Behavior of LiOH:H2O crystals obtained by evaporation and by drowning out

Cryst. Res. Technol. 43, No. 6, 616 – 625 (2008) / DOI 10.1002/crat.200711110

Behavior of LiOH?H2O crystals obtained by evaporation and by
drowning out
T. A. Graber*, J. W. Morales, P. A. Robles, H. R. Galleguillos, and M. E. Taboada
Centro de Investigación Científica y Tecnológica para la Minería (CICITEM), Departamento de Ingeniería
Química, Universidad de Antofagasta, Av. Angamos 601, Antofagasta, Chile
Received 29 August 2007, revised 8 February 2008, accepted 18 February 2008
Published online 20 March 2008
Key words lithium hydroxide, crystallization, crystal polymorphism.
PACS 81.10.-h
In the present work, the behavior of crystals derived from two different crystallization methods applied in a
concentrated aqueous lithium salt solution was studied. The LiOH?H2O crystals obtained by a simple
evaporation (Crystal I) differed in terms of morphology and solubility from those precipitated from lithium
hydroxide solutions by addition of ethanol as a co-solvent (Crystal II). Solubility of Crystal II at different
temperatures (15 to 35°C) and mass ratios of ethanol to water (0 to 0.1) was determined. Polymorphic like
behavior of these crystals was evidenced from X-ray diffraction patterns. Measurement of density, refractive
index, absolute viscosity and electrical conductivity of saturated solutions are reported. A thermodynamic
analysis in terms of the Chen model for the calculation of activity coefficients, indicate that the polymorphic
system in water and in water + ethanol (ethanol/water ratio 0.1) is enantiotropic.
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1 Introduction
Lithium hydroxide monohydrate (LiOH?H2O) is produced industrially from the reaction of lithium carbonate
with calcium hydroxide in aqueous solution. This compound is used in the production of lubricating greases
which are resistent to elevated working temperatures, and is also used in the production of inks. The anhydrous
form of lithium hydroxide is produced by heating the monohydrated form to above 100 °C is used for
purification of gases and air (as a carbon dioxide absorbent), as a heat transfer medium, as a storage-battery
electrolyte, as a catalyst for polymerization, in ceramics, manufacturing other lithium compounds and
esterfication especially for lithium stearate.
This study show the possibility of producing LiOH?H2O in two crystalline forms, known as polymorphism
or ability of a solid material to exist in more than one form or crystal structure [1]. Resulting from different
arrangements of molecular packing, polymorphs are different crystalline forms (at different free energy states)
of the same molecule or molecules [2]. This condition in various compounds has become of interest in recent
years due to its economic significance in various industrial processes, with important impacts in
pharmaceuticals, paints and pigments, explosives, and in the food sector [3]. In pharmaceuticals fields,
knowledge of crystalline polymorphism is important since different crystalline forms of a drug can vary in
bioavailability, chemical stability and industrial properties [4]. Different methods exist for identification of
crystalline polymorphism such as determination of solubilities, thermal analysis, X-ray diffraction,
thermomicroscopy, Raman spectroscopy, infrared spectroscopy, and others [5, 6]. In the present study, crystals
of LiOH·H2O, obtained by simple evaporation (Crystal I) and by precipitation from an aqueous solutions of the
hydroxide by a co-solvent (Crystal II), demonstrated different solubilities which presuppose polymorphism. A
subsequent analysis of the crystals by X-ray diffraction appeared to support this assumption.
____________________
* Corresponding author: e-mail: tgraber@uantof.cl


© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---

Cryst. Res. Technol. 43, No. 6 (2008)
617

In previous studies [7], our group reported solubility values and physical properties of Crystal I in different
water/ethanol mixtures at 25°C, and also in water/ethanol mixtures with ethanol/water ratios of 0 and 0.1 at 15,
20, 25, 30, and 35°C [8]. As indicated in Taboada et al. [8], the properties which are significantly altered under
ethanol addition are: density that decreases from 1.1120 to 1.0720 g/cm3 at 25°C, when changing the solvent
from water to water-ethanol (ethanol water ratio of 0.1), and ionic conductivity decreases from 395 to
263 µS/cm at 25°C, when changing the solvent from water to water-ethanol (ethanol/water ratio of 0.1) This
paper presents measurements on Crystal II similar to those made on Crystal I in previous studies.
Burger and Ramberger [9,10] developed a procedure for the classification of two different types of
polymorphic systems (monotropic and enantiotropic) based on thermodynamic rules. Recently, Park et al. [11]
classified three polymorphic materials as enantiotropic, plotting the data of solubility vs. 1/T following the
van't Hoff equation. These authors, and general observations in this type of analysis, consider that the
polymorphic system behaves as an ideal system. It is, however, unreasonable to assume the preceding in a
system containing electrolytes. For this reason the present study included application of similar methodology to
the used by Park et al. [11], but included the quantification of the activity coefficients in a thermodynamic
analysis of the solubilities. The Chen model [12,13] was used to estimate LiOH?H2O activity coefficients of
both crystalline forms.
2 Experimental
Crystallization by evaporation The crystals for the study were obtained by evaporation in a Heidolph
OB2000 rotary evaporator working at 60°C, 26.6 mm Hg and 240 rpm. Reagents used in the experiments
included analytical grade lithium hydroxide monohydrate (+99 mass %) Merck and ultrapure water
(conductivity 0.054 ?S/cm). Initially, a saturated solution was prepared with the hydroxide at 40°C. Solubility
data of LiOH?H2O at 40 and 60°C were obtained from the literature, and used to calculate the water/hydroxide
composition entering and exiting the evaporation system. The type I crystals were screened from the solution
using # 18 (1 mm) and 200 (0.075 mm) Tyler meshes. The evaporation rate recorded for this experiment varied
between 4 and 5 cm3/min.
Crystallization by addition of a co-solvent This method involved the precipitation of LiOH?H2O crystals
(type II) by a drowning out process in which ethanol was added to aqueous solutions of LiOH. Preliminary
trials were carried out in a 2000 cm3 jacketed crystallizing flask equipped with temperature control at 25 +
0.1°C and mechanical stirring. Measurements were carried using 70 and 90 % ethanol (+99.5 mass %, Merck),
and aqueous solutions prepared with 11 % by weight LiOH?H2O (+99 mass %, Merck). Each measurement was
carried out in duplicate. The total concentration of the hydroxide was 8 % by weight LiOH in all the
measurements. Stirring speed was 500 rpm in all tests, and the speed of alcohol addition was 7, 30 or
60 cm3/min. Once the ethanol solution had been added to the crystallizer, stirring was maintained for 15 min.
followed by a decantation period of 15 min. Samples of supernatant liquid were taken through a filter-equipped
syringe, and analyzed for concentration of the hydroxide. The supernate was filtered off by vacuum, and the
crystals were dried and screened. Tyler meshes ranging from # 20 (0.85 mm) to # 200 (0.075 mm) were used in
determining the crystal size distributions.
Phase equilibrium studies Given that the crystals obtained from simple evaporation (Crystal I) were
different from those obtained by the addition of a co-solvent (Crystal II), phase equilibrium studies were
carried out on the Crystal II, as well as measurements in the properties of the saturated solutions; similar
studies on Crystal I were reported previously [7,8]. Five (duplicate) measurements were carried out on the
LiOH + C2H5OH + H2O ternary system at mass ethanol/water ratios between 0 and 0.25 in 20 cm3 flasks. Type
II LiOH?H2O crystals were added to each flask in excess. The samples were stirred in a thermostatic bath
controlled at 25 + 0.1°C for three days until reaching equilibrium. The samples were then allowed to decant a
further day at 25°C. Solubility of the hydroxide was determined by chemical analysis, and samples for further
determinations were removed from the flask using a filter-equipped syringe acclimated to the working
temperature to avoid crystallization in the subsamples. A second set of measurements was carried out using all
the same procedures, but at working temperatures of 15, 20, 25, 30 and 35 ºC, and at mass ethanol/water ratios
of 0.0 and 0.1.
Measurements of physical properties The density of each sample was determined using a Mettler
Toledo model DE50 vibrating tube densimeter, operating in the static mode, having a precision of
+5 x 10-5 g/cm3.
www.crt-journal.org
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---

618
T. A. Graber et al.: LiOH?H2O crystals obtained by evaporation and by drowning out

Refractive indexes for the sodium D line were measured on a Mettler Toledo RE40 refractometer with
precision of + 1 x 10-4. Measurements of electrical conductivity were carried out using an Orion model 170
conductimeter, calibrated with a 0.2 M standard KCl solution having a precision of ± 0.6 S.m-1.
The kinematics viscosities of the saturated solutions were measured with an automatic Schott-Gerate AVS
310 laser viscosimeter which is based in transit time determination of the liquid meniscus through a capillary,
with a precision of ± 0.1 s. A calibrated Micro-Oswald viscosimeter was immersed in a transparent Schott-
Gerate CT 52 thermostatic bath, with a temperature precision of ±0.05 K. The dynamic viscosity was obtained
by multiplying the kinematic viscosity by the corresponding density. The uncertainty of the measured
viscosities was better than ±5 x 10-3 mPa.s.
3 Results and discussion
Analysis of the crystal II X-ray diffraction is the most useful technique for studying the crystalline
polymorphism of a sample since the diffraction patterns of the diverse crystalline forms always demonstrate
differences [14-16]. Figure 1 illustrates the X-ray diffraction pattern of the two types of LiOH?H2O crystals
studied in this research. Differences for ? = 23.5° and 30.5° are observed. Differential thermal analysis (DTA)
of the crystal II was used to determine the weight loss in the crystal II. Water as the only source for weight loss
was concluded (no ethanol detected). Infrared spectroscopy analysis of the crystal II revealed absence of C-H
groups in the range band of 400 to 4000 cm-1. Therefore both DTA and IR analysis, demonstrates the absence
of ethanol in crystal II mass.



Fig. 1 X-ray diffraction pattern of re-crystallized LiOH?H2O (Crystal II) ?, and crystalline LiOH?H2O
(Crystal I) ?. (Online color at www.crt-journal.org)

Crystallization of LiOH?H2O by evaporation Table 1 presents the crystal size distribution CSD results.

Table 1 Accumulated and retentions mass percentages in crystallization by evaporation.

Mesh #
Aperture L (mm)
% Accumulated
% Retained
18 1.000 0.12 0.12
20 0.850 0.52 0.39
30 0.600 1.52 1.01
40 0.425 9.38 7.86
50 0.300 31.43
22.05
60 0.250 56.29
24.85
70 0.212 68.57
12.28
100 0.150 79.10 10.54
120 0.125 92.85 13.75
140 0.106 96.68 3.83
170 0.090 98.31 1.62
200 0.075 99.34 1.03
Bottom 0.000
100.00 0.66

© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.crt-journal.org

---

Cryst. Res. Technol. 43, No. 6 (2008)
619





Fig. 2 Accumulated mass percent v/s crystal size.
Fig. 3 Solubility of LiOH?H2O vs ethanol-water ratio at
25°C. ? = LiOH?H2O Crystal I, ? = LiOH?H2O Crystal II.

The mean sizes crystals were obtained by adjusting the accumulated percentage data as a function of the mesh
aperture L (mm), applying an asymmetrical exponential equation [17] as follows:

i
i
[100 ? (100L) ]h
% Accumulated =
(

(1)
h?i? )
1
100

The experimental data fit gave h = 26.3949; i =2.7056. Figure 2 shows the experimental data and the obtained
by equation 1. Using the equation 1 at 50 % crystal accumulation, a mean crystal size was estimated of 0.259
mm.
LiOH?H2O crystallization by co-solvent addition (ethanol) Table 2 presents the chemical analyses
results and the crystal masses obtained, by different rates of 7-60 cm3/min of ethanol solution (70 and 90 %
ethanol) added to the master solution.

Table 2 Mean crystallization results by addition of co-solvent.

Measurement
% Ethanol
Ethanol Flow (cm3/min) (%
LiOH)
Sol Crystal
Mass
(g)
1 70
60
5.71 52.50
2 70
30
5.12 45.10
3 70
7
5.65 50.99
4 90
60
4.66 68.32
5 90
30
5.19 68.41
6 90
7
5.13 70.20

For all (duplicate) measurements, the mean standard deviations of chemical analyses and masses of crystals
(dried in warm air) were 0.17 and 2.11 respectively. Table 3 to 6 present results of crystal size distribution,
CSD measurements.

Table 3 Accumulated percentages for measurements made with 70 % ethanol. Note: cm3/min refers to rate
of addition of solvent to master solution.

% Accumulated mean
% Accumulated mean
% Accumulated mean
Mesh #
Aperture, mm
60 cm3/min*
30 cm3/min*
7 cm3/min*
20 0.850
0.58
0.72
1.16
30 0.600
2.12
1.22
3.88
40 0.450
16.97
21.60
41.95
50 0.300
63.92
48.71
69.61
60 0.250
78.52
82.02
83.41
70 0.212
90.46
90.56
91.77
100 0.150
97.17
97.24
97.44
120 0.125
99.21
99.02
98.77
140 0.106
99.80
99.59
99.51
170 0.090
99.89
99.84
99.86
200 0.075
99.95
99.93
99.95
400 0.038
99.99
100.00
100.00
Bottom 0.000
100.00
100.00
100.00

www.crt-journal.org
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---

620
T. A. Graber et al.: LiOH?H2O crystals obtained by evaporation and by drowning out

Table 4 Retained percentages for measurements made with 70 % ethanol. Note: cm3/min refers to rate of
addition of solvent to master solution.

% Mean Retained
% Mean Retained
% Mean Retained
Mesh #
Aperture, mm
60 cm3/min*
30 cm3/min*
7 cm3/min*
20 0.850
0.58
0.72
1.16
30 0.600
1.54
0.50
2.72
40 0.450
14.85
20.39
38.07
50 0.300
46.50
27.11
27.66
60 0.250
14.60
33.31
13.80
70 0.212
11.93
8.54
8.36
100 0.150
6.72
6.68
5.66
120 0.125
2.03
1.77
1.33
140 0.106
0.59
0.57
0.74
170 0.090
0.09
0.26
0.35
200 0.075
0.06
0.09
0.09
400 0.038
0.04
0.06
0.05
Bottom 0.000
0.01
0.00
0.00

Table 5 Accumulated percentages for measurements made with 90 % ethanol. Note: cm3/min refers to rate
of addition of solvent to master solution.

% Accumulated mean
% Accumulated mean
% Accumulated mean
Mesh #
Aperture, mm
60 cm3/min*
30 cm3/min*
7 cm3/min*
20 0.850
0.75
0.76
0.83
30 0.600
7.87
2.05
3.28
40 0.450
45.41
7.47
19.1
50 0.300
79.31
48.96
64.07
60 0.250
89.66
77.25
78.26
70 0.212
95.33
87.01
87.44
100 0.150
98.83
94.89
96.18
120 0.125
99.45
97.18
98.63
140 0.106
99.74
98.23
99.48
170 0.090
99.89
99.13
99.75
200 0.075
99.96
99.66
99.95
400 0.038
100
100
100
Bottom 0.000
100
100
100

Table 6 Retained percentages for measurements made with 90 % de ethanol. * Note: cm3/min refers to rate
of addition of solvent to master solution.

% Mean retained
% Mean retained
% Mean retained
Mesh #
Aperture
60 cm3/min*
30 cm3/min*
7 cm3/min*
20 0.850
0.75
0.76
0.83
30 0.600
7.12
1.29
2.46
40 0.450
37.55
5.42
15.82
50 0.300
33.91
41.49
44.98
60 0.250
10.35
28.29
14.19
70 0.212
5.67
9.76
9.19
100 0.150
3.51
7.88
8.73
120 0.125
0.62
2.30
2.46
140 0.106
0.29
1.05
0.85
170 0.090
0.14
0.90
0.27
200 0.075
0.07
0.53
0.20
400 0.038
0.05
0.33
0.05
Bottom 0.000
0.00
0.00
0.00

Table 7 Coefficients for Equation (1).

Measurement Flow
cm3/min h
i
70 % Ethanol
1 60
28.8175
3.5016
2 30
27.9426
3.3617
3 7
9.7868
2.9232
90 % Ethanol
4 60
12.4493
3.4321
5 30
42.1504
3.5175
6 7
21.2313
3.2388
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.crt-journal.org

---

Cryst. Res. Technol. 43, No. 6 (2008)
621

The mean standard deviations were 2.53 and 2.98; the maximum standard deviations were 5.97 and 6.06, for
CSD expressed as retained and accumulated percentages, respectively. The mean sizes of the crystals were
determined using a procedure similar to that employed for the crystals obtained by simple evaporation. Table 7
presents the results of parameters "h" and "i" obtained by fitting the data to equation (1). Using the equation 1
at 50 % crystal accumulation, a mean crystal size was estimated, as presented in table 8.

Table 8 Mean sizes obtained for a 50 % accumulation of crystals.

Measurement L
(mm)
1 0.34
2 0.33
3 0.40
4 0.43
5 0.31
6 0.35

In all the measurements, microscopic examinations revealed crystals agglomeration above # 20 mesh in size.
The maximum deviation for the retained fraction with a significant crystal agglomeration corresponded at mesh
# 50 (aperture 0.3 mm) it’s using 90 % ethanol and a flow of 60 cm3/min. This also explains the biggest crystal
mean size (Lav) obtained in the experiences with drowning-out with alcohol (Lav > 0.30 mm) in comparation
with evaporation (Lav = 0.259 mm).
The main advantage of the co-solvent process is crystal production without vacuum application at room
temperature is. As described by Pina et al. [18], the addition of ethanol contributed to dielectric constant
reduction of the system and also produced changes in the mobility and ions solvation. The decrease in the
dielectric constant favors ionic association and the formation of insoluble ionic species. Thus the effects
produced by the addition of co-solvent leads to increasing the forces of ionic cohesion, promoting precipitation
and agglomeration.
Physical properties Experimental solubilities of Crystal II are listed in table 9 for different ethanol-water
mixtures at 25 °C. Also values for density, refractive index, viscosity and ionic conductivity of the saturated
solutions are listed. Previously obtained values for Crystal I [7] have been included in the table for comparison.

Table 9 Properties of the LiOH?H2O (1) + ethanol (2) + H2O (3) ternary system at 25 ºC.

Solubility
Density
Refractive index
Absolute Viscosity
Conductivity
100w2
w2/w3
(g/100 g solution)
(g/ cm3)
nD
µ
?S/cm
Crystal II
0.0 0.00
10.56
1.11175
1.3734
3.647
396.2
9.1 0.11
7.97
1.07205
1.3707
3.884
268.6
18.8 0.25
6.12
1.03487
1.3694
4.065
169.6
27.3 0.40
4.30
1.00109
1.3689
3.998
107.0
38.8 0.67
3.40
0.96794
1.3685
3.852
58.6
Crystal I
0.00 0.00
11.03
1.11192
1.3735
3.614
395.1
9.10 0.11
8.48
1.07215
1.3708
3.803
264.2
18.70 0.25
6.35
1.03537
1.3695
4.060
171.0
27.20 0.40
4.97
1.00572
1.3690
4.091
111.9
38.60 0.67
3.66
0.96911
1.3686
3.918
67.0

This table demonstrates that the properties of saturated solutions of both crystals show no significant
differences, although this does not hold for the solubility. Figure 3 shows the differences in solubility between
the two crystal types. Experimental solubilities values and physical properties of crystals I and II, are presented
in tables 10 and 11 respectively, for water and ethanol mixtures at 15, 20, 25, 30 and 35°C. The data for Crystal
I were previously published in [8]. As in the above comparison (binary systems) the only differences were
noted in solubility, suggesting that polymorphism was also present in the ternary systems.
Thermodynamic analysis Park et al. [11], in classifying diverse polymorphic crystalline systems, plotted
experimental data of solubility versus the reciprocal of the absolute temperature (1/T), following the van´t Hoff
equation. This procedure considers that the solution behaves as an ideal solution and agrees with an assumption
by which the activity coefficient of the solute is equal to unity. Nevertheless, systems containing electrolytes
have a behavior very divergent from ideal, and thus in the thermodynamic analysis an estimation of the activity
coefficient of the solute is required. In the present study we used the Chen model [12,13] to determine the
www.crt-journal.org
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---

622
T. A. Graber et al.: LiOH?H2O crystals obtained by evaporation and by drowning out

mean activity coefficient (?±) for the two crystalline forms of LiOH·H2O. The present study uses a graphic
representation of the van´t Hoff equation for estimation of the enthalpies of the solutions, while the diagram of
?G versus temperature has been employed to classify the polymorphic system.

Table 10 Saturated solutions of binary (Crystal I) and ternary (Crystal II) LiOH (1) + ethanol (2) + H2O (3)
systems at different temperatures.

Temperature
Solubility
Density
Refractive Index
(ºC)
(g/100 g solution)
(g/ cm3)
nD
Binary System (water)
15 11.00
1.11437
1.3744
20 11.07
1.11298
1.3739
25 11.14
1.11196
1.3734
30 11.24
1.11093
1.3730
35 11.33
1.11023
1.3727
Ternary system: ethanol/water = 0.1
15 8.48
1.07499
1.3721
20 8.51
1.07345
1.3716
25 8.67
1.07193
1.3709
30 8.64
1.07067
1.3703
35 8.94
1.06940
1.3697

Table 11 Saturated solutions of binary and ternary: LiOH (1) + ethanol (2) + H2O (3) systems for different
temperatures for Crystal II.

Temperature
Solubility
Density
Refractive index
(ºC)
(g/100 g solution)
(g/ cm3)
nD
Binary system (water)
15 10.40
1.11426
1.3743
20 10.51
1.11311
1.3739
25 10.66
1.11175
1.3734
30 10.77
1.11040
1.3730
35 10.97
1.10982
1.3726
Ternary system ethanol/water = 0.1
15 7.58
1.07443
1.3717
20 8.06
1.07300
1.3712
25 8.14
1.07205
1.3707
30 8.63
1.07076
1.3705
35 8.82
1.06877
1.3699

Binary systems The thermodynamic equilibrium for the solubility reaction of the LiOH·H2O can be
expressed through the solubility product (Kps):

2
Kps = (m ?± ) a
s
w , (2)

where ms and aw represent the molality at saturation and the water activity, respectively. In table 12, Chen
model parameters (?ij ?ji and ?ij) are presented for LiOH + H2O system at 25 ºC, with parameters expressed in
molality units. The values of ?± and aw was estimated from Chen model and Kps was calculated following
equation (2). The same method to obtain the values at other temperatures was used. The Chen parameters are
only known for 25°C, and in the range used (15-35°C) the temperature effect was neglected.

Table 12 Binary parameters at 25 ºC, and nonrandomness factors.

Binary pair
? 12
? 21
? 12
Crystal
I
Water (1)-Ethanol (2)a 0.5702
-0.1640
0.1803
LiOH (1) - Water (2)b -4.4000
9.0080
0.2000
LiOH (1) - Ethanol (2)c -2.3628 7.5632
0.0366
Crystal
II
Water (1)-Ethanol (2)a 0.5702
-0.1640
0.1803
LiOH (1) - Water (2)b -4.4000
9.0080
0.2000
LiOH (1)- Ethanol (2)d -2.4158 7.5871
0.0330

a From Barata and Serrano [19], b From Chen et al. [12],
c From Taboada et al. [7], d Optimized in this work.
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.crt-journal.org

---

Cryst. Res. Technol. 43, No. 6 (2008)
623

Figure 4 presents a ln(Kps) versus 1/T plot for the two types of crystals, according to the van't Hoff equation.
This figure shows that Crystal II has the lowest Kps values (less soluble), and therefore these crystals are more
stable than Type I crystals. It follows that transformation of the Type I crystal into Type II is possible. The
van't Hoff curves correlated with the temperature function as shown by equation [20]:

2
? 1 ?
? 1
ln Kps A
B
?
=
+
+ C
?
(3)
T ?
? T ?
? ?
? ?





Fig. 4 Van´t Hoff plot of Crystal I and Crystal II between
Fig. 5 The ?G versus temperature diagram for
15 - 35 ºC in water.
LiOH·H2O polymorphs in water.

A, B and C parameters for Crystal I and Crystal II obtained by regression are shown in table 13.

Table 13 Values for the coefficients in Eq. (3).

Crystal A
B C
Binary
System
I 212825
-1502.8
4.3808
II 609608
-4410.7
9.5561
Ternary
System
I -265071
1535.8
-0.3326
II -355388
1150.9
1.8521

The slope of the van't Hoff curves is given by

vH
d lnKps
?HSoln
=
, (4)
2
dT
RT

where
vH
?HSoln is van´t Hoff solution enthalpy. By derivation of equation 4 and insertion into equation 5, the
values of
vH
?HSoln can be estimated for each temperature, using the following expression:

vH
?
1
?H
R 2A
B?
= ?
+
Soln
?
(5)
T
?
?
?

Table 14 presents the values of
vH
?HSoln for the two types of crystals. Based on the values in this table it is
observed that the enthalpy values of the solution for Crystal II are greater than those for Crystal I for the five
temperatures studied. This indicated that the crystal less soluble in water had a greater solution enthalpy.
The difference in Gibbs free energy associated with the transformation of Crystal I (less stable) to Crystal II
(more stable) is given by:

? (?
?
± ms )II
?G = ?RTln ?

(6)
? (
?
? ?
?
± ms )I ?

?G for each temperature was determined using equation 6. The results are presented in figure 5. Extrapolating
these data to reach a value of ?G = 0 (reversible transition) allows us to estimate the transition temperature;
following this procedure the transition temperature in the present case was 344 K.
www.crt-journal.org
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---

624
T. A. Graber et al.: LiOH?H2O crystals obtained by evaporation and by drowning out

Monnin and Dubois [21] reported a fusion temperature for LiOH·H2O of 383 K, and thus the present
polymorphic crystalline system can be classified as enantiotropic.

Table 14
vH
?H
values for Crystals I and II in water and water + ethanol (ethanol/water ratio = 0.1).
Soln

vH

?H
(J·mol-1)
Soln
Crystal
288.15 K
293.15 K
298.15 K
303.15 K
308.15 K
Binary
System
I 208.38 413.31 611.38 802.90 988.22
II 1460.17 2047.18 2614.51 3163.12 3693.92
Ternary
System
I 2472.84 2217.59 1970.91 1732.36 1501.56
II 10702.62
10360.40
10029.66
9709.84 9400.39

Ternary systems The enthalpies of the two crystal solutions in ethanol + water ( ethanol/water ratio =
0.1) are determined as for binary systems, taking into account that in Eq. 2, both ?± and aw must be evaluated in
the ternary system. The latter suggests that nine interaction parameters must be known, which are associated
with the LiOH + water, water + ethanol and LiOH + ethanol binary systems. The parameters for the first two
systems are available in the literature as shown in table 12. Parameters ?ij, ?ji and ?ij for the LiOH + ethanol
system can be estimated by regression. The regression is carried out on the experimental solubility data of the
LiOH in various water-ethanol mixtures at 25°C (Table 10). This procedure was previously applied [7] for
Crystal I, and its result is listed in table 13.
The values obtained for the LiOH + ethanol pair, associated with Crystal II, are determined following the
same procedure used for Crystal I and are shown in table 13. Given that the temperature interval is small (15 –
35°C), it was supposed that the 9 parameters of the Chen model at 25 ºC have the same values as for the other
temperatures. Figure 6 shows lnKps versus 1/T for the two crystals. Given that the solubility data for both
crystals showed a degree of dispersion, these were smoothed. This plot shows that the Kps values were lower
for Crystal II, similarly to what occurred in the binary systems, and thus Crystal II remained in the more stable
form. It was also observed that the transition would probably occur from Crystal I to Crystal II at about a
temperature of 308 K (1/T = 0.00325).





Fig. 6 The van´t Hoff plot of Crystal I and Crystal II between
Fig. 7 The ?G versus temperature diagram for
15 - 35 ºC in water + ethanol (ethanol/water ratio = 0.1).
LiOH·H2O polymorphs in water + ethanol
(ethanol/ water ratio = 0.1).

Following a similar procedure to that used in binary systems, the ln(Kps) data were fit to a second-order
polynomial, and then the values for
vH
?HSoln were estimated for both crystals. Table 14 presents the values of
the coefficients A, B and C, and Table 15 presents the values for the solution enthalpies for both types of
crystals. In this table it can be noted that the values of
vH
?HSoln follow the same tendency as observed in the
binary systems i.e. the values of the solution enthalpies of Crystal II are greater for all the temperatures. This
suggest what lowest solubility that the crystal in water + ethanol (ethanol/water ratio = 0.1) had the greater
solution enthalpy.
Equation (6) was used to determine the transition temperature, taking into account now that the values of ?±
are determined in the ternary system. Figure 7 shows the ?G v/s T diagram, and the transition temperature was
determined to be 308 K by process extrapolation. Since this temperature is lower than the fusion temperature, it
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.crt-journal.org

---

Cryst. Res. Technol. 43, No. 6 (2008)
625

was concluded that the relation between Crystal I and Crystal II is enantiotropic, as occurred in the binary
systems.
4 Conclusions
? In the binary and ternary systems, the solubility of lithium hydroxide increases with temperature.
? The solubility decreases considerably in the presence of ethanol, producing a drowning-out crystallization
by the addition of alcohol to an aqueous solution.
? In both systems, the density of the saturated solutions decreases when temperature increases approaching
linearity and in the presence of the alcohol the density dependence with temperature is greatly exacerbated.
? The refractive index of the saturated solution in both systems linearly decreases with an increase in
temperature. For the ternary system this behavior is partiatilly attenuated in the presence of the alcohol.
? The conductivity of the saturated solution in both systems increases with increase in temperature linearly.
For ternary system, the effects of the presence of the alcohol include a decrease in conductivity.
? The viscosity of the saturated solution of both systems linearly decreases with an increase in temperature.
Viscosity curves intercept each other upon reaching 35°C. The viscosity slightly decreases in the presence
of the alcohol. In the case of the ternary system this dependence is suppressed.
? Considering that the narrow mean sizes of type II crystals were between 0.3 and 0.4 mm it was concluded
that the effects of the both percentage of ethanol and the flow rates is negligible.
? The advantage of crystallization by co-solvent addition is that the process can be carried out at a room
temperature and does not require the use of vacuum.
? A polymorphic solid whose solubility differs from crystals produced by simple evaporation is obtained
under a crystallization process in the presence of ethanol.
? The LiOH·H2O crystals and their dissolved polymorphs in water or water + ethanol (ethanol/water ratio =
0.1) have an enantiotropic behavior.
? The solution enthalpy values were always greater for Crystal II (less soluble) for the five working
temperatures employed.

Acknowledgements The authors are grateful for financing provided by CONICYT-Chile, through Fondecyt Project Nº
1040299.
References
[1]
Y. Cui, Int. J. Pharmaceut. 45, 339 (2007).
[2]
S. Lahani and D. J. Grant, “Thermodynamics of polymorphs” In: Rolf, Hilfiker (Ed.),Wiley-VCH Verlag GmbH &
Co. KGaA Weinheim, Germany 2006, pp. 21–41.
[3]
R. D. Rogers, Cryst. Growth Des. 3, 867 (2003).
[4]
S. Maruyama and Ooshima, Chem. Eng. J. 81, 1 (2001).
[5]
D. J. W. Grant, “Theory and origin of polymorphism”, in: Brittain, H.G. (Ed.), Polymorphism in Pharmaceutical
Solids. Marcel Dekker, New York 1999, pp. 1–32.
[6]
V. P. McMahon, J. A. Bis, and J. A. Zaworotko, J. Pharm. Sci. 95, 499 (2006).
[7]
M. E. Taboada, H. R. Galleguillos, T. A. Graber, and J. Alvarez-Benedí, Fluid Phase Equil. 235, 104 (2005).
[8]
M. E. Taboada, D. M. Véliz, H. R. Galleguillos, and T. A. Graber, J. Chem. Eng. Data 50, 187 (2005).
[9]
A. Burger and R. Ramberger, Mikrochim. Acta 2, 259 (1979).
[10]
A. Burger and R. Ramberger, Mikrochim. Acta 2, 273 (1979).
[11]
K. Park, J. M. Evans, and A. S. Myerson, Cryst. Growth Des. 3, 991 (2003).
[12]
C.-C. Chen, H. I. Britt, J. F. Boston, and L. B. Evans, AIChE J. 28, 588 (1982).
[13]
C. C. Chen and L. B. Evans, AIChE J. 32, 444 (1986).
[14]
A. C. Schmidt and I. Schwarz, J. Mol. Struct. 748, 153 (2005).
[15]
B. Lazic, V. Kahlenberg, J. Konzett, and R. Kaindl, Solid State Sci. 8, 589 (2006).
[16]
G. Coquerel, Chem. Eng. Proc. 45, 857 (2006).
[17]
J. Dryzmala and A. Hussin, Int. J. Min. Proc. 76, 55 (2005).
[18]
C. M. Pina., L. Fernández-Díaz, M. Prieto, and S. Veintemillas-Verdaguer, J. Cryst. Growth 222, 317 (2001).
[19]
P. A. Barata and M. L. Serrano, Fluid Phase Equil. 141, 247 (1997).
[20]
F. L. Nordstrom and A. C. Rasmuson, Eur. J. Pharm. Sci. 28, 377 (2006).
[21]
C. Monnin and M. Dubois, J. Chem. Eng. Data 50, 1109 (2005).
www.crt-journal.org
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

---