mass was
weighed by an analytical balance (Shimadzu, Kyoto, Japan) with
precise to 0.0001 g. Each data point is an average of at least three
repetitions.

3.1. Characterization of CBZNIC

a
b

Heat flow

308.2

328.2

348.2

368.2

388.2

34

Angle 2

408.2

428.2

448.2

T/K
Fig. 3. DSC curves of (a) CBZ, (b) NIC, (c) CBZNIC.

468.2

Fig. 2 illustrates the PXRD patterns of CBZ, NIC and CBZNIC.

It indicated that the used CBZ powder was form III according to
the study of anhydrous polymorphs of CBZ by Grzesiak et al. [31].
The distinct peaks of the cocrystal conrm the formation of a new
crystalline phase. DSC thermograms of CBZ and NIC and CBZNIC
with melting points of 463.2, 401.2, and 433.2 K, respectively, are
further evidence of the formation of a new solid phase (Fig. 3). These
characteristics are in good agreement with the data reported in the
literature [21,23].
3.2. Solubility of CBZ, NIC and CBZNIC at different temperatures
3.2.1. Thermodynamic solubility of CBZ and NIC at 298.2, 303.2,
308.2, and 313.2 K
The calibration curve properties used to determine CBZ solubility in saturated solutions are the molar absorptivities of CBZ ranging
from 10,868 to 11,421 /(L mol1 cm1 ) for known CBZ concentrations ranging from 0.000021 to 0.000127 in mol L1 unit. The mass
fraction compositions of the ethanol + water solvent mixtures and
the averages of at least three experimental measurements are tabulated in Table 2. The solubilities of CBZ and NIC in mono-solvents
reported in the literature and the solubility data reported in this
study are in good agreement (Table 3). The CBZ solubilities are
increased with the addition of ethanol; they reached maximum
values at 0.80 mass fraction of ethanol and then decreased with
further increases in ethanol concentration. Nevertheless, the maximum solubility of NIC is in neat solvent. These different proles
in ethanol + water mixtures for different drugs have been reported
by Yalkowsky et al. [32]. Solutesolvent and solventsolvent interactions (varying from the weakest interactions of van der Waals
to the strongest interactions of hydrogen bonding) are possible
mechanisms for this maximum value [33].
The crystal forms of the excess equilibrated solid phase of CBZ
in the ethanol + water mixtures were determined by PXRD and DSC
which are essential instrumental analysis methods to detect dihydrate form of CBZ [18,34]. The crystal form of each solid phase is
listed in Table 4. Figs. 4 and 5 show the PXRD patterns and DSC
thermograms of the anhydrate and dihydrate forms of CBZ. The
DSC curve of the dihydrate sample exhibits a broad endotherm at
313.2363.2 K. This peak was not observed in the anhydrate sample, and it can indicate that the dihydrate form of CBZ and the PXRD
pattern of this form shows sharp and distinctive peaks at 2 = 8.7 .
The results of PXRD and DSC of dihydrate form of CBZ are consistent with previous studies [15,18]. In low fractions of ethanol
in solvent mixtures, the anhydrate CBZ is converted to dihydrate

100

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

Table 2
Experimental solubility (Sexp /mol L1 ) of carbamazepine (CBZ), nicotinamide (NIC) and CBZNIC in ethanol + water at different temperatures and relative standard deviation
(RSD) at atmospheric pressure.
Mass fraction of ethanol

Ta (K)

Sexp, CBZ b

RSD

Sexp, NIC c

RSD

Sexp, CBZNIC d

RSD

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

298.2
298.2
298.2
298.2
298.2
298.2
298.2
298.2
298.2
298.2
298.2
303.2
303.2
303.2
303.2
303.2
303.2
303.2
303.2
303.2
303.2
303.2
308.2
308.2
308.2
308.2
308.2
308.2
308.2
308.2
308.2
308.2
308.2
313.2
313.2
313.2
313.2
313.2
313.2
313.2
313.2
313.2
313.2
313.2

8.00 104
1.90 103
4.00 103
1.24 102
2.96 102
5.30 102
9.58 102
1.15 101
1.56 101
1.40 101
1.08 101
1.00 103
2.10 103
4.90 103
1.53 102
3.57 102
6.53 102
1.08 101
1.40 101
1.78 101
1.57 101
1.24 101
1.10 103
2.50 103
6.80 103
2.04 102
4.41 102
8.29 102
1.33 101
1.56 101
2.00 101
1.73 101
1.28 101
1.20 103
3.10 103
8.00 103
2.27 102
5.16 102
1.00 101
1.52 101
1.83 101
2.13 101
1.87 101
1.34 101

2.1
6.4
8.1
0.8
6.4
6.4
7.8
7.6
6.3
3.2
5.0
7.2
5.7
4.6
4.1
0.9
3.8
4.1
7.8
5.3
5.9
4
7.1
1.0
6.7
6.6
5.0
4.2
8.4
4.4
5.6
5.0
8.2
7.4
2.3
6.1
4.3
6.2
7.0
3.5
8.6
2.5
2.0
4.1

4.06
3.96
3.87
3.81
3.71
3.59
3.13
2.75
2.24
1.60
0.80
4.33
4.22
4.20
3.97
3.95
3.71
3.37
3.07
2.48
1.75
0.94
4.73
4.60
4.42
4.43
4.15
3.97
3.74
3.28
2.75
1.96
0.99
5.22
5.14
4.99
4.83
4.66
4.48
4.12
3.73
2.88
2.22
1.21

1.4
2.2
2.7
3.6
2.1
2.1
6.0
2.3
2.7
7.7
7.1
3.1
0.5
2.0
5.2
2.6
0.7
5.9
5.0
4.6
6.9
5.3
3.0
1.7
4.7
1.8
0.8
4.4
8.2
4.6
1.8
3.7
3.5
6.8
2.3
2.9
5.7
2.8
2.1
6.4
4.1
4.0
8.2
2.5

8.00 104
2.00 103
4.00 103
1.17 102
3.00 102
5.41 102
8.45 102
1.58 101
1.75 101
1.55 101
1.13 101
9.00 104
2.30 103
5.10 103
1.50 102
3.29 102
6.65 102
8.92 102
1.83 101
1.96 101
1.72 101
1.57 101
1.10 103
2.70 103
6.50 103
1.89 102
4.09 102
8.22 102
1.03 101
1.88 101
2.31 101
1.88 101
1.66 101
1.30 103
3.00 103
8.40 103
2.42 102
5.72 102
9.80 102
1.17 101
2.22 101
2.51 101
2.04 101
1.67 101

5.4
8.1
6.4
5.6
3.1
2.0
2.0
5.3
7.9
7.3
3.1
6.4
2.9
8.4
8.6
9.0
6.5
6.8
8.2
9.3
7.0
7.6
7.2
8.4
8.7
3.2
3.9
6.4
4.2
4.7
7.0
1.9
1.2
8.9
2.4
7.3
4.5
5.8
6.8
6.7
8.9
6.6
4.9
6.8

a
b
c
d

Standard uncertainties are: u(T) = 0.1.

Standard uncertainties ranged from  n 1 = 0.0001 to 0.0131 mol L1 .
Standard uncertainties ranged from  n 1 = 0.02 to 0.32 mol L1 .
Standard uncertainties ranged from  n 1 = 0.0004 to 0.0183 mol L1 .

form, because its formation depends on water content. The results

of this study are in good agreement with the literature data [15,35].
The correlation solubility data of CBZ and NIC in ethanol + water
mixture shown in Table 5, demonstrates that there is a linear
relation with an acceptable regression coefcient (R > 0.96) with
increasing temperatures, based on Eq. (2). (The cocrystal is partially
(or totally) unstable in ethanol-water mixture (CBZ and cocrystal are available in solution, simultaneously) and polymorphic
changes were observed in neat ethanol therefore these data in
which the crystal forms were not identical were excluded from
analysis.)

3.2.2. Apparent solubility of CBZNIC at 298.2, 303.2, 308.2, and

313.2 K
The solubility values of CBZNIC and the equilibrated solid phase
of CBZNIC are listed in Tables 2 and 4, respectively. The characteristics of the resulting precipitated solid phase of CBZNIC in
ethanol + water could help to determine the solution stability of
CBZNIC in the studied solvent mixtures. The PXRD patterns (Fig. 6)
and DSC thermograms (Fig. 7) of the studied solid phases show
that CBZNIC was dissociated and converted to dihydrate CBZ from
0.0 to 0.50 mass fractions of ethanol, because NIC is a highly soluble coformer in water; therefore, CBZNIC becomes unstable and

Table 3
Comparison of solubility data in monosolvents studied in this work and from the literature.
Solutea

Solvent

T (K)

Sexp (mol L1 ) in this study

Sexp (mol L1 ) from the literature

Refs.

CBZ
CBZ
NIC
NIC
CBZNIC

Ethanol
Water
Ethanol
Water
Ethanol

298.2
298.2
298.2
298.2
298.2

0.108
0.001
0.797
4.064
0.113

0.108
0.001
0.841
4.080
0.116

[24]
[41]
[24]
[22]
[24]

Carbamazepine (CBZ), nicotinamide (NIC) and CBZ in cocrystal form with NIC (CBZNIC).

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

101

Table 4
Solid states of carbamazepine (CBZ) and CBZ + nicotinamide cocrystal (CBZNIC) in
the saturated solutions.
Solid state in
saturated solution
of CBZ

Solid state in saturated

solution of CBZNIC

0.0
0.1
0.2
0.3
0.4
0.5
0.6

Hydrate
Hydrate
Hydrate
Hydrate
Hydrate
Hydrate
Hydrate

0.7

Hydrate

0.8a

Anhydrate

0.9

Anhydrate

1.0

Anhydrate

Dihydrate CBZ
Dihydrate CBZ
Dihydrate CBZ
Dihydrate CBZ
Dihydrate CBZ
Dihydrate CBZ
Dihydrate
CBZ + cocrystal
Dihydrate
CBZ + cocrystal
Anhydrate
CBZ + cocrystal
Anhydrate
CBZ + cocrystal
Cocrystal

Intensity

Mass fraction
of ethanol

Solid states of carbamazepine (CBZ), alone and in cocrystal form in 0.8 mass
fraction of ethanol was converted to dihydrate form at 298.2.

12

16

20

24

28

32

36

40

Ang le 2
a

Heat flow

Fig. 6. PXRD patterns of (a) CBZ anhydrate form, (b) CBZ dihydrate form, (c) mixture
of CBZ anhydrate + cocrystal (collected solid materials after equilibrating of cocrystal
at 0.9 mass fraction of ethanol), (d) CBZ dihydrate form + cocrystal (excess solid of
CBZNIC after equilibrating of cocrystal at 0.60.7 mass fraction of ethanol, and (e)
CBZNIC cocrystal.

308.2 328.2 348.2 368.2 388.2 408.2 428.2 448.2 468.2

T/K

Heat flow

Fig. 5. DSC curves of (a) anhydrate CBZ, (b) dihydrate CBZ.

converts to CBZ dihydrate in polar environments such as water or

water-rich mixtures.
From 0.60 to 0.90 mass fractions of ethanol, however, a part
of CBZNIC is changed to the dihydrate or anhydrate form of CBZ.
The DSC curve showed three peaks: the rst peak at 313.2364.2 K
indicated the dihydrate form, and the second and third peaks at
433.2 K and 453.2 K corresponded to the CBZNIC and CBZ,
respectively. In 0.90 mass fraction of ethanol, peaks of anhydrate
CBZ and CBZNIC were observed. The DSC curves were compatible with those from the literature [18,35]. A sharp peak in
2 = 8.9 (dihydrate form) and distinct peak at 2 = 7 , 10 , 13 ,
20.5 (CBZNIC) indicated dihydrate CBZ + cocrystal, and the peak
in 2 = 25 (CBZ) and the distinct peak at 2 = 7 , 10 , 13 , 20.5 in
the fraction of 0.9 of ethanol indicated anhydrate CBZ + cocrystal.
CBZNIC is completely stable in ethanol, and there was no
signicant difference in the PXRD patterns of the cocrystal after
equilibrating in ethanol at 298.2, 303.2, and 308.2 K. Signicant
changes in the PXRD pattern of the saturated solution (Fig. 8) might
be indicative of a new polymorph of CBZNIC at this temperature.
Polymorphism in carbamazepine cocrystals such as CBZNIC, CBZ
with saccharin and isonicotinamide have been reported in the literature earlier [16,36,37].
Reported solubility data of CBZNIC are not thermodynamic solubility data, because CBZNIC is not stable in 0.00.9 mass fraction
of ethanol. Rodrguez-Hornedo and coworkers recommend that,

308.2

328.2

348.2

368.2

388.2

408.2

428.2

448.2

468.2

T/K
Fig. 7. DSC curves of (a) anhydrate CBZ + cocrystal, (b) dihydrate CBZ + cocrystal.

Table 5
Regression coefcient (R) of tting solubility data of carbamazepine (CBZ), nicotinamide (NIC) in ethanol + water at different temperatures to Eq. (1).
Mass fraction of ethanol

CBZ
R

NIC
R

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

0.974
0.985
0.993
0.988
0.999
0.999
0.995
0.995
0.986a
0.998
0.949

0.994
0.991
0.985
0.985
0.978
0.961
0.997
0.993
0.991
0.996
0.979

a
Solubility datum of CBZNIC at 298.2 K, was excluded from analysis because CBZ
was in dihydrate form.

102

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

Intensity

12

16

20

24

28

32

36

Fig. 10. Solubility of CBZNIC in 60% of ethanol at 298.2 K in the presence of different
excess concentrations of NIC.

40

Angle 2

Fig. 8. PXRD patterns of (a) cocrystal, (b) cocrystal after equilibrating in ethanol at
313.2 K. The changes in PXRD pattern were shown with arrow.

due to solution conditions, cocrystal stability, and the lack of control

over cocrystal purity, these cocrystal solubility data in this study
referring to non-equilibrium conditions [38] and they are apparent
solubility data.

0.30

0.30

0.25

0.25

-1
Concentration (mol.L )

-1
Concentration (mol.L )

3.2.3. Comparison of solubility of CBZ with CBZNIC in

ethanol + water mixtures
Fig. 9 illustrates the solubility prole of CBZ and/or CBZNIC in
ethanol + water mixtures at different temperatures. From 0.0 to 0.5
mass fractions of ethanol, there are no signicant changes in the
solubility of CBZ. Additional investigations on the excess solids collected from the saturated solutions of these mixtures revealed that
the cocrystal was dissociated and transformed to CBZ dihydrate

because of the instability of CBZNIC, and no differences in solubility are expected. In fractions of 0.71.0 of ethanol, however, the
apparent solubility of CBZNIC was increased because of the partial (in the fraction of 0.70.9 ethanol) or total cocrystal (in neat
ethanol) in the solutions. A previous study by Nehm et al. [24]
showed that CBZNIC has a signicantly higher solubility than CBZ
in ethanol, and their reported solubility data is in agreement with
our ndings (see Table 2). The important point is the decreased
value of apparent solubility in the 0.60 fraction of ethanol compared with those of the thermodynamic solubility of CBZ in all
studied temperatures. To understand the reason for this decrease,
the solubility of CBZNIC was determined in the presence of different excess concentrations of NIC. The results are shown in Fig. 10.
The solubility of CBZ was reduced in 1030 g L1 and increased in
40, 50 and 100 g L1 of NIC, respectively.

0.20
0.15
0.10
CBZ
0.05

CBZ-NIC

0.20
0.15
0.10
CBZ
0.05

CBZ-NIC

0.00
0.0

0.2

0.4

0.6

0.8

0.00

1.0

0.0

Mass fraction of ethanol

0.2

0.6

0.8

1.0

(b)
0.30

0.25

0.25

-1
Concentration (mol.L )

-1
Concentration (mol.L )

(a)
0.30

0.20
0.15
0.10
CBZ
0.05

0.4

Mass fraction of ethanol

CBZ-NIC

0.20
0.15
0.10
CBZ
0.05

CBZ-NIC

0.00

0.00
0.0

0.2
0.4
0.6
Mass fraction of ethanol

(c)

0.8

1.0

0.0

0.2

0.4
0.6
0.8
Mass fraction of ethanol

(d)

Fig. 9. Solubility of CBZ and CBZNIC in ethanol + water mixtures at (a) 298.2 K, (b) 303.2 K, (c) 308.2 K, (d) 313.2 K.

1.0

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

103

40
35

0.2
0.5

-1

H (KJ.mol )

30

0.4

25

0.3

0.7

0.6

0.1

20
15

0.9

0.0

0.8
1.0

10
5
0
5.0

10.0

15.0

20.0

25.0

30.0

-1

G (KJ.mol )
(a)
25

-1

H (KJ.mol )

20

1.0

15
0.6

0.9
0.7
0.8

0.20.1 0.0
0.3
0.4
0.5

10
5
0
5.0

5.5

6.0

6.5

7.0

7.5

8.0

-1

G (KJ.mol )
(b)
Fig. 11. Enthalpyentropy compensation plots H vs. G for solubility of (a) CBZ, and (b) NIC in ethanol + water mixtures.

The solidliquid equilibrium can be dened for the cocrystal of

CBZNIC as [39]:
K

CBZ NICCBZsolution + NICsolution

The solubility of the cocrystal depends on the concentration of
coformer in the solution. The dissociation of the cocrystal in the
solution is dened by the equilibrium constant (K), which is dened
as a product of drug and coformer solution concentrations. According to Le Chateliers principle [40], with increased concentrations
of NIC, a decreased solubility of CBZ will be observed [24] while
higher concentrations of NIC produce stable cocrystals (do not dissociate to CBZ) and elevated solubility [20]. The enhanced solubility

results of this study are in agreement with the literature in which

cocrystals are considered one of the most important methods for
drug solubilization. The solubility of the cocrystal and its stability
depend on the concentration of the coformer in the solution [20,24].
3.3. Thermodynamic properties of CBZ and NIC in ethanol + water
mixtures
The solubility values in mol L1 unit were used to calculate the
saturated mole fraction solubility of solute based on Eq. (8)
x1 =

m1 /Mw1
m1 /Mw1 + m2 /Mw2 + m3 /Mw3

(8)

104

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

Table 6
Thermodynamic properties of carbamazepine (CBZ), nicotinamide (NIC) and CBZ in cocrystal form with NIC (CBZNIC) solutions in ethanol + water mixtures at 305.7 K.
Mass fraction of ethanol

G (kJ mol1 )

H (kJ mol1 )

S (J mol1 K1 )

TS (kJ mol1 )

%H

%TS

CBZ
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8a
0.9
1.0

27.7
25.4
22.9
19.9
17.6
15.8
14.2
13.4
12.6
12.4
12.6

20.4
25.5
37.4
32.7
29.2
33.3
24.5
23.2
14.3
15.0
10.6

23.8
0.2
47.3
41.7
37.8
57.3
33.8
31.9
5.8
8.7
6.4

7.3
0.1
14.5
12.8
11.6
17.5
10.3
9.8
1.8
2.6
1.9

73.8
99.7
72.1
71.9
71.6
65.5
70.4
70.4
89.0
85.0
84.5

26.2
0.3
27.9
28.1
28.4
34.5
29.6
29.6
11.0
15.0
15.5

NIC
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

6.5
6.4
6.3
6.2
6.0
5.9
5.9
5.8
6.0
6.4
7.4

12.0
12.4
11.6
11.6
10.3
10.2
13.0
13.7
12.1
15.5
19.3

18.0
19.6
17.2
17.9
13.9
14.0
23.3
25.6
19.9
29.9
38.7

2.0
2.0
1.9
1.9
1.8
1.8
1.8
1.8
1.8
1.9
2.3

68.7
67.5
68.7
68.0
70.8
70.4
64.6
63.6
66.5
62.9
61.9

31.3
32.5
31.3
32.0
29.2
29.6
35.4
36.4
33.5
37.1
38.1

Thermodynamic data is calculated at 308.2 K.

where m1 , m2 and m3 represents the mass and Mw1 , Mw2 and

Mw3 are the molecular weights of the solute, ethanol and water,
respectively.
Table 6 shows the thermodynamic properties including H, S,
and G in ethanol + water mixtures at 305.7 K. Because of the dihydrate form of CBZ at 298.2 K in 0.80 mass fraction, the instability of
CBZNIC in 0.0 to 0.9 mass fractions of ethanol, and the possibility
of polymorphic change of the CBZNIC in ethanol, these data were
excluded from thermodynamic analysis.
The enthalpy and entropy of solution are positive in all cases
except neat ethanol and water (for CBZ) which the entropy values are negative. The relative contribution of the enthalpy and
entropy of CBZ (all fractions of ethanol + water solvent mixtures)
shows that the main contribution to free energy of the solution process is enthalpy. The enthalpyentropy compensation of CBZ and
NIC were applied to investigate the mechanism of solubilization in
ethanol + water mixtures from the thermodynamic points (Fig. 11).
The results indicate complex and non-linear enthalpyentropy
compensation with negative and positive slopes for CBZ and NIC
in the studied solvent system.

4. Conclusion
Maximum solubility value of a drug in binary solvent mixture
is dependent upon solutesolvent and solventsolvent interactions. The results of this study conrm that solvent composition
and temperatures are critical parameters in drug transformation
in solution. CBZNIC cocrystal is unstable in water, because NIC is
highly soluble in water-rich regions of ethanol + water mixtures.
With increases in the ethanol fraction, the instability of CBZNIC is
decreased and a more apparent solubility than CBZ solubility in solvent mixtures was observed. Moreover, the solubility of CBZNIC
depends on the coformer solution concentration and cocrystal stability in solvent mixtures. Overall, it is concluded that coformer
solution concentration is a crucial parameter to the alteration of
the solubility and stability of cocrystals.

Acknowledgments
This article is a part of the results of the PhD thesis No. 64 submitted to the Faculty of Pharmacy, Tabriz University of Medical
Sciences, Tabriz, Iran. The authors would like to thank PXRD lab,
Institute of Mineralogy, North Western Regional Ofce, Tabriz, Iran
for providing PXRD patterns.

References
[1] L. Di, P.V. Fish, T. Mano, Drug Discov. Today 17 (2012) 486495.
[2] N. Blagden, M. de Matas, P.T. Gavan, P. York, Adv. Drug Deliv. Rev. 59 (2007)
617630.
[3] A.T.M. Serajuddin, Adv. Drug Deliv. Rev. 59 (2007) 603616.
[4] S. Aitipamula, R. Banerjee, A.K. Bansal, K. Biradha, M.L. Cheney, A.R. Choudhury, G.R. Desiraju, A.G. Dikundwar, R. Dubey, N. Duggirala, P.P. Ghogale, S.
Ghosh, P.K. Goswami, N.R. Goud, R.R.K.R. Jetti, P. Karpinski, P. Kaushik, D.
Kumar, V. Kumar, B. Moulton, A. Mukherjee, G. Mukherjee, A.S. Myerson, V. Puri,
A. Ramanan, T. Rajamannar, C.M. Reddy, N. Rodriguez-Hornedo, R.D. Rogers,
T.N.G. Row, P. Sanphui, N. Shan, G. Shete, A. Singh, C.C. Sun, J.A. Swift, R. Thaimattam, T.S. Thakur, R. Kumar Thaper, S.P. Thomas, S. Tothadi, V.R. Vangala, N.
Variankaval, P. Vishweshwar, D.R. Weyna, M.J. Zaworotko, Cryst. Growth Des.
12 (2012) 21472152.
[5] N.J. Babu, A. Nangia, Cryst. Growth Des. 11 (2011) 26622679.
[6] N. Shan, M. Zaworotko, Drug Discov. Today 13 (2008) 440446.
[7] N. Schultheiss, A. Newman, Cryst. Growth Des. 9 (2009) 29502967.
[8] A. Jouyban, Handbook of Solubility Data for Pharmaceuticals, CRC Press, Boca
Raton, 2010.
[9] T. Rager, R. Hilker, Cryst. Growth Des. 10 (2010) 32373241.
[10] R.G. Strickley, Pharm. Res. 21 (2004) 201230.
[11] A. Nokhodchi, N. Bolourtchian, R. Dinarvand, J. Cryst. Growth 274 (2005)
573584.
[12] N. Stieger, M.R. Caira, W. Liebenberg, L.R. Tiedt, J.C. Wessels, M.M. De Villiers,
Cryst. Growth Des. 10 (2010) 38593868.
[13] K. Greco, R. Bogner, J. Pharm. Sci. 101 (2011) 29963018.
[14] K.J. Chavez, R.W. Rousseau, Cryst. Growth Des. 10 (2010) 38023807.
[15] H. Qu, M. Louhi-Kultanen, J. Kallas, Int. J. Pharm. 321 (2006) 101107.
[16] J.H. Ter Horst, P.W. Cains, Cryst. Growth Des. 8 (2008) 25372542.
[17] P. Zakeri-Milani, M. Barzegar-Jalali, M. Azimi, H. Valizadeh, Eur. J. Pharm. Biopharm. 73 (2009) 102106.
[18] Y. Kobayashi, S. Ito, S. Itai, K. Yamamoto, Int. J. Pharm. 193 (2000) 137146.
[19] H. Qu, M. Louhi-Kultanen, J. Rantanen, J. Kallas, Cryst. Growth Des. 6 (2006)
20532060.
[20] D.J. Good, N. Rodri guez-Hornedo, Cryst. Growth Des. 9 (2009)
22522264.

A. Shayanfar et al. / Fluid Phase Equilibria 363 (2014) 97105

[21] Z. Rahman, C. Agarabi, A.S. Zidan, S.R. Khan, M.A. Khan, AAPS PharmSciTech 12
(2011) 693704.
[22] J. Wouters, R. Sandrine, L. Quere, Monographs of most frequent co-crystal formers, in: J. Wouters, L. Quere (Eds.), Pharmaceutical Salts and Co-crystals, Royal
Society of Chemistry, London, 2011, pp. 338382.
[23] N. Chieng, M. Hubert, D. Saville, T. Rades, J. Aaltonen, Cryst. Growth Des. 9
(2009) 23772386.
[24] S.J. Nehm, B. Rodrguez-Spong, N. Rodrguez-Hornedo, Cryst. Growth Des. 6
(2006) 592600.
[25] S.L. Childs, N. Rodrguez-Hornedo, L.S. Reddy, A. Jayasankar, C. Maheshwari, L. McCausland, R. Shipplett, B.C. Stahly, CrystEngComm 10 (2008)
856864.
[26] H. Qu, T. Munk, C. Cornett, J.X. Wu, J.P. Btker, L.P. Christensen, J. Rantanen, F.
Tian, Pharm. Res. 28 (2011) 364373.
[27] D.J.W. Grant, M. Mehdizadeh, A.H.L. Chow, J.E. Fairbrother, Int. J. Pharm. 18
(1984) 2538.
[28] M. Gantiva, F. Martnez, Fluid Phase Equilib. 293 (2010) 242250.
[29] V. Panahi-Azar, A. Shayanfar, F. Martnez, W.E. Acree Jr., A. Jouyban, Fluid Phase
Equilib. 308 (2011) 7277.

105

[30] S.B. Murdande, M.J. Pikal, R.M. Shanker, R.H. Bogner, Pharm. Dev. Technol. 16
(2011) 187200.
[31] A.L. Grzesiak, M. Lang, K. Kim, A.J. Matzger, J. Pharm. Sci. 92 (2003) 22602271.
[32] S.G. Machatha, P. Bustamante, S.H. Yalkowsky, Int. J. Pharm. 283 (2004) 8388.
[33] W.E. Acree Jr., Thermochim. Acta 198 (1992) 7179.

Hadzidedic,
G. Betz, S.
S.K. El-Arini, H. Leuenberger, Int. J. Pharm. 386
[34] S. Sehi
c,
(2010) 7790.
[35] Y. Li, P.S. Chow, R.B.H. Tan, S.N. Black, Org. Proc. Res. Dev. 12 (2008) 264270.
[36] W.W. Porter Iii, S.C. Elie, A.J. Matzger, Cryst. Growth Des. 8 (2008) 1416.
[37] A.B. Buanz, G.N. Parkinson, S. Gaisford, Cryst. Growth Des. 11 (2011)
11771181.
[38] R. Thakuria, A. Delori, W. Jones, M.P. Lipert, L. Roy, N. Rodrguez-Hornedo,
Pharmaceutical cocrystals and poorly soluble drugs, Int. J. Pharm. 453 (2013)
101125.
[39] K.C. Lee, K.J. Kim, Chem. Eng. Technol. 34 (2011) 619623.
S.L. Childs, S.A. Rizvi, W. Jones, CrystEngComm 11 (2009) 418426.
[40] T. Frisc ic,
[41] S.H. Yalkowsky, Y. He, P. Jain, Handbook of Aqueous Solubility Data, second ed.,
CRC Press, Boca Raton, 2010.