Cryst. Res. Technol. 51, No. 6, 405–408 (2016) / DOI 10.1002/crat.
201600092
Influences of ethanol on the thermodynamics and kinetics in
the crystallization of xylitol
Anne Hartwig and Joachim Ulrich∗
Received 22 March 2016, revised 13 May 2016, accepted 26 May 2016
Published online 17 June 2016
The metastable zone and crystal growth rates of xylitol in
water were investigated under the influence of small vol-
ume fractions of ethanol. The utilization of low amounts
of ethanol is necessary to help adding slightly soluble in-
gredients during the fabrication of xylitol products, e.g. in
food industry. Therefore, the metastable zone and crystal
growth rates were determined by means of an ultrasound
measurement technique and batch crystallization exper-
iments. It can clearly be seen that the addition of ethanol
is lowering the solubility of xylitol in water. The higher the
volume fraction of ethanol, the lower the solubility. The
metastable zone width is increased whereas the smallest in-
vestigated ethanol content results in the widest metastable
zone. The crystal growth of xylitol is inhibited by the ad-
dition of ethanol whereby the crystal growth rates are de-
creasing with increasing ethanol content.
1 Introduction
Xylitol is a pentavalent sugar alcohol that has many dif-
ferent benefits like its applicability for diabetic prod-
ucts due to its insulin independent metabolism. Further-
more, it has an anticariogenic effect that was proven by
the reduction of plaque and tooth decay while apply-
ing xylitol products. Those products, like tooth paste or
chewing gum, also profit by the negative heat of solution
of xylitol which creates a refreshing taste during the ap-
plication of the products [1].
The solubility and crystallization of xylitol in the com-
monly used aqueous solution is of great importance for
the fabrication of xylitol products. It is necessary to know
how the applied solution behaves to control the different
process steps. Hao et al. [2] have shown that the addition
of a further solvent, in this case methanol, has a great in-
fluence on the solubility of xylitol in water. The addition
of e.g. alcohol during the fabrication of xylitol products
Original Paper
would yet be necessary to dissolve poorly soluble com-
ponents (e.g. menthol) in the production solution. Due
to the application of xylitol products in food and phar-
maceutical industry and the frequently used low concen-
trations (ࣘ 5 vol%) of ethanol, such cases were inves-
tigated concerning their influence on the solubility and
crystallization behavior of xylitol in water.
2 Experimental
Diverse aqueous xylitol solutions with a mass fraction
between 60 and 82 wt% were prepared for the exper-
iments. To investigate the influence of ethanol 1, 3 or
5 vol% ethanol (purity ࣙ 99.8 %) were added during the
preparation of the solutions. To obtain seed crystals with
the required particle size, industrially provided xylitol
crystals were sieved and the corresponding sieve frac-
tions were used.
2.1 Ultrasound measurements
The solubility and metastable zone width of xylitol in
water were gathered by means of an ultrasonic device,
which was previously described by e.g. Omar et al. [3].
Therefore, the solutions were prepared in a double-
walled beaker which was temperature controlled by a
thermostat. The crystallization was induced by seeding
[4] by adding 500 mg xylitol seed crystals with a particle
size between 500 and 1000 μm. The crystals were added
2 K below the saturation temperature and then the solu-
tion was cooled and heated up again with the same rate
of 2 K/h. The differences in ultrasound velocity as a func-
tion of the temperature were measured by the ultrasonic
∗ Corresponding author: E-mail: Joachim.ulrich@iw.uni-halle.de,
Phone: +49 345 55 284 00, Fax: +49 345 55 273 58
Center for Engineering Sciences, Thermal Process Engineering, Martin
Luther University Halle-Wittenberg, D-06099 Halle, Germany

C 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 405
 Anne Hartwig and Joachim Ulrich: Influences of ethanol on the thermodynamics and kinetics in the crystallization of xylitol
Original Paper
probe. To prevent a disturbance of the detected signal by
the seed crystals, the probe was enclosed by a sieve cage.
Thus only the nucleation and not the seed crystals them-
selves could be registered.
2.2 Crystal growth
Previous growth experiments were often carried out us-
ing a microscopic cell as described by e.g. Gougazeh et al.
[5]. The crystal growth rates were calculated as the length
increase of a growing crystal face as function of time. This
method was proven to be inaccurate due to the varying
growth rates of each individual crystal. Even though a
large number of crystals has been investigated this ex-
periment led to very high standard deviations and thus
incomparable results [6]. This phenomenon known as
growth rate dispersion is explicitly described in the lit-
erature [7, 8].
To bypass this problems the crystal growth rates were
determined by batch experiments considering a high
quantity of crystals. Therefore, 50 mL of a xylitol solution
was saturated at 25°C in a double-walled beaker. Super-
saturated states were achieved by cooling the solution
to 24, 22 and 20°C, respectively. After that, 500 mg xyl-
itol crystals with a size between 355 and 400 μm were
added and magnetically stirred for five minutes. After a
seed growth period, filtering and washing with ethanol
(purity ࣙ 99,8 %), the crystals were dried at 60°C in a
compartment dryer. The growth rates were calculated
as linear surface growth rates (equation (1)) referring to
Buchfink et al. [9]. This relation derives from the linear
length growth of the crystals related to their mass in-
crease over time.
  13 
L1 m2
G= −1 (1)
t m1
G is the averaged crystal growth rate, L1 is the arithmetic
mean of the seed crystal size, t is the retention time in
the batch crystallizer, m2 is the crystal mass after their
growth and m1 is the initial seed crystal mass.
3 Results and Discussion
3.1 Metastable zone
The solubility and corresponding nucleation points of
xylitol in water and ethanol, gathered with an ultrasonic
probe, are shown in figure 1.
406 
C 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
nucleation 0 vol% ethanol
80 nucleation 1 vol% ethanol
nucleation 3 vol% ethanol
nucleation 5 vol% ethanol
75
w [wt%]
70
65
solubility 0 vol% ethanol
solubility 1 vol% ethanol
60 solubility 3 vol% ethanol
solubility 5 vol% ethanol
0 10 20 30 40 50 60
T [°C]
Fig. 1 Metastable zone of xylitol in water and ethanol.
It can be seen that the solubility of xylitol in water
is increasing with increasing temperatures, independent
of the ethanol content of the solution. But the addition
of ethanol has an apparent influence on the solubility.
The highest applied ethanol amount of 5 vol% is shifting
the solubility curve of xylitol about 3.5 K to higher tem-
peratures. Therefore, with increasing ethanol content the
solubility is decreasing or in other words, the solubil-
ity temperatures are increasing. Though this thermody-
namic effect can already be observed for a low ethanol
content of 1 vol%, ethanol has to be seen as a co-solvent
rather than as impurity or additive.
Furthermore, it is clearly recognizable that the addi-
tion of different amounts of ethanol has different effects
on the nucleation points and thus on the width of the
metastable zone. To clarify the differences, the solubility
and nucleation data plotted in a three-dimensional dia-
gram as a function of concentration and ethanol content
are shown in figure 2.
Overall it can be seen that the metastable zone be-
comes more narrow with increasing temperature and
concentration, independent of the ethanol content. By
adding ethanol the metastable zone width increases.
Omar et al. [10] have shown that there is a strong cor-
relation between solvent effect on the solubility and the
metastable zone width. The lower the solubility in a sol-
vent the wider is the metastable zone due to the in-
creased energy that is necessary to initiate the nucle-
ation.
Though the addition of ethanol is decreasing the
solubility, the metastable zone is enlarged for all the
tested ethanol amounts. However, the widening of the
metastable zone itself is depending on the ethanol con-
centration. The data for the metastable zone widths are
summarized in table 1. It was found that the lowest used
www.crt-journal.org
Cryst. Res. Technol. 51, No. 6, 405–408 (2016)

Original Paper
60 5x10-7
0 vol% ethanol, 1 K supersaturation

-7
50 4x10

40
3x10-7

T [°C]

G [m/s]
30
2x10-7
20

1x10-7
10
80
5
75
eth 3 0
ano 70 ] 0 10 20 30 40 50
l co 12 t%
nte
nt [ 65 [w solubility data t [min]
w
vol%
] 0 60 nucleation data
Fig. 3 Crystal growth rate of xylitol versus the time.
Fig. 2 Three-dimensional plot of the metastable zone of xylitol in
water as function of the ethanol content. lower the viscosity and the lower the metastable zone
width.
The effect in the solid phase (increase of metastable
Table 1 Metastable zone widths of xylitol in water depending
zone width) and the effects in the liquid phase (decrease
on the ethanol content.
of metastable zone width) act contrary. The superposi-
Ethanol Metastable Standard tion of these opposing phenomena leads to a minimum
content [vol%] zone width [K] deviation [K] effect at low ethanol concentrations and results in a lo-
0 5.63 1.54 cal maximum of the metastable zone width (table 1). The
addition of ethanol increases the metastable zone width
1 9.11 1.61 due to the inhibited nucleation. However, the interfering
3 8.61 1.68 effects in the liquid phase have obviously more impact
on the metastable zone. Thus, an increase of the ethanol
5 7.52 1.61
content results overall in a decrease of the metastable
zone width.

ethanol amount of 1 vol% is enlarging the metastable

zone from 5.63 to 9.11 K. As can be seen in figure 2 and
table 1, respectively, this represents a local maximum
of the metastable zone width as function of the ethanol
content.
The widening of the metastable zone can be ex-
plained by different effects in the solid and liquid phase.
Regarding the solid phase, ethanol inhibits the nu-
cleation. The dissolved ethanol molecules adsorb to the
clusters from which the nuclei (solid particles) will form.
This will result in an increase of the metastable zone
width (hence a suppression of nucleation).
Regarding the liquid phase, two different effects can
be described. On the one hand, the structure of the solu-
tion and the interaction between the molecules change
by adding ethanol [11]. On the other hand, a decrease
of the viscosity leads to a faster diffusion process and
reduces the metastable zone width. The addition of
ethanol is lowering the viscosity of the saturated so-
lutions. Therefore, the higher the ethanol content, the
3.2 Crystal growth rates
The relation between the crystal growth rates and the re-
tention time in the batch crystallizer was investigated.
The exemplarily progression for a 1 K subcooled pure xyl-
itol solution is shown in figure 3.
It can clearly be seen that the crystal growth rates
are decreasing with increasing time. This is caused by
the degradation of the supersaturation. When the xylitol
seed crystals are added to the supersaturated solution,
they start to grow immediately. As a result, the concen-
tration of the solution and therefore the degree of super-
saturation is decreasing. A lower supersaturation results
in a decreased driving force for the growth process and
thus, the crystal growth rates are dropping as well.
As an outcome, the retention time for the following
growth experiments was fixed to five minutes. The influ-
ence of ethanol on the growth rates depending on the su-
persaturation are shown in figure 4.

www.crt-journal.org 
C 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 407
 Anne Hartwig and Joachim Ulrich: Influences of ethanol on the thermodynamics and kinetics in the crystallization of xylitol
Original Paper
3x10-7
2x10-7
G [m/s]
1x10-7
0 vol% ethanol
1 vol% ethanol
3 vol% ethanol
5 vol% ethanol
0 ethanol is decreasing the growth rates. This is caused by
0 1 2 3 4 5
ΔT [°C]
Fig. 4 Crystal growth rates of xylitol in water and ethanol versus
the supersaturation.
It can be seen that the crystal growth rates of xylitol
are increasing with increasing supersaturation, indepen-
dent of the ethanol content. Due to the increased driving
force, the crystals grow faster with an increased degree of
subcooling.
Furthermore, the growth rates are decreasing with in-
creasing ethanol content. A thermodynamic effect can-
not be seen as a reason for that due to the experimen-
tal conditions. By using saturated solutions at the same
temperatures, the shift of the solubility curves was elimi-
nated. Thus, this phenomenon can be explained by a ki-
netic effect. The crystal growth involves different steps
like the mass transfer of the dissolved molecules from
the solution to the crystal surface and the integration
into the growth site [5]. These steps can be influenced by
adding additives or solvents. In this case, the dissolved
ethanol molecules adsorb to the surface of the xylitol
crystals and block certain crystal faces. Consequently,
with increasing ethanol content the growth rates are de-
creasing further. In general, it could be shown that al-
ready small amounts of ethanol are inhibiting the crystal
growth due to this described kinetic effect.
4. Conclusion
The solubility of xylitol in water is influenced by adding
ethanol. A higher ethanol concentration in the solution
408 
C 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
results in a lower solubility due to thermodynamic ef-
fects. In comparison the nucleation curve is, however, in-
fluenced in a different way. The addition of ethanol en-
larges the metastable zone of xylitol in water. Regarding
low ethanol concentrations the increase of the ethanol
amounts leads to a decrease of the metastable zone
width, but not smaller than the metastable zone width
of a pure xylitol solution. This phenomenon can be ex-
plained by the change of the viscosity and the inhabita-
tion of the nucleation.
Furthermore, the crystal growth rates of xylitol has
been investigated. The addition of small amounts of
the hindering of the integration step of the growth pro-
cess. The higher the ethanol content, the more crystal
faces are blocked and the slower the crystals are growing.
Key words. xylitol, metastable zone, crystal growth rate, growth
inhibition, additives.
References
[1] R. C. Rowe, P. J. Shesky, and S. C. Owen, Handbook of
pharmaceutical excipients (Pharmaceutical Press and
American Pharmacist Association, London, 2006),
p. 824.
[2] H. Hao, B. Hou, J.-K. Wang, and G. Lin, J. Cryst. Growth
290, 192 (2006).
[3] W. Omar and J. Ulrich, Cryst. Res. Technol. 34, 379
(1999).
[4] J. Ulrich and C. Strege, J. Cryst. Growth 237–239, 2130
(2002).
[5] M. Gougazeh, W. Omar, and J. Ulrich, Cryst. Res. Tech-
nol. 44, 1205 (2009).
[6] A. Hartwig and J. Ulrich, in: Proceedings 22nd Inter-
national Workshop on Industrial Crystallization, Dae-
jeon, South Korea, 2015 (K.-J. Kim, K. Lee, Daejeon, ),
pp. 122–129 (2015).
[7] J. W. Mullin, Crystallization (Butterworth Heinemann,
Oxford,), p. 254 (2001).
[8] J. Ulrich, Cryst. Res. Technol. 24, 249 (1989).
[9] R. Buchfink, C. Schmidt, and J. Ulrich, CrystEng-
Comm 13, 1118 (2011).
[10] W. Omar and J. Ulrich, in: Proceedings 11th In-
ternational Workshop on Industrial Crystallization,
Gyeongju, South Korea, 2004 (K.-J. Kim, Daejeon), pp.
227–236 (2004).
[11] S. Titiz-Sargut and J. Ulrich, Cryst. Growth Des. 2, 371
(2002).
www.crt-journal.org